Keith V. Guinn

Quantitative chemical topography of polycrystalline Si anisotropically etched in Cl2/O2 high density plasmas

The spatially resolved surface chemical composition or ‘‘chemical topography’’ of submicron features [polycrystalline silicon (poly‐Si) masked with photoresist (PR) lines] etched in high density, low pressure helical resonator Cl2/O2 plasmas has been quantitatively determined using angle resolved x‐ray photoelectron spectroscopy (XPS). The chemical topography of plasma etched microelectronic materials is important in understanding how impinging ions and neutrals interact with surfaces to influence etched profiles. The spatial origin of XPS signals was determined from a combination of geometric shadowing of photoelectrons by adjacent features, electrostatic charging of insulating surfaces, XPS signal calibration versus take‐off angle, x‐ray attenuation, and geometric modeling. Equal line and space width (0.75–2.0 μm) features, unmasked poly‐Si, and unpatterned PR surfaces were examined following plasma etching and vacuum sample transfer. For pure Cl2 plasmas, Cl surface concentration was found to be simila...

Mechanism for anisotropic etching of photoresist‐masked, polycrystalline silicon in HBr plasmas

After etching in a helical resonator HBr plasma, polysilicon (poly‐Si) micron‐size features masked with photoresist (PR) were investigated with angle‐resolved x‐ray photoelectron spectroscopy and scanning electron microscopy to spatially resolve surface chemical compositions. The poly‐Si sidewalls and trench bottoms, and the PR sidewalls are covered with one to two monolayers of SiBr and SiBr2. No SiBrx species are present on top of the PR surfaces. No carbon or oxygen were detected on the poly‐Si sidewalls, suggesting that line‐of‐sight deposition of carbon from sputter erosion of the corners and sides of the PR does not occur. Because Br atoms react very slowly with poly‐Si, relative to the ion‐assisted etch rates, anisotropic etching does not require sidewall passivation from products of PR erosion. A SiBrx layer does form on the PR sidewall, however, and could play a role in suppressing lateral erosion of the mask and improving profile control with HBr versus Cl2 plasma etching.

Surface analysis during plasma etching by laser-induced thermal desorption

The use of laser desorption (LD) to desorb species from the surface and laser-induced fluorescence (LIF) to detect them in the gas phase during etching of Si(100) in a high-charge-density plasma of Cl 2 and Cl 2 /HBr mixtures is reviewed. The LD-LIF intensities of SiCI and SiBr are used to track the surface coverages of SiCl x(ads) and SiBr x(ads) , respectively, as a function of RF power, DC bias, and partial pressure, and as a function of time when the plasma is turned on and off. In-line X-ray photoelectron spectroscopy (XPS), the use of which is validated by these in situ LD-LIF studies, is employed after etching to calibrate the surface coverages obtained from the LD-LIF measurements.

Surface chemistry during plasma etching of silicon

Angle-resolved x-ray photoelectron spectroscopy (XPS) and laser- induced thermal desorption (LD), combined with laser-induced fluorescence (LIF) detection, were used to study the etching of olycrystalline Si (poly-Si) and single crystal Si(lO0) in high density (1-2 x 10fl ions/cm3), low presswe (0.5- 10 mTorr) C12MBr-containing, helical resonator plasmas. The XPS measurements on both unmasked Si(100) and fine-line masked poly-Si samples were performed after the sample was etched and then transferred under high vacuum from the plasma reactor to the ultrahigh vacuum 0 analysis chamber. The LD-LIF measurements on unmasked Si(lO0) samples were performed in-situ during etching. In these experiments, XeCl excimer laser pulses rapidly heat the Si surface to near the melting point, causing thermal desorption of SiCl. The tail of the same laser pulse excites Sic1 to the (B2C+) state in the gas-phase near the surface. The subsequent fluorescence signal from this state is proportional to C1-coverage, verified by XPS. In HBr- containing plasmas, analogous LD-LIF detection was used for SiBr, providing a measure of Br coverage. The major findings of these studies are that Si surfaces rapidly become covered with a stable (in vacuum), saturated layer of about 2 monolayers of halogens during plasma etching. The layer consists of silicon mono-, di-, and tri-halides. In C12 plasmas, the C1 coverage increases with increasing ion energy, but is nearly independent of pressure (0.5-20 mTorr). Chlorination occurs rapidly with respect to the time required to etch one monolayer, at pressures as low as 0.5 mTorr. Consequently, the etching rate is limited by the ion flux, and not the neutral flux under these conditions. In mixed C12/HBr plasmas, the coverages of C1 and Br are simply proportional to the total respective halogen content of the feed gas. Other implications for etching mechanisms are discussed.

Cl2 Plasma — Si Surface Interactions in Plasma Etching: X-ray Photoelectron Spectroscopy After Etching, and Optical and Mass Spectrometry Methods During Etching

The interaction of a chlorine plasma with a Si(100) surface has been investigated by angle resolved x-ray photoelectron spectroscopy (XPS), laser-induced thermal desorption with laser-induced fluorescence detection (LD-LIF), optical emission, and mass spectrometry. From XPS, it was found that the amount of chlorine incorporated at the Si surface increases with ion energy, and doesn’t change with long exposure to the plasma. Chlorine is present solely as SiClx (x = 1−3) with average relative coverages of [SiCl]: [SiCl2]: [SiCl3] ≌ 1: 0.33: 0.1. These coverages don’t depend strongly on ion energy between ∼50 and 300 eV. Moreover, there is a substantial amount of disordered Si within the chlorinated layer at high ion energy, reflected in a broadening of the 99.4 eV Si peak and the appearance of a shoulder at 98.8 eV, ascribed to Si with a dangling bond. From modeling of the angle-resolved signal intensities of the Si-chloride species as a function of the XPS take-off angle, thicknesses of 20–35 A and 6–10 A were derived for the SiClx layer at bias voltages of −240 and 0 VDC, respectively. The total Cl content of these layers increased from 1.6×l015 Cl/cm2 at 0V to 3.0×l015 Cl/cm2 at -240 VDC bias. This shows that the top surface layer is predominantly SiCl2 and SiCl3, while just below the surface, mainly disordered Si and SiCl are present. Laser-induced thermal desorption was used to measure Cl-coverage in real time. These measurements, in addition to real time ellipsometry measurements, showed that the layer present during etching is stable when the plasma is extinguished and the gas pumped away.