Top Banner
#426 RBS Study of the Behavior of PMMA as a Negative Resist for Particle Beam Lithography A. Brita Brudvig, Paul T. Campbell, John L. Davenport, T. Clarke DeMars, Wilson C. Fricke, Zachery W. Goodwin, Bertrand N. Irakoze, Caroline A. Roberts, Joel A. Stewart, Jean Shirimpaka, Frank K. Odom, Michael R. ONeil, Randolph S. Peterson University of the South, Department of Physics and Astronomy Abstract RBS Results Conclusions Explanation of the Negative Resist Behavior RBS Experiment with PMMA on Silicon Particle Lithography with Resists PMMA (Poly(methyl methacrylate)) is widely known as a positive resist for low-dose particle beam lithography, and as a negative resist for high-dose lithography. The model often mentioned for characterizing the behavior of PMMA as a negative resist is one of cross- linking of broken polymer chains at high lithographic doses from electron or particle beams. However, there are a few experiments that do not find evidence for such cross-linking. A Rutherford backscattering experiment was performed with a 2 MeV helium beam on a PMMA resist to observe the oxygen and carbon in the polymer chains as the dose was increased from the positive resist to the negative resist behavior of the PMMA. For a sub-micron and a 1-2 µm layers of PMMA on silicon, only carbon was observed and the thickness of the PMMA appeared to decrease with beam dose until a final thickness of carbon was reached. References Preparing patterned carbonaceous nanostructures directly by overexposure of PMMA using electron-beam lithography,Huigao Duan, Jianguo Zhao, Yongzhe Zhang, Erqing Xie and Li Han, Nanotechnology, Vol 20, 135306 (2009). And references therein. This schematic diagram illustrates the behavior of PMMA as a positive and a negative resist. On the left the beams of particles damage the PMMA molecules, breaking the long molecular chains into shorter chains, causing the exposed PMMA volume to decrease. Rinsing with a mild solvent will remove only the shortened molecular chains. Removal of a resist that is exposed to a beam is the definition of a positive resist. Continued exposure to the beams of particles further reduces the volume of the exposed resist. Rinsing with a mild solvent will not remove anything, and the use of a stronger solvent removes only the unexposed resist. This is the definition of a negative resist. PMMA shows both the positive and the negative resist behaviors for all energetic charged particles and some wavelengths of electromagnetic waves. Only a few resists have do so. Si Si PMMA Resist Positive resist Negative Resist Some authors write that as the energy dose (beam dose) to the resist increases, there is a re-linking of the broken molecular chains. The longer and more entwined the molecular chains are, the more resistant is the material to mild solvents. There are a few experimental tests of this negative resist behavior that suggest that the process is not just the breaking and re-linking of molecular chains, but the continuous breaking of molecular bonds, producing a carbonaceous material that is highly resistance to ordinary solvents. This behavior has been observed with electron beams on PMMA, with the gradual extinction of Oxygen K x rays along with the continuous production of Carbon K x rays. The molecular basis for PMMA is C 5 O 2 H 8 . The absence of Oxygen implies the absence of PMMA and not the re- linking of PMMA chains. PMMA is often used as a resist for high-energy (MeV) beams of protons and alpha particles and may produce the same behavior. A beam of 2 MeV He + ions was collimated and directed into a scattering chamber with a silicon surface barrier detector at 150 o or 160 o to the beam. Beam currents were kept as low as possible, but not so low as to cause beam instabilities on target. Beam currents were typically about 10-20 nA. A fine grid before the last collimator was used to monitor beam current during the experiment and the also proved to be a template for our beam lithography. See Figure below. Targets were made by spinning a solution of 950K PMMA (MicroChem) onto a (100) silicon wafer to thicknesses of about 200 nm and 2 µm. The resulting pattern of the beam current grid is visible in the undeveloped PMMA as a negative image below made with an optical profilimeter. RBS spectra were recorded every 20 seconds and the target was protected from beam exposure while spectra were saved. There were no obvious peaks due to the oxygen in the individual spectra. But the energy at which the helium ions were scattered from the silicon substrate increased with time, forming a plateau shown in the figure below for the 200 nm PMMA on silicon with a detector angle of 150 o . In the figure above, the black line is an empirical fit to guide the eye. The spectrum above is a sum of all the 20-second spectra. Carbon appears to be present with a weak peak between 400-500 keV. Even though the counting statistics are low, note the near absence of an oxygen peak, which should be located near 700 keV. The backscatter energy for He on uncoated silicon is 1144 keV and the observed energy above is 950 keV. This indicates a residual covering that is almost entirely carbon. For the thicker PMMA coating on Silicon, about 1-2 µm, the results were similar. The RBS spectra (below) are the sums for individual spectra for the first 140 seconds of exposure (top figure) and the first 500 seconds (bottom figure). There if little evidence for oxygen in either spectrum, and an obvious carbon peak after 500 seconds of exposure to the beam. The RBS spectra from PMMA resists on silicon substrates have been measured. The predictions of the standard model of cross-linking at high doses has been shown to be inconsistent with the results of these experiments. A model of continuous breaking of bonds until all volatile molecules have left and just carbon remains is more consistent with the experimental observations of this paper. RBS Results
1

PMMA poster_CAARI

Apr 12, 2017

Download

Documents

Frank Odom III
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: PMMA poster_CAARI

#426 RBS Study of the Behavior of PMMA as a Negative Resist for Particle Beam Lithography

A. Brita Brudvig, Paul T. Campbell, John L. Davenport, T. Clarke DeMars, Wilson C. Fricke, Zachery W. Goodwin, Bertrand N. Irakoze, Caroline A. Roberts, Joel A. Stewart, Jean Shirimpaka, Frank K. Odom, Michael R. O’Neil, Randolph S. Peterson

University of the South, Department of Physics and Astronomy

Abstract RBS Results

Conclusions

Explanation of the Negative Resist Behavior

RBS Experiment with PMMA on Silicon

Particle Lithography with Resists

PMMA (Poly(methyl methacrylate)) is widely known as a positive resist for low-dose particle beam lithography, and as a negative resist for high-dose lithography. The model often mentioned for characterizing the behavior of PMMA as a negative resist is one of cross-linking of broken polymer chains at high lithographic doses from electron or particle beams. However, there are a few experiments that do not find evidence for such cross-linking. A Rutherford backscattering experiment was performed with a 2 MeV helium beam on a PMMA resist to observe the oxygen and carbon in the polymer chains as the dose was increased from the positive resist to the negative resist behavior of the PMMA. For a sub-micron and a 1-2 µm layers of PMMA on silicon, only carbon was observed and the thickness of the PMMA appeared to decrease with beam dose until a final thickness of carbon was reached.

References “Preparing patterned carbonaceous nanostructures directly by overexposure of PMMA using electron-beam lithography,” Huigao Duan, Jianguo Zhao, Yongzhe Zhang, Erqing Xie and Li Han, Nanotechnology, Vol 20, 135306 (2009). And references therein.

This schematic diagram illustrates the behavior of PMMA as a positive and a negative resist. On the left the beams of particles damage the PMMA molecules, breaking the long molecular chains into shorter chains, causing the exposed PMMA volume to decrease. Rinsing with a mild solvent will remove only the shortened molecular chains. Removal of a resist that is exposed to a beam is the definition of a positive resist. Continued exposure to the beams of particles further reduces the volume of the exposed resist. Rinsing with a mild solvent will not remove anything, and the use of a stronger solvent removes only the unexposed resist. This is the definition of a negative resist. PMMA shows both the positive and the negative resist behaviors for all energetic charged particles and some wavelengths of electromagnetic waves. Only a few resists have do so.

Si Si

PMMA Resist

Positive resist Negative Resist

Some authors write that as the energy dose (beam dose) to the resist increases, there is a re-linking of the broken molecular chains. The longer and more entwined the molecular chains are, the more resistant is the material to mild solvents. There are a few experimental tests of this negative resist behavior that suggest that the process is not just the breaking and re-linking of molecular chains, but the continuous breaking of molecular bonds, producing a carbonaceous material that is highly resistance to ordinary solvents. This behavior has been observed with electron beams on PMMA, with the gradual extinction of Oxygen K x rays along with the continuous production of Carbon K x rays. The molecular basis for PMMA is C5O2H8. The absence of Oxygen implies the absence of PMMA and not the re-linking of PMMA chains. PMMA is often used as a resist for high-energy (MeV) beams of protons and alpha particles and may produce the same behavior.

A beam of 2 MeV He+ ions was collimated and directed into a scattering chamber with a silicon surface barrier detector at 150o or 160o to the beam. Beam currents were kept as low as possible, but not so low as to cause beam instabilities on target. Beam currents were typically about 10-20 nA. A fine grid before the last collimator was used to monitor beam current during the experiment and the also proved to be a template for our beam lithography. See Figure below. Targets were made by spinning a solution of 950K PMMA (MicroChem) onto a (100) silicon wafer to thicknesses of about 200 nm and 2 µm. The resulting pattern of the beam current grid is visible in the undeveloped PMMA as a negative image below made with an optical profilimeter.

RBS spectra were recorded every 20 seconds and the target was protected from beam exposure while spectra were saved. There were no obvious peaks due to the oxygen in the individual spectra. But the energy at which the helium ions were scattered from the silicon substrate increased with time, forming a plateau shown in the figure below for the 200 nm PMMA on silicon with a detector angle of 150o.

In the figure above, the black line is an empirical fit to guide the eye.

The spectrum above is a sum of all the 20-second spectra. Carbon appears to be present with a weak peak between 400-500 keV. Even though the counting statistics are low, note the near absence of an oxygen peak, which should be located near 700 keV. The backscatter energy for He on uncoated silicon is 1144 keV and the observed energy above is 950 keV. This indicates a residual covering that is almost entirely carbon.

For the thicker PMMA coating on Silicon, about 1-2 µm, the results were similar. The RBS spectra (below) are the sums for individual spectra for the first 140 seconds of exposure (top figure) and the first 500 seconds (bottom figure). There if little evidence for oxygen in either spectrum, and an obvious carbon peak after 500 seconds of exposure to the beam.

The RBS spectra from PMMA resists on silicon substrates have been measured. The predictions of the standard model of cross-linking at high doses has been shown to be inconsistent with the results of these experiments. A model of continuous breaking of bonds until all volatile molecules have left and just carbon remains is more consistent with the experimental observations of this paper.

RBS Results