Next Article in Journal
Novel Methodology Based on Thick Film Gas Sensors to Monitor the Hydraulic Oil Ageing
Previous Article in Journal
Analysis of Single- and Double Core Planar Fluxgate Structures
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Proceeding Paper

Performance Degradations of MISFET-Based Hydrogen Sensors with Pd-Ta2O5-SiO2-Si Structure at Long-Time Operation †

by
Boris Podlepetsky
1,*,
Marina Nikiforova
2 and
Andrew Kovalenko
2
1
Micro- and Nanoelectronics Department, National Research Nuclear University MEPhI (Moscow Engineering Physics Institute), 115409 Moscow, Russia
2
Induko Ltd., 32/2 Seslavinskaia str., 121309 Moscow, Russia
*
Author to whom correspondence should be addressed.
Presented at the Eurosensors 2018 Conference, Graz, Austria, 9–12 September 2018.
Proceedings 2018, 2(13), 777; https://doi.org/10.3390/proceedings2130777
Published: 10 December 2018
(This article belongs to the Proceedings of EUROSENSORS 2018)

Abstract

:
There are presented the generalized results of studies of performance degradation of hydrogen sensors based on MISFET with structure Pd-Ta2O5-SiO2-Si. It was shown how responses’ parameters change during long-term tests of sensors under repeated hydrogen impacts. There were found two stages of time-dependence response’ instability, the degradation degree of which depends on operating conditions, hydrogen concentrations and time. To interpret results there were proposed the models, parameters of which were calculated using experimental data. These models can be used to predict performances of MISFET-based devices for long-time operation.

1. Introduction

The gas sensors based on the metal-insulator-semiconductor devices (MIS-capacitors and field-effect transistors called as MISFETs) have been studied by many investigators (e.g., [1,2,3,4,5,6,7,8,9,10]). A great contribution to the developments of gas-sensitive MIS devices has been made by the researchers at Linköping University [5]. MIS sensors with different gates’ material (palladium, platinum and iridium), with dielectric films SiO2, Si3N4-SiO2, TiO2-SiO2, Ta2O5-SiO2 have been investigated. Semiconductors Si, GaAs [3] and SiC [7] were used in MIS gas sensors to detect the low concentrations of gases H2 [2,3,4], NH3 [5], H2S [6] and CO [7]. The studies have shown that performances characteristics of MISFET-based hydrogen sensors depend on technological parameters [2], electrical modes [8], chip temperature [10] and external factors (other gases, irradiation [9]). The researchers at National Research Nuclear University MEPhI have developed and investigated the number of hydrogen sensors based on MIS-capacitors and MISFETs with structures Pd (or Pt)-SiO2-Si, Pd/Ti-SiO2-Si, Pd (or Pt)-Ta2O5-SiO2-Si. The experiments have demonstrated, that among developed MIS-sensors the integrated cells, containing on single chip MISFET with Pd-Ta2O5-SiO2-Si-structure, heater-resistor and temperature sensor, possess the best stability and reproducibility of characteristics [4].
Performances characteristics (sensitivity, stability and speed) are determined by sensor responses’ parameters, the repeatability of which becomes an important characteristic at long-term operation of gas-analytical devices. The motivations of this paper are to generalize data on researches of responses’ parameters changes of MISFET-based sensors on structure Pd-Ta2O5-SiO2-Si at long-term operation and to propose models, taking in to account the performance degradations.

2. Materials and Methods

The testing n-channel MISFET element based on Pd-Ta2O5-SiO2-Si structure was fabricated on single chip (2 × 2 mm2) together with (p–n)-junction temperature sensor and heater-resistor by means of conventional n-MOS-technology using laser evaporation Pd-films. To measure the sensor’s hydrogen responses there was used the special circuitry shown in Figure 1. The circuit provides the constant drain current ID = 0.1 mA and source-drain voltage VD = 1.0 V. In this circuitry the voltage V is equal to the gate voltage VG. The constant chip temperature (T = 130 ± 2 °C) was supported by the temperature-stabilization circuitry with feedback loop using on-chip thermo-sensor and heater.

3. Results

We have tested the sensors for 5 or 8 weeks (5 days a week in a row with 2 day breaks) by 5 of repeated hydrogen j-impacts at values δV0j and maximum response amplitudes δVCM of vs. C for each j-cycle (Figure 4). In each j-impact sensors were consecutively exposed to hydrogen pulses of different concentrations Ci (Figure 2). The indexes i, j, k and l are serial numbers of responses, ordinary, day and weak cycles respectively. There were measured the parameters of each i-response (Figure 3). There were calculated the sums of residual characteristics, instability responses’ parameters and its designations are presented in Table 1. Responses’ and models’ parameters changes are demonstrated in Figure 4, Figure 5, Figure 6 and Figure 7 and in Table 2.

4. Discussion

It was found that all MISFET-sensors at long-term operation have the following degradations’ features: (1) so-called “zero-line drift” (ZLD) – the changes of output voltage V at zero hydrogen concentration C; (2) reduction of hydrogen sensitivity S being equal |dV/dC|, if hydrogen exposition time tC increases. The following 4-component model was used to interpret the results:
V (C, t, tC) = V00 + ΔV00(t) − ΔV0S(C, tC) − ΔVC(C, tC)
ΔV00(t) =ΔV0M∙[1 − exp(−t/τ)]; ΔV0S(C, tC) = ΔV0SM(C, tC)·[1 − exp(−D/D0)];
ΔVC(C,tC) = ΔVCM(C, tC)∙[1 − exp(−kC×C)]; SdM = kC(C, tC) × ΔVCM(C, tC),
where V00 is a primary voltage being about 1.4 V (at T ≈ 130 °C). There are two sorts of ZLD: the initial time drift ΔV00 (t) and the drift ΔV0S(C,tC) associated with the total hydrogen dose D = ∫C(t)dt (C− time factor). The first one appears immediately after the turning the sensor in the operating mode (e.g., maximum changes of value ΔV00(t) can be equal ΔV0M in ranges from ±10 to ±50 mV during 1–5 min, and τ ≈ 75 s). The second sort of ZLD occurs due to operation of the sensor in a hydrogen environment. Summary ZLD (ΔV0Slkj) increases, if hydrogen exposition time tC is rising. These changes depend on time tC, the primary operating conditions (preliminary temperature-hydrogen treatments), chip temperature and electrical modes. Maximum changes of responses’ parameters are manifested in the first stages at values C-time factor D less than 25 (% vol.)×min, at D0 being equal to about 8 (% vol.)×min.

5. Conclusions

This paper generalized data on researches of responses’ parameters changes of MISFET-based sensors on structure Pd-Ta2O5-SiO2-Si at long-term operation and proposed models, taking into account the performance degradations. The all tested MISFET-sensors at long-term operation have degradations’ features: reduction of hydrogen sensitivity and “zero-line drift” (ZLD), which depend on operating conditions and accumulated hydrogen dose D (C-time factor). This can be explained by the effects of palladium aging (accumulation of Pd-H compounds and irreversible palladium swelling). Basic degradations’ parameters are manifested in the first stages at values D less than 25 (% vol.)×min. These effects for practical applications of hydrogen sensors were taken into account as the additive errors of “zero” (basic line drift).

Acknowledgments

Author acknowledges support from the MEPhI Academic Excellence Project (Contract No. 02.a 03.21.0005).

References

  1. Lundström, I.; Armgarth, M.; Spetz, A.; Winquist, F. Gas sensors based on catalytic metal-gate field-effect devices. Sens. Actuators 1986, 3–4, 399–421. [Google Scholar] [CrossRef]
  2. Fomenko, S.; Gumenjuk, S.; Podlepetsky, B.; Chuvashov, V.; Safronkin, G. The influence of technological factors on hydrogen sensitivity of MOSFET sensors. Sens. Actuators B Chem. 1992, 10, 7–10. [Google Scholar] [CrossRef]
  3. Lin, K.W.; Cheng, C.C.; Cheng, S.Y.; Yu, K.H.; Wang, C.K.; Chuang, H.M.; Liu, W.C. A novel Pd/oxide/GaAs metal–insulator–semiconductor field-effect transistor (MISFET) hydrogen sensor. Semicond. Sci. Technol. 2001, 16, 997–1001. [Google Scholar] [CrossRef]
  4. Podlepetsky, B.I.; Nikiforova, M.Y.; Gumenyuk, S.V. Stability investigation of the characteristics of integral hydrogen sensors. Instrum. Exp. Tech. 2001, 44, 257–258. [Google Scholar] [CrossRef]
  5. Lundström, I.; Sundgren, H.; Winquist, F.; Eriksson, M.; Krants-Rülcker, C.; Lloyd-Spets, A. Twenty-five years of field effect gas sensor research in Linköping. Sens. Actuators B Chem. 2007, 121, 247–262. [Google Scholar] [CrossRef]
  6. Kalinina, L.; Litvinov, A.; Nikolaev, I.; Samotaev, N. MIS-Field Effect Sensors for low concentration of H2S for enviromental monitoring. Procedia Eng. 2010, 5, 1216–1219. [Google Scholar] [CrossRef]
  7. Andersson, M.; Pearce, R.; LloydSpetz, A. New generation SiC based field effect transistor gas sensors. Sens. Actuators B Chem. 2013, 179, 95–106. [Google Scholar] [CrossRef]
  8. Podlepetsky, B.I. Integrated Hydrogen Sensors Based on MIS Transistor Sensitive Elements: Modeling of Characteristics. Autom. Remote Control. 2015, 76, 535–547. [Google Scholar] [CrossRef]
  9. Podlepetsky, B.I. Effect of irradiation on hydrogen sensors based on MISFET. Sens. Actuators B 2017, 238, 1207–1213. [Google Scholar] [CrossRef]
  10. Podlepetsky, B.I.; Nikiforova, M.Y.; Kovalenko, A.V. Chip temperature influence on characteristics of MISFET hydrogen sensors. Sens. Actuators B 2018, 254, 1200–1205. [Google Scholar] [CrossRef]
Figure 1. The simplified structure of the sensor characteristic measuring circuitry.
Figure 1. The simplified structure of the sensor characteristic measuring circuitry.
Proceedings 02 00777 g001
Figure 2. Time diagram of the typical j- ordinary cycle at different Ci.
Figure 2. Time diagram of the typical j- ordinary cycle at different Ci.
Proceedings 02 00777 g002
Figure 3. The typical i- hydrogen response and its parameters at Ci = 0.05% vol.
Figure 3. The typical i- hydrogen response and its parameters at Ci = 0.05% vol.
Proceedings 02 00777 g003
Figure 4. Experimental (symbols) and approximations (lines) of the responses’ residual sum values ΔV0 (ZLD) and amplitudes ΔVCvs. concentration C.
Figure 4. Experimental (symbols) and approximations (lines) of the responses’ residual sum values ΔV0 (ZLD) and amplitudes ΔVCvs. concentration C.
Proceedings 02 00777 g004
Figure 5. SZLD changes during the different day’s lk-cycles. A and P are active and passive stages of ZLD.
Figure 5. SZLD changes during the different day’s lk-cycles. A and P are active and passive stages of ZLD.
Proceedings 02 00777 g005
Figure 6. The changes of sensitivities SdM during the different day’s lk-cycles.
Figure 6. The changes of sensitivities SdM during the different day’s lk-cycles.
Proceedings 02 00777 g006
Figure 7. The changes of sensitivities SdM during the weak l-cycles.
Figure 7. The changes of sensitivities SdM during the weak l-cycles.
Proceedings 02 00777 g007
Table 1. Designations and nlkji-responses’ parameters (Ylkji) for the characterization of instabilities and performance degradations’ parameters of output voltage V(C,t,tC).
Table 1. Designations and nlkji-responses’ parameters (Ylkji) for the characterization of instabilities and performance degradations’ parameters of output voltage V(C,t,tC).
i-Responsej-Cyclek-and l-CyclesGeneral Values
V00—primary voltage Maximum amplitude ΔVCMAmplitude’ changes:
δVCM lk = ΔVCM − ΔVCMlk
N—number of sensors
τi—H2 pulse duration Amplitude’ changes:
δVCM j= ΔVCM − ΔVCMj
Y0—primary value of Y
Ci—H2 concentration break times tbk and tblaverage of Y, variation indices, degradation degree:
V0i—initial voltagedifferential sensitivity:
Sdj = dVC)/dC
cycle time tk= tj × jmax+ tbk Proceedings 02 00777 i001
ΔVCi—response amplitudecycle time tl = tk × kmax + tbl
δV0i—residual valuemaximum sensitivity SdM lk summary ZLD (j = 1; k = 1)
τ1i—response timebreak time tbjΔV0Sk= V0kV0(k+1)
τ2i—relaxation timecycle time tj = tji × imax+ tbjΔV0Sl= V0lV0(l+1);
ti—response periodZLD ΔV0j= V0j1V0j i maxΔV0S= V0111V0( lkji) max
Si = ΔVCi /Ci − sensitivitysummary SZLD (I = 1):C-time factor D = nknlDjmax
Di= (Ci·τi) is
Ci-time factor
ΔV0Slkj = V0lk1V0lk(j+1)Degradation rate vY = dY/dt
Table 2. Average values of responses’ and model parameters of conversion function ΔVC (C) for lkj-cycles.
Table 2. Average values of responses’ and model parameters of conversion function ΔVC (C) for lkj-cycles.
↓parameters/lkj→111155255355455555655755855Total ΔYD,%
ΔVCM, V/ρVCM, %0.46/4.30.41/3.70.39/3.80.38/3.90.37/4.00.36/4.221.7
kC, 1/%/ρkCM, %12/8.210/8.88.5/10.68.2/10.68.1/11.58.0/11.5 33.3
SdM, V/%/ρSdM, %5.52/12.54.1/12.53.32/14.43.12/14.52.96/15.52.88/15.747.8
vS, V/(% × day)-0.280.160.0240.0160.08-
Sj5, V/%/δSj5, %2.1/2.41.8/2.81.6/3.11.55/3.21.50/3.31.45/3.4531.0
ΔV0Slkj, mV33371673-
vΔV0, mV/(day)-7.43.21.40.6-
τ1i, s/τ2i, s (i = 2)10/159/158/157/167/1530/6.6
τ1i, s/τ2i, s (i = 5)7/177/166/166/167/1514.3/11.8
SdM, V/%/ρSdM, % (for Ci max = 1.0%)5.5/8.73.9/9.23.65/9.83.5/10.23.4/9.5-38.2
D, % × min0.266.51319.52632.53945.552-
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Podlepetsky, B.; Nikiforova, M.; Kovalenko, A. Performance Degradations of MISFET-Based Hydrogen Sensors with Pd-Ta2O5-SiO2-Si Structure at Long-Time Operation. Proceedings 2018, 2, 777. https://doi.org/10.3390/proceedings2130777

AMA Style

Podlepetsky B, Nikiforova M, Kovalenko A. Performance Degradations of MISFET-Based Hydrogen Sensors with Pd-Ta2O5-SiO2-Si Structure at Long-Time Operation. Proceedings. 2018; 2(13):777. https://doi.org/10.3390/proceedings2130777

Chicago/Turabian Style

Podlepetsky, Boris, Marina Nikiforova, and Andrew Kovalenko. 2018. "Performance Degradations of MISFET-Based Hydrogen Sensors with Pd-Ta2O5-SiO2-Si Structure at Long-Time Operation" Proceedings 2, no. 13: 777. https://doi.org/10.3390/proceedings2130777

APA Style

Podlepetsky, B., Nikiforova, M., & Kovalenko, A. (2018). Performance Degradations of MISFET-Based Hydrogen Sensors with Pd-Ta2O5-SiO2-Si Structure at Long-Time Operation. Proceedings, 2(13), 777. https://doi.org/10.3390/proceedings2130777

Article Metrics

Back to TopTop