Gas sensor with negligible response to humidity and temperature

Abstract

In one example, a gas sensor is provided, including a substrate, an isolation region formed on an outer edge of the substrate, a microheater formed on the isolation region, a sensing layer formed on the substrate within the isolation region, and a source and a drain formed around the sensing layer and within the isolation region.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. Provisional Patent Application Serial No. 62 / 837,839, filed April 24, 2019, which is incorporated herein by reference in its entirety. Technical Field

[0003] The present disclosure generally relates to gas sensors. Background Art

[0004] Selectivity refers to the ability of a gas sensor to selectively respond to one or more analytes and depends in part on the insensitivity of the gas sensor to changes in ambient relative humidity and temperature. Sensitivity to these ubiquitous changes is currently the main limiting factor for important gas sensing applications such as air quality monitoring and medical diagnostics. The fact that most active sensing materials of gas sensors interact with water makes it challenging to achieve humidity insensitivity. The response of chemical or gas sensors to even the slightest environmental changes can lead to inaccurate gas detection and / or gas concentration interpretation, which in turn leads to sensor signal deviation. Therefore, eliminating humidity and temperature sensitivity is critical to obtaining a robust and accurate sensor signal.

[0005] Research on making gas sensors insensitive to humidity and temperature can be divided into two strategies: computational methods and experimental methods. Computational methods involve signal processing using data from humidity and temperature sensors and sensor data of different concentrations at different humidity and temperature levels to calculate the exact gas concentration. Multivariate calibration methods, such as principal component regression (PCR), partial least squares (PLS) and artificial neural networks (ANN), have been used to compensate for the sensor response to humidity changes. Due to the calibration complexity required for different combinations of relative humidity and temperature, the need for large data sets to train ANNs and linear data sets in other methods make computational methods unfavorable. In addition, both humidity sensors and temperature sensors need to have selective signal responses relative to each other, which is experimentally impractical and requires further post-processing, which in turn increases the complexity of calibration.

[0006] Experimental approaches include functionalizing with hydrophobic materials and modifying the active sensing material through annealing, doping, or special growth conditions. These techniques reduce the available sites for water interaction, thereby reducing the sensor's response to humidity. However, previous work investigating these techniques has not been able to completely eliminate the response to humidity changes, but has focused on the response to gases at different humidity levels or the response to different humidity. Previous work has also not considered the response to gases at different humidity levels and the response to different humidity as a combined problem. The dominant technology for gas sensing, metal oxide semiconductor (MOS) sensors, are insensitive to humidity due to their high operating temperatures (e.g., >200°C), which evaporate any water molecules on the active sensing portion; however, high temperatures make MOS sensors unfavorable in consumer electronics applications due to power consumption and safety issues. Summary of the Invention

[0007] According to aspects described herein, a gas sensor is provided. One disclosed feature of an embodiment is a gas sensor comprising a substrate, an isolation region formed on an outer edge of the substrate, a microheater formed on the isolation region, a sensing layer formed on the substrate within the isolation region, and a source and a drain formed around the sensing layer and within the isolation region.

[0008] In another aspect, the present disclosure provides a method for manufacturing a gas sensor. The method includes providing a substrate, forming a silicon dioxide isolation region on an outer edge of the substrate, doping a source region and a drain region on the substrate, patterning a sensing layer region between the source region and the drain region, doping the sensing layer region, defining source and drain contacts on the source and drain regions, wherein the source and drain contacts include layers of nickel and tungsten, forming a tungsten microheater on the silicon dioxide isolation region, and depositing a sensing layer on the sensing layer region. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The teachings of the present invention may be readily understood by considering the following detailed description in conjunction with the accompanying drawings, in which:

[0010] Figure 1 illustrates an example cross-sectional view of a gas sensor of the present disclosure;

[0011] Figures 2A-2B An example process flow diagram illustrating a method of manufacturing a gas sensor of the present disclosure is illustrated;

[0012] Figure 3 illustrates an example graph illustrating H2 sensor response characteristics of an example gas sensor having a FOTS / Pt sensor layer of the present disclosure;

[0013] Figure 4 illustrates an example graph plotting chip temperature versus microheater power of the present disclosure;

[0014] Figure 5 illustrates example sensor responses of an example gas sensor of the present disclosure to hydrogen gas at various humidity levels;

[0015] Figure 6 illustrates an example graph of sensor response versus humidity of the present disclosure;

[0016] Figure 7 illustrates an example graph illustrating the effect of a heater on the response of a gas sensor of the present disclosure;

[0017] Figures 8A-8D illustrates example graphs of various characteristics of the gas sensor of the present disclosure with respect to various hydrogen gas concentrations;

[0018] Figure 9 An example graph illustrating sensor response of the gas sensor of the present disclosure to hydrogen gas with the microheater turned on and off; and

[0019] Figure 10 An example graph illustrating the sensor response of the gas sensor of the present disclosure to nitrogen gas is shown.

[0020] To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures. DETAILED DESCRIPTION

[0021] The present disclosure provides a gas sensor that is insensitive to humidity and / or temperature changes and a method for manufacturing the same. As described above, eliminating humidity and / or temperature sensitivity can result in a better gas sensor. Current methods and sensors can use computational or experimental methods to minimize humidity and temperature sensitivity. However, these current methods can be computationally intensive, have high power consumption, and / or present safety issues at the high temperatures required to eliminate humidity sensitivity.

[0022] The present disclosure provides a gas sensor that eliminates the sensor's response to changes in relative humidity and / or temperature. The gas sensor can be fabricated using a chemically sensitive field-effect transistor (CS-FET) with a microheater. A CS-FET is a nanoscale silicon transistor in which the electrical gate is replaced by a chemical sensing layer composed of nanoparticles.

[0023] The work function and / or morphology of the sensing layer changes upon exposure to the target chemical, resulting in strong output drain current modulation and achieving high detection sensitivity. In the past, microheaters were pulsed immediately upon detection of the target gas, significantly improving recovery time. Unlike previous approaches or sensor designs with bulk silicon CS-FETs, the present disclosure utilizes localized on-chip microheater integration around the sensor.

[0024] Figure 1 An example gas sensor 102 of the present disclosure is illustrated. Figure 1 The diagram shows a top view 100 of a gas sensor 102, a cross-sectional view 120 of the gas sensor 102, a close-up view 140 of the sensor layer 106, and a close-up view 160 of capillary condensation of water 162 occurring in the cavities between nanoparticles 164. In one embodiment, one or more of the gas sensors 102 can be formed on a substrate 108. The substrate 108 can be a silicon substrate. The silicon can include high-quality silicon wafers having a sheet resistivity in the range of 10-20 ohm-centimeters (Ω·cm). The combination of the gas sensor 102 and the substrate 108 may also be referred to herein as a "chip," "gas sensing device," "CS-FET," etc.

[0025] In one embodiment, each gas sensor 102 may include a microheater 104 (shown in view 120) formed on top of an isolation region 110. The isolation region 110 may include grown and patterned silicon dioxide. The microheater 104 may be formed of tungsten (W) and operated at a constant voltage to maintain a desired power output and heat the substrate 108 and gas sensor 102 to a desired temperature. The desired temperature may allow the gas sensor 102 to be insensitive to humidity changes and / or ambient temperature changes. The operating parameters are discussed in more detail below.

[0026] In one embodiment, the controller 150 can be electrically coupled to the microheater 104 to control the operation of the microheater 104. In addition, a temperature sensor (not shown) can be added to the chip and connected to the controller 150. The controller 150 can adjust the amount of power in response to changes in ambient temperature (e.g., by adjusting the current to maintain a constant voltage). Thus, the controller 150 can help regulate the temperature or maintain a constant operating temperature of the gas sensor 102 (as measured by the temperature sensor) to ensure that the gas sensor 102 is insensitive to changes in humidity and / or temperature.

[0027] In one embodiment, the gas sensor 102 may include source and drain contacts 112 (shown in view 120) formed on top of source and drain regions 114. The source and drain regions 114 may be n-type doped regions or p-type doped regions. In one embodiment, the source and drain regions 114 may be n-type doped with phosphorus. The source and drain contacts 112 may be formed from layers of nickel (Ni) and tungsten (W).

[0028] The sensor layer 106 can be formed on top of the sensing layer region 116 located between the source and drain regions 114. In one embodiment, the sensor layer 106 can be formed by depositing nanoparticles 164. View 140 illustrates the dispersion of the nanoparticles 164 in the sensor layer 106. As discussed in further detail below, the nanoparticles 164 can form cavities between the nanoparticles 164 that allow capillary condensation of water 162 in the cavities.

[0029] The type of material used for the nanoparticles 164 may depend on the type of gas that the gas sensor 102 detects. For example, for hydrogen gas detection, the sensor layer 106 may include platinum (Pt) nanoparticles on trichloro(1H,1H,2H,2H-perfluorooctyl)silane (FOTS). For nitrogen oxide gas detection, the sensor layer 106 may include indium oxide (InO x ) nanoparticles. It should be noted that the sensor layer 106 may also use other materials to detect other gases not described herein. The examples provided should not be considered limiting.

[0030] In one embodiment, multiple gas sensors 102 may include the same type of sensing layer 106 to detect the same gas. In one embodiment, multiple gas sensors 102 may include different sensing layers 106 to detect different gases. For example, a first gas sensor 102 may have a sensor layer 106 with Pt / FOTS to detect hydrogen, and a second gas sensor 102 may have a sensor layer 106 with InO x The sensor layer 106 can be configured to detect nitrogen oxide gas, etc. Thus, different gases can be detected by a single gas sensor device.

[0031] Figures 2A-2B An example process flow diagram of a method 200 for producing or manufacturing the gas sensor 102 of the present disclosure is illustrated. The method 200 can be performed by various tools within a manufacturing facility under the control of a central processor or controller that coordinates the operation of the tools. Examples of tools that can be used are described herein, but it should be noted that other methods for performing each block of the method 200 are possible and are within the scope of the present disclosure.

[0032] Method 200 may begin at block 202. At block 202, a substrate 108 is provided. The substrate 108 may comprise a high-quality silicon {100} wafer having a sheet resistivity in the range of 10-20 Ω·cm. Prior to processing, the silicon wafer may be cleaned in a standard piranha bath (e.g., 1:4 hydrogen peroxide / sulfuric acid) at 120 degrees Celsius (°C). A 10-second soak in a 1:10 hydrofluoric acid bath may remove any native oxide.

[0033] At block 204, isolation regions 110 may be formed on substrate 108. Isolation regions 110 may be thermally grown silicon dioxide. The silicon dioxide may be grown to a thickness of approximately 350 nanometers (nm). The silicon dioxide may be grown at 1000° C. and atmospheric pressure using a three-step dry (5 minutes), wet (55 minutes), and dry (5 minutes) oxidation process for 55 minutes. Fixed-angle ellipsometry may be used to verify oxide thickness.

[0034] At block 206, the trenches 120 may be etched away to form source and drain regions. A photolithography process (e.g., patterning, exposure, baking, and etching processes) may be used to form the trenches 120. In one embodiment, the photolithography process may be a standard i-line photolithography process (e.g., Fujifilm, photoresist: OiR 906-12, developer: OPD-4262) and wet etching of the isolation oxide (e.g., in 5:1 buffered hydrofluoric acid for 5 minutes).

[0035] At block 208, an ion implantation process 122 may be performed. The ion implantation process 122 may be used to dope portions of the exposed portion of the substrate 108 to form source and drain regions 114. In one embodiment, the ion implantation may use 4.5e 14 cm -2 , phosphorus, 15KeV.

[0036] At block 210, source and drain regions 114 may be formed. In one embodiment, to complete the formation of n+2 doped regions (e.g., source and drain regions 114), phosphorus is driven in and activated in the substrate 108 in the source and drain regions 114 by performing a rapid thermal anneal (RTA) at 1050° C. in nitrogen (N2) for 30 seconds. It should be noted that phosphorus is used for n-type doping. However, other chemicals or compounds may be used for p-type doping.

[0037] At block 212, a channel region may be formed using photolithography and etching processes. For example, a mask 122 may be deposited or formed over the source and drain regions 114. The mask 122 may be thermally grown silicon dioxide similar to the silicon dioxide grown for the isolation regions 110.

[0038] At block 214, the mask 122 may be etched to expose undoped portions of the substrate 108 between the source and drain regions 114. The exposed portions of the substrate 108 may be doped to form a channel region. Ion implantation (e.g., 5e 11 cm -2 , phosphorus, 18KeV) and then doping was performed by performing RTA at 900° C. for 1 second in N2.

[0039] At block 216 , the remaining portions of the mask 122 may be removed or etched away. Additionally, a nickel silicide layer 130 may be formed on the channel region 116 .

[0040] At block 218, source and drain contacts 112 may be formed. In one embodiment, a separate source-drain metallization mask may be used to define the source and drain contacts 112. The source-drain metallization mask may underlap the doped source and drain regions by 11 micrometers (μm). Thereafter, argon may be sputtered to etch the native oxide. A sputtering tool may then be used to deposit 20 nm of nickel and 50 nm of tungsten in the source and drain regions 114, followed by stripping in acetone.

[0041] At block 220, an annealing process may be performed to complete the nickel silicide layer 130 and the source-drain metallization of the source and drain contacts 112. For example, to achieve ohmic source and drain contacts, nickel silicide may be performed using RTA in forming gas (e.g., 5% hydrogen (H2) in N2) at 400°C for 5 minutes.

[0042] At block 222, a microheater 104 may be formed on the isolation region 110. For example, photolithography and tungsten sputtering may be used to pattern the microheater 104. In one embodiment, 200 nm of tungsten may be sputtered.

[0043] At block 224, the sensor layer 106 may be deposited. As described above, the type of material deposited for the sensor layer 106 may depend on the type of gas to be detected by the gas sensor 102. For example, for a hydrogen sensor, FOTS may be deposited using an AMST molecular vapor deposition MVD 100. Following this, platinum may be deposited by electron beam evaporation of 1 nm of platinum.

[0044] In one embodiment, for a nitrogen oxide sensor, InO x Can pass 1.5nm InO x The gas sensor 102 may then be annealed in forming gas at 150° C. for 1 hour after deposition, which may complete the fabrication process.

[0045] As described above, in one embodiment, the sensor layer 106 may be Figure 1 The one shown in Figure 1 is Pt / FOTS. Due to the strong interaction between Pt and hydrogen and the enhanced sensor performance characteristics of FOTS under Pt, as shown in Figure 2, the sensor performance characteristics of FOTS under Pt are enhanced. Figure 3 As shown in the graphs 302 and 304 in FIG, the layer can be used to detect hydrogen. Graph 302 shows the V D=0.1 volts (V) for 0.5% H2, where the sensing layer is a 1 nm Pt CS-FET. Graph 304 illustrates the sensor response at room temperature at V D = Example sensor response to 0.5% H2 at 0.1 volt (V), where the sensing layer is a 1 nm Pt CS-FET with FOTS underneath, as used in the present disclosure. While the example of graph 304 can enable detection of hydrogen at ppm levels, it is also highly sensitive to relative humidity changes, making it a good candidate for demonstrating technology for the gas sensor 102 of the present disclosure.

[0046] One approach to achieving negligible cross-sensitivity to relative humidity changes using the CS-FET platform is to operate the microheater 104 in a constant voltage mode so that the chip (e.g., the gas sensor 102 formed on the substrate 108, as described above) is at a temperature slightly above room temperature. Infrared imaging of the chip at various microheater powers between 0 and 560 mW shows that the chip temperature increases linearly with power, as shown in FIG. Figure 4 As shown in the curve diagram 402 in FIG.

[0047] For a relative humidity change from 50% to 90% and a drain bias of 0.8 V, the CS-FET exhibits a response of 3844%, as shown in Figure 5 The sensor response is calculated as the percentage change relative to the baseline current value (I peak -I baseline ) / I baseline ) x 100. The ambient temperature of the gas sensor 102 is regulated at 25°C, and since the relative humidity level does not increase above 95%, the chip is operated above the dew point to allow condensation to occur.

[0048] Figure 6 An example graph 602 is shown, which illustrates an example plot of sensor response versus humidity. The graph 602 may be obtained from Figure 5 The graph 502 shown in FIG is extracted to reveal the hysteresis of the adsorption and desorption curves at room temperature, which is characteristic of capillary condensation. Figure 1 A schematic diagram of the phenomena occurring on the nanoparticles is shown in view 160 of FIG.

[0049] To find the approximate relative humidity level at which capillary condensation will occur at room temperature, it can be assumed that the walls of the nanoparticles 164 are much taller in height than the nanoparticle cavity dimensions. Figure 1 In view 140 of FIG. 1 , the cavity size between most nanoparticles 164 may be approximately 2 nm. It can be assumed that the validity of the Kelvin equation for sub-10 nm is sufficient to generate Figure 1A rough approximation of the relative humidity level at which capillary condensation 162 begins is valid as shown in the diagram 160. An approximate form of the Kelvin equation can be applied, as shown in equation (1) below:

[0050] Equation (1)-

[0051] where p sat is the saturated vapor pressure, p v is the vapor pressure, γ is the surface tension of water, V m is the molar volume of water, θ is the contact angle of water with the nanoparticle wall surface, d is the diameter of the capillary, R is the universal gas constant, and T is the temperature. Given that pure water completely wets platinum without contaminants, it can be assumed that the platinum nanoparticles are pure in mass and exhibit a contact angle with water of zero. Using the standard value of the surface tension of water at room temperature of 72 dynes / cm and the molar volume of 18 cm 3 , the relative pressure p, which is equivalent to the relative humidity, can be calculated v / p sat The value was 59%, which indicates condensation between cavities formed in the nanoparticle assembly at pressures below the dew point.

[0052] However, the sensor’s response to such relative humidity changes decreases exponentially with increasing microheater power, e.g. Figure 5 As shown in graph 504 in FIG, at a power of 372 mW and a corresponding chip temperature of 37 ± 3°C, there is a negligible sensor response of 11.6%. As the surface temperature increases, the evaporation rate of any condensed water tends to increase. In addition, since adsorption is an exothermic reaction, the physical adsorption rate of water molecules on the sensing layer 106 may decrease with increasing chip temperature (this can be explained by Le Chatelier's principle). For the same reason, the overall hydrogen response may also decrease with increasing chip temperature, as shown in FIG. Figure 7 702 shown in FIG. However, the reduction in the overall hydrogen response may be offset by the elimination of the humidity response. Relatively low chip temperatures (less than <100°C) can eliminate the humidity response because water can evaporate rather than boil away. Furthermore, unlike active "thick" films in MOS and field-effect transistors, the high surface area to volume ratio of nanoparticles 164 allows for high evaporation rates.

[0053] Furthermore, a slight decrease in sensitivity to hydrogen can be achieved at relative humidity levels of 50% and 90%, where the sensor responses are 490% (250 ppm) and 1488% (1000 ppm) and 464% (250 ppm) and 1539% (1000 ppm), respectively. Figure 5This measurement can be performed using a different CS-FET sensor with a drain bias of 0.65V (to match the baseline current to a CS-FET with a drain bias of 0.8V), for which the chip temperature to eliminate the humidity response can be 64±8°C. Due to the linear sensor characteristics observed between 100ppm and 1000ppm, hydrogen concentrations of 250ppm and 1000ppm can be selected for these tests, as shown in FIG. Figure 8A As shown in graphs 802 and 804 of FIG. For example, graph 802 illustrates an example sensor response at various hydrogen concentrations. Graph 804 illustrates the sensor response versus various hydrogen concentrations. Graphs 802 and 804 allow for the assumption that sensitivity (e.g., sensor response per ppm) is constant across hydrogen concentration levels. The measurements shown in graph 802 can be performed at a chip temperature of 37±3°C to eliminate response to humidity changes.

[0054] Figure 8B The response (t 90 ) and recovery (t 10 ) The trend of time changing with hydrogen concentration. In the graph 806, the symbol t 90 It is the time it takes for the sensor to reach 90% of its peak response value from the baseline current. Symbol t 10 is the time it takes for the sensor to recover from its peak value to 10% of its baseline current.

[0055] exist Figure 8C , graph 810 shows that the sensor proves to be highly selective to other gases such as methane, carbon dioxide, ammonia, nitrogen dioxide, and sulfur dioxide, but not to hydrogen sulfide. Figure 8D Graph 812 in FIG. 1 illustrates the change in chip temperature over time when the microheater 104 is turned on for ten days and the chip temperature is maintained at 51° C. ± 6° C. Graph 812 shows that the temperature exhibits negligible drift, which means that the active material in the sensor layer 106 remains intact and is unaffected by the continuous heater operation.

[0056] In one embodiment, another advantage of using the microheater 104 is that the gas sensitivity does not change significantly with changing ambient temperature. Figure 9 Graphs 902 and 904 are illustrated. Graph 902 illustrates the sensor response of the gas sensor 102 when the microheater 104 is off. Graph 904 illustrates the sensor response of the gas sensor 102 when the microheater 104 is on.

[0057] In one embodiment, three hydrogen pulses with concentrations of 100 ppm, 600 ppm, and 1000 ppm may be injected at ambient temperatures of 15° C., 25° C., and 35° C. with a drain bias of 0.6 V. It can be seen that as the ambient temperature decreases by 20° C., the sensitivity decreases by a factor of approximately 5, from 2.2% / ppm at 35° C. to 0.4% / ppm at 15° C., as shown in graph 902.

[0058] However, when the microheater 104 was kept on at 372 mW of power (a chip temperature of 35 ± 3°C), the sensitivity remained approximately constant (1.6% / ppm at 35°C, 1.7% / ppm at 25°C, and 1.8% / ppm at 15°C, as shown in graph 904). In one embodiment, the benefit of maintaining constant sensitivity regardless of ambient temperature can greatly simplify the calibration process and ensure that low concentration levels are detected as the ambient temperature decreases. This can allow the gas sensor 102 of the present disclosure to be deployed for practical use.

[0059] In addition to demonstrating humidity-selective hydrogen detection using a Pt / FOTS system, the gas sensor 102 of the present disclosure can utilize active sensing materials for other gases using a CS-FET platform. Figure 10 Graphs 1002 and 1004 illustrating the ability of the gas sensor 102 to detect nitrogen are shown. For example, graph 1002 illustrates the ability of the gas sensor 102 to detect nitrogen when the drain bias is 4V and InO x With a thin film (e.g., approximately 1.5 nm) as the active material, the CS-FET exhibited a response of 895% to a relative humidity change from 50% to 90%. With the microheater 104 turned on to maintain the chip temperature at 70±8°C, the gas sensor 102 exhibited negligible sensor response to humidity changes, as well as a constant sensor response (-56% and -52% for 100 ppb NO2 at 50% and 90% humidity levels, respectively). A graphical representation of the negligible sensor response is shown in graph 1004.

[0060] In summary, the present invention demonstrates that by using a microheater 104 to maintain the temperature of the CS-FET slightly above the ambient temperature, the effects of relative humidity changes can be eliminated. The present disclosure also demonstrates the additional benefit of constant gas sensitivity using this technology at different ambient temperatures. This paper demonstrates hydrogen sensing using Pt / FOTS as the active material and its application in InO x Results of the application of CMOS technology for nitrogen dioxide sensing.

[0061] The microheater material and design can be optimized for various applications. For example, with different materials, it is expected that the power required to reach the chip temperature level required to eliminate the humidity response may be lower, thereby expanding the applicability of this technology to gas sensors in consumer electronics. As mentioned above, the temperature sensor can also be fabricated on the same chip, so that the microheater 104 can be looped with a proportional-integral-derivative (PID) controller to maintain a constant chip temperature regardless of the ambient temperature to achieve a constant sensor response to the same gas concentration level.

[0062] While the present disclosure provides examples of sensing layers for detecting various gases such as hydrogen and nitrogen dioxide, it should be noted that other sensing materials can be used for other gases. As long as the microheater 104 is used to maintain the desired temperature above room temperature, different sensing materials can be used to detect different types of gases.

[0063] In one embodiment, to perform the measurements of the values described herein, a CS-FET device chip can be wire bonded to an 84-pin J-bend lead chip carrier. Pure dry air can be used as the diluent gas. For H2 (e.g., Figure 5 , 8 and 9) and NO2 (e.g. Figure 10 ) sensing experiments, 1% H2 in N2 (Gasco) and 1 ppm NO2 in N2 (Gasco) can be used as sources, respectively. The selectivity measurements in graph 810 of FIG8 can be performed using 2.5% CH4, 100 ppm CO2, 50 ppm NH3, 5 ppm NO2, 50 ppm SO2, and 50 ppm H2S in N2 (Mesa gas) as sources. Typical gas flow rates can be from 1 to 100 sccm, and the diluent (air) flow rate can be approximately 1000 sccm. Gas delivery can be controlled by mass flow controllers (Alicat Scientific).

[0064] Measurements involving changes in relative humidity and temperature can be performed in an ESPEC humidity and temperature cabinet LHU-113 with a gas outlet 1-2 cm from the sensor chip, or otherwise in a walk-in fume hood. The CS-FET sensor can be biased using a Keithley 428 current preamplifier, and the current signal can be acquired using a LabVIEW-controlled data acquisition unit (National Instruments, NI USB-6211). The microheater can be powered by an Agilent E3631ADC power supply, and all measurements can be performed using a microheater placed on a die adjacent to the die with the CS-FET.

[0065] Although various embodiments have been described above, it should be understood that they are presented by way of example only and not limitation. Therefore, the breadth and scope of the preferred embodiment should not be limited by any of the above exemplary embodiments, but should be limited only in accordance with the appended claims and their equivalents.

Claims

1. A gas sensor comprising: substrate; an isolation region formed on an outer edge of the substrate; a micro heater formed on the isolation region; a controller coupled to the microheater to adjust an amount of voltage to the microheater in response to changes in ambient temperature measured by a temperature sensor to maintain the gas sensor at a constant operating temperature that is insensitive to changes in ambient relative humidity; a sensing layer formed on the substrate inside the isolation region; as well as A source and a drain are formed around the sensing layer and inside the isolation region.

2. The gas sensor of claim 1, wherein the microheater comprises a tungsten layer.

3. The gas sensor of claim 2, wherein the tungsten layer comprises approximately 200 nanometers.

4. The gas sensor of claim 1, wherein the sensing layer comprises platinum nanoparticles on trichloro(1H,1H,2H,2H-perfluorooctyl)silane (FOTS) for hydrogen sensor.

5. The gas sensor of claim 4, wherein the platinum nanoparticles in the sensing layer are approximately 1 nanometer thick. 6 . The gas sensor of claim 1 , wherein the sensing layer comprises indium oxide particles for a nitrogen oxide sensor.

7. The gas sensor of claim 6, wherein the indium oxide particles in the sensing layer are approximately 1.5 nanometers thick.

8. The gas sensor of claim 1, wherein the source and drain electrodes each further comprise nickel and tungsten contacts.

9. The gas sensor of claim 1, wherein the microheater is powered with a constant voltage to heat the gas sensor to a temperature higher than room temperature.

10. The gas sensor of claim 9, wherein the temperature comprises 37 degrees Celsius + / - 3 degrees.

11. The gas sensor of claim 9, wherein the microheater operates at a power of approximately 372 milliwatts. 12 . The gas sensor of claim 1 , wherein a drain bias voltage of 0.6 V to 0.7 V is applied to the source and drain.

13. A method for manufacturing a gas sensor, comprising: providing a substrate; forming a silicon dioxide isolation region on an outer edge of the substrate; doping a source region and a drain region on the substrate; patterning the sensing layer region between the source region and the drain region; doping the sensing layer region; defining a source contact and a drain contact on the source and drain regions, wherein each of the source and drain contacts comprises a layer of nickel and tungsten; forming a tungsten micro heater on the silicon dioxide isolation region; depositing a sensing layer on the sensing layer region; as well as A controller is provided that is coupled to the microheater to adjust the amount of voltage to the microheater in response to ambient temperature changes to maintain the gas sensor at a constant operating temperature that is insensitive to changes in ambient relative humidity.

14. The method of claim 13, wherein the sensing layer comprises platinum nanoparticles on trichloro(1H,1H,2H,2H-perfluorooctyl)silane (FOTS) for a hydrogen sensor or indium oxide particles for a nitrogen oxide sensor.

Images