0071549226 ar012

12.1 CLASSIFICATION AND SOIL BEHAVIOR

12.1.1 Classification and Engineering Properties

Soil properties that are of most concern in engineering are strength and volumechange under existing and future anticipated loading conditions. Various testshave been devised to determine these behaviors, but the tests can be costly andtime-consuming, and often a soil can be accepted or rejected for a particular useon the basis of its classification alone.

For example, an earth dam constructed entirely of sand would not only leak, itwould wash away. Classification can reveal if a soil may merit furtherinvestigation for founding a highway or building foundation, or if it should berejected and either replaced, modified, or a different site selected. Important cluescan come from the geological and pedological origin, discussed in precedingchapters. Another clue is the engineering classification, which can be useful even ifthe origin is obscure or mixed, as in the case of random fill soil.

12.1.2 Classification Tests

Engineering classifications differ from scientific classifications because they focuson physical properties and potential uses. Two tests devised in the early 1900s by aSwedish soil scientist, Albert Atterberg, are at the heart of engineeringclassifications. The tests are the liquid limit or LL, which is the moisturecontent at which a soil become liquid, and the plastic limit or PL, which isthe moisture content at which the soil ceases to become plastic and crumbles inthe hand.

Both limits are strongly influenced by the clay content and clay mineralogy,and generally as the liquid limit increases, the plastic limit tends to decrease.

12 Soil Consistencyand EngineeringClassification

246

Source: GEOTECHNICAL ENGINEERING

Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com) Copyright © 2007 The McGraw-Hill Companies. All rights reserved. Any use is subject to the Terms of Use as given at the website.

The numerical difference between the two limits therefore represents a range inmoisture contents over which the soil is plastic, and is referred to as the plasticityindex or PI. By definition,

PI ¼ LL� PL ð12:1Þ

where PI is the plasticity index and LL and PL the liquid limit and plastic limit,respectively. This relationship is shown in Fig. 12.1. Because the plasticity indexis a difference in percentages and not in itself a percentage, it is expressed as anumber and not a percent. Also shown in the figure is the shrinkage limit, which isdiscussed later in the chapter.

12.1.3 Preparation of Soil for Testing

As discussed in relation to clay mineralogy, drying a soil can change itsadsorptive capacity for water and therefore can change the liquid and plasticlimits. If the soil contains the clay mineral halloysite, dehydration fromair-drying is permanent, so to obtain realistic data the soil must not be driedprior to testing. A similar change can occur in soils that have a high content oforganic matter.

Air-drying nevertheless is still an approved method because it is more convenientfor storing soil samples and for dry sieving, because only the portion of a soilpassing the No. 40 (425 mm) sieve is tested. Also, many existing correlations weremade on the basis of tests of air-dried samples. If a soil has been air-dried it shouldbe mixed with water for 15 to 30 minutes, sealed and stored overnight, andre-mixed prior to testing. Details are in ASTM D-4318.

Figure 12.1

Schematicrepresentation oftransitions betweensolid, plastic, andviscous liquidbehaviors definedby liquid andplastic limits. Thesetests are basic toengineeringclassifications andemphasizeinfluences of claymineralogy andcapillarity.

Soil Consistency and Engineering Classification 247

Soil Consistency and Engineering Classification


12.1.4 Liquidity Index

The liquidity index indicates how far the natural soil moisture content hasprogressed between the plastic and liquid limits. If the soil moisture content is atthe plastic limit, the liquidity index is 0; if it is at the liquid limit, it is 1.0.

The formula for the liquidity index is

LI ¼w� PLLL� PL

ð12:2Þ

where LI is the liquidity index, w is the soil moisture content, PL is the plastic limit,and LL is the liquid limit. The liquidity index also is called the relative consistency.

12.2 MEASURING THE LIQUID LIMIT

12.2.1 Concept

The concept of the liquid limit is simple: keep adding water to a soil until it flows,and measure the moisture content at that point by oven-drying a representativesample. Two difficulties in application of this concept are (1) the change fromplastic to liquid behavior is transitional, and (2) flow can be prevented bythixotropic setting.

In order to overcome these limitations, Atterberg suggested that wet soil be placedin a shallow dish, a groove cut through the soil with a finger, and the dishjarred 10 times to determine if the groove closes. While this met the challengeof thixotropy, it also introduced a personal factor. Professor A. Casagrandeof Harvard University therefore adapted a cog arrangement invented byLeonardo da Vinci, such that turning a crank drops a shallow brass cupcontaining wet soil 10mm onto a hard rubber block, shown in Fig. 12.2.The crank is turned at 2 revolutions per second, and the groove is standardized.

Casagrande defined the liquid limit as the moisture content at which the groovewould close after 25 blows, which increased the precision of the blow countdetermination. Different amounts of water are added to a soil sample and stirredin, and the test repeated so that the blow counts bracket the required 25. As it isunlikely that the exact number will be achieved at any particular moisture content,a graph is made of the logarithm of the number of blows versus the moisturecontent, a straight line is drawn, and the liquid limit read from the graph wherethe line intersects 25 blows (Fig. 12.3).

12.2.2 Procedure for the Liquid Limit Test

A quantity of soil passing the No. 40 sieve is mixed with water to a pasteconsistency and stored overnight. It is then re-mixed and placed in a standardized

248 Geotechnical Engineering



round-bottomed brass cup, and the surface is struck off with a spatula sothat the maximum thickness is 10mm. The soil pat then is divided into twosegments by means of a grooving tool of standard shape and dimensions.The brass cup is mounted in such a way that, by turning a crank, it can beraised and allowed to fall sharply onto a hard rubber block or base. The shockproduced by this fall causes the adjacent sides of the divided soil pat to flowtogether. The wetter the mixture, the fewer shocks or blows will be requiredto cause the groove to close, and the drier the mixture, the greater will be thenumber of blows.

The number of blows required to close the groove in the soil pat is determinedat three or more moisture contents, some above the liquid limit and somebelow it. The logarithm of the number of blows is plotted versus the moisturecontent and a straight line is drawn through the points, as shown in Fig. 12.3.The moisture content at which 25 blows cause the groove to close is defined

Figure 12.2

Casagrande-daVinci liquid limitdevice.

Figure 12.3

Semilogarithmicplot fordetermining a soilliquid limit.




as the liquid limit. Tests usually are performed in duplicate and average resultsreported.

A ‘‘one-point’’ test may be used for routine analyses, in which the number ofblows is between 20 and 30 and a correction that depends on the departurefrom 25 is applied to the moisture content. See AASHTO Specification T-89 orASTM Specification D-423 for details of the liquid limit test.

12.3 MEASURING THE PLASTIC LIMIT

12.3.1 Concept

Soil with a moisture content lower than the liquid limit is plastic, meaning thatit can be remolded in the hand. An exception is clean sand, which falls apart onremolding and is referred to as ‘‘nonplastic.’’ It is the plasticity of clays that allowsmolding of ceramics into statues or dishes. At a certain point during drying,the clay can no longer be remolded, and if manipulated, it breaks or crumbles;it is a solid. The moisture content at which a soil no longer can be remoldedis the plastic limit, or PL.

The standard procedure used to determine the plastic limit of a soil is deceptivelysimple. The soil is rolled out into a thread, and if it does not crumble it is thenballed up and rolled out again, and again, and again . . . until the thread falls apartduring remolding. It would appear that a machine might be devised to performthis chore, but several factors make the results difficult to duplicate. First, the soilis continuously being remolded, and second, it gradually is being dried while beingremolded. A third factor is even more difficult—the effort required to remold thesoil varies greatly depending on the clay content and clay mineralogy. Despitethese difficulties and the lack of sophistication, the precision is comparable to orbetter than that of the liquid limit test.

12.3.2 Details of the Plastic Limit Test

The plastic limit of a soil is determined in the laboratory by a standardizedprocedure, as follows. A small quantity of the soil-water mixture is rolled out withthe palm of the hand on a frosted glass plate or on a mildly absorbent surface suchas paper until a thread or worm of soil is formed. When the thread is rolled to adiameter of 3mm (18 in.), it is balled up and rolled out again, the mixture graduallylosing moisture in the process. Finally the sample dries out to the extent that itbecomes brittle and will no longer hold together in a continuous thread. Themoisture content at which the thread breaks up into short pieces in this rollingprocess is considered to be the plastic limit (Fig. 12.4). The pieces or crumbstherefore are placed in a small container for weighing, oven-drying, andre-weighing. Generally at least two determinations are made and the results




averaged. See AASHTO Specification T-90 or ASTM Specification D-424 fordetails of the plastic limit test.

12.4 DIRECT APPLICATIONS OF LL AND PL TO FIELD SITUATIONS

12.4.1 When a Soil Moisture Content Exceedsthe Plastic Limit

The liquid limit and plastic limit tests are more diagnostic than descriptive ofsoil behavior in the field because the tests involve continual remolding. However,there are some important situations where remolding occurs more or less con-tinuously in the field. One example is soil in the basal zone of a landslide. As alandslide moves, it shears and mixes the soil. This mixing action can occur if thesoil moisture exceeds the plastic limit. If through chemical treatment such as withdrilled lime (quicklime) the plastic limit is increased, the landslide stops.

12.4.2 When a Soil Moisture Content Exceeds theLiquid Limit

Exceeding the soil liquid limit in the field can generate harmful and potentiallydevastating results, as the soil may appear to be stable and then when disturbedcan suddenly break away, losing its thixotropic strength and becomingtransformed into a rapid churning, flowing mudslide that takes everything in itsway. The rate of sliding depends on the slope angle and viscosity of the mud; thelower the viscosity and steeper the slope, the faster the slide. The most devastatingmudslides in terms of loss of life therefore occur in mountainous terrain where themud moves faster than people can get out of the way and escape almost certain

Figure 12.4

The plastic limittest. As the soilthread crumbs,pieces arecollected in a metalcontainer for oven-drying to determinethe moisturecontent.




death. Quick clays that appear stable can turn into a soup that can be pouredlike pancake batter.

12.4.3 Liquefaction

Another example where a liquid limit may be exceeded is when a saturated sandor silt suddenly densifies during an earthquake so that all of its weight goes topore water pressure. This is liquefaction, which can cause a sudden and completeloss of shear strength so that landslides develop and buildings may topple. Theconsequences, diagnosis, and prevention of liquefaction are discussed in moredetail in a later chapter.

12.5 THE PLASTICITY INDEX

12.5.1 Concept

The plasticity index, or PI, is the numerical difference between the liquid andplastic limit moisture contents. Whereas the two limits that are used to define a PIare directly applicable to certain field conditions, the plasticity index is mainlyused to characterize a soil, where it is a measure of cohesive properties. Theplasticity index indicates the degree of surface chemical activity and hence thebonding properties of clay minerals in a soil. The plasticity index is used alongwith the liquid limit and particle size gradation to classify soils according to theirengineering behavior.

An example of a direct application of the plasticity index is as an indicator of thesuitability of the clay binder in a soil mixture used for pavement subgrades, basecourses, or unpaved road surfaces. If the PI of the clay fraction of a sand-clay orclay-gravel mixture is too high, the exposed soil tends to soften and becomeslippery in wet weather, and the road may rut under traffic. On the other hand,if the plasticity index is too low, the unpaved road will tend to ‘‘washboard’’ inresponse to resonate bouncing of wheels of vehicular traffic. Such a road willabrade under traffic and antagonize the public by producing air-borne dust inamounts that have been measured as high as one ton per vehicle mile per day peryear. That is, a rural unpaved road carrying an average of 40 vehicles per day cangenerate up to 40 tons of dust per mile per year. Most collects in roadside ditchesthat periodically must be cleaned out.

12.5.2 A PI of Zero

Measurements of the LL and PL may indicate that a soil has a plasticityindex equal to zero; that is, the numerical values of the plastic limit and theliquid limit may be the same within the limits of accuracy of measurement.Soil with a plasticity index of zero therefore still exhibits a slight plasticity, but




the range of moisture content within which it exhibits the properties of a plasticsolid is not measured by the standard laboratory tests.

12.5.3 Nonplastic Soils

Drying and manipulating a truly nonplastic soil such as a clean sand will cause itto abruptly change from a liquid state to an incoherent granular material thatcannot be molded. If it is not possible to roll soil into a thread as small as 3mm indiameter, a plastic limit cannot be determined and the soil is said to be nonplastic,designated as NP in test reports.

12.6 ACTIVITY INDEX AND CLAY MINERALOGY

12.6.1 Definition of Activity Index

The activity index was defined by Skempton to relate the PI to the amount of clayin a soil, as an indication of the activity of the clay and therefore the clay mineral-ogy: the higher the activity index, or AI, the more active the clay. The activity indexwas defined as the PI divided by the percent 0.002mm (or 2mm) clay:

AI ¼PI

C002ð12:3Þ

where AI is the activity index, PI the plasticity index, and C002 the percent 2 mmclay determined from a particle size analsysis. The basis for this relationshipis shown in Fig. 12.5.

12.6.2 Relation to Clay Mineralogy

The relationship between activity index and clay mineralogy is shown inTable 12.1. Clay mineral mixtures and interlayers have intermediate activities.Data in the table also show how the activity of smectite is strongly influenced bythe adsorbed cation on the plasticity index.

12.6.3 Modified Activity Index

The linear relationships in Fig. 12.5 do not necessarily pass through the origin.This is shown in Fig. 12.6, where about 10 percent clay is required to generateplastic behavior. This also has been found in other investigations (Chen, 1988).

It therefore is recommended that for silty soils eq. (12.3) be modified as follows:

A ¼PI

C002 � kð12:4Þ

where A is the activity, C is the percent of the soil finer than 0.002mm, andk is a constant that depends on the soil type. For silty soils k¼ 10. If this equation




is applied to the data in Fig. 12.6 with k¼ 10, then A¼ 1.23� 0.04 wherethe � value is the standard error for a number of determinations n¼ 81.This identifies the clay mineral as calcium smectite, which is confirmed by X-raydiffraction.

12.7 LIQUID LIMIT AND COLLAPSIBILITY

12.7.1 Concept

A simple but effective idea was proposed in 1953 by a Russian geotechnicalengineer, A.Y. Denisov, and later introduced into the U.S. by Gibbs and Bara of

Figure 12.5

Linear relationsbetween PI andpercent clay.Ratios weredefined bySkempton (1953)as activity indices,which are shownin parentheses.The Shellhavensoil clay probablyis smectite.

Table 12.1

Activity indices of

selected clay

minerals (after Grim,

1968)

Naþ smectite (montmorillonite) 3–7

Ca2þ smectite (montmorillonite) 1.2–1.3

Illite 0.3–0.6

Kaolinite, poorly crystallized 0.3–0.4

Kaolinite, well crystallized 50.1




the U.S. Bureau of Reclamation. Denisov argued that if the moisture contentupon saturation exceeds the liquid limit, the soil should be collapsible—that is,it should collapse and densify under its own weight if it ever becomes saturated.The most common collapsible soil is loess, which is a widespread surficial depositin the U.S., Europe, and Asia. Because loess increases in density with depthand with distance from a source, only the upper material close to a source may becollapsible, so this is a valuable test.

12.7.2 Moisture Content Upon Saturation

The soil unit weight and specific gravity of the soil mineral grains are requiredto calculate the moisture content upon saturation, which are entered into thefollowing equations:

SI: ws ¼ 100 9:807=�d � 1=Gð Þ ð12:5Þ

English: ws ¼ 100 62:4=�d � 1=Gð Þ ð12:5aÞ

where ws is the percent moisture at saturation, �d is the dry unit weight in kN/m3

or lb/ft3, and G is the specific gravity of the soil minerals. Solutions of thisequation with G¼ 2.70 are shown in Fig. 12.7.

12.8 CONSISTENCY LIMITS AND EXPANSIVE SOILS

12.8.1 Measuring Expandability

Expandability can be determined with a consolidometer, which is a device that wasdeveloped to measure compression of soil but also can be used to measureexpansion under different applied loads. Samples are confined between porousceramic plates, loaded vertically, wet with water, and the amount of expansionmeasured. An abbreviated test measures expansion under only applied pressuresthat can simulate a floor or a foundation load.

Figure 12.6

Data suggesting amodification toeq. (12.3) for siltysoil.




Many investigations have been made relating expansion to various para-meters, including activity, percent finer than 0.002mm, percent finer than0.001mm, plasticity index, and liquid limit. For the most part the studieshave used artificially prepared soil mixtures with varying amounts of differentclay minerals.

12.8.2 Influence of Surcharge Pressure

Generally the higher the vertical surcharge pressure, the lower the amount ofexpansion. This leads to a common observation in buildings founded onexpansive clay: floors in contact with the soil are lifted more than foundationsthat are supporting bearing walls and columns and therefore are more heavilyloaded. Partition walls that are not load-bearing are lifted with the floor.

12.8.3 Lambe’s PVC Meter

A rapid method for measuring clay expandability was developed by T.W. Lambeand his coworkers at MIT. In this device, soil expands against a spring-loadedplate and the expansion is measured. Because the vertical stress increases as thesoil expands, results are useful for classification but do not directly translate intoexpansion amounts that may be expected in the field.

12.8.4 Influence of Remolding

Chen (1988) emphasizes that expansion is much lower for undisturbed thanfor disturbed soil samples subjected to the same treatment, indicating an impor-tant restraining influence from soil fabric. Therefore the expansive clay thatis inadvertently used for fill soil, as sometimes happens, may expand much

Figure 12.7

Denisov criterionfor collapsibilitywith G¼2.70.Data are for loessat 3, 40, and 55 ft(1, 12, and 16.8 m)depths in HarrisonCounty, Iowa. Thedeepest soil wasmottled gray,suggesting ahistory of wetconditions, and isindicated to benoncollapsible.




more than if the clay were not disturbed. The reason for this has not beeninvestigated, but an argument may be made for a time-related cementation effectof edge-to-face clay particle bonding, which would prevent water from enteringand separating the clay layers.

12.8.5 Relation to PI

Figure 12.8 shows the conclusions of several researchers who related clay expand-ability to the plasticity index or PI. Curves A and B show results for remoldedsamples, and curves C and D are from undisturbed samples where surcharges wereapplied to more or less simulate floor and foundation loads, respectively. It will beseen that the lowest expandability is shown by curve D, which is for undisturbedsoil under the foundation load.

12.8.6 Relation to Moisture Content

Generally expansion pressure decreases as the soil moisture content increases,and expansion stops when the smectite clay is fully expanded. This depends on therelative humidity of the soil air and occurs well below the point of saturation ofthe soil itself. This is illustrated in Fig. 12.9, where seasonal volume changeoccurs between 30 and 70 percent saturation. Expansion will not occur in a claythat already is wet, which is not particularly reassuring because damagingshrinkage still can occur if and when the clay dries out.

Figure 12.8

Data indicating thatremolding and aloss of structuregreatly increaseswelling pressures.All but curve D,which has asurcharge pressureof 1000 lb/ft2

(44 kPa) have asurchargepressure of 1 lb/in.2

(6.9 kPa).(Modified fromChen, 1988.)




12.8.7 Summary of Factors Influencing Expandability

The amount of expansion that can be anticipated depends on at least seven

variables: (1) clay mineralogy, and (2) clay content, both of which are reflected in

the consistency limits; (3) existing field moisture content; (4) surcharge pressure;

(5) whether or not the soil is remolded; (6) thickness of the expanding layer; and

(7) availability of water.

12.8.8 Thickness of the Active Layer

One of the most extensive expansive clay areas in the world is in India, but

detailed field investigations indicate that only about the upper 90 cm (3 ft) of the

expansive soil actually experiences seasonal volume changes. Below that depth

the clay is volumetrically stable, even though, as seen in the second graph

of Fig. 12.9, the moisture content is not. As previously indicated, saturation is

not required for full expansion of Ca-smectite, which is the most common

expansive clay.

The surficial layer involved in seasonal volume change is called the active layer,

and determines the depth of shrinkage cracking and vertical mixing, which by

disrupting the soil structure tends to increase its expandability.

Example 12.1Calculate the seasonal ground heave from data in Fig. 12.9.

Answer: If the soil is divided into three layers, 0–30, 30–60, and 60–90 cm, average increases

in density from the left-hand graph are approximately 0.5/1.22¼ 41%; 0.2/1.3¼ 15%; and

0.1/1.3¼ 8%, respectively. Multiplying these percentages by the layer thicknesses gives total

volume changes from the dry to the wet seasons of (0.41þ 0.15þ 0.08)� 30¼ 19 cm

(7.5 in.). However, part of this will go toward closing open ground cracks, in which case

one-third of the volume change will be directed vertically, about 6 cm or 2.5 in. The answer

Figure 12.9

Seasonal volumechanges in Poonaclay, India. Leftgraph shows thatexpandability islimited to the uppermeter despitedeeper variationsin moisturecontent. (FromKatti and Katti,1994.)




therefore is between 6 and 19 cm, depending on filling of the shrinkage cracks andamount of lateral elastic compression of the soil.

12.8.9 Controlling Volume Change with aNonexpansive Clay (n.e.c.) Layer

One of the most significant discoveries for controlling expansive clay was byDr. R. K. Katti and his co-workers at the Indian Institute of Technology, Mumbai.Katti’s group conducted extensive full-scale laboratory tests to confirm fieldmeasurements, such as shown in Fig. 12.9, and found that expansion can becontrolled by a surficial layer of compacted non-expansive clay. A particularlysevere test for the design was the canal shown in Fig. 12.10. The most commonapplication of Katti’s method is to stabilize the upper meter (3 ft) of expansiveclay by mixing in hydrated lime, Ca(OH)2. If only the upper one-third, 30 cm(1 ft), is stabilized, volume change will be (0.15þ 0.08)� 30¼ 7 cm (3 in.), areduction of about 60 percent. If the upper 60 cm (2 ft) is stabilized, the volumechange will be 0.08� 30¼ 2.4 cm (1 in.), a reduction of over 85 percent.Stabilization to the full depth has been shown to eliminate volume changealtogether. The next question is, why?

An answer may be in the curves in Fig. 12.8, as a loss of clay structure greatlyincreases clay expandability. As a result of shrink-swell cycling and an increase inhorizontal stress, expansive clays are visibly sheared, mixed, and remolded, so bydestroying soil structure expansion probably begets more expansion. According tothis hypothesis, substituting a layer of nonexpansive clay for the upper highlyexpansive layer may help to preserve the structure and integrity of the underlyinglayer. It was found that using a sand layer or a foundation load instead of densely

Figure 12.10

The MalaprabhaCanal in India wassuccessfully builton highlyexpansive clayusing Katti’smethod, byreplacing the upper1 m of soil withcompactednonexpansive clay(n.e.c.) toreplicate theconditions shownin Fig. 12.9 for theunderlying soil.




compacted clay is less effective, perhaps because it interrupts the continuum (Kattiand Katti, 2005).

12.9 PLASTICITY INDEX VS. LIQUID LIMIT

12.9.1 Concept

The relationship between PI and LL reflects clay mineralogy and has an advan-tage over the activity index because a particle size analysis is not required. Becausethe liquid limit appears on both sides of the relationship, data can plot only withina triangular area defined by the PL¼ 0 line shown in Fig. 12.11.

12.9.2 The A-Line

A line that approximately parallels the PI versus LL plot for particular soil groupsis called the A-line, which was proposed by A. Casagrande and for the most partseparates soils with and without smectite clay minerals. However, the separation isnot always consistent, as can be seen in Fig. 12.11 where loess crosses the line.At low clay contents loessial soils also are more likely to show collapse behaviorinstead of expanding.

12.10 A SOIL CLASSIFICATION BASED ON THE A-LINE

12.10.1 Background

During World War II, Arthur Casagrande devised a simplified soil classificationsystem for use by the armed forces. The objective was a system that could be usedto classify soils from visual examination and liquid/plastic behavior. In 1952

Figure 12.11

Representativerelationshipsbetween PI andLL. (Adapted fromU.S. Dept. ofInterior Bureau ofReclamation,1974.)




the system was adopted for civilian uses by the U.S. Bureau of Reclamation andthe U.S. Army Corps of Engineers and became known as the UnifiedClassification. The ASTM Designation is D-2487. The system applies not onlyto fine-grained soils but also to sands and gravels, and is the most widely usedsystem for soil investigations for building foundations and tunneling.

12.10.2 ‘‘S’’ Is for Sand

One advantage of the Unified Classification system is its simplicity, as it usescapital letters to represent particular soil properties: S stands for sand, G forgravel, and C for clay. Because S already is used for sand, another letter, M, wasselected for silt, from the German word Moh.

A sand or gravel can either be well graded, W, or for poorly graded, designated byP, respectively indicating broad or narrow ranges of particle sizes. Thus, SP is apoorly graded sand, GW a well-graded gravel.

Fine-grained soils are characterized on the basis of liquid limit and the PI andLL relationships to the A-line. A silt or clay with a liquid limit higher than50 percent is designated by H, meaning high liquid limit, and if the data plot abovethe A-line the soil is CH, clay with a high liquid limit. If the liquid limit is higherthan 50 percent and the data plot below the A-line, the designation is MH, siltwith a high liquid limit.

The Unified Classification system therefore distinguishes between silt and clay noton the basis of particle size, but on relationships to the liquid limit and plasticityindex. In order to avoid confusion, clay and silt that are defined on the basis ofsize now usually are referred to as ‘‘clay-size’’ or ‘‘silt-size’’ material.

If the silt or clay liquid limit is lower than 50, the respective designations areML and CL, silt with a low liquid limit or clay with a low liquid limit. However, ifthe plasticity index is less than 4, silt dominates the soil behavior and the soil isdesignated ML. This is shown in the graph in Table 12.2. Soils with a plasticityindex between 4 and 7 show properties that are intermediate and are designatedCL-ML.

12.10.3 Details of the Unified Classification System

Letter abbreviations for the various soil characteristics are as follows:

G¼Gravel O¼Organic

S¼Sand W¼Well graded

M¼Nonplastic or low plasticity P¼Poorly graded

C¼Plastic fines L¼ Low liquid limit

Pt¼Peat, humus, swamp soils H¼High liquid limit




Table 12.2

Unified soil classification system







These letters are combined to define various groups as shown in Table 12.2.This table is used to classify a soil by going from left to right and satisfying theseveral levels of criteria.

12.10.4 Equivalent Names

Some equivalent names for various combined symbols are as follows. Thesenames are appropriate and should be used in reports that may be read by peoplewho are not familiar with the soil classification symbols. For example, describinga soil as an ‘‘SC clayey sand’’ will be more meaningful than to only refer to it as‘‘SC,’’ and illustrates the logic in the terminology. The meanings of the symbolsfor coarse-grained soils are fairly obvious. Names used for fine-grained soils are asfollows:

CL¼ lean clay CH¼ fat clay

ML¼ silt MH¼ elastic silt

OL¼ organic silt OH¼ organic clay

If a fine-grained soil contains over 15 percent sand or gravel, it is referred to as‘‘with sand or with gravel;’’ if over 30 percent, it is ‘‘sandy’’ or ‘‘gravelly.’’ If over50 percent it goes into a coarse-grained classification. If a soil contains anycobbles or boulders it is referred to as ‘‘with cobbles’’ or ‘‘with boulders.’’ Othermore detailed descriptors will be found in ASTM D-2487.

12.10.5 Detailed Descriptions

Descriptions of the various groups that may be helpful in classification are asfollows:

GW and SWSoils in these groups are well-graded gravelly and sandy soils that contain lessthan 5 percent nonplastic fines passing the No. 200 sieve. The fines that arepresent do not noticeably affect the strength characteristics of the coarse-grainedfraction and must not interfere with its free-draining characteristic. In areassubject to frost action, GW and SW soils should not contain more than about3 percent of soil grains smaller than 0.02mm in size.

GP and SPGP and SP soils are poorly graded gravels and sands containing less than5 percent of nonplastic fines. The soils may consist of uniform gravels, uniformsands, or nonuniform mixtures of very coarse material and very fine sand withintermediate sizes lacking, referred to as skip-graded, gap-graded, or step-graded.

GM and SMIn general, GM and SM soils are gravels or sands that contain more than12 percent fines having little or no plasticity. In order to qualify as M, the




plasticity index and liquid limit plot below the A-line on the plasticity chart.Because of separation of coarse particles gradation is less important, and bothwell-graded and poorly graded materials are included in these groups. Some sandsand gravels in these groups may have a binder composed of natural cementingagents, so proportioned that the mixture shows negligible swelling or shrinkage.Thus, the dry strength is provided either by a small amount of soil binder or bycementation of calcareous materials or iron oxide. The fine fraction ofnoncemented materials may be composed of silts or rock-flour types havinglittle or no plasticity, and the mixture will exhibit no dry strength.

GC and SCThese groups consist of gravelly or sandy soils with more than 12 percent fines thatcan exhibit low to high plasticity. The plasticity index and liquid limit plot abovethe A-line on the plasticity chart. Gradation of these materials is not important, asthe plasticity of the binder fraction has more influence on the behavior of the soilsthan does variation in gradation. The fine fraction is generally composed of clays.

Borderline G and S ClassificationsIt will be seen that a gap exists between the GW, SW, GP, and SP groups, whichhave less than 5 percent passing the No. 200 sieve, and GM, SM, GC, and SCsoils, which have more than 12 percent passing the No. 200 sieve. Soils containingbetween 5 and 12 percent fines are considered as borderline and are designated bya dual symbol such as GW-GM if the soil is a well-graded gravel with a siltcomponent, or GW-GC if well-graded with a clay component. Many other dualsymbols are possible, and the meaning should be evident from the symbol. Forexample, SP-SC is a poorly graded sand with a clay component, too much clay tobe SP and not enough to be SC.

ML and MHML and MH soils include soils that are predominantly silts, and also includemicaceous or diatomaceous soils. An arbitrary division between ML and MH isestablished where the liquid limit is 50. Soils in these groups are sandy silts, clayeysilts, or inorganic silts with relatively low plasticity. Also included are loessial soilsand rock flours.

Micaceous and diatomaceous soils generally fall within the MH group but mayextend into the ML group when their liquid limit is less than 50. The same is truefor certain types of kaolin clays and some illitic clays having relatively lowplasticity.

CL and CHThe CL and CH groups embrace clays with low and high liquid limits,respectively. These are mainly inorganic clays. Low-plasticity clays are classifiedas CL and are usually lean clays, sandy clays, or silty clays. The medium-plasticityand high-plasticity clays are classified as CH. These include the fat clays, gumbo




clays, certain volcanic clays, and bentonite. The glacial clays of the northern U.S.cover a wide band in the CL and CH groups.

ML-CLAnother type of borderline classification that already has been commented on iswhen the liquid limit of a fine-grained soil is less than 29 and the plasticity indexlies in the range from 4 to 7. These limits are indicated by the shaded area on theplasticity chart in Fig. 12.11, in which case the double symbol, ML-CL, is used todescribe the soil.

OL and OHOL and OH soils are characterized by the presence of organic matter and includeorganic silts and clays. They have plasticity ranges that correspond to those of theML and MH groups.

PtHighly organic soils that are very compressible and have very undesirableconstruction characteristics are classified in one group with the symbol Pt. Peat,humus, and swamp soils with a highly organic texture are typical of the group.Particles of leaves, grass, branches of bushes, or other fibrous vegetable matter arecommon components of these soils.

12.11 FIELD USE OF THE UNIFIED SOIL CLASSIFICATION SYSTEM

12.11.1 Importance of Field Identitication

A detailed classification such as indicated in Table 12.2 requires both a gradationand plasticity analysis. Even after this information is available from laboratorytests of soil samples, it is important to be able to identify the same soils in thefield. For example, if a specification is written based on the assumption that a soilis an SC, and the borrow excavation proceeds to cut into ML, it can be veryimportant that somebody serves notice and if necessary issues a stop order. This isa reason why all major construction jobs include on-site inspection. Suggestionsfor conduct of a field identification using the Unified Classification system are inASTM D-2488.

12.11.2 Granular Soils

A dry sample of coarse-grained material is spread on a flat surface to determinegradation, grain size and shape, and mineral composition. Considerable skill isrequired to visually differentiate between a well-graded soil and a poorly gradedsoil, and is based on visual comparisons with results from laboratory tests.

The durability of coarse aggregate is determined from discoloration of weatheredmaterials and the ease with which the grains can be crushed. Fragments of shale or




other rock that readily breaks into layers may render a coarse-grained soilunsuitable for certain purposes, since alternate wetting and drying may cause it todisintegrate partially or completely. This characteristic can be determined bysubmerging thoroughly dried particles in water for at least 24 hours and observingslaking or testing to determine a loss of strength.

12.11.3 Fine-Grained vs. Coarse-Grained Soils

As shown in Table 12.2, fine-grained soils are defined as having over 50 percentpassing the No. 200 sieve. The percentage finer can be estimated without the useof a sieve and weighing device, by repeatedly mixing a soil sample with water anddecanting until the water is clear, and then estimating the proportion of materialthat has been removed. Another method is to place a sample of soil in a largetest tube, fill the tube with water and shake the contents thoroughly, and thenallow the material to settle. Particles retained on a No. 200 sieve will settle out ofsuspension in about 20 to 30 seconds, whereas finer particles will take a longertime. An estimate of the relative amounts of coarse and fine material can be madeon the basis of the relative volumes of the coarse and fine portions of thesediment.

12.11.4 Fine-Grained Soils

Field identification procedures for fine-grained soils involve testing for dilatancy,or expansion on shaking, plasticity, and dry strength. These tests are performedon the fraction of soil finer than the No. 40 sieve. In addition, observations ofcolor and odor can be important. If a No. 40 sieve is not available, removal of thefraction retained on this sieve may be partially accomplished by hand picking.Some particles larger than this sieve opening (0.425mm, or nominally 0.5mm)may remain in the soil after hand separation, but they probably will have only aminor effect on the field tests.

DiIatancyFor the dilatancy test, enough water is added to about 2 cm3 (12 in.

3) of theminus-40 fraction of soil to make it soft but not sticky. The pat of soil is shakenhorizontally in the open palm of one hand, which is struck vigorously against theother hand several times. A fine-grained soil that is nonplastic or has very lowplasticity will show free water on the surface while being shaken, and thensqueezing the pat with the fingers will cause the soil structure to dilate or expandso that the soil appears to dry up. The soil then will stiffen and finally crumbleunder increasing pressure. Shaking the pat again will cause it to flow together andwater to again appear on the surface.

A distinction should be made between a rapid reaction, a slow reaction, or noreaction to the shaking test, the rating depending on the speed with which the patchanges its consistency and the water on the surface appears or disappears. Rapid




reaction is typical of nonplastic, uniform fine sand, of silty sand (SP or SM), ofinorganic silt (ML), particularly the rock-flour type, and of diatomaceous earth(MH). The reaction becomes more sluggish as the uniformity of gradationdecreases and the plasticity increases, up to a certain degree. Even a small amountof colloidal clay will impart some plasticity to the soil and will materially slowthe reaction to the shaking test. Soils that react in this manner are somewhatplastic inorganic and organic silts (ML or OL), very lean clays (CL), and somekaolin-type clays (ML or MH). Extremely slow reaction or no reaction to theshaking test is characteristic of typical clays (CL or CH) and of highly plasticorganic clays (OH).

Field Estimate of PlasticityThe plasticity of a fine-grained soil or the binder fraction of a coarse-grained soilmay be estimated by rolling a small sample of minus-40 material between thepalms of the hand in a manner similar to the standard plastic limit test.The sample should be fairly wet, but not sticky. As it is rolled into 1

8-inch threads,folded and re-rolled, the stiffness of the threads should be observed. The higherthe soil above the A-line on the plasticity chart (CL or CH), the stiffer the threads.Then as the water content approaches the plastic limit, the tougher are the lumpsafter crumbling and remolding. Soils slightly above the A-line (CL or CH) form amedium-tough thread that can be rolled easily as the plastic limit is approached,but when the soil is kneaded below the plastic limit, it crumbles.

Soils below the A-line (ML, NH, OL, or OH) form a weak thread and with theexception of an OH soil, such a soil cannot be lumped into a coherent mass belowthe plastic limit. Plastic soils containing organic material or much mica formthreads that are very soft and spongy near the plastic limit.

In general, the binder fraction of a coarse-grained soil with silty fines (GM or SM)will exhibit plasticity characteristics similar to those ofMLsoils. The binder fractionof a coarse-grained soil with clayey fines (GC or SC) will be similar to CL soils.

Field Estimate of Dry StrengthDry strength is determined from a pat of minus-40 soil that is moistened andmolded to the consistency of putty, and allowed to dry in an oven or in the sunand air. When dry the pat should be crumbled between the fingers. ML or MHsoils have a low dry strength and crumble with very little finger pressure. Also,organic siIts and lean organic clays of low plasticity (OL) and very fine sandy soils(SM) also have low dry strength.

Most clays of the CL group and some OH soils, as well as the binder fraction ofgravelly and sandy clays (GC or SC), have medium dry strength and requireconsiderable finger pressure to crumble the sample. Most CH clays and someorganic clays (OH) having high liquid limits and located near the A-line have highdry strength, and the test pat can be broken with the fingers but cannot becrumbled.




Color and OdorDark or drab shades of gray or brown to nearly black indicate fine-grained soilscontaining organic colloidal matter (OL or OH), whereas brighter colors,including medium and light gray, olive green, brown, red, yellow, and white, aregenerally associated with inorganic soils.

An organic soil (OL or OH) usually has a distinctive odor that can be helpful forfield identification. This odor is most obvious in a fresh sample and diminishes onexposure to air, but can be revived by heating a wet sample.

The details of field identification are less imposing and more easily remembered ifthey are reviewed in relation to each particular requirement. For example, if aspecification is for SC, the criteria for an SC soil should be reviewed andunderstood and compared with those for closely related soils, SM and SP andrespective borderline classifications.

12.12 THE AASHTO SYSTEM OF SOIL CLASSIFICATION

12.12.1 History

A system of soil classification was devised by Terzaghi and Hogentogler for theU.S. Bureau of Public Roads in the late 1920s, predating the Unified Classificationsystem by about 20 years. The Public Roads system was subsequently modifiedand adopted by the American Association of State Highway Officials (nowHighway and Transportation Officials) and is known as the AASHTO system(AASHTO Method M14S; ASTM Designation D-3282).

As in the Unified Classification system, the number of physical properties of a soilupon which the classification is based is reduced to three—gradation, liquid limit,and plasticity index. Soil groups are identified as A-1 through A-8 for soilsranging from gravel to peat. Generally, the higher the number, the less desirablethe soil for highway uses.

12.12.2 Using the AASHTO Chart

The process of determining the group or subgroup to which a soil belongs issimplified by use of the tabular chart shown in Table 12.3. The procedure is asfollows. Begin at the left-hand column of the chart and see if all these knownproperties of the soil comply with the limiting values specified in the column. Ifthey do not, move to the next column to the right, and continue across the chartuntil the proper column is reached. The first column in which the soil properties fitthe specified limits indicates the group or subgroup to which the soil belongs.Group A-3 is placed before group A-2 in the table to permit its use in this mannereven though A-3 soils normally are considered less desirable than A-2 soils.




The ranges of the liquid limit and the plasticity index for fine-grained soils ingroups A-4, A-5, A-6, and A-7 are shown in Fig. 12.12, which has been arrangedto be comparable to the Unified chart.

Example 12.2Classify a soil containing 65% of material passing a No. 200 sieve and having a liquid limitof 48 and a plasticity index of 17.

Answer: Since more than 35% of the soil material passes the No. 200 sieve, it is a

silt-clay material and the process of determining its classification can begin by examiningthe specified limits for group A-4, where the maximum is 40. Since the liquid limit ofthe soil being classified is 48%, it cannot be an A-4 soil so we proceed to the columns to

the right, where it will be seen that it meets the liquid limit requirement of A-6 and A-7,but meets the plasticity index requirements of A-7. The soil therefore is A-7. This procedureis simplified by reference to Fig. 12.12, which also is used to separate A-7 soils into two

subgroups, A-7-5 and A-7-6.

12.12.3 Size Grade Definitions

AASHTO definitions of gravel, sand, and silt-cIay are as follows:

GravelMaterial passing a sieve with 75mm (3 in.) square openings and retained on aNo. 10 (2mm) sieve.

Coarse SandMaterial passing the No. 10 sieve and retained on the No. 40 (425 mm) sieve.

Fine SandMaterial passing the No. 40 sieve and retained on the No. 200 (75 mm) sieve.

Figure 12.12

Chart forclassifyingfine-grained soilsby the AASHTOsystem.




Tab

le12.3

Soil

cla

ssific

ation

by

the

AA

SH

TO

syste

m

Genera

l

Cla

ssific

ation

Gra

nula

rM

ate

rials

(35%

or

less

passin

gN

o.

200)

Silt

-Cla

yM

ate

rials

(More

than

35%

passin

gN

o.

200)

A-7

A-1

A-2

A-7

-5

Gro

up

Cla

ssific

ation

A-1

-aA

-1-b

A-3

A-2

-4A

-2-5

A-2

-6A

-2-7

A-4

A-5

A-6

A-7

-6

Sie

ve

analy

sis

,perc

ent

passin

g:

No.

10

50

max.

No.

40

30

max.

50

max.

51

min

.

No.

200

15

max.

25

max.

10

max.

35

max.

35

max.

35

max.

35

max.

36

min

.36

min

.36

min

.36

min

.

Chara

cte

ristics

of

fraction

passin

g

No.

40:

Liq

uid

limit

40

max.

41

min

.40

max.

41

min

.40

max.

41

min

.40

max.

41

min

.b

Pla

sticity

index

6m

ax.

NP

10

max.

10

max.

11

min

.11

min

.10

max.

10

max.

11

min

.11

min

.

Usualty

pes

of

sig

nific

ant

constitu

ent

mate

rials

Sto

ne

fragm

ents

,

gra

veland

sand

Fin

esand

Silt

yor

cla

yey

gra

veland

sand

Silt

ysoils

Cla

yey

soils

Genera

lra

ting

as

subgra

de

Excelle

nt

togood

Fair

topoor

aC

lassific

ation

pro

cedure

:W

ith

required

test

data

availa

ble

,pro

ceed

from

left

toright

on

above

chart

and

corr

ect

gro

up

will

be

found

by

the

pro

cess

of

elim

ination.

The

firs

tgro

up

from

the

left

into

whic

hth

ete

st

data

will

fit

isth

ecorr

ect

cla

ssific

ation.

bP

lasticity

index

of

A-7

-5subgro

up

isequalto

or

less

than

LL

min

us

30.

Pla

sticity

index

of

A-7

-6subgro

up

isgre

ate

rth

an

LL

min

us

30

(see

Fig

.12.1

2)




Silt-Clay or Combined SiltþClayMaterial passing the No. 200 sieve.

BouldersBoulders retained on the 75mm (3 in.) sieve are excluded from the portion of thesample being classified, but the percentage of such material is recorded.

The term ‘‘silty’’ is applied to fine material having a plasticity index of 10 or less,and the term ‘‘clayey’’ is applied to fine material having a plasticity index of 11 ormore after rounding to the nearest whole percent.

12.12.4 Descriptions of Groups

The following generalized observations may be applied to the various AASHTOsoil groups:

A-1Typical of this group are well-graded mixtures of stone fragments or gravel,volcanic cinders, or coarse sand. They do not contain a soil binder or have anonplastic or feebly plastic binder. Subgroup A-1-a is mainly stone fragments orgravel, and A-1-b is mainly coarse sand.

A-3Typical of this group is fine beach sand or fine desert blow sand without silty orclayey fines, or with a very small amount of nonplastic silt. The group alsoincludes stream-deposited mixtures of poorly graded fine sand with limitedamounts of coarse sand and gravel.

A-2This group includes a wide variety of granular materials that are at the borderlinebetween A-1 and A-3 and silt-clay materials of groups A-4 through A-7. A-2includes materials with less than 35 percent passing a No. 200 sieve that do notclassify as A-1 or A-3, because either the fines content or plasticity, or both, are inexcess of the amounts allowed in those groups.

Subgroups A-2-4 and A-2-5 include various granular materials with not morethan 35 percent passing a No. 200 sieve and containing a minus No. 40 portionthat has characteristics of the A-4 and A-5 groups, respectively. These subgroupsinclude such materials as gravel and coarse sand with silt content or plasticityindex in excess of those allowed in A-1, and fine sand with nonplastic silt contentin excess of the limitations of group A-3.

Subgroups A-2-6 and A-2-7 include materials similar to those described undersubgroups A-2-4 and A-2-5, except that the fine portion contains plastic clay




having the characteristics of the A-6 or the A-7 group. The group index, describedbelow, is 0 to 4.

A-4The typical material of this group is a nonplastic or moderately plastic silty soil,75 percent or more of which passes the No. 200 sieve. However, the group also canincludemixtures of fine silty soil with up to 64 percent retained on theNo. 200 sieve.

A5Typical of this group is soil that is similar to that described under group A-4, buthas a diatomaceous or micaceous content that makes it highly elastic, indicated bya high liquid limit. These soils are ‘‘springy’’ and may be difficult to compact.

A-6The material of this group typically is plastic clay soil with 75 percent or morepassing the No. 200 sieve, but can include fine clayey soil mixtures with up to64 percent retained on the No. 200 sieve. Materials of this group usually have highvolume change between wet and dry states.

A-7A-7 soils are similar to A-6 but have higher liquid limits. Subgroup A-7-5materials have moderate plasticity indexes in relation to liquid limit, and whichmay be highly elastic as well as subject to considerable volume change on wettingor drying. Subgroup A-7-6 materials have high plasticity indexes in relation toliquid limit, and are subject to very high volume changes.

A-8A-8 soil is peat or muck soil in obviously unstable, swampy areas. A-8 soil ischaracterized by low density, high compressibility, high water content, and highorganic matter content. Attention is directed to the fact that the classification ofsoils in this group is based largely upon the character and environment of theirfield occurrence, rather than upon laboratory tests of the material. As a matter offact, A-8 soils usually show laboratory-determined properties of an A-7 soil, butare properly classified as group A-8 because of the manner of their occurrence.

12.12.5 Group Index

The group index gives a means for further rating a soil within its groupor subgroup. The index depends on the percent passing the No. 200 sieve, theliquid limit, and the plasticity index. It is computed by the following empiricalformula:

Group index ¼ F� 35ð Þ 0:2þ 0:005 LL� 40ð Þ½ � þ 0:01 F� 15ð Þ PI� 10ð Þ ð12:6Þ

in which F is the percent passing the No. 200 sieve, expressed as a whole numberand based only on the material passing the 75mm (3 in.) sieve, LL is the liquid




limit, and PI is the plasticity index. When the calculated group index is negative itis reported as zero (0).

The group index is expressed to the nearest whole number and is written inparentheses after the group or subgroup designation. A group index should begiven for each soil even if the numerical value is zero, in order to indicate that theclassification has been determined by the AASHTO system instead of the originalPublic Roads system. A nomograph has been devised to solve eq. (12.6), but itnow is more conveniently solved with a computer spreadsheet.

12.13 LIMITATIONS AND COMPARISONS OFSOIL CLASSIFICATION SYSTEMS

The classification systems described above use disturbed soil properties andtherefore do not take into account factors such as geological origin, fabric,density, or position of a groundwater table. The classifications nevertheless doprovide important information relative to soil behavior so long as the limitationsare recognized. Classification is no substitute for measurements of important soilproperties such as compressibility, shear strength, expandability, permeability,saturation, pore water pressure, etc.

Boundary lines for fine soils in the Unified and AASHTO classification systemsdo not precisely coincide, but the systems are close enough that there isconsiderable overlapping of designations, so a familiarity with one system willpresent at least a working acquaintance with the other.

Some approximate equivalents that will include most but not all soils are asfollows:

A-1-a or GWWell-graded free-draining gravel suitable for road bases or foundation support.

A-1-b or SWSimilar to A-1-a except that it is primarily sand.

A-2 or SM or SCSand with appreciable fines content. May be moderately frost-susceptible.

A-3 or SPSand that is mainly one size.

A-4 or MLSilt that combines capillarity and permeability so that it is susceptible to frostheave. Low-density eolian deposits often collapse when wet.




A-5 or Low-Plasticity MHIncludes micaceous silts that are difficult to compact.

A-6 or CLModerately plastic clay that has a moderate susceptibility to frost heave and islikely to be moderately expansive. All A-6 is CL but not all CL is A-6.

A-7-5 or Most MHSilty clay soils with a high liquid limit, often from a high mica content.

A-7-6 or CHHighly plastic clay that is likely to be expansive. Low permeability reduces frostheave. All A-7-6 soils also classify as CH.

A-8 and PtPeat and muck.

12.14 OTHER DESCRIPTIVE LIMITS

Other tests and descriptive terms have been devised or defined that are not aswidely used or have fallen into disuse. Some are as follows:

12.14.1 Toughness

Toughness is defined as the flow index from the liquid limit test, which is thechange in moisture content required to change the blow count by a factor of 10,divided by the plasticity index.

12.14.2 Shrinkage Limit

The shrinkage limit test was suggested by Atterberg and has been used as acriterion for identifying expansive clay soils. However, the test involves completedestruction of the soil structure and drying from a wet mud, which makescorrelations less reliable. The shrinkage limit generally is lower than the plasticlimit, and the transition from the intermediate semisolid state to a solid isaccompanied by a noticeably lighter shade of color due to the entry of air.

The shrinkage limit test also fell into disfavor because it used a mercurydisplacement method to measure the volume of the dried soil pat. An alternativemethod now coats the soil pat with wax for immersion in water (ASTM D-4943).

In order to perform a shrinkage limit test a soil-water mixture is prepared as forthe liquid limit but with a moisture content that is considerably above the liquidlimit, and the moisture content is measured. A sample is placed in a shallow dish




that is lightly greased on the inside and struck off even with the top of the dish,which has a known weight and volume. The soil then is oven-dried at 1108C andthe weight recorded. The soil pat then is removed and suspended by a thread inmelted wax, drained and allowed to cool, and re-weighed.

During oven-drying the volumetric shrinkage equals the volume of water lost untilthe soil grains come into contact, which is defined as the shrinkage limit. Then,

SL ¼ w�V � Vd

ms

� �� 100 ð12:7Þ

where w and V are the soil moisture content and volume prior to drying, Vd is thevolume of the pat after oven-drying, and ms is the mass of the dry soil in grams.The determination assumes that the density of water that is lost during drying is1.0 g/cm3.

A so-called ‘‘shrinkage ratio’’ equals the dry density of the soil at the shrinkagelimit:

SR ¼ ms=Vd ð12:8Þ

where SR is the shrinkage ratio and other symbols are as indicated above.

12.14.3 COLE

The ‘‘coefficient of linear extensibility’’ (COLE) test is used by soil scientists tocharacterize soil expandability, and has an advantage over the shrinkage limit testin that the original soil structure is retained, which as previously discussed cangreatly reduce the amount of soil expandability. No external surcharge load isapplied. A soil clod is coated with plastic that acts as a waterproof membrane butis permeable to water vapor. The clod then is subjected to a standardized moisturetension of 1/3 bar, and after equilibration its volume is determined by weighingwhen immersed in water. The volume measurement then is repeated after oven-drying, and the volume change is reduced to a linear measurement by taking thecube root:

COLE ¼ 3p Vm=Vdð Þ � 1 ð12:9Þ

Where Vm is the volume moist and Vd is the volume dry. Volumes are obtainedfrom the reduction in weight when submerged in water, which equals the weight ofthe water displaced. For example, if the reduction in weight is 100 g (weight), thevolume is 100 cm3. A COLE of 53 percent is considered low, 3 to 6 percentmoderate, and46 percent high for residential construction (Hallberg, 1977).

12.14.4 Slaking

Shale may be subjected to a slaking test that involves measuring the weight lossafter wetting and tumbling in a rotating drum (ASTM D-4644). Dry clods of soilalso may slake when immersed in water as the adsorptive power may be so great




that air in the pores is trapped and compressed by water entering the capillaries,causing the soil clod literally to explode and disintegrate. The same soil will notslake when saturated. Slaking therefore can provide an immediate clue that a soilhas been compacted too dry, discussed in the next chapter.

12.15 SUMMARY

This chapter describes laboratory tests relating the plastic behavior to moisturecontent, which form the basis for engineering classifications. Two classificationmethods are presented, one that is more commonly used in highway soilengineering and the other in foundation engineering. Soils may be classified byeither or both methods as part of a laboratory testing program. Classification isuseful for determining appropriate uses of soils for different applications, but isnot a substitute for engineering behavioral tests.

Results of classification tests can be influenced by air-drying, so soil samplespreferably are not air-dried prior to testing. If they are air-dried, considerablemixing and aging are required to ensure complete hydration of the clay mineralsprior to testing. As soils used in classification tests are remolded, the results arenot directly applicable to most field situations, exceptions being soils that arebeing remolded in the base of active landslides, in mudflows, and soils that havebeen liquefied by vibrations such as earthquakes. Classification therefore is morecommonly a diagnostic than a performance tool.

Problems

12.1. Define liquid limit, plastic limit, plasticity index, and activity index.

12.2. Four trials in a liquid limit test give the following data. Plot a flow curveand determine the liquid limit.

Number of blows Moisture content, %

45 29

31 35

21 41

14 48

12.3. If the plastic limit of the soil in Problem 12.2 is 13%, what is the plasticityindex?

12.4. If the soil in Problem 12.2 contains 30% 2 mm clay, what is the activityindex?

12.5. The liquid limit of a soil is 59%, the plastic limit is 23, and the naturalmoisture content is 46%. What is the liquidity index? What is itssignificance?




12.6. The liquid limit of a soil is 69% and its natural moisture content is 73%.

Is this soil stable, metastable, or unstable? What is the dictionary

definition of metastable?

12.7. Define shrinkage limit and shrinkage ratio.

12.8. The volume of the dish used in a shrinkage limit test is measured and

found to be 20.0 cm3, and the volume of the oven-dry soil pat is 14.4 cm3.

The weights of the wet and dry soil are 41.0 and 30.5 g, respectively.

Calculate the shrinkage limit and shrinkage ratio.

12.9. Describe a nonplastic soil and explain how this characteristic is

determined in the laboratory.

12.10. Distinguish clearly between a nonplastic soil and one that has a PI equal

to zero.

12.11. Can you think of a reason why a fine-grained binder soil should be close

to or below the plastic limit when it is added to a coarse-grained soil to

form a stabilized soil mixture?

12.12. If the PI of a stabilized soil pavement is too high, what adverse

characteristics are likely to develop under service conditions? What may

happen if the PI is too low?

12.13. Is soil containing water in excess of the liquid limit necessarily a liquid?

Explain.

12.14. A soil clod coated with a semipermeable plastic membrane and

equilibrated at 1/3 bar moisture tension weighs 210 g in air and 48 g

submerged in water. After oven-drying, the corresponding weights

are 178 g and 47 g. (a) What is the COLE? (b) Rate the expansive

potential of this soil. (c) If you have no choice but to put a light

slab-in-grade structure on the soil, what precautions might be taken

to prevent damage?

12.15. What is the significance of the group index in connection with the

AASHTO system of classification?

12.16. State the broad general character of soils included in groups A-1,

A-2, and A-3 of the AASHTO system and give approximate equiv-

alents in the Unified Classification system. What are the specific

differences?

12.17. What are the principal differences between two soils classified as A-4

and A-5 in the AASHTO system?

12.18. What are the approximate Unified Classification equivalents of AASHTO

groups A-4, A-5, A-6, A-7, and A-8? Which pairs are most nearly

identical?

12.19. Give the major characteristics of soils included in (a) the GW, GC, GP,

and GF groups of the Unified Classification system; (b) the SW, SC, SP,

and SF groups; (c) the ML, CL, and OL groups; (d) the MN, CH, and OH

groups.




12.20. Give four examples of borderline classifications in the Unified Classifica-tion system and explain what each means.The following problems are with reference to data in Table 7.5 of Chapter 7.

12.21. Classify soils No. 1, 2, and 3 in Table 7.5 according to the AASHTO andUnified systems.

12.22. Classify soils No. 4, 5, and 6 in Table 7.5 according to the AASHTOand Unified systems.

12.23. Classify soils No. 7, 8, and 9 in Table 7.5 according to the AASHTO andUnified systems.

12.24. Classify soils No. 10, 11, and 12 in Table 7.5 according to the AASHTOand Unified systems.

12.25. Which soil in each system is most susceptible to frost heave? Whatcharacteristics contribute to this susceptibility?

12.26. Which soil in each system is most expansive? Which is moderatelyexpansive?

12.27. A loess soil changes from A-4 to A-6 to A-7-6 depending on distance fromthe source. Predict the volume change properties including expansion andcollapsibility.

12.28. State the Denisov criterion for loess collapsibity. Does it take into accountthe increase in density with depth?

12.29. Seasonal changes in moisture content of an expansive clay depositextend to a depth of 4m (13 ft). Does that depth coincide with thethickness of the active layer? Why (not)?

12.30. Why classify soils?

References and Further Reading

American Society for Testing and Materials. Annual Book of Standards. ASTM,Philadelphia.

Chen, F.K. (1988). Foundations on Expansive Soils, 2nd ed. Elsievier, Amsterdam.

Grim, R. E. (1968). Clay Mineralogy. McGraw-Hill, New York.Hallberg, G. (1977). ‘‘The Use of COLE Values for Soil Engineering Evaluation.’’ J. Soil

Sci. Soc. Amer. 41(4), 775–777.

Handy, R. L. (1973). ‘‘Collapsible Loess in Iowa.’’ Soil Sci. Soc. Amer. Proc. 37(2),281–284.

Handy, R. L. (2002). ‘‘Geology, Soil Science, and the Other Expansive Clays.’’ Geotechnical

News 20(1), 40–45.Katti, R.K., Katti, D.R., and Katti, A.R. (2005). Primer on Construction in Expansive

Black Cotton Soil Deposits with C.N.S.L. (1970 to 2005). Oxford & IBH PublishingCo., New Delhi.

Skempton, A.W. (1953). ‘‘The Colloidal Activity of Clays.’’ Proc. 3rd Int. Conf. on SoilMech. and Fd. Engg. 1, 57.

U.S. Department of Interior Bureau of Reclamation (1974). Earth Manual, 2nd ed.

U.S. Government Printing Office, Washington, D.C.




0071549226 ar012

Documents

soil behavior

soil samples

wet soil

preparation of soil

natural soil moisture

aswedish soil scientist

liquid limit increases

liquid limits