'.■.;.■ - << LIBRARY OF THE UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN no. 13 - \(o CONF. ROOM ►NfJINFFRlN^ LIBRARY f EB 27 1978 The person charging this material is re- sponsible for its return to the library from which it was withdrawn on or before the Latest Date stamped below.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 AT URBANA-CHAMPAIGN L161 — O-1096 Digitized by the Internet Archive in 2013 http://archive.org/details/effectofchemical16ghos £-», $"**" • 0C1 CIVIL ENGINEERING STUDIES SANITARY ENGINEERING SERIES NO. 16 ENGINEERING LIBRARY **n*™ UNIVERSITY OF ILLINOIS EWX P URBANA, ILLINOIS . V THE EFFECT OF CHEMICAL COMPOSITION OF ALKYLBENZENE SULFONATES ON ADSORPTION BY SOILS By SAMBHUNATH GHOSH Supported By DIVISION OF WATER SUPPLY AND POLLUTION CONTROL U. S. PUBLIC HEALTH SERVICE RESEARCH PROJECT WP-18 DEPARTMENT OF CIVIL ENGINEERING UNIVERSITY OF ILLINOIS URBANA, ILLINOIS JUNE, 1963 THE EFFECT OF CHEMICAL COMPOSITIONS OF ALKYLBENZENE SULFONATES ON ADSORPTION BY SOILS by SAMBHUNATH GHOSH Supported by Division of Water Supply and Pollution Contro U. S Public Health Service Research Project WP-18 Department of Civil Engineering Universi ty of 111 inoi s Urbana, II 1 inois May, 1963 I I I ACKNOWLEDGMENTS This research has been supported by a research grant, WP-18 (C2) , from the Water Supply and Pollution Control Division, Public Health Service, Department of Health, Education and Welfare, and carried out at the Sanitary Engineering Laboratory of the University of Illinois. The research was performed under the direction of Dr. Benjamin B. Ewing, Professor of Sanitary Engineering, University of Illinois. The author wishes to take this opportunity of thanking Dr. Ewing for his valuable guidance, enthusiastic assistance and constant encouragement in performing the work. He is particularly indebted for the invaluable suggestions of Dr. Ewing in preparation of this thesis. Acknowledgments are gratefully made to Mr. Shankha K. Banerj i , Mr. Michael J. Suess and Mr. Chu-Chuan Hsu, all co-workers in the ABS project, for their help during some of the experimental work. Members of the Sanitary Engineering Laboratory who helped the author some time or other in connection with this research are also thanked. The radioactive alkylbenzene sulfonates used in this study were furnished free of cost by the Colgate-Palmolive Company and the California Research Corporation. The author wishes to extend his gratitude to Dr. Rich- ard B. Wearn, Director of Research and Development, Household Products Divi- sion, Colgate-Palmolive Company, Jersey City, N. J., for some of his valuable suggestions and for furnishing data on the chemical analysis on the alkylben- zene sulfonates. I V ABSTRACT This research was undertaken to determine the effects of chemical composition and particularly the effects of lengths of hydrocarbon chains of alkylbenzene sulfonates on their adsorption by soils. To achieve the de- sired objectives, adsorption isotherms were run using batch techniques with 35 six typical aquifer materials and ABS of four different compositions, namely (i) pure sodium dodecy lbenzeneparasu 1 fona te , (ii) pure sodium dodecy 1 - benzeneparasul fona te , (iii) sodium dodecy lbenzenepa rasu lfonate mixed with other homologs having a 1 ky 1 chains of 7 to 18 carbons, and (iv) sodium salt of pentadecy 1 benzeneparasu 1 fonate mixed with other ABS having a 1 ky 1 chains of 8 to 19 carbon atoms. The sulfonated benzene was attached to the fourth carbon in cases of the pure compounds and predominantly to the third carbon in cases of the mixed ABS. Column studies were undertaken with the coarse siliceous soils in order to obtain qualitative verification of the results of the isotherms of such soils with the pure alkylbenzene sulfonates. The results of this investigation have indicated that the pure C ]f - benzene sulfonate is absorbed less than the pure C, 9 benzene sulfonate by all the soils examined. Out of the blended ABS, the one with C . ,. benzene sulfonate as the major frac- tion was adsorbed more than the one having C, 9 benzene sulfonate as the pre- dominant constituent. The results of the column studies were in agreement with those of the isotherms in that both the methods have shown increased ad»- sorption of pure C . _ ABS over the pure C homolog. ABS adsorption was also found to substantially reduce the cation exchange capacities of clayey soils, such reduction being proportional to the amounts of adsorptions in cases of soils containing montmor i 1 Ion i te clay mineral. The adsorption of ABS on soils was of the physical type brought about by the weak dispersion forces and the adsorbate would be leached out by detergent free waters. The process of adsorp- tion would only temporarily retard the movement of ABS in ground waters. CONTENTS Page ACKNOWLEDGMENTS i i i ABSTRACT iv LIST OF TABLES vi i i LIST OF FIGURES ix CHAPTER 1: INTRODUCTION 1 1 . 1 Genera 1 1 1.2 Characteristics and Importance of ABS 1 1.3 Problems Attributed to ABS 2 1.4 Movement of ABS in Ground Waters and the Probable Mechanisms of Its Retardation 6 1.5 Objectives of the Present Study 11 CHAPTER 2: REVIEW OF CERTAIN ASPECTS OF SURFACE CHEMISTRY RELATED TO THE THEORY OF ADSORPTION 13 2 . 1 Int roduct ion 13 2.2 Forces between Atoms and Molecules 13 2.3 Phase Interfaces 16 2 A Interfacial Energy and Surface Tension 16 2.5 Relationship between Energy and Surface Tension 17 2.6 Properties of the Surfaces of Liquids and Solids 18 2.7 The Process of Adsorption 19 2.8 Types of Adsorption 19 2.9 Adsorption from Solution 20 2.10 Adsorption Isotherms 22 2.11 Effect of Polarity and Solubility on Adsorption 27 CHAPTER 3: THE CHEMISTRY OF SURFACE ACTIVE AGENTS 29 3.1 Surface Active Agents Defined 29 3.2 Classification 29 VI CONTENTS (Continued) Page 3.3 Characteristics of the Anionic Surfactants 30 3.4 Properties of Solutions of Surface Active Agents 31 3.5 Surface Tension of Solutions of Surface Active Agents 35 3.6 Wetting of Surfaces and Penetration into Capillaries by Surface Active Agents 38 3-7 Adsorption of Anionic Detergents on Solid Mineral Surfaces 40 3.8 Orientation of Adsorbate Molecules on the Surfaces of Adsorbents 47 3.9 Kinetics of Adsorption 48 CHAPTER 4: MATERIALS USED IN THIS STUDY 49 4. 1 Introduction 49 4.2 Properties of Soils Used 49 4.3 Types of Al ky 1 benzenepa rasu 1 fonates - Sodium Salt Used 53 CHAPTER 5: LABORATORY PROCEDURE 57 5 . 1 Introduct ion 57 5.2 Analytical Techniques for Determination of ABS Adsorption 58 5-3 Determination of ABS Retention by Batch Studies 59 5.4 Determination of ABS Retention by Column Studies 66 5.5 Determination of Cation Exchange Capacities of Soils ~jk CHAPTER 6: EXPERIMENTAL RESULTS 76 6.1 Adsorption Isotherms of the Soils in Batch Techniques 76 6.2 ABS Adsorptions in the Earlier Set of Columns (Nos. 1A through 6A) 76 6.3 ABS Adsorptions in the Newer Set of Columns (Nos. IB through 6B) 96 6.4 Results of the Cation Exchange Studies 97 CHAPTER 7: DISCUSSION OF THE RESULTS 1 04 7.1 ABS Adsorptions on Soils by Batch Studies 104 CONTENTS (Continued) v I I Page 7.2 Adsorption of ABS on Soils Contained in Saturated Soi 1 Columns 120 7.3 Effect of ABS Adsorptions on Cation Exchange Capacities of Clayey Soils 128 7.4 The Methylene Blue Test 134 7.5 Sanitary Engineering Significance 135 CHAPTER 8: CONCLUSIONS 137 REFERENCES 142 APPENDICES 147 Appendix A - Procedure for Determination of Cation Exchange Capacities of Soils Before Adsorption of ABS by the "Ammonium Acetate Method" 147 Appendix B - Modified Procedure for Determination of Cation Exchange Capacities of Soils after Adsorption of ABS 150 VI I I LIST OF TABLES Table No. Page 1 Physical and Mi nera log ica 1 Properties of Soils Used 51 2 Chemical Characteristics of the Cjc (blend) and C|2 (blend) A lkyl benzene Sulfonates 5^ 3 Details of the Batches Set Up for the Adsorption Isotherms for Soi 1 s 60 k Characteristics of the Earlier Set of Columns (Nos. 1A through 6A) 70 5 Characteristics of the Newer Set of Columns (Nos. 1A through 6A) 73 6 Details of Batches Set Up for Studies of the Effect of Adsorption on B.E.C. 75 7 ABS Adsorption on Silty Clays in Batch Studies 77 8 ABS Adsorption on Siliceous Soils in Batch Studies 78 9 Adsorption of ABS in Saturated Columns of Glauconitic Sandstone 90 10 Adsorption of ABS in Saturated Columns of Mi ssi ssi ppian Sandstone 91 11 Adsorption of ABS in Saturated Columns of Ottawa Sand 92 12 Reduction of Base Exchange Capacities of Clayey Soils due to Adsorption of ABS 101 13 Comparison of the Values of *a ! and 'n 1 of the Freundlich Isotherms for Different Soil -ABS Systems 111 14 Theoretical Monolayer Adsorptions of ABS, A , Maximum ABS Adsorption Obtained from Isotherms, A , and A m /A t in Percent LIST OF FIGURES Figure No, Page 1 Attractive and Repulsive Orientations of Polar Molecules 14 2 Schematic Diagram Showing Variation of Physical Properties of Typical Detergent Solution 32 3 Schematic Diagram Showing Variation of Concentration of Individual Species Present in Colloidal Electrolyte Solution 32 4 Micelle Structure 32 5 Schematic Diagram Showing Variation of Surface Tension with Concentration of Surfactant 37 6 Contact Angle 37 7 Schematic Diagram Showing Adsorption of Surfactants (Having Sulfonated Benzene Attached to a Straight Hydrocarbon Chain) in Charcoal from Mixed Solvents Containing Methanol, Benzene, Water S- Ammonia 37 8 Composition of the C|2 (blend) and C]c (blend) Compounds (the Figure Shows Percentage Distribution of ABS Molecules Having Varying Numbers of Carbon Atoms in the Alkyl Groups) 56 9 Photograph of the Rotating Discs Used for Running the Ottawa Sand Isotherms 63 10 Photograph of the Batching Box Containing 50-Mi 11 Miter Centrifuge Tubes Used for Running the Isotherms of Mississip- pian Sandstone, Glauconitic Sandstone, Peoria Clay, 1 1 1 i te and Benton i te 65 11 Schematic Diagram Showing Feeding Arrangements of a Typical Column 68 12 Photograph of the Earlier Set of Six Columns (Nos. 1A through 6A) 69 13 Ottawa Sand Isotherms 79 14 Isotherms of Mi ss issi ppian Sandstone 80 15 Isotherms of Glauconitic Sandstone 81 16 Isotherms of Peoria Clay 82 17 Isotherms of 1 1 1 i te 83 18 Isotherms of Bentonite 84 35 19 Breakthrough of Chloride and ABS in Saturated Columns 1A & 4A of Glauconitic Sandstone 86 LIST OF FIGURES (Continued) Figure No. Page 35 20 Breakthrough of Chloride and ABS in Saturated Columns 2A 6- 5A of Mi ssi ssi ppian Sandstone 87 35 21 Breakthrough of Chloride and ABS in Saturated Columns 3A & 6A of Ottawa Sand 88 22 Comparison of Adsorptions of Cjc (pure) and C|2 (pure) in Saturated Columns of Glauconitic Sandstone 93 23 Comparison of Adsorptions of Cir (pure) and C^ (pure) in Saturated Columns of Miss i ssi ppian Sandstone Sk 2k Comparison of Adsorptions of C]c (pure) and C ] 2 (pure) in Saturated Columns of Ottawa Sand 95 35 25 Breakthrough of Chloride and ABS in Saturated Columns IB & 4B of Glauconitic Sandstone 98 2.6 Breakthrough of Chloride and ABS in Saturated Columns 2B & 5B of Mi ss i ssi ppian Sandstone 99 35 27 Breakthrough of Chloride and ABS in Saturated Columns 3B & 6B of Ottawa Sand 100 28 Relationship between Percentage Reduction in B.E.C. and ABS Adsorption 103 29 Freundlich Isotherms of Peoria Clay 105 30 Freundlich Isotherms of II lite 1 06 31 Freundlich Isotherms of Bentonite 107 32 Freundlich Isotherms of Ottawa Sand 1 08 33 Freundlich Isotherms of Mi ss i ss i ppian Sandstone 109 3^+ Freundlich Isotherms of Glauconitic Sandstone 110 35 Relationship between Reduction in B.E.C. of Soil and the Fraction of Soil Surface Covered by Adsorbed ABS 130 36 Schematic Diagram Showing Mechanism of Reduction of B.E.C. of Bentonite by Adsorbed ABS Molecules 132 37 Schematic Diagram Showing Mechanism of Reduction of B.E.C. of 1 1 1 i te and Peoria Clay (Containing 1 11 i te , Chlorite, Quartz, etc) by Adsorbed ABS Molecules 133 CHAPTER 1: INTRODUCTION 1 . 1 Genera 1 Shortly before World War II a new type of surface active agent, the synthetic detergent, was developed to overcome the age-old obstacle present to some degree in virtually all water supplies, the hard water salts, and to provide the housewife with unhindered cleaning performances in spite of the hardness of the water supply. Because of their improved cleaning efficiency, the synthetic detergents achieved rapid public acceptance and today constitute about 70 percent of the volume of the entire soap and de- tergent industry in the U. S. (1). Synthetic detergents have now occupied almost 90 percent of the household shelf once occupied by soap made from animal fats, and upwards of 3 billion pounds per year or 100 pounds per family are now used (2). 1 ,2 Characteristics and Importance of ABS Synthetic detergents are manufactured from the by-products of the petrochemical and chemical industries. Packaged synthetic detergents consist of; (a) a surface active agent or surfactant, (b) phosphate builder compounds, (c) miscellaneous builder compounds, (d) perfumes, (e) colors, (f) optical dye, etc. The surfactants have a type of molecular structure which is respon- sible for lowering the surface tension at any interface and loosening the dirt particles from the underlying fibers or other surfaces to be cleaned. The total quantity of surfactant fractions contained in the package products is about 10 to 15 percent of the total volume (1). Today the surfactant material most commonly used is the anionic surfactant alkylbenzene sulfonate which probably accounts for some 70 percent of the surfactant volume likely to be disposed of in waste water of one kind or another (l). Alkylbenzene 2 sulfonate is the most important of a family of petrochemical compounds more broadly referred to as a 1 ky 1 aryl sulfonates because they combine in one molecule both an a 1 ky 1 side chain and an aryl grouping of carbon and hydrogen atoms. When the aryl group is a sulfonated benzene, the surfactant is known as the alkylbenzene sulfonate. Dodecy Ibenzene sulfonate which is very com- monly found in package detergents is actually a mixture in which the alkyl side chain has an average of 12 to 13 carbon atoms (3). The hydrocarbon portion used for the production of alkylbenzene is polymerized propylene and benzene, which have their origins in an oil refinery mainly as a by-product from catalytic cracking of heavy stock for the production of gasoline (k) . Benzene is also derived from the coking of coal. The general features of the process of manufacturing of alkylbenzene sulfonates may be summarized as (3) (5): 1. Preparation of a satisfactory alkyl side chain component which may be an olefin or an alkyl ha 1 i de . 2. The condensation with benzene in a "Fr iede 1 -Crafts" type reaction using any of the several acidic catalysts (usually ALC 1 3 or HF) . 3. Sulfonation of the alkylbenzene. k . Purification of the resulting sulfonate, conversion to the desired salt (usually Na-salt), isolation and finishing opera t ion . It has become common practice to refer to alkylbenzene sulfonate and its sodium sa 1 t as ABS . 1 -3 Problems Attributed to ABS The problems attributed to the ABS are numerous and most of them stem from the fact that this surfactant fraction of the synthetic detergents 3 is difficult to be disposed of by conventional sewage disposal plants, since these petroleum derivatives are not easily subject to b iodegradat ion . The surfactants, after they have done their job in the household sink, find their way into the sewage disposal plants or septic tanks, cesspools, seepage fields, etc., and eventually into the wells, cisterns, reservoirs, streams, etc., and final', y back to household wate 1 " supply systems. It has been reported by Mctauhey and Klein (6) that the removal of ABS by primary sedimentation is disappointingly small and only 50 to 60 per- cent removal can be expected by the activated sludge treatment while a maxi- mum amount of removal of 25 percent was obtained in the trickling filters. These authors have also reported that a maximum ABS removal of 70 to 75 per- cent could be effected in a two-stage activated sludge plant (7). Various other special methods have been suggested and tried in laboratory models but none of them have yet proved to be economical. McGauhey and Klein (6) have shown that the process of "surface stripping" which in- volves the two steps of: (1) induced frothing and (2) froth removal and dis- posal could reduce the ABS concentration in sewage to 1 mg/1 if used as a tertiary process. Abrams and Lewton (8; t^.ave shown that 99 percent of the influent ABS could be removed in an ion exchange column containing strongly basic type of resin and operating on chloride cycle. In their expe-iments the concentration of ABS in the influents ranged from 2 tc 100 pom. The principal regenerant used was NaCl with dosages ranging from 5 to 20 pounds per cubic foot of res m usea in the bed Another study was made by Chambon and Giraud (9) on the possibility of eliminating anionic syndets by oxidation by ^2^2 ' n P resence ° f catalysts like CuSO . ammonium meta vanada te , MnSO, , CoNO , Hohr ! s salt ; ^e 9 ^S0 ) etc., prior to discharge of sewage in river waters. The autho-s observed that disintegration of the anionic syndet by oxidation was directly related :o the amount of ruO- used and the type of catalyst employed (MnSO, and Mohr's salt were most effective). One of the more popular methods which has been studied by some workers involved the application of activated carbon or activated charcoal for removal of ABS by adsorption. Renn and Barada (10) observed that nearly 50 ppm of water- treat- ment grade activated carbon would be required to bring 1 ppm of ABS down to 0.5 ppm and around 100 ppm to reduce the ABS by 90 percent. Morris and Weber (1!) estimated that the cost of removal of ABS by activated charcoal would be $800,00 per million gallons of water treated assuming no regenera- tion. This cost is prohibitive at the present time. In an attempt to elimi- nate anionic syndets from sewage prior to discharge into rivers by adsorption on activated carbon and charcoal, Chambon and Giraud (9) found that these adsorbents could completely adsorb the syndets, but the duration of their effectiveness was rather short. Professor W. W. Eckenfelder has also done a lot of wck on removal of ABS by activated carbon. As a result of investigations by various individuals and agencies the following major problems were alleged to be caused due to the presence of ABS- (a) Aesthetic objections due to appearance of foams in rivers, lakes, sea beaches and, most importantly, in private wells in areas where sewage treatment plants a r e not available and wastes are treated in septic tanks, seepage fields, cesspools, etc. ( 1 ) (2) ( 12) ( 13) ( 1*0 . The development of foam is dependent on the mineral and organic content of the carriage waters and upon its pH and temperature (10). Most frothing appears at ABS concentrations in the range of 05 to 1 ppm or higher ( 1 0) ( 1 2) ( 1 3) ( 1 5) . (b) Contamination of ground waters. The first report about con- tamination of ground waters came from Flynn (12) and many similar reports followed afterwards ( 16) (1 7) (18) (19) . Most of these cases occurred in r egions served by individual water supply and waste disposal systems and where the plot density is high ; distance between well and septic tank is less than 150 feet and the wells were shallow, in depth as was the case in Portsmouth, Rhode Island (13) (17) and Long Island, New York (12) (14). [z) Suds back-up in the waste wate r systems of multistoried apartment buildings (l)(2). (d) The presence of ABS in water is reported to cause an off -taste which can be detected by most people at concentrations of 1.5 ppm or more (12), but the Soap and Dete-gent Association ( 1 j| claims that taste and odor cannot be caused at less than 16 ppm ABS. (e) Other minor allegations like the toxic effects of ABS on plant and fish life, chronic toxicity on human beings drinking water containing low level ABS, damages to recreational facilities, difficulties in the operation of septic tanks receiving ABS with wastes have been brought to the fore from time to time but it has not been conclusively proved that normal levels of ABS present in waste waters are directly responsible for these adverse effects (1). In fact, in a recent article (20) it has been reported that irrigation with sewage is beneficial to plants in spite of the presence of ABS in any amount likely to occur in sewage at the present time, McGauhey and Klein [7) report that there is no evidence that ABS found in domestic sewage threatens to render the sedimentation, activated sludge or filiation units o f a treatment plant incapable of performing the functions for which the/ are designed. Gcwdv (21) had also mentioned that the small concentration of syndets present in sewage can hardly produce any frothing and settling problem in disposal plants, provided the proper concen- tration of m"xed liquor suspended solids is maintained, He did not observe any frothing in the aeration tanks of an activated sludge plant even after adding 250 ppm o f syndets ove- and above that normally present in sewage. On the other hand, frothing couid easily be produced at will in the absence 6 of added synthetic detergents by reducing mixed liquor solids below 1400 ppm. The problems attributed to ABS are of great concern to the deter- gent manufacturers, sanitary engineers and public health authorities. Public opinion against the use of alkylbenzene sulfonate based detergents has been vigorously increasing in the U. S. and legislative action intended to curb the manufacture of such detergents seems to be imminent. Two federal bills are proposed and five state legislatures, namely those of Maryland, Wisconsin, Nebraska, California and Connecticut have seen the introduction of some type of bills calling for the ban of non-degradab le detergents (22). It should be noted that the higher population density and more widespread reuse of river waters in some European countries, notably England and West Germany, have aggravated the problems ascribed to ABS to a much greater extent and governmental controls are imposed in some cases (1)(6), A new law passed by the Federal Republic of Germany prohibits the use of a 1 1 detergents and washing compounds not capable of biological degradation or decomposition in water purification plants, in natural waters and/or in the soil. German detergent manufacturers must switch over to products which are at least 80 percent degradable by October 1, 1964 (22). As a matter of fact, the unesthetic effects of ABS, viz foaming and off -tastes are perhaps the reasons why the limit of ABS is recommended to be 0.5 ppm in the U. S. P. H. S. Drinking Water Standards of 1962 (23). It is interesting to note, however, that water containing as much as 0.5 mg/1 of ABS would contain at least 5 percent water of sewage origin and the taste and odor difficulties are likely to arise from other wastes and their degradation products rather than from ABS (23) (2k) . 1 -k Movement of ABS in Ground Waters and the Probable Mechanism of Its Retardat ion Syndets are much more resistant to bacterial degradation in contrast 7 to soaps, and this high resistance plays a great role in the build-up of syn- dets in polluted ground water. Syndet-pol lu t ion has been found in various types of soils ranging from fractured rocks to sands and gravels. It has been said that chemical pollutants (such as ABS) travel much faster than bacterial pollutants and only part of the waste detergent may be removed by the sewage digestion and percolation through ground. Once detergents reach the ground water reservoir they will continue unchanged for a long time (14). In the small town of Portsmouth, Rhode Island, investigation by the Rhode Island State Department of Health revealed that syndets were very stable and travel appreciable distance through the ground and into the water table (13). Examination of the investigations in different places (12) (14) revealed that there is a decrease in syndets with increase in well -depths and, more importantly, with increase in the separating distances between the wells and the disposal systems. Investigations by Flynn et al. (12) on syndet pollution in Suffolk County, New York revealed that of the 99 wells less than 65 feet from the cesspools, 46 showed traces of syndets. They also concluded that although some protection from pollution may be obtained by deepening the wells, the only gain would be time and syndets wi 1 1 ultimately appear. With continued addition of syndets at a rate of about 100 lbs/year by an average home owner year after year, the amount wils build up and the problem would become more acute each year. Experience with wells in Suffolk County, and in other areas as will be mentioned presently, substantiates this conclusion. A study of ground-water pollution in Long Island, New York by launderette wastes showed that at Mastic, New York, ABS had travelled an overall distance of 1 1 00 feet downstream from a launderette. At Peoria, Illinois ABS has been found as far as 1800 feet from a recharge pit utilizing Illinois River water (25). The greatest distance reported was a sewage 8 oxidation pond at Kearny, Kansas, where a well 4000 feet away contained ABS (25). Although synthetic detergents have been increasingly used in the last 15 years or so, ABS - pollution has been found at distances of the order of a hundred feet. If, however, the ABS had moved unretarded with ground water, then this distance could have been covered in a matter of a few months, which was never the case. A single example will perhaps be enough to clarify this point. If the ground water in the area of the launderette at Mastic, New York has traversed at a rate of 1 to 3 feet per day and the ABS moved with the same velocity, then the distance of 1100 feet could have been covered in 1 to 3 years. But the launderette had been in operation for 12 years. It can therefore be concluded that there must be some mechanism by which the movement of ABS through ground water is retarded. It has been stated by Ewing et al. (25) that the movement of a pollute through ground water is a displacement process in which the pollute is retained by a zone of earth receiving the same. When the capacity of the earth to retain the pollute is exceeded and further inflow of the latter is continued, the pore-retained pollute, and perhaps that retained on the solid phase, is displaced to a new zone. The mean velocity of the pollute front i s expressed as 5-S£ (1.1) where, S is the velocity of the pollute front, v is the average velocity of transporting water, c is the concentration of the pollute in the incoming water, f is the porosity of the soil zone and r is the retention capacity for the soil zone per unit gross volume of earth. If the pollute has to move unretarded, i.e., if S has to be equal to v, then r = cf, i.e., retentive capacity is equal to the amount of pollute contained in the pore volume. But, generally retentive capacity of the soils is more than the amount stored in the pore volume, due to phenomena like ion-exchange, adsorption, absorption, chemical reactions, precipitation, etc., with the result that velocity, S, of pollute front is less than the velocity, v, of the transporting water and the diffusion of the contaminant through the soil is retarded. It is known that ABS does not form an insoluble precipitate by reacting with Ca or Mg contained in earth deposits. According to Grim (26), anion exchange capacity of soils are of two types: (a) replace- ment of OH ions of clay mineral surfaces; (b) anions such as borate, phos- phates, sulfates, arsenates, etc., may be adsorbed by fitting onto the edges of the silicate tetrahedral sheets and growing as extensions of these sheets. Schell and Jordan (27) suggested that phosphates and silicates are adsorbed 3+ k+ by holloysite and bentonite by substitution in the lattices for Al and Si and the amount of anion exchanged dependes on the surface areas of clays (bentonite as such had the largest anion exchange capacity). Mitra and Dharam Prakash (28) observed that the amount of P0> ions adsorbed by kao- linite and montmor i 1 loni te depended on pH , the adsorption being higher at lower pH due to a reaction of the adsorbate with the surface films of alumin- ium ions or hydrated aluminium oxides. It should, however, be noted that the above types of anion exchange capacities of the soils due to its clay mineral fractions would be very small, and only a few particular types of anions as mentioned above can be exchanged. Although ABS is an anion, it cannot pre- sumably be exchanged by the clay minerals due to any of the mechanisms of anion exchanges suggested by Grim (26) or Mitra and Dharam Prakash (28). This conclusion is very well supported by the studies of Klein et al. (15) who have shown that 80 percent of the ABS adsorbed by soils in laboratory columns could be leached out easily by detergent-free water. The authors 10 have also concluded that the residual ABS on the soil after leaching was held by molecular entrapment between particles of clay. Thus it is quite evident that the ABS molecules were loosely associated with the soil particles and definitely not held by ion exchange which requires strong bonding. Having the ion exchange and the chemical precipitation phenomena thus ruled out, the most probable mechanism responsible for the retention of ABS in soils and ground water appears to be that of adsorption and absorption combined with saturation of the pore spaces. In fact, it has been shown by Renn and Barada (10) (29), Ewing et al. (25), University of California (30), and Wayman (31) that sand, clay, talc diatomite and calcium carbonate adsorb measurable amounts of ABS. The results of this work as presented later in this report show'that clayey soils, particularly bentonite can adsorb a large fraction of ABS from the influent. The development of a biological slime on soil particles within the first few feet of top soil, where sufficient substrate and nutrients are usually available to support the growth of microorganisms may cause further adsorption of ABS on bacterial surfaces. Banerji (32) has shown that 7 times as much ABS from a feed solution of 50 mg/1 was retained on Ottawa sand on which slime was grown. At 10 mg/1 ABS in feed solution the retention was only about 3 times as much. However, he found that ABS adsorption on sterile slime was of the same magnitude as in a column without any slime. Studies at the California University (30) have shown that under conditions prevalent in an actual aquifer the function of soil is further supplemented by biologi- cal degradation of ABS. The adsorptive capacity of soil grains for the ABS is expressed generally as the weight of the adsorbate, x, per unit weight of adsorbent, m. So the retentive capacity of the solid phase would be x/m times p, the bulk density of the earth material. The total retentive capacity, r, of the bulk 11 of earth material would then be the sum of the ABS stored in liquid phase and that in the solid phase, r = cf + - p (1,2) m Substitution of this value of r in equation (1.1) yields an expression for the relative velocity of the pollute front, (1.3) s 1 V X m p or S v 1 + D (1 A) and S = 1 + D (1.5) where, D is the ratio of the pollute on the solid phase to that in the liquid phase. Thus the rate of movement of the ABS is dependent on the percolation rate of the transporting water and the <-atio of the ABS retained on the solid phase to that in the liquid phase 1 .5 Objectives of the Present Study It is apparent from what has been discussed so far that there is a good possibility of removal of ABS by adsorption on soil particles during its passage through earth. Accordingly, a study of the adsorptive capacities of some of the typical earth materials is necessary in making predictions regarding the movement of ABS in ground water. 12 The objective of this study has, therefore, been to: (a) evaluate the effect of the chemical structure and particularly the effect of variation in the length of alkyl chains of the ABS molecules on the mechanism and ex- tent of adsorption; (b) obtain quantitative information on retention of ABS by adsorption on some of the typical water-bearing earth materials; (c) iden- tify the characteristics of the adsorption isotherms; and (d) find out if there is any correlation between the base exchange capacity and the ABS- adsorptive capacity of a soil. To achieve the above objectives, it was planned to conduct adsorp- tion studies on six different soils, viz., Ottawa Sand, Mi ss i ss i ppian sand- stones, glauconitic sandstones, bentonite (clay), i 1 1 i te (clay) and Peoria clay with alkylbenzene sulfonates of four different chemical compositions, details of which are furnished in Chapter k. Adsorption isotherm for any ABS-soil system was determined on the basis of the results of batch studies. Column techniques were later used to check some of the results of batch studies. Determination of the cation exchange capacities of some of the soils having significant clay mineral content were performed before and after ABS adsorption had reached equilibrium, in order to evaluate the effect of adsorption on the residual base exchange capacity of the soil. 13 CHAPTER 2: REVIEW OF CERTAIN ASPECTS OF SURFACE CHEMISTRY RELATED TO THE THEORY OF ADSORPTION 2 . 1 Introduction As has been mentioned in section 1.4, the retention of ABS by soils is a surface phenomenon and the process of adsorption is specifically and primarily responsible for effecting this retention. A brief review of some of the pertinent aspects of surface chemistry, which are intimately related to the process of adsorption, may therefore be justifiable at this point. 2 ,2 Forces between Atoms and Molecules Surface phenomena can be related to forces between atoms and mole- cules; and particularly unbalance of such forces at the interface between two phases The forces are electrical in nature and can be either repulsive or attractive forces. The repulsive forces may be due to Coulombic repulsion between like cha-ged ions or due to reluctance of the electron clouds to overlap each other, while the forces of attraction are due to electrostatic attraction between two unlike charges and may be of the following types, (i) covalent bonding, (ii) d i pole -d i pole interaction, (iii) ion-dipole interaction, and (v) non-polar van der Waals forces which are also referred to as dispersion forces or London forces (33) (3*0 • All the afore- said forces are short range forces, i.e., they are effective only over few angstroms distances. For the purpose of this study only the last three types of attractive forces are im- portant and should be discussed in some detail. 14 2.2,1 D i pole -D i pole Interaction A molecule is said to be polar because the center of negative charge does not coincide with the center of positive charge (5). Molecules like HC1, HF, hLO are polar molecules and have a dipole moment given by the product of the magnitude of charge, e, and the distance, d, between the centers of positive and negative charges. Much of the attraction between polar molecules, such as water, is due to di pole-di pole interaction which is the net attractive force between two polar molecules. The attractive force must be summed over all the orientations, since thermal agitation will pre- vent the dipole from aligning itself continuously in the attractive orienta- tion, shown in Figure 1. However, the two molecules are more likely to be 3CZ3 Attractive orientation Repulsive orientation Figure 1. Attractive and Repulsive Orientations of Polar Molecules in the attractive orientation rather than in a repulsive one (Figure l), be- cause energy of the former is lower than is that of the latter (34) and it is the tendency of any system to stay in its low-energy states as long as pos- sible. The "polarization factor" is more important than the "orientation factor" just mentioned. When two polar molecules are very close to each other, each molecule to some extent polarizes (i.e., alters the distribution of charge in) the adjacent one and> in consequence, the dipole moment in both the molecules are increased or decreased depending onwhether the dipoles are in the attractive or repulsive orientation respectively. The electrostatic attraction or repulsion between two molecules for a given orientation and 15 distance is proportional to the product of the dipole moments of the two molecules (3^) • Hence the attractive force is increased while the repulsive force is decreased due to polarization. Therefore, it is evident that even if the orientation factor did not exist and even if the numbers of attractive and repulsive orientations are equal, the attraction must still predominate, since, in consequence of polarization, the attractive orientations on the average are more attractive than the repulsive orientations are repulsive. The net attraction is, however, so weak that no bond is said to be formed. 2.2.2 lon-Dipole Interaction Forces of attraction also exist between a dipole and an ion and here, also, the polarization factor plays an important part. 2.2.3 van der Waals Forces London in 1930 showed that forces of attraction also exist between non-polar molecules. These forces are always attractive and very often re- ferred to as van der Waals forces or dispersion forces or London forces. The van der Waals forces are also due to electrostatic interactions, and the polarization factor that leads to d i pole-d i pole interaction can operate here too in essentially the same way (3^), and so again can lead to attraction. The van der Waals forces are also said to be due to modification in the motions of electrons in each molecule; these modifications are caused by a very rapidly moving electric field which is itself caused by the rapidly moving charged particles in all the nearby molecules. These forces increase with the areas of molecular surfaces (3^) and vary inversely as the seventh power of the i ntermolecu lar distance (33). Highly symmetrical molecules have smaller surfaces than do unsymmetr ica 1 ones, and consequently the former molecules attract one another less strongly than do the latter. 16 2 .3 Phase Interfaces Surface phenomena are confined at the dividing surfaces, commonly referred to as phase-interfaces, between two or more phases. Out of the five types of possible interfaces, only the solid-liquid, liquid-gas and liquid- liquid interfaces are important in the study of the surface phenomena in which surface active agents are involved. The three fundamental characteristics of a phase interface are (33): (a) the interfacial boundary is not more than one or two molecules thick due to the short range of the van der Waals forces; (b) each unit area of interfacial area possesses a definite quan- tity of interfacial energy as will be explained later, and (c) there is an electrical potential across the interface. 2 .4 Interfacial Energy and Surface Tension The molecules at the interior of a phase are attracted by other molecules due to the van der Waals forces and the resultant of the attractive forces on the molecule at any instant is zero. The molecules at the inter- face, on the other hand, are bounded on one side by molecules of the same kind but on the other side by a phase of different molecular density. There exists, therefore, a resultant force on the molecules at the interface tending to pull them towards the bulk of the phase of higher molecular density. The existence of such an unbalanced force field results in the possession of free 2 energy, F (ergs/cm ) by each unit area of the interface. From whan has been stated above it is obvious that a liquid-gas interface is subjected to inward pulls towards the bulk of the liquid and, as a result, it will have a tendency to assume a shape of minimum possible area. The surface is thus in a state of tension and the free energy, F , is mathematically and dimens iona 1 1 y equivalent to what is called the surface 17 tension, y Q , which again is regarded as a force per unit length of the surface (dyne cm ) and exerted tangential ly to the surface. It should be noted that surface tensions of a 1 1 pure liquids fall with increase of temperature and become zero at what is called the "critical temperature." 2 ,5 Relationships Between Interfacial Energy and Surface Tensions Like a liquid-gas interface, all the other types of phase interfaces also possess a contractile tendency which is manifested as the interfacial energy. The interfacial energy between two liquids is less than the surface tension of the liquid with higher surface tension, because the attractive forces between one liquid phase and the other will be greater than between the liquid and a gas. This may be mathematically expressed as (35): 7 AB = 7 A " 7 B (2,1) where, y AD = interfacial energy between liquid A and liquid B /. = surface tension of 1 i qu i d A 7 D = surface tension of liquid B b Davies and Rideal (36) refer to equation (2.1) as "Antonoff's relationship" which states that the interfacial energy of two mutually saturated liquids is equal to the difference between their surface tensions, the latter being measured when each liquid has become saturated with the other: The interfacial energy of a solid-liquid interface is given by a similar relationship as (35): 7 SL = 7 SA ' 7 LA (2.2) 18 where 7^ = interfacial energy of the solid-liquid interface, 7_. = surface free energy of the solid-air interface, and 7 . = surface tension of the liquid It should, however, be noted that satisfactory methods for directly measuring 7 and 7_. are not available. That y.. is greater than 7 is proved by the fact that a marked temperature rise (due to release of heat of immersion or heat of solution) occurs when a finely divided insoluble solid is immersed in a liquid since a solid-air interface of high free energy is disappearing and is being replaced by the solid-liquid interface of relatively low energy. 2.6 Properties of the Surfaces of Liquids and Solids The surface tension of liquid-air interface is equal at all points in the interface since liquids are in dynamic equilibrium with the interior and the surface is being continually renewed. Under usual conditions most solids never approach true isotropy (35) and the molecules remain fixed in position. The atoms in the surface of the solid are not equivalent in nature in that those atoms in the rugged asperi- ties are more energy-rich and hence have more surface energy. Moreover the free energy of the solid-air interface varies from point to point, because the immobility of the surface region does not allow the atoms to rearrange themselves (33). As a consequence, the surface free energy and the surface tension of the nonequ i 1 i br ium surfaces of solids are not equal as in the cases of the surfaces of liquids. It should be noted that the surface energy of finely divided solids (of larger surface area) is much more than that of coarse ones, as has been proved by experiments (33) in which the heat of solution of the former was observed to be greater than that of the latter. 19 2. 7 The Process of Adsorption The adsorption process can be described phenomeno logical ly as the distribution of a species between the fluid phase and the surface region of a solid phase (33)- The distribution is usually expressed as moles (or micrograms) of the species adsorbed per gram of solid versus moles (or mg) of the species per liter of the fluid phase, or versus the pressure of the adsorbate if the fluid phase is gaseous. The phase interfaces are associated with the existence of some amount of surface free energy and there will consequently be a tendency for the free energy of the interface to decrease, and it is this tendency which is ultimately responsible for the phenomenon of adsorption (37) (38) . It can therefore be concluded that the more is the free energy of the interface, the more will be its tendency to return to the state of lower energy by ad- sorption. The amount adsorbed increases with increasing pressure or concen- tration of the adsorbate and decreases with increasing temperature (as is to be expected due to release of heat of adsorption). 2 .8 Types of Adsorption The process of adsorption can be divided into two broad classes, namely physical adsorption and chemi sorpt ion . Physical adsorption is a re- versible process and the adsorbate may be recovered unchanged chemically by lowering the pressure or increasing the temperature. It is supposed that this type of adsorption occurs due to the weak short range forces of attrac- tion between the adsorbent and the adsorbate such as van der Waals forces and that chemical bonds are not involved. Chemi sorpt ion is the process in which the adsorbate and the ad- sorbent are joined by covalent bonding The process is more or less irre- versible and is characterized by higher energy of adsorption. It is of less 20 importance in surfactant application than physical adsorption. Chemisorp- tion is believed to be limited to monolayer of adsorbate, whereas physical adsorption can lead to multilayer formation. Much work has been done on the above two easily distinguishable types of adsorption. Unfortunately there is quite a dearth of literature on systems having behaviors intermediate between the two extreme types of ad- sorption. This indicates that these intermediate cases are merely less well studied . 2 . 9 Adsorption from Solution There is a tendency for the free energy of the surface of a solution to decrease and so a solute, having a force field weaker than that of the solvent molecules, will be pulled toward the surface, i.e., they will be posi- tively adsorbed on the interface and the surface tension of the solution will be lowered. For similar reasons a solute having stronger force fields than the molecules of the solvent will be pulled toward the bulk of the solution and it is said to be negatively adsorbed at the surface (35) (37). In general, most inorganic salts in aqueous solution raise the surface tension of the solutions and are negatively adsorbed, while most organic compounds are posi- tively adsorbed on the surface and thereby lower the surface tension of solu- tion. The surface-active agents having the polar-nonpolar type of molecules show this effect to an extreme degree. If a solute such as the ABS comes in contact with air-water interface then they will be held there by energy barriers which oppose both the removal of the polar sulfonate group from the water and the immersion of the hydrocarbon chain in the water. The approach of the ABS molecules toward the surface will be governed by molecular and eddy diffusion only, while the desorption of this adsorbate is controlled by quite a high energy barrier. The amount of ABS at the interface thus 21 increases, tiP the chemical potentials in the surface and bulk are equal; this corresponds to a greater concentration of solute in the surface than in the bulk (36) as expressed by equations (2.3) and (2.k) below; u. = u° + RT In C (2.3) s |i = s u° + RT ln s C (2.4) where, u. = chemical potential of the solute in solution s u. = chemica.l potential of the solute at surface u° = chemical potential of the solute in solution at standard state S M° = chemical potential of the solute at surface at standard state C = concentration of the solute in bulk S C = concentration of the solute at surface R = universal gas constant T = absolute temperature The function of the interface is thus to provide a large area at which solute molecules capable of lowering the surface tension in a liquid-air interface or the interfacial energy in a solid-liquid interface can be ad- sorbed. Since the 7-. for finely divided solids is more than that of coarse ones (Section 2.6), it follows from equation (2.2) that 7-. for the former will also be greater than that of the latter, 7.. remaining the same irre- spective of the size of the solids. The greater 7.. of the finer particles is responsible for greater adsorption of solutes from solution on the finer soils. This explains why the adsorption of ABS from solution on clayey soils is greater than that on the coarser grains of sands and sandstones (see Chapter 6). Under equilibrium conditions the rates of adsorption and desorp- tion of solutes on the solid-liquid interface are equal, just as the rate of 22 evaporation equals the rate of condensation in case of adsorption of gases on sol ids (33) . As the adsorption of gases increases with pressure, so also the extent of adsorption of a solute from a solution increases with concentrations and there is sometimes a limit for adsorption by the interface corresponding to monomolecular layer formation, after which no further increase in adsorption occurs with further increase in concentration. The amounts adsorbed on solids from solution under equilibrium conditions depend on the concentration of the solute, the available surface areas of the adsorbent, temperature and charac- teristics of the solute. 2.10 Adsorption Isotherms 2.10.1 Def ini tion When equilibrium has been reached during an adsorption process the surface concentration of the adsorbate builds up to a level where s^ = n (2-5) as mentioned in the preceding section. Substituting the values of s u. and u, from equations (2.4) and (2.3) respectively in the equation (2.5) one obtains: C GXP [ ~ TF J =exp(-) (26) whe!-e 3 x, the standard free energy of desorption (36) = u° - s u° Since three variables, namely (a) moles of adsorbates per gram of solids, (b) equilibrium pressure or concentration of the adsorbate and (c) the tem- perature, are involved in the process of adsorption, only two are ordinarily plotted, the third being held constant. The isothermal relationship between the equilibrium surface and bulk concentrations is called an "adsorption 23 isotherm," which is very satisfactory to account for the variation of adsorp' tion with pressure or concentration, while temperature remains constant. Normally one determines the adsorption isotherm by measuring the amount of adsorptions at various pressures or concentrations and then plotting the amounts of adsorbate adsorbed per unit weight of adsorbent as a function of the equilibrium pressure or concentrations. In our study, the adsorptions of ABS by any given soil were measured at eight different concentrations in order to obtain sufficient points on the adsorption isotherm (at constant room temperature) for the selected type of soil -ABS system (see Table 3, Chapter 5) . 2.10.2 Lanqmuir and Freundlich Isotherms The classical Langmuir and Freundlich isotherms have been very widely applied to adsorption from solution, the latter as an empirical equation and the former in terms of the concept of a uniform surface with adsorption limited to a monolayer. The Langmuir equation may be written as follows (33): x abC m 1 + bC (2.7) v where — denotes the amount of adsorbate, x, per unit amount, m, of solids (e.g., mi 1 1 imoles/gm or micrograms/gm, etc.), C is the concentration of the adsorbate expressed in terms of volume fraction, weight fraction, etc., a, is a constant given by the specific surface area of the sol id, 22, to the actual area per unit of adsorbate molecules a , b, is related to the heat of adsorption Q, by the equation: b = b« exp (^r) (2.8) 2k and b ' ■ ^° r o y™>* (2.9) In the above equations, N is the Avogadro's number, M is the molecular weight -13 x in qm/mole and i n , is the molecular vibration time of 10 seconds. If — 3 ° m is in moles per gram, then using the usual units for X and a° of square centi' meters per gram and square centimeters per molecule respectively, it follows that (33) 8= NG o (2-10) Qualitatively, 'a' is a measure of the surface area of the solid and 'b' is one of the intensity or strength of the adsorption. The Langmuir isotherm y in its simplest form given by equation (2.7) can be obtained if — is plotted as a function of C as has been done in this study. It is convenient, when testing the fit of the equation to data and evaluating 'a' and 'b' values, to put Langmuir's equation in its linear form: C + — x/m ab a (2 . 1 1 ) C 1 Thus a plot of — t~ versus C should qive a straiqht line of slope —and x/m a intercept -j-. The principal postulates of the Langmuir isotherm are: (a) no interaction between adsorbate molecules, (b) the adsorption is on localized sites and (c) the maximum adsorption possible corresponds to a complete mono- molecular layer. The Freundlich isotherm is usually written in the form: }_ m " a C (2,12) or log - = log a + - log C (2.13) 3 m n 3 25 where, 'a' and ' n ! are characteristic constants of a system and C is the equilibrium concentration of the solution ( 1 5) (33) (37) (38) . A plot of x 1 log — versus log C should give a straight line of intercept, a, and slope, — . Roughly speaking, 'a' is a measure of the surface area of the solid and 'n 1 a measure of the intensity of adsorption, although these constants are not susceptible to as definite a physical interpretation as in the case of the Langmuir equation (33). The value of 'n' is greater than 1 if the isotherm given by equation (2.7) is convex to the concentration axis (33) whereas it should be less than 1 if the isotherm is concave to the concentration-axis. Unlike the Langmuir isotherm, the Freundlich isotherm as given by equation (2.12) does not reach any limiting value indicating complete mono- layer formation. Also, unlike the Langmuir isotherm, which can be derived from the kinetic theory of gases and by statistical methods, there does not seem to be any theoretical basis for the derivation of the Freundlich iso- therm. It is possible, however, to obtain the Freundlich equation as the net of superimposed Langmuir equations for a heterogeneous surface (33). Attempt has been made to fit the adsorption data of ABS on soil (as reported in Table 13, Chapter 7) to the linear form of Freundlich equation for three reasons: (a) most of the adsorption isotherms obtained by plotting — vs C (Figures 13 through 18) did not conform to the typical shape (convex to the C-axis) of the Langmuir isotherm and some of them were actually concave to the C-axis which do not indicate an approach to a limiting adsorption corre- sponding to monolayer formation; (b) the Freundlich equation also includes the Langmuir isotherm till the formation of monomolecu la r layer and at the same time fits nicely when there is absence of linearity in the low concentration region and saturation in the high concentration region (33), and (c) other investigators ( 1 5) (25) (39) (40) have also found that ABS adsorption on soils could be very well represented by the Freundlich equation. As will be seen 26 later in Chapter 7, almost all the isotherms for adsorption of ABS on soils could be fitted satisfactorily to the Freundlich equation (2.13). 2.10.3 Gibb's Adsorption Equation The importance of this classical (I878) equation in adsorption studies calls for at least a brief mentioning, although it is not directly applicable in this study. The equation gives the relationships between sur- 2 face excess, 1 , in moles per cm of interface and the surface tension, 7 (in case of liquid-gas interface) or surface free energy (in case of solid- liquid interface) and is expressed as (33) (36) (41 ) : 1 RT dlnC (2.14) In the above equation, C, is the concentration of the adsorbate (or more properly the activity of the adsorbate). The Gibb's equation can be applied 2 for computing the amount of ABS adsorbed per cm of the soil surface if one can determine the activity of the solute and the surface free energy of the interface between the soils and the ABS solution. Evaluation of the surface free energy is rather complicated and therefore a much simpler technique in- 35 volving the use of ABS tagged with isotope S (Section 5-2) was used for determination of the amount of ABS adsorbed on soil surfaces, 2.10.4 BET I so then Adsorption isotherms are by no means all of the Langmuir type and some of the unusual types corresponding to multilayer adsorption, or those reflecting capillary condensation, hysteresis effects, etc., can be fitted to the adsorption isotherm proposed by Brunauer, Emmett, and Telier, which is commonly known as the BET isotherm (33)- The equation has usually been 27 applied in case of adsorption of gases and vapors on solids and is also said to be applicable in case of adsorption from solution. The linear form of the equation for easy plotting may be written as (42): v'0-x 1 ) = v~\; + v l m -c' ( '7 r " ) (2J5) where, v' = adsorption at relative pressure or concentration, x', x 1 = C/C (in case of adsorption from solution) = P/P (in case of adsorption of gases or vapors on solids) 1-x 1 = fraction of solute adsorbed C Q = saturation concentration P Q = saturation pressure v' = adsorption value corresponding to monolayer c ! =a constant Use of BET equation is not required in our case since the adsorption of ABS by the soils used in this study were in most cases not enough to form multi- layers and also because the data can be fitted satisfactorily to the Freund- 1 ich i sotherm. 2.11 Effect, of Polarity and Solubility on Adsorption The behavior of a given system in which an organic compound is being adsorbed on solids from solution in water (or organic solvents), may be pre- dicted very qualitatively by thinking of adsorption as constituting a distribu' tion of an adsorbate between two phases, the solution phase and the solid. The solution, adsorbent and adsorbate are then classified qualitatively accord- ing to their polarity, i.e., as to whether they are polar or nonpolar. The r ule (known as Traube's rule) is that a polar adsorbate will tend to prefer that phase which is more polar; i.e., it will be strongly adsorbed by a polar 28 adsorbent from a nonpolar solution. Similarly a nonpolar adsorbate should be strongly adsorbed by a nonpolar adsorbent from a polar solution. Appli- cation of the above rule will not be straightforward in a system such as ours, where a polar-nonpola r adsorbate is being adsorbed on rather polar surfaces of soils from a very strongly polar aqueous solution. High molecular weight materials such as sugars, dyes and polymers tend to be more strongly adsorbed than low molecular weight species (33)- This is understandable because the van der Waals forces responsible for physical adsorption increase with increase in the size of the molecules. Still another experience is that there is usually an inverse relationship between extent of adsorption of a species and its solubility in the solvent used. That is, the less soluble the material, the more st r ongly it will tend to be adsorbed. Ermolenko and Lemets (43) studied the relation between solubility and adsorption of organic acids from mixed solvents such as "C^H^. - CC1, (both non-polar), CAi, - EtOH (pola r-nonpola r) and EtOH - Me 2 C0 (both polar)" and developed the equation: A = L where, A, is adsorption, L,, is solubility and — , is the slope of the logarithmic plot of the adsorption isotherm. The above statements regarding dependence of adsorption on polarity, molecular weight and solubility will be very much applicable (see Section 7-0 in interpreting the results of this research. 29 CHAPTER 3: THE CHEMISTRY OF SURFACE ACTIVE AGENTS 3 . 1 Surface Active Agents Defined Certain solutes have the property of altering the surface energy of their solvents, even when present in very small concentration. Such solutes are known as surface active agents. They are characterized by a typical molecular structure which is essentially linear, i.e., considerably longer than it is wide. Usually one end of such compounds is comprised of a hydrocarbon radical of hydrophobic and non-polar nature characterized by weak residual valence forces, whereas the other end is hydrophilic and polar in nature with strong residual valence forces (35). The polar portion or the hydrophilic group may involve such functional groups as carboxylic acids, esters or ethers, sulfonic acids, their esters, or sulfates or their esters, etc . Perhaps the best known and oldest surface active agent is soap. Most of the early cleansing agents prepared from animal fats suffered from the disadvantage of forming insoluble salts upon reaction with the hardness present in water supplies, and the synthetic detergents were developed to overcome this defect and provide a surfactant which forms soluble Ca and Mg salts after reacting with the hardness of the water. 3 • 2 C lassi f i ca t ion Surface active agents are classified as anionic, cationic or nonionic. The active portion of the anionic and cationic species will re- spectively carry a negative or positive electrical charge after ionization in water (1)(21). The nonionic surfactants do not ionize at ali on solution in water, although they have the other characteristics of a surfactant. The use of the nonionic detergents is rather limited as they are only employed 30 in certain low-sudsing agents. The total volume of synthetic cationics is also relatively very small and since their structure is such that they react with the anionics, their surfactant properties will be short lived in mixed sewage (l). The following discussions will therefore be devoted more on anionic surfactants with which we are concerned. 3 ,3 Characteristic of the Anionic Surfactants The most important ionogenic groups in anionic surfactant are carboxy (-C00H), sulfonic acid (-HS0_ ) and sulfuric esters (-0S0 H) while those for the cationic surfactant are primary, secondary and tertiary amino- groups and the quarternary groups. The aromatic sulfonic acids in which the aromatic nucleus is an integral part of the hydrophobic group are currently and most intensively used of all the synthetic detergents in this country. The sulfonates of un- substituted aromatic hydrocarbons such as benzene, naphthalene, etc,, have little or no surface active character. The substitution of an aliphatic, cycloa 1 i phat ic or aryl a 1 ky 1 side chain for one or more of the nuclear hydro' gen atoms confers surface activity on these sulfonates provided the substi- tuent is sufficiently large. As would be expected from their structure, sulfonic acids have the physical characteristic of highly polar compounds. The sulfonic acids and their salts are soluble in water and, in fact, they are introduced into large molecules to bring about water solubility (5). Being strong acids, they and their salts such as the alkylbenzene sulfonates are completely ionized in aqueous solutions. It should be noted that the balance between the hydrophilic and hydrophobic portions of the molecule of a surface active agent is of para- mount importance. A compound which is heavily weighted on the hydrophilic side is exceedingly water soluble and displays no surface active properties, 31 The sodium soaps of more than 18 carbon atoms on the other hand are so heavily weighted on the hydrocarbon side as to be too insoluble at ordinary tempera- tures for effecting display of surface activity. On the other hand, the marked surface activity of sodium laurate is due to the fact that there exists the proper balance between the hydrophobic and hydrophilic groups of this compound. The surface activity of the a 1 ky 1 aromatic sulfonates is influenced by the number of substituents in the aromatic nucleus, the size of the sub- stituents, the size of the aromatic nucleus, the number of side chains, the number of sulfonic acid groups and the location of the hydrophilic group (35)- 3 M Properties of Solutions of Surface Active Agents The essential features of surface active solutions in bulk is the existence of colloidal size particles, or micelles, formed by the spontaneous association of ultimate molecules or ions of the solute. They are thus repre- sentative of a class of compounds known as "colloidal electrolytes." The micelles exist in thermodynami ca 1 ly stable equilibrium with the simpler ions and molecules and behave like colloids. A general schematic display of the various physical properties of the solution of a typical colloidal electrolyte, such as sodium dodecyl sul- fate, is shown in Figure 2 (33) (^) • It is seen that striking alterations in the various physical properties occur in the region of what is called the citical micelle concentration (c.m.c.) which decreases with temperature. It is believed that in the region of the c.m.c. aggregation of the long chain electrolyte into fairly large, charged units begin to occur. The units are commonly cal'ed micelles and they may contain about a hundred monomer units ( 33^ Figure 3 (M+) indicates what happens when the concentration of an anionic detergent in water is increased. At first the concentration of the <4- — o c \ A, / / Legend . Oetergencv ,„---' Surface tension "y Conductivity Osmotic pressure Interfacial tension / .063 .083 Molar Concentration of Sodium Dodecy 1 Sulfate Figure 2. Schematic Diagram Showing Variation of Physical Properties of Typical Detergent Solution 0) o c O o c o o I ^J 1 ■ ' — r I - c .m.c J _ Leqend Na + ions Long cha in — ions L^-^T^ Micel les - " — : / ^ — - 1 l*-^ - 1 1 ' 1 - Concentration of Detergent Figure 3- Schematic Diagram Showing Variation of Concentration of Individ ua 1 Species Present in Colloidal Electrolyte Solution OOQOQOQQOOO sly ?A A y A 9 A. s> A 9 "9 ~0 ~0 "o'o'qqV'V Water Lamel lar F i gu re k S phe r * ca 1 Micel le Structure 33 positive and the negative ions increases in direct proportion to the increase in total concentration of the detergent til] the c.m.c. is reached, when micelles begin to appear and then increase rapidly. After the c.m.c, the rate of increase of the anions becomes zero while the rate of increase of the Ha ions also decreases greatly since most of them remain attached as "gegen ions' 1 (41 ) . Micelles carry a considerable net charge if the monomer unit is an electrolyte, for the reason that it is the long chain ions which aggregate and the ions of opposite charge remain as counter ions or gegen ions unaggre- gated. The net charge may not be so large as the degree of aggregation would indicate since some of the counter ions remain associated with the micelle, presumably as a diffuse layer (33). 3.4.1 Mice 1 le Structure Different theories have been advanced regarding the micelle struc- ture. Hartly proposed a spherical shape, whereas McBain believed that a lamellar form also existed. The two typical shapes are shown in Figure k. In addition, Harkins considered a cylindrical micelle. Schwartz and Perry (3) (35) mention, "current views on micelle structure appear to be shared by most of the leading investigators in the field, and the major concept is that several different micellar structures are possible and in fact exist. Each type is a phase and the transition from one type to another does in fact resemble a classical phase change in many respects." Mankowich (kS) reports that sodium dodecy lbenzene sulfonate has a micellar molecular weight of 1000 at concentration of 0.2 to 0.5 percent while the same increased to 1700 at concentration of 0.5 to 1.0 percent pro- vided no builders are added. The micellar molecular weight of built (meaning addition of different concentrations of builders) sodium dodecy 1 benzene ma me 34 sulfonate varied from 93,300 to 193,000, the molecular weight generally increasing with increased concentration of builders. 3.4.2 Solubilizinq or Hydrotropic Effect Solubilization by an aqueous system may be defined as the sponta- neous dissolving of a normally water-insoluble substance such as benzene or organic dye by an aqueous solution of colloidal electrolyte. Solubilization begins to be noticeable at the c.m.c. and it appears that the solubilized terial is incorporated into the micelle itself. In fact, a very popular thod for determination of c.m.c. depends on the solubilization of dye ions of charges opposite to that of the micelles (3) (46) . 3.4.3 Effects of Chemical Composition and Salts on Micelle Formation One way in which micelle formation can be reduced is by branching the hydrophobic chain or by using two short chains rather than a single longer one (35). Hartly (47) explains the situation by stating that, "if the same amount of paraffin in an ion is retained, but the straight chain is replaced by two approximately equal branches then we shall greatly reduce the maximum size of micelle of the normal type which can be formed, because the size is limited by the necessity of some chains to being able to reach from the surface to the center." In other words the c.m.c. of a double or branched chain surfactant is more than that of a straight chain surfactant. The c.m.c. is shifted to lower values as the length of the hydrocarbon part of the molecule increased. For example, the c.m.c. for a C. ? compound is 0.008 molar while that of the corresponding C,g compound may be 0.00017 (41). Mixtures of two or more surfactants, for example mixtures of C,~, C., , C,,- and C.q fatty a 1 ky 1 sulfates, with one another have a lower c.m.c, than would be predicted by linear interpolation from the corresponding values of the 35 pure compounds (3). At the c.m.c. of a mixture, the micelles are strongly enriched with the longer chain components, i.e., with the components of the 1 owe r c.m.c. Inorganic salt lowers the c.m.c. and increases the micelle size and at higher concentrations tends to salt out the surfactant (3). Messrs. Corrin and Harkins (48) concluded that, "the extent of lowering of the c.m.c by a salt exhibits independence of the number of charges on the ion of the salt which has the same sign of charge as the ion aggregate of the micelle. The significant feature is that the c.m.c. is affected only by the concen- tration of that ion opposite in charge to that of the colloidal aggregate. The lower the c.m.c., i.e., the greater the tendency toward aggregation and the greater is the lowering of the c.m.c. by equal amount of salt." The converse of this is also true, i.e., the higher the c.m.c, the smaller will be the lowering of the same. The contrast between the critical micelle concentrations of C._ and C,n compounds is striking and it should be noted that addition of 6 numbers of "-CH '' groups increase the c.m.c, by a factor of kj . Assuming that the increase in c.m.c. is directly proportional to the increase in the number of carbon atoms in a 1 ky 1 chain, the c.m.c. of the pure C ]9 benzene sulfonate should be about 23 times higher than that of the pure C 1C - benzene sulfonate. This indicates that micelles will be formed in C , q solution at a concentration which is 1/23 of concentration at which micelles will appear in a solution of C,_ benzene sulfonate, 3-5 Surface Tension of Solutions of Surface Active Agents A typical curve of surface tension versus concentration in case of surface active agent is shown in Figure 5 (33) (35) (*+0 • In case of surface active agents as in the case of organic solutes in general, there is a 36 decrease in surface tension with increasing concentration as shown by the theoretical curve (Figure 5). In the case of a colloidal electrolyte there occurs a sharp decrease of surface tension for a quite small concentration of the solute and this is then followed by a short rise and gradual flattening of the surface tension-concentration curve (Figure 5> experimental curve). From Gibb's adsorption (see Section 2.10.3) equation which correlates the surface tension to adsorption in the form of: 1 67 ( r = . i_ _Eic_ (3 10) 1 RT d InC °' ' it can be seen that in the portion 'ax' of the surface tension-concentration curve — ™- is negative and I , surface excess per square centimeter, is positive, i.e., there will be positive adsorption. On the other hand, on . . . d7o ... !""" the rising portion of the same curve —ft is positive and consequently I , is negative, i.e., there is negative adsorption. The flattened portion of the curve corresponds to a region of concentration over which the surface tension varies only slowly and as a result the adsorption will also vary slightly in the same region of concentration. Alexander (*+9) pointed out that the point * on the surface tension-concentration curve corresponds to the c.m,c. The minimum at 'x' is explained differently by various authors. Schwartz and Perry (35) attribute this to lowering of the bulk concentration of the individual molecules of surfactant which are the true surface-active species. These authors also quote Miles and Shedlovsky who found that once the c.m.c. is passed, the impurities or other constituents of the mixture containing surface-active agent may leave the surface because of being solubilized or held within the micelle structure, thus bringing about an increase in surface tension and in effect the desorption. Davies and Rideal (36) also attributed o u- u 1/1 Le gend Ex per imenta 1 Theoret i ca 1 Concentration of Surfactant Figure 5- Schematic Diagram Showing Variation of Surface Tension with Con- centration of Surfactant = Angle between the liquid-gas interface ond the solid-liquid interface Gas or another Liquid y |\ liquid Sol id Figure 6. Contact Angle 1_ O XI k. o T3 < 0) o 2: en o Legend 1-n- 15 1 -n-C 1/4 2-n-C]5 8-n-C)5 l-n-C 12 1-n-Cio Log (Concentration in moles/1) Figure J. Scher.wtic Diagram Showing Adsorption of Surfactants (Having Sul- fonated Benzene Attached to a Straight Hydrocaron Chain) on Charcoal from Mixed Solvents Containing Methanol, Benzene, Water d Ammonia 38 the minimum at 'x' to traces of impurities. In a recent article Wayman et al, (50) have shown that: (i) non- ionic compounds are more surface active than either anionic or cationic compounds, which were found to be about equally surface active at 20 9 C; d7 (ii) the surface tension-concentration gradient —£■ is much steeper at 35 a C than at 20 a C : (iii) in a detergent, compounds other than ABS increase surface tension (and decrease adsorption or cause desorption in consequence); and (iv) bacteria reduce the surface activity of surfactants in solution. 3 .6 Wetting of Surfaces and Penetration into Capillaries by Surface Active Agents Usually wetting means that the contact angle (Figure 6) between a liquid and a solid in a so 1 i d -1 iqu i d-gas interface is zero and non-wetting, that the angle is more than 90° . According to Dupre's equation 7 SA = 7 SL " 7 LA COs6 (3 ' 2) whe-e, / = surface tension in the solid-gas interface 7_. = surface tension in the solid-liquid interface 7 |A = surface tension in the liquid-air interface From equation (3 -2) ?SA " y SL coso = ?LA (3.3) For wetting to occur then 7 SA " 7 SL 7 LA (3» Qualitatively speaking, however, 7-. and 7 . should be made as small as 39 possible if good spreading is to occur. From a practical viewpoint this is best done by adding a surfactant to the liquid phase which is adsorbed at the solid-liquid and the liquid-air interfaces and therefore lowers these inter- facial tensions. If the surfactant is non-volatile, it may be presumed not to affect 7ca • From the above discussions it can be established that wetting action is generally accomplished by the use of surfactant additives. The aforementioned mechanism of wetting by surfactants will be operative in a loose mesh work of fibers, etc. In such cases the contribu- tion to wetting made by capillary rise is at a minimum and the condition is that of a very small solid areas for which wetting by spreading is the major factor. On the other hand, wetting of a porous mass of rock, wood, etc., presents a different situation in that in such cases the liquid is required to penetrate into the capillaries, In this case the phenomenon is that of a capillary rise, where the driving force is that of the pressure difference across the curved surface of the meniscus. Adamson (33) expresses the pres- sure difference "AP" across the curved surface of the meniscus as: _ ( 7 SA - ' SL) AP " 2 r, (3.5) so that the surface tension of the liquid-air interface is not involved at all. The principal requirement for a large AP (i.e., for large penetration) is simply that y _. be made as small as possible by addition of a surfactant since it is not possible to modify 7ca- The rate of penetration into a capillary mass is given by the relationship (5): dl 7 LA r ' COse dt kr\] (3.6) kO where, r, = capillary radius, t^ = coefficient of viscosity and 1 = length of capillary. From equation (3-6) it is evident that rate of penetration is aided by a high rather than a low surface tension of the solution. Addition of a surfactant, which is a surface tension depressant, consequently is of no pract i ca 1 va 1 ue . 3 . 7 Adsorption of Anionic Detergents on Solid Mineral Surfaces Adsorption and desorption of anionic surfactants (like ABS which is of importance in this study) on solid mineral surfaces from solution de- pends on (i) particle size, (ii) temperature, (Mi) solute concentration and bulk composition of the solution, (iv) pH , (v) mi nera logi ca 1 composition of the adsorbent, (vi) structure of the solute. 3,7-1 Effect of Particle Size Studies on this aspect have recently been completed in our laboratory and it has been established that the intensity of adsorption in- creases with increase in grain size (39)- Intensity of adsorption is defined as the adsorption of ABS per square meter of the surface area of the soil. Morris and Weber ( P ) concluded from the data on adsorption of ABS on activated charcoal that the amount of adsorption increased with decrease in the square of particle size. Presumably, this conclusion holds good in accounting for the variation of total adsorption with variation in the grain sizes of the same type of adsorbent, namely activated charcoal. The surface area factor is not included in the relationship. On the other hand, the equation developed after studies (39) in this laboratory relates the varia- tion in the adsorption per unit surface area with the particle sizes of ten different varieties of soils. Both the studies show increase in adsorption with decrease in particle size, but while Morris and Weber showed that total 41 adsorptions by activated carbon increase with decrease in the square of the particle size of charcoal, the results of our studies (39) on diverse frac- tions of Pennsy 1 vanian II (siliceous soil) indicate that total adsorption is almost inversely proportional to cube root of the grain sizes of this soil. The adsorption by activated charcoal will thus increase at a much greater rate with decrease in grain size than do the adsorption by the Pennsy lvanian sandstone with decrease in its particle size. This should be the case since the specific surface area of activated charcoal would be much more than that of the sandstone for the same grain size due to the large internal surface area of the former material, compared to none in the latter type. 3-7.2 Effect of Temperature This aspect has been discussed in Section 2.7 and is of little importance in this study since the experiments were carried out at constant room temperature. 3.7.3. Effects of Solute Concentration and Bulk Composition of the Solut ion The effect of concentration on adsorption has been discussed in general under Sections 2.9 and 2.10 and therefore only some of the special aspects of adsorption of anionic surfactants will be discussed here. Messrs. Morris and Weber (11) report that the adsorption of ABS on activated charcoal increased with increase of concentration and that the in- crease was not linear with concentration, but approximately with the square root of the concentration. Also ABS was adsorbed much faster in dilute so- lutions than in more concentrated ones. It has been reported by various authors that the adsorption isotherms for surfactants show irregularities near the c.m.c. From the study of the 42 adsorption of sodium dodecyl sulfate on polystyrene, Corrin et al. (51) had demonstrated the existence of discontinuity in the isotherm at the critical concentration of micelle formation. The minimum area per molecule correspond- ing to maximum adsorption was 51° Messrs. Swartz and Perry (3) also mention that adsorption isotherms of ionogenic detergents on carbon black, graphite and activated charcoal were generally irregular up to the c.m.c. and that at this point many of the iso- therms showed discontinuities and then with increasing concentrations beyond c.m.c. followed the normal isotherm pattern. Header and Fries (52) report the following distinctive features of the adsorption isotherms of alkyl aryl sulfonate on cloth; There occurred a pronounced maximum adsorption at the concentration range of 10 to 25 x 10 molar (0.03 to 0.07 percent) and the maximum is followed either by a steady decline in adsorption from a solution of alkyl aryl sulfo- nate or by a subsequent rise in the case of adsorption from a solution of a mixture of alkyl aryl sulfonate and sodium sulfate. Also, adsorption on cloth from a mixture of alkyl aryl sulfonate and tetrasodium pyrophosphates indicated that increased phosphate in the detergent composition resulted in increased adsorption. For the same percentage of active ingredients, compo- sitions containing tetrasodium pyrophosphate show lower adsorption than com- positions containing Na^SO, . The concentrations at the maximum adsorption were in every case greater than the c.m.c. All the adsorption isotherms showed breaks and changes of slope at equilibrium concentration almost equal to the c.m.c, The sharp rise of the isotherm at the c.m.c. was attributed to adsorption of micelles. At concentrations more than c.m.c. adsorption of micelles begins to level off while adsorption of single ions tends to de- crease, and consequently total adsorption reaches a maximum. A decrease in ^3 adsorption from solution of a 1 ky 1 a ry 1 sulfonate beyond the maximum value can be accounted for by continuous decrease in single ion activity (see Figure 3) The increase in adsorption due to increase in the percentage of Na^SO, or phosphates in the mixture can be attributed to the increased molecular weight of micelles due to high salt concentration. The explanation based on compe- titive adsorption of sulfonates and phosphates was ruled out by experiments 32 performed, in which P labelled tetrasodium pyrophosphate was shown not to be preferentially adsorbed on cotton under any conditions from sulfonate solutions. These authors have also shown that considerable precipitation of calcium and magnesium sulfonates occu r red when adsorption of detergents from solutions o f alkyl aryl sulfonate and mixture of 40 percent alkyl aryl sul- fonate and 60 percent Na_S0, in 300 ppm hard water was attempted. The preci- pitation was very much less when phosphates were present in solution. Void and S i va ramakr j shnan (53) while studying the adsorption of sodium dodecyl sulfate on "Stealing N„S„ carbon" obtained almost the same pattern of isotherm as Meader and associates (52) got and details of which are discussed in the preceding paragraph.. The former group of workers (53) observed that adsorption of sodium dodecyl sulfate increased regularly as the concentration of the surfactant increased from zero to the c.m.c. and then from this point the slope of the isotherm increases sharply, passes through a maximum and then declines asymptot i ca 1 !y toward a limiting value. They also recognized that the increased adsorption commencing at c.m.c, is due to the adsorption of micelles. It was concluded that sodium dodecyl sulfates are preferentially adsorbed on the polar part of the surface which was the ash layer on non-polar carbon and as the concentration of sodium dodecyl sulfate (SDS) increase, the polar surface is covered making the ex- terior of the particles hydrophobic. The decline in adsorption after peak adsorption is reached, is attributed to a different mechanism by these authors. It is postulated that as concentration of SDS is increased beyond the point of maximum adsorption, the potential of the layer of ions at the solution-carbon interface becomes such that there is no longer any excess electrical repulsion of the adsorbed ions over their van der Waals attraction and as a result sudden aggregation of the adsorbed ions into surface micelle occurs. The attainment of such an electrical state is independent of the nature of the adsorbent. When such a phenomenon occurs the desorption of single ions ceases but the desorption of surface micelles by collision with micelles in solution become much more enhanced and hence the isotherm drops down beyond a maximum point. That salt content increases adsorption is also supported by the findings of research in Geological Survey, Denver, Colorado (5^, which showed that in presence of more than 1000 ppm of salt, ABS ad- sorption on kaolinite is much greater due to lowering of the c.m.c. 2.7A Effect of pH Not much is known about the effect of pH on adsorption of ABS on soils. Studies in this area are being conducted at the Sanitary Engineering Laboratory, University of Illinois and the results of the same are not available at present. However, in a study undertaken by Water Resources Division, Geological Survey, Denver, Colorado, it was observed that adsorp- tion of ABS on kaolinite is more in acid solutions (pH - k.0) as compared to those at pH 7 and pH 10 (5*+) . The enhanced adsorption in acid media is attributed to an increase in the number of positive sites on the clay surfaces from decomposition of kaolinite High pH is also said to be responsible for repressing the hydrolysis of soap in solution thereby reducing adsorption (3). Renn and Barada (10) concluded that the efficiency of ABS adsorption by activated carbon in acid waters is appreciably higher than in alkaline waters. 3-75 Mi nera loqica 1 Composition of the Adsorbent Investigation on this aspect has recently been completed in our laboratory and the details are reported by Mr. Suess (39). 3.7.6 Effects of Structure of the Solute That the chemical structure and molecular weight of the surfactants are significant in affecting the amount and rates of adsorption has been pointed out by some authors. It has been reported by Morris and Weber (11) that there is a decline in the rate of adsorption on activated charcoal with increase in molecular weight of the ABS . The authors pointed out that this trend should be expected since sodium 2-octyl benzene sulfonate should diffuse much faster than sodium 2-tetradecyl benzene sulfonate. The rate of adsorp- tion of sodium 2-dodecyl benzene sulfonate is greater than for sodium 6-dodecyl benzene sulfonate. It was postulated by these authors that the former com- pound with benzene ring attached near one end of the chain can coil and dif- fuse more freely than the latter in which the benzene ring is attached near the center of the chain. This latter postulation is in agreement with the generalization made in 19^6 by the New York Academy of Sciences (41) that the location of the hydrophilic group (sulfonate in case of ABS) in the middle of the hydrocarbon chain strongly favor wetting and penetration power at the expense of other surface active properties. The above observations of Morris and Weber are not in agreement with the work carried out at the Water Resources Division, Geological Survey, Federal Center. Denver, Colorado (5k) where it is found out that "ABS adsorp- tion on kaolinite is qreater for C , ,. benzene sulfonate than for C 10 benzene 3 15 <*■ sulfonate". Mr. Wayman explains the disagreement by postulating that "appar- ently charcoal and kaolinite adsorption mechanisms are different." Mysels 5 Biswas and Tuvell (kO) have clearly shown that the effect of structure is even mo-e important than that of the molecular weight of the surfactant-solute. Figure 7 (^0) clearly shows that compound of shorter chain lengths reduces adsorption by charcoal but the effect of the position of the branching in reducing adsorption is more pronounced. This latter conclusion regarding effect of branching is again in agreement with the ob- servations of Morris and Weber (11) that adsorption of sodium 2-dodecyl benzene sulfonate is greater than that of sodium 6-dodecyl benzene sulfonate. Cason and Gillies (55), in a study of adsorption of saturated and unsaturated branched chain fatty acids on ' Da rco G-60' charcoal with 95 per- cent ethanol as solvent, had obtained isotherms which indicated that branching of the chain lowers adsorbabi 1 ity and a larger branch or several methyl branches lower adsorbab i 1 i ty more than a single methyl branch. The position of single methyl branch had only a small effect, Messrs. Meader and Griddle (56; determined the effect of molecular weight and structure of sodium a 1 ky 1 benzene sulfonate on the cross sectional areas of molecules after they are adsorbed at interfaces from solution. The area per adsorbate molecule was found to decrease to a limiting minimum value c 2 of 20 A (wh i ch is close to the cross-sectional area of the hydrocarbon chain (36)) as the molecular weight of a compound and the film pressure in- creased. This is due to the fact that while the electrostatic repulsive forces between the head groups remain the same s the attractive van der Waals forces between the hydrocarbon tails increase with increase in the length of the chain. The overall result is that the longer chain compounds are more easily compressed to their limiting value than are the short chain compounds. On the other r• 03 >^ 03 X> 03 t- -D 03 o CO 03 Q. o U Q- 03 O cn O - x a. O CO in 3 O u oj o co kD 0) — H L- 03 O — 03 <_> a. u •— 03 4-1 c • — o c 4-J o i/> u X) 3 c 03 03 i — CO CD c 03 — 03 CL C a o — 4-» Vt 1/1 i/i XI — c in 03 t/) co 51 03 $ XI 01 c ■<-> 03 +-J co o o j- o r-~ CM o j- o r^ CM o -d- o P-. CM o cn o j- O CT\ O J" o o CM J- 00 cn c c — in a) if) im 03 — o_ co C I 03 • C I 03 " a. i c i 03 • Q. I o j- o P*« CM o -d- O 1^ CM o o rA CA LA a> > a> — c i/i O 1_ c o o — E x 03 c c — CO 03 +J N QJ — QC CO LA O LA O LA O L_ OV C 03 03 vO cn r^ o o CM r— o oo vO oo vO CA r^ CA cn CM LA cn p*. 5 cn CM _ O J- LA vO LA j- r-» LA CM vf> -j- M5 LA CM CA -d- LA [^ PA M3 r-^ CM CA CA CA 4-1 E c o 03 >v OA •— E E O cn o >— "N. 4-1 L|- » E c <4- 4-J cn 03 03 X o O cn ** s_ O .— >- 03 a> 4-1 Q. >> 3 •— 4-1 l/l »* • — u c >• E •— 03 4-1 i_ «4- X • — O .— 1/3 <4- o -* o • — 03 • — 1_ c Q. 3 o ^ CO CO a. o o o vO LA O CO CM -3- o CA M3 X — O O X S_ 4-) 03 03 O E >- O LA 00 vO vO vO cn o LA CM -3" O cn rA O LA O o I 03 E L- X 03 O Q-X 4-1 1- Q3 — E 03 03 4-> C — ro — — -Q 00 03 E i 0) 03 cn L U (1)S 3 03 L-CM CO W- < E — c 03 3 CM CM If) o a. l/l E — O 03 O u 03 • — C 03 •— O E cn i/i O 3 •— -jc o 03 i — 03 L_ O 03 c •— a o — ■z. 4-1 co 03 3 o- I/) 1- 0) XI E 3 C cn c c c 3 1_ >- XI X u 1_ 03 E O 1/3 cn c i/i o Q. E O o 03 X 03 X l_ o cn c 03 C i o X 03 X 52 Q) o l/> -i-J oj L D 0> C Q. •— o 4-J L c Q_ u — ' — 03 O f— 9 — CT> 0) o ~— ~— -Q 03 03 1_ i- 0) c 2: ■o c m , — fD O i/> >- c o ■4-> c 0) CO ^D fD • — > i_ 03 O i — (L) <_) Q_ LP, o .— <-> C — o c +-> o i/i u T3 D C fD fD i — co •_3 — c — D fD O E E l/l fD — c o 1 — D ■a fD o E E O i/) 03 "D c 03 »— 0) CL c D_ o — +J LO — C I/) fD in cO s: 03 l_ Q. 03 in Q. T3 l/l , — X> 0) ' — Ll, CD U- (D 1/1 E 03 a •— U i/> O 10 • — 03 CD ■t-j fD O . — a- Q_ o i-|Q 03 (0 TO 03 C c o o E 1 — 1/1 . — 1 — E •— fD D 1_ i_ .— o fD 53 4,3 Types of A 1 kylbenzenepa rasu 1 fonate Sodium Salts Used The following four types of ABS were used; 1 -propyl nonyl benzene pa ra su 1 fonate, sodium salt henceforth referred to as C. ? (pure); 1-propyldo- decy 1 benzeneparasu 1 fona te , sodium salt henceforth referred to as C lt - (pure); sodium pentadecy lbenzenepa rasu 1 fona te mixed with other a 1 ky 1 benzeneparasu 1 - fonates having alkyl chain lengths of 8 to 19 carbons henceforth referred to as C 1£ - (blend); and sodium dodecyl benzenepa rasu 1 fona te mixed with other a 1 kylbenzenepa rasul fona tes having alkyl chain lengths of 7 to 18 carbons henceforth referred to as C,„ (blend). The chemical properties of the above a 1 ky lbenzenesul fonates , as analysed and furnished by the Colgate Palmolive Company (62.) are compared in Table 2. Figure 8 compares the percentage distribution of carbon atoms in the alkyl chains of the C . Q (blend) and C . ~ (blend) compounds. The structural formulae for the pure a 1 ky 1 benzenesu 1 fona tes are as fol lows : CH Structural formula for C (pure)- H C .-CH- f > S-0Na 15 Vf ""~' "23^1 -\_y v Wl ON a Structural formula for C ]2 (pure): h 1? C R -CH- ^ \ s-i The sulfur atoms in some molecules of each of the above four ABS 35 compounds was exchanged with the radioisotope S which serves as a tracer. 35 35 The S -tagged ABS was manufactured by sulfonating a 1 ky lbenzenes with H_S 0, 35 and will hereafter be referred to as ABS . The radioactive alkylbenzene sulfonates permitted determination of ABS adsorption by the radioassay technique (see Section 5.2). 5k -o c 0) oi 4J 1 — T) n C • — • 4- LA- 3 CS CO QJ (U SZ c +J 01 N M- c o OJ X3 lf> f— u >» — ^: v/l 10 u "D •— C E m 01 l_> ^ OJ i_ d Q. co •> — J- OA CM »— <_) 0) L- 3 Q. o » — ' "O C (U MD .Q CT\ *■— ' P^ oa LA > — (_> T3 0) in >- 4-» i — -C fD CD £= • — m 1 v i_ fD *J i— L. 3 (Li u O. fD (— — cO — O o o o O LA CT\ CO — r- — J- CA CM — \0 \£> LA CO OA J" CM — — O o LA O I I o • oo o r-« LA r^~ oo CT* o <_> c o +J D -O s_ ■4-1 c L_> O L_) o <_> <_> 55 TD c — flj XI c ^ o n- lA— — 3 O I/) 0) CD TJ X C 0) D N C W- C •— O CD ■>-> X C i/> — O O >« o — -* s ^-' +J • — CSl (/) <; i_ ^.^ 0) CD "O F— +-> C X 03 O "O — C E CO CD X O cni o LA > CO C CO L0 I O I I o ' I O I I O ' I r— I o CNI LA — OO LA o . >. •>. >. >^ > c c a c C c: CD CD CD CD CD OJ x x x x x x 0-0-Q.Q-CL.0_ I I I f I I — CNI CA -J LA NO -- o — >- — L0 c ■•-• >-. 0) — -- £ in H o_ o c Q- ro LA CT> o en c X (J c 0) -o cd c OJ "O O I • — I I I I o LA I vO l"— CNj O 00 — — X X -J o CO 5^ X C X cd en o — L CO o_ CO E cc -o "O < QJ C d) C H 13 en*--- c — C JD T3 COn-- C •— 1/1 C\i — • 0) — GO ■a o o (NJ re Shows <".a rbon CO 3 M- (the ig Numbers o on 10 CO T3 en \0 < c c ■ — ZJ — M- o >- o E TJ to > c u •— Ol a) m -J- 1 — "~ X — >- XI -*: — • i/l 1 — 0) < LA— c <_; o • — a) 10 c O E (0 '" o CNI *J »—» u~> 1 < ■a c < c a) o — <+- J3 x o — u — ' ex 1) C 3 C_) NO O u- O 4-> (J o 3 o 0) jQ- — !_ X — >» a) ■W U J^ n .*-> — E u- irt i -;c {, •*-v ^-^ *■— *s •— N X x X) X C c c C a 0) o> 0) !_ X C -O 3 Q. XI C a) =5 Q. .— v X X 0) C C V_ 0) 0) D — — a n j] ,—.,—. x x <5J 0) C C i- i- a) a> 0- D. _D -Q CM LA (_> O CM LA CM LA CM LA CM LA CM LA CM LA CM LA (_) (_} O (_) O O CJ <_)C_)OOC_>(_)(_> LA CM LA CM c c c c 05 05 05 05 0) X) Xi — a) — U d) O a) o 0) >~ >- >- > Q. c C a c Q- C Q. C a. c .— c •— c •— c • — c 05 05 05 05 5s 05 05 a. o a. o Q. O Q- O 4-J O 4-J o 4-1 o 4J o ■ — f— h- co CO .— 4J — +j • — 4-J — 4-1 • — 4J « — 4-J •— 4-J • — 4-J o O O m in in i/i in in i/i c in c VI c in c in >— 05 05 1/1 "O in x in -q m -a o X o X O X O X 05 05 05 05 * 5 Z — c — c — c — c o c o c o c o c O 05 05 m 05 in (g in (u m re o 05 => 05 =J 05 3 05 i_ l_ l_ i_ CO 4-1 4-1 in i/) in m i/i in in in 05 in 05 in 05 i/) 05 in o O O O CJ C5 C5 a> a> 0) 0) — — Q_ Q_ r-H I— I — CM CA LA VD 0O cr» O — CM CA -d" LA VO 62 o c Ifl o v_ .— CD fJ 4-J 3 • — ^— I — o • — oo i — l/) • — CO s: < o to O cr o o o o o o -J- J- J- -d- -d- -3- a) a) c c 4-1 4-1 4-J c c c c 01 — ' 1 — ' CM QJ CM U ~— -C — O 4-J O O 10 10 in 10 c c c c O o o O 4-> 4-1 4-1 4J CO — <0 — CO •— C3 C 4-1 l/> — -a — c — CO Irt c C 1/1 .— .— 4-) C \f) c — in z: c "O m- C < 4-J m ro 0) 1 — CQ J= 4-1 CD «o -C CD • — 4-> C i_ • — <+- c CD O c Q_ 3 -c a: n Ql CL> O 0) C -C l/> f0 Q- id 00 3 CD 66 every day for about 10 minutes to ensure intimate contact between the ABS solution and all the surfaces of the soil grains. After a few weeks had elapsed since the date of starting of the isotherm, the tubes were centri- fuged for 10 minutes at 13,000 rpm to get a clear supernatant which could be sampled in a manner described under section 5-2. This procedure was repeated about once in 10 to 15 days until the equilibrium conditions were reached and 35 ... the change in total S activities due to adsorption and desorption of the same became insignificant. Altogether 22 batches, as reported in detail in Table 3 had been set up in order to collect data on the 22 different adsorption isotherms which one could have due to various combinations between 6 types of soils and k types of ABS. The pH of the solutions of the blended ABS varied from 6.6 to 6.8, while the pH of the pure ABS solutions were J A to 7. 5- 5 A Determination of ABS Retention by Column Studies 5.4.1 Preparation of Laboratory Columns A set of six columns were prepared out of 1/2-inch diameter and 12-inch long polyethylene tubes. The tubes were closed at the bottom with rubber stoppers having outlet tubes introduced into the same. The columns were then filled with distilled water and some glass wool packed at the bottom of the columns to provide a mat for preventing escape of any soil grains during actual operation of the columns. The soils were then packed into the columns by allowing the grains to fall freely into the tubes while the latter were being gently tapped with a small rubber mallet. This procedure was fol- lowed in the preparation of all the columns in order to ensure a uniform and close packing of soils and at the same time prevent the trapping of air bubbles within the medium. A gap of about 5 cm was left above the top of the soil medium in each column so that this space would remain filled up with 67 the influent solution to ensure uniform rate of application of the same on the top of the media. Large amounts of distilled water were flushed through each column to wash out extraneous dusts that might have been present in the med ia . The feeding arrangement of a typical column (see Figure 11) com- prised a 2-liter feed bottle, a syphon for delivering the influent solution on to the top of the column, a clamped outlet in the influent tubing for ex- pelling air bubbles entrapped within the closed system, outlet tubing provided with a clamp for percolation through the column and preventing draining of the same. All the columns were mounted on a steel framework as shown in F i gure 12 . Table k compares the columns on the basis of physical dimensions, permeability, pore volume, etc. It would be apparent from the data in Table k that for all practical purposes, the physical conditions as well as the permeability of the pair of columns (e.g., column IA and column A-A or column 2A and column 5A or column 3A and column 6A) of the same type of soil were similar and their ABS retention would therefore be comparable. 5.^.2 Operation of the Columns In order to measure the dynamic pore volume of the media, the influent-effluent tubings and the under drain systems, a dilute calcium chloride solution was percolated through each column (in order to displace the distilled water in pores, etc.) and effluents were collected in test tubes. Fi ve-mi 1 1 i 1 i ter samples were pipetted out from each test tube holding the effluent from the column and tested for chloride concentration by the Mohr method (63). The percolation of the chloride solution was allowed to continue until the chloride concentrations in the feed and the effluent so- lution became equal. The volume of effluents collected in each test tube Air p i pe Feed bottle containing ABS solution Hydraul ic head luent col lector Movable platform Syphon Rubber tube Pinch cock »1. Schematic Fiagram Showing Feeding Arrangement of a Typical Colu mn 69 c o 4-> ^-^ y"v — i cn - — / D c >. O tO *-s — i_ — CM +-> Q_ 1 — — C_> 4-1 < t/l O i/> CD — "D Q. i/l 0) i/l in i/> .* c C CD h, E T> — 5 D i/> *j Q- . — 4-> ' o 03 O O jn in oj — X 4-» "O o •— ■w a) co O CD X3 u_ c 1-1- XI 03 O CD c • j-j — >-0 cu 03 - « CO +J OJ *" c > 5 !_ O - CL 0) O 4J — o > — - (U CNJ 1_ CD Q- — 03 (/)(_) LJJ CD X> t- T3 CD C CD JC 03 CD C ■w c — - O 03 4- Q_ 4-> +-< o - (/3 C X) O -C -CO Q. - 03 03 O ui — - L < cn O 1 o X) — i-t 4-J OJ +-< i— ( o _*: ._ ^ -C 1- c Q. m O — v £ U LA =3 >— (/) 03 ' ,-N C — o i — E cn^- =3 (13 — X> X3 l_ O C C 3 O 03 03 cn CD 70 (0 l_ O O 2 x 3 N\£ CO MD 00 O ^D O < fD C Q. o . • O O vD 4-> fD ' o j- — -J- CM VO CM CO LA 4-> in CM CA CM LA CA CA LA < — o O QJ 4-> c 4- fD ^— V O — * — — (T\ o 3 lA i/l "P o* ^t — v£> r^ CA CM 00 LA in fo CM CA CM -3" CM CM ~- <— X • — c i/) in O 05 k CD E ro l/) o s—^ — a) a) CD 4-> c i_ x: — o 3 o CM 4J C 4-> a. r*>. o ca CA 00 MD O < O m » — ' cm r^ o CD • — o in O c <_> t ( E 3 4- fD 0) o CM . — o < 3 x fD c 3 r-^ ca ca CM CT* LA CM o o O o 4-1 r*> 4-» fD CL o -3- — -3- • — — -cj- CO r^ cd 4-> m **■ ** ca cm LA CA J" LA i— CD co o la 4-> 1_ <_> 0) ■— •— c I •— fD s— V fO J- L. .- 0) cu rrj CL C 1_ 4- CL r-~ o ca v£> CM CM o _Q -C in x o -d- — v£> CP\ -3" CA OO r-» c 1- 4-1 4- o in u 4-J I/) U — c in fD i/i in X. o •- (U 4-» C — o LA O CA CM -3" CM CM O CM — o 4-1 3 o in -M c 3 u C 4J =i Nr^oo O J- CM o — 4-1 < O in Q. ^A -J- o 14- u • — U 13 ■ — * O -3- — rv. o •— CA 00 r^ 4- f0 3 C CA CM J- CM CN r— i — CD u fD fD LA ft) — in • — X .C o C_> C O "e CO ^E 1- 4-1 c CD i fD ^~- i- l/l O E 4J E CL — ' U 0) i — QJ C 4- 4- — — =5 3 < E — ' E O X X C D O o \ L. i — 4- ■z. -C 3 c — C , — cr (U o O LA *J — CO H- — 3 D 4J X c E Cl O C CA D1TJ 0) o . — >»4- fD O ^ JC LA E E CO .. C Q) J- • (U •— cu at CL • 3 3 CO m CD O XI O 4- — en — o 4-. i- c E J= 4- o CD 4- — — O ro .— o ^L JD L. C 4- 1 O — 4- 1- O 21 o l/> — c a) > 0) cr fD Q) 3 — E in O D L- C . (D •— CL E 0) Q- +- 1 > 1- • — -Q 1- >» c C — CL O CQ X O >> — > a) c i_ o o 3 4J o_ x o O l/l < +- 1 2 O 71 and also the chloride concentrations in the feed and the effluents were noted for plotting the chloride breakthrough curves. The CaCl_ solutions in the 2-liter feed bottles were then replaced by the feed solution of the desired type of tagged ABS and percolation of the same permitted to proceed at a flow rate of about 9 ml per hour by ad- justing the hydraulic heads. It was expected that the selected flow rate would be moderate in magnitude and satisfactory in that chances of short cir- cuiting were improbable at such low flow rates and more intimate contact be- tween the soil surfaces and the solution would be ensured. As reported in Table k, columns 1A, 2A and 3A were loaded with C . c (pure) solution while C,~ (pure) solution were applied through columns 4A , 5A , and 6A . ABS concen- trations in the influent solutions, determined color imetri ca 1 ly (62) were 8.2 mg/1 and 8.6 mg/1 for the C . ,. (pure) and C.~ (pure) compounds respective- ly. The effluents from the columns were collected in test tubes or bottles, their volumes measured and recorded. Samples were taken from the test tubes and bottles containing the effluents from columns for determination of ABS concentrations by the radioassay technique described under Section 5-2. Percolation through the columns was interrupted during holidays and sometimes at nights. Daily records of the duration of ABS percolation through the columns were maintained and the total periods of operation of the columns are reported in Table k. After the above columns (which will be subsequently referred to as the earlier set of columns) were operated for some time it was observed that 35 the initial ABS breakthrough in columns 1A, 2A , and 3A were very rapid in that the effluent concentrations became about 50 percent of the influent con- centrations when only 30 to 50 ml of solution were passed. This was rather unexpected. It was suspected that some leakages or short circuiting in the flow might have occurred at the time of starting of the runs and also, that 72 the greater slenderness ratio of the columns might be responsible for this. It was therefore decided to verify this apparently unexpected phenomena by duplicating the studies on a new set of columns containing the same media and loaded with the same type of ABS but differing from the first set in that glass tubes of smaller slenderness ratios were used for construction. 5.4.3 Studies on ABS Adsorption by the New Set of Columns The same procedure as outlined in section 5.^.1 above was used in preparation of the new set of columns and the arrangements for introducing the influent solution and collection of the effluents were also similar. The details of the new columns are presented in Table 5- The construction of the new column was similar to what is shown on Figure 11 excepting that the effluent collector was placed directly on table top and the 500-ml feed bottles could be moved up and down in order to adjust the hydraulic head. Before application of the chloride solution for pore-volume deter- minations, the distilled water remaining in the inlet tubes to the columns were drained until the meniscus just reached the top of the medium inside the column. The feed bottles, the influent tubing and the empty sections of the columns above the media were next filled with the chloride solutions before allowing percolation of the same. The pore volume obtained by following this procedure includes the volume of the underdrain systems and a correction had to be made to account for this. The net pore volumes of the media used in the columns have been reported in Table 5- The specific weight of the soils could then be calculated from the known weights of the materials used in the column and their corresponding pore volumes and the results obtained by this method compare very well with those obtained for the same soils and reported in reference 39. A similar procedure was followed for draining the chloride solutions 73 c E 3 O o +J 05 CO 2 0) LA .z 05 0) ■— -C -Q 4-1 03 1— M- irt 05 o 00 3 T3 t) C *-> A3 4-1 01 O C 05 ' — 05 a c ex. O •— +j 1/1 1/5 t/5 "O •— C 1/5 ro 1/5 1/5 — 05 4-J c .— c 4-J ui T3 3 c 05 to 1 — i/) cj ID 3 x> 03 c 4-J no 4-J in O c TO • — QJ O- C a O «— t-J 1/5 (/5 t/5 TJ 0— C Ifl nj X) CL Q. < lA PA CO GQ < 05 Cl r-. i->. O vO — CM ■— O CTi roOO O J" — CM — O 1^ o S» O \0 — CM — O O o r^ — cm — o r-« v£5 r^ o j- — CM — o r>> o r-. CM — E Ecm O O E — — O -C 05 j-j — aj cn-o 05 C J) 1- 1/5 05 E 03 C ■ — O 4- — » — — O 03 t/5 — C C 03 -C S- 05 1. jj d E 05 D. 4-J ■ — > 05 C a O O i— 1 vD CM la o O CM -3- o 00 LA -3- o o CM LA O O CM o CA LA -3" E CT5 4J .c 05 CTi rA CTi OO TD 05 4J C O — 05 i- 05 ■»-» s- E — O 3 E O — s — ' o 05 > */5 E c 3 T5 — — — 03 O 13 L. > — T5 <+- L. 05 05 U i_ T3 O O C Cj_ <4- 3 LA CM LA O -3- 1^ CM LA LA \£5 CM CM -o 05 03 C_> XI 05 CTi 4-J fA l_ O • Q.4- 05 05 C£. L. 0A 4- E — O O ^. 05 £ q. cn CO ~ — CM — o J- CM LA — VO CM O v£3 CA CM LA J- v£5 O 1^ J" 1_ 05 Q. j) 'e — i- o 03 05 ^ CT) XI C <4- C 13 O 4-J -a >-4- 03 4-1 • 05 — CT -C — tf» .— 4-J jO L. C 03 05 03 05 Q- 4-» E (fl s- ^ C 05 03 O Q. 03 O O CA o CA ^D vD O O — CA LA CA o CA CA CM LA CA CM rA CA O CM CA 1 03 i_ 05 a. ^-^. 14- 1/5 O *- D -D O O -C ■ ■■ «w^ L. 05 C O- E 3 05 — 4-J O 03 O E — 4- X 00 O M_ s- C O Q. O a.— n: < +- 1 Q- -d- cm o o CA LA LA o 00 r^ v£5 I — CA LA \£) LA OO LA c c 05 05 D 3 C 4- .- 05 7^ 35 from the influent tubes, etc., before application of the ABS solutions All the other steps of operations of the new set of columns were essentially similar to what has been outlined under Section 5-4.2 for running the earlier set of columns. 5 .5 Determination of the Cation Exchange Capacities of the Soils As mentioned earlier in Section 5.1, only three soils, Peoria clay, i 1 1 i te and bentonite, were chosen for this aspect of the investigation. The cation exchange capacity of these soils before adsorption of ABS were determined by the ammonium acetate method as described in Appendix A. 35 Each of these three soils were then allowed to adsorb ABS from solutions of C ]7 (pure) and C (pure) of different concentrations by setting up 12 different batches in glass stoppered bottles as presented in detail in Table 6. The bottles were shaken by hand daily, and samples taken for determination of ABS adsorption at intervals of several days after the date of starting. During sampling, about 10 ml of the turbid solution from each bottle were transferred into a 50 ml centrifuge tube, the aliquot centrifuged at 13,000 rpm for 10 minutes and then two samples of 2 ml each were pipetted onto planchets for determination of ABS adsorbed by the soils in a manner outlined in Section 5.2. Sampling from each batch was continued until equilibrium between adsorption and desorption on the soils was reached and maximum possible ad- sorption of ABS by each type of soil was obtained. The soils with the ABS adsorbed on them were then used for re-assessing their base exchange capaci- ties in a manner described in Appendix B. 75 CD CD a. i_ o < o CD Q. ID CD CD CO <4- o - co H— CO < , o CO •I*"— N <4- E O en * — 4J 3 , O UO *+- O a) CL > 1- u 1 — ° 4-> O 4J 2 o CD O O O O O O O O O O LTV LT\ LA LA LA LA LA LA LA LA LA LA CM CM CNi CNi CM CNJ CN) CNI CNI CNI CNI CNI O O O O O O O r^. r^ CM CM r^ r- CM CM r-^ r-. CM CM a) CD u CD CD L. CD CD CD l_ CD QJ 1_ cu CD a. 3 Q. CL a. Q. 3 Q. 13 Q. Q. 3 Q. 3 Q. Q. 3 Q. CM LA CM LA CM LA CM LA CM LA CM LA LA CM LA CM LA CM LA CNI LA CM LA CNI LA CN| LA CNi LA CM LA CM LA CM LA CM > >- ^ > CD CD CO CD (J O O O J-" CD CD CD CD CD CD c C c c CD CD CD CD • — ' — O 4-1 4-> 4-1 4-1 L. L, i_ s_ 4-J 4J i-> 4-1 « — • — ■— ■— O O O c c c c ■ 1 — ■ — • — CD CD CD CD CD CD (D CD 1 1 — ' — ' — Q_ Q_ Q_ Q- 0Q CO CO CO 1—4 ^H 1 — 1 i— 1 LA vD CX) cn o — 76 CHAPTER 6: EXPERIMENTAL RESULTS 6. 1 Adsorption Isotherms of the Soils by Batch Techniques 35 The amount of ABS adsorbed per gram of soil from solutions was calculated by equation (5.1) s Chapter 5. The concentration of ABS in the supernatant, C , in mg/1 could also be computed easily from the following equation, after the specific activity of the feed solution in CNCPM/mg and the residual beta activity of the clear supernatant were determined by methods outlined in Section 5-2. C' = (r^) 1000 (6.1) b a Afte r equilibrium had been reached between adsorption and desorption on soil surfaces, the amount of ABS adsorbed attained its maximum value and did not appreciably change with time, even though the soils were allowed to remain in contact with the surfactant solution. The concentration of ABS remaining in solution under such a condition would be the equilibrium concentrati on ,C . The amount of ABS adsorbed in u,g/gm and the corresponding equilibrium concen- tration of the ABS in supernatant in mg/1 are presented in Tables 7 and 8 and used for plotting the adsorption isotherms for the 22 different batches de- tailed in Table 3- The adsorption isotherms for the Ottawa sand, Mississip- pian sandstone, glauconitic sandstone, Peoria clay, illite and bentonite are shown in Figures 13, 1^, 15, 16, 17, and 18 respectively. 6.2 ABS Adsorption in Earlier Set of Columns (Nos. 1A through 6A) As has already been mentioned in Table k, Chapter 5, columns 1A, 2A and 3A contained glauconitic sandstone, Mi ss i ssi ppian sandstone and Ottawa sand respectively. The concentration of 1 -propyl dodecyl benzene sodium 77 X c X C CD CM XI 0) in XI c J o a E o o to CO < 13 Q- 1/1 ■> a <-A D CN 1 £J » — E c — ^» 3 (J "V. •— <>— C CT D l_ O E CT O n— UJ 1 a. 1" i_ C CT) o o >-*. m .— CT) x> 4-> 3. < X c — 0) -J "N, Li- o E ft -Q CJ •— E c — ■ — 3 u "^ ■— C CT 3 u O E cr <_> >—< LU 1 a 1* i_ c CT) o o \ in ,_ CT) X •4-1 n < X c . — 0J o s i — D u — •— •— C ^ 3 1_ O CT cr (-> E LxJ 1 Q. IT 1_ C CT) O o \ in .— CT) X (J 3. < X C . — OJ u ^^ (L) c CT) u_ o E l_> *■ — 1 • <— V ft— E c — ft— 3 o \ • — •— C CT 3 l_ O E ex o ■— * UJ i a. 1" u. C CT) o O ^v in 4 — CT) X 4-> n. < X C . — UJ u — CD c CT) Ll_ o ^ l_> 0) E '4- •— CO o O 2 Ol X X a> a) vO — — — OLALACM Cm J- CA vO LA LA CO -d" 0A00 OA — CM — o oo o — o oJ-r^j-o a. CL — CM in in OOO — CMLAOr*. o O O CM OA vO CM \D « • "O T3 • ■ ■ • LA O (D (1)00 O \D r^ N La — — OA O LA OA — ft- CM LA O CO o o — — Q_ Q. in in CM r— CM LALAO LAO oa n -d" o o oo — no — CM 0A LA CM O — CM ■d- csj O vO LA OA O VJD CM oo r-» \£> — — CM LA 1220. 2080. olaoolaolao OLAOOLAOLAO O LA O O LA O LA O — cm -4" vO -NJ-NOO-tT\ — cm -J- vO ~ CM -4 N O O — CM — CA vDJ-OOJ- I--.OO — rALALAfsl LA O vO -d" OA OA O vO -4 vO LA CM o ^ in -t vD I s . com — CO LTV OO^N-JO-^Jj — CM PA O O — CM -3" — — -4" CM OA oa O \£) oa CM — -d" 00 — oA-d" CO CMOO la vO — CM LTV r~- \OvOr--OOCMLALA cMv£>cr>r^ — j- o>4 — CM LA J" O O P»«. -4 — LA — r». — — CM -d" 1000. 1598. OOOLAOOLAO OOOLAOOLAO O O O LA O O LA O — N -J lO O J" 1^ LTl — cm -d - r»- - (M 4" vD O 4 1 r^LA — cm -d" ^ CM -4 vO o -d" r^ LA — CM -d" I s *" r^ oo uM — r«» r-» cm PAOvjO CT\ -d" LA— r>A OA — CO -J" LA CA oo r-~ ft— CM LA I"-- CM 00 CM — ft— (VN, 0--N400VDO — CA O — — oa-4 co r-^-d- — OA pa ■— oo o mroo — -4^00 — J- CM r^CMLA— CMCMOO CA LA — -d" — VO 00 — — OA LA CM r-. r-^ CM O00 O lO\D CM LA CT> OA O CT\ O — oA -d~ — OO O LA O LT\ O LA laooolaolaOla LA 00 O LA O LA O LA — OA LA 00 VJD O CM — OAvO O — OALAOOVOOCM — PA vjO O — OA LA 00 VO O CM OA \£> VO OA -d" \JO o cm cm N N^O N-J NO O OA 0A OA r^ CO viD 00 CM — OA LA 00 CM OA-d" oooo — — oa -d" O o o o o — — CM CM CM CM CAOO — — - raCO (^CT\ LA CM CA N IAO 0\04 N o CM LA— CA OA O 00 vO ■ — i — OA \J0 CM OA O cM CM O O -d" o LA CM O LA O — CM OA v£> VjD CM OA OA OA O — CM \Q) LA LA — O O O J" o r-^vOLALAOOOO -cr o P-- \D LA LA o o O O >— OA LA O NLA0O — — CA LA O — OALAOr^LAOA — — OA LA o — oA la o r-- LA OA OA LA AeiQ eijoaj 95! 1 II 8} juo^uag 78 in "O C 3 O a E O o 01 < O l/l CD a >4 LA CM O la .-EC — - 3 US — —CO" 3 L- O E cr o — ' LU a- "e l- c cn o O s in — en •o *j i < ^ T> C — (0 U S a) c en Li. o E (_> — _Q • •— — EC — — 3 U "V. — — c cr 3 1- O E cr o — • 1 .—^ Q- E l~ c cn O O S u> — cn "D +j "± < "-^ x> c — J) u\ CO c 01 u.. O E (_) "-^ I I .a • *— « — EC — j— 13 O — — — c cr 3 1- O E cr o ■ LU Q- "e 1- c cn O O S in •— cn t: w l < "-^ "O C — CD O \ a) c cn Ll O E O N -' I XI ■ ^— — Ec — — 3 CM 3 O C cr) O E <_> s — - 1 ^ Q- E w c cn O O \ in — cn tj +j i < ^ "O C — 0) o \ cu c cn Ll- o E a> — Em- — 03 O O Z CO v£> -3" O CM -J- OOvD o — co la md -d - ■ — cn — CO LA OOOlACNNlAUMN 1 — PO LA I s - LA MM LA — CM CO CO OLAOOLAOLAO — t\ j- so o - cn — cm Cn • UO O O LA O LO O 1 cm -j- r-. o o — cn o OM-tvO r^-oooo — COS© pues eMezuo LAOOOLALAOO CMJ-OOSooOO cm co -3" OOOLAOOLAO — N4"vD04-rviA J-J" COCM CM 00 -3" -4" O — cM-3-Lnocno colacm o r^r^co-d - ' — CO vO O • — 1 — CO — CM J" 00 LA00OLAOLAOLA O — CO LA 00 MO O CM — co vD N (M N^O WO O O OOOOOCMOO — CM OLALAOOOOLA CM lAO O la cn r~^ — ■ — CN CO LA Cn CO C^LALALAOOOO O ■ — COLAO r~-LAcO — — CO LA auo}spues uejdd 1 ss j ss jn vO N ro^D NroO O O — CMcoLAOOvO — CM CO -d-vo^t-d-oo-3- r^-LA — — COCO CM O LAOOLAOLAO -N^NOO-O — CM J" v£> r^\A)OCMLAOOO O — CMANO CT»MD CM co vO CM CM -3" O CM NmO — CM CO LA CA vO CM -3" — CO CM OOOLAOOLAO -CM^vDOJ-NCA -N4- N -3" coooocor^oo O — CMCOLAOO00 — CM CO la cn o rv r^vD vc la -J-NOCMOCO — cm -j- cn LAOOOLAOLAOLA O — COLAOOOOCN — C0\0 I — O CO O \£> O I ...... I — — — co o j- • LA O LA O LA O 1 cn cn co la 00 \o — CO LA CO LA I I LA LA O O O O II ...... 1 1 co la o r^ LA CO — — CO LA auoispues 01 } 1 uoonei 9 9 o p-. , l/l V, E CTl u E d) o „ •w ITS i/> o c yi o h-l ■nJ T3 ID C i_ fD ■u LO o c -? a> r«-\ .0 — _ cr en LiJ •— _ o ui6/6t/ 'S9V J° uoi^djospv 1/1 >- 01 TJ >~ -o fD XI LA r~« VD v£> L. O U M- o u- c a c l_ =5 U E u fc: .c o 4-J in " tfl -o <<-». c (U (U u. *— D JD a v -' — ' un un .— ^~ o CJ c o 1/1 -o c c T5 Q. Q. ui i/l U1 UI lu6/6T7 'sg\/ ^o uoiicUospv "I — r~ — r — r - ., p I _._. - — 1 \ | | ' lA >- '■ O ro ■ — i — O LP. TJ_ . — i — ^ \ in V en E en E PA E C / o «J L-' LJ E O >• i/i rrj — TJ < 0> , — ■ CC J- ■ — -Cf _c 0) oa - +j !_ — c C C C O «/» 3 1- E E E E in >- a o 3 3 _J 3 — f0 ' u_ ■ — ■ — i— ■ — TJ -o o O O o ,— . LAC — c a o o o TJ LA — 3 — 0) c r- CJ l_ cr Q) - 4_ M C - D CJ L. •* 1 — C\ 1 •_ " "TC ^T ~ L i | I 1 ' 1 ^1 <^7 -^s o o OJ c o 4-1 o in LA T> c ■ — T> s 00 en E o „ 4-1 in .— c c o o O •— o -cr 4-1 ro u • — 4-) c 9 CO 4 1! \ M- ' '. O . 6 < \ -d- 9 E ? _ 9 O / u 9 c \ ._ 3 JZ I en V _j O !5 s... x: V +j J* 03 9 o J- LA o CA Q. SL CD o i_ ~ vC O 4 4-1 i D 9 1 a -C i en i D i o 9 LA i- i i LP, - i i ? i i o i — 9 ■ i IT\ o o o c o (%) ^J,' 3 v . hr o LT\ cw LT\ CM O J" 87 O LA l- • 0) CM 4J 3 L. x: (%) J 3/ 9 -) v ^/ S J -rr 88 ? 4-1 OO -* t^ II QJ L. <4- XI . O 0) < L. vX) 3 Q. c ** E r> ^— C o E o , — "D O . — u u L. O o U- M- r- cn cn z> 3 O O i_ 1_ x: X 4-1 4-1 -^ _* ro TJ 0) tu '- o CD i- O --- 4~> f0 ._ 1- o — 4-J r &5 I- < N -' o o X (T> -4- s 6 CJ O o o co — - CD ro - i +j o > -' o o en c E 3 o o c E a) en E -n. CD en J- i u — ' c 1—1 CO — . en ro — ^v ■w o en O o i I- ^" a> w E »- +j en o c ^ c a) en t-H e 3- — E o en i/i — - CD C -M > ._ c — cu ■<-> lA 3 CD no — — CO 4- 3 CQ C E < >— I 3 <[ (_> LA CO CO- CO > < r- ^D LA in M d en O H E ^ -4 — O o ^ o CO CO — en co r-» vo r^. r»~ vo vX> r-^ — co — r-. n o -t moo o lt\ CO -4" CO — 00 O O CO \£> Cn CNl -4" CO — _ — — eg \£) O - J" N 00 VD CM CO -4" O CO — LA CM O -4" o o o o o o o nj- — oo lacm en — CO LA \0 00 O — COvO-J (N O O^N — CM CO J" J" LA - M ra-J LA\D I s - CD l_ _ 3 en E LA CM i— _) CO CM 0O Cn vO ro— co CT\ cn Cn oo oooooo r-»r-.r-«.r- 0O — — OOOOOCMvO O 00 CM — vOvor^oo r^OO LA— CMOCTV — NJ ^004- - rsj — CM CM ro J- -J- LA 0O roO ff\OD 04-4- o sjooLnoo- r-» — vo la — 00 cn cm rvrsvDvDvOvOvOvO OCOLAOvOvOCOLA P^-CM r^cM LAO LA CT\ — ro4" ^O rv (T\ O — OCO MAM30 CM N r-^ la -4" -4" roLALAro o ro LA -4- 00000000 rOvOCncMLACO — LA vOcMCOLA— r-^J-ro 00 cnLAJ- ro— 000 — CM ro J" LAvDvD CM en 3 a. en E CM M3 ■ — c_> co < -4- o ro o — en en en ro CO CM CM CM vo en en cm -4" vo en en cm cm — -4- — -4- o co vo en cm co ^O CO O CO CM \£) co O CM O 1 — co r-«. cm vO co co cm ro O -4" -4" LA 00 S4 O N0O vO — CM CM CM O -4" O — ro i-~- r-- v£> r»» — o ro LA -4- o o o o r^ ro vO en cm cm vO CM CO LA ro O — — CM -4" — CM CO -4" -4" r-. — N rA44 en \£> CM CO -4- CM CO LA — CM CO CO CD 1 — 1_ ^^ 3 en Q. E O CM 1 — CO O 1 — CQ 91 i < ' O o CO CD < o co < Cnl 4" o o O 00 fC . CA o v — O O Cn 4-> .> — s c E CD Cn E ^ CD CD o ^-- c oo LA O CM O O LA CM 1^ CA CO vO r-» -t 00 LAN CT*CT* 00'— 1^ ca — c\i(T\CTvCr> CM CM — CM CM — — — cc. | O MD 1" r— * cn to p— ■-v. 4-1 O en <_) i I— 1 — ■ ^^ CD CO E 1_ 4-J cn O c \ c CD CD — 1 E i o — .— — E • o cn +J CO - - C — a 3 id A — — CO 4- D CO C E < H-. D < in c_ > ID CO Q. CQ >- < h- — O O z: o CM la — 00 cm ro -J -J 4" -J CACACMCMCMCMCMCMCM LA ■4" 00 CNJ- (T\NOOO m <0 cA -3" ro^Mroro LA LA I — r^-. r~- cm -4 — mo 004-vD0004^)vDiA COO r-.— CM CA -4" 4" -4" (Ni LA vO l"~- CT\ — (ALA N CM ^ 00 iAIA— CM LA ca ■ — ca O^ cm O ■ — 0a CO 044 - LA-JON IA(J\'- CA v£> O^l CA -4 — — — — CM CM CM fflNrsO^D - CA CA 1^- CM tArA N-CO C0^4"Ov0caCAlA lA (ACM CM CM (AIA'- cmcolacm O^r-^oo — 4- v004 iaO^j-oo (Arv -'-'--MNrOM cMvO' — r^.r^-00' — caia vO -4" ca -4 -4 -t -4 -4" o vO vO -4" LA \£> OOOOOOOOcA l-^-d" — OO CACM CTvvOvO ■ — rA lA^D OO O ■ — CACM 00 vOJ" CM O CAIV iacM — CM CA 4 4 IA\£) I s - LA •4" CO — CM CA -4 LA^Or^OOOO Q. CJ1 E CM 0O < CM LCMAlACnLAN- LA CM CT\ 4 0O 4 - O ca ^ ^ CM CM CA4 LALA IA0O cm 4 vO CM44 CM VO LA4 IA h-OO CA o -4- OOOOOOOfA (AvO CT\CM IA0O — O vD CMCO LA— r~-J"\I> CO NiA4 CA' — OCA ' — CM CA -4 LA vO VO CA — CMCA-d"CA\£>r^C^ CD L. w— 3 "•"X o_ cn E CM v£> C_) 0O < CA IM ^) LC\ IA 4 I*-* CM — — . — r>- cx> cm — o o 00 ca — 00 MO CA-4" 0O -J - LA 00 cm ca-4" vO r^CA — \D N CA CA O^v o 00 lanvD r-- CA ' — CA LA •— CM CM — — — — CM LA CA LA CM O -4" LA 00 — LA 00 CA — — — CM la 00 cm r--oo -4" LA CM CA CA CM LA O O CM -4 00000— CA vO CA CM IAN vO CM OO LA — LA O — — CM CA00 — CM CA -4" LA LA O LA — CM CA-4" LA LA CD — CD CM LA -4" CM O -4" 00 VO LA ^O LA -4 CA CA CA ca — -4 vO 00 C^ CTi LA f-~. CM vO CA cm — cm — 00 Cn CO caNO ca-4" — — CM CM CM CA 0O CA CM CM ■ — CA LAN LA -4" I""*. N 00 -00 r^o 00 -4 J - CM CA — r^ ca — Ca 00 -4- CA O O vO LA 00 — CA-4 -4 LA LA r^vooooo cr> vo -4" CA MO CA cm la • — ■ — CM IALC\vD I s - ■4" CM ' — CM CA -J LA LA CD — =s cn 13 cn Q. E Q. E vO O LA CM ■ • — O 1 — CA O ■ — i_) ' — LA 92 -Q C OJ CO 03 3 A3 o < s - O o X trdj- co ■ • CO < o O o < O CO _ .'o- ro — a. .*-> o • O O ca I/) ■>-> ^-. c E ._ c — LA 3 05 CA — — CO M- 3 CO C E < »-< 3 < (_> ^O LA l/l C CT 3 3. o • E la ca CMCOCOCAOO CACMCM — — — CMCM -3" CO — LA O O CA CA CO CA oACO CA CO LA vO J- N M(Ti4- O (N N CM-d" LA LA N CA — — — J- CA -3" LAO CA O CO — CA LA -3" LAN- J-COOvD-d"vO'-LA CM — — — — CM vO t^oo o n n n vo ~3- I — CA — ca \£> O — — — — CM CM vO ca CA CM n o o ca -C|- ca — — CM CA -3" O CM LA oooooooj- r~-j" — CO LACM CA CM — CA LA v£> CO O — CO CO vO -3" CM O CA NoO — CM CA J 1 -3" LA LA — cm ca-3" la\x> n r-» 3 0_ CT) E co < CA o CA OCM CACO CO CO CA CA -3"CACMCMCMCMCMCM N^O LA — CA — v£> CM COW O MOO N-* N -d- lalalaca^t) caco CA LA I — CA — _j- n co C-- CA CA \£> CO CM LA vO COCA— CMLACONCM 4 O (TiO J-^)vD la CA CM — CM CM CM CM ' — -4 CM r-» j- caco co v£> Ml^' — vO • — -3" — — CM CM CA CA -3" CO LANIA(T\OCO VO CA CA CA -3" J" LACM O vO -3- LA OOOOOOO^O CA vO CA CM LA CO — CA v£> CM CO LA — r-»-4"LA co r^LAj- pa — o -3r — CM 0A J" LA VO v£> CA ■ — CM CA -4 LA\D Nl — - 3 CM O CT E CO 3 CO CM CM -3" vO r-». CA CM CA -3" co -d- co — ca r-- CM CA — — mO — o n — CO vO J" CM CO — CM — — — n — o vO CO CM O v£) — CM CM — O v£) -4 CA vO cm -3" r-» la O o CA O O O O — CA \£) CA CM r-. vO CM CO LA O O — — CM LA -N (A4 r-. — CM CA -3" -J" Q) — 3 CT ID- E vO LA — O O ■ — CQ O CA -O^DNCO -3" CA CM CM CM LA LA \0 CA v^) - lAvD NO 4 NO CON IAN04 N LA O — N CA — -d" — O CA -J" CA CA CA CO LA CM CM -3" CM VO N N N (T> r-» LA CM \*D LA — CM CA-4 LA vD\D IAO N n r-. r— _3" CA o o CA O O O O CM CA v£) CA CM CA — CM CA LA — CA v^> CA CM v£> ' — CM CA LA vO — CM CA -4" J" d) . — 1_ \ 3 CT) Q. E o CM c CA CQ vO 93 1200 000 en 800 en c o a o 600 -o < CO < *+00 200 C]2 (pure) in column 4A , Cf 8.6 mg/1- C|2 (pure) in column 4B , C f = 13.0 mg/1 Cjr (pure) in column 1A, C f = 8.2 mg/1 3 ~T- 5 Throughput Volume, liters Figure 22. Comparison of Adsorptions of C)c, (pure) and C]2 (pure) in Saturated Columns of Glauconitic Sandstones 9k I -r ! I — 1 1 "1 — 1 — 1 ' I i < CM * C \ £ A o / \ — < ° / V CD — LA c — \ cm C en \ c - E „ ^ E \ E — 3 CX CD \ 3 o /\ l. cm \ — 3 • \ o — ° o. CLOO fc. o — c — \ II \ i C \ — V. \ LA \ — en en \ . i <4- \ E ure) .6 m \ C_) C_> \ 0)\O \ "- \ 3 O Q.0O ii CM — 4- k CL — \ » \ LA C_> O - - CO / LA / C / E / 3 I o / o / c \ — cr> E (pure 13.0 n CSJ C_> CJ 1 I I 1 ! i 1 i ^^^ o\ r- vD i/i LA E 3 3 Cl CD I- CN o o O O LA O o -d- o o CA O o CM O o ui6/6t/ui uoj^djospy S av 95 ... , 1 1 « 1 1 1 ' 1 T I 1 < 1 I i vD C E < ca c O o E e ■— \ — ^ \ 3 o o cn \ — E \ C — •- s . en 3 CLOD II CM pure) 8.2 m <_> CJ \ " \ LA \ — u - cr \ \ 1 ca \ E \ 3 V o -X — 'pure) in 10.6 mg/ — / vO / c / ii LA E / 3 / ° / C -^ •- cn E pure) 13.0 T^. ii CM — <4- U c Cfc • - I ! ! 1 , 1 I ^-3 O C: 00 . O cr o s^ to o <|< E CD s en cn 00 — E c - — 3 O O •— — C - 3 S- O cr o d Q. c E i_. O - CD o — < "v. in ■M - CD X) i < W c c u ■v. c CD o E <_J '■ — ' ko LA CO ■o O- > < 1- tn co < ^^ E CD CD CDO ro O V r— CU N > cr < CO E -J- E CD o O • o UJ • — • ^N. co cr co E v — ' ,__ CO 0_ Q. O > la LA o CO o o o o o LA PA lA o o J- co o O CM OA O r^ LA CT\ vO vD vD LA vD CT\ CM OA O O O O O I J" I I ■ I » vjO i TJ O Q) CO "D Q_ ._ E O (D Q. O o CO CO Q_ — E O CO CL 1/1 UI o o o o o o o o o O LA LA LA CM CM v£> r-- r-- oo o vjO O OA — — o oo vO vD vO — CM LA LA r-~ cm vO r— v£> — CO CO L. !_ 3 3 Q. Q. 3 a. 3 Q. CM CM LA LA O OA OA CM o ai | uo;uag CM LA o o LA CM CTi O OA CTi cr» OA LA CM OA a> n la cn ' — OA O o o LA CM o o o o o o OA OA O O — v£> v£> CM CA LA CM LA vO O — OA — LA LA r^ cm vO —- vO — CO !_ 3 Q- a> co 3 a. 3 Q. O MD vO o 00 md 3}] I II CO CO D. — E O fD Q. ui 1/1 O CTi O CT> O CM O CM LA O VO O O LA 00 LA CM 0A CTl CM OA LA 3 Q. o CM O O LA O O CM — CM 0A J" CM CO CO 3 CL o oo oo oo o CM cr* cti o~\ LA O -4" VO LA — o j- o o o CM 00 J" j- -d - LA LA -J" r^ v£> — OA — oo r^ cm vO — ^O — CO 3 3 a. q. CM LA LA O O O O Ae[3 euoaj 102 The percentage reductions in b.e.c. of bentonite and Peoria clay are plotted as functions of corresponding amounts of ABS-adsorption in Figure 28. A straight line is fitted by the method of least squares (66) through the k points plotted from results of experiments on bentonite, and the following relationship between percentage reduction of b.e.c., p, and amount of adsorption, A (u.g/gm) , was obtained. P = 1 .59 + 0.42 A (6.2) 103 o o O o LA o 1/1 < < o c o E '"■ cr. c ■v. o CT • — -*- o ~ 3 < * o C c o 0) m • — en -^J TJ CL •»-> 1_ o <1) in o "D l- < 0) tT o CO (~ o rea of the solid under a monolayer coverage of adsorbate (see Section 2.10.2) and more particularly it is a measure of the potential capacity of the soil for adsorption, when a complete monomolecu la r film of the adsorbate is formed. In our case, it serves as a very useful indication about the relative adsorp- tive capacities of a particular soil for the k different types of a Iky 1 ben- zene sulfonates, as the values of, n, remain relatively constant. The adsorp- tive capacities of the different soils for a particular variety of ABS may also be compared on the basis of these values. Visual comparisons of the adsorption isotherms in Figures 13 through 18, and the Freundlich isotherms in Figures 29 through 3^ enable one to generally conclude that for any given soil: (i) C._ (blend) is adso-bed more than C,„ (blend), (ii) C ]2 (pure) is adsorbed more than C lr (pure), (, i i i ) C lr (blend) is adsorbed more than C . ,. (pure), and (iv) C,- (pure) is adsorbed more than C,„ (blend). The same con- clusion as above may also be drawn by comparison of the 'a' values for the Freundlich isothe-ms presented in Table 13 05 3000 ! ' MM E cn en oo < O C/l < 1000 00 ! . M? fpi re x 2. 15 ' lend) 3. Cl2 f lend) 4. r; 15 -lend) I I Equilibrium Concentration, tng/1 re 2 I. Freundllch Isotherms of r eoria 1 J-LL 00 06 4000 i — r 1000 E cn CD (/I < Q. !». O in < J I i I I I I I 1_ J I I I I I 1 10 Equilibrium Concent ra t i on , mg/1 J I L 50 Figure 30. Freundlich Isotherms of 1 1 1 i te 5000 1 1 — i — i i i i ooc E CD to < u- o o u o in "D < 00 Equilibrium r oncentr-.it ion , mg/1 re 31. Freundlich Isotherms of ertonite 10 : 00 I I I 1 1 1 — I — i — r i 1 — i — l I l I J E cn cn Q. o < 0.5 C|5 f blend) J 1111 J 1 I I 00 Equilibrium oncentration , ng/1 Figure 32. Freundlich Isotherms of Ottawa Sand 09 2000 Equilibrium Concentration, mg/ 1 Figure 33. Freundlich Isotherms of Hississipp Sandstone 2000 "i 1 1 — i — i — r -i 1 r T 1 f -i 1 r 1000 — E en en 00 a i_ o -a < Equi li ri loncentration, mg/1 Figure 3^- ' Mich Isotherms of Glauconitic Sandstone Ill x i/i 4-> >- CO M- O co CD _ *? (A c ' 1 — - ' — (- c (TJ Q o 1- 0) u co • — CO 1 — < co . h- — O co c (TJ l/l ' — o 0) co a. > > l— co (TJ co ■ — < <_> a> a >^ h- — o CO O r^ r-» CTi O 00 LA r^ r-. O CO J- CM 00 CM l-«- CM co CM cn O — o o CM o CM v£> co r«» en CO CT\ -d- vO CO . — , — P^. CM CM ■ — TJ TJ TJ TJ TJ TJ c C 0) C C a) 0) c C 'aT O <_> o c_> O- l/> C (_) QJ C o <_> O J-J c CJ 0) c o 4-1 O (TJ !/> CO 4-> o l/l 5 TJ .— .— 10 o TJ fD c 1/1 Q. TJ 3 c 4-> ro l/l c (D CD 4-1 i/i • — CD V) O s: i/l O r^ Cn LTV r^ ■4" vo r-^ o LA CA LA CA LA r-~ O v£> r^ — CM ON LA a\ — cn _ o ~ O o ■ ' O O o ■"" o r^ en On CM CM — CA O r-> 00 vO ^o LA CO i — co LA CO CA CM en v£» CM CM CM LA TJ TJ TJ TJ TJ TJ a) c C 0) 0) C C a) a) C C a) u OJ QJ i_ i— a> X) Q. Q. XI X) Q. Q. X) XI Q_ CM LA CM LA CM LA C\ LA CM LA CM LA C_) C_) fD i_ o 4-> i— ( O (_) O OJ 4-1 c o 4-J c CO <_> CJ 112 The fact that C . _ (blend) ABS having an average alkyl chain length of 14.2 carbons (62) is adsorbed more than C.« (blend), can be attributed to the lower solubility and higher molecular weight of the former surfactant (see Table 2, Chapter 4). The van der Waals forces of attraction between the ABS molecules and the soil surfaces increase with increase in chain length, and as such the C . _ (blend) with an average of alkyl chain of 14.2 carbons will be adsorbed more than will be the C.~ (blend) compound of shorter alkyl chains. Due to its low solubility the C , ,. (blend) is more hydrophobic and has a greater tendency to escape from the aqueous environment. The result is that this compound will be adsorbed more strongly on the soil surfaces, if other factors do not hinder the diffusion of the molecules towards the soi 1 -solution interface. In case of the pure alkylbenzene sulfonates, the C , ,. (pure) is seen to be less adsorbed than C . „ (pure) by all the soils, in spite of the fact that the former compound is less soluble and has a higher molecular weight. The following factors are considered to be the probable causes for the lower amounts of the pure C . _ benzenesu 1 fonate adsorbed: (i) Higher concentration of pentadecy 1 benzene sulfonate at the liquid-air interface. The energy of desorpt ion ,-A. , (see equation 2,6, Section 2.10.1) of a surfactant from the liquid-air interface increases at the rate of 640 cal/mole (33) (36) , as the alkyl chain is lengthened each time by the addition of one ' -CHL ' group. Also, 7^, becomes greater as the hydrocarbon chain in the molecules become longer due to increase in inter- chain cohesive forces (36). It is therefore evident that the energy of de- sorption of the pentadecy lbenzene sulfonate from the liquid-air interface is much higher than that of the pure compound with 12 carbons in the alkyl chains, and the diffusion of the C (pure) molecules toward the solid-liquid interface will be hindered more strongly. Consequently, most of the C 1£ -(pure) 113 molecules remain adsorbed at the liquid-air interface and only a small number of them could be adsorbed on the soil surfaces by the weak van der Waa 1 s forces, which became effective when the soil grains were brought in contact with surfactant molecules during shaking, etc. (ii) One of the important factors which influence the total amount of adsorption is the relative abilities of the adsorbate molecules to diffuse to the adsorption sites present on the surfaces of the adsorbent. The rate of diffusion of the pure C.,. benzene sulfonate will be slowest due to its bulkiest structure and as such it may be expected to be adsorbed in least amounts. This is in agreement with the postulations of Morris and Weber (11) (see Section 3-7.5) that the rate of diffusion of the ABS with shorter alkyl chain will be higher and as a result adsorption of a long chain ABS on activated charcoal is lower than the adsorption of the short chain ABS. The effect of the rate of molecular diffusion on adsorption will be more pronounced in case of clayey soils which have internal surface areas (39) . (iii) The pure C 1C compound will have a very low c.m.c, due 1 5 to its longer alkyl chain (4l) and higher unsymmetry in the molecular struc- ture (35) (^+7) . The c.m.c. of the C,,. (pure) may be as low as 1/23 of the corresponding figure for C ]? (pure) (Section 3-^.3) and as such, micelles will be present at practically all the equilibrium concentrations encountered (column 7 of Tables 7 and 8) for C . _ (pure) isotherms, while practically no micelle formation will start at the range of equilibrium concentrations of the C.- (pure) isotherms (column k, Tables 7 and 8). Micelles are not sur- face active (Section 3.k) and being negatively charged colloids they will not ordinarily be adsorbed on soils which also bear negative charges. Also when micelles are formed, the concentration of the surface-active ABS ions decrease (Figure 3) and consequently the adsorption of the same would also decrease. 114 Increased adsorption due to attachment of the micelles can be ruled out for this case from the following considerations. Adsorption of micelles is reported by some authors (see Section 3.7-3) and presumably such adsorption occurs only when the Coulombic repul- sion between micelle and soils is overcome and van der Waals forces take over. This may be possible due to contact of the micelles with the soil grains due to eddy diffusion created during manual shaking of the batches. But the weak van der Waals forces will be comparatively less in this case since the micelles are symmetrical and have minimum surface areas (vide Section 2.2.3). Consequently, even if adsorbed, the micelles would be very loosely attached to the soil grains and might be easily desorbed by collision with other micelles as had been reported by Void and S i varamakr i shnan (53). Davies and Rideal (36) have also mentioned that the probability of a micelle to bounce back upon collision with the soil is greater than the chance of its being attached to the adsorbent. Thus increased uptake of C , ^ (pure) due to micellar adsorption is hardly possible in this case. (iv) The smaller amounts of adsorption of pure C.,. benzene sulfonates may also be attributed to the phenomenon of steric hindrance (34) or steric effect (36). According to this theory, the bulky alkyl chains of ABS molecules adsorbed on the solid-liquid interface makes it difficult for the other molecules to approach the adsorption-sites close enough to be ad- sorbed by the weak dispersion forces. The effect is one of shielding the adsorption sites and the C (pure) molecules will demonstrate this effect to a greater extent due to their longer alkyl chains. Such effect was also observed by van der Waarden (57) in studying adsorption of alkyl aromatic compounds on the polar oxide films of carbon black. He concluded that alkyl chains block certain surface area and keep other alkyl aromatic molecules from attaching to the surface. The effect is stronger as the chains become longer . 115 While investigating the feasibility of using the activated charcoal for the removal of ABS frorr water, Morris and Weber (11) noticed that sodium 2-octy lbenzene sulfonate was adsorbed at a lesser rate than sodium 2-tetra- decylbenzene sulfonate. The increased rate of adsorption of the C,. ABS over the C,n ABS was attributed to the faster diffusion of the former surfact- ant. The results obtained by the above authors are in agreement with those obtained in course of this study, since the pure C ABS with longer alkyl chain was found to be adsorbed less than the pure C. 9 ABS by a 1 1 the soils used in the study. The work carried out at the Water Resources Division, U. S. Geological Survey, Denver, Colorado (5^) has shown that both the rate and amount of adsorption of ABS on Kaolinite are greater for C . _ benzene sulfonate than for C 9 benzene sulfonate, which is exactly opposite to the conclusion drawn from our study. C,_ (blend) is adsorbed more than C (pure) since the latter ABS is composed entirely of pentadecyl benzene sulfonates, and will as such have a lower c.m.c, slower rate of molecular diffusion, higher adsorption at the liquid-air interface and more pronounced steric effect. All these factors, which hinder the adsorption of C, q (pure) more than they hinder the adsorp- tion of C lq (blend), are really the effect of the bulkier alkyl chains of the former type of ABS. The C, ? (blend) had been adsorbed less than C, ? (pure), probably due to the slightly lower molecular weight and more particularly due to the presence of higher isomers (having alkyl chains of 13 to 18 carbon atoms) in the case of the former type of ABS. As mentioned in Section 3-3, a surfactant with twelve-carbon alkyl chain display maximum detergent property (which re- quires maximum adsorption of surfactant molecules on soil surfaces) due to the fact that there exists a proper balance between the hydrophobic and the hydrophilic groups of this compound. The surfactant having alkyl chains with 116 more than 12 carbons will be heavily weighted towards the hydrophobic side and become more and more insoluble at ordinary temperatures, with the result that these molecules wiH be increasingly adsorbed on the liquid-air interface. On the other hand, a surface active compound having a 1 ky 1 chains composed of less than 12 carbons will be heavily weighted on the hydrophilic side and become increasingly water-soluble and, as such, less and less surface active. Thus, a surfactant with a 12 carbon alkyl chain, such as the pure dodecylben- zene sulfonate, will have a maximum surface activity toward the solid-liquid interface. This explains why the C, 7 (pure) ABS is adsorbed in maximum amounts by the soils. The C,~ (blend) ABS was not adsorbed as much as the C , ,. (pure) ABS since the former compound contained only 46 percent pure dodecyl benzene sulfonate and the remaining 5^ percent (being partly C 7 , Co, C Q C, n and C.. benzene sulfonates and partly C,_, C,. , C 1£; , C./-, C ]7 and C l0 benzene sulfonates) will not be as readily adsorbed as the C._ benzene lo \ Z sulfonates due to reasons discussed above. On examination of Tables 7 and 8 as well as from comparison of the a values of the FYeundlich isotherms (see Table 13), it becomes evident that the three clayey soils were more adsorptive than the sand and sandstone and, of all the soils examined, bentonite is most adsorptive, while Ottawa sand adsorbs least. Amongst the clayey soils, bentonite adsorbs most, fol- lowed by i 1 1 i te and Peoria clay. The fact that bentonite proved to be the most adsorptive of the six soils tested can be easily understood on considering that the total (internal + external) surface areas of the solids, as determined by the gly- cerol technique (Table l) also decrease as their capacity to adsorb ABS de- creases. It is known that the higher the surface area of a soil, the more will be the interfacial energy of the solid-liquid interface which in turn will cause greater amount of adsorption of ABS in order that the said 117 interface may remain in the low-energy state. It is observed from the results of batch studies that Ottawa sand could adsorb nearly 6.2 pg/gm of C.~ (blend) from a solution containing 10.0 mg of ABS per liter of solution. This figure is higher than those ob- tained during some of the earlier studies in our laboratory (25) (32) when it was observed that the same sand could adsorb about 5 Pg per gram of C ]9 (blend) from a 10 mg per liter solution. The higher adsorption obtained in this recent investigation is due to a prolonged period of contact (35 days) of the soils with the solution compared to only 1 hour's soaking in the pre- vious study. Differences in temperatures under which the isotherms were run in the two studies might also be a factor contributing to the differences in the adsorptive capacities of the sand. It would be of some interest now to examine whether the maximum adsorptions, A , of C . ,. (pure), C. ? (pure), C.j. (blend) and C.^ (blend) by the different soils as obtained from the adsorption isotherms, exceeded or were less than the theoretical monolayer adsorption, A t in pg/gm. In order 2 to estimate, A,, it is necessary to first compute the area in cm , X, likely to be covered by 1 u.g of ABS from the following equation: x=^(io- 22 ) (7.0 23 In equation (7-0, N =Avogadro's number =6.03 x 10 molecules/mole and M is the molecular weight. The area of an ABS molecule, o° , is assumed to be 50 A (see Section 3-7-5) when a complete monolayer has been formed, and it will be the same for all the four types of ABS since the cross-sectional areas of the molecules under a condition of close packing are independent of the length of the hydrocarbon chain. The values of 'X' for the h different alkylbenzene sulfonates are shown in rows 2, 6, 10 and ]k of Table 14, The 18 o If) TJ < 4-1 c CO a> 00 o < l_ QJ E o_ D E c • — e — X (TJ ■1 2: < ^S ^ 4-> < < «* TJ CO C GO <0 -* < M- 1 0) O < ^— «q in - (TJ c in 1- O E •— L. 4-> at a x i_ +j o O m i/j TJ i—i < E 1_ O CD i_ >- H- (TJ i — X) O a> c c o •— 2: (TJ 4-> i — X (TJ O O 4-1 0) L c — o C 4-1 O in O TJ 3 C (TJ O — CO o c CO — CD Q- C Q. O — 4-> in ui in tj — c l/l fD i/l CO (TJ 2 TJ (TJ C 4-1 f0 4-> CO o (TJ — >- L- (TJ O — (U o O oa O O LA O 00 CM E IT) CM CO aj L. (TJ CD O 03 LA OO LA MD OO LA 00 LA OO LA VO oo LA ^o CO CT> i X TJ 0) L. > o o <4- TJ L, OJ Z3 u CO < o ■J- OA O LA CO O o o o LA O O LA CM CA O O LA OO E en \ cn ^^ (0 — 4-> o < c o - E c 4^ O OJ in l. TJ O (TJ OJ X o vO LA LA LA OA CM PA O CM O ^O CA ca cn o 1/1 TJ <-i (TJ E E ' D cr E^ X (D 2: o CM (sjnd) OA -3" 3 LA XI TJ OJ 1_ OJ > o o - oo CA LA CM -* CM r^ CM vO OA CM CO CM CA 00 CM J- o -d- LA CA O O O 00 LA -d- ^o CA CA Q. v£> CM o 00 O CM CM J" • — -^ , CA OA CM 1 1 1 1 CO CA 00 OA O r*« vO LA CA 1 CM CM OA CA ■J- LA 00 CM OA OA O O O O O vO CO O 00 CO O o CM 0A « CA J- 00 J- 00 -J- LA • -tf LA ~ O CM r— O CM LA . — . — CM O r-~ r ~~ J" • LA o o O O O o 0O O LA 00 O O • 00 CM OA J" -3" -3- 00 OA MD 0O OA CM " LA • ~ O ■ — •— CM J" ■— CM OA CM CM ^~ 1 vO "~~ r-^ o O o o CO O 00 00 r-^ CM o CM CM OA O OA 00 LA OA 0O 00 -tf i — CM CM J" CM 00 1 — J" LA TJ QJ CM E O OA TJ OJ E O OA TJ OJ ,— s C — ^— *s c _ ,<-N C E • — X E ._ X E .— cr (TJ 4J — cn (TJ 4J — cn (TJ 4-1 L. cr X •« l. cn -Q •- l. cn X QJ 3 O cn ^ N — 3. >- — ' =L >- — (TJ C (TJ C (TJ c i — 4 ■>o ■ — . — 4-> O 1 — 1 — 4J O < O < O < — c 4-1 ,-~v ^ c 4J ^-*x >« c 4-J *-«N o • O- ~vP X O • CL ->p -Q - CL •VP E C u , — . E c L- . — s E c W-~s O O E TJ O O E TJ O OE . — .— l/l O" O QJ . — .— in cn O OJ . — .— in cr O (TJ 4-1 TJ \ O 1_ (TJ 4-> TJ \ O 1_ (D +-> TJ\ O O Q (TJ O" . 0) a. (D cn t— QJ O O- nj cr • — l_ H > .— i_ 3 > • — '\_ =1 4-> O E ■— X +j E*— ' X O 4-1 O E?^ X QJ l/l u (J oj m D U - — u — - CO u c (!) - - 0) c o C 0) - ^^ - I 3 - 1 [ ! , ; 1 -d- O o l/> o o TJ i/-> s_ c o ■o ■ — < • — o o CO < o > • ■D 0) CD o l_ CD u i_ 3 CD en QJ U C • - to o < CD o < 0) .-» a -o • - QJ 1/1 o > o o . - C 1 n) 0> -o c 4-J 2 • — 13 ™ l_ in .jO OJ 0) 4-J C 4-1 C c .— — OJ in o i/i c O in I- 3 _o i- • — . — 0J 4- ro o XI — aj 'O 1- > L. o o o < O JJ ~— o V- OJ (0 E J- E - o 4-1 4-J in S_ L. . — _D 00 1/1 ■— C 1- 4-1 c -o c o O <0 c ro (0 OJ OJ < . £3 OJ en c -o E 4-J / \ C 1 1 1 o o i — i/i 4-1 • — 0) c »— 4-J 0) i_ • — o in E 4- 4-) OJ O C en o c E : E <4_ o 4-J 2 OJ X> 14- ^51 i\ -^^ c OJ O u - OJ c T3 > o ,\ r 3 s 0J o ■ — V s_ o 4-J : 0J i c I o OJ c u OJ x: en c 2 O CO E 13 !_ en fO jn •— 'J < — 3 o (J •— -J 4-J . 2 o £ "'*'-. OJ <— o u (/) < V u 33 T3 c/1 en d) 0) 0) c u , - rj L. 1 X _ E > o X o r. o u- o CL 4- ■0 "D (U O 4-< O 0) cn l/l I ') u_ - - ■ — c • — ; '. 0) 1/1 "D \fl <+- •— in c 0) QJ a) |< E - o e J l 1 "" 1 < u U J 1_ — O S3 cn C +j n si m / l< 3 °l 0) / 1 — 1 cn C QJ T) J-" SZ — O — x — 1 LU-I 4- o l/l i_ QJ 05 V. - . (U > — c_ L. O QJ a. T> c (5 QJ - — 1/1 ■ — U V— \ ■— 3 4- u O 0) — - • I • * - u_ -3 qj . 1_ .~ in ♦-> X U < T5 QJ - , . Lf_ u — E in » — »* • L. u C u 3 2: cn . c U - — -- 3 .- O L. _C 01 1— E t O L. •. cn ".' .13 4_l 4J cn c E — QJ c jr • — - C/1 4- r-^ ci 0) L. 3 cn 134 some more ABS molecules are adsorbed they might have remained either on those parts of surfaces where there was no exchange sites, or near the already adsorbed molecules to form 'islands' as suggested by Davies and Rideal (36). According to these authors the adsorbed film of molecules will consist of 'islands' up to several millimeters in diameter, when amount of adsorption is less than what is required for complete monolayer coverage. It is also be- lieved that the adsorbate molecules within the 'islands' will be held together by the van der Waals forces. Thus, if the additional molecules of ABS ad- sorbed are held near the already adsorbed molecules, then the size of the island will be increased and yet no other adjacent exchange spot will be blanketed due to the very wide spacing of the exchange sites. The result is that the reduction in base exchange capacities of these soils, having sparcely distributed exchange positions, will remain constant even though the adsorption of ABS on them increases up to as high as 242 u.g/gm. J .k The Methylene Blue Test It is felt that a comment on the methylene blue test should be made regarding the reliability of the results obtained by the above-mentioned test. The author had to perform about 200 methylene blue tests in connection with this investigation and only in a few instances could the results of tests on similar samples be closely reproduced. Presentation of all the data and discussion of the errors are beyond the scope of this report, but an idea regarding the inaccuracy of this method may be formed from the mag- nitude of standard deviations and coefficients of variations reported in Sections 5-3.1 and 6.2. The necessity of developing a better method of determination of ABS concentration is, therefore, of utmost importance. 135 7 , 5 Sanitary Engineering Significance This study has conclusively proved that a surfactant composition, comprising of 100 percent pure dodecyl benzene sulfonate is adsorbed more than any of the other three types of ABS examined. Consequently, the travel of the C ]7 (pure) ABS in soil formations will be retarded most. The use of the pure dodecy lbenzene sulfonate as the only surfactant fraction of package detergents for the reason that it can be retained by soil strata at maximum rate after being discharged with waste water, is not justifiable due to the high cost of production of this compound. In actual practice the surfactant used with the household detergent consists of a mixture of closely related isomers, and the ABS with a C._ side chain forms the highest single surfact- ant-component. If, however, the ABS with a C. alkyl chain is used as the predominant fraction in the mixture of surfactants, to be added to the com- mercial detergent, then the same could be more efficiently held by soil strata receiving waste waters containing detergent, as revealed in this study. But the pentadecy lbenzene sulfonate is heavily weighted towards the hydrophobic side, and as such it should be examined whether a composition containing this ABS as the major surfactant fraction will have a balanced wetting, penetration, dispersive, suspending and surface active properties as are demanded of an efficient detergent. The clayey soils, especially bentonite, are more adsorptive than the sand and sandstones because of the higher surface areas available with the former group of soils. The coarse Ottawa sand having very low surface area would also be somewhat effective in retarding underground movement of ABS, since it can build up mu 1 t imolecu la r layers of ABS on its limited surface area. It may be of some interest to examine the feasibility of applying bentonite in place of activated carbon for removal of ABS from waste waters conta in ing ABS . The bentonite, after it has been saturated 136 with ABS, may perhaps be disposed of by incineration, etc., to destroy the organic surfactant. On the other hand, incineration of used activated char- coal has the disadvantage in that the charcoal itself would be destroyed with the ABS during such operation. The forces which are believed to be responsible for bringing about dsorption of ABS on soils, are short-range and weak forces. Desorption of the ABS retained on the underground soils would thus easily take place with the inflow of detergent-free ground waters. A complete breakthrough of the ABS front through any width of soil would eventually take place and the only gain due to adsorption on soils would be time. a 137 CHAPTER 8: CONCLUSIONS The following conclusions are drawn from the discussions and results presented in the preceding chapters. 1. The adsorption isotherms of alkylbenzene sulfonates on soils such as sands, sandstones and clays can be fitted to the linear form of the Freundlich isotherm given by the equation: 1*1 1 ! loq — = loq a + — . loq a 3 m 3 n 3 According to Freundlich equation, adsorption of ABS per gram of soil is a function of ! a' and 'n'. The values of 'n' remained practically the same for all the isotherms obtained in this study, and as such the values of 'a 1 are indicative of: (a) the relative rates of adsorption of the k different a 1 ky 1 ■ benzene sulfonates by a given soil, and also (b) the relative adsorptive capacities of the different soils for the same type of ABS. 2. The pure pentadecy lbenzene sulfonate was adsorbed less than the pure dodecy lbenzene sulfonate by all the soils examined, due to the lower c.m.c, slower rate of molecular diffusion, more pronounced steric effect and higher adsorption at the liquid-air interface in the case of the former type of ABS. All the four factors considered to be responsible for the lower adsorption of C. (pure) ABS are in fact the effect of the very long alkyl chain of this compound. 3. C. (blend) having an average of 14,2 carbon atoms in the alkyl chains are adsorbed more than C,„ (blend) by all the soils, probably due to the lower solubility and higher molecular weight of the former type of ABS. 138 k. Pure dodecyl benzene sulfonate was adsorbed in maximum amounts by all the soils due to the fact that there exists a proper balance between the hydrophilic and the hydrophobic groups of this 12 carbon benzene sulfon- ate, and it is neither too soluble nor too hydrophobic to be adsorbed on solid-liquid interfaces. The C.„ (blend) contained only 46 percent pure dodecy lbenzene sulfonate, while the remaining 5^ percent consisted of closely related isomers of shorter and longer alkyl chains, which are not as greatly adsorbed as the pure dodecy lbenzene sulfonate due to their higher solubility and greater affinity for the liquid-air interface respectively. It there- fore becomes apparent that C . 9 (blend) should be adsorbed less than C._ (pure) , as has been observed during this investigation. 5. C,- (blend) is adsorbed more than the pure pentadecy lbenzene sulfonate in all cases, as the latter compound has a great affinity for the liquid-air interface (due to very low solubility), and is characterized by a very low c.m.c, a slower rate of diffusion and more pronounced steric effect 6. The isotherm studies have shown that equilibrium between ad- sorption and desorption of ABS on soil surfaces can be attained only after prolonged contact of the soils and the ABS solution. This is attributable to: (i) the weak nature of the van der Waals forces responsible for the physical adsorption and also (ii) the slow rate of molecular diffusion of adsorbate + molecules through the electrical barrier formed by the Na ions surrounding the soi 1 gra ins . This conclusion is in agreement with the observations of Wayman (30, Schwartz et al. (35), Ross (58), and Weber & Morris (69). 7. Results of column studies on the coarser siliceous soils and the pure compounds were in agreement with those of batch studies, in that 139 both studies showed that pure dodecy lbenzene sulfonates could be adsorbed more than the pure pentadecy 1 benzene sulfonates. 8. In batch studies, the clayey soils proved to be more adsorptive than the sand and sandstones because of the higher specific surface areas of the clay minerals. The adsorption isotherms have indicated that in case of clayey soils the adsorbed ABS could cover insignificant fractions of the available surface areas (0.2 percent to 0.4 percent for bentonite, 1.2 per- cent to 2.8 percent for i I 1 i te and 2.2 percent to k.8 percent for Peoria clay), while surface coverages amounted to as much as 31 to 71 percent for Mi ss i ssi ppian sandstones and from 9-8 percent to kS percent in case of glauconitic sandstones. Ottawa sand adsorbed two to three molecular layers of the blended ABS in batch studies, but in column studies the same soil was observed to adsorb 13 molecular layers of C. (pure) and 32 molecular layers of C ]2 (pure) . 9. None of the soil columns could be saturated with ABS during the tenure of the experiments. All soils removed maximum amount of ABS from that available in the first liter of ABS solution put through the column. The average rates of removal of pure dodecy 1 benzene sulfonates from the total quantity available in the influent solution were found to be approxi- mately 83 percent, kS percent and 30 percent respectively for the glauconitic sandstone, Mi ss i ssi ppian sandstone and Ottawa sand. The corresponding rates for the pure pentadecylbenzene sulfonate were about 3*+ percent, 21 percent and 18 percent respectively for glauconitic sandstone, Mi ssi ss i ppian sand- stone and Ottawa sand. 10. The results of the batch and column studies have indicated that the extent of ABS adsorption by a soil strata depends on: (a) amount of ABS available for adsorption, (b) equilibrium concentration, (c) amount of soil present and its surface area, 140 (d) flow rate and (e) static contact period. 11. ABS adsorption can substantially reduce the cation exchange capacities of soils. The reduction is directly proportional to the amounts of adsorption of ABS in case of soils containing high percentages of mont- morillonite clay mineral. Such reduction, however, remains constant in case of soils of low cation exchange capacity and not containing clay minerals characterized by expanding lattice structures. 12. Percentage reduction in base exchange capacities of the clayey soils were much higher than the percentage reduction of surface area of the soils due to ABS adsorption, which indicate that the adsorbed ABS films remain on the exchange spots in the form of 'islands'. As the amount of adsorption increases, the size of the 'islands' also increase without affecting the adjacent exchange sites which are very widely spaced. The reduction in b.e.c., therefore, remains constant even though ABS adsorption on i 1 1 i te and Peoria clay increases. In case of bentonite, the large ABS ions permeate some distance inside the basal planes of montmor i 1 Ion i te , and in doing so, they serve as barriers to the entry of inorganic cations into the basal planes in addition to covering up some exchange sites. Since about 80 percent of the exchangeable cations remain in the basal planes of montmor i 1 Ion i te , a large number of exchange sites lying at the central region will be made inaccessible due to the adsorbed ABS molecules. The effect will increase as ABS adsorption in basal planes increases and the percentage reduction in b.e.c. was proportional to the amount of ABS adsorbed . 13- This study has shown that the adsorption of pure dodecylben- zene sulfonate by soils exceeds the adsorptions of the other three ABS, namely the C . c (pure), the C 1r (blend) and the C 10 (blend). But, the use of 1 lb 1^ 141 pure dodecyl benzene sulfonate in package detergents is prohibitive due to the high cost of manufacture of this compound. Of the two blended materials examined, the one containing the pentadecy 1 benzene sulfonate as the major fraction (46 percent) will be adsorbed on soils more than will be the other compound with dodecy 1 benzene sulfonate as the biggest single constituent. Consequently, movement of C ]C . (blend) ABS in ground waters will be more retarded if this compound is used in package detergents in lieu of C,„ (blend). Nevertheless, the fact remains that adsorbed ABS of any chemical composition are loosely held by weak dispersion forces and it will be eventually leached out by detergent-free water. The use of C (blend) ABS in detergents will not therefore solve the problem of contamination of ground waters, wells, etc. 142 REFERENCES 1. Synthetic Detergents in Perspective , Technical Advisory Council, The Soap and Detergents Association (1962) 2. Compost News, "Detergent Cocktails in New York," Compost Science , Vo 1 . 2 , No. k (Winter, 1962) 3. Schwartz, A. M., Perry, J. W., and Berch, Julian, Surface Active Agents and Detergents , Volume II, Interscience Publishers, Inc., New York (1958) k. Feighner, G. C, "Structure of ABS Molecule," Jr. Am. Oil Chem. Soc . , 35 > 520 (1958) 5. Morrison, R. T., and Boyd, R. N., Organic Chemistry , Allyn and Bacon, Inc. , Boston (1961) 6. McGauhey, P. H., and Klein, Stephen A., "The Removal of ABS from Sewage," Publ [c Works Magazine , Vol. 92, No. 5 (May, 1961) 7. McGauhey, P. H., and Klein, Stephen A., "Removal of ABS by Sewage Treatment," Sewage and Industrial Wastes, Vol. 31 , No. 8, pp. 877 (August, 1959) 8. Abrams, I. M., and Lewton, S. M., "Removal of A 1 ky 1 benzenesu 1 f onate from Water by Chloride Cycle Anion Exchange," 138th Meeting of Am. Chem. Soc , New York (September, i960) 9. Chambon, M. , and Giraud, Andr'e, "Esperiments for Removing Anionic Syndets Discharges into River Waters," Bu 1 1 .Aced .Nat 1 .Me ' d . (Paris) 144 , 631 (I960) 10. Renn. Charles E., and Barada, Mary F., "Removal of ABS from Heavily Polluted Works," Jr. AWWA , 53, 129 (Feb, I96I) 11. Morris, J. C., and Weber, W. J, a Harvard University, "Removal of Biologic ally-Resistant Pollutants from Waste Waters by Adsorption," Paper pre- sented at the First International Conf. on Water Pollution Research at London, U.K. (1962) 12. Flynn, J. H., et al., "Study of Synthetic Detergents in Ground Water," Jr. AWWA ., 50, 1551 (December, 1958) 13. Compenni, Louis G., "Synthetic Detergents in Ground Waters," Water and Sewage Works , 210 (June, I96I) 14. "Ground Water Deserves Sanitary Protection," Jr. Johnson National Drl 1 lers (Nov. -Dec. 1959) 15- Klein, Stephen A., Jenkins, David, and McGauhey, P. H., "Travel of Syn- thetic Detergents with Percolating Water," First Annual Report, Sanitary Engineering Research Laboratory, College of Engineering and School of Public Health, University of California, Berkeley (December, 1 96 1 ) 143 16. AA.GSP Committee, "ABS and the Safety of Water Supplies," Jr. AWWA , 52, 786 (June, I960) 17- Deluty, J., "Synthetic Detergents in Well Waters," Public Health Reports , 75, (January, i960) 18. Lauman, C. W., Co., "Effect of Synthetic Detergents on the Ground Waters of Long Island, N.Y.," N.Y. Stare Water Pollution Control Board Research Report No. 6 (i960) 19. Schmidt, 0. J., "Significance of Detergents in Water Pollution Control," Publ ic Works , ]_2_, 93 (December, I96I) 20. Klein, S. A., Jenkins, D. I., and McGauhey, P. H., "The Fate of ABS in Soils and Plants," Paper presented at the Annual Meeting of the WPCF , Toronto, Ontario, Canada (Oct, 1962) 21. Gowdy, W. R., "Chemical Structure and the Action of Synthetic Detergents," Sewage and Industrial Wastes , 2_5_ s 19 (Jan, 1953) 22. Chemical and Engineering News, special report on "B iodegradab i 1 i ty of Detergents - A Story About Surfactants," Chemical and Engineering News , 4j_, 102 (1963) 23. U. S. Department of Health, Education and Welfare, Public Health Service, "Drinking Water Standards (Revised 1962)," USPHS Publication No. 956 , U. S. Govt. Print ing Off ice, Washington (1962) 24. Hopkins, 0. C, and Gullans, 0., "New USPHS Standards," Jr. AWWA , 52, 1161 (I960) 25. Ew'ng, B. B., et al„, '"Synthetic Detergents in Soils and Ground Waters," Sanita-y Engineering Series No. 8, Department of Civil Engineering , University of Illinois, Urbana, 111. (Jan, 1961) 26. Grim, R. E . , Clay Min eralogy, McGraw-Hi 1 '! Book Ha 1 1 , Inc. , New York (1955) 27. Schell., to. R., and Jordan, J, V . , "Anion Exchange Pure Clays," Plant and Soi 1 , J_0, 303 (1959) 28. Mitra, S, P., and Dharam Prakash, "Adsorption of Phosphates by Indian Soils (Kaolinite and Montmor i 1 loni te) at Different pH Values," Z, Physik, Chem. (Leipzig), 207 , 205 (1957) 29. Renn, Charles E., and Barada, Mary F., "Adsorption of ABS on Particulate Materials in Water, Sewage and Industrial Waste , 3J_, 85O (July, 1959) 30. Report on the Studies of Synthetic Detergents, News Quarterly , San i tary Engineering Research Laboratory. Col'ege of Engineering and School of Public Health, University of California, Berkeley, Volume XI, No. 3 (1961) 31. Wayman, Cooper H., "Adsorption of Anionic Detergents on Solid Mineral Surfaces," Short Papers on Geology and Hydrology, Article 117 s PP- 0137. U.S. Geological Survey, Washington (1962) 144 32. Bane-ji, S. K., "Effect of Biological Slime on the Retention of A 1 ky 1 Benzene Sulfonate on Granular Media, 11 Civil Engineering Studies, Sani - ta-y Engineering Series No. 10, University of Illinois, Urbana, 111 . (Jan, 1962) 33- Adamson, A. W., Physical Chemistry of Surfaces , Interscience Publishers, Inc. , New York (I960) 34. Wheland, G. W. s Advanced Organic Chemistry , John Wiley and Sons, Inc., New York (June 1949) 35- Schwartz, A. M., Perry J. W., and Berch, J., Surface Active Agents, Their Chemistry and Technology , Volume II, Interscience Publishers, Inc. New York (1949) 36, Davies, J. T., and Rideal, E. K., Interfacial Phenomena , A. P. Academic Press, New York (1961) 37, Glasstone, S., Textbook of Physical Chemistry , D, Van Nostrand Co., New York (Jan, 1950) 38, Moore, Walter J., Physi ca 1 Chemi stry , Prentice Hall, Inc., New York (1955) 39- Suess, M. J., "Effects of Soil Properties on che Adsorption of Alkyl Benzene Sulfonate," Sanitary Engineering Series No. 14, Univ. of 111. , Urbana, 111. (Jan, 1963) 40. Mysels, K. J., et al., "The Adsorption and Chromatography of Alkyl Benzene Sulfonates on Charcoal," Jr . Am. Oil Chem. Soc . , 39., 66 (1962) ^1 . Roy, Waldo Miner, Editor, "Surface Active Agents," Vol , XIVI , Art. 6 , 347 , Annals of the New York Academy of Sciences (March, 1946) 42. Keii, T., et al„, Dept. of Chem. Eng., Tokyo Institute of Technology, Tokyo, Japan, "New Plotting of the BET Method," Anal . Chem. , 33_, 1965 (1961) *+3 . Ermolenko, N. F., and Lemets, N. L . , "Dependence of the Adsorption on the Solubility of Organic Acids," Uchenye Zapiski, Belorus Gosudarst , Univ. im V.I. Lenia, Ser. Khem., No. 29, pp. 139 (1956) (In Russian) 44. Preston, W. C,, "Some Correlating Principles of Detergent Action," Jr. Phys. and Colloid Chemistry , 5_2_, 84 (1948) 45. ManKowich, A. M., "Micellar Molecular Weights of Selected Surface Active Agents," Jr. Phys. Chem, 58 , 1027 (1954) 46. Corrin, M. L. and Harkins, W. D., "Determination of the Critical Concen- tration for Micelle Formation in Solutions of Colloidal Electrolytes by the Special Change of Dye," Jr. Am. Chem. Soc. , 69, 679 (1947) 47„ Hartley, G. S., "Interfacial Activity of B ranched-Paraf f i n-Cha i n Salts," Tr ans. Faraday Soc. , 32, 130 (1941) 145 48 Cor<-in, M. L., and Harkins, W. D., "The Effect of Salts on the Critical Concentration for the Formation of Micelles in Colloidal Electrolytes," Jr. Am. Chem. Soc , 69, 683 (19^7) 49. Alexander, A. E., "The Structure of the Surfaces of Solutions," Trans . Faraday Soc. , 38, 54 ( 1 9^2 ) 50. Wayman, C. H., Robertson, J. B., and Page, H. G., "Surface Tension of Detergent Solutions," U.S. Geological Survey Prof, Paper 450-D , Article 179, PP. D 1 90 » Geological Survey Research (1962) 51. Corrin, M. L., et al., "Adsorption of Long Chain Electrolytes from Aqueous Solution on Graphite of Known Area and on Polystyrene," Jr. Col loid Scl . , 4, 485 (1949) 52. Meader, A. L. (Jr), and Fries, B. A., "Adsorption in the Detergent Process," Ind. Eng. Chem. , kk, No. 7, I636 (1952) 53- Void, R. D., and Si va ramakr i shnan , N. H., (Bangalore, India), "The Origin of the Maximum in the Adsorption Isotherms of Association Colloids," Jr. Phy. Chem. , 62, 984 (1958) 54. Personal correspondence between Dr. Ben B. Ewing, Professor of Sanitary Engineering, University of Illinois and Cooper H. Wayman, Research Chemist, Geological Survey, Federal Center, Denver 25, Colorado, Dated Oct. 1962 55- Cason, J., and Gillies, G. A., "Adsorption and Chromatography of Fatty Acids," Jr. Org. Chem. , 20, 419 (1955) 56. Meader, A. L., and Criddle, D. W., "Force-Area Curves of Surface Films of Soluble Surfaces Active Agents," Jr. Col loid Science, 8, 1 70 (1953) 57. Vander Waarden, M., "Adsorption of Aromatic Hydrocarbons in Nonaromatic Media on Carbon Black," Jr. Col loid Science , 6, 443 (1955) 58. Ross, Sidney, "The Change of Surface Tension with Time I. Theories of Diffusion to the Surface," Jr. Am. Chem. Soc. , 67 , 990 (1945); and "The Change of Surface Tension with Time II," Jr . Phys . Chem. , 49 , 377 (19^5) 59- Doss, K. S., Gururaja (Bangalore, India), "Alterung der Oberflaechen von Loesungen IV," Kol loid Zeitschrift , 86, 205 (1939) 60. Grim, R. E., et al., "The Mica in Argellaceous Sediments," Am. Mineralo - gist , 22, 813 (1937) 61. Berry, L. G., and Mason, Brian, Minera logy , W. H . Freeman and Co., San Francisco, California (1959) 62. Colgate-Palmolive Company, Research and Development Department, 1 05 Hud- son Street, Jersey City 2, New Jersey, personal communications with Dr , B . B . Ewing, Prof, of Sanitary Engineering, University of Illinois 146 63 . "Standard Methods for the Examination of Water and Wastewater Including Bottom Sediments and Sludges ," 11th edition, Second Printing, March 1961 , APHA, Inc. , New York 64. "Radiological Health Handbook ," U. S. Department of Health, Education and Welfare, Public Health Service, Bureau of State Services, Div. of San. Eng. Services, Robert A. Taft San. Eng. Center, Cincinnati, Ohio (Jan, 1957) 65. "Mathematical Tables from Handbook of Chemistry £■ Physics ," 11th edition, Chemical Rubber Publishing Co., Cleveland, Ohio (1961) 66. Chow, Ven Te, "Frequency Analysis of Hydrologic Data with Special Application to Rainfall Intensities," Bulletin Series No. 414 , page 29 , University of Illinois, Eng. Expt. Stn. (July, 1953) 67- Grim, R. E., "Ion Exchange in Relation to Some Properties of Soil- Water System," Sumposium on Exchange Phenomena in Soils, ASTM Publi- cation No. 142, 3 (June, 1952) 68. Hendricks, S. B., "Base Exchange of the Clay Mineral Montmor i 1 loni te for Organic Cations and Its Dependence Upon Adsorption due to van der Waals Forces," Jr. Phys . Chem. , 45, 65 (1941) 69. Weber, W. J., and Morris, J. C., "Kinetics of Adsorption on Carbon from Solution," Jr. San. Eng. Div., Proc. ASCE , 89_ (April, 1 963) 70. Davidson, D. T., and Sheeler, John B., "Cation Exchange Capacity of Loess & Its Relation to Engineering Properties," Symposium on Exchange Phenomena Soils , ASTM Publication No. 142 (June, 1952) 71. Ewing, B. B., "Development of the Design Criteria for the Underground Disposal of Radioactive Waste," Ph.D. Dissertation, Univ. of California , Berkeley, Ca 1 i forn ia (June, 1959) 147 APPENDICES Appendix A Procedure for Determination of Cation Exchange Capacities of Soils Before Adsorption of ABS by the "Ammonium Acetate Method" (70) (71) Step 1. A sample of about 30 grams of the soil to be tested was air-dried at 1 03 s C in an electric oven for at least 2k hours. 25 grams of this soil were than weighed accurately to 0.01 grams. Step 2. The soils were placed in a rubber stoppered bottle of 500 ml capa- city and shaken by hand with 500 ml of neutral IN ammonium acetate solution for at least 5 minutes. Step 3- The contents of the bottle were next transferred into a 400 ml beaker. The bottle was washed several times with IN neutral am- monium acetate solutions and washings poured in the beaker to ensure that all the soil particles are completely transferred from the bottle to the beaker. Step 4. The beaker was covered with a watch glass and heated over a steam bath for at least one hour with occasional stirring. Step 5. The contents of the beaker were then transferred quantitatively (by several washings of the beaker with IN neutral ammonium acetate solution) into a Buechner funnel and suction-filtered through Whatman No. 40 filter paper, Two filter papers were used for re- taining the fine soil particles. Step 6. The soil retained on filter was wetted with 5 ml of neutral IN NH-C1, sucked dry and the filtrate was discarded. 148 Step 7. The soil cake was next washed with neutral 70 percent methyl alcohol and the filtrate tested for chloride by adding few drops of AgNCL. This was repeated several times till the filtrate was found to be free from chlorides. About 150 to 250 ml of methyl alcohol was necessary to wash down all the excess ammonium acetate originally used. Pi st i 1 lat ion Step 8. Immediately following the alcohol washing the sample containing ammonium ion that has been taken up by ion exchange, is transferred to a 800 ml Kjeldahl flask. The transfer of the soil from the Buechner funnel is best accomplished by moistening the cake slightly and then rolling up the soil sample in the filter paper and trans- ferring the paper and sample into the flask. Soil grains clinging to the sides of the funnel were transferred by wetting a clean sheet of filter paper with distilled water and wiping the inside of the funnel clean; the filter paper was then placed in the flask. Step 9. About 5 grams of powdered MgO , 20 ml of phosphate buffer and few boiling chips were added to the flask and the contents diluted to about 500 ml by distilled water. Step 10. Prior to starting distillation of the sample, about 100 ml of distilled water was distilled through the Kjeldahl apparatus to wash out any ammonium ion present in the condenser. Step 11. The distillation of the actual sample was then started and a total of about 250 ml of the distillate collected in an Erlenmeyer flask containing exactly 50 ml of 1 percent boric acid. 149 Step 12. A blank sample was also run with 500 ml of distilled water and reagents and 250 ml of distillate collected similarly. Ti t rat ion Step 13- The Erlenmeyer flask with 250 ml distillate was removed after washing the boric acid off the receiving tube into the flask. Step 14, The distillate was titrated with IN sulfuric acid in an electric pH meter back to the predetermined pH of 50 ml of boric acid diluted to about 250 ml. The titrant volume was noted by reading the burette to the nearest 0.05 ml. Step 15. The distillate from the blank was also titrated similarly to determine the correction factor to be deducted from the titrant vol ume . Ca leu lat ions Step 16. The cation exchange capacity of the soil is calculated from the following equation: Cation exchange capacity in meq/100 grams (ml of IN hLSO, for sample - ml of IN H SO. for blank) x 100 Wt of soil in grams 150 Appendix B Modified Procedure for Determination of the Cation Exchange Capacities of Soils After Adsorption of ABS 19275 grams of ammonium acetate crystals were accurately weighed and added to the glass stoppered bottle containing soil and 250 ml of ABS solution after equilibrium between adsorption and desorption on the soils was reached. 19-275 grams of ammonium acetate was added in order to give a IN solution of the same. The bottle was shaken by hand for 5 minutes and thereafter allowed to stand for 2k hours during which period a few more manual shakings were done in order to ensure good contact between the ammo- nium ion and the exchange sites. The mixture of soil, ABS and ammonium acetate solutions could not be heated over steam bath as in step k of the procedure outlined in Appendix A since the adsorbed ABS molecules would be desorbed as temperature is raised above the room temperature, After 2k hours of soaking in ammonium acetate solution, the contents of the bottle were transferred entirely into a Buechner funnel and suction filtered through a pa i r of Whatman No. kO filter papers. Steps 5 through 16 of the procedure presented in Appendix A were then followed to obtain the cation exchange capacity after adsorption of the ABS molecules. One major difficulty which was experienced during distillation was vigorous frothing inside the Kjeldahl f laskand soil particles and froth tended to be carried over to the Erlenmeyer flask along with the distillate. An attempt was made to control the foaming by adding anti -foaming agents, but this was not successful in several instances. The rate of application of heat to the Kjeldahl flask was reduced and this step proved to be effec- tive when combined with the application of anti -foaming agent. Unfortu- nately some four samples were spoiled during distillation due to the above difficulties and before the method of controlling the frothing could be perfected . mm