Seebeck Effect: Turn Heat into Electricity, Then Measure It with a Thermocouple Thermometer
AbstractEveryone knows electricity can create heat, especially because our electrical appliances tend to warm up when turned on. But wouldn't it be cool to do the reverse — generate electricity from heat? In this science project, you will study why it happens, measure the effect, and then use the phenomenon to build your own device, a thermocouple thermometer, that will enable you to convert heat into electrical energy.
Sabine De Brabandere, Ph.D., Science Buddies
Wendell Wiggins, Science Buddies volunteer
Explore thermoelectricity, specifically the phenomenon called the Seebeck effect, to determine the electrical potential generated at the junction of leads made of different conductive materials and how this varies with temperature.
Mankind has known about thermal energy longer than most other forms of energy. For centuries, people have created thermal energy fairly easily as an end product or byproduct of other energy-generating processes. Think of a fire where chemical energy is transformed into thermal energy, or friction where mechanical energy is transformed into heat, and Joule heating where part of the electrical energy is dissipated as heat. Can we reverse these processes? Not all of them — but in 1821, Thomas Johann Seebeck discovered how to transform thermal energy into electrical energy, and that is precisely what we are going to do in this science project.
Because converting heat to electricity is quite unusual, make sure you understand some fundamental parts of physics before diving into this topic. Do you feel confident answering the following questions?
- What is heat (or thermal energy) and how does it relate to temperature?
- What is electric current and how does it flow through an electrical conductor?
- What is an electrical potential difference and how is it measured?
For the concepts that you are not sure about, do some research. You will need to know these concepts to understand what's going on in your project. Check your textbooks, local libraries, and see the Bibliography for some suggested tutorial materials. You can also learn more about electricity basics in the Electricity, Magnetism, & Electromagnetism Tutorial.
Your background research should have helped you visualize electrons moving freely in the spaces between the atoms of an electrical conductor. Consider a different view of these electrons in which we don't show the individual atoms of the conductor, but rather we consider the overall environment that those atoms produce. The atomic framework of the conductor interacts with the free electrons to give them an average energy. Even though they are free to move within the conductor, their electric charge is still pulled on by the electric forces of the framework. Just as you are pulled down by the net gravitational field of the Earth without noticeable contributions from each grain of sand, the electrons are held by the electrical forces from the framework of the conductor without noticeable contribution from each individual atom.
The magnitude of the electron-framework attraction varies from one conductor to another, measured by the total attractive energy on each electron and stored as potential energy. Some conductors hold their free electrons tight; it is as if the electron is in a deep valley and lots of energy is needed for the electron to overcome the valley's slope and free itself from the attractive force of the conductor. Other conductors hold electrons less tight, as if the electrons are in a shallower valley . The energy per electron needed to pull an electron completely out of a conductor (or over the slope) is called its work function.
Now, let's manipulate our conductor(s) in a couple of ways and observe what happens to this potential energy.
As our goal is to transform thermal energy into electrical energy, let's add some thermal energy, heating up one end of an electrical conductor, and see what happens. Heat will make the free electrons move around faster, meaning it will be easier in general for them to escape the attractive forces of the conductor. In other words, raising the temperature of a conductor will lower the work function. If we can keep a temperature difference across the conductor, free electrons at the warm end will roll down the energy slope to the cooler end, where they are held by a stronger attraction to the positive framework. The transfer creates an accumulation of negatively charged electrons at the cooler end, and thus an electrical potential difference (ΔV) across the conductor. Once enough of the electrons roll down the energy hill, the electrical potential difference they create opposes any more electrons coming down the hill, resulting in equilibrium.
The formation of an electrical potential difference as a result of a temperature difference across a conductor is exactly what Seebeck discovered in 1821. It is called the Seebeck effect, measured by the Seebeck coefficient, or S, and defined by the following formula:
S is a measure of the electrical potential difference induced over a conductor by one degree of temperature difference for a particular conductor at a particular temperature. This effect is not linear with temperature, so the Seebeck coefficient is temperature dependent. It is generally expressed in microvolts per degree kelvin (µV/°K).
Earlier, we mentioned electrons moving within a conductor. So did we just create a current by heating one end of a conductor? Did we transform thermal energy into electrical energy? Yes — the Seebeck effect does express how a temperature difference created by adding thermal energy to one end of a conductor creates a potential difference (or electrical energy) across the conductor. Is this enough to make a device like a thermometer based on the Seebeck effect? Come back to this question later; let us first discuss how we will measure the effect.
To measure the electrical potential difference (or the electrical energy created) , we need to put a voltmeter in the circuit — this will create connections (or junctions) between conductors. Let us first examine what happens to the average free energy per electron when we connect conductors together. Figure 1 shows this free energy per electron at the contact, or junction, between two conductors with different work functions. Any idea what would happen?
Figure 1. The average energy of free electrons depends on their host material. Inside conducting metals, the electrons have a lower free energy per electron than in free space because part of the energy is stored as potential energy, due to the attraction of the host atomic framework. The electrons may have different average free energies in different metals. This figure shows the average free energy per electron at a junction of two metals.
When such a contact is formed, some of the electrons in conductor 2 (the conductor where the electrons are less tightly bound — Metal 2 in Figure 1) will "roll down the hill" to conductor 1 because their average free energy is lower there; in other words, they feel more attraction by the framework of conductor 1. The electrons that roll over contribute an excess negative electric charge to conductor 1 that can be measured as an electrical potential difference over the junction. Once enough of the electrons have moved to conductor 1, the electrical potential they create prevents any more electrons coming down the hill, resulting in an equilibrium.
Now — let us get back to heat. What would happen if we raise the temperature of a junction? Would more electrons now have enough energy to cross over and roll down the energy hill, creating a larger electrical potential difference? In most cases, yes! And how many more electrons roll over (or how much electrical potential is created) depends on the difference between the Seebeck coefficients of the conductors in the junction.
Do you see how it all connects, how we can convert heat into electricity? Adding or removing heat from the conductor junction will change its temperature, which causes the electrons to redistribute and cause a small but measurable change in electrical potential difference at the junction.
Great — so what can we make with this? How can we apply it? One common application of the Seebeck effect is thermocouple thermometers. These thermometers measure the potential difference at the junction to calculate the temperature at the junction. The junction (or connection between the two conductors) in these devices is called a thermocouple. Let's explore the Seebeck effect in some more detail by making a thermocouple thermometer ourselves, first in theory and then in practice.
To create a thermometer, we need a measuring probe. A probe is a device where two different metals are joined together and are subject to the ambient temperature. As noted before, the induced electrical potential difference created at the probe will depend on the materials used. In this science project, you will use the probe of a commercial thermocouple as the sensor. As a follow-up, you can explore different conductors or semiconductors.
Second, we need to create a device to measure the electrical potential difference over the probe, because the measured electrical potential difference will be an indicator of the temperature of the probe. Figure 2 shows a setup to measure the electrical potential difference.
A thermocouple junction connects an aluminum and iron thermocouple lead. Both of those leads then connect to junctions at room temperature which are connected to copper plugs inserted in a voltmeter.
Figure 2. Figure showing how a junction of two different metals, here aluminum (Al) and iron (Fe), can be connected to a voltmeter to form a thermocouple thermometer. In the example, the voltmeter connectors are made of copper (Cu). To use the thermocouple thermometer, the voltmeter and its connections need to be maintained at a known temperature (here room temperature). The temperature at the probe thermocouple junction is the unknown temperature to be measured.
Will the voltmeter output (referred to as ) be the same as the potential difference at the junction? Or do we need to add more terms, as more junctions have been introduced in the loop? The potential difference observed by the voltmeter is the sum of the potentials generated by the junctions of metals. We consider the electrical potential outside the metals to be zero, so in adding up the potentials, we count them positive when entering the metal and negative when leaving the metal. Starting inside the left meter contact, the sum around the loop (assuming the unknown temperature is 100°C) in Figure 2 is then:
The subscript identifies the metal type, and the number inside the parentheses is the temperature. Can you see which terms cancel? Rearranging the other terms provides:
Reading the equations, can you see how the observed electrical potential at the voltmeter ( ) depends on how much the potential in each metal changes with temperature and on the metals used?
Great, but how do I translate the electrical potential reading on my voltmeter to a temperature? It's fairly simple: Just as a fluid thermometer uses the measurement of how much a fluid expands for a given temperature change to translate a volume measurement to a temperature measurement, a thermocouple thermometer uses the expected electrical potential change at the junction for a given temperature change to translate the measured electrical potential to a temperature value. Just as we have to calibrate our fluid thermometer (e.g., indicate the volume for a temperature of 0°C), we will need to calibrate our thermocouple thermometer (e.g. electrical potential reading for 0°C). A thermocouple thermometer is a bit more difficult than a fluid thermometer, because the expected electrical potential at the junction does not change linearly with temperature. Because of this, data lists of junction electrical potentials for a set of temperatures are needed. These lists have been published for the different thermocouple metals commercially used; an example for type K thermocouples can be found in Type K thermocouple data sheet.
Does it sound complicated? Perhaps. But go ahead, follow the instructions, and it will all fall into place.
Terms and Concepts
- Thermal energy
- Electrical current
- Electrical conductor
- Electrical potential difference
- Seebeck effect
- Seebeck coefficient
- Thermocouple thermometer
- What does a negative Seebeck coefficient mean?
- For two metals to form a thermocouple probe suited for a thermocouple thermometer, do they need to have a largely different work function or a largely different Seebeck coefficient or both?
- What are some applications of the Seebeck effect? If I never see an application, why is it important for me to learn about it?
- What characteristics of a thermocouple thermometer would make it interesting to use in applications? Some things to think about are:
- Ability to measure the temperature of gas, liquid, and/or solids
- Reaction time (how fast can an accurate measurement be obtained)
Following site is a good place to refresh your knowledge on thermal physics and current electricity:
- Henderson, T. (n.d.). Physics.Topics The Physics Classroom. Retrieved October 23, 2012.
- Laube, P. (19 September 2012). Fundamentals: Conductors, semiconductors, insulators. Semiconductor Technology from A to Z. Retrieved October 23, 2012.
- Winder, E.J., A.B. Ellis, and G.C. Lisensky (1996, October). Thermoelectric devices: Solid-state refrigerators and electrical generators in the classroom. Journal of Chemical Education, vol. 73, no. 10. Retrieved October 23, 2012, from /science-fair-projects/project_ideas/Phys_p070.pdf
- Yarris, L. (February, 2007). Taking the measure of the Seebeck effect. Science @ Berkeley Lab. Retrieved October 31, 2012.
Materials and Equipment
- Type K bare thermocouple, not encapsulated in a protective case, e.g. the following Type K Thermocouple available from Amazon.com.
- Wire strippers available from Amazon.com
- Sandpaper, 220 grit (2 inch square)
- Set of two banana plugs or connector pins (depending on the multimeter used), attaching wires with a screw (not with solder). These can be found at an electronics store or banana plugs or pins available from Amazon.com.
- Multimeter with 1/10 millivolt (mV) resolution, e.g. the Extech MN35 available from Amazon.com.
- A thermometer suitable for reading indoor room temperature
- Drinking glass
- Ice cubes
- Water boiler or burner and pot for boiling water
- Cup to hold hot water
- Oven and/or refrigerator/freezer
- Oven or refrigerator/freezer thermometer, or other device to measure temperature in the oven and/or refrigerator
Disclaimer: Science Buddies participates in affiliate programs with Home Science Tools, Amazon.com, Carolina Biological, and Jameco Electronics. Proceeds from the affiliate programs help support Science Buddies, a 501(c)(3) public charity, and keep our resources free for everyone. Our top priority is student learning. If you have any comments (positive or negative) related to purchases you've made for science projects from recommendations on our site, please let us know. Write to us at email@example.com.
Make a Thermocouple Thermometer
For this science project, the bare thermocouple plug will be substituted by two banana plugs or connector pins. (Note that in the following text, you can substitute any reference to banana plug with connector pin if your multimeter uses pins). This will enable a direct connection between the probe and the multimeter. As a side note, connectors (such as alligator clips) could be used to connect the thermocouple outlet blades to the multimeter. This science project intentionally does not use such connectors, to ensure more accurate readings.
- Use the wire cutter to cut the plug from your thermocouple close to the plug, and discard plug. Make the cut close to the plug, as the wire needs to be fairly long. Why? The probe (or one end of the wire) will need to be in a place to pick up the unknown temperature (e.g., an oven or a refrigerator). The meter will be attached to the other end of the wire and needs to be at room temperature to make accurate measurements. A longer wire will make the meter more practical.
- Strip about 1 inch of coating from the thermocouple wire near the cut. See the Science Buddies Wire Stripping Tutorial if you need to learn how to strip wires.
- An outer coating holds together two electrical wires called the thermocouple leads, each with its own coating, as shown in Figure 3. Strip the outer coating with your wire strippers to see the inner wires and coating.
Figure 3. This figure shows the thermocouple wire divided into two thermocouple leads.
If there is still coating left on one or both wires, use the sandpaper to remove the inner coating. Cut out a 2-inch square piece of sandpaper and fold it in half, rough side facing the inside of this "sandwich." Put the wire in the fold of the sandwich and rub the wire back and forth between the sandwich, applying a light pressure. A few rubs should be enough to remove all remaining coating.
- Attach a banana plug to each wire, making sure there is a good electrical connection (i.e., bare wire connects to the conducting part of the banana plug). Figure 4 shows the setup for one particular type of banana plug.
Figure 4. Thermocouple leads attached to banana plugs. In this case, a good electrical connection was achieved by putting the lead through the hole and around the banana plug back into the hole. The screws hold the wire in place and secure the connection.
- Push the banana plugs in the multimeter to read the electrical potential difference between the plugs. If you are new to multimeters, the Science Buddies resournce How to Use a Multimeter, together with your multimeter manual, will help you make your first measurements. Note that this science project deals with very small electrical potential differences and currents; it is a safe place to experiment with your meter. Figure 5 shows how your thermocouple thermometer will look when not in use.
Figure 5. Example of a thermocouple thermometer as built in this science project, with a thermocouple probe attached to an electrical multimeter with banana plugs. The multimeter will be set to measure millivolts (mV) when in use. The measured electrical potential difference will be translated to a temperature offline. It is essential to maintain the meter and its connections at a known room temperature when using them as thermometer. Only the thermocouple probe should be placed where it can be used to sense an unknown temperature.
Calibrate the Thermocouple Thermometer
As with any thermometer, the thermocouple thermometer needs to be calibrated. It will measure electrical potential relative to the electrical potential at room temperature (assuming the multimeter connections are at room temperature).
- Use your thermometer to measure the room temperature and write it down in a data table like Table 1.
|Room Temperature Potential Difference
|Junction Potential Difference
|Measurement||From Data Sheet||Measurement||Calculated:
ΔVJunc = ΔVTotal + ΔVRoom
|From Data Sheet|
at 250° F
|Refrigerator or Freezer|
- Look up the thermoelectric electrical potential difference for your observed room temperature for a Type K thermocouple in this Type K thermocouple data sheet. Note the value in your data table. Pay attention to the sign (negative or positive)!
Test the Thermocouple Thermometer
Now that you have calibrated the thermocouple thermometer, you are ready to test it.
- Take a drinking glass, fill it with ice cubes and water (preferably small size ice cubes).
- Insert the probe in the ice water.
- Put your multimeter in the volt measurement mode.
- What range of measurement do you expect? Microvolts (µV), millivolts (mV), or volts (V)? Refer to the Type K thermocouple data sheet to get an idea.
- Adjust the range of measurement of your multimeter to the expected measurement range.
- Observe the meter reading and note the electrical potential reading for 0° Celsius (the temperature or your water/ice mixture) in your data table, column .
- Calculate the junction electrical potential difference (
How? As explained in the Introduction (see Equation 2 in the Background tab), the following equality holds (assuming the meter connections are at room temperature):
orDo a little algebra and you are there. Make sure, though, to pay attention to the sign of the potential differences. Note the result in your data table.
- Look up for which temperature a K thermocouple produces the thermoelectrical potential you just calculated using the Type K thermocouple data sheet. Write this temperature down in your data table.
- Is the observed temperature of ice water what you expected? If not, here are some hints to help you figure out what went wrong:
- Did you set your multimeter to measure millivolts (mV)?
- Check the signs of the electrical potentials. As you can see from the data sheet, the thermoelectrical potential difference is negative for negative temperatures, positive for positive temperatures.
- Go back to the data sheet and see if you understand how to read it. Example: for a room temperature of 19° C, you should read a 0.758 mV thermoelectrical potential from the sheet.
- Check if there is a good electrical connection at both banana plugs.
Take Measurements with the Thermocouple Thermometer
- Measure the temperature of boiling water or other cold or hot liquids, placing the probe in the liquid. Do not forget to record your observations in a data table like Table 1.
- Measure the temperature of the air in the refrigerator, the freezer, or the range oven set at 250° F. In these cases, the probe needs to be in the refrigerator, freezer, or range oven, while the multimeter and its junctions need to be maintained at room temperature. Keep the limitations of your thermocouple probe in mind; the wire coating might melt if used at very high temperatures. Do not forget to record your observations in your data table like Table 1.
- Before you start, one more hint: Do not forget to let the meter come to room temperature, and measure the room temperature every time you move your meter to a different room or start again at a different time.
Food for Thought
Now that you know how to use your thermometer, here are some things to explore:
- Set your voltmeter to measure in millivolts.
- Heat up one banana plug with your finger or a cloth drenched in hot water. Do you see a reading? Why would heating up the junction where thermocouple leads connect to the metal in the banana plug create a measurement? Is this a thermoelectric couple as well?
- Heat up the other banana plug in the same way. Do you get a reading? If so, is it the same? If not, why would it be different? Is this a different thermoelectric couple?
Ask an Expert
- Make thermocouples from other metals (like iron or plain steel wire, copper, or aluminum) and compare the magnitude of the effect for the different combinations. Note that the effects might be small depending on the metals used. See if you can compute the Seebeck coefficient from the metals used.
If you like this project, you might enjoy exploring these related careers:
- Science Fair Project Guide
- Other Ideas Like This
- Electricity & Electronics Project Ideas
- My Favorites