Self-assembled monolayers of alkyl-thiols on InAs: A Kelvin probe force microscopy study

Surface Science 633 (2015) 53–59

Contents lists available at ScienceDirect

Surface Science journal homepage: www.elsevier.com/locate/susc

Self-assembled monolayers of alkyl-thiols on InAs: A Kelvin probe force microscopy study A. Szwajca a,b,⁎, J. Wei a,c, M.I. Schukfeh a, M. Tornow a,d,⁎ a

Institut für Halbleitertechnik, Technische Universität Braunschweig, 38106 Braunschweig, Germany Adam Mickiewicz University in Poznań, 61-614 Poznań, Poland Institute of Nanochemistry and Nanobiology, Shanghai University, 200444, Shanghai, China d Department of Molecular Electronics, Technische Universität München, 80333 München, Germany b c

a r t i c l e

i n f o

Article history: Received 29 September 2014 Accepted 25 November 2014 Available online 3 December 2014 Keywords: Self-assembled monolayers Thiols InAs Surface analysis Kelvin probe force microscopy

a b s t r a c t We report on the preparation and characterization of self-assembled monolayers from aliphatic thiols with different chain length and termination on InAs (100) planar surfaces. This included as first step the development and investigation of a thorough chemical InAs surface preparation step using a dedicated bromine/NH4OHbased etching process. Ellipsometry, contact angle measurements and atomic force microscopy (AFM) indicated the formation of smooth, surface conforming monolayers. The molecular tilt angles were obtained as 30 ± 10° with respect to the surface normal. Kelvin probe force microscopy (KPFM) measurements in hand with Parameterized Model number 5 (PM5) calculations of the involved molecular dipoles allowed for an estimation of the molecular packing densities on the surface. We obtained values of up to n = 1014 cm−2 for the SAMs under study. These are close to what is predicted from a simple geometrical model that would calculate a maximum density of about n = 2.7 × 1014 cm−2. We take this as additional conformation of the substrate smoothness and quality of our InAs–SAM hybrid layer systems. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Self-assembled monolayers (SAMs) of functional molecules grafted to solid surfaces have raised much interest in recent decades due to their manifold applications in surface protection against corrosion, as bio-sensing interface, for hybrid photovoltaics or in organic electronics [1–5]. While there exists a large amount of experimental work on SAMs on metals [6], in particular thiols on Au, less work has been devoted to the study of the electronic properties of SAMs deposited on semiconductors. Exploring these substrate materials is of growing interest due to their direct relevance in organic–inorganic integrated microelectronics, offering the possibility to tailor their properties over a larger range, via the doping level (conductance, work function) and material composition (band gap). Among the reported work on hybrid SAM/ semiconductor systems the most prominent semiconductors include silicon or Si/SiO2 and certain compound semiconductors such as GaAs [7–9]. From the perspective of using the semiconductor as contact in molecular electronics the narrow band gap semiconductor indium arsenide (InAs) is one material with exceptional electronic properties. The ⁎ Corresponding authors at: Adam Mickiewicz University in Poznań, 61-614 Poznań, Poland and Technische Universität München, 80333 München, Germany. E-mail addresses: [email protected] (A. Szwajca), [email protected] (M. Tornow).

http://dx.doi.org/10.1016/j.susc.2014.11.023 0039-6028/© 2014 Elsevier B.V. All rights reserved.

surface Fermi level is usually pinned in the conduction band, leading to a surface electron accumulation or inversion layer [10]. Therefore, at the surface, an InAs substrate usually features 2D quasi-metallic properties. In a recent study, we have made use of this property to investigate the electronic properties of a novel type of ‘all-semiconductor’, nanoelectronic device structure, made from InAs/InP heterostructure nanowires that were functionalized with oligo(phenylene-vinylene) molecular wires [11]. Thiolate based SAMs have been studied on InAs surfaces before, partly arising from the long-known possibility to electrically passivate InAs surfaces with sulfur chemistry [12–15]. For the characterization of alkylthiol SAMs on InAs mostly X-ray photoelectron spectroscopy (XPS), infrared (IR), atomic force microscopy (AFM) and contact angle (CA) measurements have been reported [16–18]. These studies have pointed to a prevailing covalent bond of the molecule terminal sulfur to the surface indium atoms. Here, we demonstrate the successful coating of (100) InAs surfaces with alkylthiol molecules over a range of chain lengths, and for different terminal groups, after removal of the native oxide layer present at the surface. We characterized the monolayers' topological and electronic properties using complementary analytical techniques including ellipsometry, contact angle, AFM (nanolithography) and in particular Kelvin probe force microscopy (KPFM). Combining different experimental techniques with model simulations allowed for an estimation of the molecular packing densities on surface.

54

A. Szwajca et al. / Surface Science 633 (2015) 53–59

2. Materials and methods 2.1. Materials Undoped, (100) oriented InAs one-side polished wafers were purchased from Wafer Technology Ltd. (UK). 1 cm × 1 cm samples were cleaved off the wafer. Backside ohmic contacts were prepared by e-beam evaporation of a Cr (10 nm)/Au (300 nm) thin film. 1-Dodecanethiol (HS\(CH2)11\CH3, abbreviated “C12”, purity 98%), 1-tetradecanethiol (HS\(CH2)13\CH3, “C14”, 98%), 1-octadecanethiol(HS\(CH2)17\CH3, “C18”, 98%), 1,9nonanedithiol (HS\(CH2)9\SH, “C9SH”, 95%), and mercaptohexanol (HS\(CH2)6\OH, “C6OH”, 97%) were purchased from Aldrich. 1-Hexanthiol (HS\(CH2)5\CH3, “C6”, 95%) and 1-decanethiol (HS\(CH2)9\CH3, “C10”, 96%) were purchased from Fluka. All alkanethiols were used without further purification. Anhydrous solvents (CH2Cl2 methylene chloride, CH3CH2OH ethanol and CH3CH2OHCH3 isopropanol) were de-oxygenated by bubbling a stream of nitrogen through them. 2.2. Monolayer formation InAs samples were sonicated in acetone, iso-propanol and methanol for 10 min each. Different recipes for oxide removal were tested (see, Supporting Information). Among these, a novel bromine/NH4OHbased etch was found to yield optimal surface conditions: the freshly cleaned samples were etched in a 2% bromine solution in deoxygenated water for 2 s, rinsed in boiling water, then dipped for 60 s in a 10% NH4OH solution and finally rinsed in redistilled water. Finally, the samples were dried under a flow of nitrogen (N2). For molecular assembly, the freshly etched InAs samples were immersed in a 5 mM ethanol solution of the appropriate alkanethiol with a small addition (10%) of dichloromethane to improve solubility and one drop (0.2 μl) of NH4OH [19]. Chemical adsorption was allowed for 24 h in a 5 ml glass vial. Following organic deposition, the samples were rinsed with ethanol and isopropanol and dried under a stream of N2. Removal of the native oxide and SAM formation on InAs samples took place in a nitrogen flushed glovebox. 2.3. Ellipsometry and contact angle measurements Monolayer thicknesses were measured with an EL X-02C rotatinganalyzer ellipsometer (DRE Dr. Riss, Ratzeburg) equipped with a HeNe laser (632.8 nm) at an angle of incidence of 70°. For data analysis, a three-layer (SAM/InAs-oxide/InAs) model was used. We assumed refractive indices of n = 1.45 [20] for the monolayers and n = 1.74 for InAs-oxide [21]. Layer thicknesses were obtained by averaging five measurement points for both the bare and the SAM-covered InAs samples. Water contact angle measurements were made using an OCA 15 plus instrument (DataPhysics, Stuttgart) on 1–2 samples per functionalization, and five angles per sample were taken at different locations on each sample, to be averaged. For this purpose, a drop (0.2 μl) of water was deposited using a microsyringe, and the contact angle was measured approximately 30 s after deposition of the water drop. 2.4. AFM/KPFM Atomic force microscopy (AFM) measurements were carried out in ambient air, using a Veeco Dimension V equipped with a Nanoscope V controller in tapping or contact mode, using Si cantilever tips (Veeco Instruments; OTESPA7). AFM pictures were taken at at least three different positions on each sample. Kelvin probe force microscopy (KPFM) measurements were carried out with the same instrument using Pt/Ir-coated silicon cantilever tips (PPP-EFM-tip, Nanosensors). Topography and KPFM data were obtained simultaneously using a

standard two-pass procedure, where a topographic line is first acquired in tapping mode and a KPFM line is secondly acquired in a lift mode. In the lift mode, the tip scans at a constant distance of 50 nm above the sample surface to ensure that electrostatic forces are dominating. In KPFM mode, the applied AC voltage has an amplitude of 800 mV at a frequency close to the resonance frequency of the cantilever (about 70 kHz). KPFM images of the sample surface were acquired at a probe scan rate of 2 Hz. For calibration, the contact potential difference VCPD was at first measured between the tip and a substrate with known work function (gold). This way, the work function of the tip was determined. The subsequently measured, averaged VCPD values between the tip and the bare and functionalized InAs samples were used to calculate their respective work functions. 2.5. PM5 calculation Dipole moment vectors (absolute value and angle with respect to molecular main axis) of all alkyl thiolates were estimated by carrying out a PM5 (“Parameterized Model number 5”) calculation (MO-G Version 1.1, Fujitsu Limited, Tokyo, Japan, 2008) using the Scigress software (Fujitsu Ltd.) on a Windows workstation. Each surface-bound thiolate was approximated by replacing the distal H atom of the thiol with an In atom, resulting in an imaginary In\S\R molecule, where R represents the alkyl chain including the terminal group of the molecule under study. The prevailing presence of In\S rather than As\S bonds on InAs surfaces that have been modified with alkanethiols has been described in the literature [12,15,16,19,22–24]. Every molecular structure under study was initially optimized with respect to its geometrical conformation. 3. Results and discussion 3.1. Surface oxide removal and SAM formation III–V compound semiconductors are known to easily react with atmospheric oxygen, forming a thin native surface oxide [25]. The oxide layer of InAs generally consists of mixed In-oxides and As-oxides [26]. The formation of thiol SAMs on the bare InAs surface requires the thorough removal of these native oxides before monolayer growth, preferably resulting in a flat, defect-free substrate surface. We have carried out all surface processing steps such as etching and SAM formation with deoxygenated solvents and in nitrogen atmosphere to prevent immediate re-growth of an oxide layer. Different methods for the removal of InAs oxides have been reported, while many of them, especially those based on HCl, H2O2–Br2, Br2–HBr or HF would lead to the formation of layered or structural defects [14,27–29]. Our own experiments with HCl led to similar effects, see Supporting Information. Surface preparation by etching in a 0.1% bromine solution in methanol for ~ 3 min was reported to result in a measured surface roughness of 0.54 nm over 2 × 2 μm2 [30]. We have obtained best results by using a dilute solution of NH4OH and 2% Br2 in water as etchant. Such alkaline solution may remove the oxides from the sample surface without residual side products. The use of bromine solution endows the surface with a polishing effect, leading to a very low surface roughness (rms value) of 0.11 nm over a 5 × 5 μm2 area as determined by AFM. Fig. 1a) displays the results of such AFM analysis for the bare, freshly etched InAs surface. The surface was found to be smooth and free from etch pit formation even for large areas, as confirmed by SEM images over several 100 μm (Fig. S.I.3, Supporting Information). Ellipsometry measurements using a two-layer InAs-oxide/InAs model verified the removal of the native InAs-oxide within experimental error (0.1 ± 0.1 nm). The bromine/NH4OH recipe was used throughout the paper in advance to all SAM depositions.

A. Szwajca et al. / Surface Science 633 (2015) 53–59

55

Fig. 1. AFM images (left) and averaged line profiles (right) of a) an InAs surface, freshly etched using the recipe described in the text, and b) an InAs surface prepared in the same way, after subsequent coating with a C6OH SAM. The positions and averaging regions of both line scan profiles are indicated as white dashed line boxes in the respective images.

3.2. Surface analysis 3.2.1. Contact angle and ellipsometry measurements The contact angles (CAs) measured on all monolayer coated InAs samples are listed in Table 1, while two exemplary water droplet images used for the measurement are displayed in Fig. 2. For the alkylthiol

SAMs CAs were around or even slightly above 100° (Fig. 2b), confirming the formation of a relatively hydrophobic surface. SH- and OH-group termination lowered the CAs down to 66° and 40° (Fig. 2a), respectively, in agreement with the expectation for polar head groups. Ellipsometry data have been used to calculate the thickness of the SAMs on the InAs substrates. All measured thicknesses (Table 1) were

Table 1 Summary of characterization data for all studied thiol-SAM coated InAs samples, together with one InAs bare reference. The (theoretical) molecular contour length was determined using a PM5 calculation (see Methods section). SAM thickness was measured by ellipsometry, and based on this the molecular tilt angle α was estimated (cf. Fig. 6). The theoretical, absolute value for the dipole moment μ and the dipole angle (between dipole vector and molecular main axis) was obtained using PM5 simulation of the respective molecule with In\S terminal group. The estimated, mean component of the dipole moment perpendicular to the surface was used to extract the molecular density n, according to the Helmholtz formula (Eq. (1)). Substrate

Φ/e ΔΦSAM/e μ (D) Dipole μ⊥ (D, ±0.2 D) n (1014 cm−2, Molecule SAM thickness Molecular Contact VCPD (V, ±0.005 V) (V, ±0.005 V) (V, ±0.005 V) angle ±0.06 × 1014 cm−2) contour length (nm, ±0.1 nm) tilt angle angle β(°) (nm) α(°, ± 10°) (°, ±5°)

InAs InAs\C6 InAs\C10 InAs\C12 InAs\C14 InAs\C18 InAs\C6OH InAs\C9SH

– 1.0 1.4 1.7 2.0 2.5 1.1 1.5

– 0.8 1.3 1.6 1.8 2.3 1.0 1.4

– 37 22 20 26 23 25 21

98 100 101 101 106 40 66

0.550 0.410 0.390 0.405 0.440 0.335 0.180 0.360

4.460 4.610 4.630 4.615 4.580 4.685 4.840 4.660

– 0.140 0.160 0.145 0.110 0.215 0.370 0.190

– 4.58 4.58 4.58 4.58 4.58 3.70 6.59

– 57 57 60 58 57 40 60

– 2.0 2.3 2.2 2.2 2.3 2.6 3.1

– 0.50 0.50 0.48 0.36 0.67 1.03 0.44

56

A. Szwajca et al. / Surface Science 633 (2015) 53–59

Fig. 2. Photographs of water droplets on the surfaces of two InAs/SAM samples, used for determination of the contact angles: a) C6OH (CA = 40°) and b) C18 (CA = 106°).

found smaller than the respective molecule contour lengths, indicating tilt angles α of the molecules with respect to the surface normal in the range of 20–40° (Table 1) which is typical for alkylthiol SAMs, as reported on various substrates [6]. 3.2.2. AFM measurements AFM topography measurements indicated smooth and surface conforming monolayers, with little to no multilayer formation. The surface roughness increased only slightly with respect to the bare InAs surface, typically to about 0.3 nm (see, Fig. 1b for the example of a C6OH SAM). Successful monolayer formation was further verified by AFM nanolithography for some of the samples: A 5 × 5 μm2 surface was first scanned in tapping mode. Then, a 1 × 1 μm2 square within the pre-scanned area was ‘scratched’ free of molecules by using contact mode, thereby moving the molecules to the square edges. Finally, the initial scanning area was measured again in tapping mode, imaging the exposed, free InAs square, cf. Fig. 3a) for a C6 SAM. Averaged line scans (height profiles) allowed for an alternative verification of the molecular layer thickness with respect to the clean InAs surface (Fig. 3b). For the shown example, a C6 film thickness of around 0.9 nm was estimated, in good agreement to the result obtained by ellipsometry. 3.2.3. Contact potential difference measurements All contact potential differences reported in this work were measured with respect to the Pt/Ir tip used for KPFM. For this reason, we adopt hereafter the notation VCPD (sample) for a contact potential difference between this tip and a sample under study. At first, based on the measured contact potential difference VCPD (Au) between the tip and a clean Au surface (ΦAu = 5.1 eV) [31], we calibrated our KPFM measurements by determining the tip work function as Φtip = ΦAu + VCPD(Au) = 5.1 eV ‐ (90 ± 5) meV = (5.010 ± 0.005) eV.1 This work function of the tip served as reference to determine the work functions of the (coated) InAs samples, by measuring the respective contact potential differences VCPD, see Fig. 4. This way, we obtained the work function of a freshly etched InAs bare sample as ΦInAs = Φtip − VCPD(InAs) = 4.460 ± 0.005 eV, in excellent agreement to the value reported in Ref. [34]. We hereby anticipate that our freshly etched InAs reference sample had not formed any substantial native oxide layer within the time of CPD measurement, as confirmed by ellipsometry measurements (see, Section 3.1). After SAM deposition, the work function of the functionalized InAs sample had increased with respect to the bare InAs substrate by up to 370 meV for InAs/C6OH. Fig. 5 shows both, a full VCPD areal image scan—plotted as variation from the average value all over that area— together with a VCPD line scan for this sample. The average VCPD values for all samples are listed in Table 1. When comparing the individual values for different terminal groups, the largest differences appear between the nonpolar CH3, and the

1 We note that the real work function of our gold reference surface may differ by up to a few 100 meV from the assumed value ΦAu = 5.1 eV, because measurements were done in ambient air [32,33]. However, this would only result in a constant offset for all extracted absolute work function values. This in turn would not affect our estimation of the molecular densities, because for that calculation, only relative changes of work functions were used.

mentioned OH terminal group. E.g., a relative difference of 230 meV is measured between InAs/C6 and its analog InAs/C6OH. Together with all VCPD mean values, the extracted work functions Φ and the relative changes in work function ΔΦSAM with respect to the bare InAs are listed in Table 1, too. 3.2.4. Molecular dipoles and surface densities In order to extract information about the molecular surface density we set out to relate the measured change in work function to the surface dipole layer that has induced it. As a first approximation, we have calculated the dipole moments of the individual alkylthiols with an In\S termination, using PM5 simulations. The obtained dipole moments μ (absolute values) are listed in Table 1. All dipole vectors were found to be oriented towards the surface (pointing from a negative to a positive center of charge) in agreement with our finding that the work function of all SAM coated samples has increased with respect to bare InAs. This increase of the work function is in contrast to previous reports on related systems involving (methyl-terminated) alkyl-thiol monolayers on, e.g., GaAs [35], gold [36] or silver [37], which appears to originate from a contrary interface dipole formation at the S\In bond: our alternative simulations of molecules with As\S terminal groups, or of free thiols, resulted in dipole moments pointing in the opposite direction. As expected, a zero dipole moment resulted for all symmetric molecules. The molecular dipoles of the SAM can be related to the change in work function of the coated substrate according to the Helmholtz equation [38,39]. jΔΦSAM j ¼

μ⊥ Aε SAM ε 0

ð1Þ

Here, A is the area per molecule, εSAM = 2.7 [40] is the relative permittivity of the monolayer, and ε0 is the permittivity of vacuum. We note that recent theoretical studies have pointed to intriguing dependencies of the permittivity itself from the area per molecule, owing to molecular depolarization effects [41,42]. Taking these extensions of Eq. (1) into account however would exceed the scope of this paper. Further in Eq. (1), μ⊥ is the normal component of the molecular dipole, hence its projection perpendicular to the surface. To extract this value, we first calculated the projection of the dipole absolute value along the molecular axis, i.e., μ cos (β), cf. Fig. 6, to obtain an averaged value. Depending on the rotational angle of the molecule around its anchoring point, all dipole orientations along a cone surface could be equally possible. Finally, we calculated the normal component of this projected dipole with respect to the surface, by taking into account the estimated tilt angle α: μ⊥ = μ cos (β) cos (α), cf. Table 1. Based on the calculated dipole moments (normal component) and the measured changes in work function ΔΦSAM induced by the monolayer coatings we estimated the molecular densities as n = 1/A using Eq. (1). The obtained surface densities are all of the order of a few 1013 cm− 2, with a maximum value for C6OH of about 1014 cm−2. Hence, most values are by about one order of magnitude smaller than the total density of (100) surface indium atoms of 5.44 × 1014 cm−2, as based on—in the simplest approximation—the truncated bulk InAs zinc blende structure with lattice constant 6.06 Å. InAs surfaces have

A. Szwajca et al. / Surface Science 633 (2015) 53–59

57

Fig. 3. a) AFM image of an InAs surface after deposition of a C6 monolayer. A 1 μm × 1 μm square was scratched free from molecules, which have visibly piled up at the edges of the square. b) Height profile line scan, averaged over the area within the white dashed line box of a). The red arrows indicate the step height (SAM vs. cleaned InAs). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

been studied before to investigate the interplay of lattice geometry and atom spacing with (organic) adsorbate layer formation on top [34,43]. The chemical adsorption of alkanethiols on the surface of III–V materials is an interesting structural problem in terms of steric commensurability. Knoben et al. reported a relatively large molecular density of 7.3 × 1014 cm−2 for SAMs of octadecanoic acids deposited on InAs and related this finding to possible unspecific molecule binding and surface roughness effects [43]. For an ideally flat, (100) truncated bulk InAs surface, the spacing between two neighboring indium atoms (4.29 Å) is too small to accommodate for an alkylthiol binding to every indium atom, considering the Van-der-Waals radius of alkyl chains of ca. 2.3 Å. Rather,

a commensurate super-lattice of alkyl chains could be formed by assuming molecules that would occupy at least every second indium site. This yields a theoretical maximum density of n2nd = 2.73 × 1014 cm−2. All our estimated densities (Table 1) are a factor of 3–8 smaller than predicted from this simple geometrical argument. To keep the inter-molecular distance as small as possible (= maximum packing) the lattice-imposed, minimal distance of molecules in SAMs is usually compensated by their tilt from the surface normal, which is additionally influenced by the terminal groups of the molecules. In our studies, we do not observe any expected increase of molecular density for smaller tilt angles though, within the experimental uncertainty range for α of ±10°. A number of reasons may be responsible for our findings: 1) we have not taken into account any, most likely present, surface reconstruction on the InAs (100) surface [44]. This would require a more detailed knowledge and analysis of the microscopic InAs surface following our chemical surface treatment (oxide removal). 2) A residual oxide layer may still be present which might have started forming at possible point defects in the layer, immediately after having removed the samples from the glovebox to measure their KPFM profiles. Such oxide layer may both result in a direct variation of SAM density or induce shifts in the work function changes ΔΦSAM. 3) The effective ε of the monolayers may be larger than assumed from literature values obtained on other substrates. 4) The real molecular dipole moments may be well smaller than estimated by our simulations which modeled an imaginary, “free” In\S\R molecular structure. Notwithstanding the simplicity of our model and a number of assumptions we note that the agreement of our quantitative estimations with our experimentally obtained values is still remarkably good. 4. Conclusions

Fig. 4. Energy levels for the InAs/SAM system, illustrating qualitatively the work functions of bare InAs (left), of a SAM coated InAs sample with increased work function (center), and of the reference Pt/Ir tip (right). Using KPFM the difference in work function between tip and sample is measured as contact potential difference, as illustrated for both samples.

We have carried out a systematic study of different alkylthiolate monolayer films on InAs surfaces that were prepared based on a novel oxide removing, bromine/NH4OH-based etchant method. Thickness determination, together with KPFM measurements and theoretical model calculations of the molecular dipoles allowed for an estimation of the molecular surface density. Our findings indicate the presence of a smooth InAs substrate functionalized with surface conforming monolayers which reach packing densities close to the geometrical expectation for an ideally flat surface, from a basic geometrical model. In future studies, we plan to include XPS (X-ray photoelectron spectroscopy) and XRR (X-ray reflectivity) measurements to further support our

58

A. Szwajca et al. / Surface Science 633 (2015) 53–59

a)

b)

VCPD (mV)

190

0

5 µm

180

0

1500

3000

4500

Scan length (nm)

Fig. 5. (a) Surface potential difference (VCPD) areal image scan—plotted as variation from the average value all over that area (=180 mV), and (b) line scan profile of the total value of VCPD, both for an InAs/C6OH sample.

conclusions of both, the clean, oxide free InAs surface after chemical etching and the quality of the formed monolayers. We believe that our molecular functionalization strategy may well find applications in InAs and related III–V device applications including optics, sensors and molecular electronics.

Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.susc.2014.11.023.

References Acknowledgments This work was partially financed by the BMBF under grant 03X5513 (Junior Research Group ‘NanoFutur’). Part of this work was supported by the National Science Center Poland (grant N N204 444740 to A. Sz.).We thank V. Bandalo for useful discussions and D. Rümmler for help with sample preparation.

Fig. 6. Sketch of the molecular structure of the model molecule C6OH bound to the InAs surface, illustrating the molecular tilt angle α, the direction and magnitude of the molecular dipole moment μ, indicated as arrow, and the angle of this dipole vector with respect to the molecular main axis, β.

[1] C. Vericat, M.E. Vela, G. Benitez, P. Carro, R.C. Salvarezza, Chem. Soc. Rev. 39 (2010) 1805. [2] F. Davis, S.P.J. Higson, Biosens. Bioelectron. 21 (2005) 1. [3] H. Ma, H.L. Yip, F. Huang, A.K.Y. Jen, Adv. Funct. Mater. 20 (2010) 1371. [4] N. Koch, ChemPhysChem 8 (2007) 1438. [5] D. Astruc, E. Boisselier, C. Ornelas, Chem. Rev. 110 (2010) 1857. [6] J.C. Love, L.A. Estroff, J.K. Kriebel, R.G. Nuzzo, G.M. Whitesides, Chem. Rev. 105 (2005) 1103. [7] A. Vilan, O. Yaffe, A. Biller, A. Salomon, A. Kahn, D. Cahen, Adv. Mater. 22 (2010) 140. [8] D.K. Aswal, S. Lenfant, D. Guerin, J.V. Yakhmi, D. Vuillaume, Anal. Chim. Acta. 568 (2006) 84. [9] D. Cahen, A. Kahn, E. Umbach, Mater. Today 8 (2005) 32. [10] L.Ö. Olsson, C.B.M. Andersson, M.C. Håkansson, J. Kanski, L. Ilver, U.O. Karlsson, Phys. Rev. Lett. 76 (1996) 3626. [11] M.I. Schukfeh, K. Storm, A. Mahmoud, R.R. Søndergaard, A. Szwajca, A. Hansen, P. Hinze, T. Weimann, S.F. Svensson, A. Bora, K.A. Dick, C. Thelander, F.C. Krebs, P. Lugli, L. Samuelson, M. Tornow, ACS Nano 7 (2013) 4111. [12] Y. Fukuda, S. Ichikawa, M. Shimomura, N. Sanada, Y. Suzuki, Vacuum 67 (2002) 37. [13] D.Y. Petrovykh, M.J. Yang, L.J. Whitman, Surf. Sci. 523 (2003) 231. [14] V.N. Bessolov, M.V. Lebedev, Semiconductors 32 (1998) 1141. [15] Q. Hang, F. Wang, P.D. Carpenter, D. Zemlyanov, D. Zakharov, E.A. Stach, W.E. Buhro, D.B. Janes, Nano Lett. 8 (2008) 49. [16] R. Stine, D.Y. Petrovykh, J. Electron Spectrosc. Relat. Phenom. 172 (2009) 42. [17] J.W.J. Slavin, D. Zemlyanov, A. Ivanisevic, Surf. Sci. 603 (2009) 907. [18] T.A. Tanzer, P.W. Bohn, I.V. Roshchin, L.H. Greene, J.F. Klem, Appl. Phys. Lett. 75 (1999) 2794. [19] D.Y. Petrovykh, J.C. Smith, T.D. Clark, R. Stine, L.A. Baker, L.J. Whitman, Langmuir 25 (2009) 12185. [20] J.P. Folkers, P.E. Laibinis, G.M. Whitesides, Langmuir 8 (1992) 1330. [21] Refractive index database, in, M. N. Polyanskiy, http://refractiveindex.info 2014. [22] D.Y. Petrovykh, J.M. Sullivan, L.J. Whitman, Surf. Interface Anal. 37 (2005) 989. [23] D.Y. Petrovykh, J.P. Long, L.J. Whitman, Appl. Phys. Lett. 86 (2005) 1. [24] S. Jewett, D. Zemlyanov, A. Ivanisevic, J. Phys. Chem. C 115 (2011) 14244. [25] G.P. Schwartz, Thin Solid Films 103 (1983) 3. [26] F.L. Lie, W. Rachmady, A.J. Muscat, Microelectron. Eng. 86 (2009) 122. [27] L.J. Singh, R.A. Oliver, Z.H. Barber, D.A. Eustace, D.W. McComb, S.K. Clowes, A.M. Gilbertson, F. Magnus, W.R. Branford, L.F. Cohen, L. Buckle, P.D. Buckle, T. Ashley, J. Phys. D. Appl. Phys. 40 (2007) 3190. [28] O.E. Tereshchenko, D. Paget, P. Chiaradia, J.E. Bonnet, F. Wiame, A. Taleb-lbrahimi, Appl. Phys. Lett. 82 (2003) 4280. [29] Z.F. Tomashik, N.V. Kusyak, V.N. Tomashik, Inorg. Mater. 38 (2002) 434. [30] J.W.J. Slavin, U. Jarori, D. Zemlyanov, A. Ivanisevic, J. Electron Spectrosc. Relat. Phenom. 172 (2009) 47. [31] T. Heim, D. Deresmes, D. Vuillaume, J. Appl. Phys. 96 (2004) 2927. [32] H. Ishii, K. Sugiyama, E. Ito, K. Seki, Adv. Mater. 11 (1999) 605. [33] S. Braun, W.R. Salaneck, M. Fahlman, Adv. Mater. 21 (2009) 1450. [34] W. Knoben, S.H. Brongersma, M. Crego-Calama, J. Phys. Chem. C 113 (2009) 18331. [35] T. Aqua, H. Cohen, A. Vilan, R. Naaman, J. Phys. Chem. C 111 (2007) 16313.

A. Szwajca et al. / Surface Science 633 (2015) 53–59 [36] D.M. Alloway, M. Hofmann, D.L. Smith, N.E. Gruhn, A.L. Graham, R. Colorado Jr., V.H. Wysocki, T.R. Lee, P.A. Lee, N.R. Armstrong, J. Phys. Chem. B 107 (2003) 11690. [37] B. De Boer, A. Hadipour, R. Foekema, T. Van Woudenbergh, M.M. Mandoc, V.D. Mihailetchi, P.W.M. Blom, Tuning of metal work functions with self-assembled monolayers, Proceedings of SPIE—The International Society for Optical Engineering, 2004, p. 18. [38] N. Saito, K. Hayashi, H. Sugimura, O. Takai, N. Nakagiri, Chem. Phys. Lett. 349 (2001) 172.

59

[39] D.M. Taylor, G.F. Bayes, Phys. Rev. E. 49 (1994) 1439. [40] S. Lenfant, D. Guerin, F. Van Tran, C. Chevrot, S. Palacin, J.P. Bourgoin, O. Bouloussa, F. Rondelez, D. Vuillaume, J. Phys. Chem. B 110 (2006) 13947. [41] L. Romaner, G. Heimel, C. Ambrosch-Draxl, E. Zojer, Adv. Funct. Mater. 18 (2008) 3999. [42] A. Natan, N. Kuritz, L. Kronik, Adv. Funct. Mater. 20 (2010) 2077. [43] W. Knoben, S.H. Brongersma, M. Crego-Calama, Surf. Sci. 604 (2010) 1166. [44] W. Mönch, Semiconductor Surfaces and Interfaces, 3rd ed. Springer, 2001.