PB-235 567
A TEST METHOD FOR VOLATILE COMPONENT
STRIPPING OF WASTE WATER
Louis J. Thibodeaux
Arkansas University
Prepared for:
Pacific Northwest Environmental Research
Laboratory
Office of Water Resources Research
May 1974
DISTRIBUTED BY:
NaiiGttJ Technical tnfonnaiisa Service
U. S. DEPARTMENT OF COMMERCE
5285 Port Royal Road, Springfield Va. 22151
-------
SI-LI rir.i) w.vn R
RILSOURl'l-S ABSTRACTS
INPUT TR\NSACTION FORM
PB 235 567
W
A TEST METHOD FOR VOLATILE COMPONENT STRIPPING
OF WASTE WATER,
Thlbodeaux. Letrfs J.
ege _o f J ng i n e.eMn cQ; .
">
_
Fayetieville, AR
^ifc tfY
R-801876
Environmental Protection Agency^repcrt number EPA-660/2-74-044, May 197!*
-This-work-1s conce.rned-*a4A-the airrstripoable
Industrial 'wastewaters. The primary purpose was to develop laboratory apparatus
and procedures that may be employed to assess the desirability of air-stripping in
cooling towers as a treatment operation for removal of a portion of the orgam'cs
from industrial wastewater. The apparatus developed consists of a short packed
(Intalox Saddle)' section with liquid recirculation and single pass countercurrent air
flod. Desorption is performed in the apparatus at 25?C and ambient pressure conditions.
Desorption experiments were performed on single pure components in water,
simulated wastewater preparations and actual industrial wastewater samples. Industrial
wastewater samples were representative of: poultry, metal, oil-field, canning,
pharmaceutical, paper, food, fibers, petroleum refinery and petrochemical industries.
BOD, COD, TOC and gas chromatographic analysis were employed wittrtne experiments-.
Industrial wastewaters were found to contain a non-volatrle organic fraction that
remains in the aqueous phase and a volatile organic fraction that can be transferred
to the air phase. Results of one hour desportion runs indicate that the range of
the volatile organic fraction is 0% to 70% TOC. The ultimate volatile range was
calculated to be 0% to 98% TOC. The volatile organic fractions displayed a range
of relative volatilization rates 4.4 to 41.6 times greater than water. The net result
of the desorption experiments is that some industrial wastewaters can be effectively
treated bv air-stripping a sizeable portion of the dissolved orgam'cs. ( G " " "*
j. ^Waste Water Treatment^, Cooling Tower/, Testing Procedures*, Volatility*.
Industrial W? ces«7 Air Pollution, Mass Transfer, Biochemical Oxygen Demand,
Chemical Oxy)*..i Demand, Gas Chromatograpny, Laboratory Tests, Self-Purification,
Safety.
Air Stripping*, Desorption*', laduatrJel-Haoto U»io» TroatnentA^ Relative
Volatilization Rate, Total Organic Carbon, Intalox Saddle, Volatile Organics,
acetone, propanol, methanol, butanol, furfural, phenol. University of Arkansas.
'- '- *:- .:: ,'M I./
NATIONAL TECHNICAL
^ ^~ . INFORMATION SERVfCE
_ Snrim;<-tld VA mi!
.. . -i.i.v frf,m rrn
VilililUtC' LI UIU O f \J
,. ....
U[3 UUC
I
microfiche from NT13 ad PB
i If O
Send To:
laoriK Moowaca acicimpie IHFOHMATUJW CCNTCH
IXS. OOMMITMCMT or THE IMTCMIO*
MJMNOTON. -JX, 16149
Louis J. Thibodpaux
University of Arkansas
U.S. GOVERNMENT PRINTING OFFICE: 1*7* in411:41
-------
KPA-660/2-74-044
May 1974
A TEST METHOD FOR VOLATILE COMPONENT
STRIPPING OF WASTE WATER
By
Louis J. Thibodeaux
College Of Engineering
University Of Arkansas
Fayetteville, Arkansas 72701
Project No. R-801876
Program Element 1BBO37
Project Officer
John S. Ruppersberger
Pacific Northwest Environmental Research Laboratory
National Environrrental Research Center
Corvallis, Oregon 97330
Prepared for
OFFICt OF RESEARCH AND MONITORING
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON.D.C. 20460
t
M
For »»!» by the iiuprrinirnilmt o' l>oc(im"iit.«. t'-*. \r
-------
ABSTRACT
This work is concerned with the air strippable volatile organic fraction of industrial
wastewiters. The primary purpose was to develop laboratory apparatus and procedures that
may be employed to assess the desirability of air stripping in cooling towers as a treatment
operation for removal of a portion of the organics from industrial wastewater. The
apparatus developed consists of a short packed (Intalox Sadd'.i) section with liquid
recirculation and single pass countercurrent air flow. Desorption is performed in the
apparatus at 25°C and ambient pressure conditions.
Desorption experiments were performed on single pure components in water,
simulated wastewater preparations and actual industrial wastewater samples. Industrial
wastewater samples were representative of: poultry, metal, oil-field, canning,
pharmaceutical, paper, food, fibers, petroleum refinery and petrochemical industries. BOO,
COD. TOC and gas chromatographic analysis were employed with the experiments.
Industrial wastewaters were found to contain a non-volatile organic fraction that
remains in the aqueous phase and a volatile organic fraction that can be transferred to the
air phase. Results of one hour desorption runs indicate that the range of the volatile organic
fraction is 0% to 70% TOC. The ultimate volatile range was calculated to be 0% to 98%
TOC. The volatile organic fractions displayed a range of relative volatilization rates 4.4 to
41.6 times greater than water. The net result of the desorption experiments is that some
industrial wastewaters can be effectively treated by air-stripping a sizeable portion of the
dissolved o.-ganics.
This report was submitted in fulfillment of Project Number R-801876 by the
University of Arkansas under the partial sponsorship of the Environmental Protection
Agency. Work was completed as of April, 1974.
-ii-0/
-------
CONTENTS
Page
Abstract ii
List of F igures iv
List of Tables vi
Acknowledgments vii
Sections
I Conclusions 1
II Recommendations 2
III Introduction 3
IV Items of Project Study 9
V Theory of Volatile Component Desorption From An Aqueous Phase 11
VI Experiment Fabrication, Materials and Methods 22
VII Exoerimental Phase and Results 33
VIII Discussion 63
IX References 67
X Publications and Patents 69
XI Glossary of Symbols 70
XII Appendices 73
in
-------
FIGURES
No. Page
1 Three Major Fractions ol Industrial Wastewater 4
2 Schematic of Volatile Desorption Apparatus 14
3 Hypothetical Packed Column Dcsorption Experiment Results 20
4 Raschig Ring Desorption Apparatus 23
5 Intalox Saddle Desorption Apparatus 26
6 Desorption of Methanol in Water with Raschig Ring Column - COO Analysis .... 35
7 Desorption of Metiunol in Water with Raschig Ring Column - GC Analysis 36
8 Desorption of Acetone in Water with Raschig Ring Column 3ODR Analysis .... 37
9 Desorption of Phenol in Water with Raschig Ring Column - FOC Analysis 38
10 Desorption of Simulated Wjstewater No. 1 with Raschig Ring Column 41
11 Desorption of Simulated Wastewater No. 2 with Ra.chi'j Ring Column 42
12 Desorption of Simulated VVastewater No. 3 with Raschig Ring Column 43
13 Desorption of n-Butanol with Intalox Saddle Column Repetitive Runs 44
14 Desorption of Sucrose Solution with Intalox Saddle Colun.n 45
15 Desorption of Poultry Processing Wastewater with intalcx Saddle Column 47
16 Desorption of n Butanol with Intalox Saddle Column Interference Tests 48
17 Desorption of Simulated Wastewater No. 4 with Intalox Saddle Column
GC Analysis 49
18 Water Evaporation Calculations in the Raschig Ring Column 51
19 n-Butanol Desorption Calculations with Intalox Saddle Device
Effect of Liquid Rate 52
20 n-Butanol Desorption Calculations with Intalox Saddle Device
Effect of Air Rate 53
21 Desorption Results for Industrial Wastewater Samples 1 58
22 Desorption Results for Industrial Wastewater Samples - ii 59
23 Desorption Results for Industrial VVastewater Samples III bu
iv
-------
No. Page
24 Relative Volatilization Rates of Industrial Wastewaters 61
25 Petroleum Refinery Wastewater - TOC and GC Experimental Results 62
C-1 The Dow Hollow Fiber Beaker Gas Permeator b/HFG-1 127
C2 Schematic Diagram of Equipment Arrangement for Gas Permeator
Oesorption Experiment 128
-------
TABLES
No. Page
1 Vapor-Liquid Equilibria of Selected Gases and L iquids in Water @ 25°C 12
2 Reldtive Volatilisation Rates and Relative Volatilities > 25°C -
Calculated Results for Eleven Pure Components 16
3 Typical Raw Data From a Desorption Experiment 29
4 Experimental Relative Volatilisation Rates for Selected Pure Components
with Rasctiig Ring Desorption Apparatus 34
5 Raschig Ring Experiments Linear Regression Statirtical Data for
Pure Components 39
6 Air Volatile Fractions in Industrial Wastewaters 56
7 Published Air Volatile Fraction Data 57
C-1 Test Results - Volatile Component Desorption in Gas Permeator 129
C-2 Relative Desorption Rates in 1-Hour Run Time 129
vi
-------
ACKNOWLEDGMENTS
The design, construction, and operation of the laboratory units were performed by a
team headed by Mr. Jack R. Jones and consisting of Messrs. Charles Cheng, Rick Jones and
Mike Gray all of the Department of Chemical Engineering, University of Arkansas. This
report was prepared with the assistance of Mrs. Susan Gray.
I wish to acknowledge Drs. J. R. Couper and L. R. Heiple of the College of Engineering
and Drs. R. E. Babcock and A. E. Harvey of the Graduate School for their support and
encouragement in this project.
The coordination and support provided °y Mr. John Ruppersbeiger (Project Officer)
and Mr. Ralph Scott, Environmental Protection Agency, Corvallis, Oregon was greatly
appreciated.
VII
-------
SECTION I
CONCLUSIONS
Based on the theoretical and experimental results of this study, it is concluded that:
1. A countercurrent air flow, recirculating liquid, packed stripping column has been
developed to perform desorption studies on wastewater samples.
2. The Intalox Saddle Column described on pp.24 and the test method presented on
pages 25 and 27 are recommended.
3. This laboratory device can be employed as a laboratory simulation of the air-stripping
operation for removal of volatile organics in cooling towers and similar units.
4. Th's desorption device can be employed as a laboratory test for studies of the volatile
character of industrial wastewater.
5. A single desorption study on this device ;s capable of quantifying the fraction
maximum degree of treatment (Fv) of air-volatile organics in a wastewater sample and
the relative rate (K/a) at which the organic fraction can be removed from the water.
6. Nineteen samples of industrial wastewater were desorbed and 'ound to contain an
average of Fv = 33% TOC volatile fraction with a range of Fv = [0%, 98%).The
volatiles in these fractions were 4.4 to 41.6 times more volatile than water.
-1-
-------
SECTION II
RECOMMENDATIONS
1. A comprehensive project should be initiated and the desorption apparatus employed to
study the volatile character (i.e.. Fv and K/a) of industrial wastewater. Only nineteen
samples were desorbed in this project but the volatile fraction seems to be sizable.
2. The desorbed species should be identified and quantified individually to assess the
impact of these components upon the environment.
-2-
-------
SECTION III
INTRODUCTION
i' .-. project is concerned with the development of a laboratory apparatus and
experimental procedures for the measurement of the fraction of the dissolved constituents
in a wosiewater that can be removed by desorbing into air. This research is not concerned
specifically with the air stripping of ammonia from wastewater. Many industrial wastewaters
contain dissolved volatile organics which are easily removed by contact with air. and it is
this type of wastewater constituent which this project proposes to study. This project is also
concerned with the measurement of the relative ease at whidi these dissolved, volatile
constituents can be transferred :o the air phase. The rate of desorption is compared to the
rate of evaporation of water. The rate of volatilization of a component must be greater than
the evaporation rate of water to effect its removal from water.
All aqueous wastes are immediately acted upon by two ambient pnysical, natural
phenomena which are capable of producing significant changes in the character of the
wastewater. These two phenomena are the force of gravity and the cl.emical potential
between phases. The Earth's gravitational force acts upon the differences in density between
suspended (solid) particles and water causing somr fractions to float to the surface and
others to settle to the bottom. The gravitational phenomena effect on wastewater is well
known to chemists and engineers. The interphase mass transfer of chemical species caused
by a chemical potential between the water and the (ever present) air phase is a more subtle
natural phenomena that is invisible and not easily detected by the other senses. Figure 1 is
an illustrated representation of the above idea.
The following are some applications to which volatile constituent information can be
applied.
Air-Stripping As A Treatment Operation
Quantifying the volatile fraction will place the upper bound on the effectiveness of air
stripping as a treatment operation. These operations are typically carried out in a cooling
tower type device (induced or forced draft) or in an air sparged (bubble) vessel. The major
thrust of this research is volatile organic constituents. Air-stripping has bean demonstrated
to be a feasible technique for removing a portion of the organics from wastewater. McAlister
et al (1) reported 65-85% treatment efficiencies for pulp mill condensates and 40 50% for
combined decker filtrates and pulp mill condensates in a cooling tower. These waste
streams, in combination with condenser waste from a baiometric-type evaporator
condenser, are cooled in the tower and reused. Prather (2) reported that an
oxidation aeration cooling tower process offers an economical method of removing small
-3-
-------
A
AIR VOLATILE FRACTION
INTERPHASE MASS TRANSFER
{II
Character and behavior
1 of this fraction is unknown
for the most part.
GRAVITATIONAL FORCES
J1J
REMAINING
FRACTION
V
SETTLEABLE AND
FLOATABLE FRACTION
'Contains the soluble fraction plus some low volatile and slow settling constituents. This
Is the fraction that typically receives secondary treatment.
Character and behavior of
these fractions is '/veil
established.
Figure 1. The three major Tractions of industrial wastewater.
-------
organic residuals by auto-oxidation and stripping of ammonia and other unwanted gases
from refinery wastewater at abort the cost of cooling the water. Mohler et al (3) reported
experience with reuse and bio-oxidation of refinery westewater in a cooling tower system.
He reported 99.9% efficiencies in removing phenolic-type compounds by bio-oxidation and
does not report on the role of air-stripping.
Volatile Losses From Cooling Towers
.Information on the volatile fraction in water will aid in the assessment of organic losses
during cooling tower operation with recycled process water or wastewater. Once-used water
or wastewater can be used as make-up water to a cooling tower. Burns and Eckenfelder !*)
report on the adapt jtion of a cooling tower to improve aeration in an activated s'udge plant.
A c^unterflow cooling tower was installed to reduce the influent waste temperature from
120" F to 100° F and..!o increase the oxygen uptake rate in adjoining aeration tanks.
Previous pilot evalnaiion of plastic trickling filters in -cries with activated sludge for
treatment of pulp and paper waste showed 25% BOD removal in 54 seconds resident times
with DQWPAC and 24% BOD removals in 38 seconds with POLYGRID. Cohn and Tonn (5)
report oV» the use of a cooling tower in black liquor evaporation. No reference to organic
removal is made but private communications revealed that similar removals as reported by
McAlister (1) were being achieved.
Desorption Of Organics During Secondary Treatment
Information on the volatile fraction will aid in the assessment of the magnitude of
organic desorption occuring during conventional secondary treatment operations. Gaudy et
al (6. 7, 8) performed diffused air-stripping experiments on volatile waste components of
petrochemical nature and reported 50% butanone removal in four hours, 50%
propionaldehyde in one hour and 50% acetone in 4.5 hours all at 25° C. Eckenfelder (9)
reports 20% removal of acetone in four hours in a similar apparatus. The variability of
sparged vessel results is due to variations in gas rate/liquid volume ratios. Hiser (10) reported
90% styrene removal ir. 20 minutes from an aerated flask. He also reported similar total
carbon removals from a composite industrial wastewater sample.
Goswami (11) reported that significant removals can occur under quiescent conditions
in a batch laboratory vessel. He observed the following removals in twelve hours from a one
liter reactor at 25° C; 17.2% removal of acetone, 18% methylethylketone, 27%
propionaldehyde. 32.8% butrylaldehyde and 35.3% removal of vale, aldehyde.
An analytical simulation model of desorption in aerated stabilization basins by
Thibodeaux and Parker (12) indicates that significant removals of selected industrial
chemicals are occurring. A study of ten common industrial chemicals in eleven full-scale
-5-
-------
aerated basins showed that 20% to 60% removal efficiencies were possible without
biochemical oxidation. Detention times ranges from 1.7 to 14.2 d.iys. Laboratory
observation of surface agitator desorbers support this data.
Morton et al (13) report the results of .-> three year study on the changes in tritium
content of a large nonsceping outdoor basin. 90% of the tritium contained in an aqueous
stream flowing at 15,000 liters per day will be lost to the atmosphere from a shallow basin
with a surface area of 6000 m2. Mackay and Wolkoff (14) derived equations to predict the
rate of evaporation of hydrocarbons and chlorinated hydrocarbons (PCBs). These
compounds have high rates of evaporation even though the vapor pressure is low.
Evaporation "half lives" of minutes and hours is due to the high activity coefficient
displayed by these components in aqueous solution.
From the above it appears that a significant amount of volatile components are being
stripped and/or desorbed in conventional secondary treatment operations involving the use
of air or the presence of large air water interfaces.
Accumulation Of Volatiles In Enclosures
Information on the volatile fractions in wastewater and tneir combustible properties
may aid in the prediction of potential explosion hazards in closed sewers, manholes, storage
vessels, treatment vessels, sumps, etc. Volatile components may accumulate in semi-air tight
enclosures and upon combination with cir an explosive mixture may result.
Information on the volatile fraction along with data on the maximum safe
concentration for breath
-------
Wastewater As An Odor Source
Information on the volatile fraction in wastewater may pin-point odors emanating
from the plant area. Odors can originate from wastewater in the same manner as air
pollutants. Hydrogen sulfide is easily desorbed from anaerobic wastewater. The absence of
odors from a plant area does not preclude that significant amounts of dissolved gases and
liquids are not being desorbed. since many common organics have a high recognition odor
threshold in air.
Loss Of Volatile* Effect Wastewater Test Results
Showing the existence of volatiles and quantifying the amount wi' .'ead to better
sampling and analysis procedures. The loss of the volatile fractions may lead to large errors
and widely fluctuating wastewater test results. At present little care is taken to assure that
the volatiles do not escape during sampling and handling of industrial wastewater samples in
the field or during the analysis in the laboratory. VVeimer and Lee (15) reported losses of
methane from lake water samples resulted in variable data. Succeeding analysis from the
same sample bottle indicated decreasing amounts of methane.
Volatile Losses Effect Wastewater Audits
The proper accounting of all the components (i.e., volatile and non-volatile fractions)
will allow more exact material balances and industrial wastewater audits. By-product
recovery and accurate design for treatment is based on reliable waste concentration
measures. A consciousness of the volatile nature of selected components in the wastewater
can lead to better accounting of the waste loads.
Information en the volatile fraction in wastewater, the volatile components, the
industry types which are likely to discharge volatiles, and the processes that are likely to
produce volatiles, increase our knowledge and understanding of industrial wastewater in
general.
Fundamental Oesorption Concepts
The volatile character of dissolved constituents in wa.tewater can be adequately
quantified by the experimental determination of two parameters. The symbols and
definition of these two parameters and the fundamental concepts revealed by the
parameters regarding volatile wastewater constituents are:
Volatile Fraction (Fv) this measure denotes the maximum amount (in %) of the
original organic pollutants in a water sample that can be removed by air contact.
-7-
-------
This is also the maximum efficiency of treatment that can.be achieved by
stripping the wastewater with dir. The organic pollutants in the wastewater can be
expressed as BO06. COD. TUC and other gross pollutant measures and/or
concentration of individual constituents (by gas/liqu:d chromatography).
Relative Volatilization Rate (K/a) - this measure denotes the ratio of the rate of
volatile removal by air contact to the rate water is evaporated in the same
apparatus. If the experimental value of K/a is greater than unity, stripping with air
may be a feasible treatment operation for this wastewater. If K/a is unity,
stripping will have no effect on removing volatile constituents from this
wastewater and if K/a is less than unity or zero, stripping will result in an increase
of this constituent in the wastewater. This parameter, like Fv, is dimensionless
and both ore determined from a single desorption experiment.
-8-
-------
SECTION IV
ITEMS OF PROJECT STUDY
The following is a list of the individual items of study considered to be necessary to
accomplish the objectives of this project.
A. Develop a laboratory desorption apparatus for a volatile component test method.
1) Test a laboratory bench scale raschig ring packed column as a laboratory
desorption apparatus. This column is to consist of a glass tube 5.08 cm. (2 inch)
in diameter packed to a height of 45.72 err.. ("8 inch) with 0.318cm (1/8 inch)
raschig rings. Air will be introduced at the bottom and liquid into the top. A
batch of wastewater will be continuously pumped over the packing. Samples of
the liquid taken at predetermined times will be tested for BOD5, COD. TOC and
GasChromatograph. The quantity of water lost (weight) will be measured also.
2) Test a laboratory bench scale glass-shot packed column. This column is to consist
of a glass tube 2.54 cm. in (1 .nc.Vi diameter packed to a height of 20.32 cm. (8
inch) with 0.159 cm. (V16 in.) glass-shot. This column will be operated just as
the raschig ring packed column.
. 3) Miniaturize and simpl'fy the desorption apparatus. The original apparatus
consisted of a packed column, liquid catch basin, pump, and laboratory air.
Control and flow measuring means are necessary for the flowing phases. Attempts
will be made to devise a method to operate the apparatus (liquid and air
movement) with air only. The operation of the apparatus under conditions of
natural convection of air and therefore requiring only a liquid pump will be
investigated. The lower limit of column size will also be established.
4) Establish the minimum and ni cessary equipment and apparatus needs. The above
column tests, miniaturization and simplification techniques will be aimed at
establishing the minimum and necessary equipment, apparatus services, etc.
desirable for conducting a reliable volatile component test.
B. Develop laboratory procedures for volatile component test method.
1) Establish the amount of a wastewater aliquot necessary to conduct the volatile
test. This volume is charged to the column catch basin and recycled over the
packing continuously. Samples for analysis are obtained from the catch basiri
volume.
2) Establish liquid and gas rates for the apparatus. There is some latitude on the flow
rates to the packed section, however, excessive high rates will cause flooding and
low rates will prolonc the experiment time.
3) Establish the time of test. It is important that the test proceeds long enough that
Fv can be quantified. Tests on a number of components should help estabiish the
- 9 -
-------
run time that will likely result in the removal of all the volatile constituents.
4) Analysis of experimental data. The raw experimental data will consist of organic
concentration (i.e. 8005. COD. TOC, G.C.,etc.) and water content (grams) with
experiment run time (minutes). Analytical expressions have been developed for
computing tho volatile component parameters from the raw data. Data handling
and processing procedures will be developed to compute Fv, a, K and K/a. These
procedures will be a combination of analytical processes involving algebra and
geometry.
C. Apply the proposed desorption test to determine:
1) Which of the several available wastewater test techniques for organic carbon is the
most accurate and/or desirable for establishing Fv,
2) Whether or not all the test techniques should result in identical values of F.
3) Fv values for simulated and actual industrial wastewaters,
4) Experimental values of K/a for selected pure components common to industrial
wastewcters.
5) Which o* the several wastewater test techniques is most accurate and desirable for
measuring K 'a.
6) K/a values for simulated and actual industrial wastes and selected pure
component., common to industrial wastewater, and
7) Whether or not all the test techniques should result in identical values of K/a.
D. Other studies important to the volatile desorption test.
1) Compute theoretical K/a for selected pure components. Analytical expressions
have been formulated for computing K/a. Theoretical calculations will be related
to experimental values.
2) Determine the effect of air temperature and relative humidity upon the K/a
measurement.
3) Develop charts and tables bv which laboratory measurements of K/a can be
corrected to the standard conditions.
4) Decide upon a standard (inlet) air temperature and relative humidity so that all
volatile wastewater K/a values can be compared on a common basis.
5) Study possible mass transfer interferences caused by surfactants, inorganic salts,
polymers, etc.. upon the desorption test results.
-10-
-------
SECTION V
THEORY OF VOLATILE COMPONENT DESORPTION
FROM AN AQUEOUS PHASE
Thcrmodynamic Basis For Desorption
A necessary condition for the transfer of a species from a liquid phase to a gas phase is
a favorable chemical potential. The thermodynamic vapor liquid equilibrium of dilute
soiir.ions is conveniently expressed bv the relative volatility of species A in water:
where ,j w = relative volatility
YJ w = activity coefficient
P] = pure component vapor pressure of species i
Pw = pure component vapor pressure of water
An alternate method of representing this equilibria is Henry's Law:
vf = MxiXj (2)
where YJ = mole fraction in the gas (i.e., air) phase for species i
Mxj = Henry's Law Constant for species i
Xj = mole fraction in the liquid (i.e., water) phase for species i
The relative volatility reveals more information about the desorbing ability of the species
since its removal from water is of main concern here. Table 1 to .tains a brief summary of
relative volatilities and Henry's Law Constant for common industrial gases and liquids. The
components 10 through 20 comprise a representative cross section of industrial organic
chemicals in wastewater. Many of the common liquid species are more volatile than water
and will therefore desorb readily as indicated by a relative volatility greater than unity.
Although some high molecular weight organics may have low pure component v«por
pressures and high boiling points compared to water they nevertheless exhibit large relative
volatilities due to large activity coefficients in water. Pierotti et. al. (16) report activity
coefficients of 103 to 107 for n acids, n primary alcohols, sec-a!:ohols, n-aldehydes,
n-ketones, n-esters, n-ethers, n-chlorides, n-paraffins and n-alkyl benzenes.
- 11 -
-------
TABLE 1
Vapor-Liquid Equilibria of Salactad Gases and Liquids in Water at 25° C
Component Normal Boiling
Point, °C
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Nitrogen
Hydrogen Sulfide (H2S)
Oxygen (©2)
Ethane (C2H6)
Prooylene (C3HQ)
Carbon Dioxide (C02)
Acetylene (C2H2>
Bromine (Bf2)
Ammonia (NH3>
Acetaldehyde
Acetone
iso-propanol
n-propanol
Ethanol
Methanol
n-butanol
Acetic acid
Formic acid
Propionic acid
Phenol
-195.8
- 59.6
-183
- 88.6
- 48
- 78.5
- 84
- 58.8
- 33.4
20.2
56.5
82.5
975
78.4
64.7
117
113.1
10CB
141.1
181.4
MX!
86500
54500
43800
30200
5690
1640
1330
73.7
.843
5.88
1.99
1.19
.471
.363
.300
.182
.0627.
.0247
.0130
.0102
1.*
2,768,000
1.744,000
1.402,000
966,400
182.100
52,480
42,560
2,358
27.0
188
63.7
38.1
15.1
11.6
9.60
5.82
2.01
790
.416
.326
-12-
-------
Descrption Of Dilute Aqueous Solutions In A Packed Column
An analytical model of the desorption of dilute solutions of gases and liquids in a
packed column has been developed by Thibodeaux, Estridge and Turner (17). The specific
volatilization rate for single component i in the desorption apparatus is:
(SRi- 1)
K'=1~
SRjExP[(ZT/HOGi)(SRj-l)] -1
where Kj = the specif'? volatilization rate of species i
SRJ = stripping factor for species i
Zj = height of the packing, cm.
HOG i ~ ne'9ht °' 9as phase (overall) transfer unit for species i, cm.
The specific volatilization rate definition is:
Kj - {LmjXjj LmoXjol/L-mjXjj (4)
where Lmj = molar liquid flow rate into packed section, moles/time area
Lmo = molar liquid flow rate out of packed section, moles/time area
Xjj = concentration of specie* i entering packed section, mole fraction
Xjo = concentration of species i leaving packed section, mole fraction
The stripping factor definition is:
SRJ = Mxi Gm/Lm (5)
where Gm = molar gas flow rate in packed section, moles/time area
From equ.-;ion (3) it is evident that Kj and hence the fraction of species i removed in a
packed section (Figure 2} is dependent on Henry's Law Constant, gas flow, liquid flow,
height of packed section, and the efficiency of the packing (i.e., HQQ j! but not on the
concentration of species i in the liquid. Assumptions necessary for use of equation (3) are
that the operation be isothermal, steady countercurrent flow, inlet air contains no species i,
and concentration of species i in water is very dilute (i.e., Xj-,-1).
The derivation of equation (3) assumes that the liquid concentration in the packed
section changes from top to bottom. Removal of this last assumption results in a simpler
equation:
-13-
-------
RECYCLE
AIR OUT PLUS
VOLATI2ED FRACTIONS
H20. a L,,
..
PACKED/
/COLUMN
AIR IN
I L -
I mo
v
_} CATCH
I BASIN M
Figure 2. Schematic of volatile desorption apparatus.
-14-
-------
Kj--:SRi [1- 1/EXP(ZT/HOGiH (6)
This development is exact for water evaporation
a = SRw[1-1/EXP(ZT/HOGw)l (7)
where a = fraction of water lost in column.
Water is most abundant in wastewater and changes little in concentration in the packed
section.
If the quotient Kj/a (defined as the relative volatilization rate) is created from
equations (5;. (6) and (7) and Zy -'"--the thermodynamic relative volatility results:
sRi Mxi
Mxw
= 'imit (Kj/a) (8)
This limit is immediate in the case of equations (6) and (7) but a numerical proof was
necessary in the case of equations (5) and (7). Calculations were performed with a 3 cm. x
15 cm. Intalox packed desorption apparatus to compare K/a using the exact equations (5,
7), the simplified equations (6. 7) and the relative volatility (8). Data for computing height
of a transfer unit for the packing (i.e., HQQ j) was from Treybal (18). Table 2 shows the
calculated results for eleven common industrial chemicals. The simplified equaf.on values of
Kj/a 's less than tj w as expected. Roughly correlated the relative volatility is related to K/a
by:
u j/w = (1. 032 ±. 108) x (Kj/a) (9)
This estimating equation will yield low values of a, w for highly volatile components,
however this result shows Kj/a is closely related to the actual relative volatility of the pure
components.
Most wastewaters contain several dissolved species. The above development is for single
species only. Desorption occurs simultaneously for each species and equations (5) and (7)
remains valid in a multicomponent mixture provided no synergystic effects are present.
Whereas K and a are highly dependent on the operating characteristics of the desorption
apparatus, the parameter K/a is somewhat insensitive to actual operating characteristics.
Equation (8) indicates that the experimental K/a values for pure components or mixtures
can be a useful parameter to quantify the volatility of constituents dissolved in wastewaters.
-15 -
-------
o>
I
TABLE 2
Relative Volatilization Rates and Relative Volat lity @ 25°C
(calculated results)
Component
Acetaldehyde
Acetone
Isopropanol
Normal Propanol
Ethanol
Methanol
Normal Butanol
Acetic Acid
Formic Acid
Propionic Acid
Phenol
HOG/
38.7 (cm.)
35.8
35.4
34.7
34.4
34.3
34.2
34.1
34.0
34.0
34.0
°i,w
188.
64.1
37.8
15.J
11.6
9.58
5.82
2.00
.789
.416
.326
Kj/a1*
159.
60.0
36.4
14.8
11.4
9.49
5.77
2.00
.788
.415
.326
if 1 *
Kj/a
169.
61.4
36.9
14.9
11.5
9.52
5.78
2.00
.789.
.415
.326
Error*
+6.3
+2.3
+ 1.4
+ .7
+ .9
.5
. .5
.0
.0
.0
.0
/ - Zy = 15 cm.
2 - Error of Kj/a wrt K,Va , percent
3 - Exact value
4 - Simplified equation value
-------
Experimental and Data Interpretation Model (or Packed Column
A desorption apparatus consisting of a packed column was the only device employed in
this work. Desorption apparati consisting of packed towers and sparged vessels were
employed by McAlister. Turner and Estridge (1). It was discovered early in the above
research that excessive foaming made sparged vessel desorption unpractical for pulp and
paper waste. It has been the authors experience that the sparged vessel also requires longer
experiment times than the packed column. Both types of desorption apparatus employ a
batch of liquid and therefore the behavior of each is transient in nature. The batch-sparged
vessel is doubly transient in nature because the interfacial area for mass-transfer is
dependent on the quantity of water remaining and water is readily vaporized in this
apparatus.
The schematic of the packed column desorption apparatus shown in Figure 2 is helpful
in visualizing the mathematic model of the packed column desorption experiment. A
volume of sample is charged to the catch basin and pumped uround and recycled over the
packing continuously. The model further assumes a constant air rate containing no volatiles,
constant wet bulb temperature, constant operating temperature and that the quantity of
volatiles removed is a constant multiple of the quantity of this component entering the
tower (i.e., molar desorption rate = KjXjLmj|.
A differential component balance for water and species i yields tv\o simultaneous
equations (17). If the concentration of water is much greater than the concentration of
species i (i.e., Xj-j Q » >Xj) an analytical solution is immediate:
log Xit = log Xio - (Kj/a - 1) log [Mo/Mtj (10)
where Xjt = concentration of species i in the desorption apparatus at
sample time, t, mg/l.
Xjo = concentration of epecies i in the desorption apparatus at start
of experiment, mg/l.
MI = quantity of water in the desorption apparatus at sample
time t, grams.
Mo = quantity of wattr in the desorption apparatus at start of
experiment, grams.
A log log graph of concentration of species i, X;t, and the quantity of water remaining, Mj,
resul.s in a straigh: line with slope equal to (Kj/a-1). A single desorption experiment thus
yields a value of the relative volatilization rate. Equation (1C) fhows that as water is lost the
concentration of species i decreases for Kj/a > 1, remains constant for Kj/a = 1 and increases
- 17-
-------
for Kj/a ^1. Water is vaporized at a constant rate and is related to run time:
Mt = M0-aLmt (11)
where t <= run time in minutes.
This relation must be used to compute a.
Industrial wastewaters may contain dissolved species of different relative volatilities.
Some dissolved components may be more volatile than water and others may be less volatile
than water. Gross organic pollutant measures (i.e., BOD5, COD, TOC, etc.) quantify the
concentrations of the combined volatile and non-volatile components. Gross concentration
measures can be expressed as the sum of the concentrations of the volatile and non-volatile
portions:
C°=C°v + Cnv (12)
where C° = the concentration of organic pollutants as measured by the
BOD5. COD, TOC tests, etc., mg02/l, mgC/l.
Cv = the concentration of organic pollutants which are more volatile
than water, mgO2/I, mgC/l.
Cnv = the concentration of organic pollutants which are less or of
equal volatility of water, mg02/l, mqC/l.
A single desorption experiment on a sample of wastewater will yield Cv and the relative
volatilization rate of the dissolved volatiles.
A detailed theoretical development for desorption operations involving
multicomponents is presented by Thibodeaux, Estridge and Turner (17). The following is a
brief summary of the working relationships needed for data analysis associated with the
desorption apparatus.
The relationship of the volatile and non-volatile factions in a wastewater consisting of
m non-volatile components and n volatile components is:
m
(13)
where Ft = the fraction of the total amount of the gross constituents
remaining in the wastewater at time t (Ft E CtMt/C°Mo).
-18-
-------
Fj = the original fraction of volatile component i in the
wastewater (F° H Xj0/C°).
F; = the original fraction of non-volatile component j in thp
wastewater (F? H
'This equation shows that as run time proceeds and water is vaporized (i.e., Mt -0) the
non-volatile fraction remains. Although a gross concentration measure cannot single out
individual constituents, a slight change in the form of equation (13) allows further study of
the volatile fraction. If the volatile fraction is assumed to be made of a single "pseudo
volatile component" then equation (13) can be reinterpreted as
Ft=Fv[Mt/M0]Ks/a + Fnv (14)
where Fv = the original fraction of all the volatile components in the
wastewater (Fy : CV/C°).
Fnv = the original fraction of all the non-volatile components in the
wastewater (F^v = Cnv/C°).
Ks = the relative volatilization rate representative cf all the volatile
components in the wastewater.
The "pseudo volatile component" is a fictitious lumped single component with a relative
volatilization rate of Ks/a which is equivalent, in gross combined behavior, to the n volatile
components with individual Kj/a 's. Figure 3 shows calculated results of the behavior of the
organic fraction, Ft, for various hypothetical relative volatility values, as a function of the
fraction of water remaining. This hypothetical wastewater has a volatile organic fraction of
80%.
Now. if only the volatile fraction is considered, equation (14) may be transformed to
yield
log *t = (K^a) log [Mo/Mt] (15)
where 4>t B (Ft Fnv)/Fv is the fraction of the original volatile fraction
remaining at run time t.
A log-log plot t vs [Mo/Mt)experimental data yields information about Ks/a. Reinhardt
(19) has demonstrated that Kg/a is not constant if the wastewater contains several
components with different relative volatilities. By the use of a graphical interpretation
technique, as shown in Figure 3, and equation (15) it is theoretically possible to establish
-19-
-------
0.2
-20
1.0
.95
.90
Mt/M0, WATER FRACTION,
.85
0.0
Figure 3. Hypothetical packed column desorption experiment results.
-------
the volatile and non-volatile organic concentrations and an estimate of the relative
volatilization rate by a single packed column desorption experiment on a sample of
wastewater.
-21-
-------
SECTION VI
EXPERIMENT FABRICATION, MATERIALS AND METHODS
Packed Column Volatile Component Desorption Apparatus
Initial packed column desorption studies were performed by McAlister, Turner and
Estridge (1). The column employed was of pilot scale dimensions. It was a square columnar
gas liquid contactor of 232 square cm. (0.25 square ft.) cross sectional area and Plexiglas
construction. Overall height was approximately 3.66 meters (12 feet). 3.05 meters (10 feet)
of which were packed with Poly Grid media stacked vertically on 2.86 cm. (1 1/8 inch)
centers. The experimental work performed on this pi.'ot scale tower gave excellent
desorption results and was an aid in predicting the effect of air stripping as a wastewater
treatment operation.
The above described tower was much too large to be practical as a routine laboratory
test apparatus for desorption studies. As has been pointed out previously a packed column
desorption device has been demonstrated to be the best possible desorption apparatus from
an operational point-of-view. Two laboratory or bench scale, packed column desorption
devices were employed in the present study.
Raschig Ring Column Desorption Apparatus
A schematic of this apparatus is shown in Figure 4. The apparatus was bench scale and
was placed in a standard laboratory hood enclosure.
Glass tubing was used as the column shell. The inside diameter was 4.76 cm. (1 7/8
inches) and the wal! thickness was .318 cm. (1/8 inch). The length of the column was .914
m. (3 feet) with a bed support 15.24 cm. (6 inches) from the bottom of the tube. The
bottom 15.24 cm. (6 inches) of the tube was graduated in inch marks. The bed support was
a circular piece of stainless steel wire screen which rested upon glass knobs on the inside of
the tube wall. Ceramic Raschig rings, 1.27 cm. (1/2 inch) size, were used as the packing
material; the rings were placed on top of the bed support to a height of .610 m. (2.0 feet). A
clearance of 15.24 cm. (6 inches) remained above the top of the packing. The column was
secured in a vertical position with the aid of a bubble level.
A four liter glass beaker was used as the solution catch basin. The bottom of the
column was elevated above the counter top so that the beaker could be positioned
underneath. The beaker was adjusted so that the bottom of the column was submerged in
the solution held by the beaker. This offered a liquid seal for the bottom of the column.
This seal could only withstand a pressure of approximately 15.24 cm. (6 inches) of water.
Tygon tubing and a variable speed pump were used to pump the liquid from the catch basin
-22-
-------
LIQUID
TUBING
LIQUID PUMP
GLASS BEAKER
ORIFICE
METER
MAIN
HOOD
VALVE
WATER
TRAP
Figure 4. Raschig ring desorption apparatus
-------
to the top of the column. The solution was allowed to exit from the Tygon tubing and fall
approximately 2.54 cm. (1 inch) to the top of the packing. From the top of the packing, the
solution sought its own path down -through the packing, finally, into the catch basin. A
thermometer placed in the catch basin measured the temperature of the solution.
The countercurrent flowing desorption gas used was compressed air which was
obtained from an air-outlet within the hood. Before entering the column the air passed
through the following devices: a water trap; a drying agent (DRIERITE) to remove water
vapor; a pressure regulator; a pressure gauge; a valve; and an orifice plate (.254 cm. = 1/10
inch). The orifice differential pressure was measured with mercury manometers. From the
orifice outlet, the air traveled through Tygon tubing which ended inside the tower about
1.27 cm. (1/2 inch) beneath the wire screen support. A thermometer placed in the air line
after the orifice meter measured the dry bulb temperature of the entering air.
Solution samples from the catch oasin were withdrawn with a volumetric pipette. The
samples were placed in rubber stoppered test tubes and stored in a refrigerator at -35°F
(~2°C) until time for analysis. A low temperature is needed to minimize loss of volatiles
during sample storage.
Operating conditions:
Solution or wastewater volume placed in catch basin beaker = 2 1/2 to 3 liters.
Solution flow rate = 700 ml./min.
Air flow rate = 52.8 I. at STP/min. (1.83 SCF/min.).
Pressure at column outlet was ambient atmospheric.
The quantity of wiitei lost by evaporation was obtained by initial and final weight
of water in the catch l)asin beaker with allowances made for water retained
in tubing, packing and column.
15 ml. samples were obtained from the beaV^t every 15 minutes for a period of 3
1/4 hours {14 samples total).
Intalox Saddle Column Desorption Apparatus
Experience obtained with the above device led to the fabrication of a still smaller
column desorption apparatus. An illustration of this device is shown in Figure 5. Three
models were attempted before this one was decided upon. Glass shot packing proved to be
inadequate at the high gas/liquid ratios desirable for a short experiment time.
The column consisted of a male ground glass joint (-20 cm.) topped with a glass funnel
with a 10 cm. (large) opening. The internal diameter of the column was 3 cm. and a stainless
steel screen, supported on glass indentures, was placed near the bottom but above the
ground glass joint. The column was packed to a height of 15 cm. with .953 cm. (3/8 inch)
Intalox saddles. The top inlet nipple was fitted with a sample aperature (see illustration).
The catch basin consisted of a erlenmeyer flask (500 ml.) fitted with a ground glass female
joint. An air inlet nipple and liquid outlet nipple completed the desorption apparatus. The
-24-
-------
relative positions of the nipples is shown in Figure 5.
Auxiliary equipment necessary to operate the apparatus consisted of a parastaltic
pump (threaded tubing type with removable pump heads), magnetic stirrer and hot-plate
combination, bottled dry air and pressure regu'ator. Interconnecting tubing was Tygon.
Consult Figure 5 for arrangement of the pieces of equipment. A thermometer (Hg type)
placed in the funnel opening under the downspout measured the liquid temperature.
Operating conditions and procedure:
Assure apparatus is clean and dry and contains the required quantity of packing in
the joint and the stirrer bar in erlenmeyer flask. Assemble rr edified joint and
erlenmeyer flask as shown in Figure 5.
1. Calibrate oarastaltic pump with plastic tubing (Tygon or equivalent) and set
flow rate at about 150 mlVmin. with tap water (130 to 160 ml/min. range).
2. Connect a .9 - 1.2 m. (3 - 4 ft.) length of this tubing to bottom nipple on
erlenmeyer and nipple on joint.
3. Add 300 to 350 ml. of wastewater to the apparatus through the funnel
opening. The sample added should contain a minimum of floating,
suspended or settleable solids. Record the exact quantity of sample added.
4. Weigh assembled apparatus, i.:tached tubing and wastewater as a single unit.
Record this initial weight.
5. Thread the recycle tubing through the parastaltic type (squeeze) pump.
6. Connect dry, oil free laooratory (or bottled) air to the top nipple of the
erlenmeyer with rubber tubing.
7. Adjust apparatus and contents to 25°C.
8. Start magnetic stirrsr and adjust pump to deliver water to packing at preset
(150 ml./min.) rate. Maintain this liquid rate (130 - 160" ml./min.)
throughout the experiment.
9. Upon starting pump take an initial sample (1 ml.) immediately. Carefully
record the quantity of sample withdrawn and the time.
10. Immediately after starting pump adjust and maintain air flow rate so that
flooding is incipient (- 12.1 l./min. at STP). Maintain this air flow rate
throughout the experiment. Flooding is an undesirable phenomena in
countercurrent packed columns that is characterized by excessive liquid
hold-up in the packing, so that the gas begins to bubble through the liquid
and eject liquid droplets out the top of the apparatus.
11. Operate in this fashion for a period of one to three hours. Withdraw and
record volume of samt-les (usually 1 ml.) at convenient time intervals (5
10 min.). Adjust the temperature of the liquid to 25°C ± 1°C by means of
hot plate or heating tape pla-x-d around the base of the erlenmeyer. The
entering dry air will cause the water in the apparatus to cool significantly
-25-
-------
THERMOMETER
FUNNEL
ERLENMEYER
FLASK
HOv.ac
PARASTALTIC
PUMP
MAGNETIC
STIRRER AND HOTPLATE
Figure 5. Intalox Saddle desorption apparatus
-26-
-------
due to the evaporative loss of water. Heating will be necessary to maintain a
25"C operating temperature.
12. Upon completion of the experiment stop the air flow, shut down the pump
and turn the heater off. Record the final run time.
13. Withdraw the final sample.
14. Disconnect the air tubing from erlenmeyer and unthread the liquid tubing
from the pump.
15. Re-weigh the assembled apparatus, attached tubing and remaining
waitewater as a single unit. Record this weight.
16. Disassemble apparatus and clean parts in preparation for next run.
17. Keep samples under refrigeration (-35°F = 2°C).
Wastewater Test Methods
Samples obtained from the desorption jpparati were analyzed for their organic
concentration. Analysis performed were: 5 day biochemical oxygen demand !BOD5).
chemical oxygen demand (COD), gas chromatograph (GO and total organic carbon (TOO.
For the BOD5 and COD tests, the apparatus and procedures described in Standard
Methods (20) was employed. For the GC analysis a Perkin i Imer Mode! 800 equipped with
1.83 m. (6 foot). .318 cm. (1/8 inch) O.D. columns was used. Separation columns used
were:
amine and K-400 on 35 mesh size particles
1C% Chrom-W on 80 - 100 mesh size particles
The flame ionization unit on the chromatograph was used exclusively with a Speedmatic G
recorder. One microliter to ten microliter samples were employed.
TOC analysis were performed by instrument also. An Oceanographic International
Model 0524 (ampule type) was employed for the bulk of the experimental work. A
Beckman Model 915 (injection type) Total Organic Carbon Analyzer was used on samples
from eight desorption experiments.
A single inorganic desorption experiment was performed with ammonia. Standard
Methods (20) was employed in the ammonia analysis.
Experimental Data Handling Techniques
The raw data obtained from the desorption experiment consisted of:
a) independent variable time (t) in minutes
b) dependent variable water content (Mt) in grams and
c) dependent variable organic concentration (C() in mg02/l, mgC/l, or area
units/I.
-27-
-------
The water content at eac'. tims is proportional to the run time and the total quantity of
water vaporized during the run. The total quantity of water vaporized is computed by:
A MT = Wj- Wf + n« A m (16)
where A My = total mass of water vaporized during experiment, grams
Wj = initial weight of desorption apparatus plus water sample, grams
Wf = final weight of desorption apparatus and remaining water at end
of run (tf ), grams
n = total number of samples withdrawn from apparatus
£ m = individual sample size withdrawn from apparatus, grams.
The water content at any time, t, is computed by:
M, = M0- A M(t/tf) (17)
where tf = experiment run time, minutes.
Water in the "as procured" condition (i.e., with interfering chemical species if originally
present) is vaporized at a constant rate if temperature, air humidity, air rate, liquid rate
remain constant throughout the run. Table 3 contains an example of the raw data.
The raw experimental data appears in the Appendix.
Data analysis techniques with individual species tests or single pure components in water
The quantity of species i >n the wastewater sample (Xjo) is obtained from the initial sample
withdrawn before the desorption experiment is performed. Equation (10) is the
mathematical tool employed to obtain the relative volatilization rate (i.e., Kj/a) of species i.
Kj/a can be obtained graphically by preparing a log-log plot of Xjt and (Mo/Mt) as directed
by Equation 10. The slope of a straight line through these raw data points yields (Kj/a 1).
Approximately the same result can be achieved by the use of linear regression. Equation
(10) is linear and of the form:
(18)
where Y ; log Xjt,
a H logXj0,
b = -(Kj/a- 1),
X = log lMo/Mt] .
-28-
-------
TABLE 3
Typical Desorption Experiment Raw and Transformed Data-Fibers. Chemical and Plastic Industry Sample
A.
t.time
(min.)
0
5
10
20
30
40
50
55
60
c-»
Raw Data
M{, water
(grams)
300.
296.5
293.
286.
279.
272.
265.
261.5
258.
Ct, organic
(mgC./l)
452
432
413
397
379
375
361
365
366
Mt/M0
1.000
.9883
.9767
.9533
.9300
.9067
.8833
.8717
.8600
.0
8. Transfo
Ct/C°
1.000
.9558
.9137
.8783
.8385
.8296
.7987
.8075
.8097
rmcd Data
Ft
1.000
.9446
.8924
.8373
.7798
.7522
.7055
.7039
.6964
F°nv=.643
A
'" t
1.0uC
.8448
.6986
.5443
.3832
.3059
.1751
.1706
.1496
(O
I
-------
Multiple linear regression computer software packager can be employed to yield Kj/a plus
statistical information related to the best fit straight line correlation (i.e.. t-test on
significance of the slope, standard error of the estimate, multiple correlation coefficient,
etc.). Such a software package was employed in all computations of Kj/a in this work.
>t should be pointed out that the above procedures can also be employed with gross
pollutant measures (i.d., BGL>5, COO and TOC) on simulated wastewater consisting of single
pure components in distilled water (i.e., methanol ir> water). The above procedures should
not be used if the simulated waste consist of a mixture of pure components added to
distilled water.
Data analysis techniques with gros; pollutant measures of pure component mixtures and
wastewater samples A pseudo relative volatilization rate (i.e., Kj/a) and the original volatile
fraction (Fv) of a simulated or actual wastewater sample may be obtained from the data of a
single packed column desorption experiment. Equations (14) and (15) are the directives by
which this data can be obtained. A graphical data analysis technique or an analytical data
analysis technique may be employed. Prior to employing either technique the raw concen-
tration data (i.e., Ct) must be transformed to Ft by:
Ft = (Ct/C°) (Mt/M0) (19)
The graphical let hnique employs the directive of equation (14) and a plot of Ft vs (Mt/Mo)
is made on Cartesian coordinate graph paper. The resulting plot should be similar to the one
shown in Figure 3. The non-volatile fraction (i.e., Fnv) can be estimated by visually
extrapolating the data curve to Mt/Mo = 0. The volatile and non-volatile fractions are
related:
Fv + F°nv = 1 (20)
Cnce the non volatile fraction has been obtained equation (15) can be employed to obtain
Ks/a. A log-log plot of vt vs. (MQ/Mt) yields a line from which Ks/a can be obtained.
Reinhardi <19) points out that the Ks/a is not necessarily constant if the waste contains a
mixture of cc.':1;: vnenis of widely varying relative volatilities. He further points out that a
slope obtained at Mt/M0 near unity is indicative of the highly volatile components while a
slope obtained at the low value of M,/MO is representative of the less volatile components. A
non-linear plot, therefore is an indication of more than one volatile organic component in
the wastewater sample.
An analytical technique was developed to obtain Fnv and K^'a. This analytical
technique was improvised to alleviate the extrapolation step necessary in the graphical
procedure. If Ks/a is small, the extrapolated estimate of Fnv is very difficult to estimate
-30-
-------
with any decree ot certainty. The calculated results shown in Figure 3 reveal that in a
normal sixty minute run (i.e.. Mt/Mo - .85) Ft is far removed from Fnv = .2 for Kj/a <5.
The analytical technique is based upon the heuristic notion that there is a unique value
of Ks/a and equation (15) wilt be a "best fit" to the transformed experimental data (>t vs.
MO/MI) in a least squares sense. A computer algorithm has been developed which performs a
one dimensional search (Golden Section) for Fnv on the range (0.0. Ft (min.)) .Thealgorithm
produces the unique values of Fnv and Kj/a when the coefficient of variation is a minimum.
The coefficient of variation is given by:
CV = SE/(Ks/a) (21)
where CV = coefficient of variation
SE = standard error of estimate of correlating equation (15)
Ks/a = least squares slope of correlating equation (15)
This "non-volatile fraction" computer algorithm is reproduced in the appendix along with
documentation for its use.
Wastewater Samples Employed
Simulated wastewater samples A total of fifteen (15) pure cnemicals were employed en
this project. These pure components were combined with water singly or in mixtures to
create wastewater of various volatilities and volatile fractions. The starting concentrations
were adjusted roughly to 1000 ppm total carbon content. The pure components employed
were:
acetaldehyde phenol
acetone formic acid
iso-propanol ethanol
methanol ammonia
normal propanol benzene
furfural styrene
propionic acid sucrose
acetic acid
Industrial wastewater samples A total of nineteen (13) industrial wastewater samples
repres- n....g fifteen different industry types were tested ir, the laboratory desorption
apparatus. These samples were obtained directly from industry personnel, from local city
-31-
-------
wastewater treatment plant operators and a local engineering consultant. The industry types
and major product represented by samples are:
poultry, liquid egg
poultry, turkey processing
poultry, broiler processing
metal, hand tool manufacture
petroleum production, oilfield blowpit
canning, grape products
pharmaceutical, specific product unknown
paper, unbleached kraft
paper, tissue and plywood
food, margarine-shortening
petroleum, refined petroleum products
petrochemical, 1, 3 butadiene
petrochemical, 1. 2 dichloroethane
paper (bleached), pulp and kraft paper
fibers, chemicals and plastics
Combined wastewater samples A total of two samples were obtained of combined
industrial and municipal wastewater. These samples were obtained prior to the primary
treatment unit from:
City of Fayetteville, Arkansas Pollution Contiol Plant
City of Springdale, Arkansas Wastewater Treatment Plant
-32-
-------
SECTION VII
EXPERIMENTAL PHASE AND RESULTS
Physical Experimental Results with Pure Components and Simulated Wastewater
Desorption in packed columns is a well established process in both the theoretically
and practical aspects. Since this knowledge is readily available only selected experiments
need he performed to confirm the predicted behavior of this proposed desorption device.
Equations (3). (7) and (10) constitute the critical packed column, batch desorption
relationships.-The predictions of these equations must be verified by physical experiment in
order that desorption data be interpreted in the proposed fashion. Equation (3) predicts
that Kj is independent of the concentration of component i and dependent only upon the
column, flow rates and species relative volatility. Equation (7) predicts that water is
evaporated from the column at a constant rate dependent upon the same type parameters.
Equation (10) predicts a log-linear relationship between the concentration of component i
and water in a batch desorption experiment if Kj and a are indeed constant.
An extension of the above theory to aqueous systems containing more than one
dissolved component has been constructed based on the validity of the single component
model. In general, wastewater 'in be env'sioned to contain volatile components (i.e , KpO)
of a quantity fraction, Fv, and r. o-volatile components (i.e., Kj = 0) of a quantity fraction,
Fnv. Equation (14) predicts the general desorption behavior of a wastewater containing
volatile and non-volatile organic components.
Selected pure components were chosen to verify the proposed desorption analytical
model. Pure components were chosen based on relative volatilities and on the occurrence of
the species in industrial wastewater. A total of fifteen pure components were employed in
various desorption experiments. Selected pure components were combined to fabricate a
"simulated wastewater" with a volatile and non-volatile character.
Raschig Ring Packed Desorption Column A total of approximately twenty five
experiments were performed with single pure components and simulated wastewater with
the raschig ring column. Four test techniques were employed to obtain concentration of the
remaining organics in the samples. Selected experimental results are shown in Figures 6
through 9. A straight line relationship was the acminant behavior in all experiments with
single pure components when plotted as directed by equation (10). Table No. 4 contains the
relative volatilization r^te data for the pure component experiments. All experimental data
was processed on a linear regression computer program and statistical data was generated.
Table 5 contains selected statistical parameters of the various experiments.
Three simulated wastewater experiments were performed with the raschig ring column.
-33-
-------
TABLE 4
Relative Oesorption Rates of Selected Pure Components
With Raschig Ring Column
Constituent
acetaldehyde .
acetone (run 1)
acetone (run 2)
i-propanol
methanol (run 1)
methanol (run 2)
n-propanol
n-butanol
furfural
propionic acid
acetic acid
phenc1 dun 1)
phenol (run 2)
formic acid
Experimental K*/a
TOC
207.
51.8
99.7
15.2
21.2
10.98
12.4
10.4
-
-37
2.23
.733
1.49
6.99
BOD(5)
207.
98.8
36.1
2.73
7.29
13.11
22.6
11.0
-
1.51
4.81
1.91
1.13
COD
171.
85.8
45.6
10.7
8.91
8.84
14.8
18.3
10.0
0.00
1.00
.557
.977
GC
55.8
48.4
-
7.73
10.21
-
-
-
-
-
-
-
Theoretical
K/a
106.
46.3
42.1
42.3
28.4
20.7
16.0
6.18
4.00
2.29
1.00
1.13
1.12
*K is the specific desorption rate of the organic component
a is the specific desorption rate of water, both in a 45 cm ht x 5 cm diameter.
Raschig ring, packed air stripping (desorption) column.
-34-
-------
7.0
6.5
Q
O 6.0
x
"o
o
u
5.5
S.O
0.90
0.94
0.98 1.02 1.06
In (Mt), Mt IS IN liters,
1.10
Figure 6. Desorption of methanol in water with Raschig Ring Column - COO analysis
-------
o>
I
7.0
6.5
Z
-------
CO
VI
7.0
6.0
I60
£ 4.0
O
O
CO
". 3.0
O
O
m
X
f 2.0
1.0
0.0
0.96
I
I
1.00
2.02
1.04 1.08
ln(M,),M, IS IN liters.
Figure 8. Desorption of acetone in water with Raschig Ring Column - BOD6 analysis
-------
8.0
I 7.0
a
2
-------
TABLE 5
Raschig Ring Experiments Linear Regression Statistical Data for Pure Components
Analysis
COD
GC
TOC
BOD
COD
GC
TOC
BOD
COD
TOC
BOD
TOC
BOD
COD
TOC
TOC
BOD
COD
TOC
BOD
COD
TOC
BOD
COD
TOC
BOD
COD
Methanol-Watpr
Standard Error Multiple Correction
Estimate Coefficient
IExp.1) (Exp. 2) (Exp.1) (Exp. 21
0.054. 0.555 0.976, 0.974
0.049. 0.081 0.973, 0.960
0.257. 0.070 0.923. 0.974
0.101. 0.114 0.885. 0.955
Acetone-Water
1.021. 0.554 0.811. 0.940
0.704. 0.439 0.777, 0.892
0.698. 0.704 0.680, 0.916
0.885. 0.474 0.909, 0.919
Phenol-Water
0.020, 0.016 0.563. 0.043
0.028, 0.053 0.279. 0.269
0.030. 0.045 0.679. 0.089
Acetic Acid
.0463 .543
.0463 .894
.0492 >.029
Formic Acid
.152 .384
Propionic Acid
.0749 .293
.129 .142
.0467 .617
Acetaldehyde
.354 .949
.665 £07
.375 .920
n-propanol
.048, 587
20} .940
.044 .992
i-propanol
.135 513
.391 .094
.110 .832
t value
for K/a - 1
-------
Simulated wastewater No. 1 consisted of -500 ppm acetone and 500 ppm phenol added to
distilled water. Simulated wastewater No. 2 and No. 3 consisted of approximately 330 ppm
acetone, 330 ppm phenol and 330 ppm methanol in distilled water. Figures 10, 11 and 12
show the test results for these two simulated wastewaters.
The above experiments with single pure components and simulated wastewater
demonstrated that the general behavior of the experiment results are as predicted by the
desorption model equations. The log log straight line behavior of single pure components
was as predicted. The volatile and nonvolatile multiple component behavior was
qualitatively correct also. The quantitative data shows much variation and this is attributed
mainly to the water evaporation lost from the apparatus. Average water loss for a 180
minute run was. 253 grams with a range of 217 to 348 grams for a percent deviation range of
-14.2% to + 37.5%. The errors inherent in accurately quantifying the water evaporated
initiated a search for a more satisfactory desorption device. Other operational problems
included; flooding due to pressure surges in air line, uncertainty in inlet air relative
humidity, inability to control column temperature, three hour run time, and equipment
bulkiness.
Intalox Saddle Packed Desorption Column Three models of a laboratory scale desorption
column were attempted before the design shown in Figure 5 was adopted. G
-------
1.0
o
z
1 0.8
<
2
ui
cc
to
u
O
tr
o
z
o
oc
o
o
u
<
oc
0.6
0.4
0.2
0.01
0.000
I
I
I
I
LEGEND
OCOO ANALYSIS
&TOC ANALYSIS
BOD ANALYSIS
I
I
0.010
0.020 0.030 0.040 0.050
1 - M,/M0. FRACTION WATER VAPORIZED.
0.060
0.070
Figure 10. Desorption of simulated wastewater No. 1 with Raschig Ring Column.
-------
NJ
0.000
LEGEND
O COD ANALYSIS
A TOC ANALYSIS
DOD ANALYSIS
J_
0.010
0.020 0.030 0.040 0.050
1-M,/M0. FRACTION WATER VAPORIZED.
Figure 11. Dcsorption of simulated wastewater No. 2 in Raschig Ring Column,
0.060
0070
-------
1.0
O
z
z
< 0.8
LU
CC
(SI
O
O 0.6
O
cc
O
LL
O 0.4
§
cc
0.2
0.0
LEGEND
OCOO ANALYSIS
A BOD ANALYSIS
I
I
I
0.00
0.01
0.02 0.03 0.04 0.05
1-M,/M0, FRACTION WATER VAPORIZED.
0.06
Figure 12. Desorption of simulated wastewater No. 3 with Raschig Ring Column.
-------
2.6
I
*
2.4
2.3
2.2
. 2.1
5
r
Z 2.0
o
3
2.4
2.3
2.2
2.1
2.0
K/a = 9.1
12-27
RUN
1-21
K/a
RUN
1-22
1 - 1 - 1 - ,
2.40 2.42 2.44 2.46 2.48 2.40 2.42 2.44 2.46 2.48
LOG M,.M, IN grams.
Figure 13.. Dcsorption of n-Butanol in Intalox Saddle Column - TOG
2.5
-------
1.6 -
*»
tn
0.8
0.6
30
60
I
I
90 120
RUN TIMEJmin.),
ISO
180
Figure 14. Oesorption of sucrose solution in Intalox Saddle Column TOC analysis
-------
hour run. Figure 15 shows the magnitude of discrepancy encountered when concentration is
employed rather than quantity on a normal one hour experirrer.t with a poultry processing
industry wastewater. As run time increases the error increases.
A series of three experiments were undertaken to determine if certain compounds
could affect the desorption test results. Since desorption is an interphase mass transfer
process, compounds were selected which have been known to interfere with the transfer
process Certain substances tend to concentrate at the interface, and may hinder the
interphase transfer of solute. Cetyl alcohol, when spread upor water in remarkably small
concentrations reduces the rate of evaporation of water into cir by as much as 95% (18).
Surfactants, such as sodium tetradecyl suliate (NaTDS) suppresses convection by inhibiting
flow at the interface (21). Goswami (11) report: that Na2 ^P04 reduces the aldehyde
desorption rate. However, this is in conflict with a referenced report that it increased COD
removal in a refinery wastfiwater from 72 to 90%. Normal hutanol was desorbed in the
Intalox Saddle device in the presences of: 30 ppm (volume) cetyl alcohol, 50 ppm (volume)
NaTDS and 30 ppm (volume) Na2HPO4, Figure 16 contains the experimental data for these
interference compounds, Cetyl alcohol appears to effect the desorption slightly. The
desorption effect of the other two compounds is not significant.
A gas-chromatogrjph (GO may be employed to determine the concentration of some
organirs in wastewater. A simulated wastewater was fabricated consisting of known volatile
and non-volatile species. Simulated wastewater No. 4 consisted of equal parts (- 165 ppm)
sucrose, phenol, n-butanol. acetic acid, acetaldehyde and acetone. Desorption was
performed in the Intalox Saddle device and GC was used to detect the total organic
concentration. The sum total area of all components was employed; no attempt was made
to obtain the concentration of individual species. The remaining organic fraction is:
f\A = tMt/M0) (22)
where Ft ^ = gas chromatograph organic fraction remaining
At = total integrator output area at time t, cm2.
AQ = total integrator output area at time initial, cm2.
Figure 17 shows the test results obtained with the 15% Chrom-W column. Sucrose, phenol
and acetic acid are non-volatile yet >9C?o of the GC-apparent organics are desorbed in a one
hour run. Gas chromatographic analysis techniques employing total integrated area output
as a measure of organic concentration will yield high values of the volatile fraction since the
non-volatile high molecular weight components may go undetected.
-46-
-------
VJ
I
0.0
10
20
I
O - C/C,,
30 40
RUN TIME, (min.).
50
60
Figure 15. Desorption of poultry processing WdStewater in Intalox Saddle Column - IOC analysis
-------
I
&
I
2.6
2.5
2.4
!
6 2.3
x~
if 2.2
o
3
2.1
2.0
CETYL
ALCOHOL
30ppm
I
I
K/a = 6.71
SODIUM
TETRADECYL
SULFATE
50 ppm
I
K/a = 7.58
I
I
SODIUM
PHOSPHATE
DIBASIC
30 ppm
I
I
K/a = 7.89
I
I
2.40 2.42 2.44 2.46 2.48 '2.40 2.42 2.44 2.46 2.48 2.40 2.42 2.44 2.46 2.48 2.50
LOG M,, M, IN grams,
Figure 16. Desorption of n-Butanol in Intalox Saddle Column - TOC analysis
-------
I
CO
10.
20.
30. 40.
RUN TIME, (min.),
50.
60.
Figure 17. Desorption of simulated wastcwatcr No. 4 with Intalox Saddle Column -GC bnalysis.
-------
Analytical Experiment Results
It was mentioned at the beginning of this section that the science of desorption in
packed columns is well established. The foregoing experimental results verify the important
predictions of the desorption model equations. In this section the mathematical equations
of packed column desorption will be employed to study aspects of the experimental
apparatus which would require prolonged laboratory experimentation to verify. Equations
(3) and (7) constitute the important desorption model relationships and will constitute the
"experimental apparatus" in this section of the report.
Raschig Ring Analytical Model Results Model equations (3) and (7) were employed with
the apparatus dimensions and operating conditions of the Raschig Ring desorption
apparatus. This device did not have mfans of temperature control. Therefore, the
calculations of K/a were performed at actual operating temperature and not at 25°C. The
inlet relative humidity for the calculations is assumed to be 0%, however there is some
doubt whether this condition was consistently maintained during all the experimental runs.
Table 4 contains the calculated K/a values and are listed under the column titled:
"Theoretical K/a". In general the analytical model does a fair job in predicting the relative
volatilization rate of pure components.
Temperature and relative humidity is considered to have a large effect on the relative
volatilization rate. Equation (7) was employed to study the effect of column temperature
and inlet air relative humidity on the fraction of water lost in the column. Figure 18 shows
the effect of temperature and % relative humidity is significant. These predicted results
stimulated a search for a means of maintaining a high degree of control over the temperature
and relative humidity during the experiment.
Intalox Saddle Analytical Model Results Model equations (3) and (7) were employed with
the apparatus dimensions and operating conditions of the Intalox Saddle desorpiion device.
Column temperature is maintainc.-d at 25°C ±1 C by means of heating tape or a rheostat
controlled hotplate and dry bottled air is employed with the Intalox Saddle device. We
became concerned with the wide variation in liquid flow rate and the effect it may have on
the relative volatilization rate. We attempted to maintain a liquid flow rate at or near 150
ml./min. Uncontrollable changes in the pumping head, etc., caused the actual measured rate
to fluctuate between 130 and 160 ml./min. Employing the ana'ytical model of the
experimental desorption apparatus, the effect of L(l/min) on K/a for n-butanol was studied.
Figure 19 shows that there is only a slight variation in K/a (2.4%) for a sizable variation in
U33.3%). Figure 20 shows a similar study of the effect of the dry air rate. A 10% variation
in K/a was computed for n-butanol for an air flow variation of 75%. It appears that
moderate deviations in liquid and air flow rates will have a relatively insignificant effect on
-50-
-------
0.0022
0.0018
- 0.0014
0.0010
0.0006
0.0002
0.0000
10.0
50
15.6
60
21.1
70
26.7
80
% RELATIVE HUMIDITY
OF THE ENTERING AIR
10
70
80
90
I
32.2
90
TEMPERATURE OF THE WATER
Figure 18. Water evaporation calculations in the Raschig Ring Column.
I
37.8
100
t
°F
-51 -
-------
4.5
4.0
3 3.5
3.0
2.5
100
110
I
I
I
120 130 140
LIQUID FLOW RATE, (cc/min.).
150
Figure 19. n-Butanol desorption calculations - effect of liquid rate.
-------
U1
w
5.5
5.0
4.5
4.0
3.5
3.0
= 2.5
2.0
1.5
0.2
5.66
_]
0.3
8.49
I
0.4
11.33
I
0.5
14.16
I
0.6
16.09
I
0.7
19.82
I .
0.8 (cu. ftVr.iin.)
22.65 (litert/minj
AIR FLOW RATE. AT STP.
Figure 20. n-Butanol desorption calcufations - effect of air rate.
-------
K/a, however increased rates tend to decrease the experiment run time.
Calculated results of the relative volatilization rate of pure components for the Intalox
Saddle device are shown in Table 3. T..ese calculated results indicate that the K/a equals the
relative volatility for those components that are non-volatile in nature.
Calculations were performed to develop a correction factor relationship for ar.
experimental relative humidity other than 0%. The analytical model results indicated that
the air emerging from the Intalox Saddle packed column (15 cm. in height) was saturated
with water and makes possible the following equation:
(K/d)e @ 0%RH = »K/a)e @ E%RH x ( 1100 - E%RH>/100) (23)
where (K/a)e@0%RH = the relative volatili??tion rate at 0%RH.
(K/a)e@E%RH = the relative volatilization rate at the experiment
relative humidity,
E%RH = the experiment percent relative humidity.
The experimental K/a result obtained when humid air is employed will be iarger than the
standard condition (i.e., 0% RH) value and the above equation should be empioyed as a
correction to this standard.
Experiments With Industrial Wastewater Samples
Industry representatives responding favorably to our request (by letter) for wastewater
samples were instructed to procure and handle the sample as follows:
quantity of sample. 500 ml.
container: small mouth plastic bottle
sample point: total wastewater effluent prior to any treatment operation (i.e., raw)
sample type: single grab aliquot during daytime under normal operating conditions
sample handling: adjust pH to 3.0 or lower, seal cap to bottle with electrical tape or
other type of rubber tape that will stretch and form a vapor seal
sample information: i \\ in requested information and attach tag to bottle neck. Tag
contains return address and stamps on face and general sample information on
backside, (requested information was: industry type, major product and daily
wastewater flow)
packaging: no special packaging is necessary. Attach tag to bottle neck.
mailing: place sealed, tagged bottle in U.S. Mail (parcel post).
A total of twenty seven letters were sent to industry representatives. Eight positive
responses and three negative responses were obtained. No response was received from the
-54-
-------
sixteen remaining inquiries. Samples received were both grab and composite types. All
samples received arrived intact ami seemed "fresh". No septic odor was detected.
Eight industrial samples composited (24 hr) by personnel from the City of Springdale,
Arkansas, were obtained. Three industrial samples were obtained from McClelland
Consulting Engineers, Inc., of Fayetteville, Arkansas.
All industrial samples were processed on the Intalox Saddle device. Three hundred
gram samples were employed. Experiment run time was one hour. One milliliter samples
were withdrawn every five minutes. TOC analysis was performed on all wastewater. A
limited number of GC analysis were performed also. Fifty microliter J50 ul) samples were
employed with the TOC and ten microliter samples were employed with the G C. Odor
eminating from the desorption apparatus funnel was noted at various times during the run.
Other details of operation are as reported in the METHODS section. All test results are
reported at standard conditions (i.e., 25°C, ambient pressure, 0% R H ).
Tabie 6 contains the experimental results of the nineteen industrial samples. Table 7
contains colUbcrative published data from which reliable volatile information could be
obtained. Figures 21, 22 and 23 display the results of the desorption experiments for twelve
of the industrial waslewater samples. Also shown in these figures is the computer program
estimated ultimate non-volatile fraction (i.e., Fnv). This extrapolated Ft value is plotted at
Mt/M0 = 0.
Now, once Fnv is obtained, it is possible to compute the relative volatilization rate for
the pseudo single component fraction (see equation 15). Figure 24 shows representatix-e
results for four wastewater experiments. Drawn straight lines are computer generated.
Wastewater data exhibiting a non-linear function are interpreted to typify a mixture of
volatile components. Wastewater data exhibiting a linear function are interpreted to typify
single volatile components or several components with identical relative volatilities.
Gas chromatographic analysis were performed on three industrial wastewater samples.
Figure 25 shows the experimental data for a petroleum refinery wastewater: The GC data
tends to show a larger fraction of volatiles than the TOC data. This type of behavior also
occurred wu!i simulated wastewater No. 4 (see Figure 17). The GC analysis tends to "see"
only the low molecular weight constituents.
A significant odor reduction was noted as the run progressed on those samples which
had a detectable initial odor. No foaming occurred during any experimental run with pure
components or actual wastewater samples.
-55-
-------
TABLE 6
Air-Volatile Fractions in Industrial Wastewaters
1
s
1
Industry Type, Major Product
poultry, liquid egg products
poultry, turkey processing
poultry, chicken processing
poultry, chicken processing
poultry, chicken processing
poultry, chicken processing
metal, hand tool mfg.
oil field blowpit
canning, grape products
pharmaceutical (1)
pharmaceutical (2)
paper(unbleached kraft)
paper, tissue, plywood
food, margarine, shortening
fibers, chemicals and plastics
petroleum, refined pet. products
petrochemical, 1,3 butadiene
petrochemical, 1,2 ai-chloroethane
petrochemical, 1,2 di-chloroethane
paper(bleached) pulp & kraft paper
Wastewater
Flow 1000
ga/da
42.8
960.
360.
622.
547.
360.
76.2
?
335.
?
?
6000.
42000.
200.
12000.
20000.
2000.
43?.
diluted 2:1
22000.
Raw Cone.
mg/l
1545.
2.1
78.
122.
206.
80.
51.
102.
202.
1854.
4100.
142.
136.
236.
452.
110.
94.
92.
60.6
311.
Volatile
Fraction
%
15.6
0.
4.1
98.0
0.
6.6
0.
19.4
0.
49.1
< 32.
26.3/63.5 (GO2
24.5/66.8 (GO
22.8
35.8
58.0/87 (GO
50.6
57.5
58.7
93.5
Relative
Vol. Rate
K/a
5.0
-
-
10.6
-
-
9.3
6.8
2.9
4.4/8.33IGC)
8.5 (GO
5.4
12.9
10.7/20.6 (GO
41.6
25.1
31.0
5.8
Source
J-T1
J-T
J-T
J-T
J-T
J-T
J-T
J-T
J-T
J-T
J-T
J-T
J-T
J-T
J-T
J-T
J-T
J-T
J-T
J-T
Jonci-Thibodaaux this project work.
As par oa» chromatogrspri analysis.
-------
I
Nl
TABLE 7
Published Air-Volatile Fraction Data
Industry Type, Major Product
Wastewater
Flow 1000
ga/da
Raw Cone.4
mg/l
Volatile4
Fraction
Relative
Vol. Rate
K/a
Source
paper, pulp and kra't paper
I Ip, cellulose
unknown, atyrene
synthetic fiber, dacron
petrochemical, vinyl chloride
simulated, six components
domestic, Fayetteville, Ark.
domestic, Springdale, Ark.
1080. 72 -8390(800) 53-94(BOD)3 8-9
? ? > 57(800)
? ? ~ 75(800)
? 85(800)
504. 1200. >8.1
9KGC) 17.0
TOO SAMPLES CONTAMINATED
6000. 254. 35.3 14.7
Estridge(1971)
Thibodeaux (1970)
Hiser (1970)
Keen (Private Comm.)
Minott (1973)
J-T
J-T
J-T
Evap. condensates 74 & 88%, decker filtrate 53%, turpentine decanter underflow 94%.
Total organic carbon unless specified otherwise.
-------
01
00
1.0
8
| 0.8
o
*5
c
01
'6
g 0.6
O
CC
U.
u
§
§ 0.2
0.0
1.0
o -
POULTRY PROCESSING (BROILER). Ks/a 10.6
PAPER (UNBLEACHED KRAFT). K$/a = 4.4
PETROCHEMICAL (1,2 UICHLOROETHANE). K$/a - 25.1
FIBER. CHEMICAL AND PLASTIC. K$/a= 12.9
J I I
.95 .90
WATER FRACTION REMAINING (M,/MO). dimensionlew.
.85
Figure 21. Desorption results for industrial wastewater I.
-------
Ol
<0
1.0
0.8
$
1
S
01
V
i 0.6
o
0.4
O
o
z 0.2
o
e
O
0.0
1.0
- POULTRY (LIQUID EGG PRODUCTS). K,/a * 5.0
-! FOOD (MARGARINE. SHORTENING). K$/a = 5.4
A -PETROLEUM REFINERY, Ks/a = 10.7
O - PETROCHEMICAL (1,3 BUTADIENE). K5/a » 41.6
I I I
.95 .90
WATER FRACTION REMAINING (M,/MO), dimensionless.
Figure 22. Desorption results for industrial wasiewater - II.
.85
-------
Figure 23. Dssorption results for industrial wastewater - III.
0.0
O -OILFIELD eLOW-ftT.Kt/a-0.3
-j- - PHARMACEUTICAL. K,/a 6.8
& - PAPER. TISSUE. PLYWOOD. K,/« - - 0.0
O -PULP,PAPER AND KRAFT PAPER.K,/«-3.1
1.0
.90
.85
H:
WATER FRACTION REMAINING (Mt/MQ), dimensionlets.
-------
O - PETROCHEMICAL (1.3 BUTADIENE). Fnv
D- FIBER. CHEMICAL AND PLASTICS. f°nv
- PHARMACEUTICAL. F
A - POULTRY (LIQUID EGG PRODUCTS). F
.03
- LOG (M,/M0).
Figure 24. Relative volatilization rates of industrial wastewaterr.
-61-
-------
o>
I
TOTAL ORGANIC CARBON ANALYZER
GAS CHROMATOGRAPHIC ANALYSIS
30 40 50
RUN TIME.Imin.).
Figure 25. Petroleum refinery wastewater T 0 C and G C results.
60
-------
SECTION VIII
DISCUSSION
Experiments with pure components established that a batch recirculating packed
column with counter-current air flow can be an effective desorption device. Good
comparison of experimental data with theoretical equations and calculated values instilled a
degree of confidence in the analytical model developed for the desorption apparatus. Table
4 shjws the extent to which the experimental results agree with the aralytical model for the
Rdschig Ring apparatus.
The experimental results of tests run with pure components appear in Table 4. The
data is u-auiated in order of decreasing relative volatility and the experimental K/a values
display thij tr.me decreasing trend. There is general quantitative agreement on K/a results
between the various test methods. The data treatment techniques (i.e., slope from a log-log
plot) is sensitive to er ors >n the test methods and may account for the relative wide range of
K/a values. Figures 6 ,ind 7 show that the least square algorithm will yield biased results if
some data are ir strong deviation from the trend. Ideally all test methods should yield
identical experimental values of K/a for pure components.
Although the Raschig Ring apparatus displayed sizable errors in water loss, which will
in turn effect K/a. there are instances of agreement between test methods worth noting in
Table 4. The acetaldehyde results show good agreement between test methods. The BOD5,
COD and GC results for acetone (run 2) and methanol (run 1) are in fair agreement. All four
test results for methanol (run 2) are in fair agreement. The results of both phenol runs are
exceptionally good. Phenol has approximately the same relative volatilization rate as water
and all six test results indicate this with a high degree of absolute accuracy. Considering the
sensitivity of the log-log correlation to experimental test error it is doubtful that the most
error free data will yield a K/a value with a range of uncertainty less than ±1. K/a unit.
The agreement between repetitive runs for the Raschig Ring column is generally poor.
The GC data for acetone in Table 4 is in fair agreement. The COO data for methanol is in
good agreement. The phenol data is good in an absolute sense.
Statistical results (Table 5} of pure component experiments with the Raschig Ring
column for K/a reveal that no test method emerges as being superior, however the standard
error is usually less for the TOC and COD tests. Non-zero (with 95% confidence) K/a values
were observed for methanol, acetone, acetaldehyde, n-butanol and i-butanol. K/a values not
significantly different from zero were observed for phenol, acetic acid, formic acid and
propionic acid. This is in general agreement with the theoretical K/a column of Table 4. An
exhaustive set of experiments was not carried out with the Intalox Saddle device.
Recommended Test - An Intalox Saddle device was developed (Figure 5) that displayed the
same experimental behavior as the Raschig Ring device. This modified apparatus required
-63-
-------
less sample, a shorter run time, was simpler to operate and gave considerably more accurate
results. Water evaporation in a one hour run averaged 40.4 grams with a range of 38.4 to
42.4 grams for a percentage deviation range of -5.0% to 5.0%. The correspondence of
repetitive runs with n-butanol (shown in Figure 13) is indicative of the reproducibility of
the Intalox Saddle device results.
Experiments with selected pure chemicals revealed some information about desorption
of solution* containing both volatile and non-volatile components. Desorption of simulated
wastewater revealc-o that the non-volatile fraction remains in the aqueous phase and a
volatile fraction i:. stripped out by the air (see Figure 10, 11 and 12). These simu'ated
wastewater runs further suggest that the volatile fraction observed depends upon the test
method. For example "Simulated Wastewater No. 2" (Figure 11) was found to contain a
50% volati e BOD5 fraction.
-------
concentrations to be measured on the total carbon analyzer. In general all the industrial
sample results display a behavior predicted by Equation (14). Some samples contain
volatile-; which display low K/a values and a sixty minute run time is too short for
accurately determining Fv. The poultry sample in Figure 21 and the paper and
pharmaceutical samples in Figure 23 are examples. Figure 24 shows that relatively
consis ant values of Ks/a can be obtained from a uesorption experiment. This figure
rep.esents the results of only the most consistant data results. Plots of this type are possible
only if the wastewater test methods are penormed with extreme care and utmost accuracy.
No operating problems were encounterea with any of the experimental runs. Significant
odor reduction was noted during the experiment.
The Intalox Saddle device as described abo/e appears to be ideally suited to perform
desorption studies on industrial wastewater samples. It has a number of inherent advantages
over sparged vessel type devices and hollow fiber devices (see Appendix C). The foremost
advantage is that it simulates the desorption (or stripping) process as it occurs in cooling
towers. The main advantages of packed column desorption are:
a) the apparatus is simple and the elements are readily available
b) the operation is simple to control and the experiment run time is short
c) standard wastewater test methods ca.^ be employed with the experiment
d) data handling and interpretation require elementary graphical or mathematical
techniques.
e) a correction is available for air with a humidity other than 0% R H
f) the results are somewhat insensitive to operating conditions (except for
temperature)
g) samples that have a tendency to foarr. when contacted with air can be readily
handled.
The main disadvantages of the packed column desorp J?n device are:
a) column temperature must be maintained constant (i.e., at 25°C, cooling tower
inlet temperature, etc.)
b) samples should be kept in a chilled state before and after desorption
c) run times of greater than one hour may be desirable on some samples
The apparatus design and the organic concentration test method are somewhat related.
The TOC (instrumental) test is possibly the best technique for measuring Fv and Ks/a due to
the present accuracy of these devices. If COD or BOD is employed, relatively large samples
will be required from the desorption apparatus. A large flask (- 1 liter) should be employed.
COD and BOD are both useful for measuring Fv and Kj/a. The GC should not be employed
for Fv measurements. Ks/a measurements can possible be estimated from GC results.
One most important use of this device is the simulation of stripping of volatile
components from wastewater in industrial size cooling towers. An experimental value of
-65-
-------
Ks/a is obtained from the Intalox Saddle dcsorption device operated at the proposed inlet
water temperature. This relative volatilization rate is employed with an equation developed
by McAhster. Turner and Estridge (1) to obtain the percentage extent of treatment by
stripping in a cooling tower. The equation is
(Ks/a)aCT(1 +B/L)
R = (Ks/a) aCT + B/L <24)
where R = percentage of treatment by stripping
aCT = fraction of water loss in cooling tower
B - blowdown rate from tower basin, l./min.
L = water flow rate to tower, l./min.
-66-
-------
SECTION «X
REFERENCES
1. McAlister, J.A.. B.C. Turner, and R.B. Estridge. Treatment of Selected Internal Kraft
Mill Wastes in a Cooling Tower. Environmental Protection Agency, Water Pollution
Control Research Series. Publication Number 12040 EEK. August, 1971.
2. Prather, V. Advanced Treatment of Petroleum Refinery Wastewater by Autoxidation.
Journal Water Pollution Control Federation. 42 (4): 596 - 603, April 1970.
3. Mohler, E.F.. H.F. Elkin, and L.R. Kumnick. Experience with Reuse and Bio-oxidation
of Refinery Wastewater in Cooling Tower Systems. Journal Water Pollution Control
Federation. 36 (11): 1380 - 1392, November 1964.
4. Burn;, O.B., Jr. and W.W. Eckenfelder, Jr. Aeration Improvement and the Adaptation
of a Cooling Tower to an Activated Sludge Plant. TAPPI 48 (11):96A(1965). Pilot
Plant Evaluation of Plastic Trickling Filters in Series with Activated Sludge. TAPPI 48
(11:42(1965).
5. Cohn, R.G. and Tonn, FIT., Use of a Cooling Tower in Black Liquor Evaporation.
TAPPI 4Z (3): 163AH964).
6. Engelbrecht. R.S., A.F. Gaudy. Jr. and J.M. Cederstrand. Diffused Air Stripping of
Volatile Waste Components from Petrochemical Wastes. Jo. W.P.C.F.^3. (2): 127-135
Feb., 1961.
7. Gaudy, A.F. Jr., R.S. Engelbrecht, and B.C. Turner. Stripping Kinetics of Volatile
Components of Petrochemical Wastes. Jo. W.P.C.F..33. (4) 382 - 392, April, 1961.
8. Gaudy, A.F., Jr., B.C. Turner and S. Pusztaszeri. Biological Treatment of Volatile
Waste Components. Jo. W.P.C.F..3JL(1): 75 - 93. January, 1963.
9. Eckenfelder. W.W., Jr. Industrial Water Pollution Control. McGraw-Hill Book
Company, New York, 1966, p. 82.
10. Hiser, L.L. Selecting a Wastewater Treatment Process. Chemical Engineering Nov. 30,
1970.p. 79.
-67-
-------
11. Goswan.i, S.R.. Treatment of Strippable and Nonstrippable Substrates by the
Activated Sludge Process. P!i.D. Thesis. Oklahoma State University. 1969.
12. Thibodeaux. L.J. and D.G. Parker. Desorption of Selected Gases and Liquids from
Aerated Basins. (Presented at 76th National Meeting American Institute Chemical
Engineers. Tulsa. Oklahoma. March 10-13. 1974.) Paper No. 30d.
13. Horton, J.H., J.C. Corey and R.M. Wallace. Tritium Losses from Water Exposed to the
Atmosphere. Environmental Science and Technology_5_<4): 388 - 343, April, 1971.
14. Mackay, D. and A.W. Wolkoff. Rate of Evaporation of Low-Solubility Contaminants
from Water Bodies to the Atmosphere. Environmental Science and Technology J7 (7)
611 -614, July, 1973.
15. Weimer, W.C. and G.F. Lee. Method for the Storage of Samples for Dissolved Gas
Analyses. Environmental Science and Technology 5_ (11): 1136 - 1138, November.
1971.
16. Pierotti, G.J.. C.H. Deal and E.L. Derr. Activity Coefficients and Molecular Structure.
Industrial and Engineering Chemistry, jjl (1): 95102, January, 1959.
17. Thibodeaux. L.J.. R.B. Estridge and B.G. Turner. Measurement of the Relative
Volatilization of the Water-Miscible Fractions in an Aqueous Effluent. New York,
American Institute of Chemical Engineers, Symposium Series 124. Volume 68, 1972.
18. Treybal, R.E. Mass-Transfer Operations, 2nd Edition. McGraw-Hill Book Company,
New York. 1968. p. 164-171.
19. Reinhardt. J.A. An Experimental Apparatus for Measurement of Volatile Components
in Aqueous Phases. Thesis (M.S.). University of Arkansas, Fayetteville, 1972.
20. American Public Health Association, Standard Methods for the Examination of Water
and Waste Water, 13th Edition, New York, APHA Inc., 1971.
21. Emory, S.F. The Effect of Surface Active Agents on Interfacial Convection. New York,
American Institute of Chemical Engineers Student Bulletin, p. 37, 1972.
-68-
-------
SECTION X
PUBLICATIONS AND PATENTS
Two pending publications have been written at a result of this study.
1. Jones, Jack Ray; "Air Stripping of Volatile Components .- Industrial
Wastewater" University of Arkansas. Thesis, May. 1974.
2. Thibodeaux, LJ. rnd Jones, J.R.; "A Desorption Test for Quantifying the
Volatile Organic* in Industrial Wastewater" (proposal accepted for presentation at
AlChE Meeting, Salt Lake City. Utah. Aug.. 1974)
No patents have been produced or applied for under this project.
-60-
-------
SECTION XI
GLOSSARY OF SYMBOLS
FREQUENTLY USED ABBREVIATIONS
ac = alternating current
BOD6 = biological oxygen demand (5 i ay, 20"C)
°C = degrees Celsius
cm = centimeters
COD = chemical oxygen demand, mgOj/l.
da = day
Exp = experiment
EXP = exponent (e = 2.71828)
°F = degrees Fahrenheit
ft = feet
ga = gallons
G C = gas chromatograph
Hg = mercury
in = inches
I = liters
log = common logarithm (i.e., base ten)
m = meters
mg = milligrams
mgC/l = milligrams carbon per liter
mgO2/l = milligrams oxygen per liter
min = minutes
ml = milliliters
n = normal
PCBS = polychlorinated biphenyls
pH = log[hydrogen ion concentration in gram ions per liter]
ppm = parts per million
RH = relative humidity
SCF = standard cubic feet (i.e.. ft3 at 32°F. and 1 atmosphere)
STP = standard temperature and pressure (i.e., 0°C and 760 mmHg or 32°F and 1
atmosphere)
TOC = total organic carbon
v = volts
wrt = with respect to
-70-
-------
% = percent
> = greater than
» > = much greater than
- = approximately
~ - approximately equal
H = a definition
<> » infinity
SYMBOLS
Roman
A = integrator area, cm2.
a - fraction water lost in column (dimensionless), regression equation constant
B = blowdown rate from cooling tower, l./min.
b = regression equation slope
C = concentration of gross organic pollutant measures (i.«;., BOD5, COO, and TOO,
mg/l.
CV = coefficient ol variation (standard error -:- mean)
E = experimental percent relative humidity, %
F = organic fraction {0 < F < 1), dimensionless
G = gas flow rate through column, (g-nioles)/{min)(cm2)
H = height of a transfer unit, cm.
K = fraction of a volatile component lost in column, dimensionless
K/a = relative volatilization rate of a volatile component wrt water, dimensionless
L = liquid flow rate through column, (s-moles)/(min)(cm2) or (g moLs)/(min)
M = quantity of liquor, grams. &M = change in quantity of liquor, grams.
m = sample size, grams. Am = individual sample size, grams.
n = number of samples
p = vapor pressure, torr. or mmHg.
R - removal of BOD, COD, TOC in a cooling tower as a fraction or as a percent of it in
feed liquor.
S = stripping factor, dimensionless
SE = standard error of estimate of regression equation
t = time, min. or statistical t-value.
W = weight, grams
X = independent linear regression variable
x = mole fraction in liquid, dimensionless. liquid phase concentration (same dimensions
-------
Y - dependent linear regression variable
y = note fraction in gas, dimensionless
Z = iieiyht of packed section of desorption column, cm.
Greek
« = relative volatility, dimensionless
Y = activity coefficient, dimensionless
<*> = fraction of the original volatile fraction remaining, dimensionless
SUBSCRIPTS
CT = cooling tower
e = experimental
f = final
i = volatile component
ii = inlet concentration of component i
io = outlet concentration of component i
j = non-volatile component
jo = initial concentration
m = total number of non-volatile components or designates a molar quantity (i.e.,
moles)
n = total number of volatile components
nv = non-volatile
o = original or initial
OG = overall gas phase
R = ratio (i.e., stripping factor ratio)
s = a pseudo volatile component
T = tower or column
t = at time t (i.e., one of many sample times)
v = volatile
w = water
x = liquid phase
SUPERSCRIPTS
o = original or pure component
* = in equilibrium
-72-
-------
SECTION XII
APPENDICES
Page
A. Raw Experimental Data 74
B. Computer Program to Calculate F^ and Kj/a fof Industrial Wastewater
Samples 119
C. A Hollow Fiber Device as a Desorption Apparatus 126
-73-
-------
APPENDIX A
RAW EXPERIMENTAL DATA
- 74 -
-------
Experiment 1
component(s) methanol-water
initial concentration approx. 1000 ppm methanol in H_0
initial ci^xrge volume 2876 ml
water vaporized 245 m^
^\
liquid flow rate 700 ral/min (4t340 Ibs/HR-ft )
^ac flow rate 1.82.8 ftV^in AP = 11" Hg
1:1 xet dry bulb tomperaturc of air 77 °?
inlet wet buib temperature of air 68 °P
iitruid temperature 68 °K
?ime into axperiment Analysis Concentrations
COD GC TCC BOD
(min) (ppm HeOH) (ppm MeOH) (ppai MeOH) (ppm MeCH)
0
15
30
45
60
75
90
105
120
135
150
165
130
805
734
718
685
636
615
598
528
496
474
442
421
399
405
550
538
522
492
442
432
423
401
390
365
340
325
i'X)
312
570
424
320
251
251
251
214
181
165
101
67
120
120
67
520
530
493
560
487
477
447
453
333
420
350
400
300
300
-75-
-------
Experiment 2
components . nethanol-water
initial concentration approx 100 ppm methanol in water
initial charge voxune 2913 nil.
water vaporized 245 ml.
liquid 2'iow rate 700 ml/mm. (4,840 lb3/HR-ft2)
(,-as flow rate 1.828 ft^/min.
inlet dry bulb temperature of air 78 F
ii; let wet bulb temperature of air 66 F
liquid temperature 66 F
rj-ne ir. V.Q Experiment Analysis Concentrations
CCD CC TOG BOD
(min) (ppm MeOH) (ppm MeOH) (ppm MeOH) (ppm MeOH)
0
15
30
''.5
75
)0
105
120
135
150
165
130
195
916
826
787
723
686
632
611
630
563
546
499
4V3
429
434
1030
970
925
842
812
820
595
710
630
610
560
512
470
438
4U
395
347
347
288
288
238
248
237
206
197
206
181
160
580
510
440
410
460
340
340
330
280
220
260
220
180
200
-76-
-------
Experiment 3
components
initial concentration
initial charge volume
water vaporized
liquid flow rate
gas flow rate
inlet dry bulb temperature
inlet wet bulb temperature
liquid temperature
Time into Experiment
(minutes)
0
15
30
45
60
75
90
105
120
135
150
165
180
195
acetone-water
approz 1000 ppm acetone in water
2912 ml
174 ml
700 ml/rain (4,840 lbs/Hr-ft2)
1.828 ft
COD
1223
734
470
258
179
111
89
11
14
0
11
0
0
0
75°
63°
63°
Analysis
GC
1000
866
410
281
160
108
68
83
0
0
0
0
0
0
P
P
P
Concentration
TOC
851
515
298
185
106
0
0
0
0
0
C
0
0
0
(ppm ACETONE)
BOD
554
412
256
142
106
62
31
27
10
3
2
2
0
2
-77-
-------
Lbcperiment 4
co;rporier.tD
initial concci.tration
initial charge volurw
water vaporised
liquid flow rate
air flow rate
inlet dry bulb temperature
.nlet wet bulb temperature
liquid temperature
acetone-water
appro* 1000 ppm Acetone in water
3013 m.
268 ml
700 mi/min (4,840 lbs/Hr-ft2)
1.828 ft*
82° F
64° F
64° :-'
Time into bxnenment
Analysis Concentrations (pp:n Acetone)
(min)
0
15
30
45
60
75
90
105
120
135
150
165
130
1^5
(XD
757
433
225
133
79
55
0
14
0
14
12
13
11
11
GC
940
545
300
210
145
85
100
0
0
0
0
0
0
0
TOC
563
245
145
56
24
8
3
3
0
0
0
0
0
0
BOD
353
126
149
85
51
41
34
24
24
7
17
17
7
14
-76-
-------
iixperiment 5
components
initial concentration
initial charge volume
water vaporized
liquid flow rate
air flow rate
inlet dry. bulb temperature
inlet wst bulb temperature
liquid temperature
Phenolwater
approx 1000 ppra Phenol in water
2937 ml
251 ml
700 ml/min
1.575ft3/min. 9-5" Hg
" F
F
83°
67°
67° F
Time into i^cneriment
Analysis Concentrations (ppm Phenol)
(min. )
0
15
30
45
60
75
90
105
120
135
150
165
130
195
COD CC
973
950
949
922
937
969
960
967
979
953
994
975
993
1002
TOG
509
477
490
490
473
465
496
487
487
487
514
487
490
509
BOD
636
636
611
623
598
579
611
579
566
611
579
611
579
578
-79-
-------
£xperiment 6
initial concentration
ii.itiul char^ vola'.io
water vaporizeu
liquid flow rate
air flow rate
ir.let ury bulb temperature
Pher.oi-water
approx 1000 ppm Phenol in water
2355 n>i
241 ral
700 ml/mir.
1.545 ft
80° F
ir.iet wet bulb temperature
liquid temperature
Time into i^xpem.nent
(min )
O
15
30
45
60
75
90
105
120
135
150
165
130
195
Col.
X3
we
940
^34
>32
;J17
?40
941
'934
913
943
944
932
964
67° P
67° V
Analysis Concentrations
GC TOG
336
400
433
409
419
419
426
419
354
386
401
401
400
400
(ppm Phenol)
BCD
648
604
679
686
611
661
686
673
626
635
641
667
617
654
-80-
-------
Experiment 7
components
initial concentration
initial charje voiii,T»
i;ater vaporised
liquiu flow rate
air flow rate
ir.-et ary bulb temperature
inlet wet bulb temperature
liquid te.-.iperature
Acetone-Phenol-V.'ater
approx 500 PPrt Acetone + 5^0 PPm Phenol in water
3027 nl
215 ml
700 nl/nin
1.828 ft
75° P
64° V
64° K
Tine into iixperimer.t
Analysis Concentrations
(nir.)
0
15
30
45
60
75
20
105
120
135
150
165
130
195
COD
^ppin ^xy£cn)
2064
1741
1564
1412
1332
1255
1259
1263
1340
1296
1244
1276
1292
1268
TOC
(ppm C)
378
310
275
250
253
253
250
248
233
232
225
225
200
208
BOD
(ppm Oxygen)
1450
1275
1030
935
955
930
890
840
860
840
865
855
825
845
-81-
-------
Kxperimnt 8
components
ir.itial concentration
initial vclu,Tic charts
water vaporized
liquid flow rate
air flow rate
inlet dry buib temperature
iniet wet bulb tcnperaturc
liquid temperature
Acetone, Phenol, Hethanol, and Water
approx 332 i-pm of each organic co**pon«nt In
0 -., , the water
2-jM ml
211 nl
7OO ml/mm
1.828 ft
74° y
1'inc i-.'.tc K
(nin)
0
'5
30
45
60
75
oo
105
120
135
150
165
130
195
(ppr. oxygen)
2032
1650
1521
1335
1304
1242
1137
1022
1007
960
934
920
91?
923
Anaii'sis Concentrations
TCC
(pnm carbon)
350
670
590
570
545
470
451
435
396
355
375
330
330
305
BCD
(pprn oxygen)
1330
1155
1020
915
900
795
310
795
765
720
680
675
675
705
-82-
-------
licperiroent
components
initial concentration
initial volume charge
water vaporized
liquid flow rate
air flow rate
ir.let dry bulb temperature
inlet wet bulb temperature
liquid temperature
Acetone, Phenol, I-'ethanol, and Water
appro* 333
2970 mi
134 nl
700 ml/min
1.828 ft
73° P
68° y
68° 1'
of each orcar.ic compond in water
ir.to Experiment
Analysis Concentration
0
5
10
15
20
25
30
35
40
45
50
55
CO
65
70
75
30
35
(ppm
COD
i oxyGcn)
1943
1313
1695
1433
1446
M30
1332
1333
1238
1267
1134
1197
1173
1138
1135
11C1
1100
1075
1104
BOD
(ppm oxygen)
1350
1250
1140
1050
360
330
810
870
750
810
750
750
630
690
720
780
660
630
810
-83-
-------
toperime.it 1-18-73
components formic acid - water
initial concentration approx 1000 ppm
initial volume charge 3?QO ml
water vaporized 223 ml
liquid flow rate 610 ml/min
air flow rate 1.65 ft /rain
in lot air humidity lees than 30$
Time into i^xperiment Ana lysis Concentrations
1WP COD TCC BOD
(min) ( F° } (ppm 02) (ppm c) (ppm 02)
0 68 967.7
15 66 652.0
30 66 757.0
45 64 767.0
60 64 . 748.0
-84-
-------
Experiment 2-26-73
components
initial concentration
initial volume charge
water vaporized
liquid flow rate
air flow rate
inlet air humidity
acetic acid - water
approx 1000 ppm
3367 ml
233 ml
570 ml/min
1 .65 ftV»in
less than 30J&
Time into Experiment
Analysis Concentrations
(rain)
0
15
30
45
60
75
90
105
120
135
150
165
180
TtMP
( P° )
77
69.5
68
65-5
64
63
63
63
63
63
63
63
63
COD
(ppm 02)
1223.0
1079.3
1092.4
1076.7
1063.6
1208.1
1120.9
1178.0
1064.4
1131.3
1157.2
1131.3
1120.9
TOC
(ppm C)
950.0
787.5
850.0
837.5
837.5
775.0
787.5
812.5
800.0
800.0
806.3
800.0
800.0
301)
(ppm C
1410
1305
1305
1335
1350
1335
1260
1290
1200
1065
1080
1125
1065
-86-
-------
Experiment 3-3-73
components
initial concentration
initial volaae
water vaponrea
liquid flow ra.lt
air flow rate
inlet air
propionic acid - water
approx 1000 ppra
3723 ml
348 ml
570 mJ/min
1 .65 ft3/min
less than 30$
Tiae into
Analysis Concentrations
(mm)
0
15
30
45
60
yO
120
150
1iiO
TEMP
( 1'° )
75
70
63
67
66
66
66
66
66
COD
(ppn 02)
1149.1
1210.8
1237.9
1313.8
13*5.0
1324.3
1279.7
130y.4
1354.1
TOC
(ppn C)
662.9
713.7
353.2
713.7
745.1
796.5
750.3
765.3
755-4
BOD
(ppm (
960
810
990
1020
810
735
990
975
810
-86-
-------
Experiment 3-23-73
compeonents
initial concentration
initial volume charge
water vaporized
liquid flow rate
air flow rate
inlet air humidity
Time into Experiment
acetaldehyde - water
approx 1000 ppm
3501 ml
260 ml
570 ml/min
1 .65 ft
less than
Analysis Concentrations
(min)
0
3
.6
9
12
15
20
25
30
Temp
(F°)
75
74
72
72
71
70
69
63
63
COD
(ppm 02)
1467.3
1287.4
1139.3
992.2
844.6
721.6
590.4
475.6
377.2
TOC
(ppm C)
599.5
462.0
416.2
397.8
297.0
242.0
150.3
141.2
95-3
BGJ
(ppm <
1050
810
690
630
450
510
630
330
30
-87-
-------
Kxperiment 4-13-73
components
initial concentration
initial volume charge
water vaporized
liquid flow rate
air flow rate
inlet air humidity
Time into Experiment
propar.ol water
approx 1000 ppm
3341 ml
237 ml
540 ml/min
A £.£. C*A. ~> II «
1 .op it /mm
less than 30$
Analysis Concentrations
(min)
0
5
10
15
25
35
45
60
75
yo
105
120
135
150
165
180
' TEMP
( F° )
72
68
68
67
67
66.5
66
66
66
66
66
66
66
66
66
66
COD
(ppm 02)
1822.2
1802.2
1762.1
1722.1
162.0
1533.5
1441.7
1261.5
1121,3
1061.3
JQO.O
920.0
060.0
7CO.O
720.0
720.0
TOG
(ppm C)
663
600
567
617
583
517
51?
467
457
417
400
317
333
317
200
267
BOD
(ppm 0,
1125
1035
1095
1035
975
975
960
900
840
795
705
570
375
330
270
180
-88-
-------
Experiment 4-18-73
compor.cr.tc
in. cial concentration
initial volume charge
water vaporized
liquid flow rate
air flow rate
inlet air humidity
t'ime into Experiment
isopropanol - water
appro* 1COO ppm
3546 ml
211 ml
540 ml/iin
1 .65 ffVniin
less than 30jS
Analysis Concentrations
(sain)
0
5
10
15
25
35
45
60
75
50
105
120
135
150
165
130
TiMP
( I'° )
80
74
72
70
53
63
67
66
66
66
66
66
66
66
66
COD
(ppm U2)
1360.0
15.10.0
1520.0
1380.0
131/4.0
1316.5
1220.0
1142.2
1006.7
1232.4
1175.5
1175.5
10W.7
1030.7
777.4
815.3
TCC
(ppa C)
617
600
550
500
533
500
667
467
450
350
330
300
267
317
250
317
BOD
(ppm (
150
60
150
60
90
90
60
150
90
60
60
90
60
60
90
150
-89-
-------
Experiment 6-1-73
components
initial concentration
initial volume charge
water vaporised
liquid flow rate
air flow rate
inlet air humidity
Time into Kxperitncr.t
(nin)
0
30
60
jo
120
150
130
210
?70
300
330
360
3;o
420
420
450
4PO
510
540
'J70
TKHP
( F°
72
$>
63
63
68
f:&
63
~f-
66
65
66
^ *
66
66
74
68
66
56
66
(5
sucrose water
approx 2000 ppm
3612 ml
542 ml
540 ral/min
1 .65 ft 3/min
lesc than
537
Analysis Concentrations
CCD TOC BOD
(ppm 02) (ppm C) (ppm 02)
-90-
-------
Experiment 7-17-73
industry
initial concentration
initial weight charge
water vaporized
liquid flow rate
air flow rate
ary air
liquid temperature
i'lrie into Experiment
(minutes)
0
5
10
15
20
25
30
35
40
poultry - broiler processing
pure waste
300 gm
41.1 gm
112 ml/rain
0.6 ftV
-------
Experiment 7-18-73
ir.ciu-jtry
initial concentration
initial weight charce
v;ater vaporised
liquid I'low rate
air flov; rate
ary air
temperature
Tine into Experiment
0
60
metal - force and plating
pure waste
300 gin
33.7 em
125 mi/min
0.6 ft3/mn
air bottle
25° c
.alycis Concentrations
TOG
(ppm Carbon)
367
378
-92-
-------
iicperi.-ncnt 7-19-73
uuiustry
initial concentration
initial weight charge
water vaporized
liquid flow rate
air fiov; rale
dry air
^.iquiu temperature
P:rr,e into hxperi:no:'.t
(ninutec)
0
60
feed grape products
pure waste
300 gm
33«5 5*"
122 ml/rain
0.6 ftV-nin
air bottle
25° c
Analysis Concentrations
TOC
(ppm Carbon)
6*5
722
-93-
-------
Experiment 7-23-73
inductry
initial concentration
initial weight charge
water vaporized
liquid flow rate
air flow rate
dry air
liquid temperature
Time into Experiment
(min)
O
5
10
15
20
2?
30
35
40
45
50
55
60
poultry - liquid egg products
pure waste
300 gro
37.2 gm
112 ml/min
0.6 ft3/min
air bottle
25 °C
Analysis Concentrations
TOC
(ppm Carbon)
1545
: 1540
1560
1620
-94-
-------
Experiment 7-31-73
components
initial concentration
initial weight charcc
water vaporized
liquid flow rate
air flow rate
dry aid
liquid tenperature
Time into Llxperi^er.t
(seconds)
0
45
>o
150
210
235
360
-35
530
660
720
730
340
otyrene - water
approx 1000 ppra
300.3 gm
10.6 gin
1?^.5 ml/nin
0.6 ft /min
air bottle
25° c
Analysis Concentrations
TOG
(ppra Carbon)
76
137
111
65
U7
455
41
24
325
-95-
-------
jJxpenner.t 8-1-73
components
initial concentration
ir.itiai weic«t ciuir^je
water vaporized
liquid flow rate
air Mow rate
dry air
~i':ui
-------
Experiment 8-16-73
..components
initial concentrations
initial weight charge
water vsporized
liquid flow rate
air flow rate
dry air
liquid temperature
Tine into rlxperinct'it
(nin)
0
5
10
15
20
25
30
35
40
46
50
55
60
sucrose, phenol, acetone,
acetaldehyde, styrene,
n-butanol, acetic acid,
water
50/tl of each component
303.5 gm
42.3 cm
144 ml/min
0.6 ft/fain
air bottle
25 °c
Analysis Concentrations
GC
(average weight, gm)
0.018
0.013
0.011
0.008
0.009
0.008
0.008
0.009
0.009
0.009
0.011
0.009
0.012
-97-
-------
Experiment 8-22-73
industry
intitai concentration
initial weight charge
water vaporized
liquid flow rate
air flow rate
ary air
iiqu.iu temperature
Time into _Expenreer.t
(minutes)
0
5
10
15
20
25
30
poultry broiler processing
pure waste
300 gm
37.8 m
129 ml/rain
0.6 i
air bottle
25° C
Ar. 'lysis Concentrations
TOC
(ppm Carbon)
122
33
111
84.5
135
72
40
45
50
55
60
60
44
40
47
49
-98-
-------
Experiment 8-23-73
industry
initial concentration
initial weight charge
water vaporized
liquid flow rate
air flow rate
dry air
liquid temperature
poultry - broiler processing
pure waste
300 gm
41.7 gm
153 nil/min
0.6 ft^/min
air bottle
25 °c
Tine into Ilxpcrr.rent
(min)
0
5
10
15
20
25
30
35
40
45
50
55
60
Analysio Concentrations
TOG
(ppm Carbon)
80
62
78
91
93
37
109
82
98
96
-99-
-------
Kxpcriment 3-24-73
co.-.ipcnents
initir.i concentration
initial weight charge
water vaporised
liquid flow rate
air flow rate
dry air
liquid temperature
Tirnc into Experiment
(minutes)
0
3
at.iraor.iun chloride, sodium
. 11 hydroxide, water
301.3 pa
3^.5 6"
123.rni/nin
0.6 ft3/min
air bottle
25° C
Analysis Concentrations
Spectrophotometer
(pprn annnonia)
'J
12
15
20
25
30
35
40
45
50
55
60
380
363
852
325
733
735
710
630
431
45?
-100-
-------
lixperiraent 8-23-73
components ;
'initial concentration
initial weight charge
water vaporized
liquid flow rate
air flow rate
dry air
liquid temperature
Time into Experiment
(minutes)
0
15
33
45
60
75
90
105
120
135
150
165
180
sucrose - water
appro* 2000 ppm
3CD.6 gm
138 gm
138 ml/min
0.6 ft^/min
air bottle
25° c
Analysis Concentrations
TOG
(ppm Carbon)
J210
1310
1360
1460
1520
1640
1640
1730
1790
1910
- 101-
-------
Experiment ->-1073
i;iuoj ury
initial concentration
initial weirrht charge
water vaporizeu
. iquiJ flov; rate
air flow rale
tiry air
ii<[Uiu temperaturo
Time into h'xperincr.t
(minutes)
O
10
?0
paper pulp, unbleached kraft paper
pure waste
300.1 em
33.6 cm
132 nl/nin
0.6 ft
50
55
60
air bottle
25° c
Analysis Concentration
TOC r,c
(ppm Carbon) (average weight, grams)
311 0.027
282 0.021
224 0.021
216 0.020
233 0.019
244 0.017
2-J6 0.016
13,' 0.014
222 0.012
218 0.012
240 0.011
187
153
-102-
-------
experiment ^-11-73
i:iuiu; try
initial concentration
initial weight ct\as r;e
water vaporised
± i.iuiii flow rate
air flow rate
dry air
temperature
i'i:,ic ii! to Kxperi.-.ner.t
(minute;;)
0
5
10
1';
30
35
50
55
paper - unbleached kraft paper
pure wacte
300 pn
40.5 cfit
132 ml/iiiir.
0.6 ft /rain
air bottle
25° c
Analysis Concentrationa
GC
(average area, grama)
0.018
0.018
0.016
0.017
0.016
0.015
0.015
0.020
0.013
0.0V;
0.018
0.015
0.015
- 103-
-------
industry
initial concentration
initial weight charge
water vaporized
liquid flow rate
air flow rate
dry air
liquid temperature
Time into Experiment
(minutes)
0
5
10
15
20
25
30
35
40
45
50
55
60
food shortening & oargerin*
37*3 g» pur* w*at«
300*4 fin
40.2 em
147 ml/mm
0.6 ft^/ain
air bottle
25° C
Analysis Concentration
OC
(average heights, units)
96
85
82
7?
70
68
54
53
47
44
42
40
40
-104-
-------
Lxperimcnt 9-17-73
i nciuc t ry
iniliaj. concentratioi:
ir.itial wei,-ht charge
water vaporized
liquid flow rate
air flow rate
dry air
paper tissue, paper, & plywood.
pure waste
300.2 gin
41 gm
154 ml/.7iin
0.6 ft
air bottle
liquid temperature
Tine ir.to Lxperirr.snt
(minutec)
0
5
10
15
20
25
30
35
40
45
50
55
60
fuC
(ppm Carbon)
173
73
162
1*5
224
156
167
124
3#>
238
v;8
25° c
Analysis Concentrations
CC
(average area, units)
0.020
0.018
0.015
0.014
0.013
0.012
0.012
0.011
0.010
0.010
0.003
0.007
0.00?
-105-
-------
Experiment 10-29-73
i;:du^try
initial concentration
initial weight, charge
water vaporized
liquid flow rate
air flow rate
ury air
liquid temperature
petroleum - refined petroleum producta
pure waste
300 gn
40.1 em
136 nl/njin
0.6 ft
air bottle
25° c
Tine into Experiment
(minutes)
0
5
10
15
20
25
30
35
40
45
50
55
60
TOC
(ppn Carbon)
74
'I)
y6
60
68
-
54
69
Analysis Concentrations
CC
(average height, cm.)
7.63
5.07
4.23
5.77
6.10
4.73
5.67
5.60
3.73
4.50
4.60
5.30
5.20
- 106-
-------
Experiment 12-18-73
indoetry
initial concentration
initial weight charge
water vaporized
liquid flow rate
air flow rate
dry air
liquid temperature
i'ime into .Lxpcrinent
(minutes)
0
5
10
petrochemical - fibers( chemicals, A plastics
pure waste
300 gra
42 gra
: .
143 ml/min
0.6 ftVnin
air bottle
25° c
Analysis Concentrations
TOC
(ppra Carbon)
464
430
413 .
20
25
30
33
<0
'5
50
55
CO
393
380
374
362
364
368
-- 107 -
-------
iixperinent 12-21-73
industry
initial concentrations
initial weight charje
water vaporized
ii'Miic. TJ.CV; rate
air I'.vv; rate
dry air
temperature
I'line into r^cpcr incut,
(ninutcc)
0
s
10
petrochemical - 1,3 butadiene
135 0^ pure ivacte
300 cjm
42 gm
133 mi/in in
0.6 ft /inin
air Lottie
25° c
Ar.al.ysia Concentrations
TCC
(ppm Carbor.)
53
50
40
34
34
33
50
55
60
51
44
33
-108-
-------
i^cperiment 12-27-73
components
initial concentration
initial weight charge
water vaporized
-i^uiu i'low rate
air fiov; rate
cry air
tiquid temperature
Time ir.to K
(minutes)
0
5
10
15
n-butonol water
approx 1000 ppm
300.3 em
40.6 gm
144 ml/min
0.6 ftV^m
dry air
25° c
Analysis Concentrations
I\.C
(ppn Carbon)
302
266
216
25
30
35
40
45
50
55
60
156
120
116
94
-109-
-------
Experiment 1-2-74
i i. due try
initial concentr.ition
initial weight charce
water vaporised
iiquiu flow rate
air l'iow rate
dry air
liquid temperature
Time into Experiment
(.niimtc'j)
0
r,
J
10
petrochemical *,2 dichloroethane
920.7 gm pure waste
300.4 gm
40 gm
1 S6 ml/min
0.6 ft /min
air bottle
25° c
Analyais Concentrations
TOO
(ppn Carbon)
60.6
48.9
33.5
20
30.6
30
35
40
45
50
55
60
30.4
29.7
28.8
30.4
34.1
- 110-
-------
iixper intent 1-16-74
industry
initial concentration
initial weight char£b
water vaporized
liquid flow rate
air flow rate
dry air
liquid temperature
J.TM into
(minutec)
0
5
10
15
20
25
30
35
40
pharmaceutical waste (?)
pure waste
300.7 en
40.7 gm
123 ml/rain
0.6 ft3/min
air bottle
25° C
Analysis Concentrations
TOG
(ppra Carbon)
1854
1846
^760
1680
1586
1564
50
55
60
1.538
1516
1470
- Ill -
-------
Experiment 1-18-74
industry
initial concentration
initial weight charge
water vaporized
liquid flow rate
air flow rate
dry air
liquid temperature
Time into experiment
(minutes)
0
5
10
oilfield - natural gas blow pit waste
pure waste
300 gm
41.5 era
1 30 rai/min
0.6 ft
air bottle
25° c
Ana lysis Concentrations
TOC
(ppm Carbon)
102.4
*5.4
97.2
20
25
30
35
40
45
50
55
60
i/8.6
99.0
101.4
102.0
101.4
100.6
-112-
-------
Experiment 1-20-74
compor.onts
Initial concentration
initial weight charge
water vaporized
liquid flow rate
eir flow rate
dry air
liquid temperature
Tine into ^
(min)
0
5
10
20
30
40
50
55
60
n-butanol water
tpprojc 1000
300.6 gm
46*6 gn
144 nl/min
0.6 t
air bottle
25 cc
An*ly»ie Concentrations
TOO
(cpn Carbon)
388
302
268
224
190
156
130
112
96
-113-
-------
Experiment 1-21-74
components
initial concentration
initial wej^ht charge
water vaporised
liquid fj.ow rate
air 1'low rate
ury air
liquid te.Tipe rat lire
1'ime into Ocperi.~e;.t
(mir.utec)
0
5
10
15
20
25
30
35
40
45
50
55
60
n-butanol - water
approx 1000 ppra
300.3
-------
Hxperinent 1-22-74
concentrations
initial concentration
initial weight charge
water vaporized
liquid flow rate
air flow rate
dry air
liquid temperature
Tine into Experiment
(min)
0
5
10
20
30
40
50
55
60
n butanol water
apprcx 100O ppm
300.3 gm
54.1 em
158 ml/min
0.6 ft
air bottle
25 °c
Analysis Concentrations
TOC
(ppm Carbon)
396
353
328
284
258
210
178
144
108
-115-
-------
tixpericcnt 1-25-74
component:;
initial concentration
initial weight charge
i/ater vaporised
.Liquid flow rate
air flow rate
dry air
liquid temperature
Time into IX
(nin)
0
5
10
20
30
40
50
55
60
n-butanol - acetyl alcohol
- water
approx 1000 ppm (n-butanol)
50 ppm (acetyl alcohol)
300.3 gm
40.6 c"i
154 ml/min
0.6 ft
air bot»le
25 °c
AnaXvais Concentrations
TOO
(ppm Carbon)
342
370
338
266
250
216
183
150
150
- 116-
-------
iicpcriment 1-28-74
components
initial concentration
initial weight charge
water vaporised
liquid f.'.ow rate
air flow rate
dry air
liquid temperature
Tine into rlxperiner.t
(min)
0
5-5
10
20
30
40
50
55
60
n-butanol - water
- sodium tetradecyl Bulfate
approx 100U ppm (n-C.HQOH)
50 ppm (NaTIE) 4 *
300.7 en
40.1 gn
156 ml/nin
0.6 ft
air bottle
25 °c
Analysis Concentrations
TOC
(ppn Carbon)
398
366
352
300
254
220
180
172
156
- 117-
-------
Experiment 1-29-74
components
initial concentration
initial weight charge
water vaporized
liquid flow rate
air flow rate
dry air
liquid tenperature
into Experiment
(min)
0
5
10
20
30
40
50
55
60
n butanol water
sodium phosphate
- dibasic htpt a pi
appro* 1000 p^ra
10/*1 (sodiun phos
(DB
te),
300.3 gm
40.1 gm
154 ml/min
0.6 tt
air to i tie
25 °c
AnaXysis Concentrations
TOG
(ppm Carbon)
374
360
338
288
236
222
168
166
146
-118-
-------
APPENDIX B
COMPUTER PROGRAM TO CALCULATE F° AND K$/a
FOR INDUSTRIAL WASTEWATER SAMPLES
Reproduced from
best available copy. \3Sjp
-119-
-------
r *
T t \i~
r <~ ' P" r r <; THf-~ l'| Tj'i
f--,' t M >'."-. A .| » r p-> «r * T,-v( T . " . , f "* Tfvr ptc»-||iy) « *> Tl'~ c
ATIVr V 11 AT »|. 1 7 «T' TO "AT= | <»- ».T \( .-) '.
PUT ;y T.|cr) rt-r-i ncr ,->-.«OT j ~v sync? » vcs T J|TL< ^ »: «!>!' "''
!?'rIJ'\T;M M rri '«f<"ci^r v^<; T r«-; *s TCO ,
r = T;>?- n|_ T ' ' T r f ' .*" . . "3» TMII-J) » V|-U;t| r 1 1 r r?^r T t '
T-.|r. \fi|4Tf|C £: f, r r i*?i vr.y nr. roT»j\;cr -»/ ^t | JT " > r » T r
p r-jrki II-ITTV, T!-^ WI'^IL1" Fc-*r.T»f'» ' "i TM'- 1 1| 1 t -. ' c
O'I\*ITITV -"I1" Ti.r '">r,-.-1>|r j-r» ]M~or;\»n«- .-/- .«_.^ T j 7- tr « T c t-i
C = ro;r> T- r^ir T '. v CT'/i
L-
A
.IT 1 «; 11 -;° Tf -i --e. trr r .1
| «n/oi »*«" I-.-.HT ,jr T in'-. T|.i
fx/rTr-T ei. M'l*"". T r> i >- m t -jr pti_
'»»^ !'!;, 'l-'l » Mf, T".jr-j«;f r T " . )
7 K «i o.».Tr. Tijt^ |*,M.'' -c.
fS fW" '. * I ' '- T -«" P'lT- -T»-" '"!F^ronT| ,-:, :ir t-L.-r , o r > * I
TV T'! T .>.:!' rr voi ii CMT T^» TI.C O'.TC -ip r /.>''" *.T » r ir
u.^Tfo^ r.ji; ;» » r "^ Tr '> t 5 fv
ir K / « i -_ r>r«Tf.% TJ*.. .,MiTv
: iv^i'ii T:. r I\T .jE'iT -'. JPu - T--» P.--;; t>T\- >i/f».--tr » »- ^ >.' -| *"»i -
" fl.l.HT \' T -, -'i^o<;:': -'in . f ' *v IT t !"» r »T, ) -iy f- '. t " -' / '.
i^ rr.-iM TI n»-tTV ' lr«<: **-*.»: :l%llTy '-4'-»i />.!«> n .> F > o t >.;
''|'-.; ' <;> | ^ ^ -nr p."fTi
t -i A--V inc i- A T T v.i TU-T
vt-n -.70 M.r.. '"r-
f«-rr-j «; T.;-I ? i ' / T' -?
.
vT f : i: : | '.VT f "!? c. "T f *"; r
:!: '; ..... : : ft « 'r » ' * f 1^
i \'fi / T «
» r-tr r>-«rr c r , ' i Tic T I -' IT ^ Tl r- » * - .» > Y "'
r
r
r
r
, .. t . i ,
r f "..^T -c -i ( -" i :*, rr ';i~ f « >«» ( cr 5-«.\T : I ^ . /« I . "I' r. ' t " |
li^r ^r^.^tv/, *-i| * T ! f. T j c f i_<; ^M'* r>.-rric..^ r>i: -"liTntiT frxll|
t r i" T ''Ti r.o"; »<-: tf<"--\T <"f5f* i-'.'j T1-" v j**.'t r. > 'T <'. TA
"-!<; in cr.nri. ( ^rs 4» r : f i ) , TM= cas-.'c. -._ ...f = -> v^»''»" T 7«-f>
r:|Tfvr, rue ryntn |vr.,r (riCMJT. riT.M, Tut J.Mf'tM <-.'-ft"s
re :'\rct rn^rrr, T" TMC r»p cnppr j r -\f *>» V ^Ti|<; |r~,r>-.«iT; C 1
«» t -\.f oi|^ T|MC T»I 'ifiMT.--^ fctr>4»r« c\
riitT *\r» cniirapir; c^en r IMT,\ |-- TU=
:rt. T t M= T*T^ IHTitVr" «=*"« ^»»ii|-.c
r ^YOCI? tMrf. t , THr TT.»t \iifjut > -v
r.- fv TI.IC
CA.ru y»"-
TMC
PI
,vf
-120-
-------
t t .
/. i 7 r » 7 ' r (", i
r, >r> ? T - r .-.,.:-, u ^ «" i i ». i - i . '«»)
f, . s = ": \r t i»i , i -.'.»
7 -- ''< S.n i 1 i, «.> ' .--r . -JT
'r.^fi \~; t.--t '*'> ": *« i rrr*1. 4\pi I.;AT-^
J : » "> | i T < " » I T J i r; f c i»c y , r ~r; r I
M Tt
1 1
ve; I Tc»r . A7 )
rno-«»7(/jv«
fin <:/. t .-. 1 . !-
-1 »« t I T f C f '
Tu- > ^.M
, /\1 ro f t -r > [ -->-( T1 --( 1 ) /:-T J
v - » T r. ( .-, . .5 > \ r ;<- % I I , » T : f> f » ) . ' T "«? ( I »
rr.rt-i AT i «ri » . » )
'o t '-.: f /,,/. r ,
n.vn.-'i / i , t ii- -.> \rr, M(-"-v.?l. . I
.- !! IT i ' .!.> '!; i -if r.« <\r T j r v J'"M_>
If
I'*
'2
'V! : » ' -1 ,' "
; T ( i | - j.'. r.-" » t )/..«' r -<( i )
,-;,.. r ( . l , .. , ,- r ) ) ( r r I I 1
" ( I ) -. I " -T ( I ) /1' < ' t 1 » » {' - r I ' ) )
'?'-,,!'> T '"r I ' I . r M t
i I r t
r ." r .«: r > = 1 I.'.
=1
O'<.'. v -rr ( ,> l
O-.« I'-lr.l .T
r."| rMIOe. #,1 J
0 = <". M i 'ii i n -.\ r < A <
wp t T £ ( /, , * ; t
r ( / l y f> if r . nn«--\tY «',<,
r nu
1ft
37
40
nfi i- P 'T.', M ?
X ( t t -T r.iur l t )
CI ?) = (=» H-.T
Y{ T»=M f.-.i^oc: I ( T
S( rn^.-.c| no^ +x M |
y f I )
- c f £
T »
-121-
-------
V v">
T «; T f. "T/i ;*
f>r, »>-. 7 -> ,:!
s '-, I ; T r ; - > : ' "> i ( > f ' > - vi - * -. * y !'))' -I '' I T >-'.'/ c" '. '.= X I . I
» 7 '"'?:«» t >/[»-.» l l >*. r
rn v ; <. ' w , / \ \.- ..
>'>! 'i ;' ; -> ) 1 .->./
vo j *((.. r i i o . <^M -,'.(. ».\' ::< ^
71 r T.-M ' » '. -'I .: . "I J
." : .: .'-. n rt -A 5; * A - ;- *; ': T %-.-':.; -S : n V f. -.*::. *r :- -*l *: i -'» " ^ * * -r *- .*'- t »^ T ( .V *: "i t\ ~,
r i'i>i -; r ,-r r ''. , ; > t '-» ^ I n''ftl <;L' ' ' r M PT' TM^ < it r t < ' :
-,.<> _->' r T; f>, r MJ I |
r r , . ? »i r '"»»/'. | " ^ " \ >'''' T . J '.-T V ( ^ '! T' lr t O T '«i . r v r r - c
' » . 7.11' r -| - :. - r., i .- i t ,< r u ] PM T«^ v i "( -\ r t :] \i <' r '.< \t
r
*~ r. * t: & * -^ *-*-;-:' r * -r A t .-. \ t ': * * i- : -^ ^ <- * ' A **!* ft i^r t- :.- ^r s £ k r- ': - : r- * ^r .-^ r- -1- ^: ': ft * ,\ ^ 1
r» ') .* ~ f- * 1
r i r P r', - ' i) -, -. .'.:.-'>
-,7 ',".. '- l r -/.-,, T 5 )
c; -! T f» ::(!(/)/! <- ; i ; f I : > f- | f | - f. I '~ " A " 1 "<.! r J I J
c, -, -s/, ,' t ' ' i , :" ) < i t , , y ( T i
.; /. T ;. .; .1 v f I * ,!'.'.. I V , i" > ' . '. »
; ' /.. > t r f - I I >, 1 i . 1 l
S ' I ' ' r ' 4 I
/, -, " I - " 'I I'"
-, /, 1 ! :>J -. - l !
f, >: i > - i
» / »-'*'>...
A 7 ! - I 1 > , I f . I
f, 'l ) 7 f'1 I'M*-''1'')/ « ''.!>
f*» r^T^"")
7't M: I -!-,»», ) .;.
7;. 1 -i f, < ! ."i '.,>
71 ^. 'V'/- / - 1 >
7'i f.-, 1" '.'>
~> f, J »« » '"
77 <-; '- ( r ' v-rvi
.-i i
-122-
-------
r
r
r
r
r
r
r
O tr civ1 f ,>!«.'.
r.n T 'i
T ( '. V -C^ ! -r! )
fuc fin* r>i -T rr:-""T'T ."c r<-ir> r'r -3 "?.»»= N'fTf >«.<;. TUT
«ITI p ir .'. * tr \i <;y«"fi r. R=> 1? TT ;T ~.tr. T-,'* r»r T|,T o»
rcrj^ TUT. . l''.iC.,-> ; f | "I!TP||T ffT'i.
t «>_i i. v ii t r I «: I. T .\| r\ t T « . *uc - » i c T n _«~r " (*c "i( ro- 1 r
?< ^MC .: « . '. y f>C " 1 "" *'T A | !"'. TA '.C " - » "> *'"» T'-?^ P I " ^ T min"
!<; TL. - r^'-r^'Ti'^'t'-ni^ rut <^r/-p«n T<; r-|r ; >.> c; r~f j* rr }
>t « ^ i i t ' [ ( fi i « 'uoc r j c T> ^ 'HI' T i ..- | -i t nnT v <: .
'' »|'.v_ <. i T ' I " r r .'r"T \ T j (» . T t ." '. c .-»» (i oj rr v ic --i :T >! i -
| .- j!.,-.- r '- s r T "" rr ' rf-'-»/f|_ < T ii c j--^:ir TM « j »ii M<^
i [ *_! - i* ' ' ' r . ' ' ~ " I " r T Tin"' | r, ->'! r t ,". ; » t > IT rr
V n "i-.c r r -...- /<(;.<> ic. TMC n; j cc,ij, (»?,': »!-,: -\mi A r T i r
i:>'-.«i». i r r-f.rT''"f ^ ;;.? !«f. l«i 'M" ^ OD ' <\ Tin; ^ T-J ; ^ r^i^r-f
) ; .» i i 1 "i r » T T f *". 'Me «-f >;_>» i-.j »T f | r r-r c U|_T r. , ~ rn\'}r\ -> i ->r v - -
i 1 1 T fi i T J c t 1 1 1 r r < 1 1 ( t r. r .- T u c f- ( r- _ T t v - « >c J r1 r t * L c- r ' ' r u .
,--,- r . T . : \. r n. r i,( i i « i c »,(-M_,(-.) / r i i ;-. {::, «.r T ! "" . r <~| (»
T .' i ,! i> . Y . . v i 11'. 11 > T f r'| r > . '"'M '" T '-"' r r t s T ><
r-«rt.-frt~"*_rr_.''"'»T'"V ° C " / \. rrl ''*' 5- ''I I"' 1 c r"r
*/ «/ i > -. r^t !:« r»ir- '-;n r n|.- \ , -u.- |/T:>->T^i.r
T i r D i' i t ^ - i » \- ' . i 'i c ) r T 11 ;i < I i f. 11 ~ >/ 11 » T 11 .- r r . * T
r ;;" fir*''' *'^|i(** *r- TM~ ^."**'^% l^'* r>i ^"iifT ?«*!/ I'll'-i'*
.- / -.151 -_ 'ii i r ^ I r T| '.: 17 \ 7 * O| T c / c
;. r> ' ' 11 r r ' .'. I ''{ r '' r I. r » "> I ' T I I r "'' ' ~ *
'""'rttf'"" '111 ' '' " 'Trrc. »-)y\t TC fMi P *. T ^ I
T'l'M/':"1 i ' ' Tn"|- TM" ;'i r (»i T».f«; 'jri3 M i y I'-'ft '~c TH» T
TMf-
-123-
-------
>r, r /i
t, t ? , n r
«?». oo
'M.OQ
».<; 7. no
?7"..'-c
>7'».oo
>/!. ^0
'' 'i 'i . r» !":
>./<... in
' 1 f.'
P.fllfV.
« . f -TV.
l<-..Crr.->
VT.'^rr
'. 0. ^PO'"
so.r.r-.^
* s . f o r r
f P . < 0 n r
'.(i ; « n >!-. » i » i ; r
^-^('J.^O ^*f-l
7 «./,.«;- r;.OT
?C ».(;) | "i.r"'
>n-,.vi ->n.r.-.
.77". I". T> .«. T
?7'.''l '.(..:'
? 6 »."'! '»"..'«.
?A l.'.r: c'. ..- -~
? ' ° . " " ' " . ' ' "
r& ^ r T . (»»- v,ii .
l.^r^-*
i.'i.-i /./.(.
"> , CO 36
n . 7 / <-. »
o. ",;->
n.70'*
i>. 7iT>'>
n..'-^/-'.
,- r t C'' , nil" «V ' t. y , " »» . /
r . A 1 I'l
r: . 7 f /
0 u '" ''
-,} f r r:
1 ) . ' '- /.
' . 1 <- r-
"* . ' ! '
:: . i- ' "'
\,i '.
<\ . ( > <-'
t.'-'. l
.»./'?.
.').'''*
n ./,/.*
«1./:'. "
n . /. /. ;i
0 . ' ' ?
O.#,A-J.
) . ^, /, <
o.r/.^
Q.^ * ~
O.-S''--'
T ./'''
O.^'V
n . ^, ^ x
n.*r
-o. »!«;""
ic^.o--, O.r y;
J«o. ;'.-./. -).PV ".-'
J OO , n <" /. ">.">)'. " . '
I'"'.' r » O..''.' r . ' '
1 '"i. ' '. >.''',' 1 ".''
1 '.H. :' ? 1 . - V I/..'."-'
1 "..« 7 1 . - 1 '. '»'.»'.
i -J.T . i.r i > i '.-/.
)-.'<' ). ' " 1 .''.
| '',.' !. 1 ».,-!' 1".' "I
I'"'.-''- I.'-*- 1 ". ' '
1' ' .' - "). '. ^ i v.
i i r r » ' »^ g ** " \ ? ', - ,
»';.''' i. ^ ! '.-'--
irrj^-., >.-.-' 1 '.'".
j'-o.rtp ,., ).- 1,7.. ./.i
ic'i.cr, -, !..->: I'.""./.
ico.c''.ni"1 1 >.-'.-.
jo»i.«-r _ I.I T 1 -».''.'
j o o , r, r T o . ;^ T) 1 ,> , < /. /.
I "' , ' r " .^ . i" '! ' 1 ',''..'
i'-' '.< < => oloir r?.* <'
j oo. «:» rt .n't' i ?.-'.*
| c o t r f a ' .I.'"'? '>.''.'
J c o . r: r .t o . .0 ) "> \ 1 , ' ', ">
]oc,.r<.. j >.'>>? 1 .'.''. f
jcn.cro 1.CT» I.'.''.'. 7
ri.ir-.f.-nr j-rev,.-,
-0) -P.lSS'.r. -0
-C! ~."/. ."". <~r TO
-01 -0.'-l/:lc CC
-124-
-------
r../.?«;-,«:_f)l --,.-; ^^c. 10
». S «f"' = _n? _.-« >-r. »c ,-r
-125-
-------
APPENDIX C
Hollow Fiber Device As A Desorption Apparatus
Preliminary investigative tests were performed on a device that may be an alternative to
the packed column desorption apparatus. A Dow Hollow fiber Gas Permeator was
purchased. The device consists of a U-shaped bundle of hollow silicone (dimethyl silicone)
rubber copolymer fibers in a 100ml. poly-methyl-pentene beaker (Fig. C1). This device is
designed for bubble free transfer of a wide variety of common gases in gas-liquid transfer
operations. Common applications include: O2 enrichment from an air stream, oxygenation
or carbonation of fermentation broth, and removal of waste gases (i.e., CO2) from sealed
environments.
The gas permeator has several characteristics that makes it attractive for desorbing
gases from wastewater samples. The device is compact and can be weighed on laboratory
scales to obtain water loss information. A simple apparatus is possible since o pump is not
necessary for the liquid. A stirrer-bar keeps the liquid sample well mixed at all times. Figure
C2 shows the equipment arrangement employed to test the gas permeator as a gas
desorption apparatus for wastewater samples.
The operating procedure employed was as follows:
1) The test solution was prepared with an approximate concentration of 1000 to 1500
parts per million ot the desired volatile component in distilled water.
2) The permeator was filled with test solution (-75 ml.)
3) The permeator, with magnetic stir-bar, was weighed.
4) The permeator was set in place, ano inlet and outlet air lines were connected.
5) The magnetic stirrer was activated to a reading of four to five.
6) The air was turned on and regulated to a pressure of ten to twelve psig. The rotarreter
was set at a constant rate of 80 on a scale yf 100.
7) One mitliliter samples were taken initially, after 15 minutes, 30 minutes, one hour, and
at appropriate intervals thereafter (usually one or one and a half hours). ''
8) Finally, the permeator and contents wore weighed again to find the water loss.
Three components were employed in the preliminary test on the gas permeator. Table
C-1 shows the total organic carbon results for the three components tested. It is apparent
from these results that this device may have limited use on gas desorption studies of
wastewater. Experimental results on these three volatile components with the 46 cm. X 5
cm. raschig ring packed column allow a comparison of desorption rites between the gas
permeator and a packed column (Table C-2'i.
It is also apparent from this data that the tube wall (40 micron) may be providing an
added resistance to the transfer ot me volatile cpecies from the liquid to the gas phase. High
molecular weight components may not be able to permeate the tube wall or may react or he
-126-
-------
Polyurethane 1 ube Sheet
Recirculation Port (Open to Beaker)
Silicons Rubber Copolymer
Hollow Fiber Bundle
Nominal Diameters
180x260 Microns
Nominal Wdi Thickness
40 Microns
Nominal Area500 cm1
Beaker Container
(Poly Methyl Pentene)
Nominal Volume - 100 ml
Protective Screen
Magnetic Stir-bar Chamber
Recommended Stir-bar
I'/z" Cylindrical
(Stir-bar not included)
Gasket
Bottom Cap
Figure C 1. The Dow hollow fiber beaker gas permeator b/HFG-1.
-127-
-------
PRESSURE PRESSURE GAUGE
REGULATOR
I , , 1
AIR SOURCE
DRYING
FILTER
*- ROTAMETER
» TO HOOD
A A
SAMPLE PORT
__ HOLLOW FIBER
APPARATUS
MAGNETIC STIR-BAR
MAGNETIC STIRRER
Figure C2. Schematic diagram of equipment arrangement for gas permeator desorption experiment.
-------
TABLE C-1
Test Results Volatile Component Desorption in Gas Permeator
Component
Acetaldehyde
-
Methanol
Propanol
"
Time (hrs)
0.0
0.25
0.5
1.0
2.0
3.0
4.0
0.0
0.25
0.5
1.0
2.0
3.0
4.0
5.0
6.5
8.0
0.0
0.25
0.5
1.0
2.0
3.0
4.0
5.0
6.5
8.0
Per Cent Stripped Water Evaporated (g)
0.0
47.1
535
57.8
60.7
85.1
100.0
0.0
11.8
11.8
125
19.0
22.9
26.9
32.1
37.4
46.0 19.1
rate * 2.39 cc/hr
0.0
7.05
7.8
155
26.1
35.0
45.6
59.4
64.7 95
rate- 1.15 cc/hr
TABLE C-2
Relative Dteorptton Rates in 1-Hour Run Tim*
Component
acetaldehyde
n-propanol
methanol
Gn Permeetor
575%
7.8%
12.5%
Packed Column
73.5%
155%
21.0%
-129-
-------
absorbed by the silicone rubber copolymcr. Another disadvantage is that the tube device
does not simulate industrial stripping operations, as dees a desorption apparatus for
wastewater samples, however further detailed testing is indicated.
-130-
-------
------- |