<<

The Science of Stabilization HANS F. WINTERKORN, Director, Laboratory, Princeton University This paper defines the science of soil stabilization and views soil systems from many different angles in order to lay the proper foundation for an understanding of the dillerent encountered by the engineer, of the desirable and undesir­ able properties of these soils, and of the possible supplementation of these pro­ perties in order to change the soils into construction materials. This intro­ ductory paper is limited to the pointing out of important fundamentals and also to the presentation of certain approaches and data that are not normally found in soil-stabilization literature. e THE science of soil stabilization is that physico-chemical, or chemical terms; (2) body of principles that explains and guides tl'anslation of the supplementary require­ the practice of soil stabilization. Soil sta­ ments into available materials and pro­ bilization, in its widest meaning, com­ cesses and decision on use of specific prises every physical, physico-chemical, method (or choice of method) on the basis and chemical method employed to make a of economy, practical feasibility, or spe­ soil serve better its intended engineering cial (military or other emergency) con­ purpose. In its specific meaning, as com­ siderations; (3) construction, cons,isting monly understood in highway and airport normally of comminution, mixing with engineering, soil stabilization is the name stabilizing material, and densification; and given to those methods of construction in (4) economic considerations relating to which soils are treated to provide base the total cost composed of cost of mate - courses, and occasionally surface courses, rials, construction, and maintenance for which can carry the applied traffic loads the service life of the structure. under all normal conditions of moisture There exists a tendency on the part of and traffic for an economic service life of laboratory workers to overlook the im­ the paved area. The paved areas may be portance of Items (3) and (4). This may roads, airport aprons and runways, park­ lead to a dangerous self-deception andmay ing and loading places, feeding courts or even impede the proper development of the other surface structures of comparable science of soil stabilization. This science stability and durability requirements. is not a pure but an applied one, and the The major established uses of soil sta­ actual processes of application must not bilization are: (1) lifting a countryor re­ only be considered but must be analyzed gion out of the mud or out of the sand for scientifically in order that the most ra­ better economic development, now espe­ tional and most effective, i.e., the most cially important for under-developed economical, method of construction be areas; (2) providing bases and surfaces used in each particular case. In many for secondary and farm-to-market roads, cases, the use of a chemical construction where good primary roads are already in aid may increase only slightly the cost of existence; (3) providing bases for high­ stabilizing materials but decrease greatly type pavements where high-type rock and the cost of construction by facilitating the crushed gravel normally employed for mixing and compaction process. such bases are not economicallyavailable; (4) for city and suburban streets where the noise-absorbing and elastic properties of SOILS AND THEIR PROPERTIES certain stabilized soil systems possess The term "soil" covers a large variety definite advantages over other construction of materials existing under widely differ­ material:;;; and (5) for military and other ing conditions. For a thorough under­ emergencies where an area must be made standing of soils and their properties it is trafficable within a short period of time. well to look at them from different angles. Soil stabilization involves: (1) diagno­ A first step is to list definitions of the sis of the resistance properties of a given term soil as employed by different dis­ soil and required supplementation of these ciplines that deal intensively with the properties for the intended use in physical, material covered by this term. 1 2 . . ,

DEFINITIONS OF THE TERM "SOIL" With respect to scientific content and general usefulness, the pedologic soil con­ Highway Engineering cept is the most important. It has resulted Soil consists of disintegrated rock and in a natural system of organic matter found on the surface of the and soil mapping which, though originally earth, the particles of which may range in qualitative, can be easily supplemented diameter from less than O. 0001 inch to a with semi-quantitative and quantitative few inches and in which the fines are a engineering information. General pedol­ product of natural, weathering forces. ogical soil maps exist for practically every Soil may or may not contain organic part of the world. For many localities, matter. Engineering soils include bank soil maps are available on a scale as low gravel, bank sand, blow and dune sand, TABLE l agricultural soils ranging from those of Size Fractions of Mineral Porllon~ of Solle predominatly sandy texture to colloidal ...... lnlnches InMll11ru~tn clays, and mixtures and combinations of these (1). Gravel .. o. oe 76.2 to 2.0 3-ln ~ lo No. 10 o.oa'·' 0. 002 l.O io o.05 No. 10 kl NO. 270 The different size fractions of the min­ Slit·~· 0. 002 .. 0. 0002 O. fl to o. 005 eral portions of soils are named and de - Cb.y <0.0002 .(0.005 Col~d• <0. 00004 .(o , 001 fined in Table 1. Textural classification of soil is based as one inch to the mile. on gradation. The scheme of classifica­ Pedological soil types can be rec­ tion and naming is shown by Figure 1. ognized easily from air photographs (7) and certain conclusions with respect fo Geology soil stabilization can be drawn solelyfrom recognition of the pedological of Soil is the superficial unconsolidated a certain area. The principal soil areas mantle of disintegrated and decomposed of the world are shown in Figure 2. The rock material, which when acted upon by climatic and vegetational soil types of the organic agencies, and mixed with varying United States are shown in Figure 3. The amomtts of organic matter, may furnish general and specific application of pedol­ conditions necessary for the growth of ogy to engineering is treated in detail by plants. In its broadest sense the term Wooltorton in the third paper of this "soil" has been used to include all the symposium. mantle of rock decay (~). SOIL AS A POLYDISPERSE SYSTEM Soil is a polydisperse system composed Soil is the climatically conditioned of (1) solid inorganic and organic par­ petrologic and biogenic transformation ticles, (2) an aqueous phase carrying product of the outermost.layer of the solid matter in solution (and sometimes in dis­ earth crust (3); it is a natural body,

TABLE 2 simpler hy«rochemical compounds; (2) Mechanlcal Composition or the Principal Soil Classes complete or partial removal of .soluble re - action products or their COflcentration in specific .layers of the soil pr:ofUe; and (3)

Sandy SO- BO 0- 50 0-20 dislocation -Of coUotdal and clay-sized lO - 50 30 - 50 0-20 particles and their em\Centration in spe­ Silty loam 50 50-100 b - 20 Sandy clay loam so·-. BO 0- 30 2:0 . :rn cific layers of the son profile. Clay loam ,. - 50 20 - 50 :ro - 30 The most general ,chemical soil clas­ Silty clay loam .. 30 50 - BO 10 - 30 Sandy clay SO- 70 0- 15 30 - 45 sification wa.sma4.e b.y 'Marbut JV) by.

m TUNDRA Bmfil (WITH MUCH BOGI ~ GRAY-BR9WN PODZOUC SOILS R LATERITIC SOILS PRAIRIE sells a • DEGRADED ~ - ~~ftsNOZEMS a REDPl6H CHESTNUT

- CHESTNUT a BROWN SOILS [Iii SIEROZEMS a DESERT SOILS SOl~S OF TliE MOUNTAINS ~ a MOUNTAIN VALLEYS

.....

Figure 5 6 . . ' cipitation >evaporation, therefore, water for peat and bog soils. It is usually below percolation and elutriation) possessing a 20 percent, but a heavy clay soil may con­ relative concentration of iron and alu­ tain over 10 percent of water and still may minum compounds in one of the profile be dry and dusty to the eye and to the touch. horizons and (2) Pedocals, soils of semi-· is a mixture of arid and arid c limates (precipitation~ many different compounds, the more im­ evaporation) possessing a CaC03 con­ portant of which are carbohydrates, pro­ centration zone within or on the surface of tiens, fats and resins, and waxes. The the soil profile. absolute amounts and relative proportions The general shift from the chemical of these compounds vary with the climate composition of parent rock to that of the and with macrobiologic and microbiologic A and B horizons of soils and the difference activity in the soil. 'the organic matter is in chemical composition of the A and B concentrated in the surface layers and de­ horizons is shown in Table 3. This table creases with increasing depth in the soil. presents ana lytical data reported by Clarke It ranges from less than 1 percent in in­ (10) as oxides but herein tr ansposed into organic soils and sands to almost 100 per­ elementary composition. cent of the solid matter in peat bogs.

gim' $OIL$ OflHl Ir.I MOllTltERlt 1'11A llll£S

R111 tri1.f fl.nt. ~t( C't'lllKO Zn.1 llYl1lJI '!OILS • IOIL~ OPl'l1EP/IClflC Y/\LL('(S t§""1nte11N •AA'i Dfl>RT SOM j; ::·:·/sruTl

Figure 3 The elementary composition given in In the semiarid soils, the Table 3 refers only to the solid soil mineral organic matter may run up to 10 percent matter and does not contain the contribu­ of the weight of the dry soil. The average tion made by the variable soil moisture carbon content of humus is 56 per cent. and the soil organic matter. Soil moisture The average carbon-nitrogen ratio is 10 provides additional 0 and H atoms, soil to 1. A value of 10 percent of organic organic matter mainly C,N,0, H,S, and P matter, therefore, corresponds to 5. 6 per­ atoms. Soil moisture is usually expressed cent of carbon and 0. 56 percent of nitro­ as a weight percentage of the solid soil gen. Lateritic soils of the wet tropics matter dried at 110 C. in accordance with despite their often brilliant r ed and yellow the formula: inorganic appearance , have been found to Percentage moisture contain up to 2 percent and more of organic matter. = ~E'..t ~~i !! h.!-~i:,y _w~ ii;~tx 100 dry weight The total amount of carbon in a soil is the sum of that contributed by the organic Soil moisture may vary from less than 1 matter and that present in the profile as percent for dry sand to over 100 percent inorganiC carbonates, especially those of 7

Ca and Mg. Horizons enriched in Ca and chemical analysis of the clay, and it be­ Mg are typical of Pedocal soils. came a widespread practi ce to character ­ The words sand, silt, and clay have ize natural clays by this r atio. In been introduced and defined ear lier as soils this ratio falls below two; in other designating certain size fractions of the soils it is above two; it is two in the clay ultimate particulate components of soils. mineral kaolinite. Typical ratios (12) It is of interest to see what conclusions are given in Table 5. - with respect to soil chemical composition During the last 25 years it has become may be drawn from recognition or deter­ possible to extract and recognize mor e or mination of the granulometry of a soil. less pure crysta llized minerals from the clayfraction. This development is treated TABLE:! in an excellent manner by Hauser in the Avenge Elementary Composlllon of the Earth Cru1:1t and o( the A and B Horlzon11 or Soil1:1 fourth of this symposium. However, in Ea rth Soil £:Lemont orust A i1 01i.1011t DHor1~ view of the expected further development (percent) (perlC:UI) (per.cent)

Al 'l.13 !5,30 6.24 of nuclear meters that probe into atomic ¥• 4.38 2. 35 4 ~ 05 Ca 3.S2 .71 and nuclear compositions, it was con­ ... .2.35 ·" K 2.54 1·, "43 1.71·" sidered worthwhile to present in Table 6 Na 2. 6Z . Bl Mn ... .·.." ... the elementary chemical composition of Tl ,., 54 54 p , 13 ... the principal clay mineJ·a.ls, a:s calculated s . 13 .·os" ' ... 1 28.60 35. '10 35.70 ~ a l \IB.18 52 , 78 4. 9.73 from the oxide data given in (13). a) O]{fgen percentage does not Include contrlbutL.on Ii-om free water content. The data on chemical composition of soil constituents are summarized in Table Sand (2. 0 - O. 02 mm.) is predominatly quartzic and silicic in humid climates but 7. may be any kind of miner al in dry cli­ Characteristics of the liquid phase mates . The white sands of New Mexico The chemical formula for water isH20, a r e gypsum sands; coral and shell beach which represents a composition of 88. 8 sands may consist almost completely of percent (by weight} of Oxygen and 11. 2 per­ calciW11 carbonate· the black sand of Yellowstone Park and some of the blue cent Hydrogen. The H20 molecules, how­ and purple beaches of the Pacific Islands ever., because 0f their electri structure (Figure 4) ar e associated as well as dis­ are obsidianite sands as are some of the Alaskan and Aleutian beach sands (!.!_). sociated as a function of tempera ture. Quantitative values are given below:

TABLE 4 £temenlary Cofl:Jpositlon of Slit Components Concentration of H/and Temperature ( °C) P ercentages Elementary Compo!lillon 0 H-ions in pure water 0 Si Al N• Feldspars: ~KAlSbO. 41 . 5 33.0 10.4. 15 . 0 0 2. 8 x 10-9 Alb:lt.e Na.AlSL,Oe 4.4.3 J5.0 U.1 9. 5 3 ~ (OR,F)1KAl.(Sl,Al0111) H.O 21.0 20. 2 9.8 0 o. 2~ 4. T 8 7. 8 x 10- 7 ~ SIOi 53.2 4.6 . 8 • 25 1.0x10- 34 1. 45 x 10-7 7 Silt (0. 02 to 0. 002 mm.) particles re­ 50 2.3x10- semble quite closely the composition of . the parent rock with feldspar, muscovite, Association ofWaterMolecules as a Func­ and quartz usually well r01>resented. Re­ tion of Temperature. presentative formulas and compositional From Raman spectra (.!i) data are given in Table 4. We may con­ clude that in the silt fraction the silicon Temperatu.re ( °C) H20 content ranges between 21 percent and 47 percent and the oxygen content between 42 percent and 53 percent. Ice 0 41 59 The older soil literature contains many Water at 0 19 58 23 excellent data on chemical composition of Water at 38 29 50 21 the clay fraction (0. 002 mm.) of dif.ferent Water at 98 36 51 13 soils and soil categories. Many physical properties of clay systems could be re­ Although the molecular size of a com­ lated to ratio SiOz/A1203 + FeaOs found in pound is indeterminate in the solid state 8

Tammann (15) and Bridgman have shown the existenceof five different types of solid water that are functions of pressure and temperature conditions. Four of these types have corresponding liquid phases (see Figure 5). The temperatures and pressures in the triple points of the one:..component H20- system are given below:

Phases in Tempera­ Pressure Equilibrium ture ° C (kg/cm2)

Water-Ice I-Ice ID -22 2050 Ice II-Ice-I-Ice III -34.7 2170 Water-Ice V -Ice HI -17.0 3530 Water-Ice V -Ice VI - 0.16 6380

TRIANC:iULAR 5HAP£ OF THE H~O-MOLECULE The melting point of water decreases AS A RESULT OF THE POLARIZABILITY with increasing pressure to a minimum at 1 OF THE OXYC:iEN ION } -22 C. for 2, 050 kg. per sq. cm; sub­ 1. Potential Energy when H-ions and sequently it increases,· passing the 0 C. oxygen nucleus fallon a straight line. Re­ mark around 6, 300 kg. per sq. cm. and pulsion of positive parts: passing through 45 C. at about 12, 000 kg. per sq. cm.

~>--~~~~~~~~~~..--~~~~-.

r 2r 40 PHASE DIAGRAM FOR THE ONE COMPONENT 5YSTEM-H20 eo charge of one electron ( A(hr Tamm1111n ancl 6r-id9manJ Attraction between electron Shell and H­ ions: - 2 x 8eo2 r

4. 77 10-lO electrostatic units. ·2. Potential energy when H-ions are moved as indicated by arrows. -to 1Z'. Repulsion: 2 x 6e2 e2 + - r" r' ·JO ' \ I \ Attraction: Same as above. -40 ' lI \ '\ At small dislocations r" increases at \ greater rate than r' decreases. Equilib­ - ~o 4 6 a 10 ll rium = lowest potential energy if H+ posi­ PR£55URE IN Ko/cM' • 101 tions are at an angle between 100° and ll!f. Figure S l After Born, Heisenberg and Hund, Zsch. These data have been presented be­ f. Physik 23, 388; 26, 196 (1924') 31, 81; cause: (1) The peculiar behavior of water 32, 1 (1925f; Eucken, Arnold "Lerhbuch is due to the polarity of its molecule; that der Chemischen Physik," Leipzig, 929, is, its electric and geometric properties. (1930 . (2) The large adsorption forces exerted on water molecules by the surfaces of 9

TABLE 5 Relationship Between Silica-Sesquioxide Ratio and Base Exchange Capacity

Base Exchange capacity Clay ME/100 gm Source

Cecil 13 An Alabama clay loam 1. 3 Susquehanna 47 Well-oxidized Alabama soil 2.3 Putnam 65 Heavy Missouri silt-loam 3.2 Wabash 78 Missouri alluvial clay 3.2 Lufkin 82 Black Belt soil from Alabama 3.8 Montmorillonite 95 From Wyoming Bentonite 5.0

Nqte: The base exchange capacity represents the milliequivalents of cations adsorbed on the surface of 100 grams of colloid and exchangeable for other cations that may be introduced into the system. solid soU particles act similar to exter­ such as exist on the surfaces of the partic - nally applied pressures; that is, they may ulate matter in soils. liquify solid water or solidify liquid water. Water in soil is never pure but holds (3) Adsorption forces of the order of mag­ materials in solution and dispersion. Dis­ nitude that solidify water at 45 C. (12, 000 solved materials are mainly salts and kg. per sq. cm.) are not at all uncommon acids. ln saline soils the solutions may

10 0 0 FROM I I I ! I "FACTORS RELATED TO I v I / I I 0 9 o THE DESIGN OF 1 v MIXTURE JNSTABL~ IN ~ET ~· STABILIZED MIXTURES" WEATHER.-BITUJ,llNOUS OR ~ I CEMENT STABILIZATION MIXTURES SUITABLE FOR o BY F. R. OLMSTEAD (1943) I l!O 8 • DESIRA9LE ..._ 11 LIGHT TRAFFIC I '! "SAND-CLAY RANGE" / EASY _TO SHAPE AND I / COMPACT. P, I, '1-12 I/ I 7 0 I I/ v [,,. 0 1/ / J / 40 0 / ~ / v v 17 l/ ~ ,....,,..- , v / v /' . .. 0 HEAVY TRAFFIC "STAB. AGGV I./' 50"'.... WEARING COURSE~ , z /P.I. 4-9 / .. 0 v "' ' HARSH MIXTURES - 60 "' V v V v / DIFFICUU' TO COM- ... ~Iv V v [J PACT AND SHAPE ... HIGH CLAY CONTENT UNSTA91. v / \ DURING CONSTRUCTION JO1-- 70 IN WET WEATHER DUE TO , v v I/ EXCESSIVE VOLUME CHANOE , t_;:/ vv M 6\JPPERY WHEN WET. _, v v HEAVY TRAFFIC 20 / ,... "STAB, A

Figure 6 on the surfaces of soil particles. (4) All actually be saturated with different salts. physical properties (mechanical, elec- In humid climates the solute concentration trical, optical, acoustical) of cohesive soils is relatively low. The water layers next are functionally connected with the be- to the solid particles are under high ad­ havior of the water substance in strong sorption pressures, which may be larger pressure, electrical and magnetic fields than 25, 000 kg. per sq. cm. The water 10 . 1 maybe in a solidconditionat temperatures dition is called the plastic limit (17); that above 50 C. The adsorption forces de­ at which the soil passes into a liquid con­ crease exponentially to about 50 kg. per dition is called the liquid limit. These sq. cm. at the so-called hygroscopic mois­ limits, like the hygroscopic water content, ture content, and hence more slowly to increase with increasing clay content and zero for the water content at which the with increasing SiOz/R203 ratio of the clay soil-water system behaves essentially as and are, in addition, functions of the type a liquid (liquid limit). The course from and amount of exchange ions on the clay hygroscopic moisture content to the liquid and .of the ions in the aqueous phase. limit is illustrated by the following data The mechanical resistance of cohesive (13). soils increases with decreasing water con­ tent. Physical properties that increase Adsorption pressure with increasing water content are: per­ Condition in kg. per sq. cm. meability to water, dielectric constant, heat conductivity, and sound velocity. llygroscopicity 50 Properties that rise to a maximum and Permanent wilting point 12.5 subsequently fall with increasing water Wilting point (dead water) 6.25 content are: electric conductivity, e"lectro­ Vacuum moisture equivalent 0.55 osmoti.c water transmission (16), thermo­ osmotic water transmission "{r8, 19), and ease of compaction or density achieved The hygroscopicity of a soil increases with a certain compactive effort. with increasing clay content and with in­ cteasing SiOz/R:iOs ratio of the clay. At the interphase between the surface of the Properties of soil-air solid soil components and the water film The portion of the soil porosity not an electric potential is established, which filled with water represents the soil-air. gives rise to electrokinetic phenomena in Soil-air is in constant exchange with the t1oils. The magnitude of this potential and atmosphere and its composition reflects the thickness of the interphase or electric that of the atmosphere except for the con­ double layer is a function of the surface- centration of those components that are

TABLE 6 Percentage of Elements in Common Clay Minerals

Mineral Mica-like Element Kaolinite Halloysite Montmorillonite Beidellite Nontronite Minerals

Si 20:6-25. 3 18. 9-21. 7 22. 6-24. 2 21. 4-22. 3 14.7-22.5 23.7-24.5 Al 15.9-21.3 17.9-20.8 10. 6-14. 4 6. 4-14. 7 0.21-12 11.5-17.4 Fe o. 22-1. 46 0-0. 29 0. 15-1. 02 0.58-13.5 11.0-29. 8 0-4.6 Mg 0-0.6 0-18 1. 26-3. 96 0. 12-1. 8 0.06-2.4 1. 2-2. 7 Ca o. 021-1. 1 0.06-0.57 0.6-2.6 0.35-2.0 0.43-3.2 0-0. 42 K 0-1. 25 0.25 0.17-0.5 0.08 0.08-0.32 5. 1-5. 7 Na 0-0. 89 0. 074-1. 48 0.22-0.59 0.07-0.74 0-0.15 o. 07-0. 37 Ti 0-0. 86 0.48 0-0.06 0.3 Ha 1. 2-1. 6 1. 5-2. 6 1. 9-2. 63 1. 9-2. 5 0. 56-1. 44 0. 71-0. 78 oaav. 53.79 56.62 57.18 56. 06 50.45 50.22 H,O contents included in 11. 0-14. 3 13. 4-23. 7 17.1-23.7 17.3-22.6 5. 1-13. 0 6.4-7.0 H andO above

Note: a This includes the contribution from hydratation and other strongly absorbed -- water which does not make the clay wet. chemical composition of the solid, of the used up or produced by microbiological amount of water present, and the type and activity in the soil. Such substances are ~trtount of ions carried in solution (16). mainly oxygen, which is used up, and car- The water content at which a soil passes bon dioxide, which is produced. Atmos­ froni an essentially solid to a plastic con- pheric air contains 20. 939 .:!: 0. 004 volume " 11 garden soil ·· 1. 638 TABLE 7 Humus-rich, fel!tilized, Average Elementary Composition of the Different Size Fractions of Soils loamy sand soil 1. 876 Moor soil 6. 649 Percentage of dominant elements Dominant Garden soil, very rich in Elements in Clay in Silt in Sanda (average comp. humid humus and lime 6.880 climate) Fertilized with stable 0 50 - 57 41 - 53 41 - 53 manure 7.020 Si 15 - 25 21 - 47 21 - 47 Al 02 - 21 0 - 20 0 - 20 The oxygen content of soil-air decreases Fe 0 - 30 0 - 5 0 - 5 as a function of the increase of the C02 content, since the C02 is derived from the Note: a F or special sands (des ert, cor al, shell, a nd other Oxygen. More detailed data on soil-air -- beach sands) the chemical composition must be individually ascertain ed. can be found in the pertinent chapters of References 13 and 13a. percent of oxygen and 0. 031 .:!:. O. 0016 vol­ ume percent of C02. According to Eber­ It is often desirable to know the actual meyer (20) the proportion of C02 in the volume of solids in a given volume of nat­ soil atmosphere is a function of all the ural soil. Representative values for dif­ physical and chemical factors that condi­ ferent soils are given below: tion soil activity and productivity. Typical Volume of data found by him for the C02 content at Soil Pore Type space solid 70 cm. depth in volume percent of the soil particles atmosphere are given below: % % Humus-free, fine granular Moor soil 84. 0 16.0 quartz sand 0.340 Sand soil 39.4 60.6 Humus-free, coarse granular loam 45.1 54.9 quartz sand 0.340 Clay soil 52.7 47.3 .. Humus-free, fine granular • lime sand o. 548 Natural soils always possess air spaces Humus-free, heavy, clayey even if allowed to take in all the water they loam 0.726 can. Of course, after long-time flooding Humus-free, loess loam 0.745 this air space may be rather small. The Humus-and lime-containing air capacity necessary for the normal loam 1. 314 growth of different plants is quoted in (13) Humus-containing sandy loam (p. 281). TABLE 8 COLLAMERITICS The Science of Composition and Properties of Non-Metallic Construction Materials based on Comb!Jlat!on of Aggregates Properties of the aggregates Properties of Cementing Agents and Cementing Agents.

A Physlc:U A Inorganic A Design of I Mortar'!_ with In­ I Granulometry I Simple: Gypsum and organic or or ganic liiiiel>tasters. Laws of Arrangement and cements. Packing as a function of II Complex: Sorel and hy­ 1) Size and gradation II Concretes: Portland draulic cements . 2) Shape factor s Cement- 1 bituminous .. a) spherical B Organic resinous-, clay- , etc. b) cubic ID Plastics: Povider-, c) plate-like Bituminous: asphalts, fiber-filled. d) needles and fibers pitches, ~ d tars. II Mechanical l ) Strength and toughness II Natural and synthetic Resins, 2) Abrasion resistance elastomers and r elated sub­ stances. B P bys,lco-Chemtca.l and Cl1omlcnl ID Gums and Glues I Reactivity and Bonding with cementing materials

II Reactivity with deleterious substances 12

Plant Air Capacity Soil Dynamics % The periodic daily and seasonal warm­ Sweet grasses 6 - 10 ing and cooling of the earth surface result Wheat 10 - 15 in a temperature wave penetrating into the Oats 10 - 15 soil. The physical picture is as follows: Barley 15 - 20 (1) the normal daily and yearly tempera­ Sugar beets 15 - 20 ture fluctuations on the surface of the earth can be expressed as sinusoidal waves; (2) because of the heal. conductivity of the THE SOIL IN SITU earth, these surface waves are transmitted into the interior; however, the amplitudes Soils are creations of climatic forces (differences in temperature) are dampened and may be considered attuned to them. because of the heat capacity of the earth These forces derive from daily and sea­ substance; and (3) because of the time sonal temperature variations, from fluc­ necessary for heat transmission, the tuations in moisture content, from the maximum of the dampened temperature annual swell and sink of the biologic waves become increasingly retarded with potential, and from any other periodic increasing depth of penetration until at the phenomenon that affects the surface layer depth of one wave length, the retardation of the earth. As a result of these factors, equals the time pe:riod. For the same soils in situ are not mixtures of their reason, the time of highest temperature at components but are natural organized the soil surface is the time of lowest tern - systems, as was shown in the section on perature at a depth of one-half wave length. soil genesis. These rehlionships can be expressed These systems continue to be exposed mathematically as follows: The speed of to the forces that formed them and their propagation of the temperature wave is: properties are in a continuous state of flux. As a result of this situation, soils where in situ share many essential properties v with "living" systems and may almost be considered as living even if we disregard the large microflora and microfauna dis­ Since X persed in soils that render soils actual liv­ ing systems. Important consequences of this fact are discussed by Wooltorton in his excellent and comprehensive contribu­ tion to this symposium (paper No. 3 o'n Engineering Pedology and Soil Stabiliza­ tion). The speed of progression of the tem­ perature wave is therefore inversely pro­ Soil Aggregations portional to the square root of the period, while the wave length is directly propor­ The primary particles of a soil are very tional to the square root of the period. In rarely encountered as independent individ­ other words, a lorig period has a long wave ual constituents but are cemented together with a small speed of propagation, while a into secondary aggregations and crumbs short period has a short wave with a great by means of inorganic or organic binders speed of propagation. (21). The stronger secondary particles even persist in disturbed and molded spec­ Since the square rootof365 is about19, imens, which results in relatively large the length of the daily temperature wave is angles of friction of clay soils · in the only X.9 of that of the yearly wave. molded wet or in coherent air-dry condi­ The maximum amplitude ex at a dis­ tion. The aggregation increases the size tance x from the surface is the following of the soil pores which results in a greater function of the maximum amplitude e on permeability for water and air (in accord­ that surface. ance with Poiseuille' s law) than the soil would have in single-grain structure even at the same porosity value. 13

-2 'Ir If x X, 1 movement determines the climatic soil = then 0 X = Be 0 530 type encountered. -'Ir 1 The following potentials (22) are avail­ If x - A then e X 0 -2, = ee = 23 able for water movement in soils: (1) x gravity potential acting on the water mass; The damping effect is usually expressed (2) potential due to hydration energy of by the ratio of the amplitudes at a distance the ions, related to the heat of wetting, of one-half wavelength. The damping ratio which can be determined directly or cal­ is therefore 23, which means that an am­ culated from vapor absorption experi­ plitude of 23 deg. on the surface is reduced ments; (3) potential due to the osmotic to 1 deg. at a distance of one-half wave­ energy of ions held in a sort of Donnan length. equilibrium at the internal surface of the Employing average values in the cgs. system; (4) the so-called capillarypoten­ system for physical soil constants, name­ tial due to the surface tensionofwater; (5) ly: the thermo-osmotic potential due mainly to k 0. 004 the change in water affinity of the internal p 1. 6 soil surface with change in temperature c = 0.4 (19); and (6) vapor-pressure differences we obtain for the length of the yearly wave: causing diffusion in the vapor phase (This phenomenon is of little practical impor­ = 1, 560 cm. tance in temperate climates butappears to and for the length of the daily wave: be important in arid and monsoon cli­ mates). = 82. 4 cm. Although the order of magnitude of L.A. J. Quetelet (1796-1874) who made these potentials can be approximately cal­ the first reliable soil-temperature deter­ culated from theoretical considerations minations gave 17 m. as the length of the and with the help of experimental data, yearly wave and 1 m. as the length of the little is known concerning the resistance daily wave. He also noted that, because of to flow under these potentials. Except heat convection, the actual amplitudes at perhaps for the pure capillary flow under the lower depths were somewhat greater a pure capillary potential, flow under the than calculated. The magnitude of the am­ named potentials is primarily of a film and plitudes at the lower depths is illustrated not of a capillary character. Since prac­ by the following data by Quetelet: tically nothing is known concerning the structure and shear resistance of such Depth Observed Calculated films, much theoretical and experimental m. m. m. W?r~ needs to be done in this field (~). 0.19 13.28 13.28 0.37 11. 35 10.28 MECHANICAL RESISTANCE OF 1. 95 7.6 . 9 SOIL SYSTEMS 3.9 4.5 Assemblies of atoms, ions, and simple 7.8 1. 4 . 6 molecules may be in the solid, liquid, or For actual soil conditions, heat pro­ gaseous state depending upon the relative duced by microbial decomposition of soil magnitude of the mutual attractive forces organic matter must be taken into account and the kinetic dispersive forces. The in addition to heat convection as modifying latter increase with increasing tempera­ elements of the straight heat-conduction ture. At low temperatures the attractive concept. The predominance of heat con­ forces prevail and the material is in the duction in the soil-temperature picture is solid state; at intermediate temperatures sufficiently obvious, however, to make its the attractive and dispersive forces are basic theory an important tool in the hands more or less balanced resulting in the • of the highway and airport engineer. liquid state, and at high temperatures the dispersive forces prevail bringing the Moisture moves in and out of the soil profile and up or down in the profile as a system to the gaseous state. result of meteorological and soil physical Typical for pure compounds is their factors. As has been seen previously the definite melting point; the only difference dominant direction and pattern of mois'ture between the solid and the liquid at the 14 , . ' melting point is the lesser density or the terns is normally expressea as shear re­ larger volume per unit mass of the liquid. sistance S. For granular soils the shear This means that in the melting process, resistance is usually expressed by the the heat of fusion introduced into the sys­ equation: tem has performed the work of pushing apart the constituent atoms or molecules; S , in which in otherwords, holes have been introduced into the system. = coefficient of friction then be due essentially to the presence of The coefficient of friction is not a con­ these holes. This concept is strengthened stant for a certain material but is a func­ by the fact that for most atoms and simple tion of the voids ratio and also of

The most-striking example of the in­ bedded in the finer material. This bedding fluence of density or packing on resistance makes the mixture more workable and properties is the difference between the allows a modification of the properties of hardness of diamond and graphite. the entire system by modifying the bedding The direct analogy between atomic and material. The latter may be done by molecular assemblies, on one hand, and introducing air bubbles or particles pos­ macromeritic assemblies on the other, sessing special elastromeric, plastic, or holds not only for uniformly sized particu­ other desired properties. late components such as chemical elements Since the resistance properties of and compounds and corresponding uni.-. granular materials are dependent solely formly sized glass beads, beach sands, or on friction and since the frictional resist­ similar materials, but also for mixtures ance is directly proportional to the normal of m~terials of different sizes, and thus pressure on the shearing plane, the re­ for the graded materials so important in sistance properties are very low in the road construction. surface layers of granular masses where The principles of granulometry and the the normal pressure on the shear plane is extent to which they can be economically small. Therefore, such systems require applied to construction materials have cementation for ·their stabilization. The been studiedmost thoroughly for portland­ particles can be cemented together by any cement concrete, (26, 27, 28). From these one of the large number of inorganic and studies much can beiearned for analogous organic cements or by combinations of in­ soil systems which so far have been organic cements with organic cements or treated more or less on an empirical basis with waterproofing agents. If the granular (29). A most important new application of system contains both coarse and fine ag­ granulometric principles i.s described in gregates the cemented system is called a the paper by Pimentel dos Santos, which concrete. If the largest granules are of if the fifth paper of this symposium. sand size, then the cemented system will For particles of nonuniform and non­ be akin to a mortar. Such systems have spherical size the densest packing is de­ been called collameritic systems, (colla = terminedexperimentally. In these experi­ glue; and meros = particle) and include ments samples of the material are densi­ portland cement concrete, bituminous fied in a container of known volume either concrete, clay concrete (granular sta­ by rodding or by vibration. From the bilized soil), water-proofed clay concrete weight of the material required to fill the and sand-clay as well as water-proofed container and from its specific gravity, sand clay (see Table 8). Since the cement­ the absolute volume of the solid materials ing substance is the most active and often is calculated. Its subtraction from the also the most expensive component, it is total volume of the container gives the vol­ advisable to use as small a: proportion of ume of voids. it as possible. This means, that the In mixtures of materials of two size cementing substance should be used pri­ ranges such as gravel and sand the smaller marily for cementing and not for filling size components will interfere with the purposes. More economical fillers are close packing of those of larger si:ze. usually available. Consequently, the absolute volume of the The type of cementing substance which coarse aggregate contained in a unit vol­ can be used for the purpose of glueing ume of a mixture of coarse and fine aggre­ coarse aggregate and sand particles to­ gate is always smaller than the absolute gether depends primarily on the prevailing volume of the coarse aggregate that can climate. While under American condi­ be packed by itself into a certain space. If tions, main~Y portland cement, flyash­ the latter is called bo and if h is the vol­ lime combiilati.ons, and bituminous mate­ ume of the coarse aggregate in the densi­ rials have bee11 ell)Pi\oyed, in drier cli­ fied mixture of coarse and fine aggregate mates, gypsu.m plastl:irs (31) sorel- , and then the fraction b is always smaller than other w~~el' ~µ,~c~P,ti~le cements can be bo used su~ressfully. With proper water­ one. Its actual magnitude is a function of proofing U1e la,tt~::r cem,ents are suitable the size range of the coarse aggregate and for soil stabilization even in moist cli- of the fineness modulus of the sand (30). mates.· (32). · This means that the coarse aggregate is So far we bave dealt with systems com- 16 posed mainly of granular soils having less 3 5 percent keep the granular partides than 20 percent of silt-clay material. In from close contact and interlocking, with the range of granular cohesive materials marked lowering of the friction of the sys­ which contain up to 35 percent of silt and tem and with the loss of the skeleton clay, we may have, at the higher silt and effect. Such systems often possess great clay contents, well-graded material that cohesive strength in dry condition but lose corresponds to the established require­ the strength in the presence of moisture ments of granular stabilization. With which they absorb avidly under swelling. lesser contents in silts and clay, there 'is There is, therefore, a dividing line be­ a lack of natural binder which must be tween non-swelling and swelling soils made good by addition of binder soil or of which occurs at silt- clay contents of ab©ut cementing and waterproofing substances. 35 percent. In the non-swelling soils, It is of considerable interest that bind­ even at the highest moisture content, the ers composed of both silt-clay materials volume of the swelling constitue!!ts does and of organic or inorganic cements have not exceed the pore space left by the mu­ definite, ~nd not only economical, advan­ tually contacting larger grains. In the tages over ':.Ising the pure cementing mate­ swelling soi.ls the coarse particles are rial. The work of Silbergh (33) had indi­ isolated in a matrix of finer particles. The cated the existance of an optimum clay actual limiting proportion of swelling com - content (depending on the activity of the ponents depends upon the granulometry of clay) in the stabilization of sandy soil by the system and on the water affinity of the various inorganic and organic stabilizers~ silt-clay material.

TABLE 9 MECHANICAL RESISTANCE PROPERTIES OF TWO SOILS AND EIGHT OF THEIR HOMOIONIC MODIFICATIONS AS DERIVED FROM TENSILE AND COMPRESSION TESTS ON AIR-DRY SPECIMENS AND FROM CONSISTENCY MEASUREMENTS IN THE MOIST STATE

Strength . Internal Angle Angle "II Exchange Tensile Compress. Pressure 11 Plasticity (from P. Soil cations (psi) (psi) (psi) (deg) Index I.) (deg)

Cecil H 61 73 369 5. 17 34 20. 7 Na 52 86 132 14.23 30 21. 8 K 63 110 148 15. 72 38 19 . 7 Mg 70 102 76 10. 72 34 20. 7 Ca 51 125 86 24. 83 34 20. 7 Ba 75 121 198 13 . 53 36 20. 2 Al 72 147 141 20.20 31 21. 5 Fe 60 142 103 24.00 36 20. 2 Natural 65 124 137 18. 18 37 20.0

Hage rs- town H 105 133 38. 17 21 24. 0 Na 135 3 2 233 25.79 40 19. 2 K 88 4f3 1 119 36.15 25 23 . 0 Mg 110 575 136 42.85 27 20. 5 Ca 182 357 381 18. 97 23 23. 5 Ba 153 341 27 8 22.33 21 24.0 Al 93 368 124 36.65 16 25. 4 Fe 98 302 145 30.67 16 25 . 4 Natural 160 312 349 18.78 26 22. 8

Condition Air dried Moist to wet

The paper by Reinhold in this bulletin THE MECHANICAL RESISTANCE PROP­ shows for stabilization of a sandy soil with ERTIES OF SILT-CLAY MATERIALS portland cement an optimum clay content The shear resistance of cohesive soils of about 25 percent, an excellent construc­ is usually expressed by the equation: tion material being obtained with 25 per­ cent of clay and 10 percent of portland S = 1Ttan cp + C cement. where S -= shear resistance, IT normal stress on the shear plane, Silt-Clay Materials tan cp coefficient of internal friction Silt-clay constituents iri excess of about and 17

C = Cohesion formula was assimed to hold, than they do for the solid state for which the values had Actually, friction and cohesion in such been obtained from compressive and ten­ systems are not independent bf each other sion data by use of the Mohr circle method. but are both functions of the density, the Soils can be considered as (1) solids moisture content, and of the manner in if their moisture content is below the which the particular moisture content' has shrinkage limit and if they have been dried been attained. The friction portion of the to this point from higher moisture con - shear resistance may be expected to follow tents; as (2) being in the plastic state if the same type of volume law that has been their moisture content falls between the previously described for the internal fric­ plastic and liquid limit; and (3) as liquids tion of liquids and of granular materials. if their moisture content is above the liquid The first step in understanding soil cohe­ limit. sion is to consider it as a result of the The location of these limits depends on attraction forces acting from the mineral the clay and colloid content of the soils, on surfaces through the water fihns and of the type and the activity of tlie clay frac­ geometric factors that have the functions tion, and on the granulometry of the entirll and therefore the properties of an angle of soil including its clay fraction. The friction. Accordingly, one may write: greater the tendency to structure forma­ .Cohesion = internal pressure x tan . tion inherent in the clay the higher is its shrinkage limit. Also, the presence of For soils at very low moisture content ions furthering structure formation in­ this internal pressure is theoretically of creases the shrinkage limit, while the the order of magnitude of 106 pounds per presence of dispersing ions decreases the square inch. It falls logarithmically to shrinkage limit. zero at a moisture content at which the The plastic state may be considered to soil flows under its own weight, i. e. ' at begin as soon as the active surfaces. in the which the soil behaves as a liquid (liquid limit). Because of secondary aggregation soil system are covered with films of water that are sufficiently thick to be con­ and non ideal arrangements of the cohesion­ gtving water films, the actual strength of tinuous and to have lubricating properties. Therefore, other factors being equal, the cohesive soils is much smaller than the larger the amount of internal surface, theoretical. It has' been found (34) that even monocrystals of pure minerals such i.e. , the larger the clay content the higher as sodium chloride give tensile strength the plastic limit. values that are two to three orders of However, dispersed clays may require magnitude smaller than those calculated thinner moisture film for lubrication than from theory. It is, therefore, not sur­ aggregated clay particles. For this reason prising that experimentally determined the plastic limit sometimes falls with in­ strength values for carefully dried clay creasing dispersion of the clay fraction. soil specimens and the related internal The liquid limit represents theoretically a pressures are s-ystematically lower than moisture volume which permits the in­ 1,,000 psi. and usually lie between 100 and dependently acting primary or secondary 400 psi. soil particies to rotate freely. In fact, the In Table 9 actual data are presented for possibility of free rotation of the constit­ compressive and tensile strength of spec - uent particles is considered as the most imens of nine ionic modifications of two important index of liquidity. Therefore, different clay soils. The specimens were the larger the proportion of plate-shaped molded in the plastic range and carefully particles and the larger the ratio of the j air dried (35). From the experimental volume of the rotation ellipsoids to the data, anglesof internal pressures were actual volume of the particles, the higher derived using the Mohr circle method. will be the liquid limits. l The Table also contains angles of friction An increase in the dispersion of a clay I derived for the same soils from the plas­ may, on one hand, decrease the plastic ticity index according to Ktlgler and limit and, 0n the other hand, increase the I Scheidig (36). It is of interest that the liquid limit. The presence of silt and sand angles of friction derived for the different fractions may modify these corislusions be­ ionic modifications of the same soil vary cause of their interference and interaction less for the plastic state for which Ktlgler' s with the clay particles. It has been found 18 . ' J quite often that increasing amounts of clay which are then cemented together by were inactivated in part -by increasing means of inorganic binder!> such as Port­ amounts of silt particles on the surface of land cement, flyash-lime combinations, which they were strongly adsorbed (37). gypsum plasters, sorel cements, etc. Finally, there is a great influence of A less-positive but still effective meth­ the granulometry. The latter is treated od, if properly used, is to stabilize the by Pimentel dos Santos. The scientific moisture content of a soil system by pre­ study of the resistance properties of silt­ venting the intake of excess water. This c lay-water systems falls within the realm water-proofing may be achieved by various of the science of rheology, and there exists materials such as asphaltic and pyro­ a great need for more fundamental studies genous bitumens, natural and artificial following the concepts of Eyring (38), resins, fats, waxes, etc. Mack (39), Ruiz (40), Nijboer (41), and The scientific factors involved in water­ others '142). - - proofing of cohesive soils have been stud­ ied in considerable detail (46, 47). sin-ce it is usually desirable thatthe stabilized Stabilization of Silt-Clay Materials system contain a certain minimum volume The crucial difficulty involved in the of stabilizer, combinations of highly active stabilization of cohesive soils that do not materials with less active ones are often contain granular skeletons is the fact that indicated as well as combinations of in­ the seat of the resistance providing co­ organic and organic stabilizers. hesion is also the seat of the water affinity. Finally, systems may be synthesized by The cohesion increases with decrease in low-temperature polymerization in which. water content. Theoretically, the simplest the actual soil constituents are held as method of stabilizing such materials is to particulate inclusions in a fibrous network (1) deprive the component particles of or felt. The fibers may or may not be their water affinity and (2) to cement attached to the clay particles by ionic, these particles together. resonance or other forces. However, this This exactly is achieved by the oldest method requires a rather large amount of method of soil stabilization, namely the water which brings the system to the sticky burning of clay soils for the making of point and ofter makes mixing extremely pathways, which has been used extensively difficult. Also the excess water is likely by the Australian aborigines (43). How­ to soften the subgrade and rob it of its ever, the practical difficulties involved in supporting power. In addition, a rather this method are obvious; despite much large amount of expensive stabilizer (us­ recent experimentation with large road ually above 10 percent by weight) is re­ burners the most-promising use of this quired. Nevertheless, this is an interest­ method is the manufacture of coarse ce­ ing development and has certain uses in ramic aggregate and its subsequent use in fields other than road construction. granular soil stabilization. The scientific principles involved in the The economics of this problem have stabilization of cohesive soils, including been well explored in Argentina where the the microbial problems, have been treated process was concluded to be feasible under quite thoroughly in the literature (48) certain conditions of absence of natural and need not be discussed here in greater granular materials. The relationship be­ detail. Jn all these methods, except per­ tween pedologic soil type and effectiveness haps the thermal one, it is important that, of thermal stabilization has been studied at the time of treatment, the soil has suffi:. especially in Russia (44). cient moisture to satisfy its hydration re­ Another use of this principle is to re­ quirements and that the particles are sur­ place the water attractive inorganic ex­ rounded by continuous moisture films with­ changeable cations by means of water re­ out excess or free water, since the latter pellent organic cations arid to cement the may lead to undesirable stickiness of the soil constituents, that have now lost their system. Table 10 shows the relationship water affinity, together by means of an between i:;oil stabilization and its support­ organic cement. This can be done using ing sciences. Figure 6 is a reproduction aniline-furfural and similar resins (45). of a chart published in 1943 by Olmstead Another method is to create strongly (49) in a paper entitled "Factors Related water resistant secondary soil aggregates tOthe Design of Stabilized Mixtures." The 19 chart exposes most clearly the connection shortened or even avoided by the use of between and stabilization re­ "shortening agents." Among the latter quirements. aniline-furfural combinations have proved themselves as most versatile and most ef­ SCIENTIFIC FACTORS INVOLVED IN fective. While some of our modern com­ THE CONSTRUCTION OF minution machinery has been developed STABILIZED COURSES with sufficient power to breakup even dry, hard-caked soils to sizes sufficient for The science of soil stabilization covers subsequent stabilizing· treatment, it is not only compositional, structural and re­ better science and better practice to bring sistance properties of stabilized systems the soil first to that degree of moistness at but also the problems involved in their which it is most easilycomminuted to par­ manufacture in the field. Construction- ticles of the proper size. wise, soil stabilization involves the three Energy of Comminution unit processes: (1) comminution, (2) mix­ ing, and (3) compaction. At the present time, the energy of com­ TABLE 10 The Science of Soil Stabilization rests on the three pillars of

Body of Knowledge on Tradi­ Physical and Chemical Sciences Pedology tional Construction Materials

The science of phys­ These allow judgement of soil de­ A stabilized soil, being a con­ ical and chemical ficiencies and indicate supple­ struction material, must re­ soil characteristics mentation in physical and chem~ semble traditional construc­ as functions of: ical terms that are translated into tion materials in Us essen­ economically available materials, tial resistance properties 1) parent material such as and structure. Especially 2) climate important are granulometric 1) inorganic cements 3) topography considerations. Most non­ 2) organic cements 4) organisms metallic construction· mate- 3) waterproofing agents with or 5) time rials and most s i-zed sl.'lil without improvement of gran­ s~s eins M ategocy ulometry of the soil. .to of cqtlamer.i~.,.)'J~\

In the case of purely granular soils the n-1.a;te~ etaeuY-f.rom comminution factor is, of course, absent. wledge, yen if The farmer who has a similar problem oL 0 y a:pproxim t~ o . thlS energy 1 -of comminution when he prepares the proper -grill~ 6Go : m.ic a~ w~ al3 sctentific m­ seed bed for his crop has learned that (1) p·ortanae. Ff?r this .ea a an.approachto for cohesive soils the plow is a most ef­ th-e scientiflc sol\) . o o s preb1em ts fective instrument and that (2) there ex­ gt.van ln the f9U0Witng. U)'.nount of ists a moisture range in which comminu­ energy X>equired for comminiltlon depends ~ tion to the desired particle size is most among other factors, upon the ultimate easy to achieve, The latter is within the particle size required and upon the extent plastic range but below the point where the to which granules of this stze are already water is already sufficiently free to stick preformed in the natural soil .. _Acc0rd­ to the plow metal (sticky point). ingly, it should require relatively' little In regions where it is normally dry dur­ energy to comminute soils to granules of ing the construction season it is easy to gravel and sand sizes but very consid­ reach the most advantageous moisture con­ erably more to break a clay soil down to tent by addition of water. A real problem the ultimate clay particle units. exists when soils have natural moisture There exist three general approaches contents above the sticky point; this may for the calculation of comminution energy. result in very costly delays of weeks and In each approach certain assumptions must even months. These delays can often be be made because of lack of pertinent sci- 20 r entific data. (1) The work performed in From these data, the maximum strain creating two unit surfaces equals one half energy that can be stored in pure tension of tne product of tensile strength ana the can be calculated approximately as the distance over which the tensile force must product of half of the failure stress and the be moved. (2) The work of comminution failure strain. Accordingly, we obtain: equa1s the maximum strain energy that can be stored by the body under consid,eration Elasllc Energy Stored per Unit Volume up t.o FaHUre

cmkg/cm3 rnkg/crn' mkg/ln3 m'kg/ft1 HP-ltr under the condition of strain application. Soll Jqq:toftt' ~ u~

Comminution is into "natural" structural lJaalooll• 0.. 003 57 35.7 x to·' 58. 5 :i:l0- 5 101 x 10·1 9.9 3.89 x io-9 5 UI).its. (3) The work of comminution equals XAollaU.. 0,00009 O.IJ x 10-$ 1.4.7 x 10- 2.~xlO-" 0. 25 o. 093 x 10-• Hille 0, 0016 16.0 x 10-• 26, 2 x 10 -~ 45.3 x 10 · " 4. 45 1, 64 x 10-• 1 the surface energy of the new surfaces Clay"Pr'' 0, 00032 3. z l( 10-• :;. 24 x 10-' 9.os x 10· 0.89 O.J3 x 10-• created. The required assumptions are for (1) Size of natural breakage units the effective action radius of the cohesive Assuming that the entire stored strain forces, for (2) the modulus of elasticity, energy is used for the creation of new sur­ and for (3) the surface energy. faces without actual breaking down of pri­ We shall employ Approach 2 and utilize mary soil particles but having the new Approach 3 to calculate the dimensions of surface going through the film water be­ i:iasily prod,uced natural soil units. tween the particles, and assuming that ad­ Using an average tensile strength value sorption forces on the surfaces of the solid of 200 psi. for a dry clay soil (strongest particles raise the surface energy of water state) and a proQable modulus of elasticity to about 150 ergs per sq. cm., we obtain value of 92, t\00 psi. (from German dynamic a surface energy of 6. 45 x 150 ergs = 968 measurements), wemaycalculate a failure ergs per sq. in. Dividing 39x107 ergs per strain of: sq. cu. ft. by 968 ergs per sq. in. gives 2 • = ~g = O. 00216 inch per inch. 5 92 , 0 4. 03 x 10 sq. in. per cu. ft. Multiplying this strain by half of the The surface in sq. in. per cu. ft. is 6x144 failure stress, we obtain: = 864 sq. in. for a cube of 1 foot side O. 00216 x 100 ,. O. 216 in. -lb. per cu. in. length and 6 x 1728 = 10, 368 sq. in. if sub­ divided into cubes of 1 inch side length. = O. 018 ft. -lb. per cu. in. The general formula = o. 018 x 1. 356 x 107 ergs per cu. in. . . 10, 368 = 224 x 103 ergs per cu. in. Surface m sq. m. = side length in inches = 39 joules per cu. ft. For a surface of 4. 03 x 105 sq. = 1. 45 x 10-0 horsepower­ ft. we obtain a side length of: hours per cu. ft. 10.368 0. 026 inch. The last value represents the maximum 4. 03 x 105 strain energy that can be stored per cubic = 0. 066 cm. foot of tb.e soil if employed as a baz: in :: 0. 66 mm. tension. If employed in pure bending, the value falls to a third of that given, and if This value indicates that the resulting used as a cantilever, to a ninth. particles are of the order of magnitude of Similar .calculations may be Jllade for a medium sand. This reasonable result soils in the plastic range using data pub­ indicates that the extrapolated assumptions lished by Professor Tschebotarioff (50). made in its calculation are also reason­ These data are given in the next table. - able. This fact is important not only for estimating power requirements in soil P\a.stic Liquid H,O Tensile Strain Soils Limit I~im it Content Strength at failure comminution but also with respect to soil % % % Kg/ cm' % response to dynamic loading by blasting and forced mechanical and sonic vibrations. Bentonie 53 540 101 o. 21 3. 4 Kaolinite 32 70 37. 6 0. OQ o. 2 The calculations just presented are, of II lite 26 60 31. 5 0.40 0. 0 course, only a first approach. However, Clay "Pr" 23 30 tu. R 0. 08 0. 0 this approach should be followed up and should finally lead to a more rational de- 21

sign of comminution equipment than is investigated on the basis of fundamental possible at the present time. theoretical considerations.

Mixing Step Economic Considerations The problem of the energy and power Soil stabilization represents an en­ requirements for adequate mixing of sta - gineering science and as such must include bilizers with the soil material in its nor­ economics. Economic considerations are mal moist condition has as yet not been not restricted to the cost of materials properly studied. While some extrapola­ which represents only a portion of the total tion can be made from the experience in cost but also must cover the costs of con­

mixing probl!=!ms of the chemical industry1 struction and of maintenance during the the special properties of the soil mi:xtures service life of the structure. It would be involved in stabilization indicate the need impossible to make here a complete anal­ for specific studies with such materials. ysis of all the important economic factors, but it can be pointed out that often a slight Densification Step increase in cost of materials caused by the use of "construction aids" can'be balanced Because of the very intimate relation­ by a lesser construction cost due to easier ship between the density of a stabilized soil mixing and .:ompaction of the treated soil. system and its resistance properties, If such "construction aids" also improve compaction is probably the most important the final quality of the system, so much the single feature in soil stabilization. In better. studies performed for the Civil Aero­ The costs of traditional cementing and nautics Administration it has been shown water proofing materials are relatively that the difference between normal and stable within certain regions, and it is not modified Proctor compaction in certain too difficult to decide which of several has stabilized soils was equivalent to the effect economic preference in a particulat loca­ of approximately 2 percent of portland tion. • cement. This means that systems sta­ bilized at the higher density with 2 percent However, the relative costs vary for less portland cementwere equal in resist­ different areas. In one region portland ance to those stabilized at the more of cement may be more economical than bitu­ Proctor density with 2 percent more of men as is the case in Portuguese East portland cement. Africa, while in other regions, bitumen However, for this to be the case, the may be more economical than portland higher density had to be reached with a cement. The same holds true with respect moisture content sufficient to satisfy the to lime, flyashes, and other materials. hydration needs of both the soil and the The economics are not as obvious in portland cement. the case of chemicals that have not been In view of the large amount of compac - used previously for soil stabilization pur­ tion tests that have been performed in soil poses since the unit price of chemicals laboratories under well standardized con­ usually decreases with increasing indus­ ditions, it would seem that enough know­ trial production. Therefore, it would be ledge is available for judging the work and unreasonable to base economic conclusions power requirements in compaction. Ac - solely on present prices of certain chemi­ tually the wall effect of the small molds cals. However, it is very much to the that are used in the laboratory has never purpose to point out certain fundamental been definitely evaluated, and rational ex­ economic facts. trapolation from laboratory tests to field 1t is obviously an economic waste to conditions has as yet not been accom­ employ an expensive cementing material plished. not only as a cement but also as a filler. In addition, the moldtng compaction ac­ As it is an economic waste to have one tually used in the field differs often quite man build an entire automobile by him­ essentially from the compaction by means self, the same way it is uneconomical to of the impact of the falling hammer have one chemical alone remedy all the employed in the laboratory test. These various deficiencies of a soil. It is us­ and other compaction problems should be ually more economical to employ several 22

mono- or bi-functional substances instead are important for scientific soil stabiliza­ of one alone that is polyfunctiortal. tion, to indicate how these properties must Even in the case of beach sand sta­ be supplemented or modified in order to bilization for military purposes, where produce construction materials, and to normal economics are· out of the picture, point out material and economic facts of it is certainly advisable not to use the importance in actual construction. more expensive resinous substance to do If we accept Poincare' s definition that the entire job of cementing the sand grains the essence of a science is to give the together, but to dilute the synthetic resin same name to different things, then it is by means of pitch, asphalt, or other evident that this condition is fulfilled in organic or even inorganic cementing mate­ the case of scientific soil stabilization. rials which can be easily employed in the The methods of stabilizing granular soils process (11). have been included in the collameritic Also, ilcertain elastomeric properties system which embraces a large number of are desired, these can be obtained by in­ traditional construction materials as well corporation of elastomeric powders or dis - as an important group of stabilized soils. persions. This possibility· resides in the At the same time it has been demonstrated fact that the large bearing systems created how physico-chemic.al, chemicaland phys­ in soil stabilization are quasi- or prac­ ical principles are important in the proper tically isotropic. stabilization of soils especially of those There is, of course, no such thing as a that do not possess a granular skeleton. completely and truly isotropic system in Also, basic principles with reference to nature since even the atoms and the nuclei the economy of producing stabilized com­ are nonisotropic. Practical isotropicity positions have been pointed out. depends upon the size of the largest com­ ponent of the system with respect to the This being an introductory paper, it was size of the area of stress ;ipplication. If intended to indicate but not to anticipate the latter is much larger than the coarsest the information which is presented in the components of the resisting material then subsequent papers on more specific sub­ we have practical isotropicity. jects. This permitted concentration and In portland-cement concrete we allow, emphasis on the principles involved. At for instance, sizes of the large aggregate the same time certain basic considerations up fo a third of the smallest dimension of and data relative to the complex system the structural system. "Soil" could be presented which are not From the. fol'egoing considerations it usually found in engineering literature. can be seen easily that it is a sound economical principle to employ several lower cost mono- or bi-functional compo­ ACKNOWLEDGE.MENTS nents to achieve certain combined proper­ GI'.ateful acknowl~qgem.ents are made to ties than to employ an eJCI).ensive polymer Adm! al w. Ma.ck Ang-as,, cha:i.J'man Of the in which several functions have been em­ De}?a»tment of Civil Engrineer-illg of Prince­ bodied. ton Un'iversity for hif! great interest in the science of soil stabilization and for effec - CONCLUSIONS tively aiding our work. Further acknow­ ledgements are due to Mrs. M. Hill who It has been intended in this introductory typed the manuscript and to Ivo Kurg who paper to expose those soil properties that did the drawings.

References 1. Soil Bituminous Roads; Cur rent Road Formation. Schweiz, Landw. Monatshefte Problems, Bulletin No. 12, Highway Re- 20, No. 6, 7, p. 24 (1942). search Board, Washington 25, D. C. (1946). 4. Joffe, J. S.: Pedology. Rutgers Uni- 2. Ries, H. and Watson, T. L.: En- versityPress. 1st ed. 1936, 2nd ed.(1949). gineering Geology, 5thEdition. John Wiley 5. Stebut, Aleksandr: Lehrbuch der and Sons (1936). See also: Legget, Robert Allgemeinen Bodenkunde, der Boden als F.: the term "Soil", Nature, Vol. 171, Dynamisches System. GebrUder Born- p. 574 (1953). traeger, Berlin (1930). 3. Pallmann, H.: Fundamentals of Soil 6. Laatsch, Willi: Dynamik der De~t- ,· 23

schen Acker-und Waldboeden. Dresden Moisture Migration. Proc. Building Re­ und Leipzig (1944). search Conference, National Research 7. Jenkins, D. S., Belcher, D.J., Council, Canada, Ottawa (1953). Gregg, L. E., and Woods, K. B.: The 20. Ebermayer, E.: Untersuchungen Origin, Distribution, and Airphoto Identi­ Uber die Bedeutung des Humus als Boden­ fication of United States Soils, Tech. Dev. bestandtheil und Uber den Einfluss des Report No. 52, CAA, Washington, D. C. Waldes, verschieder Bodenarten und (1946). Bodendecken auf die Zusammensetzung 8. Davis, R.O.E., andBennett,H.H.: der Bodenluft. Forschungen auf dem Ge­ Grouping of Soils on the Basis of Mechani­ biete der Agrikultur Physik, Vol. 13, pp. cal Analysis. U.S. Dept. of Agric. Dept. 15-45 (1890). Circular 419, Washington, D. C. (1947). 21. Winterkorn, Hans F. and Tsche­ 9. Marbut, C. F.: The Contribution of botarioff, Gregory P.: Sensitivity of Clay Soil Surveys to . Proc. Soc. to Remolding and its Possible Causes. for Prom. Agr. Sci. Vol. 41, pp. 116-142 Proc. Highway Res. Board., Vol. 27, pp. (1920). 435-443 (1947). • 10. Clarke, F. W.: The Data of Geo­ 22. Winterkorn, Hans F. and Eyring, chemistry. U. S. Geol. Survey Bull. 330 Henry: Theoretical Aspects of Water Ac­ (1908). cumulation in Cohesive Subgrade Soils. 11. Winterkorn, Hans F.: Desarrollo Proc. Highway Res. Board., Vol. 25, pp. del Metodo Anilina - Furfural para la 422-443 (1945). Estabilisacion de Playas. Bulletin No. 2, 23. Winterkorn, Hans F.: Water Move­ Winterkorn Road Research Institute , ment through Porous Hydrophilic Systems Princeton, N.J.; also in Proc. 6th Annual under Capillary, Electric and Thermal Asphalt Congress, Cordoba, Argentina Potentials A. S. T. M. Symposium on Per­ (1952). meability of Soils, June (1954). 12. Winterkorn. H;\.nS F. and Baver, 24. Batschinski, A. J.: Journ. Phys. L. D.: Sorption of Liquids by Soil Colloids: Chemie, 84, p. 643 (1913). I. Liquid Intake and Swelling by Soil Col­ 25. Winterkorn, Hans F.: Micromeritic • loidal Materials. Soil Science Vol.38, pp. Liquids. Symp. on Dynamic Testing of 291-298 (1934). Soils, A. S. T. M. Special Publication No. 13. Blanck, F. (Ed): Handbuch der 156, pp. 77-89 (1953). • Bodenlehre, Vol. 6, Julius Springer, Berlin 26. Feret, R.: Summary of Papers (1930). Also: First Supplementary Vol. published since 1892 in "Allllales des Ponts (1938). et Chauss~es" and other journals. Revue 14. Rae, I. Ramakrishna: Constitution de l'Ingenieur, Paris, France (1923). of Water in Different States, Nature, Vol. 27. Fuller, W. B. and Thompson, S. E.: 132, No. 3334, p. 480 (1933). The Laws of Proportioning Concrete. 15. Tammann, G.: The States of Aggre­ Trans. Am. Soc. Civil Eng. 59, pp. 67- gation, (Translated by Robert F. Mehl) 143, Discussion 144-72 (1907).~ D. Van Nostrand Co. (1925). 28. Rothfuchs, G.: Particle Size Dis­ 16. Winterkorn Hans F.: Surface­ tribution of Concrete Aggregate to obtain Chemical Properties of Clay Minerals and Greatest Density. Zement, 24 (1), pp. 8- Soils from Theoretical and Experimental 12 (1935). Also: How are aspfiil.t and bitu­ Development in Electroosmosis. Sympo­ minous mixes of maximum density to be sium on Exchange Phenomena in Soils, obtained? Bitumen, 5 (3) pp. 57-61 (1935). 1952. A. S. T. M. Spec. Puhl. No. 142 29. Granular Stabilized Roads: Wartime (1953). Road Problems Bulletin No. 5, Highway Re­ 17. A. S. T. M. Committee D-18 on Soils search Board, Washington, D. C. (1943). for Engineering Purposes: Procedures for 30. Goldbeck, A. T. and Gray, J , E.: Testing Soils. A. S. T. M. Philadelphia A Method of Proportioning Concrete for (1950). Strength, Workability and Durability. Bul­ 18. Winterkorn, HansF.: Fundamental letin No. 11 (revised November 1953) Similarities between Electrobsmotic and National Crushed Stone Association, Wash­ Thermoosmotic Phenomena. Proc. High­ ington, D. C. way Res. Board, Vol. 27, pp. 443-455 31. Forbes, R. J.: Notes on the History (1947). of Ancient Roads and their Construction. 19. Winterkorn, Hans F.; Potentials in Amsterdam, (1934). 24

32. Winterkorn, Hans F. : Final Reports 167, 197, London (1951). to Bureau of Yards and Docks, (1949; 43. Eckert, George W.: Soil Stabiliza­ 1950; 1951; 1952). tion by Heat Treatment. Journ. Soc. 33. Silbergh, Michael: Saline Soils Chem. Ind. Vol. 58, pp. 846-851 (1939). and their Stabilization. Master Thesis. 44. Filatoff, M. M.: Stabilization of Princeton University (1949). Road Soils by Heating and by Means of As - 34. Schmid, Erich and Boas, W. :Plas­ phaltic and Pyrogenous Bitumens, and ticity of Crystals with Special Reference other Substances. Soil Stabilization, A to Metals. F. A. Hughes & Co.(transl. collection of Papers published by the from German 1935 ed.) London, (1950). Central Road Administration, Moscow, 35. Winterkorn, Hans F.: Surface­ u.s.s.R. (1938). Chemical Factors Influencing the Engineer­ 45. Winterkorn, Hans F.: Job Expe­ ing Properties of Soils. Proc. 16th Ann. rience with Exchange Phenomena Involving Meet. Highway Research Board, pp. 293- Inorganic and Organic Ions. Symposium 302 (1936). on Exchange Phenomena in Soils. Spec. 36. Koegler, Franz and Scheidig, Techn. Publ. No. 142, A.S.T.M. pp. 29- Alfred: Baugrund und Bauwerk, p. 60, 42 (1952). Wilhelm Ernst and Son, Berlin (1938). 46. Winterkorn, Hans F., and Eckert. 37. Winterkorn, Hans F. and Eckert, George W.: Physico-Chemical Factors in George W.: Consistency and Physico­ Bituminous Soil Stabilization. Proc. Chemical Data of a Loess Pampeano Soil. Assoc. Asphalt Pav. Techn. Vol. 11, pp. I. Soil Science Vol. 49, pp. 73-82 (1940); 204-257 (February 1940). II. Soil Science, Vol. 49, pp. 479-488, 47. Winterkorn, Hans F.: Chandrasek­ (1940). haran, E. C., and Fehrman, R. G.: Ciertos 38. Glasstone, S., Laidler, K. J., and Aspectos de la Estabilizacion de Suelos con Eyring, Henry: The Theory of Rate Proc - Resinas. Proc. 5th Asphalt Congress, pp. esses. McGraw-Hill Book Co. , New York 25-65, Buenos Aires, Argentina (1950). . (1941). 48. Winterkorn, Hans F. : Fundamental 39. Mack, Charles: Series of Articles Approach fo the Stabilization of Cohesive in Journal of Applied Physics. soils. Proc. 28th Annual Meeting. High­ 40. Ruiz, C. L.: "Theory of the Struc­ way Research Board, pp. 415-422 (1948). ture and Deformation of Asphaltic Mix­ 49. Olmstead, F. F.: Factors re­ tures". First Argentine Asphalt Congress, lated to the Design of Stabilized Mixtures Proc. pp. 365-408, (1946). Journal of Asphalt Technology (August 41. Nijboer, L. W.: Plasticity as a 1943). Factor in the Design of Dense Bituminous 50. Tschebotarioff. G. P.: The Tensile Road Carpets. Elsevier Publishing Co. Strength of Disturbed and Recompacted (1948). Soils. Proc. 3rd Int. Conf. Soil Mech. 42. Rigden, P.J.: Rheology of Suspen­ and F. Engineering, Vol. I. Session 2/28, sions of High Solid Concentration. Nature pp. 207-210 (1953).