Enhanced UV detection by transparent graphene oxide/ZnO composite thin films

Abstract

All solution processed transparent thin films of graphene oxide (GO) and zinc oxide (ZnO) in different compositions prepared by a simple two-step chemical synthesis method have been studied for their UV detection properties. The preparation of GO through oxidation of graphite flakes is followed by sol–gel spin coating deposition of the GO–ZnO composite films on glass substrates. The surface morphology, microstructure and composition of the samples have been studied to confirm the formation of composite thin films comprising wurtzite-ZnO nanocrystallites and GO flakes. Optical studies demonstrate that both the transparency and optical band gap of the samples as estimated from wavelength dependent transmittance curves decrease with the increase of GO content in the films, while the charge carrier concentration increases by 5 fold. The in-plane current–voltage (I–V) measurements with two silver electrodes on the GO–ZnO film show a significant enhancement of the photosensitivity in comparison to ZnO films when they are exposed to UV light of different intensities. The response time (t₉₀-response) is nearly three times smaller for GO–ZnO composite films as compared to that of pure ZnO. This improvement is attributed to the defect state modulation and carrier density improvement of the thin films with incorporation of GO, which is encouraging to propel optical, electrical and hence optoelectronics applicability of ZnO composite based transparent devices.

---

Introduction

Graphene is a unique transparent material of sp²-carbon atoms arranged in a hexagonal network in two-dimensions. Recently, graphene itself and its functionalized and/or composite materials have been extensively investigated owing to their highly active surfaces and excellent carrier mobilities besides other unique properties.^1,2 When oxidized, a form called graphene oxide (GO), it becomes more reactive for possessing the abundant oxygen-containing functional groups which facilitate the synthesis process of GO derivatives.³ GO contains a mixture of sp² and sp³ hybridized carbon atoms which are attributed to the creation of new sp² clusters by removing the oxygen atoms providing an easy reduction pathway.⁴ GO additives are often semiconducting, metallic or even polymeric in nature.^5–7 On the other hand, zinc oxide (ZnO) is promising for a broad range of applications such as in UV-light emitters or UV-detectors, chemical sensors and transparent electronics.^8–12 ZnO is a semiconducting material possessing prominent photoconductivity in the UV region of the electromagnetic spectrum because of its wide band gap (3.34 eV at RT).¹³ The efficiency of ZnO based UV detectors, with its various microstructures in low dimensions, is hindered by slow response and weak photoresponsivity due to fast recombination of the photogenerated electron–hole pairs in ZnO.^14,15 The improvement in ZnO based photo-detection is proposed through, (i) the enhanced separation of photo-generated electrons and holes on the ZnO surface which in turn effectively slows down the recombination rate and (ii) the modification of carrier transport by introducing conductive frameworks in ZnO matrix.⁵

In recent times, several efforts have been made to solve this issue by combining ZnO with other functional materials like graphene, reduced graphene oxide (rGO) or GO.^5,15–17 The working mechanism lays behind the controlling of defect states, carrier concentration, carrier mobility and lifetime which are golden parameters in manipulating the photoconductivity of ZnO.¹⁸ Recently researchers have tried to modify the spectral response of ZnO so that it becomes more efficient in the visible spectral regime by altering its defect states.^19,20 The composite structure formation of ZnO with graphene or graphene oxide is a novel, cost effective and efficient approach to control the morphology, surface defect states, band gap and photoresponse of ZnO nanocrystals.^20–22 Moreover, graphene-semiconductor hybrid/composite materials have been identified as suitable candidates for various photonic and optoelectronic applications like photovoltaic cells, optical filters, photodetectors and ultrafast lasers.²³ This is due to the excellent electronic properties of graphene and its prominent capability to store and shuttle photogenerated electrons from the band gap excitation of semiconductors upon light irradiation.²⁴

Graphene–ZnO, rGO–ZnO or GO–ZnO hybrid structures are generally prepared by two methods. In first method, ZnO with required morphology is synthesized and then it is bonded to the surface of graphene or rGO. This is performed mainly by epitaxial growth, chemical deposition or self-assembly technique.^25,26 In the other technique, GO is used as the precursor and the nucleation of ZnO occurs in presence of GO where GO is reduced automatically in a non-toxic way to rGO which inhibits the photogenerated carriers from recombination and transports them between electrodes.²⁷ This is cost effective and easier process to control the morphology, surface defect states and hence the optoelectronic properties of ZnO. Presence of GO, may hinder the preferential c-axis growth of ZnO because of the interaction between the O-groups of GO and Zn²⁺ ions.²⁷ Pan et al. have reported that, with the variation of the added amount of GO (upto 10 wt%) the ZnO morphology changes from a 1D prismatic rod with a pointed tip to a hexagonal tube architecture with a hollow interior and flattened tip, while GO is converted to reduced GO.²⁷ Here the GO sheets play a critical role by tuning the morphology of ZnO and creating oxygen vacancies in the ZnO. These oxygen vacancies are favourable for modulating opto-electronic properties in ZnO. Therefore, GO addition may promote the opto-electronic properties and conductivity of GO–ZnO composite materials through defect manipulation and engineering. Hence it is important to perform systematic study for understanding the modulating effect in such composites.

Chang et al. have demonstrated the enhanced UV photoresponsivity for their ZnO–SWNTs hybrid thin films compared to that of pure ZnO films.²⁸ This improvement was attributed to rapid transportation of UV excited electron–hole (e/h) pairs by conductive SWCNTs. Concerning the effects of graphite allotrope in UV photoresponsivity, in another work, the hybrid photodetector based on polymer–ZnO quantum dots (QDs)–graphene layer shows an amazing photoresponsivity.²⁹ More recently, it is shown that significant improvement of UV photodetection performance of integrated ZnO–rGO composite films can be attained through synergetic effects of rGO in electron/hole recombination inhibition and providing an easy passage way for photogenerated carriers.²⁹ Therefore, in order to obtain rGO–ZnO nanocomposite having such synergistic effect, a good interfacial combination between ZnO and rGO is required to promote the charge transfer between ZnO and rGO. However, very few systematic studies on carbon–ZnO hybrid structure (in a layer by layer system) or composite thin-film synthesized through sonochemical technique for UV detection are reported which brings forward the motivation behind this work.^5,30 Safa et al. synthesized rGO–ZnO composite with maximum 7.5 wt% rGO in ZnO by a solution process followed by reduction using thermal annealing in nitrogen.^5,30 Graphene flakes or rGO however are difficult to mix with ZnO uniformly due to their hydrophobic nature. Hence it remains as a challenge to make uniform composite films based on ZnO and rGO or graphene. But GO being hydrophilic in nature for its available functional groups, is easier to incorporate in ZnO matrix as compared to rGO or graphene. Hence it would be easier to synthesis good quality of composite thin films using GO and ZnO in a simple solution processing technique. A systematic study of the effect of GO addition on the UV detection properties of ZnO in such composite films has yet not been reported. Moreover, it is necessary to optimize the simple synthesis procedure like sol–gel spin coating to fabricate smart composite materials like GO–ZnO as promising transparent UV detector that certainly meets the requirement of cost effective green technology.

Here we report the synthesis and characterization of transparent GO–ZnO composite thin films produced through cost effective and two step all solution processed coating on glass with different amount of GO systematically, demonstrating the modulation of UV photosensitivity. We have also demonstrated the modulation of the generation and transport of charge careers through GO incorporation and reduction into ZnO.

Experimental

GO preparation

Graphite flakes (Sigma-Aldrich, cat #332461, ∼150 μm flakes) were oxidized using following improved method.³¹ In this improved method, a mixture of concentrated H₂SO₄ and H₃PO₄ (180 : 20 ml) was added to a mixture of graphite flakes (1.5 g) and KMnO₄ (9.0 g) in a beaker. The mixed solution was then heated to 50 °C and stirred for 16 h using a magnetic pedal and stirrer. Then it was cooled at first to room temperature and then to ice bath temperature. Afterwards, the solution was kept into ice bath for several hours after mixing with 30% H₂O₂ (3 ml). The colour of the mixture was found to change to brown. For purification, the mixture was washed by rinsing and centrifuging with 10% HCl, deionized water and ethanol separately several times. After filtration and drying under vacuum at room temperature, the graphene oxide (GO) was obtained as a brown powder.

GO–ZnO composite films preparation

0.1 gm of the obtained GO powder was dispersed in 50 ml of isopropyl alcohol (Sigma-Aldrich). The GO solution was vibrated rigorously using an ultrasonic bath sonicator for 2 h to disperse the GO flakes very well. Then it was filtered to remove any unwanted large graphitic lumps. A separate solution was prepared using 0.4 M zinc acetate dihydrate (Zn(CH₃COO)₂·2H₂O) and 50 ml isopropyl alcohol (CH₃CH₃CH–OH). The chemicals were taken in a conical flask and stirred rigorously using a magnetic stirrer at constant temperature, 60 °C. After 5 min, 2.5 ml diethanolamine (DEA) was added to the solution with constant stirring that leads to the formation of transparent solution. The transparent solution was stirred continuously for one hour and then cooled to room temperature. In this transparent solution different amount of GO solution in isopropanol was added and the solution became light brownish to dark brownish in colour depending on the amount of GO in the solution. The solution was mixed homogeneously using ultrasonic bath for 1 h at room temperature.

The mixed solutions containing 0%, 5%, 10% and 25% GO (v/v) were spin-coated onto clean glass substrate to obtain GO–ZnO nanocomposite thin films. The substrate was set to rotate at 3000 rpm for 30 s using a programmable spin coater when 10 drops of the solution was drop casted onto it. Then the substrate having a thin uniform layer of solution was dried in an air oven at 120 °C for 10 min. The coating and drying process was repeated for 6 times to get the required thickness with good uniformity. Then the films were annealed finally at 400 °C in an air oven.

Material characterizations

The surface morphology of the GO–ZnO thin films were recorded by FEI Quanta FEG 200 – high resolution scanning electron microscope (SEM) operated at 20 kV. Transmission electron microscope (TEM) image is recorded using Joel 200 kV on Cu grid. For chemical compositional analysis, EDAX spectra were recorded with the set-up attached to the SEM. A PANalytical X'PERT PRO X-ray diffractometer (XRD) using Cu K_α radiation (0.154 nm) was used to obtain the micro-structural information. Renishaw inVia micro-Raman spectrometer was utilized to record Raman spectra for the films using 514 nm Ar laser. Optical studies were performed by measuring the transmittance in the wavelength region of λ = 300–1100 nm at room temperature using a UV-visible spectrophotometer (Hitachi-U3410). The spectra were recorded with a resolution of λ ∼ 0.07 nm along with a photometric accuracy of ±0.3% for transmittance measurements. FTIR spectra were recorded in the range of 400–4000 cm⁻¹ by using a PerkinElmer FTIR spectrometer by using powder sample obtained after scratching the films.

Photoluminescence measurement was carried out using a 300 W Xe arc lamp as the emission source, Hamamatsu photomultiplier with 3/4 meter monochromator (Oriel Instruments, Model no. 77200) as the detector. Excitation was varied with another 3/4 meter monochromator (Oriel Instruments, model no 77200). An optical chopper with controller (SR 540, Standford Research Systems) and a lock-in-amplifier (SR 530, Standford Research System) were used at the output for minimizing the noise for measuring the output signal. The lock-in-amplifier along with both the detectors and excitation monochromators were monitored with remote control software. The spectra were digitized and collected by a computer. The detection range could be varied from 200 nm to 2000 nm by changing the optical grating in the detector monochromator. All the photoluminescence measurements were recorded at 300 K with 250 nm excitation for detecting the PL peaks in the range 300–700 nm. For I–V characterization pure silver (Ag) dots were deposited on top of all the thin films by high vacuum thermal evaporation system with proper masking under a vacuum of base pressure better than 10⁻⁵ Pa. The thickness and cross-sectional area of Ag top contacts were ∼150 nm and 0.78 × 10⁻² cm⁻², respectively. The in-plane I–V characteristics under UV illumination of fixed wavelength 374 nm with different power densities (150 and 750 μW cm⁻²) of UV illumination were measured by a set-up comprising of constant voltage source and pico-ammeter. The photoresponsivity characterizations were performed at similar illumination of 750 μW cm⁻² and by using 374 nm UV excitation from Xe arc lamp at 25 V. A Keithley 2420 sourcemeter was interfaced with computer through LabVIEW program for this purpose.

Results and discussion

ZnO films with different amount of GO have been prepared under same experimental conditions within laboratory limits, on glass substrates for different characterizations. Different amount of GO solution (0%, 5%, 10% and 25% by volume) are mixed with ZnO solution to get four composite films, namely 0% GO, 5% GO, 10% GO and 25% GO films. All the four films are found to be uniform and highly transparent. The film thickness is measured to be from 80 to 100 micrometer by cross-sectional SEM images.

Fig. 1 represents the FESEM images of ZnO films with different amount of GO incorporated. Fig. 1(a) represents the surface morphology of pure ZnO film without any GO content whereas Fig. 1(b)–(d) show the ZnO films with incorporation of 5%, 10% and 25% GO. The ZnO thin film comprises of uniformly distributed nanocrystallites which are compact in nature. The morphology variation with increasing amount of GO incorporation is evident from the FESEM micrographs. Inset of Fig. 1(c) shows a representative TEM image of 10% GO–ZnO composite film with GO flake and ZnO nanoparticles. Flakes of GO is observed in 25% GO film (Fig. 1(d)) and the inset of Fig. 1(d) shows the cross-sectional SEM image of composite film with 25% GO. Clearly the graphene layers are incorporated into ZnO film.

Fig. 1 FESEM images of thin films of (a) pure ZnO (b) 5% GO–ZnO (c) 10% GO–ZnO (inset: TEM image) and (d) 25% GO–ZnO (inset shows cross-section of the composite layer).

To estimate the chemical compositions of the films EDAX measurements are performed. Fig. 2(a) shows representative EDAX spectra of 0%, 5% and 25% GO films. Beside the peaks for Zn and O, presence of other peaks like Si, Al, Na, Sb and Ca, for all the samples, are due to the glass substrate used. Presence of small C peak for GO–ZnO composite films confirms the incorporation of graphene oxide in ZnO films. The amount of carbon (C) content in 5%, 10% and 25% GO–ZnO films are 4.1, 9.2 and 16.4% (at) respectively as shown in Table 1. Fig. 2(b) indicates the variation of C and O content in GO–ZnO composite films.

Fig. 2 (a) Representative EDAX spectra of pure ZnO, 5% GO–ZnO and 25% GO–ZnO thin films, (b) variation of carbon content (at%) with different amount (v/v %) of GO incorporated in GO–ZnO composite films.

Table 1 Physical parameters of pure ZnO and GO–ZnO composite films with different GO contents

Sample	C amount (at%)	Crystallite size (nm)	Band gap (eV)	ε∞	ωp × 10^14 (S^−1)	N × 10^19 (cm^−3)	t90-Response (s)	t90-Recovery (s)	Photoresponsivity, R (A W^−1)
ZnO	0	27.7	3.27	3.45	2.15	0.63	124.63	95.84	0.005
5% GO–ZnO	4.1	24.2	3.23	3.40	4.35	2.53	56.65	65.47	0.261
10% GO–ZnO	9.2	22.6	3.20	3.29	4.83	3.03	41.94	51.63	0.493
25% GO–ZnO	16.4	21.3	3.17	3.10	5.19	3.30	36.03	44.57	0.827

---

XRD patterns of pure GO, pure ZnO and GO–ZnO composite films are shown in Fig. 3(a). XRD pattern for ZnO film having no GO content shows diffraction peaks due to reflections from (100), (002), (101), (102), (110), (103) and (112) planes that matches very well with the standard data for wurtzite structure of ZnO (JCPDS 36-1451). All the GO–ZnO composite films with different GO incorporation show the same diffraction peaks as pure ZnO which demonstrates that the presence of GO does not affect the crystallinity or the preferred orientation of ZnO. The absence of the GO peak suggests the complete exfoliation and dispersion of GO due to incorporation into the ZnO.³² The XRD pattern of GO has a strong diffraction peak at 2θ ∼ 9.6° due to reflection from (001) plane of graphene oxide³³ which is absent in case of GO–ZnO composite films mainly due to low diffraction intensity of graphene oxide in dispersed state.¹⁶ Efficient dispersion of GO during ultrasonication reduces the diffraction intensity of GO in ZnO.

Fig. 3 (a) X-ray diffraction pattern of ZnO, GO–ZnO and 100% GO films, (b) modulation of crystallite size with carbon content in GO–ZnO composite films (inset shows GO flakes in 25% GO–ZnO film) and (c) XRD pattern of 35% GO–ZnO film showing presence of graphitic phase.

Modulation of lattice constants (a and c) of h-ZnO for different films with GO content are evaluated using following equation for hexagonal space lattice:

The lattice constants as evaluated for the ZnO from eqn (1) are a ∼0.323 nm and c ∼ 0.518 nm that remained invariant for the GO–ZnO nanocomposite thin films. As the lattice constants remain unchanged with the increased GO content in the composites, the possibility of C atom incorporation in the ZnO host matrix by replacing Zn or O is negligible. Although the peak positions of the reflections from hexagonal ZnO lattice is remained same for the GO–ZnO nanocomposites, the slight broadening of the XRD peak indicates the decrease in crystallite size of ZnO nanoparticles. The average crystallite size (D) was extracted from the XRD pattern using Scherrer relation:³⁴

D = Kλ/β cos θ (2)

where K is a constant close to unity (K = 0.94 is used here), β corresponds to the full width at half maximum (FWHM) of the diffraction peak and θ is the Bragg angle of reflection from the (h k l) planes. The average crystallite size decreased with the increase in GO content in the GO–ZnO composite films as tabulated in Table 1. This modulation of crystallite size with C content (at%) in GO–ZnO composite films is presented in Fig. 3(b). This observation could be explained as a synergic effect where the presence of GO flakes influence or disturb the ZnO crystallite formation during GO–ZnO composite film synthesis. Hence, the presence of GO effected the growth of ZnO crystallites without affecting the crystal structure. The inset of Fig. 3(b) shows a FESEM image of 25% GO–ZnO composite film where the exfoliation of GO is evident. The structural details of 35% GO containing ZnO film can be attributed to the formation of graphitic particles (at 2θ ∼ 27.7°) due to the aggregation of graphene sheets as indicated by its XRD pattern as shown in Fig. 3(c) which may arise from the strong van der Waals interaction between the reduced graphene sheets.³⁵ Differently heat treated graphene oxide phases at 2θ ∼ 13.9 and 20.8° are also present.³⁶ Hence the GO amount in ZnO for this study is not increased further.

Raman spectroscopy is non-destructive technique to investigate the bonding environment and defect related disorder, especially in carbon based materials. Fig. 4 shows the Raman spectra of ZnO, GO and two representative GO–ZnO composite films with 5% and 25% GO. The only peak at 438 cm⁻¹ for ZnO thin film arises due to E₂ (high) vibrational mode which is characteristics of wurtzite phase of ZnO.³⁷ Two strong peaks for GO at 1360 cm⁻¹ and 1599 cm⁻¹ are assigned as D-band and G-band, respectively. The D-band is a measure of defects and disorders present in the materials whereas the G-band originates due to E_2g symmetric stretching of sp² hybridized C=C bond.^38,39 The Raman spectra of ZnO–GO composites have both the D and G peaks along with the characteristic vibrational mode of ZnO at 438 cm⁻¹. Although both the D and G peaks are strong in the ZnO–GO composite, the intensity ratio of D-peak and G-peak (I_D/I_G) increases in the composites (0.88 for 5% GO–ZnO and 0.96 for 25% GO–ZnO) compared to that of the pure GO (0.68). The increase in I_D/I_G ratio for ZnO–GO composite films can be attributed to the composite formation of GO with ZnO.³⁹ The intensity of 2D peak around 2930 cm⁻¹ (not shown here) of GO–ZnO composite films is low which indicates the presence of multilayer GO flakes.

Fig. 4 Raman spectra of ZnO, GO and representative GO–ZnO composite films.

Optical properties of GO–ZnO nanocomposite thin films were studied by UV-Vis-NIR spectroscopy. Fig. 5(a) shows the transmittance spectra of ZnO thin films having different GO content. All the films are highly transparent in the visible region of electromagnetic spectrum. The transmittance spectra of pure ZnO shows high transparency (96–98%)in the visible region with a very sharp fall at the band edge indicating the formation of highly crystalline ZnO with less impurity. With the increase in GO percentage in the nanocomposite thin films the absorption is more in the visible region that reduces the transparency upto 85%. Also with the increasing GO content the broadening nature of the band edge with slight red-shift is the salient feature of GO–ZnO nanocomposite films.

Fig. 5 (a) Optical transmission spectra, (b) variation of band gap with C content, (c) plot of ε₁ vs. 1/ω² for different films and (d) variation in carrier concentration with C content in GO–ZnO composite films.

The optical absorption coefficient (α) for a semiconductor as a function of the energy of incident radiation can be expressed as⁴⁰

α = (A/hν) × (hν − E_g)^m (3)

where, A is a constant which depends on the nature of transition whether it is direct or indirect. E_g is the optical band gap. ZnO being a well-known direct band gap semiconductor material, the value of m is 0.5, which is for allowed direct transition. The band gap was estimated by conventional method of extrapolating the linear portion of the plot of (αhν)² versus hν to (αhν)² = 0 (not shown here). The variation of the optical band gap of the GO–ZnO nanocomposite thin films having different GO content is shown in Table 1. The variation of optical band gap with the amount of C present in the GO–ZnO thin films is shown in Fig. 5(b).

The band gap decreased with the increase in GO in the composite films. This observation is consistent with the results obtained by Akhavan et al.⁴¹ for ZnO/CNTs. The modulation of optical transparency as well as the band gap can be attributed to the ZnO–C bond formation in GO–ZnO composite films.^30,41,42 Besides, the presence of GO in the GO–ZnO composite increases the surface charge and an electronic coupling between ZnO nanoparticle and GO sheet that leads to the shift of band gap towards the higher wavelength resulting in narrowing of the band gap.

The GO sheets act as the good electron acceptor in the GO–ZnO nanocomposite whereas the semiconductor nanoparticle ZnO as the good electron donor. The excited electrons in the conduction band transfer to the surface of GO through the π electrons. The bonding between ZnO nanoparticles and GO through the carboxyl groups act as a shorter electron transfer pathway to the GO. ZnO, being a n-type semiconductor the Fermi level is closer to the conduction band and the band gap tends to increase due to Burstein–Moss effect for which the Fermi level shifts to higher level than conduction level when transition of electron occurs from valence band to conduction band. But the presence of GO in the composite suppresses the Burstein–Moss effect as it transfers the excited electrons to the GO sheet through π electrons and empty the conduction band. Therefore, the Fermi energy level is lowered and moved to closer to valence band that reduces the energy band gap.

The charge carrier concentration (N) in a semiconductor is related to the plasma frequency (ω_p) as follows:⁴³

ω_p² = (4πNe²)/(m^*_eε_∞) (4)

where, m^*_e is the effective mass of the electrons and ε_∞ is the limiting value of the high frequency dielectric constant of the semiconductors. Now, the values of m^*_e for different GO–ZnO thin films were estimated from the following relation:⁴⁴

1/m* = 1 + p²/2mE_g (5)

where m* = m^*_e/m_e (m_e is the free electron mass), E_g is the optical band gap of the semiconductor and p = ħG (G being the smallest reciprocal lattice vector). Thus, p is related to a as p ∼ ħ/a, a being the lattice constant.

The real part of dielectric constant (ε₁) as a function of energy of incident radiation were evaluated from the transmittance spectra of the ZnO–GO thin films using KK theory.⁴⁵ In the infrared region, for ωτ ≪ 1:^43,45

ε₁ = ε_∞ − (ε_∞ω_p²)/ω² (6)

where τ is the relaxation time. Thus the values of ω_p and ε_∞ were determined from the slope and the intercept of the linear portion of the ε₁ vs. 1/ω² plots (Fig. 5(c)). Using these values of m^*_e and ω_p, the carrier concentration for GO–ZnO thin films were calculated from eqn (4). The carrier concentration increased (Table 1) with the increase in GO content in ZnO which indicates that films are more conductive. Fig. 5(d) shows the steady increase in career concentration with C at% in the films.

In order to examine the bonding environments in GO, GO–ZnO and pure ZnO films Fourier transform infrared (FTIR) spectroscopy is carried out and has been presented in Fig. 6(a). Typical transverse optical stretching modes of ZnO is observed at 507, 1415 and 1590 cm⁻¹.⁴⁶ In the IR spectrum of GO, presence of different type of oxygen functionalities are confirmed at 3400 cm⁻¹ (O–H stretching vibrations), at 1727 cm⁻¹ (stretching vibrations from C=O), at 1625 cm⁻¹ (skeletal vibrations from un-oxidized graphitic domains), at 1226 cm⁻¹ (C–OH stretching vibrations), and at 1066 cm⁻¹ (C–O stretching vibrations).⁴⁷ FTIR peak of GO–ZnO composite films demonstrates that O–H stretching vibrations observed at 3400 cm⁻¹ is almost disappeared due to deoxygenation/reduction. However, stretching vibrations from C=O at 1727 cm⁻¹ are still observed and C–O stretching vibrations at 1066 cm⁻¹ disappears for GO–ZnO composite films may be due to reduction of GO. Instead two new peaks appear around 1122 and 1282 cm⁻¹ which can be attributed to the Zn–C stretching vibrations. The peak at 890 cm⁻¹ may be arising from SO₃ group of the glass substrate during scratching the films.

Fig. 6 (a) FTIR spectra and (b) photoluminescence spectra of different GO–ZnO composite films.

The modulation in optical luminescence properties of ZnO films n GO incorporation was investigated by PL measurements with excitation wavelength 250 nm at room temperature. PL spectroscopy is well-known efficient and nondestructive technique that gives significant information about the defects present in the film. PL spectrum of ZnO and GO–ZnO composite films in the range 300–700 nm is presented in Fig. 6(b). A strong UV luminescence peak at 384 nm (∼3.23 eV) is observed for ZnO thin film along with a blue emission peak centered at ∼470 nm (∼2.64 eV) and a broad green luminescence peak centered at ∼531 nm (2.34 eV). The origin of UV emission in ZnO is attributed to the near band-edge (NBE) excitonic emission whereas the PL peaks in the visible region are due to the presence of impurities or structural defects in the form of oxygen vacancies, zinc vacancies, oxygen interstitials, and zinc interstitials (known as deep level emission, DLE), respectively.^48–50 With the increase in GO amount in ZnO films, the ratio of the intensity of BE peak at 470 nm to the intensity of excitonic UV emission peak increases, which reflects the increase in defect related transition in the composite film that surpasses the excitonic transition. The decrease in excitonic emission intensity with GO incorporation ensures that the photo induced electron–hole recombination is effectively inhibited in the composite films which essentially increases their photoresponsive behavior. The broad and very less intense green luminescence peak located around 531 nm (∼2.34 eV) can be assigned to the transition from oxygen vacancy related defect states to the valance bands^48,50 which is most intense for pure ZnO and reduces with increased incorporation of GO into ZnO films. Thus GO incorporation passivates the existing oxygen vacancies in the ZnO. In case of ZnO synthesis together with GO, the oxygen vacancies are also known to be passivated by the diffusion of oxygen from graphene oxide to ZnO.^20,27,51

In order to investigate the effect of GO incorporation on the photoconducting properties of ZnO films, the current–voltage (I–V) characteristics were recorded. Fig. 7(a) and (b) represents the I–V characteristics of pure ZnO and GO–ZnO composite films for UV light power density of 150 and 750 μW cm⁻² respectively. The linear nature of I–V curve suggests that GO and ZnO have ohmic contact. GO nano-sheets facilitate the carrier transportation in composite films with ZnO. Watcharotone et al.⁵² also observed a similar decrease in electrical resistance for SiO₂/rGO hybrid. Safa et al. reported reduction in resistivity of rGO films.³⁰ With increased incorporation of GO, the composite films show enhanced current production (Fig. 7) under UV irradiation (λ = 374 nm). This can be attributed to the increase in carrier concentration with increased GO incorporation as estimated from optical measurements.

Fig. 7 Current–voltage characteristics of different GO–ZnO composite films under 374 nm illumination with (a) 150 μW cm⁻² and (b) 750 μW cm⁻² power density (inset: I–V curves with increasing and decreasing voltage for different films).

As shown in the inset of Fig. 7(b), the addition of GO in ZnO films increases the difference in current during increasing and decreasing voltages. This phenomenon is more prominent with under UV illumination of 750 μW cm⁻² as compared to 150 μW cm⁻². This can be explained as the GO incorporation provides conducting pathways in ZnO films and the defect related electron transition is more prominent with increasing amount of GO in ZnO.

The UV wavelength selectivity of photoresponse for the GO–ZnO films lies in between 372 and 377 nm. Fig. 8(a) represents the selectivity of the photoresponse from the 25% GO–ZnO composite film with respect to various illumination wavelengths. UV-detection performance of the pure ZnO and GO–ZnO composite films under the same power density of illumination (750 μW cm⁻² at 25 V) has been compared in Fig. 8(b) for two cycles. Under the UV illumination (374 nm), photocurrent increases to a certain value and then saturates. When the UV light is turned off, the photocurrent exponentially decreases to its initial value. The response to UV illumination is more prominent for the films with increased amount of GO in ZnO. In order to compare the effect of GO incorporation into ZnO on the efficiency of UV response, some physical parameters, such as response time (t₉₀-response) and recovery time (t₉₀-recovery) are calculated. The time interval over which resistance acquires 90% of its final value is called response time whereas recovery time means the time interval in which resistance reduces to 10% of the fixed value at plateau. These values are shown in Table 1. Fig. 8(c) demonstrates the exponentially decreasing response and recovery time with C content (at%) in the GO–ZnO composite thin films. The time of response and recovery of samples were significantly improved by increasing the GO content. The response time of 25% GO–ZnO composite is only about 36 s and the recovery time is 45 s. It is worthy to note that the response and recovery speed of the previous reported conventional ZnO based UV detectors is from several seconds to several minutes.⁵³ Presence of GO in ZnO helps as the conducting medium to increase the photocurrent in GO–ZnO films. As increased GO incorporation increases the defect density in GO–ZnO films, the UV interaction is more prominent in the composite films.

Fig. 8 (a) Selectivity in UV photoresponse of 25% GO–ZnO film with different illumination wavelengths, (b) UV photoresponsivity of ZnO and GO–ZnO films, (c) variation of photo response and recovery time with different C content in different films.

Photoresponsivity (R) is another important parameter to ascertain the efficiency of the UV detectors which is defined as the ratio between photocurrent (I) and incident UV power (P).⁵⁴ Hence,

R = I/P (7)

Table 1 shows that the UV photosensitivity of ZnO gradually increases from 0.0051 to 0.82 A W⁻¹ with incorporation of GO.

The increase in the photoresponsivity and the modification of response or recovery time with GO incorporated transparent ZnO composites thin films can be explained by UV induced carrier generation and transport through formation of ZnO–C bonding. The Fermi level at the ZnO–C interface is matched through the generation of a space charge region (Fig. 9). The UV induced carriers are generated in presence of the essential driving force obtained from the electrical field corresponding to the charge space.^41,55 Besides, the electron affinity of GO is low as compared to pure ZnO (Fig. 9(a)) which drives the UV excited electrons from the ZnO conduction band to the surface of GO (through ZnO–C bonding interactions) as shown in Fig. 9(b). On the other hand, the graphene sheets offer a low resistive pathway in the ZnO thin film which might promote the direct transportation of induced charge carriers and hence the photocurrent considerably increases.^41,55,56 The ZnO films containing more than 35% (v/v) GO were found to perform worse than ZnO having 25% GO in the UV photodetection. Similar observations have been made by Akavan et al. where 5 and 10 wt% of CNT content in CNT-doped ZnO thin films were found to be crucial in degrading E. coli bacteria through UV illuminated photocatalytic mechanism.⁴¹ Recently Safa et al. have reported rGO doped ZnO films obtained by sol–gel technique.³⁰ They UV sensitivity of 2.5 is received with t₉₀-response time of 47 s. Besides, Basak et al. also used sol–gel technique to synthesize ZnO film which demonstrates 0.040 A W⁻¹ photoresponsivity with 350 nm UV light.⁵⁴ Xu et al. also used similar technique for Al dopes ZnO and used 350 nm UV for photo-generated current of 58.05 μA at a bias of 6 V.⁵⁷ Therefore, the GO–ZnO composite films studied here shows a simple solution processed pathway for advancement of ZnO film based UV detection capability.

Fig. 9 Mechanism of (a) UV detection for ZnO and (b) enhanced UV detection for GO–ZnO composite films.

Conclusions

Highly transparent graphene oxide–zinc oxide (GO–ZnO) composite thin films have been synthesized by simple and cost-effective all solution processed spin coating technique. In the presence of GO, the wurtzite ZnO nanocrystallite size reduces but the hexagonal crystal structure of ZnO remains unaffected. The narrowing of optical band gap in GO–ZnO composite is attributed to the transfer of photoexcited electrons in the conduction band to the surface of GO across the GO–ZnO interface through carboxyl groups that provides the shorter electron transfer pathway to the GO. FTIR study indicates that GO is partially reduced in GO–ZnO composite films. The photoluminescence studies on the GO–ZnO composite films reveal that the relative intensity of excitonic transition decreases compared to the intensity of blue emission peak as the amount of GO increases which effectively inhibit the electron–hole recombination and enhance UV detection capability. The in-plane I–V characteristics of GO–ZnO composite thin films under UV illumination of wavelength 375 nm show that the films become more photosensitive with the increase in GO content. The UV photo-response time significantly decreases from 125 s to 36 s as well as the recovery time reduces from 96 s to 45 s for the GO–ZnO composite films as compared to pure ZnO film. This modulation in UV detection and photoresponsivity of GO–ZnO composite thin films are very much unique and technologically significant for ZnO based transparent opto-electronic modules.

Acknowledgements

One of the authors, RNG wishes to thank the Department of Science and Technology (DST), Government of India, for financial support (File No. YSS/2014/000038). Authors would like to thank the Staffs of Department of Physics, Presidency University and Department of Instrumentation Science, Jadavpur University. SVB would like to acknowledge Nanotechnology Research Centre, SRM University, Chennai for the help with material characterization. We thank A. Kumar and A. Datta of Purdue University, USA for discussion.

Notes and references

1. S. Stankovich, D. A. Dikin, G. H. B. Dommett, K. M. Kohlhaas, E. J. Zimney, E. A. Stach, R. D. Piner, S. T. Nguyen and R. S. Ruoff, Nature, 2006, 442, 282–286.
2. X. Huang, X. Qi, F. Boey and H. Zhang, Chem. Soc. Rev., 2012, 41, 666–686.
3. J. Li, X. Zeng, T. Ren and E. V. Heide, Lubricants, 2014, 2, 137–161.
4. K. P. Loh, Q. L. Bao, G. Eda and M. Chhowalla, Nat. Chem., 2010, 2, 1015–1024.
5. S. Safa, R. Sarraf-Mamoory and R. Azimirad, J. Sol-Gel Sci. Technol., 2015, 74, 499–506.
6. T. S. Sreeprasad, S. M. Maliyekkal, K. P. Lisha and T. Pradeep, J. Hazard. Mater., 2011, 186(1), 921–931.
7. Y. Mao, S. Wen, Y. Chen, F. Zhang, P. Panine, T. W. Chan, L. Zhang, Y. Liang and L. Liu, Sci. Rep., 2013, 3(2508), 1–7.
8. V. Avrutin, D. Silversmith and H. Morkoc, Proc. IEEE, 2010, 98(7), 1269–1280.
9. G. S. Devi, V. B. Subrahmanyam, S. C. Gadkari and S. K. Gupta, Anal. Chim. Acta, 2006, 568, 41–46.
10. C. H. Wang, X. F. Chu and M. W. Wu, Sens. Actuators, B, 2006, 113, 320–323.
11. J. I. Song, J. S. Park, H. Kim, Y. W. Heo, J. H. Lee, J. J. Kim and B. D. Choi, Appl. Phys. Lett., 2007, 90, 022106.
12. K. Nomura, H. Ohta, A. Takagi, T. Kamiya, M. Hirano and H. Hosono, Nature, 2004, 432(7016), 488.
13. R. N. Gayen, S. Majumder, S. Varma and A. K. Pal, Mater. Chem. Phys., 2010, 123, 138–146.
14. S. Xu and Z. L. Wang, Nano Res., 2011, 4(11), 1013–1098.
15. Y. Yang, L. Ren, C. Zhang, S. Huang and T. Liu, ACS Appl. Mater. Interfaces, 2011, 3(7), 2779–2785.
16. X. Liua, L. Pana, Q. Zhaob, T. Lva, G. Zhua, T. Chena, T. Lua, Z. Suna and C. Sun, Chem. Eng. J., 2012, 183, 238–243.
17. O. Akhavan, Carbon, 2011, 49, 11–18.
18. X. Bai, L. Wang, R. Zong, Y. Lv, Y. Sun and Y. Zhu, Langmuir, 2013, 29(9), 3097–3105.
19. P. Kumar, L. S. Panchakarla, S. V. Bhat, U. Maitra, K. S. Subrahmanyam and C. N. R. Rao, Nanotechnology, 2010, 21, 385701.
20. M. K. Kavitha, P. Gopinathb and H. John, Phys. Chem. Chem. Phys., 2015, 17, 14647–14655.
21. O. Akhavan, ACS Nano, 2010, 4, 4174–4180.
22. D. I. Son, B. W. Kwon, D. H. Park, W. S. Seo, Y. Yi, B. Angadi, C. L. Lee and W. K. Choi, Nat. Nanotechnol., 2012, 7, 465–471.
23. Q. Bao and K. P. Loh, ACS Nano, 2012, 6, 3677–3694.
24. G. Williams and P. V. Kamat, Langmuir, 2009, 25, 13869–13873.
25. K. Chung, C. H. Lee and G. C. Yi, Science, 2010, 330, 655–657.
26. S. Ameen, M. S. Akhtar, M. Song and H. S. Shin, ACS Appl. Mater. Interfaces, 2012, 4, 4405–4412.
27. X. Pan, M. Q. Yang and Y. J. Xu, Phys. Chem. Chem. Phys., 2014, 16, 5589–5599.
28. J. Chang, C. K. Najeeb, J. Lee, M. Lee and J. Kim, J. Phys. D: Appl. Phys., 2011, 44, 095101–095106.
29. H. Y. Yang, D. I. Son, T. W. Kim, J. M. Lee and W. I. Park, Org. Electron., 2010, 11, 1313–1317.
30. S. Safa, R. S. Mamoory and R. Azimirad, Adv. Mater. Res., 2014, 829, 577–582.
31. D. C. Marcano, D. V. Kosynkin, J. M. Berlin, A. Sinitskii, Z. Sun, A. Slesarev, L. B. Alemany, W. Lu and J. M. Tour, ACS Nano, 2010, 4(8), 4806–4814.
32. S. V. Kumar, N. M. Huang, N. Yusoff and H. N. Lim, Mater. Lett., 2013, 93, 411–414.
33. S. Pei, J. Zhao, J. Du, W. Ren and H. M. Cheng, Carbon, 2010, 48, 4466–4474.
34. R. Paul, N. Bhadra, A. K. Mukhopadhyay, R. Bhar and A. K. Pal, Eur. Phys. J.: Appl. Phys., 2014, 68(20402), 1–11.
35. M. Fang, K. Wang, H. Lu, Y. Yanga and S. Nutt, J. Mater. Chem., 2009, 19, 7098–7105.
36. D. W. Wang, A. Du, E. Taran, G. Q. Lu and I. R. Gentle, J. Mater. Chem., 2012, 22, 21085–21091.
37. R. N. Gayen and R. Paul, Nano-Structures & Nano-Objects, 2016 DOI:10.1016/j.nanoso.2016.03.007.
38. A. C. Ferrari, Solid State Commun., 2007, 143(1–2), 47–57.
39. A. N. Fouda, A. B. E. Basaty and E. A. Eid, Nanoscale Res. Lett., 2016, 11(13), 1–8.
40. J. C. Manifacier, M. De Murcia, J. P. Fillard and E. Vicario, Thin Solid Films, 1977, 41, 127–135.
41. O. Akhavan, R. Azimirad and S. Safa, Mater. Chem. Phys., 2011, 130, 598–602.
42. O. Akhavan, R. Azimirad, S. Safa and M. M. Larijani, J. Mater. Chem., 2010, 20, 7386–7392.
43. R. Paul, S. Hussain and A. K. Pal, Appl. Surf. Sci., 2009, 255, 8076–8083.
44. D. Bhattacharya, S. Chaudhuri and A. K. Pal, Vacuum, 1995, 46, 1–3.
45. S. R. Bhattacharyya, R. N. Gayen, R. Paul and A. K. Pal, Thin Solid Films, 2009, 517, 5530–5536.
46. K. S. Babu, A. R. Reddy, C. Sujatha, K. V. Reddy and A. N. Mallik, J. Adv. Ceram., 2013, 2(3), 260–265.
47. Y. Xu, H. Bai, G. Lu, C. Li and G. Shi, J. Am. Chem. Soc., 2008, 130, 5856–5857.
48. K. Matsuyama, K. Mishima, T. Kato and K. Ohara, Ind. Eng. Chem. Res., 2010, 49(18), 8510–8517.
49. Q. P. Luo, X. Y. Yu, B. X. Lei, H. Y. Chen, D. B. Kuang and C. Y. Su, J. Phys. Chem. C, 2012, 116(14), 8111–8117.
50. R. N. Gayen and R. Paul, Thin Solid Films, 2016, 605, 248–256.
51. D. H. Yoo, T. V. Cuong, S. Lee, W. S. Hwang, W. J. Yoo, C. H. Hong and S. H. Hahn, J. Phys. Chem. C, 2014, 118, 17638–17642.
52. S. Watcharotone, D. A. Dikin, S. Stankovich, R. Piner, I. Jung, G. H. B. Dommett, G. Evmenenko, S. E. Wu, S. F. Chen, C. P. Liu, S. B. T. Nguyen and R. S. Ruoff, Nano Lett., 2007, 7, 1888–1892.
53. X. W. Fu, Z. M. Liao, Y. B. Zhou, H. C. Wu, Y. Q. Bie, J. Xu and D. P. Yu, Appl. Phys. Lett., 2012, 100(22), 223114.
54. D. Basak, G. Amin, B. Mallik, G. K. Paul and S. K. Sen, J. Cryst. Growth, 2002, 256, 73–77.
55. D. Shao, L. Qin and S. Sawyer, IEEE Photonics J., 2012, 4(3), 715–720.
56. Q. P. Luo, X. Y. Yu, B. X. Lei, H. Y. Chen, D. B. Kuang and C. Y. Su, J. Phys. Chem. C, 2012, 116(14), 8111–8117.
57. Z. Q. Xu, H. Deng, J. Xie, Y. Li and X. T. Zu, Appl. Surf. Sci., 2006, 253, 476–479.

---