Removal Of Soluble Bod5 In Primary Clarifiers

EPA-60Q/2-76-221 September 1976 REMOVAL OF SOLUBLE BOD IN PRIMARY CLARIFIERS George A. Dubey Averill J. Wiley John W.•Collins Effluent Processes Group The Institute of Paper Chemistry Appleton, Wisconsin Grant Number R-803119 Project Officer Ralph H. Scott Food & Wood Products Branch Industrial Environmental Research Laboratory-Cincinnati Corvallis, Oregon 97330 INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY OFFICE OF RESEARCH AND DEVELOPMENT U.S. ENVIRONMENTAL PROTECTION AGENCY CINCINNATI, OHIO LIBRARY S. ENVIfi...,.,i..-.,;\L P - '« \ . 08817 ------- DISCLAIMER This report has been reviewed "by the Industrial Environmental Research Laboratory, U.S. Environmental Protection Agency, and approved for publica- tion. Approval does not signify that the contents necessarily reflect the vievs and policies of the U.S. Environmental Protection Agency, nor does mention of trade names or commercial products constitute endorsement or re- commendation for use. ii ------- FOREWORD When energy and material resources are extracted, processed, converted, and used, the related pollutional impacts on our environ- ment and even on our health often require that new and increasingly more efficient pollution control methods be used. The Industrial Environmental Research Laboratory - Cincinnati (lERL-Ci) assists in developing and demonstrating new and improved methodologies that will meet these needs both efficiently and economically. Removal of soluble BODc in primary clarifiers presents research findings related to means and methods for improving the efficiency of primary sedimentation systems treating pulp and paper wastes. The possibility of greater BOD removal was investigated by examining effects of flocculants and coagulants in removing soluble and colloidal constituents of such wastes. The report also discloses that certain process waste streams may be antagonistic to efficient sedimentation and should be treated separately from total mill wastes. This information will be of interest to segments of the industry wishing to upgrade primary treatment efficiency to reduce loadings on secondary treatment. The findings are also applicable to treatment of waste streams not containing significant additives, as the tissue in- dustry, wherein effluent guide-line allowances may possibly be satis- fied by efficient primary treatment. For further information contact the Food and Wood Products Branch, Industrial Environmental Research Laboratory-Cincinnati. David G. Stephan Director Industrial Environmental Research Laboratory Cincinnati 111 ------- ABSTRACT Primary clarifiers used in the pulp and paper industry show wide varia- tions in BOD removal. This project was directed to evaluating means for in- creasing BOD removal and developing an understanding of the mechanism. Such understanding should permit optimization of BOD removal during primary clarifi- cation . A survey involved twelve mills, nine with primary clarifiers, to obtain data on total and soluble BODs, COD, suspended solids, and color. These data were used to select mill effluents for additional study. Laboratory studies showed that, with the proper flocculating agent and operating conditions, the BOD5 concentration of some mill effluents could be markedly reduced during clarification. These findings were confirmed by pilot- scale clarifier field trials at two mills. Measurement of sludge volumes and sludge dewatering characteristics re- sulting from chemical treatment were outside the objectives of this study. Gel chromatography studies showed that considerable low molecular weight biodegradable residues along with colloidal material were also flocculated and removed. Studies with model compounds indicated that increased removal is apparently related to the pH of the solution and to the functional groups (hy- droxyl, phenolic, and carboxylic), chain length, branching, and solubility of the compound. Chemical costs may range from 3<£ to 10$ per 1000 gallons of feed to the clarifier. Decreased cost and increased operating efficiency in total treat- ment systems, as a result of improved clarification, could more than justify this chemical charge. Minimum levels of chemical costs will depend upon mini- mizing or eliminating overflows or spills of colloids and dispersants detri- mental to clarifier efficiency. This report was submitted in fulfillment of Project No. R803-119 by The Institute of Paper Chemistry Effluent Processes Group under the partial sponsorship of the Environmental Protection Agency. Work was completed as of August, 1975. IV ------- CONTENTS Page Abstract iv List of Figures vi List of Tables x Acknowledgments xi-v I Introduction 1 II Summary and Conclusions 3 III Recommendations 5 IV Mill Surveys 6 V Laboratory Study of Processes for Higher Levels of Soluble BOD5 Removal 20 VI Pilot Scale Clarifier Studies 1;9 VII Jar Tests with Individual Sewer Discharges 67 VIII Discussion of the Mechanisms of Soluble BODs Removal 77 IX Areas for Cost Reductions 96 X References 98 XI Glossary 100 XII Appendices 101 ------- FIGURES Number Page 1 Schematic for processing clarifier samples for soluble BOD5 study ........................ 11 2 The separation of insoluble from soluble BODs by filters of various pore sizes ................... 13 3 Total biochemical oxygen demand in clarifier influent and effluent samples ..................... l6 U Total chemical oxygen demand in clarifier influent and effluent samples ..................... IT 5 Suspended solids in clarifier influent and effluent collected on a 0.^5 ym filter ............... 18 6 Suspended solids and BODs in various concentrations of influents treated with Fe3 or alum ............ 33 7 In jar tests with Mosinee mill effluent, pH 8.5s ferric chloride and ferric sulfate at 25 and 50 mg/1 of Fe3+ (aj , (b_) , ( ------- 15 Jar tests with Combined Locks individual sewer effluents. . . 7^ l6 Jar tests on wastes from individual sewers in the Combined Locks mill 75 17 Jar tests on special composites of Combined Locks individual waste streams 76 18 TOG and molecular weight distribution of gel-chromatographed chemimechanical mill effluent before and after coagulation with 70 mg/1 Fe3+ as ferric chloride 78 19 Phenol sulfuric acid test for aldoses and molecular weight determination of gel-chromatographed chemimechanical mill effluent before and after coagulation with 10 mg/1 Walco 73C32 79 20 Ultraviolet light transmittance profile and molecular weight distribution of a gel-chromatographed solution of sodium gluconate and ferrous ammonium sulfate, pH 2 8l 21 Transmittance and molecular weight distribution of a gel- chromatographed solution of sodium gluconate and ferrous ammonium sulfate, pH 7-0 82 22 Phenol sulfuric acid test for aldoses and molecular weight distribution of gel-chromatographed chemimechanical pulp mill effluents 83 23 Effect of pulp fibers on removal of soluble compounds .... 86 2ka. Effect of kraft pulp fibers at 350 mg/1 on removal of soluble compounds with 100 mg/1 of Fe3+ (as FeCls) at various pH's (acetic acid, methyl acetate, ethyl acetate, oxalic acid, lactic acid, and methyl lactate) 88 Effect of kraft pulp fibers at 350 mg/1 on removal of soluble compounds with 100 mg/1 of Fe3 (as FeCls) at various pH's (raffinose, hexanedioic acid, 2-pentanol, and tertiary amyl alcohol) 89 Effect of kraft fibers at 350 mg/1 on removal of soluble compounds with 100 mg/1 of Fe3+ (as FeCls) at various pH's (trans-butenedioic acid, cis-butenedioic acid, propanedioic acid, hydroxymalonic acid, benzylmalonic acid, and 1,U- butanedioic acid) 90 Effect of kraft fibers at 350 mg/1 on removal of soluble compounds with 100 mg/1 of Fe3+ (as FeCls) at various pH's (benzoic acid, p_-hydroxybenzoic acid, o_-hydroxybenzoic acid, p_-methylbenzoic acid, 3,^,5-trihydroxybenzoic acid, and 3-methoxy-U-hydroxy-benzaldehyde) 91 vii ------- 2Ue Effect of kraft fibers at 350 mg/1 on removal of soluble compounds with 100 mg/1 of Fe3+ (as FeCls) at various pH's (m-dihydroxy-benzene , o-dihydroxy-benzene, p_-hydroxy cinnamic acid, ^-hydroxy-3-methoxy cinnamic acid, D-galacturonic acid, and polygalacturonic acid) ...... 92 25 The model compounds shown in Figures 23-2ke cluster in two groups on the basis of solubility in water and removability with Fe3+ ......................... 95 B-l BODs reduction during clarification with Hercofloc ...... 107 B-2 Color and suspended solids reduction during clarification with Hercofloc ...................... 108 B-3 BODs reduction during clarification with alum ........ 109 B-4 Color and suspended solids reduction during clarification with alum ......................... 110 B-5 BODs reduction during clarification with ferric chloride. . . m B-6 Color and suspended solids reduction during clarification with ferric chloride ................... 112 B-7 BODs reduction during clarification with ferric sulfate . . . 113 B-8 Color and suspended solids reduction during clarification with ferric sulfate .................... Ill; B-9 BODs reduction during clarification with sulfuric acid, Hercofloc, or ferrous sulfate plus Hercofloc ....... 115 B-10 Color and suspended solids reduction during clarification with sulfuric acid, Hercofloc, or ferrous sulfate plus Hercofloc ......................... 116 C-l Total and soluble BODs in base-line study (no additives) at Locks Mill ....................... C-2 Color and suspended solids in base-line study (no additives ) at Locks Mill ................. 125 C-3 Effect of ferric chloride — 75 mg Fe3+ (without polymer) on BOD5 at Locks Mill ................... 126 C-k Effect of ferric chloride — 75 mg Fe3+ (without polymer) on color and suspended solids at Locks Mill ........ 127 C-5 Effect of ferric chloride (75 mg Fe3+/liter) + Nalcolyte 73C32 (l ing/liter) on BODs at Locks Mill ......... 128 Vlll ------- C-6 Effect of ferric chloride (75 mg Fe3+/liter) + Nalcolyte T3C32 (l mg/liter) on color and suspended solids at Locks Mill 129 C-7 Effect of ferric sulfate (100 mg Fe3+/liter) on BOD5 at Locks Mill 130 C-8 Effect of ferric sulfate (100 mg Fe3+/liter) on color and suspended solids at Locks Mill 131 C-9 Effect of alum (300 mg/liter) on BOD5 at Locks Mill 132 C-10 Effect of alum (300 mg/liter) on color and suspended solids at Locks Mill 133 C-ll Effect of alum (300 mg/liter) + Hereofloc 812.3 (0.75 mg/liter) on BOD5 at Locks Mill 13k C-12 Effect of alum (300 mg/liter) + Hereofloc 8l2.3 (0-75 mg/liter) on color and suspended solids at Locks Mill 135 C-13 Total and soluble BODs in second base line (gravity sedimentation) at Locks Mill 136 C-lh Color and suspended solids in second base line (gravity sedimentation) at Locks Mill 137 IX ------- TABLES Number Page 1 Description of Mills Surveyed 7 2 Analysis of Filtrates Prepared with and without Filter Aid. . 9 3 Comparison of Ultrafiltrates from a 100-A Membrane with Filtrates Prepared with Filter Aid over 0.45 and 0.10 Urn Filters 9 k Analytical Data for Treated and Untreated Samples During Evaluation of Method for "Soluble" BOD5. . 12 5 BOD5 in Clarifier Influent and Effluent Samples ("As Received" and Filtered Through a 0.^5 urn Filter) 1.^ 6 COD in Clarifier Influent and Effluent Samples ("As Received" and Filtered Through a 0.^5 ym Filter) ik 7 Suspended Solids in Clarifier Influent and Effluent Samples Removed by a 0.15 Vtm Filter 15 8 The Variability of pH, BODs and COD in Clarifier Influent and Effluent Samples from Different Sources 19 9 Screening Trials of Flocculants and Polymers by Analysis of Soluble Organic Carbon and Soluble BOD5 22 10 COD and BODs of Various Polymers used as Floceulation Aids. . 23 11 Comparison of Phenol Sulfuric Acid Test with BODs Test to Evaluate Flocculants in Test Tube Tests 2k 12 Comparison of Phenol Sulfuric Acid Test with BODs Test to Evaluate Flocculants in Jar Tests 25 13 A Comparison of Zeta Potential, Soluble COD and BODs, and Floe Form in Jar Tests of Flocculants 26 ik Jar Tests of Primary Flocculants with Mill A and Mill D Influents and Effluents 28 15 Jar Tests of Primary Flocculants with Mill F Total Clarifier Influent 29 16 Jar Tests of Polymers and Additives with Mill F Total Clarifier Influent 30 ------- IT Jar Tests of the Effect of Activated Carbon and Various Additives on Soluble COD and BOD5 in Mill F Clarifier 18 19 20 21 22 23 2k 25 26 27 28 29 30 31 32 33 34 Influent Jar Tests of the Effect of Activated Carbon and Additives on COD and BODs of Mill F Chemimechanical Pulping Effluent. . Jar Tests of the Effect of Flocculants on BODs, COD, Organic Carbon, and Color in Mosinee Clarifier Influent Jar Tests of FeCls, Alum, and Additives as Flocculants for Jar Tests of the Effects of Primary Flocculants and Polymers on pH and Clarity of Mosinee Clarifier Influents Jar Tests of the Effect of Flocculants on pH, BODs, COD, and Color of Effluents from Mosinee Clarifiers ...... 4 . Effect of pH on Jar Test Flocculation of Mosinee Clarifier Influent Volume of Sludge from Mosinee Clarifier Influents Treated with Various Flocculants and Polymer Removal of BOD5 from Effluents of Combined Locks Mill by Sources of Soluble and Total COD and BODs in Effluents of Mill F Chemical, Biological, and Physical Analyses of Various Process Streams within the Mosinee Mill (Mill K) Release of Solids Soluble COD and BODs ^y Beating Jar Tests of Various Flocculation Treatments of Filtered 90— Minute Beater Samples •, . Soluble COD and BODs in Mill F Effluents Treated with Flocculants and Reverse Osmosis or Ultrafiltration .... Laboratory Evaluation of Pilot— Scale Clarifier Analysis of Individual Samples to Establish Range Analysis of Special and Regular Samples of Clarifier Influent, Trial No .1 Analysis of Special and Regular Samples of Clarifier Influent, Trial No. 2 31 32 3S 36 37 38 3Q 41 4? 43 44 4s 46 47 s? S3 SS s6 XI ------- 35 Various Treatments of Mosinee Mill Effluent in Pilot-Scale and Mill Clarif iers .................... 60 36 Analysis of Special and Regular Samples of the Clarifier Influent of the Combined Locks Mill ............ 6k- 37 Clarifier Study — Combined Locks Paper Corporation Average Effluent Quality and Removal Values for IPC and Mill Clarifiers ........................ 66 38 BOD of Individual Effluent Streams and Total Mill Effluent of Mosinee Mill ...................... TO 39 Removal of Different Molecular Weight Fractions of Organic Residues by Flocculation ................. 80 ho Removal of Carbohydrates and Weak Acids from Mosinee and Combined Locks Waste Streams by Pilot-Scale Clarification. 85 Ul TOG Removal from Solutions of Model Compounds Treated in Jar Tests with Aluminum Sulfate ................ 87 U2 Solubility in Water of Various Model Compounds and Their Removal in Jar Tests with Fe3+ and Pulp Fibers U3 Chemical Costs for Clarification ............... 97 B-l Clarifier Studies with Hereof loc 812.3 (0.75 mg/l) at Mosinee Paper Corporation .................. 102 B-2 Clarifier Study with Alum (200 mg/l) plus Hereof loc 812.3 (0.75 mg/l) at Mosinee Paper Corporation ......... 103 B-3 Clarifier Study with Ferric Chloride (100 mg/l Fe3+ Plus Hereof loc 812.3 (0.75 mg/l) at Mosinee Paper Corporation . 10U E-k Clarifier Study with Ferric Sulfate and Hercofloc at Mosinee Paper Corporation ................. 105 B-5 Clarifier Study with Sulfuric Acid, Sulfuric Acid Plus Polymer, and Ferrous Sulfate Plus Polymer at Mosinee Paper Corporation ..................... 106 C-l Clarifier Study at Combined Locks — Base Line (Gravity Clarification after 0.010 Inch Sieve) Without Additives . . 117 C-2 Clarifier Study at Combined Locks — Ferric Chloride (75 mg/l Fe3+) Without Polymer .............. 118 C-3 Clarifier Study at Combined Locks — Ferric Chloride (75 mg/l Fe3+) Plus Nalcolyte 73C32 (0.75 mg/l) ...... 119 XII ------- C-U Clarifier Study at Combined Locks — Ferric Sulfate (100 mg/1 Fe3+) 120 C-5 Clarifier Study at Combined Locks —Alum (300 mg/l) 121 C-6 Clarifier Study at Combined Locks — with and Without Alum (300 mg/l) Plus Hercofloc 812.3 (0-75 mg/l) 122 C-7 Clarifier Study at Combined Locks — Second Base Line (Gravity Sedimentation After 0.010 Inch Sieve) Without Additives 123 Kill ------- ACKNOWLEDGEMENTS The original objectives of this project in terms of evaluating the degree of BOD removal attained in primary clarification of pulp and paper effluent waters were conceived by Dr. Everett Cann of the Wisconsin State Department of Natural Resources. The project was organized and conducted by the staff of The Institute of Paper Chemistry. George Dubey, Project Supervisor, had responsibility for detailing and conducting the laboratory and field studies and for reporting the results of this research in coopera- tion with Averill Wiley, Project Director, and Hardev Dugal, Director of the Institute's Division of Industrial and Environmental Systems. Gerald Hovind, Arthur Webb, and Gay Fraundorf were principally involved in the extensive analytical program. The field studies were greatly facili- tated by help from John Baumann who installed and operated The Institute of Paper Chemistry pilot equipment, by James Albrecht and Al Kleist of the Mosinee Paper Corporation, and by Lawrence Weyenberg and James Hoovers of the Combined Locks plant at National Cash Register. Ralph H. Scott, Chief of Pulp, Paper and Wood Staff, Food and Wood Products Branch, lERL-Ci, EPA, as Project Officer, has had responsibilities for the immediate supervision of this project in accordance with the objectives of the program. xiv ------- SECTION I INTRODUCTION The overall degree of biochemical oxygen demand (BOD) removal actually observed in most conventional out-plant clarifiers treating total mill ef- fluents has been relatively small (5-15 percent) and has seldom extended to levels that would significantly improve effluent quality. A search of the literature disclosed few published investigations of the mechanisms for the BOD removal in clarifiers. Only two (l.,2_) gave data indicating possible sources of BOD that is removed during clarification. The State of the Art Review of Pulp and Paper Waste Treatment by Gehm (l_) attributes the BOD to "the high content of suspended organics and small quantities of dissolved organic matter" in certain types of waste and to the "fact that the oxygen uptake rate of fiber is slower than that of dissolved materials, since it must first be liquefied before oxidation can take place." Das and Lomas (2_) point to glucuronoxylans and araboglucuroxylans as water soluble hemicellu- loses (from the beating of pulp) that reduce the BOD when removed with the floe. One of the first goals of Phase I of this research program was, there- fore, to evaluate by a controlled analytical study the removal performance for total and "soluble" BODs in different clarifiers at pulp and paper mills. Surveys of mill waste flows and clarifier effluents were conducted at 12 mills owned by 11 companies. The analytical data developed during this sur- vey established the basic program of the investigation and the definitions of the terms "soluble" and "total 5-day biochemical oxygen demand (BODs)" to be used throughout the balance of this study. Three different classes of BOD-forming components in the various waste flows were identified: 1. Substantial quantities of the BOD being removed in clarifiers are obviously suspended fibers and related solids in various stages of physical, chemical, and biological degradation. These suspended particles may originally have low levels of solubility and low BOD but can undergo degradation to produce soluble substances responsible for BOD. This fraction com- prises the principal source of BOD readily removed in conven- tional clarifiers. 2. Advanced degradation of cellulose and hemicellulose particles and the presence of papermaking additives such as starch may provide a substantial amount of soluble, high molecular weight components with biochemical demand for oxygen. Most of these kinds of substances can be surface-adsorbed and removed by advanced clarification procedures. The further development and field testing of routes to removal of this fraction com- prised a principal area for development under Phase II of this project. ------- 3. A third fraction of the BOD contained in pulp and paper waste waters is comprised of low molecular weight degradation prod- ucts of carbohydrates, such as acetic acid, aldehydes and alcohols, which are not readily amenable to physical adsorp- tion and flocculation. Removal of these components has not been reported for primary treatment if we except those cases of prolonged holding time or other conditions where advanced stages of microbiological action are evident in the clarifier. The fundamental physicochemical mechanisms of flocculation of pulp mill white water fines studied by Williams (3.) and the work by Back (h) on the nature of wood components dissolved in the pressure refining of mechanical pulps provide information on a possible approach to the development of an understanding of the mechanisms involved in the removal of BOD components from waste streams. Their ideas were used to develop the jar testing proce- dures used in our laboratory for the optimization of the removal of BODs constituents from various total mill effluents. Phase I studies were funded jointly by a group of U.S. pulp and paper companies (Appendix A), the U.S. Environmental Protection Agency (EPA) (Grant Bio. 803 119) s and the Wisconsin Department of Natural Resources (Ref. No. 8100). For a supplementary and confirming Phase II program, small-scale pilot equipment was used for on-site comparative studies, paralleling commercial clarifiers operating at two mills representative of a broad segment of the pulp and paper industry. Additional funds were supplied by the U.S. Environ- mental Protection Agency under an extension of the above grant and by The Institute of Paper Chemistry. This confirming program of study was designed to evaluate the operating parameters developed in the earlier Phase I labo- ratory research under mill conditions. The first trial was conducted at an integrated kraft mill and the second at an integrated mill producing pulp by a recently developed chemimechanical (CM) pulping process. Laboratory jar tests were also used to evaluate the flocculation and clarification characteristics of individual sewer discharges at these two mills. We studied the effects of processing each of these streams separate- ly and we could also check the effect each had on the overall BODs efficien- cy of the clarifier when processed as a part of the total mill effluent. Additionally, the jar tests were used to test the removal of soluble model compounds (acetic acid, aromatic dicarboxylic acids, etc.) from solution with the various flocculating agents in order to help elucidate the mecha- nism of BODs reduction. An economic evaluation of the data of two field trials and the results of the jar tests conducted on individual streams show the savings which might be realized if the relatively small volumes of stronger pulping efflu- ents, containing lignin compounds with dispersant properties, were not treated in the clarifier. ------- SECTION II SUMMARY AND CONCLUSIONS 1. A large portion of the BOD entering the primary clarifiers process- ing pulp and paper manufacturing waste waters are solubles that can pass through a 0.^5 ym filter. 2. These components are generated from wood in pulp and papermaking steps, including wood handling. They may also come from papermaking addi- tives such as starch or sizes. These materials can be degraded within the mill or during out-plant treatment. 3. Laboratory jar tests with various chemical flocculating agents and operating parameters showed that significant quantities, ranging to 20 per- cent or more, of the soluble BODs and color could be removed from wastes from mills surveyed in Phase I studies. These findings were confirmed by pilot scale field trials at Mosinee Paper Company at Mosinee, Wisconsin, and at the Combined Locks mill of the NCR Corporation in Combined Locks, Wiscon- sin. k. Jar tests with the individual waste streams making up the total mill discharge have shown that certain streams contribute both high BODs levels and chemicals that are detrimental to the efficient operation of the clarifiers. Elimination of these streams from the out-plant clarifiers could provide increased quality to the mill discharge. It would also de- crease the load on the secondary treatment plant. The streams that were re- moved from the influent to the clarifier could be treated separately in small clarifiers or by reverse osmosis, ultrafiltration, or other small unit pro- cesses . 5. The mechanism for the removal of soluble BODs is very complex due to the heterogeneity of the wastes treated in the clarifiers. There was no ap- parent relationship between the initial soluble BODs concentration and the ultimate removal of the soluble BODs components by the clarification process. 6. Gel chromatography analysis of effluents before and after coagula- tion with iron and pulp fines showed a significant number of low molecular weight compounds were removed. Model compound studies showed that many or- ganic acids including some quite soluble could be partially coagulated in an iron-pulp fines system as well as substances which do not form salts with iron. 7- Evaluation of the benefits and costs of achieving improved perfor- mance of clarifiers requires detailed engineering studies to fit individual mill situations. Such studies were beyond the scope and objectives of this study. However, it was apparent that total chemical costs, including the chemical additives which may be used in existing practices at these mills, ------- might range from 3<£ to 10$ per 1000 gallons treated. Efficient use of these chemicals would be balanced by benefits and cost reductions in the overall systems of waste treatment. 8. Good clarifier performance and low chemical costs could only result from adequate control of mill overflows and spills so that strong colloids or dispersants do not enter the clarifier. ------- SECTION III RECOMMENDATIONS These studies have shown that vith the proper selection of flocculating agents and operating conditions, soluble BODs concentrations in waste streams can be markedly reduced (10-50 percent) in the primary clarifier. Collapse of the double layer surrounding the charged molecule, bridging, and decreased solubility after treatment with the flocculating agent are mechanisms which could be expected to contribute to the removal of both total and soluble BODs components, as well as color, from the effluents. 1. We need to know more about the constituents and characteristics of different kinds of sludges — those from high and low levels of soluble BODs removal — and their effect on the efficiency of clarification in mills pro- cessing under different conditions. This could add considerably to our knowledge of the mechanisms involved. The influence of individual mill waste streams in clarifier sludge performance also needs further study. 2. Many of the individual sewer discharges were found to have adverse effects on the efficiency of the clarifiers and also to have a toxic effect on the 5-day BOD test. Removal of these streams from the clarifier inflow could markedly increase the efficiency of both the primary and secondary (biological-type) treatment systems. Diversion of these flows, which are usually of small volume, to other treatment systems is recommended. 3. Selection of the streams to be "eliminated" could be undertaken under laboratory conditions with a combination of jar and BODs tests. Pro- cesses for treating the eliminated streams by separate in-plant clarifica- tion, reverse osmosis, ultrafiltration, or the training of plant personnel to reduce the discharge of the detrimental stream would follow the identifi- cation of these major sources of difficulty. ------- SECTION IV MILL SURVEYS EQUIPMENT AND PROCEDURES Weekly samples from mills using clarifiers consisted of two gallons of the influent to the clarifier and a single gallon of the clarifier effluent. Where possible, both samples were collected as 2^-hour composites. Mills not using clarifiers sent in two gallons of the effluent they were discharging into ponds or aeration lagoons. Table 1 describes the types of mills and their facilities for pulping, bleaching, and effluent treatment; this table is based on data supplied by the mill staff. As soon as the samples were received in the Institute laboratories at Appleton, Wisconsin, the influent samples were composited, well mixed and immediately transferred to one-gallon plastic bottles. One bottle was filled to the brim, set aside to settle for one hour, and the top 6 inches of super- natant was withdrawn using a light vacuum and a short U-bend at the end of a glass tube. This permitted removal of the top layer of "clear" fluid without drawing the "settled" material from the bottom layer. A second bottle of the influent and also a bottle of clarifier effluent were separately mixed and analyzed "as received." Samples of the "as received" and "settled" material were pressure fil- tered through various porosity filters. These filters were 1.2-ym and O.U5- um pore size Metrical membranes from Gelman Instruments of Ann Arbor, Michi- gan and a 0.10-um plastic filter from Nuclepore of Pleasant, California. Suspended solids, BODsj and chemical oxygen demand (COD) were determined on the initial samples and on all fractions. Samples for suspended solids were processed on 1.2- and O.U5-um filters on a Gelman pressure funnel at 6.9 kg/cm2 (lOO psig) with NZ and dried at 103-105°C (2l8-221°F) for one hour. For biochemical oxygen demand we used the standard APHA Method (%), keeping the samples at 20°C (69°F) in a water bath and determining the dissolved oxygen with a Weston and Stack probe. The COD was determined by the dichromate oxidation method of APHA (6_) . Color was determined at pH 7-6 using the National Council for Air and Stream Improvement method (20_). Some waste samples filtered slowly through 0.^5- and 0.10-ym filters. To avoid filter clogging, a pad of filter aid (Celite, AR, Johns-Manville) was used on 1.2-, 0.^5- and 0.10-um filters for these wastes. The filter-aid pad was formed from a 1-percent slurry of Celite AR poured onto the filters and then dewatered. Analysis of these samples with and without the use of the filter aid (Table 2) indicated that the filter aid did not adversely af- fect the analytical accuracy. Therefore, in later work, the filter aid was ------- 0 f£J ^H p g w w . "I M «y o a o EH PH H K O M Q • rH W <£ EH !H (D 01 CH "H •H -P a -d rH "d o o) CU \|cd H bO 0 0 CO d) > 0) CO -p C 0) H ^-t q_i M hO S 0 S c O D a 2 CU O H (U M co M a? o d EH O 0) O ft o o H H S (U h H £H OJ 0> O CU >vQ .rH ft rH -H H <H ^ "w H -H O bO -rl •H~bO d '3 B id fn B rH O t— H •d o -H • cu o D K OJ -P rH C C CU O <8 O O O o a o a a a -rt 8 ^30 H OOOcdrHC CUCU H cdh -Pti'HCd S -P+3 •H E-pcdfiH O HrH CUSsCUcd OCUO^CU fn M) LT\tH p < ocdo r-DHcd-H cuSocdoHcu (L, a opnffl HopqfL,s ^oasaoa 1 0) O >j W) ft -P cd >, -H ^2 -P fl CO <; <; co cu c! cu a a s -p cu d CU P -H "d 'H H -rl fn rl MOO & O C H H bB ^ •H Cd ^5 -rf d) WOO W ft d 3 -d ft^D^t - CO O oj co _ LA _s ro ^t rH O O O O O O M fn d) (1) (3 P -P •H CO Cd d • — • fi d) > CU • 0 CU r- SO cd (3 -H cu -p +J a -H -p -p cd d) ,CJ H CO CU c! d) cd >d > -p d) •H ft 0 cd bO fi-dft^EO cup •H d) O -H -P rH cu^j^t ftO-P cd co r^j u o3 fn cd ^ O 08 p \| JlJ >»_ +3 SH -f3 £H 'H d Qj W CO i-f~\ QJ JH " cd X ^i HJ O 0 H o O • • cd H co ft CU O O O d) fi -H Q pq a a PQ o s o •H -p -p cd cd ^ pq ON o rH H rH •H S q S q •H d) T3 CU R o o • CO -p o c -p 0 o o ------- 1 n CRIPTIOH OF MILLS SURVE t/J o i j-* •H +> C o o H 3 i D 1% rH bO o s Effluent sewers to C r^ S £p p . \ o (U w Q) CO t^ XA ^t o IAVO on c\i o o o OrHO OO OO O g -p •H-p rl-PCUH-PO^> ri C to B 01 W to Pi «fc i-HI^^^dbDobDxlI Hi -H 3 O O -rl -H -p 1) c o o on (M 01 -p C w Pi s y; . ^ (U •d m O 01 •H -P a s •rH Oj I — 1 01 01 ^i 0) 01 3 -H rH d o H .r( 0) C a cd •a » 3 o « 'd 0 C s and detergent added. silicates added. ngth resin, defoamer, t! g agent, dryer release, chlorine added. IL, -d P oj 8 m p O £3 11 @ S3l 53 ------- used as a standard practice with the precaution of first checking the effect on one or two samples of each waste stream. Eventually the exploratory studies with various grades of filters resulted in adoption of the 0.45-um porosity as a standardized procedure. TABLE 2. ANALYSIS OF FILTRATES PREPARED WITH AND WITHOUT FILTER AIDa BQD5> msA COD, mg/1 1.2 um Q.45 umc 1.2 urn 0.45 ym Sample With Without With Without With Without With Without F 190 74 H 32 48 B 770 1246 c 166 184 178 82 34 42 778 1230 192 181 182 70 31 44 672 1218 188 168 176 80 32 46 675 1184 184 174 464 212 133 105 3741 5363 352 316 436 216 123 126 3650 5320 362 316 4 09 199 108 118 3636 5190 329 275 369 191 114 106 3615 4890 329 292 ^Celite (AR) BOD5 = 0.04 mg/g. 1.2-ym BOD 5 =1.92 mg/sheet. C0.45-um BOD5 =0.62 mg/sheet. A few samples were also processed through a Dia-Flo (Amicon Scientific Systems) stirred ultrafiltration (UF) unit with a nominal 100-A (0.01 Urn) pore filter. The results were compared with those from the 0.45- and 0.10- pm filters (Table 3). Since no differences were apparent in either BODs or COD, the more time consuming UF procedure was not routinely used. TABLE 3. COMPARISON OF ULTRAFILTRATES FROM A 100-A MEMBRANE WITH FILTRATES PREPARED WITH FILTER AID OVER 0.45- AND 0.10-um FILTERS Sample F H JJ SUFa 259 272 30 59 40 35 BOD, mg/1 S045F 310 256 43 48 38 40 S010F 266 267 32 50 35 40 aSUF — Dia-Flo stirred ultrafiltration cell with 100-A membrane. ------- DATA AND DISCUSSION In order to establish the various total and soluble components of BODs > the incoming samples were analyzed "as received," "settled," and after pas- sage through either 1.2-, OA5-, or 0.10-um filters (Figure l). Throughout this section of the report, the designations are as follows: AR As received AR045F As received, filtered through O.U5-ym filter with a Celite overlay S Settled S12 Settled and passed through a 1.2-pm filter S12F Same as S12, except with Celite overlay Settled and passed through a 0.^5-ym filter Same as 30^5, except with Celite overlay S010F Settled and passed through a 0.01-um filter with a Celite overlay SUF Settled sample filtered through a 100-A UF unit Data in Table k are averages of the six samples from each mill, along with high and low values (range) for each set. The values in parentheses are the percentages of the constituent in the sample after treatment. BODs de- creases as filter pore size decreases (Figure 2). The data in Table k and Figure 2 can be summarized as follows : 1. Comparing "as received" and "settled" values, we find a marked reduction in BODs (to 62 percent) after simple settling (JJ). 2. The 0.45-Um filter was, for all practical purposes, the best filter for the removal of "insoluble" BODs. This filter was used for further studies of "soluble" versus "total" BODs- 3. Seven percent additional BODs could be removed from the JJ mill samples by passing them through the O.U5-um filter. k. The COD values paralleled, but did not directly correlate with, the BODg values. In order to evaluate the operation of the mill clarifiers we analyzed clarifier influent and effluent samples in the same manner. Data for these samples "as received," and filtered through 0.^5-Um filters, are given in Table 5 (BOD), Table 6 (COD) and Table 7 (suspended solids). In order to simplify the presentation, the "as received" values are plotted in the same sequence in Figures 3, ^, and 5. Marked reduction in total BOD5 was achieved in clarifier JJ. All clarifiers except those of Mills I and J achieved 50-90 percent removal of suspended solids. Unfortunately, the values for constituents and removals are quite errat- ic, probably due to a time lag for arrival of these constituents at the sam- pling points. Although most samples were 2^-hour composites, compensation for the effect of the clarifier "holding time," in some cases on the order of 18 hours and in most cases at least k hours, was not built into the sampling system. While this would have little effect on the monthly mill averages, it had a marked effect on the samples sent to our laboratories (Table 8) . We found major differences in pH, BODs, and COD for the two streams in spite of 10 ------- •d H-> H 'd 4-> • 0) 4) K M to •H P a o as -p C M O -d 3 rH 4) O O H rH CO •rH CO S •d 5 i CQ H I B 4) LA O CO JH • CO > 4J H 0) H rH H £> Cd •rl O O OS C fe O i-3 <. J- fJJ ^^» ^3 +> fe pq •P -H B CM ^^ •H H 3. H > 4) CO to O CM -H H • H tfl Q> W) rH CL> i>> +30 43 rH •H O O iJ fi c^ o ^ •* o o o •d 5 fa m Q O I LA O 'd C trj CM rH rCj +J •H > r?H •H o to •d cu T* 4) to CO 1 "^ ^^ to •H CO H jj O O o •d c in O o pq 1 "pq — •* to •H to rH 05 .§ 01 in O W 0) rH jj ,— \| O 01 o 3 CO •rH 01 >> H 05 11 ------- -* vo O ON I---* H O t/J C— CO .* UN O 1 M H OJ -* cy cy OJ UN 5 o w OO CM h PON o vo "V CO VO OJ O ON HCO "LA m fe £3 c5 o o W OJ VO H ON ON &* VO H CM -^7 H CO 0 CO HO OJ O -~» OJ VO H OJ O H H vo CO g ooS ^-» OJ Pn voco o LA on s £3 ^- o Ct, C?\ ON tA IA rO o on oj OOCO frj ^"7 H H H w H on on oj t— tA UN LA ion oj •— - OJ fc _* OJ LA VO -* O LA LA « H H •4 on OJ VO LA CO C7\ UN CO °* VO UN vo vo H \|d *• ,— 1 m I H O CVI rH O VO cy - i H -* cy LA t-co vO OJ &s oo IA VO ON UN vo H OJ VO UN a LA o CM UN 1 H CM H UN LA ol H£ CO H OJ CO HOD H vo H C- 1 ONVO 0 t- H vo •— cy ^.-^ on H • — •CO- CM o OJ UN VO H VO ON OO rH OO OJ OJ UN C— CM CO H '— • CM HVO ^MCO oj on UN OJ CO vo CM OJ CO CM VO O on oj O 1 Jt OJ H t- Jvo" H C- OO O - CO UN ON H t— ^. LA CT\ ^ 1 HCO on ON H UN OJ ON t- 2^ UN OO h oo IA OJ vO OJ H tA OO 00 OO OJ O 1 VO UN H UN CO o UN on •35 -* OJ tA VO 1 -* on CO UN ON-* on CM ON OJ vo 1 H t— H «N ^-- u-\ H VO ^t UN -— OJ 1-1 ^ •""•• OO ^* VO J- t OJ I?- CM ON $& <%£ 0 8! CO CO CM H 0 ON (— "0^100 vo VO CO UN ON UN ^r CM s~* H vO vo OOOO 1A OJ CO 0\ vO vo CO O on t— LA OJ f. t-- CO CO "o vo on co vo OJ t— I— CM CO CO UN H ONVO ONVO o vo H -V CVI CO t- OO •§ H vo 1 O C>J J- CO OJ rH OJ OJ ££ ^. 00 H 'o T1 O UN H t— ff^ ON CM -=r on eg i o on H t— ON O m on • — • H o on cy!^ CO OJ vo o H H VO H 1 CM-* Po on CM CM VO on oj ^^ OJ OJ ro Oj 5? -* OJ o J ^* ON t— vo a CO 1 vo CO OJ ON -* H UN t CO -* -* vo on H CO OJ vo on o UNVO co on vo LA H 1 OO O OOVO VO CVI UN ON •-- OJ CO 1 CO H OO ON H - - — • o _* UN UN OO -— OJ -* CO HCO H LA --. LA t— ON -^ CM VO ON - — LA O ON ON H ^- OO 0 1 H - ON H H 0\ on 1 CM LA -• o ON OJ O CM -* CM IA cy VO P- _* UN O CM -* CM ON OJ VO t- OJ -* OOCO -* H — - LA II u\ on. CO 1 CM H OJ UN CO VO H OJ IA UN OJ LA on co OOCO CO o\| VO OJ on o H VO OJ -— . 00 •- t VO cO on t— CO CM — 0 vO -3- -d- LA oj LA on OJ OJ CO UN UN -—• 1 O LA t— t— O UN OJ t— O LA LA OO CO on H H £3 ON CM UNVO vO 1 onvo HCO SvS ,-t CO t^ t- LA VO OJ 1 H"ON — - m CO W O OJ OJ H o 1 o on on LA H £-i CO t— ^ o ON on,- - H CM ON on i on co LA vo VD CO O t— VO 1 CO ^O VO CO UN on To r-j H H co vo — OJ on H CO U\ co vo - — • on -* 04 H H ON"H oj VD OO OJ CO 1 rH VO CM CO 8 OO 1 H -* ON O H-* oo cy on on H C— on CM .^ VO VO 3-J, 8 x m H - s CO OJ H ^— - 1 t> VO 0 P .* on j~- -d <0 41 H •H 4-< o 'k— ' -d vo 0) ^ 3 OJ ON V \ CO h C CO T St C -* •• « h- \| PL r5 12 ------- 500r 400- c z 300 u o > x o o UJ X o o GO 200 100 As Received PAPER MILL Settled 1.20 0.45 Figure 2. The separation of insoluble from soluble BOD5 by filters of various pore sizes. -xC -A D H 0.10 FILTER PORE DIAMETER, 13 ------- TABLE 5. BOD 5 IN CLARIFIER INFLUENT AND EFFLUENT SAMPLES ("AS RECEIVED" AND FILTERED THROUGH A 0.15 ym FILTER) Mill A B C D E EE F G H I J JJ As received (total) Filtered (soluble) Influent , Effluent , Removal , mg/1 mg/1 % 195 311 No clarifier 261 192 156 90 No clarifier No clarifier 378 322 787 615 76 51 622 557 230 152 122 37 31 27 12 15 18 33 10 33 70 Influent mg/1 303 115 75 179 536 39 371 131 38 , Effluent , Removal , mg/1 % 30l* -0.3 138 5 59 22 200 -12 132 15 33 15 2l7 3l 121* 8 32 16 TABLE 6. COD IN CLARIFIER INFLUENT AND EFFLUENT SAMPLES ("AS RECEIVED" AND FILTERED THROUGH 0.15 JM FILTER) Mill A B C D E EE F G H I J JJ As received (total) Filtered (soluble) Influent, Effluent, Removal, mg/1 mg/1 % 1863 1328 No clarifier 712 129 583 386 No clarifier No clarifier 1695 612 2918 1606 272 188 2132 2013 715 133 691 126 29 10 31 62 15 31 6 10 82 Influent mg/1 1207 283 216 108 1201 108 1111 351 88 , Effluent, Removal, mg/1 % 1236 -2 286 -1 25U -3 399 2 1119 1 98 9 IIOU 1* 28U 19 8U 5 14 ------- the fact that both were collected within the same time period. This lack of continuity in the two samples could explain not only the lack of removal ob- served at times but could also be the reason that our studies showed the higher or lower removal values than did general mill experience. TABLE 7. SUSPENDED SOLIDS IN CLARIFIER INFLUENT AND EFFLUENT SAMPLES REMOVED BY A O.U5 um FILTER Mill A B C D E EE F G H I J JJ Influent , Effluent , mg/1 mg/1 UUl No clarifier 1126 235 No clarifier No clarifier 1187 Ui65 2UO 853 690 581 63 2U8 77 213 103h 99 587 3^0 50 Removal , 86 78 67 82 75 59 31 ^7 91 15 ------- 800 76o 720 6I»o 600 56o 520 Wo 1.1,0 YJ\ Influent (ng/1) II Effluent tmg/1) RES J Removal 320 28o 21,0 200 160 120 80 I S -90 -80 -70 -60 -50 -1)0 -30 -20 ,-10 Mill Figure 3- Total biochemical oxygen demand in clarifier influent and effluent samples. 16 ------- 3000 2850 2700 2550 21400 2250 2100 1950 1800 1650 1500 -350 1200 1050 900 750 600 1450 j)00 150 Influent (mg/1) j Effluent (mg/1) \| % Removal i 1 i 90 I 80 70 60 30 Jio G Mill JJ Figure k. Total chemical oxygen demand in clarifier influent and effluent samples. 17 ------- ItOOO 3800 _ 3600 _ 3"lOO - 3200 _ 3000 2800 - 2600 aii oo 2200 - 2000 1800 1600 11(00 1200 1000 800 600 1<00 200 R73 Influent (ma/1) V/A \~\ Effluent (mg/1) IK] Removal, % Itl85 I I I 90 80 . 70 60 50 ItO 30 20 10 G Mill Figure 5- Suspended solids in clarifier influent and effluent collected on a O.h^-ian filter. 18 ------- TABLE 8. THE VARIABILITY OF pH, BOD5 AND COD IN CLARIFIER INFLUENT AND EFFLUENT SAMPLES FROM DIFFERENT SOURCES8" pH 4.81 8.28 10.42 11.53 10.18 7.35 7.11 Influent BOD 5 27^ 884 482 634 836 476 168 COD 2268 3899 1702 21498 3388 1413 411 pH 6.88 10.68 11.05 9.88 10.80 9.80 10.78 Effluent BOD 5 268 555 489 777° 1104C 822C 258C COD 552 1548 812 2202 2478 2293 869 These values are from all of the samples (54 sets) where pH differences and/or negative "removal" vas noted. Influent and effluent samples were taken during the same time period. 12 Prolonged holding time probably explained the apparent negative removal. 19 ------- SECTION V LABORATORY STUDY OF PROCESSES FOR HIGHER LEVELS OF SOLUBLE BOD5 REMOVAL We approached the problems of improving pulp and paper mill liquid waste treatment by two differing routes. A first route studied possibilities for improving the methods for conventional, out-plant primary clarification treatment of the total mill discharge. The second, a more innovative route, was directed to the principle of separation of flows, particularly spills of strong process water, and was also directed to individual, high-efficiency treatments, preferably within the plant, for those flows detrimental to clarifier performance. Various treatment methods, such as sedimentation, flotation, filtration, membrane separation (reverse osmosis, ultrafiltration), carbon adsorption, and ion exchange could be used. The results of some of these methods are described in this section. New and relatively more expensive methods of processing would be feasi- ble for only parts of the total flow in order to prevent buildup of solubles in streams, especially those being recycled. These fractions could be treat- ed in-plant in small, highly efficient systems or, if the streams are not to be recycled, they could be treated before they are mixed with other process waters entering the total sewer system. However, since many of the mills involved in this study were already using clarifiers or had plans to use them, the laboratory study was care- fully planned to develop a better understanding of the conventional, out- plant clarification process. SCREENING TESTS Following the survey of mill discharges and clarifier efficiencies (Section IV), a study to optimize the removal of soluble BODs from both total mill waste waters and individual streams was begun in the laboratory. We preferred to evaluate clarification efficiency by some means other than the BODs test because the 5-10 days time to run each test, depending upon the day of the week that the samples were prepared, made systematic laboratory studies very difficult. Also, we wanted a procedure based on small sample volume. TEST TUBE TRIALS We had developed in our laboratory a simple test tube flocculation test using a 15-ml sample in an 18 x 150 mm test tube mounted on a 12-inch diam- eter, vertically oriented, platform rotating at 3 rpm (unpublished). The 20 ------- slow transfer of fluid from one end of the tube to the other produced a floe form and supernatant clarity that correlated well with jar tests with the same sample/flocculant combination. This simple, small-volume system was used in the early part of the work for screening the incoming samples. JAR TESTS At the same time, jar tests were carried out on the same waste material using a Phipps and Bird laboratory flocculator. Various flocculating agents, alone or in combination, were added under carefully standardized conditions of chemical addition, rapid mixing, flocculation mixing, settling and sampl- ing. Briefly, these can be described as: 1. The treatment of 1.0 or 1.5 liters of waste in 3-liter jars; 2. The addition of the primary flocculant, followed by one minute of mixing at 100 rpm; 3. If a second agent (e.g., polymer, silica, activated carbon, etc.) was added, it was followed by a second rapid mix of one-half minute at 100 rpm; k. The paddle speed was reduced to 20 rpm for 20 minutes to provide time for flocculation. The flocculation step was done after the final addition of chemical, no matter how many additives were involved; 5- At the end of the flocculation period, the jar was removed from the unit and the sample allowed to settle for one hour; 6. Samples were withdrawn into 250-ml bottles under slight vacuum, with the tip of the sample tube one-half inch under the surface of the liquid in the jar; T. An untreated jar of the waste was carried through the entire test from rapid mix to sampling. All "removal" data are based on the comparison of the analyses of the control sample with the treated samples. All samples were stored at k°C (hQ°"F) prior to analysis. ORGANIC CARBON ANALYSIS Use of a Beckman Process Carbonaceous Analyzer permitted relatively rapid (15 samples per day), small-volume analysis of solutions for total organic carbon (TOC). The test tube trials, after the floe was removed with the 0.^5-ym filter, provided sufficient sample for the determination of sol- uble organic carbon. The use of larger test tubes (25 x 250 mm) and addi- tional sample volume on the same flocculator table, or jar tests directly, allowed determination of both total and soluble BODs and TOC, and soluble organic carbon (OC) on the same sample. We found little correlation between the reduction of soluble organic carbon and the reduction of soluble BODg for a series of flocculation studies (Table 9)• We also found no relationship of COD to BODs during the mill sur- vey; these results are probably due to the high ratio of nonbiodegradable to biodegradable carbon in these waste streams. Total and soluble organic car- bon and BODs components removed during flocculation would be both degradable 21 ------- and nondegradable substances, with the percentage removal of degradable be- ing overshadowed by nondegradables. While the polymers did have both bio- chemical and chemical oxygen demands (Table 10), the amounts added to the solutions were so small (l to 10 mg/liter) that they did not contribute sig- nificantly to COD or BODs. The addition of the polymer at the 10 mg/liter level would add less than 2 percent COD and 1 percent BODs to most samples if all of the polymer remained in solution and none was Jfemoved by floccula- tion and/or filtration. TABLE 9. SCREENING TRIALS OF FLOCCULANTS AND POLYMERS BY ANALYSIS OF SOLUBLE ORGANIC CARBON AND SOLUBLE BOD5 Reduction , % Flocculant Type Alum Alum FeCl3 Lime mg/1 200 200 250 250 250 100b 150 200 200 200 200 250 25 25 25 Polymer (0.5 mg/1) None Nalco 607 Nalco 627 Nalco 634 Nalco 73C32 None None None Nalco 607 Nalco 609 Nalco 627 None None Nalco 634 Nalco 73C32 Soluble OC 18.2 34.3 26.5 15-7 6.0 20.7 17.4 25-4 21.3 23.1 -7.4 20.5 1.8 -6.2 2.3 Soluble BOD5 -9-7 -20.9 8.1 5.4 0 20.3 -20.5 21.2 16.8 20.7 15-7 19.7 -1.9 8.4 0 values on "as received" samples: Soluble OC = 125-425 mg/1. Soluble BOD5 = 100-200 mg/1. bAs Fe3+. 22 ------- TABLE 10. COD AND BOD5 OF VARIOUS POLYMERS USED AS FLOCCULATION AIDS& Manufacturer Nalco 607 609 63h T3C32 Dow PAA Tydex 12 Hercules Hereof loc 812.3 Soluble COD, mg/1 231 7^0 333 102 15kh 1665 — Soluble BOD 5, mg/1 0 lit 0 0 2lh 0 66 Polymer concentration 1 g/1. PHENOL SULFURIC ACID TEST The phenol-sulfuric acid test (£) is a rapid method for the estimation of carbohydrates, including virtually all classes of sugars, sugar deriva- tives, oligo- and polysaccharides, ketoses, aldoses and aldehydes. This method has been used for several years in these laboratories for monitoring gel chromatographic columns and for the study of aldehydes in solution, both carbohydrates and noncarbohydrates (8). Although the method appeared to be useful for measuring the removal of soluble BOD5 during trials in the test tubes (Table 11), it did not give re- producible results in the jar tests; therefore, it was not useful as a re- placement for the BOD5 test (Table 12). ZETA POTENTIAL "If the zeta potential of the colloid is lowered below a critical value, the colloids tend to coalesce, resulting in coagulation" (£). Many of the soluble" BOD5 components in the waste stream are colloids, medium and low molecular weight organics, and in some cases simpler carbonaceous compounds, which should be amenable to colloidal coagulation and coprecipitation (10.). Zeta potential changes and final values were compared with the removal of COD and BOD5 components by flocculation (Table 13). Wo correlation could be drawn between the zeta potential and the reduction of either COD or BOD5 for the waste stream after flocculation and filtration. There was, however, the expected correlation of near zero zeta potentials with good floe formation and supernatant clarity. 23 ------- TABLE 11. COMPARISON OF PHENOL SULFURIC ACID TEST WITH BOD5 TEST TO EVALUATE FLOCCULANTS IN TEST TUBE TESTS Flocculant Type FeCl3 Lime mg/1 50 100 150 200 250 300 300 300 300 300 300 300 hOO Polymer (1 mg/l) None None None None None None Nalco 609 Nalco 627 Nalco 63 ** Nalco 73C32 Dow PAA Dow Tydex 12 None Q Reduction, Soluble glucose equivalent 28.9 U5.6 50.0 16.7 17-7 6.3 25.6 10.5 10.5 20.0 39-1 32.6 28.1 % Soluble BOD 5 22.8 32.1 36.2 33-9 5.0 18.1 16.1* 12.9 25-9 12.9 23.2 31.7 7.1 values "as received": Soluble glucose equivalent = 2l6 mg/1. Soluble BOD5 = 200 mg/1. bAs Fe3+. As the study progressed we found little relation of the visual clarity of the supernatant or the rate of floe settling with the reduction of BODs. A finely divided floe seemed to result in the greatest reduction in soluble BOD5 after filtration with the 0.15-um filter, whether or not the floe had settled. JAR TESTS FOR OPTIMIZING BOD 5 REMOVAL BY CLARIFICATION General Following the lack of success in developing a rapid screening test to correlate flocculation with the removal of soluble BODs. a number of jar tests were made with several mill effluents. We used the standard BODs anal- ysis to test these flocculants. While this resulted in a rather delayed evaluation, it could be used for: (l) a study of many flocculating agents and variables or (2) careful consideration and selection of both flocculants and process variables, with BODs determinations on either (a) all samples or (b) selected samples from each series. With frequent reevaluation of the procedure as BODs results became available, the (2b) scheme was selected for the majority of the jar tests. 24 ------- TABLE 12. COMPARISON OF PHENOL SULFURIC ACID TEST WITH BOD5 TEST TO EVALUATE FLOCCULANTS IN JAR TESTS Reduction , % Flocculant Manufacturer mg/1 Nalco 607 609 627 631 73C32 Dow PAA Tydex 12 values "as Soluble Soluble Soluble 6.0 6.0 6.0 10.0 6.0 6.0 10.0 6.0 10.0 6.0 10.0 20.0 6.0 6.0 6.0 10.0 received": PH 7-5 3.0 7.5 7-5 3.0 7-5 7-5 7-5 7-5 7-5 7-5 5-0 3.0 5-0 7.5 7-5 glucose equivalent COD = 550 mg/1. BOD 5 = 220 mg/1. Soluble glucose equivalent -1.5 12.0 11.5 16.5 U.5 7-3 15-3 7-3 11.5 15-3 15.3 10.8 3.2 -1.2 7.3 Ik. 2 = 250 mg/1. Soluble BODS 2.6 11.3 6.8 9.8 Q.k 6.8 10.0 -2.1 23.8 2.1 12.8 -1-7 k.6 -2k. 8 8.1 5-1 Soluble COD -1.8 10.2 7-2 7-2 5.9 2.7 16.6 1.8 5.k 8.6 9.3 U.2 1.3 3.3 3.8 6.8 25 ------- TABLE 13. A COMPARISON OF ZETA POTENTIAL, SOLUBLE COD AND BOD5, AND FLOG FORM IN JAR TESTS OF FLOCCULANTS Flocculant Type Alum Fe3+ Lime mg/1 200 395 632 50 75 8U 8U 200 200 Effluenta A B B C C B B A A Zeta potential, pH mv -9-3 0 0 0 to -2U.5 0 3.0 0 5.0 -17.8 3.0 0 11.6 -20.2 Reduction13, % Soluble COD 2.1 17-3 12.2 16.8 23-2 30.5 33.3 U5.3 U.O Soluble BOD 5 0.0 -1.5 3.6 5.3 11.9 21.9 17-2 51.0 5.8 FlocC form __ U+ u+ — — k+ 3+ — — A = special sample of beaten sulfite pulp. B = total mill effluent (Mill F). C = digester room effluent (Mill F) . Values for "as recieved": A — soluble COD = 1+62 mg/1. soluble BOD 5 = 128 mg/1. B - soluble COD = ^53 mg/1. soluble BOD 5 = 190 mg/1. C - soluble COD = 128U mg/1. soluble BOD 5 = 585 mg/1. Q Floe form — 0 = supernatant clarity equal to untreated control. 5+ = clear supernatant without suspended solids. To reduce the problem involved in setting up sampling schedules , sample transportation and storage, it vas further decided to select mills by two criteria: (l) their proximity to our Appleton laboratories and (2) those having high soluble BOD 5 values in the initial survey. Others could be add- ed, as time permitted, to allow studies of special effluents, of effluents with lower soluble BODs, or of effluents from more distant mills. 26 ------- TOTAL MILL EFFLUENTS WITH PRIMARY FLOCCULANTS Tables lit, 15 and 16 summarize the large volumes of data gathered from jar tests with effluents from Mills A, D, and F, which are using convention- al clarifiers or flotation systems. These data can be summarized as follows: 1. Ferric chloride in the range of 75-150 mg/1 Fe3+ was the best flocculating agent for these waste streams. The optimum pH was dependent upon the particular stream being processed and the concentration of the iron salt added. 2. Lime, at higher concentrations (200-500 mg/l) also removed soluble BODs (21-10 percent). 3. Alum did not remove more than IT percent of the soluble BOD5) although there was evidence that, with some of the waste streams, careful pH control and optimized alum con- centrations could result in the removal of substantial quantities of soluble BODs. 1. The polymers plus bentonite or Celite to provide high surface area did not appear any more effective than the cheaper iron salts. 5. Some trials had shown that considerable amounts (22 percent) of soluble BODs could be removed from the clarifier effluents with ferric chloride or lime (Table lU). 6. When the concentration of the flocculating agent and the pH are carefully controlled, from 25 to 50 percent of the soluble BODs would probably be removed from the clarifier influents. 7- In recent similar studies in Finland (ll_), fly ash obtained from the burning of calcium-base spent liquor was used as a flocculating agent, it was used to raise the pH during alum treatment, and it was used as a sludge conditioner. Reductions of 50 percent for BODs and 70 percent for COD are reported in the treatment of kraft effluents. The authors also report that the sludge is easier to handle. ADDITION OF ACTIVATED CARBON Studying pilot-scale columns of activated carbon processing a chemime- chanical (CM) pulping waste, we found that 83-90 percent of the total BOD5 was removed (±2). We have no information, however, on the degree of soluble BODs removed by adsorption on activated carbon. Therefore, we used jar tests to measure the reduction in soluble BODs by powdered (300 mesh) activated carbon, with and without the addition of cationic polyelectrolytes, for the treatment of the total mill effluent (Table 17) and CM sewer discharge from Mill F (Table 18). Soluble BODs could be reduced by 32 percent in the total mill effluent and 17 percent in the CM streams. The variability of the results for the various levels of activated carbon was quite high. This was most likely due to changing conditions in the mill — high hourly variations in the flow and concentration of substances in these streams — rather than to differences in the effects of the activated carbon and polymers. 27 ------- Q 1 -a! a 3 is M w g § § 3 j O fe a ^ i EC TO S ^ H W ^ PM O H M P t"1 W H^ "£ 3 >-5 H 3 9 EH -P a H 4-4 (3 •H P H •H S •g M CO CM O •H - ON -H •p CO CM .p •H O •O II II 3 O ^3 aj u) (it o § § K & O PQ CM H H H ->»^ •H 0 6 •9 VO CO C! C CM 00 O 0) H CM -H OH -p 0 HUH 3 ^- >d t^ O fl) CM o PQ CM vo ^>^ •O OO OO C n t- CM o • o oo o •rl CM CM -H -P H -P •H 0 -d II II 3 gj in -P OOOOOOO OOOOOOOOOOOOO 0) ^ f- t- ^ f- t— t— ITvlTNlAOOOOOOOOOO ^ H H CM CM CM 00 ^ -d- IA in P. ------- TABLE 15. JAR TESTS OF PRIMARY FLOCCULANTS WITH MILL F TOTAL CLARIFIER INFLUENTa Flocculant Type Lime Bentonite Fe3+ {as FeCl3) Alum mg/1 50 100 200 800 100 500 50 100 150 200 12.5 25, 28* 5Qb 70 70* 70° b 81** 81* 81** 100, 112* ll? 150 200 25 50 100 150 158* 200 250 385* 385* 385* 385* 385* 385* 385* 632* 632* Final pH 8.8 9.2 10.6 11.1* 11.6 11.7 — — __ 6.8 6.5 2.7 5-7 5-2 1.6 l.l 8.6 3.5 5.1 1.2 3.2 3.3 6.0 3.8 2.9 2.7 — — — 3.5 — 8.9 6.8 6.2 5.3 1.6 1».3 3.9 10.6 6.9 Sample 8/13 COD = 508 mg/1 BOD 5 = 223 mg/1 Reduction, % COD -7.2 0 7-2 11.8 12.9 — 5-1 7.6 6.5 7-6 __ __ — — — — — — __ _ _ __ — __ __ — — — __ — — __ — .— ... — __ — __ ^^ BOD 5 -10.2 -10.2 -1.6 3-7 15-5 — 1.1* 1.0 -3.6 -1.1* __ — — — — — — __ __ __ — — __ — — — __ — — __ — — — — — __ __ _H "•"• Sample 8/21 COD = 209 mg/1 BOD 5 =102 mg/1 Reduction, % COD -3.8 -1.9 10.0 lit. 8 17-1 20.0 — — — 5-1* 8.9 — 19-8 — — — — — __ __ — 15.8 — 13.9 15.8 9.0 1+.3 3.8 5-2 10.0 10.0 — __ __ __ •~— BOD 5 0 0 21.1* 18.1* 36.7 10.0 — — — — 1.1+ 1.1 — 7.2 — — — — — — — — — 8.7 — — 10.1 18.8 -1.0 U.9 3.9 2.0 — 2.0 2.0 — — — __ — Sample 9/10 COD = 110 mg/1 BOD 5 = 191* mg/1 Reduction, % COD __ — — — — — — __ — — — — 12.9 — 31.3 35.8 31. 8 33-7 28.0 33.3 33-7 30.5 — 32.1 30.7 — — — — — 8.2 — 8.7 19.2 21.9 21.9 23.9 19-2 11.9 12.2 17.8 BOD 5 __ — — — — — — __ — — — — 7.5 — 5-2 2U.O 22.1* 8.8 6.2 12.2 18.8 21.9 — 17-5 11.3 — — — — — — 3.0 __ -5.2 -2l.2 12.0 -18.3 16.7 11.1 7-3 3.6 -2.1 ^Soluble COD and BODS were measured. Added on molar basis of Fe3+ and Al3"1" (0.5, 1.25, 2.0 millimolar). 29 ------- TABLE 16. JAR TESTS OF POLYMERS AND ADDITIVES WITH MILL F TOTAL CLARIFIER INFLUENT5 Flocculant Manufacturer Nalco 607 607 607 609 609 609 609 609 627 627 627 631* 63U 63lt 63U 73C32 73C32 73C32 73C32 73C32 Dow Tydex 12 Tydex 12 Tydex 12 Tydex 12 Tydex 12 Tydex 12 PAA PAA PAA PAA PAA PAA mg/1 1.0 1.0 6.0 1.0 1.0 6.0 6.0 10.0 1.0 1.0 6.0 1.0 1.0 6.0 10.0 1.0 1.0 6.0 10.0 10.0 1.0 1.0 6.0 6.0 6.0 10.0 1.0 1.0 6.0 10.0 10.0 20.0 Additive Type Bentonite A/C None Bentonite A/C None None None Bentonite A/C None Bentonite A/C None None Bentonite A/C None None Celite Bentonite A/C None None None None Bentonite A/C None None Celite None mg/1 200 200 — 200 200 — — — 200 200 — 200 200 — — 200 200 — — 200 200 200 — — — — 200 200 — — 200 — Final pH — 7-5 — 7-5 3-0 7-5 — 3.0 — 7-5 7-5 — 7-5 7-5 7-5 — — 7-5 5-0 3.0 7-5 7-5 7-5 — 5.0 Sample 7/31 COD = 552 mg/1 BOD5 = 23it mg/1 Reduction, % COD — — -1.8 _ — 6.8 10.2 7.2 — 5.9 ~._ — 2.7 10.0 — 1.8 5.1* — — — 3.8 3.3 1.3 6.8 — 8.6 9-3 — It. 2 BOD 5 — — 2.6 — 7.2 11.3 9-8 — — 8. it — 6.8 16.6 — -2.1 23-8 — — 8.1 -2U.8 it. 6 5-1 — 2.1 12.8 — -1.7 Sample 8/13 COD = 508 mg/1 BODS = 223 mg/1 Reduction, % COD 8.2 7.0 — 9-U 8.2 — — — 9-2 9.6 — 8.0 5.1 — 6.7 7.1* 8.0 — 6.1 3.0 12.lt 9-0 — — — — 14.7 8.6 — 6.3 7.1 — BOD5 11-9 -it. 6 — 3.2 18.7 — — — 1.8 16.9 — -1.U l.lt — 23.8 9.1 19.6 — 29.U 21.2 -1.U 20.5 — — — — -U.5 -ll.lt — 21.8 lit. 2 — Soluble COD and BODs were measured. A/C = activated carbon (Filtrasorb 300, powdered). 30 ------- TABLE IT. JAR TESTS OF THE EFFECT OF ACTIVATED CARBON AND VARIOUS ADDITIVES OH SOLUBLE COD AND BOD5 IN MILL F CLARIFIER INFLUENT Activated carbon Typea (1) (2) mg/1 50 100 150 200 200 200 200 200 200 200 200 50 100 200 •200 200 200 200 200 300 koo Additives Type None None None None Nalco T3C32 Dow Tydex Cellulose Nalco 609 Malco 627 Nalco 63>* Dow PAA None None None Nalco T3C32 Dow Tydex Cellulose Cellulose + Tydex Cellulose + 73C32 None None mg/1 __ — ____ 1.0 1.0 500 1.0 1.0 1.0 1.0 1.0 1.0 500 500 1.0 500 1.0 — — Final pH 7.1* 7.5 7.5 7."* 7-5 7.1* — 7.9 7.9 7.8 7.9 — — — — — — — — — - — Sample 8/13 COD = 508 mg/1 BOD5 = 223 mg/1 Reduction. % COD 6.1 5-7 5.9 5-7 8.0 9-0 — 8.2 9.6 5-1 8.6 — — — _— — — — — __ — BODS 19-2 8.7 12.3 -5.5 19.6 20.5 — 18.7 16.9 1.1* — __ — -- — — — — — — — — — Sample 8/21 COD = 209 mg/1 BODS = 102 mg/1 Reduction. % COD l.l 6.9 13. 1 — — — __ — — — ll. 7 17.1 11.9 — — — — — — — — 17.9 17.9 BOD5 3.1* ^_ __ 11.5 __ _^' — __ — — — 17.0 7-5 27.1* — — — — — — — 13.2 32.1 Sample 9/10 COD = 426 mg/1 BOD5 = 196 mg/1 Reduction. % COD BOD 5 __rl n^ — — __ 3.5 6.3 5-2 5.2 — — — — _«. — — l6.lt 7-5 8.0 It.O — 5.2 _ «... __ 2.0 23.5 6.1 12.2 _.. — — — _w — 12.2 9-2 11.2 19-1* — ll.3 — — — a(l) Filtrasorb 300, powdered. (2) Same alcohol and water washed (oven dried). In all of the jar tests the samples were first screened through a 60- mesh stainless steel screen, the activated carbon (and polymers, if used) was added, and the sample was processed through the mixing and settling cycles. No attempt was made to correlate the degree of soluble BODs removal with the presence of fiber fines or other carbonaceous adsorbing materials. Although the activated carbon-soluble BODs complex could be readily fil- tered by a 0.15-um filter, it did not settle in the jars with or without the use of polymer flocculants. Activated carbon, therefore, seems useful as an additive to the primary system for the reduction of BODs if the carbon could be removed in either the primary or secondary system. Larger carbon particles might be easier to remove; however, they would have less surface area and, therefore, probably be less sorptive. EFFECT OF THE CONCENTRATION OF THE EFFLUENT If the total water use in the mill were reduced through recycling, the wastes would be more concentrated. In order to study the flocculation and soluble BODs removal characteristics of such wastes we concentrated by re- verse osmosis a sample of the total mill effluent (clarifier influent) from 31 ------- the Mosinee mill. We have been evaluating and developing uses for reverse osmosis (RO) in our laboratories since 1961 (13). This process can concen- trate solutions and at the same time remove some of the soluble BODs • TABLE 18. JAR TESTS OF THE EFFECT OF ACTIVATED CARBON AND ADDITIVES ON COD AND BOD5 OF MILL F CHEMIMECHANICAL PULPING EFFLUENTa Carbon, mg/1 50 100 125 150 200 200 200 •200 200 300 None None None Additive Type None None None None None Nalco 63^ Nalco T3C32 Dow Tydex 12 Dov PAA None Nalco 63^ Nalco 73C32 Dow PAA mg/1 — — — — — 10.0 10.0 10.0 10.0 — 10.0 10.0 10.0 Reduction COD 2.5 3A 5-9 ^.5 5-2 lt.3 5-9 10.0 10.5 10.2 -1.0 1.0 -k.I of 5 % BOD5 15-9 — 11.6 5.2 12.0 2.0 13.1 9-0 16.8 9.8 -1-5 5-9 -5.1 Final PH 7.^ 7-5 7.6 7.7 7-5 7.5 7-5 7.5 7-5 7.5 7.U 7.U 7.6 Q Soluble COD and BODs were measured. Initial values: Soluble COD = 1105 mg/1 Soluble BODs = ^90 mg/1. Activated carbon — Filtrasorb 300, powdered, alcohol/water washed, oven dried. Reverse osmosis was used to make a fourfold v/v concentrate of the total mill effluent. This solution was diluted with the permeate (fluid passing through the RO membrane during the concentration process) to prepare 3:1, 2:1, and 1:1 concentrations. Jar tests were made with all concentrations, and the samples were analyzed for suspended solids, total and soluble BODs, and color before and after treatment with ferric chloride-polymer and alum-polymer combinations (Figure 6). These data indicate that treatment with ferric chloride at 100 mg Fe3 per liter and Hereofloc 812.3 at 0.75 mg per liter could produce an acceptable effluent for discharge, in terms of suspended solids and BODs, up to concen- tration of approximately 3:1 and reduce both hydraulic and gravimetric loads. 32 ------- -o •V f* •< :'•'.- > [ j * t '."•"•'."..• ^2* 1 •'•••'.•'• ------- With effluents more concentrated than 3:1, flocculating agents did not pro- duce adequate effluents. While this was a rough study and practical means of processing were not worked out — we did not study spill control, the effect of spills on water reuse within the mill, or economics — it certainly indicates that wastes can "be successfully treated at higher concentrations. JAR TESTS ON WASTE STREAMS PRIOR TO PILOT SCALE TRIALS Mosinee Paper Company of Mosinee, Wisconsin and the Combined Locks mill of Appleton Papers Division of NCR, Combined Locks, Wisconsin were selected for the two pilot-scale clarifier trials. Jar tests were made on the total mill effluents. These preliminary tests were followed periodically by other jar tests while the field trials were in progress. Tests on Mosinee Paper Company Effluent Samples from these jar tests were, in some cases, analyzed completely for suspended solids, total and soluble BODs, organic carbon,-COD and color (Tables 19 and 20) in order to measure the removals that might be attained under different conditions of treatment. Others were merely used to estab- lish flocculating characteristics (Table 21) and/or sludge volumes that would develop with the addition of the chemical combinations. In the early trials we established that alum at 200-300 mg/1 and ferric chloride at 100 mg/1 Fe3+ could be used in jar tests to remove BODs ~ 35-55 percent (total) and 12-25 percent (soluble) ~ from the wastes. The addition of 0.75 mg/1 Hereofloc 812.3 appeared to provide additional removal of solu- ble BODs: removal increased from 13 to 35 percent when Hercofloc was added with 100 mg/1 alum, and from 12 to 28 percent with 200 mg/1 alum (Table 19). The use of the Hercofloc polymer alone at 0.75 and 1.5 mg/1 markedly in- creased the removal of soluble BODs: we found lU and 21 percent removal, respectively. Lime as a primary flocculating agent and activated silica as an additive (Table 20) appeared to be relatively effective (^4—16 percent) in the removal of soluble BODs from the Mosinee mill effluent. Of great interest in these studies was the high removal of COD, color, and total and soluble BODs that could be achieved with ferric chloride and alum from the clarifier effluent, from this mill (Table 22). Twenty-five to forty percent removal of the soluble BODs components was obtained with these primary flocculants added to either the north or south clarifier effluents. Lime and "additional" Hercofloc 8l2.3 did not enhance the removal of the "solubles" but did increase the removal of total BOD5, COD and color. Adjustment of the pH from the initial 11, or higher, to 8 with sulfuric acid prior to the addition of the chemicals, markedly increased the amount of suspended solids removed. This only slightly affected the amount of soluble BODs remaining in solution (Table 23). The ferric ion was equally effective as the chloride or sillfate salt when they were used at the same concentration 34 ------- TABLE 19. JAR TESTS OF THE EFFECT OF FLOCCULANTS ON BOD5, COD, ORGANIC CARBON, AND COLOR IN MOSINEE CLARIFIER INFLUENT& % Reduction in Flocculant Type FeCl3 Alum Lime mg/1 50d 100 150 100 200 300 200 300 500 Final PH 6.4 6.0 5-0 7-2 7.0 6.8 11.1 11.6 12.0 Total BOD 5 51 37 58 49 56 56 22 39 41 COD 71 70 76 60 68 73 52 68 71 Soluble0 OC 31 44 47 14 24 31 -13 -1 1 BOD 5 13 25 24 13 12 12 -16 -14 -9 Totalc color 52 43 89 24 43 47 15 58 11 Hercofloc 812.3 0.75 1.5 8.4 8.2 18 19 8 11 3 6 14 21 0 15 Hercofloc 812.3 at 0.75 mg/1 was added after 1 min mixing with the following primary flocculant: Fed 3 10 7-0 No floe formation 25 6.8 No floe formation 50 6.4 56 72 30 30 47 Alum 50 100 200 7.4 7.2 7-0 No floe formation 54 52 76 71 14 21 35 28 18 32 a Initial control values for sample taken 11/21/74: 711 mg/1 215 mg/1 109 units. 87 mg/1 122 mg/1 Total COD BOD 5 Color Soluble OC BOD 5 Filtered through a 0.45 ym Metricel filter. 'rCo-Pt units at 465 nm (NCASI Method). dAs Fe3+. 35 ------- TABLE 20. JAR TESTS OF FeCl3, ALUM, AND ADDITIVES AS FLOCCULANTS FOR MOSINEE CLARIFIER INFLUENT* Floe cul ant Reduction, % Primary, mg/1 Polymer, mg/1 Final pH Total Soluble BOD5 COD OC BOD5 Color Analytical grade primary chemicals: -.c FeCl3 50 Here. Here. Act.Si Here. Act.Si Here. Act.Si Commercial grade primary chemicals: Alum 50 100 200 0.75 1.5 0.75 1.5 10 0.75 1.5 10 0.75 1.5 10 5.8 5-8 5-8 7-4 7-5 7.7 7-7 7.1 6.8 7.1 7.4 7.4 7.2 7-0 6.9 38 37 38 11 10 6 5 13 15 10 11 23 23 21 23 56 57 58 -0.7 -0.7 0.7 4 3 16 3 4 54 54 47 57 49 51 51 20 21 24 34 33 29 42 40 39 39 29 27 27 9 15 10 4 5 8 16 12 8 4 4 4 88 89 90 39 47 45 4o 69 69 72 53 86 88 87 84 FeCl3 50 c Alum 100 Here. Here. 0.75 0.75 5-5 6.5 32 20 50 9 28 13 — alnitial values (for untreated influent control) for sample taken 12/12/74: Total COD BOD 5 Color Soluble OC BOD 5 711 mg/1 215 mg/1 109 Pt-Co units 87 mg/1 122 mg/1. .tx w 3 .-.__ •"Of -i- • Here., Hercofloc 812.3; Act.Sci., activated silica. cAs Fe3+. 36 ------- of Pe3+ and when the pH of the influent was not above 9.0 (Figure 7). Sludge volumes, determined with separate jar tests and the Imhoff cone, were also markedly reduced by the addition of the Hereofloc 8l2.3 (Table 20 and for this reason it was included in all but one of the studies at this mill. TABLE 21. JAR TESTS OF THE EFFECTS OF PRIMARY FLOCCULANTS AND POLYMERS ON pH AND CLARITY OF MOSINEE CLARIFIER INFLUENTS Flocculant Primary mg/1 Fe3+ Alumb Alumc Alum > 25 50 100 150 100 200 300 200 300 100 200 300 100 Polymer , mg/1 „ — — Here. 0.75 — — Here. 0.75 — — Here. 0.75 __ _« Here. 0.75 — — — __ __ — — — __ — w. — — — — PH „ 6.6 6.8 5-0 5.6 — — __ — — — 7.1 7.1 6.5 7.0 6.9 6.5 1/16/75 .. Claritya __ 0 0 0 3 — — __ — — — 1 5 1* 1 5 1* Influent samples pH _. 6.9 6.6 5-7 5.6 U.5 4.3 9-0 8.6 8.1* 7-2 __ — — — — — — 3/5/75 Claritya __ 0 0 4 ^ i i 0 0 1 k __ — — ^_ — — pH 6.8 5.9 — U.2 U.3 3.4 3.4 ^^ — — — _ — — M_ — — 3/11/75 Claritya 0 5 — h k 3 3 _ _ — — — __ — — _._ — — Q Clarity: 0 = untreated control 1-5 = partial to complete clearing of the supernatant after one hour settling. Analytical grade. Commercial grade with precipitate in suspension. Commercial grade decanted to remove the precipitate. Tests on Combined Locks Mill Effluent Since Mill F had been one of those in the preliminary jar test studies on the optimization of soluble BODs removal, only a few additional tests were conducted prior to the pilot-scale field trials. Data from these tests (Table 25) substantiated the earlier findings (Tables 15 and 16) that ferric chloride, ferric sulfate and alum were all effective in removing total and soluble BODs from this waste stream. We found, however, marked variations in the amounts of removal with the different concentrations and combinations of the flocculants used, apparently depending upon the amounts of spent cooking liquor in the mill effluent. 37 ------- TABLE 22. JAR TESTS OF THE EFFECT OF FLOCCULANTS ON pH, BOD5, COD, AND COLOR OF EFFLUENTS FROM MOSINEE CLARIFIERSa Reduction. # Flocculant Type mg/1 pH From North Clarifier: FeCl3 5013 5.1 100 3.U 150 3.0 Alum 100 6.0 200 6.2 300 6.2 Lime 200 9-9 300 11.1 500 11.6 Hercofloc 812.3 0.75 8.8 1.5 8.6 From South Clarifier: FeCl3 50b 5.7 100 3.8 150 3.1 Alum 100 6.9 200 6.5 300 5-6 Lime 200 11.3 300 11.6 500 12.0 Hercofloc 812.3 0.75 1.5 Q Initial values (control) BODs 61 63 6k k3 61 6l 25 33 kl 12 15 63 65 59 No 60 62 No k3 k3 No Total Soluble COD Color 72 72 68 56 70 70 32 U8 55 k 11 68 70 6k flocculation 6k 66 flocculation kk 50 flocculation 65 79 85 k2 65 68 28 k2 k2 7 k6 70 75 27 k6 2 11 BOD5 39 ko 38 26 25 26 -lU -19 0.8 -6 3 38 39 38 31 26 -13 3 of samples taken 11/21/7^: Soluble BOD (mg/l) North=12l* Total BOD (mg/1) North=2ll* COD (mg/1) North=628 , Color (units) North=108 South=130 South=24o South=6l2 South=63 . Fe 38 ------- TABLE 23. EFFECT OF pH ON JAR TEST FLOCCULATION OF MOSINEE CLARIFIER INFLUENTa Primary flocculant (mg/l) Final BOD5 reduction. % pH Total Soluble Reduction in suspended solids, Fe3+ 50 100 150 200 10. 7 10.3 9-9 9.J* -3 -5 1 — -5 -3 6 8 22 55 72 50 Alum 200 300 Uoo 500 600 10.8 10.6 10. k 10.2 10.1 -0.2 -7 U 2 2 1 u 12 7 5 35 18 2 15 30 Influent adjusted to pH 8 with sulfuric acid prior to jar test: Fe3* Alum 50 100 200 300 Uoo 7.2 6.8 7-5 7.3 7.1 6 6 -9 -5 -8 11 U 5 -7 20 75 72 76 8U 80 Influent sample of 2/11/75 had an initial pH of 11.2. The effect of these spent liquors and other mill discharges on the overall efficiency of the clarifier is discussed in Section VII. LABORATORY SURVEY OF SOURCES OF SOLUBLE BOD5 IN MILL SYSTEMS In a mill system with multiple sewers there are certain streams that carry the bulk of the BOD and COD. Some of the BOD and COD can be more easily removed from concentrated streams than from diluted waste, and some substances may alter the properties of others or interfere with their removal. Therefore, isolation and treatment of these more concentrated streams would be a logical way to improve the removal of BOD$ by the clarifiers and thereby reduce the BODs load on the out-plant treatment system. Importantly also, the degree of recycling of the process water could be substantially increased. Separate _Sewer Discharges In our study samples from the separate sewer discharges were obtained from two of the mills and analyzed for various types of solids, for soluble and total BODs, as well as soluble and total COD components (Tables 26 and 27). 39 ------- Figure 7. In jar tests with Mosinee mill effluent, pH 8.5, ferric chloride and ferric sulfate at 25 and 50 mg/1 of Fe3+ (a,), (b_), (d_), and (e_), precipitated more solids than when they were used at a higher concentration, 100 mg/1 of Fe3+ (c_) and (f_) . If^one or more of the high BOD5 streams (the discharges from the CM sew- er in Mill F or the pulp mill, soda recovery or fifth stage brownstock washer sewers in Mill K) could be treated separately by flocculation, reverse osmo- sis, ultrafiltration or some other process, the overall efficiency of the treatment plant would increase, there would be less of a load on secondary facilities, and there would be a greater potential for water reuse (details discussed in Section VII). Breakdown of Fibers Another source of soluble BOD5 was fibers: saccharides and other micro- bial nutrients are released during the beating of pulp or during the recycl- ing of fibers in water recovery systems. In order to study this release of soluble BODs, a beater test was ar- ranged with a sample of unbleached sulfite pulp in a laboratory beater using 40 ------- a conventional 1-1/2 pound (0.68 kg) test with 12 pounds (5.5 kg) on the bed- plate. One-liter samples were removed at half-hour intervals during the 90- minute test period and freeness, total solids, total and soluble BODs, and total and soluble COD were analyzed on each sample (Table 28). TABLE 2k. VOLUME OF SLUDGE FROM MOSINEE CLARIFIER INFLUENTS TREATED WITH VARIOUS FLOCCULANTS AND POLYMER Flocculant Sludge Volume, Final Clarityb Primary, mg/1 Polymer,a mg/1 ml/l pH 0-5 Influent 12/21/74: Fe3+ 50 None — 122 5.U k 50 Here. 0.75 115 5-5 5 Alum 100 None — 6k 6.2 2 100 Here. 0.75 5^ 6.5 2 Influent 12/31/71: Fe3+° 50 None — 92 6.3 k 50 Here. 0.75 80 6.2 5 Fe3+d 50 None — 96 6.3 5 50 Here. 0.75 85 6.3 5 Influent 1/16/75 = Fe3 Alum 50 50 100 100 300 300 None Here . None Here. None Here . — 0.75 — 0.75 __ 0.75 36 30 18 60 96 80 6.6 6.8 5-0 5.6 6.3 6.3 0 0 0 3 k 5 fflerc., Hereofloc 812.3. Clarity: 0 = control (no apparent flocculation) 5 = clear supernatant. Analytical reagent grade of Fed3. Commercial grade, l+2°Be, of FeCl3. Soluble BODs and soluble COD markedly increased as the beating progress- ed and the fibers were fibrillated, releasing materials which became micro- bial nutrients in the BODs test. We ran several flocculation tests on portions of the filtrate from a 90- minute beaten sample (D in Table 28). This sample vas passed through a 0.^5- Um filter to remove fibers and insoluble BODs and COD components. Ferric chloride was again the best "single" flocculating agent (51 percent removal 41 ------- TABLE 25- REMOVAL OF BOD 5 FROM EFFLUENTS OF COMBINED LOCKS MILL BY FLOCCULANTS AND POLYMERS Floe cul ant Primary , d FeCl3 Fe2 (SO i» ) 3 Ferri-floc Alum Lime None mg/1 10 25 50 50 50 75 75 75 100 25 37.5 50 50 50 75 75 75 75 75 50 100 200 300 200 200 — ?H3l2.3, Hercofloc 4/23 BOD5: C4/24 BOD5: dAs Fe3+. Total Soluble Total Soluble Polymer , "" None None None H812.3 N73C32 None N73C32 H813.2 None None None None H812.3 N73C32 None H812.3 N73C32 H812 . 3 N73C32 None None None None H812.3 N73C32 H812 . 3 812.3; N73C32 = 274 mg/1 = 220 mg/1. = 222 mg/1 = 194 mg/1. mg/1 — — — 0.75 1.0 — 1.0 0.75 — — — — 0.75 1.0 — 0.75 1.0 0.75 1.0 — — — — — 0.75 , Nalco Effluent 4/23 BOD 5 removal, % Total Soluble No flocculation No flocculation — — — — 38.3 30.9 36.9 21.8 — — 42.0 30.4 No flocculation No flocculation No flocculation — — — 26.1 20.6 — — __ — — No flocculation No flocculation — — — No flocculation No flocculation — 73C32. Effluent 4/24c BOD 5 removal, % Total — — 18.9 18.5 30.6 29-7 27.5 26.1 — — — 36.9 26.6 29.3 22.5 38.7 32.0 26.6 20.7 — — 8.1 25.2 — — . 5.4 Soluble — — 15-5 14.4 26.3 20.1 23-7 18.6 — — — 27.8 16.0 20.1 9-8 28.4 35-6 14.4 15-5 — — 13.9 10.3 — — -6.7 42 ------- cd M S Pn O CO EH § 1 H 13 M in Q S s -s o o o 3 o EH O S «3j H HH1 ffl ^ O CO p^ o CO o K s co . CM E3 t-3 t— LA \O VO • • • • • t— t— t- t— ^- LA OO OJ ON ^J- 00 H H O t— 1 ^ !H O CU -P 1> IM ft ^ C CU ------- CO o H ft, ^-^ 0 W CO h3 CO M X S n3 •— ' S i-3 ^a §B CO S £ CO P-i 0 ^ s n EH A -3 r^ W 1 — I EH O M 0 ps O CO PP CU •d ^ a Digeste foul co ^^f oo H vo oo vo o o OJ LA o t— oo t— t— o CM CM VO H OO 00 00 H • O H • 1 CO ^s !H CU 0 O CU ?-l cc) •d 0 CO ^O oo vo J- o o H CM ON vo o H t— H OO H O OJ o\ o • OO +3 a cu ^f f— \ w « •H ^4 (U Clarifi cu N •H CO O -P Pi •H ' S Vi -— - 3. H O LA CD ,JCJ o P 'd •H CU C JH S OJ -P CU M LA S co -3- ^J • s o o 5. n-, J? g ° • S co 0 fn LA 0) oJ vo -p -=!• H ^H -H M -p 4H P CO O rt ^ CO O ^3 -P . -P -H Tf S fcjO 3 -p •H -P CU CO PH H CO 1 H 0) O O P-i O O 05 ,0 O 44 ------- of soluble BOD5) (Table 29). We added cellulose powder (500 mg/1, Whatman, Standard grade) to provide particles of high surface area and enhance floc- culation (No. 3 series in Table 29); it slightly increased the removal of soluble BODs (l-9 percent). At pH 7-7 with alum, the removal was 2 percent better than at pH k.2 (2A versus 1A). The addition of cellulose powder and a pH of 7.7 further increased (to 9 percent) the soluble BODs that was re- movable with 200 mg/1 alum. A combination of cellulose, lime and alum at pH 10-7 (^C) was 23 percent better than lime alone at pH 11.8 (1C) or lime and alum without cellulose (2C) at pH 10.7 (29 percent increase). Additional work on the effect of these high BODs streams on the efficiency of clarifiers was done later in the study and is discussed in greater detail in Section VII. TABLE 28. RELEASE OF SOLIDS, SOLUBLE COD, AND BOD5 BY BEATING9" Sample no. A B C D Beating, min 0 30 60 90 Freeness, ml5 750 370 100 30 2U Hr solids , 6/1 0.29 0.32 0.35 0.38 Soluble COD Increase, mg/1 % 33^ 396 18.6 116 2l.6 162 38 . 3 Soluble BODs mg/1 77 77 118 128 Increase, __ 0 53.2 66.2 Q 1-1/2 Ib beater method with 5-5 kg load on the bed plate. Canadian standard. ALTERNATE METHODS FOR REDUCTION OF BOD5 Ultrafiltration is a process for the separation of soluble low and high molecular weight (or size) components into two separate streams through the use of semiselective membranes under low pressure (less than 300 psig or 21 kg/sq. cm). We used Ultrafiltration in two trials with effluents from Mill F. In the first trial a sample of the total mill discharge to the clarifier was processed through a Westinghouse module using a polysulfone membrane that had high pH and temperature limits (pH 11.5 and 95°C). Removal of 59-65 per- cent of the soluble BODs was attained in the permeate (that portion of the solution passing through the membrane). The permeate apparently contained low molecular weight components, such as acetic acid, that had not been re- moved by flocculating agents (Table 30, Part A). The flocculating agents, lime with the polymers Nalco 609 or Dow PAA (polyacrylamide) did not remove additional soluble BODs from the permeate of the total mill effluent. The second trial (Table 30, Part B) was conducted with a sample of the CM sewer effluent, which had been shown in previous work to contain 33 per- cent of the soluble BODs in 18 percent of the flow going to the clarifier. 45 ------- Ultrafiltration without additives removed 69 percent of the soluble BODs, and activated carbon did not remove any more BODs. Apparently the membrane was the controlling factor in the removal of soluble BOD. The carbon did not ad- sorb any of the transferrable components probably because its adsorptive ca- pacity was used up by larger, more heavily charged molecules. TABLE 29. JAR TESTS OF VARIOUS FLOCCULATION TREATMENTS OF FILTERED 90-MINUTE BEATER SAMPLES3" Treatment and Flocculant sample Type 1. Flocculant added: A Alum B Fed 3 C Lime 2 . Above samples treated with alum and flocculated: A Alum C Alum 3. Cellulose powder (500 mg/l) added to No . 1 samples : A Alum B Fe013 C Lime D Control 4 Alum added to No . 3 samples : A Alum C Alum mg/l 200 200 200 200 200 200 200 200 None 200 200 Soluble COD Soluble BODs Reduction, Reduction, pH mg/l % mg/l % 4.2 458 2.1 2.7 256 45-3 11.8 449 4.0 7.7 336 28.2 10.7 449 4.0 4.2 400 14.5 2.7 248 47-0 11.8 422 9.8 468 7.7 355 22.0 10.7 4o6 13.2 105 0.0 51 51.0 98 5.8 102 1.9 109 0.0 95 8.6 50 51-9 90 13.5 104 88 15.4 74 28.8 Q Initial values: pH = 6.7 Soluble COD = 462 mg/l Soluble BODs = 128 mg/l. 46 ------- EH Q HH EH £3 a w 5 K « o o a o CO O o o 0 fo on W ^ ?Q EH If! P § 0) H H 0 CO P o o 0) H fl H O CO * fl o •H t) TSS. TJ 0) K H S c o •H -p O tsSi. jj i3 0) ffi H M d 0) H ft CD CO -p ffi -p £>OS -P ct? o5 CD cd cd ccJ CD d) d)d)" CUCJ (U \|~^ J^J ^) J — 1 ^ JT^ ^- ^J pf \| pT-[ h3 tlD _f~^ LA O <" • • 0) OOO O O O pt< 1 O H 1 O 1 1 \| [ 0 0 i on i CM i i i i CM H en CD •^^p o\ O rr} + ^ ^ 1 g O 0-1 f> ,Q m O.HS O Ot)O C5 O OO ^^ d) -p C_j J^ 0) 0) ^ •H 3 d) •H ------- The ultimate value of ultrafiltration might be in concentrating the high molecular weight soluble BODs components and removing them from process streams. It has been demonstrated elsewhere (2l) that ultrafiltration could be carried to a concentration of 25-35 percent solids without complications from osmotic pressure buildup which directly affects RO flux rates and con- centrations. This would be sufficiently high for final disposal by discharge into the strong pulping liquor streams for evaporation and utilization or burning. Still another use for ultrafiltration is in reducing the buildup of high molecular weight solubles in recycled process water within the mill; this could reduce water consumption and the volume of overflow to the out- plant clarifier. Reverse osmosis, similar to ultrafiltration but operating at higher pressures (UOO-800 psig or 28-56 kg/sq cm) with tighter membranes which re- ject smaller molecules, was tried both as a possible treatment method and as a further refinement for the permeate from the ultrafiltration process (see Table 30). The soluble BODs removal of 89 percent for reverse osmosis alone was increased to 92 and 9^ percent, respectively, by either the ultrafiltra- tion pretreatment or the addition of activated carbon to the reverse osmosis feed. The advantageous use of activated carbon in the reverse osmosis feed, which had been pretreated by ultrafiltration, substantiates the previous con- clusion that the high molecular weight materials of the ultrafiltration feed were utilizing or plugging all of the available adsorptive sites on the car- bon. Ferric chloride was added to the ultrafiltration feed in order to de- velop a floe and possibly increase membrane rejection and soluble BODs re- moval. Ferric chloride actually worked in reverse and increased the transfer of soluble BODs through the membrane. This loss of membrane rejection re- duced the removal of the soluble BODs components by h percent. Subsequent tests indicated that the ferric ion did not damage the membrane: after the module was washed with detergent, it gave salt (Had) rejection comparable to those for a new membrane. 48 ------- SECTION VI PILOT SCALE CLARIFIER STUDIES GENERAL The information from jar test studies of the effluents from the Mosinee and Combined Locks mills were used to prepare a scheme for the treatment of these mill effluents with a pilot-scale clarifier. This clarifier had been developed at The Institute of Paper Chemistry under Project 3029, jointly sponsored by member companies of the Institute, the Environmental Technology Corporation (ENCOTEC) and the Upper Great Lakes Regional Commission (12,). The stainless steel cone of the clarifier had a center height of 1.3 meters (4.25 feet) and a diameter of 1.1 meters (3-5 feet). The inlet was downflow through a 5-1-cm (2-inch) stainless steel pipe with an opening ^0 cm (1.25 feet) from the bottom of the cone. It was equip- ped with an overflow weir which discharged into a 5-1-cm (2-inch) trough around the entire circumference of the top. This section of the report covers the initial laboratory "shakedown" studies, the on-site studies at the mill in Mosinee, Wisconsin from January 15 to April U, 1975> and the on-site studies at the mill in Combined Locks, Wisconsin from April 8 to June 26, 1975- The problems encountered and some of the solutions developed for the correction of these problems are discussed. Data are given and discussed for the removal rates attained for suspended solids, total and soluble BODs and color. TRIALS WITH THE PILOT CLARIFIER UNDER LABORATORY CONDITIONS Three thousand gallons of Mosinee clarifier influent were delivered by truck to the Institute early in January 1975 and were used to check the oper- ation of the clarifier system (Figure 8) without the samplers or the submers- ible pump (P-l). The system consisted of the following items and arrangement: 1. A tank (T-l) provided a constant supply of waste to be treated. 2. A feed control pump (P-2) maintained a constant feed rate. 3. A 35-liter (9-gallon) reactor tank (T-2) with a high speed mixer provided rapid mixing of the waste with the primary flocculating agent and permitted gravity flow to the balance of the stream. U. The flow from the reactor tank was into an open pipe extending to the bottom of the flocculator tank (T-3). If polymer was to be added, it was introduced at the top of this open pipe, 49 ------- o CM T3 'd bO O H CO I I O O H i ^ PH i i i i i i i -P -P O cd 9 -P P 3 H C! O O a3 O O O > O O O O rH H d -rl H O rH PH CQ a o •H -P 03 cl H a (U H aJ o M I ft O O •H -P ao bO •H 50 ------- into a quiescent zone approximately one-quarter of the dis- tance down the pipe in order to provide adequate mixing with- out excessive turbulence or floe dispersion. In the center of this tank was mounted a slow speed (20 rpm =1.6 fps periph- eral velocity) paddle that mixed without excessive shear. 5. The overflow from the flocculator tank passed downflow into the clarifier (l-l\|), with the flow from the weir trough into a small sampling pot (T-7) and then to the sewer. 6. In these laboratory studies the sludge was removed from the bottom of the clarifier by gravity flow through a vented overflow pipe (not shown) that could be raised and lowered to control the rate of flow. Both gravity flow and a pumping system were used in the field trials and are discussed in the following part of this section. 7. The flocculating agents were stored in small tanks (T-5,6) and were metered into the system with tubing pumps (P-3»M. We used this equipment to do a brief laboratory study of the effects of waste flow and sludge removal rates, at a constant level of flocculating agent(s), on the quality of the discharge produced. An excellent quality discharge (Table 3l) appeared to be produced if the waste feed and the sludge discharge rate resulted in an overflow rate of 1^0-150 liters per square meter per day'(^00-^25 gallons per square foot per day). The removal also appeared to be dependent upon the type of sludge blanket developed in the clarifier. Best results were obtained with the level at least five centime- ters (2 inches) from the top to prevent floe carryover, but not less than one meter (3 feet) from the bottom; i.e., 60 cm (2 feet) above the waste inlet. This arrangement seemed to promote the development of larger floe particles and a filtration effect to markedly enhance the quality of the discharge. These parameters were, therefore, selected for the operation of the unit during the field trials. EQUIPMENT AND OPERATION AT MOSINEE PAPER CORPORATION The IPC system was to be operated in parallel with the two mill clari- fiers and was, therefore, installed in the waste treatment plant in a con- crete building. This building was immediately adjacent to the two commercial clarifiers and was unheated, except for heat from the incoming waste stream. Due to the arrangement of the waste stream flow pattern, it was neces- sary to use a submersible pump (P-l in Figure 8), mounted in the waste treat- ment trench immediately downstream from the bar and traveling screens, to lift the waste to floor level (2-2.5 meters or 6-8 feet) at a rate of 19-30 liters (5-8 gallons) per minute. The flow from the submersible pump was directed into the bottom of a 190-liter (50-gallon) stainless steel tank (T-l) for supplying a centrifugal feed pump (P-2) with the excess overflowing to the sewer near the influent sampler. This tank also contained the pH recording equipment. The feed pump (P-2) maintained a waste flow of 11.U liters per minute (3 gpm) into the reactor tank (T-2)} and the only other modification of the 51 ------- equipment, over that described in the laboratory trials, was the installation of a pump in the sludge removal line during some of the trials. TABLE 31. LABORATORY EVALUATION OF PILOT-SCALE CLARIFIERa Over- flowb rate 369 395 406 425 526 579 58? 632 Sludge removal, at 1° 20 13 10 10 15 12 10 Ik Flocculant, mg/1 Primary Fe3+ 87 6k 61 52 51 51 50 50 Polymer H812.3 1.3 0.78 0.77 0.77 0.77 0.77 0.77 0.75 Suspended solids 99 99 99 97 19 -52 -29 -35 % Removal of Color 55 79 — k6 49 k9 48 k9 Total 2k 3k 29 36 19 5 0 9 BOD 5 Soluble 8 17 9 10 13 16 0 9 alnitial (control) values: Suspended solids = 234 mg/1 Total BOD5 = 134 mg/1 Soluble BODs = Ilk mg/1 , Color = 130 units . Overflow in gallons /square foot/day based on surface area of clarif ier SAMPLING AT MOSINEE PAPER CORPORATION During the first part of the mill study, influent and effluent samples were taken with two Model WM-5-24R refrigerated Sigmamotor samplers (Sigma- motor, Inc. of Middleport, New York) set to take individual samples hourly around the clock. In order to reduce the analytical load, these were com- posited into 12-hour samples daily. It soon became apparent that the rapid changes in mill discharge (Table 32 and Figure 9) would not permit such an infrequent sampling scheme. A third sampler, a CVE refrigerated unit (Quality Control Equipment Company of Des Moines , Iowa) capable of taking small samples at short intervals (l-60 minutes) for compositing in a 4-liter (l-gallon) bottle, was placed in the influent stream. This unit was set to take 4o-ml samples at 15-minute in- tervals and the 12-hour composites were compared with those from one of the Sigmamotor units taking 4oO-ml aliquot s from the same influent. Poor correlation was evident (Table 33) and a second trial was made. This was with four samplers in the following arrangement : a. The two Sigmamotor samplers on the influent and effluent streams set at 30-minute intervals , taking 2k 400-ml samples in 12 hours . 52 ------- The CVE sampler on the influent line taking a 40-ml sample every 10 minutes. A polystaltic tubing pump (Buchler Instruments of Fort Lee, New Jersey) set to take 9 ml per minute continuously was used on the effluent line. TABLE 32. ANALYSIS OF INDIVIDUAL SAMPLES TO ESTABLISH RANGE3 Milligrams /liter No. 1 2 3 k 5 6 7 8 9 10 11 12 13 PH 10.12 10.07 9.57 9.63 9.11 9.kB 8.86 9.kO 9.19 9-30 9.78 9.60 9.59 Suspended solids 210 628 k9k 1^5 85 Ikk 265 152 97 76 9k 115 267 Total BOD 5 2kl 206 166 155 129 135 1U3 163 ihB 127 138 132 lU8 Soluble BOD 5 203 178 138 138 116 114 lUo iko ikO 117 135 12k 118 a_ . , 1/16/75). Analytical comparisons were excellent (Table 3^) and indicated that the Sigmamotor samplers could be used to take representative samples if the 30- minute cycle was used. In order to compensate for the "system holdup time" of 90 minutes, the samples were composited to obtain representative samples from the same 12- hour test period. This was readily accomplished by compositing the individu- ally numbered bottles in groups so that those from the effluent were three bottles "behind" those for the influent. Samples of the mill clarifier influent and effluent were taken at 7:30 a.m. daily by the mill staff and analyzed by both mill and IPC personnel. 53 ------- CM ft -P a ------- CQ [-3 Q_J s H H ll pn. • O oo ro W 9 § o u ^J CM r-l ^ft?I ^y M O , \| 0 o in Q 0 m (!) rH ,0 3 t-H O Crt ul § PQ r-H CO O H CQ •a •r-l r-H o en T3 CD •0 0) p. CQ CO EC a Cfl r-l 3 60 r4 o 0) p. CO i-l cfl rH 3 00 0) i~H cfl •H O fl\ U/ W .^ cfl rH §, 00 iH CJ\ rH CM m 00 r- CM CM "^ rH 0 CO i— 1 m rH r-t 00 CO I— 1 CJ^ 0 r-H ^~ ^j- CM r- r~- rH CM 00 rH in i~» oo CN 00 "^ CM CO CT> CM O\ i-H i-H rH 0 CM t-H CM CM CO rH r-- m r-- oo m o 0 C (\) d> rM ?H 0} (D -P H 1 H tn ft a W CO P> to r ^ H nj a -C] O -P 0 -P cfl 4) •H H to p ^ 0 bD V ft 0) H a n ft o a en tn 0 0 -• A -P cu o H CM H ft H CO a s tn o -H «H CQ H cO C cd •H -H O £5 ,^"1 0) -P p4 LT\ H CD r-l r 0} 55 ------- P4 o CO P3 )-J Q-J Sj 3j 9 OJ \|~> • O O H S PS p <3 S M §H g - w w P j ^> W J h PH s O M CO K H M W M >H fc ^ S § 5 0 _^ oo W i_^ PQ EH 'c? •H C o o 4J P-i O O O rH ~5\| a m n O PQ 01 t— 1 F^ O CO rH rH CO S3 ^ h to rH CO m D T^ vJ" 00 CM r-H 01 PS rH •H oo in O vO ON 01 CM a CO rH rH ON •H -3" S rH to <-< •$ 3 d* vo CN in 04 in i— « ^D 1 rH 1 rH r-. co m <\|- oo \o CO CX) CO CO C^ 00 i— f t-H rH ^ O CN O ^H O iH O\ sj- rH O ON iH rH iH r-l i r. i rH CM kO lT» O\ ^ ON oo -d* co o rH m , p J-J cu co cd •H P CO td co ,0 CD H h ft fi to rH 9 CO O ^J- n G_I ^ ° 0) 0) •H -rH 01 CO O 0 ft ft O O o o J^ CD P (3 •H p B •H T o j__i _p CO 01 H ft CO ES o CO ^1 B 01 •8 p CO 01 H • ft •P « £3 CO 0) CO ft \| Ed p g 0 JH «H a "O \ ^ OS ft ft •H p H CO CO ^ H o ft >> c 0) •3 p 0) H ft 1 U] ^J o C •H -P O , CQ pq 5,2 i O O -H O O CO • • •• O t— t— ft n -P P 0 ass Ctj ft CO O O O o o oo .. .. .. 00 -J H 08 08 >rH H OJ B »^- i? ""* 56 ------- For analysis all samples for BODs were stored at it°C (ifO°F) and the others were maintained at room temperature. TESTING SCHEDULE AT MOSIWEE CORPORATION In order to establish a comparison between the mill and IPC clarifiers, the first trial was the operation of the pilot unit under conditions as near those of the mill as practical. Since we were not attempting to develop op- timum clarification at high hydraulic loadings, but merely to study the in- creased removal of the soluble BODs from the waste stream while operating under different chemical treatment conditions, it was not necessary to show that we could exactly duplicate mill operation. Rather, it was more impor- tant to show that any increase in BODs removal was due to the chemical addi- tives and not to the clarifier design alone. We developed a "base line" by operating the IPC unit with Hercofloc 812.3 at O.T5 mg/1 and with a waste flow rate of 11.k liters/min (3 gpm) in parallel with the two mill clarifiers. The Hercofloc addition duplicated mill treatment. If a 10 percent sludge withdrawal could be maintained, the pilot clarifier overflow would be 1.6 liters per sq, cm per day (UOO gallons per sq. ft per day), approximately one-half that of the commercial unit. The efficiency for removal of BODs in the small unit was slightly less than for the mill units under the same conditions and time period. This in- dicated that the design of the IPC clarifier did not promote more removal and that any increases found in later trials using chemical additive would be di- rectly attributable to the chemical treatment. Following the development of the base line, other chemical treatment schemes based on the laboratory studies were tested: A. Base line using 0.75 mg/1 Hercofloc 812.3- B. Alum at 200 mg/1 plus 0.75 mg/1 Hercofloc 8l2.3- C. Ferric chloride at (100 mg/1 as Fe3+) plus 0-75 mg/1 Hercofloc 812.3- D. Ferric sulfate (50-100 mg/1 Fe3+) without the polymer. E. A second base-line study with the Hercofloc. F. Sulfuric acid for pH adjustment only. G. Sulfuric acid for pH adjustment plus the Hercofloc. H. Ferrous sulfate (100 mg/1 Fe2+) plus 0.75 mg/1 Hercofloc. The original plan was to have tests of two weeks' duration. Unfortu- nately, the several periods of problem and solution development we encoun- tered early in the study delayed the program and resulted in shortened test periods (for F, G, and H, above) near the end of the study. Also, throughout almost the entire trial period, the Mosinee mill was on a reduced work week (h days), which limited the operation of the small clarifier to the period between 9:00 a.m. Tuesday and U:00 a.m. Saturday, thereby limiting the number of samples per week. 57 ------- RESULTS AND DISCUSSION OF WORK AT MOSINEE In our first attempt to establish the base line, we immediately encoun- tered three problems: The first problem related to the submersible, centrifugal sump pump in- stalled in the waste trench. It had a small 2.5-cm (l-inch) inlet hole in the baseplate and due to the high fluid velocity in the waste trench, an "aspirator" effect developed across the inlet when the pump was vertically mounted, and we were unable to pump sufficient waste. Mounting the pump at a 90° angle pointed downstream increased the aspiration effect, while mount- ing it pointed upstream resulted in rapid plugging of the inlet hole with pitch and debris not removed by the bar or traveling screens. This problem also extended to the centrifugal pump used for feed control. The second problem was due to the high suspended solids content of the Mosinee waste. A centrifugal pump was the only available pump that was con- sidered suitable for control of the flow at the 11. ij- liters per minute (3 gpm) rate. A one-half inch Eastern centrifugal pump was, therefore, used to control the rate of waste fed to the clarifier unit, and at first the rate was controlled with a 1/2-inch valve on the pressure side of the pump. This rapidly plugged with pitch and fibers and resulted in erratic flow patterns. The third problem developed in the gravity sludge removal system. Set- tling of the high suspended solids content rapidly overwhelmed the available space in the clarifier and resulted in solids carryover, even in the absence of chemical agents, other than the Hereofloc 812.3. In order to reduce these problems: 1. A conical screen (20 mesh), 18 inches long by 7 inches in diameter, was installed on the submersible pump. When installed with the apex of the cone mounted upstream, the pumping action was excellent and the cone was somewhat self-cleaning. When mounted with downstream orientation, which should increase the self-cleaning aspect of the cone shape, the flow vas re- duced below acceptable levels. Periodic cleaning, with high pressure water, was required to maintain an adequate supply to the system. 2. The centrifugal feed pump was fitted with a smooth-bore orifice plate for flow control and with periodic cleaning this provided better flow control. 3. A multihead tubing pump (Brosites) was installed in place of the gravity system for sludge removal. In the last four trials a throttled cen- trifugal pump was tried in this position and found to work the best of any of the schemes attempted. Although none of these worked perfectly and problems were encountered from time to time, failure of the system could be minimized with careful cleaning and operation procedures. 58 ------- Data, for the trials with the various flocculating agents and for the base lines are summarized in Table 35. The more detailed analysis of the in- dividual samples are in Appendix B (Figures B-l to B-10, Tables B-l to B-5). These data were first used to establish confidence limits and the following summary was developed. 1. Both the initial and final trials (A and E of Table 35) with 0.75 mg/1 Hercofloc 812.3 indicated, as previously noted, that the IPC unit was equal to or slightly less efficient than the mill clarifiers in removing sus- pended solids, soluble BODs and color. 2. The reason for the marked difference noted in the total BOD5 reduc- tions for the mill clarifier during the second trial is unknown; throughout the study period there had been a slow, progressive increase in the effi- ciency of the mill clarifier in removing BODs; i.e., from lk to 35 percent between Trials A and H. This was not readily traceable to changes in the concentration of BODs going to the clarifiers during this period, since these were on the average l6l, 120, 138, l8l, l6l, 211, 11^ and 1?8 mg/liter for Trials A to H, respectively. 3. Alum with polymer and without pH adjustment, as predicted by the laboratory trials, did not increase the removal of the BODs components but did markedly increase the efficiency of removal of suspended solids and color. h. The ferric ion, both as chloride and sulfate salts, was an excellent flocculating agent for removing BODs, "the chloride form being more effective than the sulfate for suspended solids. The chloride could have been better because the final pH was nearer to the optimum for the Fe3 produced floes (pH U-5). 5. Ferrous ions, at the level of 100 mg Fe2+/l, appeared to reduce the efficiency of the system even with the addition of polymer. While this level of ion was below that which would be theoretically required by the Schulze- Hardy rule, which states that the concentration of the counterion required to collapse the double layer and produce rapid flocculation is inversely propor- tional to the 6th power of the valence, it was the highest level practical during our trials due to the low solubility of the commercial grade ferrous sulfate. Based on the effectiveness of the ferric ion at 100 mg/1, the level of ferrous ion theoretically required would have been approximately 1100 mg/liter. 6. The use of sulfuric acid to decrease the pH was one process under consideration by the mill for use in the large clarifiers. A trial was, therefore, designed to test this procedure for the removal of some of the components from the waste stream in the small system; but due to the mechani- cal difficulties we had encountered earlier in the trial period, only a short period was available near the end of the study. Although the data were in- sufficient to provide reliable comparisons with mill operation, laboratory studies have shown that pH reduction to below ^.5 resulted in marked reduc- tion in suspended solids and total BODs, with little effect on the soluble BOD5 level in the effluent. 59 ------- E"i £H fV] 5 j PH prj W M CO H PH S F""! H W P^c W H H ^ H LA on 1 ,3* EH V1, • H cd o a CO O H 0 0 co H p H O f> CO R o PQ , — \| cd -P O EH 0) Tj M S3 "d 0) -H ft H co O 2 CO CO rl CO O -P H -rl O S3 O P £>, [ T •H H •i « a fi jj pQ S3 0) 4n „ CD H H ^ to ,— ( ^ o CO «\ P bp o a EH W -d CU •> "d co S3 -d H CU -H — ft H M CO 0 8 p! M CO Jn O cd o oj oo • ^3 OJ -P H •H CO > O •P O •H H p! 0 CJ O J-i PH CO J- c{ p •H £j CO t— 4J H H -H rt 3 H H •H cj CTscO H H -d S3 cd n O o H > 5^ •HO O -P 0 •H H P O 0 0 h PH 0> H W • P) CO OJ -•3" O OJ H -3- \D OO OJ CO VO t~-vo co vo _=)- LA H _^j* V^> O\ H H LA OO O OO H H -3 VO -X MD H 0 H PH -H £. (Y") H CO O H • • H '-v* bD 00 O\ a H oj LA t— • o -p 0) O CO OO CO t— • OJ H CO O o H	00 H H H A -P •H CO H 0 VD CO CO cd S3 0 H 0 OH 0> PH -H CO H S • W • CO S3 O to •H cd ft o O •d 0 O bO (0 •H > 0 ft o -p cd i -p S3 CO •H O •H CM "g CO S3 H W 08 ft e> r* W 60 ------- These trials, with ferric, ferrous, and aluminum ions, and especially with the iron salts, have shown that an increase in clarifier efficiency (12 percent for total BODs, 15 percent for soluble BODs and up to 66 percent for color) could "be attained with the Mosinee waste. The sulfate form might "be preferable to the chloride, particularly if the sludge were to be burned, be- cause it is less corrosive. One of the principal practical observations we made in the field study at this mill was on the effect of strong digester room spent liquors and washes on clarifier performance. Visual, analytical, and other evidence showed that heavy slugs and spills of these wastes — some of which lasted several hours — adversely affected clarification. A single program of grab sample collection was undertaken to help indicate the sources of strong pro- cess waters. The analytical data from the study of that one set of grab sam- ples are summarized in Table 27, previously described, and showed the rela- tive strength of the various soluble materials in these samples. The effects of these streams will be discussed in greater detail in Section VII. EQUIPMENT AND OPERATION AT COMBINED LOCKS, APPLETON PAPERS DIVISION OF NCR The clarifier equipment that had been used in the Mosinee field trials was cleaned and transferred to Combined Locks on April 7> 1975- Since it was again to be operated in parallel with the mill system, it was placed in the treatment plant building which was immediately adjacent to the two mill clar- ifiers. A tee was installed in the main influent line to these two clarifi- ers and this stream was fed into the 50-gallon day tank (T-l in Figure 8). This eliminated the need for the submersible pump. At Mosinee we had encoun- tered some difficulty with uneven flow over the weir of the small clarifier. Although the unit had three leveling legs, the top rim was not level around the entire periphery. Spot overflow occurred; the floe subsequently "stream- ed" at these points and efficiency decreased. An adjustable leveling device for the top of the clarifier at Combined Locks was made of small strips of slotted plastic installed along the entire overflow edge. Carefully manipulated (up and down), these could be used to establish an overflow around the entire circumference. This helped to main- tain a smooth top on the sludge blanket and virtually eliminated spot over- flow except when the strips were disturbed during the weekly cleanup. Two other basic changes were made for operation at the Combined Locks mill to provide better control. The high fiber content of this waste stream overwhelmed the sludge disposal system on the IPC clarifier. In order to re- duce the fiber content and still retain in the influent the fine particles required for good floe formation and BODs removal, we installed a side hill screen with 60-mesh wire over the top of the day tank. Later, when we en- countered difficulties in maintaining a clean surface in the base-line study (A), we changed this to a Hydrasieve (C. E. Bauer, Springfield, Ohio) with 0.010 inch slots (Figure 10). A centrifugal pump with electrical speed control was used to control the sludge level in the clarifier and the hose was clamped on the outlet side. 61 ------- This was changed to run at low speed with the hose clajnp removed, and the time of operation was controlled with a timed OW/OFF switch (Trial D). This permitted relatively close control of the sludge blanket level that had var- ied both with the "quality" of the waste fed to the system as well as with the type of flocculating agent. Figure 10. Bauer Hydrasieve used to remove the longer fibers from the Combined Locks mill effluent 62 ------- Since the level of the sludge blanket in the clarifies was apparently very Important to the efficiency of the system, thought should be given to provision of a sensing type of control system for any future work with this unit. SAMPLING AT COMBINED LOCKS The two Sigmamotor samplers were used on the influent and effluent streams. Comparisons were again made with these two samplers, the CVE sam- pler and the polystaltic pump. The Sigmamotor samplers were set to take 1+00- ml samples every 30 minutes, the CVE sampler a 50-ml influent sample every 8 minutes, and the tubing pump a continuous 9 ml/min sample of the effluent. Samples from each were composited every 12 hours. Data in Table 36 indicate that representative samples were taken by the Sigmamotor samplers set at the 30-minute sampling frequency. To provide for the "system hold-up" we used the same "lag period" in compositing that we used at Mosinee. Samples of the mill clarifier effluent were taken by the mill staff every morning and were analyzed at the IPC laboratories for com- parison with those from the small unit. The mill had no provisions for tak- ing 24-hour composites of the mill clarifier influent, so the influent values for the IPC clarifier for the same test period were averaged for comparison of percentage removal values for total and soluble BODs and color. Since the Hydrasieve removed considerable quantities of fiber from the influent pro- cessed by the IPC unit, the removal of suspended solids could not be compared with that of the mill system. TESTING SCHEDULE AT COMBINED LOCKS The Combined Locks clarifier operates without chemical additives and de- pends upon gravity sedimentation for clarification. The "base-line" trial (Trial A) of the IPC unit was, therefore, made without chemicals. This was followed by trials: B. Ferric chloride at 75 mg/1 Fe3+. C. Ferric chloride at 75 mg/1 Fe3+ with 0.75 mg/1 Nalco 73C32. D. Ferric sulfate at 100 mg/1 Fe3+. E. Alum at 300 mg/1. F. Alum at 300 mg/1 with 0.75 mg/1 Hereofloc 812.3- G. A second base-line study with no chemical additives. Because the IPC system was not automatically controlled and because clarifier influent quality changed rapidly on Saturdays and Sundays, we usu- ally limited our trials to 8:00 a.m. Monday through 7'-00 a.m. Saturday; several trials, however, extended through the weekend. RESULTS AND DISCUSSION The Hydrasieve, cleaned frequently, provided good influent for the IPC clarifier. The basic problems at this site were with plugging of the influ- ent line between the mill main line and the 50-gallon per day tank and with the removal of sludge from the small clarifier to maintain proper sludge- 63 ------- o Jn oj H 3 W) 01 o1 w H O 1 O £) i — ! CtJ •H O CO 0) t^l ft H^ CO 1 — 1 S H °b o "-N. 3 3t* H 9 a P "CD H "i K H O S r^ O -H W P 0 H HO) &^ O ft CO CO 11 0 EH O W ' H n3 31 ~5 y fc a bo SB 41 H • « ul gj g P4 y H m ^3 En H H H oj PC ' Oj H H CO 08 H \| -H CO O M OJ 1-3 ft <\| CO 3 CJ ^ VO CD on H H • bD a o oj PQ HP* <£ EH OJ H P .0 S H a di CO 'H 0 11 Pi CO >. * p O f— t— on H OJ O t— r— on H OJ IA H t-- LTN OJ OJ o o t— LA OJ OJ on vo r— o\> OJ OJ H -^t o\ o OJ CO H J- f_ t— OJ t— \O ^~ 0 O ir\ t— CO f— ON CO IT\ND CO t— H OJ PO H OJ 00 0 0 OJ VO OJ H OJ OJ OJ co rn OJ O OJ OJ -=f i— \0 J- OJ OJ J- -3- L^-3- OJ OJ H -* u^ J- on co OJ O CO 1— rn on J- H OJ O m oo on o on oj on oj ON O CO ON t — ^ — O -=f O ON CO t— H OJ 00 H OJ 1 1 P P P P m C— CO CO IA OJ OJ LA O CO -J- OJ OJ \o o cdoJ H -3" H OJ OJ OJ LA^£) CO LA OJ OJ LA O t— CO OJ OJ -=t OJ O CO H H -=f H CO ON H H O OJ t— I— t— t— CO O\ t— CO b- t— H OJ 00 H OJ §§ O Q .j- OJ ON H O O OJ OJ H 0 0 OJ OJ CO H ON OJ H OJ t-VO OJ l/N OJ OJ -3- J- OJ \D OJ OJ 0 t-- VO ON CO IA \D O H ON H CO t^ t— ^— O 0 CO CO t— t— H OJ 00 H OJ OP p p l/N l/N O LTN t/N H H 1 1 1 1 NO -=I ON CO H H 1 O 1 t— H CO t— OJ ON OJ H 1 H 1 O OJ t- 1^ j- on I O i -» CO O VO ON CO t— H 1 O CO H OJ PO H OJ oi Q O CO on o CO t— H H 1 1 1 1 CO CO H OJ OJ OJ -3- ON H IA OJ OJ O \D ir\CO OJ OJ O J- 1/NVJD OJ OJ H H LA-rJ- ON OJ LfN-^f O H OJ O CO CO H on on o co co H OJ PO H OJ 1 1 P Q P O ON U"N O O IA H H 1 1 I 1 ovo -=f t— H H CO V£) m co H H OJ OO NO O H OJ CO -=f -=f H H OJ ON-S H -=t ON-=f H-3- OJ IA CO IA C— t— IA OJ VO NO t— t— rH 01 PP H OJ PO 0 0 o H t-— OJ H O oj on i i i i H CO H on OJ OJ VO ON H -^t OJ OJ O OJ -3- OO OJ OJ ^t OJ on t— OJ OJ NO t— vo on O LA t— on OJ H VO t^ f- t— H OJ r— co t— i— H OJ PP H OJ 1 1 P P Q O H H LA on t— CO IA C— CO CO CO -3- OJ LA H H LA O OJ C— H H O OJ -3- VO H H CO VO OJ VO H H ON CO H H ON O H OJ O t— LAVO t— L~- OJ LA VO CO c— e— H OJ PO H OJ «P Q P OJ H O LA LA O O LA OJ OJ OJ LA O H OJ H -3- OJ OJ OJ ~J H O LA h- LA OJ OJ OJ VO LA LA UN LA on OJ OJ OJ o vo on c— co ^a- OJ OJ OJ -=t -zt OJ vo co LA oj OJ OJ -=J- H CO -3- oo on OJ H LA on -d- vo LA H CO vo t— ir\ t- t— t— ro co on t— VO ^D c— t— t— H OJ H PP 0 H OJ H 00 0 00 P on j- H H CO • H tn oj H > TO ?-[ > a a h ^i OJ 0) H H ft ft 8 3 CQ • W in t* •H S O M 3 -P Hi S 0 -° o a ij > a P a O W -H J3 5- CO 1 O OJ cB H 03 0! !»,& ^3 c a O C OJ OJ M •a M a OJ ai -p -p -P •H H S^1 ft 0 goo 0 IAJ- o 01 .. •• Sn W W 0) 01 OJ * H H w S S H oj co ft 9 H h CD W Co W -H H 0 ^J H a> bo H Oi ------- blanket levels. The first remained a recurring problem throughout the study. The sludge blanket was fairly adequately maintained with the installation of the time-controlled centrifugal pump. Data from these trials (Tables C-l to C-7 and Figures C-l to C-lU in Appendix C) were checked for validity (mathematical confidence limits) on th- IPC computer. Averaged data for the seven trials are given in Table 37 and are summarized as follows: 1. The first base line (Trial A) showed that the IPC clarifier removed slightly more total BODs than the mill unit did, but removed equal amounts of soluble BODs and no color. In the second base line (Trial G) at the end of the series the units operated at equal efficiency. 2. Ferric ion, either as the chloride or sulfate salt (Trials B and D) provided excellent removal for total and soluble BODs as well as color. 3. The addition of the polymer Nalco 73^32 (Trial C) to the ferric chloride slightly reduced the efficiency for total and soluble BODs removal but had no effect on the color removal. U. Alum, as predicted by the laboratory jar tests, did not appear as effective as the ferric ion in removing BODs but was equiv- alent to it in reducing color. 5. The addition of the Hereofloc 812.3 polymer to the 300 mg/1 alum markedly increased the removal of total BODs to well above the level achieved with the other flocculating agents but had no effect on the reduction of the soluble BODs. The high "negative" soluble BODs removal in the two IPC base-line trials (Appendix C-l and C-7)5 as well as throughout the test period for the mill clarifier, could not be explained. This finding, however, did correlate with the darker color of the sludge from the mill clarifiers (lighter fibrous ma- terial was in the influent waste stream) and might indicate development of both color and soluble BODs from the "particulate matter" during clarifica- tion. Spent liquor entering the influent system through spills or deliberate discharges had the same deleterious effect in this study at Combined Locks as it did at Mosinee. This, according to the Combined Locks mill staff, has been one of the causes of clarifier upsets that have occasionally plagued their operation. Operation of the pulp and paper mill under strict control of spills and similar discharges (i.e., good housekeeping) has minimized these upsets. In our laboratories we studied the effect of the various sewer dis- charges on clarification, and the data developed in these studies are dis- cussed in detail in Section VII. 65 ------- TABLE 37. CLARIFIER STUDY - COMBINED LOCKS PAPER CORPORATION AVERAGE EFFLUENT QUALITY AND REMOVAL VALUES FOR IPC AND MILL CLARIFIERS Effluent quality Suspended BOD5 solids, Total, Clarifier mg/1 mg/1 A. B. Soluble, Color mg/l units Both units without chemical flocculating IPC Mill IPC unit k6 kQ with 232 25k ferric chloride 190 188 (75 mg/l Fe agents : 205 227 3 ),and Removal , % BOD5 Total Soluble 22 13 .1 .3 mill using 8. 8. no 6 3 Color -20 -33 .6 .5 chemical treatment : C. D. E. F. G. IPC Mill IPC unit and mill IPC Mill IPC unit treatment IPC Mill IPC unit IPC Mill IPC unit using no IPC Mill k5 7k with using 66 146 with 37 50 with 60 5^ with 169 21k ferric chloride 1^7 187 (75 mg/l Fe 112 207 ko 2k .1 .1 3+) + Nalco 73C32 25- 5- (0. 8 6 75 32 -2k mg/l), .5 .7 no chemical treatment: 196 21k ferric sulfate 136 166 168 209 (100 mg/l Fe 129 152 92 199 3 ), and 77 130 alum (300 mg/l), and mill using no 205 2ko alum (300 mg/l) 197 216 + Hercofloc 123 24l 812.3 3k 28 .k .k mill using la 29 • 9 .0 20. 1. no 25- 12. k 0 31 -k7 .8 .It chemical k 1 18 -38 .1 .3 chemical treatment: 32 21 (0. .6 .0 75 mg/l) 13. k. 2 8 , and 31 -33 mill .7 .9 chemical treatment: 85 56 237 262 208 212 Both units without chemical flocculating only): IPC Mill 50 52 239 236 21k 208 129 229 agents 252 25k 50 30 • 9 • 9 13. 11. 3 7 30 -23 .3 .8 (gravity sedimentation 27 28 .8 .7 -1. 1. U 4 -26 -27 .6 .6 did not sample influent; values are based on average of IPC influent 66 ------- SECTION VII JAR TESTS WITH INDIVIDUAL SEWER DISCHARGES GENERAL In Section V of this report, the BODs sources making up the total dis- charge of the mill to the clarifier were listed and briefly discussed. In the field studies (Section VI) we had noted a marked decrease in the effi- ciencies of both the mill and IPC clarifiers when larger than usual amounts of spent liquor were in the waste stream being processed. In this section we detail and discuss the studies made on the efficiency of the soluble BODs removals realized when individual sewer streams were treated. These individual sewer effluents were also used to make a "synthet- ic" mill effluent by compositing each on the basis of its volume to the over- all volume of the total mill discharge. Using these same individual streams, we prepared various composites, omitting one stream at a time, to determine the effect the omission had on the overall efficiency of BODs removal by clarification. Ferric chloride (25, 50, 75, 100, and 150 mg/1 Fe3+), alum (50, 100, 150, 200, 250, and 300 mg/l) and combinations of these two primaries with 0.75 nig/1 of either Hercofloc 812.3 or Nalco 73C32 were used with each stream and composite. Values for total BODs and, when time permitted, soluble BODg for the treated samples were compared with untreated (settled) control samples for the determination of BODs removal. TESTS ON MOSINEE INDIVIDUAL SEWER SAMPLES Some of the mill discharge streams, especially the "high density," pulp mill and paper machine sewers, could be effectively treated with the proper concentration of the primary flocculants (Figures 11 and 12, Legends 1-11) or primary flocculant-polymer combinations (Legends 12-23). These bar graphs display that in others, such as the foul evaporator condensate, soda recovery wastes, and digester room wastes, the BODs level was not substantially af- fected by any of the flocculating agents. Those bars shown below the zero line in the lower section of the figures indicate negative removal of BOD, probably due to prolonged holding time. These waste flows should be treated by routes other than clarification (i.e., steam stripping of the conden- sates). In our earlier work with the individual streams we found that toxic ma- terials were in some samples, and they interferred with the BODs test (Table 38). Therefore, we could not make even good estimates of the concentration 67 ------- ^D ON CM OJ O II CO 0) \|\| rH •§•3 rH 4^ W B Lf\ ^t CM rH t— 11 H rOH 3 a) H -p CT\ r-H CO It H (U II ^H rr- 3d T 0 $ 4, CO EH T H * 0, « -* Q II OJ pq ra H a CO T-J rH C o OJ 0 H (H O <•£ t. t- 0 p. ^> ? Wr3 Hr° CO CM OJ CM H CM 03 CM .H t- d H ft VO 3 H (5 H H O H LfN IA H >> $ rH M vn «> ^ T) -^ row ro CM CM CM CTs rH CO S H -H CO O rH Qj IOJ rH ^1 (11 3 S1 r-H CO VO OJ H ) n 1 CM CO o oo t— 0 0 o d r- 3 a rH t S c ir\ i « 0 & •f + 01 ft. & rH r §" f l^ U CM f 00 - rH r + + 01 0 ft. p. rH >- 3> 1 tn c H o i 1 D J H 3 CM OO J CJ 3 00 H t- -\| 3 O 1 CJ u id 3 S3 + U 01 -, f^ H rH i "i ~\ i/> - 1— t u^ - rH + a l/N J CO \| 0 H oo £ cofloc s + + fo 0> fc H If o u\ rH vo H -r 01 PK H "9 o H J- \| o-l CM rH CO C 0 0 C 0 f rH t s> ' 3 0 0 C H rH OJ O - CO 0\ C r* 1 1 1 1 1 0 -* !H 0) ft ff • ft^~~ ^ (U -P rH 0} H ra • ^ . S a- H cs p § - . id H 0 -H H t> «H 1 S * ^ g 3 3 "> crj Q. H rH 0) " T < P 3 0 in " 5 J 00 W 'H ------- f 1 1 1 f"~" <~\ 1 H H , H -P 0 0 W ^ IT- 1 1 1 ° r^ oo o ' H 3 ' d) II , ^ 1 H -P ' << £ OJ «s [^— Q OJ VD O SLA .H \|\| ,3- JKJ O 3 1 " ^ •H ,0 H J -P 3 aj ! •H H -P \| i 1 1 1 1 1 1 O O O O O C LA ^t 00 OJ H 00 OJ OJ ^ OJ 4) H to oj 0 > OJ OJ 0 (3 H 4) i-l C \£> 0) O H -p H M •rH O ir\ oo 5? OJ 0) i >. VD O r-i O •a Lf\ 'O o > C 1 1 1 1 _^ r 1 j on H cy oo oj ro ro O CJ O m o to t— H t- o oj on o o oj o • 0 o • H m oj in H m Oj rH H 0) dj f-H PSCOOJCOOJSGSSCOOJ H OO OOMH OOOJ "v. o fn o co "^ ~^. o co • M H t— r-l f- \|bpMHt~-ro OOOO OOCOOJ i/\ooouLr\mocj ro f-^H h H b- t- JH r-J OO 0> oj ff lAOirvQOOOOOOOLTv OJ ITit— OlAlAOlAOlAO OJ H H rH H OJ OJ CO r-fOJOn-^t UA VO t— CO ONO r-fOJ H H H 1 1 1 1 D O O O H OJ p"> 13 0) § •H +> e o O T) H H " •H !H a « 01 01 0) CD C bO •H vH to TJ £ - IIH 03 ^> o O) O <1J -P <1J M fc ffl (U a ^ M O ,S m -p -^ UA 69 ------- TABLE 38. BOD OF INDIVIDUAL EFFLUENT STREAMS AND TOTAL MILL EFFLUENT OF MOSINEE MILL Source Dilution mg/1 Digester sewer Pulp mill sever 5th Stage brownstock washer High density sewer Foul evaporator condensate Digester blowdown foul condensate Soda recovery sewer Total mill effluent 120 600 150 300 600 1500 300 600 1000 2000 30 100 150 120 300 600 300 600 750 600 1500 60 150 637 633 1210 1^90 1629 1620 9l0 1089 1353 1310 lll 171* 172 570 762 751 1023 1325 1300 1236 1188 271 280 of BODs in these streams without careful dilution techniques. Excessive amounts of these same toxic agents would also interfere with the operation of a biological secondary treatment system and for this reason, if for none other, their presence in the total effluent should be held to a minimum or completely eliminated. Discharges into individual mill sewers were used to prepare a "synthetic" total mill effluent in the following volume relation- ships : Paper machine High density Pulp mill Foul evaporator condensate Soda recovery Million gal/day 10 1-5 0.6 0.3 0.28 % of Total 77-9 11.7 l.6 2.3 2.2 70 ------- Digester blowdown 0.07 Digester 0.05 5th Stage brownstock washer 0.035 Total 12.835 100.0 To find the conditions for maximal BODs reduction when individual wastes were omitted from the system, we made combinations of seven of these eight wastes and treated them also in jar tests. Data in Figure 13 indicate that higher levels of BODs removal could be achieved with less chemical if certain individual discharges were eliminated from the flow. That is: 1. The "total" effluent required 250 mg/1 of alum for 17 percent reduction or 75 mg/1 Fe for \2. percent reduction in BODs. 2. With the soda recovery sewer flow eliminated, a 250 mg/1 alum level resulted in 26 percent reduction and the Fe3+ at 50 mg/1 resulted in a 51 percent removal of BODs• 3. The absence of the pulp mill sewer from the total flow, with all other flows present, permitted a 60 and 63 percent reduc- tion in the BODs concentration in the waste stream for 100 mg/1 alum and 50 mg/1 Fe3+, respectively. When 1 percent v/v of the digester room waste was added to a mixture of wastes having relatively good clarification characteristics (paper machine and high density wastes), higher concentrations of flocculating agent were needed to produce an effluent of adequate discharge quality (Figure lU). Attempts to calculate BODs balances for these various discharge combi- nations, and to correlate these with the actual BODs values obtained, was apparently hindered by some interaction(s) between these wastes. This could have been due to changes in the toxicity of the mixtures for the microorga- nisms in the BODs test. The difficulty in calculating material balances for the total or soluble BODs of the individual streams and composites is due to the fact that all of the initial BODs values for these samples were on "settled" jar samples. Settling removes BODs and the removed BOD could not be accounted for in the calculations of BODs balances. In these samples, the interaction of the ad- sorbing character of fibers and particles with the dispersing character of the various mixtures could markedly affect BODs removal during the settling step, thereby modifying the "initial" BODs of the mixture. The importance of this interaction in the overall flocculation-clarification process should be investigated further in any attempt to improve the process. TESTS ON INDIVIDUAL STREAMS FROM COMBINED LOCKS MILL The five streams making up the bulk of the flow to the clarifiers at Combined Locks ~~ from the main, "CM," wet room, Number 5 and Number 6 sewers — were tested individually for the removal of BODs in jar tests with ferric chloride, alum and combinations of these with either Hereofloc 812.3 or Nalco 73C32. We also prepared a composite "total" mill effluent and compos- ites lacking individual sewer outfalls, as we did with the Mosinee wastes. 71 ------- o 00 r-t 00 O II rH H -P O O OO rH VO oo HOO 9 ^ W Q SO) \|\| 1 1 H "T 1 ,0 H H 2 d : •H O O -P CO EH 1 1 1 1 1 o o o o o o c \T> LA -3- OO CM H 0) 60 t— u3 H IS f! D VO 4J to H LA O -P 1-t t— rH VD H H H LA a O H 0) H 3 M ^ +3 £ CM g H H \£> 0) CM O ! 00 5 > t-- ^ 0 1) CM t— r-H VD rH H 00 H H a! •p MD -* OO > ZD CO i OO OO OO 00 J CM CM CM CM H H H rH H H0 OO OO OO OOCJOOOO CD i — 1 C i — 1 t — i — 1 t — i — 1 CQ OOCJOOOO " J-lH?-\|rH^rH^ i WSWSWSW * W)bpbpw)W)bflW)CM J> i'S'i'aug'^aJHaj.H LALALALALALALA °°J3 oooodooHt-., + + + + + + + oo5i o o C (uciajaiaicur-Hius^o H H ^ rHHHHrHHH-^ "~~- CH LALAOO LA LA O t— f— (11 CM CM LA IA t— t— O • --p to ^^ ^^ rt x«^ .^. r> O 1-HrHrHrHHrHHrHg o O •H O ------- As predicted by our earlier work, those streams having low spent liquor con- tent, i.e., Number 5 and main sewers, had good supernatant quality (Figure 15). The BODs removal was good with relatively low concentrations of Fe? or alum (Figure l6). .3+ Figure lU. Jar tests show mixed effluents with addition of, left to right, 25, 50, 75, 100, and 150 mg/1 Fe3+; the con- trol, no addition, is jar f. A Mosinee effluent was used composed of: (bottom row) paper machine and high density sewer effluent, and (top row) the above with 1 percent v/v digester room effluent. The chemical concentrations of ferric chloride and alum required to ef- fectively remove both suspended solids (for high supernatant clarity) and BODs was higher for the "total" effluent than for the various composite streams (Figure IT)- As the individual streams were removed from the system, the quantity of flocculants was reduced, especially for those mixtures with- out the "CM," wet room or Number 6 sewers. The Number 5 sewer stream had good removal characteristics, with low chemical requirements, as an individ- ual stream, and omitting it from the mixture did not appear to enhance the BODs removal. 73 ------- Figure 15. Jar tests with Combined Locks individual sewer effluents, a-e (main, "CM," wet room, machine No. 5 and machine No. 6). Top row: with 25 mg/1 Fe3+; bottom row: with 20 mg/1 alum. Control wastes did not settle without additives and are not shown. 74 ------- o LA II H cd o EH ON II H 0 1 [ 0 \| H 1 II cd "p H 0\ II 1 1 L " r'" H Cd rH .p +J •H O M 1 1 I I O\ CO \O ^ h 4) OJ » H H O H ON^ ^H 00 I -* Si OJ rH H H 10 r-2 ^ fc "* & co^ H LA j — t CO rH OJ rH ON CO -H 1 LA OJ H O O O O O O LA ^t CO OJ rH OJ CO o CO t— o 0 3 rH t— O m 0) rH bO o H t— H OJ rH J O CO OJ mg/1 Hercofloc 8] t— o 00 i — ! LA OJ co 5 0 OJ 0 H 1 OJ o co t- o o rH cd H £ 0 01 1 — 1 hO 01 ON 5! H o CO co OJ mg/1 Hercofloc 8] t— o 5! i i i o LA O rH H o 0 rH ^ 0 OJ OJ CO o CO t— o o H cd a rH t— o H 1 — 1 tiO o IA H H H rH 0 LA rH LA Hercofloc 812.3 £ o H i o o rH OJ rH CO OJ H co 0 o 1 Hereof] t— o « rH 0 r-l VO OJ CO o co t— o o Is H LA t— o 5! ^J o o H co rH rH rH •H & 01 O o T) O si •H S 5 01 •H •H g 01 ------- $ 1 1 1 c OJ I ^ 1 fi 1 1 ?! ' 1 rH a 1 J ( i. ' -p f O 1 i H •»— £ "> H Q -=T •H «J •rt O 1 1 1 1 1 1 O O O O O O O VD LA J- on OJ H H LA rH JL\| OJ ,Q H* -i "s S OJ Ivo H LA LA rH fa -1 "g ts OJ \D H •H 6 ,~ P LA tj H OJ 4> rH ^ 33 o on£ r-l on H = oo e "~3 LA* ^S rH H LAH H oJ OJ ( on on on on CM CM OJ OJ 01 rH H H rH £3 QOOJCOOJCOOJCOOJ 3 on on on on m oooooooo , oonoonoonoon >H Ht-Ht-H^Ht- -P HH H-t Vl ^H 01 OOOOOOOO oooooooo MMMMIMhOMM _j aaaaaaaa 'j LALALALALALALALA !^ t— t— r— ^- t- t— t— t— T) jOCDCDOCDCDod '^! + + + + + + + + ^ fl CJ\OrHOJOn-4-LA\D 0 HHHHHrHrH "3 o o o •? s _. CM -P 1 rH .H 1 CO OJ 0) J on JIJ 0 0 § o on H< ------- SECTION VIII DISCUSSION OF THE MECHANISMS OF SOLUBLE BOD5 REMOVAL The studies with jar and pilot-scale clarifiers have readily confirmed that substantial quantities (as much as 20 to 50 percent) of the soluble BODs are removed in clarifiers by coagulation aids and that the removal can be a significant factor in the treatment of pulp and paper mill waste waters. Our data show that, at least to some extent, the reactions leading to improved BODs removal can be optimized in well operated and controlled clari- fier systems. Several studies were made in an attempt to better define the mechanism for reduction of soluble BODs during the treatment of process waters in primary clarifiers. It had been evident from the beginning of the study that some of the soluble BODs could be removed from the waste stream by several processes (membrane, ion-exchange) and that no one mechanism could be credited for the entire effect. GEL CHROMATOGRAPHY STUDIES The components responsible for the soluble BODs i-n clarifiers can be di- vided into two classes, wood-derived chemicals and additives. The wood de- rived chemicals include degraded cellulose, hemicellulose, hemicellulose degradation products (reducing sugars, acetic acid, methanol and uronic acids), extractives and in small, perhaps minor amounts, metabolites common to living tissues (adipic acid, oxalacetic acid, etc.). Additives used in papermaking include starches, polysaccharides, gums, latexes, resins, dyes, etc. Some mills will also have organic solvents such as methanol, acetone, etc., which are biodegradable. The organic solutes that pass the 0.15-um filter probably comprise colloids, low molecular weight polymers, oligomers, monomers and simple organic compounds. To better understand the removal of soluble BODs by coagulation, we treated the effluent from a chemimechanical mill (Combined Locks mill) with the various flocculating agents. Using gel chromatography we measured mo- lecular size distribution before and after flocculation and clarification. Samples were filtered on a 0.^5-ym filter prior to chromatography. The gel chromatography was carried out on Sephadex G-50, a gel which excludes (allows no equilibration with the gel pores) molecular weights 10,000 and above and includes (allows complete equilibration with the gel pores) molec- ular weights of 500 and below. Molecular weights between 500 and 10,000 are fractionated according to the number of pores with which the molecules equil- ibrate. 77 ------- Figure 18 shows the TOC (total organic carbon) of 2 ml of a tenfold con- centrate as it elutes from a 1 x 55-cm column. The second curve shows that after flocculation with TO mg/1 ferric chloride not only are large molecular weight residues removed, but also low molecular weight molecules. The curves in Figure 19 were obtained when a similar sample (l-ml sample of fivefold concentrate) was fractionated on Sephadex LH 20 (mol. wt. 5000 to 200) before and after coagulating with 10 mg/1 of Nalco 73C32, a cationic poly-electrolyte and analyzed for aldoses by the phenol-sulfuric acid test (8). The low mo- lecular weight peak almost disappeared. Molecular Weight 10,000 5,000 200 1000 800- 600 200- Before 10 20 30 UO Column Eluate, ml Figure 18. TOC and molecular weight distribution of gel-chromato- graphed chemimechanical mill effluent before and after coagulation with 70 mg/1 Fe3+ as ferric chloride. 78 ------- Molecular Weight 10,000 5,000 200 o GO l.U 1.2 1.0 0.8 S 0.6 w O.U 0.2 .After Before 10 20 30 Column Eluate, ml Figure 19. Phenol sulfuric acid test for aldoses and molecular weight determination of gel-chromatographed chemi- mechanical mill effluent before and after coagula- tion with 10 mg/1 Nalco T3C32. One-mi samples of eluate were analyzed. The phenol-sulfuric acid test measures aldoses and polymers of aldoses when borate interference is observed. The solutions analyzed in this report showed enough borate sensitivity to indicate the test was responding primar- ily to aldoses and polymers of aldoses and not to noncarbohydrate aldehydes. When a sample fractionated on Sephadex G-50 was divided into three frac- tions corresponding to high, medium and low molecular weights and analyzed for TOG, aldose and BOD5, the results summarized in Table 39 were obtained. The percentage change for the soluble BOD5 was not as great as that observed for the aldoses or TOG. Apparently not only low molecular weight biodegrad- able materials were removed but also low molecular weight nonbiodegradable organics, Das and Lomas explored the different mechanisms of the flocculation of cellulose fines with the cationic polymer polyethylenimine (2). They were 79 ------- able to show that bridging (the adsorption of two sections of a polymer to two colloids) was a significant part of the cellulose fines flocculation. Such a mechanism could also occur with the other polyelectrolytes and with ferric chloride which hydrolyzes and polymerizes to a polyelectrolyte (l4). The mechanisms for the coagulation of cellulose fines involved collapse of their double layer followed by adsorption and bridging. Lower molecular weight residues were generally difficultly adsorbed. Their participation in coagulation probably depends on salt formation and chelation. Such salt for- mation by hydrolyzed iron can be demonstrated with gel chromatography. Figure 20 shows the elution curve for a mixture of ferrous ammonium sulfate and so- dium gluconate. The complexes absorb ultraviolet light and are, therefore, detectable with the UV absorptiometer at 280 nm. Iron values were obtained by the o_-phenanthroline test. Because the curve shows that at a pH of 2.0, most of the iron elutes from Sephadex LH-20 and G-15 (fractionation range 150 to 10,000) in the highest molecular weight fraction, the complexes prob- ably have molecular weights of 10,000 and above. Figure 21 shows the dis- tribution of the complexes when the pH of the solution was 7.0. At higher pH more of the iron was in the lower molecular weight fractions. TABLE 39- REMOVAL OF DIFFERENT MOLECULAR WEIGHT FRACTIONS OF ORGANIC RESIDUES BY FLOCCULATION Q Sample Effluent treatment UNF HMW MMW LMW Phenol-sulfuric acid test; total absorbance, U80 nm Filtered -- 7-5 3-9 5.2 Floccedb and filtered — 3.0 2.U 2.5 Reduction, % — 59-9 38.6 51-9 TOG, mg/1 Filtered 3^2U 77^ ^57 2301 Floccedb and filtered 2150 320 272 1387 Reduction, % ^3-9 58.6 10.5 39-7 BOD5, mg/1 Filtered 381+0 552 ^87 2689 Flocced° and filtered 3080 379 368 2131 Reduction, % 19-8 31.3 2U.U 20.7 aUNF, unfractionated; HMW, high molecular weight (10,000 and above); MMW, middle molecular weight (500-10,000); LMW, low molecular weight (500 and below). b Flocced with ferric chloride, 70 mg/1 Fe. 80 ------- 3000 Molecular Weight 1500 150 H300 o 00 C\J -p 0) -p -p ra C 03 Ultraviolet transmittance a o 100 100 125 Eluant, ml 150 175 Figure 20. Ultraviolet light transmittance profile and molecular veight distribution of a gel-chromatographed solution of sodium gluconate and ferrous ammonium sulfate, pH 2. If these biodegradable lov molecular weight residues which participate in the coagulation are carboxylic acids, then the mechanism for their removal could be the complexing and/or chelating of the Fe3 of the ferric chloride. As the ferric chloride hydrolyzes, more and more insoluble oligomeric forms of the hydroxide are formed. Whether or not the iron system would flocculate the low molecular weight residues without concomitant flocculation of poly- mers and/or suspended solids was not determined. The types of carboxylic acids which might be derived from wood and form complexes with iron include the Krebs cycle acids (citric, adipic, oxal- acetic, etc.), oxalic, aldonic, and uronic acids, and aromatic acids. Since little ultraviolet absorption was found associated with the low molecular 81 ------- weight gel chromatography fractions, aromatic residues have not been consid- ered in this phase of the study. 3000 Molecular Weight 1500 150 -200 Ultraviolet transmittance I o ^ H 105 Eluent, ml Figure 21. Transmittance and molecular weight distribution of a gel-chromatographed solution of sodium gluconate and ferrous ammonium sulfate, pH 7.0. The sewer to the Combined Locks clarifier has five principal tribu- taries two of which gave aldose test responses strong enough for analysis by gel chromatography after the effluent was concentrated fivefold. The column elution curves are shown for the digester room effluent and the paper mill effluent (Figure 22). The digester room effluent showed predominantly high molecular weight residues and the paper mill effluent, low molecular weight residues, 82 ------- o oo 01 a a) •s o 0) 3.0 - 2.6 2.2 l.U 1.0 0.6 0.2 10,000 Molecular Weight 5,000 500 Digester room effluent I Paper mill effluent _ 20 30 50 50 Column Eluate, ml Figure 22. Phenol sulfuric acid test for aldoses and molecular weight distribution of gel-chromatographed chemimechanical pulp mill effluents 83 ------- Since this particular mill uses starch as one of its additives, one of the aldoses could be glucose and/or related starch degradation products. STUDIES ON INFLUENT AND EFFLUENT SAMPLES FROM THE FIELD TRIALS Clarifier influent and effluent samples from the two trials, at Mosinee and Combined Locks mills, were sent to the Analytical Section of the Insti- tute for carbohydrate and weak acid determinations. Data are given in Table ko for the results of these analyses on samples from high, intermediate and low soluble BODs removal levels during clarification. The carbohydrates were determined by gas chromatography after acid hy- drolysis and derivative formation (15). The weak acids were determined by passing the sample through a cation-exchange resin column and conductimetri- cally titrating it with 0.1N_ sodium hydroxide. No consistent correlation was apparent between the soluble BODs and the carbohydrate removals, indicating that in the high soluble BODs removals some precursor(s), in addition to other than the oligo- and polysaccharides, were being removed by flocculation and clarification. There was slight correlation between the soluble BODs and weak acid re- moval values for these samples, particularly those from the Mosinee study. STUDIES WITH MODEL COMPOUNDS On a modest scale we screened model organic compounds for removal from solution by flocculation in order to learn more about the chemical nature of susceptible compounds. The compounds were selected on the following basis: 1. Those compounds having at least some structural relation to substances found in mill effluents, either as atypical compounds or "families of compounds" (16-1.8); 2. Those compounds with some BODs without marked "toxic" effects; and 3. Compounds having sufficient solubility so that gas chromatography, total organic carbon or some other convenient analytical procedure could be used to monitor the results of jar tests. k. Removals of TOC from pure solutions were early recognized to be critically affected by pH. This variable was evaluated carefully and the interesting and probably important results provided in the following figures are discussed with the realization that the highest levels of TOC removal were often noted at pH levels outside the ranges normal for mill wastes or discharged to treatment systems. The jar tests were made with added fiber in the form of unbleached kraft pulp which had been disintegrated in an Oster blender for 5 minutes to pro- vide a surface area for adsorptive bridging. Fibers had increased the remov- al of BODs in some of our earlier work (Section V). This effect was confirm- ed with two runs with hydroxybenzoic acid and p_-methylbenzoic acid with Fe3 84 ------- a 0 s o 3 H 0 o g 2 (HJcfl W Fi S o M M CO EH O < S o M 2 f 1 s s ° H W ° 3 V £ W CO W EH t5 O J_3 P H >H CO PP w rf S OS ^E P W JM pq o co pq •3 EH o co pr, ^ o g o Ej S 0H • o ppT 3 9 EH •3 (U 5: ^ 1 • o H ,0 cd £H I> co o o s (U K o w; P S £> W O g •i to •H O OJ •3 £ H "a s 0] 2 § in « S jj o H •d a co « tn g -d P -H •H H -d o 0) [fl g O 85 ------- (as FeCla) at various pH's, with and without the addition of fibers (Figure 23). TOG analyses were used to monitor changes in chemical concentrations of the model compounds before and after clarification. 0) K 50 ko. 30 20 10 £-Hydroxybenzoic acid Removal 50 ko 30 20 10 n p_-Methylbenzoic acid /^^x • / TV - X ' y / A O - V 1 1 1 1 pH pH X With 350 mg/1 kraft fiber added O Without added fiber Figure 23. Effect of pulp fibers on removal of soluble compounds with 100 mg/1 Fe3+ (as Fe Cla) at various pH's. , 350 mg/1 kraft fiber added; , no fiber added. Solutions were made from an amount of the model compound calculated to give a TOG level of 90-95 mg/1 in the final volume to be used in the study. This was first dissolved in approximately k liters of distilled water; the correct amount of blended fiber was added (350 mg/1 final volume), and the volume was brought to either seven or nineteen liters (final volume), depend- ing upon whether one or two sets of jars were to be used for the evaluation. One liter of the mixture containing the model compound and fibers was added to each of six jars. Five of the jars were treated with flocculating agent (various concentrations) and the sixth was used as an untreated "con- trol." All jars were settled, sampled and filtered (0.^5 Vim). TOG was de- termined on the filtrates from the "control" and the five treated samples. All "removal" values are based on a comparison with the TOG concentration of the "control" after filtration. In this way the adsorption of the model compound on the fibers without flocculant addition could be monitored. In all tests the adsorption onto fibers of the controls was less than 8 percent (based on weight calculation), except for the arrowroot starch and dextrose which were found to be 65 and 2k percent, respectively. Ferric chloride and alum were the primary flocculants used. Alum (Table kl) did not remove significant amounts of model compounds except polygalact- uronic acid (87 percent removed at 200 mg/1 alum concentration). The highest removal of model compounds obtained with lime up to 500 mg/1 was 2 percent. 86 ------- TABLE hi. TOG REMOVAL FROM SOLUTIONS OF MODEL COMPOUNDS TREATED IN JAR TESTS WITH ALUMINUM SULFATEa Removal of Compounds TOG, % 3,lt-Dihydroxycinnamic acid 0.0 Hydroxymalonic acid (tartronic acid) 1.0 Methylmalonic acid 1.0 D-Galacturonic acid 1.1 p_-Hydroxybenzoic acid 2.2 Propanedioic acid (malonic acid) 2.6 p_-Methylbenzoic acid 3.2 Dextrose lt.0e Isopentyl alcohol (isoamyl alcohol) It.7 l,lt-Butanedioic acid (succinic acid) 7^ lt-Hydroxy-3-inethoxycinnamic acid (ferrulic acid) 8.3e 3-Methoxy-lt-hydroxy benzaldehyde (vanillin) 9.1e trans-Butenedioic acid (fumaric acid) 10.le Arrowroot starch ll.lt p_-Hydroxycinnamic acid 13.3° Benzylmalonic acid 3lt.lte Polygalacturonic acid 87.0 Q All compounds were treated with all levels of alum. Only the best removals are listed in this table. Alum = 50 mg/1. ^Alum = 100 mg/1. Alum = 200 mg/1. eAlum = 300 mg/1. Data for the TOG removed when other compounds were similarly tested are given in Figure 2ka,-e. These data, are very complex, with low molecular weight shortr-chained compounds (acetic acid, ethyl acetate) being removed at a 20 percent level, 2-pentanol and tertiary amyl alcohol showing less than 5 percent removal and hexanedioic acid, ItO percent removal. Several aromatic carboxylic acids had high removals, while galactouronic acid reductions were less than 5 percent (Figure 2ke). 87 ------- 0) * -P O CS CM -p X o <; 1 1 1 1 1 O 0 O O O C7\ GO t— ^O IA X / J \ X \ " \ \ \ X - 1 1 1 1 x- X \ Lf\ (U \J- 0) X P <•> ^ 0 W « X O O X -=r H o— o — o - •? * on — 1 1 1 1 1 1 1 1 1 -p •H 0) W -p U~"\ f^ QJ S -p P 0) woo »• ft cd ^ OH : cup. H Q)-~- rQ S (1) 0-, Pi +3 H " cd O Ti -p rn .,-4 n ooooo oooooooooo Ocd -a- "n 0) -p O 0 0) P ^ 5 X X X X 4 1 I 1 1 X Tj X •H X 1 0 O W "> 1 X < O O X \| P. 0—0—0 x 0 32 '^ 5 \- o > a) 5 K~i — I I 1 I I i i i i •^^ r^ V-* a -m cd o w ) — 1 Tj bD Hi O g O cd LA Cd -H j- oo !> -p 4^> -P cd cd cd H w - — - " 00 rH m T^J 0) H -H •r-l 0) Cd OOOOO OOOOOOOOOO VifJH -3~ oo c\j H CT\ co ^ \o LT> _j- ro c\j ^ Q ft m -H 51U3H * (001) ItJAOTOa » r-l cd H p — cd ft « + 0 43 oo o3 piq Q \£) — ^ X 1 1 {- X \. X \ \ \ 2 1 1 1 1 ? y x tc f m TJ o O xy •H O O ** o o— O I <* 1 0. 0 -H 1 H 1 -=t a 1,- X X O •oo — 1 1 1 1 1 1 1 1 1 v£) ^ i \| ^ ^ Vi cd O -P , M ------- w w O CM OJ c\l ^ ^ X — 1 1 1 1 1 1 1 oooooooo vo -i O <•> X O "> O eg "> J- 0 W 1 W W >< 1A H O— O— O— O \- 4^ (O — 1 1 1 1 1 1 1 f* CO H •^ CJ O 0 ft-rl a o 0 -H U >t) (U (3 K H cd ft r13 X e H^ ra « J- 0) O O S3 H -H CO <(H !> -d- ro c\j ^-1 t— VDLrx^troojrH OW ft M ? H a o o •H XJ O ^ o LPv Cd O -p -p cd cd H — > X x- 1 (U O X CD CM v 0 " \ C ffi \ •H CO J «H - «H O CtJ K - 1 1 1 1 1 1 1 ,V£) — t X \ r— ( m CM CM M « / W OWWWOW / ft d o— u— u— o — o o cd W V ^ "c a5 1 P4 X i \ \ 1 I I 1 1 1 1 I •H 0) ^ Vi PL, cd rH vo PJ w j_> H cd fn P3 v_x (1) O, 13 + ^ «H OJ fl jirs g fe ~~IA J- 00 CM H t-^OiA^J-fy-iOJH 3 (OCXL) (001) bO 89~~ ------- •H O •^1 ' — * O -r! •H O o <; -d o CO _ X \ X 1 S3 33 Jf O O pj rrH \| 1 W -H 0) >d o ^ 3 •>». O O LSJ ft -H fl g O tt) - 85-° (1) « (1) fl Tj oooooooo oooooooooo i— I m M g - p H PM M 0) ft TJ ^ tn 5" C 3 33 O O « - W) !> oJ a -P O O 03 -H LPv O C^ ' — H 00 —^ 05 O fl TJ OOOOOOOO OOOOOOOOOO j CH o -p O -H ^ ' Vl O ^3 O H vH I LTN "^ -d - -p bO <1) •» 0 0 fl H , S CH O -p id CH 0 3 q W H ,b CD "* a OJ ro IA j- m CM i-H at) (OOL) (OOL) T^AOBiaa 3! 90 ------- \ o to o t— o oo (OOI) -A-^ B3 ft (3OL) o o\ (OCXL) o co o t— (001) O •H O O -H N O rH fl N -p > -P H O O -H 03 O i a x 0) O TJ •H O ------- •d •H 0 < O 9 a •H U S? o 4 >> 33 Aj 1 1 O O O\ CO 0) a 01 N C3 0) .? 1 K.. £~ O H B ° tj X3 >? ° ^3 »•-) S3 d 'H T) 0 >1 M J3 0 •rH tn p» j> 1 0 1 o (^ x " 1 1 g ^ s ^! ^! 1 -p -H O ^ -H TJ !s 1 •> ^ 0\|TJ " W -H Tj » O __. d 0) aJ p a o j-H rH M 0 oooooooo ooorQO r— \.o i^ j fn (M rH ooNCopi^Sj (OOi) TUAOraaa »/ ^ "n "E o (001) TBAOUISH % m ff g •H -P tH TJ 0 p \|~ ° C 1 W 0 ^5 iJ S3 X 0 O / J d 0 33 /- ^ o_L« o-y- w / ? o -:wifi-x \ Q \ >l m ji 1 1 1 I 1 1 1 I 1 H •— 00^— ^ cd 1 tf > W !>j -H O - X 0 0 W 0 cd j -H fl 3 rCf CJ W O O 1 O ft -H -3- C <-* b 3 ^^~ CO » -p J M {* TJ CJ g -H c3 +3 a H o ctf oj cd LA bfl oo--- o •>! 00 > » S ^H f?-0 « + O -H -P10 M O 0 ^H d> TJ 03 33 p x w/ \ / \ / / T / X X - 1 1 1 1 1 1 1 ^£) -H — H ?"> ^ 5? !-< O .c3 fri 4:) OJ X in S « \| 1 -H X rn o ; 1 < X ^ &o / O -H / S3 Jd 9 vJ ° -3" T3 ro X O - S C / 33 .. Xi A / O / \ I'H ' O33W/ \S3 -3" O X O~-O"--O"Vi /) O \ ro X 1 1 1 1 1 1 1 1 1 VO CD PH f>j L_, jg O lj f. , 1 , r=q H-H 1 , -H o pj a 'H O OH » Jj ir\ -~~, CU 3 -p tip (1 -p o g cu o X 0) N ct! a In 0 a H «H O CU CC3 -=r H H £> bO 0) J- fi C\J a> J r . OOOOOOOO OOOOOOOOOO S t--^Dir\-=tcvic'jH ^cor-vDu-Nj-cocMr-j •-> a) (OOOi) T^Aonrag % (001) T^Aomsu ^ jrj 92 ------- Some of the greater reductions in TOG could be traced to the formation of insoluble salts with the ferric ion, such as for the 3,^,5-trihydroxy- benzoic acid (Figure 2Ud). This would not account, however, for the removal of p_-hydroxybenzoic acid from solution since the ferric salt is soluble. Nor would the esters and alcohols, which do not form salts, be removed by this reaction. The water solubility of the compounds, however, roughly correlates with their removal from solution with Fe3 (Table U2). The compounds group into two classes, those with greater than 10 percent solubility in water and those with less than or equal to 1.5 percent solubility; the former group tends to have removals of 15 percent or less and those in the second class tend to have removals of 25 percent or greater (Figure 25)• We must also consider not only reactions such as the cross-linking of starch phosphates with various cations and the increased complexing tendency observed for K+, Ca , Zn++, Cu"1"* and Fe that Wettstein, Neukom and Deuel (19) studied extensively but also the oxalate or salicylate complexes with aluminum or iron as discussed by Stumm and Morgan (lU). Stumm and Morgan studied the formation of ferric-hydroxo-complexes and found that they tended to polymerize with other charged molecules at rather exact pH's. Such in- teractions might be occurring in our jar tests. Additional work would be required to pinpoint the effects on soluble BODs removal by the organic compound chain lengths or branching, the presence of phenolic or carboxylic groups and the solubility of the individual com- pounds, all of which appear to have some important relationship to the over- all soluble BODs removal mechanism. 93 ------- TABLE 12. SOLUBILITY IN WATER OF VARIOUS MODEL COMPOUNDS AND THEIR REMOVAL IN JAR TESTS WITH Fe3+ AND PULP FIBERS8" Reference Q No . Compound 1 2 3 1 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 2l* 25 26 27 28 Acetic acid Methyl acetate Ethyl acetate Oxalic acid Lactic acid Methyl lactate Raffinose Adipic acid 2-Pentanol Tertiary amyl alcohol Fumaric acid Maleic acid Malonic acid Tartronic acid Benzylmalonic acid Succinic acid Benzoic acid p_-Hydroxybenzoic acid o_-Hydroxybenzoic acid p_-Methylbenzoic acid 3,l,5-Trihydroxybenzoic acid Vanillin Resorcinol Catechol p_-Hydroxycinnamic acid Ferulic acid D-galacturonic acid Polygalacturonic acid b a Solubility, Removal, % % Miscible Soluble 10 15 Miscible Decomp. in HaO 15 1.11* Soluble 12 0.63 Freely soluble 66 Very soluble Not given; probably less than 1$ 8 0.3 0.8 0.2 Slightly soluble 1.2 1 50 30 Not given; probably less than ~L% 1 Soluble ? 11* 15 18 1* 15 10 8 10 5 5 37 2k 8 9 25 37 30 31 13 50 50 10 8 >3 58 73 6 99 "Model compounds and % removal shown in Fig. 23- 2le. Solubility in water, from Merck Index, 7th edition. 94 ------- ^_ ill +3 at rH C eu ,Q 01 CO 3 o ti O OJ rH ^ CJ -P Jf O CU CO ft CU TJ cu = a G H 05 O & 13 IA CQ ft cd cu CO EH a CQ c! V S Ctl -H h C cu l-l -H +5 TH J3 O 0> bo O CO CO -p p TJ t3 -P O co C C O XI m 3 P H +3 CU O O ft f ft ft C a a a cu cu 3 o o h > K O O CO CD Cfl cu -p 0 S5 C\J • OJ. rH r ~i en . ^«-, S3 . -r . I 9 . fl Ijt ' ,. -• h' iS ' o 6' I 1 1 1 1 1 1 1 1 DOOOOOO O O O D ON CO t— V^3 LA -^ m C\J rH . CO Cfl Hi -P 0) £ CU k .H S- "8, Co • QJ J3 >> -P +> cd 4J -p a a " CO -H CU 0 cd CO CQ H O .H T3 CU CO -H h T3 H CU rO 1) (3 >H rH P ft 'HO CU H CM a > ft cu 4J X bO " 9 ft O O C! OH ^t CH P 0 rH ft « o ^ . rH 1 1 1 1 1 1 1 1 O\ co t— vo IA j rn oj -, — — LA . OJ H 0 C\J • OJ— 2 CO • — rH rH 5 1— 1 t:- 1 — 1 !A H OJ oj — H & -P eg 0 OJ Pi 3 r-J O M mj r- 1 ° > -P • + Cm •H CU PH £_, 0 <^ rC S -p -P CQ -H 3 0 >> -P (1) -H J- H n "^ 'H ° 1 rQ CO 03 CM > O W. OT 3 ------- SECTION IX AREAS FOR COST REDUCTIONS This research study on the mechanisms for the BODs removal in clarifiers was not intended to cover comparative process economics. However, evaluation of the data obtained in the laboratory and practical observations in the field studies indicate that improvements in the clarification process will reduce costs. Studies on the flows from individual process sewers demon- strated that treatment which removes soluble BODs prior to the clarifier is likely also to reduce the overall cost for the total system of treatment. SEPARATION OF FLOWS ADVERSELY AFFECTING CLARIFICATION EFFICIENCY There are several ways to reduce costs, and careful control of the qual- ity of the in-flow to the clarifier is one of them. Proper clarifier perfor- mance (reduction of suspended solids, COD, BODs, etc.) is dependent upon sedimentation and flocculation properties of the influent flow. The input of lignin and other pulping and bleaching constituents having colloidal and dis- persant properties is of particular concern. The reduction or elimination of flows containing such materials, and particularly the containment and sepa- rate treatment of large spills of them, should substantially improve the clarification efficiency and should have high priority in programs for cost reduction in waste treatment. Revised handling should produce the following substantial economies: 1. The capital investment for a clarification system will be less when the clarifier size is reduced; 2. Reagent and other operating charges will be less as clarifier effi- ciency improves; 3- Clarifier performance in all categories, including the amount of BODs removal, should be greatly improved; k. Although no measurements were undertaken in this study, sludge de- watering, sludge volume and sludge disposal problems were apparent and should be subject for further development; 5. Needs and costs for subsequent (secondary biological) treatment may be significantly reduced in terms of capital cost and operating charges, and the need for secondary clarification may also be reduced. OPTIMIZING THE TYPE AND AMOUNT OF CHEMICAL ADDITIVE This project also showed the importance of optimizing the amounts and types of flocculating agents for various clarifier inflows. Different chemi- cal additives are needed for different mills. Process fluctuations and changes affecting clarifier performance were apparent within the short term of each mill field trial. 96 ------- Operating experience may make possible the development of routines for addition of chemicals vhen the effluents change. The rising trend toward de- velopment and use of computerized systems to control the manufacture of pulp and paper may logically extend to automated changes during effluent treat- ment. Of the two mill clarifiers used in the field studies, one was utilizing chemical additives routinely and the other clarifier was well equipped to use flocculating chemicals. Capital costs for additional chemical feeding and mixing equipment would not be a major expense. Operating charges for the chemical additives that we used was about 3<£ to 10^ per 1000 gallons of clar- ifier inflow (Table 1+3). Since these chemicals might logically replace less efficient additives already being used, cost could be considerably less in actual practice. The efficient use of chemicals would, moreover, be expected to give a return in reduced overall waste treatment costs. The actual cost of chemical additives to remove soluble BODs cannot be adequately assessed short of detailed engineering cost studies to fit the situations in each mill. TABLE 13. CHEMICAL COSTS FOR CLARIFICATION a Type of Required, Cost, flocculant mg/1 ^/lOOO gallons Fe3+ 25 2.72 50 5.M 100 10.88 Alum0 100 3.13 200 6.26 300 9-39 aAddition of 0.75 mg/1 Hereof loc 812.3 would add 0.5<£. bAs ferric chloride (FeCl36H20) at $90/ton. CAluminum sulfate«l6 H20 at $75/ton. 97 ------- SECTION X REFERENCES 1. Gehm, H. State of the Art Review of Pulp and Paper Waste Treatment. Environmental Protection Technology Series, EPA-R2-73-1814, 1973. 2. Das, B. S. , and H. Lomas . Flocculation of Paper Fines. I. Adsorption of a Flocculation "by Polyelectrolytes . II. Study of the Nature of the Solid Surface and Soluble Impurities. Pulp Paper Mag. Can. 7^(8): 95-100, 1973. 3. Williams, D. G. Minimizing Chemical and Fines Buildup in White Water by Chemical Means. Tappi, 56(12) :lll-7, 1973. k. Back, E. L. Note on Dissolution of Wood Material During Pressurized Refining and Water Pollution Consequences. Svensk Papperstid. 77(ll): 39^-6, 5- Standard Method for the Examination of Water and Waste Water. 13th Edition, APHA Method 219, 1971- 6. Standard Method for the Examination of Water and Waste Water. 13th Edition, APHA Method 220, 1971. 7. Hodge, J. E. , and B. T. Hofreiter. In: Methods in Carbohydrate Chemis- try, Whistler and Wolfgram, Academic Press, New York, 1962. 388 p. 8. Collins, J. W. , A. A. Webb, and L. A. Boggs . Characterization of lignin and Carbohydrate Residues Found in Bleach Effluents. Tappi, 5^(l):105- 110, 1971. 9. Chamberlain, N. S. , and R. J. Keating. Water Technology in the Pulp and Paper Industry. TAPPI Monograph Series No. l8: 15-7 » 1957- 10. Gould, M. , H. Lundgren, and J. Walzer. Primary Treatment by a Flotation System. Tappi 9th Annual Environmental Conference, Houston, Texas, May 15-17, 1972. 11. Jensen, W., and E. Meloni. Use of Waste Chemicals in Kraft Mill Efflu- ent Treatment. Paper World Research and Development Number, 1971. 12. Lueck, B. F. , and B. McCuaig. Continued Study of Advanced Waste Treat- ment Systems for Combined Municipal and Pulp and Paper Wastes . The Institute of Paper Chemistry, Report No. 5 for Project 3029, Upper Great Lakes Regional Commission Assistance No. 103201183, July, 1971. pp. 15- 5U. 98 ------- 13. Wiley, A. J., G. A. Dubey, and I. Q. Bansal. Reverse Osmosis Concentra- tion of Dilute Pulp and Paper Effluents. The Institute of Paper Chemis- try, Environmental Protection Agency, Washington, DC. Publication Number 1020^EEL 02/72, February 1972. 358 p. 14. Stumm, W., and J. J. Morgan. Chemical Aspects of Coagulation. Jour. AWWA, 5^(8):971-92, 1962. 15- Borchardt, L. G., and C. V. Piper. A Gas Chromatographic Method for Carbohydrates as Alditol-Acetates. Tappi, 53(2):257-61, 1970. 16. Hrutfiord, B. F., T. S. Friberg, D. F. Wilson, and J. R. Wilson. Organic Compounds in Pulp Mill Lagoon Discharges. Tappi 12th Annual Environmental Conference, Denver, Colorado, May 12-1)4, 1975. 17- Brauns, F. E. The Chemistry of Lignin, Academic Press, New York, 1952. 8oh p. 18. Brauns, F. E., and D. A. Brauns. The Chemistry of Lignin, Supplement Volume, Academic Press, New York and London, I960. Qok p. 19- Wettstein, F., H. Neukom, and H. Deuel. Cation Exchange Equilibrium with Starch Phosphate. Helv. Chem. Acta U: 1919, 196l. 20. National Council for Air and Stream Improvement Technical Bulletin 253, Dec., 1971- 21. Collins, J. W., L. A. Boggs, A. A. Webb, and A. J. Wiley. Tappi 56(6): 121-3, 1973. 99 ------- SECTION XI GLOSSARY Alum — Aluminum sulfate. BOD — Biochemical oxygen demand is based on the oxygen requirement of living organisms while utilizing components of the waste stream for growth and/or reproduction. BOD5 — The biochemical oxygen demand in a 5-day test period at 20°C. CM — Chemimechanical pulping; this is based on a short chemical cook followed by mechanical refining to separate the fibers in the softened chips. GOD — Chemical oxygen demand is the measurement of the oxygen equivalent of that portion of the organic matter in a sample that is susceptible to oxidation by strong chemical oxidants (e.g., chromic acid). Colloid ~ A phase dispersion to such a degree that the surface forces become an important factor in determining its properties. General particles of colloid dimension are 0.001 to 1 micron. Gel chromatography — A column chromatographic technique which fractionates molecules on the basis of molecular weight. Hemicellulose ~~ That fraction of plant stems which is made up of carbohydrate polymers other than cellulose. Molecular exclusion ~~ The mechanism of fractionation of molecules during gel chromatography. Larger molecules are restricted from diffusing into the gel matrix, thus taking less time to travel through the gel than smaller molecules. Permeate — That portion of a solution passing through a membrane during re- verse osmosis or ultrafiltration. Reverse osmosis (RO) — Osmosis in reverse flow through a semipermeable membrane when external pressure in excess of the osmotic pressure is applied. Soluble ~ As used in this report is that portion of the solution (solvent and solute) passing through a 0.^5 ym filter under 100 psig nitrogen pres- sure . Zeta potential ~ That portion of the total potential drop between the surface of the solid and the suspending liquid that is contributed by the charge potential between the liquid adhering to the wall of the particle and the movable liquid. 100 ------- SECTION XII APPENDICES APPENDIX A: COMPANIES AND AGENCIES SUPPORTING PHASE I STUDIES Bergstrom Paper Company Hammermill Paper Company Hoerner Waldorf Corporation Kimberly-Clark Corporation The Mead Corporation NCR-Appleton Papers Division Nekoosa Edwards Paper Company, Inc. Potlatch Corporation The Proctor & Gamble Company Scott Paper Company Wausau Paper Mills Company Department of Natural Resources, State of Wisconsin U.S. Environmental Protection Agency 101 ------- i s o H EH < ffi g o o ejj S £ s I— i CO o £H ___^ bO e LJA t— O " ' CO CO M CO o 3 0 o 1 EH M •^ w H O PH t-H M 5 0 i — \| 1 m &q 1-3 c 0 •H oj § •0 (U 0 ^ m c 3 UH w x-x ^- 4-1 (U i-H C M t I 1 i i p 0 m 14 Q 1-H 0 U ,_, "i \r Q 5 CQ " o 03 co c/3 O O r-H tX S , IT P o w « r-H O CO CO c/} o T-H 0 u I O I-H 3;r-o ^mo Oj ,— ( CM i— i CM CMOJCMCM fnm^MCM $• (} -H rHCMJC1"1 r-HCMjIi-H i-HCM^1"1 ^ d M id id td « « s s s s s s u 1 -u en -J~ oo os O I-H D CM CM CM CM «n m p i-t CM en -j- ^ -^ ~^! ^ ^ Cl CM CM CM CM CM CM CO •3 H C CD O fl "8 -P H PH W) ft •H rH ^ O a £H N K. d O o3 ^ 03 S 0 P O a e to .. &, * < ^ O O ^ CD on ro a; -p •' •• -p -H H a3 O -H O O ft g S-i O -P a a o ° d «5 ft H OJ O -H N o o s a s m IH ^j t— t— I ^ c O a) a a ------- O 4J pH -,H O C O 4J r-l -iH O C . 1 . 1 1 1 \c o m i o m r>- i CM m ^o i so -l i—i rH t^C» rHrHi— IC^J O-JCNCM rH rHCSlXrH rO ,0 rH rH I rH rH CM & rH •s a -H S m m EH g ^—- I OJ CO ' I I 103 ------- o M EH O P-. § O e 1 H CO i EH H W) a LT\ t — O --- m OJ rH CO 3 11 o o ;=> -P Q) H If O o H P K 3 W O o H £5 M ^ O e g n ffi 5 O m I pq \| g (3 O -H •U O 13 £>CO^Ol CNOLOI rH CO CT\ I CT* CO U~l 1 OO CO CO OO CNI CN CO 1 i— 1 CNIp-)CNf> ool o r. oo I CMOJC^J coQOt-Hl l CMOOrHt -CJ-CNOI vO 1^- . • a a » « M j ^ ------- O 4J i—I -iH o c -Q - rH < o 00 2 6 EH H 5= ITl rH CO r.r^.ooi \£)t-Hr^mr-i-H CMOOCOO mo-^cy> CO-vfCMCM co o rH i—I CM CM OCT>O>\| CTtOCO) CO i—I i-H f CMincMl I-H r— QO r^oor- oooooo COCOON 1 -p +J Qj S >i H .. » h < t>- t-- P W O cd I a s oj ^ O O W I QJ OJ (in CM 73 OJ 03 •Hi I § I " W I O rH OJ M I -H 105 ------- EH fe g cn ;=> o w o 5 „ K S 5 ga o CO M Q-> CG O H ^ U 0 <[J o 0 K j— 1 pq p 5* § W « M P CO H O 0 E o rH "i H If P o » rH 0 en 3 en O o o ;H 00 i ir CQ rH O CO Q O eu i— i 3 rH O cn 2 o H rH o en CO CO CO 4_) •H Cl JZ) ^ 43 3 O CO rH 4J rH OO R ffi CO 4-1 •H ^ CU rH \| O C/J rH crj 4J O H "bp ffi tf 0) •H CO o 1 U >i P CU 4J ffl P ^O O ON 00 (JN CO • . . 1 • . • 1 m co vo I oo co r-. I OO 00 ^O CO ^D CO 1 co oo cj- o m oo . • • i • . . i -vj- m ON. i rH m ON i r-t CM CM CM CM m oo vo o ^ ^ i CM OO CO 1 CO VO -sf 1 CM CN CO CN CM "^ °. "^ "^ "^ ^ ^ ^ OO^COON O OCOvD cNvocor^. -J3 cor^co CM CN 00 1 rH cn rH 1 ON in rH \| J3 ONrHinvo co miDcn rH eji »H oo in o m IH m i rH \o , OCNf-CN rH i O OO rH ON 1 in rH 1 O O O Kf CO O CO 1 CO rH r-l CM rH CM vD vO 1 -J- OO 00 1 ON O ON 1 CT\ f^ O 1 rH ,H vo co r^ rH ^o vo vj 1 tjD . • • . 1 g r- oo oo oo r- r- o 0 rH rH i— 1 CU rHCMlSrH rHCMjE-1 « S S MH rH 3 CO <0 3 O ON O rJ [•• 00 o CO <1" ffi ^- >* OJ • e co -H ,0 -p rt H d c QJ -HO CJ i — [ QJ S 0 rH 0 rC ft ft cu E < -p tt) 3 -p -p S cn cfl •H >s O ^ N ^ . . o >> ° ft rf ^ S ^ CM 0 0 S e CQ fvi !>-} O (U .. .• ' — f -t-3 ^> T^ 1 f. 1 1 -H rQ C1^1^ n5 ffl 5 S w ft a; «+J -^ s PQ p. rf ^J U r-t •^ T C cd • • PO rn ON S ^ [11(11 1 ^"^ O CQ 1 a S CU ,£ o3 O O P^ 1 ii ------- 1 i • ^ ,,„_ ,.i 1 ! H H II Pt P) I1 i ' • i • nj ,j Mo i m m-~-s t— t— 03 "I1 -p -P -p ' -P -P 01 53 5 r odd i Seg££ I i ci ft o H H i 0 0 rH 1 1 ' ro p-> H ' t-c-ikj^ ; i i i » j , ' .1 li 1 E 1 i •"O . ' cO 1 i rH CO O O rH ------- l~ 1 1 1 1 1 L_ 1 1 I LZ 0 ro o o •• -y .... d t— ^- O ^> -P -P -P !3 U W O O -H OJ 0) -p -p 01 pi 3 0) Q H H i fi • • pi tn 1 a v a w o 1 1 n 1 « rH 1 1 I w \| 1 « 0 1 ------- O O o OJ a, o •H -p •H Tl 01 ------- t- [-- ro O O CM H papuadsng J3 O •H •P 03 O •H H O bO 0 O ^ •H H O CO 0) 03 O U 110 ------- o s o o ~H g ------- 01 rH P< s •H at ------- 00 CM d O o ------- 1 1 ro I CM H f CO / O 0 r! i o' 1 £ i .p -p 1 -p d d O O -H 0) 0) .p -p W ^J 2 O H rH 0 fi i'rt vl • • O M W di P< 0 oo ro <-i < M [_ — S 1 i i i U 01 , \| •H CO £ "^ 1 1 1 •H ( t) \| «H -P+ \| ^H pq L.,. 3 w ' c5 a" i " ° •H 0 H P( m o i i pp w <; w co « ------- ft Ferrous Sulfate Hercofloc 812.3 G. With Hercofloc •d CO -p to 3 o Tl (H < -P JH Vi - 3 m o a o a o < o BOD Tota Soluble CH (U (U H H ft ft to to s ^ ^ o ft 0} 00 o o t— m oo —- t— t— ID -p -p -p C C o o -H a) ------- o o -p ft. O H O O kOQr- 300 - 200- 100 - F. Sulfuric Acid Mosiuee pH Adjusted 6 - Uc — — H. Ferrous Sulfate with Hercofloc 812.3 Without G. With Hercofloc Polymer 812.3 812. 960 — M/ fh flf ABABAB AB ABA 3 — B 1 2 K 1 2 K 77th 78th — fcl — ABABAB ABABAB 1 2 K 1 2 K 79th 80th H 300 If M tf •H ra 200 •a ------- 1O\| O O ""> COCO CO ON CO O ON CO ON rH CO ON r. r- I co I-H ON ror^-J CO i-llTl CMi-t^O M ,-t oo r- CM r-i\| co o o! m vO • -* ON rH 1 CM COi F-^VnJ CM rH miOCol CNCO CO O\U-CM\| COrHt-H O>ON«O o r. -* o oo oo r^. r- o r- o en v % £ ^o^ I-n ^str. rHi-Hr-4 CMCMCM r-< (MCM CM CMCMCM i-HCMCM r-lfOCO 3 01 O O*00 f-rHO CMCOOO "^ CM r-H ON ONOOON -JiHrH OO CMOSin rHlTlt-l kO^O<-< W .,_... ... _ or^OO irip^ 00 O p ^H ol r^ecenl oo o\oo ml -• - - ' p £ vo 'tf o r-\| CM co col r. CM -\| co CM ol o n •iQ I-H _ r_ , . CM CO CM I CM CO CM \| i-H CM i-<\| fl CO CO\| CO CO CO CO' CM CO m\| CM CO < %o co vj- CM CN CM! en o in CM r. oo CM! SJ ?n i-i\| ff 117 ------- m o o o o~i m cMvocsi CM cn »M i-«o\en cMinc rH in •—i CMCMI-H rH CM CM (-H PJ rHinen in \c en -5 in id r-lCM CMi-HI CMCM mCM mCMi-H CMCnCM CMCO °'~ ' 33 at H ot" in CM" W en cn io '- co in r-." r^- » \o \o vo g t-i « JL, o-u **>? ^^y? c'OiD 5K?°°. ^5^? r^r?^ ®r^? 4- £ co^o* coi->£ CMO\O oo r- co - oo TH r-iocn -4- -± m H ^ co ^o co o i oor-jrH f^-**^ cspo ^rrrrH COOCM PP0 COCMI^ 1 a Q COOCM CMO»r-i r^O^O CO-^O Tl col co co ml .... _> in en] o CMO> »no> I-H\O _._ :•. i ^^ i ^ ------- O OJ sr ^ O. O oo r-i i-»cMi-«! CMCMCM! CMCMCM co en m m r-* m I \o r- oo cc r- r» r-l CM ;£ I-H ( fl K ^! O. CO rH CC O O C m nn 3 ' t-^- t^- . .. -H O •-> t~- t— S s •s i i i > ; (0 J3 119 ------- O %£J t-l U-l r-l • (0 p< O •a o o rH a> « on ro r-< L< i> H o 4^ t- r- S s-s i i i w S,^ nj s p o 120 ------- oj l-l "i 0 ° \| I S § i 0 1 EH CO 1 M 3 O 1 0 * i s i** s \| M /-> i rH C 8 8 ,3 O U fJ M i CD * rH O 3 CO rH O rH M g § pa CO 3 CO i g rH O CJ rH ft •§ tH O CO 1-4 5 o H r-4 O CO J 3 CO s rj s o \|H O CO !_, 4J & d f X a * § o A 3 o CO a 4J 1 as A 3 tt JJ •H i u o P^ . . 00 CO 00 O rH O , O 1* O If) rH CM ll 1 ^f CO rH 1 i-t 00 O ' CO CO 1 rH CO rH CM CO O \O t^- in CM rH in mmr^ m^nr mo m\oor** rH rH rH oo incoaoi o m m t co co 1 co oo oo I O in rH O 1 OrHinl CMvDl rH \f> vO 1 CM CM CO h >^\| -4" rH O\ ml \O O OOl \O CO V> rHl CM CMCMCMCMl CMCNCMCMj rHCM rH\| rH CM CM CM,\| •d Oh CMCMOOMfl r>.vOr^.O\| CN CO ml OCMrH^I rH COCO^O^I rH CO VD -J CM %£> s* ^T^COxol CO g, CO CO CO CO\| CO CO CO -H in [ irtr»u>< oo r— i so >n oo I rH CM j: ^ScMjE rSrHCMX rHCMX rHCMC^S O rH CM CO ^ m m ft m »n CO a> rH m c § v3 C 01 •rt 41 H u &4 3 M O !H Vi • 4J O fl a o. S H 1 d m w o P* h 11 0) fl, > -P m Qt oJ r-( QJ O O -H > W -P +> U O S> 0) Q, 0 <3 O •rt O O H 1> m m on H o h 0) •• •• -d O o +3 1— t— B H O -H . . . M n n 1 1 1 MO p> ft) pt H W 33 > •a a d CO «) So E da o 121 ------- to e 0 o CO 1 § EH O §«„ <2 WJ EH LfN M r- 0 o ro o • J OJ i—) Q CO H 5 0 o o o o < a §J-3 PH CO M i o I o w EH G O •H U •o ^ P» CM II rH rH rH ON i-H -* CM P^ st — 1 CM P- o oo CM m so o m co co m Pi Pi C mm O 'o ri-J g g [<1 > K ft o! H cd cd ,0 O O C en on of d,, .... Oj -P O O -r4 OJ O H a a 3 oj • . o > a & o •t) O O r-l 4) tn m OO H 1* 01 •• •• 'H o +J t— t— S CO 'in 1 1 1 JH O ID p,H C\J S TJ & a o 122 ------- a o ta o u i-i i-f o c. l? \|^ CM 00 rH \| CM CO 1 1 vo co m O O CO rH 1 1 •4- co m r- ON co CO CM o r- CO CM I CM VO CM CM vO CO VO O CM in in ON rH vO CO rH vO 1 CO CO 1 1 CM CM CO ^ ^ rH 1 1 1 rH CO CO CO vD ** CO CM CO vD CO 1 CO OO O 00 CM CM CN CM o m CM O r» co CM CM CM vO CM st O O CO CM CO CM ON o m vD vD --d" in CM CM CO - CN ON t P^ rH 1 i-t CM CO rH O in \o vo CM CM CM rH CO CM CO r-0 i 1 CO O sl- ON O CM rH m co I t 1 oo r-» o vO vo O CO vD Sf P^ P- CM -Ct 1 ON ON 3CN CM CO ui rH rH ON O O CO CO ON rH 00 00 0 1 r-t m i in 1 in 3 j ON r-« S CO vO Ul ft 1 to -p H 0) H C •H ° i V rH bD rf w m o ^ Q) V fa QJ H H H > ft ft O T3 T3 6 B N at • • ^» Ul ft OJ H cd o3 ,0 O O C m m 3 0) .. •• -. f-t- to cd -P O O -H to .p +3 to 1 sal > cd ft o T) O O iH ------- i ; . i : i : i ' i : i : i : i : i : i i : i : i i : i ; i i : _«j jjj atctnres ofi — ail ! "• II M- 1 : Ml W W II 7 oS H s R " a i • • o CB r^ — i tf Pt O V tl \| \| OOH H A \|\| 11111 1 ! r-l CM fc, 0 O \| 1 i : 1 i : i •P H £ M (~\| I 1 I 1 ' - 1 i i : i ; i ; i i i i i i i i 8 °8°§ooc>0 Q Q Q CJ ft. u Q rH O O a w t- U H a Q -» 0 a o o °H O a ft. o a H u 0 o O H o H O a 0 a H 0 .« § rH O ft Figure C-1. Total and soluble BODs in base-line study (no additives) at Locks Mill. 5(103 124 ------- m « • a e e * P, oJ O ro O O •• ro m t— So H -fi 01 s al t-l- 3C I I I i-t OJ fr, ogf r r r a 4> V( O H O -H cks Mill at no additives t -p CQ ------- 1 H i : i : i : i ; i ; i ; i ; i : s \ ; 2 i : w \| i i : C - Influent D - Effluent 1 0) V H H If. m 00" t— t— a •p O O -rf 4> -p m a a 0 d Pi o com M • - •• 'i-t t~ t-se i i i M CM fc i ; i j i : I ! i ; i : i : i ; i f i : I : p 1 ! i ; i : i : i i ; i : i : i j i : i j ( ; ( o ITS m i ; i : 1 I 1 1 1 1 Q cu m o m Q o Q CU CU O PO Q H O P (K O Q CM H o fi a 0 n 0 Q CM 0 O CO Q 0 a o Q CM O"v U OJ r-l Q fc O O CM ------- f ' - 1 1 1 1 1 1 1 1 1 J J _. 1 p< Pi t ' « « r5 I a a a 1 p. oj o CO 0 O •• rrl ro (— ^2 1 1 +J +> 01 1 , . .a i sag 4> ^> fJ d p, o u a) t — I 0 0 rH rJ ~l 1 III II H CM I* O Cl 1 , 1 1 1 1 1 1 Q CM ro a H u a EX 0 ... CM CM Q a Q H a rH a Ex a Q H Q Q CM O\ a o a 0s* o °rH O V Vi -H rH O -P 1 ft BJ 1 o o o o 3 S" S o O O 0 CQ Q O, O ro CM rH O CO c r i P L 1 r d i C r c r , i 1 r 1 r r~ i i i i o o o o o o o VO -* OJ Q cu ro u (") Q rH Q Q a rH CJ Q fc. a H o Q CM ON Q rH a CM ro a o d. a ------- 1+00 350 300 250 D •H \|> 200 U) Q S 150 100 50 B°DS i - 7:30 a.m. to 7: 30 p.m. sample n2 - 7:30 p.m. to 7-'30 a.m. sample Total F _ J^-Q composite 7:30 a.m. Soluble C - Influent D - Effluent __ — p- 1 •- -- — — -- .M — ,— i . _ ._ r- — .- "~ .. •- 7 ~- -- — .. ~ — t- 1— CDCDCD CDCDCD CDCDCD CDCDCD CDCDCD Sample 1 2, F 12F 12F 1 2 F 12F Day of 36 37 38 39 10 Operation Figure C-5. Effect of ferric chloride (75 mg Fe3+/liter) + Nalcolyte T3C32 (l mg/liter) on BOD5 at Locks Mill. 128 ------- sea "8 C c c 1— c •** c r , 8 8 8 § CO " - CO 0) •P •rl H A H o ** H o -* CVJ CO 0 O •-" m : D c D 2 F 9 Falcolyte r °H + ?H H V H -P -H 0 -H S t- H 3 D C D C D < 2 F 1 36 37 Df ferric chloride (75 mg F r and suspended solids at L o « -P H o 00 0) O g ------- 350 300 250 ------- 300 01 -p •H O o O O 200 100 0 1 - 7=30 a.m. to 7=30 p.m. sample 2-7:30 p.m. to 7=30 a.m. sample F - Mill composite 7:30 a.m. C - Influent D - Effluent 1h it CDCDCD Sample 1 2 F Day of kk Operation CDCDCD 12F CDCDCD 12F CDCDCD 1 2 F 17 8001— 600 -P •H en •d o CQ -s C! koo 20° 0 n n CDCDCD 1 2 F C D C D CD 1 2 F C D C D CD 1 2 F CDCDCD Sample 1 .2 ™ Da7 of kh l\|s U6 ^7 Operation Figure C-8. Effect of ferric sulfate (100 mg Fe3+/liter) on color and suspended solids at Locks Mill. 131 ------- Q 2 s 9 2 ° CO f-1 W i ; ~T CU » ( 1 H H ~" to to • I & a a d PI it <~> , . , , ro I o o •• 1 (M <•> C— f— [~— (U -P C, 0 .J •p -p to \| Q 1 • • Ui «^. a a a -p -p • • o a a a) Pi u 0)0) 0 0 H H £ ^ t^ g 5 W III II [• 1 \ i: i i i i i i : j i: 1 i: i : i : i : i : i; i i : 1 i i i : i i i 00 0 l/\ O ITv oo n c\j i i i P P R O P O P O p p o p o R O O 0 p p o p o o • o o o o O L^ o ir\ (M iH H • > H H CM- VH IA g to o 3 •8 fa in „§ "^ E! 0) -P •H i-t - 1? O fa -P a CVJ H ^ i/\ H 1 0 AJ 0 § m ------- 100 300 200 100 0 C D Sample 1 Day of 50 Operation 1 - 7=30 a.m. to 7:30 p.m. sample 2 - 7:30 p.m. to 7:30 a.m. sample F - Mill composite 7:30 a.m. C - Influent D - Effluent 1 C D C D C D 1 2 F 51 C D 1 C D C D 2 F 52 C D C D C D 1 2 F 53 C D C D C D 1 2 F 8001~ 5} 6oo - r-l •3 3 uoo o CQ Tl 3 C 01 p< to 3> 200 0 _ ~~ n n n n - i -, n ^^ -i n CD D CDCDCD CDCDCD CDCDCD CDCDCD Sample 1 F 12F 12F 1 2 F 12F Day of 50 51 52 53 5k Operation Figure C-10. Effect of alum (300 rag/liter) on color and suspended solids at Locks Mill. 133 ------- 100 350 300 250 t, ------- c a « H •-» p- Pk 1 3 f ' o o ro no t— t— 0 O I o5 ft to «J p 0 0 H &0 p 1 1 1 H CM [x, 1 1 o o o o CO CM _ ^ 1 1 f 1 8 0 p « O p r> P o o P o p p c c 1 c c „ vi-S 1 : : H O +> P, «! i^^ i i i i o o o o o o O O O 0 CD VO -^ CM °r<. o t- o Q OJ CM O P P Q U P P &H 0 P CM C— $ C L O H 0 O a 0 0) -p •H H CO o 3 m pepusdsng •H 135 ------- § 3 I o 3 H on I! 68 cks at gravity sedimentation in second base line d soluble BOD C - Influe D - Efflue sag * 0 0 0 ° on m ° ^ t— [j »— i 0 0 '£) -p -p * . . 1 2 0 Sample Operation Total OO H I O OJ tjQ •H P^ 'saoa 136 ------- Influent Effluent a.m. to 7:30 p.m. p.m. to 7:30 a.m. composite 7:30 1 - 7: 2 - 7: F - Mi 5g t& c i c r r C r i c c r C [ i i i i o o o o o o o o o co \o ^t c\j p (^ p fe o p OJ Q p ft* o Q P rH O °c. p H P Ph O Q O OJ p rH CO Day of 59 60 61 65 66 67 68 Operation Figure C-lU. Color and suspended solids in second base line (gravity sedimentation) at Locks 137 ------- TECHNICAL REPORT DATA (Please read Instructions on the reverse before completing) 1. REPORT NO. 2. EPA-600/2-76-221 4. TITLE AND SUBTITLE REMOVAL OF SOLUBLE BOD IN PRIMARY CLARIFIERS 7. AUTHOR(S) George A. Dubey, Averill J. Wiley, John W. Collins 9. PERFORMING ORGANIZATION NAME AND ADDRESS Effluent Processes Group The Institute of Paper Chemistry Appleton, Wisconsin 5^*911 12. SPONSORING AGENCY NAME AND ADDRESS Industrial Environmental Research Laboratory Office of Research and Development U.S. Environmental Protection Agency Cincinnati, Ohio ^5268 3. RECIPIENT'S ACCESSION-NO. 5. REPORT DATE September 19?6 (issuing Date) 6. PERFORMING ORGANIZATION CODE 8. PERFORMING ORGANIZATION REPORT NO. 10. PROGRAM ELEMENT NO. 1BB610 11. CONTRACT/GRANT NO. R-803119 13. TYPE OF REPORT AND PERIOD COVERED Final 7/20/7U-9/30/75 14. SPONSORING AGENCY CODE EPA-ORD 15. SUPPLEMENTARY NOTES 16. ABSTRACT This project was directed to evaluating means for increasing BOD removal from pri- mary treatment systems treating pulp and paper wastes. An improved understanding of the optimal conditions for soluble and colloidal BOD removal should permit increasing efficiency in total organic removal. The initial phase surveyed 12 mills to obtain data on total and soluble BOD^, COD, suspended solids and color removal from sedimentation systems. This data was used to select mill effluents for additional study. These laboratory studies showed that, with proper flocculating agents, soluble BOD removal could be markedly increased. Soluble BOD was defined as that organic load passing through a 0.^5 Um filter. Measurement of sludge volumes and sludge dewatering characteristics produced from use of iron salts, alum and polymers was outside the objective of the study. Gel chromatography studies showed that low molecular weight biodegradable residues and colloidal materials were flocculated and removed. Studies with model compounds indicated that increased removal is apparently related to pH of the solution and to functional groups, chain length, branching and solubility of the compound. Chemical costs may range from 3$ to 10$ per 1000 gallons. Cost reduction and improved clarifier performance can be achieved by elimination of overflows and spills within the mill that are antagonistic to efficient sedimentation due to dispersant action. 17. KEY WORDS AND DOCUMENT ANALYSIS a. DESCRIPTORS Precipitation (chemistry) Coagulation Flocculating Oxygen demand 18. DISTRIBUTION STATEMENT Release to Public b.lDENTIFIERS/OPEN ENDED TERMS Primary treatment Water pollution control Pulp waste treatment Paper waste treatment Chemical degradation 19. SECURITY CLASS (This Report) UNCLASSIFIED 20. SECURITY CLASS (This page) UNCLASSIFIED c. COS AT l Field/Group 13B 21. NO. OF PAGES 152 22. PRICE EPA Form 2220-1 (9-73) 138 : 1976 — 657-695/6129 Region 5-11 ------- Z TJ m £ 0 -------


                                          EPA-60Q/2-76-221
                                          September 1976
               REMOVAL OF SOLUBLE BOD

                IN PRIMARY CLARIFIERS
                   George A. Dubey
                  Averill J. Wiley
                   John W.•Collins
              Effluent Processes Group
          The Institute of Paper Chemistry
             Appleton, Wisconsin
                Grant Number R-803119
                   Project  Officer

                   Ralph H. Scott
             Food & Wood Products Branch
Industrial Environmental Research Laboratory-Cincinnati
              Corvallis, Oregon  97330
     INDUSTRIAL ENVIRONMENTAL RESEARCH LABORATORY
         OFFICE OF RESEARCH AND DEVELOPMENT
        U.S. ENVIRONMENTAL PROTECTION AGENCY
               CINCINNATI, OHIO
                  LIBRARY
                    S. ENVIfi...,.,i..-.,;\L P
                    - '«  \  .  08817

-------
                                 DISCLAIMER
     This report has been reviewed "by the Industrial Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for publica-
tion.  Approval does not signify that the contents necessarily reflect the
vievs and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or re-
commendation for use.
                                      ii

-------
                          FOREWORD
       When energy and material resources are extracted, processed,
converted, and used, the related pollutional impacts on our environ-
ment and even on our health often require that new and increasingly
more efficient pollution control methods be used.  The Industrial
Environmental Research Laboratory - Cincinnati (lERL-Ci) assists
in developing and demonstrating new and improved methodologies that
will meet these needs both efficiently and economically.

       Removal of soluble BODc in primary clarifiers presents
research findings related to means and methods for improving the
efficiency of primary sedimentation systems treating pulp and paper
wastes.  The possibility of greater BOD removal was investigated by
examining effects of flocculants and coagulants in removing soluble
and colloidal constituents of such wastes.  The report also discloses
that certain process waste streams may be antagonistic to efficient
sedimentation and should be treated separately from total mill wastes.
This information will be of interest to segments of the industry
wishing to upgrade primary treatment efficiency to reduce loadings on
secondary treatment.  The findings are also applicable to treatment of
waste streams not containing significant additives, as the tissue in-
dustry, wherein effluent guide-line allowances may possibly be satis-
fied by efficient primary treatment.  For further information contact
the Food and Wood Products Branch, Industrial Environmental Research
Laboratory-Cincinnati.
                                     David G. Stephan
                                         Director
                       Industrial Environmental Research Laboratory
                                        Cincinnati
                                  111

-------
                                    ABSTRACT
     Primary clarifiers used in the pulp and paper  industry show wide varia-
tions in BOD removal.  This project was directed to evaluating means  for in-
creasing BOD removal and developing an understanding of the mechanism.  Such
understanding should permit optimization of BOD removal during primary clarifi-
cation .

     A survey involved twelve mills, nine with primary clarifiers, to obtain
data on total and soluble BODs, COD, suspended solids, and color.  These data
were used to select mill effluents for additional study.

     Laboratory studies showed that, with the proper flocculating agent and
operating conditions, the BOD5 concentration of some mill effluents could be
markedly reduced during clarification.  These findings were confirmed by pilot-
scale clarifier field trials at two mills.

     Measurement of sludge volumes and sludge dewatering characteristics re-
sulting from chemical treatment were outside the objectives of this study.

     Gel chromatography studies showed that considerable low molecular weight
biodegradable residues along with colloidal material were also flocculated and
removed.  Studies with model compounds indicated that increased removal is
apparently related to the pH of the solution and to the functional groups (hy-
droxyl, phenolic, and carboxylic), chain length, branching, and solubility of
the compound.

     Chemical costs may range from 3<£ to 10$ per 1000 gallons of feed to the
clarifier.  Decreased cost and increased operating efficiency in total treat-
ment systems, as a result of improved clarification, could more than justify
this chemical charge.  Minimum levels of chemical costs will depend upon mini-
mizing or eliminating overflows or spills of colloids and dispersants detri-
mental to clarifier efficiency.

     This report was submitted in fulfillment of Project No. R803-119 by The
Institute of Paper Chemistry Effluent Processes Group under the partial
sponsorship of the Environmental Protection Agency.  Work was completed as
of August, 1975.
                                        IV

-------
                                  CONTENTS

                                                                        Page



Abstract                                                                 iv

List of Figures                                                          vi

List of Tables                                                            x

Acknowledgments                                                         xi-v

I       Introduction                                                      1

II      Summary and Conclusions                                           3

III     Recommendations                                                   5

IV      Mill Surveys                                                      6

V       Laboratory Study of Processes for Higher Levels
        of Soluble BOD5 Removal                                          20

VI      Pilot Scale Clarifier Studies                                    1;9

VII     Jar Tests with Individual Sewer Discharges                       67

VIII    Discussion of the Mechanisms of Soluble BODs Removal             77

IX      Areas for Cost Reductions                                        96

X       References                                                       98

XI      Glossary                                                        100

XII     Appendices                                                      101

-------
                                   FIGURES

Number                                                                  Page

   1    Schematic for processing clarifier samples for soluble
           BOD5 study ........................    11

   2    The separation of insoluble from soluble BODs by filters
           of various pore sizes ...................    13

   3    Total biochemical oxygen demand in clarifier influent and
           effluent samples .....................    l6

   U    Total chemical oxygen demand in clarifier influent and
           effluent samples .....................    IT

   5    Suspended solids in clarifier influent and effluent
           collected on a 0.^5 ym filter ...............    18

   6    Suspended solids and BODs in various concentrations of
           influents treated with Fe3  or alum ............    33

   7    In jar tests with Mosinee mill effluent, pH 8.5s ferric
           chloride and ferric sulfate at 25 and 50 mg/1 of Fe3+
           (aj , (b_) , (
-------
15    Jar tests with Combined Locks individual sewer effluents.  . .     7^

l6    Jar tests on wastes from individual sewers in the Combined
         Locks mill	     75

17    Jar tests on special composites of Combined Locks individual
         waste streams	     76

18    TOG and molecular weight distribution of gel-chromatographed
         chemimechanical mill effluent before and after coagulation
         with 70 mg/1 Fe3+ as ferric chloride	     78

19    Phenol sulfuric acid test for aldoses and molecular weight
         determination of gel-chromatographed chemimechanical mill
         effluent before and after coagulation with 10 mg/1 Walco
         73C32	     79

20    Ultraviolet light transmittance profile and molecular weight
         distribution of a gel-chromatographed solution of sodium
         gluconate and ferrous ammonium sulfate, pH 2 	     8l

21    Transmittance and molecular weight distribution of a gel-
         chromatographed solution of sodium gluconate and ferrous
         ammonium sulfate, pH 7-0	     82

22    Phenol sulfuric acid test for aldoses and molecular weight
         distribution of gel-chromatographed chemimechanical pulp
         mill effluents	     83

23    Effect of pulp fibers on removal of soluble compounds ....     86

2ka.   Effect of kraft pulp fibers at 350 mg/1 on removal of
         soluble compounds with 100 mg/1 of Fe3+ (as FeCls) at
         various pH's (acetic acid, methyl acetate, ethyl acetate,
         oxalic acid, lactic acid, and methyl lactate)	     88

      Effect of kraft pulp fibers at 350 mg/1 on removal of
         soluble compounds with 100 mg/1 of Fe3  (as FeCls) at
         various pH's (raffinose, hexanedioic acid, 2-pentanol,
         and tertiary amyl alcohol)	     89

      Effect of kraft fibers at 350 mg/1 on removal of soluble
         compounds with 100 mg/1 of Fe3+ (as FeCls) at various pH's
         (trans-butenedioic acid, cis-butenedioic acid, propanedioic
         acid, hydroxymalonic acid, benzylmalonic acid, and 1,U-
         butanedioic acid)	     90

      Effect of kraft fibers at 350 mg/1 on removal of soluble
         compounds with 100 mg/1 of Fe3+ (as FeCls) at various pH's
         (benzoic acid, p_-hydroxybenzoic acid, o_-hydroxybenzoic
         acid, p_-methylbenzoic acid, 3,^,5-trihydroxybenzoic acid,
         and 3-methoxy-U-hydroxy-benzaldehyde)	     91
                                   vii
-------
2Ue   Effect of kraft fibers at 350 mg/1 on removal of soluble
         compounds with 100 mg/1 of Fe3+ (as FeCls) at various pH's
         (m-dihydroxy-benzene ,  o-dihydroxy-benzene, p_-hydroxy
         cinnamic acid, ^-hydroxy-3-methoxy cinnamic acid,
         D-galacturonic acid, and polygalacturonic acid) ......    92

25    The model compounds shown in Figures 23-2ke cluster in two
         groups on the basis of solubility in water and removability
         with Fe3+ .........................    95

B-l   BODs reduction during clarification with Hercofloc ......   107

B-2   Color and suspended solids reduction during clarification
         with Hercofloc ......................   108

B-3   BODs reduction during clarification with alum ........   109

B-4   Color and suspended solids reduction during clarification
         with alum .........................   110

B-5   BODs reduction during clarification with ferric chloride. .  .   m

B-6   Color and suspended solids reduction during clarification
         with ferric chloride ...................   112

B-7   BODs reduction during clarification with ferric sulfate  . .  .   113

B-8   Color and suspended solids reduction during clarification
         with ferric sulfate ....................   Ill;

B-9   BODs reduction during clarification with sulfuric acid,
         Hercofloc, or ferrous sulfate plus Hercofloc  .......   115

B-10  Color and suspended solids reduction during clarification
         with sulfuric acid, Hercofloc, or ferrous sulfate plus
         Hercofloc .........................   116

C-l   Total and soluble BODs in base-line study (no additives)
         at Locks Mill .......................
C-2   Color and suspended solids in base-line study (no
         additives ) at Locks Mill .................   125

C-3   Effect of ferric chloride — 75 mg Fe3+ (without polymer)
         on BOD5 at Locks Mill ...................   126

C-k   Effect of ferric chloride — 75 mg Fe3+ (without polymer)
         on color and suspended solids at Locks Mill ........   127

C-5   Effect of ferric chloride (75 mg Fe3+/liter) + Nalcolyte
         73C32 (l ing/liter) on BODs at Locks Mill  .........   128
                                  Vlll
-------
C-6   Effect of ferric chloride (75 mg Fe3+/liter)  + Nalcolyte
         T3C32 (l mg/liter) on color and suspended solids at
         Locks Mill	    129

C-7   Effect of ferric sulfate (100 mg Fe3+/liter)  on BOD5 at
         Locks Mill	    130

C-8   Effect of ferric sulfate (100 mg Fe3+/liter)  on color
         and suspended solids at Locks Mill	    131

C-9   Effect of alum (300 mg/liter) on BOD5 at Locks Mill	    132

C-10  Effect of alum (300 mg/liter) on color and suspended solids
         at Locks Mill	    133

C-ll  Effect of alum (300 mg/liter) + Hereofloc 812.3 (0.75 mg/liter)
         on BOD5 at Locks Mill	    13k

C-12  Effect of alum (300 mg/liter) + Hereofloc 8l2.3 (0-75 mg/liter)
         on color and suspended solids at Locks Mill	    135

C-13  Total and soluble BODs in second base line (gravity
         sedimentation) at Locks Mill	    136

C-lh  Color and suspended solids in second base line (gravity
         sedimentation) at Locks Mill	    137
                                   IX
-------
                                   TABLES

Number                                                                  Page

   1    Description of Mills Surveyed 	     7

   2    Analysis of Filtrates Prepared with and without Filter Aid.  .     9

   3    Comparison of Ultrafiltrates from a 100-A Membrane with
           Filtrates Prepared with Filter Aid over 0.45 and 0.10
           Urn Filters	     9

   k    Analytical Data for Treated and Untreated Samples During
           Evaluation of Method for "Soluble" BOD5. .	    12

   5    BOD5 in Clarifier Influent and Effluent Samples ("As Received"
           and Filtered Through a 0.^5 urn Filter)	    1.^

   6    COD in Clarifier Influent and Effluent Samples ("As Received"
           and Filtered Through a 0.^5 ym Filter)	    ik

   7    Suspended Solids in Clarifier Influent and Effluent Samples
           Removed by a 0.1*5 Vtm Filter	    15

   8    The Variability of pH, BODs and COD in Clarifier Influent
           and Effluent Samples from Different Sources	    19

   9    Screening Trials of Flocculants and Polymers by Analysis of
           Soluble Organic Carbon and Soluble BOD5	    22

  10    COD and BODs of Various Polymers used as Floceulation Aids.  .    23

  11    Comparison of Phenol Sulfuric Acid Test with BODs Test to
           Evaluate Flocculants in Test Tube Tests	    2k

  12    Comparison of Phenol Sulfuric Acid Test with BODs Test to
           Evaluate Flocculants in Jar Tests	    25

  13    A Comparison of Zeta Potential, Soluble COD and BODs, and
           Floe Form in Jar Tests of Flocculants	    26
  ik    Jar Tests of Primary Flocculants with Mill A and Mill D
           Influents and Effluents	    28

  15    Jar Tests of Primary Flocculants with Mill F Total Clarifier
           Influent	    29

  16    Jar Tests of Polymers and Additives with Mill F Total Clarifier
           Influent	    30
-------
IT    Jar Tests of the Effect of Activated Carbon and Various
         Additives on Soluble COD and BOD5 in Mill F Clarifier

18
19
20
21
22
23
2k
25
26
27
28
29
30
31
32
33
34
Influent 	
Jar Tests of the Effect of Activated Carbon and Additives on
COD and BODs of Mill F Chemimechanical Pulping Effluent. .
Jar Tests of the Effect of Flocculants on BODs, COD, Organic
Carbon, and Color in Mosinee Clarifier Influent 	
Jar Tests of FeCls, Alum, and Additives as Flocculants for
Jar Tests of the Effects of Primary Flocculants and Polymers
on pH and Clarity of Mosinee Clarifier Influents 	
Jar Tests of the Effect of Flocculants on pH, BODs, COD, and
Color of Effluents from Mosinee Clarifiers ...... 4 .
Effect of pH on Jar Test Flocculation of Mosinee Clarifier
Influent 	
Volume of Sludge from Mosinee Clarifier Influents Treated with
Various Flocculants and Polymer 	
Removal of BOD5 from Effluents of Combined Locks Mill by
Sources of Soluble and Total COD and BODs in Effluents of
Mill F 	
Chemical, Biological, and Physical Analyses of Various
Process Streams within the Mosinee Mill (Mill K) 	
Release of Solids Soluble COD and BODs ^y Beating 	
Jar Tests of Various Flocculation Treatments of Filtered
90— Minute Beater Samples •, 	 . 	
Soluble COD and BODs in Mill F Effluents Treated with
Flocculants and Reverse Osmosis or Ultrafiltration ....
Laboratory Evaluation of Pilot— Scale Clarifier 	
Analysis of Individual Samples to Establish Range 	
Analysis of Special and Regular Samples of Clarifier Influent,
Trial No .1 	
Analysis of Special and Regular Samples of Clarifier Influent,
Trial No. 2 	
31
32
3S
36
37
38
3Q
41
4?
43
44
4s
46
47
s?
S3
SS
s6
                                    XI
-------
35    Various Treatments of Mosinee Mill Effluent in Pilot-Scale
         and Mill Clarif iers ....................    60

36    Analysis of Special and Regular Samples of the Clarifier
         Influent of the Combined Locks Mill ............    6k-

37    Clarifier Study — Combined Locks Paper Corporation Average
         Effluent Quality and Removal Values for IPC and Mill
         Clarifiers ........................    66

38    BOD of Individual Effluent Streams and Total Mill Effluent
         of Mosinee Mill ......................    TO

39    Removal of Different Molecular Weight Fractions of Organic
         Residues by Flocculation .................    80

ho    Removal of Carbohydrates and Weak Acids from Mosinee and
         Combined Locks Waste Streams by Pilot-Scale Clarification.    85

Ul    TOG Removal from Solutions of Model Compounds Treated in Jar
         Tests with Aluminum Sulfate ................    87

U2    Solubility in Water of Various Model Compounds and Their
         Removal in Jar Tests with Fe3+ and Pulp Fibers
U3    Chemical Costs for Clarification ...............    97

B-l   Clarifier Studies with Hereof loc 812.3 (0.75 mg/l) at
      Mosinee Paper Corporation ..................   102

B-2   Clarifier Study with Alum (200 mg/l) plus Hereof loc 812.3
         (0.75 mg/l) at Mosinee Paper Corporation  .........   103

B-3   Clarifier Study with Ferric Chloride (100 mg/l Fe3+ Plus
         Hereof loc 812.3 (0.75 mg/l) at Mosinee Paper Corporation  .   10U

E-k   Clarifier Study with Ferric Sulfate and Hercofloc at
         Mosinee Paper Corporation .................   105

B-5   Clarifier Study with Sulfuric Acid, Sulfuric Acid Plus
         Polymer, and Ferrous Sulfate Plus Polymer at Mosinee
         Paper Corporation .....................   106

C-l   Clarifier Study at Combined Locks — Base Line (Gravity
         Clarification after 0.010 Inch Sieve) Without Additives  .  .   117

C-2   Clarifier Study at Combined Locks — Ferric Chloride
         (75 mg/l Fe3+) Without Polymer  ..............   118

C-3   Clarifier Study at Combined Locks — Ferric Chloride
         (75 mg/l Fe3+) Plus Nalcolyte 73C32  (0.75 mg/l) ......   119
                                   XII
-------
C-U   Clarifier Study at Combined Locks — Ferric Sulfate
         (100 mg/1 Fe3+)	    120

C-5   Clarifier Study at Combined Locks —Alum (300 mg/l)	    121

C-6   Clarifier Study at Combined Locks — with and Without Alum
         (300 mg/l) Plus Hercofloc 812.3 (0-75 mg/l)	    122

C-7   Clarifier Study at Combined Locks — Second Base Line
         (Gravity Sedimentation After 0.010 Inch Sieve) Without
         Additives	    123
                                   Kill
-------
                              ACKNOWLEDGEMENTS
     The original objectives of this project in terms of evaluating the
degree of BOD removal attained in primary clarification of pulp and paper
effluent waters were conceived by Dr. Everett Cann of the Wisconsin State
Department of Natural Resources.  The project was organized and conducted
by the staff of The Institute of Paper Chemistry.  George Dubey, Project
Supervisor, had responsibility for detailing and conducting the laboratory
and field studies and for reporting the results of this research in coopera-
tion with Averill Wiley, Project Director, and Hardev Dugal, Director of the
Institute's Division of Industrial and Environmental Systems.

     Gerald Hovind, Arthur Webb, and Gay Fraundorf were principally involved
in the extensive analytical program.  The field studies were greatly facili-
tated by help from John Baumann who installed and operated The Institute of
Paper Chemistry pilot equipment, by James Albrecht and Al Kleist of the
Mosinee Paper Corporation, and by Lawrence Weyenberg and James Hoovers of
the Combined Locks plant at National Cash Register.

     Ralph H. Scott, Chief of Pulp, Paper and Wood Staff, Food and Wood
Products Branch, lERL-Ci, EPA, as Project Officer, has had responsibilities
for the immediate supervision of this project in accordance with the objectives
of the program.
                                      xiv
-------
                                  SECTION I

                                INTRODUCTION
     The overall degree of biochemical oxygen demand (BOD) removal actually
observed in most conventional out-plant clarifiers treating total mill ef-
fluents has been relatively small (5-15 percent) and has seldom extended to
levels that would significantly improve effluent quality.  A search of the
literature disclosed few published investigations of the mechanisms for the
BOD removal in clarifiers.  Only two (l.,2_) gave data indicating possible
sources of BOD that is removed during clarification.  The State of the Art
Review of Pulp and Paper Waste Treatment by Gehm (l_) attributes the BOD to
"the high content of suspended organics and small quantities of dissolved
organic matter" in certain types of waste and to the "fact that the oxygen
uptake rate of fiber is slower than that of dissolved materials, since it
must first be liquefied before oxidation can take place."  Das and Lomas (2_)
point to glucuronoxylans and araboglucuroxylans as water soluble hemicellu-
loses (from the beating of pulp) that reduce the BOD when removed with the
floe.  One of the first goals of Phase I of this research program was, there-
fore, to evaluate by a controlled analytical study the removal performance
for total and "soluble" BODs in different clarifiers at pulp and paper mills.
Surveys of mill waste flows and clarifier effluents were conducted at 12
mills owned by 11 companies.  The analytical data developed during this sur-
vey established the basic program of the investigation and the definitions
of the terms "soluble" and "total 5-day biochemical oxygen demand (BODs)" to
be used throughout the balance of this study.

     Three different classes of BOD-forming components in the various waste
flows were identified:

     1.  Substantial quantities of the BOD being removed in clarifiers
         are obviously suspended fibers and related solids in various
         stages of physical, chemical, and biological degradation.
         These suspended particles may originally have low levels of
         solubility and low BOD but can undergo degradation to produce
         soluble substances responsible for BOD.  This fraction com-
         prises the principal source of BOD readily removed in conven-
         tional clarifiers.

     2.  Advanced degradation of cellulose and hemicellulose particles
         and the presence of papermaking additives such as starch may
         provide a substantial amount of soluble, high molecular weight
         components with biochemical demand for oxygen.  Most of these
         kinds of substances can be surface-adsorbed and removed by
         advanced clarification procedures.  The further development
         and field testing of routes to removal of this fraction com-
         prised a principal area for development under Phase II of
         this project.
-------
     3.  A third fraction of the BOD contained in pulp and paper waste
         waters is comprised of low molecular weight degradation prod-
         ucts of carbohydrates, such as acetic acid, aldehydes  and
         alcohols, which are not readily amenable to physical adsorp-
         tion and flocculation.  Removal of these components has not
         been reported for primary treatment if we except those cases
         of prolonged holding time or other conditions where advanced
         stages of microbiological action are evident in the clarifier.

     The fundamental physicochemical mechanisms of flocculation of pulp  mill
white water fines studied by Williams (3.) and the work by Back (h) on the
nature of wood components dissolved in the pressure refining of mechanical
pulps provide information on a possible approach to the development of an
understanding of the mechanisms involved in the removal of BOD components
from waste streams.  Their ideas were used to develop the jar testing proce-
dures used in our laboratory for the optimization of the removal of BODs
constituents from various total mill effluents.

     Phase I studies were funded jointly by a group of U.S. pulp and paper
companies (Appendix A), the U.S. Environmental Protection Agency (EPA)
(Grant Bio. 803 119) s and the Wisconsin Department of Natural Resources (Ref.
No. 8100).

     For a supplementary and confirming Phase II program, small-scale pilot
equipment was used for on-site comparative studies, paralleling commercial
clarifiers operating at two mills representative of a broad segment of the
pulp and paper industry.  Additional funds were supplied by the U.S. Environ-
mental Protection Agency under an extension of the above grant and by The
Institute of Paper Chemistry.  This confirming program of study was designed
to evaluate the operating parameters developed in the earlier Phase I labo-
ratory research under mill conditions.  The first trial was conducted at an
integrated kraft mill and the second at an integrated mill producing pulp by
a recently developed chemimechanical (CM) pulping process.

     Laboratory jar tests were also used to evaluate the flocculation and
clarification characteristics of individual sewer discharges at these two
mills.  We studied the effects of processing each of these streams separate-
ly and we could also check the effect each had on the overall BODs efficien-
cy of the clarifier when processed as a part of the total mill effluent.
Additionally, the jar tests were used to test the removal of soluble model
compounds (acetic acid, aromatic dicarboxylic acids, etc.) from solution
with the various flocculating agents in order to help elucidate the mecha-
nism of BODs reduction.

     An economic evaluation of the data of two field trials and the results
of the jar tests conducted on individual streams show the savings which
might be realized if the relatively small volumes of stronger pulping efflu-
ents, containing lignin compounds with dispersant properties, were not
treated in the clarifier.
-------
                                 SECTION II

                           SUMMARY AND CONCLUSIONS
     1.  A large portion of the BOD entering the primary clarifiers process-
ing pulp and paper manufacturing waste waters are solubles that can pass
through a 0.^5 ym filter.

     2.  These components are generated from wood in pulp and papermaking
steps, including wood handling.  They may also come from papermaking addi-
tives such as starch or sizes.  These materials can be degraded within the
mill or during out-plant treatment.

     3.  Laboratory jar tests with various chemical flocculating agents and
operating parameters showed that significant quantities, ranging to 20 per-
cent or more, of the soluble BODs and color could be removed from wastes
from mills surveyed in Phase I studies.  These findings were confirmed by
pilot scale field trials at Mosinee Paper Company at Mosinee, Wisconsin, and
at the Combined Locks mill of the NCR Corporation in Combined Locks, Wiscon-
sin.

     k.  Jar tests with the individual waste streams making up the total
mill discharge have shown that certain streams contribute both high BODs
levels and chemicals that are detrimental to the efficient operation of the
clarifiers.  Elimination of these streams from the out-plant clarifiers
could provide increased quality to the mill discharge.  It would also de-
crease the load on the secondary treatment plant.  The streams that were re-
moved from the influent to the clarifier could be treated separately in small
clarifiers or by reverse osmosis, ultrafiltration, or other small unit pro-
cesses .

     5.  The mechanism for the removal of soluble BODs is very complex due to
the heterogeneity of the wastes treated in the clarifiers.  There was no ap-
parent relationship between the initial soluble BODs concentration and the
ultimate removal of the soluble BODs components by the clarification process.

     6.  Gel chromatography analysis of effluents before and after coagula-
tion with iron and pulp fines showed a significant number of low molecular
weight compounds were removed.  Model compound studies showed that many or-
ganic acids including some quite soluble could be partially coagulated in an
iron-pulp fines system as well as substances which do not form salts with
iron.

     7-  Evaluation of the benefits and costs of achieving improved perfor-
mance of clarifiers requires detailed engineering studies to fit individual
mill situations.  Such studies were beyond the scope and objectives of this
study.  However, it was apparent that total chemical costs, including the
chemical additives which may be used in existing practices at these mills,
-------
might range from 3<£ to 10$ per 1000 gallons treated.  Efficient use of these
chemicals would be balanced by benefits and cost reductions in the overall
systems of waste treatment.

     8.  Good clarifier performance and low chemical costs could only result
from adequate control of mill overflows and spills so that strong colloids
or dispersants do not enter the clarifier.
-------
                                 SECTION III

                               RECOMMENDATIONS
     These studies have shown that vith the proper selection of flocculating
agents and operating conditions, soluble BODs concentrations in waste
streams can be markedly reduced (10-50 percent) in the primary clarifier.

     Collapse of the double layer surrounding the charged molecule, bridging,
and decreased solubility after treatment with the flocculating agent are
mechanisms which could be expected to contribute to the removal of both
total and soluble BODs components, as well as color, from the effluents.

     1.  We need to know more about the constituents and characteristics of
different kinds of sludges — those from high and low levels of soluble BODs
removal — and their effect on the efficiency of clarification in mills pro-
cessing under different conditions.  This could add considerably to our
knowledge of the mechanisms involved.  The influence of individual mill
waste streams in clarifier sludge performance also needs further study.

     2.  Many of the individual sewer discharges were found to have adverse
effects on the efficiency of the clarifiers and also to have a toxic effect
on the 5-day BOD test.  Removal of these streams from the clarifier inflow
could markedly increase the efficiency of both the primary and secondary
(biological-type) treatment systems.  Diversion of these flows, which are
usually of small volume, to other treatment systems is recommended.

     3.  Selection of the streams to be "eliminated" could be undertaken
under laboratory conditions with a combination of jar and BODs tests.  Pro-
cesses for treating the eliminated streams by separate in-plant clarifica-
tion, reverse osmosis, ultrafiltration, or the training of plant personnel
to reduce the discharge of the detrimental stream would follow the identifi-
cation of these major sources of difficulty.
-------
                                 SECTION IV

                                MILL SURVEYS
EQUIPMENT AND PROCEDURES

     Weekly samples from mills using clarifiers consisted of two gallons of
the influent to the clarifier and a single gallon of the clarifier effluent.
Where possible, both samples were collected as 2^-hour composites.  Mills not
using clarifiers sent in two gallons of the effluent they were discharging
into ponds or aeration lagoons.   Table 1 describes the types of mills and
their facilities for pulping, bleaching, and effluent treatment; this table
is based on data supplied by the mill staff.

     As soon as the samples were received in the Institute laboratories at
Appleton, Wisconsin, the influent samples were composited, well mixed and
immediately transferred to one-gallon plastic bottles.  One bottle was filled
to the brim, set aside to settle for one hour, and the top 6 inches of super-
natant was withdrawn using a light vacuum and a short U-bend at the end of a
glass tube.  This permitted removal of the top layer of "clear" fluid without
drawing the "settled" material from the bottom layer.  A second bottle of the
influent and also a bottle of clarifier effluent were separately mixed and
analyzed "as received."

     Samples of the "as received" and "settled" material were pressure fil-
tered through various porosity filters.  These filters were 1.2-ym and O.U5-
um pore size Metrical membranes from Gelman Instruments of Ann Arbor, Michi-
gan and a 0.10-um plastic filter from Nuclepore of Pleasant, California.

     Suspended solids, BODsj and chemical oxygen demand (COD) were determined
on the initial samples and on all fractions.

     Samples for suspended solids were processed on 1.2- and O.U5-um filters
on a Gelman pressure funnel at 6.9 kg/cm2 (lOO psig) with NZ and dried at
103-105°C (2l8-221°F) for one hour.  For biochemical oxygen demand we used
the standard APHA Method (%), keeping the samples at 20°C (69°F) in a water
bath and determining the dissolved oxygen with a Weston and Stack probe.
The COD was determined by the dichromate oxidation method of APHA (6_) . Color
was determined at pH 7-6 using the National Council for Air and Stream
Improvement method  (20_).

     Some waste samples filtered slowly through 0.^5- and 0.10-ym filters.
To avoid filter clogging, a pad of filter aid (Celite, AR, Johns-Manville)
was used on 1.2-, 0.^5- and 0.10-um filters for these wastes.  The filter-aid
pad was formed from a 1-percent slurry of Celite AR poured onto the filters
and then dewatered.  Analysis of these samples with and without the use of
the filter aid (Table 2) indicated that the filter aid did not adversely af-
fect the analytical accuracy.  Therefore, in later work, the filter aid was
-------




















0
f£J
^H

p
g
w

w
. "I
M
«y
o
a
o
EH
PH
H
K
O
M

Q

•
rH

W

<£
EH


























!H (D
01 
CH "H
•H -P
a -d
rH "d
o o)



CU
|cd

H bO
0 0





CO
d)
>
0)
CO
-p
C
0)
H
^-t
q_i
M








hO
S 0
S c
O D
a 2
CU O*
H (U
M co




M a?
o *d
EH











O
0) O
ft o
o



H
H
S
(U

h H £H
OJ 0> O CU
>vQ .rH ft rH -H
H <*H ^ "w
H -H O bO -rl
•H~bO d '3 B id
fn B rH O t— H
•d o -H • cu o D
K OJ -P rH C C
CU O <8 O O O
o a o a a a
 -rt 8 ^30

H OOOcdrHC CUCU
H cdh -Pti'HCd S -P+3
•H E-pcdfiH O HrH

CUSsCUcd OCUO^CU fn M) LT\tH
p < ocdo r-DHcd-H cuSocdoHcu
(L, a opnffl HopqfL,s ^oasaoa






1
0) O >j
W) ft -P
cd >, -H
^2 -P fl CO
<; <; co cu c! cu
a a s -p cu *d
CU P -H "d 'H
H -rl fn rl
MOO & O
C H H bB ^
•H Cd ^5 -rf d)
WOO W ft

d 3 -d ft^D^t - CO O
oj co _* LA _s ro ^t
rH O O O O O O
M fn
d) (1)
(3 P -P
•H CO Cd d • — •
fi d) > CU •
0 CU r- SO
cd (3 -H cu -p +J
a -H -p -p cd d)
,CJ *H CO CU

c! d) cd >d > -p d)
•H ft 0 cd bO
fi-dft^EO cup
•H d) O -H -P rH
cu^j^t ftO-P cd co
r^j u o3 fn cd ^
O 08 p | JlJ >»_* +3
SH -f3 £H 'H d
Qj W CO i-f~\ QJ JH " cd
X ^i HJ O 0 H
o O • • cd H co ft
CU O O O d) fi -H
Q pq a a PQ o s







o
•H
-p
-p cd
cd ^

pq  ON

o

rH



H
rH
•H
S

q
S
q
•H
d)
T3
CU
R
o





o






















































•
CO
-p
o
c
-p
0
o

o
-------

















1
n
CRIPTIOH OF MILLS SURVE
t/J
o


i
j-*
•H
+>
C
o
o
H
3
i


























D 

1%
rH bO
o s




Effluent sewers




to
C 
r^ S
£p
p .
*\ o
(U
w





Q) CO t^ XA ^t
o IAVO on c\i o o o
OrHO OO OO O




g -p
•H-p rl-PCUH-PO^>
ri C to B 01 W to
Pi «fc i-HI^^^dbDobDxlI*
Hi -H 3 O O -rl -H -p




1)
c
o






o
on
(M













01
-p C
w Pi
s







y;

.
^ (U

•d m
O 01
•H -P
a s
•rH Oj
I — 1 01
01 ^i
0) 01
3 -H
rH *d

o
H .r(
0) C
a cd
•a »
3 o
« 'd
0 C

s and detergent added.
silicates added.
ngth resin, defoamer, t!
g agent, dryer release,
chlorine added.
IL, -d  P oj
8 m p
O £3 11 @
S3l 53
-------
used as a standard practice with the precaution of first checking the effect
on one or two samples of each waste stream.  Eventually the exploratory
studies with various grades of filters resulted in adoption of the 0.45-um
porosity as a standardized procedure.

                  TABLE 2.  ANALYSIS OF FILTRATES PREPARED
                        WITH AND WITHOUT FILTER AIDa


          	 BQD5> msA	      	COD, mg/1	
             1.2 um           Q.45 umc          1.2 urn           0.45 ym
 Sample   With   Without   With   Without   With   Without   With   Without
F 190
74
H 32
48
B 770
1246
c 166
184
178
82
34
42
778
1230
192
181
182
70
31
44
672
1218
188
168
176
80
32
46
675
1184
184
174
464
212
133
105
3741
5363
352
316
436
216
123
126
3650
5320
362
316
4 09
199
108
118
3636
5190
329
275
369
191
114
106
3615
4890
329
292
 ^Celite (AR) BOD5 = 0.04 mg/g.
  1.2-ym BOD 5 =1.92 mg/sheet.
 C0.45-um BOD5 =0.62 mg/sheet.

     A few samples were also processed through a Dia-Flo (Amicon Scientific
Systems) stirred ultrafiltration (UF) unit with a nominal 100-A (0.01 Urn)
pore filter.  The results were compared with those from the 0.45- and 0.10-
pm filters (Table 3).  Since no differences were apparent in either BODs or
COD, the more time consuming UF procedure was not routinely used.

                   TABLE 3.  COMPARISON OF ULTRAFILTRATES
                    FROM A 100-A MEMBRANE WITH FILTRATES
                        PREPARED WITH FILTER AID OVER
                          0.45- AND 0.10-um FILTERS
Sample
F
H
JJ

SUFa
259
272
30
59
40
35
BOD, mg/1
S045F
310
256
43
48
38
40

S010F
266
267
32
50
35
40
                   aSUF  — Dia-Flo stirred ultrafiltration
                    cell with 100-A membrane.
-------
DATA AND DISCUSSION

     In order to establish the various total and soluble components  of BODs >
the incoming samples were analyzed "as received," "settled," and after pas-
sage through either 1.2-, OA5-, or 0.10-um filters (Figure l).   Throughout
this section of the report, the designations are as follows:

     AR        As received
     AR045F    As received, filtered through O.U5-ym filter with a Celite
               overlay
     S         Settled
     S12       Settled and passed through a 1.2-pm filter
     S12F      Same as S12, except with Celite overlay
               Settled and passed through a 0.^5-ym filter
               Same as 30^5, except with Celite overlay
     S010F     Settled and passed through a 0.01-um filter with a Celite
               overlay
     SUF       Settled sample filtered through a 100-A UF unit

     Data in Table k are averages of the six samples from each mill, along
with high and low values (range) for each set.  The values in parentheses are
the percentages of the constituent in the sample after treatment.  BODs de-
creases as filter pore size decreases (Figure 2).  The data in Table k and
Figure 2 can be summarized as follows :

     1.  Comparing "as received" and "settled" values, we find a marked
         reduction in BODs (to 62 percent) after simple settling (JJ).
     2.  The 0.45-Um filter was, for all practical purposes, the best
         filter for the removal of "insoluble" BODs.  This filter was
         used for further studies of "soluble" versus "total" BODs-
     3.  Seven percent additional BODs could be removed from the JJ
         mill samples by passing them through the O.U5-um filter.
     k.  The COD values paralleled, but did not directly correlate with,
         the BODg values.

     In order to evaluate the operation of the mill clarifiers we analyzed
clarifier influent and effluent samples in the same manner.  Data for these
samples "as received," and filtered through 0.^5-Um filters, are given in
Table  5 (BOD), Table 6 (COD) and Table 7 (suspended solids).  In order to
simplify the presentation, the "as received" values are plotted in the same
sequence in Figures 3, ^, and 5.  Marked reduction in total BOD5 was achieved
in clarifier JJ.  All clarifiers except those of Mills I and J achieved 50-90
percent removal of suspended solids.

     Unfortunately, the values for constituents and removals are quite errat-
ic, probably due to a time lag for arrival of these constituents at the sam-
pling points.  Although most samples were 2^-hour composites, compensation
for the effect of the clarifier "holding time," in some cases on the order of
18 hours and in most cases at least k hours, was not built into the sampling
system.  While this would have little effect on the monthly mill averages, it
had a marked effect on the samples sent to our laboratories  (Table 8) .  We
found major differences in pH, BODs, and COD for the two streams in spite of


                                      10
-------
•d
    H->  H
'd  4->    •
 0)  4)  K
 M  to
•H      P
 a  o   as
    -p
 C       M
 O  -d   3
rH  4)   O
O  H  rH
                 CO
                •rH
                 CO
                 S

                •d

                5

                 i

                CQ

                H


                I
      B  4) LA O CO

JH • CO
>
4J H 0) H
rH H £> Cd
•rl O O OS C
fe O i-3 <.
J-










fJJ ^^»
^3 +> fe pq
•P -H B CM ^^
•H H 3. H
> 4) CO to
O CM 	 	 -H
*H • H tfl
Q> W) rH CL> i>>
+30 43 rH
•H O O iJ fi
c^ o ^
•*










o
o
o
•d
5
fa
m
Q
O

I
LA
O
'd
C
trj

CM

rH
rCj
+J
•H
>

r?H
•H

o
to

•d
cu
T*
4)

to
CO
1

"^
*^^
to
•H
CO
H

jj





























O
O
o

•d
c

in
O
o
pq
1

"pq
*— •*
to
•H
to
rH
05
.§


01
in
O
W
0)
rH
jj
,— |
O
01
o


                3
             CO

            •rH

             01

             >>

            H
             05
                                                 11
-------



-* vo
O ON
I---*


H O
t/J C—

CO
.* UN
O 1
M H OJ
-* cy



cy
OJ UN
5 o w
OO CM

h PON
o vo "V
CO VO OJ
O ON
HCO

"LA m
fe £3
c5 o o
W OJ VO
H ON
ON

&* VO H
CM -^7
H CO 0
CO HO
OJ O
-~» OJ

VO H
OJ O
H H
vo
CO
g
ooS
^-» OJ
Pn voco
o LA on
s £3

^- o
Ct, C?\ ON
tA IA rO
o on oj

OOCO
frj ^"7
H H H
w H on
on oj

t— tA
UN LA
ion oj

•— - OJ
fc _* OJ
LA VO -*
O LA LA
« H H
•4 on OJ



VO
LA CO
C7\ UN

CO


°* VO UN
vo vo
H
|d **•


,— 1
m
I
H O
CVI rH


O
VO
cy
-* i
H -*

cy
 LA
t-co
vO OJ

&s




oo
IA VO
ON UN

vo


H OJ
VO UN

a


LA
o
CM
UN 1
H CM
H UN


LA
ol
H£
CO

H
OJ CO
HOD

H
vo

H
C- 1
ONVO
0 t-
H vo

•— cy
^.-^
on H

• — •CO-
CM o
OJ UN
VO H
VO ON
OO rH


OO OJ
OJ UN
C— CM
CO H

'— • CM
HVO
^MCO
oj on
UN OJ
CO
vo
CM
OJ
CO CM
VO O

on oj
O 1
Jt OJ
H t-

J*vo"
H C-

OO O
-* CO
UN ON
H t—
^.
LA CT\
^ 1
HCO








on

ON H
UN OJ

ON

t-
2^
UN OO

h
oo

IA
OJ
vO OJ
H tA
OO
00
OO
OJ
O 1
VO UN
H UN
CO
o
UN on
•35
-* OJ
tA


VO
1
-* on
CO UN
ON-*
on CM

ON OJ
vo 1
H t—
H «N

^-- u-\
H VO
^t UN
-— OJ


1-1 ^

•""•• OO
^* VO
J- t
OJ I?-

CM ON
$&
<%£
0
8!
CO
CO CM
H 0
ON (—

"0^100
vo VO
CO UN
ON UN
^r CM

s~* H
vO vo
OOOO
1A OJ

CO 0\
vO vo
CO O
on t—
LA OJ
f. t--
CO CO
"o vo
on co
vo OJ








t—

I— CM
CO CO
UN

H

ONVO
ONVO

o


vo
H
-V
CVI CO
t- OO


•§
H

vo
1
O C>J
J- CO
OJ rH



OJ
OJ
££

^.
00 H
'o T1
O UN
H t—

ff^
ON CM
-=r on
eg i
o on
H t—


ON O
m on
• — • H
o on

cy!^
CO OJ
vo o
H H
VO
H
1
CM-*

Po
on CM

CM VO
on oj

^^
OJ OJ
ro Oj

5?
-* OJ









o
J




^* ON
t— vo

a

CO

1
vo CO
OJ ON
-* H


UN
t
CO -*
-* vo
on H
CO
OJ
vo
on o
UNVO
co on

vo
LA

H
1
OO O
OOVO
VO CVI

UN ON
•*-- OJ
CO 1
CO H
OO ON
H -*

- — • o
_* UN
UN OO
-— OJ
-* CO
HCO
H LA

--*. LA
t— ON
-^ CM
VO ON
- — LA
O ON
ON H
^- OO
0 1
H -*
ON H
H 0\
on
1
CM LA
*-*• o

ON OJ
O CM
-* CM

IA cy
VO P-
_* UN
O CM
-* CM

ON OJ
VO t-
OJ -*
OOCO
-* H
— - LA
II
u\ on.










CO
1
CM


H
OJ UN
CO VO

H

OJ
IA
UN OJ


LA
on co
OOCO
CO

o|
VO OJ


on
o
H
VO OJ

-— . 00
•*-* t
VO cO
on t—
CO CM

— 0
vO -3-
-d- LA

oj LA
on OJ


OJ CO
UN UN
-—• 1
O LA
t— t—

O UN
OJ t—
O LA
LA OO
CO
on
H
H
£3

ON CM
UNVO
vO 1
onvo
HCO

SvS
,-t CO

t^ t-
LA VO
OJ 1
H"ON
— - m
CO W
O OJ
OJ H








o

1
o on
on LA

H


£-i
CO t—

^



o
ON
on,*-
-* H


CM
ON
on i
on co

LA
vo
VD
CO O



t—

VO
1
CO ^O
VO CO
UN on


To r-j
H H
co vo


— OJ
on H

CO U\
co vo


- — • on
-* 04
H H
ON"H

oj VD
OO OJ
CO 1
rH VO
CM CO
8
OO
1
H -*
ON O

H-*
oo cy
on on

H C—
on CM

.^
VO VO
3-J, 8
x m H
- s
CO OJ H
^— - 1 t>
VO 0 P
.* on j~-
-d
<0
41

H
•H
4-<

o
'k— '
-d
vo 0)
^ 3
OJ ON V
\
CO h
C
CO T
St C
-* •*•
« h- |
PL
r5 *
12
-------
   500r
   400-
c
z
   300
u
o
>
x
o
o
UJ
X
o
o

GO
   200
    100
         As

      Received
PAPER MILL
   Settled
1.20
0.45
        Figure 2.  The  separation of insoluble from soluble BOD5

                  by filters of various pore sizes.
                                                              -xC
                                                              -A D
                                                                 H
0.10
                                  FILTER PORE  DIAMETER,
                                13
-------
TABLE 5.  BOD 5 IN CLARIFIER INFLUENT AND EFFLUENT SAMPLES ("AS
       RECEIVED" AND FILTERED THROUGH A 0.1*5 ym FILTER)
Mill
A
B
C
D
E
EE
F
G
H
I
J
JJ
As received (total)
Filtered (soluble)
Influent , Effluent , Removal ,
mg/1 mg/1 %
1*95 31*1
No clarifier
261* 192
156 90
No clarifier
No clarifier
378 322
787 61*5
76 51
622 557
230 152
122 37
31

27
1*2


15
18
33
10
33
70
Influent
mg/1
303

11*5
75


179
536
39
371
131*
38
, Effluent , Removal ,
mg/1 %
30l* -0.3

138 5
59 22


200 -12
1*32 1*5
33 15
2l*7 3l*
121* 8
32 16
TABLE 6. COD IN CLARIFIER INFLUENT AND EFFLUENT SAMPLES ("AS
RECEIVED" AND FILTERED THROUGH 0.1*5 JM FILTER)
Mill
A
B
C
D
E
EE
F
G
H
I
J
JJ
As received (total)
Filtered (soluble)
Influent, Effluent, Removal,
mg/1 mg/1 %
1863 1328
No clarifier
712 1*29
583 386
No clarifier
No clarifier
1695 61*2
2918 1606
272 188
2132 2013
715 1*33
691 126
29

1*0
31*


62
1*5
31
6
1*0
82
Influent
mg/1
1207

283
21*6


1*08
1201
108
111*1*
351
88
, Effluent, Removal,
mg/1 %
1236 -2

286 -1
25U -3


399 2
111*9 1*
98 9
IIOU 1*
28U 19
8U 5
                               14
-------
the fact that both were collected within the same time period.   This lack of
continuity in the two samples could explain not only the lack of removal ob-
served at times but could also be the reason that our studies showed the
higher or lower removal values than did general mill experience.

              TABLE 7.  SUSPENDED SOLIDS IN CLARIFIER INFLUENT
                        AND EFFLUENT SAMPLES REMOVED
                             BY A O.U5 um FILTER
Mill
A
B
C
D
E
EE
F
G
H
I
J
JJ
Influent , Effluent ,
mg/1 mg/1
UUl
No clarifier
1126
235
No clarifier
No clarifier
1187
Ui65
2UO
853
690
581
63

2U8
77


213
103h
99
587
3^0
50
Removal ,
86

78
67


82
75
59
31
^7
91
                                    15
-------
800


76o


720
6I»o


600


56o


520


Wo


1.1,0
YJ\ Influent (ng/1)


II Effluent tmg/1)


RES J Removal
320


28o


21,0


200



160


120


 80


                                                I


                                                               S
                                                                                    -90
                                                                                    -80
                                                                        -70
                                                                        -60
                                                                                    -50
                                                                        -1)0
                                                                        -30
                                                                        -20
                                                                                   ,-10
                                         Mill
               Figure  3-   Total biochemical  oxygen  demand in
                             clarifier influent and effluent
                             samples.
                                          16
-------
3000

2850

2700

2550

21400

2250

2100

1950

1800

1650

1500

-350

1200

1050

 900

 750

 600

 1450

 j)00

 150
     Influent (mg/1)

    j Effluent (mg/1)

    | % Removal
    i

1
i

90
                                                                           I 80
                                                                            70
                                                                            60
                                                                            30
                                                                           Jio
                                 G
                                Mill
                                                                     JJ
Figure  k.   Total chemical oxygen demand in  clarifier
              influent and effluent samples.
                                17
-------
ItOOO


3800 _


3600 _


3"lOO -


3200 _


3000


2800 -


2600


aii oo


2200 -


2000


1800


1600


11(00


1200


1000


 800


 600


 1<00


 200
    R73 Influent (ma/1)
    V/A


    \~\ Effluent (mg/1)


    IK] Removal, %
Itl85
I
I

                                    I
                                              90
                                                                        80
                                                                       . 70
                                                                        60
                                                                        50
                                                                        ItO
                                                                        30
                                              20
                                                                        10
                                            G
                                           Mill
              Figure  5-   Suspended solids  in  clarifier influent
                           and  effluent  collected on a  O.h^-ian
                           filter.
                                           18
-------
     TABLE 8.  THE VARIABILITY OF pH,  BOD5 AND COD
          IN CLARIFIER INFLUENT AND EFFLUENT
           SAMPLES FROM DIFFERENT SOURCES8"

pH
4.81
8.28
10.42
11.53
10.18
7.35
7.11
Influent
BOD 5
27^
884
482
634
836
476
168

COD
2268
3899
1702
21498
3388
1413
411

pH
6.88
10.68
11.05
9.88
10.80
9.80
10.78
Effluent
BOD 5
268
555
489
777°
1104C
822C
258C

COD
552
1548
812
2202
2478
2293
869
 These values are from all of the samples (54 sets)
 where pH differences and/or negative "removal" vas
 noted.

 Influent and effluent samples were taken during the
 same time period.
12
 Prolonged holding time probably explained the apparent
 negative removal.
                          19
-------
                                  SECTION V

                LABORATORY STUDY OF PROCESSES FOR HIGHER LEVELS
                           OF SOLUBLE BOD5 REMOVAL


     We approached the problems of improving pulp and paper mill liquid
waste treatment by two differing routes.   A first route studied possibilities
for improving the methods for conventional, out-plant primary clarification
treatment of the total mill discharge.  The second, a more innovative route,
was directed to the principle of separation of flows, particularly spills of
strong process water, and was also directed to individual, high-efficiency
treatments, preferably within the plant,  for those flows detrimental to
clarifier performance.

     Various treatment methods, such as sedimentation, flotation, filtration,
membrane separation (reverse osmosis, ultrafiltration), carbon adsorption,
and ion exchange could be used.  The results of some of these methods are
described in this section.

     New and relatively more expensive methods of processing would be feasi-
ble for only parts of the total flow in order to prevent buildup of solubles
in streams, especially those being recycled.  These fractions could be treat-
ed in-plant in small, highly efficient systems or, if the streams are not to
be recycled, they could be treated before they are mixed with other process
waters entering the total sewer system.

     However, since many of the mills involved in this study were already
using clarifiers or had plans to use them, the laboratory study was care-
fully planned to develop a better understanding of the conventional, out-
plant clarification process.

SCREENING TESTS

     Following the survey of mill discharges and clarifier efficiencies
(Section IV), a study to optimize the removal of soluble BODs from both
total mill waste waters and individual streams was begun in the laboratory.

     We preferred to evaluate clarification efficiency by some means other
than the BODs test because the 5-10 days  time to run each test, depending
upon the day of the week that the samples were prepared, made systematic
laboratory studies very difficult.  Also, we wanted a procedure based on
small sample volume.

TEST TUBE TRIALS

     We had developed in our laboratory a simple test tube flocculation test
using a 15-ml sample in an 18 x 150 mm test tube mounted on a 12-inch diam-
eter, vertically oriented, platform rotating at 3 rpm (unpublished).  The
                                     20
-------
slow transfer of fluid from one end of the tube to the other produced a floe
form and supernatant clarity that correlated well with jar tests with the
same sample/flocculant combination.  This simple, small-volume system was
used in the early part of the work for screening the incoming samples.

JAR TESTS

     At the same time, jar tests were carried out on the same waste material
using a Phipps and Bird laboratory flocculator.  Various flocculating agents,
alone or in combination, were added under carefully standardized conditions
of chemical addition, rapid mixing, flocculation mixing, settling and sampl-
ing.  Briefly, these can be described as:

     1.  The treatment of 1.0 or 1.5 liters of waste in 3-liter jars;
     2.  The addition of the primary flocculant, followed by one minute
         of mixing at 100 rpm;
     3.  If a second agent (e.g., polymer, silica, activated carbon, etc.)
         was added, it was followed by a second rapid mix of one-half
         minute at 100 rpm;
     k.  The paddle speed was reduced to 20 rpm for 20 minutes to provide
         time for flocculation.  The flocculation step was done after the
         final addition of chemical, no matter how many additives were
         involved;
     5-  At the end of the flocculation period, the jar was removed from
         the unit and the sample allowed to settle for one hour;
     6.  Samples were withdrawn into 250-ml bottles under slight vacuum,
         with the tip of the sample tube one-half inch under the surface
         of the liquid in the jar;
     T.  An untreated jar of the waste was carried through the entire
         test from rapid mix to sampling.  All "removal" data are based
         on the comparison of the analyses of the control sample with
         the treated samples.

All samples were stored at k°C (hQ°"F) prior to analysis.

ORGANIC CARBON ANALYSIS

     Use of a Beckman Process Carbonaceous Analyzer permitted relatively
rapid (15 samples per day), small-volume analysis of solutions for total
organic carbon (TOC).  The test tube trials, after the floe was removed with
the 0.^5-ym filter, provided sufficient sample for the determination of sol-
uble organic carbon.  The use of larger test tubes (25 x 250 mm) and addi-
tional sample volume on the same flocculator table, or jar tests directly,
allowed determination of both total and soluble BODs and TOC, and soluble
organic carbon (OC) on the same sample.

     We found little correlation between the reduction of soluble organic
carbon and the reduction of soluble BODg for a series of flocculation studies
(Table 9)•  We also found no relationship of COD to BODs during the mill sur-
vey; these results are probably due to the high ratio of nonbiodegradable to
biodegradable carbon in these waste streams.  Total and soluble organic car-
bon and BODs components removed during flocculation would be both degradable


                                     21
-------
and nondegradable substances, with the percentage removal of degradable be-
ing overshadowed by nondegradables.   While the polymers did have both bio-
chemical and chemical oxygen demands (Table 10), the amounts added to the
solutions were so small (l to 10 mg/liter) that they did not contribute sig-
nificantly to COD or BODs.  The addition of the polymer at the 10 mg/liter
level would add less than 2 percent COD and 1 percent BODs to most samples
if all of the polymer remained in solution and none was Jfemoved by floccula-
tion and/or filtration.

                 TABLE 9.   SCREENING TRIALS OF FLOCCULANTS AND
                        POLYMERS BY  ANALYSIS OF SOLUBLE
                        ORGANIC CARBON AND SOLUBLE BOD5
Reduction , %
Flocculant
Type
Alum
Alum
FeCl3




Lime


mg/1
200
200
250
250
250
100b
150
200
200
200
200
250
25
25
25
Polymer
(0.5 mg/1)
None
Nalco 607
Nalco 627
Nalco 634
Nalco 73C32
None
None
None
Nalco 607
Nalco 609
Nalco 627
None
None
Nalco 634
Nalco 73C32
Soluble
OC
18.2
34.3
26.5
15-7
6.0
20.7
17.4
25-4
21.3
23.1
-7.4
20.5
1.8
-6.2
2.3
Soluble
BOD5
-9-7
-20.9
8.1
5.4
0
20.3
-20.5
21.2
16.8
20.7
15-7
19.7
-1.9
8.4
0
           values on "as received" samples:

             Soluble OC   = 125-425 mg/1.
             Soluble BOD5 = 100-200 mg/1.
          bAs Fe3+.
                                     22
-------
                   TABLE 10.  COD AND BOD5 OF VARIOUS POLYMERS
                          USED AS FLOCCULATION AIDS&
Manufacturer
Nalco
607
609
63h
T3C32
Dow
PAA
Tydex 12
Hercules
Hereof loc 812.3
Soluble COD,
mg/1

231
7^0
333
1*02

15kh
1665

—
Soluble BOD 5,
mg/1

0
lit
0
0

2lh
0

66
               Polymer concentration 1 g/1.
 PHENOL  SULFURIC  ACID  TEST

      The phenol-sulfuric  acid test (£)  is  a rapid method for the estimation
 of  carbohydrates,  including virtually all  classes of sugars, sugar deriva-
 tives,  oligo-  and  polysaccharides, ketoses, aldoses  and aldehydes.  This
 method  has been  used  for  several  years  in  these  laboratories for monitoring
 gel chromatographic columns and for the study of aldehydes  in solution,  both
 carbohydrates  and  noncarbohydrates (8).

      Although  the  method  appeared to be useful for measuring the removal of
 soluble BOD5 during trials  in the test  tubes (Table  11),  it  did  not give re-
 producible results in the jar tests;  therefore,  it was  not useful as a re-
 placement for  the  BOD5 test (Table 12).

 ZETA  POTENTIAL

      "If the zeta potential of the colloid  is  lowered below  a critical value,
 the colloids tend to coalesce,  resulting in coagulation"  (£).  Many of the
 soluble" BOD5 components in the waste  stream  are  colloids, medium and low
molecular weight organics, and in  some cases simpler  carbonaceous  compounds,
which should be amenable  to  colloidal coagulation  and coprecipitation (10.).
 Zeta potential changes and  final values  were compared with the removal of
COD and BOD5 components by  flocculation  (Table 13).  Wo correlation could be
drawn between the zeta potential and the reduction of either  COD  or  BOD5 for
the waste stream after flocculation and  filtration.  There was, however, the
expected correlation of near  zero  zeta potentials with good floe  formation
and supernatant clarity.
                                     23
-------
           TABLE 11.  COMPARISON OF PHENOL SULFURIC ACID TEST WITH BOD5
                  TEST TO EVALUATE FLOCCULANTS IN TEST TUBE TESTS
Flocculant
Type
FeCl3



Lime








mg/1
50*
100
150
200
250
300
300
300
300
300
300
300
hOO
Polymer
(1 mg/l)
None
None
None
None
None
None
Nalco 609
Nalco 627
Nalco 63 **
Nalco 73C32
Dow PAA
Dow Tydex 12
None
Q
Reduction,
Soluble glucose
equivalent
28.9
U5.6
50.0
1*6.7
17-7
6.3
25.6
1*0.5
1*0.5
20.0
39-1
32.6
28.1
%
Soluble
BOD 5
22.8
32.1
36.2
33-9
5.0
18.1*
16.1*
12.9
25-9
12.9
23.2
31.7
7.1
            values "as received":

               Soluble glucose equivalent = 2l6 mg/1.
               Soluble BOD5 = 200 mg/1.
           bAs  Fe3+.

     As the study progressed we found little relation of the visual clarity
of the supernatant or the rate of floe settling with the reduction of BODs.
A finely divided floe seemed to result in  the  greatest reduction in soluble
BOD5 after filtration with the 0.1*5-um filter, whether or not the floe had
settled.

JAR TESTS FOR OPTIMIZING BOD 5 REMOVAL BY CLARIFICATION

General

     Following the lack of success in developing  a rapid screening test to
correlate flocculation with the removal of soluble BODs. a number of jar
tests were made with several mill effluents.  We  used the standard BODs anal-
ysis to test these flocculants.  While this resulted in a rather delayed
evaluation, it could be used for:  (l) a study of many flocculating agents
and variables or (2) careful consideration and selection of both flocculants
and process variables, with BODs determinations on either (a) all samples or
(b) selected samples from each series.  With frequent reevaluation of the
procedure as BODs results became available, the  (2b) scheme was selected for
the majority of the jar tests.
                                      24
-------
TABLE 12.  COMPARISON OF PHENOL SULFURIC ACID TEST WITH
           BOD5 TEST TO EVALUATE FLOCCULANTS
                     IN JAR TESTS
Reduction , %
Flocculant
Manufacturer mg/1
Nalco
607
609
627
631*
73C32
Dow
PAA
Tydex 12
values "as
Soluble
Soluble
Soluble

6.0
6.0
6.0
10.0
6.0
6.0
10.0
6.0
10.0

6.0
10.0
20.0
6.0
6.0
6.0
10.0
received":
PH

7-5
3.0
7.5
7-5
3.0
7-5
7-5
7-5
7-5

7-5
7-5
5-0
3.0
5-0
7.5
7-5

glucose equivalent
COD = 550 mg/1.
BOD 5 = 220 mg/1.
Soluble glucose
equivalent

-1.5
12.0
11.5
16.5
U.5
7-3
15-3
7-3
11.5

15-3
15.3
10.8
3.2
-1.2
7.3
Ik. 2

= 250 mg/1.
Soluble
BODS

2.6
11.3
6.8
9.8
Q.k
6.8
10.0
-2.1
23.8

2.1
12.8
-1-7
k.6
-2k. 8
8.1
5-1


Soluble
COD

-1.8
10.2
7-2
7-2
5.9
2.7
16.6
1.8
5.k

8.6
9.3
U.2
1.3
3.3
3.8
6.8


                         25
-------
       TABLE 13.   A COMPARISON OF ZETA POTENTIAL,  SOLUBLE  COD AND BOD5,
                   AND FLOG  FORM IN JAR TESTS  OF FLOCCULANTS
Flocculant
Type
Alum


Fe3+




Lime
mg/1
200
395
632
50
75
8U
8U
200
200
Effluenta
A
B
B
C
C
B
B
A
A
Zeta
potential,
pH mv
-9-3
0
0
0 to -2U.5
0
3.0 0
5.0 -17.8
3.0 0
11.6 -20.2
Reduction13, %
Soluble
COD
2.1
17-3
12.2
16.8
23-2
30.5
33.3
U5.3
U.O
Soluble
BOD 5
0.0
-1.5
3.6
5.3
11.9
21.9
17-2
51.0
5.8
FlocC
form
__
U+
u+
—
—
k+
3+
—
—
 A = special sample of beaten sulfite pulp.
 B = total mill effluent (Mill F).
 C = digester room effluent (Mill F) .

 Values for "as recieved":

    A — soluble COD  = 1+62  mg/1.
        soluble BOD 5 = 128  mg/1.
    B - soluble COD  = ^53 mg/1.
        soluble BOD 5 = 190 mg/1.

    C - soluble COD  = 128U mg/1.
        soluble BOD 5 = 585 mg/1.
Q
 Floe form — 0 = supernatant clarity equal to untreated control.
            5+ = clear supernatant without suspended solids.

     To reduce the problem involved in setting up sampling schedules , sample
transportation and storage, it vas further decided to select  mills by two
criteria:  (l) their proximity to our Appleton laboratories and (2) those
having high soluble BOD 5 values in the initial survey.  Others could be add-
ed, as time permitted, to allow studies of special effluents, of effluents
with lower soluble BODs, or of effluents from more distant mills.
                                     26
-------
TOTAL MILL EFFLUENTS WITH PRIMARY FLOCCULANTS

     Tables lit, 15 and 16 summarize the large volumes of data gathered from
jar tests with effluents from Mills A, D, and F, which are using convention-
al clarifiers or flotation systems.  These data can be summarized as follows:

     1.  Ferric chloride in the range of 75-150 mg/1 Fe3+ was the
         best flocculating agent for these waste streams.  The
         optimum pH was dependent upon the particular stream being
         processed and the concentration of the iron salt added.
     2.  Lime, at higher concentrations (200-500 mg/l) also removed
         soluble BODs (21-1*0 percent).
     3.  Alum did not remove more than IT percent of the soluble
         BOD5) although there was evidence that, with some of the
         waste streams, careful pH control and optimized alum con-
         centrations could result in the removal of substantial
         quantities of soluble BODs.
     1*.  The polymers plus bentonite or Celite to provide high surface
         area did not appear any more effective than the cheaper iron
         salts.
     5.  Some trials had shown that considerable amounts (22 percent)
         of soluble BODs could be removed from the clarifier effluents
         with ferric chloride or lime (Table lU).
     6.  When the concentration of the flocculating agent and the pH are
         carefully controlled, from 25 to 50 percent of the soluble BODs
         would probably be removed from the clarifier influents.
     7-  In recent similar studies in Finland (ll_), fly ash obtained
         from the burning of calcium-base spent liquor was used as a
         flocculating agent, it was used to raise the pH during alum
         treatment, and it was used as a sludge conditioner.  Reductions
         of 50 percent for BODs and 70 percent for COD are reported in
         the treatment of kraft effluents.  The authors also report
         that the sludge is easier to handle.

ADDITION OF ACTIVATED CARBON

     Studying pilot-scale columns of activated carbon processing a chemime-
chanical (CM) pulping waste, we found that 83-90 percent of the total BOD5
was removed (±2).  We have no information, however, on the degree of soluble
BODs removed by adsorption on activated carbon.  Therefore, we used jar tests
to measure the reduction in soluble BODs by powdered (300 mesh) activated
carbon, with and without the addition of cationic polyelectrolytes, for the
treatment of the total mill effluent (Table 17) and CM sewer discharge from
Mill F (Table 18).

     Soluble BODs could be reduced by 32 percent in the total mill effluent
and 17 percent in the CM streams.  The variability of the results for the
various levels of activated carbon was quite high.  This was most likely due
to changing conditions in the mill — high hourly variations in the flow and
concentration of substances in these streams — rather than to differences in
the effects of the activated carbon and polymers.
                                     27
-------














Q
1
-a!

a
3
is
M
w g
§ §
3 j
O fe
a
^ i
EC TO
S ^
H W

^ PM
O H

M P
t"1
W H^
*"£

3
>-5

H

3
9
EH














-P
a

H
4-4
(3
•H

P

H
•H
S


•g
M
 CO CM O
•H -* ON -H
•p CO CM .p
•H O
•O II II 3
*O ^3
aj u) (it

o § § K
& O PQ
CM
H H H
->»^
•H 0 6
•9 VO CO C!
C CM 00 O
0) H CM -H
OH -p
0
HUH 3
^- >d
t^ O fl)

CM o PQ
CM
vo
^>^
•O OO OO C
n t- CM o
• o oo o
•rl CM CM -H
-P H -P
•H 0
-d II II 3
gj in 
-P OOOOOOO OOOOOOOOOOOOO
0) ^ f- t- ^ f- t— t— ITvlTNlAOOOOOOOOOO
^ H H CM CM CM 00 ^ -d- IA in
P.


-------
            TABLE 15.  JAR TESTS OF PRIMARY FLOCCULANTS WITH MILL F
                           TOTAL CLARIFIER INFLUENTa
Flocculant
Type
Lime





Bentonite



Fe3+
{as FeCl3)















Alum











mg/1
50
100
200
800
1*00
500
50
100
150
200
12.5
25,
28*
5Qb

70
70*
70°
b
81**
81*
81**
100,
112*
ll?
150
200
25
50
100
150
158*
200
250
385*
385*
385*
385*
385*
385*
385*
632*
632*
Final
pH
8.8
9.2
10.6
11.1*
11.6
11.7
	
—
—
__
6.8
6.5
2.7
5-7
5-2
1*.6
l*.l
8.6
3.5
5.1
1*.2
3.2
3.3
6.0
3.8
2.9
2.7
	
—
—
—
3.5
—
8.9
6.8
6.2
5.3
1*.6
1».3
3.9
10.6
6.9
Sample 8/13
COD = 508 mg/1
BOD 5 = 223 mg/1
Reduction, %
COD
-7.2
0
7-2
11.8
12.9
—
5-1
7.6
6.5
7-6
__
__
—
—
—
—
—
—
__
_ _
__
—
__
__
—
—
—
__
—
—
__
	
— .—
...
—
__
—
__
^^
BOD 5
-10.2
-10.2
-1.6
3-7
15-5
—
1.1*
1.0
-3.6
-1.1*
__
	
—
—
—
—
—
—
__
__
__
—
—
__
—
—
—
__
—
—
__
	
— —
— —
—
__
__
_H
"•"•
Sample 8/21
COD = 209 mg/1
BOD 5 =102 mg/1
Reduction, %
COD
-3.8
-1.9
10.0
lit. 8
17-1
20.0
	
—
—
—
5-1*
8.9
—
19-8
—
—
—
—
—
__
__
—
15.8

—
13.9
15.8
9.0
1+.3
3.8
5-2

10.0
10.0
—
__
__
__
•~—
BOD 5
0
0
21.1*
18.1*
36.7
1*0.0
—
—
—
—
1.1+
1.1*
—
7.2
— —
—
—
—
—
—
—
—
8.7
—
—
10.1
18.8
-1.0
U.9
3.9
2.0
—
2.0
2.0
—
—
—
__
—
Sample 9/10
COD = 1*1*0 mg/1
BOD 5 = 191* mg/1
Reduction, %
COD
__
—
—
—
—
—
—
__
—
—
—
—
12.9
—
31.3
35.8
31*. 8
33-7
28.0
33.3
33-7
30.5
—
32.1*
30.7
—
—
—

—
—
8.2
—
8.7
19.2
21.9
21.9
23.9
19-2
11.9
12.2
17.8
BOD 5
__
—
—
—
—
—
—
__
—
—
—
—
7.5
—
5-2
2U.O
22.1*
8.8
6.2
12.2
18.8
21.9
—
17-5
11.3
—
—
—
—
—
—
3.0
__
-5.2
-2l*.2
12.0
-18.3
16.7
11.1*
7-3
3.6
-2.1
^Soluble COD and BODS were measured.
 Added on molar basis of Fe3+ and Al3"1" (0.5, 1.25, 2.0 millimolar).
                                      29
-------
                  TABLE 16.  JAR TESTS OF POLYMERS AND ADDITIVES WITH
                          MILL F TOTAL CLARIFIER INFLUENT5
Flocculant
Manufacturer
Nalco
607
607
607
609
609
609
609
609
627
627
627
631*
63U
63lt
63U
73C32
73C32
73C32
73C32
73C32
Dow
Tydex 12
Tydex 12
Tydex 12
Tydex 12
Tydex 12
Tydex 12
PAA
PAA
PAA
PAA
PAA
PAA
mg/1

1.0
1.0
6.0
1.0
1.0
6.0
6.0
10.0
1.0
1.0
6.0
1.0
1.0
6.0
10.0
1.0
1.0
6.0
10.0
10.0

1.0
1.0
6.0
6.0
6.0
10.0
1.0
1.0
6.0
10.0
10.0
20.0
Additive
Type

Bentonite
A/C
None
Bentonite
A/C
None
None
None
Bentonite
A/C
None
Bentonite
A/C
None
None
Bentonite
A/C
None
None
Celite

Bentonite
A/C
None
None
None
None
Bentonite
A/C
None
None
Celite
None
mg/1

200
200
—
200
200
—
—
—
200
200
—
200
200
—
—
200
200
—
—
200

200
200
—
—
—
—
200
200
—
—
200
—
Final
pH

—

7-5
	
—
7-5
3-0
7-5
	
—
3.0
	
—
7-5
7-5
	
—
7-5
7-5
7-5

—
—
7-5
5-0
3.0
7-5
	
	
7-5
7-5
—
5.0
Sample 7/31
COD = 552 mg/1
BOD5 = 23it mg/1
Reduction, %
COD

—
—
-1.8
_ —
	
6.8
10.2
7.2
	
—
5.9
~._
—
2.7
10.0
	
—
1.8
5.1*
—

—
—
3.8
3.3
1.3
6.8
—
	
8.6
9-3
—
It. 2
BOD 5

—
—
2.6
	
—
7.2
11.3
9-8
— —
	
8. it
	
—
6.8
16.6
	
—
-2.1
23-8


—
—
8.1
-2U.8
it. 6
5-1
	
—
2.1
12.8
—
-1.7
Sample 8/13
COD = 508 mg/1
BODS = 223 mg/1
Reduction, %
COD

8.2
7.0
—
9-U
8.2
—
—
—
9-2
9.6
—
8.0
5.1
—
6.7
7.1*
8.0
—
6.1
3.0

12.lt
9-0
—
—
—
—
1*4.7
8.6
—
6.3
7.1*
—
BOD5

11-9
-it. 6
—
3.2
18.7
—
—
—
1.8
16.9
—
-1.U
l.lt
—
23.8
9.1
19.6
—
29.U
21.2

-1.U
20.5
—
—
—
—
-U.5
-ll.lt
—
21.8
lit. 2
—
Soluble COD and BODs were measured.

A/C = activated carbon (Filtrasorb 300,  powdered).
                                      30
-------
                  TABLE IT.  JAR TESTS OF THE EFFECT OF ACTIVATED CARBON
                      AND VARIOUS ADDITIVES OH SOLUBLE COD AND BOD5
                            IN MILL F CLARIFIER INFLUENT
Activated
carbon
Typea
(1)










(2)











mg/1
50
100
150
200
200
200
200
200
200
200
200
50
100
200
•200
200
200
200

200

300
koo
Additives
Type
None
None
None
None
Nalco T3C32
Dow Tydex
Cellulose
Nalco 609
Malco 627
Nalco 63>*
Dow PAA
None
None
None
Nalco T3C32
Dow Tydex
Cellulose
Cellulose
+ Tydex
Cellulose
+ 73C32
None
None
mg/1
__
—
	
____
1.0
1.0
500
1.0
1.0
1.0
1.0

	
	
1.0
1.0
500
500
1.0
500
1.0
—
—
Final
pH
7.1*
7.5
7.5
7."*
7-5
7.1*
—
7.9
7.9
7.8
7.9
	
	
—
—
—
—
—
—
—
—
— -
—
Sample 8/13
COD = 508 mg/1
BOD5 = 223 mg/1
Reduction. %
COD
6.1
5-7
5.9
5-7
8.0
9-0
—
8.2
9.6
5-1
8.6

	
—
—
—
_—
—
—
—
—
__
—
BODS
19-2
8.7
12.3
-5.5
19.6
20.5
—
18.7
16.9
1.1*
—

__
—
--
—
— —
—
—
—
—
—
—
Sample 8/21
COD = 209 mg/1
BODS = 102 mg/1
Reduction. %
COD
l*.l
6.9
	
13. 1*
—
—
—
__
—
—
—
ll*. 7
17.1*
11.9
—
—
— —
—
—
—
—
17.9
17.9
BOD5
3.1*
^_
__
11.5
__
_^'
—
__
—
—
—
17.0
7-5
27.1*
—
—
—
—
—
—
—
13.2
32.1
Sample 9/10
COD = 426 mg/1
BOD5 = 196 mg/1
Reduction. %
COD BOD 5
__rl n^
— —
__
3.5
6.3
5-2
5.2
—
—
—
—
_«.
—
—
l6.lt
7-5
8.0
It.O
—
5.2
	
	
	
_
«...
__
2.0
23.5
6.1
12.2
_..
—
—
—
_w
	
—
12.2
9-2
11.2
19-1*
—
ll*.3
—
—
—
  a(l) Filtrasorb 300, powdered.
   (2) Same alcohol and water washed (oven dried).

     In all of the  jar tests the samples were first  screened through a 60-
mesh stainless steel  screen, the activated carbon  (and polymers,  if used) was
added, and the sample was  processed through the mixing and settling cycles.

     No attempt was made to correlate the degree of  soluble BODs  removal with
the presence  of fiber fines or other carbonaceous  adsorbing materials.

     Although the activated carbon-soluble BODs complex  could be  readily fil-
tered by a 0.1*5-um  filter, it did not settle in the  jars with or  without the
use of polymer flocculants.  Activated carbon, therefore,  seems useful as an
additive to the primary system for the reduction of  BODs if the carbon could
be removed in either  the primary or secondary system. Larger carbon particles
might be easier to  remove; however, they would have  less surface  area and,
therefore, probably be less sorptive.

EFFECT OF THE CONCENTRATION OF THE EFFLUENT

     If the total water use in the mill were reduced through recycling, the
wastes would  be more  concentrated.  In order to study the  flocculation and
soluble BODs  removal  characteristics of such wastes  we concentrated by re-
verse osmosis  a sample of  the total mill effluent  (clarifier influent) from
                                       31
-------
the Mosinee mill.  We have been evaluating and developing uses for reverse
osmosis (RO) in our laboratories since 196*1 (13).   This process can concen-
trate solutions and at the same time remove some of the soluble BODs •

           TABLE 18.   JAR TESTS OF  THE EFFECT  OF ACTIVATED  CARBON
                   AND ADDITIVES ON COD AND BOD5 OF MILL F
                      CHEMIMECHANICAL PULPING  EFFLUENTa
Carbon,
mg/1
50
100
125
150
200
200
200
•200
200
300
None
None
None
Additive
Type
None
None
None
None
None
Nalco 63^
Nalco T3C32
Dow Tydex 12
Dov PAA
None
Nalco 63^
Nalco 73C32
Dow PAA

mg/1
—
—
—
—
—
10.0
10.0
10.0
10.0
—
10.0
10.0
10.0
Reduction
COD
2.5
3A
5-9
^.5
5-2
lt.3
5-9
10.0
10.5
10.2
-1.0
1.0
-k.I
of
5 %
BOD5
15-9
—
11.6
5.2
12.0
2.0
13.1
9-0
16.8
9.8
-1-5
5-9
-5.1
Final
PH
7.^
7-5
7.6
7.7
7-5
7.5
7-5
7.5
7-5
7.5
7.U
7.U
7.6
       Q
        Soluble COD and BODs  were measured.
           Initial values:  Soluble COD  = 1105 mg/1
                           Soluble BODs  =  ^90 mg/1.
        Activated carbon — Filtrasorb 300, powdered, alcohol/water
        washed, oven dried.

     Reverse osmosis was used to make a fourfold v/v concentrate of the total
mill effluent.  This solution was diluted with the permeate (fluid passing
through the RO membrane during the concentration process) to prepare 3:1, 2:1,
and 1:1 concentrations.  Jar tests were made with all concentrations, and
the samples were analyzed for suspended solids, total and soluble BODs, and
color before and after treatment with ferric chloride-polymer and alum-polymer
combinations (Figure 6).

     These data indicate that treatment with ferric chloride at 100 mg Fe3*
per liter and Hereofloc 812.3 at 0.75 mg per liter could produce an acceptable
effluent for discharge, in terms of suspended solids and BODs, up to concen-
tration of approximately 3:1 and reduce both hydraulic and gravimetric loads.
                                      32
-------
                                                                             -o
                                                                             



•V
f*





•<


:'•'.-
>

[

j
*
t
'."•"•'."..•
^2*

1
•'•••'.•'•
-------
With effluents more concentrated than 3:1, flocculating agents did not pro-
duce adequate effluents.

     While this was a rough study and practical means of processing were not
worked out — we did not study spill control, the effect of spills on water
reuse within the mill, or economics — it certainly indicates that wastes can
"be successfully treated at higher concentrations.

JAR TESTS ON WASTE STREAMS PRIOR TO PILOT SCALE TRIALS

     Mosinee Paper Company of Mosinee, Wisconsin and the Combined Locks mill
of Appleton Papers Division of NCR, Combined Locks, Wisconsin were selected
for the two pilot-scale clarifier trials.  Jar tests were made on the total
mill effluents.  These preliminary tests were followed periodically by other
jar tests while the field trials were in progress.

Tests on Mosinee Paper Company Effluent

     Samples from these jar tests were, in some cases, analyzed completely
for suspended solids, total and soluble BODs, organic carbon,-COD and color
(Tables 19 and 20) in order to measure the removals that might be attained
under different conditions of treatment.  Others were merely used to estab-
lish flocculating characteristics (Table 21) and/or sludge volumes that would
develop with the addition of the chemical combinations.

     In the early trials we established that alum at 200-300 mg/1 and ferric
chloride at 100 mg/1 Fe3+ could be used in jar tests to remove BODs ~ 35-55
percent (total) and 12-25 percent (soluble) ~ from the wastes.  The addition
of 0.75 mg/1 Hereofloc 812.3 appeared to provide additional removal of solu-
ble BODs:  removal increased from 13 to 35 percent when Hercofloc was added
with 100 mg/1 alum, and from 12 to 28 percent with 200 mg/1 alum (Table 19).
The use of the Hercofloc polymer alone at 0.75 and 1.5 mg/1 markedly in-
creased the removal of soluble BODs:  we found lU and 21 percent removal,
respectively.

     Lime as a primary flocculating agent and activated silica as an additive
(Table 20) appeared to be relatively effective (^4—16 percent) in the removal
of soluble BODs from the Mosinee mill effluent.

     Of great interest in these studies was the high removal of COD, color,
and total and soluble BODs that could be achieved with ferric chloride and
alum from the clarifier effluent, from this mill (Table 22).  Twenty-five to
forty percent removal of the soluble BODs components was obtained with these
primary flocculants added to either the north or south clarifier effluents.
Lime and "additional" Hercofloc 8l2.3 did not enhance the removal of the
"solubles" but did increase the removal of total BOD5, COD and color.

     Adjustment of the pH from the initial 11, or higher, to 8 with sulfuric
acid prior to the addition of the chemicals, markedly increased the amount of
suspended solids removed.  This only slightly affected the amount of soluble
BODs remaining in solution (Table 23).  The ferric ion was equally effective
as the chloride or sillfate salt when they were used at the same concentration
                                      34
-------
     TABLE 19.  JAR TESTS OF THE EFFECT OF FLOCCULANTS ON BOD5,
                  COD, ORGANIC CARBON, AND COLOR IN
                     MOSINEE CLARIFIER INFLUENT&
% Reduction in
Flocculant
Type
FeCl3


Alum


Lime


mg/1
50d
100
150
100
200
300
200
300
500
Final
PH
6.4
6.0
5-0
7-2
7.0
6.8
11.1
11.6
12.0
Total
BOD 5
51
37
58
49
56
56
22
39
41
COD
71
70
76
60
68
73
52
68
71
Soluble0
OC
31
44
47
14
24
31
-13
-1
1
BOD 5
13
25
24
13
12
12
-16
-14
-9
Totalc
color
52
43
89
24
43
47
15
58
11
Hercofloc
  812.3
             0.75
             1.5
         8.4
         8.2
18
19
 8
11
3
6
14
21
 0
15
Hercofloc 812.3 at 0.75 mg/1 was added after 1 min mixing with the
following primary flocculant:
Fed 3
10       7-0      No floe formation
25       6.8      No floe formation
50       6.4      56      72    30
                                                      30
                                                            47
Alum 50
100
200
7.4
7.2
7-0
No floe formation
54
52
76
71
14
21
35
28
18
32
a
 Initial control values for sample taken 11/21/74:

                 711 mg/1
                 215 mg/1
                 109 units.
                  87 mg/1
                 122 mg/1
 Total COD
       BOD 5
     Color
Soluble OC
       BOD 5
 Filtered through a 0.45 ym Metricel filter.
'rCo-Pt units at 465 nm (NCASI Method).
dAs Fe3+.
                                 35
-------
        TABLE 20.   JAR TESTS OF FeCl3,  ALUM,  AND ADDITIVES AS
             FLOCCULANTS FOR MOSINEE CLARIFIER INFLUENT*
          Floe cul ant
                                                Reduction, %
 Primary, mg/1  Polymer,  mg/1
        Final
         pH
                                           Total
                                                      Soluble
                                        BOD5  COD    OC   BOD5  Color
 Analytical  grade  primary
  chemicals:
           -.c
    FeCl3  50
               Here.



               Here.

               Act.Si

               Here.

               Act.Si

               Here.

               Act.Si
Commercial grade primary
 chemicals:
   Alum  50
         100
         200
                          0.75
                          1.5
 0.75
 1.5
10

 0.75
 1.5
10

 0.75
 1.5
10
5.8
5-8
5-8

7-4
7-5
7.7
7-7
7.1
6.8
7.1
7.4
7.4
7.2
7-0
6.9
38
37
38

11
10
 6
 5
13
15
10
11
23
23
21
23
56
57
58

-0.7
-0.7
 0.7
 4
 3
16
 3
 4
54
54
47
57
49
51
51

20
21

24
34
33

29
42
40
39
39
29
27
27

 9
15
10
 4
 5
 8
16
12
 8
 4
 4
 4
88
89
90

39
47
45
4o
69
69
72
53
86
88
87
84
FeCl3 50 c
Alum 100
Here.
Here.
0.75
0.75
5-5
6.5
32
20
50
9
28
13
—
alnitial values (for untreated influent control) for sample taken
 12/12/74:
   Total COD
         BOD 5
       Color
  Soluble OC
         BOD 5
                711 mg/1
                215 mg/1
                109 Pt-Co units
                 87 mg/1
                122 mg/1.
         .tx w 3   .-.__ •"Of -i- •
 Here., Hercofloc 812.3; Act.Sci., activated silica.
cAs Fe3+.
                                   36
-------
 of Pe3+  and when  the pH of  the  influent was not above  9.0  (Figure  7). Sludge
 volumes,  determined with separate jar tests and the Imhoff  cone, were also
 markedly  reduced by the  addition of the Hereofloc 8l2.3 (Table 2*0  and for
 this  reason it was included  in all but one of the studies at this mill.

          TABLE 21.  JAR TESTS OF THE EFFECTS OF PRIMARY FLOCCULANTS
                       AND POLYMERS ON pH AND CLARITY OF
                          MOSINEE CLARIFIER INFLUENTS
Flocculant
Primary
mg/1
Fe3+






Alumb



Alumc


Alum


>
25
50

100

150

100

200
300
200
300
1*00
200
300
1*00
Polymer ,
mg/1
„
— —
Here. 0.75
— —
Here. 0.75
— —
Here. 0.75
__ _«
Here. 0.75
— —
—
__ __
— —
—
__ — w.
— —
— —

PH
„
6.6
6.8
5-0
5.6
—
—
__
—
—
—
7.1*
7.1
6.5
7.0
6.9
6.5
1/16/75 ..
Claritya
__
0
0
0
3
—
—
__
—
—
—
1
5
1*
1
5
1*
Influent samples

pH
_.
6.9
6.6
5-7
5.6
U.5
4.3
9-0
8.6
8.1*
7-2
__
—
—
— —
—
—
3/5/75
Claritya
__
0
0
4
^
i
i
0
0
1
k
__
—
—
^_
—
—

pH
6.8
5.9
—
U.2
U.3
3.4
3.4
^^
—
—
—
_ 	
—
—
M_
—
—
3/11/75
Claritya
0
5
—
h
k
3
3
_ _
—
—
—
__
—
—
_._
—
—
Q
 Clarity:  0 = untreated control
         1-5 = partial to complete clearing of the supernatant after
               one hour settling.
 Analytical grade.
 Commercial grade with precipitate in suspension.
 Commercial grade decanted to remove the precipitate.

Tests on Combined Locks Mill Effluent

     Since Mill F had been one of those in the preliminary jar test studies
on the optimization of soluble BODs removal, only a few additional tests were
conducted prior to the pilot-scale field trials.   Data from these tests
(Table 25) substantiated the earlier findings (Tables 15 and 16)  that ferric
chloride, ferric sulfate and alum were all effective in removing  total and
soluble BODs from this waste stream.   We found, however, marked variations in
the amounts of removal with the different concentrations and combinations of
the flocculants used, apparently depending upon the amounts of spent cooking
liquor in the mill effluent.
                                     37
-------
TABLE 22.  JAR TESTS OF THE EFFECT OF FLOCCULANTS ON pH,
            BOD5, COD,  AND COLOR OF EFFLUENTS
                FROM MOSINEE CLARIFIERSa
Reduction. #
Flocculant
Type mg/1 pH
From North Clarifier:
FeCl3 5013 5.1
100 3.U
150 3.0
Alum 100 6.0
200 6.2
300 6.2
Lime 200 9-9
300 11.1
500 11.6
Hercofloc
812.3 0.75 8.8
1.5 8.6
From South Clarifier:
FeCl3 50b 5.7
100 3.8
150 3.1
Alum 100 6.9
200 6.5
300 5-6
Lime 200 11.3
300 11.6
500 12.0
Hercofloc
812.3 0.75
1.5
Q
Initial values (control)

BODs

61
63
6k
k3
61
6l
25
33
kl
12
15

63
65
59
No
60
62
No
k3
k3
No
Total
Soluble
COD Color

72
72
68
56
70
70
32
U8
55
k
11

68
70
6k
flocculation
6k
66
flocculation
kk
50
flocculation

65
79
85
k2
65
68
28
k2
k2
7

k6
70
75
27
k6
2
11

BOD5

39
ko
38
26
25
26
-lU
-19
0.8
-6
3

38
39
38
31
26
-13
3

of samples taken 11/21/7^:
Soluble BOD (mg/l) North=12l*
Total BOD (mg/1) North=2ll*
COD (mg/1) North=628
, Color (units) North=108
South=130
South=24o
South=6l2
South=63 .


  Fe
                             38
-------
               TABLE 23.  EFFECT OF pH ON JAR TEST FLOCCULATION
                         OF MOSINEE CLARIFIER INFLUENTa
    Primary flocculant
          (mg/l)
         Final     BOD5 reduction. %
          pH       Total     Soluble
                                  Reduction in
                                suspended solids,
Fe3+



50
100
150
200
10. 7
10.3
9-9
9.J*
-3
-5
1
—
-5
-3
6
8
22
55
72
50
    Alum
200
300
Uoo
500
600
10.8
10.6
10. k
10.2
10.1
-0.2
-7
 U
 2
 2
 1
 u
12
 7
 5
35
18
 2
15
30
    Influent adjusted to pH 8 with sulfuric acid prior to jar test:
Fe3*

Alum


50
100
200
300
Uoo
7.2
6.8
7-5
7.3
7.1
6
6
-9
-5
-8
11
U
5
-7
20
75
72
76
8U
80
     Influent sample of 2/11/75 had an initial pH of 11.2.

     The effect of these spent liquors and other mill discharges on the
overall efficiency of the clarifier is discussed in Section VII.

LABORATORY SURVEY OF SOURCES OF SOLUBLE BOD5 IN MILL SYSTEMS

     In a mill system with multiple sewers there are certain streams that
carry the bulk of the BOD and COD.  Some of the BOD and COD can be more
easily removed from concentrated streams than from diluted waste, and some
substances may alter the properties of others or interfere with their removal.
Therefore, isolation and treatment of these more concentrated streams would
be a logical way to improve the removal of BOD$ by the clarifiers and thereby
reduce the BODs load on the out-plant treatment system.  Importantly also,
the degree of recycling of the process water could be substantially increased.

Separate _Sewer Discharges

     In our study samples from the separate sewer discharges were obtained
from two of the mills and analyzed for various types of solids, for soluble
and total BODs, as well as soluble and total COD components (Tables 26 and
27).
                                      39
-------
     Figure 7.  In jar tests with Mosinee mill effluent, pH 8.5, ferric
                chloride and ferric sulfate at 25 and 50 mg/1 of Fe3+
                (a,),  (b_),  (d_), and (e_), precipitated more solids than
                when  they were used at a higher concentration, 100 mg/1
                of Fe3+ (c_) and (f_) .

     If^one or more of the high BOD5 streams (the discharges from the CM sew-
er in Mill F or the pulp mill, soda recovery or fifth stage brownstock washer
sewers in Mill K) could be treated separately by flocculation, reverse osmo-
sis, ultrafiltration  or some other process, the overall efficiency of the
treatment plant would increase, there would be less of a load on secondary
facilities, and there would be a greater potential for water reuse (details
discussed in Section VII).

Breakdown of Fibers

     Another source of soluble BOD5 was fibers: saccharides and other micro-
bial nutrients are released during the beating of pulp or during the recycl-
ing of fibers in water recovery systems.

     In order to study this release of soluble BODs, a beater test was ar-
ranged with a sample of unbleached sulfite pulp in a laboratory beater using

                                     40
-------
 a conventional 1-1/2 pound (0.68 kg)  test with 12 pounds (5.5 kg)  on the bed-
 plate.   One-liter samples were removed at half-hour intervals during the 90-
 minute  test period and freeness, total solids, total and soluble BODs,  and
 total and soluble COD were analyzed on each sample (Table 28).

         TABLE 2k.  VOLUME OF SLUDGE FROM MOSINEE CLARIFIER INFLUENTS
                 TREATED WITH VARIOUS FLOCCULANTS AND POLYMER

             Flocculant                Sludge Volume,     Final     Clarityb
  Primary, mg/1     Polymer,a mg/1          ml/l           pH         0-5


  Influent 12/21/74:

  Fe3+      50      None       —           122            5.U         k
            50      Here.     0.75          115            5-5         5

  Alum     100      None       —            6k            6.2         2
           100      Here.     0.75           5^            6.5         2

  Influent 12/31/71*:

  Fe3+°     50      None       —            92            6.3         k
            50      Here.     0.75           80            6.2         5

  Fe3+d     50      None       —            96            6.3         5
            50      Here.     0.75           85            6.3         5

  Influent 1/16/75 =
Fe3*



Alum

50
50
100
100
300
300
None
Here .
None
Here.
None
Here .
—
0.75
—
0.75
__
0.75
36
30
1*8
60
96
80
6.6
6.8
5-0
5.6
6.3
6.3
0
0
0
3
k
5
  fflerc., Hereofloc 812.3.
   Clarity:  0 = control (no apparent flocculation)
             5 = clear supernatant.
   Analytical reagent grade of Fed3.
   Commercial grade, l+2°Be, of FeCl3.

     Soluble BODs and soluble COD markedly increased as the beating progress-
ed and the fibers were fibrillated, releasing materials which became micro-
bial nutrients in the BODs test.

     We ran several flocculation tests on portions of the filtrate from a 90-
minute beaten sample (D in Table 28).  This sample vas passed through a 0.^5-
Um filter to remove fibers and insoluble BODs and COD components.  Ferric
chloride was again the best "single" flocculating agent (51 percent removal
                                     41
-------
TABLE 25-  REMOVAL OF BOD 5 FROM EFFLUENTS OF COMBINED LOCKS
             MILL BY FLOCCULANTS AND POLYMERS




Floe cul ant
Primary ,
d
FeCl3








Fe2 (SO i» ) 3







Ferri-floc

Alum



Lime

None
mg/1

10
25
50
50
50
75
75
75
100
25
37.5
50
50
50
75
75
75
75
75
50
100
200
300
200
200
—
?H3l2.3, Hercofloc
4/23 BOD5:

C4/24 BOD5:

dAs Fe3+.
Total
Soluble
Total
Soluble

Polymer , ""

None
None
None
H812.3
N73C32
None
N73C32
H813.2
None
None
None
None
H812.3
N73C32
None
H812.3
N73C32
H812 . 3
N73C32
None
None
None
None
H812.3
N73C32
H812 . 3
812.3; N73C32
= 274 mg/1
= 220 mg/1.
= 222 mg/1
= 194 mg/1.

mg/1

—
—
—
0.75
1.0
—
1.0
0.75
—
—
—
—
0.75
1.0
—
0.75
1.0
0.75
1.0
—
—
—
—
	
—
0.75
, Nalco





Effluent 4/23
BOD 5 removal, %
Total Soluble

No flocculation
No flocculation
— —
—
—
38.3 30.9
36.9 21.8
— —
42.0 30.4
No flocculation
No flocculation
No flocculation
— —
—
26.1 20.6
—
—
__ 	
— —
No flocculation
No flocculation
—
— —
No flocculation
No flocculation
—
73C32.





Effluent 4/24c
BOD 5 removal, %
Total

—
—
18.9
18.5
30.6
29-7
27.5
26.1
—
—
—
36.9
26.6
29.3
22.5
38.7
32.0
26.6
20.7
—
—
8.1
25.2
—
—
. 5.4






Soluble

—
—
15-5
14.4
26.3
20.1
23-7
18.6
—
—
—
27.8
16.0
20.1
9-8
28.4
35-6
14.4
15-5
—
—
13.9
10.3
—
—
-6.7






                            42
-------
















cd
M
S
Pn
O
CO
EH

§
1
H
13
M

in
Q
S

s
*-s
o
o
o
3

o
EH

O
S
«3j

H
HH1
ffl
^
O
CO

p^
o

CO
o
K
s
co

.
CM

E3
t-3
 t— LA \O VO
• • • • •
t— t— t- t— ^-


LA
OO OJ ON ^J-

00 H H O t—

1 ^
!H O CU -P
1> IM ft ^ C

CU 
-------














CO
o
H


ft, ^-^
0 W

CO h3
CO M
X S
n3 •— '
S i-3
^a
§B
CO S
£ CO
P-i 0
^ s
n
EH
A
*-3 r^
 W
1 — I EH
O M
0 ps
O CO
PP  CU
•d
^ a
Digeste
foul co


^^f
oo
H




vo
oo
vo





o
o
OJ





LA
o
t—
oo


t—
t—
o
CM



CM
VO
H




OO
00
00





H
•
O
H



•
1

CO

^s
!H
CU
0
O
CU
?-l
cc)
•d
0
CO


^*O
oo
vo




J-
o
o
H




CM
ON






vo
o
H


t—





H
OO
H




O
OJ
o\






o
•
OO





+3
a
cu
^f
f— \
*w
«
•H
^4
(U
Clarifi

































cu
N
•H
CO

 O
-P Pi
•H ' S
Vi -— - 3.
H O LA
CD ,JCJ
o P 'd
•H CU C
JH S OJ
-P
CU M LA
S co -3-
^J •
s o o
5. n-,
J? g °
• S co
0 fn
LA 0)
oJ vo -p
-=!• H
^H -H
M -p 4H
P CO
O rt
^ CO O
^3 -P .
-P -H Tf
S
fcjO 3 -p
•H -P CU
CO PH H
CO 1 H
0) O O
P-i O O
05 ,0 O
44
-------
of  soluble BOD5)  (Table 29).  We added cellulose powder (500 mg/1, Whatman,
Standard grade) to provide particles of high surface area and enhance floc-
culation (No. 3 series in Table 29); it slightly increased the removal of
soluble BODs  (l-9 percent).  At pH 7-7 with alum, the removal was 2 percent
better than at pH k.2 (2A versus 1A).  The addition of cellulose powder and
a pH of 7.7 further increased (to 9 percent) the soluble BODs that was re-
movable with  200 mg/1 alum. A combination of cellulose, lime and alum at pH
10-7 (^C) was 23 percent better than lime alone at pH 11.8 (1C) or lime and
alum without  cellulose (2C) at pH 10.7 (29 percent increase).  Additional
work on the effect of these high BODs streams on the efficiency of clarifiers
was done later in the study and is discussed in greater detail in Section
VII.

                TABLE 28.  RELEASE OF SOLIDS, SOLUBLE COD,
                           AND BOD5 BY BEATING9"
Sample
no.
A
B
C
D
Beating,
min
0
30
60
90
Freeness,
ml5
750
370
100
30
2U Hr
solids ,
6/1
0.29
0.32
0.35
0.38
Soluble COD
Increase,
mg/1 %
33^
396 18.6
1*16 2l*.6
1*62 38 . 3
Soluble BODs
mg/1
77
77
118
128
Increase,
__
0
53.2
66.2
Q
 1-1/2 Ib beater method with 5-5 kg load on the bed plate.

 Canadian standard.

ALTERNATE METHODS FOR REDUCTION OF BOD5

     Ultrafiltration is a process for the separation of soluble low and high
molecular weight (or size) components into two separate streams through the
use of semiselective membranes under low pressure (less than 300 psig or 21
kg/sq. cm).  We used Ultrafiltration in two trials with effluents from Mill F.

     In the first trial a sample of the total mill discharge to the clarifier
was processed through a Westinghouse module using a polysulfone membrane that
had high pH and temperature limits (pH 11.5 and 95°C).  Removal of 59-65 per-
cent of the soluble BODs was attained in the permeate (that portion of the
solution passing through the membrane).  The permeate apparently contained
low molecular weight components, such as acetic acid, that had not been re-
moved by flocculating agents (Table 30, Part A).  The flocculating agents,
lime with the polymers Nalco 609 or Dow PAA (polyacrylamide) did not remove
additional soluble BODs from the permeate of the total mill effluent.

     The second trial (Table 30, Part B) was conducted with a sample of the
CM sewer effluent, which had been shown in previous work to contain 33 per-
cent of the soluble BODs in 18 percent of the flow going to the clarifier.
                                     45
-------
Ultrafiltration without additives removed 69 percent of the soluble BODs, and
activated carbon did not remove any more BODs.  Apparently the membrane was
the controlling factor in the removal of soluble BOD.  The carbon did not ad-
sorb any of the transferrable components probably because its adsorptive ca-
pacity was used up by larger, more heavily charged molecules.

           TABLE 29.  JAR TESTS OF VARIOUS FLOCCULATION TREATMENTS
                    OF FILTERED 90-MINUTE BEATER SAMPLES3"
Treatment and Flocculant
sample Type
1. Flocculant
added:
A Alum
B Fed 3
C Lime
2 . Above samples
treated with
alum and
flocculated:
A Alum
C Alum
3. Cellulose powder
(500 mg/l) added
to No . 1 samples :
A Alum
B Fe013
C Lime
D Control
4 Alum added to
No . 3 samples :
A Alum
C Alum
mg/l

200
200
200

200
200

200
200
200
None

200
200
Soluble COD
Soluble BODs
Reduction, Reduction,
pH mg/l % mg/l %

4.2 458 2.1
2.7 256 45-3
11.8 449 4.0

7.7 336 28.2
10.7 449 4.0

4.2 400 14.5
2.7 248 47-0
11.8 422 9.8
468

7.7 355 22.0
10.7 4o6 13.2

105 0.0
51 51.0
98 5.8

102 1.9
109 0.0

95 8.6
50 51-9
90 13.5
104

88 15.4
74 28.8
Q
 Initial values:  pH = 6.7
         Soluble COD = 462 mg/l
        Soluble BODs = 128 mg/l.
                                     46
-------











EH


Q
HH
EH £3

a w
5 K

« o
o a
o 
CO O
o
o
0 fo
on
W
^
?Q
EH














If!
P
§

0)
H
H
0
CO




P
o
o

0)
H
fl
H
O
CO



*
fl
o
•H

t) TSS.
TJ
0)
K

H
S

c
o
•H
-p
O tsSi.
jj
*i3
0)
ffi


H

M
d



0)
H
ft

CD
CO




-p
ffi -p £>OS -P
ct? o5 CD cd cd ccJ CD
d) d)d)" CUCJ (U |~^ J^J ^) J — 1 ^ JT^ ^- ^J

pf | pT-[
h3 tlD


_f~^
LA O <"
• • 0)
OOO O O O pt<
1 O H 1 O 1 1 | [ 0 0
i on i CM i i i i CM H en
CD*
•^^p


o\
O rr}
+ ^ ^ 1 g
O 0-1 f> ,Q m

O.HS O Ot)O C5 O OO *^^ d)
-p
C_j J^
0) 0) ^
•H 3 d)

•H 
-------
     The ultimate value of ultrafiltration might be in concentrating the high
molecular weight soluble BODs components and removing them from process
streams.  It has been demonstrated elsewhere (2l) that ultrafiltration could
be carried to a concentration of 25-35 percent solids without complications
from osmotic pressure buildup which directly affects RO flux rates and con-
centrations.  This would be sufficiently high for final disposal by discharge
into the strong pulping liquor streams for evaporation and utilization or
burning.  Still another use for ultrafiltration is in reducing the buildup of
high molecular weight solubles in recycled process water within the mill;
this could reduce water consumption and the volume of overflow to the out-
plant clarifier.

     Reverse osmosis, similar to ultrafiltration but operating at higher
pressures (UOO-800 psig or 28-56 kg/sq cm) with tighter membranes which re-
ject smaller molecules, was tried both as a possible treatment method and as
a further refinement for the permeate from the ultrafiltration process (see
Table 30).  The soluble BODs removal of 89 percent for reverse osmosis alone
was increased to 92 and 9^ percent, respectively, by either the ultrafiltra-
tion pretreatment or the addition of activated carbon to the reverse osmosis
feed.  The advantageous use of activated carbon in the reverse osmosis feed,
which had been pretreated by ultrafiltration, substantiates the previous con-
clusion that the high molecular weight materials of the ultrafiltration feed
were utilizing or plugging all of the available adsorptive sites on the car-
bon.

     Ferric chloride was added to the ultrafiltration feed in order to de-
velop a floe and possibly increase membrane rejection and soluble BODs re-
moval.  Ferric chloride actually worked in reverse and increased the transfer
of soluble BODs through the membrane.  This loss of membrane rejection re-
duced the removal of the soluble BODs components by h percent.  Subsequent
tests indicated that the ferric ion did not damage the membrane:  after the
module was washed with detergent, it gave salt (Had) rejection comparable
to those for a new membrane.
                                     48
-------
                                 SECTION VI

                        PILOT SCALE CLARIFIER STUDIES
GENERAL

     The information from jar test studies of the effluents from the Mosinee
and Combined Locks mills were used to prepare a scheme for the treatment of
these mill effluents with a pilot-scale clarifier.

     This clarifier had been developed at The Institute of Paper Chemistry
under Project 3029, jointly sponsored by member companies of the Institute,
the Environmental Technology Corporation (ENCOTEC) and the Upper Great Lakes
Regional Commission (12,).  The stainless steel cone of the clarifier had a
center height of 1.3 meters (4.25 feet) and a diameter of 1.1 meters (3-5
feet).  The inlet was downflow through a 5-1-cm (2-inch) stainless steel pipe
with an opening ^0 cm (1.25 feet) from the bottom of the cone.  It was equip-
ped with an overflow weir which discharged into a 5-1-cm (2-inch) trough
around the entire circumference of the top.

     This section of the report covers the initial laboratory "shakedown"
studies, the on-site studies at the mill in Mosinee, Wisconsin from January
15 to April U, 1975> and the on-site studies at the mill in Combined Locks,
Wisconsin from April 8 to June 26, 1975-

     The problems encountered and some of the solutions developed for the
correction of these problems are discussed.  Data are given and discussed for
the removal rates attained for suspended solids, total and soluble BODs and
color.

TRIALS WITH THE PILOT CLARIFIER UNDER LABORATORY CONDITIONS

     Three thousand gallons of Mosinee clarifier influent were delivered by
truck to the Institute early in January 1975 and were used to check the oper-
ation of the clarifier system (Figure 8) without the samplers or the submers-
ible pump (P-l).

     The system consisted of the following items and arrangement:

     1.  A tank (T-l) provided a constant supply of waste to be treated.
     2.  A feed control pump (P-2) maintained a constant feed rate.
     3.  A 35-liter (9-gallon) reactor tank (T-2) with a high speed
         mixer provided rapid mixing of the waste with the primary
         flocculating agent and permitted gravity flow to the balance
         of the stream.
     U.  The flow from the reactor tank was into an open pipe extending
         to the bottom of the flocculator tank (T-3).   If polymer was
         to be added, it was introduced at the top of this open pipe,
                                     49
-------
                        o
                        CM


                    T3  'd

                     
                     bO O
                        H
                        CO

                     I   I
                         O  O  *H

                         i ^
                                           PH
                     i   i    i   i   i   i   i
                            -P  -P

                         O  cd  9

                        -P  P  3 H
                         C!  O  O  a3
                         O  O  O  >
                         O  O  O  O
                            rH  H  d
                            -rl  H
                                O rH
                                PH CQ
                                                     a
                                                     o
                                                    •H
                                                    -P
                                                     03
                                                     cl
                                                    H
                                                     a
                         (U
                         H
                         aJ
                         o
                         M
                          I
                         ft


                         O

                         O
                        •H
                        -P
                                                   ao
                                                    bO
                                                   •H
50
-------
         into a quiescent zone approximately one-quarter of the dis-
         tance down the pipe in order to provide adequate mixing with-
         out excessive turbulence or floe dispersion.   In the center
         of this tank was mounted a slow speed (20 rpm =1.6 fps periph-
         eral velocity) paddle that mixed without excessive shear.
     5.  The overflow from the flocculator tank passed downflow into
         the clarifier (l-l|), with the flow from the weir trough into
         a small sampling pot (T-7) and then to the sewer.
     6.  In these laboratory studies the sludge was removed from the
         bottom of the clarifier by gravity flow through a vented
         overflow pipe (not shown) that could be raised and lowered to
         control the rate of flow.  Both gravity flow and a pumping
         system were used in the field trials and are discussed in  the
         following part of this section.
     7.  The flocculating agents were stored in small tanks (T-5,6) and
         were metered into the system with tubing pumps (P-3»M.

     We used this equipment to do a brief laboratory study of the effects of
waste flow and sludge removal rates, at a constant level of flocculating
agent(s), on the quality of the discharge produced.  An excellent quality
discharge (Table 3l) appeared to be produced if the waste feed and  the sludge
discharge rate resulted in an overflow rate of 1^0-150 liters per square
meter per day'(^00-^25 gallons per square foot per day).  The removal also
appeared to be dependent upon the type of sludge blanket developed  in the
clarifier.  Best results were obtained with the level at least five centime-
ters (2 inches) from the top to prevent floe carryover, but not less than one
meter (3 feet) from the bottom; i.e., 60 cm (2 feet) above the waste inlet.
This arrangement seemed to promote the development of larger floe particles
and a filtration effect to markedly enhance the quality of the discharge.
These parameters were, therefore, selected for the operation of the unit
during the field trials.

EQUIPMENT AND OPERATION AT MOSINEE PAPER CORPORATION

     The IPC system was to be operated in parallel with the two mill clari-
fiers and was, therefore, installed in the waste treatment plant in a con-
crete building.  This building was immediately adjacent to the two  commercial
clarifiers and was unheated, except for heat from the incoming waste stream.

     Due to the arrangement of the waste stream flow pattern, it was neces-
sary to use a submersible pump (P-l in Figure 8), mounted in the waste treat-
ment trench immediately downstream from the bar and traveling screens, to
lift the waste to floor level (2-2.5 meters or 6-8 feet) at a rate  of 19-30
liters (5-8 gallons) per minute.

     The flow from the submersible pump was directed into the bottom of a
190-liter (50-gallon) stainless steel tank (T-l) for supplying a centrifugal
feed pump (P-2) with the excess overflowing to the sewer near the influent
sampler.  This tank also contained the pH recording equipment.

     The feed pump (P-2) maintained a waste flow of 11.U liters per minute
(3 gpm) into the reactor tank (T-2)} and the only other modification of the


                                     51
-------
equipment, over that described in the laboratory trials, was the installation
of a pump in the sludge removal line during some of the trials.

         TABLE  31.  LABORATORY EVALUATION OF PILOT-SCALE CLARIFIERa
Over-
flowb
rate
369
395
406
425
526
579
58?
632
Sludge
removal,
at
1°
20
13
10
10
15
12
10
Ik
Flocculant, mg/1
Primary
Fe3+
87
6k
61
52
51
51
50
50
Polymer
H812.3
1.3
0.78
0.77
0.77
0.77
0.77
0.77
0.75
Suspended
solids
99
99
99
97
19
-52
-29
-35
% Removal of
Color
55
79
—
k6
49
k9
48
k9

Total
2k
3k
29
36
19
5
0
9
BOD 5
Soluble
8
17
9
10
13
16
0
9
   alnitial (control) values:   Suspended solids = 234 mg/1
                                     Total BOD5 = 134 mg/1
                                   Soluble BODs = Ilk mg/1
   ,                                       Color = 130 units .
    Overflow in gallons /square foot/day based on surface area of clarif ier
SAMPLING AT MOSINEE PAPER CORPORATION

     During the first part of the mill study, influent and effluent samples
were taken with two Model WM-5-24R refrigerated Sigmamotor samplers (Sigma-
motor, Inc. of Middleport, New York) set to take individual samples hourly
around the clock.  In order to reduce the analytical load, these were com-
posited into 12-hour samples daily.

     It soon became apparent that the rapid changes in mill discharge (Table
32 and Figure 9) would not permit such an infrequent sampling scheme.  A
third sampler, a CVE refrigerated unit (Quality Control Equipment Company of
Des Moines , Iowa) capable of taking small samples at short intervals (l-60
minutes) for compositing in a 4-liter (l-gallon) bottle, was placed in the
influent stream.  This unit was set to take 4o-ml samples at 15-minute in-
tervals and the 12-hour composites were compared with those from one of the
Sigmamotor units taking 4oO-ml aliquot s from the same influent.

     Poor correlation was evident (Table 33) and a second trial was made.
This was with four samplers in the following arrangement :

     a.  The two Sigmamotor samplers on the influent and effluent streams
         set at 30-minute intervals , taking 2k 400-ml samples in 12 hours .
                                     52
-------
         The CVE sampler on the influent line taking a 40-ml sample every
         10 minutes.
         A polystaltic tubing pump (Buchler Instruments of Fort Lee, New
         Jersey) set to take 9 ml per minute continuously was used on the
         effluent line.

                  TABLE 32.  ANALYSIS OF INDIVIDUAL SAMPLES
                            TO ESTABLISH RANGE3
Milligrams /liter
No.
1
2
3
k
5
6
7
8
9
10
11
12
13
PH
10.12
10.07
9.57
9.63
9.11
9.kB
8.86
9.kO
9.19
9-30
9.78
9.60
9.59
Suspended
solids
210
628
k9k
1^5
85
Ikk
265
152
97
76
9k
115
267
Total
BOD 5
2kl
206
166
155
129
135
1U3
163
ihB
127
138
132
lU8
Soluble
BOD 5
203
178
138
138
116
114
lUo
iko
ikO
117
135
12k
118
a_ . ,
                1/16/75).

     Analytical comparisons were excellent (Table 3^) and indicated that the
Sigmamotor samplers could be used to take representative samples if the 30-
minute cycle was used.

     In order to compensate for the "system holdup time" of 90 minutes, the
samples were composited to obtain representative samples from the same 12-
hour test period.  This was readily accomplished by compositing the individu-
ally numbered bottles in groups so that those from the effluent were three
bottles "behind" those for the influent.

     Samples of the mill clarifier influent and effluent were taken at 7:30
a.m. daily by the mill staff and analyzed by both mill and IPC personnel.
                                     53
-------
                                           CM  ft
                                                    -P
                                                    a
                                                    
-------




















CQ
[-3
Q_J
s
H H
ll
pn.
• O
oo
ro

W
9
§



























o
u
^J
CM


























r-l
^ft?I
^y













































M
O
, |
0
o







in
Q
0
m
(!)
rH
,0
3
t-H
O
Crt








ul
§
PQ

r-H
CO

O
H






CQ
•a
•r-l
r-H
o
en
T3
CD
•0
0)
p.
CQ
CO







EC
a














Cfl
r-l
3
60
r4
o
0)
p.
CO



i-l
cfl
rH
3
00
0)

i~H
cfl
•H
O
fl\
U/
W




.^
cfl
rH
§,

00
iH



CJ\
rH
CM





m
00
r-




CM
CM

"^






rH


0
CO
i— 1




m
rH
r-t




00
CO
I— 1



CJ^
0
r-H







^~
^j-
CM



r-
r~-








rH



CM
00
rH





in
i~»
oo




CN
00

"^






CM


CO  CT>
CM O\
i-H i-H



rH 0
CM t-H
CM CM





CO rH
r-- m
r-- oo




m o
0  C
(\) d>
rM
?H 0}
(D -P
H
1 H
tn ft
a
W CO
P> to
r ^
H
nj a
-C] O
-P 0


-P cfl 4)
•H H
to p ^
0 bD V
ft 0) H
a n ft
o a
en tn

0 0
-• A -P
cu o
H CM H
ft H CO
a s

tn o -H
«H CQ
H
cO C cd
•H -H
O £5 ,^"1
0) -P
p4 LT\ *H
CD r-l r*
0}
55
-------



















P4
o
CO
P*3
)-J
Q-J
Sj
*3j
9 OJ
|~> •
O O
H S
PS

p <3
S M

§H
g -
w w
P j ^>
W J
h
PH s
O M
CO K
H M
W M
>H fc
^ S
§ 5
0

_^
oo
W
i_^
PQ
EH























'c?
•H
C


o
o

4J
P-i
O

O
O


rH
~5|
a
m
n
O
PQ

01
t— 1
F^
O
CO

rH
rH CO
S3 ^
h
to
rH CO m
D T^ vJ"
00 CM r-H
01
PS

rH
•H oo in
O vO ON
01 CM
a
CO
rH
rH ON
•H -3"
S rH

to
<-<  •$
3 d* vo CN in 04 in
i— «


^D
1 rH 1
rH

r-. co m <|- oo \o
CO CX) CO CO C^ 00
i— f t-H rH





^ O CN O ^H O
iH O\ sj- rH O ON
iH rH iH r-l







i r*. i
rH

CM kO lT» O\ *^ ON

oo -d* co o rH m
,
p
J-J
cu
co cd
•H P
CO
td co
,0 CD
H
h ft
fi to
rH 9

CO O
^J-
n G_I
^ °
0) 0)

•H -rH
01 CO
O 0
ft ft

O O
o o


J^
CD
P
(3
•H
p

B
•H

T
o
j__i

_p
CO
01
H
ft
CO

ES
o
CO

^1
B
01
•8
p
CO
01
H
• ft
•P «
£3 CO
0) CO
ft | Ed

p
g
0
JH
«H

a

"O
\
^
OS
ft

ft
•H
p
H
CO
CO
^
H
o
ft


>>
c
0)
•3
p

0)
H
ft
1
U]
^J
o
C
•H
-P

O
,

CQ
pq






















































5,2 i
O O -H
O O CO
• • •• O
t— t— ft
n
-P P 0
ass

Ctj ft CO
O O O
o o oo
.. .. ..


00 -J H

08 08 >rH
H OJ

B

»^- i?
""*
56
-------
     For analysis all samples for BODs were stored at it°C (ifO°F) and the
others were maintained at room temperature.

TESTING SCHEDULE AT MOSIWEE CORPORATION

     In order to establish a comparison between the mill and IPC clarifiers,
the first trial was the operation of the pilot unit under conditions as near
those of the mill as practical.  Since we were not attempting to develop op-
timum clarification at high hydraulic loadings, but merely to study the in-
creased removal of the soluble BODs from the waste stream while operating
under different chemical treatment conditions, it was not necessary to show
that we could exactly duplicate mill operation.  Rather, it was more impor-
tant to show that any increase in BODs removal was due to the chemical addi-
tives and not to the clarifier design alone.

     We developed a "base line" by operating the IPC unit with Hercofloc
812.3 at O.T5 mg/1 and with a waste flow rate of 11.k liters/min (3 gpm) in
parallel with the two mill clarifiers.  The Hercofloc addition duplicated
mill treatment.  If a 10 percent sludge withdrawal could be maintained, the
pilot clarifier overflow would be 1.6 liters per sq, cm per day (UOO gallons
per sq. ft per day), approximately one-half that of the commercial unit.

     The efficiency for removal of BODs in the small unit was slightly less
than for the mill units under the same conditions and time period.  This in-
dicated that the design of the IPC clarifier did not promote more removal and
that any increases found in later trials using chemical additive would be di-
rectly attributable to the chemical treatment.

     Following the development of the base line, other chemical treatment
schemes based on the laboratory studies were tested:

     A.  Base line using 0.75 mg/1 Hercofloc 812.3-
     B.  Alum at 200 mg/1 plus 0.75 mg/1 Hercofloc 8l2.3-
     C.  Ferric chloride at (100 mg/1 as Fe3+) plus 0-75 mg/1 Hercofloc
         812.3-
     D.  Ferric sulfate (50-100 mg/1 Fe3+) without the polymer.
     E.  A second base-line study with the Hercofloc.
     F.  Sulfuric acid for pH adjustment only.
     G.  Sulfuric acid for pH adjustment plus the Hercofloc.
     H.  Ferrous sulfate (100 mg/1 Fe2+) plus 0.75 mg/1 Hercofloc.

     The original plan was to have tests of two weeks' duration.  Unfortu-
nately, the several periods of problem and solution development we encoun-
tered early in the study delayed the program and resulted in shortened test
periods (for F, G, and H, above) near the end of the study.   Also, throughout
almost the entire trial period, the Mosinee mill was on a reduced work week
(h days), which limited the operation of the small clarifier to the period
between 9:00 a.m. Tuesday and U:00 a.m. Saturday, thereby limiting the number
of samples per week.
                                     57
-------
RESULTS AND DISCUSSION OF WORK AT MOSINEE

     In our first attempt to establish the base line, we immediately encoun-
tered three problems:

     The first problem related to the submersible, centrifugal sump pump in-
stalled in the waste trench.  It had a small 2.5-cm (l-inch) inlet hole in
the baseplate and due  to the high fluid velocity in the waste trench, an
"aspirator" effect developed across the inlet when the pump was vertically
mounted, and we were unable to pump sufficient waste.  Mounting the pump at
a 90° angle pointed downstream increased the aspiration effect, while mount-
ing it pointed upstream resulted in rapid plugging of the inlet hole with
pitch and debris not removed by the bar or traveling screens.  This problem
also extended to the centrifugal pump used for feed control.

     The second problem was due to the high suspended solids content of the
Mosinee waste.  A centrifugal pump was the only available pump that was con-
sidered suitable for control of the flow at the 11. ij- liters per minute (3
gpm) rate.  A one-half inch Eastern centrifugal pump was, therefore, used to
control the rate of waste fed to the clarifier unit, and at first the rate
was controlled with a 1/2-inch valve on the pressure side of the pump.  This
rapidly plugged with pitch and fibers and resulted in erratic flow patterns.

     The third problem developed in the gravity sludge removal system.  Set-
tling of the high suspended solids content rapidly overwhelmed the available
space in the clarifier and resulted in solids carryover, even in the absence
of chemical agents, other than the Hereofloc 812.3.

     In order to reduce these problems:

     1.  A conical screen (20 mesh), 18 inches long by 7 inches in diameter,
was installed on the submersible pump.  When installed with the apex of the
cone mounted upstream, the pumping action was excellent and the cone was
somewhat self-cleaning.  When mounted with downstream orientation, which
should increase the self-cleaning aspect of the cone shape, the flow vas re-
duced below acceptable levels.  Periodic cleaning, with high pressure water,
was required to maintain an adequate supply to the system.

     2.  The centrifugal feed pump was fitted with a smooth-bore orifice
plate for flow control and with periodic cleaning this provided better flow
control.

     3.  A multihead tubing pump (Brosites) was installed in place of the
gravity system for sludge removal.  In the last four trials a throttled cen-
trifugal pump was tried in this position and found to work the best of any
of the schemes attempted.

     Although none of these worked perfectly and problems were encountered
from time to time, failure of the system could be minimized with careful
cleaning and operation procedures.
                                     58
-------
     Data, for the trials with the various flocculating agents and for the
base lines are summarized in Table 35.  The more detailed analysis of the in-
dividual samples are in Appendix B (Figures B-l to B-10, Tables B-l to B-5).
These data were first used to establish confidence limits and the following
summary was developed.

     1.  Both the initial and final trials (A and E of Table 35) with 0.75
mg/1 Hercofloc 812.3 indicated, as previously noted, that the IPC unit was
equal to or slightly less efficient than the mill clarifiers in removing sus-
pended solids, soluble BODs and color.

     2.  The reason for the marked difference noted in the total BOD5 reduc-
tions for the mill clarifier during the second trial is unknown; throughout
the study period there had been a slow, progressive increase in the effi-
ciency of the mill clarifier in removing BODs; i.e., from lk to 35 percent
between Trials A and H.  This was not readily traceable to changes in the
concentration of BODs going to the clarifiers during this period, since these
were on the average l6l, 120, 138, l8l, l6l, 211, 11^ and 1?8 mg/liter for
Trials A to H, respectively.

     3.  Alum with polymer and without pH adjustment, as predicted by the
laboratory trials, did not increase the removal of the BODs components but
did markedly increase the efficiency of removal of suspended solids and
color.

     h.  The ferric ion, both as chloride and sulfate salts, was an excellent
flocculating agent for removing BODs, "the chloride form being more effective
than the sulfate for suspended solids.  The chloride could have been better
because the final pH was nearer to the optimum for the Fe3  produced floes
(pH U-5).

     5.  Ferrous ions, at the level of 100 mg Fe2+/l, appeared to reduce the
efficiency of the system even with the addition of polymer.  While this level
of ion was below that which would be theoretically required by the Schulze-
Hardy rule, which states that the concentration of the counterion required to
collapse the double layer and produce rapid flocculation is inversely propor-
tional to the 6th power of the valence, it was the highest level practical
during our trials due to the low solubility of the commercial grade ferrous
sulfate.  Based on the effectiveness of the ferric ion at 100 mg/1, the level
of ferrous ion theoretically required would have been approximately 1100
mg/liter.

     6.  The use of sulfuric acid to decrease the pH was one process under
consideration by the mill for use in the large clarifiers.  A trial was,
therefore, designed to test this procedure for the removal of some of the
components from the waste stream in the small system; but due to the mechani-
cal difficulties we had encountered earlier in the trial period, only a short
period was available near the end of the study.  Although the data were in-
sufficient to provide reliable comparisons with mill operation, laboratory
studies have shown that pH reduction to below ^.5 resulted in marked reduc-
tion in suspended solids and total BODs, with little effect on the soluble
BOD5 level in the effluent.
                                     59
-------
















E"i
£H
fV]
5
j
P*H
prj
W

M CO
H PH
S F*""!
H
W P^c
W H

H  ^
H
LA
on

1

,3*
EH
























V*1,

•
H
cd

o
a
CO












O
H
0
0



co
H
p

H
O
f> CO
R
o
PQ
, — |
cd
-P
O
EH



0)
Tj M
S3 "d
0) -H
ft H
co O
2 CO
CO





rl CO
O -P
H -rl
O S3
O P




£>,
[ T
•H
H
•i «
a fi

jj pQ
S3
0)
4n
„
CD
H H

^ t*o
,— ( ^
o
CO



«\
P bp
o a
EH
W

-d
CU •>
"d co
S3 -d H
CU -H —
ft H M
CO 0 8
p! M
CO


Jn

O cd
o
oj oo
•
^3 OJ
-P H
•H CO
>
O
•P O
•H H
p! 0
CJ
O J-i
PH 
CO J-


c{
p
•H
£j

CO t— 4J
H H -H
rt
3

H
H
•H
cj
CTscO
H H -d
S3
cd

n
O
o
H
>
5^
•HO
O
-P 0
•H H
P O
0
0 h
PH 0>
H W

•
P)

CO OJ








-•3" O
OJ H







-3- \D
OO OJ







CO VO
t~-vo










co vo
_=)- LA
H





_^j* V^>
O\ H
H





LA OO
O OO
H H






-3 VO
-X MD





H
0 H
PH -H




£. 	 (Y")
H







CO O
H




• •
H
'-v*
bD 00 O\
a H oj

LA
t—
•
o
-p
0)
O CO
OO CO t—
•
OJ
H
CO

O
o
H

00 H H H A -P •H CO H 0 VD CO CO cd S3 0 H 0 OH 0> PH -H CO H S • W • CO S3 O to •H cd ft o O •d 0 O bO (0 •H > 0 ft o -p cd i -p S3 CO •H O •H CM "g CO S3 H W 08 ft e> r* W 60

-------
     These trials, with ferric, ferrous, and aluminum ions, and especially
with the iron salts, have shown that an increase in clarifier efficiency (12
percent for total BODs, 15 percent for soluble BODs and up to 66 percent for
color) could "be attained with the Mosinee waste.  The sulfate form might "be
preferable to the chloride, particularly if the sludge were to be burned, be-
cause it is less corrosive.

     One of the principal practical observations we made in the field study
at this mill was on the effect of strong digester room spent liquors and
washes on clarifier performance.  Visual, analytical, and other evidence
showed that heavy slugs and spills of these wastes — some of which lasted
several hours —  adversely affected clarification.  A single program of grab
sample collection was undertaken to help indicate the sources of strong pro-
cess waters.  The analytical data from the study of that one set of grab sam-
ples are summarized in Table 27, previously described, and showed the rela-
tive strength of the various soluble materials in these samples.  The effects
of these streams will be discussed in greater detail in Section VII.

EQUIPMENT AND OPERATION AT COMBINED LOCKS, APPLETON
  PAPERS DIVISION OF NCR

     The clarifier equipment that had been used in the Mosinee field trials
was cleaned and transferred to Combined Locks on April 7> 1975-  Since it was
again to be operated in parallel with the mill system, it was placed in the
treatment plant building which was immediately adjacent to the two mill clar-
ifiers.  A tee was installed in the main influent line to these two clarifi-
ers and this stream was fed into the 50-gallon day tank (T-l in Figure 8).
This eliminated the need for the submersible pump.  At Mosinee we had encoun-
tered some difficulty with uneven flow over the weir of the small clarifier.
Although the unit had three leveling legs, the top rim was not level around
the entire periphery.  Spot overflow occurred; the floe subsequently "stream-
ed" at these points and efficiency decreased.

     An adjustable leveling device for the top of the clarifier at Combined
Locks was made of small strips of slotted plastic installed along the entire
overflow edge.  Carefully manipulated (up and down), these could be used to
establish an overflow around the entire circumference.  This helped to main-
tain a smooth top on the sludge blanket and virtually eliminated spot over-
flow except when the strips were disturbed during the weekly cleanup.

     Two other basic changes were made for operation at the Combined Locks
mill to provide better control.  The high fiber content of this waste stream
overwhelmed the sludge disposal system on the IPC clarifier.  In order to re-
duce the fiber content and still retain in the influent the fine particles
required for good floe formation and BODs removal, we installed a side hill
screen with 60-mesh wire over the top of the day tank.  Later, when we en-
countered difficulties in maintaining a clean surface in the base-line study
(A), we changed this to a Hydrasieve (C. E. Bauer, Springfield, Ohio) with
0.010 inch slots (Figure 10).

     A centrifugal pump with electrical speed control was used to control the
sludge level in the clarifier and the hose was clamped on the outlet side.
                                     61

-------
This was changed to run at low speed with the hose clajnp removed, and the
time of operation was controlled with a timed OW/OFF switch (Trial D).  This
permitted relatively close control of the sludge blanket level that had var-
ied both with the "quality" of the waste fed to the system as well as with
the type of flocculating agent.
        Figure 10.  Bauer Hydrasieve used to remove the longer fibers
                    from the Combined Locks mill effluent
                                    62

-------
     Since the level of the sludge blanket in the clarifies was apparently
very Important to the efficiency of the system, thought should be given to
provision of a sensing type of control system for any future work with this
unit.

SAMPLING AT COMBINED LOCKS

     The two Sigmamotor samplers were used on the influent and effluent
streams.  Comparisons were again made with these two samplers, the CVE sam-
pler and the polystaltic pump.  The Sigmamotor samplers were set to take 1+00-
ml samples every 30 minutes, the CVE sampler a 50-ml influent sample every 8
minutes, and the tubing pump a continuous 9 ml/min sample of the effluent.
Samples from each were composited every 12 hours.

     Data in Table 36 indicate that representative samples were taken by the
Sigmamotor samplers set at the 30-minute sampling frequency.  To provide for
the "system hold-up" we used the same "lag period" in compositing that we
used at Mosinee.  Samples of the mill clarifier effluent were taken by the
mill staff every morning and were analyzed at the IPC laboratories for com-
parison with those from the small unit.  The mill had no provisions for tak-
ing 24-hour composites of the mill clarifier influent, so the influent values
for the IPC clarifier for the same test period were averaged for comparison
of percentage removal values for total and soluble BODs and color.  Since the
Hydrasieve removed considerable quantities of fiber from the influent pro-
cessed by the IPC unit, the removal of suspended solids could not be compared
with that of the mill system.

TESTING SCHEDULE AT COMBINED LOCKS

     The Combined Locks clarifier operates without chemical additives and de-
pends upon gravity sedimentation for clarification.  The "base-line" trial
(Trial A) of the IPC unit was, therefore, made without chemicals.  This was
followed by trials:

     B.  Ferric chloride at 75 mg/1 Fe3+.
     C.  Ferric chloride at 75 mg/1 Fe3+ with 0.75 mg/1 Nalco 73C32.
     D.  Ferric sulfate at 100 mg/1 Fe3+.
     E.  Alum at 300 mg/1.
     F.  Alum at 300 mg/1 with 0.75 mg/1 Hereofloc 812.3-
     G.  A second base-line study with no chemical additives.

     Because the IPC system was not automatically controlled and because
clarifier influent quality changed rapidly on Saturdays and Sundays, we usu-
ally limited our trials to 8:00 a.m. Monday through 7'-00 a.m. Saturday;
several trials, however, extended through the weekend.

RESULTS AND DISCUSSION

     The Hydrasieve, cleaned frequently, provided good influent for the IPC
clarifier.  The basic problems at this site were with plugging of the influ-
ent line between the mill main line and the 50-gallon per day tank and with
the removal of sludge from the small clarifier to maintain proper sludge-


                                     63

-------
o
Jn
oj
H
3
W)
01
o1 w
H
O
1 O £)
i — !
CtJ
•H
O
CO 0)
t^l ft
H^ CO
1 — 1
S H °b
o "-N. 3
3t* H
9 a
P "CD
H "i K
H O
S r^
O  -H
W P 0
H HO)
&^ O ft
CO CO
1*1
0
EH O

W ' H n3
31 ~5 y
fc a bo
SB 41
H • «
ul
gj g
P4 y
H m ^3
En H
H H oj
PC ' Oj *H

H
CO 08
H | -H
CO O
M OJ
1-3 ft
<| CO
3

CJ
^
VO CD
on H
H • bD
a o oj
PQ HP*
<£
EH OJ
H
P .0
S H
a di
CO 'H
0
11
Pi
CO



>.
*
p



O f—
t— on
H OJ





O t—
r— on
H OJ




IA H
t-- LTN
OJ OJ




o o
t— LA
OJ OJ





on vo
r— o\>
OJ OJ







H -^t
o\ o
OJ CO






H J-
f_ t—







OJ t—
\O ^~






0 O
ir\ t—
CO f—






ON CO
IT\ND
CO t—






H OJ
PO





H OJ
00
0 0






OJ




VO OJ
H OJ
OJ OJ





co rn
OJ O
OJ OJ




-=f i—
\0 J-
OJ OJ




J- -3-
L^-3-
OJ OJ





H -*
u^ J-
on co







OJ O
CO 1—
rn on






J- H
OJ O
m oo







on o
on oj
on oj





ON O
CO ON
t — ^ —






O -=f
O ON
CO t—






H OJ
00





H OJ
1 1
P P
P P






m




C— CO
CO IA
OJ OJ





LA O
CO -J-
OJ OJ




\o o
cdoJ




H -3"
H OJ
OJ OJ





LA^£)
CO LA
OJ OJ







LA O
t— CO
OJ OJ






-=t OJ
O CO
H H







-=f H
CO ON
H H





O OJ
t— I—
t— t—






CO O\
t— CO
b- t—






H OJ
00





H OJ
§§
O Q






.j-




OJ
ON
H





O
O
OJ




OJ H
0 0
OJ OJ




CO H
ON OJ
H OJ





t-VO
OJ l/N
OJ OJ







-3- J-
OJ \D
OJ OJ






0 t--
VO ON







CO IA
\D O
H





ON H
CO t^
t— ^—






O 0
CO CO
t— t—






H OJ
00





H OJ
OP
p p






l/N




l/N O
LTN t/N
H H





1 1
1 1





NO -=I
ON CO
H H




1 O
1 t—
H





CO t—
OJ ON
OJ H







1 H
1 O
OJ






t- 1^
j- on







I O
i -»






CO O
VO ON
CO t—






H
1 O
CO






H OJ
PO





H OJ
oi
Q O






CO




on o
CO t—
H H





1 1
1 1





CO CO
H OJ
OJ OJ




-3- ON
H IA
OJ OJ





O \D
ir\CO
OJ OJ







O J-
1/NVJD
OJ OJ






H H
LA-rJ-







ON OJ
LfN-^f






O H
OJ O
CO CO






H on
on o
co co






H OJ
PO





H OJ
1 1
P Q
P O






ON




U"N O
O IA
H H





1 1
I 1





ovo
-=f t—
H H




CO V£)
m co
H H





OJ OO
NO O
H OJ







CO -=f
-=f H
H OJ






ON-S
H -=t







ON-=f
H-3-






OJ IA
CO IA
C— t—






IA OJ
VO NO
t— t—






rH 01
PP





H OJ
PO
0 0






o
H



t-— OJ
H O
oj on





i i
i i





H CO
H on
OJ OJ




VO ON
H -^t
OJ OJ





O OJ
-3- OO
OJ OJ







^t OJ
on t—
OJ OJ






NO t—
vo on







O LA
t— on






OJ H
VO t^
f- t—






H OJ
r— co
t— i—






H OJ
PP





H OJ
1 1
P P
Q O






H
H



LA on
t— CO






IA C—
CO CO





CO -3-
OJ LA
H H




LA O
OJ C—
H H





O OJ
-3- VO
H H







CO VO
OJ VO
H H






ON CO
H H







ON O
H OJ






O t—
LAVO
t— L~-






OJ LA
VO CO
c— e—






H OJ
PO





H OJ
«P
Q P






OJ
H



O LA LA
O O LA
OJ OJ OJ





LA O H
OJ H -3-
OJ OJ OJ




~J H O
LA h- LA
OJ OJ OJ




VO LA LA
UN LA on
OJ OJ OJ





o vo on
c— co ^a-
OJ OJ OJ







-=t -zt OJ
vo co LA
oj OJ OJ






-=J- H CO
-3- oo on







OJ H LA
on -d- vo






LA H CO
vo t— ir\
t- t— t—






ro co on
t— VO ^D
c— t— t—






H OJ H
PP 0





H OJ H
00 0
00 P






on j-
H H











































CO
• H
tn oj
H >
TO ?-[
>  a
a
h
^i OJ
0) H
H ft
ft 8
3 CQ
• W
in t*
•H S O
M 3 -P
Hi S 0
-° o a
ij > a
P a
O W -H
J3 5- CO
1 O
OJ cB
H 03
0! !»,&
^3
c a
O C OJ
OJ M
•a M a
OJ ai -p
-p -P
•H H
S^1
ft 0
goo
0 IAJ-
o
01 .. ••
Sn W W
0) 01 OJ
* H H

w S S
H oj co
ft
9 H h
CD W Co
W -H H
0 ^J
H a> bo
H Oi 
-------
blanket levels.  The first remained a recurring problem throughout the study.
The sludge blanket was fairly adequately maintained with the installation of
the time-controlled centrifugal pump.

     Data from these trials (Tables C-l to C-7 and Figures C-l to C-lU in
Appendix C) were checked for validity (mathematical confidence limits) on th-
IPC computer.  Averaged data for the seven trials are given in Table 37 and
are summarized as follows:

     1.  The first base line (Trial A) showed that the IPC clarifier
         removed slightly more total BODs than the mill unit did, but
         removed equal amounts of soluble BODs and no color.  In the
         second base line (Trial G) at the end of the series the units
         operated at equal efficiency.
     2.  Ferric ion, either as the chloride or sulfate salt (Trials B
         and D) provided excellent removal for total and soluble BODs
         as well as color.
     3.  The addition of the polymer Nalco 73^32 (Trial C) to the ferric
         chloride slightly reduced the efficiency for total and soluble
         BODs removal but had no effect on the color removal.
     U.  Alum, as predicted by the laboratory jar tests, did not appear
         as effective as the ferric ion in removing BODs but was equiv-
         alent to it in reducing color.
     5.  The addition of the Hereofloc 812.3 polymer to the 300 mg/1
         alum markedly increased the removal of total BODs to well
         above the level achieved with the other flocculating agents
         but had no effect on the reduction of the soluble BODs.

     The high "negative" soluble BODs removal in the two IPC base-line trials
(Appendix C-l and C-7)5 as well as throughout the test period for the mill
clarifier, could not be explained.  This finding, however, did correlate with
the darker color of the sludge from the mill clarifiers (lighter fibrous ma-
terial was in the influent waste stream) and might indicate development of
both color and soluble BODs from the "particulate matter" during clarifica-
tion.

     Spent liquor entering the influent system through spills or deliberate
discharges had the same deleterious effect in this study at Combined Locks as
it did at Mosinee.  This, according to the Combined Locks mill staff, has
been one of the causes of clarifier upsets that have occasionally plagued
their operation.  Operation of the pulp and paper mill under strict control
of spills and similar discharges (i.e., good housekeeping) has minimized
these upsets.

     In our laboratories we studied the effect of the various sewer dis-
charges on clarification, and the data developed in these studies are dis-
cussed in detail in Section VII.
                                     65
-------
  TABLE 37.   CLARIFIER STUDY - COMBINED LOCKS  PAPER  CORPORATION
              AVERAGE EFFLUENT QUALITY AND  REMOVAL
               VALUES FOR IPC AND MILL CLARIFIERS
Effluent quality
Suspended BOD5
solids, Total,
Clarifier mg/1 mg/1
A.


B.
Soluble, Color
mg/l units
Both units without chemical flocculating
IPC
Mill
IPC unit
k6
kQ
with
232
25k
ferric chloride
190
188
(75 mg/l Fe
agents :
205
227
3 ),and

Removal , %
BOD5


Total Soluble

22
13

.1
.3
mill using

8.
8.
no

6
3
Color

-20
-33

.6
.5
chemical
treatment :


C.



D.



E.


F.



G.



IPC
Mill
IPC unit
and mill
IPC
Mill
IPC unit
treatment
IPC
Mill
IPC unit
IPC
Mill
IPC unit
using no
IPC
Mill
k5
7k
with
using
66
146
with

37
50
with
60
5^
with
169
21k
ferric chloride
1^7
187
(75 mg/l Fe
112
207
ko
2k
.1
.1
3+) + Nalco 73C32
25-
5-
(0.
8
6
75
32
-2k
mg/l),
.5
.7

no chemical treatment:
196
21k
ferric sulfate

136
166
168
209
(100 mg/l Fe

129
152
92
199
3 ), and

77
130
alum (300 mg/l), and mill using no
205
2ko
alum (300 mg/l)
197
216
+ Hercofloc
123
24l
812.3
3k
28
.k
.k
mill using

la
29

• 9
.0
20.
1.
no

25-
12.
k
0
31
-k7
.8
.It
chemical

k
1

18
-38

.1
.3
chemical treatment:
32
21
(0.
.6
.0
75 mg/l)
13.
k.
2
8
, and
31
-33
mill
.7
.9

chemical treatment:
85
56
237
262
208
212
Both units without chemical flocculating
only):
IPC
Mill

50
52

239
236

21k
208
129
229
agents

252
25k
50
30
• 9
• 9
13.
11.
3
7
30
-23
.3
.8
(gravity sedimentation

27
28

.8
.7

-1.
1.

U
4

-26
-27

.6
.6
did not sample influent;  values are based on average of IPC influent
                               66
-------
                                 SECTION VII

                 JAR TESTS WITH INDIVIDUAL SEWER DISCHARGES
GENERAL

     In Section V of this report, the BODs sources making up the total dis-
charge of the mill to the clarifier were listed and briefly discussed.  In
the field studies (Section VI) we had noted a marked decrease in the effi-
ciencies of both the mill and IPC clarifiers when larger than usual amounts
of spent liquor were in the waste stream being processed.

     In this section we detail and discuss the studies made on the efficiency
of the soluble BODs removals realized when individual sewer streams were
treated.  These individual sewer effluents were also used to make a "synthet-
ic" mill effluent by compositing each on the basis of its volume to the over-
all volume of the total mill discharge.

     Using these same individual streams, we prepared various composites,
omitting one stream at a time, to determine the effect the omission had on
the overall efficiency of BODs removal by clarification.

     Ferric chloride (25, 50, 75, 100, and 150 mg/1 Fe3+), alum (50, 100,
150, 200, 250, and 300 mg/l) and combinations of these two primaries with
0.75 nig/1 of either Hercofloc 812.3 or Nalco 73C32 were used with each stream
and composite.

     Values for total BODs and, when time permitted, soluble BODg for the
treated samples were compared with untreated (settled) control samples for
the determination of BODs removal.

TESTS ON MOSINEE INDIVIDUAL SEWER SAMPLES

     Some of the mill discharge streams, especially the "high density," pulp
mill and paper machine sewers, could be effectively treated with the proper
concentration of the primary flocculants (Figures 11 and 12, Legends 1-11)
or primary flocculant-polymer combinations (Legends 12-23).  These bar graphs
display that in others, such as the foul evaporator condensate, soda recovery
wastes, and digester room wastes, the BODs level was not substantially af-
fected by any of the flocculating agents.  Those bars shown below the zero
line in the lower section of the figures indicate negative removal of BOD,
probably due to prolonged holding time.  These waste flows should be treated
by routes other than clarification (i.e., steam stripping of the conden-
sates).

     In our earlier work with the individual streams we found that toxic ma-
terials were in some samples, and they interferred with the BODs test (Table
38).  Therefore, we could not make even good estimates of the concentration
                                     67
-------



^D
ON
CM OJ
O
II CO
0) ||
rH
•§•3
rH 4^
W B



Lf\
^t CM
rH t—
11 H
rOH
3 a)
H -p


CT\
r-H
CO
It H
(U II
^H rr-
3d T
0 $ 4,
CO EH T




H
* 0,
« -*
Q II OJ
pq  ra
H a
CO T-J
rH C
o
OJ 0
H
(H
O
<*•£
t.
t- 0
p.
^> ?
Wr3
Hr°
CO
CM
OJ
CM
H
CM
03
CM .H
t- d
H ft
VO 3
H (5
H
H
O
H
LfN
IA
H >>
*$
rH M
vn «>
^ T)
-^
row

ro
CM
CM
CM
CTs
rH
CO S
H -H
CO O
rH Qj
IOJ
rH ^1
(11
3 S1
r-H CO
VO
OJ
H

)
n


1
CM
CO
o
oo
t—
0 0
o
d r-
3 a
rH t
S c
ir\ i
«
0 &
•f +
01
ft. &
rH r
§" f
l^ U
CM f
00 -
rH r


+ +
01 0
ft. p.
rH >-
3> 1
tn c

H o








i
1


D





J
H
3 CM
OO
J CJ
3 00
H t-
-|
3 O
1 CJ
u id
3 S3
+
U 01
-, f^
H rH
i "i
~\ i/>
- 1—
t u^
- rH


+
a  l/N

J CO










|
0
H





oo
£
cofloc
s
+
+
fo
0>
fc
H
If
o
u\
rH
vo
H


-r
01
PK
H
"9
o
H
J-













|

o-l
CM
rH
CO C
0
0 C
0 f
rH t
 s> '
3 0 0 C
H rH OJ O
- CO 0\ C
r*
1
1

1
1







1
0
-*



!H
0)
ft
ff •
ft^~~
^ (U
-P
rH 0}
H ra
• ^  . S
a- H cs
p § -
. id H
0 -H H
t> «H
1 S * ^ g
3 3 "> crj Q.
H rH 0) " *T
<  P
3 0 in " 5*
J 00 W 'H
-------
f
1
1
	 1 f*"~"
<~\ 1
H  H ,
H -P
0 0
W ^ IT-
1
1
1

° r^
oo o '
H 3 '
d) II ,
^ 1
H -P '

<< £
OJ
«s [^—
Q OJ VD O
SLA .H
|| ,3- JKJ
O
3 1 " ^
•H ,0 H J
-P 3 aj !
•H H -P |* i 1

1 1 1 1 1
O O O O O C
LA ^t 00 OJ H
00
OJ
OJ ^
OJ 4)
H to
oj 

0 > OJ
OJ 0 (3
H 4)
i-l C
\£> 0) O
H -p

H M
•rH
O
ir\
oo 5?
OJ 0)
i >.
VD O
r-i O
•a
Lf\ 'O
o

> C








1
1
1
1
_^


r
1
	 j
on
H
cy oo oj
ro ro
O CJ O
m o to
t— H t-
o oj on o o oj
o • 0 o •
H m oj in H m
Oj rH H 0) dj f-H
PSCOOJCOOJSGSSCOOJ
H OO OOMH OOOJ
"v. o fn o co "^ ~*^. o co •
M H t— r-l f- |bpMHt~-ro
OOOO OOCOOJ
i/\ooouLr\mocj ro
f-^H h H b- t- JH r-J OO
0> oj ff
lAOirvQOOOOOOOLTv
OJ ITit— OlAlAOlAOlAO OJ
H H rH H OJ OJ CO
r-fOJOn-^t UA VO t— CO ONO r-fOJ
H H H
1 1 1 1
D O O O
H OJ p">
                                                      13
                                                       0)

                                                       §
                                                      •H
                                                      +>
                                                       e
                                                       o
                                                       O T)
                                                      H
                                                      H   "
                                                      •H  !H
                                                       a  «

                                                       01  01
                                                       0)  CD
                                                       C  bO
                                                      •H vH
                                                       to TJ

                                                      £   -
                                                           
IIH   03
       ^>
                                                           o
                                                       O)   O  <1J
                                                      -P   <1J  M
                                                           fc  ffl
                                                       (U
                                                      a
                                                          ^
                                                              M
    O  ,S
    m  -p
    -^  UA
69
-------
               TABLE 38.  BOD OF INDIVIDUAL EFFLUENT STREAMS
                 AND TOTAL MILL EFFLUENT OF MOSINEE MILL
                      Source
        Dilution
mg/1
Digester sewer

Pulp mill sever



5th Stage brownstock washer



High density sewer


Foul evaporator condensate


Digester blowdown
foul condensate


Soda recovery sewer

Total mill effluent

120
600
150
300
600
1500
300
600
1000
2000
30
100
150
120
300
600

300
600
750
600
1500
60
150
637
633
1210
1^90
1629
1620
9l*0
1089
1353
131*0
ll*l
171*
172
570
762
751

1023
1325
1300
1*236
1*188
271
280
of BODs in these streams without careful dilution techniques.   Excessive
amounts of these same toxic agents would also interfere with the operation
of a biological secondary treatment system and for this reason, if for none
other, their presence in the total effluent should be held to  a minimum or
completely eliminated.  Discharges into individual mill sewers were used to
prepare a "synthetic" total mill effluent in the following volume relation-
ships :
     Paper machine
     High density
     Pulp mill
     Foul evaporator condensate
     Soda recovery
Million gal/day

      10
       1-5
       0.6
       0.3
       0.28
% of Total

   77-9
   11.7
    l*.6
    2.3
    2.2
                                     70
-------
     Digester blowdown                      0.07
     Digester                               0.05
     5th Stage brownstock washer            0.035

                                   Total   12.835           100.0

     To find the conditions for maximal BODs reduction when individual wastes
were omitted from the system, we made combinations of seven of these eight
wastes and treated them also in jar tests.  Data in Figure 13 indicate that
higher levels of BODs removal could be achieved with less chemical if certain
individual discharges were eliminated from the flow.  That is:

     1.  The "total" effluent required 250 mg/1 of alum for 17 percent
         reduction or 75 mg/1 Fe   for \2. percent reduction in BODs.
     2.  With the soda recovery sewer flow eliminated, a 250 mg/1 alum
         level resulted in 26 percent reduction and the Fe3+ at 50 mg/1
         resulted in a 51 percent removal of BODs•
     3.  The absence of the pulp mill sewer from the total flow, with
         all other flows present, permitted a 60 and 63 percent reduc-
         tion in the BODs concentration in the waste stream for 100 mg/1
         alum and 50 mg/1 Fe3+, respectively.

     When 1 percent v/v of the digester room waste was added to a mixture of
wastes having relatively good clarification characteristics (paper machine
and high density wastes), higher concentrations of flocculating agent were
needed to produce an effluent of adequate discharge quality (Figure lU).

     Attempts to calculate BODs balances for these various discharge combi-
nations, and to correlate these with the actual BODs values obtained, was
apparently hindered by some interaction(s) between these wastes.  This could
have been due to changes in the toxicity of the mixtures for the microorga-
nisms in the BODs test.

     The difficulty in calculating material balances for the total or soluble
BODs of the individual streams and composites is due to the fact that all of
the initial BODs values for these samples were on "settled" jar samples.
Settling removes BODs and the removed BOD could not be accounted for in the
calculations of BODs balances.  In these samples, the interaction of the ad-
sorbing character of fibers and particles with the dispersing character of
the various mixtures could markedly affect BODs removal during the settling
step, thereby modifying the "initial" BODs of the mixture.  The importance
of this interaction in the overall flocculation-clarification process should
be investigated further in any attempt to improve the process.

TESTS ON INDIVIDUAL STREAMS FROM COMBINED LOCKS MILL

     The five streams making up the bulk of the flow to the clarifiers at
Combined Locks ~~ from the main, "CM," wet room, Number 5 and Number 6 sewers
— were tested individually for the removal of BODs in jar tests with ferric
chloride, alum and combinations of these with either Hereofloc 812.3 or
Nalco 73C32.  We also prepared a composite "total" mill effluent and compos-
ites lacking individual sewer outfalls, as we did with the Mosinee wastes.
                                     71
-------
o
00
r-t 00 O
II rH

H -P
O O



OO
rH VO
oo
 HOO 9 ^ W
Q
SO) || 1 	 1 	
H "T 	 1 	
,0 H
H 2 d :
•H O O
-P CO EH

1 1 1 1 1
o o o o o o c
\T> LA -3- OO CM H
0)
60
t— u3
H IS
f! D
VO 4J to
H LA
O -P

1-t

t—
rH
VD H
H H
LA a
O H 0)
H 3 M
^ +3 £


CM g
H
H
\£> 0)
CM O !*


00 5 >
t-- ^ 0
1)
CM

t—
r-H
VD
rH
H
00 H
H a!
•p
MD
-*
OO

>
ZD
CO
i
OO OO OO 00 J
CM CM CM CM *H
H H rH H H0
OO OO OO
OOCJOOOO CD
i — 1 C i — 1 t — i — 1 t — i — 1 CQ
OOCJOOOO "*
J-lH?-|rH^rH^ i
WSWSWSW *
W)bpbpw)W)bflW)CM J>
i'S'i'aug'^aJHaj.H
LALALALALALALA °°J3

oooodooHt-.,
+ + + + + + + oo5i
o o C
(uciajaiaicur-Hius^o
H H ^
rHHHHrHHH-^ "~~- CH
LALAOO LA LA O t— f— (11
CM CM LA IA t— t— O • --p
to
^^ ^^ rt x«^ .^. r> O
1-HrHrHrHHrHHrHg
o
O
•H
O
-------
As predicted by our earlier work, those streams having low spent liquor con-
tent, i.e., Number 5 and main sewers, had good supernatant quality (Figure
15).  The BODs removal was good with relatively low concentrations of Fe?
or alum (Figure l6).
.3+
     Figure lU.  Jar tests show mixed effluents with addition of, left
                 to right, 25, 50, 75, 100, and 150 mg/1 Fe3+; the con-
                 trol, no addition, is jar f.  A Mosinee effluent was
                 used composed of:  (bottom row) paper machine and high
                 density sewer effluent, and  (top row) the above with
                 1 percent v/v digester room  effluent.

     The chemical concentrations of ferric chloride and alum required to ef-
fectively remove both suspended solids (for high supernatant clarity) and
BODs was higher for the "total" effluent than for the various composite
streams (Figure IT)-  As the individual streams were removed from the system,
the quantity of flocculants was reduced, especially for those mixtures with-
out the "CM," wet room or Number 6 sewers.  The Number 5 sewer stream had
good removal characteristics, with low chemical requirements, as an individ-
ual stream, and omitting it from the mixture  did not appear to enhance the
BODs removal.
                                      73
-------
Figure 15.  Jar tests with Combined Locks  individual sewer effluents,
            a-e  (main, "CM," wet room,  machine No.  5 and machine
            No. 6).  Top row:  with 25 mg/1 Fe3+;  bottom row:   with
            20 mg/1 alum.  Control wastes  did not  settle without
            additives and are not shown.
                                  74
-------

o
LA
II
H
cd
o
EH




ON
II
H
0 1

[
0 |
H 1
II
cd
"p
H

0\
II 1
1 L
" r'"

H
Cd rH
.p +J
•H O
M
1 1 I I
O\
CO
\O
^ h
4)
OJ »
H
H
O
H
ON^
^H
00 I
-* Si
OJ
rH
H
H
10
r-2
^ fc
"* &
co^
H
LA
j — t
CO
rH
OJ
rH
ON
CO -H
1
LA
OJ
H

O O O O O O
LA ^t CO OJ rH


OJ
CO
o
CO
t—
o
0
3
rH
t—
O
m
0)
rH
bO
o
H
t—






*H
OJ
rH

J

O
CO
OJ
mg/1 Hercofloc 8]
t—
o
00
i — !
LA
OJ
co






5
0
OJ



0
H
1


OJ
o
co
t-
o
o
rH
cd
H
£
0
01
1 — 1
hO
01
ON






5!
H
o
CO




co
OJ
mg/1 Hercofloc 8]
t—
o
5!
	 i
i i
o
LA
O
rH






H
o
0
rH
^



0
OJ


OJ
CO
o
CO
t—
o
o
H
cd
a
rH
t—
o
H
1 — 1
tiO
o
IA
H
H






H
rH
0
LA
rH
LA






Hercofloc 812.3
£
o
H
i
o
o
rH
OJ
rH
CO
OJ
H
co
0
o
1 Hereof]
t—
o
«
rH
0
r-l
VO






OJ
CO
o
co
t—
o
o
Is
H
LA
t—
o
5!
^J

o
o
H
co
rH













                                                     rH
                                                     rH
                                                     •H
                                                      &

                                                      01

                                                      O
                                                      o


                                                     T)
                                                     O
                                                      si
                                                     •H
                                                      S
                                                      5
                                                      01
                                                     •H

                                                     •H
                                                     g
                                                     01
                                                     
-------
$
1

1


1

c

OJ I
^ 1

fi
1

1
?! '
1
rH
a



1
J* (
i. '

-p f
O 1
i
H
•»—
£
"> H
Q -=T

•H «J
•rt O

1 1 1 1 1 1
O O O O O O O
VD LA J- on OJ H
H
LA
rH JL|
OJ ,Q
H*
-i
"s
S
OJ
Ivo
H
LA
LA
rH fa

-1
"g
ts
OJ

\D H
•H 6
,~ P
LA tj
H
OJ 4>
rH ^
33
o
on£

r-l
on
H =
oo e
"~3
LA*
^S
rH

H
LAH
H oJ

OJ

(

on on on on
CM CM OJ OJ 01
rH H H rH £3
QOOJCOOJCOOJCOOJ 3
on on on on m
oooooooo ,
oonoonoonoon >H
Ht-Ht-H^Ht- -P
HH H-t Vl ^H 01
OOOOOOOO
oooooooo 
MMMMIMhOMM _j
aaaaaaaa 'j
LALALALALALALALA !^
t— t— r— ^- t- t— t— t— T)
jOCDCDOCDCDod '^!
+ + + + + + + + ^

fl
CJ\OrHOJOn-4-LA\D 0
HHHHHrHrH "3
o
o
o
•? s
_. CM -P
1 rH .H
1 CO OJ 0)
J on JIJ
0 0 §
o on H<
-------
                                SECTION VIII

             DISCUSSION OF THE MECHANISMS OF SOLUBLE BOD5 REMOVAL
     The studies with jar and pilot-scale clarifiers have readily confirmed
that substantial quantities (as much as 20 to 50 percent) of the soluble BODs
are removed in clarifiers by coagulation aids and that the removal can be a
significant factor in the treatment of pulp and paper mill waste waters.

     Our data show that, at least to some extent, the reactions leading to
improved BODs removal can be optimized in well operated and controlled clari-
fier systems.  Several studies were made in an attempt to better define the
mechanism for reduction of soluble BODs during the treatment of process
waters in primary clarifiers.

     It had been evident from the beginning of the study that some of the
soluble BODs could be removed from the waste stream by several processes
(membrane, ion-exchange) and that no one mechanism could be credited for the
entire effect.

GEL CHROMATOGRAPHY STUDIES

     The components responsible for the soluble BODs i-n clarifiers can be di-
vided into two classes, wood-derived chemicals and additives.  The wood de-
rived chemicals include degraded cellulose, hemicellulose, hemicellulose
degradation products (reducing sugars, acetic acid, methanol and uronic
acids), extractives and in small, perhaps minor amounts, metabolites common
to living tissues (adipic acid, oxalacetic acid, etc.).  Additives used in
papermaking include starches, polysaccharides, gums, latexes, resins, dyes,
etc.  Some mills will also have organic solvents such as methanol, acetone,
etc., which are biodegradable.  The organic solutes that pass the 0.1*5-um
filter probably comprise colloids, low molecular weight polymers, oligomers,
monomers and simple organic compounds.

     To better understand the removal of soluble BODs by coagulation, we
treated the effluent from a chemimechanical mill (Combined Locks mill) with
the various flocculating agents.  Using gel chromatography we measured mo-
lecular size distribution before and after flocculation and clarification.

     Samples were filtered on a 0.^5-ym filter prior to chromatography.  The
gel chromatography was carried out on Sephadex G-50, a gel which excludes
(allows no equilibration with the gel pores) molecular weights 10,000 and
above and includes (allows complete equilibration with the gel pores) molec-
ular weights of 500 and below.  Molecular weights between 500 and 10,000 are
fractionated according to the number of pores with which the molecules equil-
ibrate.
                                     77
-------
     Figure 18 shows the TOC (total organic carbon) of 2 ml of a tenfold con-
centrate as it elutes from a 1 x 55-cm column.  The second curve shows that
after flocculation with TO mg/1 ferric chloride not only are large molecular
weight residues removed, but also low molecular weight molecules.  The curves
in Figure 19 were obtained when a similar sample (l-ml sample of fivefold
concentrate) was fractionated on Sephadex LH 20 (mol. wt. 5000 to 200) before
and after coagulating with 10 mg/1 of Nalco 73C32, a cationic poly-electrolyte
and analyzed for aldoses by the phenol-sulfuric acid test (8).  The low mo-
lecular weight peak almost disappeared.

                                   Molecular Weight

                              10,000      5,000     200	
          1000
           800-
           600
           200-
                                                               Before
                          10           20           30           UO
                                    Column  Eluate, ml
     Figure 18.  TOC and molecular weight distribution of gel-chromato-
                 graphed chemimechanical mill effluent before and after
                 coagulation with 70 mg/1 Fe3+ as ferric chloride.
                                     78
-------
                             Molecular Weight
                             10,000      5,000
                 200
     o
     GO
       l.U
       1.2
       1.0
       0.8
     S 0.6
     w
       O.U
       0.2
                                     .After
                                                         Before
                     10
  20          30
Column Eluate, ml
     Figure 19.  Phenol sulfuric acid test for aldoses and molecular
                 weight determination of gel-chromatographed chemi-
                 mechanical mill effluent before and after coagula-
                 tion with 10 mg/1 Nalco T3C32.  One-mi samples of
                 eluate were analyzed.

     The phenol-sulfuric acid test measures aldoses and polymers of aldoses
when borate interference is observed.  The solutions analyzed in this report
showed enough borate sensitivity to indicate the test was responding primar-
ily to aldoses and polymers of aldoses and not to noncarbohydrate aldehydes.

     When a sample fractionated on Sephadex G-50 was divided into three frac-
tions corresponding to high, medium and low molecular weights and analyzed
for TOG, aldose and BOD5, the results summarized in Table 39 were obtained.
The percentage change for the soluble BOD5 was not as great as that observed
for the aldoses or TOG.  Apparently not only low molecular weight biodegrad-
able materials were removed but also low molecular weight nonbiodegradable
organics,

     Das and Lomas explored the different mechanisms of the flocculation of
cellulose fines with the cationic polymer polyethylenimine (2).   They were
                                     79
-------
able to show that bridging (the adsorption of two sections of a polymer to
two colloids) was a significant part of the cellulose fines flocculation.
Such a mechanism could also occur with the other polyelectrolytes and with
ferric chloride which hydrolyzes and polymerizes to a polyelectrolyte (l4).
The mechanisms for the coagulation of cellulose fines involved collapse of
their double layer followed by adsorption and bridging.  Lower molecular
weight residues were generally difficultly adsorbed.  Their participation in
coagulation probably depends on salt formation and chelation.  Such salt for-
mation by hydrolyzed iron can be demonstrated with gel chromatography. Figure
20 shows the elution curve for a mixture of ferrous ammonium sulfate and so-
dium gluconate.  The complexes absorb ultraviolet light and are, therefore,
detectable with the UV absorptiometer at 280 nm.  Iron values were obtained
by the o_-phenanthroline test.  Because the curve shows that at a pH of 2.0,
most of the iron elutes from Sephadex LH-20 and G-15 (fractionation range
150 to 10,000) in the highest molecular weight fraction, the complexes prob-
ably have molecular weights of 10,000 and above.  Figure 21 shows the dis-
tribution of the complexes when the pH of the solution was 7.0.  At higher pH
more of the iron was in the lower molecular weight fractions.

              TABLE 39-   REMOVAL OF DIFFERENT MOLECULAR WEIGHT
                FRACTIONS OF  ORGANIC RESIDUES BY FLOCCULATION
                                                        Q
                                                  Sample
      Effluent treatment              UNF      HMW      MMW       LMW
      Phenol-sulfuric acid test;
        total absorbance,  U80 nm

        Filtered                        --      7-5      3-9       5.2
        Floccedb and filtered           —      3.0      2.U       2.5
          Reduction, %                  —     59-9     38.6      51-9

      TOG, mg/1

        Filtered                      3^2U      77^      ^57      2301
        Floccedb and filtered         2150      320      272      1387
          Reduction, %                ^3-9     58.6     1*0.5      39-7

      BOD5, mg/1

        Filtered                      381+0      552      ^87      2689
        Flocced° and filtered         3080      379      368      2131
          Reduction, %                19-8     31.3     2U.U      20.7

      aUNF, unfractionated; HMW, high molecular weight (10,000 and
       above); MMW, middle molecular weight (500-10,000); LMW, low
       molecular weight (500 and below).
      b
       Flocced with ferric chloride, 70 mg/1 Fe.
                                     80
-------
              3000
Molecular Weight
  1500
 150
                                                                     H300
 o
 00
 C\J
 -p
 0)
 -p
 -p
 ra
 C
 03
                          Ultraviolet
                          transmittance
                                                                            a
                                                                            o
    100

                          100
        125
   Eluant, ml
150
                                                               175
     Figure 20.  Ultraviolet light transmittance profile and molecular
                 veight distribution of a gel-chromatographed solution
                 of sodium gluconate and ferrous ammonium sulfate, pH 2.

     If these biodegradable lov molecular weight residues which participate
in the coagulation are carboxylic acids, then the mechanism for their removal
could be the complexing and/or chelating of the Fe3  of the ferric chloride.
As the ferric chloride hydrolyzes, more and more insoluble oligomeric forms
of the hydroxide are formed.  Whether or not the iron system would flocculate
the low molecular weight residues without concomitant flocculation of poly-
mers and/or suspended solids was not determined.

     The types of carboxylic acids which might be derived from wood and form
complexes with iron include the Krebs cycle acids (citric, adipic, oxal-
acetic, etc.), oxalic, aldonic, and uronic acids, and aromatic acids.  Since
little ultraviolet absorption was found associated with the low molecular
                                      81
-------
 weight  gel  chromatography fractions, aromatic residues have not been consid-
 ered in this phase of the study.
                      3000
Molecular Weight
       1500
150
                                                                     -200
                                  Ultraviolet
                                  transmittance
                                                                            I
                                                                            o
                                                                            ^
                                                                            H
                                    105
                               Eluent, ml

     Figure 21.  Transmittance and molecular weight distribution of
                 a gel-chromatographed solution of sodium gluconate
                 and ferrous ammonium sulfate, pH 7.0.

     The sewer to the Combined Locks clarifier has five principal tribu-
taries two of which gave aldose test responses strong enough for analysis by
gel chromatography after the effluent was concentrated fivefold.  The column
elution curves are shown for the digester room effluent and the paper mill
effluent (Figure 22).  The digester room effluent showed predominantly high
molecular weight residues and the paper  mill effluent, low molecular weight
residues,
                                     82
-------
o
oo
 01
 a
 a)
 •s
 o
 0)
   3.0 -
    2.6
   2.2
l.U
    1.0
    0.6
    0.2
                       10,000
                               Molecular Weight

                                     5,000
500
                                Digester room
                                effluent
                   I
                                                                  Paper mill
                                                                  effluent
                  _         20         30           50           50
                                  Column Eluate, ml
    Figure 22.  Phenol  sulfuric acid test for aldoses and molecular weight
    distribution of gel-chromatographed chemimechanical pulp mill  effluents
                                       83
-------
     Since this particular mill uses starch as one of its additives, one of
the aldoses could be glucose and/or related starch degradation products.

STUDIES ON INFLUENT AND EFFLUENT SAMPLES FROM THE FIELD TRIALS

     Clarifier influent and effluent samples from the two trials, at Mosinee
and Combined Locks mills, were sent to the Analytical Section of the Insti-
tute for carbohydrate and weak acid determinations.   Data are given in Table
ko for the results of these analyses on samples from high, intermediate and
low soluble BODs removal levels during clarification.

     The carbohydrates were determined by gas chromatography after acid hy-
drolysis and derivative formation (15).  The weak acids were determined by
passing the sample through a cation-exchange resin column and conductimetri-
cally titrating it with 0.1N_ sodium hydroxide.

     No consistent correlation was apparent between the soluble BODs and the
carbohydrate removals, indicating that in the high soluble BODs removals some
precursor(s), in addition to other than the oligo- and polysaccharides, were
being removed by flocculation and clarification.

     There was slight correlation between the soluble BODs and weak acid re-
moval values for these samples, particularly those from the Mosinee study.

STUDIES WITH MODEL COMPOUNDS

     On a modest scale we screened model organic compounds for removal from
solution by flocculation in order to learn more about the chemical nature of
susceptible compounds.

     The compounds were selected on the following basis:

     1.  Those compounds having at least some structural relation to
         substances found in mill effluents, either as atypical compounds
         or "families of compounds" (16-1.8);
     2.  Those compounds with some BODs without marked "toxic" effects; and
     3.  Compounds having sufficient solubility so that gas chromatography,
         total organic carbon or some other convenient analytical procedure
         could be used to monitor the results of jar tests.
     k.  Removals of TOC from pure solutions were early recognized to be
         critically affected by pH.  This variable was evaluated carefully
         and the interesting and probably important results provided in
         the following figures are discussed with the realization that
         the highest levels of TOC removal were often noted at pH levels
         outside the ranges normal for mill wastes or discharged to
         treatment systems.

     The jar tests were made with added fiber in the form of unbleached kraft
pulp which had been disintegrated in an Oster blender for 5 minutes to pro-
vide a surface area for adsorptive bridging.  Fibers had increased the remov-
al of BODs in some of our earlier work (Section V).   This effect was confirm-
ed with two runs with hydroxybenzoic acid and p_-methylbenzoic acid with Fe3
                                     84
-------












a
0
s

o
3
H
0
o

g
2

(HJcfl
W F*i
S o
M M
CO EH
O <
S o
M
2 f*
1 s
s °
H W
° 3

V £
W CO
W EH
t5 O
J_3
P H

>H
CO PP
w
rf S
OS ^E
P W
JM pq

o co
pq

•3 EH
o co

pr, ^
o

g
o
*Ej
S
0H


•
o


ppT
3
9
EH















•3
(U
5:
^
1
• o
H ,0
cd £H
I> co
o o
s
(U
K

o
w;
P
S



£>
W
O

g

•i
to

•H
O
OJ

•3
£


H

"a
s
0]

2
§
in
«
S

jj
o
H

•d
a
co
« tn
g -d
P -H
•H H
-d o
0) [fl
g
 O
85
-------
(as FeCla) at various pH's, with and without  the addition of fibers  (Figure
23).  TOG analyses were used to monitor changes  in chemical  concentrations of
the model compounds before and after clarification.
 0)
K
50

ko.

30

20

10
             £-Hydroxybenzoic acid


Removal


50
ko
30
20
10
n
p_-Methylbenzoic acid
/^^x
• / TV
- X ' y
/ A
O
- V
1 1 1 1
                   pH                                        pH

         X With 350 mg/1 kraft fiber added

         O Without added fiber

     Figure 23.  Effect of pulp fibers on removal of soluble compounds
                 with 100 mg/1 Fe3+ (as Fe Cla) at various pH's.
                 	, 350 mg/1 kraft fiber added;	, no fiber added.

     Solutions were made from an amount of the model compound calculated to
give a TOG level of 90-95 mg/1 in the final volume to be used in the study.
This was first dissolved in approximately k liters of distilled water; the
correct amount of blended fiber was added (350 mg/1 final volume), and the
volume was brought to either seven or nineteen liters (final volume), depend-
ing upon whether one or two sets of jars were to be used for the evaluation.

     One liter of the mixture containing the model compound and fibers was
added to each of six jars.  Five of the jars were treated with flocculating
agent (various concentrations) and the sixth was used as an untreated "con-
trol."  All jars were settled, sampled and filtered (0.^5 Vim).  TOG was de-
termined on the filtrates from the "control" and the five treated samples.
All "removal" values are based on a comparison with the TOG concentration of
the "control" after filtration.  In this way the adsorption of the model
compound on the fibers  without flocculant addition could be monitored.  In
all tests the adsorption onto fibers of the controls was less than 8 percent
(based on weight calculation), except for the arrowroot starch and dextrose
which were found to be 65 and 2k percent, respectively.

     Ferric chloride and alum were the primary flocculants used.  Alum (Table
kl) did not remove significant amounts of model compounds except polygalact-
uronic acid (87 percent removed at 200 mg/1 alum concentration).  The highest
removal of model compounds obtained with lime up to 500 mg/1 was 2 percent.
                                      86
-------
                  TABLE hi.   TOG REMOVAL FROM SOLUTIONS OF
                       MODEL COMPOUNDS TREATED IN JAR
                        TESTS WITH ALUMINUM SULFATEa
                                                      Removal of
                      Compounds                         TOG, %

            3,lt-Dihydroxycinnamic acid                    0.0
            Hydroxymalonic acid (tartronic acid)          1.0
            Methylmalonic acid                            1.0
            D-Galacturonic acid                           1.1
            p_-Hydroxybenzoic acid                         2.2
            Propanedioic acid (malonic acid)              2.6
            p_-Methylbenzoic acid                          3.2
            Dextrose                                      lt.0e
            Isopentyl alcohol (isoamyl alcohol)           It.7
            l,lt-Butanedioic acid (succinic acid)          7^
            lt-Hydroxy-3-inethoxycinnamic acid
              (ferrulic acid)                             8.3e
            3-Methoxy-lt-hydroxy benzaldehyde
              (vanillin)                                  9.1e
            trans-Butenedioic acid (fumaric acid)        10.le
            Arrowroot starch                             ll.lt
            p_-Hydroxycinnamic acid                       13.3°
            Benzylmalonic acid                           3lt.lte
            Polygalacturonic acid                        87.0
            Q
            All compounds were treated with all levels of alum.
            Only the best removals are listed in this table.
            Alum = 50 mg/1.
            ^Alum = 100 mg/1.
            Alum = 200 mg/1.
            eAlum = 300 mg/1.
     Data for the TOG  removed when other  compounds  were  similarly tested are
given in Figure 2ka,*-e.  These data, are very complex,  with low molecular
weight shortr-chained compounds  (acetic acid,  ethyl  acetate)  being removed at
a 20 percent level, 2-pentanol  and tertiary amyl  alcohol showing  less  than
5 percent removal and  hexanedioic acid, ItO  percent  removal.   Several aromatic
carboxylic acids had high removals, while galactouronic  acid reductions  were
less than 5 percent (Figure  2ke).
                                      87
-------



0) *
-P O
CS CM
-p X

o
<;



1 1 1 1 1
O 0 O O O
C7\ GO t— ^O IA

X
/
J
\
X
\ "
\
\
\
X -

1 1 1 1
x-
X
\
Lf\ (U \J-
0)
X P <•>
^ 0 W « X
O O X
-=r H o— o — o -
•? *


on —

1 1 1 1 1 1 1 1 1
-p
•H
0)
W -p
U~"\ f^ QJ
S -p
P 0)
woo
»• ft cd

^ OH
: cup.
H Q)-~-
rQ S (1)
0-, Pi +3
H " cd
O Ti -p
rn .,-4 n
ooooo oooooooooo Ocd
-a- "n  0) -p
O 0 0)
P ^ 5

X

X
X
X


4

1 I 1 1


X
Tj X
•H X 1
0 O W "> 1
X < O O X |
P. 0—0—0 x
0 32 '^
5 \-
o >
a) 5


K~i —

I I 1 I I i i i i
•^^ r^ V-*
a -m cd
o w
) — 1 Tj
bD Hi O
g O cd
LA Cd -H
j- oo !> -p
4^> -P cd
cd cd H
w - — - "
00 rH m T^J
0) H -H
•r-l 0) Cd
OOOOO OOOOOOOOOO VifJH
-3~ oo c\j H CT\ co ^ \o LT> _j- ro c\j ^ Q
ft m -H
51U3H * (001) ItJAOTOa » r-l cd H
p — cd
ft «
+ 0
43 oo
o3 piq Q
\£) — ^


X
1
1
{-
X
\.
X
\
\
\
2
1 1 1 1


?
y
x tc f
m TJ o O xy
•H O O **
o o— O I
<* 1
0. 0
-H 1
H 1
-=t a 1,-
X X
O


•oo —
1 1 1 1 1 1 1 1 1
v£) ^ i | ^
^ Vi cd
O -P
, 
M 
-------
w w
O CM OJ c\l ^
^ X
—
1 1 1 1 1 1 1
oooooooo


vo -i

O <•> X
O "> O eg "> J-
0 W 1 W W ><
1A H O— O— O— O \-

4^

(O —
1 1 1 1 1 1 1
f*
CO *H
•^ CJ

O 0
ft-rl
a o
0 -H
U >t)
(U (3
K H cd
ft r13 X
e H^
ra «
J- 0)

O O
S3
H -H
CO <(H
!>  -d- ro c\j ^-1 t— VDLrx^troojrH OW
                                                                                        ft
                                                                                     M  ? H
                                                                                     a  o o
                                                                                       •H XJ
                                                                                     O  ^ o
                                                                                     LPv  Cd O
                                                                                    -p -p
                                                                                     cd cd H

—
>

X
x-
1
(U O X
CD CM v
0 " \
C ffi \
•H CO J
«H -
«H O
CtJ
K
-

1 1 1 1 1 1 1

,V£) —
t

X
\
r— ( m CM CM M « /
W OWWWOW /
ft d o— u— u— o — o o
cd W V
^ "c
a5 1
P4 X
i \
\
1

I I 1 1 1 1 I
•H 0) ^
Vi PL, cd
*rH
vo PJ w j_>
H cd fn
P3 v_x (1)
O, 13
+ ^
«H OJ fl
jirs g fe  IA J- 00 CM H t-^OiA^J-fy-iOJH 3
(OCXL)
(001)
                                                                                    bO
                                       89
-------




•H
O
•^1 ' — *
O -r!
•H O
o <;
-d o
 CO





_
X
\
X
1
S3 33 Jf
O  O
pj rrH | 	 1
W -H 0)
>d o ^
3 *•>».
O O LSJ
ft -H fl
g O tt)
- 85-°
(1) «
(1) fl Tj
oooooooo oooooooooo i— I  m M
g - p
H PM M
0) ft TJ
^ tn 5"
C 3
33 O O «

-* W) !> oJ
a
-P O
O 03 -H
LPv O
C^ ' — *H
00 —^
05 O fl TJ
OOOOOOOO OOOOOOOOOO j CH o -p
O -H ^
' Vl O ^3
O H vH I
LTN "^ -d -*
-p bO <1) •»
0 0 fl H
, S CH O -p id
CH 0 3 q
W H ,b CD
"*
a
OJ
ro
 IA j- m CM i-H at)
(OOL)
(OOL)  T^AOBiaa 3!
                                    90
-------
                          \
      o
      to
o
t—
o
oo
            (OOI)

     -A-^
                                           B3
                                           ft
           (3OL)
o
o\
                                                 (OCXL)
                                                      o
                                                     co
                                               o
                                               t—
                                                 (001)
                                                                                                O
                                                                                               •H O
                                                                                                O -H
                                                                                                N O
                                                                                             r*H  fl N
                                                                                             -p  >
                                                                                                  -P
H O
O -H
03 O

i
                                                 a  x
                                                 0)  O
TJ
•H
O
-------
•d
•H
0
<
O

9
a
•H
U
S?
o
4
>>
33
Aj

1 1
O O
O\ CO





0)
a
01
N
C3
0)
.?
1
K..
£~
O H
B °
tj X3
>? °
^3 »•-)
S3
*d 'H
T) 0
>1 M
J3 0
•rH tn
p» j> 1 0 1
o
(^ x
"
1 1
g
^
s
^! ^! 1
-p -H O
^ -H TJ
!s 1 •>
^ 0|TJ
" W -H
Tj » O
__. d 0) aJ
p a
o j-H
rH M 0
oooooooo ooorQO
r— \.o i^ j fn (M rH ooNCopi^Sj
(OOi) TUAOraaa »/ ^ "n "E o
(001) TBAOUISH % m ff g
•H -P
tH TJ  0 p *|~ ° *
C 1
W 0
^5
iJ S3 X
0 O /
J d 0 33 /-
^ o_L« o-y- w /
? o -:wifi-x \
Q \
>l
m ji

1 1 1 I 1 1 1 I 1
H •— 00^—
^ cd 1 tf
> W !>j -H
O - X 0
0 W 0 cd
j -H
fl 3 rCf CJ
W O O 1 O
ft -H -3- C
<-* b 3
^^~ CO » -p
J M {* TJ CJ
g -H c3
+3 a H
o ctf oj cd
LA bfl
oo--- o •>!
00 > »
S ^H f?-0
« + O -H
-P10 M O
0 ^H d> TJ 03


33
	 p x
w/ \ /
\ / /
T
/
X

X -
1 1 1 1 1 1 1
^£) -H —
H
?"> ^
5? !-<
O 
.c3 fri
4:) 	
OJ X
in S « |
1 -H X
rn o ;
1 < X
^ &o /
O -H / S3
Jd 9 vJ °
-3" T3 ro X O -
S C / 33 	 ..
Xi A / O / \
I'H ' O33W/ \S3
-3" O X O~-O"--O"Vi /) O
\
ro X
1 1 1 1 1 1 1 1 1
VO CD PH f>j
L_, jg O
lj f. , 1 ,
r=q H-H 1 , -H
o pj a
'H O
OH » Jj
ir\ -~~, CU 3
-p tip (1 -p
o g cu o
X 0) N ct!
a In 0 a H
«H O CU CC3
-=r H H £> bO

0)
J-
fi C\J
a>
J r .
OOOOOOOO OOOOOOOOOO S
t--^Dir\-=tcvic'jH ^cor-vDu-Nj-cocMr-j •->
a)
(OOOi) T^Aonrag % (001) T^Aomsu ^ jrj
92
-------
     Some of the greater reductions in TOG could be traced to the formation
of insoluble salts with the ferric ion, such as for the 3,^,5-trihydroxy-
benzoic acid (Figure 2Ud).  This would not account, however, for the removal
of p_-hydroxybenzoic acid from solution since the ferric salt is soluble.  Nor
would the esters and alcohols, which do not form salts, be removed by this
reaction.

     The water solubility of the compounds, however, roughly correlates with
their removal from solution with Fe3  (Table U2).  The compounds group into
two classes, those with greater than 10 percent solubility in water and those
with less than or equal to 1.5 percent solubility; the former group tends to
have removals of 15 percent or less and those in the second class tend to
have removals of 25 percent or greater (Figure 25)•

     We must also consider not only reactions such as the cross-linking of
starch phosphates with various cations and the increased complexing tendency
observed for K+, Ca  , Zn++, Cu"1"* and Fe    that Wettstein, Neukom and Deuel
(19) studied extensively but also the oxalate or salicylate complexes with
aluminum or iron as discussed by Stumm and Morgan (lU).  Stumm and Morgan
studied the formation of ferric-hydroxo-complexes and found that they tended
to polymerize with other charged molecules at rather exact pH's.  Such in-
teractions might be occurring in our jar tests.

     Additional work would be required to pinpoint the effects on soluble
BODs removal by the organic compound chain lengths or branching, the presence
of phenolic or carboxylic groups and the solubility of the individual com-
pounds, all of which appear to have some important relationship to the over-
all soluble BODs removal mechanism.
                                     93
-------
     TABLE 1*2.  SOLUBILITY IN WATER OF VARIOUS MODEL COMPOUNDS AND THEIR
                REMOVAL IN JAR TESTS WITH Fe3+ AND PULP FIBERS8"
Reference
Q
No . Compound
1
2
3
1*
5
6
7
8
9
10
11
12
13
14
15

16
17
18
19
20
21
22
23
2l*
25

26
27
28
Acetic acid
Methyl acetate
Ethyl acetate
Oxalic acid
Lactic acid
Methyl lactate
Raffinose
Adipic acid
2-Pentanol
Tertiary amyl alcohol
Fumaric acid
Maleic acid
Malonic acid
Tartronic acid
Benzylmalonic acid

Succinic acid
Benzoic acid
p_-Hydroxybenzoic acid
o_-Hydroxybenzoic acid
p_-Methylbenzoic acid
3,l*,5-Trihydroxybenzoic acid
Vanillin
Resorcinol
Catechol
p_-Hydroxycinnamic acid

Ferulic acid
D-galacturonic acid
Polygalacturonic acid
b a
Solubility, Removal,
% %
Miscible
Soluble
10
15
Miscible
Decomp. in HaO
15
1.1*1*
Soluble
12
0.63
Freely soluble
66
Very soluble
Not given; probably
less than 1$
8
0.3
0.8
0.2
Slightly soluble
1.2
1
50
30
Not given; probably
less than ~L%
1
Soluble
?
11*
15
18
1*
15
10
8
1*0
5
5
37
2k
8
9

25
37
30
31*
1*3
50
50
10
8
>*3

58
73
6
99
"Model compounds and % removal shown in Fig. 23- 2l*e.
 Solubility in water, from Merck Index, 7th edition.
                                      94
-------


^_
ill +3 at
rH C eu
,Q 01 CO
3 o ti
O OJ rH ^ CJ
-P Jf O CU
CO ft CU
TJ cu = a
G H 05
O & 13 IA CQ
ft cd cu
CO EH a CQ c!
V S Ctl -H
h C cu
l-l -H +5 TH J3
O 0> bo
O CO CO -p p
TJ t3 -P O
co C C O XI
m 3 P H +3
CU O O ft
f ft ft C
a a a cu cu
3 o o h >
K O O CO CD

Cfl
cu
-p
0
S5







C\J •











OJ.
rH

r ~i
en .
^«-,
S3 .
-r .
I

9 .
fl Ijt ' ,.
-• h'
iS ' o
6'

I 1 1 1 1 1 1 1 1
DOOOOOO O O O
D ON CO t— V^3 LA -^ m C\J rH

.
CO Cfl
Hi -P
0) £
CU k .H
S- "8,
Co • QJ
J3 >> -P +> cd
4J -p a a "
CO -H CU 0 cd
CO CQ H O .H T3
CU CO -H h T3
H CU rO 1) (3 >H
rH P ft 'HO
CU H CM
a > ft cu 4J X bO
" 9 ft O O
C! OH ^t CH
P 0 rH ft « o














^ .
rH





















1 1 1 1 1 1 1 1
O\ co t— vo IA j rn oj



-,





—





—

LA .
OJ




H 0
C\J • OJ—




2
CO • —

rH
rH

5
1— 1

t:-
1 — 1

!A
H

	





OJ
oj —





H
&
-P

eg 0
OJ
Pi
3
r-J
O
M
mj
r- 1
° >
-P •
+
Cm
•H CU
PH
£_,

0 <^ rC
S -p -P
CQ -H
3*
0 >>
-P
(1) -H
J- H
n "^ 'H
° 1 rQ
CO 03
CM >
O
W. OT 3
-------
                                 SECTION IX

                          AREAS FOR COST REDUCTIONS
     This research study on the mechanisms for the BODs removal in clarifiers
was not intended to cover comparative process economics.  However, evaluation
of the data obtained in the laboratory and practical observations in the
field studies indicate that improvements in the clarification process will
reduce costs.  Studies on the flows from individual process sewers demon-
strated that treatment which removes soluble BODs prior to the clarifier is
likely also to reduce the overall cost for the total system of treatment.

SEPARATION OF FLOWS ADVERSELY AFFECTING CLARIFICATION EFFICIENCY

     There are several ways to reduce costs, and careful control of the qual-
ity of the in-flow to the clarifier is one of them.  Proper clarifier perfor-
mance (reduction of suspended solids, COD, BODs, etc.) is dependent upon
sedimentation and flocculation properties of the influent flow.  The input of
lignin and other pulping and bleaching constituents having colloidal and dis-
persant properties is of particular concern.  The reduction or elimination of
flows containing such materials, and particularly the containment and sepa-
rate treatment of large spills of them, should substantially improve the
clarification efficiency and should have high priority in programs for cost
reduction in waste treatment.  Revised handling should produce the following
substantial economies:

     1.  The capital investment for a clarification system will be less when
the clarifier size is reduced;
     2.  Reagent and other operating charges will be less as clarifier effi-
ciency improves;
     3-  Clarifier performance in all categories, including the amount of
BODs removal, should be greatly improved;
     k.  Although no measurements were undertaken in this study, sludge de-
watering, sludge volume and sludge disposal problems were apparent and should
be subject for further development;
     5.  Needs and costs for subsequent (secondary biological) treatment may
be significantly reduced in terms of capital cost and operating charges, and
the need for secondary clarification may also be reduced.

OPTIMIZING THE TYPE AND AMOUNT OF CHEMICAL ADDITIVE

     This project also showed the importance of optimizing the amounts and
types of flocculating agents for various clarifier inflows.  Different chemi-
cal additives are needed for different mills.  Process fluctuations and
changes affecting clarifier performance were apparent within the short term
of each mill field trial.
                                      96
-------
     Operating experience may make possible the development of routines for
addition of chemicals vhen the effluents change.  The rising trend toward de-
velopment and use of computerized systems to control the manufacture of pulp
and paper may logically extend to automated changes during effluent treat-
ment.

     Of the two mill clarifiers used in the field studies, one was utilizing
chemical additives routinely and the other clarifier was well equipped to use
flocculating chemicals.  Capital costs for additional chemical feeding and
mixing equipment would not be a major expense.  Operating charges for the
chemical additives that we used was about 3<£ to 10^ per 1000 gallons of clar-
ifier inflow (Table 1+3).  Since these chemicals might logically replace less
efficient additives already being used, cost could be considerably less in
actual practice.  The efficient use of chemicals would, moreover, be expected
to give a return in reduced overall waste treatment costs.  The actual cost
of chemical additives to remove soluble BODs cannot be adequately assessed
short of detailed engineering cost studies to fit the situations in each
mill.
                   TABLE 1*3.   CHEMICAL COSTS FOR CLARIFICATION
                                                         a
                    Type of       Required,          Cost,
                   flocculant       mg/1        ^/lOOO gallons
                      Fe3+           25              2.72
                                     50              5.M

                                    100             10.88
                     Alum0          100              3.13
                                    200              6.26

                                    300              9-39
                   aAddition of 0.75 mg/1 Hereof loc  812.3  would
                    add 0.5<£.

                   bAs  ferric chloride  (FeCl3*6H20)  at $90/ton.

                   CAluminum sulfate«l6 H20 at $75/ton.
                                     97
-------
                                  SECTION X

                                 REFERENCES
 1.   Gehm, H.   State of the Art Review of Pulp and Paper Waste Treatment.
     Environmental Protection Technology Series,  EPA-R2-73-1814, 1973.

 2.   Das, B.  S. ,  and H. Lomas .   Flocculation of Paper Fines.   I.  Adsorption
     of a Flocculation "by Polyelectrolytes .   II.   Study of the Nature  of
     the Solid Surface and Soluble Impurities. Pulp Paper Mag. Can. 7^(8):
     95-100,  1973.

 3.   Williams, D.  G.  Minimizing Chemical and Fines Buildup in White Water
     by Chemical  Means.  Tappi, 56(12) :ll*l*-7, 1973.

 k.   Back, E.  L.   Note on Dissolution of Wood Material During Pressurized
     Refining and Water Pollution Consequences.  Svensk Papperstid. 77(ll):
     39^-6,
 5-  Standard Method for the Examination of Water  and Waste Water.   13th
     Edition, APHA Method 219,  1971-

 6.  Standard Method for the Examination of Water  and Waste Water.   13th
     Edition, APHA Method 220,  1971.

 7.  Hodge, J. E. , and B. T. Hofreiter.   In:  Methods in Carbohydrate Chemis-
     try, Whistler and Wolfgram,  Academic Press, New York,  1962.   388 p.

 8.  Collins, J.  W. , A. A. Webb,  and L.  A. Boggs .   Characterization of lignin
     and Carbohydrate Residues  Found in Bleach Effluents.  Tappi, 5^(l):105-
     110, 1971.

 9.  Chamberlain, N. S. , and R. J.  Keating.  Water Technology in the Pulp and
     Paper Industry.  TAPPI Monograph Series No. l8: 1*5-7 » 1957-

10.  Gould, M. , H. Lundgren, and J. Walzer.  Primary Treatment by a Flotation
     System.  Tappi 9th Annual  Environmental Conference, Houston, Texas,
     May 15-17, 1972.

11.  Jensen, W.,  and E. Meloni.  Use of Waste Chemicals in Kraft Mill Efflu-
     ent Treatment.  Paper World Research and Development Number, 1971*.

12.  Lueck, B. F. , and B. McCuaig.   Continued Study of Advanced Waste Treat-
     ment Systems for Combined Municipal and Pulp  and Paper Wastes .  The
     Institute of Paper Chemistry,  Report No. 5 for Project 3029, Upper Great
     Lakes Regional Commission Assistance No. 103201183, July, 1971*.  pp. 1*5-
     5U.
                                      98
-------
13.  Wiley, A. J., G. A. Dubey, and I. Q. Bansal.  Reverse Osmosis Concentra-
     tion of Dilute Pulp and Paper Effluents.  The Institute of Paper Chemis-
     try, Environmental Protection Agency, Washington, DC.  Publication
     Number 1020^EEL 02/72, February 1972.  358 p.

14.  Stumm, W., and J. J. Morgan.  Chemical Aspects of Coagulation.  Jour.
     AWWA, 5^(8):971-92, 1962.

15-  Borchardt, L. G., and C. V. Piper.  A Gas Chromatographic Method for
     Carbohydrates as Alditol-Acetates.  Tappi, 53(2):257-61, 1970.

16.  Hrutfiord, B. F., T. S. Friberg, D. F. Wilson, and J. R. Wilson.
     Organic Compounds in Pulp Mill Lagoon Discharges.  Tappi 12th Annual
     Environmental Conference, Denver, Colorado, May 12-1)4, 1975.

17-  Brauns, F. E.  The Chemistry of Lignin, Academic Press, New York,
     1952.  8oh p.

18.  Brauns, F. E., and D. A. Brauns.  The Chemistry of Lignin, Supplement
     Volume, Academic Press, New York and London, I960.  Qok p.

19-  Wettstein, F., H. Neukom, and H. Deuel.  Cation Exchange Equilibrium
     with Starch Phosphate.  Helv. Chem. Acta U*: 19*19, 196l.

20.  National Council for Air and Stream Improvement Technical Bulletin 253,
     Dec., 1971-

21.  Collins, J. W., L. A. Boggs, A. A. Webb, and A. J. Wiley.  Tappi 56(6):
     121-3, 1973.
                                     99
-------
                                  SECTION XI

                                   GLOSSARY
Alum — Aluminum sulfate.

BOD — Biochemical oxygen demand is based on the oxygen requirement  of living
     organisms while utilizing components of the waste stream for growth
     and/or reproduction.

BOD5 — The biochemical oxygen demand in a 5-day test period at 20°C.

CM — Chemimechanical pulping; this is based on a short chemical cook  followed
     by mechanical refining to separate the fibers in the softened chips.

GOD — Chemical oxygen demand is the measurement of the oxygen equivalent of
     that portion of the organic matter in a sample that is susceptible to
     oxidation by strong chemical oxidants (e.g., chromic acid).

Colloid ~ A phase dispersion to such a degree that the surface forces become
     an important factor in determining its properties.  General particles
     of colloid dimension are 0.001 to 1 micron.

Gel chromatography — A column chromatographic technique which fractionates
     molecules on the basis of molecular weight.

Hemicellulose ~~ That fraction of plant stems which is made up of carbohydrate
     polymers other than cellulose.

Molecular exclusion ~~ The mechanism of fractionation of molecules during gel
     chromatography.  Larger molecules are restricted from diffusing  into the
     gel matrix, thus taking less time to travel through the gel than smaller
     molecules.

Permeate — That portion of a solution passing through a membrane during re-
     verse osmosis or ultrafiltration.

Reverse osmosis (RO) — Osmosis in reverse flow through a semipermeable
     membrane when external pressure in excess of the osmotic pressure is
     applied.

Soluble ~ As used in this report is that portion of the solution (solvent and
     solute) passing through a 0.^5 ym filter under 100 psig nitrogen pres-
     sure .

Zeta potential ~ That portion of the total potential drop between the surface
     of the solid and the suspending liquid that is contributed by the charge
     potential between the liquid adhering to the wall of the particle and
     the movable liquid.

                                     100
-------
                                 SECTION XII

                                 APPENDICES


APPENDIX A:  COMPANIES AND AGENCIES SUPPORTING PHASE I STUDIES


                     Bergstrom Paper Company

                     Hammermill Paper Company

                     Hoerner Waldorf Corporation

                     Kimberly-Clark Corporation

                     The Mead Corporation

                     NCR-Appleton Papers Division

                     Nekoosa Edwards Paper Company, Inc.

                     Potlatch Corporation

                     The Proctor & Gamble Company

                     Scott Paper Company

                     Wausau Paper Mills Company
                     Department of Natural Resources, State of Wisconsin
                     U.S. Environmental Protection Agency
                                   101
-------












i


s
o
H
EH
<
ffi
g
o
o



ejj
S
£
s
I— i
CO
o
£H

___^


bO
e
LJA
t—
O
" '
CO
CO
M
CO

o
3

0
o
1
EH
M
•^

w
H
O



PH
t-H
M

5
0

i — |
1
m
&q
1-3





























c
0
•H
oj
§
•0
(U
0













^
m


c

3
UH
w



















x-x
^-

4-1
(U
i-H

C
M

t
I
1
i



i

























p
0
m








14
Q
1-H
0
U


,_,
"i

\r
Q
5
CQ


"
o
03
co

c/3





O
O





r-H
tX
S
,
IT
P
o
w


*«
r-H
O
CO

CO
c/}



















o
T-H
0
u
I

O I-H 3;r-o ^mo

Oj ,— ( CM i— i CM CMOJCMCM fnm^MCM *$• (*} -H rHCMJC1"1 r-HCMjIi-H i-HCM^1"1
^ d M id id td «
« s s s s s s
u
1


-u
en -J~ oo os O I-H
D CM CM CM CM «n m
p
i-t CM en -j-
^ -^ ~^! *^ ^ Cl
CM CM CM CM CM CM















































CO
•3
H

C
CD O
fl "8
-P H
PH
W) ft
•H
rH ^
O

a
£H N
K. d

O o3
^
03 S
0
P O

a e to ..
&, * < ^
O O ^ CD
on ro a; -p
•' •• -p -H H
a3 O -H
O O ft g
S-i O -P
a a o ° d
«5 ft H OJ
O -H N
o o s a s
m  IH ^j
t— t— I ^ c
O a)
a a 
-------
O  4J

pH  -,H

O  C
O  4J

r-l  -iH

O  C
                          .  1
                                           . 1
                                                                                                                   1
                                                                                                                                     1
\c  o m  i      o  m r>- i       CM m  ^o  i      so  -l     i—i   rH     t^C»
                                   rHrHi— IC^J     O-JCNCM
                                             rH     rHCSlXrH
                                                                              rO                ,0
                                                                                rH                rH
                                                                              I rH     rH  CM &  rH

                                                                                •s                a
   -H S
                                                                                                                                              m  m EH  g ^—-
                                                                                                                                                       I  OJ  CO


                                                                                                                                                       '  I    I
                                                                               103
-------









o
M
EH
O
P-.
§
O
e

1
H
CO
i

EH
H
W)
a
LT\
t —
O
---
m
OJ
rH
CO
3
1*1
o
o

;=>
-P
Q)
H
If

O
o
H
P
K
3
W
O
o
H
£5



*
M
^
O
e
g
n
ffi
5
O
m
I
pq
|
g







(3
O
-H
•U
O

13
£>CO^Ol     CNOLOI
                                 rH  CO CT\  I      CT* CO U~l  1
                                 OO  CO CO  OO
                                                             CNI     CN CO  1  i— 1    CNIp-)CNf>    ool     o  r*. oo   I      CMOJC^J      coQOt-Hl
                                                              l      CMOOrHt     -CJ-CNOI      vO 1^- .   •

                                                                                                                                            a  a  »  « M

                                                                                                                                         j  ^    
-------
         O  4J
         i—I -iH
         o  c
            -Q

          - rH
          <  o
             00
         2  6
EH
H
5=
                                                             ITl  rH CO
                                                                              r*.r^.ooi
                                           \£)t-Hr^mr-i-H    CMOOCOO     mo-^cy>
                              CO-vfCMCM


                                                                                                         co o

                                                                                                                                                         rH i—I  CM CM
                                                                                                      OCT>O>|     CTtOCO)     CO  i—I  i-H   f     CMincMl

                                                                                                      I-H r— QO        r^oor-        oooooo        COCOON
                                                                                                                                                                            1  -p  +J  Qj  S  >i
                                                                                                                                                                          H  ..  »      h
                                                                                                                                                                          < t>-  t--     P   W
                                                                                                                                                                                        O   cd
                                                                                                                                                                          I    a  s  oj  ^
                                                                                                                                                                              O  O  W  I   QJ

                                                                                                                                                                          OJ (in  CM  73  OJ   03

                                                                                                                                                                          •Hi   I    §  I    "
                                                                                                                                                                          W                I
                                                                                                                                                                          O rH  OJ  M


                                                                                                                                                                          I               -H
                                                                                          105
-------


















EH
fe
g
cn
;=>
o



w
o
5
„
K
S
5
ga
o
CO M

Q-> CG
O
H ^
U 0
<[J o
0 K
j— 1 pq

p 5*
§ W
« M
P CO
H O
0 E
o 



rH
"i
H
If
P
o



»

rH
0
en

3
en




O
o
o




;H
00
i


ir

CQ





rH
O




CO





















Q

O

eu
i— i
3
rH
O
cn



2
o
H
rH
o
en

CO
CO


CO
4_)
•H
Cl
JZ)

^
43
3
O
CO

rH
4J






rH
OO
R


ffi


CO
4-1
•H

^



CU
rH
|
O
C/J



rH
crj
4J
O
H






"bp





ffi


tf
0)

•H
CO
o

1
U
>i
P


CU
4J
ffl
P
^O O ON 00 (JN CO
• . . 1 • . • 1
m co vo I oo co r*-. I
OO 00 ^O CO ^D CO
1

co oo *cj- o m oo
. • • i • . . i
-vj- m ON. i rH m ON i
r-t CM CM CM CM





m oo vo o ^ ^ i
CM OO CO 1 CO VO -sf 1
CM CN CO CN CM

"^ °. "^ "^ "^ ^ ^ ^
OO^COON O OCOvD
cNvocor^. -J3 cor^co




CM CN 00 1 rH cn rH 1
ON in rH | J3
ONrHinvo co miDcn


rH eji »H oo in o
m IH m i rH \o , OCNf*-CN rH  i
O OO rH ON 1 in
rH 1

O O O Kf  CO O CO 1
CO rH r-l CM rH



CM vD vO 1 -J- OO 00 1
ON O ON 1 CT\ f^ O 1
rH ,H



vo co  r*^ rH ^o vo vj 1
tjD . • • . 1
g r- oo oo oo r- r-
o
0
rH

rH i— 1
CU rHCMlSrH rHCMjE*-1
« S S
MH
rH
3
CO
<0
3
O ON O
rJ [•*• 00

o

CO <1"

ffi ^- >*

























































OJ •
e co
-H ,0
-p rt
H

d c
QJ -HO

CJ i — [ QJ
S 0 rH
0 rC ft
ft
cu E <
-p tt)
3 -p -p
S cn cfl
•H >s

O ^ N
^ . . o >>

° ft rf ^ S ^
CM 0 0 S e
CQ fvi !>-} O (U
.. .• ' — f -t-3 ^*>
T^ 1 	 f. 	 1 	 1 -H rQ
C1^1^ n5 ffl
5 S w ft a;
«+J -^ s
PQ p. rf ^J U r-t
•^ T C cd


•
• PO rn ON S
^ [11(11 1 ^"^
O CQ
1 a S CU ,£ o3
O O P^ 1
ii 
-------

1 i

•
^
,,„_ 	 ,.i
1 !


H H II
Pt P) I1 i '

• i
• nj 	 ,j
Mo i

m m-~-s
t— t— 03 "I1
-p -P -p '
-P -P 01 53 5 r
odd i
Seg££ I i
ci ft o H H i
0 0 rH 1 1 '
ro p-> H '
t-c-ikj^ ;
i i
i
» j , '
.1 li
1 E
1 i
•"O . '
cO 1 i
rH
CO
O
O
rH
-------

l~
1


1

1

1

1
L_

1
1
I

LZ
0
  ro
o o ••
-y ....
d t— ^- O ^> -P
-P -P !3 U
W O O -H OJ 0)
-p -p 01 pi 3
0) Q H H i 	
fi • • pi tn 1
a v a
w o
1 1
n
1
« rH 1

1
I

w |
1
« 0 1
-------
 O
O
o
OJ
 a,
 o
•H
-p
•H
Tl
        01  
-------
       t- [-- 
                                                                                                  ro
O
O
                                                        CM        H

                                                            papuadsng
                                                                                         J3
                                                                                         O
                                                                                        •H
                                                                                        •P
                                                                                         03
                                                                                         O
                                                                                        •H
                                                                              H
                                                                               O
                                                                                                         bO
                                                                                                         0
                                                                                         O
                                                                                         ^
                                                                                         •H
                                                                                         H
                                                                                         O
                                                                                         CO
                                                                                                          0)

                                                                                                          03
                                                                                          O
                                                                                         U
                                                   110
-------
o
s
o
o
 ~H
 g
-------
                                            01

                                            rH

                                            P<
s
•H
at
-------
00


CM


d


O

o
 
-------

1

1
ro I
CM
H f
CO /
O
0
r! i
o' 1
£ i
 .p -p 1 	
-p d d
O O -H 0) 0)
.p -p W ^J 2
O H rH
0 fi i'rt vl
• • O M W
di P< 0

oo ro <-i < M [_
— S 1
i i i U 	
01 , |

•H
CO
£
"^ 1
1 1

•H (

t) |
«H
-P+ |
^H pq L.,.
3 w '
c5 a" i
" °
•H 0

H
P(
m o
i i
pp w
<; w co
«
-------
                                                                                                                           ft
Ferrous Sulfate

Hercofloc 812.3
G. With

Hercofloc
•d
CO
-p
to
3        o

Tl       (H
< -P  JH Vi -
    3  m  o
a  o  a  o
<

o
BOD
Tota
Soluble
CH
                                          (U  (U
                                         H H
                                          ft ft


                                          to  to  s



                                          ^  ^ o
                                          ft 0} 00

                                         o o t—
                                         m oo —-

                                         t— t—  ID  -p  -p
                                               -p  C  C
                                          o  o -H  a)  
-------
o
o

-p
ft.
O
H
O
O
   kOQr-
    300 -
200-
   100 -
F. Sulfuric Acid Mosiuee
pH Adjusted




6
-
Uc


—
—







H. Ferrous Sulfate with
Hercofloc 812.3
Without G. With
Hercofloc Polymer
812.3 812.
960

—
M/




fh flf
ABABAB AB ABA
3


—
B
1 2 K 1 2 K
77th 78th








—






fcl






—




ABABAB ABABAB
1 2 K 1 2 K
79th 80th
H 300
If
M
tf
•H
ra 200
•a
-------
                                                   1O|    O O "">    COCO        CO     ON CO       O ON CO    ON rH CO


                                                                                    ON     r*. r-  I     co I-H ON    ror^-J
                                                                                    CO     i-llTl       CMi-t^O       *M ,-t
oo    r- CM r-i|   co o o!    m vO •*    -* ON
rH     1 CM COi   F-^VnJ    CM     rH







                                                           miOCol   CNCO       CO    O\U-*CM|   COrHt-H    O>ON«O
















                                                                       o                   r*. -*                    o oo

                                                                       oo                   r^. r-                    o r-
              o
              en




v            %  £                 ^o^    I-n    ^str.

                                   rHi-Hr-4    CMCMCM          r-<    (MCM       CM    CMCMCM    i-HCMCM    r-lfOCO



3

                 01

                                                                                    O    O**00    f-rHO    CMCOOO
                                                                                    "^    CM r-H ON    
                                                           ONOOON    -JiHrH    OO    CMOSin    rHlTlt-l    kO^O<-<



                                                                              W
                                   .,_...    ...   _     or^OO    irip^       00    O p  ^H ol   r^ecenl    oo    o\oo ml   -•   -  -  '
              p £






                             vo     'tf o r-|    CM co col   r*. CM -*|   co CM ol


                 o
                 n



            •iQ I-H


           _     r_           , .     CM CO CM I    CM CO CM |   i-H CM i-<|   fl CO CO|    CO    CO CO CO'   CM CO  m|    CM CO <






                                                           %o co vj-    CM CN CM!    en    o in CM    r*. oo  CM!    SJ ?n i-i|
                                                                                                                                  ff
                                                                     117
-------
                       m o *o    o o~i m    cMvocsi    CM cn »M    i-«o\en    cMinc


                       rH in •—i    CMCMI-H    rH    CM    CM (-H PJ    rHinen    in \c en    -5 in id








                       r-lCM      CMi-HI    CMCM       mCM       mCMi-H    CMCnCM    CMCO


       °'~  '
         33




      at    H




                       ot" in      CM" W       en  cn       io '-*       co in       r-." r^-      »    \o    \o    vo



 g


         t-i «
 JL,      o-u          ***>?    ^^y?    c'OiD    5K?°°.    ^5^?    r^r?^    ®r^*?



4-


 £










                       co^o*    coi-*>£    CMO\O    oo r- co    -* oo TH    r-iocn    -4- -± m


            H





                       ^ co ^o    co o i




                       oor-jrH    f^-***^    cspo    ^rr*rrH    COOCM    PP*0    COCMI^

 1

 a
 Q                    COOCM    CMO»r-i    r^O^O    CO-^O    Tl col   co co ml
         ....          _> in en]   o       CMO>       »no>       I-H\O       _._

                                    :•*.   i    ^^   i    ^
-------
O OJ
sr
                                       ^ O. O

                                         oo r-i
                                      i-»cMi-«!  CMCMCM!  CMCMCM
                                               co en     m m
                                               r-* m  I   \o r-

                                               oo cc     r- r*»
                    r-l CM ;£   I-H (
                                                                         fl K



                                                                         ^!
                                                                         O. CO rH
                                                                             CC
                                                                         O O C
                                                                         m nn 3 '

                                                                         t-^- t^-   . .. -H
                                                                    O •*-> t~- t— S

                                                                    s •s i   i  i
                                                                 > ;
                                                                 (0 J3
                                                  119
-------
O %£J t-l    U-l r-l •    (0  p<  O
                        •a   o o rH
                        a> « on ro r-<
                        L< i>	H
                        o 4^ t- r- S
                        s-s  i  i   i
                        w S,^ nj s

                        p o
                                          120
-------




















oj
l-l
"i
0
°

|

I

S

§
i
0
1
EH
CO
1
M
3
O
1
0



*























i
s
i**














s

|
M












/->



i
rH
C






















8









8
,3
O
U
fJ
M
i

CD

*
rH
O

3
CO



rH
O


rH
M
g



§
pa




CO

3
CO



i




g
rH
O
CJ
rH
ft
•§
tH
O
CO

1-4
5
o
H
r-4
O
CO
J
3
CO
s

rj
s
o
|H
O
CO
!_,
4J
&


d
f

X
a


*
§

o
A
3
o
CO



a
4J




1




as
A
3
tt
JJ

•H
i
u
o
P^ . . 00 CO 00 *O rH O , O 1** O If) rH
CM ll 1 ^f CO rH 1 i-t 00 O ' CO CO 1  rH CO rH CM CO O \O t^- in CM rH in
mmr^ m^*nr* mo m\oor**
rH rH rH
oo incoaoi o m m t co co 1 co oo oo I
*O in rH O 1 OrHinl CMvDl rH \f> vO 1



CM CM CO h* >^| -4" rH O\ ml \O O OOl \O CO V> rHl
CM CMCMCMCMl CMCNCMCMj rHCM rH| rH CM CM CM,|
•d



Oh CMCMOOMfl r>.vOr^.O| CN CO ml OCMrH^I
rH COCO^O^I rH CO VD -J CM %£> s* ^T^COxol
CO g, CO CO CO CO| CO CO CO -H in [ irtr*»u>< oo r— i so >n oo I


rH CM j: ^ScMjE rSrHCMX rHCMX rHCMC^S

O rH CM CO ^
m m ft m »n

































CO
a>
rH
m
c
§
v3
C 01
•rt 41
H
u *
&4 3
M O
!H
Vi • 4J
O fl
a o.
S H 1
d m w o P*
h 11 0) fl,
> -P
m Qt oJ r-(
 QJ
O O -H >
W -P +> U O
S> 0) Q, 0 <3
O
•rt O O H
1> m m on H o
h 0) •• •• -d O
o +3 1— t— B H
O -H . . . *M
n n 1 1 1
MO p>
ft) pt H W 33 >
•a a d
CO «)
So E
da o
121
-------




















to
e
0
o
CO

1
§
EH

O
§«„
<2 WJ

EH LfN
M r-
0
o ro
o •
J OJ
i—)
Q CO
H
5 0
o o
o o

< a
§J-3
PH
CO

M
i
o



I
o
w



EH



































G
O
•H
U

•o
^ P*» CM
II rH rH rH




ON i-H -* CM P^ st — 1 CM P-
o oo CM m so o
m co co m  Pi Pi


C mm
O 'o

ri-J g g [<1
>
K ft o! H
cd cd
,0 O O C
en on of
d,, ....
Oj -P
O O -r4
OJ O

H a a 3
oj • . o
> a & o

•t) O O r-l
4) tn m OO H
1* 01 •• •• 'H
o +J t— t— S
CO 'in 1 1 1
JH O
ID p,H C\J S
TJ &
a o



122
-------
a
o
ta
o u
i-i i-f
o c.
l?
|^ CM
00 rH |
CM CO
1 1
vo co m
O O CO
rH 1
1
•4- co m
r- ON co
CO CM
o r-
CO CM I

CM VO CM
CM vO CO
VO O CM
in in ON
rH 
vO CO
rH vO 1
CO CO
1 1
CM CM CO
^ ^ rH
1 1 1

rH CO CO
CO vD **
CO CM CO

vD CO 1
CO OO
O 00 CM
CM CN CM
o m CM
O r» co
CM CM CM
vO CM st
O O CO
CM CO CM
ON o m
vD vD --d"
in CM CM
CO -

CN ON t
P^ rH 1
i-t CM



CO rH O
in \o vo
CM CM CM
rH CO CM
CO 
r-0 i
1

CO O sl-
ON O CM
rH m co
I t 1
oo r-» o
vO vo O
CO vD Sf
P^ P-
CM -Ct 1
ON ON
3CN CM
CO ui
rH rH
ON O O
CO CO ON
rH
00 00 0 1
r-t





m i
in 1
in




3 j
ON


r-« S




CO
vO
























Ul
ft
1
to

-p
H
0)
H
C
•H

° i
V rH
bD
rf w m o
^ Q) V fa
QJ H H H
> ft ft


O T3
T3 6 B N
at • • ^»
Ul ft OJ H
cd o3
,0 O O C
m m 3
0) .. ••
*-. f-t- to
cd -P
O O -H
to .p +3 to
1 sal
> cd ft o
T) O O iH
-------
i ; .
i :
i :

i '
i :

i :
i :
i :
i :

i

i :
i :
i
i :

i

; i
i :
_«j jjj atctnres ofi —
ail !
"• II
M- 1 :
Ml W W II

7 oS
H s R* " a 	 	 i
• • o CB r^ — i
tf Pt O V tl | |
OOH H A ||

11111 1 !
r-l CM fc, 0 O | 1

i :
1

i :

i
•P H
£ M (~|
I 	 1 	 I 1
' - 1 i
i :
i ;
i ;
i i i i i i i i
8 °8°§ooc>0
Q
Q
Q
CJ
ft.
u
Q
rH
O
O
a
w t-
U H
a
Q -»
0
a
o
o
°H
O
a
ft.
o
a
H
u
0
o
O H
o
H
O
a
0
a
H
0
.« §
rH O ft
Figure C-1. Total and soluble BODs in base-line study (no additives) at Locks Mill.
  5(103
124
-------
m  «  •
     a

e  e *

P, oJ O
     ro
O  O ••
ro  m t—



   So H
   -fi 01


s  al
t-l- 3C


 I  I   I


i-t OJ fr,
                                                  ogf

                                                                                                                          r
                                                                                                                          r
                                                                                                                         r
                                                                                                                                     a
                                                                                                                                4> V( O
                                                                                                                                H  O -H
cks Mill
at
no additives
t
-p
 CQ

 
-------
1 H
i :
i :

i :
i ;
i ;

i ;
i ;
i :
s \ ;
2 i :
w | i

i :
C - Influent
D - Effluent
1
0) V
H H
If.
m
00"
t— t— a
•p
O O -rf
4> -p m
a a 0
d Pi o
com M
• - •• 'i-t
t~ t-se
i i i
M CM fc
i ;
i j
i :
I !
i ;


i :
i :
i ;
i f
i :
I :

p
1 !
i ;
i :
i :
i

i ;

i :
i :

i j

i :

i j
( ;
(
o
ITS
m
i ;
i :
1 I 1 1 1 1
Q
cu m
o m
Q
o
Q
CU CU
O PO
Q
H
O
P
(K
O
Q
CM H
o fi
a
0
n
0
Q
CM 0
O CO
Q
0
a
o
Q
CM O"v
U OJ
r-l
Q
fc
O
O
CM 
-------

f 	 ' 	 -
1 	
1

1
1 	
1
1
1

1
1 	 J
J _. 1
p< Pi t 	 '
« « r5 I
a a a 1 	 	
p. oj o
CO
0 O ••
rrl ro (—
^2 1 	 1 	
+J +> 01 1 ,
. .a i
sag 4> ^> fJ 	
d p, o u a) t — I
0 0 rH rJ ~l 1
III II
H CM I* O Cl
1 ,
1
1

1


	
1 1 1
Q
CM ro
a
H
u
a EX
0
... CM CM
Q
a
Q H
a
rH
a
Ex
a
Q
H
Q
Q
CM O\
a
o
a
0s*


o

°rH
O
V Vi -H
rH O -P 1
ft BJ 1
o o o o 3 S" S o
O O 0 CQ Q O, O
ro CM rH O CO
c
r


i
P 	 L
1 	 r
d
i

C
r

	
c
r


, i
1 	 r
1 	 r
r~

i
i i i
o o o o
o o o
VO -* OJ
Q
cu ro
u (*")
Q
rH

Q
Q
a
rH
CJ
Q fc.
a
H
o
Q
CM ON
Q
rH

a
CM ro
a
o
d.
a
-------
1+00



350
300
250
D
•H
|> 200
U)
Q
S
150
100
50

B°DS i - 7:30 a.m. to 7: 30 p.m. sample
n2 - 7:30 p.m. to 7-'30 a.m. sample
Total F _ J^-Q composite 7:30 a.m.
Soluble
C - Influent
D - Effluent








__


















—

p- 1






•-



















--









—





—



--





.M
















—






,— i
. _












._











r-















—

.-







"~




..


















•-







7 ~-

















--































—


..






~













—











t-











1—









CDCDCD CDCDCD CDCDCD CDCDCD CDCDCD
Sample 1 2, F 12F 12F 1 2 F 12F
Day of 36 37 38 39 1*0
Operation
Figure C-5.  Effect of ferric chloride (75 mg Fe3+/liter)  + Nalcolyte
             T3C32 (l mg/liter) on BOD5 at Locks Mill.
                               128
-------
sea
                          "8
C
	
c

c
1—

c •**
c
r
, 	

8 8 8 §
CO *" -* CO
0)
•P
•rl
H
A H
o ** H
o -* CVJ
CO
0 O
•-" m
: D c D
2 F
9
Falcolyte r
°H +
?H H
V H
-P -H
0 -H S
t- H
3 D C D C D <
2 F 1
36 37
Df ferric chloride (75 mg F
r and suspended solids at L
o
« -P H
o 00
0) O
g 
-------
350

300
250
-------
   300
01
-p
•H
O
o
O
O
   200
   100
     0
1 - 7=30 a.m. to 7=30 p.m. sample
2-7:30 p.m. to 7=30 a.m. sample
F - Mill composite 7:30 a.m.
                                                         C - Influent
                                                         D - Effluent
                                                 1h         it
        CDCDCD
 Sample   1    2    F
 Day of       kk
Operation
                              CDCDCD
                               12F
                              CDCDCD
                               12F
CDCDCD
 1    2    F
      1*7
   8001—
    600
-P
•H
 en
•d
 o
 CQ
-s
C!
    koo
    20°
      0
                     n
          n
                              CDCDCD
                               1    2    F
                              C  D C D  CD
                               1   2    F
C D C D  CD
 1    2    F
        CDCDCD
 Sample   1   .2    ™
 Da7 of        kh               l|s                 U6                ^7
Operation
      Figure C-8.   Effect of ferric sulfate (100 mg Fe3+/liter)  on color
                   and suspended solids at Locks Mill.
                                      131
-------
Q 2 s
9 2 °
CO f-1 W
i ;

~T
CU » ( 1
H H ~"
to to • I 	
&
a a d
PI it <~> , . , ,
ro I
o o •• 1
(M <•> C—
f— [~— (U
-P
C, 0 .J 	
•p -p to |
Q 1
• • Ui «^.
a a a -p -p
• • o a a
a) Pi u 0)0)
0 0 H H £
^ t^ g 5 W
III II [• 1


\



i:
i
i

i

i
i
i :

j



i:
1
i:
i :



i :


i :
i :
i;
i


i :

1 i

i
i :
i i i
00 0
l/\ O ITv
oo n c\j
i i i
P
P
R
O
P
O
P
O
p
p
o
p
o
R
O
O
0
p
p
o
p
o

o • o o o o
O L^ o ir\
(M iH H
• >
H
H
CM-* VH
IA g
to
o
3
•8
fa in
*„§
"^ E!
0)
-P
•H
i-t
- 1?
O
fa -P
a
CVJ H ^
i/\ H
1
0
AJ 0
§
m 
-------
  1*00
  300
  200
  100
    0  	
          C D
    Sample  1
    Day of       50
   Operation
               1 - 7=30 a.m. to 7:30 p.m. sample
               2 - 7:30 p.m. to 7:30 a.m. sample
               F - Mill composite  7:30 a.m.
               C - Influent
               D - Effluent
                                                                                 1
C D C D C  D
 1   2   F
     51
C D
 1
C D C D
 2   F
 52
C D C D C D
 1   2   F
    53
C D C D C D
 1   2   F
  8001~
5} 6oo -
r-l
•3
3 uoo
o
CQ
Tl
3
C
01
p<
to
3> 200
0

_







~~











n n





















n n










-





i










-, n









^^








-i n
          CD        D      CDCDCD      CDCDCD     CDCDCD     CDCDCD
    Sample  1         F       12F        12F       1     2   F        12F
    Day of      50              51                52              53                 5k
   Operation

             Figure C-10.  Effect  of alum  (300 rag/liter) on  color and
                             suspended solids  at Locks Mill.
                                           133
-------
1*00
350

300
250
t,
-------
c a
« 
H •-»
p- Pk 1
3 f '

o o
ro no
t— t—
0 O I
o5 ft to «J p
0 0 H &0 p
1 1 1
H CM [x,
1 1
o o
o o
CO CM

	
	
_ 	 ^


1
1
f 	 	
1
8
0
p
«
O
p
r>

P
o

o

P
o
p
p

c



c
1
c

c

„ vi-S 1 	 	 	 : 	 	 	 :
H O +>
P, «!
i^^ i i i i
o o o o o o
O O O 0
CD VO -^ CM
°r<.
o t-
o
Q OJ CM
O
P
P
Q
U
P

P
&H
0
P
CM C—

$
C
L
                                                           O
                                                           H
                                                           0
                                                           O

                                                           a
                                                           0
                                                           0)
                                                           -p
                                                           •H
                                                           H
                                                           CO

                                                           o
                                                              3
                                                               m
                                                           
                            pepusdsng
                                                           •H
135
-------
      §
3  I
o  3
H  on
                                                                                                         I!
68
cks
at
gravity sedimentation
in second base line
d soluble BOD
C - Influe
D - Efflue
sag

* * 0
0 0 °
on m °
^ t— [j
     »— i
0  0 '£)
-p -p *
 .  . 1
2
0
Sample
Operation
Total
                                                                                                                      OO
                                                                                                                      H
                                                                                                                       I
                                                                                                                      O

                                                                                                                       OJ
                                                                                                                       tjQ
                                                                                                                      •H
                                                                                                                      P^
                                                             'saoa
                                                         136
-------
Influent
Effluent
a.m. to 7:30 p.m.
p.m. to 7:30 a.m.
composite 7:30
1 - 7:
2 - 7:
F - Mi
5g
                            t&
c
i
	
c
r

r

C
r
i
c

c
r

C
[ 	

i i i i
o o o o o
o o o o
co \o ^t c\j
p (^
p
fe
o
p
OJ
Q
p
ft*
o
Q
P
rH
O
°c.
p
H
P

Ph
O

Q
O
OJ
p
rH
CO
Day of 59 60 61 65 66 67 68
Operation
Figure C-lU. Color and suspended solids in second base line (gravity sedimentation) at Locks
                               137
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. 2.
EPA-600/2-76-221
4. TITLE AND SUBTITLE
REMOVAL OF SOLUBLE BOD IN PRIMARY CLARIFIERS
7. AUTHOR(S)
George A. Dubey, Averill J. Wiley, John W. Collins
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Effluent Processes Group
The Institute of Paper Chemistry
Appleton, Wisconsin 5^*911
12. SPONSORING AGENCY NAME AND ADDRESS
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio ^5268
3. RECIPIENT'S ACCESSION-NO.
5. REPORT DATE
September 19?6 (issuing Date)
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
1BB610
11. CONTRACT/GRANT NO.
R-803119
13. TYPE OF REPORT AND PERIOD COVERED
Final 7/20/7U-9/30/75
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT
   This project was directed to evaluating means for increasing BOD removal from pri-
mary treatment systems treating pulp and paper wastes.  An improved understanding of
the optimal conditions for soluble and colloidal BOD removal should permit increasing
efficiency in total organic removal.
   The initial phase surveyed 12 mills to obtain data on total and soluble BOD^, COD,
suspended solids and color removal from sedimentation systems.  This data was used to
select mill effluents for additional study.  These laboratory studies showed that, with
proper flocculating agents, soluble BOD removal could be markedly increased.  Soluble
BOD was defined as that organic load passing through a 0.^5 Um filter.
   Measurement of sludge volumes and sludge dewatering characteristics produced from
use of iron salts, alum and polymers was outside the objective of the study.
   Gel chromatography studies showed that low molecular weight biodegradable residues
and colloidal materials were flocculated and removed.  Studies with model compounds
indicated that increased removal is apparently related to pH of the solution and to
functional groups, chain length, branching and solubility of the compound.
   Chemical costs may range from 3$ to 10$ per 1000 gallons.  Cost reduction and
improved clarifier performance can be achieved by elimination of overflows and spills
within the mill that are antagonistic to efficient sedimentation due to  dispersant
action.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
Precipitation (chemistry)
Coagulation
Flocculating
Oxygen demand
18. DISTRIBUTION STATEMENT
Release to Public
b.lDENTIFIERS/OPEN ENDED TERMS
Primary treatment
Water pollution control
Pulp waste treatment
Paper waste treatment
Chemical degradation
19. SECURITY CLASS (This Report)
UNCLASSIFIED
20. SECURITY CLASS (This page)
UNCLASSIFIED
c. COS AT l Field/Group
13B
21. NO. OF PAGES
152
22. PRICE
EPA Form 2220-1 (9-73)
138
                                                                  : 1976 — 657-695/6129 Region 5-11
-------
Z   TJ
m   £
0
-------