Maybe some of us have ever heard the concept of CVD (Chemical Vapor Deposition),which represents a chemical process used to produce solid materials with high-purity and high-performance. Based on CVD, Initiated chemical vapor deposition (iCVD) is a novel technique used for deposition of polymeric thin films. Through iCVD, it can be achieved in one step of deposition for thin films of application-specific polymers without any liquid medium.
Figure 1. Reaction chamber of iCVD.
Figure 2. Mechanism of iCVD (Initiated chemical vapor deposition)
As we can see from Fig 1 and Fig 2, the iCVD will process in a reaction chamber, which should be vacuum. The monomers and initiator will flow into the reaction chamber, then they contact resistively heated filaments. Here the initiator will break down into radicals, which results in a beginning of a free-radical polymerization of the specific monomers at the surface of varias substrates, such as silicion wafer, glass, and aluminium. Based on such mechanism, deposition of a wide variety of polymeric thin films can be achieved. Several kinds of monomers can be introduced into the iCVD system to produce polymer films, the films I observed in this project was deposited with Poly(glycidyl methacrylate) (PGMA) as the monomer and aluminum as the substrate under different cooling temperatures. Here I aim at finding whether there are differences between films fabricated under distinct cooling conditions. All the films were synthesized by Ran Tao, PhD student of professor Michell Anthamatten's group, Chemical Engineering Department, University of Rochester.
As mentioned before, the PGMA (Poly(glycidyl methacrylate)) films were deposited at aluminum substrate. The cooling temperatures for deposition were -10℃，-5℃ and 5℃. The polymer films were soft, brittle white thin films. When observing the films with human eyes, films deposited at -5℃ and 5℃ seemed to be uniform. However, the film with cooling temperature of -10℃ appeared to be non-uniform obviously. So I chose different parts of this films to image. Since there are no liquid medium during the depostion process of the films, there was no need to do dehydration. But all the films were insulators, so before imaging, they were sputter coated with gold or platinum in order to make them conductive to avoid the charging effects, which may results in deformation of the images. Both the surface of the samples and the cross-sections of them were imaged. When observing the cross-sections, carbon tapes and graphite was used to fix the samples.
Since the thickness of the polymer films are about 100um, which is too thick for the TEM, here I mainly imaged them under SEM (Scanning Electron Microscope), to get the information of morphologies and the cross-section of them. I also observed the films under an optical microscope;but the depth of field of it is really small. So the images got from optical microscope are only for comparison.
When imaging the samples through SEM, three detectors were tried at first; the Inlens detector, SE2 detector and BSD detector. However, since the polymer films mainly consist of Carbon and Oxygen, whose atomic weights are quite similar, the BSD images are not as good as the ones of Inlens and SE2 detectors. Here the SE2 images seem to be the best one, so most of the images I posted on this webpage are got through SE2 detector.
Figure 3. Images taken under optical microscope of samples deposited at cooling temperature of 5℃ (left) and -5℃(right);
Figure 4.Images taken under optical microscope of samples deposited at cooling temperature of -10℃. Different areas were selected for imaging.
It's clear that the morphologies of -5℃ sample is not smooth, there are lots of ups and downs on the surface, while the 5℃ sample looks much more smooth than the -5℃ one.While for the film deposited at cooling temperature of -10℃, the areas selected appear differently. Sample 1(up left) appears to have some tube-like features at the surface; sample 2 (up right) seems to have close packing droplets on the surface; sample 3 (down left) has a relatively smooth surface; sample 4 (down middle) and sample 5 (down right) seem to have some droplets on the surfaces.We cannot figure out whether all these features are just exist on the surface or everywhere of the films, both on the surface and in the interior. To observe the features in detail, I imaged both the surfaces and the cross-sections of the films under SEM.
Figure 5. SEM images of PGMA film deposited at 5℃;(left) surface of the film, (right) cross-section of the film
Figure 6. SEM images of PGMA film deposited at -5℃;(left) surface of the film,(right) cross-section of the film
Figure 7. SEM images of PGMA films deposited at -10℃(sample 1);(left) surface of the film, (right) cross-section of the film
Figure 8. SEM images of PGMA films deposited at -10℃(sample 2);(left) surface of the film, (right) cross-section of the film
Figure 9. SEM images of PGMA films deposited at -10℃(sample 3);(left) surface of the film, (right) cross-section of the film
Figure 10. SEM images of PGMA films deposited at -10℃(sample 4);(left) surface of the film, (right) cross-section of the film
Figure 11. SEM images of PGMA films deposited at -10℃(sample 5);(left) surface of the film, (right) cross-section of the film
Here we can see that all the features of the films appear only on one side of them, this is because the other side, which is quite smooth, contact with the aluminum substrate. From comparison of the surfaces and the cross-sections of the films, it's obvious features appearing on the surface do not exist in the internal area of the films. This phenomenon indicates that all the features are due to the surface condition, instead of the polymerization itself. A possible explaination for it is that the surface tension causes formation of these features. The surface tension can be caculated through Young's equation,
where where γsv, γSL, γLV is solid-vapor,solid-liquid, and liquid-vapor interfacial tension,respectively. In Young’s equation, θ is solely determined by the chemical composition of solid, liquid, and vapor involved.  To obtain the detail datas, the contact angle or the surface engergy of the materials used should be mearsured and caculated. And we can see from the images of -10℃ that different area appear distinct morphologies, which is really interesting. It's probably because the temperature of substrate is so low that the monomer vapor condensed very quickly on the surfac of the substrate, which caused the nonuniform distribution of the films. To demonstrate this possible conclusion, other cooling temperatures should be chosen to deposite films. And their morphologies should be compared.
According to the obtained images, we can determine that the rough morphology of the films aren't due to the polymerization process but the surface conditions of the films. However, more measurements should be done to acquire the information of the contact angles to caculate the surface tensions. Films deposited at different cooling temperatures should be formed. And maybe other substrates should be chosen to compare the morphology.
 "Initiated Chemical Vapor Deposition (iCVD) of Poly(alkyl acrylates):A Kinetic Model" Kenneth K. S. Lan, Karen K. Gleason.Macromolecules, v 39, p3695-3703 (2006)
 "Conformal Coverage of Poly(3,4-ethylenedioxythiophene) Films with Tunable Nanoporosity via Oxidative Chemical Vapor Deposition" Sung Gap Im,David Kusters,Wonjae Choi,Salmaan H. Baxamusa,M. C. M. van de Sanden,and Karen K. Gleason.ACS Nano, 2008, 2 (9), pp 1959–1967