Journal of Civil Engineering, JKUAT, Vol. 7, March 2002, pp.

1-8

Engineering properties of common subgrade soils below pavement structures in Kenya

Sixtus K. Mwea 1) and Francis J. Gicbaga 2)

1) Senior Lecturer, Department of Civil Engineering, University of Nairobi

2) Professor of Civil Engineering, Department of Civil Engineering University of Nairobi, and Chairman, Kenya National Council of Science and Technology

ABSTRACT

This paper reports findings of laboratory testing of common subgrade soils in pavement structures in Kenya. The materials were collected from different parts of the country. The results are intended to form a database for use by practicing engineers and researchers in the field of pavement engineering. The investigated materials include the expansive black soils, silty clayey soils, sandy loamy clayey soils and lateritic gravels. In addition the properties of lime improved gravels are reported. For the fine-grained soils the plasticity index was found to be linearly proportional to the liquid limit. In all the investigated soils the optimum moisture content was found to be inversely proportional to the maximum dry density. The increase in the strength of lime improved materials was found to increase with lime content at a diminishing effect.

Key words ~ pavements, expansive black soils, silty clayey soils, sandy loamy clayey soils and lateritic gravels

INTRODUCTION

Flexible pavement design revolves around obtaining a pavement cover over a subgrade soil. The cover thickness for a particular pavement material is dependent on the subgrade strength and its plasticity. Subgrade materials in Kenya comprise of natural soils found in different geological state and conditions varying across the breadth and width of the country (Roads Department, 1987). Some of the previous studies aimed at studying the pavement material properties include Gichaga et al (1987) who investigated the properties of Nairobi red coffees soils and Mwea and Ngware (1996) who investigated the properties of Uashin Gichu red coffee soils. Gichaga (1982) studied the deformation modulus - time relationships for various subgrade soils under road pavements. Under the study the basic engineering properties of the investigated soils were established. The significant properties are summarized on Table 1

T bilE .

f b d 'I

a e ngineenng properties 0 su igra e SOl s
Murram from Red soil from Black cotton soil
Thika road Langata from Langata
Liquid Limit (%) 34 53 83
Plasticity Index (%) 10 17 32
Maximum Dry Density (kg/m '] 1630 1480 1330
Optimum Moisture content (%) 23 29 31 Source: Gichaga (1982)

1

FIELD SAMPLING

Pavement subgrade soils were collected from various parts of the country (Figure 1). Expansive black clays commonly known as black cotton soils were sampled from Bunyala division of Western province. In this area, roads were being designed to serve a large irrigation project covering a large area. The area sampled is on the River Nzoia flood plains. The soils here are predominantly black cotton soils. To fully identify and characterize the soils for the roads in the irrigated areas a sampling grid of one kilometer was adopted.

The red silty clays commonly known as red coffee soils were sampled at two locations in Meru district in Eastern Province. The locations of sampling were Liutu and Likiundu. The samples were taken on the proposed alignment of access roads to an earth-dam that was under design. The area here is generally hilly and well drained. The depth of the red soils cover was found to be over ten meters, but sampling was done in the top one and half meters. This is the depth where engineering activities with respect to the access roads were confined.

Subgrade soils were also sampled along the Wote - Makindu road. This earth/gravel road passes through a semi-arid part of Makueni district in Eastern Province. The sixty fivekilometer road passes through predominantly sandy loamy silt clays. Sampling was carried out at one-kilometer intervals to depth of one and half meter depth. The road was being designed with an aim of improving it to a bitumen-surfaced road.

Lateritic gravels were also obtained from Western province divisions of Mumias and Amagoro. These were to be used in the construction of the irrigation scheme roads. Several gravel quarries along the Wote-Makindu road were also investigated. The quarries were to be the source of base and subbase materials for the improved road .

./

Figure 1 Location sites for pavement building materials

2

LABORATORY TESTING

Samples from the field were subjected to identification, compaction and strength tests. Lateritic gravels are usually used in the subbase and the base layers of the pavement structures. For these soils therefore further testing was required to assess their suitability in the upper pavement layers. The lateritic gravel samples obtained from Wote-Makindu road were therefore tested after lime improvement in addition to the above tests.

CLASSIFICATION OF FINE-GRAINED SOILS (BLACK COTTON SOILS AND RED COFFEE SOILS).

The results of the plasticity test are plotted on the plasticity chart for the unified classification system (Smith and Smith, 1998) on Figure 2a and 2b.

50

.....

! 40

~

"

.s

30

~ 20 ~ 10

"

a: 0

o 10 20 30 40 50 60 70 Liquid Limit (%)

a) Black cotton soils (WestemKenya)

(b) Red coffee soils (Meru)

50

!4O j 30

~ 20 .!!

1 10

L

o

o 10 20 30 40 50 60 70

Liquid limit (%)

Figure 2 Classification of the fine grained soils

For the black cotton soils from Western Kenya it was found that the soils band around the A line. The majority of the sam,ples were found to lie above the 50% liquid limit (LL). These soils are therefore, clays or silts of high compressibility (CH and MH) Equation 1 shows the correlation of plasticity index (PI) and LL for the investigated soils. The red coffee soils from Meru district were found to plot below the A Iine also above the 50% LL. These soils are therefore silts of high compressibility. Equation 2 shows the correlation of PI and LL for the red coffee soils.

PI = -0.5 + 0.45LL (r= 0.64) PI= -15.62 + 0.64LL (r=0.84)

{1/ {2/

Comparison with Nairobi red coffee soils, which were, investigated by Gichaga et al (1987) and Kipwen red coffee soils (Mwea and Ngware 1996) in Uashin Gichu district of the Rift valley shows that these soils are similar as can be seen on Figures 3a and 3b.

3

-------~-----------------,

40 ~ 35 -; 30

-8 25

c

- 20 b

U 15

i 10

I'J

Ii:

40

--

rft

-; 30

i

._ 20

b

u

i 10 Ii:

o

o 10 20 30 40 50 60 70 Liquid Limit (%)

o 10 20 30 40 50 60 70 80 Liquid limit (%)

(b) Uashln Gichu

SOUTCB: MwtUI and Ng_,. (1996)

(a) Nairobi SOUTCfll Glchaga fit al (1987)

~----------------------~ L ~

Figure 3 Classification of Nairobi and Uashin Gichu red coffee soils

CLASSIFICATION OF THE SANDY LOAMY SOILS ALONG THE WOTEMAKINDU ROAD

The variation of the geotechnical properties of the sub grade soils along the sixty fivekilometer road corridor was found to be large. The fine portions of soils were found to be of low plasticity with an average PI of 14%. Overall, the sand proportion was found to be high (58%) with high proportion of fines plotting above the A line and below the 50% liquid limit mark (Figure 4). The soils can therefore be classified as clayey SANDS (SC).

o

10

20

10

30

4q

50

60

Liquid Limit (%)

Figure 4 Classification of sandy loamy soils on the Wote - Makindu road

CLASSIFICATION OF THE LATERITIC GRAVEL SOILS

Unified soils classification (USC) of soils with fine portions less than fifty percent depends on the particle size distribution, and the plasticity of the fines. Using the mean values for the test results, the classification for the various investigated gravel soils is shown in Table 2

4

T bi 2 CI 'fi

fl t .,

a e ass! ication 0 a enuc grave s. ' _(
Location LL(%) PI (%) Gravel (%) Sand (%) Fines (o/<0_ USC
Mumias 42 13 70 17 13 GM
Amagoro 35 11 73 16 11 GM-P
Wote-Makindu 38 15 50 32 18 GC Legend: - LL - Liquid limit, PI -- Plasticity index

Gravel from Western Province (Mumias and Amagoro) was found to have lower percentage of fines than the gravel from quarries located along Wote-Makindu road in Eastern Province. The Mumias gravel can be classified as silty GRAVEL (GM), while that from Amagoro had less than 12% fines but more than 5%. It is therefore described as a poorly graded silty GRAVEL (GM-P). The Wote-Makindu quarries had a higher percentage of sand and fines. These gravels are classified as silty GRAVEL (GM).

COMPACTION CHARACTERISTICS

The optimum moisture content (OMC) is the water content at which the maximum dry density (MDD) has been achieved. In all the tested soils the variation of OMC with MDD was found to be of the form given in Equation 3. Statistical analysis of OMC versus MDD was done. Table 3 shows the statistical constants related to Equation 3. The results for the general equation is shown graphically on Figure 5

OMC=A -MDDIB

Table 3 Statistical constants for MDD Vs OMC (Equation 3)

Soil TVlJe A B r
Black cotton Soils 71 32 -.85
Red coffee soils 84 27 -.92
Lateritic gravel soils 66 38 -.89
Clayey sands 54 47 -.92
General equation 66 37 -.93 Legend - A & B -constants, r - coefficient of correlation

-~--~.-----,------------

50

40

~30

-

(J

::E 20 o

10

o 1200 1400 1600 1800 2000 2200 2400

MDD (kg/m3)

Figure 5 OMC Vs MDD for Kenyan subgrade soils

13J

PROPERTIES OF THE FINE-GRAINED SOILS WITH RESPECT ro PAVEMENT PERFORMANCE

Black cotton soils

The black cotton soils tend to be problematic soils. Their low strength, high plasticity and swell characteristics affect their performance under traffic load. It is for this reason that the Roads Department (1987) recommends that the black cotton soils be placed in the lower sections of the fill where the weight counteracts the expansion of the soil. This treatment of expansive soils compares with the recommendations of the State of California (1992) handling of expansive clays where a surcharge is placed over expansive soils before placement of he pavement structures. Ibi (1992) also recommends the treatment of the expansive soils with lime.

The strength of the black cotton soils was investigated by the measurement of CBR and plasticity. The mean CBR for the test samples was four. This placed them in the S 1 subgrade class (Roads Department, 1987). The use of such weak soils as a sub grade material is to be avoided. They should be either replaced or an improved sub grade is used as cover. The CBR was found to be linearly related with plasticity index (PI). This can be explained by the fact that CBR is a measure of the shearing strength of the soil. On the other hand PI is a measure of the range of water contents when the soil is in the plastic stage. This is a weak stage of the soil and regression analysis of the experimental data gave Equation 4

eBR = 11- 0.33P1 (r= 0.82)

[4J

Red coffee soils

The investigated red coffee soils had their CBR values in the range of 9-49. This places them in the subgrade regions of S3-S5 (Roads Department, 1987). These fine-grained soils are suitable pavement foundation. The pavement design over these soils revolves around the selection of a pavement structure depending on the traffic loading and available pavement materials.

Unlike the black cotton soils these soils were found to have the CBR not well correlated to their PI. The poor correlation is due to their coarser particles and the gradation becoming increasingly important in the strength characteristics. Indeed for the lateritic gravels it was found out that strength was now inversely related to the percentage of fines as discussed in the next section.

Lateritic gravels

The results for the tested gravel soils show considerable variation in plasticity and strength characteristics. The strength of the soils was found to be inversely proportional to the plasticity modulus (PM). The plasticity modulus is defined as a product of PI and % passing 0.425mm sieve size. The strength was also found to be inversely proportional to the percentage of fines. Figure 6 shows the trend of CBR with PM and percentage of fines. The relationship shows a high scatter and is indicative of the difficulty in linking of the strength and the plasticity modulus.

Table 4 shows the Roads Department (1987) requirements for gravel when used as base or subbase material. From Figure 6 it can be seen that a large percentage of gravel in the area of investigation does not meet the plasticity and strength characteristics. Indeed for the twenty quarries investigated along the Wote-Makindu road none met the design requirements for the base while only 60 % were suitable for the subbase.

6

T bl 4 G

I h

t . t'

. d f

bb

db

a e rave c arac ens lCS require or su ase an ase
Layer Minimum CBR (%) Maximum PI Maximum PM:
Base 80 15 250
Subbase 30 15 250 Legend: PI - Plasticity index, PM - Plasticity modulus Source: Roads Department, 1987

Lime improvement was performed on the samples from Wote-Makindu road. In all the cases the effect of addition of lime was to reduce the plasticity and plasticity modulus. This was accompanied by an increase in the strength of the gravel. All the samples had their PI and PM values falling considerably. For the four and six percentage contents, the samples were non-plastic and hence PM reduced to zero. Overall the increase in lime content enhanced the strength as assessed by CBR but with diminishing effect as the percentage of the lime increases. This can be seen in Figure 7 that was developed from the test results. The trend in the figure is polynomial relationship with a coefficient of determination (r2) of 0.69.

Deformation Modulus of the fine grained soils

Gichaga (1982) established that the deformation modulus of the fine-grained sub grade soils decreases with time of loading. Table 5 shows the decrease in loading time from 10-2 seconds the time of loading equivalent to a vehicle at 100 km per hour to a loading time of 12 days (l06 seconds). The overall result is that soils undergo substantial loss in modulus with loading time and that the loss in stiffness varies with the type of soil.

T bl 5 M d I (MNI 2) T

1 ti hi s for Kenyan sub grade soils

a e o u us m - ime re a ions 11)
Loading Time (Sec) lO-L 10L lOb
Lateritic Gravel 637 274 25
Red Coffee Soil 12 5.4 0.98
Black Cotton Soil 16 6.4 0.39 Source Gichaga (1982)

l100

ri

o

20

30

40

50

% Fine.

Wote Maklndu road lateritic gravel.

o

10

20

30

40

PM

(b) Wote Maklndu road lateritic

gravel.

Figure 6 Variation of CDR with plasticity modulus and % fines

7

300
250
- 200
~
-- 150
&II:
III
U 100
50
0
0 Lime Content (LC) 0/0

---~-----

Figure 7 Effect of lime on CDR for lateritic gravels

CONCLUSIONS

The properties of pavement materials have been presented. Relationships of optimum moisture content and maximum dry density have been are presented. Some of the factors affecting pavement strength have been identified as plasticity and percentage of fines. The increase in strength as measured by CBR of lime improved materials have been quantified. The specific conclusions, which can be drawn from the study, are:

1 The plasticity index of the soils increases with the liquid limit

2 The optimum moisture content of the soils decreases with maximum dry density

3 The strength of the soils decreases with the plasticity index

4 The increase in strength of lime improved gravels increases with lime content with diminishing effect.

RECOMMENDATIONS

It is recommended that the various subgrade soils in the country be tested under repeated loading to mode'! out the actual field conditions

REFERENCES

Gichaga FJ (1982). Laboratory study of deformation modulus - time relationships for various subgrade soils under road pavement structures. Kenya Journal of Science and Technology series A Vol 3, No 2 pp 63-73

Gichaga FJ Sahu BK and Visweswaraiya TG (1987) Compaction and compression characteristics of Red coffee soils in Kenya In proceedings of the eighth international conference for soil mechanism and foundations engineering. Cartagena, Colombia pp 295-304.

Mwea SK and Ngware R (1996). Suitability of the Kipwen River basis soils with respect to earth-dam construction. Discovery and innovation Volume & Number 2 pp .119-131 Roads Department (1987). Road design manual part Ill. Materials and pavement design for new roads. Ministry of public Works Nairobi.

Smith GN and Smith IGN (1998) Elements of soil mechanics seventh edition pp494 Blackwell science Ltd. UK.

State of California, Department of Transportation (1992). Highway design manual, chapter 600. Design of the pavement section pp. 1-74

8