Resist Strip Process Research Articles

Megasonic Enhanced Photo Resist Removal Using Unique Solvent Free Phase-Fluids on an Advanced Single Wafer Resist Processing System

Removal of photoresist materials (resist stripping) is very straightforward in theory, but can be difficult and complicated in practice. Classical methods have included SPM (sulfuric peroxide module) /piranha, ozone and UV ozone, solvent based fluids and plasmas. All of the legacy methods include hazardous and toxic materials and/or high equipment expense as well as multi-step complexity. The stripping process becomes even more challenging by factors such as thick resist, cross-linked resist or sensitive metals or materials under the resist. This study was carried out In order to determine feasibility of the use of a unique phase-fluid based photoresist-stripper in commercial semiconductor resist removal process steps. Due to their highly dynamic inner structure, phase-fluids penetrate into the polymer network of photoresists and lift the material from the surface. As these water-based stripping fluids are non-aggressive, non-toxic, and require no special handling, the ability to apply them to industry standard resists will result in a reduction of toxic and dangerous chemistries and the environmental impact of their disposal. To investigate the impact of megasonic energy on the process time in both the active and rinse steps and therefore throughput, the study included the introduction of megasonics as a process variable in all test combinations. The study was broken down into two phases. Phase one consisted of a broad spectrum screening process incorporating nine standard photoresists Fig. 1 with various standard thermal treatments as well as a selection of five phase-fluid formulations. The tests were carried out in a single substrate stagnant immersion tank incorporating a megasonic transducer illustrated in Fig. 2. Process time was defined as the time from immersion of the substrate until the photoresist was completely removed. Process variables tested for each resist/phase-fluid combination included three process temperatures, ambient, 40c and 60c as well as the addition of megasonic energy at ambient with a frequency of 925KHz at a dosage of 1.5W/cm2 directly coupled. The results achieved in this phase one screening were used to select the best case phase-fluid, and process conditions for each of the screened photo resists. These conditions became the baseline parameters for phase two. Phase two of the study was carried out in actual single wafer production conditions on an EVG®100 Series Resist Processor equipped with a MegPie® large area megasonic transducer. Process recipes were established using phase one parameters and then refined for optimum total process time, including the actual phase-fluid process time and rinse time. Spin off and dry cycles were kept constant in all process recipes. Complete resist strip process times of <20 seconds were achieved with several photoresist phase fluid combinations while other combinations involving BCB based resist resulted in little to no removal. The results achieved in phase two testing demonstrate the feasibility of a phase-fluid based photo resist strip in a single wafer platform with megasonic enhancement for specific photo resists. Figure 1

Characterization and Development of High Dose Implanted Resist Stripping Processes

With the increase of implantation dose in new technologies, implanted photoresist stripping is even more challenged in terms of efficiency and substrate consumption. In this work, the effect of implantation parameters (energy and implanted specie) on the photoresist modifications are studied and several plasma chemistries are evaluated to remove it. A good removal efficiency with a low substrate consumption has been found with H2-based processes especially N2H2.

Characterization of half-pitch 15-nm metal wire circuit fabricated by directed self-assembly of polystyrene-block-poly(methyl methacrylate)

We developed a novel but simple sub-15-nm line-and-space (L/S) patterning process that requires neither special pinning guide materials to control the surface free energy nor a resist strip process after the guideline pattern fabrication. To perform process verification for practical semiconductor device manufacturing, we demonstrated electrical yield verification of the process and analyzed the causes of the process defects. One of the causes of open defects is narrower spaces in the upper layer. Improvements in the narrow space points originated from the critical dimension (CD) uniformity, and the roughness at the upper layer is effective for reduction of open defects. The radiofrequency (RF) power in pulsed mode and the depo-etch sequence in the reactive ion etching (RIE) process are effective in improving the surface roughness at the upper layer.

A novel simple sub-15 nm line-and-space patterning process flow using directed self-assembly technology

In order to evaluate process performances of directed self-assembly (DSA) lithography for semiconductor device manufacturing, we developed a novel simple sub-15nm line-and-space (L/S) patterning process using polystyrene-block-poly(methyl methacrylate) (PS-b-PMMA) lamellar block copolymer (BCP). The new process with grapho- and chemo-hybrid coordinated line epitaxy, using trimming resist and shallow etching spin-on-glass (SOG) as pinning guide, requires neither special pinning guide materials to control surface free energy on guideline patterns nor any resist strip process after guideline pattern fabrication. In this process, insolubility and thermal resistance of pinning guide are provided by shallow SOG etching deposition residue consisting of SOG-occurring Si and CF4 gas-occurring CF compound. By optimization of guide resist critical dimension (CD) and resist film thickness, the half-pitch (HP) 15nm L/S patterns after dry development were fabricated with the maximum dose margin of approximately 15%. The HP 15nm L/S DSA patterning after SOG full etching was demonstrated on 300mm wafer with 85 chips. The presence of characteristic 3-dimensional (3D) BCP internal defects related to PS short defects was suggested by observation at SOG half-depth etching condition, and defects could be minimized by optimization of neutral layer (NL) material and sufficient phase-separation annealing. The 15nm PS line patterns were transferred to SOG and spin-on-carbon (SOC) stacked substrate.

Characterisation of C–F Polymer Film Formation on the Air‐Bearing Surface Etched Sidewall of Fluorine‐Based Plasma Interacting with AL2O3–TiC Substrate

C–F polymer redeposition is generated on the etched sidewall of the patterned air‐bearing surface (ABS). This C–F polymer is a by‐product from fluorine‐based plasma using a Surface Technology Systems multiplex‐pro air‐bearing etch (ABE). The morphology of the re‐deposition and the composite element was observed by a scanning electron microscope (SEM). The chemical bonding results were characterised via X‐ray photoelectron spectroscopy, attenuated total reflected infrared spectroscopy and visible Raman spectroscopy. The purpose of this work is to demonstrate a modification of AlF3 re‐deposition to C–F polymer re‐deposition, which is easily stripped out by an isopropyl alcohol‐based solution. The benefit of this research is the removal of the re‐deposition in the resist strip process without additional cleaning process steps.

Non-Oxidizing Solvent-Based Strip of Ion Implanted Photoresist

The removal of ion implanted photoresist (II-PR) after implantation of ultra shallow extension and halo regions is considered as one of the most challenging front-end-of-line (FEOL) processing steps for 32nm and beyond CMOS technology nodes. Commonly used resist strip processes such as fluorine-based dry plasma ash and hot sulfuric/peroxide mixtures induce unacceptable levels of oxidation and material loss [1-.

Low Silicon and SiGe Loss in High Dose Implant Resist Strip

The loss of both silicon and silicon germanium in resist strip process has attracted much attention for generic device development. Such loss mainly stems from the substrate re-oxidation during strip. In this paper, we focus on the high-dose implant resist strip to investigate the re-oxidation of substrate Si and SiGe without compromising the removal capabilities of photoresist (PR) and residue. Besides, the impact of ashing gas ratio on SiN offset spacer critical dimension (CD) is also addressed. Specifically, we studied the substrate re-oxidation performance under various ash process parameters such as pressure, gas ratio, pin position, process time in crust and over-ash (OA) steps. Aside from the re-oxidation effect, we also compared the defect removal capabilities with the above ashing conditions and the dependence of SiN offset spacer CD on ash gas ratio. Results indicate A) The high pressure in preheat step coupled with the low pressure in main-step can effectively preclude Si/SiGe loss. B) For SiGe substrate, the lower ratio of O2/forming gas (FG) exhibits the different impacts between N-MOS and P-MOS implant condition. C) Less OA time brings to better performance for residue removal. D) Continuous offset spacer CD shrink is noticed in each step of lightly doped drain (LDD) loop. The optimized strip scheme delivers not only a residue-free process with the minimum substrate loss but also the improved device performance on both N-MOS and P-MOS.

Demonstration of Modification from AlF&lt;sub&gt;3&lt;/sub&gt; Re-Deposition to Polymer Re-Deposition on Air Bearing Surface (ABS) Etched Sidewall of Fluorine-Based Plasma Etch on Al&lt;sub&gt;2&lt;/sub&gt;O&lt;sub&gt;3&lt;/sub&gt;-TiC Substrate

In this work, C-F polymer rich re-deposition are generated on the etched side wall of the patterned Air Bearing Surface (ABS). This C-F rich polymer is a by product from using a Surface Technology Systems Multiplex-Pro Air Bearing Etch (ABE) tool that utilizes fluorine base plasma. The morphology of the re-deposition and elemental composition are investigated via scanning electron microscope. The chemical bonding information is characterized via attenuated total reflected infrared spectroscopy. The purpose of this work is to demonstrate the modification of AlF3 re-deposition to polymer rich re-deposition allows complete re-deposition removal with isopropyl alcohol base solution. This offers advantage as the re-deposition removal is incorporated during the resist strip process offering superior cleanliness of slider head without additional process steps.

USJ Dopant Bleaching and Device Effects in Advanced Microelectronic Plasma Enhanced Resist Strip Processing

ABSTRACTThe impact of low temperature plasma resist strip on doped silicon surface and microelectronic device performance is investigated using different chemical gas mixtures. In this investigation, different plasma treatments where applied on non-structured and structured silicon on insulator (SOI) wafers post ultra shallow surface implants .The main surface impacts dopant bleaching and oxide loss in conjunction with plasma enhanced re-oxidation are analyzed by time of flight secondary ion mass spectrometry (TOF-SIMS) and electrical measurements of microelectronic test devices. As the result, a long range plasma radiation induced dopant activation and deactivation is separated as the main effect from surface oxide loss and re-oxidation processes. This implies further optimization of plasma resist strip processes for device improvements.

E112 高濃度オゾン液膜を用いた回転平盤上のレジスト除去に関する研究(OS-10 混相流動(1),一般講演,地球温暖化防止と動力エネルギー技術)

The photo resist stripping process in photo lithography plays an important role in manufacturing large-scale integration. In the present study, experimental information is obtained to establish high-concentrated ultra pure ozonized water technology for stripping the photo resist, in stead of chemicals such as sulfuricacid or hydrochloric acid. In the experiment, rotating flow of the ozonized water on the photo resist is irradiated by excimer light, wave length is 222 nm. Ozonized water supplied by cleansing nozzle on the rotating wafer forms liquid film flow. After the process, the thickness of unremoved photo resist on the wafer is measured by ellipsometer. As the result, it is clarified that adjusting the irradiation of excimer light and the control of flow on silicon wafer is inevitable to accomplish high efficiency of photo resist removal rate.

OS10-3 円盤上の回転液膜流れに関する研究(OS10 混相流の計測技術と解析,循環型社会における動力エネルギー技術)

The photo resist stripping process in photo lithography plays an important role in manufacturing large-scale integration. In the present study, experimental information is obtained to establish highly concentrated ultra pure ozonized water technology for stripping the photo resist, in stead of sulfuric acid, hydrochloric acid etc. In the experiment, rotating flows of the ozonized water on the photo resist on silicon wafer is established with the irradiation by excimer light. Ozonized water on the rotating wafer forms Radial Liquid-film flow. After the process, the thickness of unremoved photo resist on the wafer is measured by ellipsometer. And wettability on the photo resist is also measured. As the result, it is clarified that the irradiation of excimer light and the rotating rate of silicon wafer contribute high efficiency of photo resist removal rate compared to merely ozonized water.

Resist Stripping Process on Germanium : a Basic Post-Implant Study

Germanium and Germanium-On-Insulator (GeOI) MOSFETs with high-k gate dielectrics have received recent attention for the advanced technology nodes, because of the better carrier transport properties in Ge compared to Si. For Ge or GeOI CMOS, it is mandatory to determine Ge dedicated resist stripping processes, because of the Germanium non-compatibility with actual cleaning solutions. An initial compatibility study shows a passivation effect on germanium during dry step for high N2/ (O2+N2) plasma ratio. For the post active area etching, dry stripping performed on patterned Poly-Ge-On-Insulator (PolyGeOI) wafers shows good compatibility. The lateral Ge consumption due to the water rinse step is minimized by dry process, indicating a plasma passivation effect. Post implant stripping is especially difficult because the Si typical solutions are highly aggressive for Ge, and also because of the resist graphitization. Using a ramping temperature process, a good resist removal efficiency has been achieved.

Dendrimer‐Scaffold‐Based Electron‐Beam Patterning of Biomolecules

Microarrays have revolutionized the field of genomics and more recently proteomics and have proven to be an asset in high-throughput screening. As the demand for improved sensitivity and throughput of biomolecular assays increases, considerable research effort has been put into developing microelectronic and nanophotonic biosensors, which are presumably more sensitive than conventional fluorescencebased assays and have a faster response time. The lithographic technology for making the densely packed microelectronic devices in a high-throughput manner is already quite advanced and may be integrated to form biosensors. This is expected to increase the pace of research in early detection of disease biomarkers, discovering cell-signal transduction pathways, and in drug discovery. Denser arrays are also important for reducing reagent volume consumption and to improve sensitivity. Successful biomolecule patterning on sensor chip circuitry requires a number of important steps. First, selective immobilization of the probe and reduction in non-specific binding should be achieved for higher signal-tonoise ratios. Second, reduction in the sensor size reduces the background, enhancing the signal-to-noise ratio, and therefore biomolecule patterns on the same order as the sensors are desirable. Third, the biomolecule pattern should be aligned with the sensor circuitry, which becomes more difficult as sensor size decreases. Finally, a fabrication process should be formulated to ensure that the biomolecules are intact and functional, which is a challenge given the harsh microor nanofabrication processing steps. Here, we have demonstrated an electron-beam (e-beam)-based approach fulfilling the above requirements for patterning biological macromolecules that does not involve the use of resist, hence eliminating the exposure of these biomolecules to harsh resist-stripping processes that are normally employed to remove the resist. A non-biofouling poly(ethylene glycol) self-assembled monolayer (PEG-SAM) was selectively removed by e-beam and patterned with aldehyde-terminated polyamidoamine dendrimer (ald-PAMAM-SAM) in a layer-by-layer (LbL) manner to covalently immobilize the aminated oligonucleotide, which bind only to their complementary sequence targets and can be stripped and reprobed. The Generation-6 (G-6) PAMAM molecule, terminated with 256 primary amine groups and 6.7 nm in diameter, was used to increase the surface density of aldehyde functional groups to increase the oligonucleotide-immobilization efficiency. Current techniques for patterning biomolecules involve the use of polymer-based templates, which can be removed mechanically without the use of organic solvents after biomolecule immobilization. However, serious limitations exist in each case. Poly(dimethylsiloxane) (PDMS)-based soft-lithographic techniques cannot be used to create high-resolution patterns, as aligning the PDMS pattern with sub-micrometer features has been shown to work in a mix-and-match manner with an accuracy of only 2 lm. Although alignment is not an issue for biomolecule patterning based on polymer liftoff, as it is an integrated process, this method involves extra steps of deposition and etching a polymer film, which increases the complexity of the process. Challenges are also encountered as the size of the photolithographic patterns decrease due to the increase in line-edge roughness (LER) and the isotropic nature of oxygen plasma etch. Patterned gold has been used for creating protein patterns using thiolbased linkers, but gold surfaces cannot be tolerated in some biosensors as gold interferes with the optical signal or conductivity of the sensor. This technique also includes extra photolithographic and lift-off processing steps for patterning gold. Protein patterning using fluorescence-tagged proteins C O M M U N IC A IO N S

Impact of reducing resist stripping processes at elevated temperature on ULK and HM materials

Ultra low- k materials for interconnects in chip manufacturing require new or adapted procedures for integration into copper damascene metallization, including patterning processes. This work deals with the impact H 2 containing stripping processes at elevated temperatures on porous ultra low- k materials and low- k hard mask/capping materials. Porous MSQ and SiO 2 and a low- k hard mask have been prepared by spin on deposition (SOD) and for comparison dense SiOCH by PECVD deposition. Resist stripping was done with a downstream microwave discharge in an advanced strip passivation chamber (ASP) of Applied Materials using mixtures of H 2/N 2 and H 2/He. Shrinkage (change of thickness) and electrical properties ( k-value and field break down) were investigated. FTIR spectra were used to support the investigation. Ashing parameters, like time of plasma treatment, temperature and gas composition, have been characterized. It was found that the advantage of nitrogen admixture is the high rate of ashing, but shrinkage and impact on k-values will be higher than with He admixture. Helium admixture reduces shrinkage and k-values will be only slightly impacted, but stripping time has to be increased for resist removal.

Qualification of Resist Strip Process for Ultra Low-k/Cu Interconnect

Qualification of Resist Strip Process for Ultra Low-k/Cu Interconnect

Dewetting of resist/metal bilayers in resist stripping processes

We report the observations of dewetting of resist/metal bilayers in a resist stripping process of nanofabrication in O2 plasma. The initiation of the dewetting process is tentatively associated with local heating caused by surface plasmon induced in metallic nanoparticles or nanowires. The surface patterns thus formed differ substantially from all the dewetting patterns reported so far, and they resemble trees at micrometer scale. The possible mechanism for the formation of this kind striking patterns is discussed and its implication to future nanoelectronics manufacturing is addressed.

Direct Resist Removal Process from Copper-Exposed Vias for Low-Parasitic-Capacitance Interconnects

A resist stripping process from Cu-exposed vias is developed to reduce the thickness of high permittivity (high-k) SiN in Cu/low-k interconnects. A low power, low temperature reactive ion etch is proposed to suppress the significant Cu oxidation. Additionally, removing Si contents from the Cu surface is found to be critical. This resist stripping process is successfully applied to the fabrication of Cu/low-k interconnect test devices using silicon oxycarbide Low resistance, 0.25 μm diam via connections were achieved. Time dependent dielectric breakdown lifetime tests showed that even 25 nm thick SiN has a sufficient barrier property against Cu diffusion, showing that the SiN thickness reduction is also feasible from the reliability viewpoint. By thinning SiN to less than 25 nm, over 5-10% reduction in k can be achieved even when using the same low-k interlevel dielectric. © 2001 The Electrochemical Society. All rights reserved.

Advanced silver-based metallization patterning for ULSI applications

Silver metallization is being investigated for potential use in future integrated circuits. Unlike the proposed copper metallization, Ag thin films can be reactive ion etched at reasonable rates using a CF 4 plasma. This etch technology is an atypical ‘dry-etch’ process since the formation of volatile products is not the main removal mechanism. The primary film removal mechanism, however, is the subsequent resist strip process. The effects of process conditions on the etch rate and post-etch surface roughness is also characterized. Our study shows that the silver etch process in the CF 4 plasma depends strongly on the reactive neutrals and the removal rate is enhanced significantly by the presence of energetic ions as well.

Structural Changes in a Resist Resulting from Plasma Exposure during the Reactive Ion Etching Process

Structural changes in a resist resulting from plasma exposure during the reactive ion etching (RIE) process were studied using 13C-NMR analysis and Fourier transform IR spectroscopy (FT-IR). Two types of resist were chosen containing different amounts of diazonaphthoquinone derivative as an initiator with the same type of base polymer, cresol novolak resin. It was found that a slightly soluble layer was formed by the cross-linking reaction of the main chain of novolak resin as a result of plasma exposure in the case of the resist with a low amount of photosensitizer. The molecular weight concurrently decreased, probably due to the chain-scission of the main chain of novolak resin. On the other hand, in the resist containing a larger amount of photosensitizer, the molecular weight of the material increased, probably due to the occurrence of some type of cross-linking reaction such as thermal condensation. From these results, we obtained some guidelines for developing a resist stripping process without a residue.

Novel technique to pattern silver using CF[sub 4] and CF[sub 4]/O[sub 2] glow discharges

Silver is being studied as a potential metal for future interconnects in integrated circuits. Under appropriate conditions, Ag thin films can be removed at reasonable rates using a CF4 or CF4/O2 glow discharge followed by a conventional resist strip process. This is not a typical “dry-etch” process where the formation of volatile products is the main removal mechanism. Rather, it is novel in that the glow discharge is employed to form reactive byproducts of Ag. There are simultaneous sputtering and desorption of the byproducts during the plasma exposure; but the primary film removal mechanism is in the subsequent resist strip process. Two-level-factorial design, and one-factor-at-a-time experiments are employed to study the effects of process conditions on the etch rate, and postetch surface roughness. It is found that Ag removal rate in CF4 or CF4/O2 glow discharges depends strongly on applied power. Postetch surface roughness depends on both power and pressure.