Influence of the properties of organic media in back-side silver pastes on the electrical performance of polycrystalline silicon solar cells

Abstract

In this paper, the best composition and ratio of solvents for the organic medium in back-side silver pastes was selected through a series of orthogonal experiments. The results showed that the volatility of the organic medium mentioned above was better than the others. The back-side silver paste prepared using this organic medium had appropriate viscosity (4.36 Pa s) and less residual ash content (0.75 wt%) when the composition of the solid substances in the organic medium was: acrylic resin 3.50 wt%, phenoxy resin 2.30 wt%, and ethyl cellulose 2.40 wt%. The properties of this paste were investigated after it was printed and further sintered to form back-side silver electrodes on solar cells. Scanning electron microscopy images showed that these back-side silver electrodes had a smoother and more dense surface, a more appropriate width and thickness, and closer contact with the silicon wafer and a larger welding tension. As a result, the photoelectric conversion efficiency of the solar cells printed by this back-side silver paste reached 18.06%.

---

Introduction

Front-side silver electrodes, back-side silver electrodes and back-side aluminum electrodes are the three main components of polycrystalline silicon solar cells. Among them, the back-side silver electrode is very important though the share of it on the back-side of one battery is small, because they can be welded directly while back-side aluminum electrodes cannot; they can also output photo-generated current collected by back-side aluminum electrodes from silicon wafers.¹ Furthermore, a back-side silver electrode is made up of back-side silver paste by screen printing and fast firing; therefore, back-side silver paste is also responsible for the properties of polycrystalline silicon solar cells.² Meanwhile, the properties of the organic medium, which is an important part of back-side silver pastes, greatly influence the electrical quality of the rear Ag metal contacts formed on the solar cells.

Viscosity is an important measurement index of organic media.³ If the viscosity of the organic medium is too low, the silver paste is diluted and flows after printing; while if the viscosity of the organic medium is too high, the silver paste is too viscous to be printed. The drying temperatures of well-printed back-side silver electrodes are usually from 150 °C to 250 °C, so the boiling point of the solvent should not be too high in the preparation of organic media. In order to store back-side silver pastes at room temperature, the organic medium is required to not volatilize at room temperature, but it can volatilize gradually if the temperature rises slowly from 70 °C to 250 °C. Further, the efficiency of solar cells reduces as the residual ash content of the organic medium on the sintered electrode increases. Meanwhile, back-side silver electrodes do not have thin grid lines, so thixotropy of the organic media is not demanded. Thus, organic media for better performance of back-side silver pastes should have appropriate viscosity, uniform volatilization, and reduced residual ash content.

In this paper, the influence of the composition and ratio of solvents and the ratio of different solid additives on the viscosity, volatilization and residual ash content of organic media were studied. Finally, a kind of organic media with a more appropriate viscosity, more uniform volatilization and less residual ash content, was obtained and used in back-side silver pastes . The thick film morphologies and the electrical properties of back-side silver electrodes printed using the above-mentioned silver pastes were tested. The experimental results showed that the back-side silver electrode made of the organic medium whose properties were better than the others had a smoother and more dense surface, a more appropriate width and thickness, closer contact with the silicon wafers and a larger welding tension. The solar cell printed using the corresponding silver paste also had a higher photoelectric conversion efficiency.

Experimental

Measuring the viscosity of the organic solvents

Five frequently-used organic solvents were selected to carry out a contrast experiment to determine the influence of organic solvents (butyl butyrate, diethylene glycol butylether acetate, tributyl citrate, terpineol and diethyl ethanedioate) on the properties of the organic medium. The viscosity of each organic solvent was tested, with the contents of other ingredients kept the same through the experiment, as shown in Table 1.

Table 1 The dependence of the viscosity on the differently proportioned solvents

No.	Butyl butyrate (wt%)	Diethylene glycol butylether acetate (wt%)	Tributyl citrate (wt%)	Terpineol (wt%)	Diethyl ethanedioate (wt%)	Other ingredients (wt%)	Viscosity (Pa·s)
1	80	0	0	0	0	20	0.92
2	0	80	0	0	0	20	2.45
3	0	0	80	0	0	20	22.07
4	0	0	0	80	0	20	23.24
5	0	0	0	0	80	20	1.89
6	40	40	0	0	0	20	1.48
7	40	0	40	0	0	20	3.71
8	40	0	0	40	0	20	4.01
9	40	0	0	0	40	20	1.08
10	0	40	40	0	0	20	6.45
11	0	40	0	40	0	20	6.58
12	0	40	0	0	40	20	2.05
13	0	0	40	40	0	20	13.92
14	0	0	40	0	40	20	1.84
15	0	0	0	40	40	20	2.51
16	26.67	26.67	26.67	0	0	20	2.76
17	26.67	26.67	0	26.67	0	20	1.79
18	26.67	26.67	0	0	26.67	20	1.13
19	26.67	0	26.67	26.67	0	20	2.92
20	26.67	0	26.67	0	26.67	20	1.59
21	26.67	0	0	26.67	26.67	20	1.23
22	0	26.67	26.67	26.67	0	20	5.99
23	0	26.67	26.67	0	26.67	20	2.82
24	0	26.67	0	26.67	26.67	20	2.36
25	0	0	26.67	26.67	26.67	20	4.45
26	20	20	20	20	0	20	3.07
27	20	20	20	0	20	20	2.10
28	20	20	0	20	20	20	1.93
29	20	0	20	20	20	20	3.41
30	0	20	20	20	20	20	3.92
31	16	16	16	16	16	20	3.07

---

Isothermal volatilization tests of organic solvents with different proportions

According to the proportions in Table 1, the isothermal volatilization experiments of the differently proportioned organic solvents were implemented in a program control temperature drying oven. The volatilization temperature was increased gradually from 70 °C to 210 °C, and results are shown in Fig. 1 and 2.

Fig. 1 Evaporation curves of organic media with pure solvents (no. 1–5).

Fig. 2 Evaporation curves of organic media with mixed solvents (no. 6–31).

Orthogonal experiments using organic solvents with different proportions

The best composition and appropriate dosage of solvents were selected on the basis of the isothermal volatilization tests as butyl butyrate 3–6 wt%, diethyl ethanedioate 20–23 wt%, terpineol 28–34 wt% and tributyl citrate 23–29 wt%. Then the best organic solvent composition was adjusted by an orthogonal experiment (the dosage of other ingredients was kept the same), as shown in Table 2.

Table 2 Four factors and four levels of organic solvents

Level	Butyl butyrate (wt%)	Diethyl ethanedioate (wt%)	Terpineol (wt%)	Tributyl citrate (wt%)
1	3.00	20.00	28.00	23.00
2	4.00	21.00	30.00	25.00
3	5.00	22.00	32.00	27.00
4	6.00	23.00	34.00	29.00

---

Preparation of organic media

Butyl butyrate, diethyl ethanedioate, terpineol and tributyl citrate were evenly mixed in a three-necked flask according to the selected ratio. Then a certain amount of thickener (ethyl cellulose), plasticizer (dimethyl oxalate) and bonding agent (hydrogenated rosin pentaerythritol ester) were added into the flask sequentially. The mixture^4,5 was mechanically stirred at 90 °C for 1 h and then cooled naturally to room temperature after the reaction. Finally, four different kinds of organic media were obtained by adjusting the dosage of the solid additives.

Preparation of back-side silver pastes

First, the four different kinds of organic media were separately mixed with spherical silver particles (D₉₀ = 1.50 μm, tap density = 4.1 g cm⁻³) and glass powder with the composition Bi₂O₃–TeO₂–PbO–B₂O₃–Al₂O₃ (the softening point of the glass powder was about 650–680 °C, D₉₀ = 4.38 μm). In these mixed samples, the content of silver particles was 54.72 wt%, the content of glass powder was 3.28 wt% and the content of the organic medium was 42 wt%. These mixtures were then rolled.^6–9 Finally, four different kinds of back-side silver pastes were obtained.

Measurements

The viscosities of the solvents and organic media were tested using a Viscometer (DV-II+PR, American Brookfield co., LTD.). The volatilization of solvents and organic media were measured using a program control temperature drying oven (TD-1, Suzhou Ta De Oven Equipment Manufacturing co., LTD.). The thermogravimetric curves of the organic media were characterized using a Thermogravimetric Analyzer (Netzsch STA 449C). The average particle sizes (D₉₀) of the silver particles and glass powder were tested by laser particle size distribution (BT-9300-H, Dandong city Baxter instrument co., Ltd.). The tap density of the silver particles was tested using a particle density tester (Hylology, HY-100). The softening point of the glass powders was measured using a PCY-SP-1200 fully automatic glass soft point tester (Xiangtan Huafeng instrument manufacturing Co., Ltd.). Silver pastes were made using a three roller grinding machine (Puhler, PTR65C). The surface morphologies and the cross-sectional microstructures of the back-side silver electrodes were observed by scanning electron microscopy (SEM) (Quanta 400 FEG instrument, Oxford INCA 35 detector, 25 kV). The welding tension between the back-side silver electrodes and silicon wafers was characterized using a FDV-50 force apparatus (Wagner Instruments, USA). The electrical performance of solar cells was measured by battery testing equipment (Baccini, ITA).

Results and discussion

The influence of solvents and solid additives on the properties of the organic media

The viscosity values of the organic solvents in different proportions were measured by a viscometer, as shown in Table 1. It was found that the viscosity increases with the presence of terpineol or tributyl citrate, while the viscosity decreases with the presence of butyl butyrate or diethyl ethanedioate.

The evaporation curves of solvents in each group were obtained according to the data in the isothermal volatilization test. First, evaporation curves of five kinds of pure solvents (no. 1–5 in Table 1) are shown in Fig. 1. The ready-prepared organic media should ideally not volatilize at room temperature, but they should volatilize gradually while the temperature rises (from 70 °C to 210 °C). Therefore, we commonly selected the volatility at 150 °C to determine whether the volatility of a solvent was better or not. The volatility at 150 °C (from high to low) of the five solvents was: butyl butyrate, diethyl ethanedioate, terpineol, diethylene glycol butylether acetate and tributyl citrate. These experimental results are in accordance with the fact that the higher the boiling point of the solvent, the more difficult it is to volatilize. According to the experience we have acquired, the volatility of an organic medium made of a single solvent cannot be adjusted in the drying process. The organic medium will volatilize during the storing and printing process, which thickens the back-side of the silver paste. On one hand, if the volatility of the organic medium at a certain temperature is higher, there will be more air bubbles in the back-side silver paste, which will lead to a rougher sintered membrane and reduce the stability of the paste. On the other hand, if the volatility is lower, it takes too long for the back-side silver paste to dry. Therefore, the sintered membrane layer will be a little wet after drying and the edge of it will not be smooth, which may possibly result in more holes and poor performance of the back-side silver electrode. So mixed solvents are generally adopted to adjust the volatilization of organic media.

Volatilization of organic media composed of mixed solvents is in accordance with Henry’s law; that is, the relative content of each solvent in one mixed solvent decides the relative ratio of this mixed solvent’s vapour pressure. Therefore, the volatilization of each organic medium can be adjusted by changing the content of each solvent in the mixed solvent. By doing so, organic media will not volatilize at the room temperature under normal circumstances; however, they will volatilize gradually when the temperature increases slowly from 70 °C to 210 °C. Fig. 2 displays evaporation curves of the organic media with mixed solvents (no. 6–31 in Table 1).

As seen from Fig. 2, the volatility of the organic medium with mixed solvent no. 29 is less at low temperature and reaches 55.88 wt% at 150 °C. At the same time, the volatility also increases gradually as the temperature rises, which meets the demands for organic media in the process of production. But the evaporation curves of the organic media with other mixed solvents are not so gentle or suitable as no. 29.

On the basis of the composition of the mixed solvent no. 29, the best content range of each solvent was selected through multiple tests respectively: butyl butyrate 3–6 wt%, diethyl ethanedioate 20–23 wt%, terpineol 28–34 wt%, tributyl citrate 23–29 wt%. According to the ratio of every solvent in Table 2, an orthogonal experiment was implemented as detailed in Table 3.

Table 3 Orthogonal experiment

No.	A	B	C	D	volatility at 150 °C (wt%)
	Butyl butyrate	Diethyl ethanedioate	Terpineol	Tributyl citrate
29-1	1	2	3	3	41.15
29-2	2	4	1	2	38.09
29-3	3	4	3	4	36.67
29-4	4	2	1	1	40.14
29-5	1	3	1	4	39.75
29-6	2	1	3	1	41.33
29-7	3	1	1	3	51.02
29-8	4	3	3	2	51.42
29-9	1	1	4	2	41.13
29-10	2	3	2	3	39.30
29-11	3	3	4	1	40.01
29-12	4	1	2	4	37.28
29-13	1	4	2	1	40.46
29-14	2	2	4	4	38.94
29-15	3	2	2	2	41.11
29-16	4	4	4	3	39.31
K1	162.49	170.76	169.00	161.94
K2	157.66	170.34	158.15	171.75
K3	168.81	161.34	170.57	170.78
K4	168.15	154.53	159.84	152.64
k1	40.62	42.69	42.25	40.49
k2	39.42	42.59	39.54	42.94
k3	42.20	40.34	42.64	42.70
k4	42.04	38.63	39.96	38.16
Range	2.78	4.06	3.10	4.78
Sequence	D > B > C > A
Optimal level	A3	B1	C3	D2
Optimal combination	A3B1C3D2

---

From Table 3, it can be seen that the effect order of solvents on the evaporation quantity of organic media at 150 °C is: tributyl citrate > diethyl ethanedioate > terpineol > butyl butyrate. The optimal combination (A₃B₁C₃D₂) of mixed solvents in the organic medium is as follows: butyl butyrate 5.00 wt%, diethyl ethanedioate 20.00 wt%, terpineol 32.00 wt%, and tributyl citrate 25.00 wt%. Meanwhile, the organic medium with the best volatilization performance has the optimal combination of mixed solvents mentioned above and is named as no. 29–17.

The evaporation quantities of the organic media (no. 29-1–29-17) were measured at different temperatures, and the evaporation curves are shown in Fig. 3. As seen from Fig. 3, the volatility of the organic medium no. 29–17 is less at low temperature and reaches 53.02 wt% at 150 °C. In the meantime, its volatility also increases gradually as the temperature rises, going beyond 80.00 wt% when the temperature reaches 210 °C. The evaporation curve of the organic medium no. 29-17 is more gentle and flat than the others and the related performance of it is also better, and so it meets the requirements for organic media in the process of production.

Fig. 3 The evaporation curves of the organic media (no. 29-1–29-17).

The type and content of solvent were kept the same with the organic medium no. 29-17, then different kinds of organic media were prepared by adjusting the content of solid additives, such as acrylic resin, phenoxy resin and ethyl cellulose. The viscosity and the residual ash content of these organic media were tested as shown in Table 4. From Table 4 we can see that the viscosity of the organic medium no. 29-17-6 is 4.36 Pa s and the residual ash content of no. 29-17-6 is 0.75 wt%. These two performance parameters above are appropriate for the application of the organic medium in back-side silver pastes. Therefore, the organic medium no. 29-17-6 is better than the others by comparing the viscosities and residual ash contents. If the viscosity of the organic medium is smaller, the silver paste will be too dilute; on the contrary, the silver paste will be too viscous when the viscosity of the organic medium is bigger. At the same time, the residual ash content of the organic medium is the lower the better, because the electrical performance of the solar cell will be reduced when the residual ash content is increased.

Table 4 The effect of different solid additive contents on the performances of organic media

No.	Viscosity (Pa s)		Residual ash content (wt%)	Acrylic resin (wt%)	Phenoxy resin (wt%)	Ethyl cellulose (wt%)
29-17-1		3.08	0.79	2.50	1.30	2.00
29-17-2		3.12	0.93	3.00	1.80	2.00
29-17-3		3.26	0.95	3.50	2.30	2.00
29-17-4		3.76	0.99	2.50	1.30	2.40
29-17-5		4.18	1.01	3.00	1.80	2.40
29-17-6		4.36	0.75	3.50	2.30	2.40
29-17-7		5.76	1.26	2.50	1.30	2.80
29-17-8		6.64	1.38	3.00	1.80	2.80
29-17-9		7.32	1.59	3.50	2.30	2.80

---

The influence of different organic media on the performances of back-side silver pastes

Comparison experiments of organic media were done by selecting the proportion of four solvents according to the finished experiments (as shown in Table 5, and the solid additives were kept the same as no. 29-17-6), in order to show the performances of the four different organic media.

Table 5 The proportion of the four solvents in organic media

No.	Viscosity (Pa s)	Butyl butyrate (wt%)	Diethyl ethanedioate (wt%)	Terpineol (wt%)	Tributyl citrate (wt%)
a	5.29	3.0	16.0	35.0	28.0
b	4.36	5.0	20.0	32.0	25.0
c	3.61	7.0	24.0	29.0	22.0
d	3.02	9.0	28.0	26.0	19.0

---

As seen from Table 5, the viscosity of the organic medium reduces as the content of butyl butyrate and diethyl ethanedioate increases, while the viscosity of the organic medium increases with the content of terpineol and tributyl citrate. In short, the organic medium (b) has an optimal viscosity of 4.36 Pa s, which is the same as the previous conclusion.

Fig. 4 shows the TG curves of organic media (a–d). We can observe from Fig. 4(b) that the TG curve shows a horizontal line below 60 °C which indicates that the organic medium would not lose any weight at room temperature. The weight percentage of the organic medium (b) decreases slowly from 60 °C to 250 °C, and nearly 90 wt% of the organic medium has already volatilized before 250 °C, which satisfies the demand of the volatilization performance of organic media for the drying process. From 250 °C to 450 °C, the 9 wt% of weight loss is caused by the combustion of ethyl cellulose and other agents. At temperatures above 450 °C, the TG curve shows a horizontal line and the incinerated residue content is less than 1 wt%, resulting from the incomplete combustion of solid agents. Therefore, the organic medium (b) would avoid cracks in the surface of the back-side electrode grid line.

Fig. 4 TG curves of organic media (a–d).

The TG curve of Fig. 4(a) shows a horizontal line at the higher temperature of 78 °C, which means that the organic medium (a) is more difficult to volatilize at low temperatures. It has less volatilization before 200 °C but volatilizes altogether after 200 °C because of higher amounts of terpineol and tributyl citrate. The TG curves of Fig. 4(c) and (d) show horizontal lines with lower temperature and the TG curves are both steep before 200 °C because there are higher amounts of butyl butyrate and diethyl ethanedioate in these two organic media.

The influence of different organic media on the performances of the sintered membranes of back-side silver electrodes

The above-mentioned organic media (a–d) were used to prepare back-side silver pastes and then the pastes were printed on the same polycrystalline silicon wafers (156 mm × 156 mm) by screen printing for ten wafers, separately. After being dried and sintered at a peak temperature of 900 °C, the solar cells were named as SCa, SCb, SCc and SCd.

The sintering surfaces of the back-side silver electrodes of solar cells SCa, SCb, SCc and SCd were observed by SEM^10–12 as shown in Fig. 5. It can be seen that the surface of the back-side silver electrode of SCb is more dense, more smooth and has fewer holes; while the surfaces of the back-side silver electrodes of SCa and SCc have many holes and are not dense enough; moreover, the surface of the back-side silver electrode of SCd is more rough and has more holes than those of both SCa and SCc.

Fig. 5 Sintering surface SEM images of the back-side silver electrodes of SCa, SCb, SCc and SCd (at 900 °C).

The reason behind the above experimental results is that the volatilization amount of each boiling point solvent in organic media is different at the same temperature during the sintering process. The content of low boiling point solvents in the organic medium of SCa is very little, so there are fewer holes on the surface of the electrodes at low temperatures. The incidence of holes increases with the temperature rising, because the content of solvents with high boiling point in the organic medium of SCa is higher and the solvents volatilize together at the higher temperature; so there are many holes on the surface of the back-side silver electrode of SCa, as shown in Fig. 5(a). While the content of low boiling point solvents in the organic media of SCc and SCd is higher, the holes on the surface of these back-side silver electrodes increase more and more due to the low boiling point solvents volatilizing together at low temperature at the same time, as shown in Fig. 5(c) and (d). In general, the holes on the surface of the back-side silver electrodes become fewer and the electrode surface gets more dense with the decreasing content of low boiling point solvents and the increasing content of high boiling point solvents in the paste. That is to say, when the proportion of low boiling point solvents and high boiling point solvents in the organic medium is appropriate, the surface of the back-side silver electrode of the solar cell will have fewer holes and be more dense, as shown in Fig. 5(b). This phenomenon is consistent with the TG curves in Fig. 4.

Fig. 6 shows line width and cross-sectional microstructures of back-side silver electrodes of solar cells SCa, SCb, SCc and SCd. The back-side silver electrode of the solar cell SCa has a narrower grid line and the cross-sectional height of the back-side silver electrode also seems much taller, as shown in Fig. 6(a). The back-side silver electrode grid line of the solar cell SCd is wider and thinner than the others, as shown in Fig. 6(d).

Fig. 6 Line width and cross-sectional microstructures of back-side silver electrodes of SCa, SCb, SCc and SCd (at 900 °C).

This is because the viscosity of the organic medium (a) is bigger, so the paste made by it will be too viscous to be printed through the screen printing plate. As the viscosity of the organic medium decreases, the back-side silver paste becomes dilute and has good liquidity, which makes the paste more easily to be printed and the solid powders can be fully dispersed in the paste. But if the viscosity is too small, the silver paste will be so dilute that it will flow after being printed, and the electrode made by it will become wider and thinner, which is harmful to the electrical performance of the solar cells. By comprehensive comparison, we found that the electrode width and thickness of the solar cell SCb are most appropriate, as shown in Fig. 6(b).

Cross-sectional microstructures of the contact interface between back-side silver electrodes and silicon wafers SCa, SCb, SCc and SCd are shown in Fig. 7. As seen from Fig. 7(b), the back-side silver electrode of the solar cell SCb was closely in contact with the silicon wafer and there are fewer holes between the electrode and the silicon wafer, which will export the electronic directly and quickly. However, there are more holes between the electrodes and silicon wafers of solar cells SCa, SCc and SCd, as shown in Fig. 7(a), (c) and (d), respectively. Therefore, the back-side silver electrode of the solar cell SCb has the most excellent performance, which is in accordance with the conclusions above in Fig. 5 and 6.

Fig. 7 Cross-sectional microstructures of the contact interface between back-side silver electrodes and silicon wafers SCa, SCb, SCc and SCd (at 900 °C).

The influence of different organic media on the welding tension of back-side silver electrodes

Welding performance is one of the important factors affecting the service life of solar cells. The welding tension of back-side silver electrodes not only depends on the binding force between the solder strip and the electrode, but also depends on the adhesion strength between the electrode and the silicon wafer. The weaker the adhesion strength between the electrode and silicon wafer is, the more easily the back-side silver electrode falls off after solder strip peeling, and so does the binding force. The welding tension of solar cells SCa, SCb, SCc and SCd was tested with FDV-50 force apparatus, and the results are shown in Fig. 8.

Fig. 8 The welding tension of back-side silver electrodes of SCa, SCb, SCc and SCd.

As seen from Fig. 8, the back-side silver electrode of SCb has the largest welding tension (7.7 N). This is because silver particles and glass powders can be mixed adequately by the organic medium (b), which makes the back-side silver pastes have better printing performance. Another reason is that the viscosity and the volatilization of organic medium (b) are both better, which make the electrodes maintain good performance in the process of high temperature sintering. The above experimental results show that the back-side silver electrodes of SCb contact much more closely with the silicon wafer, and they also have the better welding performance.

The influence of different organic media on the electrical performances of back-side silver electrodes

The electrical performances of solar cells SCa, SCb, SCc and SCd were tested as shown in Table 6. As seen from Table 6, the series resistance of SCb is smaller than that of SCa, SCc and SCd, and the fill factor of SCb is bigger than that of SCa, SCc and SCd. Therefore, the back-side silver paste which was printed on the electrodes of the solar cell SCb has the optimal performances. That is to say, the back-side silver paste corresponding to organic medium (b) provides the solar cell SCb with the higher conversion efficiency of 18.06%.

Table 6 Electrical performance parameters of solar cells SCa, SCb, SCc and SCd

No.	Electrical performance parameters
	Voc (V)	Isc (A)	FF (%)	Rs (Ω)	Eff
SCa	0.623737	8.614004	76.92045	0.003489	0.169824
SCb	0.635712	8.726632	79.22516	0.002308	0.180603
SCc	0.618846	8.816563	77.98797	0.003001	0.174848
SCd	0.618380	8.829535	74.95213	0.003716	0.168162

---

Conclusions

In this paper, the influence of the composition and ratio of different solvents and the ratio of different solid additives on the viscosity, volatilization and residual ash content of organic media were studied, and an organic medium with more appropriate viscosity, more uniform volatilization and less residual ash content used in back-side silver pastes was obtained. Back-side silver pastes were mixed by the same silver particles, the same glass powders and different organic media, and then they were printed on the surface of the solar cells. The sintered membrane SEM images and the electrical properties of back-side silver electrodes made by the above-mentioned silver pastes were both tested. Experimental results showed that the properties of the back-side silver electrode made of organic medium (b) is the optimal. It not only has a more smooth and more dense surface, a more appropriate width and thickness, a more closely contact with silicon wafers and larger welding tension, but also provides the solar cell SCb with higher photoelectric conversion efficiency.

Acknowledgements

The authors are grateful to the financial supports of the National Hi-Tech Research and Development Program (863) Key Project of China (No. 2012AA050301-SQ2011GX01D01292), and Xi'an Industrial Technology Innovation Project-technology transfer promoting program (No. CXY1421 and CXY1511(9)).

Notes and references

1. I. Zanesco, V. A. Gonçalves and A. Moehlecke, Energy Procedia, 2014, 57, 47–55.
2. A. S. Ionkin, B. M. Fish, Z. R. Li, M. Lewittes, P. D. Soper, J. G. Pepin and A. F. Carroll, ACS Appl. Mater. Interfaces, 2011, 3, 606–611.
3. J. Zhang, Y. Cui and H. Wang, J. Renewable Sustainable Energy, 2013, 5, 023117.
4. G. J. Zheng, Y. P. Tai, H. Wang and J. T. Bai, J. Mater. Sci.: Mater. Electron., 2014, 25, 3779–3786.
5. G. Wang, H. Wang, Y. B. Cui and J. T. Bai, J. Mater. Sci.: Mater. Electron., 2014, 25, 487–494.
6. A. Kalio, M. Leibinger, A. Filipovic, K. Krüger, M. Glatthaar and J. Wilde, Sol. Energy Mater. Sol. Cells, 2012, 106, 51–54.
7. Q. Che, H. Yang, L. Lu and Y. Wang, J. Alloys Compd., 2013, 549, 221–225.
8. N. Moudir, Y. Boukennous, N. Moulaï-Mostefa, I. Bozetine, M. Maoudj, N. Kamel, Z. Kamel and D. Moudir, Energy Procedia, 2013, 36, 1184–1191.
9. Z. Liu, X. L. Qi and H. Wang, Adv. Powder Technol., 2012, 23, 250–255.
10. Y. P. Tai, G. J. Zheng, H. Y. Wang, H. Wang and J. T. Bai, RSC Adv., 2015, 5, 92515–92521.
11. J. H. Kim, H. Y. Koo, Y. N. Ko and Y. C. Kang, J. Alloys Compd., 2010, 497, 259–266.
12. J. Zhou, N. Xu, H. Yang and Q. Zhang, Procedia Eng., 2014, 94, 1–5.

---