Polypyrrole Coated Thermally Exfoliated Graphite Nanoplatelets and

ACS2GO © 2019. ← → → ←. loading. To add this web app to the home screen open the browser option menu and tap on Add to homescreen...
0 downloads 0 Views 2MB Size
ARTICLE pubs.acs.org/IECR

Polypyrrole Coated Thermally Exfoliated Graphite Nanoplatelets and the Effect of Oxygen Surface Groups on the Interaction of Platinum Catalysts with Graphene-Based Nanocomposites Burcu Saner Okan,† Alp Y€ur€um,‡ Neylan Gorg€ul€u,† Selmiye Alkan G€ursel,† and Yuda Y€ur€um*,† † ‡

Faculty of Engineering and Natural Sciences, Sabanci University, Tuzla, Istanbul 34956, Turkey Grand Water Research Institute, Technion Israel Institute of Technology, Technion City, Haifa 32000, Israel ABSTRACT: Expanded graphite oxide (GO) synthesized by graphite oxidation and thermal expansion was exposed to ultrasonic vibration to obtain graphite nanoplatelets (GNFs). Then, expanded GO as conductive filler was dispersed in polypyrrole matrix in order to be utilized as catalyst support. Electrical conductivities and polymer thickness of GNF-based composites were tailored at different feeding mass ratios. Thermal expansion led to the removal of oxygen functional groups on the surface, and the C/O ratio increased to 6. The highest C/O ratio had less hydrophilic carbon surface, and this decreased the interaction of Pt particles with support. When comparing Pt deposition behavior of GO, expanded GO, graphene nanosheets, and their composites, the results showed that Pt dispersion increased with increasing amount of oxygen functional groups on the surface of the samples. This work was the first comprehensive and quantitative investigation on the relationship between Pt dispersion and surface oxygen functional groups of graphene-based nanocomposites.

1. INTRODUCTION Expanded graphite (EG) is a well-known material obtained from intercalated and oxidized graphite exposed to a thermal shock. Thermal shock provides vaporization of the intercalants and expansion of the crystal lattice of the planes of graphite flakes, and wormlike or vermicular-type structures are obtained at the end of the process.1 Furthermore, rapid heating of graphite oxide (GO) sheets results in superheating and volatilization of the intercalants, embedded solvent, such as water, and the evolution of gas, such as CO2, from chemical decomposition of oxygencontaining species in the GO sheets.2 Zhu et al.3 mixed natural graphite with a mixture of concentrated sulfuric acid and hydrogen peroxide and heated graphite intercalated compounds between 200 and 1000 °C for the decomposition of intercalating acids. Furthermore, EG consists of graphene layers in a hexagonal arrangement bonded together by weak van der Waals forces. The sonication process causes these bonds to break and leads to the formation of graphitic nanoplatelets (GNPs).4 There have been several attempts for the production of EG-filled polymer nanocomposites to strengthen thermal, electrical, and mechanical properties of polymers. Chen et al.5 prepared polymer/graphite conducting composites using EG by a process of in situ polymerization. They demonstrated how the conductivity of composites can be changed in the presence of EG. Challenges in catalyst dispersion, charge transport, and stabilization of catalyst particles in polymer electrolyte membrane fuel cells (PEMFC) have led to the development of novel catalyst supports playing a significant role in anchoring the noble metal catalysts such as Pt and Pd in fuel cells.68 Nanostructured carbons such as single-walled nanotubes, multiwalled nanotubes, and carbon nanofibers have been widely used as the primary catalyst supports for PEMFCs.9 However, there are still obstacles in the commercialization of these materials as PEMFC catalyst r 2011 American Chemical Society

support due to their high production costs. The possibility of employing two-dimensional materials as the conductive sheets to anchor electrocatalysts in PEMFC has been explored with the recent emergence of graphene.10 Graphene sheets can act as a promising catalyst support material for fuel cell electrocatalysts due to their 2D structure, high surface area, superior thermal and electrical conductivities, and potential low manufacturing cost.11 Furthermore, surface modification of catalyst supports provided understanding of the interaction of Pt particles with carbon supports. For instance, the oxidation process led to the formation of surface acidic sites and the destruction of surface basic sites.12 Herein, oxygen functional groups on the surface of catalyst supports have great influence on Pt dispersion. Prado-Burguete et al.13 demonstrated that the presence of oxygen functional groups on the carbon support favors the Pt dispersion. At this point, novel catalyst support materials by combining the exceptional properties of graphene sheets with the structural properties of conducting polymers as a catalyst support have a great importance for the catalyst activity enhancement in PEMFCs. In the present work, expanded GO was fabricated in large quantities by oxidation and thermal expansion. Before polymerization, ultrasonic treatment was applied in order to receive GNPs. GNPs were coated with polypyrrole (PPy) via in situ polymerization of pyrrole (Py) monomer with different feeding mass ratios of Py and expanded GO. In addition, Pt particles were deposited on expanded GO and the role of oxygen surface groups in the interaction of Pt-catalyst support and the effect of thermal treatment were investigated by comparing the Received: July 27, 2011 Accepted: October 12, 2011 Revised: September 12, 2011 Published: October 13, 2011 12562

dx.doi.org/10.1021/ie201613f | Ind. Eng. Chem. Res. 2011, 50, 12562–12571

Industrial & Engineering Chemistry Research

ARTICLE

oxygen content of GO, expanded GO, and graphene nanosheetsbased (GNS-based) composites. To the best of our knowledge, this is the first comprehensive and quantitative investigation in the literature about the relationship between Pt dispersion and oxygen functional groups on the surface of graphene-based nanocomposites.

2. EXPERIMENTAL SECTION 2.1. Materials. The following materials were used: graphite flakes (Sigma-Aldrich); acetic anhydride (Merck, extrapure); sulfuric acid (Fluka, 9597%); potassium dichromate (K2CrO4, Chempur, 99.9%); hydroquinone (Acros, 99%); sodium hydroxide (Merck, 97%); pyrrole (Merck, 98%); ferric chloride (FeCl3, Aldrich, 97%); chloroplatinic acid (Sigma-Aldrich, H2PtCl6, 8 wt % solution in water); sodium borohydride (Sigma-Aldrich, NaBH4, >98.5%). 2.2. Expanded GO. Graphite flakes were oxidized by using K2Cr2O7 and H2SO4 as the oxidizing agents and acetic anhydride as an intercalating agent.1 The solution was stirred at 45 °C for 10 days in order to increase the oxidation degree.14 The sample was filtered and neutralized with 0.1 M NaOH and washed with distilled water several times until the solution became neutral. After the filtration process, GO was dried in a vacuum oven at 80 °C overnight. For the fabrication of GNS, GO sheets were exposed to ultrasonic vibration in water for 1 h at room temperature.1,14 Then, sonicated GO was reduced through refluxing by hydroquinone in distilled water under N2 atmosphere for 24 h.1,14 At the final stage, GNS were washed with distilled water several times and dried in a vacuum oven at 60 °C overnight.1,14 GO sheets were expanded by heating to 1000 °C rapidly in a tube furnace and kept for 5 min at this temperature under an argon atmosphere. 2.3. Synthesis of PPy/Expanded GO Nanocomposites. Expanded GO as a conductive filler was coated with PPy via in situ polymerization of Py at room temperature under N2 atmosphere for 24 h. The precipitated sample was filtered and washed with ethanol and distilled water several times in order to eliminate excess unreacted and side products. The powder was dried under vacuum at 80 °C for 24 h. Before polymerization, expanded GO samples were treated in an ultrasonic bath for 2 h. This allowed expanded GO to break apart into thinner GNPs.4 During PPy polymerization, the Fe3+/Py molar ratio was adjusted to 2.4.15 The feeding mass ratios of expanded GO and Py were 1:1, 2:1, 3:1, 1:2, and 1:4. 2.4. Anchoring Pt Nanoparticles on the Surface of Composites. Pt deposition on the surface of nanocomposites was conducted under ultrasonic vibration for 2 h at room temperature. First, 8% H2PtCl6 solution (0.5 mL) was added into an expanded GOdistilled water mixture (10 mL). Then, 1 M NaBH4 (5 mL) was added into the mixture simultaneously and sonicated about 1 h. Our previous study demonstrated the synthesis and characterization of PPy/GO and PPy/GNS nanocomposites.16 To relate the Pt dispersion with the amount of oxygen surface groups, the functional groups on the surface of GO, expanded GO, and GNSbased composites were analyzed in details. 2.5. Characterization Techniques. The surface morphologies of nanocomposites were investigated by a Leo Supra 35VP field emission scanning electron microscope (SEM). Elemental analyses were performed by an energy-dispersive X-ray (EDX)

Figure 1. SEM images of (a) expanded GO and (b) Pt deposited expanded GO.

analyzing system. Powder X-ray diffraction (XRD) experiments were carried out on a Bruker AXS Advance powder diffractometer fitted with a Siemens X-ray gun, using Cu Kα radiation (λ = 1.5406 Å). The Raman spectra of samples were measured by a Renishaw InVia Reflex Raman microscopy system (Renishaw Plc., New Mills, Wotton-under-Edge Gloucestershire, U.K.) using a 514 nm argon ion laser in the range of 100 to 3200 cm1. Functional groups on the surface of samples were determined by a Nicolet iS10 instrument for Fourier transform infrared spectroscopy (FT-IR). X-ray photoelectron spectroscopy (XPS) measurements were performed using a Physical Electronics Quantum 2000 scanning ESCA microprobe. This system used a focused monochromatic Al Kα X-rays (1486.7 eV) source and a spherical sector analyzer. The electronic conductivities of nanocomposites were measured by a standard four-probe technique at room temperature. Before the measurements, composites were prepared in pellet forms under adjusted pressure.

3. RESULTS AND DISCUSSION 3.1. SEM and EDX Studies. Intercalated graphite with sulfuric acid, potassium dichromate, and acetic anhydride was heated to yield an increase of c-axis direction and form an accordion or 12563

dx.doi.org/10.1021/ie201613f |Ind. Eng. Chem. Res. 2011, 50, 12562–12571

Industrial & Engineering Chemistry Research

ARTICLE

Table 1. EDX Results of Expanded GO and PPy/Expanded GO Composites at Different Feed Ratios of Expanded GO and Py samples

carbon (wt %)

nitrogen (wt %)

oxygen (wt %)

other elements (wt %)

3.5

8.9

expanded GO

87.6

expanded GO:Py = 1:1

54.6

23.0

15.4

7.0

expanded GO:Py = 1:2

53.7

18.0

16.6

11.7

expanded GO:Py = 2:1

56.2

14.1

20.4

9.3

Expanded GO:Py = 3:1

63.6

7.8

24.3

4.3

Expanded GO:Py = 1:4

47.7

17.5

13.4

21.4

Table 2. EDX Results of Pt Deposited Expanded GO and PPy/Expanded GO Composites at Different Feed Ratios of Expanded GO and Py samples

carbon (wt %)

nitrogen (wt %)

oxygen (wt %)

platinum (wt %)

other elements (wt %)

Pt/expanded GO

82.6

10.3

1.2

5.6

Pt/expanded GO:Py = 1:1

46.1

22.8

28.0

0.7

2.4

Pt/expanded GO:Py = 1:2

54.7

17.6

25.7

0.3

1.7

Pt/expanded GO:Py = 2:1

57.5

12.5

23.0

1.0

6.0

Pt/expanded GO:Py = 3:1 Pt/expanded GO:Py = 1:4

70.5 23.1

10.3 9.6

14.7 44.2

0.5 0.2

4.0 22.9

Figure 2. SEM images of (a) Py:expanded GO = 1:1 and (b) Pt deposited Py:expanded GO = 1:1.

Figure 3. SEM images of (a) Py:expanded GO = 1:2 and (b) Pt deposited Py:expanded GO = 1:2.

“wormlike” structure. In addition, heat treatment of these samples caused the thermal decomposition of acetic anhydride

into CO2 and H2O vapors which further swelled the layered graphitic structure. Furthermore, the volume of expanded 12564

dx.doi.org/10.1021/ie201613f |Ind. Eng. Chem. Res. 2011, 50, 12562–12571

Industrial & Engineering Chemistry Research

ARTICLE

Figure 6. Raman spectrum of expanded GO after 10 days of oxidation.

Figure 4. XRD patterns of (a) pristine PPy and (b) expanded GO obtained after 10 days of oxidation. Figure 7. Raman spectra of expanded GO/PPy composites as a function of increasing Py amount.

Figure 5. XRD patterns of PPy/expanded GO composites at different feed ratios of expanded GO and Py.

graphite can increase with the enhancement of ultrasonic vibration.17 Therefore, expanded GO samples were exposed to ultrasonic treatment before polymerization. During ultrasonic

vibration, Py monomer was also added into the solution including expanded GO in order to increase the diffusion of Py monomer through layers. The SEM image of expanded GO exhibited that these layers were crumbled due to the thermal expansion, as depicted in Figure 1a. Then, Pt nanoparticles were successfully deposited on the surface of expanded GO, and the size distribution of Pt particles was in the range of 10 and 20 nm, Figure 1b. EDX results showed that expanded GO included 87.6% C and 3.5% O in its structure, Table 1. The higher C content indicated that most of oxygen functional groups were eliminated from the GO surface during thermal shock. After Pt deposition, 1.2% Pt was deposited on the surface of expanded GO, Table 2. PPy was coated on the surface of expanded GO by in situ chemical oxidative polymerization of Py. Polymer coating and irregular spherelike PPy formation was observed clearly in the SEM image, Figure 2a. Pt dispersion on the laminated structure of expanded GO sheets was not seen clearly in the SEM image of the Py:expanded GO = 1:1 composite, Figure 2b. EDX results of Py:expanded GO = 1:1 composite have been shown in Table 1. PPy/expanded GO composites consisted of 54.6% C, 23.0% N, 12565

dx.doi.org/10.1021/ie201613f |Ind. Eng. Chem. Res. 2011, 50, 12562–12571

Industrial & Engineering Chemistry Research

ARTICLE

Figure 8. Raman spectra of expanded GO/PPy composites as a function of increasing expanded GO amount.

Table 3. Four-Probe Electrical Conductivity Results of Expanded GO and Its Composites sample pristine PPy

Figure 9. FT-IR spectra of GO sheets, expanded GO, and reduced GO (GNS).

conductivity (S/cm) 7.6  104

expanded GO

0.75

expanded GO:Py = 1:1

0.15

expanded GO:Py = 2:1

1.00

expanded GO:Py = 3:1

4.00

expanded GO:Py = 1:2

0.04

and 15.4% O. Nitrogen content proved the PPy polymerization. After Pt deposition, the Pt content on the composite surface was much lower, as about 0.7%, Table 2. During the thermal shock of GO sheets, most of the functional groups degraded and small amounts of functional groups in the structure of expanded GO combined with PPy, and some PPy chains agglomerated with each other as seen in Figure 2a. Therefore, some parts of PPy/ expanded GO composites do not provide the electronic conduction pathway which is required to sink the electrons generated from the chemical reduction of H2PtCl6 solution, and thus Pt dispersion decreases.18 The amount of PPy in the composites were increased in order to observe the polymer thickness and the Pt dispersion. PPycoated GO layers were seen in the SEM image of the Py: expanded GO = 1:2 composite, Figure 3a. In Figure 3b, Pt particles did not adhere to the surface. This stemmed from the agglomeration of some PPy particles instead of layer coating, and thus the reduction of the Pt precursor was hindered.19 EDX results showed that the Py:expanded GO = 1:2 composite contained 53.7% C, 18.0% N, and 16.6% O, Table 1. As the Py content increased in the composite, the amount of anchored Pt on the surface decreased. The EDX analysis of Pt-deposited Py: expanded GO = 1:2 composite included 0.3% Pt, Table 2. Table 1 and Table 2 show the EDX analyses of all composites at different feeding ratios of expanded GO and Py before and after Pt impregnation. The amount of oxygen surface groups plays an important role in the carbonPt interaction.20 Oxygen groups make the carbon surface more hydrophilic, and thus maximum catalyst dispersion is achieved.20 On the other hand,

Figure 10. XPS survey scan spectra of GO sheets, expanded GO, and GNS.

thermal treatment leads to an increase in the basicity of the carbon and eliminates oxygen functional groups. The basicity of the carbon surface can be estimated by the number of the π sites of the carbon basal plane acting as anchoring centers for Pt catalysts.12 Leon et al.21 demonstrated the interaction of oxygenfree basic carbon sites leads to an electron-donorelectronacceptor complex described in the following equations: Cπ þ H3 Oþ T ½Cπ  H3 Oþ

ð1Þ

Cπ þ 2H2 O T ½Cπ  H3 Oþ þ OH

ð2Þ

Therefore, the metalsupport interaction depends on the amount of surface oxygen functional groups and the strength of the π sites on carbon support. However, in PPy/expanded GO composites, the relationship of Pt catalyst with the oxygen 12566

dx.doi.org/10.1021/ie201613f |Ind. Eng. Chem. Res. 2011, 50, 12562–12571

Industrial & Engineering Chemistry Research

ARTICLE

Table 4. XPS Spectra Results for C1s and O1s in the Samples of GO Sheets, Expanded GO, and GNS C1s sample GO sheets expanded GO GNS

group

binding energy (eV)

O1s fwhm

at. %

group

binding energy (eV)

fwhm

at. %

CC

284.3

0.6

56.0

CdO

531.7

1.8

9.4

CO

285.2

1.7

27.3

HO-CdO

533.1

2.2

7.3

CC

284.4

0.5

64.9

CdO

532.2

2.4

3.3

CO

285.1

1.9

31.6

OH

536.6

1.6

0.2

CO

532.4

2.7

11.0

CC

284.3

0.6

59.2

CO, COH

285.3

2.1

29.8

Figure 11. XPS survey scan spectra of composites: Py:GO = 1:1, Py: expanded GO = 1:1, and Py:GNS = 1:1.

amount was not observed due to nonuniform layer coating and the random destruction of graphene layers after thermal treatment, Table 2. On the other hand, there was a correlation between PPy amount and Pt dispersion. As the amount of PPy increased in the composite, Pt dispersion minimized, Table 2. Ptdeposited expanded GO:Py = 1:4 contained 0.2% Pt, whereas Ptdeposited expanded GO:Py = 2:1 consisted of 1.0% Pt. Other elements mentioned in the tables were Al, Cu, Cr, Fe, and their alloys resulting from SEM specimen holders and the specimen stage used during SEM/EDX analyses. 3.2. XRD Studies. PPy had an amorphous structure, Figure 4a. Due to the thermal treatment, the crystal structure of expanded GO had a broad and intense 001 peak near 2θ = 13° and a sharp and lower intense 002 peak near 2θ = 26° in the XRD pattern, Figure 4b. As the content of PPy increased in composite, the 002 peak intensity decreased, and as the amount of expanded GO increased in composite, the intensity of the 002 peak became more intense, Figure 5. The maximum intensity of the 002 peak of expanded GO:Py = 3:1 was about 453 au at 2θ = 26.5. This indicated that the structure of the composite became more crystalline by increasing the amount of expanded GO. In addition, ultrasonic treatment before the polymerization provided easy diffusion of Py monomers through layers, and after the polymerization, all layers were stacked and

formed a crystalline structure. For instance, the intensity of the 002 peak of expanded GO was about 98 au, Figure 5. On the other hand, the coverage of expanded GO by PPy in the feeding mass ratio of expanded GO and Py as 3:1 led to an increase in 002 peak intensity and provided well-ordered structure, Figure 5. 3.3. Raman Spectroscopy Studies. Raman spectroscopy is one of the sensitive techniques for the characterization of crystalline perfection. There are three prominent peaks at about 1404, 1582, and 2729 cm1 which correspond to the Raman shifts of D, G, and D0 of graphene, Figure 6. The structural change of expanded GO-based composites could be investigated by the intensity ratio of the D and G bands, ID/IG. This ratio inversely changed with the size of the crystalline grains or interdefect distance.22 As the amount of Py content increased, the ID/IG ratio increased due to the relative increase of defects (chemical and structural changes, the agglomeration of PPy chains). Raman spectra of expanded GO/PPy composites as a function of increasing Py amount depicted in Figure 7. ID/IG ratios of expanded GO:Py = 1:1, expanded GO:Py = 1:2, and expanded GO:Py = 1:4 were estimated as 0.37, 0.51, and 0.52, respectively. The D0 band can be used to determine the number of graphene layers, and the D0 band became more dominant than the G band if the number of graphene layers was smaller than 5.23 This proved that when ID/IG of the composites increased, the coverage of expanded GO layers by PPy increased. In contrast to layer-by-layer coated GO and GNS nanocomposites, nonuniform polymer dispersion on the surface of expanded GO stemmed from the lack of most of the oxygen functionalities. ID/IG ratios of expanded GO:Py = 1:1, expanded GO:Py = 2:1, and expanded GO:Py = 3:1 were also estimated as 0.37, 0.35, and 0.31, respectively. As the amount of expanded GO in the composite increased, the ID/IG ratio decreased due to the decrease in polymer coating and thus the decrease in thickness. Also, Raman spectra of expanded GO/PPy composites as a function of increasing expanded GO amount are exhibited in Figure 8. 3.4. Electrical Conductivity Measurements. Expanded GO can provide percolated pathways for electron transfer, making the composites electrically conductive.24 The electrical conductivity of samples in the pellet form was measured by the conventional four-probe method. The electrical conductivity results of samples (pristine PPy, expanded GO, and expanded GO/PPy composites) were given in Table 3. Pristine PPy has relatively poor conductivity because of weak compactness and randomly orientation of PPy nanostructures and weak bonding between the polymer particles through the boundaries.25 The conductivity of expanded GO was measured as 0.75 S/cm. After 12567

dx.doi.org/10.1021/ie201613f |Ind. Eng. Chem. Res. 2011, 50, 12562–12571

Industrial & Engineering Chemistry Research

ARTICLE

Table 5. XPS Spectra Results for C1s and O1s in the Samples of Py:GO = 1:1, Py:Expanded GO = 1:1, and Py:GNS = 1:1 Composites C1s sample Py:GO = 1:1

group

O1s

binding energy (eV)

fwhm

at. %

group

binding energy (eV)

fwhm

at. % 6.0

CC

284.8

1.9

46.7

CO

531.2

1.7

CdO, CN

287.3

3.4

27.0

CdO

532.7

2.4

7.4

Py:expanded GO =1:1

CC COC

284.7 286.2

1.6 2.1

46.4 21.0

CO, CdO HO-CdO

531.6 533.1

1.8 1.7

8.0 6.6

CdO, CdN, NdCO

288.4

1.3

7.9

Py:GNS = 1:1

CC

284.4

2.1

44.0

CO

530.9

1.5

4.8

CdO, CN

287.4

3.4

24.6

CdO

532.3

3.1

13.0

Table 6. XPS Spectra Results for N1s in the Samples of Py: GO = 1:1, Py:Expanded GO = 1:1, and Py:GNS = 1:1 Composites N1s sample Py:GO = 1:1

group

binding energy (eV) fwhm at. %

CN, NH

399.9

1.7

12.9

Py:expanded GO = 1:1 CN, NH

399.9

1.5

10.1

Py:GNS = 1:1

399.7

1.9

13.6

CN, NH

the addition of expanded GO into polymer matrices in adjusted feeding ratios, the conductivity of composites was enhanced. The conductivity of a composite synthesized with the feeding mass ratios of expanded GO and Py as 3:1 was measured as 4 S/cm. Therefore, the composites exhibited a comparably higher electrical conductivity than pure PPy because the layered and network structure of expanded GO formed large fractions of electron conducting paths in the composite. However, as the amount of PPy in the composite increased, the conductivity values decreased. 3.5. Effect of Surface Oxygen Groups on Pt Deposition. FTIR, XPS, and EDX analyses were used in order to estimate the amount of surface oxygen functional groups in GO, expanded GO, and GNS and investigate the effect of surface oxygen groups and thermal treatment on the Pt dispersion. The characteristic peaks in the FTIR spectrum of the GO sheets were two sharp CH stretching bands at 2850 and 2916 cm1 and a sharp CH2 bending band near 1480 cm1, Figure 9. Also, there were a broad band at around 1100 cm1 due to the aromatic CO stretching and two small peaks due to the CdO stretching. Expanded GO had high carbon content as seen in Figure 9. This indicated that thermal expansion eliminated oxygen functional groups. After the chemical reduction of GO, the intensity of CdO stretching peaks at 1500 °C decreased comparably, Figure 9. XPS is a quantitative surface analysis technique that evaluates the elemental composition, empirical formula, chemical state, and electronic state of the elements. For the identification of oxygen-containing functional groups, the C1s, O1s, and N1s signals were measured and groups were assigned on the basis of the differences in their binding energy of carbon atoms.26 The atomic ratios and surface functional groups of GO sheets, expanded GO, GNS, and their composites were determined by using the XPS elemental analysis. The intensities of O1s and C1s

Table 7. EDX Results of GO Sheets, Expanded GO, and GNS samples

carbon (wt %)

oxygen (wt %)

other elements (wt %)

GO sheets

46.5

42.8

10.7

expanded GO

87.6

3.5

8.9

GNS

70.3

27.9

1.8

peaks for GO, expanded GO, and GNS were compared in the XPS survey scan spectra, Figure 10. The C/O ratios of GO, expanded GO, and GNS were measured as 2.3, 6.0, and 3.2, respectively. These results indicated that thermal expansion led to the removal of oxygen functional groups on the surface, and thus carbon content increased in the structure of expanded GO. Table 4 has summarized the functional groups, binding energies, full width at half-maximum (fwhm) values, and atomic percentages which were estimated from the C1s and O1s XPS spectra of GO sheets, expanded GO, and GNS. The fwhm of the peak assigned to CO in the XPS O1s spectrum of GNS with a value of 2.7 appeared much larger than all other carbonoxygen containing bonds. Herein, the large fwhm and the broad tail toward higher binding energy indicate that different types of carbonoxygen containing bonds overlap each other.27 In Table 4, the C1s envelope of GO sheets contained two peaks at 284.3 and 285.2 eV which were assigned to the nonoxygenated ring C and the C in CO.28 There were two deconvoluted O1s peaks of GO sheets at 531.7 and 533.1 eV which were attributed to CdO and HO—CdO groups, respectively.26 Although the C1s peaks of expanded GO had the same oxygen functional groups as GNS, the atomic weight percent of its CC peak was much larger than that in the C1s XPS spectrum of GO. This again proved the removal of oxygen functional groups from the surface during thermal treatment. The C1s peaks of GNS also belonged to the same carbonoxygen containing functional groups instead of COH bonds. This indicated that, also, CdO bonds in the structure of GO sheets were converted to CO bonds due to the chemical reduction of GO sheets by hydroquinone. In this reaction, hydroquinone loses either one H+ from one of its hydroxyls to form a monophenolate ion or two H+ from both hydroxyls to form a diphenolate ion and thus GO sheets gain H+.29 Furthermore, the deconvoluted O1s XPS spectrum of GNS exhibited that there was a CO peak near 532.4 eV, and the atomic weight percentage of its O1s peak was 11%, considerably lower than that of GO sheets. 12568

dx.doi.org/10.1021/ie201613f |Ind. Eng. Chem. Res. 2011, 50, 12562–12571

Industrial & Engineering Chemistry Research

ARTICLE

Table 8. EDX Results of Pt Deposited GO Sheets, Expanded GO, and GNS other carbon

oxygen

platinum

elements

(wt %)

(wt %)

(wt %)

(wt %)

Pt/GO sheets

41.8

44.8

3.5

9.9

Pt/Expanded

70.9

10.9

1.2

16.7

57.0

31.8

1.4

9.8

samples

GO Pt/GNS

Table 9. EDX Results of PPy/GO Sheets, PPy/Expanded GO, and PPy/GNS Composites carbon

nitrogen

oxygen

other elements

samples

(wt %)

(wt %)

(wt %)

(wt %)

Py:GO = 1:1 Py:expanded GO =

40.2 60.6

16.7 21.6

32.9 11.0

10.2 6.8

42.3

7.9

44.3

5.5

1:1 Py:GNS = 1:1

Table 10. EDX Results of Pt Deposited PPy/GO Sheets, PPy/Expanded GO, and PPy/GNS Composites carbon

nitrogen

oxygen

samples

(wt %)

(wt %)

(wt %)

(wt %)

(wt %)

Pt/Py:GO = 1:1

37.4

20.2

34.2

6.1

2.1

Pt/expanded: GO = 1:1

46.1

22.8

28.0

0.7

2.4

39.1

3.6

40.9

7.9

8.5

Pt/Py:GNS

=

platinum other elements

1:1

The XPS peaks of C1s, O1s, and N1s for Py:GO = 1:1, Py: expanded GO = 1:1, and Py:GNS = 1:1 composites were shown in the XPS survey scan spectra, Figure 11. The C/O ratios of Py: GO = 1:1, Py:expanded GO = 1:1, and Py:GNS = 1:1 were 1.4, 1.6, and 1.2, respectively. Py:expanded GO = 1:1 composite with the highest C/O ratio had less hydrophilic carbon surface, and this decreased the interaction of the catalyst particles with the support.30 Table 5 and Table 6 have summarized functional groups, binding energies, fwhm values, and atomic percentages which were estimated from the C1s, O1s, and N1s XPS spectra of Py: GO = 1:1, Py:expanded GO = 1:1, and Py:GNS = 1:1 composites. Py:GNS = 1:1 composite with the largest fwhm indicated that different types of carbonoxygen and carbon nitrogen containing bonds are superimposed.27 The deconvoluted C1s spectrum of Py:GO = 1:1 nanocomposite had two main peaks near 284.8 and 287.3 eV which were assigned to the nonoxygenated ring C and the overlapped CdO and CN peaks, respectively.31 The XPS O1s peaks of Py:GO = 1:1 nanocomposite at 531.2 and 532.7 eV belonged to CO and CdO bonds, respectively. Moreover, the N1s peak around 399.9 eV was assigned to the overlap of the NH and CN components which indicated a considerable PPy coating, Table 6.32 The C1s chemical shifts at 284.7, 286.2, and 288.4 eV of Py:expanded GO = 1:1 composite were attributed to CC,

COC, and the overlap of CdO, CdN, and NdCO bonds, respectively, Table 5. The higher amount of carbon groups in the structure of Py:expanded GO = 1:1 nanocomposite was also supported by its deconvoluted C1s peaks. Moreover, the XPS O1s spectrum of Py:expanded GO = 1:1 nanocomposite had two main peaks at 531.6 and 533.1 eV which were assigned to the overlapped CO and CdO bonds, and OH—CdO bond, respectively. The XPS N1s peak near 399.9 eV corresponded to the overlapping peaks of NH and CN bonds, Table 6.32 Although Py:GNS = 1:1 nanocomposite included the same carbon-containing functional groups as Py:GO = 1:1 nanocomposite, the atomic weight percent of CdO peak on the surface of Py:GNS = 1:1 nanocomposite was slightly smaller than that of the Py:GO = 1:1 nanocomposite because chemical reduction led to the transformation of the sp2 carbonoxygen bonds to sp3 hybridized carbon bonds. Furthermore, the deconvoluted O1s peaks of Py:GNS = 1:1 nanocomposite at 530.9 and 532.3 eV were assigned to CO and CdO bonds, respectively, Table 5. This indicated that most of the CdO bonds in the structure of GO were converted to CO bonds using hydroquinone as a reducing agent. The XPS N1s spectrum of Py:GNS = 1:1 contained the overlapping peaks of NH and CN components due to the PPy coating on the surface of GNS, Table 6.32 Table 7 showed EDX results of GO sheets, expanded GO, and GNS. The C/O ratios of GO sheets, expanded GO, and GNS were 1.0, 25.0, and 2.6, respectively. Expanded GO had high C/O ratio, and thus a low amount of oxygen functional groups decreased metal dispersion. Table 8 exhibited EDX results of Pt deposited GO sheets, expanded GO, and GNS. The results proved that the higher oxygen amount in GO structure hindered agglomeration and promoted Pt dispersion on the surface of sheets. The effect of oxygen surface groups on Pt dispersion was also investigated for PPy/GO sheets, PPy/expanded GO, and PPy/ GNS composites. The EDX analyses of composites have been exhibited in Table 9. The C/O ratios of Py:GO = 1:1, Py: expanded GO = 1:1, and Py:GNS = 1:1 were calculated as 1.22, 5.50, and 0.95, respectively. Table 10 shows the EDX analyses of Pt deposited composites. These results supported that Pt dispersion increased with an increasing amount of oxygen surface groups.

4. CONCLUSIONS Expanded GO was used as a conductive filler in conducting PPy matrix. Before the polymerization, expanded GO samples were exposed to ultrasonic vibration to receive GNPs. Instead of a layer-by-layer coating, nonuniform polymer dispersion on the surface of expanded GO occurred due to the removal of oxygen functional groups on the surface during thermal expansion of GO sheets. In addition, the electrical conductivity of PPy/expanded GO increased up to 4 S/cm, as the amount of expanded GO increased in the composite. Both XRD and Raman spectroscopy results indicated that although PPy was agglomerated with each other, the thickness of composites changed linearly with the increase of the PPy amount. With this study, the Pt deposition behavior on the surfaces of GO, expanded GO, GNS, and their composites were compared regarding surface oxygen functional groups. XPS and EDX results showed that Pt dispersion increased with 12569

dx.doi.org/10.1021/ie201613f |Ind. Eng. Chem. Res. 2011, 50, 12562–12571

Industrial & Engineering Chemistry Research increasing the amount of oxygen functional groups on the surface of samples. However, the random destruction of graphene layers after thermal treatment and the lack of most of the oxygen functional groups decreased the Pt dispersion on the surface. Since thermal expansion led to the removal of oxygen functional groups on the surface, the C/O ratio increased up to 6.0 in the structure of expanded GO. Therefore, the highest C/O ratio had less hydrophilic carbon surface, and this decreased the interaction of Pt catalysts with the support. Consequently, the relationship between surface oxygen functional groups and catalyst particles carries a significant importance on the durability of fuel cells. This comprehensive study will open up new ways for the production of novel catalyst support materials to be utilized in fuel cells.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. Tel.: +90 216 4839512. Fax: +90 216 4839550.

’ ACKNOWLEDGMENT The authors acknowledge to Prof. Dr. Levent Toppare from Middle East Technical University, Turkey for his help in fourprobe measurements, Assoc. Prof. Mustafa Culha from Yeditepe University, Turkey for his help in using their Raman spectroscopy, and Prof. Dr. S-efik S€uzer from Bilkent University for his help in XPS characterization. ’ REFERENCES (1) Saner, B.; Okyay, F.; Y€ur€um, Y. Utilization of multiple graphene layers in fuel cells. 1. An improved technique for the exfoliation of graphene-based nanosheets from graphite. Fuel 2010, 89 (8), 1903– 1910. (2) Prudhomme, R. K.; Aksay, I.; Adamson, D.; Abdala, A. Tire containing thermally exfoliated graphite oxide. U.S. Patent 0054581 A1, 2009. (3) Gu, W. T.; Zhang, W.; Li, X. M.; Zhu, H. W.; Wei, J. Q.; Li, Z.; Shu, Q. K.; Wang, C.; Wang, K. L.; Shen, W. C.; Kang, F. Y.; Wu, D. H. Graphene sheets from worm-like exfoliated graphite. J. Mater. Chem. 2009, 19 (21), 3367–3369. (4) Li, B.; Zhong, W. H. Review on polymer/graphite nanoplatelet nanocomposites. J. Mater. Sci. 2011, 46 (17), 5595–5614. (5) Chen, G. H.; Wu, D. J.; Weng, W. G.; Yan, W. L. Preparation of polymer/graphite conducting nanocomposite by intercalation polymerization. J. Appl. Polym. Sci. 2001, 82 (10), 2506–2513. (6) Shao, Y. Y.; Liu, J.; Wang, Y.; Lin, Y. H. Novel catalyst support materials for PEM fuel cells: Current status and future prospects. J. Mater. Chem. 2009, 19 (1), 46–59. (7) Gubler, L.; Beck, N.; Gursel, S. A.; Hajbolouri, F.; Kramer, D.; Reiner, A.; Steiger, B.; Scherer, G. G.; Wokaun, A.; Rajesh, B.; Thampi, K. R. Materials for polymer electrolyte fuel cells. Chimia 2004, 58 (12), 826–836. (8) G€ursel, S. A.; Dogan, H. D. C. Preparation and characterisation of novel composites based on a radiation grafted membrane for fuel cells. Fuel Cells 2011, 11 (3), 361–371. (9) Natarajan, S. K.; Cossement, D.; Hamelin, J. Synthesis and characterization of carbon nanostructures as catalyst support for PEMFCs. J. Electrochem. Soc. 2007, 154 (3), B310–B315. (10) Novoselov, K. S.; Geim, A. K.; Morozov, S. V.; Jiang, D.; Zhang, Y.; Dubonos, S. V.; Grigorieva, I. V.; Firsov, A. A. Electric field effect in atomically thin carbon films. Science 2004, 306 (5296), 666–669.

ARTICLE

(11) Seger, B.; Kamat, P. V. Electrocatalytically active grapheneplatinum nanocomposites. Role of 2-D carbon support in PEM fuel cells. J. Phys. Chem. C 2009, 113, 7990–7995. (12) Yu, X. W.; Ye, S. Y. Recent advances in activity and durability enhancement of Pt/C catalytic cathode in PEMFC—Part I. Physicochemical and electronic interaction between Pt and carbon support, and activity enhancement of Pt/C catalyst. J. Power Sources 2007, 172 (1), 133–144. (13) Prado-Burguete, C.; Linaressolano, A.; Rodriguez-Reinoso, F.; Delecea, C. S. The effect of oxygen-surface groups of the support on platinum dispersion in Pt/carbon catalysts. J. Catal. 1989, 115 (1), 98–106. (14) Saner, B.; Dinc, F.; Y€ur€um, Y. Utilization of multiple graphene nanosheets in fuel cells: 2. The effect of oxidation process on the characteristics of graphene nanosheets. Fuel 2011, 90 (8), 2609–2616. (15) Armes, S. P. Optimum reaction conditions for the polymerization of pyrrole by iron(iii) chloride in aqueous-solution. Synth. Met. 1987, 20 (3), 365–371. (16) Saner, B.; Alkan-G€ursel, S.; Y€ur€um, Y. Layer-by-layer polypyrrole coated graphite oxide and graphene nanosheets as catalyst support materials for fuel cells. Fullerenes, Nanotubes, Carbon Nanostruct. 2011in press. (17) Li, J. H.; Li, J.; Li, M. Preparation of expandable graphite with ultrasound irradiation. Mater. Lett. 2007, 61 (28), 5070–5073. (18) Lee, H. Y.; Kim, J. Y.; Park, J. H.; Joe, Y. G.; Lee, T. H. Performance of polypyrrole-impregnated composite electrode for unitized regenerative fuel cell. J. Power Sources 2004, 131 (12), 188–193. (19) Sepulveda-Escribano, A.; Coloma, F.; Rodriguez-Reinoso, F. Platinum catalysts supported on carbon blacks with different surface chemical properties. Appl. Catal., A 1998, 173 (2), 247–257. (20) Coloma, F.; Sepulveda-Escribano, A.; Fierro, J. L. G.; Rodriguez-Reinoso, F. Gas phase hydrogenation of crotonaldehyde over Pt/ activated carbon catalysts. Influence of the oxygen surface groups on the support. Appl. Catal., A 1997, 150 (1), 165–183. (21) Leon, C. A. L. Y.; Solar, J. M.; Calemma, V.; Radovic, L. R. Evidence for the protonation of basal-plane sites on carbon. Carbon 1992, 30 (5), 797–811. (22) Ferrari, A. C.; Robertson, J. Interpretation of Raman spectra of disordered and amorphous carbon. Phys. Rev. B 2000, 61 (20), 14095– 14107. (23) Gupta, A.; Chen, G.; Joshi, P.; Tadigadapa, S.; Eklund, P. C. Raman scattering from high-frequency phonons in supported n-graphene layer films. Nano Lett. 2006, 6 (12), 2667–2673. (24) Wang, W. P.; Pan, C. Y.; Wu, J. S. Electrical properties of expanded graphite/poly (styrene-co-acrylonitrile) composites. J. Phys. Chem. Solids 2005, 66 (10), 1695–1700. (25) Sahoo, N. G.; Jung, Y. C.; So, H. H.; Cho, J. W. Polypyrrole coated carbon nanotubes: Synthesis, characterization, and enhanced electrical properties. Synth. Met. 2007, 157 (89), 374–379. (26) Zhang, L.; Li, Y.; Zhang, L.; Li, D. W.; Karpuzov, D.; Long, Y. T. Electrocatalytic oxidation of NADH on graphene oxide and reduced graphene oxide modified screen-printed electrode. Int. J. Electrochem. Sci. 2011, 6 (3), 819–829. (27) Yang, D.; Velamakanni, A.; Bozoklu, G.; Park, S.; Stoller, M.; Piner, R. D.; Stankovich, S.; Jung, I.; Field, D. A.; Ventrice, C. A.; Ruoff, R. S. Chemical analysis of graphene oxide films after heat and chemical treatments by X-ray photoelectron and micro-Raman spectroscopy. Carbon 2009, 47 (1), 145–152. (28) Park, S.; Lee, K. S.; Bozoklu, G.; Cai, W.; Nguyen, S. T.; Ruoff’, R. S. Graphene oxide papers modified by divalent ions—Enhancing mechanical properties via chemical cross-linking. ACS Nano 2008, 2, 572–578. (29) Jia, Y. F.; Demopoulos, G. P. Adsorption of silver onto activated carbon from acidic media: Nitrate and sulfate media. Ind. Eng. Chem. Res. 2003, 42, 72–79. (30) Coloma, F.; Sepulveda-Escribano, A.; Fierro, J. L. G.; RodriguezReinoso, F. Preparation of platinum supported on pregraphitized carbonblacks. Langmuir 1994, 10, 750–755. 12570

dx.doi.org/10.1021/ie201613f |Ind. Eng. Chem. Res. 2011, 50, 12562–12571

Industrial & Engineering Chemistry Research

ARTICLE

(31) Shen, W.; Li, Z.; Liu, Y. Surface chemical functional groups modification of porous carbon. Recent Pat. Chem. Eng. 2008, 1, 27–40. (32) Zhang, J. P.; Zhang, L. D.; Zhu, L. Q.; Zhang, Y.; Liu, M.; Wang, X. J.; He, G. Characterization of ZnO:N films prepared by annealing sputtered zinc oxynitride films at different temperatures. J. Appl. Phys. 2007, 102 (11), 114903–1149037.

12571

dx.doi.org/10.1021/ie201613f |Ind. Eng. Chem. Res. 2011, 50, 12562–12571