(Color online) Schematic of experimental setup and electrode. (a) The cryochamber consists of an outer compartment (blue) that can be evacuated to insulate the inner compartment (red)5 adiabatically from the environment. The gas of the inner compartment is cooled by liquid nitrogen ( ) flowing through a loop. The concentrations of the plasma gases Ar, , and are controlled by mass flow controllers (MFC), while the conditions inside the chamber are monitored by pressure and temperature indicators (PI and TI). (b) The microplasma reactor consists of two ITO-covered quartz substrates (thickness of 0.8 mm), a glass slide (thickness of 0.12–0.17 mm) that serves as a shadow mask, and the sample, which is a photoresist-covered Si substrate (thickness of 0.5 mm). To facilitate alignment of the electrode, shadow mask, and substrate, the parts are placed in a custom-made holder and pressed together (the upper part of the holder is not shown). Plasma is generated using a function generator and high-voltage power supply and characterized by current–voltage and optical emission spectroscopy measurements. The optical emission spectra are acquired by placing an optical fiber coupled to a spectrograph on top of the viewport of the cryochamber (not shown in the figure).
(Color online) Photographs (exposure time 30 s) and I–V curves ofplasmas during photoresist ashing. (a) mixture (applied voltage ). (b) mixture. Experimental conditions: . (c) Variation of ; measured and discharge currents ( ) in the mixture during ashing at 170 K. (d) Applied voltage and current waveforms of a discharge generated in at 170 K. Here, shows a single large current peak onto which current spikes were superimposed.
(Color online) Optical emission spectra of two ashing gas mixtures at . (a) gas ( ). The inset shows a detailed view of the optical emission spectrum between 765 and 805 nm, which contains the line of atomic oxygen (777 nm). (b) mixture ( ) With the addition of nitrogen, the emission of Ar-excited neutrals is greatly reduced, and molecular bands become dominant: namely, the second positive (2p: ) and first positive systems (1p: ) of . The inset shows that, when is added, the intensities of atomic oxygen emission at 777 nm and the Ar radicals decrease drastically.
(Color online) XPS spectra of C 1s, O 1s and Si 2p for samples ashed at and 170 K and the and gas mixtures. The intensities of the respective scan windows have been scaled so that they can be represented on the same graph. In the C 1s and Si 2p windows, the lines indicate the positions of the binding energies corresponding to different chemical bonds of the components in the sample: C-C/C-H, C-OH, C = O/O-C-O a shake-up satellite for C 1s, and two peaks that can be attributed to bulk Si ( ) and .
(Color online) Box plot of ashing rates estimated from the variation in the C 1 s composition at , 240, and 290 K. To estimate the ashing rates, the C 1s peak intensities of the individual experiments were treated as independent samples. The horizontal bar inside the boxes indicates the median value, the box boundaries are the 0.25 and 0.75 quantiles, and the horizontal bars below and above the boxes are the minimum and maximum ashing rates, respectively.
(Color online) Variation of the XPS C 1s peak under different ashing conditions. The spectra were acquired for an ashing duration of 20 min, at and 170 K, and for the and gas mixtures to compare with the nonashed sample. To model the C 1 s peak, the contributions of C–H [label (1)] and and bonding [labels (2) and (3)] that were expected to be present in an ESCAP-type monomer (cf. the inset in the nonashed XPS spectrum) were considered. Peak (4) at 291.94 eV is a shake-up satellite that was caused by valence electron transitions in the benzene rings. The black lines indicate the components of the C 1s peak according to the peak model depicted for the nonashed sample, and the colored lines (white for and red for ) indicate the envelope of the peak.
Experimental conditions adopted for ashing experiments.
Article metrics loading...
Full text loading...