See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/339324682

Behavior of Precast Concrete Columns Bases Embedded in Socket Foundations

with Different Bonding Interfaces

Conference Paper · February 2020

CITATIONS READS

0 266

3 authors:

Hayder Khudhair Haitham Muteb

1 PUBLICATION   0 CITATIONS   
University of Babylon
22 PUBLICATIONS   5 CITATIONS   
SEE PROFILE
SEE PROFILE

Haider Al-Baghdadi
University of Babylon
8 PUBLICATIONS   7 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Ultimate Strength Capacity and the bond -slip Behavior of Composite Ultra-High Performance Concrete-Steel Beams View project

EXPERIMENTAL AND FINITE ELEMENT INVESTIGATION OF STEEL PLATE GIRDERS WITH CORRUGATED WEBS SUBJECTED TO BIAXIAL SHEAR AND BENDING View project

All content following this page was uploaded by Haitham Muteb on 18 February 2020.

The user has requested enhancement of the downloaded file.

 Behavior of Precast Concrete Columns Bases Embedded in Socket
Foundations with Different Bonding Interfaces
a*
Hayder Khudhair Hassan , Haitham H. Muteb b , Haider M. Al-Baghdadi c

a
M.Sc. Student, College of Engineering, University of Babylon, Hilla, Babil, Iraq.([email protected])
b
Professor, Ph.D., College of Engineering, University of Babylon, Hilla, Babil, Iraq.( [email protected])
c
Asst.Prof. College of Engineering, University of Babylon, Hilla, Babil, Iraq.( [email protected])

Abstract

This paper presents an experimental investigation of the behavior and performance of the socket

connection used to stabilize the precast concrete column with the foundation under the lateral

load effect, using a new test method developed based on the principle of symmetry axis. The

experimental program included testing of five specimens representing the state of the precast

concrete column bases embedded in the external socket foundation. The five specimens were

arranged in tow groups based on the embedment length. Group one consists of two specimens

with smooth interface and with different embedded lengths. Group two consists of three

specimens with constant embedment length, two of them with smooth interface and the other

with rough interface. Self-compacting mortar (SCM) was used as a bonding material to fill the

gap between the concrete column and the socket for the first group specimens, whereas the non-

shrinkage grout and self-compacting mortar (SCM) was used a bonding material to fill the gap

between the concrete column and the socket in the second group. The practical results showed

the possibility of using the developed test method as an alternative test method for comparing

and representing a relationship between the different cases of socket connections. It is also found

that the roughness of the interface does not affect the ultimate load capacity of socket walls

when the embedded length is equal to the depth of the concrete column.

Key word: Precast concret column ; Socket foundation ; Self-compacting mortar ; Non-

shrink grout.

1
Introduction

The necessity to improve the construction industry is growing effectively as the demands for

quality buildings are increasing with the rapid infrastructural growth. One of the ways that used to

achieve this improvement is by utilizing a precast concrete construction. The application of the

precast concrete has increased around the world due to its proven advantages like higher quality,

durability, faster construction, and low cost. It is however recognised that the performance of the

precast concrete structures is affected by the performance and properties of the connections

between the precast elements.

Regarding to the buildings that are exposed to lateral loads, such as wind loads and

earthquakes, adequate connections represent particular challenge in these buildings where the

greatest forces are transferred from beam to column and from column to foundation. Therefore, it

is important to develop economic links which are strong enough to resist these loads, and easy to

build with high quality.

Socket connections are one of the techniques used to stabilize the concrete pre-cast concrete

column with the foundation. The procedure for erected precast concrete column in socket can be

summarized as the following: Firstly, a socket is created in the reinforced concrete foundation.

After that, the precast concrete column is inserted into the socket and set in position by using a

centralization device located in the inside bottom of the socket. In this process, wooden wedges are

inserted to facilitate the temporary fixation and to correct any deviations that may happen. Lastly,

the gap between the precast concrete column and the socket walls is filled with non-shrink grout or

self-compacting concrete to obtain a rigid connection between the precast concrete column and the

foundation. While a concrete column cast in place acts on a monolithic way. The monolithic

degree of precast column depends on the efficiency of the connection such as the design strength

of the walls of socket, embedment length of column in socket, the design strength of the grout and

the friction coefficient between column and socket to ensure a good transmission of stress between

the precast concrete column and the socket.

2
 According to the Brazilian Code ABNT NBR 9062:2006 [1], the socket connections can be

classified as follows in three main categories depending on the location of the socket with respect

to the foundation as illustrated in Figure 1. Socket embedded connection, where the socket is

completely inside the foundation depth. The external socket connection, in which the socket is

completely above the foundation depth as reinforced concrete pedestal walls forming the socket is

made above the foundation. Socket partially embedded connection, in which a part of the socket is

located within the foundation depth and the other part above that of the foundation.

Figure 1.Different types of socket connections.

The model of the transfer mechanism of forces from the concrete column to the walls of the

socket was initially suggested by Leonhardt and Mönnig [2] and the following proposed models

are derived from this model. The German institute for standardization DIN-1045 [3] used two

equations to compute the horizontal force Htop depending on the condition of the interface between

the precast column and socket walls(smooth or rough) and neglected the effect of vertical reaction;

see Figure 2. Osanai et al. [4] proposed a model for the mechanism of transmission of stresses in

the embedment region, taking into account the impact of friction and vertical reaction and its

eccentricity. Elliott [5] presented two models, one of these did not include the effect of side loads

and the other includes it. The effect of vertical reaction and friction on the lower face of the precast

concrete column was neglected in the first model, while all horizontal and vertical reactions and all

friction forces between the column and the socket were considered in the second model. El Debs

[6] suggested a model taking into account the effect of friction between the column and socket as

well as taking into accounts the vertical reaction and its eccentricity. Canha et al. [7] proposed

3
model proposes design models and recommendations for column-foundation connection through

socket with rough interfaces, including the shear key configuration, the socket and the precast

column base.

Figure 2. Forces mechanism transmission according to DIN-1045.

The main objective of current study is to investigate the effect of the embedment length of

precast concrete column, type of bounding material (grouting) between precast concrete column

and socket and type of interface between precast column and socket walls on the behavior of the

external socket base connection subjected to lateral load by using a test method was developed

based mainly on the principle of the symmetry axis, which lies in the middle distance between the

two supports. The developed test method is useful for comparing and representing a relationship

between the different cases of socket connections.

2. Behavior of transverse and longitudinal walls of socket foundation

Figure 3 shows the stresses transmission between the precast concrete column and the transverse

walls of socket for a smooth interface surface, as observed by El Bebs[6]. The moment (Md) and

the horizontal force (Vd) produced by external loads on the column are transported through the

concrete cast between them as a stress to the transverse walls of the socket (Htopf) resultant of

stress on front transverse wall and (Hbot) resultant of stress on rear transverse wall). These

resultants mobilize frictional forces on the interfaces between the column and the transverse walls.

The direction of the friction force on the front wall (Fftopf) is always downward, while, the

direction of the friction force acting on the rear transverse wall (F fbot) depends on the relative

4
eccentricity and the socket geometry; where it could be downward for small relative eccentricity

and vice versa in the case of large eccentricity. The value of the resultant stress (Fnb) between the

lower face of column and the base of the socket is equal to the vertical axial force minus the

algebraic total of the friction forces generated on the transverse walls. Also, the friction force

between the lower face of column and foundation is mobilized by the resultant force (Fnb).

The resultant pressure on the top of the front transverse wall (Htopf) is transmitted evenly to the

longitudinal walls. The longitudinal walls transmit the forces to the foundation in the manner of a

tensile force at the corners of the longitudinal walls (Fvd) (resulting from the stress in the vertical

reinforcing steel) and the compression force of the concrete strut (Rc). The resultant bottom

pressure on the rear transverse wall (Hbot) is transferred immediately to the foundation. Thus, the

longitudinal walls of the socket behaved like corbels fixed at the bottom with the foundation [1, 2].

Figure 4 shows the behavior of longitudinal socket walls.

Figure 3. Behavior of transverse walls [6].

Figure 4. Behavior of longitudinal walls.

5
3. Experimental Program

The practical program consists of two groups of specimens. Each specimen of these groups

consists of three pieces (two columns and socket) fix together by grout or self-compact mortar to

test as a simply supported beam (benefiting from the axis of symmetry to represent two symmetric

column embedded in socket foundation subjected to lateral load )

3.1. Materials Properties

3.1.1 Concrete

Normal strength concrete was used for casting of all specimens in the present research. The

type of cement used in this study was ordinary Portland cement. The mix of proportions was

(1:1.66:2.12) (by weight) cement, sand and gravel, respectively and (0.44) water cement ratio. The

compressive strength of concrete after 28 days was 32 MPa and the splitting tensile strength was

3.09 MPa.

3.1.2 Reinforcement

In this study, two sizes of reinforcing steel bars, deformed steel bars of diameter Ø10 mm for

main longitudinal reinforcement and deformed steel bars of diameter Ø6 mm for closed stirrups

were used. The yield strength was 463 MPa for Ø10 mm diameter steel bars and 420 MPa for Ø6

mm diameter steel bar.

3.1.3 Flo-grout 2

Flo- Grout 2 manufactured by (DCP) Company, England was used as a bonding material in one

specimen. It contains cement, selected additives, well graded, and non-reactive aggregates and is

designed to give excellent flow properties, shrinkage compensation, frost resistance, and high

compressive strength. Flo-Grout 2 complies with ASTM C1107, Grade A [8]. The compressive

strength test was carrying out according to (ASTM C109/109M-11) [9], see table 1

6
3.1.4 Self-compacting mortar (SCM)

SCM was used as a bonding material in four specimens. The same reference mix proportions

designed by Tuaum, et al. [10] were adopted and then the water and superplasticizer ratio was

modified to obtain the same fresh properties, see Table 2. Test of Fresh SCM was carrying out

according to EFNARC [11], see figure 5. The slump flow test using mini slump con was

evaluated for the deformability of the SCM mixture and must be kept within 250 ± 10 mm. The

viscosity and flow ability of SCM mixtures were investigated by conducting mini V-funnel flow

time test and the permissible flow time value is 9±2 second. In this study, the slump flow of SCM

was 25.25mm and the flow time was 9.4 second. The compressive strength test of SCM was

carrying out according to (ASTM C109/109M-11) [9], see table 1.

(a) Mini slump flow test (b) Mini V-funnel flow time test.

Figure 5. Test of fresh SCM.

Table 1. Compressive strength of Flo-grout 2 and SCM.

Compressive strength
Type of
(MPa)
material
7Days 14 Days 28 Days

Flo-Grout 2 47.16 54.4 59.6

SCM 49.31 57.44 61.73

7
 Table 2. Mix proportions of self-compacting mortar.

Limestone Fine SP/Cement

Water Cement SP
W/P W/C (kg/m3) aggregate
(kg/m3) (kg/m3) (kg/m3)
(kg/m3

0.27 0.33 208 623 138 1186 0.0103 6.4

3.1.5 Superplasticizer (SP)

Sika ViscoCrete -5930L, manufactured by Sika Company, Swiss, was a superplasticizer used to

produce self-compacting mortar. It is a superplasticizer of the third generation for concrete and

mortar

3.2. Specimen Details

All specimens have the same span length 1620 mm, thickness of socket wall 70 mm, inner

socket dimension 200×200 mm, gap thickness for grouting was 25 mm, length of socket 620 mm

and column cross section 150×150 mm. The main variables considered in the experimental

program were the embedment length of column inserted into socket, the type of interface surface

(smooth or rough) and the type of bonding material at interface surfaces. In order to obtain the

property of roughness, shear keys are worked in both the concrete column and the internal surface

of the socket walls at the embedment region. The shear keys in the embedment region have been

designed according to the requirements of the Brazilian Code NBR 9062:2006 [1]. Figure 6 shows

the details of shear key. The five specimens were arranged in two groups based on the type of

bonding interface surface between the column and the embedment length. Table 3 describes the

details of each specimen.

The method of DIN 1045 [3] on how to distribute the stresses in the embedment region for

smooth interface has been adopted to compute the vertical reinforcement (Asv). All columns were

designed with a flexural capacity greater than that of the socket walls. The reinforcement of

column and socket is kept the same in all specimens. The longitudinal column reinforcement
8
consisted four deformed bars of 10 mm diameter are extended along the whole of column, with

clear cover of 10 mm. The closed stirrups of Ø6 mm reinforcing bar were placed at distance 65

mm along the column. While, 8 Ø10 mm deformed steel reinforcement bar were distributed along

the parameter of the center cross section of the socket walls along the socket and the close stirrups

(Asv) of Ø6 mm at 80 mm. Figure 7 shows details of the geometry and reinforcement of the

specimens.

Figure 6. Details of shear key.

Figure 7. Details of the geometry and reinforcement of the specimens.

9
 Table 3. Details of tested specimens.

Embedment Lx/Column Type of

Group Specimen Type of
length (Lx) depth bonding
No. designation interface
mm mm/mm material

Group L270SCST70 270 1.8 SCM Smooth

one L225SCST70 225 1.5 SCM Smooth

L150SCRT70 150 1.0 SCM rough

Group
L150SCST70 150 1.0 SCM Smooth
two
L150FGST70 150 1.0 Flo-grout 2 Smooth

3.3. Test mechanism

A new test technique is proposed based on the fact that the main contributor to the value of the

resultant horizontal pressure on the socket walls is the bending moment and shear force

transmitted from the column to the socket walls. Thereafter, the resultant pressure is transmitted to

the foundation through the socket walls, while most of the vertical load transferred directly from

the column to the foundation.

In the most theoretical studies, the effect of vertical force has been cancelled to facilitate the

calculation process and add a high safety factor, considering that the mechanism of transmission of

internal stresses in the embedment portion is very complex. Also, the effect of punching on the

foundation was not considered in this study. As mentioned earlier, the objective of this study is not

to calculate the resultant of horizontal pressure generated on the socket walls, but to representing a

relationship between the different cases of bonding between precast concrete column and socket

Therefore, vertical force negligence in testing all samples achieves the objective of this study.

It is well known that the simply supported beam is considered as a geometrically symmetrical

around the vertical axis that passes halfway between the supports when the loads and the structures

are also symmetrical around this axis, therefore, half of this beam can be considered as a cantilever

beam. Adopting this principle, it is possible to represent two symmetrical columns, each of which
10
is embedded in the socket as shown in Figure 8.

Figure 8. Test mechanism.

3.4. Test Setup

All specimens were tested under a concentrated line loading at mid – span by using a hydraulic

testing machine with 600 kN capacity. The supporting system was simply supported (roller and

hinge) with a distance equal to 1620 mm between them. For each specimen, three vertical dial

gauges of accuracy 0.01mm were used to measure the vertical deflection at the middle span and at

the end of columns. Two horizontal dial gauges of accuracy 0.01mm were used in each specimen

in order to measure the horizontal slip between the column and socket. At first, specimen was

loaded by 4 kN to seat the support and the loading system, then unloading to zero. After that, the

load was applied in stages by a load control mode at a load rate of 0.1 kN/sec. The crack formation

and propagation were examined at each load step, as well as recorded the first and ultimate

cracking load. Figure 9 shows the specimen test layout.

Figure 9. specimen test layout.

11
4. Results and Discussion

Table 4 summarizes the load capacity and failure modes of the specimens tested. Specimen

observation during the tests and detailed discussions were described below.

Table 4. Test results of the tested specimens

Failure Max. Horizontal Failure

Group Specimen Cracking
load Deflection slip position
No. designation load* (kN)
(kN) (mm) (mm)

Group L270SCST70 10 61.77 10.77 0.0 In column

one L225SCST70 10 63 11.91 0.0 In column

L150SCRT70 11 50 5.7 0.0 In socket

Group
L150SCST70 10 50.4 6.1 0.0 In socket
two
L150FGST70 10 48 5.82 0.0 In socket

* represents the corresponding load of the first crack in the concrete column.

4.1. Ultimate Load and Modes of Failure

As can be seen in Table 6, there are two types of failure in the testing of specimens. The first is

the failure of the concrete column of all test specimens in group one. These specimens failed by

yielding the longitudinal column reinforcements in the tension zone near the contact zone with

socket followed by crushing concrete in the compression zone at the ultimate load (61.77-63

kN).The second is that failure of socket walls at ultimate load for all specimens in group two.

These specimens failed at ultimate load (48-50.4 kN) due to the arrival of the vertical reinforcing

steel of the longitudinal socket walls (Asv) distributed over the first third of the embedded length

(𝑳𝒙 /𝟑) to ultimate stress. Figure 10 shows the failure modes.

It is observed that there is no effect of roughness between the column and socket walls on the

ultimate load when the embedment length is equal to the concrete column depth. The reason for

this is that the roughness is affected by several factors, where it decreases when the embedded

length of column decreases and the roughness effect decreases when the number of shear keys in

12
the embedment region decreases [7], while, the test results appeared that the ultimate load-carrying

capacity increased slightly when using SCM as a bonding material compared to Flo-grout 2.

(a) Failure of socket (b) Failure of column

Figure 10. Failure modes of tested specimens.

It is observed that there is no effect of roughness between the column and socket walls on the

ultimate load when the embedment length is equal to the concrete column depth. The reason for

this is that the roughness is affected by several factors, where it decreases when the embedded

length of column decreases and the roughness effect decreases when the number of shear keys in

the embedment region decreases [7]. While, the test results appeared that the ultimate load-carrying

capacity increased slightly when using SCM as a bonding material compared to Flo-grout 2.

4.2. Cracking Behaviour

Cracks begin to appear in specimens when the applied load causes stresses exceed the tensile

strength of concrete in the tension region of the cross-section. The location of the first crack of all

specimens was at the bottom of the concrete column near the contact zone with the socket, where

the column had the largest positive moment. The first cracking load in all specimens was very

close because the stiffness and length of the column from the support to the socket were equal in

all specimens.

Figure 11 illustrates the crack pattern at failure of the specimens tested. All test specimens

group one showed no cracks in the sockets. It is noted that the cracks spread along the columns

13
only where the position of failure. While, the specimens of group two showed cracks in the socket

walls. The cracks begin to appear on the socket at load (48-50 kN) in the corners of socket at an

angle of about (45°) due to the concentration of stresses in that region. Then the cracks are spread

along the socket walls, particularly longitudinal ones, and increase their width and length without

any rising in the applied load until failure due to the arrival of the vertical reinforcing steel of the

longitudinal socket walls (Asv) to ultimate stress. The compression strut can be deduced from the

crack form of longitudinal walls. Based on the failure and crack patterns, it can therefore be

concluded that the behavior of longitudinal walls is similar to the behavior of the corbels [1,2].

The occurrence of failure in the socket walls for the specimens with embedment length of 150

mm and its non-occurrence in the socket walls for the specimens with embedment length of 225

mm and 270 mm, expresses the phenomenon that the value of the reaction (Htop) is inversely

proportional to the value of the embedment length, whereas the area of effect of this force is

directly proportional to the value of the embedded length [1, 3, 4], see figure 2.

Figure 11. Cracks patterns at failure for tested specimens.

14
4.3. Horizontal Slip between Column and Socket

By generating a horizontal shear stress at the interface zone, slip generally occurs in composite

components after the loading increase. Thus, if the shear strength between the composite members

in the interface is less than the shear force generated, slippage occurs. In this study, the lab results

showed that the horizontal slip value of all specimens tested was zero.

4.4. Load – Deflection Behavior

The deflections were measured at three points in each test specimen to evaluate the deflection

in the middle of the socket and at the end of each concrete column. In all specimens of group on, it

was observed that the amount of deflection of the center of the socket is very close to the

deflection of the concrete column, i.e. the behavior of the socket in these specimens is like a rigid

body. While, the specimens with length embedment length of 150 mm the behavior of socket as a

rigid body from the beginning of the load until the first crack occurs in the socket. Figure 12 shows

the load deflection curves for specimens L270SCST70 and L150SCRT70.

L270SCST70 L150SCRT70
70 70

60 60

50 50
Load kN

40 40
Load kN

30 30

20 20
column deflection column deflection
10 10
middel socket deflection middel socket deflection
0 0
0 5 10 15 0 2 4 6 8
Deflection mm Deflection mm

Figure 12. Load-deflection curves at the middle of the socket and at the end of the column.

As can be seen from Figure 13, which shows the load-deflection curves of all specimens, the

roughness slightly increases the initial stiffness when the embedment length equal to the depth of

column. It can also be noted that there is no effect of the embedment length on the initial stiffness

15
of the specimens of the first group due to the failure of the column and the behavior of the socket

as a rigid body.

70

60

50

40
Load kN

30

L270SCST70

20 L225SCST70

L150SCST70

10 L150FGST70

L150SCRT70

0
0 2 4 6 8 10 12
deflection mm

Figure 13. Load-deflection curves of all specimens

4.5. Ductility ratio

Ductility is described as its ability to withstand inelastic deformation without decreasing its

load-bearing capacity to failure. In other words, the ductility factor can be defined as the maximum

deformation divided by the corresponding deformation when yielding occurs. Table 5 shows the

ductility ratios for all specimens tested.

The results in the above table show that the average ductility ratio of the specimens dominated

by the socket failure is lower than that of the specimens dominated by column failure by about

(42%). The reason is that the amount of reinforcement used (Asv) is very low. For this reason, the

16
failure of the walls of the socket is brittle, which is the nature of shear fracture in reinforced

concrete.

Table 5. Ductility ratio for the tested specimens

Deflection (mm) Ductility ratio

Group No. Specimen designation
Δy Δu (Δu/Δy)

L270SCST70 5.3 10.77 2.03

Group one
L225SCST70 5.5 11.91 2.16

L150SCRT70 4.7 5.7 1.21

Group two L150SCST70 5.3 6.1 1.15

L150FGST70 4.5 5.82 1.29

4. CONCLUSIONS

The important conclusions of the experimental study for the specimens tested can be set out as

follows:

 The specimens tested in the alternative testing method showed a similar behavior to the

testing method adopted by the other literature [1, 2, 3], independent of the values of the

horizontal reaction at the transverse socket walls.

 The roughness of the interface surface does not affect the ultimate strength of the socket

walls when the embedded length is equal to the depth of the concrete column.

 The use of (SCM) to fill the gap between the concrete column and the socket foundation

caused a slight increase in the ultimate load compared to the use of Flo-grout 2 when the

embedded length is equal to the concrete column depth.

 Both Flo-grout 2 and self-compacting mortar were given the same efficiency to resist the

horizontal slip between the concrete column and the socket. Where the results showed

that there is no horizontal slippage for all tested specimens.

17
View publication stats

 The use of self-compacting mortar to fill the gap between the concrete column and the

socket foundation is considered more economical than using non-shrinkage grout.

 The occurrence of failure in the concrete column and its disappearance in the socket walls

in the specimens that have an embedded length of (1.8hc and 1.5hc) gives the impression

that the German Code method DIN 1045[3] for computation the reactions on the concrete

column in the embedment region is conservative even with no external axial force

affecting the concrete column.

6. Acknowledgments

The completing of the current work was in the Department of Civil Engineering at Engineering

College of Babylon University. Therefore, the moral support that was provided is gratefully

acknowledged.

References:

[1] Brazilian Association of Technical Standards. NBR 9062 - Design and fabrication of precast concrete structures.
Rio de Janeiro, ABNT.2006.

[2] F. Leonhardt and E. Mönnig, "Vorlesungen über Massivbau,"Springer-Verlag, Berlin, 1973. (in English:
Lectures on solid construction)

[3] "German Institute for Standardisation (Deutsches Institut für Normung)", DIN 1045, 1981.

[4] Y. Osanai, F. Watanabe, and S. Okamoto, "Stress transfer mechanism of socket base connections with precast
concrete columns," Structural Journal, Vol. 93. No. 3, 1996, pp.266-276.

[5] K. S. Elliott, "Multi-story precast concrete framed structures," Blackwell Science, Oxford, 1996.

[6] M. K. El Debs, "Precast concrete: foundations and applications. Published EESC-USP," São Carlos, 2000.

[7] R. M. F. Canha, G. M. Campos, and M. K. El Debs, "Design model and recommendations of column-
foundation connection through socket with rough interfaces," Revista IBRACON de Estruturas e Materiais, Vol.
5, No. 2, 2012, pp. 182-218.

[8] ASTM C1107-02, "Standard Specification for Packaged Dry, Hydraulic-Cement Grout (Nonshrink)," ASTM
International: West Conshohocken, PA, USA, 2002.

[9] ASTM C 109/C 109M-11, "Standard Test Method for Compressive Strength of Hydraulic Cement Mortars
(Using 2-in. or [50-mm] Cube Specimens)," ASTM International: West Conshohocken, PA, USA, 2011.

[10] A. Tuaum, S. Shitote, and W. Oyawa, "Experimental Study of Self-Compacting Mortar Incorporating Recycled
Glass Aggregate", Buildings, Vol. 8, No. 15, 2018, pp.1-18.

[11] European Federation for Specialist Construction Chemicals and Concrete Systems (EFNARC)," Specification
and Guidelines for Self-Compacting Concrete", European Federation for Specialist Construction Chemicals and
Concrete Systems, Norfolk, UK, 2002,

18