BENEFITS OF PORTLAND-LIMESTONE CEMENT FOR CONCRETE 1 WITH ROUNDED GRAVEL AGGREGATES AND 2 HIGHER FLY ASH REPLACEMENT RATES 3 4 Jay Shannon 5 Graduate Research Assistant 6 Civil and Environmental Engineering 7 Mississippi State University (MSU) 8 501 Hardy Road-Mail Stop 9546, Mississippi State, MS 39762 9 662-325-3050 (ph) 662-325-7189 (fax) [email protected]10 11 Isaac L. Howard, PhD, PE 12 Associate Professor 13 Materials and Construction Industries Chair 14 Civil and Environmental Engineering 15 Mississippi State University (MSU) 16 501 Hardy Road-Mail Stop 9546, Mississippi State, MS 39762 17 662-325-7193 (ph) 662-325-7189 (fax) [email protected]18 Corresponding Author 19 20 V. Tim Cost, PE, F.ACI 21 Senior Technical Service Engineer 22 Holcim (US) Inc. 23 121 Hampton Hills Blvd., Canton, MS 39046 24 601-856-2487 (ph) [email protected]25 26 Wayne M. Wilson, PE, LEED AP 27 Senior Technical Service Engineer 28 Holcim (US) Inc. 29 4678 Arbor Crest Place, Suwanee, GA 30024 30 770-789-3254 (ph) [email protected]31 32 Paper Prepared for Consideration for Presentation and Publication at the 94 th Annual Meeting of 33 the Transportation Research Board. 34 35 Original Submission: August 1, 2014 36 Revised Submission: November 9, 2014 37 38 5,490 Words, 5 Figures (1250 words), 3 Tables (750 words) = 7,490 Total Equivalent Words 39 40 41 42 43 44 45 46
17
Embed
1 BENEFITS OF PORTLAND-LIMESTONE CEMENT FOR …docs.trb.org/prp/15-4049.pdf · 1 BENEFITS OF PORTLAND-LIMESTONE CEMENT ... 1 40% SCM replacement of cement with Class C fly ash was
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
BENEFITS OF PORTLAND-LIMESTONE CEMENT FOR CONCRETE 1
WITH ROUNDED GRAVEL AGGREGATES AND 2
HIGHER FLY ASH REPLACEMENT RATES 3
4
Jay Shannon 5 Graduate Research Assistant 6
Civil and Environmental Engineering 7
Mississippi State University (MSU) 8
501 Hardy Road-Mail Stop 9546, Mississippi State, MS 39762 9
Concrete specimens were 10.2 cm by 20.3 cm cylinders fabricated in accordance with 25
ASTM C192. Concrete was mixed in batches of 0.05 m3 in a laboratory concrete mixer, each 26
batch producing 12 cylinders. Immediately after mixing, concrete was tested for slump, air 27
8
Shannon et al.
content, and unit weight in accordance with C143, C231, and C138, respectively. Cylinders 1
were then fabricated at the mixing site, covered, and initial curing began. A small portion of the 2
batch was used to conduct time of setting testing according to C403. After initial curing, the 3
specimens were removed from molds and stored in a curing room meeting the curing 4
environment requirements of C192 until testing. Concrete compressive strength (fc) testing, using 5
the unbonded caps as described in C1231, was conducted at 7, 14, 28, and 56 days in accordance 6
with C39. 7
8
Petrographic Investigation 9
Four specimens that were subjected to compressive testing after a 56 day cure were evaluated via 10
petrography, with special attention to distinctions in the ITZ that might suggest differences in 11
paste-aggregate bond of OPC vs. PLC mixtures. These test cylinder specimens were prepared by 12
removing any sections damaged in testing using a standard block saw, and cutting to a sample 13
size of 9.5 cm (3.75 in) by 12.7 cm (5 in), 2.5 cm (1 in) thick. Each specimen was prepared for 14
optical microscopic examination according to ASTM C856. Observations were made and 15
reference images collected using a digital microscope with magnification up to 200X. 16
17
TEST RESULTS 18
19
Fresh Concrete Property Trends 20 Comparisons of fresh concrete properties of otherwise similar OPC and PLC mixtures provide 21
some insight into PLC vs. OPC early performance distinctions. Concrete slump and air for the 22
baseline mixtures were predicted to be about 20.3 cm (8.0 in) and 2.0%, respectively, based on 23
mix designs. In total 15 matched pairs were used in t-tests to determine significant differences in 24
properties. These matched pairs included multiple sources, multiple replacement levels, and 25
multiple admixture dosages as indicated in Figures 1-4. Mean values for slump were 20.5 cm for 26
OPC and 20.1 cm for PLC. Test results found a p-value of 0.3280 indicating that these slumps 27
were not statistically different. Mean air contents were 2.53% for both OPC and PLC. 28
Time of setting trends were evaluated using the same 15 concrete pairs and an additional 29
14 CP pairs. Mean concrete time of setting was 6.56 hr for OPC and 5.87 hr for PLC, and test 30
results found a p-value of 0.0003, indicating that the times of setting were indeed statistically 31
different. CP setting indication results yielded means of 15.57 hr for OPC and 12.89 hr for PLC, 32
with a p-value of 0.0092, again indicating that the CP setting indication was statistically 33
different. 34
35
SCM Replacement Rate Effects for Class C Fly Ash and Slag Cement 36 One cement source (source C) was used for OPC vs. PLC comparisons in mixtures using Class C 37
fly ash or slag cement at varying replacement rates (40%, 50%, and 60% for fly ash and 50%, 38
60%, and 70% for slag cement). Figure 1 shows concrete fc and CP fcp data for fly ash mixtures. 39
Parts (a) and (c) illustrate fc and fcp differences between fly ash replacement rates at test days of 40
7, 14, 28, and 56 days. In total 14 concrete mixtures (168 specimens) and 14 CP mixtures (252 41
specimens) were used to produce the data shown in these bar charts. An equality plot of OPC to 42
PLC trends is included in parts (b) and (d). Note that part (b) includes all data from part (a) as 43
well as the small sample of 0.52 w/cm mixtures with the alternate [2] admixture dosage. Part (d) 44
also includes more data than part (c) in the form of fcp from the additional paste test ages (1 and 45
180) not included in bar charts and the small set of mixtures with admixture dosage [2]. 46
9
Shannon et al.
1
2 a) Concrete fc b) Concrete Equality 3
4 c) Cement Paste fcp d) Cement Paste Equality 5
FIGURE 1. Incremental Replacement Rate Class C Fly Ash Results 6
7 Figure 1 parts (b) and (d) show that on average, all 3 fly ash replacement rates resulted in 8
higher compressive strengths with PLC than OPC. The overall percent increase, as illustrated in 9
the equality plots, was similar in both concrete and CP specimens, though there are clearly 10
different trends when similar replacement rates are compared, concrete vs. CP. In concrete 11
mixtures, 40% replacement produced the greatest fc values and ratio of PLC to OPC fc. As 12
replacement levels increased, both fc and the ratio of PLC to OPC fc decreased. In CP mixtures 13
this trend was essentially reversed, on average, as higher replacement mixtures generally 14
outgained lower replacement in both areas. This observation (different trends, concrete vs. CP) 15
may suggest different concrete paste-aggregate bond effects as influenced by the percentage of 16
fly ash in the mix. The mixtures with different w/cm and admixture dosage fell within a 17
reasonable range of the other PLC to OPC ratios portrayed in the equality plots. 18
Figure 2 shows concrete fc and CP fcp data for slag cement mixtures in the same format as 19
Figure 1. While the focus of the paper is on Class C fly ash replacement effects, slag cement 20
comparison trends may also be of interest and may help add to the understanding of performance 21
synergies of SCM-PLC systems in concrete as influenced by both chemistry and physical 22
0
10
20
30
40
50
60
7 14 28 56 7 14 28 56 7 14 28 56
f c(M
Pa)
Test Day
OPC
PLC
50% Ash 60% Ash40% Ash
y = 1.28xR² = 0.71
0
10
20
30
40
50
60
0 10 20 30 40 50 60
PL
C f
c(M
Pa)
OPC fc (MPa)
w/cm 0.43, Admix 1w/cm 0.52, Admix 2
0
10
20
30
40
50
60
70
80
90
7 14 28 56 7 14 28 56 7 14 28 56
f cp
(MP
a)
Test Day
OPC
PLC
50% Ash 60% Ash40% Ash
y = 1.23xR² = 0.90
0
10
20
30
40
50
60
70
80
90
0 10 20 30 40 50 60 70 80 90
PL
Cf c
p(M
Pa)
OPC fcp (MPa)
w/cm 0.50, Admix 1
w/cm 0.50, Admix 2
10
Shannon et al.
(fineness) cementitious properties. In this case only a single w/cm ratio (0.43 for concrete) and 1
admixture dosage [1] were used. Concrete and CP with slag cement reflected PLC strength 2
benefits at 7 days (note circled data points on equality plots, parts (b) and (d)), but at later ages 3
the benefits were usually less pronounced. Concrete performance at later ages was actually quite 4
similar, PLC vs. OPC. There are still some interesting trends and distinctions in trends between 5
concrete and CP performance, however. 6
Concrete strengths are on average noticeably greater for slag cement mixtures than fly 7
ash mixtures, especially at higher replacement rates, even though CP strengths are generally 8
lower. This suggests better inherent paste-aggregate bond with slag cement in all cases than with 9
fly ash, possibly related, in part, to the higher fineness of slag cement and relative coarseness of 10
fly ash particles. These impacts in fly ash mixtures are somewhat mitigated with PLC, which 11
contributes a high proportion of very fine (limestone) particles that enhance the overall particle 12
size distribution. 13
14
15 a) Concrete fc b) Concrete fc Equality 16
17 c) Cement Paste fcp d) Cement Paste fc Equality 18
Multiple Cement Sources Compared in 0% and 40% Class C Fly Ash Mixtures 21 The question may be posed whether the beneficial trends observed for PLC (vs. OPC) from one 22
source will be common to other cement sources. To address this and to contrast general 23
0
10
20
30
40
50
60
7 14 28 56 7 14 28 56 7 14 28 56
f c(M
Pa)
Test Day
OPCPLC
70% Slag60% Slag50% Slag
y = 1.00xR² = 0.49
0
10
20
30
40
50
60
0 10 20 30 40 50 60
PL
C f
c(M
Pa)
OPC fc (MPa)
0
10
20
30
40
50
60
7 14 28 56 7 14 28 56 7 14 28 56
f cp
(MP
a)
Test Day
OPCPLC
70% Slag60% Slag50% Slag
y = 1.12xR² = 0.78
0
10
20
30
40
50
60
70
80
90
0 10 20 30 40 50 60 70 80 90
PL
C f c
p(M
Pa)
OPC fcp (MPa)
7 Day
7 Day
11
Shannon et al.
performance trend differences of mixtures with no SCMs and those with 40% Class C fly ash, 1
mixtures with OPC and PLC samples from each of the 4 sources have been used to develop the 2
comparisons shown in Figures 3 and 4. 3
Figure 3 shows OPC and PLC data for all cement sources with no SCM (100% of the 4
cementitious content is cement) and 40% fly ash for both concrete and CP. A total of 16 5
concrete mixtures (192 specimens) and 16 paste mixtures (288 specimens) are represented. All 6
mixtures were made at a w/cm of 0.43 and admixture dosage [1]. In “No SCM” concrete 7
mixtures, part (a), OPC fc was slightly greater than PLC for cement sources A and C, but very 8
slightly lower for cement sources D and E. Overall, these differences (without fly ash) were 9
essentially negligible, which is consistent with other published data sets. Similar mixtures in CP 10
specimens, as seen in part (c), favored OPC with source A and PLC with source C and somewhat 11
with sources D and E. Though CP trends show more variability, again these overall differences 12
are not especially meaningful. In 40% fly ash mixtures, PLC strengths clearly excelled beyond 13
those of OPC in all CP (part (d)) and concrete (part (b)) comparisons, and by similar, meaningful 14
margins, in most cases. 15
16
17 a) Concrete fc No SCM b) Concrete fc 40% Ash 18
19 c) Cement Paste fcp No SCM d) Cement Paste fcp 40% Ash 20
FIGURE 3. 0% and 40% Class C Fly Ash Strength Results, 4 Cement Sources 21 22
23
0
10
20
30
40
50
60
7 14 28 56 7 14 28 56 7 14 28 56 7 14 28 56
f c(M
Pa)
Test Day
OPC PLC
"D""C""A" "E"
0
10
20
30
40
50
60
7 14 28 56 7 14 28 56 7 14 28 56 7 14 28 56
f c(M
Pa)
Test Day
OPC PLC
"D""C""A" "E"
0
10
20
30
40
50
60
70
80
7 14 28 56 7 14 28 56 7 14 28 56 7 14 28 56
f cp
(MP
a)
Test Day
OPC PLC
"D""C""A" "E"
0
10
20
30
40
50
60
7 14 28 56 7 14 28 56 7 14 28 56 7 14 28 56
f cp
(MP
a)
Test Day
OPC PLC
"D""C""A" "E"
12
Shannon et al.
Figure 4 presents equality plots for the mixtures depicted in Figure 3, with results for all sources 1
shown without differentiation. In parts (a) and (c), concrete and CP mixtures without SCMs 2
show little or no difference in strength performance on average, OPC vs. PLC. Figure 4 (a) 3
shows a modest favoring toward PLC, but with considerable scatter, this isn’t believed to be 4
especially meaningful. In parts (b) and (d), 40% fly ash mixtures indicate considerable 5
advantages with PLC, with much greater benefits in concrete (equality line slope of 1.46 and 6
every data point favoring PLC) than CP. Again, this is thought to be somewhat related to the 7
particle size contributions of PLC to the fly ash concrete mixtures and possibly to associated 8
improvements in paste-aggregate bond. 9 10
11 a) Concrete No SCM b) Concrete 40% Fly Ash 12
13 c) Cement Paste No SCM d) Cement Paste 40% Fly Ash 14
FIGURE 4. 0% and 40% Class C Fly Ash Equality Plots, 4 Cement Sources 15
16
Concrete Petrography Results 17 Concrete Petrography was performed on 4 specimens (No SCM OPC, No SCM PLC, 40% fly 18
ash OPC, and 40% fly ash PLC) from mixtures using cement source C, in the interest of 19
exploring observed strength trends thought to be possibly related to paste-aggregate bond 20
differences. Results are presented in Figure 5 along with an example specimen in part (a). In the 21
“No SCM” mixtures, the OPC paste portion was generally darker in color with a less uniform, 22
more mottled appearance than the PLC paste portion. OPC paste appeared coarser with a 23
medium texture, while PLC paste looked finer with a more medium fine texture ((b) and (c)). 24
25
26
27
y = 0.98xR² = 0.74
0
10
20
30
40
50
60
0 10 20 30 40 50 60
PL
C f
c(M
Pa)
OPC fc (MPa)
y = 1.46xR² = 0.64
0
10
20
30
40
50
60
0 10 20 30 40 50 60
PL
C f
c(M
Pa)
OPC fc (MPa)
y = 1.07xR² = 0.67
0
10
20
30
40
50
60
70
0 10 20 30 40 50 60 70
PL
C f p
(MP
a)
OPC fcp (MPa)
y = 1.22xR² = 0.91
0
10
20
30
40
50
60
70
0 10 20 30 40 50 60 70
PL
C f c
p(M
Pa)
OPC fcp (MPa)
13
Shannon et al.
1
2 a) Example Specimen b) No SCM OPC 20X c) No SCM PLC 20X 3