Thin film absorbers for visible, near-infrared, and short-wavelength infrared spectra

Sensors and Actuators A 162 (2010) 210–214

Contents lists available at ScienceDirect

Sensors and Actuators A: Physical journal homepage: www.elsevier.com/locate/sna

Thin film absorbers for visible, near-infrared, and short-wavelength infrared spectra M. Laamanen ∗ , M. Blomberg, R.L. Puurunen, A. Miranto, H. Kattelus VTT Technical Research Centre of Finland, Tietotie 3, Espoo, P.O. Box 1000, FI-02044 VTT, Finland

a r t i c l e

i n f o

Article history: Received 30 September 2009 Received in revised form 22 January 2010 Accepted 18 February 2010 Available online 4 March 2010 Keywords: Absorber Visible light Infrared Mo–Si–N Refractive index MEMS

a b s t r a c t Two thin film absorbers are presented in this paper: one for the visible (VIS) and the near-infrared (NIR) spectra in the wavelength range of 350. . .1000 nm, the other for the short-wavelength infrared (SWIR) spectrum in the wavelength range of 1200. . .2000 nm. First, the refractive indices were determined for Al2 O3 films prepared with atomic layer deposition (ALD), and amorphous Mo–Si–N films prepared with reactive sputter deposition. The measurements were made by spectroscopic reflectometry, ellipsometry, gonioreflectometry, and double-beam transfer standard spectrometry. The results were utilised in the design of the absorbers. The absorbers were manufactured as well, and they proved to have high absorption over their whole working spectra varying from 93.4% at the minimum to 99.9% at the maximum. The absorbers are applicable, e.g. to MEMS thermopile detectors. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Optical spectrometers are widely used when the composition of a gas or a fluid must be measured. Such measurements are needed in industry, science, medicine, pharmacy, agriculture and biology with a wide range of applications [1]. Conventional spectrometers are bulky laboratory equipment with high performance and high cost. The demand for small-sized spectrometers for portable measuring devices and system integrated sensors has resulted in increasing interest in microspectrometers. A thermal infrared (IR) detector with sufficient absorption efficiency is a central part of many spectrometers. The operational principle of such detector is simple. First, incident radiation is absorbed. Second, the energy of the absorbed radiation is transformed into heat. Third, the consequent temperature change is measured. In order to achieve high sensitivity, high absorption efficiency combined with a small thermal mass is needed [2]. In general, infrared radiation absorbers can be divided into three groups: thin metallic films, porous metal blacks and thin film stacks [2]. Thin metallic films act as IR absorbers over a wide spectral range. Absorption of 50% can be achieved [3,4]. If a metal film is deposited on a solid substrate, however, its absorbing properties may be strongly dependent on the dielectric function of the substrate [4]. The metal black coatings are of, e.g. gold or platinum.

∗ Corresponding author. Tel.: +358 20 722 6687; fax: +358 20 722 7012. E-mail address: mari.laamanen@vtt.fi (M. Laamanen). 0924-4247/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.sna.2010.02.015

They are low-density deposits with a very porous structure. The metal blacks have been demonstrated to work as IR absorbers at wavelengths up to 50 ␮m, and they have a low thermal mass [5]. Due to their fragile nature, however, they may get easily crushed or displaced by any touch [6]. Hence, they cannot be patterned with lithography and etching. The metal black absorbers have also been reported to be sensitive to baking or changes in relative humidity, and to suffer from aging phenomena [5]. There are several studies on the third absorber group, the thin film stacks. The reported approaches include, e.g. the use of the passivation layers of a CMOS process for an integrated absorber [7,8], and thin metals with a dielectric spacer [3,9]. Also IR absorbers based on highly boron doped silicon slabs have been proposed [10]. The silicon slabs work simultaneously as heat spreaders but, on the other hand, their thermal mass is large resulting in low response time. We have focussed on the absorbers based on thin film stacks because of their high performance together with feasible integration into microelectromechanical systems (MEMS). Our previous work was a thin film blackbody structure designed for wavelengths of 4. . .8 ␮m, and applied to MEMS thermopile detectors used in IR gas sensors [11]. In this paper, two new absorber structures are presented: one for visible and NIR wavelengths (350. . .1000 nm), the other for wavelengths at the SWIR region (1200. . .2000 nm). The absorbers employed Al2 O3 films fabricated by ALD, and amorphous Mo–Si–N films fabricated by reactive sputtering. ALD is a chemical vapour deposition (CVD) technique that is suitable for the fabrication of inorganic thin films at rather low temperatures. It is based on the sequential use of self-terminating gas-solid reactions

M. Laamanen et al. / Sensors and Actuators A 162 (2010) 210–214

211

Table 1 The specification of the thin film stacks that composed the absorbers. The absorber VIS/NIR was designed for wavelengths of 350. . .1000 nm, and the absorber SWIR for wavelengths of 1200. . .2000 nm. Layer #

Absorber VIS/NIR Material

1 2 3 4

Mo Al2 O3 Mo–Si–N Al2 O3

Absorber SWIR Thickness (nm) 275 38 14 60

Deposition method

Material

Thickness (nm)

Deposition method

Sputter deposition ALD Sputter deposition ALD

Mo Si–N Mo–Si–N Si–N

100 40 42 169

Sputter deposition PECVD Sputter deposition PECVD

which result in material layers with thickness down to a fraction of a monolayer [12]. The advantages include precise thickness control as well as very high uniformity and conformity of the deposited films [13]. Mo–Si–N films, for one, belong to thin ternary films of the generic composition (early transition metal) –Si– (nitrogen or oxygen), also known as mictamict alloys. The early transition metals and nitrogen form metallic compounds with simple crystal structures, while silicon and nitrogen form Si3 N4 which is a covalent compound with a complex crystal structure. If nitrides of these two types are intermixed by a deposition process with a low atomic mobility (like sputter deposition), the result is a ternary film with an amorphous microstructure [14]. The mictamict alloys are interesting candidates for the structural material in MEMS because of their good mechanical properties but still a low deposition temperature [15]. However, the reported studies on optical or IR properties of mictamict alloys are very few in number. The scarce existing studies concern the use of mictamict films in phase-shift masks [16,17], the effects of annealing on W–Si–N films [18], and ellipsometry measurements of Ta–Si–N films [19]. Because of the limited existing information, the dispersion of the refractive index of amorphous Mo–Si–N films was measured in this study. The refractive index of Al2 O3 films was measured as well. The measuring data suggested that the optical properties of those films might well suit for IR and visible light absorbers. To demonstrate the suitability, we designed two absorber structures whose performance was first simulated and then, after fabrication, measured. 2. Experimental The Al2 O3 films were prepared in a SUNALETM R150 ALD reactor manufactured by Picosun. It was a showerhead-type reactor where the reactant gases enter the reaction space from above the substrate. The substrates for the determination of the refractive indices were phosphorus-doped polished silicon wafers with (1 0 0) crystal orientation. Before the deposition, they were cleaned with SC-1 (NH3 :H2 O2 :H2 O) and SC-2 (HCl:H2 O2 :H2 O) and rinsed in deionised (DI) water. The clean leaves the silicon surface terminated with a SiO2 layer and further with hydroxyl (OH) groups. After the clean, the wafers were dried with a spin-dryer. A trimethylaluminium/water (AlMe3 /H2 O) process at 300 ◦ C was used for the deposition. Nitrogen served both as the purging gas and the carrier gas for the AlMe3 and H2 O reactants. After loading a sample, the reactor was purged with nitrogen by three refilling–evacuation cycles. Before starting a process run, the chamber and the wafer were allowed to stabilise at the process temperature for about 20 min. A deposition process of 5000 cycles resulted in a film whose thickness was 473 nm. Each reaction cycle consisted of a trimethylaluminium pulse, nitrogen purge, a water pulse, and another nitrogen purge. When the dependence of the refractive index of Al2 O3 films on the deposition temperature was studied, the temperature was lowered stepwise from 300 ◦ C down to 80 ◦ C. The Mo–Si–N films were prepared by reactive sputter deposition in a Provac LLS 801 which is a carousel-type sputtering system. The wafers are placed on the sample holder carousel in a vertical position. As the carousel revolves, the wafers pass by the sputtering

targets placed on the walls of the chamber. The silicon substrates were cleaned with SC-1 and rinsed in DI water before the deposition. Some substrates were thermally oxidised to have a 100 nm thick film of silicon oxide on them. The wet oxidation took place at 850 ◦ C. The samples were not cleaned again between the thermal oxidation and the sputter deposition. The Mo–Si–N films were deposited by co-sputtering of molybdenum and silicon targets in a gas mixture of argon and nitrogen. The molybdenum target was biased with DC power, while the silicon target had to be biased with a pulsed power supply in order to avoid arcing across the insulating surface nitride. Inert argon was used to control the pressure, while reactive nitrogen incorporated into the depositing film. The flow rate was 50 sccm for argon, and 25 sccm for nitrogen. The process pressure was 5.3 ␮bar. During one rotation of the carousel, about 1.5 nm of Mo–Si–N was deposited. Two absorbers employing Al2 O3 and Mo–Si–N films were designed. One absorber (Absorber VIS/NIR) was designed for the wavelength range of 350. . .1000 nm, while the other absorber (Absorber SWIR) was designed for the wavelength range of 1200. . .2000 nm. Polished silicon wafers served as substrates. The wafers were cleaned with SC-1 before their use. The absorbers were based on thin film stacks of four different layers like defined by Table 1. Besides the Al2 O3 and Mo–Si–N films, they included conventional Mo and Si–N films. The molybdenum films were deposited by sputtering, and the Si–N films by PECVD. The appropriate value for the thickness of each film was calculated with TFCalc thin film design software (a product of Software Spectra Inc.). The dispersions of the refractive indices were determined based on reflectance measurements performed with spectroscopic reflectometry (FilmTek 4000), ellipsometry (1410 Plasmos SD 2300), gonioreflectometry [20], and double-beam transfer standard spectrometry (PerkinElmer). The reflectance measurements of the thin film absorbers were made with spectroscopic reflectometry and transfer standard spectrometry. The wavelength regions for the measurements were 190. . .1650 nm (spectroscopic reflectometry), 1000. . .1650 nm (gonioreflectometry), and 400. . .2400 nm (transfer standard spectrometry). The ellipsometer operated with a He/Ne laser at 633 nm and a measuring angle of 70◦ . The spectroscopic reflectometer (FilmTek) operated with two fixed angles (0◦ and 70◦ ). The gonioreflectometer employed a silicon detector on the visible spectral range, and an InGaAs detector on the NIR range being capable of measuring up to 1650 nm. A measuring interval of 50 nm was used. The transfer standard spectrometer, for one, measures regular spectral reflectance. The instrument in question employed halogen-tungsten and deuterium-arc light sources. The size of the incident beam was roughly 5 mm × 10 mm, and its angle relative to the sample normal 6◦ . The regular reflectance was determined twice for both orthogonal polarisation states (p and s), and with wavelength intervals of 5.0 nm. An aluminium mirror served as a reference. For process monitoring purposes, the thickness of each dielectric film being part of an absorber structure was measured with NanoSpec AFT 4150 (manufactured by Nanometrics Inc.) which is a spectroscopic reflectometer. Finally, the surface roughness was measured by atomic force microscopy (AFM).

212

M. Laamanen et al. / Sensors and Actuators A 162 (2010) 210–214

Fig. 1. The refractive indices of Al2 O3 films of different thicknesses measured at the wavelength of 633 nm by two means: spectroscopic reflectometry (FilmTek) and ellipsometry (Plasmos). The films were deposited by ALD with an AlMe3 /H2 O process at 300 ◦ C. Roughly 0.1 nm of Al2 O3 was deposited by one reaction cycle. Both measuring methods have their own film thickness limits outside which they are no longer reliable.

3. Results and discussion 3.1. Refractive index of Al2 O3 films For measuring the dispersion of the refractive index of a dielectric film like Al2 O3 , spectroscopic ellipsometry is often used (e.g. [21]). Because we did not have such a system, we measured the refractive index of Al2 O3 films by two means: spectroscopic reflectometry (FilmTek 4000) and, as a reference, single-wavelength ellipsometry (1410 Plasmos SD 2300). With reflectometry measurements, the Lorentz oscillator model was used as the dispersion model. It included an assumption about zero optical absorption (the extinction coefficient k = 0) which improved the resolution of how accurately the refractive index n could be determined. The ellipsometry measurements were performed for a reference only at one wavelength (633 nm) but on several samples which had Al2 O3 films of different thicknesses. The refractive index of Al2 O3 films at the wavelength of 633 nm is plotted by Fig. 1. The values measured by reflectometry and ellipsometry agreed when the film thickness was between 100 and 300 nm (corresponding roughly to 1000 and 3000 reaction cycles at deposition) but differed from each other elsewhere. A closer look revealed that the refractive index did not really change but a separate determination of the film thickness and the refractive index was not feasible with reflectometry for Al2 O3 films thinner than 100 nm. On the other hand, the refractive index determined with reflectometry did not depend on the film thickness as long as the film thickness exceeded 100 nm. The refractive index determined with ellipsometry, instead, did not depend on the film thickness down to a thickness of a few tens of nanometers. However, there was a considerable spread in the measuring results when the film thickness was 500 nm or more. In conclusion, ellipsometry appeared to be the better choice for measuring the refractive index of Al2 O3 films of 50. . .100 nm, and reflectometry for measuring the refractive index of films thicker than 100 nm. With appropriate measuring ranges (as regards the film thickness), the values obtained with reflectometry and ellipsometry agreed like they should. The dispersion of the refractive index of an Al2 O3 film was measured with reflectometry. The film was deposited with 5000 reaction cycles, which resulted in thickness of 473 nm. The measuring results at wavelengths between 190 and 1650 nm are shown by Fig. 2. The refractive index decreased when the wavelength increased. The values obtained agree with the dispersion reported by Boher et al. [21] who also deposited their Al2 O3 films by ALD.

Fig. 2. The dispersion of the refractive index n of an Al2 O3 film measured by spectroscopic reflectometry (FilmTek). The extinction coefficient k was assumed to be zero. The film was 473 nm thick corresponding to 5000 reaction cycles.

Finally, the refractive indices of Al2 O3 films deposited at different temperatures were measured. The samples were deposited with 1000 reaction cycles corresponding to film thicknesses between 80 and 100 nm, depending on the deposition temperature. The measurements were made with ellipsometry at the wavelength of 633 nm. Like Fig. 3 shows, the refractive index decreased with decreasing deposition temperature. Between 200 and 300 ◦ C, the changes were moderate, but below 200 ◦ C the slope got steeper. The cause of the decreased refractive index is most likely impurities (mainly hydrogen and carbon) that incorporate into the films deposited at lower temperatures. 3.2. Refractive index of Mo–Si–N films The dispersion of the complex refractive index of amorphous Mo–Si–N films was determined by reflectance measurements performed with gonioreflectometry (1000. . .1650 nm), double-beam transfer standard spectrometry (400. . .2400 nm) and spectroscopic reflectometry (190. . .1650 nm). With the use of these three sets of overlapping measuring data, the complex refractive index of a Mo–Si–N film on top of a silicon substrate was modelled (Fig. 4a). The thickness of the Mo–Si–N film was 33 nm. The measuring data and the modeling result fitted quite well. Both the refractive index n and the extinction coefficient k increased with increasing wavelength. For a reference, the reflectance of a 33 nm thick Mo–Si–N film on top of an oxidised silicon substrate was measured, too. The thickness of the thermal oxide film was 100 nm. The measurement was made with spectroscopic reflectometry at wavelengths of 230. . .1650 nm, and the complex refractive index was modeled

Fig. 3. The refractive indices of Al2 O3 films deposited at different temperatures. The samples were deposited with 1000 reaction cycles corresponding to film thicknesses between 80 and 100 nm, depending on the deposition temperature. The measurements were made by ellipsometry (Plasmos) at the wavelength of 633 nm.

M. Laamanen et al. / Sensors and Actuators A 162 (2010) 210–214

213

Fig. 4. (a) The dispersion of the complex refractive index (n, k) of a Mo–Si–N film deposited on top of a silicon substrate. The measurements were made with gonioreflectometry, double-beam transfer standard spectrometry and spectroscopic reflectometry. (b) The dispersion of the complex refractive index (n, k) of a Mo–Si–N film deposited on top of an oxidised silicon substrate. The measurements were made with spectroscopic reflectometry.

Fig. 5. The simulated and measured absorption spectra of (a) Absorber VIS/NIR designed to work at wavelengths of 350. . .1000 nm, and (b) Absorber SWIR designed to work at wavelengths of 1200. . .2000 nm.

based on the data (Fig. 4b). The result was not fully identical to the earlier one (Fig. 4a) although the two graphs resemble each other. The reason for the difference or whether it has any physical origin (instead of being a measuring artifact) is not known for sure. We tend to regard the result of the first sample more reliable, because the related reflectance data was obtained by three separate measuring methods with their wavelength ranges overlapping each other. Additional information could be sought, e.g. through determining the refractive index of Mo–Si–N films with different thicknesses. To the best of our knowledge, no one else has measured the dispersion of the refractive index of sputter-deposited Mo–Si–N thin films.

SWIR, a small deviation of the measured data from the simulated one occurred at the lower end of the spectrum but the absorption yet remained well above 90% even at its worst. For understanding the reason behind the small deviation, the roughness of a Mo–Si–N film of about 40 nm was measured by AFM. The AFM scan proved, however, that the film was very smooth with the root mean square (RMS) value of just few Ångströms. Thus the surface roughness of Mo–Si–N films could not be the reason for the deviation. Instead, simulations revealed that the absorption is rather sensitive to the thickness tolerances of thin film depositions. In conclusion, the deviations between the simulated and measured data in Fig. 5 are likely a consequence of thickness inaccuracy of one or more of the constituent films.

3.3. Thin film absorbers On the basis of the known refractive indices, two absorbers were designed and their structures optimised with the help of a simulation software (TFCalc): one for wavelengths of 350. . .1000 nm (Absorber VIS/NIR), the other for wavelengths of 1200. . .2000 nm (Absorber SWIR). Thereafter, the absorbers were fabricated on top of polished silicon substrates, and their reflectance spectra were measured (Fig. 5). The absorption on the y axes of the graphs was calculated assuming A() = 1 − R() where A() is the absorption and R() the reflection as functions of the wavelength . It was assumed that no transmission through the structures occurs because the molybdenum films at the bottom of the structures were fairly thick. The graphs show that both absorbers were broadband. The absorption was higher than 96% for Absorber VIS/NIR, and higher than 93% for Absorber SWIR over the whole spectra of interest. For Absorber VIS/NIR, the absorption fluctuated between few local minima and maxima at wavelengths below 350 nm, while it decreased smoothly and slowly at wavelengths above 1000 nm. For example, the absorption was still 89.9% at 1200 nm. For Absorber

4. Conclusions The dispersions of the refractive indices in the visible and near-infrared spectra were determined for Al2 O3 films deposited with ALD, and amorphous Mo–Si–N films deposited with reactive sputtering. Based on the information, two absorbers were designed and manufactured for wavelength ranges of 350. . .1000 and 1200. . .2000 nm. They were broadband with absorptions well above 90% over the whole spectra of interest. The absorbers may find practical use in MEMS thermopiles, bolometers and pyranometers. Acknowledgements Meeri Partanen and Antti Tolkki are acknowledged for cleanroom processing, Silja Holopainen for measurements with gonioreflectometry and transfer standard spectrometry, Kimmo Solehmainen and Sami Ylinen for measurements with spectroscopic reflectometry, and Markku Kapulainen for optical modelling.

214

M. Laamanen et al. / Sensors and Actuators A 162 (2010) 210–214

Silja Holopainen is at MIKES (Finnish Centre for Metrology and Accreditation) while the others are at VTT. The work was done within a EUREKA project (EUR 07-408). References [1] R.F. Wolffenbuttel, MEMS-based optical mini- and microspectrometers for the visible and infrared spectral range, Journal of Micromechanics and Microengineering 15 (2005) S145–S152. [2] A. Graf, M. Arndt, M. Sauer, G. Gerlach, Review of micromachined thermopiles for infrared detection, Measurement Science & Technology 18 (2007) R59–R75. [3] K.C. Liddiard, Application of interferometric enhancement to self-absorbing thin film thermal IR detectors, Infrared Physics 34 (4) (1993) 379–387. [4] S. Bauer, S. Bauer-Gogonea, W. Becker, R. Fettig, B. Ploss, W. Ruppel, W. von Münch, Thin metal films as absorbers for infrared sensors, Sensors and Actuators A 37–38 (1993) 497–501. [5] D.J. Advena, V.T. Bly, J.T. Cox, Deposition and characterisation of far-infrared absorbing gold black films, Applied Optics 32 (7) (1993) 1136–1144. [6] N. Nelms, J. Dowson, Goldblack coating for thermal infrared sensors, Sensors and Actuators A 120 (2005) 403–407. [7] R. Lenggenhager, H. Baltes, J. Peer, M. Forster, Thermoelectric infrared sensors by CMOS technology, IEEE Electron Device Letters 13 (9) (1992) 454–456. [8] N. Schneeberger, O. Paul, H. Baltes, Optimized structured absorbers for CMOS infrared detectors, in: Proceedings of The 8th International Conference on Solid-State Sensors and Actuators, and Eurosensors IX, Stockholm, Sweden, June 25–29, 1995, pp. 648–651. [9] A. Roncaglia, F. Mancarella, G.C. Cardinali, CMOS-compatible fabrication of thermopiles with high sensitivity in the 3–5 ␮m atmospheric window, Sensors and Actuators B 125 (2007) 214–223. [10] L. Fonseca, J. Santander, R. Rubio, N. Sabaté, E. Figueras, M. Duch, I. Gràcia, C. Cané, Use of boron heavily doped silicon slabs for gas sensors based on freestanding membranes, Sensors and Actuators B 130 (2008) 538–545. [11] M. Blomberg, A. Lehto, A thin film blackbody structure, in: Proceedings of Eurosensors XIV. 14th European Conference on Solid-State Transducers, Copenhagen, Denmark, August 27–30, 2000, pp. 301–302. [12] R.L. Puurunen, Surface chemistry of atomic layer deposition: a case study for the trimethylaluminum/water process, Journal of Applied Physics 97 (12) (2005) 1–52. [13] J. Dekker, K. Kolari, R.L. Puurunen, Inductively coupled plasma etching of amorphous Al2 O3 and TiO2 mask layers by atomic layer deposition, Journal of Vacuum Science & Technology B 24 (5) (2006) 2350–2355. [14] M.-A. Nicolet, P.H. Giauque, Highly metastable amorphous or near-amorphous ternary films (mictamict alloys), Microelectronic Engineering 55 (2001) 357–367. [15] H. Kattelus, M. Ylönen, M. Blomberg, Amorphous Mo–N and Mo–Si–N films in microelectromechanical systems, Fatigue & Fracture of Engineering Materials & Structures 28 (2005) 743–749. [16] C.-M. Lin, W.-A. Loong, TiSix Ny and TiSix Oy Nz as embedded materials for attenuated phase-shifting mask in 193 nm, Microelectronic Engineering 46 (1999) 93–96. [17] W.-A. Loong, C.-M. Lin, S.-P. Tseng, W.-L. Yeh, The correlation between the chemical compositions and optical properties of TiSix Ny as an embedded layer for AttPSM in 193 nm, Microelectronic Engineering 57–58 (2001) 481–487.

[18] O.H. Gokce, S. Amin, N.M. Ravindra, D.J. Szostak, R.J. Paff, J.G. Fleming, C.J. Galewski, J. Shallenberger, R. Eby, Effects of annealing on X-ray amorphous CVD W–Si–N barrier layer materials, Thin Solid Films 353 (1999) 149–156. [19] M. Meier, D. Bertsch, N. Onda, M. Etter, M. Gutsche, A. Dommann, V. Romano, M.-A. Nicolet, Laser trimming of amorphous Ta42 Si13 N45 thin films with ultrashort pulses, Microelectronic Engineering 83 (2006) 2234–2237. [20] S. Nevas, F. Manoocheri, E. Ikonen, Gonioreflectometer for measuring spectral diffuse reflectance, Applied Optics 43 (2004) 6391–6399. [21] P. Boher, C. Defranoux, P. Heinrich, J. Wolstenholme, H. Bender, VUV spectroscopic ellipsometry applied to the characterization of high-k dielectrics, Materials Science and Engineering B 109 (2004) 64–68.

Biographies Mari Laamanen received her MSc degree in Electrical Engineering from Helsinki University of Technology, Finland, in 2001. She joined VTT Technical Research Centre of Finland in 2001 and has worked with MEMS fabrication and devices since then. During her years at VTT, she has developed LPCVD processes, metal micromachining, pressure sensors, audio microphones, RF devices, Fabry-Pérot interferometers, and thermopile detectors. Martti Blomberg obtained his Licentiate’s degree in Electrical Engineering in 1991 from Helsinki University of Technology. He currently works as a Senior Research Scientist at VTT. He has worked over 25 years on silicon technology. He has over 15 years’ experience in MEMS device design and fabrication technology development. His recent activities include device and process development regarding various MEMS and MOEMS devices such as electrically tunable surface micromechanical Fabry-Pérot filters and thermopile detectors. He has written several publications from the field and is the inventor of a number of patents. Riikka Puurunen completed her doctoral degree in the Department of Chemical Technology at Helsinki University of Technology in 2002. Her main technological and scientific interest is in the atomic layer deposition (ALD) technique. Her doctoral research (1999–2002) dealt with the preparation of catalysts by ALD and her postdoctoral research (2003–2004, IMEC, Belgium) with modelling ALD especially for high-k gate dielectric applications. In October 2004, she joined the MEMS technology group at VTT, implementing ALD in MEMS fabrication. She is the author or co-author of about 30 scientific articles. At present, she is a Senior Research Scientist and a Project Manager in Microsystems and Nanoelectronics Centre, VTT. Akseli Miranto joined the research staff at Optical Sensors Team at VTT in 2007, and received the MSc (Tech) in Electrical Engineering from the Helsinki University of Technology in 2009. His research interests are in optical instruments, micro-optoelectromechanical systems and imaging sensors. Hannu Kattelus received his doctoral degree from the Department of Electrical Engineering at Helsinki University of Technology in 1988. He has been working at VTT since 1980 developing thin film processes and devices for various applications. His activities have covered display technologies, CMOS, integrated passive devices, and MEMS. He has authored or co-authored about 100 scientific or technical papers. Presently he is a Technology Manager in Microsystems and Nanoelectronics at VTT and a Research Professor in Microsystems Technology.