KSCE Journal of Civil Engineering

Surface area model for the assessment of plastic aggregate concrete properties
--Manuscript Draft--

Manuscript Number:

Full Title: Surface area model for the assessment of plastic aggregate concrete properties

Article Type: Research Paper

Corresponding Author: Frya Jafr, MSc

Sulaimani Polytechnic University
Sulaymaneyah, IRAQ

Corresponding Author Secondary

Information:

Corresponding Author's Institution: Sulaimani Polytechnic University

Corresponding Author's Secondary

Institution:

First Author: Frya Jafr, MSc

First Author Secondary Information:

Order of Authors: Frya Jafr, MSc

Azad A. Mohammed, PhD

Hemn U. Ahmed, MSc

Order of Authors Secondary Information:

Funding Information:

Abstract: In this study, a model was proposed based on the calculation of total surface area of
aggregate to assess fresh density, compressive and splitting tensile strengths of plastic
aggregate (PA) recycled concrete. The key factor is the change of total surface area of
natural aggregate by the PA, and for a given PA volume the change in the property
could be assessed. The prediction is well depend on size distribution, specific gravity
and bulk density of natural and plastic aggregates. The proposed model prediction was
found accurate when applied on high strength and lightweight concretes, mainly
because of relatively good PA-cement paste bond. For the majority of concrete mixes
investigated, the model is moderately underestimates strength loss. The reason could
be attributed to the low PA-cement paste bond efficiency. An attempt was made to
assess the bond deficiency parameter to obtain more accurate prediction.

Author Comments: Dear Respected Editor,

I hope you are very well.

This is a work that is unique of its kind and I believe it would benefit the scientific
community.

Thank you so much and have a great time,

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
 Click here to access/download;Manuscript;Manuscript.docx

1
2
3
4
5 Surface area model for the assessment of plastic aggregate
6
7 concrete properties
8
9 Frya S. Jafr1*, Azad A. Mohammed2, Hemn U. Ahmed3
10 1
11 Technical College of Applied Science, Sulaimani Polytechnic University,
12 Sulaimani, Kurdistan Region, Iraq.
13
14
15
16 2,3
17
College of Engineering, University of Sulaimani, Sulaimani, Kurdistan Region,
18 Iraq.
19
20 1
21 E-mail: fryahawrami@gmail.com
2
22 E-mail: azad.mohammed@univsul.edu.iq
23 3
24
E-mail: hemn.ahmed@univsul.edu.iq
25 * Corresponding author
26
27
28
29
Abstract: In this study, a model was proposed based on the calculation of total surface area of aggregate
30
31 to assess fresh density, compressive and splitting tensile strengths of plastic aggregate (PA) recycled
32 concrete. The key factor is the change of total surface area of natural aggregate by the PA, and for a given
33 PA volume the change in the property could be assessed. The prediction is well depend on size
34 distribution, specific gravity and bulk density of natural and plastic aggregates. The proposed model
35 prediction was found accurate when applied on high strength and lightweight concretes, mainly because
36 of relatively good PA-cement paste bond. For the majority of concrete mixes investigated, the model is
37 moderately underestimates strength loss. The reason could be attributed to the low PA-cement paste bond
38 efficiency. An attempt was made to assess the bond deficiency parameter to obtain more accurate
39
40
prediction.
41
42
43 Keywords: Aggregate surface area, Compressive strength, Fresh density, Natural aggregate, Plastic
44 aggregate, Splitting tensile strength.
45
46 1. Introduction
47 Currently used plastics are mainly categorized to seven most popular types and commonly used
48 plastics are acrylic or polymethyl methacrylate (PMMA), polycarbonate (PC), polyethylene
49
50 (PE), polypropylene (PP), polyethylene terephthalate (PETE or PET), polyvinyl chloride (PVC),
51 acrylonitrile-butadiene-styrene (ABS). The highly consumed plastics worldwide usually
52 accompanied with a huge amount of solid waste needs to be recycled to avoid environmental
53 pollution. This pollution will create many problems; it reduces the natural beauty of the
54
55
environment, confinement and death of aquatic organisms, and blockage of sewage systems in
56 cities, especially in developing countries, causing other illnesses. Researchers estimate that more
57 than 40% of the world’s waste is incinerated. However, burning of plastics and other wastes
58 release toxic substances such as heavy metals, persistent organic pollutants (POPs), and other
59
60
61
62 1
63
64
65
 1
2
3
4 toxic chemicals into the atmosphere. To ovoid this, researchers have worked to find out the best
5
6 rout of plastic waste management to save the environment against further pollution.
7 Recycling virgin plastic to make a new product is, however, accompanied with a fate known as
8 down-recycling, which results in the plastic being used for low quality materials when compared
9 to its original use [1]. It must be emphasized that in order to recycle a solid plastic, there is a
10
11
need for plasticizers. In developing countries, low cost plasticizer is attempted, and this sort of
12 material is categorized as cancerous one and may cause health problems. As an alternative,
13 recycling shredded plastic in the form of aggregate or fiber for concrete production may has a
14 promised future.
15 If plastics are used as an aggregate in concrete, because of absorbing energy by the plastic
16
17 particle, post peak of the response becomes more ductile [2]. In general, using plastic aggregate
18 instead of natural sand will lead to strength loss depending on the level of replacement.
19 However, tests show that compressive strength loss is not always take place, in which there are
20 records indicate a very low compressive strength loss, and even strength enhancement when low
21 ratio of plastic aggregate is added to concrete [3,4]. Knowing that due to the low density of
22
23 plastic particle, the produced concrete will tends to be lightweight concrete. Further, there is a
24 good occasion to construct composite section based on two different concretes; using plastic
25 aggregate concrete as an infill core between two strong outer skins (or sandwich panel). This
26 novel composite section may have a promised future in precast concrete industry. However, this
27
28
topic is relatively new and the relevant researches are quite limited. Furthermore, records [5]
29 have shown that using plastic aggregate instead of natural sand has the ability to enhance impact
30 strength of concrete.
31
32 2. State-of-art review
33
34 It is important here to discuss the effect of plastic aggregate shape and size on the residual
35 properties of concrete as an introduction to the proposed aggregate surface area model. Effect of
36 different plastic ratios as fine aggregate replacement for a given type of plastic aggregate used
37 was investigated by the past researchers such as Mohammed et al. [2], Al-Hadithi and Alani [6]
38
39
and Kou et al. [7]. However, there are several experimental attempts [8,9,10,11,12] to investigate
40 the role of plastic aggregate particle size and shape on some important properties of recycled
41 concrete. Below, results of experimental works performed by the past researchers on some
42 important properties of recycled concrete are given and discussed.
43 One of the early attempts was that made by Marzouk et al. [13] in 2007, in which they
44
45 investigated compressive strength, flexural strength, and elastic modulus of concrete of natural
46 sand replaced by 2%, 5%, 10%, 15%, 20%, 30%, 50% 70% and 100% PET aggregate of three
47 maximum sizes: 0.1, 0.3 and 0.5 cm. There is some compressive strength enhancement when
48 larger sizes of plastic aggregate are added by 5% followed by strength loss with increasing
49
plastic aggregate (PA) volume. Their results indicate that using smaller size PA lead to higher
50
51 compressive strength loss regardless of the PA volume in the mix. The behavior of the PA
52 concrete in flexure was similar to that in compression with regard the effect of particle size on
53 the strength loss. Elastic modulus of concrete followed the same rule of compressive and flexural
54 strengths. Also, using PA aggregate of the maximum size of 5 mm had better action in concrete
55
56
compared with the lower sizes aggregate.
57 In an experimental study, Saikia and de Brito [10] tested concrete with sand replaced by three
58 types of plastic aggregate: coarse particle (PC), fine particle (PF) and heat-treated pellet (PP).
59 Their test data show that with increasing plastic content in the mix up to 15% there was a need to
60
61
62 2
63
64
65
 1
2
3
4 increase w/c ratio to maintain workability in terms of slump, indicating the reduction of slump
5
6 with increasing plastic content. There was a need to use higher w/c ratio for coarser plastic
7 particles as compared with finer particles to maintain nearly the same workability, and w/c ratio
8 for pellet particles was lower than that of flaky PET particle aggregate. These results clearly
9 indicate the change of workability with changing the shape and size of plastic aggregate, in
10
11
which using pellet particle aggregate will give higher workability compared with flaky PET
12 particles. They observed a continuous reduction of compressive strength with increasing plastic
13 aggregate content, being increased with increasing size of plastic and flakiness of particles. Their
14 strength loss was relatively high in which there was 65% strength loss of concrete on using 15%
15 coarse plastic particles tested at 28 days. The performance of heat-treated pellets having thick
16
17 cross-section was better than that of flaky plastic particles to control strength loss regardless of
18 plastic ratio and curing time.
19 Further, Albano et al. [11] investigated slump of concrete with natural sand replaced by 10% and
20 20% PET aggregate of 1.14 cm, 0.26 cm and combination of both sizes. Their results showed a
21 reduction of slump with increasing PA ratio. In general, using coarse particle plastic aggregate
22
23 will lead to higher slump loss. They also investigated compressive strength of concrete at the
24 ages of 7,14,28 and 60 days. Results showed a compressive strength loss with increasing PET
25 aggregate ratio in concrete, being increased with increasing size of plastic particles for concrete
26 mixes with w/c ratios of 0.5 and 0.6. Furthermore, Hannawi et al. [9] tested cement mortar made
27
28
of natural sand partially replaced with polycarbonate (PC) and polyethylene terephthalate (PET).
29 Various volume fractions of sand (3%, 10%, 20%, and 50%) were replaced by the same volume
30 of PA. There was a gradual compressive strength loss with increasing plastic aggregate in the
31 mortar. Although the maximum size of PET aggregate was 10 mm and that of PC was 5 mm, test
32 data show that there is a negligible difference between the two compressive strength losses.
33
34 From the foregoing presentation, one can find the effect of plastic particle geometry or shape and
35 size on the residual compressive strength of the recycled concrete. On this base, any proposed
36 equation for calculating compressive strength and other concrete properties will be accurate after
37 taking into account the effect of shape and size of aggregate particle. This effect could be
38
39
collected in the physical properties of the plastic particle, in particular specific gravity and bulk
40 density in addition to particle size distribution. In the sections to follow, the outline of aggregate
41 surface area model for the assessment of residual properties of recycled concrete is described.
42
43 3. Aggregate surface area model for PA concrete properties
44
45 In this proposed model, the effect of plastic aggregate particle’s size and shape and even surface
46 smoothness are included to assess accurately the residual fresh density, compressive strength and
47 splitting tensile strength of PA concrete. The author thinks that the proposed model is a new
48 attempt to assess properties of recycled concrete containing PA via calculating surface area of
49
natural and PAs. The change of a surface area of a relatively strong particle aggregate with the
50
51 weak plastic particle one is a key factor representing the change takes place in the structure of
52 hardened concrete leading to the change of the residual properties. The predictions are compared
53 with the test data to check the validity of the model. There was necessary to make some rational
54 assumptions to work on the model because some experiments contain no sufficient data related
55
56
to PA properties, in particular PA thickness.
57
58
59
60
61
62 3
63
64
65
 1
2
3
4 3.2 Calculation of aggregate surface area
5
6 3.2.1 Surface area of natural aggregate
7 Several methods for calculating total surface area or specific surface area (SSA) of aggregate
8 particles are available. Assessment of SSA has a significant effect on its workability of concrete
9
10
in fresh state [14]. Another goal behind calculating total surface area of aggregate is basically
11 related to the design of asphalt mixture [15]. The mathematical approximation of surface area is
12 based on the assumption of a spherical shape for the particles. This approximation will lead to
13 failure of taking into account the effect of shape and the square–cube law. For this purpose,
14 Ghasemi et al. [14] worked on the possibility of replacing the assumption of a spherical shape
15
16 with that of Platonic solids as the representative shape to account for the angularity of
17 aggregates. Carr et al. [16] developed an empirical method based on fractional dimension for
18 estimating surface area of aggregate particles not smaller than 1 mm. However, one of the
19 simplest procedures for the assessment of SSA is that given by Panda et al. [15]. They have
20
21
proposed a simple procedure for calculating total surface area of dust, coarse and fine aggregates
22 for the purpose of hot mix asphalt (HMA) design. This procedure is based on: (a) calculating
23 mean size of the three components, (b) assuming shape of particle, cylindrical for coarse and fine
24 aggregates and spherical for dust, (c) calculating volume of particle, (d) calculating theoretical
25 and actual number of particles in one m3 of aggregate, and (e) calculating surface area of each
26
27 particle and calculating total surface area in m2. Below, steps followed for calculating total
28 surface area of natural aggregate are presented.
29 The surface area of the aggregate blend is calculated as per empirical formula given below:
30
31
SA-Total = SA-CA + SA-FA + SA-DUST (1)
32
33
34 in which SA-CA is total surface area of coarse aggregate, SA-FA is total surface area of fine
35 aggregate, and SA-DUST is total surface area of dust (particles below 75 microns) all in one cubic
36
37
meter of blend. Panda et al. [15] assumed cylindrical shape for coarse and fine aggregate
38 particles and spherical shape for dust particle (see Fig. 1), this to make a simplification for
39 calculating total surface area aggregate in m2/kg base on surface area of one particle.
40 In plastic aggregate recycled concrete, fine aggregate is mostly replaced with the plastic one, and
41 for this reason calculation of total surface area of fine aggregate is given herein. Surface area of
42
43 fine aggregate having particle size above 75 micron and below 4.75 mm is calculated as follows:
44
45 SAFA = NFAMA-FAFFA (2)
46
47
48
where NFA is number of fine aggregate particle in one cubic meter of concrete, MA-FA is area of
49 mean size of fine aggregate and FFA is multiplying factor for unevenness of fine aggregate. For
50 all i having sieve above 4.75 mm, mean size of fine aggregate (MSFA) is given by:
51
52
𝑀𝑖 𝑀𝑆𝑖2
53 MSFA = √ (3)
54 ∑𝑀𝑖
55
56 where Mi is the percentage mass of aggregates passing in (i +1)th sieve minus ith sieve, MSi is
57 the mean size of (i +1)th sieve and ith sieve. Mean length of fine aggregate particle is given by:
58
59
60 MLFA = MSFA (1+Total flakiness & Elongation) (4)
61
62 4
63
64
65
 1
2
3
4
5
6 Fine aggregate particle is assumed to be cylindrical; particle’s surface area and volume are given
7 by the following equations:
8
2 𝜋
9 MAFA = πMSFAMLFA + 2 𝑀𝑆𝐹𝐴 (5)
10
11
2 𝜋
12 MVFA = 4 𝑀𝑆𝐹𝐴 𝑀𝐿𝐹𝐴 (6)
13
14
15 Number of particles per volume is related to the voids ratio in the following equation:
16
17 1
18 NFA = 𝑀𝑉 (1 − 𝑒)𝑊𝐹𝐴 (7)
𝐹𝐴
19
20
21 where WFA is the weightage of fine aggregate in the mix.
22 The same procedure is followed for calculating surface area of coarse aggregate (SA-CA) having
23 particle size above 4.75 mm.
24 Surface area of dust (particles below 75 microns) is given by
25
26
27 SA-DUST = NDUSTMA-DUSTFDUST (8)
28
29 where NDUST is number of dust particle in one cubic meter of concrete , MA-DUST is area of mean
30 size of dust particle and FDUST is multiplying factor for unevenness of dust.
31
32 The main difference between dust and aggregate particles is that the shape of dust is assumed to
33 be spherical (see Fig. 1). On this base, area and volume of dust particle is given in the following
34 equations respectively
35
36 2
37
MADUST = π 𝑀𝑆𝐷𝑈𝑆𝑇 (9)
38
3 𝜋
39 MVDUST = 6 𝑀𝑆𝐷𝑈𝑆𝑇 (10)
40
41
42 where MSDUST is the mean size of dust.
43 Number of dust particles per volume is related to the voids ratio (e) in the following equation:
44
45 1
46 NDUST = 𝑀𝑉 (1 − 𝑒)𝑊𝐷𝑈𝑆𝑇 (11)
𝐷𝑈𝑆𝑇
47
48
49 where WDUST is the weightage of dust in the mix. It will be noted that number of particles
50 depends on the void ratio of the bulk aggregate and dust and could be calculated as follows [17]
51
52 𝜌
53
e = 1- 𝐺.𝜌𝐴 (12)
𝑤
54
55 where ρA is bulk density of aggregate (kg/m3), G is specific gravity of aggregate and ρw density of
56
57 water (kg/m3).
58 Panda et al. [15] have used the following unevenness factors for the dust and aggregate:
59 FDUST =1.6 (additional sixty percent towards unevenness),
60 FFA = 1.6 x 2 = 3.2 (additional 100 percent towards unevenness as compared to dust), and
61
62 5
63
64
65
 1
2
3
4 FCA = 3.2 x 2 = 6.4 (additional 100 percent towards unevenness as compared to fine aggregates).
5
6 Knowing that these factors are for the case of aggregate used for hot asphalt mixture, which is
7 mainly crushed stone aggregate.
8 Based on the foregoing presentation, the steps below are followed for calculating surface area of
9 dust particles:
10
11
a- Average size of dust passing 75 micron (MSDUST) is taken as 0.04 mm, assuming spherical
12 particles, theoretical volume of dust particle (MVDUST) is 3.35238 x 10-5 mm3.
13 b- Calculate voids ratio from Eq. 12.
14 c- From sieve analysis, dust content percentage (WDUST) is determined, and from Eq. 11 number
15 of particles in one cubic meter (NDUST) is determined.
16
17 d- Having surface area of 0.04 mm dust particle, which is equal to 0.005028571 mm2 using Eq.
18 9, total area of dust is determined (in m2). Assume multiplying factor as 60%, higher on dust
19 particle for uneven surface texture (which is 1.60), the total area of dust (after correction) is
20 calculated. The same procedure given in steps a to d is followed for fine aggregate of total
21 weight retained on fine aggregate sieves (WF) and retained on coarse aggregate sieves (WC) to
22
23 calculate the two surface areas, later from Eq. 1, total surface area is determined.
24 For the recycled concrete made of natural river aggregate, however, calculation of surface area of
25 dust seems to be not required, because the existence of dust particles is related to the crushed
26 stone aggregate. Nearly, all researchers that worked on the properties of concrete containing PA
27
28
give no information about dust particle in their sieve analysis.
29
30 3.2.2 Surface area of PA
31
Shape of heat treated plastic pellet could be considered as cylindrical similar to that of natural
32
33 coarse and fine aggregates. Also, any flaky particle of mean size not larger that the thickness of
34 plastic flakes is treated as pellets. The main difference between plastic pellet and natural
35 aggregate is related to their surface texture. As mentioned before, unevenness factors of 3.2 for
36 fine aggregate is recommended, but due to the smooth surface of plastic particle, as the author
37
38
thinks, there is no need to magnify the calculated area such as that done for natural aggregate. On
39 this base, there is high surface area reduction on replacing natural aggregate with plastic one, and
40 consequently there is lower total bond area of embedded plastic-natural particles with cement
41 paste as compared with that of natural aggregate with cement paste.
42 Using plastic flakes of thickness smaller than its mean size (see Fig. 1c) need to be treated
43
44 separately. In this way, the effect of the particle’s shape could be well included to assess the
45 surface area accurately. In general, plastic flakes used by some researchers such as Kou et al. [7]
46 (obtained from shredding PVC pipe of thickness about 3 mm) could be treated as pellet particles
47 because the thickness is relatively large. For flake particles surface area and volume are given by
48
49
the two equations:
50
51 MAF = 2 MLF*MSF (13)
52
53 MVF = tF* MLF*MSF (14)
54
55
56 Where MAF = mean surface area of flake, MVF = mean volume of flake, tf = thickness of flake,
57 MSF = mean size of flake and MLF = mean length of flake.
58
59
60
61
62 6
63
64
65
 1
2
3
4 3.3. Calculation of fresh density loss
5
6 First application of the surface area model is the prediction of density of fresh concrete. Initial
7 unit weight of control mix without plastic aggregate (in kg/m3) is given by
8
9 Wi = Ww + Wc + WCA + WS (15)
10
11
12 Final weight of fresh concrete containing PA as fine aggregate replacement (in kg/m3) is given
13 by
14
15 𝑅𝑁𝑝 𝑤𝑝𝑝 +(1−𝑅)𝑁𝑠 𝑤𝑝𝑠
16 Wf = Ww + Wc + WCA + WS [ ] (16)
𝑁𝑠 𝑤𝑠𝑝
17
18
19 Where Ww is weight of water, Wc is weight of cement, WCA is weight of coarse aggregate and Ws
20 is weight of sand, all in kg/m3. R is ratio of sand replacement with PA. Np is number of PA
21 particles in one cubic meter of aggregate, wpp is weight of one PA particle, Ns is number of sand
22
23
particles in one cubic meter of aggregate, and wps is weight of one particle of sand. Results are
24 represented in terms of fresh density ratio which is calculated by dividing Wf by Wi.
25 3.4. Calculation of strength loss
26 In this proposed model, it is assumed that there is no compressive strength enhancement due to
27 natural aggregate replacement with plastic one, and there is always a strength loss. Calculation of
28
29 the strength loss is based on the fact that a portion of active surface area of natural aggregate is
30 replaced with the non-active weak surface area provided by the PA being increased with
31 increasing replacement ratio. For calculating plastic surface area, important properties of plastic
32 material of specific gravity and bulk density are included via calculation of voids ratio given in
33
Eq. 12. On this base, normalized residual compressive strength loss is assumed to depend on the
34
35 change of surface area (modified surface area to basic surface area ratio) given by:
36
′
𝑓𝑐𝑝 𝑆𝐴𝑓
37 𝑆𝐴𝑁 ∗ 𝑉𝑁 + 𝑆𝐴𝑃 ∗ 𝑉𝑃
= = (17)
38 𝑓𝑐′ 𝑆𝐴𝑖 𝑆𝐴𝑁
39
40 ′
41 where𝑓𝑐𝑝 is compressive strength of concrete containing plastic aggregate, fc’ is compressive
42 strength of control concrete, SAf is final surface area, SAi is initial surface area, SAN is surface
43 area of natural aggregate, SAP is surface area of PA, VN is volume ratio of natural aggregate and
44 VP is volume ratio of PA.
45
46 With regard the splitting tensile strength loss, it is assumed that the tensile strength loss due to
47 plastic aggregate addition is similar to that of compressive strength loss, and on this base
48 splitting tensile strength ratio is given by
49
50 𝑓𝑠𝑝𝑝 𝑆𝐴𝑓 𝑆𝐴𝑁 ∗ 𝑉𝑁 + 𝑆𝐴𝑃 ∗ 𝑉𝑃
51 = = (18)
𝑓𝑠𝑝 𝑆𝐴𝑖 𝑆𝐴𝑁
52
53
54 where fspp is splitting tensile strength of concrete containing PA and fsp is splitting tensile strength
55 of control concrete. However, there may be a difference between compressive strength loss and
56 splitting tensile strength loss and the source of difference could be related to the effect of PA
57
58
particle’s smoothness and hydrophobicity on the bond between cement paste and PA discussed
59 later.
60 In order to work on the proposed surface area model:
61
62 7
63
64
65
 1
2
3
4 (a) grading of natural and plastic aggregates obtained from sieve analysis must be given,
5
6 (b) test data on specific gravity and bulk density of the two aggregates must be given, and
7 (c) thickness of flaky aggregate must be given.
8 It should be noted that necessary test data on aggregate properties to be used for calculating total
9 surface area have not been given by some authors. These experiments are omitted in the
10
11
comparison between test and predicted property in question. However, a limited assumption
12 related to the plastic thickness was made.
13
14 4. Validity of the proposed model
15 It is important herein to check the accuracy of the three predictions, and for this purpose, test
16
17 data from published literature were utilized. Comparison with test data will indicate the accuracy
18 of the proposed model.
19
20 4.1 Fresh density
21 First of all, there is a limited occasion to compare the predictions of fresh concrete density with
22
23 the test one measured by some researchers, because there are limited works on concrete
24 containing PA that attempted to measure fresh density. Secondly, majority of works conducted
25 contain no necessary data lead us to work on the surface area model because of missing
26 information about grading of natural and plastic aggregates used and/or information on specific
27
28
gravity and bulk density of both aggregates. However, an attempt was made to collect those
29 works contain necessary data to check the accuracy of the model when applied on the density of
30 fresh concrete.
31 Based on test data given by Rahmani et al. [4], number of particles of fine and PET aggregates
32 was calculated using Eq. 7 and tabulated in Table 1. This table also shows results of calculated
33
34 fresh density using Eq. 16 accompanying with the test one, from which one can find an accurate
35 prediction of mean test/calculated value equal to 1.004.
36 da Silva et al. [18] worked on cement mortar of sand partially replaced with two types of plastic
37 aggregates: shredded PET particles and heat-treated pellet. Necessary calculations steps were
38
39
followed to obtain number of natural and plastic aggregate particles and fresh density and the
40 results of calculation in addition to the test density measured are given in Table 1. One can find
41 that the quality of predictions when applied on mortar of fine aggregate replaced with PF or PP
42 aggregate is not good as compared with that on concrete. The test/calculated ratio of fresh
43 density is equal to 1.089 higher than that obtained for the case of concrete tested by Rahmani et
44
45 al. [4]. As an average, results of analysis show us that the test/calculated ratio of fresh density is
46 1.046, which is reasonably accurate.
47
48 4.2. Compressive and splitting tensile strengths
49
Here, an attempt is made to calculate the residual compressive and splitting tensile strengths of
50
51 concrete due to fine aggregate replacement with PA. Eqs. 17 and 18 are utilized to predict the
52 two properties based on data input taken from some past researchers worked on different types of
53 PA in their concrete mixes. Table 2 shows results of calculated number of particles, total surface
54 area and predicted compressive and splitting tensile strengths in addition to the test values. Figs.
55
56
2 and 3 show test and calculated compressive and splitting tensile strengths based on data on
57 high strength concrete and lightweight concrete respectively. The data given in Table 2 are for
58 concrete mix tested in the age of 28 days. Below, in some detail, results of predictions are
59 discussed.
60
61
62 8
63
64
65
 1
2
3
4
5
6 Normal strength concrete
7 By Rahmani et al. [4], no information have given for the thickness of PET particles, and it is
8 assumed here to be 0.2 mm as an average, based on visual observation of the PET picture given.
9 For the case of normal strength mixes, results indicate that mean test/calculated compressive
10
11
strength ratio is equal to 1.02 (only 2% error) indicating the accuracy of the proposed model
12 when applied on concrete containing PET aggregate tested by Rahmani et al. [4]. In contrast,
13 predictions of splitting tensile strength is relatively not good, since the test/calculated tensile
14 strength ratio is 0.923, and accordingly the predictions overestimate test data by 7.7%.
15 Since there is no information about thickness of PET particles given by Saikia and de Brito [10]
16
17 it is assumed to be 0.2 mm as an average. One can observe that the predicted splitting tensile
18 strength is quite close to that measured by these researchers for the mixture containing PP
19 aggregate, but appreciably overestimates test data for the mix containing flaky PET aggregate.
20 This will inform us that the particle’s flakiness has additional effect not been included just by
21 considering total surface area of aggregate. Results indicate that mean test/calculated
22
23 compressive strength ratio is equal to 0.739 and 0.913 for concrete mix containing PET and PP
24 aggregate respectively. Accordingly, the predicted model underestimates compressive strength
25 loss because of neglecting the effect of plastic-hardened cement paste bond deficiency. It will be
26 noted that this effect is more important for PET aggregate because of particle’s flaky nature as
27
28
compared with PP aggregate particles. For concrete tested at the age of 7 days, mean
29 test/calculated compressive strength ratio was found to be 0.834 and was found to be 0.77 for
30 concrete tested at 91 days.
31 Results of Table 2 indicate that mean test/calculated splitting tensile strength ratio is equal to
32 0.778 and 1.003 for concrete mix containing PET and PP aggregate respectively. Accordingly,
33
34 the predicted model underestimates splitting tensile strength loss for concrete with PET
35 aggregate, and consequently there is a need to account for the effect of aggregate plastic-
36 hardened cement paste bond deficiency to obtain more accurate prediction.
37 Results of data analysis indicate that mean test/calculated compressive strength ratio is equal to
38
39
0.967 when applied on data by Mohammed et al. [2] accompanying with non-safety of the model
40 since the calculated strength ratio overestimating the test ratio. According to the obtained results,
41 test/calculated splitting tensile strength ratio is 0.949 and again the model predictions
42 overestimates test ratio. On this base, there is similarity between using PVC aggregate and PET
43 aggregate in concrete in which there is another parameter governing the strength loss which is
44
45 the aggregate plastic-hardened cement paste bond deficiency, mainly because of hydrophobicity
46 of plastic particle embedded in concrete.
47 Based on data by Ghernouti et al. [19], results indicate that mean test/calculated compressive
48 strength ratio is equal to 0.925. This ratio is relatively high as compared with that of concrete
49
containing plastic flake particles. On this base, the effect of plastic aggregate-hardened cement
50
51 paste bond deficiency is low in heat-treated plastic pellets as compared with than of flake PA.
52
53 Mortar
54 da Silva et al. [18] results indicate that mean test/calculated compressive strength ratio is equal to
55
56
0.716 and 0.948 for concrete with PF and PP respectively, indicating that the model prediction
57 overestimates test strength ratio. It will be noted that the predictions has more overestimation for
58 the case of PF aggregate supporting our conclusion about the bond deficiency because of
59
60
61
62 9
63
64
65
 1
2
3
4 hydrophobicity of plastic particle being higher when plastic flakes (PF) are used in concrete as
5
6 compared with plastic pellets (PP).
7
8 High strength concrete
9 Results of Table 2 indicate that mean test/calculated compressive strength tested by Mohammed
10
11
and Mohammed [8] is equal to 1.06 and 1.022 for concrete with fine PVC and coarse PVC
12 aggregates respectively, tested at the age of 28 days. One can note a different prediction as
13 compared with the previous cases in which the model underestimates test data. On this base the
14 nature of bond between hardened cement paste and PVC particle is different and the transition
15 zone seems to be relatively strong. The author thinks that this behavior is because of using lower
16
17 water/binder ratio able to reduce the action of hydrophobicity of plastic embedded in cement
18 paste, and also because of using silica fume able to improve the interfacial bond between plastic
19 particle and cement paste. However, to learn more about these two actions, there is a need for
20 further researches. Fig. 2 shows calculated and test compressive strength and splitting tensile
21 strength at different ages. When we apply the model prediction on specimens tested at 7 days the
22
23 ratio of test/predicted compressive strength ratio will be 1.112 and 1.076 for concrete mix with
24 fine PVC aggregate and coarse PVC aggregate respectively, and equal to 1.012 and 0.989
25 respectively for concrete tested at the age of 56 days. One can well observe that the model
26 prediction is fairly accurate when applied on high strength concrete mix containing PVC
27
28
aggregate. The nature of prediction is quite good when applied on splitting tension, of
29 test/calculated tensile strength equal to 0.974, 0.979 and 1.004 for concrete containing fine PVC
30 aggregate tested at 7,28 and 56 days respectively. For the mix with coarse PVC aggregate, the
31 above values will be 0.966, 0.969 and 0.991 respectively. One can find better prediction with
32 increasing age of concrete, and this may be due to improving plastic aggregate-cement paste
33
34 bond because of silica fume used with increasing age.
35 Based on Al-Hadithi and Alani [6] tests, results of analysis indicate that mean test/calculated
36 compressive strength ratio is relatively low and equal to 0.642, and accordingly the prediction is
37 well overestimates compressive strength loss. In contrast, the prediction is slightly overestimates
38
39
splitting tensile strength since the ratio is equal to 0.956. It should be noted that for specimens
40 tested at the age of 7, 56 and 91 days the test/calculated compressive strength ratio is 0.855, 0.86
41 and 0.879 respectively. As shown in Table 2, the test/calculated splitting tensile strength ratio is
42 0.938, 0.991 and 0.991 for specimens tested at 7,56 and 91 days respectively, indicating
43 increasing the accuracy of the model prediction with increasing age of concrete tested. This
44
45 could be attributed to the action of silica fume used by Al-Hadithi and Alani [6] with increasing
46 age of concrete. The action of SF is also appear in data given by Jafr [20] in which the nature of
47 prediction is better for those mixes containing SF as shown in Table 2-2. It will be noted that for
48 the mixes containing pellet PA the predicted compressive strength is better since mean
49
test/calculated ratio is 0.964, but not good for the mixes with IM and PET aggregate (mean
50
51 test/calculated ratio is 0.85 and 0.895 respectively) mainly because these mixes contain high
52 ratio of flaky particles.
53
54 Lightweight concrete
55
56
Fig. 3 shows calculated and test compressive strength and splitting tensile strength at different
57 ages tested by Kou et al. [7]. Results indicate that mean test/calculated compressive strength ratio
58 is equal to 1.129 for concrete containing PVC aggregate tested at 28 days respectively. One can
59 note that the predictions are different from those discussed in other cases in which that the model
60
61
62 10
63
64
65
 1
2
3
4 predictions overestimates test data, since the predicted strength loss is lower than that of test one.
5
6 In contrast, when one applies the predicted model on specimens tested at 7 days and 56 days the
7 ratio of test/predicted compressive strength ratio will be 0.917 and 0.96 respectively. Results of
8 splitting tensile strength prediction showed that mean test/calculated ratio is equal to 1.053 for
9 concrete samples tested at 28 days. It should be noted that for specimens tested at the age of 7
10
11
and 56 days the ratio will be 0.977 and 1.056 respectively. As an average, test/calculated ratio of
12 compressive strength is equal to 1.003 and that of splitting tensile strength is equal to 1.029, and
13 consequently there is no need to use the bond deficiency coefficient (to be done later) for the
14 case of lightweight concrete containing PVC aggregate based on results of Kou et al. [7]. This
15 may be because of the fact that the weak point in the lightweight concrete mix tested is the
16
17 lightweight coarse aggregate particle causing final failure and not the interfacial bond between
18 plastic aggregate and cement paste. Secondly, the PVC aggregate used by Kou et al. [7] was
19 thick and tends to be pellet particles having different action in concrete as compared with that of
20 plastic flakes.
21
22
23 Self-compacting concrete
24 Results indicate that mean test/calculated compressive strength ratio is equal to 0.962, 0.834,
25 0.834 and 0.908 and test/calculated splitting tensile strength is equal to 0.894, 0.823, 0.899 and
26 0.925 for concrete with PVC, pellet (PEL), irregular mixed (IM) and PET aggregates
27
28
respectively. These ratios are smaller than unity identical to the most types of concrete
29 previously mentioned in this study. So, the effect of bond deficiency is also available for the case
30 of self-compacting concrete based on test data given by Abdulqadir [21].
31
32 5. Bond deficiency parameter (Db) and research need
33
34 Since for concrete mixes, except high strength and lightweight ones, mean test/calculated value
35 is lower than unity, there must be additional parameter to account for more strength loss. This
36 parameter is related to the plastic aggregate particle-cement paste bond deficiency, mainly
37 because of hydrophobicity of plastic particle embedded in cement paste able to create weak
38
39
transition zone. In this study, an attempt was made to assess this parameter, and herein this
40 parameter is called bond deficiency termed as Db. The author thinks that the bond deficiency of
41 the plastic aggregate particle could be improved via surface treatments to be done on the plastic
42 aggregate. Several attempts were made by the past researchers to investigate the effect of plastic
43 surface treatment aimed at improving plastic aggregate-cement paste bond for the purpose of
44
45 controlling strength loss. According to tests by Abu-Saleem et al. [22], interfacial transition zone
46 between plastic surface and cement paste is improved due to PET plastic surface treatment using
47 microwave radiation pre-treatment (MRP) leading to an improvement of compressive strength,
48 splitting tensile strength, flexural strength and modulus of elasticity of concrete. As mixed plastic
49
(PET+HDPE+PP 5% each) was treated and used in concrete no improvement was observed. This
50
51 evidence supports the conclusion drawn by the authors which is the shape and size of plastic
52 particle has an effect on the residual properties of concrete.
53 Therefore, to assess Db utilizing MRP procedure there is a need for tests on concrete containing
54 different PAs, with or without surface treatment, to enhance interfacial plastic aggregate-cement
55
56
paste bond. Measuring residual property of concrete in question containing PA with or without
57 surface treatment could be made and the difference will lead to the assessment of Db parameter
58 experimentally.
59
60
61
62 11
63
64
65
 1
2
3
4 Another way to assess the value of Db parameter accurately is via testing concrete containing PA
5
6 with or without mineral admixture, such as fly ash, silica fume, metakaolin, etc. As stated before,
7 the existence of silica fume has an action to enhance PA-cement paste bond. Therefore, using
8 mineral admixture via enhancing the interfacial bond will enhance the property of concrete in
9 question. The difference between the two measurements (with or without mineral admixture)
10
11
will lead to the assessment of Db parameter.
12 On considering Db, final form of the residual compressive and splitting tensile strength equations
13 will be
14
15 ′
𝑓𝑐𝑝 𝑆𝐴𝑁 ∗ 𝑉𝑁 + 𝑆𝐴𝑃 ∗ 𝑉𝑃
16 = 𝐷𝑏 (19)
𝑓𝑐′ 𝑆𝐴𝑁
17
18
𝑓𝑠𝑝𝑝 𝑆𝐴𝑁 ∗ 𝑉𝑁 + 𝑆𝐴𝑃 ∗ 𝑉𝑃
19 = 𝐷𝑏 (20)
20 𝑓𝑠𝑝 𝑆𝐴𝑁
21
22 In this study, an attempt was made to assess Db parameter approximately for concrete
23 compressive strength and splitting tensile strength utilizing those test data on mixes with and
24
25 without SF. These data are given by Jafr [20] for concrete mixes containing four different PAs
26 given in Table 2 tested in compression, while for the case of splitting tensile strength test data are
27 related to PVC aggregate (tests by Mohammed et al. [2] and Mohammed and Mohammed [8]).
28 As compressive strength is regarded, for mixes with PVC and pellet PA no bond deficiency is
29
required, while for mixes with IM and PET aggregate, Db is taken as 1.09 and 1.02 respectively.
30
31 For mixes with PVC aggregate tested for splitting tensile strength, Db could be taken as 1.03.
32 Indeed, further tests are required to assess better the role of PA-cement paste bond improvement
33 using different treatments which lead to better assessment of Db parameter.
34
35
36 6. Conclusion
37 From this research study the following conclusions are made
38 1- Based on published experimental works, shape and size of plastic aggregate particle has an
39 influence of fresh density and residual compressive and splitting tensile strengths of concrete.
40 Consequently, their effects should be regarded for the prediction of residual properties of such
41
42 type of recycled concrete.
43 2- A model based on calculating aggregate surface area was proposed for the fresh density,
44 compressive and splitting tensile strengths and the author noted that the accuracy of the model
45 prediction will increase with increasing curing age of concrete, and better for concrete containing
46
47
plastic pellet as compared with that containing plastic flakes. The non-accuracy of the prediction
48 is attributed to the plastic particle-cement paste bond deficiency, mainly because of
49 hydrophobicity of plastic embedded in concrete.
50 3- In contrast to the case of conventional concrete, the model prediction is accurate for the case
51 of high strength and lightweight concretes. The authors think that the existence of silica fume in
52
53 high strength concrete will lead to the interfacial bond enhancement leading to accurate
54 prediction. Bond deficiency parameter (Db) was assessed approximately and found to be 1.09
55 and 1.02 for mixes with irregular and PET PA respectively tested in compression and found to be
56 1.03 for mixes containing PVC aggregate subjected to splitting tension.
57
58
59 References
60
61
62 12
63
64
65
 1
2
3
4 [1] K.S. Rebeiz, D.W. Fowler, D.R. Paul, Recycling plastics in polymer concrete for
5
6 construction applications. ASCE J Mater Civ Eng 5(2) (1993) 237–248.
7 [2] A.A. Mohammed, I.I. Mohammed, S.A. Mohammed, Some properties of concrete with
8 plastic aggregate derived from shredded PVC sheets. Constr. Build. Mat. 201 (2019) 232-
9 245.
10
11
[3] I. Almeshal, B.A. Tayeh, R. Alyousef, H. Alabduljabbar, A.M. Mohamed, Eco-friendly
12 concrete containing recycled plastic as partial replacement for sand, J. Mat. Res. Tech. 9 (3)
13 (2020) 9631-9643.
14 [4] E. Rahmani, M. Dehestani, M.H.A. Beygi, H. Allahyari, I.M. Nikibin, On the mechanical
15 properties of concrete containing waste PET particles, Constr. Build. Mat. 47 (2013) 1302-
16
17 1308.
18 [5] A. Jain, S. Siddique, T. Gupta, R.K. Sharma, S. Chaudhary, Utilization of shredded waste
19 plastic bags to improve impact and abrasion resistance of concrete, Environ. Dev. Sust. 22
20 (2018) 337-362.
21 [6] A. I. Al-Hadithi, M.F.A. Alani, Mechanical Properties of High-Performance Concrete
22
23 Containing Waste Plastic as Aggregate, J. Engineering 21 (8) (2015) 100-115.
24 [7] S.C. Kou, G. Lee, C.S. Poon, W.L. Lai, Properties of lightweight aggregate concrete prepared
25 with PVC granules derived from scraped PVC pipes, Waste Manage. 29 (2009) 621–628.
26 [8] T.S. Mohammed, A.A. Mohammed, Physical and Mechanical Properties of High Strength
27
28
Concrete containing PVC Waste as a Sand Replacement, Sulaimani J. Eng. Sci. 7 (3) (2020)
29 132-148.
30 [9] K. Hannawi, S. Kamali-Bernard, W. Prince, Physical and mechanical properties of mortars
31 containing PET and PC waste aggregates, Waste Manage. 30 (2010) 2312–2320.
32 [10] N. Saikia, J. De Brito, Waste polyethylene terephthalate as an aggregate in concrete, Mater.
33
34 Res. 16 (2) (2013) 341–350.
35 [11] C. Albano a, N. Camacho, M. Hernández, A. Matheus, A. Gutiérrez, Influence of content
36 and particle size of waste pet bottles on concrete behavior at different w/c ratios, Waste
37 Manage. 29 (2009) 2707-2716.
38
39
[12] J.R. Correia, J.S. Lima, J. de Brito, Post-fire mechanical performance of concrete made with
40 selected plastic waste aggregates, Cem. Conc. Compos. 53 (2014) 187-199.
41 [13] O. Y. Marzouk, R.M. Dheilly, M. Queneudec, Valorization of post-consumer waste plastic
42 in cementitious concrete composites, Waste Manage. 27 (2007) 310–318.
43 [14] Ghasemi Y, Rajczakowska M, Emborg M and Cwirzen A Shape-dependent calculation of
44
45 specific surface area of aggregates versus X-ray microtomography. Magazine of Concrete
46 Research, https://doi.org/10.1680/jmacr.18.00121
47 [15] R.P. Panda, S. S. Das, P.K. Sahoo, An empirical method for estimating surface area of
48 aggregates in hot mix asphalt, J. Traffic Trans. Eng. (English Edition) 3 (2) (2016) 127-136.
49
[16] J.R. Carr, M. Misra, J. Litchfield, Estimating surface area for aggregate in the size range 1
50
51 mm or larger, Transportation research Record 1362, University of Nevada, Reno Nev.
52 89557.
53 [17] A.M. Neville, Properties of concrete, 5th Ed. Pearson, 2011.
54 [18] A.M. da Silva, J. de Brito, R. Veiga, Incorporation of fine plastic aggregate in rendering
55
56
mortar, Constr. Build. Mat. 71 (2014) 226-236.
57 [19] Y. Ghernouti, B. Rabehi, B. Safi, R. Chaid, Use of recycled plastic bag waste in the
58 concrete, J. Int. Scientific Publ.: Materials, Methods and Technologies 8 (2014) 480-487.
59
60
61
62 13
63
64
65
 1
2
3
4 [20] F.S. Jafr, The effect of shape and size of recycled plastic aggregate on the properties of high
5
6 strength concrete, MSc. Thesis, University of Sulaimani, 2022 (Unpublished).
7 [21] Z. M. Abdulqadir, Impact of plastic waste geometry on the properties of self-compacted
8 concrete, MSc. Thesis, University of Sulaimani, 2022 (Unpublished).
9 [22] M. Abu-Saleem, Y. Zhugea, R. Hassanli, M. Ellis, M. Rahmana, P. Levett, Microwave
10
11
radiation treatment to improve the strength of recycled plastic aggregate concrete, Case
12 Studies in Const. Mat. 15(2021) e00728
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62 14
63
64
65
(a) Cylindrical particle (b) Spherical particle (Dust) (c) Flaky plastic aggregate
Fig. 1 Aggregate particle shape

Fig. 2 Test and calculated strength (Mohammed and Mohammed [8])

Fig. 3 Test and calculated strength (Kou et al. [7])

Table

Table 1 Measured and calculated fresh concrete ratio

Number of particles
Calculated Measured density Test / calculated
Reference (in m3)
density ratio ratio density ratio
Ns Np
0.989 0.989 1.000
0.979 0.982 1.003
0.968 0.972 1.004
Rahmani et al. [4]
1.21*108 1.78*108 0.988 0.993 1.005
0.976 0.981 1.005
0.963 0.968 1.005
Mean 1.004
0.941 0.983 1.045
6.76*108 96955480 0.884 0.957 1.0833
0.830 0.940 1.132
da Silva et al. [18]
0.947 0.989 1.044
31544977 0.895 0.970 1.084
0.846 0.966 1.142
Mean 1.089
Table 2 Results of predicted and test compressive strength and splitting tensile strength
′ 𝑓𝑠𝑝𝑝
𝑓𝑐𝑝
Natural Agg. Plastic Agg.
Type of PA 𝑓𝑐′ 𝑓𝑠𝑝
References Type of PA
concrete (%) NFA x 106 SAN NP x 106 SAP Test / Test /
(per m3) (m2/m3) (per m3) (m2/m3)
Calculated Test Calculated Test
Calculated Calculated
5 0.983 0.964 0.981
10 0.967 0.915 0.947
15 0.95 0.844 0.888
- - -
5 0.983 1.09 1.11
10 0.967 0.971 1.01
15 0.95 0.937 0.986
Rahmani et al. [4] PET 128 6276 178 4190
5 0.983 1.09 1.11 0.983 0.98 0.996
10 0.967 1.03 1.06 0.967 0.927 0.959
15 0.95 0.949 1.0 0.95 0.851 0.896
5 0.983 1.12 1.14 0.983 0.95 0.966
10 0.967 1.01 1.04 0.967 0.862 0.892
15 0.95 0.915 0.96 0.95 0.785 0.826
5 0.982 0.838 0.854 0.982 0.88 0.896
PET
10 310 3350
0.965 0.715 0.741 0.965 0.75 0.778
Normal
15 0.95 0.588 0.621 0.947 0.624 0.659
strength Saikia and de Brito [10] 103 5180
5 0.961 0.878 0.914 0.961 0.921 0.959
PP
10 26.15 1137
0.922 0.856 0.928 0.922 0.972 1.054
15 0.883 0.792 0.897 0.883 0.88 0.996
5 0.964 0.955 0.991 0.964 0.895 0.928
15 0.891 0.912 1.024 0.891 0.986 1.107
30 0.782 0.774 0.99 0.782 0.675 0.864
Mohammed et al. [2] PVC 62.65 5048 64.46 1372
45 0.672 0.58 0.862 0.672 0.621 0.924
65 0.527 0.472 0.896 0.527 0.481 0.914
85 0.381 0.395 1.037 0.381 0.365 0.957
10 0.925 0.893 0.965
20 0.851 0.746 0.877
Ghernouti et al. [19] PP 152 5641 45.67 1426
30 0.776 0.71 0.915 - - -
40 0.701 0.66 0.942
5 0.956 0.962 1.006
PET
10 96.96 1316 0.912 0.516 0.566
- - -
15 0.869 0.500 0.576
Mortar da Silva et al. [18] 676 10545
5 0.956 0.897 0.938
PP
10 31.54 1192 0.911 0.891 0.978
- - -
15 0.869 0.804 0.928
5 0.96 1.06 1.104 0.96 0.996 1.037
10 0.92 0.897 0.975 0.92 0.919 0.999
756 1446
Mohammed and Fine PVC 20 136 7407 0.839 0.856 1.02 0.839 0.839 0.889
Mohammed [8] 40 0.678 0.774 1.142 0.678 0.678 0.993
High 5 0.959 1.026 1.07 0.959 0.991 1.033
 strength Coarse 10 0.918 0.877 0.955 0.918 0.914 0.996
PVC 20 90.31 1341 0.836 0.839 1.003 0.836 0.723 0.865
40 0.672 0.712 1.059 0.672 0.661 0.983
5 0.986 0.675 0.684 0.986 1.036 1.051
Al-Hadithi and Alani
15 0.972 0.636 0.655 0.972 0.884 0.909
[6] PET 152 6935 164 3017
30 0.958 0.563 0.587 0.958 0.868 0.907
5 0.967 0.985 1.019
10 0.933 0.956 1.024
15 0.900 0.886 0.985
20 0.866 0.834 0.963
PVC 212 2410
5 0.967 0.891* 0.922
10 0.933 0.830* 0.889
15 0.900 0.81* 0.900
20 0.866 0.738* 0.852
5 0.957 0.973 1.017
10 0.914 0.895 0.980
15 0.870 0.834 0.958
20 0.827 0.766 0.926
PP 15.55 992
5 0.957 0.924* 0.966
10 0.914 0.896* 0.981
15 0.870 0.807* 0.927
20 0.827 0.788* 0.953
Jafr [20] 127 7272 - - -
5 0.956 0.930 0.973
10 0.912 0.718 0.787
15 0.868 0.695 0.800
20 0.824 0.580 0.704
IM 15.33 858
5 0.956 0.914* 0.956
10 0.912 0.883* 0.968
15 0.868 0.712* 0.821
20 0.824 0.654* 0.793
5 0.960 0.898 0.935
10 0.920 0.847 0.920
15 0.880 0.731 0.830
20 0.841 0.721 0.858
PET 55.62 1475
5 0.960 0.942* 0.981
10 0.920 0.840* 0.913
15 0.880 0.775* 0.880
20 0.841 0.707* 0.841
5 0.953 0.909 0.955 0.953 0.944 0.991
15 0.858 0.895 1.043 0.858 0.856 0.998
Lightweight Kou et al. [7] PVC 830 12350 10 638
30 0.716 0.961 1.343 0.716 0.843 1.178
45 0.573 0.674 1.176 0.573 0.598 1.043
Self- 2.5 0.982 0.930 0.948 0.982 0.756 0.771
Abdulqadir [21] PVC 150 6925 173 1532
compacting 5 0.961 0.929 0.966 0.961 0.911 0.948
 7.5 0.942 0.925 0.982 0.942 0.898 0.954
10 0.922 0.878 0.952 0.922 0.833 0.903
2.5 0.978 0.856 0.875 0.978 0.817 0.836
5 0.956 0.791 0.828 0.956 0.724 0.757
PP 13.18 847
7.5 0.934 0.772 0.826 0.934 0.778 0.833
10 0.912 0.729 0.799 0.912 0.791 0.867
2.5 0.978 0.926 0.947 0.978 0.887 0.907
5 0.957 0.844 0.882 0.957 0.837 0.875
IM 16.17 914
7.5 0.935 0.730 0.781 0.935 0.820 0.877
10 0.913 0.664 0.727 0.913 0.855 0.936
2.5 0.983 0.932 0.949 0.983 0.947 0.964
5 0.965 0.897 0.929 0.965 0.907 0.939
PET 79.23 2135
7.5 0.948 0.832 0.878 0.948 0.872 0.919
10 0.931 0.815 0.875 0.931 0.816 0.877
* Mix containing 10% silica fume as cement replacement
Supplementary Material

Click here to access/download

Supplementary Material
declaration-of-competing-interests.docx