Neville, Adam M. and Aïtcin, Pierre-Claude High-performance concrete: an
overview. Materials and structures/Matériaux et constructions, vol. 31, n.º 206,
March, ...
CEMENT-SUPERPLASTICIZER INTERACTION Assessing behaviour stability
Júlia Paulo Vieira
Extended abstract
Supervisors: António Carlos Bettencourt Simões Ribeiro Augusto Martins Gomes
October, 2010
Cement-superplasticizer interaction: assessing behaviour stability
1. INTRODUCTION The use of water-reducing admixtures in concrete is a widespread practice in the building industry. Indeed these products help to increase the strength and quality of concrete since the water/cement (W/C) ratio can be reduced and the same workability achieved. A number of new products have been developed in the last few years and these highly efficient superplasticizers are being increasingly used in the industry. Though there are advantages in using these new materials their action is depends to a great extent on the characteristics of the cement. It is still not possible to foresee the compatibility of the materials, even knowing the superplasticizer and cement that are used, because a mere change in the batch of the cement supplied can alter the adsorption of the admixture and, therefore, its efficiency. Such change is not necessarily an impediment to the use of admixtures in concrete, but it makes it harder to optimise mixtures and can lead to unexpected behaviour, given the sensitivity to proportioning. The specific goal of the work described in this paper is to assess the change in the rheology of the cement pastes in the fresh state caused by using cement samples CEM I and CEM II, taken from separated batches. The same water/cement (W/C) ratio and cements from the same origin were used. The aim was to determine the influence of the cement-superplasticizer interaction on the rheology by measuring the spread diameter and flow time. The influence of the batch of cement was compared with the change due to altering the W/C ratio, so as to facilitate interpretation of the results and make the study more useful in practical terms. The results presented here show the most relevant conclusions of a study that also includes assessment of bleeding [1].
2. MATERIALS AND TEST PROCEDURES 2.1 Materials 2.1.1. Portland cement Six cements were chosen for the experimental work and given the following designations: S, CA and CS; the first letter of each tag represents the supplier and the second is the plant that produced the cement. The full list of cements used is, therefore: S I 42,5 R, CA I 42,5 R, CA I 52,5 R, CS I 42,4 R, CA II/A-L 42,5 R and CS II/A-L 42,5 R. Ten samples of first two cements, hereunder designated
results, while only 3 samples of the others were requested. The samples were taken on dates far enough apart to ensure they were separate productions. Tables 1 to 6 present the chemical and mineralogical composition and the Blaine fineness of the cements. The mineralogical composition was calculated using the Bogue equations [2]. This composition was also determined by the Rietveld method for cements CA I 42,5 R and CA II/A-L 42,5 R (information provided by the manufacturer), since the Bogue equations were not applied to the CEM II cements and the aim was to compare the relative performance of the two types of cement based on mineralogical composition. The filler content marked with an asterisk in these
1
Extended abstract
tables indicates that this was not supplied by the producer of the cement and it has therefore been determined by estimation. These figures were arrived at by multiplying the mean of the ratio of the filler content and the loss to fire of each sample for which the filler content was to be calculated. The 3 filler content figures in Table 2 and the 10 in Table 1 marked with an asterisk were thus determined from the 7 samples of cement CA I 42,5 R, whose filler content was known. The filler content of cement CS I 42,5 R indicated in Table 4 were calculated from the filler content of the samples of cement CS II/A-L 42,5 R.
Table 1
Chemical (Bogue method) and physical characteristics of samples of cement S I 42,5 R. Sample
Characteristics
Chemical (%)
04/06 12/06 18/06 25/06 02/07 16/07 23/07 30/07 06/08 13/08
Loss to fire
3.66
4.31
4.58
4.63
3.51
3.37
3.11
3.02
2.95
2.76
SO3
2.94
2.78
2.31
2.85
2.58
2.71
2.87
2.80
2.74
2.97
Na2O eq
0.25
0.37
0.36
0.45
0.36
0.31
0.32
0.36
0.35
0.32
Filler*
6.55
7.71
8.20
8.29
6.28
6.03
5.57
5.41
5.28
4.94
Mineral (%)
C 3S
65.74 66.19 64.96 65.68 63.40 67.15 69.66 59.87 71.86 71.41
C 2S
7.08
8.33
8.59
8.79
12.38
7.34
3.67
12.46
2.95
6.05
C 3A
6.08
6.13
6.10
6.33
5.80
5.64
5.68
5.01
5.43
5.78
9.89
9.95
9.86
9.83
10.01 10.35 10.22 10.68 10.68 10.77
473
420
425
488
C4AF 2
Physical
Table 2
Blaine (m /kg)
430
427
413
416
412
Chemical, mineralogical (Bogue method) and physical characteristics of samples of cement CA I 42,5 R. Sample
Characteristics
Chemical (%)
20/07 27/07 03/08 10/08 17/08 24/08 31/08 07/09 14/09 21/09
Loss to fire
2.49
2.56
2.79
2.68
2.50
2.74
2.73
2.39
2.64
2.61
SO 3
2.84
2.70
2.73
2.82
2.94
2.95
2.98
2.90
2.96
2.95
Na2Oeq
0.88
0.80
0.82
0.87
0.87
0.88
0.87
0.89
0.84
0.81
Filler
4.60
4.70
4.90
4.90
4.48* 4.90* 4.89*
4.20
4.60
4.60
C3 S
56.52 55.05 56.57 57.15 54.82 54.25 54.57 54.40 55.56 56.27
C2 S
12.32 14.00 12.97 10.50 14.29 14.12 13.76 14.27 13.59 12.71
C3 A
9.15
Mineral (%)
C4AF Physical
Table 3
2
Blaine (m /kg)
Mineral (%)
9.59
9.86
10.02
9.40
9.27
10.38 10.04 10.16 10.01 10.53 10.44
9.52
9.68
10.16
9.92
303
317
314
313
313
9.07
309
9.03
328
8.74
316
9.65
303
314
Mineralogical characteristics (Rietveld method) of samples of cement CA I 42,5 R. Sample
Characteristics
20/07 27/07 03/08 10/08 17/08 24/08 31/08 07/09 14/09 21/09 C 3S
75.05 74.01 74.53 74.27 68.21 62.80 70.69 65.02 66.46 68.31
C 2S
4.85
5.11
4.98
5.05
11.70 15.29
9.52
13.08 12.36 10.18
C 3A
5.47
5.68
5.58
5.63
6.82
5.16
8.02
C4AF
2
422
8.41
7.81
6.87
11.97 12.75 12.36 12.56 12.17 10.95 12.97 10.99 10.59 11.45
Cement-superplasticizer interaction: assessing behaviour stability
Table 4
Chemical, mineralogical (Bogue method) and physical characteristics of samples of cements CS I 42,5 R and CA I 52,5 R. Type of cement and sample Characteristics
Chemical (%)
CS I 42,5 R 30/09
21/10
25/11
12/10
02/11
09/11
Loss to fire
1.81
2.16
0.89
1.35
1.31
1.26
SO3
3.16
3.11
3.31
3.47
3.28
3.13
Na2Oeq
0.97
0.89
1.02
0.83
0.85
0.85
Filler
4.45*
5.31*
2.19*
3.80
2.90
4.30
C 3S
57.49
58.62
55.94
46.80
49.86
50.34
C 2S
12.34
11.25
13.73
21.77
19.38
18.24
C 3A
8.59
8.14
8.46
8.89
9.12
8.87
C4AF
8.67
8.86
9.10
10.41
10.10
9.83
291
315
302
417
421
421
Mineral (%)
Physical
Table 5
CA I 52,5 R
2
Blaine (m /kg)
Chemical and physical characteristics of samples of cements CA II/A-L 42,5 R and CS II/A-L 42,5 R. Type of cement and sample Characteristics
CA II/A-L 42,5 R 20/07
17/08
21/09
29/09
07/10
04/11
Loss to fire
7.30
7.87
8.19
6.06
6.44
5.41
SO3
2.99
2.90
3.31
3.66
3.60
3.43
17.00
18.00
18.80
15.20
18.50
10.80
388
394
391
447
439
420
Chemical (%)
Filler Physical
Table 6
CS II/A-L 42,5 R
2
Blaine (m /kg)
Mineralogical characteristics (Rietveld method) of the samples of cement CA II/A-L 42,5 R.
Characteristics
Mineral (%)
Sample 20/07 17/08 21/09
C 3S
68.36 66.18 67.82
C 2S
11.17 13.16 12.21
C 3A
6.12
C4AF
11.80 12.18 11.43
6.53
6.19
2.1.2. Superplasticizers One of the superplasticizers chosen for the study is 2
nd
generation, based on naphthalene sulfonate
rd
(tagged R in this work) and the other two are 3 generation, with polymer chains based on carboxylic ether (here tagged G and GS). Table 7 shows some characteristics of the superplasticizers used in the study. Throughout this experimental work the superplasticizers were used in the maximum dose recommended by the manufacturer.
3
Extended abstract
Table 7
Main characteristics of the superplasticizers used [3, 4, 5].
Characteristic
G
GS
R
Mass of solids content in mass (%)
20.5
21.5
36
Relative Density (20%)
1.05 ± 0.02 3 g/cm
Appearance
brown liquid
1.053 ± 0.02 3 g/cm yellowish cloudy liquid
pH
7.3 ± 1
Chloride ion content
None
recommended dose (%)
1.6
1.18 ± 0.03 g/cm
3
dark brown liquid
5±1
7±1
1.5
1.4
2.2 Test methods Compatibility between the superplasticizers and the cements can be assessed by testing the pastes [6, 7] with the March cone or with the mini-slump test, which is usually performed with the Kantro minicone [8]. This paper presents the results of the mini-slump test (hereunder called the spread test) and of an adaptation of the Marsh cone test (hereunder called the flow test) yielded by each cementsuperplasticizer pair used. The spread and flow tests for each paste made were carried out 30 minutes after mixing started. 2.2.1. Mixing Procedure All the mixes in this study were produced in a mixer that met the requirements set out in [9]. Mixes were prepared as follows: 1. the temperature of the water was taken; 2. the cement was placed in the pan; 3. about 5/6 of the water was added and the time count started; 4. it was mixed at slow speed for 30 seconds; 5. the pan was removed and the material was scraped from the sides of the pan and the paddle for 30 seconds; 6. the rest of the water and the admixture were added (the water was used to rinse out the superplasticizer container); 7. the pan was replaced and mixing resumed at slow speed for 30 seconds; 8. the procedure described in (5) was repeated; 9. there was 1 minute of high-speed mixing. 2.2.2. Establishing the W/C ratio of the pastes The W/C ratio was determined for each cement-superplasticizer pair, which would lead to a spread value in the 120 to 150 mm range, which was regarded as crucial to the methodology used, so as to be able to detect significant variations in behaviour, i.e. higher or lower fluidity, within the detection capacity of the method. Table 8 shows the reference W/C ratios determined and the spread obtained.
4
Cement-superplasticizer interaction: assessing behaviour stability
Table 8
Reference W/C ratios found for each reference cement, the composition of the pastes concerned and the results of the spread tests that established the ratios.
Producer of reference cement
S
Superplasticizer Sample used
G
GS
C R
G
06/08/2009 25/06/2009 12/06/2009 20/07/2009
(W/C)reference
0.35
0.27
0.29
GS
R
20/07/2009
27/07/2009
0.55
0.42
0.3
Cement (g)
750
750
750
750
750
750
Water (g)
262.5
202.5
217.5
412.5
315
225
(%) (g) Spread (mm) at 30 minutes
1.6 12
1.5 11.25
1.4 10.5
1.6 12
1.5 11.25
1,4 10,5
149.3
137.5
145.3
127.5
137.0
140.0
SP
In order to find an order of magnitude for the variation in the results obtained, pastes with both a positive (+0.03) and negative (-0.03) change in the W/C ratio relative to the reference cement were prepared using a sample of each reference cement and the 3 superplasticizers studied. The figures obtained for the pastes made with the ± 0.03 change in the W/C ratio were used to find the upper (Ls) and lower (Li) limits [1] to accommodate the results obtained for the 10 samples of each reference cement. 2.2.3. Spread test This was the first test carried out, at 30 minutes. The measurements of the mini-cone used were: base diameter 38.1 mm; top diameter 19.0 mm and height 57.0 mm. The test was carried out as follows: 1. a glass panel was placed on a levelled horizontal surface, and the centre of the mini-cone and two orthogonal diagonals were marked; 2. the mini-cone was place in the centre of the glass panel and it was slowly filled with the paste, without compacting; 3. the mini-cone was raised carefully so as not to disturb the movement of the material. The result of this test is the mean of the two orthogonal diameters of the spread paste. 2.2.4. Flow test After the spread test the paste was placed in a 220 ml glass funnel with a tube with an internal tube diameter 7 mm and mouth diameter of 95 mm. The test was carried out as follows: 1. the funnel was placed in its stand, mouth upwards, and a beaker on a balance was placed under the tube outlet; 2. the outlet was covered with a finger while the funnel was filled with paste (220 ml), with care being
outlet was uncovered and the time taken by the paste to flow out was measured. The flow time was determined for the following amounts: 200 g, 300 g and Mf where Mf is the total mass that flowed continuously, but the analysis of the results will only consider the figures for 300 g of flow material.
5
Extended abstract
3. RESULTS AND THEIR ANALYSIS 3.1. Spread and flow 3.1.1. Efficiency of the superplasticizers Analysis of Table 8 shows that it can be said that pastes made with reference cement CA needed higher W/C ratios than those made with reference S cement, to obtain a spread within the 120 to 150 mm range. It is thus concluded that the efficiency of the superplasticizers depends on the characteristics of the cement, with the superplasticizers studied being more efficient when combined with the cement from plant S. The table further shows that superplasticizer G requires a higher W/C ratio, with both cements. The water-reducing efficiency of the superplasticizers is not proportionately the same in the two cements, however, which again signifies the importance of the cementsuperplasticizer interaction. 3.1.2. Influence of the cement batch The results of the spread and flow tests on the pastes prepared with the reference cements, for the reference W/C ratios, are given in Figures 1 and 2, respectively. With respect to the stability of the effect of the superplasticizers on different cement batches, the figures show that there is greater dispersion of the results for reference cement CA than for cement S, so it can be concluded that the characteristics of the cement are important. It can also be noted in Figure 1 (on the right) that the direction of the more relevant changes, when higher than 20 mm, for instance, are similar for the 3 admixtures. This indicates a change linked to the characteristics of the cement, not to a random effect or one related to the properties of the superplasticizer. But when the changes are smaller (i.e. below around 20 mm), the differences are not consistent in the 3 superplasticizers. The importance of the W/C ratio is also obvious from the results provided earlier, since the results are very different for superplasticizers G and R, most clearly seen in cement CA I 42,5 R.
Figure 1 - Results of spread tests carried out at 30 minutes on cement S I 42,5 R (left) and for cement CA I 42,5 R (right).
6
Cement-superplasticizer interaction: assessing behaviour stability
Figure 2 - Results of flow tests carried out at 30 minutes on 300 g of flow material made with cement S I 42,5 R (left) and cement CA I 42,5 R (right).
Examining the influence of changing the W/C by ±0.03 (Table 9), it can be seen that the influence of a change in the W/C ratio is naturally more significant in pastes with a lower W/C ratio. But the type of superplasticizer also has an influence, since it can be seen, for example, that even for low W/C ratios the change of spread and of the flow time for pastes with admixture G is relatively small. Comparing the variations obtained for different cement batches with those due to varying the W/C ratio it can be seen that the influence of the cement batch on pastes made with S reference cement was always less than the influence of a ±0.03 change in the W/C ratio. With respect to the CA reference cement, the changes due to changing the cement batch can be greater than those prompted by altering the W/C ratio by ±0.03.
Table 9
Upper (Ls) and lower (Li) limit for the spread and flow tests, for a ±0.03 variation in the (W/C) reference.
Cement-superplasticizer pair S reference cement
CA reference cement
Spread (mm) at 30 minutes Ls Li
Flow time at 30 minutes for 300 g of paste (s) Ls Li
No. of tests outside the variation range Spread Flow
G
168.0
144.9
23.1
22.7
49.5
26.8
0
0
GS
160.3
81.6
78.7
73.0
n/flow*
677.0*
0
0
R
165.0
75.5
89.5
59.4
n/flow*
690.6*
0
0
G
160.0
141.1
18.9
8.5
12.0
3.5
4
5
GS
164.0
135.7
28.3
15.7
45.1
29.4
1
1
R
144.6
88.5
56.1
77.9
728.8
650.9
0
1
Maximum variation amplitude; * there was no flow in these pastes, and 750 s was considered for the purposes of determining .
3.1.3. Influence of fineness and C 3A and alkali content It can be seen in 3.1.1 and 3.1.2 that the pastes made with the two cements behave differently. The main differences between the S and CA reference cements are related to fineness and the C 3A and alkali content. The results of the two new cements, CA I 52,5 R and CS I 42,5 R, were then analysed
7
Extended abstract
to try and directly individualise the influence of two of the characteristics indicated, fineness and alkali content, and infer that of the third, the C3A content. Table 10 gives the results obtained for the two new cements, mean, coefficient of variation and amplitude of variation ( reference cement and the respective cement (a positive value represents a higher value in the pastes made with the new cement). Table 10 shows that, with respect to CA reference cement, the CA I 52,5 R cement, with greater fineness and similar content of alkali and C 3A, may or may not lead to significant variations in spread and in flow time, according to the superplasticizer used. This indicates from the start that fineness does not exert the same influence on all the superplasticizers. But it should be noted that the pastes have a different W/C ratio, viz. 0.55, 0.42 and 0.30, respectively, for the pastes with G, GS and R admixtures. The paste with admixture R has a greater concentration of solids and so the increased fineness may imply a change in the distance between particles that is more relevant than it is in the other pastes, hence the reduction in fluidity. As cement S I 42,5 R has greater fineness than cement CA I 42,5 R but similar to CA I 52,5 R, if this were the predominant characteristic in the behaviour of the pastes it might be expected that the pastes made with cement CA I 52,5 R would exhibit much better workability than the pastes made with cement CA I 42,5 R, in light of the W/C ratios of the reference pastes. This was not the case, however, and so it cannot be said that the effect of fineness on the cement-superplasticizer interaction is predominant.
Table 10 Mean, coefficient of variation and amplitude of variation ( ) between the mean of the cements indicated (CA I 52,5 R or CS I 42,5 R) and the CA reference cement for the spread and flow tests. Spread Cement-superplasticizer pair
CA I 52,5 R
CS I 42,5 R
Mean (mm)
(mm)
Flow Coefficient of variation (%)
Mean (s)
(s)
Coefficient of variation (%)
G
146.1
-1.5
6
11.65
1.46
24
GS
168.3
15.1
6
39.90
15.06
12
R
59.9
-63.5
6
750*
607.44
0
G
144.3
-3.2
11
9.19
-1.00
19
GS
138.9
-14.3
16
38.13
13.29
58
R
117.1
-6.3
18
216.60
74.04
89
Now comparing the CA reference cement with the CS I 42,5 R one, and following the same analytical reasoning, but for the results of the pastes made with cement CS I 42,5 R, comparing the results of the pastes made with cements CA I 42,5 R, S I 42,5 R and CS I 42,5 R, taking the content of C 3A to be reduced in cements CA I 42,5 R and CS I 42,5 R, the main difference between these two cements is the alkali content. As the alkali content of cement CA I 42,5 R is much higher than it is in S I 42,5 R, but lower than it is in cement CS I 42,5 R, then if this content were predominant for workability there should be a reduction of spread and an increase in flow time. This effectively occurred, which suggests that this content is predominant for the cement-superplasticizer interaction.
8
Cement-superplasticizer interaction: assessing behaviour stability
Despite the conclusion on the influence of the alkali content of cement on the cement-superplasticizer interaction, it was not large enough to explain the different behaviour of the reference cements on its own, especially when these were combined with superplasticizer G, where the reference W/C ratios were 0.55 and 0.35. Since it was not found that the interaction was not systematically affected by fineness, therefore, it can be inferred that the variation in the results of the two reference cements could have been due to their different C3A content. In relation to this, the effect was as expected, that is, poorer efficiency of the superplasticizer with increased C 3A content [10], This effect does depend on the type of superplasticizer used, however, as is easily seen from the reference W/C ratios obtained with the 3 products used. 3.1.4. Influence of the filler content Table 11 gives the essential figures for the spread and flow tests performed on the cement samples CA II/A-L 42,5 R and CS II/A-L 42,5 R with the aim of ascertaining the importance of filler content. The value shown for
is the difference between the means obtained with the reference CA or CS I 42,5 R cement
and the CA II/A-L 42,5 R or CS II/A-L 42,5 R cement, respectively (positive values signify higher results in the pastes using the CEM II cements). Workability tends to increase with the introduction of filler. The harmful effect of filler on workability in the case of cement CS II/A-L 42,5 R with admixture R is due to the higher solid content of these pastes, since these are the ones with a lower W/C ratio. Table 11 Mean, coefficient of variation and amplitude of variation ( ) between the mean of the cements indicated (CA II/A-L 42,5 R or CS II/A-L 42,5 R) and reference cement CA or CS I 42,5 R for the spread and flow tests. Spread Cement-superplasticizer pair
CA II/A-L 42,5 R
CS II/A-L 42,5 R
Mean (mm)
(mm)
Flow Coefficient of variation (%)
Mean (s)
(s)
Coefficient of variation (%)
G
170.6
23.0
12
7.70
-2.49
28
GS
203.1
49.9
10
16.98
-7.86
23
R
129.9
6.5
5
84.40
-58.16
13
G
156.8
12.5
7
8.50
-0.69
16
GS
155.5
16.6
4
27.57
-10.56
9
R
90.4
-26.7
9
325.40
108.80
76
3.4 Analysis in the statistics programme R Analysis with a statistical tool was performed to improve assessment of the results, with particular reference to the influence of the simultaneous change of two parameters. R was the programme used for the statistical analysis by multivariate linear regression. This programme is available at http//www.rproject.org. The most relevant variables for the spread and flow results were estimated (setpAIC function of the programme). The data used in each analysis were for just one cement, where there were 10 samples, and the 10 samples of the two reference cements were assembled so as to get the widest variation of the characteristics considered in the analysis. The analysis of the CA I 42,5 R cements was done jointly with the CA II/A-L 42,5 R cements, since there were only 3 samples of the latter. The cement characteristics evaluated in the analyses carried out for one superplasticizer and
9
Extended abstract
the same cement, were the alkali, C3A, C3S, C2S, C 4AF, SO 3, and filler content and Baline fineness. In the analyses on cements from different producers, for the same kind of admixture, the W/C variable was added. The analysis consisted of evaluating the hypothesis that the spread and flow results could be explained by a multiple linear model with Xi variables for observation y, in accordance with the 0,
below:
y In equation (1) intercept.
0
1
X1
2
X2
.....
n
Xn
, for i
(1)
0
is the
represents the model error, i.e. it is the difference between what is explained by the
systematic part of the model and what is observed. It should be noted that the parameters,
, are
estimated by the least squares method, thus minimising the error, . The stepAIC function performs a statistical analysis of the model, including all the chosen variables and only chooses those that allow minimisation of the error, eliminating the rest. The multiple linear model with the chosen variables has a p-value for an F test that expresses the greater or lesser significance of the model, and it is regarded as significant if the value is less than a determined significance level, which in this work is 5%, p 0.05. From the analyses we can see that workability is basically affected by the C3S, C 2S and clinker alkali content, and by the W/C ratio. Of all these characteristics of the pastes only the alkali content has a negative effect on workability, which is consistent with the initial analysis. The filler content also had a positive effect, albeit less clear, on workability and the fineness of the cement had a negative effect. Finally, it should be noted that an unexpected tendency was found for the C 3A content, which was shown to have a positive effect on workability. This finding was not rated, however, since it was felt that it could have been caused by the choice of W/C ratios in the reference cement pastes, which implied greater workability in the pastes with a higher C3A content.
4. CONCLUSIONS With respect to the water-reducing efficiency of the superplasticizers, it was found that this depends on the cement used, which indicates the importance of the cement-superplasticizer interaction. So it is not possible to classify the water-reducing capacity of a superplasticizer in absolute terms, and two different superplasticizers may be classified in order of opposite efficiency if they are combined with different cements. The results of the tests also showed that the new-generation superplasticizers are more sensitive to the chemical composition of the cements than the earlier ones. The older superplasticizers were often manufactured from naphthalene.
10
Cement-superplasticizer interaction: assessing behaviour stability
The spread and flow test results showed that the changes in workability as a function of cement batch may be greater than those that resulted from a ± 0.03 variation in the W/C ratio, and they appear consistently, regardless of the type of admixture, when the changes are significant. This is therefore relevant to the quality control of concrete on work sites. The analyses were intended to determine the potential causes of the variation in the results for different cements in general, and of small variations in the content of the components of an individual cement, but the overall outcome was inconclusive. However, when cements of different types or from different origins were analysed simultaneously, systematic trends of determined parameters of the pastes did emerge, which could be related to the fact that a broader range of variation of their content had been achieved. The results clearly show the influence of the C3S, C2S and clinker alkali content, and the W/C ratio, on workability. With respect to the influence of the C3A content, results were contradictory, but, in light of the poor consistency of the results, the negative influence on workability attributed to this ingredient should not be called into question.
REFERENCES [1]
VIEIRA, J.
Interacção
cimento-superplastificante:
Avaliação
da
estabilidade
do
comportamento. Dissertação para obtenção do Grau de Mestre em Engenharia Civil, Instituto Superior Técnico, Lisbon, 2010. [2]
ASTM C150
Standard specification for Portland cement. Annual Book of ASTM Standards,
American Society of Testing Materials, 2007. [3]
BASF Construction Chemicals Portugal, S.A.
Glenium 26 SCC
Superplastificante de
nova geração para betão rheodinâmico com grande tempo de trabalhabilidade. Ficha técnica. Rio de Mouro, Julho 2006. [4]
BASF Construction
Chemicals
Portugal, S.A.
Glenium
Sky 512
Adjuvante
superplastificante/redutor de água de alta actividade, à base de éter policarboxílico para betão preparado em central. Ficha técnica. Rio de Mouro, Fevereiro 2009. [5]
BASF Construction Chemicals Portugal, S.A.
Rheobuild 561
Superplastificante, redutor
de água para betões rheoplásticos, com longo tempo de trabalhabilidade, isento de cloretos. Ficha técnica. Rio de Mouro, Julho 2006. [6]
Aïtcin, Pierre-Claude
High-performance concrete. Modern concrete technology 5, E & FN
SPON, London and New York, 1998. [7]
Neville, Adam M.
Properties of concrete. Fourth edition, Longman Group Limited, England,
1995.
11
Extended abstract
[8]
Kantro, D.L.
Influence of Water-Reducing Admixtures on Properties of Cement Paste - A
Miniature Slump Test. Cement, Concrete and Aggregates, Vol.2, N.2, pp.95-102, 1980. [9]
Comité Européen de normalisation (CEN) NP EN 196-1:2006
Methods of testing cement
part 1: determination of strength. Portuguese version of European Standard EN 196-1:2005, IPQ, 2006. [10]
Neville, Adam M. and Aïtcin, Pierre-Claude
High-performance concrete: an overview.
Materials and structures/Matériaux et constructions, vol. 31, n.º 206, March, pp. 111-117, 1998.
12
