奈米碳管的儲氫能力

(1)

行政院國家科學委員會補助專題研究計畫

█ 成果報告

□期中進度報告

（計畫名稱）

Hydrogen storage in carbon nanotubes

計畫類別：█ 個別型計畫 □ 整合型計畫

計畫編號：NSC － 96 － 2112 － M － 011 － 002 執行期間：2007 年 8 月 1 日至 2008 年 7 月 31 日

計畫主持人：Ju-Yin Cheng (鄭如茵)

計畫參與人員：Ching-Lin Liu (劉慶霖)、Chang-Chih Kuo (郭昶志)、

Huimin Lee(李慧敏)、Chao Tang(唐超)

成果報告類型(依經費核定清單規定繳交)：█精簡報告 □完整報告

執行單位：Department of Polymer Engineering, National Taiwan University of Science and Technology

中華民國 97 年 7 月 1 日

(2)

Final report

1. Introduction

Carbon nanotubes have been used to store alien gases such as nitrogen [1]. In so doing, carbon nanotubes are first bonded with carboxyl radicals on the surface, and then heated at about 300°C for pore generation on the attachment sites. The whole process includes two typical oxidation reactions: chemical oxidation and thermal oxidation. Chemical oxidation uses basic or acidic solutions to eliminate either small, unstable nanotubes or carbonaceous impurities such as amorphous carbon. If this oxidation step is not completed, then hydrogen has no way to enter carbon nanotubes except the entry of already open tube ends. Therefore, it is difficult to make feasible hydrogen storage media out of carbon nanotubes. It is indeed necessary to carry out such an oxidation treatment and thus create pores on the tube surface prior to hydrogen

permeation. In this report, we will discuss some preliminary results from chemical oxidation of multi-walled carbon nanotubes in five aqueous solutions.

2. Experimental

Five aqueous solutions, that is, hydrochloric acid, nitric acid, sulfuric acid, peroxide, and potassium hydroxide, are made in concentration of 1 M or 1.5 M and in one liter for chemical oxidation of multi-walled carbon nanotubes (MWNTs). 100 mg MWNTs obtained from a CVD method are submersed in such solutions, and the whole system is thermally agitated at 80°C on a hot plate. Oxidation reactions soon reach equilibrium at 35°C within twenty-four hours. After oxidation, MWNTs are filtered and collected on a holey aluminum thin foil covered by a holey waxed paper, and then dried in an oven at 60°C for a day.

As-grown MWNTs as well as chemically treated MWNTs of about 10 mg carried by a small platinum holder are sent to thermogravimetric analysis (TGA). The TGA instrument is heated up to 900°C at the heating rate of 10°C/min and purged by a nitrogen flow of 60 c.c./min during the measurement. Meanwhile, the same samples carried by a piece of glass or a waxed paper are characterized for resonance effects using Raman spectroscopy. The incident radiation of the Raman instrument has the wavelength of ultraviolet light, and scans our samples from 100 to 2,400 cm^-1. We observe a typical Raman spectrum of MWNTs which contains two RBM (radial breathing mode) peaks at about 108 cm^-1 and 215 cm^-1, one D (defect) band at 1380 cm^-1, and one G (graphite) band at 1580 cm^-1.

3. Results and discussion

Figures 1 and 2 show the TGA curves of all MWNTs samples after oxidation in 1 M and 1.5 M solutions. We determine the temperature of the first time thermal decomposition for each curve in this way: differentiate each curve at the first turn and find the maximum absolute value of the differential or the slope of the curve. Using the data, we calculate the difference in the thermal

(3)

difference in temperature divided by 900°C indicates the consumption fractions of some carbon material in as-grown MWNTs. The results from 1 M and 1.5 M solutions as well as as-grown MWNTs are listed in tables 1 and 2. The conclusion inferred from TGA data is that selective oxidation of small, unstable MWNTs occurs in solutions with hydroxyl radicals, yet oxidation of amorphous carbon sp² species is primarily found in solutions with oxygen-based or halogen radicals.

The Raman spectrum labels different carbon structures after oxidation in 1 M solutions as well as those in as-grown MWNTs, as illustrated in figure 3. Under each RBM peak and characteristic peaks such as D band and G band, the ratio of integrated areas from such oxidized MWNTs to as-grown MWNTs reveals the change of representative carbon structures induced by various oxidation reactions. We calculate the integration and summarize the ratios δ in table 3. We can make a conclusion from Raman data that amorphous carbon sp² and sp³ species altogether take 40 % of as-grown MWNTs, surface defects on MWNTs take 30%, and the rest is MWNTs consisting of graphene sheets. If one wants to eliminate amorphous carbon sp² structures use hydrochloric acid, nitric acid, and sulfuric acid. If amorphous carbon sp³ structures are to be terminated, use hydrochloric acid, nitric acid, sulfuric acid as well as potassium hydroxide. If one attempts to remove surface defects from MWNTs, use nitric acid, sulfuric acid, and peroxide.

Once MWNTs are purified, small, unstable MWNTs can be furthermore deleted by peroxide and potassium hydroxide.

Figure 1. TGA results before and after oxidation in 1 M solutions: (a) HCl (b) H2SO4 (c) KOH (d) HNO3 (e) H2O2 (f) as-grown MWNTs [2].

(4)

Figure 2. TGA results before and after oxidation in 1.5 M solutions: (a) HNO3 (b) H2SO4 (c) as-grown MWNTs (d) KOH (e) HCl (f) H2O2 [2].

Figure 3. Raman results (A) RBM: (a) KOH (b) H2O2 (c) H2SO4 (d) HNO3 (e) HCl (f) as-grown MWNTs; (B) D band and G band: (a) KOH (b) H2O2 (c) H2SO4 (d) HNO3 (e) HCl (f) as-grown

(5)

Table 1. The first thermal decomposition temperature and consumption fraction from 1 M solutions as well as as-grown MWNTs [2].

Sample

First thermal

decomposition temperature (°C)

Difference in thermal

Consumption fraction (％)

as-grown 661.26 0 0

HCl 885.21 223.95 24.88

H2SO4 705.04 43.78 4.86

HNO3 280.43 38.63 4.29

H2O2 253.53 380.83 42.31

KOH 699.89 407.73 45.30

Table 2. The first thermal decomposition temperature and consumption fraction from 1.5 M solutions as well as as-grown MWNTs [2].

Sample

First thermal

Difference in thermal

Consumption fraction (％)

as-grown 661.26 0 0

HCl 900＋ 238.74＋ 26.53＋

H2SO4 841.85 180.59 20.07

HNO3 889.3 228.04 25.34

H2O2 273.72 387.54 43.06

KOH 138.43 522.83 58.09

Table 3. Areas integrated under each Raman peak for all samples and ratios δ of oxidation data to as-grown data [2].

RBM1 RBM2 D band G band

as-grown 6308.8 [100,163.9]

71221.4 [163.9,450]

373848.9 [1049.6,1450.8]

346629

[1450.8,1683.1]

HCl 3081.4 [100,171.3]

69868 [171.3,450]

259662.8 [1047.1,1461.1]

226892.9 [1461.1,1680]

H2SO4

3461.6 [100,165.8]

31662.5 [165.8,450]

127971.5 [1019,1460.3]

115429.5 [1460.3,1685.4]

HNO3

5240.8 [100,163.9]

53986.3 [163.9,450]

225652.6 [1048,1474.5]

210268.3 [1474.5,1683.1]

H2O2

6760 [100,166.7]

44798.1 [166.7,450]

200137.9 [1027.9,1473.7]

205412.1 [1473.7,1693.2]

(6)

KOH 6320.6 [100,164.8]

32981.4 [164.8,450]

220043.3 [1031.2,1474.5]

196129.7 [1474.5,1693.9]

δ　RBM1 δ　RBM2 δ　D band δ　G band

HCl 0.488428861 0.98099728 0.694566174 0.65456987

H2SO4 0.548693888 0.44456441 0.342308082 0.33300589 HNO3 0.830712655 0.75800672 0.603593056 0.60660908 H2O2 1.071519148 0.62899775 0.53534436 0.59259929 KOH 1.001870403 0.46308273 0.588588866 0.56582023

4. Conclusions

Before hydrogen storage in carbon nanotubes, they must be chemically treated for a tremendous increase in surface porosity and hence gas adsorption. We use five aqueous solutions on the purpose of oxidizing multi-walled carbon nanotubes (MWNTs). We find that the maximum oxidizing power comes from hydrochloric acid, then nitric acid, and sulfuric acid is the minimum.

The order follows the acidic strength (or acidity) of the acids. Nonetheless, the oxidizing power of potassium hydroxide is grater than that of peroxide. This is because they react with small, unstable MWNTs rather than other acids. These fascinating findings are obtained from both TGA and Raman measurements.

5. References

[1] A. Kuznetsova, D. B. Mawhinney, V. Naumenko, J. T. Jr. Yates, J. Liu, and R. E. Smalley, Chem. Phys. Lett. 321, 292 (2000).

[2] Chen-Yo Kao, Liquid phase oxidation of multi-walled carbon nanotubes in five aqueous solutions, Master thesis, National Taiwan University of Science and Technology, Taiwan ROC, July 2008.