Materials Science in Semiconductor Processing 24 (2014) 193199

Contents lists available at ScienceDirect

Materials Science in Semiconductor Processing

journal homepage: www.elsevier.com/locate/mssp

Band gap reduction of titania thin films using

graphene nanosheets
Haroon Mahmood n, Amir Habib, Mohammad Mujahid, Muhammad Tanveer,
Sofia Javed, Asif Jamil
School of Chemical and Materials Engineering, National University of Sciences and Technology, Islamabad, Pakistan

a r t i c l e i n f o

abstract

Available online 5 April 2014

Titania possesses a band gap of ! 3.2 eV, whereas the work function of graphene ! 4.5 eV
exists near the lowest unoccupied molecular orbital (LUMO) of titania. The photoexcitation of
electrons in titania (TiO2) is possible only under the ultraviolet radiation. If such conversion is
required in visible light range, band gap has to be lowered below 3.1 eV. This research work
hence aims at lowering the band gap of titania thin films by incorporation of graphene
nanosheets. Graphene was synthesized using modified Hummer's method. Oxidation of
graphite was confirmed by X-ray diffraction (XRD) revealing a peak at 11.81 with interplanar
distance of 7.5 and energy dispersive X-ray spectroscopy (EDS) analysis. Atomic force
microscopy (AFM) analysis revealed exfoliation of single layer graphene oxide nanosheets
having thickness of 0.61.0 nm. Fourier transform infrared spectroscopy (FT-IR) analysis
confirmed the reduction of graphene oxide by revealing removal of carbonyl group (CQO).
Titania sol was obtained using hydrolysis and condensation of titanium tetraisopropoxide. The
TiO2graphene composite was prepared by liquid phase dispersion of graphene oxide into
titania sol and reduction of deposited films was carried out by exposing hydrazine hydrate
vapors. In the second method, nanoparticles of reduced graphene oxide were dispersed in
titania sol; thin films were deposited on indiumtin oxide (ITO) coated glass using spin coating
and subsequently heat treated. A scanning electron microscope (SEM) was used to reveal
dispersion of graphene nanosheets among homogenously distributed titania nanoparticles of
uniform size distribution (! 3050 nm). Surface roughness of about 16 nm was observed by
AFM topographic images. The highest occupied molecular orbital (HOMO) and LUMO levels
were calculated by cyclic voltammetry and subsequent band gap was calculated which came
out to be as low as 2.9 eV. TiOC chemical bonding between titania and graphene sheets
resulted in enhanced electron transport in the obtained composite films.
& 2014 Elsevier Ltd. All rights reserved.

Keywords:
Titania thin film
Graphene
Band gap
Photovoltaics

1. Introduction
Titania, either in anatase or rutile phase, is considered to
be one of important materials for various applications on
account of its property of a photoexcitation phenomenon.
n
Correspondence to: SCME, National University of Sciences and
Technology (NUST), Sector H-12, Islamabad, Pakistan.
Tel.: 92 51 90855204.
E-mail address: haroon88@gmail.com (H. Mahmood).

http://dx.doi.org/10.1016/j.mssp.2014.03.038
1369-8001/& 2014 Elsevier Ltd. All rights reserved.

Among the applications, efficiency of photocatalytic devices

and photovoltaic solar cells could be enhanced if the band
gap of the films made up of titania may be narrowed down
[1,2]. Till now, several methods have been reported in
which doping of titania is done to reduce the band gap
and enhance the photocurrent by preventing photoexcitons recombination [35].
Carbon nanostructures have superior electrical properties and hence they can provide a medium for transport
of electrons in semiconductor thin films. Use of carbon

194

H. Mahmood et al. / Materials Science in Semiconductor Processing 24 (2014) 193199

nanotubes in TiO2, CdS and CdSe thin films has resulted in

increase in the efficiency of photocurrent in such structures
[68]. Graphene, a 2D carbon nanostructure, has high
electron mobility (exceeding 2000 cm2/V). Incorporation
of graphene has shown enhancement in the charge transfer
rate of electrons. This increased charge transfer rate
resulted in increase in the photocurrent by preventing
recombination of photo-generated electronhole pair
because of interactions of graphene [911].
Originally, titania possesses a band gap of ! 3.2 eV.
Lowering of band gap is essential for maximum harvesting
of sunlight and thus improvement of efficiency of devices
hence produced. Graphene, having a work function of
4.5 eV, is explored in this research work for its effect on
the band gap of titania thin films and also on photocurrent
generated in such structures. In the past, much of the
research have been performed for enhancement of photocatalytic activity of titania using graphene nanosheets,
such as preparing composites through thermal reaction
[5], solvothermal reaction [9], hydrothermal method
[10,12,13], solgel method [14] etc. These methods have
shown success in increment of photocatalytic activity by
the composite; however none of the work actually
reported the effect of graphene on the band gap of the
semiconductor material. This work was hence developed
to report the change in electrical properties of titania by
incorporation of graphene nanosheets in two forms, i.e.,
graphene oxide (GO) and reduced graphene oxide (RGO).
Solgel route was employed to obtain titania sol and
graphene was dispersed in the sol. Thin films were
obtained using spin-coating composite sol on the glass
and ITO substrates. After heat treatment of the films,
characterization and testing were performed.
2. Experimental work
Graphene oxide was synthesized from flake graphite
(average100 mesh (Z75% min), SigmaAldrich) by a
modified Hummers method reported elsewhere [15]. All
other chemicals were of analytical reagent grade and used
without further purification. Sonication of the resultant
solid was carried out in water to exfoliate graphene oxide
nanosheets. Chemical reduction of GO was carried out by
adding hydrazine hydrate (1.00 ml, 32.1 mmol) (Merck) in
the exfoliated GO dispersion and the solution was heated
at 100 1C for 24 h under a water-cooled condenser for 24 h.
The reduced GO slowly precipitated out as a black solid.
The subsequent nanoparticles were washed with water
and methanol and finally vacuum dried.
For thin film fabrication, titania sol was prepared by
hydrolysis and condensation of 97% titanium tetraisopropoxide (TTIP) (Aldrich) in the acidic mixtures of glacial
acetic acid and 37% hydrochloric acid. The solution, whose
composition is shown in Table 1, was heated at 60 1C
under vigorous stirring for 2 h [16].
Two approaches are used for composite thin film
processing. In one method, graphene oxide was mixed
with various concentrations (Table 2) in titania sol and
resulting spin coated composite films were exposed to an
atmosphere that has been saturated with vapor of hydrazine hydrate at 100 1C for 2 min. In another approach,

Table 1
Titania sol composition.
Chemical

Amount (ml)

TTIP
Glacial acetic acid
HCl
Water

10
5
1.4
83.6

Table 2
TitaniaGO composite sols.
Precursor
composite sol

Titania
sol (ml)

GO (5 mg ml $ 1)
(ml)

TGO-1
TGO-2
TGO-3
TGO-4
TGO-5

50
50
50
50
50

0
2
4
6
8

already reduced graphene oxide was mixed in various

amounts (Table 3) in titania sol and films were prepared
using a spin coating method. In both cases, after addition
of graphene precursor in titania sol, the solution was
sonicated for 30 min before film deposition. Films were
pre-annealed at 80 1C for 2 min followed by annealing at
450 1C for 2 h at 2 1C/min of heating rate.
3. Results and discussion
Fig. 1 shows a comparison of the X-ray diffractograms
of precursor graphite and graphite oxide. Pristine graphite
shows a high degree of crystallinity with a sharp diffraction peak at 226.471, that has an interplanar distance of
3.37 of (002) diffraction plane. However the degree of
crystallinity of graphite oxide is very low with a minor
broad-diffraction peak at 11.81, which is due to a conserved
graphene-like honeycomb lattice in graphite oxides [17].
Hence the shift of the diffraction peak position from 26.471
to 11.81 is due to increase in the interplanar distance from
3.37 to 7.5 . This is confirmed from Braggs equation
which inversely relates interplanar distance d to the
diffraction angle 2 [18]. The increase in interplanar
distance is due to trapping of oxygen atoms between the
graphene sheets in graphite structure [19].
Scanning electron microscope (SEM) images (Fig. 2a
and b) reveal exfoliation of graphite oxide into individual
(single or multi-layer) graphene sheets. A closely packed
tilting of graphene oxide platelets having an even contrast was detected on the substrate. The lateral dimensions
of the sheets ranged from several hundred nanometers to
several micrometers. Fig. 2c shows the thickness of exfoliated graphene oxide sheets investigated by atomic force
microscopy in a tapping mode. The graphene oxide height
profiles typically ranged between 0.6 nm and 1 nm which
is reported to be single layer graphene oxide sheet [20].
From the AFM topographical image (Fig. 3), graphene
nanosheets can be seen lying on glass slide. Overall, roughness (RMS) is about 0.5 nm indicating that the features are
very small in dimension along the z axis. The only factor

H. Mahmood et al. / Materials Science in Semiconductor Processing 24 (2014) 193199

contributing to the roughness values is intrinsic height of

carbon sheets and their lifting up against the base plane due
to crossovers of neighboring flakes. Line plots and their
Table 3
TitaniaRGO composite sols.
Precursor composite sol

Titania sol (ml)

RGO (mg)

TRGO-1
TRGO-2
TRGO-3
TRGO-4

50
50
50
50

0
5
10
15

Fig. 1. Comparison of XRD diffractographs of graphite and graphite oxide.

195

analysis support the above statement. Three lines have very

small difference in roughness values due to difference in
regions selected for the analysis. Red line has maximum
roughness of 0.5 nm and has maximum height feature of
about 3.4 nm. It is because the line has passed over two
flakes and got lifted upward due to many crossovers and
buckling. While some of the portion of green line indicates
that it has passed over single sheet or some bilayers as
evident from the line plot and corresponding z values. More
interesting part is the blue line which has only passed over a
monolayer graphene flake and has minimum roughness
value of 0.2 nm. The maximum height feature is observed
to be 1.12 nm relative to mean plane of overall image, while
the z plot of this line indicates that the height of flake is
even less than 1 nm.
Reduction of graphene oxide was confirmed using FT-IR
and Raman Spectroscopy. Fig. 4a shows FT-IR spectra
which indicates that the attached H2O molecules and
much of the oxide groups (Carbonyl CQO) of GO are
detached after chemical reduction by hydrazine hydrate.
Nevertheless, hydroxyl group (OH) and cyanide (CN)
remain on GO after chemical reductions due to poor
reactivity of hydrazine hydrate with these groups. Fig. 4b
shows Raman spectrum of the reduced GO which contains
both G and D bands (at 1589 and 1325 cm $ 1, respectively).
G band is related with stretching of bond of the sp2 carbon
pairs in both rings and chains, whereas the D band is
because of the breathing mode of aromatic rings with
dangling bonds in plane terminations. Furthermore, the
intensity ratio of the D band to the G band is usually
employed to determine the disorder degree of structure
for carbon. High D/G ratios obtained in Raman spectra of
GO indicate a high degree of exfoliation/disorder [21].

Fig. 2. (a and b) SEM images of single layer and multi-layer graphene oxide and (c) AFM image with height profile of graphene oxide nanosheets.

196

H. Mahmood et al. / Materials Science in Semiconductor Processing 24 (2014) 193199

Fig. 3. AFM topographical image of graphene nanosheets. (For interpretation of the references to color in this figure, the reader is referred to the web
version of this article.)

Fig. 4. (a) FTIR spectra of graphene oxide (GO) and reduced graphene oxide (RGO) and (b) Raman spectra of reduced graphene oxide.

Fig. 5a shows the scanning electron microscope (SEM)

images of 110125 nm thick graphenetitania composite
thin films. SEM images (Fig. 5b and c) reveal embedding of
titania nanoparticles of size ranging from 30 to 50 nm
embedded on graphene nanosheets. The film produced

has homogeneity of titania nanoparticles to appreciable

lengths. Some pinholes and agglomerated nanoparticles
have been observed on films as well.
AFM analysis confirms the dispersion of titania nanoparticles and embedding of titania nanoparticles on

H. Mahmood et al. / Materials Science in Semiconductor Processing 24 (2014) 193199

197

Fig. 5. (a) Thickness of graphenetitania composite thin film, and (b) low magnification image of graphene embedded in titania thin film. (c) High
magnification image of titania particles with size ranging from 30 nm to 50 nm.

graphene nanosheets (Fig. 6). Homogenous structure as

seen in SEM images is confirmed by AFM analysis of titania
nanoparticles with height profile ranging 1525 nm.
Using cyclic voltammetry (CV), HOMO/LUMO levels
were evaluated by the standard formula EHOMO $ Eonset
ox
$4:8 eV; ELUMO $Eonset
red $ 4:8 eV which gives us the
band gap information. The device structure for the CV
measurement consists of three electrodes: a working electrode which is the test sample, a counterelectrode made of a
platinum wire while saturated calomel as the reference
electrode. The range of current was set to auto while voltage
was varied from $2.5 V to 2.5 V. An assessment of the
obtained band gap and current of synthesized films is shown
in Fig. 7a and b.
The films containing reduced graphene oxide (RGO) in
titania matrix exhibited enhanced reduction of band gap
as compared to films developed using dispersion of
graphene oxide (GO) in titania sol. This can be related to
the higher conductivity of reduced graphene oxide as
compared to graphene oxide reduced after film fabrication. The lowest band gap found to be 2.90 eV for TRGO-4.
The overall trend for both types of composites films (GO
and RGO embedded films) is linear, apart from couple of
exceptions i.e. TGO-8 and TRGO-3. The distinctiveness
could be either because of defects in the thin films or
improper embedment of graphene nanosheets. Moreover
in both graphs (Fig. 7a and b), it is also clear that electron
mobility or current increases linearly with increasing
amount of graphene in the composite. This also verifies
the hypothesis that with increasing graphene amount, the
movement of electron becomes easier due to electron
capture as well as higher conductivity of graphene.

Fig. 8 displays a model diagram which shows the

synergistic effect of graphene on the electron transport
mechanism generated by titania nanoparticle. The work
function of graphene ranges between 4.2 and 4.5 eV [22].
After getting excited, the electron jumps from HOMO to
LUMO level within titania. Since LUMO (3.23.5 eV) of
titania is close to work function of graphene, the excited
electron in the LUMO level finds an intermediate energy
level and gets trapped hence causing charge separation.
This prevents recombination of electron and holes. Thus
effective electron transfer from lower band gap film
increases the current generation. This complements the
results obtained by cyclic voltammetry as shown above.
4. Conclusion
In this study, graphenetitania composite thin films were
synthesized using a spin coating technique. The dispersion of
graphene nanosheets in titania thin films, homogeneity of
films, HOMO/LUMO levels and band gap were evaluated. XRD
results indicated the oxidation of graphite while AFM analysis
revealed the exfoliation of graphite oxide into single or
multilayer graphene oxide nanosheets. Using SEM, the length
of graphene nanosheets ranged between several hundred
nanometers to several micrometers. Raman Spectroscopy
revealed higher D/G ratio suggesting high degree of exfoliation/disorder of graphene nanosheets. Reduction of graphene
oxide nanosheets was confirmed by FTIR displaying removal
of functional groups containing oxygen. Using SEM, graphenetitania composite thin films were analyzed which
revealed dispersion of graphene nanosheets. Single layers
graphene was fully embedded in the titania matrix while

198

H. Mahmood et al. / Materials Science in Semiconductor Processing 24 (2014) 193199

Fig. 6. AFM images of titania thin film with graphene nanosheets.

Fig. 7. Comparison of band gap of composite films: (a) titaniagraphene oxide and (b) titania-reduced graphene oxide.

H. Mahmood et al. / Materials Science in Semiconductor Processing 24 (2014) 193199

199

Fig. 8. Model energy diagram effect of graphene on electron transport.

multilayered graphene or irregular shaped graphene did not

fully embed in the film due to having feature size greater than
film thickness. Both SEM and AFM confirmed the titania
nanoparticles size in the range of 3050 nm. Cyclic voltammetry confirmed the effect of graphene on reduction of band
gap and enhancement of current by titania thin film. The
lowest band gap obtained was 2.9 eV of graphenetitania
composite thin film as compared to pure titania thin film
having 3.18 eV. The work function of graphene being closer to
LUMO energy level of titania and superior electrical properties
of graphene nanosheets resulted in the trapping of electrons
generated by titania nanoparticles which resulted in increased
efficiency of electrons generation and lowering of band gap.
References
[1]
[2]
[3]
[4]

A.L. Linsebigler, G. Lu, J.T. Yates Jr., Chem. Rev. 95 (1995) 735758.
B. ORegan, M. Gratzel, Nature 353 (1991) 737740.
X. Chen, S.S. Mao, Chem. Rev. 107 (2007) 28912959.
G. Liu, L. Wang, H.G. Yang, H.-M. Cheng, J. Mater. Chem. 20 (2010)
831843.
[5] Y. Zhang, C. Pan, J. Mater. Sci. 46 (2011) 26222626.
[6] A. Kongkanand, R.M. Dominguez, P.V. Kamat, Nano Lett. 7 (2007)
676680.

[7] F. Vietmeyer, B. Seger, P.V. Kamat, Adv. Mater. 19 (2007) 29352940.

[8] M. Wang, S. Kumar, A. Lee, N. Felorzabihi, L. Shen, F. Zhao, et al.,
J. Am. Chem. Soc. 130 (2008) 94819491.
[9] K. Zhou, Y. Zhu, X. Yang, X. Jiang, C. Li, New J. Chem. 35 (2011)
353359.
[10] H. Zhang, X. Lv, Y. Li, Y. Wang, J. Li, ACS Nano 4 (2009) 380386.
[11] Y.H. Ng, I.V. Lightcap, K. Goodwin, M. Matsumura, P.V. Kamat, J. Phys.
Chem. Lett. 1 (2010) 22222227.
[12] S.D. Perera, R.G. Mariano, K. Vu, N. Nour, O. Seitz, Y. Chabal, et al.,
ACS Catal. 2 (2012) 949956.
[13] Y. Zhang, Z.-R. Tang, X. Fu, Y.-J. Xu, ACS Nano 4 (2010) 73037314.
[14] X.-Y. Zhang, H.-P. Li, X.-L. Cui, Y. Lin, J. Mater. Chem. 20 (2010)
28012806.
[15] D.C. Marcano, D.V. Kosynkin, J.M. Berlin, A. Sinitskii, Z. Sun,
A. Slesarev, et al., ACS Nano 4 (2010) 48064814.
[16] W.S. Tung, W.A. Daoud, S.K. Leung, J. Colloid Interface Sci. 339 (2009)
424433.
[17] G. Wang, J. Yang, J. Park, X. Gou, B. Wang, H. Liu, et al., J. Phys. Chem.
C 112 (2008) 81928195.
[18] B.D. Cullity, S.R. Stock, Elements of X-ray Diffraction, Prentice-Hall,
Upper Saddle River; NJ, 2001.
[19] A. Lerf, H. He, M. Forster, J. Klinowski, J. Phys. Chem. B 102 (1998)
44774482.
[20] G. Eda, G. Fanchini, M. Chhowalla, Nat. Nanotechnol. 3 (2008)
270274.
[21] H. Wang, Y.H. Hu, Ind. Eng. Chem. Res. 50 (2011) 61326137.
[22] R. Czerw, B. Foley, D. Tekleab, A. Rubio, P. Ajayan, D. Carroll, Phys.
Rev. B 66 (2002) 033408.