Weebit Nano 3FQ19 Activities Update

Key highlights for 3Q19

• Formed partnership with Silvaco to create ReRAM models and tools
• Demonstrated excellent test results ahead of move to 300mm wafers at 28nm
• Extended commercial agreement with Leti into artificial intelligence
• Collaborations with leading research institutes Technion and Politecnico di Milano
• Dr Nissan-Cohen to provide additional services as Weebit prepares to transfer its technology to a production fab and progress discussions with potential partners.

Weebit Nano Ltd provides an update for the quarter ending 31 March 2019, along with its Appendix 4C cash flow results.

Development Program Partnership with Silvaco
In January, Weebit Nano formed a development program partnership with Silvaco, Inc., a provider of software, IP and services for designing chips and electronic systems for semiconductor companies.

The two companies will work together to develop a Technology Computer-Aided Design (TCAD) solution to accurately model the electrical behavior of Weebit’s ReRAM devices. The models, based on firm’s physical samples and data, will be the basis for creating a Silvaco TCAD solution for joint customers employing ReRAM technology.

Silvaco’s TCAD simulation will speed up the design, fabrication and commercial use of ReRAM technology through the elimination of expensive and time-consuming experimental wafers during technology adoption by OEMs.

Test results ahead of move to 300mm wafers
The company demonstrated successful endurance results of its ReRAM cells in January, at levels competitive to production non-volatile memories. Reaching these technical parameters is a key step towards moving to 300mm wafers at 28nm.

Weebit performed the tests with its partner Leti, the French research institute in the field of micro-electronics, which demonstrated Array-level endurance above 100,000 cycles – on par with expectation in the storage memory market, and a significant improvement over flash memories.

In addition to technical parameter improvements, Weebit created a more flexible manufacturing base, ensuring the SiOx ReRAM layer will be compatible with different tools and technologies used by different production fabs. This is crucial for transferring Weebit’s technology to different commercial manufacturers.

Extended commercial agreement with Leti into artificial intelligence
During the reporting period, Weebit extended its existing partnership with Leti to develop a system which will demonstrate AI algorithms implemented using silicon oxide (SiOx) ReRAM.

The new agreement will combine Weebit’s SiOx ReRAM technology and Leti’s development of AI technologies and architectures to build a neuromorphic demonstration system.

The system will be based on a Spiking Neural Network (SNN) test vehicle for implementing synapses, which will showcase the capabilities of the technology by performing precise object recognition tasks in an energy-efficient manner.

The system will demonstrate a technology disrupting the object recognition models currently being used.

Collaborations with research institutes Technion & Politecnico di Milano
During the quarter, Weebit partnered with research institutes Politecnico di Milano (Polimi), an European university for Industrial and Information engineering, technology and industrial design and the Technion (Israel Institute of Technology) on ReRAM projects.

In January, the company launched a joint Neuromorphic ReRAM project with Polimi to test, characterise and implement its developed algorithms using Weebit’s ReRAM chips to demonstrate the capability of ReRAM-based hardware in neuromorphic and artificial intelligence applications.

The research at Polimi will be led by Professor Daniele Ielmini. He is a Fellow of the IEEE (Institute of Electrical and Electronics Engineers), has held visiting positions at Intel and Stanford University, and received the Intel 2013 Outstanding Researcher Award.

In February, Weebit and the Technion agreed to examine the possible use of ReRAM memories in a novel computing architecture that could speed up processing, memory transfer rate and memory bandwidth, and decrease processing latency – while using much less power.

Weebit and the Technion will perform characterisation and implementation of logic operations using SiOx ReRAM test chips, demonstrating basic logic operations on a ReRAM array using the Technion’s MAGIC (Memristor Aided Logic) technique implementing ‘Real Processing in Memory’, or Real PIM. Research at the Technion will be led by Professor Shahar Kvatinsky, a pioneer in the field of circuits and architectures with emerging memory technologies and design of energy-efficient architectures.

Agreement with Dr Yoav Nissan-Cohen and share consolidation

During the reporting period, Weebit executive director Dr Yoav Nissan-Cohen agreed to increase his time commitment with the company. He will be providing additional services as Weebit prepares to transfer its technology to a production fab, along with progressing discussions with potential partners, delaying the need to hire a VP of R&D.

In line with the increase in his time commitment, consideration paid to Nissan-Cohen’s consulting company, Nicohy Ltd, under the Consulting Agreement dated January 1, 2018, increased from NIS20,000 to NIS40,000 per month.

In February, Weebit Nano underwent a share consolidation at a ratio of 25:1. The consolidation improves the company’s investment proposition to partner prospects as it moves closer to commercialisation.

Looking ahead
In collaboration with research partner Leti, Weebit Nano continues to improve the baseline parameters of its SiOx ReRAM technology, and is moving closer to producing 300mm wafers at 28nm.

It continues to progress discussions with potential customers/partners, while the company also explores other emerging applications for its ReRAM technology, such as neuromorphic computing and real PIM. The assistance of executive director Nissan-Cohen is invaluable in advancing these negotiations and preparing for the move to a production fab.