enhanced vertical charge transport ...hilke/publications/pdf... · when in contact with a...

FULL P

APER

© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 1wileyonlinelibrary.com

when in contact with a semiconducting organic material such as P3HT. Strong photoluminescence and fl uorescence quenching has been recently reported between graphene and P3HT, indicating electron transfer at the interface between the two materials. [ 5,6 ] Already, examples of graphene/P3HT composite devices such as organic memories, photovoltaic devices, and organic fi eld-effect transistors have been demonstrated. [ 5–9 ]

Charge transport in conjugated poly-mers occurs either in the direction of π–π stacking, or along the chain backbone which is the fastest transport direction inside an aggregate. Although processing conditions (e.g. spin speed, solvent) can alter its crystalline orientation, P3HT has a strong tendency to form edge-on lamellae with π–π stacking in the plane of the fi lm on weakly interacting surfaces (e.g. silicon

oxide). For this reason, P3HT has been one of the most studied materials for organic based FETs which require high in-plane mobility. [ 10–13 ] The strong light absorbance of P3HT in the vis-ible range has also generated a lot of interest as a donor mate-rial for photovoltaics (PV). However, in this case the ability to effi ciently transport charges vertically (from top to bottom of the fi lm) is a necessary condition. Unfortunately, edge-on orien-tation does not produce effi cient vertical charge transport, and mobility is most often measured in the plane of the fi lm, which does not tell us how effi ciently charges can be transported out of plane. Face-on or chain-on orientations would be preferable for vertical charge transport in photovoltaics. [ 11,14,15 ] Besides the huge interest generated by both graphene and P3HT for applications, there is still very little known about P3HT crys-tallization on a graphene surface, and how it affects crystalline orientation and charge transport. A better understanding of this process is therefore of strong interest for both fundamental and applied science.

In this paper, we report the crystallization of highly regio-regular P3HT on single layer graphene, and we compare it to a P3HT layer deposited on a silicon wafer with native oxide. The crystalline structure of fi lms with two different thicknesses (10 and 50 nm) was characterized by 2D grazing incidence X-rays diffraction (2D GIXD). The optical and electrical properties of the fi lms were measured by several complementary methods. The fi lms deposited on graphene had a very different distribution of crystallite orientations, and exhibited a much higher degree of π–π stacking perpendicular to the plane of the fi lm, compared to

Enhanced Vertical Charge Transport in a Semiconducting P3HT Thin Film on Single Layer Graphene

Vasyl Skrypnychuk , Nicolas Boulanger , Victor Yu , Michael Hilke , Stefan C. B. Mannsfeld , Michael F. Toney , and David R. Barbero *

The crystallization and electrical characterization of the semiconducting polymer poly(3-hexylthiophene) (P3HT) on a single layer graphene sheet is reported. Grazing incidence X-ray diffraction revealed that P3HT crystallizes with a mixture of face-on and edge-on lamellar orientations on graphene com-pared to mainly edge-on on a silicon substrate. Moreover, whereas ultrathin (10 nm) P3HT fi lms form well oriented face-on and edge-on lamellae, thicker (50 nm) fi lms form a mosaic of lamellae oriented at different angles from the graphene substrate. This mosaic of crystallites with π–π stacking oriented homogeneously at various angles inside the fi lm favors the creation of a continuous pathway of interconnected crystallites, and results in a strong enhancement in vertical charge transport and charge carrier mobility in the thicker P3HT fi lm. These results provide a better understanding of poly-thiophene crystallization on graphene, and should help the design of more effi cient graphene based organic devices by control of the crystallinity of the semiconducting fi lm.

DOI: 10.1002/adfm.201403418

V. Skrypnychuk, N. Boulanger, Prof. D. R. Barbero Department of Physics Umeå University Umeå 90187 , Sweden E-mail: [email protected] V. Yu, Prof. M. Hilke Department of Physics McGill University, Montréal Québec H3A 2T8 , Canada Prof. S. C. B. Mannsfeld, Dr. M. F. Toney Stanford Synchrotron Radiation Lightsource Menlo Park , CA 90187 , USA

1. Introduction

Graphene, a two dimensional semi-metal made of sp 2 hybrid-ized carbon, is an outstanding material which exhibits high mechanical and chemical stability. Due to its bal-listic charge transport and high electron mobility reaching 140 000 cm 2 V −1 s −1 at room temperature, graphene has been hailed as one of the most exciting new materials for electronic devices. [ 1,2 ] The high transparency of single and few layers gra-phene is advantageous for optoelectronic applications where light must enter the active layer of the device. [ 3 ] Recently, devices such as fi eld-effect transistors (FETs), and organic photo voltaics have been produced using graphene. [ 1,4,5 ] Moreover, due to its large surface area and highly delocalized electrons, it is expected that graphene can produce effi cient charge transfer

Adv. Funct. Mater. 2014, DOI: 10.1002/adfm.201403418

www.afm-journal.dewww.MaterialsViews.com

http://doi.wiley.com/10.1002/adfm.201403418

FULL

PAPER

2 wileyonlinelibrary.com © 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

the strong in plane π stacking in fi lms on silicon. Surprisingly, we also observed a clear difference in crystallite orientations between ultrathin (10 nm) and thin (50 nm) fi lms on graphene. A lower amount of face–on lamellae, and a mosaic of π–π ori-entations, was observed in the 50 nm fi lm on graphene, which resulted in a strong enhancement in vertical charge transport and charge–carrier mobilities. Raman and UV–vis spectroscopy also revealed a slight increase in correlation length and in chain planarity on graphene. These results suggest that graphene could help improve the vertical charge transport, and the effi ciency of organic and hybrid light emitting diodes (LEDs) and PVs.

2. Results

Large-area single layer graphene was synthesized on a copper foil by chemical vapor deposition (CVD) using a ver-tical quartz tube as explained in detail in the Supporting Information. [ 16 ] Raman spectroscopy showed full monolayer coverage of the graphene on copper. [ 17 ] A free standing gra-phene layer was then made by etching of the copper foil, and transferred onto a solid substrate (either silicon or glass) by fl oating in DI water, and dried overnight. Raman spec-troscopy of the graphene transferred to silicon ( Figure 1 A) revealed that the characteristic G peak at ≈1580 cm −1 and the G′ band at ≈ 2785 cm −1 were very sharp and well defi ned in our samples. The position and the small full width at half maximum (FWHM ≈ 30.8 cm −1 ) of the G′ band corresponds to single layer graphene. [ 18 ]

Figure 1 B shows the optical transmittance of a glass slide, and of a layer of graphene deposited on the same glass slide.

Glass refl ects ≈8% of the incident light which results in a trans-mittance of ≈92% in the visible spectrum. By contrast the glass/graphene sample transmits slightly less (≈88–90% in the same range of wavelength) due to the slight absorbance of graphene. This can be seen in Figure 1 C which shows a monolayer of gra-phene deposited in the center of a glass substrate, and which appears slightly darker than the glass. The values measured are indeed consistent with the transmittance of a monolayer of graphene reported in the literature. [ 3 ] This refl ects the high quality and the single layer character of the graphene used in this study.

2.1. Electrical Characterization

Thin fi lms of regioregular P3HT were deposited on silicon and on the graphene/silicon substrate by spin-coating from a dilute solution and annealing at 170 °C. Conductivity and mobility measurements in P3HT fi lms are usually performed in the plane of the fi lm, which is the direction in which the current fl ows in a FET device. However, the knowledge of the vertical conductivity (across the fi lm thickness) is important for some applications (e.g. photovoltaics), and it is not necessarily the same as in plane. In this study we therefore measured current fl ow perpendicularly to the fi lm surface, as shown in Figure 2 A, where a potential difference was applied between the top sur-face of the fi lm and the conductive substrate, and the resulting current measured across the fi lm thickness. Figure 2 B,C reports the current density J (current divided by the surface area of the electrode) as a function of applied voltage U. The meas-urements were repeated several times for consistency, and the



Figure 1. A) Raman spectrum of the graphene layer on silicon where the two characteristic G and G′ peaks are clearly visible. B) UV–vis transmittance of a glass slide and of a graphene/glass sample showing high transmittance and supporting the fact that the graphene is a single layer. C) Graphene layer deposited at the center of a glass sample.

FULL P

APER

3wileyonlinelibrary.com© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

data points were collapsed onto one curve for each substrate.

The conductivity σ JL

Uσ = ⋅⎛

⎝⎜⎞⎠⎟ , where L is the fi lm thickness was

measured in the diffusion regime (low voltage). The charge–carrier mobilities µ in the thin P3HT fi lms were extracted by fi tting the J-U characteristics with the classical Mott-Gurney equation which is used to describe space–charge–limited cur-rent measured in the drift regime (at higher voltages) in organic semiconductors [ 19 ] :

4 2

2

3JkT

q

U

Lπ μ= ε (1)

where J is the current density, U the applied voltage, L the sample thickness, ε the dielectric constant of the fi lm, k the Boltzmann constant, T the temperature, and q the elementary charge.

As shown in Figure 2 both P3HT fi lms deposited on gra-phene were much more conductive than the fi lms on sil-icon for the same fi lm thickness (either ≈10 nm or ≈50 nm). The P3HT fi lms on silicon reached a maximum conduc-tivity σ ≈ 2.2 × 10 −4 S m −1 , which is consistent with previous results obtained in our group, and also as measured by other groups. [ 20–22 ] However, when deposited on graphene, the conductivity in the P3HT fi lms increased to σ ≈ 4.1 × 10 −4 S m −1 in the 50 nm fi lm. Room temperature vertical mobili-ties were also higher in the fi lms on graphene and reached µ ≈ 6.5 × 10 −6 cm 2 V −1 s −1 and ≈2.8 × 10 −4 cm 2 V −1 s −1 in the 10 and 50 nm fi lms respectively, as shown in Figure 2 D. The measurements of vertical conductivity and mobility on gra-phene are the fi rst to be reported, and these are expected to

help better design applications where charges must move verti-cally as opposed to in–plane.

In order to understand the reason for the increased cur-rent density and mobility measured in the P3HT fi lms on graphene, we must look at both the energy barrier for charge injection between silicon, graphene and P3HT, and at the crys-talline order in the P3HT fi lms. The work functions (WF) of the p-doped silicon substrate (p-Si, WF ≈ –5.1 eV), graphene (WF ≈ –4.6 eV) and P3HT (WF ≈ –4.8 eV) indicate that there is a charge injection barrier between p-Si and graphene. [ 23,24 ] Therefore, charge injection is not enhanced by the presence of graphene, unless graphene’s WF is modifi ed. Modulation of graphene’s work function has been shown by application of a bias, oxygen defects or water adsorption, leading to a reduc-tion in energy barrier. [ 25,26 ] This can effectively change the work function of graphene by ≈ 0.2 eV, and enable a slightly better alignment of Fermi levels at the silicon/graphene interface. In agreement with previous studies, hole injection from both p-doped Si and p-doped Si/graphene into P3HT was observed in our experiments. [ 24 ] However, based on work functions alone, and since we had not intentionally doped graphene, it is not clear that the graphene layer in our samples helps charge transfer into the organic semiconductor compared to the sil-icon substrate. [ 24,27 ] Moreover, a better energy alignment at interfaces does not explain the strong enhancement in mobili-ties measured in the 50 nm fi lm compared to the 10 nm on graphene, which have the same energy barriers to charge trans-port. Another explanation could be that the crystallinity of the P3HT layer is different in the two fi lms. It is indeed well known that crystallinity strongly infl uences charge transport in organic semiconductors such as P3HT, and we therefore must look in detail at the crystallinity of the P3HT fi lms on both surfaces.



Figure 2. Vertical current density and mobility in the P3HT fi lms. A) Schematics of the measurement where a voltage difference is applied across the fi lm's thickness, and the resulting current is measured in the direction perpendicular to the fi lm. B) Current density–voltage characteristics of 10 nm thick fi ms, and C) 50 nm thick fi lms on silicon (Si) and graphene (G). D) Room-temperature charge–carrier mobilities in the 10 nm and 50 nm fi lms on graphene.

FULL

PAPER


2.2. X-Ray Diffraction and Crystallinity of the Films

Since conjugated polymers are semi-crystalline materials, whose degree of crystallinity and crystallite orientation is critical for charge transport, we performed a study of the crystallinity in these fi lms by the commonly used technique of 2D grazing inci-dence X-rays diffraction. [ 28–30 ] This method gives accurate infor-mation about the crystallinity, the crystallites orientation and their coherence length inside the fi lms. The three different pos-sible orientations commonly found in regioregular P3HT fi lms are illustrated in Figure 3 A. Edge-on where the (100) direction is perpendicular to the substrate and hence both the backbones and the π–π stacking lie in the plane of the fi lm. The second orientation, chain-on, provides the fastest vertical charge trans-port but is rarely observed with the chains backbones vertically aligned along the z direction. [ 31 ] The third possibility is called face-on where π–π stacking is perpendicular to the substrate, which also helps to enable charges to be transported vertically.

The diffraction patterns of 10 nm and 50 nm thin fi lms of highly regioregular P3HT deposited either on silicon or on gra-phene are shown in Figure 3 and Figure 4 respectively. These were recorded on a planar area detector which limits slightly the polar angles accessible at large q due to the curvature of the Ewald sphere. This is seen in Figure 3 B and Figure 3 C where no diffraction is visible along the vertical z axis. On

silicon, P3HT fi lms either 10 or 50 nm thick formed clearly visible edge-on lamellae distinguishable by strong diffraction peaks at (100), (200) and (300) along the z axis, as is widely observed in the literature. [ 14,32–34 ] The absence of (010) peak near the z axis and of π stacking perpendicular to the sub-strate on silicon is also in accordance with other studies. [ 14,35 ] The ultrathin (10 nm) fi lm showed a strong lamellar alignment in the plane of the fi lm as can be seen by the (h00) diffraction peaks showing little spreading away from the z axis. The 10 nm fi lm deposited on graphene (Figure 3 C) also shows strong (h00) alignment along z, but also exhibited a well defi ned (010) dif-fraction peak along the z axis, which is characteristic of π–π stacking perpendicular to the plane (face-on in Figure 3 A). This result is consistent with binding energy calculations performed between a P3HT chain and a graphene surface, which showed that face-on orientation is favored at the surface of graphene. [ 36 ]

One might think that this lamellar arrangement in P3HT would simply extend to thicker fi lms, and show a similar dif-fraction pattern as for the ultrathin fi lm. However, Figure 4 B shows an intense (100) ring in the 50 nm fi lm on graphene, which extends from in-plane (χ≈ 90°) to out-of-plane angles (χ≈ 2°). The (100) diffracted intensity is ≈10 times higher on graphene compared to silicon at high χ angles (Figure 4 C). The (010) diffraction on graphene is also more than one order of magnitude more intense than on silicon along the z direction



Figure 3. Crystallinity in the 10 nm thick P3HT fi lms at a shallow beam incident angle α = 0.13°. A) Possible lamellar orientations in a P3HT fi lm (x–y plane). From left to right: edge-on lamellae with π–π stacking in the plane of the fi lm; chain-on orientation with the P3HT chain backbone vertically oriented along the z axis; face-on lamellae with π–π stacking perpendicular to the plane of the fi lm along the z direction. B,C) 2D grazing incidence diffraction patterns of P3HT on B) silicon, and C) graphene. D,E) Variation of the integrated diffracted intensity as a function of the χ angle for the D) (100), and the E) (010) diffraction. The fi lms were processed and annealed in the same conditions (see text for additional information).

FULL P

APER

5wileyonlinelibrary.com© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

(Figure 4 D), which is characteristic of π–π stacking perpendic-ular to the plane (face-on in Figure 3 A), instead of in–plane in the fi lm on silicon. The existence of (100) and (010) rings in the 50 nm fi lm on graphene indicates that π–π stacking is oriented uniformly from in-plane to out-of-plane directions inside this fi lm on the graphene substrate. Moreover, the overall amount of face–on lamellae in the 50 nm fi lm is only half of that of the 10 nm (relative to its thickness) on graphene. Since vertical charge transport in P3HT is often associated with face–on lamellae, one would expect the 10 nm fi lm to be much more conductive, which goes against the electrical measurements shown here.

Since charges must be transported across the whole fi lm thickness, it is useful to know how large crystalline domains are in the thicker fi lms, and how charges can be effi ciently transported from one ordered aggregate to the next in the ver-tical direction. The coherence length of the crystalline aggre-gates was estimated from the FWHM of the (100) diffraction peaks using the Debye-Scherrer equation:

cosl

Kλβ θ

= (2)

where l is the coherence length, K is a dimensionless shape factor (typically taken as K = 0.9), λ is the X-ray wavelength, β is the peak width at half maximum intensity (FWHM) in radians, θ is the position at the center of the peak. [ 37–39 ] This analysis revealed that the out of plane (100) coherence length in the thicker fi lms was ≈20.4 nm on silicon, and ≈12.2 nm on graphene. The out of plane (010) face-on lamellae in P3HT had a coherence length of ≈20.7 nm on graphene. Therefore in the thicker fi lms, several ordered aggregates must be inter-connected from top to bottom of the fi lm in order to create a percolated pathway for vertical charge transport. Charges could be transported vertically if several face-on aggregates were con-nected to each other from top to bottom of the fi lm.

2.3. Crystallinity at the Top of Films

To answer this question, we performed 2D GIXD at a very shallow incident beam angle (0.08 °) which probes only the top (≈10 nm) of the thicker P3HT fi lm instead of the whole thickness (50 nm). The 2D diffraction pattern (Figure S3, Sup-porting Information) showed a similar crystalline orientation in the top layer compared to the whole fi lm on both substrates. Interestingly, the rings in (100) and (010) are also formed in the top layer on the graphene sample, which indicates that the π stacking also exists in all directions in the top layer on graphene. The ratio of diffracted intensities for the two peaks (010) and (100) were calculated to give an idea of the relative proportion of face-on to edge-on lamellae in the top portion of the fi lm and inside the whole fi lm on both substrates. This data is shown in Table 1 where the intensity ratio I(010)/I(100) was estimated at two χ angles (χ ≈ 8 ° near out of plane, and χ ≈ 86 ° in plane). From this analysis, it appears that in the 50 nm fi lm on graphene the proportion of in-plane π stacking is much less than on silicon, and there is a non- negligible amount of out of plane π stacking throughout the fi lm.



Figure 4. Crystallinity in the 50 nm thick P3HT fi lms at a shallow beam incident angle α = 0.13°. A,B) are 2D grazing incidence diffraction patterns of P3HT on A) silicon and B) graphene. The magnifi ed view in (B) shows the (100) ring in P3HT on graphene. C,D) show the variation of the integrated diffracted intensity as a function of the χ angle on C) silicon and on D) graphene.

Table 1. Intensity ratios I(010)/I(100) in the 50 nm (whole fi lm and top 10 nm), and the 10 nm thick P3HT fi lms. The χ angles are 8° (near out of plane) and 86° (in plane).

Thickness probed χ Silicon Graphene

10 nm 8° 0.02 0.10

10 nm 86° 0.31 0.46

50 nm (Top 10 nm) 8° 0.00 0.10

50 nm (Top 10 nm) 86° 3.28 0.61

50 nm (Whole fi lm) 8° 0.00 0.28

50 nm (Whole fi lm) 86° 3.07 0.25

FULL

PAPER


Remarkably, the intensity ratios are nearly identical (0.25 and 0.28) in the plane and out of plane on graphene, which confi rms the even spread of π stacking orientations at all angles observed previously. Moreover, the ultrathin (10 nm) fi lm and the top 10 nm of the thicker fi lm exhibits the same proportion of face–on to edge–on lamellae near z on graphene, but the top 10 nm of the thicker fi lm has ≈ 25 % less face–on lamellae than in the ultrathin 10 nm fi lm.

2.4. Creation of an Interconnected Path for Charge Transport

These results indicate that the strong enhance-ment in charge carrier mobility in the 50 nm fi lm on graphene is not due to more face-on orientations, but instead it is likely due to a mixture of face-on and edge-on aggregates which creates a continuous pathway (not nec-essarily the shorter path) for charges to move from top to bottom of the fi lm. It seems rea-sonable to assume that a high degree of inter-connection between crystalline regions, even small aggregates, is a necessary condition for effi cient charge transport, as outlined in a recent study. [ 40 ] Charges can indeed be trans-ferred from one crystalline region to another by a mechanism of charge hopping across amorphous regions, and charge mobility in P3HT has been shown to only decrease slightly when crossing a small number (5 or less) of grain boundaries. [ 41 ] The particular crystalline arrangement of P3HT observed on graphene, which consists of a mosaic of crystalline domains with a large spread of orientations, is likely to favor charge transport vertically across the whole fi lm thickness by forming a continuous pathway of interconnected crystallites and aggregates for charge transport.

Figure 5 depicts the difference in crystalline orientations, and possible charge transport pathways between the samples. On the silicon surface, P3HT forms mainly edge-on lamellae which are parallel to the surface in both the 10 nm and 50 nm fi lms. On the other hand, on graphene P3HT forms both edge-on and face-on crystallites, which help with vertical charge transport compared to only edge-on or face-on. The strong ordering of crystallites (sharp peaks along z) in the 10 nm fi lm is likely due to strong π interactions with the substrate, and to confi nement effects since the crystallite size is almost the same as fi lm thick-ness. However, in the 50 nm fi lm on graphene, several crystal-lites are necessary to bridge the bottom to the top of the fi lm. The crystallites next to the substrate are likely well ordered with face–on and edge–on oriented along z due to strong interactions with the graphene layer, as shown in the 10 nm fi lm. However, as the thickness increases there is less interaction with the sub-strate, and crystallites can adopt different orientations, giving rise to a spread of orientations. This spread creates an intercon-nected pathway for charge transport across the fi lm along the π–π stacking (and possibly also along the chain backbone) from

0 and 90° angles. From these results, it appears that graphene is a material of choice for obtaining crystalline orientations that favor vertical charge transport in P3HT, and it might help improve the effi ciency of organic and hybrid OLED and OPV devices.

3. Conclusion

In conclusion, we have shown that graphene induces the crys-tallization of both face-on and edge-on lamellae on its surface. In very thin layers (10 nm) the lamellae are mostly parallel to the plane, whereas in thicker (50 nm) fi lms, a large spread of crystalline orientations is formed. This particular crystalline mosaic with π–π stacking oriented homogeneously at various angles inside the fi lm favors the creation of a continuous pathway of interconnected crystallites for charge transport. Con-ductivity measurements showed a strong increase in mobility and charge transport in P3HT when deposited on graphene.

4. Experimental Section Materials : Poly-3-hexylthiophene (P3HT) (98% RR, Mw = 32 kD) was

purchased from American Dye Source Inc. Single layer graphene was produced by CVD on a copper foil and subsequently transferred to a solid substrate (silicon or glass). Details of the synthesis and transfer procedure of graphene, and other materials used can be found in the Supporting Information.



Figure 5. Schematics of charge transport in P3HT on A,C) silicon and B,D) graphene. A,C) P3HT crystalline aggregates on silicon are formed of mainly edge-on lamellae, which results in non-ideal vertical charge transport. B,D) On the graphene surface, aggregates form both edge-on and face-on lamellae. In the 10 nm fi lm, lamellae are well oriented with π–π stacking perpendicular to the plane of the fi lm, whereas in the thicker 50 nm fi lm, a mix of edge-on and face-on lamellae, with various orientations is present. The arrows in (B,D) represent effi cient vertical charge transport pathways across the fi lm. The close-ups show the preferential crystal-line orientation on both surfaces.

FULL P

APER

7wileyonlinelibrary.com© 2014 WILEY-VCH Verlag GmbH & Co. KGaA, WeinheimAdv. Funct. Mater. 2014, DOI: 10.1002/adfm.201403418


Sample Preparation : The P3HT fi lms were prepared as described in the Supporting Information. The fi lms were then annealed in inert nitrogen atmosphere at 170 °C for 15 min, followed by slow cooling (≈1 °C · min −1 ) to room temperature.

Sample Characterization : The samples were characterized by AFM (Veeco Multimode) for the determination of fi lm thickness, and by Raman (Renishaw InVia) and UV–vis spectroscopy (Perkin-Elmer) for determination of their optical and physical properties. Crystallinity of the P3HT fi lms was measured by 2D Grazing-incidence X-ray diffraction carried out at the Stanford Synchrotron Radiation Lightsource (SSRL) (Menlo Park, CA, USA). See Supporting Information for more details.

Supporting Information Supporting Information is available online from the Wiley Online Library or from the author.

Acknowledgements Portions of this research were carried out at the Stanford Synchrotron Radiation Lightsource, a Directorate of SLAC National Accelerator Laboratory and an Offi ce of Science User Facility operated for the US Department of Energy Offi ce of Science by Stanford University. Some sample preparation was performed at MyFab (Ångström Laboratory) at Uppsala University. The authors thank the Baltic Foundation, MyFab and the Kempe Foundation for fi nancial support. D.R.B. also thanks a Young Researcher Career Award from the Faculty of Science and Technology at Umeå University.

Received: September 30, 2014 Published online:

[1] K. S. Novoselov , A. K. Geim , S. V. Morozov , D. Jiang , Y. Zhang , S. V. Dubonos , I. V. Grigorieva , A. A. Firsov , Science 2004 , 306 , 666 .

[2] L. Wang , I. Meric , P. Y. Huang , Q. Gao , Y. Gao , H. Tran , T. Taniguchi , K. Watanabe , L. M. Campos , D. A. Muller , J. Guo , P. Kim , J. Hone , K. L. Shepard , C. R. Dean , Science 2013 , 342 , 614 .

[3] F. Bonaccorso , Z. Sun , T. Hasan , A. C. Ferrari , Nat. Photonics 2010 , 4 , 611 .

[4] B. Obradovic , R. Kotlyar , F. Heinz , P. Matagne , T. Rakshit , M. D. Giles , M. A. Stettler , D. E. Nikonov , Appl. Phys. Lett. 2006 , 88 , 142102 .

[5] Z. Liu , Q. Liu , Y. Huang , Y. Ma , S. Yin , X. Zhang , W. Sun , Y. Chen , Adv. Mater. 2008 , 20 , 3924 .

[6] L. Zhang , Y. Li , J. Shi , G. Shi , S. Cao , Mater. Chem. Phys. 2013 , 142 , 626 .

[7] Y. C. Lai , Y. X. Wang , Y. C. Huang , T. Y. Lin , Y. P. Hsieh , Y. J. Yang , Y. F. Chen , Adv. Funct. Mater. 2013 , 24 , 1430 .

[8] Y. Wu , X. Zhang , J. Jie , C. Xie , X. Zhang , B. Sun , Y. Wang , P. Gao , J. Phys. Chem. C 2013 , 117 , 11968 .

[9] J. Huang , D. R. Hines , B. J. Jung , M. S. Bronsgeest , A. Tunnell , V. Ballarotto , H. E. Katz , M. S. Fuhrer , E. D. Williams , J. Cumings , Org. Electron. 2011 , 12 , 1471 .

[10] G. Wang , J. Swensen , D. Moses , A. J. Heeger , J. Appl. Phys. 2003 , 93 , 6137 .

[11] H. Yang , T. Shin , L. Yang , K. Cho , C. Ryu , Z. Bao , Adv. Funct. Mater. 2005 , 15 , 671 .

[12] M. Surin , P. Leclère , R. Lazzaroni , J. D. Yuen , G. Wang , D. Moses , A. J. Heeger , S. Cho , K. Lee , J. Appl. Phys. 2006 , 100 , 033712 .

[13] D. Gargi , R. J. Kline , D. M. DeLongchamp , D. A. Fischer , M. F. Toney , B. T. O’Connor , J. Phys. Chem. C 2013 , 117 , 17421 .

[14] L. H. Jimison , S. Himmelberger , D. T. Duong , J. Rivnay , M. F. Toney , A. Salleo , J. Polym. Sci., Part B: Polym. Phys. 2013 , 51 , 611 .

[15] W. Porzio , G. Scavia , L. Barba , G. Arrighetti , S. Milita , Eur. Polym. J. 2011 , 47 , 273 .

[16] D. R. Cooper , B. D’Anjou , N. Ghattamaneni , B. Harack , M. Hilke , A. Horth , N. Majlis , M. Massicotte , L. Vandsburger , E. Whiteway , V. Yu , ISRN Condens. Matter Phys. 2012 , 2012 , 501686 .

[17] V. Yu , E. Whiteway , J. Maassen , M. Hilke , Phys. Rev. B 2011 , 84 , 205407 .

[18] A. C. Ferrari , J. C. Meyer , V. Scardaci , C. Casiraghi , M. Lazzeri , F. Mauri , S. Piscanec , D. Jiang , K. S. Novoselov , S. Roth , A. K. Geim , Phys. Rev. Lett. 2006 , 97 , 187401 .

[19] N. F. Mott , R. W. Gurney , Electronic Processes in Ionic Crystals Dover Publications , New York , 1964 .

[20] D. R. Barbero , N. Boulanger , M. Ramstedt , J. Yu , Adv. Mater. 2014 , 26 , 3111 .

[21] J. Obrzut , K. A. Page , Phys. Rev. B 2009 , 80 , 195211 . [22] M. Osaka , H. Benten , L. T. Lee , H. Ohkita , S. Ito , Polymer 2013 , 54 ,

3443 . [23] A. Novikov , Solid-State Electron. 2010 , 54 , 8 . [24] R. Yan , Q. Zhang , W. Li , I. Calizo , T. Shen , C. A. Richter ,

A. R. Hight-Walker , X. Liang , A. Seabaugh , D. Jena , H. Grace Xing , D. J. Gundlach , N. V. Nguyen , Appl. Phys. Lett. 2012 , 101 , 022105 .

[25] H. Yang , J. Heo , S. Park , H. J. Song , D. H. Seo , K. E. Byun , P. Kim , I. Yoo , H. J. Chung , K. Kim , Science 2012 , 336 , 1140 .

[26] M. G. Lemaitre , E. P. Donoghue , M. A. McCarthy , B. Liu , S. Tongay , B. Gila , P. Kumar , R. K. Singh , B. R. Appleton , A. G. Rinzler , ACS Nano 2012 , 6 , 9095 .

[27] B. R. Matis , J. S. Burgess , F. A. Bulat , A. L. Friedman , B. H. Houston , J. W. Baldwin , ACS Nano 2012 , 6 , 17 .

[28] J. Rivnay , S. C. B. Mannsfeld , C. E. Miller , A. Salleo , M. F. Toney , Chem. Rev. 2012 , 112 , 5488 .

[29] D. M. DeLongchamp , R. J. Kline , D. A. Fischer , L. J. Richter , M. F. Toney , Adv. Mater. 2011 , 23 , 319 .

[30] J. L. Baker , L. H. Jimison , S. Mannsfeld , S. Volkman , S. Yin , V. Subramanian , A. Salleo , A. P. Alivisatos , M. F. Toney , Langmuir 2010 , 26 , 9146 .

[31] Y. K. Lan , C. I. Huang , J. Phys. Chem. B 2008 , 112 , 14857 . [32] T. Salammal Shabi , S. Grigorian , M. Brinkmann , U. Pietsch ,

N. Koenen , N. Kayunkid , U. Scherf , J. Appl. Polym. Sci. 2012 , 125 , 2335 . [33] G. Scavia , W. Porzio , S. Destri , L. Barba , G. Arrighetti , S. Milita ,

L. Fumagalli , D. Natali , M. Sampietro , Surf. Sci. 2008 , 602 , 3106 . [34] E. Verploegen , R. Mondal , C. J. Bettinger , S. Sok , M. F. Toney ,

Z. Bao , Adv. Funct. Mater. 2010 , 20 , 3519 . [35] D. M. DeLongchamp , R. J. Kline , Y. Jung , E. K. Lin , D. A. Fischer ,

D. J. Gundlach , S. K. Cotts , A. J. Moad , L. J. Richter , M. F. Toney , M. Heeney , I. McCulloch , Macromolecules 2008 , 41 , 5709 .

[36] D. H. Kim , H. S. Lee , H. J. Shin , Y. S. Bae , K. H. Lee , S. W. Kim , D. Choi , J. Y. Choi , Soft Matter 2013 , 9 , 5355 .

[37] A. Salleo , R. J. Kline , D. M. DeLongchamp , M. L. Chabinyc , Adv. Mater. 2010 , 22 , 3812 .

[38] J. Rivnay , R. Noriega , R. J. Kline , A. Salleo , M. F. Toney , Phys. Rev. B 2011 , 84 , 045203 .

[39] W. Chen , M. P. Nikiforov , S. B. Darling , Energy Environ. Sci. 2012 , 5 , 8045 .

[40] R. Noriega , J. Rivnay , K. Vandewal , F. P. V. Koch , N. Stingelin , P. Smith , M. F. Toney , A. Salleo , Nat. Mater. 2013 , 12 , 1038 .

[41] E. J. W. Crossland , K. Tremel , F. Fischer , K. Rahimi , G. Reiter , U. Steiner , S. Ludwigs , Adv. Mater. 2012 , 24 , 839 .

Copyright WILEY-VCH Verlag GmbH & Co. KGaA, 69469 Weinheim, Germany, 2014.

Supporting Information

for Adv. Funct. Mater., DOI: 10.1002/adfm.201403418

Enhanced Vertical Charge Transport in a SemiconductingP3HT Thin Film on Single Layer Graphene

Vasyl Skrypnychuk, Nicolas Boulanger, Victor Yu, MichaelHilke, Stefan C. B. Mannsfeld, Michael F. Toney, and DavidR. Barbero*

Supporting Information

for Adv. Funct. Mater., DOI: 10.1002/ad m.

Enhanced vertical charge transport in a semiconducting P3HT thin film on sin-

gle layer graphene

Vasyl Skrypnychuk, Nicolas Boulanger, Victor Yu, Michael Hilke, Stefan C. B. Mannsfeld,

Michael F. Toney and David R. Barbero∗

1 Graphene layer preparation

Large-area single layer graphene were synthesized on a copper foil by chemical vapour depo-

sition (CVD) using a vertical quartz tube.[1] The copper was pre-treated in acetic acid for

at least 30 minutes to smoothen the ridges of the foil. This step is necessary to minimize

structural defects in the graphene growth by forming less nucleation sites.[2] This was fol-

lowed by a brief clean in acetone and isopropanol. Next, it was annealed for 90 minutes in

a flow of 10 sccm H2 at 1050 C and a pressure of 120 Torr. A flow of 0.15 sccm CH4 was

introduced for 30 minutes while maintaining the same H2 flow. The tube was removed from

the oven and cooled rapidly to room temperature under H2 flow. The graphene on copper

was characterized by Raman, which shows full monolayer coverage.[3] The sample was then

spin-coated with PMMA, and the copper was etched in ammonium persulfate. Once fully

etched, the graphene was placed in a 10% HCl solution for 10 mins to remove remaining

Cu residues. Finally, the graphene was transferred in DI water before being placed onto a

1

201403418f

substrate. The transfer of graphene on hydrophobic substrates is more challenging compared

to conventional hydrophilic substrates like SiO2/Si and glass.[4] For a successful transfer, di-

luted isopropanol is used to reduce the repulsive force between the liquid and the substrate.

A mix of 35 mL of DI water and 15 mL of IPA helps minimize the repulsion between the

liquid and substrate. It provides enough surface tension to keep the graphene floating and to

prevent the graphene from wrinkling. The hydrophobic substrate was placed at the bottom

of the beaker, and the water was slowly removed to let the graphene settle onto the substrate.

The sample was dried overnight before removing the PMMA with acetone.

2 P3HT film preparation

Highly P-doped (100) silicon wafers (resistivity 0.001–0.002 Ω · cm; University Wafer) with

a thin native oxide layer were cleaned by ultrasonication in acetone and IPA, blown dry

with nitrogen, and cleaned once more in a stream of carbon dioxide ice crystals (snow-jet) to

remove dust particles. The same procedure was used to clean the glass samples. A freshly

prepared solution of P3HT 0.5% (mass.) in 50/50 (vol.) chloroform/o-dichlorobenzene mix-

ture was filtered through a syringe filter (pore size 0.2 µm) and used for the film deposition.

The thin P3HT films on silicon, graphene-coated silicon (graphene/Si), glass and graphene-

coated glass (graphene/glass) were prepared by spin coating at 1000 revolutions per minute

(rpm) for 90 s (until the films were dry), resulting in films of 50–55 nm thick. The films were

then annealed in inert nitrogen atmosphere at 170 C for 15 min, followed by slow cooling

(≈1 C ·min−1) to room temperature.

2

3 Conductivity measurement

The electrical conductivity of the samples was measured using a home-made flexible elec-

trode in order to obtain good conformal contact with the film, and accurate measure of

the distance between the two electrodes (Figure S2). The flexible and conducting electrode

was made of a smooth polydimethylsiloxane (PDMS) layer covered by silver to make the

surface conductive. The liquid PDMS solution was mixed on a 10:1 ratio with its precursor,

homogenized, degasified in a vacuum desiccator, and cured at a temperature between 150 C

for 5 hours. Then, a square ∼4.2 × 4.2 mm electrode was cut out of the cured PDMS layer,

covered with a layer of silver paste and smoothened by putting in contact with a smooth

Si wafer. Polydimethylsiloxane (PDMS) was purchased from Dow Corning Corp. (Sylgard

R© 184). A voltage was applied with an external power source (GW GP3030) between the

top and the bottom of the films, and the current was measured using an ammeter (Agilent

34410A). The measurements of conductivity were made using equation 1 in the manuscript,

and required knowledge of the surface area in contact with the top electrode, the height of

the patterns, the voltage applied and the current measured. The conductivity was measured

using equation 1 in the manuscript, and required knowledge of the surface area in contact

with the top electrode, the film’s thickness, the voltage applied and the current measured.

The film thickness was measured within less than 1–2nm by AFM (less than 4% error of

measurement). The accuracy of measurement on the current measured and voltage applied

are within 1% each. The overall error of measurement on the conductivity, including the

error on the area covered by the electrode, was calculated to be less than 15%.

3

1.6 2.0 2.4 2.8 3.20

20

40

60

80

100Absorbance(a.u.)

Photon energy (eV)

P3HT / glasstotal absorbanceaggregated chains

1.6 2.0 2.4 2.8 3.20

20

40

60

80

100

Absorbance(a.u.)

Photon energy (eV)

P3HT / Graphene-glasstotal absorbanceaggregated chainsA B

2.057 eV 2.046 eV

Figure S1: UV–vis absorbance of the P3HT films on glass (a) and on graphene (b). An anal-ysis using the Spano model shows a fit of the region corresponding to crystalline aggregatesin the films, and the small downshift in maximum absorbance observed on the sample ongraphene.

4 Characterization by Raman, UV–vis Spectroscopy

and grazing incidence X-ray diffraction

Optical spectra were acquired in transmittance mode using a Perkin-Elmer UV–vis spectrom-

eter equipped with an integrating sphere detector (Labsphere Inc.). The Raman spectra were

collected using a Renishaw InVia microscope with 488 nm laser excitation, 1 s exposure time

and 50 accumulations for each measurement. The UV-Vis spectra were converted to eV scale

and interpolated using the Spano model [5, 6]. Based on this model, the energy of the E0−0

transition, as well as the contribution of the aggregated chains to the absorbance spectra

was calculated for each of the samples. The UV vis absorbance of the films deposited on

both substrates (Figure S1) revealed a small upshift in the absorbance of P3HT on graphene

from 602.8 nm to 606.1 nm, and a corresponding decrease in the energy band gap from 2.057

eV on glass to 2.046 eV on graphene. This likely suggests that the P3HT chains are more

planar, and that their conjugation length increases on the graphene surface.

Following an analysis by Gao et al., we performed resonant Raman spectroscopy and we

4

1 3 5 0 1 4 0 0 1 4 5 0 1 5 0 00 . 0

5 . 0 x 1 0 31 . 0 x 1 0 41 . 5 x 1 0 42 . 0 x 1 0 42 . 5 x 1 0 4 P 3 H T / g r a p h e n e

Inten

sity (a

.u.)

W a v e n u m b e r ( c m - 1 )

P 3 H T / S i

Figure S2: Raman absorbance of the P3HT films on silicon and on graphene. A higher ratiobetween the C-C band (1380 cm−1) and the C=C band (1445 cm−1) is observed for thesample on graphene.

estimated the ratio of double vs. single C bonds symmetric stretching modes in both films.[7]

The data shown in Figure S2 gave a slightly higher ratio on silicon (0.143 vs 0.140), which

indicates more planar chains on graphene and confirms the results obtained by UV–vis. The

Raman bands of P3HT in the range 1300-1500 cm−1 were interpolated with a Gaussian

function. Namely, the C-C symmetric stretching band (around 1380 cm−1) and the C=C

symmetric stretching band (around 1445 cm−1).

2D Grazing-incidence X-ray diffraction (2D GIXD) experiments were carried out at Stan-

ford Synchrotron Radiation Lightsource (SSRL) (Menlo Park, CA, USA) using beamline

11-3. The beam energy was 12726 eV, and the grazing incidence angle was set to 0.13 and

0.08 (Figure S3). Figure S3 shows the crystallinity at the top of the 50 nm films.

References

[1] D. R. Cooper, B. D’Anjou, N. Ghattamaneni, B. Harack, M. Hilke, A. Horth, N. Majlis,

M. Massicotte, L. Vandsburger, E. Whiteway, V. Yu, ISRN Condens. Matter Phys. 2012,

5

Si 50 nm Graphene 50 nm

A B

C D

300

200

100

qZ (Å-1)

qxy (Å-1)

2

1.5

1

0.5

02 1 0

300

200

100

qZ (Å-1)

qxy (Å-1)

2

1.5

1

0.5

02 1 0

010

0 30 60 90

χ (deg)z axis xy plane

G

2°

Si

0 30 60 90

χ (deg)z axis xy plane

Si

G

8° 50 nmα=0.08°

102

103I 010

(a.u.)

101

102

103

104

105

I 100

(a.u.)

50 nmα=0.08°

010

Figure S3: Crystallinity in the 50 nm P3HT films at a very shallow beam incident angleα=0.08 . Figures (a) and (b) are 2D grazing incidence diffraction patterns of P3HT on(a) silicon, and (b) graphene. Figures (c) and (d) show the variation of the integrateddiffracted intensity as a function of the χ angle on silicon (c), and on graphene (d). Notethat intensities were normalized to a 50 nm thick film in (c) and (d) for easier comparisonwith Figure 4.

6

2012, 501686.

[2] M. Massicotte, V. Yu, E. Whiteway, D. Vatnik, M. Hilke, Nanotechnology 2013, 24 (32),

325601.

[3] V. Yu, E. Whiteway, J. Maassen, M. Hilke, Phys. Rev. B 2011, 84, 205407.

[4] S. Bernard, E. Whiteway, V. Yu, D. G. Austing, M. Hilke, Phys. Rev. B 2012, 86,

085409.

[5] F. C. Spano, J. Chem. Phys. 2005, 122 (23), –.

[6] J. Clark, J.-F. Chang, F. C. Spano, R. H. Friend, C. Silva, Appl. Phys. Lett. 2009,

94 (16), –.

[7] Y. Gao, J. K. Grey, J. Am. Chem. Soc. 2009, 131 (28), 9654–9662.

7

enhanced vertical charge transport ...hilke/publications/pdf... · when in contact with a...

Documents