LIGAND SELECTION FOR TERNARY OXIDE THIN FILMS

    公开(公告)号:US20220328302A1

    公开(公告)日:2022-10-13

    申请号:US17851100

    申请日:2022-06-28

    摘要: Embodiments of the present invention are directed to forming a ternary compound using a modified atomic layer deposition (ALD) process. In a non-limiting embodiment of the invention, a first precursor and a second precursor are selected. The first precursor includes a first metal and a first ligand. The second precursor includes a second metal and a second ligand. The second ligand is selected based on the first ligand to target a second metal uptake. A substrate is exposed to the first precursor during a first pulse of an ALD cycle and the substrate is exposed to the second precursor during a second pulse of the ALD cycle, the second pulse occurring after the first pulse. The substrate is exposed to a third precursor (e.g., an oxidant) during a third pulse of the ALD cycle. The ternary compound can include a ternary oxide film.

    Ligand selection for ternary oxide thin films

    公开(公告)号:US11462398B2

    公开(公告)日:2022-10-04

    申请号:US16514351

    申请日:2019-07-17

    摘要: Embodiments of the present invention are directed to forming a ternary compound using a modified atomic layer deposition (ALD) process. In a non-limiting embodiment of the invention, a first precursor and a second precursor are selected. The first precursor includes a first metal and a first ligand. The second precursor includes a second metal and a second ligand. The second ligand is selected based on the first ligand to target a second metal uptake. A substrate is exposed to the first precursor during a first pulse of an ALD cycle and the substrate is exposed to the second precursor during a second pulse of the ALD cycle, the second pulse occurring after the first pulse. The substrate is exposed to a third precursor (e.g., an oxidant) during a third pulse of the ALD cycle. The ternary compound can include a ternary oxide film.

    SUB-STOICHIOMETRIC METAL-OXIDE THIN FILMS

    公开(公告)号:US20210020427A1

    公开(公告)日:2021-01-21

    申请号:US16516423

    申请日:2019-07-19

    摘要: Embodiments of the present invention are directed to forming a sub-stoichiometric metal-oxide film using a modified atomic layer deposition (ALD) process. In a non-limiting embodiment of the invention, a first precursor and a second precursor are selected. The first precursor can include a metal and a first ligand. The second precursor can include the same metal and a second ligand. A substrate can be exposed to the first precursor during a first pulse of an ALD cycle. The substrate can be exposed to the second precursor during a second pulse of the ALD cycle. The second pulse can occur directly after the first pulse without an intervening thermal oxidant. The substrate can be exposed to the thermal oxidant during a third pulse of the ALD cycle.

    SUB-STOICHIOMETRIC METAL-OXIDE THIN FILMS

    公开(公告)号:US20210272796A1

    公开(公告)日:2021-09-02

    申请号:US17323178

    申请日:2021-05-18

    摘要: Embodiments of the present invention are directed to forming a sub-stoichiometric metal-oxide film using a modified atomic layer deposition (ALD) process. In a non-limiting embodiment of the invention, a first precursor and a second precursor are selected. The first precursor can include a metal and a first ligand. The second precursor can include the same metal and a second ligand. A substrate can be exposed to the first precursor during a first pulse of an ALD cycle. The substrate can be exposed to the second precursor during a second pulse of the ALD cycle. The second pulse can occur directly after the first pulse without an intervening thermal oxidant. The substrate can be exposed to the thermal oxidant during a third pulse of the ALD cycle.

    Sub-stoichiometric metal-oxide thin films

    公开(公告)号:US11081343B2

    公开(公告)日:2021-08-03

    申请号:US16516423

    申请日:2019-07-19

    摘要: Embodiments of the present invention are directed to forming a sub-stoichiometric metal-oxide film using a modified atomic layer deposition (ALD) process. In a non-limiting embodiment of the invention, a first precursor and a second precursor are selected. The first precursor can include a metal and a first ligand. The second precursor can include the same metal and a second ligand. A substrate can be exposed to the first precursor during a first pulse of an ALD cycle. The substrate can be exposed to the second precursor during a second pulse of the ALD cycle. The second pulse can occur directly after the first pulse without an intervening thermal oxidant. The substrate can be exposed to the thermal oxidant during a third pulse of the ALD cycle.

    LIGAND SELECTION FOR TERNARY OXIDE THIN FILMS

    公开(公告)号:US20210020426A1

    公开(公告)日:2021-01-21

    申请号:US16514351

    申请日:2019-07-17

    摘要: Embodiments of the present invention are directed to forming a ternary compound using a modified atomic layer deposition (ALD) process. In a non-limiting embodiment of the invention, a first precursor and a second precursor are selected. The first precursor includes a first metal and a first ligand. The second precursor includes a second metal and a second ligand. The second ligand is selected based on the first ligand to target a second metal uptake. A substrate is exposed to the first precursor during a first pulse of an ALD cycle and the substrate is exposed to the second precursor during a second pulse of the ALD cycle, the second pulse occurring after the first pulse. The substrate is exposed to a third precursor (e.g., an oxidant) during a third pulse of the ALD cycle. The ternary compound can include a ternary oxide film.

    ASSIGNING DNN WEIGHTS TO A 3D CROSSBAR ARRAY
    10.
    发明公开

    公开(公告)号:US20240202275A1

    公开(公告)日:2024-06-20

    申请号:US18085011

    申请日:2022-12-20

    IPC分类号: G06F12/02 G06F17/16

    CPC分类号: G06F12/0207 G06F17/16

    摘要: A system, method and computer program product for assigning deep neural network (DNN) weight matrices to a Compute-in-Memory (CiM) accelerator system, and particularly, efficient allocation strategies for assigning DNN model weight-layers to two-dimensional (2D) tiers of three-dimensional (3D) crossbar array tiles. Such efficient allocation strategies for assigning DNN model weight-layers to tiers and tiles of a CiM accelerator are optimized to minimize contention, latency and dead-time, and to maximize accelerator throughput. In one scenario, efficient allocation strategies include assigning DNN weight matrices to the 2D tiers of a 3D crossbar array tile to maximize throughput and minimize completion latency for a finite-batch-size example of an incoming workflow. In a further scenario, efficient allocation strategies assign DNN weight matrices to the 2D tiers of a 3D crossbar array tile to minimize dead-time-latency-before-next-batch-member-can-be-input in an infinite-batch-size or a continuous workflow scenario.