Technical Note

Improving Mechanical Properties of Sand Using Biopolymers

Hamid Reza Khatami1 and Brendan C. O’Kelly, Ph.D.2

Abstract: Natural polymers (biopolymers) are discussed as environmentally friendly and sustainable grouting chemicals. This paper presents
guidelines for selecting potentially useful biopolymers for strengthening cohesionless soil. Agar and six modified starches were identified for
further study over a range of concentrations (1–4% agar and 0.5–1% starch). Experimental results demonstrated the compatibility of agar and
starch. Depending on the biopolymer concentration, the unconfined compressive strength of the sand treated with agar and starch biopolymers
Downloaded from [Link] by University Of Auckland on 05/16/15. Copyright ASCE. For personal use only; all rights reserved.

ranged from 158 to 487 kPa. Triaxial compression tests over a range of confining pressures also indicated that the biopolymers effectively
increased the cohesion intercept and stiffness of the treated sand. DOI: 10.1061/(ASCE)GT.1943-5606.0000861. © 2013 American Society of
Civil Engineers.
CE Database subject headings: Compressive strength; Soil stabilization; Cohesionless soils; Shear strength; Stiffness; Polymer;
Mechanical properties.
Author keywords: Biopolymer; Compressive strength; Ground improvement; Sand; Shear strength; Stiffness.

Introduction 2% by weight to a silty sand and found that xanthan gum performed
better in lowering permeability. The application of agar, in con-
The deliberate process of incorporating a dispersed phase capable of centrations of typically less than 1%, to stabilize soil temporarily by
binding particles and filling pore voids in the soil matrix can bring coating the soil grains with a biopolymer film, has been described
about an overall improvement in the mechanical properties and (or) by Schneider et al. (1989), Frost (1989), and Sutterer et al. (1996) in
a reduction in the hydraulic conductivity of treated ground. Over the the development of undisturbed sampling procedures.
past decades, numerous types of chemicals (e.g., sodium silicate, The current research investigates the potential of two bio-
epoxy, polyurethane, and acrylamides) have been considered for polymers, namely agar and starch, to enhance the strength of sand.
geotechnical application. Chemical grouts are generally toxic and
hazardous, with a few notable exceptions (e.g., sodium silicate).
This necessitates undertaking comprehensive research to find al- Biopolymer Selection
ternative ecofriendly materials for ground treatment. It appears that
several natural polymers (biopolymers) can be used as substitutes The biopolymer selection was a two-stage process that considered
for the conventional components of grouting materials, especially surface properties and mechanical properties, which are responsible
for temporary works and relatively dry conditions. The widespread for providing adhesion and cohesion, respectively. Cohesion refers
application of biopolymers in this manner would provide a sustain- to the intermolecular attraction by which the elements of a body are
able industry because they are obtained mainly from the nonfood held together. In a solid, resistance to an outward force or a tension
parts of plants being cultivated year after year. Previous research has that operates to prevent separation of the atoms of a solid it is com-
focused mainly on the effect of biopolymers on soil permeability. monly known as cohesion. Adhesion, on the other hand, is defined as
Martin et al. (1996) and Karimi (1998) investigated the effects of the physicochemical attraction forces between dissimilar surfaces
biopolymers, specifically xanthan gum and sodium alginate, mainly that make them stick together. Depending on their structure, bio-
on permeability, but also on the shear strength of a silty soil, polymers possess various chemical functional groups, such as hy-
reporting a reduction of two orders of magnitude in permeability and droxyl, ester, or amines. Their long-chain structure also provides
a 30% increase in shear strength. Khachatoorian et al. (2003) showed more sites at which the characteristic chemical reaction of a given
that some biopolymers, when flooded into the matrix of a petroleum functional group can occur. Chemical bonding corresponds to the
reservoir, decrease the permeability relative to the fluids present in adhesive forces whose function is to hold the soil particle and gel
the reservoir. In addition, Bouazza et al. (2009) applied xanthan together at their surfaces. At a microscopic scale, the effectiveness
gum, guar gum, and sodium alginate in concentrations of up to of the bonding depends mainly on the type of forces present at the
interface of the particle and the gel. The forces operating at such a
1
Ph.D. Candidate, Dept. of Civil, Structural and Environmental Engi- phase interface include ionic/electrostatic or covalent bonds (chemi-
neering, Trinity College Dublin, Dublin 2, Ireland (corresponding author). sorption), hydrogen bonding (strong polar attraction), and van der
E-mail: khatamih@[Link] Waals forces (physical absorption). Short-range ionic/electrostatic
2
Associate Professor, Dept. of Civil, Structural and Environmental and covalent bonds have the highest bond energy in terms of KJ/
Engineering, Museum Building, Trinity College Dublin, Dublin 2, Ireland. mol and therefore give the strongest bond. Van der Waals forces,
E-mail: bokelly@[Link]
which are the interaction between dipoles within the bulk material,
Note. This manuscript was submitted on February 18, 2012; approved on
October 31, 2012; published online on November 3, 2012. Discussion develop the weakest bonds over a long range. In selecting a bio-
period open until January 1, 2014; separate discussions must be submitted polymer, a cationic biopolymer together with soil particles would
for individual papers. This technical note is part of the Journal of Geo- provide strong electrostatic bonding throughout the treated ground
technical and Geoenvironmental Engineering, Vol. 139, No. 8, August 1, because most natural particles in soil show a slight negative surface
2013. ©ASCE, ISSN 1090-0241/2013/8-1402–1406/$25.00. charge. Alternatively, a chemical capable of forming hydrogen

1402 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2013

J. Geotech. Geoenviron. Eng. 2013.139:1402-1406.

 bonding with the soil particles (still quite a strong chemical attraction adsorption of biopolymers. The sand has a specific gravity of 2.66,
force) would be used ideally in the case of nonionic biopolymers. a minimum density of 1:39 Mg=m3 , and a maximum density of
In terms of mechanical properties, polymers vary extensively in 1:75 Mg=m3 , with D10 5 0:15 and Cu 5 1:6, where D10 is the
mechanical performance, depending on their degree of crystallinity, equivalent spherical grain size corresponding to 10% finer by
molecular weight, and degree of cross-linking (Feldman 1989). weight, D60 corresponds to 60% finer by weight, and Cu is D60 =D10 .
High strength and rigidity are obtained with a high degree of crys- The biopolymers used in the current study were microbiological-
tallinity and/or cross-linking. A biopolymer of adequately regular grade agar from VWR International and six commercial modified
chemical structure is believed to be capable of undergoing crys- starches (Staramic 105 and 747, along with Starpol 136, 469, 600,
tallization to some extent. Also, the higher its molecular weight, the and 700) from Tate and Lyle. These starches were supplied as a
greater the chance of ordering and crystallization of its macromol- pregelatinized powder, showing at least 20% solubility in cold
ecule chain. The selection of a biopolymer of appropriate molecular water.
weight that is also of a viscosity feasible for injection is usually Complete dissolution was achieved by placing containers with
a matter of trade-off because the apparent viscosity in water usually the biopolymer mixtures in an oven set at a temperature of 95
C for
increases as the molecular weight increases. The flexibility of the a period of 2 h. The containers were sealed to prevent loss of water,
Downloaded from [Link] by University Of Auckland on 05/16/15. Copyright ASCE. For personal use only; all rights reserved.

polymeric chain is also believed to be particularly important for which would otherwise have changed the final concentration of the
permeation of the grout. Once placed within the soil matrix, it is solution.
desirable that the biopolymer undergo some form of cross-linking The dry sand was air pluviated into specimen molds, 38 mm in
to enhance its strength and reduce its mobility in the ground. Cross- diameter by 76 mm long, with the free-fall height of the sand grains
linking connects polymeric chains through chemical reactions (ini- adjusted to produce dry densities of 1:50 to 1:54 Mg=m3 (i.e., rel-
tiated by temperature rise, change in pressure, or pH) and can form ative densities of 36–47%). The biopolymer solution was poured
a comprehensive lattice in the soil matrix. This rigidifies the whole into the molds and was allowed to permeate through the sand and fill
polymeric structure, thereby enhancing its mechanical strength. the voids. The specimens were cured at a laboratory temperature of
Flexibility can be regarded as an initial property of a polymeric 21
C over a 3-day period, followed by oven drying at 35
C over a
chain to facilitate the penetration of a polymeric grout into the soil. 4-day period, after which strength tests were performed in triaxial
On the other hand, it is desirable that cross-linking take place after compression. The mass ratio of biopolymer to dry sand for all the
the grout has permeated the soil void spaces or while it is curing. test specimens ranged from 0.3 to 1.2%.
Taking the preceding factors into account, the authors decided The research was carried out in two stages. First, trials were per-
to focus their efforts on two different types of biopolymers, namely formed on a range of biopolymer combinations and concentrations to
agar and modified starch. It is understood that agar provides the identify which of the treatments were more effective in improving the
highest mechanical strength among biopolymers and that modified strength properties of the sand. Second, unconfined compression and
starch is positively charged. Hence, together they have the potential unconsolidated-undrained triaxial compression tests were performed
to generate significant cohesion and adhesion between the nega- at a strain rate of 0.01%/min, with confining pressures of 100, 200,
tively charged sand particles. and 400 kPa, to identify a suitable modified starch to use in combi-
Agar is contained in the cell walls of several species of red algae, nation with the agar. The test results also provided the data necessary
including Gelidium, Pterocladia, and Gracilaria. Natural agar con- to determine the Mohr-Coulomb parameter values.
sists of a mixture of predominantly agarose, but also agaropectin, with
the former showing good gelling properties. Agarose starts to dissolve
at about 85
C, the exact dissolution temperature depending on the Experimental Results and Discussion
agar source. After dissolution, the viscosity of the agar solution in-
creases constantly with the reducing temperature until finally solid- A series of triaxial compression tests were performed on the dry sand
ifying between 34
C and 40
C. From the range of polysaccharides, and on sand that had been treated with solutions containing 1, 2, and
agarose is the only one that can form a three-dimensional gel network 4% of agar by weight.
in the soil matrix, which would produce a dense film, coating the The data indicated that higher agar concentrations produced
sand grains, after curing. Because of its complex structure, very few higher compressive strength and stiffness in the treated sand (Fig. 1).
bacteria can digest agar gel and thus, biodegradation is minimized. The maximum deviatoric stress, mobilized for a given confining
Agar is also sensitive to polymeric additives and, unless a suitable pressure, also increased at higher agar concentrations (Fig. 2).
additive is used, can significantly reduce strength and chemical re- Because agar gel and sand particles are both slightly negative in
sistance. From the authors’ perspective, it appears that positively charge, it would appear that a valuable practice would be to add
charged starch, which can be manufactured by physical, chemical, a positively charged intermediate agent, such as modified starch, to
or even enzymatic processes, is compatible with agar. In contrast, further enhance the biopolymer treatment. Fig. 3 shows deviatoric
natural starch is either nonionic or slightly anionic, depending on the stress–strain responses of the sand treated with mixtures of 1.0%
material source. Starch is a white, tasteless, odorless powder that agar and 0.5% of the different starches.
does not dissolve in cold water or ethanol. It consists of linear and From Fig. 3, the following are inferred for the starches con-
helical amylose, along with branched amylopectin, which consists sidered in this study:
of glucose monomers. • In general, most starches remarkably increased compressive
strength and stiffness;
• Starpol 136 produced the greatest increase in strength, followed
Materials and Methods by Starpol 469, although the latter demonstrated significantly
higher ductility; and
Fontainebleau sand, with a particle size range of 0.06 to 0.4 mm, was • Staramic 105 and Starpol 700 were found to marginally reduce
used in all the experiments. Approximately 90% of the sand grains compressive strength and also the strain at failure, compared with
by dry mass were in the size range of 0.15 to 0.3 mm. It is a white agar alone.
siliceous sand (SiO2 . 99:8%) with subangular grains, whose rel- Fig. 4 shows the deviatoric stress–strain responses of the sand
atively high specific surface area and clean surfaces are ideal for the treated with 1.0% agar and either 1.0% Starpol 136 or 0.5% Starpol

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2013 / 1403

J. Geotech. Geoenviron. Eng. 2013.139:1402-1406.

Downloaded from [Link] by University Of Auckland on 05/16/15. Copyright ASCE. For personal use only; all rights reserved.

Fig. 1. Unconfined compression tests of sand treated with agar Fig. 3. Unconfined compressive tests of sand treated with 1% agar
and 0.5% starch

Fig. 4. Triaxial compression tests of sand treated with 1% agar and

0.5% Starpol 600/1% Starpol 136

The test specimens failed in compression either by bursting and

disintegrating into sand clusters or by forming a rough shear plane
or an intermediate state for sand treated with 1.0, 2.0, and 4.0%
agar solution, respectively [Figs. 5(a–c)]. Also, the specimen
treated with 1.0% agar and 0.5% Starpol 600 sheared along
a distinct failure plane at a slightly lower axial strain [Fig. 5(d)]. It is
postulated that because agar gel by itself behaves plastically, in-
creasing its concentration introduces some degree of ductility to
the treated soil matrix. The addition of 0.5% starch had the effect of
increasing brittleness, which is consistent with the nature of a
Fig. 2. (a) Triaxial compression tests on untreated sand and sand treated starch solution, which sets into a brittle solid at concentrations
with 1% agar grout; (b) triaxial compression tests on untreated sand and greater than 15%.
sand treated with 2 and 4% agar grout Two mathematical frameworks were used to estimate the
Mohr-Coulomb parameters of cohesion intercept, c, and friction
angle, w, namely nonlinear optimization and linear least-squares
analyses of the maximum deviatoric stress [Eqs. (1) and (2),
600, which were two of the better-performing starches, mobilizing respectively]
greater deviatoric stress compared with sand treated with agar alone
(Fig. 1). Starpol 136 produced a higher deviatoric stress for a given
confining pressure, although the concentration of the former was s1f 2 × c × cos w 1 þ sin w
¼ þ (1)
twice that of the latter. s3f s3f ð1 þ sin wÞ 1 2 sin w

1404 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2013

J. Geotech. Geoenviron. Eng. 2013.139:1402-1406.

 Table 1. Mohr-Coulomb Parameters for Sand Treated with Biopolymers
Method
Nonlinear Linear
optimization least squares
Conditions c (kPa) w (degree) c (kPa) w (degree)
Untreated sand 0 32.3 0 33.1
1% agar 62 24.7 49 25.4
2% agar 111 25.6 105 26.4
4% agar 190 26.3 222 23.6
1% agar and 0.5% Starpol 600 187 17.5 187 17.5
1% agar and 1% Starpol 136 240 17.6 245 17.4
Downloaded from [Link] by University Of Auckland on 05/16/15. Copyright ASCE. For personal use only; all rights reserved.

1 2 sin w
c ¼ b× (2a)
2 cos w

 
w ¼ arcsin m 2 1 (2b)
mþ1

where s1f and s3f denote the major and minor principal stresses at
failure, respectively.
For the nonlinear optimization, the Solver tool of Microsoft
Excel was used to find the responses for c and w of this mathe-
matically overdetermined system (Table 1).
In the linear least-squares analysis, data of (s3f , s1f ) were plotted
in a s3 2s1 plane and fitted by the best line of

s 1 ¼ m × s3 þ b (3)

The relationships between c and w with slope, m, and intercept, b, of

this best-fit line are given by Eqs. (2a) and (2b).
Table 1 indicates that the cohesion intercept was directly pro-
portional to the concentration of agar. Furthermore, the addition of
starch at the same agar concentration substantially increased the
cohesion intercept (for instance, threefold for 0.5% Starpol 600 and
fourfold for 1% Starpol 136). However, the biopolymer treatment
was found to produce a step reduction in w from 32
to 33
for the
untreated sand to 25
to 26
for sand treated with 1–4% agar
solution. The addition of starch produced a further step reduction in
w to 17.5
. It is postulated that the coating effect of the biopolymer
on the grain surfaces smoothened the microscale roughness, thereby
reducing the asperity interlocking of the sand grains.

Summary

Biopolymers can effectively improve the strength characteristics of

sand without causing environmental toxicity. The improvement in
performance of sand treated with agar and modified starch was
found to be directly dependent on the concentration of agar as the
main component and starch as the additive. The addition of Starpol
600 and 136 at the same agar concentration was observed to sig-
nificantly increase the value of cohesion intercept and also to en-
Fig. 5. Shearing modes and strain at failure (peak deviator stress) in hance stiffness.
triaxial compression under a confining pressure of 100 kPa: (a) 1.0% In conclusion, biopolymer treatment shows promise as a tool to
agar at 2% strain; (b) 2.0% agar at 2.5% strain; (c) 4.0% agar at 3% modify and engineer soil behavior to have specific strength and
strain; (d) 1.0% agar and 0.5% Starpol 600 at 1.6% strain deformation characteristics in terms of level of stiffness or ductility.
The sustainability and ecofriendliness of biopolymers also add to
their attractiveness for use in engineering applications.

JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2013 / 1405

J. Geotech. Geoenviron. Eng. 2013.139:1402-1406.

 Acknowledgments Frost, J. D. (1989). “Studies on the monotonic and cyclic behavior of
sands.” Technical Rep., Purdue Univ., West Lafayette, IN.
The first author acknowledges the Irish Research Council (IRC) for Karimi, S (1998). “A study of geotechnical applications of biopolymer
financial support under the Embark Postgraduate Research Scholar- treated soils with an emphasis on silt.” Ph.D. thesis, Univ. of Southern
California, Los Angeles.
ship Scheme 2010 (RS 2010/2571). Special thanks are also due to
Khachatoorian, R., Petrisor, I. G., Kwan, C. C., and Yen, T. F. (2003).
Mr. Marin Carney and Mr. Eoin Dunne for assistance in performing “Biopolymer plugging effect: Laboratory-pressurized pumping flow
the laboratory tests. studies.” J. Pet. Sci. Eng.., 38(1–2), 13–21.
Martin, G. R., Yen, T. F., and Karimi, S. (1996). “Application of bio-
polymer technology in silty soil matrices to form impervious barriers.”
References Proc., 7th Australia-New Zealand Geomechanics Conf., Institution of
Engineers, Barton, ACT, Australia, 814–819.
Bouazza, A., Gates, W. P., and Ranjith, P. G. (2009). “Hydraulic con- Schneider, H. R., Chameau, J.-L., and Leonards, G. A. (1989). “Chemical im-
ductivity of biopolymer treated silty sand.” Geotechnique, 59(1), pregnation of cohesionless soils.” ASTM Geotech. Test. J., 12(3), 204–210.
71–72. Sutterer, K. G., Frost, D. J., and Chameau, J.-L. A. (1996). “Polymer
Downloaded from [Link] by University Of Auckland on 05/16/15. Copyright ASCE. For personal use only; all rights reserved.

Feldman, D. (1989). Polymeric building materials, Elsevier Science Publishers, impregnation to assist undisturbed sampling of cohesionless soils.”
Belfast, U.K. J. Geotech. Eng., 122(3), 209–215.

1406 / JOURNAL OF GEOTECHNICAL AND GEOENVIRONMENTAL ENGINEERING © ASCE / AUGUST 2013

J. Geotech. Geoenviron. Eng. 2013.139:1402-1406.