Available online at www.sciencedirect.

com

Procedia - Social and Behavioral Sciences 53 (2012) 1001 – 1010

SIIV - 5th International Congress - Sustainability of Road Infrastructures

The use of lightweight materials in road embankment construction

Abstract

Construction overload on high compressive soils can cause significant settlements as to require the design of an
appropriate soil treatment, including load compensation, aimed to hold the stress-strain state present in the soil
when undisturbed. Lightweight materials, both natural and/or derived from industrial process, can be used to
solve the problem of replacement of large volumes of material. This paper presents some preliminary results
obtained from comparative tests on both natural lightweight materials and geofoam samples. Results focus on the
possibility to reach adequate performance, allowing lightweight materials to be used for innovative applications
such as airport pavement subgrades and embankments.

© 2012
© 2012 The
The Authors.
Authors. Published
Published by
by Elsevier
Elsevier Ltd.
Ltd. Selection and/or peer-review under responsibility of SIIV2012
Scientific
Selection Committee Open access
and/or peer-review underunder CC BY-NC-ND
responsibility license. Scientific Committee
of SIIV2012
Keywords: Settlement;Lightweight Aggregates;Gyratory Compactor;Geofoam;Deflectometric Tests

1. Introduction

The settlement of the embankment base is one of the critical problems encountered during construction of the
road structure, often causing severe damage on the overlying layers of the pavement structure. Every types of soil
are subjected to settlement due to compression loads with, consequently, reduction of volume filled in the ground.
The evaluation of settlement evolution should be carried out during the design process with the aim to limit the
amount of residual settlement occurring at the end of construction. Allowable settlement depends on the
importance of the infrastructure, the pavement type (flexible, semi-rigid or rigid), the risk to have differential
settlement etc. In Italy the Technical Specifications provided by the CIRS “Centro Sperimentale Interuniversitario
di Ricerca Stradale” (2001) state that residual settlement, at the end of construction, should remain under of 10%
of the total estimated settlement and never higher than 5 cm. Ground settlement can be defined as a void

* Corresponding author. Tel.: +39-050-2217745; fax:+39-050-2217762.

1877-0428 © 2012 The Authors. Published by Elsevier Ltd. Selection and/or peer-review under responsibility of SIIV2012 Scientific Committee
Open access under CC BY-NC-ND license. doi:10.1016/j.sbspro.2012.09.949
1002 Marradi A. et al. / Procedia - Social and Behavioral Sciences 53 (2012) 1001 – 1010

reduction, due to a load application, which forces the particles closer together and change the orientation of the
structural framework (Budhu M. 2010). Referring to the design height of a road embankment surface, settlement
should be considered as the amount of vertical deformations that occurs immediately due to the elastic settlement
of the fill mass or foundation soil, for consolidation and secondary compression of the foundation soil and long-
term creep of the fill mass. Settlement caused by lateral deformation of the foundation soil at the edge of an
embankment is generally considered small if compared to the four previous mentioned settlement mechanisms
(Terzaghi et al 1996). All the components of total settlement can be explained as follow:

S total S i  S p  S s  S cf S p  Ss | S p  Ss (1)

where:
Si immediate or elastic settlement of the soil;
Sp end-of-primary consolidation of the soil;
Ss secondary consolidation of the soil;
Scf long-term vertical deformation (creep) of the soil.

As reported by Stark D.T. et al (2004), the total amount of post-construction settlement of 0.3÷0.6 m during
the economic life of a roadway are generally considered tolerable provided that the settlements are uniform, occur
slowly over a period of time and do not occur next to a pile-supported structure. If post-construction settlement
occurs over a long period of time, any pavement distress can be repaired when the pavement is resurfaced.
Although rigid pavements have performed well after 0.3÷0.6 m of uniform settlement, flexible pavements are
usually selected where doubts exist about the uniformity of post-construction settlements. Generally, flexible
pavements are selected when predicted settlement exceed 0.15 m (NCHRP Synthesis of Highway Practice 29).
In Italy Technical Specifications provided by CIRS suggest that when predicted settlement exceeds 0.15 m,
contractor has to monitor the time evolution by placing control instrument above the embankment. When the total
amount of predicted settlement reaches values not complying with the constructing structure, is necessary to
reduce their effects on embankment stability and on pavement evenness. There are many suitable solutions to
prevent excessive settlement of the embankment foundation depending on type and thickness of materials to be
reclaimed. In general, cohesionless soils can be treated improving density by means of mechanical action (roller
compaction, drop forge treatment, vibroflot etc.) while with cohesions soils is necessary to force the water to
drive out the soil reducing the consolidation path (drainage and overload). More recently a new possibility
emerged to reduce, or in some cases stop, the settlement of the ground plane, consisting in the lightening of the
embankment weight using lightweight aggregates, both natural or coming from industrial process, following the
load compensation principle. The situation where lightweight fill may be appropriate include conditions where
the construction schedule does not allow the use of staged construction, where existing utilities or adjacent
structures are present and cannot tolerate the magnitude of settlement induced by placement of typical fill, and at
locations where post-construction settlements may be excessive under conventional fill. Lightweight can consist
of a variety of materials including expanded polystyrene blocks (EPS – geofoam), lightweight aggregate like
(expanded clay, pumice of other type of effusive rocks), crumb rubber tire and foam glass aggregates. In this
study the mechanical characteristics of natural aggregates and geofoam have been analysed with the aim to
investigate their response to (reproduced) actual loads coming from road surface. The paper presents different
assessments regarding natural (pumice) and industrial (geofoam) lightweight materials. Natural lightweight
material performances have been evaluated analysing a mixture of pumice aggregates properties and also the
relative densification due to the post-compaction process. The use of Geofoam is actually considered a suitable
alternative to natural aggregates to be used for construction of embankments standing on soft subgrade
(Montepara et al 2001). Geofoam offers the possibility to have a very light structure avoiding the most part of the
problem related to the settlements of embankment foundation. In this paper a preliminary mechanical
characterization has been performed. Results hereinafter presented regard the performance behavior evidenced
inducing different stress/strain levels in order to reproduce the embankment weight and traffic loads.
 Marradi A. et al. / Procedia - Social and Behavioral Sciences 53 (2012) 1001 – 1010 1003

2. Experimental Results

2.1. Natural lightweight aggregates

Pumice aggregates come from rapid cooling of lava outside of the earth’s shell (effusive rocks). The ones used
in this study are quarried in the central region of Italy in different grain-size fractions, by the grinding of bigger
pumice blocks. All the physical and mechanical characteristics, carried out by means of AASHTO compaction
test and Californian Bearing Ratio, are reported in Table 1 and Table 2 (tests executed following standards UNI
EN 1097-6, UNI EN 1936).
Table 1 Physical characteristics

Grain size fraction Bulk density ρa Real Density ρr Water absorption WA24 Porosity
(mm) (Mg/m3) (Mg/m3) (%) (%)
0/6.3 1.208 2.55 105 52.63
0/4 1.206 2.55 104 52.71
1/12.5 1.212 2.55 108 52.47
3/7 1.254 2.55 112 50.82
7/12 1.276 2.55 115 49.96
15/20 1.302 2.55 128 48.94

Table 2 Mechanical characteristics

Grain size fraction Dry density ϒa Optimum moisture CBR dry CBR wet
(mm) (Mg/m3) content W% (%) (%)
0/6.3 0.757 9.0 63 60
0/4 0.768 9.3 65 62
1/12.5 0.713 9.5 60 58
3/7 0.665 9.8 58 57
7/12 0.603 9.8 56 54
15/20 0.550 10 54 53

Results reported in Table 1 and Table 2 show that no excessive differences can be underlined between the
physical characteristics of the different grain sizes. Moreover, compaction and bearing capacity tests evidenced
no significant influence of the moisture content on compaction level reached by this kind of materials for all the
grain size fractions reported in Tables 2. Compaction is mostly due to the attitude of porous aggregates to crush,
reducing the interparticle voids between the aggregates, an example for the grain size fraction 15/20 is reported in
Figure 1.
1004 Marradi A. et al. / Procedia - Social and Behavioral Sciences 53 (2012) 1001 – 1010

Figure 1 Example of grading curve evolution before and after compaction

Standard compaction tests, like AASHTO modified compaction test, can be improved by the use of new
compaction methods able to better simulate the real stress conditions occurred on site during roller compaction.
In this context, the use of the Shear Gyratory Compactor (SGC) is a new approach that is focus increasing
worldwide, also for unbound natural and Construction and Demolition (C&D) recycled materials (Lancieri 2006),
thanks to their capability to be closer to the effect of the on-site compaction machines and to allow deeper
investigation about the life cycle of the soil in an accelerated way. SGC consists in a cylindrical mould where a
tested specimen is compacted with a constant consolidation pressure, at a constant angle of gyration and at a
fixed rotation speed. All these parameters can be changed in order to simulate the on-site compaction pressure,
the shear stress induced by the roller during subsequent passes and their working speed. Test results are usually
outputs as the height of the specimen versus the number of gyrations, allowing to calculate the material density
for each gyration. The effect of the compaction parameters have been deeply investigated in the past. Ping W.V.
et al (2003) suggest as standard procedure to compact unbound soils with SGC the use of a vertical pressure of
200kPa, gyratory angle of 1.25°, a gyration rate of 20 gyrations/minute until 90 gyrations. McGarvey K. et al
(2010) states that the vertical pressure should be selected in the range 600÷1000 kPa. In this study pumice
specimens have been compacted using a Pine Gyratory Compactor, set with 600 kPa vertical pressure, 1.25°
gyratory angle and a rotation rate of 30 gyration/minute. In the first step of compacting tests the compaction
effort was assumed to be simulated by 90 gyration, but in the subsequently tests this parameter have been
changed in order to investigate the different behaviour of specimens compacted with a different number of
gyrations. Graph reported in Figure 2, referred to grain size fraction 15/20, shows the evolution of the specimen
density plotted versus the number of gyrations. Three different moisture contents have been used: one close to the
optimum value, one very low and the third very high. No significance difference can be underlined in the density
evolution neither in the bearing capacity (Californian Bering Ratio test CBR carried out immediately after
compaction), confirming what stated during evaluation of the mechanical tests carried out on AASHTOmod
compacted specimens. Furthermore, CBR values appear to be very high, always greater than 28%. Grading curve
evolution after compaction seems to be similar to one evidenced after the previous mentioned AASHTOmod
compaction test, even if the fine production appear to be less intensive. These results can be attributed to the
differences in type of compaction method; under impulsive compaction the aggregates are destroyed while the
effect of the shear stresses induced by the gyratory compactor can be mostly attributed to the interparticle
attrition and abrasion. Results reported below also underline the effect of the number of gyrations. As reported by
Ping W.V. et al (2003), specimen compaction evidences a higher increase during the first 30 gyration than the
following ones, confirming the similarity assumption of the gyratory compactor with on-site roller compaction.
 Marradi A. et al. / Procedia - Social and Behavioral Sciences 53 (2012) 1001 – 1010 1005

Figure 2 Density evolution (grain size fraction 15/20) for different moisture content and resultant grading curve after
compaction

As reported by many authors (McGarvey K. et al (2010), Kumpel et al (2012)), mechanism that mainly
contribute to compaction is the reduction of the aggregates angularity, produced by the abrasion and attrition
process induced in the aggregates during compaction. This process reduces the aggregates interlocking and
increases the tendency to move into a denser configuration. For aggregates traditionally used in road construction
it was found that compaction led to a 1% increase in the fines, which indicate attrition and abrasion of the
aggregates (Rammatury et al 1974). This quantification, compared to wash sieve analysis presented in Figure 2,
seems to be reliable for common natural aggregates, and too high conservative for the natural porous aggregates,
like pumice aggregate, which shows a more pronounced fines production reaching values close to 20%. It is
believed that traffic load, especially for airport infrastructure (Donovan P et al (2008)), can have a dangerous
effect on stability of unbound layers due to the particles constant rearrangement and compaction. The
densification of this material can lead the pavement to rut (Archilla 2006). For all these reasons, the use of porous
lightweight aggregates depends on the possibility to use them without having stability problems during the
pavement service life related to the higher densification induced by traffic. One of the aim of this research is to
investigate the densification attitude of a pumice mixture, composed using all the grain size fraction previous
mentioned (Figure 3), exploiting the ability of the SGC machine to simultaneously apply a vertical load in
addition to a self-adjusting kneading action which can simulates the moving traffic load experienced by a flexible
pavement system (Ping W. et al 2003). For this purpose compaction tests were organized into two different
phases: first step simulates the effect of compaction roller.

100%
Pumice mix
90%
Reference mix
80% Subgrade gradation
band
70%

60%
% Passing

50%

40%

30%

20%

10%

0%
0.01 0.1 1 10 100
Grain saize (mm)

Figure 3 Grain size distribution of pumice and reference mixes

Second step, carried out strongly reducing the vertical consolidation pressure, was performed to investigate
the long term behaviour of unbound aggregates, simulating with great number gyrations the effect of traffic
vehicle. This type of test needs to adopt a specific procedure, setting the SGC compaction parameter (N° of
gyration and compaction pressure) in different manner for the first and the second step. The first compaction step,
which simulates the effect of compaction roller on site, was characterized by a 600 kPa vertical pressure. For this
step the number of gyrations was set on two different values 30 and 60 rotations, in order to simulate a different
on-site material compaction level. The second step was performed compacting the specimens again up to the
maximum number of gyrations allowed by the machine, adopting a 200 kPa of vertical pressure as conservative
value to simulate a the stress level induced to a subgrade layer in a road pavement by traffic loads. This
procedure was carried out to evaluate post-compaction attitude, in terms of densification due to the traffic loads,
of mixture compacted applying a different compaction effort during construction phase. Results were then
compared with that founded out with a limestone mixture (Figure 3 – reference mix) which can be considered as
a reference mixture for subgrade mixture. Compaction trend is expressed as the evolution of the height of the
specimen inside the mould during the gyratory compaction. In Figure 4 are reported the graphs related to the two
different compaction settings, following the previous described procedure. It is possible to notice that compaction
trend of the two type of materials, pumice and limestone mix, is not the same; pumice mix shows a better
compactability compared to the reference material, confirming the attitude of this materials to be abraded during
compaction, producing a very high quantity of fines, and resulting in a better compaction after a small number of
gyrations. This behaviour is especially true for compaction with 30 gyrations than with 60 gyrations.
The second step of the compaction curve (Figure 4), referred to the evolution of the specimen height after
decreasing of vertical pressure down to 200 kPa, have been further analysed in order to evaluate the post-
compaction attitude of the pumice mix (Figure 5). Results, expressed as the percentage of specimen height at the
end of compaction phase (first step 30 or 60 gyration), have been compared to the ones founded out with the
reference mix.

Hk  Hi
Post Compaction [%] (2)
Hk
where:
Hk specimen height at the end of first step (30 or 60 gyration – compaction);
Hi specimen height at i-th gyration of second step;
Results obtained for the two different test settings show that there are no significant differences between post
compaction evolution of pumice mix and reference mix, confirmed by similar slope of regression curve equation
in Figure 4, even if compaction (first test step) showed different trend for both materials. Analysing post
compaction data (Figure 5) it’s possible to state that average difference between values undertaken for pumice
and reference mix is always lower than 0.2%. Furthermore, aiming to quantify the settlement occurred during the
pavement service lice due to the traffic loads, maximum post compaction values have been evaluated: specimens
compacted with 60 gyrations show a post compaction settlement close to 4% of the height reached after
compaction.
 Marradi A. et al. / Procedia - Social and Behavioral Sciences 53 (2012) 1001 – 1010 1007

Figure 4 Evolution of the specimen height during compaction tests

Figure 5 Post Compaction trend

2.2. Natural lightweight aggregates

Geofoam is defined by the ASTM D6817 Standard as block or planar rigid cellular foam polymeric material
used in geotechnical engineering applications. It is also defined (EPS – Expanded Polystyrene) as a type of
foamed plastic formed by the expansion of polystyrene resin beads in a moulding process. Geofoam is produced
as a block and is characterized by specific dimensions tolerances (UNI EN 13163). The bulk density varies from
10 to 40 daN/m3 with water absorption of about 2÷5% of the specimen volume. As reported by many authors,
(Montepara et al 2001) the use of geofoam is a suitable alternative to natural aggregates to be used for
construction of embankment standing on soft subgrade. Geofoam offers the possibility to have a very light
structure avoiding all the problems related to the settlement of the embankment foundation. This is especially true
when rapid road rehabilitation is required in order to solve the severe problems related to landslide phenomena.
Geofoam mechanical properties are the main goal of different research provided by many authors. Horvat (1995)
performed different tests in order to evaluate the stress strain behaviour of the geofoam. Results of this research
allow to state that stress-strain behaviour can be divided into four subsequent parts: an initial linear response until
strains between 1÷1.5%, a yielding zone, a linear and work hardening zone and a non-linear but still hardening
work zone. More recently, Montepara et al (2000) reported that geofoam can be considered as homogeneous and
isotropic, showing an elastic behaviour until 2.5% of deformation level. Shear resistance curves for varying
values of applied slipping velocity (0.1, 0.5 and 1 mm/min) show a typical trend of shear resistance increasing
1008 Marradi A. et al. / Procedia - Social and Behavioral Sciences 53 (2012) 1001 – 1010

with applied deformation, without peak values and, hence, without showing a clear rupture point. Specimens of
geofoam subjected to bending with a stroke rate of 5 mm/min show a linear behaviour up to the distinct rupture.
Furthermore, it seems that the higher is the density, the lower will the strain at break. In this study different
mechanical testes have been performed on geofoam EPS CS(10)150 (hereinafter called EPS 150), assuming the
EPS a common material for embankment construction, as suggested by Stark et al (2004). Preliminary results are
reported in terms of CBR% (Californian Bearing Ratio) (Figure 6) and Surface Modulus (Figure 7), carried out
with LWD (Light Weight Deflectometer) device. CBR tests have been carried in dry and wet condition, without
enter the specimen in the Proctor mould. This test setting have been chosen according to previous tests, showing
that confinement have no influence on specimens bearing capacity. Results, reported in the graph below, show
that no significant differences can be underlined between wet and dry tests and that CBR% remain close to 3%,
according to Stark et al (2004).

Figure 6 (a),(b) CBR test device, (c) CBR test results

LWD tests are usually used for Qc/Qa activities on unbound, cement bound or recycled materials (Marradi
2011). Surface Modulus can be calculated using the Bossinesq theory. Poisson ratio, according to Montepara et al
(2001), has been assumed to be 0.02. Results, reported in the graph below, shows the Surface Moduli for two
different applied pressure levels, 70 kPa and 50 kPa.

Figure 7 (a) LWD test device, (b) LWD test results

Results from CBR and LWD tests underline the weakness of this material, even if a small stress hardening
tendency can be highlighted by LWD tests executed with different applied pressure.
As reported by previous mentioned researches (Montepara et al., 2001) the most important parameter that must
be taken into account for design purpose is the vertical stress applied to the geofoam. This should be limited at
value lower than those determining the elastic limit of the material, in order to avoid the problems related to the
viscous deformations. To prevent these problems, ASTM D7180-05 (2009) suggests placing a concrete slab, with
suitable reinforcement, to distribute the load to levels below the compressive resistance at 1% strain.
In this research creep behaviour has been investigated by means of Load Plate Test device, applying a constant
pressure and measuring the settlements for approximately 7 hours. Four different levels have been chosen:
18 kPa, 35 kPa, 100 kPa and 150 kPa. Results are reported in the graph below (Figure 8).

Figure 8 (a) LPT device using for creep test, (b) Creep test results

The aim of this test is to undertake a preliminary investigation on the effect of the vertical pressure on an
EPS 150 to be used for construction of the backfill below an airport pavement, in the same configuration
successfully used for the reconstruction on the Luis Armstrong Int’l Airport in New Orleans. In this research, the
pavement structure above the EPS layer was supposed to be composed by 25 cm of asphalt mixture (2350
daN/m3), 25 cm of foam stabilized (2150 daN/m3) layer and a 25 cm concrete slab (2500 daN/m3). The first
stress level was chosen as the one induce by pavement weight, the other values were chosen with the aim to
progressively analyse the creep behaviour of EPS150. Results show that until 100 kPa stress level no significance
creep deformations occur.

3. Conclusions

In this research two different types of lightweight materials (both natural and derived from industrial process)
have been analysed; the first is a mix of pumice aggregates and the second is an EPS150 geofoam. Experimental
analysis regarding the pumice mix was carried out with the aim to simulate and investigate the post compaction
trend after the compaction process. The grading curve carried out after compaction, both impulsive (Proctor
modified) and gyratory (Shear Gyratory Compactor), have shown a marked attitude to crush and produce a very
high quantity of fine, prompting to believe that this kind of “structural weakness” of the aggregates will reduce
the stability of the mixture to the exercise traffic loads. For this reasons a post compaction test have been carried
out by means of the SGC, analysing the compaction process in two distinct phases: the first tries to simulate the
compaction process (30÷60 gyrations at 600 kPa vertical pressure) while the second (700 gyrations at 200 kPa
1010 Marradi A. et al. / Procedia - Social and Behavioral Sciences 53 (2012) 1001 – 1010

vertical pressure) has been performed with the aim to reproduce post compaction process due to the traffic loads.
The same process has been carried out with a mix of limestone aggregates, composed in order to be considered as
a reference material. Results show that no significant differences can be recognized between the post compaction
trend of the pumice mix and the reference mixture, remaining always lower than 6%, even if the density
evolution during the first part of the test (compaction) is significantly different. Further on going tests will
analyse the shear resistance of pumice mix by means of GLPA (Gyratory Plate Load Assembly). Preliminary
results regarding the performance of an EPS 150 show the weakness of this material, evidencing a CBR% of 3%
and a Surface Modulus of about 10 MPa, even if a small stress-hardening behaviour can be underlined. Creep
test, carried out by means of LPT device, shows that until 100 kPa of applied stress no marked creep
deformations occur. Further on-going researches will investigate the durability characteristics of EPS geofoam
under dynamic traffic loads effects.

REFERENCES

[1] CIRS Centro Sperimentale Interuniversitario di Ricerca Stradale (2001). Studio a Carattere pre Normativo delle Norme Tecniche di Tipo
Prestazionale per Capitolati Speciali d’Appalto. Document approved by the Commission study for the Specification on road materials and
design, construction and maintenance of the CNR (Consorzio Nazionale Ricerche).
[2] Stark D.T., Arellano D., Horvath J.S., Leshchinsky (2004). Guideline and Recommended Standard for Geofoam Applications in Highway
Embankments. National Cooperative Highway Research Program NCHRP Report 529, Transportation Research Board Washington DC.
[3] Budhu M. (2010). Soil Mechanics and Foundation – 3rd Edition” John Wiley & Sons Inc., ISBN 978-040-55684-9
[4] Terzaghi K., Peck R.B., Mesri G. (1996). Soil Mechanics in Engineering Practice. 3rd edition” John Wiley and Sons Inc. New York.
[5] AA.VV. (1994). Settlement analysis. Technical Engineering and Design Guide as Adapted from the U.S. Army Corps of Engineering
N°9 ASCE.New York.
[6] AA.VV.(1994). Treatment of Soft Foundation for Highway Embankment. NCHRP Synthesis of Highway Practice 29, Transportation
Research Board .Washington DC.
[7] Montepara A., Giuliani F. (2001). Rilevati Alleggeriti in Polistirene Espanso Sinterizzato per Costruzioni Stradali su Suoli Difficili.
Proceedings of 11th SIIV conference. Verona, Italy.
[8] Lancieri F., Marradi A., Mannucci S. (2006). C&D waste for road construction: long time performance of road constructed using recycled
aggregate for unbound pavement layers. Waste Management and the Environment III,Vol 1, pp 559-569, Malta, ISBN 978-1-84564-173-3.
[9] Ping W.V., Xing G., Leonard M., Yang Z. (2003). Evaluation of Laboratory Compaction Techniques for Simulating Field Soil
compaction. Florida department of Transportation Final Report, March 2003.
[10] McGarvey K., Panko M. Hurt C. Metha Y., Sukumaran B. (2010). Use of the Superpave Gyratory Compactor as a Predictor of Field
Performance of Unbound Material. 2010 FAA Worldwide Airport Technology Transfer Conference. Atlantic City, New Jersey, USA.
[11] Kumpel C., Panko M., McGarvey K., Hurt C., Melici A., Rossi S., Yurick C., Daouadji A., Sukumaran B. (2012). Compaction
Characteristics of Subbase Material Using the Superpave Gyratory Compactor. Transportation System Workshop. Austin, Texas.
[12] Rammatury T., Kanitkar V.K., Prakash K. (1974). Behaviour of coarse-grained soils under high stresses. Indian Geotechnical Journal.
[13] Donovan P., Tutumler E. (2008). Effect of Aircraft Load Wander on Unbound Aggregates Pavement Layer Stiffness and Deformation
Behaviour. Proceedings of the Airfield and Highway Pavenebts Conference. Bellevue, Washington.
[14] Archilla A.R. (2006). Use of Superpave Gyratory Compaction Data for Rutting Prediction. Journal of Transportation Engineering., Vol.
132, N°. 9, 734-741.
[15] Ping W., Leonard M., Putcha S.(2002). Laboratory Simulation of Field Compaction Characteristics on Sandy Soil. Trasportational
Research Board 81st Annual Meeting. Washington DC.
[16] ASTM D6817 (2007). Standard Specifications for Rigid Cellular Polystyrene Geofoam.
[17] ASTM D7180 (2009). Standard Guide for Use of Expanded Polystyrene (EPS Geofoam in Geotechnical Projects.
[18] UNI EN 13163 (2009). Prodotti di Polistirene Espanso Ottenuti in Fabbrica.
[19] Montepara A., Giuliani F.(2000). Design of road embankments lightened by Expanded Polystyrene (EPS) laying on low-bearing capacity
grounds. Proceedings of the 10th REAA Conference, Road Development for 21th Century. Tokyo, Japan.
[20] Horvath J.S.(1995). Geofoam Geosynthetic. Scarsdale, New York.
[21] Marradi (2011). Dynamic Field Assessment of Short Term Bearing Capacity for Cold Recycled Layers. Pavement Engineering and
Asphalt Technology, volume 1 2011, pp12-26, ISSN 1464-8164