Electrochemical impedance spectroscopy of polycrystalline boron doped diamond layers with hydrogen and oxygen terminated surface

Diamond & Related Materials 55 (2015) 70–76

Contents lists available at ScienceDirect

Diamond & Related Materials journal homepage: www.elsevier.com/locate/diamond

Electrochemical impedance spectroscopy of polycrystalline boron doped diamond layers with hydrogen and oxygen terminated surface Zuzana Vlčková Živcová a,⁎, Václav Petrák b,c, Otakar Frank a, Ladislav Kavan a a b c

J. Heyrovsky Institute of Physical Chemistry of the AS CR, v.v.i., Dolejškova 3, 182 23 Prague 8, Czech Republic Institute of Physics of the AS CR, v.v.i., Na Slovance 2, 182 21 Prague 8, Czech Republic Czech Technical University in Prague, Faculty of Biomedical Engineering, Sítná 3105, 272 01 Kladno, Czech Republic

a r t i c l e

i n f o

Article history: Received 6 December 2014 Received in revised form 23 February 2015 Accepted 2 March 2015 Available online 4 March 2015 Keywords: Boron doped diamond Electrochemical impedance spectroscopy Aqueous electrolyte solution Surface terminations Acceptor concentration

a b s t r a c t This work is a systematic study of electrochemical impedance spectroscopy of high quality polycrystalline boron doped diamond films with varying boron content (from semiconducting to metallic behavior) and with different surface terminations (hydrogen or oxygen) in aqueous electrolyte solution. The films were grown by microwave plasma enhanced chemical vapor deposition. The concentration of acceptors (NA) was determined from the MottSchottky plots and the values were compared with those from neutron depth profiling and Raman spectroscopy. The NA values are in a good agreement across the different techniques, ranging from ca. 1.2 · 1020 cm−3 for semiconducting samples up to ca. 2 · 1021 cm−3 for heavily doped films with metallic conductivity. The films with hydrogen terminated surface exhibit lower values of both the flat band potentials (Efb) and the NA values, compared to the films with oxygen terminated surface. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Polycrystalline boron doped diamond (BDD) is a complicated system whose electrochemical properties are influenced by many factors, such as boron doping level, morphological features (grain boundaries and point defects), non-diamond impurity content (especially sp2 carbon), crystallographic orientation and surface termination [1–6]. Diamond can be modified by different surface functional groups ranging from simple heteroatom-termination of the carbonaceous skeleton to targeted chemical modifications for achieving specific electrochemical [1,4] and photoelectrochemical [7–9] functions. The most common surface terminations include oxygen (OT) and hydrogen (HT) atoms, which is associated with different wettability of the diamond (HT surface is hydrophobic, OT surface is hydrophilic). Further significant difference between hydrogen and oxygen terminated surface is in an electron affinity. The negative electron affinity (NEA) is a result of the C–H dipole at the hydrogenated diamond surface and the positive electron affinity (PEA) is caused by the C–O dipole at the oxygenated surface [5,10–12]. The electrochemical properties of BDD with hydrogen or oxygen terminated surface have been studied by several authors [1, 11–20]. Electrochemical impedance spectroscopy (EIS) is a useful technique to study the double-layer structure and charge transfer at surface modified BDD films. The impedance spectra can be fitted using ⁎ Corresponding author at: J. Heyrovsky Institute of Physical Chemistry, v.v.i., Academy of Sciences of the Czech Republic, Dolejškova 3, 182 23 Prague 8, Czech Republic. E-mail address: [email protected] (Z.V. Živcová).

http://dx.doi.org/10.1016/j.diamond.2015.03.002 0925-9635/© 2015 Elsevier B.V. All rights reserved.

equivalent circuits that correspond to the processes occurring at the electrolyte/BDD interface. The simplest equivalent circuit is the socalled Randles circuit, in which the Helmholtz capacitance and its corresponding parallel resistor [1,5,8,15,21–23], as well as the Warburg element [15,20] are neglected. From the Mott-Schottky plots, obtained from EIS measurements, it is possible to determine the concentration of active acceptors (NA) of BDD films. There are also other techniques for determining the acceptor concentration of BDD such as the neutron depth profiling (NDP), secondary ion mass spectroscopy (SIMS), Raman spectroscopy and Hall-effect measurements [24] which also provide the concentration of acceptors in bulk or in the surface. However, a specific feature of EIS is that it selectively maps the active acceptors in a thin space-charge (accumulation) layer underneath the surface. Although there is a bulky literature about impedance spectra and Mott-Schottky plots on diamond electrodes, the subject is far from being fully understood and consistent. Individual experimental works show large spread of results, such as flatband potentials (which range from about 0 to 4 V vs. SHE) [1,17,20,25–28] and also several different equivalent circuits were used for spectra fitting (see Fig. S1 in Supplementary data). Quite often, the actually applied circuit is not specified in the source works [17–19,29–31] but we may assume that they employed the simplest Randles-type RC circuit with pure capacitor like in [23]. Denisenko et al. [26] proposed a circuit with two RC elements in series, to account for a separation of space-charge layer and double layer at the interface. A replacement of simple capacitor with constant-phase element (CPE) and inclusion of Warburg impedance has been presented in several works [15,20,32]. A variant of Deniseko's

Z.V. Živcová et al. / Diamond & Related Materials 55 (2015) 70–76

model [26] with one CPE instead of capacitor was used in [21,22]. Yet more complex circuits with elements representing surface states are discussed in [27]. However, a systematic comparison of these circuits is less frequent in the literature. One exception is the work by Garrido at al. [15]. They compared the quality of fitting to two different R-CPE circuits: with or without the Warburg element, the latter provided more accurate matching with experimental spectra. To address these questions, we present here a comparative study using several different circuits, pointing at possibilities and limitations of this approach. More specifically, we have carried out an electrochemical impedance study of high quality polycrystalline boron doped diamond films prepared by microwave plasma enhanced chemical vapor deposition with various boron doping level and with hydrogen and oxygen terminated surface, respectively. Acceptor concentration (NA) of BDD films from impedance measurements (EIS), neutron depth profiling (NDP) and Raman spectroscopy were compared. 2. Experimental section 2.1. Preparation of BDD layers The polycrystalline boron doped diamond (BDD) films were deposited on fused silica substrates in an ASTeX 5010 series Microwave Plasma Enhanced Chemical Vapor Deposition (MPECVD) reactor. The samples were grown in a conventional CH4/H2 plasma and doping was induced by trimethyl boron gas B(CH3)3. The growth conditions were: B/C ratio in the gas phase 500, 2000 and 8000 ppm, pressure 47.7 mBar, temperature 710 °C, methane content 0.5% and deposition time 480 min. These synthetic conditions are identical to those used in our previous work [2]. The surface of BDD films was terminated with hydrogen (HT BDD) and oxygen (OT BDD), respectively. Surface termination was realized by plasmatic treatment in hydrogen (oxygen) plasma at pressure of 30 mBar and temperature of 500 °C for 10 min. Samples were cooled down to room temperature under constant flow of hydrogen (oxygen). The BDD films in work [2] were as-grown, i.e. with hydrogen terminated diamond surface. For a contact angle measurements 1.5 μl volume of distilled water was used to make a droplet on the surface of BDD film using a micropipette. The images of sessile droplets were taken using a commercial digital camera (Canon IXUS 500 HS). The contact angle was determined by fitting of droplet using the open source image processing software ImageJ with Contact Angle plug-in. 2.2. Methods for BDD films characterization Electrochemical measurements (impedance spectroscopy and cyclic voltammetry) were performed using a three electrode system in aqueous phosphate buffer solution pH 7.00 (PBS, Sigma Aldrich). The BDD films with hydrogen or oxygen terminated surface were used as working electrodes (Ag contact with Au wire insulated by epoxy coating), platinum mesh was a counter electrode and an Ag/AgCl electrode (sat. KCl) was a reference. The electrochemical impedance spectra were measured in the frequency range from 100 kHz to 0.1 Hz and in the bias voltage range from 0.9 V to −0.9 V using an AUTOLAB PGSTAT128N potentiostat with the frequency response analyzer (EcoChemie). The measurement was controlled by the GPES4 and FRA software. All

71

electrochemical measurements were carried out in Ar atmosphere. The physical surface area of the BDD electrodes was determined from the roughness factor of ca. 2.5 estimated from the double-layer capacitance. The total boron concentration (depth profile) in BDD films was determined with a neutron depth profiling (NDP) method with an error of ± 10% at standard ambient conditions [33]. Raman spectra were excited using 633 nm (1.96 eV) or 457 nm (2.71 eV) laser wavelength and recorded by a Labram HR spectrometer (Horiba JobinYvon) interfaced to an Olympus microscope (objective 100×). Alternatively, a WITEC spectrometer with a 532 nm (2.33 eV) laser excitation in a highly confocal setup (100 × objective and 25 μm fiber optics) was used. The spectrometers were calibrated by the F1g mode of Si at 520.2 cm−1. 3. Results Polycrystalline BDD thin films with an approximate thickness of 2 μm were grown by MPECVD on quartz substrates. The surfaces of these as-grown films can be considered mostly as H-terminated owing to the reduction conditions during growth, however, the surface termination cannot be regarded as unambiguously defined for surfacesensitive studies like EIS. The surface morphology evaluated by AFM and the structural properties (presence of graphitic or amorphous phases, boron doping level) analyzed by micro-Raman spectroscopy of the as-grown BDD films have been detailed in our previous work [2]. The BDD films show well-defined and faceted crystals (ca. 450 nm in size), and almost no graphitic or amorphous carbon impurities. For further studies, the BDD samples were first treated in hydrogen plasma to achieve fully H-terminated surface. Subsequently, after the completion of all relevant experiments (see below), the BDD samples were treated in oxygen plasma to acquire the O-terminated surface. The surface contact angles were 92° for H-terminated BDD and 12° for O-terminated BDD, respectively. This confirms high efficiency of hydrogen and oxygen treatments (Fig. 1). Electrochemical impedance spectra were measured on boron doped diamond (BDD) films with hydrogen or oxygen terminated surface. Fig. 2 shows experimental data corresponding to selected potentials for the whole frequency range (AC impedance Nyquist plots; imaginary part vs. real part of complex impedance for different frequencies) of BDD films with hydrogen (HT BDD) and oxygen terminated surface (OT BDD). Fig. 3 shows the equivalent circuit used for fitting the experimental data which takes into account the diffusion phenomena [15,20]. In this circuit the ohmic resistance Rs of the electrolyte solution, electrodes, contacts etc. is in series with the parallel combination of the space charge capacitance (BDD/electrolyte interface) represented by constant phase element (CPE) and its associated resistance (R1) in series with diffusion impedance Zw, the so-called Warburg element. This equivalent circuit (Fig. 3) provided the best fits to experimental spectra; other possible circuits are shown in Supplementary content (Fig. S1) with the corresponding fits to Nyquist plots in Fig. S2. By comparing the individual fits in Figs. 2 and S2, we can conclude that the circuit with CPE and Warburg impedance (Fig. 3) gives the best matching. The impedance of a CPE equals: −β

ZCPE ¼ BðiωÞ

Fig. 1. Contact angle of A) as-grown, B) hydrogen terminated and C) oxygen terminated surface of boron doped diamond film.

ð1Þ

Z.V. Živcová et al. / Diamond & Related Materials 55 (2015) 70–76

72

Fig. 2. Nyquist plots from electrochemical impedance spectra of the polycrystalline boron doped diamond (B/C ratio 500, 2000 and 8000 ppm) films with hydrogen (HT BDD; measured data as a red symbols, fit as a line) and oxygen (OT BDD; measured data as a black symbols, fit as a line) terminated surface measured at different potentials (−0.9 V — □, −0.5 V — ○, 0 V — ✵, 0.5 V — ●, 0.9 V — ■) vs. Ag/AgCl reference electrode in PBS electrolyte solution. The spectra were fitted (solid lines) to the equivalent circuit shown in Fig. 3. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

where ω is the EIS frequency, and B, β are frequency-independent parameters of the CPE (0 ≤ β ≤ 1; when β = 1, then the CPE corresponds to the capacitance C). In general, the CPE accounts for situations when the interfacial capacitance cannot be described by a single capacitor, e.g. when a frequency dispersion of double-layer capacitance occurs [34,35]. For diamond electrodes, the issue is even more controversial; but the CPE is usually justified by a complicated surface morphology, such as the presence of grain boundaries, edges and facets with each capacitance interfaced to a serial resistance [14]. Fig. 4 shows the CPE exponent β as a function of the applied potential for HT BDD and OT BDD. Hernando et al. [14] reported on the effect of surface termination on the value of CPE parameter β. As the capacitance value increases the β parameter decreases. The capacitance of the diamond with oxygen terminated surface is higher than that of the diamond with hydrogen termination surface due to the presence of electro-active oxygen-containing surface groups [14]. Also, we observe larger values of β parameter for the OT BDD than for the HT BDD. The origin of this trend is not clear and is subject to further investigation. The capacitance, C is obtained from ZCPE as follows: C¼

ðR1  BÞ1=β : R1

ð2Þ

An obvious advantage of this evaluation protocol is that it removes the virtual ‘frequency dispersion’ of Mott-Schottky plots [8,36,37]. This dispersion was often pronounced also on diamond electrodes [15,29, 30,38]. For the equivalent circuit shown in Fig. 3, the capacitance C is equal to the capacitance of space charge layer in the diamond electrode, that is C = Csc. In the potential range of water stability, the values of R1 are very high (of the order of 1012 Ω), therefore this resistance can be neglected in the equivalent circuit in Fig. 3 [26] and the response is almost perfectly capacitive. However, at more positive potentials the

resistances of R1 decrease to values up to ca. 105 Ω. Such smaller values of R1, observed at potential close to depletion regime, can indicate a faradaic leakage at the interface. However, the issue is quite complex, because even a non-degenerately doped diamond does not behave like an ideal semiconductor electrode obeying the classical Marcus– Gerischer model (with a faradaic hole injection from electrolyte redox couple into the valence band). A discrepancy between the expected diode-like characteristics for the faradaic charge-transfer and the experimentally observed ‘metal-like’ electrochemistry was attributed to interfacial surface states (presumably B-atoms) which mediate the charge transfer by hoping [27,28]. The Nyquist plots (Fig. 2) show a clear potential-dependent impedance evolution, similar for all boron concentrations as well as for both surface terminations. In all cases the curvature of the impedance data points increases with the absolute value of the potential. In other words, the Nyquist plot is almost straight at the potential of 0 V (the charge-transfer resistivity, R1, is the highest), while it changes towards a semi-arc at 0.9 and − 0.9 V (the charge-transfer resistivity, R1, decreases). The effect of increasing boron doping level is to reduce the overall impedance and the film resistivity. Fig. 5 shows the Mott-Schottky plots (effect of the applied potential on the space-charge layer in the semiconductor [1,15,16,29,30,38]) for HT BDD (left) and OT BDD (right). The plots are not linear over the

Fig. 3. Equivalent circuit used to fit the electrochemical impedance spectra for boron doped diamond (BDD) films; Rs is the ohmic serial resistance, CPE is the constant phase element, R1 is the associated charge transfer resistance and Zw is the Warburg element.

Z.V. Živcová et al. / Diamond & Related Materials 55 (2015) 70–76

73

The acceptor concentration (NA) of highly-doped BDD films (2000 and 8000 ppm) is higher for OT BDD than for HT BDD (Table 1) as well as the values of flat band potential (Efb). For example, in the case of 8000 ppm-doping, the intercept on bottom axis equals 2.9 V vs. Ag/AgCl for the oxygen-terminated surface, while it reaches only 1.2 V vs. Ag/AgCl for the hydrogen-terminated surface. This finding is consistent with other works [17–20]. However, we have to stress that the Mott-Schottky plots for highly-doped diamonds (8000 ppm) hardly account for pure space-charge capacitance only, i.e. the assumption in Eq. (3) that the Helmholtz capacitance (CH) can be omitted is incorrect in this case. Although reasonably linear Mott-Schottky plots are sometimes reported even for heavily doped diamonds [15,17,20] and the ‘flatband potentials’ as high as 4 V can be extrapolated from these plots [20,26,27] we have to consider these large values with care. At the condition of strong doping, the potential drop occurs primarily on the solution part of the interface, i.e. the impedance data do not describe solely the semiconductor properties of diamond. More specifically, the extrapolated potential from Eq. (3) is shifted from the ‘true’ value of Efb by a potential of ΔE [27]: ΔE ¼ Fig. 4. Frequency-independent parameter β of the constant phase element, (0 ≤ β ≤ 1) from the polycrystalline boron doped diamond films with hydrogen (HT BDD — circles) and oxygen (OT BDD — triangles) surface termination with different boron doping level (500, 2000 and 8000 ppm). The spectra were fitted to the equivalent circuit shown in Fig. 3.

whole measured potential range which is caused by the heterogeneous character of the polycrystalline interface [11,16–20]. Our HT electrodes shows typical S-shaped plots similar to some other works [17,20] but the linear central part is applicable for fitting. The Mott-Schottky equation used to calculate the acceptor concentration NA and flatband potential, Efb is: 1 C sc 2

 ¼

2 eε0 εr NA

  kT E−E fb − e

ð3Þ

where Csc is the capacitance of the space-charge (accumulation) region normalized to the electrode area, e is the electron charge, ε0 is the permittivity of free space, εr is the dielectric constant of the semiconductor, NA the number of acceptors per unit volume, E the applied voltage, Efb is the flatband potential, k is Boltzmann's constant and T is the temperature.

eεr ε0 NA : 2C 2H

ð4Þ

Assuming a value of CH ≈ 5 μF/cm2 and εr = 5.5 for diamond, we can estimate ΔE ≈ 1.5 V for NA ≈ 1021 cm−3. Consequently, the data for our 8000 ppm doping (Fig. 5, Table 1) should be considered with reservations, because they do not represent the space charge layer only. The positive shift of Efb for oxygen-terminated surface is due to the increase in the potential drop in the Helmholtz layer by the presence of oxygen functional groups and the associated reduction of the surface and subsurface hydrogen [20]. Conversely, in the case of semiconducting BDD film with boron doping level 500 ppm, the reverse trend of NA and Efb values is caused by the hydrogen termination leading to an increased free carrier density near the surface. This difference in electrochemical behavior of low doped BDD films is the result of a change from nondegenerate to degenerate doping [10]. In general, each negative Efb shift induces a better covering between the Fermi level of the redox system and the valence band of the diamond leading to an increase of charge transfer, while each positive flat band potential shift induces a decrease of the charge transfer [12]. Eventually, the actual shape of MottSchottky plots is also dependent on the equivalent circuit, which was chosen for EIS fitting. To illustrate this effect, we present in Fig. S3 (Supplementary content) the corresponding plots for such alternative circuits.

Fig. 5. Mott-Schottky plots for the polycrystalline boron doped diamond films with hydrogen (HT BDD — left) and oxygen (OT BDD — right) surface termination with different boron doping levels. The acceptor concentration (NA) was determined from the slope. This fit was carried our using the equivalent circuit in Fig. 3. Inset in the figure on the left shows where the slope crosses the x-axis for y = 0.

Z.V. Živcová et al. / Diamond & Related Materials 55 (2015) 70–76

74

Table 1 Flat band potential (Efb vs. Ag/AgCl) and concentration of acceptors (NA) of boron doped diamond films with hydrogen (HT) and oxygen (OT) surface termination determined by MottSchottky plots (MSP). For comparison are given the values of NA determined by the neutron depth profiling (NDP) and the Raman spectra on the same samples with the “as-grown surfaces”.

HT BDD

OT BDD

B/C (ppm)

Efb (V) vs. Ag/AgCl

NA (cm−3) (MSP)

500 2000 8000 500 2000 8000

1.19 ± 0.12 1.05 ± 0.03 1.16 ± 0.08 1.40 ± 0.12 2.25 ± 0.42 2.90 ± 0.83

1.9 × 1020 2.8 × 1020 4.9 × 1020 1.4 × 1020 4.6 × 1020 1.6 × 1021

“As-grown surface”(HT BDD)

All the BDD samples have been tested also by Raman microscopy employing several laser excitation wavelengths: 633, 532 and 457 nm. It should be noted that there are no visible differences between samples with the same B content but with a different surface chemistry (Fig. 6A), owing to the low sensitivity of the Raman spectroscopy to the surface compared to the optically transparent bulk. Standard Raman features associated with the increasing boron content can be observed both in Fig. 6A (for OT and HT BDD) and 6B (“as-grown”). The intensity of the zone-center phonon diamond peak (1332 cm−1) decreases and its frequency exhibits a shift towards lower wavenumber upon increasing B content. Its lineshape also changes from a symmetric Lorentzian to an asymmetric Fano appearance due to interference with the continuum of electronic states induced by the presence of boron [39]. The origin of the two broad features at ~1225 and 500 cm−1 is still a subject of disputes [39,40]. Recent theoretical and experimental studies suggest the peaks to be a superposition of C–C, B–C and B–B vibrations with varying surrounding (and thus varying bond strengths), with the vibrations of the boron pairs being dominant in the 500 cm−1 feature [41–43]. However, regardless of the particular origin of the 500 cm−1 peak, its frequency, ω, was shown to downshift with the increasing boron content, following an empirical relation [39,44]: 30

N A ¼ 8:44  10

expð−0:048ωÞ



−3

for N A in cm

and ω in cm

−1



:

ð5Þ The analysis relies on a decomposition of the broad 500 cm−1 feature using one Lorentzian and one Gaussian component, where only

B/C (ppm)

NA (cm−3) (NDP)

NA (cm−3) (Raman, 633 nm)

NA (cm−3) (Raman, 457 nm)

500 2000 8000

4.0 × 1020 1.3 × 1021 5.8 × 1021

9.1 × 1020 9.6 × 1020 4.3 × 1021

3.5 × 1020 6.3 × 1020 2.9 × 1021

the Lorentzian component shifts with the B content and its frequency is thus used for the B quantification. However, such a fitting is subjected to uncertainties, which might be reflected in the obtained NA values. An error of 2 cm−1 was shown to lead to approximately doubling the NA [44]. The results in Table 1 were obtained using two different excitations. They partially confirm this observation, showing their difference to be lower than by a factor of 2. The measurement error thus corresponds to the spectral resolution of the Raman spectrometer. As can be seen further from Table 1, the difference between results obtained using NDP and Raman spectroscopy fall within a factor of 2 as well. In a previous study of similar polycrystalline BDD films, the boron concentrations obtained by NDP were compared to acceptor concentrations determined by Hall measurements, and the concentrations were found to correlate [41]. Interestingly, the acceptor concentration obtained using impedance spectroscopy is underestimated by close to one order of magnitude compared to the values of boron concentration obtained both by NDP and Raman spectroscopy, when comparing the Hterminated and “as-grown” surface, i.e. virtually identical samples. On the other hand, for O-terminated surface, the NA values are within the experimental errors of NDP and Raman spectroscopy on the “asgrown” samples. This observation highlights the difference between the bulk concentration of boron atoms and the surface states generated by the interplay of boron atoms and surface functionalization. Fig. 7 shows the cyclic voltammograms of the HT BDD (left) and OT BDD (right) with different B/C ratio in the gas phase. We have selected here a relatively narrow potential window to demonstrate an unusual effect that our voltammetric background currents do not depend significantly on the surface termination. Usually, the oxygen termination

Fig. 6. (A) Raman spectra of BDD films with hydrogen (HT BDD — full lines) or oxygen (OT BDD — dashed lines) surface termination with different boron doping levels; excited by 633 nm laser radiation. (B) Raman spectra of “as-grown” BDD films (i.e., mostly H-terminated), excited by 633 nm (full lines) and 457 nm (dashed lines) laser radiations. The spectra are offset for clarity but the intensity scale is identical.

Z.V. Živcová et al. / Diamond & Related Materials 55 (2015) 70–76

75

Fig. 7. Cyclic voltammograms of polycrystalline boron doped diamond films with hydrogen (HT BDD-left) and oxygen (OT BDD-right) terminated surface. Electrolyte solution PBS (pH 7); scan rate 100 mV/s; potential range from −0.8 V to 1 V. Electrode potentials are given vs. the Ag/AgCl electrode.

(which makes the surface more hydrophilic, cf. Fig. 1) also causes significant enhancement of these currents, due to the presence of electroactive oxygen-containing surface groups [1]. For instance, Hernado et al. [14] reported almost one order of magnitude large currents on their OT samples as referenced to their HT samples [14]. (However, the latter work used ultra-nanocrystalline diamond made in high nitrogen-containing plasma without B-doping). Our finding about comparable voltammetric response of HT and OT samples matches the work of Liu et al. [13] reporting that the voltammetric background currents on the OT sample can be similar or even smaller than those on the HT sample, but this effect was not explained in the source work [13]. We can only speculate that oxygen plasma causes defect removal or surface smoothening similar to that upon bombardment with Ar+ ions [45] but deeper discussion would be unsupported at this stage of our research.

(from semiconducting to metallic behavior) and with different surface terminations (hydrogen or oxygen) in aqueous electrolyte solution. From the Mott-Schottky plots was determined the concentration of acceptors (NA) and the values were compared with those from neutron depth profiling and Raman spectroscopy carried out on the same samples. Acknowledgement This work was supported by the Grant Agency of the Czech Republic (contract No. 13-31783S). Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.diamond.2015.03.002.

4. Conclusions Polycrystalline boron doped diamond films were synthesized by microwave plasma enhanced chemical vapor deposition with different boron contents and with hydrogen or oxygen surface termination. The electrochemical impedance was studied in aqueous electrolyte solution. Data were fitted using the RC equivalent circuits, which takes into account the diffusion and where the space charge capacitance (BDD/electrolyte interface) is represented by constant phase element (CPE). Nyquist plots show a potential-dependent impedance evolution similar for all boron concentration as well as for both surface terminations, and in all cases the curvature of the impedance data increases with the potential. The acceptor concentration (NA) and the flatband potential (Efb) determined from Mott-Schottky plots of metallic BDD films are higher for OT BDD than for HT BDD. On the other hand, in the case of semiconducting BDD films, the surface-dependence of the NA and Efb values is reversed because the hydrogen termination leads to an increased free carrier density near the surface. Neutron depth profiling and Raman spectroscopy were also used for the determination of NA values of the same BDD films. In general, the NA values obtained by electrochemical impedance measurements NDP and Raman spectroscopy are in a good agreement across the different techniques and their corresponding measurement error. Prime novelty statement We present a comparative systematic study of electrochemical impedance spectroscopy (EIS) using several different circuits of high quality polycrystalline boron doped diamond films with varying boron content

References [1] A. Fujishima, Y. Einaga, T.N. Rao, D.A. Tryk, Diamond Electrochemistry, Elsevier, Tokyo, 2005. [2] Z. Vlckova-Zivcova, O. Frank, V. Petrak, H. Tarabkova, J. Vacik, M. Nesladek, L. Kavan, Electrochemistry and in situ raman spectroelectrochemistry of low and high quality boron doped diamond layers in aqueous electrolyte solution, Electrochim. Acta 18 (2013) 518–525. [3] I. Duo, A. Fujishima, C. Comninellis, Electron transfer kinetics on composite diamond (sp3)-graphite (sp2) electrodes, Electrochem. Commun. 5 (2003) 695–700. [4] J.H.T. Luong, K.B. Male, J.D. Glennon, Boron-doped diamond electrode: synthesis, characterization, functionalization and analytical applications, Analyst 134 (2009) 1965–1979. [5] C.E. Nebel, B. Rezek, D. Shin, H. Watanabe, Surface electronic properties of hterminated diamond in contact with adsorbates and electrolytes, Phys. Stat. Sol. (b) 203 (2006) 3273–3298. [6] Y.V. Pleskov, Y.E. Evstefeeva, M.D. Krotova, V.Y. Mishuk, V.A. Laptev, Y.N. Palyanov, Y.M. Borzdov, Effect of crystal structure on the behavior of diamond electrodes: electrochemical characteristics of individual crystal faces, J. Electrochem. Soc. 149 (2002) E260–E264. [7] S.W. Yeap, D. Bevk, X. Liu, H. Krysova, A. Pasquarelli, D. Vanderzande, L. Lutsen, L. Kavan, M. Fahlman, W. Maes, K. Haenen, Diamond functionalization with lightharvesting molecular wires: improved surface coverage by optimized suzuki cross-coupling conditions, RCS Adv. 4 (2014) 42044–42053. [8] M. Zukalova, M. Bousa, Z. Bastl, I. Jirka, L. Kavan, Electrochemical doping of compact TiO2 thin layers, J. Phys. Chem. C 118 (2014) 25970–25977. [9] W.S. Yeap, X. Liu, D. Bevk, A. Pasquarelli, L. Lutsen, M. Fahlman, W. Maes, K. Haenen, Functionalization of boron-doped nanocrystalline diamond with N3 dye molecules, ACS Appl. Mater. Interfaces 6 (2014) 10322–10329. [10] B. Rezek, C. Sauerer, C.E. Nebel, M. Stutzmann, J. Ristein, L. Ley, E. Snidero, P. Bergonzo, Fermi level on hydrogen terminated diamond surfaces, Appl. Phys. Lett. 82 (2003) 2266–2268. [11] A.F. Azevedo, M.R. Baldan, N.G. Ferreira, Doping level influence on chemical surface of diamond electrodes, J. Phys. Chem. Solids 74 (2013) 599–604. [12] N. Simon, H. Girard, D. Ballutaud, E. De La Rochefoucauld, A. Etcheberry, Evolution of surface terminations and electrochemical behavior of boron doped diamond

76

[13]

[14]

[15] [16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26] [27]

[28]

[29]

Z.V. Živcová et al. / Diamond & Related Materials 55 (2015) 70–76 electrodes induced by controlled anodic and cathodic pretreatments, ECS Trans. 3 (2007) 59–66. F.B. Liu, J.D. Wang, B. Liu, X.M. Li, D.R. Chen, Effect of electronic structures on electrochemical behaviors of surface-terminated boron-doped diamond film electrodes, Diamond Relat. Mater. 16 (2007) 454–460. J. Hernando, S.Q. Lud, P. Bruno, D.M. Gruen, M. Stutzmann, J.A. Garrido, Electrochemical impedance spectroscopy of oxidized and hydrogen-terminated nitrogeninduced conductive ultrananocrystalline diamond, Electrochim. Acta 54 (2009) 1909–1915. J.A. Garrido, S. Nowy, A. Hartl, M. Stutzmann, The diamond/aqueous electrolyte interface: an impedance investigation, Langmuir 24 (2008) 3897–3904. A.F. Azevedo, F.A. Souza, P. Hammer, M.R. Baldan, N.G. Ferreira, The influence of hydrogen plasma pretreatment on the structure of BDND electrode surface applied for phenol detection, J. Nanoparticle Res. 13 (2011) 6133–6139. P. Actis, A. Denoyelle, R. Boukherroub, S. Szunerits, Influence of the surface termination on the electrochemical properties of boron-doped diamond (BDD) interfaces, Electrochem. Commun. 10 (2008) 402–406. N. Simon, H. Girard, D. Ballutaud, S. Ghodbane, A. Deneuville, M. Herlem, A. Etcheberry, Effect of H and O termination on the charge transfer of moderately boron doped diamond electrodes, Diamond Relat. Mater. 14 (2005) 1179–1182. H. Girard, N. Simon, D. Ballutaud, M. Herlem, A. Etcheberry, Effect of anodic and cathodic treatments on the charge transfer of boron doped diamond electrodes, Diamond Relat. Mater. 16 (2007) 316–325. M.R. Baldan, A.F. Azevedo, A.B. Couto, N.G. Ferreira, Cathodic and anodic pretreated boron doped diamond with different sp2 content: morphological, structural, and impedance spectroscopy characterizations, J. Phys. Chem. Solids 74 (2013) 1830–1835. W. Yang, J.E. Butler, J. Russell, R.J. Hamers, Interfacial electrical properties of DNAmodified diamond thin films: intrinsic response and hybridization-induced field effects, Langmuir 20 (2004) 6778–6787. H. Gu, X.D. Su, K.P. Loh, Electrochemical impedance sensing of DNA hybridization on conducting polymer film-modified diamond, J. Phys. Chem. B 109 (2005) 13611–13618. J. Scharpf, A. Denisenko, C. Pietzka, E. Kohn, Effect of surface defects by rf oxygen plasma on the electrical properties of thin boron-doped diamond layers in electrolyte, Diamond Relat. Mater. 20 (2011) 1250–1254. W. Gajewski, P. Achatz, O.A. Williams, K. Haenen, E. Bustarret, M. Stutzmann, J.A. Garrido, Electronic and optical properties of boron-doped nanocrystalline diamond films, Phys. Rev. B 79 (2009). T.N. Rao, D.A. Tryk, A. Fujishima, Band gap movement of semiconducting diamond in aqueous electrolyte induced by anodic treatment, J. Electrochem. Soc. 146 (1999) 680–684. A. Denisenko, C. Pietzka, A. Romanyuk, H. El-Hajj, E. Kohn, The electronic surface barrier of boron-doped diamond by anodic oxidation, J. Appl. Phys. 103 (2008) 14904. J. Van De Lagemaat, D. Vanmaekelbergh, J.J. Kelly, Electrochemistry of homoepitaxial CVD diamond: energetics and electrode kinetics in aqueous electrolytes, J. Electroanal. Chem. 475 (1999) 139–151. M.N. Latto, G. Pastor-Moreno, D.J. Riley, The influence of doping levels and surface termination on the electrochemistry of polycrystalline diamond, Electroanalysis 16 (2004) 434–441. G. Pastor-Moreno, D.J. Riley, Electrochemical studies of moderately boron doped polycrystalline diamond in non-aqueous solvent, Electrochim. Acta 47 (2002) 2589–2595.

[30] L. Boonma, T. Yano, D.A. Tryk, K. Hashimoto, A. Fujishima, Observation of photocurrent from band-to-band excitation of semiconducting p-type diamond thin film electrodes, J. Electrochem. Soc. 144 (1997) L142–L145. [31] M. Wang, N. Simon, C. Decorse-Pascanut, M. Bouttemy, A. Etcheberry, M. Li, R. Boukherroub, S. Szunerits, Comparison of the chemical composition of borondoped diamond surfaces upon different oxidation processes, Electrochim. Acta 54 (2009) 5818–5824. [32] L. Mosinska, M. Kowalska, P. Popielarski, K. Fabisiak, K. Paprocki, M. Szybowicz, A. Wrzyszczynski, G.K. Zhusupkalieva, Cyclic voltammetry and impedance studies of undoped diamond films, Mater. Sci. Poland 31 (2013) 146–150. [33] V. Hnatowicz, J. Vacik, D. Fink, Deconvolution of charged particle spectra from neutron depth profiling using simplex method, Rev. Sci. Instrum. 81 (2010). [34] G.J. Brug, A.L.G. Van Den Eeden, M. Sluyters-Rehbach, J.H. Sluyters, The analysis of electrode impedances complicated by the presence of a constant phase element, J. Electroanal. Chem. 176 (1984) 275–295. [35] P. Zoltowski, On the electrical capacitance of interfaces exhibiting constant phase element behaviour, J. Electroanal. Chem. 443 (1998) 149–154. [36] L. Kavan, M. Zukalova, O. Vik, D. Havlicek, Sol–gel titanium dioxide blocking layers for dye-sensitized solar cells: electrochemical characterization, ChemPhysChem 15 (2014) 1056–1061. [37] L. Kavan, N. Tetreault, T. Moehl, M. Grätzel, Electrochemical characterization of TiO2 blocking layers for dye sensitized solar cells, J. Phys. Chem. C 118 (2014) 16408–16418. [38] M.D. Krotova, Y.V. Pleskov, A.A. Khomich, V.G. Ralchenko, D.N. Sovyk, V.A. Kazakov, Semiconductor properties of nanocrystalline diamond electrodes, Russ. J. Electrochem. 50 (2014) 101–107. [39] P.W. May, W.J. Ludlow, M. Hannaway, P.J. Heard, J.A. Smith, K.N. Rosser, Raman and conductivity studies of boron-doped microcrystalline diamond, facetted nanocrystalline diamond and cauliflower diamond films, Diamond Relat. Mater. 17 (2008) 105–117. [40] M. Bernard, C. Baron, A. Deneuville, About the origin of the low wave number structures of the raman spectra of heavily boron doped diamond films, Diamond Relat. Mater. 13 (2004) 896–899. [41] P. Ashcheulov, J. Sebera, A. Kovalenko, V. Petrak, F. Fendrych, M. Nesladek, A. Taylor, Z. Vlckova-Zivcova, O. Frank, L. Kavan, M. Dracinsky, P. Hubik, J. Vacik, I. Kraus, I. Kratochvilová, Conductivity of boron-doped polycrystalline diamond films: influence of specific boron defects, Eur. Phys. J. B 86 (2013) 443–4439. [42] P. Achatz, F. Omnes, L. Ortega, C. Marcenat, J. Vacik, V. Hnatowitcz, U. Koster, F. Jomard, E. Bustarret, Isotopic substitution of boron and carbon in superconducting diamond epilayers grown by MPCVD, Diamond Relat. Mater. 19 (2010) 814–817. [43] L. Niu, J.Q. Zhu, X. Han, M.L. Tan, W. Gao, S.Y. Du, First principles study of structural, electronic and vibrational properties of heavily boron-doped diamond, Phys. Lett. A 373 (2009) 2494–2500. [44] M. Bernard, A. Deneuville, P. Muret, Non-destructive determination of the boron concentration of heavily doped metallic diamond thin films from raman spectroscopy, Diamond Relat. Mater. 13 (2004) 282–286. [45] Y. Einaga, R. Sato, H. Olivia, D. Shin, T.A. Ivandini, A. Fujishima, Modified diamond electrodes for electrolysis and electroanalysis applications, Electrochim. Acta 49 (2004) 3989–3995.