658
M. C. Bouleklab et al.
and the L/L 0 shows the maximum value near of 2.1·10
−3 at T ~ 770 °C (Fig. 12,
curve 1).
If small amount (0.5 vol. %) of the MWCNTs is added to the Fe–Cu sample, the
tendency of the L/L 0 (T) dependence is practically the same as for the described
before Fe–Cu sample, but the absolute L/L 0 values are higher for overall temperature range that was applied. It is also important to note, that this increase and its
temperature behavior are different for the above-mentioned temperature ranges. So,
the L/L 0 is on average higher than 0.5·10
−3 and its dependence is similar to that
for Fe–Cu sample in the range of 35–375 °C. Further temperature increase leads to
a sharp change of the L/L 0 value for the Fe–Cu-MWCNT(0.5) composite from
the value of ~0.6·10
−3 up to ~4.5·10
−3 at 600–630 °C. A further increase in temperature to 710–715 °C leads to a change in the L/L 0 value in the same range T,
as it was found for the Fe–Cu sample in this temperature range and that is =
1.3·10
−3 . Unlike the case of the Fe–Cu sample, the further increase in temperature in
the case of the Fe–Cu-MWCNT(0.5) sample leads to a decrease in the L/L 0 value
(Fig. 12, curve 2). As a result, the absolute maximum of the L/L 0 temperature
dependence occurs at a temperature near 716 °C and L/L 0 maximum is 5.85·10
−3 .
Consequently, we can state that the effect of the MWCNT additive is different for
different ranges of the temperature behavior of the relative linear expansion of the
Fe–Cu-MWCNT(0.5) composite. Namely, a) MWCNT only affect the value without
changing the character of the L/L 0 temperature dependence in the range 35–375
°C; b) the MWCNT addition highlights the presence of two parts of the L/L 0
dependence in the range 200–630 °C (it concerns 200–375 and 400–630 °C ranges)
which were hidden in the case of Fe–Cu composite; c) the plateau of the L/L 0 ,
which took place for the Fe–Cu sample for temperatures after ~720 °C, was changed
to the decrease of the L/L 0 when MWCNTs were introduced into the composite
(the sample Fe–Cu-MWCNT(0.5)).
Two peculiarities, in our opinion, should be emphasized when considering changes
in the dilatometric behavior of the Fe–Cu-MWCNT composites with a further
increase in the content of carbon nanotubes (please compare curves 1 and 2 in Fig. 12
for the Fe–Cu-MWCNT(0.5) and Fe–Cu-MWCNT(1.0) composites). It is easy to see
that the increase of the L/L 0 value is continued in the 35–375 °C range, whereas in
the temperature range 400–800 °C the L/L 0 values decrease markedly—approximately by 2.5 times. At the same time, we even have a decrease in the relative linear
expansion L/L 0 with increasing temperature in the range 400–630 °C. As a result,
the L/L 0 values and their temperature dependences in the range 650–800 °C are
close to ones measured for the samples without MWCNTs. The L/L 0 (T) riches the
peak (2.9·10
−3 ) at T ≈ 790 °C for the Fe–Cu-MWCNT(1.0) composite.
When the MWCNT concentration reaches 2%, the L/L 0 (T) values do not practically depend on temperature if to regard the L/L 0 (T) curve in the same scale as
other curves (see Fig. 12a). In fact, mentioned dependence is similar to measure
for the Fe–Cu-MWCNT(1.0) composites, but if it is observed in other scales (see
Fig. 1b). So, the maximal L/L 0 value for the Fe–Cu-MWCNT(2.0) is near to 1·10
−4
at T = 790 °C. Thus, the set of the maximal L/L 0 values is 2.1·10
−3 , 5.85·10
−3 ,
M. C. Bouleklab et al.
and the L/L 0 shows the maximum value near of 2.1·10
−3 at T ~ 770 °C (Fig. 12,
curve 1).
If small amount (0.5 vol. %) of the MWCNTs is added to the Fe–Cu sample, the
tendency of the L/L 0 (T) dependence is practically the same as for the described
before Fe–Cu sample, but the absolute L/L 0 values are higher for overall temperature range that was applied. It is also important to note, that this increase and its
temperature behavior are different for the above-mentioned temperature ranges. So,
the L/L 0 is on average higher than 0.5·10
−3 and its dependence is similar to that
for Fe–Cu sample in the range of 35–375 °C. Further temperature increase leads to
a sharp change of the L/L 0 value for the Fe–Cu-MWCNT(0.5) composite from
the value of ~0.6·10
−3 up to ~4.5·10
−3 at 600–630 °C. A further increase in temperature to 710–715 °C leads to a change in the L/L 0 value in the same range T,
as it was found for the Fe–Cu sample in this temperature range and that is =
1.3·10
−3 . Unlike the case of the Fe–Cu sample, the further increase in temperature in
the case of the Fe–Cu-MWCNT(0.5) sample leads to a decrease in the L/L 0 value
(Fig. 12, curve 2). As a result, the absolute maximum of the L/L 0 temperature
dependence occurs at a temperature near 716 °C and L/L 0 maximum is 5.85·10
−3 .
Consequently, we can state that the effect of the MWCNT additive is different for
different ranges of the temperature behavior of the relative linear expansion of the
Fe–Cu-MWCNT(0.5) composite. Namely, a) MWCNT only affect the value without
changing the character of the L/L 0 temperature dependence in the range 35–375
°C; b) the MWCNT addition highlights the presence of two parts of the L/L 0
dependence in the range 200–630 °C (it concerns 200–375 and 400–630 °C ranges)
which were hidden in the case of Fe–Cu composite; c) the plateau of the L/L 0 ,
which took place for the Fe–Cu sample for temperatures after ~720 °C, was changed
to the decrease of the L/L 0 when MWCNTs were introduced into the composite
(the sample Fe–Cu-MWCNT(0.5)).
Two peculiarities, in our opinion, should be emphasized when considering changes
in the dilatometric behavior of the Fe–Cu-MWCNT composites with a further
increase in the content of carbon nanotubes (please compare curves 1 and 2 in Fig. 12
for the Fe–Cu-MWCNT(0.5) and Fe–Cu-MWCNT(1.0) composites). It is easy to see
that the increase of the L/L 0 value is continued in the 35–375 °C range, whereas in
the temperature range 400–800 °C the L/L 0 values decrease markedly—approximately by 2.5 times. At the same time, we even have a decrease in the relative linear
expansion L/L 0 with increasing temperature in the range 400–630 °C. As a result,
the L/L 0 values and their temperature dependences in the range 650–800 °C are
close to ones measured for the samples without MWCNTs. The L/L 0 (T) riches the
peak (2.9·10
−3 ) at T ≈ 790 °C for the Fe–Cu-MWCNT(1.0) composite.
When the MWCNT concentration reaches 2%, the L/L 0 (T) values do not practically depend on temperature if to regard the L/L 0 (T) curve in the same scale as
other curves (see Fig. 12a). In fact, mentioned dependence is similar to measure
for the Fe–Cu-MWCNT(1.0) composites, but if it is observed in other scales (see
Fig. 1b). So, the maximal L/L 0 value for the Fe–Cu-MWCNT(2.0) is near to 1·10
−4
at T = 790 °C. Thus, the set of the maximal L/L 0 values is 2.1·10
−3 , 5.85·10
−3 ,