The summary of the result of the geotechnical properties of the natural soil is presented in the Table 1.
laboratory tests.
Fig. 1. Al O -SiO -Fe O
ternary plot for the soil sample
International Journal of Scientific & Engineering Research, Volume 5, Issue 6, June-2014 631
ISSN 2229-5518
Plasticity, Strength and Permeability of
Reclaimed Asphalt Pavement and Lateritic Soil
Blends
I.I. Akinwumi
Abstract— This paper presents the results of laboratory evaluation of the effects of the addition of reclaimed asphalt pavement (RAP), to an A-2 lateritic soil, on the plasticity, strength and permeability of the soil. The natural soil was classified as A-2-6(1), according to AASHTO classification system. RAP was added to the soil in 0, 4, 8 and 12%, by dry weight of the soil. Specific gravity, Atterberg limits, compaction, California bearing ratio (CBR), unconfined compression and permeability tests were conducted on each of the soil-RAP blends. Results obtained show that as RAP content in the blend increased, the plasticity index, optimum moisture content, maximum dry unit weight, swell potential, unconfined compressive strength and permeability decreased while the specific gravity, soaked and unsoaked California bearing ratios increased. These results indicate that RAP effectively improved, especially, the plasticity and permeability of the soil. It also indicates that deformation should be a major design criterion while planning the use of lateritic soil-RAP blend as a road pavement layer material.
Index Terms— laterite, recycled asphalt, soil modification, subgrade, sub-base, tropical soil, unbound granular materials
—————————— ——————————
N today’s world, issues of reusing and recycling of non- renewable resources as a means of minimizing waste and environmental pollution; and for economic reasons are a priority to governments, organizations, researchers and the general public. The road construction industry, being a large consumer of non-renewable natural resources, is not left out. Solutions for reuse and recycle of wastes resulting from earthmoving, construction, rehabilitation and maintenance
operations of road infrastructure are now being favoured [1].
The rehabilitation of hot mix asphalt (HMA) roadways that
results into milling of the existing asphalt concrete surfacing
and resurfacing of the roadway with new HMA accounts for a
large quantity or stockpiling of reclaimed asphalt pavement
(RAP) annually, worldwide. It is estimated that about 100 mil-
lion tons of asphalt pavement is reclaimed each year in the
United States [2]. Though some government agencies respon-
sible for making policies that relate to highway and transpor-
tation permit a certain percentage of RAP to be recycled into
new HMA, most of these stockpiling of milled asphalt ends up being disposed off as waste, especially in developing coun- tries.
McGarrah [3] reported that transportation agencies of gov- ernments in the US received many request from road con- struction contractors seeking permission to use RAP as a base course and sub-base materials and this led to various research works and field testing [4], [5], [6], to investigate the proper- ties and performance of wholly using RAP as a base course material. The properties of RAP made it worth considering for use as a pavement layer material [7], [8]. The use of RAP as a base material increases the rutting resistance and stability; and decreases temperature susceptibility of pavements [9]. However, Taha et al. [10] found out that blending of RAP with
————————————————
virgin aggregates gave more satisfactory results than wholly using RAP. Their results showed that the dry density and Cal- ifornia bearing ratio (CBR) values decreased; the optimum moisture content (OMC) was unchanged; and the permeability increased, as the RAP in the blends increased. They recom- mended limiting the substitution of natural aggregate with RAP to 10% for road base applications. Taha et al. [11] dis- couraged the use of 100% RAP as a base material.
Bennert and Maher [12] investigated the effects of blending RAP with virgin aggregates on the mechanical properties of these blends for use as base course and sub-base materials. They found out that as RAP increased in the blend, the CBR and permeability of the RAP-virgin aggregate blends de- creased.
In the tropics, lateritic soils occupy about 23 percent of the land surface and are the single most extensive kind of soil in this region [13]. Consequently, their selection for use as a con- struction material becomes an economically-viable choice. However, some of the lateritic soils are unsuitable for use as road construction materials because their properties do not comply with existing standard requirements. Some of these soils exhibit high plasticity, poor workability, low strength, high permeability, tendency to retain moisture and high natu- ral moisture content [14].
Mustapha et al. [15] compared the CBR and unconfined compressive strength (UCS) of an A-6 lateritic soil and that of a mixture of the soil and 60% RAP, by weight of soil. These researchers found out that adding 60% RAP to the A-6 lateritic soil slightly increased its CBR and UCS. Edeh et al. [16] deter- mined the plasticity and CBR of a ternary blend of a lateritic soil, RAP and Cement.
This paper presents the results of laboratory evaluation of the
effects of adding RAP to an A-2 lateritic soil on its plasticity,
strength and permeability.
IJSER © 2014 http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 5, Issue 6, June-2014 632
ISSN 2229-5518
The RAP used for this research work was collected from a stockpile along Benin-Ore Road, during its roadway rehabili- tation. Sieve analysis was performed to determine the grada- tion of the RAP. About 4% of its particles are finer than BS No.
200 Sieve (0.075 mm). The RAP sample contained an average bitumen content of 4.8% by weight of the mix. A typical bitu- men content range for RAP obtained from wearing surface mixes is 4.5-6.0% [9].
The soil sample was collected, by method of bulk disturbed sampling, from Covenant University borrow pit (latitude
06o40′24"N and longitude 03o 09′12"E) behind the university student hostels. A top soil layer of 0.2 m was removed before digging to a depth of 0.5 m to collect the sample. Samples to be used for natural moisture content determination were stored in a watertight bag before this laboratory determina- tion. The remaining samples were air-dried in the Soil Me- chanics/Geotechnics laboratory of the Department of Civil Engineering, Covenant University, before carrying out the
The summary of the result of the geotechnical properties of the natural soil is presented in the Table 1.
laboratory tests.
Fig. 1. Al O -SiO -Fe O
ternary plot for the soil sample
Gradation and specific gravity tests were conducted on the soil and RAP samples. The oxide composition of the soil sam-
2 3 2 2 3
Table. Geotechnical Properties of Natural Soil
ple was determined using atomic absorption spectrophotome- ter. RAP was added to the soil in 0, 4, 8 and 12%, by dry
weight of the soil. Specific gravity, Atterberg limits, compac- tion, CBR, unconfined compression and permeability tests were conducted on each of the soil-RAP mixtures. The proce- dures for the various tests were carried out in accordance with BSI [17], [18].
Pearson’s correlation coefficient (r) was used to identify the
Gravel (>4.75 mm), % 12.5
Sand (0.075 - 4.75 mm), % 58.4
Silt and Clay (<0.075 mm), % 29.1
AASHTO Soil Classification System A-2-6 (1) SC - Clayey
extent to which the values of each of these engineering proper-
ties are correlated with the increase in RAP content in the soil-
RAP blends. The probability that the effects of changes in each
of these engineering properties with the addition of RAP are
not due to just chance alone was determined based on pre-
specified probability threshold (p-value) to a significance level
of 5% (0.05).
Unified Soil Classification System
Sand
In order to determine the level/extent of laterization of the soil used, the concentration of oxides of silica (SiO2 ), iron (Fe2 O3 ) and aluminum (Al2 O3 ) in the soil was determined by atomic absorption spectroscopy. A ternary or tri-plot of this composi- tion is shown in Fig. 1. The soil contains a higher silica content than iron oxide content. This suggests that this soil was formed from laterite on an acidic rock and it contains some quartz. The soil also contains a higher proportion of sesquiox- ide of aluminum than the sesquioxide of iron. Thus, the soil is bauxitic. The ratio of silica-sesquioxides was determined to be
1.35. Thus, confirming that the soil is lateritic. According to the Schellmann [19] scheme of classification of weathering prod- ucts, this soil sample was classified as being taken from a kao- linized profile.
Unsoaked CBR (%) 56
Soaked CBR (%) 29
Unconfined Compressive Strength
(kN/m2) 130.4
The soil has a natural moisture content of 15.7% and it is clas- sified as A-2-6(1), according to the AASHTO soil classification
IJSER © 2014 http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 5, Issue 6, June-2014 633
ISSN 2229-5518
system. The particle size distribution curves for the soil and
RAP aggregate are shown in Fig. 2.
Fig. 2. Particle size distribution of natural soil and RAP
It showed that the percentage passing BS No. 200 sieve (0.075 mm) is 29.1% and according to AASHTO, it is a granu- lar soil since it is less than 35%. Since this fine content is 29.1%, it indicates that it considerably influenced the properties of the soil. The soil was in its plastic state at the time of collection. Its plasticity index was found to be greater than 11% and thus according to AASHTO, the fines are clayey. The activity of the soil was determined to be 1.08 and using the table for activity of clay-rich soils provided by [20], this soil can be described as normal.
The specific gravity of the soil sample and RAP is 2.54 and
2.93, respectively. Variation of the specific gravity of soil with
RAP content is presented in Fig. 3. There is a positive correla-
tion between the percent of RAP added to the soil and the spe-
cific gravity of the blend.
The specific gravity of the natural soil increased by 3.9% af-
ter adding 12% RAP to the soil. The increase in specific gravi-
ties of the soil-RAP blends as the RAP content increased was
strongly correlated, r = 0.983, p = 0.017. The p-value obtained
indicates that there is moderate evidence against the null hy-
pothesis (no difference between the specific gravity of the nat-
ural soil and those of the soil-RAP blends).
The changes in the liquid, plastic and shrinkage limits; and
the plasticity index of the lateritic soil sample, as various per-
centages of RAP were blended with it, are presented in Fig. 4.
There is a negative correlation between the plasticity indices of
the soil-RAP blends and the addition of RAP to the soil.
The plasticity index of the natural soil decreased by 42.3%
after adding 12% RAP to the soil. This decrease in the plastici-
ty indices of the soil-RAP blends with increasing RAP content
was strongly correlated, r = −0.976, p = 0.024.
Fig. 3. Variation of specific gravity with RAP content
The p-value obtained indicates that there is moderate evidence against the null hypothesis (no difference between the plastici- ty index of the natural soil and those of the soil-RAP blends). This decrease in the plasticity index makes the soil-RAP blend more workable. After the addition of 12% RAP, the linear shrinkage of the natural soil increased by 95%. This shows that deformation of the soil increases with increasing RAP content in the soil.
Fig. 4. Variation of Atterberg limits with RAP content
Fig. 5 is the plasticity chart showing the variation of plastic- ity index with liquid limit. It shows that the natural soil pro- gressively changed from being clay of low plasticity (CL) to silt of low plasticity (ML). This change is attributed to the ag-
IJSER © 2014 http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 5, Issue 6, June-2014 634
ISSN 2229-5518
glomeration of clay particles facilitated by the bituminous
(binder) content of the RAP.
60
0 % RAP
50 4 % RAP
hypothesis (no difference between the OMC of the natural soil and those of the soil-RAP blends).
The decrease in OMC may also be resulting from the clumping-together of clay particles, which reduce the surface area of the soil and its water-holding capacity. The clumped clay particles begin to behave like silt-sized particles; which is
40
30
20
10
CL - ML
0
8 % RAP
12 % RAP
CL
ML & OL
CH
MH & OH
corroborated by Fig. 5. The coarser the grain of a soil becomes, the lesser the water it requires to reach optimum [21], [22], [23].
Variation of unsoaked and soaked CBR with the addition of RAP to the soil is shown in Fig. 7. There is a positive corre- lation between the unsoaked CBR value and the percent of RAP added to the soil. This positive correlation was found to be strong, r = 0.996, p= 0.004. The p-value obtained indicates that there is strong evidence against the null hypothesis (no difference between the unsoaked CBR of the natural soil and
0 20 40 60 80 100
Liquid limit (%)
Fig. 5. Plasticity chart showing the variation of the soil plastici- ty with its RAP content
Variation of OMC and maximum dry unit weight with the addition of RAP to the soil is shown in Fig. 6. After adding
12% RAP to the soil, its maximum dry unit weight and OMC decreased by 2.5% and 10.3%, respectively. The negative corre- lations of each of the maximum dry unit weight and OMC with the addition of RAP to the soil were strong, r = −0.961, p
= 0.039; and r = −0.933, p = 0.067, respectively.
Fig. 6. Variation of compaction characteristics with RAP con- tent
The p-value for the maximum dry unit weight indicates that there is moderate evidence against the null hypothesis (no difference between the maximum dry unit weight of the natu- ral soil and those of the soil-RAP blends) while that for the OMC indicates that there is weak evidence against the null
those of the soil-RAP blends). After the addition of 12% RAP
to the soil, the unsoaked CBR of the soil increased by 14.3%.
This increase is slight. Thus, the unsoaked CBR of the soil is generally low.
Fig. 7. Variation of CBR with RAP content
The increase in soaked CBR of the soil-RAP blends as the RAP content increased was moderately correlated, r =
0.303, p = 0.697. The p-value obtained indicates that there is no evidence against the null hypothesis (no difference be- tween the soaked CBR of the natural soil and those of the soil-RAP blends). This means that the soaked CBR results appear to be consistent with the null hypothesis. After the addition of 12% RAP to the soil, the soaked CBR of the soil increased by 10.3%. Though the addition of RAP did not appreciably increase the soaked CBR of the natural soil, which is less than 30%, the soaked CBR became greater than 30% for all the blends. Consequently, the natural soil that only met TRL [24] requirements for use as a subgrade material became suitable for use as a sub-base material.
The swell potential of the natural soil decreased with in- crease in RAP content, as can be seen in Fig. 8. The decrease in swell potential of the soil-RAP blends as the RAP con-
IJSER © 2014 http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 5, Issue 6, June-2014 635
ISSN 2229-5518
tent increased was strongly correlated, r = −0.954, p=0.046. The p-value obtained indicates that there is moderate evi- dence against the null hypothesis (no difference between the swell potential of the natural soil and those of the soil- RAP blends). After the addition of 12% RAP to the soil, it decreased by 44.6%. The swell potential of the soil was generally low suggesting that kaolinite is the predominant clay mineral in the soil and this is in alignment with its classification in Fig. 1.
Fig. 8. Variation of swell potential with RAP content
The unconfined compressive strength (UCS) of the soil and that of the soil-RAP blends are presented in Fig. 9. The de- crease in UCS of the soil with an increase in its RAP content was strongly correlated, r = −0.881, p = 0.119. The p-value ob- tained indicates that there is no significant evidence against the null hypothesis (no difference between the UCS of the nat- ural soil and those of the soil-RAP blend). The UCS of the soil decreased by 88.4% after adding 12% RAP to the soil. This result affirms that increasing RAP content in lateritic soil-RAP blends can cause substantial irrecoverable deformation. Ben- nert and Maher (2005) also found out that high RAP content in RAP-virgin aggregate blends caused large permanent defor- mation.
The variation of permeability with the addition of RAP to the soil is shown in Fig. 10. There is a negative correlation be- tween the percent of RAP added to the soil and the permeabil- ity of the soil. The decrease in permeability of the soil with an increase in its RAP content was strongly correlated, r = −1.000, p < 0.001. The p-value obtained indicates that there is very strong evidence against the null hypothesis (no difference be- tween the permeability of the natural soil and those of the soil- RAP blends). Using the results of specific gravity and compac- tion characteristics, the void ratio and porosity of the natural soil and soil-RAP mixtures were calculated. The results ob- tained showed that the void ratio and porosity both progres- sively decreased with increasing RAP content. This decrease in void ratio and porosity may be attributed to the clogging of the pores within the soil by the bituminous content of the
RAP. This may be responsible for the decrease in permeability with increasing RAP content in the soil.
Fig. 9. Variation of UCS with RAP content
Fig. 10. Variation of permeability with RAP content
From the results obtained, the following conclusions were made:
(i) RAP, when added to a lateritic soil, can be used to re- duce the plasticity and swell potential of its clay frac-
tion. Thus, making the soil more workable.
(ii) The addition of RAP to the lateritic soil did not signif- icantly improve its strength.
(iii) The addition of RAP to the lateritic soil reduced the ease with which water permeates the soil and its moisture-holding capacity. This makes it suitable for reducing the permeability of soils to be used as road pavement layer material and earth dam material.
IJSER © 2014 http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 5, Issue 6, June-2014 636
ISSN 2229-5518
(iv) Deformation should be a major design criterion while planning the use of lateritic soil-RAP blend for use as a pavement layer material.
(v) The natural soil, prior to application of RAP, was only suitable for use as subgrade/fill material for flexible
pavement construction, according to TRL [24]. This is because its plasticity index is greater than 12% and its soaked CBR is less than 30%. However, upon applica- tion of 4, 8, and 12% RAP, the plasticity index and soaked CBR became less than 12% and greater than
30%, respectively; making the lateritic soil-RAP
blends suitable for use as sub-base material.
The author would like to acknowledge the contribution of Miss Winifred Umane towards the samples collection and la- boratory investigations carried out during this research work.
[1] P. Danzanvilliers, Building sustainable roads: A voluntary commitment of road con- struction and maintenance companies, Association for European Transport and Contributors, 2009 [Cited 28 May, 2014]. Available at:
<http://abstracts.aetransport.org/paper/download/id/3136>.
[2] PAPA, Paving the way, Pennsylvania Asphalt Pavement Association, April-June
2010 [Cited 28 May, 2014]. Available at:
<www.pahotmix.org/Paving_The_Way/PAPA_AprMayJun10.pdf>.
[3] E.J. McGarrah, Evaluation of current practices of reclaimed asphalt pavement/virgin aggregate as base course material, Report No. WA-RD 713.1, Washington State Department of Transportation, Olympia, WA, 2007 [Cited 28 May, 2014]. Available at: <www.wsdot.wa.gov/research/reports/fullreports/713.1.pdf>.
[4] S.M. Sayed, J.M. Pulsifer, and R.C. Schmitt, “Construction and performance of shoulders using UNRAP base”, Journal of Materials in Civil Engineering, 5(3),
321–338, 1993, doi: 10.1061/(ASCE)0899-1561(1993)5:3(321).
[5] N. Garg, and M.R. Thompson, “Lincoln Avenue reclaimed asphalt pavement base project”, Transportation Research Record, 1547, 89-95, 1996, doi:
10.3141/1547-13.
[6] B.D. Trzebiatowski, and C.H. Benson, Saturated hydraulic conductivity of com- pacted recycled asphalt pavement. Geotechnical Testing Journal 28(5): 514-
519, 2005, doi: 10.1520/GTJ12698.
[7] J.A.C. MacGregor, W.H. Highter, and D.J. DeGroot,. “Structural numbers for reclaimed asphalt pavement base and subbase course mixes”, Transportation Research Record, 1687, 22–28, 1999, doi: 10.3141/1687-03.
[8] S.M. Sayed, J.M. Pulsifer, and N.M. Jackson, “UNRAP: Are we ready for it?”
Journal of Materials in Civil Engineering, 23(2): 188–196, 2011, doi:
10.1061/(ASCE)MT.1943-5533.0000157.
[9] P.T. Sherwood, Alternate materials in road construction: A guide to the use of recycled and secondary aggregates, second edition, Thomas Telford, London, 2001.
[10] R. Taha, G. Ali, A. Basma, and O. Al-Turk, “Evaluation of reclaimed asphalt pavement aggregate in road bases and subbases”, Transportation Research Rec- ord, 1652, 264-269, 1999, doi: 10.3141/1652-33.
[11] R. Taha, A. Al-Harthy, K. Al-Shamsi, and M. Al-Zubeidi, Cement stabilization of reclaimed asphalt pavement aggregate for road bases and subbases, Journal of Materials in Civil Engineering, 14(3), 239–245, 2002, doi: 10.1061/(ASCE)0899-
1561(2002)14:3(239).
[12] T. Bennert, and A. Maher, The development of performance specification for granular base and subbase material, 2005 [Cited 28 May, 2014]. Available at: < http://www.state.nj.us/transportation/refdata/research/reports/FHWA- NJ-2005-003.pdf>.
[13] F.H. Beinroth, H. Eswaran, F. Palmieri, and P.F. Reich, Properties, classification and management of oxisols, Soil management collaborative research support program, Washington DC, 1996.
[14] R. Maignien, Survey of research on laterites, United Nations Educational, Scientific
and Cultural Organization, Paris, 1964.
[15] A.M. Mustapha, R. Jibrin, N.M. Etsuworo, and M. Alhassan, “Stabilization of A-6 lateritic soil using cold reclaimed asphalt pavement”, International Journal of Engineering and Technology, 4(1), 52–57, 2014.
[16] J.E. Edeh, A.O. Eberemu, and O. Agnes, “Lateritic soil stabilization of reclaimed asphalt pavement as flexible highway pavement materials”, Advanced Materi- als Research, 367, 3–11, 2012, doi: 10.4028www.scientific.net/AMR.367.3.
[17] BSI, Methods of test for soils for civil engineering purposes, British Standards Institu- tion, BS1377, London, 1990a.
[18] BSI, Stabilised materials for civil engineering purposes: General requirements, sampling, sample preparation and on materials before stabilisation, British Standards Institu- tion, BS 1924: Part 1, London, 1990b.
[19] W. Schellmann, A new definition of laterite. Geological Survey of India Memoir,
120, 1-7, 1986.
[20] M. Budhu, Soil mechanics and foundations, third edition, John Wiley & Sons, Inc., New Jersey, 2011.
[21] I.I. Akinwumi, “Soil modification by the application of steel slag”, Periodica
Polytechnica Civil Engineering, Submitted for publication.
[22] I.I. Akinwumi, Utilization of Steel Slag for the Stabilization of a Lateritic Soil, LAP Lambert Academic Publishing, GmbH & Co. KG, Saarbrücken, 2013.
[23] I.I. Akinwumi, J.B. Adeyeri, and O.A. Ejohwomu, “Effects of steel slag addition on the plasticity, strength and permeability of lateritic soil”, Proceedings of Se- cond International Conference of Sustainable Design, Engineering and Construction, Texas, 457-464, 2012, doi: 10.1061/9780784412688.055.
[24] TRL, Overseas Road Note 31: A guide to the structural design of bitumen-surfaced roads in tropical and sub-tropical countries, Transport Research Laboratory (TRL), Berkshire, 1993.
IJSER © 2014 http://www.ijser.org