sensors

Article
Investigation of Radiation Effects on FD-SOI Hall
Sensors by TCAD Simulations
Linjie Fan 1,2 , Jinshun Bi 1,2, *, Kai Xi 1 and Gangping Yan 1,2
1 Institute of Microelectronics, Chinese Academy of Sciences, Beijing 100029, China;
fanlinjie19@mails.ucas.ac.cn (L.F.); xikai@ime.ac.cn (K.X.); yangangping@ime.ac.cn (G.Y.)
2 School of Microelectronics, University of Chinese Academy of Sciences, Beijing 100049, China
* Correspondence: bijinshun@ime.ac.cn; Tel.: +86-10-82995940




Received: 1 July 2020; Accepted: 12 July 2020; Published: 16 July 2020


Abstract: This work investigates the responses of the fully-depleted silicon-on-insulator (FD-SOI) Hall
sensors to the three main types of irradiation ionization effects, including the total ionizing dose (TID),
transient dose rate (TDR), and single event transient (SET) effects. Via 3D technology computer aided
design (TCAD) simulations with insulator fixed charge, radiation, heavy ion, and galvanomagnetic
transport models, the performances of the transient current, Hall voltage, sensitivity, efficiency,
and offset voltage have been evaluated. For the TID effect, the Hall voltage and sensitivity of the
sensor increase after irradiation, while the efficiency and offset voltage decrease. As for TDR and SET
effects, when the energy deposited on the sensor during a nuclear explosion or heavy ion injection
is small, the transient Hall voltage of the off-state sensor first decreases and then returns to the
initial value. However, if the energy deposition is large, the transient Hall voltage first decreases,
then increases to a peak value and decreases to a fixed value. The physical mechanisms that produce
different trends in the transient Hall voltage have been analyzed in detail.

Keywords: Hall sensors; FD-SOI CMOS; radiation effects; total ionizing dose; transient dose rate;
single event transient; 3D TCAD simulations

1. Introduction
Hall sensors are the most common converters used to turn a magnetic field into an
electric signal. Owing to their advantages of non-contact, strong anti-interference, high linearity,
robustness, and versatility [1], Hall sensors are used in scientific detecting, brushless DC
motors, contactless measurements and so on [2–4]. Compared to the conventional bulk Si
CMOS (complementary metal-oxide-semiconductor) for Hall sensor manufacturing, the choice
of fully-depleted silicon-on-insulator (FD-SOI) technology brings several important advantages.
The FD-SOI structure not only has the advantages of less noise generation, lower biasing voltage,
and higher integration density [5–7], but it has also been confirmed that—compared with the bulk
structure—the characteristics of the sensors (such as sensitivity and efficiency) are improved because
of its thin thickness and low doping concentration [8].
When used in aerospace systems or nuclear weapon control systems, sensors must function
normally while being strongly irradiated. In general, the radiation environment that the sensors are
exposed to mainly space radiation and man-made nuclear radiation [9]. For instance, there are cosmic
rays, Van Allen Belt, and solar flares in the space environment, which may result in total ionizing
dose (TID) effects and single event transient (SET) effects on the sensors [10,11]. Moreover, transient
and high-energy X-ray and gamma-ray produced by nuclear explosions and dayglow could lead to
a transient dose rate (TDR) effect [12]. Therefore, it is necessary to study the radiation effects on the
performance of sensors used in harsh radiation conditions.

Sensors 2020, 20, 3946; doi:10.3390/s20143946 www.mdpi.com/journal/sensors

Sensors 2020, 20, 3946 2 of 12

Previously published literature mainly focused on a single type of radiation effect on the Hall sensor
under a certain radiation source, such as gamma [13–15], proton [16,17], and neutron radiation [18,19].
Karlova et al. [14] studied the influence of gamma radiation on sensors based on GaAs. However,
this research was limited to the analysis of the volt-ampere (IV) characteristics and noise spectral
density before and after gamma irradiation. Therefore, the analysis of the effects of irradiation
on magnetoelectric properties such as the Hall voltage and sensitivity has not been investigated.
Abderrahmane et al. [16] have irradiated GaN-based Hall sensors by protons with the energy of 380 keV
and fluence from 1014 p/cm2 to 1016 p/cm2 . The changes in the electron mobility, sheet resistance,
and sensitivity of the sensors before and after irradiation have been observed. Moreover, most published
studies focus on III-V semiconductor-based Hall sensors. There are few reports on the radiation effects
of silicon-based—especially FD-SOI—Hall sensors aimed at sensor and peripheral circuit monolithic
integration [20].
This work investigates the radiation effects on the performance of the FD-SOI Hall sensors
with the use of 3D technology computer aided design (TCAD) tools. It focuses on the impacts of
radiation effects (such as the TID, TDR, and SET effects) on the performance of the FD-SOI Hall
sensors. The figure-of-merits (FOMs) includes the Hall voltage, sensitivity, efficiency, and offset voltage.
The remainder of this paper is organized as follows. Section 2 introduces Hall sensor theory, the FD-SOI
Hall sensor structure, and the TCAD simulation methodology. Section 3 explores the performance of
the FD-SOI Hall sensor under different irradiation effects via physical simulations and discusses the
physical mechanism. Finally, Section 4 summarizes the radiation effects of FD-SOI Hall sensors.

2. Methodology and Physical Models

2.1. Hall Sensor Theory

This section briefly introduces the figure-of-merits (FOMs) definitions for Hall sensor evaluation.
The specific relations of the FOMs are shown in Table 1. Moreover, in the following sections, the changes
in these parameters during or after radiation will be evaluated.

Table 1. The figure-of-merits (FOMs) definitions for Hall sensor 1 .

FOMs Relations References

rH
Hall voltage (VH ) VH = G nqt Ibias B [1]
VH rH
Absolute sensitivity (SA ) SA = B = G nqt Ibias [1]
SA VH
i
V
Efficiency factor (η) η = P = Vbias ·Ibias ·B [ W·T [21]
Offset voltage (Voffset ) N/A [20]
1 where G is the geometrical correction factor, rH is the scattering factor, Ibias is the bias current, B is the magnetic field
induction, n is the carrier density, q is the electron charge, t is the thickness of the active region, P is the dissipated
power and Vbias is the bias voltage.

When the semiconductor with current flowing through is placed in a magnetic field, the carriers
in the semiconductor is biased to one side by Lorentz force, and then Hall voltage (VH ) is generated.
The sensitivity is one of the most significant FOMs related to a Hall sensor. Generally speaking,
the sensitivity is defined as the change in output concerning a given change in input. In addition,
a FOM related to the power consumption of the sensor should also be noted. An efficiency factor (η),
which is also known as power-related sensitivity, indicates how much Hall voltage will be generated by
consuming 1 W under the magnetic induction of 1 T. It is well known that even if the sensor is biased at
zero magnetic field, there is also a parasitic voltage named the offset voltage (Voffset ). The Voffset can be
generated by misalignment of contacts, asymmetry of the geometric dimensions and non-uniformity
of the active region material [20]. In practical applications, it is obvious that the Voffset should be as
small as possible.
Sensors 2020, 20, x FOR PEER REVIEW 3 of 12

uniformity
Sensors 2020, 20,of the
active region material [20]. In practical applications, it is obvious that the3 of
3946 Voffset
12
should be as small as possible.

2.2.
2.2. Device
Device for
for Simulation
Simulation
The
Theoptimal
optimalFD-SOI
FD-SOIHall Hallsensor
sensorstructure
structure has been
has demonstrated
been demonstrated in the authors’
in the previous
authors’ work
previous work[8].
There is a brief summary in Appendix A. The 3D geometric model and
[8]. There is a brief summary in Appendix A. The 3D geometric model and cross-section of the FD- cross-section of the FD-SOI
Hall sensor
SOI Hall are illustrated
sensor are illustratedin Figure 1. Moreover,
in Figure 1. Moreover, thethe
specific geometrical
specific geometricaldimensions
dimensionsandanddoping
doping
concentrations
concentrations of the FD-SOI Hall sensors are included in Table 2. The doping concentrationof
of the FD-SOI Hall sensors are included in Table 2. The doping concentration ofthe
the
50 nm silicon film above the buried oxide is 1 × 10 16 cm−3 , thereby forming the FD state. Four heavily
50 nm silicon film above the buried oxide is 1 × 10 cm , thereby forming the FD state. Four heavily
16 −3

doped
dopedcontacts
contactsare arelocated
locatedininthe thecenter
centerofofthe
thefour
foursides
sidesofofthe
thesilicon
siliconfilm.
film.When
Whenaabias
biascurrent
current(I(Ibias
bias))
isisapplied to the bias contacts, the Hall voltage can be detected at the Hall contacts
applied to the bias contacts, the Hall voltage can be detected at the Hall contacts under the magnetic under the magnetic
induction
inductionintensity
intensity(B)(B)ininthe
thenegative
negative direction
direction of of
thethez-axis. In particular,
z-axis. In particular, by applying a voltage
by applying to the
a voltage to
gate (V ) on the gate oxide layer, the sensors can be turned
the gate (Vg) on the gate oxide layer, the sensors can be turned on or off.
g on or off.

(a) (b)
Figure1.
Figure 1. (a)
(a) 3D
3Dgeometric
geometricmodel,
model,and (b)(b)
and cross-section of the
cross-section fully-depleted
of the silicon-on-insulator
fully-depleted (FD-
silicon-on-insulator
SOI) Hall sensor for the technology computer aided design (TCAD) simulation.
(FD-SOI) Hall sensor for the technology computer aided design (TCAD) simulation.

Table2.2.The
Table TheFD-SOI
FD-SOIHall
Hallsensors
sensorsdesign
designfeatures.
features.
Geometrical
GeometricalDimensions
Dimensions Doping
Doping Concentrations
Concentrations
L L 15 15
µmµm
W W 15 µm
15 µm
NAN=A1=×110
Siliconfilm
film P-type: × 16
10cmcm 16 −3
La 1 µm Silicon P-type: −3
La 1 µm
Wb 7.5 µm
Gate oxide W b
100 nmµm
7.5
SiliconGate
film oxide 50100
nm nm
Contacts N-type: ND = 1 × 1021 cm−3
BuriedSilicon
oxide film 40050nmnm
Substrate Contacts N-type: ND = 1 × 1021 cm−3
Buried oxide 5 400µm nm
Substrate 5 µm
2.3. Simulation Models
2.3. Simulation Models
The ionization damages of ionizing radiation in the materials lead to the severe degradation of
deviceThe
performance, triggeringofthree
ionization damages ionizing
ionizing radiation
radiation effects:
in the the TID,
materials leadTDR,
to theand SETdegradation
severe effects. of
Theperformance,
device TID effect refers to thethree
triggering effectionizing
of radiation damage
radiation related
effects: to time
the TID, TDR, accumulation
and SET effects.by X-ray,
gamma-ray,
The TIDand charged
effect refersparticles. Electron-hole
to the effect of radiation(e-h) pairsrelated
damage are generated in the insulator
to time accumulation byregion
X-ray,
during the TID
gamma-ray, radiation.
and charged Most electrons
particles. are quickly
Electron-hole swept
(e-h) out
pairs areof generated
the oxide under
in the the electricregion
insulator field
because of the
during the TIDlarge mobility
radiation. of electrons.
Most electrons are In this process,
quickly sweptsome
out ofelectrons
the oxidewill be recombined
under the electricwith
field
holes.
becauseAt of
thethesame
largetime, holes will
mobility also be relatively
of electrons. slowly transported
In this process, some electrons to thewill
Si/SiO interface under
be 2recombined with
the electric
holes. At thefield.
same Parts
time,ofholes
the holes are trapped
will also in theslowly
be relatively oxide transported
layer, forming a net
to the positive
Si/SiO oxide under
2 interface layer
trap charge [22].
the electric field.The
Parts final effect
of the is the
holes arenegative
trappeddrift of oxide
in the the threshold voltageainnet
layer, forming n-channel
positive transistor:
oxide layer
trap charge [22]. The final effect is the negative drift of the threshold voltage in n-channel transistor:
q
∆V ot = − tox ∆Not (1)
εox
Sensors 2020, 20, 3946 4 of 12

where εox is the dielectric constant of the oxide, tox is the thickness of the oxide layer, and ∆Not is the
net positive trap charge density of the oxide layer. In this work, the insulator fixed charge model in
Sentaurus TCAD activated in oxide layer is adopted to simulate the above TID effect [23].
The TDR effect occurs in the environment of transient large dose radiation, such as with
nuclear explosions and solar storms. The TDR radiation deposits a large amount of energy in the
semiconductor device, generating a huge number of e-h pairs. Different radiation sources (gamma-ray,
X-ray, or electron) lead to different charge generation and recombination rates [24–26]. In this work,
the main source taken into account for the radiation simulation is gamma-ray, which is also the main
method for evaluating the TDR effect of semiconductor devices and integrated circuits. The gamma
radiation model in Sentaurus TCAD is used to simulate the TDR effect. The generation of e-h pairs
caused by gamma radiation depends on the electric field (E), described by:

Gr = g0 D·Y(E) (2)

where g0 is the generation rate of e-h pairs, D is the dose rate, and Y(E) is a function related to the
electric field. Gr is derived as a linear function of the dose rate.
The SET effect refers to the generation of a large number of e-h pairs along the track of incident
high-energy particles hitting a semiconductor device. The electrodes collect the e-h pairs, causing soft
errors in the circuits and permanent damage in severe cases [27,28]. The heavy ion model in Sentaurus
TCAD has an input parameter linear energy transfer (LET), which describes the capability of energy
deposition in case of heavy ions penetrating a semiconductor. The generation rate of the e-h pairs
caused by heavy ion incidence is computed by:

G(l, w, t) = GLET (l)R(w, l)T (t) (3)

where GLET (l) is the linear energy transfer generation density, and R(w,l) and T(t) are the functions
describing the spatial and temporal variations of the generation rate, respectively.
The galvanic transport model handles the magnetic field acting on the semiconductor Hall sensor.
The galvanic transport model is based on the common drift-diffusion transport model enhanced with
magnetic field-dependent terms taking Lorentz force into account. The following equations for holes
and electrons govern its behavior:

→ → 1 → → → → →
Jα = µα gα + µα 2
[µ∗α B × gα + µ∗α B × (µ∗α B × gα )] (4)
1 + (µ∗α B)

→ →
where α = n or p, gα is the current vector without mobility, µ∗α is the Hall mobility, B is the magnetic field
→
vector, and B is the magnitude of the vector B [29]. Other than the above-mentioned radiation-related
and magnetic-field related models, the TCAD simulation also takes many physical models into account,
such as SRH generation/recombination and mobility degradation due to high doping concentration,
rough surface scattering, and high field saturation.

3. Experimental Results and Discussion

In this section, with the support of the TCAD simulation, the effects on the performance of
the FD-SOI Hall sensor during and after radiation are evaluated. Based on the simulation results,
the physical processes of the radiation effects are analyzed.

3.1. TID Effect

As described in Section 2.3, the effect of the TID on the device can be made equivalent by setting
the fixed positive charge in the oxide layer. Under low-dose radiation conditions, the trap charge
density (i.e., fixed charge density in the simulation) in the oxide induced by radiation is linearly
 Sensors 2020, 20, x FOR PEER REVIEW 5 of 12

charge densities is presented in Figure 2. The fixed charge-induced off-state leakage current (Ioff,
Ibias@Vg = 0 V) and the on-state current (Ion, Ibias@Vg = 3 V) are extracted and displayed in the inset.
As shown
Sensors 2020, 20, 3946 in Figure 2, an increase in the fixed charge density, which means an increase5of the
of 12
TID radiation, causes an increase in the magnitude of the off-state leakage current. When the charge
density reaches 5 × 1015 cm−2, the off-state leakage current is four orders of magnitude higher than the
related to the radiation dose, and the trap charge density tends to saturate at medium to high radiation
pre-radiation case, signifying that the sensor loses the capacity of a normal switch. In addition, since
doses [30]. Therefore, different fixed charge densities (Qf ) are equivalent to different radiation doses.
the fixed charge affects the threshold voltage, the on-state current also increases as the fixed charge
The transfer characteristic curve of the FD-SOI Hall sensor with different fixed charge densities is
density increases. With the change of Qf from 0 to 1 × 1016 cm−2, Ion increases from 90.2 µA to 177.5
presented in Figure 2. The fixed charge-induced off-state leakage current (Ioff , Ibias @Vg = 0 V) and the
µA (an increase of 96.78%).
on-state current (Ion, Ibias @Vg = 3 V) are extracted and displayed in the inset.

Qf increasing
10-4

10-5

10-6 16
Vbias=2 V
200
Qf (cm-2)
Ibias (A)

10-7 12
150
0

Ion (μA)
Ioff (μA)

8
-8 1x1014
10 4
100
5x1014
50 1x1015
-9 0
10 0 15 15 15 15 16
5x1015
2x10 4x10 6x10 8x10 1x10
1x1016
Qf (cm-2)
-10
10
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Vg (V)

Figure2. 2.Simulation
Figure Simulationresultsresultsofofthe Ibias
the –V
Ibias –Vg gcurve
curveofofaaFD-SOI
FD-SOIHall
Hallsensor
sensorwith
withchanges
changesininthe thefixed
fixed
charge
chargedensity
density (Q(Q
f ) fin
) inthe
theoxide.
oxide.TheTheinset
insetshowsshowsthe
thevariation
variationofof
the leakage
the current
leakage (Ioff
current (I)offand
) andthetheonon
current (Ion
current ) )with
(Ion withfixed
fixedcharge
chargedensity.
density.

AsToshown in Figure
evaluate 2, an
the effect of increase in theonfixed
fixed charge charge
sensor density, which
performance, means an
the magnetic increaseisofswept
induction the
TID radiation, causes an increase in the magnitude of the off-state leakage current.
from 0 T to 1 T with Vg = 3 V and Vbias = 2 V. Figure 3 demonstrates that the Hall voltage increases as When the charge
density reaches 5 ×density −2 , the off-state leakage current is four orders of magnitude higher than
1015 cmincreases.
the fixed charge At the same time, by the calculation of equation in Table 1, the
the pre-radiation
absolute case,
sensitivity signifying
also increases. that
Thethe sensor sensitivity
absolute loses the capacity
increasesoffrom a normal
86.49 switch.
mV/T toIn addition,
111.75 mV/T
since the fixed charge affects the threshold voltage, the on-state current
after adding 1016 cm−2 fixed charge. As expressed in the equation in Table 1, the Hall voltage also increases as the fixed
will
charge density increases. With the change of Q from 0 to 1 × 10 16 cm−2 , Ion increases from 90.2 µA to
f
increase as the bias current increases. The bias current is increased because of the fixed charge present
177.5 (an increase
µAoxide
in the of 96.78%).
layer, which leads to an increase in absolute sensitivity.
To evaluate the effect of fixed charge on sensor performance, the magnetic induction is swept from
0 T to 1 T with Vg = 3 V and Vbias = 2 V. Figure 3 demonstrates that the Hall voltage increases as the
fixed charge density increases. 125At the same time, by the calculation of equation in Table 1, the absolute
Qf (cm-2)
sensitivity also increases. The absolute sensitivity increases from 86.49 mV/T to 111.75 mV/T after
Qf increasing
0
adding 1016 cm−2 fixed charge. 100 As expressed 14 in the equation in Table 1, the Hall voltage will increase
1x10
as the bias current increases. The bias5x10 current
14 is increased because of the fixed charge present in the
oxide layer, which leads to an75 increase1x10 15
in absolute sensitivity.
VH (mV)

15
In addition to absolute sensitivity, 5x10
efficiency and 115offset voltage are also extracted, as shown in
1x1016 110
Figure 4. Both the Hall voltage 50and the bias current increase 105 because of the existence of a fixed charge,
SA (mV/T)

but the bias current growth rate is greater, which therefore leads to a decrease in the efficiency of the
100

95
25
sensor. The efficiency of the sensor drops from 479.2 V/WT 90
to 314.7 V/WT after adding 1016 cm−2 fixed
charge, while the offset voltage of the sensor decreases 85from 4.45 mV to 2.1 mV. The offset voltage is
15 15 15 15 16
0 2x10 4x10 6x10 8x10 1x10
related to the symmetry, uniformity, 0 resistance, and other factors Q cmof the sensor [20]. The existence of f
(
-2
)

the fixed charge in the oxide layer0.0 causes0.2 0.4 0.6 0.8 1.0 1.2field
the changes of the electric 1.4and even the resistivity of the
silicon film, so that the offset voltage changes.
B (T)
 from 0 T to 1 T with Vg = 3 V and Vbias = 2 V. Figure 3 demonstrates that the Hall voltage increases as
the fixed charge density increases. At the same time, by the calculation of equation in Table 1, the
absolute sensitivity also increases. The absolute sensitivity increases from 86.49 mV/T to 111.75 mV/T
after adding 1016 cm−2 fixed charge. As expressed in the equation in Table 1, the Hall voltage will
increase as the bias current increases. The bias current is increased because of the fixed charge present
Sensors 2020, 20,oxide
in the 3946 layer, which leads to an increase in absolute sensitivity. 6 of 12

125
Qf (cm-2)
Q increasing
0 f
1001x1014
Sensors 2020, 20, x FOR PEER REVIEW
Sensors 2020, 20, x FOR PEER REVIEW 5x1014 6 of 12
6 of 12
75 1x1015

VH (mV)
5x1015
Figure 3. Hall voltage (VH) versus magnetic induction
115
(B) simulated for sensors with different fixed
Figure 3. Hall voltage (VH) versus magnetic induction (B) simulated for sensors with different fixed
1x1016 110
50 oxide.
charge densities (Qf) in the
charge densities (Qf) in the oxide.
The inset shows the variation of the absolute sensitivity (SA) with
The inset shows the variation
105 of the absolute sensitivity (SA) with

SA (mV/T)
fixed charge density. 100
fixed charge density.
95
25 90

In In addition
addition to absolute
to absolute sensitivity,
sensitivity, efficiency
efficiency andand offset
offset
85
voltage
voltage areare also
also extracted,
extracted, as as
shownshownin in
Figure 4. Both the Hall
Figure 4. Both the Hall voltage and voltage and the bias current increase
0 the bias current increase because because of
0 2x10 4x10 6x10 8x10 1x10
the existence of a fixed charge, 15 15 15 15 16

Q cm of the existence of a fixed charge, f

(
-2
)

butbut
thethe
biasbias current
current growth
growth raterate is greater,
is 0.0
greater, which
which
0.2 from
therefore
therefore
0.4 479.2
0.6 0.8
leads
leads
1.0 to to a decrease
1.2a decrease
1.4 after
in the
in the efficiency
efficiency of the
of the
sensor. The efficiency of the sensor drops V/WT to 314.7 V/WT
sensor. The efficiency of the sensor drops from 479.2 V/WT to 314.7 V/WT after adding 1016 cm−2 fixed adding 10 16 cm−2 fixed

charge, while B (T) from 4.45 mV to 2.1 mV. The offset voltage is
charge, while thethe offset
offset voltage
voltage of the
of the sensor
sensor decreases
decreases from 4.45 mV to 2.1 mV. The offset voltage is
related
Figure 3. to
Hallthe symmetry,
voltage (V
related to the symmetry, uniformity, ) uniformity,
versus resistance,
magnetic and
induction
resistance, and other other
(B) factors
simulated
factors of
of theforthe sensor
sensors
sensor [20].
with
[20]. The The existence
different
existence of of
fixed
H
the
charge fixed charge
densities in
(Qthe the oxide layer causes the changes of the electric field
andand eventhethe resistivityof of
the fixed charge in f ) inoxide
the oxide.
layer The inset
causes theshows the of
changes variation
the of the
electric absolute
field sensitivity
even (SA ) with
resistivity
the the silicon
silicon
fixed charge film,
film, so that the offset voltage
so that the offset voltage changes.
density. changes.

480 4.5
480 4.5
4.0
440 4.0
440
Voffset (mV)

3.5
Voffset (mV)
η (V/WT)

3.5
η (V/WT)

400
400
3.0
3.0
360
360 2.5
2.5
320 2.0
320 2.0
15
0 2x10 4x1015 6x1015 8x1015 1x1016 0 2x10 15
4x1015 6x1015 8x10 15
1x1016
0 2x1015 4x1015 6x1015 8x1015 1x1016 0 2x1015 4x1015 6x1015 8x10
-2
15
1x1016
-2
Qf (Q
cm
f
(cm
-2
)
)
Qf (Q
cm
f
(cm
-2
)
)

(a) (a) (b)(b)

Figure 4. Simulated (a) efficiency factor (η), and (b) offset voltage (Voffset) as a function of fixed charge
Figure
Figure 4. Simulated
4. Simulated (a)(a) efficiencyfactor
efficiency factor(η),
(η),and
and (b)
(b) offset
offset voltage
voltage(V(V
offset) as a function of fixed charge
offset ) as a function of fixed charge
densities (Qf).
densities (Q
densities (Qf ). f).

3.2. TDR Effect

3.2. TDR Effect
3.2. TDR Effect
The The TDR effect investigates the changes in sensor characteristics at the moment of transient
TDR effect investigatesthethechanges
changes in
in sensor
sensor characteristics at at
thethe
moment of transient
The TDR
radiation
effect
like
investigates
a nuclear explosion. For the TCAD
characteristics
simulation, a high dose rate from
moment
510×810
of transient
8 rad(Si)/s
radiation
radiation like like a nuclear explosion. For the TCAD simulation, a high dose rate from 5 × rad(Si)/s
5 ×the to to
108negative
5 × 10 12 arad(Si)/s
nuclearwith explosion.
20 ns For the is
duration TCAD simulation,
established. a highthe
Meanwhile, doseB rate
of 2 from
T along rad(Si)/s
5 × 10 12 rad(Si)/s with 20 ns duration is established. Meanwhile, the B of 2 T along the negative
12 rad(Si)/s
to 5 × 10
direction of the with
z-axis 20isns duration
applied. is established.
TDR-induced Meanwhile,
photocurrents under the B of 2(VTg =along
off-state 0 V) thedetected
negative
direction of the z-axis is applied. TDR-induced photocurrents under off-state (Vg = 0 V) areare
detected
direction of the
atbiasV=bias2=V, z-axis
2 V, is
as shownapplied. TDR-induced
in Figure photocurrents under off-state (V = 0 V) are detected at
at V as shown in Figure 5. 5. g
Vbias = 2 V, as shown in Figure 5.

0.8 Vbias=2 V B=2 T

0.8 Vbias=2 V B=2 T
Irradiation time=20 ns
0.8 Irradiation time=20
Dose ns
rate [rad(Si)/s]
0.8 Dose rate [rad(Si)/s]
0.6
5x108
0.6 5x108
0.6 1x109
0.6
Ibias max (μA)

1x109
Ibias max (μA)

0.4 5x109
0.4 5x109
Ibias (μA)

10
1x10
Ibias (μA)

0.2 10
1x10
0.4 0.2 5x1010
0.4 0.0
5x1010
1x10
11
11
0.0 1x10
10
8
10
9
10
10
10
11
10
12
10
13
5x1011
10
8
10
9 10
10 Dose
11 12
10rate [rad(Si)/s]
10 10
13
5x1011 12
1x10
0.2 Dose rate [rad(Si)/s]
1x10
12

0.2 5x10
5x10
12
12

Start Irradiation
Start Irradiation

0.0
0.0
0.0 0.5 1.0 1.5 2.0 2.5
0.0 0.5 1.0 1.5 2.0 2.5
Time (μs)
Time (μs)
5. Transient
Figure Figure dose rate
5. Transient dose(TDR)-induced currents
rate (TDR)-induced currents in the FD-SOI
in the FD-SOIHall
Hallsensors. Theinset
sensors. The insetshows
showsthe
the
Figure 5. Transient dose rate (TDR)-induced currents in the FD-SOI Hall sensors. The inset shows the
change change
of the detected maximum bias current
of the detected maximum bias current(I ) with dose rate.
(Ibias) with dose rate.
bias
change of the detected maximum bias current (Ibias) with dose rate.
Sensors 2020, 20, 3946 7 of 12

As can be observed from Figure 5, at the moment of the start of transient radiation, a current is
detected at the bias contact. This is because a large number of e-h pairs are generated at the moment of
Sensors 2020, 20, x FOR PEER REVIEW 7 of 12
irradiation, and some e-h pairs that do not have enough time to recombine move under the electric field
generated by the As canbias be contacts
observed from to formFigure
a 5, at the moment
current. of the start
Moreover, of transient
as the dose rate radiation,
increases,a current
theismaximum
detected at the bias contact. This is because a large number of e-h pairs are generated at the moment
bias current (Ibias max ) detected is larger. When the dose rate reached 5 × 1012 rad(Si)/s, the detected
of irradiation, and some e-h pairs that do not have enough time to recombine move under the electric
Ibias max even fieldexceeded
generated 0.5 by µA.
the bias contacts to form a current. Moreover, as the dose rate increases, the
Figuremaximum
6. shows biasthecurrent
changes in) the
(Ibias max Hallisvoltage
detected during
larger. When theirradiation at different
dose rate reached dose rates.
5 × 1012 rad(Si)/s, the Since the
detected I even exceeded 0.5 µA.
sensor is in off-state, the initial Hall voltage value before irradiation is the superposition value of Voffset
bias max

Figure 6. shows the changes in the Hall voltage during irradiation at different dose rates. Since
and Hall voltage
Sensors 2020, 20,formed
x FOR PEER Ioff under the B. On the one hand, when the irradiation dose
byREVIEW rate is at
7 of 12
the sensor is in off-state, the initial Hall voltage value before irradiation is the superposition value of
a smaller value range (Figure 6a), the Hall voltage will first decrease and
Voffset and Hall voltage formed by Ioff under the B. On the one hand, when the irradiation dose rate is then return to the initial
As can be observed from Figure 5, at the moment of the start of transient radiation, a current is
value during the irradiation
at a smaller value rangeprocess.
(Figure 6a), Atthethis
Halltime, the
voltage minimum
will Hall
first decrease andvoltage
then return (Vto min )initial
H the will decrease
detected at the bias contact. This is because a large number of e-h pairs are generated at the moment
as the dose value
rate during
increases.the irradiation
On process.
the other At hand,
this time, the minimum
when the Hall voltagedose
irradiation (VH minrate
) willisdecrease
at a as
larger value
of irradiation, and some e-h pairs that do not have enough time to recombine move under the electric
the dose rate increases. On the other hand, when the irradiation dose rate is at a larger value range
range (Figure
field(Figure 6b),
generated the Hall
byHall voltage
the voltage
bias contacts will rapidly
to form decrease,
a current. then increase
Moreover, as the to reach the maximum value,
6b), the will rapidly decrease, then increase to reach thedose
maximumrate increases,
value, andthe
and finally
maximumdecrease bias to a fixed
current (I biasvalue. At
max) detected thisis time,
larger. the
Whenmaximum
the dose Hall
rate voltage
reached
finally decrease to a fixed value. At this time, the maximum Hall voltage (VH max) will increase 5 ×(V10H ) will
as theincrease
12 rad(Si)/s,
max the as
the dosedetected
rate Ibias max
doseincreases.
rate even exceeded 0.5 µA.
increases.
Figure 6. shows the changes in the Hall voltage during irradiation at different dose rates. Since
the sensor is in off-state, the initial Hall voltage value before irradiation is the superposition value of
6 10
Voffset and Hall voltage formed by Ioff under the B. On the one hand, when the irradiation dose rate is
at a smaller5 value range (Figure 6a), the Hall voltage will 8 first decrease and then return to the initial
6
value during 4
the irradiation process. At this5
time, the minimum Hall voltage (VH min) will decrease as
VH (mV)

V =2 V B=2 T
6
VH (mV)

bias

the dose rate increases. On the other hand, when

Irradiation time=20 ns
4 the irradiation dose rate is at a larger value range
VH min (mV)

Dose rate [rad(Si)/s]

(Figure 6b), 3 the Hall voltage will rapidly decrease,

3
5x10
8 V =2 V B=2 T
then4 increase to reach the maximum value, and bias

2 Irradiation time=20 ns
9
1x10
finally decrease
2 to a fixed value. At this time, 1
9
the maximum Hall
Dose voltage (VH max) will increase as the
rate [rad(Si)/s]
5x10
11
dose rate increases. 0
0
1x10
10 2 5x10
2x1010 4x1010 6x1010 8x1010 1x1011

1 Dose rate [rad(Si)/s]

5x10
10 1x10
12
11
0 12
1x10 5x10

6 0 0.0 0.5 1.0 1.5 2.0 2.5 10 0.0 0.5 1.0 1.5 2.0 2.5
Time (μs) Time (μs)
5 8
6
(a) (b)
4 5
VH (mV)

V =2 V B=2 T
6
VH (mV)

bias
Irradiation time=20 ns
Figure4 6. The changes in Hall voltage (VH) during the irradiation of (a) dose rate = 5 × 108~1 × 10118
of (a) dose rate = 5 × 10 ~1 × 1011
VH min (mV)

Figure 6. The changes in Hall voltage (V

Dose ) during
rate [rad(Si)/s]
H the irradiation
3rad(Si)/s and (b) dose rate = 5 × 10 ~5 × 10
3 11 12 V =2
8rad(Si)/s. The inset inV (a)
B=2 Tis the change in minimum Hall bias

with=dose 1011 ~5 × 101x10

rad(Si)/s and (b)2 dose 5 ×rate. 12 rad(Si)/s. The
5x10
4 Irradiation
voltage (VH min)rate 9 insettime=20
in (a) ns is the change in minimum Hall
2 1
voltage (VH min0) with dose rate. 5x10
9
Dose rate [rad(Si)/s]
11
0 2x10 4x10 6x10 8x10 1x10 10
1x10
10
10 10 2
10 11 5x10
8 Dose rate [rad(Si)/s]
1 B =2.5 T 5x10
10 6 1x10
12
From the equation in BTable =2 T
1, the Hall voltage 1x10
11 is related
05 to the 5x10
bias
12 and magnetic induction. Figure 7
0 6 V decreasing
depicts the variations
0.0 of
0.5the 1.0 Hall voltage
1.5 2.0during 2.5 irradiation with 0.5 a dose of 1 ×2.010112.5
rate1.5 rad(Si)/s under bias

4 0.0 1.0
VH (mV)

different bias voltages 4 and Time magnetic(μs) inductions. The magnetic induction and bias voltage mainly
VH (mV)

B =1 T V =0.5 V Time (μs) 3 bias

affect the initial and2minimum B =0.5(a)

T
Hall voltages. The larger theVbias =1 V voltages and magnetic inductions
(b) bias
2 V =2 V bias
are, the greater the initialB Hall voltage, which will result in a V larger
=2.5 V built-in electric field. Therefore,
Figure 6. The changes =0 T in Hall voltage (VH) during the irradiation of (a) dose rate = 5 × 108~1 × 1011
1
bias

0
the larger built-in electric
rad(Si)/s and (b)Dosedose field creates
raterad(Si)/s
rate= 1x10 a greater V H The
= 5 × 10 ~5 × 10 rad(Si)/s.
11 12 and a faster
Dose rate= 1x10 rad(Si)/s
min inset rate at which
in (a) is the change Vin minimum
11
increasing the
HallHall voltage
11

Irradiation time=20 ns V =2 V 0 Irradiation time=20 ns B=2 T bias

returns to its initial

voltage (VHvalue
min) with
-2 after
dose irradiation.
rate.
0.0 0.5
bias

1.0 1.5 2.0 2.5

0.0 0.5 1.0 1.5 2.0 2.5
Time (μs) Time (μs)
8
B =2.5 T (a) 6 (b)
B =2 T irradiation with a dose rate of 1 × 1011 5
6 7. During
Figure rad(Si)/s, the variations of Hall voltage (VH)
Vbias decreasing
under different (a) magnetic inductions (B), and (b) bias 4voltages (Vbias).
VH (mV)

4
VH (mV)

Vbias=0.5 V
3.3. SET Effect B =1 T 3
Vbias=1 V
2 B =0.5 T
The SET effect influences the characteristics of a2 sensorVbias =2 V
after a single high-energy particle
Vbias=2.5 V
B =0 T
injection. In the TCAD simulation, at the time of 1.52 µs,1heavy ions strike in the middle of the sensor
0 Dose rate= 1x1011 rad(Si)/s
Dose rate= 1x1011 rad(Si)/s Irradiation time=20 ns B=2 T Vbias increasing
Irradiation time=20 ns Vbias=2 V 0
-2 0.0 0.5 1.0 1.5 2.0 2.5
0.0 0.5 1.0 1.5 2.0 2.5
Time (μs) Time (μs)
(a) (b)
Figure
Figure 7. 7. During
During irradiation
irradiation with
with a doserate
a dose rateof
of 11 ×
× 10
1011rad(Si)/s, thethe
rad(Si)/s, variations of Hall
variations ofvoltage (VH)
Hall voltage (VH )
11

under different (a) magnetic inductions (B), and (b) bias voltages (Vbias).
under different (a) magnetic inductions (B), and (b) bias voltages (Vbias ).
3.3. SET Effect
The SET effect influences the characteristics of a sensor after a single high-energy particle
injection. In the TCAD simulation, at the time of 1.52 µs, heavy ions strike in the middle of the sensor
Sensors 2020, 20, 3946 8 of 12

3.3. SET Effect

The SET effect influences the characteristics of a sensor after a single high-energy particle injection.
Sensors 2020, 20, x FOR PEER REVIEW 8 of 12
In the TCAD simulation, at the time of 1.52 µs, heavy ions strike in the middle of the sensor along the
negative the z-axis 2 /mg. The ion trajectories
along thedirection
negativeofdirection ofwith the LET
the z-axis with ranging
the LET from 0 to 100
ranging MeV·cm
from 0 to 100 MeV·cm2/mg. The ion
have a Gaussian radial distribution with a characteristic
trajectories have a Gaussian radial distribution with a characteristic radius of 20 radius of 20 nm. The sensor
nm. Thein the TCAD
sensor in
simulation is biased to off-state (V g = 0 V) with V bias = 2 V. Meanwhile, the
the TCAD simulation is biased to off-state (Vg = 0 V) with Vbias = 2 V. Meanwhile, the B of 2 T along B of 2 T along the negative
direction
the negativeof z-axis is applied.
direction of z-axis is applied.
After
After heavy ions enter
heavy ions enter thethe sensor,
sensor, aa large
large number
number of of e-h
e-h pairs
pairs are
are generated
generated alongalong the
the particle
particle
track, which are collected by the bias contacts to form the transient current.
track, which are collected by the bias contacts to form the transient current. Moreover, with the Moreover, with the increase
of the LET,
increase of the
the peak
LET, value
the peakof Ibias
value(Ibias
ofmaxIbias) keeps
(Ibias maxincreasing, as shownasinshown
) keeps increasing, Figurein8.Figure
This is8.because
This is
the larger the LET is, the more e-h pairs are generated. So the
because the larger the LET is, the more e-h pairs are generated. So the charge collected charge collected by the bias
bycontacts
the bias
is increased
contacts accordingly.
is increased Moreover,
accordingly. it can beit seen
Moreover, can be from
seenFigure 8 that when
from Figure the LET
8 that when theisLET
lower than
is lower
10 MeV·cm 2 /mg, the I increases faster with the increase of LET. Then, after 10 MeV·cm 2 /mg,
bias max
than 10 MeV·cm2/mg, the Ibias max increases faster with the increase of LET. Then, after 10 MeV·cm2/mg,
and after/mg
the 2
the increasing
increasingtrendtrendis gentler. Through
is gentler. Throughfitting,fitting,
the slopes the ofslopes
the curves before
of the and after
curves before10 MeV·cm 10
2/mg are 3.38 /mg) and 0.077 /mg),
are 3.38 µA/(MeV·cm 2 and 0.077 µA/(MeV·cm 2 respectively.
MeV·cm µA/(MeV·cm 2/mg) µA/(MeV·cm 2/mg), respectively.

30

20 Heavy ion incident

Ibias max (μA)

Silicon film
10
Buried oxide

0 20 40 60 80 100
2
LET (MeV·cm /mg)

Figure 8. The change of the detected maximum bias current (Ibias ) with linear energy transfer (LET).
Figure 8. The change of the detected maximum bias current (Ibias) with linear energy transfer (LET).
The inset shows a schematic diagram of heavy ion incidence.
The inset shows a schematic diagram of heavy ion incidence.
Figure 9 depicts the changes in the Hall voltage with time under different LET ranges. Firstly,
whenFigure
the ions9 depicts
incidentthe changes
with low LETs in (Figure
the Hall9a),
voltage with
the Hall time under
voltage different
will first decreaseLET
andranges. Firstly,
then return to
when the ions incident with low LETs (Figure 9a), the Hall voltage will first decrease
the initial value. Secondly, when the LETs are at medium values (Figure 9b), the Hall voltage will and then return
first
to the initial
decrease, thenvalue. Secondly,
increase to the when
maximum the LETs are
value (Vat medium values (Figure 9b), the Hall voltage will
H max ), and finally decrease to a fixed value. It can
first
be decrease,
observed then
that theincrease to the maximum
VH max increases value (VHofmaxthe
with the increase ), and
LET,finally decrease tends
and gradually to a fixed value. It
to saturation.
can be observed that the V H max increases with the increase of the LET, and gradually tends to
Thirdly, when the LETs are in a higher range, the change of the Hall voltage during the heavy ion
saturation.isThirdly,
incidence basicallywhen the LETs
the same, that are
is, itindecreases
a higher range,
rapidly,the change
then of thetoHall
increases voltage, and
the V during the
finally
H max
heavy ionto
decreases incidence is basically the same, that is, it decreases rapidly, then increases to the VH max, and
a fixed value.
finally decreases to a fixed value.
Sensors 2020, 20, 3946 9 of 12
Sensors 2020, 20, x FOR PEER REVIEW 9 of 12

5 16 16
4
VH (mV)

14

V H max (mV)
3 LET (MeV·cm2/mg) LET increasing
2
0
0.005 12 12

0.01 10
1

VH (mV)
Vbias=2 V B=2 T

0 8
0.0 0.5 1.0 1.5 2.0
Time (μs)
2.5 3.0
8 0 2 4 6 8
LET (MeV·cm2/mg)
LET (MeV·cm2/mg)
(a) 0.1 0.2

16
4 0.4 0.6
0.8 1.0
12
2.0 4.0
VH (mV)

8
LET (MeV·cm /mg) 2 0 6.0 8.0
10 20 Vbias=2 V B=2 T
4 40 60
80 100
0 Vbias=2 V B=2 T 0.0 0.5 1.0 1.5 2.0 2.5 3.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
Time (μs) Time (μs)
(c) (b)
Figure 9. Time
Timevariations
variationsofofthe
theHall
Hallvoltages (VH(V
voltages ) with (a) low,
H ) with (b) medium,
(a) low, and and
(b) medium, (c) high LETs.LETs.
(c) high The
The
insetinset inis(b)
in (b) is change
the the change in maximum
in maximum HallHall voltage
voltage (V)Hwith
(VH max max ) LETs.
with LETs.

3.4. Summary and

3.4. Summary and Discussion
Discussion
During
During or after irradiation,
or after irradiation, the
the performance
performance of the FD-SOI
of the FD-SOI Hall
Hall sensors,
sensors, such
such as
as the
the Hall
Hall voltage,
voltage,
sensitivity, efficiency, and offset voltage, changes are worthy of investigation. Table 3 summarizes
sensitivity, efficiency, and offset voltage, changes are worthy of investigation. Table 3 summarizes the
effects of three
the effects types
of three of radiation
types effects
of radiation on the
effects onFD-SOI sensor
the FD-SOI performance.
sensor performance.

Table 3. Radiation effects on performance of FD-SOI Hall sensor 11.

Table 3. Radiation effects on performance of FD-SOI Hall sensor .
Effect on Performance
Effects Time Changed
Changed Values Effect on Performance
Effects Time Change Ibias VH SA η Voffset
Values Change Ibias VH SA η Voffset
total ionizing dose (TID) After irradiation Total dose ↑ ↑ ↑ ↑ ↓ ↓
total ionizing dose
After irradiation Total dose ↑ ↑ ↑
Transient ↑H
V ↓ ↓
(TID)
transient dose rate (TDR) Dose rate
↑ ↑ &%
single event transient During irradiation Transient VH
LET
(SET)
transient dose rate
1 ↑ and(TDR)
↓ indicate an increase or decrease Dose
in value while &% ratethat the value decreases first and then increases.
indicates
During
↑ ↑ ↘↗
single event transient irradiation
LET
The TID effect mainly explores the changes in the characteristics of the sensor after irradiation.
(SET)
After1 irradiation, a net
↑ and ↓ indicate positiveorcharge
an increase decreasein in
the oxide
value layer
while ↘↗ mainly
indicatesaffects
that thethe off-state
value leakage
decreases current
first and
and on-state bias
then increases. current. Since the sensitivity and efficiency of the sensor are both related to the bias
current, they will also change after irradiation. At the same time, the holes trapped in the oxide layer
affectThe
the TID effect of
resistance mainly explores
the silicon film the changes
of the sensor,inleading
the characteristics
to the reductionof the sensor
of the after
sensor irradiation.
offset voltage.
AfterThe
irradiation,
TDR andaSET net effects
positive chargethe
explore in changes
the oxide oflayer mainly
transient affects
current the off-state
generated leakage
and the current
Hall voltage
and
in theon-state bias
off-state current.
sensor Since
during the sensitivity
irradiation. The and
doseefficiency
rate in the ofTDR
the sensor are both related
effect simulation to the
and the LETbias
in
current,
the they will
SET effect also change
simulation bothafter irradiation.
determine At the same
the irradiation time, deposited
energy the holes trapped in the oxide
on the sensor during layer
the
affect the resistance
irradiation process. Theof the siliconmechanism
physical film of theofsensor, leading
the effect to theirradiation
of different reduction energy
of the sensor
levels on offset
the
voltage.
Hall voltage change will be analyzed next.
The TDR
First, andthe
under SET effects
effect explore
of bias the changes
voltage of transient
and magnetic currenta generated
induction, Hall voltage andisthe Hall voltage
generated and
ainbuilt-in
the off-state sensor
electric field during
is formed irradiation. The dose
in the direction of rate
Hallin the TDRaseffect
contacts, shown simulation
in Figureand 10a.the LET in
Secondly,
the SET
fewer e-heffect
pairssimulation bothbecause
are generated determine thelow-energy
of the irradiation energy
nucleardeposited
explosionsonorthe sensor
heavy ionduring the
incidence.
irradiation
They process.
will move underThethe
physical
action mechanism of the
of the built-in effect
electric of different
field, thereby irradiation
weakeningenergy levelselectric
the built-in on the
Hall voltage
field, change
which will causewill
thebetransient
analyzed next.
Hall voltage to decrease. Finally, when all the generated e-h pairs
First, under
are collected, the the
Halleffect
voltageof bias
returnsvoltage
to theand magnetic
initial value.induction, a Hall voltage is generated and a
built-in electric field is formed in the direction of Hall contacts, as shown in Figure 10a. Secondly,
fewer e-h pairs are generated because of the low-energy nuclear explosions or heavy ion incidence.
 Secondly, because of the high irradiation energy, a mass of e-h pairs are generated, and they mainly
move toward the bias contacts. During this process, electrons and holes move toward the Hall
contacts under the influence of the Lorentz force. At this time, the built-in electric field will be
strengthened, and the transient Hall voltage will increase. Finally, after the irradiation is completed,
a new dynamic balance is reached, and the built-in electric field and Hall voltage are slightly larger
Sensors 2020, 20, 3946 10 of 12
than the initial value.

× B × B
Irradiation or Irradiation or
Heavy ion incidence Heavy ion incidence

Ibias Ibias

Built-in Built-in
electric field electric field
Hole Hole
Electron Electron

(a) (b)
Figure
Figure 10. Illustration
10. Illustration of the
of the physical
physical mechanism
mechanism during(a)
during (a)low-
low-and
and(b)
(b)high-energy
high-energy irradiation
irradiationatat the
the X-Y cross section of the silicon
X-Y cross section of the silicon film. film.

Therefore,
As for under different
the irradiation irradiation
in the case of highenergies,
energy, the two electric
as shown fields 10b,
in Figure (the the
electric
e-h field
pairsformed
generated
by the bias contacts and the built-in electric field formed by the Hall contacts) that dominate the
by irradiation still weaken the built-in electric field first, so the transient Hall voltage first decreases.
movement of the e-h pairs will determine how the transient Hall voltage changes.
Secondly, because of the high irradiation energy, a mass of e-h pairs are generated, and they mainly
move4.toward the bias contacts. During this process, electrons and holes move toward the Hall contacts
Conclusions
under the influence of the Lorentz force. At this time, the built-in electric field will be strengthened,
This work investigated the impacts of the irradiation ionization effects (the TID, TDR, and SET
and the transient Hall voltage will increase. Finally, after the irradiation is completed, a new dynamic
effects) on the performance of the FD-SOI Hall sensors. The sensor FOMs were evaluated in terms of
balance is reached, and the built-in electric field and Hall voltage are slightly larger than the initial ®value.
the Hall voltage, sensitivity, and efficiency via the TCAD simulations based on Synopsys Sentaurus .
Therefore,the
Moreover, under different
physical irradiation
mechanism of theenergies, the two electric
effect of irradiation fields (thewas
on performance electric field formed by
also analyzed.
the bias contacts
For the and
TID the built-in
effect, electric field
the sensitivity formed the
increases, by the Hall contacts)
efficiency thatand
decreases, dominate thevoltage
the offset movement
of thedecreases
e-h pairsbecause
will determine how the transient Hall voltage changes.
of the trapped holes in the oxide layer after irradiation. As for the TDR and SET
effects, the transient Hall voltage changes in the off-state are also different due to either nuclear
4. Conclusions
explosion or heavy ion incidence at different energies. This is mainly due to the difference in the
moving
This work direction of e-h pairs
investigated generatedofunder
the impacts different irradiation
the irradiation ionizationenergy
effectsunder the bias
(the TID, TDR,electric
and SET
field and built-in electric field.
effects) on the performance of the FD-SOI Hall sensors. The sensor FOMs were evaluated in terms of
To the authors’ best knowledge, these results comprehensively summarized the response of the
the Hall voltage, sensitivity, and efficiency via the TCAD simulations based on Synopsys Sentaurus® .
FD-SOI Hall sensors in terms of irradiation ionization effect for the first time. These results provide
Moreover, the physical
deep insight into themechanism of the effect
design, processing, of irradiation
and testing on performance
of the FD-SOI Hall sensorswas also
used inanalyzed.
radiation
For the TID
environments. effect, the sensitivity increases, the efficiency decreases, and the offset voltage decreases
because of the trapped holes in the oxide layer after irradiation. As for the TDR and SET effects,
Author Contributions:
the transient Hall voltage Funding acquisition,
changes J.B.; methodology,
in the off-state are alsoL.F., J.B., and
different G.Y.;
due tosoftware, L.F., J.B., explosion
either nuclear and K.X.; or
heavyinvestigation,
ion incidence L.F. at
and J.B.; writing—original
different energies. Thisdraft preparation,
is mainly due toL.F.;
the writing—review
difference in theand editing,direction
moving J.B.;
supervision, J.B.. All authors have read and agreed to the published version of the manuscript.
of e-h pairs generated under different irradiation energy under the bias electric field and built-in
electric field.
To the authors’ best knowledge, these results comprehensively summarized the response of
the FD-SOI Hall sensors in terms of irradiation ionization effect for the first time. These results
provide deep insight into the design, processing, and testing of the FD-SOI Hall sensors used in
radiation environments.

Author Contributions: Funding acquisition, J.B.; methodology, L.F., J.B. and G.Y.; software, L.F., J.B. and K.X.;
investigation, L.F. and J.B.; writing—original draft preparation, L.F.; writing—review and editing, J.B.; supervision,
J.B. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the Key R&D Plan of Guangdong Province (2019B010145001), the National
Natural Science Foundation of China (NSFC) (61634008 and 61704188), and the Youth Innovation Promotion
Association CAS (2014101).
Acknowledgments: The software cited in this document is furnished under a license from Synopsis International
Limited. Synopsis and Synopsis product names described herein are trademarks of Synopsis, Inc. (Mountain View,
CA, USA). The authors would like to thank all the participants in this study.
Conflicts of Interest: The authors declare no conflict of interest.
Sensors 2020, 20, 3946 11 of 12

Appendix A
This appendix briefly summarizes the previous work of [8] and focuses on the selection of
the optimal sensor structure. Compared with the bulk structure, the characteristics of the FD-SOI
Hall sensors (such as sensitivity and efficiency) are improved because of its thin thickness and low
doping concentration.
If the FD-SOI structure is to be obtained, the thickness of the silicon film needs to be less than twice
the thickness of the maximum depletion layer (Xdmax ) under the gate oxide. The Xdmax is determined
as follows: s
4εsi ϕF
Xdmax = (A1)
qNA

where NA is the doping concentration of the silicon film, εsi is the dielectric constant of silicon, and ϕF
is the Fermi potential. Therefore, the silicon film thickness of 50 nm matches the doping concentration
of 1016 cm−3 , which can make the sensor reach FD-state.
In addition to FD-state, other structural features also play an important role in improving the
characteristics of the sensor. High-symmetry square sensors demonstrate the best efficiency, which is
the best trade-off between sensitivity and power consumption. The shorter length (1 µm) of the Hall
contacts and the width of the 0.5× W bias contacts can improve the sensitivity of the sensor.

References
1. Ramsden, E. Hall-Effect Sensors–Theory and Applications, 2nd ed.; Elsevier: Amsterdam, The Netherlands, 2006.
2. Ripka, P. Magnetic sensors for industrial and field applications. Sens. Actuators A Phys. 1994, 42, 394–397.
[CrossRef]
3. Boero, G.; Demierre, M.; Besse, P.A.; Popovic, R.S. Micro-Hall devices: Performance, technologies and
applications. Sens. Actuators A Phys. 2003, 106, 314–320. [CrossRef]
4. Ripka, P.; Janošek, M. Advances in magnetic field sensors. IEEE Sens. J. 2010, 10, 1108–1116. [CrossRef]
5. Vitale, S.A.; Wyatt, P.W.; Hecka, N.; Kedzierski, J.; Keast, C.L. FDSOI process technology for
subthreshold-operation ultralow-power electronics. Proc. IEEE 2010, 98, 333–342. [CrossRef]
6. Dolgyi, L.N.; Lovshenko, I.Y.; Nelayev, V.V. Technology and electric characteristics of the field-effect hall
sensor based on SOI structure. Russ. Microelectron. 2013, 42, 368–372. [CrossRef]
7. Paun, M.A. Three-dimensional simulations in optimal performance trial between two types of hall sensors
fabrication technologies. J. Magn. Magn. Mater. 2015, 391, 122–128. [CrossRef]
8. Fan, L.; Bi, J.; Xi, K.; Majumdar, S.; Li, B. Performance Optimization of FD-SOI Hall Sensors Via 3D TCAD
Simulations. Sensors 2020, 20, 2751. [CrossRef]
9. Chen, P. Radiation Effects of Semiconductor Devices and Integrated Circuits, 1st ed.; National Defense Industry
Press: Beijing, China, 2005.
10. Zhao, X.; Mei, B.; Bi, J.S.; Zheng, Z.S.; Gao, L.C.; Zeng, C.B.; Luo, J.J.; Yu, F.; Han, Z.S. Single event transients
in a 0.18 µm partially-depleted silicon-on-insulator complementary metal oxide semiconductor circuit.
Acta Phys. Sin. 2015, 64, 136102.
11. Bi, J.S.; Zeng, C.B.; Gao, L.C.; Liu, G.; Luo, J.J.; Han, Z.S. Estimation of pulsed laser-induced single event
transient in a partially depleted silicon-on-insulator 0.18-µm MOSFET. Chin. Phys. B 2014, 23, 088505.
[CrossRef]
12. Xue, H.; Zhang, M.; Yang, G. Numeric simulation of NMOS transient dose rate effect based on sub-micro
SOI. Appl. Electron. Technol. 2019, 45, 59–62.
13. Oh, S.J.; Le, T.T.; Kim, G.W.; Kim, C. Size effect on NiFe/Cu/NiFe/IrMn spin-valve structure for an array of
PHR sensor element. Phys. Status Solidi 2007, 204, 4075–4078. [CrossRef]
14. Karlova, G.F.; Gradoboev, A.V. Radiation characteristics of hall-sensor on the base of gallium arsenide.
In Proceedings of the 2013 International Siberian Conference on Control and Communications (SIBCON),
Krasnoyarsk, Russia, 12–13 September 2013; pp. 1–3.
15. Gradoboev, A.V.; Karlova, G.F. Stability of the GaAs based Hall sensors irradiated by gamma quanta.
IOP Conf. Ser. Mater. Sci. Eng. 2015, 81, 012027. [CrossRef]
Sensors 2020, 20, 3946 12 of 12

16. Abderrahmane, A.; Tashiro, T.; Takahashi, H.; Ko, P.J.; Okada, H.; Sato, S.; Ohshima, T.; Sandhu, A. Partial
recovery of the magnetoelectrical properties of AlGaN/GaN-based micro-Hall sensors irradiated with protons.
Appl. Phys. Lett. 2014, 104, 023508. [CrossRef]
17. Boudinov, H.; Kucheyev, S.O.; Williams, J.S.; Jagadish, C.; Li, G. Electrical isolation of GaN by MeV ion
irradiation. Appl. Phys. Lett. 2001, 78, 943–945. [CrossRef]
18. Hristoforou, E.; Bolshakova, I.; Holyaka, R.; Duran, I.; Stockel, J.; Viererbl, L.; Vayakis, G. Sensors and
instrumentation for the magnetic field measurement under high neutron fluences. In Proceedings of the
IEEE Sensors, Vienna, Austria, 24–27 October 2004; Volume 3, pp. 1075–1077.
19. Bolshakova, I.; Hristoforou, E. Direct Hall measurements of InSb thin films during their irradiation with fast
neutrons. Sens. Actuators A Phys. 2006, 129, 192–196. [CrossRef]
20. Popovic, R.S. Hall Effect Devices, 2nd ed.; CRC Press: Abingdon, UK, 2004.
21. Crescentini, M.; Biondi, M.; Romani, A.; Tartagni, M.; Sangiorgi, E. Optimum design rules for CMOS hall
sensors. Sensors 2017, 17, 765. [CrossRef]
22. Sanchez Esqueda, I.; Barnaby, H.J.; King, M.P. Compact modeling of total ionizing dose and aging effects in
MOS technologies. IEEE Trans. Nucl. Sci. 2015, 62, 1501–1515. [CrossRef]
23. Lee, M.S.; Lee, H.C. Dummy gate-assisted n-MOSFET layout for a radiation-tolerant integrated circuit.
IEEE Trans. Nucl. Sci. 2013, 60, 3084–3091. [CrossRef]
24. Seo, D.; Trang, L.D.; Han, J.W.; Kim, J.; Lee, S.; Chang, I.J. Total Ionizing Dose Effect on Ring Oscillator
Frequency in 28-nm FD-SOI Technology. IEEE Electr. Dev. Lett. 2018, 39, 1728–1731. [CrossRef]
25. Lalumondiere, S.D.; Dillingham, E.C.; Scofield, A.C.; Bonsall, J.P.; Karuza, P.; Brewe, D.L.; Schrimpf, R.D.;
Sternberg, A.L.; Wells, N.P.; Cardoza, D.M.; et al. Application of a Focused, Pulsed X-ray Beam for Total
Ionizing Dose Testing of Bipolar Linear Integrated Circuits. IEEE Trans. Nucl. Sci. 2018, 65, 478–485.
[CrossRef]
26. Le Roch, A.; Virmontois, C.; Paillet, P.; Warner, J.H.; Belloir, J.-M.; Magnan, P.; Goiffon, V. Comparison of
X-Ray and Electron Radiation Effects on Dark Current Non-Uniformity and Fluctuations in CMOS Image
Sensors. IEEE Trans. Nucl. Sci. 2020, 67, 268–277. [CrossRef]
27. Bi, J.S.; Liu, G.; Luo, J.J.; Han, Z.S. Numerical simulation of single-event-transient effects on ultra-thin-body
fully-depleted silicon-on-insulator transistor based on 22 nm process node. Acta Phys. Sin. 2013, 62, 208501.
28. Bi, J.; Chen, L.; Han, Z.; Wang, Y.; Liu, M. Body Bias Effects on the Single-Event-Transient Response of PDSOI
Devices. In Proceedings of the 2015 15th European Conference on Radiation and Its Effects on Components
and Systems (RADECS), Moscow, Russia, 14–18 September 2015; pp. 1–4.
29. Synopsys. Sentaurus Device User Guide; Synopsys: Mountain View, CA, USA, 2010.
30. Chen, W.; Du, L.; He, L.; Zhang, T.; Zhang, X.; Zhuang, Y.; Bao, J. A model considering the ionizing radiation
effects in MOS structure. Acta Phys. Sin. 2009, 58, 4090–4095.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).