See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/250191790

Evaluation of gravel for concrete and road aggregates, Rapti River, Central
Nepal Sub-Himalaya

Article in Bulletin of the Department of Geology · November 2008

DOI: 10.3126/bdg.v10i0.1425

CITATIONS READS

7 1,385

2 authors:

Surendra Maharjan Naresh Kazi Tamrakar

irrigation nepal Tribhuvan University
4 PUBLICATIONS 19 CITATIONS 86 PUBLICATIONS 495 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Surendra Maharjan on 24 August 2018.

The user has requested enhancement of the downloaded file.

 Bulletin of the Department of Geology

Bulletin of the Department of Geology, Tribhuvan University, Kathmandu, Nepal, Vol. 10, 2007, pp. 99–106
Cen

gy
olo
t

lD
Ge
ra

epar
tment of
Kirtipur

Evaluation of gravel for concrete and road aggregates, Rapti River,

Central Nepal Sub-Himalaya
*Surendra Maharjan and Naresh Kazi Tamrakar
Central Department of Geology, Tribhuvan University, Kathmandu, Nepal

ABSTRACT
Several local mining sites of sand and gravel have been operating for decades in the Rapti River. River gravels are consumed
in Hetauda, Narayanghad and areas in India near the border between Nepal and India. Until now little known about consumers
concerning about quality of gravel. The present study was carried out to evaluate quality of river gravel to know its suitability for
aggregate (raw material for concrete and road). The samples of river gravel were analysed for petrographic, physical, mechanical
and chemical properties. Samples were categorised as quartzite group, carbonate group and granite group according to British
Standard Institution (BSI). Among these, samples of quartzite group were found dominant. Image analysis of gravel showed that
clasts were well graded. The majority of the samples had rounded, high sphericity and oblate triaxial clasts. The surface texture of
clasts was rough to smooth. In terms of shape, workability of gravel was satisfactory. Gravel samples possessed low water absorption
value (0.69 to 1.12%) and low effective porosity. Dry density of samples ranged from 2460 to 2680 kg/m3. Aggregate impact values
of samples (14.2 to 16.1%) showed good soundness. Los Angeles abrasion test also showed consistent hardness of each of the
samples as uniformity factor did not exceed 0.2. Magnesium sulphate values ranged between 4.46% and 7.29% suggesting good
resistance against chemical weathering and frosting. Comparing with the existing Nepal Standard, British Standard and American
Standard of Testing Material, the studied samples were suitable for concrete and road aggregates.

et al. 2006). From this region the construction material

INTRODUCTION
of worth Rs. 53 million has been exported to India
The Rapti River is known for carrying large in the year 2000/2001 (DMG 2004). Therefore, the
amount of sediments from the Lesser and the Sub- sediments of the Rapti River are of economic
Himalayas. The river sediments are chiefly gravel importance of the Central Development Region.
with few sand and mud. Gravel (coarse aggregates Although coarse aggregates have been used for
greater than 4.75 mm) can be easily collected from concrete and filling material for road embankment
the river area after deposition during each flood in the area, there is little concern about quality.
season. Hence, the river is perennial source of coarse Therefore, authors obtained samples from the river
aggregates. The use of coarse aggregate from the elements (side and mid bars) and analysed to evaluate
Rapti River as construction material has been quality of coarse aggregate.
extensive because of high demand in major cities
like Hetauda and Narayanghad. Gravel is also STUDY AREA
exported to nearby border towns in India. About 3.4 The study area lies in Hetauda Dun Valley, Central
million m3 of boulders, gravel and sand are reported Nepal (Fig. 1). The Rapti River flows northeast to
to be present for the depth of one meter in the major southwest and contributes the Narayani River. It is
rivers of the Central Development Region (Duvadi a meandering river in the upper stretch and becomes
braided in the middle to downstream stretches. It
*Corresponding author: includes different bar deposits containing huge amount
E-mail address: nerus00977@hotmail.com of gravel and minor sand. The East-West Highway
 S. Maharjan and N. K. Tamrakar/ Bulletin of the Department of Geology, Vol. 10, 2007, pp. 99–106

80
30

C H I N A

NEPAL

INDIA Hetauda
26
86

MA
er HA
i Riv SI BH
an DU N VA LL EY
ray W AR

27o 32’41”
NaMEGHAULI A
LI AT
R10 R7 RA
R9 R8 K
NG
R13 R12 R11 E
27o 32’41”

Rapti R
ive R6 RA
SIWA
r NG
LIK R E R2
ANGE R5 R3
R4 HETAUDA
I N
27o 20’41”

D
N I A

27o 20’41”
Main Boundary Thrust Watershed boundary
Fig. 1 Location map of study
0 4 8 16 24 K m R2 River bar sample International boundary area showing sampling points
84o 6’42” 84o 18’42” 84o 30’42” 84o 42’42” 84o 54’42” 85o 6’42”
of river sediment

runs along the right bank of the river which is the Test was made based on ASTM (1978 and 1989).
main access of the study area. Gravel roads and foot
RESULTS
trails link the river from the highway.
Composition and texture
Material and method
The major constituents of gravel samples were
Altogether 12 samples from bar deposit (surface)
quartzite, sandstone, granite gneiss and limestone.
of the river were taken for analysis (Fig. 1). Sampling
Among them, quartzite is the dominant constituent
was made following the method of Howard (1993).
in all the samples (Table 1). The clasts of amphibolite,
About 10 kg of each sample was collected. The
schist, phyllite, slate, quartz and feldspar were
photographs of the sampling spots were obtained and
commonly found in all the samples. The samples
image analyser was used for photo sieving. The gravel
were categorised as quartzite group, carbonate group,
samples were utilised in determining composition,
schist group and granite group (BSI 1975) (Table 1).
texture, aggregate impact value, abrasion value and
chemical durability. Surface texture defines the periphery of aggregates
whether it is smooth, rough, honeycombed, granular
The samples were washed to remove surface
and crystalline (BSI 1989). From the observation,
coating of organic material and clay, if any present
quartzite clasts had smooth surface texture whereas
for easy identification of clast type. Then, the samples
granite, sandstone, carbonate, gneiss, schist, phyllite
were grouped into categories after BSI (1975). About
and slate clasts had rough surface texture. Therefore,
100 clasts of each sample were measured for their
the samples had mixed type of surface texture which
three mutually perpendicular axes; longest (a),
should provide sufficient bonding with cement.
intermediate (b) and shortest (c) axes to determine
shapes of clasts (Fig. 2). The physical properties such The clast shape was quantified in terms of
as water absorption and dry density were determined elongation ratio (q = b/a) and flatness ratio (p = c/b)
after ISRM (1979). Aggregate Impact Test and Los (Table 2). The shape factor (F = p/q) ranged from
Angeles Abrasion Test were made to obtain 0.82 to 1.03 showing nearly equant shape of clasts.
mechanical soundness and hardness of samples. To The plots of elongation and flatness ratios on the
obtain soundness of aggregates against frosting and modified Zingg’s (1935) shape classification diagram
chemical weathering, Sodium Sulphate Soundness after Lees (1964) showed that the clasts were oblate

100
 Evaluation of gravel for concrete and road aggregates, Rapti River, Central Nepal Sub-Himalaya

(a) (b)

Fig. 2 Clasts used for measuring the longest, the intermediate and the shortest axes to determine shape of clasts in samples; (a) sample
R5 and (b) sample R7

triaxial (Fig. 3). Roundness of clasts was calculated after Janoo

(1998) as:
Sphericity (Y) was calculated using the expression
of Aschenbrenner (1956) as:
R = 4pA/P2 (2)

Y = {12.8 (p2q)0.3}/{1+p (1+q) where, A is area and P is perimeter of a clast. The

average roundness (R) varied between 0.81 and 0.86
(Table 2) and were subrounded to rounded.
+ 6 (1+ p2 (1+q2))0.5} (1)
The river gravel had dominantly rounded, prolate
Sphericity ranged from 0.84 to 0.91 (Table 2), triaxial and high sphericity clasts. These shape
therefore the clasts possessed high sphericity (Fig.3). attributes suggest that the gravel when it will be used

Table 1: Composition of gravel of the Rapti River

Samples
Composition (%) R2 R3 R4 R5 R6 R7 R8 R9 R10 R11 R12 R13
Sandstone 9 6 6 12 12 10 13 15 15 17 9 13
Limestone, dolomite and marble 25 22 26 28 11 6 2 4 12 7 8 6
Quartzite 28 21 29 42 50 55 60 66 48 62 62 44
Slate and phyllite 12 20 15 2 6 2 2 0 2 1 0 0
Schist 15 11 6 3 1 1 0 0 2 0 0 3
Gneiss 5 11 6 2 2 2 4 1 4 1 1 7
Granite and pegmatite 2 5 5 4 10 15 18 9 15 10 18 25
Amphibolite 2 3 4 0 2 3 1 1 1 2 0 1
Vein quartz 2 1 2 6 5 5 0 3 1 0 1 0
Feldspar 0 0 0 0 0 0 0 0 0 0 0 1
Undifferentiated grains 0 0 1 1 1 1 0 1 0 0 1 0
*Compositional group QG- SG- QG- QG- QG QG- QG- QG QG- QG QG- QG-
SG-CG QG-CG CG CG GG GG GG GG GG
*(BSI 1975): QG = Quartzite group; SG = Schist group; CG = Carbonate group; GG = Granite group

101
 S. Maharjan and N. K. Tamrakar/ Bulletin of the Department of Geology, Vol. 10, 2007, pp. 99–106

Table 2: Results of form, sphericity, roundness and size of clasts of the gravel samples

Samples Elongation ratio Flatness ratio Shape factor Sphericity Roundness Median grain size
q p SF Y R d50 (mm)
R2 0.7 0.58 0.82 0.85 0.86* 34.3
R3 0.7 0.57 0.82 0.84 0.84* 34.3
a
R4 0.68 0.62 0.91 0.86 0.81 13.9
R5 0.72 0.64 0.89 0.88 0.82* 15.5
R6 0.71 0.66 0.94 0.88 0.83* 21.1
R7 0.73 0.63 0.86 0.88 0.85* 15.5
R8 0.72 0.69 0.96 0.89 0.83* 23.4
R9 0.74 0.76 1.03 0.91 0.85* 19.7
R10 0.72 0.66 0.92 0.88 0.86* 39.4
R11 0.71 0.63 0.88 0.87 0.82* 22.6
R12 0.74 0.66 0.88 0.89 0.84* 21.9
R13 0.74 0.67 0.9 0.89 0.84* 59.7
a
subrounded; * rounded; interpretation of roundness after Folk's roundness ( r): r = 10.83 log (R) + 6.18

as aggregate will provide good workability with and rod shaped particles require more water to produce
cement. In fact the shape of coarse aggregates affects workable concrete than rounded and compact particles
workability and mobility of concrete (Lees 1963). in aggregate. In this case cement should also be
Workability decreases if the majority of clasts are increased to maintain the water-cement ratio.
disc and rod shaped (Neville 1996). If the clasts are
The average median grain size varied from 59.71
angular, they can decrease workability by 10%
to 13.93 mm (Table 2). The river gravel had well
(Kaplan 1958). Particle shape and surface texture
graded clasts. The grading curves of the samples
influence the properties of freshly mixed concrete
were compared with that of the coarse aggregate, 40
more than the properties of hardened concrete. Angular
mm of NS (1994) as shown in Table 3 and Fig. 4.
The gradation lines of the samples do not lie within
Oblate biaxial the gradation zone and are coarser compared with
1.0 the standard grading. The river gravel may therefore
DISC CUBIC y
O

= be crushed and screened to obtain appropriate grading

bl
at

0.
e

95 before using as coarse aggregate.

tri
ax

0.8
ia
l

Water absorption and dry density

y
=0
0.6 .9 Water absorption ranged from 0.69 to 1.12%
Pr

Prolate biaxial
I/L = q

ol
bl

(Table 4). It is less than 3% indicating very low

at
at

e
e

tri
bl

effective porosity of clasts (BSI 1987). The aggregates

ax
ad

y=
ia
e

0.4 0 .8 having clasts with low effective porosity are strong

Pr
.33

ol

y= enough to withstand chemical decomposition against

at

0 .7
e
F= 0

bl

cement and weathering fluid.

6

ad
0.

0
1.
e

y = 0.6
F=

0.2
F= 1.5 Dry density varied from 2460 to 2680 kg/m3
F= 3.0 y = 0.5
F= BLADE y = 0.4 ROD (Table 4). This range falls into the normal density
0.0 category of aggregates (NF 1983). The density ranging
0.0 0.2 0.4 0.6 0.8 1.0
between 2000 and 3000 kg/m3 can also be considered
S/I = P
medium weight aggregate, and the samples which
Fig. 3 Shape of clasts on modified Zingg’s diagram after Lees
have dry density exceeding 2600 kg/m3 are more
(1964); the parameters are after Aschenbrenner (1959) suitable for aggregate (Zafir amd Majid 2000).

102
 Evaluation of gravel for concrete and road aggregates, Rapti River, Central Nepal Sub-Himalaya

Aggregate impact value AIV = (W3/W1) 100% (3)

Basically the aggregate impact value (AIV) is the where, W3 = weight of aggregate passing through
percentage of fines produced from the aggregate 2.36 mm sieve and W1 = initial weight of the sample.
sample after subjecting it to a standard amount of
impact (Fig. 5). The standard amount of impact was Aggregate Impact Values (AIV) below 10% are
produced by a known weight, i.e. a steel cylinder of regarded as strong, and AIV's above 35% would
14 kg, falling a set height, a prescribed number of normally be regarded as too weak for using road
times (15 times), onto an amount of aggregate of surfaces. For the samples tested, AIV ranged from
standard size (10–14 mm) and weight retained in a 14.1 to 16.1% (Table 4). These values lie in the range
mould. After impacting, the aggregate was removed of BSI (1992) where strong aggregates possess AIV
from the cylinder and sieved through a 2.36 mm between 10 and 20%. According to NS (1994), AIV
sieve. The AIV was then computed using the of aggregate which should be used for wearing
following expression (ASTM 1979): surfaces should not exceed 30% and for other
application, it should not exceed 45%. Therefore,

Table 3: Grading limits of coarse aggregate (NS 1994)

Sieve size (mm) Percentage passing for graded aggregate of nominal size
40 mm 20 mm 16 mm 12.5 mm
80 100 - - -
40 95 to 100 100 - -
20 35 to 70 95 to 100 100 100
16 - - 90 to 100 -
12 - - - 90 to 100
10 10 to 35 25 to 55 30 to 70 40 to 85
4.75 0 to 5 0 to 10 0 to 10 0 to 10

100

90

80

70
Cumulative passing %

60

50

40

30

20

10

0
1 10 100 1000

Grain size (mm)

Fig. 4 Grain size distribution of the samples from the Rapti River represented by the shaded portion compared with gradation curves
of 40 mm of NS (1994) as shown by dark curves

103
 S. Maharjan and N. K. Tamrakar/ Bulletin of the Department of Geology, Vol. 10, 2007, pp. 99–106

AIV of the studied samples lies within the standards soundness value (SSV) was obtained as below:
of both NS (1994) and BSI (1992).
SSV = {(W1–W2)/W1} . 100% (7)
Los Angeles abrasion value
Hardness (wearing property) and toughness where, W1 = initial weight of the sample and W2
(breaking property) of aggregates associated together = weight retained on 10 mm after the test.
are often carried out in Los Angeles test. The principle Minimum allowance of SSV for coarse aggregate
of the test is to obtain percent wear due to relative is 10% for concrete (ASTM 1978). The SSV of the
rubbing action between aggregates and steel balls tested samples varied from 4.46 to 7.29% (Table 4).
used as an abrasive charge. Uniform factor and wear The samples fall within specified range and are
of gravel were determined by Los Angeles test resistant against chemical weathering and frosting.
(ASTM 1989). Los Angeles abrasion value was
calculated as: EVALUATION OF GRAVEL
The Rapti River samples of gravel are dominantly
Wr100 = {(W–W100)/W}.100% (4) quartzite group with few carbonate, schist and granite
groups. Surface texture is smooth to rough. Clasts
are mostly subrounded, oblate triaxial and sphered
Wr500 = {(W–W500)/W}.100% (5)
suggesting good workability. The samples are well
where, Wr100 = %wear for 100 revolutions, Wr500 graded but do not lie within the grading zone of
= %wear for 500 revolutions, W = initial weight of standard.
sample, W100 = weight of sample retained on 1.7 mm Absortion values of the studied samples do not
after 100 revolutions and W500 = weight of sample exceed the standard value of 3% as suggested by BSI
retained on 1.7 mm after 500 revolutions. Uniformity
factor (UF) was obtained as:

UF = (W–W100)/(W–W500) (6)

UF should not exceed 0.2 for material of uniform

hardness (ASTM 1989).
Mixture of samples R5, R6 and R7 was crushed
to produce Grade A sample (Table 5) for Los Angeles
test because there was no large variation in proportion
of constituents and textures of clasts in these samples.
UF did not exceed 0.2 indicating that the hardness
of sample was uniform (Table 4). The Los Angeles
abrasion value of 500 revolutions (Wr500) was 29.83%.
This value falls within the specification of 30% of
NS (1994) for coarse concrete aggregate.

Sodium sulphate soundness value

Sodium sulphate test was carried out to test for
chemical durability of samples. Sample of size 10–14
mm and weight 425±5 grams were immersed in the
sodium sulphate solution of specific gravity 1.162
for 48 hours and were subsequently dried. The process
was repeated five times and sodium sulphate Fig. 5 Diagram showing aggregate impact test

104
 Evaluation of gravel for concrete and road aggregates, Rapti River, Central Nepal Sub-Himalaya

Table 4: Results of water absorption, dry density, AIV, Los Angeles value and SSV

Samples rdry (kg/m 3) Water absorption Samples Los Angeles Value SSV
(%) Wr 100 (%) Wr 500 (%) UF (%)
R2 2630 0.74
R3 2680 0.69 15.3 5.39
R4 2590 0.75
R5 2460 0.84
R6 2470 0.96 14.1 6.44
R7 2460 1.01 16.1 6.43
R8 2620 1.10 4.46
R9 2500 1.12 15.3 7.29
R10 2620 0.96
R11 2610 1.10
R12 2560 1.11 14.2 7.27
R13 2630 0.95
Mixture of R6, R7,
R8, R9 and R10 5.35 29.83 0.19

(1987). The samples also have medium density allowance of SSV for coarse aggregate is 10% for
varying within 2000 and 3000 kg/m 3 and are concrete (ASTM 1978). As the SSV of the test samples
considered good for medium weight concrete is less than 10%, aggregate of the Rapti River gravel
aggregates. Similarly, AIV varies between 10 and offer good resistance against chemical weathering.
20% of BSI (1992) and lies below 30% of NS (1994).
Therefore, the gravel of the Rapti River can be used CONCLUSIONS
as aggregate for heavy duty concrete floor finishes, The gravel samples from the Rapti River are
pavement wearing surfaces, sub-base, roadbase and dominantly of quartzite group with few carbonate,
for other concrete. schist and granite groups. They are mostly subrounded,
The Los Angeles abrasion value (Wr500) of the oblate triaxial ellipsoid with high sphericity and rough
test sample does not exceed 30%. The sample also to smooth surface texture. Workability of gravel is
abrades uniformly with time having UF around 0.2. good considering the textural attributes. Grading is
Based on the specification of NS (1994), the Rapti coarser compared with the Nepal standard.
River gravel is suitable for road and concrete AIV and Los Angeles values suggest that gravels
aggregates. are mechanically sound. They have normal density
Regarding chemical durability, the minimum of medium weight aggregates. Water absorption value

Table 5: Grading of test samples of coarse aggregates for Los Angeles abrasion test (ASTM 1989)

Sieve size (mm) Grade of sample (Wt. of fraction by sieving), gram

Opening (mm) Retaining (mm) Grade A Grade B Grade C Grade D
40 28 1250± 25
28 19 1250± 25
19 12.5 1250± 25 2500± 10
12.5 9.5 1250± 25 2500± 10
9.5 6.3 2500± 10
6.3 4.75 2500± 10
4.75 2.36 5000±10
Total weight 5000±25 5000±10 5000±10 5000±10

105
 S. Maharjan and N. K. Tamrakar/ Bulletin of the Department of Geology, Vol. 10, 2007, pp. 99–106

is also low and is less than the standard, 3% (0.69 to B. and Manandhar, S.P., 2006. Engineering and
1.12%). SSV falls below 10% suggesting that gravel environmental geology mapping of Hetauda and
samples are competent against frosting and chemical surrounding areas, Annual report of Department of
decomposition. The results from different tests fall Mines and Geology, v. 3, pp. 50–59.
within the specified values of standards, suggesting Howard, J.L., 1993. The statistics of counting clast in
rudites: a review with example from the Upper
that the Rapti River gravel materials are appropriate
Paleogene of southern California USA, Jour.
for concrete and road aggregates.
Sedimentology, v. 40, pp. 157–174.
ACKNOWLEDGEMENTS ISRM, 1979. Suggested methods of determining water
content, porosity, density, absorption and related
We thank Central Department of Geology, properties and swelling and slake durability index
Tribhuvan University, Kathmandu, and Soil and Rock properties: Intl. Soc. Rock Mech. Comm. on
Testing Laboratory, Nepal Electricity Authority, Standardization of Laboratory and Field Tests, Intl.
Swayambhu, Kathmandu for providing facilities. We Jour. Rock Mech. Min. Sci. and Geomech. Abstract,
also thank Raghubanshi U. and Otto for their helps v. 16, pp. 141–156.
during sampling. Janoo, V.C., 1998. Quantification of shape, angularity and
surface texture of base-coarse materials, US Army
REFERENCES Corporation of Engineers, Special Report 98-1, pp.
1–22.
Aschenbrenner, B.C., 1956. A new method of expressing Kaplan, M.F., 1958. The effects of the properties of coarse
particle sphericity. Jour. . Sed. Petrology, v. 26, pp.
aggregates on the workability of concrete, Magazine
15–72.
of Concrete Research, v. 10, pp. 63–74.
ASTM, 1978. Book of American Standard of Testing
Lees, G., 1963. A new method for determining the
Materials Standards, Part 14, Concrete and Mineral
angularity of particles, Jour. Sedimentology, v. 3, pp.
Aggregates specification C33, pp. 15–22.
2–22.
ASTM, 1989. Test for resistance to abrasion of coarse
Lees, G., 1964. The measurement of particle shape and
aggregate particle by use of the Los Angeles machine,
specification C131, American Standard of Testing its influence in engineering materials. Jour. British
Materials Standards. Granite and Whinstone Federation, London 4 and 5,
BSI, 1975. Methods of sampling and testing of mineral pp. 1–22.
aggregates, sands, fillers. BS 812, Part 1, British Neville, A.M., 1996. Properties of concrete. Fourth Edition,
Standard Institution (BSI), London. Addison Wesley Longman, Essex, England.
BSI, 1987. Code of practice for design of concrete structures NF, 1983. L’Association Francaise de Normalisation,
for retaining aqueous liquids, BS 8007, British Standard Franulates, Vocabulaire-Defenations-Classification,
Institution (BSI), London. pp. 18–101.
BSI, 1989. Description of surface texture, BS 812, Part NS, 1994. Specification for aggregate, Nepal Standard
102, British Standard Institution (BSI), London. (NS) 297–2054, U.D.C. 620, 113, NBSM, Balaju,
BSI, 1992. Specification for aggregates from natural Kathmandu, pp. 1–5.
sources of concrete, BS 882, British Standard Institution Zafir, N. and Majid, A., 2000. The influence of aggregates
(BSI), London. properties on strength of concrete, Series on K-economy,
DMG, 2004. Mineral resources of Nepal, Department of Civil and Structural Engineering Works, Malayasia,
Mines and Geology (DMG), Government of Nepal, pp. 1–22.
Lainchaur, Kathmandu, pp. 37–38. Zingg, T., 1935. Beitrage Zur Schott enanlyse Mineral,
Duvadi, A.K., Koirala, A., Rimal, L.N., Nepali, D., Piya, Schweiz, Petrog. Mitt. Schweiz, v. 15, pp. 39–140.

106