Nanoimprint lithography technology and its development status

1 Introduction

Due to economic reasons, the semiconductor industry is moving towards shrinking feature sizes, and subsequent technological advances have led to an exponential increase in equipment costs. Due to the increase in costs, people are paying more and more attention to nanoimprint lithography, a low-cost pattern transfer technology. By avoiding the use of expensive light sources and projection optical systems, nanoimprint lithography greatly reduces costs compared to traditional lithography methods.

Research on nanoimprint lithography began with Professor Stephen Y. Chou of the Nanostructure Laboratory at Princeton University, where a template with nano-patterns was mechanically (high temperature, high pressure) pressed on a silicon substrate coated with a polymer material in equal proportions Printing and copying nanopatterns, the processing resolution is only related to the size of the stencil pattern, and is not subject to the physical limitation of the shortest exposure wavelength of optical lithography. At present, NIL technology can already produce patterns with line widths below 5nm [1]. The cost of optical lithography reticle and the use of optical imaging equipment is eliminated. Therefore, NIL technology has the economic advantages of low cost and high output. In addition, NIL technology can be applied in a wide range, covering nanoelectronic components, biological or chemical wafer laboratories, micro-channel devices (micromixers, microreactors), ultra-high storage density magnetic disks, micro-optical components and other fields .

2. The basic principles and processes of nanoimprint technology

In the past decade, research on various innovative NIL processes has been carried out successively, and its experimental results have become more and more satisfactory. At present, four representative technologies can be summarized: hot imprint lithography technology, ultraviolet hardening imprint lithography technology, Soft imprinting, laser assisted direct lithography.

2.1 Hot stamping (HE-NIL)

The hot stamping process includes the following steps:

â‘  First, use electron beam direct writing technology (EBDW) to make a nano-patterned Si or SiO2 stencil, and prepare a silicon substrate that is uniformly coated with thermoplastic polymer photoresist (usually PMMA as the main material).

â‘¡ Heat the photoresist on the silicon substrate to the glass transition temperature (Glass Transfer
Temperature), using mechanical force to press the template into the photoresist layer softened at high temperature, and maintain high temperature and high pressure for a period of time, so that the thermoplastic polymer photoresist is filled into the nanostructure of the template.

â‘¢After the photoresist is cooled and solidified, release the pressure and release the stencil from the silicon substrate.

â‘£ Finally, the reactive ion etching (Reactive Ion Etching) is performed on the silicon substrate to remove the remaining photoresist, that is, the nano pattern with the same proportion as the template can be copied.

2.2 Ultraviolet hardening imprint lithography technology (UV-NIL)

Thermoplastic polymer photoresists using hot imprint lithography must undergo a phase change process of high temperature, high pressure, and cooling. The pattern imprinted after demolding often deforms, so it is not easy to perform multiple times using hot imprint technology. In order to solve this problem, some people began to develop some imprint lithography techniques that can be used at room temperature and low pressure.

M. Bender and M. Otto proposed an imprint lithography technology that uses ultraviolet light to harden polymers at room temperature and low pressure [4, 5]. The pretreatment is similar to hot imprinting. The pattern template, and the UV-NIL template material must use quartz that can penetrate ultraviolet light, and a layer of low viscosity, UV sensitive liquid polymer photoresist is coated on the silicon substrate, and the alignment between the template and the substrate is completed After that, the stencil is pressed into the photoresist layer and irradiated with ultraviolet light to cause the photoresist to undergo polymerization and hardening, and then the mold is released and the remaining photoresist on the substrate is etched to complete the entire UV-NIL.

A new development in UV imprinting recently is the step-flash imprinting. Step-flash imprint was invented at the University of Texas in Austin, and it can achieve a resolution of 10 nm. First, the monomer solution with low viscosity is dropped on the imprinted substrate, and the stencil is pressed onto the wafer with very low pressure to disperse the liquid and fill the cavity in the stencil. The ultraviolet light irradiates the monomer through the back of the template, and after curing and forming, the template is removed. Finally, the residual layer is etched and pattern transfer is performed to obtain a structure with a high aspect ratio.

Obviously, compared with thermal imprinting, UV imprinting does not require high temperature and high pressure conditions. It can obtain high-resolution graphics at the nanoscale at a low cost. Its process can be used to develop nanodevices. Flash imprinting not only leads to a significant reduction in process and tool costs, but is also as good or better than optical lithography in other aspects. These other aspects include tool life, mold life (without reticle), mold cost, process yield, Yield and size reproduction accuracy. But its disadvantage is that it needs to be operated in a clean room environment.

2.3 Microcontact Imprint Lithography (Microcontact-NIL)

The micro-contact imprint lithography technology must first obtain the template through optical or electron beam lithography. The chemical precursor of the mold material is cured in the stencil, and is detached from the stencil after polymerization and molding, and the mold required for micro-contact printing is obtained. The mold often obtained is PDMS. Next, the PDMS mold is immersed in a reagent containing thiol, and then the mold soaked in the reagent is pressed onto a gold-plated substrate. The substrate can be in various forms such as glass, silicon, and polymer. In addition, a thin layer of titanium can be plated on the substrate and then plated with gold to increase adhesion. Mercaptan reacts with gold to form a self-assembled monolayer SAM. There are two processes to deal with it after printing. One is to use wet etching. For example, in the hydride solution, the hydride ions promote the dissolution of the gold that is not covered by the SAM layer. Since the SAM can effectively block the hydride ions, the gold covered by the SAM is retained. This transfers the monolayer pattern to gold. You can further use gold as a mask to etch the areas that are not covered by gold to achieve pattern transfer again. The other is to link certain organic molecules on the gold film by self-assembled single-layer thiol molecules to achieve Self-assembly, such as the surface of the biosensor can be processed by this method.

Micro-contact printing not only has the advantages of being fast and cheap, but it also does not require the harsh conditions of the clean room, or even an absolutely flat surface. Micro-contact printing is also suitable for a variety of different surfaces, and has the characteristics of flexible and variable action methods. The disadvantage is that in the sub-micron scale, the diffusion of thiol molecules during printing will affect the contrast and make the printed graphics wider. By optimizing the immersion method and immersion time, especially controlling the amount and distribution of reagents on the mold, the diffusion effect can be reduced.

3. Current status of nanoimprint lithography technology

At present, a large number of research institutions have carried out research work on this technology. HP researchers Stan Williams used nanoimprint lithography technology to make experimental memory chips with a line width of 45nm. Stan Williams is the research direction of quantum science in the HP laboratory. Senior Research Director, he said: "We are using nanoimprint lithography to steadily produce usable integrated circuits with a line width of 30nm."

Research institutes and universities like Hewlett-Packard are using imprint lithography to manufacture a large number of molecular-scale devices. As the experimental results reported in their papers say, nanoimprinting, the most promising manufacturing method, has gradually become Mainstream industrial technology, which can not only produce extremely tiny graphics but also greatly simplify many production processes, its cost is extremely low, maybe one-tenth of optical lithography, but first it is necessary to develop production preparation tools and manufacturing infrastructure, and Potential customers should also believe that imprint lithography technology has technical and economic advantages over other technologies. Although HP ’s prototype chips are still far from the industrialization of their methods, Williams is convinced that nano-imprint lithography technology may bring semiconductor industry Revolution, "We think this is a very meaningful technology, this is the most promising technology that can be manufactured in the field of nanotechnology," he said. "No other technology can make our circuits now."

HP is not the only enthusiast of nanoimprinting. The other five major companies—EV Group, Molecular Imprints, and Nanonex are shown in Table 1. Obducat and Suss MicroTec are both working on this technology. They sell imprint lithography equipment, or many other companies that are committed to other aspects of this technology, including stencils, polymer materials, and inspection tool manufacturing companies. Although imprint equipment Soon after the rise of industry, there is no data to show how big this market is. Both Molecular Imprints and Nanonex claim to have sold about a dozen devices at prices ranging from 100,000 to a million dollars.

Companies are also creating a good environment in the early stages. EV recently established an industry association to support the commercialization of embossing technology for more than a dozen companies, universities and research institutes by granting NILcom. Helge Luesebrink, director of EV's Advanced Lithography Business Department, estimates that there will be approximately 160 embossing equipment worldwide by the end of 2004, mainly in universities and research institutes, and their companies account for about a quarter of the market share of these equipment.

Molecular Imprints also claims that its R & D partners include the famous Motorola, KLA-Tencor, Lam Research, Photronics and Carl Zeiss. In 2004, Molecular Imprints added $ 12 million in venture capital, and NIST (National Institute of standards and Technology) funded $ 36.8 million to prove the possibility of using nanoimprinting for 65nm and below processes.

Although there is at least a few years away from the commercial production of imprint technology, industrial personnel hope that this technology can be quickly used in other markets. Norm Schumaker, CEO of Molecular Imprints, believes that this technology is still in many markets besides silicon wafers. For applications, he was a researcher at Bell Labs and helped set up two semiconductor equipment companies.

In the digital camera market, imprint technology can be used to make high-quality image sensor microlenses. Imprint equipment can make filters and photonic band gap structures to make TVs and light-emitting diodes brighter. Using imprinting technology is currently the most economical way to make these devices. Disk drive manufacturers are watching imprinting technology to make patterned media. Small magnetic dots can replace the consistent layer of today's magnetic recording materials. Disk manufacturer Komag recently obtained pressure from EV Printing technology license, and hope to expand the storage capacity is higher than 160Gb per disk, and EV's Luesebrink believes that by 2007 the storage industry will use imprinting to start mass production.

The biological industry is another potential market. Schumaker said the university is using imprint lithography to create a patterned surface on which cells cannot grow or grow in a specific direction. Waseda University and a Japanese medical device company are working on a The cell array device uses an imprinting device to quickly analyze the liquid and identify specific target cells.

The best proof of the survivability of imprint lithography technology is Nanonex, a four-year-old private company in New Jersey, who has commercialized this technology. Nanonex was founded by Professor Stephen Chou of Princeton University. He recently started to use his own embossing equipment to produce various optical devices including DVDs and CDs to improve the optical signal of the wave disc. These products are being shipped to customers.

Despite these advances, the semiconductor industry that uses expensive optical lithography seems reluctant to accept this revolutionary change in technology, and invests in increasingly difficult downward optical lithography, market leaders in lithography equipment ASML, Nikon and Canon has not expressed interest in imprint technology, but some people think that this is not necessarily the case, as EV ’s Luesebrink said: "I am sure that optical lithography companies are working on this technology."

Imprint lithography is still controversial in the semiconductor manufacturing industry, and it is understandable that they are cautious about immature technologies. "Nothing is more conservative than the manager of a semiconductor manufacturing plant," said Schumaker of Molecular Imprints. "He didn't dare to block it with a three-million-dollar facility." Bernie Roman, manager of advanced lithography technology research at Freescale Semiconductor, thinks Imprint lithography technology has a great challenge to make high-density chips. He added: "We don't think this is the best solution." VLSI research analyst Risto Puhakka was asked how long imprint lithography can be used. When producing in the semiconductor industry, he replied: "It's still far away from that day, maybe it won't be realized at all."

Why can't it be achieved? Puhakka said that the most fundamental reason is that the imprinting tool and the material to be manufactured need to be in contact. "There is something to contact the silicon wafer, which is taboo in the semiconductor industry." Contact imprinting was commonly used 30 years ago, but it was gradually better The optical lithography method was eliminated.

Other reasons include the difficulty of obtaining high-quality templates and the need for other tools to achieve alignment between the various layers of the circuit. Of course, the template is the most challenging one. "No one has really started commercial production of templates," HP's Williams said, and their templates were manufactured by Lawrence Berkeley Laboratories. This is a "chicken raw egg, egg raw chicken" problem. He said that the scarcity of templates has led to the slow development of technology, and template manufacturers are reluctant to expand investment before the market becomes larger.

Franklin Kalk, the chief technical expert of Toppan Photomasks, said his company manufactures limited imprint lithography templates, but the process of inspecting their defects is very difficult. "We do n’t want to spend millions of dollars to purchase e-beam inspection tools. "It seems that imprint lithography technology has a long way to go before real industrialization.

Abs Pc Zip Luggage

Abs Travel Luggage,Trolley Luggage,Hardshell Spinner Luggage,Abs Pc Zip Luggage

Yongxin Juanhua Leather & Bags Co., Ltd , https://www.luggagegld.com