Previous Article in Journal
Thermal Stability and Flammability Behaviors of Phosphorus/Graphene Oxide Co-Modified Waterborne Polyurethane Coatings: An Experimental Study
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Coatings
Volume 16
Issue 6
10.3390/coatings16060692
coatings-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Development of a Synthetic Optical Coating for Efficient UV Light Conversion and Enhanced Transmittance

by
Daolong Xu
1,2,3,
Daruo Cao
1,2,3,
Zihan Shan
1,2,3 and
Liang Fang
1,2,3,*
1
State Key Laboratory of Materials-Oriented Chemical Engineering, College of Materials Science and Engineering, Nanjing Tech University, Nanjing 210009, China
2
Jiangsu Collaborative Innovation Center for Advanced Inorganic Function Composites, Nanjing Tech University, Nanjing 210009, China
3
Jiangsu National Synergetic Innovation Center for Advanced Materials (SICAM), Nanjing Tech University, Nanjing 210009, China
*
Author to whom correspondence should be addressed.
Coatings 2026, 16(6), 692; https://doi.org/10.3390/coatings16060692 (registering DOI)
Submission received: 10 May 2026 / Revised: 3 June 2026 / Accepted: 8 June 2026 / Published: 10 June 2026
(This article belongs to the Section Composite Coatings)
Browse Figures
Versions Notes

Abstract

Photovoltaic modules require efficient sunlight modulation, including enhanced visible transmittance and conversion of unused ultraviolet light. This study develops a synthetic optical coating that achieves both functions by integrating down-conversion BAM (BaMgAl10O17:Eu2+, Mn2+) nanophosphors into a silica anti-reflection sol. The key novelty lies in a synergistic surface engineering strategy that decouples dispersion stabilization from luminescence protection. Five dispersants are systematically compared under combined ball and sand milling. The polyester-modified acrylic long-chain dispersant (DK062) yields a stable nanodispersion with an average particle size of 228 nm and a Zeta potential of −7.61 mV, effectively suppressing re-agglomeration while retaining high photoluminescence. Subsequent surface modification with KH570 grafts a dense silane passivation layer via Si–O–M covalent bonds, further increasing the photoluminescence intensity by 1.39-fold. The optimized nanophosphors are incorporated into a commercial anti-reflection sol and dip-coated onto photovoltaic glass. At a doping concentration of 2‰ and a withdrawal speed of 8 mm/s, the resulting DCSAR coating exhibits an average transmittance of 91.16%—slightly higher than that of the pure anti-reflection coating (90.96%)—while showing strong green emission at 515 nm. Industrial on-site testing further demonstrates an average transmittance of 94.20%–94.31% with uniform green emission. This work provides a scalable route to fabricate highly transparent, light-converting anti-reflection coatings by combining dispersant-assisted milling and silane passivation.

1. Introduction

In PV modules, the transmittance of the encapsulating glass directly affects the absorption and utilization efficiency of sunlight by the solar cells [1]. Conventional photovoltaic glass suffers from approximately 4% Fresnel reflection loss due to the refractive index mismatch at the air-glass interface [2]. To address this, porous silica anti-reflection coatings are developed, which effectively reduce reflectivity by creating a refractive index gradient [3,4]. In addition, short-wavelength photons in the solar spectrum (e.g., ultraviolet light, UV) possess high energy that conventional silicon-based cells cannot efficiently utilize, leading to spectral mismatch losses [5,6]. Converting high-energy ultraviolet/blue light into green light, to which solar cells respond more favorably, represents a potential route to enhance PV efficiency [7,8]. Therefore, integrating down-conversion phosphors with anti-reflection coatings to fabricate light-conversion anti-reflection coatings that offer both spectral modulation and anti-reflection functions holds significant research value [9,10,11]. Several comprehensive reviews have summarized the potential of down conversion and downshifting phosphors for enhancing silicon solar cell efficiency, particularly by improving UV response and reducing thermalization losses [10,12,13].
Currently, several approaches have been explored to combine anti-reflection and light-conversion functionalities. These include first, stacking separate anti-reflection and phosphor-containing layers, for example a porous SiO2 layer coated with a phosphor-doped film; second, directly embedding organic fluorescent dyes into the anti-reflection sol; and third, surface grafting of luminescent complexes onto the porous coating framework [14,15,16]. However, each method suffers from critical drawbacks. Multilayer stacking increases fabrication complexity and cost, and introduces interfacial reflections and adhesion issues [17]. Organic dyes often exhibit poor photostability and are prone to photobleaching under prolonged UV exposure, limiting device lifetime [18,19]. Direct incorporation of untreated inorganic phosphors, especially micron-sized particles, into the anti-reflection sol leads to severe light scattering, compromising the coating’s transparency and anti-reflection effect; moreover, particle agglomeration causes coating inhomogeneity and pinhole defects [20,21].
Given these challenges, direct blending of down-conversion inorganic phosphors into a commercial anti-reflection sol, provided the phosphors can be downsized to the nanoscale, stabilized, and surface-engineered, offers a cost-effective and industrially viable pathway to fabricate light-conversion anti-reflection coatings [22,23]. Yet, this strategy requires systematic investigation into several key issues: how to achieve stable nanodispersion of phosphors without severe fluorescence quenching; how to minimize light scattering while maintaining high transmittance; how to ensure compatibility between the phosphor particles and the sol matrix for defect-free coating formation; and how the phosphor doping concentration and coating parameters affect the final optical performance [24,25]. Addressing these questions is essential for developing efficient, low-cost, and scalable light-conversion anti-reflection coatings.
In addition, inorganic phosphors are typically prepared by high-temperature solid-state reaction, resulting in coarse particles with severe agglomeration, making them difficult to directly apply in optical coatings [26,27]. To obtain nanoscale dispersed phosphor particles, high-energy ball milling and sand milling are effective physical fragmentation methods [28,29]. Nevertheless, mechanical milling introduces a large number of surface defects and dangling bonds, which severely quench the fluorescence emission [30,31]. Meanwhile, the high surface energy of the nanoparticles tends to cause re-agglomeration, compromising coating uniformity [32]. Selecting an appropriate dispersant to provide steric hindrance and electrostatic repulsion, thus, is critical for achieving a stable nanodispersion [33,34]. Furthermore, chemical modification of the particle surface with a silane coupling agent is expected to passivate defects, enhance fluorescence, and improve compatibility between the particles and the sol matrix [35,36].
In this work, to prepare the synthetic optical coating with efficient UV light conversion and enhanced transmittance, we first compared the effects of different dispersants and milling processes (ball milling, sand milling, and their combination) on the particle size, dispersion stability, and fluorescence performance of BaMgAl10O17:Eu2+, Mn2+ (BAM), and identified the optimal dispersant and processing conditions. Subsequently, we performed surface silane modification on phosphors with different particle sizes using a silane coupling agent and investigated its effectiveness in defect passivation and fluorescence enhancement. Finally, the optimized nano-phosphors (both unmodified and modified) were incorporated into a commercial anti-reflection sol, and light-conversion anti-reflection coatings were fabricated on photovoltaic glass via the dip-coating method. The effects of phosphor doping concentration and withdrawal speed on the coating microstructure, transmittance, and fluorescence emission were systematically examined, and the application potential was validated through industrial on-site testing. This study aims to provide experimental evidence and theoretical guidance for the development of efficient and low-cost light-conversion anti-reflection coatings.

2. Experiment and Material

2.1. Materials and Characterizations

BaMgAl10O17:Eu2+, Mn2+ (BAM) was purchased from Shenzhen Yaode Sheng Technology Co., Ltd. DK059, DK061, DK062, 2020 was purchased from Guangdong Zhongke Hongtai New Materials Co., Ltd. YCK-2601 was purchased from Shanghai Jufeng Chemical Co., Ltd. γ-methacryloxypropyltrimethoxysilane (KH570) was purchased from Beijing Vokai Biotechnology Co., Ltd. Hydrochloric acid, ethanol and deionized water were both purchased from Sinopharm Group Chemical Reagent Co., Ltd. The anti-reflective liquid was purchased from Chenguang (Changzhou) New Material Technology Co., Ltd.
The surface morphology and microstructure of the samples were characterized using a field-emission scanning electron microscope (FE-SEM, SU8100, Hitachi, Japan) operated at an acceleration voltage of 5-10 kV. The surface topography was analyzed by atomic force microscopy (AFM) on a Bruker Dimension Icon system (Bruker, Germany) in tapping mode. Chemical structures were analyzed by Fourier transform infrared spectroscopy (FTIR) using a PerkinElmer spectrometer (Waltham, MA, USA). Optical properties were measured using an Agilent Cary 5000 UV-Vis-NIR spectrophotometer (Agilent Technologies, USA). Photoluminescence (PL) spectra were recorded using a Horiba Jobin Yvon FL3-221 fluorescence spectrometer (Jobin Yvon Inc., USA).

2.2. Preparation of Down-Conversion Dispersion

Firstly, 10 g of the BAM down-conversion phosphor (synthesized via high-temperature solid-state reaction; for its preparation and luminescence properties, see [37]) was added to a mixed solution consisting of 140 mL of ethanol and 10 mL of deionized water. Then, 3 g of different grades of dispersants were separately weighed and uniformly dispersed into the above mixed solution. The resulting mixture was ultrasonicated for 5 min. Ball milling and sand milling times were set to different values. The ball milling times were 2, 4, and 6 h, respectively. The mass ratio of large (10 mm), medium (8 mm), and small (5 mm) ball milling sands in each jar was approximately 20:20:100. Ball milling was performed in a unidirectional mode at a rotation speed of 200 rpm [38]. After ball milling, samples were collected and allowed to stand for observation. The samples obtained with different dispersants and ball milling times were designated as WFSJ-Q(h), 2020-Q(h), YCK-Q(h), DK059-Q(h), DK061-Q(h), and DK062-Q(h), respectively (see Table 1). The sand milling times were set to 15 and 30 min. For sand milling, 1 kg of zirconia sand (0.3 mm) was fed into the grinding chamber of a nano sand mill, and the rotation speed was set to 2500 rpm. After sand milling, samples were collected and allowed to stand for observation. The samples after sand milling were designated according to the dispersant and the milling time as WFSJ-Q(h)-S[min], 2020-Q(h)-S[min], YCK-Q(h)-S[min], DK059-Q(h)-S[min], DK061-Q(h)-S[min], and DK062-Q(h)-S[min] (see Table 2).

2.3. Modification of Down-Conversion Phosphors with Different Particle Sizes

Based on the processing window diagrams and particle size distributions, the phosphors selected for preparation in this chapter were obtained from the DK062-Q(6), EAK2020-Q(6)-S[30], and DK062-Q(6)-S[30] down-conversion dispersions via centrifugation and subsequent drying. The dried powders were redispersed in absolute ethanol, and dilute hydrochloric acid (1 mol/L) together with KH570 was added. The mixture was reacted in a water bath at 70 °C for 10 h, then naturally cooled to room temperature. After cooling, the mixture was centrifuged, and the solid precipitate was washed and dried to obtain the silane-modified inorganic down-conversion powder. The mass ratio of inorganic nanoparticles, absolute ethanol, γ-methacryloxypropyltrimethoxysilane (KH570), and dilute hydrochloric acid was 0.5:140:6:3.
For clarity, samples labeled with the suffix “-g-k” indicate that the phosphor powder has been surface-modified with the silane coupling agent KH570 (where “g” stands for “grafting” and “k” stands for “KH570”). Specifically, samples such as “2020-Q(6)-S[30]-g-k” and “DK062-Q(6)-S[30]-g-k” denote the corresponding phosphor dispersions after ball milling and sand milling, followed by KH570 grafting. Unmodified samples (without “-g-k”) were used as controls.

2.4. Preparation of Light-Conversion Anti-Reflection Solution and Light-Conversion Anti-Reflection Coating

To ensure the compactness of the light-conversion anti-reflection surface and in consideration of the fluorescence emission spectra, two down-conversion dispersions were selected and added to a commercially available anti-reflection solution. The solid content of the commercial down-conversion dispersion was determined to be 10%. The down-conversion dispersion was then incorporated into the commercial anti-reflection solution at mass ratios of 2‰, 5‰, and 10‰. The light-conversion anti-reflection coating prepared from the unmodified phosphor was designated as DCAR, while that from the modified phosphor was designated as DCSAR. Photovoltaic glass substrates were sequentially immersed in 10% hydrochloric acid and 10% ammonia cleaning solutions, each subjected to ultrasonic treatment at 60 W for 0.5 h. Subsequently, the substrates were ultrasonically rinsed with absolute ethanol and deionized water, then air-dried. The treated photovoltaic glass was clamped onto a fixture using the dip-coating method. Light-conversion anti-reflection coatings were prepared at withdrawal speeds of 5 mm/s, 8 mm/s, and 10 mm/s [39]. The coated samples were then placed in a muffle furnace, heated from room temperature to 650 °C at a ramp rate of 5 °C/min, and held at this temperature for 2 min [40].

3. Result and Discussion

3.1. Mechanical Regulation of Morphology and Size of Aluminate Inorganic Particles and Their Dispersion Performance

Figure 1a shows the XRD pattern of the pristine BAM phosphor. It can be observed that the multiple diffraction peaks of the pristine BAM phosphor are consistent with the standard BAM PDF card No. JCPDS 26-0163, indicating the high purity of the pristine BAM phosphor. Figure 1b presents four SEM micrographs of the pristine BAM phosphor at different magnifications. Collectively, these images reveal that the pristine BAM particles are severely agglomerated, forming large, dense clusters with sizes ranging from several micrometers to over ten micrometers. At higher magnifications, individual particles within the agglomerates can be seen to have irregular shapes and fused grain boundaries, indicating partial sintering. This agglomeration behavior is attributed to the high-temperature solid-state synthesis method used to prepare the BAM phosphor, during which grain melting, mutual adhesion, and uncontrolled grain growth readily occur. The four images together provide a clear baseline morphology: without any milling or dispersion treatment, the phosphor exists as hard agglomerates that would cause severe light scattering if directly incorporated into an optical coating. Furthermore, Figure 1c presents the particle size distribution histogram and its Gaussian fitting curve for the pristine BAM phosphor. The statistical results indicate that the particle size distribution of this phosphor is relatively broad, mainly ranging from 2 μm to 12 μm, with a calculated average particle size of 5.95 μm.
Figure 1d shows the infrared (IR) spectra of different grades of dispersants. First, in the wavenumber range of 3400–3500 cm−1, all five dispersants exhibit broad and pronounced absorption peaks, which are typically attributed to the stretching vibrations of –OH or –NH– groups within the molecules. In the region of 2800–3000 cm−1, each spectrum displays distinct sharp absorption peaks, characteristic of saturated C–H stretching vibrations, indicating that these dispersant molecules possess a carbon chain backbone of a certain length. Additionally, a strong characteristic absorption peak appears around 1700–1750 cm−1, which is generally assigned to the C=O stretching vibration, suggesting that the molecular structures of these dispersants commonly contain polar functional groups such as esters, carboxyl, or amide groups. In the fingerprint region from 1000 to 1450 cm−1, numerous absorption peaks are densely distributed, mainly corresponding to C–O stretching vibrations and C–H bending vibrations. Overall, these IR spectral features strongly indicate that the five dispersants are all polymeric surfactants or polymers rich in hydrophilic polar groups and hydrophobic nonpolar carbon chains. During the preparation of slurries or suspensions, these polar functional groups act as anchoring groups, enabling strong physical or chemical adsorption onto the surface of the BAM phosphor particles mentioned earlier. Meanwhile, their nonpolar carbon backbones extend as solvated chains into the dispersion medium, providing a strong steric hindrance effect that effectively prevents particle agglomeration caused by van der Waals forces.
Based on the experimental results shown in Figure 2, the dispersion and stability performance of BAM phosphor under different dispersants and milling processes can be summarized as follows. In the systems using ball milling alone (Figure 2a): As the ball milling time increased from 2 h to 6 h, the particle size decreased and the settling time prolonged. The blank group (WFSJ) and the system with EAK2020 exhibited significantly larger particle sizes (5045 nm and 4577 nm, respectively, at 2 h) and completely separated within 30–50 min, indicating extremely poor stability. In contrast, the DK series and YCK-2601 effectively reduced the particle size (controlled at 1600–2300 nm at 6 h). Among them, DK062, with its long polymeric chains of polyester-modified acrylate, constructed a thick steric hindrance layer and maintained a uniform milky-white state even after 12 h of standing, demonstrating the best overall performance.
After the introduction of high-energy sand milling (Figure 2b, ball milling for 2 h followed by sand milling for 15/30 min): Most systems achieved nanoscale particle sizes of 200–350 nm. However, a smaller particle size did not necessarily guarantee stability. The blank group settled and separated within only 20 min. EAK2020 exhibited an abnormal increase in particle size (to 549 nm) after 30 min of sand milling, attributed to poor wettability and dispersant desorption-induced agglomeration, resulting in a clear supernatant after 24 h of standing. In contrast, DK062 achieved a finely controlled particle size of 229 nm after 30 min of sand milling and remained uniformly milky-white after 24 h, demonstrating its superior anchoring strength and steric hindrance capability under high shear conditions.
When the preliminary ball milling time was extended to 4 h before sand milling (Figure 2c): Most systems reached particle sizes of 200–300 nm after 30 min of sand milling. Although the blank group achieved a particle size of 227 nm, it completely separated within 20 min. EAK2020 again showed an abnormal particle size increase (from 330 nm to 421 nm) after sand milling, with a settling time of only 30 min and clear solid–liquid separation after 24 h. The DK062 system exhibited a particle size of 230 nm after 30 min of sand milling and maintained a highly uniform milky-white suspension after 24 h.
When the preliminary ball milling time was further extended to 6 h before sand milling (Figure 2d): The stability of the blank group and the EAK2020 system deteriorated sharply, both showing clear supernatant after 24 h. EAK2020 also exhibited an abnormal particle size increase due to severe bridging flocculation. This abnormal increase can be explained by the weak anchoring of EAK2020 onto the BAM surface. Under the high shear force of sand milling (2500 rpm), the dispersant molecules partially desorb from the particle surface, exposing fresh, high-surface-energy sites. These bare surfaces rapidly re-aggregate via van der Waals attraction, forming either “soft agglomerates” (reversible) or, upon further mechanical compaction, “hard agglomerates” (irreversible). The measured size increase from 330 nm to 421 nm after 30 min sand milling is direct evidence of such desorption-induced flocculation. This behavior is consistent with the DLVO theory, where insufficient steric or electrostatic stabilization leads to a net attractive potential between particles. In contrast, DK062 continued to perform excellently, achieving a low particle size after 6 h of ball milling followed by 30 min of sand milling, with no separation after 24 h. The thorough ball milling pretreatment dissociated hard agglomerates, creating favorable conditions for the DK062 molecules to form a dense and continuous steric hindrance layer, thereby endowing the high-solid-content BAM slurry with long-term suspension stability.
From the SEM images of the unmodified phosphors shown in Figure 3, it can be clearly observed that the introduction of dispersants has a significant impact on the dispersibility of the phosphor, with marked differences among the various dispersants. First, Figure 3a shows the WFJS-Q(6) sample without any dispersant. The particles exhibit severe large agglomerates with tight adhesion between particles, making it nearly impossible to distinguish individual particle boundaries. This indicates that without the aid of a dispersant, mechanical ball milling alone cannot effectively overcome the high surface energy between particles, leading to rapid re-agglomeration after mechanical fragmentation and the formation of dense secondary aggregated structures. In contrast, Figure 3b shows the 2020-Q(6) sample with the addition of the 2020 dispersant. The agglomerate size is somewhat reduced, but noticeable adhesion between particles remains, indicating limited improvement in dispersibility. Figure 3c shows the DQ062-Q(6) sample with the addition of the DQ062 dispersant. Here, the particle distribution is the most uniform, large agglomerates are significantly reduced, and particle boundaries become clearer, demonstrating the best dispersion effect among the three. These results indicate that the choice of dispersant plays a decisive role in the final dispersion state of the phosphor. The DQ062 dispersant, under the ball milling conditions employed, is more effective at reducing the surface energy of the particles and providing sufficient steric hindrance, thereby inhibiting re-agglomeration between particles.
Figure 3d–i shows SEM morphologies of phosphors after mechanical fragmentation with different dispersants under the same pretreatment conditions. Based on ball milling for 6 h, the introduction of sand milling enhanced fragmentation, but the final dispersion state still depended on the dispersant type. In Figure 3d,e, the sample with dispersant 2020 (2020-Q(6)-S[15]) exhibits angular fragments with extensive fine powder adhering to large particle surfaces, forming a coating-like agglomerate structure. After 30 min of sand milling, the particles become finer, but surface adhesion shows little improvement, indicating that the 2020 dispersant forms a weak or easily detachable adsorption layer that cannot inhibit re-agglomeration of fresh surfaces. In contrast, Figure 3f,g shows the sample with DQ062 dispersant (DK062-Q(6)-S[15]), where particle distribution is relatively uniform, less fine powder adheres to large particles, and particle boundaries become clearer. After 30 min of sand milling, the surface shows almost no fine powder adhesion. This demonstrates that DQ062 not only reduces surface energy but also forms a stable adsorption layer during continuous fragmentation, effectively preventing secondary particle aggregation.
Figure 4 reveals the mechanisms by which different dispersants affect the stability of nano-BAM slurries. Figure 4a presents the Zeta potential results: the blank group (WFSJ) shows only −0.46 mV, near the isoelectric point, leading to easy agglomeration and sedimentation; the EAK2020 system gives +0.62 mV, indicating weak electrostatic protection and a tendency to induce bridging flocculation; in contrast, the DK062 system exhibits a significantly increased absolute Zeta potential of −7.61 mV, indicating that the polyester-modified acrylate forms a dense negatively charged layer on the particle surface, enhancing electrostatic repulsion and inhibiting agglomeration. Figure 4b shows the XRD patterns: after mechanical fragmentation, the phosphor exhibits decreased diffraction peak intensities, peak broadening, and the disappearance of some characteristic peaks, indicating reduced crystallinity and increased lattice defects. However, the main peak positions remain unchanged, confirming that the crystal structure is preserved.
To evaluate the influence of the combined milling process and different dispersants on the luminescence performance of BAM phosphors, photoluminescence emission spectra of the phosphors at each milling stage were analyzed (Figure 5). The decrease in photoluminescence intensity with increasing milling intensity is attributed to two main factors. First, high-energy mechanical fragmentation introduces a large number of lattice defects (such as vacancies, dislocations, and surface dangling bonds) that act as non-radiative recombination centers, as confirmed by the peak broadening and decreased intensity in the XRD patterns (Figure 4b). These defects provide alternative pathways for excited carriers to relax without emitting photons, thus reducing the radiative quantum efficiency. Second, the choice of dispersant critically affects both milling efficiency and the degree of surface protection.
In the blank group (WFSJ), the absence of a dispersant leads to severe hard agglomeration during milling. Such agglomeration concentrates mechanical impact at particle contacts, exacerbating lattice damage and generating more surface defects. Consequently, its luminescence intensity drops to 2.62 × 105. The EAK2020 system, although containing a dispersant, suffers from weak adsorption: during sand milling, the dispersant molecules partially desorb from the particle surface, leaving fresh surfaces unprotected and promoting re-agglomeration. As a result, its intensity drops to 1.39 × 106.
In contrast, the DK062 dispersant—a polyester-modified acrylic long-chain polymer—strongly adsorbs onto the BAM particle surface through multiple polar anchoring groups (as indicated by the FTIR peaks in Figure 1d, particularly the C=O stretch at ~1730 cm−1 and the broad O–H/N–H band). This adsorbed layer provides two critical benefits: it creates a thick steric hindrance layer that prevents particle re-agglomeration, ensuring uniform milling. it acts as a physical buffer that absorbs and dissipates part of the mechanical impact during ball and sand milling, thereby reducing the generation of new lattice defects. Under the optimal process (DK062-Q(6)-S[30]), although particle refinement to ~228 nm inevitably causes some luminescence loss due to the increased surface-to-volume ratio, the emission intensity remains significantly higher than that of the WFSJ and EAK2020 systems. This demonstrates that DK062 not only enables nanoscale dispersion but also maximally preserves the luminescent lattice integrity by mitigating mechanical damage—a critical advantage over dispersants that only provide steric stabilization without impact protection.

3.2. Surface Chemical Modification and Fluorescence Enhancement of Aluminate Particles

While high-energy mechanical milling effectively reduces the particle size of BAM phosphors, it inevitably introduces a large number of lattice defects and dangling bonds on their surfaces, leading to a significant decrease in luminescence intensity. These surface defects, acting as non-radiative recombination centers, severely limit the practical application efficiency of the phosphors. To repair the mechanical damage and enhance the luminescence performance, chemical modification and passivation of the phosphor surface are essential. In this study, BAM phosphors with different particle sizes were selected as the research objects, and their surfaces were modified using KH570.
Figure 6 shows the infrared spectra of BAM phosphors with different particle sizes after silane modification. From the overall absorption characteristics of the infrared spectra, all three modified phosphors exhibit significant organic functional group vibration peaks, directly confirming the success of the surface modification. Specifically, a broad absorption peak in the range of 3350–3415 cm−1 is attributed to the stretching vibrations of residual –OH groups or adsorbed water on the phosphor surface. A newly emerged sharp absorption peak near 2975 cm−1 is clearly assignable to the C–H stretching vibrations of the organic carbon chains in the silane coupling agent molecules. This characteristic peak, which is absent in unmodified inorganic BAM crystals, provides the most direct evidence for the successful introduction of silane long chains onto the phosphor surface. Furthermore, a strong and broad asymmetric absorption peak around 1050 cm−1 corresponds to the overlapping vibrations of Si–O–Si network crosslinking bonds formed by polycondensation of hydrolyzed siloxanes, as well as Si–O–M (where M represents Al, Ba, or Mg) chemical bonds generated by dehydration condensation between silanol groups and surface hydroxyl groups. This indicates that the silane molecules are not merely physically adsorbed on the phosphor surface but are chemically anchored via robust covalent bonds.
Figure 6b–d shows the SEM morphology of silane-modified phosphors with different particle sizes. After KH570 modification, the surface characteristics and modification effects differ significantly among the three phosphors, showing good consistency with the subsequent fluorescence performance data. EDS elemental analysis confirmed the successful grafting of KH570. C and Si elements were detected on the sample surfaces. The C element originates from the organic functional groups in the KH570 molecule, while the Si element comes from its silane end, indicating that KH570 has been uniformly coated onto the phosphor surface via chemical bonding. The presence of matrix elements such as O, Ba, and Mg confirms that the modification is only surface decoration and does not alter the bulk composition of the phosphor.
Figure 7 shows the fluorescence emission spectra of silane-modified phosphors with different particle sizes. Under the same sand milling conditions, the phosphor treated with the 2020 dispersant exhibits a fluorescence intensity of 1.39 × 106 at 515 nm, while that treated with the DK062 dispersant reaches 2.73 × 106, approximately twice as high. This result indicates that DK062 not only improves dispersibility but also provides significantly better protection of the luminescence performance compared to 2020. This is likely because, after adsorbing onto the particle surface, the dispersant molecules buffer the mechanical impact during ball/sand milling, reducing the generation of new surface defects from lattice damage. Additionally, the functional groups of the dispersant may weakly interact with surface dangling bonds, providing a preliminary passivation effect that suppresses the formation of non-radiative recombination centers. DK062, with its superior molecular structure and surface affinity, more effectively covers and protects the particle surface, thus exhibiting stronger luminescence protection capability. After silane modification, the fluorescence intensity of the 2020-treated phosphor increases to 1.05 × 107, an enhancement of approximately 7.55 times. This significant increase indicates that although 2020 provides some protection, a large number of defect sites remain on the surface. KH570 chemically grafts onto the phosphor surface, effectively occupying these defects and forming a dense passivation layer, which greatly suppresses non-radiative recombination and markedly improves the radiative recombination efficiency. For the DK062-treated phosphor, the fluorescence intensity after modification is 3.78 × 106, an enhancement of about 1.39 times. This is because DK062 itself already offers excellent surface protection, resulting in a low initial defect density, leaving limited room for further passivation by KH570. For the phosphor treated with DK062 without sand milling, its initial fluorescence intensity is as high as 1.79 × 107, and after KH570 modification it further increases to 2.30 × 107, an enhancement of about 1.28 times. This result again demonstrates the protective effect of DK062, and also shows that even without the sand milling process, KH570 can moderately enhance fluorescence by passivating residual defects or improving dispersion.

3.3. Application and Research of Inorganic Down-Conversion Nano-Phosphors in Anti-Reflection Coatings for Photovoltaic Modules

The BAM nano-phosphor DK062-Q(6)-S[30], prepared under the protection of the optimally performing DK062 dispersant, and its silane-modified counterpart DK062-Q(6)-S[30]-g-k were selected as functional units and physically blended into a commercial photovoltaic anti-reflection solution. By adjusting the phosphor doping concentration and the dip-coating process, a series of light-conversion anti-reflection coatings were fabricated on photovoltaic glass. The coatings prepared with the unmodified phosphor were designated as DCAR, while those prepared with the modified phosphor were designated as DCSAR.
Figure 8 illustrates the optical properties of the DCAR coatings. As shown in Figure 8a–c (transmittance spectra of DCAR coatings), an appropriate amount of DK062-Q(6)-S[30] nano-phosphor does not compromise the anti-reflection characteristics of the coating, but excessive addition leads to significant optical losses. From Table 3, the average transmittance of uncoated bare glass in the visible range is 88.97%. After applying the pure anti-reflection coating, as the withdrawal speed increases from 5 mm/s to 10 mm/s, the transmittance gradually improves. The AR10 coating achieves 90.96%, indicating that the optical thickness of the film at this withdrawal speed best matches the anti-reflection condition for the visible spectrum. With the introduction of a small amount of DK062-Q(6)-S[30] nano-phosphor, the coating still maintains excellent anti-reflection performance. The average transmittance of the 2‰ DCAR coatings at different withdrawal speeds remains between 90.2% and 91.1%. Notably, the 5‰ DCAR10 system attains an average transmittance of 91.07%, which is higher than that of the pure AR10 coating. This suggests that the low concentration (5‰) of nano-phosphor not only avoids significant Mie/Rayleigh scattering but may also further optimize the refractive index gradient on the coating surface, thereby enhancing visible light transmittance. When the addition of DK062-Q(6)-S[30] nano-phosphor is increased to 1%, the transmittance drops markedly. At a withdrawal speed of 8 mm/s, the 1% DCAR8 coating exhibits an average transmittance of 88.37%, which is lower than that of bare glass, and even the optimal 1% DCAR10 reaches only 89.81%. This confirms that an excessive amount of down-conversion nanoparticles may disrupt coating compactness, introducing scattering and thus reducing average transmittance.
From the fluorescence emission spectra of DCAR coatings (Figure 8d–f), it is evident that the luminescence intensity is positively correlated with both the phosphor concentration and the film thickness. All coatings retain the characteristic green emission peak at 515 nm. As the phosphor concentration increases from 2‰ to 5‰ and then to 1%, the absolute intensity of the green peak increases significantly. Although the absolute intensity increases with concentration, the relationship is not strictly linear. From 2‰ to 1%, the intensity only increases by a factor of 3.7. This sub-linear behavior is characteristic of concentration quenching. When the inter-particle distance becomes sufficiently small (≤10 nm), dipole–dipole interactions between Eu2+ or Mn2+ ions in neighboring particles provide alternative non-radiative decay pathways. Additionally, agglomeration at high concentrations creates local regions of high particle density, further reducing inter-particle distances and enhancing energy migration to quenching sites. At a withdrawal speed of 10 mm/s, the green peak intensity of 2‰ DCAR10 is approximately 1.3 × 105, that of 5‰ DCAR10 rises to 1.7 × 105, and that of 1% DCAR10 reaches 4.8 × 105. This indicates that the surge in fluorescence intensity within the coating is mainly due to the increased number of luminescent centers. At the same phosphor concentration, increasing the withdrawal speed effectively enhances the luminescence intensity. Comparing the 1% concentration group, the fluorescence intensity of 1% DCAR10 (4.8 × 105) is significantly higher than that of 1% DCAR8 and 1% DCAR5. This is because a higher withdrawal speed typically results in a thicker coating, which loads more phosphor particles per unit area, thereby maximizing the overall down-conversion light output of the coating.
To investigate the effect of incorporating unmodified DK062-Q(6)-S[30] nano-phosphor on the microstructure and optical properties of DCAR coatings, systematic characterization of the cross-sectional and surface morphology was performed using SEM. As shown in Figure 9a,b, the pure anti-reflection coating AR10 exhibits a stable cross-sectional thickness of 407 nm, with its surface densely packed by hollow silica nanoparticles, forming a flat and smooth coating. This coating effectively establishes a refractive index gradient between air and the glass substrate, significantly suppressing interfacial Fresnel reflection and thus achieving the desired broadband transmittance enhancement. As shown in Figure 9c,d, when the BAM nano-phosphor doping concentration is 5‰, the cross-sectional thickness of the 5‰ DCAR10 coating decreases to 282 nm. From the surface morphology, it is clearly observed that the phosphor nanoparticles are uniformly distributed within the anti-reflection matrix. The change in physical film thickness and the introduction of a high-refractive-index inorganic phase synergistically optimize the equivalent refractive index matching of the coating system. This moderate microstructural intervention not only preserves the continuity of the original matrix but further enhances the overall macroscopic transmittance. However, as shown in Figure 9e,f, when the BAM nano-phosphor concentration is increased to 1%, the film formation behavior of the 1% DCAR8 coating becomes significantly destabilized, with its cross-sectional thickness increasing to 391 nm. Severe secondary agglomeration of the unmodified phosphor at the high loading is clearly observed. The agglomerated phosphor particles hinder the continuous contraction of the hollow silica microsphere network, leading to the formation of distinct structural voids on the coating surface, which seriously compromise the film compactness. This microscopic imperfection, arising from both phosphor agglomeration and pore defects, induces strong light scattering losses at the macroscopic level, ultimately resulting in a marked decrease in the overall transmittance of the coating.
Figure 10 illustrates the optical properties of the DCSAR coatings after phosphor modification. From Figure 10a–c (transmittance spectra of DCSAR coatings), it can be seen that increasing the amount of DK062-Q(6)-S[30]-g-k nano-phosphor has a diminished effect on transmittance. Table 3 shows that the silane-modified nano-phosphor, under specific processing conditions, not only preserves the anti-reflection characteristics of the base coating but also provides an additional antireflective effect. Specifically, the bare glass substrate has an average transmittance of 88.97%, and the pure AR10 coating achieves 90.96%. When a low concentration of the modified DK062-Q(6)-S[30]-g-k nano-phosphor is introduced, the transmittance is further enhanced: at a BAM phosphor content of 2‰, the 2‰ DCSAR8 coating reaches 91.16%; and the 5‰ DCSAR10 coating reaches 91.02%. This phenomenon is attributed to the silane modification, which greatly improves the dispersibility of the inorganic particles in the sol system, enabling the formation of a uniform coating on the glass surface and achieving antireflective gain by optimizing the interfacial refractive index gradient. Moreover, even when the doping concentration is increased to 1%, the transmittance of the 1% DCSAR series coatings remains between 90.02% and 90.58%, still superior to that of the uncoated bare glass in all cases. This fully confirms that the silane coupling coating on the phosphor surface effectively suppresses secondary agglomeration of particles at high concentrations, greatly reducing Mie scattering and optical path obstruction losses caused by nano-agglomerates.
Figure 10d–f shows the fluorescence emission spectra of the DCSAR coatings. The fluorescence emission results indicate that, due to the effective passivation of surface lattice defects by the silane coating and its thermal buffering effect during high-temperature sintering, the luminescence performance of the modified coatings is significantly enhanced. The absolute intensity of the main green emission peak at 515 nm exhibits a strong positive correlation with the phosphor concentration and the withdrawal speed. As the phosphor concentration increases, the density of radiative transition centers within the coating increases correspondingly. For the 2‰ concentration series, the green peak intensity ranges from 1 × 105 to 1.27 × 105. For the 5‰ series, the intensity increases markedly to the range of 1.92 × 105–8.37 × 105. At a doping level of 1%, the emission intensity further rises to the range of 7.74 × 105–1.15 × 106. At the same phosphor concentration, a higher withdrawal speed increases the physical thickness of the coating, thereby increasing the effective loading of luminescent centers per unit area. Taking the 5‰ system as an example, when the withdrawal speed increases from 5 mm/s to 10 mm/s, the peak intensity dramatically increases from 1.92 × 105 to 8.37 × 105. The 1% system follows the same trend, with its luminescence intensity stably maintained around 1.09 × 106 at higher withdrawal speeds.
To verify the improvement in dispersion and optical performance of DK062-Q(6)-S[30]-g-k nano-phosphor in coatings after silane surface modification, systematic SEM characterization was performed on the modified DCSAR coatings. As shown in Figure 11a,b, the cross-sectional physical thickness of the 2‰ DCSAR8 coating is 377 nm. From the surface morphology, the coating exhibits a highly dense and smooth appearance. Furthermore, as shown in Figure 11c,d, when the modified phosphor concentration is greatly increased to 1%, the cross-sectional thickness of the 1% DCSAR10 coating increases with withdrawal speed to 476 nm, a dimensional evolution consistent with hydrodynamic film-formation behavior. Notably, even at such a high doping level, the nano-phosphor remains uniformly dispersed within the silica matrix, and the shrinkage cavities observed in the unmodified high-loading system have largely disappeared. These microstructural improvements unequivocally demonstrate that the silanized modification network on the phosphor surface successfully constructs a strong steric hindrance layer, which not only effectively shields the high van der Waals attraction between inorganic particles but also improves the compatibility between the phosphor and the silica sol matrix. This excellent compatibility ensures that the phosphor maintains an ideal dispersion state throughout the sol stage and subsequent high-temperature sintering, effectively suppressing micro-pore defects caused by framework agglomeration and hindered shrinkage. It is this dense, uniform, and large-scale-scattering-center-free optical interface that fundamentally reduces Mie scattering, further explaining the intrinsic mechanism by which DCSAR coatings achieve high-loading down-conversion luminescence while maximally avoiding macroscopic transmittance loss.
To investigate the structure–property relationship between coating surface roughness and anti-reflection performance, Figure 12 presents AFM images and corresponding roughness parameters of different samples. The results indicate that surface roughness and uniformity are key factors influencing the final transmittance of the coatings. As shown in Figure 12a,b, the AR10 sample exhibits the smoothest surface with the lowest roughness (Rq = 4.16 nm) and a height variation range of only −16.8 nm to 16.9 nm. Although the surface is not doped or modified, its flat structure helps reduce light diffusion scattering, resulting in an average transmittance of 90.96%, significantly higher than that of bare glass. This demonstrates that a dense, smooth film alone already possesses good anti-reflection properties, laying the foundation for achieving higher-performance coatings.
In contrast, the 5% DCAR10 sample in Figure 12c,d shows a marked increase in surface roughness to 5.32 nm after doping, along with the largest height variation range. Notably, its average transmittance reaches 91.07%, exceeding that of the smoother AR10. This phenomenon can be attributed to the moderately increased surface roughness, which helps form a graded refractive index profile on the film surface. Such a gradient structure effectively reduces Fresnel reflection and thus enhances light transmission. However, the presence of visible agglomerates and large-scale undulations on the sample surface may introduce additional light scattering centers, limiting further improvement in transmittance. Figure 12e,f shows the optimal modified sample, 2% DCSAR8. Its roughness Rq = 5.58 nm is slightly higher than that of AR10, but importantly, its height variation range is significantly smaller than that of 5% DCAR10. This indicates that after phosphor modification, the coating surface forms a more uniform nanostructure. Its excellent uniformity minimizes scattering losses caused by surface agglomeration or large particles. It is this ideal surface structure—combining moderate roughness with high uniformity—that enables the 2% DCSAR8 coating to achieve the highest average transmittance of 91.16% while reducing reflection and minimizing scattering losses.

3.4. Industrial On-Site Testing

Figure 13a–d shows the optical photographs and fluorescence emission spectra of the single-layer light-conversion anti-reflection photovoltaic glass under different roller speeds. The fluorescence intensity of the CG-5 series (roller speed 5 mm/s) is significantly higher than that of the CG-15 series (roller speed 15 mm/s). This is mainly because a lower coating speed allows the coating to remain on the glass surface for a longer time, leading to better slurry leveling and more uniform dispersion of the phosphor particles, indicating that a lower roller speed helps enhance the light-conversion effect of the coating. Table 4 lists the average and peak transmittance values of the two groups measured by the industrial partner. The data show that the average transmittance of the CG-5 series (94.20%–94.31%) is significantly better than that of the CG-15 series (93.15%–93.29%). This is because the low-speed coating produces a moderate thickness and a smoother surface, which not only reduces light scattering loss but also enhances the interference anti-reflection effect of the film. In contrast, high-speed coating tends to result in a thinner or uneven coating, thereby reducing the anti-reflection capability. Taken together, the results indicate that at a roller speed of 5 mm/s, the single-layer light-conversion anti-reflection coating achieves higher transmittance and stronger fluorescence emission, demonstrating superior overall optical performance.
Regarding long-term durability, the high-temperature sintering step at 650 °C for 2 min drives complete polycondensation of the silica sol–gel network, forming a dense Si–O–Si framework that is chemically inert and mechanically robust. The covalent grafting of KH570 further anchors the phosphor particles to the matrix, preventing interfacial delamination.

4. Conclusions

In summary, we have developed a synthetic optical coating that combines efficient UV-to-green down-conversion with enhanced visible transmittance for photovoltaic applications. A systematic comparison of five dispersants under combined ball and sand milling reveals that the polyester-modified acrylic long-chain dispersant (DK062) achieves the best performance: an average particle size of 228 nm, a Zeta potential of −7.61 mV, and the highest retained photoluminescence. Subsequent surface grafting with KH570 forms a dense Si–O–M passivation layer, further increasing the photoluminescence intensity by 1.39-fold. Remarkably, even at a doping concentration of 2‰, the DCSAR coating maintains high transparency with an average transmittance of 91.16%—which is slightly higher than that of the pure AR coating (90.96%)—while exhibiting strong green emission at 515 nm. These results demonstrate that dispersant-assisted milling combined with silane surface passivation enables the fabrication of highly transparent down-conversion anti-reflection coatings without compromising optical performance.

Author Contributions

Investigation, D.C.; Resources, L.F.; Writing—original draft, D.X.; Supervision, Z.S.; Funding acquisition, L.F. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by Jiangsu Agricultural Science and Technology Innovation Fund (CX (24) 1025).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data are contained within the article.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Law, A.M.; Jones, L.O.; Walls, J.M. The performance and durability of anti-reflection coatings for solar module cover glass—A review. Sol. Energy 2023, 261, 85–95. [Google Scholar] [CrossRef]
  2. Freiburger, B.M.; Thompson, C.S.; Fleming, R.A.; Hutchings, D.; Pop, S.C. High efficiency anti-reflective coating for PV module glass. In Proceedings of the 2017 IEEE 44th Photovoltaic Specialist Conference (PVSC), Washington, DC, USA, 25–30 June 2017; IEEE: New York, NY, USA, 2017; pp. 1869–1872. [Google Scholar]
  3. Wu, G.; Wang, J.; Shen, J.; Yang, T.; Zhang, Q.; Zhou, B.; Deng, Z.; Fan, B.; Zhou, D.; Zhang, F. A novel route to control refractive index of sol-gel derived nano-porous silica films used as broadband antireflective coatings. Mater. Sci. Eng. B. 2000, 78, 135–139. [Google Scholar] [CrossRef]
  4. Tao, C.; Zou, X.; Reddy, K.M.; Zhang, L.; Jiang, B. A hydrophobic ultralow refractive-index silica coating towards double-layer broadband antireflective coating with exceptionally high vacuum stability and laser-induced damage threshold. Colloids Surf. A 2019, 563, 340–349. [Google Scholar] [CrossRef]
  5. Mirershadi, S.; Sattari, F.; Hadi, S.; Rasoulzadehfard, S.; Mortazavi, S.Z. Efficiency enhancement of silicon-based solar cells by solar energy harvesting in WS2 nano-flake/perovskite layer. J. Mater. Sci. Mater. Electron. 2023, 34, 1794. [Google Scholar] [CrossRef]
  6. Yu, J.H.; Cho, D.-H.; Lee, W.-J.; Kim, W.-J.; Kang, S.J.; Chung, Y.-D. Application of quantum dot down-conversion layer in thin-film solar cells to increase short-wavelength spectral response. ECS J. Solid State Sci. Technol. 2021, 10, 055012. [Google Scholar] [CrossRef]
  7. Ho, W.-J.; Yang, G.-C.; Shen, Y.-T.; Deng, Y.-J. Improving efficiency of silicon solar cells using europium-doped silicate-phosphor layer by spin-on film coating. Appl. Surf. Sci. 2016, 365, 120–124. [Google Scholar] [CrossRef]
  8. Gu, S.; Lu, Z.; Zou, S.; Wu, C.; Peng, C.; Ni, M.; Chen, Z.; Huang, H.; Sun, H.; Wang, H.; et al. In situ generating YVO4: Eu3+, Bi3+ downshifting phosphors in SiO2 antireflection coating for efficiency enhancement and ultraviolet stability of silicon solar cells. Sol. RRL 2023, 7, 2300215. [Google Scholar] [CrossRef]
  9. Zheng, J.; Mehrvarz, H.; Liao, C.; Bing, J.; Cui, X.; Li, Y.; Gonçales, V.R.; Lau, C.F.J.; Lee, D.S.; Li, Y.; et al. Large-area 23%-efficient monolithic perovskite/homojunction-silicon tandem solar cell with enhanced UV stability using down-shifting material. ACS Energy Lett. 2019, 4, 2623–2631. [Google Scholar] [CrossRef]
  10. De La Mora, M.B.; Amelines-Sarria, O.; Monroy, B.M.; Hernández-Pérez, C.D.; Lugo, J.E. Materials for downconversion in solar cells: Perspectives and challenges. Sol. Energy Mater. Sol. Cells 2017, 165, 59–71. [Google Scholar] [CrossRef]
  11. Wang, P.; Yan, X.; Wang, H.; Luo, C.; Wang, C. Study on improving the efficiency of crystalline silicon photovoltaic module with down-conversion chlorophyll film. Opt. Mater. 2022, 132, 112821. [Google Scholar] [CrossRef]
  12. Mayavan, A. Comprehensive review on downconversion/downshifting silicate-based phosphors for solar cell applications. ACS Omega 2024, 9, 3c08806. [Google Scholar] [CrossRef]
  13. Yu, I.-S.; Wu, S.-C.; Dumont, L.; Cardin, J.; Labbé, C.; Gourbilleau, F. Monolithic crystalline silicon solar cells with SiN layers doped with Tb3+ and Yb3+ rare-earth ions. J. Rare Earths 2019, 37, 515–519. [Google Scholar] [CrossRef]
  14. Hung, W.-B.; Chen, T.-M. Efficiency enhancement of silicon solar cells through a downshifting and antireflective oxysulfide phosphor layer. Sol. Energy Mater. Sol. Cells 2015, 133, 39–47. [Google Scholar] [CrossRef]
  15. Jung, E.D.; Kim, C.U.; Noh, Y.W.; Seo, S.K.; Noh, Y.I.; Choi, K.J.; Song, M.H. Aesthetic and efficient perovskite/si tandem solar cells using luminescent down-shifting textured anti-reflection films. EcoMat 2023, 5, e12399. [Google Scholar] [CrossRef]
  16. Meier, T.; Solares, S.D. Rhodamine-doped nanoporous polymer films as high-performance anti-reflection coatings and optical filters. Nanoscale 2016, 8, 17675–17685. [Google Scholar] [CrossRef]
  17. Hou, G.; García, I.; Rey-Stolle, I. High-low refractive index stacks for broadband antireflection coatings for multijunction solar cells. Sol. Energy 2021, 217, 29–39. [Google Scholar] [CrossRef]
  18. Griffini, G.; Bella, F.; Nisic, F.; Dragonetti, C.; Roberto, D.; Levi, M.; Bongiovanni, R.; Turri, S. Multifunctional luminescent down-shifting fluoropolymer coatings: A straightforward strategy to improve the UV-light harvesting ability and long-term outdoor stability of organic dye-sensitized solar cells. Adv. Energy Mater. 2015, 5, 1401312. [Google Scholar] [CrossRef]
  19. Pintossi, D.; Iannaccone, G.; Colombo, A.; Bella, F.; Välimäki, M.; Väisänen, K.-L.; Hast, J.; Levi, M.; Gerbaldi, C.; Dragonetti, C.; et al. Luminescent downshifting by photo-induced sol-gel hybrid coatings: Accessing multifunctionality on flexible organic photovoltaics via ambient temperature material processing. Adv. Electron. Mater. 2016, 2, 1600288. [Google Scholar] [CrossRef]
  20. Lee, S.; Kim, C.U.; Bae, S.; Liu, Y.; Noh, Y.I.; Zhou, Z.; Leu, P.W.; Choi, K.J.; Lee, J.-K. Improving light absorption in a perovskite/si tandem solar cell via light scattering and UV-down shifting by a mixture of SiO2 nanoparticles and phosphors. Adv. Funct. Mater. 2022, 32, 2204328. [Google Scholar] [CrossRef]
  21. De Boer, D.K.G.; Broer, D.J.; Debije, M.G.; Keur, W.; Meijerink, A.; Ronda, C.R.; Verbunt, P.P.C. Progress in phosphors and filters for luminescent solar concentrators. Opt. Express 2012, 20, A395. [Google Scholar] [CrossRef] [PubMed]
  22. Yao, H.; Tang, Q. Luminescent anti-reflection coatings based on down-conversion emission of Tb3+-Yb3+ co-doped NaYF4 nanoparticles for silicon solar cells applications. Sol. Energy 2020, 211, 446–452. [Google Scholar] [CrossRef]
  23. Huang, J.Y.; Wang, Y.; Fei, G.T.; Xu, S.H.; Wang, B.; Zeng, Z. Dual-functional antireflection and down-shifting coating for si solar cells. Colloids Surf. A. 2022, 652, 129907. [Google Scholar] [CrossRef]
  24. Li, C.; Yang, J.; He, J. Mechanically robust, high-transmittance bilayer sol–gel anti-reflective coating with down-shifting function. Appl. Surf. Sci. 2026, 724, 165716. [Google Scholar] [CrossRef]
  25. Zhang, X.; Sun, Z.; Zhu, Z.; Luo, J.; Wu, Z.; Wang, Z. High-efficient, spherical and thermal-stable carbon dots@silica fluorescent composite as rare earth-free phosphors for white LED. Ceram. Int. 2020, 46, 14706–14712. [Google Scholar] [CrossRef]
  26. Rahman, J.U.; Khan, S.; Jain, V.; Rajiv, A.; Dasi, S.; Fawy, K.F.; Jindal, P.K.; Sivaranjani, R. Exploring inorganic phosphors: Basics, types, fabrications and their luminescence properties for LED/WLED/displays. Rev. Inorg. Chem. 2025, 45, 55–76. [Google Scholar] [CrossRef]
  27. Song, Z.; Liao, J.; Ding, X.; Liu, X.; Liu, Q. Synthesis of YAG phosphor particles with excellent morphology by solid state reaction. J. Cryst. Growth 2013, 365, 24–28. [Google Scholar] [CrossRef]
  28. Yang, H.K.; Lee, J.H.; Hong, W.T.; Jang, H.; Moon, B.K.; Jeong, J.H.; Je, J.Y. Preparation and photoluminescence properties of nano-sized SrZnO2:Sm3+ phosphor powders obtained by high-energy ball milling. Ceram. Int. 2015, 41, 991–994. [Google Scholar] [CrossRef]
  29. Hong, W.T.; Yang, H.K.; Jeong, J.H. Fine yellow GdSr2AlO5:Ce3+ phosphor for white LEDs prepared by high energy ball milling process. Ceram. Int. 2016, 42, 4594–4599. [Google Scholar] [CrossRef]
  30. Baláž, P.; Achimovičová, M.; Baláž, M.; Billik, P.; Cherkezova-Zheleva, Z.; Criado, J.M.; Delogu, F.; Dutková, E.; Gaffet, E.; Gotor, F.J.; et al. Hallmarks of mechanochemistry: From nanoparticles to technology. Chem. Soc. Rev. 2013, 42, 7571. [Google Scholar] [CrossRef] [PubMed]
  31. Gu, F.; Wang, P.; Yu, H.; Guo, B.; Tong, L. Optical quenching of photoconductivity in CdSe single nanowires via waveguiding excitation. Opt. Express 2011, 19, 10880. [Google Scholar] [CrossRef]
  32. Ma, P.-C.; Mo, S.-Y.; Tang, B.-Z.; Kim, J.-K. Dispersion, interfacial interaction and re-agglomeration of functionalized carbon nanotubes in epoxy composites. Carbon 2010, 48, 1824–1834. [Google Scholar] [CrossRef]
  33. Anastasiadis, S.H.; Chrissopoulou, K.; Stratakis, E.; Kavatzikidou, P.; Kaklamani, G.; Ranella, A. How the physicochemical properties of manufactured nanomaterials affect their performance in dispersion and their applications in biomedicine: A review. Nanomaterials 2022, 12, 552. [Google Scholar] [CrossRef] [PubMed]
  34. Böck, N.C.; Sundermann, J.; Koziolek, M.; Keller, B.-L.; Mäder, K. Impact of stabilizers on particle size and dispersion behavior in biorelevant media in solid nanocrystal formulations. Eur. J. Pharm. Biopharm. 2025, 208, 114651. [Google Scholar] [CrossRef]
  35. Ahangaran, F.; Navarchian, A.H. Recent advances in chemical surface modification of metal oxide nanoparticles with silane coupling agents: A review. Adv. Colloid Interface Sci. 2020, 286, 102298. [Google Scholar] [CrossRef]
  36. Urakawa, K.; Miyasato, K.; Totani, K.; Watanabe, T. Emission increase in persistent inorganic phosphor improving absorption property by surface chemical modification. J. Lumin. 2020, 217, 116772. [Google Scholar] [CrossRef]
  37. Yang, P.; Yao, G.-Q.; Lin, J.-H. Energy transfer and photoluminescence of BaMgAl10O17 co-doped with Eu2+ and Mn2+. Opt. Mater. 2004, 26, 327–331. [Google Scholar] [CrossRef]
  38. Gusev, A.I.; Kurlov, A.S. Production of nanocrystalline powders by high-energy ball milling: Model and experiment. Nanotechnology 2008, 19, 265302. [Google Scholar] [CrossRef] [PubMed]
  39. Tamar, Y.; Kahanov, M.; Haspel, C.; Sasson, Y. Size selectivity during dip coating of sol–gel silica-based antireflective coatings and its effect on the porosity of the coatings. J. Coat. Technol. Res. 2016, 13, 1103–1113. [Google Scholar] [CrossRef]
  40. Reda, S.M. Synthesis and optical properties of CdS quantum dots embedded in silica matrix thin films and their applications as luminescent solar concentrators. Acta Mater. 2008, 56, 259–264. [Google Scholar] [CrossRef]
Coatings 16 00692 g001
Figure 1. (a) XRD pattern of the BAM phosphor; (b) SEM image showing the morphology of the pristine BAM phosphor; (c) Particle size distribution of the BAM phosphor; (d) Infrared spectroscopy of different grades of dispersants.
Figure 1. (a) XRD pattern of the BAM phosphor; (b) SEM image showing the morphology of the pristine BAM phosphor; (c) Particle size distribution of the BAM phosphor; (d) Infrared spectroscopy of different grades of dispersants.
Coatings 16 00692 g001
Coatings 16 00692 g002
Figure 2. Processing window diagrams: (a) different ball milling times; (b) different sand milling times after ball milling for 2 h; (c) different sand milling times after ball milling for 4 h; (d) different sand milling times after ball milling for 6 h. Note: In each diagram, color represents settling time and circle size represents particle size.
Figure 2. Processing window diagrams: (a) different ball milling times; (b) different sand milling times after ball milling for 2 h; (c) different sand milling times after ball milling for 4 h; (d) different sand milling times after ball milling for 6 h. Note: In each diagram, color represents settling time and circle size represents particle size.
Coatings 16 00692 g002
Coatings 16 00692 g003
Figure 3. SEM images of the phosphor after ball milling with different dispersants: (a) WFJS-Q(6); (b) 2020-Q(6); (c) DQ062-Q(6); (d) 2020-Q(6)-S[15]; (e) 2020-Q(6)-S[30]; (f) DK062-Q(6)-S[15]; (g) DK062-Q(6)-S[30]; (h) DK059-Q(6)-S[15]; (i) DK059-Q(6)-S[30].
Figure 3. SEM images of the phosphor after ball milling with different dispersants: (a) WFJS-Q(6); (b) 2020-Q(6); (c) DQ062-Q(6); (d) 2020-Q(6)-S[15]; (e) 2020-Q(6)-S[30]; (f) DK062-Q(6)-S[15]; (g) DK062-Q(6)-S[30]; (h) DK059-Q(6)-S[15]; (i) DK059-Q(6)-S[30].
Coatings 16 00692 g003
Coatings 16 00692 g004
Figure 4. Under the same pretreatment conditions: (a) Zeta potential of different light-conversion dispersions; (b) XRD patterns of the phosphors.
Figure 4. Under the same pretreatment conditions: (a) Zeta potential of different light-conversion dispersions; (b) XRD patterns of the phosphors.
Coatings 16 00692 g004
Coatings 16 00692 g005
Figure 5. Fluorescence emission spectra of the BAM phosphor after sand milling with different dispersants under identical process conditions. Note: Excitation was performed at 365 nm, and the spectra were measured on a Horiba Jobin Yvon FL3-221 spectrometer. (a) Ball mill for 2 h and sand mill for 15 min; (b) Ball mill for 2 h and sand mill for 30 min; (c) Ball mill for 4 h and sand mill for 15 min; (d) Ball mill for 4 h and sand mill for 30 min; (e) Ball mill for 6 h and sand mill for 15 min; (f) Ball mill for 6 h and sand mill for 30 min.
Figure 5. Fluorescence emission spectra of the BAM phosphor after sand milling with different dispersants under identical process conditions. Note: Excitation was performed at 365 nm, and the spectra were measured on a Horiba Jobin Yvon FL3-221 spectrometer. (a) Ball mill for 2 h and sand mill for 15 min; (b) Ball mill for 2 h and sand mill for 30 min; (c) Ball mill for 4 h and sand mill for 15 min; (d) Ball mill for 4 h and sand mill for 30 min; (e) Ball mill for 6 h and sand mill for 15 min; (f) Ball mill for 6 h and sand mill for 30 min.
Coatings 16 00692 g005
Coatings 16 00692 g006
Figure 6. (a) Infrared spectra of the converted powder after silane modification at different particle sizes; (b–d) SEM morphology and elemental distribution of silane-modified phosphors with different particle sizes: (b) 2020-Q(6)-S[30]-g-k, (c) DK062-Q(6)-g-k, (d) DK062-Q(6)-S[30]-g-k.
Figure 6. (a) Infrared spectra of the converted powder after silane modification at different particle sizes; (b–d) SEM morphology and elemental distribution of silane-modified phosphors with different particle sizes: (b) 2020-Q(6)-S[30]-g-k, (c) DK062-Q(6)-g-k, (d) DK062-Q(6)-S[30]-g-k.
Coatings 16 00692 g006
Coatings 16 00692 g007
Figure 7. Fluorescence emission spectra of silane-modified phosphors with different particle sizes: (a) 2020-Q(6); (b) DK062-Q(6); (c) DK062-Q(6)-S[30]. Note: Excitation was performed at 365 nm, and the spectra were measured on a Horiba Jobin Yvon FL3-221 spectrometer.
Figure 7. Fluorescence emission spectra of silane-modified phosphors with different particle sizes: (a) 2020-Q(6); (b) DK062-Q(6); (c) DK062-Q(6)-S[30]. Note: Excitation was performed at 365 nm, and the spectra were measured on a Horiba Jobin Yvon FL3-221 spectrometer.
Coatings 16 00692 g007
Coatings 16 00692 g008
Figure 8. Optical properties of the DCAR coatings: (a–c) Transmittance spectra of the DCAR coatings; (d–f) Fluorescence emission spectra of the DCAR coatings. Note: Excitation was performed at 365 nm, and the spectra were measured on a Horiba Jobin Yvon FL3-221 spectrometer.
Figure 8. Optical properties of the DCAR coatings: (a–c) Transmittance spectra of the DCAR coatings; (d–f) Fluorescence emission spectra of the DCAR coatings. Note: Excitation was performed at 365 nm, and the spectra were measured on a Horiba Jobin Yvon FL3-221 spectrometer.
Coatings 16 00692 g008
Coatings 16 00692 g009
Figure 9. Cross-sectional and surface SEM morphology of the DCAR coatings: (a,b) AR10 coating; (c,d) 5‰ DCAR10 coating; (e,f) 1% DCAR8 coating.
Figure 9. Cross-sectional and surface SEM morphology of the DCAR coatings: (a,b) AR10 coating; (c,d) 5‰ DCAR10 coating; (e,f) 1% DCAR8 coating.
Coatings 16 00692 g009
Coatings 16 00692 g010
Figure 10. Optical properties of the DCSAR coatings: (a–c) Transmittance spectra of the DCSAR coatings; (d–f) Fluorescence emission spectra of the DCSAR coatings. Note: Excitation was performed at 365 nm, and the spectra were measured on a Horiba Jobin Yvon FL3-221 spectrometer.
Figure 10. Optical properties of the DCSAR coatings: (a–c) Transmittance spectra of the DCSAR coatings; (d–f) Fluorescence emission spectra of the DCSAR coatings. Note: Excitation was performed at 365 nm, and the spectra were measured on a Horiba Jobin Yvon FL3-221 spectrometer.
Coatings 16 00692 g010
Coatings 16 00692 g011
Figure 11. Cross-sectional and surface SEM morphology of the DCSAR coatings: (a,b) 2‰ DCSAR8 coating; (c,d) 1% DCSAR10 coating.
Figure 11. Cross-sectional and surface SEM morphology of the DCSAR coatings: (a,b) 2‰ DCSAR8 coating; (c,d) 1% DCSAR10 coating.
Coatings 16 00692 g011
Coatings 16 00692 g012
Figure 12. Atomic force microscopy images of the pure anti-reflection coating and the light-conversion anti-reflection coatings: (a,b) AR10 coating; (c,d) 5‰ DCAR10 coating; (e,f) 2‰ DCSAR8 coating.
Figure 12. Atomic force microscopy images of the pure anti-reflection coating and the light-conversion anti-reflection coatings: (a,b) AR10 coating; (c,d) 5‰ DCAR10 coating; (e,f) 2‰ DCSAR8 coating.
Coatings 16 00692 g012
Coatings 16 00692 g013
Figure 13. Optical photographs and fluorescence emission spectra of the single-layer light-conversion anti-reflection photovoltaic glass: (a,b) Prepared at a roller speed of 5 mm/s; (c,d) Prepared at a roller speed of 15 mm/s. Note: Excitation was performed at 365 nm, and the spectra were measured on a Horiba Jobin Yvon FL3-221 spectrometer.
Figure 13. Optical photographs and fluorescence emission spectra of the single-layer light-conversion anti-reflection photovoltaic glass: (a,b) Prepared at a roller speed of 5 mm/s; (c,d) Prepared at a roller speed of 15 mm/s. Note: Excitation was performed at 365 nm, and the spectra were measured on a Horiba Jobin Yvon FL3-221 spectrometer.
Coatings 16 00692 g013
Table 1. Preparation formula of down-conversion dispersion by ball milling.
Table 1. Preparation formula of down-conversion dispersion by ball milling.
Sample NameDispersant
Content/g		Dispersant
Type		Ball Milling Time/h
WFSJ-Q(2)3Without dispersant2
WFSJ-Q(4)3Without dispersant4
WFSJ-Q(6)3Without dispersant6
2020-Q(2)3EAK20202
2020-Q(4)3EAK20204
2020-Q(6)3EAK20206
YCK-Q(2)3YCK-26012
YCK-Q(4)3YCK-26014
YCK-Q(6)3YCK-26016
DK059-Q(2)3DK0592
DK059-Q(4)3DK0594
DK059-Q(6)3DK0596
DK061-Q(2)3DK0612
DK061-Q(4)3DK0614
DK061-Q(6)3DK0616
DK062-Q(2)3DK0622
DK062-Q(4)3DK0624
DK062-Q(6)3DK0626
Note: Ball milling at 200 rpm, ball mass ratio large:medium:small = 20:20:100. Dispersant 3 g/150 mL. Sample name: “dispersant-Q(h)”. Blank (WFSJ): no dispersant.
Table 2. Preparation formula for grinding the down-conversion dispersion liquid.
Table 2. Preparation formula for grinding the down-conversion dispersion liquid.
Sample
Name		Dispersant Content
/g		Dispersant
Type		Ball Milling Time/hSand
Milling
Time/min
WFSJ-Q(2)-S[15]3Without dispersant215
WFSJ-Q(2)-S[30]3Without dispersant230
WFSJ-Q(4)-S[15]3Without dispersant415
WFSJ-Q(4)-S[30]3Without dispersant430
WFSJ-Q(6)-S[15]3Without dispersant615
WFSJ-Q(6)-S[30]3Without dispersant630
2020-Q(2)-S[15]3EAK2020215
2020-Q(2)-S[30]3EAK2020230
2020-Q(4)-S[15]3EAK2020415
2020-Q(4)-S[30]3EAK2020430
2020-Q(6)-S[15]3EAK2020615
2020-Q(6)-S[30]3EAK2020630
YCK-Q(2)-S[15]3YCK-2601215
YCK-Q(2)-S[30]3YCK-2601230
YCK-Q(4)-S[15]3YCK-2601415
YCK-Q(4)-S[30]3YCK-2601430
YCK-Q(6)-S[15]3YCK-2601615
YCK-Q(6)-S[30]3YCK-2601630
DK059-Q(2)-S[15]3DK059215
DK059-Q(2)-S[30]3DK059230
DK059-Q(4)-S[15]3DK059415
DK059-Q(4)-S[30]3DK059430
DK059-Q(6)-S[15]3DK059615
DK059-Q(6)-S[30]3DK059630
DK061-Q(2)-S[15]3DK061215
DK061-Q(2)-S[30]3DK061230
DK061-Q(4)-S[15]3DK061415
DK061-Q(4)-S[30]3DK061430
DK061-Q(6)-S[15]3DK061615
DK061-Q(6)-S[30]3DK061630
DK062-Q(2)-S[15]3DK062215
DK062-Q(2)-S[30]3DK062230
DK062-Q(4)-S[15]3DK062415
DK062-Q(4)-S[30]3DK062430
DK062-Q(6)-S[15]3DK062615
DK062-Q(6)-S[30]3DK062630
Note: All samples were first ball-milled for 2, 4, or 6 h (200 rpm, unidirectional, ball mass ratio large:medium:small = 20:20:100), followed by sand milling for 15 or 30 min (2500 rpm, 0.3 mm zirconia beads). Sample names follow the pattern “dispersant-Q(ball milling time)-S[sand milling time]”, where “Q” stands for ball milling and “S” for sand milling. Dispersant content was fixed at 3 g per 150 mL of ethanol/water mixture (140 mL ethanol + 10 mL deionized water). The blank group (WFSJ) contained no dispersant.
Table 3. Average transmittance of the DCAR and DCSAR coatings.
Table 3. Average transmittance of the DCAR and DCSAR coatings.
Sample
Number		Sample
Name		Average Transmittance from 380 nm to 780 nm/%
1Glass88.97
2AR589.97
3AR890.37
4AR1090.96
52‰ DCAR590.27
62‰ DCAR890.93
72‰ DCAR1090.68
85‰ DCAR590.52
95‰ DCAR890.29
105‰ DCAR1091.07
111% DCAR589.91
121% DCAR888.37
131% DCAR1089.81
95‰ DCAR890.29
142‰ DCSAR590.80
152‰ DCSAR891.17
162‰ DCSAR1091.02
175‰ DCSAR590.60
185‰ DCSAR890.05
195‰ DCSAR1091.02
201% DCSAR590.58
211% DCSAR890.02
221% DCSAR1090.12
Note: AR5/8/10 = pure anti-reflection coatings dip-coated at withdrawal speeds of 5, 8, and 10 mm/s, respectively. DCAR = coatings containing unmodified DK062-Q(6)-S[30] nanophosphors. DCSAR = coatings containing KH570-modified DK062-Q(6)-S[30]-g-k nanophosphors. Doping concentrations (2‰, 5‰, 1%) are mass ratios of phosphor to anti-reflection sol. Transmittance was measured using an Agilent Cary 5000 UV-Vis-NIR spectrophotometer.
Table 4. Average and peak transmittance of the single-layer light-conversion anti-reflection photovoltaic glass.
Table 4. Average and peak transmittance of the single-layer light-conversion anti-reflection photovoltaic glass.
Sample No.Average Transmittance (%)Peak Transmittance (%)
CG-5-194.2095.16
CG-5-294.3195.28
CG-5-394.2595.34
CG-15-193.1694.19
CG-15-293.2994.32
CG-15-393.1594.26
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Xu, D.; Cao, D.; Shan, Z.; Fang, L. Development of a Synthetic Optical Coating for Efficient UV Light Conversion and Enhanced Transmittance. Coatings 2026, 16, 692. https://doi.org/10.3390/coatings16060692

AMA Style

Xu D, Cao D, Shan Z, Fang L. Development of a Synthetic Optical Coating for Efficient UV Light Conversion and Enhanced Transmittance. Coatings. 2026; 16(6):692. https://doi.org/10.3390/coatings16060692

Chicago/Turabian Style

Xu, Daolong, Daruo Cao, Zihan Shan, and Liang Fang. 2026. "Development of a Synthetic Optical Coating for Efficient UV Light Conversion and Enhanced Transmittance" Coatings 16, no. 6: 692. https://doi.org/10.3390/coatings16060692

APA Style

Xu, D., Cao, D., Shan, Z., & Fang, L. (2026). Development of a Synthetic Optical Coating for Efficient UV Light Conversion and Enhanced Transmittance. Coatings, 16(6), 692. https://doi.org/10.3390/coatings16060692

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Article metric data becomes available approximately 24 hours after publication online.
Back to TopTop