Cooling Rate Dependent Ellipsometry Measurements to Determine the Dynamics of Thin Glassy Films

The overall goal of this procedure is to provide a simple method to quickly and accurately measure the glass transition temperature, apparent expansion coefficient, and average dynamics of ultra-thin glassy films. This method can help answer key questions in a polymer glass field, such as how the dynamics of thin-films relate to those of the bulk. The main advantage of this technique is that it can calculate the glass transition temperature, apparent activation coefficients, and fragilities of polymer ultra-thin films in a single high-throughput experiment.

While the example provided here is mostly focused on polymer glasses, this method can be applied more broadly to study the properties of thick and thin films of other types of glasses, such as organic molecule glasses and nanocomposites. Start preparing the film a day before it is needed for glass transition temperature measurements. Have a scale ready to accurately weigh materials for the film.

This experiment will use polystyrene and toluene to create the film. Begin by measuring 40 milligrams of polystyrene. Add two grams of toluene to produce an approximately 100 nanometer film.

After the solution has sat overnight, use it to create a film. Move the vial to a spin coater that is in a fume hood. At the fume hood, put the solution aside for later use.

Also, ensure that there is toluene ready for use with the spin coater. Next, get a silicon wafer to create the film on by spin coating. The silicon wafer for this protocol is one centimeter by one centimeter.

Place the wafer in the spin coater, and spin it at 8, 000 rpm for 45 seconds. As the wafer spins, drop approximately one milliliter of toluene onto it. Stop the spin coater after 45 seconds of spinning.

Use the prepared polystyrene and toluene solution, and begin to add it drop-wise to the silicon surface. Stop when the entire surface is covered. Before the solution dries, spin the wafer at 4, 000 rpm for 20 seconds.

Stop the spin coater, and remove the wafer to measure the film thickness. Determine the film thickness using an ellipsometer. Begin by placing the film on the ellipsometer stage and securing it.

Continue by checking the angle of incidence, acquisition time, and other settings. Then begin the scan. After collecting data, use the ellipsometer software to fit the ellipsometric angles.

Use a three-layer model. The first layer is the substrate, in this case, silicon. The second layer, layer number one, is a native oxide.

The layer has a thickness of 1.5 nanometers. The third layer, layer number two, is a Cauchy Model, corresponding to the optical properties of polystyrene film. The Cauchy Model for the index of refraction is given by this formula.

Lambda is the wavelength, and A and B are constants to be determined from data. K equals zero for a transparent material. The model returns a determination of the film thickness, in this case, about 105 nanometers.

Remove the wafer from the ellipsometer and proceed to the next step. If the film is of the desired thickness, take the wafer to a vacuum oven. Place the wafer in the oven and anneal it for 15 hours at 393 Kelvin.

Return to the ellipsometer with the annealed film. Put the film aside while preparing the sample stage. The ellipsometer should be equipped with a variable temperature stage.

Use a thermal paste to coat the surface of its heating element. Next, place the annealed polystyrene film onto the heating element. Then clamp it tightly into place.

Start a flow of 100%dry nitrogen through the temperature stage. Turn to the computer and the temperature stage software to create a temperature profile. This figure provides an idea of the temperature profile.

Temperature is along the vertical axis. Time is along the horizontal axis. The sample is alternately heated to 393 Kelvin and cooled to 293 Kelvin.

The temperature ramp-up grade is always a constant 150 Kelvin per minute, as indicated by the constant steep upward slope. The ramp-down varies with each cycle. It starts fast, then slows down, as indicated by the changing downward slopes in the figure.

The sample is held for 20 minutes after it first reaches 393 Kelvin. All later temperature holding times, at both 393 Kelvin and 293 Kelvin, are for five minutes. Stay at the computer to complete setting up the experiment.

Use the ellipsometer software to create a temperature-dependent ellipsometry model. The substrate layer is a temperature-dependent silicon model. Layer number one is a 1.5 nanometer native oxide layer.

Layer number two, the third layer, is the Cauchy Model for polystyrene film. Work with the temperature-dependent silicon layer settings. Turn on the Use Ext Temp from Parm Log"to allow use of the temperature stage temperature.

Now navigate to be able to edit the hardware configurations. Set the fast acquisition time to one second. Also, select high-accuracy zone averaging.

Set the normal acquisition time to three seconds. Once again, use the high-accuracy zone averaging. Click the In Situ"tab of the ellipsometer software.

Check the Fast Acquisition Time Mode"box. Press Start Acquisition"to begin collecting data. Monitor the data collection as the temperature profile is followed.

Just before the three Kelvin per minute cooling ramp, uncheck the Fast Acquisition"time box. Use measurements of the thickness versus temperature for a given cooling ramp to calculate the glass transition temperature. In this sample curve, the region highlighted in red corresponds to the supercooled liquid.

The region highlighted in blue corresponds to the glassy regime. The point where linear fits to these regions intersect defines the glass transition temperature. This sample is 110 nanometer polystyrene film of 342 kilogram per mole polystyrene.

The cooling rate is 10 Kelvin per minute. Measurements at one Kelvin per minute give a temperature of about 372 Kelvin. Here is data of the glass transition temperature versus the cooling rate for a 110 nanometer film of polystyrene.

This same data is plotted with black circles using the log of the cooling rate on the left vertical axis. The horizontal axis is 1000 over the transition temperature, as suggested by an empirical relation. For comparison, the red squares are the bulk dynamics of polystyrene, as determined by dielectric spectroscopy.

For this data, refer to the right axis of the log of the bulk alpha relaxation time and the horizontal axis, 1000 over temperature. Once mastered, this technique will take five hours if performed properly. While attempting this procedure, it's important to remember to use an appropriate material model, so that you accurately fit the ellipsometry values for the thickness and the index of refraction of the film.

I first tried this method of studying polymer thin-films in James Forrest's lab in the University of Waterloo. However, with recent advances in ellipsometry technology, we are now able to use this as a high-throughput method to study broad systems, such as organic thin-films and newly-synthesized molecules. After this video, you should have a good understanding of how to make a polymer thin-film and calculate the glass transition temperature, expansion coefficient, and apparent activation barrier to the relaxation time near the glass transition.