Presentation Overview
Download
Report
Transcript Presentation Overview
Detection of glycerides in biodiesel
Team Members:
Ben Wilson
Matt Yourek
Megan Templeton
Mentors: Dr. Tom Hess
Dr. Dev Shrestha
Client: Dr. Dev Shrestha
1
Opportunity Statement
Current Testing Method
Product Specifications
Housing Design
Circuit Design
Membrane
Experimentation
Experimental Design
Parameter Measurements
Results
Recommendations
Design, fabricate, and test a handheld biosensor that
uses a modified pH meter combined with an
immobilized enzyme electrode to detect the amount of
free and total glycerol in a sample of biodiesel.
3
Currently biodiesel makers use a gas chromatograph to
measure the amount of glycerol within their biodiesel. While
the GC provides a very accurate measurement, it is a time
consuming process to prepare and test a sample.
4
Light weight – total weight of less than 5lbs
Small – have a footprint smaller than half a sheet of paper
Durable – able to withstand drops and bumps
Reusable – can test multiple samples with one membrane
Fast – easily calculates concentration percent in seconds
Simple – minimal buttons and good user feedback
Accurate – Consistent voltage signal for each sample
Easy to Calibrate – a single button push resets reference
voltage
Fabricated using the Mechanical
Engineering Department ABS
plastic rapid prototype printer
3/16
” wall thickness
1” x 6” x 3.5” internal space
Six attachment points
Separate battery compartment
Circuit board attachment points
Major Components
Op – Amps
RC Filter
Arduino UNO
LM741CN
Simple operational amplifier
Forces voltage signal to be positive by adding a constant
voltage to the measured signal
AD623AN
Variable gain operational amplifier
Amplifies voltage signal so small changes can be
recognized by the Arduino
RC Filter
Uses a resistor and a capacitor in series to filter the signal
Removes high frequency signal noise added during
amplification
The cutoff frequency (Hz) is calculated using the equation
Arduino UNO
Registers user input through buttons
Reads voltage from probe and assigns bit value to
corresponding variable
Blinks LED to verify that a button has been pushed
Writes valuable information to LCD screen
Processes input voltage from probe by removing the added
voltage and converting to concentration percent
The Biosensor was
calibrated using a DC
voltage source and a
digital multimeter
A line was then fit to the
data using the Matlab
fitting tool
The calibration equation
was then programmed
into the Arduino, so the
true signal voltage is
displayed
Initially going to test three
types of membranes
PVA
PVDF
Water emulsion
Membrane Failure
In the beginning the
membrane easily
degraded
Then even the thinnest
layers of membrane were
too thick to record accurate
measurements
Determined to add lipase
directly to solution
Ran 32 experiments testing
four variables:
Feedstock
Lipase
Temperature
Soap
Variables
Feed
- Oil
- Biodiesel
Temp.
- 20°C
- 40°C
Lipase
- 20 mg
- 10 mg
Soap
- 1 ml
- no soap
ANOVA
- Identify significant
variables
Regression Analysis
- Identify optimal
combinations
Calibration
Point Measurements:
Time-based Sampling:
• Initial pH
• Initial mV
• Final pH
Potential (mV)
mV Response to Lipase Catalyzed Rxn
0
-20
-40
-60
-80
-100
-120
-140
-160
0
• Final mV
• Δ mV
400
600
Time (s)
800
1000
600 Sample Moving Average
moving average (mV)
• Δ pH
200
-40
-50
-60
-70
-80
-90
-100
-110
-120
-130
0
200
400
600
800
1000
1200
1400
1600
Significant Variable:
Optimal Combination:
Feedstock
20 mg lipase, 20°C, 1 ml soap
Sample
1
2
3
4
5
6
Feed
Canola Oil
Canola Oil
30 min. Biodiesel
30 min. Biodiesel
60 min. Biodiesel
60 min. Biodiesel
ΔmV
15.9
8.0
46.7
49.0
91.1
101.0
Final mV - Initial mV
120
Dimensionless
Value
0
0
0.5
0.5
1
1
y = 84.1x + 9.9
R² = 0.9815
100
80
60
40
20
0
0
0.5
1
Construct calibration curve from biodiesel samples with
known concentration of total glycerides. True concentration
is determined by gas chromatography (GC)
Validate curve with GC analysis
Gain better understanding of reaction mechanism driving
the pH change
Determine membrane composition and thickness having
the least impact on probe sensitivity