Search

The authors evaluated photoresist (PR) stripping processes that are compatible with ultralow dielectric constant (ULK) materials using H2-based remote plasmas generated in an inductively coupled plasma reactor. The materials used were 193 nm PR and nanoporous SiCOH-based ULK (JSR LKD 5109). PR ashing rates and ULK damage (carbon depletion) were measured for H2, H2/N2, and H2/Ar discharges as a function of substrate temperature over the range of 200–275 °C. They employed ellipsometry, x-ray photoelectron spectroscopy (XPS), optical emission spectroscopy, and time-of-flight secondary ion mass spectroscopy (ToF-SIMS) for analysis. For their H2 remote plasmas and a substrate temperature in the range of 200–275 °C, the PR ashing rate varied from 270 to 880 nm/min, whereas 3–5 nm of ULK damage was measured for 20 s remote plasma exposure. As a useful process metric, they defined ashing efficiency as the thickness of PR removed over the thickness of ULK simultaneously damaged. PR stripping processes can be optimized to an ashing efficiency of ∼60 for substrate temperatures above 250 °C, if pure H2 discharges are employed. The addition of N2 or Ar to H2 did not improve the ashing rate and, especially for N2, such additions dramatically increased ULK damage. This resulted in reduced ashing efficiency for these cases. To clarify the impact of etching/ashing process interactions on ULK modification, they exposed blanket ULK film to C4F8/Ar-based etching plasmas employing a dual frequency (40.68/4 MHz) capacitively coupled plasma (CCP) reactor. Plasma exposures of the ULK were performed utilizing a silicon roof, which shielded the ULK film located underneath from direct ion bombardment. Since the aspect ratio of the small gap structure was selected to be equal to that of an actual trench structure, trench sidewall-like surface modifications induced by etching processes along with their impact on ashing damage that were introduced during a subsequent PR stripping process can be simulated and studied on blanket films with appropriate size. XPS revealed fluorocarbon (FC) deposition together with ∼3 nm of ULK damage on the ULK film surface after the FC plasma etching process. Most of the deposited FC material was removed during a subsequent H2-based remote plasma treatment at 275 °C. The influence of surface modifications introduced by the prior C4F8/Ar-based etching exposure on hydrogen permeation of the ULK material during a subsequent H2 remote plasma ashing process was studied by substituting deuterium (D2) for H2 in the remote plasma process and performing ToF-SIMS analysis. ToF-SIMS depth profiling of ULK films exposed to D2 plasma showed reduced D permeation in the ULK films with C4F8/Ar etching plasma exposure relative to that without such FC plasma exposure. Photoresist patterned ULK structures were also processed, employing the same ashing conditions after prior FC plasma etching in the CCP reactor. The ULK damage results measured with trench structures were consistent with the above findings obtained with blanket ULK films. [ABSTRACT FROM AUTHOR]

As the device dimensions scale to 100 nm, the use of photoresist materials is suitable for lithographic patterning at 193 nm. The molecular structure of 193 nm photoresist materials is significantly different from that of 248 nm photoresist materials [H. Ito, IBM J. Res. Deu. 45, 683 (2001), T. Kajita et al., Proc. SPIE 4345, 712 (2001)], which leads to a number of undesirable consequences, including pronounced surface and line edge roughness during plasma etching [H. Ito, IBM J. Res. Deu. 41, 69 (1997), [E. Reichmanis et al., J. Vac. Sci. Technol. B 15, 2528 (1997), [L. Ling et al., ibid. 22, 2594 (2004)]. In this article, we present an investigation of the mechanisms for the surface/line edge roughening of photoresist materials during plasma etching using C4F8/90% Ar discharges. We emphasized in our study short exposure times (the first few seconds) of the photoresist materials and structures to the plasma, a time regime that has not been well studied. Rapid modifications were observed for both 193 and 248 nm photoresists during short time exposure. During the first seconds of plasma exposure, photoresist material densification and hydrogen depletion are important processes. It is also found that rough surfaces develop within a few seconds of exposure to the C4F8/90% Ar discharges. Plasma exposure leads to the formation of rough edges on the top of trench sidewalls in photoresist trench and line structures. During prolonged exposure to the plasma, the roughness is transferred to produce striations on the sidewalls. After an initial stage, the roughening rate remains constant for 193 nm photoresist, whereas for 248 nm photoresist the roughening rate is negligible. This difference is possibly related to the preferential removal of carbonyl groups for the 193 nm photoresist material, which has been revealed by x-ray photoelectron spectroscopy and seconday ion mass spectroscopy. [ABSTRACT FROM AUTHOR]

Plasma-based ashing of photoresist masks after pattern transfer is a common processing step in the fabrication of integrated circuits. In this work we investigated damage mechanisms of nanoporous ultra low k (ULK) materials with different overall porosities due to the ashing process. Oxygen-, nitrogen- and hydrogen-based photoresiststripping using direct and remote plasma processes were examined. Ellipsometry, x-ray photoelectron spectroscopy, secondary ion mass spectroscopy, and transmission electron microscopy were utilized to study the damage layer thickness, physical (pore morphology), and chemical modifications of the nanoporous silica thin films after exposure to the O2-, N2- or H2-based ashing processes. As a result of the plasma exposure, carbon groups in nanoporous silica can be removed from the ULK layers which is also accompanied by material densification. We find severe ashing damage of ULK materials after O2-based ashing using both direct and remote discharges. N2 and H2 discharges also damage ultralow k materials for direct plasma ashing processes which are accompanied by low energy ion bombardment of the substrates. The introduction rate and degree of the ULK materials modifications correlates with the overall porosity. We show that the pore interconnectivity is one of the key parameters that determine ashing damage. ULK damage is greatly reduced for remote N2 or H2 discharges, but the resist removal rates are impractically low if the substrate is at room temperature. We show that both acceptable photoresist stripping rates and ULK damage levels can be achieved for remote H2 plasma ashing processes if the substrate temperature is 250 °C and higher. [ABSTRACT FROM AUTHOR]

Recent experimental studies of fluorocarbon (FC) plasmas etching various substrates suggest that ions will transport initially bound fluorine (F) through overlying FC films, thereby defluorinating these films and inducing fluorination reaction with the underlying substrate material. Simulations of thermal CF[sub 2] on Si with simultaneous bombardment by energetic Ar[sup +] demonstrate this defluorination phenomenon, showing that F is separated from adsorbed CF[sub 2] and mixed into the underlying Si, initiating etching. Additionally, this creates dangling bonds on the surface where CF[sub 2] may adsorb. Thus, our simulations show that F and C uptake is enhanced by energetic rare gas ion impact, the number of Si–F bonds is greatly increased, and the resultant Si etch rate is higher than expected from physical sputtering alone. The results are compared to experimental measurements made under similar conditions, and the mechanisms of ion-induced F transport are identified. © 2004 American Institute of Physics. [ABSTRACT FROM AUTHOR]