With the increasing amount of waste generation from different processes, there has been a growing interest in the use of waste in producing sustainable building materials to achieve potential benefits. This study investigated the influence of waste polypropylene carpet fibers and palm oil fuel ash (POFA) on the microstructure and residual properties of concrete composites exposed to elevated temperatures. Four mixes containing carpet fibers (0% and 0.5%) and POFA (0% and 20%) were prepared. The specimens were exposed to high temperatures (200, 400, 600 and 800 C) for 1 h. The fire resistance of the concrete specimens was then measured in terms of mass loss as well as both residual ultrasonic pulse velocity (UPV) and compressive strength. The role of carpet fibers and POFA was investigated through the analysis of the microstructure in terms of scanning electron microscopy (SEM), thermogravimetric analysis (TGA) and differential thermal analysis (DTA). The results revealed that the addition of waste polypropylene carpet fibers to the concrete matrix significantly enhanced the fire resistance and residual compressive strength in addition to eliminating the explosive spalling behavior of the concrete composites at elevated temperatures. The fire resistance of the concrete mixtures was further enhanced by the inclusion of POFA. The study revealed that the utilization of waste carpet fiber and palm oil fuel ash in the production of sustainable green concrete is feasible both technically and environmentally.
Trang 1Microstructure and residual properties of green concrete composites
temperatures
Department of Structure and Materials, Faculty of Civil Engineering, Universiti Teknologi Malaysia, 81310, UTM Skudai, Johor, Malaysia
a r t i c l e i n f o
Article history:
Received 23 October 2016
Received in revised form
31 December 2016
Accepted 31 December 2016
Available online 2 January 2017
Keywords:
Green concrete composites
Elevated temperatures
Waste carpet fibers
Palm oil fuel ash
Microstructure
Residual properties
a b s t r a c t
With the increasing amount of waste generation from different processes, there has been a growing interest in the use of waste in producing sustainable building materials to achieve potential benefits This study investigated the influence of waste polypropylene carpet fibers and palm oil fuel ash (POFA) on the microstructure and residual properties of concrete composites exposed to elevated temperatures Four mixes containing carpetfibers (0% and 0.5%) and POFA (0% and 20%) were prepared The specimens were exposed to high temperatures (200, 400, 600 and 800C) for 1 h Thefire resistance of the concrete specimens was then measured in terms of mass loss as well as both residual ultrasonic pulse velocity (UPV) and compressive strength The role of carpetfibers and POFA was investigated through the analysis
of the microstructure in terms of scanning electron microscopy (SEM), thermogravimetric analysis (TGA) and differential thermal analysis (DTA) The results revealed that the addition of waste polypropylene carpetfibers to the concrete matrix significantly enhanced the fire resistance and residual compressive strength in addition to eliminating the explosive spalling behavior of the concrete composites at elevated temperatures Thefire resistance of the concrete mixtures was further enhanced by the inclusion of POFA The study revealed that the utilization of waste carpetfiber and palm oil fuel ash in the production
of sustainable green concrete is feasible both technically and environmentally
© 2017 Elsevier Ltd All rights reserved
1 Introduction
There is no doubt that cleaner and more efficient management
of various forms of waste generation is receiving more attention in
order to maintain sustainability in green construction The
utili-zation of waste materials is one of the fundamental issues of waste
management strategies in many parts of the world According to
Guo et al (2014)andSalesa et al (2017), the advantages of recycling
include reducing environmental pollution, reducing landfilling and
disposal of wastes and preserving natural resources Fire represents
one of the most severe potential risks to which structures may be
subjected The behavior of structures exposed to elevated
temper-atures is mostly associated to stress distribution, cracking, spalling
and surface micro cracking In some circumstances, the concrete
structure is exposed to elevated temperatures and pressures
throughout its service for a substantial period, for example,
concrete in a reactor vessel, coal gasification, nuclear plant and other applications.Noumowe et al (1994)andKalifa et al (2001), reported that the significant impacts of high temperature on con-crete structure are the dehydration of cement paste, variation in water content, increase in porosity, thermal expansion and cracking, modification of pore pressure and decrease in strength and thermal spalling owing to extreme pore pressure
A great deal of attempt has been made and various practices have been used to manage high temperatures as well as evaluate the residual performance of concrete structures.Guo et al (2014)
stated that to develop concrete properties,fibrous materials can
be added into the concrete mixture The purpose of such addition is
to enhance its toughness, tensile andflexural strengths, resistance against impact loads and other mechanical properties, reported by
Rashad (2015a,b) In their studies,Silva et al (2014)andMugume and Horiguchi (2014) ascertained that fibrous materials have exhibited good performance in developing thefire resistance ca-pacity of concrete components Recently, the detection and recog-nition offibers for the reinforcement and improvement of concrete have rapidly increased the need for practice in research,
* Corresponding author.
E-mail address: hofa2018@yahoo.com (H Mohammadhosseini).
Contents lists available atScienceDirect Journal of Cleaner Production
j o u r n a l h o m e p a g e :w w w e l s e v i e r c o m / lo c a t e / j c l e p r o
http://dx.doi.org/10.1016/j.jclepro.2016.12.168
0959-6526/© 2017 Elsevier Ltd All rights reserved.
Journal of Cleaner Production 144 (2017) 8e21
Trang 2development and concrete industries According toSanchayan and
Foster (2016), various kinds offibers, either polymeric or metallic,
are generally utilized in concrete mixture for their benefits.Shuaib
and Mativenga (2016)also stated that, the most common fibers
used in concrete are steel, glass and syntheticfibers such as nylon
and polypropylene (PP) as well as naturalfibers and fibers from
pre- and post-consumer wastes Technical developments brought
forward the advancement offibers with different materials,
geo-metric forms and properties to increase the advantages in concrete
constructions Modern manufacturing methods and demands on
fibers, which are to be used in concrete have since been developed
Therefore, different features offiber-reinforced concrete have been
introduced to the market globally (Yu et al., 2016)
In general, syntheticfibers are industrialized to supply the high
demand for textile and carpet products Nylon and polypropylene
are some of the most commonly employed types of syntheticfibers
in the said industries In waste streams, carpets are classified as
textiles and are generated from either pre- or post-consumer
products According to Carpet Recycling UK, cited inSotayo et al
(2015), 400,000 tons of carpet are sent to landfills annually In
the USA alone, approximately 1.9 million tons of textile waste were
generated in 2007, accounting for 4.7% of the total municipal solid
waste Of this, 15.9% of the textile waste was recovered Industrial
carpet wastes are from back and face yarns AsMohammadhosseini
and Awal (2013)point out, the back yarn is mainly in the form of
woven sheets while the face yarn is usually polypropylene or nylon
fibers These fibers are 50e70% nylon and 15e25% polypropylene
Waste carpetfibers can be potentially used in the manufacturing of
concrete as doing so is a hypothetically effective method to reduce
the disposal of waste materials and at the same time decrease the
amount of raw materials used in concrete industries.Awal and
Mohammadhosseini (2016)ascertained that, the concrete
manu-factured containing waste carpetfibers would be lightweight and
possess good acid and alkali resistance
With the growing demand for supplementary cementing
ma-terials, smart and efficacious conservation of construction materials
comprising several by-product waste have received more attention
for the sustainability of green construction (Sua-iam and Makul,
2014) Alsubari et al (2016) point out, the utilization of
pozzo-lanic ashes as supplementary cementing materials in concrete is an
effective way to develop the properties of concrete composites In
recent periods, a great deal of attention is being focused on the
potential use of pozzolanic ashes in concrete These pozzolanic
materials are used in all corners of the world for their technical,
economic and ecological benefits According to
Mohammadhosseini et al (2015,2016), one of the latest inclusions
in the ash group is palm oil fuel ash (POFA), which is obtained by
burning palm oil husks and palm kernel shells as fuel in palm oil
mills.Khankhaje et al (2016)andMujah (2016)reported that in
2007, approximately three million tons of POFA were produced in
Malaysia, and this production rate is expected to rise due to the
increased size of the oil palm tree plantation in the country.Lim
et al (2015)stated that the ash, which is disposed of without any
profitable return, is now considered as a valuable material with
good performance in improving the strength and durability of
concrete mixtures
Concrete has been presented to have a number of benefits when
used in constructions However, it suffers from a main weakness,
which is its high brittleness Due to the importance of concrete
performance at elevated temperatures and infire, several studies
byArioz (2007), Behnood and Ghandehari (2009), and Ates¸ and
Barnes (2012)have been previously carried out in regards to the
subject of fiber-reinforced concrete at high temperatures The
addition of pozzolanic materials in concrete was also reported by
Rashad (2015a,b)forfly ash,Xiao and Falkner (2006)for silica fume,
andAwal et al (2015)for POFA with satisfactory performance at elevated temperatures Amongst fibers, the inclusion of poly-propylene (PP)fibers in concrete mixtures was found to perform very efficiently.Kalifa et al (2001)andPoon et al (2003)stated that steel and polypropylenefibers could be used to decrease cracking and spalling in addition to enhancing the residual strength of concrete at elevated temperatures According to thefindings, it was ascertained that most properties of concrete reduced with an in-crease in temperature, especially for polypropylene fiber-reinforced concrete mixtures
As the addition of polypropylenefibers and pozzolanic materials has been recommended byNoumowe (2005)andBonakdar et al (2013)for the possible decreasing of spalling of concrete at high temperatures, it paves the way for the application of waste carpet fibers and POFA to develop enhanced performance of concrete at elevated temperatures However, research on the utilization of such waste in concrete has not yet been conducted Taking into account the availability of the waste materials and pozzolanic activities of the ash, POFA in particular, extensive research work has been car-ried out in the Department of Structure and Materials of Universiti Teknologi Malaysia (UTM) to explore the potential benefits of producing sustainable building materials
Given the aforementioned argument, the purpose of this study was to investigate the combined effects of waste carpetfibers and POFA on the performance of concrete at elevated temperatures in addition to understanding the way carpetfibers contribute to the reduction in spalling in comparison to plain concrete without any fibers Although this research includes an investigation of industrial waste carpetfibers available, the conducted experiments and an-alyses are based on one single type offiber, namely polypropylene carpetfiber The work has been focused on performance of concrete containing carpet fibers exposed to elevated temperatures, but rather it is believed that technical issues have to be understood and fixed right before utilization of any type of waste fibers in concrete
In this study, a comparison was made amongst the compressive strength and ultrasonic pulse velocity (UPV) of both concrete mixtures containing carpetfibers and plain concrete when exposed
to high temperatures Thermogravimetric analysis (TGA), differ-ential thermal analysis (DTA) and scanning electron microscopy (SEM) were carried out
2 Materials and experimental study
2.1 Materials
Type I ordinary Portland cement (OPC), which achieved the requirements of ASTM C 150-07, was used in this research The palm oil fuel ash (POFA) was collected from a palm oil mill in Malaysia The raw POFA was subsequentlyfinely ground in a Los Angeles milling device containing ten steel bars that were 800 mm long and 12 mm in diameter for a period of 2 h for each 4 kg of POFA The ash conformed to the requirements of BS3892: Part
1-1992 and according to ASTM C618-15, may be categorized as in between class C and F However, considering the source and sort, the ash was neither of class C nor F The specific gravity and Blaine fineness of the used POFA were 2.42 and 4930 (cm2/g) The chemical analysis of both OPC and POFA was conducted using en-ergy dispersive spectrometry The obtained results along with the physical properties are given inTable 1
Mining sand with saturated surface dry condition passing through a 4.75 mm sieve, withfineness modulus of 2.3, specific gravity of 2.6 and 0.7% water absorption, was used as the fine aggregate On the contrary, crushed granite with a maximum size of
10 mm, specific gravity of 2.7 and 0.5% water absorption was used
as the coarse aggregate Throughout the study, supplied tap water
H Mohammadhosseini, J.M Yatim / Journal of Cleaner Production 144 (2017) 8e21 9
Trang 3was used for both mixing and curing purposes A polymer-based
superplasticizer (RHEOBUILD 1100) at 1.0% by weight of
cementi-tious materials was employed to increase the concrete workability
The required waste carpetfibers were collected from ENTEX Carpet
Industries SDN BHD., Selangor, Malaysia The multi-filament
polypropylene carpetfibers were cut into lengths of 20 mm with
an aspect ratio (l/d) of 44 The general properties of the carpetfiber
used are presented inTable 2
2.2 Mix proportions
Four concrete mixtures were prepared in two series, with and
without POFA contents, in addition to having 0 and 0.5% of carpet
fibers contents Series A mixes were made with OPC only, while
series B mixes were made with POFA at the replacement level of
20% by weight of cement The water/binder (w/b) ratio of 0.47 was
kept constant in all mixes The details of the mix proportions are
summarized inTable 3
2.3 Specimen preparation and test methods
For each concrete mix, 100-mm cube specimens were cast and
cured for 24 h in accordance with BS EN 12390-2:2009 and BS EN
12390-3:2009 Subsequently, the cube specimens were demolded
and kept in a water tank until they are required on the day of the
test After 90 days of water curing, the fully saturated concrete
specimens were taken out and dried at room temperature Prior to
testing, all cube samples were weighed The control samples were
tested for ultrasonic pulse velocity (UPV) following ASTM C597-09
and the compressive strength was set at the ambient temperature
of 27 C The concrete cubes were heated in an electric furnace
(Fig 1) to progressive temperature rise of 200, 400, 600 and 800C
The peak temperature was maintained for a period of 1 h The
timeetemperature curve of the furnace is illustrated inFig 2 The
temperature graph revealed similar trend to those of ISO 834 and
ASTM E119 Previous studies byXiao and Falkner (2006)as well as
Awal et al (2015), presented a comparable heating configuration
The thermally preserved concrete cubes were separated into
two sets, which were air- and water-cooled samples The concrete
samples were allowed to cool naturally in the air at the laboratory
temperature of 25 ± 2 C in the air-cooled set while the other
concrete cubes were exposed to water-spray to reflect
fire-combating actions in the water-cooled set The residual weight,
UPV values and cube compressive strength were then recorded for
both cooling categories of all specimens
Scanning electron microscopy (SEM) was used to investigate the morphology and microstructure of the concrete samples at room temperature and elevated temperatures Small particles of concrete samples were prepared for the SEM investigation of the specimens
A thermogravimetric analysis (TGA) and differential thermal anal-ysis (DTA) were also carried out on both unheated and heated concrete specimens TGA determined the equivalent mass loss owing to the thermal decomposition of water within the concrete samples while DTA measured the heatflow in the concrete exposed
to heating Samples taken from concrete cubes of both unheated and heated were subsequently crushed into powder form for testing and analysis purposes Each specimen was heated at a rate
of 20C/min up to 900C under an inert argon condition The heat flow was measured to determine the temperature at which phase changes occurred
3 Results and discussion
3.1 Mass loss
For weight loss assessment, the weights of the concrete cubes were measured before and after the exposure to elevated temper-atures The impact of high temperature on the mass loss of both plain concrete and concrete containing carpetfibers for the air-cooled and water-air-cooled regimes is shown in Fig 3 The mass loss of all the investigated specimens is expressed as a percentage
of the original mass at the ambient temperature to the mass after exposure to a specific elevated temperature.Fig 3further displays that at different temperatures, the weight loss of the concrete mixtures containing carpetfibers and POFA had the tendency to increase
Temperature influence can be separated into three phases, in accordance to the difference of the residual mass obtained at high temperatures In thefirst phase, 27e200C, small mass loss was
observed for all mixes, as the extra amount of free water was pre-sent in the concrete samples Given that the melting point offibers
is at approximately 170 C, this range of temperature did not significantly affect the inner fibers in the concrete specimens However, the outerfibers, which were exposed to temperatures up
to 200C, melted In the second phase, where the temperature increased from 200 to 400C, the weight loss was considerable due
to the complete melting of thefibers as well as the release of both gel and capillary water
Table 1
Chemical composition and physical properties of OPC and POFA.
SiO 2 Al 2 O 3 FeO 3 CaO MgO K 2 O SO 3 LOI Specific gravity Blaine fineness (cm 2 /g)
Table 2
Properties of waste carpet fibers.
Fiber Length (mm) Diameter (mm) Density (kg/m 3 ) Tensile strength (MPa) Melting point (C) Reaction with water
Waste carpet fiber used in this research
H Mohammadhosseini, J.M Yatim / Journal of Cleaner Production 144 (2017) 8e21 10
Trang 4Beyond 400C, the rate of weight loss slowed down moderately.
It should be noted that for both cooling regimes, the mass loss rate
was lesser for the water-cooled regime compared to that of
air-cooled The highest loss of 10.24% for air-cooled and 9.24% for
water-cooled were found in the concrete specimens containing
0.5%fibers and 20% POFA The mass loss in POFA-based concrete
mixtures could be due to the higher moisture content absorbed by
the ash According toAwal et al (2015), the comparatively lesser
mass loss in the water-cooled concrete specimens could be the
consequence of water being absorbed from the surface of the
concrete during the spraying of water in the attempt of bringing
down the temperature In theory, the mass loss in the concrete
samples at high temperatures could be attributed to the
decom-position of calcareous aggregates, liberation of carbon dioxide (CO2)
and sloughing off of the concrete surface, which therefore altered
the mechanical properties of the concrete, stated byDügenci et al
(2015)andMa et al (2015).Xiao and Falkner (2006)interpreted the
structural integrity offiber-reinforced concrete in terms of mass
loss They perceived that the cement matrix losses its binding
properties owing to the vaporization of free water in the calcium
silicate hydrate (CeSeH) gel and decomposition of calcium
hy-droxide Ca(OH)2
To demonstrate the effects of the waste carpetfibers and POFA
on the mass loss of the concrete composites, all the obtained weight
loss data of the concrete specimens are illustrated inFig 4 To
correlate the experimental data, linear regression method was
used, resulting in Eqs.(1)e(4), with a coefficient of determination
(R2) The R2values ranged from 0.90 to 0.95 for all samples, which
signified good confidence for both air- and water-cooled regimes
They are as the following:
A1: mT/m20¼ 0.0113T þ 0.31 (R2¼ 0.9354) (1)
A2: mT/m20¼ 0.0128 T þ 0.0435 (R2¼ 0.9508) (2)
B1: mT/m20¼ 0.0127 T þ 0.1141 (R2¼ 0.9066) (3) B2: mT/m20¼ 0.0134 T þ 0.3326 (R2¼ 0.9178) (4) where m20 is the mass of concrete at 20 C, T represents the exposure temperature (C) and mTsignifies the mass at T 3.2 Spalling and surface color
No notable explosive spalling was observed in the concrete cubes containing carpet fibers throughout the fire testing This finding reinforced the notion that carpet fiber is able to enhance the resistance of concrete against spalling at elevated temperatures significantly Sancak et al (2008) stated that, the main cause of concrete spalling at high temperatures is related to the internal pore pressure build-up, which is due to the evaporation of both free and bound water In the plain concrete specimens without carpet fibers, this inner pressure was not released and thus resulted in explosive spalling of the concrete surface
As aforementioned, spalling was not observed in the concrete samples with carpet fibers at different temperatures This phe-nomenon could be due to the low melting point of carpetfibers
Sideris and Manita (2013)point out, polypropylenefibers melt at approximately 170C while spalling occurs beyond 190C When thefibers melt and are partly absorb by the matrix, the bed of the fibers acts as an additional pathway for gases Therefore, the fibers contribute to the formation of a network along the matrix, which subsequently permits the outward migration of gases and as a consequence, the decrease in pore pressure
At the ambient temperature, the surface color was grey for OPC and dim grey for POFA concrete specimens with smooth surfaces (Fig 5) These appearances were retained up to a temperature of
200C However, at 800C, a whitish grey color for OPC and light grey color for POFA concrete specimens were observed Hairy cracks began to develop at 800C in OPC and POFA mixtures for both air- and water-cooled samples.Xiao and Falkner (2006)stated
Table 3
Mix proportions and the properties of the different concrete mixes.
Mix Cement (kg/m 3 ) POFA (kg/m 3 ) Water (kg/m 3 ) Fine agg (kg/m 3 ) Coarse agg.(kg/m 3 ) V f
(%)
V f
(kg/m 3 )
Slump (mm) VeBe (sec)
0 200 400 600 800 1000 1200
oC)
Time (min)
ASTM E119-14 Experimental ISO 834-12
H Mohammadhosseini, J.M Yatim / Journal of Cleaner Production 144 (2017) 8e21 11
Trang 5that the color variation of the concrete specimens could be
attrib-uted to the changes in the composition and texture as well as both
development and crystal destruction whilefiring
The said observation demonstrated that the variation in the
surface color of the concrete specimens had no clear relationship
with the inclusion of the carpetfibers This could be because the
melting point of the carpet fibers is at a temperature of below
200C However, the alterations in the concrete containing POFA
was more obvious at elevated temperatures This observation
probably owed to chemical transformations, which took place in
the specimens at high temperatures The amount of Fe2O3in the
amorphous state of POFA was higher than that of OPC The iron
oxide in POFA oxidized at temperatures beyond 250C, therefore
created an appearance with a severe variety of colors, as the heating
increased
3.3 Residual ultrasonic pulse velocity (UPV)
The ultrasonic pulse velocity (UPV) test is a non-destructive test
that measures the quality and homogeneity of concrete specimens
to determine the existence of pores and cracks.Fig 6displays the
variations in the UPV of the concrete mixtures containing carpet
fibers and POFA exposed to the designated temperatures It can be
seen that the polypropylene carpetfibers produced no notable
ef-fects on the UPV values of the concrete At the ambient
tempera-ture, the UPV values of the concrete mixtures were high while at
higher temperatures, lower values were recorded in all the test
samples At room temperature, the UPV values of OPC concrete, for
instance, were 4570 m/s and 4580 m/s for 0 and 0.5% carpetfibers
content, which were excellent in terms of concrete quality, as stated
byNeville (1995) UPV values of 4559 m/s and 4540 m/s were also recorded in POFA concrete samples
Higher UPV values of between 4100 and 4550 m/s were found at temperatures of 200e600C in the specimens containing carpet
fibers for both OPC and POFA mixtures in contrast to that of the plain concrete mixture without anyfibers, which could be classified
as good quality concrete However, at a temperature of beyond
600C, the UPV values recorded for the concrete mixes containing carpetfibers significantly dropped
The decrease in the UPV values could be due to the melting of thefibers, which creates an additional porous network along the bed of the meltedfibers In theory, it could also be a result of the deterioration of microstructure of the matrix.Zheng et al (2012)
andAwal et al (2015)indicated that the said form of variation is due to the degradation of CeSeH at temperatures beyond 450C,
which increases the volume of pores in the concrete, therefore reducing the UPV values of the concrete specimens The UPV results illustrated inFig 6also showed that the overall UPV values of the air-cooled samples were higher than that of the water-cooled concrete specimens
3.4 Residual compressive strength
The experimental results of the cube compressive strength of the concrete mixtures at the ambient temperature and upon heating to 200, 400, 600 and 800C in addition to exposure to air-and water-cooled regimes are illustrated in Fig 7 The results showed that the room-temperature compressive strength of the concrete decreased with the addition of the carpet fibers Comparing the value of the control mix (concrete without any
0
2
4
6
8
10
12
0 200 400 600 800 1000
Temperature (oC)
A1
Air cooled Water cooled
0 2 4 6 8 10 12
0 200 400 600 800 1000
Temperature (oC)
A2
Air cooled Water cooled
0
2
4
6
8
10
12
0 200 400 600 800 1000
Temperature (oC)
B1
Air cooled Water cooled
0 2 4 6 8 10 12
0 200 400 600 800 1000
Temperature (oC)
B2
Air cooled Water cooled
Fig 3 Mass loss of different concrete mixtures.
H Mohammadhosseini, J.M Yatim / Journal of Cleaner Production 144 (2017) 8e21 12
Trang 6fibers or POFA), the addition of fibers at 0.5% volume fraction
decreased the compressive strength by 6.9% Further decrease in
compressive strength of 3.3% in contrast to OPC concrete was
observed in the concrete containing 20% POFA In the fibrous
mixtures containing POFA and 0.5%fiber, the compressive strength
value decreased by 5.4% in comparison to that of OPC concrete,
which possessed the same amount offibers The said reduction was
attributed to the slow hydration and low pozzolanic activity of
POFA, which negated the increase in compressive strength, stated
byAwal et al (2015)
After heating up to 200C and subsequent cooling either by air
or water, the cube compressive strength of all four mixtures reduced by 3.35%e9.76% compared to the strength values at the ambient temperature The strength loss in this phase could be attributed to the initial moisture loss in the concrete mixtures Higher compressive strength losses were observed in both the OPC and POFA concrete specimens without carpetfibers Given this, the positive effects of carpet fibers on the residual compressive strength of concrete mixtures was clearly shown As aforemen-tioned, carpetfibers melt at temperatures between 160 and 180C.
As such, the meltedfibers, which are in liquid form, fill the holes and subsequently contribute to better performance under loads
Behnood and Ghandehari (2009) found that, the high residual compressive strength of the concrete mixtures containing fibers was therefore attributed to its dense microstructure in comparison
to that of the plain concrete
The significant influence of moisture in the concrete specimens
at elevated temperatures was established byNoumowe (2005) For partial loss of moisture until 200C, owing to the vaporization of free water, the loss in compressive strength was not significant However, with full water loss, the compressive strength dropped sharply at 400C Beyond 400C, the compressive strength loss became gradual with the increase in temperature for all mixes This phenomenon was observed in both air- and water-cooled regimes The steady degradation of compressive strength could be a result of
A1: y = 0.0113x + 0.31 R² = 0.9354 A2: y = 0.0128x + 0.0435
R² = 0.9508
0 2 4 6 8 10 12
B1: y = 0.0127x + 0.1141 R² = 0.9066 B2: y = 0.0134x + 0.3326 R² = 0.9178
0 2 4 6 8 10 12
Fig 4 Regression for the mass loss of the concrete mixtures.
H Mohammadhosseini, J.M Yatim / Journal of Cleaner Production 144 (2017) 8e21 13
Trang 70 1000
2000
3000
4000
5000
6000
27 200 400 600 800
Temperature (oC)
A2 B1 B2
0 1000 2000 3000 4000 5000 6000
27 200 400 600 800
Temperature (oC)
A2 B1 B2
Fig 6 Variation in UPV values of the concrete mixtures exposed to high temperatures.
0
10
20
30
40
50
0 200 400 600 800 1000
Temperature (oC)
W/C
0 10 20 30 40 50
0 200 400 600 800 1000
Temperature (oC)
W/C
0
10
20
30
40
50
0 200 400 600 800 1000
Temperature (oC)
W/C
0 10 20 30 40 50
0 200 400 600 800 1000
Temperature (oC)
W/C
H Mohammadhosseini, J.M Yatim / Journal of Cleaner Production 144 (2017) 8e21 14
Trang 8the complete melting process offibers as well as slow evaporation
of chemically bound water owing to the disintegration and
dehy-dration process of C-S-H gel and decomposition of Ca(OH)2, which
occurs at temperatures beyond 400 C in concrete, reported by
Noumowe et al (1994)
For higher temperatures, the residual compressive strength of
the concrete mixtures containing carpetfibers was higher than that
of OPC and POFA mixtures without anyfibers This was due to the
presence of thefibers that resulted in the reduction of spalling on
the concrete specimens At elevated temperatures, the addition of
discontinuous carpetfibers in the concrete mixtures decreased the
uneven propagation of macro-cracks and thus supported a ductile
performance Accordingly, the fibers were capable of providing
adequate loads to repress cracks opening and redistribute the
stresses against the neighboring matrix (Xiao and Falkner, 2006)
The concrete mixtures containing carpetfibers were more ductile
than that of the plain concrete with a slow decrease in strength
Therefore, the results indicated that the addition of the carpet
fi-bers resulted in the increase of ductility of concrete, with a higher
energy absorption and a well-distributed cracking, as shown in
Fig 8
In their study,Xiao and Falkner (2006)ascertained that the
re-sidual compressive strength of concrete mixtures containing
polypropylene was slightly higher than that of the mixtures
withoutfibers They observed that PP fibers melt at high
temper-atures and create networks to release thermally induced pressures
and consequently, avoid excessive loss of strength Similar results
were also reported byBehnood and Ghandehari (2009)
3.5 Relationship amongst the residual compressive strength and
ultrasonic pulse velocity
It was observed that the ultrasonic pulse velocity (UPV) values
could be correlated with the corresponding residual cube compressive strength Figs 9 and 10display a good relationship amongst the residual compressive strength and UPV values of all four concrete mixtures at high temperatures for both air-and water-cooled regimes To explain further,Fig 9 illustrates the relation-ships between the compressive strength and UPV values of the concrete mixtures containing carpetfibers and POFA for the air-cooled regime
The obtained residual cube compressive strength values were used as a response factor with the UPV values as their predicator parameter To correlate the experimental data, linear regression method was applied, resulting in Eqs.(5)e(8), with R2values of between approximately 0.72 and 0.77 for all samples, which signified good confidence for the relationships They are as the following:
frcuA¼ 0.0137VA 20.944 (R2¼ 0.7264) (5)
frcuA¼ 0.0111VA 10.187 (R2¼ 0.7749) (6)
frcuA¼ 0.0136VA 21.705 (R2¼ 0.7396) (7)
frcuA¼ 0.0106VA 10.038 (R2¼ 0.7725) (8) where frcuAis the residual cube compressive strength and VA
sig-nifies the residual UPV for the air-cooled regime at high temperatures
In addition,Fig 10presents the relationships between the re-sidual compressive strength and UPV values of OPC and POFA mixtures with and without carpetfibers for water-cooled regime Linear regression method was used to correlate the experimental, which resulted in Eqs.(9)e(12), with R2values of between of 0.77 and 0.81 for all specimens, as follows:
frcuW¼ 0.0141VW 20.982 (R2¼ 0.7705) (9)
frcuW¼ 0.0115VW 10.574 (R2¼ 0.8103) (10)
frcuW¼ 0.0139VW 21.016 (R2¼ 0.779) (11)
frcuW¼ 0.0109VW 9.7035 (R2¼ 0.8155) (12) where frcuWrepresents the residual cube compressive strength and
VW is the residual UPV for the water-cooled regime at elevated temperatures
The empirical parameters of the equations attained in this study were almost comparable to those stated bySuhaendi and Horiguchi (2006)for concrete containing polypropylenefibers andAwal et al (2015)for plain concrete containing POFA The correlations in this study showed that the application of UPV measurement could be applied in the inspection of the properties offire-damaged concrete
in terms of compressive strength in a faster and more efficient way However, the establishment of solid correlations wouldfirst require more essential experimental data from the process
3.6 Scanning electron microscopy (SEM)
Scanning electron microscopy (SEM) investigations demon-strated different variations in the morphology of OPC and POFA concrete mixtures with and without carpet fibers at designated temperatures Fig 11 reveals the SEM of unheated and heated concrete specimens exposed to 200 and 800C The SEM of OPC and POFA concrete mixtures at the ambient temperature (27 C) showed the C-S-H gel formation and a continuous structure without micro cracks and pores As seen inFig 11a, on the 90 days
27
200
800
H Mohammadhosseini, J.M Yatim / Journal of Cleaner Production 144 (2017) 8e21 15
Trang 9curing period, the C-S- H gel was more evenly spared in POFA
concrete in comparison to OPC Thefinely spared of C-S-H gel and
development of extra C-S-H gel due to the consumption of
por-tlandite by pozzolanic action of POFA resulted in better
perfor-mance in the concrete mixtures According toAwal et al (2015), it is
due to the fact that POFA modified the concrete matrix through the
pozzolanic reaction and reduced the Ca(OH)2content
Fig 11b shows a slight number of micro cracks, which were
detected at 200 C, while a fairly large amount of micro cracks
occurred in the cement matrices at the temperature of 800C At
the latter temperature, the OPC and POFA matrices turned into an
amorphous structure and thus large amount of cracks appeared
throughout the concrete specimens as illustrated inFig 11c It can
be seen that at elevated temperatures, the matrices of POFA was
more compact than that of OPC However, at the highest
temper-ature of 800 C, the microstructures of all specimens were
extremely damaged, which led to the deterioration of C-S-H The
results of the present work were in agreement with that described
byNoumowe (2005)
Fig 12displays the general views of the carpetfibers dispersed
in unheated concrete mixtures SEM analysis demonstrated that the
carpetfibers acted as bridges across cracks and pores It was also
shown that the fibers were tightly wrapped by the C-S-H gels
Fig 12a also reflects a strong bond between the carpet fibers and cement matrix The SEM image further revealed that the carpet fibers along with cement paste provided a strong interfacial bonding, which resulted in smaller crack size on the interface
In addition,Fig 12b shows the concrete containing carpetfibers after exposure to high temperatures At the ambient temperature, it can be seen that the fibers had star cross section However, at
200C, the carpetfibers had lost their solid structure in both OPC and POFA concrete mixtures A significant change in the bond be-tween the carpetfibers and cement matrix of the concrete mixtures upon exposure at 200C was found The said finding could be attributed to the formation of the micro cracks and therefore reduction in the bonding between thefibers and cement matrix Upon heating the concrete specimens up to 800C, the carpetfibers melted and evaporated, formed an additional network in the mixture that could act to release high internal pressures.Fig 12c presents the effects of the meltedfibers The use of PP fibers clearly affected the porosity of the concrete at elevated temperatures In addition, it could even reduce the pore pressure inside the concrete Similar observations have been found byNoumowe (2005) and
S¸ahmaran et al (2011)
y = 0.0137x - 20.944 R² = 0.7264
0
10
20
30
40
50
60
1000 2000 3000 4000 5000
UPV (m/s)
A1-A/C
y = 0.0111x - 10.187 R² = 0.7749
0 10 20 30 40 50
1000 2000 3000 4000 5000
UPV (m/s) A2-A/C
y = 0.0136x - 21.705 R² = 0.7396
0
10
20
30
40
50
1000 2000 3000 4000 5000
UPV (m/s)
B1-A/C
y = 0.0106x - 10.038 R² = 0.7725
0 10 20 30 40 50
1000 2000 3000 4000 5000
UPV (m/s) B2-A/C
Fig 9 Correlation amongst the residual UPV and compressive strength of the air-cooled regime.
H Mohammadhosseini, J.M Yatim / Journal of Cleaner Production 144 (2017) 8e21 16
Trang 103.7 Thermogravimetric analysis (TGA) and differential thermal
analysis (DTA)
When a composite material such as concrete is exposed to
elevated temperatures of between 100 and 900 C, numerous
chemical and physical phenomena occur The reactions take place
throughout the heating of the concrete mixtures.Fig 13a reveals
DTA results of the concrete specimens at the ambient temperature
It can be seen that some physical phenomena had occurred in the
temperature series between 80 and 180 C The evaporation of
water in the cement matrix below 110C, C-S-H dehydration as
well as shrinkage and melting offibers at approximately 170 C
were the most important issues during the said phase The
obser-vations made in this study were comparable to those found by
Noumowe (2005)andFares et al (2015)
Few endothermic peaks were observed in the unheated
speci-mens, which were 85e130, 470, 555, 660 and 750C The detected
peaks of heat flow were related to the temperatures of phase
transition of the hydrates in the cement paste as well as the melting
of carpetfibers There was no significant distinction between OPC
and POFA concrete mixtures comprising waste carpetfibers for the
entire heating process A dual peak at 80e130C was also observed,
which was attributed to the vanishing of free and bound water in hydrates like C-S-H gel (Fares et al., 2015)
Noumowe (2005) stated that the small variation of heatflow between 170 and 480 C could be attributed to a continuous dehydration of the C-S-H gel and melting of PPfibers According to
Sun and Xu (2008), the said difference owes to the decomposition
offibers at 200e300C, release of free water of hydrates,first phase
of dehydration and failure of C-S-H gel structure.Sun and Xu (2008)
also found that between 600 and 700C, the hydroluminate de-composes and formsb-C2S and at approximately 720C, the cal-cium carbonate (CaCO3) decomposes, thus permitting CO2 to liberate from the concrete There were some differences observed
in OPC and POFA concrete mixtures during the heating process At below 110C, both B1 and B2 mixtures containing POFA offered a higher peak than both A1 and A2 specimens with OPC only Ac-cording toAwal et al (2015), this could be attributed to the lower unit weight of POFA compared to OPC that resulted in the increased volume of mixtures and therefore, more free water in the specimens
Fig 13b illustrates TGA results of the concrete specimens at the ambient temperature Various reductions in mass, equivalent to the dehydration,fibers melting and phases variations are presented
y = 0.0141x - 20.982 R² = 0.7705
0
10
20
30
40
50
60
1000 2000 3000 4000 5000
UPV (m/s)
A1-W/C
y = 0.0115x - 10.574 R² = 0.8103
0 10 20 30 40 50
1000 2000 3000 4000 5000
UPV (m/s) A2-W/C
y = 0.0139x - 21.016 R² = 0.779
0
10
20
30
40
50
1000 2000 3000 4000 5000
UPV (m/s)
B1-W/C
y = 0.0109x - 9.7035 R² = 0.8155
0 10 20 30 40 50
1000 2000 3000 4000 5000
UPV (m/s) B2-W/C
Fig 10 Correlation amongst the residual UPV and compressive strength of the water-cooled regime.
H Mohammadhosseini, J.M Yatim / Journal of Cleaner Production 144 (2017) 8e21 17