Sajini
Vadukumpully
,
Jhinuk
Gupta
,
Yongping
Zhang
,
Guo Qin
Xu
and
Suresh
Valiyaveettil
*
Department of Chemistry, National University of Singapore (NUS), 3 Science Drive 3, 117543, Singapore. E-mail: chmsv@nus.edu.sg; Fax: +65 67791691; Tel: +65 65164327
First published on 4th November 2010
A facile and simple approach for the covalent functionalization of surfactant wrapped graphene sheets is described. The approach involves functionalization of dispersible graphene sheets with various alkylazides and 11-azidoundecanoic acid proved the best azide for enhanced dispersibility. The functionalization was confirmed by infrared spectroscopy and scanning tunneling microscopy. The free carboxylic acid groups can bind to gold nanoparticles, which were introduced as markers for the reactive sites. The interaction between gold nanoparticles and the graphene sheets was followed by UV-vis spectroscopy. The gold nanoparticle-graphene composite was characterized by transmission electron microscopy and atomic force microscopy, demonstrating the uniform distribution of gold nanoparticles all over the surface. Our results open the possibility to control the functionalization on graphene in the construction of composite nanomaterials.
Given the interest in graphene chemistry, it is desired to have convenient dispersion routes to increase the availability and processability of graphene. To improve the solution processability of graphene, a few methods that have been employed for the functionalization and stabilization of carbon nanotubes (CNTs) can be adopted.26,27 Moreover, chemical functionalization of the carbon network in graphene by grafting atoms or molecules is very important because this provides a method for the introduction of interesting photoactive molecules for the development of optoelectronic materials. An interesting functionalization to try on graphene is azide addition because many organic azides have been used for the covalent modification of single walled and multi walled CNTs.28
Based on this concept, a mild covalent functionalization of cetyltrimethylammonium bromide (CTAB) wrapped graphene sheets using a series of azides, which include alkylazides (hexyl and dodecyl), 11-azidoundecanol (AUO), and 11-azidoundecanoic acid (AUA), is developed. Besides the azido group, other functional groups such as hydroxyl and carboxylic groups are incorporated to enhance the dispersibility in common organic solvents. The functionalized sheets are characterized by Fourier transform infrared spectroscopy (FTIR), Raman spectroscopy and various imaging techniques. In addition, the reactive sites on the graphene sheets were marked using gold nanoparticles. Free carboxylic acid groups present in the AUA functionalized sheets bind with gold nanoparticles, indicating that the reaction has taken place not just at the edges but also at the internal C
C bonds of graphene. In short, our results show that the covalent functionalization using azides provides a useful platform for the synthesis of functional graphene nanocomposites using gold nanoparticles.
![]() | ||
| Scheme 1 Schematic representation of the covalent functionalization of graphene sheets with various alkylazides. | ||
The presence of polar functional groups on the graphene surface is desirable because it assists in enhancing the dispersibility. Besides, functional groups such as –COOH and –OH can anchor metal nanoparticles onto the surface, which may help in visualizing the reactive sites. Hence, gold nanoparticle synthesis was attempted in the presence of both AUO and AUA functionalized graphene sheets. In the presence of AUO functionalized sheets, the gold nanoparticles were aggregated on their surface. Transmission electron microscopic (TEM) images of the AUO functionalized sheets and the gold nanocomposites are given in Fig. S3 of the ESI.† A possible reason for this may be the poor stabilizing/coordinating effect of hydroxyl groups. However, gold nanocomposites with AUA functionalized graphene sheets were found to be stable without any visible aggregation. Hence, we focused our studies on AUA functionalized graphene owing to the fact that the presence of –COOH groups on the surface might enhance the metal nanoparticle-graphene composite stabilities and dispersibility. The dispersibility of the resultant product was poor when a 1
:
1 weight ratio of graphene to AUA was used for functionalization. It could be due to minimal functionalization on the graphene sheets. The TEM image of the dispersed sample is given in Fig. S4 of the ESI.† Treatment of graphene with AUA at a 1
:
10 ratio led to the formation of a solid considerably dispersible in common organic solvents. These sheets were found to be easily dispersible in polar solvents with a mild sonication. These samples were used for subsequent experiments. The dispersibility was found to be high in DMF. In the case of CTAB stabilized graphene sheets, the maximum dispersibility achieved was ∼0.005–0.01 mg mL−1, but for AUA functionalized sheets, it could reach up to a maximum of ∼0.05–0.1 mg mL−1.
Fig. 1a and 1b show the TEM micrographs of the CTAB stabilized and AUA functionalized graphene sheets, respectively. Due to the distortions caused by the covalent functionalization, sheets attain a crumpled topology (Fig. 1b). As reported previously, crumpling and scrolling are intrinsic properties of thin graphene sheets, which is due to the extra thermodynamic stability of the 2D membranes arising from microscopic crumpling, either by bending or buckling.30
![]() | ||
| Fig. 1 TEM images of (a) CTAB stabilized and (b) AUA functionalized graphene sheets. (c) FTIR and (d) Raman spectra of the AUA functionalized graphene sheets. | ||
To verify the covalent functionalization of graphene sheets with AUA, FTIR was performed (Fig. 1c). Traces A and B correspond to the IR spectrum of AUA and AUA functionalized graphene sheets (1
:
10 w/w of graphene to AUA), respectively. The peak at ∼2100 cm−1 corresponds to the azido stretching31 and that at 1730 cm−1 to the carbonyl stretching from the –COOH group. Absence of an azido stretching peak in AUA functionalized graphene sheets implies covalent functionalization.28bAzides are added onto C
C bonds on graphene with the elimination of N2. The IR spectrum of the AUA functionalized graphene sample showed the presence of C–N (∼1128 cm−1) and C
C stretching (∼1630 cm−1).
Fig. 1d shows the Raman spectra of CTAB stabilized graphene sheets (trace 1) and the AUA functionalized sheets with 1
:
1 and 1
:
10 w/w of graphene to AUA (traces 2 and 3, respectively). The peak centered at ∼1350 cm−1 corresponds to the disorder induced D band and that at ∼1580 cm−1 to the E2g phonon of sp2 atoms.32 The intensity of the D band is directly proportional to the amount of disorder present in the sheets. The D band intensity increased considerably with azide functionalization, owing to the disorders induced by the addition of nitrenes to the double bonds. Another prominent feature in the Raman spectra of graphene samples is the 2D (or G′) peak. No defects are required for the existence of this band and it is seen even when no D band is present.24 The number of layers present in the graphene sheets can be evaluated from the shape of 2D peak.33Graphene sheets with less than five layers have very broad 2D peaks, which can be clearly seen in the case of both CTAB stabilized and AUA functionalized graphene samples.
The thickness and morphology of the functionalized graphene sheets was evaluated using atomic force microscopy (AFM) (Fig. 2a). From the cross section analysis of one of the flakes, the thickness was found to be 1.12 nm (Fig. 2b). Analysis of a large number of AFM images revealed graphene flakes with lateral dimensions of 500–2000 nm and thicknesses in the range of 1–1.3 nm (statistical analysis on the thickness is given in Fig. S5 of the ESI†).
![]() | ||
| Fig. 2 AFM images of (a) the AUA functionalized graphene sheets and (b) the section analysis of one of the flakes. | ||
The thickness of a monolayer of graphene sheets is 0.34 nm as predicted by theory,3 while our samples showed an average thickness of 1.3 nm, indicating the presence of 2–3 layered graphene sheets. A slight increase in the thickness of the sheets as compared with the CTAB stabilized sheets29 can be attributed to the presence of long alkyl chains present on both sides of the sheets.
Finally, scanning tunneling microscopy (STM) analysis was carried out on the samples to detect the disorders caused by functionalization. A highly ordered hexagonal lattice of graphene was seen in the case of CTAB stabilized graphene flakes (Fig. 3a). Small nonuniformities arose from the presence of loosely bound residual surfactants. However, the STM image of AUA functionalized sheets clearly showed many bright regions, lacking the hexagonally ordered arrangement. These disorders could be due to the functionalization on certain areas (Fig. 3b). The functionalized regions are highlighted with white contours.
![]() | ||
| Fig. 4 (a) UV-vis spectrum of the AUA functionalized graphene sheet (trace A) and that of the gold-graphene nanocomposite (trace B) solutions. Inset of Fig. 4a shows the photograph of (A) functionalized graphene solution and that of (B) the gold-AUA graphene composite in DMF. (b) TEM image and (c) the AFM image with (d) the cross sectional analysis of the gold-graphene nanocomposite sheets. | ||
Homogeneous grafting of gold nanoparticles on the graphene film is visible in the TEM image (Fig. 4b).
From the TEM images, the average size of the nanoparticles was found to be 9.5 ± 2.2 nm. The AFM image also depicts fine distribution of nanoparticles on the graphene nanosheets (Fig. 4c). From the cross sectional analysis (Fig. 4d), the height from background was found to be ∼1.26 nm, corresponding to the AUA functionalized graphene sheet thickness. The smaller peaks seen in the cross section were from the gold nanoparticles anchored on the graphene flakes. Even though the reaction mechanism is not clear at this stage, it is believed that the nanoparticles are adhered onto the functionalized graphene sheets via electrostatic interactions and –COOH groups.
The uniform distribution of the nanoparticles on the AUA functionalized graphene indicates that the reaction has taken place not just at the edges of the graphene sheets but also at the central C
C bonds. This is very interesting because this method of functionalization can open up new reaction routes for the synthesis of graphene based functional composite materials.
Primary investigations on the applicability of the graphene-gold nanoparticle composite as a conducting composite material were carried out using I–V measurements. Fig. 5 shows the I–V plots of drop-casted films of AUA functionalized graphene sheets and the gold-graphene nanocomposite. It is evident from the graph that AUA functionalized graphene sheets show low conductivity due to the lack of extended π-conjugation. The film showed a conductance of 22 nA at an applied voltage of 10 V (trace A). In the case of gold-graphene nanocomposites, the conductivity increased approximately three fold as compared to pure functionalized sheets. Gold-graphene nanocomposite films showed a conductance of 76 nA at an applied voltage of 10 V (trace B). The increase in conductance for the gold-graphene nanocomposites can be attributed to the presence of attached gold nanoparticles. Both the drop-casted samples contained the same concentration of graphene sheets. Hence, it can be concluded that the increased conductivity is due to the embedded gold nanoparticles. Some of the earlier works on GO based composites showed 6 fold35 and 270
000 fold36,37 increases in conductance on reduction of GO by chemical means. This increase can be attributed to the restoration of π–π electronic conjugation or presence of metallic impurities on the graphene surface. In our case, there is no restoration of the extended conjugation or use of metallic reducing agents. Therefore, the increase in conductivity is mainly due to the embedded Au nanoparticles, which create the conducting pathways. The conductivity obtained is lower than that of bulk graphite since the measurement was done on drop-casted films with multiple discontinuities. At the present stage, it will be difficult to comment on the mechanism of electrical conduction, despite the great potential of such materials. Attempts to anchor other metal nanoparticles such as Pt and Fe onto the functionalized graphene sheets and their electrical property investigations are currently underway in our lab.
![]() | ||
| Fig. 5 I–V plots of films of (A) AUA functionalized graphene sheets and (B) gold-graphene nanocomposites. | ||
000 rpm for 45 min. The resultant black residue obtained could be re-suspended in DMF.
:
1 and 1
:
10) were used to check the changes in dispersibility with the extent of functionalization. The weight of graphene was kept constant for convenience. Briefly, 20 mg of the CTAB assisted exfoliated graphene29 was dispersed in 50 mL of O-dichlorobenzene (ODCB). The solution was purged with nitrogen and preheated to 160 °C. Different amounts of AUA (20 mg and 200 mg, respectively) in 5 mL of ODCB were then added drop-wise to the reaction mixture over a period of 20–30 min. The temperature was maintained at 160 °C for 45 min, after which the product was cooled to room temperature. The suspension was diluted with acetone to flocculate the functionalized graphene sheets. Subsequent washing and centrifugation resulted in the removal of all by-products. The graphene sheets were decanted and dried.
Footnote |
| † Electronic Supplementary Information (ESI) available: Synthesis and characterization details of dodecylazide, hexylazide, 11-azidoundecanol (AUO), micrographs (SEM and TEM images) of the various azide functionalized samples and the statistical analysis of the graphene thickness. See 10.1039/c0nr00547a. |
| This journal is © The Royal Society of Chemistry 2011 |