Fabrication and photovoltaic performance of aligned ZnO nanorod arrays on silicon heterojunction solar cell

Shiguang Shanga, *, Rui Wanga, Xiaobao Jub, Xiaoxian Lib, Xinxia Zhengb

aSchool of Electronic Engineering, Xi`an University of Posts and Telecommunications, Xi`an 710121, China

bXi`an Huanghe Photovoltaic Technology Co., Ltd

Abstract: Heterojunction solar cells (HSCs) with the structure of p-Si/n-ZnO nanorod arrays were prepared by synthesizing an aligned ZnO nanorod arrays on patterned p-type silicon substrate through a low temperature hydrothermal method. ITO and Al films using as the front and back contact electrode layers were deposited by DC-magnetron sputtering, respectively. The influences of various process parameters, such as seed layer annealing temperature and hydrothermal growth time, on the crystal structure, surface morphology, and optical property of the aligned ZnO nanorod arrays were investigated by using X-ray diffraction (XRD), field emission scanning electron microscopy (FESEM), high resolution transmission electron microscopy (HRTEM), selected area electron diffraction (SAED) and UV-vis diffuse reflectance spectra (DRS). The optimized short-circuit current density and allover energy conversion efficiency of the p-Si/n-ZnO nanorod array HSCs were 11.475 mA·cm-2 and 2.0%, respectively.

Keywords: Heterojunction solar cells; Seed layer; ZnO nanorod; Conversion efficiency

1. Introduction

Zinc oxide (ZnO) is a semiconductor material with a direct band gap of 3.3eV and large exciton binding energy of 60 meV at room temperature [1], high transparency in the visible wavelength region [2] and nontoxicity [3]. Due to the superior properties of ZnO nanomaterials, such as high surface-to-volume ratios and quantum confinement effects, over those of bulk materials, a variety of nanostructures including nanoparticles [4], nanorods(NRs) [5], nanowires [6], nanoflowers [7] and nanowalls [8] have been investigated extensively for various application purposes depending on their morphologies and fabrication methods. Among them, the nanorods or nanowires have received the greatest attention for their potential applications in various fields, such as photocatalysts [9], gas sensor [10], dye-sensitized solar cells [11, 12], field emission [13] and ultraviolet photodetector [14]. Besides, single crystalline 1D ZnO nanostructure arrays have much better electron transportation capability by providing a direct conduction path for electron transport and reducing the number of grain boundaries greatly [15].

To achieve high-efficiency solar cell, the front side with high surface area and low reflectivity is essential, because it allows adsorption of sufficiently large sun light. Therefore, the surface of the front side is crucial to the photovoltaic performance of p-Si/n-ZnO nanorod array HSCs and many research groups devote to decreasing the reflectivity of the front side or building hierarchical structures [16, 17]. As previously reported, the nanorod structures were used as an antireflective layer to improve photovoltaic conversion efficiency for film solar cells [18, 19]. Moreover, ZnO material was dominated to extend the solar spectrum from visible light to ultraviolet light due to their different bandgap energies [15]. Base on the benefits of nanostructure and properties, a version of the novel solar cells in which the traditional film structures are replaced by an array of vertically aligned crystalline ZnO nanorods collocated with p-type silicon single crystal were introduced.

2. Experimental details

Fig.1 shows the schematic diagrams of fabrication procedure for p-Si/n-ZnO nanorod array HSCs. As shown in Fig.1(a-b), in order to obtain ohmic contact between p-type (100) silicon with a resistivity of 1-5 Ω·cm and Al metal electrode, the heavily boron-doped (p+) rear surface as low as 0.032 ohm/sq was obtained by semiconductor manufacturing process, such as oxidation and photolithography, etching and thermal diffusion. Then, the photoresist with a thin layer was coated on the purity surface of silicon dioxide covered p-type silicon substrate by the spin-coating technique, and a purpose-made patterned shadow mask was placed on the surface of photoresist, the mask was subsequently irradiated by UV light, as shown in Fig.1(e). The patterned photoresist and silicon dioxide layer with an active area 0.5×0.5 cm2 were respectively obtained by using development and wet-etching processes in Fig.1(f). ZnO target was employed to deposit seed layer on the p-type silicon surface with the patterned photoresist and silicon dioxide layer by RF-magnetron sputtering at room temperature (RT), and the patterned ZnO seed layer was obtained after the photoresist was lifted off, as shown in Fig.1(g). Prior to the deposition of ZnO seed layer, the surface of p-type silicon substrate was sequentially cleaned by deionization water, and dried with nitrogen purge. The annealing temperatures of the ZnO seed layer were 250℃, 350℃ and 450℃ for 6 hours, respectively. The annealing condition for seed layers was employed to improve the crystalline quality with well-aligned c-axis crystal orientation for the subsequent NR growth [3].

Fig.1 Schematic diagrams of fabrication procedure for p-Si/n-ZnO nanorod array HSCs

The aqueous chemical solution containing equal molar (0.05 molar) zinc acetate dehydrate(Zn(CH3COO)2) and hexamethylenetetramine (C6H12N4) were dissolved in 50 mL deionization water and stirred at room temperature for 30 minutes to ensure complete mixing. The hydrothermal ZnO nanorod growth was carried out in a high pressure reaction kettle by immersing the ZnO seeded silicon substrate in the growth solutions and heating at 80 for 3h and 5h, respectively. Particularly, the ZnO seed layer was faced downwards and leaned about 45° against the reaction kettle in order to achieve a minimum temperature gradient growth effect [10]. Then, as-grown aligned nanorod arrays were removed from solution, rinsed with deionization water, and dried with nitrogen blow. ITO and Al films were respectively deposited on the front (n-ZnO nanorod arrays) and back (p+-type silicon) sides as ohmic contact electrodes for better electrical performance by DC magnetron sputtering, as shown in Fig.1(h). The main parameters of the fabrication process for p-Si/n-ZnO nanorod array HSCs are given in table 1.

Table.1 The main parameters for fabricating p-Si/n-ZnO nanorod array HSCs

Parameters

Oxidation

Deposition

Seed layer

Al

ITO

Temperature

960 °C

RT

RT

RT

Volume ratio of Ar/O2

0/1

2/1

1/0

1/0

Pressure (Pa)

In air

0.9

1.2

1.2

Gas flow (sccm)

100

30

40

40

Time(minute)

90

10

15

15

Powder (W)

 

150(RF)

100(DC)

100(DC)

The crystalline phases of the ZnO seed layers and ZnO nanorod arrays were characterized by small angle X-ray diffraction in Shimadzu diffractometer (XRD-6000, Japan) using Cu-Ka radiation. The morphologies and structures of the as-grown nanorod arrays were recorded by field emission electron microscope (FESEM, JSM-6700F) operating at 5.0 kV and high resolution transmission electron microscopy (HRTEM, JEOL, JEM-2010). The UV-Vis diffuse reflectance spectra (DRS) were measured at room temperature by PerkinElmer LAMBDA 1050. The current-voltage characteristics of solar cells were measured by precision source/measure unit (Keysight, B2902A) under standard AM1.5 illumination (100 mW·cm2) using a solar simulator (Abet Technologies Inc., 10500) equipped with a xenon arc light source.

3. Results and discussion

Fig.2 shows X-ray diffraction patterns of seed layers and as-grown ZnO nanorod arrays on p-type silicon substrates. All the X-ray diffraction patterns of ZnO nanorod arrays have the approximate same peak locations of 34.62, 36.36, 47.70, 63.02 and 68.20, corresponding to (002), (101), (102), (103) and (112) peaks, and the diffraction peaks can be well indexed to hexagonal wurtzite structures (JCPDS card No. 36-1451). And no diffraction peaks of any other impurities are observed. A comparison of X-ray diffraction patterns between seed layer at annealing temperature 450 and grown ZnO nanorod arrays shown in Fig.2(a) reveals that the preferred orientation along (002) polar surface is enhanced with increasing growth time of ZnO nanorod arrays, which implies that most of the ZnO nanorod arrays grow with c-axis vertical to the silicon substrate, and affirms that the crystallinity of the samples increases and the ZnO nanorods exhibit some random growth, but the nanorods with c-axis being vertical to the substrates are still dominated. As shown in Fig.2(b), the intensities of (002) and (103) peaks of ZnO nanorod arrays with growth time 5h become stronger and sharper with increasing the seed annealing temperature. The result indicates that the crystallinity of ZnO nanorod arrays is improved and depends on the ZnO seed annealing temperature.

Fig.2 X-ray diffraction patterns of ZnO nanorod arrays at different growth conditions (a) growth time, and (b) seed annealing temperature

Fig.3 shows the FESEM images of ZnO seed layers and as-grown nanorod arrays. As shown in Fig.3(a-c), a significant change in surface morphology and grain size is found with the increase of seed annealing temperature. The seed layers annealed at 250 are composed of many small crystalline grains with similar morphologies and the corresponding ZnO particles have uniform sizes. With increasing annealing temperatures in sequence, the size of the ZnO grains increases, and the average sizes annealed at 350 and 450 are approximately 25 nm and 30 nm, respectively. Fig.3(d-f) is the microstructure morphology of ZnO nanorods arrays grown on above three different seed layers for about 3h (marked sample 1-3). The seed layer area is coated with highly uniform and densely ZnO nanorod arrays, where diameters were measured using the top view images and lengths were measured using the cross-sectional images (inset in Fig.2(f)). The average diameter of ZnO nanorods are 80nm, 90nm and 120nm with length about 2.5 μm, respectively. Their growth mechanism could be attributed to the substrate-induced oriental nucleation and fast growth under thermodynamic equilibrium state. An increase of the nanorod diameter as a function of the growth time can be inferred from the top view FESEM observations in Fig.3(g-i), and the corresponding diameters of ZnO nanorod arrays grown for about 5h (marked sample 4-6) are 260 nm, 270nm and 350 nm with the length about 4μm (inset in Fig.2(i)), respectively. The results indicate that the morphology of ZnO nanorods strongly depends on the annealing temperature of seed layers and the growth time of ZnO nanorod arrays.

Fig.3 FESEM images of ZnO seed layers annealed at various temperatures (a) 250, (b) 350 min, (c) 450, and ZnO nanorod arrays grown on above three seed layers for (d-f) about 3h and (g-i) about 5h

Fig.4 shows the HRTEM images of the individual ZnO nanorods, which again confirm the detailed morphology and structure of FESEM image in Fig.3(e). A typical low magnification TEM image of the individual ZnO nanorods is shown in Fig.4(a-c). It reveals that the ZnO nanorod had an average diameter of around 90 nm and that an outer boundary of the ZnO nanorod was distinctly different from the ZnO nanorod core. The outer layer, with a thickness of about 7.8 nm (Fig.4(c)), is almost similar to the previous reference [20]. The atomic structure of the individual ZnO nanorod is shown in Fig.4(d), confirming not only the nanorod growth along the [002] direction (the lattice spacing of ~0.266 nm along the longitudinal axis direction), but also the single crystalline character of the nanorods. The result is in good correlation with the X-ray diffraction in Fig.2. The crystalline nature of the sample was further determined using selected area electron diffraction (SAED) analysis, and the result in inset of Fig.4(d) confirms the ZnO nanorods have a single crystalline nature.

Fig.4 HRTEM images of ZnO nanorod arrays (a-c) low magnification image, (d) high magnification image. Inset: corresponding SAED pattern

Fig.5 shows the FESEM images of ZnO nanorod arrays covered with ITO film. As shown in Fig.5(a), the ZnO nanorod arrays were connected with each other by ITO film and the isolated ZnO nanorod arrays can’t be observed, which provides an uninterrupted conduction paths for electron transport. In Fig.5(b), the ZnO nanorods with uniform diameter became asymmetry with match rod shape, and the diameter of the nanorods increases several times from bottom to top of the nanorod arrays. The inset in Fig.5(b) reveals that the ZnO nanorod arrays were covered with conductive ITO film, which was composed with many small particles of ITO grains, which could lead to the increase in the special surface area and promote the sunlight absorption of the ZnO nanorod arrays.

Fig.5 FESEM images of ZnO nanorod arrays coated with ITO (a) top view, (b) cross section

Fig.6 shows the UV-Vis diffuse reflectance spectra (DRS) of as-grown ZnO nanorod arrays at room temperature. For the samples 1-6, the average reflectance of ZnO nanorod arrays is 49.8%, 26.6%, 37.7%, 27.7%, 21.1%, and 24.2% in the wavelength range of 200-1100nm, respectively. As shown in Fig.6 (a), the absorption wavelength for the samples 1-6 is obtained at a wavelength about 390nm, 389nm, 395nm, 388nm, 393nm, 392nm, respectively. The optical band gap (Eg) is determined from the Kubelka-Munk mode, the corresponding band gap is 3.18 eV, 3.19 eV, 3.14 eV, 3.20 eV, 3.16 eV, and 3.16 eV, which agree with the reference [21]. Compared with the samples 1-3, the samples 4-6 show a lower reflectance in the wavelength range of 380-1100nm, which indicates that the absorption of ZnO nanorod arrays increase with the growth time increasing. The result in Fig.6 (b) shows that the average reflectance of samples decrease at first and then increase with increasing the seed layer annealing temperature, and the ZnO nanorod arrays with growth time about 5h on seed layers annealing at 350℃ has the lowest average reflectance of 21.1%.

 

Fig.6 (a)DRS of ZnO nanorod arrays at room temperature, (b)Variation of average reflectance as a function of annealing temperature  

Fig.7 presents the current-voltage (J-V) characteristics of the heterojunction solar cell with the Al/p+-Si/p-Si/n-ZnO/ITO structure operated under standard AM1.5 illumination. As shown in Fig.7 (samples 1-3), the average short-circuit current density (JSC) 10.59 mA·cm-2 and open-circuit voltage (VOC) 0.327V are observed for the p-Si/n-ZnO nanorod array HSCs. And the corresponding results observed from Fig.7(curves 4-6) are about 11.437 mA·cm-2 and 0.345V for the p-Si/n-ZnO nanorod array HSCs. The results indicate that the major trends in the data are a small increase in open voltage Voc and short-circuit Jsc with increasing the diameter and length of ZnO nanarod arrays originated from the growth time of the nanorod arrays, but the continuous increase of the diameter and length of ZnO nanorod arrays will lead to the decrease of the properties of the p-Si/n-ZnO nanorod array HSCs..

Fig.7 Current-voltage characteristics of the p-Si/n-ZnO nanorod array HSCs

The best stabilized conversion efficiency (h) of 2.0% (VOC=0.355V, JSC=11.47 mA·cm-2, fill factor (FF) =0.490) was achieved for the p-Si/n-ZnO nanorod array HSCs with an area of 1.0 cm2 without any antireflection coatings (Fig.7, curve 5).The photovoltaic parameters of p-Si/n-ZnO nanorod array HSCs were shown as table.2. The allover energy conversion efficiency of p-Si/n-ZnO nanorod array HSCs is still below the state of the art for crystalline silicon solar cell and the deep research are still carried out.

Table.2 Photovoltaic parameters of p-Si/n-ZnO nanorod array HSCs

Sample

VOC(V)

JSC(mA·cm-2)

FF

h

Sample1

0.305

10.619

0.445

1.44%

Sample2

0.340

10.228

0.494

1.72%

Sample3

0.335

10.925

0.422

1.54%

Sample4

0.355

10.664

0.451

1.71%

Sample5

0.355

11.475

0.490

2.00%

Sample6

0.325

12.172

0.454

1.80%

4. Conclusion

In the paper, a new solar cell model based on n-ZnO nanorod array film as front layer and p-type silicon as rear region was proposed. The ZnO nanorod array film will act as an active n-type layer as well as antireflection coating saving considerable processing cost. The structural, morphological, and optical properties of the ZnO nanorod arrays strongly depend on the seed layer property resulted from thermal annealing temperature and the growth time of the ZnO nanorod arrays. The ZnO nanorod arrays have a prominent wurtzite hexagonal structure with preferred orientation along (002) polar surface. The average reflectance of samples decrease at first and then increase with increasing the seed layer annealing temperature. The best conversion efficiency of the heterojunction solar cells was 2.0% with fill factor 0.49.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No.61874087, 61634004), and Foundation of Shaanxi Educational Committee (No.16JK1693). We would like to thank Ping Zhao for the help in making the precision source/measure unit available to us.

References

[1] E. Pourshaban, H. Abdizadeh, M.R. Golobostanfard. ZnO nanorods array synthesized by chemical bath deposition: effect of seed layer sol concentration [J]. Procedia Materials Science, 2015, 11: 352-358.

[2] Z. Ben Ayadi, L. El Mir, K. Djessas, et al. Effect of the annealing temperature on transparency and conductivity of ZnO: Al thin films [J]. Thin Solid Films, 2009, 517(23): 6305-6309.

[3] W. Khan, S.D. Kim. Ultra-violet photo-response characteristics of p-Si/i-SiO2/n-ZnO heterojunctions based on hydrothermal ZnO nanorods [J]. Materials Science in Semiconductor Processing, 2017, 66: 232-240.

[4] K. Zuo, B.R.K. Blackman, J.G. Williams, et al. The mechanical behaviour of ZnO nano-particle modified styrene acrylonitrile copolymers [J]. Composites Science and Technology, 2015, 113(5): 9-18.

[5] M. Shukla, Pramila, T. Dixit, et al. Influence of aspect ratio and surface defect density on hydrothermally grown ZnO nanorods towards amperometric glucose biosensing applications [J]. Applied Surface Science, 2017, 422(15): 798-808.

[6] T. Demes, C. Ternon, F. Morisot, et al. Mechanisms involved in the hydrothermal growth of ultra-thin and high aspect ratio ZnO nanowires [J]. Applied Surface Science, 2017, 410(15): 423-431.

[7] A. Umar, H. Algarni, S.H. Kim, et al. Time dependent growth of ZnO nanoflowers with enhanced field emission properties [J]. Ceramics International, 2016, 42(11): 13215-13222.

[8] N. Islavath, D. Das, S.V. Joshi, et al. Seed layer-assisted low temperature solution growth of 3D ZnO nanowall architecture for hybrid solar cells [J]. Materials & Design, 2017, 116(15): 219-226.

[9] F. Solis-Pomar, A. Jaramillo, J. Lopez-Villareal, et al. Rapid synthesis and photocatalytic activity of ZnO nanowires obtained through microwave-assisted thermal decomposition [J]. Ceramics International, 2016, 42(16): 18045-18052.

[10] C.H. Tan, S.T. Tan, H.B. Lee, et al. Automated room temperature optical absorbance CO sensor based on In-doped ZnO nanorod [J]. Sensors and Actuators B, 2017, 248: 140-152.

[11] T. Bai, Y. Xie, J. Hu, et al. Novel one-dimensional ZnO nanorods synthesized through a two-step post-treatment for efficiency enhancement of dye-sensitized solar cells [J]. Journal of Alloys and Compounds, 2015, 644(25): 350-353.

[12] Y. Chae, J.T. Park, J.K. Koh, et al. All-solid, flexible solar textiles based on dye-sensitized solar cells with ZnO nanorod arrays on stainless steel wires [J]. Materials Science and Engineering B, 2013, 178(17): 1117-1123.

[13] X.F. Su, J.B. Chen, R.M. He, et al. The preparation of oxygen-deficient ZnO nanorod arrays and their enhanced field emission [J]. Materials Science in Semiconductor Processing, 2017, 67(15): 55-61.

[14] D. Zhang, Y. Sheng, J. Wang, et al. ZnO nanowire photodetectors based on Schottky contact with surface passivation [J]. Optics Communications, 2017, 395(15): 72-75.

[15] S. Bai, L. Chen, D. Li, et al. Different morphologies of ZnO nanorods and their sensing property [J]. Sensors and Actuators B, 2010, 146(1): 129-137.

[16] K. Kim, S.K. Dhungel, S. Jung, et al. Texturing of large area multi-crystalline silicon wafers through different chemical approaches for solar cell fabrication [J]. Solar Energy Materials and Solar Cells, 2008, 92(8): 960-968.

[17] K.A. Salman. Effect of surface texturing processes on the performance of crystalline silicon solar cell [J]. Solar Energy, 147(Supplement C): 228-231.

[18] Y. Qu, X. Huang, Y. Li, et al. Chemical bath deposition produced ZnO nanorod arrays as an antireflective layer in the polycrystalline Si solar cells [J]. Journal of Alloys and Compounds, 2017, 698: 719-724.

[19] A. Bai, Y. Tang, J. Chen. Efficient photon capturing in Cu(In,Ga)Se2 thin film solar cells with ZnO nanorod arrays as an antireflective coating [J]. Chemical Physics Letters, 2015, 636: 134-140.

[20] W. Yang, Y. Wang, Q. Zhen, et al. Effect of growth time on morphology and photovoltaic properties of ZnO nanowire array films [J]. Rare Metals, 2011, 30(6): 676-680.

[21] Caglar Y, Ilican S, Caglar M. FESEM, XRD and DRS studies of electrochemically deposited boron doped ZnO films [J]. Materials Science- Poland, 2017, 35(4): 824-829.

 

 
Posted by: Shiguang Shang, Associate Professor, Xi`an University of Posts and Telecommunications, China (05-Jan-2020)