Influence of the Carbon Allotropes on Dilatometric Properties …
671
at ~1160 cm
−1 is a confirmation of the MWCNT destruction. However, the main
confirmation of the last fact is the data on the D and G bands characteristics. We see
that these characteristics are close to magnitude for the Fe–Cu-MWCNT(0.5) and
Fe–Cu-MWCNT(2.0) nanocomposites. As for their values (half-width in the range
of 104–204 cm
−1 , and the ratio of I D /I G (Integral)—within the range of 0.86–1.11),
they indicate a significant defect and destruction of the MWCNT. We used integral
intensity values here since the D and G bands for our samples are a superposition of
several components (see below), and therefore, in our opinion, in this case, we should
operate precisely with the integral values of intensity. Indeed, the macro-Raman
spectra, obviously, reflect the state of MWCNT and Fe–Cu matrix in areas with a
cross section up to ~500 microns. This is quite larger in comparison to the size of
metal grains and MWCNT sizes. Consequently, the analyzed macro-Raman data are
a certain averaging. The temperature behavior of the characteristics of the sample is
the result of a set of processes at the micro- and nanoscale levels. In order to highlight
this issue, we conducted a Raman scattering study on a microlevel, when the Raman
scattering cross section was only ~1 μm. This type of measurement was performed
for different areas of the sample under study. The typical micro-Raman spectroscopy
data for the areas (points) p1(0.5), p2(0.5), and p3(0.5) (Fe–Cu-CNT(0.5) sample)
and for the areas (points) p1(2.0), p2(2.0), p3(2.0), and p4(2.0) (Fe–Cu-CNT(2.0)
sample) are shown in Figs. 18 and 19. Quantitative data from this study are presented
in the Table 2. The data on micro-Raman spectroscopy of the initial MWCNT are
shown on the same Figures and given in Table 2, for comparison.
The main result of these measurements is the conclusion that there is significant
heterogeneity in the distribution of MWCNTs on the surface of the samples. This
follows, firstly, with a significant Raman intensity difference of the nanotubes spectra
measured at different areas (points) of the samples, as well as the different ratio of the
intensity of the spectrum associated with MWCNT (spectral range 900–1800 cm
−1 )
and the wing of Rayleigh scattering (range 100–250 cm
−1 ) due to the reflection of
light from the metal components of the sample. It is obvious, that the higher values of
this ratio correspond to those sample areas where the carbon material content is higher
(see curves p1(0.5), p3(2.0), and p4(2.0) in Fig. 18). In addition, the composition
of the bands, their shape, the values of width and position of the maxima are also
evidence of the diversity of carbon forms in the investigated composites. The detailed
view and characteristics of the breakdown for the Gaussian components of bands from
the MWCNT-scattering spectral range are the illustration of this statement (Fig. 19
and Table 2).
Comparing the spectra of nanocomposites with the spectrum of initial carbon
nanotubes (Fig. 10, MWCNT curve), we see that there are areas of composites
(p1(0.5) and p4 (2.0)), where the spectra are quite similar to the MWCNT spectrum.
In these areas, as noted above, the contents of the tubes are higher. However, in the
case of nanocomposites, the D and G bands have considerably larger half-widths
(especially the band D) and, unlike the MWCNT spectrum, they lie on the background of a very wide scattering band (band W in Table 2). In addition, the G band
for these areas of the composites is shifted toward higher frequencies by ~10 cm
−1 .
671
at ~1160 cm
−1 is a confirmation of the MWCNT destruction. However, the main
confirmation of the last fact is the data on the D and G bands characteristics. We see
that these characteristics are close to magnitude for the Fe–Cu-MWCNT(0.5) and
Fe–Cu-MWCNT(2.0) nanocomposites. As for their values (half-width in the range
of 104–204 cm
−1 , and the ratio of I D /I G (Integral)—within the range of 0.86–1.11),
they indicate a significant defect and destruction of the MWCNT. We used integral
intensity values here since the D and G bands for our samples are a superposition of
several components (see below), and therefore, in our opinion, in this case, we should
operate precisely with the integral values of intensity. Indeed, the macro-Raman
spectra, obviously, reflect the state of MWCNT and Fe–Cu matrix in areas with a
cross section up to ~500 microns. This is quite larger in comparison to the size of
metal grains and MWCNT sizes. Consequently, the analyzed macro-Raman data are
a certain averaging. The temperature behavior of the characteristics of the sample is
the result of a set of processes at the micro- and nanoscale levels. In order to highlight
this issue, we conducted a Raman scattering study on a microlevel, when the Raman
scattering cross section was only ~1 μm. This type of measurement was performed
for different areas of the sample under study. The typical micro-Raman spectroscopy
data for the areas (points) p1(0.5), p2(0.5), and p3(0.5) (Fe–Cu-CNT(0.5) sample)
and for the areas (points) p1(2.0), p2(2.0), p3(2.0), and p4(2.0) (Fe–Cu-CNT(2.0)
sample) are shown in Figs. 18 and 19. Quantitative data from this study are presented
in the Table 2. The data on micro-Raman spectroscopy of the initial MWCNT are
shown on the same Figures and given in Table 2, for comparison.
The main result of these measurements is the conclusion that there is significant
heterogeneity in the distribution of MWCNTs on the surface of the samples. This
follows, firstly, with a significant Raman intensity difference of the nanotubes spectra
measured at different areas (points) of the samples, as well as the different ratio of the
intensity of the spectrum associated with MWCNT (spectral range 900–1800 cm
−1 )
and the wing of Rayleigh scattering (range 100–250 cm
−1 ) due to the reflection of
light from the metal components of the sample. It is obvious, that the higher values of
this ratio correspond to those sample areas where the carbon material content is higher
(see curves p1(0.5), p3(2.0), and p4(2.0) in Fig. 18). In addition, the composition
of the bands, their shape, the values of width and position of the maxima are also
evidence of the diversity of carbon forms in the investigated composites. The detailed
view and characteristics of the breakdown for the Gaussian components of bands from
the MWCNT-scattering spectral range are the illustration of this statement (Fig. 19
and Table 2).
Comparing the spectra of nanocomposites with the spectrum of initial carbon
nanotubes (Fig. 10, MWCNT curve), we see that there are areas of composites
(p1(0.5) and p4 (2.0)), where the spectra are quite similar to the MWCNT spectrum.
In these areas, as noted above, the contents of the tubes are higher. However, in the
case of nanocomposites, the D and G bands have considerably larger half-widths
(especially the band D) and, unlike the MWCNT spectrum, they lie on the background of a very wide scattering band (band W in Table 2). In addition, the G band
for these areas of the composites is shifted toward higher frequencies by ~10 cm
−1 .