Novel Techniques for Observing Structural Dynamics of Photoresponsive Liquid Crystals

This method can help address a common problem in the liquid crystal research. Which is that peak molecular motion during the photoresponse cannot be determined by conventional methods. The main advantage of these techniques is that the structure and the ultrafast dynamics of photoexcited liquid crystal are obtained both from differential IR spectrum and differential electron diffraction patterns.

I will be demonstrating the time-resolved IR vibration spectroscopy. To prepare a solution phase sample for time-resolved IR vibrational spectroscopy, dissolve 0.025 millimoles of pi-COT in 25 milliliters of dichloromethane, to obtain a 1 millimole per liter solution. To prepare a liquid crystal phase sample, cover a 3 millimeter thick calcium fluoride substrate in pi-COT powder.

Place the substrate on a hot plate, and melt the powder at 100 degrees Celsius. Turn off the hot plate, and allow the sample to cool to room temperature. After preparing the sample, turn on the titanium-sapphire laser and the chirped pulse amplifier and allow them to thermally stabilize.

Set the pulse duration to 120 femtoseconds, the repetition rate to 500 Hertz, and the incident fluence to 1 millijoule per square centimeter. Check the power and stability of the UV pump, and mid-IR probe pulses, and realign the optical paths as needed. Cool the mercury-cadmium-tellurium IR detector with liquid nitrogen.

Position the spectrometer in line with the optical path and calibrate the spectrometer. Then, mount a 1 millimeter thick silicone wafer on the sample holder as a test sample. Set the pump-probe delay to a positive value.

Optimize the pump-probe overlap by adjusting the mirror directing the pump beam to the sample, to obtain the maximal transient signal intensity. Set up a scan, starting at minus 100 picoseconds and ending at 1000 picosends, with a step time of 5 picoseconds. Scan the test sample, and identify the temporal position where the transient signal starts to emerge or time zero.

Then, if using a solution phase sample, mount a flow cell with barium fluoride windows in the instrument and begin pumping the sample through the flow cell. If using a liquid crystal phase sample, mount the sample on a motorized stage to allow continuous movement of the laser spot position on the sample, to minimize laser induced damage. Confirm the time zero temporal position with the sample, and set the start, end and step times.

Select a directory for the data file, and run the data acquisition process. First, etch and clean a silicone nitride coated wafer, to obtain a substrate with silicone nitride windows. Place the substrate on a spin coater chuck with the silicone nitride windows facing up.

Spin coat the substrate with a 10 milligram per milliliter solution of pi-COT in chloroform. Place the spin coated substrate on a hot plate, and heat the hot plate to 100 degrees Celsius to melt the pi-COT. Turn off the hot plate when it reaches 100 degrees Celsius and let the sample cool in place, to obtain a pi-COT liquid crystal film.

To begin the measurement, mount the sample on the instrument sample holder. Place the sample holder in the vacuum chamber and close the chamber. Use a rotary pump to evacuate the chamber to less than 1000 pascals or 10 millibars.

Then, use a turbomolecular pump to evacuate the electron gun chamber to about 10 to the minus 6 pascals. Next, turn on the titanium-sapphire laser, and the chirped pulse amplifier, and allow them to thermally stabilize for one hour. Set the repetition rate to 500 hertz.

Turn on the CCD camera chiller, and cool the device to 10 degrees Celsius. Then, turn on the electrical power supply, and set the voltage to 75 kilovolts. Open the special overlap software, and set the exposure time to 50 milliseconds.

To find the probe electron beam position, first set the start type to z-overlap, and click start. When that process finishes, select y-overlap and click start again. Then, set the electron beam to the sample holder pinhole position, and align the pump laser with the reflected pump light by the pinhole.

Next, switch to the time-resolved electron diffraction program. Set the start type to time-resolved and the pump influence to 2 millijoules per square centimeter. Click start to measure the time zero position based on an inorganic standard material fixed on the sample holder.

Then, insert the Faraday cup into the electron beam path and measure the fluence with the picoammeter. Rotate the neutral density filter to adjust the electron beam fluence to 3 picoamperes. Adjust the pump pulse fluence to 1.67 milliwatts by rotating the wave plate on the pump beam line.

Then, in the time-resolved electron diffraction software, move to the sample position. Set the CCD camera exposure time to 1 second. Set the start type to single, and click start to obtain a static electron diffraction image.

Then, turn on the Peltier element of the CCD camera and cool the device to minus 20 degrees Celsius. Once the CCD camera has cooled, set the number of patterns to collect at each step, and the number of steps for the time-resolved measurements. Set the start type to time-resolved, and click start to collect the time-resolved electron diffraction images.

When the experiment has finished, turn off the electron acceleration power supply. Collect the time-resolved image in the same way to obtain a background. The differential IR vibrational spectrum of a pi-COT based liquid crystal thin film had peaks corresponding to the stretching modes of the COT and thiazole rings, and of the biphenyl moyades, which are strongly IR active in the flat T1 form of pi-COT.

The time-dependent evolution of the peak intensity had a wave number of 1338 reciprocal centimeters. It's consistent with the saddle to flat confirmational change within 2 picoseconds. Followed by relaxation to the initial saddle form in 10 to 20 picoseconds for single molecules, or 150 picoseconds for stacked molecules.

The electron diffraction peaks from the photoresponsive moyades were extracted from the long pattern dominated by long carbon chains, by subtracting the initial diffraction pattern from a pattern obtained 500 picoseconds after UV pulse radiation. The positive and negative peaks indicated the formation and loss, respectively, of structural features by 500 picoseconds after irradiation. Th formation of a new ordered structure began about 200 picoseconds after photoexcitation, consistent with the relaxation to the initial saddle form in stacked molecules.

The loss of the pi-pi stacking order on a 300 picoseconds timescale, was attributed to a small number of flat conformers, with 300 to 1000 picoseconds lifetimes, twisting to minimize steric hindrance. We first had the idea for this method when we discussed liquid crystals, which are not usually examined by times of diffraction because they do not provide good electron diffraction patterns. Our methodology allows changes in liquid crystal diffraction patterns to be identified, allowing access to important structure information and ultrafast molecule dynamics.

Once mastered, these techniques can be done in a half day if they are performed properly. These techniques can potentially be used to answer further questions about the structure and ultrafast dynamics of more complex biomaterials.