Construction and Building Materials 154 (2017) 176–190

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Preparation and tests for workability, compressive and bond strength of

ultra-fine slag based geopolymer as concrete repairing agent
Sulaem Musaddiq Laskar, Sudip Talukdar ⇑
Department of Civil Engineering, Indian Institute of Technology Guwahati, Guwahati 781039, India

h i g h l i g h t s

Prepared ultra-fine blast furnace slag based geopolymer for use as concrete repairing agent.

Workability, compressive and bond strength tests are conducted.

1 day strength of geopolymer concrete is about 60% of 28 days strength.

Addition of flyash and superplasticizer at low dosage level shows satisfactory results.

Bond of geopolymer concrete with rebar and old concrete shows promising results.

a r t i c l e i n f o a b s t r a c t

Article history: In modern era, there is a need to develop sustainable concrete repairing material. This study aimed to
Received 23 March 2017 develop ultra-fine blast furnace slag based geopolymer as concrete repairing agent. Sodium hydroxide
Received in revised form 15 June 2017 activated geopolymer concrete mixes with various admixtures were prepared and tested to study fresh
Accepted 25 July 2017
and hardened state properties including bond strength. The 1 day strength of ultra-fine blast furnace slag
Available online 3 August 2017
based geopolymer concrete was found to be about 60% of that at 28 days. Addition of flyash up to certain
quantity level contributed towards achieving satisfactory workability, compressive and bond strength of
Keywords:
the geopolymer concrete. Similar improvement in fresh and hardened state properties was also observed
Ultra-fine ground granulated blast furnace
slag
due to addition of superplasticizer at low dosage level. High alkali activator concentration led to loss of
Flyash workability and strength of the geopolymer concrete.
Superplasticizer Ó 2017 Elsevier Ltd. All rights reserved.
Workability
Compressive strength
Bond strength

1. Introduction metakaolin (MK) based concrete repairing geopolymer pastes.

The geopolymer pastes possessed high early strength. The mechan-
Blast furnace slag (BFS) based geopolymer concrete (GPC) has ical properties were found to be better than PC based pastes. Addi-
the potential of replacing the conventional Portland cement con- tion of steel slag to the geopolymer pastes improved the
crete (PCC) in the construction industry. Its use can reduce the performance. Phoo-ngernkham et al. [8] attempted to develop con-
CO2 emission that otherwise results due to manufacturing of Port- crete repairing mortar using high calcium FA. Satisfactory bond
land cement (PC), the prime component in any construction pro- strength was observed in the GPM compared to the commercial
ject. The speciality of BFS based GPC is that unlike flyash (FA) repair binders. On adding PC to the GPM, the bonding behaviour
based GPC, it can gain strength even when cured at ambient tem- tended to improve. Vasconcelos et al. [9] used MK based GPM for
perature [1,2]. Of late, numerous researchers have dedicated their repairing concrete beams and found it to be effective as repairing
attention towards the study of fresh and hardened state properties layer than as adhesive material for applying carbon fibre reinforced
of BFS based GPC [3–6]. polymer strips on the beams. Investigation on the mechanical
Many researchers in the past have tried to use geopolymer mor- properties of ambient temperature cured blended and unblended
tar (GPM) and GPC as concrete repairing agent. Hu et al. [7] inves- FA and slag based GPC was carried out by Manjunatha et al. [10].
tigated the compressive, bond strength and abrasion resistance of GPC consisting of only slag exhibited superior performance among
others including conventional PCC. Increase in slag content in the
⇑ Corresponding author. mixes progressively increased the strength and improved bonding.
E-mail address: staluk@iitg.ernet.in (S. Talukdar).

http://dx.doi.org/10.1016/j.conbuildmat.2017.07.187
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
 S.M. Laskar, S. Talukdar / Construction and Building Materials 154 (2017) 176–190 177

Zanotti et al. [11] prepared and tested MK based geopolymer for surface area of UGGBS and FA were evaluated by Laser Particle Size Analyzer and
specific gravity by laboratory experiment as per IS 1727 1967 [19]. The oxide com-
using as repair mortar. Significant strength was achieved by curing
position was determined by X-ray Fluorescence Spectrometer (XRF).
the specimens at elevated temperature. Addition of polyvinyl alco- Sodium hydroxide (SH) pellets having 97–98% purity were used to prepare the
hol fibers to the geopolymer improved its cohesion with the sub- alkali activator solution. SH solution of required concentration was prepared by
strate. MK based geopolymer was also employed for preparing mixing SH pellets with distilled water 24 h before the casting of specimens. Alluvial
PCC pavement repair mortar. The 3 days strength of MK based sand conforming to zone III of IS 383 1970 [20] having specific gravity of 2.69 and
water absorption of 1.7% was used as fine aggregate. Well-graded crushed coarse
GPM was found to be as high as 80% of its 28 days strength. It also
aggregate of maximum 20 mm size having aggregate crushing value of 17.63, speci-
outperformed the other commercial materials available for con- fic gravity of 2.61 and water absorption of 0.7% was used. Both sulfonated naph-
crete repairing [12]. Duan et al. [13] prepared a novel concrete thalene (SN) based SP namely Conplast SP430SRV having specific gravity of 1.260
repairing agent using MK based geopolymer. It possessed proper- and polycarboxylate ether (PE) based SP namely Structuro 201 having specific grav-
ity of 1.090 were used. SPs were supplied by Fosroc Chemicals (India). Rebar of size
ties such as water resistance, fast setting, hydrophobic surface,
16 mm diameter having yield stress, fy of 547.40 MPa was used for preparing spec-
high compressive and bond strength. Sarker [14] found that com- imens for performing pull-out test. OPC 43 grade [21] was used for preparation of
pared to PCC, GPC shows better bonding capacity with rebars. the PCC substrate for preparing specimens to perform slant shear test.
The cracking patterns of GPC specimens under the bond test were
similar to that of PCC specimens. 2.2. Mix proportions
Increase in fineness of binding material leads to early strength
gain in GPM and GPC. Early age strength is a desirable property Mix proportion of the UGGBS based GPC is presented in Table 2. Constant alkali
activator solution/total binding agent (a/b) ratio of 0.66 was maintained in GPC
of concrete repairing agent [15]. Collins and Sanjayan [3] prepared
throughout the study. The mixes were prepared to study the effect of addition of
slag based geopolymer concrete specimens by replacing 10% of slag FA of amounts of 20, 30, 40 and 50% by weight of total binding agent; addition of
by ultra-fine FA (UFA), ultra-fine slag (UFS) and condensed silica SP of amounts of 0.5, 1.5 and 3% by weight of total binding agent; variation of SH
fume (CSF). From test results, it was observed that remarkable concentrations of 8, 10, 12 and 14 M (M); and variation in the time of addition of
improvement in workability occurred due to addition of UFA. Nota- SP on workability, compressive strength and bond strength of GPC. SH as the alkali
activator, the lower and upper limits of FA and SP content; and SH concentrations
ble improvement in strength occurred even at early ages due to were set based on the study of past works by various authors in the same field of
addition of UFS and CSF. Enhanced ultimate strength of BFS based research mentioned elsewhere [22]. SP of both SN and PE types were added to
GPM was observed by Oner et al. [16] due to increase in fineness of the GPC in following three types:-
binder. Finer binder contributes to accelerated strength gain and
Type I: SP added after the addition of SH to the mix.
improved durability of the geopolymeric systems [4,17,18]. How-
Type II: SP was mixed with SH and then added to the mix.
ever, increase in fineness of the binder leads to accelerated setting Type III: SP added before the addition of SH to the mix.
and low workability of mortar or concrete compared to that with
The difference in all the three types lies in the time of SP addition. In Type I, the
same binder of lower fineness [3]. In case of ultra-fine ground gran-
SP was added to the wet mix which consisted of UGGBS, FA, fine and coarse aggre-
ulated blast furnace slag (UGGBS) based geopolymer mixes, high gates; and SH solution. After the addition of SP to the wet mix, it was blended thor-
fineness and angular shape of slag particles contribute towards loss oughly so that the SP gets homogenously distributed. In Type II, SP was added to the
in setting time and workability of the mixes. SH solution and stirred properly. Later, the SP contained SH solution was added to
the dry mix. In Type III, SP was added to the dry mix prior to the addition of SH solu-
In this study, the authors attempted to develop UGGBS based
tion. On addition of the SP, it was thoroughly mixed for homogenous distribution.
GPC possessing such fresh and hardened state properties that make SH solution was added to the mix at the end of this step of SP addition.
the GPC advantageous to be used as concrete repairing agent. PCC substrates for preparing slant shear test specimens were cast using cement
Admixtures such as FA and superplasticizer (SP) were added to of amount of 380 kg/m3, fine aggregate of 662 kg/m3, coarse aggregate of 1146 kg/
3
the GPC to alter the properties and arrive at satisfactory workabil- m and water/cement ratio of 0.5. The maximum size of coarse aggregate was
restricted to 16 mm. The compressive strength of PCC substrate at 28 days was
ity, compressive and bond strength. Effect of variation of alkali
31.78 MPa.
activator concentration and time of addition of SP in GPC have also
been noted. Slump, compressive strength, rebar pull-out and slant
2.3. Specimen preparation and curing
shear tests were performed on total 21 numbers of mixes in this
study. For preparing UGGBS based GPC, initially dry mix was manually composed by
thoroughly mixing UGGBS, FA, fine and coarse aggregates for 2 min. To this dry
mix, SH was added and the mixing processes was further continued for 3 min. SP
2. Experimental program
was added to the mix by either of Type I, Type II or Type III method. On completion,
the fresh GPC was placed into the respective moulds for compressive strength, pull-
2.1. Materials
out and slant shear tests as per the codal provisions [23–25]. The moulds were kept
at ambient temperature of 20 ± 2 °C. After 24 h of casting, the specimens were
Table 1 presents the physical and chemical properties of UGGBS and FA. UGGBS
demolded and submerged inside water tank, maintaining temperature of
was the primary binding agent in the GPC. Class F FA obtained from thermal power
20 ± 2 °C and stored till the arrival of test day.
plant at Farakka, India was used as an additive. Hereafter, UGGBS and FA are
The size of cube specimens for compressive strength and pull-out test was
together referred as total binding agent. The median particle size and specific
150 mm. Fig. 1 shows the special arrangements that were made for casting of the
specimens for pull-out test. Provisions mentioned in IS 2770 1967 [24] were fol-
lowed for the preparation of the specimens. The size of the rebar in each specimen
was of 16 mm diameter. The embedded length of rebar inside the cube was 50 mm.
Table 1 The remaining part of the rebar inside the cube was covered using plastic tube to
Chemical composition and physical properties of UGGBS and FA. break the contact between the GPC and debonded length of the rebar.
Chemical composition (% mass) UGGBS FA For the slant shear test, the GPC constituted half of the specimen volume. The
other half, i.e. the substrate was prepared using PCC (see Fig. 2). The PCC substrates
Silicon dioxide (SiO2) 33.6 55.47 were prepared with the help of dummy specimens which were cast, cured for
Aluminium oxide (Al2O3) 22.5 25.37 28 days and hand finished to give the perfect shape and slant angle of 30° along
Ferric oxide (Fe2O3) 1.3 6.2 the surface ‘ab’ as shown in Fig. 2. The volume of each dummy specimen was half
Calcium oxide (CaO) 34.0 6.24 of that of cylindrical mould of size 75 mm diameter and 150 mm height [25]. Prior
Magnesium oxide (MgO) 6.8 1.55 to the casting of the PCC substrate for slant shear test, the dummy specimens were
Sulphur oxide (SO3) 0.15 0.9 placed into the cylindrical moulds with the surface ‘ab’ covered with polyvinyl
Physical properties sheet which acted as debonding media between the PCC substrates and dummy
Specific gravity 2.84 2.42 specimens. The fresh concrete for PCC substrate was prepared and placed into the
Median particle sized d50(lm) 3.579 26.33 moulds such that it occupied the other half of volume of the mould and kept for
Specific surface area (cm2/g) 30100 8940 24 h at temperature of 20 ± 2 °C. The size of the moulds were taken as 75 mm
diameter and 150 mm height to economize the cost of the experimental study.
178 S.M. Laskar, S. Talukdar / Construction and Building Materials 154 (2017) 176–190

Table 2
Mix proportion for GPC (kg/m3).

Mix UGGBS FA (% of total SH concentration SP (% of total SP type w/s SP addition type Mix group
binding agent) (Molarity, M) binding agent)
S1 358 0 (0) 10 – – 0.40 Type I G1
S2 286.4 71.6 (20) – –
S3 250.6 107.4 (30) – –
S4 214.8 143.2 (40) – –
S5 179 179 (50) – –
S6 250.6 107.4 10 0.5 SN 0.40 Type I G2 (including S3)
S7 250.6 107.4 0.5 PE
S8 250.6 107.4 1.5 SN
S9 250.6 107.4 1.5 PE
S10 250.6 107.4 3 SN
S11 250.6 107.4 3 PE
S12 250.6 107.4 8 1.5 SN 0.43 Type I G3 (including S8 and S9)
S13 250.6 107.4 8 1.5 PE 0.43
S14 250.6 107.4 12 1.5 SN 0.37
S15 250.6 107.4 12 1.5 PE 0.37
S16 250.6 107.4 14 1.5 SN 0.34
S17 250.6 107.4 14 1.5 PE 0.34
S18 250.6 107.4 10 1.5 SN 0.40 Type II G4 (including S8 and S9)
S19 250.6 107.4 1.5 PE Type II
S20 250.6 107.4 1.5 SN Type III
S21 250.6 107.4 1.5 PE Type III

Note: Fine aggregate = 611 kg/m3, coarse aggregate = 1208 kg/m3, SH = 236.28 kg/m3, w/s = water to total solid ratio.

Fig. 1. Details of specimens for pull-out test (All dimensions are in mm).

Moreover, this study deals with the comparative behaviour of various mixes. Hence, Bond strength of GPC mixes was evaluated on the basis of bond strength with
the size effect on strength due to use of small moulds for slant shear test has not rebar (BSr) and with 2 months old PCC (BSc). The pull-out test and slant shear tests
been taken into account. The specimens were later demoulded and cured till were conducted as per IS 2770 1967 [24] and ASTM C882/882M-13a [25] respec-
28 days from date of casting by submerging inside water tank, maintaining water tively on the GPC mixes at 3 and 28 days. The BSr and BSc were evaluated using
temperature at 20 ± 2 °C. After 28 days, the PCC substrates were placed in the open the following equations:-
air till 2 months from the date of casting. At the end of 2 months, the substrates’
surfaces ‘ab’ were carefully cleaned to peel off the dust layer, placed into the moulds Fr
and then the new GPC was poured for each mix. The specimens were again kept into BSr ¼ ð1Þ
Ar
the mould for 24 h at 20 ± 2 °C. After 24 h, these were demoulded and cured by sub-
merging inside water tank maintaining temperature of 20 ± 2 °C till the arrival of
test day. Fc
BSc ¼ ð2Þ
Ac
2.4. Experiments
Here, Fr and Fc are the loads carried by the specimen at failure in pull-out and slant
The workability of GPC was determined with the help of slump test as per IS shear tests respectively. Ar and Ac are the areas of surface of the embedded length of
1199 1959 [26]. The test was performed at 5, 15 and 30 min from the time of mixing the bar and of PCC substrate shown by ‘ab’ in Fig. 2.
of dry GPC mix and SH. Compressive strength test was conducted on the GPC spec- In each type of test, 3 numbers of specimen per mix for each selected day were
imens at 1, 3, 7, 28, 56 days as per IS 516 1959 [23] to evaluate the strength of GPC prepared and tested. The results presented are the average of the test result values
mixes at early and later ages. of each specimen per mix for each selected day.
 S.M. Laskar, S. Talukdar / Construction and Building Materials 154 (2017) 176–190 179

Fig. 2. (a) Dummy specimen and (b), (c) details of specimens for slant shear test (All dimensions are in mm).

Microscopic images of UGGBS, FA, combination of 80% UGGBS and 20% FA; and formation of the geopolymerisation products led to loss in worka-
50% UGGBS and 50% FA were obtained using field emission scanning electron
bility of the mix [3,27,28]. Addition of FA in the mix caused mono-
microscope (FESEM) and presented in Fig. 3. FESEM images of microstucture of
S1 mix (consisting of UGGBS as binder) and S3 mix (consisting of 70% UGGBS and
tonous increase in slump value. The slump retaining property of
30% FA as binder) were also captured at 3 and 28 days. the mixes improved because of two prime factors: (i) FA particles
improved the mobility of the mix components because of its
smooth spherical shape (see Fig. 3) which eventually contributed
3. Results and discussion to delay of setting, (ii) FA at ambient temperature is unable to form
geopolymerisation products instantly, thus allowing more time for
3.1. Workability setting of the mix [1,2,29].
Addition of SP in the mixes contributed towards the release of
Fig. 4 presents the results from the slump test performed at var- the liquid component in the mixes which were trapped in the flocs
ious time interval. UGGBS as the only binding agent resulted in formed by the solid particles. This eventually increased the amount
GPC mix with very poor workability. The slump value at 5 min of available liquid component in the mix for better mobility of par-
was found to be only 25 mm which eventually became 0 at ticles and hence improved the workability. PE based SP improved
15 min. The mix acquired the shape of the slump cone with no sign the workability of the mixes more efficiently compared to SN based
of movement. High liquid demand due to high fineness and instant SP. Increase in workability and slump retention capacity was

Fig. 3. FESEM images of (a) UGGBS; (b) FA; (c) 80% UGGBS and 20% FA; (d) 50% UGGBS and 50% FA.
180 S.M. Laskar, S. Talukdar / Construction and Building Materials 154 (2017) 176–190

Fig. 4. Results from slump test.

monotonous due to increase in SP content. However, at very high strength at early ages. About 50, 80 and 90% of 28 days strength
SP content (3%), segregation was observed in the fresh mixes. This were attained by these mixes at 1, 3 and 7 days respectively.
resulted in erroneous slump value of mixes with high SP dosage. Variation of compressive strength was observed in the UGGBS
Depletion in slump value was observed due to increase in SH based GPC mixes with constant FA content of 30% and varying
concentration in mixes with 30% FA and 1.5% SP dosage. Increase dosage of both types of SP (see Fig. 6). GPC mix with 0.5% PE based
in SH concentration resulted in the increase in the total amount SP dosage attained maximum strength at all selected days. The
of solids in the solution (seen as w/s in Table 2). This caused the strength of this mix at 1 and 3 days was around 60 and 80% of that
viscosity of the solution to increase and hence obstruction to at 28 days. The improvement of strength of the mixes due to the
mobility of the particles. Increase in SH concentration also led to addition of PE based SP was because of the fact that it improved
deterioration in the slump retention capacity of the fresh mix. Sim- mobility of the particles in the mixes leading to their even distribu-
ilar observations were made by Lee and Lee [30] and Karakoca et al. tion and thus forming of homogenous concrete mixes. Mixes with
[31] in their investigations on geopolymer mixes. Results showed PE based SP attained higher strength than SN based ones for each
that reduction in slump value and retention property was monoto- SP dosage. Moreover, the rate of strength gain was also higher in
nous due to increase in SH concentration regardless of SP types in mixes with PE based SP. Drop in strength at early and later ages
the mixes. was observed in case of mix with dosage higher than 0.5% PE based
SP addition type also played significant role in controlling the SP. In case of GPC mixes with SN based SP, addition of SP and the
workability of the mixes. Type I PE based SP addition outperformed increment in the quantity led to monotonous loss in compressive
all other types. Type II and III SP addition resulted in mixes with strength. At 3% SP dosage, due to segregation the mix particles
almost similar workability. The performance of PE based SP when were not distributed homogenously resulting into GPC with very
mixed with SH activator (Type II) was lower than SN based SP. This low strength. This phenomenon was pronouncedly observed in
is in contradiction with the performance trend of SP when mixed mix with 3% SN based SP.
by other two methods. A change in appearance of the SH solution The effect of concentration of SH on compressive strength of
from transparent to whitish colour was noted on addition of the PE UGGBS based GPC at constant FA content of 30% and SP of 1.5%
based SP. This indicated a possible chemical change in the solution. can be seen in Fig. 7. GPC mixes with 10 M SH exhibited maximum
Detailed study in this regard is presented by Palacios and Puertas strength at early and later ages. Moreover, the rate of strength gain
[32]. was also observed to be higher irrespective of SP type. At 8 M SH
concentration, the rate of strength gain of mixes were lower than
3.2. Compressive strength that of mixes with 10 M SH concentration. This was in line with
results obtained in earlier studies [37,38] where the rate of
The compressive strengths of UGGBS based GPC mixes with var- strength gain was explained to be dependent on alkali concentra-
ious amounts of FA content are presented in Fig. 5. With the addi- tion. Lower SH concentration led to slow rate of formation of poly-
tion of FA, considerable loss in compressive strength has been merisation product with lower structural strength. However, the
observed though the workability improved. The rate of strength results for mixes with 12 and 14 M SH concentration were contra-
gain also reduced. Since, FA has low Ca content, the total Ca con- dicting with the previous studies on geopolymeric systems. While
tent of the mix got reduced as the FA content increased. Therefore, other studies revealed that rate of strength gain and strength at
the contribution of Ca towards strength attainment of mixes was specified days increase with alkali concentration [39,40], but in
reduced. Moreover, the specimens were cured at ambient temper- the present study, the compressive strength and rate of strength
ature. But in case of FA based geopolymer, elevated temperature gain were lowered due to increase in SH concentration. At high
curing is required for SiO2 and Al2O to form the geopolymerisation SH concentration, the pH of the liquid component in the mixes
products and hence, the contribution towards attaining of strength was high. The dissolution of Si, Al, Ca ions were enhanced. This
was reduced on increase of FA content in the mixes [2,29,33–36]. increased the concentration of the ions in the mixes, limiting the
The mixes containing high amount of FA exhibited very low early ions’ mobility and hence delayed the formation of coagulated
strength. At later ages, the mixes attained quite higher strength structures, i.e. delayed the polymerisation process [37]. Moreover,
compared to that attained at early ages. The mixes without any due to loss of workability at high SH concentration, a non-
FA content or low FA content (less than 40%), could attain high homogenous GPC matrix was formed. Hence, low rate of strength
 S.M. Laskar, S. Talukdar / Construction and Building Materials 154 (2017) 176–190 181

Fig. 5. Effect of FA content on compressive strength of mixes.

Fig. 6. Effect of SP type and content on compressive strength of mixes.

Fig. 7. Effect of alkali concentration on compressive strength of mixes.

gain and low strength were exhibited by the mixes with higher SH and III SP addition exhibited low strengths. In all, mixes with PE
concentration. The behaviour was in conformity with the results based SP showed superior performance in all types of SP addition
obtained by the authors in their previous works on UGGBS based methods.
GPM [22]. The rate of strength gain was lower for mixes with SN
based SP compared to the PE based ones, irrespective of SH 3.3. Bond strength (pull-out)
concentration.
UGGBS based GPC with Type I SP addition exhibited better rate Results from pull-out test on UGGBS based GPC mixes are pre-
of strength gain than other methods (see Fig. 8). This was observed sented in Table 3. It is evident that the strength of mix with 30% FA
in case of mixes with both SN and PE based SP. Mixes with Type II content is highest compared to others. Addition of FA provided
182 S.M. Laskar, S. Talukdar / Construction and Building Materials 154 (2017) 176–190

Fig. 8. Effect of SP addition type on compressive strength of mixes.

Table 3
Results from compressive, pull-out and slant shear tests (MPa).

Mix Compressive strength test Pull-out test (BSr) Slant shear test (BSc)
3 days 28 days 3 days 28 days 3 days 28 days
S1 42.33 51.78 6.03 7.86 7.03 8.00
S2 39.78 50.22 9.21 12.04 7.15 8.55
S3 38.22 49.22 10.58 14.30 6.67 8.19
S4 28.33 38.44 9.87 12.69 5.70 6.67
S5 20.22 31.56 7.62 9.97 3.76 5.34
S6 30.89 42.00 10.35 13.45 6.49 7.64
S7 40.44 51.89 11.96 15.48 7.88 9.40
S8 33.33 44.89 9.97 12.75 6.12 7.09
S9 37.11 46.44 12.67 16.37 7.15 8.79
S10 22.44 31.78 9.03 10.48 5.21 5.88
S11 28.22 40.89 10.11 13.11 5.94 6.85
S12 27.22 42.44 7.80 11.74 4.61 6.49
S13 32.89 42.89 10.72 14.11 6.43 7.88
S14 26.44 34.67 6.47 9.29 5.09 6.73
S15 28.33 38.00 7.74 10.88 6.18 7.40
S16 14.89 20.67 3.02 4.95 3.76 4.85
S17 22.33 28.00 4.62 7.34 4.97 6.37
S18 24.22 32.67 8.83 10.66 5.15 6.31
S19 27.78 34.11 9.39 11.66 5.88 7.22
S20 25.78 33.33 8.95 11.42 5.46 6.97
S21 29.56 38.22 9.89 12.51 6.24 7.28

better mobility to mix particles and resulted in homogenous Variation of bond strength with change of SH concentration in
matrix. This led to better bonding of the matrix with the rebar sur- UGGBS based GPC mixes with constant FA and SP content has been
face. However, addition of high amount of FA (>30%) led to reduced observed. Lower SH concentration of 8 M had similar effect on the
bond strength. The slow rate of strength gain and reduced strength mixes as that of 12 M concentration. The mixes with 8 M concen-
of FA based geopolymer compared to BFS based ones, may be tration attained low strength because of the fact that lower con-
attributed to occurrence of such phenomenon. centration cause slow polymerisation process [38] and formation
The enhanced bond strength was observed due to addition of PE of end products with low structural strengths. On the other hand,
based SP in the GPC mixes with constant FA content of 30%. PE slow strength gain and attainment of low bond strength with
based SP when added to the mixes produced homogenous mix mixes activated by 12 M SH was primarily related to the fact that
due to better workability and mobility to the mix particles leading high SH concentration produced high pH in the liquid component
to better bond between the matrix and the rebar. However, dosage and delayed the polymerisation process. This eventually led to
of PE based SP higher than 1.5% caused segregation in the mix and low strength gain in the mixes [37]. Moreover, higher SH concen-
hence it attained significantly low bond strength. The behaviour tration led to lower workability and hence formation of non-
observed was rather different for mixes with SN based SP. Addition homogenous matrix resulting in poor bond between the GPC
of SN based SP caused monotonous reduction in bond strength of matrix and rebar. The fall of strength was drastic for mixes with
the mixes with the rebar surface asserting the fact that such type 12 M SH onwards. Mixes with 14 M SH were poor in terms of
of SP is not suitable for UGGBS based geopolymer mixes. The test workability and setting time and hence, resulted in specimens with
results of mixes with SN based SP were in agreement with that large pores (see Fig. 9) and exhibited poor results. PE based SP con-
of compressive strength test results. tained mixes showed better bond strength with rebar at both early
 S.M. Laskar, S. Talukdar / Construction and Building Materials 154 (2017) 176–190 183

Fig. 9. Pull-out test specimens with visible cracks after failure.

Fig. 10. Pull-out test specimens without any visible crack after failure.

and later ages compared to SN based SP ones, irrespective of the SH Fig. 9) that originated from the interface of GPC matrix and rebar
concentration. and travelled to the surface of the specimen indicating strong bond
Type I SP addition exhibited better bond strength with rebar between both. On the other hand, specimens in Fig. 10 with lower
than type II and III. This was in conformity with the compressive slump value and bond strength failed without any visible cracks
strength test results. Maximum bond strength at 3 and 28 days development. This gave an indication that failure of such speci-
was observed in mixes with PE based SP added by Type I. mens occurred due to weak bond between GPC matrix and rebar
On examining the specimens after failure in the pull-out surface.
strength test, it was observed that in the specimens with higher To observe the influence of workability and compressive
slump value and bond strength wide cracks were visible (see strength on the bond strength of UGGBS based GPC with rebar,
184 S.M. Laskar, S. Talukdar / Construction and Building Materials 154 (2017) 176–190

Fig. 11. Influence of workability on bond strength of GPC with rebar for mix groups G1-G4.

results from pull-out test have been plotted as function of slump not hold good and consistent (R2 values range from 0.1168 to
value and compressive strength in Figs. 11(a)–(d) and 12(a)–(d) 0.6972).
respectively. Linear trend-lines have been plotted along with their The trend-lines for various GPC mix groups show increasing
coefficients of determination (R2). The trend lines show difference effect in bond strength of UGGBS based GPC with rebar due to
in bonding behaviour due to workability and compressive increase in compressive strength. Pronounced effect in increase
strength. No definite relation could be concluded between work- of bond strength with increase in compressive strength was
ability and the bond strength. Increase in slump value in some observed in case of G3 GPC mixes (R2 = 0.916). In case of G1 GPC
mixes led to improved bond strength. While in others especially mixes, a slight increase in bond strength of GPC with rebar was
when high quantity SPs were added, bond strength of GPC with observed due to higher value of compressive strength. Neverthe-
rebar reduced due to increase in slump value. The relation less, the relation in G1 GPC mixes was not found to be consistent
between workability and bond strength of GPC with rebar does (R2 = 0.0152).
 S.M. Laskar, S. Talukdar / Construction and Building Materials 154 (2017) 176–190 185

Fig. 12. Influence of compressive strength on bond strength of GPC with rebar for mix groups G1-G4.

3.4. Bond strength (slant shear) Improvement of bond strength was observed in the mixes at
certain amount of PE based SP. The bond strength of mixes reduced
Table 3 presents the results from slant shear test on UGGBS monotonously as the amount of SN based SP in the mixes
based GPC. Insignificant variation in strengths of the mixes have increased. Maximum bond strength was attained by mix contain-
been observed due to addition of FA up to 30% content. Amount ing 0.5% PE based SP. Amount higher than this though increased
higher than 30% FA in the mixes caused pronounced and monoto- workability but reduced the bond strength. The results from the
nous decrease in the bond strength. Higher amount of FA in the slant shear test are in line with the compressive strength test
mix caused pronounced reduction in strength due to the reason results. Very high amount of SP caused segregation in the mix that
specified in the Section 3.3. The slight improvement of strength eventually restricted formation of proper bonds within the
at 20% FA content was because FA addition improved workability geopolymer matrix and bond between the GPC and PCC surface.
that eventually produced homogenous mix and better bonding of However, on inspection of the specimens after failure in the test,
the mix matrix to the PCC substrate surface. it was noticed that in mixes with low bond strength, the bond
186 S.M. Laskar, S. Talukdar / Construction and Building Materials 154 (2017) 176–190

Fig. 13. Slant shear test specimens with failure plane along GPC-PCC interface.

Fig. 14. Slant shear test specimens with cracks in GPC and PCC region.

between the GPC and PCC interface was weaker than the bond It can be observed from Table 3 that, increase in bond strength
within the GPC or PCC matrix. The failure surface was along the occurred in the UGGBS based GPC with constant FA and SP content,
interface of the GPC and PCC (see Fig. 13). On the other hand, as on increasing the SH concentration. However, this was true only up
seen in see Fig. 14, the failure surface for GPC with higher bond to 10 M concentration. Higher concentration led to reduction in the
strength was not only along the interface of the GPC and PCC but bond strength. At higher SH concentration, the strength tend to
also within the GPC and PCC matrix. decrease monotonously. The phenomenon occurred in mixes
 S.M. Laskar, S. Talukdar / Construction and Building Materials 154 (2017) 176–190 187

Fig. 15. Influence of workability on bond strength of GPC with PCC for mix groups G1-G4.

containing both the types of SP. Due to high pH and delay in poly- pull-out strength test results. Type I SP addition outperformed
merisation, the bond formation of GPC matrix with the PCC was other types. Type II and III addition showed insignificant difference
not proper. Mixes containing SN based SP showed lower strength in the results for a given type of SP. PE based SP performed better in
compared to those containing PE based SP regardless of SH each type of addition.
concentration. Figs. 15(a)–(d) and 16(a)–(d) present the plot of results from
Bond strength results for mixes with various methods of slant shear test of UGGBS based GPC mixes as function slump value
SP addition are found to be in line with the compressive and and compressive strength respectively. From the plots, no definite
188 S.M. Laskar, S. Talukdar / Construction and Building Materials 154 (2017) 176–190

Fig. 16. Influence of compressive strength on bond strength of GPC with PCC for mix groups G1-G4.

relation could be established between workability and bond The trend-lines for various UGGBS based GPC mix groups show
strength. Increase in slump value in some mixes led to improved that increase in compressive strength increases the bond strength
bond strength while in other mixes especially when high quantity of GPC with PCC. Effect was prominent in case of G1 GPC mixes
of FA and SPs were added, bond strength deteriorated with the (R2 = 0.972). In case of G2 and G4 GPC mixes, relation between
increase in slump value. The relation between workability and the bond strength of GPC with PCC and compressive strength
bond strength of UGGBS based GPC with PCC does not hold good was found to be moderately consistent (R2 range from 0.711 to
and is inconsistent (R2 values range from 0.289 to 0.593). 0.785).
 S.M. Laskar, S. Talukdar / Construction and Building Materials 154 (2017) 176–190 189

Fig. 17. FESEM images of geopolymer mixes: (a) S1 at 3 days; (b) S1 at 28 days; (c) S3 at 3 days; (d) S3 at 28 days.

3.5. Microstructure study of geopolymer mixes 4. High workability of GPC mixes is observed when low concentra-
tion SH is used. Increase in concentration of SH lead to deterio-
On studying the microscopic images of the S1 and S3 GPC mixes ration of workability property which is pronounced when SH
presented in Fig. 17, it was observed that at early age, S1 mix con- concentration is higher than 10 M. Mixes with 10 M SH concen-
sisted of denser matrix than S3 mix. The unreacted FA particles as tration are found to exhibit highest compressive and bond
observed proved that at ambient temperature, the reaction process strengths at early and later ages compared to mixes with SH
for the formation of geopolymeric products is slow in case of FA as concentration of 8, 12 and 14 M. Higher SH concentration cause
binder. However, at 28 days such unreacted FA particles were not monotonous decrease in the compressive and bond strength.
visible, confirming its contribution towards geopolymerisation 5. Time of addition of SP shows significant influence on fresh and
process. No significant change was observed in the matrix of S1 hardened state properties of UGGBS based GPC. Type I SP addi-
mix. Hence, it can be concluded that the rate of strength gain of tion prove to be reasonable for the mix process as it contribute
the mixes was dependent on the amount of FA. towards better performance in terms of workability and
strength. The other types of SP addition are not beneficial in
4. Conclusion improving the properties of the mixes.

Results of laboratory experiments performed on 21 different

UGGBS based GPC mixes to evaluate the workability, compressive
and bond strength have been presented in this paper. The impor- Acknowledgement
tant observations from the study are as follows:-
The authors gratefully acknowledge the help and support
1. UGGBS based GPC has strong potential to be used as concrete extended by Counto Microfine Products Private Limited, India
repairing agent. The speciality of UGGBS based GPC is that at and Fosroc Chemicals India Private Limited, India by providing
1 day it can develop strength of about 60% of that at 28 days. materials required for the laboratory tests in the current research
The bond strength of the GPC with rebar and PCC surface is at IIT Guwahati, India.
found to be satisfactory at both early and later ages. However,
UGGBS alone as binder creates workability issues.
2. Addition of FA to the UGGBS based GPC contribute towards References
monotonous improvement in workability. The strength on the
[1] G. Gorhan, G. Kurklu, The influence of the NaOH solution on the properties of
other hand reduces. However, such reduction in strength is sig- the fly ash-based geopolymer mortar cured at different temperatures,
nificant for mixes with high FA content. In such mixes, both Composites Part B 58 (2014) 371–377.
early and later strengths are negatively affected. In case of bond [2] K.H. Yang, J.K. Song, A.F. Ashour, E.T. Lee, Properties of cementless mortars
activated by sodium silicate, Constr. Build. Mater. 22 (2008) 1981–1989.
strength results, the addition of FA improves the performance of [3] F. Collins, J.G. Sanjayan, Effects of ultra-fine materials on workability and
the mixes but up to low FA content level. strength of concrete containing alkali-activated slag as the binder, Cem. Concr.
3. Workability of the UGGBS based GPC improves due to addition Res. 29 (1999) 459–462.
[4] A.R. Brough, A. Atkinson, Sodium silicate-based, alkali-activated slag mortars
of SP. PE based SP perform better than SN based SP at all dosage
Part I. Strength, hydration and microstructure, Cem. Concr. Res. 32 (2002) 865–
level. At high dosage of 3%, segregation occurs in the mixes and 879.
leads to erroneous workability results. The compressive and [5] Y.J. Kim, S.C. Kim, Y.T. Kim, Corrosion resistance and hydration heat of concrete
bond strength of the mixes improve till lower level of PE based containing ground granulated blast furnace slag, KSCE J. Civil Eng. 7 (4) (2003)
399–404.
SP dosage. In case of mixes with SN based SP, the addition of SP [6] D.W. Law, A.A. Adam, T.K. Molyneaux, T. Patnaikuni, Durability assessment of
cause monotonous reduction in compressive and bond strength. alkali activated slag (AAS) concrete, Mater. Struct. 45 (2012) 1425–1437.
190 S.M. Laskar, S. Talukdar / Construction and Building Materials 154 (2017) 176–190

[7] S. Hu, H. Wang, G. Zhang, Q. Ding, Bonding and abrasion resistance of [25] ASTM C882/C882M – 13a, Standard Test Method for Bond Strength of Epoxy-
geopolymeric repair material made with steel slag, Cem. Concr. Compos. 30 Resin Systems used with Concrete by Slant Shear, vol. 04.02, Annual Book of
(2008) 239–244. ASTM Standard, United States, 2005.
[8] T. Phoo-ngernkham, V. Sata, S. Hanjitsuwan, C. Ridtirud, S. Hatanaka, P. [26] IS: 1199 – 1959, Methods of Sampling and Analysis of Concrete, Bureau of
Chindaprasirt, High calcium fly ash geopolymer mortar containing Portland Indian Standard, New Delhi, India, 2004.
cement for use as repair material, Constr. Build. Mater. 98 (2015) 482–488. [27] J. Zhu, Q. Zhong, G. Chen, D. Li, Effect of particle size of blast furnace slag on
[9] E. Vasconcelos, S. Fernandes, J.L. Barroso de Aguiar, F. PachecoTorgal, Concrete properties of Portland cement, Procedia Eng. 27 (2012) 231–236.
retrofitting using metakaolin geopolymer mortars and CFRP, Constr. Build. [28] F.A. Memon, M.F. Nuruddin, N. Shafiq, Effect of silica fume on the fresh and
Mater. 25 (8) (2011) 3213–3221. hardened properties of fly ash-based self-compacting geopolymer concrete,
[10] G.S. Manjunatha Radhakrishna, K. Venugopal, S.V. Maruthi, Strength Int. J. Miner. Metall. Mater. 20 (2) (2013) 205–213.
characteristics of open air cured geopolymer concrete, Trans. Ind. Ceram. [29] P. Nath, P.K. Sarker, Effect of GGBFS on setting, workability and early strength
Soc. 73 (2) (2014) 149–156. properties of fly ash geopolymer concrete cured in ambient condition, Constr.
[11] C. Zanotti, P.H.R. Borges, A. Bhutta, N. Banthia, Bond strength between concrete Build. Mater. 66 (2014) 163–171.
substrate and metakaolin geopolymer repair mortar: effect of curing regime [30] N.K. Lee, H.K. Lee, Setting and mechanical properties of alkali-activated fly ash/
and PVA fiber reinforcement, Cem. Concr. Compos. 80 (2017) 307–316. slag concrete manufactured at room temperature, Constr. Build. Mater. 47
[12] H. Alanazi, M. Yang, D. Zhang, Z. Gao, Bond strength of PCC pavement repairs (2013) 1201–1209.
using metakaolin-based geopolymer mortar, Cem. Concr. Compos. 65 (2016) [31] M.B. Karakoca, I. Turkmen, M.M. Maras, F. Kantarci, R. Demirboga, M.U. Toprak,
75–82. Mechanical properties and setting time of ferrochrome slag based geopolymer
[13] P. Duan, C. Yan, W. Luo, A novel waterproof, fast setting and high early paste and mortar, Constr. Build. Mater. 72 (2014) 283–292.
strength repair material derived from metakaolin geopolymer, Constr. Build. [32] M. Palacios, F. Puertas, Stability of superplasticizer and shrinkage-reducing
Mater. 124 (2016) 69–73. admixtures in high basic media, Mater. Constr. 54 (276) (2004) 65–86.
[14] P.K. Sarker, Bond strength of reinforcing steel embedded in fly ash based [33] K. Somna, C. Jaturapitakkul, P. Kajitvichyanukul, P. Chindaprasirt, NaOH-
geopolymer concrete, Mater. Struct. 44 (2011) 1021–1030. activated ground fly ash geopolymer cured at ambient temperature, Fuel 90
[15] K. Yun, P. Choi, Causes and controls of cracking at bridge deck overlay with (6) (2011) 2118–2124.
very-early strength latex-modified concrete, Constr. Build. Mater. 56 (2014) [34] J.C. Swanepoel, C.A. Strydom, Utilisation of fly ash in a geopolymeric material,
53–62. Appl. Geochem. 17 (2002) 1143–1148.
[16] M. Oner, K. Erdogdu, A. Gunlu, Effect of components fineness on strength of [35] S. Kumar, R. Kumar, S.P. Mehrotra, Influence of granulated blast furnace slag on
blast furnace slag cement, Cem. Concr. Res. 33 (2003) 463–469. the reaction, structure and properties of fly ash based geopolymer, J. Mater.
[17] H. Binici, H. Temiz, M.M. Kose, The effect of fineness on the properties of the Sci. 45 (2010) 607–615.
blended cements incorporating ground granulated blast furnace slag and [36] F. Puertas, S. Martinez-Ramirez, S. Alonso, T. Vazquez, Alkali-activated fly ash/
ground basaltic pumice, Constr. Build. Mater. 21 (2007) 1122–1128. slag cement strength behaviour and hydration products, Cem. Concr. Res. 30
[18] S. Teng, T.Y.D. Lim, B.S. Divsholi, Durability and mechanical properties of high (10) (2000) 1625–1632.
strength concrete incorporating ultra fine ground granulated blast-furnace [37] S. Alonso, A. Palomo, Alkaline activation of metakaolin and calcium hydroxide
slag, Constr. Build. Mater. 40 (2013) 875–881. mixtures: influence of temperature, activator concentration and solids ratio,
[19] IS: 1727-1967, Methods of Test for Pozzolanic Materials, Bureau of Indian Mater. Lett. 47 (2001) 55–62.
Standard, New Delhi, India, 2004. [38] C.D. Atis, E.B. Gorur, O. Karahan, C. Bilim, S. Ilkentapar, E. Luga, Very high
[20] IS: 383-1970, Specification for Coarse and Fine Aggregates from Natural strength (120 MPa) class F fly ash geopolymer mortar activated at different
Sources for Concrete, Bureau of Indian Standard, New Delhi, India, 2002. NaOH amount, heat curing temperature and heat curing duration, Constr.
[21] IS: 8112-1989, 43 Grade Ordinary Portland Cement – Specification, Bureau of Build. Mater. 96 (2015) 673–678.
Indian Standard, New Delhi, India, 2005. [39] E. Altan, S.T. Erdogan, Alkali activation of a slag at ambient and elevated
[22] S.M. Laskar, S. Talukdar, Development of ultrafine slag-based geopolymer temperatures, Cem. Concr. Compos. 34 (2012) 131–139.
mortar for use as repairing mortar, ASCE J. Mater. Civil Eng. 29 (5) (2017). [40] M.I. Khan, K. Azizlin, S. Sufian, Z. Man, Sodium silicate-free geopolymers as
[23] IS: 516 – 1959, Methods of Tests for Strength of Concrete, Bureau of Indian coating materials: Effects of Na/Al and water/solid ratios on adhesion strength,
Standard, New Delhi, India, 2004. Ceram. Int. 41 (2015) 2794–2805.
[24] IS: 2770 – 1967, Methods of Testing Bond in Reinforced Concrete, Bureau of
Indian Standard, New Delhi, India, 2002.