Improved Color Purity of Monolithic Full Color Micro-Leds Using Distributed Bragg Reﬂector and Blue Light Absorption Material

coatings

Article Improved Color Purity of Monolithic Full Color Micro-LEDs Using Distributed Bragg Reﬂector and Blue Light Absorption Material

Shao-Yu Chu 1, Hung-Yu Wang 1, Ching-Ting Lee 1,2, Hsin-Ying Lee 1,* , Kai-Ling Laing 3, Wei-Hung Kuo 3, Yen-Hsiang Fang 3 and Chien-Chung Lin 3,4 1 Department of Photonics, National Cheng Kung University, Tainan 701, Taiwan; [email protected] (S.-Y.C.); [email protected] (H.-Y.W.); [email protected] (C.-T.L.) 2 Department of Electrical Engineering, Yuan Ze University, Taoyuan 320, Taiwan 3 Electronic and Optoelectronic System Research Laboratories, Industrial Technology Research Institute, Hsinchu 310, Taiwan; [email protected] (K.-L.L.); [email protected] (W.-H.K.); [email protected] (Y.-H.F.); [email protected] (C.-C.L.) 4 Institute of Photonic System, National Chiao Tung University, Tainan 711, Taiwan * Correspondence: [email protected]; Tel.: +886-6-2082368

Received: 26 February 2020; Accepted: 28 April 2020; Published: 29 April 2020

Abstract: In this study, CdSe/ZnS core-shell quantum dots (QDs) with various dimensions were used as the color conversion materials. QDs with dimensions of 3 nm and 5 nm were excited by gallium nitride (GaN)-based blue micro-light-emitting diodes (micro-LEDs) with a size of 30 µm 30 µm to × respectively form the green and red lights. The hybrid Bragg reflector (HBR) with high reflectivity at the regions of the blue, green, and red lights was fabricated on the bottom side of the micro-LEDs to reflect the downward light. This could enhance the intensity of the green and red lights for the green and red QDs/micro-LEDs to 11% and 10%. The distributed Bragg reflector (DBR) was fabricated on the QDs color conversion layers to reflect the non-absorbed blue light that was not absorbed by the QDs, which could increase the probability of the QDs excited by the reflected blue light. The blue light absorption material was deposited on the DBR to absorb the blue light that escaped from the DBR, which could enhance the color purity of the resulting green and red QDs/micro-LEDs to 90.9% and 90.3%, respectively.

Keywords: blue light absorption material; color conversion layer; distributed Bragg reﬂector; hybrid Bragg reﬂector; micro-light-emitting diodes; quantum dots

1. Introduction In recent years, display screens have been widely used in daily life, and liquid crystal display (LCD) and organic light-emitting diode (OLED) displays are the mainstream of display technology [1]. However, breakthrough progress has been made in semiconductor manufacturing technology, leading to the abrupt rise of micro-light-emitting diodes (micro-LEDs). In particular, the micro-LEDs have advantages of long lifetime, high luminous efficiency, and smaller volume [2–4]. On the other hand, micro-LEDs can effectively reduce energy consumption and improve pixel characteristics, which can be expected to become the mainstream of next-generation display technology. In general, each pixel of the full color micro-LED display is constructed by red, green, and blue light sources; then different light source colors can be obtained by controlling the ratio of the three primary colors. However, this method includes disadvantages such as different lifetime and complex driver circuits. To overcome the above-mentioned problem, the blue light emitted from the gallium nitride (GaN)-based micro-LEDs was used to excite the quantum dots (QDs) with various dimensions to form

Coatings 2020, 10, 436; doi:10.3390/coatings10050436 www.mdpi.com/journal/coatings Coatings 2020, 10, 436 2 of 9 the monolithic red and green QDs/micro-LEDs [5–7]. In this work, the QDs were filled in the regions around the black photoresist with very low transmittance in the visible light region. The function of the black photoresist is to prevent the light emitted from the sidewall of the micro-LEDs to reduce the crosstalk between the micro-LEDs. However, since the excitation efficiency of the red and green QDs was still low, a large amount of non-absorbed blue light would respectively blend with the green light and red light and be simultaneously emitted from the green and red QDs/micro-LEDs. Consequently, the color purity of red and green QDs/micro-LEDs would be not good. The distributed Bragg reflector (DBR) with high reflectivity in the blue light region was deposited on the QD color conversion layer to solve the problem of the non-absorbed blue light [8,9]. However, since the DBR reflectivity depends on the incident angle of the light, the DBR reflectivity will be reduced if the light is not incident perpendicularly into the DBR structure [10,11]. Consequently, in this study, the blue light absorption layer was utilized and deposited on the DBR to absorb the escaped blue light from the DBR, which could improve the color purity of the monolithic full color micro-LEDs. To further enhance the output light intensity, a hybrid Bragg reflector (HBR) was deposited on the bottom side of the micro-LEDs to reflect the downward propagation of red, green, and blue lights.

2. Materials and Methods

2.1. Materials In this work, the epitaxial wafers of the GaN-based blue micro-LEDs were supported by Epistar Co., Hsinchu, Taiwan. CdSe/ZnS core-shell quantum dots with an average dimension of 5 nm and 3 nm were purchased from Taiwan Nanocrystal Inc., Tainan, Taiwan. The granules of titanium dioxide (TiO2) (99.9%) and silicon dioxide (SiO2) (99.999%) were purchased from Admat Inc., Pennsylvania, PA, USA. The black photoresist (model: ABK408X) was supported by Daxin Materials Co., Taichung, Taiwan. The blue light absorption material (model: Eusorb UV-1995) was supported by Eutec Chemical Co., Taipei, Taiwan.

2.2. Experimental Procedure Figure1a illustrates the schematic configuration of the monolithic full color micro-LEDs. A metal organic chemical vapor deposition system was used to epitaxy the GaN-based blue micro-LEDs structure on c-plane sapphire substrates. The structure of the micro-LEDs was constructed by a 2.8 µm-thick GaN buffer layer, a 4 µm-thick n-GaN layer, an undoped InGaN/GaN (3/7 nm, 10 pairs) multiple quantum wells (MQW) active layer, a 33 nm-thick p-AlGaN layer, and a 150 nm-thick p-GaN layer. A 500 nm-thick Ni metal was deposited on the p-GaN layer as the metal mask to protect the mesa (30 µm 30 µm) of the micro-LEDs through an electron-beam evaporator. After the mesa dry etching, × the remaining Ni metal was removed by using aqua regia. The Ti/Al/Pt/Au (25/100/50/150 nm) metals were deposited on the patterned n-GaN layer by an electron-beam evaporator and then annealed in a pure N2 environment at 850 ◦C for 2 min using a rapid temperature annealing (RTA) system to obtain n-electrode ohmic contact [12]. Afterward, the thin Ni/Au (3/3 nm) metals were deposited on the p-GaN surface treated by (NH4)2Sx solution (S = 6%) for 30 min [13] as the current spreading layer, and the Ni/Au (20/100 nm) metals were deposited as the p-electrode. To obtain the p-electrode ohmic contact, the samples were annealed in an air environment at 500 ◦C for 10 min using the RTA system. In this work, an extended p-electrode pattern was designed to facilitate electrical measurement of the micro-LEDs (as shown in Figure1b). The 2.5 µm-thick black photoresist was spun on the samples and patterned between the monolithic full color micro-LEDs through a photolithography method. The black photoresist was solidified through the RTA system in a N2 environment at 150 ◦C for 10 min. Since black photoresist had a low average transmittance of 0.56% in the visible light region, it could prevent crosstalk among the full color micro-LEDs. Subsequently, the HBR composed of m-pairs (TiO2/SiO2) (48/77 nm) and a Ag (200 nm) metal reflector was deposited on the bottom side of the samples through the electron beam evaporator. The red (green) CdSe/ZnS core-shell QDs with Coatings 2020, 10, 436 3 of 9 an average dimension of 5 nm (3 nm) and the poly(methyl methacrylate) (PMMA) were added into Coatings 2020, 10, x FOR PEER REVIEW 3 of 9 toluene, and then the red (green) QD slurry was filled into the openings in the black photoresist layer to form the red (green) QD color conversion layer. Subsequently, the electron beam evaporator was photoresist layer to form the red (green)n-pairs QD color conversion layer. Subsequently, the electron beam used to alternately deposit (TiO2/SiO2) (48/77 nm)/TiO2 (48 nm) on the red (green) QDs at 100 ◦C evaporator was used to alternately deposit (TiO2/SiO2)n-pairs (48/77 nm)/TiO2 (48 nm) on the red (green) as the DBR structure. Finally, a 100 nm-thick blue light absorption layer was formed on the DBR QDs at 100 °C as the DBR structure. Finally, a 100 nm-thick blue light absorption layer was formed by spin-coating. on the DBR by spin-coating.

(a) (b)

FigureFigure 1.1. (a) TheThe schematicschematic conﬁgurationconfiguration ofof thethe monolithicmonolithic fullfull colorcolor micro-light-emittingmicro-light-emitting diodesdiodes (micro-LEDs).(micro-LEDs). ((bb)) TopTop viewview photographphotograph ofof thethe blueblue micro-LEDs.micro-LEDs.

3.3. Results and Discussion Figure2 a2a and and the the inset inset figure figure show show the electroluminescencethe electroluminescence (EL) spectrum(EL) spectrum and the and current–voltage the current– (voltageI–V) characteristics (I–V) characteristics of the blue of the micro-LEDs, blue micro-LEDs, respectively. respectively. In this work, In this an work, EL measurement an EL measurement system wassystem installed was installed in a black in box a black and equippedbox and equipped with an Agilentwith an 4156C Agilent semiconductor 4156C semiconductor parameter parameter analyzer, ananalyzer, optical an fiber, optical and fiber, a spectrometer and a spectrometer (Ocean Optics (Ocean USB2000, Optics USB2000, Florida, FL,Florida, USA) FL, system. USA) system. In the EL In measurementthe EL measurement process, process, the resulting the resulting samples samples were placed were onplaced the stage on the in stage the black in the box black and box the and optical the fiberoptical was fiber fixed was above fixed theabove micro-LEDs the micro-LEDs with a with distance a distance of 1 mm of 1to mm catch to catch the light the light emitted emitted from from the micro-LEDs.the micro-LEDs. The The maximum maximum EL intensityEL intensity of all of the all micro-LEDsthe micro-LEDs at the at appliedthe applied current current of 1.2 of mA 1.2 wasmA measuredwas measured by adjusting by adjusting the sample the sample stage. Thestage. emission The emission wavelength wavelength and the and threshold the threshold voltage ofvoltage the blue of micro-LEDsthe blue micro-LEDs was 451 nm was and 451 2.5 nm V, and respectively. 2.5 V, respec Thetively. emission The wavelengths emission wavelengths of 625 and 550of 625 nm and for the550 rednm andfor the green red QDs and excitedgreen QDs by a excited tunable by laser a tunable with a wavelengthlaser with a ofwavelength 451 nm were of 451 obtained, nm were respectively obtained, (Figurerespectively2b). The (Figure full width 2b). The at halffull maximumwidth at half (FWHM) maximum of the (FWHM) blue, green, of the and blue, red green, lights and emitted red fromlights theemitted blue micro-LEDs,from the blue red micro-LEDs, and green QD red micro-LEDs and green wasQD 27.0,micro-LEDs 32.2, and was 34.5 27.0, nm, respectively.32.2, and 34.5 nm, respectively.Generally, the QD density in the color conversion layer would affect the color conversion efficiency of the QDs. Consequently, the various weight ratios (10:20, 10:10, 20:10, and 40:10 mg, hereafter referred to as 1:2, 1:1, 2:1, and 4:1) of the greenBlue and micro-LEDs red QDs:PMMA were added into toluene (1 mL) to form

the QD slurry with various QD densities.20 Figure3a,b presentgreen QDs the emissionred QDs spectra of various green QDs:PMMA ratios and red QDs:PMMA15 ratios excited at the tunable laser with a wavelength of 451 nm.

As shown in Figure3, the emission10 intensity of the green QDs:PMMA = 1:1 and red QDs:PMMA 5 Current (mA) Current

= 1:1 was larger than that of the other0 green and red QDs:PMMA ratios. This phenomenon can be 012345 attributed to the excessive QDs in the QDsVoltage slurry (V) with QDs:PMMA ratios of 2:1 and 4:1, which would reabsorb the converted green and red lights, reducing the output intensity of the converted green and red lights [14]. In contrast, the fewer QDs in the QD slurry with the QDs:PMMA ratio of 1:2 would lead to the absorbed amount of the blue light being too low. Consequently, the emission intensity of the QD slurry with QDs:PMMA ratio of 1:2 was smaller in comparison with the QD slurry with the Emission Intensity (a.u.) Intensity Emission Emission Emission (a.u.) Intensity QDs:PMMA ratio of 1:1. 400 500 600 700 800 400 500 600 700 800 Wavelength (nm) Wavelength (nm) (a) (b) Figure 2. (a) The electroluminescence (EL) spectra of the GaN-based blue micro-LEDs operated at a bias of 3 V. The inset figure shows the current–voltage characteristics. (b) The emission spectrum of the green and red QDs.

Coatings 2020, 10, x FOR PEER REVIEW 3 of 9 photoresist layer to form the red (green) QD color conversion layer. Subsequently, the electron beam evaporator was used to alternately deposit (TiO2/SiO2)n-pairs (48/77 nm)/TiO2 (48 nm) on the red (green) QDs at 100 °C as the DBR structure. Finally, a 100 nm-thick blue light absorption layer was formed on the DBR by spin-coating.

(a) (b)

Figure 1. (a) The schematic configuration of the monolithic full color micro-light-emitting diodes (micro-LEDs). (b) Top view photograph of the blue micro-LEDs.

3. Results and Discussion Figure 2a and the inset figure show the electroluminescence (EL) spectrum and the current– voltage (I–V) characteristics of the blue micro-LEDs, respectively. In this work, an EL measurement system was installed in a black box and equipped with an Agilent 4156C semiconductor parameter analyzer, an optical fiber, and a spectrometer (Ocean Optics USB2000, Florida, FL, USA) system. In the EL measurement process, the resulting samples were placed on the stage in the black box and the optical fiber was fixed above the micro-LEDs with a distance of 1 mm to catch the light emitted from the micro-LEDs. The maximum EL intensity of all the micro-LEDs at the applied current of 1.2 mA was measured by adjusting the sample stage. The emission wavelength and the threshold voltage of the blue micro-LEDs was 451 nm and 2.5 V, respectively. The emission wavelengths of 625 and 550 nm for the red and green QDs excited by a tunable laser with a wavelength of 451 nm were obtained, respectively (Figure 2b). The full width at half maximum (FWHM) of the blue, green, and red lights emittedCoatings from2020 the, 10 ,blue 436 micro-LEDs, red and green QD micro-LEDs was 27.0, 32.2, and 34.54 of 9nm, respectively.

Blue micro-LEDs

20 green QDs red QDs 15

Coatings 2020, 10, x FOR PEER REVIEW 10 4 of 9

5 Current (mA) Current

0 Generally, the QD density 012345in the color conversion layer would affect the color conversion Voltage (V)

efficiency of the QDs. Consequently, the various weight ratios (10:20, 10:10, 20:10, and 40:10 mg, hereafter referred to as 1:2, 1:1, 2:1, and 4:1) of the green and red QDs:PMMA were added into toluene (1 mL) to form the QD slurry with various QD densities. Figure 3a,b present the emission spectra of various green QDs:PMMA ratios and red QDs:PMMA ratios excited at the tunable laser with a wavelength of 451 nm. As shown in Figure 3, the emission intensity of the green QDs:PMMA = 1:1 Emission Intensity (a.u.) Intensity Emission Emission Emission (a.u.) Intensity and red QDs:PMMA = 1:1 was larger than that of the other green and red QDs:PMMA ratios. This phenomenon can be attributed400 500to the 600 excessive 700 800QDs in 400the QDs 500 slurry 600 with 700 QDs:PMMA 800 ratios of 2:1 and 4:1, which would reabsorbWavelength the converted (nm) green andWavelength red lights, reducing (nm) the output intensity of the converted green and red lights(a) [14]. In contrast, the fewer(b) QDs in the QD slurry with the QDs:PMMA ratio of 1:2 would lead to the absorbed amount of the blue light being too low. Figure 2. (a) The electroluminescence (EL) spectra of the GaN-based blue micro-LEDs operated at a FigureConsequently, 2. (a) The the electroluminescence emission intensity (EL) of the spectra QD slurry of the withGaN-based QDs:PMMA blue micro-LEDs ratio of 1:2 operatedwas smaller at a in bias of 3 V. The inset ﬁgure shows the current–voltage characteristics. (b) The emission spectrum of the biascomparison of 3 V. The with inset the figure QD slurry shows with the thecurrent–voltage QDs:PMMA ratiocharacteristics. of 1:1. (b) The emission spectrum of green and red QDs. the green and red QDs.

Green QDs:PMMA Red QDs:PMMA 1:2 1:2 1:1 1:1 2:1 2:1 4:1 4:1

Emission Intensity (a.u.) Emission Intensity Emission Intensity (a.u.) Intensity Emission

400 450 500 550 600 650 700 400 450 500 550 600 650 700 Wavelength(nm) Wavelength (nm) (a) (b)

Figure 3.3.The The emission emission spectra spectra of various of various (a) green (a) core-shell green core-shell quantum dots:quantum poly (methyldots: poly methacrylate) (methyl methacrylate)(QDs:PMMA) (QDs:PMMA) ratios and (b) redratios QDs:PMMA and (b) red ratios QDs:PMMA excited byratios a tunable excited laser by a with tunable a wavelength laser with ofa wavelength451 nm. of 451 nm.

ToTo improve improve the the output output light light intensity intensity of ofthe the mono monolithiclithic fullfull color color micro-LEDs, micro-LEDs, the HBR the HBRwith high with reflectivityhigh reflectivity for the for red, the green, red, green, and blue and lights blue lights was deposited was deposited on the on bottom the bottom side of side the of samples. the samples. The The HBR was constructed by (TiO /SiO )m-pairs (48/77 nm) and Ag (200 nm) metal. Figure4 shows HBR was constructed by (TiO2/SiO2 2)m-pairs2 (48/77 nm) and Ag (200 nm) metal. Figure 4 shows the the reflectivity spectra of the Ag metal and (TiO /SiO )m-pairs (m = 1, 2, and 3)/Ag measured using an reflectivity spectra of the Ag metal and (TiO2/SiO2 2)m-pairs2 (m = 1, 2, and 3)/Ag measured using an UV– UV–Visible spectrophotometer. The reflectivity of all of the (TiO /SiO )m-pairs/Ag HBR structures was Visible spectrophotometer. The reflectivity of all of the (TiO2/SiO2)m-pairs2/Ag HBR structures was higher higher than that of the single Ag metal. Furthermore, the highest reflectivity of (TiO /SiO )2-pairs/Ag than that of the single Ag metal. Furthermore, the highest reflectivity of (TiO2/SiO2)2-pairs2/Ag was2 97.5%, 98.9%,was 97.5%, and 98.9%,98.8% andat the 98.8% wavelengths at the wavelengths of 451, 550, of 451, and 550, 625 and nm, 625 respectively. nm, respectively. Consequently, Consequently, the downwardthe downward propagation propagation of the of thered, red, green, green, and and blue blue lights lights could could be be effectively effectively reflected reflected by by the the HBR, HBR, which could improve the output light intensity.

Coatings 2020, 10, x FOR PEER REVIEW 5 of 9 Coatings 2020, 10, x FOR PEER REVIEW 5 of 9 Coatings 2020, 10, 436 5 of 9

100 10090 8090 7080 Ag metal 6070 Agm-pairs metal (TiO2/SiO2) /Ag 5060 m-pairs (TiO2/SiO m2) = 1 /Ag 4050 m m = = 2 1 3040 m m = = 3 2 2030 m = 3 Reflectivity (%) 1020 Reflectivity (%) 100 300 350 400 450 500 550 600 650 700 750 800 0 300 350 400Wavelengh 450 500 550 600(nm) 650 700 750 800

Wavelengh (nm) m-pairs Figure 4. The reflectivity spectra of the Ag metal and (TiO2/SiO2) m-pairs (m = 1, 2, and 3)/Ag. Figure 4. The reflectivity spectra of the Ag metal and (TiO2/SiO2) (m = 1, 2, and 3)/Ag. Figure 4. The reflectivity spectra of the Ag metal and (TiO2/SiO2)m-pairs (m = 1, 2, and 3)/Ag. InIn this this work, work, the the green green and and red red QD QD color color conversion conversion layers layers were were used used to to convert convert the the blue blue light light intointo theIn the thisgreen green work, and and thered red greenlights, lights, and respectively. respectively.red QD color To avoi Toconversion avoidd the red the layers and red andgreenwere green usedlights to lights that convert emitted that the emitted fromblue light fromthe QDs/micro-LEDsintothe QDsthe green/micro-LEDs and blended red blendedlights, with respectively. the with non-absorbed the non-absorbed To avoi blued the light, blue red light,andthe DBRgreen the with DBR lights high with that reflectivity high emitted reflectivity from at the theat wavelengthQDs/micro-LEDsthe wavelength of 451 ofblended nm 451 should nm with should be the deposited benon-absorbed deposited on the on blue thered light, redand and thegreen green DBR QDs QDswith color colorhigh conversion reflectivity conversion layers. at layers. the Simultaneously,wavelengthSimultaneously, of 451the the nm DBR should withwith high highbe transmittancedeposited transmittance on atthe wavelengthsat red wavelengths and green of 550 ofQDs nm 550 andcolor nm 625 andconversion nm 625 was nm requested. layers. was requested. The DBR was composed of (TiOn-pairs2/SiO2)n-pairs (48/77 nm)/TiO2 (48 nm). Figure 5 presents the Simultaneously,The DBR was composed the DBR of with (TiO 2/highSiO2 )transmittance(48/77 nm) at/ TiOwavelengths2 (48 nm). Figureof 5505 presentsnm and the625 reflectivity nm was reflectivity spectra of the (TiOn-pairs2/SiO2)n-pairs (n = 2, 3, n-pairs4, and 5)/TiO2 DBR structure measured using an requested.spectra of theThe (TiODBR2 /wasSiO2 composed) (n = of2, (TiO 3, 4,2 and/SiO 5)2)/TiO2(48/77DBR nm)/TiO structure2 measured(48 nm). Figure using 5 an presents UV–Visible the UV–Visiblereflectivityspectrophotometer. spectrophotometer.spectra of The the number(TiO2 The/SiO of 2number)nn-pairs-pairs (n =of would 2, n -pairs3, 4, obviously and would 5)/TiO obviously aff2ect DBR the structure reflectivityaffect the measured reflectivity of the DBR using ofat the thean DBRUV–Visibleblue at light the range.blue spectrophotometer. light The range. DBR reflectivity The The DBR number reflectivity at the wavelengthof n-pairs at the would wavelength of 451 obviously nm increased of 451affect nm with the increased reflectivity an increase with of in anthe the increaseDBRnumber at inthe of the bluen -pairs.number light Although ofrange. n-pairs. The the Although DBR highest reflectivity the DBR highest reflectivity at theDBR wavelength reflec at thetivity wavelength ofat the451 wavelengthnm of increased 451 nm of was 451with 98.6% nm an wasincreaseas n98.6%= 6, in theas the n DBR number= 6, reflectivitythe ofDBR n-pairs. reflectivity at wavelengths Although at wavelengths the of highest 550 and DBR of 625 550 reflec nm and wastivity 625 31.9% atnm the was and wavelength 31.9% 15.8%, and respectively, of 15.8%,451 nm respectively, which adversely affected the emitting of the green light. The reflectivity5-pairs of the waswhich 98.6% adversely as n = a6,ff ectedthe DBR the emittingreflectivity of theat wavelengths green light. Theof 550 reflectivity and 625 ofnm the was (TiO 31.9%2/SiO 2and) 15.8%,/TiO2 5-pairs (TiOrespectively,DBR2/SiO structure2) which/TiO at the2 DBR adversely wavelengths structure affected at of the 451, wavelengthsthe 550, emitting and 625 of ofnm451, the was 550, green 96.4%, and 625light. 13.7%, nm Thewas and 96.4%,reflectivity 7.6%, 13.7%, respectively. of and the 7.6%, respectively.5-pairs Consequently, the 5-pairs (TiO2/SiO2) were more suitable for the DBR structure (TiOConsequently,2/SiO2) /TiO the2 5-pairsDBR structure (TiO2/SiO at the2) were wavelengths more suitable of 451, for550, the and DBR 625 nm structure was 96.4%, than 13.7%, the 6-pairs and than the 6-pairs (TiO2/SiO2). 7.6%,(TiO2 /respectively.SiO2). Consequently, the 5-pairs (TiO2/SiO2) were more suitable for the DBR structure than the 6-pairs (TiO2/SiO2). 100 10090 n-pairs (TiO2/SiO2) /TiO2 8090 n-pairs (TiO2/SiO 2n) = 2 /TiO2 7080 n n = = 3 2 6070 n n = = 4 3 n = 5 5060 n = 4 n n = = 6 5 4050 n = 6 3040

Reflectivity (%) 2030

Reflectivity (%) 1020 100 300 350 400 450 500 550 600 650 700 750 800 0 300 350 400Wavelengh 450 500 550 600(nm) 650 700 750 800

Wavelenghn-pairs (nm) Figure 5. The reﬂectivity spectra of the (TiO2/SiO2) (n = 2, 3, 4, and 5)/TiO2 distributed Bragg Figure 5. The reflectivity spectra of the (TiO2/SiO2)n-pairs (n = 2, 3, 4, and 5)/TiO2 distributed Bragg reﬂector (DBR) structure. reflectorFigure 5.(DBR) The structure.reflectivity spectra of the (TiO2/SiO2)n-pairs (n = 2, 3, 4, and 5)/TiO2 distributed Bragg reflectorThe EL spectra(DBR) structure. of the resulting blue micro-LEDs and the resulting QDs/micro-LEDs were measured to verify the function of the HBR, DBR and blue light absorption layer. Figure6 shows the EL spectra of

Coatings 2020, 10, x FOR PEER REVIEW 6 of 9 Coatings 2020, 10, x FOR PEER REVIEW 6 of 9 Coatings 2020, 10, 436 6 of 9 The EL spectra of the resulting blue micro-LEDs and the resulting QDs/micro-LEDs were The EL spectra of the resulting blue micro-LEDs and the resulting QDs/micro-LEDs were measured to verify the function of the HBR, DBR and blue light absorption layer. Figure 6 shows the measured to verify the function of the HBR, DBR and blue light absorption layer. Figure 6 shows the theEL resulting spectra of green the andresulting red QDs green/micro-LEDs and red QDs/micr and the blueo-LEDs micro-LEDs and the with blue HBR. micro-LEDs As shown with in FigureHBR. 6As, EL spectra of the resulting green and red QDs/micro-LEDs and the blue micro-LEDs with HBR. As theshown output in intensityFigure 6, of the the output green and intensity red QDs of/micro-LEDs the green and with red HBR QDs/micro-LEDs was respectively with increased HBR 11%was shown in Figure 6, the output intensity of the green and red QDs/micro-LEDs with HBR was andrespectively 10% in comparison increased 11% with and the 10% QDs /inmicro-LEDs comparison without with the HBR. QDs/micro-LEDs This proved that without the HBR HBR. could This respectively increased 11% and 10% in comparison with the QDs/micro-LEDs without HBR. This reflectproved the that downward the HBR emittedcould reflect red, green,the downward and blue lights.emitted Furthermore, red, green, and it was blue worth lights. noting Furthermore, that only it proved that the HBR could reflect the downward emitted red, green, and blue lights. Furthermore, it partwas of worth the blue noting light that was only absorbed part of bythe the blue green light and was red absorbed QDs, which by the leads green to and non-absorbed red QDs, which blue lightleads was worth noting that only part of the blue light was absorbed by the green and red QDs, which leads blendingto non-absorbed with the blue green light light blending and red with light the simultaneously green light and being red light emitted simultaneously from the green being and emitted red to non-absorbed blue light blending with the green light and red light simultaneously being emitted QDsfrom/micro-LEDs the green and with red HBR. QDs/micro-LEDs Therefore, the non-absorbedwith HBR. Therefore, blue light the emitted non-absorbed from the blue QDs /lightmicro-LEDs emitted from the green and red QDs/micro-LEDs with HBR. Therefore, the non-absorbed blue light emitted withfrom HBR the QDs/micro-LEDs could be effectively with reflected HBR could back be into effectiv the QDely reflected color conversion back into layerthe QD by color using conversion the DBR from the QDs/micro-LEDs with HBR could be effectively reflected back into the QD color conversion withlayer high by using reflectivity the DBR at the with wavelength high reflectivity of 451 nmat the and wavelength excite more of red 451 and nm greenand excite lights. more Compared red and layer by using the DBR with high reflectivity at the wavelength of 451 nm and excite more red and withgreen the lights. QDs /Comparedmicro-LEDs with with the HBR QDs/micro-LEDs and without DBR,with HBR the output and without intensity DBR, of the the output green andintensity red green lights. Compared with the QDs/micro-LEDs with HBR and without DBR, the output intensity QDsof the/micro-LEDs green and red with QDs/micro-LEDs HBR and DBR was with increased HBR and 20% DBR and was 23%, increased respectively. 20% and However, 23%, respectively. since the of the green and red QDs/micro-LEDs with HBR and DBR was increased 20% and 23%, respectively. reflectivityHowever, since of the the DBR reflectivity was not 100%of the andDBR the was reflectivity not 100% and of the the DBR reflectivity was decreased of the DBR as was the lightdecreased was However, since the reflectivity of the DBR was not 100% and the reflectivity of the DBR was decreased notas the perpendicularly light was not incidentperpendicularly into the DBRincident structure, into the a smallDBR structure, portion of a bluesmall light portion still escaped,of blue light which still as the light was not perpendicularly incident into the DBR structure, a small portion of blue light still wouldescaped, aff ectwhich the colorwould purity affect of the the co QDslor purity/micro-LEDs. of the QDs/micro-LEDs. In this work, the blueIn this light work, absorption the blue layer light escaped, which would affect the color purity of the QDs/micro-LEDs. In this work, the blue light wasabsorption deposited layer on thewas DBR deposited to reduce on thethe escapedDBR to bluereduce light. the The escaped absorptivity blue light. spectrum The absorptivity of the blue absorption layer was deposited on the DBR to reduce the escaped blue light. The absorptivity lightspectrum absorption of the layerblue light is shown absorption in Figure layer7. Theis show absorptivityn in Figure of 7. the The blue absorptivity light absorption of the blue layer light at spectrum of the blue light absorption layer is shown in Figure 7. The absorptivity of the blue light wavelengthsabsorption layer of 451, at 550,wavelengths and 625 nmof 451, was 550, 77.2%, and 1.3%, 625 nm and was 1.8%, 77.2%, respectively. 1.3%, and Since 1.8%, the respectively. blue light absorption layer at wavelengths of 451, 550, and 625 nm was 77.2%, 1.3%, and 1.8%, respectively. escapedSince the from blue the light DBR escaped could befrom absorbed the DBR by could the blue be absorbed light absorption by the blue layer, light the intensityabsorption of layer, the blue the Since the blue light escaped from the DBR could be absorbed by the blue light absorption layer, the lightintensity for the of QDsthe blue/micro-LEDs light for wasthe QDs/micro-LEDs reduced, as shown was in Figurereduced,6, whichas shown could in improveFigure 6, thewhich red could and intensity of the blue light for the QDs/micro-LEDs was reduced, as shown in Figure 6, which could greenimprove color the purity. red and green color purity. improve the red and green color purity.

Blue micro-LEDs with HBR Blue micro-LEDs with HBR Blue micro-LEDs with HBR Blue micro-LEDs with HBR Green QDs/micro-LEDs Red QDs/micro-LEDs Green QDs/micro-LEDs Red QDs/micro-LEDs without HBR and DBR without HBR and DBR without HBR and DBR without HBR and DBR with HBR and without DBR with HBR and without DBR with HBR and without DBR with HBR and without DBR with HBR and DBR with HBR and DBR with HBR and DBR with HBR and DBR

with HBR, DBR, and with HBR, DBR, and

with HBR, DBR, and with HBR, DBR, and blue light absorption layer blue light absorption layer blue light absorption layer blue light absorption layer Relative Intensity (a.u.) Relative Relative Intensity (a.u.) Relative Intensity (a.u.) Relative Relative Intensity (a.u.) 400 450 500 550 600 650 700 400 450 500 550 600 650 700 400 450 500 550 600 650 700 400 450 500 550 600 650 700 Wavelength (nm) Wavelength (nm) Wavelength (nm) Wavelength (nm) (a) (b) (a) (b) Figure 6. The EL spectra of the resulting (a) green QDs/micro-LEDs and (b) red QDs/micro-LEDs. Figure 6. The EL spectra of the resulting ( a) green QDs/micro-LEDs QDs/micro-LEDs and (b)) redred QDsQDs/micro-LEDs./micro-LEDs.

100 100 90 90 80 80 70 70 (%)

(%) 60 60 50 50 40 40 30 Blue light absorption layer 30 Blue light absorption layer

Absorptivity 20

Absorptivity 20 10 10 0 0300 350 400 450 500 550 600 650 700 750 800 300 350 400 450 500 550 600 650 700 750 800 WavelengthWavelength (nm) (nm)

Figure 7. The absorptivity spectrum of the blue light absorption layer. Figure 7. The absorptivity spectrum of the blue light absorption layer. Figure 7. The absorptivity spectrum of the blue light absorption layer.

Coatings 2020, 10, 436 7 of 9

Coatings 2020, 10, x FOR PEER REVIEW 7 of 9 Figure8 shows the international commission on illumination (CIE) chromaticity coordinates of the resultingFigure micro-LEDs. 8 shows the Compared international with thecommission green and on red illumination QDs/micro-LEDs (CIE) withchromaticity HBR and coordinates without DBR, of thethe CIEresulting chromaticity micro-LEDs. coordinate Compared of the greenwith the and gr redeen QDs and/ micro-LEDsred QDs/micro-LEDs with HBR with and HBR DBR shiftedand without from (0.211,DBR, the 0.221) CIE to chromaticity (0.271, 0.531) andcoordinate from (0.311, of the 0.109) green to (0.564,and red 0.226), QDs/micro-LEDs respectively. Thewith CIE HBR chromaticity and DBR coordinateshifted from of (0.211, the green 0.221) and to red(0.271, QDs 0.531)/micro-LEDs and from with (0.311, HBR, 0.109) DBR, to and (0.564, blue 0.226), light respectively. absorption layer The wasCIE (0.292,chromaticity 0.632) andcoordinate (0.646, 0.279),of the green respectively. and red The QDs/micro-LEDs color purity of thewith LEDs HBR, could DBR, be and estimated blue light by Equationabsorption (1) layer as follows was (0.292, [15–17 0.632)]: and (0.646, 0.279), respectively. The color purity of the LEDs could be estimated by Equation (1) as follows [15–17]: q 2 2 (X X ) (Y Y ) ( −− i) −−(−− i) color purity= = q ×100% % (1) (1) × ((X −X)2 −((Y −Y ))2 d − i − d − i where (X, Y) is the CIE chromaticity coordinate of the resulting QDs/micro-LEDs, (Xi, Yi) is the CIE where (X, Y) is the CIE chromaticity coordinate of the resulting QDs/micro-LEDs, (X ,Y ) is the CIE chromaticity coordinate of the blue micro-LEDs, (Xd, Yd) is the CIE chromaticity coordinatei i of green chromaticitylight and red coordinate light at color ofthe purity blue of micro-LEDs, 100%. The CIE (Xd,Y chromaticityd) is the CIE coordinate chromaticity of the coordinate green light of green with lightwavelength and red of light 550 atnm color and purity the red of light 100%. with The wavelength CIE chromaticity of 625 coordinatenm was (0.301, of the 0.692) green and light (0.701, with wavelength0.299), respectively. of 550 nm Compared and the red with light the with green wavelength and red QDs/micro-LEDs of 625 nm was (0.301, with 0.692)HBR and and without (0.701, 0.299), DBR, respectively.the color purity Compared of the green with and thegreen red QDs/micro- and red QDsLEDs/micro-LEDs with HBR with and HBR DBR and was without improved DBR, from the 28.2% color purityto 75.4% of theand green from and27.7% red to QDs 74.2%,/micro-LEDs respectively. with Consequently, HBR and DBR the was DBR improved could effectively from 28.2% reflect to 75.4% the andnon-absorbed from 27.7% blue to 74.2%, light to respectively. improve the Consequently, color purity. theThe DBR color could purity eﬀ ofectively the green reﬂect and the red non-absorbed QDs/micro- blueLEDs light with to HBR, improve DBR, the and color blue purity. light Theabsorpti coloron purity layer ofwas the further green andimproved red QDs to/ micro-LEDs90.9% and 90.3%, with HBR,respectively. DBR, and blue light absorption layer was further improved to 90.9% and 90.3%, respectively.

Blue micro-LEDs with HBR and without DBR 0.9 with HBR and DBR 0.8 with HBE, DBR, and blue light absorption layer 0.7

0.6 s D

0.5 -L o

c 0.4 i m / s CIE Y

0.3 Q

e e s 0.2 r ED G o-L icr 0.1 s/m QD ed 0.0 R 0.0 0.2 0.4 0.6 0.8 CIE X Figure 8. The international commission on illumination (CIE) chromaticity coordinates of the resulting micro-LEDs. Figure 8. The international commission on illumination (CIE) chromaticity coordinates of the 4. Conclusionsresulting micro-LEDs. In this study, monolithic full color micro-LEDs were successfully fabricated by using blue 4. Conclusions micro-LEDs to excite various dimensions of CdSe/ZnS core-shell QDs. To increase the output intensity and improveIn this study, the color monolithic purity, full HBR color and micro-LEDs DBR were were used successfully as the bottom fabricated reflector by for using the blue full micro- colors andLEDs the to topexcite reflector various for dimensions the blue light,of CdSe/ZnS respectively. core-shell The QDs. downward To increase emitted the output full colors intensity could and be reflectedimprove bythe the color bottom purity, reflector HBR and HBR, DBR which were could used increaseas the bottom the probability reflector for of bluethe full light colors exciting and thethe redtop andreflector green for QDs. the blue Consequently, light, respectively. the output The intensity downward of the emitted green full and colors red QDs could/micro-LEDs be reflected with by HBRthe bottom was respectively reflector HBR, increased which11% could and increase 10% in the comparison probability with of blue the QDslight/ micro-LEDsexciting the red without and green HBR. Furthermore,QDs. Consequently, the output the intensityoutput intensity of the green of th ande green red QDsand /micro-LEDsred QDs/micro-LEDs with HBR with and DBRHBR waswas respectively increased 11% and 10% in comparison with the QDs/micro-LEDs without HBR. Furthermore, the output intensity of the green and red QDs/micro-LEDs with HBR and DBR was further increased 20% and 23%, respectively, as the blue light reflected by the DBR could excite QDs

Coatings 2020, 10, 436 8 of 9 further increased 20% and 23%, respectively, as the blue light reﬂected by the DBR could excite QDs to form more red and green lights. Finally, the blue light absorption layer with absorptivity of 77.2% at the wavelength of 451 nm was developed to absorb the blue light that escaped from the DBR. The color purity of the green and red QDs/micro-LEDs with HBR, DBR, and blue light absorption layer was further improved to 90.9% and 90.3%, respectively.

Author Contributions: Conceptualization, H.-Y.L., C.-T.L., K.-L.L., W.-H.K., Y.-H.F., and C.-C.L.; Data curation, S.-Y.C. and H.-Y.W.; Funding acquisition, H.-Y.L., C.-T.L., K.-L.L., W.-H.K., Y.-H.F., and C.-C.L.; Investigation, S.-Y.C., H.-Y.W., H.-Y.L., and C.-T.L.; Writing—original draft, S.-Y.C. and H.-Y.W.; Writing—review and editing, H.Y.L. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by the Ministry of Science and Technology (MOST), Taiwan under Nos. MOST-108-2221-E-006-196-MY3 and MOST 108-2221-E-006-215-MY3, and the Industrial Technology Research Institute, Taiwan. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References

1. Yeh, C.H.; Lo, K.S.; Lin, W.S. Visual-attention-based pixel dimming technique for OLED displays of mobile devices. IEEE Trans. Ind. Electron. 2019, 66, 7159–7167. [CrossRef] 2. Lu, X.; Yang, S.; Jiang, H.X.; Wu, J. Monolithic integration of GaN LEDs with vertical driving MOSFETs by selective area growth and band engineering of the p-AlGaN electron blocking layer though TCAD simulation. Semicond. Sci. Technol. 2019, 34, 064002. [CrossRef] 3. Crawford, M.H. LEDs for solid-state lighting: Performance challenges and recent advances. IEEE J. Sel. Top. Quantum Electron. 2009, 15, 1028–1040. [CrossRef] 4. Shur, M.S. Solid-state lighting: Toward superior illumination. Proc. IEEE 2005, 93, 1691–1703. [CrossRef] 5. Chen, K.J.; Chen, H.C.; Shih, M.H.; Wang, C.H.; Kuo, M.Y.; Yang, Y.C.; Lin, C.C.; Kuo, H.C. The influence of the thermal effect on CdSe/ZnS quantum dots in light-emitting diodes. J. Lightwave Technol. 2012, 30, 2256–2261. [CrossRef] 6. Talapin, D.V.; Steckel, J. Quantum dot light-emitting devices. MRS Bull. 2013, 38, 685–695. [CrossRef] 7. Lin, Y.; Zhang, Y.; Zhao, J.; Gu, P.; Bi, K.; Zhang, Q.; Chu, H.; Zhang, T.; Cui, T.; Wang, Y.; et al. White-light-emitting diodes using GaN-excited CdSe/CdS/ZnS quantum dots. Particuology 2014, 15, 90–93. [CrossRef] 8. Lee, C.T.; Cheng, C.J.; Lee, H.Y.; Chu, T.C.; Fang, Y.H.; Chao, C.H.; Wu, M.H. Color conversion of GaN-Based micro light-emitting diodes using quantum dots. IEEE Photonics Technol. Lett. 2015, 27, 2296–2299. [CrossRef] 9. Chen, G.S.; Wei, B.Y.; Lee, C.T.; Lee, H.Y. Monolithic Red/Green/Blue micro-LEDs with HBR and DBR structures. IEEE Photonics Technol. Lett. 2018, 30, 262–265. [CrossRef] 10. Dong, W.J.; Lo, N.T.; Jung, G.H.; Ham, J.; Lee, J.L. Efficiency enhancement and angle-dependent color change in see-through organic photovoltaics using distributed Bragg reflectors. Appl. Phys. Lett. 2016, 108, 103902. [CrossRef] 11. Leem, J.W.; Guan, X.Y.; Yu, J.S. Tunable distributed Bragg reflectors with wide-angle and broadband high-reflectivity using nanoporous/dense titanium dioxide film stacks for visible wavelength applications. Opt. Express 2014, 22, 18519–18526. [CrossRef][PubMed] 12. Lee, C.T.; Kao, H.W. Long term thermal stability of Ti/Al/Pt/Au ohmic contacts to n-type GaN. Appl. Phys. Lett. 2000, 76, 2364–2366. [CrossRef]

13. Lin, Y.J.; Tsai, C.D.; Lyu, Y.T.; Lee, C.T. X-ray photoelectron spectroscopy study of (NH4)2Sx-treated Mg-doped GaN layers. Appl. Phys. Lett. 2000, 77, 687–689. [CrossRef] 14. Chung, W.; Park, K.; Yu, H.J.; Kim, J.; Chun, B.H.; Kim, S.H. White emission using mixtures of CdSe Quantum dots and PMMA as a Phosphor. Opt. Mater. 2010, 32, 515–521. [CrossRef] 15. Pramod, A.G.; Nadaf, Y.F.; Renuka, C.G. A combined experimental theoretical approach for energy gap determination, photophysical, photostable, optoelectronic, NLO, and organic light emitting diode (OLED) application: Synthesized coumarin derivative. J. Mol. Struct. 2019, 1194, 271–283. [CrossRef] Coatings 2020, 10, 436 9 of 9

16. Wu, Y.T.; Chena, M.; Qiub, K.; Zhanga, W.; Tanga, Q. Photoluminescence enhancement of 3+ 3+ Ca3Sr3(VO4)4:Eu ,Al red-emitting phosphors by charge compensation. Opt. Laser Technol. 2019, 118, 20–27. [CrossRef] 17. He, A.; Xi, Z.; Li, X.J.; Long, W.; Fang, P.Y.; Zhang, J. Temperature dependence of upconversion luminescence and sensing sensitivity of Ho3+/Yb3+ modiﬁed PSN-PMN-PT crystals. J. Alloys Compd. 2019, 803, 405–455. [CrossRef]

© 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/).