**3.3 Optical properties on periodically textured 304BA stainless steel substrate**

Fig. 14 shows the OM images of the striped texture on the 304BA SS substrate. There are four patterns (i.e. the period/depth of 12/0.1, 12/0.3, 6/0.1, and 6/0.3 μm) which are designed and

Enhanced Diffuse Reflection of Light by

Using a Periodically Textured Stainless Steel Substrate 49

In our previous study (Lee et al. 2009), it was found that for a textured 430BA SS substrate the DR rate increased with the increased effectiveness of the etch-pit regions compared to that of the smooth regions. Thus, the large and deeply etched areas of the textured 304BA SS indicated that they can improve the DR rate of a textured 304BA SS substrate. In order to improve the DR rate even further, we design two other kinds of textured 304BA SS substrates, the ridged-stripe and the pyramid texture. 3D images of the ridged-stripe and pyramid texture are shown in Figs. 16(a) and (b), respectively. The etching depth and the width for both textured 304BA SS substrates were estimated to be ~6.5 μm and ~22.5 μm, respectively. The aspect ratio (i.e. depth/width) was ~1/3.5 indicating that the opening angle of the textured surface was about ~120o. It should be noted that the etching depth is controlled by the PR thickness and the etching time. In general, a thick PR and a long

The TR and DR rates of the ridged-stripe and pyramid textured 304BA SS substrates are shown in Fig. 17. We found that the DR rate at the wavelength of 600 nm increased from 3.5 % for the untreated 304BA SS substrate to 60.1% for the pyramid and 63.1% for the ridgedstripe textured 304BA SS substrate. In addition, the DR rate also increased 1.5 times at the period/depth of 6/0.3 μm for the stripe-textured 304BA SS substrate. However, the textured substrates had a lower TR rate compared to the untreated 304BA SS substrate. The lowering of the TR rate for the textured surface of the 304BA SS substrate can be explained as follows (a) the multiple scattering is the result of the multiple reflections from the ridged-stripe or pyramid textured surface of the 304BA SS substrate, and the etching pit reduction in light intensity at each reflection is due to the finite value of the reflectance for the 304BA SS substrate, (b) light trapping occurs in the indentations of a highly textured surface. Therefore, the results show that the textured 304BA SS substrate can generate a random

It is well known that the incident light is reflected back into the cell for a second pass and subsequent passes. This phenomenon results in enhanced absorption in the cell. Thus, a back reflector must possess high reflectance in the solar part of the spectrum, making Ag or Al good candidates. However, Al films absorb the incident light wavelength of 800 nm and reduce the light conversion efficiency. On the other hand, the reflection of Ag film can achieve 99% from the visible to the IR wavelength (Jenkins and white 1957). Thus, we also used an Ag coating on a textured 304BA SS substrate to study the TR and DR rates of incident light. The TR and DR rates versus the wavelength of ridged-stripe and pyramid textured 304BA SS substrates with a silver film thickness of 150 nm are shown in Fig. 18. The

Fig. 16. The 3D images of (a) ridged-stripe and (b) pyramid 304BA SS substrate.

etching time can create the deeper textured 304BA SS substrate.

(a) (b)

distribution of light through reflection from a textured surface.

used to study the TR and DR rates of the 304BA SS substrate. The stripe width and depth of the samples were measured by a surface profiler. The TR and DR rates versus the wavelength curves for untreated and the stripe-textured 304BA SS substrates are shown in Fig. 15. The "P" and "D" indicate the period and the depth for the periodically textured 304BA SS substrates, respectively. It was found that the DR rate at the wavelength of 600 nm increased substantially, from 3.5% of an untreated 304BA SS substrate to 10.5%, 21.8%, 18.2% and 39.4% for textured 304BA SS substrates with a period/depth of 12/0.1, 12/0.3, 6/0.1, and 6/0.3 μm, respectively. In addition, the TR rate of the untreated 304BA SS was 67.7% and increased to ~97% for the striped textured 304BA SS substrate due to the high reflection of the Ag film on its surface. It was evident that for the same areas of analysis, the smaller the period and the larger the depth, the better the DR rate would be, resulting in a better diffuse reflection rate.

Fig. 14. The OM images of the stripe-textured 304BA SS substrate with a period/depth of (a) 12/0.1 μm (b) 12/0.3 μm (c) 6/0.1 μm (d) 6/0.3 μm.

Fig. 15. The TR and DR rates versus the wavelength curves for untreated and stripe-textured 304BA SS substrates.

used to study the TR and DR rates of the 304BA SS substrate. The stripe width and depth of the samples were measured by a surface profiler. The TR and DR rates versus the wavelength curves for untreated and the stripe-textured 304BA SS substrates are shown in Fig. 15. The "P" and "D" indicate the period and the depth for the periodically textured 304BA SS substrates, respectively. It was found that the DR rate at the wavelength of 600 nm increased substantially, from 3.5% of an untreated 304BA SS substrate to 10.5%, 21.8%, 18.2% and 39.4% for textured 304BA SS substrates with a period/depth of 12/0.1, 12/0.3, 6/0.1, and 6/0.3 μm, respectively. In addition, the TR rate of the untreated 304BA SS was 67.7% and increased to ~97% for the striped textured 304BA SS substrate due to the high reflection of the Ag film on its surface. It was evident that for the same areas of analysis, the smaller the period and the larger the depth, the better the DR rate would be, resulting in a better diffuse reflection rate.

(d)

 Fig. 14. The OM images of the stripe-textured 304BA SS substrate with a period/depth of (a)

0

Fig. 15. The TR and DR rates versus the wavelength curves for untreated and stripe-textured

10

20

30

Diffuse reflection rate (%)

40

<sup>50</sup> (b)

400 450 500 550 600 650 700

 304BA P6H0.1 P12H0.1 P6H0.3 P12H0.3

Wavelength (nm)

12/0.1 μm (b) 12/0.3 μm (c) 6/0.1 μm (d) 6/0.3 μm.

(a) (b)

400 450 500 550 600 650 700

 304BA P6H0.1 P12H0.1 P6H0.3 P12H0.3

Wavelength (nm)

304BA SS substrates.

Total reflection rate (%)

<sup>100</sup> (a)

(c)

In our previous study (Lee et al. 2009), it was found that for a textured 430BA SS substrate the DR rate increased with the increased effectiveness of the etch-pit regions compared to that of the smooth regions. Thus, the large and deeply etched areas of the textured 304BA SS indicated that they can improve the DR rate of a textured 304BA SS substrate. In order to improve the DR rate even further, we design two other kinds of textured 304BA SS substrates, the ridged-stripe and the pyramid texture. 3D images of the ridged-stripe and pyramid texture are shown in Figs. 16(a) and (b), respectively. The etching depth and the width for both textured 304BA SS substrates were estimated to be ~6.5 μm and ~22.5 μm, respectively. The aspect ratio (i.e. depth/width) was ~1/3.5 indicating that the opening angle of the textured surface was about ~120o. It should be noted that the etching depth is controlled by the PR thickness and the etching time. In general, a thick PR and a long etching time can create the deeper textured 304BA SS substrate.

Fig. 16. The 3D images of (a) ridged-stripe and (b) pyramid 304BA SS substrate.

The TR and DR rates of the ridged-stripe and pyramid textured 304BA SS substrates are shown in Fig. 17. We found that the DR rate at the wavelength of 600 nm increased from 3.5 % for the untreated 304BA SS substrate to 60.1% for the pyramid and 63.1% for the ridgedstripe textured 304BA SS substrate. In addition, the DR rate also increased 1.5 times at the period/depth of 6/0.3 μm for the stripe-textured 304BA SS substrate. However, the textured substrates had a lower TR rate compared to the untreated 304BA SS substrate. The lowering of the TR rate for the textured surface of the 304BA SS substrate can be explained as follows (a) the multiple scattering is the result of the multiple reflections from the ridged-stripe or pyramid textured surface of the 304BA SS substrate, and the etching pit reduction in light intensity at each reflection is due to the finite value of the reflectance for the 304BA SS substrate, (b) light trapping occurs in the indentations of a highly textured surface. Therefore, the results show that the textured 304BA SS substrate can generate a random distribution of light through reflection from a textured surface.

It is well known that the incident light is reflected back into the cell for a second pass and subsequent passes. This phenomenon results in enhanced absorption in the cell. Thus, a back reflector must possess high reflectance in the solar part of the spectrum, making Ag or Al good candidates. However, Al films absorb the incident light wavelength of 800 nm and reduce the light conversion efficiency. On the other hand, the reflection of Ag film can achieve 99% from the visible to the IR wavelength (Jenkins and white 1957). Thus, we also used an Ag coating on a textured 304BA SS substrate to study the TR and DR rates of incident light. The TR and DR rates versus the wavelength of ridged-stripe and pyramid textured 304BA SS substrates with a silver film thickness of 150 nm are shown in Fig. 18. The

Enhanced Diffuse Reflection of Light by

and must be carefully investigated in future study.

0

20

40

Diffuse reflection rate (%)

textured 304BA SS substrates.

**4. Conclusions** 

60

80

100

Using a Periodically Textured Stainless Steel Substrate 51

nine ridged-surfaces within an area measuring 100×100 μm2. For the pyramid textured 304BA SS substrate, the total effective area was calculated by adding 25 pyramid-textured surfaces to the no-pyramid-coverage areas. Since the high reflection property of Ag films, the TR rate was almost higher than 90% after Ag-film coating of the textured 304BA SS substrates. It is worth noting that the DR rate increased linearly with the increase in total effective area of the stripe-textured 304BA SS substrate. However, the increase of the DR rate with the increase in the total effective area for the ridged-stripe and pyramid textured 304BA SS substrate was much more dramatic. We believe that the dramatic increase in the DR rate was due to the fact that the textured surface generated a random distribution of light by reflection from the textured surface. The aspect ratio for the ridged-stripe and pyramidal textured 304BA SS substrate were about 1/3.5 with an opening angle of 120o. In addition, the diffuse rate was defined when the incident light angle was zero, and the reflection light of that angle was larger than 80 over the incident light. Thus, the increased light diffuse due to the 120o opening angle of the texture surface caused the dramatic increase of the DR rate for the ridged-stripe and pyramid textured 304BA SS substrate. In addition, weakly absorbed light is totally reflected internally at the top surface of the cell as long as the angle of incidence inside the a-Si at the a-Si/TCO interface is greater than 160 (Banerjee and Guha 1991). It was indicated that the tilt angle of the V-shaped light trapping configuration substantially increases the photocurrent generation efficiency (Rim et al. 2007). The photocurrent increased with the increase of the tilt angle of the V-shaped configuration and is believed to enhance the number of ray bounces per unit cell area over that in a planar structure at each point in the V-fold structure. Therefore, the tilted angle of the textured surface is related to the DR and TR rate,

9900 10200 10500 10800 11100 11400 11700

Total effective area (m2

Fig. 19. The TR and DR rates as a function of the total effective area for Ag films coated on

We have demonstrated that a large diameter or a small interval of a concave shaped structure made from textured 430BA SS substrate can improve the DR rate of light.

0

)

20

40

Total reflection rate (%)

60

80

100

DR rates at the 600 nm wavelength were 95.6% and 96.8%, for the ridged-stripe and pyramid Ag film coated/texture 304BA SS substrates, respectively. The DR rate increased about 15-fold in comparison with the Ag coated untreated 304BA SS substrate. In addition, the TR rates at the 600 nm wavelength were 96.7% and 96.8%, for the ridged-stripe and pyramidal Ag film coated/texture 304BA SS substrates, respectively.

Fig. 17. The TR and DR rates versus the wavelength curves for ridged-stripe and pyramid textured 304BA SS substrates.

Fig. 18. The TR and DR rates versus the wavelength curves for Ag films coated/untreated 304BA SS substrate and Ag film coated/ridged-stripe and pyramid textured 304BA SS substrates.

Fig. 19 shows the relationship between the DR/TR rates and the total effective area of the Ag film coated/textured 304BA SS substrate. It should be noted that the total effective area was defined by the incident light reaching the textured 304BA SS substrate in an area of 100×100 μm2. For example, the total effective area of the stripe textured 304BA SS substrate was calculated by the etched side wall area added to the untreated area of 10000 μm2. For the ridged-stripe textured 304BA SS, the total effective area was calculated by summing the

DR rates at the 600 nm wavelength were 95.6% and 96.8%, for the ridged-stripe and pyramid Ag film coated/texture 304BA SS substrates, respectively. The DR rate increased about 15-fold in comparison with the Ag coated untreated 304BA SS substrate. In addition, the TR rates at the 600 nm wavelength were 96.7% and 96.8%, for the ridged-stripe and

pyramidal Ag film coated/texture 304BA SS substrates, respectively.

 Ridged-stripe Pyramid

Fig. 17. The TR and DR rates versus the wavelength curves for ridged-stripe and pyramid

45

0

20

40

Diffuse reflection rate (%)

Fig. 18. The TR and DR rates versus the wavelength curves for Ag films coated/untreated 304BA SS substrate and Ag film coated/ridged-stripe and pyramid textured 304BA SS

Fig. 19 shows the relationship between the DR/TR rates and the total effective area of the Ag film coated/textured 304BA SS substrate. It should be noted that the total effective area was defined by the incident light reaching the textured 304BA SS substrate in an area of 100×100 μm2. For example, the total effective area of the stripe textured 304BA SS substrate was calculated by the etched side wall area added to the untreated area of 10000 μm2. For the ridged-stripe textured 304BA SS, the total effective area was calculated by summing the

60

80

<sup>100</sup> (b)

50

55

Diffuse reflection rate (%)

60

65

<sup>70</sup> (b)

400 450 500 550 600 650 700

Wavelength (nm)

400 450 500 550 600 650 700

Wavelength (nm)

Ridged-stripe

Pyramid

 304BA 304BA/Ag ridged-stripe/Ag pyramid/Ag

400 450 500 550 600 650 700

Wavelength (nm)

400 450 500 550 600 650 700

 304BA 304BA/Ag ridged-stripe/Ag pyramid/Ag

Wavelength (nm)

55

50

substrates.

60

70

Total reflection rate (%)

80

90

100 (a)

textured 304BA SS substrates.

60

65

Total reflection rate (%)

70

<sup>75</sup> (a)

nine ridged-surfaces within an area measuring 100×100 μm2. For the pyramid textured 304BA SS substrate, the total effective area was calculated by adding 25 pyramid-textured surfaces to the no-pyramid-coverage areas. Since the high reflection property of Ag films, the TR rate was almost higher than 90% after Ag-film coating of the textured 304BA SS substrates. It is worth noting that the DR rate increased linearly with the increase in total effective area of the stripe-textured 304BA SS substrate. However, the increase of the DR rate with the increase in the total effective area for the ridged-stripe and pyramid textured 304BA SS substrate was much more dramatic. We believe that the dramatic increase in the DR rate was due to the fact that the textured surface generated a random distribution of light by reflection from the textured surface. The aspect ratio for the ridged-stripe and pyramidal textured 304BA SS substrate were about 1/3.5 with an opening angle of 120o. In addition, the diffuse rate was defined when the incident light angle was zero, and the reflection light of that angle was larger than 80 over the incident light. Thus, the increased light diffuse due to the 120o opening angle of the texture surface caused the dramatic increase of the DR rate for the ridged-stripe and pyramid textured 304BA SS substrate. In addition, weakly absorbed light is totally reflected internally at the top surface of the cell as long as the angle of incidence inside the a-Si at the a-Si/TCO interface is greater than 160 (Banerjee and Guha 1991). It was indicated that the tilt angle of the V-shaped light trapping configuration substantially increases the photocurrent generation efficiency (Rim et al. 2007). The photocurrent increased with the increase of the tilt angle of the V-shaped configuration and is believed to enhance the number of ray bounces per unit cell area over that in a planar structure at each point in the V-fold structure. Therefore, the tilted angle of the textured surface is related to the DR and TR rate, and must be carefully investigated in future study.

Fig. 19. The TR and DR rates as a function of the total effective area for Ag films coated on textured 304BA SS substrates.

### **4. Conclusions**

We have demonstrated that a large diameter or a small interval of a concave shaped structure made from textured 430BA SS substrate can improve the DR rate of light.

Enhanced Diffuse Reflection of Light by

350., ISSN: 0378-7788.

29, pp. 71., ISSN: 2158-3226

522. ISBN: 0070-8534-60 John A. E. St., U. S. Patent No. 3 487 223

ISSN: 0254-0584

pp. 2861., ISSN: 0009-2673

*Lett.* Vol. 91, pp. 243501. ISSN: 1077-3118

0927-0248

, ISSN: 1079-7114

1010-6030

Using a Periodically Textured Stainless Steel Substrate 53

Fung Taddy Y. Y. and H. Yang. (2008). Study on thermal performance of semi-transparent

Hartstein A., J. R. Kirtley, J. C. Tsang. (1980). Enhancement of the Infrared Absorption from

Hatta A., T. Ohshima, W. Suëtaka. (1982). Observation of the enhanced infrared absorption

Huang B. R., K. H. Chen, W. Z. Ke. (2000). Surface-enhanced Raman analysis of diamond films using different metals. *Mater. Lett*., Vol.42, pp. 162., ISSN:0167-577X He Chun, Ya Xiong, Jian Chen, Changhong, Xihai Zhu. (2003). Photoelectrochemical

Jenkins F. A. and H. E. White. (McGraw Hill, New York, 1957). *Fundamentals of Optics*., P.

Llopis F. and I. Tobías. (2005). The role of rear surface in thin silicon solar cells. *Sol. Energy* 

Lee Shuo Jen, Shiow Long Chen, Cheng Wei Peng, Chih Yuan Lin, Wen Cheng Ke. (2009).

Müller J., B. Rech, J. Springer and M. Vanecek (2004). TCO and light trapping in silicon thin

Nishikawa Y., T. Nagasawa, K. Fujiwara, M. Osawa. (1993). Silver island films for surface-

Rech B., O. Kluth, T. Repmann, T. Roschek, J. Springer, J. Müller, F. Finger, H. Stiebig, and

Rim Seung-Bum, Shanbin Zhao, Shawn R. Scully, Michael D. McGehee and Peter Peumans.

Selvan J. A. Anna., A. E. Delahoy, S. Guo and Y. M. Li. (2006). A new light trapping TCO for nc-Si:H solar cells. *Sol. Energy Mater. Sol. Cells*., Vol. 90, pp. 3371., ISSN: 0927-0248

absorption enhancement. *Vib. Spectrosc*., Vol. 6, pp. 43., ISSN: 0924-2031 Otte K., L. Makhova, A. Braun, I. Konovalov. (2006). Flexible Cu(In,Ga)Se2 thin-film solar cells for space application. *Thin Solid Films*., Vol. 511, pp. 613., ISSN: 0040-6090 Osawa M. (1997). Dynamic Processes in Electrochemical Reactions Studied by Surface-

film solar cells. *Sol. Energy*. Vol. 77, pp. 917., ISSN: 0038-092X

Enhanced diffuse reflection of light into the air using silver coating on periodically textured 430BA stainless steel substrate. *Mater. Chem. Phys.,* Vol. 118, pp. 219-222.,

enhanced infrared absorption spectroscopy: effect of island morphology on the

Enhanced Infrared Absorption Spectroscopy (SEIRAS). *Bull. Chem. Soc. Jpn*., Vol. 70,

H. Wagner. (2002). New materials and deposition techniques for highly efficient silicon thin film solar cells. *Sol. Energy Mater. Sol. Cells.,* Vol. 74, pp. 439., ISSN:

(2007). An effective light trapping configuration for thin-film solar cells. *Appl. Phys.* 

*Mater. Sol. Cells.,* Vol. 87, pp. 481., ISSN: 0927-0248

building-integrated photovoltaic glazings. *Energy and Buildings.,* Vol. 40, pp. 341-

Molecular Monolayers with Thin Metal Overlayers. *Phys. Rev. Lett*., Vol. 45, pp. 201.

of p-nitrobenzoate on Ag island films with an ATR technique. *J. Appl. Phys. A.*, Vol.

performance of Ag–TiO2/ITO film and photoelectrocatalytic activity towards the oxidation of organic pollutants. *J. Photochem. Photobiol. A.*, Vol. 57, pp. 71., ISSN:

However, the textured surface of a 430BA SS substrate led to a lower TR rate compared to a specular surface of raw 430BA SS substrate. This was due to the trapping of light in the hollows of the highly textured surface. Moreover, coating the textured 430BA SS substrate with an Ag film substantially improved not only the DR rate but also the TR rate of the incident light. The slow increase of the TR and DR rates versus the wavelength in the IR region of the Ag coated/textured 430BA SS substrates was due to the Ag absorption effect. We believe that Ag coated/textured 430BA SS substrates can generate a random distribution of light, increase the light trapping efficiency and be applied in thin films solar cells.

In addition, the DR and TR rate of the stripe, ridged-stripe and pyramid textured 304BA SS substrate were investigated to determine the optimal surface for increasing their light trapping efficiency. The DR rate increased with the increase in the total effective area of the Ag film coated/stripe textured 304BA SS substrate. It is believed that the tilt angle of the textured 304BA SS substrate increases the DR rate. The experimental results showed that the DR rate and the TR rate of the Ag film coated/ ridged-stripe textured 304BA SS substrate can achieve up to ~97% and 98% efficiency, respectively. The DR and TR rate of the Ag film coated/ridged-stripe textured 304BA SS substrates increased 28-fold and 1.4-fold, respectively, compared with the untreated 304BA SS substrate. The drastically increased DR rate is due to not only the increase in total effective area, but also to the decrease in the opening angle of the ridged textured substrate which generates a more random distribution of light by scattering.
