MEDICAL INTELLIGENCE
Chip-Scale Sensor System Integration for Portable Health Monitoring
Nan M. Jokerst, PhD*,
Martin A. Brooke, PhD*,
Sang-Yeon Cho, PhD*, and
Allan B. Shang, MD
From the Departments of *Electrical and Computer Engineering, and Anesthesiology, Duke University, Durham, North Carolina.
Address correspondence and reprint requests to Nan M. Jokerst, PhD, Department of Electrical and Computer Engineering, PO Box 90291, Duke University, Durham, NC 27708-0291. Address email to nan.jokerst{at}duke.edu.
 |
Abstract
|
|---|
The revolution in integrated circuits over the past 50 yr has produced inexpensive computing and communications systems that are powerful and portable. The technologies for these integrated chip-scale sensing systems, which will be miniature, lightweight, and portable, are emerging with the integration of sensors with electronics, optical systems, micromachines, microfluidics, and the integration of chemical and biological materials (soft/wet material integration with traditional dry/hard semiconductor materials). Hence, we stand at a threshold for health monitoring technology that promises to provide wearable biochemical sensing systems that are comfortable, inauspicious, wireless, and battery-operated, yet that continuously monitor health status, and can transmit compressed data signals at regular intervals, or alarm conditions immediately.
In this paper, we explore recent results in chip-scale sensor integration technology for health monitoring. The development of inexpensive chip-scale biochemical optical sensors, such as microresonators, that are customizable for high sensitivity coupled with rapid prototyping will be discussed. Ground-breaking work in the integration of chip-scale optical systems to support these optical sensors will be highlighted, and the development of inexpensive Si complementary metal-oxide semiconductor circuitry (which makes up the vast majority of computational systems today) for signal processing and wireless communication with local receivers that lie directly on the chip-scale sensor head itself will be examined.
|
Introduction
|
|---|
Increasing the measurement and monitoring of critical physiological variables could have a tremendous effect upon health care from the hospital to the home setting for patients from infants to the elderly. For example, blood oxygenation is a critical physiological variable that indicates life, death, and disability. The ability to pervasively monitor oxygenation could have a profound effect in cardiac and pulmonary disease, asthma, and sudden infant death syndrome monitoring and diagnosis, and in elderly home care. In 1999, 720,000 deaths in the United States were directly related to cardiac disease, fully 30% of all deaths in the United States that year, with the majority (462,000 or 64%) due to sudden cardiac death (1). Congestive heart failure affects more than 3 million people in the United States, costing more than $10 billion in treatment annually (2). Chronic obstructive pulmonary disease affects an estimated 10 million adults in the United States, and costs approximately $32 billion annually (3). The thread that binds all of these diseases and syndromes together is that they all hinge on the failure of the cardiopulmonary system to effectively exchange gases and deliver oxygen to vital organs. All of these diseases can be monitored in at-risk patients, from the hospital to the home, using a pulse oximeter. Next generation pulse oxygenation systems may be so small, comfortable, and inexpensive that they may be worn pervasively and continuously to monitor for alarm conditions and for disease progression. These data could be transmitted without user intervention to a wireless station (cell phone, computer), and could be monitored by a central alarm alert center. Innovations such as these could have a transformative effect on at-risk populations and the elderly, producing critical early response warnings, improving health-care delivery through a quantitative disease progression database, and enabling the elderly to remain at home, improving quality of life, and saving health care funds in an ageing population. This is one example of the impact that applying state-of-the-art semiconductor and optical technology to the miniaturization and portability of sensing systems could have on revolutionizing health care.
Home monitoring is an emerging area of health care that will significantly affect the quality of life of ageing populations through the potential delay of entry into health care facilities. The elderly population of the United States and the entire world continues to grow at ever increasing rates. By 2025, the world population of individuals older than 65 yr is expected to be more than 800 million (4). By the year 2050, approximately 20% of the world’s population will be 65 yr or older, upwards of 2.5 billion individuals (5). The US population older than 65 yr is now around 35 million, and is expected to reach 55 million by 2020, and 80 million by 2040, comprising more than 20% of the total population (6). Although currently 13% of the total US population, the elderly consume one-third of health care spending and one-half of all physician time (7). In 2003, total health care expenditures were $1.7 trillion, more than 15% of the US gross domestic product (8). This amounts to more than $500 billion spent on health care for the elderly in 1 yr. Miniaturized, portable, integrated monitoring systems, again, may have a tremendous impact upon the availability and cost of such home health monitoring systems.
Health monitoring systems can be grossly partitioned into the functional subgroups that include the sensor, the signal processing, and the communications units. The identification and integration of appropriate, inexpensive sensors, the low power, compact signal processing that will enable portability, and a low power communication approach are all critical to the realization of chip-scale health monitoring systems. Sensing technology is under intense research investigation, and many types of sensors have been reported. However, many of these sensors are bulky, insensitive, and consume a great deal of power. Optical sensors, which are among the most sensitive and can be miniaturized, will be the focus of this paper. For health monitoring, we will discuss two types of optical sensors: noninvasive optical sensors, such as pulse oximeters, and fluidic sensors, such as microresonators, that use one to a few drops of human fluid input. This is a subset of optical monitoring systems; miniaturization and integration will also strongly impact the next generation of imaging and optical probe (fluorescence, Raman, etc.) technologies as well by significantly increasing portability and reducing power consumption. The goals of both areas of research are to leverage current engineering technology breakthroughs in semiconductors and in integration to create miniaturized, portable, inexpensive sensor systems for health monitoring.
|
NONINVASIVE OPTICAL MONITORING MICROSYSTEMS
|
|---|
Pulse oximeters measure arterial blood oxygenation, which are critical physiological data, since insufficient oxygen can quickly lead to irreversible brain and vital organ damage. Pulse oximetry is currently the standard of care for noninvasive arterial oxygen monitoring. Most current pulse oximeters are bulky, limit user mobility due to wires, have limited accuracy ranges, and have limited sensor sites that are often poorly perfused. The next generation of miniaturized, integrated pulse oximetry systems may be more accurate, wearable, wireless, and offer more sensor site options. In the future, there may be pulse oximetry systems that have optics, signal processing, wireless communication, and a battery, all integrated onto a sensor head that continuously monitors blood oxygenation, with alarms and data that are wirelessly transmitted.
By leveraging inexpensive micro- and nanotechnologies, it will be possible to create an accurate oxygenation monitoring system with wireless communication integrated into an unobtrusive, wearable package. Critical technical aspects of next generation pulse oximeters include the integration of low power, geometrically optimized optical emitters and photodetectors, dedicated low noise, low power signal processing, and wireless communication implemented in inexpensive Si complementary metal-oxide semiconductor (CMOS), and packaging that will enable continuous wear. By integrating the signal processing with the communication, the oxygenation can be calculated directly on the signal head, enabling data compression and intermittent transmission for data within the safe monitoring range, with the option for immediate alarm transmission if the data are outside of the acceptable range.
The integration of multiple light emitting diodes (LEDs) with geometrically optimized photodetectors may enable pulse oximeters that are smaller and use less power through improved collection efficiency. Figure 1 is a photomicrograph of an integrated optical emitter/photodetector pair that has been implemented in Si CMOS that constitutes first step in this direction. Figure 1a is a close-up of the integrated optoelectronic system, in which a thin film (1-µm-thick) AlGaAs/GaAs/AlGaAs LED (that emits in the near infrared) has been integrated into the center of a Si CMOS photodetector array (9). Analog emitter driver and photodetector receiver circuitry are implemented on the chip in Si CMOS, and the digital circuitry on the integrated circuit provides the signal processing capability. Figure 1b is a photomicrograph of the integrated circuit shown in Figure 1a on the back of an American coin (dime) to show the size of the integrated system. Figure 1c shows the LED (integrated onto the same chip shown in Fig. 1a) emitting in the red wavelength region, under power. The Si CMOS integrated circuit was fabricated through the MOSIS foundry, and is a standard Si CMOS integrated circuit. The advantage of using foundry Si CMOS is that the photodetector array geometrical format can be arbitrarily defined (which can be optimized for pulse oximetry applications, which Fig. 1 is not), and that the interconnections between the analog photodetector output signal and the receiver circuitry are short, and thus less prone to noise pick-up. Analog and digital signal processing for noise reduction (and signal to noise improvement) can be implemented on the chip, which is on the sensor head, and wireless communication has previously been implemented in Si CMOS.
View larger version (52K):
[in this window]
[in a new window]
|
Figure 1. Photomicrographs of integrated LED/Si PD/Si CMOS circuit microsystem: (a: left) near infrared light emitting diode in the center of Si PD array with integrated Si CMOS analog and digital signal processing circuitry; (b: center) integrated chip from (a) on the back of a dime (shows size); (c: right) Red light emitting diode integrated onto the same Si CMOS chip, emitting under power.
|
|
Integrated chip-scale sensor systems that integrate multiple types of function are newly emerging technologies. Mixed signal microsystems have traditionally integrated analog and digital electronics onto the same chip ("System on a Chip") or into a single package ("System in a Package"). New, "ultra" mixed signal systems may integrate optics/photonics, microfluidics, micro-electromechanical structures, and/or radio frequency (wireless) capabilities into the system, as well. It is this type of integration that enables sensors, such as electrical sensors or optical sensors, to be integrated or packaged with chip-scale control and signal processing electronics and communication (e.g., wireless) functions. One emerging technology that enables these ultramixed signal integrated systems is the heterogeneous integration of compound semiconductor devices (for optical and wireless functions) with foundry Si CMOS analog and digital circuitry.
|
OPTICAL MONITORING MICROSYSTEMS USING HUMAN FLUIDS
|
|---|
More invasive sensing, such as technologies that use human fluids as sensor inputs, offers biological and chemical sensing opportunities that are often not available in a noninvasive probe format. Chip-scale sensors have been demonstrated using a variety of sensing mechanisms, including electrochemical, thermal, optical, and mass-sensitive transducers. Optical sensors are characterized by their versatility, sensitivity, and potential for miniaturization (10–13). The discrimination of sensed analytes using optical index of refraction changes are commonly used in laboratory instruments such as surface plasmon resonance sensing systems. However, miniaturized optical sensors with high sensitivity are just emerging. Optical sensors such as high-quality microresonators are attractive candidates for realizing customizable sensing in a miniaturized format.
Microresonators, which include microrings and microdisks, are attractive optical sensors, since they have a large quality factor (Q factor) and compact size (14,15). These devices are becoming widely published for use as lasers (16) and filters (17), and have been demonstrated in typical inexpensive materials for chip-scale sensors: silica (18), polymers (19), and III–V compound semiconductor materials (20). These resonators are highly sensitive to optical index of refraction changes in the surrounding environment, and reports using these resonators for sensors as gyros (18), for pressure sensing (21), fluor signal concentration sensing (22), and as distributed Bragg reflectors (23) are emerging. Several researchers have reported microcavity sensors in various formats including mirroring, microdisk, and modified mirrors (14,24,25). Polymer microcavity sensors with surface modification and microcavities measuring glucose concentration in aqueous solutions (25) have been reported, and enhancement of a fluorescence signal was demonstrated using an integrated optical microcavity (26). Essentially, microresonator sensors sense index of refraction changes on the surface of the sensor, which causes a change in the resonator to waveguide coupling. These sensors are typically 20–200 µm in diameter in dinner plate (disk) or ring shapes that are about 1–3-µm-thick.
The microdisk sensor shown in Figure 2 is an orthogonal microresonator, since the input and output coupling regions are orthogonal (27). This microdisk consists of three parts: the input waveguide/coupling section, microdisk cavity (sensor portion), and output waveguide/coupling section. Part of the input beam is coupled into the microdisk by the input coupling section. This coupled optical signal is trapped in the microdisk at the wavelengths of resonance, and circulates in the disk section. Optical resonance occurs when the circulating beam satisfies the cavity resonant condition: the total phase shift per roundtrip of the optical beam in the resonator is equal to an integer multiple of 2[ ]. The output beam is at a maximum when the resonance condition is met for any particular wavelength. If the optical structure (refractive index) of the microdisk or of the surface above the microdisk is changed (for example, through a surface chemical or biological binding/hybridization sensing event), the resonance condition will be altered, thus changing the output power. By monitoring the output power variation in the output waveguide (for example, with an integrated photodetector), orthogonal microresonators can be used as optical sensors.
There are several practical issues involved in the design and implementation of microresonator sensors. First, as shown in Figure 2, the input waveguide and output waveguide of the microresonator sensors should be perpendicular. This reduces undesired coupling of the input beam into the output waveguide or external collection system (e.g., photodetector or optical fiber), which can occur if the input and output waveguides are parallel to one another. The orthogonal approach also eliminates additional bent output waveguides for exciting and collecting the optical beam into the cavity, so the total size of the sensor system, which includes the input/output waveguides, can be significantly reduced in comparison to sensors with nonorthogonal configurations. Second, the manufacturability of the microresonator sensor is important, since this dictates cost. Also, the ability to format arrays of microresonator sensors on a single substrate is important if discrimination through overlapping signatures is desired. If there are multiple microresonator sensors, each with a different surface customization, then the different sensor outputs, with different signatures for a single sample, can be signal-processed to more precisely determine the sample content.
Both the manufacturability and the ease of fabricating arrays of microresonators lead to vertically coupled microresonators (VCMRs). The distance between the input waveguide and the microresonator and between the output waveguide and the microresonator determines the coupling, and is typically submicron in distance. Lateral separation between these waveguides and the microresonators that are this small do not use typical, inexpensive, high-throughput photolithography tools but, rather, more expensive, lower throughput electron beam lithography. In contrast, if the input/output waveguides are stacked above or below the microresonator with a submicron spacer layer in between, then a VCMR is created, and standard photolithography can be used to fabricate inexpensive microresonators in array format. Using VCMRs, inexpensive, standard microelectronics fabrication can be used to fabricate the sensor arrays, thus leveraging the microelectronics industry for integrated optical sensing systems.
To integrate microresonator sensors into a self-contained chip-scale microsystem, the inputs and outputs of the microresonator must be well understood. The microresonator input is a single-mode laser input. The output power of the microdisk sensor is oscillatory as a function of wavelength (spectral output), and reaches maximum values when the resonance condition is satisfied (for multiple wavelengths). As the refractive index of the microresonator surface changes, the resonant condition changes, and the peak wavelengths shift. At any particular monitored wavelength, there is a change in output power (either increasing or decreasing, depending upon where the monitored wavelength lies on the oscillatory spectrum). Figure 3 shows the input and output of a linear (not orthogonal) microresonator sensor. Since microresonators are highly sensitive to index of refraction changes at the surface of the sensor, surface customization can be used to sensitize the sensor surface. Thus, DNA hybridization and antibody–antigen binding can be sensed on the microresonator surface for biological sensing, and chemically selective membranes can be used for selective chemical detection. The microdisk resonator shown in Figure 2 was implemented in polymer material on a Si substrate and demonstrated in operation as a glucose sensor.
View larger version (11K):
[in this window]
[in a new window]
|
Figure 3. Input (Ein) and output (Eout) of a microresonator sensor with a single input/output waveguide. The circulating power in the resonator is also shown in the upper right.
|
|
Optical sensors are attractive due to their high sensitivity and miniature size; however, the integration of chip-scale optical sensing systems dictates the implementation of an optical source, sensor, and photodetector. In the case of microresonator structures, waveguides (which confine the light) act as input/output to the sensor, and so a planar lightwave integrated circuit that combines both active and passive optical elements at the chip scale is an ideal, and emerging viable technical, solution. To realize this integration, the waveguide and passive devices are deposited or spin coated using traditional chemical vapor deposition (nitrides, dioxides) or spin coating (polymers) techniques. The active devices are thin film optoelectronic devices, in which the substrate has been removed, thus resulting in devices on the order of 1–5-µm-thick, which are ideal for chip-scale planar lightwave integrated circuit integration. The entire structure can also be designed for encapsulation, with only the sensor exposed to the sensed analyte.
The integration of thin film active optoelectronic devices (emitters, photodetectors) onto Si, Si CMOS (28), and FR-4 (29) substrates have been demonstrated using spin-coat planarization, thin film device bonding (also called heterogeneous integration), and spin coat/photolithographically defined passive waveguide structures. Thin film devices embedded in polymer waveguides on silicon (30), ceramic (31), and FR-4 (32) substrates have also been demonstrated. For example, thin film InGaAsP-based edge emitting laser, polymer waveguide, and independently optimized thin film InGaAs photodetector were all co-integrated onto a SiO2/Si substrate (33). Figure 4 illustrates that more than one thin film compound semiconductor device can be integrated onto a single chip-scale Si substrate. Figure 4 (left) is a photomicrograph of an integrated laser/waveguide/photodetector chip-scale system (approximately 4 mm x 1 mm) and the resultant data from that system (34). This integrated planar optical system consists of a thin film edge emitting laser, a polymer waveguide, and a thin film waveguide photodetector. The data to the right of Figure 4 show the output of the photodetector under three laser output conditions. These data show the photodetector dark current (when the laser is off), and the photodetector current when the laser is biased below threshold (very little light output from the laser) and above threshold (high light output from the laser). As expected, the photodetector current output tracks the laser operation, demonstrating that the laser and photodetector are optically connected by the waveguide. This system is a significant step toward chip-scale integrated optical sensing systems; to realize an optical sensing system, an optical sensor can be inserted into the waveguide section of this integrated substrate. The laser will act as the sensor optical source, and the photodetector can monitor the output of the sensor. Currently, our group is integrating a microresonator sensor into this system toward a complete chip-scale optical sensor system.
View larger version (30K):
[in this window]
[in a new window]
|
Figure 4. Photomicrograph and measured data from thin film edge emitting laser, polymer waveguide, and thin film photodetector all integrated onto a SiO2/Si substrate.
|
|
This integration means that highly sensitive optical sensing systems, previously addressed by external lasers and optics, can now be fully portable, and that many waveguide-based systems (sensor and other) that currently use an external laser will now be able to use an on-chip laser, freeing the system from the wall power plug and significantly enhancing system portability. In addition to the laser/waveguide/detector integrated chip-scale system and numerous waveguide/detector and optoelectronic device/Si CMOS circuit integrated systems, a polymer microring has been integrated with a thin film InGaAs photodetector, another step toward chip-scale optical sensing systems, as shown in the photomicrograph in Figure 5 (35). In this research, a thin film active photodetector was embedded in the output waveguide of the microresonator, as shown in Figure 5, to provide optical to electrical for the output, replacing external photodetectors with integrated photodetectors, thus eliminating external coupling from the waveguide to an external fiber, which creates optical loss in the system. As discussed, this photodetector monitors the output of the waveguide, which changes in optical intensity (per the oscillatory spectrum) when sensing occurs on the microresonator surface.
View larger version (33K):
[in this window]
[in a new window]
|
Figure 5. Planar polymer microring integrated with an thin film InGaAs photodetector embedded in the polymer output waveguide of the microresonator. The entire system has been on fabricated on a SiO2/Si substrate.
|
|
|
CONCLUSIONS
|
|---|
Next generation miniaturized, portable, inexpensive integrated sensing systems will have a profound effect on health monitoring. Technologies for fabricating and integrating sensors into chip-scale portable systems stand at a threshold that promises to provide wearable biochemical sensing systems that are comfortable, inauspicious, wireless, and battery-operated, yet that continuously monitor health status, and that can transmit compressed data signals at regular intervals, or alarm conditions immediately.
Recent research into chip-scale sensor integration for health monitoring includes the demonstration of small optical sensors, as well as the integration of the associated optical input and output necessary to drive these systems. Leveraging inexpensive Si CMOS circuitry fabrication technology and the circuitry itself for signal processing and wireless communication is key to the realization of inexpensive systems that have a high level of signal capture and signal processing capability for portable, miniature health monitoring systems.
|
Footnotes
|
|---|
Accepted for publication June 11, 2007.
Supported by the National Science Foundation, the Army Research Office (ARO MURI W911NF-05-1-0262) and the Defense Advanced Research Projects Agency.
|
REFERENCES
|
|---|
- Arcing D, Chugh S, Chakravarti A, Spooner P. Genomics in sudden cardiac death. Circulation. Available at: http://www.circresaha.org
- Jerant A, Azari R, Nesbitt T. Reducing the cost of frequent hospital admissions for congestive heart failure: a randomized trial of a home telecare intervention. Med Care 2001;39:1234–45[Web of Science][Medline]
- Mapel D, Chen J, George D, Halbert R. The cost of chronic obstructive pulmonary disease and its effects on managed care. Manag Care Interface 2004;17:61–6[Medline]
- http://www.merck.com/mrkshared/mmg/sec1/ch2/ch2b.jsp
- http://www.cdc.gov/ncidod/eid/vol7no2/strausbaugh.htm
- http://www.postgradmed.com/issues/2002/06_02/editorial_jun.htm
- http://www.agingresearch.org/press/news_test_022802.cfm
- http://www.reason.com/0411/fe.rb.mandatory.shtml
- Cross J, Buchanan B, Carastro L, Lopez-Lagunas A, Wang T, Jokerst N, Brooke M, Wills D, Ingram M. A single fiber bi-directional optical link using co-located emitters and detectors. IEEE Photonics Technol Lett 1996;8:1385–7
- Janata JM, Josowicz P, Van
 sek DM, Vaney D. Chemical sensors. Anal Chem 1998;70:179R - Wolfbeis OS. Fiber-optic chemical sensors and biosensors. Anal Chem 2000;72:81R–89R[Medline]
- Wolfbeis OS. Fiber-optic chemical sensors and biosensors. Anal Chem 2002;74:2663[Medline]
- Mizaikoff B. Mid-infrared fiberoptic sensors. Anal Chem 2003;75:258A–267A[Medline]
- Chao CY, Guo LJ. Biochemical sensors based on polymer microrings with sharp asymmetrical resonance. Appl Phys Lett 2003;83:1527–9
- Boyd RW, Heebner JE. Sensitive disk resonator photonic biosensor. Appl Opt 2001;40:5742–7
- McCall S, Levi A, Slusher R, Pearton S, Logan R. Whispering-gallery mode microdisk lasers. Appl Phys Lett 1992;60:289–99
- Little B, Chu S, Haus H, Foresi J, Lain J. Microring resonator channel dropping filters. IEEE J Lightwave Technol 1997;15:998–1005
- Suzuki K, Takiguchi K, Gotate K. Monolithically integrated resonator microoptic gyro on silicon planar lightwave circuit. IEEE J Lightwave Technol 2000;18:66–72
- Lee H, Park J, Ryoo H, Lee S, Hoan O, Lee E. Resonance characteristics of waveguide coupled polyimide microring resonator. Opt Mater 2002;21:535–41
- Gmachl C, Faist J, Capasso F, Sirtori C, Sivco DL, Cho AY. Long-wavelength (9.5–11.5 mm) microdisk quantum-cascade lasers. IEEE J Quantum Electron 1997;33:1567–73
- Brabander GD, Boyd J, Beheim G. Integrated optical ring resonator with micromechanical diaphragm for pressure sensing. IEEE Photonics Technol Lett 1994;6:671–3
- Blair S, Chen H. Resonant enhanced evanescent wave fluorescence biosensing with cylindrical optical cavities. Appl Opt 2001;40:570–82
- Hennig O, Mendes S, Fallahi M, Peyghambarian N. Chemical sensors based on the modification of a resonator cavity. Proc SPIE Int Soc Opt Engl 1999;3537:336–41
- Krioukov E, Klunder D, Driessen A, Greve J, Otto C. Sensor based on an integrated optical microcavity. Opt Lett 2002;27:512–14[Web of Science][Medline]
- Chao C, Guo L. Polymer microring resonators fabricated by nanoimprint technique. J Vac Sci Technol B 2002;20:2862–6
- Krioukov E, Klunder D, Driessen A, Greve J, Otto C. Integrated optical microcavities for enhanced evanescent-wave spectroscopy. Opt Lett 2002;27:1504–6[Web of Science][Medline]
- Cho SY, Jokerst N. A polymer microdisk photonic sensor integrated onto silicon. IEEE Photonics Technol Lett 2006;18:2096–8
- Jokerst N, Brooke M, Laskar J, Wills DS, Brown AS, Vrazel M, Jung S, Joo Y, Chang JJ. Microsystem optoelectronic integration for mixed multisignal systems. IEEE J Sel Top Quantum Electron 2000;6:1231–9
- Suzuki T, Nonaka T, Cho SY, Jokerst N. Tummala R. Embedded optical interconnections on printed wiring boards. In: Proceedings of the 53rd Electronic Components Technology Conference, New Orleans, LA, 2003:1153–9
- Cho SY, Hall JD, Chellappa A, Jokerst N, Brooke M. High speed optical interconnection using embedded PDs on electrical boards. In: Proceedings of the 53rd Electronic Components Technology Conference, New Orleans, LA, 2003:1046–52
- Huang Z, Ueno Y, Kaneko K, Jokerst N, Tanahashi S. Embedded optical interconnections using thin film ingaas msm photodetectors. IEE Electron Lett 2002;38:1708–9
- Suzuki T, Nonaka T, Ogawa N, Cho SY, Seo SW, Jokerst N. Embedded optical interconnections in printed wiring boards. In: Proceedings of the 54th Electronic Components and Technology Conference, Las Vegas, NV, 2004:1524–9
- Camperi-Ginestet C, Hargis M, Jokerst N, Allen M. Alignable epitaxial lift off of gaas materials with selective deposition using polyimide diaphragms. IEEE Photonics Technol Lett 1991;3:1123–6
- Seo SW, Cho SY, Jokerst N. A thin film laser, polymer waveguide, and thin film photodetector co-integrated onto a silicon substrate. IEEE Photonics Technol Lett 2005;17:2197–9
- Cho SY, Jokerst N. Polymer microrings integrated with thin film ingaas msm photodetectors for sensor-on-a-chip applications. In: Conference Proceedings - International Conference on Indium Phosphide and Related Materials, Princeton, NJ, 2006:WB1.2
|
Top
Abstract
Introduction
