Figure 1. Fabrication of Janus membranes using physicochemical
inhibition. (a) A schematic of a three-step process for the fabrication
of a Janus membrane. Step (i): the hydrophilic support is filled with
glycerol/inhibitor (HQ) mixture. Step (ii): the filled membrane is
placed on the back-cooled stage (15 oC) of the iCVD
reactor, and the PTFE deposition is performed for a specific time (5-15
min). Step (iii): the glycerol/inhibitor mixture is removed from the
sample by placing it inside an isopropanol bath. (b) A top-down
representative scanning electron microscopy (SEM) image of the
hydrophilic PVDF support. (c) A top-down SEM image of the Janus membrane
surface showing the PTFE structure deposited on the top surface of the
support without blocking the pores. (d) A cross-section SEM image of the
Janus membrane surface showing the PTFE structure deposited on the
hydrophilic PVDF support. The dashed line shows the PVDF-PTFE interface.
Here, the pressure and time of deposition were 900 mTorr and five
minutes, respectively.
Although it has been shown that, 28–30 by changing
the iCVD condition and moving from a diffusion-limited process toward a
reaction-limited process, we are able to apply polymeric coating within
the entire porous substrates, selective area deposition of polymers has
not yet been achieved. Here, using the developed approach, we show that
the PTFE deposition can be limited to the solid domains on the top
surface of the substrate. As a result, the surface porosity of the
substrate does not decrease after the coating. In the developed method,
we note two different inhibition mechanisms: physical and chemical. Once
the primary radicals are absorbed into the glycerol mixture, the
relatively faster rate of diffusion, compared to that of the surface
diffusion, prevents the formation of a polymer film on the liquid
surface. As a result, the deposition of PTFE on the liquid domain will
be delayed. The proposed mechanism can be found in Section S2,
Supporting Information. In addition to the physical inhibition, the
chemical inhibition also exists because of the quenching of the absorbed
radicals in glycerol through the HQ present in the liquid mixture.
To unravel the difference between physical and chemical inhibition, we
performed similar experiments on liquid films. For this purpose, the
liquid films were prepared via spin-coating of the glycerol/inhibitor
mixture on pieces of the silicon wafer with comparable areas. The
detailed information for sample preparation can be found in Section S3,
Supporting Information. Figure 2 (a) shows the Fourier transformed
infrared spectroscopy (FTIR) of liquid samples before and after five
minutes of PTFE deposition. As shown, the FTIR spectra of
glycerol/inhibitor mixture remained unchanged after five minutes of iCVD
deposition, indicating that no detectable PTFE was deposited on the
surface of the glycerol/inhibitor. On the other hand, the PTFE
characteristic peaks, located between 1150-1225 cm-1,
were salient in the spectra of the pure glycerol sample processed in the
iCVD chamber. This observation indicates that, during the first five
minutes of reaction, HQ molecules present in glycerol quenched the
adsorbed primary radicals. Increasing the deposition time to seven
minutes, we found that the inhibition process becomes ineffective, and
the characteristic peaks of PTFE were again observed in the spectra. The
data showing the FTIR of the samples processed for a longer time are
shown in Section S3, Supporting Information.