Carbon functionalization

Fig. 5.5 demonstrates the Raman spectra for the electrodes after CV scans and H2SO4

immersion. As shown, both samples revealed characteristic peaks which were defined as D-band (1310 cm⁻¹) and G-band (1596 cm⁻¹), respectively. The D-band represented the presence of defects and disorder in the carbon structure while the G-band reflected the graphitic in-plane vibrations with E_2g symmetry.[6] Hence, the ratio of D/G signals suggested the degree of crystallinity in the carbon structure. As mentioned earlier, the electrode undergoing the H₂SO₄ immersion was selected for comparison purpose and it exhibited a D/G value of 2.57. In contrast, the sample after CV scans revealed a D/G value of 2.67. This moderate variation in the D/G ratio inferred that the carbon structure was reasonably maintained after CV treatments. The XPS was adopted to obtain variations on the signals from carbon, oxygen, and fluorine for the electrodes under CV scans with and without the supply of ambient oxygen. We also performed the XPS analysis on the as-prepared electrode without CV scans for comparison. As shown in Fig. 5.6, relevant peaks on the XPS profiles (resolution in 1 eV) were labeled properly and they were identified as F1s, Fkll, O1s, and C1s, respectively. Table 5.1 lists their respective atomic ratios. It can be seen that there was negligible difference in the atomic ratios between samples in the as-prepared state and after CV scans without the supply of ambient oxygen. However, the sample after CV scans with the supply of ambient oxygen revealed a similar carbon amount but its atomic ratio for the oxygen was increased considerably in conjunction with a notable reduction in the fluorine content. These results suggested that the CV scans coupled with the supply of ambient oxygen were able to produce oxygen-rich functional groups on the electrode surface while the Nafion ionomer was partially decomposed.

100

1000 1200 1400 1600 1800 2000

G-band

CV scans with O₂

Int ensity (a.u.)

Raman shift (cm

^-1

)

H₂SO₄ immersion D-band

Figure 5.5. Raman spectra for electrodes after CV scans with ambient oxygen and H2SO4

immersion only. These electrodes contain carbon cloth, XC-72R, and Nafion ionomer.

800 600 400 200 0

F1s

O1s

(b) (c)

Intensity (a.u. )

Binding energy (eV)

(a)

F_KLL C1s

Figure 5.6. XPS surveys for (a) as-prepared electrode, as well as electrodes after CV scans (b) without ambient oxygen and (c) with ambient oxygen. These electrodes contain carbon cloth, XC-72R, and Nafion ionomer.

101

Table 5.1. The atomic ratios for carbon, oxygen, and fluorine from XPS profiles for as-prepared electrode, as well as electrodes after CV scans with and without the supply of ambient oxygen.

C (at%) O (at%) F (at%)

as-prepared 61 4.3 34.7

CV scans without O2 62 3.5 34.5

CV scans with O₂ 61.6 10.5 27.9

Fig. 5.7(a) presents the C1s XPS profiles (resolution in 0.1 eV) for the as-prepared electrode and electrodes after CV scans with and without the supply of ambient oxygen, respectively.

Apparently, the as-prepared sample and the one after CV scans without the supply of ambient oxygen displayed similar patterns as expected. In contrast, the sample after CV scans with the supply of ambient oxygen demonstrated a notable peak around 286-288 eV in addition to the typical C1s signal at 284.5 eV. To understand its nature, this C1s profile was subjected to curve-fitting with known functional groups to determine their relative amounts. Fig. 5.7(b) illustrates the curve-fitting results and the atomic ratios for individual functional groups are listed in Table 5.2. These functional groups were selected from earlier literature reports and were presumed to be present in the functionalized electrodes.[148, 150-152] From Table 5.2, the sample after CV scans with the supply of ambient oxygen revealed a notable reduction in the amount for C−F group. In addition, the oxidized −C=O and –COOH groups were substantially increased along with considerable reduction in the C−C backbone.

102

295 290 285 280

Intensity (a.u. )

Binding energy (eV)

as-prepared

CV scans without O₂ CV scans with O₂

(a)

292 288 284

Int ensity (a.u.)

Binding energy (eV)

(b)

C-C sp²/sp³

C=O C-OH C-OOH

C-F

Figure 5.7. (a) C1s XPS profiles for as-prepared electrode, as well as electrodes after CV scans without ambient oxygen and with ambient oxygen. (b) Curve fitting for the C 1s XPS profile from electrode after CV scans with ambient oxygen. These electrodes contain carbon cloth, XC-72R, and Nafion ionomer.

103

Table 5.2. The atomic ratios for the C−C, −OH, −C=O, −COOH, and C−F from XPS curve fitting for as-prepared electrode, as well as electrodes after CV scans with ambient oxygen and without ambient oxygen and this process resulted in the formation of oxygenated functional groups on the carbon surface. To validate our premise, we attempted to obtain the S2p signal but the 0.5 M H2SO4

aqueous solution provided unnecessary background noises. Hence, we prepared several electrodes (carbon cloth/XC-72R/Nafion ionomer) and subjected them to CVs in 0.1 M HCl aqueous solution instead. The purpose for these CV scans was to decompose the Nafion ionomer so the HCl solution with concentrated residues was formed. According to Teranishi et al., the degradation of Nafion produced F⁻, SO32−

, CO2, SO2, and some fluoro carbons.[163] Subsequently, we immersed the electrode made of XC-72R and carbon cloth in the HCl solution containing concentrated Nafion ionomer residues to allow sufficient adsorption of the decomposed species. As shown in Fig. 5.8, signals from the ion chromatography were attributed to SO₄²⁻ and F⁻ in different intensities. Similar constituents were observed in earlier work by Chen and Fuller for Nafion membrane degradation.[61] In their work, a rather strong CF₃COO⁻ peak was identified on the cathode side associated with the oxygen reduction reaction. Unfortunately, in our case the amount of CF₃COO⁻ was below the detection limit. This notable absence of CF₃COO⁻ was possibly due to its immediate readsorption onto the carbon surface after detachment from the Nafion backbone.

104

Cl-Figure 5.8. Ion chromatogram for Nafion ionomer degradation in 0.1 M HCl aqueous solution.

To monitor the extent of Nafion ionomer degradation, we recorded the signal for SO42−

upon CV cycles and the resulting data are displayed in Fig. 5.9. The value for the 0 cycle was obtained from the sample with immersion in the 0.1 M HCl aqueous solution for 17 min, which served as the reference because the sample of 20 CV cycles experienced the same amount of time in the 0.5 M H2SO4 aqueous solution. As shown, the reference sample revealed sulfate concentration of 0.35 ppm. This reduced amount was not unexpected as the Nafion ionomer likely maintained reasonable chemical stability against the 0.1 M HCl aqueous solution at 25°C. However, once CV scans were applied, the sulfate anion concentrations became larger considerably reaching a plateau after 20 cycles at 4.3 ppm. Apparently, within the first 20 cycles, there appeared a linear increase in the sulfate concentration with cycling number. This indicated that a desirable amount of Nafion decomposition and its subsequent carbon functionalization was possible by selecting appropriate CV scans.

105

0 20 40 60 80 100

0 1 2 3 4 5

2

^rd

cycle

sul fate co ncer tra ction (p pm)

CV cycle numbers

Immersion in HCl 1

^st

cycle

Figure 5.9. Variation of sulfate concentration as a function of CV scans with ambient oxygen. The data at 0th cycle is obtained from the electrode immersed in 0.1 M HCl aqueous solution.

Fig. 5.10 provides the C1s XPS profiles (resolution in 0.1 eV) for the as-prepared electrode (carbon cloth/XC-72R/Nafion ionomer) as well as electrodes (carbon cloth/XC-72R) with and without immersion in the HCl solution containing concentrated residues from Nafion ionomer decomposition. Apparently, the electrode of XC-72R/carbon cloth demonstrated a single C1s peak at 284 eV while the as-prepared electrode exhibited an additional C−F peak around 291 eV.

However, the electrode of XC-72R/carbon cloth immersed in the HCl solution with concentrated decomposed Nafion ionomer residues revealed a strong signal around 289 eV which was attributed to the oxygenated groups on the carbon surface. Table 5.3 lists the atomic ratios for the individual functional groups from the curve-fitting results of Fig. 5.10. Apparently, the sample after immersing in the HCl solution showed a large amount of oxygenated functional groups. We surmised that the Nafion ionomer residue in the HCl solution was likely present as CF₃COO⁻. After chemical adsorption, these residues were transformed to the oxygenated functional groups on the carbon surface.

106

The chemical adsorption of Nafion ionomer residues can also be confirmed from the S2p XPS profile (resolution in 0.1 eV) displayed in Fig. 5.11. The electrode of XC-72R/carbon cloth revealed a characteristic S2p signal near 164 eV which was attributed to the impurity intrinsic to the carbon material. However, the electrode of XC-72R/carbon cloth/Nafion ionomer demonstrated an additional peak around 168 eV which was caused by the HSO₃ from the Nafion ionomer.

Interestingly, the XC-72R/carbon cloth sample immersed in the HCl solution with concentrated Nafion ionomer decomposed residues also exhibited the HSO₃ signal in addition to the S2p from impurity. This further supported our premise that the decomposed Nafion ionomer residues were able to chemically adsorb onto the carbon surface.

107

292 288 284 280

(c)

(b)

Inte n s ity (a .u.)

Binding energy (eV) (a)

Figure 5.10. C1s XPS profiles for (a) as-prepared electrode (carbon cloth/XC-72R/Nafion ionomer), as well as electrodes (carbon cloth/XC-72R) (b) before and (c) after immersion in HCl solution ontaining concentrated residues from Nafion ionomer decomposition.

108

180 175 170 165 160 155 150 Binding energy (eV)

(c)

(b)

(a)

Inte n s ity (a .u.)

HSO S

3

Figure 5.11. S2p XPS profiles for (a) as-prepared electrode (carbon cloth/XC-72R/Nafion ionomer), as well as electrodes (carbon cloth/XC-72R) (b) before and (c) after immersion in HCl solution containing concentrated residues from Nafion ionomer decomposition.

109

Table 5.3. The atomic ratios for C−C, −OH, −C=O, −COOH, and C−F from C1s XPS curve fitting for as-prepared electrode, as well as electrodes made of XC-72R/carbon cloth with and without immersion in HCl solution containing concentrated residues from Nafion ionomer decomposition.

C−C

Fig. 5.12(a) demonstrates the TEM image for Pt nanoparticles deposited on the functionalized electrode followed by electrochemical reduction. As shown, there were plenty Pt nanoparticles uniformly distributed with notable aggregations. The primary particle size from the image analysis software was 2.68±1.62 nm. The TEM image on the reference sample (simple H₂SO₄ immersion followed by electrochemical reduction) is presented in Fig. 5.12(b). Apparently, the amount of Pt nanoparticles was substantially reduced, a fact consistent with earlier findings from ICP-MS. In addition, their size was slightly smaller at 2.20±1.45 nm. These results confirmed that the functionalized electrode enabled a larger amount of Pt deposits, albeit with moderate coalescence.

Fig. 5.13 presents the CV profiles of hydrogen desorption and adsorption for the functionalized and reference electrodes, respectively. As expected, there appeared stronger responses in hydrogen desorption and adsorption for the functionalized electrode because of its relatively larger amount of Pt deposit. Estimation on the ECSA was conducted by the integral area for hydrogen desorption in the anodic scan and the resulting ECSA values were 82.2 and 60.9 cm² for the functionalized and reference electrodes, respectively. This ratio of 1.35 was smaller to that of 1.70 for the Pt loading from ICP-MS. We attributed the reduced ESCA ratio to the observed Pt aggregation on the functionalized electrode.

在文檔中 合成核殼鉑釕奈米顆粒及官能基化碳載體 (頁 117-127)