Module 6: Glucose Sensor

Measuring Glucose with Op-Amps!

Objectives

  • Design a voltage divider circuit that delivers a fixed voltage to the glucose sensor.
  • Build and analyze a circuit that allows you to determine the unknown resistance value of a resistor and the current flowing through it.
  • Build and analyze an integrator Op-Amp circuit.
  • Build and test the circuit for a blood glucose meter using several glucose concentrations.
  • Analyze the output from the glucose sensor to create a calibration curve.

Procedure 1: Pre-Lab Instructions

Diabetes is a serious disease which affects nearly 400 million people worldwide, it is the 8th leading cause of death, and costs the US nearly billion a year. These glucose meters play a very important role in managing diabetes as they provide the patient the ability to monitor their glucose levels from home and at any time of the day. If you look in the pharmacy you will find a number of hand-held meters and accompanying test strips. Typically, the patient would insert one of the disposable test strips into the meter, prick their finger, load a droplet of blood into the test strip, wait a few seconds and get an instant reading of the current glucose level.

The basic principle of operation is as follows:

  • A set of electrodes are printed on the test strips.
  • A fixed voltage is applied across the electrodes (either constant or in a known sequence) and the resulting current and its response is monitored. In our case the voltage applied across the electrodes is -0.4 V or -400 mV.
  • The current response is then related to the glucose concentration through calibration.
  • Typically the electrodes are coated such that an enzymatic chemical reaction occurs at the electrode surface and this reaction dictates the resulting current across the electrodes.

The details of the electrochemistry can be quite complex. Since we are using commercial strips, the details are somewhat unknown as the companies do not typically release detailed data about their particular test strip operation. The basic principles you will learn in this lab are similar to a number of electrochemical detection sensors which are common in biochemistry. Here are some examples of the measurements made by glucose sensors:

  • Some devices watch the current after a short initial transient (the current will level out to some degree) and then report this steady state current after a fixed time.
  • Another principle looks at the total amount of reaction which has occurred and thus integrates the current with respect to time to obtain the total amount of chemical reaction which has occurred.

We will use a particular test strip, similar to the One Touch Ultra and will examine both types of measurements (direct current and the integration of the current with respect to time) and see which relates to the glucose level more strongly, the “steady-state” current or the total integrated current.

We will build the circuit in stages, testing each piece as we go along. The Op-Amp that was introduced last week, the LMC6484 Op-Amp (shown in Figure 1) will be used again in this lab.

Figure 1: Op-Amp Pinout for LM6484

Remember that you need to connect your 5V power supply to pin 4, and ground to pin 11. Try orienting your Op-Amp so that pin 4 is facing your 5V power supply rail. Also, to refresh your memory, these are the 3 assumptions for an ideal Op-Amp:

  1. The inputs into the Op-Amp draw no current (i.e. I = 0 into both the non-inverting and the inverting inputs).
  2. The two input voltages are equal to each other.
  3. When wired with negative feedback (i.e. the output of the Op-Amp is connected to the inverting input, with no resistors in between), then the output voltage is also equal to the two input voltages, which is fixed by the non-inverting input – buffer or follower circuit

Procedure 1 Results:

- None for this section. Please make sure you understand the manner in which measurements are made for glucose sensors (see underlined section above).

Procedure 2: Create the voltage source

The applied voltage or input voltage that we want to apply across the electrodes is about -400 mV. We can start by creating a voltage source of about 2.1 V measured relative to ground or (2.1 - 0)V = 2.1V.) If we try to measure the same 2.1 V source relative to 2.5 V instead of ground, we will measure (2.100 – 2.500) V = - 0.400 V or -400 mV. Let’s begin first by creating the 2.1 V source.

Figure 2: Voltage divider providing 2.1 V input voltage.

Select two or more resistors to make a voltage divider that will provide about 2.1 V (measured relative to ground.) You can use more than two resistors if we do not have a convenient value.

For example, say you wanted an 11 kΩ resistor and we do not have that value – you could place a 1kΩ and 10kΩ in series.

Build the circuit shown in Figure 2 with your choices of R1 and R2. Check that the output of the Op-Amp is providing 2.1 Volts relative to ground. Remember that by using the Op-Amp, you can estimate the resistors required to generate a 2.1 V signal by using the voltage divider method.

Procedure 2 Results:

- Recreate the circuit in Figure 2 with the appropriate values of resistance that you used to create your 2.1 volt source. You can use the software available by Digi-Key to create your circuit --http://www.digikey.com/schemeit/. Most of the symbols you need are under “Schematic Symbols” and “Passives.” You can label your resistors using this software, or in your word document, whichever is easier for you.

.

- Include your voltage divider calculations in word below your circuit diagram.

Procedure 3: Resitor value between Op-Amps

You will now add to the circuit from Procedure I (shown in Figure 2). Note that since we are using a chip with four Op-Amps, we are showing the power supply for all the Op-Amp in the lower left corner of the schematic, since the circuit diagram becomes a little simpler if we don’t draw the power supply for every Op-Amp.

Figure 3: Current through 100KΩ resistor is a function of the glucose concentration.

The above circuit can also be used as a resistance meter to determine the value of R3. The voltage at the negative input of Op-Amp A is 2.1 V and the voltage at the negative input of Op-Amp B is 2.5 V (remember our Op-Amp rule about inputs being equal?). Therefore, -400 mV is applied across resistor R3 (though the exact value will depend on your selection of R1 and R2). NOTE: WE WILL BE REPLACING THIS RESISTOR WITH THE GLUCOSE SENSOR!!

The -400 mV known voltage drop across R3 will have an associated current flowing through it. The current flowing through R3 must be the same as that flowing through the 100 kΩ resistor (since no current can go into the Op-Amp inputs). If we measure the voltage at the output of Op-Amp B relative to 2.5 V, we can easily infer the current flowing through the 100 kΩ resistor

….twinkle, twinkle, little star… V=IR….

Since the current flowing through R3 can be determined, and the voltage across it is set, we can calculate the value of R3. Build the circuit and test it with a few values of R3. Note that if R3 is too small (relative to 100kΩ) the Op-Amp will saturate. If your Vout is close to 2.5 V, you may be in the saturation region of the Op-Amp. Report the results of your resistance testing. Convince us the circuit works and you understand how/why. Once you connect the sensor between Op-Amp "A" and Op-Amp "B" (in Procedure IV), the output will result in a current and voltage proportional to the amount of glucose in solution!

Procedure 3 Results:

- Complete the handout for Procedure III for the checklist only. No need to include these results in your report.

Procedure 4: Build an Integrator Op-Amp

The circuit in Image 4 integrates the input voltage (which is proportional to glucose concentration) with respect to time.

Figure 4: Integrator Op-Amp Circuit.

To see that this circuit integrates, just use all our laws for the resistor, capacitor, and Op-Amp. The Op-amp rules tell us the voltage on the negative input must be the same as the positive input. Therefore, the voltage at the negative input is 2.5V. Ohm’s law says that the current through the resistor is

$${V_{in} - 2.5V \over R} = I$$

Since the input to the Op-Amp draws no current, this is the same amount of current flowing through the capacitor. The capacitor can be modeled by

$$i(t) = C{dV \over dt}$$

This indicates that the voltage cannot change instantaneously through a capacitor. Therefore,

$$i(t) = C{d(2.5V - V_{out})\over dt} = -C{d(V_{out}) \over dt}$$

and this is equal to the current through the resistor,

$$i(t) = {C{d(2.5V - V_{out})\over dt}} = {-C{dV_{out} \over dt}} = {V_{in} - 2.5V \over R}$$

Therefore,

$$\int_{V_{out}(0)}^{V_{out}(t)} {-CdV_{out}} = \int_{0}^{t} {V_{in}-2.5V \over R}dt$$

$$V_{out}(t) - V_{out}(0) = {{-1 \over RC}(V_{in}-2.5V)t} $$

$$V_{out}(t) = {V_{out}(0) - {1 \over RC}(V_{in} - 2.5V)t} $$

The above model describes how the integrator works, the output voltage is related to the time integral of the input voltage, though we have to be careful that with integrators, the initial state of the capacitor matters. In this module, you will focus on the actual behavior of the circuit.

Build the integrator circuit with a 100kΩ resistor and 1 microFarad (µF) capacitor shown in Figure 4, and analyze the output by using your Waveform generator as the input voltage as described below (READ EACH VERY CAREFULLY):

  1. Test the circuit with a low frequency square wave. The square wave should be centered at 2.5 volts (adjust the offset) and you may want to experiment with changing the amplitude.
  2. Monitor the input and output voltage with the Scope on your Analog Discovery and prove that this circuit is a reliable integrator (i.e. squares integrate to triangles).
  3. Note that the frequency you select should be low enough that there is sufficient time for the integrator to completely “fill up” and reach either the 0 or 5 volt rail during each cycle. The circuit will work fine at higher frequencies, though small amounts of noise can cause the integrator to drift on a slow time scale – so it is easier to get it to work if you saturate the integrator each cycle.
  4. Provide at least 3 plots of this experiment to show that you know how the integrator works. You decide exactly what you want to describe in your report, so long as you use multiple graphs.

Procedure 4 Results:

- Include the 3 plots for this section, and include the following information at the top of each plot:

Frequency = ?, Amplitude = ?

Procedure 5: Build the Glucose Circuit

Now, we will put these pieces together and plug in the glucose sensor. This circuit is just combining the pieces you already built. Note the changes. Resistor R3 was removed and replaced with the 3 wires connected to the glucose sensor. The test strip will stand in for the resistor. We added a small capacitor in parallel to the 100kΩ resistor – this adds a bit of filtering and provides a smoother signal. Also, the capacitor value on the integrator is changed to 100 µF

Figure 5: Glucose Sensor Circuit

The principle with the test strip is common for electrochemical measurements. Electrochemical measurements usually use three electrodes, named the counter, working and reference electrode. Current flows between the counter and working electrode. The reference electrode is held at a known potential with no current. The reference electrode fixes the potential of the liquid on the test strip.

From Figure 5, you can see that based on Op-Amp rules, the reference electrode is held at 2.1 volts and the working electrode is at 2.5 volts. Thus the reference electrode is 400 mV lower than the working electrode. Our circuit then monitors the current flow between the counter and working electrodes

If we measure the voltage at the output of Op-Amp B we get a signal proportional to current, and if we monitor the output of op-amp C we see a voltage proportional to the integral of the current.

NOTE: in order to test the integral, you need to reset the capacitor voltage to zero by shorting out the capacitor with a small piece of wire.

Procedure 5 Results:

- Include the circuit in Figure 5 in your report, and explain each stage, starting with the voltage divider and buffer circuit, the sensor connections, and ending with the filter and integrator.

- Take a picture of your final circuit, and include it in your report. You will be graded on the quality of your circuit on the breadboard. No loopy wires!!!

Procedure 6: Test the Glucose Circuit

At this point you have tested your circuit in pieces. Be careful to plug in the wires from the glucose test strip connector in the proper place. Note that the red, green, and black wires are attached to the counter, reference, and working electrodes respectively. We have limited test strips (they are not super cheap), thus each student can use 6 strips. No more, please. If something does not work perfectly, and you do not get 6 good measurements, do not panic. You should be able to interpret the data with 3 or 4 good measurements. However, if the first experiment does not work, do not try another strip expecting a different result.

The test strips look like the picture below. The sample loads on the left and by touching a droplet, the liquid will wick in automatically. The electrodes are inside the little box on the far left. The black lines on the right are the connections to the electrodes. The black lines printed on the paper are conductive. The electrode connections from top to bottom are Reference, Counter, and Working.

Figure 6: Glucose Strip

Follow each of the instructions listed below carefully in order to measure the glucose concentration of 3 solutions.

  • MAKE SURE YOUR BREADBOARD POWER SUPPLY IS PLUGGED IN!
  • Place your Scope measurements at the output of op-amps B and C. Measure relative to 2.5 volts.
  • Set the time scale to 4 seconds per division
  • Set the voltage scale for each channel to 500 mV per division.
  • Make sure the Scopes voltage offset is zero for both channels.
  • Place a small droplet in one of the sample cups. You only need one small drop.
  • Reset your integrator by shorting out the 100 uF capacitor with a wire.Both Ch1+ and Ch2+ should measure zero volts before you test your solutions.
  • Hold the test strip against the connector and squeeze for about 20 seconds. The pattern on the test strip must mate with that on the connector. Yes, this is a little hokey but it seems to work. You need to be careful that the electrodes are aligned and making good contact with the connector.
  • Refresh the scope screen by hitting “stop” then “run” (or just wait until the screen loops around).
  • Touch the test strip to the droplet.
  • The current should rise and then decay over the course of several seconds. After about 20 seconds since touching the fluid has passed, hit stop and export the data. Since a whole screen will last for 40 seconds you have 20 seconds of margin to conduct the experiment and stop the scope before the data scrolls back over itself!
  • Repeat for different glucose concentrations. Since we have limited strips, if you mess one up don’t panic. Just get as many good data captures as possible.

Procedure 6 Results:

a. Current vs time for 3 concentrations and 1 unknown.

b. Integral of current vs time for 3 concentrations and 1 unknown.

c. Table with data (from either (a) or (b).