Li 2017

Subscriber access provided by University of Newcastle, Australia
Letter
A comprehensive photonic approach for solar cell cooling
Wei Li, Yu Shi, Kaifeng Chen, Linxiao Zhu, and Shanhui Fan
ACS Photonics, Just Accepted Manuscript • DOI: 10.1021/acsphotonics.7b00089 • Publication Date (Web): 14 Mar 2017
Downloaded from http://pubs.acs.org on March 14, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Photonics is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.
Page 1 of 26 ACS Photonics
1
2
3
4 A comprehensive photonic approach for solar cell cooling
5
6 Wei Li1, Yu Shi1, Kaifeng Chen1,2, Linxiao Zhu2 and Shanhui Fan1*
7
8 1
9 Department of Electrical Engineering, Ginzton Laboratory, Stanford University, Stanford, California
10 94305, USA
2
11 Department of Applied Physics, Stanford University, Stanford, California 94305, USA
*
12 email: shanhui@stanford.edu
13
14
15 Abstract: The heating of solar cell has significant adverse consequences on both its efficiency
16
17
18 and its reliability. Here to fully exploit the cooling potential of solar cells, we experimentally
19
20 characterized the thermal radiation and solar absorption properties of current silicon solar cells,
21
22
and based on such experimental characterization, propose a comprehensive photonic approach by
23
24
25 simultaneously perform radiative cooling while also selectively utilizing the sunlight. In
26
27 particular, we design a photonic cooler made of one-dimensional photonic films that can strongly
28
29
30
radiate heat through its thermal emission while also significantly reflect the solar spectrum in
31
32 sub-band gap and ultraviolet regime. We show that applying this photonic cooler on solar panel
33
34 can lower the cell temperature by over 5.7 degree Celsius. We also show that this photonic
35
36
37 cooler can be used in concentrated photovoltaic system to significantly reduce the solar cell
38
39 temperature or required cooling power. This photonic cooler can be readily implemented in
40
41 current photovoltaic modules as a retrofit to improve both efficiency and lifetime. Our approach
42
43
44 points to an optimal photonic approach for thermal management of solar cells.
45
46
47
48
49
Key words: photonics; solar cells; radiative cooling; thermal radiation; radiative heat transfer.
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
ACS Photonics Page 2 of 26
1
2
3
A single-junction solar cell is typically designed to efficiently absorb most incident solar
4
5
6 irradiance with photon energy above its semiconductor band gap. However, under the AM 1.5
7
8 solar spectrum, there is an intrinsic theoretical power conversion efficiency limit of 33.7% based
9
10
11
on Shockley and Queisser’s analysis1 and nowadays most commonly used silicon solar cells
12
13 possess efficiency around 15% - 22%. Therefore, a significant portion of solar energy is
14
15 converted into heat and in turn, heats up the solar cell. In practice, in outdoor conditions, the
16
17
18 typical operating temperature of a solar cell is around 50 °C - 55°C or higher2–4. This heating of
19
20 the solar cell has adverse consequences on both its performance and reliability5. The conversion
21
22 efficiency of the solar cell deteriorates at elevated temperatures. For crystalline silicon solar
23
24
25 cells, every 1°C temperature rise leads to a relative efficiency decline of about 0.45%6. In
26
27 addition, the aging rate of a solar cell array doubles for every 10 °C solar cell temperature
28
29
increase7. Therefore, it is extremely important to develop solar cell cooling techniques to keep
30
31
32 the cell operating temperature as low as possible.
33
34 Conventional solar cell cooling strategies, including conduction of heat to dissipation
35
36
37 surfaces8, forced air flow9, water cooling10 and heat-pipe-based systems11,12, have been mainly
38
39 focused on engineering the non-radiative heat transfer accesses of solar cells using heat
40
41 conduction or convection approaches. Some of these conductive or convective cooling
42
43
44 techniques either need extra energy input or increase the system complexity. On the other hand,
45
46 radiative heat exchange plays a significant role in the thermal balance of the solar cell. A solar
47
48 cell is heated up by the sun, and moreover it naturally faces the sky, and therefore can radiate
49
50
51 some of its heat out as infrared radiation. Therefore, complementary to non-radiative approaches,
52
53 it is also of importance to explore the possibilities of using the photonic approach to engineer the
54
55
56
radiative heat transfer access of solar cells for thermal management.
57
58
59
60
2
1
2
3
There currently exist two major photonic approaches in thermal management of solar
4
5
6 cells, focusing on controlling either the solar absorption13–15 or the thermal radiation16,17. A solar
7
8 cell cannot convert sub-band gap photons to electricity. However, a practical solar cell structure
9
10
11
in fact typically has substantial absorption in the sub-band gap wavelength range. Such sub-band
12
13 gap absorption does not contribute to the generation of photo-current and represents a parasitic
14
15 heat source. To address this issue, infrared filter design has been proposed to selectively reflect
16
17
18 solar spectrum in the sub-band gap regime13–15 to minimize the parasitic heat source. More
19
20 recently, there is also an emerging interest to design the thermal radiation property of the solar
21
22 cell for cooling purposes16–20. This radiative cooling approach is based on the fact that any sky-
23
24
25 facing structure naturally has radiative access to the cold universe through the Earth’s
26
27 atmosphere’s transparency window between 8 and 13 µm16–32. And moreover, since solar cells
28
29
under sunlight are typically operating above the ambient air temperature, thermal emissions
30
31
32 outside the atmosphere transparency window in the entire wavelength range between 4-30
33
34 microns also help in radiative cooling. Thus, Refs. [16] and [17] designed and demonstrated a
35
36
37 photonic structure that is transparent in the solar wavelength range, yet behaves as a blackbody
38
39 in the thermal wavelength range. Such a photonic structure, when placed on a solar cell, can
40
41 radiatively cool a solar cell underneath it without affecting the solar absorption.
42
43
44 As evidenced from the description above, there are significant opportunities in
45
46 developing photonic approaches for solar cell cooling. However, the existing approaches, which
47
48 focused solely on either engineering solar absorption or thermal radiation, can only exploit part
49
50
51 of the opportunities that photonic engineering can provide for solar cell cooling purposes. In this
52
53 paper, we propose a comprehensive photonic strategy for passive cooling of solar cells that can
54
55
56
be implemented as a retrofit to the current photovoltaic modules (Figure 1). The idea is to place a
57
58
59
60
3
1
2
3
photonic coating layer on top of the existing encapsulated solar panels. The coating is designed
4
5
6 to simultaneously reflect part of the solar spectrum that does not contribute to photo-current and
7
8 to maximize thermal radiation. In addition, such a coating serves as broadband anti-reflection
9
10
11
coating, preserving or even enhancing the optical absorption characteristics of the solar cells in
12
13 the part of the solar spectrum that contributes to photo-current. To illustrate the efficacy of this
14
15 approach, we first provide experimental characterizations of the absorption properties of solar
16
17
18 panels in both the solar and thermal wavelength ranges. Based on the experimental data, we then
19
20 numerically show that a photonic coating structure made of planar multilayer dielectric stack
21
22 placed on top of the such solar panels can provide a temperature reduction for the solar cell of
23
24
25 over 5.7K as compared to the same panel without the coating, taking into account the non-
26
27 radiative heat dissipation processes in a typical terrestrial operating condition.
28
29
30
31
32
33
34
35
36
37
38
39
40
41 Figure 1 a, Schematic of the solar absorption and thermal radiation properties of existing solar
42
43
panels. Suboptimal radiative cooling and strong parasitic sub-band gap absorption are observed
44 in existing solar panels. b, With a photonic cooler coated on top of solar panel, the radiative
45 cooling can be enhanced and the sub-band gap solar radiation can be strongly reflected.
46
47
48
49
50 Experimental Characterization of the Absorptivity Spectra of Solar Cells in the Solar and
51
52 Thermal Wavelength Range
53
54 To illustrate our photonic approach for solar cell cooling, we first experimentally
55
56
57 examine solar absorption (Fig. 2a) and thermal emission (Fig. 2b) properties of typical
58
59
60
4
1
2
3
commercial solar cells. As examples, we consider crystalline silicon solar cells, which represent
4
5
6 the solar cells with the highest market penetration33. We obtained two types of state-of-the-art
7
8 bare silicon solar cells: cell I is a silicon solar cell with interdigitated back contacts34 and cell II
9
10
11
is a silicon solar cell with screen-printed silver front and aluminum rear contacts35 (Fig. 2c).
12
13 Since in practice in a photovoltaic module a solar cell is always encapsulated, we also
14
15 encapsulated cell I by sandwiching it with two 0.46 mm thick ethylene-vinyl acetate (EVA) joint
16
17
18 interlayers, with a 3.2 mm thick glass as front cover and a 0.5 mm thick back sheet layer made of
19
20 Tedlar (polyvinyl fluoride), as shown in Fig. 2d. The materials and parameters for the
21
22 encapsulation layers were chosen according to commercial solar panel standards.
23
24
25 The solar absorption and thermal emissivity spectra of the bare cells and the encapsulated
26
27 cell are experimentally characterized and shown in Figs. 2a and b, respectively. In the solar
28
29
wavelength region, both of the bare cells show strong light absorption from 0.3 to 1.1 µm, where
30
31
32 the photon energy is above the silicon band gap, as expected. In the wavelength range between
33
34 1.1 to 1.8 µm, both of the bare cells show significant absorption, even though photons at these
35
36
37 wavelength ranges have energy below the silicon band gap. Such a strong sub-band gap
38
39 absorption is in contrast with measured data on a bare silicon wafer with planar anti-reflection
40
41 coating and metallic back reflector17, most likely due to the presence of metal contacts and
42
43
44 heavily doped regions in the cells. The sub-band gap absorption may be further enhanced by the
45
46 light trapping effects induced by surface textures36. From the measured absorption spectra, under
47
48 AM1.5 illumination, the sub-band gap absorption of cell I and II are calculated to be 85 W/m2
49
50
51 and 150 W/m2, respectively. Such sub-band gap absorption represents a parasitic heat source that
52
53 does not contribute to current generation.
54
55
56
57
58
59
60
5
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50 Figure 2 a, Experimentally measured solar absorption spectra of two representative solar cells
51 and an encapsulated cell, with the normalized AM1.5 solar spectrum plotted for reference. b,
52
Experimentally measured emissivity spectra of two representative solar cells and an encapsulated
53
54 cell over the infrared wavelength, with a realistic atmosphere transmittance model from Ref. [37]
55 for winter in California plotted as the shaded area. c, Photo of the front side of the two measured
56 solar cells. I: a silicon solar cell with interdigitated back contacts. II: a screen-printed silicon
57 solar cell with silver front and aluminum rear contacts. d, Photo (left) and a cross-section
58
59
60
6
1
2
3
schematic (right) of a solar cell with encapsulation, made of cell I, with a 3.2 mm front glass
4
5 cover, top and bottom 0.46 mm EVA film and a 0.5 mm back sheet.
6
7
8 Compared to the bare cell, the encapsulated cell shows an even stronger absorption in the
9
10
11
sub-band gap as well as UV wavelength range. Our fabricated encapsulated cell has a sub-band
12
13 gap absorption of 110W/m2 for the AM 1.5 illumination, which is significant higher than that of
14
15 the bare cell I (Fig. 2a). This is due to the fact that EVA strongly absorbs UV light with
16
17
18 wavelength shorter than 0.375 µm and slightly absorbs in the near infrared regime38. Such
19
20 absorption can lead to excess heat generation while also degrade the EVA, reducing the lifetime
21
22 of the solar panels. Therefore, for cooling purposes, and also to enhance the lifetime of
23
24
25 encapsulated cell, it is beneficial to suppress the sub-band gap and UV absorption of the solar
26
27 panel while preserving or even enhancing the absorption of the solar cell in the wavelength range
28
29
of 0.375-1.1 µm.
30
31
32 In the infrared wavelength range between 2 – 25 µm, the emissivity spectra of the bare
33
34 solar cells and encapsulated cell are shown in Fig. 2b. Both the bare and the encapsulated cells
35
36
37 show strong absorption and hence strong thermal emission in the thermal wavelength range.
38
39 Such an absorption spectrum again differs significantly as compared with planar silicon wafer
40
41 measured in Ref. [17]. For the bare cells, although silicon is a weak absorbing material in this
42
43
44 regime, the presence of highly doped silicon region, metal contacts and anti-reflection layers,
45
46 usually made with SiN or SiO2, all of which have non-zero absorption in the thermal wavelength
47
48 range, together with the light trapping effect, result in higher absorptivity/emissivity of the solar
49
50
51 cells in the thermal wavelength range as compared to the planar silicon wafers measured in Ref.
52
53 [17]. For the encapsulated cell, its thermal emissivity is even higher in the most of the
54
55
56
wavelength range between 2-25 µm, and is mainly determined by the 3.2mm thick front glass
57
58
59
60
7
1
2
3
layer which contains 70%-80% silica39. As shown in Fig. 2b, the measured emissivity spectrum
4
5
6 of the encapsulated cell is very similar to the measured or calculated emissivity spectra of fused
7
8 silica16,17.
9
10
11
To summarize this section, a commercial bare solar cell has strong sub-band gap
12
13 absorption in the solar wavelength range, as well as substantial thermal emission in the
14
15 wavelength range of 2-25 µm. Encapsulation of solar cells, as is required in photovoltaic
16
17
18 modules, further enhances the sub-band gap solar absorption and the thermal emission, and in
19
20 addition enhances the solar absorption in the UV wavelength range.
21
22 From the perspective of thermal management, the presence of strong solar absorption in
23
24
25 the sub-band gap and UV wavelength range is detrimental since this absorption represents a
26
27 parasitic heat source. The presence of high emissivity in the thermal wavelength range is
28
29
beneficial. It is important to note, however, that the thermal emissivity spectra of commercial
30
31
32 solar cells, even in the best-case scenario with the encapsulation layer, is sub-optimal. As was
33
34 pointed out in Refs. [16] and [17], the emissivity exhibits a large dip near 9 µm due to the bulk
35
36
37 phonon-polariton excitation of silica. This dip coincides with the transparency window of the
38
39 atmosphere. Refs. [16] and [17] have sought to maximize the thermal emissivity of the silica
40
41 layer by surface patterning. Such an approach is difficult to be implemented as a retrofit on
42
43
44 existing panels, and also does not address the parasitic solar absorption issue that exist in
45
46 commercial solar cells.
47
48
49
50
51 Our approach
52
53 Based on the experimental characterization of commercial solar cells as presented in the
54
55
56
previous section, we now develop a comprehensive photonic approach for the thermal
57
58
59
60
8
1
2
3
management of solar cells. We propose to add a multilayer dielectric stack on top of an existing
4
5
6 encapsulated solar panel. This approach can be used as a retrofit: it does not require any
7
8 modification of existing structures or materials in standard encapsulated cell.
9
10
11
For optimal thermal management purposes, the added photonic structure needs to satisfy
12
13 the following design criteria as shown in Fig. 3a. First, in the solar wavelength range, the
14
15 photonic structure needs to have a tailored transmission/reflection characteristic. In the
16
17
18 wavelength range of 0.3 – 0.375 µm and 1.1 – 4 µm, the photonic structure needs to have
19
20 maximized reflection to reduce the parasitic heat generation. Whereas in the wavelength range of
21
22 0.375 – 1.1 µm where photons can be converted into photocurrent, the photonic structure needs
23
24
25 to have minimum reflection to reduce the reflection loss. Second, to maximize the radiative
26
27 cooling performance of the solar cell, the photonic structure needs to have maximized emissivity
28
29
in the thermal wavelength range beyond 4 µm. In order to satisfy these criteria, we note the
30
31
32 following considerations: First, the constituent materials all need to be transparent over the solar
33
34 wavelength yet some need to be lossy over the thermal wavelength. Second, to create large solar
35
36
37 reflection over the sub-band gap wavelength range, materials with large index contrast need to be
38
39 used. Third, over the thermal wavelength range, many lossy dielectrics such as silica have strong
40
41 phonon-polariton response which results in a negative permittivity that leads to large reflectivity
42
43
44 and hence low emissivity. Therefore, one needs to combine dielectrics with both positive and
45
46 negative permittivity to avoid the large reflection. Finally, all the materials should be commonly
47
48 used dielectrics for cost consideration and should be amenable to large-area fabrication. Taking
49
50
51 into account all these considerations, for our multilayer design we choose to use the dielectric
52
53 materials of silica, alumina, titania and silicon nitride40–42.
54
55
56
57
58
59
60
9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28 Figure 3 a, The ideal reflection (blue) and emissivity (red) spectra of a photonic cooler, with the
29 normalized AM1.5 solar spectrum and a realistic atmosphere transmittance model for winter in
30 California from Ref. [37] plotted as the orange and the light blue shaded areas. The wavelength
31 is plotted with a logarithmic scale for clarity. b, Schematic of a photonic cooler made of
32 multilayer dielectric stack, with n sub-layers. Each sub-layer is made of Al2O3/SiN/TiO2/SiN,
33
34 and a single SiO2 layer on top. The photonic cooler is on a glass substrate. The structure
35 thickness is aperiodic for optimized performance. c, Reflection spectrum of the photonic cooler
36 (blue) with n = 11 sub-layers. The reflection spectrum of a glass substrate without a photonic
37 cooler (dashed green) is plotted for reference. The wavelength is plotted with a logarithmic scale
38 for clarity. d, Emissivity spectra of the photonic cooler over the infrared wavelength, with a
39
40
realistic atmosphere transmittance model for winter in California from Ref. [37] plotted as the
41 shaded area. e, Angle-dependent emissivity spectra of the photonic cooler over the infrared
42 wavelength. f, The total solar power (black dot), the transmitted solar power in the wavelength
43 range of 0.375 – 1.1 µm (blue dot), the reflected parasitic solar power in the wavelength range of
44 0.3 – 0.375 µm and 1.1 – 4 µm (green dot) and the average emissivity ε (red dot) between 8 and
45 13 µm, as a function of n. Data points are connected by lines for visualization.
46
47
48 We perform an extensive numerical optimization procedure to achieve the photonic
49
50 cooler design schematically shown in Fig. 3b. The photonic cooler is on a glass substrate
51
52
53
representing the top layer of an encapsulated solar cell, consists of alternating layers of
54
55 Al2O3/SiN/TiO2/SiN with aperiodic arrangement of thickness, with a single top layer of SiO2 for
56
57 anti-reflection purpose. The use of aperiodic structure, as opposed to a periodic structure, serves
58
59
60
10
1
2
3
to minimize the reflection oscillation in the wavelength range of 0.375 to 1.1 µm. The layer
4
5
6 thickness of the design is presented in Supporting Information Table S1. The calculated solar
7
8 reflectivity and thermal emissivity spectra of this structure are shown in Fig. 3c and 3d,
9
10
11
respectively. Over the solar spectrum, the structure exhibits strong reflection in the sub-band gap
12
13 and UV wavelength range. In the wavelength range of 0.375 to 1.1 µm, this cooler also serves as
14
15 broadband anti-reflection coating and enhances the solar transmission, as compared to the case
16
17
18 without coating (dashed green curve in Fig. 3c). In this wavelength range, the transmitted power
19
20 through this photonic cooler is calculated as 772.6 W/m2, which is more than the case without
21
22 the photonic cooler 755.5 W/m2. In the wavelength range between 1.3 and 1.8 micron, the
23
24
25 cooler structure shows a reflectivity that is near unity, which is important in suppressing the
26
27 parasitic absorption and heat generation from part of the solar spectrum that is below the silicon
28
29
band gap. Over the thermal wavelength, this cooler also shows a broadband high emissivity
30
31
32 (Fig. 3d). In particular, over the 8-13 µm atmosphere’s transparency window, this cooler shows
33
34 remarkably high emissivity, as opposed to silica, and this high emissivity persists to large angles
35
36
37 (Fig. 3e), a useful feature to maximize the radiative cooling power and reminiscent of the
38
39 behavior of hyperbolic metamaterials43. The performance of this cooler can be even further
40
41 enhanced by adding more layers (Fig. 3f) or using other materials19. It should also be noted that,
42
43
44 here the photonic cooler is optimized for photovoltaic modules with solar irradiance from normal
45
46 or with small incident angles. However, it can also be designed to work for photovoltaic modules
47
48 with large incident angles ranges without having reflection loss in the useful solar spectrum from
49
50
51 0.375 to 1.1 µm (Supporting Information Fig. S1).
52
53
54
55
56
57
58
59
60
11
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44 Figure 4 a, Schematic of thermal simulation for a solar panel. h1 and h2 are the non-radiative
45 heat transfer coefficients at the top and bottom surfaces, respectively. b, Under AM1.5
46 illumination, simulated temperature distribution across a solar panel without a photonic cooler
47
(purple), with a photonic cooler (red), and in the ideal situation (green), respectively. The non-
48
49 radiative heat transfer coefficient in this calculation are h1 = 10 W/m2/K and h2 = 5 W/m2/K. The
50 top and bottom ambient temperature are both 298 K. c, Under AM1.5 illumination, operating
51 temperature of the solar panel, without a photonic cooler (purple), with a photonic cooler (red),
52 and in the ideal situation (green), as a function of h1, with a fixed h2 = 5 W/m2/K. d, Under
53 AM1.5 illumination, operating temperature of the solar panel, without a photonic cooler (purple),
54
55
with a photonic cooler (red), and in the ideal situation (green), as a function of h2, with a fixed h1
56 = 10 W/m2/K.
57
58
59
60
12
1
2
3
We then perform a thermal analysis to calculate the cooling effect from this photonic
4
5
6 cooler when it is integrated on top of a real solar panel (see the Methods section). Here, since the
7
8 in-plane temperature variation over the solar cell is sufficiently small, we can use a one-
9
10
11
dimensional thermal model for simplicity. It should also be noted that, in the wavelength range
12
13 of 0.375 to 1.1 µm, the refractive index of the EVA layer is closely matched to the glass and the
14
15 refection from solar cell surface is sufficiently small due to the surface texturing. As a result, the
16
17
18 solar reflection characteristics of this photonic cooler can be largely preserved when it is
19
20 integrated on top of the solar panel. Therefore, we can calculate the absorbed solar power in the
21
22 solar cell based on the solar reflection characteristics of this photonic cooler and experimentally
23
24
25 measured solar cell absorptivity and use it as the heat input in the thermal simulation (see the
26
27 Methods section). We then setup a finite-difference-based thermal simulator on a solar panel
28
29
model (Fig. 4a), with which we can simulate the temperature distribution across the 5 layers in
30
31
32 the solar panel along the vertical direction (Fig. 4b) by solving the steady-state heat diffusion
33
34 equation. In particular, we choose the non-radiative heat transfer coefficient as h1 = 10 W/m2/K
35
36
37 and h2 = 5 W/m2/K, and the ambient temperature as 298 K to mimic a typical outdoor
38
39 condition44. As shown in Fig. 4b, the temperature across a solar panel can be significantly
40
41 reduced by applying a photonic cooler on the top surface. We also calculate the cell operating
42
43
44 temperature by spatially averaging the temperature inside the solar cell region. As a result, the
45
46 cell operating temperature can readily be lowered by 5.7K with current photonic cooler design
47
48 and by 8.6K with an ideal photonic cooler (Fig. 3a). For a 22% efficiency silicon solar cell with a
49
50
51 temperature coefficient of 0.45%6, this 5.7K temperature reduction can provide an absolute
52
53 efficiency improvement of 0.56%, without the need of modifying the current solar panel
54
55
56
configuration. Together with the solar transmission enhancement in the wavelength range of
57
58
59
60
13
1
2
3
0.375 to 1.1 µm, the overall absolute efficiency improvement is estimated to be around 1%. In
4
5
6 solar cell industry, an improvement of the absolute efficiency by one percentage point is very
7
8 significant. For example, the record module efficiency for commercial crystalline silicon panel
9
10
11
increases from about 22.7% in 2006 [45] to about 23.8% in 2016 [46], or about one percentage
12
13 point over a decade of intensive investments. Our work in highlighting a new theoretical
14
15 pathway for one percentage point improvement in module efficiency therefore represents a
16
17
18 significant advancement. It should also be noted that, here the calculation is based on the
19
20 encapsulated cell made of cell I, which has a much smaller sub-band gap absorption (85 W/m2)
21
22 compared to cell II (150 W/m2). For solar panels made of cell II, a much more significant cooling
23
24
25 effect with temperature reduction of 7.7K and hence an even larger efficiency gain can be
26
27 achieved (Supporting Information Fig. S2).
28
29
To further understand the cooling effects under various heat transfer conditions, we also
30
31
32 calculate the cell operating temperature as a function of non-radiative heat transfer coefficient at
33
34 the top (Fig. 4c) and bottom surface (Fig. 4d). This photonic cooling can still have a significant
35
36
37 impact even in the presence of significant non-radiative cooling. For instance, when the back
38
39 side of solar panel has a strong non-radiative heat transfer coefficient h2 = 30 W/m2/K, which
40
41 corresponds to a wind speed of 9 m/s (Ref. [25]), our photonic cooler can still achieve a
42
43
44 temperature reduction of 3.1K (Fig. 4d), indicating its potential of being used in conjunction with
45
46 other non-radiative cooling techniques.
47
48
49
50
51
52
53
54
55
56
57
58
59
60
14
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51 Figure 5 a, Schematic of photonic cooling in low concentration photovoltaic system. b, The
52 reflection spectrum of the photonic cooler as a function of half incident angle up to 30°. c, Under
53
54
10 Suns intensity, operating temperature of the solar panel, without a photonic cooler (purple),
55 with a photonic cooler (red), and in the ideal situation (green), as a function of h2, with a fixed h1
56 = 30 W/m2/K. d, To maintain the solar cell in ambient temperature, the require cooling power for
57 a solar panel, without a photonic cooler (purple), with a photonic cooler (red), and in the ideal
58
59
60
15
1
2
3
situation (green), as a function of solar concentration.
4
5
6 So far we have shown that our photonic cooling approach can be used for cooling of non-
7
8 concentrated, flat solar panels. Interestingly, this technique can be even more beneficial in
9
10
11
concentrated photovoltaic systems (CPV)47 where the thermal management is more critical. In
12
13 this case, the primary benefit of our photonic cooling approach arises from suppressing the sub-
14
15 band gap solar absorption. Here we focused on low and medium concentration systems (2X-
16
17
18 100X). In this concentration range, silicon is a commonly used cell material since it leverages the
19
20 huge investment already made in silicon supply chain and reduces the development cost47. As
21
22 schematized in Fig. 5a, here we consider a concentration system with a half incidence angle of
23
24
25 up to 30°, corresponding to roughly an upper limit of typical high efficiency Fresnel-based
26
27 concentrator48. In this off-normal incident angle range, the thermal emissivity spectra of the
28
29
photonic cooler remain nearly unchanged (Fig. 3e) and the solar reflectivity spectra will slightly
30
31
32 blue shift. However, this shift can be easily compensated by modifying the photonic design
33
34 (Supporting Information Table S1). As we shown in Fig. 5b, the redesigned photonic cooler can
35
36
37 efficiently reduce the sub-band gap solar absorption while preserve the high transmission from
38
39 0.375 µm to 1.1 µm for all the incident angles up to 30°. For this cooler structure design, we then
40
41 perform the thermal analysis to see its cooling ability. We consider a CPV system with 10 Suns
42
43
44 intensity concentration and incident angle range from 0° to 30° (Supporting Information Fig. S3).
45
46 In this case, we consider both strong top and bottom non-radiative heat transfer coefficient to
47
48 take into account other conventional cooling mechanisms such as backside finned heat exchanger
49
50
51 or forced convection. We calculate the cell operating temperature as a function of bottom non-
52
53 radiative heat transfer coefficient h2, as shown in Fig. 5c. Remarkably, this photonic cooler can
54
55
56
reduce the cell temperature by 15.4K when h2 is 30 W/m2/K. Even when h2 is over 200 W/m2/K,
57
58
59
60
16
1
2
3
it can still significantly reduce the cell temperature by as much as 6.2K, indicating its potential of
4
5
6 been used in conjunction with other cooling techniques. For a 22% efficiency silicon solar cell
7
8 with a temperature coefficient of 0.45%, the temperature reduction of 15.4K and 6.2K can
9
10
11
provide absolute efficiency improvement of 1.52% and 0.61%, respectively. Alternatively, in a
12
13 CPV system where the operating temperature need to be maintained at a fixed value and active
14
15 cooling techniques such as water cooling10 need to be employed. This photonic cooler can be
16
17
18 used to significantly save the required cooling power (Fig. 5d). And remarkably, the cooling
19
20 effect will linearly scale up with the system concentration ratio (Fig. 5d) and therefore can
21
22 potentially play a significant role in thermal management of CPV system.
23
24
25
26 Discussion
27
28 We have introduced a comprehensive photonic approach for cooling solar cell
29
30
31 temperature in a solar panel scheme by simultaneously enhancing the radiative cooling while
32
33 also reducing the parasitic solar absorption. We propose a cost-effective photonic design made of
34
35 multi-layer dielectric stack to fulfill this requirement. We show that this photonic cooler can
36
37
38 retrofit into the existing solar panel and efficiently reduce the cell temperature. This technique
39
40 can passively cool both non-concentrated and concentrated photovoltaic systems and can be used
41
42
43
in conjunction with other conventional cooling mechanisms. As a proof of principle, we have
44
45 designed a multi-layer structure made of Al2O3, SiN, TiO2 and SiO2. However, this photonic
46
47 cooler can potentially be realized with other materials that can be even more cost-effective. For
48
49
50 instance, many polymers are strongly absorbing in the thermal wavelength and may naturally
51
52 have good radiative cooling abilities19. And all-polymer multilayer films49,50 have been employed
53
54 as mirrors in CPV systems. Therefore, designing our photonic cooler with all-polymer multilayer
55
56
57
58
59
60
17
1
2
3
films49,50 may even further improve the cooling performance and be more suitable for mass
4
5
6 production.
7
8 Although here we have focused on silicon solar cells, this photonic cooler in principle
9
10
11
can be redesigned and applied to any solar cells, and is especially beneficial for the cells with
12
13 larger band gap energy where the amount of sub-band gap solar energy is more significant and
14
15 hence the parasitic heating effect from the sub-band gap energy can be significant. In addition,
16
17
18 the cooling effect of our approach can be much more significant on solar cells in space
19
20 applications where the radiative access to the sun and the universe are the dominant heating
21
22 exchange mechanism.
23
24
25 More generally, our strategy provides a universal approach for cooling any outdoor
26
27 objects where the sunlight utilization is intrinsic either for functional or aesthetic purposes. For
28
29
applications where the object color needs to be preserved, only the visible transparency is
30
31
32 required and over a half of the total solar power serves as parasitic heat source. In this case, our
33
34 photonic cooler can be easily redesigned to selectively transmit the visible spectrum while also
35
36
37 radiatively cool the object. Hence our approach here can potentially be very important for
38
39 applications such as automobile cooling where the need for air-conditioning can reduce fuel
40
41 economy by 20%51. Finally, previous radiative cooling work have shown the ability of
42
43
44 controlling the thermal radiation and solar absorption properties independently, for example
45
46 completely reflects the solar spectrum, making no use of sunlight29, or completely transmits the
47
48 solar spectrum16,17. Here we show that radiative cooling can be used in conjunction with a more
49
50
51 carefully engineered solar energy harvesting. Our approach represents a step forward in
52
53 simultaneously exploiting both radiative cooling and solar energy harvesting.
54
55
56
57
58
59
60
18
1
2
3
Methods
4
5
6 Solar Cell Encapsulation and Characterization. To perform the solar cell encapsulation, a
7
8 solar cell is sandwiched with two 0.46 mm thick EVA films and then covered with a 3.2 mm
9
10
11
thick glass on the top and a 0.5 mm thick back sheet on the bottom, as schematized in Fig. 2d.
12
13 Then the sample is heated up to 150°C in vacuum condition for 10mins and naturally cooled
14
15 down to get the final encapsulated cell. In the wavelength range of 0.3 – 1.8 µm, the solar
16
17
18 reflectance and transmittance of both bare cells and the encapsulated cell was characterized using
19
20 a spectrophotometer (Agilent Cary 6000i) with an unpolarized light source and a high
21
22 Lambertian reflectance standard (Labsphere SRS-99-020). A diffuse reflectance accessory (DRA
23
24
25 1800) with a 150-mm-diameter integrating sphere was used to collect both specular and diffuse
26
27 components of reflection at 8° angle of incidence. In the wavelength range of 2 – 25 µm, a
28
29
Fourier transform infrared spectrometer (Thermo Scientific Nicolet 6700) accompanied with a
30
31
32 diffuse gold integrating sphere (PIKE Technologies), was used to characterize the reflectance and
33
34 transmittance of both bare and encapsulated cells. The absorptivity/emissivity spectra were
35
36
37 obtained by subtracting the reflectance and transmittance from unity.
38
39 Thermal Calculation of Solar Cells. We perform a thermal analysis to calculate the cooling
40
41 effect from this photonic cooler in a real solar panel scheme. Here, since the in plane temperature
42
43
44 variation over the solar cell is sufficiently small, we can use a one-dimensional thermal model
45
46 for simplification. We set up a finite-difference-based thermal simulator over a 5-layer solar
47
48 panel model (Fig. 4a), where we can simulate the temperature distribution across the solar panel
49
50
51 in the vertical direction by solving the steady-state heat diffusion equation:
52
53
54
55 (1)
56
57
58
59
60
19
1
2
3
where T (z) is the temperature distribution across the solar panel. Here the thermal conductivity
4
5
6 values of glass, EVA, silicon solar cell, and Tedla are 0.98 W/m/K, 0.24 W/m/K, 148 W/m/K and
7
8 0.36 W/m/K, respectively, according to Ref. [52]. We apply the thermal boundary condition at
9
10
11 the top surfaces of the solar panel as:
12
13 dT (z)
14 −k(z) top = Pcooling (Ttop ) + h1 (Ttop − Tamb ) (2)
15 dz
16
17
to take into account both the radiative cooling effect Pcooling (Ttop ) , as well as additional non-
18
19 radiative heat dissipation due to convection and conduction, as characterized by h1 (Ttop − Tamb ) .
20
21
22 At the lower surface, we assume a boundary condition
23
24
25 dT (z)
k(z) bottom = h2 (Tbottom − Tamb ) (3)
26 dz
27
28
29 to characterize the non-radiative heat loss of the lower surface.
30
31 Based on the solar absorptivity and thermal emissivity spectra of the encapsulated cell
32
33
34 and the photonic cooler, we can first calculate the total solar absorption and radiative cooling
35
36 power of the solar panels by:
37
38 ∞
39 Psun = ∫ d λ ⋅ I AM 1.5 (λ )⋅ ε (λ ,θ sun )⋅(1− r(λ ,θ sun ))⋅ cosθ sun (4)
40 0
41
42
43 where Psun is the solar absorption power, I AM 1.5 (λ ) is the AM1.5 spectrum, ε (λ ,θ sun ) is the solar
44
45 absorptivity of the solar cell. r(λ ,θ sun ) is the reflectivity spectrum of the photonic cooler. θ sun is
46
47
48 the solar incidence angle. Here we assume the heat is uniformly generated in the solar cell and
49
50 we can obtain in Eq.1. Meanwhile, the top surface radiative cooling power Pcooling can be
51
52
53 calculated as:
54
55
56 Pcooling (Ttop ) = Prad (Ttop ) − Patm (Tamb ) (5)
57
58
59
60
20
1
2
3
4
Prad is the total thermal radiation power by the solar panel and can be calculated as:
5
6 ∞
7 Prad (Ttop ) = ∫ dΩ ⋅ cosθ ∫ d λ ⋅ I BB (Ttop , λ )ε (λ ,Ω) (6)
0
8
9 π /2 2π
10 where ∫ dΩ = ∫ 0
dθ sinθ ∫ dφ
0
is the angular integral over the hemisphere.
11
I BB ( T , λ ) = ( 2hc 2 / λ 5 ) /  ehc/λ K BT −1 is the spectral radiance of a blackbody at temperature T ,

12
13
14
15
16 where h is the Planck’s constant, c is the velocity of light, K B is the Boltzmann constant.
17
18
19 Prad is the absorbed thermal emission power from the atmosphere and can be calculated as:
20
21 ∞
22 Patm (Tamb ) = ∫ dΩ ⋅cosθ ∫ d λ ⋅ I BB (Tamb , λ )ε (λ ,Ω)ε atm (λ ,Ω) (7)
0
23
24
where ε atm (λ ,Ω) = 1− t(λ )1/cosθ is the angular-dependent emissivity of atmosphere and t(λ ) is the
25
26
27 atmosphere’s transmittance in the zenith direction.
28
29
30 The solution of the heat equation results in a temperature distribution T (z) shown in Fig.
31
32
4b. The operating temperature of the solar cell (Fig. 4c and d) is then defined as the spatially
33
34
35 averaged temperature inside the solar cell region.
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
21
1
2
3
Supporting Information
4
5
6 Supporting Information Available: This material is available free of charge via the Internet at
7
8 http://pubs.acs.org
9
10
11
Supporting Information Fig. S1: Photonic cooler design for large incident angles. Supporting
12
13 Information Fig. S2: Photonic cooling effects on photovoltaic modules made of cell II.
14
15 Supporting Information Fig. S3: A concentrated photovoltaic system design with 10x
16
17
18 concentration factor and 30 degree half incident angle. Supporting Information Table S1: Layer
19
20 thickness parameters for various photonic cooler designs.
21
22
23
24
25 Competing Financial Interests
26
27 The authors declare no competing financial interests.
28
29
30
31
32 Acknowledgements
33
34 This work was supported by the Department of Energy Grants No. DE-FG-07ER46426 and DE-
35
36
37 EE0007544. The authors thank John Stayner for the assistance of solar cell encapsulation
38
39 process.
40
41
42
43
44
45 References
46
47 1. Shockley, W. & Queisser, H. J. Detailed Balance Limit of Efficiency of p-n Junction Solar
48
Cells. J. Appl. Phys. 32, 510 (1961).
49
50 2. Davis, M. W., Fanney, A. H. & Dougherty, B. P. Prediction of Building Integrated
51 Photovoltaic Cell Temperatures. J. Sol. Energy Eng. 123, 200-210 (2001).
52 3. Ingersoll, J. G. Simplified Calculation of Solar Cell Temperatures in Terrestrial
53 Photovoltaic Arrays. J. Sol. Energy Eng. 108, 95-101 (1986).
54 4. Jones, A. D. & Underwood, C. P. A thermal model for photovoltaic systems. Sol. Energy
55
56
70, 349–359 (2001).
57 5. Dupré, O., Vaillon, R. & Green, M. A. Thermal Issues in Photovoltics and Existing
58
59
60
22
1
2
3
Solutions in Thermal Behavior of Photovoltaic Devices 1–28 (Springer International
4
5 Publishing, 2017).
6 6. Skoplaki, E. & Palyvos, J. A. On the temperature dependence of photovoltaic module
7 electrical performance: A review of efficiency/power correlations. Sol. Energy 83, 614–
8 624 (2009).
9 7. Otth, D. H. & Ross, R. G. Assessing photovoltaic module degradation and lifetime from
10
11
long term environmental tests. Proceedings of the 1983 Institute of Environmental
12 Sciences 29th Annual Technical Meeting. pp. 121–126 (1983).
13 8. Royne, A., Dey, C. J. & Mills, D. R. Cooling of photovoltaic cells under concentrated
14 illumination: a critical review. Sol. Energy Mater. Sol. Cells 86, 451–483 (2005).
15 9. Teo, H. G., Lee, P. S. & Hawlader, M. N. A. An active cooling system for photovoltaic
16
modules. Appl. Energy 90, 309–315 (2012).
17
18 10. Moharram, K. A., Abd-Elhady, M. S., Kandil, H. A. & El-Sherif, H. Enhancing the
19 performance of photovoltaic panels by water cooling. Ain Shams Eng. J. 4, 869–877
20 (2013).
21 11. Akbarzadeh, A. & Wadowski, T. Heat pipe-based cooling systems for photovoltaic cells
22 under concentrated solar radiation. Appl. Therm. Eng. 16, 81–87 (1996).
23
24
12. Zhang, Y., Du, Y., Shum, C., Cai, B., Le, N., Chen, X., Duck, B., Fell, C., Zhu, Y., Gu,
25 M. Efficiently-cooled plasmonic amorphous silicon solar cells integrated with a nano-
26 coated heat-pipe plate. Sci. Rep. 6, 24972 (2016).
27 13. Mullaney, K., Jones, G., Kitchen, C. A. & Jones, D. P. Infra-red reflective coverglasses:
28 the next generation. in Photovoltaic Specialists Conference, 1993., Conference Record of
29
the Twenty Third IEEE 1363–1368 (1993).
30
31 14. Beauchamp, W. & Tuttle-Hart, T. UV/IR reflecting solar cell cover. UV/IR reflecting
32 solar cell cover. US Pat. 5, 449, 413 (1995).
33 15. Yoon, H., Joslin, D. E., Law, D. C., Krut, D., King, R. R., Vijayakumar, P., Peterson, D.,
34 Hanley, J., Karam, N. H. Application of infrared reflecting (IRR) coverglass on
35 multijunction III-V solar cells. in 2006 IEEE 4th World Conference on Photovoltaic
36
37 Energy Conference 1861–1864 (2006).
38 16. Zhu, L., Raman, A., Wang, K. X., Anoma, M. A. & Fan, S. Radiative cooling of solar
39 cells. Optica 1, 32 (2014).
40 17. Zhu, L., Raman, A. P. & Fan, S. Radiative cooling of solar absorbers using a visibly
41 transparent photonic crystal thermal blackbody. Proc. Natl. Acad. Sci. USA 112, 12282–
42
43
12287 (2015).
44 18. Safi, T. S. & Munday, J. N. Improving photovoltaic performance through radiative
45 cooling in both terrestrial and extraterrestrial environments. Opt. Express 23, A1120
46 (2015).
47 19. Wu, S.-H. & Povinelli, M. L. Solar heating of GaAs nanowire solar cells. Opt. Express 23,
48 A1363 (2015).
49
50 20. Bermel, P., Boriskina, S. V., Yu, Z. & Joulain, K. Control of radiative processes for
51 energy conversion and harvesting. Opt. Express 23, A1533 (2015).
52 21. Catalanotti, S., Cuomo, V., Prio, G., Ruggi, D., Silvestrini, V., Troise, G. The radiative
53 cooling of selective surfaces. Sol. Energy 17, 83–89 (1975).
54 22. Harrison, A. W. & Walton, M. R. Radiative cooling of TiO2 white paint. Solar Energy 20,
55
56
(Pergamon, 1978).
57 23. Granqvist, C. G. & Hjortsberg, A. Radiative cooling to low temperatures: General
58
59
60
23
1
2
3
considerations and application to selectively emitting SiO films. J. Appl. Phys. 52, 4205-
4
5 4220 (1981).
6 24. Nilsson, T. M. J., Niklasson, G. A. & Granqvist, C. G. A solar reflecting material for
7 radiative cooling applications: ZnS pigmented polyethylene. Sol. Energy Mater. Sol. Cells
8 28, 175–193 (1992).
9 25. Nilsson, T. M. J. & Niklasson, G. A. Radiative cooling during the day: simulations and
10
11
experiments on pigmented polyethylene cover foils. Sol. Energy Mater. Sol. Cells 37, 93–
12 118 (1995).
13 26. Gentle, A. R., Aguilar, J. L. C. & Smith, G. B. Optimized cool roofs: Integrating albedo
14 and thermal emittance with R-value. Sol. Energy Mater. Sol. Cells 95, 3207–3215 (2011).
15 27. Zhu, L., Raman, A. & Fan, S. Color-preserving daytime radiative cooling. Appl. Phys.
16
Lett. 103, 223902 (2013).
17
18 28. Rephaeli, E., Raman, A. & Fan, S. Ultrabroadband Photonic Structures To Achieve High-
19 Performance Daytime Radiative Cooling. Nano Lett. 13, 1457-1461 (2013).
20 29. Raman, A. P., Anoma, M. A., Zhu, L., Rephaeli, E. & Fan, S. Passive radiative cooling
21 below ambient air temperature under direct sunlight. Nature 515, 540–544 (2014).
22 30. Tong, J. K., Huang, X., Boriskina, S. V., Loomis, J., Xu, Y., Chen, G. Infrared-
23
24
Transparent Visible-Opaque Fabrics for Wearable Personal Thermal Management. ACS
25 Photonics 2, 769–778 (2015).
26 31. Shi, N. N., Tsai, C.-C., Camino, F., Bernard, G. D., Yu, N., Wehner, R. Keeping cool:
27 Enhanced optical reflection and radiative heat dissipation in Saharan silver ants. Science
28 349, 298-301 (2015).
29
32. Chen, Z., Zhu, L., Raman, A., Fan, S. Radiative cooling to deep sub-freezing temperatures
30
31 through a 24-h day–night cycle. Nat. Commun. 7, 13729 (2016).
32 33. Saga, T. Advances in crystalline silicon solar cell technology for industrial mass
33 production. NPG Asia Mater. 2, 96–102 (2010).
34 34. Smith, D., Cousins, P., Masad, A., Westerberg, S., Defensor, M., Ilaw, R., Dennis, T.,
35 Daquin, R., Bergstrom, N., Leygo, A., Zhu, X., Meyers, B., Bounre, B., Shields, M., Rose,
36
37 D. SunPower's Maxeon Gen III solar cell: high efficiency and energy yield. Proceedings
38 of IEEE 39th Photovoltaic Specialists Conference (PVSC) 16-21 June 2013, Tampa,
39 Florida. pp. 0908-13 (2013).
40 35. Neuhaus, D. H. & Münzer, A. Industrial silicon wafer solar cells. Adv. Optoelectron.
41 2007, 1–15 (2007).
42
43
36. Campbell, P. & Green, M. A. Light trapping properties of pyramidally textured surfaces.
44 J. Appl. Phys. 62, 243 (1987).
45 37. Berk, A., Anderson, G. P., Acharya, P. K., Bernstein, L. S., Muratov, L., Lee, J., Fox, M.,
46 Adler-Golden, S. M., Chetwynd, J. H. Jr., Hoke, M. L., Lockwood, R. B., Gardner, J. A.,
47 Cooley, T. W., Borel, C. C., Lewis, P. E., Shettle, E. P. MODTRAN5: 2006 update. in
48 Proceedings of SPIE 6233, 62331F–62331F–8 (2006).
49
50 38. Miller, D. C. Analysis of transmitted optical spectrum enabling accelerated testing of
51 multijunction concentrating photovoltaic designs. Opt. Eng. 50, 13003 (2011).
52 39. Kurkjian, C. R. & Prindle, W. R. Perspectives on the History of Glass Composition. J.
53 Am. Ceram. Soc. 81, 795–813 (2005).
54 40. Palik, E. D. & Ghosh, G. Handbook of optical constants of solids (Academic press).
55
56
(1998).
57 41. Kischkat, J., Peters, S., Gruska, B., Semtsiv, M., Chashnikova, M., Klinkmüller, M.,
58
59
60
24
1
2
3
Fedosenko, O., Machulik, S., Aleksandrova, A., Monastyrskyi, G., Flores, Y., Masselink,
4
5 W. T. Mid-infrared optical properties of thin films of aluminum oxide, titanium dioxide,
6 silicon dioxide, aluminum nitride, and silicon nitride. Appl. Opt. 51, 6789–6798 (2012).
7 42. Devore, J. R. Refractive Indices of Rutile and Sphalerite. J. Opt. Soc. Am. 41, 416 (1951).
8 43. Jacob, Z., Kim, J., Naik, G. V., Boltasseva, A., Narimanov, E. E., Shalaev, V. M.
9 Engineering photonic density of states using metamaterials. Appl. Phys. B: Laser Opt.
10
11
100, 215–218 (2010).
12 44. Armstrong, S. & Hurley, W. G. A thermal model for photovoltaic panels under varying
13 atmospheric conditions. Appl. Therm. Eng. 30, 1488–1495 (2010).
14 45. Green, M. A., Emery, K., King, D. L., Hisikawa, Y. and Warta, W. Solar cell efficiency
15 tables (version 27). Prog. Photovolt: Res. Appl. 14, 45–51 (2006).
16
46. Green, M. A., Emery, K., Hishikawa, Y., Warta, W., and Dunlop, E. D. Solar cell
17
18 efficiency tables (version 48). Prog. Photovolt: Res. Appl. 24, 905–913 (2016).
19 47. Kurtz, S. Opportunities and Challenges for Development of a Mature Concentrating
20 Photovoltaic Power Industry Opportunities and Challenges for Development of a Mature
21 Concentrating Photovoltaic Power Industry. Tech. Rep. NREL/TP-52, 43208 (2012).
22 48. Benítez, P., Minano, J. C., Zamora, P., Mohedano, R., Cvetkovic, A., Buljan, M., Chaves,
23
24
J., Hernandez, M. High performance Fresnel-based photovoltaic concentrator. Opt.
25 Express 18, A25 (2010).
26 49. Gilbert, L. R., Jonza, J. M., Ouderkirk, A. J., Stover, C. A., Weber, M. F., Merrill, W. W.
27 Multilayer Polymer Film With Additional Coatings or Layers. US Pat. 6,368,699 (2002).
28 50. Weber, M. F., Stover, C. A., Gilbert, L. R., Nevitt, T. J. & Ouderkirk, A. J. Giant
29
Birefringent Optics in Multilayer Polymer Mirrors. Science 287, 2451–2456 (2000).
30
31 51. Farrington, R. & Rugh, J. Impact of Vehicle Air-Conditioning on Fuel Economy, Tailpipe
32 Emissions, and Electric Vehicle Range. (NREL/CP-540-28960, 2000:
33 http://www.nrel.gov/docs/fy00osti/28960.pdf )
34 52. Lee, B., Liu, J. Z., Sun, B., Shen, C. Y. & Dai, G. C. Thermally conductive and
35 electrically insulating EVA composite encapsulant for solar photovoltaic (PV) cell.
36
37 Express Polym. Lett. 2, 357–363 (2008).
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
25
1
2
3
For Table of Contents Use Only
4
5
6
7
8 A comprehensive photonic approach for solar cell cooling
9
10
11
Wei Li1, Yu Shi1, Kaifeng Chen1,2, Linxiao Zhu2 and Shanhui Fan1*
12 1
13 Department of Electrical Engineering, Ginzton Laboratory, Stanford University, Stanford, California
14 94305, USA
2
15 Department of Applied Physics, Stanford University, Stanford, California 94305, USA
*
16 email: shanhui@stanford.edu
17
18
19
20
21
22 Table of Contents Graphic
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
26

Li 2017

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Li 2017

Uploaded by

Copyright:

Available Formats

Subscriber access provided by University of Newcastle, Australia

I BB ( T , λ ) = ( 2hc 2 / λ 5 ) /  ehc/λ K BT −1 is the spectral radiance of a blackbody at temperature T ,

You might also like