Optimization of the Secondary Optical Element of a Hybrid Concentrator Photovoltaic Module Considering the Eﬀective Absorption Wavelength Range

applied sciences

Article Optimization of the Secondary Optical Element of a Hybrid Concentrator Photovoltaic Module Considering the Eﬀective Absorption Wavelength Range

1,2, 1,2, 1 1,2 Woo-Lim Jeong † , Kyung-Pil Kim †, Jung-Hong Min , Jun-Yeob Lee , Seung-Hyun Mun 1,2, Jeong-Hwan Park 1, Jang-Hwan Han 1, Won-Kyu Park 3, Sewang Yoon 4 and Dong-Seon Lee 1,2,*

1 School of Electrical Engineering and Computer Science, Gwangju Institute of Science and Technology, Gwangju 61005, Korea; [email protected] (W.-L.J.); [email protected] (K.-P.K.); [email protected] (J.-H.M.); [email protected] (J.-Y.L.); [email protected] (S.-H.M.); [email protected] (J.-H.P.); [email protected] (J.-H.H.) 2 Research Institute for Solar and Sustainable Energies, Gwangju Institute of Science and Technology, Gwangju 61005, Korea 3 Korea Advanced Nano Fab Center, Suwon 16229, Korea; [email protected] 4 194 Central-ro, Yeonsu-gu, Incheon 22003, Korea; [email protected] * Correspondence: [email protected]; Tel.: +82-62-715-2248 These authors contribute equally to this work. † Received: 27 February 2020; Accepted: 10 March 2020; Published: 18 March 2020

Abstract: Hybrid concentrator photovoltaic (CPV) architectures that combine CPV modules with low-cost solar cells have the advantage of functioning well in modest direct normal irradiance (DNI) regions as well as high-DNI regions, where these architectures allow for higher performance in a limited space. For higher performance of a hybrid CPV module, we optimized the secondary optical element (SOE) using raytracing software and conducted experimental measurements that consider the eﬀective wavelength range. Our experiments with the optimized SOE (θ = 30◦, h = 15 mm) demonstrated a maximum output power on the triple-junction cell and polycrystalline silicon cell of 212.8 W/m2 and 5.14 W/m2, respectively.

Keywords: hybrid concentrator photovoltaic; secondary optical element; Fresnel lens

1. Introduction The origin of the technique to obtain solar energy effectively using a concentrator is considered to be a combination of reflector and thermal collector, since the late 1970s [1,2]. The technique has been further advanced, and concentrator photovoltaic (CPV) modules have been actively studied to reduce costs and improve performance [3–12]. CPV modules are highly efficient power generators in high-direct normal irradiance (DNI) regions near the equator, but less effective in the global normal irradiance (GNI) mid-latitude regions, where these modules do not perform as well as flat photovoltaic (PV) modules. To overcome these limitations, several researchers have studied hybrid CPV modules that combine a CPV module with low-cost PV modules [13–16], since these hybrid CPV architectures can collect sunlight in modest-DNI regions, which allows the same solar module design regardless of region. In addition, hybrid CPV modules can maximize power generation in the high-DNI regions without requiring as much volume, since they absorb scattered light more efficiently [11]. Lee et al. experimentally analyzed two types of hybrid CPV modules with concentration ratios of 18X and

Appl. Sci. 2020, 10, 2051; doi:10.3390/app10062051 www.mdpi.com/journal/applsci Appl. Sci. 2020, 10, 2051 2 of 9

Appl. Sci. 2020, 10, 2051 2 of 9 1000X and showed high power generation eﬃciency [12]. Yamada et al. combined the two-terminal triple-junctiontwo-terminal triple-junction (TJ) solar cell and (TJ) bifacial solar cell crystalline and bifa siliconcial crystalline solar cell forsilicon their solar hybrid cell CPV for modules, their hybrid and performedCPV modules, analyses and performed based on the analyses (GNI–DNI) based/GNI on ratiothe (GNI–DNI)/GNI [13]. While many ratio studies [13]. haveWhile been many conducted studies onhave the been secondary conducted optical on the elements secondary (SOEs) optical used elem in conventionalents (SOEs) used CPV modulesin conventional [17–19 ],CPV we havemodules not seen[17–19], a study we have on the not SOEs seen ofa study hybrid on CPV the modules.SOEs of hybrid In this CPV paper, modules. we analyzed In this the paper, mirror-type we analyzed SOE forthe themirror-type hybrid CPV SOE modules for the and hybrid optimized CPV itmodules through simulationand optimized and experimentalit through simulation measurements and underexperimental high-DNI measurements conditions. under Especially high-DNI analyzed conditions. were the Especially shading lossesanalyzed of thewere silicon the shading solar cells losses by theof the SOEs silicon on the solar hybrid cells CPV by the modules. SOEs on In the addition, hybrid the CPV two-terminal modules. In TJ solaraddition, cell, mainlythe two-terminal used in CPV TJ modules,solar cell, cannotmainly convertused in CPV light modules, of wavelength canno abovet convert about light 1.3 ofµ wavelengthm, due to the above current about limit 1.3 of μ them, due top cellto the [20 current,21]. This limit longer of the wavelength top cell [20,21]. light is This also lo notnger absorbed wavelength by the light silicon is also solar not cells absorbed in the hybridby the CPVsilicon modules solar cells around in the the hybrid TJ solar CPV cell. modules For this around reason, the we TJ considered solar cell. theFor ethisﬀective reason, wavelength we considered range forthe theeffective optimization wavelength of the range SOE. Thesefor the hybrid optimizati CPVon modules of the areSOE. promising These hybrid as the nextCPV generationmodules are of thepromising thermal as collector the next system. generation of the thermal collector system.

2. Simulation Setup The angle angle at at which which sunlight sunlight refracts refracts from from the the primary primary optical optical element element (POE) (POE) of the of hybrid the hybrid CPV CPVmodule module depends depends on the on wavelength, the wavelength, which whichcreates createschromatic chromatic aberration aberration [22]. In [addition,22]. In addition,infrared infraredrays with rays wavelengths with wavelengths above 1.3 above μm are 1.3 notµm absorbed are not absorbedin either TJ in sola eitherr cells TJ solar or silicon cells orsolar silicon cells, solar due cells,to current due tolimits current in the limits top cell in the and top limited cell and semi limitedconductor semiconductor bandgap energy, bandgap respectively. energy, respectively. Therefore, Therefore,we considered we considered the effective the e ffwavelengthective wavelength range rangewhile whilewe optimized we optimized the theSOE SOE of ofthe the hybrid hybrid CPV module. ToTo analyze analyze the the hybrid hybrid CPV architecture,CPV architecture, we used we the used TracePro the (LambdaTracePro Research (Lambda Corporation, Research Littleton,Corporation, Massachusetts, Littleton, Massachusetts, USA) raytracing USA) software, raytracing based software, on a Monte based Carlo on methoda Monte with Carlo the method xenon lampwith the spectrum xenon lamp (1000 spectrum W/m2), to (1000 create W/m the2), three-dimensional to create the three-dimens model shownional model in Figure shown1. Assuming in Figure modest-DNI1. Assuming regions,modest-DNI sunlight regions, was designedsunlight withwas designed 80% DNI andwith 20% 80% di DNIffuse an light.d 20% We diffuse selected light. a 254 We × 254selected3 mm a 2543 Fresnel × 254 × lens 3 mm made3 Fresnel of polymethyl lens made methacrylate of polymethyl (PMMA; methacrylate refractive (PMMA; index = refractive1.49 at 500 index nm) × and= 1.49 an at Al 500 mirror nm) (92%and an reflective Al mirror at 500(92% nm) reflective as the POEat 500 and nm) SOE, as the respectively. POE and SOE, The focal respectively. length of The the POEfocal waslength 250 of mm, the POE and thewas body 250 mm, size and was the 254 body254 size253 was mm 2543 .× We 254 placed × 253 mm the3 silicon. We placed solar the cell silicon 8 mm × × awaysolar cell from 8 themm TJ away solar from cell in the position TJ solar #2, cell and in the position spacing #2, between and the the spacing silicon between solar cells the was silicon 3 mm. solar For easecells ofwas cleaning 3 mm. theFor module,ease of cleaning the flat side the ofmodule, the Fresnel the flat lens side faced of the sun,Fresnel and lens the faced stepped the side sun, faced and the solarstepped cells. side We faced placed the a high-performance solar cells. We placed TJ solar a cellhigh-performance (10 10 mm2; concentration TJ solar cell ratio(10 ×= 10645X) mm at2; × theconcentration center of the ratio hybrid = 645X) CPV at modulethe center to of collect the hy thebrid concentrated CPV module DNI. to collect Around the the concentrated TJ solar cell, DNI. we spreadAround 12 the silicon TJ solar solar cell, cells we (80 spread55 mm 122) silicon to collect solar light cells that (80 was × scattered55 mm2) from to collect the concentrator light that was and × discatteredffuse solar from irradiance. the concentrator and diffuse solar irradiance.

Figure 1. (a) The raytracing model and (b) raytracing results of the hybridhybrid concentrator photovoltaic (CPV) modulemodule combinedcombined withwith low-costlow-cost siliconsilicon solarsolar cells.cells.

Appl. Sci. 2020, 10, 2051 3 of 9 Appl. Sci. 2020, 10, 2051 3 of 9

3.3. Simulations

3.1.3.1. Optimization ofof SOESOE DesignDesign

FigureFigure2 2 shows shows our our SOE SOE designs designs with with di differentfferent anglesangles whenwhen placedplaced onon thethe TJTJ solarsolar cells.cells. AtAt thisthis point,point, wewe fixedfixed thethe SOESOE heightheight atat 1515 mmmm andand maintainedmaintained aa constantconstant distancedistance betweenbetween thethe FresnelFresnel lenslens andand thethe TJ solar cell. Using Using these these SOEs, SOEs, we we calculated calculated the the irradiance irradiance on on the the TJ TJ solar solar cell cell and and on on the the 12 12silicon silicon solar solar cells cells (positions (positions #1 #1 to to #12 #12 in in Table Table 1).1). We We also analyzedanalyzed thethe irradianceirradiance byby dividingdividing the the resultsresults intointo anan eeffectiveffective wavelengthwavelength rangerange (0.3–1.299(0.3–1.299 µμm)m) andand a lossy wavelength range (1.3–2.5 μµm). TheThe irradianceirradiance on on the the TJ solar TJ solar cell in cell the einffective the wavelengtheffective wavelength range was minimizedrange was for minimized a perpendicular for a 2 2 SOEperpendicular (326.8 W/m SOE), and (326.8 maximized W/m2), whenand maximizedθ = 30◦ (606.2 when W/m θ );= when30° (606.2 the angle W/m is2); greater when thanthe angle 30◦, the is lightgreater is reflectedthan 30°, and the thelight intensity is reflected of the and light the is inte reduced.nsity of Although the light theis reduced. sum of power Although in the the eff sumective of andpower lossy in wavelengththe effective ranges and lossy was higherwavelength in the θranges= 40◦ wascase higher than the inθ the= 30 θ◦ =case, 40° thecase E TJ,lossthan /theETJ,e θff =ratio 30° indicates that the loss ratio of the θ = 40 (6.9%) case was higher than that of the θ = 30 (5.7%) case, so case, the ETJ,loss/ETJ,eff ratio indicates that◦ the loss ratio of the θ = 40° (6.9%) case was higher◦ than that θof= the30 ◦θseems = 30° (5.7%) more optimal.case, so θ In = addition,30° seems higher more optimal. SOE angles In addition, in the silicon higher solar SOE cell angles will decreasein the silicon the irradiance,solar cell will due decrease to shading the fromirradian thece, SOE. due to shading from the SOE.

Figure 2. Secondary optical element (SOE) designs at a 15 mm height with different mirror angles of (a) Figure 2. Secondary optical element (SOE) designs at a 15 mm height with different mirror angles of θ = 0◦,(b) θ = 10◦,(c) θ = 20◦,(d) θ = 30◦,(e) θ = 40◦, and (f) θ = 50◦. (a) θ = 0°, (b) θ = 10°, (c) θ = 20°, (d) θ = 30°, (e) θ = 40°, and (f) θ = 50°. Table 1. Irradiance on the triple-junction and silicon solar cells of the hybrid concentrator photovoltaic (CPV)Table 1. module Irradiance with on di thefferent triple-junction mirror angles and andsilicon a 15sola mmr cells height, of the as hybrid calculated concentrator using the photovoltaic TracePro raytracing(CPV) module software. with Wedifferent separated mirror the angles results and into a the 15 emmffective height, (0.3–1.299 as calculatedµm) and using lossy the (1.3–2.5 TraceProµm) wavelengthraytracing software. regions. We separated the results into the effective (0.3–1.299 μm) and lossy (1.3–2.5 μm) wavelength regions. 2 2 2 2 SOE Angle ETJ,eff (W/m )ETJ,loss (W/m ) ETJ,loss/ETJ,eff ratio (%) ESi,eff (W/m )ESi,loss (W/m ) SOE Angle ETJ,eff (W/m2) ETJ,loss (W/m2) ETJ,loss/ETJ,eff ratio (%) ESi,eff (W/m2) ESi,loss (W/m2) 0◦ (Perpendicular) 326.8 7.2 2.2 135.8 7.3 0° (Perpendicular)10◦ 495.6326.8 15.17.2 3.1 2.2 130.6 135.8 7.3 7.1 20◦10° 588.6495.6 24.115.1 4.1 3.1 129.8 130.6 7.1 13.5 30◦20° 606.2588.6 34.524.1 5.7 4.1 127.4 129.8 13.5 9.9 40 604.7 41.7 6.9 125.8 10.2 ◦30° 606.2 34.5 5.7 127.4 9.9 50◦ 603.4 26.1 4.3 125.1 5.2 40° 604.7 41.7 6.9 125.8 10.2 50° 603.4 26.1 4.3 125.1 5.2 Next, we analyzed the irradiance with a fixed 30◦ SOE angle as we varied the SOE height from 0 mm toNext, 100 mm,we analyzed as shown the in irradiance Figure3 and with Table a 2fixed. Since 30° the SOE conversion angle as we e ffi variedciency the of theSOE TJ height solar cell from is better0 mm thanto 100 the mm, low-cost as shown silicon in solarFigure cells, 3 and the Table hybrid 2. CPVSince module the conversion configuration efficiency should of the be adjustedTJ solar cell for 2 betteris better light than concentration. the low-cost Whensilicon the solar height cells, was thezero hybrid (i.e., CPV no SOE), module we configuration calculated powers should of 590.8be adjusted W/m 2 andfor better 131.2 light W/m concentration.on the TJ solar When celland the height the silicon was solarzero (i.e., cells, no respectively. SOE), we calculated Although powers larger heightsof 590.8 ledW/m to2 higherand 131.2 power W/m values2 on the on TJ the solar TJ solarcell and cell, the the silicon irradiance solar withincells, respectively. the lossy wavelength Although rangelarger increasedheights led faster to higher than the power irradiance values within on the the TJ esolarffective cell, wavelength the irradiance range, within as shown the lossy by the wavelength increasing ErangeTJ,loss /increasedETJ,eff ratio. faster Moreover, than the since irradiance taller SOEs within increase the effective the production wavelength cost andrange, maximize as shown shading, by the 2 moderatelyincreasing E tallTJ,loss SOEs/ETJ,eff are ratio. more Moreover, desirable. since While taller the irradianceSOEs incr onease the the silicon production solar cells cost was and 127.4 maximize W/m 2 2 forshading, h = 15 moderately mm, the irradiances tall SOEs are on themore silicon desirable. solar While cells were the irradiance 124.8 W/m onand the 85.4silicon W /solarm for cells h = was20 and127.4 100 W/m mm,2 for respectively, h = 15 mm, duethe irradiances to the shading on the on thesilicon SOEs. solar Therefore, cells were we 124.8 focused W/m our2 and subsequent 85.4 W/m2 simulationsfor h = 20 and and 100 experimental mm, respectively, measurements due to on the the shading case of on h = the15 mmSOEs. and Therefore,θ = 30◦. we focused our subsequent simulations and experimental measurements on the case of h = 15 mm and θ = 30°. Appl. Sci. 2020, 10, 2051 4 of 9 Appl. Sci. 2020, 10, 2051 4 of 9

Figure 3. SOE designs with different mirror heights of (a) 0 mm (no SOE), (b) 5 mm, (c) 10 mm, (d) 15 Figure 3. SOE designs with different mirror heights of (a) 0 mm (no SOE), (b) 5 mm, (c) 10 mm, (d) 15 mm, (e) 20 mm, and (f) 100 mm. mm, (e) 20 mm, and (f) 100 mm. Table 2. Irradiance on the triple-junction and silicon solar cells of the hybrid CPV module with different mirrorTable heights,2. Irradiance as calculated on the usingtriple-junction TracePro. and We againsilicon separated solar cells the of results the hybrid into the CPV effective module (0.3–1.299 with µdifferentm) and lossymirror (1.3–2.5 heights,µm) as calculated wavelength using regions. TracePro. We again separated the results into the effective (0.3–1.299 μm) and lossy (1.3–2.5 μm) wavelength regions. 2 2 2 2 SOE Height (mm) ETJ,eff (W/m )ETJ,loss (W/m ) ETJ, loss/ETJ,eff Ratio (%) ESi,eff (W/m )ESi,loss (W/m ) SOE Height (mm) ETJ,eff (W/m2) ETJ,loss (W/m2) ETJ, loss/ETJ,eff ratio (%) ESi,eff (W/m2) ESi,loss (W/m2) 0 (No SOE) 590.8 20.4 3.5 131.2 7.1 0 5(No SOE) 601.8 590.8 26.1 20.4 4.3 3.5 131.2 131.5 7.1 9.8 105 605.2601.8 30.926.1 5.1 4.3 131.5 130.3 9.8 10.4 1510 606.2605.2 34.530.9 5.7 5.1 130.3 127.4 10.4 9.9 2015 608.3606.2 37.934.5 6.2 5.7 127.4 124.8 9.9 9.3 100 609.7 47.4 7.8 85.4 1.6 20 608.3 37.9 6.2 124.8 9.3 100 609.7 47.4 7.8 85.4 1.6 3.2. Irradiance Distributions 3.2. Irradiance Distributions Next, we analyzed the light distribution of a hybrid CPV module with a pyramidal SOE (θ = 30◦, h = 15Next, mm). we Figureanalyzed4 shows the light the distribution simulation of results a hybrid for CPV the six module silicon with solar a pyramidal cells using SOE the ( layoutθ = 30°, fromh = 15 Figure mm). 1Figurea. Since 4 shows the hybrid the simulation CPV module results is symmetric, for the six we silicon only solar needed cells to using identify the six layout of the from 12 2 2 2 2 siliconFigure solar1(a). cells.Since Thethe hybrid irradiances CPV on module the Si #7–12is symmetric, were 10.8 we W /onlym , 13.8needed W/m to, 11.0identify W/m six, 9.2of Wthe/m 12, 2 2 9.2silicon W/m solar, and cells. 9.3 WThe/m irradiances, respectively. on the We analyzedSi #7–12 were the irradiances 10.8 W/m2, and 13.8 irradiance W/m2, 11.0 distributions W/m2, 9.2 W/m using2, the9.2 W/m effective2, and wavelength 9.3 W/m2, range.respectively. Although We singleanalyzed crystal the siliconirradiances modules and areirradiance more effi distributionscient than smaller using polycrystallinethe effective wavelength silicon solar range. cells, weAlthough simulated single polycrystalline crystal silicon silicon modules solar cellsare more to better efficient match than the experimentalsmaller polycrystalline measurements silicon to solar follow. cells, Figure we 4simulatedb shows that,polycrystalline although the silicon silicon solar solar cells cell to placed better nextmatch to the the experimental SOE was affected measurements by the shadow to follow. of the Figu SOE,re that 4b shows solar cell that, still although collected the the silicon highest solar power cell 2 (13.9placed W next/m ). to In the addition, SOE was the affected silicon solarby the cells shadow placed of awaythe SOE, from that the solar CPV cell also still showed collected irradiance the highest of at 2 leastpower 9.2 (13.9 W/m W/mby absorbing2). In addition, diffuse the light. silicon Considering solar cells that placed hybrid away CPV from modules the CPV are constructed also showed as arrays,irradiance we usedof at least a square 9.2 W/m Fresnel2 by lens, absorbing which diffuse seems tolight. have Considering affected the that irradiance hybrid distribution.CPV modules are constructed as arrays, we used a square Fresnel lens, which seems to have affected the irradiance distribution. Appl. Sci. 2020, 10, 2051 5 of 9 Appl. Sci. 2020, 10, 2051 5 of 9

Figure 4. IrradianceIrradiance distributions distributions of of the the six six silic siliconon solar solar cells cells for for a apyramidal pyramidal SOE SOE with with θ θ= =30°30 and◦ and h =h 15= 15mm: mm: (a) (Sia) #1, Si #1, (b) ( bSi) #2, Si #2, (c) (Sic) #3, Si #3, (d) ( dSi) #4, Si #4, (e) ( Sie) #5, Si #5, and and (f) (Sif) Si#6. #6. The The size size of each of each cell cell was was 80 80× 55 2 × mm55 mm2. .

To identify the irradiance distribution of the TJ and silicon solar cells according to the design of the SOE, we examined the three cases shown in Figu Figurere5 5.. InIn thethe perpendicularperpendicular SOE,SOE, thethe concentratedconcentrated 2 power on the TJ solar cell was the smallest (326.8 W/m W/m2),), while while the the irradiance irradiance on the the silicon silicon solar solar cells cells 2 was the largest (135.8 WW/m/m2).). Without Without an an SOE, SOE, the the irradiance irradiance on on the the TJ TJ and silicon solar cells were 2 2 590.8 W/m W/m2 andand 131.2 131.2 W/m W/m2,, respectively. respectively. When When using using the the pyra pyramidalmidal SOE SOE optimized above, we saw 2 2 the largest power (606.2 W/m W/m2)) concentrated concentrated on the TJ solar solar cell, cell, and and the the smallest smallest power power (127.4 (127.4 W/m W/m2)) on the silicon solar cell. The solar cell conversion efficiency eﬃciency relation is:

PMAX ηη== × 100 (%)100 (%) (1)(1) Pinput × where η is the conversion efficiency, PMAX is the maximum output power, and Pinput is the input optical where η is the conversion efficiency, PMAX is the maximum output power, and Pinput is the input optical power. Therefore, Therefore, assuming a a TJ solar cell conver conversionsion efficiency efficiency of 38.5% when using a pyramidal SOE, we calculate PMAX as 233.4 W/m22. In addition, assuming an 18% conversion efficiency for the SOE, we calculate PMAX as 233.4 W/m . In addition, assuming an 18% conversion efficiency for the 2 polycrystalline silicon solar cell, we expect to generate 22.9 WW/m/m2 of electric power. Using the same 2 2 assumption, we predict maximum output powers of 125.8 WW/m/m2 and 24.4 W/m W/m2 forfor the the perpendicular perpendicular 2 2 SOE, andand maximummaximum output output powers powers of 227.5of 227.5 W/m W/m2 and 23.6and W23.6/m 2W/mfor the for no the SOE no case. SOE We case. will compareWe will comparethese predictions these predictions with experimental with experiment measurementsal measurements in the next in section. the next section. Appl. Sci. 2020, 10, 2051 6 of 9 Appl. Sci. 2020, 10, 2051 6 of 9

Figure 5. IrradianceIrradiance distributions distributions of of the the triple-junct triple-junctionion and and 12 12 silicon silicon solar solar cells cells with with three three SOE configurations:conﬁgurations: (a (a,b,b) )perpendicular perpendicular (θ (θ = =0°,0 ◦h, = h 15= 15mm), mm), (c,d ()c ,nod) noSOE SOE (h = (h 0),= and0), and(e,f) (pyramidale,f) pyramidal (θ = 30°,(θ = h30 =◦ 15, h mm).= 15 mm).

4. Experimental Measurements 4. Experimental Measurements To supplement the simulation results with experimental measurements, we constructed the hybridTo CPVsupplement module the analyzed simulation above, results and show with the experimental results in Figure measurements,6 and Table we3. Weconstructed used a solar the hybridsimulator CPV (WACOM, module analyzed WXS-450S-65), above, xenon and show lamp the (6500 results W), Fresnelin Figure lens 6 and (PMMA, Table 254 3. We254 used3 a mm solar3), × × simulatorGaInP/GaInAs (WACOM,/Ge triple-junction WXS-450S-65), solar xenon cell (Spectrolab, lamp (6500 CDO-100-C3MJ,W), Fresnel lens 10(PMMA,10 mm 2542), × polycrystalline 254 × 3 mm3), × GaInP/GaInAs/Gesilicon solar cells (Zhiwang, triple-junction 80 55 solar mm cell2), and (Spectrolab, a heat sink CDO-100-C3MJ, in the setup for 10 this × 10 measurement. mm2), polycrystalline Figure6g × siliconshows solar the I-V cells and (Zhiwang, P-V curves 80 × of55 themm TJ2), solarand a cellsheat sink and in the the hybrid setup CPVfor this module, measurement. demonstrating Figure 6(g)that shows the pyramidal the I-V and SOE P-V had curves the highest of the maximumTJ solar cells output and the power hybrid of 212.8 CPV Wmodule,/m2. The demonstrating hybrid CPV thatmodule the pyramidal without an SOE SOE had had the the highest next highest maximum power output (181.5 power W/m2 ),of followed 212.8 W/m by2 the. The perpendicular hybrid CPV 2 2 moduleSOE (149.0 without W/m an). TheSOE perpendicular had the next highest SOE had power the lowest (181.5P W/mMAX due), followed to its narrow by the entrance, perpendicular but it SOEwas expected(149.0 W/m to2 act). The as aperpendicular homogenizer SOE well. had Although the lowest these PMAX values due areto its slightly narrow lower entrance, than thebut resultsit was expectedcalculated to above, act as theya homogenizer show a similar well. relationship. Although these The I-Vvalues and are P-V slightly curves lower for the than polycrystalline the results calculatedsilicon solar above, cells inthey Figure show6h demonstratea similar relationship. that the sum The of I-V maximum and P-V output curves power for the for polycrystalline the pyramidal siliconSOE was solar the cells lowest in Figure at 5.14 6h W demonstrate/m2. The perpendicular that the sum of SOE maximum and the output no SOE power cases for attained the pyramidal the sum SOEof maximum was the lowest output at power 5.14 W/m values2. The of perpendicular 5.67 W/m2 and SOE 5.99 and W/ mthe2, no respectively. SOE cases attained From Figure the sum4, the of 2 2 MAX maximumPMAX of the output polycrystalline power values silicon of 5.67 solar W/m cells and with 5.99 an W/m 18%, conversionrespectively. eﬃ Fromciency Figure were 4, found the P to beof the1.98 polycrystalline W/m2, 2.5 W/m silicon2, 1.96 solar W/m cells2, 1.66 with W/ man2 ,18% 1.67 conversion W/m2, and efficiency 1.67 W/m were2, respectively. found to be On 1.98 the W/m other2, 2 2 2 2 2 2 2 2 2 2.5hand, W/m the, experimental1.96 W/m , 1.66 measurements W/m , 1.67 obtainedW/m , andPMAX 1.67of W/m 0.42 W, respectively./m , 0.76 W/m On, 0.4 the W other/m , 0.33hand, W /them , experimental measurements obtained PMAX of 0.42 W/m2, 0.76 W/m2, 0.4 W/m2, 0.33 W/m2, 0.32 W/m2, and 0.33 W/m2, respectively, which were lower than expected. These values are lower than the Appl. Sci. 2020, 10, 2051 7 of 9

0.32Appl. W Sci./m 22020, and, 10, 0.332051 W/m2, respectively, which were lower than expected. These values are7 lowerof 9 than the simulation results, due to the lower proportion of diﬀuse light in the measurements using the simulation results, due to the lower proportion of diffuse light in the measurements using the solar solar simulator. Since the measured output power values were smaller than those predicted by the simulator. Since the measured output power values were smaller than those predicted by the simulation, we need to include diﬀuse light in the experimental measurement setup for modest-DNI simulation, we need to include diffuse light in the experimental measurement setup for modest-DNI conditions in future tests. Overall, the experimental measurements show similar results to those we conditions in future tests. Overall, the experimental measurements show similar results to those we calculatedcalculated using using the the raytracing raytracing simulations,simulations, andand the hybrid CPV CPV module module with with the the optimized optimized SOE SOE designdesign performed performed better better than than the the other other casescases wewe tested.tested.

FigureFigure 6. 6. PhotographsPhotographs of of experiment experimentalal measurement measurement setup: setup: (a) the ( asolar) the simulator solar simulator (WACOM, (WACOM, WXS- WXS-450S-65),450S-65), (b) the (b) CPV the CPVmodule module with a with Fresnel a Fresnel lens and lens a heat and sink, a heat (c) the sink, perpendicular (c) the perpendicular SOE (θ = 0°, SOE h (θ = 150◦ ,mm), h = 15 (d) mm), the pyramidal (d) the pyramidal secondary secondary optical element optical (SOE) element (θ = (SOE)30°, h = ( θ15= mm),30◦, h(e=) the15 mm),laminated (e) the laminatedtriple-junction triple-junction solar cell solar with cellbypass with diode, bypass and diode, (f) the and polycrystalline (f) the polycrystalline silicon solar silicon cell. solarAlso cell.shown Also shownare I-V are and I-V P-V and curves P-V curves of the of (g the) triple-junction (g) triple-junction solar cell solar and cell (h and) silicon (h) siliconsolar cells solar for cells the forthree the SOE three SOEconfigurations. conﬁgurations.

Table 3. Device characteristics of the triple-junction and 12 silicon solar cells of the hybrid CPV module, Table 3. Device characteristics of the triple-junction and 12 silicon solar cells of the hybrid CPV usingmodule, three using SOE conﬁgurations.three SOE configurations. 2 2 SOESOE Conﬁguration Configuration Cell Cell PMAX PMAX(W (W/m/m ) 2) VVOCOC (V)(V) ISC ISC (A)(A) Fill Fill facto factorr AreaArea (cm (cm2) ) PerpendicularPerpendicular Triple-junctionTriple-junction 149.0149.0 3.04 3.04 3.78 3.78 83.61 83.61 645.2 645.2 (θ =(0θ◦ ,= h 0°,= 15h = mm) 15 mm) siliconsilicon 5.67 5.67 (sum (sum of of #1–12) #1–12) ------645.2 645.2 Triple-junctionTriple-junction 181.5181.5 3.03 3.03 4.73 4.73 81.8 81.8 645.2 645.2 No SOENo SOE siliconsilicon 5.99 5.99 (sum (sum of of #1–12) #1–12) ------645.2 645.2 PyramidalPyramidal Triple-junctionTriple-junction 212.8212.8 3.04 3.04 5.57 5.57 81.16 81.16 645.2 645.2 (θ =(30θ ◦=, 30°, h = 15h = mm) 15 mm) siliconsilicon 5.14 5.14 (sum (sum of of #1–12) #1–12) ------645.2 645.2

5. Conclusions5. Conclusions WeWe demonstrated demonstrated a a hybrid hybrid CPV CPV modulemodule thatthat combines a a TJ TJ solar solar cell cell and and low-cost low-cost silicon silicon solar solar cellscells by by comparing comparing various various SOE SOE designs. designs. When When we we varied varied the the angle angle of of the the SOE SOE between between 0◦ 0°and and 50 50°,◦, our our analysis showed an optimized irradiance at 30° on the TJ solar cell in the effective wavelength analysis showed an optimized irradiance at 30◦ on the TJ solar cell in the effective wavelength range. Higherrange. SOE Higher angles SOE increased angles increased the irradiance the irradiance within within the lossy the wavelengthlossy wavelength range range more more than than the power the withinpower the within effective the wavelengtheffective wavelength range, which range, decreased which decreased the performance the performa of thence hybridof the hybrid CPV module. CPV module. Likewise, taller SOEs also caused larger ETJ,loss/ETJ,eff ratios. We concluded that an SOE with Likewise, taller SOEs also caused larger ETJ,loss/ETJ,eff ratios. We concluded that an SOE with an angle of an angle of 30° and a height of 15 mm was the most desirable, with the irradiances of 626.2 W/m2 and Appl. Sci. 2020, 10, 2051 8 of 9

2 2 30◦ and a height of 15 mm was the most desirable, with the irradiances of 626.2 W/m and 127.4 W/m in the effective wavelength range, and a ETJ,loss/ETJ,eff ratio of 5.7%. Next, we compared three SOE configurations for the hybrid CPV module by using simulations and experimental measurements. The perpendicular SOE and the no SOE case generated maximum output powers of 149.0 W/m2 and 181.5 W/m2, respectively, in the TJ solar cell of the fabricated hybrid CPV module. The optimized pyramidal SOE, however, showed a higher maximum output power of 212.8 W/m2 in the TJ solar cell. The hybrid CPV module has potential applications in next-generation power production systems and, therefore, requires more research in the future.

Author Contributions: Conceptualization, W.-L.J.; software, W.-L.J.; writing – original draft, W.-L.J.; writing – review & editing, W.-L.J.; experiments, K.-P.K.; visualization, J.-H.M.; validation, J.-Y.L., S.-H.M., J.-H.P. and J.-H.H.; resources, W.-K.P. and S.Y.; supervision, D.-S.L. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the Korea Institute of Energy Technology Evaluation and Planning (KETEP), the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea, grant number 20183010014310; the GIST Research Institute (GRI), no grant number. Acknowledgments: The authors would like to thank Eddie Chung at Hanul Solar Energy for his technical support. This work was supported by the Korea Institute of Energy Technology Evaluation and Planning the Ministry of Trade, Industry & Energy of the Republic of Korea (No. 20183010014310), and the GIST Research Institute (GRI) grant funded by the GIST in 2020. Conﬂicts of Interest: The authors declare no conﬂicts of interest.

References

1. McDaniels, D.K.; Lowndes, D.H.; Mathew, H.; Reynolds, J.; Gray, R. Enhanced solar energy collection using reflector-solar thermal collector combinations. Sol. Energy 1975, 17, 277–283. [CrossRef] 2. Baker, S.; McDaniels, D.K.; Kaehn, H.D.; Lowndes, D.H. Time integrated calculation of the insolation collected by a reflector-collector system. Sol. Energy 1978, 20, 415–417. [CrossRef] 3. Armenta, C.; Vorobieff, P.; Mammoli, A. Summer off-peak performance enhancement for rows of fixed solar thermal collectors using flat reflective surfaces. Sol. Energy 2011, 85, 2041–2052. [CrossRef] 4. Sharaf, O.Z.; Orhan, M.F. Concentrated photovoltaic thermal (CPVT) solar collector systems: Part I—Fundamentals, design considerations and current technologies. Renew. Sustain. Energy Rev. 2015, 50, 1500–1565. [CrossRef] 5. Herrero, R.; Victoria, M.; Domínguez, C.; Askins, S.; Antón, I.; Sala, G. Concentration photovoltaic optical system irradiance distribution measurements and its effect on multi-junction solar cells. Prog. Photovolt. Res. Appl. 2012, 20, 423–430. [CrossRef] 6. Gordon, J.M.; Feuermann, D.; Young, P. Unfolded aplanats for high-concentration photovoltaics. Opt. Lett. 2008, 33, 1114–1116. [CrossRef] 7. Sansoni, P.; Fontani, D.; Francini, F.; Jafrancesco, D.; Pierucci, G.; Lucia, M.D. Technique for outdoor test on concentrating photovoltaic cells. Int. J. Photoenergy 2015, 2015, 1–9. [CrossRef] 8. Sun, Y.; Wang, Y.; Zhu, L.; Yin, B.; Xiang, H.; Huang, Q. Direct liquid-immersion cooling of concentrator silicon solar cells in a linear concentrating photovoltaic receiver. Energy 2014, 65, 264–271. [CrossRef] 9. Narasimman, K.; Selvarasan, I. Design construction and analysis of solar ridge concentrator photovoltaic (PV) system to improve battery charging performance. Ecotoxicol. Environ. Saf. 2016, 127, 187–192. [CrossRef] [PubMed] 10. Wei, A.; Chen, Z.; Sze, J. Planar solar concentrator with a v-groove array for a side-absorption concentrated photovoltaic system. Optik 2016, 127, 10858–10867. [CrossRef] 11. Chen, Y.; Chiang, H. Design of the secondary optical elements for concentrated photovoltaic units with Fresnel Lenses. Appl. Sci. 2015, 5, 770–786. [CrossRef] 12. Nakatani, M.; Yamada, N. Characterization of core-shell spherical lens for microtracking concentrator photovoltaic system. Energies 2019, 12, 3517. [CrossRef] 13. Yamada, N.; Hirai, D. Maximization of conversion efficiency based on global normal irradiance using hybrid concentrator photovoltaic architecture. Prog. Photovolt. Res. Appl. 2016, 24, 846–854. [CrossRef] Appl. Sci. 2020, 10, 2051 9 of 9

14. Lee, K.T.; Yao, Y.; He, J.; Fisher, B.; Sheng, X.; Lumb, M.; Xu, L.; Anderson, M.A.; Scheiman, D.; Han, S.; et al. Concentrator photovoltaic module architectures with capabilities for capture and conversion of full global solar radiation. Proc. Natl. Acad. Sci. USA 2016, 113, E8210–E8218. [CrossRef][PubMed] 15. Yamada, N.; Okamoto, K. Experimental measurements of a prototype high concentration Fresnel lens CPV module for the harvesting of diffuse solar radiation. Opt. Express 2014, 22, A28–A34. [CrossRef][PubMed] 16. Araki, K.; Ota, Y.; Lee, K.; Nishioka, K.; Yamaguchi, M. Is It CPV? Yes, But It Is a Partial CPV. In Proceedings of the 13th Conference on Concentrator Photovoltaic Systems, Ottawa, ON, Canada, 1–3 May 2017; Volume 1881, p. 080001. 17. Kreske, K. Optical design of a solar flux homogenizer for concentrator photovoltaics. Appl. Opt. 2002, 41, 2053–2058. [CrossRef][PubMed] 18. Fontani, D.; Sansoni, P.; Francini, F.; DeLucia, M.; Pierucci, G.; Jafrancesco, D. Optical tests on a curve Fresnel Lens as secondary optics for solar troughs. Int. J. Photoenergy 2017, 2017, 1–11. [CrossRef] 19. Benítez, P.; Miñano, J.C.; Zamora, P.; Mohedano, R.; Cvetkovic, A.; Buljan, M.; Chaves, J.; Hernández, M. High performance Fresnel-based photovoltaic concentrator. Opt. Express 2010, 18, A25–A40. [CrossRef] 20. Cotal, H.; Fetzer, C.; Boisvert, J.; Kinsey, G.; King, R.; Hebert, P.; Yoon, H.; Karam, N. III–V multijunction solar cells for concentrating photovoltaics. Energy Environ. Sci. 2009, 2, 174–192. [CrossRef] 21. Zhang, L.; Niu, P.; Li, Y.; Song, M.; Zhang, J.; Ning, P.; Chen, P. Investigation on high-efficiency

Ga0.51In0.49P/In0.01Ga0.99As/Ge triple-junction solar cells for space applications. AIP Adv. 2017, 7, 125217. [CrossRef] 22. Hayashi, N.; Inoue, D.; Matsumoto, M.; Matsushita, A.; Higuchi, H.; Aya, Y.; Nakagawa, T. High-eﬃciency thin and compact concentrator photovoltaics with micro-solar cells directly attached to a lens array. Opt. Express 2015, 23, A594–A603. [CrossRef][PubMed]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).