Effectiveness of Different Types of Surface Protection Materials Applied in Concrete Structures M. Q. Oliveira Escola Politécnica de Pernambuco Universidade de Pernambuco
[email protected]
S. M. S. Chaves Escola Politécnica de Pernambuco, Universidade de Pernambuco,
[email protected]
D. P. Morais Neto Escola Politécnica de Pernambuco, Universidade de Pernambuco,
[email protected]
C. S. T. Lima Escola Politécnica de Pernambuco, Universidade de Pernambuco,
[email protected]
E. C. B. Monteiro Universidade de Pernambuco and Universidade Católica de Pernambuco,
[email protected]
ABSTRACT
Nowadays one of the main problems encountered in concrete structures is corrosion, usually caused by the ingress of aggressive agents from the outside environment. This has led to the development of products designed to prevent the penetration of these agents, one of which is the surface treatment of concrete. Within this context, we tried to analyze three types of surface protection systems (water repellent, pore blockers and film formers), usually applied in reinforced concrete structures located in marine environments. Water absorption tests by immersion and by capillary action were used, and accelerated corrosion tests in order to compare the performance of materials. With this adopted procedure, clear advantages were found for using the water repellent as a surface product protection in the structures. It also showed that the water repellent performance was comparatively better than other materials.
KEYWORDS:
corrosion, water repellents, film formers, pore blocker, water absorption.
INTRODUCTION Numerous cases of reinforced concrete structures are often found to have structural, aesthetic or functional problems. The structures must not only have the capacity to withstand forces acting on them, but must also be able to withstand the action of aggressive agents in the environment, which have been their main causes of pathologies. - 3761 -
Vol. 20 [2015], Bund. 9
3762
Mehta apud Helene (1993) mentions the ruin of 27 buildings in England between 1974 and 1978, eight (30%) of which were the result of corrosion. Wayers apud Calçada (2004), provides 1991 data showing that 32% of the 574,671 American bridges had structural problems, in which USD 28 billion was spent on recovery from corrosion. Various researchers state that the main cause of corrosion is the penetration of chlorides (ANDRADE; WHITING (1996); MANGAT; MOLLOY (1992)). Calçada (2004) mentions a study by the UK National Audit Office that showed that 142 structures had corrosion problems, 20 of which were caused by chloride action resulting in 17% of the cases and cost to recover 22% of the total. Since Brazil has a vast coastline along which most of the population lives, it has serious problems due to the action of chloride ions from salt spray on its concrete structures. The penetration of chloride ions is the key cause of corrosion in the main Brazilian capitals. This finding encouraged the development of products designed to prevent the penetration of such agents, one of which is the surface treatment of concrete, which will be the subject of study of this dissertation. At the São Paulo Polytechnic School five PhD theses have already been written on topics relating to the subject herein: Figueiredo (1994), Repette (1997), Uemoto (1998) and Medeiros (2008). The assessment of the effectiveness of the surface treatment system proves to be very relevant due to the need to learn more about the functioning of this type of product in the different kinds of existing situations, thereby looking to fill the gaps left by the studies undertaken so far.
Objectives To analyze the comparative performance of three surface protection products, usually applied to recovery projects, and to assess their effectiveness and behavior under the action of chloride irons in accelerated corrosion tests.
MATERIALS AND METHODS Independent variables
In this study it was decided to consider as independent the variables listed below: Protection groups – surface water repellents, film formers and pore blockers – [benchmark (unprotected concrete) + 3 protection systems)];
●
State of the concrete substrate to be protected (contaminated and non-contaminated by chlorides) and;
●
●
Microstructure of the concrete – variation in the w/c ratio (0.4 and 0.7).
Dependent variables They are the variables influenced by the variation in independent variables and, in certain situations, by other dependent variables. Those in this study are as follows: ●
Corrosion potential (Ecorr);
Vol. 20 [2015], Bund. 9
3763
●
Water absorption by immersion;
●
Water absorption by capillary action;
Materials employed in the confection of test samples Hydraulic binding agent The hydraulic binding agent used to make the substrate of the test samples was CPII Z-32, cement consisting of pozzolana, chosen because it is the cement most frequently used in the region.
Fine aggregate
The fine aggregate used in the experiment was fine river sand, whose particle size analysis and physical properties are given below. Table 1 describes the main physical properties of the fine aggregate. Table 1 – Physical properties of fine aggregate COMPLETED TESTS Specific mass (g/cm3) Loose aggregate unit mass (g/cm3) Clay content in clumps (%) Powder content (%) Maximum aggregate size(mm) Fineness modulus
NBR 9776 (ABNT, 1987) NBR 7251 (ABNT, 1982) NBR 7218 (ABNT, 1987) NBR 7219 (ABNT, 1987) NBR 7217 (ABNT, 1987) NBR 7217 (ABNT, 1987)
2.59 1.59 8.31 0.57 1.18 2.28
The grading test was carried out as described in standard NBR 7217 (ABNT, 1987). The particle size composition of the fine aggregate fits into zone 2, considered a fine sand, according to the Brazilian standard NBR 7211 (ABNT, 1983).
Water In the experimental procedure drinking water was used supplied by the Pernambuco water and sewage company (COMPESA).
Steel The steel used was CA-60, obtained by wire rod drawing and produced as specified in NBR 7480 (ABNT, 1996). The diameter was 5.0 mm.
Vol. 20 [2015], Bund. 9
3764
Surface protection materials Three protection systems will be adopted, all available on the national market and designed for surface treatment of reinforced concrete structures. The choice of products was based on comparing samples of three categories of surface protection systems: surface water repellents, film formers and pore blockers. Table 2 provides the consumption and characteristics of the examined materials.
Table 2: Description of systems comprising the study Protection Manufacturer’s Consumption Number Type of Identification system description (l/m²) of coats curing Simple Water Silane0.25 2 Water repellent siloxane evaporation + reaction Simple Film former Water-based 0.20 1 Coalescence acrylic varnish Simple Pore blocker Liquid sodium 0.20 3 Reaction silicate
Density (g/cm³) 0.76
Drying time(h) 6
1.26
2
1.02
½-1
Definition of test series The experimental program was designed to compare the influence of the three types of surface protection systems for concrete surfaces, with two varying parameters: ● ●
Water/cement factor; Substrate: with/without chlorides.
Water/cement factors were chosen with significant variation: w/c = 0.4 and w/c = 0.7. The test samples were all in the same fresh conditions; that is, they were made within the same workability range: 260-320 mm, using the mortar consistency test in NBR 7215 (ABNT, 1991). Test samples with and without added chloride were made only for the test samples for the corrosion potential. The purpose of this procedure was to assess the behaviour of the surface protection products with regard to the starting periods and spread of corrosion in new structures (no chloride iron contamination) and old structures (contaminated by chloride ions).
Water Absorption Absorption by immersion This test used cylindrical mortar test samples, 50mm in diameter and 100mm in height, and they were cast according to standard NBR 7215 (ABNT, 1991). The test samples were dried in an oven at 100ºC until the mass was constant and they were kept in the laboratory for 24 hours to cool down. The protection systems were applied to all surfaces. For this test four test samples were used for each material under study. The absorption was measured by weighing the test
Vol. 20 [2015], Bund. 9
3765
samples after their surface was dry. The mass variation was accompanied for 30 days from the start of the immersion period.
Absorption by capillarity action Cylindrical mortar test samples were used in this test, 50mm in diameter and 100mm in height, and were cast according to standard NBR 7215 (ABNT, 1991). The test samples were dried in an oven at 100ºC until the mass was constant and they were kept in the laboratory for 24 hours to cool down. The protection systems were applied to the circular surface corresponding to the mold bottom. Four test samples were used for each material under analysis. Silicon protection was applied over 4cm in height from the surface where the surface protection was applied. The water level was kept constant at 5mm throughout the test. Monitoring the water absorption was accompanied by weighing the test samples throughout the experiment. The variation in mass of the test samples was accompanied for 30 days from the first contact with the water line.
Accelerated corrosion tests Test samples measuring 60x80x25mm were made, reinforced with two longitudinal steel rods 5.00mm in diameter, 100m in length and a 10mm cover as indicated in Figure 1. The water/cement factors were used with significant variation of 0.4 and 0.7. The rods were longer so that their end was exposed to enable the power connection of the reinforcement rods for corrosion monitoring. The rod cleaning procedure was based on standard ASTM C- 1152 (1992).
Figure 1: Vertical and horizontal section of test sample (Morais Neto, 2015).
Vol. 20 [2015], Bund. 9
3766
There are two series of test samples: one with a substrate previously contaminated by chlorides (1% in relation to the cement mass) and the other of concrete with no added chlorides. Both underwent drying and moistening half-cycles. The methodologies involving the drying and moistening cycles or contact with the NaCl solution correspond to the strictest regime of chloride penetration (PAGE et al 1991). The half-cycles are characterized in Table 3.
Table 3: Stages in the cyclical test
Half-cycle Drying Moistening
Condition Oven at 50°C Partly submersed in a 3% NaCl solution
Duration 5 days 2 days
The studied variables were measured at the end of each half-cycle. The measurements were taken during eight full cycles (drying + moistening).
Corrosion monitoring and assessment Corrosion potential (Ecorr) can be considered an indicator of deterioration and assess the thermodynamic aspects associated with corrosion. Standard ASTM C – 876 (1992) associates a higher probability than 95% of corrosion for potential values below –350 mV, for a copper/copper sulphate electrode (Table 4).
Table 4: Assessment criteria of corrosion potential measurements according to ASTM C876 (1992)
Corrosion potential in relation to benchmark copper/copper sulphate electrode - Cu/CuSO4 (mV) More positive than –200 More negative than –350 Between –200 and –350 DISCussion
Corrosion probability (%) 5 95 Uncertain
Water absorption
The water absorption criteria table of the Comité Euro-International du Béton - CEB – 192 below was used to analyze the results of this test.
Table 5: Assessment criteria of water absorption according to CEB – 192
Absorption(%) 5.0
Absorption of concrete Low Average` High
Quality of concrete Good Average Poor
Vol. 20 [2015], Bund. 9
3767
Water absorption by immersion Figures 2 and 3 refer to the results obtained for the benchmark test samples and the three products applied to the test sample surface (water repellent, film former and pore blocker).
Figure 2: Comparison of the results of absorption by immersion (%) for the benchmark, and water repellent, film former and pore blocker for the w/c factor 0.4. As can be seen from Figure 3 for the w/c factor 0.4 the film former product gave the best performance of all the products applied. The water repellent gave the worst performance compared to the other products at the end of the test. With the techniques used, it was possible to classify the materials studied in decreasing order of performance with regard to absorption by immersion for the w/c factor 0.4: Film former > Pore blocker > Water repellent
Vol. 20 [2015], Bund. 9
3768
Figure 3: Comparison of results of absorption by immersion (%) for the benchmark and the water repellent, film former and pore blocker for w/c factor 0.7. In the case of the 0.7 w/c factor, the product showing the best performance was the water repellent, which succeeded in retarding the high water absorption for the longest period, while the film former gave the worst performance of the three products used, and compared to the benchmark. By the techniques used, it was possible to classify the materials under analysis in decreasing order of performance with regard to absorption by immersion for the 0.7 w/c factor: Water repellent > Pore blocker > Film former
Water absorption by capillarity action Figures 4 and 5 refer to the results obtained for the benchmark test samples and the three products applied to the test sample surface (water repellent, film former and pore blocker). As can be seen for both w/c ratios analyzed herein, the water repellent material showed the best performance while the pore blocker was the product with the worst performance. With the techniques used, it was possible to classify the materials under study in decreasing order of performance regarding absorption by capillary action for both water/cement factors:
Vol. 20 [2015], Bund. 9
3769
Water repellent > Film former > Pore blocker
Figure 4: Comparison of the results of absorption by capillary action (g/cm²) for the benchmark and the water repellent, film former and pore blocker for the w/c factor 0.4.
Figure 5: Comparison of results of absorption by capillary action (g/cm²) for the benchmark and water repellent, film former and pore blocker for the 0.7 w/c factor.
Vol. 20 [2015], Bund. 9
3770
Corrosion Potential Test samples with no added chlorides Figures 6 and 7 refer to the results obtained for the benchmark test samples and the three products applied to the surface of the test samples (water repellent, film former and pore blocker). As can be seen from Figure 6 for the 0.4 w/c factor, the water repellent and film former products behaved similarly with regard to corrosion potential, with no occurrence of depassivation of its reinforcement rods until the end of monitoring. The pore blocker, however, had the worst performance, with depassivation of its reinforcement rods by the end of the 6th cycle. With the techniques used, it was possible to classify the materials under analysis in decreasing order of performance with regard to the corrosion potential for new structures and the 0.4 w/c factor: Water repellent > Film former > Pore blocker
Figure 6 - Comparison of the corrosion potential results (mV) for the benchmark test samples, water repellent, film former and pore blocker, 0.4 w/c factor, with no added chlorides (potential relating to the copper/copper sulphate electrode). For the 0.7 w/c factor, the product with the best performance was also the water repellent. The film former, which had a good performance for the 0.4 w/c factor, had a wider range of Ecorr values for the 0.7 w/c factor, and its reinforcements underwent depassivation by the end of the 7th cycle. The pore blocker also behaved similarly to the benchmark for the 0.7 w/c factor. With the techniques used, it was possible to classify the materials under analysis in decreasing order regarding the corrosion potential for new structures and the 0.7 w/c factor: Water repellent > Film former > Pore blocker
Vol. 20 [2015], Bund. 9
3771
Figure 7: Comparison of the corrosion potential results (mV) for the benchmark test samples, water repellent, film former and pore blocker, 0.7 w/c factor, with no added chlorides (potential relating to the copper/copper sulphate electrode).
Tests samples with added chlorides Figures 8 and 9 refer to the results obtained for the benchmark test samples and the three products applied to the surface of the test samples (water repellent, film former and pore blocker). As can be seen for the 0.4 w/c factor, the water repellent and film former products are more efficient with regard to restricting the water access, preventing the kinetics of reinforcement corrosion. The pore blocker had the worst performance of the three products. With the techniques used, it was possible to classify the materials under analysis in decreasing order of performance with regard to the corrosion potential for structures already contaminated by chlorides for the 0.4 w/c factor: Film former > Water repellent > Pore blocker
Vol. 20 [2015], Bund. 9
3772
Figure 8: Comparison of the corrosion potential results (mV) for the benchmark test samples, water repellent, film former and pore blocker, 0.4 w/c factor, with added chlorides (potential relating to the copper/copper sulphate electrode). For the 0.7 w/c factor, the product with the best performance was the water repellent, which prevented corrosion from happening during the 56-day test period. The film former only managed to retard the start of the corrosion process but did not prevent it from happening by the end of the 7th cycle, as can be seen in Figure 9. The pore blocker also had a poorer performance than the other materials for this w/c factor, showing Ecorr values below -350mV immediately at the start of the 1st cycle. With the use of the techniques, it was possible to classify the materials under analysis in decreasing order of performance regarding the corrosion potential for structures already contaminated by chlorides for the 0.7 w/c factor: Water repellent > Film former > Pore blocker
Vol. 20 [2015], Bund. 9
3773
Figure 9: Comparison of the corrosion potential results (mV) for the benchmark test samples, water repellent, film former and pore blocker, 0.7 w/c factor, with added chlorides (potential relating to the copper/copper sulphate electrode).
FINAL COMMENTS Considering the test conditions described herein we can conclude the following for the materials under analysis: • Reduction of the water/cement ratio helped improve the performance of the three types of
materials used with regard to water absorption and, consequently, the corrosion of reinforcement by chloride ions.
• The test samples made with the water repellent material had a comparatively better performance than those made with the other materials with regard to water absorption and, consequently, corrosion by chlorides, principally for the 0.4 water/cement ratio. • With the technique used, it was possible to classify the materials under analysis in decreasing order of performance with regard to chloride-contaminated environments:
Water repellent > Film former > Pore blocker
1. 2. 3.
REFERENCES
AMERICAN SOCIETY FOR TESTING AND MATERIALS. Standard test method for acid-soluble chloride in mortar and concrete. ASTM C 1152: 1992. Philadelphia: Annual Book of ASTM Standards. ____. Standard test method for half cell potential of uncoated reinforcing steel in concrete. ASTM C 876: 1992. Philadelphia: Annual Book of ASTM Standards. ANDRADE, C.; WHITING, D. A. Comparison of chloride ion diffusion coefficients derived from concentration gradients and non-steady state accelerated ionic migration. Materials and Structures, v. 29, p. 476 – 484, 1996.
Vol. 20 [2015], Bund. 9 4. 5. 6. 7. 8. 9. 10. 11. 12.
13. 14. 15. 16. 17. 18. 19. 20. 21.
3774
ASSOCIAÇÃO BRASILEIRA DE NORMAS TÉCNICAS. NBR 7211: Agregado para concreto. Rio de Janeiro, 1983. ______. NBR 7215: Cimento Portland- Determinação da resistência à compressão. Rio de Janeiro, 1991. ______. NBR 7217: Determinação da composição granulométrica. Rio de Janeiro. 1987. ______. NBR 7218: Agregados: determinação do teor de argila em torrões e materiais friáveis. Rio de Janeiro. 1987. ______. NBR 7219: Agregados: determinação do teor de materiais pulverulentos. Rio de Janeiro. 1987. ______. NBR 7251: Agregados em estado solto: determinação da massa unitária. Rio de Janeiro. 1982. ______. NBR 7480: Barras e fios de aço destinados a armadura para concreto armado - Rio de Janeiro. 1996. ______. NBR 9776: Agregados: determinação da massa específica de agregados miúdos por meio do frasco de Champman. Rio de Janeiro, 1987. CALÇADA, L.M.L. Estudo da Eficácia do Ensaio de Migração em Condições de Fluxo Não Estacionário na Previsão da Penetração de Cloretos em Concreto. Florianópolis, 2004. 214p. Thesis (PhD in Civil Engineering) – Post-graduation Civil Engineering Program, Federal University of Santa Catarina. FIGUEIREDO, E. J. P. (1994). Avaliação do desempenho de revestimentos para proteção da armadura contra corrosão através de técnicas eletroquímicas – contribuição ao estudo de reparo de estruturas de concreto armado. PhD thesis – Polytechnic School, University of São Paulo, São Paulo, 423p. HELENE, P.R.L. Contribuição ao estudo da corrosão em armaduras de concreto armado. 1993. 231p. Thesis (Professorship) – Polytechnic School, University of São Paulo. São Paulo. MANGAT, P.S.; MOLLOY, B. T. Factors influencing chloride-induced corrosion of reinforcement in concrete. Materials and Structures, v. 25, p. 404 – 411, 1992. MEDEIROS, M. H. F. de. Contribuição ao estudo da durabilidade de concretos com proteção superficial frente à ação de íons cloretos. Thesis (PhD), Polytechnic School, University of São Paulo – São Paulo, 2008. MORAIS NETO, D. P. Avaliação de Eficiência de Inibidores de Corrosão em Estruturas de Concreto Armado. 2015. Dissertação (Mestrado) - Polytechnic School, University of Pernambuco, Recife, 2015. REPETTE, W. L. Modelo de previsão de vida útil de revestimento de proteção da superfície do concreto em meios de elevada agressividade. 1997. 230f. Thesis (PhD in Civil Construction Engineering). Polytechnic School of São Paulo, São Paulo. UEMOTO, K. L. Influencia da formulação das tintas de base acrílica como barreira contra a penetração de agentes agressivos. 1998. 178p. Thesis (PhD) – Polytechnic School, University of São Paulo, São Paulo, 1998. WINSLOW, D. and LIU, D. The pore structure of paste in concrete. Cement and Concrete Research. 1990. Vol. 20, pp.227-235. PAGE, C. L., LAMBERT, P., & VASSIE, P. R. Investigations of reinforcement corrosion. 1. The pore electrolyte phase in chloride-contaminated concrete. Materials and Structures, 1991. Vol. 24, pp. 243252.
© 2015 ejge