Page 73 - Zmist-n4-2015
P. 73

of the wires are shown in the Table. The hot rolled bar was marked E0 and the follo-
                  wing steels as E1…E7 according to a cold drawing step.
                                                                                   p
                       Diameter D of the steels E and accumulated plastic deformation level e ee accum
                                                                                  e
                       Steel       E0      E1     E2     E3      E4     E5     E6      E7

                       D, mm      11.03   9.90   8.95    8.21   7.49   6.80    6.26   5.04
                        p
                       e accum     0.00   0.22   0.42    0.59   0.78   0.97    1.13   1.57
                                                                                 –1
                                                                              –7
                      Mechanical investigations consisted in slow strain rate testing (10  s ) in air and
                  in a model environment using smooth cylindrical specimens with diameters equal to
                  wires thickness and with length 300 mm. Surface of the tested wires was not grinded
                  but only degreased by acetone and washed with water to approach the real working
                  conditions. Specimens were tested on the MTS Alliance RT/100 testing machine with
                  software TESTWORKS 4. The initial distance between grips was 220 mm.
                      For the study of hydrogenation effect on the mechanical behaviour of the steel an
                  electrochemical cell of 8 mm height was fixed around a specimen. In this electrochemi-
                  cal three-electrode scheme a tested wire (working electrode) was connected to a poten-
                  tiostat  by  its  negative  pole  and  served  as  cathode.  The  platinum  spiral  as  a  counter
                  electrode was used for polarization providing uniform distribution of current along the
                  specimen surface. Constant cathodic potential –1.2 V was maintained by the potentio-
                  stat AMEL VOLTALAB PGP 201. Reference electrode was saturated calomel – SCE
                  (Hg|Hg 2Cl 2). Tests were performed in the solution containing 1 g/l Ca(OH) 2 + 0.1 g/l
                  NaCl (pH 12.5) with free oxygen access modelling a pore solution in concrete [6, 11].
                  At least three specimens were tested in air and for each “metal–environment” system.
                      The object of the analysis was the true stress– true strain curves σ–e and reduction
                  in area (RA), y. Curves in air were recorded using an extensometer and presented up to
                  the moment of reaching the ultimate tensile strength σ UTS (the stage of uniform elon-
                  gation).  For the tests in hydrogenating medium  the whole tensile curves are shown.
                  Percentage of RA was calculated after fracture of the specimens. The commercial wire
                  was not taken into consideration because of its thermal treatment after cold drawing to
                  remove residual stresses, which modified its plasticity characteristics. It did not allow
                  the comparison of the final stage of cold drawing with the previous ones. Macrofrac-
                  ture maps were obtained using scanning electron microscope JEOL JSM-5610 LV for
                  the identification of characteristic fracture zones, namely, crack initiation, subcritical
                  crack growth and final fracture area.
                      Results  and  discussion.  Uniform  elongation  e u  of  the  specimens  tested  in  air
                  decreased sequentially with cold drawing degree with improving the strength characte-
                  ristics (Fig. 1, curves 0–6). Such mechanical behaviour corresponds to conventional
                  notion about strain hardening of materials. Concerning the tests with cathodic polariza-
                  tion (Fig. 1, curves 0¢–6¢), it should be noted that no visible transformations were fixed
                  related to subcritical crack growth, because it could be reflected in the curves shape. It
                  can be explained by a very low strain rate. In this case even if the stage of crack propa-
                  gation is prolonged it could be visible in the negligible increment of ε on the stress–
                  strain diagram. Hydrogenated material revealed divergent behaviour: firstly the para-
                  meter e increased with cold drawing reaching the maximum value for the steel E3 and
                  then it reduced. The possible explanation will be done later involving another plasticity
                  parameter, reduction in area.
                      Reduction in area for the test in air, in contrast to relative elongation, is nonmono-
                                                          p
                  tonic  function  of  accumulated  plastic  strain  e accum  exhibiting  maximum  at  the  later
                  stages of cold drawing (Fig. 2, curve 1). It means that two plasticity parameters, e u and

                  72
   68   69   70   71   72   73   74   75   76   77   78