New NVM Architecture Paves the Way for Expanding 3D Xpoint Market

Article By : Gary Hilson 

A new 3D vertical NVM memory architecture will allow customers to design chips for high density, high performance computing applications at affordable costs...

A new combination of materials may enable 3D vertical non-volatile memory (NVM) architecture for customers to design chips for high density, high performance computing applications at affordable costs and even shape the second iteration of 3D Xpoint technology.

Intermolecular Inc., a wholly-owned subsidiary of Merck KGaA, Darmstadt, Germany, recently announced it has developed what it believes is the first quaternary atomic layer deposition (ALD) GeAsSeTe OTS device for 3D vertical memory arrays. This device overcomes the inability to stack tens of layers in a 3D structure, which limits memory density and results in higher costs, said Intermolecular device engineer Mario Laudato. The company’s new material combination can help to realize these architectures, which would support emerging use cases such as artificial intelligence (AI) and neuromorphic computing, and other semiconductor designs necessary for faster and more affordable digital applications.

Founded in Silicon Valley in 2004, Intermolecular develops workflows that physically and electrically characterize thin films and film stack for material innovation in the memory, logic, ferroelectrics and quantum computing fields. It became part of Merck KGaA last year. In a telephone interview with EE Times, Laudato said its device for 3D vertical memory arrays could potentially help evolve 3D Xpoint technology. “The semiconductor industry is spending huge amounts of resources in R&D to fill the gap between DRAM and NAND flash,” he said, “and probably we will add more companies working in this market and releasing product with 3D Xpoint technology.”

Intermolecular’s new approach involves the use of ALD chalcogenides in lieu of the current PVD process for 3D Xpoint, which limits film conformality and homogeneity on a large scale, he said, and as a result, precludes the integration of tens of decks in a 3D Xpoint architecture. By employing ALD chalcogenides, Intermolecular’s process enables future 3D vertical integration with higher density.

According to Intermolecular, Inc., its ALD process allows for future 3D vertical integration with higher density and reduced cost, paving the way for the expansion of the 3D Xpoint market. (Source: Intermolecular)

The company isn’t the first to delve into the development of ALD OTS films, said Laudato, as several research institutes and universities have tried to develop them in recent years but weren’t able to demonstrate electrical performances in terms of leakage and endurance on par with PVD OTS. He said Intermolecular has been able to develop a quaternary ALD OTS GeAsSeTe film with electrical performances similar or better than PVD OTS. “The ALD process could speed up the development for this technology and possibly maybe in the next year we will have a lot of development and release new product with higher density and we will see a bigger market.”

As outlined in a recent research paper, Intermolecular was able for the first time to integrate an ALD quaternary GeAsSeTe OTS device that demonstrated high selectivity, excellent endurance, low V drift and fast switching, while outlining a path for the future development of ALD chalcogenide films, according to Laudato. This paves the way for the use of ALD chalcogenide-based selectors as leading technology for multiple stacks integration of 3D Xpoint memory arrays, he said.

Thomas Coughlin, president of Coughlin Associates, said that as a boutique development laboratory that’s part of Merck, Intermolecular can do a whole bunch of experiments at once in an environment they can control and avoid contamination. “They deal with weird materials and processes,” he said. “They’re willing to take on experiments with stuff that a big facility wouldn’t want to tackle, or they would rather farm it out and then work with it themselves.”

Coughlin said if Intermolecular can develop new intellectual property that Micron and Intel don’t have, it’s possible that other companies could see the 3D Xpoint market as being viable for them. With Micron having released very little product based on the technology, Intel is really the only player right now.

The prospect of making 3D Xpoint technology less expensive comes at a time where both the SSD and DIMM form factor of the PCRAM-based NVM is only just starting to see some volume sales, said Coughlin. “The DIMMs are probably the most interesting aspect because they cut down on expensive DRAM. And since Intel’s willing to sell below cost to keep it below the price of DRAM, I’m seeing evidence that there is a pickup.”

Leave a comment