Abstract:
A data retention reliability screen of integrated circuits including ferroelectric random access memory (FRAM) arrays. Sampled groups of cells in the FRAM array are tested at various reference voltage levels, after programming to a high polarization capacitance data state and a relaxation time at an elevated temperature. Fail bit counts of the sample groups at the various reference voltage levels are used to derive a test reference voltage, against which all of the FRAM cells in the integrated circuit are then tested after preconditioning (i.e., programming) and another relaxation interval at the elevated temperature, to determine those cells in the integrated circuit that are vulnerable to long-term data retention failure.
Abstract:
A data retention reliability screen of integrated circuits including ferroelectric random access memory (FRAM) arrays. Sampled groups of cells in the FRAM array are tested at various reference voltage levels, after programming to a high polarization capacitance data state and a relaxation time at an elevated temperature. Fail bit counts of the sample groups at the various reference voltage levels are used to derive a test reference voltage, against which all of the FRAM cells in the integrated circuit are then tested after preconditioning (i.e., programming) and another relaxation interval at the elevated temperature, to determine those cells in the integrated circuit that are vulnerable to long-term data retention failure.
Abstract:
A reliability screen of integrated circuits including ferroelectric random access memory (FRAM) arrays for stuck bits. The FRAM devices are subjected to a high temperature bake in wafer form. A “shmoo” of the reference voltage is performed, at an elevated temperature, for each device to identify a first reference voltage at which a first cell in the device fails a read of its low polarization capacitance data state, and a second reference voltage at which a selected number of cells in the device fail the read. The slope of the line between the first and second reference voltages, in the cumulative fail bit count versus reference voltage plane, is compared with a slope limit to determine whether any stuck bits are present in the device.
Abstract:
A system includes a ferroelectric random access memory (FRAM) array having one or more memory elements. A cycle controller cycles data to be fixed in a subset of the one or more memory elements by reading or writing the data a predetermined number of times to fix the data to a non-volatile stable state.
Abstract:
A system includes a ferroelectric random access memory (FRAM) array having one or more memory elements. A cycle controller cycles data to be fixed in a subset of the one or more memory elements by reading or writing the data a predetermined number of times to fix the data to a non-volatile stable state.
Abstract:
A method of setting the reference voltage for sensing data states in integrated circuits including ferroelectric random access memory (FRAM) cells of the one-transistor-one capacitor (1T-1C) type. In an electrical test operation, some or all of the FRAM cells are programmed to a particular polarization state. A “shmoo” of the reference voltage for sensing the data state is performed, at one or more worst case electrical or environmental conditions for that data state, to determine a reference voltage limit at which the weakest cell fails to return the correct data when read. A configuration register is then written with a reference voltage based on this reference voltage limit, for example at the limit plus/minus a tolerance.
Abstract:
A data retention reliability screen of integrated circuits including ferroelectric random access memory (FRAM) arrays. A reference voltage level is determined for each integrated circuit being tested, corresponding to the read of a high polarization capacitance data state. A number of FRAM cells in the integrated circuit are programmed to that data state, and then read at an elevated temperature, with the number of failing cells compared against a pass/fail threshold to determine whether the integrated circuit is vulnerable to long-term data retention failure.
Abstract:
A method of setting the reference voltage for sensing data states in integrated circuits including ferroelectric random access memory (FRAM) cells of the one-transistor-one capacitor (1T-1C) type. In an electrical test operation, some or all of the FRAM cells are programmed to a particular polarization state. A “shmoo” of the reference voltage for sensing the data state is performed, at one or more worst case electrical or environmental conditions for that data state, to determine a reference voltage limit at which the weakest cell fails to return the correct data when read. A configuration register is then written with a reference voltage based on this reference voltage limit, for example at the limit plus/minus a tolerance.
Abstract:
Curing of a passivation layer applied to the surface of a ferroelectric integrated circuit so as to enhance the polarization characteristics of the ferroelectric structures. A passivation layer, such as a polyimide, is applied to the surface of the ferroelectric integrated circuit after fabrication of the active devices. The passivation layer is cured by exposure to a high temperature, below the Curie temperature of the ferroelectric material, for a short duration such as on the order of ten minutes. Variable frequency microwave energy may be used to effect such curing. The cured passivation layer attains a tensile stress state, and as a result imparts a compressive stress upon the underlying ferroelectric material. Polarization may be further enhanced by polarizing the ferroelectric material prior to the cure process.
Abstract:
Disclosed embodiments include a testing system that electrically connects to an integrated circuit (IC) having ferroelectric memory (FRAM) cells. The testing system programs the FRAM cells to a first data state and then iteratively reads the programmed cells at a plurality of reference voltages to identify a reference voltage limit that indicates a first occurrence at which at least one of the cells fails to return the first data state when read. Iteratively reading the cells includes reading each cell at an initial reference voltage at which all the cells return the first data state, and then reading each of the programmed cells at each of the remaining reference voltages by incrementally changing the initial reference voltage in one direction until the reference voltage limit is identified. The testing system sets the reference in the IC at an operating level based on the reference voltage limit.