The Virginia Innovation Partnership Corporation (VIPC) has awarded $100,000 to Virginia Commonwealth University (VCU) to expand the MRAM research conducted by Dr. Jayasimha Atulasimha

Atulasimha’s group is developing skyrmion-mediated MRAM (SkMRAM), a nanomagnet-based RAM technique that builds upon STT-MRAM technology while significantly improving energy efficiency. By adding a layer of heavy metal to STT-MRAM, energy consumption is reduced by 100-1000x, or 2-3 orders of magnitude. The resulting product is non-volatile, meaning that it can retain data even when the device is powered off, and has a very low write-errors rate, which means that it saves energy while writing information. It also doesn’t require standby power to retain information and is reliable. One patent currently covers this technology.

Vertןcal Compute, a new startup that was spun-off from the imec institute to commercialize its MRAM technology for AI applications, has raised 20 million Euro. The seed round was led by imec.xpand and supported by Eurazeo, XAnge, Vector Gestion, and imec.

The company says that it will develop an MRAM in-memory computing chiplet technology. The chiplet, according to the company, can reduce power consumption by 80% and speed up the execution of large language models for AI by 100X. It is believed that the company's MRAM is based on SOT-MRAM technology.

Researchers from Osaka University developed a new MRAM device architecture that enables an electric-field-based writing scheme with reduced energy consumption compared to the present current-based approach.

The researchers have developed a new component for electric field control of MRAM devices, with the key innovation being a multiferroic heterostructure with magnetization vectors that can be switched by an electric field. The researchers also demonstrate that two different magnetic states can be reliably realized at zero electric field by changing the sweeping operation of the electric field. This means a non-volatile binary state can be intentionally achieved at zero electric field.

Here are some recent and popular news that you may find of interest:

• Researchers from UNIST design a high-efficiency graphene-based MRAM device
• Samsung details its eMRAM roadmap with 5 nm expected by 2027
• Everspin and QuickLogic to bring MRAM technology to radiation hardened FPGAs
• MRAM-based memory architecture could accelerate AI by a factor of 1000
• Everspin awarded a $14.55 million project to provide stable MRAM manufacturing
• Leading the Charge in MRAM: An Interview with Everspin's CEO
• Researchers from IST and Western Digital develop a promising new SOT-MRAM material
• Purdue University team to develop energy efficiency MRAM-based AI
• Kioxia and Hynix co-develop the world's smallest 1Selector-1MTJ cell
• Renesas achieves the world's fastest STT-MRAM random access speed
• IBM's next-gen FlashCore modules to feature Everspin's 1 Gbg STT-MRAM
• Researchers say MnPd3 will enable a breakthrough in SOT-MRAM memory devices
• NXP and TSMC to offer 16 nm FinFET automotive embedded MRAM
• Samsung first to demonstrate MRAM-based in-memory computing

Researchers from Korea's Ulsan National Institute of Science & Technology (UNIST) have designed a new MRAM structure, based on graphene, that offers higher efficiency (and lower heat generation) compared to existing MRAM solutions.

The design of the MRAM device is based on a graphene layer sandwiched between a magnetic insulator (yttrium iron garnet) and a ferroelectric material (PVDF-TrFE). Upon application of a voltage pulse, the current flow through the graphene is altered, enabling the storage of binary data based on this current direction.

Hprobe, a developer of testing equipment for magnetic devices, announced a strategic collaboration with NY CREATES, a lab-to-fab bridge for advanced electronics, to advance testing capabilities for next-generation semiconductor memory technologies. This joint development project will focus on the co-development of advanced testing equipment at the 300mm wafer scale.

NY CREATES and Hprobe will work together to support new testing solutions for MRAM, RRAM and selector devices. The joint project will focus on initial testing of MRAM and RRAM wafers from NY CREATES using Hprobe’s equipment, including the full-scale implementation of Hprobe’s IBEX wafer-level magnetic tester. Development of breakthrough testing procedures and algorithms for MRAM, RRAM, and selector devices, will also seek to advance the state-of-the-art in semiconductor memory testing.

pMTJ STT-MRAM developer Avalanche Technology announced a new partnership with Trusted Semiconductor Solutions (TSS) to provide valued DDR3/4 non-volatile memory products hardened to strategic levels for the aerospace & defense community. 

Building on prior collaborations between Avalanche and TSS, this new product family will address the critical need for radiation resilient, high reliability execution memory with instant-on capability for Strategic radiation environments. Leveraging Avalanche's broadly adopted modular Gen 3 MRAM product architecture TSS will add specialized survivability features uniquely required for this application and provide the storefront for this tailored solution.

Everspin Technologies reported its financial results for Q3 2024, with revenues reaching $12.1 million (down from $16.5 million in Q3 2023), and a net profit of $2.3 million. 

For the fourth quarter 2024, Everspin expects total revenue in a range of $12 million to $13 million. At the end of Q3 2024, the company had $39.6 million in cash and equivalents, up $2.8 million from the previous quarter. 

Kioxia Corporation (Toshiba's memory unit that was spun-off in 2021), in collaboration with SK Hynix, announced that it has developed the world's smallest 1Selector-1MTJ cell.

Kioxia says that the two companies have demonstrated a reliable 1 selector-1 MTJ (1S1M) cell read/write operation with low read disturb rate of <1E^-6 in 64 Gb cross-point array architecture. They have implemented cross-point 1S1M chips integrated in Half Pitch (HP) of 20.5 nm and MTJ CD of 20 nm using As-doped SiO2 selector and perpendicularly magnetized MTJ (p-MTJ).

Researchers from  Purdue University were awarded with a new project to advanced AI using research performed at Purdue for over a decade. The CHEETA:CMOS+MRAM project will utilize a CMOS+X approach specifically leveraging the unique capabilities of MRAM memory to design efficient in-memory computing hardware fabrics.

The CHEETA project aims to deliver impactful improvements in energy efficiency and sensor-to-decision latency compared to the current commercial standard. The project received funding from the Applied Research Institute.