On-going Projects
4. Sensors of Piezoelectrets
and Biomedical Systems Based on Piezoelectrets
|
Smart insole with sensors of piezoelectrets for measuring
movements. |
|
Schematic representation of the layer sequence for the 3D
printing process: (a) Bottom layer, (b) Central layer with a fill factor of
80% (triangular pattern) and (c) Top layer (partially cut away to expose
layer b). |
|
(a) Top view of the bottom and central layers, showing the
triangular patterns of the central polypropylene layer. (b) SEM image (top
view). (c) Cross section showing two of the 3D-printed cavities. |
3. Chemical Sensors Based on RF
and Optics
2. Irradiation System for Changing
Electronic Materials
FE-SEM images of (a–e) the pellet surface and (f–j) the cross sections of all samples with different time irradiation 1, 2, 4 and 8 min, respectively.
FE-SEM images of the surface and
cross-section of pristine carbon and (aeb) pellets irradiated by
electron beam for different times of exposure dose: (ced) 2 min, (eef)
4 min, (geh)
8 min, (iej) 16 min and (kel) 32 min, respectively.
(A colour version of this figure can be viewed online.)
1. Electronics Modules in CMOS for
New Functions in Deep-Brain Stimulation (DBS)
|
|
Sponsored by FAPESP under the grant 2019/05248-7 Principal Investigator (PI, USP/EESC): João Paulo Carmo Started
on May 2020 Total
budget: (a)
4-channels oscilloscope and 2-channels AWG: 26292 BRL. (b) Microelectronics
fabrication and packaging: 15797.47 USD. (c)
Total for consumables: 8999.99 BRL. |
Summary
|
This Project aims the development, in CMOS technology, ACQUISITION,
SIGNAL PROCESSING, CONTROL AND STIMULATION MODULES for microsystems in new
therapies for deep brain stimulation, with specific objectives being the
development of microelectronics modules that enable: - the simultaneous acquisition of a large number of channels; - multi-channel neurostimulation; - integration of the radio frequency system and energy management in the
proposed microsystem, using commercial components for these two functions
(COTS - Commercial off-the-shelf); - enhance future implantation, making it less invasive and minimizing
discomfort; - enhance the reuse of modules and application in future active
implantable tips. Figure illustrates a block diagram containing the acquisition,
neurostimulator and control modules to be developed in CMOS technology for
integration in a microsystem for new DBS therapies. These modules will be
developed using low-power (or low-power) CMOS technology so as not to
compromise future implantation in the brain. Technologies appropriate to the
proposed objectives are 0.18mm CMOS technogies and 65nm CMOS both
from TSMC. Therefore, the main contribution to the Electronic Instrumentation
and Microsystems area is the project of acquisition and neurostimulation
modules for integration in microsystems for new DBS therapies. The great
advantage of the proposed microsystem project is the possibility of using the
same CMOS technology to have the complete DBS system, interface, energy
management and RF transceiver. |
Chip layout and photo
of the seventh generation of a microdevice on 0.18mm CMOS
|
This microdevice were fabricated to the foundry and will
comprise electronic modules for
Deep-Brain Stimulation (DBS), e.g., Low-Noise Amplifier, Notch-Filter,
Howland Charge-Pump, 8-bits ADC, Operational Amplifier, and a random number
generators. |
Chip layout and photo
of the seventh generation of a microdevice on 65nm
CMOS
Finished Projects and Prototypes
12. Portable Platform and Sensors for Detection
of Hepatitis-C
11. New functions on endoscopy
|
|
Sponsored by the Brazilian National Council of Scientific Research
(CNPq) under the grant 400110/2014-8 Principal Investigator (PI, USP/EESC): João Paulo Carmo Started
on July 2014 and finishes on July 2017 Total
budget: 122 kR$ |
Summary (in Portuguese)
|
This Project aims the development of new technologies and technologies
based on photonics for the integration of new functions in endoscopic
capsules, namely: - Narrow Band Imaging (NBI); - Confocal endomicroscopy (ENM); - Photodynamic therapy (PDT); - and 3-D (R3D) reconstruction of
the gastrointestinal mucosa. The social-economic impacts for Brazil are many, namely: - make new preventive diagnostic
techniques available that drastically reduce the number of expensive and / or
ineffective surgeries against the disease as a result of late diagnoses; - boost the market related to the
area of medical instrumentation and healthcare; - training of new staff (doctors,
technical and auxiliary staff) linked to health care; - investment costs with techniques
will have a medium / long term impact on substantial savings for the state
and citizens, as it will result (1) in the reduction of costs in medical
resources (improvement of the working conditions of the medical and auxiliary
personal work, reduction in the number of hospital admissions, reduction of
processed consumable waste, rationalization in the allocation of physical
resources, reduction of the bill related to energy consumption, etc.), (2)
professional absenteeism, ( 3) in the reduction of avoidable mortality, and
(4) spread to an increasing number of patients at an ever lower cost.. |
Fabricated prototypes
Sixth generation of a microdevice on 0.7mm CMOS process from
the on-semiconductor
|
This microdevice is composed by several photonic, analog and digital IP blocks. One of the blocks comprises a second generation of an optical transceiver for integration on minimum invasive medical devices to provide intra-body communication (red-shading in the figure). This transceiver is intended to exchange data at high speed from/to the interior of the human body and was designed to operate near the 850nm. The selection behind the 850nm was due to low attenuation of the living tissue and to get high data-rates without the need to use radiofrequency waves at high frequencies, avoiding the high losses by the tissue. The transceiver is fully integrated in CMOS, excepting the optical source. Inside is composed by a digital modem with a parallel interface with the controller of the medical devices or with a microcontroller. It is also composed by two PLLs, one to provide the desired and programmable clock and other to recover the clock from the received bitstream. The difference to the first generation (also in 0.7μm from the on-semiconductor) is the first PLL that was fully converted to allow the frequency selection and the current-to-voltage converter from the CMOS photodiode used to convert light into a photocurrent. Moreover, this block also allows the selection between the option to encoding or not the bitstream to send, in order to allow an easier clock recovery in the receiver. Another photonic
block within this microdevice is a sensor to accurately read the angle of
incidence of the light (yellow-shading). This block was designed to be
integrated on photovoltaic sub-modules to track the sun, thus allowing the
photovoltaic panels to move to the optimal position to maximize the converted
power. The photodetectors generate photocurrents proportional to the angle of
incidence thanks to the metal and polysilicon structures placed on top. These
photocurrents are converted on output voltages for sampling and
analog-to-digital conversion. The microdevice contains two sets of the former
photodetectors, each one to a specific type of readout circuit. The
architecture of one of the read-out circuits is based on the 3T-pixel, used
on CMOS images, while the other is based on current mirroring combined with
transmission ports for reading and for resetting. This microdevice also
contains a sample-and-hold circuit for interfacing with this photonic block.
The photodetectors used in this photonic block are n+/p sub photodiodes,
which are also individually provided for testing, for characterization
purposes and to use with other external circuits (pink-shaded). A low-power
operational amplifier was also fabricated to prototype amperometric circuits
for application with functionalized electrodes for the detection of the
C-hepatitis (light green-shading). The target is the future integration of a
developed prototype of a portable laboratory platform. A
voltage-controlled-oscillator (VCO) was also fabricated to generate signals
in the range [370, 510] MHz, fully covering the LPD433 band (low power device
433 MHz) whose frequencies are located
within the range [433.050, 434.790] MHz (dark green-shading).
This VCO was de-signed to test radio-frequency circuits working in this band.
It was also
fabricated a circuit to test a new and ESD-robust anti-latch structure to
interface digital circuits fabricated in this technology with external
circuits (wine-colored-shading). Finally, the
blue-shading in the
figure highlights a
second generation of
an arbitrary/programmable PN
sequence generator. This
generator is intended
to be used
on chemical sensors based on radio-frequency. |
Fifth generation of a microdevice on 0.7mm CMOS process from
the on-semiconductor
|
Photo of the full microdevice on on-semiconductor 0.7mm CMOS. This chip contains 2 biopotencial amplifiers, 16 photodetectors with diffraction structures on top and respective readout electronics, 1 optical transceiver for intra-corporal communications, 1 radiofrequency modulator, 1 arbitrary/programmable PN sequence generator up to 28 bits, 1 digital controllable PWM generator for photodynamic therapy (PDT), few test structures of photodetectors, 1 plasmonic photodetector and 1 test structures of a 3T active-pixel. |
Photodynamic Therapy
(PDT)
Narrow-Band Imaging
(NBI)
10. Technologies
for photonic devices on advanced optical transceivers
|
Principal Investigator (PI): Giovanni Beninca de Farias (CPqD). Project name (in Portuguese): Tecnologias de dispositivos fotónicos para transceptores ópticos avançados. Project name (in English): Technologies for photonic devices on advanced optical transceivers. Partners: Centro de Pesquisa em Telecomunicações (CPqD), School of Engineering of São Carlos - EESC/USP. Project members: João Paulo Carmo (EESC/USP), Jacklyn Dias Reis (CPqD). Sponsored by FAPESP: FAPESP Regular. Total budget: 200kR$. |
9. Solar
Photovoltaic Sub-Modules (PV-Submodule)
Second prototype of
PV-Submodule on 0.7mm CMOS process from
the on-semiconductor
Fourth generation of a
microdevice on 0.18mm CMOS process from
the TSMC foundry
|
Photo of the complete CMOS microdevice on TSMC 0.18mm CMOS. This chip contains a matrix of 4´4 N+/P‑sub photodiodes and respective readout circuit; it must be noted that on few photodiodes metal structures were placed above in order to provide optical diffraction and/or blocking; few metal structures will allow to measure angles and wavelenght resolving; a fully programable PN sequence generator, by (right part) a PLL for generating clock signals within the range 83-132MHz, and by (top part) a RF switch for the generation of RF pulses with durations downwards to 2ns; floating N+/P‑sub photodiodes, P+/N-well photodiodes, and a SPAD (Single-Photon Avalanche Photodiodes); a P+/N-well and SPAD photodiodes with readout electronics; floating N+/P‑sub photodiodes, P+/N-well photodiodes, and SPAD (Single-Photon Avalanche Photodiodes). |
8. High-Resolution
Image Sensor in CMOS Technology for Minimally Invasive Surgeries
|
|
Sponsored by the Portuguese Foundation for Science and Technology
(FCT) under the grant PTDC/EEA-ELC/109936/2009 Principal Investigator (PI): João Paulo Carmo; Research Members: Rui Pedro
Rocha (BOSCH), João Filipe Ribeiro (DEI/UMinho), José Higino Correia
(DEI/UMinho). Started
on February 2011 and finishes on August 2014 Total
budget: 140 k€ |
Summary
|
Endoscopes are used to examine the inside of live organisms, as well as cavities in technical structures, by means of image acquisition and presentation. Originally developed for medical diagnostics in humans, today they generally are used as advanced tools in the visual examination of difficult-to-enter cavities. A rigid endoscope is composed of a direct-sight endoscope with glass lenses and adjustable ocular, a lateral connection for a light conductor, an adaptor that allows focusing, and a CCD camera head. This configuration is also called a video endoscope; it has been in use for more than a quarter century. But it exhibits three significant disadvantages: lack of image quality, the need for sterilization and high manufacturing cost [1]. Extending further this concept, the ‘camera in a pill’ is one recent development that is generating considerable interest. Until recently, only the proximal and the distal portions of the gastrointestinal tract were easily visible using available technology [2]. The twenty feet or so of small intestine in between these two portions was essentially unreachable. Since news about wireless capsule endoscopy first began to appear about seven years ago when a paper in Nature described the development of a disposable capsule that takes pictures during its course through the digestive tract after being swallowed [2,3]. Proprietary software loaded onto a desktop or laptop computer enables interpretation and storage of this data and generation of a report [2]. Improvements in the growing digital imaging world continue to be made with two main image sensor technologies: charge coupled devices (CCD) and CMOS sensors. The continuous advances in CMOS technology for processors and DRAMs have made CMOS sensor arrays a viable alternative to the popular CCD sensors. New technologies provide the potential for integrating a significant amount of VLSI electronics into a single chip, greatly reducing the cost, power consumption and size of the cameras. This advantage is especially important for implementing full image systems requiring processing such as digital cameras and computational sensors [4]. Beyond this, CMOS processes have the advantage to integrate the sensing element and the processing electronics on the same substrate [5]. Moreover, in contrast to CCD image converters, sensor chips based on energy-saving CMOS technology offer several advantages for video endoscopes. CMOS circuits are highly immune to magnetic fields generated by medical RF equipment and therefore do not need shielding. A single supply voltage of 1.8 to 2.5 V for light loads is sufficient. All of this simplifies both surgery help and capsule camera endoscope designs. This is of big importance, because in capsule camera applications, CMOS image sensors chip requires much less current than a CCD for comparable image quality and can be integrated together with advanced ASIC video transmitters and white-light-emitting diode light sources. Synchronous switching of the light source, CMOS chip and ASIC transmitter further minimised power consumption so that the battery power was sufficient for the duration of the capsule’s trip through the gastrointestinal tract [2]. The surface interface of a
classical photodiode is the main contributor to dark current. Moreover, this
process is the one that leads to an acceptable compromise between the fill
factor and the charge transfer capacity of transistors [7]. The use of the
three-transistor pixel architecture in conjunction with a p-substrate
photodiode is the one that helps to achieve the biggest possible fill factor.
However, the use of other potentially architectures that can leads to a best
pixel behavior will not be discarded. The CMOS image sensors currently
present on the market have average performances such as a dynamic range of
60dB [8] and a signal-to-noise ratio about 70dB [9]. The use of a logarithmic
photoreceptor architecture can push the dynamic range up to 120dB or more
[10], which is much greater than the 80dB found in a good CCD image sensor
[11]. Microlenses are usable for this application because they collect light
over the entire pixel surface and concentrate it on the light-sensitive
barrier layer. Basically, the optical system is based on an objective with
two opens (left and right) and a CMOS array composed by an array of
microlenses, making the rays from each open to arrive into the sensor plane
at a defined angle. The microlenses focus the rays in the respective pixels. |
|
|
[1] |
Endoscopes use CMOS image sensors,
http://www.vision-systems.com/display_article/301875/19/none/none/SPFEA/Endoscopes-use-CMOS-image-sensors. |
|
[2] |
W. Qureshi, "Current and
future applications of the capsule camera", Nature Reviews Drug
Discovery, Vol. 3, pp. 447-480, May 2004. |
|
[3] |
G. Iddan, et al, "Wireless
capsule endoscopy", Nature, Vol. 405, pg. 417, May 2000. |
|
[4] |
J. Dubois, et al, "A 10000
fps CMOS sensor with massively parallel image processing", IEEE Journal of
Solid-State Circuits, Vol. 43, No. 3, pp.706-717, March 2008. |
|
[5] |
S. Heini, et al, "A new
fotoFET for monolithic active pixel sensors using CMOS submicron
technology", in Proc. 13th IEEE International Conference on Electronics
Circuits, and Systems, |
|
[6] |
W. Yang, et al, "An
integrated 800×600 CMOS imaging system", in Proc. Solid-State Circuits
Conference 1999 (ISSCC), |
|
[7] |
H. Rhodes, et al, "CMOS imager
technology shrinks and image performance", in Proc. IEEE Workshop on
Microelectronics and Electron Devices, pp. 7-18, 2004. |
|
[8] |
T. Yamada, et al, "A
140dB-dynamic-range MOS image sensor with in-pixel multiple-exposure
synthesis", in Proc. Solid-State Circuits Conference 2008 (ISSCC), San
Francisco, USA, pp. 50-52, February 2008. |
|
[9] |
E. Labonne, et al, "A 100dB
dynamic range CMOS image sensor with globel shutter", in
Proc. 13th IEEE International Conference on Electronics Circuits, and Systems,
Nice, France, pp. 1133-1136, December 2006. |
|
[10] |
M. Loose, et al, "A
self-calibrating single-chip CMOS camera with logarithmic response",
IEEE Journal of Solid-State Circuits, Vol. 36, No. 4, pp. 586-596, April
2001. |
|
[11] |
T. Yamada, et al, "Trench
CCD image sensor", IEEE Transactions on Consumer Electronics, Vol. 35,
No. 3, pp. 360-367, August 1989. |
|
[12] |
G. Agranov, et al, “Crosstalk and
microlens study in a color CMOS image sensor”, IEEE Transactions Electron
Devices, Vol. 50, No. 1, pp. 4–11, January 2003. |
The research plan and experimental methods
include (but not limited to):
|
(1) |
Fabrication
of photodetectors |
|
|
The photodetector is a n+/p-epilayer junction photodiode fabricated in
a CMOS process, because it provides the best possible quantum efficiency in
the desired spectral range of photodiodes available in a CMOS process, at the
same time yielding the highest possible fill factor, since a deep n-well is
not required for every pixels. A shared-pixel architecture to be used will be
the 1.5-transistor concept to maximise the fill factor. The pixel consists of
transfer transistors (M1, 2, 3, & 4), reset transistor (M5), amplifying
transistor (M6) and four photodiodes. By using M5 as a row select transistor,
a conventional row select transistor becomes unnecessary. In this
architecture, the reset transistor (M5) and the amplifying transistor (M6)
are shared by four photodiodes (PD1, 2, 3 & 4). As a result, a minimum
number of transistors/pixel is realized. In the selected CMOS process, the junction depth of the photodiodes
is fully defined and cannot be altered. However, the quantum efficiency can
be improved by a suitable arrangement of dielectric layers on top of the
photodiode surface. These act as a thin-film interference filter and
influence the optical transmittance for each wavelength independent of the
CMOS process. In the CMOS process (taking the Austria Semiconductor 0.35 um CMOS
process as example), there are three major dielectric layers above the
photodiode p-n junction. The first oxide (i.e., boron phosphor silicate glass
or BPSG is the metal1-poly oxide) is above the photodiode, and its thickness
is typically 645 nm. The second major oxide (SiO2) comprises those between the first and
third metal layers, and has a total (typical) thickness of 2 um. The last
silicon nitride layer (i.e., the overlayer) used for scratch protection is
1000 nm thick. Previous simulations shown that the photodiode structure without two
of the three dielectric layers above the pn-junction, will provide the best
possible quantum efficiency in the desired spectral range of photodiode
structures available in a CMOS process.
Thus, this requires the removal of the SiO2 oxide and the silicon nitride layers. This
removable process will be performed using the Reactive Ion Etching (RIE)
method. |
|
|
|
(2) |
Fabrication
of Microlenses |
|
|
The AZ4562 photoresist will be combined with lithography in such a way
that it will be produced arrays containing a million or more MLs of good
optical quality in just a few minutes. The first stage is to coat a glass substrate with a layer of
photoresist. This then exposed to a pattern of an array of circular masks and
developed to form a series of cylindrical islands. These are then heated in
an oven or on a hotplate until the resist melts and surface tension draws
them into the shape of MLs. These then will be used as they stand or may be
processed further. In order to calculate the thickness of resist necessary to produce
the MLs of a given focal length, it is assumed that the MLs will have the
form of a cap of a sphere and that the volume of the resist remains constant. Thus, the paraxial focal length (the 'f' number) 'f' of a MLs
consisting of a single spherical surface of radius 'R' in a medium of refractive index 'n' is given by f=R/(n-1),
and the height 'h' of the surface
undulation of a MLs with an aperture radius 'r' is h=R-(R2 - r2)1/2. The volume of a cylinder of resist is T.PI. r2, where 'T' is the thickness before melting, and the volume of the MLs is
1/3.PI. h2(3R-h). It therefore follows that the
necessary thickness is given by T=h/6[3+(h/r)2]. Good-quality lenses will be fabricated with relatively crude
lithography although greater consistency and reliability is achieved only if
the process is carried out efficiently. The precise form of the MLs, and hence their focal properties, are
determined by the effects of surface tension. In particular the contact angle
of the softened resist with the surface of the substrat will strongly
influence the shape of the MLs. In order to produce a good-quality MLs of a given specification, it
is necessary either to exercise some control over the process or modify the
MLs after they have been formed. |
|
|
|
(3) |
FabriCAtion
of optical filters |
|
|
The thin film deposition of a given material in a transparent
substract, such as the glass or the quartz, is a common method employed in the
fabrication of optical filters. It will be used dielectric materials instead
of metallic materials for the depositions of the films, because the formers
don't absorb the light, thus, the losses inside the material are negligible. The amount in the amplitude of light that is reflected in a border
between two transparent mediums (the light goes from 1 to 2) is given by kr=(1-N21)/(1+N21), where 'N21' is
the ratio of refractive indexes of the second medium 'n2' to
the first medium 'n1'. The in the amplitude of light that is
transmitted to the second medium is kt=1-kr. The fabrication of filters with a desired characteristic (low, high,
or band-pass around a given wavelength) is done by successively depositing
different materials in order to obtain a dielectric multilayer structure able
to achieve the desired characteristic. For each of the primary colours (Red, Green and Blue), the optical
filters will be designed to yield a narrow passband around the respective
wavelengths. The dielectrics containing in the optical filter will be composed of
a stack of (titanium dioxide) TiO2 and (silicon dioxide) SiO2 thin films (materials with high and low refractive index, in the
visible spectrum, about 3.0 and 1.5, respectively). These dielectric films
are also hard materials; thus it is extremely difficult, or even almost
impossible, to remove them from the substrate. SiO2 is a selected material because the
wavelength dependence of its refractive index for the spectral band between
480 and 700 nm is almost constant (1.465 to 1.457, respectively). TiO2 has been selected due to fabrication
constraints (the deposition process is well characterized). In order to have the best filtering/transmission results, the
multilayer will be structurally optimized together with the transmission
through the two dielectric layers on top of the p-n junctions provided by the
CMOS process used to fabricate the photodiodes. The optical filter will be postprocessed, by reactive plasma
sputtering deposition of thin films, on top of the photodetector using the
same mask that was used for the selection of the dielectric layers available
in the CMOS process. The filter fabrication starts with the deposition of a
TiO2 layer after completion
of the standard CMOS process, including the metalization and the etching of
the two oxide layers on top of the photodiode. Then, the eight subsequent
layers of SiO2 and
TiO2 are deposited
with the suitable thicknesses. Then, a commercially available passband
optical filter on top of the microsystem is used to block the nonvisible part
of the spectrum. |
|
|
|
(4) |
Readout
electronics |
|
|
The development of the readout electronics will met the target to reduce
the number of analog readout circuit as much as possible to increase the
overall noise performance of the CMOS image sensors |
|
|
Few prototypes
First generation of a microdevice on 0.7mm CMOS process from
the on-semiconductor
|
|
|
|
(a) |
(d) |
|
|
|
|
(b) |
(e) |
|
|
|
|
(c)
|
|
|
(f) |
(g) |
(a) A
photograph showing the CMOS microdevice prototype,
which is composed by several photodetectors. SEM
images of a selected sample taken from few arrays of microlenses
(b) before and (c) after the thermal reflow. (d) SEMs
of few functional prototypes before (on top) and after (on middle) using the
reflow process, and few photographs (on bottom) with projections of few laser
beams after being spread from the impinging one by the array of microlenses. It is possible to note interference patterns
between beams, especially in the middle region. (e) (left) General and (right)
zoomed views of a microlenses array made of AZ4562 on
top of a TiO2 thin film after doing the thermal reflow (it must be
noted that the upper layer of the optical filters is made of this material and,
parallel the filtration purposes, has the additional function to support the microlenses array). (f) (top photo) on the left is an
optical photograph of the die and on the right is a zoom-in of the 24 um square
hotodiode (signalled with a
blue circle) with a ML already fabricated on top of it; (bottom photo) zoom-in
optical microscope photograph of the die placed on the PCB and with the wire
bonding already performed and (g) a more zoomed photography for a better view
of the post-reflow microlenses on top of the CMOS microdevice.
Second generation of a
microdevice on 0.7mm CMOS process from
the on-semiconductor
Full view of the layout and photo,
detailing the structure of a P+/N-well photodiode (top and cross-section
views).
|
(a) Full view of the packaged microdevice. (b) Full view of the microdevice area. (c) Detail showing two arrays of pixels, based on 1.5T- (left array) and 3T-architectures (right array), a P+/N-well photodiode (right top corner) and few test structures. (d) Detail showing an array of pixels, based on the 3T-architecture (the array), and P+/N-well photodiodes (right) and few test structures. (e) Readout electronics to access both arrays of pixels, and a P+/N-well photodiode with a smaller area than the previous ones (right bottom corner). (f) A detail of the readout electronics to access both arrays of pixels, and a P+/N-well photodiode with equal (right above the 6 internal PADs) and a smaller area than the previous ones (right bottom corner). |
|
|
|
|
(a) |
(b) |
|
|
|
|
(c) |
(d) |
|
|
|
|
(e) |
(f) |
Third generation of a microdevice on 0.7mm CMOS process from
the on-semiconductor
|
Photo of the full microdevice, and a SEM on right part showing the readout circuit and respective schematic above (and respective position within the microdevice in red on the microdevice photo). |
|
|
|
|
(a) |
(b) |
|
|
|
|
(c) |
(d) |
|
|
|
|
(e) |
(f) |
|
(a) Layout of the full microdevice showing on left an array (bottom) of P+/N-well photodiodes and respective readout circuit (the vertical column in the middle of layout), a N+/P-well photodiode and respective readout circuit (right bottom corner), a readout circuit for two P+/N-well photodiodes and respective readout circuit (right top corner), and a 2´2 matrix of P+/N-well photodiodes (left top corner). (b) Photo of the full microdevice showing on left an array (bottom) of P+/N-well photodiodes and respective readout circuit (the vertical column in the middle), a N+/P-well photodiode and respective readout circuit (right bottom corner), a readout circuit for two P+/N-well photodiodes and respective readout circuit (right top corner), and a 2´2 matrix of P+/N-well photodiodes (left top corner). (c) Top SEM of the full microdevice showing on left an array (bottom) of P+/N-well photodiodes and respective readout circuit (the vertical column in the middle), a N+/P-well photodiode and respective readout circuit (right bottom corner), a readout circuit for two P+/N-well photodiodes and respective readout circuit (right top corner), and a 2´2 matrix of P+/N-well photodiodes (left top corner). (d) Top SEM of the N+/P-well photodiode and respective readout circuit. (e) Layout of the N+/P-well photodiode and respective readout circuit. (f) Schematic SEM of the N+/P-well photodiode and respective readout circuit. |
Finished PhD Thesis
with direct relation to this project
|
(1) |
Rui Pedro
Leitão da Silva Rocha PhD in Leaders for the
Technology Industries (LTI), from the MIT-Portugal
Program, Engineering Design and Advanced Manufacturing (EDAM) focus area.
Started in September 2009. Finished on 11th November 2013. Thesis: Microlenses for optical
microsystems. Final accounting on 11th
November 2013: 6 papers published on journals and 5 submitted (under
evaluation yet) |
Main results (up to
now)
|
|
|
|
|
|
7. Flexible Energy
Harvesting Film for Stand-Alone Microsystems
|
|
Sponsored by the Portuguese Foundation for Science and Technology
(FCT) under the grant PTDC/EEA-ELC/114713/2009 Principal Investigator (PI): Luís Miguel Gonçalves (DEI/UMinho); Research
Members: João Paulo Carmo, Jorge Martins (DEM/UMinho), Francisco Brito
(DEM/UMinho), Pedro Alpuim (INL). Started
on February 2011 and finishes on August 2014 Total
budget: 83 k€ |
Summary
|
The increasing energy demand of
battery-powered wireless devices requires new energy scavenging systems, capable
of harvesting energy from environment when available and deliver it when
necessary. Energy scavenging is mainly based on thermoelectrics, vibration
and photovoltaic energy sources. In a photovoltaic scavenger, the output
current and power of the photovoltaic cells vary much as function of
illumination intensity and spectra and an energy-efficient electrical power
supply from this source is difficult to obtain under the strongly varying
conditions of illumination. On the other hand, the usual backup batteries
provide voltages which decrease during discharging of the battery. During
charging, the applied voltage should also be adapted to the evolution of the
electrical potential and the stored charge. An efficient energy scavenging
system is proposed to overcome these limitations. We propose a new approach
offering an autonomous power source: a flexible thin film device for
photovoltaic (PV) energy scavenging that integrates a solar cell, a lithium
battery and electronics for maximum power point tracking (MPPT) and battery
charge. This flexible thin film stack (see annexed figure) can be used in
sensing and monitoring applications, in particular human body applications.
The device is flexible enough to be applied on a curved surface like the
human body and supply energy for autonomous wireless microsystems, which can
also be integrated in the film. The film is composed of three parts: |
|
|
(1) |
A flexible thin-film photovoltaic
cell |
|
(2) |
A flexible rechargeable
solid-state lithium battery, fabricated by planar thin-film technology |
|
(3) |
A flexible surface-mount (SMD)
electronic circuit, for DC-DC conversion and power management that can also
include application electronics for monitoring purposes |
|
The inclusion of a thin-film
battery in the system adds the possibility of powering the device when light
is not available. Since many of wireless sensors are powered in a peak basis,
the battery can supply this current. Electronics and materials engineering
are involved in this project. Considering materials engineering, two main
research areas are considered: Thin-film Si photovoltaic cells and
solid-state lithium rechargeable batteries. The main key challenges in each
area are: |
|
|
(1) |
Fabrication of thin-film Si
photovoltaic cells on flexible substrates, with efficiency higher than 5% |
|
(2) |
Fabrication of flexible
solid-state rechargeable batteries with fast charging-time. A layered lithium
battery is proposed to decrease charge time and decrease capacity of
conventional solid-state lithium batteries |
|
(3) |
Electronic circuits to charge
battery with maximum efficiency using MPPT algorithm, considering the voltage
and current supplied by photovoltaic cell and provide power (in a peak basis)
and information about remaining battery charge to application electronics |
|
The target properties of this
device are: |
|
|
(1) |
Area of 10 cm2 |
|
(2) |
Thickness below |
|
(3) |
Unlimited number of bending
actions over a curved surface with a radius of |
|
(4) |
Highly flexible amorphous
silicon-based photovoltaic cells with AM 1.5 conversion efficiency above 5% and
fill factor of 0.65 |
|
(5) |
|
|
(6) |
Charge and discharge rates up to |
|
(7) |
Electrolyte materials with an
ionic conductivity of 2×10-6 S/cm at room temperature |
|
(8) |
Dynamic optimization of the
voltages and currents of photovoltaic cell harvesting and battery charging |
|
(9) |
Supply voltage available for the
electronic application 3.3 V |
|
(10) |
Electrical power conversion
efficiency of 90% in average from the solar cells to the battery and 95% from
battery to application electronics |
|
(11) |
Power management unit uses the
direct battery voltage |
|
(12) |
Charging time of full battery
capacity in less than 15 minutes, under direct sunlight |
Few prototypes
|
(a)
(b) |
(c)
(d) |
(a)
Electronic circuit, including MPPT step-up conversion, battery power management
and under/over voltage protection. Photographs of the FTF prototype: (b) just after
the PDMS application, (c) after the PV cell placement, and (d) showing the
maximum thickness of the FTF.
Few experimental
results
|
(a)
(b) |
(c)
(d) |
(a) Power
and efficiency of the FTF in a typical clear spring day at latitude of 41ºN. (b)
FTF versus a PV connected to battery (with and without diode). (c) FTF behaviour when supplying a variable load. (d) Power versus
the angle of curvatures imposed to the FTF.
Main results (up to
now)
|
|
|
6. RF CMOS Transceiver at 5.7 GHz on UMC CMOS 0.18 um
|
|
|
|
(a) |
(b) |
For the frequency of 5.7 GHz: (a) a photograph and (b) the schematic
block of the RF part of a wireless interface at 5.7 GHz presented.
5. RF CMOS Transceiver at 2.4 GHz on UMC CMOS 0.18 um
|
|
|
|
(a) |
(b) |
(a)
Photography and (b) block schematic of a RF CMOS transceiver at 2.4 GHz
especially designed for standalone wireless instruments in biomedical
applications.
4. Super-Regenerative Receiver at 433 MHz on AMI Semiconductor CMOS
0.7 um (formerly AMIS foundry, currently on-semiconductor)
|
|
|
|
(a) |
(b) |
The block diagram of the super regenerative receiver at 433 MHz and (b) a
die photography containing the first prototype of the super regenerative
receiver (the shaded area in green).
3. Thermoelectric
Energy Scavenging Microsystem
|
|
Sponsored by the Portuguese Foundation for Science and Technology (FCT)
under the grant PTDC/EEA-ENE/66855/2006 Principal Investigator (PI): José Higino Correia; Research Members: João Paulo Carmo (DEI/UMinho), Luís Miguel
Gonçalves (DEI/UMinho), Luís Rocha (DEI/UMinho/INL), Pedro Alpuim (INL), José
Augusto Afonso (DEI/UMinho). Started
on 2007 and finished on 2010 Total
budget: 120 k€ |
Summary
|
This project purposes a new
global approach for power generation, by associating a Seebeck thermogenerator with thin-film integrated batteries used as
energy reservoir in an ultra-low consumption power management device. The
expected original contributions are (but not limited to): |
|
|
(1) |
The fabrication of thermoelectric
microgenerators in a reproducible way with standard microsystem technologies |
|
(2) |
The use co-deposition of the
tellurides avoiding more complex and expensive systems (MOCVD) |
|
(3) |
To achieve a power density of
2-20 mW.cm-2, with a temperature gradient of 3-10 ºC |
|
(4) |
Require a minimum temperature
gradient of 3 ºC for powering wireless sensors |
|
(5) |
The integration of thermoelectric
microgenerators, microelectronics and solid-state battery in the same device |
|
(6) |
To use on-chip rechargeable
thin-film Li-ion battery and DC-DC converter |
|
(7) |
Use Bi2Te3
and Sb2Te3 materials for thermoelectric energy
harvesting rather than for conventional Peltier
cooling |
Few prototypes
|
(a)
|
|
(d) |
|
(b) |
(c) |
(e) |
(a) Photography of a n-type and p-type elements,
before deposition of top contacts (left) and a photography of a thermoelectric microconverter with eight pairs of thermoelectric elements,
fabricated with bottom contacts (right). (b) Cross-sectional SEM image of a microbattery prototype obtained by thin-films depositions.
(c) Developed thermoelectric devices in polyimide foil of 50 μm of thickness, using the co-evaporation technique
with the help of shadow masks. (d) Schematic of a simple steup-up
converter. (e) A photography of a charge-pump prototype, followed by a step-up,
both based on commercial off-the-shelf components.
Few experimental
results
|
(a) |
(b)
|
|
|
(c) |
(a) The power-factor of Bi2Te3
(top) and Sb2Te3 (bottom) thin films as a function of the
Te/Bi evaporation flow rate ratio, R,
and the respective curve fittings (solid lines). (b) SEM top view (left) and
cross-sectional (right) images of Bi2Te3 (top) Sb2Te3
(bottom) thin-films and the respective XRD analysis of Bi2Te3
(top plot) and Sb2Te3 (bottom plot) thin film analysis;
where it can be observed that in the both plots, the peaks agree well with the
respective power diffraction spectrums (doted lines and diamonds).
2. Wireless Biomedical Microsystem for Neural Implants
|
|
Sponsored by the Portuguese Foundation for Science and Technology
(FCT) under the grant PTDC/SAU-BEB/100392/2008 Principal Investigator (PI): José Higino Correia (DEI/UMinho coordinator)
; Research Members: João Paulo Carmo
(DEI/UMinho), Paulo Mateus Mendes (DEI/UMinho), Luís Jacinto (ICVS/UMinho),
Carlos Lima (DEI/UMinho), Nuno Dias (ICVS/UMinho). Started
on 2009 and finished on 2012 Total
budget: 180 k€ |
Summary
This project proposes to develop 3-D button-size
microelectrodes array with embedded signal processing electronics and a
wireless interface. Microelectrodes will be designed for insertion in any part
of cerebral cortex. They will integrate features as wireless communications and
can provide electrical recording and stimulation. Having 3-dimensional
recording capability by itself will permit the most complete mapping of local
circuitry in the brain. In order to fabricate thin-film electrode arrays
capable of recording from neurons in vivo on a chronic basis, we need a
microprobe structure that can be fabricated reproducibly and is electronically
stable in vivo.
Few prototypes
|
|
|
|
|
(a) |
(b) |
(c) |
A subject (a) with a conventional EEG system based on wires and (b) with
a wireless data acquisiton, logging and transmission
system prototype fixed with the help of an elastic band (the white strap) on
the thorax region. (c) Photography of na electrode
composed by a matrix made with 16 pyramidal microneedles,
coated by a thin layer of iridium oxide for electrical conduction purposes and
to aid the biocompatibility of these electrodes.
1. Development of
Integrated Systems for Smart Interiors
|
|
Sponsored by the Portuguese Foundation for Science and Technology (FCT)
under the grant MIT-Pt/EDAM-SI/0025/2008 in the context of the MIT-Portugal
R&D program Principal Investigator (PI): José Higino Correia (DEI/UMinho coordinator)
; Research Members: João Paulo Carmo
(DEI/UMinho), Paulo Mateus Mendes (DEI/UMinho), Alexandre Silva
(MIT-Portugal/DEI/UMinho), Rui Pedro Rocha (BOSCH). Started
on 2008 and finished on 2011 Total
budget: 198 k€ |
|
|
|
Summary
|
One of the current focuses, of
the engineering community, is the development of a new generation of high-performance
mechanical systems that have integrated sensing, diagnostics and control
capabilities while continuing to perform their intended functions. The
automotive industry, among many others, is keen on increasing the use of
these integrated systems in their vehicles with the aim of saving weight,
increasing the number of functions and reducing both the component and
assembly costs. To this end, this project seeks to develop smart devices and
materials for automotive interiors that incorporate sensor and actuator
capabilities for both conventional and new functions in terms of: safety,
comfort, performance, aesthetic and information processing. The design of
novel integrated systems for smart interiors, proposed in this project, is
extremely challenging and requires the development of scientific
understanding in several areas together with their practical application. To
address such a multidisciplinary topic, a consortium with several partners,
which provide complementary skills in the field of research, was specifically
chosen. The project participants comprise members from three Universities
together with four industrial partners. The three major industrial partners
are well established automotive suppliers in complementary areas for
automotive interiors: artificial leathers and soft tissues (TMG), technical
molded components (Iber-Oleff) and seat related components (Sunviauto). In
addition, a company that specializes in sensoring technologies (FiberSensing)
is also part of the consortium. The objectives of the work program are to
utilize and further develop existing experimental and computational
instruments and expertise available to the partners in order to provide the
knowledge necessary for the automotive industry to develop large-scale production
processes of integrated systems for smart interiors. Some of the key aspects
of the work include: |
|
|
(1) |
Embedding optical fibers and
other sensors in laminated structures for automotive interior applications and
trimming (fabrics, coatings or structural elements of automotive components).
The laminated structures are composed of layers, comprising textile and/or
polymeric materials |
|
(2) |
Development of optimized interfaces between humans and both electronic and mechanical devices. This includes the design and performance assessment of these interfaces |
|
(3) |
Design of new multi materials
solutions for car seats that incorporate intelligent features in terms of
safety and comfort |
|
(4) |
The main project output will take the form of a body of research knowledge based on validated computational decision support instruments and experimental test data, complemented by the results of validation of the prototypes. |
Few prototypes
|
|
|
|
|
(a) |
(b) |
(c) |
|
|
|
|
|
(d) |
(e) |
(f) |
(a) A set of four photographs showing cross-section views of the layers
configuration of the carrier and the supported FBG. (b) In a) the elastic knee with
the pressure buttons signaled with green ellipses. In b) the sensing part
attached to a standard elastic knee band and in c) a close-up of the PVC foil
with the embedded FBG signaled with a red circle. (c) A concept of a PVC
flexible carrier with FBG sensors for placing in the chest to measure the
cardio-respiratory frequency and a photography of the respective prototype,
where it is possible to observe the velco sections
for fixing the flexible structure (e.g., FBG+PVC carrier). (d) CAD model of the
volume which is the “mold” for embedding the FBG within the PDMS. The two fiber
holders are also illustrated. (e) It can also be observed a finished mold with
the two fiber holders already screwed to it. (f) A
photography of a prototype, composed by an FBG sensor embedded within a
flexible carrier in PDMS, after being removed out from the mold.
Few results
|
|
|
|
|
(a) |
(b) |
(c) |
|
|
|
(f)
|
|
(d) |
(e) |
(g) |
(a). Measured values for a walk @