• 沒有找到結果。

Effects of indium concentration on the efficiency of amorphous In-Zn-O/SiOx/n-Si hetero-junction solar cells

N/A
N/A
Protected

Academic year: 2021

Share "Effects of indium concentration on the efficiency of amorphous In-Zn-O/SiOx/n-Si hetero-junction solar cells"

Copied!
6
0
0

加載中.... (立即查看全文)

全文

(1)

Effects of indium concentration on the ef

ficiency of amorphous

In

–Zn–O/SiO

x

/n-Si hetero-junction solar cells

Hau-Wei Fang

a

, Tsung-Eong Hsieh

a,n

, Jenh-Yih Juang

b a

Department of Materials Science and Engineering, National Chiao Tung University, 1001 Ta-Hsueh Road, Hsinchu 30010, Taiwan, ROC

bDepartment of Electrophysics, National Chiao Tung University, 1001 Ta-Hsueh Road, Hsinchu 30010, Taiwan, ROC

a r t i c l e i n f o

Article history: Received 22 May 2013 Received in revised form 31 October 2013

Accepted 1 November 2013 Available online 30 November 2013 Keywords:

Indium zinc oxide Pulsed laser deposition

Hetero-junction structure solar cells

a b s t r a c t

Semiconductor–insulator–semiconductor (SIS) hetero-junction solar cells comprising of the amorphous indium zinc oxide (a-IZO) layer directly deposited onto the n-type Si substrates by pulsed laser deposition were fabricated. Characterizations on the physical properties of the a-IZO layer and the a-IZO/SiOxinterface as a function of In/(ZnþIn) ratio were carried out to delineate their influences on the photovoltaic performance of SIS solar cells. The optical and electrical analyses indicated that the resistivity of a-IZOfilms decreased with increasing In concentration, reaching 4.5  104Ω-cm for In/ (ZnþIn)¼0.5, which also exhibited a transmittance higher than 80% in the visible-light wavelength range. Moreover, combining with an optimally controlled insulating SiOxlayer (about 2.0 nm), the device exhibited excellent SIS solar cell performance with open-circuit voltage of 0.38 V, short-circuit current density of 45.1 mA cm2,fill factor of 49.7% and a conversion efficiency of 8.4% under the AM1.5 illumination condition. The dramatic performance enhancement was attributed to the reduction of effective interface trap densities at the a-IZO/SiOxinterface and the increase of carrier mobility in the a-IZO layer resulted from the increase of In/(InþZn) ratio.

& 2013 Elsevier B.V. All rights reserved.

1. Introduction

The photovoltaic devices based on transparent conducting oxides (TCOs) such as semiconductor–insulator–semiconductor (SIS) heterojunction solar cells have attracted extensive research interests due to the advantages of relatively high conversion efficiency (η), simple device structure, easy manufacturing process and low production cost. Various TCOs, e.g., indium tin oxide (ITO)

[1–3], tin oxide (SnO2) [4] and zinc oxide (ZnO)[5] have been deposited on silicon (Si) wafer substrates clad with a thin SiOx layer to form the SIS devices and theη values in those reports were 10%, 7.8% and 6.9%, respectively. However, in abovementioned studies, the TCOs were all deposited at relatively high tempera-tures (4400 1C) to yield the required polycrystalline structure. This is adverse to the thermal budget of manufacturing process.

Recently, amorphous TCOs such as amorphous indium zinc oxide (a-IZO) films have received tremendous attention because of their wide optical bandgap (Eg43.3 eV), high carrier mobility (410 cm2

V1s1) and high carrier concentration (41020 cm3) properties, which are suitable for the optoelectronic applications. In particular, the relatively low In content and deposition temperature have made a-IZO a promising alternative for the fabrication of SIS

solar cells in comparison with ITO. In the TCO/SiOx/Si SIS solar cells, it has been pointed out that the thickness of the SiOxlayer and the presence of interface states are the key factors affecting the SIS device performance[1]. In general, the thickness of the SiOxlayer has to be limited to about 2 nm in order to efficiently separate the TCO layer and Si substrate and serve as the tunneling layer of electrons from Si substrate to the TCO layer. In our previous studies of a-IZO/ SiOx/Si SIS solar cells[6,7], various growth methods of SiOxlayers including the wet process (immersing the Si substrate in a hot H2O2 solution to form the SiOx layer) and the dry thermal oxidization (forming the SiOxlayer by in-situ heating the Si substrate in vacuum chamber prior to the a-IZO deposition) were performed. It was found that indeed the quality of SiOxlayer plays a critical role in the SIS solar cell performance. For instance, the dry method was found to result in denser SiOxlayer with lower interface trap densities (Dit) at the a-IZO/SiOxand SiOx/Si interfaces, which in turn leads to over 50% improvement in theη value of the SIS devices from 2.2%[6]to 3.4%

[7]. The above results also suggest thatη may be further improved if the effective Dit at the SiOx/Si and a-IZO/SiOx interfaces can be reduced by pre-occupied part of the interface states with free carriers, such that more photo-excited electrons can be harvested. In practice, the thickness of SiOxlayer and Ditat the SiOx/Si interface are determined by the condition of the dry thermal oxidization and can be regarded as afixed parameter. On the other hand, it might be possible to manipulate the effective Dit at a-IZO/SiOxinterface by changing the carrier density in the a-IZO layer.

Contents lists available atScienceDirect

journal homepage:www.elsevier.com/locate/solmat

Solar Energy Materials & Solar Cells

0927-0248/$ - see front matter& 2013 Elsevier B.V. All rights reserved.

http://dx.doi.org/10.1016/j.solmat.2013.11.003

nCorresponding author. Tel.:þ886 3 5712121x55306; fax: þ886 3 5724727.

(2)

In this work, we investigate the effects of carrier concentration in the a-IZO layer on the photoelectric characteristics of a series of a-IZO/SiOx/Si SIS devices by varying the In/(ZnþIn) ratio while keeping the SiOx/Si parts identical. Systematic examinations on the microstructures, electrical, optical properties were carried out on the a-IZOfilms. In particular, the photo-responsive capacitance– voltage (C–V) measurement[8,9]was performed to evaluate the Ditproperty at a-IZO/SiOxinterface and its influences on the SIS solar cell performance are discussed as follows.

2. Experimental

The a-IZO films were separately deposited on Corning 1737 glass plates and n-type Si(1 0 0) wafers (resistivity¼2–5 Ω-cm) by pulsed laser deposition (PLD). The mixtures of In2O3 and ZnO powders (supplier: Gredmann; purity: 99.999%) with In/(ZnþIn) ratios of 0.25, 0.4 and 0.5 were used for preparing the PLD targets. In order to obtain the compact pellets, the In2O3–ZnO powder mixtures were homogenized by attrition milling, pre-calcined at 6001C for 6 h in air, and then pressed at 10 psi in disc form with a diameter of 1 in. Finally, the pellets were sintered at 12501C for 2 h to form the PLD targets.

Before being transferred into the vacuum chamber, the glass substrates were cleaned sequentially in de-ionized water, acetone and ethanol (10 min for each step). The Si wafers were cleaned with a buffered oxide etching solution comprised of 40% NH4F and 49% HF in a volume ratio of 6:1 for 3 min to remove the native oxide layer. The KrF excimer laser (Compex 201,λ¼248 nm) was used to ablate the PLD target with an energy density of 3.8 J cm2per pulse and a repetition rate of 5 Hz. The background pressure of the PLD vacuum chamber was 4.0 106Torr. The distance between the target and the substrate was kept at 4 to 5 cm and the substrate temperature was kept at 2501C during the deposition. In order to obtain the low-resistivity a-IZO films, high-purity argon (Ar; 99.999%) with a flow rate of 6.0 sccm was introduced into the chamber and the working pressure was maintained at 4.2 m Torr. The thickness of a-IZO films determined by the ellipsometry (SOPRA, GES-5) and alpha-step profilometer (KOSAKA, ET300) was in the range of 25 to 35 nm. Afterward, 100-nm thick aluminum (Al) metal layers were deposited by e-beam evaporation to form the bottom electrodes on the backside of Si substrates and the topfinger electrodes on a-IZO layers, respectively.

The crystallinity of a-IZOfilms was examined by X-ray diffrac-tion (XRD; REGAKU) withλCu-Kαof 0.15405 nm and a transmission electron microscope (TEM; JEOL JEM-2100F) operated at 200 kV. The transport properties of a-IZOfilms were determined by the Hall measurement (ECOPIA HMS-3000) in a constant magnetic field of 0.58 T. The optical transmittance of a-IZO films was measured in the wavelength range of 300 to 1600 nm by an UV– vis–NIR spectrometer (JASCO V-650). The Ditvalues were deduced from the photo-responsive C–V measurements equipped with a 20-W halogen lamp as the light source and a Wayne Kerr 6520B precision LCR meter operating at the frequency of 100 kHz. In order to differentiate the influence of bulk defects in a-IZO layer from that of interfacial defects at a-IZO/SiOx interface, the C–V characteristics of Al/a-IZO/Al and Al/a-IZO/SiOxsamples were also analyzed by using a Wayne Kerr 6520B precision LCR meter at 100 kHz. The performance of the a-IZO/SiOx/n-Si SIS solar cells were measured in a solar cell efficiency measurement system equipped with a xenon lamp and a Keithley 2400 I–V source meter under the AM1.5 illumination condition. The incident photon-to-current conversion efficiency (IPCE) was measured with a xenon lamp and a Keithley 2400 source meter in the wavelength range of 400 to 1400 nm.

3. Results and discussion

Fig. 1 presents the XRD profiles of a-IZO films deposited on glass substrates at 2501C using the IZO targets with various In/ (ZnþIn) ratios. It is evident that, in the detected diffraction angle (2θ) ranging from 20 to 601, no characteristic peak can be identified except for a broad low-intensity peak occurring in the range of 2θ¼30 to 351, indicating that the IZO films are amorphous or of the nano-crystalline structure [10,11]. Moreover, the broad XRD peak shifts toward lower diffraction angles with the increase of In content. It is noted that 2θ of 30.61 and 34.41 can be assigned to the diffraction peaks of 2 2 2ð ÞIn2O3and (0 0 2)ZnO, respectively.

The peak shift might thus be ascribed to the In3þ (r0.080 nm) substitution for the Zn2þ (r0.074 nm) in IZO lattice, which is expected to result in a lattice expansion. Using the targets with different In contents to prepare IZOfilms, Minami et al. [12]and Naghavi et al.[13]also reported a similar lower-angle-shift of XRD peak with the increase of In content.

Fig. 2 summaries the transport properties of a-IZO films obtained by the Hall measurement. The resistivity of a-IZO film decreases with the increase of In/(ZnþIn) ratio and the sample with In/(ZnþIn) of 0.5 exhibits the lowest resistivity of 4.5 104Ω-cm. All a-IZO films exhibit n-type conducting beha-viors and the carrier concentration increases from 2.0 1020to 8.5 1020cm3when the In/(ZnþIn) ratio increases from 0.25 to 0.5, indicating that the resistivity decrement is correlated to the increase of carrier concentration. The enhancement of n-type carrier concentration is a natural consequence of doping higher valence In3þ into ZnO[14]. Moreover, it is also evident that the increase of In/(ZnþIn) ratio results in a moderate increase of Hall mobility from 12 to 16 cm2V1s1. Previous studies [15,16] reported that the Hall mobility of a-TCOs comprising (n1) d10ns0 (nZ4) cation dopants are insensitive to the structural

20 25 30 35 40 45 50 55 60 0 20 40 60 80 100 120

In/(Zn+In) = 0.25

In/(Zn+In) = 0.4

(222)

In2O3

(002)

ZnO

In/(Zn+In) = 0.5

2θ (degrees) Intensity

Fig. 1. XRD profiles of a-IZO films prepared on glass substrates using the IZO targets with different In/(ZnþIn) ratios.

0.0 0.5 1.0 1.5 2.0 2.5 3.0

Ν

μμ

ρρ

Resistivity (10 -3 Ω -cm) 0.5 0.4 0.25 Carrier Concentration (10 20 cm -3) In/(Zn+In) (at.%) 0 2 4 6 8 10 5 10 15 20 25 Mobility (cm 2 V -1sec -1)

Fig. 2. Resistivity (ρ), carrier concentration (N), and mobility (μ) of a-IZO films prepared on glass substrates using the IZO targets with different In/(ZnþIn) ratios.

(3)

disorder because of the large overlap between the ns orbitals. Thus, the improved Hall mobility observed in the present a-IZO films can be attributed to more abundant percolated conduction paths arising from the overlapping of In 5s orbitals.

Optical transmittances of the a-IZO films with various In/ (ZnþIn) ratios are presented inFig. 3. The average transmittances exceed 80% in the visible-light to NIR wavelength region for all a-IZOfilms, indicating that In content has negligible effects on the film transmittance. However, the absorption edge appears to shift toward the short wavelength side with the increase of In content. This can be explained by the Burstein–Moss effect[17]induced by the increase of the carrier concentration in the a-IZOfilm. Alter-natively, it might be due to the blunted curvature of conduction band resulted from the amorphous structure [18]. For a direct bandgap semiconductor [19], the Eg of sample can be obtained from the Tauc plot[20], i.e., the plot of (αhv)r

against hv, whereα is the absorption coefficient, h is Plank's constant, v is the frequency of the incident photon and r¼2 for amorphous oxide semicon-ductors, respectively[19]. As shown in the inset ofFig. 3, the Egof a-IZO films increases from 3.37 to 3.54 eV when the In/(ZnþIn) ratio increases from 0.25 to 0.5.

Fig. 4shows the cross-sectional TEM image of the SIS device containing a-IZO film with In/(ZnþIn)¼0.5. An ultra-thin SiOx layer with the thickness about 2.03 nm can be seen at the interface between the Si substrate and a-IZO layer. In addition to the microstructure and properties of a-IZO layers, the geometry of SiOxlayer is known to affect the tunneling current and, hence, is correlated to the SIS device performance[6,21]. The thickness of SiOxlayer in our SIS sample is in agreement with the optimum thickness of about 2 nm obtained from our previous studies on the

Si-based SIS solar cells [7]. Thus, the SIS device characteristics presented in the following should be able to delineate the effects arising from the properties of the top a-IZO layer. Moreover, the distribution of lattice fringes in a-IZO region (seeFig. 4) reveals that the a-IZO layer is in nano-crystalline form with grain size of a few nanometers. This is in agreement with the XRD results presented inFig. 1, indicating that the a-IZO layers deposited on both the Si and glass substrates are of the same microstructure.

The current density-applied voltage (J–V) characteristics of the SIS devices measured in dark environment at room temperature are shown inFig. 5. The results depict a rectifying characteristic with a turn-on voltage at a forward bias (positive bias on a-IZO) of about 0.48 V for all samples. To explain this, we schematically plot the corresponding band diagram of the SIS device inFig. 6. In this plot,ϕI–Sof 4.6 eV andϕS–Iof 4.1 eV are the electron affinities for a-IZO and Si, respectively. According to Ponpon et al. [22], the difference betweenϕS–IandϕI–Sgives the theoretical upper limit of the barrier height,ϕBE0.5 eV, in good agreement with forward turn-on bias of 0.48 V seen inFig. 5. Consequently, the interfacial SiOx layer at the a-IZO/Si interface might serve as a double Schottky barrier between both n-type layers. Under the forward bias, the band bending at Si/SiOxinterface will gradually induce an inversion region for SIS devices due to the presence of the built-in electricfield. Moreover, the current is dominated by the electrons in n-Si tunneling through the SiOxlayer into the a-IZOfilm due to the relatively largeϕS–I. In such a case, the increase of the electron concentration arising from the In-doping in the a-IZO layers enhances the band bending at the a-IZO/SiOx interface. Giving that the structure is consisted of three layers and at least two pivotal oxide/semiconductor interfaces, the 10% variation of the

400 600 800 1000 1200 1400 1600 0 10 20 30 40 50 60 70 80 90 100 2.0 2.5 3.0 3.5 4.0 0 2 4 6 8 10 (a) In/(Zn+In) = 0.25 (b) In/(Zn+In) = 0.4 (c) In/(Zn+In) = 0.5 (b) (c) (a) hυ (eV) ( α h ν ) (1 0 cm eV ) Transmittance (%) Wavelength (nm)

Fig. 3. Transmittance spectra of a-IZOfilms prepared at various In content ratios. The inset shows the determination of Egby extrapolating the Tauc plot of (αhv)2

against hv.

Fig. 4. Cross-sectional TEM micrographs of the SIS device with In/(ZnþIn)¼0.5.

-3 -2 -1 0 1 2 3 -0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 In/(Zn+In) = 0.25 In/(Zn+In) = 0.4 In/(Zn+In) = 0.5 Current Density (A cm -2) Voltage (V)

Fig. 5. J–V characteristic of a-IZO/SiOx/n-Si devices measured in the dark condition.

Fig. 6. Schematic energy band diagram of a-IZO/SiOx/n-Si hetero-junction

(4)

tunneling current density (namely in the range from 0.58 to 0.70 A cm2observed inFig. 5) is considered to be reasonable.

Theoretical studies by Shewchun et al.[1–3]indicated that the value ofϕBis intimately correlated with the open-circuit voltage (Voc) of SIS devices. Consequently, the presence of interfacial states may affect the carrier tunneling process and suppress the Vocof SIS devices. In order to evaluate the properties at the a-IZO/SiOxand SiOx/Si interfaces, photo-responsive C–V measurements were per-formed and the magnitude of Ditwas estimated by the following equation[8,9]:

Dit¼ CiΔV=qEg ð1Þ

where Ciis the insulator layer capacitance,ΔV is the voltage shift caused by the photo-induced change in charges at the a-IZO/SiOx and SiOx/Si interfaces, and q is the electron charge, respectively. As shown inFig. 7, the representative photo-responsive C–V profile of SIS devices with In/(ZnþIn)¼0.25 in the a-IZO layer gives ΔV¼0.13 V and, according to Eq.(1), Dit¼3.05  1011cm2eV1 for such a sample is obtained. Similar measurements on SIS devices with In/(ZnþIn)¼0.4 and In/(ZnþIn)¼0.5 in the a-IZO layer give Dit¼8.85  1010and Dit¼3.26  1010cm2eV1, respec-tively. The suppression of Ditdue to the increase of In/(ZnþIn) ratio can be readily seen. Note that the SIS sample in fact contains the a-IZO/SiOx and SiOx/Si interfaces. It is inferred that the variation of Dit delineated above should mainly correlate with the a-IZO/SiOxinterface since the growth condition of SiOxlayer on Si substrate is fixed in this study. The properties of SiOx/Si interfaces in all samples could thus be regarded as the same and their contribution to the change of Ditcould be neglected.

In order to clarify the roles of the bulk defects in a-IZO layer and that of the interfacial traps at a-IZO/SiOx interface, C–V measurements were also performed on the IZO/Al and Al/a-IZO/SiOxcapacitor samples.Fig. 8(a) and (b) separately shows the C–V profiles of the Al/a-IZO/Al MIM capacitor sample and a-IZO/ SiOx/Si MOS capacitor samples with the In/(ZnþIn) ratios of 0.25 and 0.5. The insets in thesefigures show the device configurations and we note that the thickness of SiOxlayer in a-IZO/SiOx/Si MOS capacitor samples has been increased to about 40 nm in order to obtain appropriate capacitance data.Fig. 8(a) indicates that both samples exhibit very similar C–V behaviors with relatively low capacitance densities, about 0.01 fFμm2at an applied bias of 1 V, implying that the bulk defect density in the a-IZO layer is relatively insensitive to the In/(ZnþIn) ratio. Thought the zinc vacancies in ZnO is escalated by the increase of In2O3content via the defect reaction In2O3

-ZnO

2InZnþ3OOþV″Zn, such an effect might be smeared out since the nano-crystalline IZO layer might be abound of the crystalline defects. Hence, the change of

In/(ZnþIn) ratio affects the bulk defect density of a-IZO layer insignificantly.

Fig. 8(b) depicts a shaper transition of capacitance from the depletion region to the accumulation region for the MOS capacitor sample of In/(ZnþIn)¼0.5 in comparison with that for sample of In/(ZnþIn)¼0.25. This implies a well-formed a-IZO/SiOxinterface in the sample of In/(ZnþIn)¼0.5[23]. Moreover, the addition of In2O3 in ZnO has been demonstrated to promote the n-type conduction by increasing the number of free carriers (electrons) in ZnO. For sample of In/(ZnþIn)¼0.5, the larger electron con-centration obtained in such an a-IZO layer (Fig. 2) may have, at least, partially alleviated the carrier trapping and hence reduced the effective Ditat the a-IZO/SiOxinterface. Consequently, increas-ing the In content of the a-IZO layer would benefit the Voc and short-circuit current density (Jsc) simultaneously and should play as the dominant factor in enhancing the performance of the SIS devices prepared in this study.

Fig. 9 presents the J–V characteristics for the SIS devices measured under the AM1.5 illumination condition and the Voc, Jsc, fill factor (FF) and η deduced from these J–V profiles are summarized in Table 1. It is evident from Table 1that both Voc and Jscare increased with increasing the In/(ZnþIn) ratio for these a-IZO/SiOx/n-Si SIS solar cells. In particular, for the In/(ZnþIn)¼0.5 sample, a remarkableη of 8.4% was obtained with the correspond-ing Voc, Jsc and FF being of 0.38 V, 45.1 mA cm2 and 49.7%, respectively. Fig. 10 presents the IPCE profile for the SIS device with In/(ZnþIn)¼0.5. The IPCE values exceed 80% at the wave-length range of 400 to 900 nm, decrease with the increase of wavelength, and drop to about 10% when the wavelengths are greater than 1100 nm. This indicates the incident light is mainly absorbed by Si with Eg¼1.1 eV (¼1127 nm), that is, the charge carriers are generated by the irradiation of light with energy

-2.0 -1.5 -1.0 -0.5 0.0 0.5 0 5 10 15 20 25

V

Δ

= 0.13 V

V

Δ

Dit = 3.05x1011 cm-2eV-1 Dark Light Capacitance (nF) Applied Voltage (V)

Fig. 7. Photoresponsive C–V illustration for SIS devices prepared at In/(ZnþIn)¼ 0.25.

Fig. 8. (a) C–V profiles of the Al/a-IZO/Al capacitor samples. The insets show the plot of the sample structure and the J–V curves of samples with In/(ZnþIn)¼0.25 and 0.5. (b) C–V characteristics of a-IZO/SiOx/Si MOS capacitor samples obtained at

(5)

greater than 1.1 eV and transported from Si to TCO layer via the tunneling process[1,2].

Since in the present study the only tuning parameter was the In/(ZnþIn) ratio used for preparing the a-IZO layer, it should be interesting to elaborate the possible underlying mechanisms in more details. For a solar cell, in general, theη value is determined mainly by three factors: (i) extensive photon harvesting; (ii) efficient separation of photo-excited carriers and (iii) the swiftness of carrier passage to external electrodes. Thefirst factor has to do with the properties of the light absorption layer (i.e., the n-Si substrate in our case), which is relatively irrelevant to the present study. We have addressed the second factor in our previous studies and found that the SiOxthickness giving rise to the best performance of the a-IZO/SiOx/n-Si SIS devices should be about 1.8 to 2.0 nm[6,7]. The thickness of this insulating layer was optimized to obtain a balance between the built-infield and the tunneling efficiency to drive the photo-excited carriers from the n-Si substrate to the a-IZO layer. Again, in the present study, we had fixed the condition of growing this layer in the PLD chamber under exactly the same condition. Thus, it should not be the primary reason for the observed differences. For the third factor, one would like to have smaller numbers carrier trap sites in both the bulk of

the a-IZO layer and at the a-IZO/SiOx interface (to prevent the photo-excited carriers from being trapped along their journey to the electrodes) as well as larger carrier mobility in the a-IZO layer (to allow the photo-excited carriers to move toward the electrode once they tunnel through the SiOx insulating layer). Within the context of this scenario, one should be able to further improve the performance of SIS devices by reducing Dit. In fact, the increase in Vocand Jscobtained by increasing the In/(ZnþIn) ratio in the a-IZO layer described above can also be explained within the same context. First, the C–V measurements shown in Figs. 7 and 8

clearly indicated that the increased In/(ZnþIn) ratio in the a-IZO layer dramatically lowered the effective Ditat a-IZO/SiOxinterface, presumably due to charge compensation on the interface sites. This not only would lead to the increment of built-in electricfield near the interface and change ϕB, but also would reduce the recombination centers at a-IZO/SiOx interfaces to provide addi-tional tunneling current and lower the series resistance (Rs) of the device, giving rise to an improvement in Jscas seen in a-IZO layer with high In/(ZnþIn) ratio. Finally, the slightly increased carrier mobility with increasing In/(ZnþIn) ratio (see Fig. 2) may also contribute to the overall enhancement of the performance of the present a-IZO/SiOx/n-Si solar cells.

4. Conclusions

In summary, we demonstrated a simple process to fabricate a-IZO/SiOx/n-Si solar cells by directly depositing a-IZO layers on the n-type Si substrates subjected to appropriate dry oxidation treatment. The ultra-thin SiOx layer formed on the Si surface during PLD deposition has been found to be crucial for the performance of the SIS devices in our previous studies. Present study further illustrates that the increase of the In content in a-IZO layer suppresses the effective Dit and Rs of the SIS devices, resulting in a substantial improvement of device performance. For the a-IZO layer with In/(ZnþIn)¼0.5 deposited at 250 1C, a corresponding SiOx layer about 2-nm thick and low Dit value around 1010cm2eV1were obtained. Moreover, the SIS device containing such a-IZO layer exhibited aη value of 8.4% with the Voc, Jscand FF being of 0.38 V, 45.1 mA cm2 and 49.7%, respec-tively. Our results evidently point an efficient way of fabricating and improving the performance of SIS-type solar cells.

Acknowledgment

This work is supported by the National Science Council, Taiwan, ROC, under the contracts No. NSC100-2221-E-009-055-MY3 and No. NSC101-2112-M-009-015-MY2. JYJ is supported in part by the MOE-ATU program operated at NCTU.

References

[1]J. Shewchun, J. Dubow, A. Myszkowski, R. Singh, The operation of the semiconductor–insulator–semiconductor (SIS) solar cell: theory, J. Appl. Phys. 49 (1978) 855–864.

[2]J. Shewchun, D. Burk, R. Singh, M. Spitzer, J. Dubow, The semiconductor– insulator–semiconductor (indium tin oxide on silicon) solar cell: character-istics and loss mechanisms, J. Appl. Phys. 50 (1979) 6524–6533.

[3]M. Spitzer, J. Shewchun, D. Burk, The operation of the semiconductor– insulator–semiconductor solar cell: barrier height lowering through interface states, J. Appl. Phys. 51 (1980) 6399–6404.

[4]A.K. Ghosh, C. Fishman, T. Feng, SnO2/Si solar cells-heterostructure or Schottky-barrier or MIS-type device, J. Appl. Phys. 49 (1978) 3490–3498. [5]H. Kobayashi, H. Mori, T. Ishida, Y. Nakato, Zinc oxide/n-Si junction solar cells

produced by spray-pyrolysis method, J. Appl. Phys. 77 (1995) 1301–1307. [6]H.W. Fang, S.J. Liu, T.E. Hsieh, J.Y. Juang, J.H. Hsieh, Fabrication and

character-ization of amorphous In–Zn–O/SiOx/n-Si heterojunction solar cells, Sol. Energy 85 (2011) 2589–2594. 0.0 0.1 0.2 0.3 0.4 0 10 20 30 40 50 In/(Zn+In) = 0.25 In/(Zn+In) = 0.4 In/(Zn+In) = 0.5

Photoc

urr

e

n

t D

e

n

sity (m

A cm

-2

)

Voltage (V)

Fig. 9. J–V characteristic of a-IZO/SiOx/n-Si SIS devices measured at the AM1.5

illumination condition. Table 1

Photovoltaic characteristics of SIS solar cells with different In contents. In/(ZnþIn) (at%) SiOxthickness (nm) Voc(V) Jsc (mA cm2) FF (%) η (%) 0.25 1.88 0.29 30.86 41.50 3.69 0.4 1.99 0.37 31.58 44.46 5.16 0.5 2.03 0.38 45.05 49.67 8.38 400 600 800 1000 1200 1400 0 10 20 30 40 50 60 70 80 90 100

900 nm

IPCE (%) Wavelength (nm)

(6)

[7]H.W. Fang, T.E. Hsieh, J.Y. Juang, Influences of SiOxlayer thickness on the characteristics of In–Zn–O/SiOx/n-Si Hetero-junction structure solar cells, Surf. Coat. Technol. 231 (2013) 214–218.

[8]J. Tan, M.K. Das, J.A. Cooper Jr., M.R. Melloch, Metal–oxide–semiconductor capacitors formed by oxidation of polycrystalline silicon on SiC, Appl. Phys. Lett. 70 (1997) 2280–2281.

[9]T. Hashizume, E. Alekseev, D. Pavlidis, K.S. Boutros, J. Redwing, Capacitance– voltage characterization of AlN/GaN metal–insulator–semiconductor struc-tures grown on sapphire substrate by metalorganic chemical vapor deposition, J. Appl. Phys. 88 (2000) 1983–1986.

[10]M.P. Taylor, D.W. Readey, M.F.A.M. van Hest, C.W. Teplin, J.L. Alleman, M.S. Dabney, L.M. Gedvilas, B.M. Keyes, B. To, J.D. Perkins, D.S. Ginley, The remarkable thermal stability of amorphous In–Zn–O transparent conductors, Adv. Funct. Mater. 18 (2008) 3169–3178.

[11]N. Ito, Y. Sato, P.K. Song, A. Kaijio, K. Inoue, Y. Shigesato, Electrical and optical properties of amorphous indium zinc oxidefilms, Thin Solid Films 496 (2006) 99–103.

[12]T. Minami, T. Kakumu, S. Takata, Preparation of transparent and conductive In2O3–ZnO films by radio frequency magnetron sputtering, J. Vac. Sci. Technol., A 14 (1996) 1704–1708.

[13]N. Naghavi, A. Rougier, C. Marcel, C. Guery, J.B. Leriche, J.M. Tarascon, Characterization of indium zinc oxide thinfilms prepared by pulsed laser deposition using a Zn3In3O6target, Thin Solid Films 360 (2000) 233–240.

[14]C. Huang, M. Wang, Q. Liu, Y. Cao, Z. Deng, Z. Huang, Y. Liu, Q. Huang, W. Guo, Physical properties and growth kinetics of co-sputtered indium–zinc oxide films, Semicond. Sci. Technol. 24 (2009) 095019.

[15]H. Hosono, M. Yasukawa, H. Kawazoe, Novel oxide amorphous semiconduc-tors: transparent conducting amorphous oxides, J. Non-Cryst. Solids 203 (1996) 334–344.

[16]D.S. Ginley, H. Hosono, D.C. Paine, Handbook of transparent conductors, in: H Hosono (Ed.), Transparent Amorphous Oxide Semiconductors for Flexible Electronics, Springer, New York, 2010, pp. 459–487.

[17] J.W. Jeon, D.W. Jeon, T. Sahoo, M. Kim, J.H. Baek, J.L. Hoffman, N.S. Kim, I.H. Lee, Effect of annealing temperature on optical band-gap of amorphous indium zinc oxidefilm, J. Alloys Compd. 509 (2011) 10062–10065.

[18]T. Moriga, Y. Hayashi, K. Kondo, Y. Nishimura, K. Murai, I. Nakabayashi, H. Fukumoto, K. Tominaga, Transparent conducting amorphous Zn–Sn–O films deposited by simultaneous dc sputtering, J. Vac. Sci. Technol., A 22 (2004) 1705–1710.

[19]A.J. Leenheer, J.D. Perkins, M.F.A.M. van Hest, J.J. Berry, R.P. O'Hayre, D. S. Ginley, General mobility and carrier concentration relationship in transpar-ent amorphous indium zinc oxidefilms, Phys. Rev. B: Condens. Matter 77 (2008) 115215.

[20] J. Tauc, R. Grigorovici, A. Vancu, Optical properties and electronic structure of amorphous germanium, Phys. Status Solidi A 15 (1966) 627–637.

[21]K. Ramamoorthy, M. Jayachandran, K. Sankaranarayanan, P. Misra, L.M. Kukreja, C. Sanjeeviraja, Epi-n-IZO thinfilms/(1 0 0) Si, GaAs and InP by L-MBE-a novel feasibility study for SIS type solar cells, Sol. Energy 77 (2004) 193–201.

[22] J.P. Ponpon, P. Siffert, Open-circuit voltage of MIS silicon solar cells, J. Appl. Phys. 47 (1976) 3248–3251.

[23] H. Wang, Y. Wang, J. Zhang, C. Ye, H.B. Wang, J. Feng, B.Y. Wang, Q. Li, Y. Jiang, Interface control and leakage current conduction mechanism in HfO2film prepared by pulsed laser deposition, Appl. Phys. Lett. 93 (2008) 202904.

數據

Fig. 2 summaries the transport properties of a-IZO films obtained by the Hall measurement
Fig. 4. Cross-sectional TEM micrographs of the SIS device with In/(ZnþIn)¼0.5.
Fig. 8 (b) depicts a shaper transition of capacitance from the depletion region to the accumulation region for the MOS capacitor sample of In/(Zn þIn)¼0.5 in comparison with that for sample of In/(Zn þIn)¼0.25
Fig. 10. IPCE profile for a-IZO/SiO x /n-Si SIS devices prepared at In/(ZnþIn)¼0.5.

參考文獻

相關文件

We are importers in the textile trade and would like to get in touch with ______ of this line.(A)buyers (B)suppliers (C)customers

This discovery is not only to provide a precious resource for the research of Wenxuan that has a long and excellent tradition in Chinese literature studies, but also to stress

6 《中論·觀因緣品》,《佛藏要籍選刊》第 9 冊,上海古籍出版社 1994 年版,第 1

The first row shows the eyespot with white inner ring, black middle ring, and yellow outer ring in Bicyclus anynana.. The second row provides the eyespot with black inner ring

Reading Task 6: Genre Structure and Language Features. • Now let’s look at how language features (e.g. sentence patterns) are connected to the structure

Now, nearly all of the current flows through wire S since it has a much lower resistance than the light bulb. The light bulb does not glow because the current flowing through it

[This function is named after the electrical engineer Oliver Heaviside (1850–1925) and can be used to describe an electric current that is switched on at time t = 0.] Its graph

Figure 6 shows the relationship between the increment ∆y and the differential dy: ∆y represents the change in height of the curve y = f(x) and dy represents the change in height