PRESSURE-INJECTED LIME FOR TREATMENT
OF SWELLING SOILS
Marshall R. Thompson and Quentin L. Robnett,
University of Illinois at Urbana-Champaign
The pressure-injected lime technique for treating swelling soils is de-
scribed and evaluated. Basic mechanisms of soil-lime reactions and
pressure-injected lime are considered, and the effects of treatments with
pressure-injected lime are discussed. Typical field experiences with
pressure-injected lime are summarized, and the factors that appear to in-
fluence the effectiveness of the technique are identified. There are con-
flicting reports concerning the effectiveness of pressure-injected lime
treatment of expansive soils. The condition most favoring the achieve-
ment of successful pressure-injected lime treatment of expansive soils is
the presence of an extensive fissure and crack network into which the lime
slurry can be successfully injected. The proposed treatment mechanisms
(prewetting, development of soil-lime moisture barriers, and effective
swell restraint with the formation of limited quantities of soil-lime reaction
products) have validity. The relative significance of the prewetting and
soil-lime pozzolanic reaction aspects of pressure-injected lime treatment
has not been established. The various statements and reports in the lit-
erature and the information presented in the paper suggest that pressure-
injected lime may not be effective under all circumstances but that in ap-
propriate conditions it can be satisfactorily and economically used. It is
indicated that appropriate guidelines and principles should be developed for
evaluating (on a site-by-site basis) the potential effectiveness of pressure-
injected lime treatment.
•EXPANSIVE clays in the design and construction of transportation facilities present
a major problem. Many techniques (compaction control, prewetting, heating, and
various additive stabilization procedures) have been used for controlling volume
changes in expansive soils (1).
Pressure injection of lime-water slurry into the expansive soil deposit is a recently
developed procedure that is attracting considerable attention. The purpose of this
paper is to describe and evaluate the pressure-injected lime (PIL) technique based on
information currently available.
BACKGROUND
Lime has been widely and successfully used as a stabilizing agent for fine-grained
soils. When lime is added to a fine-grained soil and intimately mixed, several
reactions are initiated. Cation exchange and agglomeration-flocculation reactions
take place rapidly and produce immediate changes in soil plasticity, workability,
and swell properties. Plasticity and swell are reduced, and workability is substan-
tially improved because of the low plasticity and the friable nature of the mixture. A
soil-lime pozzolanic reaction may commence depending on the characteristics of the
soil being stabilized. The pozzolanic reaction results in the formation of various types
of hydrated calcium silicate and calcium aluminate cementing agents or both. The
cementing agents increase mixture strength and durability. Pozzolanic reactions are
time dependent, and strength development is gradual but continuous for a long period
24
25
of time (several years in some instances).
Extensive studies (2) have shown that practically all fine-grained soils react with
lime (cation exchange;- agglomeration-flocculation) to effect beneficial changes in work-
ability, plasticity, and swell properties. The extent to which the lime-soil pozzolanic
reaction proceeds is influenced primarily by natural soil properties (3). With some
soils, the pozzolanic reaction is inhibited, and cementing agents are not extensively
formed.
Thompson (3) has ter med th ose soils that react with lime to produce s ubstantial
strength increase as r eactive and those that display limited pozzolanic reactivity as
nonreactive. Properties of compacted and cured soil-lime mixtures will be different
for reactive and nonreactive soils.
For the nonreactive soils, plasticity, workability, and swell properties are altered,
but strength increases are nominal. Reactive soils initially experience similar plas-
ticity, workability, and swell changes and will attain substantial strength because of the
pozzolanic reaction.
In the treatment of subgrade soils with high swell potential, the major objective of
lime stabilization is swell control. If the soil and lime can be intimately mixed, - it is
highly probable that the soil swell potential can be drastically reduce!}. Holtz (4) sum-
marized the beneficial effects of lime treatment relative to the control of volume change
in expansive clay soils. Not all soils respond favorably to lime treatment as a control
of swell potential. Plummer (5) reported that certain of the Red River, North Dakota,
clays will expand around 10 percent even after lime treatment. Mitchell (1) indicated
that lime was the most effective additive for stabilization. -
Even though the efficacy of lime for swell control treatment has been conclusively
demonstrated, the reality of field conditions has limited the use of conventional lime
stabilization. Holtz (4) has suggested that for effective swell control the expansive
soil should be stabilized to a depth of approximately 5 ft (1.5 m). The costs associated
with using conventional soil-lime stabilization procedures for treating the subgrade of
a typical roadway section to a 5-ft (1.5-m) depth are substantial and in many instances
prohibitive.
Thus, other techniques such as drill-hole lime and pressure-injected lime have
been devised in an attempt to lime stabilize, clays in their in situ state.
DRILL-HOLE LIME
The drill-hole lime technique basically consists of introducing quicklime or hydrated
lime into a soil mass by placing the lime in holes drilled in the soil mass. When placed
in the holes,· the lime (usually hydrated lime in a slurry form) migrates or diffuses into
the soil system thereby initiating the soil-lime reactions. If sufficient migration or
diffusion occurs, it is possible that the properties of a sizable quantity of the soil mass
around the drill hole will be improved. However, lime migration or diffusion is a very
slow process, and substantial time may be required before a substantial quantity of
soil is affected.
In highway applications, small-diameter [6 to 12 in. (15 to 31 cm)] holes are ad-
vanced through the pavement into the-subgrade soil by using a suitable apparatus such
as a power post-hole digger equipped with a continuous flight auger. In highway ap-
plications, provision must be made to construct a hole in the pavement structure to
gain access to the subgrade. Typica,lly, the depths of the drill holes range from 30
to 50 in. (76 to 127 cm). The exact hole depth depends largely on the depth and nature
of material to be treated. Typical hole spacings are about 4 to 5 ft (1.2 to 1.5 m)
center-to-center.
After the hole has been made, it is partially filled with either quicklime or hydrated
lime. In some cases, water is added to the lime to create a slurry, or a lime-water
slurry is placed directly in the drill hole. However, dry lime (especially quicklime)
is thought to act as a drying agent that absorbs soil moisture, thereby reducing the
moisture content of the surrounding soil. The use of a lime-water slurry may, how-
ever, tend to increase the mobility of the lime since water acts as a medium for
migration.
26
Backfilling the hole and patching the pavement are normally required. Both soil
(from the drill hole) and aggregate have been used. The backfill should be tamped
into the hole. The holes in the pavement may be patched with portland cement con-
crete or asphaltic concrete.
Several studies (6, 7, 8, 9) have considered the drill-hole procedure. Although in
many instances the mafor-stabilization objective was not swell control, the background
information developed in the studies is still applicable.
In general, drill-hole lime treatment results have been erratic. Some report suc-
cess, but others indicate that little or no improvement has been achieved. The various
investigations have shown that the zone of influence in which soil-lime reactions have
taken place is limited to the areas immediately adjacent to the drill hole.
It is apparent that the major factor limiting the effectiveness of the drill-hole lime
procedure is the inability to achieve lime distribution throughout the soil mass.
PRESSURE-INJECTED LIME
In an attempt to achieve better lime distribution in the soil mass, the PIL procedure
was developed. In this procedure, a lime-water slurry is pumped under pressure
through hollow injection rods into the soil. Generally, the injection rods are pushed
into the soil in about 12-in. (31-cm) intervals. At each depth, the lime slurry is in-
jected to refusal. Refusal occurs when
1. Soil will not take additional slurry,
2. Slurry is running freely on the surface either around the injection pipe or out of
previous injection holes, or
3. Injection has fractured or distorted the pavement surface.
Although there is substantial variability in the amow1t of slurry that can be injected,
a normal take is about 10 gal/ft (124 liters/m) of injected depth. Obviously, the nature
of the soil being treated will influence the quantity of slurry that can be injected.
The normal lime-water slurry composition is 2% to 3 lb of lime/gal (0.3 to 0.4 kg
of lime/ lite1·) of water with a wetting age,1t added in accordance with the manufacturer's
recommendation. Based on extensive field experience, the above slurry composition
ha.s proved to be satisfactory.
Although injection pressures as high as several hundred pounds per inch2 (pascals)
can be developed with most lime slurry injection equipment, the maj<?rity of the work
is injected in the pressure range of 50 to 200 psi (345 to 1380 kPa). In this pressure
range, it is normally possible to disperse the maximum amount of slurry into the soil.
Spacings of 3 to 5 ft (0.9 to 1.5 m) on centers are common in pressure injection
treatment for building foundation work. Spacings of 4 ft (1.2 m) were used in the deep-
layer stabilization flexible pavement test sections at Altus Air Force Base, Oklahoma.
Spacings of 5 ft (1.5 m) are also typical for PIL treatment of railroad subgrades. Var-
ious pressure injection contractors in the Dallas-Fort Worth, Texas, area indicated
that the amount of lime slurry that could be injected per unit volume of soil was in-
dependent of injection probe spacing within the range of 3 to 6 ft (0.9 to 1.8 m).
Injection depths are variable, but current equipment is capable of injecting to depths
of approximately 10 ft (3 m). Wright (10) has indicated that a treatment depth of 7 ft
(2.1 m) is normally sufficient for foundation treatments. This depth compares reason-
ably well with the 5-ft (1.5-m) depth suggested by Holtz (4). The general guide is to
inject to a depth sufficient to be below the zone of critical moisture change in the ex-
pansive soil deposit.
If the surface of the FIL-treated soil deposit is exposed, it is common practice to
mix the free surface lime available into the soil to a depth of 6 to 8 in. (15 to 20 cm).
The stabilized layer further contributes to the process of retarding moisture loss from
the underlying soil. Teng, Mattox, and Clisby (11) have shown the effectiveness of a
lime-treated Yazoo clay layer to act as an effective capillary barrier for preventing
desiccation. Similar results have been found in a wide variety of soil-lime stabiliza-
tion applications.
27
Field studies (12, 13) in which PIL-treated soils have been excavated show that the
PIL slurry is forced along fracture zones, cracks, fissures, bedding planes, root lines,
coarse-textured seams in varved clays, seams and fractures effected by the pressure
slurry injection process, or other passages in the soil mass. The field observations
and a recent laboratory investigation (14) have conclusively demonstrated that the lime-
water slurry will not permeate an intact fine-grained soil mass.
Hillel (_!2) states,
The hydraulic conductivity (of the soil mass) is obviously affected by structure as well as by
texture, being greater if the soil is highly porous, fractured, or aggregated than if it is tightly com-
pacted and dense. The conductivity depends not only on the total porosity, but also, and pri-
marily, on the sizes of the conducting pores.
Lytton (16) in characterizing the geomorphological aspects of expansive clay indi-
cated that in gilgai land forms the "soil is fractured to great depths." In the same
paper, Lytton also discusses the crack pattern formation process in expansive soils.
The fact that expansive soil deposits are typically cracked or fractured in the near sur-
face depths that are of concern in swell control procedures is helpful when the problem
of trying to pressure-inject lime-water slurry into the soil mass is considered.
It is apparent that the presence of openings in the soil mass is requisite for obtain-
ing adequate slurry distribution. Wright (10) has described the final lime distribution
pattern frequently obtained in swelling clays as "a network of horizontal, sheet lime
seams, often interconnected with vertical or angular veins."
TREATMENT MECHANISMS
From the previously presented information, it is apparent that there are two major
treatment mechanisms of concern relative to PIL. The first is the ability to perme-
ate the soil mass with the stabilizing additive (in this case a lime-water slurry), and
the second is the process whereby, following PIL treatment, the lime translocates
and modifies the soil adjacent to the lime seams.
Lime Injection
When the basic theory of permeability is combined with Darcy's law of fluid flow in a
soil mass, the total quantity of fluid that can be forced into a soil mass during a given
interval of time can be approximated by equation 1:
Q= p; (t) KAt (1)
where
A= cross-sectional area over which pressure acts;
ri = viscosity of fluid;
g = acceleration due to gravity;
p = pressure head;
K = Cd2 , intrinsic permeability of medium, where C is a shape factor and dis
average pore size of mdeium;
.i = length over which pressure head acts;
Q =quantity of fluid flow;
p = density of fluid; and
t = time of pressure application.
28
Equation 1 indicates that the following major factors control the quantity of fluid
that is injected: (a) fluid viscosity, (b) injection pressure and time, and (c) intrinsic
permeability of the soil medium.
Since lime slurry is not an ideal fluid but rather a particulate suspension, the pore
size distribution of the soil mass is an important consideration in the permeation pro-
cess. Successful injection of the lime slurry into the soil mass would require that
channels larger than the lime particles be present.
The inherent pore size of most fine-grained soils is quite small relative to the lime
particle size. Thus, appreciable lime slurry movement through these pores is ques-
tionable.
Johnson (17) recommends that the groutability ratio, as calculated by using equation
2, be greaterthan 20 to 25 for successful cement grouting.
Groutability ratio = gcisJ soil t
<ssJ grou
(2)
where
D<1sJ soil =particle size for which 15 percent of the soil fraction is finer, and
D<as) grout =particle size for which 85 percent of the cement grout is finer.
Many commercial hydrated lime specifications for soil stabilization purposes re-
quire a minimum of 85 percent passing the No. 200 sieve. Using 0.0029 in. (0.074 mm)
as D<asl in Johnson's equation indicates the D<isl for the soil must be approximately
0.059 in. {1.5 mm) or larger to meet a groutability ratio criterion of 20. It is obvious
that for fine-grained expansive soils the lime slurry cannot be effectively forced
through the soil pore system.
Laboratory studies conducted at the University of Illinois (14) indicate that it is
almost impossible to force a typical lime slurry (30 percent byweight) into fine-
grained soils even when pressures of up to 1,000 psi (6.9 MPa) are applied for 20 min.
Typical results from this limited study are shown in Figure 1. In general, slurry
penetrations averaged less than 1/; in. (12. 7 mm) into the silty materials, and almost
no penetration was achieved in the clayey materials.
Teng, Mattox, and Clisby (11) found in their water flooding studies of a remolded
and compacted Yazoo clay embankment that little moisture penetration was achieved.
The flooding of the undisturbed Yazoo clay in the cut sections was successful. These
findings caused them to conclude, "The fissure system plays a very significant role
in allowing water intrusion into the soil mass thus causing swell.'' The significance
of the fissure system would be equally important for the PIL procedure. It is im-
portant to note that the fissure structure in the Yazoo clay study (11) was present in
the in situ cut sections, but the fissure structure was destroyed int he embankment
construction process.
It is apparent that to successfully pressure-inject lime slurry into a fine-grained
soil natural channels and passages larger than the lime particles must be present in
the soil mass. Such channels may be present as a result of (a) inherent pore structure
of the soil mass; {b) cracks, fissures, seams, and root holes present in the soils; or
(c) jetting or tearing of the soil effected by the pressure injection process.
Even though the permeability of the soil resulting from the inherent soil pore struc-
ture may be quite low, the mass permeability or conductivity may be substantially
higher as a result of the presence of seams, fissures, cracks, varves, and so on.
When this condition exists, the potential for successful lime slurry injection of a soil
mass is greatly enhanced.
29
Diffusion-Migration of Lime
Based on the preceding discussion, it is apparent that
1. Intimate permeation of fine-grained soils through PIL is virtually impossible,
and
2. Lime slurry can be pressure-injected into certain fine-grained soil masses if
varves, seams, fissures, and cracks exist but the distribution of the slurry is strati-
fied or of a network type.
However, the lime will tend to be translocated in the fine-grained soil mass as time
progresses because of diffusion and migration phenomena.
Space limitations do not allow presentation and discussion of diffusion-migration
theory. In general, however, factors such as differences in clay content, clay min-
erals, density, absorbed cations, and temperature have been found to affect the rate
of diffusion (18, 19, 20, 21).
In an early study ;Davidson, Demirel, and Handy (22) suggested that the diffusion
of calcium cations in a soil-lime water system is an example of the diffusion phenomena.
The processes accompanying lime diffusion may include (a) transfer of lime into the
soil, (b) chemical reaction between the lime and the soil, (c) formation of nuclei and
growth of the reaction product, and (ct) further diffusion of the lime into the soil from
the reaction product layer.
The following equation has been suggested for determining the rate of growth of a
product layer from the lime source (22, 23):
(3)
where
.(. = dis tance of lime m igration fo r a time t in inches (millimeter s),
kd =diffu sion c onst ant in inc hes/ day112 [ reported values range fr om 0.081 to 0.63
in. / day 1! 2 (2 to 16 mm/ day1/ 2 ) (22 , 23 )], and
t = elapsed time of diffusion in dayS. -
Limited laboratory and field studies have been conducted to evaluate the rate and
extent of lime migration. In a controlled laboratory study conducted by Fohs and
Kinter (24), about 0 .8 per cent lime was found to migrate approximately 1% to 2 in.
(3.8 to 5cm) after 180 days. They concluded that the migration process for effecting
translocation of lime in a soil system is very slow and that only very small amounts
of lime can be translocated. This makes this process impractical for effecting sub-
stantial soil mass strength increases (24). Robnett, Jamison, and Thompson (14) con-
ducted a limited laboratory lime migration study. Typical results are shown illFigure
2. It is apparent that the amount of lime translocated by the migration process is
small.
In a field s tudy, Lundy, Jr. , and Greenfield (13) found that after 1 year approx-
imately 3 4 to 11/:i in. (19 to 38 mm) of lime migration had occurred away from the lime
seams. A Louisiana Department of Highways study (12) found that about % to 11/z in.
(13 to 38 mm) of lime migration occurred after 4 years.
It is evident that lime translocation by the diffusion-migration process is very slow.
If equation 3 is used to estimate the required time for various distances of migration,
the following values are found, if kd is assume d to be 0.10 in. / day 1l 2 (2.5 mm/ day 1l 2 )
(1 in . = 25.4 mm):
30
Figure 1. Typical lime-water slurry Figure 2. Typical lime migration data
pressure penetration data for Fayette C for Altus subgrade soil [AASHO
soil [AASHO A 4(8)]. A6(13)].
Equivalent Percent Lime Equivalent Percent Lime
0.0 OJ2' Q,2, O.!K> a.a 0.12' 0.2, o.~
pH of Mi•lure
1. 0 o.o
1000 psi, 20 min ~
.!:
.!:
L'
~
.
Q.
0
ii.
• 2.0
0
120 Day
>.O •.o
•.0-----------------
l (in.) t (days)
1 100
6 3,600
12 14,600
Recent comprehensive studies by Stocker (25) have led to the development of an in-
tegrated theory of soil-lime stabilization reactions termed diffusion and diffuse ce-
mentation. Diffuse cementation theory as proposed by Stocker describes a process in
which lime (from a FIL-deposited lime seam in the situation of interest) will diffuse
into a natural soil lump. Based on his studies, Stocker stated,
The diffused lime is shown to react with all the clay present, including that within unpulverized
lumps, leading first to volume-stabilization (against wetting and drying) and increase in soaked
strength, and later to remarkable increases in. even unsoaked strength (for relatively high sta-
bilizer contents). This cementation is diffuse.... It [diffuse cementation] is the dominant mech-
anism whereby included lumps of unpulverized soil are made impotent with respect to differ-
ential volume change and finally by which the lumps are increased in mechanical strength.
Stocker indicated that, in a lime-covered, soil-lump system in which diffuse cemen-
tation is occurring, the earliest physical property change effected is the "suppression
of swell or wetting from the as-cured state." His studies further suggest that the de-
velopment of limited cementing material is sufficient to prevent swelling but does not
contribute to the development of substantial compressive or shear strength.
Stocker's diffuse cementation theory (25) suggests that in lime-reactive soils soil-
pozzolanic reaction products may form inregions of low calcium concentration remote
from the lime source. The applicability of the diffuse cementation theory depends on
the soil being lime reactive (soil will react with lime to form calcium silicate and cal-
cium aluminate hydrates).
31
EFFECTS OF PRESSURE-INJECTED LIME
As a consequence of PIL treatment, major changes are effected in the soil (assuming
that there are sufficient soil passageways to facilitate lime slurry distribution in the
soil mass). The soil moisture content is increased, and an initial network of lime
seams is formed in the soil.
It is well established that initial water content has a significant influence on volume
change in expansive soils. In general, the lower the initial water content is, the higher
the swell will be. In some PIL jobs the moisture content of the treated soil is specified.
Wright (10) indicates that a typical final moisture content requirement is 1 to 2 percent
above theplastic limit. It is not uncommon to achieve several inches (millimeters) of
swelling in the deposit following PIL, depending on the moisture content of the expan-
sive soil before injection (10, 26). Mitchell and Raad (1) have considered the applica-
tion of prewetting as a methodof controlling volume change in expansive earth mate-
rials. Many engineers are prone to overlook this important aspect of PIL treatment,
which is in reality a form of prewetting.
Following PIL treatment, soil-lime reactions may occur in the areas adjacent to the
lime seam network. Lime diffusion-migration takes place as previously discussed.
If the soil is reactive, soil-lime cementing products will form. The stabilized zones
serve as moisture barriers. Stocker (25), as previously discussed, indicated that the
formation of limited accumulations of soil-lime reaction products at or near the edges
of the clay particles would effectively restrain moisture-related volume expansion.
Even if only the boundaries of a soil mass have reacted with lime, Stocker (25) suggests
that the nonswelling shell of modified soil may constrain the relatively unaffected core.
Stocker's work with lime treatment of lumps of montmorillonitic soil even suggests that
the swell characteristics of the soil had been substantially modified ahead of the pene-
trating diffusion-reaction front.
It is important to recall that, in a practical field application, the potential for soil
swell is previously reduced by the PIL treatment prewetting effect so that the preserva-
tion of the in situ soil moisture content alone will ensure volume constancy. The re-
straint developed from the formation of reaction products is an added beneficial effect.
It was noted by Stocker that higher soil moisture contents increase the rate of develop-
ment of diffuse cementation. Thus the water content increase effected by PIL will en-
hance the diffusion cementation process.
The preceding discussion indicates that there are several potentially beneficial pro-
cesses that can occur in a FIL-treated expansive soil deposit. Under the proper cir-
cumstances, PIL treatment should serve as an effective swell control procedure. At
this time it is not possible to accurately define what conditions are required to ensure
a high probability of success.
EVALUATION OF PRESSURE-INJECTED LIME
PIL has gained acceptance as a possible procedure for treating expansive soils in
building foundation applications, but has not yet achieved the same degree of acceptance
in transportation facility construction. PIL was not considered in any depth at the re-
cent Denver conference, but some comments were made concerning lime treatment
techniques.
1. Kelley and Kelly (27) referred to the successful use of PIL to.treat the swelling
soils under the Dallas-Fort Worth Regional Airport terminal buildings.
2. Krazynski (28) indicated that he did not think PIL treatments were very effective.
3. Teng, Mattox, and Clisby (11) thought that research and expel'ience have shown
that lime stabilization by pressure injection has not lessened the swell potential.
4. Blacklock (26), a PIL contractor, indicated that, based on results, PIL was
definitely a solution to the swelling problem.
5. Gerhardt (29) reported on the successful use of drill-hole lime stabilization in
a Colorado test section and concluded that potential swell can be reduced to almost
32
nothing for any depth.
6. Brakey (30) suggested that the drill-hole lime procedures were not effective in
providing protection against expansion in the Mancos shale.
Ingles and Neil (31) have evaluated the effects of PIL treatment for typical Austra-
lian soil conditions:-Their studies indicated that PIL treatment of expansive soils was
an effective procedure, but only when the treatment was carried out at the time of max-
imum desiccation when the soil is fissured.
Robnett, Jamison, and Thompson in recent studies (32) of deep-layer stabilization
procedures for pavement systems also considered the various aspects of PIL. Of the
various procedures evaluated, the PIL procedure was determined to be one of the more
promising methods.
Wright (10) cited the rapid growth of PIL building foundation treatment in Fort Worth,
Texas, and indicated that the use of the procedure was spreading to other parts of Texas
as well as Arkansas, Tennessee, Louisiana, and Oklahoma. It is apparent that PIL is
an accepted expansive soils treatment procedure in some geographic areas.
The various statements and reports in the literature suggest that PIL may not be
effective under all circumstances, but under appropriate conditions it can be satis-
factorily and economically used. It appears that higher probabilities of success are
achieved when the job conditions permit the satisfactory injection of lime slurry and
the development of a comprehensive network of lime seams throughout the soil mass.
The presence of an extensive fissure and crack system in the soil seems to be neces-
sary.
If the lime slurry can be successfully injected throughout the fissures or cracks in
the soil mass, the prewetting phase of the PIL treatment can then occur. The soil-
lime pozzolanic reaction will also commence (reaction rate is influenced by time and
temperature) if the soil is reactive. It is thus possible that in some PIL applications
only the prewetting effect is achieved and that the soil-lime pozzolanic reaction does
not occur. In such a case, the moisture barrier developed in the areas of the lime
seams may not be as effective because any soil property change will be due to cation
exchange and perhaps limited changes in soil structure but will not show the benefit
of soil-lime reaction product formation. Whether or not the soil-lime pozzolanic re-
action is essential for successful long-term PIL treatment has not been established.
However, Stocker's theory (25) indicates that a lime-reactive soil is essential if ef-
fective diffuse cementation atpoints remote from the lime source is to be achieved.
Fortunately, many expansive soils are lime reactive.
SUMMARY
There are conflicting reports concerning the effectiveness of PIL treatment of ex-
pansive soils. The proposed treatment mechanisms (prewetting, the development of
soil-lime moisture barriers, and effective swell restraint with the formation of limited
quantities of soil-lime reaction products) have validity. It therefore seems logical to
conclude the PIL may be an effective swell control procedure under certain circum-
stances. The condition most favoring the achievement of successful PIL treatment of
expansive soils is the presence of an extensive fissure and crack network into which
the lime slurry can be successfully injected. The relative significance of the pre-
wetting and soil-lime pozzolanic reaction aspects of PIL treatment has not been es-
tablished. If soil-lime pozzolanic reactions are essential to achieving an effective ap-
plication, perhaps that fact can be used to evaluate the potential success of an antici-
pated treatment and explain the apparent conflicting reports on PIL experience.
It is suggested that future research and development activities focus on the following
areas:
1. The relative significance of prewetting and soil-lime pozzolanic reactions in PIL
treatments,
2. Consideration of Stocker's diffusion cementation theory and its application to
33
PIL treatment, and
3. Development of appropriate guidelines and principles for evaluating (on a site-
by-site basis) the potential effectiveness of PIL treatment.
REFERENCES
1. J. K. Mitchell and L. Raad. Control of Volume Changes in Expansive Earth
Materials. Proc., Workshop on Expansive Clays and Shales in Highway Design
and Construction, Federal Highway Administration, Vol. 2, May 1973.
2. M. R. Thompson. Factors Influencing the Plasticity and Strength of Lime-Soil
Mixtures. Engineering Experiment Station, Univ. of Illinois at Urbana-Champaign,
Bulletin 492, 1967.
3. M. R. Thompson. Lime Reactivity of Illinois Soils. Journal, Soil Mechanics and
Foundations Division, ASCE, Vol. 92, No. SM5, 1966.
4. W. G. Holtz. Volume Change in Expansive Clay Soils and Control by Lime Treat-
ment. Proc., 2nd International Research and Engineering Conference on Expan-
sive Clay Soils, Texas A&M Univ., 1969.
5. D. D. Plummer. Swelling Soils in North Dakota. Proc., Workshop on Expansive
Clays and Shales in Highway Design and Construction, Federal Highway Adminis-
tration, Vol. 2, May 1973.
6. Subgrade Improved With Drill-Lime Stabilization. Rural and Urban Roads, Oct.
1963.
7. R. C. Perez. Drill-Lime Stabilization. Materials Research Division, Puerto
Rico Department of Public Works, Research Study 5, July 1966.
8. C. M. Higgins. Lime Treatment at Depth. Louisiana Department of Highways,
Final Rept., Research Rept. 41, June 1969.
9. Lime and Lime-Fly Ash Soil Stabilization. Alabama Department of Highways,
Highway Research HPR Rept. 26, June 1967.
10. P. J. Wright. Lime Slurry Pressure Injection Tames Expansive Clays. Civil
Engineering, ASCE, Oct. 1973.
11. T. C. P. Teng, R. M. Mattox, and M. B. Clisby. Mississippi's Experimental
Work on Active Clays. Proc., Workshop on Expansive Clays and Shales in High-
way Design and Construction, Federal Highway Administration, Vol. 2, May
1973.
12. C. M. Higgins. High Pressure Lime Injection. Louisiana Department of High-
ways, Research Rept. 17, Interim Rept. 2, Aug. 1965.
13. H. L. Lundy, Jr., and B. J. Greenfield. Evaluation of Deep In-Situ Soil Stabili-
zation by High-Pressure Lime-Slurry Injection. Highway Research Record 235,
1968, pp. 27-36.
14. Q. L. Robnett, G. F. Jamison, and M. R. Thompson. Stabilization of Deep Soil
Layers. Air Force Weapons Laboratory, Kirtland Air Force Base, New Mexico,
Technical Rept. AFWL-TR-71-90, Jan. 1972.
15. D. Hillel. Soil and Water. Academic Press, Inc., New York, 1971.
16. R. L. Lytton. Expansive Clay Roughness in the Highway Design System. Proc.,
Workshop on Expansive Clays and Shales in Highway Design and Construction,
Federal Highway Administration, Vol. 2, May 1973.
17. S. J. Johnson. Grouting With Clay-Cement Grouts. Journal, Soil Mechanics
and Foundations Division, ASCE, Vol. 84, No. SMl, Feb. 1958.
18. T. M. Lai and M. M. Mortland. Self-Diffusion of Exchangeable Cations in
Bentonite. Proc., 9th National Conference on Clays and Clay Minerals, 1960,
Vol. 9, pp. 229-247.
19. N. Jost. Diffusion in Solids, Liquids, Gases. Academic Press, Inc., New York,
3rd Ed., 1960.
20. O. A. Talme. Clay Sensitivity and Chemical Stabilization. Department of Quater-
nary Research, Univ. of Stockholm, Rept. 56, 1968.
21. H. Jenny and R. Overstreet. Surface Migration of Ions and Contact Exchange.
Journal of Physical Chemistry, Vol. 43, 1939, pp. 1185-1196.
34
22. L. K. Davidson, T. Demirel, and R. L. Handy. Soil Pulverization and Lime
Migration in Soil-Lime Stabilization. Highway Research Record 92, 1965, pp.
103-126.
23. R. L. Handy and W. W. Williams. Chemical Stabilization of an Active Land-
slide. Civil Engineering, ASCE, Aug. 1967.
24. D. G. Fobs and E. P. Kinter. Migration of Lime in Compacted Soil. Public
Roads, Vol. 37, No. 1, June 1972.
25. P. T. Stocker. Diffusion and Diffuse Cementation in Lime and Cement Stabilized
Clayey Soils. Australian Road Research Board, Special Rept. 8, 1972.
26. J. R. Blacklock. Discussion Comments on Design Against Detrimental Volume
Change. Proc., Workshop on Expansive Clays and Shales in Highway Design and
Construction, Federal Highway Administration, Vol. 2, May 1973.
27. C. Kelley and J. E. Kelly. Uses of Hydrated Lime. Proc., Workshop on Ex-
pansive Clays and Shales in Highway Design and Construction, Federal Highway
Administration, Vol. 2, May 1973.
28. L. M. Krazynski. Discussion Comments on Design Against Detrimental Volume
Change. Proc., Workshop on Expansive Clays and Shales in Highway Design and
Construction, Federal Highway Administration, Vol. 2, May 1973.
29. B. B. Gerhardt. Soil Modification Highway Projects in Colorado. Proc., Work-
shop on Expansive Clays and Shales in Highway Design and Construction, Federal
Highway Administration, Vol. 2, May 1973. ·
30. B. A. Brakey. Moisture Stabilization by Membranes, Encapsulation and Full
Depth Paving. Proc., Workshop on Expansive Clays and Shales in Highway De-
sign and Construction, Federal Highway Administration, Vol. 2, May 1973.
31. O. G. Ingles and R. C. Meil. Lime Grout Penetration and Associated Moisture
Movements in Soil. Division of Applied Geomechanics, Commonwealth Scientific
and Industrial Research Organization, Australia, Research Paper 138, 1970.
32. Q. L. Robnett, G. F. Jamison, and M. R. Thompson. Technical Data Base for
Stabilization of Deep Soil Layers. Air Force Weapons Laboratory, Kirtland Air
Force Base, New Mexico, Technical Rept. AFWL-TR-70-84, April 1971.
