| With the advent of the new era of information and the rise of the metaverse,the construction of high-speed Internet of Things puts forward higher requirements for real-time acquisition of information data,high-speed storage and reading,and high-efficiency processing capabilities.New sensors with power consumption advantages are used for information collection.Due to the advantages of high detection sensitivity,small size,and easy integration,the giant magnetoresistance(GMR)sensor with spin valve structure has extremely high application value in weak magnetic detection,biological detection,flexible electronic skin and other fields.In this thesis,a spin-valve structure GMR magnetic sensor with high magnetoresistance rate of change is developed,and based on this,the hierarchical local exchange bias regulation of the sensor is studied.With the help of magnetic annealing furnace and current annealing method,the exchange bias regulation between the layers of the GMR strip structure and the Wheatstone bridge structure realizes the linearized response of the sensor to the magnetic field.The optimization of material system and thickness parameters of spin valve structure GMR device is studied in this thesis.And the GMR samples with high magnetic resistance rate of6.36 % are prepared with the growth parameters as: substrate/ Ta(1.5)/Ni Fe(3)/Co Fe(4)/Cu(4.5)/Co Fe(4)/Ir Mn(8)/Ta(2)(unit: nm).Then the temperature and magnetic field parameters required for the linearization of the sensor are determined through theoretical analysis and practical experiments of the GMR material system.The magnetic moment of the pinned layer is perpendicular to the magnetic moment of the free layer by annealing along the short direction and long direction of the reluctance strip successively at 300 ℃,1k Oe and 145 ℃,300 Oe.The resulting reluctance strip structure alone demonstrated a linear response with a sensitivity of 0.63 m V/V/Oe,while the Wheatstone bridge sensor assembled with the device through this linearization process has a sensitivity of 0.31 m V/V/Oe in the range of-51.71 Oe to 44.58 Oe,and the linearity is about 7.01 %.In subsequent studies,current annealing,which is a more efficient and low-cost way,is selected to realize local annealing treatment of the same substrate.Under the driving current parameters of 17 m A and 13 m A,the vertical structure of the magnetic moment of the pinned layer and the magnetic moment of the free layer is realized,and the sensor response with linear sensitivity of about 0.43 m V/V/Oe is obtained.The Wheatstone bridge structure with sensitivity of 0.21 m V/V/Oe measured from-33.11 Oe to 38.01 Oe with a linearity of 6.23 % is obtained by linearization of four groups of reluctance strips on the same substrate by current annealing.In this thesis,starting from the linearization requirements of GMR magnetic sensors,through the analysis and comparison of the current general linearization processing methods,the traditional magnetic annealing furnace and current annealing methods are innovatively used to realize the exchange bias regulation between GMR layers.Based on this,the linear control of the magnetic sensor is completed,the magnetic field linear response of the separated GMR strip is realized,and the Wheatstone full-bridge sensor with linear output is obtained.This thesis briefly analyzes the advantages and limitations of the two linearization processing methods,and puts forward the direction and ideas for further optimization.The research results of this thesis provide a feasible technical method for solving the current large-scale industrial production problems of GMR linear magnetic sensors in the field of small-sized and high-precision integrated sensors,and promote the further application and development of GMR magnetic sensors. |
