USE OF CONCRETE ADMIXTURES TO PRODUCE “WATERPROOF” CONCRETE– ASIA RESULTS Zhang Shu Qiang*, Grace Construction Products, Singapore Chen Hong Fang, Grace Construction Products, Singapore Zhong Hua, Grace Construction Products, Singapore Leung Fuk Ming, Grace Construction Products, Singapore Nick Peng, Grace Construction Products, Singapore 33rd Conference on OUR WORLD IN CONCRETE & STRUCTURES: 25 - 27 August 2008, Singapore

  Article Online Id: 100033056  The online version of this article can be found at: http://cipremier.com/100033056

This article is brought to you with the support of  Singapore Concrete Institute  www.scinst.org.sg    All Rights reserved for CI‐Premier PTE LTD  You are not Allowed to re‐distribute or re‐sale the article in any format without written approval of  CI‐Premier PTE LTD  Visit Our Website for more information  www.cipremier.com  

33rd Conference on OUR WORLD IN CONCRETE & STRUCTURES: 25 – 27 August 2008, Singapore

USE OF CONCRETE ADMIXTURES TO PRODUCE “WATERPROOF” CONCRETE– ASIA RESULTS Zhang Shu Qiang*, Grace Construction Products, Singapore Chen Hong Fang, Grace Construction Products, Singapore Zhong Hua, Grace Construction Products, Singapore Leung Fuk Ming, Grace Construction Products, Singapore Nick Peng, Grace Construction Products, Singapore

Abstract Waterproof concrete has reduced capillary water absorption properties as well as low permeability to water under pressure. Such qualities allow “waterproof” concrete to be used in water-resisting construction below ground in place of membranes, particularly where application of membranes is complicated by details and irregularities on the surface areas, thereby decreasing installation time and labour costs. In this paper a novel admixture to provide waterproof concrete is discussed. This waterproofing admixture provides added reduction in drying shrinkage to mitigate the development of cracks, thus eliminating a major concern with the concrete-only approach to waterproofing. In addition to enhanced resistance to cracking, concrete produced with the new admixture has excellent performance in the reduction of permeability and capillary absorption as well as good strength development [1]. Various laboratory test results and field projects mainly from Asian countries will be presented and discussed in this paper. Keywords: Absorption, admixtures, permeability, shrinkage, waterproofing 1. Introduction One of the major uses of concrete is in the basements of buildings. Often these basements are in environments where preventing the penetration of water is of major importance. Membranes, primarily in sheets, have been extensively used in these applications. However, ever increasing details and irregularities in the surface areas requiring protection lead to increased labour costs and installation time. For these applications, “waterproof “ concrete is an attractive alternative [1]. Waterproof concrete has reduced capillary absorption properties as well as low permeability to water under pressure. One problem with the use of waterproof concrete, using conventional waterproofing admixtures in lieu of a membrane, is cracks in concrete [2,3]. Cracking is often induced by restraining the autogenous and drying shrinkage of concrete [4]. Cracking due to plastic shrinkage can be addressed by the addition of fibers or good curing practices, and they are not addressed in this paper. Shrinkage-reducing admixtures (SRA’s) are an effective method of reducing shrinkage in concrete and subsequent cracking [5-7]. In addition, long-term research shows that at least one SRA based upon a glycol ether blend lowers the diffusion of chloride into concrete [8] indicating that permeability is reduced [9]. In this paper, we discuss the combination of a glycol ether SRA with hydrophobic materials to produce a water-free, waterproofing admixture for concrete. Data will be presented showing excellent performance in the reduction of permeability and capillary absorption as well as good strength

development. In addition, tests conducted in different countries under various conditions showed that this combination can also reduce drying shrinkage effectively. 2. Experimental 2.1 Laboratory Concrete in Singapore and Hong Kong Portland Blast Furnace Cement (PBFC B), natural sand, and crushed granite with maximum size of 25 mm were used for all concrete mixes. Admixtures, where applicable, met the requirements of one or more of the types in BS 5075 Pt3 Concrete Admixture Specification for high range water reducer. The waterproofing admixture is a new product formulated in North America (Adprufe® 100 – referred to as WPA in this paper) and modified in Asia (Adprufe AP1 and AP3 – referred to as WPA-AP in this paper) based upon a glycol ether shrinkage reducing admixture (SRA) and additional hydrophobic elements providing enhanced hydrophobic action. Laboratory specimens were made according to BS 1881–108: Testing Concrete Method for making test cubes from fresh concrete. Slumps were determined by BS 1881–102 and BS 1881-107: Testing Concrete Method for determination of the density of compacted fresh concrete. Compressive strengths were determined at the times noted according to BS 1881 Pt 116 at 3, 7 and 28 days. Three cubes were used for each testing age and the average values were calculated and presented. Shrinkage beams were tested according to ASTM C 157: Standard Test Method for Length Change of Hardened Hydraulic-Cement Mortar and Concrete with curing times as indicated in this paper. Water absorption was measured according to BS 1881: Part 122:1983, Testing Concrete Method for determination of water absorption. Water penetration was measured according to DIN 1048–5: 1991, Testing Concrete, Testing of Hardened Concrete (specimens prepared in mould). 2.2 Laboratory Concrete in Beijing, China P1 cement, natural sand, and crushed limestone with maximum size of 25 mm were used for all concrete mixes. Admixtures, where applicable, met the requirements of one or more of the types in GB 8076-1997: Concrete Admixture, Specification for high range water reducer. The waterproofing admixture is WPA-AP. Laboratory specimen-making and testing were done in accordance with GB 8076-1997: for concrete admixtures, GBJ82-85: Test Methods for Concrete Durability and JC 474-1999: Water-repellent Admixture for Mortar and Concrete. 2.3 Field Concrete and Laboratory Testing in Singapore Laboratory and field tests were conducted in Singapore for various projects. For these mixes, cube strengths and other concrete properties were measured according to BS 1881 (concrete mixing, fresh properties testing, strength and water absorption), DIN 1048 (water penetration) and ASTM C157 (drying shrinkage). 3. Results and Discussion 3.1 Laboratory Concrete in Singapore and Hong Kong Table 1 provides information on concrete mixtures used to evaluate WPA-AP. The dosage of WPAAP was less than 5 litres/m3, and is easily added using normally available concrete plant admixture dispensing systems. This facilitates autographic recording of its addition via computerized concrete batching systems. Mixtures with WPA-AP and a NSFC superplasticizer were highly workable and did not entrain air. It can be seen that in this case there is an increase in compressive strength relative to the 0.45 w/c typical mixture for this application and that strengths were comparable with a w/c of 0.38. The lower w/c is desired to lower permeability to water under pressure and reduce water absorption [2]. The water permeability of mixtures with WPA-AP at 28 days are approximately one-third that of the control at 0.45 w/c and even lower than the 0.38 w/c control indicating that the WPA-AP reduces permeability beyond that achieved by just lowering the w/c [9].

Table 1

Test Results from SETSCO Singapore

Property WRA-AP*, litre/m3 Superplasticizer**, litre/100kg cement Portland blast furnace cement content, kg/m3 Free w/c Concrete slump, mm Compressive strength @ 3 days, MPa Compressive strength @ 7 days, MPa Compressive strength @ 28 days, MPa Water penetration @ 28days, mm Drying shrinkage @ 28 days, % * Adprufe AP1/AP3 Waterproofing Admixture ** Daracem® 130 Superplasticizer

Plain Concrete 1 1.5 400 0.38 175 39.5 57.5 72.0 9.7 -0.029

Plain Concrete 2 0.9 400 0.45 150 33.0 46.0 64.5 18.5 -0.030

WPA- AP1

WPA- AP3

3.5 1.5 400 0.38 200 35.5 52.0 73.0 6.0 -0.011

5.0 1.6 400 0.38 150 37.5 51.5 68.5 4.9 -0.016

One feature of WPA-AP is its ability to reduce drying shrinkage as well as water absorption. Water absorption testing in accordance with BS 1881 Part 122 demonstrated significant improvements over the typically used 0.45 w/c control as well as the control made to the same 0.38 w/c as used in the WPA-AP mixtures, as shown in Figure 1. Figure 1 Absorption Performance of WPA-AP versus Control Mixtures in Table 1. BS 1881122 at 30 minutes. BSI Absorption Rate

Absorption Rate, %

3 2.5 2 7 days 28 days

1.5 1 0.5 0 Control-1

Control-2

W PA-AP1

WPA-AP3

Figure 2 shows reductions in drying shrinkage in accordance with ASTM C157 with 28 days of water curing followed by drying at 50% RH. There is a significant decrease in drying shrinkage with the addition of WPA-AP. The drying shrinkages of mixtures with WPA-AP at 28 days are approximately half that of the control at 0.45 and 0.38 w/c indicating that the WPA-AP reduces drying shrinkage significantly. Comparison tests were also carried out in SETSCO Singapore on WPA-AP versus competitive product by using slag cement, and the details are given in Table 2. In this case the w/cm was the same for both mixtures, however their dosage rates are significantly different (the dosage of WPA-AP is 3.75 liters/m3 vs the dosage of the competitive product of 30 liters/m3). As expected, the BS 1881 absorption values of mixture with WPA-AP are slightly higher, however its penetration and shrinkage are lower. More importantly, the compressive strengths of the mixture with WPA-AP are significantly higher for both early and late strength development. These results indicate that WPA-AP is a highly effective and well-balanced waterproofing admixture with multiple benefits.

Figure 2 Drying Shrinkage for Mixtures in Table 1. Drying started after 28-day water curing.

% Shrinkage

ASTM C 157 Drying Shrinkage

0.01 0.005 0 -0.005 -0.01 -0.015 -0.02 -0.025 -0.03 -0.035 -0.04

7

14

28

56

Control-1 Control-2 W PA-AP1 W PA-AP3

Days

Table 2

Test Results from SETSCO Singapore

Mix Proportion Portland Blast Furnace Cement, kg/m3 Fine Aggregate kg/m3 Coarse Aggregate kg/m3 Water kg/m3 Superplasticizer*, litre/100kg cement

WPA-AP 370 870 1010 140 2.2

Competitive Product 370 870 1010 117 2.8

3.75 0.38

30 0.38

Slump, mm Air content, %

125 2.6

130 3.4

Unit weight, kg/m3

2325

2290

Water absorption @ 7 days, %

1.0

0.7

Water absorption @ 28 days, %

0.6

0.5

Compressive strength @ 7 days, MPa

56.0

46.5

Compressive strength @ 28 days, MPa

73.5

65.5

Water penetration @ 28 days, mm

5.5

7.4

Drying shrinkage @ 28 days, %

-0.013

-0.015

Drying shrinkage @ 56 days, %

-0.021

-0.026

Waterproofing admixture, W/C ratio

litre/m3

* Daracem 130 Superplasticizer Additional comparison tests were carried out in Grace Hong Kong Laboratory as well as Materialab commercial laboratory in Hong Kong, between WPA-AP and competitive product using Portland cement and fly ash. Details of these comparison tests are given in Table 3. These mixtures were chosen to be in conformance with BS 1881 requirements, in which all the mixtures have the same w/c of 0.40. The dosage of WPA-AP was less than 5 litres/m3, while the dosage of competitive product was 30 litres/m3. Table 3 gives the mixture proportions, plastic concrete properties, compressive strength, water absorption and drying shrinkage results.

Table 3 Property

Test Results from Grace Hong Kong and Materialab Hong Kong Plain WPA-AP1 WPA-AP3 Concrete Waterproofing admixture, litre/m3 3.5 5.0 Superplasticizer*, litre/m3 4.2 4.2 4.2 3 Portland cement, kg/m 340 340 340 Fly ash, kg/m3 110 110 110 Free water/cement ratio 0.40 0.40 0.40 Air content, % 1.1 0.9 0.9 Concrete slump, mm 180 175 180 Initial Setting Time, hr:min 12hrs 11h 30min 12h 30min Compressive strength @ 3 days, MPa 36.9 36.5 35.8 Compressive strength @ 7 days, MPa 52.9 51.1 51.1 Compressive strength @ 28 days, MPa 77.8 74.5 72.4 Water absorption @ 7 days, % 3.4 1.3 0.8 Drying shrinkage @ 28 day, % -0.031 -0.025 -0.027 * MIRA® 218 polycarboxylate-based superplasticizer

Competitive Product 30 5.0 340 110 0.40 2.8 190 22hrs 25.8 40.8 61.5 0.9 -

Results show that the mixtures with WPA-AP and a polycarboxylate-based superplasticizer were highly workable, did not entrain air and have no effect on the setting time of the concrete. It can also be seen that in this case there was a slight lowering of compressive strength for the mixture containing high WPA-AP dosage, while there is significant strength reduction for the mixture containing the competitive product. Water absorption testing in accordance with BS 1881 Part 122 demonstrated significant improvements for the mixtures containing WPA-AP compared to the same 0.38 w/c control mix. Also, it shows that there is some decrease in drying shrinkage with the addition of WPA-AP. Although the materials and mix designs are quite different for Hong Kong and Singapore, the test results from Hong Kong corresponds very well with the Singapore test results, showing again that WPA-AP is a highly effective waterproofing admixture with multiple benefits. 3.2 Laboratory Concrete in Beijing, China Comparison tests were also carried out in Grace China Laboratory on WPA-AP by using P1 cement and the details are given in Table 4. These mixtures were chosen to be in conformance with GB 8076, GBJ82-85 and JC 474 requirements, in which the control mix has w/c of 0.68, while the mixtures with WPA-AP have the same w/c of 0.46. The dosage of WPA-AP was 3.5 litres/m3. Table 4 gives the mixture proportions, plastic concrete properties, and compressive strength, water absorption rate and permeability rate results. Table 4 Test Results from Grace China Lab Property Plain Concrete Waterproofing admixture, litre/m3 Superplasticizer*, kg/100kg cement Daravair AEA**, kg/100kg 3 P1 cement, kg/m 330 Free water/cement ratio 0.68 Air content, % 1.0 Concrete slump, mm 175 Compressive strength @ 3 days, MPa 15.7 Compressive strength @ 7 days, MPa 21.1 Compressive strength @ 28 days, MPa 25.3 Absorption ratio @ 28 days for 48 hrs, % Permeability ratio @ 28 days, % Drying shrinkage ratio @ 28 days after 3-day moist curing, %

WPA-AP1

WPA-AP3

3.5 1.4 0.9 330 0.46 3.2 205 37.6 (239%) 47.1 (223%) 53.2 (210%) 30 12 91

3.5 1.4 0.9 330 0.46 2.8 205 41.7 (265%) 51.0 (242%) 60.2 (238%) 22 14 79

JC474 Requirement > 100% > 110% > 110% < 65 < 30 < 125

* ADVA® polycarboxylate-based superplasticizer ** Daravair® AEA air-entraining admixture Results show that the mixtures with WPA-AP and a polycarboxylate-based superplasticizer were compatible with air-entrained concretes. It can also be seen that in this case the strength development is very good. Water absorption testing in accordance with JC 474 demonstrated significant improvements for the mixtures containing WPA-AP over the control mix. Also, it shows that there is a significant reduction for permeability and drying shrinkage for the mixtures containing WPAAP. Once again, WPA-AP demonstrates very good dosage efficiency, strength development, water absorption, permeability and drying shrinkage reduction, and it meets JC 474 requirement as an waterproofing admixture excellently. 3.3 Field Concrete and Laboratory Testing in Singapore Field Project 1 – Hospital Basement Diaphragm Wall in Singapore Delays in this project had resulted in the push to severely tighten construction schedule. Therefore the speed for various jobs became a top priority. In considering waterproofing for the project, the specification team was mindful of the fact that the diaphragm wall, right next to a drain, posed additional leakage risk. The waterproofing specifications for the basement slab were especially stringent, and it required the highest level of protection (Grade Four Waterproofing according to Singapore’s local standards) and an upward, fully bonded system. Having taken the time to understand the project requirements, the Grace team proposed a system backed by detail drawings, data and test results. The recommended waterproofing system included the dual tanking combination of Preprufe® 300R, a unique, pre-applied, fully bonded membrane system, and WPA-AP3, an innovative liquid admixture for waterproof concrete. The main requirements for the waterproofing concrete contains hydrophobic and pore blocking admixture that includes (1) less than 1.0% water absorption at 7days in accordance with BS 1881 Pt 122, and (2) water penetration depth less than 15mm at 28 days in accordance with DIN 1048, or the coefficient of water permeability less than 5x10-13 in accordance with HDB Singapore Method. The Grace solution offered not only a full and complementary system, it also met project schedule requirements and cost considerations. In all, the applicator team took three months instead of the usual five to complete an area of about 28,000 m2. Field Project 2 – Drainage Project in Singapore This project required (1) less than 1.0% water absorption at 28 days in accordance with BS 1881 Pt 122, and (2) water penetration depth less than 15mm at 28 days in accordance with DIN 1048. Concrete was produced by local ready-mixed concrete company by using OPC cement and w/c of 0.38. The test results are summarized in Table 5, and they show that Grace WPA-AP met the project requirements very well. Table 5

Field Test Results from Singapore

Property Waterproofing admixture, litre/m3 Superplasticizer*, litre/100kg cement OPC cement, kg/m3 Free water/cement ratio Concrete slump, mm Compressive strength @ 3 days, MPa Compressive strength @ 7 days, MPa Compressive strength @ 28 days, MPa Water absorption @ 7 days, % Water absorption @ 28 days, % Water penetration @ 28 days, mm *Daracem 130 NSFC-based superplasticizer

WPA-AP Concrete 3.5 1.5 400 0.38 220 29.7 45.8 52.9 1.02 0.86 6.0

Field Project 3 – Commercial Building Diaphragm Wall Project in Singapore Due to high seepage risk, dual protection with waterproof concrete and cavity drainage are designed to minimize leakage at the wall. The estimated concrete volume is 24,000 m3. This project required less than 1.0% water absorption at 28 days in accordance with BS 1881 Pt 122. Concrete was produced by local ready-mixed concrete company using both OPC and PBFC cements with w/c of 0.40. The dosage of WPA-AP3 was 2.5 litre/ m3. Good water absorption result (~0.5% at 28 days) was obtained. 3.4 Additional Discussion Table 6 is a summary of the water absorption test results from various Asian countries at different WPA-AP dosage levels and w/c ratios. It shows that water absorption result is a function of both WPA-AP dosage and w/c ratio. Generally, higher WPA-AP dosage results in lower absorption result and lower w/c, and can also result in lower absorption rates. However, the effect of WPA-AP dosage seems to be more significant. Further analysis can be made in the future. However, other factors such as type of cementitious materials, aggregates and their quality may also have some effect on water absorption results, though the effects are less significant. Table 6

Summary for Lab and Field Test Results

No. 1

W/C 0.5

WPA-AP Dosage (litre/m3) 2.5

Water Absorption @ 28 days (%) 1.30

2

0.5

3.5

1.15

3

0.5

2.5

1.00

4

0.5

3.5

0.80

5

0.4

2.5

1.00

6

0.4

3.5

0.70

7

0.4

3.5

1.00

8

0.4

5.0

0.60

9

0.38

5.0

0.40

10

0.38

5.0

0.70

11

0.38

3.75

0.60

12

0.38

3.5

0.50

13

0.38

3.75

0.60

14

0.38

2.5

1.01

15

0.38

3.5

0.86

The various test results from both the laboratory and field projects on a similar WPA admixture from North America and the UK were very similar to Asia results, especially if one looks at improvements versus the controls. Improvements were noted at the same w/c, clearly demonstrating that the addition of WPA is making the concrete more impermeable. In addition, the reduction in shrinkage and cracking with the use of WPA is an important feature. To demonstrate the effects of reduced shrinkage in terms of cracking resistance, restrained shrinkage performance was measured in North America according to ASTM C 1581 Standard Test Method for Determining Age of Cracking and Induced Tensile Stress Characteristics of Mortar and Concrete under Restrained Shrinkage. The results showed that the mixture with WPA had significant improvement over the control mix with respect to time to cracking and crack size, with improved performance related to reduced drying shrinkage. Cracking of concrete due to autogenous or drying shrinkage can clearly compromise the waterproofing properties of the concrete. Crack mitigation is thus an important feature that will help to minimize future maintenance costs [1]. Significant waterproof concrete projects that contain WPA in UK include (1) Kings Waterfront, Liverpool: basement; (2) Slateford Road, Edinburgh: Ring Beam; (3) Kendal, Cumbria: car park basement slab at riverside.

4. Conclusions Chemical admixtures can be used to significantly reduce the water permeability and water absorption properties of concrete. The WPA and WPA-AP admixtures discussed in this paper have the added benefit of reducing drying shrinkage, which was shown to significantly mitigate cracking under drying conditions. In summary this waterproofing admixture has the following beneficial properties: (1) Significant reduction in water permeability and water absorption at the same water-to-cementitious ratio as the control. (2) Significant reduction in drying shrinkage that corresponds to improved resistance to cracking under drying conditions. (3) Reductions in water absorption, shrinkage, and water permeability were observed with typical OPC, fly ash or GGBFS concretes. (4) No adverse affect to concrete’s air content and setting time. (5) Highly dosage-effective and show good compressive strength development. 5. Acknowledgment The authors would like to thank Neal S. Berke et. Al., of W. R. Grace & Co.’s global R&D Centre, for their preliminary paper and guidance to this paper. The authors would also like to thank Shaun Tan of W. R. Grace (Singapore) Ltd. for his assistance in the collection of field project information and results.

References [1] Antonio J. Aldykiewicz, Jr., Neal S. Berke, et al, Use Of Concrete Admixtures To Produce “Waterproof” Concrete. To be published. [2] Roy, S. K. and Northwood, D. O., “Admixtures to Reduce the Permeability of Concrete,” in Durability of Concrete, Proceedings Fourth CANMET/ACI International Conference, Sydney, Australia, 1997, SP-170, Malhotra, V. M., ed., American Concrete Institute, Farmington Hills, MI (1997), pp. 267-284. [3] Mathieu, G. and Sari, J., “Survey of Water Towers, Reservoirs, Tanks, and Basins: Their Conditions and the Watertightness of the Waterproofing,” in Durability of Concrete, Proceedings Third CANMET/ACI International Conference, Nice, France, 1994, SP-145, Malhotra, V. M., ed., American Concrete Institute, Farmington Hills, MI (1994), pp. 1033-1050. [4] Early Age Cracking in Cementitious Systems, RILEM Report 25, RILEM (2003), 337 pp. [5] Folliard, K. J. and Berke, N. S., “Properties of High Performance Concrete Containing ShrinkageReducing Admixture,” Cement and Concrete Research 27(9) (1997), pp. 1357-1364. [6] Weiss, J. W., and Shah, S. P., “Restrained Shrinkage Cracking: The Role of Shrinkage Reducing Admixtures and Specimen Geometry”, Pre-Proceedings of the RILEM International Conference on Early Age Cracking in Cementitious Materials (EAC’01), Haifa, Israel, March 12-14, 2001, pp. 145157. [7] Bentur, A., Berke, N. S., Dallaire, M.P., and Durning, T. A., “Crack Mitigation Effects of Shrinkage Reducing Admixtures,” Design and Construction Practices to Mitigate Cracking, SP-204, American Concrete Institute (2001), pp. 155-170. [8] Berke, N. S., Dallaire, M.P., Hicks, M. C., and Kerkar, A., “New Developments in ShrinkageReducing Admixtures,” Superplasticizers and Other Chemical Admixtures in Concrete, SP-173, American Concrete Institute (1997), pp. 971-998. [9] Nokken, M. R. and Hooton, R. D., “Development of Early-Age Impermeability in High-Performance Concretes,” PCI Bridge Conference and PCI/FHWA International Symposium on High Performance Concrete, Orlando, FL, October 19-22, 2003, (CD ROM Proceedings) Paper 104, 16 pp.