Field of invention The present invention relates to capacitive-based sensing devices, in particular printed capacitive gas sensing devices.

State of the art

Capacitive-based sensing is especially appealing as sensing principle due to its ultra-low power consumption compared with other methods such as resistive.

Due to their advantageous properties such as very low-cost, flexible and conformal, compatible with large arrays, optical transparency, etc., printed gas sensors on plastic foils offer new routes in applications related to consumer electronics (smart phones), wearable systems and monitoring of the environment and goods (wireless sensors and smart RFID labels).

Using printed/plastic electronic processes offers benefits compared to standard cleanroom thin films processes because of the unique properties of plastic electronics (potentially low cost, light weight, thinness, mechanical flexibility and stretchability), and the advantages of printing methods (efficiency in material use and low thermal and chemical budget requirements). However, printing processes present limited resolution compared to standard optical photolithography, resulting in large capacitive devices, especially those utilizing typical interdigitated electrodes (IDE). Other drawbacks of printed/plastic devices are that the standard plastic - and organic in general - substrates exhibit sensitivity to some volatile organic compounds and humidity; and the fact that printable functional materials are solvent-based liquids, so that chemical incompatibilities arise between different stacked layers, reducing the range of utilizable materials. Printed electronics methods offer limitations in resolution, which makes difficult the fabrication of small capacitive devices with a capacitance value large enough for detection with standard electronics. This is an important issue when dealing with low cost capacitive gas sensing applications, because the smaller the surface area of a device, the smaller the amount of material needed to fabricate it (ink and substrate material) and the related fabrication cost. Besides, reduced dimensions involve improved performances such as response time or sensitivity.

The chemical interaction between the printed inks (conductors, sensing layers, dielectrics) and the substrate is challenging due to the nature of the printing techniques and solvent-based materials. The latter could directly influence the integrity of the layers as well as the adhesion among them.

Different structures can be found in literature, which seek for the optimization of sensitivity and response time in capacitive gas sensors based on sorption phenomena. Among these structures, we can find:

· Parallel -plates (PP) devices: PP devices were reported where different-shaped gas admittance openings are fabricated on the sensor to optimize sensor performances. The openings are included on top electrode [1, 2], or both bottom and top electrodes [2] and dielectric to favor gas flow inside the structure. The alignment between top and bottom patterned electrodes is also considered (see Figure 1, 2 and 3). The fabrication process and materials are basically vacuum deposited metals for electrodes, spin coated polyimide and Reactive Ion Etching (RIE) or 0 ₂ plasma etching for patterning [2].

• Interdigitated electrodes (IDE) devices: IDE devices have been studied before. As an example, we can find in literature some papers dealing with Si-based sensors fabricated with standard microfabrication techniques, allowing for feature sizes of few microns. Small size is not a challenge for this established technology and the substrate is inert. Therefore, isolation between both branches of electrodes is not needed to prevent overlap of the electrodes or as a barrier, but to avoid bad sensor performance when condensation happens in the sensor at high values of analyte concentration. In order to increase sensitivity, it is possible to pattern one branch of electrodes higher than the other, to shift up the lines of electric field to the area that will be occupied by the sensing layer (see Figures 4 and 5) [3, 4]. • Combination of both PP and IDE devices: By using spin coating and evaporation, some groups have beneficiated from the advantages of PP and IDE devices by combining them somehow. A very simple example can be found in [5], where the two electrodes of the device are in plane and a floating electrode is placed on the other extent of the dielectric layer (see Figure 6). The whole structure has the resulting capacitance value of two parallel-plates capacitors in series.

Finally, different electrode geometries have been compared using a theoretical model. The model was previously assessed with the data obtained from a PP gas sensor. In this work, the authors studied the influence of stacking two IDE layers for gas sensing [6]. Figures 7 and 8 show the structure used for evaluation as well as one of the studied configurations.

Similar strategy is proposed in the patent [7], where the structure shown in Figure 9 allows for a higher implication of the lateral wall component of the capacitance in the sensing mechanism.

The rigid substrates presented above does not beneficiate from the polymeric electronics methods, which result especially appealing for low-cost, smart packaging due to the ability of the sensors of being arbitrary shaped (see [8]), as well as the possibility of printing them directly on the package.

Above was reviewed some works dealing with rigid substrates and standard micro-fabrication processes. Some work was also published on the development of flexible / polymeric gas sensors using standard clean room and also printing technology. A part of them relies in changes on resistivity in presence of gas [9-11]. Considering the type of device addressed here, only low- power capacitive-based devices are reviewed in this section.

• Flexible and polymeric IDE devices: In 2008, temperature and humidity sensors were developed using gold patterned on parylene and polyimide as sensing layer [12]. One year later, interdigitated capacitive humidity sensors based on MWCNTs/polyelectrolyte interfaces on flexible substrates were also reported [13]. This same year Oprea et al., in collaboration with EPFL-SAMLAB fabricated low-power temperature and volatile organic compounds (VOC) sensors on flexible substrates [14]. They measured the capacitance using a differential method to subtract the parasitic effect of the substrate. The sensing materials were common polymers, like PEUT or PDMS. Recently, another humidity sensor based on cellulose-polypyrrole nanocomposite has been reported [15].

• PP structure: Also PP and flexible capacitive structures have been reported for gas sensing using standard processes. In 2006, Satyanarayana et al. showed an array of vertical sensors, which sensing principle was based on surface stress [16]. In the last years, Zampetti and his group have been developing stacked PP structures based on flexible substrates and different gas sensing dielectrics; using standard clean room processes (see Figure 10) [17, 18].

• Printed devices: Some work has been reported in the fabrication of printed capacitors for other than gas sensing applications describing interesting fabrication processes. In 2003, a PEDOT: PSS (conductive polymer) inkjet-printed parallel-plate capacitor for RC filters was published [19]. In 2009, an interdigitated capacitive and resistive strain sensor using inkjet-printed PEDOT:PSS was also reported [20]. Our group has recently gathered some experience in inkjet printer capacitive gas sensors [21-23] using a standard IDE configuration in which the size has not been optimized yet, leading to large devices of tens of millimeters squares of surface area. Recently some IDE devices using gravure and inkjet printing techniques on PET and PI foil have been presented [24-26].

Even though some of the capacitive gas sensing approaches presented above have been developed on flexible organic foil, they were mainly developed with standard clean room technology with processes such as evaporation, lithography, etching, spin coating... The incompatibilities with polymeric and cellulose based substrates introduced by the use of standard processes are linked to the low maximal temperature and bad tolerance to aggressive chemicals typical of organic flexible materials. Large improvements in fabrication costs and large scale production can be achieved by the use of printing electronics methods on a large area.

Even for the mentioned examples of printed devices, the size of the proposed sensors remains still too large to obtain their peak of performances, especially for inkjet-printed devices. Despite its low resolution, inkjet printing is recently getting special attention due to its maskless and contactless character, great flexibility of prototyping, potential availability of materials to be used as printed solutions, and suitability for the fabrication of arrays of sensors with different gas sensitive materials. Apart from inkjet printing, other printing methods such as micro-contact, gravure, flexography, screen or offset printing are also compatible with this invention and could benefit for the proposed approaches for size reduction of printed capacitive gas sensing devices. Finally, the interaction between the printed inks and the polymeric substrates presents disadvantages, due to the solvent-based nature of the inks and the curing and sintering processes involved.

Because the operation of this kind of capacitive gas sensors is based on the diffusion of the analyte in the sensing layer, a big improvement in the sensor performance can be obtained by including a heater in the structure [27, 28]. At higher temperature, the diffusion of the analyte in the sensor occurs more rapidly, decreasing the response time of the sensor. Higher temperature also improves the linearity of the sensor signal by reducing the chances of condensation of certain analytes, such as humidity, in the sensing layer and it makes the sorption process more reversible by reducing the hysteresis of the sensor signal. Pulse temperature modulation can also be applied to regenerate/reset the sensors.

General description of the invention

The problems mentioned in the previous chapter are solved by the invention which relates to a capacitive-sensing device and a process for manufacturing such a device. Both objects are defined in the claims.

Among other things the invention includes two geometries and their corresponding manufacturing processes, which are able to solve the main issues associated to printed capacitive gas sensors on plastic foils; namely decrease their surface area for the same capacitance value, and eliminate the parasitic influence of the substrate. Both geometries can be patterned so that a heater or a thermoresistor may be formed in the device for improved sensing capabilities and/or temperature control.

Smaller gas sensing devices are more performing and cost efficient. In order to increase the capacitance per surface area, two main approaches may be followed:

• The first one is to decrease the gap in between electrodes for planar or quasi planar IDE devices, by adding a dielectric interlayer between electrodes.

· The second one involves the use of a parallel-plates (PP) structure.

Both solutions solve the problem of substrate parasitic signal in a different way: the interlayer in the first one acts as barrier for the analyte that isolates the substrate, whereas the geometry of the second confines the electric field exclusively in the sensing layer. Finally, the first solution involves the deposition of the sensing layer onto the IDE, avoiding in this way any chemical compatibility problem. For the second solution, the use of micro-contact printing process minimizes the appearance of chemical incompatibilities between the different stacked layers. Each kind of configuration presents its own advantages depending on the specific application. In terms of materials, every flexible organic material can be used as substrate regardless its thermal properties, since printed process do not require high temperature. Among the most common materials used in printed electronics, one can find: polyethylene terephthalate (PET), polyethylene naphthalene (PEN), polyimide (PI), parylene, polyethyleneimide (PEI), cellulose based materials such as paper, dry photoresist and biodegradable polymers such as poly(lactic acid) (PLA) and derivate. This invention is compatible with every printing electronics method such as micro-contact, inkjet, gravure, flexography, offset or screen printing. A wide variety of materials can be deposit by means of printed methods. Among them, we can find conductive materials: metals and conductive polymers such as Poly(3,4-ethylenedioxythiophene) poly(styrenesulfonate) (PEDOT:PSS), polypyrrole (PPY) or polyaniline (PANI)); or dielectrics, which can be printed or evaporated to be used as passivation layer (parylene-C, AI ₂ O ₃ or others) or gas sensing material: Cellulose acetate butyrate (CAB), poly(methylmethacrylate) (PMMA), cellulose acetate (CA), polyetherurethane (PEUT), polydimethylsiloxane (PDMS) or polyvinylpyrrolidone (PVP), dry foil or Poly(2-hydroxyethyl methacrylate (pHMEA) and their combinations to name some of them.

Detailed description of the invention

The invention will be presented in a more detailed manner below, with non-limiting examples and with some figures.

Brief description of the figures:

Figure 1 : from references [1, 2] parallel -plate configuration. The top electrodes with humidity admittance openings, a) rectangular stripes, b) square holes, and the bottom electrodes, c) conventional and d) & e) patterned to reduce the less sensitive capacitive area.

Figure 2: from reference [2] parallel-plate configuration. Cross section of various sensor designs: a) conventional, b) with patterned and aligned bottom electrodes and c) with patterned and displaced bottom electrode. Figure 3 : from reference [2] parallel -plate configuration. The cross section of a humidity sensor with patterned humidity admittance opening formed by RIE to increase sensitivity.

Figure 4: from reference [3]. Electrode width and spacing are 1.6 mm. The total size of both sensor and reference is 800 x 800 μπι . Note that the stacked electrodes have same polarity.

Figure 5: from reference [4]. The upper electrode consists of 32 aluminum the two electrodes fingers and the lower of polysilicon. The fingers are each 3 μπι wide and 3 μπι apart. CVD oxides and nitrides guarantee an electric insulation between. Surface: 380 x 380 μπι .

Figure 6: from reference [5]. Structure of the floating electrode capacitor.

Figure 7: from reference [6]. Configuration of fabricate sensor for evaluation of the model. Figure 8: from reference [6]. Sketch of simulated models (no fabrication). Sizes of W and G of 3 μπι.

Figure 9: from reference [7]. Sketch of a capacitive structure combining properties from PP and IDE devices.

Figure 10: from reference [17, 18]. Sketch of a flexible parallel plates gas sensor.

Figure 11: example of printing (inkjet) of an IDE device with a passivation layer in between the two electrodes to prevent short circuit. The sensing layer is deposited as last step on top of the printed electrodes. The top and bottom electrodes could be designed as resistor types to use as RTD or heater.

Figure 12: cross-sectional schematic of non-overlapping capacitive sensor with a passivation layer that separates both electrodes to prevent short circuits and minimize the gap between them, and to prevent influence from the substrate. The sensing layer comes on top of the structure while the substrate on the bottom. Figure 13: cross-sectional schematic of non-overlapping capacitive sensor with a passivation layer between the substrate and the electrodes to prevent influence from the substrate, and a sensing layer in-between the electrodes.

Figure 14: cross-sectional schematic of non-overlapping capacitive sensor with a passivation layer between the substrate and the electrodes to prevent influence from the substrate. The structure presents two sensing layers: one in-between the electrodes and one on top of the structure.

Figure 15: prevention of short circuit between two overlapping printed electrodes (of different polarity), thanks to their separation with a passivation layer (dielectric layer) in between (Figure 11). Figure 16: comparison between theoretically estimated and measured capacitance per surface area versus inter-electrodes space (a), showing optical pictures of different scenarios of overlapping between the two electrodes (b).

Figure 17: comparison of transient response of the sensor (right axis) versus pulses of relative humidity (left axis), showing a good agreement between the two curves.

Figure 18: cross-sectional schematic of parallel -plate capacitive sensor with a sensing layer in- between the electrodes, on top of an organic substrate. Figure 19: cross-sectional schematic of parallel -plate capacitive sensor with a sensing layer in- between the electrodes, on top of an organic substrate. The top electrode is patterned to use it as resistive sensor or resistive heater.

Figure 20: cross-sectional schematic of parallel-plate capacitive sensor with a sensing layer in- between the electrodes, on top of an organic substrate. The top and bottom electrodes are patterned to use them as resistive sensors or resistive heaters. Figure 21: optical pictures and magnification of a fully inkjet-printed parallel-plate capacitive humidity sensor. The rough surface and porous nature of the top electrode allows the analyte to reach the sensing layer without the need of patterning a grid.

Figure 22: comparison between the parallel plate device measurements with and without the grid on the top electrode versus relative humidity values. The plot shows the feasibility of having a top electrode without patterning and still having the analyte reaching the sensing layer. Figure 23: comparison between parallel plate device measurements without the grid on the top electrode (left axis) versus a commercial sensor (right axis), showing a good agreement between the two curves.

Figure 24: fabrication process of a printed vertical parallel plate capacitive gas sensor with a complete and grid top electrode. The sensing layer is deposited after the first metal layer onto the organic substrate. Then using micro-contact printing two approaches could be followed: complete or grid mold. The grid mode increases the exposure area of the sensing layer to the analyte. Figure 25: example of micro contact printing process, showing the process flow to fabricate a flexible complete or grid mold.

Figure 26: example of a micro contact printed top Au electrode onto an inkjet printed bottom Au electrode. Both electrodes are separated by a cellulose acetate butyrate layer.

Figure 27: schematic of single drop size capacitive sensor with a printed bottom electrode and sensing layer, and micro contact printed top electrode using a flexible stamp. IDE coplanar capacitors for gas sensing

In IDE coplanar capacitive gas sensors, the sensing layer is placed on the top of the device electrodes allowing for direct interaction between the target gas and the sensing layer as depicted in Figure 11. On the other hand, the patterning of the electrodes can result more complicated than in PP structures, requiring high precision to define closely spaced electrodes, which is usually difficult to achieve with standard printing methods (such as inkjet, screen or gravure printing). Closely packed electrodes, without short-circuits between them, are mandatory to obtain high values of capacitance per surface area. Another parameter playing an important role in the sensor performance is the thickness of the sensing layer(s). As commented above, the operation of this type of sensors is diffusion-based, thus a very thick sensing layer implies a longer sensor response [23, 29]. In general, thick sensing layers provide also higher sensitivity, but this factor becomes less important for the highly packed electrodes that we propose in this report [30]. Finally, because the substrate is also polymeric, it would act as a secondary sensing layer, so special care has to be put in its characteristics. In case that the substrate introduces parasitic effect, differential measurements need to be performed to eliminate this undesirable component [14].

In order to increase area density and the fabrication yield at the same time, one IDE-based capacitor is proposed utilizing a polymeric substrate where we print a first set of electrodes / heater / thermoresistor and passivate them with a thin layer of insulator material such as parylene or other [31]. Then we inkjet print the second half of the electrodes on top and finally we add a dedicated sensing layer as depicted in Figure 11 and Figure 12. Alternatively, the insulator material can be deposited directly on the substrate. Then, after the first set of electrodes / heater / thermoresistor is deposited, a sensing layer is added as electrodes-separating layer prior to the printing of the second set of electrodes (see Figure 13). In the last step, the sensing layer can be completed if needed with a second sensing layer as depicted in Figure 14. Even if one electrode overlaps with the contiguous one, they will not be exactly on the same plane and there will not be in contact, avoiding short circuits (see Figure 15). Therefore, and despite IDE is the only electrode geometry shown in the figures, any other shape of coplanar electrodes (such as meanders or grids) could be used in the present invention without loss of generality. The described devices present several advantages for capacitive sensing compared with less compact structures. For instance, the sensitivity of such printed sensors improves for two reasons: on one hand, the nominal capacitance increases for the same area, and accordingly the absolute sensitivity. On the other hand, being the electrodes more packed, there would be a maximization of the interaction between the electric field and the thin sensing layer, increasing also the relative sensitivity [23] since it is known that the electric field of such combed structures extends basically to a height equal to the electrodes pitch [29, 30, 32, 33]. In principle an intermediate layer, considered insensitive to the target gas as is the case of the passivation, would decrease the sensitivity because it occupies space filled by the electric field lines. Nevertheless its influence results negligible in practice, due to its thickness, very small compared to the electrodes pitch. The previous fact has been supported theoretically by removing the parylene layer in a theoretical model and observing a very small capacitance difference. In the same way, overlapping electrodes do not contribute to the shift in capacitance in the presence of the target gas, since the electric field associated to them would be confined in the insensitive insulating layer. Another sensing advantage of the presented devices comes from the fact that both capacitor polarities are insulated. Therefore, condensation of analyte (such as water) between electrodes, responsible of an undesired modification of the sensor signal at high concentration, does not appear [34-36]. Last but not least, the isolation layer acts as a barrier for the analyte, so that the substrate does not absorb analyte, making possible the operation of one single device, instead of differential mode measurements [14, 23].

The value of capacitance per surface area obtained for well aligned (see Figure 16) devices was as high as 0.45 pF/mm , for an electrode pitch of 60 μπι and an inter-electrodes gap as small as 12 μπι. A theoretical model based on conformal mapping and partial capacitance is proposed to estimate the nominal capacitance value of the different fabricated geometries, providing a very good fitting with experiments. The potential of the aforementioned devices for sensing has been evaluated for relative humidity capacitive sensing, by inkjet printing a test layer of cellulose acetate butyrate (CAB), whose electrical permittivity and thickness is sensitive to relative humidity, but the concept can be extended to any analyte/sensing layer couple. The characterization of the sensors in both steady and transient state offered devices with linear behavior up to 70% r.h., an improved absolute sensitivity per area of 5.46 fF/1% r.h. and stable and reproducible signal (see Figure 17). Parallel-plates capacitors for gas sensing

Parallel-plates capacitors for gas sensing present the main advantage of having the electric field confined between the two parallel plates electrodes, avoiding parasitic effect of the polymeric substrate. This configuration also maximizes the nominal value of capacitance and the relative sensitivity, maximizing the overlapping between sensing layer and electric field.

In principle, one or both parallel electrodes can be patterned as shown in Figures 18, 19 and 20. We have fabricated a parallel-plate device consisting of polymeric substrate, inkjet-printed bottom electrode, inkjet-printed humidity sensing layer of cellulose acetate butyrate (CAB) and inkjet-printed top electrode (see Figure 21) [37]. We demonstrated that the material of the top electrode is non-homogeneous and porous enough to let the analyte (water in this case) reach the sensing layer as depicted on the right side of Figure 21 and in the capacitive response in Figure 22. This fact makes the challenging patterning of the top electrode step unnecessary. Good response to humidity is appreciated in Figure 23 for an un-patterned top electrode.

In principle, stacked electrodes capacitors allow high capacitance value per surface area and very fast response by utilizing a very thin dielectric in between the two electrodes. Inkjet printing allows for a tight control of the thickness of very thin printed layers, which makes it a good process for the fabrication of this configuration. However, it is very challenging to print the top electrode onto a very thin sensing layer using conventional printing methods, without short- circuiting the device. The reason is that conventional printing techniques use solvent-based inks that are prone to react with the dielectric material, damaging it or dissolving it in the worst case.

To outcome the previous challenges, the device proposed is an improvement of the developed in [37] and consists of a dielectric that acts as the substrate, a printed (with any method and material) conductor that works as bottom electrode and another inkjet-printed (or printed in general) dielectric that is the sensing layer between electrodes (see Figure 24). The challenging top electrode is to be developed using micro-contact printing, μ-contact printing is an emerging fabrication method in which a mold with sizes down to tens of nanometers is obtained by e-beam or photo lithography. The mold is dipped on self-assembled metallic nanoparticles or other conductive paste and stamped on the substrate transferring the desired layout as shown in Figure 25. The advantage of using this printing method is that it makes possible the transfer of a dry conductive layer which would not interact with the sensing layer underneath, expanding the range of material to be used as sensing layer, as well as allowing the use of thinner dielectric layers with no risk of shortcircuits. We have some previous experience using PDMS stamps to transfer layers of gold-nanoparticles from the surface of deionized water onto a sensitive polymer CAB without observing interaction between both (see Figure 26). The gold layer is sintered using low-frequency Ar plasma, which is a room temperature process compatible with the materials in the structure. Very thin sensing layers would permit the shrinking of the device surface area down to single drops devices maintaining yet a measurable nominal value (see Figure 27). With μ-contact printing, it is possible to achieve the same resolution than with photolithography, so that patterning the top electrode is not a problem. Nonetheless, it is likely that the transferred material is also porous enough as in [37] or have certain self-assembled pattern that make this patterning step unnecessary. In any case, it is possible to transfer a complete and a grid-shaped mold on top of the sensing layer with the described micro-contact printing technique, as shown in Figure 25.

Finally, the bottom electrode can be shaped as a meander-shaped heater / RTD (as in Figure 20) if needed, at expenses of having larger area device, but benefiting from the advantages of an integrated heater.

References

1. Hong S. M., Kim K. N., Jo Y. C, Kim W. FL, Humidity sensor of capacitance type and method of fabricating same, Patent WO 2010056049 A2, 2009.

2. Ya Ling Y., Lieh Hsi L., Huang I. Y., Chen H. J. H., Wen Shen H., Huang S. R. S., Improvement of polyimide capacitive humidity sensor by reactive ion etching and novel electrode design, Proceedings of IEEE Sensors, June 12-14, 2002, Orlando, Florida, USA, pp. 511-514.

3. Hierlemann A., Lange D., Hagleitner C, Kerness N., Koll A., Brand O., et al. Application-specific sensor systems based on CMOS chemical microsensors. Sensors and

Actuators B: Chemical, 2000, 70 (1-3) 2-11.

4. Cornila C, Hierlemann A., Lenggenhager R., Malcovati P., Baltes H., Noetzel G., et al. Capacitive sensors in CMOS technology with polymer coating. Sensors and Actuators B: Chemical, 1995, 25 (1-3) 357-361.

5. Roman C, Bodea O., Prodan N., Levi A., Cordos E., Manoviciu I. A capacitive-type humidity sensor using crosslinked poly(methyl methacrylate-co-(2 hydroxypropyl)- methacrylate). Sensors and Actuators B: Chemical, 1995, 25 (1-3) 710-713.

6. Sen A. K., Darabi J. Modeling and Optimization of a Microscale Capacitive Humidity Sensor for HVAC Applications. Sensors Journal, IEEE, 2008, 8 (4) 333-340.

7. Aurelle Humbert M. M., Sensor has combined in-plane and parrallel-plane

configuration, Patent WO 2010/029507 Al, 18 2010.

8. Adamec R., William H., Beadle P.Johnson P., Suzuki T., Tanner P., Thiel D., Environmental sensor, Patent WO 2004109238 Al, 2004.

9. Pecora A., Maiolo L., Zampetti E., Pantalei S., Valletta A., Minotti A., et al., Chemoresistive nanofibrous sensor array and read-out electronics on flexible substrate, Solid- State Sensors, Actuators and Microsystems Conference (TRANSDUCERS 2009), June 21-25, 2009, Denver, Colorado, USA,pp. 144-147.

10. Parikh K., Cattanach K., Rao R., Suh D.-S., Wu A., Manohar S. K. Flexible vapour sensors using single walled carbon nanotubes. Sensors and Actuators B: Chemical, 2006, 113 (1) 55-63. 11. Su P.-G., Wang C.-S. Novel flexible resistive-type humidity sensor. Sensors and Actuators B: Chemical, 2007, 123 (2) 1071-1076.

12. Lee C.-Y., Wu G.-W., Hsieh W.-J. Fabrication of micro sensors on a flexible substrate. Sens Actuators A Phys, 2008, 147 (1) 173-176.

13. Arena A., Donato N., Saitta G. Capacitive humidity sensors based on MWCNTs/polyelectrolyte interfaces deposited on flexible substrates. Microelectronics Journal, 2009, 40 (6) 887-890.

14. Oprea A., Courbat J., Barsan N., Briand D., de Rooij N. R, Weimar U. Temperature, humidity and gas sensors integrated on plastic foil for low power applications. Sens Actuators B Chem, 2009, 140 (1) 227-232.

15. Mahadeva S. K., Yun S., Kim J. Flexible humidity and temperature sensor based on cellulose-polypyrrole nanocomposite. Sensors and Actuators A: Physical, 2011, 165 (2) 194- 199.

16. Satyanarayana S., McCormick D. T., Majumdar A. Parylene micro membrane capacitive sensor array for chemical and biological sensing. Sensors and Actuators B: Chemical, 2006, 115

(1) 494-502.

17. Zampetti E., Pantalei S., Pecora A., Valletta A., Maiolo L., Minotti A., et al. Design and optimization of an ultra thin flexible capacitive humidity sensor. Sens Actuators B Chem, 2009, 143 (1) 302-307.

18. Zampetti E., Maiolo L., Pecora A., Maita F., Pantalei S., Minotti A., et al. Flexible sensorial system based on capacitive chemical sensors integrated with readout circuits fully fabricated on ultra thin substrate. Sensors and Actuators B: Chemical, 2011, 155 (2) 768-774. 19. Chen B., Cui T., Liu Y., Varahramyan K. All-polymer RC filter circuits fabricated with Inkjet printing technology. Solid-State Electron, 2003, 47 (5) 841-847.

20. Al-Chami PL, Cretu E., Inkjet printing of microsensors. Mixed-Signals, Sens, and Syst Test Workshop, 2009 IMS3TW '09 IEEE 15th Int, 10-12 June, 2009, Scottdale, AZ, USA, pp. 1- 6.

21. Courbat J., Kim Y. B., Briand D., de Rooij N. F. Inkjet printing on paper for the realization of humidity and temperature sensors, Solid-State Sensors, Actuators and Microsystems Conference (TRANSDUCERS 2011), June 5-9, 2011, Beijing, China, pp. 1356- 1359. 22. Molina-Lopez F., Courbat J., Briand D., de Rooij N. F. Inkjet Printing of Silver on Flexible Substrates for Sensing Applications, Proceeding LOPE-C 2011, June 28-30, 2011, Frankfurt, Germany, pp. 278-282.

23. Molina-Lopez F., Briand D., de Rooij N. F. All additive inkjet printed humidity sensors on plastic substrate. Sens Actuators B Chem, 2012, 166-167 (0) 212-222.

24. Reddy A. S. G., Narakathu B. B., Atashbar M. Z., Rebros M., Rebrosova E., Joyce M. K. Fully Printed Flexible Humidity Sensor. Proc Eng, 2011, 25 (0) 120-123.

25. Weremczuk J., Tarapata G., Jachowicz R. S. The ink-jet printing humidity sorption sensor— modelling, design, technology and characterization. Meas Sci Technol, 2012, 23 (1) 014003.

26. Starke E., Turke A., Krause M., Fischer W. J., editors. Flexible polymer humidity sensor fabricated by inkjet printing, Solid-State Sens, Actuators and Microsyst Conf (TRANSDUCERS 2011), June 5-9, 2011, Beijing, China, pp. 1152-1155.

27. Kang U., Wise K. D. A high-speed capacitive humidity sensor with on-chip thermal reset. Electron Devices, IEEE Transactions on, 2000, 47 (4) 702-710.

28. Kevin J. Engler M. F., Humidity sensor formed on a ceramic substrate in association with heating components, Patent US7635091 B2, 2005.

29. Igreja R., Dias C. J. Dielectric response of interdigital chemocapacitors : The role of the sensitive layer thickness. Sens Actuators B Chem, 2006, 115 (1) 69-78.

30. Igreja R., Dias C. J. Analytical evaluation of the interdigital electrodes capacitance for a multi-layered structure. Sens Actuators A Phys, 2004, 112 (2-3) 291-301.

31. Molina-Lopez F., Briand D., de Rooij N. F. Decreasing the size of printed comb electrodes by the introduction of a dielectric interlayer for capacitive gas sensors on polymeric foil: Modeling and fabrication. Sensors and Actuators B: Chemical, (0).

32. Kummer A. M., Hierlemann A., Baltes H. Tuning Sensitivity and Selectivity of Complementary Metal Oxide Semiconductor-Based Capacitive Chemical Microsensors. Anal Chem, 2004, 76 (9) 2470-2477.

33. Van Gerwen P., Laureyn W., Laureys W., Huyberechts G., Op De Beeck M., Baert K., et al. Nanoscaled interdigitated electrode arrays for biochemical sensors. Sens Actuators B Chem, 1998, 49 (1-2) 73-80. 34. Schonauer D., Moos R. Detection of water droplets on exhaust gas sensors. Sens Actuators B Chem, 2010, 148 (2) 624-629.

35. Tetelin A., Pellet C, Laville C, N'Kaoua G. Fast response humidity sensors for a medical microsystem. Sens Actuators B Chem, 2003, 91 (1-3) 211-218.

36. Yang B., Aksak B., Lin Q., Sitti M. Compliant and low-cost humidity nanosensors using nanoporous polymer membranes. Sens Actuators B Chem, 2006, 114 (1) 254-262.

37. Molina-Lopez F., Briand D., de Rooij N. F., Smolander M., Fully inkjet-printed parallel- plate capacitive gas sensors on flexible substrate, Sensors, 2012 IEEE, October 28-31, 2012, Taipei, Taiwan, pp. 1-4.