University of Southern California Dissertations and Theses

Colloid properties of some New Mexico clays

(USC Thesis Other)

Colloid properties of some New Mexico clays

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content COLLOID PROPERTIES OF SOM E N EW MEXICO CLAYS A D isse rtatio n Presented To The Faculty Of The Department Of Chemistry U niversity Of Southern C a lifo rn ia In P a r t i a l F u lfillm en t Of The Requirements For The Degree Doctor Of Philosophy hy E. R. Harrington A pril 1940 c / This dissertation, written by ELDHSD..HAY HARRINGTON under the guidance of h.Xs.. Faculty Committee on Studies, and approved by all its members, has been presented to and accepted by the Council on Graduate Study and Research, in partial ful fillment of requirements for the degree of D O C T O R OF P H IL O S O P H Y Secretary Dean Date JyN E ^.1940 Committee on Studies Chairman i TABLE OF CONTENTS PAGE CHAPTER I INTRODUCTION................................................................... 1 Choice Of P r o b l e m ....................................................................... . 2 A cknow ledgm ents......................................................................... . 5 CHAPTER I I SELECTION AND PREPARATION OF SAMPLES . . . 7 P reparation ........................................................... .. ... . 7 San I l d e f o n s o ...........................................................................* 6 Kinney • • • ........................ 8 Tonque-Hed ................................................. 9 Tonque-Green . ........................... . . * ........................... 9 oarthage-1 . . . 10 Las Yegas . .......................... 10 Gallup • 10 Organ Kaolin . . . . . . . . . . . . . . . . . 10 Carthage-2 ...................................... 11 San Antonio ......................................................................... 12 Santa Clara .................. . . . . . . . . . . . . . IE A lb e rh ill . . . . .............................. 12 Gala ....................... . .........................................IS S. H. 4 IS La Bajada .................................... 14 i i PAGE CHAPTER I I I EKPERIMENTAL W O RK D O N E ..........................................15 Colloid Content Determined hy E xtraction of Total bolids . . . . . . . . . . * • • • • • * • • 15 Twenty-Four Hour P ip ette Method ............................ 16 Three Hour P ip e tte Determination . . . . . . . 17 Twenty-Four Hour P erco lato r Method .... .. 18 F ifte e n Day E xtraction Method ........................20 Hydrometer and Viscosimeter Tests ..................................... 21 Adsorption Determinations .................. . .......................... 21 Adsorption of Methylene Blue . . . . . . . . . 22 Adsorption of Malachite Green . . . . . . . . 23 Adsorption of Water Vapor • • • » . . . • • • 25 Adsorption of Ammonia . . . . . . . . . . . . 25 Base Exchange . . . . . . . . . . . . . . . . . . 26 Heat of Wetting . . . . . . . » * » . . . . . . . 29 pH Determinations . .• • • • • • • « • « . • • • 29 E leotrom etric T itra tio n ........................................ 30 E lectrophoresis 31 stu d ie s With the Ultra-Microscope ..................................... 32 Experiments on Sedimentation . . . . . . . . . . . 33 B a ttlin g According to S tokes1 Law • • • • . * 33 Use of the Sedimentation Balance • • • • . • • 35 i i i PAGE CHAPTER I I I (Continued) Thixotropic Phenomena . . . . . . . . . . . . . . 36 Liesegang Rings . . . . . . . . . . . . . . . . . 37 P l a s t i c i t y . . . . . . . . . . . . . . . . . . . . 38 Analysis of Clays . . . . # 40 Physical P ro p e rtie s of Clays Before and A fter f i r i n g . . . . . . . . . . . . . . . . . . . . . . 42 CHAPTER IV RESULTS AND THEIR INTERPRETATION . . . 48 Colloids E xtracted hy Total Solids . . . . . . . 48 Tests With Visoosimeter and Hydrometer . . . . . • 55 Adsorption . . . . * . » • . . . . . • • « • ♦ • « 58 Adsorption of Methylene Blue ...... , • . 58 Adsorption of Malachite Green . . . > . « . . 64 Adsorption of Water and Ammonia . . . . . . . 69 Base Exchange . . . . . . . . . . . . . . . . . . 72 Heat of Wetting . • . • • * « • « * • • • • « » « 76 pH Determinations 80 E lectrom etric T itr a tio n • * . . . . . . . . . . . 81 E lectrophoresis . . . . . . . . . . . . . . . . . 87 Studies with the Ultra-Microscope « . » » * * . • 89 Experiments on Sedimentation 92 iv PAGE CEAPT® 3Y (Continued! fiii^otropio Phenomena * • • » » » • • » * . • . . 98 I^asegar^ Rings , • • + »-•« . . . v . » * . • . 1 0 0 P l a s t i c i t y • * • » . . . • . • # PCS Analysis .of the Clays , « * . * .. » * . * ♦ * « * * ■ 104 P hysical Propart iss: of tii® Clays Before and A fter B iring * * . . . . . * •■>...* * « * • % . ♦ .♦ % ♦ 1G8 CHAPIW T CCNCnJBIQIJS , ■ . * v * * .. * . ♦ . . 125 bM i o c ® a p p t * • . : *-y. . ♦ f ♦ * ♦ * . . . . «la? Boofts « . . « § . « , * « » * # » , '♦ f , ♦ • * ♦ IS 7 B u lle tin s . . ♦ . « ♦ / * ♦ ♦ * ♦ . * * . ... * * . 129 V LIST OF TABLES TABLE EUM BER PAGE I Bata on E xtractions by T o tal Solid Methods . . . 50 I I Adsorption of Byes, Water, and Ammonia * * . . . 60 I I I Data on Methylene Blue Colloid Ratio . . . . » • 61 IV Samples Taken From Tables of the L ite ra tu re . . • 71 V Adsorption of Water, Ammonia, and Barium Chloride 77 VI Bata on E lectrophoresis . . . . . . . . . . . . 88 VII Bata on Stokes* Law Determinations . . . . . . . 94 V III Bata on F urther S e ttlin g of Clays . . . . . . . 96 IX Bata on Thixotropy of Clays . . . . . . . . . . 99 X Bata on Chemical Analysis of Clays 107 XI Bata on Clay Samples A fter Drying and F irin g . .. 109 XII Bata Summary 124 v i LIST OF PLATES PLATE NUM BER PAGE 1 S tr a t i f i c a t i o n of Sols and Liesegang Rings . . . 19 2 Drawings of Apparatus 28 3 Tyndall E ffect and Sedimentation Balance • • . • 44 4 P erco lato r and Fifteen-Day E xtraction Graph . . . 53 5 Twenty-Four Hour and Three Hour Graphs . * . . . 54 6 Graph of Methylene Blue Adsorption 63 7 Graph of Malachite Green Adsorption . . . . . . . 67 8 Graph of Water and Ammonia Adsorption .. .. . . 73 9 Graph of Barium Chloride A d s o r p t i o n ....................* ♦ 75 10 Heat of Wetting Graph . . * . . 79 11 E leotrom etric T itra tio n Curves 1:100 Ratio . . . 82 12 E lectrom etric T itra tio n Curves 1:5 Ratio . • . . 84 13 Electrom etric T itra tio n Curves 1:5 Ratio . . . . 85 14 E lectrom etric T itr a tio n Curves 1:5 R atio . . . . 86 15 Sedimentation Curves 97 16 Graph of Shrinkages on Drying and F irin g . . . . . 110 17 Weight Lost on F irin g and Water Imbibed by the Fired B r i c k . . , 113 18 Graph of Crushing Strengths of Fired Samples . . * 117 19 P ictu res of Fired Samples . . . . . . . . . . . . 118 20 Graph of Tensile Strengths of Fired Samples . . . 121 Chapter I IBTnODUCTIOH D efinition of Colloid Chemistry. Colloid chemistry is generally considered to be that subdivision of chemistry which deals with p a r tic le s interm ediate in size between those of a true so lu tio n and those of a true suspension. Workers in th is f i e l d of chemistry do not agree, exactly, on the size lim its between which the co llo id p a r tic le must f a l l . holmes1 l i s t s fiv e a u th o ritie s who give lower lim it values from 0.001 mu to 0.005 mu and upper lim it values from 0.1 mu to 0.5 mu. (One mu is equal to 0.001 m illim eter) p The United S tate s Bureau of S oils uses an upper lim it of 3 one mu . A hydrometer method much m use would give a value as high as seventeen mu as the diameter of the la rg e r c o llo id p a r tic le s . With v a ria tio n s going from one-tenth to seventeen mu as an upper lim it fo r the diameter of the co llo id p a r tic le (1) no lines, iiarry N. introductory Colloid Chemistry, p 2. John Wiley and Son 1934. (2) Eno, £’. H. Highway Subsoil In v estig atio n s in Ohio. Ohio S tate U niversity Engineering Experiment S tatio n B u lle tin Humber 39, 1928. pp 15-16 (3) Bouyoucos, George John. The Hydrometer as a Hew and Hapid Method of Determining the Colloid Content of S o ils . S oil Science Volume 23, 1927. p 319. £ the in v e stig a to r must make an a rb itr a r y choice to r him self. In accordance with th at n ec essity the w rite r accepted the 4 Jiilgard c la s s if ic a tio n . This c la s s ific a tio n defines as co llo id a l any clay p a r tic le s which w ill stay dispersed in a water suspension eight inches in height fo r twenty-four hours. According to Stokes* Law such a p a r tic le would have a diameter of about two mu. Choice of the Problem. As a substance containing co llo id a l m atter, clay has been studied fo r f i f t y years. During that time a number of methods for determining 5 c o llo id content have been proposed. Even so, in 1935 Dearie made the statem ent: "Much fu rth e r research is needed before a s a tis fa c to ry method of determining colloid content of clay is found” . Very lik e ly D earie's statement was inspired by noting the fa c t that many of the methods advocated showed e rro rs as high as three hundred per cent. Such r e s u lts could not be classed as q u a n tita tiv e and yet such claims 4 Hilgard, A. Estim ation of C olloidal M aterial in S o ils by Adsorption" t>. s . Department of A griculture B u lle tin Lumber 1195, 1924. p 4. 5 S earle, .albert B. Chemistry and Physics of Clays Ana Other Ceramic M aterials. Benn, London. 1933, p 251. 3 g were being made many times. In 1935 the w riter took twelve C a lifo rn ia clays and te ste d th e ir adsorption of dyes seeking to find a method of determining the co llo id content of the samples. He did not succeed in finding a q u an tita tiv e method,but he determined to in v e stig a te more f u lly the p ro p e rtie s of clays to see ju s t what c o rre la tio n existed between co llo id content and the many p ro p e rtie s which were supposed to be caused by th is c o llo id a l m atter. I t was found th a t the l i t e r a t u r e abounded in claims that c o llo id content was measured by adsorption of dye, water, and ammonia; by base exchange, shrinkage, p l a s t i c i t y , heat of w etting, brownian movements, the hydrometer and so fo rth . I t was noted that the clay samples taken were usually sim ilar in nature and in almost every case the experimenter subjected these sim ilar samples to only one or two t e s ts . The w rite r wished to conduct a great many t e s ts on a larg e number of d iffe re n t clays so that he might see exactly what a c la y ’s co llo id content had to do with adsorption, base exchange, heat of wetting, pH, i t s m igration in an e le c t r ic a l f ie ld , i t s p l a s t i c i t y , i t s ultim ate stre n g th Y/hen f ire d , and other p ro p ertie s a ttrib u te d to c o llo id a l m atter. He wished to extend his in v e stig a tio n of 1935 so that i t might take in 6 G-ile, P.L. with Middleton, H.H. .Estimation of C olloidal M aterial in Boils by Adsorption. U. S. Department of A griculture B u lle tin Humber 1193, 19B4. p 7. 4 more clay samples of g rea te r v arie ty , as well as a g reater v arie ty of t e s t s . The p o s s ib ility of finding a Q uantitative method for determining co llo id content was to be kept in mind at a l l times but th is was to be only a p a rt of the main problem of finding out what c o rre la tio n ex isted between co llo id content and the various p ro p e rtie s of the clays. The New Mexico clays were near at hand, varied in character and of local importance in building and in the native a r ts and c r a f ts . Accordingly the New Mexico clays 'were taken as the samples to be used in the t e s t s to be made and the problem became a study of the COLLOID PROP-LHTIES O L 3GM L N L 'a iv iiiX . ICO C LAI S. A CK N O W LED G M EN TS Acknowledgment of appreciation i s due the members of the supervising committee fo r th e ir help, often extended when such help was most inconvenient to give. Acknowledgments are also due the follow ing: Dr. John D. Clark of the Chemistry Department of the U niversity of New Mexico fo r the loan of necessary equipment and laboratory space; P rofessor W . A. Wagner of the same school fo r use of the te s tin g laboratory and permission to use the equipment of the Hew Mexico S tate Highway Department; Mr. L. G r. H etrick of the Denver F ire Clay Company and Mr. H. W . Reyland of the Coors P orcelain Company fo r equipment and assista n c e in s i l i c a t e rock a n a ly sis; El Toro Portland Cement Company fo r i t s o ffe r of the use of i t s laboratory equipment; Dr. C. E. Needham, P resident of the S tate School of Mines for a lik e o ffe r; the Kinney Brick Company for f i r i n g the clay saxaples; Dr. A. 0. Shepard of the Carnegie i n s t i t u t i o n of Washington fo r inform ation on pre-Spanish p o ttery ; the Colorado School of Mines Experimental P lant for some analyses of r e la te d clays; and the United S ta te s Department of A griculture, S oil Conservation Laboratory in Albuquerque for suggestions and l ib r a r y f a c i l i t i e s . 6 Special app reciatio n is due the la te P rofessor L. D. Roberts who, as a teacher of freshman chemistry in 1920, in sp ired fu rth e r study of the su b ject. Ten years l a t e r he was the in s p ira tio n of graduate study in chemistry and the chairman of the supervising committee of th is problem up u n t i l h is untimely death. His a ssista n c e and frien d sh ip over a period of twenty years w ill never be fo rg o tten . Special thanks are also due Dr. Robert R. Vivian who took charge of committee reo rg an izatio n a f te r the death of Professor Roberts. Chapter I I SELECTION AND PREPARATION OF SAMPLES P re p aratio n # Three C a lifo rn ia p o tte ry and stoneware olays and twelve New Mexico clays of various types were used* The f if te e n samples included kaolins, pottery clay, b a ll clay, f i r e clay, and adobe. The three C alifo rn ia clays were obtained in the ground s ta te but the others were obtained d ir e c tly from the p i t s . They were a l l given the same prelim inary treatm ent according to a method suggested by 7 'Wilson. The method is given below. The clays were soaked in water, s t i r r e d and pugged by hand. When a th in s lip was obtained i t was poured through a 100-mesh sieve to remove sand and other larg e fragments. The screened s lip was placed in large stoneware ja rs and a rev/ cubic centim eters of hydrochloric acid added to ac ce lerate the s e ttli n g of the suspension. A fter s e ttli n g from one to six days the c le a r, supernatant liq u id was siphoned off the top and the th ic k remaining s lip poured out in la rg e f l a t pans and dried in the a i r . When completely dry the clay was ground to a coarse powder and experimentation began. Sinoe in the discussion which follow s, the various clay samples v/ill be re fe rre d to many times, a d e sc rip tio n of each sample w ill 7 Wilson, Hewitt. Keramio K iln s. Denver F ire Clay Company B u lle tin Number 560, p 6. be givbn. The order of p resen tatio n is taken, vdth one exception, to coincide with th e ir c o llo id a l content order. (1) San Ildefonso. This clay came from the south p art of the Indian v illa g e of San Ildefonso, th ir ty m iles north of Santa Be, Hew Mexico. The sample was taken from a p i t being used by the Indians for adobe to make sun-dried b rick . The clay is chocolate brown in color and is a ty p ic al adobe. I t s e t t l e s ra p id ly in water and shows the presence of considerable calcium carbonate when tre a te d vdth hydrochloric acid. The v illa g e from which the clay comes i s famous fo r the making of p o tte ry . The San Ildefonso p o tte rs obtain th e ir clay from pockets in the canyons to the north and west of th e ir v illa g e , tempering the ric h e r clay with adobe from th e ir ir r i g a ti o n d itc h es. This clay sample i s the most arenaceous of a l l the samples te sted in th is rep o rt. (2) Kinney Brick. This clay is a gray-brown in color and, unlike the preceding sample, i s a slow s e t t l e r in water. I t is a ty p ic a l Rio G-rande flood p la in adobe. This riv e r c a rrie s a tremendous amount of s i l t , the so lid content of the 8 water sometimes r is in g as high as twenty per oent. The r iv e r meanders over a narrow flood p la in and throughout the v alley th ere is an abundance of water-borne adobe. This fa c t is of great lo c a l importance in the building of adobe houses. 8 Bakin, henry M . with Brown, Carl B. S ilti n g of R eservoirs. U. S. Department of A griculture, Technical B u lle tin Humber 524, 1939. p 93. 9 The Kinney adobe is round fiv e m iles south of the c ity of Albuquerque# The clay is quite sandy ana contains considerable 9 10 calcium carbonate,-both being c h a ra c te ris tic s of adobes. * (5) Tongue Hed Adobe. -This clay i s a d u ll red and a rapid s e t t l e r in water, i t is obtained from the Tonque Brick P lant th ir ty - f i v e miles north east of Albuquerque. This sample comes from across a range of mountains from the Bio Grande v alley and i s not a v alley alluvium but ra th e r to be classed as a product of upland d i s i n t e g r a t i o n . ^ This adobe is quite high in calcium carbonate and has given trouble in the making of brick, hence has been mixed with the nearby green shale mentioned in the next paragraph. (4) Tonque Green Shale# This is a d u ll greenish shale of 12 Cretaceous age outcropping t h ir ty - f i v e m iles north east of Albuquerque. This shale contains some sand and a few s e le n ite c ry s ta ls but i t s calcium content is much le s s than 9 S alinas, L. Salazar, k in e ra le s B esultantes del A lteracion de Rocas o de^ffenomeftos de Metamorphismo. Anales d e l I n s t i t u t o de Geologia de Mexico. Tomo 4, 1959. p 81. 10 Compilacion. A nali^is Iiechos^ en e l Laboratorio de Quimioa del I n s titu t o Geologico de ivlexico. Parergones del I n s titu to de Geologia, Tomo 5, Burnero 4, 1913. ps. 182-189. 11 P irrso n , Louis V and Knopf, Adolph. Books and Bock M inerals, h iley , 1926. pp G4b-846. 12 Barton, H.H. Bed Beds and Associated Formations in Kew Mexico. U.3.G.8. B u lle tin Humber 794, 1928. p 102. 10 th at of the nearby red adobe with which i t is mixed for making brick. (5) Carthage-1 . This is a dark blue-gray clay taken from above the coal beds at Carthage, New Mexico, some twenty m iles southeast of Gocorro. The clay i s dense and compact, breaking with a conchoidal fra c tu re . The dense mass d is in te g ra te s re a d ily in water and contains considerable clay substance, very l i t t l e of i t being reta in ed on a 100-mesh sieve. No fre e calcium carbonate i s present as the clay shows no reactio n with hydrochloric acid. (6) Las Vegas Brick. This clay, or shale, was obtained two miles southwest of the city of Las Yegas, New Mexico, 15 from an outcrop of the Graneros shale . The shale i s gray to a gray-green in color, ra th e r sandy and somewhat b r i t t l e . Gome calcium carbonate i s present. (7) Gallup P ire Clay. This dark gray f i r e clay is found between two coal beds a mile southwest of the town of Gallup, New Mexico. The clay does not d is in te g ra te re a d ily when soaked in water. No calcium carbonate is indicated when acids are added. (8) Organ Mountain K aolin. This sample was obtained from the Torpedo mine at Organ, f if t e e n miles east of Las Gruces, New Mexico. The deposit is in th ic k seams and large 15 Talmage, s te r l in g B. with wootton, T.P. The Non- M etallic Mineral Hesources of New Mexico and Their economic Importance. N.M.. Bureau of Mines B u lle tin IB, 1957. p 64. 11 irre g u la r masses which Dunham-^ believes to have been formed through action of ascending acid sulphate solutions on feld sp ar and s e r i c i t e . The kaolin is white to gray in 15 color and in some places is stained with what Bentley id e n tif ie s as s tu b e lite , a hydrous copper-manganese s i l i c a t e . On soaking the kaolin d is in te g ra te s into smaller pieces but does not powder re a d ily . In true kaolin fashion th is sample s e t t l e s out of water read ily and an addition of acid shows no fre e carbonates. (9) Carthage-2. This clay was obtained from the old ra ilro a d cut near the remains of the abandoned coal camp of Carthage, mentioned above under (5). The clay is about f o rty fe e t th ic k and of a variegated purple color when a fresh surface is exposed. The samples break with a rough and sometimes conchoidal fra c tu re and f a l l apart re a d ily on being soaked in water. The clay s lip formed with water passes read ily through the 100-mesh sieve leaving no more than one per cent of sand on the screen. Acids give no in d ic a tio n s of calcium carbonate. 14 Dunham, Kingsley. The Geology of the Organ Mountains, new Mexico State Bureau of Mines B u lletin No.11, 1955, p 154. 15 Bentley, L. B. Personal Verbal Communication, November 1, 1958. 12 (IQ) San Antonio T i l e * This sample was obtained from the abandoned t i l e p lan t at San Antonio, ten miles south of the town of Socorro, new Mexico. I t is a bright red clay about ninety per cent of which passes through the 100-mesh screen when mixed with water. This is an a llu v ia l clay from the Eio Grande flood p la in . I t i s p la s tic and stick y and contains a large amount of calcium carbonate,- th is fa c t being the cause of the t i l e p l a n t ’s abandonment. (11) Santa Clara P o ttery . The sample was obtained from a canyon a mile west of the Santa Clara Indian Pueblo, t h i r t y m iles north of Santa Fe, and was dug from a p i t which the Indians had been using fo r p o tte ry clay. The region is one of T e rtia ry lake beds consisting of fria b le sandstone and arenaceous sh ales. The clay is dense and brown-red in color, breaking with a conchoidal f ra c tu r e . I t contains considerable sand and shows no fre e carbonates when tre a te d with acid. The clay is quite p la s tic and is much used fo r the making of Indian p o tte ry in the surrounding v illa g e s . (12) A lb e rh ill. This is one of the C a lifo rn ia clays used as a check. The sample was obtained from the Tudor P o ttery on Fast F ifty-E ight. S tre e t, Los Angeles. I t comes as a lig h t blue-gray "rock" that needs considerable soaking before a s lip can be made. Once the s lip is made, hov/ever, almost a l l of i t passes through the 100-mesh screen. The clay comes from a p i t south of H iverside, C alifo rn ia, 15 where i t is associated with a small deposit of l i g n i t e 16 coal . The clay has a f a i r p l a s t i c i t y and burns cream to buff, i t is used fo r a r t t i l e , stoneware, and r e f r a c to r ie s . (15) Gala or Lincoln Clay. The sample was obtained from the St. Louis Brick and Clay Company of Last Blauson Avenue, Los Angeles. I t has a lig h t cream color and a d is t i n c t soapy fe e l, a c h a ra c te ris tic which stays with i t while i t is wet and even to a c e rta in extent a f te r i t is f ir e d . The clay 17 comes from near Lake Tahoe and is shipped to many p a rts of C alifo rn ia. I t is p la s tic and forms a s lip re a d ily , being much used for p o tte ry , stoneware, t i l e , te r r a - c o tt a and various r e f r a c to r ie s . (14) S. H. 4. This is a C a lifo rn ia clay obtained from the 8 t. Louis Brick and Clay Company mentioned under (15) above. Like number (12), the A lb e rh ill Clay, th is sample 18 comes from near Riverside , also being associated with Eocene l ig n i te coal. The S. H. 4 clay is a dense, fin e grained m a te ria l of a blue-gray color, streaked with organic m atter and m ottled with white and pink spots. The sample 16 D ietrich , Waldemar Benn. The Clay Resources and the Ceramic Industry of C a lifo rn ia. C a lifo rn ia S ta te Mining Bureau B u lle tin Lumber 99, 1928. p S87. 17 Ibid. pp 18-19. IS Tbid. pages 167 and 275. 14 had to be soaked lo r some time before i t could be poured through a screen. I t i s ra th e r p la s tic and is rated as a b a ll clay on account of i t s bonding stren g th . Like the other two C alifornia clays mentioned in the two preceding paragraphs, th is one shows no calcium carbonate and lik e them i t is used in making p o tte ry , stoneware, te r r a - c o tt a and r e f r a c to r ie s . (15) La Bajada. This clay was obtained from half way up old La Bajada grade, twenty m iles southwest of Santa Fe, Dew Mexico. I t is a b rig h t red clay found in numerous pockets beneath a la y e r of volcanic ash. I t is hard and dense, breaking with a d i s t i n c t conchoidal fra c tu re . I t takes a g reat deal of soaking to form a s lip but passes re a d ily through the sieve \Yhen the s lip is formed. The clay s e t t l e s very slowly and gels e a s ily . I t i s very p la s tic and i t s shrinkage is larg e hence the Santo Domingo Indians who have used i t for p o tte ry , temper i t with the volcanic ash found above the clay. The sample does not contain free calcium carbonate. Chapter I I I EABmrl 1 tlEN I aL V j Q H K DONE In the d esc rip tio n of the experimental work which follows an attempt has been made to follow the f if te e n clay samples through a number of t e s t s . The te s ts made and the techniques employed are given b r ie f ly and more a tte n tio n is given to the r e s u lts and th e ir in te rp r e ta tio n . The experimental work i s taken up under the headings of; (1) Colloid content determined by ex tra ctio n of to ta l so lid s; (2) Tests with viscosim eter and hydrometer; (3) Adsorption of dyes, water, and ammonia; (4) Base exchange p ro p e rtie s; (5) Heat of w etting; (6) Determination of pH; (7) E lec tro m etric t i t r a t i o n and buffer actio n ; (8) E lectrophoresis; (9) Tests with the ultram icroscope; (10) Sedimentation stu d ies; (11) Thixotropy; (12) Liesegang phenomena; (13) P la s t ic i ty ; (14) Chemical an aly sis; (15) Physical p ro p e rtie s of the clays before and a f te r f ir i n g . (1) Colloid Content Determined by E xtraction of Total s o l i d s : In any determ ination of co llo id m a terial the only absolutely sure way of finding the to ta l amount would seem to be by ex tra c tin g the c o llo id a l m aterial and weighing i t . If the experimenter uses some in d ire c t method he is always open to the suggestion that perhaps the method is not valid 16 or tiiat through i t s use he may have missed some of the c o llo id a l m a te ria l. In succeeding te s ts When co llo id determ inations are made by adsorption, e t c ., i t w ill be necessary to have some standard to check against and the only absolutely valid one would seem to be that of ex tra ctin g the c o llo id a l m aterial and weighing i t . 19 liilgard c l a s s i f i e s as a co llo id those p a r tic le s which f a i l to s e t t l e out from an eight inch water column in twenty- four hours. Some standard had to be taken so th is one was used, though lik e a l l other standards of size i t was purely an a rb itra ry one. lour separate types of determ inations were made on the samples as follows: (A) Determination of M aterial in Suspension Twenty-lour Hours by P ip ette Method; (B) M aterial in Suspension Three Hours by P ip e tte Method; (C) M aterial in Suspension Twenty-Pour Hours by P ercolator Method; (D) A F ifte e n Day E xtraction of C olloidal M aterial. (A) Twenty-four hour suspension by p ip e tte method. Five gram samples of clay were placed in four hundred cubic centim eters of d i s t i l l e d water, a few drops of ammonium hydroxide added and the mixture churned by a s t i r r i n g motor fo r twenty minutes. The suspension was then allowed to stand fo r twenty-four hours. A fter s e ttli n g a one hundred cubic 19 Hilgard, A. Estim ation of Colloidal M aterial in S o ils by Adsorption. U. S. Department of A griculture B u lletin Humber 1195, 1924. p 4. 17 centim eter sample was withdrawn from the center of the supernatant column by means of a p ip e tte . This sample was evaporated down in a s i l i c a dish of known weight, dried at one hundred and fiv e degrees Centigrade and weighed. Four times th is value was taken as the amount of m aterial l e f t in suspension twenty-four hours. This is the common p ip e tte method used by s o il s c i e n t i s t s . The ammonium hydroxide was added in accord 'with the recommendations of A nderson^. Sodium hydroxide is sometimes used as a d ispersing agent but one h e s ita te s to use such a substance which can re a d ily re a c t with fin e ly divided s i l i c a or alumina. (B) M aterial in suspension three hours. The amount of m a te ria l remaining in suspension three hours was determined in the same fashion as th a t fo r the twenty-four hour determ ination, the only d ifferen ce being the time the suspension was l e f t standing. I t was noticed in the suspensions th a t on remaining standing for a few hours they often s t r a t i f i e d into bands almost lik e Liesegang rin g s. (See P late 1). From the differen ce in opacity of these rin g s i t can be seen that the concentration of so lid m a terial one place is quite d iffe re n t from that in another. I f, then, 20 Anderson, k.S. /idsorptlon of Colloids and Non- Colloid S o ils. U. S. Department of A griculture B u lle tin Dumber 1122, 1922. p 5. 18 a person were to p ip e tte h is sample from a th ic k rin g and m ultiply th a t by four he would get a r e s u lt that was too g reat. On the other hand i f he should withdraw the sample from one of the thinner rin g s he would get low r e s u lt s . I t is a noticeable fa c t th a t two samples which have seemingly had exactly the same treatm ent may sometimes s t r a t i f y in a d iffe re n t manner so the experimenter could not always expect to check him self on duplicate determ inations. In order to get around th is s t r a t i f i c a t i o n i t was thought necessary to use some means where a l l the supernatant liq u id was withdrawn and not ju st a p a r t. Method C was designed for that purpose. (C) M aterial in suspension twenty-four hours extracted by the p erco lato r method. The fiv e gram clay sample was churned with d i s t i l l e d water and ammonium hydroxide exactly as before. A fter churning, the mixture was poured into a p in t p erco lato r provided with an inverted loop glass tube as shown on P la te 2. The inverted loop tube was attached to a rubber hose f i t t e d with a pinch clamp. A fter the suspension had s e ttle d fo r twenty-four hours the pinch clamp was opened and a l l of the supernatant liq u id drawn o ff. The whole four hundred cubic centim eter ex tra c tio n was evaporated to dryness and weighed. This method was considered to be more accurate than the p ip e tte method since i t avoided the hazards of s t r a t i f i c a t i o n in the suspension. 19 P late 1 ■ Figure A Figure B Figure A. s t r a t i f i e d banding in clay sole a f te r s e ttli n g fo r twenty-four hours. Samples are in p a irs reading l e f t to r ig h t: Tonque-Green, Gallup, and Las Vegas Figure B. Bhythmic banding or Liesegang rin g s formed by various re ac tiv e substances d iffu sin g through gels of s i l i c i c acid. 20 invert the percolator method is open to c ritic is m , however. Colloid p a r tic le s are known to be aggressive adsorbers so very lik e ly they might adsorb themselves to some heavy p a r tic le which would carry thou to the bottom. I f so, we would expect the percolator method to give us low r e s u l t s . To elim inate th is e rro r in the percolator method i t was decided to use a m ultiple e x tra c tio n ♦ (13) A f i f t e e n day e x tra c tio n of c o llo id a l m a te rial. The fiv e gram samples of clay were ag ita te d in water as before and l e f t standing fo r twenty-four hours. A fter th a t length of time the suspended liq u id was decanted c a re fu lly and the sediment which was l e f t behind again s ti r r e d w ith the same amount of water that had been removed. A fter th is churning the sample was allowed to s e t t l e fo r another twenty-four hours and was again decanted. This procedure was continued fo r f ifte e n consecutive days. A fter the f if t e e n t h decanting the sediment remaining was dried and weighed. The d ifference between the remainder and the o rig in a l fiv e grams taken was judged to be the co llo id a l m aterial removed. The experiment was not run fo r more than f if t e e n days because about the ten th day fu rth e r a g ita tio n fa ile d to disperse any appreciable amount of the co llo id a l m a te ria l. Very lik e ly some co llo id a l m a terial s t i l l remained adsorbed to the coarser fragments of the remaining sediment but i t i s believed that th is amount was small. 21 (2) Tests with Yiscosimeter and Hydrometer, liv e grams of clay were dispersed in four hundred cubic centim eters of water as before. A fter the suspension had been allowed to s e t t l e fo r twenty-four hours the co llo id sol was te ste d fo r density by a Westphal balance and a portion poured through an Gstwald viscosim eter and the time of flow compared with th at fo r an equal volume of d i s t i l l e d water. (5) Adsorption of Dyes, Water, and Ammonia. Probably the subject of adsorption of dyes and other substances by co llo id s has received as much discussion as any phase of experimentation in the f ie ld of co llo id chemistry. Many very p o sitiv e and in clu siv e statem ents have been made on the su b ject, lo r instance G i l e ^ s ta te s th at: "Hon-colloid p a rts of the s o il are without appreciable adsorption or heat of 22 w etting". Holmes l i s t s three ways of measuring colloid content, they being adsorption of water, ammonia, and m alachite green. The references in the l i t e r a t u r e are often so p o sitiv e that one is led to believe th a t the case is c le a r. If so, these adsorption t e s t s should not only check themselves, 21. Gile, P. L. C olloidal S o il M aterial. S o il science Humber 25, 1928. p 861. 22 Holmes, Harry N. Laboratory Manual of Colloid Chemistry. Wiley, 1954. p 207. 22 but check other methods, such as those involving to ta l so lid s. To make a check on the adsorption methods the clays were successively tre a te d with: (A) Methylene blue, (B)Malachite green, (C) Water vapor, and (D) ammonia. (A) Adsorption of methylene blue, liv e gram samples of clays were placed in six teen ounce wide-mouthed b o ttle s with four hundred cubic centim eters of d i s t i l l e d water. The dry methylene blue was then added and the mixture churned with a s t i r r i n g motor fo r twenty minutes# The mixture was allowed to s e t t l e for twenty-four hours a f te r which a portion was s e ttle d out by a water driven centrifuge of 2160 R.P.M. A few drops of hydrochloric acid were added to aid in the s e ttli n g i t being found by t r i a l th a t the acid caused no change in color of the dye. A fter cen trifu g in g , the c le a r, blue liq u id was compared with appropriate standards in a K le tt colorim eter and the amount of dye adsorbed was calcu lated . A d is tin c t excess of dye was used for the f in a l fig u re s because experi mentation showed that a clay y&11 adsorb more dye i f i t 23 were av a ila b le . A irston suggested a method in which methylene blue was added u n t i l the supernatant liq u id acquired a f a in t blue color which remained on standing. As these clays contain sonsiderable c o llo id a l m atter which 23 A irston, Margaret. A Study of the Adsorptive P ro p e rties of Colloidal Clays. U niversity of Southern C alifo rn ia M aster's Thesis. 1937. remains in suspension twenty-four hours or more th is method would show a color long "before the p a r tic le s had taken up th e ir maximum amount of dye. By th is method a value of 1.92 m illigram s of dye per gram of* f u l l e r s earth was obtained while the d is t in c t excess method gave a value of 1482. If adsorption is to be d ealt with as such i t would seem necessary to tr e a t i t as maximum adsorption instead of p a r tia l adsorption so an excess of dye was used in a l l cases. Bince so many statem ents have been made in d ic atin g that no n -co llo id al m aterial is p r a c tic a lly without adsorptive p ro p e rtie s , a determ ination was run on th is also . The residues from the f if t e e n day a g ita tio n s were taken and th e ir adsorptions of methylene blue v/ere measured in the same way as fo r the raw clays, from the adsorption for raw clay and that fo r no n -co llo id al m a terial i t was p o ssib le to ca lcu late the adsorption per gram of c o llo id a l m a te ria l. When th is value was determined i t could be compared with the adsorption fo r non-colloidal m a terial and a co llo id -f non-colloid r a tio was obtained. These data are presented in Table I I I . (B) Adsorption of m alachite green. Five grams of clay were placed in wide-mouthed b o ttle s with four hundred cubic centim eters of water as in the determ inations for methylene blue. The so lid m alachite green oxalate was then added and the mixture churned th ir ty minutes and allowed to s e t t l e fo r twenty-four hours. A po rtio n of the liq u id was then 2 4 centrifuged with a few drops of sodium chloride solution, i t being found that t h i s s a l t would not cause a change in color. The remaining dye was then determined by comparison with standards using the K lett colorim eter. Special care had to be taken with the malachite green. If too much dye v^as added i t would foam badly and a large amount of the dye would come to the top, even acting as a f lo ta t io n substance on some of the clay. In some such cases low r e s u lts were obtained. In other determ inations th is f lo ta t io n seemed to render the clay incapable of adsorbing any appreciable amount of the dye and obvious e rro rs would r e s u lt . Because of these reasons, very carefu l manipulation was necessary and various t r i a l s had to be made u n t i l the clay had been taken with a small, but d is t in c t excess of the dye. The presence of calcium s a lts in clays supposedly causes more of the malachite green oxalate to be adsorbed. Bather, i t i s said th a t the calcium s a l t s re ac t with the dye and that true r e s u lts can be obtained only a f te r the calcium s a lts have been p re c ip ita te d by sodium oxalate. To te s t th is statement the clay samples were a g ita te d as before, but th is time in three hundred seventy-five cubic centim eters of d i s t i l l e d water and tw enty-five cubic centim eters of a satu rated solu tio n of sodium oxalate. A fter twenty minutes of s t i r r i n g the dye was added and the mixture churned another 2 5 twenty minutes, allowed to s e t t l e , centrifuged and te sted in the colorim eter as before. (C) Adsorption of water vapor. Clay samples of approximately twenty grams were placed in large low-form weighing b o ttle s having diameters of fiv e centim eters. The samples were weighed a f te r having been dried for twenty- four hours at one hundred and fiv e degrees, Centigrade. The weighing b o ttle s with covers removed were placed in a large d e ssic a to r over two inches of water and l e f t there fo r seven days, the d essicato r being placed where the temperature did not vary more than fiv e degrees from a standard of seventy- two degrees, F ahrenheit. A fter seven days exposure to the sa tu ra ted atmosphere the b o ttle s were again weighed and the water vapor adsorption calculated in per cent. (D) The adsorption of ammonia. Small cylinders of clay were made by mixing the raw clay with water and squeezing the mass through a brass die fiv e -e ig h th s inch in diameter. Two inch sections of the clay p encil were cut o ff, dried in a i r and f in a lly twenty-four hours in an oven. The cylinders were weighed and then exposed to dry ammonia vapor fo r twenty- four hours, in an apparatus shown on P late 2. The ammonia was obtained by warming ammonium hydroxide, care being taken that the temperature did not r i s e higher than seventy degrees, Centigrade. The ammonia vapor was passed through three coolers packed in ice and s a lt to freeze out water vapor. 2 6 The ammonia gas was then passed through a long tube f i l l e d with so lid sodium hydroxide and f in a ll y passed over calcium chloride before going to the clay cylinders* A fter twenty-four hours exposure to the ammonia vapor the cylinders were taken from the adsorption chamber and placed in a d e ssic a to r for another twenty-four hours a f te r which the ammonia was boiled off by heat and caught in a so lu tio n of hydrochloric acid of known stren g th . The excess of acid was then t i t r a t e d back against a standard solution of sodium hydroxide and the ammonia adsorption calculated from the r e s u lt . (4) Ease Exchange« This is sometimes mentioned as a co llo id phenomenon* Only barium chloride was tr ie d here* liv e gram clay samples were churned in four hundred cubic centim eters of d i s t i l l e d water containing ten m illigrams of barium chloride per cubic centim eter of water. The mixture was allowed to s e t t l e twenty-four hours and then f i l t e r e d through doubled f i l t e r paper by suction. The cold solution of barium chloride which remained was made up to a volume of four hundred cubic centim eters and a one hundred cubic centim eter portion withdrawn fo r a barium determ ination. A home made turbidim eter was used fo r the barium determ ination. The so lu tio n to be tested was placed in a c le a r, square g lass b o ttle six centim eters in thickness. An excess of one to ten sulphuric acid was added to i t and the mixture shaken. An apparatus sim ilar to that shorn on P late 2 was used. Light from a midget fifte e n -w a tt Mazda lamp f i t t e d with a c e llu lo id r e f le c to r was passed through the turbid solu tio n and the lig h t* s in te n s ity measured by a G-eneral E le c tric lig h t meter reading d ir e c tly in foot candles. A standard curve had previously been made containing samples whose barium chloride content varied from zero to ten m illigrams per cubic centim eter. A curve had been made l i s t i n g foot candles of illu m in atio n against m illigram s of barium chloride per cubic centim eter. The illu m in atio n s varied from t h ir ty to seven hundred and f i f t y foot candles, the graph having i t s g re a te s t curvature between one hundred and two hundred foot candles. I t was found by gravim etric check th a t the samples to be te ste d could be compared with the standard curve and good r e s u lt s obtained, esp e c ia lly if the concentration was such as to give an illum ination of one hundred to two hundred foot candles, where the curvature of 24 the graph was g re a te s t. Concentrations in the barium chloride solutions were read in the turbidim eter and base exchange figured by the differen ce in concentration from the o rig in a l solu tio n used. 24. Harrington, E.R. &uick Q uantitative Method fo r Barium. The Chemist-Analyst in an issue to appear. Notice received November 1, 1939. 1=>tate Z f ! p p a r a £ U 5 fe a v a o % w w NaoH Dryer j ce _ S<4(i 0a.in To HO Vatfs R’C. IE tecirophor«ns Ffpparaius 2 9 (5) Heat of Wetting* A fiv e hundred cubic centim eter Dewar fla s k was f i t t e d with a two-hole rubber stopper bearing a se n sitiv e thermometer and a hand s t i r r e r . Two hundred cubic centim eters of d i s t i l l e d water were placed in the f la s k and temperature measured. One hundred grams of hot water of known temperature was added and the mixture*s re s u ltin g temperature determined. By the r is e in temperature the water equivalent of the calorim eter was determined. After the water equivalent of the thermometer had been determined, two hundred cubic centim eters of cold water were placed in the f la s k and allowed to stand u n til constant temperature was reached. Twenty grams of previously dried clay were then added and s t i r r e d and the temperature r i s e noted. From the r is e in temperature and the water equivalent of the calorim eter the heat of w etting was then calcu lated in c a lo rie s per gram of clay. (6) pH Determination. The pH values of the clays were determined e le o tr io a lly , using a hydrogen electrode with a calomel h a l f - c e ll as a reference. The calomel electrode was made up with sa tu ra ted potassium chloride and the whole 25 procedure conducted according to K olthoff and Furman using 25 K olthoff, 1. M. and Furman, N.H. Potentiom etric T itra tio n . Wiley, 1926. p 138, pp 204-205. 3 0 the ta b les for conversion which appear in the in s tru c tio n s given by the Leeds-Northrup Company2? Experiments were s ta rte d using a clay-water r a t i o of one to one hundred, by weight, but t h i s was l a t e r abandoned in favor of a one to fiv e r a tio , th a t being the one used by the U. 3. S oil Conservation Service in th e ir Southwestern laboratory. In the determ ination, t h i r t y grams of clay were placed in one hundred f i f t y cubic centim eters of d i s t i l l e d water in a four hundrea cubic centim eter beaker. The calomel and hydrogen electrodes were dipped into the mixture and the s t i r r i n g motor kept the clay and water well mixed u n t i l a constant reading was obtained on the potentiom eter. By use of a tab le mentioned above the E. M. F. shown by the potentiom eter was changed into pH. (7) Electrom etric T itra tio n . This determ ination was ca rrie d on with the same potentiom etrio equipment used in the pH determ inations. O riginally, fiv e grams of clay were taken with fiv e hundred cubic centim eters of water. A slow-speed s t i r r e r was used and the calomel and hydrogen electro d es were dipped d ir e c tly into the suspension. When equilibrium had been 26 Leeds and Northrup Company. Notes on Hydrogen Ion keasurement. pp 18-26. 51 reached small additions of 0.1045 normal sodium hydroxide were made and a l t e r each addition the s t i r r i n g was continued u n til equilibrium had been reached. The procedure was continued with small additions u n t i l fo rty or more cubic centim eters of' the sodium hydroxide had been added. The t i t r a t i o n curves were superimposed upon a curve where lik e amounts of sodium hydroxide had been added to the same amount of d i s t i l l e d water, without the clay. In some of the c±ays the curves showed some buffer action but the one to one hundred r a tio did not show th is to advantage so the experiment was repeated using t h i r t y grams of clay and one hundred and f i f t y grams of d i s t i l l e d water. These t i t r a t i o n s were superimposed upon a pure water curve as before. (8) E lectrophoresis. I t is an estab lish ed fa c t th a t the c o llo id a l p a r tic le s of a sol possess a charge and so migrate in an e l e c t r ic a l f i e l d . However obscure the o rig in of the charge may be i t s presence is re a d ily demonstrated. To t e s t t h is m igration an apparatus such as that shown in P late 2 was used. A power-pack supplying d ire c t current at a p o te n tia l of one hundred and th ir ty - f i v e v o lts was used as an e l e c t r ic a l source. A glass tube was bent in the form of a f l a t and wide U with a s lig h t ra ise d place in the center. Alectrodes were made of nickle wire and were placed twenty- 5 2 four centim eters apart giving a p o te n tia l gradient of fiv e and s ix -te n th s v o lts per centim eter. liv e grams of clay were placed in two hundred cubic centim eters of water and churned by a s t i r r i n g motor, no dispersing agent being used. The suspension was allowed to s e t t l e for ten minutes and a sample withdrawn from the top by a p ip e tte and placed in the glass tube, care being taken to see that the cen tral hump of the fla tte n e d U tube contained a small bubble of a i r . The glass tube was clamped in a horizontal p o sitio n and more of the sol added to one end or the other u n t i l the in d icato r bubble was in the exact center. The electrodes were then in serted into the ends of the tubes and the current turned on and allowed to flow fo r twenty minutes or more a f te r which time the distance the bubble had moved was measured. (9) btudies With the Ultra-microscope. If a powerful source of lig h t is used to throw a beam of lig h t through a sol i t is found that the lig h t path becomes luminous and we have the Tyndall e f f e c t. If th is intense beam of lig h t is examined by a high powered microscope set a t rig h t angles to the beam of l ig h t , the sc atte re d re fle c tio n s of the colloid p a r tic le s can be seen, the p a r tic le s appearing as luminous disks. In th is in v e stig a tio n a Bausch and Lomb s l i t u l t r a microscope was used, the lig h t source being a s e lf-a d ju s tin g 33 timed carbon are. The sol to be examined was placed in a pyrex c e ll, open at the top. The microscope could be adjusted Tor height and the illu m in atin g beam could be adjusted for depth of p en etratio n of the so l. The c e ll containing the m a te ria l to be examined could be moved p a r a l le l to or at rig h t angles to the beam of l i g h t . The samples of clay were s tir r e d with water and a small addition of ammonium hydroxide as a dispersing agent. These samples were allowed to s e t t l e for a day or more and at various times a few drops were withdrawn and placed in the pyrex c e ll fo r examination. The times of s e ttli n g varied from twenty-four hours to more than three weeks. In a number of cases the sols were d ilu ted down several times to see what e ffe c t th is would have on the number of p a r tic le s present in the sol and the speed of th e ir brownian movements. (10) Axperiments on Sedimentation. Here t e s ts were made in two fashions: (A) The s e ttli n g of p a r tic le s according to Stokes Law and (B) The use of the sedimentation balance . ^ (A) S e ttlin g according to Stokes law, l iv e gram samples of clay were dispersed in four hundred cubic centim eters of water in the usual manner and set aside to s e t t l e . The s t i r r e d m a te ria l was poured into t a l l hydrometer ja rs to a 27 «are, John 0. Chemistry of the C olloidal S tate. John Wiley and Sons, 1950. pp 24-25. 3 4 depth of eight inches. N aturally in d iv id u al p a r tic le s could not he picked out and th e ir s e ttli n g ra te determined so the marks chosen had to be c e rta in standards of opacity of the so l. A General B le c tric p h o to -cell lig h t meter was used to determine th is standard of opacity. A midget f i f t e e n watt Mazda lamp and r e f le c to r were placed on one side of the hydrometer tube and the lig h t meter placed on the other. Since the lig h t meter and the l i g h t source were each in contact with the hydrometer tube, the distance between them was always the outside diameter of the tube. The lig h t was turned on and the Tight in te n sity measured through the top of the so l. Time was taken and the s e t t l i n g allowed to continue u n t i l some hours l a t e r when the lig h t was again turned on and the in te n s ity of the illu m in atio n again read. By t h i s time i t would be found that the top of the sol was becoming more c le ar and both l i g h t and lig h t meter would have to be moved f a r th e r down the sedimentation column to find an opacity equal to that formerly encountered near the top of the suspension. By noting the time that i t took fo r a c e rta in standard of opacity to move downward the experimenter arriv ed a t an estim ation of the ra te of s e t t l i n g of the suspended p a r tic le s . Borne of these readings were carried on over a period of several weeks for some of the f in e r clays. (B) Use off the sedimentation balance. This apparatus may be very elaborate and even include se lf-reco rd in g devices of go one so rt or anotner. E sse n tia lly they are a l l the same in p rin c ip le , co n sistin g of a submerged pan upon which the p a r tic le s of a suspension may s e t t l e . This s e t t l i n g pan is attached to the beam of a d e lic a te balance so that the s e t t l i n g p a r tic le s may be weighed as they f a l l . The usual balance has a s e n s itiv i ty of fiv e m illigram s. The general 29 method advocated by Holmes was used. He suggested taking h a lf a gram of m a te ria l, disp ersin g i t in a liq u id and allowing i t to s e t t l e out upon the s e ttli n g pan while the observer, under time, kept the pan balanced by the addition of fiv e m illigram weights. In th is determ ination la rg e r weights were used because the s e n s itiv i ty of the balance was too low to use the smaller fiv e m illigram ad d itio n s. In th is determ ination ten grams of clay were dispersed in four hundred cubic centim eters of water and the mixture placed in the s e ttli n g ja r . The balance weights added were b ird shot which had previously been selected, by weight, from a large number. Any shot varying as much as one ZQ Bvedburg, The. Colloid Chemistry. Chemical Catalog Company, 1928. p 180. 29 Holmes, Harry E. Laboratory Manual of Colloid Chemistry. Wiley, 1934. p 4. 3 6 m illigram was re je c te d . This accuracy was thought to be s u ffic ie n t as i t was well w ithin the s e n s itiv i ty of the balance used. The suspended p a r tic le s were allowed to s e t t l e for several days but the accuracy a f t e r the f i r s t two hours is subject to doubt. 30 (11) fh lx o tro p ic Phenomena. Jenny s ta te s ; "YYhen a su ita b le e le c tro ly te i s added to a concentrated sol of f e r r i c hydroxide, g e la tin iz a tio n takes place. Upon shaking, temporary liq u ifa c tio n takes p la c e .M This r e v e r s i b i l i t y of gels to so ls to gels again is termed "thixotropy and has been 31 mentioned by numerous in v e stig a to rs as fa r back as 1923. A ctually, under another name, i t i s a very old phenomenon. The process of ^ca stin g ’1 clay ware i s a very old procedure and i t e s s e n tia lly i s determined by the p rin c ip le s of thixotropy. Home clays are e s p e c ia lly well su ited for the casting process. The f if t e e n clays taken in t h is rep o rt were te ste d as to th e ir a b i l i t y to form re v e rsib le g els. Freundlich 30 Jenny, Hans. P ro p e rties of Colloids. Stanford U niversity P ress, 1938. p 107. 31 Houser, E rnst. Colloid Phenomena. Me Graw H ill, 1939. p 216. 32 Jenny, Hans. P ro p erties of Colloids Op. C it. p 107. 3 7 rate d a substance as a gel i f a te s t tube containing i t could be turned upside down without lo ss of m a te ria l. That c r ite r io n was used in th is experiment, except in that eight ounce b o ttle s were used instead of te s t tubes. Various clay-water r a tio s were tr ie d and several e le c tro ly te s were added to see what e ffe c t they might have on the a b ility of the clay to form a re v e rsib le gel, s u ita b le fo r casting th in ware. (12) Liesegang Rings. I f two re a c tiv e substances which form an insoluble product are allowed to re a c t in a gel of s i l i c i c acid or agar, they w ill often produce a rhythmic 53 banding known as Liesegang rin g s, being named a f te r th e ir f i r s t recorder. Home years ago a number of gels and Liesegang rings were produced by the w rite r. S i l i c i c acid was used as the gel and the re a c tiv e substances were such as to produce rin g s of copper chrornate, s il v e r bromide, calcium phosphate and o th ers. In t h i s experiment the clays were tr i e d as the g els. Since most of the clay samples were dark in color only the l i g h t colored clays were used in th is experiment. Gala, A lb e rh ill, and Organ Kaolin were lig h t enough in color to act as proper backgrounds fo r th is phenomenon. In each case 35 Holmes, Harry N. Laboratory Manual of Colloid Chemistry. wiley, 1934. p 156. two clay gels were made by s t i r r i n g the clay into a water so lu tio n of potassium iodide. After the gels had been allowed to se t fo r a few hours the reac tin g substances were poured on top of the gel and allowed to d iffu se downward through the clay. Three of these gels were tre a te d with a so lu tio n of mercuric chloride and three with s ilv e r n i t r a t e , these so lu tio n s being calculated to produce mercuric iodide and, s ilv e r iodide, re sp e c tiv e ly , when they reacted with the potassium iodide of the gel. A second method involved only the Gala clay the w hitest of the samples. Here gels of clay and water, alone, were prepared and poured into three larg e U-tubes. These tubes were allowed to stand fo r several hours a f te r which the re a c tiv e substances were placed in opposite limbs of the tubes, above the le v e ls of the clay g els. In one tube the re ac tiv e substances were copper sulphate and potassium chromate; in another, s ilv e r n i t r a t e and potassium iodide; in the th ird potassium chromate and s ilv e r n i t r a t e . These tubes were watched c a re fu lly over a time of several months. (IS) P l a s t i c i t y . P l a s t i c i t y is c e rta in ly one of the important p ro p e rtie s of clay and i t is a property v/hich is ra th e r d i f f i c u l t to determine, though various methods have been used. Many attem pts have been made to l i s t p l a s t i c i t y as having a close c o rre la tio n with co llo id content also . 3 9 3 4 Ries quotes o eg e r's d e fin itio n of p l a s t i c i t y as: 1 1 The property which so lid bodies show of adsorbing and holding liq u id s in th e ir pores and forming a mass which can be pressed or kneaded into any desired shape, which i t r e ta in s when the pressure ceases and on wi their awl of the water changes to a hard mass’1. 35 Barrera points out re la tio n sh ip between shrinkage and 36 p l a s t i c i t y m c e rta in p o tte ry clay. Von Buzagh quotes Planck as saying th a t even the del"ormability of an ideal so lid body is determined by no le s s than t h i r t y - s i x fa c to rs . Prom th a t Von Buzagh in fe rs that the p l a s t i c i t y of such a complex m a terial as clay is very lik e ly due to many fa c to rs 37 also . Dearie l i s t s a number of fa c to rs e ffe c tin g the 38 p l a s t i c i t y of clays and Davis goes s t i l l f a r th e r and gives a d etaile d analysis of each factor e ffec tin g the c la y ’s p l a s t i c i t y . whatever may be said concerning the case of p l a s t i c i t y 34 Ries, H einrich. Clays. Their Occurrence, P ro p erties and Uses. V/iley, 1937. p 165. 35 Barrera, Tomas. Las A rc illa s y la Pabricacion de Loza de Oaxaca. Anales del in s t itu to de Geologico de Mexico. Tomo 4, 1930. p 110. 36 Von Buzagh, A. A Colloid Bystem. Technical Press, 1937. p 84. 37 Dearie, Albert B. The Chemistry and Physics of Clay and Other Ceramic M a te ria ls. Benn, London. 1933. p 279. 38 Davis, D. B. The P l a s t i c i t y of Clay and I ts R elation to mode of O rigin. Transactions of the American i n s t i t u t e of Mining .engineers, Volume 51, 1915. pp 451-480. 4 0 i t c e rta in ly is an important property ancl one dependent upon a number of d iffe re n t fa c to rs and there is no special reason to expect that d iffe re n t clays might not have somewhat d if fe re n t reasons for th e ir p l a s t i c i t y . Some clay workers determine p l a s t i c i t y by the shearing stren g th of an extruded clay column. Other workers have forced a th in clay p en c il from a die and determined the length of the column extruded 39 before i t broke away from the die. This method can be used on a sm aller scale than some of the others and i t was used in th is experiment. The experimenter must bear in mind, however, that the only sure method is to tr y the clay under actual manufacturing conditions and see i f i t s p ro p e rtie s are s a tis fa c to ry f o r the use to which he seeks to put i t . (14) Analysis of the Clays. As w ill be seen la t e r under the subject of R e su lts, the p ro p erties of the clays are not a l l in accord with co llo id theory. N atu rally i t was necessary to see i f the chemical composition of the clays would throw any lig h t on th e ir p ro p e rtie s . The analyses of eleven clays were obtained in a l l , some analyses coming from the companies handling the clay and six being analyzed by the w rite r. 39 Austin, Chester R. Surface Clays and Shales of Ohio. Ohio S tate U niversity Engineering Experiment s ta tio n B u lletin Number 81, 1934. p 4. 4 1 The analysis procedure was o b ta in ed ,c h iefly , by 40 41 following the d ire c tio n s la id down by Washington and Low , 42 45 . a a though the works of Scott , Willard and Turman , Ries^* and others were consulted. The general method was as follows: The clay samples were fused in platinum with sodium carbonate. The melt v/as extracted and evaporated down twice with hydrochloric acid to dehydrate the s i l i c a . The s i l i c a was f i l t e r e d o ff, blasted to constant weight, and v o la tiliz e d with hydrofluoric acid, the Si 0^ being determined by d ifferen ce. The f i l t r a t e from the s i l i c a determ ination was tre a te d with hydrochloric acid and an excess of ammonium hydroxide, thus p r e c ip ita tin g the R_ 0„ group of aluminum, 2 5 iron, and titanium . This complete group was heated to constant weight and weighed as a group. The e n tire group 40 Washington, H.S. Analysis of hocks. Wiley, 1930. 41 Low, Albert H. Technical Methods of Ore A nalysis. Wiley, 1919. 42 Scott, Wilfred W . Standard Methods of Chemical A nalysis. Van Nostrand, 1929. 43 Willard, Lobart H and Turman, 11. Lowell. nlementary Q uantitative A nalysis. Van Nostrand, 1935. 44 Kies, Heinrich. Clays, Their Occurrence. P ro p e rtie s ana bses. Wiley, 1937. 4 2 was given a fusion w ith potassium pyrosulpnate* The solution containing the R 0 group was made up to a c e rta in volume 2 0 and an aliquot p a rt taken for determ ination of titanium by the hydrogen peroxide colorometric method. The r e s t of the so lu tio n was run through the Jones reductor to reduee the iron and th is was then determined by t i t r a t i n g i t against a standard potassium permanganate so lu tio n . The aluminum was determined by d iffe re n c e . The f i l t r a t e from the Rg 0^ group was tre a te d with ammonium oxalate in the presence of an excess of ammonium chloride, the calcium being p re c ip ita te d as an oxalate and Ig n ited and weighed as calcium oxide. The magnesium was ob tained from the calcium oxalate f i l t r a t e by p r e c ip ita tio n with sodium ammonium phosphate and th is p re c ip ita te was ig n ited and weighed as the pyrophosphate. Separate samples were used for the determ ination of the combined a lk a lie s , the hydrofluoric acid method being used. Separate samples were also used to determine the lo ss of weight on ig n itio n . (15) Physical P ro p e rtie s of Clays Before and After T iring. When a company contemplates manufacturing ware from a c e rta in clay i t is f i r s t necessary to find out something about the raw m a terial. The manufacturer must know how re a d ily his clay taues up water on mixing. I t must have a c e rta in minimum 4 3 p l a s t i c i t y , He must know how much i t shrinks on drying and whether or not i t checks badly on being Tired. I t is necessary th a t he know i t s shrinkage lo ss on Tiring, the hardness of the fire d ware and i t s strength as w ell. He must know at what point of temperature the v i t r i f i c a t i o n of h is clay takes place, at what temperature i t develops i t s maximum hardness, whether i t tends to scum or b lo a t, how porous the fin ish ed product w ill be and what the e ffe c t ‘of water on the fin ish ed product i s . The answers to some of these questions were obtained fo r the f i f t e e n clay samples used here. H irst, each clay was mixed with water and pugged by hand u n til a th ick , homogeneous mass was prepared. These masses were allowed to dry in f l a t pans u n t i l they had lo s t th e ir stic k in e ss but were s t i l l so ft enough to be molded with the hands. These p la s tic clays were then rammed into two kinds of molds, one fo r te n s ile strength t e s t s and the other fo r determ ination of crushing stren g th , shrinkage, absorption of water, e tc. The te n s ile stren g th b riq u e tte s were made in the common f l a t , brass molds used in forming Portland cement t e s t shapes. The mold was given a th in coating of o i l and the clay rammed in by hand and then pounded down by a f l a t board and a hammer. About twenty b riq u e tte s were made out of each kind of clay. The molds were broken apart and the clay se t to dry on a wire screen covered with paper towels. Hach clay 4 4 P la te 5 Figure A Figure B Figure A. Tyndall e ffe c t caused by passing beam from an are- lamp through a clay so l. This clay (Santa Clara) had s e ttle d f o r twelve days when p ic tu re was taken. Figure B, Sedimantation balance set up. Later a t a l l e r balance was used and a la rg e r amount of clay taken as a sample. 4 5 b riq u ette was stamped with an id e n tif ic a tio n number. After drying in a i r fo r several days the blocks wefe dried in an oven a t one hundred and fiv e degrees, Centigrade fo r an a d d itio n a l tv/enty-four hours. They were then ready to f i r e . The samples to be used in the crushing te s ts were made in the form of a cylinder 1.528 inches in diameter and one and one h alf inches high. The mold was made of cast iron and the two halves could be separated by removing a clamp ring. Clay was placed in a brass pipe one and one h a lf inches in diameter and a hardwood plunger f i t t e d above i t . The brass tube was placed on top of the cast iron mold and the clay forced down into the cylinder by pounding the wooden plunger with a hammer. The clay cylinders were removed, marked, and dried in the same manner as the b riq u e tte s . In addition each cylinder was weighed a f te r i t had been dried. I t s diameter was also measured to the n e a re st thousandth of an inch by use of the v ernier c a lip e rs . I'rom the diameter measurements i t was p o ssib le to t e l l how much the lin e a r shrinkage of the clay had been by comparing the fig u re with the o rig in al diameter of the sample as i t came from the mold. A fter being dried, weighed, and measured the clay samples were placed in a f i r e clay sagger and placed on top of the brick p ile in a bee hive down-draft k iln using o i l as a f u e l. The k iln was operated by the Kinney Brick Company and the samples were f ir e d to a temperature of about eighteen hundred 4 6 degrees, Fahrenheit, to r approximately a week, the temperature being recorded by a thermocouple pyrometer. A fter being cooled the samples were removed and weighed and measured as before. The weights and measures so obtained provided means of ca lc u la tin g shrinkage and lo ss of weight on f ir i n g . The diameters of the cylinders also provided means for area de term inations which were to be used la t e r when the cylinders were to be te s te d for crushing stren g th . The sm allest p a r ts of the b riq u e tte s were also measured so t h e ir areas might be used in the te n s ile stren g th determ inations. Half of the b riq u e tte s and h alf of the f ir e d cylinders were soaked in hot water for twenty-four hours and allowed to dry in the a i r fo r a lik e period. The wet and dry shapes were then put through the same t e s t s . The b riq u e tte s were pulled apart in a Biehle te s tin g machine, the tension being applied at the ra te of nine hundred pounds per minute by a steady stream of fine shot. The cylinders were dipped in wax, th e ir ends capped with p la s te r of P aris and a f t e r a day's s e ttin g were crushed in an Olsen te s tin g machine with a movable upper block provided w ith a spherical bearing p la te . The cylinders were ca re fu lly centered and the bearing block lowered by hand u n t i l so lid contact had been obtained. The e l e c t r ic drive was then engaged and pressure applied a t the r a te of one thousand pounds per second, a se lf-re c o rd in g 4 7 d ia l stopping when the sample f a ile d under compression. From previous diameter measurements the crushing strength was then figured in pounds per square inch. The f ire d samples were also examined fo r color, hardness, sp a llin g , bloating, v i t r i f i c a t i o n and so fo rth , in order that an idea might be obtained as to the s u i t a b i l i t y of the clays for commercial work;. C hapter IV KB8 ULTS A H D THHIK INTERPKETATI O N The r e s u lts of the experimentation are taken up in the same order in which they were presented in the previous chapter, the order being : (1) Colloid Content Determined by E xtraction of Total Bolias; (2) Tests With Viseosimeter and hydrometer; (5) Adsorption of Dyes, Water, and Ammonia; (4) Base Exchange; (5) Heat of Wetting; (6) pH Determination; (7) E lectrom etric T itra tio n ; (8) E lectrophoresis; (9) Studies with the Ultra-Microscope; (10) Experiments on Sedimentation; (11) Thixotropic Phenomena; (12) Liesegang Hings; (15) P l a s t i c i t y ; (14) Analysis of Clays; (15) Physical P ro p e rtie s of the Clays Before and A fter Hiring. (1) Colloid Content Determined By the E xtraction of T otal S olids. As mentioned under experimental work, the w riter is adopting H ilg a rd 's c la s s if ic a tio n of a c o llo id a l p a r t i c l e . Any p a r tic le which stays in suspension for twenty- four hours in a column of water eight inches high is judged to be colloidal* In Table I there is a p rese n tatio n of data concerning the per cent of m aterial remaining in suspension three hours and for twenty-four hours, the p ip e tte method being used. In addition the same ta b le contains lik e inform ation concerning the complete e x tra ctio n of the suspended m a terial by the p erco lato r method and also the 4 9 per Gents obtained by fifteen successive extractions by decanting# Obviously the p e rc o la to r method is superior to the p ip e tte method since i t avoids any v a ria tio n in concentration at d iffe re n t le v e ls caused by s t r a t i f i c a t i o n . I t has every advantage the p ip e tte method has as far as accuracy is concerned and in addition i t weighs a l l the suspended m a te ria l and not ju s t an aliq u o t p a rt of i t . Though the percolator method is c le a rly superior to the p ip e tte method i t is s t i l l open to c ritic is m because of the fa c t that large p a r tic le s w ill carry the smaller ones down with them as was mentioned under experimental work:. On successive e x tra c tio n s i t was found th a t considerably more m a terial could be obtained by decanting a f te r as many as ten extractions# By the ten th e x tra c tio n , however, the supernatant liq u id is almost cle a r and, th e re fo re, these successive ex tra c tio n s were not continued more than f if t e e n days, i t being assumed that the c o llo id a l m a te ria l remaining with the sediment was comparatively small. Since th is f if t e e n day e x tra c tio n seems to be by fa r the most accurate means of determining the co llo id content i t is taken as the standard throughout th is determ ination. In a number of graphical p resen tatio n s which are to follow th is f i f t e e n day ex tra c tio n i s always represented by heavy black lin e s and a l l other findings, when presented g rap h ically 50 Table I Data on E x tra c tio n s by the T o tal B olids Method Kame of Clay Used A B C D Ban Ildefonso 0.9 Q jo 6 .4 $ 3.0 $ 0 .5 fo Kinney . a .i 6.5 5.6 1 .5 Tonque-Hed 5.9 6. a 7.0 3.3 Carthage-1 4 .3 9.1 10.7 3.7 Tonque-^Green 4.6 13.0 7.7 3.4 Las Vegas 4 .8 13.3 13.4 3.3 Gallup 5,8 15.1 36.7 4 .4 K aolin 7.0 17.3 ' 15.3 6.1 C arthage-3 11.1 33.0 16.0 7.9 Ban Antonio a. 7 £3.3 6.7 3,3 Banta Clara 6.9 34-. 9 6.3 3 .5 A lb e r h ill 30.1 37.7 30.6 18.3 Gala 16.5 38.1 31,7 11.8 B. H. 4 14. a 39.5 30.0 13.1 La Bajada 7.9 35.5 35.8 7.7 (A) Twenty-four hour e x tra c tio n by th e p e rc o la to r. (B) F ifte e n day e x tra c tio n , to be taken as stan d ard . (C) M aterial l e f t in suspension th re e hours as determ ined by th e p ip e tte method. (B) M a te ria l l e f t in suspension tw enty-four hours as determ ined by th e p ip e tte method. 51 are placed ad jace n t to i t f o r comparison* The p e rc o la to r e x tra c tio n as th e next in accuracy i s p resen ted in a lik e manner to the l e f t of the f i f t e e n day e x tra c tio n and always in w hite. P la te 4 shows a re p re s e n ta tio n of d ata on m a te ria ls e x tra c te d by th ese two methods, the v e r tic a l lin e s re p re s e n tin g per cen ts of m a te ria l e x tra c te d . P la te 5 shows the tw enty-four hour p ip e tte e x tra c tio n and the th re e hour p ip e tte e x tra c tio n placed ad jacen t to the standards of p la te 4. Thus p la te 5 re p re s e n ts th e complete g rap h ic al p re s e n ta tio n of d ata l i s t e d in Table I. S ev eral in fe re n c e s can be drawn from p la te s 4 and 5. In P l^ te 4 the clay s have been placed in ascending order according to t h e i r c o llo id a l m a tte r as determ ined by the f i f t e e n day e x tra c tio n , taken as stan d ard . In o th e r words La Bajada clay co n tain s the most c o llo id a l m a te ria l and Ban Ildefonso the l e a s t . I f , th e re fo re , the p e rc o la to r method is a v a lid means of determ ining c o llo id content we would expect a p re s e n ta tio n of m a te r ia l so obtained to r i s e in an ascending order p a r a lle lin g th a t o f our stan d ard . In P la te 4 we note th a t t h i s i s not the case. Ban Antonio clay in p a r tic u la r shows a a very bad la g w hile th e highly c o llo id a l La Bajada clay shows a lower p e rc o la to r value than fo u r o th e r c la y s, a l l of le s s e r c o llo id co n ten t. A ctu ally the San Antonio and La Bajada samples o f fe r th e key to the s it u a ti o n . A fter fiv e successiv e a g ita tio n s and e x tra c tio n s 52 most of the clay s were alm ost c le a r a f t e r an hour*s s e t t l i n g . In the San Antonio and La B ajada samples the tu r b id it y was not l o s t u n t i l at l e a s t te n e x tra c tio n s , showing c le a r ly th a t in th ese clay s the c o llo id p a r t i c l e s were being taken down by the la r g e r ones. La Bajada c la y co ntains co n sid erab le coarse m a te ria l which s e t t l e s q u ick ly , but the rem aining fin e m a te ria l remains in suspension sev eral weeks. C le a rly a la rg e p a rt of t h i s very fin e m a te r ia l adsorbs i t s e l f to the coarse p a r t i c l e s and needs numerous e x tra c tio n s to d islo d g e i t . In the case of P la te 5, we fin d cases th a t are even more c o n f lic tin g . I t i s seen th a t th e th ree-h o u r e x tra c tio n and th e tw enty-four-hour e x tra c tio n (both by p ip e tte ) do not p a r a l l e l the stan d ard . Here again we fin d San Antonio and La Bajada low on tw enty-four-hour e x tra c tio n . Santa C lara, a h ig h ly c o llo id a l and very p l a s t i c clay , jo in s San Antonio and La Bajada as an extreme exception. The th ree -h o u r e x tra c tio n s s t a r t out to conform in a b e tte r way but again we have San Antonio and Santa Clara as excessive lows and Gallup f i r e clay e n te rs the exceptions w ith an ex c essiv ely high value, c le a r ly out of p la c e in the p ic tu re . The in feren c e i s c le a r ; the p ip e tte method is not an a c cu rate method of determ ining the c o llo id a l m a tte r in a clay* S t r a t i f i c a t i o n , c u rre n ts caused by w ithdraw al of liq u id , a d so rp tio n of c o llo id p a r t i c l e s to p a r t i c l e s too heavy fo r them to f l o a t , - a l l th ese ren d er the method apt to 5 3 5: . \ X * X 'o s >5 S: k >2 >2 o i ! N ^ • S o > ? £ >5 N . $ < a 3 *:! > H C ? L ______ < I ----------- t: yjDeCvg VJ k~HV f/t’D y ^ / y rx J r > L T H y? *tu»g * * 0 4 yy u v g £ & •J'/eoy c/n//t>q s*Asn r>-2/)/luo^ / - V 4 f > /b i* < ? £ .O wyy i> < W /.) v y i i ^ g f n u a jv ^ yvynocfsftg //*? '? a C /?/ 0 - f S<ss/sendz,d C/ay P l a t e 5 P esu/is o f Jests by Pipette Method Compared Id/th P/ate 4 Ptrcolaiar Extraction UP Z 4 M rs. Op 3 Mrs. 55 produce la rg e e r r o r s . I f ju s t th e r ig h t c o lle c tio n of clays should be used the method would probably give r e s u l t s in d ic a tiv e of c o llo id a l co n ten t. For in sta n c e , i f San Antonio, Santa C lara, A lb e r h ill, and La Bajada were withdrawn, the tw enty-four-hour p ip e tte method would be in rough accord with the stan d ary f i f t e e n day e x tra c tio n . I f a se p a ra te ascending s e r ie s i s made according to e x tra c tio n by the p e rc o la to r i t i s found th a t t h is s e r ie s o f f e r s ju s t as u n s a tis fa c to ry checks with m a te ria l e x tra c te d by the p ip e tte method. (2) T ests w ith V iscosim eter and Hydrometer. I t was found th a t clay so ls which had s e ttle d fo r tw enty-four hours passed through the Ostwald v isc o sim e te r in p r a c t i c a l l y the same time as did the samples of d i s t i l l e d w ater, th e v a r ia tio n being so s lig h t th a t th e method o ffe re d no check upon the c o llo id p ro p e r tie s of the c la y s. 45 Concerning the hydrom eter, Bouyoucos says: "U nquestionably one of th e g r e a te s t needs in s o il study i s a very ra p id and sim ple method fo r determ ining the c o llo id a l m a tte r in a s o i l fp r c o llo id a l m a tte r now seems to c o n s titu te the se at of a l l th e r e a c t i v i t i e s of the s o il " . He goes on to 45 Bouyoucos, George John. The Hydrometer as a Hew and Kauid Method dor Determ ining the C olloid Content of S o ils . B oil Science Volume 23, 1927. pp 319-324* d escrib e h is hydrometer method in which a s o i l is d isp e rse d in w ater, potassium hydroxide being used as a d isp e rs in g agent. The suspension i s allowed to s e t t l e f i f t e e n m inutes before being te s te d with the hydrom eter. By t h i s method he claim s clo se checks with c o llo id d eterm in atio n s by heat of w ettin g methods. The close checks are open to q u estio n , however, as one of h is ta b le s l i s t s e r r o rs as high as s ix ty per c e n t.46 The g en e ra l th eo ry seems f a u lty also as he s t a t e s 4! "The m a te ria l stay in g in suspension f i f t e e n m inutes i s not a l l c o llo id a l but the content of c o llo id a l m a te ria l th a t might be c a rrie d down or s e t t l e d is ap p a re n tly compensated by the n o n -c o llo id m a te ria l stay in g in suspension. The n o n -c o llo id m a te ria l sta y in g in suspension at the end of f i f t e e n m inutes, however, is probably very s m a ll.” This l a s t statem en t, i f taken w ith Stokes law would assume the c o llo id p a r t i c l e to have a diam eter as g re a t as seventy mu in s te a d of the upper lim it of one mu as agreed upon by the 47 U. 5. Bureau of cioils. In a l a t e r p u b lic a tio n he l i s t s a f i f t e e n minute s e t t l i n g fo r s o il s , one hour fo r conventional clay s and two hours for c o llo id a l c la y s. Bven so the lim its 4 s I b id . pages 526 and 520. 47 Bouyoucos, George John. B tudies on the D ispersion procedure Used in Hydrometer methods fo r Making Mechanical Analyses of B o ils. S o il scien ce Number 55, 1952. p 21. 57 are such th a t the siz e of the c o llo id p a r tic le would have an upper lim it as high as tw en ty -fiv e mu which is much too high i f credence i s to be placed in Stokes* law a t all* 48 Keen c r i t i c i z e d the hydrometer method and d ec lared i t s t r i c t l y q u a lita tiv e and not q u a n tita tiv e and pointed out a 49 number of v a r ia tio n s from the checks claim ed. P u ri c r i t i c i z e d the technique and pointed out th a t only f a i r checks had been made in s ix te e n of t h i r t y c a se s. In s p ite of h is severe c r itic is m of the Bouyoucos hydrometer method he, too, announced an hydrometer method but h is r e s u l t s c a rrie d e r r o r s ,- or d e v ia tio n s ,- as high as f i f t y per cent. I t would c e rta in ly seem th a t i f siz e i s accepted as a c r i t e r i o n upon which c o llo id p a r tic le s a re defined, th a t the hydrometer method as p ra c tic e d could h ard ly be recommended. I f , however, the suspension i s allowed to s e t t l e tw enty-four hours, the d e n sity i s then so close to th a t of w ater th a t i t cannot be determ ined by a Westphal balance, l e t alone the hydrom eter. One must th e re fo re conclude th a t the hydrometer method is u n s a tis f a c to ry as a means of determ ining c o llo id content of c la y s. This conclusion i s in accord with the 48 Keen, B. A. Borne Comments on th e Hydrometer method Bor Studying B o ils. B oil Bcience humber 26, 1928. pp 261-265. 49 P u ri, *miar h a th . A hew Type of hydrom eter Bor the m echanical A nalysis of o o ils . B oil Bcience number 35, 1932. p 241. views of workers in the lo c a l S o il C onservation L aboratory. ■These workers la ck f a i t h in the method although i t i s s t i l l the c h ie f one they use in th e i r own la b o ra to ry . (3) A dsorption of Dyet Water, and Ammonia. The r e s u lt s of the ad so rp tio n of methylene blue, m alach ite green, water vapor, and ammonia appear in Table I I , to g e th e r with the c o llo id r a t i o worked out from methylene blue. (A) A dsorption of methylene b lu e . This inform ation i s l i s t e d in column A in Table I I and i s p rese n ted , g ra p h ic a lly 50 in P la te 6. Ware l i s t s su rface area s of m a te ria ls on su b d iv isio n and shows th a t one cubic cen tim eter of a substance divided down to a s iz e of one mu p a r t i c l e s would A produce a t o t a l area of 6 x 10 square ce n tim e te rs. Since the clays here of course have s p e c ific g r a v iti e s of two or above, a clay volume of one cubic cen tim eter would re p re se n t a weight of approxim ately two grams. Low i f we are to assume a p a r t i c l e su b d iv isio n as low as 0 .1 mu we would 5 expect one gram of clay to have an area of 3 x 10 square ce n tim ete rs, assuming a l l the clay was so subdivided. Paneth 50 Ware, John C. Chemistry of the C o llo id a l B tate . Wiley, 1930. p 5. 51 Paneth, I r i t z and Thurman, Wilhelm. The Adsorption of Lyes by C ry s ta ls . B e rich te , Lumber 57 B, 1924. p 1221. concludes th a t in dye ad so rp tio n the amount of dye taken up i s never g r e a te r than the q u a n tity needed to farm a mono- 52 m olecular la y e r over the adsorbing substance. Holmes gives Paneth*s ad so rp tio n r u le th a t one m illig ram of dye i s s u f f ic ie n t to cover one square m eter of area or an area of 4 10 square c e n tim e te rs. I f t h is ru le was ap p lied r ig i d l y and the e n tir e gram of clay was broken down to p a r t i c l e s of 0 .1 mu we would never expect a g r e a te r ad so rp tio n of methylene blue than about s ix m illigram s o f dye per gram of c la y . The low est value found in t h i s in v e s tig a tio n was f o r ty - f iv e m illig ram s per gram of clay and the c o a rs e st sample (San Ildefonso) adsorbed s ix ty m illigram s of dye per gram of sample. According to Paneth*s system th is would in d ic a te th a t the clay is d iv id ed down to p a r t i c l e s of a s iz e 0.01 mu. This is c le a r ly out of th e q u estio n because when t h i s clay was tre a te d w ith f i f t e e n successive e x tra c tio n s only 6.4 53 per cent of th e clay was found to b© c o llo id a l. Pallmann shows diagrams of clays adsorbing more than one la y e r of m olecules, the v ario u s m olecular s h e lls forming around the p a r t i c l e in a lte r n a tin g p o la r it y of m olecules. Numerous assum ptions of t h i s s o r t have been made to account fo r the 52 Holmes, Harry N. In tro d u c to ry C olloid C hem istry. Wiley, 1934. p 143. 53 Pallmann, H, Uber Bodenblldung und Bodenferein en der Schweis. Die Hrnahrung Der P fla n z e , B e rlin , tfuli 1934. pp 225-234. 60 T ab le I I Bata on A bsorption of Dyes Water and Ammonia Name of Clay A_______ B C ________D _______ £_______ £ Ban Ildei'onso Kinney Tonque-Bed C arthage-1 Tonque-Green Las Vegas Gallup K aolin C arthage-2 Ban Antonio S anta C lara A lb e r h ill Cala B. ii • 4 La Bajada 60 6.7 88 1 .0 62 2.7 76 1 .0 53 4.1 74 1.7 48 3.8 90 1.5 98 2 .6 86 1 .2 145 1 .1 45 2 .8 46 2.9 55 2.3 117 3.8 198 92 240 280 122 225 140 142 272 230 295 307 96 99 168 112 168 123 290 292 338 183 81 143 92 107 176 135 5.5 9.7 8 .4 6 .2 5.8 9.0 7.7 4 .4 4 .5 6.6 6 .8 7.4 5.6 3.3 10.6 5.9 9.3 7.7 9.5 4.9 16.1 18.9 7.4 2.9 7.6 2 .1 7.9 4.6 11.9 9.8 250 166 A Methylene blue in m illigram s per gram of cla y . B C o llo id -ra tio f o r methylene blue. C M alachite green in m illig ram s per gram of clay which has been tr e a te d w ith sodium o x a la te . 1 ) M alachite green in m illig ram s per gram of raw clay . B Water vapor in per cen ts of the weight of the clay . B Ammonia in m illigram s p er gram of clay . 61 T able I I I Data For Methylene Blue C o llo ld -R atio Rame of Clay A B C Ban Ildefonso 278 42 6*7 Kinney 89 88 1.0 Tonque-Eed 154 57 2.7 C arthage-1 76 76 1.0 Tonque-Green 139 34 4.1 Las Vegas 116 67 1.7 Gallup 128 34 3.8 K aolin 134 82 1.5 Carthage-2 154 59 2.6 Ban Antonio 97 83 1.2 Banta C lara 158 143 1.1 A lb e rh ill 85 30 2.8 Cala 89 30 2.9 B. H. 4 95 40 2.3 La Bajada 369 97 3.8 A Methylene blue in m illigram s adsorbed per gram of c o llo id . B Methylene blue in m illigram s adsorbed per gram of re sid u e from fifte e n -d a y e x tra c tio n . C C o llo id - ra tio , A -7-B . 62 charge on c o llo id p a r t i c l e s so perhaps an extension of th a t p r in c ip le might account fo r th e la r g e r dye a d so rp tio n . R egardless of what th eo ry might seem to imply i t is apparent th a t the clay s adsorb co n sid erab le methylene b lu e. When the amounts are placed alongside th e stan d ard s as they are in P la te 6 i t i s noted th a t th e re is no constant in c re a se from l e f t to r i g h t . A lb e r h ill, Gala, and 3. H. 4 are very low in th e i r ad so rp tio n though th e y are high in c o llo id a l m a tte r. The lo w -co llo id Kinney clay has a high ad so rp tio n and the .banta C lara sample has a value roughly equal to the next th re e samples added to g e th e r. A glance at th e c o l lo id - r a tio in Tables I I and I I I is a lso of i n t e r e s t . Here we have r a t i o s varying from 1.0 to 6.7 . f o r in sta n c e , in the Han Ildefonso clay the c o llo id a l m a tte r is c re d ite d w ith adsorbing 6.7 tim es as much dye, gram fo r gram, than did the re sid u e l e f t by th e fifte e n -d a y e x tra c tio n . In Kinney, C arthage-1, K aolin, Han Antonio, and Santa C lara, th e re sid u e s l e f t a f t e r th e e x tra c tio n proved them selves to be p r a c t ic a lly as good ad so rb ers as the c o llo id a l m a te ria l which had been removed. Such r e s u l t s should c e r ta in ly throw some doubt upon the o ft-e n co u n te re d 64 statem ent th a t: ,fThe non c o llo id p a r t of the s o il i s w ithout ap p rec iab le ad so rp tio n c a p a c ity .” 54 G ile, P. L. C o llo id a l S o il M a te ria l. S o il scien ce Humber 25, 1928. p 361 . P /a / e 6 P f e //y /ene B /t/e Pc/sorp//^ Co^par>eJ WpA P/a/e 4 $ • } < f ) \ > > 3oo Pert o / a tor £E X tra c S/on I /&- Pay pxAr&cfion Methyfwe Blue % £ 2oo p T H r m i T T t 1 r~ 1 ! r I 1 IP T P I 1 1 I I ii1 .— L . iii i_ 4 < b % ^3 C s 1 v5 As a g e n e ra lity i t can be sa id th at the c o llo id s do adsorb more dye than the n o n -c o llo id s but th e n o n -c o llo id a l m a tte r i s c e r ta in ly capable of some ad so rp tio n and in some cases i s h ig h ly so. I t is ev id en t th a t a c o llo id a l f r a c tio n from some clays i s of g re a te r ad so rp tiv e power than th a t from 55 some o th e r c la y s. In a former in v e s tig a tio n the w rite r found that the amount of m ethylene blue adsorbed per gram of a c tu a l c o llo id a l m a te ria l v a rie d from s ix ty -f o u r m illigram s to more than t h i r t y - f i v e hundred, a range of about one to s ix ty . That in v e s tig a tio n was made on th e assumption th a t the c o llo id a l m a te ria l was re sp o n sib le fo r a l l the a d so rp tio n , a su p p o sitio n th a t would seem sc a rc e ly te n a b le . (B) The ad so rp tio n of m alach ite green. The ad so rp tio n of m alach ite green confirmed the conclusions reached from experim ents on methylene blue. P la te 7 shows a g rap h ic al p re s e n ta tio n of the m alach ite green a d so rp tio n s superimposed upon the stan d ard graph. Here again we fin d A lb e r h ill, Cala, and S. H. 4 as lows in the ad so rp tio n column though high in c o llo id a l m a te ria l. The lo w -c o llo id Kinney clay again has high ad so rp tiv e p r o p e r tie s as do th e Tonque-Bed, Tonque-Green and Las Vegas sam ples. A ctu a lly the graph would be s lig h tly more in keeping with theory i f i t were to be 55 H arrington, K. B. Some P re lim in ary In v e s tig a tio n s on C o llo id a l C lays. U. S. C. Beport f o r Chem istry 29G-L . 1935, p 10. 65 turned around and put in to complete re v e rs e . Santa C lara and San Antonio are again th e high adso rb ers f a r out of keeping w ith th e i r c o llo id co n ten t. Knowing th a t Kinney, Tonque-R, Tonque-G, and San Antonio were high in calcium s a l t s one im m ediately came to th e conclusion th a t th ese s a l t s were re sp o n sib le f o r th e high r a te s of a d so rp tio n . However, we must note th e high ad so rp tio n r a te s of Las Vegas and La Bajada which a re low in calcium s a l t s . in a d d itio n we fin d th a t the samples where sodium o x a la te was added to p r e c i p ita t e the calcium show no ap p rec iab le v a r ia tio n from th e values f o r the raw c la y s. I f calcium s a l t s a c tu a lly use up the dye as is o fte n s ta te d we would look f o r the raw clay values to be higher in the graph on P la te 7. In seven cases t h i s i s tru e (Kinney, Tonque-R, Las Vegas, G allup, San Antonio, A lb e rh ill and C ala). In only th re e of these seven cases do the clays have any co n sid erab le amount of calcium carbonate p rese n t and i t i s a n o tic e a b le f a c t th a t the clay with th e h ig h e st calcium content (San Antonio) has alm ost e x a c tly the same value fo r the raw clay as i t has f o r the sample tr e a te d w ith sodium o x a la te . In seven o th e r cases (San Ild efo n so , Tonque-G K aolin, C arthage-3, Santa G lara, S. H. 4 and La Bajada) the co n d itio n s were re v erse d and th e a d d itio n of sodium o x a la te produced a g re a te r ad so rp tio n in s te a d of a le s s e r one. E v id en tly in th ese l a t t e r cases the sodium o x alate was a c tin g 66 as a p e p tiz in g agent and b rin g in g more c o llo id a l m a te ria l in to r e a c tiv e form. The same s itu a tio n was encountered p re v io u sly by the w rite r who took tw elve C a lifo rn ia clays and found th a t in eleven cases the o x alate treatm en t in c re ase d ad so rp tio n in ste a d of d ecreasin g i t . He a lso found th a t the clays had g re a t d iffe re n c e s in ad so rp tio n per gram of c o llo id a l m a te ria l p r e s a n t,-th e s e values running from 102 to 6100 m illigram s of m alach ite green per gram of raw clay and from 102 to 6780 m illigram s per gram of the m a te ria l f i r s t tre a te d with sodium o x a la te . Here again the conclusions a re e s s e n ti a ll y th e same as those a rriv e d a t from the ad so rp tio n of methylene b lu e. The clay s are e v id e n tly h ig h ly s e le c tiv e in t h e ir ad so p rtio n and w hile th e re seems to be a g en e ra l q u a l ita tiv e r e la tio n s h ip between c o llo id content and ad so rp tio n of m alachite green th e c o r r e la tio n i s by no means q u a n tita tiv e . In o th e r words, th e general assumption th a t c o llo id a l m a tte r of a clay or of a s o i l can be measured by i t s ad so rp tio n of m alach ite green i s sc a rc e ly te n a b le . 56 Thies found th a t the clays which adsorbed the most dyes were the p o o rest ones to use as rubber f i l l e r s while the poor adsorbers were the b est f i l l e r s . The in feren ce was th a t 56 T hies, H. H. H ala tio n Between Dye Adsorbed by Clays and T heir Behavior in Hubber Compounds. Journal of I n d u s tr ia l and Engineering Chemistry, Humber 17, 1925. p 1166. P /ate 7 A1 a /a cf/fe Greet? /Icfsorphot? Cowparec/ 14/M P / a / < ? 4 Pm o/a f o r Exfracft'on I IS - * Day £xfrctcfio0 1 law O xuJatie Treated ■ US tij us lifJ IlM U S i c n -O c o llo id a l clay s were i n t e r i o r as f i l l e r s ana the c o llo id content was d ir e c tly p ro p o rtio n a l to the amount of dye they 57 adsorbed. Ashley , as fa r back as 1909 pointed out the f a c t th a t clays w ill adsorb d if f e r e n t amounts of dye depending upon how much i s a v a ila b le . A c e r ta in Texas k ao lin , fo r in sta n c e , when tr e a te d with one gram, two grams, and th re e grams of b r i l l i a n t green, l e f t one m illigram , th ir te e n m illig ram s, and fo rty fiv e m illigram s unadsorbed showing th a t th e clay*s ad so rp tio n was q u ite dependent upon the a v a ila b le amount of dye. His t e s t s with m alach ite green were ju s t as la ck in g in conform ity. He re p o rte d some check between dye ad so rp tio n and p l a s t i c i t y but th is l a t t e r p ro p e rty is such an in ta n g ib le one and so dependent upon o th e r f a c to rs th a t the evidence i s f a r from conclusive. 58 ware sa id concerning ad so rp tio n by c o llo id s : tt. . . I t can be seen th a t th e b a s is fo r the ad so rp tio n of the c o llo id s i s s t i l l very la rg e ly in the em p irical s ta te and th at in d u s t r ia l o p eratio n s in v o lv in g the c o llo id a l s t a t e must, of n e c e s s ity , r e s t very la rg e ly upon ex ten siv e experim entation r a th e r than exclusive- . in d u c tiv e reasoning which is based on a fundamental c o n c e p t.” This seems to be tru e and the ad so rp tio n of dyes seems to have as i t s main use the f a c t th a t i t o f fe rs a g en eral in d ic a tio n of amounts of c o llo id m a te ria l p re s e n t. 57 Ashley, H arriso n . The C olloid M atter of Olay and I t s Measurement. U. 8. G. 8. B u lle tin 388, 1909. p 43. 58 Ware, John C. Chemistry of the C o llo id al H ta te . Wiley, 1930* pp 75-74. 69 (G and D) A dsorption of w ater and ammonia* Way, as e a rly as 1850 p o in ted out the f a c t th a t clays l o s t t h e ir power of adsorbing w ater when su b jected to a red heat# 59 B le in in g e r sa id : "The w ater thus taken up from sa tu ra te d a i r may be said to be p ro p o rtio n a l to the amount of c o llo id p r e s e n t,- ie i t v a rie s in v e rs e ly as the mean diam eter of the p a r t i c l e s " . Jenny p o in ts out th a t m o n tm o rrillio n ite , a m ineral p re se n t in b e n to n ite , has a s o r t of space or la y e r l a t t i c e work th a t expands and c o n tra c ts lik e an accordion with the presence of vary in g amounts of w ater, numerous o th e r statem en ts are found in the l i t e r a t u r e in d ic a tin g th a t th e re i s a d e f in ite c o r r e la tio n between ad so rp tio n of w ater and th e q u a n tity of c o llo id a l m a tte r p re s e n t. Clays a lso adsorb co n sid erab le ammonia even when kept dry. Hueckel n o te s th a t p l a s t i c clays adsorb more ammonia 62 than clays of l e s s e r p l a s t i c i t y , Moore and o th e rs e x tra c te d 59 B le in in g e r, A. V. The P ro p e rtie s of C lays. C olloid Symposium Monograph number 2. Chemical C atalog Company, 1925. p 90. 60 Jenny, Hans. P ro p e rtie s of C o llo id s. S tanford U n iv ersity P re s s , 1938. p 22. 61 Hueckel, W alter C. Research in Pry P re ss R e f r a c to r ie s . Ohio S ta te U n iv e rsity Engineering Experiment S ta tio n B u lle tin Humber 82, 1954. p 23. 62 Moore, C harles G - and o th e rs . Methods of Determining th e Amounts of C o llo id a l M a te ria l in S o i l s . Jo u rn al of in d u s tria l' and Engineering Chemistry, number 13, 1921. p 23, the c o llo id a l m a te ria l from some s o i l and found th a t the " u lt r a - c la y ” so e x tra c te d adsorbed a l i t t l e more than fo u r tim es as much ammonia as did the n o n -c o llo id a l m atter# 63 Anderson and M attson obtained fig u re s fo r c o llo id a l a d so rp tio n of both w ater and ammonia and determ ined a r a t i o of w ater ad so rp tio n to ammonia ad so rp tio n varying from 2.93 to 4.33, a v a r ia tio n of about f o rty per ce n t. Checks even as clo se as t h i s lead some workers in the f i e l d to speak of c o llo id content as determ ined by water ad so rp tio n as checking 64 w ith d eterm in atio n s made by ammonia a d so rp tio n . Anderson in a form er a r t i c l e published r e s u lt s such as the ones l i s t e d in Table IV. These r e s u lt s check each o th e r in a few cases but e r r o rs up to f i f t y per cent can be found and the method cannot be termed q u a n tita tiv e . The inform ation on ad so rp tio n of w ater vapor and ammonia f o r the IM ew Mexico clays appears g ra p h ic a lly in P la te 8. I t w ill be n o tic ed th a t th e w ater ad so rp tio n r i s e s as h ig h .a s six te e n per cent in the case of Santa C lara clay though t h i s i s approxim ately equal to the sum of th e ad so rp tio n s fo r A lb e r h ill, Cala, and S.H .4, any of which has a g re a te r 63 M attson, Svante with Anderson, M .S .P ro p erties of the C o llo id a l S o il M a te ria l. U. S. Department of A g ric u ltu re B u ll'stin l\iumber 1452, 1926• p 12. 64 Anderson, M. S. The Meat o f w etting of S o il C o llo id s. Jo u rn al of A g ric u ltu ra l R esearch, Volume 28, 1924. p 930. 71 T able IV Samples Taken from the Adsorption Tables of the L ite r a tu r e S o il Number .......... by_e______ >.ater Ammonia 1 19.4 $ 32.4 32.1 ° / o 2 10.1 10.1 14.6 3 17.2 24.8 34.0 4 9.3 18.1 20.6 S o il Number lie a t of W etting bye v<ater N H3 1 35. 6 25.4 31.5 26.0 2 29.5 29.4 29.8 34.5 These ta b le s were taken from page 930 of the Journal of A g ric u ltu ra l Hesearch, Volume 28, Number 6, 1924* 72 colloid, co n ten t. The Organ k a o lin also shows i t s e l f to be a good adsorber of w ater though not having a high c o llo id co n ten t. The ad so rp tio n of ammonia, lik e th a t of w ater, does not in c re a se s te a d ily w ith in c re a se in c o llo id co n ten t. In comparing the ammonia and w ater ad so rp tio n s we are brought to th e conclusion th a t they are somewhat s im ila r to each o th er in th e ir g en eral in d ic a tio n s but th a t th e c o r re la tio n between them i s f a r too s li g h t to suggest th a t they check each o th e r, l e t alone the c o llo id a l m a te r ia l in the clay samples. In g e n e ra l, when summing up the in d ic a tio n s concerning ad so rp tio n of dyes, w ater, and ammonia, one must conclude th a t th e r e - is n o t a q u a n tita tiv e r e la tio n s h ip between the ad so rp tiv e p ro p e rtie s of a clay and i t s c o llo id co n ten t. ilL Base Exchange. Like a l l base exchange experim ents t h i s one i s open to the q u estio n as to ju s t how much the r e s u l t i s due to base exchange and ho?/ much i s due to a d so rp tio n . Perhaps, a la rg e p a rt of th e r e s u l t i s due to simple ad so rp tio n and i f so the procedure should be c a lle d 65 the ad so rp tio n of barium c h lo rid e in s te a d . Hies p o in ts out th e f a c t th a t some clays adsorb so lu b le s a l t s from 65 H ies, H ein rich . Clays, T heir Occurrence, P ro p e rtie s and Uses. Vviley, 1957. p 268. P e r C e s? f ///. 0 //Ported k y P ry C /a y P / a t e 3 /fc /s o r /o h o n o f k f o P r a n tf /fsn /tfo si/a f <)// ■ i V 'S 7 7 * ' I Percotaior is-Day £ t fruc+fon y ^fraction lh<> % I / / 1 1 ^ ii^ lifS ! a < 3 c > ; 7 4 s o lu tio n s and m entions such s a l t s as barium, aluminum, and lead compounds being removed in co n siderable q u a n tity while the compounds of strontium , magnesium, and calcium are removed in le s s e r amounts. he goes on to say th a t c h lo rid e s, n i t r a t e s , and a c e ta te s are adsorbed more than su lp h a tes and th a t the higher the c o n c e n tra tio n , the more th a t w ill be taken up. This would seem to in d ic a te a high degree of s e l e c t i v i t y in clays and we would h ard ly expect the ad so rp tio n of th e se s a l t s to be a means of m easuring the c o llo id a l m a tte r p re s e n t. Murray^6 found th a t a c o llo id s e t t l i n g out barium su ip n a te would tra p most of th e barium 67 but l i t t l e of the su lp h ate ion. E verhart and o th e rs , in using barium s a l t s to avoid scumming in f i r e d clay ware found that the su rface clays (presumably of g r e a te r c o llo id content) took more of these s a l t s than did the sh a le s which came from some d ista n c e beneath the ground’s su rfa c e . 68 S chollenberger determ ined re p la c e a b le bases in s o ils by the ammonium a c e ta te le ach in g but did not a r r iv e at any 66 Murray, ii. D. The C oagulation of C o llo id s by E le c tr o ly te s . Chemical Hews, Humber 123, 1921. pp 277-279. 67 E v erh art, J . 0. w ith Hueckel, W . C. and A ustin, C.R. Barium Hydroxide to P revent Scumming of Ceramic P ro d u cts. Ohio S ta te U n iv e rsity Engineering Experiment S ta tio n C irc u la r Humber 30, 1935. 68. S c h o lle n b e rg e r, 0. J . Determining the R eplaceable Bases in B o ils by the Ammonium A cetate Method. Science, Humber 65, 1927. p 552. / a e r ( f r s r / r ? < 7 S' < C f/ #y P/ate 9 Barium Ch/oriae /fdsorp+/on Compared W//t> P/ate 4 P ert o fa for J f ~ £Ext r a c f ton £ -H tra o ttc p _ 7 6 d e f in ite conclusion reg ard in g the i n t r i c a t e p ro cess as a means of determ ining the c o llo id content of s o i l s . In t h i s study the barium c h lo rid e replacem ent d a ta ,- or a d so rp tio n d ata p o s s ib ly ,- i s given in Table V to g e th e r w ith the inform ation on heat of w ettin g , ad so rp tio n of water and ad so rp tio n of ammonia. The barium ch lo rid e d ata i s also p resen ted g ra p h ic a lly in P la te 9. The r e s u l t s , w hile they show co n sid erab le d iffe re n c e in the c lay s, bear no r e l a t i o n ship to c o llo id content* (5) Heat of W etting. I t i s a known f a c t th a t w ater has a d e f in it e , though slow a c tio n on c la y s. Whitney6® t r i e s to ex p la in the form ation of s o i l c o llo id s by means of tru e chemical a c tio n and th e bombarding of the clay p a r t i c l e s by m olecules of w ater. Davis found th a t clean , c r y s ta lli n e rock powders i f ground in water f o r a long time began to have c o llo id p ro p e r tie s such as p l a s t i c i t y . He p o in ted out th a t the more p l a s t i c clay s were th e ones th a t had been subjected to the a c tio n of w ater fo r a long p erio d . W e do know th a t clay s r e a d ily imbibe water and in th e p ro cess, whatever i t 69 Whitney, M ilton. The O rigin of Boil C o llo id s and T heir Reason, fo r A xisting in th is B tate of M atter. •science, Volume 54, 1921. pp 653-656. 70 Davis, M'.B. The P l a s t i c i t y of Clay and I t s R e la tio n to Mode of O rigin. T ran sactio n of th e American I n s t i t u t e of Mining E ngineers, Volume 51, 1915, p 480. 77 Table V A dsorption of Water, Ammonia, and Barium Chloride Lame of Clay A B C D San Ildefonso 5.5 9.7 2.7 312 Kinney 8 .4 6.2 1 .5 280 Tonque-Red 5.8 9.0 2.6 256 C artbage-1 7.7 4.4 2.2 280 Tonque-Green 4 .5 6.6 2.1 264 Las Vegas 6.8 7.4 3.0 256 Gallup 5.6 3.3 1.7 312 K aolin 10.6 5.9 2.2 190 Cartiiage-2 9.3 7.7 2.6 320 San Antonio 9.5 4.9 3.2 264 Santa Clara 16.1 18.9 3.7 284 A lb e r h ill 7.4 2.9 2.1 320 Gala 7.6 2.1 1.0 312 S. H. 4 7.9 4.6 2.3 254 La Bajada 11.9 9.8 3.8 280 A A dsorption oi‘ w ater vapor in per c e n ts . B A dsorption of ammonia in m illigram s per gram of clay* C Beat of W etting in C g |o rie s per gram of clay . D A dsorption of barium c h lo rid e in m illigram s per gram. 7 8 i s , a sm all amount of las at i s given o f f. Gil© 71 c r i t i c i z e s Bouyoucos fo r in f e r r in g th a t a c o llo id i s any substance which, w ill give a heat of w etting although he agrees th a t n o n -c o llo id a l m a tte r is w ithout any 72 ap p rec iab le amount of such h e a t. S te n z e l , however, has shown th a t 200-mesh ground q u artz has an ap p reciab le heat of w ettin g , and o th e r .substances not s t r i c t l y c o llo id a l have 73 lik e w ise had the same p ro p e rty . The heat of w ettin g fo r the f i f t e e n samples i s l i s t e d in Table V and p resen ted g ra p h ic a lly in P la te 10. Yalues from one c a lo rie to 3.7 c a lo r ie s per gram were obtain ed . Borne o th e r samples were run a lso , a b e n to n ite going as high as 12.9 c a lo r ie s p er gram. Prom the g rap h ic al p re s e n ta tio n i t w ill be noted th a t La Bajada clay , the one co ntaining the most c o llo id a l m a te ria l, i s also th e one having th e g r e a te s t heat of w ettin g . However, i t i s a lso seen th a t Ban Ildefonso clay, the one co n tain in g the l e a s t c o llo id a l m a tte r ranks f i f t h from the 71 G ile, P . L. C o llo id a l B oll M a te ria l. B oil sc ie n c e , Number 25, 1928. p 359. 72 B tenzel, B. W . The Heat of Immersion of S ilic a Gels In V arious Petroleum S ubstances. Journal of the iunerican Chemical S o ciety , Number 54, 1932. p 871. 73 Holmes, H arry N. Laboratory Manual of C olloid C hem istry. Wiley, 1934. p 99. P/a H / 0 / i e a t o f W e i i in g C O m p a reo i With P/a ie 4 PerccJa fo r* tS~ D a y £ y itr a c Hon £ x Hon Heal »/ M h «j 80 top in th e ta b le while Las Vegas and Tonque-R are also high in heat of w etting though th ey co n tain com paratively sm all amounts of c o llo id a l m aterial* As in the ad so rp tio n of dye, the C a lifo rn ia c lay s, A lb e r h ill, Gala, and 8. H. 4 , are low t e s t e r s though high in c o llo id a l m a tte r. The r e s u l t s do not in d ic a te th a t heat of w ettin g can be used as a means of determ ining c o llo id content of a wide sampling of c lay s, 74 (6) PH D eterm inations, Wilson l i s t e d the pH of clays 75 from 3,0 to 8 .5 , Powers found samples running as high as 9 .2 , t h i s being due to s o i l co n tain in g co n sid erab le ”black a l k a l i ” or sodium carbonate. In the t e s t s made on the f i f t e e n samples l i s t e d in th is re p o rt the pH values were: San Ildefonso 7.4 Las Vegas 6.4 Santa Clara 6.2 Kinney 7.1 Gallup 4.1 A lb e rh ill 4 ,4 Tonque-R 7.4 K aolin 6.6 Gala 5.2 Carthage-1 6.4 C arthage-2 3.6 S. H. 4 4.6 Tonque-G 7,5 Ban Antonio 8.1 La Bajada 4.6 Statem ents have been made th a t clays u su a lly give an acid pH because the O H ion i s re a d ily adsorbed by the c o llo id 74 Wilson, H ew itt. Ceramics and Clay Technology. M e Craw H ill, 1927. p 67. 75 Powers, w. L. A Study of the C olloid f r a c tio n s of C ertain S o ils Having R e s tric te d D rainage. S o il Science Number 23, 1927. pp 487-491. 81 p a r t i c l e s of the clay# P o ssib ly t h i s i s c o rre c t, i f so we would expect the h ig h ly c o llo id a l c la y s to give sm aller pH v alu es. In our d eterm in atio n s the pH values ran from S .6 to 8 .1 , th e two samples g iv in g the extreme v a r ia tio n s being ad jacen t to each o th e r in the ta b le and of almost the same c o llo id co n ten t. The one with the high pH owes t h is to the presence of calcium s a l t s . In g eneral we can see th a t as the c o llo id content has in c re ase d the pH has become sm aller. 76 (7) H iectrom etri c T ,itrat io n s . Jenny r e f e r s to the 77 78 works of Baver and B ra d fie ld who t i t r a t e d clays w ith sodium hydroxide and found them a c tin g lik e monobasic a c id s . This a c tio n can r e a d ily be seen in P la te 11 where the t i t r a t i o n curves of two clay s are superimposed upon a curve formed by a d d itio n s of sodium hydroxide to d i s t i l l e d w ater. A ll of the clays te s te d acted very much in th is way except the C arthage-2 sample which shov^ed a d i s t i n c t b u ffer a c tio n u n t i l about s ix cubic cen tim eters of the sodium hydroxide had been added. 76 Jenny, Hans. P ro p e rtie s of C o llo id s. S tanford U n iv e rsity P re ss, 1938. p 33. 77 Baver, L. D. Research B u lle tin Humber 129. M issouri S ta te A g ricu ltu re Experiment S ta tio n , 1929. 78 B ra d fie ld , R ichard. Proceedings o f the H irst In te rn a tio n a l Congress of S o il Science, Number 3, 1928. pp 838-869. 8 2 io P/a /e Ti/ration Carves SO m i A/a OH 30 Z O so 83 Hue to th e f a c t th a t th e clays in a one to one hundred clay -w ater r a t i o did not show any g re a t d iffe re n c e in a c tio n from a l i k e q u a n tity of d i s t i l l e d w ater, the t i t r a t i o n s were rep eated using a clay -w ater r a t i o of one to f iv e . These r e s u l t s are p resen ted in P la te s 13, 13, and 14. I t w ill he n o tic e d in a l l cases th a t th e re i s considerable b u ffe r a c tio n shown. This i s s p e c ia lly w ell shown in P la te 12 with th e C arthage-2 sample. In a l l cases, however, when enough sodium hydroxide was added the c lay curves always came up to the water curves, though in se v eral in sta n c e s as much as one hundred cubic cen tim eters of the sodium hydroxide had to be added. These curves were hard to make because the hydrogen e le c tro d e was e a s ily poisoned and had to be r e - p la tin iz e d a t sh o rt in te r v a ls . Sometimes as many as th re e e le c tro d e s were used in one d eterm in atio n . A ctu ally the curves show th a t very l i t t l e of th e sodium hydroxide is adsorbed a f t e r a c e r ta in c o n c e n tra tio n of the base i s reached. When th a t p o in t i s reached th e curve r i s e s ra p id ly to the value one would expect fo r d i s t i l l e d w ater, showing th a t p r a c t ic a lly a l l of the O H ions are out a t work and are not adsorbed by the c lay p a r t i c l e s . Perhaps t h i s i s due to the c o llo id p a r t i c l e s being p r e c ip ita te d or coagulated a t a c e rta in co n c e n tra tio n and thus allow ing the adsorbed O H ions to be re le a s e d at th a t c e r ta in 1 1 i s o - e l e c t r i c " p o in t. 'T iir a iic n C u r v e s C la y TZaha ’ 3 r 30 40 Zo 40 to 8 5 It It 1 0 9 P/a te / 3 a Tf/ra lien Curves i C la y Ticrho j f 5 so m ! A/a OH 40 Jo 20 do _ _ _ _ _ _ _ _ 4o_______ so ml A/aOH p H 86 T i i r a t i o r C a r v e 5 zo JO 8 7 (8) B le c tro p h o re sis. However obscure the o rig in of the e l e c t r i c charge on the c o llo id p a r tic le may be, i t i s easy 79 tp dem onstrate th a t such a charge e x is ts . Coward , fo r in stan ce n o tes th a t b en to n ite p a r t i c l e s move toward the p o s itiv e pole in an e l e c t r i c a l f i e l d , th e re fo re th e p a r tic le s 80 must be n e g a tiv e ly charged. B ra d fie ld conducted e l e c tr o p h o re sis experim ents w ith u l tr a - c l a y using a v o ltag e of 120 and a d ista n c e between e le c tro d e s of 43*1 ce n tim e te rs. The cu rre n t was passed f o r twenty m inutes and he found m ig ratio n toward the p o s itiv e p o le . He produced a sy n th e tic c o llo id by analyzing the u ltr a - c l a y and making up a m ixture of Si 0g, 0^, and Alg 0^ , f in e ly d isp e rsed and in the same 81 p ro p o rtio n s but found th is s y n th e tic c o llo id not to have the same p r o p e rtie s as the n a tu ra l one. Among o th e r v a ria tio n s i t had a p o s itiv e charge and th e re fo re m igrated' toward the n eg ativ e p o le . He concluded th a t the c o llo id m a te ria l was a complex a lu m in o -s ilic a te and not a m ixture of th e se p a ra te oxides. In t h i s in v e s tig a tio n a l l of the clays possesed a n eg ativ e 79 Coward, H. B. Sedim entation of B e n to n ite. Journal of the Chemical S o ciety , Volume 125, 1924. p 1471. 80 B ra d fie ld , R ichard. The Chemical N ature of a C o llo id a l Clay. U n iv ersity of M issouri A g ric u ltu ra l Experiment S ta tio n R esearch B u lle tin Number 60, 1923. pp 27-29. S1 I b id . pp 56-57. 88 Table VI Data On K leo tro p h o resis Dame of Clay Minutes c u rre n t D istance th a t V elocity of was passed bubble moved p a r tic le San Ildefonso 39 25 mm. 0.64 mm/min, Kinney 21 18 0.86 Tonque-R 32 18 0.56 Carthage-1 26 18 0.69 Tonque-G 34 25 0.74 Las Vegas 24 10 0.42 Gallup 36 36 1.00 K aolin 20 10 0.50 Carthage-2 16 10 0.62 San Antonio 26 20 0.77 Santa C lara 35 20 0.57 A lb e rh ill 15 10 0.67 Cala 26 30 1.15 S * xi * 4 52 20 0.62 Xa Ba,jada_____________ 24________ 20 0.83 P o te n tia l d iffe re n c e was 15b v o lts and d ista n c e between e le c tro d e s was 24 ce n tim ete rs th e re fo re p o te n tia l g ra d ie n t was 5.6 v o lts per c e n tim ete r. 89 charge, hence m igrated toward the p o s itiv e p o le. The p o te n tia l g ra d ie n t was 5.6 v o lts per ce n tim e te r, ahout tw ice th a t used by B ra d fie ld . Data fo r the time of m ig ratio n and the d ista n c e moved 3rs given in Table VI . M igration v e lo c itie s v a rie d from 0.42 m illim e te rs to 1.15 m illim e te rs per m inute, th e re being no sp e c ia l r e la tio n s h ip between the m ig ratio n r a te and the s iz e of p a r t i c l e s . I t was n o tic e d th a t th e passage of the sm all amount of c u rre n t through the s o l acted as an e f f e c tiv e p r e c i p ita t o r , causing the p a r t i c l e s to s e t t l e out ra th e r ra p id ly . (9) S tu d ies w ith the U ltra-M icroscope. F ir s t of a l l i t was found that th e brownian movements observed did not follow th e c la s s ic a l diagram almost always found in the te x t books. B ather i t seemed th a t the p a r t i c l e s d r if t e d acro ss the f i e l d of view showing sm all zig -z ag movements as they went, the am plitude of the movements being g re a te r fo r th e 82 sm a lle r p a r t i c l e s . Wesley and France had found t h i s out when they made motion p ic tu r e s of th e brownian movements of fo u r clay s and i t had been commented on by o th e rs . They also found th a t the am plitude of the movements o ffe re d them a rough check on the p l a s t i c i t y of t h e i r fo u r c lay s. 82 Wesley, B. w ith France, G -. An U ltram ieroscopic Motion P ic tu re Study of the K elatio n of C olloid Content and P l a s t i c i t y in Clays. Journal of th e American Ceramics .Society Volume 9, Number 2, 1926. p 76. 90 In t h i s work a l l f i f t e e n clays were examined and a l l samples showed p a r t i c l e s in the u ltram icro sco p e, though some were d i f f i c u l t to lo c a te due to adjustm ent tro u b le s and also due to the f a c t th a t the d i s t i l l e d water , i t s e l f , was o p tic a lly a c tiv e . The ban Ildefonso sample was allowed to s e t t l e from a suspension f o r two days before a p a r t of the so l was examined. Many p a r t i c l e s were v is ib le , t h e i r shiny in te rfe re n c e rin g s overlapping. The so l was cut down in co n c en tratio n on 1:5 r a t i o s e ig h t su ccessiv e tim es and s t i l l p a r t i c l e s were v is i b le . On account of the o p tic a lly a c tiv e w ater ( c o llo id a l s i l i c a , probably) the d ilu tio n could not give a measurement of the number of c o llo id a l p a r t i c l e s p re se n t in any given volume of s o l but th e re seemed to be alm ost as many p a r t i c l e s in the d ilu te d s o ls as in the more concentrated ones. D ilu tio n s as low as 1:10,000 were used and every in d ic a tio n p o in te d to the in fe re n c e th a t as the so l i s d ilu te d th e c o llo id p a r t i c l e s break up a lso , p o ssib ly due to the bombardment of the w ater m olecules. Kinney clay so l showed la rg e i r i d i s c e n t d isk s under the u ltram icro sco p e, Brownian movements c o n siste d of slow re v o lu tio n s . Tonque-R and C a rth ag e-1 , showed p o o rly . P a r tic le s d r if t e d acro ss th e illu m in a te d f i e l d but no p a r tic u la r ly fin e lig h t e f f e c ts or broYvnian movements were see^n* T'onque-G showed b r i l l i a n t l i g h t p o in ts a f t e r two days s e t t l i n g . A fter fo u r days of s e t t l i n g th e Las Vegas sample showed many p a r t i c l e s 91 ana s l i g h t , though d i s t i n c t brownian movements. Gallup clay gave a fin e d isp la y of l i g h t e f f e c t s . Some p a r t i c l e s d r if te d a c ro ss the f i e l d w ith d i s t i n c t ana ra p id brownian movements. Some o th e r p a r t i c l e s were q u ite la rg e and d r if t e d through the f i e l d of view lik e la rg e crowns studded with g l i t t e r i n g jew els. The Organ k ao lin and C arthage-2 showed good brownian movements a f t e r s e t t l i n g fo r two days. The San Antonio so l was s e t t l e d fo r two days and tr ie d under the ultram icroscope at a number of d ilu tio n s down to 1:10,000 • D is tin o t p a r t i c l e s , o r r a th e r l i g h t p o in ts , were observed in a l l cases. Under th e lower c o n c en tratio n s the p a r t i c l e s seemed to be held in loose g e l- lik e aggregates and under such co n d itio n s no v ib ra tio n s were noted. Good brownian movements were seen a t g r e a te r d ilu tio n s , however. The Santa C lara sample a f t e r two days s e t t l i n g showed p a r t i c l e s but th e so l was so th ic k th a t i t needed much d ilu t io n and a f t e r too g re a t a d ilu tio n the o p tic a lly a c tiv e w ater showed up b e t te r than the suspension being stu d ie d . The A lb e r h ill clay showed good brownian movements a t 1:1000 d ilu t io n . Cala and S. H. 4 showed s im ila r perform ances^ the l a t t e r , lik e the Gallup sample a lso showing some la rg e p a r t i c l e s which looked l i k e g re a t bundles of jew els. La Bajada sample o ffe re d the b e s t medium fo r study and showed a c tio n r a th e r ty p ic a l of th e o th e r samples a ls o . On s e t t l i n g two days alm ost a c le a r liq u id was l e f t above the sedim ent, only a s li g h t cloudiness snowing• The illu m in a tin g beam from the ultram icroscope showed a f a i n t Tyndall e f f e c t but many p a r t i c l e s were v is ib le in the f a in t beam. The p a r t i c l e s moved slowly acro ss the f i e l d showing a fin e d isp la y of ir id is c e n c e . E vidently some of the p a r t i c l e s were not round or symm etrical because th e in te rfe re n c e "haloes* th a t were produced were sometimes fan-shaped and changed from complete co n c en tric c ir c le s to shining d isk s , to fan s, to pin p o in ts , back to co n cen tric c i r c le s , e tc . e tc , as the p a r t i c l e s turned over and over in the illu m in a te d f i e l d . The sm aller p a r t i c l e s moved f a s t e r and had d i s t i n c t zig-zag p a th s. An e f f o r t was made to count p a r t i c l e s in a given volume of so l but t h i s was not completed on account o f the o p tic a lly a c tiv e w ater which made i t im possible to t e l l what p a r t i c l e s came from the clay and what ones came from th e w ater. I t was noted, however, th a t d ilu tio n s down as f a r as 1:500 seemingly had no e f f e c t on th e number of p a r t i c l e s p rese n t so i t is r a th e r evident th a t th e re i s f u r th e r su b d iv isio n of p a r t i c l e s on d ilu tio n . (10) Experiments on S edim entation. S e ttlin g of p a r t i c l e s according to o to k e s 1 law was f i r s t t r i e d , the law ’s equation being l i s t e d a t the top of the next page. 9 3 2 g ( d1 - dg ) V ” ------- ----------------------------------- 9 n ¥ = v e lo c ity o f . s e t t l i n g p a r t i c l e in c e n tim e te rs/se c . g - a c c e le ra tio n of g ra v ity , 980 cen tim eters / sec /s e c . a - ra d iu s of suspended p a r t i c l e in c e n tim e te rs. d^ - d e n s ity of th© p a r t i c l e . dg - d e n s ity of the d isp e rs in g medium. n - v is c o s ity of the suspending medium in p o ise s. The work was conducted a t tw enty-four degrees C entigrade. The s p e c if ic g r a v itie s of a l l clays had been determ ined. The v is c o s ity of the w ater was taken as 0.0091 p o ise . A ll d ata f o r the s e t t l i n g according to Stokes* law are found in Table V II. I t must be borne in mind th a t these f ig u re s fo r p a r t i c l e s iz e are only in d ic a tiv e of the siz e of the p a r t i c l e s th a t s e t t l e d out during the tim es s p e c ifie d . For in sta n c e a s e t t l i n g of 1 .5 ce n tim eters in 120 hours fo r 8. H. 4 in d ic a te d a s iz e of 0 .116 mu fo r th os e p a r t i c l es doing th e------ s e t t l i n g . The so l was s t i l l very dense a f t e r a s e t t l i n g of th re e weeks so c le a r ly some of the o th e r p a r t i c l e s were much sm a lle r. .Similar, though le s s pronounced occurrences were noted in A lb e r h ill, Santa C lara, and C arthage-2. Values given are to be taken as upper lim its of s iz e fo r the co a rser fragm ents only. 94 Tab/e V// Dai a an Se/ibby N am e O f C f a y V/stance d e t ile d “ Tame O f S e l l/w j Vefoc/ty 2* Cm. ^ 5ec. Specific Gravity 7)f 5 ample 5/z.e 07 Parf/efe San TlJefonso 9.7 Cm. 19 Hours 0*0 00 f 4 Z 2 . 3 4 0*6*2 P K i n n e y 3- 0 4 9 0*00 00 t? 1*9 Z 0. 2,73 Tona/ue - ^ 13. 0 4 3 0*00 0 0 752 Z . 2 8 0 - 4 B S -----V......" " " ' ”" C arthage-/ 1 . 7 1 1 9 0*0000063 Z • 6 4 0 / 3 0 7onqn/e - 0 6 * 2 z o 6*6 0 00 770 Z * 3 2 6- 3 0 0 ----- 7 -- " ’ L as Vegas 2 * 2 4i> 0*0060/2 1 * 7 1 0. 2 S 9 Gallop 2 3 1 Z o* 0 0 0 0 0 3 9 2 * 3 8 a* / <o4 K achn z * z 9 ? Q.00000 6 3 2 * 2 8 0 . 1 4 3 Carlhaqe-Z 1*8 2 - 4 O'OOOQ / o q 2 * S i o * n o San Anion to ? . o ? 0 * 000X3 2 * 2 4 0 • 96 6 Sania Clara 2 * 4 4 8 0 • 0 0 0 0 / 4 I >91 o z s t Htberhill i s 7 2 o.ooooo s e 2 . 3 3 0 ‘ / Z p O. /48 Cala 2 9 7 4 3 0 006005 6 2*0 8 5. H 4 I S i Z O o . o o o o o s s 2. 0 8 0 . / l b La TSajaUa /(a. 0 3 Z o. o o o 14 2 * 2 4 0 6 3 Z 95 The curves of s e t t l i n g obtained from the sedim entation balance appear in P la te 15. To avoid overlapping of the curves each su ccessiv e one was s ta r t e d two m inutes l a t e r on the graph. I t w ill be n o tic e d th a t the San Ild efo n so clay , which has th e l e a s t c o llo id a l m a tte r, s e t t l e s out most of i t s m a te r ia l in f i f t e e n m inutes w hile th e A lb e r h ill and Cala c la y s, both h ig h ly c o llo id a l, s t i l l r e t a i n most of th e i r sediment af~uer two hours of s e t t l i n g . P re s e n ta tio n of curves fo r a g re a te r length of time would be too involved but fo r the sake of making the curves more complete a summation is given in Table V III showing how much more m a te r ia l s e t t l e d out a f t e r the graphs were stopped. Here we w ill see th a t the h ig h ly c o llo id a l A lb e r h ill and Gala clays s e t t l e out 75 # and 54 $ more m a te r ia l in the next se v eral days but th a t La Bajada clay s e t t l e s out only U*7 % more m a te ria l in the next two days. The m a te ria l in th e f i r s t two clay s (Gala and A lb e rh ill) seeras to be m ostly very sm all p a r t i c l e s w hile La Bajada i s e v id e n tly made up of both la rg e and sm all p a r t i c l e s , the la rg e ones taking most of the sm all ones w ith them when th ey s e t t l e , hence making a rep ea ted e x tra c tio n necessary to show a l l c o llo id a l m a te ria l. This i s c e r ta in ly an argument for th e f ifte e n - d a y e x tra c tio n as a means of a c tu a lly determ ining the c o llo id content of the clay being stu d ie d . 96 T able V III Data On F u rth e r S e ttlin g Of Clays Lame of Clay L ast Reading On Graph Days f L a te r lox f i n a l Reading f u r th e r S e ttl in g San Ild efo n so 49 3 53 8 .2 $ Kinney 26 4 31 19.2 Tonque^-R 41 2 42 2.5 C arthage-1 50 2 54 8.0 Tonque-G 44 3 45 2.3 Las Vegas 31 3 37 19.4 Gallup 25 2 32 28.0 K aolin 40 2 41 2.5 C arthage-2 44 3 51 15.9 San Antonio 34 1 35 2.9 Santa C lara 29 2 30 3.3 A lb e r h ill 20 5 35 75.0 Gala 22 4 34 54.6 S. E. 4 25 3 29 16.0 La Bajada 37 2 38 2.7 The ta b le may be explained as fo llo w s, tak in g La Bajada as a sample. The l a s t read in g p resen ted on the graph in P la te 15 shows a reading of 37. Two days l a t e r the f i n a l read in g was taken, i t being 38, a gain o f one w eight, th is re p re s e n tin g an added sedim entation of 2.7 per cent. PLftTE / S ' 5 E D IM E N T R T tO N C U R V E S u r 98 (11) T hixotropic Phenomena. P r a c tic a l th ix o tro p y was used by clay workers long before i t ever came to be known by th is name; the process was known as " c a s tin g ”. In the clay samples used here the q u estio n s asked were: (A) W ill the clay form a s li p which might be used in c a stin g ? (B) ’ What e f f e c t, i f any, does an e le c tr o ly te have on the s l i p ? (G) ’ W ill the clay s l i p become a g e l and then become a liq u id again on s t i r r i n g or shaking ? 83 Jenny quotes P reu n d lich to th e e f f e c t th a t a g el may be considered as such i f a t e s t tube of i t can be in v e rte d w ithout s p i l l i n g . In t h is experim ent a more rig o ro u s t e s t was used, the s l i p being placed in an e ig h t ounce wide-mouthed b o ttle w ith a diam eter of about one and o n e-h alf inches, approxim ately four times th a t of a sm all t e s t tube. I t was n ecessary , th e re fo re , fo r the f i f t e e n clays te s te d to have a g re a te r v is c o s ity than would have held them in an in v e rte d p o s itio n in sm all diam eter tu b e s. V arious clay r a t i o s had to be used fo r the d if f e r e n t samples but when the proper clay -w ater com bination had been found a l l samples were able to q u a lify as th ix o tro p ic . The clay -w ater r a t i o s and o th e r inform ation are given in Table XX. 83 ' Jenny, Hans. P ro p e rtie s o f C o llo id s. S tanford U niversity P re ss, 1938. p 107. 99 Table IX Data On Thixotropy Of Clays Name of Clay W ater-Clay B atio Remarks on M a te ria ls San Ildefonso 75:50 Used NaCl and Nag SiO^ Kinney 50:50 Same as above Tonque-B 50:50 Aided by Nart SiO„ 2 3 C arthage-1 75:50 same as San Ild efo n so Tonque-Gr 50:50 Bo improvement w ith NaCl Las Vegas 75:50 No e le c tr o ly te used Gallup 75:50 Used NaCl and Na0 SiO* 2 o K aolin 100:50 Same as above C arthage-2 75:50 No e le c tr o ly te used San Antonio 75:50 Na_ SiO helped 2 3 Santa C lara 100:50 No e l e c tr o ly te used A lb e r h ill 75:50 B a d added Gala 85:50 Same as above 3 • ii« 4 75:50 Na^ SiOs helped La Bajada 100:50 No e le c tr o ly te used A b en to n ite was p re p a re ! w ith w a te r-c la y r a t i o s as g re a t as 97:3 and proved th ix o tro p ic* 100 The Cala and A lb e r h ill clay s are used in c a stin g in C a lifo rn ia and La Bajada, Banta C lara, K aolin, and Carthage-2 should work eq u a lly w e ll. The o th er samples, w hile showing enough r e v e r s i b i l i t y to be c a lle d th ix o tro p ic are too coarse f o r the c a stin g of th in ware. Sodium s i l i c a t e makes the s li p s somewhat more f l u i d but otherw ise does not seem to have any s p e c ia l e f f e c t on th e p ro p e rtie s of the s l i p as they s t i l l g el on sta n d in g . Sodium c h lo rid e does not have much e f f e c t except in the case of the A lb e r h ill sample, and to a somewhat l e s s e r e x ten t w ith Santa C lara and La B ajada. The shrinkage of La Bajada and Santa C lara clays is high and i t i s d o u b tfu l i f they could be used in c a stin g very th in ware on account of the high breakage due to checking. (12) Liesegang Rings. I t i s a known f a c t that c r y s ta ls w ill grow in g e ls and under proper circum stances form rhythm ic bands. In n atu re the agate i s a good example of the p ro cess. The roughly s t r a t i f i e d bands of manganese s ta in s in the Organ k a o lin might have been formed by two re a c tin g substances w ithin the g e l. In th is work only th re e clay s were used on account of t h e i r l i g h t co lo r being th e proper background to show any bands th a t might form. The Cala clay when used as the re a c tin g stage fo r m ercuric c h lo rid e and potassium iodide formed th re e ir r e g u la r 101 heavy hands near the top follow ed by fo u r more la y e rs , uniform and d i s t i n c t but th in n e r than the th re e above* The lower rin g s were about one cen tim eter thick* When potassium iodide and s i l v e r n i t r a t e re a c te d in the same clay the bands were ir r e g u la r and close packed and the example was n o t good. The K aolin, though a darker clay than the A lb e r h ill or Cala, worked b etter* The m ercuric ch lo rid e-p o tassiu m io d id e re a c tio n produced fo u rte e n d i s t i n c t bands of m ercuric io d id e . The f i r s t th re e bands were about f iv e m illim e te rs a p a rt but the fo llo w in g ones became c lo s e r below. The potassium io d id e and s i l v e r n i t r a t e re a c te d to form two th ic k and d i s t i n c t bands, varying co n sid erab le in c o lo r, the top one being dark gray and th e bottom one a gray-green* A lb e r h ill clay formed a ready background fo r m ercuric c h lo rid e and potassium io d id e , e ig h t d i s t i n c t bands being formed. The s i l v e r n i t r a t e and potassium io d id e re a c te d r e a d ily enough but ir r e g u la r blobs of co lo r r e s u lte d and no rhythmic bands were formed. The Cala clay , alone, was used in the th re e U -tubes. E ight th in , c le a r cut bands and a mass of sm all, red c r y s ta ls were formed in the tube in which m ercuric ch lo rid e and potassium io d id e had been p laced . The r e a c tin g su rface was much n ea re r th e limb in to which the m ercuric ch lo rid e had been placed, showing th a t the potassium io d id e had d iffu se d more ra p id ly through the c la y than had the m ercuric c h lo rid e . 102 In a second U-tube copper su lp h ate and potassium chromate re a c te d to form th re e bands of copper chromate, the a c tio n tak in g place n e a re r the copper sulphate s o lu tio n , again showing the potassium s a l t to be the f a s t e r in d iffu s in g through the g e l. Avery r e s u l t seems to p o in t to clay as being as su c c e ssfu l a growing medium as th e g e ls of s i l i c i c acid , g e la tin e , or ag ar. 84 (13) P l a s t i c i t y . Bingham sa id : "A substance i s p l a s t i c which cannot only be molded or deformed under p re ssu re , or more e x a c tly shearing s t r e s s , but w ill also hold i t s shape when the- shearing s tr e s s has been removed. Clay has been conceived to be th e p l a s t i c substance fpar excellence* because i t i s e a s ily shaped in th e hands of the p o tte r and once given 85 an exact shape, r e ta in s i t . " B le in in g e r gives p l a s t i c i t y as the most c h a r a c te r i s tic p ro p e rt of cla y s. The above two statem en ts seem to be tru e enough. A ll of the clays te s te d here might be sa id to be p l a s t i c sin c e they a l l could be re a d ily molded in to shapes. There seems to be some r e la tio n s h ip between p l a s t i c i t y and c o llo id a l content a ls o . The seven clay s having the h ig h e st c o llo id content 84 Bingham, A.C. P l a s t i c i t y . Journal of P h y sical Chemistry kumber 28, 1925. p 201. 85 B le in in g e r, A. V. P reface to The C olloid m atter of Clay and I t s Measurement by Ashley. U. 3. G. 8. B u lle tin h i umber 388, 1909. pp 5-6. 103 were Tery p la s tic * However, th e seventh from the top, C arthage-3, was the most p l a s t i c of th e whole group with La Bajada, th e most highly c o llo id a l sample, about on an equal w ith Banta C lara and San Antonio which were f i f t h and s ix th from th e to p , re s p e c tiv e ly . San IldefonsQ and Kinney, the clays of lowest c o llo id content were a lso lowest in p l a s t i c i t y . Even so th e Kinney clay has a p l a s t i c i t y g re a t enough so th a t i t can be forced from a d ie and made in to hollow t i l e and w ire -c u t b ric k . In f a c t co n sid erab le grog is added to the raw Kinney clay before i t i s pugged, showing th a t i t s p l a s t i c i t y i s ample fo r i t s use in in d u stry . Tonque-H and C arthage-1 are more p l a s t i c than e ith e r Tonque-G or Las Vegas, the two clay s th a t are immediately above them in c o llo id co n ten t. P l a s t i c i t y of th e Gallup f i r e c lay i s f a i r and the same is tru e of K aolin which i s p l a s t i c enough to form o rd in ary shapes w ithout the use of any b a ll- c la y mix. Undoubtedly c o llo id s have e f f e c ts on the p l a s t i c i t i e s 86 of c lay s. Moore and o th e rs found the binding power of t h e i r M u l t r a - c l a y n to be g re a te r than th a t of P o rtlan d cement so we would expect th a t th e fin e clay f r a c tio n would 86 Moore, C harles G. et. a l . Methods of Determining the Amounts of C olloid M atter in B o ils. Journal of in d u s tr ia l and Engineering Chemistry, Lumber 13, 1921. p 327. 104 e x e rt a powerful bonding in flu en c e on the co arser p a r t i c l e s . H iegel8''’ p o in ts out the f a c t th a t the more p l a s t i c the clay i s , in g en e ra l, the stro n g e r the a r t i c l e is in to which i t i s formed. Shrinkage and p l a s t i c i t y are c lo s e ly r e la te d in some cases, a ls o , but t h i s i s a very general r e la tio n s h ip as w ill soon be determ ined. (14) A nalysis of the Clays. The a n a ly s is ta b u la tio n fo r eleven of the clay s i s p resen ted in Table X . Si 0 & v a rie s from 61.65 to 43.26 A1 0 from 17.52 to 38.40 & 5 Fe 03 from 0.75 to 5.48 ft; Ti 0 from tra c e to 0.98 ° J > ; 2 ° 2 Ca 0 from 0.13 to 7.87 M g 0 from tra c e to 3.04 fo\ and combined a lk a lie s from 0.10 % to 2.90 fo . The ty p ic a l adobes such as Sal Indefonso, Kinney, and San Antonio show high co n ten ts of Ca 0. This i s not too o b je c tio n a b le , however, i f the clay i s f in e ly ground and the Ca 0 not produced in lumps during the f i r i n g p ro cess. I f ground as fin e as 30-mesh, Ca 0 c o n ten ts as high as twelve p er cent may be used w ithout excessive weakening of the f ir e d p ro d u ct. B ricks have o ften been made from clay s co n tain in g s t i l l higher per cents of calcium but the r e s u l t s have u s u a lly been u n s a tis f a c to ry . The calcium oxide i s a powerful 87 B iegel, 11. H. I n d u s tr ia l Chemistry. Chemical C atalog Company, 1928. p 159. 105 flu x in g agent in a clay , though not so pow erful as s a l t s of 88 the a l k a li m etals . » * ■ * I f the a lk a lie s are p rese n t as s i l i c a t e s no tro u b le is encountered, but i f the o th e r s a l t s are p rese n t th e f ir e d * ware w ill scum or b lo a t and w ill fu se much q u ick er. The Santa Clara d a y is a good example of t h is as i t i s the h ig h e st in . a l k a l i content and also the clay most e a s ily fu sed . Hies shows ranges of sev eral hundred analyses and fin d s the clay s o n s titu e n ts to vary over wide ran g es. Tor in sta n c e Si 0^ goes from a high of 96.79 % to a low of 2 32.44 ° j t > and the o th e r components follow in a l i k e manner. He a lso p o in ts out the f a c t that g reat v a r ia tio n s may occur 89 in .th e same clay from the same bank and c ite s th re e samples in which Si 0 o v arie d from 51.9 to 59.1 A1 0r , from 2 2 3 17.6 to 28.84 ° / o and 0 from 1.00 to a high of 16.6 With such v a ria tio n s e x is tin g in the same clay one is led to believe th a t an a n a ly sis of a clay i s a minor fa c to r in judging whether or not the clay w ill be u se fu l in in d u stry . 90 M iddleton and o th e rs found th a t th e re were l i t t l e d iffe re n c e s in th e chemical analyses of c o llo id s e n tra e te d 88 H ies, H einrich. Clays, Their Occurrence, P r o p e rtie s and U ses. Wiley, 1937., p 144* ! 89 I b id . p 112. 90 M iddleton, H. H. and O thers. P h y sica l and Chemical C h a ra c te ris tic s of the B oils from the Hrosion Experiment S ta tio n s . U. 8. Department of A g ric u ltu re , Technical B u lle tin Humber 316, 1952. pp 20-24. 106 from s o il s and the re sid u e l e f t behind. They found, in g en eral, th a t the c o llo id f r a c tio n was h ig h er in iro n and alumina and s li g h t l y lower in s i l i c a though exceptions were found in each case. Their analyses in d ic a te th a t much of the calcium i s h eld in c o llo id a l form, Eight years before, 91 Bobinson and Holmes had found th a t th e analyses of se v eral successive e x tra c tio n s from the same s o i l showed only s lig h t v a r ia tio n s . The s li g h t v a ria tio n between th e com positions of c o llo id s in a s o il and the com position of - the re sid u e l e f t a f te r the c o llo id s have been e x tra c te d is much le s s than the v a r ia tio n between samples taken a few f e e t a p a rt in th e same clay bank so such analyses throw l i t t l e l i g h t on the c o llo id problem. In f a c t analyses of clay s are not as u se fu l as th e i r co st would 90 lead one to b e lie v e . ■ S heerer quotes Hewitt Wilson as saying th a t w hile chemical an aly ses are always d e s ira b le they are of secondary importance when compared w ith a c tu a l f i r i n g t e s t s which cost le s s in time and money. 91 Robinson, W . 0. and Holmes, R.S. The Chemical Composition of S o il C o llo id s. U. 3. Department of A g ricu ltu re B u lle tin Humber 1 5 1 1 , 1984. p 19. 98 S heerer, Leonard F. The Clays and Shales of Oklahoma. Oklahoma G eological Survey and Engineering Experiment 3ta tio n B u lle tin Humber 17, 1930. pp 6-7. 107 7~c / /* S*0 ^ PL 0, TiO*. CaO A T 9 0 K^O /Va2.0 Valahle tn Xqntfm T o ta l 5 a n X l d e h m o 6 /. 40 17 S2 3 02 068 b-ZS h 7/ 1.04 7.84 445/ M i n n e y 5142 2 0 -b0 4ol 075 4.45 M t 1-04 4 7i 100*44 Carihafe-/ 6 / U X X A 6 4 3 3 0.65 3.04 0 3 2 0 X4 8 oi loo 2 i G a llu p 54 44 2 7 00 2 / 4 0 4 8 0X 3 a-3b 1-26 8 bf 100.0/ K a o t m 4b-7 3 350$ h l k 0.13 Ir- 1 43 12 98 4 2 bo S a n f l n f a m o S* 10 21.20 3. hi a.SO 7.87 I 44 0 43 10-83 44-52. San la Clar<z 5) 41 1$ A t S48 43 b 4.// 3 0 4 Z 40 14.18 100 45 i / W P l b e r h i H 6/.*5 24 10 %%2 0-81 0-37 104 8 4 0 44-84 Cala W 5 l¥ i 33 45 X-St ---- - 5-80 0-63 0-74 10 . 30 / 06-94 & 5 . H 4 43 2b 38 44 0.75 0.70 l r . 0 . 10 lb/6 9437 Ca 3aj< ?/Xci 54.50 XSOi 4-53 0 4 ! /• 77 0 4 3 1 6 7 1/26 <94 6/ 0 ) P e n ana! C om m un / c aitort 14.C. G arrett o-f Ga/lu/o y M -M - (z) Page 6 5 /V<£*iv Alex/co B u r e a u o f A f/nes B u ile4/n //d /2, (?) Pages 36'4-?s(> Cali fornia 5M te M /n/*f Bureau Bull. /fo99 (f) P a g e 354 5ame Pub /tea4/on (5) P a g e 354 S a m e Pub/tcaf/an 108 (15) P h y sical P ro p e rtie s of th e Clays Before and A fter f i r i n g . The r e s u l t s on: (A) shrinkage on drying, (B) weight lo s s on f i r i n g , (C) shrinkage on f i r i n g , (B) w ater imbibed by the f ir e d clay, (E) wet crushing stre n g th , (F) dry crushing stre n g th , (G) wet te n s ile stre n g th and (H) dry t e n s ile s tre n g th are shown in Table XI . The f i r s t two, A and B a re p resen ted g ra p h ic a lly in P la te 16 and w ill be commented on f i r s t . The shrinkage on drying v arie d from 8 ,8 $ up to 14*9 $ . The San Ildefonso clay which is the low est in c o llo id content was also the low est in shrinkage but i t must be noted th a t t h i s low shrinkage of 2.8 i s a lso shared w ith S. H. 4 clay which i s next to the h ig h e st in c o llo id co n ten t. Santa C lara clay , f i f t h from the top in c o llo id a l m a te ria l, i s h ig h e st in shrinkage w ith 14.9 #, w hile La Bajada, the most h ig h ly c o llo id a l of a l l the samples, i s a r a th e r poor second. The shrinkages of A lb e r h ill and Cala clays are also low and would not serve as in d ic a to rs of th e i r c o llo id co n ten t. Isio p a r a lle lis m e x is ts between the shrinkage on drying and the lo s s of weight on f i r i n g . S. E. 4 which shared the low shrinkage reco rd w ith San Ildefonso was high in f i r i n g lo s s with 17.0 $ . Tonque-Green and Gallup came next in lin e of descending lo s s by f i r i n g but carbon content cannot be charged w ith the lo s s here as i t could have been in 8.H .4. 109 ~7~a6/e X I T h y sic a / P rop erf/es o f C lo y s N R M b l Shrinkage O ft Drying */o /Loss 01 Weight on Firing ^ Shrinkage on Firing a/ /o H x 0 rakrn U p ty C M y * 1 0 Crush mg S1 re/tffs Jbs. per 5 9 Ji/ore m 7en $ i ie 5/rejip% jbs. per Square m Wet P r y We 1 Pry Son Tjde/a/tso x x 4-< 7 -0-2. / 5 Z 982. ... 7 4 4 ZO 7 * 0 3 H m n e y 7‘ 3 4 - 2 0.6 /S. 0 3 5 4 8 349s 3 4 4 310 Tonqjoe - R 4.0 7.(o -0 . 2 /S.+ / S is 4 7435 306 314 C a r th a g e -I 7-0 2 4 7 0 4 4 3 5 7 0 3 4 2 0 4 5 5 1 0 4 0 Tan/floe - G S>6 l h 4 d 7 7k<r 23 20 2 3 95 344 3 8 3 L a s Vegas 6 1 10.0 2 4 74 4 78 7 5 1 4 4 5 2 4 6 308 Ga/lufo 4 -9 1 6 0 3. I 13. 1 2 5 SO 7 7 9 5 5 0 7 4 8 8 M ao h n X.JL 14-4 8 2. 27.7 1247 2090 264 141 C arth age - Z 2 1 & 4 S -4 $-2. 3560 2460 534 3 74 dan tin ton to 6U 4 - 0 1 5 4 3 1870 m o 28 0 5 9 5 Santa C/ara. 74 4 4 3 8 ? O’ 7 4 5 k 437 O O /fiber hi 11 5.JL 2 k X x 1 4 4 2 4 / 0 2.630 4 0 4 3 73 C a / a . 0 4 4 8 3-8 / S o /<* 72 1/60 4 /2 - 4 66 s . t t 4 2 2 7 TO 5 6 21. 1 2 5 4 3 2 745 80 7 2 4 L a T3aja.cLa. 12.2 6 . 5 6-0 o-4 15/0 7 5 7 5 3 31 6 32, S h r i n k a g e s P/afe k6 C o m jo o n e o ! A /t kh P/a/c 4 - I Ptrcotaior Ext rachon /S- Day £ * tract* e* On Dry/nq ()n / < * inp V ? I I ' £ I I ' 0 ^6 I % P V \ § ■ x i > . st * a * ? a , * t i vS . * XI H 1 — 1 o I l l The low est f i r i n g lo s s was th a t of Ban Ild efo n so , the lowest c o llo id member, but th is was c lo se ly follow ed by La Bajada, the h ig h e st, in d ic a tin g th a t no sp e c ia l c o r r e la tio n e x is ts between the c o llo id content of a clay and i t s lo s s of weight on f i r i n g . from a handling sta n d p o in t, shrinkage on drying i s very im portant. La Bajada and Banta C lara clays crack badly on drying in a i r and had to be d rie d very c a re fu lly in the dark with c a re fu l c o n tro l of the r e l a t i v e humidity before good specimens f o r f i r i n g were o b tain ed . Carthage-2 was a lso a source of tro u b le on drying though K aolin, w ith a s li g h tly g re a te r shrinkage, o ffe re d no tro u b le at a l l . A ll of the clay samples had s u f f ic ie n t dry stre n g th to be handled w ithout danger of breakage and only Ban Ildefonso and Tonque-Green had to be handled c a re fu lly . 93 A ustin te s te d a number of t i l e and b ric k clays and 94 found shrinkages running from two to nine per c e n t. Lyre found the average shrinkage of adobe clays to be seventeen per cent which high fig u re can be accounted f o r by the f a c t th a t the adobe mix i s q u ite f lu id and not to be confused with 93 A ustin, C hester R. S urface Clays and Bhales of Ohio. Ohio B tate U n iv ersity en g in eerin g experim ent s ta tio n B u lle tin Lumber 61, 1934. pp 32-53. 94 Lyre, Thomas T. The P h y sical P ro p e rtie s of Adobe Used as a B uilding M a te ria l. U n iv e rsity of Lew Mexico B u lle tin , »«hole Lumber 263, 1935. p 16. 112 the th ic k , p l a s t i c column extruded from a p u g -m ill. This i s 95 in accord with Ladd who mixed clay s w ith ju s t enough w ater to make them p l a s t i c and found t h e ir shrinkage to vary from one to ten p er c e n t. When he remixed the clays and d ried them from a co n d itio n o f s e l f - s a tu r a tio n he found the shrinkage v a rie d from f iv e up to t h i r t y per ce n t, P la te 16 re p re s e n ts th e shrinkage lo s s e s on drying and on f i r i n g , P la te 17 shows lo s s of weight on f i r i n g and and w ater taken up by the f ir e d samples. From the g rap h ic al re p re s e n ta tio n we come to th e conclusion th a t th e re i s no r e la tio n s h ip between fire -s h rin k a g e and c o llo id co n ten t. Now, however, we are more concerned with what l i g h t th ese data m ight c a st upon t h e i r use f o r b u ild in g m a te ria ls or r e f r a c t o r i e s . Statem ents are q u ite o fte n made th a t the clay which sh rin k s a g re a t deal on drying w ill sh rin k but l i t t l e on f i r i n g and v ice v ersa . This statem ent is not in accord w ith th e f i r i n g process used here. San Ildefonso w ith the low est of dry shrinkages showed an expansion of 0.2 on f i r i n g as did fonque-Red which is th ir d from the bottom in dry shrinkage. Santa C lara which was h ig h e st in dry 95 Ladd, George. A P relim in ary Report of a P a rt of the Clays of Georgia. Georgia G eological Survey B u lle tin 6 a, 1898. p 28. 96 R ieg el, A. R. I n d u s tr ia l C hem istry. Chemical C atalog Company, 1928. p 159. 115 \ \ S 5 < 5 I- <o .$ N \ V 5 % $ Q j £ .5* £ < o i r - t / v s/s /s's a > v 5 ,< ? S < o * ^ I -i ^ ^ a ? Jb * y v 'S 4 . " 4 1 V I v J JS r > A 4 * ' * » v - X S 7 5 3 S 2 L - w j yyj A V '/ 7 / r 7 7 7 * ' / / ? u — ■ — _ 7 7 7 7 7 * 7 7 7 / 7 7 . ' r n ---I ■ — - ■■_— - ' i_____I 1 W 7 7 7 7 7 ^ 1 12 M 77?/ / / SM / / m vm , 7J77& r m V' / ' fc» - * v • f f< K t * . r 'H i 0/0/ > * y * ' w IU f/ U & , V < ?-rtt> 0tfJJV > '^ ’ . ■7 ZT7 / y 7 7 7 Z>7 . -> n ift> £ ) n j p A h x? I X ' ? n fa u p j 77Z'VTy,77Z/.7Z:.: i “ y^',] urc , " ^ ) A v 114 shrinkage a lso was the h ig h e s t in the fire -s h rin k a g e column. In the case of A lb e r h ill, C ala, and S. H. 4, San Antonio and G allup, the shrinkages both on drying and on f i r i n g were low and only in Kinney and La Bajada were th e re d i s t i n c t in v erse r e la tio n s h ip s between the two. In the w ater a d so rp tio n fo r th e f ir e d clay s, S. H. 4 and K aolin led w ith 21.1 fo , both being f ir e d to a porous s o lid . Ban Ild efo n so , Kinney, Tonque-Red, Tonque-Green, Las Vegas, A lb e r h ill, and Cala ran as high as f i f t e e n per cent while Banta C lara and La Bajada were lows with 0 .1 and 0 .4 re s p e c tiv e ly . In these l a t t e r cases the reason i s very ap p aren t. Santa Clara clay slagged down eomple' te ly , the fused g la s s e s f i l l i n g the pore space. The o th er c la y , La Bajada, re a c te d in the same manner but to s l i ^ i t l y le s s e r degree. In both eases th e tem perature of the k iln was above the range f o r t h e i r f i r i n g . Both of th ese clays are used su c c e s sfu lly by the Indians in making p o tte ry and the low fu sin g p o in ts a re advantages because the Indians w ith t h e i r crude f i r i n g means cannot o b ta in a heat g re a te r 97 than 800 degrees C entigrade or about cone 014. On the o th e r hand, some of the clay s te s te d here are of the r e f r a c to r y v a rie ty . This i s s p e c ia lly tru e of the C a lifo rn ia c la y s, A lb e r h ill, C ala, and 8. H. 4 and the K aolin from the Organ m ountains. 97 no Lame. Btandard Pyrom etrio Cones. The Ldward Orton J r . Ceramic fo u n d atio n . No d a te , pp 16-17. 115 Q D D ie tric h gives the f i r i n g properties of Alberhill, Cala, and S. E. 4 samples as follow s: (A) A lb e r h ill. S te e l hardness a t cone 02 (1095 degrees C) v i t i r f i c a t i o n w ell along a t cone 15 (1350), maximum lin e a r shrinkage, p l a s t i c b a s is , i s 13*3 i at cone 13, b lo a tin g appears at cone 15 (1410), th e so fte n in g p o in t i s cone 29 (1640). (A ll tem peratures su p p lied by w r i te r ) . (B) C ala. S te e l hardness a t cone 1 (1125), V itr i f i c a t i o n complete a t cone 9, (1250) . Maximum t o t a l lin e a r shrinkage, p l a s t i c b a s is , i s 21.5 i at cone 11 (1285). S oftening p o in t, cones 31-32 (1680-1700). (C) S. H. 4 . S te e l hardness at cone 02 (1095), 22 % shrinkage at cone 15 (1410) • S often in g p o in t cone 34, (1760). The p ro p e rtie s of Organ K aolin are somewhat sim ila r to 99 those of the th re e r e f r a c t o r i e s l i s t e d above. Talmage gives the k a o lin ’s so fte n in g p o in t as not below cone 30, a tem perature of 1650 degrees, C entigrade. W e would n a tu r a lly expect th a t a tem perature of about 1000 degrees C entigrade 98 D ie tric h , Waldemar Benn. The Clay Hesources and the Ceramic In d u stry of C a lif o r n ia . C a lifo rn ia S ta te Mining Bureau B u lle tin Number 99, 1928. pp 287,146, 273. 99 Talmage, S te r lin g and Wootton, T. P. The N on-M etallio M ineral Resources of New Mexico and Their Economic F e a tu re s . New' Mexico S ta te Bureau of Mines B u lle tin Number 12, 1937. p 65. 116 (cone 07) would not im pair th e p o ro s ity of th es$ r e f r a c t o r i e s . In f a c t i t i s probable th a t the tem perature a t which they were f ir e d in t h i s experiment was not high enough fo r them to develop anything lik e the maximum s tre n g th of which they are capable. I n c id e n ta lly a l l of the fo u r r e f r a c to r ie s mentioned, A lb e r h ill, Gala, 3. H. 4 and Organ K aolin w ill pass the standard s p e c if ic a tio n s f o r high heat duty b ric k . Crushing s tre n g th s are presen ted g ra p h ic a lly in P la te 18. The low est value i s found in Santa C lara which, as mentioned b efo re, had com pletely slagged down and was h ard ly to be co n sid ered . The next low est was S. H. 4, a high alumina r e f r a c to r y which had not been burned a t high enough tem perature fo r i t to develop any g reat stre n g th . S tre n g th of the K aolin was low f o r the same reaso n . In a c tu a l p ra c tic e the crushing s tre n g th of a b ric k i s not of too g re a t im portance sin ce even a weak b ric k w ill have ten tim es the stre n g th of the bonding m a te ria l. S e a r le ^ ^ l i s t s crushing s tre n g th s from 74 to 583 tons per square fo o t and recommends a stre n g th of not le s s than 90 tons per 100 Committee C-8 Manual of A. S. T. M * Standards On R efrac to ry M a te ria ls . The American S o cie ty of T esting M a te ria ls, P h ila d e lp h ia , November 1937. pp 96-97# 101 S earle , A lb ert B. The Chemistry and P hysics of Clays and Other Ceramic M a te ria ls. Benn, London, 1933. pp 198-200. 0 0 0 7 000 r ^ OCCf 1? 9t » / c t Y t.'M f>zjK>a/*uoQ ZU0/^ /0 s* /- / / » 9 / * W d P la t e 19 Samples of the f if t e e n clays a f t e r being f i r e d . The m etal molds are shown in both cases. Note Santa C lara Sample f i f t h from the r ig h t. Here v i t r i f i c a t i o n has been complete and the clay das b lo a ted and slumped down. 119 square f o o t. T ra n sla ted in to pounds per square inch these v a r ia tio n s would be roughly from 1000 to 8000 with a minimum stan d ard of 1250 pounds per square inch. Only th re e of the f ir e d samples te s te d f a l l below th e recommended minimum in the dry s ta te while the o th e rs have very high crushing s tre n g th s . Water was added to the f ir e d clay samples to sla c k any f r e e calcium oxide th a t might have been formed in the f i r i n g p ro cess. I t i s w ell known th a t calcium oxide has a weakening e f f e c t upon f i r e d ware and t h i s i s a quick way to show such a weakness up. In a l l these c la y s, however, fin e g rin d in g was p ra c tic e d with th e r e s u l t th a t f a i r l y high co n ten ts of calcium had not been s p e c ia lly d e trim e n ta l. The f in e ly d iv id ed calcium s a l t s had not been burned to calcium oxide but had combined w ith Si 0 to form one of the se v e ra l calcium s i l i c a t e s found in P o rtla n d cement. The presence of such calcium s i l i c a t e s would not in ju re the s tre n g th of the f i r e d ware on w ettin g . In f a c t is was found on te s tin g th a t e ig h t of th e samples had in creased th e i r s tre n g th by being soaked in w ater. The San Antonio clay contained alm ost e ig h t per cent calcium oxide, by a n a ly s is . This sample had s l i g h t l y g re a te r stre n g th a f t e r i t had been soaked f o r one day in hot w ater and allowed to dry an a d d itio n a l two days. E vidently the calcium s i l i c a t e formed acted as a f a i r bonding medium. Tensile strengths of the fired samples are presented graphically in Flat# B O * Ear# again w © see that the soaking in water has not Impaired the strength of fit# of the fired samples though i t has done so in eight* in the ceee of can lld e f one© the tm value# are about the name and in. the ease of haata Glare the briqnettes earn # apart in the hands ana a value of zmT o we* listed * the oan /j&tonio sample, high in calcium, showed a greater dry strength than wet strength* hew the calcium arid# ©onts&t actually showed u p * I t h ad not done so in the crushing teats* th is mm probably due to th# foot that the fired cylinders were allowed to , > * '• r . " ” dry two days and the briquettes hut one* Foasibly : t ^ ^ lc ii® formed by the firin g bad more time to se t fo r the sai^ lee than for the tension samples* The te n sile strength of a brick Is usually not of such orteno© as & brick i s rarely placet in a position where tension in a h o w ev er, th e c l a y s with high te n sile -strength# w ill also have- a high transverse rupture modulus ant th is i s a property mmtimm called for in clay products* Cerml© gives tonsil© strengths varying from lib to 100$ pound# per. square inch* The clays used here, with the exception of Canta Clara, rang© from a 1GB ! * » AwMSfr P / a / e £ 0 P&/7S//6 S fn esiyf/} o f P i r e d l/da Gompoure d k/tih P/a (g tS?l /Ana IS- Day i.fcs 1 I-* JXi 12 2 dry value of 124 up to 1090 pounds per square inch. The low value fo r th e S. H. 4 i s probably due to the f a c t th a t the f i r i n g tem perature here was not s u f f i c ie n t to bring and g re a t amount of fu sio n , and hence no g re a t bonding s tre n g th . The Carthage-1 clay had an e x c e p tio n a lly high stre n g th both in compression and te n sio n . I f th is clay were in a reg io n of g r e a te r p o p u latio n i t would probably be used e x te n siv e ly as a paving b ric k and fo r many o th e r purposes where an e x c e p tio n a lly stro n g a r t i c l e is demanded. As i t i s , sin c e th is te s tin g work has s ta r t e d , some of t h i s clay is being shipped to a lo c a l b ric k p la n t. Carthage-2 is lik ew ise being used, ex p erim en tally , to produce some slip p e d b ric k s f o r fa c in g s on b u ild in g s. As s p e c ifie d under Kxperimenta1 tfork, each f ir e d product was su b jected to an exam ination fo r c o lo r, hardness, evidence of s p a llin g , b lo a tin g , e tc . A b r ie f ta b u la tio n of the r e s u l t s fo llo w s, the samples being taken up in the u su al order in which they have appeared in the graphs. A photograph of the samples arranged in th e same order is found in P la te 19. San Ild e fo n s o . This clay forms a s o f t, sandy b ric k with medium dark red c o lo r. Kinney: This i s a smooth, f a i r l y hard b ric k of salm on-red c o lo r, somewhat l i g h t e r in co lo r than the San Ildefonso sample. Tongue-Red: This forms a s o f t, sandy sample much lik e San Ildefonso but a li g h t e r 123 salm on-red c o lo r. C arth ag e-1 s Sample is dense, smooth and hard and of a d u ll salm on-red. Tongue-Green: This clay forms \ a s l i g h t l y sandy, hard b ric k of a b u ff to p ale red c o lo r. Las Vegas:This sample i s smooth, hard, and p o ssesses an even deep red co lo r much in demand as a face b r i c k .G allu p : This f i r e b ric k i s smooth and hard and of a cream c o lo r. K aolin: B rick is of a d u ll gray co lo r w ith hard s ilk y su rfa c e . Carthag®-2. This sample came near i t s fu sio n p o in t forming very hard and dense m a te ria l w ith smooth su rface and a dark red co lo r lik e the Las Vegas b ric k . Ban Antonio: This b ric k was f a i r l y hard and dense but of a m o ttled salmon red c o lo r. Santa C lara: This clay fused to a dark re d ,g la s s y mass, badly b lo ated , cracked, and scummed. A lb e r h ill: B rick i s f a i r l y hard and dense having a smooth f in is h and a lig h t cream c o lo r. Gala: This sample was much l i k e A lb e r h ill except fo r th e co lo r which is a darker cream’to buff with a yellow -red tin g e , 5. H. 4 : This sample burns to almost pure white forming a s o f t and porous b ric k . La Bajada: This clay had ju s t reached i t s fu sio n p o in t burning to a hard, heavy, g la ssy mass of dark red c o lo r. Some b lo a tin g was s t a r t i n g . & & & sr. *; i - I v r , ft r O' i r i tv 3 sr | r o Si J p ? a X O' S' ft 0 f t “ A * u ^ ft 5 0 5 ft 1 1 r r N A M E o { C U Y > 4 '-ft £ V i V O' < * ) V * ft O' -ft N ft X N C l ft O * < B « f t r f t k v S y f t V i P V J 4 V s H 2 4 hr. E*tr*cti«* hy f^rco/aior $ b K > f t Q x V i N X •b K < * > ft >s u s b * x N N M X X x U i ft X V i O K O s * H ft 0 * t S V * y /£jcfracf/on S * < V f to & ft S i ? > 4 V * ft ft o\ ft ( V >i C l S 4 4 K » -4 X ft f t N St.. “ X . ■ 4 ft O' v > ^ b 3kr. £ r firAC i/»* by P/fretfe N X V i x x 60 V i SI < V f t f t x f t r ft ft V i N * * ft v » ~ 4 V i f* ? ft ^ 2 4 hr. Sximcii** by fi pertc N -4 -ft O' * * f t O s O' $ < > o f t O b X ft < v U ) ■ 4 O ' P % **$ Metky)•v%0 QUt* per $r*m c /o v s a* v > f t N ■ f t V . 0 * X X X f t f t O' X ft £ X ft ft X X b Vi b X ft • -4 Colloid Koii° of Mofklkn* &»* ffUtorpthn V * & \ -I rv 2 f t N * V i f t * a X O ' ft X O ' 0 * o- V i ft ft v » -4 V * V i V i ft ft Malachite Orocn /U ttrph " * f/9 r Oja late Traoimmf X O . O ' « X ft > 4 f t ft & X V ) f t X £ ft V i a i o te < v H 0 * ft V I Malachite (Proem 4jfrpt>*» **f/$rm m o f Haw Clan X X •ft * ft N O - > 4 ft X ^ X ? • < > f t s O ' (o 0- O' E f t ft S « 4 < V a* X ft < v ^ ffAsorptiom of H xO in fin- Cent Rdiorrt"* °+NH3 • ■ o O f t f t f t > K i X . V i -ft § > -ft N b ft ■ o s ftl ft f t O' O ' f t f t ft ft O' H •ft • * 4 m 3 per gram S * ft* V ft ■ x . ft V i > » ft ft Jft V l O' s ft Si b N N V * V V Ki •b ff& arf o f h/eff/nf Calories per gram $ S V I X V i ft ft £ ft a f t ft V i $ ft. X X a C M N &a Clt fietsorpt'om my- per gram X V i N v * 0 * O s V < V V l $ -ft O' O' s ft* ft. f t f t O s O i r r X I a f t a - 1 b* V 4 £ s* • fhrcent Shrinkaf« on Vrginq O' < * X > 1 o o* O f t f t -$ ft ft! O' O' * X :* ft X 5 » O' £ a X ■ f t •« ♦ O' f t b* •4 far Cent W015M Los* on Firing < A ► ifi ft» V * Si 1 S > N ft 1 ft w 0 fierce** Shrink*** V 9» N t? ft ♦ f t o O ' 5 » on firm q a -V V » X x S ft X 5 O v f t x f t f t v » v* X X X ft < 1 f t -b X U V X « * % tfx 0 Heliorbcd by F ired C/a y V * 1 ■ s . ft > * 5 f t V f t % 1 * ft X ft X f t c ? f t p a X > • f t * 4 V * < * * 5f X i V * h * is g 1 § 5 -x Crushing gfrengih o f fire d h/are Pounds /* * i* X ( A 4 u V l • S . -ft fc X X ft ft V i ft ft ■ f t f t ft 4 £ ft ft tv ft 4 ft ft ft 1 X $ * 1 ft 0 1 2 a $ £ * > 1 « •a V I ft ft* ft -IV X V i ft ft ft O S f t * V * O ' ft & * 4 Vi $ a a 5 t * • a ft pi . .. I ft X Tcnu /e Sfreeaff} 0 f fir e d Ware * ft ft * 1 . V * 1 W > 1 ft § ft ft X • f t X \ s C f t *• 0 * V * l x * 0 s Ir t M ft B ' ** ’ t X F u etd sy/# i " . * O ' ;fe (V 9\ N * fti O' V 4 9 ^ o « % ■ V * O' O' O ' f t O' i > 1 ^ r 4 > X » ft n J s p H o f Harr C lay v V > i * ft ft S ' £ * i i | r ft a *£ N l ? * S' ■ f « i & f c i S ' S' 5 ? * ^ x £ 0 -.,.jL » 5 * l i a , S ' 5 ft < ^ O s ? * ft* 5 r _ a — N A M E of CLHY Chapter V CONCLUSIONS One who read s th e l i t e r a t u r e o f c o llo id chem istry, p a r tic u la r ly in th e f i e l d of clays and s o il s , w ill be stru c k by th e v a rie ty of d e f in itio n s fo r the term c o llo id * He can hard ly f a i l to agree th at re se a rc h in s o i l chem istry could be g re a tly a s s is te d by the workers in the f i e l d agreeing e x a c tly what s h a ll be considered c o llo id a l. In t h i s re se a rc h clay s have been found e x c e lle n t media fo r work in c o llo id ex p erim en tatio n . A general c o r r e la tio n between c o llo id content and th e v ario u s physical and chemical p r o p e r tie s of the clay s c le a r ly e x i s t s . I t has been shown th a t the only sure way of o b ta in in g accu rate measurement of c o llo id content i s by e x tra c tin g the c o llo id a l m a te ria l and weighing i t . This can be a tta in e d by a m u ltip le e x tra c tio n with w ater and an a p p ro p ria te d isp e rs in g agent, the procedure being c a rrie d on as long as a tu rb id so l remains a f te r tw enty-four hours of s e t t l i n g . Ten such e x tra c tio n s are u su a lly s u f f i c i e n t . As a second b e s t, though much in f e r io r , method i t is suggested that the e n tir e clay so l be evaporated to dryness and weighed in accordance with the p e rc o la to r method. P ip e tte methods or any sim ila r procedure involving only an a liq u o t p a r t of the so l are c lasse d as u n s a tis f a c to ry . A number of the advooated c o llo id methods have been te s te d as to t h e i r accuracy. A f t e r . ev a lu a tin g methods in v o lv ing hydrom eters, v isc o sim e te rs, ad so rp tio n of vario u s substances, base exchange, heat of w ettin g , p l a s t i c i t y , u ltram icro sco p es, sedim entation dev ices, ion m ig ratio n , e t c . , i t is concluded th a t w hile some p a r tle n la r r a p id .m e th o d may approxim ate a q u a n tita tiv e d eterm in atio n in a few sim ila r cases i t f a i l s to dp so when a wide s e le c tio n of \ m a te ria ls i s taken, ffo one method or com bination of: ra p id methods proved i t s worth over more than a narrow s e le c tio n of m a te ria ls . A quick method of determ ining c o llo id content of clays and s o i l s is needed,. . . . . . , but i t i s im possible to avoid the conclusion th a t none of th e quick methods advocated i s s a tis f a c to r y and i t is apparent th a t much f u r th e r work is re q u ire d i f a ra p id method i s to be found. D IB -u -L Y 1 5 12 22 25 27 28 29 SO 31 32 33 34 36 37 40 41 1 2 7 BOOKS Holmes, Harry N* In tro d u c to ry C o llo id a l Chemistry, John Wiley and Son 1934. pp 198. S e a rle , A lb ert B. The Chemistry and P hysics of Clays and Other Ceramic M a te ria ls. Benn, London, 1933. pp 738. P irrs o n , Louis V. and Knopf, Adolph. Books and Hook M inerals. John 'Wiley and Son 1926. pp 426. Holmes, Harry N. L aboratory Manual of C olloid Chemistry. John Wiley and son 1934. pp 229. K o lth o ff, I . M. and Purman, N.H. P o ten tio m etrio T i t r a t i o n , John Wiley and Son 1926. pp 345. Ware, John C. Chemistry of the C o llo id al S ta te . John Wiley and Son 1930. pp 314. Svedburg, The. C olloid C hem istry. The Chemical Catalog Company 1928. pp 302. Holmes, Harry N. Same as 22 above. Jenny, Hans. P ro p e rtie s of C o llo id s. S tan fo rd U n iv ersity P re ss 1938. Houser, E rn st. C olloid Phenomena. M o Craw H ill 1939. pp 294. Jenny, Hans, Same as 30 above. Holmes, Harry N. Same as 22 above. H ies, H ein rich . Clays, T heir Occurrence, P ro p e rtie s and U ses. John Wiley and Son 1937. pp 613* von Buzagh, A. C olloid System. T echnical P re ss 1937. pp 331. S e a rle , A lb ert B. Same as number 5 above. Washington, H. S. A nalysis of Bocks. John Wiley and Son 1930. pp 296. Low, A lbert H. Technical Methods of Ore A n aly sis. John ; ^ ile y and Son 1919. pp 388. 42 43 44 50 52 58 59 60 65 73 74 76 83 87 8 8 89 96 128 S c o tt, W ilfred W . Standard Methods of Chemical A n a ly sis. Van Nostrand 1929. pp 880. W illard, Hobart and Furman, N. H. Filamentary Q u a n tita tiv e A n aly sis. Van N o stran d ■ 1935. pp 419. H ies, H ein rich . Same as 34 on page 127. Ware, John C. Same as 27 on page 127. Holmes, Harry N. Same as 1 on page 127. Ware, John C. Same as 34 on page 127. B le in in g e r, A. V. The P ro p e rtie s of Clays in C olloid Symposium Monograph Yol. 7, 1929. Chemical Catalog Company 1925. pp 80-98. Jenny, Hans. Same as 30 on page 127. H ies, H ein ric h . Same as 34 on page 127. Holmes, Harry N. Same as 22 on page 127. Wilson, H ew itt. Ceramics and Clay Technology. M e Craw H ill 1927. pp 296. Jenny, Hans. Same as 30 on page 127. Jenny, Hans. Same as 30 on page 127. E ie g e l, H. H. I n d u s tr ia l Chem istry. Chemical C atalog Company 1928. pp649. H ies, H ein ric h . Same as 34 on page 127. H ies, H ein rich . Same as 34 on page 127. H iegel, E. H. Same as 87 above. BULLETINS 129 2 Eno, F. ii. Highway S ubsoil In v e s tig a tio n s in Ohio. Ohio S ta te U n iv ersity Engineering Experiment S ta tio n B u lle tin Number 39, 1928. pp 64. 4 H ilgard, A. E stim ation of C o llo id a l M aterial in S o ils by A dsorption. U. 8. Department of A g ric u ltu re b u lle tin Number 1193, 1924. pp 42. 6 G ile, P. L. E stim ation of C o llo id a l M a te ria l in b o ils by A dsorption. U. 8. Department of A g ric u ltu re B u lle tin Number 1193, 1924. pp 42. 7 Wilson, H ew itt. Keramic K iln s . Denver F ire Clay Company B u lle tin Number 360. Ho d a te , pp 45. 8 Eakin, Henry M . and Brown, Garl B. S ilti n g of R e se rv o irs. U. 8. Department of A g ric u ltu re Technical B u lle tin Number 524, 1939. pp 168. 9 S a lin a s, S a la z a r, M inerales R e su lta n te s d el A lte ra cio n de Booas o de Fenomenos de Metamorfismo. Anales del I n s t i t u t o de Geologia de M eiico, Tomo 4, 1930. pp 161. 1 0 .S a lin a s, S a la z a r. A nalisis^E echos en_ e l L aboratorio de ^uimioa d el I n s t i t u t o Qeologipo de Mexico. P arergornes del i n s t i t u t o Geologico de Mexico. Tomo 5, Numero 4, 1913. 11 Darton, N. H. Bed Beds and A ssociated Formations of New Mexico. U. 8. G. S. B u lle tin 794, 1928. pp 356. 13 Talmage, s t e r l i n g B. w ith Wootton, T. P. The N onm etaliic M ineral Besouroes of New Mexico and Their Economic Importance. New Mexico S ta te Bureau of Mines B u lle tin Number 12, 1937. pp 159. 14 Dunham, K ingsley. The Geology of th e Organ Mountains. New Mexico s t a t e Bureau of Mines B u lle tin Number 11, 1935. pp 272. 16 D ie tric h , Waldemar Fenn. The Clay Besouroes ana the Ceramic In d u stry of C a lifo rn ia . C a lifo rn ia S ta te Mining Bureau B u lle tin Number 99, 1928. pp 290. 17 Same as 16 above. 18 Same as 16 above. ISO 20 Anderson, M. S. A dsorption of C o llo id s and N on-C olloidal S o il. U. S. Department of A g ric u ltu re B u lle tin Number 1122, 1922. 2b Anonymous. Notes on Hydrogen Ion Measurement. Leeds- Northrup Company. No Date, pp 30. S5 B a rre ra , Tomas. Las A rc illa s y l a ffabricacion de Loza de Oaxaca. Anales d el I n s t i t u t o Geologico de Mexico Tomo 4, 19S0. pp 161. S9 A ustin, C hester R. The Surface Clays and Shales of Ohio. The Ohio s ta te U n iv e rsity Engineering Experiment S ta tio n B u lle tin Number 81, 1934. pp 53. 57 Ashley, H arrison. The C olloid M atter of a Clay and i t s Measurement. U. S. G. S. B u lle tin Number 388, 1909.pp 65. 61 Rueckel, W alter C. R esearches on D ry-Press R e fr a c to rie s . Ohio S ta te U n iv e rsity Engineering Experiment S ta tio n B u lle tin Number 82, 1934. pp 36. 63 Anderson, M.S. and M attson, Svante. P ro p e rtie s of the C o llo id a l S o il M a te ria l. U. S. Department of A g ric u ltu re B u lle tin Number 1452, 1926. pp 47. 67 E v erh art, J.O . w ith Rueckel, W.C. and A ustin C.R. Barium Hydroxide to P revent Scumming of Ceramic P ro d u c ts. Ohio S ta te U n iv e rsity E ngineering Experiment S ta tio n B irc u la r Number 30, 1934. pp 16. 77 Baver, L.D. R esearch B u lle tin Number 129. M issouri A g ric u ltu ra l Experiment S ta tio n , 1929. pp 48. 80 B ra d fie ld , R ichard. The Chemical Nature of a C o llo id a l C lay. U n iv ersity of M issouri A g ric u ltu ra l Experiment S ta tio n , Research B u lle tin Number 60, 1923. 81 B ra d fie ld , R ichard. Same as number 80, above. 85 B le in in g e r, A. V. P reface to The C olloid M atter of Clay and I t s Measurement. U.S.G.S. B u lle tin 388, 1909. pp 65. 90 M iddleton, H. E. and o th e rs . P h y sical and Chemical C h a ra c te ris tic s of the S o ils from the Erosion Experiment s t a t i o n s . U. s . Department of Agriculture Technical B u lle tin Number 316, 1932 . pp 50. 131 91 Robinson, « » . 0* and. Holmes R.S. The Chemical Composition Of S o il C o llo id s, u. S. Department of A g ricu ltu re B u lle tin Number 1311, 1924. pp 19. 90 S heerer, Leonard E. The Clays and Shales of. Oklahoma. Oklahoma G eological Survey and Engineering Experiment S ta tio n B u lle tin Number 17, 1930. pp 251. 93 A ustin, C hester R. Same as 39 on page 130. 94 Eyre, Thomas T. The P h y sical P ro p e rtie s of Adobe Used As a B u ild in g m a te ria l. U n iv ersity of New Mexico te'ul'ietin, whole Number 263, 1935. pp 32. 95 Ladd, George F9 A P re lim in ary Report of P a rt of the Georgia Clays. G eological Survey of Georgia, B u lle tin Number 6 a, 1898. pp 130 . 97 Anonymous. Standard Pyrom etric Cones. The Edward Orton J r . Ceramic Boundat io n * No d a te , pp 44. 98 D ie tric h , Waldemar Benn. Same as number 16, page 129. 99 Talmage, S te r lin g B. Same as number 13, page 129. 101 Committee C~8, Manual A. S. T. M. S tandards on R e frac to ry M a te r ia ls . American S ociety of T esting M a te ria ls, November 1937. pp 180. om sRs 122 2 Bouyoucos, George John* The Hydrometer as a New and Rapid Method of Beterraining the C olloid Content of B o ils. B oil Science Number 22, 1927. pp 239-297. 15 B entley, L. B. P erso n al Communication, November 1, 1928. 21 G ile, P. L. C o llo id a l B oil M a te ria l. B oil Science Number 25, 1928. p 361. 24 H arrington, N. R. Rapid Q u a n tita tiv e Method fo r the D eterm ination of Barium. The Chem ist-A nalyst at some fu tu re date'. Acceptance n o tic e rece iv ed December 1, 1939. 28 Davis, N. B. The P l a s t i c i t y of Clays and I t s R e la tio n to Mode of O rigin. T ran sactio n s of the American I n s t i t u t e of Mining B ngineers, Volume 51, 1915.pp 451-480. 45 Bouyoucos, George John. Same as number 2, above. 46 Bouyoucos, George John. Same as number 3, above. 47 Bouyoucos, George John. S tu d ies on the D ispersion Procedure Used in the Hydrometer Method f o r Making M echanical Analyses of' S o ilsT S o iI S cien ce, Number 33, 1932. pp 343-354. 48 Keen, B. A. Borne Comments on the Hydrometer Method fo r Studying S o i l s . SoiiHs'cience dumber 26« 1928. pp' 261-263. 49 P u ri, Arnar Nath. A New Type of Hydrometer fo r the Mechanical A nalysis of B o ils . B oil Science Number 23, 1932. pp 241-248. 51 Paneth, T r i tz and Thimann, Wilhelm. The A dsorption of Dye by C ry s ta ls . B e ric h te , 57 B, 1924. p 1221. 33 Pallmann, H. Uber Bodenbildung und Bodenferein en der Bchwels. Die Brnahrung Der P fla n ze , B e rlin , Juli' 1934. pp 225-234. 54 G ile, P. L. Same as number 21, above. 55 H arrington, B. R. Some P relim in ary In v e s tig a tio n s on C o llo id a l Clays. U n iv ersity of Southern C a lifo rn ia Report fo r Chemistry 290-L, 1935. pp 26. 56 62 64 66 68 69 70 71 72 75 78 79 82 1 5 5 T hies, H. R. R e latio n ' Between Dye A dsorption of Clays And T heir Behavior in Rubber Compounds. Jo u rn al of I n d u s tr ia l and Engineering Chemistry Number 17, 1925. pp 1165-1169. Moore, C harles G and o th e rs . Methods fo r Determ ining the Amount of C o llo id a l M a te ria l i n S o i l s T Journal of In d u s tr ia l "and E ngineering Chemistry, Number 15, 1921. pp 527-530. Anderson, M. 8. The Heat of G etting of (Soil C o llo id s ,. Jo u rn al of A g ric u ltu ra l Research, Volume 28, Number 6, 1924. pp 927-935. Murray, H.D. The C oagulation of C olloids by E le c tr o ly te s . Cneiriical News Number 125, 1921. pp 277-279. 8 ch o llen b erg er, 0. J . Determ ining the R eplaceable Bases In S o ils by the Ammonium A cetate Method. Science, Number 65, 1927. pp 552 and 1692. Whitney, M ilton. The O rigin of S o il C olloids and T heir Reason Tor E x istin g in This S ta te of M atter. Science, Volume 54, 1921. pp 652-656. Davis, N. B. Same as number 38 on page 132. G ile, P. L. Same as number 21, page 132. s te n z e l, R. W . The Heat of Immersion of S ilic a Gel in Various Petroleum S ubstances. Jo u rn al of th e American Chemical S o ciety Number 54, 1932. pp 870-876. Powers, W . L. A Study of th e C olloid F ra c tio n of C e rtain S o ils Having a R e s tric te d D rainage. S o il Science Number 23, 1927. pp 487-491. B ra d fie ld , R ichard. Proceedings of th e F i r s t I n t e r n a tio n a l Congress of S o il Science, Volume 5, 1928. pp 858-869. Coward, H. F. Sedim entation of B e n to n ite . Jo u rn al of the Chemical S o cie ty Volume 125, 1924. pp 1470-1474. F rance, Lesley G. An U ltra-M icroscopic Study of the R e latio n sh ip of C o llo id a l Content and P l a s t i c i t y o f C lays. Jo u rn al of the iuaerican Chemical S o cie ty Volume 9, Number 2, 1926. pp 67-76. 134 84 Bingham, A. C. P l a s t i c i t y . Jo u rn al of P hysical Chemistry, Number 29, 1925. pp 1201-1204. 86 Moore, C harles G. and o th e rs . Methods of Determ ining The M ount of C o llo id al M a te ria l in S o ils . Jo u rn al of Indust r i a l and E ngineering Ch emi s tr y , Number 13, 1921. pp 527-530. 103 Ceramic Age, Number 28, 1937. pp 197-207. UMI Number: DP21727 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. UMI Dissertation Publishing UMI DP21727 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code ProOuest ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106- 1346

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

An experimental investigation of amino compounds of boron hydrides

PDF

A critical study of the hydrogenation of talloel

PDF

Addition compounds of boron fluoride with certain sulfoxy-amines

PDF

A study of the reaction of 2,4-dinitrobenzene-sulfenyl halides with ketones

PDF

A study of the reaction of tetramethyldiborane with sodium in liquid ammonia and the products derived from this reaction

PDF

A study of the activation of hydrogen by cuprous salts in quinoline / by Max K. Barsh

PDF

A study of the reactions and properties of petroleum sulfuric acid sludge and some of its derivatives

PDF

Adrenalectomy and fat absorption

PDF

A new approach to the problem of measuring the properties of micelles

PDF

Calculations of electrostatic interactions in proteins

PDF

Chiroptical spectroscopy of nitrogenase

PDF

A study of the interactions of polar gases with solid proteins and some simple organic compounds

PDF

A study of the structures and properties of triaryl compounds of boron

PDF

A determination of ionic diffusion coefficients

PDF

A study of chemical tests for the vitamins

PDF

Contributions to the chemistry of the chlorides, simple and complex, of beryllium

PDF

A study of the double sulphates of some rare-earth elements with sodium, potassium, ammonium and thallium

PDF

A comparison of the electron donor properties of 1,3- and 1,4-dioxane

PDF

Competitive and sequential absorption and dissociation mechanisms in the infrared fragmentation of polyatomics

PDF

Connections of the dorsomedial hypothalamic nucleus in the rat

Asset Metadata

Creator Harrington, Eldred Ray (author)

Core Title Colloid properties of some New Mexico clays

School Department of Chemistry

Degree Doctor of Philosophy

Degree Program Chemistry

Degree Conferral Date 1940-04

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag agriculture, soil science,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Vivian, Robert E. (committee chair), [illegible] (committee member), Clements, Thomas (committee member), Copeland, C.S. (committee member), Vollrath, Richard E. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c17-19516

Unique identifier UC11347855

Identifier DP21727.pdf (filename),usctheses-c17-19516 (legacy record id)

Legacy Identifier DP21727.pdf

Dmrecord 19516

Document Type Dissertation

Rights Harrington, E. R.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA