Materials and Methods

The concrete cover quality is the most effective protection and first defense against concrete corrosion. Thus, the study looked at the comparable durability of of eight different concrete mixes that are practical and cost effective alternatives in the construction of livestock buildings and manure storage structures. One of the mixes was Portland cement type 10 with water/cementing material (W/CM) ratio of 0.5, which is considered a commonly used mix in farm building structures. Common wisdom is to lower the W/CM ratio to increase durability. So the same cement type was used with a W/CM ratio of 0.4. However, lowering the W/CM ratio invariably means increasing the cement content of the mix thereby increasing the C3A content of the hardened cement paste and thus decreasing the sulphate resistance.

On order to reduce the Portland cement content six other mixes with various supplementary cementing materials, like slag, fly ash and silica fume, were included. The use of supplementary cementing materials to replace a portion of the Portland cement in the mix contributes to the properties of the hardened concrete through hydraulic or pozzolanic activity. It reduces the concrete permeability, decreases the cement content, and the W/CM ratio. It also, in general, improves the resistance of the concrete to sulphate attack by lowering the C3A content of the hardened cement paste.

Concrete used for the construction of liquid manure handling and storage structures is subjected to sulphides, sulphates and sulphuric acid. In some locations the concrete is submerged continuously, in other places it is submerged some of the time and some locations are never submerged. Field observations indicate that the latter two situations lead to more severe corrosion than the totally submerged condition.

To reach the objectives of the research a combination of laboratory research and numerical simulation was selected. The laboratory research consisted of accelerated corrosion testing of concrete specimens by subjecting them to hydrogen sulphide gas and a sulphate solution in high concentrations. One half of the specimens was partially immersed in sodium sulphate (20,000 ppm SO42-) and also subjected to hydrogen sulphide gas (1,000 ppm H2S). This condition was chosen to simulate units of construction that are partially submerged in liquid manure. The second set was subjected to hydrogen sulphide gas only, a condition that may occur for concrete permanently above the manure. Each set consisted of the eight different mixes. The sulphuric acid corrosion was measured on those six mixes not containing slag by applying 2 cm3 of 7% sulphuric acid three times per week to disc-shaped specimens.

The corrosion of the concrete and of the embedded reinforcing steel was measured separately. For the purpose of the embedded steel corrosion study, all treatments were tested with the half-cell potential between the concrete surface and the reinforcing steel to define the corrosion state. The reinforcing steel was visually inspected at three stages of the sulphide/sulphate experiment.

For the purpose of the study on concrete corrosion due to sulphate and/or sulphide, the compressive strength of all mixes was measured at 28 days. The sulphate solutions, in which the mixes were partially immersed, were analyzed and their pH’s measured to follow the leaching of the alkalis from the concrete paste and the effect of H2S gas on the solution acidity and SO42- concentration. The actual corrosion damage to the concrete was measured using the volume loss of the concrete at the end of the test period. The corrosion of concrete is either caused by a direct chemical reaction of the corrosive agents with the concrete at its surface, or by reactions below the surface after diffusion of the corrosive gases or liquids into the concrete. To try to measure the extent of the latter the total sulphur profile was determined across the concrete cover thickness. Finally, the mineralogy of the concrete was studied using powder x-Ray test to confirm the nature of the chemical reactions that took place and to confirm that these were similar to those that happened in a field situation where a pig barn collapsed occurred due to corrosion of the concrete structure.

The concrete corrosion from exposure to sulphuric acid was measured by weight loss. That aspect of the study is in progress at the time of writing.

Liquid swine manure has sulphate concentrations in the order of 1500 to 2000 ppm. Table 1 lists the concrete requirements in Canadian concrete standards for four degrees of exposure to sulphate (Kosmatka et al. 2002).

Table 1. Requirements for concrete subjected to sulphate attack

Degree of exposure

Water-soluble sulphate in soil (%)

Sulphate in groundwater (mg/L)

Max. water­cementing materials ratio

Portland cement type to be used

Very severe

Over 2.0

Over 10,000

0.40

Type 501 plus a pozzolanic admixture

Severe

0.20 to 2.0

1500 to 10,000

0.45

Type 50

Moderate

0.10 to 0.20

150 to 1500

0.50

Type 202

Negligible

Below 0.10

Below 150

No restriction

No restriction

1 CSA A5 (eq. to ASTM C1157 type HS) 2 CSA A5 (eq. to ASTM C1157 type MS)

The replacement of 35% of type 10 cement by slag is considered sufficient to provide Type 50 equivalence. The relative improvement in sulphate resistance through the use of fly ash is greater for low cement content mixes and in high sulphate exposure. Class F fly ash is effective in improving the sulphate resistance of concrete when used at suitable replacement rates. The higher the calcium content of a Class C fly ash, the less likely it is that it will provide benefits with regard to sulphate resistance (A Publication of Lafarge Canada Inc.). Unfortunately, in Ontario the available fly ash is type C.

Another effective mix is the use of silica fume. Due to reduced permeability, silica fume cement provides excellent sulphate resistance to concrete. Recent research shows that ternary blends containing slag cement or fly ash, along with silica fume and Portland cement, can be effective for sulphate resistance concrete (A Publication of Lafarge Canada Inc.).

It is intended that the objectives of the research be attained by the combination of laboratory research and numerical simulation. The corrosion testing in the laboratory are accelerated tests using hydrogen sulphide gas, and a sodium sulphate solution as the only corrosive agents. Acceleration is achieved by using much higher concentration than those experienced in the field, 1,000 ppm H2S and 20,000 ppm SO42-.

Eight mixes, which could reasonably be used in the construction of liquid manure tanks, floors and slats, were examined for corrosion resistance. Concrete cylinders (100 mm diameter by 100 mm high) made of Portland cement, limestone, sand and water, each with a 10 mm diameter by 90 mm long reinforcing steel bar embedded in the center. Six replicates for each mix were made. In all mixes a superplasticizer (CATEXOL 1000 SP-MN) is used (625 ml/100 kg of cementitious material) to reduce the water requirements in concrete and attain the necessary workability without the use of excess water. Also an air-entraining admixture (CATEXOL A. E.260) was used in all mixes (50 ml/100 kg of cementitious material) to increase concrete durability, improve workability and reduce bleeding. A further five larger replicates of each mix, 100 mm diameter and 200 mm height, were cast without a steel bar for compressive strength determination at 28 days (three replicates) and for the sulphuric acid corrosion tests. The mix proportions and materials used for all eight mixes are provided in Table 2. The coarse aggregate was crushed limestone.

Table 2. Mix proportions for all eight mixes

PC50

PC40

SR

SC

SFC

FAC

SSFC

FASF

PC

with

W/CM

ratio

0.5

PC

with

W/CM

ratio

0.4

Sulphate

resisting

cement

Slag

cement

Silica

fume

cement

Fly ash cement

Silica fume & slag cement

Silica fume & fly ash cement

Cement type

10

10

50

10

10

10

10

10

Cement

(kg/m3 of concrete)

340

425

425

276

391

319

293

293

Water

(kg/m3 of concrete)

170

170

170

170

170

170

170

170

W/CM ratio

0.5

0.4

0.4

0.4

0.4

0.4

0.4

0.4

Additive (% of

cementitious

material

content)

35%

8%

silica

fume

25%

25% slag 6%

25% fly ash 6%

slag

fly ash

silica

fume

silica

fume

The top of all specimens, except those for the compressive strength tests, and the exposed ends of the steel bars were coated with an epoxy coating (TRU-GLAZE 4508 Chemical Resistant Epoxy Coating 4508-1000A and 4508-9999B with a ratio of 1:1), in order to prevent the diffusion of the corrosive ions through that surface.

For the sulphide/sulphate corrosion study the 48 specimens from 6 replicates were divided into two sets, each set had three replicates. The first set (i. e. 24 specimens) was tested partially (50%) immersed in sodium sulphate (20,000 ppm SO4-2) and at the same time subjected to hydrogen sulphide gas (1,000 ppm H2S) above the surface of the sodium sulphate solution. Each treatment was submerged in a separate container, and placed in the upper level of a two-storey test chamber. The second set (i. e. the remaining 24 specimens) was subjected only to the hydrogen sulphide gas, nitrogen and air (0.1% H2S, 9.9% N2, and 90% air) in the lower level of the same chamber.

To keep the gas in the sealed Plexiglas test chamber at the required concentration a control circuit consisting of H2S sensor, solenoid valves, flow meter, control program and a gas cylinder (1% H2S, 99% N2) was used. The hydrogen sulphide gas cylinder lasted about 3 weeks, keeping the concentration of the gas inside the chamber at 1000 ppm. After this the chamber was left closed for one more week in order to lower the H2S concentration back to 0 ppm. Then all specimens were taken out of the chamber for the half-cell potential measurement, and thus they were subjected to air for another week. This test cycle was repeated approximately every five weeks.

The specimens for the sulphuric acid tests were 25 mm thick discs cut from the 100 mm diameter by 200 mm high unreinforced cylinders described earlier. Six mixes that did not include fly ash were included in these tests. The reason for excluding the fly ash containing mixes was that the type C fly ash that was used is not a viable solution for corrosion resistance in an acidic environment.

Standard 28-day compressive strength test were carried out for the 100 mm diameter by 200 mm high cylindrical specimens. The splitting tensile strength test (CSA A23.2-13C 1994) was carried out after 11, 15 and 23 cycles of exposure to the corrosive environment. This allowed the removal of the reinforcing bar for inspection of corrosion and provided a rough indication of the specimen’s strength about 26, 32 and 41 months after the specimens were cast and after the exposure to a corrosive environment for most of that period.

Materials and Methods
Подпись: Figure 1 Half-cell potential measurement

The half-cell potential, Ecorr, was used to define the corrosion state of reinforcement bars. A copper-copper sulphate electrode (CSE) and a high impedance voltmeter were used to read Ecorr. The half-cell potential measurement connection is shown in Fig. 1. The first half-cell potential measurements were taken after three cycles of exposure to the corrosive environment. Measurements were made every test cycle of about 5 weeks thereafter until the end of the experiment. Six potential readings were taken every 20 mm (at the air end, 20, 40, 60, 80, and 100 mm) along three meridians at 0o, 120o, and 240o, for a total of 18 readings per specimen (Fig. 2). The 18 readings were averaged. Corrosion potential measurements provide an indication of the oxidizing power of the environment in which a specimen is exposed.

At the end of the test period (after the 23rd test cycle), specimens were split open and concrete cover was removed to inspect the steel in the specimens. A visual assessment of the corrosion of reinforcing steel was performed. The corrosion was rated as three corrosion levels. Level 1 represents very slight or no visible rust on the surface. Level 2 shows a little rust at the surface. Level 3 indicates relatively heavy corrosion.

After the exposure of concrete specimens to the sulphate solution and/or sulphide gas, for the study period, the last set of specimens were air dried and brushed to remove all loose material. The volume loss of each specimen was determined using the water displacement method.

X-Ray powder diffraction patterns provided information on the phase, chemical and crystal structure. For the XRD test, 50 mg samples of the concrete paste were obtained from the outer surface of each mix and were well-ground to a uniform particle size to below 10 pm. The test was carried out for three concrete mixes: PC40, SR, and SFC. The samples were taken from the last set of replicates for both corrosive environment exposures, a total of six samples. For the ones that were submerged the samples were taken from above the solution level. Also, samples were taken from a barn that collapsed in Innerkip in 2001, reportedly as a result of corrosion of the supporting piers. One sample was taken from the surface of a beam that supported a solid slab over the manure pit. Another sample was obtained from the outer surface of one of the piers that caused the failure.

The sulphuric acid corrosion tets were carried out on 25 mm thick by 100 mm diameter discs positioned on the flat. The present tests use discs that have the top surface of the cylinders from which they were cut on top, the cut surface on the bottom. Three times per week 2 cm3 of 7% (by volume) H2SO4 is dripped slowly on the surface of the discs. After each 10 applications (about 3 weeks) the discs are washed, scrubbed to remove loose material, dried in an oven for one week, and weighed. The weight losses are used as a comparative measure of corrosion.