Kusuma 2018
Kusuma 2018
Kusuma 2018
A R T I C L E I N F O A B S T R A C T
Keywords: The future of third generation Quantum dot sensitized solar cells (QDSSCs) can be optimistic if concerns of
Graphene nanoribbons charge recombination and stability of cells can be worked upon. In this work, we report the role of Graphene
Conducting edges materials in its different dimension in overcoming recombination and enhancing power conversion efficiency
Band gap (PCE). Composite of Graphene oxide nanoribbons (GNRs) synthesized by oxidative unzipping of multi walled
SILAR-QDSSC
carbon nanotubes (MWCNTs) was used as an effective electron transport layer with TiO2 as photo anode in
Carbon nanotubes
1D quasi structure
QDSSCs. The best performing TiO2-GNR showed a significant enhancement in PCE of 3.7% for CdS based
quantum dots, apart from increased efficiency reported earlier. Investigation of design of anode material reveals
that the enhanced surface area and active sites of partially conducting GNRs facilitated uniform deposition of
Titania particles (TiO2) with less aggregation, excellent mobility, presence of organic functional group for anchor
of TiO2, its 1D quasi structure add to enhanced electron transport (less resistance).
⁎
Corresponding author.
E-mail address: [email protected] (R.G. Balakrishna).
https://doi.org/10.1016/j.solmat.2018.01.039
Received 6 September 2017; Received in revised form 24 January 2018; Accepted 26 January 2018
Available online 06 March 2018
0927-0248/ © 2018 Elsevier B.V. All rights reserved.
J. Kusuma et al. Solar Energy Materials and Solar Cells 183 (2018) 211–219
212
J. Kusuma et al. Solar Energy Materials and Solar Cells 183 (2018) 211–219
(BET) specific surface area of the materials was obtained from N2 ad- MWCNTs with inner tube diameter ranging between 7 nm and 15 nm.
sorption/desorption isotherms recorded on model BELSORP- max MWCNTs unzipped to form GNRs is as shown in Fig. 2(d) which has
(Japan) at 77 K. Time resolved PL measurements were carried out using varying ribbon width from 80 to 300 nm. Ribbon lengths appear shorter
Horiba fluorescence photometer at 345 nm excitation source. than tubes but contribute efficiently to increase the aspect ratio.
Composite analysis was carried out using surface measurements by X- Figs. 2(b) and 2(f) shows the composites of TiO2-MWCNT and TiO2-
ray photoelectron spectra with Al Kα (1486.6 eV) as the X-ray source GNR respectively. The entanglement or intercalation of tubes in
set at 150 W and a pass energy of 50 eV. Photovoltaic performance of MWCNTs tend to trap the TiO2 nanoparticles restricting the flow,
QDSSCs were measured by integrated I–V Test station (PVIV-211 V) whereas in case of GNR composite the ribbons/strips provides a large
with Keithley 2420 source meter and 94023A Oriel Sol3A, class AAA surface area for TiO2 nanoparticles with less grain boundaries for better
Solar Simulator (power output: 100 mW/cm2, lamp power: 450 W), transport. Also the high reactive edge sites with carboxylic, epoxy and
equipped with an AM 1.5 filter. IQE measurements were carried out –OH groups allows better bonding with TiO2 than metallic nanotubes. It
using Oriel using Hg lamp (450 W, Oriel) and a monochromator with a could be possible that TiO2 spheres can act as pillars between the strips,
bandwidth of 5 nm. Impedance analysis was carried out using electro- separating the ribbons from each other due to its high active edge sites.
chemical station in Impedance mode using CHI660D model in the fre- It also allows to carry the electrons across more quickly and this could
quency range of 1 MHz to 1 Hz. For AFM, a solution of nano crystals be the reason for better quenching (observed from PL). The HR-TEM
(dispersed in DMF) were dropped on Mica sheet for imaging in air using image of nanotubes and GNR composites are shown in Fig. 2(c) and (g).
Bruker Dimension Icon AFM equipped with a Nano scope V SPM con- Inclusion of graphene materials into TiO2 via thermal treatment tend to
troller in contact mode and TUNA was used for conductivity measure- form a strong chemical interaction giving rise to new Ti-O-C bonds
ments. (confirmed from XPS and IR- Fig S5 in SI), resulting in changes in the
lattice spacing. The changes in the lattice structure brings out changes
4. Results and discussions in the lattice parameters reflecting the changes in d-spacing as well
from 0.39 nm to 0.42 nm (GNRs) and from 0.35 nm to 0.32 nm (TiO2).
4.1. X-ray diffraction studies (XRD) SAED diffraction fringes are presented in Fig. 2(d) and (h) for MWCNT
and GNR composites. It clearly depicts the material on a whole to be
Structural studies of graphene and composite materials were carried perfectly crystalline and also the lattice plane reflections match with
out using XRD as shown in Fig. 1. Peaks at ~10.8°, 11.7° and 25.7° the XRD data.
(JCPDS no 75-1621) with d-spacing of 8.1 Å, 8.9 Å and 3.4 Å (Fig. 1a)
indicate the synthesized GO [20], GNRs [21] and MWCNTs [24] re-
spectively. The appearance of (001) reflection in GNRs at 11.7° is an 4.3. Brunauer-Emmett-Teller (BET)
indication of successful unzipping of MWCNT (at 2θ = 26°) and ex-
foliation [24]. TiO2 and its composite with GNR is shown in Fig. 1(b). N2 adsorption and desorption isotherms were carried out to find the
TiO2 is in anatase phase and peaks at 2θ values of 25.5°, 37.8°, 47.9° surface area of graphene materials and their composites as listed in
and 53.9° are assigned to the crystal planes of (101), (004), (200) and Table 1 below. The highest surface area is observed for unzipped GNRs
(105) of the body-centred tetragonal TiO2 (JCPDS 21-1272). The strong (399.3 m2 g−1), which is 1.8 times higher than MWCNTs (213.7 m2
interaction caused during composite formation results in changes in the g−1) showing maximum unzipping of MWCNTs. The isotherm curve for
crystalline phase of TiO2 particles. Since the amount of dispersed GNRs (Fig S1 of SI) suggest a type IV isotherm of irreversible desorption
carbon material is less, there is no significant peak of GNRs in the patterns with H3-hysteresis loop [25], and average pore size around
composite and hence exhibit similar XRD patterns [8]. HR-TEM, IR and 1.9 nm (micro porous structure) [26]. Tri-distribution mesoporous
XPS has been further used to confirm the Ti-O-C bond formation in nature (3, 8 and 14 nm) with higher pore diameter and volume has been
composites. obtained for MWCNTs. Also TiO2-GNR has comparatively greater sur-
face area and narrow pore size distribution (Fig S2 of SI) with TiO2
nanoparticles being uniformly spread throughout (as evidenced by
4.2. Transmission electron microscopy (TEM)
EDAX mapping in Fig S3 of SI). This predicts that GNRs could serve as a
conducting floor for deposition of TiO2 particles because of its high
TEM (Fig. 2) has been further used to evidence the unzipping of
surface area and microporous structure.
MWCNTs and as well to confirm the network of graphene materials
with TiO2 as composites. Fig. 2(a) represents the entangled network of
Fig. 1. XRD graphs of (a) Graphene materials and (b) Composite of GNR.
213
J. Kusuma et al. Solar Energy Materials and Solar Cells 183 (2018) 211–219
Fig. 2. TEM micrograph of (a) MWCNTs (b), (c) and (d) TEM, HR-TEM and SAED OF TiO2-MWCNT, (e) GNRs, (f), (g) and (h) TEM, HR-TEM and SAED OF TiO2-MWCNT.
Fig. 3. Surface morphology of: (a) MWCNTs, (b) GNRs, (c) GO, (d) TiO2-MWCNT, (e) TiO2-GNR and (f) TiO2-GO.
214
J. Kusuma et al. Solar Energy Materials and Solar Cells 183 (2018) 211–219
of lamellar aggregation of TiO2 particles. Fig. 3(c) shows two dimen- Table 2
sional stacked infinite nanosheets of Graphene oxide and its composite Parameters of radiative and non-radiative lifetime of exciton.
(Fig. 3f) represents uniform deposition of TiO2 spherical particles on
Sample A1 τ1(ns) A2 τ2(ns) τavg (ns)
nanosheets [27,28] with average particle size of 25 nm.
TiO2 93 0.54 7 6.19 5.01
TiO2-MWCNT 98 0.46 2 6.25 5.11
4.5. Photoluminescence studies (PL) TiO2-GO 99 0.45 1 8.01 6.42
TiO2-GNR 99 0.46 1 6.78 4.22
PL spectroscopy was employed to follow the fate of photo-excited
charge carriers in composite materials. The irradiation of TiO2 results in
emission of photons at specific wavelength associated with direct band network to form Ti-O-C linkages (also confirmed from ATR-IR shown in
gap or defect mediated recombination [29,30]. Composites of graphene Fig S5 of SI). Also decrease of electron density around Ti atom results in
materials with TiO2 show quenching in PL intensity (Fig. 4a) suggesting increase in its binding energy because of greater electro negativity of
electron transfer from conduction band of TiO2 to Graphene materials carbon than Ti. However no Ti 3+ peaks were detected.
efficiently while reducing radiative recombination. Among all the
composites, TiO2-GNR shows better quenching due to the enhanced 4.7. Atomic Force Microscopy (AFM)
charge separation (reflected from higher Jsc in IPCE) [31]. Also peaks
of composites have slightly shifted. This could be preferably due to the AFM was used to analyze the structure, height (no. of layers), width
reduction in the dimension of the material. CNTs/GNRs are one di- and conductivity of the ribbons. Fig. 6(a) shows the topography on
mensional in nature whose absorbance are slightly blue shifted and red micron scale wherein the ribbons looks separated forming strips of
shifted for GO composites for its plane dimension (as observed in graphene. Fig. 6(b) shows the height profile of GNRs varying from 3 nm
UV–Visible spectroscopy). Further PL emission as a function of time to 7 nm (combination of bilayer to few layers) of ribbons [37]. Widths
(Time resolved photoluminescence spectra) was carried out for probing of the ribbons were observed to extend up to 300 nm which are in
the recombination dynamics in the excited states. insert 'full stop' after agreement with the results obtained from TEM. The oxidized edges and
excited state. The kinetics can be analyzed using bi-exponential decay disruption of π network allows GNRs to gain partial conductivity and
which involves modelling the data and fitting the curve into bi-ex- was confirmed with the AFM TUNA conductivity measurements. The
ponential decay (y = y0 + A1e −x / t1+ A2e −x / t2 ) [32] as shown below in measurements were carried out at many places and results showed the
Fig. 4(b) and parameters being tabulated in Table 2. The decreasing combination of both semiconducting and metallic nature of the mate-
trend in average lifetime from TiO2 to GNR is in par with PL emmision rial suggesting ribbons to be in arm chair and zigzag configurations
decay. The average lifetime in case of GNR composites suggests the respectively. Fig. 6(c) is IV curve for biased voltage of ± 4 V. The curve
greater electron transfer rate across the interface substantiating better tends to saturate beyond 3 V biased voltage suggesting GNRs being
quenching in PL emission decay. transformed into a semiconducting material [37]. This also sub-
stantiates the band gap of GNRs (~3.45 eV) obtained by band edge of
4.6. X-ray photoelectron spectroscopy (XPS) UUV–Visble spectroscopy (Fig S4 in SI).
Fig. 4. (a) Photoluminescence spectra and (b) Time-resolved photo luminescence spectra.
215
J. Kusuma et al. Solar Energy Materials and Solar Cells 183 (2018) 211–219
Fig. 5. (a) XPS whole survey spectra of TiO2, MW-TiO2 and GNR-TiO2; (b)-(d) High resolution and Gaussian curve fitting for C, O and Ti in TiO2, TiO2-MWCNT and TiO2-GNR.
interface. GNR composites with increased active sites red shifts the The IPCE spectra for graphene based composites are shown in the
absorption edge (as observed from UV–Visble spectroscopy), enhances Fig. 7(b), which signifies light harvesting, charge injection, and charge
the photocurrent and reduces the recombination of exciton by facil- collection in the device [47]. IPCE spectra show a superior photo re-
itating efficient charge separation [38,39]. The photovoltaic perfor- sponse between 320 and 520 nm. The Graphene composites show an
mances of these carbonaceous materials tend to show a quite significant enhancement in photocurrent compared to TiO2. The higher photo
difference in its performance because of its different chemical, optical, current observed in case of GNR composite show enhanced charge se-
electrical properties and dimensional frameworks. GO composites with paration, due to the wrinkled strips which could allow electrons to
TiO2 as photo anode has already been reported in case of DSSC [40,41]. migrate through the interconnected TiO2 nanoparticles. Though the
But GO when brought in contact with TiO2 under UV irradiation tends alignments of strips are not optimally oriented electrons can easily
to decrease the activity and PCE drops rapidly over continuous illu- channelize through these strips as they have less number of voids. Also
mination [42] unlike in the case of GNRs [43]. Though CNTs are good it allows the electrons to easily hop between the strips for which the
electron transporters the entanglement in MWCNTs can adversely affect FTO substrate creates a potential between working electrode and
(as observed in TEM) the electron transport and also suffer from in- junction driving the electrons enhancing efficiency. In case of TiO2-
tertube slipping which degrades their ability as structural reinforce- MWCNT composite, the intermingled nanotubes tends to trap electrons
ment additives [44]. But GNRs followed by unzipping is quite stable within them and possess relatively more obstruction for flow of elec-
and found to have lower electron impedance as shown in Fig. 7(c) that trons compared to GO and GNRs.
can enhance the separation of electron–hole pairs [45] (as reflected The electrical resistance and recombination dynamics are studied
from PL and IPCE). Also the high surface area of GNRs creates as many using Electron impedance spectroscopy (EIS). The measurements were
active sites for TiO2 nanoparticles and allows for intercalation with carried out in dark and the frequency was measured from 1 MHz to 1 Hz
host, facilitating quick electron transfer. The isotherm shows micro and the Nyquist plots for these cells are given in Fig. 7(c) along with
porous structure in GNRs which helps in channelizing the electrons their equivalent circuit. The series ohmic resistance (Rs) at higher fre-
through interconnected network of TiO2 nanoparticles. Also exfoliated quency contributes to the interfacial resistance between the electrical
ribbons with confined electrons tend to increase the photochemical contacts and FTO, which is the x-intercept at high frequency [29] and
activity. The work function of Graphene (−4.42 eV) is close to FTO are almost found to be around ~11 Ω. CPE is the constant phase ele-
(−4.7 eV) and hence there is no much change in the Voc for composites ment which is a constant factor for all. Rt is charge transfer at TiO2/
[46]. The enhancement of Jsc clearly indicates that GNRs are good QD/Electrolyte interface for which the values are tabulated for com-
electron transporters. posites in Table 3. The smaller radius of the semicircle with lesser Rt for
216
J. Kusuma et al. Solar Energy Materials and Solar Cells 183 (2018) 211–219
Fig. 6. (a) AFM image of GNRs, (b) Height profile, (c) AFM TUNA measurements and (d) IV Curve from TUNA.
Fig. 7. (a) IV characteristics, (b) IPCE spectra, (c) Nyquist plot for Graphene composites and (d) Nyquist plot for different wt. ratio of GNRs.
217
J. Kusuma et al. Solar Energy Materials and Solar Cells 183 (2018) 211–219
218
J. Kusuma et al. Solar Energy Materials and Solar Cells 183 (2018) 211–219
[19] S. Swetha, S.M. Santhosh, R. Geetha Balakrishna, Enhanced bactericidal activity of nanotubes using γ-rays, J. Appl. Phys. 109 (2011) 054303.
modified titania in sunlight against Pseudomonas aeruginosa, a water-borne pa- [35] H.L.Q. Zeng, H. Duan, Y. Guo, X. Liu, Y. Zhang, H. Liu, A green method to prepare
thogen, Photochem. Photobiol. 86 (2010) 1127–1134. TiO2/MWCNT nanocomposites with high photocatalytic activity and insights into
[20] J.Y. Chen, Bowen Li, Chun Shi, Gaoquan, An improved Hummers method for eco- effect of heat treatment on photocatalytic activity, RSC Adv. (2015).
friendly synthesis of graphene oxide, Carbon 64 (2013) 225–229. [36] F. Teng, Guozhi Wang, Youqing Wang, Caitian Gao, Lulu Chen, Peng Zhang,
[21] D.V.K. Amanda, L. Higginbotham, Alexander Sinitskii, Zhengzong Sun, James Zhenxing Zhang, Erqing Xie, The role of carbon in the photocatalytic reaction of
M. Tour, Lower-defect graphene oxide nanoribbons from multiwalled carbon na- carbon/TiO2 photocatalysts, Appl. Surf. Sci. 320 (2014) 703–709.
notubes, ACS Nano 4 (2010) 2059–2069. [37] R.S. Khare, Dhanraj B. Bansode, Sanjeewani More, Mahendra A. Majumder,
[22] Z. Du, Hua Zhang, Huili Bao, Xinhua Zhong, Optimization of TiO2 photoanode films Mainak Pillai, Vijayamohanan K. Late, J. Dattatray, Graphene nanoribbons as
for highly efficient quantum dot-sensitized solar cells, J. Mater. Chem. A 2 (2014) prospective field emitter, Appl. Phys. Lett. 106 (2015) 023111.
13033. [38] A. Adán-Más, Di Wei, Photoelectrochemical properties of graphene and its deriva-
[23] N. Fuke, Ryuzi Katoh, Ashraful Islam, Motohiro Kasuya, Akihiro Furube, tives, Nanomaterials 3 (2013) 325–356.
Atsushi Fukui, Yasuo Chiba, Ryohichi Komiya, Ryohsuke Yamanaka, Liyuan Han, [39] W. Tu, Yong Zou, Zhigang Zou, Versatile graphene-promoting photocatalytic per-
Hiroshi Harima, Influence of TiCl4 treatment on back contact dye-sensitized solar formance of semiconductors: basic principles, synthesis, solar energy conversion,
cells sensitized with black dye, Energy Environ. Sci. 2 (2009) 1205. and environmental applications, Adv. Funct. Mater. 23 (2013) 4996–5008.
[24] D.V. Kosynkin, A.L. Higginbotham, A. Sinitskii, J.R. Lomeda, A. Dimiev, B.K. Price, [40] Y.S. Li, J.L. Liao, S.Y. Wang, W.H. Chiang, Intercalation-assisted longitudinal un-
J.M. Tour, Longitudinal unzipping of carbon nanotubes to form graphene nanor- zipping of carbon nanotubes for green and scalable synthesis of graphene nanor-
ibbons, Nature 458 (2009) 872–876. ibbons, Sci. Rep. 6 (2016) 22755.
[25] M. Sahoo, S. Ramaprabhu, Enhanced electrochemical performance by unfolding a [41] I.V. Lightcap, P.V. Kamat, Fortification of CdSe quantum dots with graphene oxide.
few wings of graphene nanoribbons of multiwalled carbon nanotubes as an anode Excited state interactions and light energy conversion, J. Am. Chem. Soc. 134
material for Li ion battery applications, Nanoscale 7 (2015) 13379–13386. (2012) 7109–7116.
[26] S.K. Hong, D. Praveen Reddy, D. Amaranatha Choi, Jiha Kim, Tae Kyu, Excellent [42] S. Morales-Torres, L.M. Pastrana-Martínez, J.L. Figueiredo, J.L. Faria, A.M. Silva,
photocatalytic hydrogen production over CdS nanorods via using noble metal-free Design of graphene-based TiO2 photocatalysts–a review, Environ. Sci. Pollut. Res.
copper molybdenum sulfide (Cu2MoS4) nanosheets as co-catalysts, Appl. Surf. Sci. Int. 19 (2012) 3676–3687.
396 (2017) 421–429. [43] J.G.K. Radich, Anthony L. Zhu, Jiadong Kamat, V. Prashant, Is graphene a stable
[27] N.M.H. Zainudin, Ab. Rahim Yusoff, Mashitah Mohd Tan, Ling Ling Chong, platform for photocatalysis? Mineralization of reduced graphene oxide with UV-
Kwok Feng, Impedimetric graphene-based biosensor for the detection of Escherichia irradiated TiO2 nanoparticles, Chem. Mater. 26 (2014) 4662–4668.
coli DNA, Anal. Methods 6 (2014) 7935–7941. [44] J.M.T. Mohammad, A. Rafiee, Wei Lu, Nikhil A. Koratkar, Abhay V. Thomas,
[28] J. Wang, J. Liu, D. Chao, J. Yan, J. Lin, Z.X. Shen, Self-assembly of honeycomb-like Ardavan Zandiatashbar, Javad Rafiee, Graphene nanoribbon composites, ACS Nano
MoS2 nanoarchitectures anchored into graphene foam for enhanced lithium-ion (2010).
storage, Adv. Mater. 26 (2014) 7162–7169. [45] U.M.C.N.R. Rao, H.S.S. Ramakrishna Matte, Synthesis, Characterization, and
[29] L.W. Zhizhong Hana, Haibo Panb, Chunyan Lia, Jinghua Chena, Variant effect of Selected Properties of Graphene, 2013.
graphene sheets and ribbons on photocatalytic activity of TiO2 sheets/graphene [46] A. Badawi, N. Al-Hosiny, S. Abdallah, The photovoltaic performance of CdS
composite, J. Mol. Catal. A: Chem. 398 (2015) 399–406. quantum dots sensitized solar cell using graphene/TiO2 working electrode,
[30] Y.T.V. Liang, Baiju K. Gray, Kimberly A. Hersam, C. Mark, Minimizing graphene Superlattices Microstruct. 81 (2015) 88–96.
defects wnhances titania nanocomposite-based photocatalytic reduction of CO2 for [47] P. Zhu, A. Sreekumaran Nair, P. Shengjie, Y. Shengyuan, S. Ramakrishna, Facile
improved solar fuel production, Nano Lett. 11 (2011) 2865–2870. fabrication of TiO2-graphene composite with enhanced photovoltaic and photo-
[31] T.-Y.W. Han, Chu-Fu Hsieh, Te Chien, Hydrothermal synthesis and visible light catalytic properties by electrospinning, ACS Appl. Mater. Interfaces 4 (2012)
photocatalysis of metal-doped titania nanoparticles, J. Vac. Sci. Technol. B: 581–585.
Microelectron. Nanomater. Struct. 25 (2007) 430. [48] M.S. Khannam, Shyamalima Dolui, Swapnil Dolui, Swapan Kumar, A graphene
[32] A.C. Austin Nevin, Sara Bellei, Cosimo D.’ Andrea, Lucia Toniolo, oxide incorporated TiO2photoanode for high efficiency quasi solid state dye sen-
Gianluca Valentini, Daniela Comelli, Time-resolved photoluminescence spectro- sitized solar cells based on a poly-vinyl alcohol gel electrolyte, RSC Adv. 6 (2016)
scopy and imaging: new approaches to the analysis of cultural heritage and its 55406–55414.
degradation, Sensors (2014). [49] J. Chen, W. Lie, X.B. Zhang, Enhanced electron transfer rate for quantum dot sen-
[33] K. Ding, Z. Miao, Z. Liu, Z. Zhang, B. Han, G. An, S. Miao, Y. Xie, Facile synthesis of sitized solar cell based on CNT-TiO2 film, J. Nanosci. Nanotechnol. 12 (2012)
high quality TiO2 nanocrystals in ionic liquid via a microwave-assisted process, J. 6476–6479.
Am. Chem. Soc. 129 (2007) 6362–6363. [50] G. Zhu, T. Xu, T. Lv, L. Pan, Q. Zhao, Z. Sun, Graphene-incorporated nanocrystalline
[34] Z. Xu, Chunying Min, Lei Chen, Liangsen Liu, Guangwei Chen, Ning Wu, TiO2 films for CdS quantum dot-sensitized solar cells, J. Electroanal. Chem. 650
Modification of surface functionality and interlayer spacing of multi-walled carbon (2011) 248–251.
219