Final Report Summary - THIME (Thinfilm measurements on organic photovoltaic layers)
Executive Summary:
The THIME project developed an on-line quality control system for monitoring thinfilm thickness in roll-to-roll organic photovoltaics (OPV) production processes. The THIME system is capable of providing real time measurements across the full web width therefore obtaining a thickness map of the printed web which is not being used in the printing industry to date. The system is capable of substituting the currently used off-line thickness metrology instruments or on-line single point systems. Indeed, the THIME system is capable of measuring a layer thickness of up to 500nm with 1 standard deviation from the measurements of a commonly used industry reference method such as mechanical profilometry. The THIME system comprises two optical techniques such as spectral interferometry and spectroscopic ellipsometry and allows measurements with a standard resolution of 0.1 and 1 mm offering a reliable, stable system for monitoring in real time the thickness of the printed products. One of the most important features of the system is its versatility and possibility to be adapted to different materials and substrates with the proper calibration and modelling. Similarly, an important characteristic is the fact that the dedicated software allows the operator to analyse the output from the measurement system through a graphic interface and record its readings in order to study the thickness of a printed web during and after printing. The functioning of the THIME system has been proven to withstand the common industrial conditions and therefore achieved a readiness level close to the full industrial application of the technology.
The THIME system is comprised of two separate devices (Hyperspectral imager and ellipsometer) which can be installed at different stages of the printing line and are integrated in a single computing unit with a tailored software for analysing the output. All the readings from printing periods are stored in the computer and tagged with the date and printing name.
The THIME system was tested in 3 different installations in industrial environments and proved to fulfil the requirements of the printing operators. The system was installed after the oven step in the printing machine which allowed measuring the dry thickness which is unknown to most of the printing operators. This allowed for monitoring in real time the characteristics of the coated web therefore enhancing the control of the printing process by the operator and helping to reduce faulty products and material waste. It was demonstrated that the THIME system is able to provide more information with the same quality than off-line or point measurement systems allowing a real time and industrial compatible monitoring of this information. This functionality could result in closed loop control of the printing process in a future development allowing for a complete optimised production line.
The THIME system led to the business planning for a future commercial exploitation and could not only benefit the OPV producers but could be extended to all the printed electronics sector as thickness is a very important parameter in many different applications.
Project Context and Objectives:
Printing and other large-area roll-to-roll (R2R)-compatible processes present exciting opportunities for cost-efficient mass manufacturing of electronics, among other functionalities, on large-area and flexible substrates such as plastic, paper and fabrics. In particular, thin film Organic Photovoltaics (OPV) are generating a buzz in the industry. Printing the active components of a PV system onto flexible substrates means that solar cells could be incorporated onto a host of everyday objects, offing advantages in terms of weight, flexibility and low-cost production methods.
A major challenge for the manufacture of polymer and printed electronics is the ability to control the layer properties more precisely than with conventional colour printing. The performance of OPVs is strongly affected by the thickness and uniformity of the needed layers. Accurate information about the thickness of the thin films being deposited would prevent the production of large volumes of materials that do not perform to the standards that they should. Without online thin film thickness measurements the real thickness of the R2R deposited layer can be measured only after the deposition process.
The industry is in need of an online quality control method for thin film thickness on selected layers, which is vital for improved high quality, high volume, cost-effective production of such printed and large area electronics (OLAE) devices.
Based on this background the THIME project developed a novel optical instrument for the on-line measurement of thin film thickness during the R2R manufacturing of these devices, which is capable of measuring different OPV layers of differing characteristics values: thickness, refractive index, transparency and surface smoothness, and suited to a moving process (up to 10m/min) and is most likely not always in stable position in Z direction. No such detection method is available in the market and THIME is a breakthrough for advancing the EU industry. The system has been proven and tested at laboratory and pilot industrial scale with successful results in monitoring thickness of printing processes.
The overall objective of this project was to develop an efficient process control system that allowed for precise on-line thinfilm measurements on organic photovoltaic (OPV) layers, and that was suited for its future easy integration into roll-to-roll (R2R) machines. This THIME process control system would ensure greater uniformity and accurately deposited functional layers, which would results in higher operational performance of the final integrated circuits.
In view of these aims, the scientific and technological objectives that needed to be fulfilled within the THIME project are provided below:
1. To implement a market-driven approach whereby the requirements of SMEs from the organic thinfilm value chain will be consulted, along with the criteria for compliance with standards and regulations and to use the findings to define the specifications for the THIME system
2. To build a simple ellipsometer laboratory test rig and to perform laboratory work with this and other available ellipsometers in order to verify the parameters for scale-up of this set-up that would be suitable for providing on-line measurements of specified OPV layers with different characteristic values.
3. To build a hyperspectral measurement test rig and to perform laboratory work in order to verify the parameters for scale-up that would be suitable for proving on-line measurements of specified OPV layers in combination with the scale-up parameters defined in WP2.
4. To design and build the optical measurement hardware that fulfils the requirements for on-line measurement in a R2R environment, taking into account the parameters from the laboratory work with in WP2 and WP3 as well as the specifications defined from WP1.
5. To develop the general software for the operation of the THIME system, including the synchronized control of the different subsystems, the user application and the analytical software that will process the data obtained by the system. This analytical software will use the algorithms and needed databases developed during WP2, in order to characterize the properties of the thinfilms under analysis and readily display the results via an ergonomic User Interface.
6. To integrate the system hardware, software and User Interface in order to provide a pre-competitive prototype that can be validated in industry and to carry out pre-validation tests with the system to ensure its proper functioning before shipping to industry test-sites.
7. To install the THIME system in commercial R2R lines to test and validate the precompetitive prototype in industry, and to run trials with the system in order to demonstrate its ability for delivering on-line measurements to enable real-time control of the thickness and other properties of organic thin films, as well as its performance, usability and acceptance in the industry. The effectiveness of the on-line measurement with the THIME device will be compared to with conventional off-line methods.
8. To carefully outline scaling-up guidelines and development work for full production.
9. To facilitate the uptake of the THIME results by the participating SMEs as well as a wider audience by carrying out a comprehensive series of knowledge transfer and training activities to on the one hand show the validity for the system for performing fast, precise on-line measurements of organic thin film thickness and other parameters, and on the other hand to capacitate end-users about its usability.
Project Results:
Printing and other manufacturing technologies for large-area roll-to-roll (R2R)-compatible processes have been under extensive research during several years. Main motivation for the research has been foreseeable exciting opportunities for cost-efficient mass manufacturing of electronics, among other functionalities, on large-area and flexible substrates such as plastic, paper and fabrics. In particular, thin film Organic Photovoltaics (OPV) has generated great interest in the industry. Printing the active components of a PV system onto flexible substrates means that solar cells could be incorporated within everyday objects, offering advantages in terms of weight, flexibility and high volume production methods.
The starting point for the Science and Technology developments performed in THIME project was that printing and other large-area roll-to-roll (R2R)-compatible processes present exciting opportunities for European companies. However, a major challenge hindering the manufacture of polymer and printed electronics is the ability to control the layer properties more precisely than with conventional colour printing. As an example, the performance of OPVs is strongly affected by the thickness and uniformity of the needed layers. Accurate information about the thickness of the thin films being deposited would prevent the production of large volumes of materials that do not perform to the standards that they should. Without online thin film thickness measurements the real thickness of the R2R deposited layer can be measured only after the deposition process limiting seriously production the capacity of the manufacturing line.
THIME project was designed and performed to produce effective solutions for the industry need of an online quality control method for thin film thickness on selected layers vital for improved high quality, high volume and cost efficient production of organic and large area electronics (OLAE) devices.
The project started with WORK PACKAGE 1 by understanding of the technological needs of European OPV and printed electronics producers, as well as market needs in terms of on-line thickness measurement systems. In order to complement the findings of the questionnaire based survey among industry and the in-depth consultations (over 80 questionnaires were sent and 6 were returned filled) and face-to-face meetings, an extensive literature research was carried out. This revealed that thin films are widely used for various purposes and precise measurements of film thickness especially in a nondestructive way are crucial to ensure the intended functions of films. In the preparation of multilayered devices, real time measurement, monitoring, and control are important capabilities that allow one to: (i) obtain layer properties in the actual device configuration for correlation with device performance; (ii) develop new growth processes, assess reproducibility, and correlate variations to device performance variations; (iii) ensure that desired layer characteristics are met during the process, thus, improving yield.
Research also revealed that printing is a complex process, and therefore numerous factors may affect the thickness of printed layers. This includes the properties of the ink or varnish (e.g. viscosity), the ambient conditions (e.g. temperature), and technical factors such as the printing speed and the specific setting parameters of the printing press, which control the amount of varnish on the printing plate. Any change of one of these parameters may lead to a change of the resulting thickness of the printed layers. For many printing applications, in particular in case of top-quality products, it would be therefore desirable to monitor the thickness or the coating weight of the applied layers continuously during the printing process in order to be able to readjust the thickness and to ensure in this way a high degree of uniformity.
Based on the previously mentioned findings, a detailed list of specifications was defined with the collaboration of the SMEs in the Consortium. These included all the characteristics that the prototype should include in order to fulfill the requirements of the printing industry.
WORK PACKAGE 2 focussed on the development of the ellipsometer test rig in order to test it and optimise it at laboratory scale. This prototype was progressively improved throughout the project development but the ellipsometer used in the measurements worked with one wavelength generated with a pulsed laser-diode (currently with wavelength 905nm) and with a fixed angle of incidence (AOI, currently 75°). Both wavelength and AOI had to be chosen in advance for a given task.
The illumination part projected a line on the sample and the measurement side imaged this stripe with a bi-telecentric objective on the sensor. The measurement used laser-pulses triggered by the oscillation of the photo-elastic modulator (PEM), placed between first polarizer and sample. The PEM modulated the polarization of the through-going laser-beam, such that with a proper pulse-timing, well defined polarization states could be generated, namely 45°-linear, -45°-linear, left-circular-, and right-circular-polarization. By taking pictures with the analyzer at 45° with four different polarizations the following ratios were calculated:
R1 =(I45 – I-45)/(I45 + I-45) R2 = (IR – IL)/(IR + IL)
where I45, -45, H,R,L... are the corresponding reflected intensities. With these two ratios the ellipsometric parameters ψ and Δ, which are widely used to describe polarization or a polarization-changing sample, could be calculated. With a proper mathematical model the desired thickness could be determined from the ψ/Δ-measurement.
The modelling developed during the laboratory testing was based on assumptions about the optical constants (refractive index, absorption) of the materials and about the surface-quality. However, the sample materials were optically complex. Especially the PET may contain a primer, or added Si-particles (to avoid sticking between the PET-foils) or other sources of roughness. It was discovered that in some cases the boundaries between PET and layers might be not sharp, 1st due to roughness, 2nd due to chemical reactions which made complete modelling very demanding. The approach used here was therefore to check with a simple model (layer on PET-substrate with sharp boundaries, no inter-layers and well defined optical constants) the qualitative behaviour of ψ/Δ against layer-thickness. This led to a basic understanding of the measurement possibilities and the influence of the main measurement parameters, namely wavelength and AOI.
The laboratory trials revealed that the measurement was not unique, due to the oscillating character of ψ/Δ. So some pre-estimation of the thickness-range was necessary to determine the thickness uniquely, but as is the case with the HSI interferometry technique, the available thickness measurement range is much greater than the typical variation in thickness encountered for a manufacturing process. An advantage of this was quite a high sensitivity of the measurement to thickness-changes if the measurement was adjusted such that the average thickness is on the steepest slope of the ψ- or Δ-dependence on thickness.
The THIME-ellipsometer measurement performance was tested on two samples, namely P3HT:PCBM on PET and PEDOT on PET, the same materials used for the HSI validation. The overall result of the Work Package was an imaging ellipsometer capable of measuring thickness across the full web width in real time.
In WORK PACKAGE 3, the objective was to develop a Hyperspectral imaging (HSI) test rig and to carry out laboratory tests in order to provide the parameters for the scale up. Following this objective, a prototype of the HSI was developed which was based on hyperspectral camera with specular reflection geometry. This was considered to give an absolute thickness map, but was considered to need some calibration or extensive computation. Illumination system was targeted to be simple, low cost and to provide the required optical power for reflection interference measurements performed in VIS-NIR range for both OSC active material and PEDOT. Therefore, illumination system for precompetitive prototype was decided to be based on high power halogen lamp with specific reflector box and diffusor, although tests with LED arrays were performed. As for the imaging part, the selected camera, after extensive testing, was a SPECIM Imspector V10E in the visible wavelength range (400-1000nm). During the testing the work centered in obtaining a system capable of measuring a large surface area with high spatial resolution on a moving OPV web. In order to achieve the aforementioned a high resolution large area matrix detector or a long array detector was found to be typically required. Regardless of which detector type was used in the measurement, the Signal-to-Noise ratio (S/N) of the measurement system was one critical measure defining the overall performance of the system.
In order to increase performance of the measurement system the S/N had to be maximized. This was done by increasing signal using more power (light in optical measurements) and at the same time decreasing noise by optimizing bandwidth in the measurement.
The power performance of optical systems was evaluated by compiling and computing the so-called optical power budget. This provided information prior to the construction and purchasing of the components on the required optical power on the detector or camera element and also allowed to obtain a properly prepared optical power budget calculation which revealed where the power-critical points were in the optical design, and how the optical power budget could be improved.
Another important constraint was the resolution of the image which was highly dependant on the camera. When taking images of a moving web, the factors affecting the resolution are different depending on whether we are considering the machine direction (MD, the direction of web movement in the image) or cross direction (CD, the direction across the web, i.e. perpendicular to the movement of the web) in the image. Several considerations were tested and taken into account when choosing the camera:
General considerations:
- The higher the magnification in the camera, the better the pixel resolution in the image. However, when the magnification goes up, the field of view goes down, which means a smaller area of the web can be imaged.
- The real optical resolution in the image is often worse than the pixel resolution. The optical resolution might be affected by, among others:
o Poor focusing of the camera
o Web fluttering (going up and down)
o Depth of focus: the image is on focus at a certain distance interval only.
Line camera
- Machine direction
o The resolution is usually dependent on the speed of the web and the line frequency (how many measurement / second) of the camera. Since the image is built from many individual line snapshots, the higher the line frequency, the higher the resolution.
o The integration time of the camera (or the pulse duration of a flashing light source, whichever is shorter) also affects the resolution.
o The moving web smooths the features in the image, thereby affecting the resolution.
- Cross direction
o The resolution depends on the camera pixel size and the lens.
Matrix camera
- Machine direction
o The integration time of the camera (or the pulse duration of a flashing light source, whichever is shorter) affects the resolution.
o The moving web smooths the features in the image, thereby affecting the resolution.
- Cross direction
o The resolution depends on the camera pixel size and the lens.
After the performed evaluation and testing a line camera type was selected, therefore, the frame rate requirement was quite high for on-line measurements such as roll-to-roll applications.
An extensive array of laboratory trials for characterising materials and measuring procedure were planned and carried out with a wide combination of cameras, probes and algorithms. This allowed to define a first version of the software for evaluating thickness using the HSI that would be used in the future. The calculation of thickness based on the spectra measured was based on simple interference fringe peak / valley position and the calibration against a reference method. This was compared to a simulation of the interference patterns based on knowledge of the thin-film structure and the spectra measured made by a TFCompanion software package. These tests were performed on an ITO-PET sample, a PEDOT-PET sample and a P3HT-PCBM-PET sample. Several important conclusions were drawn after the intensive testing:
- Thickness references of the samples are needed to check the accuracy of the thickness calculation
- The complex refractive indices should be known in the full spectral range (600 – 900 nm for the OSC active material, 200 – 600 nm for PEDOT).
- The material is thought of as being isotropic in the simulation. In practice, some of the materials are birefringent and they might have a layered structure.
- The amount of information (number of fringes) might not be enough for a reliable measurement.
- The film thickness is not uniform. This can make the simulation complicated in the case of single-point measurements.
The objective of WORK PACKAGE 4 was to design and build the optical measurement hardware that fulfilled the requirements for online measurement in a roll-to-roll environment. To this end, the previously developed prototypes in WP2 and WP3 were scaled up to be installed in the selected industrial machines for validation and testing in an operational environment. The prototype was designed according to the requirements from the industrial partners based on the size and limitations of the environment where the system would be installed. An important aspect that was considered was the securing of the prototypes to the roll-to-roll machine and its isolation from potential vibrations that could affect the measurements. This was achieved by designing a solid aluminium frame structure and stabilising the web between dedicated rollers in order to measure in a stable surface. Similarly, it was important to guarantee a design that allowed to measure at least 100mm across the web length. This was achieved through optical simulation when scaling up the prototypes. Following the detailed design the construction of the THIME hardware took place. The two technologies were constructed as separated instruments following the designs created previously. The HSI was constructed on a stainless steel backplate with variable positions for the optical components in order to adjust them to the needs of the measurement. It included two rolls for bringing the web to the measurement plane and stabilising it. The whole system was covered by a stainless steel cover which conferred also optical stability and mechanical protection. Due to the high calorific power of the light source a fan was introduced inside the cover in order to cool down the air inside the measurement area and therefore protecting the web. Similarly, in order to protect the web against the heat form the light source, a tachometer measuring web movement was installed in order to stop the light source when the web movement was 0. The ellipsometer was constructed on aluminium frames with each component on a single pole and frame mounted on an optical rail. This allowed for a quick and precise adjustment of the sensible optical parts of the system during alignment. The whole system was also covered by a stainless steel cover isolating the system from the external light, and thus offering optical stability, while protecting the system from mechanical hazards and the operator from any stray laser light that could damage the eyes. The web was stabilised by using two small rolls on which the web would slide obtaining a tight tension gap where measures were taken. In both systems, the electronics were embedded into designated cabinets and powered through conventional power plugs. In addition to the scale up performed in WP4, a test-rig of a roll-to-roll system was built in order to test the system in dynamic conditions during web movement before the industrial installations.
The objective of WP5 was to develop the general software for the operation of the THIME system, including the user application and analytical software that would process the data obtained by the system and to integrate and calibrate the software, hardware and user interface. To this end, and following the requirements defined in WP1, the software was developed by integrating the algorithms created during WP2 and WP3. The software was designed to be a single package for controlling both systems but being able to individually operate with each system. The software stores all the outputs coming from the measuring systems and offers the possibility of analysing the data from previous runs in order to inspect past measurements. Similarly, the User Interface was developed taking into account the requirements from the SMEs in the project which affected the usability and experience from the operator. In this sense, the User Interface allows to graphically see the thickness variations, to control the access of operators in the system and to modify certain parameters of the system. The number of parameters that are accessible by the user depend on the type of access privileges given to the user. These range from basic user, advanced user or developer.
Finally, the hardware and software were integrated and tested on the roll-to-roll test rig constructed during WP4 to reproduce, at pilot scale, industrial printing conditions. The THIME system was tested and fine-tuned in order to optimise its performance for the approaching industrial validation tests. Optical, mechanical and electronic adjustments were made and software debugging took place during this period.
The objective of WORK PACKAGE 6 was to install the system in industrial roll-to-roll facilities and to extensively test the prototype to validate its performance. The trials took place in Germany, at COATEMA’s click’n’coat line (two installations) and in Finland in VTT’s printing facilities. In both installations, stability, accuracy and repeatability tests were carried out. All the installations involved static and dynamic testing (under non-moving and moving web conditions) and trials with different materials such as PEDOT on PET, P3HT:PCBM on PET or organic pink ink printing.
A thorough validation plan was designed and put in action in order to tests the performance and accuracy of both the HSI and ellipsometer as required by the SMEs in the Consortium. Reference thickness measurements for the OSC and PEDOT samples in this study were done with the mechanical profilometer Dektak 150 manufactured by Veeco (current supplier Bruker). This instrument is used in industry as a quick and easy to operate thickness control instrument, therefore its results would be familiar to many printing line operators.
The Dektak 150 Profilometer manufactured by Veeco (current supplier Bruker) was used as the primary reference thickness measurement instrument in this project. In this kind of a mechanical profilometer, a stylus (small radius tip) is moved to contact with the sample and then moved laterally across the sample with specific contact force. The height of the stylus generates an analog signal which is analyzed and converted to height profile or roughness information.
Four different sample sets were used in this project for validation of the THIME prototypes. First two sample sets were made at VTT with a Labratester laboratory printing device manufactured by Norbert Schläfli Maschinen. These two sample sets were the OSC active material (P3HT:PCBM) on PET foil and the printing method for these samples was gravure. The third sample set included a similar OSC active material roll, which was manufactured at VTT’s pilot-scale roll-to-roll facility with gravure printing. The fourth sample set was the PEDOT sample set on PET-foil from the company Nanolayer, SME partner of the THIME project. These samples were done with slot die coating, which produced smoother layers (small scale roughness) than gravure printing.
Three OSC sample sets were processed for tests. First sample set was processed by changing line density in gravure printing. This set contained 8 samples in total. Second sample set was processed by using same line density in printing and modifying concentration. This set contained 4 samples in total. Last sample included actually only one sample. It was an OSC stripe roll, which had been manufactured in the VTT roll-to-roll facility. All samples were made by preparing an OSC active material mixture (P3HT:PCBM 1:0.62) and dissolving it in different concentrations in 1,2-Dichlorobenzene (O-DCB). The solution was then applied on a PET foil using the Schläfli labratester.
The Dektak profilometer thickness results varied from 173.0 (#210 line density) to 554.8 nm (#70 line density) for the line density variation sample set. Standard deviation of measured thicknesses of samples varied from 12 to 26.2 nm for this sample set. Thickness results varied from 141.1 to 200.4 nm for the second sample set processed by modifying concentration. Standard deviation of measured thicknesses of samples varied from 8.4 to 12.0 nm for this sample set. Thickness result for the OSC stripe roll was 242.5 nm and standard deviation was 18.7 nm. This was a large sample of approximately 20 m, which was suitable for testing the performance of the instruments over a continuous period of up to 1 hour. The very same roll had been used in the validation trials at Coatema in Dormagen, Germany (April 2014) and at VTT in Oulu, Finland (August 2014). The Dektak thickness measurements were taken from ten different positions, and of course it only represented a very small part of the roll. However, it was expected that the thickness would be approximately on the same level throughout the roll.
Nanolayer provided a set of PEDOT samples to be used as an alternative test set for the THIME prototypes. PEDOT:PSS samples were printed on PET substrate using slot die coating. Three different material references were used in samples namely F52, F58 and S305. The reference of the sample referred to either different material supplier or different formulations (only in terms of material concentration or type of solvent used). The Dektak profilometer was used to measure the thickness of this sample set also. The F52 samples were prepared with changing the speed of the coating machine with the aim of producing four different thickness levels. This was actually achieved, due the fact that thickness values varied from 101.1 to 361.5 nm. Standard deviation of measured thicknesses of samples varied from 14.4 to 34.1 nm for this sample set. In addition, it was possible to compare different PEDOT:PSS formulations coated on PET at the same coating speed of 0.8 m/min.
The thickness calibration model used in WP3 was used in the initial validation measurements and tests for HSI system. This calibration model will be referred to as the “original calibration model” hereinafter. Based on the validation tests, three other calibration models were developed for OSC active material on PET and for PEDOT on PET.
The interference features achieved by the implemented precompetitive prototype in validation measurements were really clear, and the developed interference peak finding algorithm could easily find the interference peak maxima. Based on the peaks found, the original calibration model was able to compute the thickness of the samples. The thickness computed with the original calibration model was compared against the Dektak reference thickness.
The mean thickness was calculated from the thickness profile results, abandoning zero values where the algorithm didn’t find any peaks. The results were an average over 15 different positions on the samples, which was comparable to the Dektak measurements (Dektak measurements were measured from 10 different positions).
The results corresponded quite well with the Dektak values, and the deviation was the smaller the thinner the sample was. This was understandable, since the thicker samples are relatively non-uniform, and because the Dektak thickness was not measured from the same position as the HSI thickness profiles. The root mean square (RMS) error was 7.4 nm and 2.5 % relative to the mean thickness of the sample set.
The HSI system achieved an RMS error of 7.4 nm (2.5 % relative to the mean thickness of the sample set) for the OSC active material thickness measurement. The corresponding figures for the ellipsometer system were 16.6 nm or 5.7 %. The PEDOT samples could be measured with an RMS error of 34.6 nm (12.0 % relative to the mean thickness of the sample set) for the HSI. The corresponding figures for the ellipsometer system were 18.3 nm or 8.2%. It can be concluded that the HSI system is slightly better in measuring the OSC active material, whereas the ellipsometer system seems to be much better in measuring the PEDOT samples.
The repeatability of the THIME HSI measurement was studied by computing different standard deviation values based on stationary samples and on samples that were moved during the measurement. Several hundred measurements were taken, and the resulting thickness map was then analyzed. Clear outliers were removed from the thickness map. Similarly, the lateral resolution of both systems was studied obtaining a millimeter scale resolution for both systems over a large surface area web (100-250 mm), clearly indicating a very interesting result in comparison with the currently existing thickness measurement methods.
The main objective of WORK PACKAGE 7 was to carry out a comprehensive series of knowledge transfer and training activities in order to prove the industrial viability of THIME, to highlight its technical performance features, as well as to outline its potential and limitations.
The THIME project consortium planned three demonstration events of project results and achievements to public. The demonstration sessions were prepared as practical workshops that included an introduction to the project, a summary of the results obtained in the laboratory and the results obtained with the THIME prototype in terms of resolution, repeatability of measures, vibration resistance and advantages over other existing thickness monitoring techniques. Practical demonstrations of the operational prototype were also carried out.
Potential Impact:
There are numerous socio-economic impacts that will be derived from the results of this THIME research project. Central to the expected socio-economic impacts is the boosting of the competitiveness of OPV companies along the printed electronics value chain- by improving the production efficiency and quality. Given the fast development of this sector in the present and near future, there is a need of optimising the production processes to compete against other printed electronics companies in the market, especially from Asia and North America. Overall, the European printed electronics and OPV sector faces a delicate situation and needs to benefit from any technology or system capable of enhancing their efficiency and quality and therefore overall income.
In 2010, it was estimated that between 1,100 and 1,400 companies were operating globally in the printed electronics industry- many of the based in Asia, North America and Europe (with Germany and the UK being the leading printed electronics markets in Europe). If we consider that a Compound Annual Growth Rate (CAGR) of 58% was expected between the period 2010 and 2016, it is expected that by the time the THIME product is launched in the market, there could be at least some 4000 companies, based on growth predictions for the industry. Should THIME achieve an uptake of 3-4% among this market within 5 years of entry, some 140 units could be sold, installed and serviced in the market by 2020 valued at € 14 M. This value is based on a unit cost of €100,000 (€12 M), in addition to maintenance and service contracts (€2M). This is deemed achievable through sound marketing and in light of the unique THIME technology offering an on-line measurement method that offers multiple benefits in comparison to offline methods. Thus, especially the EU printed electronics industry but also the worldwide sector can benefit from the THIME technology in order to improve their productivity and competitiveness.
Taking into account the economic aspects and the benefits for the OPV and printed electronics industry by achieving for example a reduced active material consumption or an increased production control, the THIME technology is expected to have a significant impact on printed electronics industry, which is demanding of such improvements to reduce printing faulty production, production costs and enhance chip performance, and will therefore have a direct positive effect of the environment and on the industry competitivity.
Moreover, the impact of THIME will not only affect EU OPV or printed electronics producers, but it can have a global effect on the worldwide printing community. Production is geographically concentrated in Eastern Asia, United States, Germany and the UK. If the THIME system owners (COATEMA, ACCURION, SPECIM, NANOLAYER and SOLAR) manage to commercially exploit the system with a global patent, the sales could boost to all the producers across the printed electronics countries. This would imply a large benefit for the participating SMEs while ensuring an improved management of the production and resource efficiency worldwide.
The economic impact associated to THIME will expand to several sectors across the value chain apart from the printing end-users:
i) Producers of raw materials, components and equipment:
The electronics and optics producers will benefit from the demand of components for the TIHME system which will require the elements for the HSI and ellipsometer system, from the cameras and sensors to the lenses and electronic boards. Moreover, the steel and aluminium producers will provide the housing and structure for the instruments.
ii) Manufacturing and distribution companies:
The manufacturing companies, will have to produce and program the THIME hardware and software and deliver an off the shelf package for the final users or coating machinery companies to distribute the system to the end users. This will increase their activity and number of orders and in return increase their profit substantially.
iii) Coating machinery suppliers
The responsible companies for delivering the THIME system to the final users will be those dealing directly with OPV or printed electronics producers, mainly supplying the THIME system as a quality control instrument added onto their coating machines.
The increased productivity and quality generated by the more accurate control of thickness in the printing process will directly affect the job creation and qualified employment, offering a social benefit to the THIME system. In addition to the socio-economic benefits, the THIME system offers also environmental profits. Firstly, it will optimise raw materials usage by controlling the amount of material deposited during the printing process and resulting, in some cases, to raw material usage reduction. Secondly, it will reduce energy usage from producing non-conforming products that need to be produced again to fulfil the quality requirements. Thirdly, the waste generated by the faulty products will be reduced thus decreasing the impact over the lifecycle of the product. Finally, the concept of smart and sustainable OPV production will be enriched by the usage of THIME. This will stimulate the consumption of “greener” products in the market and will have a positive effect on the environment and in the printed electronics producers.
Finally, the THIME monitoring technology can be exported to other economic sectors such any industrial process demanding thickness measurements. This could be done by the THIME instrument and would definitely become a major source of profits for the owners of the system will offering an improved productivity and yield to those sectors using the THIME technology.
Main dissemination and exploitation activities
Each partner has expressed its intentions and wishes for the exploitation strategy. In addition, it allowed a great deal of dissemination activities to be carried out.
The current status of the protection and exploitation strategy for the THIME foreground within the Consortium can be summarized as follows:
- The HSI part of the system will be developed autonomously by the SME Specim and an external integrator company, as described in the Joint Exploitation Agreement.
- The ellipsometer part of the system will be developed autonomously by the SME Accurion for its commercial production.
- The SMEs Coatema and Nanolayer will integrate the instruments in their coating products and offer it to their clients.
- The SME Solar will benefit from preferential access to the technology such as early access or discount conditions.
All these roles and strategy are defined in the Joint Exploitation Agreement signed by all partners.
A really successful dissemination strategy has been carried out during the THIME project maximising the exploitability of the material and demonstrating its potential to provide OPV producers a tool for thickness monitoring and control in their production process and thus increasing productivity and quality of OPV and printed products.
Over the course of the project over 254 interested companies, stakeholders and policy makers were directly contacted by the THIME Consortium members in conferences, trade fairs, personal meetings and workshops. Moreover, a larger number were indirectly aware of the THIME project via 8 press releases in the general media, 2 presentations in conferences, and 1 scientific article published. The details can be found in the list annexed or in the Part A of the PUDF.
The project has attracted a large interest from the OPV, printed electronics and scientific community and has been in contact with other FP7 projects to find synergies and evaluate the possibility of uniting the different developed technologies to provide producers with a unique, complete tool.
Project Coordinator:
Mr. Pau Puigdollers de Balle
ppuigdollers@iris.cat
+34 935542500
IRIS Innovació i Recerca Industrial i Sostenible
Avda. Carl Friedrich Gauss n. 11
08860 Castelldefels
Spain
The THIME project developed an on-line quality control system for monitoring thinfilm thickness in roll-to-roll organic photovoltaics (OPV) production processes. The THIME system is capable of providing real time measurements across the full web width therefore obtaining a thickness map of the printed web which is not being used in the printing industry to date. The system is capable of substituting the currently used off-line thickness metrology instruments or on-line single point systems. Indeed, the THIME system is capable of measuring a layer thickness of up to 500nm with 1 standard deviation from the measurements of a commonly used industry reference method such as mechanical profilometry. The THIME system comprises two optical techniques such as spectral interferometry and spectroscopic ellipsometry and allows measurements with a standard resolution of 0.1 and 1 mm offering a reliable, stable system for monitoring in real time the thickness of the printed products. One of the most important features of the system is its versatility and possibility to be adapted to different materials and substrates with the proper calibration and modelling. Similarly, an important characteristic is the fact that the dedicated software allows the operator to analyse the output from the measurement system through a graphic interface and record its readings in order to study the thickness of a printed web during and after printing. The functioning of the THIME system has been proven to withstand the common industrial conditions and therefore achieved a readiness level close to the full industrial application of the technology.
The THIME system is comprised of two separate devices (Hyperspectral imager and ellipsometer) which can be installed at different stages of the printing line and are integrated in a single computing unit with a tailored software for analysing the output. All the readings from printing periods are stored in the computer and tagged with the date and printing name.
The THIME system was tested in 3 different installations in industrial environments and proved to fulfil the requirements of the printing operators. The system was installed after the oven step in the printing machine which allowed measuring the dry thickness which is unknown to most of the printing operators. This allowed for monitoring in real time the characteristics of the coated web therefore enhancing the control of the printing process by the operator and helping to reduce faulty products and material waste. It was demonstrated that the THIME system is able to provide more information with the same quality than off-line or point measurement systems allowing a real time and industrial compatible monitoring of this information. This functionality could result in closed loop control of the printing process in a future development allowing for a complete optimised production line.
The THIME system led to the business planning for a future commercial exploitation and could not only benefit the OPV producers but could be extended to all the printed electronics sector as thickness is a very important parameter in many different applications.
Project Context and Objectives:
Printing and other large-area roll-to-roll (R2R)-compatible processes present exciting opportunities for cost-efficient mass manufacturing of electronics, among other functionalities, on large-area and flexible substrates such as plastic, paper and fabrics. In particular, thin film Organic Photovoltaics (OPV) are generating a buzz in the industry. Printing the active components of a PV system onto flexible substrates means that solar cells could be incorporated onto a host of everyday objects, offing advantages in terms of weight, flexibility and low-cost production methods.
A major challenge for the manufacture of polymer and printed electronics is the ability to control the layer properties more precisely than with conventional colour printing. The performance of OPVs is strongly affected by the thickness and uniformity of the needed layers. Accurate information about the thickness of the thin films being deposited would prevent the production of large volumes of materials that do not perform to the standards that they should. Without online thin film thickness measurements the real thickness of the R2R deposited layer can be measured only after the deposition process.
The industry is in need of an online quality control method for thin film thickness on selected layers, which is vital for improved high quality, high volume, cost-effective production of such printed and large area electronics (OLAE) devices.
Based on this background the THIME project developed a novel optical instrument for the on-line measurement of thin film thickness during the R2R manufacturing of these devices, which is capable of measuring different OPV layers of differing characteristics values: thickness, refractive index, transparency and surface smoothness, and suited to a moving process (up to 10m/min) and is most likely not always in stable position in Z direction. No such detection method is available in the market and THIME is a breakthrough for advancing the EU industry. The system has been proven and tested at laboratory and pilot industrial scale with successful results in monitoring thickness of printing processes.
The overall objective of this project was to develop an efficient process control system that allowed for precise on-line thinfilm measurements on organic photovoltaic (OPV) layers, and that was suited for its future easy integration into roll-to-roll (R2R) machines. This THIME process control system would ensure greater uniformity and accurately deposited functional layers, which would results in higher operational performance of the final integrated circuits.
In view of these aims, the scientific and technological objectives that needed to be fulfilled within the THIME project are provided below:
1. To implement a market-driven approach whereby the requirements of SMEs from the organic thinfilm value chain will be consulted, along with the criteria for compliance with standards and regulations and to use the findings to define the specifications for the THIME system
2. To build a simple ellipsometer laboratory test rig and to perform laboratory work with this and other available ellipsometers in order to verify the parameters for scale-up of this set-up that would be suitable for providing on-line measurements of specified OPV layers with different characteristic values.
3. To build a hyperspectral measurement test rig and to perform laboratory work in order to verify the parameters for scale-up that would be suitable for proving on-line measurements of specified OPV layers in combination with the scale-up parameters defined in WP2.
4. To design and build the optical measurement hardware that fulfils the requirements for on-line measurement in a R2R environment, taking into account the parameters from the laboratory work with in WP2 and WP3 as well as the specifications defined from WP1.
5. To develop the general software for the operation of the THIME system, including the synchronized control of the different subsystems, the user application and the analytical software that will process the data obtained by the system. This analytical software will use the algorithms and needed databases developed during WP2, in order to characterize the properties of the thinfilms under analysis and readily display the results via an ergonomic User Interface.
6. To integrate the system hardware, software and User Interface in order to provide a pre-competitive prototype that can be validated in industry and to carry out pre-validation tests with the system to ensure its proper functioning before shipping to industry test-sites.
7. To install the THIME system in commercial R2R lines to test and validate the precompetitive prototype in industry, and to run trials with the system in order to demonstrate its ability for delivering on-line measurements to enable real-time control of the thickness and other properties of organic thin films, as well as its performance, usability and acceptance in the industry. The effectiveness of the on-line measurement with the THIME device will be compared to with conventional off-line methods.
8. To carefully outline scaling-up guidelines and development work for full production.
9. To facilitate the uptake of the THIME results by the participating SMEs as well as a wider audience by carrying out a comprehensive series of knowledge transfer and training activities to on the one hand show the validity for the system for performing fast, precise on-line measurements of organic thin film thickness and other parameters, and on the other hand to capacitate end-users about its usability.
Project Results:
Printing and other manufacturing technologies for large-area roll-to-roll (R2R)-compatible processes have been under extensive research during several years. Main motivation for the research has been foreseeable exciting opportunities for cost-efficient mass manufacturing of electronics, among other functionalities, on large-area and flexible substrates such as plastic, paper and fabrics. In particular, thin film Organic Photovoltaics (OPV) has generated great interest in the industry. Printing the active components of a PV system onto flexible substrates means that solar cells could be incorporated within everyday objects, offering advantages in terms of weight, flexibility and high volume production methods.
The starting point for the Science and Technology developments performed in THIME project was that printing and other large-area roll-to-roll (R2R)-compatible processes present exciting opportunities for European companies. However, a major challenge hindering the manufacture of polymer and printed electronics is the ability to control the layer properties more precisely than with conventional colour printing. As an example, the performance of OPVs is strongly affected by the thickness and uniformity of the needed layers. Accurate information about the thickness of the thin films being deposited would prevent the production of large volumes of materials that do not perform to the standards that they should. Without online thin film thickness measurements the real thickness of the R2R deposited layer can be measured only after the deposition process limiting seriously production the capacity of the manufacturing line.
THIME project was designed and performed to produce effective solutions for the industry need of an online quality control method for thin film thickness on selected layers vital for improved high quality, high volume and cost efficient production of organic and large area electronics (OLAE) devices.
The project started with WORK PACKAGE 1 by understanding of the technological needs of European OPV and printed electronics producers, as well as market needs in terms of on-line thickness measurement systems. In order to complement the findings of the questionnaire based survey among industry and the in-depth consultations (over 80 questionnaires were sent and 6 were returned filled) and face-to-face meetings, an extensive literature research was carried out. This revealed that thin films are widely used for various purposes and precise measurements of film thickness especially in a nondestructive way are crucial to ensure the intended functions of films. In the preparation of multilayered devices, real time measurement, monitoring, and control are important capabilities that allow one to: (i) obtain layer properties in the actual device configuration for correlation with device performance; (ii) develop new growth processes, assess reproducibility, and correlate variations to device performance variations; (iii) ensure that desired layer characteristics are met during the process, thus, improving yield.
Research also revealed that printing is a complex process, and therefore numerous factors may affect the thickness of printed layers. This includes the properties of the ink or varnish (e.g. viscosity), the ambient conditions (e.g. temperature), and technical factors such as the printing speed and the specific setting parameters of the printing press, which control the amount of varnish on the printing plate. Any change of one of these parameters may lead to a change of the resulting thickness of the printed layers. For many printing applications, in particular in case of top-quality products, it would be therefore desirable to monitor the thickness or the coating weight of the applied layers continuously during the printing process in order to be able to readjust the thickness and to ensure in this way a high degree of uniformity.
Based on the previously mentioned findings, a detailed list of specifications was defined with the collaboration of the SMEs in the Consortium. These included all the characteristics that the prototype should include in order to fulfill the requirements of the printing industry.
WORK PACKAGE 2 focussed on the development of the ellipsometer test rig in order to test it and optimise it at laboratory scale. This prototype was progressively improved throughout the project development but the ellipsometer used in the measurements worked with one wavelength generated with a pulsed laser-diode (currently with wavelength 905nm) and with a fixed angle of incidence (AOI, currently 75°). Both wavelength and AOI had to be chosen in advance for a given task.
The illumination part projected a line on the sample and the measurement side imaged this stripe with a bi-telecentric objective on the sensor. The measurement used laser-pulses triggered by the oscillation of the photo-elastic modulator (PEM), placed between first polarizer and sample. The PEM modulated the polarization of the through-going laser-beam, such that with a proper pulse-timing, well defined polarization states could be generated, namely 45°-linear, -45°-linear, left-circular-, and right-circular-polarization. By taking pictures with the analyzer at 45° with four different polarizations the following ratios were calculated:
R1 =(I45 – I-45)/(I45 + I-45) R2 = (IR – IL)/(IR + IL)
where I45, -45, H,R,L... are the corresponding reflected intensities. With these two ratios the ellipsometric parameters ψ and Δ, which are widely used to describe polarization or a polarization-changing sample, could be calculated. With a proper mathematical model the desired thickness could be determined from the ψ/Δ-measurement.
The modelling developed during the laboratory testing was based on assumptions about the optical constants (refractive index, absorption) of the materials and about the surface-quality. However, the sample materials were optically complex. Especially the PET may contain a primer, or added Si-particles (to avoid sticking between the PET-foils) or other sources of roughness. It was discovered that in some cases the boundaries between PET and layers might be not sharp, 1st due to roughness, 2nd due to chemical reactions which made complete modelling very demanding. The approach used here was therefore to check with a simple model (layer on PET-substrate with sharp boundaries, no inter-layers and well defined optical constants) the qualitative behaviour of ψ/Δ against layer-thickness. This led to a basic understanding of the measurement possibilities and the influence of the main measurement parameters, namely wavelength and AOI.
The laboratory trials revealed that the measurement was not unique, due to the oscillating character of ψ/Δ. So some pre-estimation of the thickness-range was necessary to determine the thickness uniquely, but as is the case with the HSI interferometry technique, the available thickness measurement range is much greater than the typical variation in thickness encountered for a manufacturing process. An advantage of this was quite a high sensitivity of the measurement to thickness-changes if the measurement was adjusted such that the average thickness is on the steepest slope of the ψ- or Δ-dependence on thickness.
The THIME-ellipsometer measurement performance was tested on two samples, namely P3HT:PCBM on PET and PEDOT on PET, the same materials used for the HSI validation. The overall result of the Work Package was an imaging ellipsometer capable of measuring thickness across the full web width in real time.
In WORK PACKAGE 3, the objective was to develop a Hyperspectral imaging (HSI) test rig and to carry out laboratory tests in order to provide the parameters for the scale up. Following this objective, a prototype of the HSI was developed which was based on hyperspectral camera with specular reflection geometry. This was considered to give an absolute thickness map, but was considered to need some calibration or extensive computation. Illumination system was targeted to be simple, low cost and to provide the required optical power for reflection interference measurements performed in VIS-NIR range for both OSC active material and PEDOT. Therefore, illumination system for precompetitive prototype was decided to be based on high power halogen lamp with specific reflector box and diffusor, although tests with LED arrays were performed. As for the imaging part, the selected camera, after extensive testing, was a SPECIM Imspector V10E in the visible wavelength range (400-1000nm). During the testing the work centered in obtaining a system capable of measuring a large surface area with high spatial resolution on a moving OPV web. In order to achieve the aforementioned a high resolution large area matrix detector or a long array detector was found to be typically required. Regardless of which detector type was used in the measurement, the Signal-to-Noise ratio (S/N) of the measurement system was one critical measure defining the overall performance of the system.
In order to increase performance of the measurement system the S/N had to be maximized. This was done by increasing signal using more power (light in optical measurements) and at the same time decreasing noise by optimizing bandwidth in the measurement.
The power performance of optical systems was evaluated by compiling and computing the so-called optical power budget. This provided information prior to the construction and purchasing of the components on the required optical power on the detector or camera element and also allowed to obtain a properly prepared optical power budget calculation which revealed where the power-critical points were in the optical design, and how the optical power budget could be improved.
Another important constraint was the resolution of the image which was highly dependant on the camera. When taking images of a moving web, the factors affecting the resolution are different depending on whether we are considering the machine direction (MD, the direction of web movement in the image) or cross direction (CD, the direction across the web, i.e. perpendicular to the movement of the web) in the image. Several considerations were tested and taken into account when choosing the camera:
General considerations:
- The higher the magnification in the camera, the better the pixel resolution in the image. However, when the magnification goes up, the field of view goes down, which means a smaller area of the web can be imaged.
- The real optical resolution in the image is often worse than the pixel resolution. The optical resolution might be affected by, among others:
o Poor focusing of the camera
o Web fluttering (going up and down)
o Depth of focus: the image is on focus at a certain distance interval only.
Line camera
- Machine direction
o The resolution is usually dependent on the speed of the web and the line frequency (how many measurement / second) of the camera. Since the image is built from many individual line snapshots, the higher the line frequency, the higher the resolution.
o The integration time of the camera (or the pulse duration of a flashing light source, whichever is shorter) also affects the resolution.
o The moving web smooths the features in the image, thereby affecting the resolution.
- Cross direction
o The resolution depends on the camera pixel size and the lens.
Matrix camera
- Machine direction
o The integration time of the camera (or the pulse duration of a flashing light source, whichever is shorter) affects the resolution.
o The moving web smooths the features in the image, thereby affecting the resolution.
- Cross direction
o The resolution depends on the camera pixel size and the lens.
After the performed evaluation and testing a line camera type was selected, therefore, the frame rate requirement was quite high for on-line measurements such as roll-to-roll applications.
An extensive array of laboratory trials for characterising materials and measuring procedure were planned and carried out with a wide combination of cameras, probes and algorithms. This allowed to define a first version of the software for evaluating thickness using the HSI that would be used in the future. The calculation of thickness based on the spectra measured was based on simple interference fringe peak / valley position and the calibration against a reference method. This was compared to a simulation of the interference patterns based on knowledge of the thin-film structure and the spectra measured made by a TFCompanion software package. These tests were performed on an ITO-PET sample, a PEDOT-PET sample and a P3HT-PCBM-PET sample. Several important conclusions were drawn after the intensive testing:
- Thickness references of the samples are needed to check the accuracy of the thickness calculation
- The complex refractive indices should be known in the full spectral range (600 – 900 nm for the OSC active material, 200 – 600 nm for PEDOT).
- The material is thought of as being isotropic in the simulation. In practice, some of the materials are birefringent and they might have a layered structure.
- The amount of information (number of fringes) might not be enough for a reliable measurement.
- The film thickness is not uniform. This can make the simulation complicated in the case of single-point measurements.
The objective of WORK PACKAGE 4 was to design and build the optical measurement hardware that fulfilled the requirements for online measurement in a roll-to-roll environment. To this end, the previously developed prototypes in WP2 and WP3 were scaled up to be installed in the selected industrial machines for validation and testing in an operational environment. The prototype was designed according to the requirements from the industrial partners based on the size and limitations of the environment where the system would be installed. An important aspect that was considered was the securing of the prototypes to the roll-to-roll machine and its isolation from potential vibrations that could affect the measurements. This was achieved by designing a solid aluminium frame structure and stabilising the web between dedicated rollers in order to measure in a stable surface. Similarly, it was important to guarantee a design that allowed to measure at least 100mm across the web length. This was achieved through optical simulation when scaling up the prototypes. Following the detailed design the construction of the THIME hardware took place. The two technologies were constructed as separated instruments following the designs created previously. The HSI was constructed on a stainless steel backplate with variable positions for the optical components in order to adjust them to the needs of the measurement. It included two rolls for bringing the web to the measurement plane and stabilising it. The whole system was covered by a stainless steel cover which conferred also optical stability and mechanical protection. Due to the high calorific power of the light source a fan was introduced inside the cover in order to cool down the air inside the measurement area and therefore protecting the web. Similarly, in order to protect the web against the heat form the light source, a tachometer measuring web movement was installed in order to stop the light source when the web movement was 0. The ellipsometer was constructed on aluminium frames with each component on a single pole and frame mounted on an optical rail. This allowed for a quick and precise adjustment of the sensible optical parts of the system during alignment. The whole system was also covered by a stainless steel cover isolating the system from the external light, and thus offering optical stability, while protecting the system from mechanical hazards and the operator from any stray laser light that could damage the eyes. The web was stabilised by using two small rolls on which the web would slide obtaining a tight tension gap where measures were taken. In both systems, the electronics were embedded into designated cabinets and powered through conventional power plugs. In addition to the scale up performed in WP4, a test-rig of a roll-to-roll system was built in order to test the system in dynamic conditions during web movement before the industrial installations.
The objective of WP5 was to develop the general software for the operation of the THIME system, including the user application and analytical software that would process the data obtained by the system and to integrate and calibrate the software, hardware and user interface. To this end, and following the requirements defined in WP1, the software was developed by integrating the algorithms created during WP2 and WP3. The software was designed to be a single package for controlling both systems but being able to individually operate with each system. The software stores all the outputs coming from the measuring systems and offers the possibility of analysing the data from previous runs in order to inspect past measurements. Similarly, the User Interface was developed taking into account the requirements from the SMEs in the project which affected the usability and experience from the operator. In this sense, the User Interface allows to graphically see the thickness variations, to control the access of operators in the system and to modify certain parameters of the system. The number of parameters that are accessible by the user depend on the type of access privileges given to the user. These range from basic user, advanced user or developer.
Finally, the hardware and software were integrated and tested on the roll-to-roll test rig constructed during WP4 to reproduce, at pilot scale, industrial printing conditions. The THIME system was tested and fine-tuned in order to optimise its performance for the approaching industrial validation tests. Optical, mechanical and electronic adjustments were made and software debugging took place during this period.
The objective of WORK PACKAGE 6 was to install the system in industrial roll-to-roll facilities and to extensively test the prototype to validate its performance. The trials took place in Germany, at COATEMA’s click’n’coat line (two installations) and in Finland in VTT’s printing facilities. In both installations, stability, accuracy and repeatability tests were carried out. All the installations involved static and dynamic testing (under non-moving and moving web conditions) and trials with different materials such as PEDOT on PET, P3HT:PCBM on PET or organic pink ink printing.
A thorough validation plan was designed and put in action in order to tests the performance and accuracy of both the HSI and ellipsometer as required by the SMEs in the Consortium. Reference thickness measurements for the OSC and PEDOT samples in this study were done with the mechanical profilometer Dektak 150 manufactured by Veeco (current supplier Bruker). This instrument is used in industry as a quick and easy to operate thickness control instrument, therefore its results would be familiar to many printing line operators.
The Dektak 150 Profilometer manufactured by Veeco (current supplier Bruker) was used as the primary reference thickness measurement instrument in this project. In this kind of a mechanical profilometer, a stylus (small radius tip) is moved to contact with the sample and then moved laterally across the sample with specific contact force. The height of the stylus generates an analog signal which is analyzed and converted to height profile or roughness information.
Four different sample sets were used in this project for validation of the THIME prototypes. First two sample sets were made at VTT with a Labratester laboratory printing device manufactured by Norbert Schläfli Maschinen. These two sample sets were the OSC active material (P3HT:PCBM) on PET foil and the printing method for these samples was gravure. The third sample set included a similar OSC active material roll, which was manufactured at VTT’s pilot-scale roll-to-roll facility with gravure printing. The fourth sample set was the PEDOT sample set on PET-foil from the company Nanolayer, SME partner of the THIME project. These samples were done with slot die coating, which produced smoother layers (small scale roughness) than gravure printing.
Three OSC sample sets were processed for tests. First sample set was processed by changing line density in gravure printing. This set contained 8 samples in total. Second sample set was processed by using same line density in printing and modifying concentration. This set contained 4 samples in total. Last sample included actually only one sample. It was an OSC stripe roll, which had been manufactured in the VTT roll-to-roll facility. All samples were made by preparing an OSC active material mixture (P3HT:PCBM 1:0.62) and dissolving it in different concentrations in 1,2-Dichlorobenzene (O-DCB). The solution was then applied on a PET foil using the Schläfli labratester.
The Dektak profilometer thickness results varied from 173.0 (#210 line density) to 554.8 nm (#70 line density) for the line density variation sample set. Standard deviation of measured thicknesses of samples varied from 12 to 26.2 nm for this sample set. Thickness results varied from 141.1 to 200.4 nm for the second sample set processed by modifying concentration. Standard deviation of measured thicknesses of samples varied from 8.4 to 12.0 nm for this sample set. Thickness result for the OSC stripe roll was 242.5 nm and standard deviation was 18.7 nm. This was a large sample of approximately 20 m, which was suitable for testing the performance of the instruments over a continuous period of up to 1 hour. The very same roll had been used in the validation trials at Coatema in Dormagen, Germany (April 2014) and at VTT in Oulu, Finland (August 2014). The Dektak thickness measurements were taken from ten different positions, and of course it only represented a very small part of the roll. However, it was expected that the thickness would be approximately on the same level throughout the roll.
Nanolayer provided a set of PEDOT samples to be used as an alternative test set for the THIME prototypes. PEDOT:PSS samples were printed on PET substrate using slot die coating. Three different material references were used in samples namely F52, F58 and S305. The reference of the sample referred to either different material supplier or different formulations (only in terms of material concentration or type of solvent used). The Dektak profilometer was used to measure the thickness of this sample set also. The F52 samples were prepared with changing the speed of the coating machine with the aim of producing four different thickness levels. This was actually achieved, due the fact that thickness values varied from 101.1 to 361.5 nm. Standard deviation of measured thicknesses of samples varied from 14.4 to 34.1 nm for this sample set. In addition, it was possible to compare different PEDOT:PSS formulations coated on PET at the same coating speed of 0.8 m/min.
The thickness calibration model used in WP3 was used in the initial validation measurements and tests for HSI system. This calibration model will be referred to as the “original calibration model” hereinafter. Based on the validation tests, three other calibration models were developed for OSC active material on PET and for PEDOT on PET.
The interference features achieved by the implemented precompetitive prototype in validation measurements were really clear, and the developed interference peak finding algorithm could easily find the interference peak maxima. Based on the peaks found, the original calibration model was able to compute the thickness of the samples. The thickness computed with the original calibration model was compared against the Dektak reference thickness.
The mean thickness was calculated from the thickness profile results, abandoning zero values where the algorithm didn’t find any peaks. The results were an average over 15 different positions on the samples, which was comparable to the Dektak measurements (Dektak measurements were measured from 10 different positions).
The results corresponded quite well with the Dektak values, and the deviation was the smaller the thinner the sample was. This was understandable, since the thicker samples are relatively non-uniform, and because the Dektak thickness was not measured from the same position as the HSI thickness profiles. The root mean square (RMS) error was 7.4 nm and 2.5 % relative to the mean thickness of the sample set.
The HSI system achieved an RMS error of 7.4 nm (2.5 % relative to the mean thickness of the sample set) for the OSC active material thickness measurement. The corresponding figures for the ellipsometer system were 16.6 nm or 5.7 %. The PEDOT samples could be measured with an RMS error of 34.6 nm (12.0 % relative to the mean thickness of the sample set) for the HSI. The corresponding figures for the ellipsometer system were 18.3 nm or 8.2%. It can be concluded that the HSI system is slightly better in measuring the OSC active material, whereas the ellipsometer system seems to be much better in measuring the PEDOT samples.
The repeatability of the THIME HSI measurement was studied by computing different standard deviation values based on stationary samples and on samples that were moved during the measurement. Several hundred measurements were taken, and the resulting thickness map was then analyzed. Clear outliers were removed from the thickness map. Similarly, the lateral resolution of both systems was studied obtaining a millimeter scale resolution for both systems over a large surface area web (100-250 mm), clearly indicating a very interesting result in comparison with the currently existing thickness measurement methods.
The main objective of WORK PACKAGE 7 was to carry out a comprehensive series of knowledge transfer and training activities in order to prove the industrial viability of THIME, to highlight its technical performance features, as well as to outline its potential and limitations.
The THIME project consortium planned three demonstration events of project results and achievements to public. The demonstration sessions were prepared as practical workshops that included an introduction to the project, a summary of the results obtained in the laboratory and the results obtained with the THIME prototype in terms of resolution, repeatability of measures, vibration resistance and advantages over other existing thickness monitoring techniques. Practical demonstrations of the operational prototype were also carried out.
Potential Impact:
There are numerous socio-economic impacts that will be derived from the results of this THIME research project. Central to the expected socio-economic impacts is the boosting of the competitiveness of OPV companies along the printed electronics value chain- by improving the production efficiency and quality. Given the fast development of this sector in the present and near future, there is a need of optimising the production processes to compete against other printed electronics companies in the market, especially from Asia and North America. Overall, the European printed electronics and OPV sector faces a delicate situation and needs to benefit from any technology or system capable of enhancing their efficiency and quality and therefore overall income.
In 2010, it was estimated that between 1,100 and 1,400 companies were operating globally in the printed electronics industry- many of the based in Asia, North America and Europe (with Germany and the UK being the leading printed electronics markets in Europe). If we consider that a Compound Annual Growth Rate (CAGR) of 58% was expected between the period 2010 and 2016, it is expected that by the time the THIME product is launched in the market, there could be at least some 4000 companies, based on growth predictions for the industry. Should THIME achieve an uptake of 3-4% among this market within 5 years of entry, some 140 units could be sold, installed and serviced in the market by 2020 valued at € 14 M. This value is based on a unit cost of €100,000 (€12 M), in addition to maintenance and service contracts (€2M). This is deemed achievable through sound marketing and in light of the unique THIME technology offering an on-line measurement method that offers multiple benefits in comparison to offline methods. Thus, especially the EU printed electronics industry but also the worldwide sector can benefit from the THIME technology in order to improve their productivity and competitiveness.
Taking into account the economic aspects and the benefits for the OPV and printed electronics industry by achieving for example a reduced active material consumption or an increased production control, the THIME technology is expected to have a significant impact on printed electronics industry, which is demanding of such improvements to reduce printing faulty production, production costs and enhance chip performance, and will therefore have a direct positive effect of the environment and on the industry competitivity.
Moreover, the impact of THIME will not only affect EU OPV or printed electronics producers, but it can have a global effect on the worldwide printing community. Production is geographically concentrated in Eastern Asia, United States, Germany and the UK. If the THIME system owners (COATEMA, ACCURION, SPECIM, NANOLAYER and SOLAR) manage to commercially exploit the system with a global patent, the sales could boost to all the producers across the printed electronics countries. This would imply a large benefit for the participating SMEs while ensuring an improved management of the production and resource efficiency worldwide.
The economic impact associated to THIME will expand to several sectors across the value chain apart from the printing end-users:
i) Producers of raw materials, components and equipment:
The electronics and optics producers will benefit from the demand of components for the TIHME system which will require the elements for the HSI and ellipsometer system, from the cameras and sensors to the lenses and electronic boards. Moreover, the steel and aluminium producers will provide the housing and structure for the instruments.
ii) Manufacturing and distribution companies:
The manufacturing companies, will have to produce and program the THIME hardware and software and deliver an off the shelf package for the final users or coating machinery companies to distribute the system to the end users. This will increase their activity and number of orders and in return increase their profit substantially.
iii) Coating machinery suppliers
The responsible companies for delivering the THIME system to the final users will be those dealing directly with OPV or printed electronics producers, mainly supplying the THIME system as a quality control instrument added onto their coating machines.
The increased productivity and quality generated by the more accurate control of thickness in the printing process will directly affect the job creation and qualified employment, offering a social benefit to the THIME system. In addition to the socio-economic benefits, the THIME system offers also environmental profits. Firstly, it will optimise raw materials usage by controlling the amount of material deposited during the printing process and resulting, in some cases, to raw material usage reduction. Secondly, it will reduce energy usage from producing non-conforming products that need to be produced again to fulfil the quality requirements. Thirdly, the waste generated by the faulty products will be reduced thus decreasing the impact over the lifecycle of the product. Finally, the concept of smart and sustainable OPV production will be enriched by the usage of THIME. This will stimulate the consumption of “greener” products in the market and will have a positive effect on the environment and in the printed electronics producers.
Finally, the THIME monitoring technology can be exported to other economic sectors such any industrial process demanding thickness measurements. This could be done by the THIME instrument and would definitely become a major source of profits for the owners of the system will offering an improved productivity and yield to those sectors using the THIME technology.
Main dissemination and exploitation activities
Each partner has expressed its intentions and wishes for the exploitation strategy. In addition, it allowed a great deal of dissemination activities to be carried out.
The current status of the protection and exploitation strategy for the THIME foreground within the Consortium can be summarized as follows:
- The HSI part of the system will be developed autonomously by the SME Specim and an external integrator company, as described in the Joint Exploitation Agreement.
- The ellipsometer part of the system will be developed autonomously by the SME Accurion for its commercial production.
- The SMEs Coatema and Nanolayer will integrate the instruments in their coating products and offer it to their clients.
- The SME Solar will benefit from preferential access to the technology such as early access or discount conditions.
All these roles and strategy are defined in the Joint Exploitation Agreement signed by all partners.
A really successful dissemination strategy has been carried out during the THIME project maximising the exploitability of the material and demonstrating its potential to provide OPV producers a tool for thickness monitoring and control in their production process and thus increasing productivity and quality of OPV and printed products.
Over the course of the project over 254 interested companies, stakeholders and policy makers were directly contacted by the THIME Consortium members in conferences, trade fairs, personal meetings and workshops. Moreover, a larger number were indirectly aware of the THIME project via 8 press releases in the general media, 2 presentations in conferences, and 1 scientific article published. The details can be found in the list annexed or in the Part A of the PUDF.
The project has attracted a large interest from the OPV, printed electronics and scientific community and has been in contact with other FP7 projects to find synergies and evaluate the possibility of uniting the different developed technologies to provide producers with a unique, complete tool.
Project Coordinator:
Mr. Pau Puigdollers de Balle
ppuigdollers@iris.cat
+34 935542500
IRIS Innovació i Recerca Industrial i Sostenible
Avda. Carl Friedrich Gauss n. 11
08860 Castelldefels
Spain
