Systems and methods for monitoring curing progress of photocurable polymer during additive manufacturing

Abstract

Systems and methods for monitoring curing progress of curing of a photocurable polymer formation during an additive manufacturing process is are described A system for monitoring curing progress of a photocurable polymer formation during an additive manufacturing process comprises: a transparent stage configured to support the photocurable polymer formation; a source of ultra violet radiation configured to illuminate a target region of the photocurable polymer formation with incident ultra violet radiation; an ultra violet radiation detector arranged on an opposing side of the transparent stage from the source of ultra violet radiation, the ultra violet radiation detector configured to detect an intensity of transmitted ultraviolet radiation transmitted through the target region of the photocurable polymer formation; and a data storage device storing a machine learning model trained to predict a degree of cure of a photocurable polymer from input data comprising an indication of ultraviolet transmission of the photocurable polymer.

Description

[0001] SYSTEMS AND METHODS FOR MONITORING CURING PROGRESS OF PHOTOCURABLE POLYMER DURING ADDITIVE MANUFACTURING

[0002] TECHNICAL FIELD

[0003] The present disclosure relates to additive manufacturing. In particular, the present disclosure relates to monitoring of a curing progress of photocurable polymer during an additive manufacturing process.

[0004] BACKGROUND

[0005] In the field of tissue engineering, photocurable polymers including hydrogels such as Gelatin Methacryloyl (GelMA) have emerged as a critical component for bioprinting applications due to their inherent biocompatibility and tissue-like mechanical properties. Using GelMA to generate high fidelity 3D structures requires optimal application of ultraviolet light to fully crosslink and cure the structure. Yet, the optimization of hydrogel curing process presents a significant challenge. Traditional method for optimizing process parameters for GelMA curing requires iterative experimentations on interplaying parameters with strong correlations. Among these parameters, the regulation of UV light dose is crucial for photopolymerization. Curing of hydrogels by applying the UV light dynamically along specific trajectories or patterns rather than uniformly, which is often seen in bioprinter, adds to the complexity.

[0006] There is currently no method for real-time monitoring of the degree of curing for photocurable hydrogel, resulting in extensive resource allocation for post-process validation. The following methods for quantifying curing exist but have limitations.

[0007] In situ measurement of the refractive index. This method involves detecting the UV light scattered by the photopolymerizing resin. However, implementation of such a system is difficult to achieve due to the large size requirement preventing direct integration into currently available 3D printing systems. Infrared spectroscopy (NIR / MIR). Although this technique has been utilized to monitor the resin during curing, it presents substantial financial overhead due to the expensive nature of the system, which also complicates integration with 3D printers.

[0008] Electrical conductivity measurement. The introduction of an electrical current for monitoring can interfere with the print quality and may not be applicable for certain types of inks.

[0009] In-situ interferometric curing monitoring. Despite its accuracy, the setup is both large- scale and expensive, rendering it impractical to be integrated into an existing printer.

[0010] In-situ ultrasonic monitoring. This method requires ultrasonic transducers and sensors It has the potential to disrupt the printing process itself.

[0011] The collective issues with the aforementioned methods are their expensive cost, bulkiness, and the complexity associated with integrating them into an existing system. The present disclosure proposes a cost-effective alternative to the real-time measurement of the hydrogel curing process.

[0012] SUMMARY

[0013] According to a first aspect of the present disclosure, a system for monitoring progress of a photocurable polymer formation during an additive manufacturing process, the system comprising: a transparent stage configured to support the photocurable polymer formation; a source of ultra violet radiation configured to illuminate a target region of the photocurable polymer formation with incident ultra violet radiation; an ultra violet radiation detector arranged on an opposing side of the transparent stage from the source of ultra violet radiation, the ultra violet radiation detector configured to detect an intensity of transmitted ultra violet radiation transmitted through the target region of the photocurable polymer formation; and a data storage device storing a machine learning model trained to curing progress of a photocurable polymer from input data comprising an indication of ultraviolet transmission of the photocurable polymer. The system provides a UV power meter below a transparent printing stage, and is therefore simple and straightforward to integrate into an additive manufacturing system compared to other methods that typically requires an array of lenses or a complex spectrometer.

[0014] Utilizing a UV sensor and power meter substantially reduces the cost when compared to the more expensive spectrometric equipment.

[0015] As the process intrinsically uses UV light specific to the hydrogel's curing, the method does not introduce any invasive or potentially damaging elements to the printing process.

[0016] In an embodiment, the transparent stage is movable relative to the source of ultra violet radiation and the ultra violet radiation detector such that the target region of the photocurable polymer formation can be moved laterally relative to the source of ultra violet radiation and the ultra violet radiation detector.

[0017] In an embodiment, the ultra violet radiation detector comprises an aperture configured to allow ultra violet radiation transmitted through the target region of the photocurable polymer formation to enter the ultra violet radiation detector.

[0018] In an embodiment, the machine learning model is trained to predict a degree of cure of a photocurable polymer.

[0019] In an embodiment, the machine learning model comprises a random forest regression model, a support vector regression model or a deep neural network.

[0020] In an embodiment, the machine learning model is trained using one or more parameters selected from the group consisting of a formulation of the photocurable polymer, a concentration of one or more photoinitiators, a thickness of the photocurable polymer, a measured ultra violet intensity, or an applied ultra violet intensity. In an embodiment, the photocurable polymer formation comprises hydrogel.

[0021] The system for monitoring a degree of cure of a hydrogel formation may be integrated into an additive manufacturing system

[0022] According to a second aspect of the present disclosure, a method of monitoring progress of curing of a photocurable polymer formation during an additive manufacturing process is provided. The method comprises: illuminating a target region of the photocurable polymer formation with incident ultraviolet radiation; detecting an intensity of transmitted ultra violet radiation transmitted through the target region of the photocurable polymer formation; determining progress of curing of the target region of the photocurable polymer formation by inputting an indication of the intensity of transmitted ultraviolet radiation into a machine learning model trained to predict progress of curing of a photocurable polymer from input data comprising an indication of ultraviolet transmission of the photocurable polymer.

[0023] In an embodiment, the method further comprises determining an indication of the incident ultra violet radiation by illuminating an ultraviolet radiation detector without the photocurable polymer present.

[0024] In an embodiment, the photocurable polymer formation is arranged on a transparent stage and the method further comprises moving the transparent stage relative to a source of the ultra violet radiation.

[0025] In an embodiment, the photocurable polymer formation comprises hydrogel.

[0026] In an embodiment, the machine learning model is trained to predict a degree of cure of a photocurable polymer.

[0027] The method of monitoring a degree of cure of a hydrogel formation may be integrated into an additive manufacturing process.

[0028] BRIEF DESCRIPTION OF THE DRAWINGS In the following, embodiments of the present invention will be described as non-limiting examples with reference to the accompanying drawings in which:

[0029] FIG.1 shows a schematic view of a system for monitoring curing progress of a photocurable polymer during an additive manufacturing process according to an embodiment of the present invention;

[0030] FIG.2a and FIG.2b illustrate the calibration of a system for monitoring curing progress of a photocurable polymer during an additive manufacturing process according to an embodiment of the present invention;

[0031] FIG.3 is a block diagram showing the training of machine learning model used in embodiments of the present invention;

[0032] FIG.4a is a schematic diagram of a system for monitoring degree curing progress of a photocurable polymer during an additive manufacturing process in which a UV radiation source and detector are movable according to an embodiment of the present invention;

[0033] FIG.4b is a schematic diagram of a system for monitoring curing progressof a photocurable polymer during an additive manufacturing process in which a transparent stage is movable according to an embodiment of the present invention; and

[0034] FIG.5 is a flowchart showing a method of controlling curing in an additive manufacturing process according to an embodiment of the present invention.

[0035] DETAILED DESCRIPTION

[0036] The present disclosure relates to monitoring the progress of curing of a photocurable polymer. The progress of curing may be quantified by a metric such as degree of cure, gel fraction, and total dose absorbed by the polymer. Degree of cure (DoC) is a term used for the amount of crosslinking happens in the polymer. The photocurable polymer may be a hydrogel or other photocurable material. FIG.1 shows a schematic view of a system for monitoring progress of curing of a photocurable polymer during an additive manufacturing process according to an embodiment of the present invention. As shown in FIG.1 , the system 100 comprises an ultraviolet (UV) radiation source 110 which emits UV radiation towards a photocurable polymer formation 120 which is deposited on a transparent stage 130 during an additive manufacturing process. In the example, the photocurable polymer is a hydrogel. A UV radiation detector 140 is arranged below the transparent stage 130. The UV radiation detector 140 measures an intensity of UV radiation transmitted through the photocurable polymer formation 120. The UV radiation detector 140 comprises an aperture 142 which allows UV radiation transmitted through a target region 122 of the photocurable polymer formation 120 to enter the UV radiation detector 140.

[0037] The UV radiation detector 140 is coupled to a power meter 144 which determines a measured U V intensity 146. The system 100 comprises a machine learning model 150 which receives the measured UV intensity 146 as an input. The machine learning model 150 also receives input data 152 which may comprise an indication of the hydrogel formulation, an indication of a photo-initiator concentration which is added to the hydrogel, an indication of the thickness of the hydrogel and an indication of the applied UV intensity, that is the intensity of the UV radiation incident on the photocurable polymer formation 120.

[0038] As shown in FIG.1 , the machine learning model 150 provides an output 154 which indicates the curing progress of the hydrogel without prior dosage information.

[0039] By using hydrogel formulation, photoinitiator concentration, hydrogel thickness, applied UV intensity, and measured UV intensity as input, the machine learning (ML) algorithms establish a correlation between the hydrogel's UV transmittance and the received UV dosages, allowing the system to precisely calculate the received UV dosage for any specific location within the printed construct. The detection zone can be adjusted by moving the transparent stage to place the area to be measured above the UV detector. The size of the pinhole or aperture on the UV detector will determine the size of detection zone. In an embodiment, the aperture used was 1 mm x 1 mm square. There is no limitation to the possible size, however, the smaller the aperture, the better the resolution of the sensor. For example, the 1 mm x 1 mm square will just measure the 1 mm x 1 mm square shape of the gel above the pinhole. Generally, the aperture should not be too large such that the distribution of LIV power inside the aperture size is uniform (within 10% variation). The UV radiation can be focused or broad. It depends on the application. However, the aperture should be much smaller than the UV radiation, and right at the middle of the beam, if the UV radiation has a gaussian distribution, to ensure that the distribution of the UV power is uniform (most UV radiation is gaussian). For UV radiation with top hat distribution, it is fine as long as the aperture is smaller than the UV radiation.

[0040] UV radiation may refer to any electromagnetic radiation with wavelengths in the range 10-410nm. It is noted that wavelengths of 405 nm are commonly used for curing polymer. For the purposes of the present disclosure a wavelength of 405nm is referred to as UV light. It is understood for some applications 405 nm is slightly above the wavelength of UV light and is visible light, but for the purposes of curing, a wavelength of 405nm may be considered as UV light.

[0041] The importance of this setup is that it can measure the UV dosage received by the photocurable polymer without any information on how much crosslinking is received before. For example, when the photocurable polymer is exposed to uneven UV intensity such as when the UV light spot size is smaller than the photocurable polymer, when the known UV intensity is wrong due to fluctuation in machine, or when the curing is stopped abruptly due to error in operation. With this setup, the predicted UV dosage can be utilized to accurately fine tune the UV dosage received by the photocurable polymer.

[0042] In experimental validation, the predicted UV dosage received by the photocurable polymer has an error of less than 10% from the actual UV dosage received. FIG.2a and FIG.2b illustrate the calibration of a system for monitoring progress of curing of a photocurable polymer during an additive manufacturing process according to an embodiment of the present invention.

[0043] As described above, one of the inputs to the machine learning model 150 is an indication of the applied LIV intensity, that is, an indication of the UV intensity incident on the photocurable polymer formation. In order to determine the applied UV intensity, the system may be calibrated as shown in FIG.2a.

[0044] As shown in FIG.2a, during calibration, UV radiation from the UV radiation source 110 is incident directly on UV detector 140 without any photocurable polymer present. Thus, the UV radiation passes through the transparent stage 130 and through the aperture 140 to be measured by the UV detector 140. This allows an indication of the incident UV radiation to be estimated.

[0045] FIG.2b shows the system in use. As shown in FIG.2b, the UV radiation from the UV radiation source 110 is incident on the target region 122 of the photocurable polymer formation 120. The transmitted radiation then passes through the transparent stage 130 and enters the UV detector 140 though the aperture 142.

[0046] By calibrating the system in the configuration shown in FIG.2a, the incident UV radiation in the configuration shown in FIG.2b can be estimated.

[0047] In order to acquire data for training the machine learning model, hydrogel is prepared at different composition with different concentrations of photoinitiator. For example, GelMA solution is prepared at 10% w / v or 15% w / v, mixed with Lithium Phenyl(2,4,6- trimethylbenzoyl)phosphinate (LAP) solution at 0.1% w / v or 0.3% w / v. In this case, the GelMA is the hydrogel and LAP is the photoinitiator. The photocurable polymer solution will be printed or casted at different thickness on a transparent stage such as a glass slide.

[0048] UV light at the activation wavelength for the photoinitiator will be applied on the hydrogel at different intensity (and calibrated according to the configuration shown in FIG.2a), and the intensity of UV radiation transmitted through the photocurable polymer is recorded using the configuration shown in FIG.2b.

[0049] FIG.3 is a block diagram showing the training of machine learning model used in embodiments of the present invention. The machine learning (ML) model 150 may be implemented as a random forest regression (RFR) model, a support vector regression (SVR) model, or a deep neural network (DNN). RFR or SVR is preferred at low sample count, while DNN will perform better with sufficient samples.

[0050] The ML model can be implemented by collecting a dataset comprised of the input and output (any indicator for degree of cure), then train the ML model with it. Ideally a DNN should be used as lesser sample is required for an accurate model by using transfer learning, where a trained model is calibrated to a different type of sample.

[0051] As shown in FIG.3, the input 152 for model training is hydrogel formulation, photoinitiator concentration, photocurable polymer thickness, applied UV intensity, and measured UV intensity. The output 154 of the model is the UV dosage received by the hydrogel and for the purpose of training the model this may be calculated by the following formula.

[0052] Received UV dosage — Applied UV intensity X UV exposure duration

[0053] In embodiments of the present invention, the UV radiation source and UV detector may be scanned across the hydrogel formation in order to cure the photocurable polymer and monitor the degree of cure. This can be achieved by either transparent stage being fixed and the UV radiation source and UV detector being movable or by the UV radiation source and UV detector being fixed and the transparent stage being movable.

[0054] FIG.4a is a schematic diagram of a system for monitoring progress of curing of a photocurable polymer during an additive manufacturing process in which a UV radiation source and detector are movable according to an embodiment of the present invention. As shown in FIG.4a, the UV radiation source 110 and the UV detector 140 are movable and can be scanned across the photocurable polymer formation 120. This allows the target region 122 of the hydrogel formation 120 to be moved such that the U V radiation passes through a different target region 122 and enters the UV detector 140 through the aperture 142 depending on the positioning of the U V radiation source 110 and the UV detector 140 relative to the photocurable polymer formation 120 arranged on the transparent stage 130.

[0055] FIG.4b is a schematic diagram of a system for monitoring progress of curing of a photocurable polymer during an additive manufacturing process in which a transparent stage is movable according to an embodiment of the present invention.

[0056] As shown in FIG.4b, transparent stage 130 is movable and by moving the transparent stage 130, UV radiation emitted by the UV radiation source can be scanned across the photocurable polymer formation 120. This allows the target region 122 of the photocurable polymer formation 120 to be moved such that the UV radiation passes through a different target region 122 and enters the UV detector 140 through the aperture 142 depending on the positioning of the UV radiation source 110 and the UV detector 140 relative to the photocurable polymer formation 120 arranged on the transparent stage 130.

[0057] The system for monitoring degree of progress of a photocurable polymer may be integrated into an additive manufacturing system which utilizes photocurable resin. It may also be commercialized as a standalone system for measuring the curing process of photopolymerizing resin.

[0058] FIG.5 is a flowchart showing a method of controlling curing in an additive manufacturing process according to an embodiment of the present invention.

[0059] The method 500 shown in FIG.5 may be carried out as part of an additive manufacturing process in which a photocurable polymer or resin such as a hydrogel is applied on a transparent stage and then cured. In step 502, the detected UV radiation is used to determine a curing progress in an area of the photocurable polymer by inputting an indication of the transmitted UV radiation into a machine learning model as described above. In step 504, a determination is made as to whether enough curing progress has occurred. This can be achieved by comparing the curing progress output by the machine learning model with a target or threshold amount of curing progress.

[0060] If the progress is not yet enough, the method moves to set 506 and UV curing continues followed by repeating step 502.

[0061] If the progress is determined to be enough, then the method moves to step 508 in which curing in the area stops. Following step 508, a different area my be cured or the curing process may stop if all areas have been cured.

[0062] Whilst the foregoing description has described exemplary embodiments, it will be understood by those skilled in the art that many variations of the embodiments can be made within the scope and spirit of the present invention.

Claims

CLAIMS1 . A system for monitoring curing progress of a photocurable polymer formation during an additive manufacturing process, the system comprising: a transparent stage configured to support the photocurable polymer formation; a source of ultra violet radiation configured to illuminate a target region of the photocurable polymer formation with incident ultra violet radiation; an ultra violet radiation detector arranged on an opposing side of the transparent stage from the source of ultra violet radiation, the ultra violet radiation detector configured to detect an intensity of transmitted ultra violet radiation transmitted through the target region of the photocurable polymer formation; and a data storage device storing a machine learning model trained to curing progress of a photocurable polymer from input data comprising an indication of ultraviolet transmission of the photocurable polymer.

2. The system according to claim 1 , wherein the transparent stage is movable relative to the source of ultra violet radiation and the ultra violet radiation detector such that the target region of the photocurable polymer formation can be moved laterally relative to the source of ultra violet radiation and the ultra violet radiation detector.

3. The system according to claim 1 or 2, wherein the ultra violet radiation detector comprises an aperture configured to allow ultra violet radiation transmitted through the target region of the photocurable polymer formation to enter the ultra violet radiation detector.

4. The system according to any preceding claim, wherein the machine learning model is trained to predict a degree of cure of a photocurable polymer.

5. The system according to any preceding claim wherein the machine learning model comprises a random forest regression model, a support vector regression model or a deep neural network.

6. The system according to any preceding claim, wherein the machine learning model is trained using one or more parameters selected from the group consisting ot a formulation of the photocurable polymer, a concentration of one or more photoini tiators, a thickness of the photocurable polymer, a measured ultra violet intensity, or an applied ultra violet intensity.

7. The system according to any preceding claim, wherein the photocurable polymer formation comprises hydrogel.

8. An additive manufacturing system comprising the system according to any- preceding claim.

9. A method of monitoring progress of curing of a photocurable polymer formation during an additive manufacturing process, the method comprising: illuminating a target region of the photocurable polymer formation with incident ultraviolet radiation; detecting an intensity of transmitted ultra violet radiation transmitted through the target region of the photocurable polymer formation; determining progress of curing of the target region of the photocurable polymer formation by inputting an indication of the intensity of transmitted ultraviolet radiation into a machine learning model trained to predict progress of curing of a photocurable polymer from input data comprising an indication of ultra violet transmission of the photocurable polymer.

10. The method of claim 9, further comprising determining an indication of the incident ultra violet radiation by illuminating an ultraviolet radiation detector without the photocurable polymer present.11 . The method according to claim 9 or 10, wherein the photocurable polymer formation is arranged on a transparent stage and the method further comprises moving the transparent stage relative to a source of the ultra violet radiation.

12. The method according to any one of claims 9 to 11 , wherein the photocurable polymer formation comprises hydrogel.

13. The method according to any one ot ciaims 9 to 12, wherein the machine learning model is trained to predict a degree of cure of a photocurable polymer.

14. An additive manufacturing method comprising the method according to any one of claims 9 to 13.

Images