Initiated chemical vapor deposition (iCVD) is a gentle, low energy processing, non-wet chemistry scheme to grow thin polymer films directly from gas phase feeds. The technique involves feeding low molar mass gaseous initiators and monomers into a low pressure chamber. By co-depositing non-reactive molecules (porogens) with reactive monomers and crosslinkers, it is possible to induce and arrest phase separation of deposited species. Polymerization, crosslinking and phase separation are intended to occur simultaneously on the substrate, resulting in a vitrified macroporous thin film. Differences in the cohesive energy density of the polymer and the porogen should drive phase separation, and, therefore candidate porogens were chosen on the basis of their solubility parameters.
Recent experiments of our research group have successfully conducted iCVD with ethylene glycol (EG) as a porogen, and glycidyl methacrylate (GMA) as monomer. When both the porogen gas and monomer gas’s degree of saturation exceed unity, a porous polymer film is supposed to be observed, and a skin-like layer would be expected above the porous film. We are able to block the feeding of any involved species at any time by simply turning off the valve of that gas, this makes possible a facile fabrication of a layer-by-layer thin film with desired thickness of porous part and continuous solid part.
Through Scanning Electron Microscopy observation, we would be able to characterize the cross-sectional and surface morphology as well as the phase-separated bulk area with micrometer-sized pore continuously dispersed inside.
Above: Mechanism for the formation of the porous film via CVD, EG were applied as porogen to induce phase separation and porosity.
Materials and Methods:
Above is a sample thin film made through iCVD technique. Samples were provided by Ran Tao from the Department of Chemical Engineering at the University of Rochester.
The sample appears white because it is porous inside and diffracts visible light (a solid thin film tend to be transparent). The sample would be in wet state with liquid porogen once made. Either vacuum oven or Critical Point Drying (CPD) technique can be applied to receive a stable, dry sample, observable in SEM. After small pieces of sample were chosen and stuck onto a slope stage with conductive tape, sputtering coating would be applied, this together with CPD are two major preparation step required for the sample. Once the sample is loaded onto the stub and transferred into the SEM chamber, it is especially important to apply 45 degree tilt of the stage, since We were mostly interested in the cross-sectional structure of the thin film because it allows for the determination of the porosity. When doing this, special carefulness needs to be taken in order to make sure the stage does not hit the gun.
The gold sputtering coater (left) and Zeiss Auriga Crossbeam SEM System (right) at the University
of Rochester used for this project.
Results and Discussions:
Porous film with skin layers:
The expected porous thin film is as illustrated in the following. Porogen liquid phase separate from the system and is evaporated away to afford interconnected micrometer-sized pores. The thickness of the porous bulk material can be precisely determine by the feed time of the reagents. The mechanism of the formation of the skin layer is still unknown, but this structure can be essentially important in rendering a relatively tough thin film.
Illustration of the phase-separated porous thin film.
By far the most important aspect of this project is of course observing porous structure Using SEM. We applied different SEM conditions and determined SE2 detector with a working distance at accelerating voltage around 10 kV to be the optimum. From the captured pictures of cross-sectional view we can see the flat film is about 30 micro-meters thick, with continuously dispersed porous phase. The polymer thin film bulk is with a 1-10 micrometer pore-size, which is in accordance with the macrophase separated morphology of the system.
The uniform thickness distribution of the skin layer can be witnessed by the up left image, the porous bulk is connected with this solid dense layer and the interface appeared rather obvious. The surface of the skin layer is shown as bottom left, seeming ups and downs made the plane roughness rather significant.
SEM images of sample with different magnifications and features.
Sample after DSC:
The sample was subjected to some thermal analysis like Differential Scanning Calorimetry (DSC). During the test, the piece of thin film was put in a tiny stage with heat programmedly applied in the bottom. SEM pictures were taken after DSC to determine the thermal behavior of the material. A rather different structure was discovered, Although thickness of the film remained roughly unchanged, the bulk morphology was quite different from sample before heating. The size of the PGMA scaffold apparently got larger, and the bottom part of the film which was previously porous became solid. This surprising morphology may suggest that the material to some extent "melt down", since the glass-transition temperature of PGMA is relatively low around 100 ℃ while the upper temperature limit of the DSC test was 100 ℃. The formation of the thick solid layer in the bottom may be induced by gravity, although for a micrometer-thick structure, this effect should not be significant.
Porous thin film after DSC characterization, quite different structure was observed.
The top skin layer is quite a special structure in this very material. It is grown beneath the porous bulk, with a sometimes rough surface. The interfacial interaction between the two phases is not strong, so the top layer can actually be easily peeled off, exposing the inner porous scaffold material. But this structure is strong enough to hold the fragile bulk material together, as can be seen in the bottom left image. Except for the first picture, the other three pictures look really terrible, charging occurs from time to time, and there is always positions without enough coating that appear dark, thus the captured picture from a depth perspective always looked either too dark or too contrasted.
Images of the skin layer structure
Some photoshop colorization:
The below two pictures were criticized by Brian as "too contrasted and too dark". Yes they are... but I did not think of these stuff when taking them. They were taken for colorization for the special morphology displayed.
Bad SEM image... so-so PS work...
I thought of wafer when taking the image, but my roommate said Twix should work.
OK, I was hungry shooting this image.
Porous structure of thin film made with iCVD technique was characterized with SEM. Both cross-sectional and top views of the 30 micrometer thick flat material were visited. Skin layer and the bulk material displayed drastically different features. Further work can focus on the characterization of a layer-by-layer structure made by timely switching on porogen feed.
I would like to thank Brian McIntyre for his brilliant lecture and great tolerance over my switching the topic half way and repeatedly showing him ugly pictures. I would also like to show my gratefulness to Rohit Puranik who did a phenomenal work as Teaching Assistant.
PS, Brian did not really appear to know what the date it is when giving the pre-class speech!
Chemical vapor deposition of conformal, functional, and responsive polymer films.
Alf, M. E. et al. Adv Mate, 2010. V 22 (18): p. 1993-2027