Beam-shaping for Laser-based Additive and Subtractive-manufacturing Techniques (BLAST)
Lead Research Organisation:
University of Southampton
Department Name: Optoelectronics Research Centre (ORC)
Abstract
Digital Micromirror Devices (DMDs) are the heart of the image-projection technology used in the modern cinema projectors. They are a 2D array of several million, micro-sized, computer-controllable mirrors, where each mirror can flip on its axis many thousands of times per second. When combined with a RGB light source, such as in a cinema, the device enables the projection of full-colour videos onto a screen. However, in recent years this projection technology has moved out of the cinema and into laboratories across the world, where it is assisting scientists in many research fields.
At the Optoelectronics Research Centre, at the University of Southampton, scientists have been using this DMD technology to generate micron-sized intricate patterns of laser light, for the development of a range of novel subtractive (removing material) and additive (adding material) laser-based manufacturing processes.
In this 5-year project, the team will be working with a wide range of industrial and academic partners, who see the potential for new and exciting manufacturing processes, as summarised below:
SPI Lasers, a UK fibre laser company: A major advantage of using DMDs for shaping a laser beam is the extremely high speed at which light patterns can be generated, updated and modified. The team will be combining fibre laser technology with DMD technology to enable extremely high-repetition-rate beam shape and energy control, for applications in a wide range of manufacturing areas including the marking of high-value objects.
M-Solv, a UK laser-integrator: Here, the team will be testing and optimising their technology using a wide range of industrial manufacturing lasers, and will develop a range of novel additive manufacturing processes for the micro-scale. The outcome will be additional manufacturing capability for UK companies.
University Hospital Southampton: Recent scientific results have shown the ability to control the specialisation of human stem cells (e.g. to bone or to muscle) via intricately designed 2D surface structures. Working with Prof. Richard Oreffo, a founder of this field, the team will be using their technique to produce a range of bespoke surface-textured substrates that will enable biologists to further understand and control stem-cell specialisation for applications in regenerative medicine.
University of Southampton: Metamaterials are a family of materials that offer amazingly unusual properties, such as the ability to bend light (for use as invisibility cloaks) or even slow it right down. However, scientists have yet to develop a cost-effective method for making such devices on centimetre or larger size-scales. The team will be investigating whether the DMDs combined with high-repetition-rate lasers can speed up the process and enable cost-effective manufacturing of cm-sized devices.
Oxsensis, a UK company that develops sensors for extreme environments: The team intends to develop new manufacturing processes that will enable a new range of sensors for applications in industries such as Aerospace, Power Generation, and Oil and Gas. Specifically, the team will be using their recently demonstrated ability to laser-machine very accurately and rapidly in diamond, in order to develop new techniques for making sensors in a range of difficult-to-machine materials.
At the Optoelectronics Research Centre, at the University of Southampton, scientists have been using this DMD technology to generate micron-sized intricate patterns of laser light, for the development of a range of novel subtractive (removing material) and additive (adding material) laser-based manufacturing processes.
In this 5-year project, the team will be working with a wide range of industrial and academic partners, who see the potential for new and exciting manufacturing processes, as summarised below:
SPI Lasers, a UK fibre laser company: A major advantage of using DMDs for shaping a laser beam is the extremely high speed at which light patterns can be generated, updated and modified. The team will be combining fibre laser technology with DMD technology to enable extremely high-repetition-rate beam shape and energy control, for applications in a wide range of manufacturing areas including the marking of high-value objects.
M-Solv, a UK laser-integrator: Here, the team will be testing and optimising their technology using a wide range of industrial manufacturing lasers, and will develop a range of novel additive manufacturing processes for the micro-scale. The outcome will be additional manufacturing capability for UK companies.
University Hospital Southampton: Recent scientific results have shown the ability to control the specialisation of human stem cells (e.g. to bone or to muscle) via intricately designed 2D surface structures. Working with Prof. Richard Oreffo, a founder of this field, the team will be using their technique to produce a range of bespoke surface-textured substrates that will enable biologists to further understand and control stem-cell specialisation for applications in regenerative medicine.
University of Southampton: Metamaterials are a family of materials that offer amazingly unusual properties, such as the ability to bend light (for use as invisibility cloaks) or even slow it right down. However, scientists have yet to develop a cost-effective method for making such devices on centimetre or larger size-scales. The team will be investigating whether the DMDs combined with high-repetition-rate lasers can speed up the process and enable cost-effective manufacturing of cm-sized devices.
Oxsensis, a UK company that develops sensors for extreme environments: The team intends to develop new manufacturing processes that will enable a new range of sensors for applications in industries such as Aerospace, Power Generation, and Oil and Gas. Specifically, the team will be using their recently demonstrated ability to laser-machine very accurately and rapidly in diamond, in order to develop new techniques for making sensors in a range of difficult-to-machine materials.
Planned Impact
The main beneficiaries of this research lie within the healthcare and manufacturing sectors, each of which represents a
major opportunity for financial return (economic benefit), and healthcare provision (societal impact).
1. Economic benefit
Laser and system integrators (SPI, M-Solv, and others) will benefit from the generation of new markets for their industrial lasers, when coupled to DMD hardware, and the introduction of new techniques in laser-based processing. The UK can effectively exploit such an opportunity and thereby generate financial return. In an ideal scenario, these companies can then market DMD-based products either as stand-alone systems, or integrated into their current production facilities, and a spinout from Southampton is therefore also feasible, if financial backing is secured to take this forward. Areas that will benefit from precision rapid machining and marking at the micro/nanoscale span security, optoelectronics, consumer products, sensing, safety, aerospace and more, and we will engage with our impact panel to both promote our activities and also make further connections to the next set of interested parties within UK manufacturing.
2. Healthcare and societal benefit
With the continuously increasing strains on already over-burdened public health systems of developed nations such as the UK, and developing countries with the low-resource settings, the role of healthcare, and in particular cancer research, diagnostic tests, stem-cell research for regenerative medicine and other medical and clinical application areas is an increasingly important factor.
Our DMD-based micromachining technique may, for the first time, enable a step-change in any (and ideally all) of the topics above. If even one of these succeeds, the implications are profound for an already stretched public health service, whose budget of £100B per year is already inadequate. Healthcare is an easy heading to list in such an impact summary, but we are very confident (and have the full support and approval of our medical colleagues) that our technique will really lead to some radical breakthroughs.
3. Training and the next generation of researchers
It almost goes without saying, but this progamme will be of immense value to the next generation of research scientists and, in particular, those researchers who can span the 'single discipline' problem so prevalent in academia. This programme is an almost ideal vehicle for combining laser science with biological materials, and materials science with healthcare. The named researcher and PhD students should be extremely employable in either a company situation, academic appointments, or even a start-up if this were to go ahead.
major opportunity for financial return (economic benefit), and healthcare provision (societal impact).
1. Economic benefit
Laser and system integrators (SPI, M-Solv, and others) will benefit from the generation of new markets for their industrial lasers, when coupled to DMD hardware, and the introduction of new techniques in laser-based processing. The UK can effectively exploit such an opportunity and thereby generate financial return. In an ideal scenario, these companies can then market DMD-based products either as stand-alone systems, or integrated into their current production facilities, and a spinout from Southampton is therefore also feasible, if financial backing is secured to take this forward. Areas that will benefit from precision rapid machining and marking at the micro/nanoscale span security, optoelectronics, consumer products, sensing, safety, aerospace and more, and we will engage with our impact panel to both promote our activities and also make further connections to the next set of interested parties within UK manufacturing.
2. Healthcare and societal benefit
With the continuously increasing strains on already over-burdened public health systems of developed nations such as the UK, and developing countries with the low-resource settings, the role of healthcare, and in particular cancer research, diagnostic tests, stem-cell research for regenerative medicine and other medical and clinical application areas is an increasingly important factor.
Our DMD-based micromachining technique may, for the first time, enable a step-change in any (and ideally all) of the topics above. If even one of these succeeds, the implications are profound for an already stretched public health service, whose budget of £100B per year is already inadequate. Healthcare is an easy heading to list in such an impact summary, but we are very confident (and have the full support and approval of our medical colleagues) that our technique will really lead to some radical breakthroughs.
3. Training and the next generation of researchers
It almost goes without saying, but this progamme will be of immense value to the next generation of research scientists and, in particular, those researchers who can span the 'single discipline' problem so prevalent in academia. This programme is an almost ideal vehicle for combining laser science with biological materials, and materials science with healthcare. The named researcher and PhD students should be extremely employable in either a company situation, academic appointments, or even a start-up if this were to go ahead.
People |
ORCID iD |
Benjamin Mills (Principal Investigator / Fellow) |
Publications

Bruck R
(2016)
All-optical spatial light modulator for reconfigurable silicon photonic circuits
in Optica

Buchnev O
(2022)
Deep-Learning-Assisted Focused Ion Beam Nanofabrication.
in Nano letters

Courtier A
(2021)
Predictive Visualisation of Fibre Laser Machining via Deep Learning

Courtier A
(2017)
Laser-Based Fabrication of Nanofoam inside a Hollow Capillary
in Materials Sciences and Applications

Courtier A
(2023)
Predictive visualization of fiber laser cutting topography via deep learning with image inpainting
in Journal of Laser Applications

Courtier AF
(2021)
Modelling of fibre laser cutting via deep learning.
in Optics express

Feinaeugle M
(2017)
Time-resolved imaging of flyer dynamics for femtosecond laser-induced backward transfer of solid polymer thin films
in Applied Surface Science

Feinaeugle M
(2016)
Laser-induced backward transfer of nanoimprinted polymer elements
in Applied Physics A

Grant-Jacob J
(2021)
Semantic segmentation of pollen grain images generated from scattering patterns via deep learning
in Journal of Physics Communications

Grant-Jacob J
(2021)
In-flight sensing of pollen grains via laser scattering and deep learning
in Engineering Research Express
Title | Visualization 1.mp4 |
Description | The 5th laser pulse (highlighted in red) was deliberately displaced from its undisturbed location to all possible positions on the virtual workpiece. In almost all cases the RL agent was able to successfully complete machining of the target pattern regardless of this disturbance. |
Type Of Art | Film/Video/Animation |
Year Produced | 2022 |
URL | https://opticapublishing.figshare.com/articles/media/Visualization_1_mp4/19077152 |
Title | Visualization 1.mp4 |
Description | The 5th laser pulse (highlighted in red) was deliberately displaced from its undisturbed location to all possible positions on the virtual workpiece. In almost all cases the RL agent was able to successfully complete machining of the target pattern regardless of this disturbance. |
Type Of Art | Film/Video/Animation |
Year Produced | 2022 |
URL | https://opticapublishing.figshare.com/articles/media/Visualization_1_mp4/19077152/1 |
Title | Visualization 2.mp4 |
Description | The 5th laser pulse (highlighted in red) was deliberately displaced from its undisturbed location to all possible positions on the virtual workpiece. In almost all cases the RL agent was able to successfully complete machining of the target pattern regardless of this disturbance. |
Type Of Art | Film/Video/Animation |
Year Produced | 2022 |
URL | https://opticapublishing.figshare.com/articles/media/Visualization_2_mp4/19535962/1 |
Title | Visualization 2.mp4 |
Description | The 5th laser pulse (highlighted in red) was deliberately displaced from its undisturbed location to all possible positions on the virtual workpiece. In almost all cases the RL agent was able to successfully complete machining of the target pattern regardless of this disturbance. |
Type Of Art | Film/Video/Animation |
Year Produced | 2022 |
URL | https://opticapublishing.figshare.com/articles/media/Visualization_2_mp4/19535962 |
Title | Visualization 3.mp4 |
Description | The 5th laser pulse (highlighted in red) was deliberately displaced from its undisturbed location to all possible positions on the virtual workpiece. In almost all cases the RL agent was able to successfully complete machining of the target pattern regardless of this disturbance. |
Type Of Art | Film/Video/Animation |
Year Produced | 2022 |
URL | https://opticapublishing.figshare.com/articles/media/Visualization_3_mp4/19535965 |
Title | Visualization 3.mp4 |
Description | The 5th laser pulse (highlighted in red) was deliberately displaced from its undisturbed location to all possible positions on the virtual workpiece. In almost all cases the RL agent was able to successfully complete machining of the target pattern regardless of this disturbance. |
Type Of Art | Film/Video/Animation |
Year Produced | 2022 |
URL | https://opticapublishing.figshare.com/articles/media/Visualization_3_mp4/19535965/1 |
Title | Visualization 4.mp4 |
Description | The 5th laser pulse (highlighted in red) was deliberately displaced from its undisturbed location to all possible positions on the virtual workpiece. In almost all cases the RL agent was able to successfully complete machining of the target pattern regardless of this disturbance. |
Type Of Art | Film/Video/Animation |
Year Produced | 2022 |
URL | https://opticapublishing.figshare.com/articles/media/Visualization_4_mp4/19535971/1 |
Title | Visualization 4.mp4 |
Description | The 5th laser pulse (highlighted in red) was deliberately displaced from its undisturbed location to all possible positions on the virtual workpiece. In almost all cases the RL agent was able to successfully complete machining of the target pattern regardless of this disturbance. |
Type Of Art | Film/Video/Animation |
Year Produced | 2022 |
URL | https://opticapublishing.figshare.com/articles/media/Visualization_4_mp4/19535971 |
Title | Visualization 5.mp4 |
Description | The 5th laser pulse (highlighted in red) was deliberately displaced from its undisturbed location to all possible positions on the virtual workpiece. In almost all cases the RL agent was able to successfully complete machining of the target pattern regardless of this disturbance. |
Type Of Art | Film/Video/Animation |
Year Produced | 2022 |
URL | https://opticapublishing.figshare.com/articles/media/Visualization_5_mp4/19535974 |
Title | Visualization 5.mp4 |
Description | The 5th laser pulse (highlighted in red) was deliberately displaced from its undisturbed location to all possible positions on the virtual workpiece. In almost all cases the RL agent was able to successfully complete machining of the target pattern regardless of this disturbance. |
Type Of Art | Film/Video/Animation |
Year Produced | 2022 |
URL | https://opticapublishing.figshare.com/articles/media/Visualization_5_mp4/19535974/1 |
Description | The fundamental proposal hypothesis that beam shaping has the potential to unlock new laser-based manufacturing capabilities has been proven correct. The project was a huge success in that regard, and this hypothesis has been proven through a wealth of scientific publications, and through partnerships with UK industry (where the beam shaping technology is now starting to be integrated). However, the most important project discovery, is that these techniques can also be further enhanced through deep learning approaches. |
Exploitation Route | The key breakthrough of the project, namely applying deep learning to high precision laser machining, is now starting to be transferred into UK industry, through an increasing number of partnerships between the proposal PI and UK companies. This process will almost certainly continue to accelerate over the coming months/years. The fundamental results aligned with this application of deep learning have been published widely in open access journals, and hence the knowledge and understanding of these novel exciting techniques are publicly available. |
Sectors | Manufacturing including Industrial Biotechology |
Description | There has been a wide range of non-academic impacts, specifically in the transfer of beam shaping technology and artificial intelligence techniques to the UK laser-based manufacturing industrial partners involved in this project. This transfer of knowledge provides the UK with additional high-precision laser machining capability and hence assists these companies on the global manufacturing market. In the years 2021-2023, research from this grant has led to multiple invited talks at industry events, a keynote talk at an industrial conference, and several additional collaborations with UK manufacturing companies. This is clear evidence of a general and wider acceptance that artificial intelligence (and deep learning in particular) has now started to become mainstream within the UK laser based manufacturing companies, and indeed many of the world-first demonstrations in this exciting new area were demonstrated during this grant. As of 2024, there is now a start-up using the technology developed during this fellowship for next-generation data storage. |
First Year Of Impact | 2019 |
Sector | Manufacturing, including Industrial Biotechology |
Impact Types | Economic |
Description | Lasers that Learn: AI-enabled intelligent materials processing |
Amount | £777,859 (GBP) |
Funding ID | EP/T026197/1 |
Organisation | Engineering and Physical Sciences Research Council (EPSRC) |
Sector | Public |
Country | United Kingdom |
Start | 11/2020 |
End | 10/2024 |
Title | distribution_of_angles_v=15.txt |
Description | Distribution of angles for v=15 m/min, experimental and predicted. |
Type Of Material | Database/Collection of data |
Year Produced | 2021 |
Provided To Others? | Yes |
URL | https://opticapublishing.figshare.com/articles/dataset/distribution_of_angles_v_15_txt/15050136/1 |
Title | distribution_of_angles_v=15.txt |
Description | Distribution of angles for v=15 m/min, experimental and predicted. |
Type Of Material | Database/Collection of data |
Year Produced | 2021 |
Provided To Others? | Yes |
URL | https://opticapublishing.figshare.com/articles/dataset/distribution_of_angles_v_15_txt/15050136 |
Title | distribution_of_angles_v=20.txt |
Description | Distribution of angles for v=20 m/min, experimental and predicted. |
Type Of Material | Database/Collection of data |
Year Produced | 2021 |
Provided To Others? | Yes |
URL | https://opticapublishing.figshare.com/articles/dataset/distribution_of_angles_v_20_txt/15050142 |
Title | distribution_of_angles_v=20.txt |
Description | Distribution of angles for v=20 m/min, experimental and predicted. |
Type Of Material | Database/Collection of data |
Year Produced | 2021 |
Provided To Others? | Yes |
URL | https://opticapublishing.figshare.com/articles/dataset/distribution_of_angles_v_20_txt/15050142/1 |
Title | distribution_of_pixel_intensities_experimental_predicted.txt |
Description | Distribution of pixel intensities for v=15, 20 m/min, experimental and predicted |
Type Of Material | Database/Collection of data |
Year Produced | 2021 |
Provided To Others? | Yes |
URL | https://opticapublishing.figshare.com/articles/dataset/distribution_of_pixel_intensities_experimenta... |
Title | distribution_of_pixel_intensities_experimental_predicted.txt |
Description | Distribution of pixel intensities for v=15, 20 m/min, experimental and predicted |
Type Of Material | Database/Collection of data |
Year Produced | 2021 |
Provided To Others? | Yes |
URL | https://opticapublishing.figshare.com/articles/dataset/distribution_of_pixel_intensities_experimenta... |
Title | experimental_and_predicted_labels_confusion_matrix.txt |
Description | Experimental and predicted cutting speeds for a confusion matrix |
Type Of Material | Database/Collection of data |
Year Produced | 2021 |
Provided To Others? | Yes |
URL | https://opticapublishing.figshare.com/articles/dataset/experimental_and_predicted_labels_confusion_m... |
Title | experimental_and_predicted_labels_confusion_matrix.txt |
Description | Experimental and predicted cutting speeds for a confusion matrix |
Type Of Material | Database/Collection of data |
Year Produced | 2021 |
Provided To Others? | Yes |
URL | https://opticapublishing.figshare.com/articles/dataset/experimental_and_predicted_labels_confusion_m... |
Title | percentage_of_dark_spots_experimental_and_predicted.txt |
Description | Percentage of dark spots for v=15, 20 m/min, experimental and predicted. |
Type Of Material | Database/Collection of data |
Year Produced | 2021 |
Provided To Others? | Yes |
URL | https://opticapublishing.figshare.com/articles/dataset/percentage_of_dark_spots_experimental_and_pre... |
Title | percentage_of_dark_spots_experimental_and_predicted.txt |
Description | Percentage of dark spots for v=15, 20 m/min, experimental and predicted. |
Type Of Material | Database/Collection of data |
Year Produced | 2021 |
Provided To Others? | Yes |
URL | https://opticapublishing.figshare.com/articles/dataset/percentage_of_dark_spots_experimental_and_pre... |