ENVIRONMENTAL PROTECTION AGENCY OFFICE OF ENFORCEMENT E PA- 3 30/ 2-7 8-O15 Ei a Iu at ion of the Treatment CapabiI/1y of the West T re at m en t Pi ant Fit eh burv, M a s a a eh // 5 et is o NATIONAL ENFORCEMENT INVESTIGATIONS CENTER DENVER.COLORADO AND REGION I. BOSTON. MASSACHUSETTS ffc» STAfr. DECEMBER 1978 ------- Environmental Protection Agency Office of Enforcement EPA-330/2-78-015 EVALUATION OF THE TREATMENT CAPABILITY OF THE WEST TREATMENT PLANT Fitchburg, Massachusetts Arthur N. Masse December 1978 National Enforcement Investigations Center - Denver, CO and Region I - Boston, MA ------- CONTENTS I. INTRODUCTION 1 II. SUMMARY AND CONCLUSIONS 4 CHEMICAL COAGULATION AND SEDIMENTATION 4 CARBON TREATMENT 4 GENERATION OF HYDROGEN SULFIDE (H2S) 5 INDUSTRIAL LOADING 6 ABILITY TO MEET PERMIT LIMITATIONS 6 PILOT PLANT DATA - POMONA, CALIFORNIA 7 III. EVALUATION OF DESIGN AND OPERATION 9 SELECTION OF PHYSICAL-CHEMICAL TREATMENT 9 PLANT DESIGN 11 PLANT OPERATION 13 IV. PHYSICAL-CHEMICAL TREATMENT AT POMONA, CALIFORNIA .... 22 GENERATION OF HYDROGEN SULFIDE 22 PRESSURE DROP AND BIOLOGICAL ACTIVITY AS A RESULT OF NITRATE ADDITION 24 APPLICABILITY OF POMONA FINDINGS TO PROBLEMS AT FITCHBURG 25 REFERENCES 27 TABLES 1. Pilot Plant Data West Treatment Plant Fitchburg, MA . . . 10 2. Summary of BOD Carbon Column Effluent Data, West Treatment Plant, Fitchburg, MA 18 3. Allowable & Actual Loads from Paper Mills, West Treatment Plant, Fitchburg, MA 21 4. Comparison of Physical Chemical Treatment Plant Designs at Pomona and Fitchburg 23 5. Summary of Performance of Physical-Chemical Treatment Plant at Pomona, California 23 6. Carbon Capacity at Physical-Chemical Treatment Plants . . 26 FIGURES 1. Process Flow Diagram, Physical-Chemical Treatment Plant Fitchburg, MA 12 2. Activated Carbon Tests, West Plant, Fitchburg, MA ... . 19 ------- I. INTRODUCTION The City of Fitchburg, Massachusetts, with a population of about 43,000, is 64 km (40 mi) northwest of Boston on the North Branch of the Nashua River. The industries in Fitchburg produce paper and allied pro- ducts, cutting tools, plastics, cotton and yard goods, fabricated metals and machinery. These industries, particularly the paper industry, use all the river flow except during periods of high runoff.1 Fitchburg owns and operates two wastewater treatment plants. The West Plant, the subject of this report, was designed to treat 0.67 m3/sec (15.3 mgd) by 1990. The plant treats over 90% industrial wastes, mostly from the paper mills, by a physical-chemical treatment system made up of chemical clarification and granular activated carbon adsorption. The East Plant, designed to treat 0.54 m3/sec (12.4 mgd) by 1990, is a two- stage activated sludge plant which treats mostly domestic wastes. The plant was designed to remove phosphorous by chemical precipitation and to biologically oxidize ammonia and organic nitrogen compounds to the nitrate ion. The Massachusetts Department of Natural Resources has decreed that the North Branch of the Nashua River, into which both plants discharge, should be of the quality necessary for boating, fishing and general recre- ational use. The Massachusetts Division of Water Pollution Control issued an NPDES* Permit (MA0101281) to the West Plant on December 27, 1974. * National Pollutant Discharge Elimination System. ------- 2 The permit requires the effluent from the West Plant to meet limits of 8 mg/1 biochemical oxygen demand (BOD) and 8 mg/1 total suspended solids (TSS) as a monthly average, 12 mg/1 BOD and 12 mg/1 TSS as a weekly aver- age, and 15 mg/1 BOD and 15 mg/1 TSS as a daily maximum.* The West Plant has not been able to meet these limitations. From March to September 1977, the BOD of the plant effluent ranged from 40 to 71 mg/1 and the TSS ranged from 20 to 35 mg/1. Because the plant is unable to meet the permit limitations, EPA Region I requested the National Enforcement Investigations Center (NEIC) to conduct a technical evaluation of the West Treatment Plant. Specifi- cally, NEIC was requested to evaluate the design and operation of: 1. the chemical coagulation and sedimentation systems (primary treatment) 2. the carbon columns, including the column lining and backwash system 3. the carbon transfer system. NEIC was also requested to evaluate potential and existing problems i ncludi ng: 1. the generation of hydrogen sulfide in the carbon columns, 2. the effect of excessive loadings from industrial contributors on: a) the ability of the plant to meet the permit limitations, b) the carbon regeneration frequency, and c) the ability of the carbon regeneration system to meet the resultant demands, and, 3. the potential for excessive carbon loss during regeneration. Based on the plant evaluation, NEIC was asked to appraise the capa- bility of the system to meet the prescribed limitations and, as necessary, recommend changes in operation or design to achieve compliance. * Settleable solids, dissolved oxygen and total and fecal coliform are limited also, but these limits are not pertinent to this study. ------- 3 The West Fitchburg Plant was inspected on May 4 and 5, 1978 by EPA personnel from NEIC, from the Office of Research and Development in Cincinnati and from the Regional Office. The EPA visitors met with Mr. George Chretien, Plant Operator, and Mr. Jim Taylor, Chief Engineer. The information obtained through discussions with the Plant representa- tives and their timely response to a follow-up letter was used in the preparation of this report. ------- II. SUMMARY AND CONCLUSIONS The West Plant of the City of Fitchburg, Massachusetts, which receives more than 90% of its influent load from paper mills in the area, has been unable to meet the 8 mg/1 BOD and 8 mg/1 TSS limita- tions prescribed in its NPDES permit. The treatment sequence at the plant consists of chemical addition, flocculation, sedimentation and adsorption on columns of granular activated carbon. NEIC conducted an inspection of the plant to evaluate its design, operations and treatment capability. The following summarizes the inspection findings and conclusions reached. CHEMICAL COAGULATION AND SEDIMENTATION The pumps originally installed on the primary sedimentation tanks could not remove the sludge as fast as it was being produced. These pumps have been replaced and the coagulation-sedimentation system is producing an effluent with an average TSS concentration of less than 35 mg/1. If the final effluent is to meet the 8 mg/1 limitation, more rigorously controlled operation of the system is needed to re- duce the average TSS concentration of the clarified wastewater. CARBON TREATMENT The protective lining applied to the carbon columns has proven defective. In many places, the lining has split and corrosion of the base metal is occurring. The cause of this problem has not been deter- mined nor has any corrective action been taken. ------- 5 The suface wash mechanisms on all columns have been sheared off, probably by incorrect backwashing procedures. Repair or replacement of these mechanisms is awaiting recommendations of the City's consul- tants, Camp, Dresser & McKee (CDM). When a column of carbon is being regenerated, the carbon is moved hydraulically in some locations and pumped mechanically in other loca- tions. The hydraulic transport systems are working satisfactorily but the pumps used to move the carbon mechanically have experienced severe wear of the impeller linings and require expensive downtime and maintenance. Another type of pump has been recommended but is not yet in service. Insufficient data are available to support any determination of the carbon losses during regeneration. Similar systems have experi- enced losses of 5 to 10%, and the losses at Fitchburg should be in this range. GENERATION OF HYDROGEN SULFIDE (H9S) Because the wastewater entering the carbon columns has a high (~60 mg/1) contentration of BOD, bacterial growth occurs on the carbon granules. In the absence of dissolved oxygen, sulfate-reducing bacteria proliferate, resulting in the production of hydrogen sulfide, which is released to the atmosphere when the effluent from the columns is discharged, thus causing a serious odor problem in the area. The City of Fitchburg alleviates this problem by shutting down the columns periodically and adding enough caustic to raise the pH to 10.5. The carbon is allowed to soak for several hours at this high pH, thus killing the sulfate-reducing bacteria. In the winter, this procedure must be carried out every 4 to 8 weeks. In the summer, however, the bacteria reappear within a week after a column resumes operation, and the treatment must be repeated weekly. ------- 6 INDUSTRIAL LOADING Design conditions for this plant included a total BOD load in the raw wastes of 4,400 kg (9,700 lb)/day. The actual BOD load to the plant in recent months has been about 6,350 kg (14,000 lb)/day, 45% over design conditions. About 99% of the BOD is from the indus- trial sources. Increases in production and revised operating proce- dures at the paper mills were reported as the causes of the increased loads. These high loads have undoubtedly contributed to the inability of the plant to meet the permit limitations, but the extent of this contribution cannot be determined from the data available. The heav- ier loads may necessitate a higher regeneration frequency than was originally anticipated but, again, data are not available to support this. Until the regeneration frequency necessary to meet the permit limits is determined, the ability of the existing regeneration system to meet the regeneration demands cannot be ascertained. ABILITY TO MEET PERMIT LIMITATIONS When the carbon columns were on stream, they were not effective in reducing the BOD to the degree necessary to meet the limitation. The wastewater leaving the coagulation-sedimentation system had a BOD concentration of about 60 mg/1, of which the carbon columns removed only 50%, leaving a final effluent BOD concentration of 30 mg/1. The carbon column effluent has seldom, if ever, met the permit limitation of 8 mg/1. These data, plus carbon adsorption studies of Fitchburg wastewater conducted in NEIC laboratories, indicate that the wastewater contains a high percentage of non-adsorbable organic materials. An evaluation of the raw materials used by one of the paper mills indicates that many of the materials used are relatively low in molecular weight and would be biodegradable but not adsorbable. ------- 7 The data available make it apparent that, with the high loads of non-adsorbable materials being contributed to the West Plant by the paper mills, coagulation, sedimentation and carbon adsorption will not reduce the BOD to 8 mg/1. It will be necessary to provide biolo- gical treatment to degrade the low-molecular-weight materials prior to, or in conjunction with, carbon adsorption. PILOT PLANT DATA - POMONA, CALIFORNIA The physical-chemical treatment of wastes at Pomona, California was evaluated because of the many simi1iarities between the design characteristics of the Pomona and Fitchburg Plants. The Pomona Plant treats municipal rather than industrial wastes, but both provide chemi- cal coagulation and sedimentation followed by adsorption on granular carbon. Because the Pomona Plant, like Fitchburg, experienced sulfide generation problems, it conducted an extensive study of procedures to solve this problem. Most successful was the addition of the nitrate ion to the carbon column influent. Nitrate addition not only reduced sulfide concentration of the column effluent to less than the detec- table level, but also promoted biological growth on the carbon granules. These organisms degraded the organic materials that were concentrated on the carbon and, in effect, partially regenerated the carbon and decreased the frequency of regeneration necessary to maintain effluent quality. Both the elimination of the sulfide problem and the bio- degradation of organic materials provided by the addition of the nitrate ion would improve the quality of the effluent at Fitchburg. The nitrate ion is reduced to nitrogen gas so nitrogen is not added to the effluent in any form. The addition of the nitrate ion does increase the pressure drop through the carbon columns, but at Pomona this was satisfactorily handled by proper backwashing techniques. The effect of the increased ------- 8 pressure drop resulting from nitrate addition cannot be determined for Fitchburg; but the procedure has enough potential benefits that it should be evaluated. ------- III. EVALUATION OF PLANT DESIGN AND OPERATION SELECTION OF PHYSICAL-CHEMICAL TREATMENT After an initial evaluation of the characteristics of the waste- water at Fitchburg, the City consultants (CDM) recommended physical- chemical treatment (coagulation, sedimentation, carbon adsorption) rather than conventional biological treatment. CDM indicated that biological treatment: a) could not operate effectively with the wide variability in flows and concentrations experienced at Fitchburg (the paper mills shut down on weekends); b) would not remove the color bodies present in the raw wastes; and c) would require the addition of nutrients to support biological growth. To determine if adsorption on activated carbon was suitable for this application, CDM conducted pilot-scale adsorption studies. Waste- water collected from several of the contributing mills was composited in a tank trailer to simulate the feed to the municipal plant. The feed to the pilot-scale columns was supernatant from the tank trailer. During the six weeks of pilot plant operation [Table 1], the settled waste BOD ranged from 1 to 34 mg/1 (avg = 14 mg/1) and the carbon column effluent BOD (after 23 minutes' empty bed contact time) ranged from 0 to 6.3 mg/1 (avg = 2 mg/1). On the basis of the pilot plant results, the West Plant was de- signed to provide chemical coagulation and sedimentation followed by 34 minutes' contact (empty bed basis) in downflow granular activated carbon columns. ------- 10 Table 1 PILOT PLANT DATA3 WEST TREATMENT PLANT FITCHBURG, MASSACHUSETTS Date TSS (mg/1) BOD (mq/1) 1970 Carbon Feed Carbon Effluent Carbon Feed Carbon Effluent MAY 12 1.5 - 26.2 6.3 13 18.5 12.7 22.5 5.1 14 29 17.3 18.2 4.9 18 16 - 10.0 5.3 19 2.0 4.0 13.5 5.0 20 18 4 6.2 3.9 21 8 0 12.0 4.8 22 23 - - - 26 2 0 1.0 0.03 28 18 7.2 4.2 0.3 JUNE 1 12.4 1.4 8.1 0.4 2 147 3.8 - - 3 - - 34.3 7 4 10.8 0 4.8 0.9 5 15.0 5.4 10.0 0.5 8 10.8 0.2 4.8 1.0 9 15 0 5.6 0.2 10 25.6 - 27.6 1.4 11 59.0 0.2 30.2 0 12 12.4 0 17.4 1.0 15 49.2 0 15.3 0.8 16 34.0 0 15.9 0 17 52.8 0 12.3 0.7 18 4.4 0 11.0 0 19 10.4 0 10.0 0.75 22 14.8 2.0 10.3 0.8 23 - 0 14.2 0.6 24 21.6 0 15.0 1.7 25 1.6 0 17.5 2.6 26 9.6 0 4.5 0 29 9.6 0 4.8 0 30 45 0.4 21.3 1.3 a Table taken from "Supplement C, Proposed Process Revisions, West Fitchburg Wastewater Treatment Facility, August, 1970." Prepared by Camp, Dresser and McKee. ------- 11 PLANT DESIGN The West Plant receives municipal and industrial wastewaters in separate conduits [Figure 1]. The sanitary wastewaters are settled in two 9.1 m (30 ft) diameter clarifiers operating in parallel. At the 1990 design flow of 0.08 m3/sec (1.8 mgd), the overflow rate would be 30 1/min/m2 (1,060 gpd/ft2). This primary effluent is then chlori- nated and joins the industrial wastewater. This combined flow of 0.67 m3/sec (15.3 mgd) is then dosed with alum and sent through two rapid-mix tanks in parallel (6-minute residence time). Polymer is then added, and the water flows to two 2-stage flocculation basins in parallel (40-minute total detention time). After flocculation, the wastewater is settled in two 40 m (130 ft) diameter clarifiers. At 0.62 m3/sec (15.3 mgd), the clarifier overflow rate is 16.4 1/min/m2 (580 gpd/ft2). The chemical clarifi- cation system is well designed and is capable of providing the quality of effluent necessary for carbon adsorption. Following clarification, the wastewater is pumped to the carbon adsorption system which consists of twelve carbon columns containing beds of granular (8 x 30 mesh) activated carbon. The carbon columns are 6 m (20 ft) in diameter and have an overall height of 10 m (33 ft). Each column contains a carbon depth of 4.7 m (15.5 ft) (57,000 kg-126,000 lb). The carbon is supported on a perforated stainless steel plate underlain by 30 cm (12 in) of graded gravel. The gravel is supported by a Sybron-Leopold* glazed clay filter underdrain system. A 5.8 m (19 ft) diameter Sybron-Leopold rotary filter agitator (surface wash mechanism) is installed 5 cm (2 in) above the top of the carbon. This is a horizontal pipe rotating on a vertical axis with spray nozzles * Mention of commercial products does not imply endorsement by the Environmental Protection Agency. ------- Domestic Wastes Filter Backwash Polymer A1 urn Industrial Wastes to Process Flow Diagram - Physical-Chemical Treatment Plant - Fitchburg, Massachusetts Figure Lagoon Chlorination Flocculation Carbon Regeneration Clarification Primary Treatment Post Aeration Rapid Mix Carbon Adsorption hO ------- 13 directed at the carbon surface. Its function is to agitate the bed during backwash and to separate the carbon granules from the waste- water solids filtered out and from the biological solids generated. Design conditions called for ten columns to be on stream (in parallel) at any one time while one is being regenerated and one is being backwashed. When the capacity of the carbon in one column is exhausted and regeneration is necessary, the carbon is moved hydrau- lically, using water-driven eductors, from the carbon column to the spent carbon storage tank. The carbon is then mechanically pumped to a screw conveyor feeding a 3.3 m (10.75 ft) diameter six-hearth furnace for regeneration. The carbon is moved down from hearth to hearth, reaching a maximum temperature of about 870°C (1,600°F). Following this step, the carbon flows by gravity, without contact with air, to a quench tank and is then mechanically pumped to a regenerated carbon storage tank. From here, the carbon is educted hydraulically back to a column. PLANT OPERATION Data available for the West Treatment Plant indicate that the carbon columns have treated less than 10% of the clarified wastewater since startup. From November 16, 1976 to May 31, 1978 (the first and last dates for which information is obtainable), data are available for only 429 column days.* If only 6 of the 12 columns had been opera- ting continuously during this time, there would have been 3,360 column days of operation. For treating 0.67 m3/sec (15.3 mgd) (the design flow) at the design column flow rate, ten columns would have to be operating at the same time. * One column day = one column in operation for one day. ------- 14 Operating Problems Continuous operation of the carbon columns has been prevented because of the many design and operating problems. These problems are itemized below and discussed in the text following: 1. The pumps installed to remove sludge from the clarifiers could not handle the load being produced; thus excess sludge built up in the clarifiers. 2. Piping system breaks in the carbon column feed system oc- curred and required major revisions. 3. The pumps that move the activated carbon to and from the regeneration furnace have been a high maintenance problem because of excessive wear of the impeller linings. 4. The lining applied to the contactors has failed and severe corrosion of the base metal has occurred and is still occurring. 5. The surface wash mechanisms on all carbon columns have been broken and are no longer effective. Improper backwashing procedures have been reported as the cause. 6. The automatic rate controllers designed to balance the flow among the active carbon columns do not work properly. 7. The growth of sulfate-reducing bacteria on the carbon gran- ules has resulted in hydrogen sulfide concentrations up to 5 mg/1 in the column effluent, causing serious local odor problems. ------- 15 Clarifiers Since the clarifier sludge pumps were replaced with pumps that can handle the sludge, the clarifiers have operated satisfactorily. Between March and September 1977, the TSS contentration of the clari- fier overflow ranged from 20 to 38 mg/1 and averaged 32 mg/1. This resulted in a carbon column effluent that, with few exceptions, exceeded the 8 mg/1 TSS limitation [Table 2]. Improved operation of the clarifier to reduce the TSS concentration of the feed to the carbon column will be necessary if the carbon column effluent limitation of 8 mg/1 TSS is to obtained. At the same time, the BOD of the clarifier overflow ranged from 4 to 71 mg/1 and averaged 61 mg/1. A reduction in the TSS of this stream should result in a BOD reduction as well. Carbon Treatment System Piping system breaks caused start-up delays, but these problems have been solved. The pumps that move the carbon to and from the regeneration furnace, however, remain a high maintenance problem. It is necessary to replace the impellers frequently. Replacement pumps have been suggested by the contractor but have not yet been installed. Several studies have been made to determine the cause of the column lining failure. The study reports have not been made available and, to date, no action has been taken to repair or replace the lining. Severe corrosion has occurred and is still occurring where the base metal has been exposed. For successful long-term operation, it will be necessary to patch the linings or to replace them completely. ------- 16 Because of improper backwashing procedures, the surface wash mechanisms were torn loose from their mountings during operation. None of the mechanisms have yet been replaced although replacement is necessary for successful plant operation. Automatic rate controllers were installed in the influent lines to balance the flows to each column. These have not worked satisfac- torily, and the flow rate to the operating columns is now controlled manually while the automatic system is being redesigned. Generation of Hydrogen Sulfide In metabolizing biodegradable organic material, bacteria will use oxygen from, in the order of preference, dissolved oxygen, the nitrate ion and the sulfate ion. The feed to the carbon columns is high in BOD (avg = 61 mg/1) and has little, if any, dissolved oxygen or nitrate ion. The organisms, therefore, use oxygen from the sulfate ion, thus producing hydrogen sulfide (H2S). The odorous hydrogen sulfide is released into the atmosphere when the effluent is dis- charged from the columns; H2S concentrations as high as 5 mg/1 have been measured in the column effluent. The odor is so strong and per- vasive that it is necessary to shut the columns down when this problem occurs. The City officials have elected to solve the odor problem by shutting down a column when the H2S odor from that column becomes objectionable and adding caustic to the column feed until the effluent reaches pH 10.5 to 11. The carbon is soaked in this high pH water for 24 to 36 hours. During this period, the sulfate-reducing bacteria are killed and the column can be started up again free of H2S production. This procedures controls sulfide generation from 4 to 8 weeks in the winter. In the summer, however, the sulfate-reducing bacteria grow so rapidly that it is necessary to repeat the procedure every week. ------- 17 Adsorption-Resistant Materials in Column Feed During the time the carbon columns were on stream, the 8 mg/1 BOD limitation was not met. The limited BOD data available [Table 2] indicate that the columns removed only about 50% of the influent BOD, producing an effluent with about 30 mg/1 BOD. The influent, then, must contain a high concentration of organic materials that are not adsorbable on activated carbon. A sample of the carbon column influent was collected during the plant inspection and carbon adsorption studies were conducted at the NEIC laboratory. The results show that the wastewater contains a high concentration of materials that are resistant to adsorption on carbon. In the NEIC test, 500 mg/1 of activated carbon lowered the total organic carbon (TOC) content of the column feed from 27 to 13 mg/1 (52%). The addition of 1,000 mg/1 of activated carbon removed only 3 mg/1 TOC more. The low additional TOC removal obtained with double the carbon dosage is due to the high percentage of non-adsorb- able organic compounds present. The NEIC study results [Figure 2] indicate that carbon adsorption alone will not remove all the organic material. It is well known that activated carbon is not effective in ad- sorbing low-molecular-weight materials such as alcohol and formalde- hyde. An evaluation of the data supplied by one of the paper mills2'3 shows that their wastewaters contain both of these materials, and several other organic materials that would not be adsorbable on acti- vated carbon. Because these non-adsorbable organic materials have low molecular weights, it is likely that they are readily biodegrad- able. Therefore, to meet the BOD limitation required, it will be necessary to combine both carbon adsorption and biological treatment. ------- 18 Table 2 SUMMARY OF CARBON COLUMN EFFLUENT DATA WEST TREATMENT PLANT FITCHBURG, MASSACHUSETTS Column Number of BOD (mg/1) TSS (mq/1) No. Data Points Range Average Range Average 1 3 23-34 29 7-13 9 5 9 12-44 30 6-20 9 6 7 12-32 22 2-15 6 7 14 19-64 35 2.5-43 15 8 8 26-29 28 3-10 6 9 8 14-49 28 1-34 19 10 7 16-49 34 0.5-44 18 12 9 18-40 27 2.4-19 10 ------- - - - - - - _ - - - - - - ¦ - - - - - - - - v - k - ¦ " - — - - - ¦ - 1 ! 1 ... - - - - - - re vat cTT&Tj PI 29 Car jOQ-'Te >js- ;$adhusetts 2$ 9 ant 0 TOO 200 300 400 500 600 Activated Carbon Dosage, mg/1 700 800 "900~ T000 ------- 20 Carbon Regeneration Between June 1977 and June 1978, the carbon from seven of the West Plant columns was regenerated. The data provided indicate that the regeneration restored the carbon to near virgin conditions. Re- peated adsorption and regeneration cycles are necessary, however, to determine the effectiveness of the regeneration procedures. No data are available on the amount of carbon lost during the carbon transfer and regeneration processes. However, carbon has been transferred and regenerated in similar processes with losses ranging between 5 and 10%, and it should be possible to limit losses to these levels at Fitchburg. Effect of Industrial Load on Treatment Capability During the CDM pilot-scale studies, the BOD content of the input to the pilot carbon columns ranged from 1 to 34 mg/1 [Table 1] and the carbon treatment successfully reduced this to less than 8 mg/1. The average BOD input to the carbon columns under present operating conditions is 61 mg/1. The industries have increased their load several- fold over the design conditions [Table 3]. Although the City has requested that the industries reduce the load to the plant, a signifi- cant load reduction does not seem probable. The change in quantity, and probably in nature, of this load between the time of the pilot- plant study and the present time has undoubtedly affected the ability of the plant to meet the permit limitations. Further plant-scale studies combining biological oxidation and carbon adsorption will be necessary to determine if the plant has the capability to meet the limitations under the increased industrial load. ------- Table 3 ALLOWABLE AND ACTUAL LOADS FROM PAPER MILLS WEST TREATMENT PLANT FITCHBURG, MASSACHUSETTS Flow BOD TSS Plant Allowed Actual Allowed Actual Allowed Actual Name m3/sec mgd m3/sec mgd kg/day lb/day kg/day lb/day kg/day lb/day kg/day lb/day James River, Massachusetts 0.23 5. 3 0.15 3.5 993 2,190 2,650 5,840 8,527 18,800 6,577 14,500 Fi tchburg Paper Co. 0.21 4.8 0.15 3.5 1,770 3,900 2,780 6,130 9,980 22,000 13,150 29,000 Crocker Technical Papers 0.11 2.6 0.05 1.1 363 800 250 550 770 1 ,700 624 1 ,375 James River, Fitchburg 0.05 1.1 0. 06 1.3 458 1,010 615 1,355 1,590 3,500 787 1 ,735 Sanitary 0.07 1.5 0.01 0.15 816 1 ,800 68 150 910 2,000 56 125 Total 0.67 15.3 0.42 9.55 4,400 9,700 6,300 14,025 21,770 48,000 21,200 46,735 ------- IV. PHYSICAL-CHEMICAL TREATMENT AT POMONA, CALIFORNIA Between 1973 and 1975, the Environmental Protection Agency spon- sored a 27-month study of physical-chemical treatment at Pomona, Cal- ifornia. The feed to the system at Pomona was principally municipal wastes, as opposed to the 90+% industrial wastes in the feed to the Fitchburg plant. Nevertheless, the designs of the two systems are similar [Table 4] and the problem of hydrogen sulfide generation being experienced at Fitchburg, was encountered and solved at the Pomona Plant. For this reason, the results of the Pomona study4 were eval- uated and are discussed below. Table 5 shows the results of the Pomona study. An overall per- centage removal for BOD cannot be determined because the BOD of the raw wastewater was not given, but the carbon system alone reduced the average BOD from 36.2 to 7.8 mg/1, an average removal of 78.5%. It is probable that the overall BOD removal (clarification and adsorp- tion) is greater than 90%. GENERATION OF HYDROGEN SULFIDE The Pomona plant experienced difficulties with hydrogen sulfide generation in the carbon column, and total sulfides in the effluent exceeded 5 mg/1 at times, a situation almost identical to the problems experienced at Fitchburg. Backwashing alone was not adequate to solve the problem, so the addition of pure oxygen to the feed was evaluated. The sulfide content of the effluent decreased slightly as a result of oxgen addition, but this procedure was abandoned because of excessively high pressure drops which occurred in the column as a result of the increased biological activity. ------- 23 Table 4 COMPARISON OF PHYSICAL-CHEMICAL TREATMENT PLANT DESIGNS AT POMONA AND FITCHBURG Parameter Fitchburg Pomona Contact Time in Rapid Mix Tank, min 6 2.8 Contact Time in Flocculation Basin, min 40 42.5 Clarifier Overflow Rate, 1/min/m2 (gpd/ft2) 16 (580) 33 (1,180) TSS in Clarifier Overflow, mg/1 <35 28 BOD in Clarifier Overflow, mg/1 61 36 Column Feed Rate, 1/sec/m2 (gpm/ft2) 2.1 (3.2) 2.7 (4) Carbon Contact Time, Empty Bed, min. 34 30 BOD in Column Effluent, mg/1 30 7.8 Table 5 SUMMARY OF PERFORMANCE OF PHYSICAL-CHEMICAL TREATMENT PLANT AT POMONA, CALIFORNIA Parameter Raw Clarified Carbon Overall Sewage Effluent Effluent Removal (%) TSS, mg/1 199 28.3 6.7 96.6 Total COD, mg/1 321 95.8 19.3 94.0 Dissolved COD, mg/1 49.0 48.6 13.5 72.7 BOD, mg/1 - 36.2 7.8 78.5a a Does not include BOD removal by clarification. ------- 24 Continuous chlorine addition, at doses ranging from 20 to 50 mg/1, was then evaluated. It was hoped that the chlorine would kill the sulfate-reducing bacteria. Because chlorine is rapidly removed by activated carbon, this procedure was not effective and the sulfide concentration of the column effluent reached 2.6 mg/1. The addition of sodium nitrate to the column feed was then evalua- ted. At a dosage of 5.4 mg/1 of sodium nitrate (as nitrogen), this procedure was completely successful and the sulfide levels were reduced consistently to below detectable levels. PRESSURE DROP AND BIOLOGICAL ACTIVITY AS A RESULT OF NITRATE ADDITION The addition of sodium nitrate at Pomona was accompanied by an increase in the rate of pressure drop build-up during operation. On several occasions, pressure losses exceeding 3.5 kg/cm2 (50 psi) were experienced before the daily backwash. By proper backwashing proce- dures, however, the backwashing frequency was kept at about once-per-day. About 6-8% of the product water was used to backwash the column. Offsetting the headloss problems caused by the nitrate addition was the significant increase in carbon capacity, also apparently result- ing from the addition of the nitrate ion. The column, containing 2360 kg (5,200 lb) of 8 x 30 mesh activated carbon, treated 113,500 m3 (30 million gal) of wastewater during the first cycle (before regen- eration). The first cycle was ended so that experience could be gained in regenerating the carbon. The carbon was not exhausted and appeared to have leveled out at about 70% dissolved COD removal, and regeneration was not necessary to maintain product quality. It was theorized that the organisms generated by the nitrate addition metabolized some of the organic material from the feed as it proceeded down the column and, at the same time, partially regenerated the carbon by attacking the organic materials concentrated on it. ------- 25 Table 6 compares experience at Pomona with data at other physi- cal-chemical treatment plants where the nitrate ion had not been added to the columns. The carbon capacity at Pomona was over five times as high as the average obtained at the other plants. The economic bene- fits of nitrate addition at the West Treatment Plant would have to be balanced against the cost of the nitrate added and the effect of the increased pressure drop. APPLICABILITY OF POMONA FINDINGS TO PROBLEMS AT FITCHBURG The findings at Pomona appear relevant to the problems being experienced at the Fitchburg West Plant. Adding the nitration ion to the carbon column feed should reduce, or possibly eliminate, the pro- duction of hydrogen sulfide. In addition, the organisms whose growth will be promoted by the addition of the nitrate ion will metabolize some of the low-molecular-weight materials that are resisting adsorp- tion on carbon. This would result in reduced BOD in the effluent from the carbon columns, but the degree of reduction is not predict- able. The rate of pressure drop increase at the West Plant resulting from the addition of nitrate ion to the carbon column influent cannot be predicted. Every effort should be made to backwash the columns properly so that both sulfide reduction (or elimination) and biological degradation can be achieved at the plant. ------- 26 Table 6 CARBON CAPACITY AT PHYSICAL-CHEMICAL TREATMENT PLANTS4 Plant Total COD, mg/1 Influent Effluent Carbon Capacity Total COD kg total COD Removed Removal % kg carbon Lebanon, OH 67 Washington, D.C 55 27 15 59.7 72.7 0.5 0.7 Ewi nq-Lawrence, N.J. 75 Pomona, CA 98.4 20 23.6 73.3 76 0.8 3.5 ------- 27 REFERENCES 1. Callahan, W.F. and A.B. Pincence, Physical-Chemical Treatment System at Fitchburg, Massachusetts. TAPPI Environmental Conference Proceedings. April 1977. 2. Memo of September 8, 1978 from Frank J. Early (EPA, NEIC) to John F. Hackler (EPA, Region I), titled Carbon Adsorption Resistance of Chemicals Found in Untreated Effluent from Fitchburg Paper Company, Fitchburg, Massachusetts. 3. Independent Physical-Chemical Treatment of Raw Sewage. EPA-600/2-77-137, August 1977. Environmental Protection Technology Series. 4. Letter of July 10, 1978 from F.E. Pendleton, Jr., of Fitchburg Paper to John F. Hackler, EPA Region 1. ------- RECOMMENDATIONS FOR IMPROVING EFFLUENT QUALITY AT THE FITCHBURG WEST TREATMENT PLANT 1. Addition of the nitrate ion to the feed to a full-scale carbon con- tactor should be evaluated. To prepare for this evaluation, a surface wash mechanism should be installed on one of the carbon columns. The carbon should then be filled with virgin or freshly- regenerated activated carbon and backwashed as recommended by the supplier of the surface wash mechanism. 2. Clarified wastewater, to which 5 mg/1 of sodium nitrate (as nitro- gen) has been added, should be passed downflow through the column at the design rate. The nitrate ion should be added far enough upstream to assure adequate mixing. If hydrogen sulfide is detect- able in the effluent under these conditions, the nitrate addition rate should be increased. If H2S is not detectable, the nitrate ion addition rate may be decreased. 3. Backwashing should be initiated when required by pressure drop through the'column. The backwashing procedure recommended by the supplier of the surface wash mechanism should be followed closely. Samples of the backwash water should be taken frequently and back- washing stopped when a predetermined turbidity of the backwash water is reached. Shutdown of the backwashing procedure should also be done as recommended by the supplier of the surface wash mechanism. 4. During this evaluation of nitrate addition, complete records of influent BOD and TSS, influent flow, and column pressure drop should be maintained. 5. The caustic wash procedures should not be used during this evaluation. ------- 6. If this treatment procedure does not reduce the BOD to the desired level or if the rate of pressure drop increase prevents satisfac- tory operation, one of the following procedures must be followed to enable the plant to meet the prescribed limitations: a. Reduce the amount of non-adsorbable organic materials enter- ing the plant by process revisions or pretreatment at the paper mills. b. Provide a separate biological system at the treatment plant to remove non-adsorbable materials prior to carbon adsorption. ------- |