Coated microhotplate

Most microhotplate-based chemical sensors have been realized as multi-chip solutions with separate transducer and electronics chips. One example includes a gas sensor based on a thin metal film [16]. Another example is a hybrid sensor system comprising a tin-oxide-coated microhotplate, an alcohol sensor, a humidity sensor and a corresponding ASIC chip (Application Specific Integrated Circuit) [17]. More recent developments include an interface-circuit chip for metal oxide gas sensors and the conccept for an on-chip driving circuitry architecture of a gas sensor array [18,19]. 

Fig. 4.1. (a) Micrograph of a circular microhotplate without sensitive layer, (b) SEM-micrograph of a metal-oxide coated microhotplate... 

The first device is a circular microhotplate (Sect. 4.1). One important guideline was to implement the microhotplate in CMOS technology with a minimum of post-CMOS micromachining steps. Additionally the hotplate had to be optimized for drop-coating with nano crystalline tin-oxide layers. This microhotplate was cointegrated with circuitry, and the respective monolithic sensor system will be discussed in Sect. 5.1. 

The circular heater design also perfectly matches the shape of the sensitive tin-oxide droplet No excess area is heated, and the heat losses to ambient air are reduced. A SEM (Scanning Electron Microscope) micrograph of a microhotplate coated with a Sn02-droplet is shovm in Fig. 4.1. The grainy structure of the nanocrystaUine oxide material is clearly visible. 

In order to determine the thermal time constant of the microhotplate in dynamic measurements, a square-shape voltage pulse was applied to the heater. The pulse frequency was 5 Hz for uncoated and 2.5 Hz for coated membranes. The amplitude of the pulse was adjusted to produce a temperature rise of 50 °C. The temperature sensor was fed from a constant-current source, and the voltage drop across the temperature sensor was amplified with an operational amplifier. The dynamic response of the temperature sensor was recorded by an oscilloscope. The thermal time constant was calculated from these data with a curve fit using Eq. (3.29). As already mentioned in the context of Eq. (3.37), self-heating occurs with a resistive heater, so that the thermal time constant has to be determined during the cooHng cycle. 

Fig. 4.10. Measured relative temperature differences between the temperatiu e sensors T2 to and T. The microhotplate was coated with a nano-crystalline Sn02 droplet...

The microhotplate with the transistor heater was electrothermally characterized similarly to the procedures presented in Sect. 4.1.3. Special care was taken to exclude wiring series resistances by integration of on-chip pads that allow for accurate determination of Fsg and sd- With the two on-chip temperature sensors in the center (Tm) and close to the transistor (Tt) the temperature homogeneity across the heated area was assessed as well. Both sensors were calibrated prior to thermal characterization. The relative temperature difference (Tj - Tm)/Tm was taken as a measure for the temperature homogeneity of the membrane. The measured thermal characteristics of a coated and an uncoated membrane are summarized in Table 4.6. The experimental values have been used for simulations according to Eq. (4.10). 

Table4.6. Experimentally determined values of the thermal resistance, the thermal time constant, and the temperatm-e homogeneity for uncoated and coated, transistor-heated microhotplates...

The microhotplate was coated with a thick-film tin-oxide droplet as described in Sect. 4.1.2. To characterize the chemical-sensor performance, the chip was exposed to CO concentrations from 5 to 50 ppm in humidified air at 40% relative humidity (23.4 °C water vaporization temperature) (see Sect. 5.1.8 for a description of the gas test measurement setup). 

First chemical test measurements have been conducted with the array chip. Figure 6.19 shows the results that have been obtained simultaneously from three microhotplates coated with different tin-dioxide-based materials at operation temperatures of 280 °C and 330 °C in humidified air (40% relative humidity at 22 °C). The first microhotplate (pHPl) is covered with a Pd-doped Sn02 layer (0.2wt% Pd), which is optimized for CO-detection, whereas the sensitive layer on microhotplate 3 contains 3 wt% Pd, which renders this material more responsive to CH4. The material on microhotplate 2 is pure tin oxide, which is known to be sensitive to NO2. Therefore, the electrodes on microhotplate 2 do not measure any significant response upon exposure to CO or methane. The digital register values can be converted to resistance values by taking into account the resistor bias currents [147,148]. The calculated baseline resistance of microhotplate 1 is approximately 47 kQ, that of hotplate 2 is 370 kQ and the material on hotplate 3 features a rather large resistance of nearly 1MQ. 

The thermal resistance of the circular hotplate was measured to be 5.8 °C/mW for the coated and uncoated transducer. An increased thermal resistance is desirable for sensor arrays with one approach being the reduction of the heated membrane area. At the moment the smallest possible diameter of drop-deposited tin-oxide is 100 pm. A microhotplate with a heated area of 100 pm in diameter was fabricated and featured an increased thermal resistance of approximately 10 °C/mW. 

The last and most advanced system presented in this book includes an array of three MOS-transistor-heated microhotplates (Sect. 6.3). The system relies almost exclusively on digital electronics, which entailed a significant reduction of the overall power consumption. The integrated C interface reduces the number of required wire bond connections to only ten, which allows to realize a low-prize and reliable packaging solution. The temperature controllers that were operated in the pulse-density mode showed a temperature resolution of 1 °C. An excellent thermal decoupling of each of the microhotplates from the rest of the array was demonstrated, and individual temperature modulation on the microhotplates was performed. The three microhotplates were coated with three different metal-oxide materials and characterized upon exposure to various concentrations of CO and CH4. 

---