MSD1 Senior Design Project- Oxygen Gas Sensor

Download Report

Transcript MSD1 Senior Design Project- Oxygen Gas Sensor

MSD1 Senior Design ProjectOxygen Gas Sensor
P09051
Samuel Shin
Jeremy Goodman
Sponsor: RIT uE & EE department
Project Guide: Professor Slack
Agenda






Project description
High Level Customer Needs/ Eng Specs
Concept Description & Rationale
System Architecture
High Risk Assessment
Detailed Assembly
◦ Emitter and Receiver Circuit
◦ Photodiode Fabrication
Testing Results
 Future Plans

Project Description
Oxygen gas detection via fluorescence quenching.
 Based on Tris-Ruthenium[II](dichloride) material
incorporated in an oxygen-permeable polymer

◦ Responds to gaseous %Oxygen which changes
fluorescent intensity and lifetime
◦ Higher O2 conc = decreased intensity and lifetime

Method has been researched and is widely used
◦ Expensive
◦ Equipment not readily available to everyday user

Plan is to design a complete cost & size- efficient
sensor system for the measurement of % Oxygen
High Level Customer Needs / Eng Specs

Provide consistent measurement results
◦ LED pulse width at 100ms
◦ Entering wavelength at 455nm

Cost and size-effective
◦ Commercially available LED source
◦ Standard electronic components for signal
conditioning
◦ Low-cost, high performance optical filters
◦ RIT SMFL designed/built photodetector
◦ Ru(dpp) polymer created in RIT Chem dept.
Concept Description/ Rationale
Incorporate the
entire system inside
a light-tight box
 Inject fixed amounts
of nitrogen and
oxygen to exhibit an
environment with
fixed %Oxygen

System Architecture
Input Signal (100ms pulse width from function generator)
LED Pulsing Circuit (455nm)
Ru(dpp) Thin Film (fluorescent material) – emitting wavelength of 613nm
Optical & Signal Conditioning
Amplified Signal in Oscilloscope (I or V vs. Time)
High Risk Assessment

Still a proof of concept
◦ Design will have to be modified to match needs
 Unclear Parameters will exist
 Where noise is coming from, etc

Materials
◦ Creating Ru(dpp) polymer has to be done with help
from a faculty member

Funding
◦ Assembly of chamber, gas canisters needed.
◦ Difficult to obtain funds
Final Results- LED Emitter Circuit

Circuit assembled to
exhibit a steady
source of LED light,
in a set fixed pulse.
◦ Used a power
PMOSFET

Completed assembly
using vectoboard
and soldering
components.
V2
5v dc
0
M1
MbreakP
V
V1 = 5
V2 = 1
TD = 0
TR = 5n
TF = 5n
PW = 1m
PER = 10m
V1
R1
2.6
0
V
D1N4149
D1
I
0
Final Results- Receiver Circuit
Circuit assembled to receive
the light source and transfer
it into voltage output.
 Used photovoltaic amplifier
circuit configuration.
 Completed assembly using
vectoboard and soldering
components.
 Completed circuit
demonstration in lab, and
also with complete lighttight box.

◦ Used commercial photodiode
for test.
Photodiode Planning

Two Architectures – 4” n-type silicon
◦ Lateral (Finger) Diode
 Small Active Area
 Fast Response Time
◦ Planar Diode
 Large Active Area
 Slow Response Time
 Tunable Junction Depth (Wavelength Selectable)

Fabricated in the RIT SMFL
Photodiode Design
PLANAR PHOTODIODE
P-Well Implant
Finger Contacts
N+ Implant
N-Type Wafer
N-Type Wafer
P+ Implant
LATERAL PHOTODIODE
Contact Ring
Photodiode Fabrication Process
Photodiode Fabrication Process
Photodiode Fabrication Process
Photodiode Results - Responsivity
Planar responsivity >2x greater than Lateral!
0.5
Planar
Lateral
PLANAR
0.4
(A/W)
Responsivity
Power (A/W)
↑ Active Area
Tuned Junction
↑ Responsivity
>2x
Difference
0.3
GREATER SIGNAL!
0.2
LATERAL
0.1
0
BUT
↑ DARK CURRENT!
400
500
600
700
800
Wavelength (nm)
Wavelength
900
1000
1100
Photodiode Results - Capacitance
Planar capacitance much higher than Lateral
4
2
x 10
1.8
Capacitance per unit Area (pF/cm2)
↑ Surface Area
↑ Capacitance
↑ Response Time
Planar
Lateral
PLANAR
1.6
1.4
1.2
SLOWER
DIODE!
1
0.8
0.6
LATERAL
0.4
0.2
0
-20
-18
-16
-14
-12
-10
-8
Voltage (V)
-6
-4
-2
0
Photodiode Conclusion

Planar diode had increased responsivity
◦ Higher Signal from Fluorescence Signal
◦ Higher Dark Current

Lateral diode had low capacitance
◦ Fast Response Time
Planar likely candidate for Fluorescence Spec.
Testing Results
Plan was to assemble a tight flow
chamber with valves with oxygen and
nitrogen flowing in.
 Emitter and receiver circuit showed
proper required behavior as outlined in
specifications and customer needs.
 Limited testing environment available, but
still showed a change in intensity, as
specified.

Strong / Weak Points of Design/ Room for
future research & improvement

Strong points of final design
◦ Was able to exhibit a possible, more
affordable alternative.
◦ Introduced cost effective fabrication method
of photodiode.
Weak points & places for improvements

◦
◦
◦
◦
Actual testing of chamber incomplete
Abnormal behavior in emitter circuit
Needed more people in respective fields
Needed more funding
Conclusion
Project description
 High Level Customer Needs/ Eng Specs
 Concept Description & Rationale
 System Architecture
 High Risk Assessment
 Detailed Assembly

◦ Emitter and Receiver Circuit
◦ Photodiode Fabrication


Testing Results
Strengths & weakness of design, plans for future
research