Permanent Photo-patternable Epoxies
Temporary Plating Resists
Temporary Lift-off Resists
Permanent Photo-patternable Low k Dielectrics
SU-8
What is the difference between the SU-8 and SU-8 2000 resist products?
SU-8 resists were first introduced commercially by MicroChem in 1996. These resists are formulated in gamma butyrolactone (GBL) solvent. Recently we introduced the SU-8 2000 series, which are formulated in cyclopentanone solvent and have improved coating and adhesion properties. In addition they are compatible with conventional edge bead removal processes directly after spin coating.
How thick can I coat SU-8 or SU-8 2000 in a single spin coat process?
The SU-8 and SU-8 2000 resin/solvent systems have unique molecular properties, which enables us to build very high solids formulations with relatively low viscosity. This allows us to offer a broad range of products with a wider range of film thicknesses than any other commercially available resist. Our products are designed to produce uniform resist films from <1um to >300um with conventional single spin coat processes.
I often observe bubbles in my resist film after coating or during the prebake process. How can I eliminate the bubbles?
If bubbles are observed in the bottle, we recommend heating the resist bottle in an upright position (50-60 C) for 30 minutes, which will reduce the viscosity and allow the bubbles to rise out of solution. Once the bubbles have been removed, allow the resist to reach room temperature prior to use.
Bubbles are commonly observed in the resist film when dispensed with a syringe. To prevent bubble formation during dispense, increase the size of the orifice and remove the air from the syringe prior to dispensing. We are unaware of bubbles associated with auto dispense processes when dispensing from bubble-free bottles.
How do I produce SU-8 structures with near vertical sidewalls in very thick films?
SU-8 and SU-8 2000 resists are optically transparent and highly functional, which permits high aspect ratio structures in very thick films to be imaged with optimized lithographic processes. Optimization techniques include; spectral shaping of the exposure bandwidth- remove shorter wavelengths that are absorbed in the upper portion of the resist film and result in negatively sloped sidewalls; fine tuning the exposure dose and post exposure bake (PEB) process in order to obtain uniform cross-link density throughout the resist film and by optimizing the prebake and develop process. For more information about suggested baseline processes, please refer to our SU-8 and SU-8 2000 technical data sheets or your SU-8 technical sales representative.
How do I produce SU-8 structures with near vertical sidewalls in very thick films?
SU-8 and SU-8 2000 resists are optically transparent and highly functional, which permits high aspect ratio structures in very thick films to be imaged with optimized lithographic processes. Optimization techniques include; spectral shaping of the exposure bandwidth- remove shorter wavelengths that are absorbed in the upper portion of the resist film and result in negatively sloped sidewalls; fine tuning the exposure dose and post exposure bake (PEB) process in order to obtain uniform cross-link density throughout the resist film and by optimizing the prebake and develop process. For more information about suggested baseline processes, please refer to our SU-8 and SU-8 2000 technical data sheets or your SU-8 technical sales representative.
What types of developer are compatible with SU-8 and SU-8 2000 resists?
SU-8 and SU-8 2000 resists are solvent developable. A number of safer solvent systems, including SU-8 developer, ethyl lactate and diacetone alcohol work well with SU-8. A two stage immersion develop process can be used to increase throughput, especially for very thick films or cyclopentanone (SU-8 2000 thinner), followed by a second bath of SU-8 developer.
Can thick film SU-8 structures be developed in spray mode?
Yes, they can. In fact we recommend spray develop processes and use them routinely in our labs for thin as well as thick film develop applications. Optimized spray develop processes are fast, clean, compatible with high aspect ratio structures and reduce the risk of resist adhesion failure commonly associated with lengthy immersion/sonic develop processes.
How do I strip SU-8?
SU-8 is a highly functional epoxy and therefore extremely difficult to strip. Conventional stripper solvents, including Kayaku Advanced Material, Inc.’s Remover PG will swell and lift off partially cross-linked SU-8, but will not remove hard baked (cured) SU-8. However, dozens of SU-8 users have successfully developed stripping processes. Techniques include RIE plasma ashing, laser ablation, molten salt baths, CO2 crystal and water jets and pyrolysis, among others. Keep an eye on our site, we will add SU-8 stripping applications notes as we develop and or are notified about them.
LOR & PMGI
Can I use PMGI as an imaging resist?
PMGI is not usually used as an imaging resist for optically exposed lift off applications, unless an isotropic slope is needed. It can be used as the imaging resist when undercut cannot be used, such as in the manufacture of T gates with e beam exposure.
How can I strip PMGI and LOR?
Typical resist strippers can be used for PMGI and LOR. Remover PG from MCC or Remover 1165 from Shipley can be used. For other non metallized substrates, oxygen plasma or Piranha can be used. The removal rate in plasma is roughly twice that of conventional novolak based resists.
What is the best developer for PMGI?
The best developer for PMGI is dictated by the application requirements. Some developers have a high dissolution rate. Other developers have a low dissolution rate. PMGI can be used with both metal ion bearing and metal ion free developers. For development that is more selective for PMGI than the imaging resist, the MCC’s 101 Developer is used. This allows for controlled development of each resist layer.
What resolution can I attain with PMGI?
Resolution is a function of the top imaging resist. The PMGI contributes the undercut needed for a discontinuous deposited film prior to lift off which can be adjusted to the desired geometry. PMGI is currently used in conjunction with DUV resists to achieve 0.25µm geometries. It is also used in resist stacks where 0.1µm gates have been imaged.
What is the minimum feature size I can obtain with an LOR bi-layer lift-off process?
LOR A resists are currently used with advanced I-line resists in 0.35um high volume production applications. Superb control of the undercut profile can be obtained with optimized prebake and develop processes. We also offer a line of PMGI resists, which have lower dissolution (undercut) rates than LOR A. These resists are currently used with deep UV resists in 0.25um high volume production applications.
Exposure
There are 2 different ways to optically expose PMGI.
1. Coat and bake the PMGI.
2. Flood expose the entire film.
3. Coat the imaging resist and process it normally.
4. Develop the PMGI at the same time as the imaging resist.
A flood exposure of the entire film as described above, will result in a larger undercut since all PMGI beneath the imaging resist has undergone scission and has a higher dissolution rate. OR
1. Coat and bake PMGI.
2. Coat, bake, image and develop the top resist.
3. Flood expose the PMGI through the developed window in the top resist
4. Develop the PMGI at the same time as the imaging resist.
A flood exposure of PMGI through a top imaged and developed resist will result in a lesser undercut than if the entire PMGI film was flood exposed. The exposed regions develop out at a high rate, decreasing as the developer laterally reaches the unexposed region of PMGI.
To reduce the undercut even further, eliminate any exposure of the PMGI. The intrinsically high dissolution rate of unexposed PMGI will permit some undercut even though it has not been exposed.
To obtain a very large undercut without exposure, the NANO LOR resist products can be used.
Softbake
The amount of undercut is also influenced by the softbake time and temperature. PMGI should be baked at a temperature of 185C or greater, up to about 250C to drive off the maximum amount of solvent and densify the film. It can also be baked at a temperature lower than 185C. As the softbake temperature and time are reduced, more solvent is retained in the film. This results in a larger amount of undercut due to the increased dissolution rate.
Developer
Some developers are more aggressive than others and result in different undercut rates during the development. For example, Shipley Microposit Developer or AZ Developer, containing sodium silicates and phosphates, have a very low development rate. TMAH (tetra methyl ammonium hydroxide) – metal ion free – based developers, such as Shipley Microposit Developer CD26 are more aggressive and will result in higher undercut rates. The same is true for the metal ion bearing sodium hydroxide and potassium hydroxide developers such as the Shipley Microposit 350 and 450 series.PMGI Resin
Just as the undercut rate can be controlled with the parameters discussed above, the type of PMGI resin used will also affect undercut. MCC offers PMGI formulations made from resins of differing dissolution rates. There are slow, medium and fast resins offered.
PMGI Resin
Just as the undercut rate can be controlled with the parameters discussed above, the type of PMGI resin used will also affect undercut. MCC offers PMGI formulations made from resins of differing dissolution rates. There are slow, medium and fast resins offered.
How thick must the PMGI and LOR layer be for my deposition process?
The PMGI and LOR layer, which is the bottom layer in a bi-layer stack, should be greater than the target deposition thickness. This will allow discontinuity of the deposited film and a subsequent clean lift-off. If the PMGI and LOR layer is not thicker than the subsequent film deposition, the film will be continuous, which will result in tearing of the film during lift-off. Ina typical process the PMGI and LOR film is 1.33X the target deposition thickness.
How do I decide if LOR A or LOR B resists are better for my application?
LOR A resists have lower dissolution/undercut rates in aqueous developers than LOR B resists. Therefore they are designed for more aggressive developers such as 2.38% TMAH (0.26N) such as Shipley’s CD 26 & JSR’s NMD-3. Because of LOR A’s lower undercut rates they are better suited for high resolution lift-off processes where fine control of the undercut profile is required. LOR B resists are optimized for less aggressive developers such as 0.24N TMAH developers such as Shipley’s MF-319 and Clariant’s AZ 400 K 1:4. The relatively high dissolution/undercut rates make them better suited for thick film and large feature deposition processes.
Can I reproducibly control the undercut?
Yes. There are 4 different variables that can be used to “tune” the process to give the desired undercut. They are: exposure (or lack of exposure), softbake temperature/time, developer type/normality and the PMGI resin composition. It is the last parameter, the resin, that makes PMGI unique and permits a broad range of highly reproducible undercut rates.
What is a typical exposure dose?
The dose required to flood expose PMGI is dependent upon thickness and exposure source output as well as other process variables. Typical optical doses are in the 1 to 5 Joule/cm2 range. Typical e beam doses are reported to be in the 30 to 50 µC/cm2 range, or about 3 to 5 times faster than PMMA.
How do I expose PMGI?
PMGI can be exposed with different optical processes using DUV (240 to 290nm) flood exposure, as described in “Can I reproducibly control the undercut?” PMGI is also sensitive to electron beam irradiation and is currently used with e beam exposure tools.
Will an interfacial layer form using PMGI and LOR?
If PMGI and LOR are used as the bottom layer of a resist stack, and the solvent has been removed by the softbake, no or neglible intermixing with the top resist occurs, primarily because PMGI is insoluble in conventional photoresist solvents. Thus no interfacial layer is formed. If PMGI and LOR are used as a middle or top layer, it is possible that intermixing can occur due to penetration by the PMGI and LOR solvent which is capable of dissolving most resist chemistries.
How can I defeat edge bead or cobwebbing?
Both of these problems can be overcome by using MCC’s edge bead remover product EBR PG. It should be noted that acetone is not compatible with PMGI and LOR and will not remove edge bead. EBR PG is a universal formulation and will also remove other conventional resists and anti reflective coating materials. This permits a single edge bead remover to be used for many materials.
How thick can PMGI be coated?
PMGI and LOR resists are available in a range of concentrations designed for spin coating from very thin, (<50nm) to relatively thick (>8µm) in a single step. For thick films, the bake should be adjusted to accommodate the solvent evaporation. A ramped softbake is recommended for coatings above about 5 or 6 microns to prevent micro bubbles.
Will I need an adhesion promoter for PMGI and LOR?
PMGI and LOR typically requires no adhesion promoter. As with any resist, the softbake temperature should exceed the Tg (glass transition temperature) of the resist resin in order to maximize adhesion. The Tg of PMGI is a function of molecular weight and is typically about 180-190C. Therefore, to obtain the best adhesion, the softbake temperature should be greater than 185C.
How are PMGI and LOR products used?
There are different ways to use PMGI. It may be used as the bottom layer of a 2 layer system or also between other layers in a trilayer system. After the top imaging resist is exposed and developed, additional development removes the PMGI directly beneath the new imaging resist window and also develops PMGI laterally. This lateral undercutting of the top imaging resist leaves an overhang feature in the imaging resist.
Will I need an adhesion promoter for PMGI and LOR?
PMGI and LOR typically requires no adhesion promoter. As with any resist, the softbake temperature should exceed the Tg (glass transition temperature) of the resist resin in order to maximize adhesion. The Tg of PMGI is a function of molecular weight and is typically about 180-190C. Therefore, to obtain the best adhesion, the softbake temperature should be greater than 185C.
What is PMGI?
PMGI is an alkaline soluble polymer derived from PMMA. It is formulated for spin coating in a solvent primarily composed of cyclopentanone, for use as a photoresist in multilayer lithography when an undercut layer is needed, as in liftoff applications. It is used to create the retrograde profile, which provides a clean liftoff with no tearing, ripping or shredding of the deposited film. PMGI has an actinic spectral response at wavelengths from 240nm to 290nm.
PMMA
What is the difference between your PMMA A resists and PMMA C resists?
Our PMMA A resists are formulated in anisole, while our PMMA C resists are formulated in chlorobenzene. We introduced the PMMA A resists in response to the environmental concerns expressed by many of our customers with PMMA resists formulated in chlorobenzene solvent.
Will I need to redesign my process if I switch from PMMA resist formulated in chlorobenzene to an equivalent formulated in anisole?
The viscosity of PMMA resists formulated in anisole is slightly lower than those formulated in chlorobenzene. Therefore you may need to use a slightly higher solids PMMA A resist to obtain the same film thickness. Please refer to the spin speed curves located in our PMMA & copolymer technical data sheet to select the appropriate product.
Are there performance differences between PMMA resists formulated in chlorobenzene and those formulated in anisole?
It depends upon your process requirements. In most cases PMMA formulated in anisole will have virtually identical performance characteristics as with PMMA formulated in chlorobenzene. However, if the PMMA resist is placed on top of another resist, such as a DNQ based positive resist, the anisole resist may attack (intermix with) the underlayer more readily than with a chlorobenzene based PMMA resist. For more comparative data between PMMA in anisole and PMMA in chlorobenzene, please refer to the following paper, which can be obtained from our PMMA home page. Characterization of safe solvent PMMA resist variables for electron beam applications. Bruce W. Smith, Todd D. Eakin, Rochester Institute of Technology and Donald W. Johnson, MicroChem Corp., 200 Flanders Road, Westborough, MA 01581.
What is the difference between 50PMMA, 100PMMA, 200PMMA, 495PMMA, 950PMMA and 2.2PMMA?
These are polymethylmethacrylate (PMMA) resists with molecular weights (MW) varying between 50,000 and 2.2 million. 495PMMA and 950PMMA are our standard products. We offer the other MW resists upon request. In general, the higher the molecular weight, the slower it will dissolve in a solvent developer. After exposure (molecular chain scission), the develop contrast between the exposed and unexposed regions of the film becomes higher as MW increases. Thus 950PMMA is typically used in high resolution applications and 495 PMMA in less demanding applications. Proper selection of MW is also critical in bi-layer T-gate processes where contoured sidewall profiles are desired.
How do I remove PMMA and copolymer resists?
PMMA and copolymer resists can be removed by using Kayaku Advanced Materials’ Remover PG or standard cleanroom solvents, such as anisole or positive photoresist removers. Resists that have seen higher processing temperatures and/or hostile processes that have toughened the polymer will require a more aggressive removal process. This can include Remover PG at elevated temperatures.
What is the Index of Refraction (I.R.) of PMMA at the HeNe wavelength?
The I.R. of PMMA is 1.49-1.52 at 632.8nm. Please refer to our PMMA & copolymer technical data sheet for other optical parameters such as cauchy coeffiecients, dispersion curves and n & k values.
What is the glass transition temperature (Tg) of PMMA?
The Tg of 495PMMA and 950PMMA is approximately 95 – 106°C.
Can you recommend any good PMMA reference papers?
Yes we can. They are listed below.
Fabrication limits of nanometer T and G gates: Theory and experiment
B. Maile – Daimler Benz J. Vac. Sci. Tech. B 11(6), Nov/Dec 1993.
Developer Characteristics of Poly-(Methyl Methacrylate) Electron Resist
J. Greeneich – GMRL J. Electrochem. Soc., Vol 122, 970-6, 1975.
Influence of developer and developer conditions on the behavior of high molecular weight e-beam resists
D. Hasko, S. Yasin, A. Mumtaz – U. of Cambridge J. Vac. Sci. Tech. B 18(6), Nov/Dec 2000.
Electron-beam resist technology for GaAs microwave device fabrication
D. Brambley, R. Bennett – GEC GEC Journal of Research, 13,1, 42-53,1996.
Effect of molecular weight on poly (methyl methacrylate) resolution
M. Khoury and D. Ferry – Arizona State J. Vac. Sci. Tech. B 14(1), Jan/Feb 1996.
Effects of molecular properties on nanolithography in PMMA
E. Dobisz, S. Brandow, R. Bass,
J. Mitterender- NRL J. Vac. Sci. Tech. B 18(1), Jan/Feb 2000.
UniLOR® N **NEW**
What type of resist is UniLOR® N?
UniLOR® N is a chemically amplified, negative-tone resist intended for high temperature lift-off applications. It can be used as a single layer lift-off resist or as the imaging resist on top of PMGI or LOR in a bi-layer lift-off scheme.
What are common applications for UniLOR® N?
Single layer lift-off resist using metal evaporation or physical vapor deposition processes requiring < 3 microns of metallization.
What is the casting solvent of the resist?
The solvent contained in UniLOR® N is PGMEA / Cyclopentanone.
Does UniLOR® N require a primer?
HMDS priming is recommended on bare silicon.
Is an EBR removal step necessary/recommended and what EBR solvent can be used?
Customers should use EBR PG for best EBR results.
Is it necessary to do a post exposure bake? Hard bake?
The PEB step is necessary to complete the thermally driven catalytic reaction that causes the exposed resist to become insoluble to the aqueous developer. A hard bake is not required, as the resist is meant to be temporary and stripped post-processing.
How controllable is the sidewall angle?
The sidewall profile may be varied by exposure dose and post exposure bake temperature.
What is the resist lift-off process?
Kayaku Advanced Materials’s Remover PG (NMP-based) or DuPont’s INTERVIA™ 2011 Remover (DMSO/GBL based) are recommended for lift-off. Heating and ultrasonication can improve lift-off quality and rates. Lift-off times range from 10 – 30 minutes for complete lift-off depending on heat and sonication conditions.