Final Report Summary - ADOPSYS (ADvanced OPtical SYStem Design)
ADOPSYS (ADvanced OPtical SYStem design) established a network of twelve Early-Stage Researchers at ten host organisations: Delft University of Technology (NL), Vrije Universiteit Brussel (BE), The National Research University of Information Technologies, Mechanics and Optics (RU), Centre national de la recherche scientifique (FR), Universidad Politécnica de Madrid (ES), Friedrich-Schiller-Universität Jena (DE), OSRAM (DE), Carl Zeiss AG (DE), Light Prescriptions Innovators (ES) and University of Eastern Finland (FI) and three Associated partners: Datalogic (IT), Carl Zeiss SMT (DE) and KLA Tencor (BE).
The primary objectives of the ADOPSYS ITN were to develop new strategies for the inverse problem of optical design that reduce the amount of human trial-and-error effort significantly, and to apply them in close collaboration with the industrial partners to design problems that occur in modern applications, in particular energy efficient lighting systems, high resolution systems and machine vision, inspection and safety. In addition, improving the enabling methodology for the propagation of optical fields through optical systems (i.e. the direct problem in optics) was studied to increase the efficiency of the inverse design methodologies and hence to increase design productivity. The second, equally important, objective of this project was to build a unique European training infrastructure for PhDs in the field of optical design methods for a future career in European academy and photonics industry. The training network ADOPSYS trained young optical scientists and engineers in the full scope of optical modelling and design at the most advanced (i.e. PhD) level. The academic and industrial partners in ADOPSYS together covered the complete modern field of optical modelling and design and its applications, which no single European group or institute could provide alone.
The research fellows participated in the three Research Workshops, two General Scientific Trainings and the Training in Transferable Skill event organized by the consortium, and were enrolled in local training at each of the host institutes. These local trainings were often part of the Graduate Schools at the different host institutes and consisted of e.g. English courses and courses on how to write a scientific paper. Further, each fellow has carried out secondments at other full or associated partners.
The main results achieved by the ADOPSYS research programme are improved methods for both direct modelling of optical diffraction methods as for inverse design problems. The new methods for optical design concern both imaging and non-imaging optics.
In the modelling of diffraction problems, a method for cascaded diffraction was developed by ESR10. This method was studied in collaboration with Carl Zeiss and aims at incorporating in the model the effects of multiple diffraction at different limiting apertures in complicate systems. The method was first developed for scalar fields and was then extended to include polarisation effects. ESR10 has investigated a complimentary method base on the field tracing technique. A careful analysis was made to investigate the remaining errors in the obtained focused fields.
ESR1 has studied phase sensitive imaging techniques, in particular ptychography applied to soft X-ray and EUV systems. A major problem in this technique is to retrieve the probe positions, which often are not known with sufficient accuracy and this has a detrimental effect on the accuracy of the reconstructions. A new and fast method for retrieving the probe positions was developed. In the study of new methods for optical design for imaging systems, ESR2 concentrated on the design of large systems such as for lithography. A major issue is to obtain a sufficiently good starting design. The approach developed was to divide the optical systems in two parts. Both parts were first taken to consists of only a few lenses and more and more lenses were added depending on the transverse aberrations obtained for the entire field of view. Apart from publications, several patents have resulted from this work. ESR2 has collaborated with ESR3 to investigate the application of the Saddle Point Method (SPM) to complicated optical systems and a common paper is expected soon. ESR3 has further developed the SPM on challenging optical design systems of large field of view and on a-spheres. On the latter ESR3 has collaborated with ESR6 and a common paper is going to be published soon. Furthermore ESR3 has made a design during an internship at industrial partner Datalogic and a successful prototype has been made based on this design. ESR6 has investigated modifying the SMS method for application in imaging optics. SMS was developed for non-imaging optics. The SMS method was used to obtain good starting values for optical systems that consist of a-spheres. The new method was also further explored during an internship at Carl Zeiss. ESR7 has developed a new multi-fields direct design that is capable of partially coupling more than two fields with two surfaces in both two and three dimensions to obtain optical systems consisting of multiple free forms. The method was applied to designing a wide-field objective lens, achieving superior imaging performance when compared to its counterpart. Another promising design strategy was obtained by using paraxial aberration theory to get a good starting point for N > =3 freeform imaging systems. The project was carried out in a collaboration between VUB, technical University of Madrid and Carl Zeiss.
ESR9 has obtained new results on designing pupil masks for improved imaging. Improved images are obtained by deconvolution with appropriate levels of regularisation. An internship was done in KLA Tencor. ESR11 research concentrated on polarimetry, i.e. the imaging of polarisation contrast. The challenge was to image with broad band light where a trade-off must be found between accuracy in resolution and in polarisation contrast, due to the fact that polarisers are wavelength dependent. Application to microscopy was studied during an internship at Carl Zeiss.
The other ERSs worked on design methods for non-imaging optics. The challenge in the project of ESR4 was to obtain homogeneous illumination from inhomogeneous source distributions such as multicolor extended LEDs. To achieve this, the ESR proposed to use free form Shell-mixer. Furthermore, the SMS was extended to more surfaces which substantially improves the homogenuity. The work was carried out in a collaboration between the company LPI and VUB. ESR6 used novel integral equation methods as design method for free form surfaces to achieve homogeneous distributed light patterns. The surfaces were described using B-splines. This project was carried out in a collaboration between OSRAM and the Technical University Madrid. ESR7 has developed a new multi-fields direct design that is capable of partially coupling more than two fields with two surfaces in both two and three dimensions to obtain optical systems consisting of multiple free forms. The method was applied to designing a wide-field objective lens, achieving superior imaging performance when compared to its counterpart. Another promising design strategy was obtained by using paraxial aberration theory to get a good starting point for N > =3 freeform imaging systems. ESR12 has developed an interesting new algorithm for designing diffractive optical elements such as gratings which includes absorption of the materials involved. A new method to evaluate the mutual coherence function of a light beam using a diffractive element was also obtained. The research was carried out in a collaboration between UEF and FSU and an internship was done at OSRAM.
As illustration of the success of ADOPSYS, it should be mentioned that one of the ESRs, namely Yunfeng Nie (ESR7), obtained for her research work, the 2016 Michael Kidger Memorial Scholarship, selected annually to one worldwide PhD student in optical design domain. Furthermore, another ESR, ESR6, Milena Nikolic received a Cum Laude for her PhD thesis from partner UPM.
A lot of training activities which have been organised by the consortium which have been attended by all ESRs (with very few occasional assumptions). There have been three Research Workshops, two Scientific Trainings and one training on transferable skills. All activities were highly appreciated by the ESRs.
The ESR have organised regular Skype meetings between them and have elected a representative who attended the Board meeting. The atmosphere between the ESRs has been exceptionally good throughout the entire programme.
Project website: www.adopsys.eu
Project coordinator: Prof.Dr. Paul Urbach (h.p.urbach@tudelft.nl)
The primary objectives of the ADOPSYS ITN were to develop new strategies for the inverse problem of optical design that reduce the amount of human trial-and-error effort significantly, and to apply them in close collaboration with the industrial partners to design problems that occur in modern applications, in particular energy efficient lighting systems, high resolution systems and machine vision, inspection and safety. In addition, improving the enabling methodology for the propagation of optical fields through optical systems (i.e. the direct problem in optics) was studied to increase the efficiency of the inverse design methodologies and hence to increase design productivity. The second, equally important, objective of this project was to build a unique European training infrastructure for PhDs in the field of optical design methods for a future career in European academy and photonics industry. The training network ADOPSYS trained young optical scientists and engineers in the full scope of optical modelling and design at the most advanced (i.e. PhD) level. The academic and industrial partners in ADOPSYS together covered the complete modern field of optical modelling and design and its applications, which no single European group or institute could provide alone.
The research fellows participated in the three Research Workshops, two General Scientific Trainings and the Training in Transferable Skill event organized by the consortium, and were enrolled in local training at each of the host institutes. These local trainings were often part of the Graduate Schools at the different host institutes and consisted of e.g. English courses and courses on how to write a scientific paper. Further, each fellow has carried out secondments at other full or associated partners.
The main results achieved by the ADOPSYS research programme are improved methods for both direct modelling of optical diffraction methods as for inverse design problems. The new methods for optical design concern both imaging and non-imaging optics.
In the modelling of diffraction problems, a method for cascaded diffraction was developed by ESR10. This method was studied in collaboration with Carl Zeiss and aims at incorporating in the model the effects of multiple diffraction at different limiting apertures in complicate systems. The method was first developed for scalar fields and was then extended to include polarisation effects. ESR10 has investigated a complimentary method base on the field tracing technique. A careful analysis was made to investigate the remaining errors in the obtained focused fields.
ESR1 has studied phase sensitive imaging techniques, in particular ptychography applied to soft X-ray and EUV systems. A major problem in this technique is to retrieve the probe positions, which often are not known with sufficient accuracy and this has a detrimental effect on the accuracy of the reconstructions. A new and fast method for retrieving the probe positions was developed. In the study of new methods for optical design for imaging systems, ESR2 concentrated on the design of large systems such as for lithography. A major issue is to obtain a sufficiently good starting design. The approach developed was to divide the optical systems in two parts. Both parts were first taken to consists of only a few lenses and more and more lenses were added depending on the transverse aberrations obtained for the entire field of view. Apart from publications, several patents have resulted from this work. ESR2 has collaborated with ESR3 to investigate the application of the Saddle Point Method (SPM) to complicated optical systems and a common paper is expected soon. ESR3 has further developed the SPM on challenging optical design systems of large field of view and on a-spheres. On the latter ESR3 has collaborated with ESR6 and a common paper is going to be published soon. Furthermore ESR3 has made a design during an internship at industrial partner Datalogic and a successful prototype has been made based on this design. ESR6 has investigated modifying the SMS method for application in imaging optics. SMS was developed for non-imaging optics. The SMS method was used to obtain good starting values for optical systems that consist of a-spheres. The new method was also further explored during an internship at Carl Zeiss. ESR7 has developed a new multi-fields direct design that is capable of partially coupling more than two fields with two surfaces in both two and three dimensions to obtain optical systems consisting of multiple free forms. The method was applied to designing a wide-field objective lens, achieving superior imaging performance when compared to its counterpart. Another promising design strategy was obtained by using paraxial aberration theory to get a good starting point for N > =3 freeform imaging systems. The project was carried out in a collaboration between VUB, technical University of Madrid and Carl Zeiss.
ESR9 has obtained new results on designing pupil masks for improved imaging. Improved images are obtained by deconvolution with appropriate levels of regularisation. An internship was done in KLA Tencor. ESR11 research concentrated on polarimetry, i.e. the imaging of polarisation contrast. The challenge was to image with broad band light where a trade-off must be found between accuracy in resolution and in polarisation contrast, due to the fact that polarisers are wavelength dependent. Application to microscopy was studied during an internship at Carl Zeiss.
The other ERSs worked on design methods for non-imaging optics. The challenge in the project of ESR4 was to obtain homogeneous illumination from inhomogeneous source distributions such as multicolor extended LEDs. To achieve this, the ESR proposed to use free form Shell-mixer. Furthermore, the SMS was extended to more surfaces which substantially improves the homogenuity. The work was carried out in a collaboration between the company LPI and VUB. ESR6 used novel integral equation methods as design method for free form surfaces to achieve homogeneous distributed light patterns. The surfaces were described using B-splines. This project was carried out in a collaboration between OSRAM and the Technical University Madrid. ESR7 has developed a new multi-fields direct design that is capable of partially coupling more than two fields with two surfaces in both two and three dimensions to obtain optical systems consisting of multiple free forms. The method was applied to designing a wide-field objective lens, achieving superior imaging performance when compared to its counterpart. Another promising design strategy was obtained by using paraxial aberration theory to get a good starting point for N > =3 freeform imaging systems. ESR12 has developed an interesting new algorithm for designing diffractive optical elements such as gratings which includes absorption of the materials involved. A new method to evaluate the mutual coherence function of a light beam using a diffractive element was also obtained. The research was carried out in a collaboration between UEF and FSU and an internship was done at OSRAM.
As illustration of the success of ADOPSYS, it should be mentioned that one of the ESRs, namely Yunfeng Nie (ESR7), obtained for her research work, the 2016 Michael Kidger Memorial Scholarship, selected annually to one worldwide PhD student in optical design domain. Furthermore, another ESR, ESR6, Milena Nikolic received a Cum Laude for her PhD thesis from partner UPM.
A lot of training activities which have been organised by the consortium which have been attended by all ESRs (with very few occasional assumptions). There have been three Research Workshops, two Scientific Trainings and one training on transferable skills. All activities were highly appreciated by the ESRs.
The ESR have organised regular Skype meetings between them and have elected a representative who attended the Board meeting. The atmosphere between the ESRs has been exceptionally good throughout the entire programme.
Project website: www.adopsys.eu
Project coordinator: Prof.Dr. Paul Urbach (h.p.urbach@tudelft.nl)