Effect of optimum compaction moisture content formulations on the strength and durability of sustainable stabilised materials

Applied Clay Science xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Applied Clay Science journal homepage: www.elsevier.com/locate/clay

Research paper

Effect of optimum compaction moisture content formulations on the strength and durability of sustainable stabilised materials ⁎

Mohamad Nidzam Rahmat , Norsalisma Ismail Faculty of Architecture, Planning and Surveying, Universiti Teknologi MARA, 40450 Shah Alam, Selangor, Malaysia

A R T I C LE I N FO

A B S T R A C T

Keywords: Moisture content Compaction Waste Stabilisation Strength Durability

The achievement of Optimum Compaction Moisture Content (OCMC) of clay soil plays an important role in compaction as well as the durability and strength of compacted soil. This is due to its effect on the structure and orientation of the clay soil particles. Most researchers on stabilised systems involving soils and/or industrial waste by-product additives for applications in roads and buildings are faced with the problem of how to approach the establishment of OCMC, when the complex mixtures involved. This paper reports on the laboratory investigation of theoretical methods of two different approaches to establish the OCMC in the stabilisation of clay soil involving multi-binary binder in cementitious binder system. Furthermore, this research also explores the use of an industrial by-product, Pulverized Fuel Ash (PFA) as partial target material and ground granulated blastfurnace slag (GGBS), with a view to reducing the reliance on the traditional cementitious binders, such as lime and/or Portland Cement (PC), in stabilising Lower Oxford Clay (LOC) soil combining with PFA at 50:50 ratio. LOC + PFA was stabilised both in conventional manner using Lime and PC as control and using sustainable binders incorporating GGBS. The results show that there was no one particular approach to the establishment of the optimal compaction moisture content for best strength development and durability. The best approach being dependent on the period of curing, stabiliser content and whether GGBS was blended with Lime or with PC. Of the various stabilisers studied, the highest strength magnitudes were however recorded with LOC-PFA stabilised using the blended binders incorporating GGBS. For all 7, 28 and 56 days of curing periods, the PC-based stabilisers were observed to be less sensitive to the different approaches to compaction moisture content, relative to the lime-based systems.

1. Introduction Soil stabilisation is an alteration of the properties of an existing soil to meet the specified engineering requirements. The main properties that may require to be altered by stabilisation are strength, volume stability, durability and permeability. Soil stabilisation is widely used in road construction to improve sub-bases and sub-grades. Several methods are available for stabilising clay soil in order to increase the strength properties and to reduce swelling or expansion behaviour. These can be achieved by the use of chemical additives, soil replacement, compaction control, moisture control and surcharge loading. Chemical stabilisation involves the formation of strong bonds between the clay minerals and other soil particles. Lime and PC are common among earlier chemical stabilisation. Regardless of the stabilisation method, the ultimate goal is to ensure adequate strength of stabilised soil (Jagendan et al., 2010; Malhotra and Sanjeev, 2013; Kartik et al., 2014). The achievement of the optimum moisture content (OMC) and maximum dry density (MDD) of a soil plays an important role in ⁎

compaction as well as in strength and durability of the compacted soil, where the properties of the soil and its performance are influenced by the molding moisture content due to its effect on the structure and orientation of clay particles. For all stabilisation work, the extent to which air can be removed depends on the strength of the soil or the friction between soil particles which in turn be subject to the moisture content of the soil during compaction. The moisture content of a soil has a major impact on how well the soil will compact and stabilised. When a soil is completely dry it will not compact to its greatest possible density because of friction between the soil particles. As the moisture content increases, the water lubricates the soil, allowing it to move more easily into a compact state and the density increases. The effect of water content on the compaction of soil can also be explained with the help of electrical double layer theory of clay particles. At low water content, the forces of attraction in the adsorbed water layer are large, and there is more resistance to movement of the particles. Stabilised material are usually compacted at the optimum moisture content when the dry density at its maximum or at the wetter side nearly saturation

Corresponding author at: Construction, Faculty of Architecture, Planning and Surveying, University Technology MARA, 40450 Shah Alam, Selangor, Malaysia. E-mail address: [email protected] (M.N. Rahmat).

https://doi.org/10.1016/j.clay.2018.02.036 Received 5 December 2017; Received in revised form 19 February 2018; Accepted 25 February 2018 0169-1317/ © 2018 Elsevier B.V. All rights reserved.

Please cite this article as: Rahmat, M.N., Applied Clay Science (2018), https://doi.org/10.1016/j.clay.2018.02.036

Applied Clay Science xxx (xxxx) xxx–xxx

M.N. Rahmat, N. Ismail

line (Whitlow, 2001). For a given compactive effort, higher strength can be achieved by compacting the raw material at its optimum moisture content to ensure that the maximum dry density is achieved. The complex interaction between stabilising agent and water makes the relationship between water, strength and durability much harder to predict (Beckett and Ciancio, 2014). On the other hand, in recent years much research has been directed to the identification and investigation of wide range of new pozzolanic materials from industrial and agricultural source such as metakaolin, fly ash, red mud, rice husk ash, wheat straw ash or by latently hydraulic materials. The partial replacement of cementitious material, lime and cement by pozzolanic materials results in the effective reduction in cost of production. Therefore there are significant numbers of research projects on the application of lime or cement blended binders in soil stabilisation, which offer sustainability advantages. Researchers now believe that with the addition of small amount Lime or PC, the calcium present causes an ionic exchange, which results in flocculation that has dramatic effect on modification of soil's workability and strength. Stabilisation transpire as the reaction between silica and alumina within the clay structure, lime and water to form calcium silicate hydrates, calcium-aluminate-hydrates and calcium-alumino-silicate-hydrates (C-S-H, C-A-H and C-A-S-H) to bind the structure together. Regardless of stabilisation method, the ultimate goal is to ensure adequate final density and strength of the soil. The point of achieving optimum moisture content and maximum dry density of the soil plays an important role in compaction. For nearly all soil, the extent to which air can be removed depends on the strength of the clay lumps or friction between the granular particles which in turn depend on the moisture content of the soil during compaction (Barnes, 2000). The potential of using by-products such as fly ash and blastfurnace slag (GGBS) is promising, well establish, and has been investigated by several researchers. These by-products can incorporated in cementitious material to modify and improve certain properties and also to conserve non-renewable natural resources (Seco et al., 2011; Oti et al., 2014; Adam and Maria, 2015; Marcelino-Sadaba et al., 2017; Shalabi et al., 2017). GGBS is commonly activated by PC. Water hydration of PC produces mainly calcium hydroxide (Ca(OH)2) and C-S-H gel. In the hydration of blended PC-GGBS system, although minor amount of alkalis released, GGBS is mainly activated by the hydration products Ca (OH)2 (Bijen, 1996). PC-GGBS hydration produces slightly different hydrates from those formed by hydration of standard PC, as the main reaction products of GGBS hydration are C-S-H gel, C-A-H gel and small amount of calcium hydroxide (Higgins et al., 1998). Therefore PC-GGBS system is not significantly different from either the PC or activated GGBS system. It is believed that there is a gap of published work directly referring to the effect of optimum compaction water content on the strength and durability of stabilised material. Therefore this paper aims to address this vagueness by determining the effect of optimum compaction moisture content using two different formulations.

Table 1 Oxide composition of LOC, PFA, PC, Lime and GGBS. Oxide

LOCa (%)

PFAb (%)

PCc (%)

Limed (%)

GGBSe (%)

SiO2 Al2O3 CaO Fe2O3 MgO SO4 K2O Mn2O Na2O3 TiO2 FeO P2O5 CaCO3 Loss on Ignition

46.73 18.51 6.15 6.21 1.13 1.29 4.06 0.07 ≤0.52 1.13 ≤0.80 ≤0.17 – 15.79

47.6 26.2 2.4 9.4 1.42 0.86 3.02 – 1.1 – – – – –

20 6 63 3 4.21 2.3 – 1.11 – – – – – –

≤0.9 ≤0.15 95.9 ≤0.07 ≤0.46 – – – – – – – 2.2 –

35.34 11.59 41.99 0.35 8.04 0.23 – 0.45 – – – – – –

a b c d e

Hanson Brick Ltd. Ash Resources. Lafarge Cement Ltd. Calch Ty Mawr lime. Civil and Marine Slag Cement Ltd.

2.1.2. Pulverized fuel ash Pulverized fuel ash (PFA) is a by-product of thermal power plants resulting from the combustion of pulverized coal in the coal-fired furnaces. It is commonly available in the United Kingdom. The PFA used in this research was supplied by the United Kingdom Quality Ash Association (UKQAA). 2.1.3. Stabilisers Portland Cement (PC) was supplied by Lafarge Cement Ltd. UK. The PC was used as an alternative to lime and as the activator to Ground Granulated Blastfurnace Slag (GGBS) in order to stabilised the target materials. Quicklime (CaO) was supplied by Breacon Calch Tý Mawr Lime, Wales, UK. CaO is denser, less dusty and more effective as a stabiliser having a higher available lime content per unit mass where 3% CaO is normally equivalent of 4% hydrated lime. GGBS which is readily available throughout UK was supplied by Civil and Marine Slag Cement Ltd. at Llanwern, Newport, Wales. It has a latently hydraulic material that occurs as by-products of steel industry when molten slag is rapidly cooled and granulated. The use of GGBS as a cementitious material blended with PC or CaO are based on its activation with alkalis (mainly Ca(OH)2) released from hydration of PC. Table 1 shows the oxide composition of target materials and stabilisers used in this research. 2.2. Preparation of specimens and testing procedures Prior to sample preparation, target values of maximum dry density (MDD) and optimum moisture content (OMC) were established using Proctor compaction test in accordance with BS 1377 (1990) Part 2. In all systems, LOC-PFA + Lime, LOC-PFA + PC, LOC-PFA + Lime:GGBS and LOC-PFA + PC:GGBS, the MDD ranges from 1.42–1.48 Mg/m3 and OMC ranges from 19 to 24%. Therefore, for the test, the MDD of 1.42 Mg/m3 and OMC value of 24% were adopted for all specimen mixes. All specimens were expected, within experimental error, to be of approximately comparable bulk-density, since the volume was maintained as 50 mm diameter and 100 mm height was established to be about 380 g. Two approaches for establishing the compaction water content were used as presented in Table 2. Using these parameters, two formulations (F1 and F2 at OMC and 1.2OMC) were used to calculate the amount of water. Mix Design Composition are summarised in Table 3. The dry materials were mixed thoroughly before adding the precalculated amount of water. A steel mould fitted with collar, was used to compact the material into a cylinder using hydraulic jack. After

2. Experimental procedures 2.1. Materials 2.1.1. Lower Oxford Clay (LOC) LOC used in this study is currently used in the manufacture of fired clays. The characteristics and mineralogy of LOC is well established. It was supplied by Hanson Bricks Ltd., from their clay-mine at Stewartby, Bedfordshire. The clay is grey in colour and is known to have high sulfate and sulfide contents. Mineralogy studies by Hanson Brick Ltd. has established that LOC contains illite (23%), kaolinite (10%), chlorite (7%), calcite (10%), quartz (29%), gypsum (2%), pyrite (4%), feldspar (8%) and organics (7%).

2

Applied Clay Science xxx (xxxx) xxx–xxx

M.N. Rahmat, N. Ismail

fully soaked of the cured specimens in water for 4 days. The specimens were individually soaked in separate perspex containers, covered with a lid fitted with dial gauge for measuring the linear expansion (Fig. 1). Two cylinders per mix proportion were subjected to soaking process, and the mean UCS strength determined for each mix composition. The control specimens were also cured for 7, 28 and 56 days at room temperature and the UCS were determined without the cured specimens being subjected to soaking in water. The durability index (DI) was obtained by determining the compressive strength of soaked specimens as a percentage of compressive strength of control specimens. [Durability Index DI = (UCS value for Soaked Specimen/UCS value for Control Specimen) × 100%].

Table 2 Mix design composition of stabilised LOC-PFA. Target materials

LOC-PFA (50:50)

Stabiliser

Blending ratio (%)

Curing period (days)

Stabiliser dosage (%)

Lime PC Lime-GGBS PC-GGBS

100 100 30:70 40:60

7, 7, 7, 7,

10, 10, 10, 10,

28, 28, 28, 28,

56 56 56 56

20 20 20 20

Table 3 Formulation approach to optimum compaction moisture content. Formula

⁎

F1 F2

⁎⁎

Target material (T), (gm) T T

Stabiliser (s), (gm)

+ +

sT sT

Water (w), (gm) (@OMC, @1.2OMC) + +

w(T + sT) wT

Total weight (gm) = =

3. Results and discussion

380 380

3.1. Unconfined compressive strength of the LOC-PFA in lime stabiliser system

⁎ F1; the calculation of water (w) was made based on the total amount of soil and stabiliser. ⁎⁎ F2; the calculation of water (w) was made based on the amount of soil only.

Figs. 2 (a) and (b) show the effects of different methods of calculating the moisture content [(F1: OMC, 1.2OMC) and (F2: OMC, 1.2OMC)] on the UCS of LOC-PFA (50:50) as target materials, when stabilised with lime alone and with Lime-GGBS blended stabilisers at 30:70 ratios, at 10% and 20% stabiliser dosages. Three specimens were tested for each blended mixtures and the mean strength value taken. Fig. 2(a) shows the effects when 10% stabiliser level was used for stabilisation. At 7 days of curing, F1@OMC recorded the highest strength, with a UCS value of 1430 kN/m2, with the lime-GGBS 30:70 ratio. At this curing stage the lowest strength value was shown by formula [email protected] using lime only. This formula also indicates lowest strength values with lime-GGBS stabilisers. At 28 days of curing, [email protected] gives in highest strength value of 2591 kN/m2 when stabilised with lime-GGBS at (30:70) ratio. The lowest strength values

compaction, the cylinders were extruded using steel plunger, trimmed and wrapped in cling film, labelled and cured in controlled temperature at 20 ± 1 °C and 100% relative humidity. This humidity was ensured by storing the cling film wrapped specimen in a sealed polythene bag and the placed in a sealed plastic container. Unconfined Compressive Strength (UCS) of stabilised test specimens was determined using a Hounsfield testing machine capable of loading up to 10kN at a compressive rate of 1 mm/min. Durability test consists of durability index test, change in weight upon soaking and linear expansion were also investigated. The specimen were prepared in the same process as described for UCS test and were cured for 7, 28 and 56 days, followed by

100 mm

50 mm

Fig. 1. Soil cylinder specimen, glass and perspex containers for soaking specimens and hounsfield testing machine.

3

Applied Clay Science xxx (xxxx) xxx–xxx

M.N. Rahmat, N. Ismail

4500

LOC-PFA (50:50) + 10% Lime stabiliser

(a)

F1@OMC

F2@OMC

[email protected]

4500

[email protected]

F1@OMC 4000

3500

3500

3000

3000

UCS Test (kN/m2)

UCS Test (kN/m2)

4000

2500 2000 1500

F2@OMC

[email protected]

[email protected]

2500 2000 1500

1000

1000

500

500

0

0 7 days

7 days

28 days

28 days

56 days

56 days

7 days

7 days

28 days

28 days

56 days

56 days

Lime

Lime-GGBS (30:70)

Lime

Lime-GGBS (30:70)

Lime

Lime-GGBS (30:70)

PC

PC-GGBS (40:60)

PC

PC-GGBS (40:60)

PC

PC-GGBS (40:60)

LOC-PFA (50:50) + 20% Lime stabiliser

(b)

4500

4500

F1@OMC

F2@OMC

[email protected]

[email protected]

4000

4000

3500

3500

UCS Test (kN/m2)

UCS Test (kN/m2)

LOC-PFA (50:50) + 10% PC stabiliser

(c)

3000 2500 2000 1500

LOC-PFA (50:50) + 20% PC stabiliser

(d)

F1@OMC

F2@OMC

[email protected]

[email protected]

3000 2500 2000 1500

1000

1000

500

500 0

0 7 days

7 days

28 days

28 days

56 days

56 days

7 days

7 days

28 days

28 days

56 days

56 days

Lime

Lime-GGBS (30:70)

Lime

Lime-GGBS (30:70)

Lime

Lime-GGBS (30:70)

PC

PC-GGBS (40:60)

PC

PC-GGBS (40:60)

PC

PC-GGBS (40:60)

Fig. 2. Unconfined compressive strength of LOC-PFA (50:50) stabilised with (a) 10% lime stabiliser (b) 20% lime stabiliser (c) 10% PC stabiliser and (d) 20% PC stabiliser.

Mysteriously, the strength generally decreased from 28 days to 56 days of curing, at both stabiliser dosages for all methods of calculating the initial compaction water input. For lime only, the results illustrate that the strength development deteriorated more significantly with increasing stabiliser dosage from 10% to 20%. There are little variation in strength results between F1 and F2, OMC and 1.2OMC.

again were observed when lime only was used to stabilise LOC-PFA, at all different formulae. The lowest strength was 839 kN/m2 when F2 was used at OMC. Within this system at 56 days of curing, with all formulae, it shows most of the highest UCS values occurred when LOC-PFA was stabilised with lime-GGBS (30:70) compared to lime only. It reveals that the highest strength development was achieved at 2689 kN/m2 when LOC-PFA was stabilised with lime-GGBS (30:70) using formula F2@OMC. At this curing stage the lowest strength value was 1316 kN/ m2 with formula [email protected] when LOC-PFA stabilised with lime only. Fig. 2(b) illustrates the strength development of lime system when 20% stabiliser dosage was used. At 7 days of curing, LOC-PFA stabilised with lime-GGBS (30:70) indicated the highest strength values with a highest strength of 2015 kN/m2 recorded when F2@OMC was used. Meanwhile for the same F2 at OMC, the test specimens totally collapsed with no strength gain, when lime only was used as the stabiliser. The strength at 7 days with 20% lime was detrimental compared to when 10% lime was used. At 28 days of curing, LOC-PFA stabilised with lime only continued to show the lowest strength values with all formulae compared to other stabilisers, with the lowest strength value of 90 kN/ m2 being observed when formula F2@OMC was used. Incorporating GGBS at 30:70 ratios tremendously increased the strength in the system for all formulae. However, the highest strength values were obtained with lime-GGBS at 30:70, where [email protected] recorded the highest strength value of 3208 kN/m2. At 56 days curing period, results showed that the strength decreased from 28 days to 56 days with all methods of calculating compaction water input. The highest strength value was achieved at 2,765 kN/m2 when LOC-PFA was stabilised with lime-GGBS (30:70) using formula F1@OMC. LOC-PFA stabilised with lime only continued to show the lowest strength values with all formulae, with the lowest strength value at 49 kN/m2 when formula F2@OMC was used. The overall results show that there is little or no increase in strength with increasing stabiliser dosage from 10% to 20%.

3.2. Unconfined compressive strength of the LOC-PFA in PC stabiliser system Figs. 2(c) and (d) illustrate the UCS development of LOC-PFA (50:50) when stabilised with PC alone and PC-GGBS at 40:60 ratios with 10% and 20% stabiliser dosage for the different methods of calculating the compaction moisture content. Like in the lime system, the overall results show that there is increase in strength with increasing stabiliser from 10% to 20% and with increased curing period from 7 to 28 to 56 days for all methods of calculating the compaction moisture content. However, compared with the lime-GGBS system, the strength magnitudes are lower with the PC-GGBS system at both curing periods. With 10% stabiliser dosage, LOC-PFA stabilised with PC alone (Fig. 2(c)) shows higher strength increases compared to the blended stabiliser PC-GGBS. At 7 days of curing, the highest strength was observed with PC using Formula F1@OMC at 954 kN/m2. Formula F2@ OMC gives the lowest strength value of 451 kN/m2 when PC-GGBS (40:60) was used. At 28 days of curing [email protected] indicates the lowest strength value of 900 kN/m2 when stabilised with PC-GGBS (60:40) ratio. While the highest strength of 2399 kN/m2 was recorded with PC only with formula F1@OMC. [email protected] recorded the highest 56-day strength, with a UCS value of 2813 kN/m2, when LOC-PFA was stabilised with PC only. At this curing stage the lowest strength value was shown by Formula F1@OMC using blended PC-GGBS (40:60) was used, with a strength value of 1707 kN/m2. Fig. 2(d) illustrates the 4

Applied Clay Science xxx (xxxx) xxx–xxx

M.N. Rahmat, N. Ismail

the blended lime-GGBS (30:70) for stabilise action of the LOC-PFA, using formula FI@OMC. For 28 days curing, there are small differences in DI reading for both stabilisers with all formulae of calculating compaction moisture content, from 75% to 111% which, is the highest percentage was recorded with lime-GGBS (30:70) with formula F1@ 1.2OMC and the lowest with same stabiliser but this time with formula F2@OMC. For 56 days of curing, there are obvious differences in DI when lime only was used to stabilised LOC-PFA mixture, compared to DI at 7 and 28 days. At this stage with lime only, formulae F2@OMC and [email protected] showed higher DI value compared to formulae FI@ OMC and [email protected]. Highest DI was recorded at 1329% with formulae F2@OMC (see Fig. 3 (b)). However, when blended lime-GGBS (30:70) was used to stabilise LOC-PFA mixture, results showed there was slightly increased in DI when compared to DI at 7 and 28 days, with all formulae of calculating compaction moisture content. There are little variable beween F1 and F2, OMC and 1.2OMC except for lime only at 56 days.

strength development when 20% stabiliser dosage was used in the PC system. There are increases in strength development upon increasing the stabiliser dosage to 20% at both 7 and 28 days with all formulae. At 7 days of curing, F2@OMC shows the highest strength value of 1591 kN/m2 with PC only. Meanwhile, Formula F1 (1.2 OMC) indicates the lowest strength values when LOC-PFA was stabilised with PC-GGBS (60:40) (1036 kN/m2). At 28 days of curing, [email protected] shows the highest strength value of 2754 kN/m2 with PC only and F1@OMC stabilised with PC:GGBS (40:60) indicates the lowest strength at 1712 kN/m2. At 56 days of curing, again LOC-PFA stabilised with PC alone shows higher strength increment compared to blended PC-GGBS stabiliser. The highest strength was observed with Formula F1@ 1.2OMC at 3862 kN/m2. Formula [email protected] gives the lowest strength value of 2948 kN/m2 when PC-GGBS (40:60) was used. Overall the results show that there are increases in strength development with increasing in curing period with all methods of calculating the initial compaction water input, with the PC only stabiliser showing best performance. Overall, at 20% stabiliser level, LOC-PFA stabilised with PC only demonstrates highest strength improvement with all formulae compared to the blended stabilisers. In general, increased amount of stabiliser dosage resulted in increased UCS with increasing curing period for LOC-PFA target materials, except for when lime alone was used as stabiliser. Low strength values were observed with both formulae for material compacted at the OMC, probably due to good compaction but incomplete hydration. Research by Nidzam and Kinuthia (2011) has attributed the variability in strength of stabilised soil mixtures especially at high lime dosages, incomplete hydration of lime due to lack of adequate water. Overall results therefore show that lime alone is not suitable for use as stabiliser. For prolonged curing up to 56 days, the highest UCS values in LOC-PFA were recorded in the system stabilised using PC at 20% stabiliser dosage, using formula F1 (1.2OMC). The strength developments patterns were in closely similar for both PC-GGBS (40:60) and limeGGBS (30:70). There are little differences in strength development in general between the two formulae.

3.4. Durability index (DI) of the LOC-PFA in PC stabiliser system Fig. 3 (c) and (d) shows the durability index of the LOC-PFA when stabilised with PC only and with PC-GGBS (40,60). At 10% of stabiliser dosage (Fig. 3 (c)), for 7 days curing, the overall result shows that LOC-PFA stabilised with PC only recorded higher DI values compared to LOC-PFA stabilised with the blended PC-GGBS stabiliser at (40:60) ratio, except when formula F2 was applied at 1.2OMC. This formula recorded the highest DI value at 253% with the blended PC-GGBS (40:60) stabiliser. The lowest DI value at this stage was 40% recorded when LOC-PFA was stabilised with blended PCGGBS (40:60) using formula F2@OMC. There are changes in the DI pattern for 28 days curing compared to 7 days. In general, results show that a higher DI value was recorded when LOC-PFA was stabilised with the blended PC-GGBS (40:60) stabiliser using all formulae compared to PC only except for day 7, with the highest DI value of 182% using formula [email protected] and the lowest was 61% when LOC-PFA was stabilised using formula F2@OMC. Similar DI pattern was observed for 56 days of curing as for 28 days. However, curing for 56 days demonstrated overall lower DI value compared to 28 days curing with both stabilisers and with all formulae of calculating compaction moisture content. At this stage, the highest DI value at 100% was recorded when PC-GGBS (40,60) was used to stabilise the LOC-PFA mixture, using formula [email protected]. This formula also recorded the lowest DI value at 53% when LOC-PFA was stabilised with PC only using formula F2@ 1.2OMC to calculate the amount of water. At 20% stabiliser dosage (Fig. 3(d)), for 7 days of curing, the overall results showed, that when PC only was used to stabilise the LOC-PFA blend, higher DI values was observed when compared to the use of the blended stabiliser PC-GGBS (40:60), with all formulae of calculating compaction moisture content. At this stage, the highest DI value (201%) was recorded when PC only was used with formula [email protected] and the lowest was recorded at 67% when LOC-PFA was stabilised with PCGGBS (40:60) with formula FI@OMC. For 28 days of curing, in general, PC-GGBS (40:60) recorded higher DI value when compared to PC only with all formulae for calculating compaction water content. The highest DI value (140%) was recorded with PC-GGBS (40:60) using formula FI@OMC and the lowest value was 80% when stabilised with PC only with formula F2@OMC. As with the 10% stabiliser dosage, at 20% and 56 days curing, the overall DI value was lowered when compared to DI values for both 7 and 28 days of curing with all formulae. For 56 days of curing, significantly lower in DI values were observed relative to those observed for 28 days, with both stabilisers with all formulae. Formula [email protected] demonstrated the highest DI value at 96% with PC-GGBS (40:60), whereas the lowest DI was 53% observed when LOC-PFA was stabilised with PC only, using the formulae with least water (F2@ OMC). Marginally higher DI at 1.2OMC compared to OMC with both formulae, F1 and F2.

3.3. Durability index(DI) of the LOC-PFA in lime stabiliser system Fig. 3 (a) and (b) shows the durability index of the LOC-PFA mixture when stabilised with lime only and the lime-GGBS (30:70) blended at 10% stabiliser. Specimens were cured for 7, 28 and 56 days and then were fully soaked in the water for 4 days. The overall results show that at 10% stabiliser dosage, there is a slight increase in the DI patterns from 7 days to 56 days of curing, with all of the compaction formulae used. In general, the DI of the stabilised LOC-PFA mixture using lime only demonstrates higher values of DI when compared to the values for LOC-PFA mixture stabilised with the blended lime-GGBS (30:70) stabiliser at all ages. For 7 days of curing at 10% stabiliser dosage (Fig. 3(a)), the highest DI value achieved was 159% when the LOC-PFA mixture was stabilised with lime only, using formula [email protected]. The lowest reading was 85% when the LOC-PFA mixture was stabilised with lime-GGBS at (30:70) using formula FI@OMC. For 28 days of curing, formula FI@ OMC recorded the highest DI value at 173% when LOC-PFA stabilised with lime only. As happened for the 7 days curing, the lowest DI value was recorded (75%) when LOC-PFA was stabilised with blended limeGGBS (30:70) using formula FI@OMC. For 56 days of curing, the highest DI value was again recorded when the LOC-PFA was stabilised with lime only using formula [email protected] at 198%, and the lowest DI value was 94%, observed with blended lime-GGBS at (30:70) ratio using formula [email protected]. At 20% stabiliser dosage (Fig. 3(b)), for 7 days of curing, specimens with both formula F2@OMC and [email protected] collapsed when lime only was used to stabilise the LOC-PFA mixture. This was probably due to the high lime dosage, which lead to incomplete lime hydration due to lack of adequate water. The highest DI value was 142% achieved with 5

Applied Clay Science xxx (xxxx) xxx–xxx

M.N. Rahmat, N. Ismail

LOC-PFA (50:50) + 10% Lime stabiliser

(a) 1400

F1@OMC

F2@OMC

[email protected]

F1@OMC

[email protected]

1200

Durability Index (%)

Durability Index (%)

F2@OMC

[email protected]

[email protected]

250

1000 800 600 400 200

200 150 100 50

0 7 days

7 days

28 days

28 days

56 days

56 days

Lime

Lime-GGBS (30:70)

Lime

Lime-GGBS (30:70)

Lime

Lime-GGBS (30:70)

0

LOC-PFA (50:50) + 20% Lime stabiliser

(b)

7 days

7 days

28 days

28 days

56 days

56 days

PC

PC-GGBS (40:60)

PC

PC-GGBS (40:60)

PC

PC-GGBS (40:60)

LOC-PFA (50:50) + 20% PC stabiliser

(d) 300

1400 F1@OMC

F1@OMC

F2@OMC

1200 [email protected]

F2@OMC

[email protected]

[email protected]

250

[email protected]

Durability Index (%)

1000

Durability Index (%)

LOC-PFA (50:50) + 10% PC stabiliser

(c) 300

800 600 400

200

150

100

50

200 0 7 days

7 days

28 days

28 days

56 days

56 days

Lime

Lime-GGBS (30:70)

Lime

Lime-GGBS (30:70)

Lime

Lime-GGBS (30:70)

0 7 days

7 days

28 days

28 days

56 days

56 days

PC

PC-GGBS (40:60)

PC

PC-GGBS (40:60)

PC

PC-GGBS (40:60)

Fig. 3. Durability Index of LOC-PFA (50:50) stabilised with (a) 10% lime stabiliser (b) 20% lime stabiliser (c) 10% PC stabiliser and (d) 20% PC stabiliser systems.

compared to all formulae of calculating compaction moisture content when LOC-PFA was stabilised with blended stabiliser lime-GGBS (30:70) for 7 days of curing(Fig. 4(d)). For 28 days of curing, lime only recorded higher weight increase compared to lime-GGBS (30:70) with all compacting formula, whereas 56 day illustrate that there are decrease in weight gain for both stabilisers with all compacting formula. The highest reading was 11% when lime only stabilised LOC-PFA with F2@OMC and lowest was 4% recorded by both stabilisers with formula F1@OMC.

3.5. Change in weight upon soaking of the LOC-PFA in lime stabiliser system Figs. 4(a) to (d) illustrate the percentage of weight increase for stabilised specimens for the lime-based system of stabilised LOC-PFA at 10% and 20% of stabiliser dosages. Stabilised specimens were cured for 7, 28 and 56 days, before each being fully soaked in the water for 4 days. Each specimen was then weighed at day 4, before the UCS test was carried out. At 10% stabiliser dosage, using lime only (Fig. 4(a)) for 28 days curing there were slightly an increased in the percent weight gain due to fully soaked in water for both formulae (FI@OMC and [email protected]) compared to 7 days of curing. All formulae show decreased in percent weight gain for 56 days when compared to percent increase for 28 days of curing. For the blended lime-GGBS (30:70) (Fig. 4(b)) all the specimens showed decreases in percent weight increase for 28 days of curing compared to percent weight increase for 7 days of curing. With limeGGBS (30:70) all formulae showed no weight increase from 28 days to 56 days of curing. At 20% stabiliser dosage Fig. 4(c) and (d), overall results showed there were higher weight increase compared to the one at 10% dosage with both stabilisers (Lime only and lime-GGBS (30:70)) used at all method of calculating compaction moisture content. For 7 days curing, it illustrates those specimens with lime only stabilised LOC-PFA with 2 formulae of calculating compaction water content was totally collapsed (F2@OMC and [email protected]). On the other hand formulae FI@OMC and [email protected] have showed higher percent weight increase

3.6. Change in weight upon soaking of the LOC-PFA in PC stabiliser system Figs. 5(a) and (b) demonstrate the pattern of weight increased for LOC-PFA stabilised with PC only and blended PC-GGBS (40:60) at 10% stabiliser dosage with all formulae of calculating compaction water content. Formula with less water content F2@OMC illustrated higher percent weight increase compared to others formulae at all curing period, except for 7 days curing, which demonstrated the lowest percent weight increase at 2%. The highest reading for 7 days was when LOC-PFA stabilised with blended PC-GGBS (40:60) with formula F2@ OMC (Fig. 5(b)). For 28 days curing, the highest weight increased was 8% with PC-GGBS (40:60) with formula F2@OMC. The lowest was 2% with formula [email protected] with PC-GGBS at (40:60) ratio. For 56 days curing, the overall results showed, blended PC-GGBS (40:60) stabilised LOC-PFA demonstrated lower percent weight increase compared to PC stabilised LOC-PFA with all formulae of calculating compaction 6

Applied Clay Science xxx (xxxx) xxx–xxx

M.N. Rahmat, N. Ismail

(a)

LOC-PFA + Lime at (10%)

(b)

F1@OMC

F1@OMC 35

F2@OMC

30

[email protected]

25

[email protected]

Change in Weight (%)

Change in Weight (%)

LOC-PFA + L-GGBS (30:70) at (10%)

40

40

20 15

35

F2@OMC

30

[email protected]

20 15

10

10

5

5

0 7 days

2 8 days

[email protected]

25

0 7 days

5 6 days

2 8 days

Curing period

(c) 40 35

LOC-PFA + Lime at (20%)

(d)

F1@OMC

F1@OMC

F2@OMC [email protected] [email protected]

25 20 15 10 5 0 7 days

LOC-PFA + L-GGBS (30:70) at (20%)

40

Change in Weight (%)

Change in Weight (%)

30

5 6 days

Curing period

35

F2@OMC

30

[email protected]

25

[email protected]

20 15 10 5

2 8 days

0 7 days

5 6 days

Curing period

2 8 days

5 6 days

Curing period

Fig. 4. Change in weight upon soaking of LOC-PFA stabiliser with lime stabiliser system (a) 10% Lime (b) 10% Lime-GGBS (30:70) (c) 20% Lime (d) 20% Lime:GGBS (30:70).

4 days soaking period has been used. LOC-PFA stabilised with lime on its own has the highest expansion compared to PC and other blended stabilisers, especially for 28 days of curing. In general, all stabilised LOC-PFA specimens in both systems, with all method of calculating compaction moisture content either attained terminal linear expansion or continued to expand at a negligible rate of increase. In both lime and PC systems, formulae with less moisture content (e.g F2) indicated the most expansion rate. Overall formula [email protected] has less expansion rate compared to others. Addition of stabiliser dosage from 10% to 20% is not necessary beneficial, for both lime and PC systems, and for both blended and unblended systems. The system using lime is more sensitive to stabiliser dosage. This perhaps due to the sensitivity of lime-stabilisation of sulphate bearing clay soil. Even with the very robust lime-GGBS stabiliser, increase in stabiliser dosage is detrimental. Moist-curing up to 56 days does not completely eliminate the risk to expansion. This is an interesting observation. Perhaps some property of cured material-porosity, brittleness, carbonation etc. is at play. There is no consistent trend to differentiate the two formulae. However formula F2@OMC appears to predominate the high expansions for the lime system at all curing stages. This is followed by F2@ 1.2OMC. On the other hand, formula F2@OMC and F1@OMC are the most expansive for the PC system. Formula [email protected] appears most stable for both systems. Those results appear to confirm the commonly held view that compaction on the wet side of the OMC is preferable, in order to eliminate swelling of compacted clay soil in both stabilised and unstabilised states. For both lime and PC systems, blending with GGBS is beneficial.The results of weight gain corroborate those of linear expansion.

moisture content. The highest reading recorded was 8% when LOC-PFA stabilised with PC only with formula F2@OMC, and the lowest was 2% with blended PC-GGBS at (40:60) ratio with formula [email protected]. At 20% of stabiliser dosage Fig. 5(c) and (d), like in the lime system and PC based system the formula with less water content F2@OMC illustrated higher percent weight increase compared to others formulae at all curing period. For 7 days curing, highest increased in weight was 8% with both stabilisers with formula F2@OMC. The lowest was 2% when LOC-PFA was stabilised with PC only with formula [email protected]. For 28 days, formula F2@OMC indicated highest weight increase at 9% when PC was used as the stabiliser. [email protected] recorded the lowest reading at 2% when LOC-PFA was stabilised with blended PC-GGBS at 40:60 ratio (Fig. 5(d)). For 56 days, the highest weight increased was 10% with formula F2@OMC when LOC-PFA was stabilised with PC only and the lowest was 3% with formula [email protected] when blended PCGGBS (40:60) was used to stabilise LOC-PFA. In general, in contrast with the lime system, in this PC based system the percent weight increase are lower with both stabilisers at 20% stabiliser dosage compared to at 10% dosage with all formulae of calculating compaction moisture content, except for formula [email protected] with PC only. 3.7. Linear expansion of the LOC-PFA system Figs. 6 (a) to (f) illustrate the linear expansion of the LOC-PFA stabilised with the lime-based system and PC based system at 10% and 20% of stabiliser dosage. Specimens were cured for 7, 28 and 56 days days at room temperature prior to fully soaking in water for 4 days, two specimens were tested for each type of stabiliser and each formula of calculating compaction moisture content at OMC and 1.2OMC, the linear expansion was monitored on a daily basis during the 4 days of soaking. In the figures, the expansion magnitudes at the end of 7

Applied Clay Science xxx (xxxx) xxx–xxx

M.N. Rahmat, N. Ismail

(a)

(b)

LOC-PFA + PC at (10%) F1@OMC

F2@OMC

[email protected]

30

[email protected]

25

F1@OMC

35

F2@OMC

Change in Weight (%)

Change in Weight (%)

35

20 15 10 5

30

[email protected]

25

[email protected]

20 15 10 5

0 7 days

2 8 days

0 7 days

5 6 days

Curing period

(c)

LOC-PFA + PC at (20%)

(d) 40

F1@OMC

30 25

F2@OMC

35

F1(1.2OMC)

30

Change in Weight (%)

35

F2(1.2OMC)

20 15 10 5 0 7 days

2 8 days

5 6 days

Curing period

40

Change in Weight (%)

LOC-PFA + PC-GGBS (40:60) at (10%)

40

40

25

LOC-PFA + PC-GGBS (40:60) at (20%) F1@OMC F2@OMC [email protected] MC [email protected] MC

20 15 10 5

2 8 days

0 7 days

5 6 days

2 8 days

5 6 days

Curing period

Curing period

Fig. 5. Change in weight upon soaking of LOC-PFA stabiliser with lime stabiliser system (a) 10% PC (b) 10% PC-GGBS (30:70) (c) 20% PC (d) 20% PC:GGBS (30:70).

4. Discussion and conclusions

significant reduction in lime consumption. From the extensive laboratory work reported in this study, the following conclusion may be drawn:

From the observation of the current research, it is hypothesised that, at 10% dosage, approaches with more water (F1) would be expected to be associated with an increased amount of water available within the voids of the test specimens, relative to those specimens stabilised with 20% stabiliser. However, by raising the level of stabiliser from 10% to 20%, results indicate that those formulae with higher moisture content achieved better strength. This explains the currently common practice of compacting soils on the wet side of OMC is a good practice, not only for addressing evaporation losses and reduced cracking/expansion, but also for chemical hydration in stabilised systems. The higher strength development with the wetter mixes at prolonged curing to 28 days is probably due to unhindered formation of a relatively higher amount of C-S-H gel (Higgins, 2005). Reduced strength with the formula F2 or with both formulae @OMC formula (as opposed to F1, or F1 or F2 @ 1.2OMC) is probably due to poor compaction, and/or inhibited hydration, both effects due to reduced water content. The blended stabilisers had better strength development magnitudes in general compared to stabilisation with Lime or PC on their own, indicating that GGBS has a high potential as a partial replacement material for these traditional stabilisers in soil stabilisation for applications in building and general construction. This is beneficial since GGBS has environmental benefits relative to lime or cement, as a byproduct material. Its manufacture involves less energy and CO2 emissions, both of which are associated with the negative climatologically influence of the manufacture of both PC and lime (Higgins, 2005). Research by some researchers (Obuzor et al., 2011; Oti et al., 2008; Nidzam and Kinuthia, 2011; Nidzam and Kinuthia, 2010; Higgins, 2005; Wild et al., 1999) has reported that use of GGBS results in

1. In relation to the optimal compaction of stabilised soils, there are differences of varying significance between approaches that consider all the dry components in the mix, in establishing the compaction water content, and approaches that only consider the bulk target materials for this purpose. The significance of these differences will depend on, possibly among other variables, stabiliser type and dosage, degree of compaction and time of curing. However, whichever method is used, the designer must bear in mind that: a. If early strength will be key to the desired level of stabilisation, then care must be taken to ensure that for the stabiliser dosages used, there is no excess water in the system. This is however likely to result in unused raw stabiliser residues. For this reason, further research work is recommended for the purpose of establishing the possibility and effects of the late addition of water to re-start the hydration of any excess stabiliser so as to optimize performance. b. If early strength is unimportant, the results presented here suggest that the designer must ensure that the material is well hydrated with a sufficient quantity of water as any approach that results in inadequate water in the system is likely to result in low residual strength and with excess raw and unreacted stabiliser. 2. For both 7 and 28 days of curing periods, the PC-based stabilisers were less sensitive to the different approaches to establishing the compaction moisture content relative to the lime-based systems. This most probably due to the differences in hydration mechanisms, with PC being a hydraulic material and self-hydrates relatively more 8

Applied Clay Science xxx (xxxx) xxx–xxx

M.N. Rahmat, N. Ismail

LOC-PFA (50:50) + Lime stabiliser system (7 days curing - linear expansion at day 4 of soaking)

(a)

(d) 0.16

0.14

F1@OMC

F2@OMC

0.12

[email protected]

[email protected]

Linear Expansion (%)

Linear Expansion (%)

0.16

0.10 0.08 0.06 0.04

0.14

F1@OMC

F2@OMC

0.12

[email protected]

[email protected]

0.10 0.08 0.06 0.04 0.02

0.02

0.00

0.00 Lime (10wt%)

L-GGBS(30:70) (10wt%)

Lime (20wt%)

PC (10wt%)

L-GGBS(30:70) (20wt%)

LOC-PFA (50:50) + Lime stabiliser systerm (28 days curing - linear expansion at day 4 of soaking)

(b)

(e)

PC-GGBS(40:60) (10wt%)

PC (20wt%)

PC-GGBS(40:60) (20wt%)

LOC-PFA (50:50) + PC stabiliser system (28 days curing - linear expansion on day 4 of soaking)

0.16

1.00 0.80

F2@OMC [email protected]

Linear Expansion (%)

F1@OMC [email protected]

0.90

Linear Expansion (%)

LOC-PFA (50:50) + PC stabiliser system (7 days curing - linear expansion on day 4 of soaking)

0.70 0.60 0.50 0.40 0.30

0.14

F1@OMC

F2@OMC

0.12

[email protected]

[email protected]

0.10 0.08 0.06 0.04

0.20 0.02

0.10

0.00

0.00 Lime (10wt%)

(c)

L-GGBS(30:70) (10wt%)

Lime (20wt%)

PC (10wt%)

L-GGBS(30:70) (20wt%)

LOC-PFA (50:50) + Lime stabiliser system (56 days curing - linear expansion at day 4 of soaking)

(f)

F1 OMC F1(1.2OMC)

F2 OMC F2(1.2OMC)

0.14

Linear Expansion (%)

Linear Expansion %

0.12

PC (20wt%)

PC-GGBS(40:60) (20wt%)

LOC-PFA (50:50) + PC stabiliser system (56 days curing - linear expansion on day 4 of soaking)

0.16

0.16 0.14

PC-GGBS(40:60) (10wt%)

0.10 0.08 0.06 0.04 0.02

0.12

F1 OMC F1(1.2OMC)

F2 OMC F2(1.2OMC)

0.10 0.08 0.06 0.04 0.02

0.00 Lime (10wt%)

L-GGBS(30:70) (10wt%)

Lime (20wt%)

0.00

L-GGBS(30:70) (20wt%)

PC (10wt%)

PC-GGBS(40:60) (10wt%)

PC (20wt%)

PC-GGBS(40:60) (20wt%)

Fig. 6. Linear expansion at 4 days soaking of LOC-PFA (50:50) (a-c) Lime stabiliser system and (d-f) PC stabiliser system.

rapidly in the presence of water, while lime undergoes much slower pozzolanic reactions with either clay soils and/or GGBS during hydration. 3. There are technological as well as environmental advantages that are likely to accrue from the use of Lime-GGBS and PC-GGBS blended binders in soil stabilisation for general civil engineering construction such as roads, embankments and building components such as blocks and bricks. As GGBS is an industrial by-product material, there may also be economic advantages, depending on the proximity of the sources for GGBS.

strength of cement-stabilised rammed earth materials. Can. Geotech. J. 51, 583–590. Dx.doi.org https://doi.org/10.1139/cgi-2013-0339. Bijen, J.G., 1996. Blast Furnace Slag Cement. Association of the Netherlands Cement Industry (VNV) the Netherlands. Higgins, D.D., 2005. Soil Stabilisation with Ground Granulated Blastfurnace Slag. Report for UK Cementatitious Slag Makers Association (CSMA), UK. Higgins, D.D., Kinuthia, J.M., Wild, S., 1998. “Soil stabilisation using lime activated GGBS”. Proceeding of the 6th CANMET/ACI International Conference on Fly Ash, Silica Fume. In: Malhotra, V.M. (Ed.), Slag and Natural Pozzolans in Concrete. Vol 2. pp. 1057–1074 (Bangkok 31 May – 5 June). Jagendan, S., Liska, M., Osma, A.A., Al-Tabaa, A., 2010. Sustainable binders for soil stabilisation. Proc. Institution Civil Engineers ICE – Ground Inprov. 163 (1), 53–61. Kartik, S., Ashok, Kumar E., Gowtham, P., Elango, G., Gikul, D., Thangaraj, S., 2014. Soil stabilisation by using fly ash. IOSR J. Mech. and Civil Eng. (IOSR-JMCE) 10 (6), 20–26 (e-ISSN: 2278-1684, p-ISSN: 2320-334X). Malhotra, M., Sanjeev, Naval, 2013. Stabilization of expansive soils using low cost materials. Int. J. Eng. Innov. Technol. (IJEIT) 2 (11), 181–184. Marcelino-Sadaba, Sara, Kinuthia, John, Oti, Jonathan, Meneses, Andres Seco, 2017. Challenges in life cycle assessment (LCA) of stabilised clay-based construction materials. Appl. Clay Sci. 144, 121–130. Nidzam, R.M., Kinuthia, J.M., 2010. “Sustainable soil stabilisation with blastfurnace slag (ggbs) – a review.” Proceeding of the Institute of Civil Engineers (ICE). J. Construct. Mater. Vol. 163 (CM3), 157–165 (10.1680COMA-2010.163.3.157). Nidzam, R.M., Kinuthia, J.M., 2011. Compaction of fills involving stabilisation of expansive soils. In: Proceeding of the Institute of Civil Engineers (ICE); Geotechnical Engineering. 164. pp. 113–126. http://dx.doi.org/10.1680/geng.2011.164.2.113. GE2. Obuzor, G.N., Kinuthia, J.M., Robinson, R., B, R., 2011. Enhancing the durability of flooded low-capacity soils by utilizing lime-Aactivated ground granulated Blastfurnace slag (GGBS). Eng. Geol. 123 (3), 179–186. Oti, J.E., Kinuthia, J.M., Bai, J., 2008. Using Slag for Unfired-Clay Masonry-Bricks. Proceedings of the Institution of Civil Engineers (ICE). J. Construct. Mater. 161 (CM4), 147–155. http://dx.doi.org/10.1680/coma.2008.161.4.147. Oti, J.E., Kinuthia, J.M., Robinson, R.B., 2014. The development of unfired building material using brick dust waste and Mercia mudstone clay. Appl. Clay Sci. 102, 148–154. Seco, A., Ramirez, F., Miqueleiz, L., Garcia, B., 2011. Stabilisation of expansive soil for the

Acknowledgement The authors wish to thanks Universiti Teknologi MARA (UiTM) Shah Alam, Malaysia for sponsoring this research. Sincere thanks also go to Professor John Kinuthia and the technical staff of the Faculty of Computing, Science and Technology, University of South Wales UK for their technical support and the use of laboratory facilities. The Cementitious Slag Makers Association, Buxton Lime Industries Ltd., and Hanson Brick Ltd., for providing research materials such as GGBS, Lime and Lower Oxford Clay respectively. References Adam, Joe M., Maria, Rajesh A., 2015. Soil stabilisation using industrial waste and lime. Int. J. Sci. Res. Engineer. & Technol. 4 (7), 799–805 (IJSRET) ISSN 2278–0882). Barnes, G.E., 2000. Soil Mechanic Principals and Practice, 2nd Edition. Pelgrave, New York (ISBN 0-333-77776-X). Beckett, Christopher, Ciancio, Daniela, 2014. Effect of compaction water content on the

9

Applied Clay Science xxx (xxxx) xxx–xxx

M.N. Rahmat, N. Ismail

38109-6). Wild, S., Kinuthia, J.M., Jones, G.I., Higgins, D.D., 1999. Suppression of swelling associated with ettringite formation in lime-stabilised sulfate-bearing clay soils by partial substitution of lime with ground granulated blastfurnace slag (GGBS). Eng. Geol. 51, 257–277 (ISSN 0013-7952).

use in construction. Appl. Clay Sci. 51, 348–352. Shalabi, Faisal I., Asi, Ibrahim M., Qasrawi, Hisham Y., 2017. Effect of by-product steel slag on the engineering properties of clay soil. J. King Saud Univ. – Eng. and Sci. 29, 394–399. Whitlow, R., 2001. Basic Soil Mechanic, 4th Edition. Prentice Hall, England (0582-

10