**6. ZnO/Cu2O heterojunction solar cells**

Until now, we have made Schottky barrier solar cells. As we could not improve their efficiency and their stability, we decided to make heterojunction p-n solar cells based on a ptype Cu2O thin films. We selected ZnO as an n-type semiconductor. ZnO is a transparent oxide that is widely used in many different applications, including thin film solar cells. The p-n junction was fabricated by potentiostatic deposition of the ZnO layer onto SnO2 conducting glass with a sheet resistance of 14 and potentiostatic deposition of Cu2O onto ZnO, Fig.22.

## **6.1 Electrochemical depositing of ZnO**

ZnO/Cu2O heterojunction solar cells were made by consecutive cathodic electrodeposition of ZnO and Cu2O onto tin oxide covered glass substrates. Zinc oxide (ZnO) was cathodically deposited on a conductive glass substrate covered with SnO2 as cathode by a potentiostatic method (Dalchiele et al.,2001, Izaki et al.,1998, Ng-Cheng-Chin et al.,1998). Conducting glass slides coated with SnO2 films are commercial samples. The electrolysis takes place in a

Low Cost Solar Cells Based on Cuprous Oxide 71

Fig. 23. X-ray diffraction spectrum of undoped electrodeposited ZnO film at 650C

Fig. 24. SEM micrograph of undoped electrodeposited pure ZnO

)2 versus energy *h*

making a plot of (

*h*

temperature value of 3.2 to 3.4 eV.

Thin films of ZnO grown by electrochemical deposition technique on SnO2/glass substrate

The transmission is relatively low (~ 50%) in the blue region (400–450 nm) Fig.25. The transmission maximum is about 60–70% through the red light region. Probably defects and

Assuming an absorption coefficient corresponding to a direct band to band transition and

through a linear fit. It was found to be 3.4 eV , which corresponds to the documented room

, the optical band gap energy *Eg* was determined

are optically transparent in a visible spectral region, extending to 300 nm wavelength.

structural irregularities are presented in the films, indicating low transmission.

Fig. 22. Profil of ZnO/Cu2O hetrojunction solar cells

simple aqueous 0,1M zinc nitrate [Zn (NO3)2] solution with pH about 6, maintained at 700C temperature. The cathodic process possibly can be described by the following reaction equations (Izaki & Omi, 1992):

$$\text{Zn(NO}\_3\text{)}\_2 \rightarrow \text{Zn}^{2+} + 2\text{NO}\_3^-$$

$$\text{NO}\_3^- + \text{H}\_2\text{O} + 2\text{e}^- \rightarrow \text{NO}\_2 + 2\text{OH}^-$$

$$\text{Zn}^{2+} + 2\text{OH}^- \rightarrow \text{Zn(OH)}\_2 \rightarrow \text{ZnO} + \text{H}\_2\text{O} \tag{5}$$

ZnO films were electrochemically grown at constant potential of 0.8 V between the anode and cathode. For a fixed value of the potential, a current density decreased with increasing the film thickness. The deposition time was varying from 10 min to 30 min. Deposited films were rinsed thoroughly in distilled water and allowed to dry in air at room temperature. The anode was zinc of 99.99% purity.

The deposition conditions of the thin films of Cu2O have been described in 2.1.1. The deposition potential is pH sensitive. It suggests, also and it has already been reported that the Cu2O layer was formed by the following reaction:

$$2\text{Cu}^{2+} + 2\text{e}^{\cdot} + 2\text{OH}^{\cdot} \rightarrow \text{Cu}\_2\text{O} + \text{H}\_2\text{O},\tag{6}$$

even this reaction does not explain the large pH dependence of deposition potential (Izaki et al. 2007, Wang & Tao, 2007). The present study was conducted, in a first instance, on undoped zinc oxide films and cuprous (I) oxide films. The structure of the films was studied by X-ray diffraction measurements using monochromatic Cu K radiation with a wavelength of 0,154 nm operated at 35 kV and 24 mA. Morphology and grain size was determined through micrographs on a JEOL JSM 6460 LV scanning electron microscope.

Figure 23 shows the X-ray diffraction patterns of ZnO film prepared at 0.8 V potential for 10 min. The Bragg angle of 2 was varied between 200 and 700. It can be seen that the film has crystalline structure. XRD peaks corresponding to ZnO (signed as C) and the substrate material SnO2 (signed as K) were determined with JCPDS patterns. The XRD spectrum indicates a strong ZnO peak with a (0002) or (1011) preferential orientation.

Figure 24 shows a scanning electron micrograph of undoped electrodeposited ZnO film.The photograph shows small rounded grains. It is difficult to determine the grain size from the micrograph. But using Scherrer's equation ( 0,9 cos *D* ), the apparent crystallite size of ZnO

is about 20nm, which means that it is nanostructured film

simple aqueous 0,1M zinc nitrate [Zn (NO3)2] solution with pH about 6, maintained at 700C temperature. The cathodic process possibly can be described by the following reaction

Zn(NO3)2 Zn2+ + 2NO3

ZnO films were electrochemically grown at constant potential of 0.8 V between the anode and cathode. For a fixed value of the potential, a current density decreased with increasing the film thickness. The deposition time was varying from 10 min to 30 min. Deposited films were rinsed thoroughly in distilled water and allowed to dry in air at room temperature.

The deposition conditions of the thin films of Cu2O have been described in 2.1.1. The deposition potential is pH sensitive. It suggests, also and it has already been reported that

 2Cu2+ + 2e- + 2OH- Cu2O + H2O, (6) even this reaction does not explain the large pH dependence of deposition potential (Izaki et al. 2007, Wang & Tao, 2007). The present study was conducted, in a first instance, on undoped zinc oxide films and cuprous (I) oxide films. The structure of the films was studied by X-ray diffraction measurements using monochromatic Cu K radiation with a wavelength of 0,154 nm operated at 35 kV and 24 mA. Morphology and grain size was determined through micrographs on a JEOL JSM 6460 LV scanning electron microscope. Figure 23 shows the X-ray diffraction patterns of ZnO film prepared at 0.8 V potential for 10 min. The Bragg angle of 2 was varied between 200 and 700. It can be seen that the film has crystalline structure. XRD peaks corresponding to ZnO (signed as C) and the substrate material SnO2 (signed as K) were determined with JCPDS patterns. The XRD spectrum

Figure 24 shows a scanning electron micrograph of undoped electrodeposited ZnO film.The photograph shows small rounded grains. It is difficult to determine the grain size from the

*D*

cos

 ), the apparent crystallite size of ZnO

indicates a strong ZnO peak with a (0002) or (1011) preferential orientation.

+ H2O +2e NO2 +2OH

**SnO2 ZnO**

Zn2+ + 2OH

The anode was zinc of 99.99% purity.

the Cu2O layer was formed by the following reaction:

micrograph. But using Scherrer's equation ( 0,9

is about 20nm, which means that it is nanostructured film

equations (Izaki & Omi, 1992):

Fig. 22. Profil of ZnO/Cu2O hetrojunction solar cells

NO3  -**glass**


**C-graphite**

Zn(OH)2 ZnO +H2O (5)

Fig. 23. X-ray diffraction spectrum of undoped electrodeposited ZnO film at 650C

Fig. 24. SEM micrograph of undoped electrodeposited pure ZnO

Thin films of ZnO grown by electrochemical deposition technique on SnO2/glass substrate are optically transparent in a visible spectral region, extending to 300 nm wavelength.

The transmission is relatively low (~ 50%) in the blue region (400–450 nm) Fig.25. The transmission maximum is about 60–70% through the red light region. Probably defects and structural irregularities are presented in the films, indicating low transmission.

Assuming an absorption coefficient corresponding to a direct band to band transition and making a plot of (*h*)2 versus energy *h*, the optical band gap energy *Eg* was determined through a linear fit. It was found to be 3.4 eV , which corresponds to the documented room temperature value of 3.2 to 3.4 eV.

Low Cost Solar Cells Based on Cuprous Oxide 73

10 15 20 25 30 35 40 45 50 60 **I/mW/cm2**

I/mW/cm<sup>2</sup>

Cu2O/ZnO/SnO2 cell *Vb*(mV) *Voc*(mV) just made 368 330 after few days 276 240

Table 4. Values of barrier height *Vb* and open circuit voltage *Voc* for just made cell and after

Fig. 27**.** Volt-current characteristics of the Cu2O/ZnO/SnO2 solar cell upon 50mW/cm2

The values of barrier height *Vb* and the open circuit voltage *Voc* upon illumination of 100 mW/cm2 for just made cell and the cell after few days are presented in table 4. Also in this table are given their values after few days of depositing. The values of the barrier height are great than the values of open circuit voltage *Voc*. The grate *Vb* gives the great *Voc*, that

Fig. 26. Dependence of the *Voc* and *Isc* vs. solar irradiation

correspondent to the photovoltaic theory.

**jse/A/cm2**

 *j***sc/A/cm2**

Voc/mV jse/µA/cm2

*j*sc/A/cm<sup>2</sup>

0

few days.

Illumination

50

100

150

**Voc/mV**

200

250

300

350

Fig. 25. Optical transmission spectrum of ZnO film

#### **6.2 Some characteristic of the cells**

To complete Cu2O/ZnO/SnO2 heterojunction as solar cell, thin layer of carbon paste or carbon spray was deposited on the rear of the Cu2O. Front wall cells were formed. A carbon back contact was chosen because of simplicity and economy of the cell preparation and because the cells with carbon give high values of the short circuit current density despite the evaporated layer of nickel. The total cell active area was 1 cm2. Antireflectance coating or any special collection grids have not been deposited. The best values of the open circuit voltage *Vo c*= 330 mV and the short circuit current density *Isc* = 400 µm/cm2 were obtained by depositing carbon paste and illumination of 100 mW/cm2. The *Voc* increases as logarithmic function with solar

radiation, ( 0 ln 1 *sc kT <sup>I</sup> Voc e I* ). The *Isc* increases linear with solar radiation, (Fig.26).

Our investigations show that the ZnO layer improves the stability of the cells. That results in a device with better performances despite of the Schhotky barrier solar cells (Cu2O/SnO2). First, the cells show photovoltaic properties without annealing, because potential barrier was formed without annealing. The barrier fell for a few days which result in decreasing the open circuit voltage despite the values of *Voc* for just made cells. It decreases from 330 mV to 240 mV. But after that the values of *Voc* keep stabilized, because of stabilized barrier potential. It wasn't case with Schotkky barrier solar cells, because barrier potential height decreases with aging. In ZnO/Cu2O cells, thermal equilibrium exists. The *Voc* decreases and *Isc* increases with increasing the temperature, that is characteristic for the real solar cell. It could be seen from the current-voltage (*I-V*) characteristic in incident light of 50mW/cm2, Fig.27.

Barrier potential height was determined for one device from capacitance measurement as a function of reverse bias voltage at room temperature. Capacitance dependence of reverse bias voltage at room temperature was measured by RCL bridge on alternating current (HP type) with bilt source with 1000 Hz frequency. Results for (1/C2) versus voltage are shown in Figure 28. The Cu2O/SnO2 cells without the ZnO layer show a lower *Voc*. The improvement in *Voc* could be due to the increase of the barrier height using ZnO layer as ntype semiconductor.

T/ %

**6.2 Some characteristic of the cells** 

radiation, (

type semiconductor.

Fig. 25. Optical transmission spectrum of ZnO film

0

 

ln 1 *sc kT <sup>I</sup> Voc e I*

/ nm

To complete Cu2O/ZnO/SnO2 heterojunction as solar cell, thin layer of carbon paste or carbon spray was deposited on the rear of the Cu2O. Front wall cells were formed. A carbon back contact was chosen because of simplicity and economy of the cell preparation and because the cells with carbon give high values of the short circuit current density despite the evaporated layer of nickel. The total cell active area was 1 cm2. Antireflectance coating or any special collection grids have not been deposited. The best values of the open circuit voltage *Vo c*= 330 mV and the short circuit current density *Isc* = 400 µm/cm2 were obtained by depositing carbon paste and illumination of 100 mW/cm2. The *Voc* increases as logarithmic function with solar

Our investigations show that the ZnO layer improves the stability of the cells. That results in a device with better performances despite of the Schhotky barrier solar cells (Cu2O/SnO2). First, the cells show photovoltaic properties without annealing, because potential barrier was formed without annealing. The barrier fell for a few days which result in decreasing the open circuit voltage despite the values of *Voc* for just made cells. It decreases from 330 mV to 240 mV. But after that the values of *Voc* keep stabilized, because of stabilized barrier potential. It wasn't case with Schotkky barrier solar cells, because barrier potential height decreases with aging. In ZnO/Cu2O cells, thermal equilibrium exists. The *Voc* decreases and *Isc* increases with increasing the temperature, that is characteristic for the real solar cell. It could be seen from the

Barrier potential height was determined for one device from capacitance measurement as a function of reverse bias voltage at room temperature. Capacitance dependence of reverse bias voltage at room temperature was measured by RCL bridge on alternating current (HP type) with bilt source with 1000 Hz frequency. Results for (1/C2) versus voltage are shown in Figure 28. The Cu2O/SnO2 cells without the ZnO layer show a lower *Voc*. The improvement in *Voc* could be due to the increase of the barrier height using ZnO layer as n-

current-voltage (*I-V*) characteristic in incident light of 50mW/cm2, Fig.27.

). The *Isc* increases linear with solar radiation, (Fig.26).

Fig. 26. Dependence of the *Voc* and *Isc* vs. solar irradiation

The values of barrier height *Vb* and the open circuit voltage *Voc* upon illumination of 100 mW/cm2 for just made cell and the cell after few days are presented in table 4. Also in this table are given their values after few days of depositing. The values of the barrier height are great than the values of open circuit voltage *Voc*. The grate *Vb* gives the great *Voc*, that correspondent to the photovoltaic theory.


Table 4. Values of barrier height *Vb* and open circuit voltage *Voc* for just made cell and after few days.

Fig. 27**.** Volt-current characteristics of the Cu2O/ZnO/SnO2 solar cell upon 50mW/cm2 Illumination

Low Cost Solar Cells Based on Cuprous Oxide 75

Even low efficiency it may be acceptable in countries where the other alternative energy

Part of this work has been performed within the EC funded RISE project (FP6-INCO-

Dalchiele E.A., Giorgi P.,.Marotti R.E, at all. Electrodeposition of ZnO thin films on n-Si(100),

GeorgievaV. Ristov M. (2002) Electrodeposited cuprous oxide on indium tin oxide for solar

Izaki M, Ishizaki H., Ashida A. at all. (1998), *J.Japan Inst.Metals*, Vol. 62,No.11pp.1063-1068 Izaki M., Shinagawa T., Mizuno K., Ida Y., Inaba M., and Tasaka A., (2007) Electrochemically

Katayama J., Ito K. Matsuoka M. and Tamaki J., (2004) Performance of Cu2O/ZnO solar cells

Kemell M., Dartigues F., Ritala M., Leskela M, (2003) Electrochemical preparation of In and Al doped Zno thin films for CuInSe2 solar cells, *Thin Solid Films* 434 20-23 Machado G., Guerra D.N., Leinen D., Ramos-Barrado J.R. Marotti R.E., Dalchiele E.A.,

Minami T.,.Tanaka H, Shimakawa T., Miyata T., Sato H., (2004) High-Efficiency Oxide Heterojunction Solar cells Using Cu2O Sheets *Jap.J.Appl.Phys*.,43, p.917-919 Mukhopadhyay A.K.,.Chakraborty A.K, Chattarjae A.P. and.Lahriri S.K, (1992), *Thin Solid* 

Papadimitriou L., Valassiades O. and Kipridou A., (1990), *Proceeding,* 20th ICPS,

Ng-Cheng-Chin F., Roslin M.,.Gu Z.H and Fahidy T.Z., (1998) *J.Phys.D.Appl.Phys*.31 L71-L7

Rakhshani A.E (1986) Preparation, characteristics and photovoltaic properties of cuprous

Rakhshani A.E., Jassar A.A.Al and.Varghese J. (1987) Electrodeposition and characterization

Rakhshani A.E.and Varghese J., (1987) Galvanostatic deposition of thin films of cuprous

oxide-A review *Solid-State Electronics* Vol. 29.No.1. pp.7-17.

Rakhshani A.E., Makdisi Y. and Mathew X., (1996), *Thin Solid Films*, 288, 69-75

constructed p-Cu2O/n-ZnO heterojunction diode for photovoltaic device

prepared by two-step electrodeposition *Journal of Applied Electrochemistry*, 34: 687-

(2005), Idium doped zinc oxide thin films obtained by electrodeposition ,*Thin Solid* 

509161). The authors want to thank the EC for partially funding this project.

applications, *Solar Energy Materials & Solar Cells* 73, p 67-73

Jayanetti J.K.D., Dharmadasa I.M, (1996), *Solar Energ.Mat.andSolar Cells* 44 251-260

*Solar Energy Materials &Solar Cells* 70 (2001) 245-254

Olsen L.C., Bohara R.C., Urie M.W., (1979) *Appl.Phys.Lett,* 34, p. 47 Olsen L.C., Addis F.W. and Miller W., (1982-1983), *Solar Cells*, 7 247-249 Papadimitriou L., Economu N.A and Trivich, (1981), *Solar Cells*, 3 73

of cuprous oxide *Thin Solid Films*, 148,pp.191-201

Izaki M., Omi T,J*.* (1992).,*Electrochem.Soc*.1392014

*J.Phys.D:Appl.Phys*.40 3326-3329.

sources are much more expensive.

**8. Acknowledgment** 

**9. References** 

692,

*Films* 490 124-131

*Films*, 209, 92-96

Thessaloniki, 415-418

Rai B.P., (1988) Cu2O Solar Cells *Sol. Cells* 25 p.265.

oxide *Solar Energy Materials*, 15,23,

Stareck, U.S. Patents 2, 081, 121 *Decorating Metals*, 1937

Fig. 28. 1/C2 vs applied voltage of Cu2O/ZnO/SnO2 cell

The values Voc=316 mV, Isc=0,117 mA/cm2, fill factor =0,277, upon 50mW/cm2 illumination are compared with the values: Voc=190 mV, Isc=2,08 mA/cm2, fill factor = 0,295; upon 120mW**/cm2** illumination (Katayama et al. 2004) made with electrochemical deposition technique. Maybe doping of the ZnO films with In, Ga and Al (Machado et al., 2005, Kemell et al., 2003) will decrease the resistivity and increase the electro conductivity of the films, consequently and the short circuit current density of the cells.

#### **7. Conclusion**

The performance of the Cu2O Schottky barrier solar cells are found to be dependent on the starting surface material, the type of the junction, post deposition treatment and the ohmic contact material. Better solar cells have been made using an heterojunction between Cu2O and n-type TCO of ZnO. It is a suitable partner since it has a fairly low work function. Our investigation shows that the ZnO layer improves the stability of the cells. That results in a device with better performances despite of the Schhotky barrier solar cells (Cu2O/SnO2). First, the cells show photovoltaic properties without annealing, because potential barrier was formed without annealing. To improve the quality of the cells, consequently to improve the efficiency of the cells, it has to work on improving the quality of ZnO and Cu2O films, because they have very high resistivity, a factor which limits the cells performances. Doping of the ZnO films with In, Ga and Al will decrease the resistivity of the deposited films and increase their electroconductivity. SEM micrographs show that same defects are present in the films which act as recombination centers. Behind the ohmic contact, maybe one of the reason for low photocurrent is just recombination of the carriers and decreasing of the hole cocentracion with the time. The transmittivity in a visible region have to increase. Also, it is necessary to improve the ohmic contact, consequently to increase the short circuit current density (*Isc*). For further improvement of the performances of the cells maybe inserting of a buffer layer at the heterojunction between Cu2O and ZnO films will improve the performance of the cells by eliminating the mismatch defects which act as recombination centers. Also it will be protection of reduction processes that maybe exists between ZnO and Cu2O.

Even low efficiency it may be acceptable in countries where the other alternative energy sources are much more expensive.
