Photoresist Removal Research Articles

オゾンマイクロバブルを用いた高機能洗浄技術

The removal technology of single-wafer processing using ozone is proposed on photoresist removal in the semiconductor manufacturing. Ozone is friendly for the environment and no harm to the product. Before now, the authors indicated that disk-shaped nozzle forms a vortex structure between the disks which is considered to cause the reduction of photoresist removal rate. This study proposes a disk-shaped convex nozzle to suppress the vortex. In order to investigate the effects of the presence or absence of the convex, the removal experiments were conducted. As a result, disk-shaped convex nozzle had the higher removal ability than disk-shaped nozzle in the convex part. Next, as a new removal method, we suggest the ozone microbubble with a Venturi tube. To investigate the performance of the photoresist removal with it, first, the generation of ozone microbubbles is observed in the Venturi tube via a high speed video camera. Next, to clarify the effect of the superficial liquid and gas velocities, the ozone microbubble, and presence or absence of convex, ozone water concentration measurement and photoresist removal experiments were conducted. As a result, the ozone water concentration increased with increasing superficial velocity and the remaining photoresist thickness reduced. Also, ozone microbubble had the higher removal ability than ozone water single-phase. Compared presence and absence of convex, convex influenced photoresist removal efficiently. In addition, compared with the existing study, it is suggested that the superficial liquid and gas velocities and ozone water concentration affect the photoresist removal rate.

Characterization of Clean after Photoresist Removal from Wafers with Copper Pillars with Lead-Free Solder Caps

Opportunities for developing new and enabling packaging schemes are being pursued as part of device improvement strategies for electronic products. Processes such as embedded technologies in wafer level packaging and 3-D chip architecture schemes open up opportunities for realization of a variety of package configurations. As a result, there are many opportunities to impact both device performance and the processes used to create them. In the area of electroplated solder applications, there has been widespread adoption of copper pillar bumps with lead free solder caps. Solder wetting of only the copper pillars after reflow confines the spread of the solder cap in the x,y plane enabling tighter pitch layouts, while the smaller solder cap still improves the robustness of the interconnection. However, as more data is gathered using this process, one area of growing interest is the characterization of polymer photoresist residue on the side walls of plated copper pillars with lead-free solder caps. As has been seen with other processes including for example, redistribution line (RDL) cleaning processes, cleaning challenges have increased in tighter pitch applications. In RDL, cleaning challenges were increased as cleaning processes that were once taken for granted were now being applied to wafers with 10μm line/space and tighter pitches. Analogies can be found in the tight pitch copper pillar cleans. Typical characterization techniques such as optical microscopy have been used traditionally in the industry as a post process inspection method. Advantages such as ease of use, fast throughput, automated tooling, use of filters to highlight certain defect types, and the ability to non-destructively inspect whole wafers have all contributed to its success in the field. With tighter pitch copper pillar applications, inspection using optical microscopy is challenged in an extreme and it is possible to inspect a clean copper pillar wafer and not see photoresist residue on the copper side walls. This paper will describe wet cleaning processes used today for copper pillar cleans, with special focus on characterization of the Cu pillar and solder side walls. Examples using several different photoresists and copper and solder configurations will be examined. Samples with differing copper/solder bump heights and pitches will be discussed. Cleaning and compatibility results will be shown.

Enhancement of Removal Uniformity by Oxygen Addition for Photoresist Removal Using H Radicals Generated on a Tungsten Hot-Wire Catalyst

Enhancement of Removal Uniformity by Oxygen Addition for Photoresist Removal Using H Radicals Generated on a Tungsten Hot-Wire Catalyst

Lift-Off Patterning of Nano-Crystalline Quantum Dot Films

The ultra-thin films formed using colloidal solutions of nanocrystalline quantum dots (NQD) show good promise in display/LED, as well as nano-labeling applications between others (e.g. [1]). Regardless of application, to become a part of the functional device the NQD film must be patterned accordingly. The conventional top-down patterning sequence involving photolithography and etching cannot be used in this case as the process would results in the critical damage to the NQD film. Hence, alternative patterning technologies suitable for NQD films patterning need to be pursued. The goal of this contribution is to continue exploration of an approach to NQD film patterning based on the lift-off principle [2]. In electronic/photonic device manufacturing the lift-off process is a method allowing patterning of the thin-film using a sacrificial material, most commonly a photoresist. The lift-off patterning is employed to created desired geometries in the films which are either very difficult to etch (e.g. gold) or are not sufficiently structurally coherent to be subject to a conventional photolithographic process without enduring permanent damage. The thin-films comprised of nano-scale materials such as nandots, nanowires and nanotubes fall into this last category. The principle of the NQD film pattering using lift-off process is demonstrated in Fig.1. In this study NQDs deposited were CdSe/ZnS core/shell NQDs featuring average diameter of 6.5 nm and corresponding red (620 nm) emission wavelength. Two deposition methods used in this study was a mist deposition and a conventional spin coating. The method of mist deposition converts colloidal solution, in this case CdSe/ZnS NQDs suspended in toluene, into a fine mist of sub-micrometer sized particles using an atomizer. The mist droplets are carried from the atomizer to the deposition chamber by ultrapure N2carrier gas and then coalesce on the substrate in the presence of an electric field. The same colloidal solution was used in the preparation of the samples by spin coating. Silicon wafers used as substrates are first covered by spin coating with a layer of positive 1811 photoresist to the thickness of about 1 µm and then cured at 100oC. The resist covered wafers were subsequently exposed through the standard Cr-photomask in an aligner/exposure tool following which photoresist was develop using in 351 developer (1:4) diluted with Di water for 1 minute (Fig.1a). Then, CdSe/ZnS QD film was blanked deposited during 20 min. long mist deposition process and thermally cured (Fig. 1b). Then, the samples were exposed for 20 min. to 20% O2in nitrogen plasma at the pressure of 1 Torr at the power of 200 watts (Fig. 1c). In another experimental run instead of plasma ashing resulting in the partial NQD film lift-off, the photoresist removal was carried out by immersion of the wafer in acetone. The results obtained in this study demonstrate effectiveness of the lift-off process in pattern definition of NQD films. Figure 2 shows patterns created in NQD films using proposed technique. In the full account of this work a complete set of experimental results obtained in the course of this investigation is presented and possible applications of the processes under investigation is considered.

Effect of Microbubbles on Ozonized Water for Photoresist Removal

In this paper, we describe the use of ozone microbubbles in photoresist removal from silicon wafers. Ozonized water has attracted much attention as an environmental friendly cleaning method in semiconductor manufacturing. However, it would be desirable to enhance the oxidative ability of ozonized water for practical application. The existence of microbubbles in ozonized water has been shown to significantly enhance the photoresist removal rate due to an elevated dissolved ozone concentration (approximately 2.5 times that of ordinary ozone bubbling) and a direct effect of the microbubbles (removal rate is approximately 1.3 times faster than water with the same concentration of dissolved ozone without microbubbles). Additionally, the ozone microbubble solution was able to effectively remove a high-dose ion-implanted photoresist, which is extremely resistant to removal by ozonized water and other wet chemicals because of its amorphous carbon-like layer, or “crust”. Electron spin resonance experiments were al...

Photoresist and Thin Metal Film Removal by Steam and Water Mixed Spray

Photoresist and Thin Metal Film Removal by Steam and Water Mixed Spray

Enhancement of Photoresist Removal Rate by Using Atomic Hydrogen Generated under Low-pressure Conditions

Enhancement of Photoresist Removal Rate by Using Atomic Hydrogen Generated under Low-pressure Conditions

Modified microscale pattern transfer without photolithography of substrates

Micro-dimple arrays have recently become a popular feature in tribology for improving the friction and lubrication performances of various mechanical components. Microscale pattern transfer without photolithography of substrates is a common electrochemical micromachining method to generate these micro-dimple arrays with controlled size, location, and density. It is still a challenge to reduce the ratio of etched micro-dimple diameter and mask hole diameter in microscale pattern transfer without photolithography of substrates. In this paper, a modified microscale pattern transfer without photolithography of substrates, through movable dry-film mask electrochemical micromachining, is proposed to allow a reduction of the ratio. In this method, an electrochemical tool carrying a dry-film mask with the pattern of perforations keeps in close contact with the anodic workpiece surface during electrochemical machining. The proposed technology offers some unique advantages such as re-use of the mask and avoiding removal of photoresist. Simulation results indicate that the presented method might reduce the ratio, and experiments verify that this approach is effective in reducing the ratio. Finally, 22,500 micro-dimples of 109.4μm in diameter and 15.1μm in depth were prepared in 4min, with the ratio of micro-dimple diameter and mask hole diameter being only 1.09 at the etched depth of 15.1μm.

A high-speed photoresist removal process using multibubble microwave plasma under a mixture of multiphase plasma environment

This paper proposes a photoresist removal process that uses multibubble microwave plasma produced in ultrapure water. A non-implanted photoresist and various kinds of ion-implanted photoresists such as B, P, and As were treated with a high ion dose of 5 × 1015 atoms/cm2 at an acceleration energy of 70 keV; this resulted in fast removal rates of more than 1 μm/min. When the distance between multibubble microwave plasma and the photoresist film was increased by a few millimeters, the photoresist removal rates drastically decreased; this suggests that short-lived radicals such as OH affect high-speed photoresist removal.

Ultraviolet/ozone treatment to reduce metal-graphene contact resistance

We report reduced contact resistance of single-layer graphene devices by using ultraviolet ozone treatment to modify the metal/graphene contact interface. The devices were fabricated from mechanically transferred, chemical vapor deposition grown single layer graphene. Ultraviolet ozone treatment of graphene in the contact regions as defined by photolithography and prior to metal deposition was found to reduce interface contamination originating from incomplete removal of poly(methyl-methacrylate) and photoresist. Our control experiment shows that exposure times up to 10 min did not introduce significant disorder in the graphene as characterized by Raman spectroscopy. By using the described approach, contact resistance of less than 200 Ω μm was achieved for 25 min ultraviolet ozone treatment, while not significantly altering the electrical properties of the graphene channel region of devices.

The effects of supercritical CO2 formulations on the removal of high-dose ion-implanted photoresists

Highlights3wt.% EH-3 and TMN-3 are chosen to form scCO2 microemulsions.2vol.% DMSO and IPA can enhance HDI PR efficiency by 75.5% and 64.2%.DMSO and EH-3 show a positive synergistic effect on the stripping efficiency.scCO2 density tuned by temperature and pressure affects HDI PR stripping behavior.100% Removal efficiency is achieved at 60?C and 25MPa for 20min. A new effective method for high-dose ion-implanted photoresist (HDI PR) stripping is reported based on supercritical CO2 (scCO2) microemulsions consisting of branched hydrocarbon surfactant 2-ethylhexanol polyoxyethylene-polyoxypropylene (EH-3) and co-solvent dimethylsulfoxide (DMSO). The optimized concentrations of EH-3 and DMSO in scCO2 were 3wt.% and 2vol.%, respectively, and the weight ratio of H2O and EH-3 is 1:1. The HDI PRs containing crust layers were removed in the scCO2 microemulsions at 10-25MPa and 40-80?C for 20min. The results revealed that the stripping efficiency of HDI PR increased with the pressure and temperature (T) when T is lower than 60?C. The removal efficiency of 96.8% was achieved at 60?C and 20MPa and then it gradually decreased as T higher than 60?C due to the lower scCO2 density at higher T. Magnetic agitation was favorable for the HDI PR removal. The optimum stripping parameters were determined to be 60?C and 25MPa for 20min at which 100% removal efficiency was achieved in EH-3/DMSO/scCO2 microemulsions. This environmentally benign method provides a promising alternative to effectively strip HDI PR for devices with nano-sized feature dimensions in microelectronic processing.

Quantitative study on removal of SU-8 photoresist patterns by supercritical CO2 emulsion

Supercritical CO2 emulsion (SCE) was found to be effective in removal of epoxy-type SU-8 photoresist. The ability to remove SU-8 was compared between a commercially available Remover PG (RPG) solution and SCE. Both RPG and SCE were effective in removing not-hard-baked SU-8 dots, with height of [email protected] and diameter ranged from 50 to [email protected] When the SU-8 dots were hard-baked at 150^oC for 10min, RPG could not remove any of the SU-8 dots. On the other hand, SCE could still remove 42.11% of the [email protected] SU-8 dots. When the SU-8 dots were hard baked at 200^oC for 10min, 5.75% of the [email protected] SU-8 dots could be removed by SCE.

Development of Alkaline Solvent Stripper for Wafer Level Package

In this paper, the advantage and limitation of different alkali photoresist stripper is reviewed. Different from traditional dry film stripper, tetramethyl ammonium hydroxide is adopted to replace potassium hydoxide as an etchant to prevent precipitation. In details, the substrate protection, such as Al, Cu, Ag and Sn, is studied, and angstrom level etch rate is received with the carefully-designed inhibitor system; Secondly, the effective photoresist removal performance is acheived with the components balance; In the end, the cleaning mechanism and capability of this novel photoresist removal stripper are disclosed.

Comparison of the effects of downstream H2- and O2-based plasmas on the removal of photoresist, silicon, and silicon nitride

For the 45 nm technology node and beyond, there is a need to strip photoresist quickly while suppressing the loss of materials such as polycrystalline silicon (poly-Si) and silicon nitride (Si3N4). To achieve this goal, the authors characterized and compared the effects of downstream pure-H2, H2/N2, and O2/N2 plasmas on the etch behaviors of photoresist, poly-Si, and Si3N4. The addition of N2 to H2 plasma increases the photoresist ash rate to a maximum that is reached at ∼30–40% N2, and the ash rate drops with further addition of N2. At 30% N2 addition, the ash rate increases by a factor of ∼3 when compared to that obtained with pure-H2 plasma. For O2/N2 plasma, the photoresist ash rate also exhibits a maximum, which is attained with 5% N2 addition, and the ash rate drops drastically as more N2 is added. A small addition of N2 increases the H and O radical densities in the H2- and O2-based plasmas, respectively, resulting in the higher ash rates. The ash rate achieved by the O2/N2 chemistry is generally higher than that attained with the H2/N2 chemistry, and the difference becomes more significant at high temperatures. The activation energy for photoresist strip under O2/N2 plasma was measured to be ∼10 kcal/mol, which is higher when compared to the ∼5 kcal/mol measured for both the H2/N2 (30% N2) and the pure-H2 chemistries. At 300 °C, when compared to the O2-based chemistry, the H2-based chemistry was shown to remove Si3N4 with a much lower rate, ∼0.7 Å/min, highlighting the benefit of the latter in conserving material loss. The ability of the H2-based chemistry to suppress material loss and its nonoxidizing property could justify the trade off for its lower ash rates when compared to those obtained using the O2-based chemistry. For the H2-based chemistry, a small N2 addition to the H2 plasma was found to not only increase the ash rate but also suppress the Si etch rate by a factor of 8 to 22, depending on the temperature. Collectively, the H2/N2 chemistry shows a great promise for photoresist-strip applications in the advanced nodes, and it should be run at high temperatures (e.g., T ≥ 300 °C) to maximize the ash rate while still maintaining extremely low Si and Si3N4 losses.

Large-scale assembly of semiconductor nanowires into desired patterns for sensor applications

The capability to rationally assemble semiconductor nanostructures into desired patterns is vital to their applications in functional devices. Here we report a facile method to assemble silicon nanowires (SiNWs) into large-area desired patterns by applying an external alternating electrical field between two photoresist-patterned indium tin oxide (ITO) electrodes. By manipulating the electrode geometries through photolithography, pattern shapes can be tuned, from parallel strips, concentric circles, grid-like structures to square windows. The patterns can be firmly preserved on the substrate after the removal of photoresist, and can be readily transferred onto a flexible and insulating substrate for further device applications. These NW patterns formed on the polydimethylsiloxane (PDMS) flexible substrate found excellent applications as both humidity and optical sensors with outstanding device stability and reproducibility. Moreover, it was found that the silver nanoparticle decorated-SiNW (AgNPs@SiNW) patterns could serve as an excellent surface-enhanced Raman scattering (SERS) substrate for sensing Rhodamine 6G (R6G) molecules.

1204 気中ウォータジェットを用いたフォトレジスト除去に関する実験的検討 (噴流衝突角度と試料水の効果について)(GS01-1 流体工学,オーガナイズドセッション 7:「最新工作機械」)

1204 気中ウォータジェットを用いたフォトレジスト除去に関する実験的検討 (噴流衝突角度と試料水の効果について)(GS01-1 流体工学,オーガナイズドセッション 7:「最新工作機械」)

The Stripping Behavior of High-Dose Ion-Implanted Photoresists in Supercritical CO2 Formulations

A new type of high-dose ion-implanted photoresist (HDI PR) stripping method was reported based on supercritical CO2 (scCO2) microemulsions consisting of branched hydrocarbon surfactant 2-ethyl hexanol polyoxyethylene-polyoxypropylene (EH-3) and co-solvent dimethylsulfoxide (DMSO). The optimized concentration of EH-3 and DMSO in scCO2 were 3 wt% and 2 vol.%, respectively, and the ratio of H2O and EH-3 is 1:1. The post-etch HDI PRs containing crust polymer were removed in the scCO2 microemulsions under the pressure of 10–25 MPa at 40–80°C for 10 min. The results revealed that the stripping efficiency increased with the pressure and temperature when the temperature is lower than 60°C. The 94% removal efficiency was achieved at 60°C and 20 MPa and then it gradually decreased as the temperature is higher than 60°C due to the lower scCO2 density at higher temperature. The magnetic agitation and pressure pulsation were favorable to HDI PR removal. The optimum stripping parameters were determined to be 60°C and 25 MPa for 20 min at which 100% photoresist removal efficiency was achieved in EH-3/DMSO/scCO2 microemulsions. This environmentally benign method provides a promising alternative to effectively strip the HDI PR for nano-sized devices in microelectronics processing.

Stripping of High-Dose Ion-Implanted Photoresist Using a Combination of Dry and Wet Single-Wafer Processing

High dosed ion implanted (HDI) photoresist is well-known to be difficult to remove by conventional wet stripping e.g. sulfuric–peroxide mixture without substrate loss because of their carbonized crust layer. To overcome this issue, several methods are proposed such as dry etch + wet, assist of physical cleaning. In this paper, a new method to strip HDI photoresist using a combination of dry and wet processing in the same chamber is introduced to achieve HDI photoresist removal without substrate loss. A new process consists of two steps. The first process removes the carbonized crust layer by atmospheric-pressure inductively coupled hydrogen plasma, and the second process removes remaining the bulk photoresist by a sulfuric–ozone mixture. The new process enables removal of the HDI photoresist without any substrate loss, and with significantly shorter times.

Effect of Microbubbles on Ozonized Water for Photoresist Removal

In this paper, we describe the use of ozone microbubbles in photoresist removal from silicon wafers. Ozonized water has attracted much attention as an environmental friendly cleaning method in semiconductor manufacturing. However, it would be desirable to enhance the oxidative ability of ozonized water for practical application. The existence of microbubbles in ozonized water has been shown to significantly enhance the photoresist removal rate due to an elevated dissolved ozone concentration (approximately 2.5 times that of ordinary ozone bubbling) and a direct effect of the microbubbles (removal rate is approximately 1.3 times faster than water with the same concentration of dissolved ozone without microbubbles). Additionally, the ozone microbubble solution was able to effectively remove a high-dose ion-implanted photoresist, which is extremely resistant to removal by ozonized water and other wet chemicals because of its amorphous carbon-like layer, or crust. Electron spin resonance experiments were also performed without the influence of serious metal contamination and indicated the presence of hydroxyl radicals, which are thought to be formed by interaction of ozone with hydroxide ions adsorbed at the gas-water interface upon collapse of the microbubbles. The hydroxyl radicals play an important role in photoresist removal by the ozone microbubble treatment.

Characterization of hydrogen–plasma interactions with photoresist, silicon, and silicon nitride surfaces

For the 45 nm technology node and beyond, a major challenge is to achieve reasonably high photoresist ash rates while minimizing the loss of the silicon (Si) substrate and its nitride (Si3N4). Accordingly, an objective of this work is to characterize the photoresist strip rate under varying conditions of H2 plasma and the effects of these conditions on Si and Si3N4 etch rates. In addition, we discuss in detail the fundamental mechanisms of the reactions between H atoms and the above substrates and successfully reconcile the process trends obtained with the reaction mechanisms. In this work, photoresist, Si, and Si3N4 films were exposed to downstream pure-H2 discharges and their removal rates were characterized by ellipsometry as a function of the following parameters: substrate temperature, reactor pressure, H2 flow rate, and source power. The authors found that the H2-based dry ash and Si3N4 etch are both thermally activated reactions, evidenced by the steady increase in etch rate as a function of temperature, with activation energies of ∼5.0 and ∼2.7 kcal/mol, respectively. The Si substrate exhibits a rather unique behavior where the etch rate increases initially to a maximum, which occurs at ∼40 °C, and then decreases upon a further increase in temperature. The decrease in the Si etch rate at higher temperatures is attributed to the activation of competing side reactions that consume the chemisorbed H atoms on the Si surface, which then suppresses the Si-etch step. The photoresist and Si3N4 removal rates increase initially with increasing pressure, reaching maxima at ∼800 and 2000 mTorr, respectively, beyond which the removal rates drop with increasing pressure. The initial increase in removal rate at the low-pressure regime is attributed to the increased atomic-hydrogen density, whereas the decrease in ash rate at the high-pressure regime could be attributed to the recombination of H atoms that could occur by various mechanisms. At temperatures where the reaction rates are relatively fast, the photoresist and Si removal rates both increase continuously with the H2 flow rate, indicating that both reactions are in the supply-limited regime. For the range of process conditions explored here, we find that the etch rates of Si are generally much higher than those for Si3N4 with Si:Si3N4 etch-rate ratios that vary from 25 to ≫520. Based on the process trends obtained here, we have identified a process window—high temperature and intermediate pressure—that attains relatively high photoresist ash rates and low Si and Si3N4 etch rates.