Analysis of Effect of Protective Coatings on the Rate of Corrosion in Reinforced Concrete Sewers: Microscopic Study

in


Introduction
Due to microbial induced corrosion (MIC) in sewers, billions of dollars per annum losses have been incurred worldwide on sewerage system (Wei et al. 2013;Zhang et al. 2008a;E hewayde et al. 2007); more than $50 billion for rehabilitation purpose in Germany (Hewayde et al. 2006), almost $500 million for treatment of corroded and degraded pipes in Los Angeles (Apté et al. 2015) and approximately 10% of the total sewage treatment cost for treating MIC of sewers in Flanders, Belgium (Zhang et al. 2008). MIC is an immense problem in RC sewers infrastructure due to complexity of multistage processes (Sulikowski & Kozubal, 2016;Islander et al, 1992) that occurs externally (sulphate attack) and internally (sulphide corrosion) ( & Liu, 2017). Soil containing high sulphate content attacks the surface of concrete externally while sulphate-reducing bacteria, below the water line, acted on the sewage internally (Kuliczkowska, 2016;Ling et al. 2014), produce hydrogen sulphide (H2S) which enters into the moisture layer and converted into sulphuric acid (H2SO4) Parker, 1951) which attack the concrete matrix and continuously deteriorated the sewer internally (Y. Parker, 1947). Below is summarized sulphide corrosion phenomena is (Sharma et al., 2008;Wells et al. 2012): Organic matter + SO4 2-→H2S + CO2 (1) H2S + 2O2 →H2SO4 (2)

Experimental Program
RC sewer Samples were used in this research as well as sewage samples were also obtained from existing waste water streams to present the in-situ conditions. Sewer samples were placed in the sewage whose sulphate concentration was known, and then SEM and EDAX were performed to determine the durability of protective coating against MIC.

Materials
Sewer samples were procured from Public Health Engineering Department (PHED) Khyber Pakhtunkhwa (KPK), Pakistan, which were made according to ASTM C-76 standard. These sewer samples, having thickness of 1.5in and length of 8ft, were cut down into small pieces approximately 2-3ft each, as shown in fig.1, for testing and analysis purpose then these samples were divided into two categories; sample A and sample B. They were washed and stored in a dry, clean place.  Sewage was required to chemically attack the samples as in real conditions. Hence, it was collected from two different places; one sample was collected from Khwar, referred as "X" (locally used name for waste water stream) and the other was collected from Hayyatabad Canal located in industrial area, referred as "Y" as shown in fig.3.

Figure 3. Sewage samples X and Y respectively
Sewage samples were filtered through the simple filter paper to remove large suspended particles and then the sulphate content of the sewage samples was calculated using a mass spectrophotometer. The sulphate concentration of sewage sample X came out to be 50mg/L and for sewage sample Y, it came out to be 98mg/L. Sewage sample Y was used to achieve the maximum amount of chemical attacks in minimum possible time.

Coating and immersion in sewage
For the application of PEC on samples; initially RC sewer sample B was taken from the two prepared samples A and B, then sprayed with water to make the surface damp before PEC. The coating was done in two phases on sample B; in first phase, single layer of coating was applied all over the sample with the help of brush and allowed to dry for 6 hours. In second phase, after 6 hours, a second coat was applied all over the sample and allowed to dry as shown in fig

In-situ conditions
The samples prepared were subjected to chemical attack in a specially designed arrangement of plastic containers, containing sewage sample "Y", sealed with plastic covers so that an environment akin to the actual conditions can be provided. Plastic containers were used because of their inertness with the sewage. The duration of chemical attack was 30 and 60 days.

Sample formulation for microscopic study
After immersion in sewage sample "Y", sample was prepared for SEM as per the specification of ASTM E2809 (2015) and ASTM C1723 (2010) standard. The surface of samples was made smooth using grinding machine. To make the corners of the samples already placed on a conductive tape, a silver coating was applied all over the sides of the samples. To remove the excess charge, the samples were sputtered in a gold sputtering machine. After all these steps, the samples were ready to be tested under scanning electron microscope (Jana, 2006;Stutzman, 2001). The microscopic study; SEM AND EDAX, was conducted in two stages on both samples; A and B, simultaneously. Stage:1, sample A (uncoated) and sample B (coated) were placed in the plastic containers with half-filled sewage "Y" in it and chemically attacked for 30 days simultaneously. After 30 days, the microscopic study was carried out on both samples. Stage:2, sample A1 (uncoated sample A was coated after 30 days) and sample B (same sample as used in stage-1) were chemically attacked for more 30 days by placing in plastic container having sewage sample "Y". After 60 days, again the microscopic study was carried out on samples.

SEM after 30 days of first chemical exposure
SEM analysis was carried out using scanning electron microscope (JEOL, JED-2300, Japan). It can be observed from fig.5, that the crown portion of the uncoated sample A and coated sample B shows more formation of ettringite (Jiahui, Jianxin, & Jindong, 2006) as compared to portion of sewer submerged in the sewage. This is because the formation of Sulphuric acid (H₂SO₄) near the crown portion of the sewer is more. Also, the ettringite formation in the crown portion of sample A was much more as compared to the crown portion of the coated sample B as shown in fig.6 because the sample A was uncoated, and the concrete was exposed to the reaction taking place in the sewer.

EDAX after 30 Days of First Chemical Exposure
EDAX was performed simultaneously with the SEM. The graphs clearly show the peaks of precipitations of corrosive components; sulphur peak is more in the crown portion of sample A (uncoated) as compared to submerged portion because the sulphuric acid formation is more in the crown portion as shown in fig.7. The graphs of sample A show the highest peak of calcium (Ca) because sample is uncoated and calcium is the main component of concrete/cement. Also, the gold (Au) peaks were visible because the samples were spluttered in a gold sputtering machine. The sample B (coated) show relatively low peaks or negligible peaks of sulphur both for crown and submerged portions because of the coating as shown in fig.8. Also, the silicon peak was observed because it is the main constituent of epoxy coating.

SEM after 60 days
The crown portion of sample A1 showed more ettringite formation as compared to the submerged portion of the sewer as indicated by fig.9. Also, the crown portion of coated sample A1 showed more ettringite formation as compared to the crown portion of sample B (coated from the beginning; since 60 days) as shown in fig.10, because the sample A1 was initially left uncoated and the reaction already took place in first 30 days of immersion in sewage sample "Y".

EDAX after 60 days
After 60 days, the Sulphur peak of sample A1 is lower than that of sample A, showing the decrease in corrosion rate because of the epoxy coating as illustrated in fig.12. Visible peaks of gold (Au) and silicon (Si) are because of gold sputtering and epoxy coatings of the sample. The crown and submerged portion of sample B show low peaks of the sulphur because it was epoxy coated as shown in fig.12. The silicon (Si) and gold (Au) peaks appeared because of the epoxy coating and gold sputtering of sample.

Conclusion
Based on the results from SEM and EDAX analysis of RC sewer, the following conclusions can be drawn: • After 30 days of 1 st chemical attack, the sample A (uncoated) shows high peaks of corrosive products as compared to sample B (coated). • After 60 days of chemical attack, sample A1 shows less surface precipitation; Sulphur content as compared to sample A because of application of epoxy coating but greater corrosive products than sample B, showing that corrosion had already started in this sample during the initial 30 days when it was left uncoated. • The sample B has minimum rate of corrosion when compared to sample A and A1 at both stages; stage:1 and stage:2, because of the application of epoxy coating over it. • The epoxy coatings increase the service life of the sewers by enervating the corrosive environment moreover there is a likelihood that rehabilitated corroded sewers can be reused by applying epoxy coatings over it. • Because of the limited resources, the experiment was carried out using a standing sewage sample, but It is expected that the gravity flowing sewage, as in actual field conditions, increase the rate of corrosion as compared to the samples placed in standing sewage.