Technology conversion between DRAM and 3D NAND flash

In 2014, 3D NAND Flash was introduced to the industry in 24 layers and coexisted with 2D NAND Flash until 2017. 3D NAND flash has gradually become the market leader in high-density storage capacity and technology, as well as the manufacturing process. Then it completely branched off from DRAM. The synergies of using the same process and equipment for the two memory types (DRAM and NAND) have completely disappeared, changing the core of the memory industry’s competitive strategy.

The DRAM manufacturing process is slowly progressing from 1xnm to 1y, 1z, 1-alpha, and 1-beta, requiring expensive EUV equipment from 1z or 1-alpha. With the move of 3D NAND flash to 3D, the accuracy of the horizontal manufacturing process is no longer important and the rapid expansion of 3D storage capacity has been prioritized. The focus of the manufacturing process shifts from photolithography to etching and smoothness.

Improving the slow DRAM manufacturing process provides memory manufacturers with the opportunity to rethink their operational strategies. Even in the absence of competition, companies need to consider ways to generate profits to finance the research and development of new DRAM manufacturing processes. The problem is not so urgent, as most companies in the DRAM industry have been wiped out in the pre-competition phase and the market is currently dominated by a small number of major players. Undoubtedly, DRAM will continue to be a major component of most computing devices, but slow manufacturing process improvements mean that DRAM may one day lose its place in the high-tech industry lineup. Means. The strategy that a company should adopt is important. If the manufacturing process takes 10 years to advance to the next generation, companies can easily catch up with new players entering the industry. Even if market demand persists, it is unlikely that a non-tech competitive model will generate 50% gross profit for a company.

The future of 3D NAND flash is clear. Observers expect to expand 3D NAND up to 600 layers, so the linear advance strategy for the relevant manufacturing process will continue for several years.

The synergies of using the same process and equipment for two memory types may have disappeared, but memory companies need to establish new competitive strategies, even though current market competition is not intensifying. there is.

Market synergies for both memory types begin to emerge first. 3D NAND Flash Solid State Drives (SSDs) are already cheaper than mechanical hard drives, and both DRAM and 3D NAND Flash have become standard accessories for all computing devices, including mobile phones and computers. .. All memory manufacturers must be able to deliver both memory products to the same client at the same time. Companies that do not produce DRAM, such as Toshiba, need to seek partnerships to provide one-stop services to their clients. The deadlock in DRAM development is near. If the industry wants to maintain its high-tech business model, one option is to introduce new memory products that have potential for future growth and can create new economic value through improved manufacturing processes. Therefore, the development of new memory is essential. However, most memory companies are already spending a lot of money developing new manufacturing processes for DRAM and 3D NAND flash, so considering new memory development can be a heavy burden for them. As a result, memory manufacturers need to partner with partners to develop new memory. Examples include Micron and Intel co-developing 3D X-point memory, and SK Hynix co-developing MRAM with Toshiba. These new memory products currently have only niche applications and cannot dominate the market.

(Editor’s Note: This is part of a series on the memory industry written by DIGITIMES advisor Albert Lin. Lin received his PhD in physics in 1988 and taught at National Central University, after which. Moved to the technology industry. Lin was a former director. Vice President of ProMOS Technologies, President of ConDel International Technologies. He is an advisory member of the Taiwan Semicon, chair of the Huangguang Forum, and the Taiwan Semiconductor Industry. He chaired the editorial board of the Association. Currently, Lin is the Faculty of Physics, National Central University. His main research fields are new materials, new mechanisms, and basic research on quantum information. He is a Taiwanese quantum computing information technology. I am a permanent director of the association.)