Mechanical Properties of Sprayed SFRC

2. Uniaxial Tension Test

2.1 Apparatus, specimen and test method

The main specifications of the servo-controlled testing machine employed for uniaxial tension tests are as follows: (i) capacity: 500kN, (ii)frame stiffness: 1GN/m. Platen displacement and force are, respectively, measured by a LVDT and a strain gauge type of load cell.
Fig.5 shows a schematic diagram of the experimental assembly used in uniaxial tensile testing, which was prepared as follows. After liberally coating the top end of the specimen with epoxy resin, it is affixed to the surface of the upper platen and allowed to set for 24 hr. The platen is then screwed into the ram head and the specimen's bottom end is correspondingly affixed to lower platen, after which about 1kN load is placed on the specimen for 24 hr prior to tension testing. The specimen is able to accurately self-align itself by the load before the epoxy resin becomes hard.

The specimen was shaped cylindrical with the size of ƒÓ50~100 mm, top and bottom faces were ground parallel to each other(}0.01mm). All specimens were laboratory air-dried (room temperature 20}3Ž,relative humidity 70}15% being kept) for a minimum of 30 days prior to testing. Considering anisotropy of SFRC caused by steel fiber's orientations, two types of specimens, see Fig.6, were made boring parallel and normal to sprayed plane.

The testing was performed under a control of constant strain-rate set to 10-6s-1. Two pieces of cross strain gauge (gauge length 30 mm) stuck on the specimen oppositely were used to detect the lengthwise and lateral strain of pre-failure region respectively. More than 3 peaces of the same type specimens were tested under the same conditions. The ages of SFRC were 7-10 months during the testing.

2.2 Complete stress-strain curves

The result of the uniaxial tensile testing is shown in Table 3. We found that the tensile strength of type a is higher than type b. This is considered that the week layers are easy to be formed in sprayed SFRC parallel to a sprayed plane. Both Young's modulus and Poisson's ratio are close to each other. That is because Young's modulus and Poisson's ratio were taken at low stress level, say, just up to 30% of peak strength.

Fig.7 shows the complete stress-strain curves in the uniaxial tensile testing. It can be easy intuitively found through both type a and type b that there are no obvious variation caused by mix of steel fibers in the pre-failure region in, for example, tensile strength, Young's modulus and Poisson's ratio. So we can say the behaviors of SFRC prior to the strength failure point is mainly supported by the matrix of concrete. In specimens of type b, almost no evident difference between plain concrete and SFRC can be check out even in the post-failure region whatever how many steel fiber contents. That says no effective steel fiber effort is at work in the direction normal to the sprayed plane if subjected to a tensile stress. It is, however, quite different in specimens of type a. The curves become more ductile in the post-failure region when the more steel fibers contents have been mixed in its matrix.

2.3 The relationship between residual strength and content of steel fiber

By seeing fig.7(a), two facts about residual strength can be noticed: (1) The value of residual strength at the same strain seems to be in proportion to the steel fiber volume fraction in the matrix of concrete. (2) The decrease of residual strength together with growing of strain is very slow. The facts suggest that the cause of residual strength is closely related to the mixture of steel fibers. A direct proof for supporting this point is come from an observation of specimen's fractured surface caused by uniaxial tensile stress, as shown in Fig.8. First let us see the specimen of type b: no steel fiber has been found to be pulled out from matrix body, subsequently no residual strength appeared in the stress-strain curve. When we observe the specimen of type a, it is found that all peaces of steel fiber, which stand on the surface in any angles, are pulled out from specimen's body. No fiber was torn off. The more fibers appear on the surface, the higher residual strength is.

Residual strengths ƒÐRE vs. steel fiber volume fraction Vf were plotted in Fig.11. Good linear relationships are shown in any strain from 1.0% to 5.0%. The follow equation is suggested to express the relationships:
ƒÐRE = a(Y)¥Vf

where a(Y), as displacement Y's function, is a proportional parameter of the bond relationship between steel fiber and matrix. An experimental expression of a(Y) is chosen to be a(Y)=5~10-6E(12.5-Y)5 in this testing. With this equation, one can estimate the residual strength of the SFRC which has any volume fraction of steel fiber by doing a test with only one kind of SFRC.