L I B HAHY OF THE U NIVERSITY OF ILLINOIS lJL65c t! iGJWttRINC The person charging this material is re- sponsible for its return on or before the Latest Date stamped below. Theft, mutilation, and underlining of books are reasons for disciplinary action and may result in dismissal from the University. University of Illinois Library SEP 6 1968 oc p A ' ! L161— O-1096 Digitized by the Internet Archive in 2013 http://archive.org/details/syntheticdeterge08ewin jpjjjajjtiirtG UBRAW CIVIL ENGINEERING STUDIES SANITARY ENGINEERING SERIES NO. 8 £28 n«8 ENGINEERING LIBRARY^ UNIVERSITY OF ILLINOIS URBANA, ILLINOIS C/?JII?l?r 'CEBDOM SYNTHETIC DETERGENTS IN SOILS AND GROUND WATERS RU OF THE K\< -.** By BEN B. EWING LOUIS W. LEFKE and SHANKHA K. BANERJI PROGRESS REPORT FOR THE PERIOD SEPTEMBER 1, 1959 TO NOVEMBER 30, 1960 for the NATIONAL INSTITUTES OF HEALTH U. S. PUBLIC HEALTH SERVICE RESEARCH PROJECT RG-6560 DEPARTMENT OF CIVIL ENGINEERING UNIVERSITY OF ILLINOIS URBANA, ILLINOIS First Progress Report for Research Project RG-656O SYNTHETIC DETERGENTS IN SOILS AND GROUND WATERS Principal Investigator - Ben B. Ewing, Associate Professor of Sanitary Engineering Co-principal Investigator - R. S. Engelbrecht, Professor of Sanitary Engineering Department of Civil Engineering The University of Illinois Urbana, I 1 1 i noi s for the period from September 1, 1959 to November 30, I960 January, 1961 SUMMARY The purpose of this research was to determine what mechanisms may be important in retarding the movement of alkyl benzene sulfonate in ground waters. A satisfactory method of studying retention of ABS on soils and biological slimes has been developed by modification of existing procedures using a radioisotope of sulfur. At a concentration of 50 mg/1 , Ottawa sand in a column was found to re- tain 3.30 jig of ABS per gram of sand. The relative velocity of the ABS front with respect to the water front under these conditions would be 0.77- When a biologica slime was developed on the sand in a similar column, seven times as much ABS was retained on the solid phase. Under these conditions, the relative velocity of the ABS front would be 0.31. It is concluded, therefore, that the presence of a bio- logical slime grown on sewage would retard the movement of ABS through that zone of soil containing the slime. Beyond the first few feet of soil, where slime growth would be negligible, retention of the ABS would be due principally to physical adsorption on the soil and this effect would not be great for coarse clean sand. Table of Contents Page I Introduction Syndet Pollution of Ground Water 1 Underground Movement of Syndets 2 Mechanism of Retardation 3 Objectives of Present Study 6 II Analytical Techniques Methods Avai lable 7 Reasons for Selection 8 Modifications in Radioassay Procedure 8 III Preparation of Laboratory Columns Physical Description ]k Medi urn Used ]k Preparation ]k IV Physical Adsorption Batch Studies 21 Column Studies 23 V Biological Aspects Purpose of Experiment 26 Column B 26 Seeding of Column A 26 Standard Operation 27 Performance of Column 27 ABS Adsorption 28 Evaluation of Slime 3 J VI Conclusions 33 Appendix References 35 Radiochemical Determination of ABS 37 Publications, Staff, and Foreign Travel 39 Acknowledgements kO SYNTHETIC DETERGENTS IN SOILS AND GROUND WATERS Ben B. Ewing, Louis W. Lefke, and Shankha K. Banerj i I INTRODUCTION Syndet Pollution of Ground Water Synthetic detergents were developed as early as 1932, but they did not (0 become generally available until 19^8 . Since that time their use has in- creased greatly for both domestic and industrial purposes. During 1956, they (2) represented nearly 70 per cent of the market . Today approximately 90 per cent of the sales for home consumption are attributed to detergents of the syn- (3) thetic type . A recent study in Suffolk County, New York, indicated that each householder now uses and discharges to the environment about 100 pounds of syn- (4) thetic detergents annually . The first reports of syndets in waste water having any detrimental effects came from sewage treatment plants, where they were suspected of causing foaming in aeration tanks, outfall sewers and receiving streams. Later they were reported to affect surface waters and to interfere with some water treatment processes and to cause consumer complaints of foaming and an off-taste of (5) dri nki ng water . Although their use has been general in the United States since 19^8 and synthetic detergents have been discharged to the aqueous environment in in- creasing amounts since that time, there were no published reports of major ground (4) water contamination until 1958 '. The first reported incident involved housing developments on Long Island, N. Y. , in which each lot was provided a shallow well for water supply and a cesspool for sewage disposal. Since then there have been reports of similar ground water pollution in metropolitan fringe areas in 13 (6) states . Among these housing subdivisions, the most serious problems have apparently been encountered in the Mi nneapol i s-St . Paul area , Suffolk County, (4) (7) Long Island ' and Portsmouth, Rhode Island . There have also been incidents of pollution of ground water from polluted streams, sewage oxidation ponds, and (6) holding ponds for industrial waste. Walton tabulated instances of this type occurring in 9 states. It is obvious that while the adverse effect synthetic detergents have on ground waters was not recognized as early as in the case of surface waters and waste waters, the problem is real. What makes it even more serious is that ground water pollution is of long duration even though the source of pollution is removed. Synthetic detergents being marketed today contain a variety of ingre- dients including surface active agents, phosphate builders and miscellaneous (3) builders . The surface active agents most commonly used are anionic compounds and are usually of the class of compounds called alkyl benzene sulfonate (ABS) . These surfactants cause foaming in water in concentrations greater than 0.5 mg/i . Flynn, et al , ' showed that "off-taste" complaints are encountered when water contains more than 1.5 mg/,0 of surfactant (as ABS) , although it is possible that the taste is caused instead by the builder compounds associated with this amount of ABS in the packaged product. Toxicity studies on humans and other animals indicate the toxicity of ABS is so low as to cause no effect in man at concentrations likely to be encountered in drinking water . While the (9) present Public Health Service Drinking Water Standards impose no limit on the concentration of ABS in drinking water, it has been proposed that the standards be revised to incorporate a recommended limit of ABS concentration of 0.5 mg/jj as ABS for drinking water . This recommendation is based on the aesthetic effect. Underground Movement of Syndets A review of the technical literature indicates that the distance traveled by ABS in ground water, while generally short, may in some instances be fairly great, Flynn, et al , ' reported no ABS found in wells more than 65 feet from the source of pollution. They also showed less probability of finding syndets in water from wells as the depth of the well increased. The ground water in the area travels at 1 to 3 feet per day. Chromium waste from a plating plant which had been in operation for 2 1/2 years were found to have traveled about 1200 feet, which corresponds to 1.3 feet per day. Inasmuch as the detergents which must have been used in increasing amounts over the past 10 years had only traveled 65 feet, it appears there is some mechanism by which the movement of ABS is retarded. Deluty reported the maximum distance from the source of pollution to any of the polluted wells in the Portsmouth, R. I. area was 150 feet and Sk per cent of the wells were within 100 feet of the source of pollution. A recent study of pollution of ground water on Long Island by laun- derette waste showed that at Mastic, New York, ABS has traveled an overall distance of 1100 feet downstream from a launderette. If the ground water traveled 1 to 3 feet per day, this distance would be traversed in 1 to 3 years. The launderette had been in operation for 12 years, however. (G) Walton showed that in almost every instance wells contaminated with ABS by household sewage disposal systems were less than 100 feet from the source of pollution. On the other hand he cited reports that in five instances, ABS had been found more than 1000 feet from the source of pollution where the source was a municipal or industrial waste pond or recharge pit. The greatest distance reported was at a sewage oxidation pond at Kearny, Kansas where a well 4000 feet away contained ABS. At Peoria, 111., ABS has been found as far as 1800 feet from a recharge pit utilizing Illinois River water. The river water was found to contain 0.7 mg/j> ABS. A well 200 feet away contained 0.7 mg/j> ABS and the well 1800 feet away contained 0.17 mg/i ABS. It is concluded that the distance traveled by ABS depends upon many factors, one of which appears to be the amount of ABS introduced into the soil at the source. Greater area would be contaminated by municipal and industrial waste disposal facilities than by household facilities. This fact, together with the comparison between distance traveled by the ground water and distance traveled by the ABS on Long Island, lead to the conclusion that the porous medium through which ground water flows has some capacity to retain ABS and retard the movement of ABS in the water phase. Mechanism of Retardation Movement of a pollute through a ground-water formation is a displace- ment process. The vehicle of transportation of synthetic detergents, or any other pollute of waste origin, is waste-contaminated ground water. The pollute moves into a zone of earth and, until the zone has retained an amount of pol- lute equal to its capacity, the pollute is retained. When the zone has re- tained its capacity, pollute in the pore fluid is displaced by continued flow of waste-water into a new zone. There is always, then, a fringe of earth material which is being loaded with the pollute. The pollute moves as a diffuse front through the formation. The width of the front depends upon the kinetics involved in the mechanism representing the capacity of the soil to retain pol- lute and the hydraulic dispersion phenomena in ground water flow. The mean velocity of movement of a pollute front may be expressed as: c C vf s -~iT~ (1) where S is the velocity of the pollute front in feet (or meters) per day, R is the retention capacity per unit gross volume of earth for the pollute in micro- grams per cubic centimeter, C is the concentration of that pollute in the feed solution or waste-water at the source in milligrams per liter and v is average velocity of the transporting water at the same point in the earth in feet (or meters) per day and f is the porosity. Also, vf is the discharge per unit cross- sectional area at this point. Hence the velocity of the pollute front is inversely proportional to the retentive capacity of the formation for the pollute and directly proportional to the concentration of pollute in waste-water and the percolation rate. The retentive capacity would never be less than the amount of pollute contained in the amount of waste-water required to displace the pore volume. This would be the product of the concentration of pollute in the waste-water and the porosity of the earth, Cf. Substituting this value for R in equation (1) yields S = v. Under these conditions the pollute moves with the percolating water and it is not retarded. Such a pollute would be an ideal tracer of ground water if it met other conditions of detectabi 1 i ty and stability. Actually, the number of substances which fit in this category are not so great as one might suppose. It is more likely that some mechanism would increase the retentive capacity of the formation. It would appear that the mechanism most probably responsible for re- tention of anionic surfactants is physical adsorption on the soil. While soils, particularly the clay fraction, exhibit some cation exchange capacity, the anion exchange capacity is practically nonexistent. Further, the ABS is soluble in the low concentration found in waste water, particularly when diluted by the ground water. Adsorption, however, might account for considerably more reten- tive capacity than the pore volume in view of the very low concentration of these substances in the transporting water. (12) Renn and Barada investigated the use of various common adsorbents for removal of ABS from water supplies. While mineral adsorbents, being the type which prefer the hydrophilic end of the ABS molecule, are less effective than hydrocarbon-adsorbing surfaces, nevertheless they do adsorb measureable amounts of ABS. Their studies showed suspended silt could adsorb 20 to 50 mg of ABS per gram of silt. For clay, talc, diatomite, silica, and calcium car- bonate, the adsorption they obtained was in the order of magnitude of 1 mg of ABS per gram of material. They obtained higher adsorption on mineral oil and activated carbon, generally, than on any of the above-mentioned materials other than the silt. An activated carbon filter has been developed for removal of ABS from household water supplies and Eckenfelder and Barnhart ' have proposed the use of a similar device for treating laundry waste on Long Island. The retentive capacity of the solid phase for adsorption of a pollute is generally expressed as the weight of adsorbate, X, per unit weight of adsor- bent, M. The solid phase retentive capacity would be X/M times p, the bulk density of the earth material. The total retention would be the sum of the pollute stored on the solid phase and that stored in the liquid phase. R = Cf + *p (2) Substitution in equation (1) yields an expression for the relative velocity of the pol lute front. S/v=— V- 1 + M_! Cf (3) or S/v = 1 + D (3a) where D is a distribution factor equal to the ratio of pollute on the solid phase to that in the liquid phase. Thus, the rate of movement of a pollute is dependent on the distribution factor and the percolation rate of the transporting water. The amount of pollute adsorbed on the solid phase will increase until it equals the specific adsorption capacity in equilibrium with the concentration of the pollute in the liquid phase. According to the Langmuir isotherm, the equilibrium is expressed as k C X/M = k, k 2 C (k) Accordingly a study of the characteristics of the adsorption isotherm for ABS on some typical earth materials seems to be worthwhile in making predictions regarding the movement of ABS in ground water. Undoubtedly adsorption on the surface of soil particles is an important mechanism in retarding the movement of ABS, but under some conditions the develop- ment of a zoogleal slime on the soil particles may cause further loss of ABS. The large surface area provided by microbial cells may well offer considerable adsorptive capacity for ABS. In addition, it is known that there is some bac- terial decomposition of ABS. Many investigators have studied the biological degradation of ABS in order to determine its fate in biological waste treatment plants . They have shown the decomposition of ABS is slow; even in the intensively active biological systems encountered in trickling filters and activated sludge plants only about half the ABS is decomposed. Nevertheless, the time available for decomposition is much greater in the soil, and in the first few feet of a soil percolation system, it may be that a significant frac- tion of the ABS is) actually decomposed. This phenomena is important in septic tank percolation fields, sewage oxidation ponds, holding ponds for industrial waste and ground water recharge operations where the water medium contains sufficient organic matter to support the growth of biological slime* It is important, therefore, that the relative magnitude of ABS decomposition, adsorp- tion on the surface of microbial cells, and adsorption on the soil matrix and possibly other phenomena be evaluated. Objectives of Present Study One purpose of this study has been to determine the relative importance of biological slime activity and physical adsorption in retarding the movement of ABS through a soil system. The approach used to do this has been to build identical columns of earth material and apply to one of them an aqueous solu- tion of ABS to measure the physical adsorption* The other column is seeded with sewage and synthetic substrate so as to develop a zoogleal slime. An aqueous solution of ABS has then been applied to this column and the amount retained compared with the former column. These studies are being continued to evaluate the effect of the slime when it is not actively metabolizing. A second objective is to evaluate the effect of various parameters on the physical adsorption of ABS on typical water-bearing earth materials. Such parameters as mineral species, surface area, ABS concentration, and chemical structure are being studied by both batch and column techniques. The work has thus far dealt only with a coarse quartz sand which was being used for the biological studies. The purpose of this was to evaluate the additional retention in the presence of the biological growth. Studies on other typical aquifer minerals are underway also. I I ANALYTICAL TECHNIQUES Methods Avai lable Of importance to the research is the method used to analyze and deter- mine the ABS. A search of the literature disclosed that the methylene blue method , infrared spectrometry and radioassay procedures were avai 1 able., The methylene blue method, as modified by the Task Group Report and included in the 11th Edition of Standard Methods, is by far the most widely used analytical technique for detecting ABS. This method depends on the forma- tion of a blue colored salt when methylene blue reacts with anionic surfactants, which include not only ABS but alkyl sulfate as well. The salt formed is soluble in chloroform but not in water, whereas the dye and the ABS are soluble in water but not in chloroform. The color intensity of the dye-ABS complex in the chloroform is proportional to the concentration. The color intensity is mea- sured in a spectrophotometer and compared with standard solutions for apparent ABS content. Interferences due to organic and inorganic compounds do arise with this method. Positive errors are much more common than negative errors. The infrared method was developed by the syndet industry to provide a method which is specific and accurate for low ABS concentrations in water. This method also eliminates alkyl sulfates. The amount of ABS in any given water sample can unquestionably be established by the infrared technique. In this procedure, the water to be analyzed is passed through a column of activated carbon and the ABS is adsorbed on the carbon. This concentrates the ABS and separates it from many interferences. ABS can be quantitatively collected from concentrations as low as a few parts per billion by this process. The ABS is then desorbed from the carbon and subjected to several purification steps, the last of which is an amine extraction into chloroform. The amine complex is then examined in an infrared spectrophotometer and, by comparison with known standards, the amount of ABS present can be determined. The infrared method is quite lengthy and fairly complicated, and therefore, not suited to routine analysi s. For routine laboratory analysis the radioassay technique using sulfur- ic 35 labelled ABS (ABS ) as developed at the University of California was used in this study with modification. This procedure permits separate recovery and determination of ABS, of inorganic sulfur produced by degradation of ABS and of 8 intermediate products. The ABS is separated from the water sample by liquid- liquid extraction using ether. It is then concentrated by adsorption on acti- vated carbon. The ether is evaporated, the carbon is resuspended in acetone and water, and the carbon is then transferred to a planchet, dried, and counted for radioactivity. The inorganic sulfur remaining in the aqueous portion is oxidized by bromine water and the sulfate recovered by barium chloride precipi- tation and filtration for counting in a proportional counter. The remaining intermediate products, which were not ether soluble or precipitated as the sul- fate, were determined by drying an aliquot of the filtrate in a planchet and counting. The complete analytical procedure as modified is outlined in the Appendix. Reasons for Selection The methylene blue analysis has been used to determine the specific 35 activity of the ABS feed solutions. The specific activity is the amount of 35 radioactivity per milligram of ABS. Since the ABS solutions were prepared using distilled water, it was not felt that any interferences of serious mag- nitude would result using the methylene blue technique. Having established the specific activity of the radioi sotope- labelled ABS, the total sulfur in any sample of ABS solution can be determined as well as the fraction of the sulfur which is still ABS, the fraction which has been degraded to inorganic sulfur, and the intermediate products. Further- more, the interferences encountered in the methylene blue method due to the presence of other organic compounds originating in the sewage or soil will be eliminated. Inasmuch as one of the objectives of this project is to evaluate the effect of biological growth on the retention of ABS, it was considered essential that this interference be eliminated. Modifications in Radioassay Procedure The technique as described above used ether for the extraction of the 35 ABS from the aqueous solution. Because of the high volatility and undesirable vapors of the ether, an attempt was made to determine if less volatile chloro- form could be substituted. Identical and simultaneous extractions using ether and chloroform were made to compare the two solvents. The criteria used for com- 35 pari son were the reproducibility of counting and the amount of ABS recovered, or efficiency of extraction. Following the standard procedure the ether was evaporated from the carbon suspension, or slurry, and the dried carbon mixed with 50 ml. of acetone and 50 ml. of water. An aliquot suspension of acetone- water-carbon was then transferred by pipette to a planchet, dried and counted. With the chloroform extraction, the carbon-chloroform slurry was transferred to a planchet directly. The low volatility of the chloroform enabled this to be done without difficulty. The aqueous solution remaining from the ether or chloroform extrac- tions was sampled in a planchet directly for counting. The radioactivity in the water solution from the ether extraction was quite low, but that in the water solution remaining from the chloroform extraction was almost four times higher. This indicates that the ether has a higher efficiency of extraction. 35 The ABS extracted in the ether also showed higher radioactivity. The fifteen planchets containing ether-extracted samples produce a more reproducible count, as well as approximately 1 1/2 times the radioactivity than for the chloroform- 35 extracted ABS . Although it is felt that the ether is less desirable to work with, its higher efficiency of extraction and more uniform results in radio- active counting offset the objections to its use. The radioassay technique as recorded in the University of California (19) report noted that the dried carbon was packed into deep well planchets and "struck" smooth with a small metal spatula. This method was practiced in this laboratory and not found to be very satisfactory. Difficulty was encountered in packing the dried carbon in the planchet to produce an even and smooth surface. The activated carbon in the dry state is very light and disturbed by the slightest air movements. When the planchets packed with carbon were placed in the internal proportional counter, the large initial volume of gas that passed through the chamber for pref lushing caused the surface of the packed carbon to be disturbed. The loosened carbon collected on the interior of the chamber and in the gas exhaust tube, causing an abnormally high background count due to this contamination. In an effort to simplify packing in the planchets and minimize con- tamination, the acetone-water-carbon slurry was transferred by pipette into planchets and then dried. The slurry was well mixed before transferring to 35 insure uniform distribution of the ABS throughout the carbon and a represen- tative aliquot. This method proved a fast and simple method of depositing carbon in the planchet evenly. Consideration was given to a procedure that would eliminate the drying of the ether-carbon mixture and resuspension in acetone and water by pipetting the ether-carbon suspension directly in a planchet. ABS was extracted from two water samples by ether and then adsorbed on activated carbon,, One sample of the ether-carbon suspension was transferred directly into planchets for drying and counting. The other ether-carbon suspension was dried on a steam bath, and the carbon resuspended with acetone and water. This resulting slurry was pipetted into planchets to be dried and counted. The high concentration of the mixture of ether and carbon, plus the high volatility of the ether led to difficulties with this procedure. The tip of the pipette easily clogged with the ether-carbon mixture, which if used as a routine procedure would be exasperating and perhaps be dangerous due to spills in attempting to unplug the pipette. This difficulty was not encountered with the acetone-water-carbon suspension. The high volatility of the ether also presented a problem in obtaining an even dis- tribution of the carbon across the bottom of the planchet. The ether would evaporate so quickly that the carbon would collect in small spots and form a non-uniform surface. One advantage of the direct plancheting of the ether- carbon mixture was the higher radioactivity obtained. The reproducibility of samples was not very reliable and the other difficulties cited ruled out the acceptance of this method. The resuspension of the dried carbon with acetone and water produced very reproducible results and did not have any of the dis- advantages mentioned above with ether-carbon slurry. The amount of extra time expended in drying and resuspending the carbon was felt to be justified. As part of these investigations of the analytical technique, the amount of acetone and water to be added to the dried carbon was studied. As shown in Table 1, the ABS from identical samples was extracted and adsorbed on 1.000 gram of carbon. The carbon was then resuspended in various quantities of acetone water mixture. The 10 ml variation was troublesome to work with because of the con- centrated carbon slurry. Clogging of the tip of the pipette was experienced. Another drawback of using so little suspending lfuid is the limitation placed on the number of replicate samples that may be obtained for counting. Clogging of the pipette tip was also encountered with 20 and 30 ml of acetone:water . The 200 ml variation (100 ml of acetone and 100 ml water) resulted in a very dilute slurry. Upon drying the 2 ml aliquots in the planchets, the carbon did not completely cover the bottom of the planchets. This did not produce reliable counting. With the 50 ml of acetone and 50 ml of water carbon slurry no diffi- culties with clogging of the pipette were encountered. The sample completely and evenly covered the bottom of the planchet and produced consistently 11 reproducible counting. The radioactivity of this variation was the same as the 30 ml variation. A large number of replicate samples could also be obtained with this dilution if desired,, TABLE Effect of Various Amounts of Acetone and Water on the Counting Procedure Total Amount of 1:1 Volume Calculated Amount of Carbon Acetone:Water Transferred Weight of Carbon Relative Suspended Mixture Added to Planchet in Planchet Count 1. 000 gram 1. 000 gram 1. 000 gram 1. 000 gram 1. 000 gram 10 ml 20 ml 30 ml 100 ml 200 ml 1 ml 2 ml 3 ml 2 ml 2 ml 0. 100 1 . 0* 0.100 0.43 0.100 0.32 0.020 0.33 0.010 0.18 "Taken arbitrarily as one unit Because of the very weak beta emission from sulfur-35> even small variation in the amounts of moisture in the carbon would prevent reproducible radioactive counts. The preparation of the dried carbon prior to counting was thought to be an important point in establishing reproducibility and uniformity of counting. Specifically considered was the effect of air drying the samples in relation to oven drying and dessicating. Ten 2 ml samples of ether-extracted 35 carbon-adsorbed ABS were placed in planchets and allowed to dry in the labora- tory atmosphere and counted in the internal proportional counter. The air-dried planchets were then dried in an oven at 103 C. for approximately 15 hours and cooled in a dessicator for approximately 6 hours prior to recounting. The counting of the ten samples was more uniform after oven-drying. A further point investigated was the gain in weight of ten other samples that were air dried, dessicated, counted and then placed in the labora- tory atmosphere to adsorb moisture. The gain in weight was again very uniform. This confirmed the large adsorption of moisture on the carbon and its apparent 35 effect on counting. Other laboratory carbon-ABS samples were checked during other analytical work to determine the effect of oven-drying and dessicating on counting. When replicate samples were checked, the counting was found to be more uniform after the samples had been dried and dessicated. This work led to a standard procedure of drying all carbon samples for at least two hours in an oven at 103 C. and cooling the dried samples in a dessicator prior to counting. The long time required to evaporate the liquid from the planchet was found inconvenient. An effort to reduce the time was made by trying to remove much of the liquid by filtration. A precipitation apparatus consisting of a cylindrical funnel and a plunger, between which filter paper could be placed, was used to filter the slurry- In vacuum filtering a 2 ml. portion of the slurry from a pipette, the carbon collected on the filter paper did not distri- bute evenly across the surface. The carbon would form into mounds depending upon where the pipette tip was placed. This occurred with or without the vacuum being applied while the slurry was being discharged onto the filter paper from the pipette. In removing the vacuum-dried carbon and filter paper from the apparatus, the carbon would crumble and therefore present a poor surface for counting. Possible losses of the radioactive carbon in the working area also led to the disapproval of this procedure. For all of the radioassay counting of prepared samples a Nuclear Measurements Corporation Model P03A internal proportional counter was used. P-10 gas (90 per cent argon, 10 per cent methane)was used to flush the chamber. All samples were counted for a minimum of two minutes and a maximum of ten minutes. The desirable level of total counts is 5,000 in order to limit the statistical counting error to about 1.5 per cent. All counted samples were corrected for background and radioactive decay. This enables the samples to be compared for each individual analysis conducted. The counted samples have not been corrected for self absorption because infinite thickness was thought to have been obtained. In certain phases of the research, use is made of direct plating of 35 the ABS in the planchets. By this is meant the transfer of 2 ml of water sample to a planchet without any pretreatment. This is then dried and counted. The procedure is applicable when it is certain the ABS has not been degraded in that it simply measures the total sulfur associated with the original ABS. Where the possibility of degradation of ABS is being investigated, a modification of the University of California ether extraction procedure is used to determine the fraction which Is still ABS= The principal innovation in the procedure is the method of transferring the carbon to a planchet. The procedure is outlined in the Appendix B. 13 35 In order to determine the amount of ABS reduced to inorganic sul- fate by biological action, a barium sulfate precipitation is made. This pro- 35 cedure is outlined in detail in Appendix B. The exact amount of the ABS reduced to sulfate was not determined by this radioassay technique, in that 35 to date no attempt has been made to distinguish between the ABS degraded completely to sulfates from intermediate inorganic compounds. 14 Ml PREPARATION OF LABORATORY COLUMNS Physical Description Six glass columns have been prepared for laboratory study of ABS in relation to physical adsorption and biological degradation. The columns are constructed of 2" diameter Pyrex glass pipe. Each column consists of two sections of glass pipe with an overall length of 36 inches. The 18-inch sec- tions are joined with the manufacturer's standard flanges. This feature was incorporated so the columns may be separated for sterilization in an autoclave too short to accomodate the entire length. Glass tubing was inserted in rubber stoppers in each end of the column to permit solutions to flow through the co 1 umn o To prevent loss of the column contents when the columns are disassembled at the joint for sterilization, lucite plugs 1/4" thick are inserted in both the glass pipes at the flanged joint. In order to simulate through this section of the column the effect of a porous medium, numerous 1/16 inch holes were simul- taneously drilled through both lucite plugs in concentric circles. The holes are maintained in alignment by pins to insure the hydraulic continuity of liquid flowing through the column. A thin mat of glass wool is placed between the earth material and the lucite to prevent loss of material when disassembled. Figures 1 and 2 are photographs of the columns. Figure 1 also shows the pro- portional counter. Figure 3 is a line drawing showing construction of the co 1 umn . Medium used The material used in the columns is a silica sand. A very uniform Ottawa sand which has a geometric mean size of O.838 mm was utilized. The uni- formity of this sand is shown in Figure k. The specific gravity of this sand, as determined in the laboratory was 2.64. Preparation The Ottawa sand was packed into the columns by allowing the sand to fall freely into the glass pipe while the latter was being vibrated with a rub- ber mallet. This procedure was followed in the preparation of all columns to obtain as uniform and close packing of the sand as possible. The two halves of each column are packed separately, but are used as one. 15 c E 3 o 2 o s_ o fO o CD o o 03 en c °i o to CO C E 3 'o o C E 3 *o o x: Q. fD i_ en O ■i-j o -C a. 13 FIGURE 2. Photograph of Columns C through F before Seeding 17 rO CENTER KEY FLANGE JOINT GLASS FEED TUBE •RUBBER STOPPER w ^ ^^^^ ^ ^ft^Z Z +i ^HwG^iS^^Uf v vXm^^^^ GLASS WOOL MAT EARTH MATERIAL PYREX PIPE ALIGNING KEY TEFLON WASHER DRILLED LUCITE PLUGS GLASS WOOL MAT GLA SS DRAIN TUBE FIGURE 3» Schematic Drawing Showing Construction of Soil Columns, 18 100 1 1 I i y i 111 i I I ■ I i 80 \ — \ \ 60 Y \ Ol. LlI z u. ~ ■ h- 2 uj 40 — — o or UJ CL - 20 — — i l III 1 Ml 1 L .].. ... J i 1 i i 5 4 2 I .9 .8 .7 6. .5 4 DIAMETER, mm. FIGURE k. Ottawa Sand Size Distribution. 19 When both halves of the column were packed with dry sand and had been assembled for operation, distilled water was admitted to the columns at the top while a vacuum was applied at the bottom. This was done to remove all air and to prevent air binding,, Large amounts of distilled water were flushed through each column to wash out any fines or dust that may be in the sand. The pore volume was determined by displacement of the pore fluid by a chloride solution. A dilute calcium chloride solution was allowed to flow through each column with frequent sampling and chloride determination of the effluent. The chloride breakthrough curves are shown on Figure 5. The pore volume is represented by the area to the left of the chloride breakthrough curve. An effort was made to demonstrate that the columns had been constructed in a similar manner so that their ABS retention could be compared. The general dimensions were the same in all columns and the same medium was used. The packing of the columns was compared on the basis of the porosity, permeability, and dispersion coefficient as shown in Table II. TABLE I I Characteristics of Soil Columns Column No: A B C D E F Dimensions! Inside Diameter, i nches 2 2 2 2 2 2 Total Length, inches 36 36 36 36 36 36 Porous Medium: Ottawa Ottawa Ottawa Ottawa Ottawa Ottaw. Sand Sand Sand Sand Sand Sand Weight, pounds 6.55 6.68 6.55 6.75 6.72 6.69 Porosity, per cent 32.6 3^.3 32.9 33.6 31.2 30.8 Pore Volume, ml 567 595 573 584 5^2 535 Specific gravity (calculated) 2.5^ 2.65 2.55 2.65 2.55 2.52 Permeabi II ty: gal Ions per day per square foot 1210 667 811 1 850 53^ 797 Medium Dispersion: Constant Cm 0.25 0.25 — 20 Z 111 3 UJ Z 3 -I O O «> o o 4) > u a o X O) 3 o u JZ *J « en X u C9 o o (%) 8 'NO 1VW1N30N00 9 1N3D1JN O CM 'N0I1VH1N30N00 ±N3mdd3 IV PHYSICAL ADSORPTION The first laboratory studies to determine the amount of ABS which would be adsorbed on earth materials utilized the same Ottawa sand as used on the biological phase of these investigations. The reason for this choice was to gain information about the physical adsorption which would undoubtedly accompany any biological uptake. The physical adsorption was studied both by batch operations in a flask and by continuous flow of ABS solution through columns of sand. Batch Studies A preliminary batch determination was made using 25 gram replicates 35 35 of the Ottawa sand and various amounts of ABS solution. The ABS solution had a concentration of 10 mg/^. The amount of solution added to the 25 gms of sand was 25 rn£, 50 m£ and 75 rnf- The supernatant was sampled at 15, 30 35 and 60 minutes after vigorous agitation. The uptake of the ABS by the sand is shown in Table III. Because these results were inconclusive, further re- search was conducted to determine the effect of time on uptake. 35 In the next experiment 25 gms of sand and 50 m£ of ABS solution, containing 10 mg/,0, were placed in each of five flasks. The supernatant was sampled for counting at 5 minutes, 10 minutes, 15 minutes, 1 hour, 2 hours and 5 hours. The results showed an initial 5 minute adsorption of 3.8 u,g/gm and the 5 hour terminal adsorption was k.2 u.g/gm. The trend of all the samples showed the highest adsorption to occur at 1 hour. The variation of these re- sults are best shown graphically. Figure 6 is a plot of the physical adsorp- 35 tion of the ABS for the 5 samples. It is concluded that at least one hour is required for equilibrium to be attained, even for Ottawa sand, and other finer soils may require much longer. TABLE I I I Adsorption of ABS on Ottawa Sand Wei ght of Vc )1 ume of Shaki ng ABS 7 M, g Sand, grams ABS Solution, Time Adsorbed (M) m£ Mi i nutes (X) , microqram 25 25 15 30 60 77-5 80 80 3-1 3.2 3-2 25 50 15 30 60 80 75 135 3.2 3,0 6.0 25 75 15 30 60 67.5 60 172 2.7 2.k 6.9 22 J3> 5 3s V- 0. CC O , the uptake increased almost linearly with concentration. These concentrations cover a wider range than would be expected in ground water. It is to be noted that the uptake at the end of two 35 hours for the 8.23 mg/j5 concentration was 3.5 u.g ABS /gm of sand. This is 20 per cent less uptake than was determined on the above batch study. Column Studies To correlate the batch study work, a continuous flow sand adsorption determination was made using a column, designated as Column B, as described 35 above. The feed solution to Column B had a concentration of 10 mg/£ ABS . The flow through the column was controlled at approximately 6.5 m^/minute. The Ottawa sand in the column became saturated after approximately 1600 m£ of influent had passed through. The saturation curve for this adsorption is shown in Figure 35 35 8. The adsorption of the ABS on the sand amounted to 1.01 u.g ABS per gm of sand. This was obtained by converting the area between the chloride breakthrough and the ABS breakthrough curve to volume and multiplying by the concentration of the feed solution. Column A, which was fed with sewage and had a biological slime growth 35 on the sand grains was also fed an ABS solution to measure the physical and biological uptake. The solution fed to Column A had a concentration of 50 mg/|. This higher concentration was used to reach saturation because of the low total amount of flow that could be put through Column A before clogging occurred. In order to measure the uptake attributable to the biological slime, the physical adsorption on Column A was evaluated by a strictly physical adsorption on Column B. This experiment differed from the previous in that the concentration of ABS in the feed solution was 50 mg/^ instead of 10 mg/j>. Before making this experi- ment, the dilute (10 mg/j?) feed solution was thoroughly flushed from the column. The column became saturated after 1500 m£ of the concentrated ABS solution was applied as shown in Figure 8. The uptake for this more concentrated solution 35 amounted to 3.30 p.g ABS per gm of sand. The average flow through the column during the saturation period was 16.6 m^/minute. 2** E 3 o _) z ■o c £ O z u o fl) -C h- •M < o cc h- z c UJ o •M z a u o o o Jrt T3 < CO CD CO < < => CD ^ O O O B/Br^'aNVS (Jo NOIlddOS&V sav iAiniugnino3 25 LlI _J U_ u_ UJ z Z z> _i o o to c O o (A > L. 3 O 3 o u X J£ (0 (!) u, 0Q CO UJ C9 (%) ]D_ ' NOIlVdlNBONIOQ JLN3P1dlMI 'N0I±\/U1N30N0D ±N3n"ldd3 26 V BIOLOGICAL ASPECTS Purpose of Experiment The second phase of the research dealt with the uptake of ABS on a biological slime., The purpose was to determine whether any significant increase in retention would occur due to biological degradation and/or adsorption on microbial cells. This was accomplished by comparison of the ABS removed from solution in passing through a column of Ottawa sand on which a slime growth had developed with the removal of the clean columns reported in the previous section of this report. It was thought that the comparison of the two columns, one of which had active biological slime coating the sand grains and the other which did not have any growth, would give an indication of the adsorption or degrada- tion phenomenon when the quantity of labelled ABS retained by these two columns was determined. The comparison would be effective since the two columns were otherwise identical. It was also intended to find whether the ABS is degraded while passing through the column- Column B The physical adsorption of ABS on the clean sand in Column B was discussed above. It was shown that when an ABS solution containing 50 mg/^ is applied to the column, 3.30 u.g of ABS were adsorbed per gram of dry sand. Seeding of Column A To seed Column A with microorganisms and develop a growth on the sand, 115 liters of settled sewage was applied to Column A in 15 days. The slime that developed was black, as shown in Figure 1. So much growth accumulated that the column clogged almost completely. The column was then washed with 175 liters of tap water until the permeability increased. Because the amount of bacterial seed remaining in the column was questionable, additional seeding was attempted by applying 46 liters of dilute sewage (1 liter of settled sewage and 8 liters of tap water). The sewage was filtered through glass wool as well as settled. Assuming that the column then contained sufficient seed to develop a growth and being anxious to avoid any further application of suspended solids to the column which would only increase the clogging effect, the feeding of the column was thereafter continued using a synthetic waste containing only dissolved substrate and minerals. The synthetic waste consisted of 300 mg/^ of anhydrous dextrose, 50 mg/j> of ammonium chloride, and 1 m£/£ of buffer solution containing 8.5 g/iJ 27 KH 2 P0 v 21.75 g/& K^PO^, 33.^ g/£ Na^PO^ ■ 7H 2 0, and 1.7 g/| NH^. All this was made up in tap water for pH 7-2. Standard Operation The growth of slime proceeded so rapidly that in a short time the column had clogged to the point of practically stopping the flow. The only method found effective in restoring the permeability was to drain the column fluid, and by application of a partial vacuum from a water aspirator to the effluent end of the column to draw air through the column. The flow of air over the slime was thought to shrink the slime largely by dehydration. The re- duction in slime may be due in part to endogenous respiration under the aerobic condition and in the absence of substrate. The velocity of air flow through the interstices also may have sheared off parts of the slime and carried them out of the column. Dehydration seems the most likely shrinking mechanism. After the permeability was restored, the feeding of the column of synthetic substrate could be continued. No schedule was established initially, but the column was rested and aerated whenever it clogged. Under these conditions the performance of the column, as measured by permeability, reduction in biochemical oxygen demand (BOD) , etc., was very erratic. In order to stabilize the performance and establish as nearly steady state conditions as possible, a standard operational procedure was adopted. At the same time each day the application of feed solution to the column was stopped and the column was drained by applying a partial vacuum on the effluent end. This vacuum was continued, to draw an air flow through the column, for one hour. The column was then evacuated by closing off the air inlet. After the pore atmosphere was exhausted, new feed solution was admitted to saturate the porous medium with the solution. Then the aspirator was removed from the effluent of the column and the feed solution was allowed to flow by gravity at varying rates under fixed head until the same time next day. Samples of the feed solution and column effluent were taken daily for determination of dis- solved oxygen, biochemical oxygen demand (BOD) and chemical oxygen demand (COD). Performance of Column The BOD of the feed solution for Column A averaged about 200 mg/£ and the BOD load applied varied from 870 to 2710 pounds of BOD per acre per day. The variation was largely due to differences in the amount of feed solution which could be put through the column each day under fixed head conditions. 28 The average BOD loading was about 1200 lbs/acre/day, which is much higher than commonly applied to sand filters in sewage treatment. Steel reports that the loading on intermittent sand filters should not exceed 150,000 gallons per acre per day. At 200 mg/i> of BOD, this would be equivalent to 250 pounds per acre per day. The removal of BOD by this column varied from 18.4 to 84.6 per cent. In view of the very high loading, these figures indicate a high degree of biolo- gical activity in the column. The lowest per cent removal occurred on the day ABS was applied to the column in a concentration of 50 mg/j>. This fact raised the question of toxicity of ABS to the biological growth on that day. The COD removal effected by the column varied from 16.9 to 68.8 per cent. These wide variations in per cent removal of BOD and COD were not ex- pected and indicate that even with a standardized operating procedure, steady state conditions were not attained. The absence of dissolved oxygen in the column effluent at al 1 times and the appearance of the slime in the column indicated anaerobic conditions. To further evaluate the performance, volatile acids determinations were also made on the column. These were found to vary from 58 to 1190 mg/^ as acetic acid. The average volatile acids concentration in the effluent of the column was about 100 mg/j>, which is not significant in view of the accuracy of the distillation method used in the determination. ABS Adsorption Having established that an active biological slime had developed in the column, the retention of ABS was determined by applying a solution of the 35 same synthetic waste to which had been added 50 mg/i of ABS . Samples of effluent were analyzed by the direct plating technique described above for ABS. The column became completely clogged after 2.3 liters of solution had been used. The breadthrough curve showing arrival of ABS in the column effluent is shown in Figure 9- Unfortunately the column contents were not completely saturated when the flow ceased, as indicated by the fact that the last effluent sample contained only 50 per cent of the ABS in the influent. The ABS which was re- tained in the column is represented by the area to the left of the ABS break- through curve. The ABS retained in the pore fluid is represented by the area to the left of the chloride breakthrough curve multiplied by the average ABS concentration in the pore fluid. The concentration of the pore fluid at the influent end was that of the feed solution, 50 mg/i>. That of the pore fluid at 29 1 T i 1 ' X \ 1 1 1 ROUG mg EAKTH IN A 5 CO => _l CO o ^^_ CO o < - - X o v» 10 - 3 O CC X E o < CD _ \ LU CC ■z. \ V CO 3 _J \ CO < O o ^ * — V ^ A — — O CC O _l X X 3 o cc X i- < V \ < o 1 LU CC CD _L 1 1 1 1 1 1 1 <» o o 0J rO O co O CD c CM o o c O 10 o < «■„ CM c E 3 o o o o o o. «/» CM > 3 — O E JZ O O 15- z LU 3 en 3 o u _J ^ U_ «D Ll. o o CM 3 _l o CO < O o c o L. <0 O Q. O O CO o en Ui 3 O o O m. 11 32 60 375 1475 2275 13 110 199 0. 12 ~0.05 0.76 187 1009 1813 (1) Counted on activated carbon (2) Counted on BaSO. precipitate (3) Counted after drying original sample in planchet. It was concluded from these data that the breakdown of ABS into inor- ganic compounds of sulfur was not evident. Further study in this line is 31 necessary to ascertain the fate of ABS in passing through a biologically active column, however. Evaluation of SI ime The amount and nature of the slime growth on the sand in Column A was studied in order to provide a frame of reference for evaluation of the ABS reten- tion and comparison with future columns. An indication of the amount and acti- vity of the slime growth was provided by the column performance as measured by BOD and COD removal, as discussed above. The physical appearance of the slime growth was quite striking. After the initial black slime which developed on the sewage feed was washed out with tap water, the growth of slime on synthetic substrate proceeded. A very charac- teristic pink color developed on the portions of the column exposed to light. This was thought to be a growth of chromobacter i a, light sensitive anaerobes. The rest of the column was covered with a greyish slime except in the top 4 or 5 inches where the growth was black. To evaluate the slime further, samples were removed from the column at six different depths. The amount of organic matter on the sand was deter- mined by loss on ignition. Unfortunately the results were not very satisfac- tory. The number of bacteria in the slime was determined by agitating the sand sample in sterile water. The slime which was thus scrubbed off the sand was decanted with the water, diluted, plated on nutrient agar, incubated at room temperature for 48 hours, and counted. The results are presented in Table V . TABLE V Characterization of Slime Sample Depth, Volati le Sol i ds No. i nches mq/q ram of sand 1 3.75 1.48 2 10.25 38.9 3 16.25 0.30 4 21.12 12. 1 5 26.37 6 34. 12 Total Plate Count, organisms/gram sand+slime 2.94 x 10 1.00 x 10 0.88 x 10 0.84 x 10 0.81 x 10 0.59 x 10 32 The total plate count shows remarkably uniform growth of bacteria at all depths. The number of organisms has been expressed in terms of the total solids in the column on a dry weight basis. The total solids included both the si ime and the sand. 33 VI CONCLUSIONS The following conclusions are supported from the work reported herein^ 1. The use of radioi sotope-label led ABS provides a satisfactory and convenient determination- The original procedure has been prin- cipally modified by transferring the carbon to a planchet by pipette before drying. 2. The physical adsorption of ABS on Ottawa sand is very much lower (12) than obtained by Renn and Barada for finer earth materials. 3. The adsorption of ABS on Ottawa sand is time dependent. In batch studies, where intimate contact between sand and water is afforded, at least one hour is required to obtain equilibrium. The low ad- sorption obtained in columns may be due in part to failure to saturate all the surface area of the sand. Much of the surface is either in contact or nearly in contact with an adjacent sand grain. In these zones, the fluid is not moving, so the only transport of ABS molecules is by diffusion, a slow mechanism indeed. Complete saturation under column conditions may, therefore, require a very long time. k. From the comparison of the retention of ABS on two columns, one with and one without biological growth, it is concluded that the biological growth on earth material enhances the adsorption. At 50 mg/^ in the feed solution, 7 times as much ABS were retained on the solid phase of a column in which a biological slime was developed. 5. It has not been fully determined to what extent the retention was due to adsorption of the large surface area afforded by these bacterial cells, but it appears that this is an important pheno- menon. 6. It is concluded that the adsorption on biological slime is one mechanism which retards the movement of ABS in a septic tank (21) drain field. Hurwitz, et al . showed the assimilation of some ABS in the Illinois River. The studies reported here raise the question of the role played by algae and slime growths in a river in the adsorption of the ABS. 3^ APPENDIX 35 Appendix A References Snell, F. D. Transitions in Soaps and Syndets, Chem. Enq. News , 35 , 106 (Mar. 11, 1957) Task Group Report, Effects of Synthetic Detergents on Water Supplies, Jr.AWWA, 49, 1355 (Oct. 1957) Weaver, P. J., Review of Detergent Research Program, Jour. Water Pollution Control Fed . , 32, 288 (Mar. I960) Flynn, J. M., Andreoli, A., and Guerreva, A. A., Study of Synthetic Deter- gents in Ground Water, Jr.AWWA , 50, 1551 (Dec. 1958) Metzler, et al , Emergency Use of Reclaimed Water for Potable Supply at Chanute, Kansas, Jr.AWWA , 50, 1021 (Aug. 1958) Walton, Graham, ABS Contamination, Jr.AWWA , J>2, 135^ (Nov. I960) Deluty, Jerome, Synthetic Detergents in Well Water, Public Health Reports , 7£, 75 (Jan. I960) AASGP Committee, ABS and the Safety of Water Supplies, Jr.AWWA , 52 , 786 (June I960) Public Health Service Drinking Water Standards, 19^6, Publ ic Health Reports , il, 371 09^6) Hopkins, 0. C. and Gullans, 0., New USPHS Standards, Jr.AWWA , £2, 1161 (Sept. I960) Lauman, C. W. , Co., Effect of Synthetic Detergents on the Ground Waters of Long I si and, N. Y. , New York State Water Pollution Control Board Research Report No. 6 (I960) Renn, C. E. and Barada, M. F„, Adsorption of ABS on Particulate Materials in Water, Sewage and Industrial Wastes , 3J_, 850 (July 1959) Lieber, M. , Syndet Removal from Drinking Water using Activated Carbon, Water and Sewage Works , 107 , 299 (Aug. I960) Eckenfelder, W. W. and Barnhart, E., Removal of Synthetic Detergents from Laundry and Laundromat Wastes , New York State Water Pollution Control Board Research Report No. 5, (i960) McGauhey, P. H. and Klein, S. A., Removal of ABS by Sewage Treatment, Sewage and Ind. Wastes , 3J_, 877 (Aug. 1959) Sawyer, C. N. , Effect of Synthetic Detergents on Sewage Treatment Pro- cesses, Sewage and Ind. Wastes , 30 , 757 (June 1958) McKinney, R. E. and Symons, J. E., Bacterial Degradation of ABS. I. Fundamental Biochemistry, Sewage and Industrial Wastes , 31 , 5^9 (May 1959) APHA, Standard Methods for Examination of Water Sewage and Industrial Wastes , 11th Ed. (I960) Final Report on the Fate of Alkylbenzenesulfonate in Sewage Treatment , Sanitary Engineering Research Laboratory, University of California, Berkeley (July 1957) 36 (20) Steel, E. W. , Water Supply and Sewerage, Fourth Edition, page 519, McGraw- Hill Book Co.,, New York (i960) (21) Hurwitz, E., Beaudoin, R. E., Lothian, T., and Sniegowski , M. , Assimilation of ABS by an Activated Sludge Treatment Plant - Waterway System, Journal , Water Pollution Control Federation , 32 , 1111 (i960) 37 Appendix B A Modified Procedure for Radiochemical Determination of ABS and Degradation Products using Sulfur-35 Determination of ABS 35 Step 1 . 50 ml aliquots of the ABS solution are placed in a 250 ml separatory funnel and acidified with one ml of concentrated HCL. Step 2 . The 50 ml aliquots are extracted using 25 ml of ether. The extraction is performed by shaking vigorously for exactly 2 minutes. After the ether and ABS solution has separated, the ABS solution is drawn off into a second separatory funnel. The extraction is repeated three more times. The four ether portions are accumulated in the second separatory funnel. Step 3 - The collected ether is then "washed" twice with 20 ml of 2 N. HCL for exactly 2 minutes each time. Again, the ether and 2 N. HCL wash solution are separated between washings. The wash solution is col- lected in a third separatory funnel. Step 4 . loOOO gram of carbon is placed in a 300 ml Erlenmeyer flask to which the ether is added. This mixture of carbon and ether is then evapo- rated to dryness on a steam bath. Step 5 ° The dried carbon is further treated by adding 50 ml of acetone and 50 ml of distilled water. The Erlenmeyer flask is then stoppered unti 1 planchet samples are taken. Step 6 . It has been the practice to transfer two ml samples of the carbon- acetone-water mixture to a planchet by pipette. After drying, either by infra-red lamp or in the 103 C oven., the samples are cooled in a dessicator. Counting may be performed after the drying is completed. Step 7 ° The planchets are counted in an internal proportional counter and the count is corrected for background, self-absorption and decay. Step 8 . The concentration of ABS is the corrected net cou nt per minute lri 3 . AOC /, rrr. — : — r— ^^^ . — _ x 10 , in mq ABS/1 specific activity, CNCPM/mg ABS ' 3 Determination of Inorganic Sulfur Degraded from ABS Step 1 . To the original water sample after extraction, together with the wash water, 0.5 grams of sodium sulfate are added. 38 Step 2 . The solution is heated to boiling and 15 ml of concentrated bromine water are added. The bromine water will oxidize the inorganic sulfur to sulfate. Step 3 . After the bromine color is gone, 10 ml of 10% barium chloride is added and a white precipitate is formed . Step k . The solution is then vacuum filtered through Whatman No. 2 filter paper to remove the barium sulfate precipitate, using a Tracerlab Model E-8B precipitation apparatus. Hot water is used to wash the precipitate from the Erlenmeyer flask. Step 5 ° The precipitate on the vacuum filter paper is placed in a planchet for drying and counting. The count is corrected for background, self-absorption and decay. Step 6 . The concentration of inorganic sulfur is the corrected net counts per minute nn ,, .„ 7-T-. — : — r- rM ^ r>t , / TT7 x 20, mg/1 as ABS specific activity, CNCPM/mg ABS 3 Determination of Intermediate Products Step 1 . Two ml aliquots of the filtrate from step 5 are transferred to a planchet by pipette and dried under infrared lamps. Step 2 . The planchet is counted in the internal proportional counter and the count rate is corrected for background, sel f -absorption and decay. Step 3 » The concentration of intermediate products in the degradation of the 35 original ABS equals the ml of f i 1 trate corrected net counts per minute / in Step 5 above _ n ., specific activity, CNCPM/mg ABS X " 2 "arc 39 Appendix C Publ i cations: To date no publications have resulted from the work reported herein. Staff: The following professional personnel have been employed on the project. Ben B. Ewing, Associate Professor of Sanitary Engineering, Project Di rector. September 1, 1959 to June 15, I960 - time as required June 16, I960 to August 15, I960 - full time August 16, I960 to November 30, I960 - time as required R. S. Engelbrecht, Professor of Sanitary Engineering, Co-Principal I nvesti gator. September 1, 1959 to November 30, I960 - time as required Louis W. Lefke, Research Assistant November 1, 1959 to November 30, i960 - 50% S. K. Banerj i , Research Assistant September 15, I960 to November 30, i960 - 50% Travel : No foreign travel has been undertaken in connection with this project. ko Acknowledgements The radioactive alkyl benzene sulfonate used in this study was furnished without cost to the project by the California Research Corporation, The assistance of Dr. J. To Rutherford in making this possible is gratefully acknowledged. The standard spectrophotometri c absorption curve for the methylene blue determination of the specific activity for the ABS was furnished by the American Association of Soap and Glycerine Producers through the kind assistance of Mr. H. V. Moss of Monsanto Chemical Company. <