Results and discussion
4.7 TGA results
It is known that the oxidation stability correlates with the combing effect of defects and the diameter of the nanotubes [Zhang 2004-149, Xu 2007-8945, Liu 2007-5006, BOM 2002-615]
.
The presence of defects in graphite contributes to a decrease in the oxidative stability of the material [BOM2002-615]
. In addition, the larger the diameter is, the higher the oxidation stability of the nanotubes is [Zhang 2004-149, BOM 2002-615]
. Theoretically, SEM and TEM can not observe the functional groups nor precisely predict the formation of free radical bonds if the ―very small‖
defects are formed in the chemical structure of the surface. Because defects and derivatization moieties in nanotube walls result in lower structural integrity and can lower the thermal stability [Pillai 2007-3011, Liu 2007-5006]
, it is useful to confirm the results by using TGA.
(a) (b)
Fig. 4-18 TGA curves for the acid-treated MWCNTs under different treatment times, (a) 6 h,
and (b) 9 h (for Specimens A1 and A2, respectively).As revealed in Fig. 4-18, the mainly derivative peak temperatures of the MWCNTs treated by nitric/sulfuric acid process for 6 and 9 h are shown to be 635 and 638 oC, which are much higher than that of the as-purchased MWCNTs (600 oC). As supported by the results of SEM, it is suggested that the increase of the decomposition temperature may be dominated by
63
the significant diameter distribution change. Meanwhile, it is observed that there are satellite peaks presented at ~ 250 and ~ 350 oC in weight derivative TGA curves. The presence of the satellite peak of the derivative curve at ~ 350 oC might have resulted from additional carbonaceous impurities which are introduced by the destruction of nanotubes during acid treatments [Zhang 2007-149, Hu 2003-13838, Harutyunyan 2002-8671, Shi-1999-35]
. Additionally, the peak below 300 oC is typical of residual acid and absorbed water in the sample after treatment [Landi
2005-6819]
. The presence of the satellite peaks are supporting the fact that the structural damage are severe when the nitric/sulfuric acid are used to facially modify the MWCNTs and this matches the SEM characterization results in Fig. 4-1. Meanwhile, as revealed in the figures, the TGA curve depicts that the carbon impurities introduced weight approximately 20% in 6 h acid-treated sample and 30% in 9 h acid-treated sample. The weight percentages of the catalysts are also found to be increased to 7% and 13% for the MWCNTS treated by acid for 6 and 9 h respectively (for Specimen A1 and A2 respectively). This shows that the structural damage increases as the acid treatment time increases. Therefore, the graphite structure are damaged and removed by filter during cleaning process. So the share percentage of catalyst is thus raised up.
Fig. 4-19 TGA curves for the ECR ion-treated MWCNTs under different H
2/O2 gas flow ratio.Figure 4-19 shows the TGA curves of the MWCNTs treated by the ion treatment for 5 min with various H2/O2 gas flow ratios. It is found that all curves of the ion-treated MWCNTs
64
are almost overlapping on the curve of the as-purchased MWCNTS only the MWCNTS treated by H2/O2 gas flow ratio of 50/0 (sccm/sccm) are slightly deviated from that of the as-purchased nanotubes because a very few percentage of the MWCNTS treated by H2/O2 gas flow ratio of 50/0 (sccm/sccm) are decomposed from temperature at 300 oC. However, the deviation is insignificant.
Fig. 4-20 Weight-derivative TGA curves for the 5 min ion-treated MWCNTs under different
H2/O2 flow ratios (for Specimens B1 to B5).Figure 4-20 shows the weight-derivative curves of TGA analysis on the as-purchased MWCNTs and the 5 min ion-treated MWCNTs. The results show that with a main decomposition temperature of 600 oC, the as-purchased MWCNT samples are the most thermally stable with respect to oxidative degradation. Correspondingly, the MWCNTs treated by the ion treatment with a gas composition of 40/10 (sccm/sccm) have the lowest decomposition temperature (594 oC). Because the oxidation stability is a function of the combined effect of defects and the diameter of the nanotubes, with the same diameter distribution observed by the SEM characterizations in Fig. 4-3, the results are in agreement with the hypothesis that the decrease of the oxidation reaction temperature is mainly a result of the free radical bonds produced by the ion treatment. The marginal differences of the main decomposition temperature between as-purchased and ion-treated samples reflect the fact that
65
the effect of gas composition on structural integrity of the nanotubes is negligible in this case although the insignificant difference are varied with H2/O2 gas flow ratio.
Fig. 4-21 TGA curves for 0.25M HNO
3 acid-treated MWCNTsFigure 4-21 reveals the TGA results of the MWCNTs are only treated by 0.25 M HNO3
for 2 h. It is found that after the dilute acid treatment, carbon impurities including absorbed water are of approximately 15% in weight and the catalysts weight approximately 7% in the dilute acid-treated MWCNTs sample. Meanwhile, the main decomposition temperature peaks at 684 oC. This shows that this treatment is so moderate that the MWCNTS can be with high structural integrity and low free radical bonds introduced (ID/IG = 0.93). This is also supported by SEM characterization (Fig. 4-2) in which the morphology change is insignificant after the dilute acid treatment while the diameter is slightly enlarged.
In order to evaluate the effects of the ion pretreatment on the decomposition temperature of the dilute acid-treated MWCNTs, all TGA and weight derivative TGA curves of the MWCNTs treated by the two-step process with various H2/O2 gas flow ratios are presented in Fig. 4-22 and 4-23. As shown in the figures, the largest carbon impurities share percentage is approximately 22 % when the MWCNTs are pretreated by the ion treatment with H2/O2 gas flow ratio of 25/25 (sccm/sccm) (Specimen C3) but the main combustion region is apparently shifted to higher temperature region. This leads to the highest main decomposition
66
temperature at 684 oC which is the same temperature as the MWCNTS only treated by 0.25 M HNO3 acid for 2 h (Specimen A3). The results show that, with the ion pretreatment, the MWCNTs can be treated in dilute acid with high [O]/[C] ratio and thermal stability while the process time is relative low in contrast to the strong acid treatment.
Fig. 4-22 TGA curves of the 5 min ion-pretreated MWCNTs and followed by a 0.25 M HNO
3acid treatment (for Specimens C1 to C5).
Fig. 4-23 Weight-derivative TGA curves of the 5 min ion-pretreated MWCNTs and followed
by a 0.25 M HNO3 acid treatment (for Specimens C1 to C5)For comparison, TGA and weight derivative TGA curves of the MWCNTs treated by various treatments and time conditions are presented in Figs. 4-24 and 4-25. As shown in Fig.
4-24, in contrast to the TGA curves of the as-purchased MWCNTs and the MWCNTs only
67
treated by the ion treatment (Specimen B3), all TGA curves of the MWCNTs treated by any acid treatment present oxidative reaction in lower temperature region raging from 200 to 400
oC. As shown in Fig. 4-25, the satellite peaks of the weight derivative TGA curves show that this region could be contributed to carbon impurities. Meanwhile, the share percentage of the catalyst in the MWCNTs treated by HNO3/H2SO4 acid for 9 h (Specimen A2) is significant higher than the specimens treated by other acid treatments.
Fig. 4-24 TGA curves of MWCNTs for different conditions, (a) as-purchased, (b) 5 min
ion-treated (Specimen B3), (c) merely 0.25 M HNO3 acid-treated (Specimen A3), (d) 5 minion-pretreated and 0.25 M HNO3 acid-treated (Specimen C3), and (e) merely acid-treated MWCNTs (Specimen A2).
Fig 4-25 Weight-derivative TGA curves of MWCNTs for different conditions, (a)
as-purchased, (b) 5 min ion-treated (Specimen B3), (c) merely 0.25 M HNO3 acid-treated (Specimen A3), (d) 5 min ion-pretreated and 0.25 M HNO3 acid-treated (Specimen C3), and(e) merely acid-treated MWCNTs (Specimen A2).
According to the results of TGA, XPS and SEM characterization, although the
68
nitric/sulfuric acid treatment can introduce oxygenated functional groups on nanotube surface, the process is however censorable due to its high structural damage, long process time and solution wasting issues. This figure further depicts that, compared with the results of acid treatments, the ion treatment does not cause serious structural damage, and so the ion-treated sample can still be as the same thermal stability as the as-purchased sample.
According to the results of TGA, XPS in Table 4-1 and SEM characterization, the process combining the ion pretreatment applying H2/O2 gas flow ratio of 25/25 (sccm/sccm) and a 0.25 M HNO3 post acid treatment for 2 h can yield significant advantages such as a very high [O]/[C] ratio and high thermal stability MWCNTs although nanotubes with high density free radical bonds resulted from the ion pretreatment can be possibly damaged during the dilute acid treatment.
It is noted that the presence of the uncertainties in the results of XPS and TGA may slightly change the data provided above. Uncertainty analysis based on sufficient experiments data are required to revise the process parameters if the proposed method is to be applied for practical uses. Although the analysis has not been conducted yet, the present results are still reliable enough to characterize the performance of this method qualitatively.
69
Chapter 5 Conclusions
This study proposed two new processes and compared with the traditional processes to functionalize MWCNTs, in order to expand the potential applications of CNTs. The examined processes include the acid treatment by dilute or nitric/sulfuric acid solution, the ion treatment and the ion pretreatment followed by a dilute nitric acid treatment. Table 5-1 summarizes the experimental results in terms of efficiency of functionalization ([O]/[C], ID/IG), facial purification (sp2), TGA decomposition temperature, and structure damage. From the experimental results, the following conclusions can be drawn:
1. By comparing the degree of functionalization and structure damage of CNTs by nitric/sulfuric with that by dilute acid treatments, the results show that sonicating CNTs in nitric/sulfuric acid treatment leads to a greater functionalization with [O]/[C] values of 52.7% but too much structure damage (i.e. higher ID/IG ratios and lower decomposition temperatures, up to 0.96 and down to 638 oC), though the values of [O]/[C], sp2 and ID/IG merely represent the near surface features due to limitation of penetration depth of XPS and Raman probes. Meanwhile, the chemical process is found to be with the drawbacks of pollution issue and too long treating time (up to 9 h).
2. When the MWCNTs are treated by the ion treatment, the results indicate that there are existence of maximum values of [O]/[C] and sp2, and minimum values of ID/IG values at medium H2/O2 ratios. The existence of maximum functionalization is due to the competition between the amounts of free radical bonds on nanotube surface and oxygen cations in the ion stream. Meanwhile, it is also shown that the ion treatment causes no significant structure damage, and at treatment times of 5 and 20 minutes, the maximum
70
values of the [O]/[C] values are 31.1% and 59.8% , respectively, at medium H2/O2 ratio (= 25/25 (sccm/sccm)).
Table 5-1 Comparisons of the performance of the modification processes on structure damage
and the ranges of [O]/[C] ratio, sp3 percentage, and decomposition temperatureSurface acid treatment, the results show that the ion-treated MWCNTs at medium H2/O2 ratio (=
25/25 (sccm/sccm)) can be further treated by dilute acid to increase the decomposition temperature from ~ 595 oC up to 684 oC without sacrificing the functionalization ([O]/[C]=52.4%). By comparing different process methods, both nitric/sulfuric and dilute acid treatments can enhance decomposition temperature by eliminating the impurities and the smaller CNTs to vary the size distribution of the tubes but it causes
71
either too much structure damage or too long treating time. According to the results, the two-step process is relatively simple and efficient to functionalize CNTs, simultaneously enhance the decomposition temperature and cause no significant structure damage.
72