In the traditional fabrication of microelectronic devices and flat panel displays, the patterning process can be both time consuming and expensive. Patterning by photolithography involves numerous steps, including photoresist (PR) coating, PR baking, exposure, developing, material etching and PR stripping. This complex process requires a large investment in equipment, high fabrication costs, high maintenance charges, large volumes of chemicals, and long fabrication times. With the increasing size of flat panel displays, these problems are becoming much more severe.
Laser ablation is a simpler patterning method that involves stimulating photodissociation in materials by irradiating them with a laser beam. This method does not require the developing and etching steps of photolithography, and can be used with most polymeric materials that dissociate readily. However, for most inorganic materials, laser ablation is only able to produce poor quality patterns. This is because inorganic material is mostly removed thermally, rather than by photodissociation, and this requires a more powerful laser beam. In addition to producing poor quality patterns, a powerful laser beam can damage the underlying structure, shorten the lifetime of the equipment, lead to increased maintenance costs and raise safety concerns, all of which have deterred industry from using laser ablation.
To overcome this problem, we have developed a new process that merges two conventional microelectronic processes: polymer ablation and lift-off. Figure 1 illustrates the concept of the process, which we applied to the patterning of indium tin oxide (ITO) in the fabrication of thin-film-transistor liquid-crystal displays (TFT-LCDs).
Figure 1. A novel patterning process (a) first coated the substrate, which already contained thin-film-transistor (TFT) structures, with photoresist (PR). (b)The surface was ablated by excimer laser illumination through a mask. (c) Indium tin oxide (ITO) was deposited over the top, and then (d) patterned by removing the PR with acetone, in a process known as lift-off.
Figure 2 shows the patterned PR after laser ablation, demonstrating that the PR was ablated cleanly. There is some residue—called debris—on top of the remaining PR, but this will be removed with the PR during the lift-off process. Figure 3 shows the patterned ITO after the lift-off process. The linearity of the ITO pattern is sufficiently good that this patterning technique could be used in various other applications.
Figure 2. Optical microscope photo of patterns formed in a photoresist layer by laser ablation.
Figure 3. Scanning electron microscopy photo of a 60nm-thick indium tin oxide film.
Like laser ablation, this patterning process does not require developing and etching steps, which helps to increase the throughput and also decrease equipment maintenance. This process will also result in significant cost savings for industry, as production plants will need less equipment and fewer chemicals, such as developer and etchant.
This process also has other advantages over the direct (thermal) ablation of ITO. In our process, the strength of the laser beam, known as fluence, required to ablate the PR is low, typically 50–200mJ/cm2. For direct ablation of ITO, a fluence of more than 1000mJ/cm2 is required if the same kind of laser is used.1 Using a fluence this high for TFT-LCD fabrication can damage the amorphous silicon (a-Si) transistor fabricated under the ITO layer. Previous studies have reported that a-Si can be decomposed by high-fluence laser illumination.2 In our process, we ablated PR at a fluence of just 50mJ/cm2, which is low enough to prevent any damage to the a-Si transistor.
In summary, we have developed a non-lithographic patterning method based on excimer laser ablation of polymers, which can be used for fabricating microelectronic devices and flat panel displays. This new process reduces the cost of fabricating these products by eliminating two important processing steps. We demonstrated the feasibility of this process by experimentally illustrating how it can replace the conventional ITO patterning process. We are currently researching the use of this technique for patterning other kinds of materials, and a prototype TFT-LCD will be fabricated and reported in the near future.
The authors are grateful to Sreeram Appasamy and Dr Krishna Kuchibhotla of Anvik Corporation, Hawthorne, NY, for their help with the experiments.
Kanti Jain, Junghun Chae
Department of Electrical and Computer Engineering
University of Illinois at Urbana-Champaign
Kanti Jain is professor of electrical and computer engineering at the University of Illinois at Urbana-Champaign (UIUC). He received a BTech (Hons) degree from the Indian Institute of Technology, Kharagpur, in 1969, and MS and PhD degrees from UIUC in 1970 and 1975 respectively. He was a post-doctoral fellow at the Massachusetts Institute of Technology between 1975 and 1977. From 1979 to 1988, he worked for IBM, where he invented and developed the technology of excimer laser lithography, which is now used worldwide for semiconductor integrated circuit manufacturing. For this development, he received two Outstanding Innovation Awards from IBM. From 1989 to 1991, he was director of technology in advanced microelectronic packaging systems for Raychem. He is also the founder of Anvik Corporation and has been its CEO since 1992. Professor Jain is a fellow of the Institute of Electrical and Electronics Engineers, the Optical Society of America, and the SPIE, and a former member of the Board of Directors and the Executive Committee of the SPIE. He is the author of the book Excimer Laser Lithography (SPIE 1990), has published 55 papers, and holds 61 patents.
Junghun Chae is a postdoctoral research associate at UIUC. He received his BS degree from the Korea Advanced Institute of Science and Technology (KAIST) in 1995 and his PhD degree in materials science and engineering from KAIST in 2002. He was a senior research engineer at LG Philips LCD from 2002 to 2006, developing a-Si transistors, a-Si transistor processes, low temperature transistors and poly-Si transistors for TFT-LCD. He joined UIUC in 2006 and his current research interests include excimer laser processing for device fabrication, nano-electronics, and flexible electronics.