Lear Home
  for Studies

People in
Useful Links
Visitor Info

CSU Homepage

ECE Homepage

OE Research

CSU Cleanroom

Biosensors and Microfluidics Program

(c) 2006 
Kevin Lear
All rights reserved

The CSU Optoelectronics Group led by Prof. Lear is pursuing research on both  vertical cavity and planar waveguide based optoelectronic devices and systems with applications to photonic biosensors and optical communications.



Planar Waveguide

Vertical Cavity




Optofluidic intracavity spectroscopy (OFIS)

Passive Fabry-Perot cavities integrated with a microfluidic channel are being investigated for use in cell detection and differentiation using a technique we call optofluidic intracavity spectroscopy (OFIS).  Some recent publications on this project are:

Fluidic intracavity laser diode (FILD) sensor

There is also an active cavity version of OFIS using fluid filled external cavity electrically driven laser diodes called the fluidic intracavity laser diode (FILD) sensor.  Get it? "filled"?  A recent publication on this project is:

Local evanescent array coupled (LEAC) waveguide biosensor

A very promising multianalyte immunoassay biosensor is being developed based on a new concept in waveguide sensors.  The sensor design allows tens to hundreds of biomolecular targets to be recognized by binding to different regions where they will be locally detected by evanescent coupling to an array of photodetectors.  The acronym, LEAC, is meant to evoke the notion of the bound targets controlling the amount of light "leaking" into each of the photodetector array elements. Near-field scanning optical microscopy (NSOM) has been a key tool for characterizing the response of optical fields to small islands of material patterned on waveguides to mimic bound biological layers.  This work is in collaboration with Prof. David Dandy in CSU's Chemical Engineering Department and Profs. David Grainger and Alan van Orden in CSU's Chemistry Department.  Some recent publications on this project are:

On-chip optical interconnects in commercial CMOS

Another project using planar waveguide technology is aimed at demonstrating on-chip optical clock distribution to meet the challenges of growing interconnect bottlenecks with each generation of high performance processor chips.  Taking an approach that can be fabricated in commercial CMOS makes this technology more likely to be adopted in the near term.  Some recent publications on this project are:

High speed VCSELs for chip-to-chip optical interconnects (C2OI)

VCSEL technology dominates the LAN and SAN fiber optic links and is also attractive for short distance optical communication between components such as between processor chips on the same board.  Prof. Lear has been researching high speed VCSELs for a decade including developing 21 GHz ones at Sandia.  As VCSELs are the most bandwidth limited components in short distance optical interconnects, this research in this project endeavors to raise the direct modulation bandwidth of VCSELs to 40Gb/s, and thus move short distance optical interconnects well past the capabilities of copper traces.  Some recent publications on this project are:

Non-linear effects in VCSEL based multimode fiber links

While VCSEL speeds define the bandwidth of optical interconnects at the board level, non-ideal properties of legacy multimode fiber are the biggest limit for higher data rate LAN and SAN applications, particularly as VCSELs get faster.  Associated studies in the CSU Optoelectronics Group have examined the impact of non-linearities in VCSELs, specifically the pattern dependence of beam divergence, on multimode fiber links.  Some recent publications on this project are:

Fluorescence based fiber optic sensors

Research in the area of engineering optoelectronics instrumentation for fiber optic sensors is being performed in collaboration with Prof. Reardon of CSU's Chemical Engineering Department.  Prof. Reardon is investigating the use of genetically engineered bacteria or their enzymes to metabolize target pesticides that can be sensed through fluorescence.  The Optoelectronic Research Group's major role in this project is to develop high sensitivity, field-portable instrumentation and related system models for the fiber optic sensors.  The effort will accelerate in the fall of 2006 due to a recent NSF award.

Research Highlights

Record low thermal resistances show dramatic impact on VCSEL modulation bandwidth

Extending plated copper heatsinks over the mesa side wall on top-emitting VCSELs reduces the thermal resistance by 29% compared to identical devices where the copper plating is restricted to the top of the mesa, indicating the importance of lateral heat transport in mirror stacks.  Devices with 9 micron diameter oxide apertures had thermal resistances as low as 1.0 C/mW, half the best previously published results.  The reduction in thermal resistance with sidewall heatsinks increased the maximum power of the test lasers laser bandwidth by 131% with an associated 40% increase in bandwidth.  Details are available in a paper to appear in Photonics Technology Letters.  This work is sponsored by DARPA's C2OI program.

Plot of VCSEL Rth vs. diameter

SEM images of copper plated VCSELs

NSOM experiments continue to validate LEAC biosensor concept

A progression of experiments are validating expected phenomena underlying concepts for a new type of biosensor.  The local, evanescent, array-coupled (LEAC) biosensor is a label-free immunoassay device similar to existing ones that use the binding of target antigens to antibodies on a waveguide.  When anitgens bind, i.e. attach, to the sensitized surface, their presence alters the optical fields in the waveguide. While others then use observations of the optical intensity or phase measured at the end of the waveguide, the LEAC biosensor detects the shift in the confined mode locally by sensing the change in evanescent coupling to a detector buried under the waveguide.  The LEAC acronym is motivated by the image of the antigens controlling the "leakage" of light to the detector.  End-sensing allows only a single sensing region per waveguide while local sensing enables detection of many antigens using tens or hundreds of different antibody patches along the waveguide with a corresponding detector opposite each patch.  Prior work on a proof-of -concept device using an increased thickness of silicon nitride (n=1.8) as a pseudo-antigen layer appeared in a Photonics Technology Letters paper.  More recently, NSOM measurements of the evanescent field along a waveguide with a 100 nm thick, 10 micron, long low index, photoresist feature agree with BPM simulations providing further support that the LEAC waveguide sensor concept is valid.  This research program is supported by NIH. 

Cross-section, simulated, and measured fields of proof-of-concept LEAC structure

Monolithic optical interconnect with high responsivity polysilicon detectors fabricated in commercial CMOS

Optical components including waveguides and photodetectors for on-chip optical interconnects have been successfully fabricated in a commercial CMOS process through an industrial collaborator.  The technology is initially designed for clock distribution in high end CMOS processor chips. As seen in the photograph below, light is launched in a waveguide at the edge of the chip that is routed to multiple receiver nodes via an H-tree structure.  Leaky-mode coupling with the highest effective absorption coefficient of silicon waveguide detectors measured to date (0.56 dB/um) has been achieved giving good estimated responsivities of 0.34 A/W in a 10 micron long detector.  A chip with electronics for measuring the skew in the optical clock distribution system is currently being fabricated. This work is supported by NSF and is part of their GOALI programs.

Photo of lit-up on-chip waveguide H-treeLight and Dark current vs. voltage for polysilicon photodiode

Microfluidic Fabry-Perot interferometer differentiates cells

A passive cavity approach to performing optofluidic intracavity spectroscopy simplifies the sensor system relative to previous external cavity laser implementations.  The device is based on the treatment of cells in an aqueous environment as lenses due to their higher index and curved interface with the fluid.  Paul Gourley and colleagues at Sandia National Laboratories demonstrated that placing cells in the cavity of a photopumped VCSEL generated unique spectral features such as the number, spacing, and offset of multimode spectra that could be used to distininguish different types of cells.  The CSU optoelectronics group subsequently demonstrated an electrically driven laser diode version.  However, the active cavity sensors place relatively high demands on the laser to overcome the loss associated with the fluid filled external cavity, and to integrate the laser diode with the microfluidics system.  These challenges motivated the passive cavity system that relies on the transmission spectra of a cavity taken using an external LED as the continuum light source.  This system has been used to obtain spectra of red and white blood cells, yeast, and polystyrene spheres as discussed in a paper to appear in the IEEE Sensors Journal.  The spectra of red blood cells in particular were found to be well correlated and allowed their quantitative differentiation with respect to  white blood cells.  Thus, this technology may be appropriate applications such as unstained whole blood counts in a lab-on-a-chip environment.  This work was supported by DARPA's university optoelectronic centers program.

Channel and spectrum of passive cavity OFIS sensorspectra of red blood cells in microfluidic Fabry-Perot