materials
Article
Influence of Lime and Volcanic Ash on the Properties of Dune
Sand as Sustainable Construction Materials
Faisal I. Shalabi 1, * , Javed Mazher 2 , Kaffayatullah Khan 1 , Muhammad Nasir Amin 1 , Alaa Albaqshi 1 ,
Abdullah Alamer 1 , Ali Barsheed 1 and Othman Alshuaibi 1






Citation: Shalabi, F.I.; Mazher, J.;
Khan, K.; Amin, M.N.; Albaqshi, A.;
Alamer, A.; Barsheed, A.; Alshuaibi,
O. Influence of Lime and Volcanic
Ash on the Properties of Dune Sand
as Sustainable Construction Materials.
Materials 2021, 14, 645. https://
doi.org/10.3390/ma14030645
Received: 13 December 2020
Accepted: 26 January 2021
Published: 30 January 2021
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2021 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Materials 2021, 14, 645. https://doi.org/10.3390/ma14030645
1 Department of Civil and Environmental Engineering, College of Engineering, King Faisal University (KFU),
Al-Ahsa 31982, Saudi Arabia; kkhan@kfu.edu.sa (K.K.); mgadir@kfu.edu.sa (M.N.A.);
214020718@student.kfu.edu.sa (A.A.); 214010919@student.kfu.edu.sa (A.A.);
214110860@student.kfu.edu.sa (A.B.); 215018421@student.kfu.edu.sa (O.A.)
2 Physics Department, College of Science, King Faisal University (KFU), Al-Ahsa 31982, Saudi Arabia;
jkhan@kfu.edu.sa
* Correspondence: fshalabi@kfu.edu.sa; Tel.: +966-13-589-5417; Fax: +966-13-581-7068
Abstract: This study focused on evaluating dune sand stabilized with lime and volcanic ash as
base course materials in engineering construction. Dune sands are found in Saudi Arabia in huge
quantities. Due to the high demand for construction materials, this makes them highly suitable for
construction. A testing program was designed to investigate the effect of adding different percentages
by weight of lime (L: 0, 2, 4, and 6%) and volcanic ash (VA: 0, 1, 3, and 5%) on the engineering
properties of the stabilized mixture. Unconfined compressive strength (UCS) and California bearing
ratio (CBR) tests were conducted. In addition, Raman spectroscopy and laser-scanning microscopy
(LSM) tests were performed to explore the chemical characteristic, packing, and structure of the
mixture. The results showed that the UCS, CBR, and the Young’s modulus (Es ) of the treated dune
sand increased with the increase in percentage of both stabilizers. Furthermore, LSM images of
mortar blended with intermediate L-to-VA blend ratio ≈ 0.55 (L: 6% and VA: 5%) exhibit compact
packing of sand grains, indicating strong adhesion and higher cementing value. The results of the
study are promising and encourage using the treated dune sand in engineering construction even
with a low percentage use of lime (2%) and volcanic ash (1–3%) as stabilizers.
Keywords: dune sand stabilization; cement and volcanic ash; CBR; compressive strength; young’s
modulus; sustainability; construction materials
1. Introduction
Vast growth in industry and population necessitate searching for more and different
resources of good quality earth materials for construction work. The extensive use of
quarried construction materials for the construction of building and roads have led to
severe damage to the environment and increased pollution levels. Governments and
environmental agencies place colossal pressure on the construction industry to look for new
materials and assimilate sustainability. Extensive research works were carried out to serve
sustainability and to stimulate construction process and development. Most of the research
work directed toward using different additives to improve the engineering properties
of materials not meeting the required standards. In arid regions where dune sands are
abundant in huge quantities, many researchers have investigated their use as construction
materials. Since dune sands are almost classified as SP or SP-SM according to USCS (A-1 to
A-3 according to AASHTO classification) [1], and according to the World Road Association,
they are suitable in construction works if they are properly stabilized and treated [2]. Over
the last few decades, dune sands were investigated and utilized in construction purposes
after stabilization with different additives to improve their geotechnical properties and to
meet engineering requirements and specifications. Many researchers concluded that the
https://www.mdpi.com/journal/materials
Materials 2021, 14, 645 2 of 21

dune sands located in various world sites almost possess similar physical properties and
characteristics [3,4].
Many other studies have been conducted on dune sands to explore their characteristics.
Studies on chemical and mineralogical compositions showed that dune sand is mainly
composed of quartz [3–5]. Studies on the physical properties indicated that dune sands
fall in a narrow range of grain sizes (0.08–0.40 mm) [6,7]. For specific gravity of dune
sands from different places in the world, the range was found to be wide, ranging between
2.44 (sands from Egypt) [8] to 2.87 (sands from India) [3]. Morphological studies showed
that dune sands particles are almost rounded in shape with flat cleavage planes and
conchoidal breakage arrangements [3,5,9]. Early attempts to assess dune sand performance
in construction was performed by Khan [10] by studying samples from the Libya desert
and by Sanad and Bindra [11] by analyzing samples collected from Saudi Arabia. More
extensive and focused studies were started to explore the engineering properties of dune
sands and derive conclusions based on experimental work.
Many researchers performed several studies for improving the mechanical and mi-
crostructural properties of dune sand used in road construction by incorporating a different
binder such as cement, lime, ground granulated blast furnace slag, municipality waste, fly
ash, silica fume, metakaolin, bentonite, etc. [12–20]. Wahab and Asi [12] used lime and
Portland cement to accelerate the curing process of dune sand collected from the Eastern
part of Saudi Arabia, treated with emulsified asphalt. The results showed an improvement
in shear strength and resistance of the treated sand. Mohamedzein et al. [13] evaluated
the use of municipal solid waste incinerator ash to stabilize Oman dune sand for geotech-
nical engineering works. The results of the tests showed a considerable increase in shear
strength parameters and UCS of the treated sand. Wayal et al. [14] investigated the effect of
using bentonite and lime to stabilize dune sand collected from Rajasthan, India for uses in
geotechnical engineering work. The results showed a substantial increase in the UCS of the
stabilized sand by adding 3% lime and 15% bentonite. Rabbani et al. [15] investigated the
influence of ground granulated blast furnace slag (GGBFS) and lime on the geotechnical
properties of dune sands of Iran for possible usage in roadways and railways projects. The
results showed significant improvement in the CBR and UCS of the treated sand. Experi-
mental tests on Indian dune sands stabilized with waste crumb rubber showed an increase
in friction angle from 30 to 35 by using 25% of waste rubber [16]. Lahlih and Ahmed [21]
used sulfonated urea, formaldehyde, and melamine to stabilize dune sand. The results
showed a significant increase in the compressive strength of the dune sands even at very
low percentages of stabilizers. Homauoni and Yasrobi [22] used poly methyl and polyvinyl
to stabilized dune sand for road construction in Iran. The results showed an increase in
the shear strength of the stabilized sand in the dry state and a reduction in CBR value for
the saturated state compared to the dry state. The results also showed that the optimum
polymer content for best effect was 3%. Dune sand from Djelfa-Algeria was stabilized
by different percentages of Portland cement to improve its mechanical performance for
usage in roads construction. The results of the tests showed improvement in compressive
strength, tensile strength, and compressibility of the stabilized sand [17]. Querol et al. [23]
and based on an experimental study performed on cement-stabilized dune sand collected
from the western part of Saudi Arabia, found a strong relationship between the CBR value
(as a measure of bearing capacity) and cement content.
Due to the current trends in infrastructure development, the demand for construction
material is also increasing. Nowadays, the research is focused on using locally available
materials for the construction of the road network due to their economic and sustainability
benefits. Dune sands are available in huge quantities all over Saudi Arabia, and therefore,
their uses in the construction industry can be thrifty. The locally available due sand is an
attractive material to be used for the construction of roads and foundations. However,
due to the low bearing capacity of these soil, it is necessary to improve their properties
by using different stabilizers [24–26]. In addition, the locally available pozzolan called
volcanic ash is also available in large quantities in the western part of the country as natural
Materials 2021, 14, 645 3 of 21

Materials 2021, 14, x FOR PEER REVIEW 3 of 22

materials. Comparative to commercially available materials, volcanic ash could be a viable
and economical material to be used as a stabilizer.
usingTherefore, in this study,
different stabilizers. [24–26].lime and locally
In addition, available
the locally volcanic
available pozzolan ash (VA)
called vol-were used as
canic ash ismaterials
stabilizing also available in large quantities
to improve in the western
the engineering part ofof
properties thethe
country
duneassandnatural
for its potential
materials.
use Comparative
as a base to commercially
course material for roads available materials, volcanic
and foundations ash could be
construction. a viable percentages
Different
and economical material to be used as a stabilizer.
of lime (0%, 2%, 4%, and 6%, of the dry weight of the sand) and of volcanic ash (0%,
Therefore, in this study, lime and locally available volcanic ash (VA) were used as
1%, 3%, and
stabilizing 5%) and
materials their blends
to improve were selected.
the engineering Initially,
properties of the duneall materials
sand were analyzed
for its poten-
and
tial characterized by performing
use as a base course material for roadsthe and
different conventional
foundations construction.tests. In the
Different per-second step,
mechanical properties
centages of lime (0%, 2%,such as UCS,
4%, and 6%, of Young’s modulus,
the dry weight of the and
sand)CBR
and oftests wereash
volcanic performed for
all(0%, 1%, 3%,with
samples and 5%)
and and their blends
without limewere
andselected.
volcanicInitially,
ash andall materials wereIn
its blends. analyzed
the third step, the
and characterized
effect of optimized by quantities
performing the different
of lime andconventional
VA on the tests. In the second
improvement ofstep, me-
microstructure was
chanical properties such as UCS, Young’s modulus, and CBR tests were performed for all
investigated by performing laser scanning microscopy and Raman microscopy analysis.
samples with and without lime and volcanic ash and its blends. In the third step, the effect
Finally, useful
of optimized and practical
quantities relationships
of lime and were developed
VA on the improvement between the
of microstructure wasUCS,
inves- Es , and CBR
value as by
tigated a measure
performing of laser
the bearing
scanning capacity
microscopy of and
the stabilized dune sand
Raman microscopy for practical
analysis. Fi- use in
roadways and
nally, useful foundations
and construction.
practical relationships were developed between the UCS, Es, and CBR
value as a measure of the bearing capacity of the stabilized dune sand for practical use in
2.roadways
Materials
andUsed
foundations construction.
2.1. Dune Sand
2. Materials Used
Dune sand used in this study was collected from one of the stockpiles of construction
2.1. Dune Sand
materials located along Dammam road (the eastern region of Saudi Arabia). Figure 1a
Dune sand used in this study was collected from one of the stockpiles of construction
shows an located
materials aspect along
of theDammam
dune sand roadin theeastern
(the eastern partofofSaudi
region Saudi Arabia.
Arabia). According
Figure 1a to the
Unified
shows an Soil Classification
aspect of the dune System-
sand in the ASTM Dpart
eastern 6913ofand
SaudiASTM
Arabia.D2487-17
According[27,28],
to the the sand is
classified
Unified Soilas Classification
SP (poorly graded sand),Dwhile
System- ASTM according
6913 and to the AASHTO
ASTM D2487-17 system,
[27,28], the sand is AASHTO
classified[29],
M145-82 as SPit(poorly graded sand),
is classified as A3while accordingfine
(non-plastic to thesand).
AASHTO system,
Figure 1b AASHTO
shows the grain size
M145-82 [29],
distribution ofitthe
is classified
dune sand,as A3 (non-plastic
which mostlyfine sand).
falls Figure
in the 1b shows
range the grain
of 0.1–0.8 mm size
(with a medium
distribution of the dune sand, which mostly falls in the range of 0.1–0.8 mm (with a me-
grain size of D50 = 0.4 mm). Table 1 summarizes the physical properties and classification
dium grain size of D50 = 0.4 mm). Table 1 summarizes the physical properties and classifi-
ofcation
the dune
of thesand
dune according
sand accordingto standards.
to standards.

(a)

Materials 2021, 14, x FOR PEER REVIEW 4 of 22

Figure 1. (a) Dune sand in the eastern part of Saudi Arabia, and (b) grain size distribution of the dune sand.
Figure 1. (a) Dune sand in the eastern part of Saudi Arabia, and (b) grain size distribution of the
dune
Tablesand.
1. Summary of the physical properties and classification of the dune sand.

Soil Property Value/Description

Specific Gravity [30] 2.68
Color Yellow
D10 (mm) 0.18
Materials 2021, 14, 645 4 of 21

Table 1. Summary of the physical properties and classification of the dune sand.

Soil Property Value/Description

Specific Gravity [30] 2.68
Color Yellow
D10 (mm) 0.18
D30 (mm) 0.3
D60 (mm) 0.42
CC 1.2
Cu 2.3
Both, coarse and fine portions are rounded and
Shape of Particles
sub rounded in shape
Classification-AASHTO system Non-plastic fine sand (A3)
Classification-USCS system Poorly graded sand (SP)

2.2. Volcanic Ash (VA)

Basaltic volcanic ash was collected from the western part of Saudi Arabia, near the
city of Jeddah. Millions of years ago, the western part of the country was subjected to
volcanic activities, which led to the formation of basaltic flows called “Harrat” which cover
an area of about 90,000 km2 [31]. Table 2 shows the physical and chemical properties of the
used volcanic ash. In this table, it can be seen that Silicon Oxides (SiO2 ), Aluminum Oxides
(Al2 O3 ), and Ferric Oxide (Fe2 O3 ) are the most abundant oxides that form the chemical
composition of the used volcanic ash with 73.6% as a total sum by weight.

Table 2. Chemical composition of volcanic ash, VA [31].

Oxide % by Weight
SiO2 46.4
Al2 O3 14.4
Fe2 O3 12.8
CaO 8.8
MgO 8.3
Na2 O 3.8
K2 O 1.9
SO3 0.8
LOI (Loss on ignition) 2.8
Materials 2021, 14, x FOR PEER REVIEW 5 of 22

X-ray diffraction analysis of the volcanic ash powder sample was performed using
Rigaku MiniFlex II and XRD intensity peaks as shown in Figure
MgO 8.32. The XRD peaks show the
presence of both crystalline
Na2O and amorphous phases. The main3.8 crystalline phases present in
K2Oforsterite. While on the other hand, the
the VA are anorthite and 1.9 wairakite phase is calcium-
based zeolite and the SO
major
3 amorphous phase present in the 0.8volcanic ash. According to
LOI (Loss on ignition) 2.8
GSD analysis, D90 and D50 of the used ash are 25 µm and 3 µm, respectively [31].

Figure2.2.
Figure X-ray
X-ray diffraction
diffraction analysis
analysis of powder
of powder sample
sample of volcanicofash
volcanic
(VA). ash (VA).

2.3. Hydrated Lime

The hydrated lime used in this study was procured from a Saudi lime factory. The
hydrated lime has a purity level of Ca(OH)2, a minimum of 90%, and the particle size
analysis shows that 90% of the grain size is less than 90 micron. When lime is added to
Materials 2021, 14, 645 5 of 21

2.3. Hydrated Lime

The hydrated lime used in this study was procured from a Saudi lime factory. The
hydrated lime has a purity level of Ca(OH)2 , a minimum of 90%, and the particle size
analysis shows that 90% of the grain size is less than 90 micron. When lime is added to
silicon and aluminum oxides (within soils or added volcanic ash) with the presence of
water, pozzolanic reactions occur and lead to forming calcium silicate hydrate (C-S-H)
and calcium aluminate hydrate (C-A-H) gels that, after crystallization, will substantially
contribute to connecting the soil particles to a relatively strong structure [15].

2.4. Water
Water used in the testing program was tapped water. According to AASHTO T
26 specifications [32], it has less than 1000 ppm of chloride (CL−2 ) and less than 3000 ppm
of sulfates (SO4 +2 ).

3. Experimental Program and Methodology

3.1. Physical and Engineering Properties Testing Program
An experimental program was performed to attain the outcomes of this study. The
program concentrated on the investigation of the engineering properties and behavior of
the stabilized dune sand using various percentages by weight of hydrated lime (0, 2, 4, and
6%) and volcanic ash (0, 1, 3, and 5%) mixed at the maximum unit weight and optimum
moisture content of dune sand samples compacted by the standard Proctor test, ASTM
D698-07 method A [33]. The preliminary tests considered in this study are the UCS based
on ASTM D2166-85 [34] and the CBR based on ASTM D1883-07 [35]. The samples of both
CBR and UCS were tested after a curing time of 28 days. After casting, the treated dune
sand samples were firmly sealed by thin plastic film and maintained at room temperature
of around 22 ◦ C to keep the moisture content constant. Table 3 shows the performed tests
at various percentages of lime and volcanic ash. Figure 3 presents the standard Proctor
compaction curve of the dune sand with a maximum unit weight of 16.8 kN/m3 and
optimum water content of 7.8%.

Table 3. Physical and engineering testing program of the treated dune sand.

Percentage of Curing Time

Test Percentage of Lime Used Standard
Volcanic Ash (Days)
Specific gravity 0 0 - ASTM D854 [30]
Grain size analysis 0 0 - ASTM D6913 [27]
Standard proctor compaction 0 0 - ASTM D698-07 [33]
Material classification (USCS). ASTM D2487-17 [28]
0 0 -
Material classification (AASHTO) AASHTO M 145-82 [29]
ASTM D2166-85 [34]
Unconfined compressive strength (UCS) 0, 2, 4, 6 0, 1, 3, 5 28
(Method A)
ASTM D1883-07 [35]
California bearing ratio (CBR) 0, 2, 4, 6 0, 1, 3, 5 28
(Method C)

3.2. Spectroscopy and Microscopy Testing Program

A testing program was designed to investigate the morphologic, cement chemistry,
and pozzolanic reactions between sand, volcanic ash, and lime. The hydration of lime and
volcanic ash is performed in different ratios, as shown in Table 4. All measurements were
performed after a fixed aging duration of 90 days.
Materials 2021, 14, 645 6 of 21
Materials 2021, 14, x FOR PEER REVIEW 6 of 22

17
Compaction curve of dune sand

Dry unit weight (kN/m3)

16.8

16.6

16.4

16.2

16
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Water content (%)
Figure3.3.Standard
Figure Standardproctor
proctorcompaction
compactioncurve
curveofofdune
dunesand.
sand.

Table 3. Physical and engineering testing program of the treated dune sand.
Table 4. Tested samples for Raman spectroscopy and laser-scanning microscopy (LSM).
Percentage of Volcanic Curing Time
Test Percentage
Sample Numberof
(#)LimeSample Type and Percentage Used Lime/BlendUsed
RatioStandard
(L/L + VA)
Ash (Days)
a (L:VA)—(0:5) 0.0
Specific gravity 0 0 - ASTM D854 [30]
Grain size analysis b 0 (L:VA)—(6:0)
0 - 1.0
ASTM D6913 [27]
Standard proctor compaction c 0 (L:VA)—(6:1)
0 - ASTM0.86D698-07 [33]
Material classification (USCS). d (L:VA)—(2:1) 0.67
ASTM D2487-17 [28]
Material classification e 0 0
(L:VA)—(6:5) - 0.55 M 145-82 [29]
AASHTO
(AASHTO)
f (L:VA)—(2:3) 0.40
Unconfined compressiveL: Lime, VA: Volcanic Ash. ASTM D2166-85 [34]
0, 2 ,4, 6 0, 1, 3, 5 28
strength (UCS) (Method A)
3.2.1. Raman Spectroscopy ASTM D1883-07 [35]
California bearing ratio (CBR) 0, 2, 4, 6 0, 1, 3, 5 28
Macro Raman spectra were recorded from Horiba Labram (Evolution) (Method C)
spectrometer
using a He-Ne laser excitation source (633 nm, wavelength). The spectrometer was cal-
3.2. Spectroscopy
ibrated and Microscopy
by a Si standard using theTesting
Raman peak at 520.7 cm−1 . All Raman spectra were
Program
recorded using the
A testing confocal
program microscopic
was designed method keeping
to investigate thea morphologic,
hole size fixed cement
at 50 A.U. (1 airy
chemistry,
unit = 1.22 × λ/numerical-aperture) and a spectral accuracy of 0.5 cm −1 .
and pozzolanic reactions between sand, volcanic ash, and lime. The hydration of lime and
volcanic ash is performed in different ratios, as shown in Table 4. All measurements were
3.2.2. Laser Scanning Microscopy
performed after a fixed aging duration of 90 days.
We have deployed the laser scanning microscopy (LSM) method with confocal data
acquisition capabilities
Table 4. Tested samples to
forperform the morphological
Raman spectroscopy studies of microscopy
and laser-scanning sand, lime,(LSM).
ash, and their
mixtures in the various ratios under consideration. We collected the morphologies using
the 50×Number
bothSample and 5× objectives lensesType
for high Lime/Blend Ratio
(#) Sample andresolution
Percentage imaging
Used and large area imaging
respectively. The scan-areas of 200 µm and 2 mm were used for the two said(L/L + VA)
morphologies.
All the imaging a data were collected by deploying
(L:VA) — a(0:5)
confocal hole of a size of ~1 0.0A.U. and
402 nm diodeblaser for the sample illumination.(L:VA) — (6:0) 1.0
c (L:VA) — (6:1) 0.86
4. Results and Discussion
d (L:VA) — (2:1) 0.67
4.1. Raman Spectroscopy
e (L:VA) — (6:5) 0.55
Raman spectroscopy
f is a simple and effective
(L:VA) — (2:3)technique to explore cement 0.40 chem-
istry [36,37]. The active phase blends of cement show higher pozzolanic activity and has
L: Lime, VA: Volcanic Ash.
been extensively explored in the recent scientific literature of cement mortars [38–42]. The
hydrations of lime and ash blends have shown higher cementing strengths due to the for-
mation of many pozzolanic reaction compounds [43,44]. Volcanic ash is known to comprise
of active phase constituents of siliceous and aluminous pozzolans such as nanoporous zeo-
Materials 2021, 14, 645 7 of 21

lites (aluminum silicates), nanoporous silica, iron-oxides, etc. The pozzolans on hydration
with lime in the cementing mortars undergo the formation of calcium silicate hydrates
represented as C-S-H in geochemical notation [45]. It should be noted that the pozzolans
have no cementing value unless mixed with lime in the presence of water. However lime
hydration with pozzolan produces C-S-H compounds, represented by a long duration of
pozzolanic reaction Equation (1) below, which provides the required binding strength to
the cementing mortars [44].

C−H+S−H → C−S−H (1)

where, the industrial notations, C−H and S−H represent calcium hydroxide Ca(OH )2 and
silicate hydrate H4 SiO4 , which is a silica-based pozzolan. The reaction product, C−S−H,
is a pozzolanic reaction compound CaO·SiO2 ·3H2 O. The effect of the C−S−H formation
on the mortar strengths has been explained in detail for lime hydration in the recent
literature [43,46].
The Raman spectra in Figure 4a,b show the properties of volcanic ash and lime,
respectively. The volcanic ash spectrum in Figure 4a has six broad peaks positioned at 481,
551, 670, 797, 876, and 993 cm−1 representative of different pozzolans present in volcanic
ash. The peak at 481 cm−1 is broad and represent bending of various Si-O-Si linkages in the
silicates and silica [45]. The second peak at 551 cm−1 originates from the Al-O-Al stretching
mode of a C3 A-type pozzolan [47]. Slowly hydrating pozzolan (C4 AF) peak is also detected
at 670 cm−1 in the Raman spectrum of volcanic ash indicating the presence of calcium iron
aluminates (Ettringite phases) in the pozzolan [48,49]. In addition, the 670 cm−1 peak also
represents iron (III) oxide minerals (Magnetite) presence [50]. However, the presence of
aluminates are further confirmed by the 797 cm−1 peak of the AlO4 vibration [47]. Another
slowly hydrating pozzolan peak is detected at 872 cm−1 corresponding to the C2 S-type
silicate (Belite phase) present in the ash [48]. The presence of slowly hydrating pozzolans
indicate that the volcanic ash used in this study requires a long cementing time period.
The last peak at 993 cm−1 also indicates the presence of aluminum silicates (Andalusite) in
volcanic ash [50,51].
As shown in Figure 4b, the Raman spectrum of the lime sample confirms the typical
characteristic vibrations of calcium hydroxide phase in between 1497–1761 cm−1 and
portlandite modes at 355 and 719 cm−1 [52,53]. The hydration of lime and volcanic ash
results are shown in Figure 4c–f with a decreasing content of lime in the blend. The lime to
blend ratio is reduced from 0.86 to 0.40 systematically. Owing to the onset of pozzolanic
activity in the hydrating blend, we can observe a series of pozzolanic reaction compounds
of hydrates of calcium-silicates and -aluminates, indicated by a broad band (yellow box) of
Raman peaks in between 450–950 cm−1 . The pozzolanic activity of the blend and thus the
binding performance of the blend are directly related to the size of the yellow box, which
signifies the presence of new pozzolanic reaction compounds. The box area is found to
increase with the increasing content of ash, pointing towards the requirement of silicates
and aluminates to complete the pozzolanic reactions. One can see that the largest number
of pozzolanic reaction compounds form at the lime-to-blend ratio ≈ 0.55. Nevertheless, on
further decreasing the ratio ≈ 0.44, the pozzolanic activity decreases, indicating a deficiency
of lime, which is required to complete the reaction.

4.2. Laser Scanning Microscopy

Microscopic images of grains of sand dune particles, volcanic ash, and lime are shown
in Figure 5a–c, respectively. Microscopic images of the hydrated samples of the cementing
mortars are recorded after 90 days of aging at different lime-to-ash ratios. The long aging
duration of blends has been deliberately chosen to fully complete the pozzolanic reaction
because the slowly hydrating pozzolan phases (C2 S and C4 AF) present in the blends are
also proven from the Raman studies.
Materials 2021, 14, 645 8 of 21

Figure 4. Raman spectra of volcanic ash and pure lime are shown in the bottommost spectra (a) and (b), respectively.
Remaining Raman spectra represent aged cementing mortars prepared in different ratios of lime/ash; 0.86 in Figure (c), 0.67
in Figure (d), 0.55 in Figure (e), and 0.40 in Figure (f), arranged in the decreasing quantity of lime content in the mortar.
Spectral positions of the major Raman peaks are marked and labeled adjacent to the peak in the units of cm−1 .

The laser scanning microscope analysis performed on hydrated volcanic ash mortar
with sand is shown in Figure 6a,b, both at lower and higher magnifications, respectively.
The higher magnification image, Figure 6b, clearly shows a negligible cementing effect
since very little or no interaction is observed in between the grains of the sand dune. The
individual grains remain separated from each other with almost no interaction manifesting
the absence of pozzolanic reaction compound formation. Moreover, the morphology of
the hydrated blend is also phase segregated and the ash particles look like flaky structures
with sharp edges while the sand dune particles are more distinct with rounded granular
structures. The images confirm that both the phases have no binding interaction and thus
the presence of pure ash in sand dune mortar has no cementing value.

1
 The higher magnification image, Figure 6b, clearly shows a negligible cementing effect
since very little or no interaction is observed in between the grains of the sand dune. The
individual grains remain separated from each other with almost no interaction manifest-
ing the absence of pozzolanic reaction compound formation. Moreover, the morphology
of the hydrated blend is also phase segregated and the ash particles look like flaky struc-
Materials 2021, 14, 645 9 of 21
tures with sharp edges while the sand dune particles are more distinct with rounded gran-
ular structures. The images confirm that both the phases have no binding interaction and
thus the presence of pure ash in sand dune mortar has no cementing value.

Figure
Materials (a)x Laser
2021,5.14, scanning
FOR PEER microscopy image of grains of sand dunes. The red colored scale-bar in the left shows ≈200 µm10 of 22
REVIEW
length.Figure 5. (a)scanning
(b) Laser Laser scanning microscopy
microscopy imageimage of grains
of flacks of sand
of volcanic dunes.
ash. The The red colored
red colored scale-bar
scale-bar in the
in the leftleft shows
shows ≈20≈ µm
200
μm length. (b) Laser scanning microscopy image of flacks of volcanic ash. The red colored scale-bar in the left shows ≈ 20
length. (c) Laser scanning microscopy image of the lime powder. The red colored scale-bar in the right shows ≈20 µm length.
μm length. (c) Laser scanning microscopy image of the lime powder. The red colored scale-bar in the right shows ≈ 20μm
length.

Figure 6.
Figure 6. (a)
(a) Laser
Laser scanning
scanningmicroscopy
microscopyimages
imagesofofhydrated
hydratedvolcanic ashash
volcanic blended mortar
blended with
mortar sand,
with sample
sand, #a in#aTable
sample 4, rec-
in Table 4,
orded at lower 5× magnification image (red color scale-bar shows ≈ 2 mm length). (b) Morphological image
recorded at lower 5× magnification image (red color scale-bar shows ≈2 mm length). (b) Morphological image of the of the sample
at a higher magnification of 50× with scale-bar showing ≈ 200 μm length.
sample at a higher magnification of 50× with scale-bar showing ≈200 µm length.

Hydrated lime mortar’s surface morphology is depicted in Figure 7, the wide area
scan is shown in Figure 7a, and the high-resolution scan is shown in Figure 7b. Both im-
ages show that the adhesion of the mortar grains are not very strong owing to a large
number of voids present in between grains. The wider scan, Figure 7a, describes that only
some part of the mortar forms bigger lobes due to the lime related weak binding. The high
 Figure 6. (a) Laser scanning microscopy images of hydrated volcanic ash blended mortar with sand, sample #a in Table 4, rec-
Materials 2021,
orded14,at645 10 of 21
lower 5× magnification image (red color scale-bar shows ≈ 2 mm length). (b) Morphological image of the sample
at a higher magnification of 50× with scale-bar showing ≈ 200 μm length.

Hydrated lime mortar’s surface morphology is depicted in Figure 7, the wide area
Hydrated lime mortar’s surface morphology is depicted in Figure 7, the wide area
scan is shown in Figure 7a, and the high-resolution scan is shown in Figure 7b. Both im-
scan is shown in Figure 7a, and the high-resolution scan is shown in Figure 7b. Both images
ages show that the adhesion of the mortar grains are not very strong owing to a large
show that the adhesion of the mortar grains are not very strong owing to a large number of
number of voids present in between grains. The wider scan, Figure 7a, describes that only
voids present in between grains. The wider scan, Figure 7a, describes that only some part
some part of the mortar forms bigger lobes due to the lime related weak binding. The high
of the mortar forms bigger lobes due to the lime related weak binding. The high resolution
resolution
image, Figureimage, Figure
7b, also show7b, alsogranular
some show some granular
rounded edgesrounded edges
indicating thatindicating
some sand that some
dune
sandare
grains dune grains are
separated fromseparated from forming
other grains, other grains, forming
weaker weaker
clusters. The clusters.
presenceTheof apresence
large
of a large number of voids and weak bonding between sand grains observed
number of voids and weak bonding between sand grains observed in lime-sand mixtures in lime-sand
mixtures
is mainly is mainly
attributed to attributed to the
the formation of aformation of a weak
weak compound duecompound due to
to the presence ofthe
limepresence
only
of lime only as compared to mixtures having both
as compared to mixtures having both lime and volcanic ash. lime and volcanic ash.

Figure
Figure 7. (a)
7. (a) Laser
Laser scanning
scanning microscopy
microscopy image
image of hydrated
of hydrated mortar
mortar containing
containing pure
pure lime
lime and
and sand,
sand, sample
sample #b #b
in in Table
Table 4, 4,
recorded at a lower 5× magnification image (red color scale-bar shows ≈2 mm length). (b) Morphological image of thethe
recorded at a lower 5× magnification image (red color scale-bar shows ≈ 2 mm length). (b) Morphological image of
sample at a higher magnification of 50× with a scale-bar showing ≈ 200 μm length.
sample at a higher magnification of 50× with a scale-bar showing ≈200 µm length.

LSM images of the blended mortar with a lime-to-blend ratio of 0.67 are shown in
Figure 8a,b. It is clear from the wide area scan, as shown in Figure 8a, that an increasing
amount of lumping behavior is observed among the sand dune grains of the mortar due to
the presence of the pozzolanic reaction compounds, which are acting as the binding agents.
The darker regions in the image show bigger and stronger clusters that are difficult to
separate. A darker contrast of the cementing mortar in the microscopic images is indicative
of the formation of C-S-H compounds [48]. Similarly, the high resolution image, Figure 8b,
depicts larger-sized clusters with good adhesion between grains and the lime-ash blend
indicating the formation of pozzolanic compounds. Moreover, the binding of the grains is
stronger than before since no distinct granular edges are observed. In some regions the
distinct granular edges apparently indicate not very strong binding due to incomplete
pozzolanic reactions. Interestingly, the incomplete pozzolanic reaction is also confirmed in
this composition by the preceding Raman studies due to a deficiency of pozzolans.
In the hydrated and lime-to-blended mortar ratio ≈0.55, all grains are very well
connected forming larger lumps due to increased binding properties, as shown in Figure 9.
The wide area scan, Figure 9a, shows large clusters bonded strongly with each other and
appear in the dark contrast color indicating that the grains are bonded by the pozzolanic
reaction compounds. Similarly, the high resolution image, Figure 9b, also depicts void-free
morphology of strong cementing mortar. Moreover, the sand-dune’s grains are also not
distinct in the image indicating that all the granular boundaries are very well soaked
in the binder reaction compounds, forming singular and large solidifications. Thus, the
 agents. The darker regions in the image show bigger and stronger clusters that are difficult
to separate. A darker contrast of the cementing mortar in the microscopic images is indicative
of the formation of C-S-H compounds [48]. Similarly, the high resolution image, Figure 8b,
depicts larger-sized clusters with good adhesion between grains and the lime-ash blend
Materials 2021, 14, 645 indicating the formation of pozzolanic compounds. Moreover, the binding of the11grains of 21 is
stronger than before since no distinct granular edges are observed. In some regions the
distinct granular edges apparently indicate not very strong binding due to incomplete
pozzolanic reactions. Interestingly, the incomplete pozzolanic reaction is also confirmed
microscopic results areby
in this composition in the
a perfect overlap
preceding withstudies
Raman the Raman
due toresults, signifying
a deficiency the high
of pozzolans.
density of pozzolanic reaction compounds at a lime-to-blend ratio of 0.55.

Materials 2021, 14, x FOR PEER REVIEW 12 of 22

Figure 8. (a)
Figure 8. Laser scanning
(a) Laser scanningmicroscopy
microscopyimages of hydrated
images andand
of hydrated blended mortar
blended containing
mortar the the
containing lime-to-ash ratio
lime-to-ash ≈0.67,
ratio ≈ 0.67,
sample
sample #d #d
in in
thethe Table4,4,recorded
Table recordedatat aa lower 5×
5×magnification image
magnification image(red color
(red scale-bar
color shows
scale-bar ≈ 2 mm
shows ≈2 length). (b) Mor-
mm length).
phological image
(b) Morphological of the
image of sample at a at
the sample higher magnification
a higher of 50×
magnification with
of 50 scale-bar
× with showing
scale-bar ≈ 200≈μm
showing 200 length.
µm length.

In the hydrated and lime-to-blended mortar ratio ≈ 0.55, all grains are very well con-
nected forming larger lumps due to increased binding properties, as shown in Figure 9. The
wide area scan, Figure 9a, shows large clusters bonded strongly with each other and ap-
pear in the dark contrast color indicating that the grains are bonded by the pozzolanic
reaction compounds. Similarly, the high resolution image, Figure 9b, also depicts void-
free morphology of strong cementing mortar. Moreover, the sand-dune’s grains are also
not distinct in the image indicating that all the granular boundaries are very well soaked
in the binder reaction compounds, forming singular and large solidifications. Thus, the
microscopic results are in a perfect overlap with the Raman results, signifying the high
density of pozzolanic reaction compounds at a lime-to-blend ratio of 0.55.

Figure 9. (a) Laser scanning microscopy images of hydrated and blended mortar containing a lime-to-ash ratio = 0.55, sample
Figure 9. (a) Laser scanning microscopy images of hydrated and blended mortar containing a lime-to-ash ratio = 0.55,
#esample
in Table#e4,inrecorded
Table 4, at a lowerat5×
recorded magnification
a lower image (red
5× magnification color(red
image scale-bar shows ≈shows
color scale-bar 2 mm length). (b) Morphological
≈ 2 mm length). (b) Morpho-
image of the sample at higher magnification of a 50 × with scale-bar showing ≈ 200 µm length.
logical image of the sample at higher magnification of a 50× with scale-bar showing ≈ 200 μm length.

4.3.
4.3.Unconfined
UnconfinedCompressive
CompressiveStrength (UCS)
Strength (UCS)
The
Theresults
resultsofofUCS
UCS ofof the treated
treated dune
dunesandsandwithwithlime
limecontent
contentandandforfor different
different per-
percentages
centages of of volcanic
volcanic ashash
are are
shownshown in Figure
in Figure 10. The10.tests
Thewere
testsconducted
were conducted at the
at the optimum
optimum water content
water content and maximum
and maximum dry density
dry density of the of the dune
dune sand prepared
sand prepared according
according to the
tostandard
the standard Proctor compaction test (Method A). The size of the
Proctor compaction test (Method A). The size of the tested samples was 102tested samples was
mm
102 mm in diameter and 116 mm in height, as shown in Figure 11a.
in diameter and 116 mm in height, as shown in Figure 11a. The sand was mixed with The sand was mixed
with different
different percentages
percentages by weight
by weight of lime
of lime (0,4,2,and
(0, 2, 4, and
6%)6%)
andand volcanic
volcanic ashash (0,3,1,and
(0, 1, 3, and
5%).
5%).
TheThe samples
samples werewere sealed
sealed withwith thin plastic
thin plastic sheetssheets and cured
and cured for 28 for 28atdays
days at atemper-
a room room
temperature ◦ C. The results in Figure 10 show that with the absence of volcanic ash,
ature of 22 of
°C.22The results in Figure 10 show that with the absence of volcanic ash, UCS
increases from almost zero at a lime content of 0% to 0.12 MPa at a lime content of 2%,
and for a lime content above 2%, there is almost no increase in the strength of the treated
sand. As the percentage of volcanic ash increases, the results show tremendous improve-
ment in the strength with the increase in lime content. The maximum value of strength
 Materials 2021, 14, 645 12 of 21

UCS increases from almost zero at a lime content of 0% to 0.12 MPa at a lime content of
2%, and for a lime content above 2%, there is almost no increase in the strength of the
treated sand. As the percentage of volcanic ash increases, the results show tremendous
improvement in the strength with the increase in lime content. The maximum value of
strength was observed at lime and volcanic ash of 6% and 5%, respectively (Lime to blend
ratio of 0.55). The UCS results are consistent with the results observed in Raman tests
als 2021, 14, x FOR PEER REVIEW and laser microscopic images (Section 4.2). Raman results show that13 the materials in the
of 22
volcanic ash have no cementing behavior unless mixed with lime with the presence of
water, which in turn produces calcium silicate hydrates compounds (Figure 4b). In addition,
the increase in lime content in the absence of volcanic ash as a source of pozzolans will not
crumbled afterimprove
a few minutes of soaking,
sand strength (Figurewhile the results
4a). The one with lime: that
indicate 6% and VA: 5%activities
pozzolanic re- and thus
mained intact the
andbinding
strong for more than 60 days of soaking, as shown in Figure 12c. These
performance of the blend increases with the increase in volcanic ash and lime
results are consistent
content,withand the
the results
maximum obtained from
increase in the UCS is
strength tests.
observed at a lime-to-blend ratio of 0.55
Two factor(Lime 6% and VA 5%). This behavior is represented to
multi-level statistical analysis were performed byfind out theshaded
the yellow statistical
zone in Figure 4c,
significance ofwhere
the results of the effects of lime and volcanic ash on UCS [54].
the largest number of pozzolanic reaction compounds are formed. Furthermore,Coefficients
of variance calculated for lime
laser scanning and volcanic
microscopy ash treatments,
images in Figure 9 showusing that
the nine
for a test samples, ratio of 0.55,
lime-to-blend
on the UCS response
large clusters bonded strongly with each other with a void-free andmeasure
are 13.8 and 6.9, respectively. The P values, which are the dense morphological
of confidence on the truth
structure of the large
forming null hypothesis (Ho),are
solidifications, areidentified.
calculated As for aboth thethe
result, effects
newofintegrated and
lime and VA on the UCS
compact response.
structure The statistically
is expected to havecalculated Plime and strength.
high compressive PVA values for the of the treated
Samples
UCS data are dune
0.001 sand
and 0.0103,
and after respectively.
performing The P value
the UCS test for
were the lime is
soaked in by an order
water, as shownof in Figure 12.
magnitude smaller than the traditional cutoff value for the meaningful statistical
In Figure 12a it can be seen that the sample with lime: 6% and VA: 0% completely crumbled signifi-
cance (≈ 0.05)after
and athus
few the UCS of
minutes results
soaking,are while
undoubtedly dependent
the one with lime: 6% onand
limeVA:content.
5% remained intact
Whilst the PVAandvalue is smaller
strong for morebut than
comparable
60 daystoofthe order of
soaking, asthe
shownstatistical significance,
in Figure 12c. These results are
implying a weak dependence
consistent of the
with the UCSobtained
results on the volcanic
from the ash
UCScontent.
tests.

1
Unconfined compressive strength (MPa)

VA = 5%
0.9
VA = 3%
0.8
VA = 1%
0.7 VA =0%
0.6
0.5
0.4
0.3
0.2
0.1
0
0 1 2 3 4 5 6 7 8
Lime content (%)

Figure 10. Unconfined compressive strength of the treated dune sand vs. different lime content.
Figure 10. Unconfined compressive strength of the treated dune sand vs. different lime content.

Two factor multi-level statistical analysis were performed to find out the statistical
significance of the results of the effects of lime and volcanic ash on UCS [54]. Coefficients
of variance calculated for lime and volcanic ash treatments, using the nine test samples,
on the UCS response are 13.8 and 6.9, respectively. The P values, which are the measure
of confidence on the truth of the null hypothesis (Ho ), are calculated for both the effects
of lime and VA on the UCS response. The statistically calculated Plime and PVA values
for the UCS data are 0.001 and 0.0103, respectively. The P value for the lime is by an
order of magnitude smaller than the traditional cutoff value for the meaningful statistical
 0.4

Unconfined com
0.3
0.2
0.1
Materials 2021, 14, 645 13 of 21
0
0 1 2 3 4 5 6 7 8
Lime content (%)
significance (≈ 0.05) and thus the UCS results are undoubtedly dependent on lime content.
Whilst the PVA value is smaller but comparable to the order of the statistical significance,
Figure 10. Unconfined
implying compressive
a weak dependence of strength
the UCSofon
thethe
treated duneash
volcanic sand vs. different lime content.
content.

Materials 2021, 14, x FOR PEER REVIEW 14 of 22

Figure 11.
Figure (a) UCS-test,
11. (a) UCS-test, (b)
(b) CBR
CBR test,
test, and
and (c)
(c) CBR-tested
CBR-tested sample.
sample.

Figure
Figure12.
12.Samples
Samplesof
ofstabilized
stabilizeddune
dunesand
sandsoaked
soakedin
inwater.
water. (a)
(a) L:
L: 6%,
6%, VA:
VA: 0%,
0%, (b)
(b) L:
L: 6%,
6%, VA: 3%, and
VA: 3%, and (c)
(c) L:
L: 6%,
6%, VA:
VA: 5%.
5%.

4.4.
4.4. Young’s
Young’s Modulus
Modulus (E (Es))
s
Young’s modulus (E
Young’s modulus (Es)s )ofofthe thetreated
treateddune
dunesandsandwaswas alsoalso
investigated
investigated at different per-
at different
centages of lime and volcanic ash. The modulus was evaluated as
percentages of lime and volcanic ash. The modulus was evaluated as the slope of the linear the slope of the linear
elastic region of
elastic region ofthe
thestress-strain
stress-strain curve.
curve. The
The results
results in Figure
in Figure 13 show
13 show almost almost a similar
a similar trend
trend and behavior to that observed in Figure 10 for the UCS. With
and behavior to that observed in Figure 10 for the UCS. With the absence of volcanic the absence of volcanic
ash,
ash, the improvement
the improvement in Esin
withEs with the increase
the increase in lime
in lime content content is insignificant.
is insignificant. Besides Besides that,
that, there
there is almost
is almost no improvement
no improvement in Es forin lime
Es forcontent
lime content
greatergreater
than 2%. than
In 2%. In addition,
addition, the resultsthe
results show that, with the absence of lime, there is no increase in E
show that, with the absence of lime, there is no increase in Es with the addition of volcanic s with the addition of
volcanic ash. Ramon spectroscopy results in Figure 4a,b explained
ash. Ramon spectroscopy results in Figure 4a,b explained the observed behavior, which the observed behavior,
which is directly
is directly related related to the number
to the number of pozzolanic
of pozzolanic reaction reaction
compounds compounds and the for-
and the formation of
mation
dense and of dense
bondedand bondedAs
clusters. clusters. As the percentage
the percentage of the added of the addedash
volcanic volcanic
increases, ash the
in-
creases, the results
results show show a considerable
a considerable increase in Eincrease
s with the inincrease
Es with in thelime
increase
content.in lime content.
Interestingly,
Interestingly,
by using just by 2%using justand
of lime 2% 1%of lime and 1% ash,
of volcanic of volcanic
sand Eash, s of sand
the E of
treated
s the treated
sand sand
increases
increases
significantlysignificantly
from almost from almost
zero to 50zero
MPa.toThe50 MPa.
slopesThe slopes
of the of the in
Es curves Es Figure
curves13 in start
Figure to
13 start to decrease for lime content greater than 2%. This means that for lime content
greater than 2%, the effect of volcanic ash content on the Es is more significant than the
increase in the effect lime content.
Coefficients of variance calculated for lime and volcanic ash treatments, using the 9
test samples, on the Young’s modulus response are 5.1 and 6.0, respectively. The statistical
 Materials 2021, 14, 645 14 of 21

14, x FOR PEER REVIEW decrease for lime content greater than 2%. This means that for lime content 15 ofgreater
22 than
2%, the effect of volcanic ash content on the Es is more significant than the increase in the
effect lime content.

300
VA = 5%
250 VA = 3%
Young’s modulus , Es (MPa)

VA = 1%
200 VA = 0%

150

100

50

0
0 1 2 3 4 5 6 7 8
Lime content (%)

Figure
Figure 13.13. Young’s
Young’s modulus
modulus oftreated
of the the treated dunevs.
dune sand sand
limevs. lime content
content for different
for different percentages
percentages of
of volcanic ash.
volcanic ash.
Coefficients of variance calculated for lime and volcanic ash treatments, using the
9 testRatio
4.5. California Bearing samples, on the Young’s modulus response are 5.1 and 6.0, respectively. The statistical
(CBR)
analyses show that the calculated P
VA and P values for the modulus are 0.0168 and
CBR tests (Figure 11b,c) were performed on lime un-soaked samples of the treated due
0.0068, respectively. Although the Plime value is smaller but still comparable to the order of
sand at the maximum dry density and optimum water content
the statistical significance, implying a weak dependence of standard Proctoronand
of the modulus lime content.
for the same percentages of lime
On the contrary, theand
PVAvolcanic ash used
value is almost for the
an order UCS tests.
of magnitude The results
smaller than theinmeaningful
Figure 14 show that with no use of volcanic ash, the CBR value increases slightly with
statistical significance (≈0.05) pointing towards the dependence of the modulus of the mixthe
increase in lime content. As the
more so than percentage
on the of the volcanic ash increases, the CBR value
volcanic ash.
increases considerably. Apparently, the effect of volcanic ash on the CBR value is obvious
4.5. California Bearing Ratio (CBR)
even at low lime content. Using just 2% of lime and 3% of volcanic ash, the CBR value
increases tremendously CBR tests4%
from (Figure 11b,c)As
to 130%. were performed
discussed on un-soaked
before, and based samples
on theofresults
the treated due
sand at the maximum dry density and optimum water content of standard Proctor and
obtained from Raman and laser microscopy tests, pure lime mortar is not a suitable binder
for the same percentages of lime and volcanic ash used for the UCS tests. The results in
with the absence Figure
of the pozzolanic reaction compound. Moreover, the maximum number
14 show that with no use of volcanic ash, the CBR value increases slightly with
of reaction compounds are formed
the increase in lime when
content.lime
Asand ash are blended
the percentage almost in
of the volcanic ashequal pro- the CBR
increases,
portions (Lime/Blended ratio ≈ 0.55). Blends at too high or too law ratios will not complete
value increases considerably. Apparently, the effect of volcanic ash on the CBR value is
the pozzolanic reaction
obvious dueeventoat the
low lack of sufficient
lime content. Usingreactants, resulting
just 2% of lime and 3% tooflow bearing
volcanic ash, the CBR
resistance. value increases tremendously from 4% to 130%. As discussed before, and based on the
Coefficients results obtained
of variance from Raman
calculated and laser
for lime andmicroscopy
volcanic ash tests, pure lime mortar
treatments, using is not9 a suitable
the
test samples, on the CBR response are 7.6 and 8.2, respectively. The statistical analysesmaximum
binder with the absence of the pozzolanic reaction compound. Moreover, the
number of reaction compounds are formed when lime and ash are blended almost in equal
show that the Plime and PVA values for the CBR are 0.0077 and 0.0059, respectively, which
proportions (Lime/Blended ratio ≈ 0.55). Blends at too high or too law ratios will not
are much smallercomplete
than thethe traditional cutoff value for the meaningful statistical signifi-
pozzolanic reaction due to the lack of sufficient reactants, resulting to low
cance ≈ 0.05. Since the difference
bearing resistance. between the P values are also small, therefore the CBR
of the mix equally depends on both the lime and VA.
OR PEERMaterials
REVIEW 2021, 14, 645
16 of 2215 of 21

400
VA = 5%
350
VA = 3%
300
VA = 1%
CBR value (%)

250 VA = 0%

200

150

100

50

0
0 1 2 3 4 5 6 7 8
Lime content (%)

FigureFigure
14. CBR value
14. CBR of of
value thethetreated dunesand
treated dune sandvs.vs. lime
lime content
content for different
for different percentages
percentages of volcanic of vol-
ash.
canic ash.
Coefficients of variance calculated for lime and volcanic ash treatments, using the
4.6. Practical Relationshipssamples,
9 test among on the CBR
UCS, response
Es, and are 7.6 of
CBR Value andthe8.2, respectively.
Treated Dune TheSandstatistical analyses
show that the Plime and PVA values for the CBR are 0.0077 and 0.0059, respectively, which
For engineering arepractice and than
much smaller to minimize thecutoff
the traditional efforts of for
value assessing the properties
the meaningful and
statistical significance
parameters needed ≈ for theSince
0.05. design of foundations
the difference between ofthebuildings
P values are andalsoroadways, useful
small, therefore rela-of the
the CBR
mix equally depends on both the lime and VA.
tionships were developed among the CBR value, UCS, and Es of the dune sand treated
with lime and volcanic ash, asRelationships
4.6. Practical presentedamong in Figures
UCS, Es15–17.
, and CBR Figure 15theshows
Value of Treatedthat
Dunethe
SandUCS
increases with the increase in the CBR
For engineering value.
practice andThe data show
to minimize a good
the efforts linear relationship
of assessing the properties and
(UCS = 0.0033 CBR (%)), with needed
parameters correlation
for thefactor
designRof =foundations
2 0.88. Figure 16 shows
of buildings the
and relationship
roadways, useful rela-
between the UCS and tionships
Es. In were
this developed
figure, theamong the CBR value,
Es increases with theUCS, and Es ofinthe
increase dune
the UCS sand
andtreated
with lime and volcanic ash, as presented in Figures 15–17. Figure 15 shows that the UCS
the relationship can be strongly expressed as a second-degree polynomial equation with
increases with the increase in the CBR value. The data show a good linear relationship
2 = 0.95. The relationship between the Es and CBR value is presented
correlation factor, R(UCS = 0.0033 CBR (%)), with correlation factor R2 = 0.88. Figure 16 shows the relationship
in Figure 17. In this between
figure, the
it can
UCSbe seen
and Es . that
In thisthe Es also
figure, the Eincreases
s increases with theincrease
with the increase in the
in the UCS and
CBR value. The relationship can be expressed as a second-degree polynomial with a cor- with
the relationship can be strongly expressed as a second-degree polynomial equation
relation factor, R2 = correlation
0.90. factor, R2 = 0.95. The relationship between the Es and CBR value is presented
in Figure 17. In this figure, it can be seen that the Es also increases with the increase in
the CBR value. The relationship can be expressed as a second-degree polynomial with a
correlation factor, R2 = 0.90.
OR PEERMaterials
REVIEW 2021, 14, 645 17 of 2216 of 21
OR PEER REVIEW 17 of 22

1.6
1.6
UCS

UCS = 0.0033 CBR (%)

strength,UCS

1.4 UCS = 0.0033 CBR (%)

1.4 R22 = 0.88
R = 0.88
compressivestrength,

1.2
1.2
1
1
Unconfinedcompressive
(MPa)
(MPa)

0.8
0.8
0.6
0.6
Unconfined

0.4
0.4
0.2
0.2
0
0
0 50 100 150 200 250 300 350 400
0 50 100 150 CBR value
200 (%) 250 300 350 400
CBR value (%)
Figure 15. Unconfined compressive
Figure 15. Unconfined strength,
compressive UCS vs.vs.
strength, CBR value
valueof the treated dune sand.
Figure 15. Unconfined compressive strength, UCSUCS CBR
vs. CBR value of the
of the treated
treateddune sand.
dune sand.

300
300
E = 151 UCS2+114 UCS +1.15
Ess = 151 UCS2+114 UCS +1.15
250 R2 = 0.95
250 R2 = 0.95
s (MPa)
modulus,EsE(MPa)

200
200
Young’smodulus,

150
150

100
Young’s

100

50
50

0
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Unconfined compressive strength, UCS (MPa)
Unconfined compressive strength, UCS (MPa)

Figure 16. Young’s modulus, Es vs. unconfined compressive strength of the treated dune sand.
Figure 16.Figure
Young’s modulus,
16. Young’s Es vs.Esunconfined
modulus, compressive
vs. unconfined strength
compressive strength of of
thethe treated
treated dunedune
sand. sand.
OR PEERMaterials
REVIEW 2021, 14, 645 18 of 2217 of 21

300

Es = 0.0018 CBR2 +0.354 CBR

250 R2 = 0.90
Young’s modulus, Es (MPa)

200

150

100

50

0
0 50 100 150 200 250 300 350 400
CBR value (%)

Figure 17. Young’s Figure

modulus, Es vs. modulus,
17. Young’s CBR value of CBR
Es vs. the treated dune
value of the sand.
treated dune sand.

5. Conclusions 5. Conclusions
The testing program Thewas
testing program was
conducted toconducted
investigateto investigate
the effecttheof
effect
usingof using
limelime
and and volcanic
vol-
ash as stabilized materials on the engineering properties of dune sand used in the construc-
canic ash as stabilized materials on the engineering properties of dune sand used in the
tion of foundations of roadways and buildings. The program consists of two major parts.
construction of foundations of roadways
The first part investigates and buildings.properties
the engineering The program consists
of the treated of two
sand, ma- UCS,
including
jor parts. The first part investigates
Young’s modules, andtheCBR
engineering
values. Theproperties
second part of the treated
focused sand, includ-
on the chemical characteristic
ing UCS, Young’s modules, and CBR values. The second part focused on the chemical anal-
of the interaction between the additives using Raman spectroscopy and morphologic
characteristic of theysis of the lime-ash-sand
interaction betweenmortar using laser-scanning
the additives using Raman microscopy (LSM). Theand
spectroscopy results of
the testing program of the second part were very important and strongly supported the
morphologic analysis of the lime-ash-sand mortar using laser-scanning microscopy
findings obtained from the testing programs of the first one. The following conclusions
(LSM). The results were
of the testing program of the second part were very important and
derived:
strongly supported 1.the findings
The Ramanobtained from
spectroscopy the testing
analysis programs
of the samples having ofhydrated
the firstlime
one. The
and volcanic
following conclusions were derived:
ash with lime to blend ratio of 0.55 (VA = 6% and L = 5%) showed better Raman
peaks as compared to all other samples. The high Raman peaks demonstrated high
1. The Raman spectroscopy analysis
pozzolanic of the
reactivity samples
between having
lime and hydrated
volcanic ash andlime and volcanic
ultimately formed a large
number of reaction products for better strength properties;
ash with lime to blend ratio of 0.55 (VA = 6% and L = 5%) showed better Raman peaks
as compared to2. all Laser
otherscanning
samples. microscopy (LSM) results showed that dune sand treated with hy-
The high Raman peaks demonstrated high poz-
drated lime and volcanic ash with lime to a blend ratio of 0.55 were more compacted
zolanic reactivity between lime and
and had better volcanic
binding propertiesashthan
andother
ultimately
mixtures.formed a large num-
High pozzolanic reactivity of
ber of reaction products for and
both lime better strength
volcanic properties;
ash resulted in the formation of reactive compounds which
2. improved the
Laser scanning microscopy adhesion
(LSM) between
results the sand
showed particles
that duneand are ultimately
sand treated withresponsible
hy- for
the compact packing of sand particles;
drated lime and 3. volcanic
The UCS ash
and with
CBRlime
valuetoof athe
blend ratio
treated duneofsand
0.55increased
were morewith compacted
the increase in the
and had better binding properties
percentages than ash
of volcanic other andmixtures.
lime content.High
Thepozzolanic
maximum value reactivity of was
of strength
observed
both lime and volcanic ashatresulted
a lime-to-blend
in theratio of 0.55. The
formation ofUCS and CBR
reactive results are consistent
compounds which with
the results observed from Raman and laser microscopic tests for a lime-to-blend ratio
improved the adhesion between the sand particles and are ultimately responsible for
of 0.55;
the compact packing of sand particles;
3. The UCS and CBR value of the treated dune sand increased with the increase in the
percentages of volcanic ash and lime content. The maximum value of strength was
observed at a lime-to-blend ratio of 0.55. The UCS and CBR results are consistent with
Materials 2021, 14, 645 18 of 21

4. As the percentage of the added volcanic ash increased, the Es of the treated dune sand
increased with the increase in lime content. Interestingly, using just 2% of lime and
1% of volcanic ash, the Es increased significantly from almost zero to 50 MPa;
5. Practical and useful relationships were developed among the CBR value, UCS, and Es
of the dune sand treated with lime and volcanic ash. These relations can helpfully
be used for the design of foundations of roadways and structures using dune sands
stabilized with lime and volcanic ash.
The current study concluded that the combined use of volcanic ash and hydrated
lime were more effective in the stabilization of dune sand by improving its mechanical
and microstructural properties. However, to further investigate the effective use of locally
available VA as a potential dune sand stabilizer, future research is recommended by using
the blend of VA with cement and other highly reactive materials such as silica fume and
metakaolin for the stabilization of dune sand. To get a better understating of the properties
of the treated dune sand with VA and reactive materials, other aspects may need to be
considered such as the effect of water soaking in conjugation with curing time. This
would help the researcher and designer in selecting the best stabilization materials for
stronger, economical, and sustainable construction. The main challenge facing the use
of treated dune sand as construction materials for roadways in Saudi Arabia is to set
proper standards and codes that cover the entire processes, including the design phase,
construction procedures and methods, monitoring techniques and tools, and remedial
measure processes. This can be solved and improved with close cooperation between
researchers in this field and the decision makers of the roadways industry.

Author Contributions: Conceptualization, F.I.S., J.M. and K.K.; methodology, F.I.S., J.M., K.K. and
M.N.A.; software, F.I.S. and J.M.; validation, F.I.S. and J.M., project administration and resources,
F.I.S., J.M. and M.N.A.; supervision, F.I.S.; preparation and performing the tests of soil classification
and mechanical properties, A.A. (Alaa Albaqshi), A.A. (Abdullah Alamer), O.A. and A.B.; performing
and analyzing spectroscopy and microscopy tests, J.M. and K.K.; performing and analyzing XRD
tests, K.K.; discussion and conclusions, F.I.S., J.M., K.K. and M.N.A.; writing—original draft of this
manuscripts, F.I.S., J.M. and K.K.; writing—review and editing, F.I.S., J.M., K.K. and M.N.A. All
authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the Deanship of Scientific Research at King Faisal University
through its group research grant number (1811011).
Data Availability Statement: The data presented in this study are available on request from the
corresponding author.
Acknowledgments: The Authors acknowledge the Deanship of Scientific Research at King Faisal
University, Al-Ahsa, Saudi Arabia, for the financial support under the Research Group Support
Track (Grant No. 1811011). The authors also extend their appreciation to “Super Burkani Blocks
Company” for providing basaltic volcanic ash used in this research and to the Department of Civil
& Environmental Engineering—College of Engineering at King Faisal University, Saudi Arabia
for providing its lab facilities to conduct the experimental tasks. The authors also would like to
acknowledge the support provided by the Department of Physics—College of Science at King Faisal
University, Saudi Arabia for providing its lab facilities to conduct the microstructural tests.
Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design
of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or
in the decision to publish the results.
Materials 2021, 14, 645 19 of 21

Abbreviations

A3 Non-plastic fine sand

AASHTO American Association of State Highway and Transportation Officials
ASTM American Society for Testing and Materials
A.U. Airy unit
CBR California Bearing Ratio
Cc Coefficient of curvature
Cu Coefficient of uniformity
C-A-H Calcium aluminate hydrate
C-S-H Calcium silicate hydrate
C3 A Tricalcium aluminate
C4 AF Tetracalcium alumino ferrite
C2 S Tricalcium silicate
Es Young’s modulus
D10 Grain diameter (in mm) corresponding to 10% passing
D30 Grain diameter (in mm) corresponding to 30% passing
D50 Grain diameter (in mm) corresponding to 50% passing
D60 Grain diameter (in mm) corresponding to 60% passing
D90 Grain diameter (in mm) corresponding to 90% passing
GGBFS Ground granulated blast furnace slag
GSD Grain Size Distribution
L Lime
LOI Loss on ignition
LSM Laser-scanning microscopy
Plime Factor of multi-level statistical analysis for lime
PVA Factor of multi-level statistical analysis for volcanic ash
ppm Parts per million
SP Poorly graded sand
SP-SM Poorly graded sand with silt
UCS Unconfined compressive strength
USCS Unified Soil Classification System
VA Volcanic ash
XRD X-ray diffraction

References
1. Elipe, G.M.; Querol, S.L. Aeolian sands: Characterization, options of improvement and possible employment in construction- the
state-of-the-art. Const. Build. Mat. 2014, 73, 728–739. [CrossRef]
2. Al-Khanbashi, A.; El-Gamal, M. Modification of sandy soil using water-borne polymer. J. Appl. Poly. Sci. 2003, 88,
2484–2491. [CrossRef]
3. Srinivas, G.P.K.; Uday, K.V.; Iyer, K.R.; Pathak, P.; Keshava, S.M.; Singh, D.N. Characterization of aeolian sands from Indian
desert. Eng. Geol. 2012, 139–140, 38–49.
4. Abu Seif, E.S. Assessing the engineering properties of concrete made with fine dune sands: An experimental study. Arab. J. Geosci.
2013, 6, 857–863. [CrossRef]
5. Abu-Zeid, M.M.; Baghdady, A.R.; El-Etr, H.A. Textural attributes, mineralogy and provenance of sand dune fields in the greater
Al ain area. UAE. J. Arid. Environ. 2001, 48, 475–499. [CrossRef]
6. Al-Sanad, H.A.; Ismaei, N.F.; Nayfeh, A.J. Geotechnical properties of dune sands in Kuwait. Eng. Geol. 1993, 34, 45–52. [CrossRef]
7. Yuan, Y.; Wang, X.; Zhoau, X. Experimental research on compaction characteristics of aeolian sand. Front. Arch. Tech. Civ. Eng.
2008, 2, 359–365. [CrossRef]
8. Abu Seif, E.S. Performance of cement mortar made with fine aggregates of dune sand, Kharga Qasis, Western Desert, Egypt: An
experimental study. Jordan J. Civ. Eng. 2013, 7, 270–284.
9. Hongwei, Y.; Béatrice, A.B.; Ting, Y. Characterization of the surface roughness of sand particles using an advanced fractal
approach. Proc. Math Phys. Eng. Sci. 2016, 472, 20160524.
10. Khan, I.H. Soil studies for highway construction in arid zones. Eng. Geol. 1982, 19, 47–62. [CrossRef]
11. Al-Sanad, H.A.; Bindra, S.P. Soil Mechanics for Road Engineers in Arabian Peninsula; Kuwait University: Jamal Abdul Nasser St,
Kuwait, 1984.
12. Shalabi, F.I.; Mazher, J.; Khan, K.; Alsuliman, M.; Almustafa, I.; Mahmoud, W.; Alomran, N. Cement-stabilized waste sand as
sustainable construction materials for foundations and highway roads. Materials 2019, 12, 600. [CrossRef] [PubMed]
Materials 2021, 14, 645 20 of 21

13. Mohammedzein, Y.E.; Al-Aghbari, M.Y. Stabilization of desert sands using municipal solid waste incinerator ash. Geotech. Geol.
Eng. 2006, 24, 1767–1780. [CrossRef]
14. Wayal, A.S.; Ameta, N.K.; Purohit, D.G. Dune sand stabilization using bentonite and lime. J. Eng. Res. Stud. 2012, 3, 58–60.
15. Rabbani, P.; Daghigh, Y.; Atrechian, M.R.; Karimi, M.; Tolooiyan, A. The potential of lime and grand granulated blast furnace slag
(GGBFS) mixture for stabilization of desert silty sands. J. Civ. Eng. Res. 2012, 2, 108–119. [CrossRef]
16. Tiwari, S.K.; Sharma, J.P.; Yadva, J.S. Geotechnical properties of dune sand-waste tires composite. Mat. Today Proc. 2017, 4,
9851–9855. [CrossRef]
17. Ghrieb, A.; Kettab, R.M. Stabilization and utilization of dune sand in road engineering. Arab J. Sci. Eng. 2014, 39,
1517–1529. [CrossRef]
18. Wang, Y.; Yang, K.; Tang, Z. In situ effect of combined utilization of fly ash and polyacrylamide on sand stabilization in North
China. Catena 2019, 172, 170–178. [CrossRef]
19. Rahman, M.A.; Ahmed, A. Use of Silica Fume in Stabilizing Cement-Dune Sand for Highway Materials. Am. J. Civ. Eng. 2017, 51,
41–49. [CrossRef]
20. Aiban, S.A. A Study of Sand Stabilization in Eastern Saudi Arabia. Eng. Geol. 1994, 38, 65–79. [CrossRef]
21. Lahalih, S.M.; Ahmed, N. Effect of new soil stabilization on the compressive strength of dune sand. Const. Build. Mat. 1998, 12,
321–328. [CrossRef]
22. Homauoni, Z.J.; Yasrobi, S.S. Stabilization of dune sand with poly (methyl methacrylate) and polyvinyl acetate using dry and wet
processing. Geotech. Geol. Eng. 2011, 29, 571–579. [CrossRef]
23. Querol, S.L.; Trujillo, J.A.; Elipe, M.G.; Sanchez, A.M.; Cantero, B. Improvement of the bearing capactity of confined and
unconfined cement-stabilized Aeolian sand. Const. Build. Mat. 2017, 153, 374–384. [CrossRef]
24. Smaida, A.; Haddadi, S.; Nechnech, A. Improvement of the mechanical performance of dune sand for using in flexible pavements.
Const. Build. Mat. 2019, 208, 464–471. [CrossRef]
25. Aksoy, H.S.; Gor, M. Stabilization of dune sand by using various materials. In 4th Geotechnical Symposium; Cukurova University:
Adana, Turkey, 2011; pp. 622–628.
26. Fattah, M.Y.; Joni, H.H.; Al-Dulaimy, A.S.A. Compaction and collapse characteristics of dune sand stabilized with lime-silica
fume mix. Earth Sci. Res. J. 2016, 20, 11–18. [CrossRef]
27. ASTM D6913. Standard Test Methods for Particle-Size Distribution (Gradation) of Soils Using Sieve Analysis; ASTM: West Conshohocken,
PA, USA, 2017.
28. ASTM D2487-17. Standard Particle for Classification of Soils for Engineering Purpose (Unified Soil Classification System); ASTM: West
Conshohocken, PA, USA, 2017.
29. AASHTO M145-82. Standard Specification for Classification of Soils-Aggregate Mixtures for Highway Construction Purposes; AASHTO:
Washington, DC, USA, 1991.
30. ASTM D854. Standard Test Methods for Specific Gravity of Soils by Water Pycnometer; ASTM: West Conshohocken, PA, USA, 2014.
31. Khan, K.; Amin, M.N.; Saleem, U.M.; Qureshi, H.J.; Al-Faiad, M.A.; Qadir, M.G. Effect of finesses of basaltic volcanic ash on
pozzolanic reactivity, ASR expansion and drying shrinkage of blended cement mortars. Materials 2019, 12, 2603. [CrossRef]
32. AASHTO T 26. Standard Method of Test for Quality of Water to Be Used in Concrete; AASHTO: Washington, DC, USA, 2004.
33. ASTM D698-07. Standard Test Methods for Laboratory Compaction Characteristics of Soils Using Standard Effort; ASTM: West
Conshohocken, PA, USA, 2007.
34. ASTM D2166-85. Standard Test Method for Unconfined Compressive Strength of Cohesive Soil; ASTM: West Conshohocken, PA,
USA, 2013.
35. ASTM D1883-07. Standard Test Method for CBR (California Bearing Ratio) of Laboratory-Compacted Soils; ASTM: West Conshohocken,
PA, USA, 2007.
36. Bensted, J. Uses of Raman spectroscopy in cement chemistry. J. Am. Ceram. Soc. 1976, 59, 140–143. [CrossRef]
37. Torres-Carrasco, M.; del Campo, A.; Rubia, M.A.; Reyes, E.; Moragues, A.; Fernández, J.F. In situ full view of the Portland cement
hydration by confocal Raman microscopy. J. Raman Spectrosc. 2019, 50, 720–730. [CrossRef]
38. Kupwade-Patil, K.; Palkovic, S.D.; Bumajdad, A.; Soriano, C.; Büyüköztürk, O. Use of silica fume and natural volcanic ash as a
replacement to Portland cement: Micro and pore structural investigation using NMR, XRD, FTIR and X-ray microtomography.
Constr. Build. Mater. 2018, 158, 574–590. [CrossRef]
39. Sakir, S.; Raman, S.N.; Safiuddin, M.; Kaish, A. Utilization of by-products and wastes as supplementary cementitious materials in
structural mortar for sustainable construction. Sustainability 2020, 12, 3888. [CrossRef]
40. Cruz-Moreno, D.; Fajardo, G.; Flores-Vivian, I.; Orozco-Cruz, R.; Ramos-Rivera, C. Multifunctional surfaces of portland cement-
based materials developed with functionalized silicon-based nanoparticles. Appl. Surf. Sci. 2020, 531, 147355. [CrossRef]
41. Wang, Z.; Yu, Q.; Gauvin, F.; Feng, P.; Qianping, R.; Brouwers, H. Nanodispersed TiO2 hydrosol modified Portland cement paste:
The underlying role of hydration on self-cleaning mechanisms. Cem. Concr. Res. 2020, 136, 106156. [CrossRef]
42. Lemougna, P.N.; Xue-Min, C. Review on the use of volcanic ashes for engineering applications. Resour. Conserv. Recycl. 2018, 137,
177–190. [CrossRef]
43. Masmoudi, R.; Kupwade-Patil, K.; Bumajdad, A.; Büyüköztürk, O. In situ Raman studies on cement paste prepared with natural
pozzolanic volcanic ash and Ordinary Portland Cement. Constr. Build. Mater. 2017, 148, 444–454. [CrossRef]
Materials 2021, 14, 645 21 of 21

44. Zhou, Z.; Sofi, M.; Lumantarna, E.; San Nicolas, R.; Hadi Kusuma, G.; Mendis, P. Strength development and thermogravimetric
Investigation of High-Volume Fly Ash Binders. Materials 2019, 12, 3344. [CrossRef] [PubMed]
45. Elsen, J.; Mertens, G.; Sellings, R. Snellings. Supplementary cementitious materials. Rev. Mineral. Geochem. 2012, 74, 211–278.
46. Yue, Y.; Wang, J.J.; Basheer, P.M.; Bai, Y. In-situ monitoring of early hydration of clinker and Portland cement with optical fiber
excitation Raman spectroscopy. Cem. Concr. Compos. 2020, 112, 103664. [CrossRef]
47. Khan, A.S. Vibrational spectroscopy of selective dental restorative materials. Appl. Spectrosc. Rev. 2017, 52, 507–540. [CrossRef]
48. Deng, C.-S. Ageing of oilfield cement at high humidity: A combined FEG-ESEM and Raman microscopic investigation. J. Mater.
Chem. 2012, 12, 3105–3112. [CrossRef]
49. Rahhal, V.; Cabrera, O.; Delgado, A.; Pedrajas, C.; Talero, R. C4AF ettringite and calorific synergic effect contribution. J. Therm.
Anal. Calorim. 2010, 100, 57–63. [CrossRef]
50. El Mendili, Y.; Vaitkus, A.; Merkys, A.; Grazulis, S.; Chaterigner, D.; Mathevet, F.; Gascoin, S.; Petit, S.; Bardeau, J.; Zanatta, M.;
et al. Raman Open Database: First interconnected Raman–X-ray diffraction open-access resource for material identification. J.
Appl. Crystallogr. 2019, 52, 618–625. [CrossRef]
51. Winter, J.K.; Ghose, S. Thermal expansion and high-temperature crystal chemistry of the Al2 SiO5 polymorphs. Am. Mineral.
1979, 64, 573–586.
52. Shi, L.; Jian, C.; Hai-ning, C.; Sheng-nan, S. Native defect luminescence of zinc oxide films and its potential application as white
light sources. Guang Pu Xue Yu Guang Pu Fen Xi 2016, 36, 1604–1614.
53. Lafuente, B.; Downs, R.; Yang, H.; Stone, N. The Power of Databases: The RRUFF Project. In Highlights in Mineralogical
Crystallography; Armbruster, T., Danisi, R.M., Eds.; Walter de Gruyter: Berlin, Germany, 2015; Volume 1, p. 25.
54. Leonzio, G. ANOVA analysis of an integrated membrane reactor for hydrogen production by methane steam reforming. Int. J.
Hydrogen Energy 2019, 44, 11535–11545. [CrossRef]