Chapter 2 Theory
3.3 Measurement system
3.3.3 Drift measurement
On measurement of drift characteristic , Id-Vg curves at pH 7 will be extracted for every particular period we set in stress program . Then under the condition of
constant Id , we plot the diagram of Vg versus time . Drift characteristic of ISFET changes fast initially and then keeps stable several hours later . Using differential measurement of ISFET-ReFET pair , the goal to reduce this non-ideal effect can be obtained . Total measurement time is 25870 seconds and Vg depends on the Id current that we keep constant .
3.4 References
[1] U. Guth
,
“ Investigation of corrosion phenomena on chemical microsensors ＂, Electrochimica Acta 47 (2001) p.201-210[2] George T. Yu ,“ Hydrogen ion diffusion coefficient of silicon nitride thin films ＂, Applied Surface Science 202 (2002) p.68-72
[3] P. Hein ,“ Drift behavior on ISFET with nitride gate insulator ＂, Sensors and Actuators B , 13-14 (1993) p.655-656
[4] I-Yu Huang ,“ Fabrication and characterization of a new planar solid-state
reference electrode for ISFET sensors ＂, Thin Solid Films 406 (2002) p.255-261 [5] Hung-Kwei Liao,“ Multi-structure ion sensitive field effect transistor with a
metal light shield ＂, Sensors and Actuators B61 (1999) p.1-5
[6] P. Bergveld ,“ How electrical and chemical requirements for REFETs may coincide ＂, Sensors and Actuators 18 (1999) p.309-327
[7] Paul A. Hammond , Danish Ali , and David R. S. Cumming ,“ Design of a Single-Chip pH Sensor Using a Conventional 0.6-μm CMOS Process ＂, IEEE Sensors Journal , vol. 4 , No. 6 December (2004)
[8] P. A. Hammond , D. R. S. Cumming , D. Ali ,“ A Single-Chip pH Sensor Fabricated by a Conventional CMOS Process ＂, IEEE (1991)
Chapter 4
Results and Discussions
4-1 Drift of different kinds of ISFET-ReFET pairs
In our work , we choose five kinds of CMOS-compatible materials which are silicon nitride (Si3N4) 、 Hafnium oxide (HfO2) 、 Titanium oxide (TiO2) 、 Zirconium oxide (ZrO2) 、 and Aluminium oxide (Al2O3) as sensing layers of ISFET . Sputter and E-gun are used for deposition of HfO2 film . So we have twelve kinds of ISFET-ReFET pairs for demonstrate the drift characteristics of sensing layers we choose here . Table. 4-1 shows the composition of ISFET-ReFET pairs . In different kind of ISFET-ReFET pair , we keep Id constant respectively . The drift characteristics curves of every kind of ISFET-ReFET pairs in our work are shown in Fig. 4-1 ~ Fig. 4-12 . The relative drift after ten thousand seconds and the correction coefficient are listed in Table. 4-2 .
We can find that the drifts of silicon nitride film and E-gun HfO2 film are very unstable . But we can correct the shortcomings by using differential measurement setup . The correction coefficient means that the drift improving percentage after ten thousand seconds . If the ISFET-ReFET pair has compatible drift characteristics , the result of improving drift characteristics is remarkable .
4-2 To choose the proper ReFET
According to our results about ISFET-ReFET pairs , we can obtain the drift characteristics of every kind of sensing films we choose . First , we must choose a sensing layer of ReFET that has similar drift characteristics with the sensing layer of ISFET we used here . Because the ISEFT-ReFET pair has compatible drift
characteristics , we can obtain better result of improving drift characteristics . And then we take the sensitivities of sensing layers into consideration in order to define ReFET . Even if the sensitivity of ISFET-ReFET pair is not quite high after differential measurement , we still can distinguish the voltage difference due to different pH value of buffer solution .
4-3 Conclusions
In sample-1、sample-10 and sample-11 , the correction coefficient is quite high because of the compatibility of drift characteristics of two kinds of sensing layers . That is to say we can obtain a great improvement of drift characteristics . If the difference of drift between two kinds of sensing layers is large , the result of improving drift is limited (sample-3、sample-5 and sample-12) . Once the difference of drift between two kinds of sensing layers is quite large , we even may not obtain any improvement (sample-4 and sample-7) . According to our results about drift , we can choose the most appropriate ReFET that has the most compatible drift characteristics with ISFET we use here and even predict the effect of drift to sensitivity of ISFET-ReFET pair . Once the drift characteristics can be reduced as possible , the stability of ISFET-ReFET pair will get better naturally to solving a major problem of ISFET .
According to our results , the S-HfO2-Si3N4 pair has the best result of drift characteristic (0.19mv/hr after 10000 s) . And the TiO2-ZrO2 pair has the best of drift correction up to 98 % .
Chapter 5 Future Work
5-1 Future work
According to our results , we can have a data base to choose proper ReFET for the ISFET we used . We can find that some kind of ISFET-ReFET pair has great improvement to drift characteristics . In order to realize miniaturization , we can integrate reference electrode with ISFET-ReFET pair on the same chip in the future . Combining with MEMS technology , we also can fabricate a multi-channel sensor for sensing different kinds of materials at the same chip .
Besides , the external differential measurement circuit is a subject that we have to study in the future . Even we can focus on compensation of the non-ideal effect at the same time .
Figure 1-1 Schematic representation of MOSFET (a) , ISFET (b) , and electronic diagram (c) .
Figure 2-1 Potential profile and charge distribution at oxide / electrolyte interface
Sensor effect e.g. pH shift
Uat
Ua2 = Uat Ua1 = Um + Uat
Signal difference
Ua1-Ua2 = Um
ReFET ISFET
Um
Common mode disturbance : undefined metal-
liquid interface : leakage current : temperature
Uat
Figure 2-2 Differential measurement setup of an ISFET-ReFET pair
`
silicon
(a) RCA clean bare silicon
Wet oxide
(b) wet oxidation
(c) maskⅠ
silicon
(d) screening oxidation
Figure 3-1 Experimental process
silicon
(e) implantation
silicon
PE - oxide
(f) passivation layer deposition
silicon
(g) maskⅡ
silicon
(h) dry oxidation
Figure 3-1 Experimental process
silicon
(i) sensing layers deposition and mask Ⅲ
silicon
(j) Ti / Pt deposition and mask Ⅳ
silicon
(k) backside Al deposition
Figure 3-1 Experimental process
Diameter (mm): 100+/-0.5 Type / Dopant : P / Boron
Orientation : <100>
Resistivity (ohm-cm):1-10 Thickness (μm) ：505-545
Grade : Prime
Table 3-1 Specifications of wafers
TiO2 HfO2 Al2O3
Density 4.26 13.3 3.9 Z-ratio 0.4 0.36 0.336
Tooling 50.47 65 50
Current (mA) 1 ~ 60 1 ~ 60 1 ~ 50
Rate (Å/sec) 1.2 0.5 1.8
Pressure (Torr) 5*10-6 5*10-6 5*10-6
E – gun
Table 3-2-a Parameters of sensing layers deposition with E – gun
parameters of HfO2 sputter parameters of ZrO2 sputter
power : 200 W power : 200 W
Ar / O2 : 24 / 8 ( sccm ) Ar / O2 : 24 / 8 ( sccm ) Density : 13.9 Density : 6.51
Acoustic impendance : 24.53 Acoustic impendance : 14.72 Tooling factor : 0.533 Tooling factor : 0.533
Rate : 0.01 Å / s Rate : 0.01 Å / s pre sputter 60W for 10 min pre sputter 60W for 10 min
Pressure : 7.6×10-3 Pressure : 7.6×10-3
Sputter
Table 3-2-b Parameters of sensing layers deposition with Sputter
parameters of LP-nitride deposition NH3 : 17 sccm
SiH2Cl2 : 85 sccm 1000 Å for 13 min Temperature : 850 ℃
Pressure : 180 mT
LPCVD
Table 3-2-c Parameters of sensing layers deposition with LPCVD
HDP-RIE
Process pressure : 10 mTorr flow rate of CHF3 : 40 sccm
flow rate of Ar : 40 sccm ICP power : 600 W Bias power : 150 W etch time of LP-nitride : 21 sec
Table 3-3 Recipe of HDP-RIE for LP-nitride
silicon
Figure 3-2 The system of measurement
sample L R sample L R
#1 S-HfO2 LP nitride #7 S-ZrO2 LP nitride
#2 E-HfO2 LP nitride #8 S-HfO2 S-ZrO2
#3 E-TiO2 LP nitride #9 S-HfO2 E-Al2O3
#4 E-Al2O3 LP nitride #10 E-TiO2 S-ZrO2
#5 E-HfO2 S-HfO2 #11 E-TiO2 E-Al2O3
#6 E-HfO2 E-TiO2 #12 S-ZrO2 E-Al2O3
Table 4-1 Table of composition of ISFET – ReFET pairs
0 5000 10000 15000 20000 25000 30000
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7
VG (V)
Time(s)
S-HfO2
Figure 4-1-a drift characteristic of sputter HfO2 (sample-1)
0 5000 10000 15000 20000 25000 30000 1.0
1.1 1.2 1.3 1.4 1.5 1.6 1.7
VG (V)
Time (s)
LP-nitride
Figure 4-1-b drift characteristic of LP-nitride (sample-1)
0 5000 10000 15000 20000 25000 30000
0.00 0.05 0.10 0.15 0.20
VG (V)
Time (s)
sample-1
Figure 4-1-c drift characteristic of sample-1
0 5000 10000 15000 20000 25000 30000 0.6
0.8 1.0 1.2 1.4 1.6 1.8
VG (V)
Time (s)
E-HfO2
Figure 4-2-a drift characteristic of E-gun HfO2 (sample-2)
0 5000 10000 15000 20000 25000 30000
0.6 0.8 1.0 1.2 1.4 1.6 1.8
VG(V)
Time(s)
LP-nitride
Figure 4-2-b drift characteristic of LP-nitride (sample-2)
0 5000 10000 15000 20000 25000 30000
Figure 4-2-c drift characteristic of sample-2
0 5000 10000 15000 20000 25000 30000
0.8
Figure 4-3-a drift characteristic of E-gun TiO2 (sample-3)
0 5000 10000 15000 20000 25000 30000
Figure 4-3-b drift characteristic of LP-nitride (sample-3)
0 5000 10000 15000 20000 25000 30000
-0.5
Figure 4-3-c drift characteristic of sample-3
0 5000 10000 15000 20000 25000 30000
Figure 4-4-a drift characteristic of E-gun Al2O3 (sample-4)
0 5000 10000 15000 20000 25000 30000
1.0
Figure 4-4-b drift characteristic of LP-nitride (sample-4)
0 5000 10000 15000 20000 25000 30000
Figure 4-4-c drift characteristic of sample-4
0 5000 10000 15000 20000 25000 30000
-0.8
Figure 4-5-a drift characteristic of E-gun HfO2 (sample-5)
0 5000 10000 15000 20000 25000 30000
Figure 4-5-b drift characteristic of Sputter HfO2 (sample-5)
0 5000 10000 15000 20000 25000 30000
-1.60
Figure 4-5-c drift characteristic of sample-5
0 5000 10000 15000 20000 25000 30000 1.0
1.1 1.2 1.3 1.4 1.5
VG (V)
Time (s)
E-HfO2
Figure 4-6-a drift characteristic of E-gun HfO2 (sample-6)
0 5000 10000 15000 20000 25000 30000
1.0 1.1 1.2 1.3 1.4 1.5
VG (V)
Time (s)
E-TiO2
Figure 4-6-b drift characteristic of E-gun TiO2 (sample-6)
0 5000 10000 15000 20000 25000 30000
Figure 4-6-c drift characteristic of sample-6
0 5000 10000 15000 20000 25000 30000
0.0
Figure 4-7-a drift characteristic of Sputter ZrO2 (sample-7)
0 5000 10000 15000 20000 25000 30000
Figure 4-7-b drift characteristic of LP-nitride (sample-7)
0 5000 10000 15000 20000 25000 30000
1.8
Figure 4-7-c drift characteristic of sample-7
0 5000 10000 15000 20000 25000 30000 0.8
0.9 1.0 1.1 1.2 1.3 1.4 1.5
VG (V)
Time (s)
S-HfO2
Figure 4-8-a drift characteristic of Sputter HfO2 (sample-8)
0 5000 10000 15000 20000 25000 30000
1.4 1.5 1.6 1.7 1.8 1.9
VG (V)
Time (s)
S-ZrO2
Figure 4-8-b drift characteristic of Sputter ZrO2 (sample-8)
0 5000 10000 15000 20000 25000 30000 0.3
0.4 0.5 0.6 0.7 0.8
VG (V)
Time (s)
sample-8
Figure 4-8-c drift characteristic of sample-8
0 5000 10000 15000 20000 25000 30000
0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5
VG (V)
Time (s)
S-HfO2
Figure 4-9-a drift characteristic of Sputter HfO2 (sample-9)
0 5000 10000 15000 20000 25000 30000 0.7
0.8 0.9 1.0 1.1 1.2
VG (V)
Time (s)
E-Al2O3
Figure 4-9-b drift characteristic of E-gun Al2O3 (sample-9)
0 5000 10000 15000 20000 25000 30000
1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5
VG (V)
Time (s)
sample-9
Figure 4-9-c drift characteristic of sample-9
0 5000 10000 15000 20000 25000 30000 1.5
1.6 1.7 1.8 1.9 2.0 2.1
VG (V)
Time (s)
E-TiO2
Figure 4-10-a drift characteristic of E-gun TiO2 (sample-10)
0 5000 10000 15000 20000 25000 30000
1.0 1.1 1.2 1.3 1.4 1.5 1.6
VG (V)
Time (s)
S-ZrO2
Figure 4-10-b drift characteristic of Sputter ZrO2 (sample-10)
0 5000 10000 15000 20000 25000 30000 0.3
0.4 0.5 0.6 0.7 0.8
VG (V)
Time (s)
sample-10
Figure 4-10-c drift characteristic of sample-10
0 5000 10000 15000 20000 25000 30000
1.4 1.5 1.6 1.7 1.8 1.9 2.0
VG (V)
Time (s)
E-TiO2
Figure 4-11-a drift characteristic of E-gun TiO2 (sample-11)
0 5000 10000 15000 20000 25000 30000
Figure 4-11-b drift characteristic of E-gun Al2O3 (sample-11)
0 5000 10000 15000 20000 25000 30000
0.0
Figure 4-11-c drift characteristic of sample-11
0 5000 10000 15000 20000 25000 30000 1.0
1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8
VG (V)
Tim e (s)
S-ZrO 2
Figure 4-12-a drift characteristic of Sputter ZrO2 (sample-12)
0 5000 10000 15000 20000 25000 30000
0.8 1.0 1.2 1.4 1.6 1.8
VG (V)
Time (s)
E-Al2O3
Figure 4-12-b drift characteristic of E-gun Al2O3 (sample-12)
0 5000 10000 15000 20000 25000 30000 0.0
0.1 0.2 0.3 0.4 0.5
VG (V)
Time (s)
sample-12
Figure 4-12-c drift characteristic of sample-12
Type Drift (mv/hr) Difference Correction(%) L/R (10000 s ~)
Sample-1 S-HfO2 2.85369 0.190548 93.32 nitride 3.04423 0.190548 93.74 Sample-2 E-HfO2 16.1036 3.74291 76.76 nitride 19.8465 3.74291 81.14 Sample-3 E-TiO2 27.3981 5.896106 78.48 nitride 21.5025 5.8956 72.58 Sample-4 E-Al2O3 6.912363 4.090433 40.82 nitride 2.82193 4.090433 - 44.95 Sample-5 E-HfO2 5.18336 1.90321 63.28
S-HfO2 3.280151 1.90321 41.98 Sample-6 E-HfO2 5.01777 2.29111 54.34 E-TiO2 7.30888 2.291115 68.65 Sample-7 S-ZrO2 0.226389 1.4155 -525.25
nitride 1.64234 1.4155 13.81 Sample-8 S-HfO2 6.96181 0.22707 96.74 S-ZrO2 6.73497 0.22707 96.63
Table 4-2 The drift after 10000 s and the relative correction coefficient
Type Drift (mv/hr) Difference Correction(%) L/R (10000 s ~)
Sample-9 S-HfO2 1.27486 0.838866 34.2 E-Al2O3 2.11327 0.838866 60.305 Sample-10 E-TiO2 22.22541 0.431 98.06
S-ZrO2 22.65891 0.431 98.1 Sample-11 E-TiO2 18.4696 0.54397 97.05
E-Al2O3 19.014 0.54397 97.14 Sample-12 S-ZrO2 5.744575 4.61626 19.64 E-Al2O3 10.36015 4.61626 55.44
Table 4-2 The drift after 10000 s and the relative correction coefficient