利用LCD 3D打印模具制备的低成本几何可调PDMS人工晶状体用于犬眼科应用。 机翻

Summary

This study demonstrates the feasibility of fabricating geometry-tunable polydimethylsiloxane (PDMS) intraocular lens (IOL) prototypes using consumer-grade liquid crystal display (LCD) 3D-printed molds and PDMS casting. A reusable mold fabrication strategy incorporating a PMMA barrier layer successfully eliminated PDMS curing inhibition while preserving the mold geometry for repeated casting. The fabrication workflow accurately replicated a plate-haptic lens with a 6 mm optic and dimensional deviation below 1.5%, while enabling optical power to be tuned from 18 to 42 D by varying mold curvature alone. The fabricated PDMS IOLs exhibited high visible-light transparency (>95%), preliminary focusing ability, resolving power and image contrast comparable to a commercial IOL, acceptable post-treatment cytocompatibility, successful injector-assisted delivery with recovery of the original geometry, and preliminary hydrolytic stability and leachability performance. These results establish a reproducible, low-cost, and customizable fabrication platform for PDMS intraocular lenses and provide a foundation for future optimization toward customizable veterinary ophthalmic applications.

Keywords: intraocular lens, polydimethylsiloxane, PDMS, LCD 3D printing, veterinary ophthalmology, optical power tuning

Introduction

Globally, number of national organizations like Canadian National Institute for the Blind (CNIB) in Canada and Thailand Association of the Blind (TAB) in Thailand provide guide dogs to visually impaired individuals so that they can lead a more normal life in our society. However, on the flip side, these dogs are now prone to blindness, which can have significant impact in the availability of guide dogs for visually impaired individuals. Cataracts are one of the most common causes of visual impairment and blindness in dogs, characterized by progressive opacification of the crystalline lens that obstructs light transmission to the retina. In a cross-sectional study of 2,000 dogs, cataract prevalence reached 50% at 9.4 ± 3.3 years, and all dogs older than 13.5 years had some degree of lens opacity.1 Cataract formation in dogs is influenced by multiple factors, including aging, hereditary conditions, and metabolic disorders, which can accelerate the opacification of the lens.2 Progressive lens opacity can lead to severe visual impairment, lens-induced uveitis, glaucoma, and irreversible blindness, substantially affecting the quality of life and daily functioning of affected animals.2,3 Hence, there is an urgent need to address this challenge.

Cataracts are therefore considered the leading cause of preventable blindness in dogs,3 and surgical removal of the opaque lens through phacoemulsification followed by intraocular lens (IOL) implantation has been reported to restore functional vision in up to 79.2% of cases.2 IOL implantation significantly reduces postoperative glare and restores depth perception compared to aphakic outcomes, thereby improving postoperative visual performance.4 Despite these clinical advantages, the accessibility and performance of veterinary IOLs remain dependent on material selection, optical design, and manufacturing precision, all of which directly influence surgical outcomes and long-term biocompatibility.

Conventional IOL materials, including polymethyl methacrylate (PMMA) and silicone elastomers,5 and acrylic polymers,6 are widely used due to their optical clarity and established clinical performance.5,6 However, these materials present inherent trade-offs among optical transparency, mechanical flexibility, foldability, and long-term biocompatibility.6,7 In particular, rigid materials such as PMMA require larger incision sizes,6,8 while softer polymers may suffer from surface-related complications, protein adsorption, cellular adhesion,9 and inflammatory deposits, which can affect long-term optical performance.5,7,10 These limitations highlight the need for alternative materials that can balance optical quality, mechanical compliance, and biocompatibility for veterinary applications.

Current IOL fabrication relies on precision molding, lathing, or photolithography using PMMA, acrylic copolymers, or silicone elastomers. These manufacturing processes are largely constrained by material-specific requirements and limited design flexibility. Veterinary-specific designs are further constrained by limited manufacturers, long lead times, and a lack of customization options. Commercially available lenses remain limited in range and largely standardized, restricting the ability to tailor optical properties to individual anatomical variations.

Recent advances in additive manufacturing offer promising, cost-effective approaches for the fabrication of optical components.11,12 However, directly printed polymer lenses are inherently limited by the available printable photopolymer or photoresponsive resin systems.8 In addition, printed surface roughness can degrade optical quality.13,14 An alternative approach is to use 3D printing for mold fabrication, followed by casting with optically suitable materials such as polydimethylsiloxane (PDMS), thereby separating geometric design from material functionality.15,16 In the present study, this mold-assisted strategy was applied to fabricate geometry-tunable PDMS intraocular lens prototypes, allowing PDMS to be used as the final lens material while retaining the design flexibility of 3D-printed mold fabrication, which enabled optical power tuning from 18 to 42 D by changing mold curvature.

In this work, a low-cost approach for PDMS-based intraocular lenses is developed to address key challenges in surface quality and manufacturing accessibility. The study utilizes consumer-grade liquid crystal display (LCD) 3D-printed molds with spin-coated surface smoothing to improve dimensional fidelity and optical surface quality. A monofocal 6 mm plate-haptic design is investigated, with emphasis on how geometric control and material formulation influence lens performance. Comprehensive characterization is conducted, including optical evaluation, imaging performance, mechanical testing, surface wettability, and in vitro cytotoxicity assessment. The results demonstrate the feasibility of a customizable fabrication platform for PDMS intraocular lenses with tunable optical properties. The fabricated prototypes exhibited promising preliminary optical, mechanical, and biological performance, providing a foundation for further optimization and future veterinary ophthalmic applications.

Results and discussion

Selection of IOL material

Materials for intraocular lens fabrication were selected to meet requirements for optical transparency, flexibility for small-incision delivery, moldability, and biocompatibility. For mold-cast lenses, the material must reproduce curved optical surfaces while maintaining acceptable biological performance after curing. Conventional IOL materials such as PMMA, acrylic, and silicone-based polymers provide established clinical performance but may limit foldability, customization, or surface quality depending on their processing routes.5,6 Surface properties, including roughness and wettability, can also influence cellular adhesion and posterior capsule opacification, requiring smooth replication and reliable characterization.7

Polydimethylsiloxane (PDMS) was selected due to its optical transparency,17 established biomedical use,18,19 elastomeric flexibility,19,20 and compatibility with replica molding from 3D-printed molds.15,21 Its flexibility is particularly advantageous for foldable IOLs, enabling implantation through smaller incisions compared with rigid lenses.

The Sylgard 184 PDMS (Dow Corning, USA) base-to-curing-agent weight ratio was varied at 10:1, 15:1, and 20:1 to control crosslink density and mechanical properties while maintaining optical transparency. Increasing the base ratio reduces crosslink density, resulting in a softer elastomer network with lower elastic modulus and higher compliance. The 10:1 formulation is expected to provide higher stiffness and structural stability, while the 15:1 and 20:1 formulations are anticipated to exhibit increased flexibility, which is advantageous for folding during implantation. However, reducing crosslink density may also influence other material characteristics,22 including dimensional stability, surface fidelity, and biological response. Therefore, a comprehensive evaluation of optical performance, mechanical behavior, surface properties, and cytotoxicity is required to identify a formulation that provides an appropriate balance for intraocular lens applications.

Optical and geometrical design of the canine intraocular lenses

A plate-haptic configuration, in which the optic of the lens is supported by broad, plate-like extensions rather than flexible loop structures, is selected for fixation within the capsular bag to support stable positioning and optical alignment of the lens. The IOL geometry is illustrated in Figure 1.

[图：Geometry of the designed plate-haptic intraocular lens (A) Top view showing the plate-haptic geometry, 12 mm overall len]

Plate-haptic IOLs demonstrate significantly lower postoperative decentration (0.12 ± 0.06 mm) and tilt than C-loop haptic designs (0.22 ± 0.13 mm), improving optical alignment within the capsular bag.23 Haptic thickness is set at 1 mm to provide sufficient mechanical support, while maintaining the flexibility required for folding during implantation, thereby enabling insertion through a reduced surgical incision.

Sharp-edge geometries are incorporated at both the optic and haptic periphery. A sharp optic edge creates a distinct bend in the posterior capsule that mechanically impedes migration of residual lens epithelial cells (LECs), reducing the risk of posterior capsule opacification (PCO).24 For the haptic periphery, a controlled edge geometry is introduced to regulate mechanical interaction with the capsular bag. Excessively sharp edges may lead to localized stress concentration and potential tissue irritation, whereas overly rounded geometries may reduce mechanical engagement and positional stability. The proposed edge design therefore aims to achieve sufficient mechanical anchoring while maintaining stable and consistent positioning within the capsular bag. Both features are also designed to be compatible with PDMS mold-based fabrication, which requires geometries that can be reliably replicated and demolded without deformation or damage.

Mold geometry and surface optimization

A mold was developed to enable accurate replication of the intraocular lens geometry and to facilitate controlled fabrication of PDMS-based lenses. The design incorporates the 6 mm optic cavity, plate-haptic profile, and sharp-edge features. Optical cavity surfaces were modeled as spherical geometries corresponding to the target optic curvatures. All designs were generated using computer-aided design (CAD) software and exported as STEP files for additive manufacturing.

The mold design is illustrated in Figure 2A. A two-part mold configuration was used to enable complete lens cavity replication and to facilitate reliable demolding after PDMS curing. Two channels were integrated into the mold, one serves as the injection port for the PDMS precursor, and the second acts as an air vent to minimize trapped air and reduce bubble formation during cavity filling. The two parts of the mold were secured within a rigid holder to ensure consistent clamping and maintain alignment during casting as shown in Figure 2B.

[图：Intraocular lens mold design (A) Top view, cross-sectional view, and 3D rendering of the mold design, showing the optica]

The surface morphology of the 3D-printed molds, which directly influences the optical quality of the replicated PDMS lenses, was evaluated by confocal laser scanning microscopy (OLS5100, Olympus Corporation, Japan) before and after spin-coating a thin UV-curable resin layer to smooth the mold surface, as shown in Figure 3.

[图：Confocal microscopy of 3D-printed mold surfaces before and after UV resin spin coating Reflected-intensity images (top) ]

Before spin-coating with the UV-curable resin layer, the molds exhibited radially periodic surface features arising from the layer-by-layer photopolymerization process.25 These features appeared in reflected-intensity images as ring-like patterns and in height-profile images as periodic topographic variations, corresponding to microscale roughness. Such microscale roughness can degrade the optical quality of fabricated lenses.14,26

After spin-coating a thin UV-curable resin layer at 1,000, 2,000, and 3,000 rpm to reduce microscale roughness, the periodic surface features were no longer observed in the height profiles, indicating effective surface smoothing. However, stripe-like bands appeared in the reflected-intensity images. These bands are associated with thin-film optical interference arising from variations in the coating thickness, rather than surface roughness. As the spin speed increased from 1,000 to 3,000 rpm, the fringe spacing decreased, indicating the formation of a thinner resin layer at higher rotational speeds. The localized aberration observed at the center of the 3,000 rpm condition suggests insufficient coating coverage, where the coating layer was likely too thin to completely smooth residual defects from the 3D-printed mold surface.

To quantify the effect of spin-coating on surface quality and geometric fidelity, cross-sectional profiles, curvature measurements, and surface roughness were analyzed. The cross-sectional profiles in Figure 4A illustrate changes in the surface profile with spin speed, primarily reflecting variations in coating thickness and overall geometry. The curvature measurements in Figure 4B show an increase in the radius of curvature obtained from the height profiles at lower spin speeds, indicating a deviation from the designed geometry toward a flatter optical surface due to the thicker coating layer. The surface roughness (Ra) values in Figure 4C quantify the degree of smoothing, confirming that lower spin speeds yield reduced roughness. Together, these results reveal a clear trade-off between surface smoothing and curvature preservation.

[图：Quantitative analysis of mold surface and geometry before and after UV resin spin coating, and Raman characterization of]

The surface roughness was reduced after spin-coating the resin, while maintaining optical transparency, indicating that the resulting lenses are suitable for optical evaluation. A spin speed of 1,000 rpm was selected as the optimal balance between surface smoothing and curvature preservation, yielding a mold surface roughness (Ramold) of 0.40 μm and a PDMS intraocular lens roughness (RaIOL) of 0.36 μm. These results demonstrate that the selected processing conditions provide sufficient surface quality for subsequent optical evaluation while maintaining the designed lens geometry.

To further verify the PMMA barrier layer, which was introduced to prevent PDMS curing inhibition on the LCD 3D-printed resin mold, Raman spectroscopy (Xplora plus, Horiba, Japan) was performed to determine whether PMMA residue remained on the fabricated PDMS IOL after casting and demolding. Raman spectroscopy was performed on the PMMA reference, the PMMA-coated mold, and the fabricated PDMS IOL. The Raman spectrum of the fabricated PDMS IOL was dominated by characteristic PDMS bands, while the characteristic ester carbonyl (C=O) band of PMMA at approximately 1,725–1,735 cm−1 was not clearly observed, as shown in Figure 4D. This result suggests that the PMMA barrier layer was not detectably transferred to the final lens surface during casting and demolding.

Mold-to-lens replication fidelity

Following optimization of the mold surface, the fidelity of geometric and surface feature transfer to the PDMS intraocular lenses was evaluated. Geometric and surface parameters of the 1,000 rpm spin-coated mold and the replicated PDMS IOL are summarized in Table 1. In this analysis, Ra500, Ra1,000, and Ra2,000 denote surface roughness measured over evaluation lengths of 500, 1,000, and 2,000 μm, respectively, to capture surface features at different spatial scales. The fabricated lenses closely replicate the mold geometry, with thickness and curvature radius showing differences of 0.74% and 1.32%, respectively, indicating high dimensional accuracy. In addition to geometric agreement, the surface roughness of the PDMS lenses was consistently lower than that of the corresponding molds across all roughness evaluation lengths. This reduction confirms that the casting process does not introduce additional surface artifacts and instead contributes to further surface smoothing. This behavior is attributed to the conformal nature of PDMS casting, which allows the elastomer to fill surface features while reducing small-scale irregularities during curing.

Fabricated PDMS intraocular lens

The developed fabrication strategy combines LCD 3D-printed molds with surface modification and PDMS casting to produce the designed intraocular lenses, as summarized in Figure 5 the printed mold surface was first smoothed using a UV-curable resin coating to reduce surface features arising from the layer-by-layer printing process and improve the optical quality of the resulting lens.

[图：Fabrication workflow of PDMS intraocular lenses, consisting of mold fabrication, PMMA surface modification, and PDMS len]

Direct casting of PDMS against the photopolymer resin mold presents an additional challenge because phosphine oxide-based photoinitiators from the resin can interfere with the Pt-catalyzed hydrosilylation reaction of addition-cure PDMS, resulting in incomplete or non-uniform curing.27,28 PMMA interfacial barrier was therefore introduced between the printed mold and PDMS to minimize photoinitiator migration and enable complete crosslinking. Using the modified molds, PDMS IOLs were successfully obtained at base-to-curing-agent ratios of 10:1, 15:1, and 20:1. These formulations were investigated because variation in crosslink density influences the mechanical properties of the cured elastomer.29 The fabricated lenses retained the designed plate-haptic geometry after demolding and were subsequently evaluated for their dimensional, optical, mechanical, and biological characteristics.

The mold-based approach also separates the optical geometry from the final lens material. Lens geometry can therefore be modified digitally without changing the PDMS formulation, while the molds can be fabricated in parallel and reused for repeated casting. This provides a flexible platform for producing geometry-tunable PDMS IOLs.

Optical transparency

Following the validation of geometric and surface replication, the optical performance of the fabricated lenses was evaluated, with particular focus on transparency. The fabricated PDMS IOL is shown in Figure 6A, exhibiting a clear, transparent appearance with a smooth optical surface and well-defined geometry. To evaluate the effect of the base-to-curing-agent ratio on optical transparency, PDMS lenses prepared at 10:1, 15:1, and 20:1 (w/w) were examined. Optical transparency was assessed by UV-vis spectrophotometry (Lambda 850+, PerkinElmer, USA), as shown in Figure 6B.

[图：Optical transparency of fabricated PDMS intraocular lenses (A) Fabricated PDMS IOL. (B) UV-vis transmission spectra of P]

All PDMS formulations exhibited broadband visible-light transmission exceeding 95% across 400–700 nm, with no significant variation among ratios. The high transmission is consistent with the intrinsic visible-light transparency of PDMS.17 The reduced transmission observed in the UV region is an inherent characteristic of PDMS and can be advantageous for intraocular lens applications, as it can help limit UV exposure to ocular tissues. No significant trend in optical transmission with base-to-curing-agent ratio was observed, indicating that high transparency is consistently maintained across all formulations, supporting their suitability for intraocular lens applications.

Optical power and tunability

To further evaluate the optical performance of the fabricated lenses, their focusing behavior and optical power were characterized. An optical setup was developed to measure the axial intensity distribution and determine focal positions, as illustrated in Figure 7A. A collimated light source from a fiber-coupled LED (M530F3, Thorlabs, USA) was directed through a neutral density (ND) filter, which was used to control the light intensity and ensure appropriate exposure for accurate beam profiling. The beam was then passed through the PDMS lens mounted in a liquid-filled cuvette containing balanced salt solution (BSS, n = 1.336) to simulate the refractive environment of the eye. The actual experimental setup is shown in Figure 7B, with an enlarged view highlighting the lens positioning within the cuvette. The transmitted beam was recorded along the optical axis using a CMOS camera system (HY-500M, Hayear, China), enabling extraction of the focal position from the axial intensity profile.

[图：Optical characterization of fabricated PDMS intraocular lenses (A) Schematic of the optical setup for evaluating the foc]

Using this setup, PDMS lenses with five different curvature radii (R = 3.8, 5.5, 7.3, 8.7, and 9.6 mm) were evaluated to investigate the tunability and reproducibility of optical power. For each curvature radius, 5 lenses were fabricated and measured under identical conditions. Axial normalized intensity profiles were measured to determine the focal positions, as shown in Figure 7C, where the peak intensity along the optical axis was used to obtain the focal length (f) and calculate optical power (D = 1/f). A systematic shift in focal position was observed with changing curvature, where smaller radii produced focal points closer to the lens, corresponding to shorter focal lengths. The lens with R = 3.8 mm exhibited a focal position of approximately 24 mm, corresponding to an optical power of ∼42 D. Overall, the fabricated lenses exhibited optical powers ranging from 18 to 42 D, demonstrating tunability through geometric design, with smaller radii producing higher optical power, as shown in Figure 7D.

To assess the accuracy of the measured optical performance, the results were compared with theoretical predictions based on the lensmaker’s equation,30 For the symmetric biconvex PDMS IOL used in this study, the theoretical optical power (D) in balanced salt solution (BSS) can be expressed as

where nPDMS and nBSS are the refractive indices of PDMS (1.45) and BSS (1.336), respectively, R is the radius of curvature, and d is the central lens thickness (3.82 mm). As shown in Figure 7D, close agreement between measured and theoretical values confirm the reliability of both the fabrication process and the optical measurement approach. This indicates that the designed curve-mold geometry is effectively transferred to the PDMS lenses. This performance meets the clinically required range for canine pseudophakic correction. A key advantage of this fabrication approach is that optical power can be controlled through a single geometric parameter, enabling the production of patient-specific lenses spanning 18 to 42 D without modifying the material system or processing conditions. This flexibility is especially relevant for addressing variations in axial eye length and corneal curvature across different canine breeds.

To further evaluate the image-forming performance of the fabricated PDMS IOL, a USAF 1951 resolution target was employed as shown in Figure 7E. Representative images acquired through the commercial IOL and the fabricated PDMS IOL are shown in Figures 7F and 7G, respectively. Both lenses successfully resolved group 4, element 2, corresponding to a spatial frequency of 20.16 line pairs/mm, which was the highest spatial-frequency target evaluated in this study. Image contrast was quantified using the Michelson contrast (C).

where Imax and Imin are the maximum and minimum intensity, respectively. The fabricated PDMS IOL exhibited a Michelson contrast of 40.6 ± 1.05%, which was comparable to that of the commercial IOL (41.2% ± 1.55%). These results demonstrate that the fabricated PDMS IOL preserves fine spatial features while maintaining image contrast comparable to that of a commercial IOL under the tested imaging conditions.

Imaging performance

To demonstrate real-image formation, a PDMS lens was fabricated with a larger curvature radius (R = 20 mm) to provide a longer focal length suitable for bench-top imaging in air. The lens was mounted in a 3D-printed holder directly in front of the image sensor of a Sony NEX-5T APS-C camera with the camera lens removed, forming a bare-sensor imaging configuration. Four scenes were captured at different object distances, as shown in Figure 8, alongside reference images captured using a smartphone under similar lighting conditions.

[图：Comparison of imaging performance The fabricated PDMS IOL (top) and a smartphone reference (bottom).]

The PDMS IOL produced clearly recognizable images with distinguishable shapes, colors, and text. Color fidelity was preserved, consistent with the high optical transparency reported in Section 3.3. No visible haze or significant scattering was observed, confirming that the fabricated lens surfaces are suitable for image-forming applications. This qualitative experiment confirmed that the fabricated PDMS intraocular lenses can function as effective image-forming elements.

Surface wettability

Surface wettability was evaluated to assess the interfacial properties of the fabricated PDMS intraocular lenses, which can influence interactions with the ocular environment. Static water contact angles were measured as 110.59 ± 4.81°, 114.58 ± 5.14°, and 108.32 ± 3.20° for PDMS IOLs prepared at base-to-curing-agent ratios of 10:1, 15:1, and 20:1, respectively, as shown in Figures 9A–9C, indicating consistently hydrophobic surfaces. These values are consistent with the intrinsic hydrophobicity of crosslinked PDMS and are comparable to those of silicone-based intraocular lenses.7,31

[图：Static water contact angle measurements of fabricated PDMS intraocular lenses (A) 10:1, (B) 15:1, and (C) 20:1 base-to-c]

Mechanical properties

The mechanical properties of the fabricated PDMS intraocular lenses were first evaluated under compressive loading using a tensile tester (EZ-5, Shimadzu, Japan). This test assessed whether the lenses retained adequate structural stability while maintaining sufficient flexibility for folding during implantation. The results, including maximum compressive stress, strain, and elastic modulus, are summarized in Table 2. The maximum compressive stress was similar across all formulations, remaining around 22 MPa. In contrast, the elastic modulus decreased with increasing base-to-curing-agent ratio, with an overall reduction of approximately 49% from 10:1 to 20:1. Correspondingly, the maximum compressive strain increased, indicating greater flexibility at higher base ratios. These results show that the compressive mechanical response of the fabricated lenses can be tuned by adjusting the PDMS base-to-curing-agent ratio.

Although compression testing provides information on bulk mechanical behavior, implantation of a foldable intraocular lens also involves substantial shear deformation as the lens passes through the injector cartridge. Therefore, oscillatory shear rheological measurements were additionally performed to evaluate the viscoelastic response of the fabricated PDMS IOLs under implantation-relevant deformation.

To further evaluate the viscoelastic properties of the fabricated PDMS intraocular lenses under implantation-relevant shear deformation, oscillatory rheological measurements were performed using a rheometer (HAAKE MARS 40, Thermo Fisher Scientific, Germany) in shear mode at 37°C and 1 Hz. The representative rheological parameters are summarized in Table 3. At low strain, all PDMS formulations exhibited storage modulus (G′) values higher than loss modulus (G″), indicating predominantly elastic behavior. The 10:1 formulation exhibited the highest initial G′ (61.06 kPa), followed by the 15:1 (53.48 kPa) and 20:1 (24.44 kPa) formulations, reflecting the higher crosslink density associated with increasing curing-agent content. In the high-strain region (approximately 50%–60% strain), the 15:1 formulation retained the highest G′ (27.97 kPa), compared with 13.20 kPa and 15.91 kPa for the 10:1 and 20:1 formulations, respectively. The 15:1 formulation also exhibited the highest G″ (15.35 kPa) in this region, indicating a balanced viscoelastic response under large deformation, with the ability to maintain its structure while dissipating deformation energy. Together with the compression analysis, these findings demonstrate that the mechanical behavior of the fabricated PDMS IOLs can be tailored by adjusting the base-to-curing-agent ratio.

These results show that the mechanical response of PDMS IOLs can be adjusted through the base-to-curing-agent ratio. Although the 15:1 formulation provided the most stable shear response at high strain, the 10:1 formulation remains the most suitable candidate among the tested formulations when both mechanical performance and cytocompatibility are considered.

To further assess the practical deformation response of the selected 10:1 PDMS IOL during implantation, an injector-delivery demonstration was performed using a standard commercial IOL injector cartridge as shown in Figure 10, the fabricated plate-haptic PDMS IOL was first loaded into the injector cartridge, positioned within the loading chamber, folded prior to delivery, and subsequently delivered through the injector. Following injector-assisted delivery, the lens recovered its overall plate-haptic geometry without visible cracking, tearing, or gross structural damage, demonstrating its ability to tolerate the combined compression, folding, shear, and frictional stresses encountered during the implantation process. The complete injector-assisted delivery procedure is provided in Video S1.

[图：Injector-assisted delivery of the fabricated PDMS intraocular lens (A) Fabricated PDMS IOL before loading into a standar]

[内容请查看原文]

In vitro cytotoxicity

To ensure the suitability of the fabricated PDMS intraocular lenses for implantation, their biocompatibility was evaluated through in vitro cytotoxicity testing. For each PDMS formulation, five fabricated lenses (n = 5) were used for extract preparation. Extracts were prepared in accordance with ISO 10,993–12. Cell viability was interpreted according to ISO 10,993–5 guidance, where cell viability below 70% is considered indicative of a cytotoxic effect under the tested extract conditions.32 Cell viability results for the different PDMS formulations are presented in Figure 11.

[图：In vitro cytotoxicity of fabricated PDMS intraocular lenses before and after post-treatment Cell viability was evaluated]

As shown in Figure 11, the control group showed 100.00% cell viability. Among the PDMS formulations, the 10:1 base-to-curing-agent ratio showed 83.97% ± 15.35% cell viability, exceeding the 70% threshold and indicating acceptable cytocompatibility. In contrast, the 15:1 and 20:1 formulations showed lower cell viability values of 57.61% ± 8.17% and 60.48% ± 5.13%, respectively, suggesting potential cytotoxic effects under the tested conditions. The observed differences among formulations are attributed to the presence of residual uncrosslinked PDMS oligomers. At higher base-to-crosslinker ratios, reduced crosslinking leads to a greater fraction of low-molecular-weight chains in the network. These oligomers can interact with cell membranes and induce cytotoxic responses,33 which is consistent with the lower viability observed for the 15:1 and 20:1 formulations.

Following post-treatment by rinsing with isopropyl alcohol (IPA) and deionized (DI) water, all formulations exhibited cell viabilities above the ISO 10,993–5 acceptance criterion (70%), with values of 97.78% ± 3.29%, 99.68% ± 11.48%, and 86.19% ± 10.73% for the 10:1, 15:1, and 20:1 formulations, respectively. These results indicate that the initial reduction in cell viability was primarily associated with removable extractable residues or uncured surface components rather than the cured PDMS network itself.

Hydrolytic stability test

The hydrolytic stability test was performed in accordance with ISO 11,979-534 to evaluate the stability of PDMS IOLs under accelerated aqueous aging. Samples were incubated in BSS at 95°C for 30 days, corresponding to approximately five years of physiological aging.34 After accelerated aging, the PDMS IOLs maintained their overall lens structure, although localized surface alterations, including pitting-like features, dark deposits/discoloration, and line-like surface defects, were observed as shown in Figure 12. In addition, the visible-light transmittance of the aged PDMS IOLs remained 80.94% ± 11.91%, which was comparable to that of the commercial IOL (82.27%), indicating that the optical transparency was largely preserved following accelerated hydrolytic aging. These results indicate preliminary structural stability of the fabricated PDMS IOLs after five-year equivalent hydrolytic aging, although further long-term stability testing remains necessary before clinical application.

[图：Confocal microscopy images of the fabricated PDMS intraocular lens (A) before and (B) after accelerated hydrolytic aging]

Leachability test

The leachability test was performed in accordance with ISO 11,979–5 to evaluate potential compounds released from PDMS IOLs under simulated extraction conditions. Samples were extracted in distilled water, BSS, and toluene at 37°C for 72 h to represent aqueous and lipophilic extraction environments. The extracted PDMS IOLs maintained high optical transmittance after exposure to all extraction media. The visible-light transmittance remained 91.85% ± 11.47%, 99.09% ± 4.99%, and 96.75% ± 8.97% following extraction in DW, BSS, and toluene, respectively, with no statistically significant differences among the extraction media (p = 0.47). Furthermore, the extraction media exhibited consistently low UV-vis absorbance without distinct absorption peaks, indicating minimal evidence of UV-vis-detectable leachable compounds under the extraction conditions investigated. These findings suggest that the fabricated PDMS IOLs maintained their optical properties after extraction in both aqueous and lipophilic media while exhibiting minimal release of UV-vis-detectable leachable species.

Limitations of the study

The fabricated PDMS IOL presented in this work is intended as a prototype demonstrating the feasibility of a customizable fabrication platform rather than as a replacement for commercially available intraocular lenses. Compared with a commercial IOL, the fabricated PDMS IOL demonstrated comparable preliminary performance in several evaluations performed in this study, including image resolution, Michelson contrast, optical transparency following accelerated hydrolytic aging, and injector-assisted delivery. Nevertheless, several important challenges remain before clinical application, including further optimization of surface quality, long-term mechanical durability under cyclic loading, comprehensive biological evaluation, and in vivo validation. Despite these limitations, the proposed fabrication strategy enables rapid digital customization of lens geometry without dedicated tooling. Reusable molds allow repeated PDMS casting with tunable optical power (18–42 D) through simple geometric modification, facilitating rapid design iteration and providing a practical platform for producing patient-specific intraocular lenses. Overall, this work establishes a customizable and scalable fabrication platform for PDMS-based intraocular lenses and provides a foundation for the future development of ophthalmic implants.

Resource availability

Lead contact

Requests for further information and resources should be directed to and will be fulfilled by the lead contact, Charusluk Viphavakit (charusluk.v@chula.ac.th).

Materials availability

This study did not generate new unique reagents.

Data and code availability

Acknowledgments

This research is financially supported by the Second Century Fund (C2F), Thailand Science research and Innovation Fund Chulalongkorn University (IND_FF_69_091_2100_014), 111th Anniversary Engineering Research Catalyst Fund Toward U Top 100 (038/2569), National Research Council of Thailand (NRCT), and Chulalongkorn University (N42A690579), Rachadapisek Sompote Fund for Biomedical Materials and Devices for Revolutionary Integrative Systems Engineering Research Unit (BMD-RISE), Chulalongkorn University, Thailand.

Author contributions

S.R. writing – original draft, validation, methodology, investigation, formal analysis, data curation; A.C. data curation; T.L. data curation; N.T., O.J., and S.K.M. supervision, writing – review and editing; C.V. writing – review and editing, supervision, project administration, methodology, funding acquisition, conceptualization.

Declaration of interests

The authors declare no competing interests.

STAR★Methods

Key resources table

Method details

Mold fabrication and surface preparation

Molds were fabricated using an Elegoo Mars 5 Ultra LCD stereolithography printer (Elegoo, China) operating at 405 nm with an 18 μm pixel resolution and UV photopolymer resin (8K standard resin, Elegoo, China). After printing, the molds were cleaned by sequential immersion in ethanol to remove residual uncured resin. A thin layer of UV-curable resin was spin-coated onto the mold surface at 1,000 rpm for 60 s to reduce surface roughness arising from the layer-by-layer printing process. The molds were subsequently UV post-cured for 1 h and thermally treated at 100°C for 1 h to complete photopolymerization and remove residual photoinitiator.

PMMA barrier coating

Following surface preparation, a PMMA barrier layer was applied to the mold surface to prevent curing inhibition of the PDMS. A 4% (w/w) PMMA solution in anhydrous toluene was spin-coated onto the mold at 1,000 rpm for 30 s, followed by thermal treatment at 100°C for 30 min.

PDMS casting and curing

PDMS lenses were fabricated using Sylgard 184 at base-to-curing-agent ratios of 10:1, 15:1, and 20:1 (w/w). Each formulation was thoroughly mixed and vacuum-degassed (<600 mmHg) for 1 h to remove entrapped air. The PDMS precursor was then cast into the assembled mold cavity and thermally cured in a convection oven at 100°C for 2 h. After cooling to room temperature, the lenses were carefully demolded and prepared for subsequent characterization.

Cytotoxicity testing

In vitro cytotoxicity testing was performed using L929 mouse fibroblast cells at the Biomaterial Testing Center, Faculty of Dentistry, Chulalongkorn University. Five fabricated lenses from each PDMS formulation were used for extract preparation. For the post-treatment group, fabricated lenses were rinsed with isopropyl alcohol for 1 min followed by deionized water for 1 min before extraction. Cell viability was interpreted according to ISO 10993–5, where a reduction in cell viability of more than 30% relative to the control was considered indicative of cytotoxicity.

Hydrolytic stability testing

Hydrolytic stability testing was performed according to ISO 11979-5. Fabricated PDMS IOLs were incubated in BSS at 95°C for 30 days, corresponding to approximately five years of physiological aging. PDMS IOLs were extracted in 3 mL BSS, while commercial IOL controls were extracted in 0.3 mL BSS. After aging, lens morphology was evaluated by confocal microscopy at 100× magnification and compared with commercial IOLs under the same aging condition.

Leachability testing

Leachability testing was performed according to ISO 11979–5. Fabricated PDMS IOLs were extracted in distilled water, BSS, and toluene at 37°C for 72 h. PDMS IOLs were extracted in 3 mL of each medium, while commercial IOL controls were extracted in 0.3 mL of each medium. Solvent blanks were prepared under the same conditions. After extraction, the extraction media and extracted lenses were analyzed by UV/Vis spectrophotometry from 300 to 900 nm. The extraction media were assessed for UV/Vis-detectable leachable compounds, while the extracted lenses were evaluated for changes in optical transparency.

Quantification and statistical analysis

Quantitative data are reported as mean ± standard deviation (SD), unless otherwise stated. The number of replicates (n) is indicated in the corresponding figure legends, table captions, or method descriptions. For optical power measurements, n represents independently fabricated lenses measured at each curvature radius, whereas for cytotoxicity testing, n represents independently fabricated lenses used for extract preparation for each PDMS formulation. For other measurements, n represents independent samples or repeated measurements, as specified in the corresponding sections.

Footnotes

[内容请查看原文]

References

[内容请查看原文]

Associated Data

Supplementary Materials

[内容请查看原文]

Data Availability Statement