Effect of pre-surface treatments on p-Cu 2 O/Au Schottky junctions

: Cuprous oxide (Cu 2 O) is a suitable semiconducting material for fabrication of low-cost, eco-friendly semiconductor junction devices. Besides the parameterization of the growth conditions of Cu 2 O, formation of metal contacts impact the overall performance of these type of devices. The existence of unavoidable dangling bonds and/or dislocated surface atoms could lead to form imperfect contacts with metals, for example in Cu 2 O/Au junction devices. Nevertheless, modification of the Cu 2 O thin film surfaces prior to make contacts with Au has shown the capability to alter the junction properties. Here we report that, the application of surface treatments; annealing and/or sulphidation on specifically the electrodeposited p-Cu 2 O thin film surfaces, where p-Cu 2 O thin films were grown in low cupric ion concentrated acetate bath, has influenced the interfacial properties of particular p-Cu 2 O/Au Schottky junctions compared to the untreated p-Cu 2 O/Au Schottky junction. This has been well-established by the results of SEM and C-V characterizations of p-Cu junctions. The subsequent annealing and sulphidation of p-Cu 2 O thin film surfaces have lowered the built-in potential value by 121 mV compared to the untreated Schottky junction. This result reveals the possibility of employing surface treatments on electrodeposited Cu 2 O thin films in fabrication of high efficient Cu 2 O based junction devices. applied potential (V), Mott-Schottky

It is known that extracting output of a semiconductor through a proper metallic contact enhances the device performance. Therefore, it is useful to have knowledge on how a semiconductor works with a metal interface. Basically, the work functions of a semiconductor and the metal in contact tell whether it is an ohmic or Schottky junction (Sze & Ng, 2006). The Schottky nature of Cu 2 O/ metal junction is useful in fabrication of Schottky barrier solar cells and making ohmic contact is important in other type of solar cell configurations (i.e., homo-, hetero-or multi-junctions), transistors, Peltier modules etc.
In 1979s, Olsen et al. have reported the theoretical and experimental results related to Cu 2 O in contact with variety of metals such as, Yb, Mg, Mn, Al, Cu, Cr and Au (Olsen et al., 1979). Practically, the reported ohmic nature of the Cu 2 O/Au contact is not observed in electrodeposited Cu 2 O thin films (Kafi et al., 2018a;. However, the other attractive properties of electrodeposition method such as ability to grow Cu 2 O on different substrates (Abdelfatah et al., 2015;Mohra et al., 2016;Bouderbalaet al., 2018) or different orders in a cell configurations (Jayathileke et al., 2008;Wijesundera et al., 2016) and uses of Au contact such as, low resistivity, high mobility and durability (Mayer, 1984) tell us the importance of formation of low resistive p-Cu 2 O/Au junctions.

March 2021
Journal of the National Science Foundation of Sri Lanka 49 (1) Detailed interfacial studies on surface treated p-Cu 2 O/Au Schottky junction, where p-Cu 2 O thin films grown in acetate bath, are not available in the literature. Semiconductor surface reconstruction and modification change the surface properties where it is useful in optimisation of the overall junction performance (Morrison, 1977;Jayathilaka et al., 2014;Wijesundera et al., 2016) In this study, the effect on the surface properties of p-Cu 2 O thin films grown in aqueous acetate electrolyte, by changing the surface properties; annealing and/or sulphidation, in contact with Au Schottky junction has been investigated microscopically and the interfacial properties were studied with the aid of Mott-Schottky analysis. The study has shown that these pre-treatments on Cu 2 O thin film surfaces modify the interface properties not only by protecting its surfaces from corrosion and passivating the surface reactivity, but also by improving the photocurrent collection via Au contact. This result is well-established by the values obtained from the builtin potential and doping density profiles of the Cu 2 O thin films. Further, the built-in potentials relative to the Fermi level of Au have diminished for the surface treated p-Cu 2 O thin films compared to the as-grown p-Cu 2 O thin film. Thus, the reduction of the Schottky barrier height is feasible by applying surface pre-treatments. This understanding is a very important in fabrication of junction devices to minimize power losses.

METHODOLOGY
p-Cu 2 O thin films were potentiostatically electrodeposited on titanium (Ti) substrates by using a three electrode electrochemical system consisting of a reference electrode of Ag/AgCl, a counter electrode of platinum plate and a working electrode of Ti. Hokuto Denko Hub-151 potentiostat/galvanostat was used for this purpose. Table 1 represents the essential parameters of p-Cu 2 O thin film deposition used here (Jayathileke et al., 2008;Wijesundera et al., 2016). Note that, a few drops of diluted 0.01 M NaOH was used to adjust the pH of the film deposition bath.
In order to study the effect on the pre-treatments of the p-Cu 2 O thin film surfaces, one set of as-grown p-Cu 2 O thin films was annealed at 175°C for 10 min in a Carbolite heavy duty box furnace (Wijesundera et al., 2016) and another identical set of as-grown p-Cu 2 O thin films was exposed to 20 vol% of ammonium sulphide solution for 8s at room temperature (Kafi et al., 2018b;Jayathilaka et al., 2014). Also, a third identical set of as-grown p-Cu 2 O thin films was subsequently annealed and sulphided using the above pre-treatment conditions. Then, the desired Schottky junctions; as-grown p-Cu 2 O/Au, annealed p-Cu 2 O/Au, sulphided p-Cu 2 O/Au and annealed and sulphided p-Cu 2 O/Au were fabricated using the Cressington sputtered coater.
The structure and the surface morphology of the p-Cu 2 O thin films were studied using Rigaku Ultima IV X-ray diffractometer and ZESIS EVO scanning electron microscope respectively. Further, the capacitancevoltage (C-V) characterization was used for all of the above surface treated and untreated p-Cu 2 O/Au Schottky junctions using Gamry series G300 potentiostat/ galvanostat/ZRA instrument.

RESULTS AND DISCUSSION
Well-adhered qualitative p-Cu 2 O thin films were successfully electrodeposited on Ti substrates by applying the growth conditions stated in the Table 1. The thickness of the films was calculated by monitoring the current passing through the electrodeposition process (Wijesundera et al., 2016) and the calculated thickness of the film was in the order of 1-2 µm. The optimum growth conditions were used to grow the p-Cu 2 O thin films (Jayathileke et al., 2008). Since the same growth conditions were used to grow thin films, the thickness of the thin films can be assumed the same for all the films. In this study, the effect of surface properties on the surface pre-treatment to the films was explored.

Parameter
Best condition to the reflections from (110), (111), (200), (220) and (311) planes. There are no any other peaks (from impurity phases) except the peaks corresponding to the Ti substrate. Similar XRD for electrodeposited Cu 2 O thin films has been reported by Jamali et al., (2017). Therefore, it can be concluded that this XRD pattern describes the bulk properties of p-Cu 2 O thin films.
SEM images in Figure 2 visibly illustrate the changes of the p-Cu 2 O thin film surfaces under pre-described surfaces treatments. According to the Figure 2 the sharpness of the grain boundaries of the as-grown p-Cu 2 O thin films in the acetate bath is very low compared to the Cu 2 O thin films that have undergone pre-treatments. With the effect of low temperature annealing and sulphidation, the grain arrangement of p-Cu 2 O thin films on titanium substrates has reconstructed the thin films surfaces into sharp grain arrangement (Kafi et al., 2018b). Thus, it is observed as sharp pyramid shaped grains on their surfaces. The observed pyramid shaped grains may due to the orientation of cubic shaped micro grains. i.e., application of pre-treatments on the Cu 2 O film surfaces have qualitatively changed the surface properties.
In order to make a quantitative analysis on these surface-treated or untreated p-Cu 2 O/Au Schottky junctions Mott-Schottky analysis is the obvious choice. The interfacial behaviour of p-Cu 2 O/Au Schottky junctions can be easily analysed through the built-in potentials and the doping densities obtained from the C-V measurements (Mott-Schottky analysis). A Mott-Schottky plot is the linear variation of the inverse square of the capacitance of the space charge layer (1/C 2 ) versus the applied potential (V), which is explained by the Mott-Schottky theory (Sze & Ng, 2006). According to equation 1, the V-intercept and the gradient of the Mott-Schottky plot give the built-in potential and the doping density of the particular film respectively. 5 calculated by monitoring the current passing through the electrodeposition process (Wijesundera et al., 2016) and the calculated thickness of the film was in the order of 1-2 m.
The optimum growth conditions were used to grow the p-Cu2O thin films (Jayathileke et al., 2008). Since the same growth conditions were used to grow thin films, the thickness of the thin films can be assumed the same for all the films. In this study, the effect of surface properties on the surface pre-treatment to the films was explored.     theory (Sze & Ng, 2006) According to equation 1, the V-intercept and the gradient of the Mott-Schottky plot give the built-in potential and the doping density of the particular film respectively.
where C, ε, ε 0 , A, N A , V, V bi , k, T and e represent the capacitance of the space charge region, dielectric constant of the semiconductor [6.6 for Cu2O (Heltemes, 1966)], the permittivity of the free space, area in contact of the junction, acceptor density, applied potential, built-in potential, Boltzmann constant, temperature and the charge of an electron respectively.
Further, the Fermi level (EF) positions relative to the valence band edge (EV) of p-Cu2O was calculated using the following equation (2) (Sze & Ng, 2006).
Where N V is the effective density of states in valence band and for p-Cu2O this value is 1.1 × 10 19 cm -3 (calculated with an effective hole mass of 0.58 m0 (Hodby et al., 2001); Here the m0 is the electron mass) and all other symbols have their usual meaning.
... (01) where 7 Mott-Schottky plot give the built-in potential and the doping density of the particular film respectively.
where C, ε, ε 0 , A, N A , V, V bi , k, T and e represent the capacitance of the space charge region, dielectric constant of the semiconductor [6.6 for Cu2O (Heltemes, 1966)], the permittivity of the free space, area in contact of the junction, acceptor density, applied potential, built-in potential, Boltzmann constant, temperature and the charge of an electron respectively.
Further, the Fermi level (EF) positions relative to the valence band edge (EV) of p-Cu2O was calculated using the following equation (2) (Sze & Ng, 2006).
Where N V is the effective density of states in valence band and for p-Cu2O this value is 1.1 × 10 19 cm -3 (calculated with an effective hole mass of 0.58 m0 (Hodby et al., 2001); Here the m0 is the electron mass) and all other symbols have their usual meaning. and e represent the capacitance of the space charge region, dielectric constant of the semiconductor which is 6.6 for Cu 2 O (Heltemes, 1966), the permittivity of the free space, area in contact of the junction, acceptor density, applied potential, builtin potential, Boltzmann constant, temperature and the charge of an electron respectively.
Further, the Fermi level (E F ) positions relative to the valence band edge (E V ) of p-Cu 2 O was calculated using the following equation (2) (Sze & Ng, 2006). Mott-Schottky plot give the built-in potential and the doping density of the particular film respectively.
where C, ε, ε 0 , A, N A , V, V bi , k, T and e represent the capacitance of the space charge region, dielectric constant of the semiconductor [6.6 for Cu2O (Heltemes, 1966)], the permittivity of the free space, area in contact of the junction, acceptor density, applied potential, built-in potential, Boltzmann constant, temperature and the charge of an electron respectively.
Further, the Fermi level (EF) positions relative to the valence band edge (EV) of p-Cu2O was calculated using the following equation (2) (Sze & Ng, 2006).
Where N V is the effective density of states in valence band and for p-Cu2O this value is 1.1 × 10 19 cm -3 (calculated with an effective hole mass of 0.58 m0 (Hodby et al., 2001); Here the m0 is the electron mass) and all other symbols have their usual meaning.
... (02) Where 7 charge layer (1/C ) versus the applied potential (V), which is explained by the Mott-Schottky theory (Sze & Ng, 2006) According to equation 1, the V-intercept and the gradient of the Mott-Schottky plot give the built-in potential and the doping density of the particular film respectively.
where C, ε, ε 0 , A, N A , V, V bi , k, T and e represent the capacitance of the space charge region, dielectric constant of the semiconductor [6.6 for Cu2O (Heltemes, 1966)], the permittivity of the free space, area in contact of the junction, acceptor density, applied potential, built-in potential, Boltzmann constant, temperature and the charge of an electron respectively.
Further, the Fermi level (EF) positions relative to the valence band edge (EV) of p-Cu2O was calculated using the following equation (2) (Sze & Ng, 2006).
Where N V is the effective density of states in valence band and for p-Cu2O this value is 1.1 × 10 19 cm -3 (calculated with an effective hole mass of 0.58 m0 (Hodby et al., 2001); Here the m0 is the electron mass) and all other symbols have their usual meaning.
is the effective density of states in valence band and for p-Cu 2 O this value is 1.1 × 10 19 cm -3 when it is calculated with an effective hole mass of 0.58 m 0 (Hodby et al., 2001); Here the m 0 is the electron mass and all other symbols have their usual meaning.    (1) and (2). According to the table 2, it is clear that the built-in potential and doping density of the surface treated p-Cu2O/Au Schottky junctions have diminished in values compared to as-grown p-Cu2O/Au Schottky junction. It is wellknown that the p-and n-type nature of the Cu2O is due to the lattice defects. Cu2O thin film surface are very reactive in absence of surface passivation. Presence of dangling bonds and chemisorption of the Cu2O thin film surfaces lead to exist surface states in the Cu2O thin film surfaces. Thus, the bulk properties of Cu2O thin films will not be same at the surface. One possible explanation to the observed result is that, annealing and/or sulphidation of p-Cu2O Journal of the National Science Foundation of Sri Lanka 49 (1) March 2021 in comparison with the E V -E F values with the surface treatments, it is clear that the observed negative shift of the built-in potential relative to the Fermi level of Au is resulted due to the shift of its valence band edge itself. Evidently, similar result was reported for the p-Cu 2 O thin films grown in lactate bath (Kafi et al., 2018b).

CONCLUSION
SEM and C-V characterisations have revealed that the pre-treatments on the p-Cu 2 O thin film surfaces prior to the Au contacts have the ability to alter the interface properties of the p-Cu 2 O/Au junctions. This is due the surface modification and/or reconstruction of the dislocated atoms at the interface layer of the Schottky junctions. Thus, the information on the negative shift of the build-in potential at the surface treated Cu 2 O/Au interfaces relative to the Fermi level of Au implies the reduction of Schottky barrier height. This result is very useful in fabrication and designing of efficient Cu 2 O based junction devices.