ArticlePDF Available

Analog and Mixed Signal Circuit Design Techniques in Flexible Unipolar a-IGZO TFT Technology: Challenges and Recent Trends

Authors:

Abstract and Figures

Advancements in the field of flexible electronics have enabled many novel applications such as wearables, flexible large-area displays, and textile-based sensor systems. A widely used semiconductor material for flexible electronics is amorphous indium gallium zinc oxide (a-IGZO). When compared to other semiconductors suitable for flexible technologies, this material is attractive for its higher mobility, better bias stability and improved uniformity. This paper presents an overview of current trends in analog and mixed signal circuit design and in architectures for a-IGZO TFT based sensor systems. We highlight the specific design challenges associated with the main sub-blocks found in sensors and bio-signal acquisition systems based on a-IGZO TFT technologies: front-end amplifiers and data converters. We then present an overview of several state-of-the-art sensor systems based on a-IGZO TFTs and their applications. Finally, conclusions are drawn, and a roadmap for the future development of analog and mixed signal circuit design in a-IGZO TFT is presented.
Content may be subject to copyright.
Received 21 June 2021; revised 17 September 2021; accepted 9 October 2021. Date of current version 24 November 2021.
Digital Object Identifier 10.1109/OJCAS.2021.3123206
Analog and Mixed Signal Circuit Design Techniques
in Flexible Unipolar a-IGZO TFT Technology:
Challenges and Recent Trends
MOHAMMAD ZULQARNAIN (Member, IEEE), AND EUGENIO CANTATORE (Fellow, IEEE)
Department of Electrical Engineering, Eindhoven University of Technology, 5600 MB Eindhoven, The Netherlands
This article was recommended by Associate Editor X. Guo.
CORRESPONDING AUTHOR: M. ZULQARNAIN (e-mail: m.zulqarnain@tue.nl)
This work was supported by the UEBIT Project in TUe/UMC Alliance.
ABSTRACT Advancements in the field of flexible electronics have enabled many novel applications such
as wearables, flexible large-area displays, and textile-based sensor systems. A widely used semiconductor
material for flexible electronics is amorphous indium gallium zinc oxide (a-IGZO). When compared to
other semiconductors suitable for flexible technologies, this material is attractive for its higher mobility,
better bias stability and improved uniformity. This paper presents an overview of current trends in analog
and mixed signal circuit design and in architectures for a-IGZO TFT based sensor systems. We highlight the
specific design challenges associated with the main sub-blocks found in sensors and bio-signal acquisition
systems based on a-IGZO TFT technologies: front-end amplifiers and data converters. We then present an
overview of several state-of-the-art sensor systems based on a-IGZO TFTs and their applications. Finally,
conclusions are drawn, and a roadmap for the future development of analog and mixed signal circuit
design in a-IGZO TFT is presented.
INDEX TERMS a-IGZO TFT, analog circuits, mixed signal circuits, flexible electronics, flexible wearable
systems.
I. INTRODUCTION
THE FIELD of electronics has witnessed a transformation
phase in the recent years, due to the rapid growth of
wearables and ubiquitous sensors and the advent of Internet
of Things (IoT). Flexible electronics is increasingly recog-
nized as a key enabler for novel sensor applications due to
its low-cost, light weight, ultra-thin and large-area form fac-
tor, conformability and mechanical flexibility. The inherent
flexible nature of the substrates in these technologies can
be leveraged in applications requiring conformability, bend-
ability, foldability and stretchability. A typical application
example that can strongly benefit from flexible electronics
is the domain of flexible and conformable wearable devices
for wellness and healthcare monitoring.
Amorphous indium gallium zinc oxide (a-IGZO) [1]
is considered to be one of the best options for imple-
menting flexible electronic systems due to its compara-
tively higher mobility, large area uniformity, compatibility
with low temperature processing and good bias stability,
among flexible devices. The stability of a-IGZO TFTs is
much greater than that of hydrogenated amorphous sili-
con (a-Si:H) and organic TFTs and is comparable to that
of polycrystalline silicon (poly-Si) TFTs [2]. Moreover,
a-IGZO devices have shown bendability upto 25 µm radius
without damage [3]–[5] and can be stretched more than
200% [6]. Additionally, TFT scaling is possible in IGZO
due to its amorphous nature [7], as grain boundaries
will not interplay with the dimensions of the transis-
tor and hence, TFTs can be scaled down to very short
dimensions. a-IGZO TFT technology has also demonstrated
high yield and integration capability, e.g., compared to
organic TFTs. Indeed, integration of more than 50,000
devices (n-type TFTs and resistors) has been reported [8].
Thus, the a-IGZO TFT technology represents a good com-
promise between performance, technology reliability and
cost.
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME 2, 2021 743
ZULQARNAIN AND CANTATORE: ANALOG AND MIXED SIGNAL CIRCUIT DESIGN TECHNIQUES
However, when compared to mainstream silicon
electronics, the physical properties of flexible a-IGZO TFTs
still give rise to many challenges at circuit and system level,
which can be summarized as follows:
1) Large flicker noise and high noise corner frequency [9]
compared to silicon;
2) Lower power efficiency than silicon due to the lower
mobility;
3) Parameter variability;
4) Unipolar devices with only n-type TFTs available;
5) Difficulty in interfacing with silicon-based circuits.
The large flicker noise and high noise corner frequency
makes chopping less effective and increase low-frequency
noise in amplifiers. The low intrinsic gain and the unavail-
ability of complementary devices make high gain amplifiers
difficult to realize and result in reduced accuracy when using
negative feedback to overcome variability issues. These and
other issues translate into the necessity of unconventional
circuit and system design approaches for a-IGZO TFT cir-
cuits, as the well-known design techniques developed for
complementary circuits cannot be applied. Hence, a circuit
design approach considering specific technology constraints
is mandatory.
There has been a strong research and industrial effort
in a-IGZO TFT technology applications related to large
area displays [10]–[13], flexible X-ray imagers [14]–[16],
touchscreen tags [17]–[19], Radio Frequency Identification
(RFID) [20], [21], Near Field Communication (NFC)
tags [22]–[25], machine learning processing engines [26]
and flexible microprocessors [8] in the last few years. This
research was mainly focused on circuits and systems where
the TFTs are used as switches, due to various challenges
observed when trying to implement analog functions.
Recent advancements in processing technology have
resulted in improved a-IGZO TFT performance, enabling
more demanding applications [7]. As a consequence, novel
applications can be envisioned using this technology, such
as: physiological monitoring by wearable flexible health
patches [27]–[30], temperature sensors [31], [32], display-
integrated fingerprint sensors [33]–[35] and piezoelectric
pressure sensing [36], [37]. However, analog and mixed
signal circuit design in a-IGZO TFT technologies has not
reached maturity yet.
Thus, in this paper we will present an overview of chal-
lenges and the recent trends in the design of a-IGZO TFT
circuits. Several techniques that have been experimentally
validated to improve analog and mixed signal designs are
presented. In Section II, an introduction of a-IGZO TFT
technology is provided. In Section III, we explain how
technology limitations translate to challenges at the cir-
cuit and system design level when implementing analog
and mixed-signal functions. Some of the state-of-the-art
design approaches for a-IGZO TFT front-end amplifiers are
presented in Section V. Furthermore, data conversion tech-
niques recently implemented in a-IGZO TFT technology are
FIGURE 1. A simple bottom-gate a-IGZO TFT stack.
discussed in Section VI. Section VII provides an overview
of some of the sensor systems demonstrated using a-IGZO
TFTs and their applications. Finally, we draw some conclu-
sions and outline some future research directions in the field
of analog and mixed-signal circuit design with a-IGZO TFT
technology.
II. A-IGZO TFT TECHNOLOGY
Fig. 1 shows the cross section of a typical single-gate a-
IGZO TFT [38], [39]. It is based on Etch Stop Layer (ESL)
approach. The device is fabricated by depositing different
layers on top of a flexible substrate. The stack consists of
three metal layers, an a-IGZO layer, dielectric layers (gate,
ESL) and a passivation layer (interlayer). The maximum
processing temperature is 350C. a-IGZO TFTs typically
show charge carrier mobility (µ) of 15-20 cm2/V.s, while
the threshold voltage (VT) varies between 0-2 V. Better
speed performance compared to ESL TFT technology can
be achieved by using self-aligned stacks, where the overlaps
between gate, source and drain, and the parasitic capacitance
that they cause, are minimized [12], [40]–[42]. The process-
ing temperature of a-IGZO TFTs typically ranges between
150C and 350C. However, room temperature fabrication
of a-IGZO TFT has also been reported [1], [43].
The device architecture of a-IGZO TFTs is an active field
of research with many stack variants reported [44]. The device
stack can either be bottom gate [45] or top gate [46], depending
on deposition of gate electrode with respect to semiconducting
layer. In the case of bottom gate devices, the gate electrode
is below the semiconducting layer, while in top gate devices,
the gate electrode is above the semiconducting layer. The
TFT stack could further be categorized in coplanar [40] or
staggered [47] topologies depending on source/drain contacts
position with respect to the accumulated channel. In coplanar
TFT stacks, source/drain contacts are on the same plane as the
channel, while in staggered TFT stacks, source/drain contacts
are on the opposite side of the channel. The performance of
TFT device can be improved using double gate [48], [49] archi-
tectures. Indeed, the presence of an extra gate helps to control
the channel more effectively, increasing the transconductance
when front and back gate are shorted together. Alternatively,
the second gate can be used to control via electric biasing
the TFT threshold, a feature that can be extremely useful to
744 VOLUME 2, 2021
improve circuit behavior. To reduce channel length, non-planar
TFT stack, such as such as vertical TFTs (VTFTs) or quasi-
vertical TFTs (QVTFTs) have also been reported [50]–[54].
The use of different materials in the TFT stack and their
effects on the electrical and mechanical performance of the
device, along with the stability have been extensively inves-
tigated [55], [56]. The choice of the gate dielectric is an
important factor for the performance of a TFT. The quality of
the interface between gate insulator and channel has indeed
a strong influence on the stability and the carrier mobility of
the TFT, by controlling the charge trap density [55]. The most
commonly used gate dielectric is aluminium oxide (Al2O3),
which has comparatively high relative permittivity of 9.5, and
provides a high-quality interface with a-IGZO [55]. Besides
Al2O3, hafnium oxide (HfOx), titanium oxide (TiOx), silicon
oxide (SiO2) and silicon nitride (SiNx) are also used as gate
dielectrics.
TFT devices with high performance in terms of mobil-
ity and stability have been reported [57]–[59]. The typical
supply voltage of a-IGZO TFTs ranges between 5-30 V,
however recently a-IGZO TFTs operated at 1 V or less have
been described [60], [61]. The nominal transition frequency
(fT) of these devices lies in the MHz regime [62]. However,
scaling of channel length below the micrometer regime and
minimizing the overlap capacitances has made possible to
reach fTin the range of GHz [63], [64]. Recent reports have
documented further scaling of IGZO TFTs channel lengths
down to 160 nm [65] and 32 nm [66].
III. CIRCUIT AND SYSTEM CHALLENGES
The physical properties of a-IGZO TFTs and their processing
give rise to many challenges at circuit and system level,
which are summarized here below.
A. LOW FREQUENCY NOISE
The low-quality interface between dielectric and semicon-
ductor, together with mobility fluctuation results in large
low-frequency noise (1/f or flicker noise) in thin film tran-
sistors [9]. This 1/f noise component is the dominant noise
contribution for bio-signal acquisition front-ends owing to
the low frequency content of biomedical signals. Hence,
low-frequency noise mitigation is the foremost challenge
in designing bio-signal acquisition front-ends. Flicker noise
can be suppressed by using chopping [67], but at the cost
of larger bandwidth requirements. It can be decreased by
increasing device area as well, which increases the input
parasitic capacitance, reducing the input impedance. Discrete
time circuit techniques such as auto-zeroing and corre-
lated double sampling can also be exploited to mitigate 1/f
noise [67]. However, such techniques require large capacitors
which are difficult to integrate in a-IGZO TFT technology
with acceptable yield, at the state-of-the-art.
B. POWER EFFICIENCY
One of the main limitations of a-IGZO TFTs is their mobil-
ity, which is one to two order of magnitude smaller than
mainstream silicon field effect transistors. Low mobility
results in low transconductance (gm) of the devices, which
consequently limits the small-signal intrinsic gain (gmro)
and fT. Moreover, the large oxide thickness in a-IGZO TFTs
results in limited gate electrostatic control over the chan-
nel, which ultimately leads to larger supply voltages, even
lower transconductance and more pronounced channel length
modulation. Together with the considerable parasitic capac-
itances observed in non-self aligned TFTs, from the circuit
design perspective, this results in low gain-bandwidth prod-
uct (GBW) and low efficiency in terms of speed versus
needed power.
C. VARIABILITY
The low process temperature (compared to silicon fabrica-
tion), which is the enabler of mechanical flexibility, and
the low-cost manufacturing of a-IGZO TFTs, result in large
parameter variability and in poor intrinsic matching between
identical neighboring devices. This has a particularly strong
impact on analog circuit techniques and analog to digi-
tal conversion, which typically take advantage of the good
matching offered by silicon technologies. If high accuracy
a-IGZO TFT circuitry is required in an application, then
system techniques like calibration might need to be applied.
D. UNIPOLAR NATURE OF DEVICES
Complementing n-type a-IGZO TFTs with another material
p-type TFTs on flexible substrates is quite challenging due to
diffi ˛Aculties with hybrid integration [68]–[71]. The unipolar
nature of the devices in a-IGZO TFT technology makes it
difficult to exploit commonly used CMOS circuit techniques.
Consequently, active loads with simple configuration and low
transistor count are not available in unipolar technologies. The
low intrinsic gain and the unavailability of complementary
devices limit the achievable gain with unipolar a-IGZO TFT
based amplifiers, which makes feedback architectures less
useful.
E. INTERFACES WITH SILICON-BASED CIRCUITS
Circuits consisting of a-IGZO TFTs generally require a
high supply voltage due to thicker dielectrics, while state-
of-the-art silicon-based integrated circuits are operated at
supply voltages below 1 V. This inherent difference in supply
voltage domains makes interfacing of flexible circuits with
silicon based circuits in hybrid systems quite challenging.
Recent advances in technology have resulted in some
fruitful results in the direction of low voltage flexible tech-
nologies. Recently, TFTs operating at 1-V or lower supply
voltages have been reported [60], [61]. However, complex
circuits built with such devices have still not been demon-
strated. Some research groups have reported IGZO TFTs [72]
which achieve low voltage operation (< 2 V) using an
electrolyte as gate dielectric. The low voltage operation of
electrolyte-gated TFTs is due to the large capacitance pro-
vided by the electrolyte [73]. However, electrolyte-gated
VOLUME 2, 2021 745
ZULQARNAIN AND CANTATORE: ANALOG AND MIXED SIGNAL CIRCUIT DESIGN TECHNIQUES
FIGURE 2. General architecture of a sensor acquisition system.
FIGURE 3. Different types of state-of-the-art a-IGZO TFT amplifiers. (a) Pseudo CMOS load amplifier [74], (b) Diode connected load amplifier, (c) Single ended positive feedback
common source amplifier, (d) Differential positive feedback amplifier [31], (e) Bootstrap load amplifier [75], (f) Modified diode load amplifier [76].
TFTs have large parasitic capacitances, which results in
reduced device speed.
IV. ARCHITECTURE OF SENSOR ACQUISITION
SYSTEMS
A typical implementation of a sensor acquisition system is
shown in Fig. 2. The full acquisition chain comprises a sensor
or electrode, an analog front-end (AFE), an analog to digital
converter (ADC), a digital signal processing and transmission
block, and a reader. The external analog quantity is picked by
an electrode, or converted to the electric domain by a sensor.
The sensor or electrode output is generally small, hence,
amplification through an AFE is required. Conversion to
digital representation is essential before transmitting signals
to a reader device (e.g., smart phone), to increase noise
immunity during transmission. This digital conversion and
transmission process is performed by an ADC, digital signal
processing and a transmission block.
The AFE is the most crucial analog building block, which
is inevitably present in almost all sensor systems. The
next section deals with the design techniques of front-end
amplifiers based on a-IGZO TFTs.
V. A-IGZO TFT BASED AMPLIFIERS
As mentioned in Section III, a-IGZO TFTs are only n-type.
Different types of amplifiers depending on load configu-
rations have been reported in a-IGZO TFTs. They can be
broadly classified considering their load topologies among
amplifiers in which the load mimimcs a p-type device
behavior (often called pseudo-CMOS amplifiers) and ampli-
fiers that exploit an n-type load (often supported by some
output-impedance enhancement technique).
Fig. 3(a) shows the circuit schematic of a pseudo CMOS
amplifier [74]. It consists of a pseudo-CMOS load, comprising
3 TFTs as shown in the enclosed box. The basic idea in this
topology is to put an inverter in front of an n-type device
746 VOLUME 2, 2021
TABLE 1. Comparison of the state-of-the-art a-IGZO TFT amplifiers.
to obtain a transistor that behaves in a similar way to a p-
type device. Although this solution tries to circumvent the
unavailability of p-type TFTs, it requires very careful transistor
sizing, exploiting accurate models, to enable stable operation.
The most conventional n-type load amplifier topology in
a-IGZO TFT technology is the diode connected load ampli-
fier [27], [28], [32], [75], [77]. It is popular due to its
minimalistic design. A differential implementation of this
topology is shown in Fig. 3(b). As the small-signal gain is
given by a ratio of transconductances of TFTs (gm1
gm3),this
topology in general, is less sensitive to parameter variability.
In some applications, the gain of one stage may not be suffi-
cient and hence cascading two stages may be required [27],
[28], [75]. The availability of a dual gate a-IGZO TFT tech-
nology may enable further variations of the diode connected
load amplifier topology [32]. The second gate of the device
may be used to improve performance [81], e.g., the second
gate can be connected to the source of the TFT to increase
the output resistance. For input driver TFTs, the gates can
be shorted together to improve the transconductance.
Another very common circuit technique which is
employed in a-IGZO TFT amplifiers [29], [31], [78], [79],
[82] is the use of a local positive feedback, shown in
Fig. 3(c). The small-signal gain of this topology can be
written as:
G=gm1
gm2(1A)(1)
If we design Asuch that |A|<1, then, the gain of this
amplifier is enhanced by the factor 1
1Acompared to a diode
connected load amplifier.
A TFT implementation of this gain enhancement technique
is schematically presented in Fig. 3(d), for a fully differential
amplifier [31]. The auxiliary amplifier consisting of TFTs
M58and M10 is used to provide the gain A: this reduces
the vgs of the load TFTs M34and thus boosts the output
impedance, increasing the gain. Ensuring the stability of the
amplifier in presence of variations is crucial for this type of
design in a-IGZO TFT technology, and is a design concern
when using auxiliary amplifiers in positive feedback to boost
the gain.
Another variant of the gain enhancement technique shown
in Fig. 3(b) is displayed in Fig. 3(e), which is also known
as a bootstrapped load amplifier [75], [80]. It uses a load
comprising M3and M5with a bootstrapping capacitor CB.
The peculiarity of this load configuration is that, at low
frequencies, M3acts as a diode-connected load offering a
smaller gain, while at high frequencies, the capacitor CB
shorts the gate of M3with its source, providing a zero-VGS
load, and consequently higher gain. According to this sim-
ple analysis, this amplifier has a band-pass response, which
is especially suited to biomedical applications, where the
inherent DC electrode offset needs to be rejected. The band-
pass response is also beneficial to avoid amplifier saturation
due to mismatch and thus excessive offset. One major dis-
advantage of this amplifier topology is that it requires large
bootstrapping capacitors CB, and hence large area, due to
two reasons. The first reason is that a large CBis required
to realize low high-pass frequencies. Secondly, the capacitor
CBand the parasitic capacitance from the gate of M3to
ground form a capacitive divider. Hence, CBshould have a
very small impedance, to provide an effective zero-VGS load,
i.e., to effectively short the gate and source of M3.
Fig. 3(f) shows another n-type load amplifier design which
is based on load enhancement and a positive feedback tech-
niques [76]. The load configuration consists of the diodes
M3,M
5and M7with top gates of diodes M5and M7con-
nected to the source of M3. This specific arrangement of load
diodes increases the output resistance of M5, consequently
increasing the DC gain. The output resistance is given indeed
by the expression:
Rout =1
gm7+1
gm5(1+η) +1
gm3(1+η)2(2)
where, ηcorresponds to the threshold modulation coefficient
due to the second gate biasing, and gm3,gm5,gm7are the
transconductances of M3,M
5and M7respectively. In the
same amplifier, a partial positive feedback is also imple-
mented by cross coupling the top gates of the input driver
TFTs M1and M2. This enhances both transconductance and
gain. A buffer with a source follower configuration (M16
and M18) is also implemented at the output. Although, this
peculiar arrangement of the load diode TFTs increases the
gain, it also increases the sensitivity to mismatch and varia-
tions. This amplifier is also susceptible to input offset due to
large gain at DC, which may result in amplifier saturation.
Another drawback of this amplifier topology is its limited
output swing due to large stack of TFTs.
Although, a-IGZO TFT based amplifiers are generally
used in low frequency applications, there have been some
reports, where large GBW amplifiers implemented with
TFTs having few MHz fThave been demonstrated [83]–[88].
An overview of some state-of-the-art amplifiers in a-IGZO
TFT technologies which compares several performance
indicators is shown in Table 1.
VOLUME 2, 2021 747
ZULQARNAIN AND CANTATORE: ANALOG AND MIXED SIGNAL CIRCUIT DESIGN TECHNIQUES
TABLE 2. Different types of active load topologies for a-IGZO TFT based amplifiers.
(b)
(c)
(a)
Logic
AFE
Serial
Digital
Output
Vin
Counter
Tref
PWM Waveform
Reset Integrator
ADC
References
Sensor
read-in
Digital
control
Comparator
DC bias
Out Clock Vbias
Sensor
VDD
VSS
DAC
Analog to Digital Converter
FIGURE 4. a-IGZO TFT based data converter architectures. (a) Asynchronous delta sigma modulator [31], (b) Reset integrator based ADC [27], (c) C-2C SAR ADC [32]. (a)
Copyright 2017 IEEE. Reprinted with permission from [31].
Table 2 summarizes various active load topologies using
n-type devices in a-IGZO TFT technology. It also com-
pares their output impedance (Zout), the number of elements
required to realize the active load, and the main pros and
cons of each circuit block.
VI. DATA CONVERTERS BASED ON A-IGZO TFT
TECHNOLOGY
The next block in the signal acquisition chain (Fig. 2) after
the amplifier is the data converter. It is required to convert
the analog signal at amplifier’s output to a digital (multi-bit
discrete time) or binary (two-level pulse-width modulated)
representation, which ensure robustness against interferers
during transmission. As discussed in Section IV, in the for-
mer case one typically speaks of an ADC. Fig. 4 shows some
of the data converter architectures reported in a-IGZO TFT
technology. The important sub-blocks of these architectures
are further illustrated in Fig. 5.
One of the first reported data converters [31] in a-IGZO
TFT technology is shown in Fig. 4(a). It has an asyn-
chronous delta sigma modulator (ADSM) architecture. Thus,
it converts an input current signal to an output pulse-width
748 VOLUME 2, 2021
FIGURE 5. Implementation details of a-IGZO TFT based data converter architectures. (a) Schematic of the comparator used in ADSM [31], (b) Schematic of the reset
integrator [27], (c) Timing diagram of the reset integrator based ADC [27], (d) Schematic of the comparator used in C-2C SAR ADC with offset cancellation [32], (e) Schematic of
the C-2C DAC in SAR ADC [32]. (a) Copyright 2017 IEEE. Reprinted with permission from [31], (d-e) Copyright 2018 IEEE. Reprinted with permission from [32].
modulated (PWM) signal. It comprises a loop filter, a pre-
amplifier, a comparator with tunable hysteresis, and a current
steering digital to analog converter (DAC). Thanks to the
loop filter and feedback arrangement, the time average of
the output binary signal equals the input current. This data
converter is not time sampled and thus does not introduce
a quantization error. Besides, it is built using an oscilla-
tor whose frequency is modulated by the input current,
and this ensures an inherent oversampling of the input sig-
nal, which improves accuracy in the transformation to the
time domain [89]. The amplifier used as core of the loop
filter (A1) is already shown in Fig. 3(c). The filter is com-
pleted with feedback capacitors to realize a current integrator
(Fig. 4(a)). The comparator preamplifier (A2) is a diode con-
nected load amplifier. The top gates of the input pair are
connected to the output of the opposite branches, realizing
a limited positive feedback that increases the transconduc-
tance and hence enhances gain. The pre-amplifier is used to
boost the loop filter’s limited output swing. Fig. 5(a) shows
the schematic of the comparator. The top gates of the input
TFTs M1and M2are cross coupled through level shifters
M5,M
8and M6,M
9to create a positive feedback with loop
gain larger than one. The desired hysteresis of the compara-
tor, which is caused by the positive feedback, can be tuned
by the bias voltage VH. The 1-bit DAC consists of the ref-
erence current IREF and the switches S14. The ADSM can
also been considered as an asynchronous oscillator whose
frequency is modified by the input current.
The ADSM experimentally demonstrates a maximum sig-
nal to noise ratio (SNR) of 55 dB, spurious free dynamic
VOLUME 2, 2021 749
ZULQARNAIN AND CANTATORE: ANALOG AND MIXED SIGNAL CIRCUIT DESIGN TECHNIQUES
range (SFDR) of 55 dB and signal to noise-plus-distortion
ratio (SNDR) of 50 dB in 10-Hz bandwidth. A maximum
SNR of 44 dB, SNDR of 40 dB and SFDR of 55 dB,
are obtained in 300-Hz bandwidth. The performance of this
ADSM is the state-of-the-art among a-IGZO TFT technol-
ogy based data converters. The power dissipation of the
full system is 2 mW. The area occupied by the ADSM is
27.9mm2.
Another way to convert a signal to PWM representation is
the use of a reset integrator. It has the disadvantage of losing
the inherent oversampling due to the self-oscillation of the
ADSM, but it has the advantage of being far simpler, more
compact and potentially lower power. Fig. 4 (b) shows the block
diagram of a reset integrator based ADC [27]. It consists of a
reset integrator, a logic block and a counter. The reset integrator
is used to convert the input signal to a binary PWM signal,
i.e., to a time-domain representation. The rationale behind
the time domain processing is that a-IGZO TFT technology
does not have stable voltage references which are required
in most of the data converter architectures, while processing
signals in time domain requires a time reference which can be
easily obtained by suitable division of 13.56-MHz NFC (or
HF RFID) carrier. Indeed, in this work [27], the PWM signal
is transformed to a digital word quantizing the PWM with the
help of a counter and suitable logic, providing a serialized
output. The circuit schematic of the reset integrator is shown
in Fig. 5(b), while its timing waveform is shown in Fig. 5(c).
The reset integrator operates in two phases: in first phase
φ, a signal dependent discharging of the load capacitors CL
takes place, while in the other phase ¯
φ, a constant current Ib2
discharges the capacitors. The two phase integration is used
to improve the linearity of the reset integrator, by providing a
constant signal slope at the comparator inputs. The integrator
outputs are periodically reset to VDD. The discharge of the
reset integrator output branches (Int_out1,2) is monitored by
comparators. Based on the timing instants at which Int_out1,2
cross a given threshold, a PWM output of pulse width Tout is
generated (Fig. 5(c)), which is proportional to the differential
input signal.
The reset integrator based ADC in [27] has been exper-
imentally validated with the full sensor interface, which is
discussed in Section VII.
Considering now a traditional Nyquist analog to digital
converter with digital time-sampled output, Fig. 4(c) shows
a C-2C SAR ADC presented in [32], [90]. The main sub-
blocks of the ADC are a comparator, a DAC, a bias circuit
and a digital control sub-block. A sensor interface is added
at the input of the ADC. The block level diagram of the
comparator is shown in Fig. 5(d). It is built of cascading
differential amplifiers (DA) implemented with double gate
TFTs and using a diode connected load topology. Buffers
are also placed at the output stage after the comparator.
Auto-zeroing is used at each comparator stage to cancel the
offset. The operation of offset cancelation takes place in two
phases. Vaz is high in the first phase, which resets all the
input nodes of the DAs to Vcm. The capacitors store the
output offset. In the second phase, the comparator inputs are
connected to the inputs to be compared. The common mode
voltage Vcm at half rail is generated using a voltage divider
based on diode connected TFTs.
Fig. 5(e) shows the two 6-bit C-2C capacitor networks
that implement the DAC. One C-2C network is connected
to the input signal VIN from sensor while the other network
is used to obtain the comparator’s reference value from the
minimum reference (REFMIN) for the operation range of
the sensor. The C-2C DAC architecture is chosen for its
compactness and simplicity, and for the fact that ensures
sufficient linearity for the 6-bit quantizer that is discussed
in this work. The use of a second C-2C network to generate
the comparator reference alleviates inaccuracies caused by
charge injection due to switching and by interferers. A digital
control block is also implemented to control C-2C network
and comparator’s offset cancelation.
The experimental characterization of the 6-bit C-2C SAR
ADC yields an SNDR of 35.9 dB with a sampling rate of
26.67 S/s. The power dissipation of the analog part is 73 µW.
The area occupied by the ADC is 1.5cm2. A similar ADC
with different flavor of a-IGZO TFT technology is presented
in [91].
VII. SENSOR ACQUISITION SYSTEMS BASED ON
A-IGZO TFTS
The advancements in a-IGZO TFT processing technology
have resulted in applications beyond digital systems, that
exploit implementation of complex functionalities by inte-
grating different circuit blocks. In this section, we will
present some of the examples of state-of-the-art a-IGZO TFT
based systems that have successfully demonstrated various
applications. Two a-IGZO TFT based temperature sensors
have been demonstrated in [31], [32]. Furthermore, wear-
able physiological monitoring systems have been shown
in [27]–[30].
Reference [31] showed one of the first a-IGZO TFT based
temperature sensor. The architecture of the system is shown
in Fig. 4(a). The full system consists of the ADSM based
on a-IGZO TFTs presented in Fig. 4(a) and described in
Section VI, and two thin-film thermistors fabricated on a
Polyethylene Naphthalate (PEN) substrate. The micrograph
of the ADSM on flexible foil is shown in Fig. 6(a). The
experimental results are shown in Fig. 6(b). An accuracy
of 0.54C in 100 ms integration time is achieved, while
a maximum full range error of 2.3C is obtained after 3
points calibration. The total power dissipation is 2 mW. The
ADSM provides a binary PWM representation of the sensed
temperature, but no digital code.
Another temperature sensor based on a-IGZO TFTs is
presented in [32]. It is based on a 5-bit version of the C-2C
SAR ADC topology shown in Fig. 4(c) and discussed in
Section VI. Fig. 6(c) shows the foil micrograph of the full
temperature sensor. It consists of a C-2C SAR ADC and 4
reference tunable resistors using source drain metal layers to
generate the reference voltages. The system is demonstrated
750 VOLUME 2, 2021
(a)
(b)
(c)
(d)
FIGURE 6. Temperature sensors based on a-IGZO TFTs. (a) Foil micrograph of ADSM presented in [31], (b) Duty cycle vs temperature and inaccuracy vs temperature
characteristics of the temperature sensor presented in [31], (c) Foil micrograph of the system presented in [32], (d) Output digital code and temperature estimate from sensor’s
analog input. (a-b) Copyright 2017 IEEE. Adapted with permission from [31], (c-d) Copyright 2018 IEEE. Adapted with permission from [32].
TABLE 3. Comparison of the state-of-the-art a-IGZO TFT sensor systems.
with a printed Negative Temperature Coefficient (NTC)
sensor on a separate Polyimide (PI) foil. The measured tem-
perature sensing performance is shown in Fig. 6(d), which
shows the digital output code of the system in response to
NTC sensor’s input signal. It also shows the estimated non-
calibrated temperature. The power dissipation of the system
is 245 mW.
One of the first reported physiological monitoring system
based on a-IGZO TFTs is the electromyogram (EMG)
interface described in [29]. The system consists of four
chopper amplifiers driven at four different frequencies,
whose differential outputs are summed in the current
domain on a single pair of wires, to realize a frequency
division multiplexing (FDM). Fig. 7(a) shows the foil micro-
graph of one EMG acquisition circuit, while Fig. 7(b)
shows the positions of electrodes and the acquired EMG
signal during in-vivo experiment. The front-end achieves
an input referred noise of 31.4 µVrms in 500-Hz band-
width, 29.6-Minput impedance and 41-dB SNR. The
total power consumption is 1.3 mW. This design neces-
sitates a silicon IC to implement analog to digital
conversion.
VOLUME 2, 2021 751
ZULQARNAIN AND CANTATORE: ANALOG AND MIXED SIGNAL CIRCUIT DESIGN TECHNIQUES
FIGURE 7. Bio-signal monitoring systems based on a-IGZO TFTs. (a) Foil micrograph of the EMG front-end presented in [29], (b) Acquired EMG signal [29], (c) 4 ×4sEMG
sensor array [30], (d) sEMG acquired through passive MEA and active MEA, (e) Foil micrograph of the HR readout presented in [28], (f) Acquired HR signal from reconstructed
PWM [28], (g) Micrograph of the flexible ECG patch [27], (h) Acquired ECG signal from the output PWM and Manchester encoded output [27]. (a-b) Copyright 2018 IEEE.
Reprinted with permission from [29], (c-d) Copyright 2020 IEEE. Adapted with permission from [30], (e-f) Copyright 2018 IEEE. Adapted with permission from [28].
Fig. 7(c) shows a surface EMG (sEMG) multielectrode
array (MEA) based on a-IGZO TFTs [30]. It consists of 4×4
array of single ended (SE) common source (CS) amplifiers.
The MEA achieves gain of 15 dB in 480-Hz bandwidth with
supply voltage of 5 V. The power dissipation of a single CS
amplifier is 8 µW. Fig. 7(d) shows the sEMG signal acquired
during in-vivo experiment using a passive and active MEA.
The recorded SNR of passive and active MEA is 24.7 dB and
35.4 dB, respectively. This MEA demonstrates the large area
character of the technology enabling multiple electrodes, but
752 VOLUME 2, 2021
it does not provide multiplexing and requires other building
blocks such as filters, ADC, DSP etc. to complete the full
signal acquisition chain. The spatial resolution (pitch) of the
MEA is 4 mm, which might impact its clinical applicability.
Another example of physiological monitoring system is
the HR readout interface on flexible foil presented in [28]
and shown in Fig. 7(e). It is able to convert the input sig-
nal to a PWM representation using a reset integrator. The
system experimentally demonstrates a low-frequency gain
of 25 ms/V with an input referred noise of 186.3 µVrms
in 200-Hz bandwidth. The total power consumption of the
interface is 52 µW. Fig. 7(f) shows an in-vivo experiment
demonstrating measurement of the heartbeat of a person.
This interface is very power efficient but it is lacking the
digital output representation (i.e., in bits), which is required
to be compatible with most standard wireless transmission
systems. Moreover, the integrated input referred noise is too
high to perform most bio-signal measurements.
An evolution of this concept, aiming to demonstrate an
intelligent healthcare patch based on a-IGZO TFTs that is
compatible with the NFC wireless communication standard,
is demonstrated in [27]. This work exploits a digitization
strategy using a time reference, that can be derived from the
NFC carrier, without the need for a voltage reference. The
ECG signal is amplified, transformed to PWM representa-
tion using a reset integrator (Section VI) and quantized by
a counter to provide the digital output in the format of a
105.9 kb/s Manchester encoded serial bit stream (Fig. 4(b)).
The analog front-end in this system has an input-referred
noise of 8 µVrms in 100-Hz bandwidth. Moreover, the system
has 67.4-dB common mode rejection ration (CMRR), 58.9-
dB power supply rejection ratio (PSRR) and 16.5-Minput
impedance at 50 Hz with 1 kHz chopping. The power dis-
sipation of the analog subsystem is 280 µW, while the
digital subsystem consumes 15.4 mW. Fig. 7(g) shows the
foil micrograph of the system, while Fig. 7(h) demonstrates
in-vivo experiments showing the ECG reconstructed from the
PWM and from the Manchester encoded output bitstream.
To the best of authors’ knowledge, this is the first and only
demonstrated physiological signal acquisition system in flex-
ible a-IGZO TFTs on foil providing a digital serial output bit
stream compatible with the NFC standard. The system imple-
ments digital signal processing on the flexible foil, which
increases the overall power dissipation of the system con-
siderably due to the power-hungry nature of unipolar logic.
Moreover, it is a single channel system which does not fully
leverage the large area character of flexible technologies.
Apart from temperature sensing and physiological mon-
itoring applications, a-IGZO TFT based circuits have been
used in other applications such as radiation sensing [82],
magnetosensory applications [92], space applications [93],
gamma ray detection [94], NO2gas sensing [95] etc.
Moreover, IGZO TFTs have also been utilized as sens-
ing devices in an aqueous medium for applications such
as biosensing [96], virus detection [97], and artificial DNA
detection [98] etc.
VIII. CONCLUSION AND FUTURE PERSPECTIVES
Advances in flexible a-IGZO TFT processing technologies
have enabled the implementation of complex circuits and
architectures to realize various flexible sensors systems.
Emerging applications of the Internet of Things (IoT) such
as wearable textiles, wearables for diagnostics and sports,
sensing of physical quantities distributed on surfaces, NFC
tags etc. may benefit from this fabrication strategy.
a-IGZO TFT based amplifiers show a continuous trend to
realize high-gain amplifiers with novel load topologies while
using only n-type TFTs. The performance of these amplifiers
has reached the point where they can be used in demanding
low-noise applications like continuous physiological signal
monitoring. Recently, there has also been an effort to real-
ize larger bandwidth amplifiers, which could be used, e.g.,
in potential RF applications, also exploiting the impressive
improvements in terms of fTat device level.
The research in the field of data converters based on a-
IGZO TFTs is still in its infancy: only few data converters
have been presented in literature. Realizing a data converter
in the absence of complementary devices, voltage and cur-
rent references, and using a technology that is affected by
large parasitic capacitance and mismatch is indeed quite chal-
lenging. A clear trend is to prefer time-domain conversion
approaches, as time references at the state-of-the-art are eas-
ier to implement (in RFID or NFC platforms) than their
voltage or current counterparts.
Some significant achievements have been demonstrated in
complete sensor acquisition systems based on a-IGZO TFTs
as discussed in Section VII. However, there is still a long
way to go, before being able to realize the vision of fully
flexible sensors on foil integrating all parts of the signal
processing chain, including wireless data transmission. The
true potential of flexible technologies which lies in the large
area character, flexibility and conformability has thus not
been exploited entirely.
In terms of future perspective, research in the area of TFT
device processing and architecture should open up new pos-
sibilities for circuit designs offering lower supply voltage,
faster devices and better noise performance. The mobility
of present day a-IGZO TFTs is limited, however, operation
frequencies can be increased by scaling the TFTs dimen-
sions and lowering parasitic capacitance. Apart from that,
realization of reliable and well-matched integrated passive
components like capacitors and resistors would be an impor-
tant asset to enable better data conversion on foil. Low
supply voltage is also key to enable easier and efficient
integration between TFTs on flexible foil and silicon inte-
grated circuits, which is an interesting possibility to embody
applications requiring speed, power-efficient digital process-
ing and wireless communication together with flexible form
factor.
From the circuit design and system architecture perspec-
tive, a very important and rather unexplored domain in
flexible smart sensors is power management. Although many
groups are working on flexible batteries, more research on
VOLUME 2, 2021 753
ZULQARNAIN AND CANTATORE: ANALOG AND MIXED SIGNAL CIRCUIT DESIGN TECHNIQUES
voltage regulators, fast rectifiers for efficient RF energy har-
vesting in RFID platforms, current and voltage references,
etc. is essential to design and realize smart sensors that inte-
grate their energy sources, an achievement that would be of
paramount practical importance.
Non-conventional data processing approaches such as ana-
log classifiers, neuromorphic computing, machine learning,
compressed sensing and stochastic computing, which have
been reported in other flexible technologies [99]–[102], are
still unexplored in a-IGZO TFT technologies. The use of
such approaches at system level might ensure system robust-
ness against device variability [102], and also minimize the
number of physical interconnects needed to interface a large
number of sensors to the external world [101].
The demonstrators which have been shown in a-IGZO
TFT technologies so far, still need to be commercialized,
which, in turn, requires technology scalability. Nevertheless
we think that advancements in the field of a-IGZO TFT
technologies will play an important role in fulfilling societal
needs with unprecedented solutions.
REFERENCES
[1] K. Nomura, H. Ohta, A. Takagi, T. Kamiya, M. Hirano, and
H. Hosono, “Room-temperature fabrication of transparent flexible
thin-film transistors using amorphous oxide semiconductors,” Nature,
vol. 432, pp. 488–492, Nov. 2004.
[2] T. Kamiya, K. Nomura, and H. Hosono, “Present status of amorphous
In–Ga–Zn–O thin-film transistors,” Sci. Technol. Adv. Mater., vol. 11,
no. 4, Feb. 2010, Art. no. 044305.
[3] D. Karnaushenko et al., “Biomimetic microelectronics for regen-
erative neuronal cuff implants,Adv. Mater., vol. 27, no. 43,
pp. 6797–6805, Sep. 2015.
[4] P. Heremans et al., “Mechanical and electronic properties of thin-
film transistors on plastic, and their integration in flexible electronic
applications,” Adv. Mater., vol. 28, no. 22, pp. 4266–4282, Jun. 2016.
[5] Y. Kumaresan et al., “Highly bendable In-Ga-ZnO thin film transis-
tors by using a thermally stable organic dielectric layer,Sci. Rep.,
vol. 6, Nov. 2016, Art. no. 37764.
[6] N. Münzenrieder et al., “Stretchable electronics: Stretchable and con-
formable oxide thin-film electronics,” Adv. Electron. Mater.,vol.1,
no. 3, Feb. 2015, Art. no. 1400038.
[7] K. Myny, “The development of flexible integrated circuits based on
thin-film transistors,” Nat. Electron., vol. 1, pp. 30–39, Jan. 2018.
[8] J. Biggs et al., “A natively flexible 32-bit arm microprocessor,
Nature, vol. 595, pp. 532–536, Jul. 2021.
[9] T.-C. Fung, G. Baek, and J. Kanicki, “Low frequency noise in long
channel amorphous In-Ga-Zn-O thin film transistors,” J. Appl. Phys.,
vol. 108, Oct. 2010, Art. no. 074518.
[10] J.-S. Park et al., “Flexible full color organic light-emitting diode
display on polyimide plastic substrate driven by amorphous indium
gallium zinc oxide thin-film transistors,” Appl. Phys. Lett., vol. 95,
Jul. 2009, Art. no. 013503.
[11] M. Nag et al., “Novel back-channel-etch process flow based a-IGZO
TFTs for circuit and display applications on PEN foil,” J. Soc. Inf.
Display, vol. 21, pp. 369–375, Sep. 2013.
[12] M. Nag et al., “Circuits and AMOLED display with self-aligned
a-IGZO TFTs on polyimide foil,” J. Soc. Inf. Display, vol. 22,
pp. 509–517, Oct. 2014.
[13] Y. Xue et al., “31-inch 4K flexible display employing gate driver
with metal oxide thin-film transistors,” IEEE Electron Device Lett.,
vol. 42, no. 2, pp. 188–191, Feb. 2021.
[14] F. De Roose et al., “16.5 A flexible thin-film pixel array with a
charge-to-current gain of 59μA/pC and 0.3% nonlinearity and a cost
effective readout circuit for large-area X-ray imaging,” in Proc. IEEE
Int. Solid-State Circuits Conf. (ISSCC), Jan. 2016, pp. 296–297.
[15] A. J. van Breemen et al., “Curved digital X-ray detectors,NPJ
Flexible Electron., vol. 4, p. 22, Sep. 2020.
[16] T. Zou et al., “Pixellated perovskite photodiode on IGZO thin film
transistor backplane for low dose indirect X-ray detection,IEEE J.
Electron Devices Soc., vol. 9, pp. 96–101, 2021.
[17] N. Papadopoulos et al., “Touchscreen tags based on thin-film
electronics for the Internet of everything,Nat. Electron.,vol.2,
pp. 606–611, Dec. 2019.
[18] K. Wu et al., “Design of AM self-capacitive transparent touch
panel based on a-IGZO thin-film transistors,” IEEE Access,vol.8,
pp. 76929–76934, 2020.
[19] N. Papadopoulos et al., “78-4: IGZO-based identification tags com-
municating with everyday touchscreens,SID Symp. Dig. Tech.
Papers, vol. 51, pp. 1167–1170 Aug. 2020.
[20] B.-D. Yang, “A transparent logic circuit for RFID tag in a-IGZO
TFT technology,ETRI J., vol. 35, no. 4, pp. 610–616, Aug. 2013.
[21] M.-H. Hung et al., “Ultra low voltage 1-V RFID tag implement in
a-IGZO TFT technology on plastic,” in Proc. IEEE Int. Conf. RFID
(RFID), May 2017, pp. 193–197.
[22] K. Myny, A. K. Tripathi, J.-L. van der Steen, and B. Cobb,
“Flexible thin-film NFC tags,IEEE Commun. Mag., vol. 53, no. 10,
pp. 182–189, Oct. 2015.
[23] K. Myny et al., “16.3 flexible thin-film NFC tags powered by com-
mercial USB reader device at 13.56MHz,” in IEEE Int. Solid-State
Circuits Conf. (ISSCC) Dig. Tech. Papers, Feb. 2015, pp. 1–3.
[24] K. Myny and S. Steudel, “16.6 flexible thin-film NFC transpon-
der chip exhibiting data rates compatible to ISO NFC standards
using self-aligned metal-oxide TFTs,” in Proc. IEEE Int. Solid-State
Circuits Conf. (ISSCC), Jan. 2016, pp. 298–299.
[25] K. Myny et al., “A flexible ISO14443-A compliant 7.5mW 128b
metal-oxide NFC barcode tag with direct clock division circuit from
13.56MHz carrier,” in Proc. IEEE Int. Solid-State Circuits Conf.
(ISSCC), Feb. 2017, pp. 258–259.
[26] E. Ozer et al., “A hardwired machine learning processing engine fab-
ricated with submicron metal-oxide thin-film transistors on a flexible
substrate,” Nat. Electron., vol. 3, pp. 419–425, Jul. 2020.
[27] M. Zulqarnain et al., “A flexible ECG patch compatible with NFC RF
communication,” NPJ Flexible Electron., vol. 4, pp. 1–8, Jul. 2020.
[28] M. Zulqarnain et al.,“A52μW heart-rate measurement interface fab-
ricated on a flexible foil with a-IGZO TFTs,” in Proc. ESSCIRC 44th
Eur. Solid State Circuits Conf. (ESSCIRC), Sep. 2018, pp. 222–225.
[29] C. Garripoli, S. Abdinia, J.-L. J. P. van der Steen, G. H. Gelinck,
and E. Cantatore, “A fully integrated 11.2 mm2a-IGZO EMG front-
end circuit on flexible substrate achieving up to 41 dB SNR and
29 Minput impedance,” IEEE Solid-State Circuits Lett.,vol.1,
no. 6, pp. 142–145, Jun. 2018.
[30] S. Mao, J. Li, A. Guo, T. Zhao, and J. Zhang, “An active multi-
electrode array for collecting surface electromyogram signals using
a-IGZO TFT technology on polyimide substrate,” IEEE Trans.
Electron Devices, vol. 67, no. 4, pp. 1–6, Apr. 2020.
[31] C. Garripoli, J.-L. P. J. van der Steen, E. Smits, G. H. Gelinck,
A. H. M. V. Roermund, and E. Cantatore, “An a-IGZO asyn-
chronous delta-sigma modulator on foil achieving up to 43dB SNR
and 40dB SNDR in 300Hz bandwidth,” in Proc. IEEE Int. Solid-State
Circuits Conf. (ISSCC), Feb. 2017, pp. 260–261.
[32] N. P. Papadopoulos et al., “Toward temperature tracking with unipo-
lar metal-oxide thin-film SAR C-2C ADC on plastic,” IEEE J.
Solid-State Circuits, vol. 53, no. 8, pp. 2263–2272, Aug. 2018.
[33] N. Papadopoulos et al., “9-1: Invited paper: Metal-oxide readout
electronics based on indium-gallium-zinc-oxide and indium-tin-zinc-
oxide for in-panel fingerprint detection application,” SID Symp. Dig.
Tech. Papers, vol. 50, no. 1, pp. 95–98, May 2019.
[34] J. F. Mainguet et al., “A large-area curved pyroelectric finger-
print sensor,” in Proc. IEEE Int. Electron Devices Meeting (IEDM),
Dec. 2019, pp. 26.5.1–26.5.4.
[35] F. D. Roose et al., “14-1: Flexible large-area multi-fingerprint sensors
based on thermal mass detection,” SID Symp. Dig. Tech. Papers,
vol. 51, pp. 176–179, Aug. 2020.
[36] D. Geng, S. Han, H. Seo, M. Mativenga, and J. Jang, “Piezoelectric
pressure sensing device using top-gate effect of dual-gate a-IGZO
TFT,IEEE Sensors J., vol. 17, no. 3, pp. 585–586, Feb. 2017.
[37] Z. Zhang et al., “Enhanced flexible piezoelectric sensor by the inte-
gration of p(VDF-TrFE)/AgNWs film with a-IGZO TFT,” IEEE
Electron Device Lett., vol. 40, no. 1, pp. 111–114, Jan. 2019.
754 VOLUME 2, 2021
[38] M. Nag et al., “High performance a-IGZO thin-film transistors with
mf-PVD SiO2 as an etch-stop-layer,J. Soc. Inf. Display, vol. 22,
no. 1, pp. 23–28, Jan. 2014.
[39] A. K. Tripathi, K. Myny, B. Hou, K. Wezenberg, and G. H. Gelinck,
“Electrical characterization of flexible InGaZnO transistors and 8-b
transponder chip down to a bending radius of 2 mm,IEEE Trans.
Electron Devices, vol. 62, no. 12, pp. 4063–4068, Dec. 2015.
[40] D. H. Kang, I. Kang, S. H. Ryu, and J. Jang, “Self-aligned coplanar
a-IGZO TFTs and application to high-speed circuits,” IEEE Electron
Device Lett., vol. 32, no. 10, pp. 1385–1387, Oct. 2011.
[41] N. Münzenrieder et al., “Flexible self-aligned double-gate IGZO
TFT,IEEE Electron Device Lett., vol. 35, no. 1, pp. 69–71,
Jan. 2014.
[42] M. Nag et al., “20.1: Flexible AMOLED display and gate-driver
with self-aligned IGZO TFT on plastic foil,” SID Symp. Dig. Tech.
Papers, vol. 45, no. 1, pp. 248–251, Jun. 2014.
[43] Y.-H. Kim et al., “Flexible metal-oxide devices made by room-
temperature photochemical activation of sol–gel films,Nature,
vol. 489, pp. 128–132, Sep. 2012.
[44] L. Petti, G. Cantarella, J. C. Costa, and N. S. Münzenrieder,
“Bendable metal oxide thin-film transistors and circuits for ana-
log electronics applications,” in Proceedings of the Oxide-Based
Materials and Devices XII, F. H. Teherani, D. C. Look, and D. J.
Rogers, Eds. San Francisco, CA, USA: SPIE, Mar. 2021, pp. 73–78.
[45] J. Jiang, M. Furuta, and D. Wang, “Self-aligned bottom-gate
In–Ga–Zn–O thin-film transistor with source/drain regions formed
by direct deposition of fluorinated silicon nitride,” IEEE Electron
Device Lett., vol. 35, no. 9, pp. 933–935, Sep. 2014.
[46] K. Han, S. Samanta, C. Sun, J. Zhang, Z. Zheng, and X. Gong, “Top-
gate short channel amorphous indium-gallium-zinc-oxide thin film
transistors with sub-1.2 nm equivalent oxide thickness,” in Proc. 5th
IEEE Electron Devices Technol. Manuf. Conf. (EDTM), Apr. 2021,
pp. 1–3.
[47] D. Geng, D. H. Kang, and J. Jang, “High-performance amorphous
indium-gallium-zinc-oxide thin-film transistor with a self-aligned
etch stopper patterned by back-side UV exposure,” IEEE Electron
Device Lett., vol. 32, no. 6, pp. 758–760, Jun. 2011.
[48] G. Baek, K. Abe, A. Kuo, H. Kumomi, and J. Kanicki, “Electrical
properties and stability of dual-gate coplanar homojunction DC sput-
tered amorphous indium–gallium–zinc–oxide thin-film transistors
and its application to AM-OLEDs,” IEEE Trans. Electron Devices,
vol. 58, no. 12, pp. 4344–4353, Dec. 2011.
[49] M. Nag et al., “Characteristics improvement of top-gate self-aligned
amorphous indium gallium zinc oxide thin-film transistors using a
dual-gate control,” J. Soc. Inf. Display, vol. 25, no. 6, pp. 349–355,
Jun. 2017.
[50] L. Petti et al., “Mechanically flexible vertically integrated a-IGZO
thin-film transistors with 500 nm channel length fabricated on free
standing plastic foil,” in Proc. IEEE Int. Electron Devices Meeting,
Dec. 2013, pp. 11.4.1–11.4.4.
[51] Y. Liu, H. Zhou, R. Cheng, W. Yu, Y. Huang, and X. Duan, “Highly
flexible electronics from scalable vertical thin film transistors,Nano
Lett., vol. 14, no. 3, pp. 1413–1418, Feb. 2014.
[52] L. Petti et al., “Flexible quasi-vertical In-Ga-Zn-O thin-film transistor
with 300-nm channel length,” IEEE Electron Device Lett., vol. 36,
no. 5, pp. 475–477, May 2015.
[53] H.-R. Kim, M. Furuta, and S.-M. Yoon, “Highly robust flexible
vertical-channel thin-film transistors using atomic-layer-deposited
oxide channels and zeocoat spacers on ultrathin polyimide sub-
strates,” ACS Appl. Electron. Mater., vol. 1, pp. 2363–2370,
Oct. 2019.
[54] H.-R. Kim, J.-H. Yang, G.-H. Kim, and S.-M. Yoon, “Flexible
vertical-channel thin-film transistors using In-Ga-Zn-O active chan-
nel and polyimide spacer on poly(ethylene naphthalate) substrate,”
J. Vacuum Sci. Technol. B, vol. 37, Jan. 2019, Art. no. 010602.
[55] L. Petti et al., “Metal oxide semiconductor thin-film transistors
for flexible electronics,” Appl. Phys. Rev., vol. 3, Jun. 2016,
Art. no. 021303.
[56] G. Cantarella et al., “Review of recent trends in flexible metal
oxide thin-film transistors for analog applications,” Flexible Printed
Electron., vol. 5, Sep. 2020, Art. no. 033001.
[57] M. Nag et al., “P-12: High performance dual-gate dual-layer amor-
phous oxide semiconductors TFTs on PI foil for display application,”
SID Symp. Dig. Tech. Papers, vol. 50, pp. 1255–1258, May 2019.
[58] I. M. Choi et al., “Achieving high mobility and excellent stability in
amorphous In–Zn–Sn–O thin-film transistors,” IEEE Trans. Electron
Devices, vol. 67, no. 3, pp. 1014–1020, Mar. 2020.
[59] S.-L. Li, M.-X. Lee, C.-C. Yen, T.-L. Chen, C.-H. Chou, and
C. W. Liu, “Double-layer amorphous InGaZnO thin film transistors
with high mobility and high reliability,” in Proc. Int. Symp. VLSI
Technol. Syst. Appl. (VLSI-TSA), Apr. 2021, pp. 1–2.
[60] D. Kumar, A. Abdou, and J. Kettle, “Half-volt IGZO flexible
thin-film transistors with e-beam deposited Al2O3gate dielectric,”
in Proc. IEEE Int. Conf. Flexible Printable Sens. Syst. (FLEPS),
Jul. 2019, pp. 1–3.
[61] N. Mohammadian, B. C. Das, and L. A. Majewski, “Low-voltage
IGZO TFTs using solution-deposited OTS-modified Ta2O5dielec-
tric,” IEEE Trans. Electron Devices, vol. 67, no. 4, pp. 1625–1631,
Apr. 2020.
[62] N. Münzenrieder, L. Petti, C. Zysset, T. Kinkeldei, G. A. Salvatore,
and G. Troster, “Flexible self-aligned amorphous InGaZnO thin-
film transistors with submicrometer channel length and a transit
frequency of 135 MHz,” IEEE Trans. Electron Devices, vol. 60,
no. 9, pp. 2815–2820, Sep. 2013.
[63] Y. Wang et al., “Amorphous-InGaZnO thin-film transistors operating
beyond 1 GHz achieved by optimizing the channel and gate dimen-
sions,” IEEE Trans. Electron Devices, vol. 65, no. 4, pp. 1377–1382,
Apr. 2018.
[64] C. Tuckmantel, U. Kalita, T. Haeger, M. Theisen, U. Pfeiffer, and
T. Riedl, “Amorphous indium-gallium-zinc-oxide TFTs patterned by
self-aligned photolithography overcoming the GHz threshold,” IEEE
Electron Device Lett., vol. 41, no. 12, pp. 1786–1789, Dec. 2020.
[65] N. Münzenrieder, G. Cantarella, and L. Petti, “Fabrication and
AC performance of flexible indium-gallium-zinc-oxide thin-film
transistors,” ECS Trans., vol. 90, no. 1, pp. 55–63, Apr. 2019.
[66] S. Samanta, K. Han, C. Sun, C. Wang, A. V.-Y. Thean, and X. Gong,
“Amorphous IGZO TFTs featuring extremely-scaled channel thick-
ness and 38 nm channel length: Achieving record high gm,max of
125 μS/μm at VDS of 1 V and ION of 350 μA/μm,” in Proc. IEEE
Symp. VLSI Technol., Jun. 2020, pp. 1–2.
[67] C. C. Enz and G. C. Temes, “Circuit techniques for reducing the
effects of op-amp imperfections: Autozeroing, correlated double
sampling, and chopper stabilization,” Proc. IEEE, vol. 84, no. 11,
pp. 1584–1614, Nov. 1996.
[68] R. Martins et al., “Complementary metal oxide semiconductor
technology with and on paper,Adv. Mater., vol. 23, no. 39,
pp. 4491–4496, Sep. 2011.
[69] H. Chen, Y. Cao, J. Zhang, and C. Zhou, “Large-scale complementary
macroelectronics using hybrid integration of carbon nanotubes and
IGZO thin-film transistors,” Nat. Commun., vol. 5, p. 4097, Jun. 2014.
[70] W. Honda, S. Harada, S. Ishida, T. Arie, S. Akita, and K. Takei,
“High-performance, mechanically flexible, and vertically integrated
3D carbon nanotube and InGaZnO complementary circuits with a
temperature sensor,Adv. Mater., vol. 27, pp. 4674–4680, Jul. 2015.
[71] Y. Li et al., “Complementary integrated circuits based on p-type SnO
and n-type IGZO thin-film transistors,” IEEE Electron Device Lett.,
vol. 39, no. 2, pp. 208–211, Feb. 2018.
[72] S. Park et al., “Sub-0.5 V highly stable aqueous salt gated metal
oxide electronics,” Sci. Rep., vol. 5, Aug. 2015, Art. no. 13088.
[73] C. J. Wan, Y. H. Liu, L. Q. Zhu, P. Feng, Y. Shi, and
Q. Wan, “Short-term synaptic plasticity regulation in solution-gated
indium–gallium–zinc-oxide electric-double-layer transistors,” ACS
Appl. Mater. Interfaces, vol. 8, pp. 9762–9768, Apr. 2016.
[74] K. Ishida et al., “22.5 dB open-loop gain, 31 kHz GBW pseudo-
CMOS based operational amplifier with a-IGZO TFTs on a flexible
film,” in Proc. IEEE Asian Solid-State Circuits Conf. (A-SSCC),
Nov. 2014, pp. 313–316.
[75] M. Zulqarnain, S. Stanzione, J.-L. P. J. van der Steen, G. H. Gelinck,
S. Abdinia, and E. Cantatore, “Design trade-offs in amorphous
indium gallium zinc oxide thin film transistor based bio-signal
sensing front-ends,” Flexible Printed Electron., vol. 4, Jan. 2019,
Art. no. 014001.
[76] C. Garripoli et al., “Analogue frontend amplifiers for bio-potential
measurements manufactured with a-IGZO TFTs on flexible sub-
strate,” IEEE J. Emerg. Sel. Topics Circuits Syst., vol. 7, no. 1,
pp. 60–70, Mar. 2017.
VOLUME 2, 2021 755
ZULQARNAIN AND CANTATORE: ANALOG AND MIXED SIGNAL CIRCUIT DESIGN TECHNIQUES
[77] C. Zysset, N. Munzenrieder, L. Petti, L. Buthe, G. A. Salvatore, and
G. Troster, “IGZO TFT-based all-enhancement operational amplifier
bent to a radius of 5 mm,” IEEE Electron Device Lett., vol. 34,
no. 11, pp. 1394–1396, Nov. 2013.
[78] N. Papadopoulos et al., “In-panel 31.17dB 140kHz 87μW unipolar
dual-gate In-Ga-Zn-O charge-sense amplifier for 500dpi sensor array
on flexible displays,” in Proc. IEEE 44th Eur. Solid State Circuits
Conf. (ESSCIRC), Sep. 2018, pp. 194–197.
[79] R. Shabanpour et al.,“A70
phase margin OPAMP with positive
feedback in flexible a-IGZO TFT technology,” in Proc. IEEE 58th
Int. Midwest Symp. Circuits Syst. (MWSCAS), Aug. 2015, pp. 1–4.
[80] P. G. Bahubalindruni et al., “Analog circuits with high-gain topolo-
gies using a-GIZO TFTs on glass,” J. Display Technol., vol. 11,
no. 6, pp. 547–553, Jun. 2015.
[81] M. J. Seok, M. Mativenga, D. Geng, and J. Jang, “Achieving high
performance oxide TFT-based inverters by use of dual-gate con-
figurations with floating and biased secondary gates,” IEEE Trans.
Electron Devices, vol. 60, no. 11, pp. 3787–3793, Nov. 2013.
[82] P. G. Bahubalindruni et al., “High-gain transimpedance amplifier for
flexible radiation dosimetry using InGaZnO TFTs,” IEEE J. Electron
Devices Soc., vol. 6, pp. 760–765, 2018.
[83] C. Perumal et al., “A compact a-IGZO TFT model based on MOSFET
SPICE Level = 3 template for analog/RF circuit designs,IEEE
Electron Device Lett., vol. 34, no. 11, pp. 1391–1393, Nov. 2013.
[84] R. Shabanpour et al., “High gain amplifiers in flexible self-aligned a-
IGZO thin-film-transistor technology,” in Proc. 21st IEEE Int. Conf.
Electron. Circuits Syst. (ICECS), Dec. 2014, pp. 108–111.
[85] R. Shabanpour et al., “Cherry-hooper amplifiers with 33 dB gain at
400 kHz BW and 10 dB gain at 3.5 MHz BW in flexible self-aligned
a-IGZO TFT technology,” in Proc. Int. Symp. Intell. Signal Process.
Commun. Syst. (ISPACS), Dec. 2014, pp. 271–274.
[86] R. Shabanpour et al., “A transistor model for a-IGZO TFT circuit
design built upon the RPI-aTFT model,” in Proc. 15th IEEE Int. New
Circuits Syst. Conf. (NEWCAS), Jun. 2017, pp. 129–132.
[87] A. Rahaman, Y. Chen, M. M. Hasan, and J. Jang, “A high
performance operational amplifier using coplanar dual gate a-IGZO
TFTs,” IEEE J. Electron Devices Soc., vol. 7, pp. 655–661, 2019.
[88] T. Meister, K. Ishida, A. Sou, C. Carta, and F. Ellinger, “49.35 MHz
GBW and 33.43 MHz GBW amplifiers in flexible a-IGZO TFT
technology,Electron. Lett., vol. 56, pp. 782–785, Jul. 2020.
[89] S. Ouzounov, E. Roza, J. A. Hegt, G. van der Weide, and
A. H. M. van Roermund, “Analysis and design of high-performance
asynchronous sigma-delta modulators with a binary quantizer,IEEE
J. Solid-State Circuits, vol. 41, no. 3, pp. 588–596, Mar. 2006.
[90] N. Papadopoulos et al., “Flexible selfbiased 66.7nJ/c.s. 6bit 26S/s
successive-approximation C-2C ADC with offset cancellation using
unipolar metal-oxide TFTs,” in Proc. IEEE Custom Integr. Circuits
Conf. (CICC), Apr. 2017, pp. 1–4.
[91] N. Papadopoulos, S. Steudel, A. J. Kronemeijer, M. Ameys, and
K. Myny, “Flexible 16nJ/c.s. 134S/s 6b MIM C-2C ADC using dual
gate self-aligned unipolar metal-oxide TFTs,” in Proc. IEEE Custom
Integr. Circuits Conf. (CICC), Apr. 2019.
[92] N. Münzenrieder et al., “Entirely flexible on-site conditioned mag-
netic sensorics,” Adv. Electron. Mater., vol. 2, no. 8, Jun. 2016,
Art. no. 1600188.
[93] J. C. Costa et al., “Flexible IGZO TFTs and their suitability for space
applications,” IEEE J. Electron Devices Soc., vol. 7, pp. 1182–1190,
2019.
[94] M. B. Zalte, V. Kumar, S. G. Surya, and M. S. Baghini, “A solution
processed amorphous InGaZnO thin-film transistor-based dosimeter
for gamma-ray detection and its reliability,IEEE Sensors J., vol. 21,
no. 9, pp. 10667–10674, May 2021.
[95] M. T. Vijjapu, S. G. Surya, S. Yuvaraja, and K. N. Salama, “A multi-
bit fully integrated thin-film transistor NO2gas detector at room
temperature,” IEEE Sens. Lett., vol. 4, no. 8, pp. 1–4, Aug. 2020.
[96] H. W. Son, J. H. Park, M.-S. Chae, B.-H. Kim, and T. G. Kim,
“Bilayer indium gallium zinc oxide electrolyte-gated field-effect tran-
sistor for biosensor platform with high reliability,Sens. Actuators
B, Chem., vol. 312, Jun. 2020, Art. no. 127955.
[97] T.-H. Yang, T.-Y. Chen, N.-T. Wu, Y.-T. Chen, and J.-J. Huang,
“IGZO-TFT biosensors for Epstein–Barr virus protein detection,”
IEEE Trans. Electron Devices, vol. 64, no. 3, pp. 1294–1299,
Mar. 2017.
[98] S. J. Kim et al., “Low-cost label-free electrical detection of arti-
ficial DNA nanostructures using solution-processed oxide thin-film
transistors,” ACS Appl. Mater. Interfaces, vol. 5, pp. 10715–10720,
Oct. 2013.
[99] M. Ozatay et al., “Artificial intelligence meets large-scale sensing:
Using large-area electronics (LAE) to enable intelligent spaces,” in
Proc. IEEE Custom Integr. Circuits Conf. (CICC), Apr. 2018, pp. 1–8.
[100] L. E. Aygun et al., “Hybrid LAE-CMOS force-sensing system
employing TFT-based compressed sensing for scalability of tactile
sensing skins,” IEEE Trans. Biomed. Circuits Syst., vol. 13, no. 6,
pp. 1264–1276, Dec. 2019.
[101] J. C. Sturm et al., “Machine learning and high-speed circuitry in thin
film transistors for sensor interfacing in hybrid large-area electronic
systems,” ECS Trans., vol. 92, pp. 121–134, Jul. 2019.
[102] L. Shao, T. Lei, T.-C. Huang, Z. Bao, and K.-T. Cheng, “Robust
design of large area flexible electronics via compressed sensing,
in Proc. 57th ACM/IEEE Design Autom. Conf. (DAC), Jul. 2020,
pp. 1–6.
MOHAMMAD ZULQARNAIN (Member, IEEE) is
currently a Postdoctoral Researcher with the IC
Group, Eindhoven University of Technology, The
Netherlands. During his Ph.D., he worked on
design of wearable systems based on flexible elec-
tronics, in collaboration with imec, Belgium, and
Holst Centre/imec-nl, The Netherlands. He has
authored or coauthored ten papers in journals and
conference proceedings. His research interest is
in the design of biomedical circuits and systems
using flexible thin film transistors.
He serves as a Reviewer for IEEE TRANSACTIONS ON CIRCUITS
AND SYSTEMS—II: EXPRESS BRIEFS, MDPI Electronics, and many IEEE
conferences, including ISCAS, BioCAS, FLEPS, and MWSCAS.
EUGENIO CANTATORE (Fellow, IEEE) is a Full
Professor with the Integrated Circuits Group,
Eindhoven University of Technology, where he
leads the Emerging Technologies Lab. He authored
or coauthored over 200 papers in journals and
conference proceedings. He holds 13 patents. His
research focusses on the design and characteriza-
tion of electronic circuits fabricated with emerging
technologies, as well as the design of ultra-low
power micro-systems for biomedical applications.
One of his main interests is the design of flexible
electronics fabricated on plastic foils, including sensors interfaces, analog-
digital converters, and transceivers.
He was a recipient of the Beatrice Winner Award from ISSCC for
Editorial Excellence in 2006, the Philips Research Invention Award in 2007,
the Best Paper Award from ESSDERC 2012, the Distinguished Technical
Paper Award from ISSCC 2015, and nominated in the Scientific American
top 50 list. He has been twice a Guest Editor of the Journal of Solid-State
Circuits. He has been active in the Technical Program Committees of
ESSDERC, IWASI, ESSCIRC, and ISSCC. At ISSCC, he has been a Chair
of the Technology Directions Subcommittee from 2013 to 2016, a Program
Chair in 2019, and is currently a Conference Vice Chair. He is a member at
large of the SSCS AdCom, an Associate Editor of IEEE TRANSACTIONS ON
CIRCUITS AND SYSTEMS—I: REGULAR PAPERS, and the Editor-in-Chief
of IEEE Open Journal of Solid-State Circuits Society. He was nominated
IEEE Fellow in 2016.
756 VOLUME 2, 2021
... [23][24][25][26] More importantly, the amorphous nature of the IGZO enables the use of standard deposition methods compatible with CMOS back-end-of-line processes, making it feasible for large-scale preparation. 27 A stackable vertical Channel-All-Around (CAA) IGZO FETs with ultra-high-density DRAM beyond 4F 2 shows the great potential of IGZO for further density scaling. 28 IGZO-based TSM for stochastic neuron emulations has been reported in our previous work, showing remarkable stochastic resistive switching properties. ...
Article
True random number generators (TRNGs) can generate unpredictable binary bitstream by exploiting the intrinsic stochasticity in physical variables. In a threshold switching memristor, the stochastic forming/rupture of conducting pathway has been proved to be a good random source, while further improvement of high randomness and throughput is still a challenge. Here, a crossbar array of amorphous indium–gallium–zinc–oxide (a-IGZO)-based threshold switching memristors was designed for high-throughput TRNGs. The intrinsic stochasticity of Ag conductive filament in IGZO memristor and the stochastic sneak paths in the crossbar array are the two sources of randomness in our TRNGs. In our design, one input pulse train can produce multi-channel random bits, which enables a high scalability for such TRNGs. In addition, the average energy consumption of the TRNGs can be further reduced by increasing the integration scale of the memristors. Such IGZO-based TRNGs are of great significance for security applications.
... Even though several FlexIC digital circuits and mixed-signal interfaces can be found in recent literature [1][2][3][4][5], [8][9][10][11][12], the contributions on flexible analog building blocks as Operational Transconductance Amplifiers (OTAs) are still limited in number [13][14][15][16][17]. Moreover, the OTAs in IGZO technology proposed so far suffer of relatively poor performance due to the intrinsic limitations of the IGZO technology and in particular to the lack of complementary devices. ...
Conference Paper
In this paper, an Analog and a Digital-Based Operational Transconductance Amplifier (OTA) in a 800 nm Indium-Gallium-Zinc-Oxide (IGZO) Thin-Film Transistors (TFT) Flexible Integrated Circuits (FlexICs) technology are presented and compared on the basis of post-layout simulations. The analog OTA (A-OTA) and the Digital-Based OTA (DB-OTA) occupy a total area of 42,624μm2 and 25,207μm2 , respectively and - based on post-layout Monte Carlo (MC) simulations on 100 samples operated at 3.3V with 50pF capacitive load - they achieve an average gain-bandwidth product (GBW) of 58 kHz and 86 kHz, respectively, with an average power consumption of 90μW and 113μW . The simulated standard deviation of the input offset voltage is 22.3 mV for the A-OTA and 7.2 mV for the DB-OTA while the input-referred integrated noise over the entire GBW is 8.8μVRMS and 87μVRMS for the A-OTA and DB-OTA respectively.
Article
High-throughput printing-based fabrication has emerged as a key enabler of flexible electronics given its unique capability for low-cost integration of circuits based on printed thin film transistors (TFTs). Research in printing inorganic metal oxides has revealed the potential for fabricating oxide TFTs with an unmatched combination of high electron mobility and optical transparency. Here, we highlight recent developments in ink chemistry, printing physics, and material design for high-mobility metal oxide transistors. We consider ongoing challenges for this field that include lowering process temperatures, achieving high speed and high resolution printing, and balancing device performance with the need for high mechanical flexibility. Finally, we provide a roadmap for overcoming these challenges with emerging synthetic strategies for fabricating 2D oxides and complementary TFT circuits for flexible electronics.
Article
In this study, the homojunction thin-film transistors (TFTs) with amorphous indium gallium zinc oxide (a-IGZO) as active channel layers and source/drain electrodes were fabricated by RF magnetron sputtering. The effect of oxygen partial pressure on the phase, microstructure, optical and electrical properties of IGZO thin films was investigated. The results showed that amorphous IGZO thin films always exhibit a high transmittance above 90% and wide band gaps of around 3.9 eV. The resistivity increases as the IGZO thin films are deposited at a higher oxygen partial pressure due to the depletion of oxygen vacancies. In addition, the electrical behaviors in homojunction IGZO TFTs were analyzed. When the active channel layers were deposited with an oxygen partial pressure of 1.96%, the homojunction IGZO TFTs exhibited optimal transfer and output characteristics with a field-effect mobility of 13.68 cm2 V-1s-1. Its sub-threshold swing, threshold voltage and on/off ratio are 0.6 V/decade, 0.61 V and 107, respectively.
Article
Flexible electronics based on thin film transistors (TFTs) are expected to deliver important benefits in the mobile healthcare field. Nevertheless, the mechanical strain caused by the respiration or movement of organisms seriously affects the characteristics of the TFTs, and thus deteriorates the performance of TFT-based circuits and systems. Taking indium-gallium-zinc-oxide (IGZO) TFT as an example, this paper proposes a tensile-force-resilient biomedical front-end system based on omni-directional thin film transistors (OD TFTs). The proposed OD TFT is a combination of dual gate TFTs with optimized layout, whose force-insensitive axis can be adjusted by the biasing of top gate. Based on this, an auto calibration technique in front-end system is introduced, which can automatically adjust the top gate biasing of OD TFTs according to the Poisson's ratio of substrate and the direction of external force, so as to align the force-insensitive axis along the direction of external force. Employing the OD TFTs in circuits and the auto calibration technique, the force resilience of the flexible electronic system is significantly improved in practical applications. The simulations of front-end system on acquiring electrocardiogram (ECG) and electromyogram (EMG) signals are performed. The results show that the proposed front-end system with auto calibration provides an increase of SNR from initial −19.91 dB and −25.89 dB without calibration to 17.53 dB and 1.32 dB in two cases, respectively.
Article
This letter presents analysis of an inherently stable positive-feedback-based active load, first reported in our prior work, to realize high gain analog amplifiers using unipolar a-IGZO TFT technology. Additionally reported are theoretical performance bounds and design guidelines to maximally utilize a general positive-feedback scheme for the active load while maintaining stability. To demonstrate a design implementation, a single stage fully differential operational transconductance amplifier with common mode feedback has been manufactured and measured to have 40-dB open-loop gain at dc and a unity gain frequency of 61 kHz while driving a 30-pF load capacitance.
Article
Full-text available
This special Section of the IEEE Open Journal of Circuits and Systems (OJCAS) is dedicated to the recent developments in circuits and systems for flexible electronics.
Article
Full-text available
Nearly 50 years ago, Intel created the world’s first commercially produced microprocessor—the 4004 (ref. ¹), a modest 4-bit CPU (central processing unit) with 2,300 transistors fabricated using 10 μm process technology in silicon and capable only of simple arithmetic calculations. Since this ground-breaking achievement, there has been continuous technological development with increasing sophistication to the stage where state-of-the-art silicon 64-bit microprocessors now have 30 billion transistors (for example, the AWS Graviton2 (ref. ²) microprocessor, fabricated using 7 nm process technology). The microprocessor is now so embedded within our culture that it has become a meta-invention—that is, it is a tool that allows other inventions to be realized, most recently enabling the big data analysis needed for a COVID-19 vaccine to be developed in record time. Here we report a 32-bit Arm (a reduced instruction set computing (RISC) architecture) microprocessor developed with metal-oxide thin-film transistor technology on a flexible substrate (which we call the PlasticARM). Separate from the mainstream semiconductor industry, flexible electronics operate within a domain that seamlessly integrates with everyday objects through a combination of ultrathin form factor, conformability, extreme low cost and potential for mass-scale production. PlasticARM pioneers the embedding of billions of low-cost, ultrathin microprocessors into everyday objects.
Article
Full-text available
The integration of perovskite photodetectors with thin-film transistor (TFT) backplane or complementary metal-oxide-semiconductor CMOS circuit is a key step towards prototyping perovskite-based image sensors. Here, we demonstrate a pixel configuration for indirect X-ray detection comprising of IGZO TFTs and perovskite photodiodes (PDs). The perovskite photodiode is patterned by a two-step deposition method. Our integrated TFT/PD pixel shows a weak light detection capability down to 4 nW cm-2, and a fast-transient response to the pulse light and gate switching. Combining with a CsI scintillator, the integrated pixel achieves a specific X-ray sensitivity of 8.2 × 102 μC mGyair-1 cm-3. Theoretically, with a state-of-the-art scintillator, the new pixel can provide a detectable signal for X-ray imaging at a dose rate as low as 10 μGyair s-1. This work provides an advanced pixel design for high resolution, high sensitivity, and high frame-rate flat-panel imager.
Article
Metal oxide semiconductors proved their usability in environmental monitoring applications and are considered successful in detecting ionizing radiation. This work reports the feasibility of solution-processed metal oxide semiconductor thin-film transistors for radiation sensing for the first time. In particular, the effects of a wide range of gamma radiation (100Gy to 10kGy) on the performance of solution-processed amorphous Indium-Gallium-Zinc-Oxide (a-IGZO) transistors are investigated. The current-voltage (IV) characterization (both output and transfer characteristics) of IGZO-TFTs before and after irradiation are obtained to study different parameters. The radiation-induced changes in TFT are mainly observed in the threshold voltage shift ( $\sf \Delta ~\text{V}_{th}$ ) and the increase of subthreshold swing. It is observed that up to a total dose of 1kGy, threshold voltage increases negatively ( $\sf \Delta \,\,\text{V}_{th}= -1.8\text{V}$ at 1 kGy), and beyond 1 kGy, threshold voltage increases positively ( $\Delta \,\,\text{V}_{\text{th}} =0.8\text{V}$ at 10 kGy). The XRD and AFM data of IGZO thin-film suggests minor structural and morphological changes after exposure to gamma irradiation. The corresponding sensitivity obtained with gamma irradiation is 27.78 mV/Gy (100Gy-1kGy), expressed in the threshold voltage shift. The effects of radiation-induced changes in TFTs are completely removed after storing irradiated TFTs in the vacuum at room temperature for 6 months.
Article
A 31-inch 4K flexible active-matrix organic light-emitting diode (AMOLED) display integrating with gate driver on the array (GOA) technology has been proposed and fabricated on a polyimide substrate. The GOA circuits, composed of two series-connected thin-film transistors (TFTs) and dual low-voltage power structure, are introduced to decrease the leakage current in TFTs. They are arranged symmetrically along the sides of the display. Inkjet printing (IJP) OLED process is used to reduce the number of sub-pixels from four (R/G/B/W) in commercial white OLED process to three (R/G/B) and a high pixel density in the display (144 pixels per inch) is achieved. Moreover, top-gated (TG) amorphous indium-gallium-zinc oxide (IGZO) TFTs with good electrical and mechanical reliability, have been successfully utilized to decrease the capacitance load in the display and improve signal transfer speed in GOA. Finally, this 31-inch flexible AMOLED display was operated for 500h under accelerating test conditions of 60°C and 90% humidity.
Article
In this letter, we demonstrate amorphous indium-gallium-zinc-oxide (IGZO) thin-film transistors (TFTs) with a maximum transit frequency ( ${f}_{T}$ ) of 1.1 GHz and maximum oscillation frequency ( ${f}_{max}$ ) of 3.1 GHz, which set a new record for metal-oxide TFTs based on scalable and cost-effective process techniques. A key for the outstanding high-frequency performance of our TFTs is an optimized two-step self-alignment concept, which allows the rigorous minimization of parasitic elements, that typically limit high-frequency operation. We expect that our results will inspire a wide range of high-frequency circuits based on metal-oxide electronics.
Article
This paper shows the concept and design of a flexible large‐area multi‐fingerprint sensor. The sensor detects the thermal mass of different materials, e.g. the skin with high heat capacity and air with very low heat capacity. We show the device architecture and readout strategies. The performance of the 500ppi flexible sensor is not influenced by bending with a radius of 5mm.