Advances in Ma terials Physics and Che mist ry, 2012, 2, 102-105
doi:10.4236/ampc.2012.24B028 Published Online December 2012 (htt p://www.SciRP.org/journal/ampc)
Influence of Humidity on Yiel d Stress Determi nation
by Sl um p Test of Sl i p-P rone Clayey Soils and
Their Relation with the Chemical Properties
Arturo F. Méndez-Sánchez, Ana M. Paniag ua-Mercado1, Karen E. Nieto-Zepe da,
Leonor Pérez-Trejo, Elvia Diaz Valdés, Concepción Mejía García
Escuela Superior de Física y Matemáticas, Instituto Politécnico Nacional, Edif. Unidad Profesional Adolfo López Mateos, Col. Lindavista, México
Received 2012
ABSTRACT
In this work, the yield stress evaluation as a function of water content for slip-prone clayey soils is studied in order to understand how
yield st ress decreases as water conten t increases, an d their relatio n with the chemical pro perties. The clayey soil samples were t aken
from the region of Teziutlán-Puebla-Mexico. Yield stress was calculated using the slump test in cylindrical geometry. Results show
three zo nes . The first on e sho ws an exp o nen tial d ecrement o n yield stres s du e to lower water content in accord with clayey soils with
high content of illita, foll owed by a second region where yield st ress decreases d ramaticall y at a cer tain crit ical water concen tration,
and the third one where yield stress dependence is not well-defined since the clayey soil flow is seen. Finally, it is discussed how
yield st r es s var iation due to the water incr ement in fluences the landslide risk in crement.
Keywords: Clayey Soil; Yield Stress; Slump Test; Microstructure; Illita
1. Introduction
There are a few studies focused to analyze the modified physi-
cal parameters before a landslide occurrence [1-3]. Reference
[1] implemented a debris-flow monitoring system employing
real-time rain gauge data. The pre-warning for the time of
landslide triggering derives from the critical rainfall peak ob-
tained from historical events, involving regional rainfall pat-
terns and geological conditions. Reference [2] proposed equa-
tions of state of soil prone to slum-type settlement, which take
into account the degree of wetting in the initial stage. These
equations were developed using models of deformation of the
continuous and experimental results of cohesion and the angu-
lar coefficient of internal friction as well as the bulk compres-
sion and shear modulus. Those authors proposed a plasticity
function that decreases exponentially when the wetting content
in the soil is increased. It is clear that plasticity function is one
of the most important modified parameters before o f a land slid e
occurr ence by rainfall . Hence, l andsl ides can take place b ecause
of load excess generated by a water saturated soil overcoming
yield stress [ 4-6], as well as, infiltrated water excess in the soil
(decrement of the pore pressure) produces a yield stress decre-
ment and the internal load overcomes the decremented yield
stress.
In this work, the yield stress evaluation as a function of water
content for slip-prone clayey soils due rainfall is studied in
order to understand how yield stress is decremented by the
water content. Yield stress was calculated for several water
concentrations using the slump test in cylindrical geometry.
Particularly, samples of the region of Teziutlán-Puebla-Mexico
were tested and the results were analyzed and compared with
the historical daily rain data of October 1999, when a landslide
occurred in the zone. In addition, a comparison of the chemical
microstructure and the compound determination using Energy
Dispersive Spectroscopy by X-ray dispersion was performed.
As well, clayey soils were characterized by SEM observation
and X-ray diffraction .
2. Exp erimenta l Pro cedu r e
The studied clay corresponds to high risk zone located in the
Aurora neighborhood in Teziutlán-Pu e bla-Mexico, where a
landslide took place due to high rainfall in October 1999 [7].
The zone where the sample was taken corresponds to a transi-
tion zone of two physiographic units-the transversal volcanic
belt and oriental mountain chain. Andosol is the predominant
soil d erived from volcan ic materials; als o, there are ign imbrites
and cl ayey soils. Th is kind o f soil is charact erized b y a variable
high capacity of acquiring water and humidi ty.
Micro anal yses of chemic al co mpo sit ion were performed with
an energy dispersive spectroscopy technique (EDS) attach ed to
a scanning electron microscope FEI, Sirion. In addition, X-ray
diffraction exp eri ments were c arri ed o ut emplo yin g a MM A, GBC
diffractometer in order to determine the clayey compounds, by
using CoKα radiation (λ = 1.789Å) in the 2θ range of 5-120
degrees with a 0.02 step and 0.5s as step width and step count-
ing time resp ectively.
The samples were sifted with a standard mesh No. 8 (2.36
mm) mesh in order to eliminate larger debris. Samples of 0.3 kg
of clay were p rep ared at 3 0-40 wt% of water concentration, and
slump test experiments were carried out [8]. The method con-
sists of filling a cylindrical frustum with the material to be
tested in the specified way; lifting the frustum off and allowing
the material to collapse under its own weight (Figure 1). The

A. F. MÉNDEZ-SÁNCHEZ ET AL.
103
height of the final slumped material is measured and the differ-
ence between the initial and final heights is called the slump
height (s). Figure 1 outlines the experimental procedure.
Yield stress value ( τy) was calculated by the equation of Pa-
shias and coworkers, expression 1.
11 .
22
y
s
gH H
τρ

= −



(1)
where ρ is the material density, g is the gravity, s is the slump
height, and H is the frustum height. In this case, the slump
height was measured at room temperature, after 40 seconds
lifting the frustum off, as was suggested in a previous work
[9,10].
3. Results and Discussion
Figure 2 shows SEM micrographs of clayey soil. It can be
Figure 1. Slump test diagram, a) frustum filling, b) frustum lifting,
c) collapsed materia l, and d) slump height measurement.
(a)
(b)
Fi gur e 2. SEM images at a) 500x and b) 1000x, granular shape with
fiber con f ormation.
observed a granular shape with a fibrous surface of individual
particles. Table 1 shows the chemical composition determined
by EDS analysis. High contents of aluminum and silicon were
detected as expected for this type of material. Besides, a low
cont ent of Iron and Titanium was observed in this material.
Figure 3 shows the particle size distribution. It can be seen that
60% of the p articl e sizes are i n the ran ge between 3 0 0 and 1250
microns, the 10% are in the interval 1250- 2360 microns, and
the rest 30% of the particle size is shorter than 200 microns.
The X -ray diffraction analysis of clayey soil shows the pres-
ence of compounds, such as illite (39.79%), gibbsite (33.74%)
and cristobalite (26.47%). The peak identification is shown in
Figure 4 and the Percentage of mineralogical phases is shown
in Tabl e 2.
Figure 5 shows the plot of yield stress, τy, versu s water con-
centration expressed in weight percentage. In the case of con-
tents lower than 35.5 wt%, the yield stress decreases exponen-
tially with concentration. The regression equation is also shown.
These results are in agreement with Sultanov and Khusanov’s
model [2], as well as with that reported by Sánchez-Crúz [9].
Tabl e 1. Chemical compo si t ion o f c l ayey soil.
Element Wt% Int. Error
O 49.09 0.55
Al 18.17 0.59
Si 23.1 0.55
Ti 1.42 3.3 7
Fe 8.23 1.46
Total 100
20
40
60
80
100
0.1 1
0
5
10
15
20
25
30
35
Finus
Particle size distribution
Gauss fit
Frequency (%)
Particle size (mm)
Cum ulative frequency
Cum ulative frequency
Sigmoidal fit
Figure 3. Particle size distrib u t ion of th e clayey soil .
Tabl e 2. Percentage of mineralogical phases.
Element Percenta ge %
SiO2 39.79
KAl2(SiAlO10)(OH)2 33.74
Al2O3H2O 26.47
Total 100.00

A. F. MÉNDEZ-SÁNCHEZ ET AL.
104
020 40 60 80100120
0
100
200
300
400
500
600
700
3
3
1
3
1
1
3
3
(2) 2
11
11
1
3
2
1
22
2
Counts
Degrees 2-Theta
1) SiO
2
2) K Al
2
(Si
3
Al O
10
)(OH)
2
3) Al
2
O
3
!
3
3H
2
O
Figure 4. X-ray diffraction pattern of the clayey soil.
30 32 34 36 38 40
0
50
100
150
200
250
300
350
21.0 21.5 22.0 22.5 23.0 23.5 24.0 24.5 25.0
Y ield stress (Pa)
Water concentration (%)
Y ield stress measurements
τ
y
=170.7157e
-.08397 *C
R a in fall milimet e rs ( m m)
Figure 5. Y i eld st ress ve rsu s w at er p erc en t a ge con cen t r at i on .
These authors studied a clayey soil with the presence of illite,
which showed a similar behavior. In the case of contents be-
tween 35.5 wt% and 36 wt%, the yield stress shows an abnor-
mal behavi or and it decreases substan tially, up to 50 percent of
its initial value. At this point, it is possible to elucidate an in-
crement in the landslide risk, since the sample has changed
from solid-plastic to solid-viscous behavior. It is important to
mention that this decrease in yield stress was not predicted by
the Sultanov and Khusanov’s model, in spite of having in-
cluded the plastic and the viscous behavior in their model. For
higher water concentrations (>36 wt %), a non-linear d ecr ement
on yield stress is seen and it differs from the exponential or
power-law behavior. We believe that this response is due to a
combination of non-Newtonian behavior and yield stress and
this is not possible to separate them in the slump test.
Additionally, upper horizontal axis in Figure 5 shows the
variation of yield stress versus equivalent millimeters of rainfall.
In this case, it was supposed that all of the water was absorbed
by the clayey soil. Millimeters of rainfall (h) were calculated b y
using the expression 2.
.
w
V
hA =
(2)
where VW is the water volume in the frustum, and A is the fr us-
tum c r oss section .
Under this assumption, it can be seen that only 23 mm of
water are enough for the soil to start to flow. However, this
value is lower in comparison with the historical rainfall data
obtained in the studied geographical zone [7], where the maxi-
mum rainfall peak was reached (360 mm of water) the day
before the landslide and considering that in the previous ten
days, an unusual accumulated rainfall reached 908 mm (com-
pared with the medium annual rain 1593 mm). This difference
arises from the small quantity of the rainfall absorbed by the
soil ( natur e’s soil ) and by the f act th at most o f the water moves
down due to the region´s inclination (23 degrees). In order to
clarify this, it would be necessary to carry out the yield stress
determination immediately after the occurrence of a landslide
and measure the absorbed rainfall water.
4. Conclusions
Yield stress determination as function of water content by a
slump test for a clayey soil from a Teziutlán-Puebl a -Mexico
zone was performed. The results showed an exponential
decrement of yield stress followed by an abrupt reduction of it
with the increase in water concentration. From this value, an
increment of the risk of landslide was revealed. At high water
content (36%), a decrease in yield stress was observed, and a
more complex behavior was exhibited. Finally, a correlation of
yield stress with rainfall was done, but results were below the
values reported in the literatur e.
5. Acknowledgements
The authors thank to Professor V. M. López-Hirata by his use-
ful commen ts.
REFERENCES
[1] C . Ch ien -Yuan, C . Ti en -Chen, C. Y. Fan-Chieh, Y. Wen-Hui, T.
Chun-Chieh, “Rainfall duration and debris-flow initiated studies
for real-time monitoring,” Environment Geology, vol. 47, p.p.
715–724, 2005.
[2] K. S. Sultanov, B. E. Khusanov, “Stat e equation s for soils p rone
to slump-type settlement with allowance for degree of wetting,”
Soil Mec hanics and Foundation s Engineering, vol. 38, No. 3, p.p.
80-86, 2001.
[3] I.A. Caldiño-Villagómez, I. Bonola-Alonso, G.
Salgado-Maldonado, “Estudio experimental del esfuerzo de
cedencia con relación al flujo de lodos y debris,” Asociación
Internacional de Ingeniería e Investigaciones Hidro-Ambientales
vol. 8, [Memorias del XXII Congreso Latinoamericano de
Hidráu lica, Guayana, Ven ezuela].
[4] D. F. Van Dine, R. F. Rodman, P. Jordan , J. Dupas, “Kusk onook
Creek, an example of a debris flow analysis,” Lanslides vol. 2,
p.p. 2 57-265, 200 5.
[5] R. M. Iverson, “The physics of debris flows,” Reviews of Geo-
physi cs, vol. 35, No. 3, p.p. 245–296, 1997.
[6] R . P .Den linger, R . M. Iverson, “Flow of variably fluidized gra-

A. F. MÉNDEZ-SÁNCHEZ ET AL.
105
nular masses across three-dimensional terrain 2. Numerical
predictions and experimental tests,” Journal of Geophysical
Researc h, vol. 106, No. b1 , p. p. 553–566, 200 1.
[7] P. Flores Lorenzo, I. Alcántara Ayala, “Cartografía
morfogenética e identificación de procesos de ladera en
Teziutlán, Puebla,” Investigaciones geográficas Boletín, vol. 49,
p.p. 7-26, 2002.
[8] N. Pashias, J. Boger, D.V. Summers, D. J. Glenister, “A fifty
cent rheometer for yield stress measuremen t,” Jou rnal of Rheol-
ogy, vol. 4 0, No. 6, p.p. 1179-1189, 1996.
[9] P. Sánchez Crúz, “Análisis del esfuerzo de cedencia de suelos
arcillosos como posible indicador de un derrumbe,” Bachelor
Thesis, ESFM, Instituto Politécnico Nacional, Mexico, 2008.
[10] A. F. M én d ez-Sánchez, L. P ére z-Trejo, A. M. Paniagua Mercado,
“Dete rmi na ción del esfu erzo de c eden cia para s uelos vuln erab les
a deslizamientos originados por lluvias,” Boletín de la Sociedad
Geológica Mexicana, vol. 63,No 2, p.p. 345-352, 2011.