Functional graphene: synthesis, characterization and

Dissertation zur Erlangung des Doktorgrades Der Technischen Fakultät der Albert-Ludwigs-Universität Freiburg im Breisgau

Functional graphene: synthesis, characterization and application in optoelectronics Van Chuyen Pham 2015

Albert-Ludwigs-Universität Freiburg im Breisgau Technische Fakultät Institut für Mikrosystemtechnik

Dekan: Prof. Dr. Georg Lausen

Referenten: PD. Dr. Michael Krüger Prof. Dr. Stefan Weber Autor: Van Chuyen Pham

Datum der Abgabe: 2015 Datum der Disputation: 2015

II

Erklärung

Ich erkläre, dass ich die vorliegende Arbeit ohne unzulässige Hilfe Dritter und ohne Benutzung anderer als der angegebenen Hilfsmittel angefertigt habe. Die aus anderen Quellen direkt oder indirekt übernommenen Daten und Konzepte sind unter Angabe der Quelle gekennzeichnet. Insbesondere habe ich hierfür nicht die entgeltliche Hilfe von Vermittlungsoder Beratungsdiensten (Promotionsberaterinnen oder Promotionsberater oder anderer Personen) in Anspruch genommen. Niemand hat von mir unmittelbar oder mittelbar geldwerte Leistungen für Arbeiten erhalten, die im Zusammenhang mit dem Inhalt der vorgelegten Dissertation stehen. Die Arbeit wurde bisher weder im In- noch im Ausland in gleicher oder ähnlicher Form einer anderen Prüfungsbehörde vorgelegt. Ich erkläre hiermit, dass ich mich noch nie an einer inoder ausländischen wissenschaftlichen Hochschule um die Promotion beworben habe oder gleichzeitig bewerbe.

Datum/date:

Unterschrift/signature:

III

List of publications Scientific journals  Chuyen V. Pham, Michael Eck and Michael Krueger. “Thiol functionalized reduced graphene oxide as a base material for novel graphene-nanoparticle hybrid composites”. Chem. Eng. J. 231 (2013) 146–154  Chuyen V. Pham, Michael Krueger, Michael Eck, Stefan Weber, Emre Erdem. “

Comparative electron paramagnetic resonance investigation of reduced grapheme

oxide and carbon nanotubes with different chemical functionalities for quantum dot attachment”. Appl. Phys. Lett. 104, (2014) 132102  Michael Eck, Chuyen V. Pham, Simon Züfle, Martin Neukom, Martin Seßler, Dorothea Scheunemann, Emre Erdem, Stefan Weber, Holger Borchert, Beat Ruhstaller, Michael Krüger, “Improved efficiency for bulk heterojunction hybrid solar cells by utilizing CdSe quantum dot - graphene nanocomposites” Phys. Chem. Chem. Phys. 16 (24), 1225112260 Conferences  Michael Krueger, Chuyen V. Pham, Michael Eck. “Nanocomposites of quantum dots and functionalized graphenes are improving the performance of inorganic-organic hybrid solar cells”. Oral presentation, 2014 MRS Fall Meeting & Exhibit, November 30 – December 5, 2014. Boston, Massachusetts.  Chuyen V. Pham, M. Eck, M. Krueger, “Hybrid nanocomposites of quantum dots and functionalized graphenes for optoelectronic applications”. Oral presentation, Graphene Week conference 2014, 23-27 June, 2014, Chalmers University of Technology, Sweden.  Chuyen V. Pham, Michael Eck, Michael Krueger, “Thiol functionalized graphene decorated with CdSe quantum dots for hybrid solar cells”. Poster presentation, Graphene Week conference, 2-7 June, 2013, Chemnitz, Germany.  Chuyen V. Pham, Michael Eck, Michael Krueger. “Nanoparticle decorated graphene based hybrid materials for energy harvesting applications”. Poster presentation, Hybrid Materials conference, 3-7 March, 2013, Sorrento, Italy.  Chuyen V. Pham “Thiol functionalized graphene decorated with CdSe quantum dots for hybrid solar cells”. Oral presentation, Freiburg Materials Research Centre (FMF) Kolloquium, 16-17 Oktober 2012, Schuchsee, Baden Württemberg.

IV

Table of contents Abstract ..................................................................................................................................... IX 1. Introduction to graphene ........................................................................................................ 1 1.1 Graphene fundamentals .................................................................................................... 1 1.1.1 Electronic properties .................................................................................................. 3 1.1.2 Optical properties ...................................................................................................... 7 1.1.3 Mechanical properties................................................................................................ 8 1.1.4 Thermal properties .................................................................................................... 8 1.2 Graphene synthesis ........................................................................................................... 9 1.2.1 Mechanical exfoliation (ME)..................................................................................... 9 1.2.2 Epitaxial growth on silicon carbide SiC (EG) ......................................................... 10 1.2.3 Chemical vapor deposition (CVD) .......................................................................... 10 1.2.4 Solution approach to reduced graphene oxide through graphite oxide ................... 14 1.2.5 Unzipping of carbon nanotubes ............................................................................... 15 1.2.6 Other methods.......................................................................................................... 15 1.3 Applications of graphene-type structures ....................................................................... 16 1.3.1 Flexible transparent conductors (FTCs) .................................................................. 16 1.3.2 High-frequency transistors ...................................................................................... 17 1.3.3 Energy storage systems ........................................................................................... 19 1.3.4 Light –emitting devices ........................................................................................... 23 1.3.5 Photovoltaics devices (PVs) .................................................................................... 24 1.3.6 Photodetectors (PDs) ............................................................................................... 29 1.3.7 Fuel cells .................................................................................................................. 30 1.3.8 Other applications .................................................................................................... 32 1.4 Motivation ...................................................................................................................... 32 2. Graphene oxide (GO) and reduced graphene oxide (rGO) .................................................. 34 2.1 Motivation of GO and rGO synthesis ............................................................................ 34 2.2 Materials and synthesis of GO ....................................................................................... 35 2.2.1 Chemicals ................................................................................................................ 35 2.2.2 Synthesis of GO ....................................................................................................... 35 2.3 Characterization of GO, results and discussions ............................................................ 36 2.3.1 Optical microscope, Scanning electron microscopy (SEM) .................................... 36 V

2.3.2 Atomic force microscopy (AFM) ............................................................................ 37 2.3.3 X-ray photoelectron spectroscopy (XPS) ................................................................ 38 2.3.4 UV-Vis absorption spectroscopy ............................................................................. 39 2.4 Reduction of GO to rGO ................................................................................................ 40 2.4.1 Synthesis of rGO ..................................................................................................... 40 a) Reduction of GO by hydroiodic acid (HI) .................................................................... 40 b) Reduction in N2H4 vapor .............................................................................................. 40 c) Thermal reduction ........................................................................................................ 41 d) Reduction by phosphorus pentasulfilde ....................................................................... 41 2.4.2 Resistivity measurements ........................................................................................ 42 2.4.3 Results and discussions ........................................................................................... 42 2.5 Conclusions .................................................................................................................... 45 3. Nanoparticle-functionalized graphene hybrid materials ...................................................... 46 3.1 CdSe quantum dots – graphene hybrid material (CdSe QD-TrGO) .............................. 46 3.1.1 Background and motivation ........................................................................................ 46 3.1.2 Synthesis and characterizations of TrGO .................................................................... 48 3.1.2.1 Materials and synthesis ......................................................................................... 48 3.1.2.2 Characterization, results and discussions ............................................................. 48 a) XPS investigation ......................................................................................................... 49 b) FTIR spectroscopy ....................................................................................................... 51 c) UV-Vis absorption spectroscopy .................................................................................. 52 3.1.3 Synthesis of CdSe QDs ............................................................................................... 54 3.1.3.1 Materials and methods .......................................................................................... 54 3.1.3.2 Characterization .................................................................................................... 54 3.1.3.3 Results and discussions ........................................................................................ 54 3.1.4 Synthesis and characterizations of CdSe QD-TrGO hybrid material ......................... 55 3.1.4.1 Materials and synthesis ......................................................................................... 55 3.1.4.2 Characterization .................................................................................................... 56 a) TEM analysis ................................................................................................................ 56 b) UV-Vis absorption spectroscopy ................................................................................. 57 c) PL quenching experiments ........................................................................................... 58 3.2 Synthesis and application of Ag NP-graphene hybrid material ..................................... 61 VI

3.2.1 Motivation ............................................................................................................... 61 3.2.2 Synthesis .................................................................................................................. 61 3.2.3 Result and discussions ............................................................................................. 61 3.2.4 Application of Ag NP-rGO hybrid material in dye-sensitized solar cells ............... 62 3.3 Synthesis of ZnO NP-graphene hybrid materials........................................................... 65 4. EPR investigations of functionalized graphene and CNT, and CdSe QD-TrGO hybrid materials ................................................................................................................................... 68 4. 1 Introduction and motivation of graphene EPR research ............................................... 68 4.2 Materials and methods ................................................................................................... 72 4.2.1 Synthesis .................................................................................................................. 72 4.2.2 EPR experiments ..................................................................................................... 73 4.2.3 Spin-counting procedure.......................................................................................... 73 4.3 Results and discussions .................................................................................................. 74 4.3.1 Removal of Mn+2 in O-CNT and GO samples ........................................................ 74 4.3.2 EPR signals of graphene and CNT with different functionalities ........................... 75 4.3.3 Comparative EPR signals of CNT-SH and TrGO ................................................... 77 4.3.4 Quenching of the EPR signals in CdSe QD-TrGO hybrid material ........................ 78 4.3.4 Temperature-dependent EPR spectra of O-CNT and TrGO ................................... 78 4.4 Conclusions .................................................................................................................... 80 5. Applications of QD-TrGO hybrid material in photovoltaics ............................................... 81 5.1 Introduction to photovoltaics ......................................................................................... 81 5.1.1 Hybrid solar cells ..................................................................................................... 82 5.1.2 Applications of graphene in photoactive layer of solar cells. .................................. 84 5.1.3 Motivation ............................................................................................................... 86 5.2 Methods and experiments............................................................................................... 87 5.2.1 Power conversion efficiency ................................................................................... 87 5.2.2 Solar cell fabrication ................................................................................................ 88 4.2.3 TEM tomography .................................................................................................... 89 5.2.4 Solar cell characterization ....................................................................................... 89 5.2.5 Cyclic voltammetry (CV) experiments.................................................................... 90 5.3 Results and discussions .................................................................................................. 90 5.3.1 I-V Characteristics ................................................................................................... 90 VII

5.3.2 Cyclic voltammetry experiments ............................................................................. 92 5.3.3 TEM and TEM tomography .................................................................................... 93 5.3.4 AFM......................................................................................................................... 95 5.4 Conclusions .................................................................................................................... 96 6. Application of CdSe-rGO hybrid materials for PDs ............................................................ 98 6.1 Motivation ...................................................................................................................... 98 6.2 Experiments.................................................................................................................... 98 6.2.1 rGO preparation ....................................................................................................... 98 6.2.2 Photoconductor fabrication...................................................................................... 99 6.2.3 Electrical measurements .......................................................................................... 99 6. 3 Results and discussions ................................................................................................. 99 6.3.1 Contact probe fabrication ........................................................................................ 99 6.3.2 Photoconductor measurements .............................................................................. 102 7. Summary and outlook ........................................................................................................ 104 7.1 Summary ...................................................................................................................... 104 7.2 Outlook ......................................................................................................................... 106 Abbreviations ..................................................................................................................... 107 Acknowledgements ............................................................................................................ 109 References .......................................................................................................................... 110

VIII

Abstract Graphene has recently attracted enormous attention within the scientific community due to its outstanding optical, electrical and mechanical properties and is highly promising for various applications. Also, semiconducting CdSe quantum dots (QDs) with easily tunable optical and electrical properties, already demonstrated their potential for application in various applications such as e.g. photovoltaics. In this thesis, the transparency, high carrier mobility and high specific surface area of graphene are combined with the optical properties of CdSe QDs by fabricating a novel CdSe QD decorated graphene hybrid material. Based on the synergetic properties, the hybrid material has been successfully characterized by various methods and was introduced in active layers of hybrid solar cells, remarkably enhancing the power conversion efficiency of the solar cells. Graphene oxide (GO) was synthesized by a modified Hummers method. Then, oxygencontaining groups within GO were reduced, and partially transformed into thiolfunctionalities. As a result, thiol-functionalized reduced graphene oxide (TrGO) was obtained. This was achieved by refluxing GO with phosphorus pentasulfilde in dimethylformamide solvent. The thiol groups can serve as anchor points for the attachment of nanoparticles. This was realized with CdSe QDs, and ZnO nanoparticles, respectively via a self-assembly process and novel hybrid materials were obtained: nanopartice decorated TrGO. Various techniques and methods such as SEM, TEM, XPS, FTIR, UV-Vis absorption spectroscopy have been used to characterize and develop GO, rGO, TrGO and QD-TrGO hybrid materials. Graphenes with different functionalities and reduction degrees were also investigated in comparison with carbon nanotubes by electron paramagnetic resonance (EPR) spectroscopy. For application in photovoltaics, CdSe QD-TrGO hybrid material was mixed with semiconducting polymers to form photoactive thin films which were utilized in hybrid solar cells. The photovoltaic cells showed significantly enhanced solar power conversion efficiencies. The graphene-based hybrid solar cells achieved a PCE of up to 4.24% which is

IX

46% higher compared to graphene-free cells and is ranging among the best efficiencies of hybrid quantum dot-polymer solar cells so far. Additionally, CdSe QD decorated rGO was tested regarding its utilization in photoconductiors. Preliminary results proved that photo-induced charge transfer from QDs to graphene occurs resulting in a measureable photocurrent contribution. This result is consistent with results obtained from PL quenching experiments and the quenching of EPR signals for CdSe QD-TrGO hybrid materials and further demonstrates the high potential of nanoparticle-graphene hybrid materials for optoelectronic applications.

X

Deutsche Zusammenfassung Wegen seiner außerordentlichen optischen, elektrischen sowie mechanischen Eigenschaften, hat Graphen eine enorme Beachtung in der Wissenschaft gefunden, auch im Hinblick auf sein vielfältiges Anwendungspotential. Halbleitende kolloidale CdSe Quantenpunkte besitzen leicht einstellbare optische und elektrische Eigenschaften und ihr Potential für verschiedene Anwendungen, wie zum Beispiel der Photovoltaik, wurde bereits erfolgreich gezeigt. Innerhalb dieser Arbeit werden die optische Transparenz, die hohe Ladungsträgermobilität und die große spezifische Oberfläche von Graphen mit den optischen Eigenschaften von kolloidalen CdSe Quantenpunkten in Form von neuartigen CdSe Quantenpunkt-Graphen Hybridmaterialien kombiniert. Die Hybridmaterialien wurden anhand von verschiedenen Methoden charakterisiert und in photoaktiven Schichten von hybriden Solarzellen erfolgreich eingesetzt, was zu einer bemerkenswerten Erhöhung der Solarzelleneffizienz führte. Anfängliche vielversprechende Versuche, Quantenpunkt-Graphene Hybridmaterialien als Material für Photodetektoren zu verwenden wurden ebenfalls durchgeführt und zeigen das große Potential für deren Verwendung in weiteren optoelektronischen Anwendungen.

XI

Chapter 1 Introduction to graphene 1.1 Graphene fundamentals For the past six decades, almost all electronics have been based on silicon, because of its intrinsic semiconducting, and environmental friendly behavior, as well as its abundance covering almost all electronic technologies available in modern life. In 2004, Andre Geim and Konstantin Novoselov, two chemists at the University of Manchester in the United Kingdom, extracted a new type of carbon, called graphene by repeatedly pressing and pulling a Scotch tape onto graphite, peeling off single-atom thick carbon flakes in which carbon atoms are arranged in hexagons, which is the thinnest known material. Afterwards, they explored its extraordinary electron mobility by characterizing field effect transistors made out of graphene [1]. By this way, they opened up a new world of research based on this material and its fascinating properties and were awarded the Nobel Prize in Physics in 2010. Among main features, graphene is exceptional strong, flexible with impressive optical properties and it is equally transparent to light with wavelengths ranging from ultraviolet to infrared. Therefore, the scientific community could fancy a new Age of technology where graphene emerges replacing silicon to dominate electronic industries such as computer chips, displays, touch screens and solar cells. Graphene is the latest discovered allotrope of carbon, following the discovery of the fullerenes ('buckyballs') in the 1980s and the nanotubes in the 1990s [2]. Graphene, a 2D material, is a basic building block of all other known graphitic material forms. Graphene can be wrapped to 0 dimentional fullerenes, rolled-up into 1 D nanotubes and stacked together forming 3D graphite as shown in Figure 1.1 [3]. But how is a 2D crystal defined? Clearly one-atomic thick layer is a 2D crystal, while 100 layers could be considered as thin film of a 3D material. In graphene, the electronic properties substantially depend on the number of atomic layers [4]. Single atomic layer called graphene and its double layers have simple electronic band spectra, both are zero-band gap semiconductors. From 3 to 10 layers, the electronic band spectra become more complex; conduction and valence bands overlap, and as a result different electronic properties occur. Therefore, single, double and few layers (3 to < 10) of graphene should be distinguished as three different types of 2D crystals [1,3,5].

1.1 Graphene fundamentals

Figure 1.1 Graphene is a 2D building block for carbon materials of all other dimensionalities. It can be wrapped up into 0D buckyballs, rolled up into 1D nanotubes or stacked into 3D graphite (reproduced from ref. [3]). Before the discovery of graphene, it was presumed that single-atom thick crystal layers were thermodynamically unstable and could not exist in free standing state [6,7]. In fact, when the thickness of a thin film decreases, its melting point is rapidly lowered and it therefore becomes unstable and decomposes into islands, normally occurring at the thickness of dozens of atomic layers. However, the tightly binding of the hexagonal lattice structure with a basic of two atoms per unit cell in which sp2 hybridized carbon atoms bind to each other by one bond in plane and a π- bond perpendicular to plane, leads to the super mechanical strength and high charge mobility occurring in graphene. The 2D material has high crystallinity, so electrons can travel several micrometers through graphene without scattering and are behaving as massless Dirac fermions [8] at room temperature. This provides researchers a new tool to study quantum mechanics with massless activity on their lab table instead of performing experiments at extremely low temperature [9]. Graphene exists in different types depending on the synthesis approaches. It can be obtained by e.g. mechanical exfoliation (ME), chemical vapor deposition (CVD), chemical synthesis from graphite to graphene oxide (GO) and reduced into reduced graphene oxide (rGO), epitaxial synthesis (EG), and so on. More detail will be given in the synthesis section in Chapter 1.2.

2


1.1.1 Electronic properties The electronic structure of single-layer graphene (SLG) is based on a tight-binding Hamiltonian [8,10], where six carbon atoms are arranged in a hexagon, as shown in Fig. 1.2. The structure can be viewed as a triangular lattice with a basis of two atoms per unit cell. The lattice vectors are determined as following: a1 = (3,

), a2 = (3, -

),

(1.1)

where a  1.42 Å is the carbon-carbon distance, a1 and a2 are the lattice unit vectors as depicted in Fig. 1.2. The reciprocal-lattice vectors are given by b1 =

(1,

), b2 =

(1, -

).

(1.2)

The two points K and K’ at the corners of the graphene Brillouin zone (BZ) (Fig. 1.2b) are named Dirac points and are important for graphene physics. At the Dirac points, the lowest electron densities occur.

Figure 1.2 Honeycomb lattice and its Brillouin zone. Left: lattice structure of graphene, made out of two interpenetrating triangular lattices (a1 and a2 are the lattice unit vectors, and δi, i=1,2,3 are the nearest-neighbor vectors). Right: corresponding Brillouin zone. The Dirac cones are located at the K and K′ points (adopted from ref. [11]). Based on the tight-binding Hamiltonian, the energy bands of single-layer graphene obey following formula [11]: E± (k) = ±t

- t’f(k),

(1.3)

3


where, f(k) = 2 cos(

kya) + 4 cos

cos

,

(1.4)

Where E± (k) is energy bands of a single-layer graphene, t is  2.8 eV, the nearest-neighbor hopping energy (hopping between different sublattices), and t’ is the next nearest-neighbor hopping energy (hopping in the same sublattice), k is the momentum of an electron at a certain point. In equation (1.3), the plus sign is related to the upper conduction band ( ) and the minus sign is related to the lower band valence band ( ). From Eq. (1.4), one can see that the spectrum is symmetric around zero energy if t’ = 0. For a certain value of t’, the electron-hole symmetry is broken and the

and  bands therefore become asymmetric [11].

Figure 1.3 Electronic spectrum in the hexagonal lattice of graphene; Left: energy spectrum (in units of t) for certain values of t and t’, with t = 2.7 eV and t’ = -0.2t. Right: zoom in of the energy bands close to one of the Dirac points (reproduced from ref. [11]). The overlap between the hybridized electron orbitals px and py forms the -bonds, and pz the π-bond. The pz electrons are independent from the other valence electrons. With one pz electron per atom in the π- π conjugated bond, the (-) sign band in equation (1.3) is fully occupied, while the (+) band is empty. These two bands touch at the K points and the Fermi level EF is the zero-energy reference. The Fermi surface is defined by K and K’. Expanding equation (1.3) at K(K’) results into linear π- and π- bands for Dirac fermions: E±(q) = ±νF

+O

(1.5)

4


where E±(q) is the graphene carrier dispersion, q is the momentum relative to the Dirac points and νF is the Fermi velocity, given by νF =3ta /2, with the best estimate of t  2.5 eV and a = 0.14 nm resulting in νF  1× 108 cm/s [8]. The two bands intersect directly at the intrinsic Fermi level, giving rise to its zero-gap semiconducting nature and semi-metallic properties. Near the Dirac point, the density of states converges to zero and the linear dispersion relation is formally equivalent to the massless Dirac equation. So electrons in graphene behave like massless Dirac fermions resulting in an extremely high carrier mobility, thereby a ballistic transport with an electron mean free path of about one micrometer is observable at a room temperature [12]. The twoatom based graphene lattice leads to an additional type of spin, so called pseudospin which is derived from the Hamiltonian describing the chiral nature of the quasiparticles that behave like massless Dirac fermions, apparent in the Pauli spinor term. The two inequivalent lattice points result in the K and K’ reciprocal points on the Brillouin zone edge, two different valleys with opposite pseudospin. This leads to an interesting property that an electron travels in one direction into the K valley, inducing a hole with opposite pseudospin that will travel to the inverse direction into the K’ valley. The phenomenon can be utilized in spintronic applications where the degree of freedom of carrier states is controllable by this pseudospin state [13]. The single-atom thin structure of graphene confines electrons travelling within the plane, creating a so-called two-dimensional electron gas (2DEG) resulting in the observation of a quantum Hall effect (QHE) [14]. This leads to a magneto-transport at high mobility for both electron and hole carriers in graphene. The QHE may provide the basis for new applications in

carbon-based

electric

and

magnetic

field-effect

devices,

such

as

ballistic

metallic/semiconducting graphene ribbon devices and electric field spin transport devices using a spin-polarized edge state [15]. Another key property is electric field effect observed in graphene on thermally oxidized surfaces of highly doped silicon wafers (SiO2/Si) as a back gate. It was investigated shortly after the discovery of graphene [1]. Electron and hole densities are tuned using a gate voltage, which raises and lowers the Fermi energy. As a result, the conductivity decreases or increases with this energy and reaches a minimum at the Dirac points (neutrality points) K and K’. However, zero bandgap graphene gives an insufficient maximum resistance that means no distinctive off-state for graphene field effect transistors (GFETs) occurs. Table 1 summarizes the reported carrier mobilities of different types of graphene deposited on typical substrates.

5


Table 1.1 Summary of reported charge carrier mobilities of graphene deriving from different synthesis approaches and suspended or deposited on different substrates. The mobility is calculated from the field effect transconductance (FET) or quantum Hall effect (QHE) measurements. Synthesis

Substrate

Suspended

hBN

Temperature

Measurement

Mobility

(K)

method

(cm2/V.s)

 4K

FET

1.0 × 106

[16]

240 K

FET

1.2 × 105

[17]

RT

FET

1.0 × 104

[18]

4K

FET

1.4 × 105

[19]

QHE

2.5 × 104

2K

FET

6.0 × 104

RT

FET

4.0 × 104

RT

FET

2.4 × 104

Mechanical exfoliation (ME)

OTS on SiO2 SiO2 SiO2 + topgate

-

Chemical

vapor

deposition

hBN

1.6 K

1.6 K SiO2

graphene (EG)

rGO

SiC

SiO2

-

FET : FET

[20]

[21] [22]

7.0 × 104 : 2.1 × 10

+

2.7 × 104 :

[23]

2.4 × 104

QHE

2.5 × 104

4K

QHE

2.7 × 105

[24]

27 K

QHE

2.0 × 103

[25]

RT

QHE

1.8 × 103

[26]

3.6 × 102 -

[27]

RT

FET

5.0 × 103

Graphene nanoribbons

[23]

4

6.5 × 104

QHE

(CVD)

Epitaxial

FET : FET

+

Ref.

[28] SiC

RT

QHE

3

< 3.0 × 10

(GNRs)

6


1.1.2 Optical properties Graphene’s first commercial applications may be at the intersection of electronics and optics, in the form of transparent electrodes replacing indium tin oxide (ITO). Those electrodes can carry high currents but also allow light to pass through. The electrodes collect current from solar cells without blocking the light, or inject current to power organic light-emitting diodes (OLEDs) while allowing most of the photons to escape. Graphene shows a linear optical absorption: the resulting optical image contrast enables the use of an optical microscope to identify graphene sheets on top of a Si/SiO2 substrate. The contrast scales with the number of layers and can be maximized by adjusting the spacer thickness or the light wavelength [29]. The transmittance of a freestanding SLG can be derived by applying the Fresnel equations in the thin-film limit for a material with a fixed universal optical conductance [30] Go = e2/(4ħ)  6.08 × 10-5 -1, to give: T = (1 + 0.5πα)-2  1 – πα  97.7 (%)

(1.6)

where T is transmittance, α = e2/(4πεoħc) = Go/ πεoc  1/137 is the fine-structure constant [31]. Graphene reflects less than 0.1% of the incident light in the visible region [31], and up to 2% for 10 layers. Therefore, the optical absorption of graphene layers is proportional to the number of layers, each absorbing A  1 – T  πα  2.3% across all spectrum from the ultraviolet to the far infrared with a peak in ultraviolet region ( 270 nm) (Fig. 1.4 A, B).

Figure 1.4 (A) photograph of graphene and its bilayer covered on a 50 μm aperture. (B) Transmittance spectrum of SLG, (Inset) Transmittance of visible light as a function of the number of layers (reproduced from ref. [31]). (C) Schematic illustration of energy cascade with two different photo-energies hf and hf’ (hf’ > hf); photoexcitation creates a primary hot exciton (grey and orange balls) and induces a cascade of carrier-carrier scattering steps, where energy is transferred to multiple secondary hot electrons in the conduction band. (Reproduced from ref. [32])

7


This unique property arises from graphene’s electronic structure which lacks a bandgap. The wide range of absorption is very valuable for applications of graphene in photodetectors with respect to the detection of light of any wavelength, whereas a lot of materials only absorb certain wavelengths. Photoexcitation cascade and multiple hot-carrier generation in graphene were recently reported [32]. Photo-induced electron-hole pairs do not lose energy as heat, but instead transfer their excess energy for the generation of additional electron-hole pairs through carrier-carrier scattering processes (Fig 1.4 C). This phenomenon makes graphene potential for highly efficient broadband conversion of light energy into free electrons, enabling high efficiency in optoelectronic applications. 1.1.3 Mechanical properties The Young’s modulus and fracture strength of graphene have been studied by various simulations such as molecular dynamics [33,34]. Experimentally, force-displacement measurements were used for extracting the Young’s modulus of 0.5 TPa for few-layer graphene by atomic force microscopy (AFM) where a strip of graphene was suspended over trenches [35]. It was reported that defect-free graphene has a Young’s modulus of about 1.0 TPa and a fracture strength of 130 Ga [36]. A paper-like material that is made from graphene oxide platelets has the average elastic modulus and the highest fracture strength of  32 GPa and 120 MPa, respectively. These values are much higher than those reported for bucky papers from carbon nanotubes and flexible graphite foils [37]. The mechanical properties of graphene oxide paper were improved by introducing chemical cross-linking between individual graphene flakes. The introduction of divalent ions improved the mechanical stiffness (10–200%) and the fracture strength (∼50%) [38]. And polyallylamine enhanced the modulus by 30% [39]. 1.1.4 Thermal properties The thermal conductivity (k) of graphene is mainly based on phonon transport, i.e. diffusive conduction at high temperature and ballistic conduction at sufficiently low temperature. A suspended monolayer graphene theoretically has a thermal conductivity of about 6000 W m-1 K-1 at room temperature which is much higher than that of graphitic carbon [40]. Experimentally, a thermal conductivity of about 5000 W m-1 K-1 for a suspended monolayer graphene flake obtained by mechanical exfoliation was recently reported using the shift in the 8

1.2 Graphene synthesis

Raman G band [41]. With the high k value, graphene can even outperform carbon nanotubes in heat conduction, and is therefore a promising material for heat management in electronic applications. A thermal conductivity of about 2500 W m-1 K-1 at 350 K was also measured for graphene synthesized by chemical vapor deposition (CVD) and suspended on a circular hole of a thin silicon nitride membrane coated with a thin layer of gold for ensuring better thermal contacts [42]. A calculation based on the Boltzmann equation predicts a particular dependence of k on the width d of graphene nanoribbons (GNRs) and on the ‘roughness’ of the edges, when diffusive conduction dominates k [43].

1.2 Graphene synthesis 1.2.1 Mechanical exfoliation (ME)

Figure 1.5 Graphene exfoliated from graphite by using Scotch tape [44] The discovery of graphene is based on the ME method performed by A. Geim and K. Novoselov. Before that, Ruoff and coworkers created an array of small graphite islands by reactive ion etching of the highly oriented pyrolytic graphite (HOPG) surface with oxygen plasma. These islands were then manipulated with an atomic force microscope tip with the goal of isolating single-layer graphene sheets [45]. Afterwards, Ruoff and colleagues predicted some fundamental properties of graphene in relation to carbon nanotubes, inspiring physicists to discovery graphene. A. Geim and K. Novoselov used a Scotch tape to peel off single layers of graphene [1]; first, the tape was pressed onto a chunk of graphite and then pulled back. This eventually peeled off a thin flake of grey-black carbon (Fig. 1.5). Then the carbon-covered tape was repeatedly sticked against itself and peeled away: the carbon flake breaks up further into thin, faint fragments of about hundred micrometers in lateral size. After 9


a few repetitions, some flakes have been peeled away into single-atom thick graphene. The graphene flakes were then released in acetone and captured onto the surface of a Si/SiO2 wafer. By applying this technique, Geim and Novoselov obtained few- and single-layer graphene flakes of up to 10 μm in lateral size. The ME method delivers the highest quality graphene suitable for fundamental studies of electrical transport properties and other physical properties, but does not yet appear to be scalable to obtain large-area graphene needed for potential device applications. 1.2.2 Epitaxial growth on silicon carbide SiC (EG) In this method, graphene is synthesized by heating a SiC substrate at 1200 – 1600 C in an ultra high vacuum chamber (UHV) [46–48]. Under this condition, Si atoms on the surface sublimate from the substrate leaving the carbon-rich surface, and carbon atoms are then arranging and bonding together into a hexagonal structure forming graphene on the SiC surface. EG can occur on either the C-terminated (000 ) or Si-terminated (0001) surfaces, however, the layers grow much faster on the C-terminated surface [49,50] due to the weak coupling of the substrate to the growing graphene layers. Several micrometer large domains have been achieved by changing from normal UHV conditions to an argon atmosphere [25]. An advantage of the EG method is that SiC can also serve as a suitable insulating layer in graphene devices. Graphene obtained by EG can be transferred to arbitrary substrates by using a thermal release tape without a significant drop in carrier mobility [51]. However, the commercialization of graphene synthesized by EG is hindered by the use of expensive singlecrystal SiC wafers, as well as ultra-high vacuum and high-temperature synthesis conditions. 1.2.3 Chemical vapor deposition (CVD) Chemical vapor deposition is one of the most promising methods for synthesizing large-area continuous graphene films with controllable single- or few-layer graphene sheets. Carbon is supplied in gas form by pyrolysis of hydrocarbon precursors, typically methane for example. Graphene is grown on the surface of transition metal catalyst loaded substrates such as copper or nickel at either high- or low-vacuum conditions at high temperature (typically 1000 C). By applying this method in combination with roll-to-roll techniques, Bae et al. [52] produced graphene sheets as large as 30 inch in diagonal with electrical sheet resistances as low as 30 -1 and  90% optical transparency. Growth of single-crystal monolayer graphene on reusable hydrogen-terminated germanium has been recently reported [53] and is a significant step towards a potential commercialization. 10


a) Growth on nickel (Ni). Polycrystalline Ni-foils are used to grow few-layer graphene sheets. The foil is first annealed in hydrogen and then exposed to a reaction gas mixture of CH4/H2/Ar at ambient pressure for ca. 20-30 min at a temperature of 1000 C [54,55]. The samples are then cooled down to room temperature. Different cooling rates are reported, for example 10 C/s or 0.1 C/s for rapid or slow cooling, respectively. Rapid-cooling rates can suppress the formation of multiple layers and graphene layers can be easier separated from the substrate [55]. Nickel has a relatively high carbon solubility. Carbon atoms can dissolve into Ni at high temperature, and then precipitate onto the metal surface and form single- or multiple-layer graphitic films upon cooling. It was demonstrated that graphene growth on Ni follows a precipitation mechanism, which is a non-equilibrium process (Fig. 1.6) [56]. Therefore, it is difficult to control the number of layers and the uniformity of graphene as well as to obtain single-layer graphene. Ni catalysts are often used to synthesize multi-layer graphene sheets, whereas copper foil is usually used for obtaining single-layer graphene. b) Growth on copper (Cu). Typically, 25 μm thick copper foils as catalytic substrates are first annealed at 1000 C for ca. 60 min in a flow of hydrogen in vacuum to increase the size of Cu grains and smoothen grain boundaries, and then exposed to a mixture of CH4/H2 for 30 min typically at low pressure, and finally cooled down to room-temperature. Different cooling rates have been reported ranging between 40 C/min and 10 C/s [57]. It was found that graphene growth on Cu is a self-limited process and yields to similar results from different growth times ranging from 10 min to 60 min and more. For growth times less than 10 min, the Cu surface is usually not fully covered with graphene [57]. Because of the very low carbon solubility in Cu and poor carbon saturation, graphene growth on Cu is based rather on a surface adsorption mechanism than on a surface segregation mechanism which was demonstrated by Li et al. [56]. At high temperature, methane precursors are decomposing into carbon atoms which are absorbed onto the Cu surface. Carbon atoms then move onto the surface of the Cu substrate, and collide with each other forming chemical bonds resulting in graphene lattices. As depicted in Fig 1.6,

13

CH4 and

12

CH4 precursors were introduced

sequentially during the graphene growth process. In the case of Ni catalysts (segregation mechanism) (Fig 1.6a) 12C is diffusing into the whole surface of Ni, consequently 12C and 13C distribute randomly within the graphene lattice. For the case of Cu catalysts, the graphene lattice is formed with

13

C and

12

C in consequence, corresponding to the

13

CH4 and

12

CH4

precursor supply proving a surface adsorption mechanism.

11


Figure 1.6 Schematic illustration of two graphene growth mechanism: (a) Precipitation mechanism for Ni catalyst substrates and (b) Adsorption mechanism for Cu catalyst substrates.

13

CH4 and

12

CH4 precursors are introduced sequentially during the graphene

growth processes (reproduced from ref. [56]). To make the CVD approach more practical, the growth temperatures need to be reduced as much as possible. To this end, graphene growth at a temperature as low as 300 C was recently demonstrated using a liquid benzene hydrocarbon source [58]. By utilizing plasma enhanced CVD, graphene can be grown at moderate temperatures of about 700 C [59]. A laser-induced CVD process was used to directly write graphene-patterns. Here the laser beam creates a local temperature-rise enabling a precisely controlled carbon deposition at room temperature with a growth rate of one thousand times faster than for typical CVD processes [60]. A useful feature of CVD is the possibility to grow doped-graphene by introducing additional dopant molecules such as ammonia (NH3) or borane (BH3) into the gas mixtures. Nitrogen or boron replaces carbon atoms within the lattice to form N- or B-doped graphene [61]. In addition, solid state carbon sources such as poly(methylmethacrylate) (PMMA) or other carbon-based materials such as even food or insects etc [62] were used for the synthesis of 12


graphene instead of using hydrocarbon precursor gases. In another method, a thin film of PMMA deposited on a Cu catalyst substrate was prepared by spin-coating and heated up to 800 C in an Ar/H2 flow for graphene growth. As doping reagent, melamine (C3N6H6) was mixed with PMMA for growing nitrogen-doped graphene at atmospheric pressure [63].

Figure 1.7 Roll-to-roll transfer of CVD-graphene from a catalytic Cu foil to a target substrate, image was taken from ref [52]. A drawback of CVD-synthesized graphene is the need of transferring the graphene to insulating substrates for further characterization or applications. The transfer is often complicated, time-consuming, and costly and especially downgrades the graphene quality. This

step

is

normally

performed

by

coating

a

protective

polymer

such

as

polydimethylsiloxane (PDMS) or PMMA on top of the graphene thin film followed by etching the underlying copper layer in iron chloride (FeCl3 in HCl) [64] or 1 M Fe(NO3)3 solution [65]. The Cu substrate is removed by immersing the substrate with the graphene film into the etching solution until the free-standing graphene membrane is floating on top of the solution. The polymer-supported graphene sheet is sufficiently stable for transferring graphene onto arbitrary substrates. Afterwards, the polymer is removed by dissolving in acetone. Thermal release tape is also used as a support polymer adhering to the as-grown graphene on Cu substrates to enable a roll-to-roll transfer as previously reported [52] where 30 inch-graphene sheets were transferred onto flexible PET plastic substrates (Fig 1.7). Lee et al. [53] recently reported the growth of graphene on a hydrogen-terminated germanium buffer layer. There, graphene has only weak interaction with the underlying buffer layer allowing the direct and etch-free transfer of graphene and the recycling of the germanium substrate.

13


1.2.4 Solution approach to reduced graphene oxide through graphite oxide

Figure 1.8 Schematic illustration of graphene oxide synthesized from graphite via graphite oxide. Graphite oxide or oxidized graphite was firstly synthesized by Staudenmaier [66] in 1898 and Hummer [67] in 1958. They used a mixture of strong acids and oxidants to oxidize graphite into dissolvable graphite oxide. By this, oxygen-containing groups such as carboxyl and hydroxyl groups are introduced into both the edges and the plane of the graphene lattice. Graphite oxide therefore becomes strongly hydrophilic and water molecules can intercalate between the graphene layers. The interlayer distance within graphene oxide (GO) sheets increases and graphite oxide can be exfoliated forming an aqueous solution of GO by the aid of sonication as shown in Fig. 1.8. Functional groups cause GO sheets to become negatively charged and therefore repulse each other forming a stable aqueous suspension of GO. So far, great efforts have been dedicated to develop reduction methods for achieving reduced GO from GO. This can be classified as following: the use of chemical reduction agents such as hydrazine [68–70], sodium borohydride [71,72], hydroquinone [73], sulfur-containing compounds [74,75], vitamin C [76–78], protein [79]), or using alternative reducing conditions such as thermal treatment [80–82], photocatalytic reduction [83–85], laser treatment [86,87], plasma treatment[88], electrochemical reduction [89–91], or utilizing a two-step reduction process by combining two methods, i.e. performing chemical reduction, 14


followed by an thermal annealing step [92,93]. These different reduction methods result in rGO with different properties. For example, hydrazine-rGO is an aqueous dispersion without surfactant as stabilizers. However, hydrazine is highly toxic and explosive, so it is prohibited to be used for large-scale practical applications. Recently, two or more types of reduction methods have been combined to obtain high electrical conductivity of rGO [94]. A challenge for any reduction method is that rGO tends to irreversible agglomerate due to van der Waals forces, limiting further processing. Normally, a higher reduction degree leads to better electrical conductivity rGO, but also to more agglomerates resulting in significant material loss. Therefore, depending on the application which often demands high conductivity or solubility, one has to choose a suitable reduction method to obtain the desired type of rGO. 1.2.5 Unzipping of carbon nanotubes Graphene nanoribbons (GNRs) can be directly produced by longitudinally unzipping carbon nanotubes after chemical treatment. For example, multiwall carbon nanotubes (MWCNTs) are suspended in concentrated H2SO4 and then KMnO4 is added under stirring at room temperature. Thereby, MWCNTs are oxidized and lengthwise cut into oxidized GNRs. Subsequently, the oxidized GNRs are chemically reduced to restore the electrical conductivity [95]. These GNRs have an advantage in term of scalability and production cost, but bears severe lattice defects, resulting in graphene with inferior electrical properties in comparison with GNRs prepared by other methods such as lithographic procedures. 1.2.6 Other methods Molecular beam deposition: a thermal-cracker enhanced gas source molecular beam epitaxy system is used to grow layer-by-layer graphene films on a nickel substrate at 800 C. This method does not require a fast cooling rate, and is capable of subsequent growth of graphene layers on top of already existing layers. The growth follows a surface mechanism rather than a conventional precipitation observed for CVD on a nickel catalyst substrate [96]. 2-4 layer thick graphene flakes can be synthesized by arc discharge between graphite electrodes at a relatively high hydrogen pressure [97]. Colloidal graphene sheets suspended in organic solvents such as DMF, NMP were achieved by sonication of graphite powder in the respective solvents. This method leads to low yields, delivering 0.01 mg/l graphene suspensions and 1-12 wt% of single-layered graphene of high quality [98].

15

1.3 Applications of graphene-type structures

1.3 Applications of graphene-type structures Many of the outstanding properties of graphene might be utilized in future-promising applications with prospect of influencing technologies ranging from consumer electronics to energy production, replacing silicon in device applications or opening the way for completely new applications. It’s extremely strength and flexibility, for example, enables bendable devices. Electrical conductivity in combination with optical transparency renders graphene for application in optoelectronics while its super-large specific area is beneficial for sensor applications or as catalyst support. 1.3.1 Flexible transparent conductors (FTCs) Transparent conductors (TCs) are crucial components of optoelectronic devices such as display panels. They play a significant role as electrodes and allow light to pass through. Conventional TCs are made from indium tin oxide (ITO) or less popular fluorine-doped tin oxide (FTO). However, ITO TCs have some disadvantages, including the shortage and high cost of indium; brittleness of ITO makes it impossible for using it as flexible TCs; indium diffusion into adjacent film layers of devices such as solar cells or OLEDs will degrade the device performance. With respect to this, graphene is a promising candidate to complement or even replace ITO as TCs, overcoming the brittleness problem of ITO and has the additional potential to reduce the costs for TCs [99]. A single-layer graphene lets 97.7% of visible light pass through, and its electrical resistance is often larger than 0.1 k/ depending on the graphene quality and doping conditions. Meanwhile ITO has a sheet resistance of less than 100 / and an optical transparency of 90%. So a single-layer graphene sheet doesn’t reach the conductivity values of normal TCs. Some strategies have been implemented to improve the electrical conductivity: e.g. synthesizing multi-layer graphene sheets [100], the transferring layer by layer to obtain few-layer graphene films, and the doping of single-layer graphene. Bae et al. [52] used a roll-to-roll transfer technique and wet-chemical doping of a 30-inch graphene monolayer film grown by CVD and obtained single-layer films with sheet resistances of 125 / and 97.4% optical transparency. Furthermore, layer-by-layer stacking was used to fabricate a doped four-layer film with sheet resistances as low as 30 / at  90% optical transparency, which is superior to commercial available ITO TCs. Other methods for graphene TCs are the deposition of graphene out of solution onto transparent substrates, whereby graphene flakes overlap and interconnect forming graphene thin films by applying vacuum filtration [101,102], printing [103], drop-casting [104], 16


dipping [105], spraying [69] or spin-coating [106] techniques. The optical transparency is inversely proportional to the film thickness while the electrical conductivity increases with the film thickness. Thus, to get good graphene TCs, the thickness must be optimal for compromising between electrical conductivity and optical transparency. Aqueous dissolvable graphene oxide (GO) is often used as starting material to fabricate GO thin films and must be chemically reduced to increase the conductivity of the reduced graphene oxide (rGO) films. A thermal annealing step is usually used to improve the degree of reduction and solidify the percolation network of the rGO flakes, achieving rGO films with resistances as low as 800 / at 82% optical transparency [107]. The rGO thin films have conductivities and transparencies less than those of ITO TCs, but give cost effectiveness by avoiding high vacuum evaporation and high-temperature annealing steps. TCs from both, CVD graphene and graphene made out of exfoliated graphite, demonstrate reasonable quality for many promising applications. However, to fully replace ITO, it still needs much effort to further improve the quality and reduce the production costs of graphene by developing optimized growth technologies as well as suitable graphene transfer techniques applicable for large scale. The optical transparencies and electrical conductivities of rGO TCs must be significantly enhanced to meet the standard of current state of the art TCs. 1.3.2 High-frequency transistors Due to the lack of a band-gap, intrinsic graphene transistors cannot be fully switched off, which obstructs its application for digital electronics, but does not rule out the use in analog radio-frequency (RF) devices [108]. An exceptional high carrier mobility of 106 cm2 V-1 s-1 at low temperature (4 K) was found for devices based on suspended graphene [16] and room temperature carrier mobilities of 4 × 104 cm2 V-1 s-1 were observed for SiO2 supported graphene based devices [21], which facilitates the fast operating speed of graphene transistors. Furthermore, a unique feature of graphene is the carrier velocity that is theoretically calculated and has a peak intrinsic average up to 4.6 × 107 cm/s [109], twice of that of GaAs and 4 times of that of Si. This high carrier velocity is important for transistor application, because if the device size is reduced to the nanometer regime, the channel length is significantly shorter and therefore the saturated carrier velocity becomes a more significant measure of a transistor. Liao et al. [110] used a self-aligned nanowire gate to fabricate graphene transistors with a channel length as short as 140 nm and a high intrinsic cut-off (transit) frequency of fT = 100300 GHz. However, because of the large ratio between 17


parasitic pad capacitance and gate capacitance, the intrinsic fT is often in the range of a few gigahertzes without de-embedding procedure which is a mathematical process for subtracting the effects of device structure embedded in the measured data. To avoid this problem, a glass substrate was employed resulting in an extrinsic cutoff frequency up to 55 GHz [111]. Although high values of fT have been achieved, the maximum oscillation frequency fmax is still only a few tens of gigahertz. The power gain performance and switching speed need to be improved for the realization of high-performance transistors. In research, graphene transistors are usually fabricated on SiO2/Si wafers with a highly doped silicon layer on the back site which is used as a back gate electrode, requiring usually high gate voltages in the range of several dozens of volts. So, for practical applications, a top gate structure is preferably used for the realization of graphene transistors. A dielectric oxide such as Al2O3 is deposited by atomic layer deposition (ALD) on top of the graphene layer following a pattern of metal layers as the top gate. The incompatibility of graphene with most metal oxides challenges the deposition of a good dielectric layer on it, so a prior treatment of the graphene surface with NO2 for example is applied before oxide deposition. As a result, a uniform layer of less than 10 nanometer thin Al2O3 was successfully grown on graphene without pinholes by ALD and was firstly reported by a group from IBM [112]. Although, the NO2 treatment degrades the mobility of graphene down to 400 cm2/Vs, the transistor shows a typical dependence of the intrinsic current gain on 1/f. To improve the quality of top-gates, Lin et al. [113] deposited a thin layer of metal, which was then oxidized afterwards to form a thin dielectric layer as a nucleation layer for further oxide growth by ALD. By this approach, a dual-gate graphene field-effect transistor (GFET) was fabricated with improved RF performance by reducing the access resistance using electrostatic doping. Therefore, a cutoff frequency of 50 GHz was obtained for the GFET with a 350 nm long gate electrode. Epitaxial graphene grown on the Si-face of 2-inch SiC wafers with Hall mobilities up to 4000 cm2 V-1 s-1 was used for top-gated RF GFETs exhibiting a peak cutoff frequency of 100 GHz [114]. CVD methods hold the most potential for up-scalability amongst approaches which produce continuously large area graphene, so research on CVD graphene is important for the realization of graphene based applications. Wu et al. at IBM [115] reported in 2012 a significant result for a CVD GFET with intrinsic fT above 300 GHz for a 40 nm channel length and fmax of 44 GHz for a 140 nm channel device. These remarkable results were obtained due to the improvement in graphene quality as well as doping control of graphene channels with the Al2O3 dielectric gate electrode. 18


1.3.3 Energy storage systems High mechanical strength, superior conductivity and extremely large specific surface area are key features which make graphene a great candidate for battery as well as for supercapacitor applications. Graphene can serve as active and supporting material as well as current collectors for both cathodes and anodes. a) Graphene-based material for Li-batteries Lithium-ion batteries (LIBs) usually have high energy densities (120-170 W h kg-1), long life times and higher safety performance compared to traditional batteries. LIBs normally need a longer charging time (ca. 6 h) and exhibit lower power densities. They are therefore more suitable for low power devices such as mobile phones and laptops. For their usage in hybrid electricity vehicles, the power density, specific capacity and charging time must be significantly improved. The power capability critically depends on the speed that Li-ions and electrons need to migrate through the electrolyte towards the electrodes. This speed can be enhanced by improving the electrical conductivity of the electrode material and shortening the Li-ion migration length through engineering a nano-structured surface of the electrode materials. This might be realized by using graphene as electrode material due to its high carrier mobility and intrinsic nano-size. Graphene for anodes Carbon materials have been already commercially used for LIB anodes in the form of graphite with a theoretical specific capacity of 372 mA h g-1 based on the formation of LiC6 [116]. The specific capacity of a battery is the amount of electric charge that it can deliver at the rated voltage per weight unit (g, kg). As a building block of graphite, graphene has the same planar structure of the carbon lattice but a different 3D structure with the expansion of the d-spacing between graphene layers that increases the space for the accommodation of Li ions. A pioneering work [117] on the usage of graphene nanosheets (GNS) as LIB anode was reported in 2008, where a specific capacity of 540 mA h g-1 was achieved. This value was then increased up to 730 mA h g-1 and 784 mA h g-1 by incorporation of CNT and C60 to the GNS, respectively, which are superior to that of graphite. Lian et al. [118] utilized few layers of graphene with a large specific surface area in coin-style battery cells and obtained the first reversible specific capacity as high as 1264 mA h g-1 remaining at 848 mA h g-1 after 40 cycles at a current density of 100 mA g-1. The battery cells also exhibited a specific capacity of 718 mAhg-1 at a high current density of 500 mA g-1. However, the reversible capacity 19


undergoes a large capacity fluctuation at a high charge-discharge rate of 500 mA h g-1 or even higher because of surface side reactions forming solid electrolyte interphase films. Moreover, in the delithiation state, oxygen that decomposes from oxygen-containing functional groups will partly oxidize the electrolyte, leading to an electrochemical instability of the electrodes. Removal of oxygen residues from rGO might enhance the stability of graphene electrode by reduction methods such as annealing at high temperature in vacuum or chemical reduction, but decreases the wettability of the electrode with the electrolyte and therefore the overall performance of a battery. Another strategy to improve the battery performance is the introduction of heteroatoms into graphene lattices by N or B-doping, for example. The doped-graphene based battery shows a high reversible capacity of  1040 mA h g-1 at a low rate of 50 mh g-1 [119]. This also displays a fast charge-discharge of about 1 h to dozens of seconds. In addition, a high rate capability of 199 and 235 mA h g-1 was achieved for Ndoped and B-doped graphene, respectively, at 25 A g-1 (about 30 s to reach full charging) and excellent long-term cyclability was found [119]. The good result was explained by the unique two-dimensional structure of graphene, the disordered surface morphology, heteroatomic defects, a better electrode/electrolyte wettability, an increased intersheet distance, an improved electrical conductivity of the doped graphene. These are all beneficial for a rapid Li-ions absorption on the surface and ultrafast ion diffusion and electron transport. Metal oxides such as Fe2O3, Co3O4, Mn3O4 and NiO have been already proven as good anode materials for batteries due to their high theoretical Li-ion storage capacities (600 mA h g-1) [120–122]. However, they often bear the disadvantages of a low electrical conductivity and poor capacity retention caused by a pulverization process due to weak mechanical tolerance to the volume expansion/contraction during the working time. This can be solved by using graphene as a supporting material for anchoring metal NPs and as current transporter. The anchoring of NPs to graphene will protect them not only from the pulverization process due to the volume expansion/contraction, but also from aggregation during charging and discharging. In return, NPs intercalated between graphene layers can reduce the restacking of the layers and maintain the high active surface area of graphene. Zhou et al. [123] reported that graphene nanosheets (GNSs) decorated with Fe3O4 particles show a reversible specific capacity approaching 1026 mA h g-1 after 30 cycles at 35 mA g-1 and 580 mA h g-1 after 100 cycles at 700 mA g-1. Mn3O4-attached rGO hybrid material reaches a high specific capacity of 900 mA h g-1, which is close to the theoretical capacity for Mn3O4. And even at a high current density of 1600 mA g-1, the specific capacity remains at 390 mA h g-1 whic is

20


superior to that for Mn3O4 with a specific capacity of 100 mA h g-1 at a current density of 40 mA g-1 [124]. Another promising material for high performance batteries, namely rGOencapsulated Co3O4, was reported [125] in which the Co3O4/rGO composites exhibited a highly reversible capacity of  1100 mA h g-1 at a current density of 74 mA g-1 within the first 10 cycles and a remarkable discharge capacity of 1000 mA h g-1 remained even after 130 cycles. Graphene-based materials for cathodes Common cathode materials often suffer from poor electrical conductivity, sluggish kinetics of electron and Li-ion transport, low specific capacity and particle aggregation. Thus, there is still a great need for searching new materials as well as new dedicated nano-structuring techniques for materials to improve the overall battery performance. Olivine-structured LiMPO4 (M = Fe, Mn, Co, or Ni) recently attracted great interest as cathode material for rechargeable LIBs due to their outstanding properties such as high specific capacity, good cyclability, high thermal stability and low cost [126,127]. However, the low electrical conductivity hinders them from practical applications so far. Graphene was incorporated to provide conducting network amongst LiFePO4 NPs, while keeping the porous structure and allowing the penetration of the electrolyte and reducing the diffusion length of the Li-ions. Thereby, a high specific discharge capacity was improved from 125 mA h g-1 to a value of 165 mA h g-1 at a rate of 0.1 C and from 50 mA h g-1 to 88 mA h g-1 at a rate of 10 C [128]. Wang et al. [129] obtained ultrahigh-rate performance Li-ion batteries by using LiMn0.75Fe0.25PO4 nanorods grown on graphene. By this way, a fast Li-ion diffusion was achieved leading to specific capacities of 107 mA h g-1 at a rate of 50 C and 65 mA h g-1 at a rate of 100 C. Furthermore, the cell exhibited an excellent cyclability with a specific capacity of 155 mA h g-1 even at the 100th cycle at a rate of 0.5 C. In another attempt [130], vanadium dioxide (VO2) - graphene ribbons were used with unique architectures providing channels for the access of electrolyte, short diffusion lengths for Li-ions, a high electrical conductivity and a high loading of the electrode with electrochemically active material (84 wt%). As a result, a high reversible capacity (410 mA h g-1 at a current density rate of 5 C) and an ultrafast charging and discharging capacity was achieved for LIBs with full charging and discharging within 20 s, while more than 90% of the initial capacity remained after 1000 cycles at an ultrahigh rate of 190 C, which is a remarkable breakthrough in rate performance for cathodes in LIBs. Ultrathin V2O5 nanowires were uniformly incorporated into graphene paper as a cathode achieving an extremely high cycle lifetime [131]. After 100 000 cycles the composite 21


paper with 15 wt% of V2O5 still showed a capacity of 94.4 mA h g-1 at a high current density of 10 000 mA g-1. b) Graphene-based material for supercapacitors (ultracapacitors) Compared to batteries, supercapacitors (SCs) or electrical double layer capacitors (EDLCs) have higher power density and a better cyclability. They require a very simple charging circuit and have no memory effect. They therefore show a very high potential for applications in hybrid electric or electric vehicles. However, SCs suffer from a low energy density of only 5-10 W h kg-1 for commercial available products so far, which is inferior to that of 120-170 Wh kg-1 for commercially available LIBs [131]. With unique properties such as an ultrahigh specific surface area, graphene emerges as a promising material which can improve the performance of SCs by increasing the key features, i.e. the increase of the active surface area and electrical conductivity of capacitor’s electrodes. In an early report [132], chemically modified graphene mixed with 3 wt% polytetrafluoroethylene binder was used as materials for ultracapacitor electrodes. Although relatively low specific capacitances of 135 and 99 F/g in aqueous and organic electrolytes were obtained, respectively, the results illustrated the exciting potential for high performance SCs based on a low cost carbon material. The main drawback of using graphene as electrode material is aggregation and restacking between graphene sheets driven by the strong π-π interactions resulting in the reduction of the active surface area of the graphene based electrodes. Following strategy exists to overcome this restacking problem: decorating graphene with polymers or NPs. For example, Yang and coworkers [133] demonstrated that water can serve as an effective “spacer” to prevent the restacking of chemically reduced graphene. Consequently, the authors obtained a highly open pore structure of the graphene film for graphene-based SCs with a capacitance of 156.5 F g-1 at an ultrafast charge/discharge rate of 1080 A g-1. A porous graphene-based paper with a BET surface area of up to 3100 m2/g was synthesized by chemical activation of exfoliated graphite oxide. A two-electrode supercapacitor cell was constructed with this carbon material yielding to a specific capacitance of up to 165 and 166 F g-1 at current densities of 1.4 and 5.7 A g-1, respectively [134]. Impressively, N-doped rGO based SCs delivered a large specific capacitance of 282 F g-1 at a high current density of 1 A g-1, which is about 4 times larger than that of pristine rGO. The cells showed an excellent cyclability with high power capabilities still remaining after 200 000 cycles [135]. In another approach [136], vertically oriented graphene (VG) nanosheets were grown on nickel foam by plasma-enhanced chemical vapor deposition (PECVD). When the nickel foam was applied in SCs, it serves as a 22


current collector directly connected with VG-nanosheets acting as brides interlinking the current collector and the conventional graphene film active layer. The VG-bridged SCs showed a remarkable high power density of 112.6 kW kg-1 (specific capacitance of 130 F g-1) at a current density of 600 A g-1. Although some achievements have been made with SCs, there is still a large gap to overcome reaching the energy densities for commercial devices. Alternatively, pseudo-supercapacitors have recently been emerging to take advantage of redox reactions taking place on the surface of the metal oxide to store energy for capacitance enhancement. In a remarkable report [137], graphene was coated on porous textiles support structures and highly loaded with MnO2 active electrode material. The hybrid graphene/MnO2-based textile yielded to a specific capacitance up to 315 F/g at a scan rate of 2 mV s-1, which is considerably higher than that of the pure-rGO textile due to the additional redox response of MnO2 with electrolyte ions. Polymers with active functional groups are also grafted with graphene for their usage as flexible SCs. For example, polyaniline nanorods were in-situ electrodeposited on the surface of rGO micro-patterns resulting in micro-SCs with high capacitance of 970 F g-1 at a discharge current density of 2.5 A g-1 [138]. To increase the active sites, three-dimensional graphene foams were in situ grown with Co3O4 nanowires [139]. The 3D graphene/ Co3O4 free-standing electrode for SCs delivered a super high specific capacitance of  1100 F g-1 at a current density of 10 A g-1. 1.3.4 Light–emitting devices Due to high image quality, low power consumption and flexible thin device structure, organic light-emitting diodes (OLEDs) are widely utilized as display screens for ultrathin televisions, computer monitors, mobile phones and digital cameras. OLEDs contain an organic electroluminescent layer between two charge-injecting electrodes; at least one of them is transparent. During their operation, holes are injected into the highest occupied molecular orbital (HOMO) from the anode, while electrons are injected into the lowest unoccupied molecular orbital (LUMO) from the cathode. The work functions (WF) of anode and cathode have to match HOMO and LUMO energies of the light-emitting layer for a more efficient charge injection [140]. Conventionally, ITO is used as the transparent electrode with a wellmatching work function of 4.44.5 eV, but bearing high cost and its brittleness limits the usage as a flexible electrode. Moreover, indium tends to diffuse into the active layers, which degrades the device performance over time [141]. Graphene sheets with a work function of 4.5 eV, which is similar to that of ITO, and combined with its promise as a flexible and cost23


effective TCF, is an ideal candidate for an OLED anode. Graphene TCF anodes eliminate the indium-diffusion problem and enable an out-coupling efficiency comparable to ITO [142]. Matyba et al. [143] utilized chemically modified graphene as TCFs in an all-plastic sandwich-structure, called a light-emitting electrochemical cell. The report demonstrated low-voltage, inexpensive, and efficient light-emitting devices without using any metals. By using conducting polymer compositions to modify the surface and by doping with HNO 3 or AuCl3, the sheet resistance of graphene was reduced to 30 /, and the surface WF increased to 5.95 eV which is higher than the WF of the organic light-emitting layer favoring the hole injection. As a result, Han and co-workers [144] achieved flexible OLEDs with extremely high luminous efficiencies of 37.2 lm W-1 in fluorescent OLEDs and 102.7 lm W-1 in phosphorescent OLEDs which are superior to optimized ITO-based reference devices with 24.1 lm W-1 and 85.6 lm W-1, respectively. 1.3.5 Photovoltaics devices For sustainable development of the world, production of renewable energy is in urgent need. The energy generated from photovoltaics (PV) recently has been growing fast. The world’s cumulative installed PV capacity was 23, 40.3 and 70.5 GW at the end of 2009, 2010 and 2011, respectively and by 2013, almost 138.9 GW of PV was installed globally [145]. However, it still only accounts for less than 1% of the total worldwide energy consumption at present [145]. The low contribution portion of PV energy is due to its relatively higher cost compared to other energy sources such as fossil fuel. This leads to great interest in finding new materials as well as technologies to solve the cost issue of solar cells. Graphene emerges as a promising candidate for improving the power conversion efficiency (PCE) and costeffectiveness of photovoltaics devices. The applications of graphene in PV are down to three categories: as transparent conductive electrodes (TCE); as catalytic counter electrode in dyesensitized solar cells; as part of the active layers, i.e., light harvesting material, electron and hole transport layers or Schottky junctions. The third category will be addressed in Chapter 4 of this thesis and discussed in detail. a) Transparent conductive electrodes in solar cells. Graphene-based TCE is similar to FTCs which have already been discussed in section 1.3.1. However, a good FTC does not always become a good TCE for a particular solar cell, because of several issues such as work function match between TCE and the photoactive layer or the compatibility to form good electrical contacts to charge collecting electrodes. 24


These all affect the performance of the PV systems. Mostly, TCE is applied in organic photovoltaics (OPV) for development of flexible solar panels. For this purpose, graphene sheets are usually synthesized by CVD methods and then transferred onto a transparent glass or polyethylene terephthalate (PET) substrates. By this method, a graphene/PET film was achieved, showing a sheet resistance as low as 230  -1 and an optical transparency of 72% [146]. The as-obtained OPV device with a graphene/PET film as an anode exhibited a PCE of 1.18% comparable with the performance of ITO-based reference solar cells [146]. In addition, the graphene solar cells demonstrated an outstanding capability to operate under bending conditions up to 138o. In another attempt, rGO deposited PET substrates were also used for flexible OPV devices. Although, the cell performance is low with PCE of 0.78%, mainly because of a high sheet resistance (3.2 k -1) and a low optical transparency (65%) [147], the results are promising to enable cost-effectiveness compared to CVD graphenebased OPV devices. The high sheet resistance (Rs) of graphene-based TCE and the incompatibility of hydrophobic graphene with hydrophilic poly(3,4-ethylenedioxythiophene)/ poly(styrenesulfonate) (PEDOT/PSS) buffer layers are the key challenges for graphene to replace ITO as TCE. To address those challenges, there are several strategies, including the fabrication of multi-layer graphene films, doping of graphene films with chemicals such as acids, and organic molecules. For example, graphene was doped with HCl and HNO3 and layer-by-layer stacked forming a four-layer film with a Rs of 80  -1 and an optical transparency of 90%. As a result, the graphene-based OPV exhibit a PCE as high as 2.5%, which was 80% of the PCE of the ITO based reference solar cell [148]. However, the acid dopants tend to diffuse out of the graphene film due to heating and electrical stress that degrades the overall stability of the cell. Alternatively, organic molecules such as teracyanoquinodimethane (TCNQ) were used to dope graphene, and the OPVs that were fabricated based on this graphene-TCE showed a PCE of 2.58%. Graphene was also used as front TCE of thin film solar cells, where a ZnO barrier layer was included to improve the compatibility between graphene and a CdS film. Consequently, the conversion efficiency increased from 2.81% to 4.17% [149]. Although some significant results have been obtained with OPVs made out of graphene-based TCE, there remain still many challenges such as working instability, low PCE values, and scalability of CVD methods. Those problems must be resolved before graphene can be used to replace ITO as TCE in OPVs. Table 1.2 summarizes the results reported in literature of graphene-based TCE in OPVs.

25


Table 1.2. Summary of graphene-based TCE with resistivities (Rs), transparency (T) and PCEs of the corresponding OPVs Graphene

Rs (k  )

(%)

rGO

100 – 500

85 – 95

rGO

1.6

55

CVD

0.23 – 8.3

72 – 91

0.374

84.2

material

CVDmultilayer CVD-AuCl3 doped CVD-acid doped

Device structure

T -1

0.5 – 0.3

97.1, 91.2

0.08

90

0.278

92.2

0.22

84

PCE (G represents graphene and its

(%)

Ref.

derivatives) G/CuPc/BCP/Ag

0.4

[150]

0.78

[147]

G/PEDOT:PSS/P3HT:CuPc/C60/Al

1.18

[146]

G/PEDOT:PSS/P3HT:PCBM/Ca:Al

1.17

[151]

G/PEDOT:PSS/P3HT:CuPc/C60/Ag

1.63

[152]

2.5

[148]

2.58

[153]

4.17

[149]

G/PEDOT:PSS/P3HT:PCBM/TiO2/ Al

G/MoO3/PEDOT:PSS/P3HT:PCBM /LiF/Al

CVDorganic molecules

G/PEDOT:PSS/P3HT:PCBM/Ca:Al

doped CVDgraphene

Glass/graphene/ZnO/CdS/CdTe/ graphite paste

b) Graphene for catalytic electrodes of dye-sensitized solar cells Dye-sensitized solar cells (DSSCs), also known as Grätzel cells were firstly reported in 1991 [154]. They attracted enormous attention within the scientific community due to their high PCE and low-cost production potential. Its light-conversion efficiency has been continuously improved to reach  12% [155]. Therefore, DSSCs hold a potential for commercialization in near future. However, there are still several challenges that must be overcome, including the suppression of charge recombination at the electrode interface, the high cost of the Pt-based 26


counter electrode, and so on. So far, great effort has been paid to the use of graphene as catalytic counter electrodes replacing expensive Pt, because graphene has advantages in low cost, high conductivity, and corrosion resistance, as listed in table 1.3. The first report [156] on using graphene as the counter electrode was published in 2008 and proved its well catalytic activity toward I3−/I1− redox electrolyte, although the device performance was still inferior to that of Pt-based counterparts. Since then, many reports have been published in this regard, exploiting the high surface area networks of graphene materials or metal-graphene composites. Impressively, graphene based electrodes show more effectiveness for catalytic activity with Co(bpy)3 (II/III) as redox mediator than Pt-based electrodes. Thermally reduced GO contains lattice defects and oxygen-containing functional groups which serve as active sites for the catalytic activity of I3−/I1− redox mediators. As a result, the obtained DSSCs exhibited a PCE of 5% which is comparable with the PCE of 5.5% for Ptbased cells [157]. Additionally, reports have been published using of rGO as a counter electrode and achieved remarkable results for DSSCs with PCEs of 5.69% [158], 6.81% [159], and 6.93% equal 95% to that of comparable Pt-based cells [160]. However, they could not exceed the performance of plantinized FTO electrodes. There are two main strategies to overcome the limitation of the catalytic activity of graphene material: (i) improving the nano-morphology, by normally increasing the surface area and pore sizes and (ii) increasing the intrinsic activity of the material through chemical modification. As an example of the first strategy, GNS vertically embedded on activated carbon was used as a counter electrode in a DSSC delivering a PCE as high as 7.5%. This is attributed to the well dispersed and vertically embedded GNSs which provide high electrical conductivity for facilitating the electrocatalytic activity towards the redox reaction [161]. By mechanical grinding rGO in liquid polyethylene glycol (PEG), rGO-PEG gels were obtained with 3D solid networks and rGO sheets inside. Then, a porous rGO film was prepared by doctor-blading, followed by a heating step to remove PEG. The resulting DSSC showed a good performance with a PCE of 7.19% which is slightly lower than that of 7.76% for a Ptbased reference DSSC, but the graphene-based cell demonstrated a better long-term stability [162]. To enhance the intrinsic activity, graphene is usually chemically doped with other elements such as B or N. Xue et al. [163] created N-doped graphene through annealing of GO in an ammonia vapor and agon mixture. The DSSC constructed with this doped graphene showed a high PCE of 7.07% due to the higher electric conductivity resulting in a good catalytic 27


activity. Moreover, the surface hydrophilicity induced by heteroatom doping enhances the contact of electrolyte to the catalytic electrode of the DSSC. Another approach is to take advantages of the large surface area and good electrical conductivity of graphene combining with the high catalytic activity of Pt by developing Pt-graphene hybrid electrodes. Tjoa et al. used photoassisted co-reduction of chloroplatinic acid and graphene oxide to form rGO decorated with 3 nm Pt particles. The Pt NP-rGO hybrid was sprayed onto FTO as counter electrode which outperformed platinized FTO electrodes obtained from low-temperature sputtering [164]. Graphene/Pt monolayers can be fabricated by layer-by-layer self-assembly techniques and used as counter electrodes in DSSC. By this, the amount of Pt can be reduced by a factor of 1000 but the as-received DSSC exhibits a lower PCE of 7.66% compared to 8.16% of a Pt based DSSC [165]. Impressively, graphene electrodes even show a better performance than platinized FTO electrodes in the catalytic activity towards thiol-based electrolytes. DSSCs fabricated using a Co(bpy)3(II/III) redox couple and graphene nano-platelet (GNP) cathodes with 66% optical transparency revealed PCEs over 9% which is superior to Pt-based DSSCs [166]. This can be attributed to the smaller charge transfer resistance (RCT) and better electrocatalytic activity of GNP-FTO cathodes towards the Co(bpy)3(II/III) redox reaction. GNP also showed an advanced performance, especially the open-circuit voltage was exceeding 1 V, in DSSCs with Co(L)2; where L is 6-(1H-pyrazol-1-yl)-2,2’-bipyridine [167]. In conclusion, most of the works on the use of graphene for catalytic electrodes have shown that graphene’s catalytic activity is lower than that of platinized FTO, some results lead to comparable activities and only few to higher activities. The usage of Pt-graphene composites for catalytic electrodes reduces the amount of Pt significantly and therefore the cost for such electrodes. In the case of DSSC using Co(bpy)3(II/III) redox electrolyte, a graphene nanoplatelets counter electrode showed a better performance. These results in combination with lower cost compared to Pt electrodes demonstrate that graphene is a very promising material for replacing or reducing the expensive Pt in catalytic counter electrodes for DSSCs. Table 1.3 summarizes the results of applications of graphene in catalytic counter electrodes for DSSCs.

28


Table 1.3 Graphene for catalytic counter electrodes and the PCE of corresponding DSSCs Graphene type Functionalized rGO Graphene

Device structure

PCE (%)

Ref.

FTO/TiO2/dye/I3−/I1− mediated electrolyte/rGO

4.99

[157]

FTO/TiO2/dye/I3−/I1− mediated electrolyte/rGO

5.73

[168]

FTO/TiO2/dye/I3−/I1− mediated electrolyte/GNS

6.93

[160]

7.19

[162]

7.5

[161]

9.3

[167]

nano-platelets Graphene nanosheets Gel graphene

Vertical GNS

Graphene platelets

FTO/TiO2/dye/I3−/I1− mediated electrolyte/Gel graphene FTO/TiO2/dye/I3−/I1− mediated electrolyte/ GNSactivated carbon FTO/TiO2/dye/Co(III)/(II) mediated electrolyte/ graphene

1.3.6 Photodetectors Photodetectors (PDs) are devices used to measure light power by converting the absorbed photon energy into an electrical current. Their common applications are in digital cameras, televisions, remote controls, DVD players and so on. Graphene-based PDs are working mostly based on the photoelectric effect in which incident photons excite electrons to form excitons which then are separated into individual charge carriers and propelled by the external bias forming a photo-induced current, also called photocurrent. Other working principles, for example, the photo-thermoelectric effect and the photo-bolometric effect are also exploited for PDs, but are less common than the photoelectric effect. The former is based on the thermoelectric effect induced by light illumination, while the latter expresses the dependence of the electric conductivity on temperature. The spectral bandwidth of detection is typically limited by the semiconductor material’s bandgap that hinders IV or III/V semiconductors based PDs from long-wavelength detection. In contrary, graphene absorbs from the ultraviolet to terahertz range [31] leading to a much broader wavelength working

29


range. In addition, due to their high carrier mobility, graphene based PDs (GPDs) can be ultrafast. Early works on GPDs focused on to exploit graphene-metal junctions or graphene p-n junctions for measuring the photocurrent [169–172]. Because of the small area of the effective junction region contributing to the photocurrent, as well as weak optical absorption, the responsivity is therefore limited to a few mA W-1. To improve the interaction with light, graphene was utilized together with plasmonic nanostructures [173–175] or microcavities [176,177]. As a result, the responsivity was improved to tens of mA W-1. By using silicon waveguide-integrated graphene, the absorption of evanescent light that propagates parallel to the graphene sheet, is enabled and a responsivity of 0.13 A W-1 was achieved to wavelengths ranging from near to mid-infrared [178]. Defect midgap states were introduced in the band gap by engineering a monolayer of CVD graphene into quantum dot-like array structures. The midgap states can serve as electron trapping centers. This improved the photoreponsivity up to 8.61 A W-1 with high gain of 120 to a broad range from the visible (532 nm) to the midinfrared (10 μm) in a single pure graphene PD [179]. However, efficient photodetection was only achieved below 150 K due to the short electron lifetime of the midap states at higher temperatures. Another approach is to hybridize graphene with efficient light-absorber materials such as QDs. The hybrid material will take advantages out of the high carrier mobility from graphene and light-sensibility from QDs. Thus, an ultrahigh photoconductive gain of 108 electrons per photon and a great responsivity of 107 were obtained in a hybrid PD that consists of monolayer or bilayer graphene covered with a thin film of QDs [180]. Despite the excellent responsivity, the light absorption relies on the QDs instead of the graphene, thus restricting the spectral range of the photodetection. In a recent report [181], a PD was constructed by a pair of stacked graphene monolayers, (the top layer serves as gate and the bottom layer serves as channel) separated by a thin tunnel barrier. Photo-induced excitons generated in the top layer tunnel into the bottom layer, leading to a charge build-up on the gate and a strong photogating effect on the channel conductance. As a result, a responsivity greater than 1 A W -1 to light of the mid-infrared range was achieved even at room-temperature. 1.3.7 Fuel cells A fuel cell (FC) is an energy conversion device that generates energy by oxidizing a fuel catalyzed by the catalysts immobilized on electrodes. Most common fuels used in FCs 30


include hydrogen, methanol and ethanol which are converted to water and carbon dioxide while generating energy. Pt and its alloys are usually used as the most efficient catalysts and they determine the performance of FCs. However, the high cost of Pt restricts its practical application and there is therefore a strong need for reducing the amount of Pt or even replacing Pt with other materials. To this end, graphene with its high specific surface area and high electrical conductivity is an ideal supporting material for Pt catalysts. Furthermore, nitrogen or sulfur-doped graphene demonstrates high catalytic activities towards the oxygen reduction reaction (ORR), being a promising candidate to replace expensive Pt [182–184]. In addition to high cost, there are several challenges for the utilization of Pt-based catalysts, such as low durability and easy poisoning from methanol crossover. To address these problems, Pt-supported graphene materials have been extensively studied for ORR. Pt NPs (approx. 2.9 nm in diameter) supported rGO platelets exhibited much larger electrochemically active surface area and greater catalytic activity toward ORR compared to commercial Pt/C (75 wt% Pt) [185]. Besides Pt, Co or Ni alloys were also embedded on graphene forming catalytic hybrid materials for ORR. Both Co- and Ni-based hybrid materials showed a higher ORR activity than pure Pt catalysts, lower material costs, as well as a good durability in addition [186,187]. Non-precious metal-based catalysts such as Fesupported on graphene or doped graphene have been also proven as promising catalysts for ORR [188]. In another approach, Pd-supported by graphene nanosheets (Pd-GNS) was also used for ORR catalysis. The new material demonstrated higher mass and specific catalytic activities towards ORR than the Pt-GNS catalyst. Moreover, the two materials both enable efficient four-electron reduction processes [189]. The introduction of other elements such as N, B, S or P will break the symmetry of graphene lattices inducing defects which can be active sites for ORR catalysis [190]. N-doped graphene demonstrated high ORR activity through a four electron transfer process with durabilities comparable or even better than the ones with commercial Pt/C as catalyst [190]. It is proven that graphene doped with elements, which have similar electronegativity to carbon such as sulfur and selenium, can also exhibit better catalytic activities than commercial Pt/C catalysts in alkaline media. This might pave the way to fabricate new low-cost precious-metal free catalysts with high electrocatalytic activity for practical FCs [184].

31

1.4 Motivation

1.3.8 Other applications Graphene offers the best surface to volume ratio of any known material. Every carbon atom in graphene is a viable target for reactive species. The interactions between graphene and detected species can vary from weak van der Waals forces to strong covalent chemical bonding that will perturb the pristine nature of the graphene structural and electronic system. This therefore forms the basis for the detection of such interaction events, making the detection of a single event possible. The mechanism of graphene-sensors can be based on chemical sensing (e.g. gas sensors), electrochemical sensing, photoelectric sensing, biological sensing, mass sensing and strain sensing. In addition, graphene is also used in applications for non-volatile memory, in composite materials for enhancing the mechanical performance, in paints and coatings, and as well as bio-applications such as tissue-engineering or regenerative medicine [191]. However, these subjects are just mentioned and will not be further discussed in detail within this thesis.

1.4 Motivation Nowadays, we are seeing the daily change of technologies in every corner of life from electronic devices, transportation vehicles, to medicine technologies and so on. This leads to a high demand for material resources and energy supply. Moreover, the demand for high compact devices, speed processing, high density memories and high working efficiency, and so on motivates the development to integrate nanomaterials or nanotechnology into existing technologies. The basic devices become smaller and smaller and are approaching the limits of current techniques (e.g. optical lithography) and conventional materials such as silicon. Therefore, the discovery of new materials and their applications in energy conversion to complement or replace existing ones is one of the great urgencies to be addressed by the research community. With the extraordinary properties (see 1.1), graphene offers huge possibilities for various applications with the potential to revolutionize various technologies, and the age of carbon materials, which was already proposed by some scientists, might partially become a reality. The chemical approach for the synthesis of graphene (see chapter 1.2) is the method of choice for massive production of GO and rGO at low cost. GO and rGO also possess some features of merit that distinguish them from graphene produced by other methods. For example, the solubility in various polar and non-polar solvents depending on the reduction degree of rGO enables solution-processability. The availability of oxygen-containing functional groups, 32

1.4 Motivation

which can be transformed to different desired chemical functionalities for attaching different NPs or species, can be utilized to obtain nano-hybrid materials with unique physical and chemical properties. These hybrid materials often have synergetic properties out of excellent features of graphene and desirable properties of the functional NPs. As a result, great potential applications in photovoltaics or for other optoelectronic devices can be explored based on these materials. This thesis is focusing on the synthesis, functionalization, characterization and application of graphene based on the chemical solution approach with following three aspects: i)

Synthesis of GO using chemical exfoliation, and reduction of GO to rGO.

ii)

Functionalization of rGO with desired species for attaching metal NPs such as CdSe QDs to achieve functional graphene hybrid materials.

iii)

Applications of graphene and functional graphene in photovoltaics and optoelectronics.

33

2.1 Motivation of GO and rGO synthesis

Chapter 2 Graphene oxide (GO) and reduced GO (rGO) (Some parts of this section were published in Chem. Eng. J. 231, 146 (2013) entitled “Thiol functionalized reduced graphene oxide as a base material for novel graphene-nanoparticle hybrid composites”)

2.1 Motivation of GO and rGO synthesis So far, several methods have been utilized to synthesize graphene. Pristine graphene was firstly obtained by using Scotch tapes to peel off a single or few layers of graphene sheets from bulk graphite which was reported by Geim, Novoselov and co-workers [1]. This method delivers the highest quality graphene for fundamental studies in physics, but does not yet appear to be scalable to large area. Alternatively, CVD is the method of choice to produce thin and continuous large-area graphene films [192]. However, the CVD method is still facing challenges such as high cost and sophistication to realize the graphene transfer step after the synthesis. Although the solution approach to graphene through the reduction of GO cannot yet produce high quality graphene and large graphene sheets, the chemical reduction of GO is the method of choice for obtaining graphene-like structures in high quantities and at low cost [193]. Moreover, GO is dissolvable in aqueous solution and rGO is soluble in various solvents such as dimethyl formamide (DMF), chlorobenzene (CB). This is a crucial advantage for further processing the material for applications. Graphenes obtained through chemical reduction of GO inevitably suffer from many defects in their lattices resulting in a significant reduction in the charge-carrier mobility due to interruption of the

electronic

system [193]. However, the lattice defects and functional groups within GO are useful for some applications such as their use as catalyst support or electrode in dye–sensitized solar cells [194], as well as for anchoring NPs to form hybrid materials [195,196]. The hydroxyl and carboxylic groups of GO can be transformed into different functional groups for favorable attraction of specific metal NPs. For example, thiol-groups were formed by thionation of GO and used for anchoring CdSe QDs to obtain a hybrid material [197]. Therefore, the solution approach to synthesize functional graphenee through GO is a significant and promising method for many applications.

34

2.2 Materials and synthesis of GO

2.2 Materials and synthesis of GO 2.2.1 Chemicals Graphite was purchased from Merck, NaNO3 (95%) and KMnO4 (99%) were obtained from Grussing GmbH (Germany). H2SO4 (95-98%), HCl (37%) and H2O2 (30%) were obtained from Sigma-Aldrich. 2.2.2 Synthesis of GO GO was synthesized using a modified Hummers method [198]. In a typical synthesis approach, 1 g graphite, 46 ml H2SO4 and 0.5 g NaNO3 were mixed together and stirred at 35 C for 2 minutes. Then, the solution was continuously stirred in an ice bath until the temperature reached 0 C, which usually took 15 min. After that, 3 g KMnO4 was added gradually, so that the temperature was not allowed to exceed 20 C. Subsequently, the solution was held and stirred at 35 C for 6 h. In an additional step, another 3 g KMnO4 was added to the solution, and stirred again for 12 h at 35 C. Finally, 150 ml H2O containing 6 ml of 30% H2O2 was added slowly while keeping the temperature below 80 C. As a result, residual KMnO4 and MnO2 were reduced to Mn2+ dissolving in the solution. The color of the solution changed from dark brown to bright yellow. Post-synthetic cleaning procedure for GO: Firstly, the obtained bright yellow solution was sonicated for 30 min and centrifuged at 1000 rpm for 20 min to remove bigger particles and aggregates which have not been fully oxidized. Then, the remaining solution was centrifuged at 4400 rpm for 3 h to collect the GO product from the solution. Subsequently, 80 ml of 37% HCl was added to the GO powder which was dispersed in HCl by stirring for 2 min. By an additional centrifugation step at 4400 rpm for 3 h, remaining metal-ions were removed. Then, the GO precipitate was washed two times in 150 ml of DI-distilled H2O adding 50 ml absolute C2H5OH and centrifuged at 4400 rpm for 2.5 h to remove acid and Mn+2 in the sample. Hence, purified GO was obtained. Finally, the GO was dispersed in DI-distilled water followed by sonication for 1 h to obtain a homogeneous aqueous dispersion of GO with a pH ranging between 6 and 7. Fabrication of GO freestanding thin films: 40 ml of 0.05 wt% GO aqueous solution was taken into a 100-ml glass beaker, and then the solution was naturally evaporated at room temperature in ambient condition that usually lasts 24 h. After water evaporated out, GO flakes were deposited on the glass beaker bottom forming a GO thin film. The film thickness 35

2.3 Characterization of GO, results and discussions

was controlled by the amount and concentration of the GO solution and was measured by a profilometer (Bruker Dektak 150). From a certain thickness on (ca. > 5 μm), the film was stable enough and could be peeled off as free standing film.

2.3 Characterization of GO, results and discussions 2.3.1 Optical microscope (OM) and scanning electron microscopy (SEM) Samples for OM experiments were prepared by spin-coating 50 μl of 0.05 wt% GO aqueous solution onto silicon substrates at two successive speeds of 500 rpm for 1 min and 2500 rpm for 2 min. Then, the samples were imaged by an OM in the reflection mode with a magnification of 50×, and recorded by an integrated digital camera. Samples for SEM measurements were prepared according to the same procedure as OM samples, but on ITO/SiO2 substrates instead. SEM imaging of GO flakes on ITO was performed by using a Quanta 250 FEG (FEI, USA) usually at 2.00 kV accelerating voltage.

Figure 2.1 (a) OM image of GO flakes deposited onto a Si/SiO2 substrate, (b) SEM image of GO flakes distributed onto an ITO/SiO2 substrate [74]. The contrast difference between GO flakes and the Si surface makes GO visible in OM (Fig. 2.1 a). Although the images are not clear and one cannot distinguish few-layer from singlelayer GO flakes, this ability to observe graphene flakes under an OM enables the use of optical lithography techniques for graphene structuring, which is the subject of Chapter 6. During OM investigation of graphene, we found an interesting phenomenon that GO flakes are visible in OM as shown in Figure 2.1a, but after being reduced, the contrast changed and they became not distinguishable any more from the surface of the Si/SiO2 substrate. It is supposed that the presence of oxygen-containing functional groups in graphene lattices increases the optical reflectivity of graphene sheets by inducing surface plasmonic effects. 36


However, this hypothesis needs more investigation to be confirmed and is not further addressed in this thesis. SEM investigation reveals that individual exfoliated GO platelets can be deposited onto the ITO substrate out of an aqueous solution (Fig. 2.1b). No aggregated GO or bigger particles of graphite are observed in the SEM image. This means that almost all graphite has successfully been exfoliated into single- or few-layer graphene oxide flakes. From the image, one can easily observe the single-layer graphene oxide flake with good transparency, while the few-layer graphene flakes and overlapping area of single-layer graphene flakes are appearing darker. The size distribution of GO flakes varies from several hundreds of nanometers to less than 10 µm. 2.3.2 Atomic force microscopy (AFM) For the preparation of samples for AFM measurements, a 1 × 1-cm2 ITO/glass substrate was treated with O2 plasma for 2 min to form a hydrophilic surface. Then, 25 µl of a 0.05 wt% aqueous solution of exfoliated GO has been spin-coated onto the substrate at 2000 rpm during 30 s, followed by an additional spin coating step (3000 rpm for 1 min) to dry the sample.

Figure 2.2 AFM image (left) and height profiles (right) of single GO sheets extracted from position 1 and 2. An average thickness of about 0.7 nm was detected [74].

37


The AFM image of GO was recorded in the tapping mode using a Multimode AFM (Fa. Veeco) at 1.00 Hz scan rate, and was analyzed by a Nanoscope analysis program. Figure 2.2 is an AFM image of GO deposited onto ITO/glass. It reveals that single-layer GO sheets are successfully exfoliated by using the previously described modified Hummers’s method. The AFM profiles show that single-layer GO has thickness of about 0.7 nm. The thickness is a little thinner than mentioned in previous reports [70,199,200] where singlelayer thicknesses of around 1 nm have been reported for GO. It should be considered that the GO thickness in AFM measurements is established by not only the van der Waals thickness of pristine graphene, which is 0.34 nm, but also by functional groups above and below the graphene plane, a thin layer of water adsorbed on the graphene surface and the interaction between GO and the substrate. Thus, the thinner thickness is attributed to the stronger attachment of GO to ITO substrate used in this AFM measurement compared with silicon substrates in other reports. 2.3.3 X-ray photoelectron spectroscopy (XPS) A Specs X-Ray Source XR50 with an Al-Kα anode (1486.6 eV) and source (12 kV, 20 mA) was used to excite the electronic states of atoms below the surface of the sample. Electrons ejected from the surface were energy-filtered and detected via the Specs Hemispherical Energy Analyzer Phoibos 50 (SPECS Surface Nano Analysis GmbH). Free standing GO

Figure 2.3 XPS spectrum of a GO film: (a) the whole spectrum with a carbon peak at 289 eV and oxygen peak at 535 eV. Inset image: camera picture of a free standing GO thin film. (b) A high resolution XPS spectrum showing a carbon 1s compound peak at 289 eV for carbons in CO(O) and C=O bonds and 286.5 eV for CO bonds and a small peak at 285 eV for sp2 hybridized carbons in CC bonds [74]. 38


films (Fig. 2.3a inset), which were subjected for XPS measurements, were prepared by evaporation of GO aqueous solution in a glass beaker at room temperature which has been described in detail under 2.2.2. Figure 2.3a shows a XPS spectrum with two sharp peaks at 289 eV and 535 eV for carbon and oxygen, respectively. This demonstrates that a considerable amount of oxygen was introduced into GO by creating oxygen-containing functionalities such as carboxylic, carbonyl, hydroxyl and epoxy groups. The C1s XPS spectrum of GO (Fig. 2.3b) has two major peaks at 289 eV corresponding to carbon in C=O, C(O)O, and 286.5 eV in CO and a very small peak at 284.6 eV from a CC sp2 carbon. This means that most of the carbon atoms from graphite were oxidized during the transformation into GO. 2.3.4 UV-Vis absorption spectroscopy The UV-Vis absorption spectrum was obtained by using a TIDAS 100 spectrophotometer (J&M, Germany). A rectangular standard UV quartz cuvette from Starna (170-2700 nm spectral range) with a volume of 700 μl was used for the measurements. The spectrum has been recorded by measuring a 0.02 wt% solution of GO dissolved in DI-distilled water.

Figure 2.4 UV-Vis absorption spectrum of GO with a typical absorption maximum at 231 nm. The inset image represents a photograph of aqueous solution of GO.

39

2.4 Reduction of GO to rGO

In figure 2.4, an UV-Vis spectrum of a GO aqueous solution shows a characteristic absorption peak at 231 nm which is consistent with previous reports [199–201]. GO only absorbs in the violet and UV range of light, and has a maximum absorption at 231 nm. This means that GO is almost transparent to the visible light. However, when GO flakes are deposited to form a thin film, they aggregate and stack together, and the film becomes semi or non-transparent depending on the film thickness. The inset image reveals that GO is very well dissolved in water resulting in a clear homogeneous solution and can be colloidally stable for a long period of time stored at room temperature under ambient atmosphere.

2.4 Reduction of GO to rGO In this thesis, the reduction of GO to rGO has been performed by four different methods: thermal reduction, chemical reduction in hydroiodic acid vapor (HI), in hydrazine vapor (N2H4) and with phosphorus pentasulfide (P4S10). 2.4.1 Synthesis of rGO a) Reduction of GO by hydroiodic acid (HI) GO was reduced to rGO in vapor of a HI aqueous solution (57 wt%, Sigma) by a procedure according to a previous report [193]. A freestanding GO thin film (fabricated as described in 2.2.2) was placed inside a 300-ml flask containing 2.0 ml of HI and 5.0 ml of acetic acid (96 wt%, Merck). The cover of the jar was sealed with vacuum grease and placed over an oil bath at 40°C for 24 h for GO reduction. The color of the rGO thin film changed from matte brown to metallic grey after treatment with HI vapor inside the flask, indicating the reduction of GO. Subsequently, the received rGO thin film was rinsed in succession with a saturated sodium bicarbonate solution, water and methanol and dried at room temperature. b) Reduction in N2H4 vapor GO thin film deposited on the glass substrate was fabricated by drop-casting 70 μl of 0.05 wt% GO aqueous solution onto a 2 × 2-cm2 transparent glass substrate, followed by evaporation of water in ambient atmosphere at room temperature. With a GO thin layer on top, the transparent glass substrate becomes semitransparent (Fig. 2.7). The reduction of the GO thin film in N2H4 vapor was performed using a method which is based on a previously reported method with some modification [202]. 20 ml of N2H4 monohydrate (98%, Sigma) was put into a round-bottom flask (4). Next, the GO thin film 40


covered glass substrate was placed into a different flask (2). The flask (2) and flask (4) were integrated in a distillation set-up (Fig. 2.5).

Figure 2.5 Schematic illustration of the setup for reduction of GO to rGO in N2H4 vapor: (1) magnetic-stirring hot plate with an oil bath, (2) glass substrate with a thin GO film deposited on top in a round-bottom flask, (3) thermometer, (4) N2H4 liquid (98 wt%), (5) lock. Firstly, the system was evacuated to 5 × 10-3 mbar by a vacuum pump for 10 min. The lock (5) was closed after the pump stopped. In vacuum, the N2H4 evaporated and filled up the system, so that flask (2) was filled with N2H4 vapor. The flask (2) with the GO sample was then heated up to 90 C and held at this temperature for 24h for performing the chemical reduction step. After the reduction reaction, the rGO sample was washed with DI-distilled water, and dried for 30 min at 90 C in an air-floated oven. c) Thermal reduction In this method, the GO thin film covered glass substrate was placed in a 100 ml flask. The flask was then repeatedly evacuated to 5 × 10-3 mbar and refilled with nitrogen gas for 3 times. Consequently, the nitrogen gas supply was closed, and the flask was held at 5 × 10-3 mbar under continuous vacuum pumping. The temperature of the wood’s metal alloy was increased to 450 C. After the alloy was melting, the flask containing GO sample was dipped in the liquid alloy for the thermal reduction. The pressure and temperature were kept for 2 h during the reduction. d) Reduction by phosphorus pentasulfilde Materials: HPLC reagent grade dimethylformaminde (DMF), phosphorus pentasulfide (P4S10) (99%) and 0.45 µm polyamide filter membranes were purchased from Scharlau 41


Chemie S.A, Sigma Aldrich and Whatman Int.Ltd, respectively. A glass vacuum filtration system for 50 mm diameter filter was purchased from Sartorius Stedim Biotech (Germany). Synthesis: 100 mg GO was dispersed in 100 ml DMF and sonicated for 1 h. After removing undispersed GO by centrifugation at 1000 rpm for 10 min, a homogeneous solution of GO in DMF was obtained and put into a round-bottom flask. Then, 300 mg P4S10 was added to the solution and the reaction flask was evacuated to 5 × 10-3 mbar at 100 C for 2 min to remove traces of water in the flask. The thionation was performed for 12 h under vacuum and continuous stirring at 120 C. Finally, the thiol-functionalized rGO (TrGO) product was collected by filtering the reaction solution through a 0.45-µm polyamide membrane and was extensively washed first with 100 ml of water, followed by 100 ml ethanol and acetone, respectively using a glass vacuum filtration system. If the reaction was performed at the boiling point of DMF (152154 C) under reflux for 36 h, TrGO was obtained with a higher degree of reduction, but less thiol groups incorporated. The reduction degree of TrGO can be determined by measuring the UV-Vis absorption as well as the sheet resistance. After the filtration, a purified TrGO thin film was obtained on the filter membrane and could be peeled off as a free-standing thin film. 2.4.2 Resistivity measurements The sheet resistance measurements of rGO and TrGO films (usually ca. 1.0 cm in dimensions) were performed using the van der Pauw method [203]. First, the thickness of the tested rGO and TrGO films were measured by a profilometer (Bruker Dektak 150). Then, the films were fixed on a non conductive substrate by four small droplets of silver paste (Acheson Silver DAG 1415), which also acted as four contact pads with a diameter of about 2 mm for the following sheet resistance measurement. The four contact pads were contacted by four probeheads (Süss MicroTech PH100) and connected to a source-meter (Keithley 2602). Afterwards, the samples were measured to obtain the resistivities based on van der Pauw method. 2.4.3 Results and discussions There are several methods to determine the reduction degree of rGO such as XPS, FTIR which can measure the content of oxygen remaining in rGO. However, this section is focusing on evaluating the conductivity of rGO and TrGO, which is the most important feature of graphene for almost all its applications, especially in optoelectronics. The higher 42


reduction degree of rGO normally results in a better conductivity due to the restoration of πconjugated bonds in the carbon-hexagonal lattice.

Figure 2.6 Camera images of GO and rGO free-standing thin films, corresponding to before (left) and after (right) reduction with HI vapor. As shown in Figure 2.6, the GO thin film (left image) exhibits a polish and smooth surface. In contrast, the rGO thin film (right image) has a rough surface as a result of the reduction process. GO thin film is electrically insulating. After reduction by HI, the as-received rGO thin film has a conductivity of 3000 S/m as measured by the van de Pauw method as described in 2.4.2.

Figure 2.7 Semitransparent thin film of GO deposited on a glass substrate (a), after treatment in N2N4 vapor it was reduced to rGO thin film and turned to a dark brown color (b). In order to make rGO into a transparent conducting thin film, a thin film of GO has been deposited on a transparent glass substrate (Fig. 2.7 a), and reduced to an rGO thin film (Fig 2.7b) by treatment with N2N4 vapor. The reduction step increased the film conductivity from 0 (GO) to 1030 S/m (rGO), but lowered its transparency and therefore turned it to dark 43


brown. The rGO thin film deposited on glass substrate has also been fabricated by thermal reduction. However, the obtained conductivity was lower than that obtained by N2N4 vapor reduction. Typically, a commercial ITO/SiO2 transparent substrate has a conductivity of about 10+6 S/m at a 90% transparency. Therefore, the fabrication of rGO thin films for transparent conductor applications is still remaining as a big challenge. Doping of rGO with other species such as metal ions might be a solution for improving the conductivity of rGO thin films in general.

Figure 2.8 (a) Free-standing TrGO paper, (b) TrGO dispersion in DMF (black solution) synthesized from GO dispersion in DMF (yellow brown solution), (c) A 1×1 cm piece of TrGO thin film with four small droplets of silver paste as contact pads. GO in DMF is forming a homogeneous solution with yellow brown color, indicating a good solubility of GO in DMF, and it turns to a black dispersion in DMF after reduction with P4S10 to form TrGO (Fig.28b). Normally, reduction of GO is leading to an aggregation because of the stacking behavior of rGO flakes due to van de Waals forces. However, thiol groups within TrGO prevent graphene flakes from stacking together, enabling TrGO to be well dissolvable in DMF as well (Fig. 2.8b). As a result, a homogeneous free-standing TrGO paper can be fabricated by filtering the TrGO solution. The electrical conductivity of freestanding TrGO paper with a thickness of 1.3 μm is 5300 S/m measured by the van der Pauw 44

2.5 Conclusions

method. This conductivity is comparable with the best rGO reported in literature [193] synthesized by strong reductants and by high temperature annealing. The conductivity of TrGO paper is also significant higher than that of 3000 S/m for rGO synthesized by HI vapor described above, indicating the superiority of the P4S10 reduction method.

2.5 Conclusions Significant amount of oxygen-containing groups have been introduced into graphene lattices to form graphene oxide (GO). These functional groups contain negative charges that induce repulsive forces and enlarge the distance between graphene layers, enabling the exfoliation of GO single sheets by sonication. Therefore, single- and few-layer graphene oxide flakes have been successful produced by chemical exfoliation of graphite, followed by a sonication step. Single-layer GO sheets show a typical thickness of 0.7 nm as measured by AFM, and a lateral size ranging from several hundreds of nanometers to less than 10 µm proven by SEM. It is possible to observe individual GO flakes by optical microscopy which renders the possibility to structuring graphene flakes for device applications by optical lithography techniques which will be described in Chapter 5. GO is well dissolvable in water forming a stable homogeneous aqueous solution that is an advantage for further modification or application processing. GO has been reduced to rGO by several methods described above. Among those, TrGO thin films exhibit the highest conductivity of 5300 S/m for a TrGO thin film with a thickness of 1.3 μm, which is comparable with the best rGO thin film conductivity values from literature.

45

3.1 CdSe QDs – graphene hybrid material (CdSe QD-TrGO)

Chapter 3 Nanoparticle-functionalized graphene hybrid materials (Main parts of this section were published in Chem. Eng. J. 231, 146 (2013) entitled “Thiol functionalized reduced graphene oxide as a base material for novel graphene-nanoparticle hybrid composites”)

3.1 CdSe QDs – graphene hybrid material (CdSe QD-TrGO) 3.1.1 Background and motivation Already various applications utilizing graphene based materials have been reported such as transparent electrodes in solar cells [3,204,205], catalytic electrodes in fuel cells [206], SCs [132138] as well as electrodes in transistors [110,111,113,88] and sensors [9]. Due to its low catalytic activity, graphene is often decorated with catalytic NPs or QDs for applications in fuel cells [206] or optoelectronics [207], respectively, resulting in hybrid materials with both advantages of graphene and NPs often inducing novel physical or chemical properties. For example, Pt or titanium dioxide (TiO2) NPs have high specific surface areas and catalytic activities; they have been proven for applications in catalysis [208,209] and for energy conversion [210,211] [184,185]. Also, semiconducting CdSe QDs with easily tunable optical and electrical properties have been demonstrated successfully for application in hybrid photovoltaic devices [212,213]. However, the NPs are often synthesized in solution with ligand shells to prevent aggregation; this in turn decreases the effectivity of catalysis or charge transfer processes in fuel cells and solar cells [214]. Moreover, before NPs are incorporated into devices, the synthesis ligands have to be exchanged ideally by a monolayer of more conductive ligands with the disadvantage that the NPs might agglomerate [215,216] leading to a significant reduction of their active surface area. This general problem can be overcome by using a framework to support the NPs, keeping them separated and avoiding aggregation. Graphene is an excellent framework and support for NPs [217,218]. Additionally, it favors charge transfer processes at the NP-graphene interface [207] as well as charge transport processes making them ideal candidates for interlayer or electrode material in electrocatalysis or photovoltaics [195,219]. Taking advantages of the high electrical conductivity (electron mobility of 15,000 cm2 V-1s-1), the ultra-high specific area of graphene (theoretical value, 2630 m2/g) [220] and good catalytic or optoelectric properties of NPs, the obtained NP-graphene hybrid materials have high potential in energy-harvesting applications. 46

3.1.1 Background and motivation

So far, much effort has been paid to the synthesis and application of metal-NP graphene hybrid materials. Pt or Ag NP decorated graphene was successfully synthesized and exhibits highly catalytic behavior towards methanol and ethanol oxidation and oxygen reduction as electrode material in fuel cells [217,218]. They have also been used as catalytic counter electrode for reduction of triiodide in DSSCs [195,219]. Another example is graphene decorated with CdSe QDs with promising properties for applications in photovoltaics and LEDs [207,221–223]. Usually, graphene-based hybrid materials are synthesized by in-situ growth of NPs on GO sheets. However, graphene might affect the formation of NPs by hindering or influencing their growth, resulting in non-uniform NP shapes. Moreover, during the NP formation, GO is simultaneously reduced to rGO and tends to agglomerate. Therefore, it is difficult to control the NP quality (e.g. size distribution and uniformity) and at the same time the degree of GO reduction to rGO which determines important properties such as the electrical conductivity, the work function and the solubility of the resulting hybrid material. Also, rGO is often insoluble in either aqueous media, used for the synthesis of metal NPs such as e.g. Ag [224] or Pt [195,218] or organic solvents and matrices such as hexadecylamine (HDA) and tri-n-octylphosphine oxide (TOPO) often used for the synthesis of semiconductor QDs [207,223]. A major drawback of these in-situ approaches is that during the reduction of GO to rGO, graphene sheets are often irreversibly stacking together by van der Waals forces, which therefore hinders NPs to reach the graphene surface leading to a less effective NP loading. Another approach to fabricate NP-graphene hybrid materials is the functionalization of graphene with specific functional chemical groups which have a high affinity towards NPs. CdSe NP decorated poly-(diallyldimethylammonium chloride) (PDDA)-functionalized graphene was synthesized by Lu et al. [225] using positively charged PDDA to attach negatively charged CdSe NPs. The as-received hybrid material exhibits good solubility in polar solvents such as alcohol and water. Feng et al. [226] achieved cadmium sulfide (CdS) NP decorated graphene by using benzyl mercaptan as linker binding to CdS NPs via thiol group (SH) and attaching to graphene via non-covalent - stacking. Such NP-graphene hybrid materials obtained by binding NPs to rGO through non-covalent linker molecules can overcome the above mentioned solubility problem of graphene. However, they might induce less effective charge transfer when applied in fuel cell or solar cell applications. Here, a simple synthesis method towards thiol-functionalized graphene directly from GO, and its effective decoration with transition metal NPs by self-assembly is described. The selfassembly decoration is based on the fact that thiol groups strongly bind to NPs of transition 47

3.1.2 Synthesis and characterizations of TrGO

metals such as e.g. Ag, Au or metal chalcogenide semiconductors such as e.g. CdSe or CdS QDs as reported by Mann et al. [227] and Colvin et al. [228]. At the time that this work was published, it was the first report on thiol-functionalization and simultaneous reduction of GO by direct thionation of GO with P4S10. This approach is a general scheme to obtain NP-graphene hybrid material with high NP loading. The respective hybrid materials have high potential to be utilized in optoelectronics, sensors and transistors.

3.1.2 Synthesis and characterizations of TrGO 3.1.2.1 Materials and synthesis This part has been described in Chapter 2 (see 2.4.1). 3.1.2.2 Characterization, results and discussions GO is directly thionated by transforming hydoxyl (OH) and carboxylic (COOH) functional groups of GO into thiol (–SH) and thiocarboxylic (COSH) or dithiocarbonxylic (–CSSH) groups, respectively (schematically shown in Fig. 3.1) which was already demonstrated for MWCNT by Čech et al. [229]. The chemical reaction between GO and P4S10 takes place under refluxing condition at the boiling point of DMF (152154 °C). We suppose that the thionation reaction produces hydrosulfide byproduct (H2S) which in turn reduces GO partially to rGO inducing a red-shift in the UV-Vis absorption maximum. The thiol groups are also allowing the individual graphene sheets to keep distance, preventing them from aggregation and making the resulting thionated rGO well dispersible in various solvents. GO normally contains carboxylic, hydroxyl and epoxy groups and is therefore easily dissolvable in polar solvents including water and DMF, the latter one serving as a reaction medium for GO and P4S10. Based on a thionation mechanism reported by Ozturk et al. [230] and demonstrated for carbon nanotubes by Čech et al. [229], I propose a thionation mechanism of GO by P4S10 which is shown in Figure 3.1. In solution, P4S10 can be dissociated into P2S5 subunits with a pronounced polar character of the P +–S- bond, which then alternatively binds to C+=O- or O-–H + groups of GO. Nucleophilic S- will attack C+ and electrophilic P+ will, in contrast, bind to O- forming a four-ring transition state, leading to a replacement of oxygen atoms from GO by sulfur atoms to form C=S and S–H groups in TrGO.

48


Figure 3.1 Schematic illustration of the proposed thionation mechanism of GO by P 4S10 according to Ozturk et al. [230] and Pham et al. [74]. During the thionation, GO is simultaneously reduced partially to rGO by hydrosulfide (H2S) at elevated temperature, a byproduct of this thionation reaction which can be detected by its characteristic smell. The resulting TrGO is soluble in DMF which prevents the aggregation of graphene sheets. Furthermore, the obtained TrGO is dispersible in various solvents such as CB or chloroform, depending on the degree of GO reduction. Controlling the solubility is very valuable for future applications, as well as for the formation of NP-TrGO hybrid materials. a) XPS investigation Sample preparation: 10 mg TrGO was dispersed in 50 ml DMF by sonication for 1 h forming a TrGO dispersion. The TrGO dispersion was then filtered under vacuum through a 0.45-µm polyamide membrane filter using the glass vacuum filtration system. An 8-μm thick TrGO film was formed and peeled off from the filter membrane resulting in a free-standing TrGO 49


film. The experimental procedure for the XPS investigation has already been described in section 2.3.3.

Figure 3.2 XPS spectra of TrGO and GO: (a). Entire spectra with a carbon peak at 289 eV and oxygen peak at 535 eV; (b). High resolution spectra showing a carbon 1s peak at 285 eV for sp2 carbon (C=C bonds), as well as additional compound peaks at 289 eV for carbons in CO(O) and C=O bonds and 286.5 eV for carbon in CO bonds ; (c). The XPS spectra of TrGO and GO around the sulfur 2p peak at 163.6 eV and the silicon peak at 149 eV; (d). XPS spectra of TrGO and GO around the sulfur 2s peak at 230 eV [74]. The presence of sulfur containing chemical groups in TrGO is confirmed by XPS and FTIR investigations. XPS spectra of pristine GO and of TrGO are shown in Figure 3.2a (b, c and d are high-resolution measurements). The black line shown in Fig. 3.2 c indicates a compound peak at a binding energy of 163.6 eV deriving from 3 peaks of sulfur 2p3/2 that is consistent with previous reports [231,232]. The intrinsic Si peak deriving from the Si-substrate at 151 eV [233] exhibits the accuracy of the measurements. Moreover, a sulfur 2s peak at 230 eV (Fig. 3.2d red line) with a high intensity is detected and confirms a significant amount of 50


sulfur present in TrGO. The respective sulfur peaks are easily observable in direct comparison with the spectra of GO (red lines, Fig. 3.2 c, d). Before performing XPS measurements, the TrGO samples have been extensively washed by DI-distilled water, ethanol, and acetone to ensure that no non-covalent bound sulfur containing residues from P4S10 and byproducts are present in the samples. The whole XPS spectra (Fig. 3.2a) and XPS C1s spectra (Fig. 3.2b) of GO and TrGO reveal a significant reduction degree of TrGO after the thionation process. While the spectrum of GO (Fig. 3.2b red line) has a peak intensity ratio IO1s/IC1s - between the oxygen peak (O1s) at 535 eV and carbon 1s (C1s) peak at 289 eV - of about 1.80, the IO1s/IC1s ratio of TrGO (Fig. 3.2a black line) is about 0.51 and therefore much smaller. This confirms that a considerable amount of oxygen is removed by the reduction from GO to TrGO. C1s XPS spectrum of TrGO (Fig. 3.2b black line) has a sharp peak at 285 eV assigned to a carbon peak which is composed out of a major C–C sp2 carbon peak (284.6 eV) and three minor peaks: C–O (286.5 eV); C=O (287.8 eV); C(O)O (289.1 eV) [193]. In contrast, the C1s of GO (Fig. 3.2b red line) has two major peaks at 289 eV of C=O, C(O)O, and 286.5 eV of C–O and a very small peak at 284.6 representing a C–C sp2 carbon. b) FTIR spectroscopy For preparing FTIR samples, GO and TrGO were grinded into a powder and mixed with KBr powder (KBr for IR spectroscopy, Merck) and grinded again. The ratio is 1 mg TrGO and GO, respectively per 100 mg KBr. The mixture was then pressed into pellets for FTIR measurements. The FTIR spectra of TrGO and GO were recorded by a Nicolet/Thermo Magna-IR 760 spectrometer with a probe chamber using a diamond ATR-Unit (detection range: 400 cm–1–4000 cm–1). The occurrence of characteristic 2p and 2s peaks of sulfur in XPS spectra proves the presence of sulfurs in TrGO. To demonstrate that the sulfurs in TrGO samples are covalently bonded to carbon, FTIR spectra were taken in addition. Figure 3.3 represents FTIR spectra of TrGO and GO, respectively for direct comparison. The spectrum of GO (Fig. 3.2. blue line) shows three strong peak signals at 1625 cm-1, 1733 cm-1 and 1047 cm-1 assigned to C=C sp2, C=O and COC bond stretching vibration, respectively and a weak peak signal of OH at 1222 cm-1 which are consistent with previous reports [201,234]. The FTIR spectrum of TrGO (Fig. 3.3 black line) reveals not only peaks at 1625 cm-1 and 1733 cm-1 representing C=C sp2 and C=O bond stretching vibrations, respectively, but also peaks related to thiol groups. A strong 51


band at 1095 cm-1 can be assigned to the presence of thiocarbonyl groups (C=S) that are formed by replacing the oxygen element (O) in carbonyl (C=O) or carboxylic group (HOC=O) by sulfur (S). The hydroxyl groups (OH) are exchanged into thiol groups (SH) proven by the presence of the sharp CS stretching peak at 671cm-1 and the CSH bending peak at 804 cm-1 as well as a weak SH stretching peak at 2576 cm-1 [235]. The combination between XPS and FTIR signals fully demonstrates the thionation of GO to TrGO while noncovalent attachment of sulfur containing compounds to graphene was excluded.

Figure 3.3 FTIR spectra of TrGO (black line) with vibration peaks of: CS stretch at 671 cm-1, C-SH bending at 804 cm-1, C=S stretch at 1095 cm-1 and C-H stretch at 2576 cm-1; and GO (blue line) with epoxy (COC) stretching peak at 1047 cm-1, hydroxyl (O-H) stretching peak at 1230cm-1, C=O stretching peak at 1733 cm-1, C=C sp2 stretching peak at 1625 cm-1 [74]. c) UV-Vis absorption spectroscopy For sample preparation, TrGO was dispersed in DMF by sonication for 1 hour to form a homogeneous solution of 0.02 wt% TrGO in DMF (Fig 3.4 camera image). Details on the experimental procedure and the equipment have been already described in Chapter 2.

52


Figure 3.4 UV-Vis absorption spectra of GO and TrGO. The GO spectrum has a typical peak maximum at 231 nm, and the peak maximum of TrGO is red-shifted to 272 nm due to the reduction of GO to TrGO and a partial restoration of the

conjugated network. The inset

image represents a photograph of GO in aqueous solution (yellow-brown solution) and TrGO in DMF (black-brown solution) [74]. Figure 3.4 reveals a significant red-shift of the UV-Vis absorption peak maximum from 231 nm to 271 nm for GO and TrGO, respectively and the camera image inset reveals a color change from brown yellow for the GO aqueous solution to black for the TrGO dispersion in DMF, which is a typical color for reduced GO. A higher reduction state leads to more redshifted signals in the UV-Vis absorption spectrum. The absorption red-shift indicates the restoration of sp2 bonds enhancing the

conjugation within the graphene lattice. By this, the

electrical conductivity of TrGO free standing paper is enhanced significantly from electrically insulating for GO to 5300 Sm-1 for TrGO measured by the van der Pauw method and can be controlled by adjusting the degree of the chemical reduction of TrGO. This conductivity is comparable with one of the best literature values of rGO free standing paper

53

3.1.3 Synthesis of CdSe QDs

which is 7850 S/m as reported by Moon et al. [193] where GO was reduced to rGO by HI solution as a strong reducing agent.

3.1.3 Synthesis of CdSe QDs 3.1.3.1 Materials and methods The CdSe QDs were synthesized according to Yuan et al [236]. First, 2898 mg (12 mmol) of HDA (hexadecylamine, ≥95%, Merck Schuchardt), 3092 mg (8 mmol) of TOPO (trioctylphosphine oxide, 99%, Sigma-Aldrich), and 444 mg (0.4 mmol) of Cd-stearate were heated up under nitrogen atmosphere inside a 25 ml three neck flask to 300°C. After reaching 300°C, 400 µl (0.4 mmol) of a 1M solution of Selenium in TOP (trioctylphosphine, 97%, ABCR) was quickly injected into the flask. The synthesis was continued at 300°C under stirring and stopped after 30 min. 3.1.3.2 Characterization For UV-Vis absorption measurements, CdSe QDs were dissolved in chloroform forming dispersions with concentrations of about 100 mg/ml. The absorption spectra were recorded with a TIDAS 100 spectrophotometer (J&M, Germany). These samples were also used for PL measurements, where their PL spectra were recorded by utilizing a J&M FL3095 spectrometer (Horiba Jobin-Yvon Fluorolog, Germany). The excitation wavelength was 350 nm and a slit width of 1 nm for excitation and 0.5 nm for emission was used. The detector resolution was ±1 nm. TEM measurements were performed using a Zeiss (LEO) 912 Omega transmission electron microscope with an acceleration voltage of 120 kV and zero loss filtering. For sample preparation, carbon coated copper grids were dip-coated in a solution of about 10 mg/ml CdSe QDs in chloroform. 3.1.3.3 Results and discussions Figure 3.5a shows a typical TEM image of CdSe QDs after 30 min reaction time under optimal synthesis conditions. The image reveals a uniform size distribution of spherically shaped QDs with an average diameter of about 6 nm which were used for the synthesis of CdSe QDs- TrGO hybrid nanocomposites and their integration into hybrid solar cells which will be described in the following sections. The as synthesized red-emitting CdSe QDs have a first absorption peak maximum at 611 nm with a corresponding PL emission peak maximum at 622 nm. The absorption spectrum reveals that CdSe QDs absorb in the visible and 54

3.1.4 Synthesis and characterizations of CdSe QD-TrGO hybrid material

UV spectrum, but not in the infrared region. Moreover, the existence of an absorption fine structure and the narrow full width at half maximum (FWHM) of 29 nm revealed that the CdSe QDs have a narrow size distribution which was additionally confirmed by TEM.

Figure 3.5 (a) TEM image of CdSe QDs (average diameter of about 6 nm) and (b) Absorption and PL spectra of CdSe QDs dissolved in chloroform after 30 min reaction time.

3.1.4 Synthesis and characterizations of CdSe QD-TrGO hybrid material 3.1.4.1 Materials and synthesis The CdSe QDs were decorated on TrGO sheets forming the CdSe QD-TrGO hybrid material by a self-assembly attachment as schematically illustrated in Fig. 3.6. The self-assembly decoration is based on the fact that thiol groups within TrGO have strong affinity to CdSe QDs. As-received TOPO/HDA capped CdSe QDs were purified using a previously published post synthetic ligand reduction procedure [237], where the QDs were treated with hexanoic acid to remove excess of HDA ligand molecules followed by a subsequent addition of methanol and a final centrifugation step. By this method almost all organic ligands were removed. Then, 1 mg TrGO was dispersed in 10 ml DMF by sonication for 30 min. After that, the TrGO solution was centrifuged by an Eppendorf MiniSpin at 2000 rpm for 2 min to remove larger TrGO-agglomerated particles which were not well dispersed. The resulting TrGO solution was centrifuged at 14500 rpm for 10 min, the supernatant was poured away, and the TrGO deposited at the bottom of the centrifugation tube was collected. Subsequently, 2 mg CdSe QDs and 1mg TrGO were both dispersed in 4 ml CB in a 10-ml glass vial by sonication for 15 min. A homogeneous solution of TrGO and CdSe QDs was obtained and was held at 55


110 C under continuous stirring for 30 min. As a result, CdSe QD decorated TrGO was obtained.

Figure 3.6 Schematic illustration for the formation of CdSe QD-TrGO hybrid material. 3.1.4.2 Characterization a) TEM analysis For TEM measurements, the samples were prepared by dip-coating of a structured holey carbon coated-copper grid (Quantifoil Micro Tools GmbH, Germany) into a solution of CdSe QD-TrGO in CB. The CdSe QD-TrGO hybrid material was fabricated by self-assembly decoration of TrGO with CdSe QDs as described in the experimental part. TrGO and CdSe QDs were dissolved and stirred in CB and CdSe QDs self-bind to thiol groups of TrGO leading to the formation of the respective hybrid material. Interestingly, it was observed that TrGO decorated with CdSe QDs became well soluble in CB, resulting in a clear homogeneous solution. This improved solubility is caused by CdSe QDs attached to the TrGO. To investigate the morphology of the CdSe-TrGO hybrid material, a typical TEM image was taken, as shown in Figure 3.7, clearly revealing the high loading of TrGO with spherical CdSe QDs of 6 nm in diameter. The CdSe QDs were synthesized as described in the method section by using a standard synthesis procedure [236]. They were already utilized successfully in photovoltaic applications after integrating them into hybrid solar cells [238]. Because the CdSe QD-TrGO 56


hybrid material is made by a mild self-assembly decoration process, the shape and size of CdSe QDs is not affected. As a result, a highly NP decorated CdSe-TrGO hybrid materials is obtained. Strong non-covalent bonds between CdSe QDs and thiol functional groups [227,228] of TrGO are formed.

Figure 3.7 (a) TEM image of TrGO decorated with CdSe QDs and (b) a zoom in TEM image of (a) [74]. In Fig. 3.7b, wrinkles from graphene sheets and few places without CdSe QDs can be observed; showing that graphene in the CdSe QD-TrGO hybrid consists out of a single layer or few layers and does not agglomerate during the reduction and decoration process into macroscopic assemblies. b) UV-vis absorption spectroscopy The samples were made by dissolving CdSe QD-TrGO in CB to form solution with a concentration of 0.02 wt%. Due to its good solubility, a homogeneous solution of the CdSeTrGO hybrid material in DMF has been prepared that gives a smooth UV-Vis absorption spectrum, as shown in Fig. 3.8. The spectrum reveals a sharp signal at 250 nm from TrGO and a typical peak from CdSe QDs at 611 nm which demonstrates the presence of CdSe QDs in the hybrid material after unattached CdSe QDs were removed after the synthesis as mentioned in the experimental section.

57


Figure 3.8 UV-Vis absorption spectrum of the CdSe QD-TrGO hybrid material with a TrGO peak maximum at 250 nm and CdSe QD peak maximum at 611 nm [74]. c) PL quenching experiments When QDs absorb photons with sufficient energy, electrons are excited from the valence band to the conduction band, and then relax after a short time back to the valence bands inducing photoluminescence. If QDs are in contact with other conducting materials with matching work function, a charge transfer can occur before electron-hole recombination can take place and the PL signal will be quenched as a result. Therefore, PL quenching experiments are performed to demonstrate a potential charge transfer occuring between QDs and TrGO as a result of either physical or chemical attachment of QDs to TrGO. PL quenching experiments were performed according to the following description: Six samples were prepared following the procedure of the CdSe QD-TrGO hybrid material synthesis as described above. They were made by combining 1.0 mg of CdSe QDs with 0.10, 0.12, 0.14, 0.16, 0.18 and 0.20 mg of TrGO respectively, and dispersed in 4 ml CB. As a reference sample, 1.0 mg CdSe QDs dissolved in 4 ml CB was used without any addition of TrGO. The PL spectra were recorded with a J&M FL3095 spectrometer (J&M, Germany).

58


Figure 3.9 (a) PL quenching of TrGO decorated with CdSe QDs with different TrGO/CdSe QDs ratios. (b) Maximal PL intensity values as a function of TrGO content extracted from (a). PL-quenching experiments were performed to investigate whether charge transfer from CdSe to TrGO occurs. TrGO with suitable work function can be utilized as an acceptor [207,239] for extracting charge carriers from CdSe QDs resulting in a quenching of the CdSe QD PL signal. In the experiments, the PL of 1 mg CdSe QDs in 4 ml CB was compared to the PL of the same amount of CdSe QDs while adding increasing amounts of TrGO. As depicted in Fig. 3.9, the PL intensity of CdSe QDs in the hybrid material is dramatically quenched with increasing amounts of TrGO. This PL quenching thereby demonstrates a potential charge or energy transfer from CdSe to TrGO as reported in literature from similar hybrid systems [222,239], but here this phenomenon is probably enhanced by the direct binding of CdSe QDs to TrGO. So, the PL-quenching is an indication of efficient charge transfer occurring in the hybrid material from CdSe QDs to TrGO. The LUMO level of CdSe QD is at about 3.5 eV [240] and above the work function of graphene which is about 4.5-4.7 eV [241]. Therefore, the electron transfer from CdSe QDs to graphene is energetically favored as schematically illustrated in Fig. 3.10. This fact might be very promising for future application of these materials in energy harvesting e.g. in hybrid solar cells or in other applications such as transistors, photodetectors and sensors which will be addressed in detail in the application parts, presented in chapter 5 and chapter 6.

59


Figure 3.10 Schematic illustration of photo-induced charge transfer occurring in CdSe QDTrGO. The work function of TrGO is about 4.5–4.7 eV [241] and LUMO and HOMO levels of 6-nm diameter CdSe QDs were calculated as 3.5 and 5.5 eV [240,242], respectively. PL quenching experiments as a function of times were performed during the formation of CdSe QD-TrGO hybrid material to investigate the reaction kinetics. The PL signals were recorded every two minutes while the solutions were stirred at 110 °C. The PL intensity of

Figure 3.11 (a) PL quenching of a CdSe QD-TrGO solution compared to a CdSe QD-rGO solution under the same conditions as a function of incubation time. (b) PL spectra of CdSe QDs, CdSe QD-TrGO and CdSe QD-rGO after 2 min incubation time. the CdSe QD related peak in CdSe/rGO and CdSe/TrGO mixtures dropped by 66.69% and 82.47% after two minute of stirring, respectively (Fig. 3.11b). Fig. 3.11a reveals a faster drop of the PL signal for the CdSe QD/TrGO mixture (red line) as compared to the mixture of 60

3.2 Synthesis and application of Ag NP-graphene hybrid material

CdSe QDs/rGO (black line). By time, direct chemical attachment of CdSe QDs to TrGO can evolve while for the CdSe QD/rGO mixture physical attachment might occur with reduced electronic coupling. Therefore, we conclude that photo-induced charge or energy transfer is favored in CdSe QD-TrGO hybrid material as compared to CdSe QD/rGO mixtures.

3.2 Synthesis and application of Ag NP-graphene hybrid material 3.2.1 Motivation Silver NPs (Ag NPs) are well-known materials for their unique optical, electronic and catalytic properties, which are showing advancement to its bulk counterpart [243]. Therefore, they find potential applications in sensors, as catalysts, in nanoelectronic devices and optical switches, etc. However, in most cases, the NPs tend to aggregate during catalytic processes due to van der Waals forces and high surface energy. To avoid the aggregation, they are often protected by surface modification using polymers, complex ligands or surfactants [244]. But these protecting coatings in turn affect the catalytic activity of the Ag NPs. These problems might be resolved by anchoring Ag NPs directly to a support material such as graphene. Thereby, graphene not only protects the particles from aggregation, maintaining the high specific surface area, but also serves as conducting framework enhancing the catalytic activity of Ag NPs in e.g. electrocatalysis applications. 3.2.2 Synthesis Sodium borohydride (NaBH4) and silver nitrate (AgNO3) were purchased from Merck. The synthesis of Ag NPs was performed based on a modified procedure reported by Shen el al. [245]. Briefly, 340 mg AgNO3 was added to 30 ml of 0.05% GO aqueous solution in an Erlenmeyer flask. Then, a magnetic stirring bar was added and the flask was placed in an ice bath onto a stirring plate. The solution was stirred and cooled for about 20 minutes. Afterwards, 30 ml aqueous solution containing 38 mg NaBH4 was dripped into the Erlenmeyer flask under continuous stirring with approximately 1 drop per second. The reaction was finished after 1h stirring. The Erlenmeyer flask was kept in the ice bath for the whole reaction time. 3.2.3 Results and discussions In the solution, Ag+ ions tend to attach to the negatively charged carboxyl groups of GO. When NaBH4 is put into the reaction solution, Ag+ ions are reduced to Ag (0) evolving as NPs on the surface of GO flakes. Simultaneously, GO is reduced by the NaBH4 reductant. Ag 61


NPs deposited on the rGO surface prevent rGO flakes from stacking together, and rGO sheets in turn act as scaffolds keeping the NPs apart from aggregation. Moreover, the hydrophilic property of Ag NPs enables the Ag NP-rGO hybrid material being dissolvable in water or other polar solvents such as DMF. This is of high advantage allowing the hybrid material

Figure 3.12 (a) UV-Vis absorption spectrum of Ag NPs decorated rGO (Ag NP-rGO) and (b) TEM image of this hybrid material. to be solution-processable for further modification or integrated into applications. The UVVis absorption spectrum of Ag NP-rGO hybrid material (Fig. 3.12a) contains a peak at 267 nm deriving from rGO, and a typical peak at 425 nm deriving from Ag NPs. TEM samples were prepared by dip-coating a TEM grid into an aqueous Ag NP-rGO dispersion. The TEM image (Fig. 3.12b) shows that Ag NPs anchor to both surfaces of the rGO flake. No particles are present outside of the graphene sheet that proves an efficient and dedicated decoration process. However, the size of Ag NPs is inhomogeneous; it might be due to the presence of rGO which disturbs the growing process of Ag NPs. 3.2.4 Application of Ag NP-rGO hybrid material in DSSCs DSSCs also known as Grätzel cells were firstly reported in 1991 [154]. The DSSC is a nanostructured photoelectrochemical device. Light is absorbed by a dye attached to the surface of a mesoporous large band gap semiconductor such as TiO2. Typically, functional ruthenium (II)-polypyridyl complexes are used as dye. Solar energy is transformed into electricity via the photoinduced injection of an electron from the excited dye into the conduction band of the semiconductor. The electrons move through the semiconductor to a 62


current collecting electrode into the external circuit. A redox mediator such as I3−/I1− in the pores of the semiconductor ensures that oxidized dye species are continuously regenerated and that the whole process is reversible [246]. In a previous report [131], graphene demonstrated its well catalytic activity towards I3−/I1− redox electrolyte. Here, Ag NP-rGO nano-hybrid material has been applied in catalytic counter electrodes of DSSCs. Herein, preliminary experiments are shown to demonstrate the proof of concept in using Ag NP-rGO hybrid materials as catalytic counter electrode for DSSCs. a) Cell fabrication Three different DSSCs were fabricated with different counter electrodes made out of Ag NPrGO, rGO and platinum, respectively for direct comparison. Ag NP-rGO counter electrodes were fabricated by drop-casting 150 μl of 0.1 wt % solution of Ag NP-rGO dispersed in DMF onto a 2×2 cm2 glass substrate, followed by evaporation of the solvent in ambient atmosphere at room temperature for three hours. Consequently, the samples were annealed on a hot-plate at 100 C for 30 min for improving: the interconnection of rGO flakes, further reduction of rGO, and increasing the thin film conductivity as a result. The rGO counter electrode was prepared according to a similar procedure to the Ag-rGO counter electrode. For the Pt counter electrode, 2 μl of 5 mM chloroplatinic acid in isopropanol was drop-cast on a FTO electrode. The sample was then heated to 380 C for 20 min. Cells were fabricated using a procedure according to a previous report [131]. In brief, 2 g of P25 Ti NPs (Evanonik) were suspended with 66 μl of acetyl acetone and 3.333 ml DIdistilled water. Ti films were casted on TiCl4 treated FTO glass using a Scotch tape mask and a glass rod via the doctor blade technique. These films were then heated to 450 °C for 30 min in air before being placed in a 0.2 M TiCl4 solution for 12 h and heated to 450 °C for 30 min. The resulting electrode was immersed in a 0.3 mM N3 dye/ethanol (Acros) solution for 20 h to form the sensitized photoanode. Ag NP-rGO, rGO and Pt counter electrodes were formed as described above. A 25-μm Surlyn film (Solaronix) was used to separate the photoanode and the counter electrode and to seal the cell after the electrolyte (Iodolyte AN-50 from Solaronix) was added. Cells were tested immediately after fabrication.

63


b) DSSC characterization The DSSCs were characterized under ambient atmosphere by a computer-controlled Keithley 2602A source-meter in two-point probe setup. The cells were individually illuminated by a LOT-Oriel Sun Simulator, housing a xenon lamp and using an AM 1.5G filter. The light was coupled to a solar cell device holder by a liquid light guide from Lot-Oriel. The light intensity was adjusted by a calibrated silicon reference solar cell to match 100 mW/cm². c) Results and discussions

Figure 3.12 Current density-voltage (J-V) characteristics for the DSSCs with thin layer of rGO (black line), Ag NP-rGO (red line) and Pt (blue line) deposited on glass substrates as counter electrodes. Measurements were performed under AM 1.5G illumination. A thin film of Ag NP-rGO was deposited on a glass substrate by drop casting; the as-received film is semi-transparent, and conductive. The film has then been used as a catalytic counter electrode for a simple DSSC. Preliminary experiments show promising results presented in Fig. 3.12 and Table 3.1. The results reveal that Ag NP have a well catalytic activity towards I3−/I1− redox electrolyte, enhancing the DSSC PCE by 63.5% from 0.126% for a rGO based cell to a value of 0.206 % for a Ag NP-rGO based cell. This enhancement can be attributed to two factors: One is the synergetic effect of Ag NPs and rGO resulting in enhanced catalytic 64

3.3 Synthesis of ZnO NP-graphene hybrid materials

Table 2.1 Comparison of the results of three simple DSSCs with counter electrodes made out of rGO, Ag NP-rGO and Pt deposited on glass substrates. Counter electrode type

JSC [mA/cm²]

rGO

1.02

0.351 0.32 0.126

Ag NP-rGO

1.34

0.304 0.36 0.206

Pt

1.68

0.34

FF

VOC [V]

PCE [%]

0.49 0.331

activities. And the other is that Ag NPs decorated on rGO flakes enlarge the interlayer between rGO flakes, keeping them apart from aggregation, leading to the enlargement of the catalytic active surface area. However, The PCE of the Ag NP-rGO based cell is still lower than that of the Pt based cell. More work must be devoted to improve the Ag NP-rGO hybrid material quality, to adjust its parameters such as NP size, rGO reduction degree. The Ag NPrGO counter electrode fabrication also needs to be further investigated to improve the performance of Ag-rGO based cells to be competitive with Pt-based DSSCs. However, since this is not the main focus of the thesis, this proof of concept results has not been further investigated.

3.3 Synthesis of ZnO NP-graphene hybrid materials TrGO and rGO can also be decorated with ZnO NPs forming ZnO NP-TrGO and ZnO NPrGO hybrid materials, respectively by the previously introduced self-assembly decoration process. These hybrid materials have been demonstrated promising for many applications such as photocatalysis [247], gas-sensor [248], and solar cells [249]. Herein, ZnO NPs can be either bought from commercial stock or separately synthesized; therefore, their quality is easily controllable. Synthesis and results ZnO NPs and rGO or TrGO were simultaneously dispersed in ethanol, and then vigorously stirred for 30 min on a stirring bar at 60 C. As a result, ZnO NP-GO or ZnO NP-TrGO was obtained by a self-assembly decoration, as proven for CdSe QD-TrGO. Fig. 3.13 shows a

65


TEM image of a TrGO sheet decorated with ZnO NPs (the ZnO NPs were obtained from Sigma). The image reveals a high loading of the NPs on a TrGO sheet, and there are no NPs

Figure 3.13 TEM image of ZnO NPs attached to a TrGO sheet

Figure 3.14 TEM image of ZnO NPs attached to a rGO sheet

66


outside of the TrGO sheet, proving the effective decoration. Fig. 3.14 exhibits a TEM image of ZnO NPs decorated on an rGO sheet. Here, ZnO NPs were synthesized by a separate experiment. The results show a facile synthesis towards hybrid material of ZnO NPs and graphene. Each material can be separately synthesized or obtained from commercial stock. This is an advantage for controlling material properties towards specific applications. However, more work should be dedicated to improve the uniformity of the resulting hybrid material by enhancing the solubility of both ZnO NPs and graphene or other experimental parameters such as the reaction time or reaction temperature.

67

4. 1 Introduction and motivation of graphene EPR research

Chapter 4 EPR investigations of functionalized graphene and CNT, and CdSe QD-TrGO hybrid materials (Main parts of this section were published in Appl. Phys. Lett. 104, 132102 (2014) entitled “Comparative EPR investigation of graphene and carbon nanotube with different functionalities as well as CdSe quantum dot – graphene hybrid material” this work was performed in cooperation with group of Prof. Dr. S. Weber, Institute of Physical Chemistry, University of Freiburg)

4. 1 Introduction and motivation of graphene EPR research 4.1.1 Introduction to EPR spectroscopy [250–252]. Electron paramagnetic resonance (EPR) or electron spin resonance (ESR) spectroscopy is utilized for studying materials with unpaired electron spins. The concept of this method is similar to the nuclear magnetic resonance (NMR) technique, but deals with the interaction of electromagnetic radiation with electron spins instead of nuclear spins in NMR. Electrons have a magnetic moment and spin quantum number s = , with two magnetic components ms =

and ms =

, corresponding to two spin states α and β, respectively (Fig. 4.1). When

an external magnetic field with strength Bo is applied, the electron’s magnetic dipole moment aligns itself either parallel (ms =

) or antiparallel (ms =

) to the field, each alignment has

a specific energy due to the Zeeman effect [250,253]: E = msgeμBB0 where

(4.1)

ge is the electron’s g-factor and ge = 2.0023 for free electron [254] μB is the Bohr magneton. The g-factor allows distinguishing and identifying different types of samples; carboncentered radicals typically have a g-factor close to the “free electron" value. Heteroatoms may shift the g-factor. For example, benzosemiquinones containing a considerable spin density on the oxygen have g  2.004 and nitroxide radicals contain a spin density on nitrogen and oxygen giving g  2.006, and so on. The separation energy between the lower and the upper state is: 68


E = Eα - Eβ = geμBBo

(4.2)

for the unpaired electron spins, indicating the splitting of the energy states is directly proportional to the magnetic field’s strength, as shown in Fig 4.1. An unpaired electron spin can move between the two energy states by either absorbing or emitting a photon of energy h (h –Planck constant,  - frequency of radiation), the resonance occurs when E = h. This leads to the fundamental resonance condition of EPR spectroscopy: h = geμBBo

(4.3)

This equation allows arbitrary combinations of frequency and magnetic field values. However, typical EPR measurements are performed with microwave radiation in the 9–10 GHz region in combination with magnetic fields of about 3500 G (3.5 mT).

Figure 4.1 Energy-level diagram for the simplest electron spin system (e.g., the free electron) as a function of the applied magnetic field B0. Eα and Eβ represent the energies of the ms = ± states. Hyperfine interaction: Nuclear hyperfine interactions deliver additional information such as the identity and number of atoms within a molecule of the material. Hyperfine spectral features result from the interactions of the electron's magnetic dipole moment with those of nearby nuclei. In a simplified picture, the magnetic moment of the nucleus (which is weaker than that of the electron) acts like a bar magnet and induces a magnetic field at the electron. The magnetic field opposes or adds to the magnetic field from the external magnet, depending on the alignment of the magnetic moment of the nucleus (Fig. 4.2). Hyperfine interactions lead to a splitting of the EPR absorption signal into pairs of signals depending on the number and type of nuclei as depicted in Fig 4.2.

69


Figure 4.2 Schematic illustration of hyperfine interactions and the resulting splitting of energy levels and EPR signals (modified from ref. [250]) Temperature dependence of EPR signal intensity: The population of the two energy levels is given by the Boltzmann distribution equation: = exp (-

) = exp(



) = exp(



)

(4.3)

where nupper and nlower are the numbers of paramagnetic centers occupying the upper and lower energy states, respectively. k is the Boltzmann constant, T is the temperature in K. To increase the EPR intensity, the ratio

should be lowered to increase the number of

electrons in the lower energy state to absorb radiation and transfer to the upper energy state. Based on equation (4.3), this can be achieved by performing the experiments at low temperature, for example at the boiling point of liquid nitrogen or helium. Moreover, at low levels of magnetization, the magnetization of paramagnets follows the so-called Curie law. This law indicates that the susceptibility of paramagnetic materials is inversely proportional to their temperature, i.e., those materials become more magnetic at lower temperatures. In a continuous-wave EPR experiment, the microwave frequency is kept constant, and by increasing the strength of the external magnetic field, the energy difference between the upper and the lower spin state, E, is widened until it matches the energy of the microwaves. At this point the unpaired electrons can move between lower and upper states by absorbing or emitting microwave radiation. At thermal equilibrium, the population of the lower state is larger than that of the upper state, due to Maxwell-Boltzmann distribution, thus there is a net absorption of energy, which is recorded in a spectrum. Typically, EPR spectra are often recorded in the first-derivative mode due to application of magnetic-field modulation to increase the signal-to-noise ratio.

Fig. 4.3 shows the general layout of a typical EPR

spectrometer. The electromagnetic radiation source and the detector are in a box called the

70


“microwave bridge”. The sample is in a cavity, which is, simply speaking, a metal box that helps to amplify weak signals from the sample. The magnet is to adjust the energy levels of the spins in the sample. In addition, there is a console, which contains signal processing devices and control electronics, as well as a computer. The computer is used for analyzing data as well as coordinating all the units for acquiring a spectrum.

Figure 4.3 General outlay of an EPR spectrometer (reproduced from ref. [250]). 4.1.2 Motivation EPR experiments have recently provided detailed information on the defect structure of carbon-related nanomaterials [255][220]. EPR is a powerful and sensitive method to detect active intrinsic and extrinsic paramagnetic defects in a material system. In carbon-related materials, for example, EPR does not only provide insight into the spin properties, which includes conduction electrons, unpaired spins and dangling bonds [255], but also enables investigation of electronic states in different forms of carbon [256]. All those carboninherited EPR-active defect centers are probably affecting the electronic and optical properties of a material. Zaka et al. [255] indicated that high-quality pure CNTs are indeed EPR-inactive, however, they observed an EPR signal of Lorentzian line shape at g = 2.001, which was attributed to catalyst impurities in the CNTs. When the size of graphite is reduced to nano-dimensions, the localized edge states couple with itinerant electrons, thus leading to a narrow EPR signal, while antiferromagnetism develops when the temperature decreases below 23 K [256]. Although the magnetic properties of graphene have recently attracted high attention, due to graphene’s potential application in spintronics devices, there have been only few EPR studies performed on graphene and graphene-like structures. Rao et al. [257] observed two EPR

71

4.2 Materials and methods

signals of rGO: a broad signal at g = 2.0027, which can be attributed to graphitic-like carbon, and a narrower signal at g = 2.0028 associated with carbon radicals. Beckert et al. [[258] monitored an in-situ EPR spectrum to observe free-radical grafting of homo- and copolymers onto functionalized graphene during polymerization. EPR studies also revealed the existence of Mn+2 arising from the preparation procedure of rGO from GO, and indicated that Mn+2 is anchored to the graphene lattice and cannot be removed by conventional washing [235,259]. Quenching experiments through monitoring the EPR signal by Su et al. [260] revealed that the catalytic activity of porous GO is related to the localized spins generated at the edge of the -electron system. Prakash et al. [261] found that EPR signals from ZnO-rGO hybrid materials occur at g-values in a range between 2.0042 and 2.0045, which is slightly higher than for “normal” graphitic EPR signals that typically resonate at g-values between 2.0022 and 2.0035, thus indicating an interaction among carbon-inherited spin species, -electrons in condensed sp2-domains of graphene sheets and oxygen-vacancy-related defect centers of ZnO. Here, the transformation of oxygen-containing functionalities, such as hydroxyl and carboxylic groups, within GO to thiol-containing functional groups of TrGO has been monitored by EPR. Additionally, CdSe QD-TrGO hybrid material was investigated by EPR to examine the change in electronic properties of TrGO before and after attached with CdSe QDs. The goal is to demonstrate charge transfer occurring between CdSe QDs and TrGO by EPR.

4.2 Materials and methods 4.2.1 Synthesis CdSe QD-TrGO hybrid material was prepared as described in Chapter 3 (see 3.1.4). For comparison, oxidized CNT (O-CNT1) and thiol-functionalized CNT (CNT-SH) were also prepared by the same procedures as GO and TrGO (see 2.2 and 3.4), respectively, for comparative EPR investigations. Mn-free oxidized CNT (O-CNT2) and rGO synthesis: 200 mg CNT (inner and outer mean diameter: 4 and 13 nm, respectively, length: 1 μm, Bayer Material Science, >99%) was added into a mixture of 10 ml HNO3 (Merck, 63%) and 30 ml concentrated H2SO4 (Merck, 98%) and stirred at 100 °C for 4 h. The dispersion was then poured into 150 ml DI-H2O. After that, O-CNT2 was collected by centrifugation at 4400 rpm for 2 h, and then purified by 3 times washing with 100 ml DI-H2O. Finally, O-CNT2 was dried overnight in vacuum at room temperature. rGO was synthesized by reducing GO with hydrazine (N2H4) in DMF: 20 72

4.2 Materials and methods

mg GO was dispersed in 50 ml DMF by sonication for at least 30 minutes. After 20 µl N 2H4 (Sigma, 64%) was added and the dispersion was stirred at 90 C for 12 h. The resulting rGO was then purified by centrifugation and re-dispersion in DMF three times in succession. 4.2.2 EPR experiments CdSe QD-TrGO hybrid material was prepared as described in 3.1.4, and put into a glass tube for EPR measurements. X-band (9.86 GHz) electron paramagnetic resonance (EPR) measurements were performed using a Bruker EMX spectrometer. The magnetic field was measured with an NMR gaussmeter (ER 035M, Bruker), and as a standard magnetic-field marker, polycrystalline diphenylpicrylhydrazyl (DPPH) with g = 2.0036 was used for accurate determination of the resonance magnetic-field values and the g-factor. The following EPR experimental parameters were used: microwave power: 1 mW, modulation amplitude: 0.5 G (0.01 mT), time constant: 20.48 ms and receiver gain: 2  103. 4.2.3 Spin-counting procedure In order to accurately count the number of spins, there are many important issues that should be considered before and after the EPR experiment [262]. Crucial issues, which have to be carefully taken into account, are: (i) Samples should be always weighed before the experiment to avoid complications due to different sample amounts in EPR tubes. If it is not possible to have always the same amount of sample in an EPR tube, then each spectrum should be multiplied by a filling factor for normalization [262,263] (deduced from the mass of the sample in the EPR tube). (ii) The sample position should be always adjusted to the center of the microwave cavity. (iii) If there is a background EPR signal (e.g., from impurities in the resonator), it has to be subtracted from the EPR signal of the sample. (iv) One should always be careful not to saturate the EPR signal by applying too high microwave power. The microwave phase should be carefully adjusted during the critical coupling (tuning) of the resonator. (v) One has to check whether there is an offset in the magnetic field, and calibrate if necessary. (vi) The Q value of the resonator has to be measured and all spectra should be referred to the same Q value. The number of defect centers can be quantitatively determined by the aid of EPR spectra independent from the microwave frequency. In order to calculate the defect concentration, one doubly integrates each first-derivative EPR signal. By comparing the integral of the standard sample (here, MnO powder) and the measured sample one obtains the corresponding

73

4.3 Results and discussions

number of spins, and thus, the concentration of defect centers. For an exact determination of defect concentration, one has to take into account the normalised-corrected value in the following expression including experimental parameters of both the reference and the probe under investigation:

(4.1) where NS*, RG, MF, MA, CT, P, Scans, SW, and S stand for the number of spins in the reference sample, receiver-gain, modulation frequency (in kHz), modulation amplitude (in G), conversion time (in ms), microwave power (in mW), magnetic field sweep (in G), number of scans, and spin quantum number, respectively. Note that, (*) indicates the measurement parameters for the reference sample. Once the normalized-corrected value of NS* is obtained, via simple cross multiplication of the NS* and the area under the EPR signals the defect concentration, ND, of the sample under investigation is revealed:

(2) where (Area)D and ND are the area of the EPR signal of the related defect centre and the number of spins of the sample, respectively. In this work, MnO powder was used as standard reference sample, which has 3.34  1019 spins/g.

4.3 Results and discussions 4.3.1 Removal of Mn+2 in O-CNT and GO samples In Fig. 4.4(b), the EPR spectra of TrGO and CdSe QD-TrGO are shown and compared to spectra from GO and rGO. EPR spectra (Fig. 4.4(a)) of CNT with different functionalities were recorded under identical experimental conditions for direct comparison. EPR investigations of GO and O-CNT samples revealed characteristic Mn+2 (S=5/2, I=5/2) hyperfine patterns centered at g = 2.0019, both arising from the central electronic transition MS = –1/2  MS = +1/2. The typical sextet is due to Mn impurities that originate from the preparation stage of the respective materials, where KMnO4 is used as an oxidizing agent [235,259]. In addition to the Mn+2 signals, one can observe EPR signals that derive from defects caused by chemical functional groups such as OH and COOH. It has been reported previously that such well-resolved hyperfine patterns indicate the existence of Mn2+ in the 74


compound as magnetically diluted paramagnetic complexes [235]. To get rid of the Mn+2 traces and to investigate C-related defect properties by EPR, a Mn-free oxidation method by oxidizing CNT in an HNO3/H2SO4 mixture was applied to obtain O-CNT2 samples. Consequently, the EPR spectrum of O-CNT2 (Fig. 4.4. a) exhibits only one sharp signal arising from carbon-centered radical(s), and the Mn+2 hyperfine pattern is no longer observed.

Figure 4.4 First-derivative X-band EPR spectra with and without Mn+2 hyperfine patterns of GO and reduced GO (a), and of CNT (b) with different functionalities. All spectra were recorded at room temperature (reused from ref. [242]). This proves that Mn+2 ions have been successfully removed from the sample by transformation and simultaneous reduction of oxygen-containing functional groups in both CNT and graphene, thus leading to the disappearance of the Mn+2 -related EPR signal. 4.3.2 EPR signals of graphene and CNT with different functionalities In addition to the Mn+2 signal we observed single EPR lines for both O-CNT and GO at g~2.0029, which is within the EPR signal g range between 2.0022 and 2.0035 typically found for C-related dangling bonds [258]. In this range, EPR signals deriving from spins of oxygencontaining functionalities such as carboxylic, carbonyl, hydroxyl and epoxide groups in OCNT and GO can be found. After chemical reduction, the oxygen-containing functionalities were mostly removed, and the graphene hexagonal carbon lattice is mostly restored, with the consequence, that the electron mobility is increasing. Hence, the local spin-state density is

75


reduced resulting in an almost quenched EPR signal from the rGO sample as observed in Fig. 4.4(b). However, even “nearly perfect” graphene with a very low degree of defects still exhibits a certain EPR signal because of defect centers from the zigzag edges of graphene planes [257]. When GO was thiol-functionalized and simultaneously reduced into TrGO to restore the sp2carbon structure of graphene, thiol functional groups have been introduced for attachment of semiconductor NPs directly to the graphene lattice. GO is electrically insulating while the TrGO conductivity is 5300 S/m. Figure 4.4 (b) shows that in comparison with GO the TrGO signal consists out of a broader peak (larger peak-to-peak linewidth (Bpp), see Table 4.1), which indicates that the spin density of TrGO is decreased. This results from a chemical reduction process lead to a partial restoration of the sp2-carbon structure from a decrease of oxygen-containing chemical groups from the graphene surface and an increase of thiolcontaining groups such as -CSSH, and -SH. Remaining non-bonding weakly localized radicals (e.g., trapped electrons) result in an EPR signal that is found near the g-factor of free electrons, g = 2.0023 [264]. In Table 4.1, the peak-to-peak widths (∆Bpp) and the spin densities are summarized. The spin densities were calculated by applying the spin-counting procedure, which was described above. It shows that the ∆Bpp of TrGO is 5.5 mT, which is much broader than the 0.36 mT of GO. Correspondingly there is a spin-density decrease from 2.4  1017 spins/g for GO to 1.1  1017 spins/g for TrGO. This is attributed to less defects occurring in TrGO than in GO due to the restoration of the sp2-carbon structure within graphene lattices as a result of the reduction. Table 4.1 Peak-to-peak linewidth (∆Bpp) and spin densities of functional graphene and CNT. Bpp

Spin density

(mT)

( of spins/g)

GO

0.36

2.4 × 1017

rGO

-

TrGO QD-TrGO

Sample

Bpp

Spin density

(mT)

( of spins/g)

CNT

-

-

-

O-CNT1

0.23

4.3 × 1017

5.5

1.1 × 1017

O-CNT2

0.37

0.9 × 1017

22.6 (weak)

2 × 1015

CNT-SH

6.42

2 × 1016

Sample

Table 4.1 also indicates that good-quality CNT does not contain any spins. Afterwards, the oxidation of CNT into O-CNT1 and O-CNT2 introduced defects into the CNT lattices 76


resulting in spin densities of 4.3 × 1017 spins/g for O-CNT1 and 0.9 × 1017 spins/g for O-CNT2. These data indicate that the oxidation of CNT is more effective using the mixture of KMnO4, NaNO3 and H2SO4 than using a HNO3, H2SO4 mixture. Moreover, after O-CNT1 has been transformed into CNT-SH, the spin density was decreased from 4.3 × 1017 spins/g to 6.42 × 1016 spins/g, correspondingly as a result of the simultaneous reduction during the thionation. The spin density of CdSe QD-TrGO was dramatically reduced to the low value of 2 × 1015 spins/g from original value of 1.1 × 1017 spins/g due to a charge transfer and coupling between the two materials which will be addressed in 4.3.4. 4.3.3 Comparative EPR signals of CNT-SH and TrGO

Figure 4.5 First-derivative X-band EPR spectra of CNT-SH (a), and of TrGO (b) For comparison, CNT-SH was synthesized utilizing the same procedure as for TrGO, however, the EPR signal of CNT-SH has a much smaller intensity (Fig.4.5). As a result, CNT-SH has a spin density of 2 × 1016 spins/g, which is considerably lower than that of 1.1 × 1017 spins/g for TrGO, as revealed in Table 4.1. This can be tentatively explained as follows: GO was formed from graphene platelets with much more edge boundaries than in CNT; therefore GO is probably containing more carboxylic groups than O-CNT. Moreover, only the outer shell of multi-walled CNT (MWCNT) is surface-functionalized leading to fewer defects per weight. It is known that it is more difficult to reduce carboxylic groups than hydroxyl, carbonyl, or epoxide chemical groups. Consequently, after the reduction and thionation steps, TrGO might contain more thiol groups than CNT-SH, and, as a result, an EPR signal with higher peak intensity is observed in TrGO when comparing the same amounts (equal weight), as shown in Fig. 4.5.

77


4.3.4 Quenching of the EPR signals in CdSe QD-TrGO hybrid material Surprisingly, when TrGO was decorated with CdSe QDs to form a CdSe QD-TrGO hybrid nano-composite, the EPR signal of TrGO is nearly quenched as shown in Fig. 4.6. This is most probably due to direct chemical binding of the QDs to TrGO, leading to a better electronic coupling, thus favoring more efficient charge or energy transfer between QDs and TrGO. This hypothesis was also proven by PL quenching experiments presented in Chapter 3 (3.1.4.2), and also demonstrated by the shift of energy levels of the QDs in cyclic voltammetry measurements described in Chapter 5. Moreover, the EPR signal of CdSe QD-TrGO is quenched even in the dark due to the direct chemical binding between the two materials during formation of the CdSe QD-TrGO hybrid material.

Figure 4.6 EPR spectrum quenching of CdSe QD-TRGO (lower line) and the EPR spectrum of TrGO (upper line) measured under the same condition. 4.3.4 Temperature-dependent EPR spectra of O-CNT and TrGO For many studies it is necessary to vary the temperature of the samples. Temperature control may be needed to hold a sample in a well-defined state of equilibrium or to study at a series of temperatures. Moreover, as described in the introduction (see 4.1), the sensitivity of EPR is increasing with decreasing temperature. Figure 4.7 (a,b) shows temperature-dependent X-band EPR spectra of O-CNT1 and TrGO in the temperature range between 100 and 300 K. The EPR signal of O-CNT1 exhibits the

78


expected temperature dependence given by Curie’s law, while the EPR signal for TrGO does not change significantly as a function of temperature, thus indicating Pauli-type behavior. This Pauli-type paramagnetism is resulting from itinerant electrons within TrGO, and is independent of temperature. Curie-type behavior suggests that the number of defects that exist initially in the carbon matrix increases by heat treatment. It is supposed that thiol groups

Figure 4.7 Temperature-dependent first-derivative X-band EPR spectra of (a) O-CNT1 and (b) TrGO. Temperature dependence of the EPR-susceptibility of c) O-CNT1 and d) TrGO, obtained after double integration of the EPR spectra from a) and b), respectively.

79

4.4 Conclusions

are good radical trapping centers, keeping spins more localized with increasing temperature, thus leading to a TrGO EPR signal, which is independent in a temperature range below the Curie temperature of 222 K. The measured EPR spectra can be analyzed quantitatively in terms of double integration of the dispersive-type continuous-wave EPR which is given as EPR susceptibility. The related Curie-plots are given in Figure 4.5c and 4.5d. The EPR-susceptibility of O-CNT1 is proportional to the experimental temperature in a range of 222–300 K, while under 222 K the dependence becomes exponential. Obviously, the EPR line belonging to O-CNT1 follows a ferromagnetic behavior with a Curie temperature of TC = 222 K (Fig 4.6a). In contrast, TrGO shows a more complicated behavior (Fig. 4.6b,c). The EPR signal of TrGO is independent to temperature below TC (222 K), which is the characteristic for a Pauli paramagnetism. Between 222–275 K, the TrGO EPR line starts to tumble destroying the magnetic coupling of the electron spins which has been also recently reported for Fullerene anions [265].

4.4 Conclusions The EPR investigations on functionalized graphene and MWCNTs as well as CdSe-TrGO confirm that Mn+2 ions induced from potassium permanganate reduction during GO synthesis remain chemically bound in GO lattices. They were removed from TrGO during the reduction and surface functionalization steps. The EPR signal of TrGO derives from unpaired electrons localized in lattice defects and functional groups. The temperature-dependence of the EPR signals for O-CNT1 and TrGO was investigated and compared. While the temperature- dependence of EPR for O-CNT1 is based on Curie’s law, the dependence of TrGO is complicated, and shows characteristics of Pauli paramagnetism at temperatures below 222 K. The EPR signal is completely quenched in CdSe QD-TrGO hybrid material due to direct chemical binding of CdSe QDs to TrGO. More pronounced, PL quenching in CdSe QD-TrGO hybrid nano-composites might result from improved light-induced charge transfer based on the direct chemical QD attachment. This implies that the CdSe QD-TrGO nanocomposite has promising potential for various optoelectronic applications such as photodetectors and solar cells, where photo-induced charge transfer and good electric transport properties are of utmost importance for device performance.

80

5.1 Introduction to photovoltaics

Chapter 5 Application of QD-TrGO hybrid material in photovoltaics (Main parts of this chapter were published in Physical Chemistry Chemical Physics 16 (24), 1225112260 (2014) entitled “Improved Efficiency for Bulk Heterojunction Hybrid Solar Cells by utilizing CdSe Quantum Dot - Graphene Nanocomposites”. In this work, I was responsible for the CdSe QDTrGO preparation. The solar cell fabrication and characterizations were performed in cooperation with Dr. Michael Eck. The work was published under shared authorship and has also partially been presented in Dr. Eck`s PhD thesis [266])

5.1 Introduction to photovoltaics The urgent need for the use of renewable energy is more and more increasing due to the shortage of traditional energy sources such as fossil fuels and the negative consequences of their use, for example their contribution to climate change and environmental pollution. Photovoltaic (PV) energy conversion is attracting great attention as renewable energy source, besides wind, rain tides and geothermal heat energy technologies. Solar PV is nowadays the third most important renewable energy source in terms of globally installed capacity. Development of solar cells is divided into first, second and third generation solar cells. First generation solar cells are based on crystalline silicon and are still dominating the PV market. So far, in this category, the best single crystalline silicon solar cell has achieved a PCE of 25% [267]. In order to reduce the costs, second generation solar cells have been developed. These types of solar cells are mostly associated with thin film solar cells which are much cheaper than the first generation cells. They are fabricated by depositing thin films of photoactive materials on substrates by cheaper manufacturing processes, and by consuming significant less amounts of material. The most common photoactive materials of thin film solar cells are copper indium gallium selenide (CIGS), CdTe, amorphous silicon (a-Si), and Gallium arsenide (GaAs). The highest reported efficiencies of thin film solar cells are 20.1, 19.6, 13.4 and 28.8% [268] for CIGS, CdTe, a-Si and GaAs cells, respectively. Third generation solar cells include multi-junction tandem, organic, dye-sensitized solar cells, as well as QD–polymer hybrid and perovskite solar cells. They are potentially able to overcome limitations of current solar cell technologies such as limited power efficiency and high cost production. They are still in an early stage of ongoing development with no or limited commercial availability. 81


5.1.1 Hybrid solar cells This chapter was performed in cooperation with Dr. Michael Eck. It reports on a significant efficiency enhancement of hybrid solar cells by using graphene as transporting and QD-supporting material in the active layer of solar cells. Hybrid solar cells have a similar working principle to organic solar cells, where fullerenebased electron acceptors are replaced by inorganic QDs from e.g. CdSe, CdS, CdTe, PbS and so on. Moreover, QDs can additionally act as 2nd absorption material in the photoactive layer. The light spectrum that QDs absorbs can be adjusted by tuning the QDs band gap through controlling the diameter of QDs or the QD material. a) Device structure of hybrid solar cells Hybrid solar cells are thin film devices consisting of photoactive layer(s) between two electrodes of different work functions (Fig. 5.1).

Figure 5.1 Schematic structure of a typical hybrid solar cell. (Reproduced from ref. [269]) Higher work function conductive transparent indium tin oxide (ITO) on a flexible plastic or glass

substrate

is

often

used

as

anode.

Conducting

polymer

poly(3,4-

alklenedioxythiophenes):poly(styrenesulfonate) (PEDOT:PSS) is usually spin-coated on ITO as an anode buffer layer, enhancing additionally the adhesion to the upper light absorbing layer, The PEDOT:PSS layer also ensures a better energy level matching and lead to an improved device stability by hindering oxygen and indium diffusion through the anode, and 82


by blocking electrons from going to the anode, while letting hole going through [270,271]. The photoactive layer is normally made by spin-coating a NC/polymer blend solution onto the ITO substrate to form a thin film of about 100-200 nm. Finally a top metal electrode (Al, LiF/Al and Ca/Al) is vacuum-deposited onto the photoactive layer. b) Working principle Similar to organic solar cells, photo current generation is a multistep process in hybrid solar cells. Briefly, it can be summarized in four main steps which are illustrated in Figure 5.2.

Figure 5.2 Schematic diagram of the photocurrent generation mechanism in typical (a) bilayer heterojunction and (b) bulk-heterojunction organic solar cells, where photons are mainly absorbed by the donor material. Photogeneration process: exciton generation (1), exciton diffusion (2), charge transfer (3), charge carrier transport and charge collection (4). (Reproduced from ref. [269]) 1) Photon absorption First, incident photons with an energy hν are absorbed mainly by the donor material and excite the electrons from the highest occupied molecular orbital (HOMO) to the lowest unoccupied molecular orbital (LUMO) level, creating excitons with a certain binding energy (typically 200-500 meV [272,273]). 2) Exciton diffusion

83


In order to generate separated negative and positive charges, the excitons need to diffuse to the donor/acceptor (D/A) interface. Since excitons are neutral species, their motion is not affected by any electric field and they diffuse via random hoppings. Note that the exciton diffusion lengths are typically around 10-20 nm [274,275] for most conjugated polymers before recombination takes place. Excitons that do not reach the D/A interface are lost for the energy conversion and dot not contribute to the photocurrent. 3) Charge Transfer Excitons dissociate at the D/A interface resulting in a charge when the LUMO level of the QDs is lower than that of the polymer, and the HOMO level of the polymer is higher than that of the QDs. The offsets in both HOMO and LUMO levels must be larger than the exciton binding energy minus the coulomb binding energy of the charge-separated state [276]. 4) Charge carrier transport and collection In the final step, once charge transfer has occurred at the D/A interface, separated holes and electrons are distributed within the donor and acceptor phases, respectively. Holes and electrons are then transported towards their respective electrodes driven by an internal electric field deriving from the Fermi level difference of the electrodes. The charges are collected at the respective electrodes. 5.1.2 Applications of graphene in photoactive layer of solar cells. The applications of graphene as conducting transparent electrodes and catalytic electrodes have been already discussed in detail in 1.3.5. Here, the use of graphene in photoactive layers will be presented. Graphene can be functionalized with other species forming light-harvesting material for PV devices. Its high charge mobility and large specific surface area make graphene a good candidate as supporting and charge transporting material. a) Light harvesting materials Graphene itself does not absorb light very efficiently. However, with various functional groups graphene can easily be functionalized with other absorbing materials to improve the absorption property of the resulting graphene hybrid materials. For example, GO sheets were functionalized with phenyl isocyanate (SPFGraphene), and then mixed with poly(3octylthiophene) (P3OT) to form the P3OT/SPFGraphene composite [277]. This composite was then annealed to reduce GO’s functional groups, so its conductivity was improved, while P3OT was crystallized simultaneously. The P3OT/SPFGraphene was applied in the active layer of a bulk heterojunction (BHJ) OPV device [277], which achieved a PCE of 1.4%. This 84


result proves that graphene can potentially replace [6,6]-phenyl C61-butyric acid methyl ester(PCBM) as the electron acceptor for high-performance OPV devices. Anilinefunctionalized graphene quantum dots (ANI-GQDs) were also used as electron acceptor in OPV with P3HT as electron donor. A maximum PCE of 1.14% was obtained from such a cell, consisting out of 1 wt% of ANI-GQDs and P3HT, which is much higher than the value of 0.65% obtained from a device with 10 wt% of aniline-functionalized graphene sheets and P3HT [278]. b) Graphene for Schottky junctions Metallic graphene can form the Schottky junctions with other semiconductors, which can be employed as active layer for solar cells. For example, graphene was patterned by lithographical techniques and used for the fabrication of CdSe nanobelt (NB)/graphene Schottky junction solar cells [279]. The cell demonstrated an excellent photovoltaic behavior, with an open-circuit voltage (Voc) of 0.51 V, a short-circuit current density (Jsh) of 5.75 mA cm-2, and a PCE of about 1.25%. Impressively, the performance of Schottky solar cells based on the graphene-Si can be greatly enhanced after a proper doping of graphene [280] with bis(trifluoromethanesulfonyl)amide (TFSA) was applied. The TFSA doping increased the work function of graphene, thus enhancing the built-in potential between the doped-graphene and n-Si in solar cells. As a result, the doped-graphene based cell has a PCE as high as 8.6% compared to 1.9% of the non-doped graphene based cell, as a real breakthrough in graphene based solar cells. The non-doped device showed an external quantum efficiency (EQE) of  50% in the wavelength range of 400-850 nm. After the TFSA doping, the EQE was considerably enhanced up to the value of about 65% in the aforementioned wavelength range. This enhancement occurred due to the more efficient charge separation and charge collection as a consequence of the increased built-in potential and due to an overall reduced sheet resistance. Although, the efficiency of graphene based solar cells is still lower than their organic or Si counterparts, the above mentioned examples are demonstrating the high potential of graphene for PV applications. c) Graphene as supporting and charge transport material The high mobility of graphene can be exploited to enhance the charge transport within solar cells by incorporating graphene with other materials forming nanocomposites. For example,

85


rGO was combined with TiO2 NPs forming rGO-TiO2 nanocomposites. The rGO-TiO2 was then used in DSSCs demonstrating a better performance than CNT-TiO2 composite based DSSCs [281]. This is because the particles can anchor onto graphene more efficiently, therefore the photogenerated electrons can be easily captured and transferred to graphene. So far there have been only few reports on the application of graphene in solar cells. 5.1.3 Motivation

Figure 5.3 Schematic illustration of the working principle of hybrid solar cell with and without graphene. Photo-electrons are transferred from polymer to QDs, and afterwards transferred to graphene and consequently transported to the cathode. Likewise, holes are transferred from QDs to polymer and then transported to the anode. There were already certain achievements realized for the integration of graphene into applications as transparent electrodes and catalytic electrodes for DSSCs and OPV [282,283]. Graphene-based devices demonstrated competitive or even advanced performance compared with conventional devices. However, because of difficulties in material processing, there have been only few reports on the integration of graphene into the photoactive layer as supporting and charge transporting material in solar cells.

86

5.2 Methods and experiments

In this work, graphene has been incorporated into the active layer of hybrid solar cells by utilization of CdSe QD-TrGO hybrid materials as electron acceptor, charge transporter, as well as light-harvesting material (as illustrated in Fig.5.3). The expected benefits after the incorporation of graphene into the active layer of hybrid solar cells are the following: 

Graphene might play the role as electron extractor by anchoring directly to CdSe QDs, improving the interconnection between the QDs. An improved inorganic network might lead to a lower amount of QDs needed in the blend for an efficient charge extraction, making it possible to increase the fraction of polymer in the active layer for increasing the light absorption.



The QDs aggregation challenge might be overcome, because QDs are anchored to graphene sheets, which act as supporter material. Therefore a better long term device stability might be expected due to a better stability of the internal D/A.

5.2 Methods and experiments 5.2.1 Power conversion efficiency Power conversion efficiency (PCE) is one of the most import parameter to characterize solar cell performances. It is defined as the percentage of maximum output of electrical power to the incident light power. Fig.5.4 shows the current density-voltage (J-V) characteristic for a

Figure 5.4 J-V characteristic of a typical solar cell in the dark (dashed line) and under illumination (solid line). Typical solar cell parameters such as short-circuit current density Jsc, open-circuit voltage Voc, and the maximum power point Pm are illustrated in the graph. (Adopted from Ref. [269]). 87


typical hybrid solar cell in the dark and under illumination. The PCE can be described as PCE =

=

(5.1)

where Pm is maximum power point, Pin is the incident light intensity, Jsc is the short-circuit current density, and Voc is the open-circuit voltage, and FF is the fill factor, which is defined as the ratio of Pm to the product of Jsc and Voc. FF =

(5.2)

5.2.2 Solar cell fabrication PCPDTBT with a molecular weight of Mn=10-20 kDa was purchased from 1-Material. ITO substrates (Resistivity ≤ 10 Ωsq) obtained from Präzisions Glas & Optik GmbH, and WS4006NPP-Lite spin coater from Laurell Technologies. Baytron AI4083 PEDOT:PSS was bought from HC Starck. The hybrid solar cells have been fabricated according to a previous publication [212]. For the CdSe QD-TrGO/PCPDTBT solar cells, the CdSe QD-TrGO hybrid nano-composites were utilized instead of CdSe QDs only. As-received CdSe QDs have a relatively homogeneous spherical shape and size with a typical diameter of 6 nm showing an average PL peak position of 658.4 ±7.7 nm, an average full width at half maximum (FWHM) of 29.7 ±1.1 nm, and an average 1st excitonic absorption peak at 637.3 ±5 nm. For the reference cells, the CdSe QD/CB solution with a NC concentration of 24 mg/ml was mixed in a weight ratio of 88:12 with a 20 mg/ml solution of PCPDTBT in CB. For graphene based cells, the TrGO-CdSe QD/CB solution was mixed with PCPDTBT/CB solution in a weight ratio of 85:15, which were found to be the optimum ratios. The final ink was spin-coated on a pre-structured ITO substrate at 800 rpm for 30 s followed by a 60 s drying step at 1800 rpm, resulting in an active layer thickness of about 80 nm. Before spin-coating, the pre-structured ITO substrates were treated for 5 min with oxygen plasma and then spin-coated with Baytron AI4083 PEDOT:PSS at 2000 rpm for 30 s and dried for 20 min at 160 °C, to form a 70 nm thick hole blocking layer. After thermal evaporation of a 80-nm aluminum layer, the cells were annealed at 145 °C. Therein, the optimum annealing time for TrGO containing cells proved to be with an average of 14.5 min which is longer than the average of 9.5 min needed for CdSe QD/PCPDTBT solar cells to reach their optimum performance.

88


4.2.3 TEM tomography (These experiments were performed in collaboration with the “organic and hybrid photovoltaics” group of PD Dr. H. Borchert, Institute of Physics, University of Oldenburg) The samples were prepared by dissolving the PEDOT:PSS layer of a hybrid BHJ solar cell in a water bath. The active layer thereby delaminated from the ITO substrate after about 30 s and began then to float on top of the water. Subsequently, the floating layer was collected as a planar film on a carbon film coated 300 mesh copper TEM grid (Quantifoil Micro Tools GmbH, Germany). Acquisition of tilt series for TEM tomography was performed on a Jeol JEM2100F electron microscope (Jeol Ltd., Tokyo, Japan) operated at 200 kV. All tilt series were obtained in an automatic fashion by using TEMography™ microscope control software in a tilt angle range of approximately -60° to 60° in steps of 2°. The alignment and reconstruction of the data series and visualization of the 3D reconstructed volume was carried out by using the TEMography™ software package, Composer and Visualizer-Kai (System in Frontier Inc., Tokyo, Japan). 5.2.4 Solar cell characterization Solar cells were characterized inside a nitrogen filled glovebox by a computer controlled Keithley 2602A source-meter in a 2-point probe setup. The cells were individually illuminated by a LOT-Oriel Sun Simulator, housing a xenon lamp and using an AM 1.5G filter. The light is coupled to a solar cell device holder inside the glovebox by a liquid light guide from Lot-Oriel. The light intensity is adjusted by a calibrated silicon reference solar cell to match 100 mW/cm². Solar cells were transferred inside a sealed flask to the group of dye and organic solar cells of the Fraunhofer Institute for Solar Energy Systems (ISE) for testing. First, spectral response measurements were conducted for the tested solar cells and spectral mismatch factors of 0.956 for the TrGO solar cell and 1.017 for the CdSe QD reference solar cell were determined. The solar cells were measured inside a glovebox in a 4-point probe setup using a computer controlled Keithley 2400 source-meter. The solar cells were illuminated by a K. H. Steuernagel Lichttechnik GmbH sun simulator through a window at the bottom of the glovebox with light intensities that were adjusted to the respective spectral mismatch correction factor by a calibrated reference silicon solar cell. For exact determination of the active area of two best devices, photos have been taken to exactly determine the active area represented by the overlapping region between the ITO substrate and the aluminum top electrode.

89


5.2.5 Cyclic voltammetry (CV) experiments CV experiments were performed to investigate the change of LUMO and HOMO levels of QDs caused by their attachment to TrGO. This could be one of the factors that might determine the Voc of hybrid solar cells. For working electrode preparation: An ITO substrate was firstly cleaned by sonication in absolute ethanol, followed by DI-distilled H2O and dried afterwards in a nitrogen flow at RT. Then, 20 g/ml CdSe QDs in CB was drop-cast on the ITO substrate and dried by a gentle flow of N2 at RT. CdSe QD-TrGO samples were prepared in the same way as CdSe QD samples. A 0.1 M solution of tetra-n-butylammonium hexafluorophosphate (TBAP, 98%, Sigma-Aldrich) in acetonitrile (HPLC grade, EMD) was used as electrolyte. Prior to each measurement, the electrolyte solution was saturated by bubbling N2 for at least 15 minutes. Then CV measurements were performed by sweeping a potential from 0 to 2 V with a speed of 50 mV/s by using a Heka Potentiostat PG 340 with a Pt wire counter electrode and a reference electrode of Ag/AgCl in 3M KCl.

5.3 Results and discussions 5.3.1 J-V Characteristics The utilized solar cells were prepared according to the description given in the method section as depicted in Figure 5.5. They comprise a pre-structured ITO anode, a 70 nm PEDOT:PSS electron blocking layer, a ca. 80 nm thick hybrid CdSe QD/polymer or CdSe QD-TrGO/polymer active layer, and a ca. 80 nm thick aluminum cathode. No additional hole blocking layer was introduced.

Figure 5.5 Top view of the design of the utilized hybrid BHJ device containing three individual solar cells (left, middle and right) on one substrate. The active layer consists either of a blend of CdSe QDs and PCPDTBT polymer or of a blend out of the CdSe QD-TrGO hybrid material and PCPDTBT. (Reused from ref. [284]) 90


The use of CdSe QD-TrGO hybrid material as electron acceptor has delivered a considerable enhancement of PCE by 36.7% and 45.7% measured in an external laboratory as shown in Table 5.1. This improvement is mainly resulting from the increase of Voc in the graphenebased cells, while Jsc is only a little bit higher and the fill factors are similar in those cells. The Voc improvements can firstly be attributed to a better charge transfer at the D/A junctions. The CdSe QD-TrGO based cell has a Voc of 0.71 V which is 0.15 V higher than the Voc of 0.56 V for the TrGO free cell (table 5.1). Excited electrons are transferred from the polymer to QD acceptors, and then directly go to TrGO and from there they are transported to the Al electrode. The electron transfer from CdSe QDs to TrGO is favorable, because 6 nm diameter size CdSe QDs typically have a LUMO level of about 3.5 eV calculated according to Brus et al. [215] lying above the work function of TrGO which is assumed to be similar to the work function of rGO of about 4.5 eV [214]. Due to the high electron mobility, graphene offers a good percolation pathway for photo-induced electrons that in turn enhances the charge transfer between polymer chains and QDs. As a result, this contributes to the improvements of the Voc of hybrid solar cells. Moreover, the charge transfer enhancement and higher charge carrier mobility lead to an increase of Joc by 4.5% and 19.9% (table 5.1 and figure 5.6) measured by internal and ISE laboratory, respectively.

Figure 5.6 J-V graphs for the best CdSe QD/PCPDTBT and TrGO-CdSe QD/PCPDTBT solar cells measured under AM 1.5G illumination in internal laboratory (solid lines), and for comparison in the external laboratory 20 days later at ISE (dashed lines). (Reused from ref. [284]) Although, with the enhanced carrier mobility, a higher improvement of Joc can be expected, the achieved enhancement is relatively small. Tentatively, the higher carrier mobility is 91


compromised by other adverse effects caused by the presence of graphene such as the roughness of morphology in active layer and potential charge trapping. In fact, the solubility of CdSe QD-TrGO material in CB is worse than for CdSe QDs, leading to an unwell mixing of CdSe-TrGO and PCPDTBT polymer in CB, which partially contributes to a reduced junction area as consequence within the photoactive D/A layer. Table 5.1 Comparison of the results of the best solar cells measured in two laboratories. The active areas for the measured solar cells are of 0.0595 cm² for the CdSe QD/polymer and of 0.0517 cm² for the CdSe QD-TrGO/polymer device. In addition, the spectral mismatch for the AM 1.5G illumination was considered. Internal laboratory JSC [mA/cm²]

FF

External laboratory

VOC [V]

PCE [%]

JSC [mA/cm²]

FF

VOC [V]

PCE [%]

CdSe QD/ PCPDTBT

9.58

0.574

0.549

3.02

8.96

0.566

0.554

2.91

TrGO-CdSe QD/ PCPDTBT

10.02

0.575

0.713

4.12

10.74

0.548

0.721

4.24

Moreover, thiol functional groups within TrGO can act as doping agents, changing the energy levels of CdSe QDs. This is consistent with a previous report proving that thiol ligand exchange on CdSe QDs is shifting their energy levels without affecting the band gap size [252,253]. Here TrGO might also shift up the LUMO level of QDs, enlarging the offset to the HOMO level of the polymer donor, and therefore enhances the Voc of the resulting solar cells. To confirm this hypothesis, CV experiments were performed. 5.3.2 CV experiments The LUMO levels can approximately be obtained from the onset potential of the reduction waves according to following empirical formula [285]: ELUMO = - (Ered + 4.5), [eV]

(5.3)

where, ELUMO is the LUMO energy level of the measured materials used as working electrode, Ered is the reduction peak onset potential versus the potential of the normal hydrogen electrode (NHE). Based on Fig.5.7, the two values for Ered are extracted as -0.61 and -0.72 V for CdSe QDs only and CdSe QDs decorated TrGO, respectively. Applying these values of Ered to formula 5.3, the LUMO energy levels are obtained to be about - 3.76 and

92


Figure 5.7 CV spectra of CdSe QDs (red line) and QD-TrGO hybrid material (blue line) showing a reduction peak onset shift of CdSe QD-TrGO hybrid material compared to CdSe QDs. -3.87 eV for CdSe QDs and QD-TrGO, respectively. This means that thiol functionalities within TrGO induce a 0.11 eV lift of the CdSe QDs LUMO level. The energy shift can be attributed to electronic interactions or charge transfer between the materials. Thiol groups of TrGO might donate electrons to QDs during the formation of the hybrid material which is consistent with EPR results described in chapter 4 (4.3.5). The adding of electrons to QDs causes an increase of electron density in the QDs conduction band, leading to a shift of the LUMO level. The shifted-up LUMO level of CdSe QDs enlarges the difference between the LUMO level of the acceptor and the HOMO level of the donor leading to the increase in Voc of the solar cells. 5.3.3 TEM and TEM tomography The morphology of the active layer is one of the crucial factors deciding on the performance of hybrid solar cells. It forms junction between polymer donor and QDs acceptor, where charge transfer occurs, leading to the dissociation of photo-excitons into free electrons and holes. The QDs attached TrGO forms the electron percolation pathways for the cells. TEM and TEM tomography provides a deeper insight into the 3D structure of the active layers, visualizing the changes caused by the introduction of graphene compared to the QD only based devices. TEM and TEM tomography experimental details can be found in ref. [284].

93


Figure 5.8 (a): TEM images of the active layer of a CdSe/PCPDTBT (upper image) and of a CdSe QD-TrGO /PCPDTBT solar cell (lower image). The dark regions represent CdSe QDs and the bright regions the polymer phase. (b): X-z-cut through 3D reconstructions of the respective active layers obtained by TEM tomography. The bright regions correspond to the volume filled with CdSe QDs. The dimensions of the reconstructed volume were 150 nm parallel to the film (x- and y-direction) and 90 nm perpendiculars to the film (z-direction). (Re-used from ref. [284]) CdSe QDs and PCPDTBT polymer are both well dispersed in CB forming homogeneous solution resulting in a good mixture of the two materials. This maximizes the junction area between the QDs and polymer and increases the ability of interfacial charge separation processes. Figure 5.8a reveals the significant difference between active layers of CdSe QD/PCPDTBT and CdSe QD-TrGO/PCPDTBT. The latter shows a better interconnection of QDs in z-direction represented by the continuous dark regions, while the former consists of dark dots homogeneously distributed within the polymer phase (white areas) indicating no preferred direction for the charge-transporting pathway. The TEM tomography (Fig. 5.8b) provides a deeper inside into the 3D structure of the active layers. Interestingly, it shows that TrGO-CdSe QD nano-composites align along the z-direction forming good pathways for charge-transport to electrodes, whereby TrGO might improve additionally charge transfer and transport. The z-direction alignment of the nanocomposite material might also reduce the electron travelling pathway, which might result in a lower 94


charge recombination. This assumption was confirmed by Charge Extraction with Linearly Increasing Voltage (CELIV) experiments (see table 5.2 and 5.3). Table 5.2 Electron and hole mobilities extracted from CELIV measurements for CdSe QD/PCPDTBT and TrGO-CdSe QD/PCPDTBT solar cells. µe [cm²/Vs] 1.2  10-5

CdSe/PCPDTBT

8  10-5

2.3  10

-5

TrGO CdSe/PCPDTBT

µh [cm²/Vs]

The results shown in table 5.2 reveal that the electron mobility of a TrGO-based solar cell is by a factor of two higher than that of the reference cell, while hole mobilities are similar for both cells. This directly leads to the enhancement of Joc for the graphene-based hybrid solar cells. Furthermore, the presence of TrGO does not affect the hole mobility of the cells, which might demonstrate that only electrons are transferred to graphene, not holes. So, a better Table 5.3 List of: Extracted charges in dark (QD), under illumination (QL), calculated amount of additionally extracted charges due to illumination (QL-QD) and the ratio of these charges to charges extracted in dark (QL-QD) / QD of CdSe QD/polymer and CdSe QD-TrGO /polymer solar cells from CdSe QDs. 3

CdSe QD/ PCPDTBT TrGO-CdSe QD/ PCPDTBT

3

Extracted QD [e/cm ]

Extracted QL [e/cm ]

in dark

under illumination

1.97  1016 0.27  1016

3

(QL-QD) [e/cm ]

(QL-QD) / QD

5.00  1016

3.03  1016

1.54

1.12  1016

0.86  1016

3.19

charge separation can be expected, which is also an important factor for reducing the recombination of photo-electrons and holes. In addition to the higher electron mobility in TrGO containing cells, the CELIV measurements give information about charge extraction. It shows that fewer charges (86% fewer in dark and 78% fewer under illumination) are extracted from the TrGO containing cells (Table 5.3). 5.3.4 AFM AFM samples were prepared by dissolving the PEDOT:PSS layer in water and collecting the active layer on a glass substrate. The AFM measurements were performed as described in chapter 2 (2.2.3)

95

5.4 Conclusions

Figure 5.9 (a): AFM topographical images of the active layer surface of a CdSeQD /PCPDTBT (top) and of a CdSe QD-TrGO/PCPDTBT solar cell (bottom) recorded in tapping mode. (b): Representative extracted AFM height profiles of CdSe QD/PCPDTBT (top) and of CdSe QD-TrGO/PCPDTBT (bottom). (Adopted from ref. [284]]) The AFM topographical images (Fig. 5.9) show that the CdSe QD/PCPDTBT solar cells have a smoother surface with profile heights of about 5 nm, while that of TrGO-CdSe/PCPDTBT cells is about 20 nm. The coarser surface is obviously induced by graphene, which supports CdSe QDs. The surface roughness of CdSe QD-TrGO/PCPDTBT seems to be beneficial for better contacting with Al electrode enhancing the long-term stability of hybrid solar cells.

5.4 Conclusions Graphene has been successfully incorporated into hybrid solar cells by using CdSe QD-TrGO hybrid materials. TrGO changes the morphology of the active layer, in which CdSe QDs bind to graphene, resulting in an alignment of the hybrid material along z-direction between the two electrodes. This enables a shorter electron percolation pathway to the Al electrode, enhancing charge transport and leads to a higher Joc. The electron mobility of graphene-based cell is double to that of the reference cell, while the hole mobility remains unchanged. This means only electrons are transferred to graphene rather than holes, due to the favorable band energy offset. This selective charge transfer improves the charge separation and reduces the charge recombination. The better charge transfer and the up-shift of the LUMO level of CdSe 96

5.4 Conclusions

QDs caused by thiol-doping both contribute to a significant enhancement of the cell Voc. The roughness of the active layer surface allows the formation of a good electrical contact with the Al cathode, reducing the shunt resistance and improving the overall solar cell stability. Based on the introduction of graphene, the achieved cells demonstrated a PCE up to 4.2% which is amongst the highest values for state-of-the-art hybrid solar cells. In addition, the CdSe QD contents could be reduced in the graphene based cells. It is noteworthy that this was just an initial attempt for introducing of graphene into hybrid solar cells. The main target was to demonstrate the proof of concept of the graphene-based cells. There are still plenty of aspects to be investigated for improvement of the cell performance. The lateral size of graphene flakes can be optimized, as well as the degree of thiol-functionalization which can be tuned to improve the solubility of CdSe QD-TrGO hybrid nano-composites. The work function of graphene can also be varied and adjusted by tuning its reduction degree or by additional doping agents, to improve the matching with the LUMO level of CdSe QDs.

97

6.1 Motivation

Chapter 6 Application of CdSe-rGO hybrid materials for PDs (This work was performed in corporation with the group of Prof.Margit Zacharias, Institute of Microsystems and Engineering, IMTEK, University of Freiburg)

6.1 Motivation State-of-the-art applications of graphene in PDs were summarized in Chapter 1 (see 1.3.5). There, graphene has been demonstrated either as a promising light-absorber or transport material for highly sensitive, fast responding and broad wavelength detection PDs. The lightabsorption limitation of graphene can be generally improved by several strategies such as the integrations of plasmonic nanostructures [173–175], microcavities [176,177], or silicon waveguides to graphene. Impressively a monolayer of CVD graphene was engineered into quantum dot-like array structures to introduce midgap states which can serve as electron trapping centers. By that, significant results were achieved for graphene PDs with a photoreponsivity up to 8.61 A W-1 with a high gain of 120 [179]. Another strategy is to hybridize graphene with efficient light-absorber materials such as QDs. Thus, an ultrahigh photoconductive gain of 108 electrons per photon and a great responsivity of 107 were obtained in a hybrid PD that consists of monolayer or bilayer graphene covered with a thin film of CdS QDs [180]. In that report, graphene was synthesized by CVD. Alternatively, here chemically synthesized graphene flakes are investigated for potential PD applications. The graphene flakes were decorated with CdSe QDs, and used to fabricate micro-sized PDs to exploit the synergistic effect of the excellent light-absorption of CdSe QDs and the high electrical mobility of graphene.

6.2 Experiments 6.2.1 rGO preparation For rGO preparation, 50 μl of 0.05 wt% GO aqueous solution was spin coated onto SiO2 substrates at two successive speeds of 500 rpm for 1 min and 2500 rpm for 2 min. To increase its conductivity, GO was then reduced into rGO by placing the GO deposited silicon substrate in a hydrazine vapor atmosphere at 90 C for 24h (for details, see 2.4.1b).

98

6. 3 Results and discussions

6.2.2 Photoconductor fabrication The rGO deposited SiO2 substrates were used to fabricate PDs by a photolithographic process. Firstly the samples were spin-coated with photoresist to form 1.5 μm thick layer of photoresist AZ5214e. Then they were annealed on a hotplate at 110 C for 50 s to evaporate solvent from the photoresist layer. Contact patterns were established by exposing the photoresist layer to a diode laser at 405 nm wavelength in reversal mode, using a micropattern generator (μPG 101, Heidenberg). Subsequently, the photoresist was heated at 150 C for 2 min, followed by a flood exposure to UV light for 20 s. Next, the samples were developed in AZ 726 MIF developer; as a result, contact patterns were obtained in the photoresist layer. Afterwards, successive layers of 5 nm Ti and 100 nm gold were deposited by a home built evaporator in a vacuum of 5.10-6 bar at a deposition rate of 0.4 nm/s. Finally a lift-off step was performed in acetone solution to get the final two-probe contacts, fourprobe contacts, or interdigital electrode structures, respectively. The PDs were fabricated by spin-coating 50μl of CdSe QDs in chloroform (0.5 mg/ml) on the surface of the two-probe contact (named sample Ph2) (Fig. 6.2) at 2000 rpm for 2 min. The CdSe QDs solution were prepared according to the same procedure presented in 3.1.3, followed by a washing step, as described in 3.1.4. Then they were dissolved in chloroform forming CdSe QDs solution (0.5 mg/ml) for the spin-coating step. Afterwards, the samples were annealed on a hotplate in a nitrogen-filled glove box at 100 C for 10 min to evacuate the excess solvents. 6.2.3 Electrical measurements I-V characterizations were performed using a microprobe station that is connected to a device analyzer (Agilent B1500A). The microprobe station is placed in a housing box enabling the measurements to be performed in dark and under illumination. After the microprobes were carefully connected to the contact probes of the samples, the potential bias was scanned from -1 to + 1 V. I-V characteristic data were extracted by the device analyzer.

6. 3 Results and discussions 6.3.1 Contact probe fabrication The rGO flakes are visible under an OM (Fig. 6.1). This enables to manually select each rGO flake and locate the micro-contact pattern on its center with suitable direction of the pattern to fit the shape of the rGO flakes. 99


Figure 6.1 shows an rGO flake with four aluminum contact probes for electrical experiments (sample Ph1). However, the work function mismatch of rGO and Al induces Schottky barrier between them, hindering to establish a good electrical contact (Fig. 6.4a). For a better contact and avoiding a Schottky barrier, Ti-Au contact material was used (sample Ph2). Therefore, rGO was constructed with two probe contacts which are fabricated by depositing a 5-nm thick Ti followed by a 100-nm thick Au layer (Fig. 6.2). In another attempt, interdigital structured contacts were established directly on a SiO2 substrate, and then CdSe QD-TrGO hybrid material was drop-casted onto the interdigital contact electrodes (Fig. 6.3).

Figure 6.1 An rGO flake constructed with four point probe contacts for electrical measurements. The width of the contact fingers is 2 μm and distance between two fingers is 5 μm (sample Ph1).

Figure 6.2 An rGO flake with two probe contacts separated by 2 μm (sample Ph2).

100


Figure 6.3 Digital electrode structure for contacting QD-rGO flakes (sample Ph3).

Figure 6.4 I-V characteristics of rGO flakes with (a) aluminum contact pads, (b) Ti/gold contact pads measured for sample Ph1. Figure 6.4a reveals I-V characteristics of a rGO flake electrically connected by aluminum contact pads that shows a Schottky barrier between graphene and aluminum. The contact barrier is induced by the work function mismatch between rGO and Al, and interfacial insulating Al-oxide formation. The schottky barrier not only increases the contact resistance dramatically, resulting in a low current, it is also responsible for the non-ohmic I-V characteristic shown in Fig. 4a. In contrast, with Ti/Au contacts, an Ohmic contact is obtained as shown in Fig. 6.4b, leading to a linear I-V characteristic. Consequently, the current 101


increases to 4.10-6 A at 1 V which is two orders of magnitude higher than that for the sample with the Al contact probe. 6.3.2 Photoconductor measurements The sample preparation was described in 6.2.2. Sample Ph2 was covered with CdS QDs by spin-coating. A following thermal annealing step at 100 C for 10 min helped to improve the contact between QDs and graphene. Afterwards, electrical measurements were performed under illumination and in the dark. Figure 6.5 schematically shows the working principle of the photoconductor. Under illumination, electrons within QDs are excited by photons and excitons are generated. Charge separation can occur at the graphene-QD interface, changing the charge density in the graphene channel, resulting in a change of the electrical conductivity.

Figure 6.5 Schematic illustration of the working principle for a QDs-graphene photoconductor. The two red lines in Figure 6.6 reveal the I-V characteristics of the sample Ph2, which are extracted from electrical measurements under illumination. They have a smaller slope than the black line which was measured in the dark. When the photoconductor was illuminated by white light from a lamp, the conductivity of the CdSe-rGO channel is decreased, leading to the photo-response current of the photoconductor, as shown in Figure 6.6. The decrease of conductivity can be explained by a photo-induced electron transfer from QDs to graphene causing the decrease of a positive charge carrier density present in rGO. This means that the rGO channel was originally in a p-doped state and is therefore p-type conductive. This charge transfer results are consistent with the PL quenching experiments and EPR quenching presented in Chapter 3 (3.1.4.2) and Chapter 4 (4.3.4), respectively.

102


Figure 6.6 Photo-response current of a CdSe QD-rGO based photoconductor device with two probe contacts (sample Ph2) under illumination (red line) and in the dark (black line). For a better photo-responsivity, the CdSe QD-rGO PD should be modulated by applying an external gate potential and tuned into an n-doped state by an external electric field. By this, the electron density of the channel can be adjusted to a lower value. Therefore, the transfer of photo-induced electrons will significantly change the conductivity of the rGO channel. As a result, a higher photo-responsivity can be achieved. The main purpose of these results is to demonstrate the kind of charge transfer occurring between QDs and graphene. For further PD application, this preliminary result needs to be extended by more experiments to develop an optimized photo-responsivity. However, this is beyond the aim of this thesis.

103

Summary and outlook

Chapter 7 Summary and outlook 7.1 Summary Single-layer graphene oxide flakes have been exfoliated successfully by a modified Hummers method. The thickness of single-layer GO flakes is 0.7 nm measured by AMF, and the lateral size distribution of GO flakes varies from several hundreds of nanometers to less than 10 µm. Oxygen-containing functional groups were introduced into graphene lattices by oxidizing graphene into GO with strong oxidant mixtures of NaNO3, KMnO4 and H2SO4. GO becomes hydrophilic and well dissolvable in water forming homogeneous aqueous solutions. FTIR, XPS results demonstrate that a significant amount of oxygen has been introduced into graphene by this oxidation method. GO is an intermediate in the synthesis of chemically synthesized graphene. GO itself is limited for applications due to its electrical insulation. GO is only soluble in polar solvents that is an additional limitation. For the incorporation of graphene into hybrid solar cells, its conductivity and solubility in non-polar CB become key features. To this end, GO was transformed into thiol-functionalized reduced graphene oxide (TrGO). The thiolfunctionalization and chemical reduction have been performed simultaneously by refluxing GO with P4S4 in DMF. The oxygen-containing functionalities were removed from GO and partially transformed into thiol groups. Based on that, the conductivity increases remarkably to the value of up to 5300 Sm-1, while a significant amount of thiol-functional groups also were introduced as proven by XPS and FTIR. With respect to the reduction of GO, the thionation and simultaneous reduction method is advanced because of its simplicity, lowtemperature reaction, low-cost and less toxicity compared to the common hydrazine reduction method. In terms of functionalization of graphene, the under-studied method is also superior, because the thiol groups are directly bound to the carbon lattice of graphene. Meanwhile the common approach for functionalization of graphene is to use functional species such as benzyl capable of forming non-covalent π-π attachments to the graphene lattice. The degree of reduction and functionalization can be adjusted to achieve TrGO with: being dissolvable in desired polar or non-polar solvents, various electrical conductivities and work functions. EPR results show that Mn+2 hyperfine signals disappear after GO was reduced and transformed

104

7.1 Summary

into TrGO. This is because the Mn+2 remaining within GO from synthesis step had been removed by the thionation and reduction process. TrGO was then decorated with CdSe QDs by a self-assembly process to form CdSe-TrGO hybrid material. This is a general and advanced approach to NP-graphene hybrid materials compared to conventional in-situ growth which is often used to synthesize similar hybrid materials. The size and quality of CdSe QDs is easily controllable by a separate synthesis and not disturbed by the presence of graphene during CdSe QD growth. TEM investigation shows that a high loading of QDs on graphene flakes was achieved. This is due to the great number of thiol groups which were introduced into the graphene lattices and their strong affinity to CdSe QDs. From EPR experiments, we deduced that chemical attachment is occurring between QDs and thiol groups. So, photo-induced electrons from QDs can be transferred to TrGO, coupling with the single electron spins within TrGO. Consequently, a complete quenching of the EPR signal from TrGO was observed after QD attachment. The intimate coupling of QDs and TrGO facilitates the charge transfer and therefore favors the use of the hybrid materials in optoelectronic applications such as hybrid solar cells and PDs. With graphene being incorporated into hybrid solar cells they exhibited a significant improvement in Voc leading to a significant PCE enhancement by  46% to the value of 4.2%. This is amongst the highest PCE values for hybrid solar cells reported so far. The successful application of graphene in hybrid solar cell is resulting from several factors which are the following: Graphene has chemically been modified to obtain thiol-functionalization, while the conductivity is dramatically increased. The solubility of TrGO is better in DMF than in CB (CB), but interestingly after being decorated with CdSe QDs, the hybrid material becomes well dissolvable in CB, the most suitable solvent for spin coating of the active layer. This enables CdSe QD-TrGO being mixable with the semiconducting polymer forming a homogeneous solution in CB, and therefore enables the deposition of the active layer by a simple and cheap solution-based process. PDs were fabricated using CdSe QD decorated rGO as light absorber channel. Preliminary results proved a photo-induced electron transfer from QDs to graphene leading to a photoresponsivity of the photoconductor. This result is consistent with PL quenching experiments and the quenching of EPR signals for CdSe QD-TrGO hybrid materials. However, the photoresponsivity is low, and therefore more work has to devote to improve its performance.

105

7.2 Outlook

7.2 Outlook As mentioned in section 5.4, the performance of graphene-based hybrid solar cells has high potential to be further improved. By tuning the degree of the functionalization and the reduction state, the work function of TrGO can be engineered. Thus, the charge transfer might be further improved between either the QDs or polymer and graphene. The optimal situation would be achieved when photo-induced electrons from polymer and QDs are transferred to graphene, but no holes, so that charge recombination is reduced. To implement this, the work function of TrGO should be adjusted to a value right below the LUMO level of CdSe QDs. It can experimentally be done by adjusting the TrGO synthesis conditions such as time and temperature reaction, in combination with work function measurements by Kelvin probe. Graphene has purposely been functionalized with thiol groups to attach CdSe QDs. This also means that graphene has successfully been doped with sulfur. Previous report demonstrated that sulfur-doped graphene showed a highly catalytic ability towards oxygen reduction reaction (ORR), even better than that of Pt for applications as catalytic cathode in fuel cells [184]. The thionation method described in this thesis offers a low temperature synthesis of sulfur-doped graphene, and therefore a potential low-cost production pathway compared to the commonly described doping methods which are often performed at high temperature (600-1050 C) [184]. TrGO and rGO can also be decorated with ZnO NPs forming ZnO NP-TrGO and ZnO NPrGO hybrid materials, respectively by applying self-assembly decoration. These hybrid materials have been demonstrated promising for many applications such as photocatalysis [247], gas-sensors [248] and solar cells [249].

106

Abbreviations

Abbreviations Abbreviations Description 0D

zero-dimensional

1D

one-dimensional

2D

two-dimensional

3D

three-dimensional

AFM

atomic force microscopy

Ag NP-rGO

Silver nanoparticle-reduced graphene oxide hybrid material

Ag

silver

Al

aluminum

AM1.5G

air mass 1.5 global

CB

chlorobenzene

CdSe-TrGO

CdSe quantum dot –thiolated reduced graphene oxide hybrid material

CdSe-rGO

CdSe quantum dot –reduced graphene oxide hybrid material

CELIV

Charge extraction by linear increase of voltage

CNT

carbon nanotube

CVD

chemical vapor deposition

D/A

donor/acceptor

DSSC

dye sensitized solar cell

EG

epitaxial growth

EPR

electron paramagnetic resonance

EQE

external quantum efficiency

FC

fuel cell

FF

fill factor

FTCs

flexible transparent conductors

FTIR

fourier transform infrared spectroscopy

FWHM

full width at half maximum

GO

graphene oxide

HDA

hexadecylamine

HOMO

highest occupied molecular orbital

IQE

internal quantum efficiency

ITO

indium tin oxide

Jsc

short-circuit current density 107

Abbreviations

J-V

current density-voltage

LED

light-emitting diode

LiB

lithium ion battery

LUMO

lowest unoccupied molecular orbital

ME

mechanical exfoliation

NC

nanocrystal

NP

nanoparticle

NR

nanorods

NREL

National Renewable Energy Laboratory

OLED

organic light-emitting diode

OSC

organic solar cell

P3HT

poly(3-hexylthiophene)

PCBM

[6,6]-phenyl C61 butyric acid methyl ester

PCE

power conversion efficiency

PCPDTBT

Poly[2,6-(4,4-bis-(2-ethylhexyl)-4H-cyclopenta[2,1-b;3,4-b'] -dithiophene)-alt-4,7-(2,1,3-benzothiadiazole)]

PDs

photodetectors

PEDOT:PSS

poly(3,4-alklenedioxythiophenes):poly(styrenesulfonate)

PL

photoluminescence

PV

Photovoltaic

QD

quantum dot

rGO

reduced graphene oxide

rpm

revolutions per minute

SEM

scanning electron microscopy

SC

super capacitor

TC

transparent conductor

TEM

transmission electron microscopy

TOP

trioctylphosphine

TOPO

trioctylphosphine oxide

TrGO

Thiol-functionalized reduced graphene oxide

Voc

open-circuit voltage

XPS

X-ray photoelectron spectroscopy

∆Bpp

peak-to-peak widths

108

Acknowledgements

Acknowledgements First of all, I would like to thank my supervisor PD. Dr. Michael Krüger for the chance to join his Laboratory for Nanosciences at FMF (Freiburg Materials Research Center) as PhD student. His tireless guidance and support contributed the most in shaping me into the researcher I am today. I also thank Prof. Dr. Stefan Weber for being the second referee of my thesis. A special thank you goes to Dr. Michael Eck for his great supports in my research and also in the life from the day one that I came to Freiburg. I thank Dr. Frank S. Riehle for his fruitful scientific discussions in quantum dots and Dr. Gregory B. Stevens for his guidance in laboratory working skills. Moreover I would like to acknowledge all other group members: Simon Einwächter, Yin Yuan, and Alfian F. Madsuha for sharing a pleasant working time and life with me. Amongst my collaborations, I acknowledge Dr. Emre Erdem for his crucial contribution in electron paramagnetic resonance part of my thesis, described in chapter 4. His guidance and discussion were sources of inspiration and foundation of my work in this chapter. I would like to express my special thanks to Prof. Dr. Margit Zacharias and Dr. Andreas Menzel for the supports and collaboration in device microfabrication, and allowing me to perform electrical measurements in their laboratory, leading to the results presented in chapter 6. I also acknowledge Dr. Ralf Thomann for performing TEM experiments and Dr. Yi Thomann for introducing me to AFM. In addition, I thank Dr. C. Haas for XPS measurements. Finally I want to express my deepest gratitude to my wife Tran T. K. Ngan for unconditionally supporting me with love, patience and trust, and lonely taking care of our children. I gratefully acknowledge the Vietnam International Education Development (VIED) for financial support.

109

References

References [1] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, I.V. Grigorieva, A.A. Firsov, Electric field effect in atomically thin carbon films, Science 306 (2004) 666–669. [2] H. Brody, Graphene, Nature 483 (2012) S29. [3] A.K. Geim, K.S. Novoselov, The rise of graphene, Nat. Mater. 6 (2007) 183–191. [4] B. Partoens, F. Peeters, From graphene to graphite: Electronic structure around the K point, Phys. Rev. B 74 (2006) 075404. [5] S. Morozov, K. Novoselov, F. Schedin, D. Jiang, A. Firsov, A. Geim, Two-dimensional electron and hole gases at the surface of graphite, Phys. Rev. B 72 (2005) 201401. [6] L. D. Landau, Zur Theorie der phasenumwandlungen II, Phys. Z. Sowjetunion 11 (1937) 26–35. [7] R. Peierls, Quelques proprietes typiques des corpses solides, Ann. I. H. Poincare 5 (1935) 177–222. [8] P. Wallace, The band theory of graphite, Phys. Rev. 71 (1947) 622–634. [9] N. Savage, Materials science: Super carbon, Nature 483 (2012) S30. [10] J.-C. Charlier, P.C. Eklund, J. Zhu, A.C. Ferrari, Electron and phonon properties of graphene: their relationship with carbon nanotubes, in: C.E. Ascheron, A.H. Duhm, A. Jorio, G. Dresselhaus, M.S. Dresselhaus (Eds.), Carbon Nanotubes, Springer Berlin Heidelberg, Berlin, Heidelberg, 2008, pp. 673–709. [11] A. H. Castro Neto, N. M. R. Peres, K.S. Novoselov, A.K. Geim, The electronic properties of graphene, Rev. Mod. Phys. 81 (2009) 109–162. [12] A.S. Mayorov, R.V. Gorbachev, S.V. Morozov, L. Britnell, R. Jalil, L.A. Ponomarenko, P. Blake, K.S. Novoselov, K. Watanabe, T. Taniguchi, A.K. Geim, Micrometer-scale ballistic transport in encapsulated graphene at room temperature, Nano Lett. 11 (2011) 2396–2399. [13] N.O. Weiss, H. Zhou, L. Liao, Y. Liu, S. Jiang, Y. Huang, X. Duan, Graphene: an emerging electronic material, Adv. Mater. Weinheim 24 (2012) 5782–5825. [14] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, M.I. Katsnelson, I.V. Grigorieva, S.V. Dubonos, A.A. Firsov, Two-dimensional gas of massless Dirac fermions in graphene, Nature 438 (2005) 197–200. [15] Y. Zhang, Y.-W. Tan, H.L. Stormer, P. Kim, Experimental observation of the quantum Hall effect and Berry's phase in graphene, Nature 438 (2005) 201–204. [16] E.V. Castro, H. Ochoa, M.I. Katsnelson, R.V. Gorbachev, D.C. Elias, K.S. Novoselov, A.K. Geim, F. Guinea, Limits on charge carrier mobility in suspended graphene due to flexural phonons, Phys. Rev. Lett. 105 (2010) 266601. [17] K.I. Bolotin, K.J. Sikes, J. Hone, H.L. Stormer, P. Kim, Temperature-dependent transport in suspended graphene, Phys. Rev. Lett. 101 (2008) 096802. [18] X. Du, I. Skachko, A. Barker, E.Y. Andrei, Approaching ballistic transport in suspended graphene, Nat. Nanotechnol. 3 (2008) 491–495. [19] C.R. Dean, A.F. Young, I. Meric, C. Lee, L. Wang, S. Sorgenfrei, K. Watanabe, T. Taniguchi, P. Kim, K.L. Shepard, J. Hone, Boron nitride substrates for high-quality graphene electronics, Nat. Nanotechnol. 5 (2010) 722–726. [20] S.-Y. Chen, P.-H. Ho, R.-J. Shiue, C.-W. Chen, W.-H. Wang, Transport/magnetotransport of high-performance graphene transistors on organic molecule-functionalized substrates, Nano Lett. 12 (2012) 964–969. [21] J.-H. Chen, C. Jang, S. Xiao, M. Ishigami, M.S. Fuhrer, Intrinsic and extrinsic performance limits of graphene devices on SiO2, Nat. Nanotechnol. 3 (2008) 206–209. 110

References

[22] L. Liao, J. Bai, Y. Qu, Y.-c. Lin, Y. Li, Y. Huang, X. Duan, High-kappa oxide nanoribbons as gate dielectrics for high mobility top-gated graphene transistors, Proc. Natl. Acad. Sci. U.S.A. 107 (2010) 6711–6715. [23] N. Petrone, C.R. Dean, I. Meric, van der Zande, Arend M, P.Y. Huang, L. Wang, D. Muller, K.L. Shepard, J. Hone, Chemical vapor deposition-derived graphene with electrical performance of exfoliated graphene, Nano Lett. 12 (2012) 2751–2756. [24] C. Berger, Z. Song, X. Li, X. Wu, N. Brown, C. Naud, D. Mayou, T. Li, J. Hass, A.N. Marchenkov, E.H. Conrad, P.N. First, de Heer, Walt A, Electronic confinement and coherence in patterned epitaxial graphene, Science 312 (2006) 1191–1196. [25] K.V. Emtsev, A. Bostwick, K. Horn, J. Jobst, G.L. Kellogg, L. Ley, J.L. McChesney, T. Ohta, S.A. Reshanov, J. Röhrl, E. Rotenberg, A.K. Schmid, D. Waldmann, H.B. Weber, T. Seyller, Towards wafer-size graphene layers by atmospheric pressure graphitization of silicon carbide, Nat. Mater. 8 (2009) 203–207. [26] J.A. Robinson, M. Wetherington, J.L. Tedesco, P.M. Campbell, X. Weng, J. Stitt, M.A. Fanton, E. Frantz, D. Snyder, B.L. VanMil, G.G. Jernigan, R.L. Myers-Ward, C.R. Eddy, D.K. Gaskill, Correlating Raman spectral signatures with carrier mobility in epitaxial graphene: a guide to achieving high mobility on the wafer scale, Nano Lett. 9 (2009) 2873–2876. [27] G. Jo, S.-I. Na, S.-H. Oh, S. Lee, T.-S. Kim, G. Wang, M. Choe, W. Park, J. Yoon, D.-Y. Kim, Y.H. Kahng, T. Lee, Tuning of a graphene-electrode work function to enhance the efficiency of organic bulk heterojunction photovoltaic cells with an inverted structure, Appl. Phys. Lett. 97 (2010) 213301. [28] M. Sprinkle, M. Ruan, Y. Hu, J. Hankinson, M. Rubio-Roy, B. Zhang, X. Wu, C. Berger, de Heer, W A, Scalable templated growth of graphene nanoribbons on SiC, Nat. Nanotechnol. 5 (2010) 727–731. [29] C. Casiraghi, A. Hartschuh, E. Lidorikis, H. Qian, H. Harutyunyan, T. Gokus, K.S. Novoselov, A.C. Ferrari, Rayleigh imaging of graphene and graphene layers, Nano Lett. 7 (2007) 2711–2717. [30] A. Kuzmenko, E. van Heumen, F. Carbone, van der Marel, Universal optical conductance of graphite, Phys. Rev. Lett. 100 (2008) 117401. [31] R.R. Nair, P. Blake, A.N. Grigorenko, K.S. Novoselov, T.J. Booth, T. Stauber, N. M. R. Peres, A.K. Geim, Fine structure constant defines visual transparency of graphene, Science 320 (2008) 1308. [32] K.J. Tielrooij, Song, J. C. W., S.A. Jensen, A. Centeno, A. Pesquera, A. Zurutuza Elorza, M. Bonn, L.S. Levitov, F. H. L. Koppens, Photoexcitation cascade and multiple hotcarrier generation in graphene, Nat. Phys. 9 (2013) 248–252. [33] G. van Lier, C. van Alsenoy, V. van Doren, P. Geerlings, Ab initio study of the elastic properties of single-walled carbon nanotubes and graphene, Chem. Phys. Lett. 326 (2000) 181–185. [34] C.D. Reddy, S. Rajendran, K.M. Liew, Equilibrium configuration and continuum elastic properties of finite sized graphene, Nanotechnology 17 (2006) 864–870. [35] I.W. Frank, D.M. Tanenbaum, van der Zande, A. M., P.L. McEuen, Mechanical properties of suspended graphene sheets, J. Vac. Sci. Technol. B 25 (2007) 2558. [36] M. Poot, H. S. J. van der Zant, Nanomechanical properties of few-layer graphene membranes, Appl. Phys. Lett. 92 (2008) 063111. [37] D.A. Dikin, S. Stankovich, E.J. Zimney, R.D. Piner, Dommett, H. B. Geoffrey, G. Evmenenko, S.T. Nguyen, R.S. Ruoff, Preparation and characterization of graphene oxide paper, Nature 448 (2007) 457–460.

111

References

[38] S. Park, K.-S. Lee, G. Bozoklu, W. Cai, S.T. Nguyen, R.S. Ruoff, Graphene oxide papers modified by divalent ions-enhancing mechanical properties via chemical crosslinking, ACS Nano 2 (2008) 572–578. [39] S. Park, D.A. Dikin, S.T. Nguyen, R.S. Ruoff, Graphene oxide sheets chemically crosslinked by polyallylamine, J. Phys. Chem. C 113 (2009) 15801–15804. [40] S. Berber, Y.-K. Kwon, D. Tománek, Unusually high thermal conductivity of carbon nanotubes, Phys. Rev. Lett. 84 (2000) 4613–4616. [41] A.A. Balandin, S. Ghosh, W. Bao, I. Calizo, D. Teweldebrhan, F. Miao, C.N. Lau, Superior thermal conductivity of single-layer graphene, Nano Lett. 8 (2008) 902–907. [42] W. Cai, A.L. Moore, Y. Zhu, X. Li, S. Chen, L. Shi, R.S. Ruoff, Thermal transport in suspended and supported monolayer graphene grown by chemical vapor deposition, Nano Lett. 10 (2010) 1645–1651. [43] D. Nika, E. Pokatilov, A. Askerov, A. Balandin, Phonon thermal conduction in graphene: Role of Umklapp and edge roughness scattering, Phys. Rev. B 79 (2009) 155413. [44] R. van Noorden, Production: Beyond sticky tape, Nature 483 (2012) S32-3. [45] X. Lu, M. Yu, H. Huang, R.S. Ruoff, Tailoring graphite with the goal of achieving single sheets, Nanotechnology 10 (1999) 269–272. [46] C. Berger, Z. Song, T. Li, X. Li, A.Y. Ogbazghi, R. Feng, Z. Dai, A.N. Marchenkov, E.H. Conrad, P.N. First, de Heer, A. Walt, Ultrathin epitaxial graphite: 2D electron gas properties and a route toward graphene-based nanoelectronics, J. Phys. Chem. B 108 (2004) 19912–19916. [47] J. Robinson, X. Weng, K. Trumbull, R. Cavalero, M. Wetherington, E. Frantz, M. Labella, Z. Hughes, M. Fanton, D. Snyder, Nucleation of epitaxial graphene on SiC(0001), ACS Nano 4 (2010) 153–158. [48] J.K. Hite, M.E. Twigg, J.L. Tedesco, A.L. Friedman, R.L. Myers-Ward, C.R. Eddy, D.K. Gaskill, Epitaxial graphene nucleation on C-face silicon carbide, Nano Lett. 11 (2011) 1190–1194. [49] S. Shivaraman, R.A. Barton, X. Yu, J. Alden, L. Herman, M. Chandrashekhar, J. Park, P.L. McEuen, J.M. Parpia, H.G. Craighead, M.G. Spencer, Free-standing epitaxial graphene, Nano Lett. 9 (2009) 3100–3105. [50] J. Hass, de Heer, W A, E.H. Conrad, The growth and morphology of epitaxial multilayer graphene, J. Phys.: Condens. Matter 20 (2008) 323202. [51] J.D. Caldwell, T.J. Anderson, J.C. Culbertson, G.G. Jernigan, K.D. Hobart, F.J. Kub, M.J. Tadjer, J.L. Tedesco, J.K. Hite, M.A. Mastro, R.L. Myers-Ward, C.R. Eddy, P.M. Campbell, D.K. Gaskill, Technique for the dry transfer of epitaxial graphene onto arbitrary substrates, ACS Nano 4 (2010) 1108–1114. [52] S. Bae, H. Kim, Y. Lee, X. Xu, J.-S. Park, Y. Zheng, J. Balakrishnan, T. Lei, H. Ri Kim, Y.I. Song, Y.-J. Kim, K.S. Kim, B. Özyilmaz, J.-H. Ahn, B.H. Hong, S. Iijima, Roll-toroll production of 30-inch graphene films for transparent electrodes, Nature Nanotech. 5 (2010) 574–578. [53] J.-H. Lee, E.K. Lee, W.-J. Joo, Y. Jang, B.-S. Kim, J.Y. Lim, S.-H. Choi, S.J. Ahn, J.R. Ahn, M.-H. Park, C.-W. Yang, B.L. Choi, S.-W. Hwang, D. Whang, Wafer-scale growth of single-crystal monolayer graphene on reusable hydrogen-terminated germanium, Science 344 (2014) 286–289. [54] Q. Yu, J. Lian, S. Siriponglert, H. Li, Y.P. Chen, S.-S. Pei, Graphene segregated on Ni surfaces and transferred to insulators, Appl. Phys. Lett. 93 (2008) 113103. [55] K.S. Kim, Y. Zhao, H. Jang, S.Y. Lee, J.M. Kim, K.S. Kim, J.-H. Ahn, P. Kim, J.-Y. Choi, B.H. Hong, Large-scale pattern growth of graphene films for stretchable transparent electrodes, Nature 457 (2009) 706–710. 112

References

[56] X. Li, W. Cai, L. Colombo, R.S. Ruoff, Evolution of graphene growth on Ni and Cu by carbon isotope labeling, Nano Lett. 9 (2009) 4268–4272. [57] X. Li, W. Cai, J. An, S. Kim, J. Nah, D. Yang, R. Piner, A. Velamakanni, I. Jung, E. Tutuc, S.K. Banerjee, L. Colombo, R.S. Ruoff, Large-area synthesis of high-quality and uniform graphene films on copper foils, Science 324 (2009) 1312–1314. [58] Z. Li, P. Wu, C. Wang, X. Fan, W. Zhang, X. Zhai, C. Zeng, Z. Li, J. Yang, J. Hou, Low-temperature growth of graphene by chemical vapor deposition using solid and liquid carbon sources, ACS Nano 5 (2011) 3385–3390. [59] R. Vitchev, A. Malesevic, R.H. Petrov, R. Kemps, M. Mertens, A. Vanhulsel, C. van Haesendonck, Initial stages of few-layer graphene growth by microwave plasmaenhanced chemical vapour deposition, Nanotechnology 21 (2010) 095602. [60] J.B. Park, W. Xiong, Y. Gao, M. Qian, Z.Q. Xie, M. Mitchell, Y.S. Zhou, G.H. Han, L. Jiang, Y.F. Lu, Fast growth of graphene patterns by laser direct writing, Appl. Phys. Lett. 98 (2011) 123109. [61] B. Guo, Q. Liu, E. Chen, H. Zhu, L. Fang, J.R. Gong, Controllable N-doping of graphene, Nano Lett. 10 (2010) 4975–4980. [62] G. Ruan, Z. Sun, Z. Peng, J.M. Tour, Growth of graphene from food, insects, and waste, ACS Nano 5 (2011) 7601–7607. [63] Z. Sun, Z. Yan, J. Yao, E. Beitler, Y. Zhu, J.M. Tour, Growth of graphene from solid carbon sources, Nature 468 (2010) 549–552. [64] Y. Lee, S. Bae, H. Jang, S. Jang, S.-E. Zhu, S.H. Sim, Y.I. Song, B.H. Hong, J.-H. Ahn, Wafer-scale synthesis and transfer of graphene films, Nano Lett. 10 (2010) 490–493. [65] X. Li, Y. Zhu, W. Cai, M. Borysiak, B. Han, D. Chen, R.D. Piner, L. Colombo, R.S. Ruoff, Transfer of large-area graphene films for high-performance transparent conductive electrodes, Nano Lett. 9 (2009) 4359–4363. [66] L. Staudenmaier, Verfahren zur Darstellung der Graphitsäure, Ber. Dtsch. Chem. Ges. 31 (1898) 1481–1487. [67] W.S. Hummers, R.E. Offeman, Preparation of graphitic oxide, J. Am. Chem. Soc. 80 (1958) 1339. [68] C. Gómez-Navarro, R.T. Weitz, A.M. Bittner, M. Scolari, A. Mews, M. Burghard, K. Kern, Electronic transport properties of individual chemically reduced graphene oxide sheets, Nano letters 7 (2007) 3499–3503. [69] S. Gilje, S. Han, M. Wang, K.L. Wang, R.B. Kaner, A chemical route to graphene for device applications, Nano Lett. 7 (2007) 3394–3398. [70] S. Stankovich, D.A. Dikin, R.D. Piner, K.A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu, S.T. Nguyen, R.S. Ruoff, Synthesis of graphene-based nanosheets via chemical reduction of exfoliated graphite oxide, Carbon 45 (2007) 1558–1565. [71] Y. Si, E.T. Samulski, Synthesis of water soluble graphene, Nano Lett. 8 (2008) 1679– 1682. [72] R. Muszynski, B. Seger, P.V. Kamat, Decorating graphene sheets with gold nanoparticles, J. Phys. Chem. C 112 (2008) 5263–5266. [73] G. Wang, J. Yang, J. Park, X. Gou, B. Wang, H. Liu, J. Yao, Facile synthesis and characterization of graphene nanosheets, J. Phys. Chem. C 112 (2008) 8192–8195. [74] C.V. Pham, M. Eck, M. Krueger, Thiol functionalized reduced graphene oxide as a base material for novel graphene-nanoparticle hybrid composites, Chem. Eng. J. 231 (2013) 146–154. [75] T. Zhou, F. Chen, K. Liu, H. Deng, Q. Zhang, J. Feng, Q. Fu, A simple and efficient method to prepare graphene by reduction of graphite oxide with sodium hydrosulfite, Nanotechnology 22 (2011) 045704.

113

References

[76] Fernández-Merino, M. J., L. Guardia, Paredes, J. I., S. Villar-Rodil, P. Solís-Fernández, A. Martínez-Alonso, Tascón, J. M. D., Vitamin C is an ideal substitute for hydrazine in the reduction of graphene oxide suspensions, J. Phys. Chem. C 114 (2010) 6426–6432. [77] J. Zhang, H. Yang, G. Shen, P. Cheng, J. Zhang, S. Guo, Reduction of graphene oxide vial-ascorbic acid, Chem. Commun. 46 (2010) 1112–1114. [78] V. Dua, S.P. Surwade, S. Ammu, S.R. Agnihotra, S. Jain, K.E. Roberts, S. Park, R.S. Ruoff, S.K. Manohar, All-organic vapor sensor using inkjet-printed reduced graphene oxide, Angew. Chem. Int. Ed. 49 (2010) 2154–2157. [79] J. Liu, S. Fu, B. Yuan, Y. Li, Z. Deng, Toward a universal “adhesive nanosheet” for the assembly of multiple nanoparticles based on a protein-induced reduction/decoration of graphene oxide, J. Am. Chem. Soc. 132 (2010) 7279–7281. [80] M.J. McAllister, J.-L. Li, D.H. Adamson, H.C. Schniepp, A.A. Abdala, J. Liu, M. Herrera-Alonso, D.L. Milius, R. Car, R.K. Prud'homme, I.A. Aksay, Single sheet functionalized graphene by oxidation and thermal expansion of graphite, Chem. Mater. 19 (2007) 4396–4404. [81] H. Wang, J.T. Robinson, X. Li, H. Dai, Solvothermal reduction of chemically exfoliated graphene sheets, J. Am. Chem. Soc. 131 (2009) 9910–9911. [82] S. Dubin, S. Gilje, K. Wang, V.C. Tung, K. Cha, A.S. Hall, J. Farrar, R. Varshneya, Y. Yang, R.B. Kaner, A One-Step, Solvothermal reduction method for producing reduced graphene oxide dispersions in organic solvents, ACS Nano 4 (2010) 3845–3852. [83] G. Williams, B. Seger, P.V. Kamat, TiO2-graphene nanocomposites. UV-assisted photocatalytic reduction of graphene oxide, ACS Nano 2 (2008) 1487–1491. [84] O. Akhavan, E. Ghaderi, Photocatalytic reduction of graphene oxide nanosheets on TiO2 thin film for photoinactivation of bacteria in solar light irradiation, J. Phys. Chem. C 113 (2009) 20214–20220. [85] K.K. Manga, Y. Zhou, Y. Yan, K.P. Loh, Multilayer hybrid films consisting of alternating graphene and titania nanosheets with ultrafast electron transfer and photoconversion properties, Adv. Funct. Mater. 19 (2009) 3638–3643. [86] D.A. Sokolov, K.R. Shepperd, T.M. Orlando, Formation of graphene features from direct laser-induced reduction of graphite oxide, J. Phys. Chem. Lett. 1 (2010) 2633– 2636. [87] Y. Zhou, Q. Bao, B. Varghese, Tang, Lena Ai Ling, C.K. Tan, C.-H. Sow, K.P. Loh, Microstructuring of graphene oxide nanosheets using direct laser writing, Adv. Mater. 22 (2010) 67–71. [88] M. Baraket, S.G. Walton, Z. Wei, E.H. Lock, J.T. Robinson, P. Sheehan, Reduction of graphene oxide by electron beam generated plasmas produced in methane/argon mixtures, Carbon 48 (2010) 3382–3390. [89] Z. Wang, X. Zhou, J. Zhang, F. Boey, H. Zhang, Direct electrochemical reduction of single-layer graphene oxide and subsequent functionalization with glucose oxidase, J. Phys. Chem. C 113 (2009) 14071–14075. [90] Y. Shao, J. Wang, M. Engelhard, C. Wang, Y. Lin, Facile and controllable electrochemical reduction of graphene oxide and its applications, J. Mater. Chem. 20 (2010) 743–748. [91] L. Chen, Y. Tang, K. Wang, C. Liu, S. Luo, Direct electrodeposition of reduced graphene oxide on glassy carbon electrode and its electrochemical application, Electrochem. Commun. 13 (2011) 133–137. [92] W. Gao, L.B. Alemany, L. Ci, P.M. Ajayan, New insights into the structure and reduction of graphite oxide, Nat Chem 1 (2009) 403–408. [93] J. Geng, L. Liu, S.B. Yang, S.-C. Youn, D.W. Kim, J.-S. Lee, J.-K. Choi, H.-T. Jung, A simple approach for preparing transparent conductive graphene films using the 114

References

controlled chemical reduction of exfoliated graphene oxide in an aqueous suspension, J. Phys. Chem. C 114 (2010) 14433–14440. [94] W. Gao, L.B. Alemany, L. Ci, P.M. Ajayan, New insights into the structure and reduction of graphite oxide, Nat Chem 1 (2009) 403–408. [95] D.V. Kosynkin, A.L. Higginbotham, A. Sinitskii, J.R. Lomeda, A. Dimiev, B.K. Price, J.M. Tour, Longitudinal unzipping of carbon nanotubes to form graphene nanoribbons, Nature 458 (2009) 872–876. [96] N. Zhan, M. Olmedo, G. Wang, J. Liu, Layer-by-layer synthesis of large-area graphene films by thermal cracker enhanced gas source molecular beam epitaxy, Carbon 49 (2011) 2046–2052. [97] K.S. Subrahmanyam, L.S. Panchakarla, A. Govindaraj, Rao, C. N. R., Simple method of preparing graphene flakes by an Arc-discharge method, J. Phys. Chem. C 113 (2009) 4257–4259. [98] Y. Hernandez, V. Nicolosi, M. Lotya, F.M. Blighe, Z. Sun, S. De, I.T. McGovern, B. Holland, M. Byrne, Y.K. Gun'Ko, J.J. Boland, P. Niraj, G. Duesberg, S. Krishnamurthy, R. Goodhue, J. Hutchison, V. Scardaci, A.C. Ferrari, J.N. Coleman, High-yield production of graphene by liquid-phase exfoliation of graphite, Nat. Nanotechnol. 3 (2008) 563–568. [99] D.S. Hecht, L. Hu, G. Irvin, Emerging transparent electrodes based on thin films of carbon nanotubes, graphene, and metallic nanostructures, Adv. Mater. Weinheim 23 (2011) 1482–1513. [100] A. Reina, X. Jia, J. Ho, D. Nezich, H. Son, V. Bulovic, M.S. Dresselhaus, J. Kong, Large area, few-layer graphene films on arbitrary substrates by chemical vapor deposition, Nano Lett. 9 (2009) 30–35. [101] S. De, P.J. King, M. Lotya, A. O'Neill, E.M. Doherty, Y. Hernandez, G.S. Duesberg, J.N. Coleman, Flexible, transparent, conducting films of randomly stacked graphene from surfactant-stabilized, oxide-free graphene dispersions, Small 6 (2010) 458–464. [102] G. Eda, G. Fanchini, M. Chhowalla, Large-area ultrathin films of reduced graphene oxide as a transparent and flexible electronic material, Nat. Nanotechnol. 3 (2008) 270– 274. [103] L. Huang, Y. Huang, J. Liang, X. Wan, Y. Chen, Graphene-based conducting inks for direct inkjet printing of flexible conductive patterns and their applications in electric circuits and chemical sensors, Nano Res. 4 (2011) 675–684. [104] C. Gómez-Navarro, R.T. Weitz, A.M. Bittner, M. Scolari, A. Mews, M. Burghard, K. Kern, Electronic transport properties of individual chemically reduced graphene oxide sheets, Nano Lett. 7 (2007) 3499–3503. [105] X. Wang, L. Zhi, K. Müllen, Transparent, conductive graphene electrodes for dyesensitized solar cells, Nano Lett. 8 (2008) 323–327. [106] H.A. Becerril, J. Mao, Z. Liu, R.M. Stoltenberg, Z. Bao, Y. Chen, Evaluation of solution-processed reduced graphene oxide films as transparent conductors, ACS Nano 2 (2008) 463–470. [107] G. Eda, G. Fanchini, M. Chhowalla, Large-area ultrathin films of reduced graphene oxide as a transparent and flexible electronic material, Nat. Nanotechnol. 3 (2008) 270– 274. [108] T. Palacios, A. Hsu, H. Wang, Applications of graphene devices in RF communications, IEEE Commun. Mag. 48 (2010) 122–128. [109] A. Akturk, N. Goldsman, Electron transport and full-band electron-phonon interactions in graphene, J. Appl. Phys. 103 (2008) 053702.

115

References

[110] L. Liao, Y.-c. Lin, M. Bao, R. Cheng, J. Bai, Y. Liu, Y. Qu, K.L. Wang, Y. Huang, X. Duan, High-speed graphene transistors with a self-aligned nanowire gate, Nature 467 (2010) 305–308. [111] L. Liao, J. Bai, R. Cheng, H. Zhou, L. Liu, Y. Liu, Y. Huang, X. Duan, Scalable fabrication of self-aligned graphene transistors and circuits on glass, Nano Lett. 12 (2012) 2653–2657. [112] Y.-M. Lin, K.A. Jenkins, A. Valdes-Garcia, J.P. Small, D.B. Farmer, P. Avouris, Operation of graphene transistors at gigahertz frequencies, Nano Lett. 9 (2009) 422–426. [113] Yu-Ming Lin, Hsin-Ying Chiu, K.A. Jenkins, D.B. Farmer, P. Avouris, A. ValdesGarcia, Dual-gate graphene FETs With fT of 50 GHz, IEEE Electron Device Lett. 31 (2010) 68–70. [114] C. Dimitrakopoulos, Y.-M. Lin, A. Grill, D.B. Farmer, M. Freitag, Y. Sun, S.-J. Han, Z. Chen, K.A. Jenkins, Y. Zhu, Z. Liu, T.J. McArdle, J.A. Ott, R. Wisnieff, P. Avouris, Wafer-scale epitaxial graphene growth on the Si-face of hexagonal SiC (0001) for high frequency transistors, J. Vac. Sci. Technol. B 28 (2010) 985. [115] Y. Wu, K.A. Jenkins, A. Valdes-Garcia, D.B. Farmer, Y. Zhu, A.A. Bol, C. Dimitrakopoulos, W. Zhu, F. Xia, P. Avouris, Y.-M. Lin, State-of-the-art graphene highfrequency electronics, Nano Lett. 12 (2012) 3062–3067. [116] J.M. Tarascon, M. Armand, Issues and challenges facing rechargeable lithium batteries, Nature 414 (2001) 359–367. [117] E. Yoo, J. Kim, E. Hosono, H.-s. Zhou, T. Kudo, I. Honma, Large reversible Li storage of graphene nanosheet families for use in rechargeable lithium ion batteries, Nano Lett. 8 (2008) 2277–2282. [118] P. Lian, X. Zhu, S. Liang, Z. Li, W. Yang, H. Wang, Large reversible capacity of high quality graphene sheets as an anode material for lithium-ion batteries, Electrochim. Acta 55 (2010) 3909–3914. [119] Z.-S. Wu, W. Ren, L. Xu, F. Li, H.-M. Cheng, Doped graphene sheets as anode materials with superhigh rate and large capacity for lithium ion batteries, ACS Nano 5 (2011) 5463–5471. [120] H.S. Kim, Y. Piao, S.H. Kang, T. Hyeon, Y.-E. Sung, Uniform hematite nanocapsules based on an anode material for lithium ion batteries, Electrochem. Commun. 12 (2010) 382–385. [121] X. Wang, L. Yu, X.-L. Wu, F. Yuan, Y.-G. Guo, Y. Ma, J. Yao, Synthesis of singlecrystalline Co3O4 octahedral cages with tunable surface aperture and their lithium storage properties, J. Phys. Chem. C 113 (2009) 15553–15558. [122] M.-Y. Cheng, B.-J. Hwang, Mesoporous carbon-encapsulated NiO nanocomposite negative electrode materials for high-rate Li-ion battery, J. Power Sources 195 (2010) 4977–4983. [123] G. Zhou, D.-W. Wang, F. Li, L. Zhang, N. Li, Z.-S. Wu, L. Wen, G.Q. Lu, H.-M. Cheng, Graphene-Wrapped Fe3O4 anode material with improved reversible capacity and cyclic stability for lithium ion batteries, Chem. Mater. 22 (2010) 5306–5313. [124] H. Wang, L.-F. Cui, Y. Yang, H. Sanchez Casalongue, J.T. Robinson, Y. Liang, Y. Cui, H. Dai, Mn3O4-graphene hybrid as a high-capacity anode material for lithium ion batteries, J. Am. Chem. Soc. 132 (2010) 13978–13980. [125] S. Yang, X. Feng, S. Ivanovici, K. Müllen, Fabrication of graphene-encapsulated oxide nanoparticles: towards high-performance anode materials for lithium storage, Angew. Chem. Int. Ed. Engl. 49 (2010) 8408–8411. [126] A. Guerfi, S. Duchesne, Y. Kobayashi, A. Vijh, K. Zaghib, LiFePO4 and graphite electrodes with ionic liquids based on bis(fluorosulfonyl)imide (FSI)− for Li-ion batteries, J. Power Sources 175 (2008) 866–873. 116

References

[127] D. Rangappa, K. Sone, M. Ichihara, T. Kudo, I. Honma, Rapid one-pot synthesis of LiMPO4 (M = Fe, Mn) colloidal nanocrystals by supercritical ethanol process, Chem. Commun. (Camb.) 46 (2010) 7548–7550. [128] Y. Shi, S.-L. Chou, J.-Z. Wang, D. Wexler, H.-J. Li, H.-K. Liu, Y. Wu, Graphene wrapped LiFePO4/C composites as cathode materials for Li-ion batteries with enhanced rate capability, J. Mater. Chem. 22 (2012) 16465. [129] H. Wang, Y. Yang, Y. Liang, L.-F. Cui, H.S. Casalongue, Y. Li, G. Hong, Y. Cui, H. Dai, LiMn(1-x)Fe(x)PO4 nanorods grown on graphene sheets for ultrahigh-rateperformance lithium ion batteries, Angew. Chem. Int. Ed. Engl. 50 (2011) 7364–7368. [130] S. Yang, Y. Gong, Z. Liu, L. Zhan, D.P. Hashim, L. Ma, R. Vajtai, P.M. Ajayan, Bottom-up approach toward single-crystalline VO2-graphene ribbons as cathodes for ultrafast lithium storage, Nano Lett. 13 (2013) 1596–1601. [131] J.W. Lee, S.Y. Lim, H.M. Jeong, T.H. Hwang, J.K. Kang, J.W. Choi, Extremely stable cycling of ultra-thin V2O5 nanowire–graphene electrodes for lithium rechargeable battery cathodes, Energy Environ. Sci. 5 (2012) 9889. [132] M.D. Stoller, S. Park, Y. Zhu, J. An, R.S. Ruoff, Graphene-based ultracapacitors, Nano Lett. 8 (2008) 3498–3502. [133] X. Yang, J. Zhu, L. Qiu, D. Li, Bioinspired effective prevention of restacking in multilayered graphene films: towards the next generation of high-performance supercapacitors, Adv. Mater. Weinheim 23 (2011) 2833–2838. [134] Y. Zhu, S. Murali, M.D. Stoller, K.J. Ganesh, W. Cai, P.J. Ferreira, A. Pirkle, R.M. Wallace, K.A. Cychosz, M. Thommes, D. Su, E.A. Stach, R.S. Ruoff, Carbon-based supercapacitors produced by activation of graphene, Science 332 (2011) 1537–1541. [135] H.M. Jeong, J.W. Lee, W.H. Shin, Y.J. Choi, H.J. Shin, J.K. Kang, J.W. Choi, Nitrogen-doped graphene for high-performance ultracapacitors and the importance of nitrogen-doped sites at basal planes, Nano Lett. 11 (2011) 2472–2477. [136] Z. Bo, W. Zhu, W. Ma, Z. Wen, X. Shuai, J. Chen, J. Yan, Z. Wang, K. Cen, X. Feng, Vertically oriented graphene bridging active-layer/current-collector interface for ultrahigh rate supercapacitors, Adv. Mater. Weinheim 25 (2013) 5799–5806. [137] G. Yu, L. Hu, M. Vosgueritchian, H. Wang, X. Xie, J.R. McDonough, X. Cui, Y. Cui, Z. Bao, Solution-processed graphene/MnO2 nanostructured textiles for high-performance electrochemical capacitors, Nano Lett. 11 (2011) 2905–2911. [138] M. Xue, F. Li, J. Zhu, H. Song, M. Zhang, T. Cao, Structure-based enhanced capacitance: in situ growth of highly ordered polyaniline nanorods on reduced graphene oxide patterns, Adv. Funct. Mater. 22 (2012) 1284–1290. [139] X.-C. Dong, H. Xu, X.-W. Wang, Y.-X. Huang, M.B. Chan-Park, H. Zhang, L.-H. Wang, W. Huang, P. Chen, 3D graphene-cobalt oxide electrode for high-performance supercapacitor and enzymeless glucose detection, ACS Nano 6 (2012) 3206–3213. [140] J.H. Burroughes, Bradley, D. D. C., A.R. Brown, R.N. Marks, K. Mackay, R.H. Friend, P.L. Burns, A.B. Holmes, Light-emitting diodes based on conjugated polymers, Nature 347 (1990) 539–541. [141] I. Hamberg, C.G. Granqvist, Evaporated Sn-doped In2O3 films: Basic optical properties and applications to energy-efficient windows, J. Appl. Phys. 60 (1986) R123. [142] J. Wu, M. Agrawal, H.A. Becerril, Z. Bao, Z. Liu, Y. Chen, P. Peumans, Organic light-emitting diodes on solution-processed graphene transparent electrodes, ACS Nano 4 (2010) 43–48. [143] P. Matyba, H. Yamaguchi, G. Eda, M. Chhowalla, L. Edman, N.D. Robinson, Graphene and mobile ions: the key to all-plastic, solution-processed light-emitting devices, ACS Nano 4 (2010) 637–642.

117

References

[144] T.-H. Han, Y. Lee, M.-R. Choi, S.-H. Woo, S.-H. Bae, B.H. Hong, J.-H. Ahn, T.-W. Lee, Extremely efficient flexible organic light-emitting diodes with modified graphene anode, Nature Photon. 6 (2012) 105–110. [145] Global Market Outlook for Photovoltaics 2014-2018, European Photovoltaic Industry Association (EPIA), http://www.epia.org/news/publications/global-market-outlook-forphotovoltaics-2014-2018/. [146] Gomez De Arco, Lewis, Y. Zhang, C.W. Schlenker, K. Ryu, M.E. Thompson, C. Zhou, Continuous, highly flexible, and transparent graphene films by chemical vapor deposition for organic photovoltaics, ACS Nano 4 (2010) 2865–2873. [147] Z. Yin, S. Sun, T. Salim, S. Wu, X. Huang, Q. He, Y.M. Lam, H. Zhang, Organic photovoltaic devices using highly flexible reduced graphene oxide films as transparent electrodes, ACS Nano 4 (2010) 5263–5268. [148] Y. Wang, S.W. Tong, X.F. Xu, B. Ozyilmaz, K.P. Loh, Interface engineering of layerby-layer stacked graphene anodes for high-performance organic solar cells, Adv. Mater. Weinheim 23 (2011) 1514–1518. [149] H. Bi, F. Huang, J. Liang, X. Xie, M. Jiang, Transparent conductive graphene films synthesized by ambient pressure chemical vapor deposition used as the front electrode of CdTe solar cells, Adv. Mater. Weinheim 23 (2011) 3202–3206. [150] J. Wu, H.A. Becerril, Z. Bao, Z. Liu, Y. Chen, P. Peumans, Organic solar cells with solution-processed graphene transparent electrodes, Appl. Phys. Lett. 92 (2008) 263302. [151] Y.-Y. Choi, S.J. Kang, H.-K. Kim, W.M. Choi, S.-I. Na, Multilayer graphene films as transparent electrodes for organic photovoltaic devices, Sol. Energy Mater. Sol. Cells 96 (2012) 281–285. [152] H. Park, J.A. Rowehl, K.K. Kim, V. Bulovic, J. Kong, Doped graphene electrodes for organic solar cells, Nanotechnology 21 (2010) 505204. [153] C.-L. Hsu, C.-T. Lin, J.-H. Huang, C.-W. Chu, K.-H. Wei, L.-J. Li, Layer-by-layer graphene/TCNQ stacked films as conducting anodes for organic solar cells, ACS Nano 6 (2012) 5031–5039. [154] B. O'Regan, M. Grätzel, A low-cost, high-efficiency solar cell based on dye-sensitized colloidal TiO2 films, Nature 353 (1991) 737–740. [155] A. Yella, H.-W. Lee, H.N. Tsao, C. Yi, A.K. Chandiran, M.K. Nazeeruddin, E.W.-G. Diau, C.-Y. Yeh, S.M. Zakeeruddin, M. Gratzel, Porphyrin-sensitized solar cells with cobalt (II/III)-based redox electrolyte exceed 12 percent efficiency, Science 334 (2011) 629–634. [156] W. Hong, Y. Xu, G. Lu, C. Li, G. Shi, Transparent graphene/PEDOT–PSS composite films as counter electrodes of dye-sensitized solar cells, Electrochem. Commun. 10 (2008) 1555–1558. [157] J.D. Roy-Mayhew, D.J. Bozym, C. Punckt, I.A. Aksay, Functionalized graphene as a catalytic counter electrode in dye-sensitized solar cells, ACS Nano 4 (2010) 6203–6211. [158] H. Choi, H. Kim, S. Hwang, Y. Han, M. Jeon, Graphene counter electrodes for dyesensitized solar cells prepared by electrophoretic deposition, J. Mater. Chem. 21 (2011) 7548. [159] D.W. Zhang, X.D. Li, H.B. Li, S. Chen, Z. Sun, X.J. Yin, S.M. Huang, Graphenebased counter electrode for dye-sensitized solar cells, Carbon 49 (2011) 5382–5388. [160] S.-Y. Jang, Y.-G. Kim, D.Y. Kim, H.-G. Kim, S.M. Jo, Electrodynamically sprayed thin films of aqueous dispersible graphene nanosheets: highly efficient cathodes for dyesensitized solar cells, ACS Appl. Mater. Interfaces 4 (2012) 3500–3507. [161] M.-S. Wu, Y.-J. Zheng, Electrophoresis of randomly and vertically embedded graphene nanosheets in activated carbon film as a counter electrode for dye-sensitized solar cells, Phys. Chem. Chem. Phys. 15 (2013) 1782–1787. 118

References

[162] H. Zheng, C.Y. Neo, X. Mei, J. Qiu, J. Ouyang, Reduced graphene oxide films fabricated by gel coating and their application as platinum-free counter electrodes of highly efficient iodide/triiodide dye-sensitized solar cells, J. Mater. Chem. 22 (2012) 14465. [163] Y. Xue, J. Liu, H. Chen, R. Wang, D. Li, J. Qu, L. Dai, Nitrogen-doped graphene foams as metal-free counter electrodes in high-performance dye-sensitized solar cells, Angew. Chem. Int. Ed. Engl. 51 (2012) 12124–12127. [164] V. Tjoa, J. Chua, S.S. Pramana, J. Wei, S.G. Mhaisalkar, N. Mathews, Facile photochemical synthesis of graphene-pt nanoparticle composite for counter electrode in dye sensitized solar cell, ACS Appl. Mater. Interfaces 4 (2012) 3447–3452. [165] F. Gong, H. Wang, Z.-S. Wang, Self-assembled monolayer of graphene/Pt as counter electrode for efficient dye-sensitized solar cell, Phys. Chem. Chem. Phys. 13 (2011) 17676–17682. [166] L. Kavan, J.-H. Yum, M. Grätzel, Graphene nanoplatelets outperforming platinum as the electrocatalyst in co-bipyridine-mediated dye-sensitized solar cells, Nano Lett. 11 (2011) 5501–5506. [167] L. Kavan, J.-H. Yum, M.K. Nazeeruddin, M. Grätzel, Graphene nanoplatelet cathode for Co(III)/(II) mediated dye-sensitized solar cells, ACS Nano 5 (2011) 9171–9178. [168] L. Kavan, J.H. Yum, M. Grätzel, Optically transparent cathode for dye-sensitized solar cells based on graphene nanoplatelets, ACS Nano 5 (2011) 165–172. [169] J. Park, Y.H. Ahn, C. Ruiz-Vargas, Imaging of photocurrent generation and collection in single-layer graphene, Nano Lett. 9 (2009) 1742–1746. [170] F. Xia, T. Mueller, R. Golizadeh-Mojarad, M. Freitag, Y.-M. Lin, J. Tsang, V. Perebeinos, P. Avouris, Photocurrent imaging and efficient photon detection in a graphene transistor, Nano Lett. 9 (2009) 1039–1044. [171] F. Xia, T. Mueller, Y.-M. Lin, A. Valdes-Garcia, P. Avouris, Ultrafast graphene photodetector, Nature Nanotech. 4 (2009) 839–843. [172] T. Mueller, F. Xia, P. Avouris, Graphene photodetectors for high-speed optical communications, Nature Photon. 4 (2010) 297–301. [173] T.J. Echtermeyer, L. Britnell, P.K. Jasnos, A. Lombardo, R.V. Gorbachev, A.N. Grigorenko, A.K. Geim, A.C. Ferrari, K.S. Novoselov, Strong plasmonic enhancement of photovoltage in graphene, Nat. Comms. 2 (2011) 458. [174] Z. Fang, Z. Liu, Y. Wang, P.M. Ajayan, P. Nordlander, N.J. Halas, Graphene-antenna sandwich photodetector, Nano Lett. 12 (2012) 3808–3813. [175] A.N. Grigorenko, M. Polini, K.S. Novoselov, Graphene plasmonics, Nature Photon. 6 (2012) 749–758. [176] M. Furchi, A. Urich, A. Pospischil, G. Lilley, K. Unterrainer, H. Detz, P. Klang, A.M. Andrews, W. Schrenk, G. Strasser, T. Mueller, Microcavity-integrated graphene photodetector, Nano Lett. 12 (2012) 2773–2777. [177] M. Engel, M. Steiner, A. Lombardo, A.C. Ferrari, H.V. Löhneysen, P. Avouris, R. Krupke, Light-matter interaction in a microcavity-controlled graphene transistor, Nat. Commun. 3 (2012) 906. [178] X. Wang, Z. Cheng, K. Xu, H.K. Tsang, J.-B. Xu, High-responsivity graphene/silicon-heterostructure waveguide photodetectors, Nature Photon. 7 (2013) 888–891. [179] B.Y. Zhang, T. Liu, B. Meng, X. Li, G. Liang, X. Hu, Q.J. Wang, Broadband high photoresponse from pure monolayer graphene photodetector, Nat. Commun. 4 (2013) 1811.

119

References

[180] G. Konstantatos, M. Badioli, L. Gaudreau, J. Osmond, M. Bernechea, Garcia de Arquer, F. Pelayo, F. Gatti, F. H. L. Koppens, Hybrid graphene-quantum dot phototransistors with ultrahigh gain, Nat. Nanotechnol. 7 (2012) 363–368. [181] C.-H. Liu, Y.-C. Chang, T.B. Norris, Z. Zhong, Graphene photodetectors with ultrabroadband and high responsivity at room temperature, Nature Nanotech. 9 (2014) 273– 278. [182] L. Qu, Y. Liu, J.-B. Baek, L. Dai, Nitrogen-doped graphene as efficient metal-free electrocatalyst for oxygen reduction in fuel cells, ACS Nano 4 (2010) 1321–1326. [183] Y. Shao, S. Zhang, M.H. Engelhard, G. Li, G. Shao, Y. Wang, J. Liu, I.A. Aksay, Y. Lin, Nitrogen-doped graphene and its electrochemical applications, J. Mater. Chem. 20 (2010) 7491. [184] Z. Yang, Z. Yao, G. Li, G. Fang, H. Nie, Z. Liu, X. Zhou, X. Chen, S. Huang, Sulfurdoped graphene as an efficient metal-free cathode catalyst for oxygen reduction, ACS Nano 6 (2012) 205–211. [185] H.-W. Ha, I.Y. Kim, S.-J. Hwang, R.S. Ruoff, One-Pot synthesis of platinum nanoparticles embedded on reduced graphene oxide for oxygen reduction in methanol fuel cells, electrochem. Solid-State Lett. 14 (2011) B70. [186] Q. Yue, K. Zhang, X. Chen, L. Wang, J. Zhao, J. Liu, J. Jia, Generation of OH radicals in oxygen reduction reaction at Pt-Co nanoparticles supported on graphene in alkaline solutions, Chem. Commun. 46 (2010) 3369–3371. [187] K. Zhang, Q. Yue, G. Chen, Y. Zhai, L. Wang, H. Wang, J. Zhao, J. Liu, J. Jia, H. Li, Effects of acid treatment of Pt−Ni alloy nanoparticles@graphene on the kinetics of the oxygen reduction reaction in acidic and alkaline solutions, J. Phys. Chem. C 115 (2011) 379–389. [188] H.R. Byon, J. Suntivich, Y. Shao-Horn, Graphene-based non-noble-metal catalysts for oxygen reduction reaction in acid, Chem. Mater. 23 (2011) 3421–3428. [189] M.H. Seo, S.M. Choi, H.J. Kim, W.B. Kim, The graphene-supported Pd and Pt catalysts for highly active oxygen reduction reaction in an alkaline condition, Electrochem. Commun. 13 (2011) 182–185. [190] D. Geng, Y. Chen, Y. Chen, Y. Li, R. Li, X. Sun, S. Ye, S. Knights, High oxygenreduction activity and durability of nitrogen-doped graphene, Energy Environ. Sci. 4 (2011) 760. [191] K.S. Novoselov, V.I. Fal'ko, L. Colombo, P.R. Gellert, M.G. Schwab, K. Kim, A roadmap for graphene, Nature 490 (2012) 192–200. [192] S. Bae, H. Kim, Y. Lee, X. Xu, J.-S. Park, Y. Zheng, J. Balakrishnan, T. Lei, H. Ri Kim, Y.I. Song, Y.-J. Kim, K.S. Kim, B. Özyilmaz, J.-H. Ahn, B.H. Hong, S. Iijima, Roll-to-roll production of 30-inch graphene films for transparent electrodes, Nature Nanotech. 5 (2010) 574–578. [193] I.K. Moon, J. Lee, R.S. Ruoff, H. Lee, Reduced graphene oxide by chemical graphitization, Nat. Comms. 1 (2010) 1–6. [194] S. Pang, Y. Hernandez, X. Feng, K. Müllen, Graphene as transparent electrode material for organic electronics, Adv. Mater. 23 (2011) 2779–2795. [195] Y. Wang, J. Liu, L. Liu, D. Sun, High-quality reduced graphene oxide-nanocrystalline platinum hybrid materials prepared by simultaneous co-reduction of graphene oxide and chloroplatinic acid, Nanoscale Res. Lett. 6 (2011) 241. [196] Y. Wen, F.Y. Li, X. Dong, J. Zhang, Q. Xiong, P. Chen, The electrical detection of lead ions using gold-nanoparticle- and DNAzyme-functionalized graphene device, Adv. Healthc. Mater. 2 (2013) 271274.

120

References

[197] K.S. Lee, Y. Lee, J.Y. Lee, J.-H. Ahn, J.H. Park, Flexible and platinum-free dyesensitized solar cells with conducting-polymer-coated graphene counter electrodes, ChemSusChem 5 (2012) 379–382. [198] D.C. Marcano, D.V. Kosynkin, J.M. Berlin, A. Sinitskii, Z. Sun, A. Slesarev, L.B. Alemany, W. Lu, J.M. Tour, Improved synthesis of graphene oxide, ACS Nano 4 (2010) 4806–4814. [199] L.J. Cote, F. Kim, J. Huang, Langmuir−Blodgett assembly of graphite oxide single layers, J. Am. Chem. Soc. 131 (2009) 1043–1049. [200] D. Li, M.B. Müller, S. Gilje, R.B. Kaner, G.G. Wallace, Processable aqueous dispersions of graphene nanosheets, Nature Nanotech. 3 (2008) 101–105. [201] D.C. Marcano, D.V. Kosynkin, J.M. Berlin, A. Sinitskii, Z. Sun, A. Slesarev, L.B. Alemany, W. Lu, J.M. Tour, Improved synthesis of graphene oxide, ACS Nano 4 (2010) 4806–4814. [202] D. Yang, A. Velamakanni, G. Bozoklu, S. Park, M. Stoller, R.D. Piner, S. Stankovich, I. Jung, D.A. Field, Ventrice Jr. Carl A., R.S. Ruoff, Chemical analysis of graphene oxide films after heat and chemical treatments by X-ray photoelectron and micro-Raman spectroscopy, Carbon 47 (2009) 145–152. [203] L. J. van der Pauw, A method of measureing the resistivity and hall coefficient on lamellae of arbitrary shape, Philips Tech. Rev. 20 (1958) 220–224. [204] H. Wang, Y.H. Hu, Graphene as a counter electrode material for dye-sensitized solar cells, Energy Environ. Sci. 5 (2012) 8182. [205] N.G. Sahoo, Y. Pan, L. Li, S.H. Chan, Graphene-based materials for energy conversion, Adv. Mater. 24 (2012) 4203–4210. [206] Y. Xin, J.-g. Liu, Y. Zhou, W. Liu, J. Gao, Y. Xie, Y. Yin, Z. Zou, Preparation and characterization of Pt supported on graphene with enhanced electrocatalytic activity in fuel cell, J. Power Sources 196 (2011) 1012–1018. [207] Y. Lin, K. Zhang, W. Chen, Y. Liu, Z. Geng, J. Zeng, N. Pan, L. Yan, X. Wang, J.G. Hou, Dramatically enhanced photoresponse of reduced graphene oxide with linker-free anchored CdSe nanoparticles, ACS Nano 4 (2010) 3033–3038. [208] Y. Jin, Y. Shen, S. Dong, Electrochemical design of ultrathin platinum-coated gold nanoparticle monolayer films as a novel nanostructured electrocatalyst for oxygen reduction J. Phys. Chem. B 108 (2004) 8142. [209] C. Radhakrishnan, M.K.F. Lo, M.V. Warrier, M.A. Garcia-Garibay, H.G. Monbouquette, Photocatalytic reduction of an azide-terminated self-assembled monolayer using CdS quantum dots, Langmuir 22 (2006) 50185024. [210] R. Plass, S. Pelet, J. Krueger, M. Grätzel, Quantum dot sensitization of organicinorganic hybrid solar cells, J. Phys. Chem. B 106 (2002) 75787580. [211] J.L. Blackburn, D.C. Selmarten, A.J. Nozik, Electron transfer dynamics in quantum dot/titanium dioxide composites formed by in situ chemical bath deposition, J. Phys. Chem. B 107 (2003) 1415414157. [212] Y. Zhou, M. Eck, C. Men, F. Rauscher, P. Niyamakom, S. Yilmaz, I. Dumsch, S. Allard, U. Scherf, M. Krüger, Efficient polymer nanocrystal hybrid solar cells by improved nanocrystal composition, Sol. Energy Mat. Sol Cells 95 (2011) 3227–3232. [213] S. Ren, L.-Y. Chang, S.-K. Lim, J. Zhao, M. Smith, N. Zhao, V. Bulović, M. Bawendi, S. Gradečak, Inorganic–organic hybrid solar cell: bridging quantum dots to conjugated polymer nanowires, Nano Lett. 11 (2011) 3998–4002. [214] D.S. Ginger, N.C. Greenham, Charge injection and transport in films of CdSe nanocrystals, J. Appl. Phys. 87 (2000) 1361. [215] W.U. Huynh, J.J. Dittmer, A.P. Alivisatos, Hybrid nanorod-polymer solar cells, Science 295 (2002) 2425–2427. 121

References

[216] W. Huynh, J. Dittmer, N. Teclemariam, D. Milliron, A. Alivisatos, K. Barnham, Charge transport in hybrid nanorod-polymer composite photovoltaic cells, Phys. Rev. B 67 (2003) 115326. [217] N. Shang, P. Papakonstantinou, P. Wang, S.R.P. Silva, platinum integrated graphene for methanol fuel cells, J. Phys. Chem. C 114 (2010) 15837–15841. [218] J.-D. Qiu, G.-C. Wang, R.-P. Liang, X.-H. Xia, H.-W. Yu, Controllable deposition of platinum nanoparticles on graphene as an electrocatalyst for direct methanol fuel cells, J. Phys. Chem. C 115 (2011) 15639–15645. [219] F. Gong, H. Wang, Z.-S. Wang, Self-assembled monolayer of graphene/Pt as counter electrode for efficient dye-sensitized solar cell, Phys. Chem. Chem. Phys 13 (2011) 17676–17682. [220] A.K. Geim, K.S. Novoselov, The rise of graphene, Nat. Mater. 6 (2007) 183–191. [221] D. Lu, Y. Zhang, S. Lin, L. Wang, C. Wang, Sensitive detection of esculetin based on a CdSe nanoparticles-decorated poly(diallyldimethylammonium chloride)-functionalized graphene nanocomposite film, ANALYST 136 (2011) 4447–4453. [222] A.F. Zedan, S. Sappal, S. Moussa, M.S. El-Shall, Ligand-controlled microwave synthesis of cubic and hexagonal CdSe nanocrystals supported on graphene. photoluminescence quenching by graphene, J. Phys. Chem. C 114 (2010) 19920–19927. [223] Y. Wang, H.-B. Yao, X.-H. Wang, S.-H. Yu, One-pot facile decoration of CdSe quantum dots on graphene nanosheets: novel graphene-CdSe nanocomposites with tunable fluorescent properties, J. Mater. Chem. 21 (2010) 562. [224] J. Shen, M. Shi, N. Li, B. Yan, H. Ma, Y. Hu, M. Ye, Facile synthesis and application of Ag-chemically converted graphene nanocomposite, Nano Res. 3 (2010) 339–349. [225] D. Lu, Y. Zhang, S. Lin, L. Wang, C. Wang, Sensitive detection of esculetin based on a CdSe nanoparticles-decorated poly(diallyldimethylammonium chloride)-functionalized graphene nanocomposite film, Analyst 136 (2011) 4447. [226] M. Feng, R. Sun, H. Zhan, Y. Chen, Lossless synthesis of graphene nanosheets decorated with tiny cadmium sulfide quantum dots with excellent nonlinear optical properties, Nanotechnology 21 (2010) 075601. [227] J.R. Mann, D.F. Watson, Adsorption of CdSe nanoparticles to thiolated TiO2 surfaces: influence of intralayer disulfide formation on CdSe surface coverage, Langmuir 23 (2007) 10924–10928. [228] V.L. Colvin, A.N. Goldstein, A.P. Alivisatos, J. Am. Chem. Soc 114 (1992) 5221. [229] J. Čech, S.A. Curran, D. Zhang, J.L. Dewald, A. Avadhanula, M. Kandadai, S. Roth, Functionalization of multi-walled carbon nanotubes: Direct proof of sidewall thiolation, Phys. Stat. Sol. (b) 243 (2006) 3221–3225. [230] T. Ozturk, E. Ertas, O. Mert, A Berzelius reagent, phosphorus decasulfide (P4S10), in Organic Syntheses, Chem. Rev. 110 (2010) 3419–3478. [231] D. Choudhury, B. Das, D.D. Sarma, Rao, C. N. R., XPS evidence for molecular charge-transfer doping of graphene, Chem. Phys. Lett. 497 (2010) 66–69. [232] J.A. Bearden, A.F. Burr, Reevaluation of X-ray atomic energy levels, Rev. Mod. Phys. 39 (1967) 125–142. [233] J. Moulder, J. Chastain, Handbook of X-ray photoelectron spectroscopy: a reference book of standard spectra for identification and interpretation of XPS data, Perkin-Elmer, 1995. [234] A. Bagri, C. Mattevi, M. Acik, Y.J. Chabal, M. Chhowalla, V.B. Shenoy, Structural evolution during the reduction of chemically derived graphene oxide, Nature Chem. 2 (2010) 581–587.

122

References

[235] P. Cao, J. Yao, B. Ren, R. Gu, and Z. Tian, Surface-enhanced raman scattering spectra of thiourea adsorbed at an iron electrode in NaClO4 solution, J. Phys. Chem. B, 106 (39) (2002) 10150–10156. [236] Y. Yuan, F.S. Riehle, H. Gu, R. Thomann, G. Urban, M. Krüger, Critical parameters for the scale-up synthesis of quantum dots, J Nanosci. Nanotechnol. 10 (2010) 6041– 6045. [237] Y. Zhou, F. S. Riehle, Y. Yuan, H.-F. Schleiermacher, M. Niggemann, G. A. Urban, M. Kruger, Improved efficiency of hybrid solar cells based on non-ligand-exchanged CdSe quantum dots and poly(3-hexylthiophene), App. Phys. Lett. 96 (2010) 013304. [238] Y. Zhou, M. Eck, C. Veit, B. Zimmermann, F. Rauscher, P. Niyamakom, S. Yilmaz, I. Dumsch, S. Allard, U. Scherf, M. Krüger, Efficiency enhancement for bulkheterojunction hybrid solar cells based on acid treated CdSe quantum dots and low bandgap polymer PCPDTBT, Sol. Energ. Mat. and Sol. C 95 (2011) 1232–1237. [239] I.V. Lightcap, P.V. Kamat, Fortification of CdSe quantum dots with graphene oxide excited state interactions and light energy conversion, J. Am. Chem. Soc 134 (2012) 7109–7116. [240] L. Brus, Electronic wave functions in semiconductor clusters: experiment and theory, J. Phys. Chem. 90 (1986) 2555–2560. [241] J. Liu, Y. Xue, Y. Gao, D. Yu, M. Durstock, L. Dai, Hole and electron extraction layers based on graphene oxide derivatives for high-performance bulk heterojunction solar cells, Adv. Mater. 24 (2012) 2228–2233. [242] C.V. Pham, M. Krueger, M. Eck, S. Weber, E. Erdem, Comparative electron paramagnetic resonance investigation of reduced graphene oxide and carbon nanotubes with different chemical functionalities for quantum dot attachment, Appl. Phys. Lett. 104 (2014) 132102. [243] R. Jin, Y. Cao, C.A. Mirkin, K.L. Kelly, G.C. Schatz, J.G. Zheng, Photoinduced conversion of silver nanospheres to nanoprisms, Science 294 (2001) 1901–1903. [244] L. Longenberger, G. Mills, Formation of metal particles in aqueous solutions by reactions of metal complexes with polymers, J. Phys. Chem. 99 (1995) 475–478. [245] J. Shen, M. Shi, N. Li, B. Yan, H. Ma, Y. Hu, M. Ye, Facile synthesis and application of Ag-chemically converted graphene nanocomposite, Nano Res. 3 (2010) 339–349. [246] M. Gratzel, Photoelectrochemical cells, Nature 414 (2001) 338–344. [247] G. Williams, P.V. Kamat, Graphene−semiconductor nanocomposites: excited-state interactions between ZnO nanoparticles and graphene oxide†, Langmuir 25 (2009) 13869–13873. [248] G. Singh, A. Choudhary, D. Haranath, A.G. Joshi, N. Singh, S. Singh, R. Pasricha, ZnO decorated luminescent graphene as a potential gas sensor at room temperature, Carbon 50 (2012) 385–394. [249] Z. Yin, S. Wu, X. Zhou, X. Huang, Q. Zhang, F. Boey, H. Zhang, Electrochemical deposition of ZnO nanorods on transparent reduced graphene oxide electrodes for hybrid Solar cells, Small 6 (2010) 307–312. [250] G. Eaton, S. Eaton, D. Barr, R. Weber, Basics of continuous wave EPR, in: Quantitative EPR, Springer Vienna, 2010, pp. 1-14. [251] Wertz, J.E., Boton, J.R., Electron spin resonance: elementary theory and practical applications, McGraw Hill, 1972. [252] R.S. Drago, Physical Methods for Chemists, 2nd ed., Saunders, 1992. [253] P. Zeeman, VII. Doublets and triplets in the spectrum produced by external magnetic forces, Philosophical Magazine Series 5 44 (1897) 55–60.

123

References

[254] B. Odom, D. Hanneke, B. D’Urso, G. Gabrielse, New measurement of the electron magnetic moment using a one-electron quantum cyclotron, Phys. Rev. Lett. 97 (2006) 030801. [255] M. Zaka, Y. Ito, H. Wang, W. Yan, A. Robertson, Y.A. Wu, M.H. Rümmeli, D. Staunton, T. Hashimoto, J.J.L. Morton, A. Ardavan, G.A.D. Briggs, J.H. Warner, Electron paramagnetic resonance investigation of purified catalyst-free single-walled carbon nanotubes, ACS Nano 4 (2010) 7708–7716. [256] J. Kausteklis, P. Cevc, D. Arčon, L. Nasi, D. Pontiroli, M. Mazzani, M. Riccò, Electron paramagnetic resonance study of nanostructured graphite, Phys. Rev. B 84 (2011) 125406. [257] S.S. Rao, A. Stesmans, D.V. Kosynkin, A. Higginbotham, J.M. Tour, Paramagnetic centers in graphene nanoribbons prepared from longitudinal unzipping of carbon nanotubes, New J. Phys. 13 (2011) 113004. [258] F. Beckert, A.M. Rostas, R. Thomann, S. Weber, E. Schleicher, C. Friedrich, R. Mülhaupt, Self-initiated free radical grafting of styrene homo- and copolymers onto functionalized graphene, Macromolecules 46 (2013) 5488–5496. [259] A.M. Panich, A.I. Shames, A.E. Aleksenskii, A. Dideikin, Magnetic resonance evidence of manganese–graphene complexes in reduced graphene oxide, Solid State Commun. 152 (2012) 466–468. [260] A. Jorio, G. Dresselhaus, M.S. Dresselhaus, Carbon nanotubes: Advanced topics in the synthesis, structure, properties, and applications, Springer, Berlin, New York, 2008. [261] C.E. Ascheron, A.H. Duhm, A. Jorio, G. Dresselhaus, M.S. Dresselhaus (Eds.), Carbon Nanotubes, Springer Berlin Heidelberg, Berlin, Heidelberg, 2008. [262] G.R. Eaton, S.S. Eaton, D.P.Barr, R.T. Weber, Quantitative EPR, Springer Vienna, 2010. [263] G. Eaton, S. Eaton, D. Barr, R. Weber, Filling Factor, in: Quantitative EPR, Springer Vienna, 2010, pp. 89-90. [264] T.Hirade, Acta Phys Pol A 107 (2005) 615–622. [265] M.B. Boeddinghaus, W. Klein, B. Wahl, P. Jakes, R.-A. Eichel, T.F. Fässler, C 603versus C 604- /C 602- Synthesis and characterization of five Salts containing discrete fullerene anions, Z. Anorg. Allg. Chem. 640 (2014) 701–712. [266] Michael Eck, Performance enhancement of hybrid nanocrystal-polymer bulk heterojunction solar cells aspects of device efficiency, reproducibility, and stability, PhD thesis, Freiburg, Germany, 2014. [267] J. Zhao, A. Wang, M.A. Green, F. Ferrazza, 19.8% efficient “honeycomb” textured multicrystalline and 24.4% monocrystalline silicon solar cells, Appl. Phys. Lett. 73 (1998) 1991. [268] Photovoltaics report, Fraunhofer ISE, 28 July 2014, http://www.ise.fraunhofer.de/en/ downloads-englisch/pdf-files-englisch/photovoltaics-report-slides.pdf . [269] Y. Zhou, M. Eck, M. Krüger, Bulk-heterojunction hybrid solar cells based on colloidal nanocrystals and conjugated polymers, Energy Environ. Sci. 3 (2010) 1851. [270] C. Jonda, Mayer, A. B. R., U. Stolz, A. Elschner, A. Karbach, Surface roughness effects and their influence on the degradation of organic light emitting devices, J Mater. Sci. 35 (2000) 5645–5651. [271] A. Elschner, F. Bruder, H.-W. Heuer, F. Jonas, A. Karbach, S. Kirchmeyer, S. Thurm, R. Wehrmann, PEDT/PSS for efficient hole-injection in hybrid organic light-emitting diodes, Synthetic Met. 111-112 (2000) 139–143. [272] R.N. Marks, Halls, J J M, Bradley, D D C, R.H. Friend, A.B. Holmes, The photovoltaic response in poly(p-phenylene vinylene) thin-film devices, J. Phys.: Condens. Matter 6 (1994) 1379–1394. 124

References

[273] S. Barth, H. Bässler, Intrinsic photoconduction in PPV-type conjugated polymers, Phys. Rev. Lett. 79 (1997) 4445–4448. [274] J. J. M. Halls, K. Pichler, R.H. Friend, S.C. Moratti, A.B. Holmes, Exciton diffusion and dissociation in a poly(p-phenylenevinylene)/C60 heterojunction photovoltaic cell, Appl. Phys. Lett. 68 (1996) 3120. [275] D. Markov, C. Tanase, P. Blom, J. Wildeman, Simultaneous enhancement of charge transport and exciton diffusion in poly(p-phenylene vinylene) derivatives, Phys. Rev. B 72 (2005) 045217. [276] J. Halls, J. Cornil, D. dos Santos, R. Silbey, D.-H. Hwang, A. Holmes, J. Brédas, R. Friend, Charge- and energy-transfer processes at polymer/polymer interfaces: A joint experimental and theoretical study, Phys. Rev. B 60 (1999) 5721–5727. [277] Z. Liu, Q. Liu, Y. Huang, Y. Ma, S. Yin, X. Zhang, W. Sun, Y. Chen, Organic photovoltaic devices based on a novel acceptor material: graphene, Adv. Mater. 20 (2008) 3924–3930. [278] V. Gupta, N. Chaudhary, R. Srivastava, G.D. Sharma, R. Bhardwaj, S. Chand, Luminscent graphene quantum dots for organic photovoltaic devices, J. Am. Chem. Soc. 133 (2011) 9960–9963. [279] Y. Ye, L. Gan, L. Dai, Y. Dai, X. Guo, H. Meng, B. Yu, Z. Shi, K. Shang, G. Qin, A simple and scalable graphene patterning method and its application in CdSe nanobelt/graphene Schottky junction solar cells, Nanoscale 3 (2011) 1477–1481. [280] X. Miao, S. Tongay, M.K. Petterson, K. Berke, A.G. Rinzler, B.R. Appleton, A.F. Hebard, High efficiency graphene solar cells by chemical doping, Nano Lett. 12 (2012) 2745–2750. [281] N. Yang, J. Zhai, D. Wang, Y. Chen, L. Jiang, Two-dimensional graphene bridges enhanced photoinduced charge transport in dye-sensitized solar cells, ACS Nano 4 (2010) 887–894. [282] D.W. Chang, H.-J. Choi, A. Filer, J.-B. Baek, Graphene in photovoltaic applications: organic photovoltaic cells (OPVs) and dye-sensitized solar cells (DSSCs), J. Mater. Chem. A 2 (2014) 12136. [283] Z. Yin, J. Zhu, Q. He, X. Cao, C. Tan, H. Chen, Q. Yan, H. Zhang, Graphene-based materials for solar cell applications, Adv. Energy Mater. 4 (2014) 1300574. [284] M. Eck, C. van Pham, S. Züfle, M. Neukom, M. Sessler, D. Scheunemann, E. Erdem, S. Weber, H. Borchert, B. Ruhstaller, M. Krüger, Improved efficiency of bulk heterojunction hybrid solar cells by utilizing CdSe quantum dot-graphene nanocomposites, Phys. Chem. Chem. Phys. 16 (2014) 12251–12260. [285] C.M. Cardona, W. Li, A.E. Kaifer, D. Stockdale, G.C. Bazan, Electrochemical considerations for determining absolute frontier orbital energy levels of conjugated polymers for solar cell applications, Adv. Mater. 23 (2011) 2367–2371.

125