CC BY-NC-ND
0Accepted Manuscript
Title: PERFORMANCE ASSESSMENT OF NATURAL POZZOLAN ROLLER COMPACTED CONCRETE PAVEMENTS
Author: S.A. Ghahari A. Mohammadi A.A. Ramezanianpour
PII: DOI:
Reference:
S2214-5095(16)30047-X http://dx.doi.org/doi:10.1016/j.cscm.2017.03.004 CSCM 87
To appear in:
Received date: 21-6-2016
Revised date: 12-3-2017
Accepted date: 17-3-2017
Please cite this article as: <doi>http://dx.doi.org/10.1016/j.cscm.2017.03.004</doi>
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1 PERFORMANCE ASSESSMENT OF NATURAL POZZOLAN
2 ROLLER COMPACTED CONCRETE PAVEMENTS
4 S.A. Ghaharia*ghahary@hotmail.com, A. Mohammadib, A.A. Ramezanianpourc
7 a Department of Civil and Environment Engineering, Lyles School of Civil Engineering, Purdue University, Indiana, USA
8 b Department of Civil and Environment Engineering, Amirkabir University of Technology, Tehran, Iran
9 c Department of Civil and Environment Engineering, Concrete Technology and Durability Research Center, Amirkabir
10 University of Technology, Tehran, Iran
12 * Corresponding author. 610 South St., Apt 15, Lafayette, IN 47901; +1-803-603-3063
14 Abstract
15 Concrete pavement is cost effective and beneficial because of its sustainability and durability.
16 The maintenance and renovation periods for such pavement compared to other pavements are
17 relatively long; however, a significant issue with pavements, especially roller compacted
18 concrete pavements (RCCP), is salt scaling which occurs due to saline solutions such as deicer
19 salts. In the present work, the performance of RCC containing a natural pozzolan called Trass, as
20 a supplementary cementitious material, and an air-entraining agent for salt scaling was
21 investigated. Mechanical and durability tests were performed on specimens containing a water to
22 binder ratio of 0.32, with and without Trass, and an air-entraining agent. It was concluded that,
23 Trass could not improve the compressive and tensile strengths, however, the permeability was
24 improved. Moreover, the amount of mass loss due to salt scaling was not decreased. In all
25 concrete mixtures, using a suitable amount of an air-entraining agent to maintain a total air
26 content of 4.5% to 5% was found to be necessary for producing RCC containing Trass.
28 Keywords
29 Roller compacted concrete, natural pozzolan, air-entraining agent, salt scaling, supplementary
30 cementitious material, pavements.
32 Introduction
33 Advantages of using RCC, such as a high-rate of production and low cost [1], have increased
34 the incidence of its use, especially in pavement construction projects for heavy weight vehicles
35 like airports [2]. Roller compacted concrete for pavement (RCCP) has ingredients similar to
36 those found in conventional concrete. However, since it is a non-slump concrete, vibratory
37 compaction [3] has to be used in order to compact each 25 cm layer of concrete slabs; this work
38 should be done by equipment used for asphalt paving [4]. Problems reported for RCCP are the
39 rigidity and relative tendency to crack because of plastic shrinkage and low tensile strength [5].
40 To decrease the possibility of thermal cracking, RCCP is produced with low Portland cement
41 content, and consequently, with high amounts of supplementary cementitous materials such as,
42 fly ash, silica fume, and blast furnace slag [6, 7]. Much work has been conducted on improving
43 the frost resistance of RCCP through an air-entraining agent [8]. The amount of air entraining
44 admixtures that provides substantive air voids for countering the effects of freezing and thawing
45 (F-T) cycles on concrete specimens have been studied as well [9]. Production of RCCP with an
46 air-entraining agent is not feasible in some projects due to the inherent difficulty of entraining air
47 in dried concrete [10].
48 Some research projects have been performed on using pozzolans and natural pozzolans in RCCP.
49 It has been found that in specimens containing pozzolans and with a compressive strength higher
50 than 40 MPa, resistance to F-T cycles is acceptable [11]; however, the water to cement ratio
51 should be limited in order to prevent concrete bleeding. Silica fume, as one type of pozzolan, can
52 improve the resistance of concrete specimens to F-T cycles [12], meanwhile, the percentage of
53 silica fume should be limited at 5% to 7% to maintain durability requirements [13]. Higher
54 compressive strength due to adding silica fume leads to higher F-T resistance [14]. Using slag, as
55 another pozzolan, in concrete mixture shows similar results [15]. Fly ash, on the other hand,
56 reduces the resistance to F-T cycles [16, 17], which could be due to bleeding and segregation on
57 the surface of the specimens because of the pozzolans [18, 19]. Investigations of the effects of
58 supplementary cementitous materials on low-cement RCCP have shown these materials lead to a
59 reduction in compressive strength and resistance to F-T cycles [20].
60 According to a recent survey, the use of deicing salts on concrete pavements is growing,
61 especially in cold regions of the USA and Canada [21]. Therefore, the pavements should be
62 constructed not only to resist F-T cycles but also to be more resistant to scaling in the presence of
63 salt solutions and deicers. Therefore, it is a promising idea to introduce a feasible way of
64 improving the deicer salt scaling resistance of RCCP by using supplementary cementitous
65 materials. Trass, as a natural poozzolan is cost beneficial. It is a readily available natural
66 pozzolans and this natural pozzolan reacts well with air-entraining agents. Herein, in this study,
67 the effect of Trass, as a natural pozzolan, on workability, and mechanical and durability
68 properties of air-entrained and non-air-entrained RCC is investigated.
70 Experimental Program
71 Materials and Mixture Design
72 Specimens were cast with typical type I Portland cement which meets ASTM C150
73 specifications [22]. 20% by weight of cement was replaced with Trass natural pozzolan, and
74 specimens with a 0.32 water/binder ratio were cast. Local river sand as fine aggregate and
75 crushed stone as coarse aggregate with maximum aggregate size of 4.75mm and 19mm were
76 used according to the ACI recommendation. The densities of fine and coarse aggregates are
77 measured as 2520kg/m3 and 2580 kg/m3, with water absorptions of 2.8% and 1.5%, respectively.
78 The characteristics of cement and Trass are illustrated in Table. 1. In order to attain workability in
79 all mixtures, a liquid polycarboxylic ether-base as superplasticizer (SP) was used. The SP used in
80 this research had a specific gravity of 1.18 and 40% solid content.
83 In this research, an air-entraining agent with specific gravity and solid content of 1.2 and 45%
84 was used. Considering the maximum aggregate size of 19 mm and "moderate exposure", a
85 proposed dosage of suitable air-entraining agent is 0.04% to 0.1% by weight of cement [23].
86 After trial and error process in casting samples with the preferred total air content which is 4.5%
87 to 5% [24], the preferred dosage of the air-entraining agent was chosen as 0.06% by cement
88 weight. The calculation of air content percentage is discussed in the last section of this paper.
91 For each test three samples were cast. Specimens were cast regarding the mixture proportions
92 given in Table.2. R-32 denotes specimens with no air-entraining agent and pozzolan (W/C =
93 0.32), and T-A-32 denotes specimens with Trass natural pozzolan and air-entraining agent (W/C
94 = 0.32). Having batched the materials in a mixer, each mixture design was tested for workability
95 by VeBe test method, which is suitable for RCCP [25, 26].
98 1 Testing procedure and specimen preparation
99 Compressive and tensile strength tests were carried out on 150x300 mm cylindrical specimens at
100 the ages of 28, 90, and 180 days in accordance with ASTM C39 and ASTM C496 [27, 28].
101 150x150 mm cubic specimens were molded to be tested for water penetration and sorptivity
102 according to BS EN 12390-8 and BS EN 480-5 [29, 30]. Moreover, ASTM C672 [31] salt
103 scaling test was performed on two disk specimens of 450cm area and 7.5cm thickness. All
104 specimens were molded in accordance with ASTM C1176, which is designated to RCCP [32],
105 and were cured in a room with 50±5% relative humidity and 23 ±2 °C temperature. The
106 specimens were put in an F-T chamber for a long period of freezing and thawing cycle, i.e. -18
107 °C for 18 hours and 23 °C for 6 hours. In order to measure the distribution of air voids and the air
108 content percentage in specimens, the vertical profile of each 150x300 mm cylindrical specimen
109 was scanned, and the distribution and percentage of air voids were measured by Bubble Counter
110 Software using the method found in ASTM C457-12 [33].
111 112
113 2 Results and Discussion
114 3 Fresh concrete properties
115 To indicate the workability of RCC mixtures, the VeBe test, which is suitable for mixture
116 designs and specimens with no slump, was performed. The test was performed on three samples
117 for each mixture design. According to the results shown in Fig. 1, the value of the VeBe test,
118 when using Trass, is higher than that of the reference concrete. The obtained results indicate that
119 the workability has been reduced, possibly due to the high water absorption properties of natural
120 pozzolans; consequently, a larger amount of superplasticizer is needed to attain the VeBe results
121 compared with the mixtures with no Trass. When using 0.06% air-entraining agent, the VeBe
122 value for T-A-32 is 38 seconds, which is 10% lower than that of T-32, the same mixture without
123 air-entraining agent. This shows that the effect of the air-entraining agent in the workability of
124 plain concrete is similar to the specimens with Trass, and when more air bubbles or voids are
125 available in the concrete microstructure, a higher VeBe value is achieved, and this problem could
126 be alleviated by adding extra superplasticizer.
129 4 Hardened concrete properties
130 5 Compressive and tensile strengths
131 Results of compressive strength, relative compressive strength, and splitting tensile strength are
132 presented in Fig. 2, Fig. 3, and Fig. 4, respectively. The tests were performed on three samples
133 for each mixture design. Regarding the results, the compressive strength of specimens containing
134 Trass is 35% lower than that of the plain concrete at early ages which could be due to the low
135 pozzolanic activity. At early ages, low pozzolanic activity reduces the participation of cement
136 materials in hydration. However, gradually, at the late age of 90 days, the compressive strength
137 of T-32 is relatively improved and is 14% lower than that of R-32. This indicates the fact that
138 Trass has compensated its lag in production of C-S-H gel. According to Fig. 3, until the age of 7
139 days, the percentage of compressive strength improvement for the specimens which contain
140 natural pozzolan is 17% lower than that of the plain concrete; however, the percentage is
141 relatively stable from 28 days to 90 days. Moreover, due to the fact that tensile strength has a
142 direct relationship with compressive strength, the same trend can be seen from the results.
143 Tensile strength values signify that, the tensile strength of T-32 is 12% lower than that of R-32.
144 The 0.06% air-entraining agent is an important factor in the decrease of compressive and tensile
145 strengths. According to the results, the compressive and tensile strengths for T-A-32 is 7% and
146 5% lower than that of T-32 at the age of 90 days, which is probably related to a higher void
147 content due to the air-entraining agent. Furthermore, T-A-32 compared with R-A-32 has 11%
148 and 10% higher compressive and tensile strengths, respectively, at the late age of 90 days; this
149 could be due to the filling effect of natural pozzolan on the voids, and consequently, reduction in
150 their volume.
154 6 Water penetration
155 A Water penetration test was carried out in order to measure permeability. The test was
156 performed on three samples for each mixture design. Specimens with higher durability and
157 surface strength had a lower depth of water penetration because of a lower rate of chloride ion
158 penetration. The results of water penetration depths are illustrated in Fig. 5. The results indicate
159 that, at the age of 90 days, the specimen with Trass has a water penetration depth of 10.8mm,
160 14% lower than that of R-32. This could be due to the reduction of amount of cement by 20% of
161 its weight in the specimens containing supplementary cementitous materials. Dilution effect and
162 low pozzolanic activity of Trass are responsible for higher permeability at early ages; however,
163 at the age of 180 days, the water penetration depth of both plain concrete and concrete containing
164 pozzolan are relatively the same. On the other hand, when 0.06% air-entraining agent is used, the
165 water penetration depth for both types of mixtures is decreased. The lower depth could be due to
166 more porosity caused by the air-entraining agent. However, due to the filling ability and gradual
167 formation of C-S-H gel in the presence of Trass, the water penetration depth for T-A-32 is 13%
168 lower than that of R-A-32 at the age of 90 days. This could lead to less capillary porosity in the
169 meantime. No significant differences in water penetration results were seen among the 4 types of
170 mixtures at the age of 180 days, which could be due to the slow pozzolanic reaction of Trass that
171 fills the voids by making tortuous paths of water to be absorbed into.
174 7 Sorptivity
175 The test was performed on three samples for each mixture design. Results of sorptivity
176 coefficients (S), defined from BS EN 480-5 [29] equation as stated in the following, are
177 illustrated in Fig. 6:
180 Q = A*(C + S*t0.5)(1)
181 where Q is the amount of water absorbed; A is the cross section of the specimen that is in contact
182 with water; t is the time in seconds; C is the constant coefficient; and S is the sorptivity
183 coefficient of the specimen (m/s05).
184 Regarding the sorptivity coefficient as another index for the permeability of concrete, Trass
185 could considerably decrease the coefficient. The sorptivity coefficient for R-32 is significantly
186 higher than that of T-32 at the age of 90 days. Due to the pozzolanic activity of Trass, the value
187 of S for T-32 is 14% lower than that of R-32, which signifies a reduction in capillary porosity
188 and loss of connectivity in the pore structure and has been discussed in concrete technology
189 related discussions elsewhere [34-37]. Besides, after using the air-entraining agent, the value of S
190 has decreased because of the higher amount of porosity. The value of S for T-A-32 is 11.6(10-6)(
191 m/s05) which is 13% lower than that of R-A-32. This could be due to the capability of Trass to
192 reduce the capillary porosity and conductivity of the pores, and the results are in agreement with
193 the results obtained from the water penetration test.
194 Deicer Salt Scaling
195 After 50 cycles of F-T, the mass scaled off the surface of the specimens was weighted. The test
196 was performed on three samples for each mixture design. The results of the deterioration rate of
197 the deicer salt scaling test and the cumulative weight loss for the mass scaled off the surface of
198 the specimens cured for 28 days are given in Table.3 and Fig. 7, respectively. The same results
199 shown in Fig. 7 indicate that when exposed to a deicer salt solution, the mass loss for T-32 is
200 51% higher than that of R-32, as Trass has shown a low reaction with cement particles. It can be
201 concluded that while using Trass, the compressive strength is not reciprocally related to the mass
202 loss of the specimens. For R-32, R-A-32, and T-A-32, the mass loss is below the threshold of 1
203 kg/m2 which is required for good performance according to the Swedish standard [10]. On the
204 other hand, when the air-entraining agent is added, the overall mass loss for the air-entrained
205 specimens is lower than that of the non-air entrained specimens. As it goes below 4 °C, the air
206 voids may provide more spaces to accommodate expanding water, i.e. about 10% of their
207 volume, which is the main cause of the cracking, delaminating, or deteriorating of concrete
208 structures. The mass loss for T-A-32 is 0.92kg/m2 which is 56% higher than that of R-A-32;
209 therefore, R-A-32 has the highest salt-scaling resistance. In this condition, the compressive
210 strength is not reciprocally related to the results of the salt scaling test. Therefore, 0.06% air-
211 entraining agent by weight of cement, which provides the total air content of 4.5% to 5%, is
212 sufficient if using Trass; however, for each type of air-entraining agent, this value may change in
213 order to provide the total air content of 4.5% to 5%.
214 According to a recent study by Nili et al. (Nili et al., 2011), by using silica fume, higher
215 compressive strength could lead to a lower mass loss due to F-T cycles. This is reported to be
216 due to the improvement of the strength of the surface of the specimen as well as decreasing its
217 permeability; however, by using slag, higher compressive strength did not lead to a higher F-T
218 resistance [15]. Therefore, by improving compressive strength through materials that decrease
219 surface permeability, usually the mass loss due to F-T cycles can be limited. In fact, using Trass
220 leads to the creation of clustered air voids which accordingly creates more F-T and salt scaling
221 mass loss values; therefore, as the compressive strength has not been improved while using
222 Trass, the mass loss value in the F-T cycles has not been reduced either, although the overall
223 permeability has improved. All the results were in accordance with what was observed by
224 monitoring the surface of the specimens exposed to the F-T cycles as well. As can be seen from
Fig. 8 and Fig. 9, which are related to T-32 and T-A-32 respectively, the surface of T-32 deteriorated more than the surface of T-A-32.
Spacing factor analysis
To measure the air content percentage and air void distribution of the specimens, in order to support the durability results stated above, the vertical profile of each 150x300mm cylindrical specimen was provided and scanned, and then the distribution and percentage of the air voids were measured by Bubble Counter Software, according to ASTM C457-12 guidelines (ASTM C457, 2012). The test was performed on three samples for each mixture design. Fig. 10 and Fig. 11 presents the scanned figures of the profiles of T-32 and T-A-32, respectively. A lower spacing factor value means a shorter distance between air voids, which limits the distance that water must flow before reaching a void, thus allowing water to expand and freeze without causing perceptible damage. According to ASTM C-457 (ASTM C457, 2012), the air content (A) in % and the distribution of air voids known as the spacing factor (SF) can be obtained as follows:
SF = -p- (3) 4N
where Ta is the traverse length through the air void, Tt is the total length of the traverse, Tp is the traverse length through the paste, and N is the total number of air voids that are intersected. An SF value under 0.2mm could be sufficient for concrete durability [38] (ASTM C672, 2003). According to the SF values provided in Table.4, T-32 has an SF value similar to T-A-32, which could be due to the filling effect of pozzolans on the distribution of the air voids. That is, Trass has not altered the air void structure significantly. On the other hand, using the air-entraining agent decreases the SF value. The SF value for T- 32 is 32% higher than that of R-32, however, this value for T-A-32 is 0.151mm, which is 40% higher than that of R-A-32. This could be due to the heterogeneity of the specimens made with Trass, which makes the concrete structure develop more tortuous paths. These tortuous paths are suitable for larger volume of water to be absorbed into the concrete pore structure, and more voids may have been clustered compared to an ordinary concrete. In other words, it can be concluded that Trass has not led the air voids to be distributed homogeneously.
259 Conclusions
260 In this study the effect of Trass natural pozzolan and an air-entraining agent on mechanical and
261 durability properties of roller compacted concrete was investigated. Having conducted extensive
262 laboratory research, the following conclusions are drawn:
263 1. The value of the VeBe test when using Trass is higher than that of plain concrete, and
264 consequently, by using Trass, an extra amount of superplasticizer is needed to create a similar
265 VeBe value. When using an air-entraining agent, the VeBe value is 10% lower than that of the
266 specimen which contains Trass and does not contain the air-entraining agent.
267 2. The compressive strength of the specimens containing Trass is 35% lower than that of the
268 plain concrete at early ages, which could be due to the low pozzolanic activity of Trass that may
269 not participate in hydration at early ages. Gradually, at the age of 90 days, the compressive
270 strength of the specimens containing Trass is relatively improved, which indicates that Trass has
271 compensated its lag in the production of C-S-H gel.
272 3. Results signify that the tensile strength of the specimens containing Trass is 12% lower
273 than that of plain concrete. Additionally, the air-entraining agent decreased both the compressive
274 and tensile strengths.
275 4. At the age of 90 days, the specimens with natural pozzolan have a 10.8mm water
276 penetration depth, which is 14% lower than that of plain concrete. Dilution effect and low
277 pozzolanic activity of Trass are responsible for the higher permeability at early ages. Moreover,
278 when the air-entraining agent is used, the water penetration depths for both types of mixtures are
279 decreased.
280 5. The sorptivity coefficient for the specimens containing natural pozzolan is lower than that
281 of plain concrete, which signifies capillary porosity reduction and loss of connectivity in the pore
282 structure. The sorptivity coefficient of the specimens containing both Trass and the air-entraining
283 agent is lower than that of plain concrete. This could be due to Trass natural pozzolan's
284 capability to reduce the capillary porosity and conductivity of the pores.
285 6. The mass loss value for the specimens containing pozzolan, when exposed to a deicer salt
286 solution, is 51% higher than that of plain concrete. This is likely due to the lower reaction rate of
287 Trass compared with cement. In fact, using Trass leads to the creation of clustered air voids
288 which accordingly creates more F-T and salt scaling mass loss values; therefore, as the
289 compressive strength has not been improved while using Trass, the mass loss value in the F-T
290 cycles has not been reduced either, although the overall permeability has improved.
291 7. The scaling factor value for the specimens containing both Trass and the air-entraining
292 agent is 40% higher than that of plain concrete containing the air-entraining agent. This could be
293 due to the heterogeneity of the specimens made with Trass, which makes the concrete structure
294 develop more tortuous paths, suitable for a larger volume of water to be absorbed into, and more
295 voids may have been clustered compared to an ordinary concrete.
296 References
297 1. Pittman, D.W., Development of a design procedure for roller-compacted concrete(RCC)
298 pavements, 1993, University of Texas at Austin.
299 2. ACI 325.10R, State-of-the-art Report on Roller Compacted Concrete Pavement, 1991,
300 American Concrete Institute.
301 3. PIARC, Use of Roller Compacted Concrete for Roads, 1993, Technical Committee for
302 Concrete Roads, France.
303 4. Delagrave, A., et al., Deicer salt scaling resistance of roller-compacted concrete
304 pavements. ACI materials journal, 1997. 94(2): p. 164-169.
305 5. Jingfu, K., H. Chuncui, and Z. Zhenli, Strength and shrinkage behaviors of roller-
306 compacted concrete with rubber additives. Materials and structures, 2009. 42(8): p. 1117307 1124.
308 6. Pigeon, M. and V.M. Malhotra, Frost resistance of roller-compacted high-volume fly ash
309 concrete. Journal of materials in civil engineering, 1995. 7(4): p. 208-211.
310 7. Ragan, S.A. Proportioning RCC pavement mixtures. in Roller Compacted Concrete II:.
311 1988. ASCE.
312 8. Pigeon, M. and J. Marchand, Frost Resistance of Foller-Compacted concrete. Concrete
313 International, 1996. 18(7): p. 22-26.
314 9. Powers, T., Freezing effects in concrete. Special Publication, 1975. 47: p. 1-12.
315 10. Marchand, J., et al., Mixture Proportioning of Roller Compacted Concrete-A Review.
316 Special Publication, 1997. 171: p. 457-486.
317 11. Valenza, J.J. and G.W. Scherer, A review of salt scaling: I. Phenomenology. Cement and
318 Concrete Research, 2007. 37(7): p. 1007-1021.
319 12. Jacobsen, S., et al., Frost salt scaling of no slump concrete: effect of strength. Nord.
320 Concr. Res, 1991. 11: p. 57-71.
321 13. Pigeon, M., D. Perraton, and R. Pleau, Scaling tests of silica fume concrete and the
322 critical spacing factor concept. Special Publication, 1987. 100: p. 1155-1182.
323 14. Nili, M. and M. Zaheri, Deicer salt-scaling resistance of non-air-entrained roller-
324 compacted concrete pavements. Construction and Building Materials, 2011. 25(4): p.
325 1671-1676.
326 15. Bilodeau, A. and V. Malhotra, Deicing salt scaling resistance of concrete incorporating
327 supplementary cementing materials: CANMET research. RILEM PROCEEDINGS 30.
328 FREEZE-THAW DURABILITY OF CONCRETE, 1997.
329 16. Gebler, S. and P. Klieger, Effect of fly ash on the air-void stability of concrete. Special
330 Publication, 1983. 79: p. 103-142.
331 17. Assi, L., et al., Improvement of the early and final compressive strength offly ash-based
332 geopolymer concrete at ambient conditions. Construction and Building Materials, 2016.
333 123: p. 806-813.
334 18. Bilodeau, A. and V. MalhotraI, Concretes Incorporating High Volumes of ASTM Class F
335 Fly Ashes: Mechanical Properties and Resistance to De-icing Salt Scaling and to
336 Chloride-Ion Penetration. Special Publication, 1992. 132: p. 319-350.
337 19. Ramezanianpour, A.M., et al., Influence of initial steam curing and different types of
338 mineral additives on mechanical and durability properties of self-compacting concrete.
339 Construction and Building Materials, 2014. 73: p. 187-194.
340 20. Vahedifard, F., M. Nili, and C.L. Meehan, Assessing the effects of supplementary
341 cementitious materials on the performance of low-cement roller compacted concrete
342 pavement. Construction and Building Materials, 2010. 24(12): p. 2528-2535.
343 21. American Society of Civil Engineers. Report Card for Americas' Infrastructure. 2013
344 [cited 2013 12/05/13]; Available from: http://www.infrastructurereportcard.org/.
345 22. ASTM C-150, Standard Specification for Portland Cement, 2012, American Society for
346 Testing and Materials.
347 23. Kumar Mehta, P. and P. Monteiro, Concrete: Microstructure, Properties, and Materials
348 McGraw-Hill Professional, 2013, USA.
349 24. ACI 211.1-91, Standard Practice for Selecting Proportions for Normal, Heavyweight,
350 and Mass Concrete, 1991, American Concrete Institute.
351 25. ASTM C-1170, Standard Test Method for Determining Consistency and Density of
352 Roller Compacted Concrete Using a Vibrating Table, 2002, American Society for
353 Testing and Materials.
354 26. BS EN 12350-3, Testing Fresh Concrete. Vee Bee Test, 2009, British Standard European
355 Norm.
356 27. ASTM-C39, Standard Test Method for Compressive Strength of Cylindrical Concrete
357 Specimens, 2012, American Society for Testing and Materials.
358 28. ASTM C-496, Standard Test Method for Splitting for Splitting Tensile Strength of
359 Cylindrical Concrete Specimens, 2011, American Society for Testing and Materials.
360 29. BS EN 480-5, Test Methods, Determination of Capillary Absorption, 1997, British
361 Standard European Norm.
362 30. BS EN 12390-8, Depth of Penetration of Water Under Pressure, 2000, British Standard
363 European Norm.
364 31. ASTM C-672, Standard Test Method for Scaling Resistance of Concrete Surfaces
365 Exposed to Deicing Chemicals, 2003, American Society for Testing and Materials.
366 32. ASTM C-1176, Standard Practice for Making Roller-Compacted Concrete in Cylinder
367 Molds Using a Vibrating Table, 2013, American Society for Testing and Materials.
368 33. ASTM C-457, Standard Test Method for Microscopical Determination of Parameters of
369 the Air-Void System in Hardened Concrete, 2012, American Society for Testing and
370 Materials.
371 34. Ramezanianpour, A.A., et al., Laboratory study on the effect of polypropylene fiber on
372 durability, and physical and mechanical characteristic of concrete for application in
373 sleepers. Construction and Building Materials, 2013. 44: p. 411-418.
374 35. Ramezanianpour, A.A., S.A. Ghahari, and A. Khazaie. Feasibility Study on Production
375 and Sustainability of Poly Propylene Fiber Reinforced Concrete Ties Based On a Value
376 Engineering Survey. in 3rd International Conference on Sustainable Construction
377 Materials and Technologies (SCMT3). 2013. Coventry University, University of
378 Wisconsin.
379 36. Ramezanianpour, A.A., S.A. Ghahari, and M. Esmaeili, Effect of combined carbonation
380 and chloride ion ingress by an accelerated test method on microscopic and mechanical
381 properties of concrete. Construction and Building Materials, 2014. 58: p. 138-146.
382 37. Ghahari, S.A., et al., An Accelerated Test Method of Simultaneous Carbonation and
383 Chloride Ion Ingress: Durability of Silica Fume Concrete in Severe Environments.
384 Advances in Materials Science and Engineering, 2016. 2016: p. 12.
385 38. ACI 201.2R, Guide to durable concrete, 2001, American Concrete Institute.
388 Fig 1. Results of the VeBe test for fresh concrete
389 Fig 2. Results of the compressive strength for the specimens at the age of 28 days
390 Fig 3. Relative compressive strength for the specimens at the age of 7 to 180 days
391 Fig 4. Results of tensile strength for all of the specimens at the age of 90 and 180 days
392 Fig 5. Water penetration depth for the specimens at the age of 90 and 180 days
393 Fig 6. Sorptivity coefficient for the specimens at the age of 90 and 180 days.
394 Fig 7. The results of the salt scaling test
395 Fig 8. T-32 without the air-entraning agent after 50 F-T cycles
396 Fig 9. T-A-32 with the air-entraning agent after 50 F-T cycles
397 Fig 10 T-32 without the air-entraining agent
398 Fig 11 T-A-32 with the air entraining agent
Table 1 Chemical and physical characteristics of t ie cement and filler
Chemical Composition (%) Cement Trass
CaO 62.08 3.36
SiO2 21.10 67.2
AI2O3 4.18 14.14
Fe2O3 3.34 2.96
MgO 3.79 1.6
SO3 2.84 0.068
K2O 0.69 2.5
Na2O 0.14 4.3
Pozzolanic Activity at 7 days - 57%
Pozzolanic Activity at 28 days - 68%
Loss on ignition(%) 3.00 8.5
Physical properties Specific gravity 3.17 3.10
Blaine fineness (cm2/g) 3519 3200
403 Table 2 Mixture proportions for RCCP
Sample ID Binder (kg/m3) Ratio of the filler to cement (%) Aggregate (kg/m3) Air-entraining agent by Cement weight/volume (%) Super plastisizer dosage by Cement weight (%)
Cement Fille r Fine Agg. (FA) Coarse Agg. (CA) Tota l
R-32 330 0 0 1180 788 1968 0/0 0
R-A-32 330 0 0 1158 772 1930 0.06/0.15 0
T-32 264 66 25 1161 774 1935 0/0 0.8
T-A-32 264 66 25 1138 759 1897 0.06/0.15 0.6
405 Table 3 Deterioration rate and weight loss percentage that scaled off the surface
Number of cycles
Weight loss at each cycle (%)
Sampl e ID No 5 10 15 20 25 30 35 40 45 50 Cumulativ e weight los s (kg/m2)
R-32 1 1 1 1 2 2 2 2 2 2 2 0.57
2 1 1 1 2 2 2 2 2 2 2
(% ) 2.22 4.93 12.34 20.49 29.62 41.49 54.77 69.29 82.23 100
R-A- 32 1 1 1 1 1 2 2 2 2 2 2 0.41
2 1 1 1 2 2 2 2 2 2 2
(% ) 1.76 4.29 10.5 15.23 24.95 37.12 47.38 55.44 78.71 100
T-32 1 1 2 2 3 3 3 3 3 3 4 1.15
2 1 2 3 3 3 3 3 3 4 5
(% ) 2.80 6.11 19.43 32.13 37.94 54.95 69.36 78.83 90.62 100
T-A- 32 1 1 2 2 3 3 3 3 3 4 4 0.92
2 1 2 2 2 3 3 3 3 3 4
(% ) 2.23 5.17 16.56 25.08 33.65 48.95 61.05 73.17 85.47 100
408 Table 4 Spacing factor and air void percentage
Sample ID Spacing factor (mm) Air content (%)
R-32 0.103 2.9
R-A-32 0.074 4.9
T-32 0.186 3.2
T-A-32 0.151 5.5
VeBe value
R-A-32
415 Fig1. Results of the VeBe test for fresh concrete
T-A-32
Fig.2. Results of the compressive strength for the specimens at the age of 28 days
190 days 180 days
R-32 R-A-32 T-32 T-A-32
Fig.4. Results of the tensile strength for all of the specimens at the age of 90 and 180 days
à T
I ISO days 90 days
R-A-32
T-A-32
436 Fig.5. Water penetration depth for the specimens at the age of 90 and 180 days
« 0.20
o 0.15
■3 0.10
190 days ISO days
R-A-32
T-A-32
442 Fig.6. Sorptivity coefficient for the specimens at the age of 90 and 180 days.
■T-A-32 ■T-32 R-A-32 ■R-32
Number of F-T cycles
Fig.7. The results of the salt scaling test
Fig.8. T-32 without the air-entraning agent after 50 F-T cycles
Fig.9. T-A-32 with the air-entraning agent after 50 F-T cycles
Fig. 10 T-32 without the air-entraining agent
Fig. 11 T-A-32 with the air entraining agent
120 100 80 60 40 20 0
%6 %5.6 %7 %6.6
%9.3 %22 I %^2.4 %19.7 I %23.3 I %19.6 I %23.6 I
1 %17.6
%66.9 %59.9 %49.8 %50.1
R-32 R-A-32 T-32 T-A-32
0-7 days 7-28 days ■ 28-90 days ■ 90-180 days
Fig.3. Relative compressive strength for the specimens at the age of 7 to 180 days
Page 18 of 18
