CN1281986A - Integrated silicon microresistance type acceleration sensor and its manufacturing method - Google Patents

Abstract

The present invention provides an integrated silicon microresistive acceleration transducer and its production method. it includes base, shell body and detection chip consisting of eight mass blocks supported by strain beam, piezoresistor and bridge circuit, and mainly adopts the working technological processes of miniature machine special-purpose double polished silicon water, SiO2 protective layer, photoetching, phosphorus diffusion and phosphorus ion implantation, annealing (silicon wafer), EPW anisotropic corrosion and RIE corrosion. Its measurement portion is an integral structure, and its anti-overloading performance is strong, so it can use one transducer to implement acceleration detection from one-dimensional to three-dimensional and arbitrary direction.

Description

Integrated silicon micro-resistance type acceleration sensor and manufacturing and processing method thereof

The invention belongs to the technical field of test sensors, and relates to an integrated silicon micro-resistor type acceleration sensor and a manufacturing and processing method thereof.

The acceleration sensor is a common sensor for measuring motion parameters and vibration data of various machines, vehicles, ships, aviation and spacecrafts, and the traditional acceleration sensor structure has a piezoelectric type. It is composed of one or more piezoelectric sheets under the mass block and pressed by spring on a relatively thick metal base, and there is a casing protecting the piezoelectric sheets, and the quantity of electric charge presented by the piezoelectric sheets is proportional to the acceleration of the sensor in a certain frequency range. If the outer body fixed type spring and the shell are fixed together, the base and the shell are part of a spring-mass system, and the defects are that the external temperature, noise and deformation of a test piece influence the output of acceleration through the reaction of the shell and the base, the output signal is small, and the output impedance is high. In addition, a strain type acceleration sensor is provided, which adopts a mass block, an elastic beam, a strain gauge, a base and a shell structure. Although the above disadvantages of the piezoelectric acceleration sensor are overcome, the structure of the piezoelectric acceleration sensor makes the manufactured acceleration sensor larger, and affects the application range.

With the development of large scale integrated circuits, chip fabrication techniques, and micromachining techniques in recent years. The formation of a metal thin film on a silicon wafer (semiconductor) is produced, and a semiconductor acceleration sensor is manufactured by etching. Chinese patents 97114505.9 and 9711322 are a semiconductor acceleration sensor and a method of manufacturing the same. The whole measuring component is installed on a ceramic chip, the cantilever beam is made of semiconductor material and cut by micro-machining, and a mass block is additionally installed. Because the deformation beam, the mass block and other split structures are assembled, the overload resistance is poor, and the durability and the reliability are poor when the deformation beam, the mass block and other split structures are used. The three sensors are only suitable for measuring unidirectional acceleration, and when three-dimensional directions and any direction are measured, the three sensors are required to be installed with high precision, and the measurement error is large because the three sensors are not located at the same point.

The invention aims to overcome the defects of the acceleration sensor and the defects thereof, and provides an integrated silicon micro-resistor type acceleration sensor which has an integral measuring part, strong overload resistance and capability of measuring acceleration from one dimension to three dimensions.

The invention provides an integrated silicon micro-resistance type acceleration sensor, which comprises a base, a shell, a detection chip and a control chip, wherein the shell is arranged on the base and plays a role in protection, and the detection chip consists of a stress sensitive part, a mass block part, a piezoresistor and a bridge circuit; the detection chip is characterized in thatthe center of the detection chip is provided with a micro-motion mass block, two, four or eight mass blocks are symmetrically distributed around the mass block and used for supporting the mass block, and the sensitive part is a strain beam.

The integrated silicon micro-piezoresistive acceleration sensor is characterized in that the mass block and the strain beam are formed by a whole single crystal silicon wafer formed by micromachining.

The integrated silicon micro-resistor type acceleration sensor is characterized in that the mass block is a regular diamond table body, the four side faces and the bottom face of the mass block form included angles of α, the peripheral edge of the supporting strain beam is trapezoidal, and the inner side face and the bottom face form an included angle of β.

The integrated silicon micro-resistor type acceleration sensor is characterized in that the piezoresistor is a resistor structure which is formed by injecting phosphorus ions into a silicon wafer, and the piezoresistor is positioned at the maximum stress point on each beam.

The integrated silicon micro-resistor type acceleration sensor is characterized in that the piezoresistors respectively form a bridge circuit for measuring X, Y, Z accelerations in three directions, and the bridge circuit comprises: r1= R2= R3= R4= R5= R6= R7= R8= R12= R13= R, R9= R10= R11=4R and R9, R10, R11, R12, R13 are fixed resistances on the outer edge of the chip.

The invention relates to a manufacturing and processing method of a chip of an integrated silicon micro-resistance type acceleration sensor, which is characterized by comprising the following steps: (9) selecting materials, namely taking a double-polished silicon wafer special for a micro machine; (10) adopting a SiO2 protective layer; (11) photoetching, namely, conducting wires, resistors, mass blocks and diamond-shaped tables; (12) manufacturing a low-resistance wire by a phosphorus diffusion method; (13) injecting phosphorus ions to form an N-type piezoresistor; (14) annealing (silicon wafer); (15) EPW anisotropically etching to obtain quality blocks and diamond-shaped platforms; (16) RIE etching, dry etching out the strain beam.

The invention provides an integrated silicon micro-resistor type acceleration sensor. The whole measuring component is installed on a ceramic chip, the cantilever beam is made of semiconductor material and is cut by micro-machining, and the acceleration sensor is assembled by the structure. The device overcomes the defects that the overload resistance is poor, the durability and the reliability are poor when the device is used, the device is only suitable for measuring unidirectional acceleration, three sensors are needed when the device is used for measuring the three-dimensional direction, the mounting precision is high, and the measuring error is large because the sensors are not on the same point, and the defects. The invention is a sensor which has an integral measuring part, strong overload capacity and can carry out unidirectional and three-dimensional acceleration measurement.

In the acceleration sensor measurement section, simulation calculation was performed using Ansys analysis software. Although the structure is two-dimensional symmetrical, the stress is asymmetrical, and in order to analyze the separability of the structure to the triaxial acceleration, the whole structure is subjected to simulation calculation, fig. 10 shows a finite element model of the acceleration sensor, Solid 45 eight-node 3D units are selected, all the units are of standard cuboid structures, the ratio of length, width and height is close to 1, 8868 units are provided, and 10080 nodes are provided. The material properties were calculated as follows: modulus of elasticity: ex =130 × 10 ^ 9Pa, shear modulus: g=79 × 10 ^ 9Pa, density: 2.33X 10 ^ 3 Kg/m ^ 3, the poisson ratio: 0.18. fig. 11, 12 and 13 are deformation diagrams of the structure under unidirectional x, y and z loads, respectively, and fig. 14 is a deformation diagram of the structure under combined x, y and z loads. Fig. 15 is a displacement curve. Fig. 16 is a stress curve. The simulation result is similar to the detection result, the single value error is within 10 percent, and the theory that the stress state at the micron level is not related to the scale effect is verified. According to the results of the computer simulation analysis, the integrated silicon micro-resistor type acceleration sensor provided by the invention has good reliability and durability.

The invention provides a manufacturing and processing method of a chip of an integrated silicon micro-resistance type acceleration sensor. It is not exactly the same as silicon wafer used in microelectronic processing, piezoresistive sensors based on piezoresistive effect require high piezoresistive coefficient, and the piezoresistor is located in the direction of the greatest piezoresistive coefficient of the silicon wafer. The special double-polished silicon wafer for the micromachine is adopted, a P-type (100) surface is used as a substrate, a resistance layer adopts phosphorus ion implantation as an N-type piezoresistor, and the processes of photoetching, plate making, corrosion, annealing and the like are adopted. The chip of the acceleration sensor with high precision, strong overload resistance, small temperature and zero drift and good stability is manufactured according to the design requirement.

Description of the drawings:

FIG. 1 is a schematic diagram of an integrated silicon micro-resistor type acceleration sensor;

FIG. 2 is a schematic structural diagram of an acceleration sensor chip with two deformation beams;

FIG. 3 is a schematic structural diagram of an acceleration sensor chip with four deformation beams;

FIG. 4 is a front view of an acceleration sensor chip structure with eight deformation beams;

FIG. 5 is a structural sectional view of an acceleration sensor chip with eight deformation beams;

FIG. 6 is a layout diagram of resistors and wires on an acceleration sensor chip with eight deformation beams;

FIG. 7 is an x-direction bridge circuit diagram;

FIG. 8 is a y-direction bridge circuit diagram;

FIG. 9 is a z-bridge circuit diagram;

FIG. 10 is a finite element model of an acceleration sensor;

FIG. 11 is a deformation diagram of the structure under x unidirectional loads;

FIG. 12 is a deformation diagram of the structure under y unidirectional loads;

FIG. 13 is a deformation diagram of the structure under z-unidirectional load;

FIG. 14 is a deformation diagram of the structure under x, y, z three-directional composite loads;

FIG. 15 is a graph of displacement;

FIG. 16 is a stress plot;

FIG. 17 is a diagram of a mechanical model with force applied to the x-axis;

FIG. 18 is a diagram of a mechanical model with force applied to the y-axis;

FIG. 19 is a diagram of a mechanical model with z-axis force;

FIG. 20 is a schematic view of a window of the EPW system before etching a silicon wafer;

FIG. 21 is a schematic view of a window after etching a silicon wafer by the EPW system;

FIG. 22, FIG. 23 and FIG. 24 are schematic views of lobe compensation for etching a silicon wafer mask for the EPW system;

fig. 25 is a photograph showing a real object of the result after the three-dimensional acceleration sensor chip is processed;

exemplary embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a diagram of an integrated silicon micro-resistance type acceleration sensor, which comprises a base 1, a shell 2 mounted on the base for protection, and a detection chip 3 composed of a stress sensitive part, a mass part, a piezoresistor and a bridge circuit; fig. 2, 3 and 4 show the detection chip, wherein a micro-motion mass 4 is arranged in the center of the detection chip, the sensitive parts are symmetrically distributed around the mass, and two, four or eight strain beams 5 are used for supporting the mass. The mass block and the strain beam are formed by a whole single crystal silicon wafer which is formed by micro-machining.

In fig. 5, the mass in the acceleration sensor chip is shown as a regular diamond frustum (or a truncated cone) with four side faces and a bottom face all having an included angle of α, while the peripheral edge of the supporting strain beam is trapezoidal and the inner side face and the bottom face have an included angle of β, and the preferred included angles of α and β are both 54.7 °.

Fig. 6 shows a resistive acceleration sensor, and the piezoresistor 6 is a resistor structure implanted with phosphorus ions onto a silicon wafer, and is located at the maximum stress point on each beam. The maximum stress point of the eight strain beams for supporting the mass block can be determined by the following calculation method:

mechanical structure parameter calculation

The accelerometer mass is made from EPW anisotropic etching. The mass block of the acceleration sensor is a regular quadrangular frustum pyramid, and the included angles between the four side surfaces and the bottom surface of the quadrangular frustum pyramid are all 54.7 degrees, as shown in fig. 5. The length of the upper bottom edge is a1, the length of the lower bottom edge is b, the height is H, the volume of the mass block is V, the mass of the mass block is m, and the distance between the center and the xoy plane is zc. The beam length is 1, the width is w, the thickness is h, the distance between the centers of the two beams is a2, and the distance from the center of mass to the upper surface is zC. Take a1=800 μm, 1=750 μm, H =340 μm, w =68 μm, H =40 μm, a2=400 μm mass block bottom width b = a1-2 × H/tg 54.7⁰=800×10^-6-2×(300×10^-6)／tg54．7⁰=375×10^-6mMass volume V = 1/3H (a)₁ ²+b²+a₁b)=1．08×10^-10m³Mass m = 2.33 × 10 of the mass block^-3V=2．49×10^-7The position Zc = 1.02 × 10 of the mass center of the kg mass block^-4m

The acceleration a in any direction can be decomposed into three acceleration components of ax, ay and az in the directions of x, y and z. When the accelerometer is subjected to the action of the speed a in any adding direction, the force analysis is as follows:

e — young's modulus of elasticity<100>E = 1.3 × 1011pa in the following formulae;

g — shear modulus of elasticity<100>G = 7.9 × 107 Pa;

μ -poisson ratio<100>μ = 0.18; A. when only the acceleration ax in the x-direction is applied, the acceleration is decomposed into two terms (one) which is acted on by the inertial force-max in the y-plane, and the deformation is symmetrical to the x-axis, as shown in fig. 17 below.

The stress condition is symmetrical to the x axis, and the stress of the beams on the two sides of the y axis is equal in magnitude and opposite in direction. Therefore, only one fourth of the stress conditions of the structure, namely the stress conditions of the beams 1-2, 3-4 and the stress conditions of the beams 3-4, need to be analyzed: is subjected to axial pressure at the beam 3 ${\mathrm{<figure-callout\; id="3"\; label="N"\; filenames="0012617200223.png,0012617200232.png"\; state="\{\{state\}\}">N</figure-callout>}}_3=\frac{\mathrm{Ewh}}{l}\mathrm{\&Delta;x}$ Is subjected to an axial pressure N at the beam 4₄=N₃Compressive stress σ at 3, 4_3xi=σ_4xi=-N₃The maximum stress sigma 3-4 of the/wh beam 3-4_ximax=σ3_xi=σ4_xiStress condition of the beam 1-2: the beam 2 is subjected to shearing force $Q_{}$ ${}_{2\mathrm{\&pi;t}}=\frac{{\mathrm{Ehw}}^3}{{\mathrm{<figure-callout\; id="3"\; label="l"\; filenames="0012617200223.png,0012617200232.png"\; state="\{\{state\}\}">l</figure-callout>}}^3}\mathrm{\&Delta;x}$ The beam 1 is subjected to a shear force Q_1xi=Q_2xiBending moment at the beam 2 ${\mathrm{<figure-callout\; id="2"\; label="M"\; filenames="0012617200273.png"\; state="\{\{state\}\}">M</figure-callout>}}_{2\mathrm{xt}}=\frac{{\mathrm{ma}}_{x}l\frac{w^2}{{\mathrm{<figure-callout\; id="2"\; label="l"\; filenames="0012617200273.png"\; state="\{\{state\}\}">l</figure-callout>}}^2}}{8(1+\frac{w^2}{l^2})}$ Bending moment M at the beam 1_1xi=M_2xiMaximum stress sigma on beams 1, 2_1xi=M_1i／hw²=σ_2xiOccurring when the side beam 1-2 is subjected to the maximum stress sigma_{1_2ximax}=σ_1xi=σ_2xiWhere △ x is the displacement of the mass in the x-direction, $\mathrm{\&Delta;x}=\frac{{\mathrm{ma}}_x}{\frac{4\mathrm{Ewh}}{l}(1+\frac{w^2}{l^2})}$

and (II) is subjected to moment of inertia about the y-axis, mayzC, and deformation is symmetrical about the x-axis, as shown in FIG. 18 below. Therefore, only one fourth of the structure, namely the beams 1-2 and 3-4, needs to be analyzed, the stress condition of the beam 1-2 is symmetrical to the x axis, the stress of the beams on the two sides of the y axis is equal in magnitude, and the stress conditions are opposite in direction: 1, 2 part is subjected to shearing force $Q_{1\mathrm{xm}}=Q_{2\mathrm{xm}}=\frac{{\mathrm{Ewh}}^3{a}_2}{{2l}^3}{\mathrm{\&theta;}}_y$ Moment of bending at beam 2 $M_{2\mathrm{ym}}=\frac{{\mathrm{Ewh}}^3{a}_2}{{4l}^2}{\mathrm{\&theta;}}_y$ Flexural moment M of beam 1_1xm=M_2xmThe maximum stress at 1 position on the beam occurs at the beam top ${\mathrm{\&sigma;}}_{1\mathrm{xm}}=-\frac{M_{1\mathrm{xm}}}{\frac{{\mathrm{wh}}^2}{6}}$ The maximum stress at 2 places on the beam occurs at the beam top ${\mathrm{\&sigma;}}_{2\mathrm{xm}}=\frac{M_{2\mathrm{xm}}}{\frac{{\mathrm{wh}}^2}{6}}$ The beam 1-2 is subjected to the maximum stress ${\mathrm{\&sigma;}}_{1-2\mathrm{xm}\max }={\mathrm{\&sigma;}}_{1\mathrm{xm}}={\mathrm{\&sigma;}}_{2\mathrm{xm}}$ The beam 1-2 is acted on by torque $M_{\mathrm{Txl}2}=\frac{{\mathrm{Gwh}}^3}{3l}{\mathrm{\&theta;}}_y$ The shear stress is the same in the section of the beam 1-2 $t_{x1}=\frac{M_{\mathrm{Tx}12}}{\frac{1}{3}{\mathrm{wh}}^2}=t_{x2}$ The maximum shear force occurs at the beam top midpoint $t_{\max }=\frac{M_{\mathrm{Tx}12}}{{0.25\mathrm{wh}}^2}$ Stress condition of the beams 3-4: 3, 4 part is subjected to shearing force $Q_{3\mathrm{xm}}=Q_{4\mathrm{xm}}=\frac{{\mathrm{Ewh}}^3}{{2l}^3}{(\mathrm{al}+l)\mathrm{\&theta;}}_y$ Bending moment of beam 3 $M_{3\mathrm{xm}}=\frac{{\mathrm{Ewh}}^3}{{12l}^2}({3a}_1+2l){\mathrm{\&theta;}}_y$ Flexural moment of beam 4 $M_{4\mathrm{xm}}=\frac{{\mathrm{Ewh}}^3}{{12l}^2}({3a}_1+4l){\mathrm{\&theta;}}_y$ Maximum compressive stress at the beam 3 ${\mathrm{\&sigma;}}_{3\mathrm{xm}}=\frac{\left|M_{3\mathrm{xm}}\right|}{\frac{{\mathrm{wh}}^2}{6}}$ Maximum tensile stress at the beam 4 ${\mathrm{\&sigma;}}_{4\mathrm{xm}}=-\frac{\left|M_{4\mathrm{xm}}\right|}{\frac{{\mathrm{wh}}^2}{6}}$ The beams 3-4 are subjected to maximum stress ${\mathrm{\&sigma;}}_{3-4\mathrm{xm}\max }={\mathrm{\&sigma;}}_{3\mathrm{xm}}={\mathrm{\&sigma;}}_{4\mathrm{xm}}$ Where thetay is the corner of the mass, ${\mathrm{\&theta;}}_y=\frac{{3\mathrm{ma}}_x{z}_c{l}^3}{{\mathrm{wh}}^3[E({6a}_{1}l+{3a}_1^2+{4l}^2)+{3a}_2^{2}E+{4\mathrm{Gl}}^2]}$ B. the y-direction acceleration ay (I) is acted by an inertia force-may in the xoy plane, and the deformation is symmetrical to the y axis. The stress condition is symmetrical to the y axis, the stress on the two side beams on the x' axis is equal in magnitude and in the same directionAnd the reverse. Therefore, only one fourth of the structure, the beams 1-2 and 3-4 and the beams 1-2 need to be analyzed: the beam 1-2 being subjected to axial pressure $N_1=N_2=\frac{\mathrm{Ewh}}{l}\mathrm{\&Delta;y}$ The beam 1-2 being under axial compressive stress ${\mathrm{\&sigma;}}_{1\mathrm{yi}}={\mathrm{\&sigma;}}_{2\mathrm{yi}}=-\frac{N_1}{{\mathrm{wh}}^2}$ The beam 1-2 is subjected to the maximum stress ${\mathrm{\&sigma;}}_{1-2\mathrm{yim}x}={\mathrm{\&sigma;}}_{1\mathrm{yt}}={\mathrm{\&sigma;}}_{2\mathrm{yt}}$ Stress condition of the beams 3-4: 3, 4 part is subjected to shearing force $Q_{3\mathrm{yi}}=Q_{4\mathrm{yi}}=\frac{{\mathrm{Ew}}^{3}h}{l^3}\mathrm{\&Delta;y}$ Bending moment at beam 3 $M_{3\mathrm{yi}}=\frac{{\mathrm{ma}}_{y}l\frac{w^2}{l^2}}{8(1+\frac{w^2}{l^2})}$ Bending moment at beam 4 $M_{4\mathrm{yi}}=M_{3\mathrm{yi}}$ The maximum stress at the beam 3, 4 occurs at the side of the beam ${\mathrm{\&sigma;}}_{3\mathrm{yi}}={\mathrm{\&sigma;}}_{4\mathrm{yi}}=\frac{M_{3\mathrm{yi}}}{\frac{w^{2}h}{6}}$ The beams 3-4 are subjected to maximum stress ${\mathrm{\&sigma;}}_{3-4\mathrm{yi}\max }={\mathrm{\&sigma;}}_{3\mathrm{yi}}={\mathrm{\&sigma;}}_{4\mathrm{yi}}$ Where △ y is the displacement of the mass in the y-direction, $\mathrm{\&Delta;y}=\frac{{\mathrm{ma}}_y}{\frac{4\mathrm{Ewh}}{l}(1+\frac{w^2}{l^2})}$ and (II) the deformation is symmetrical to the x axis under the action of the inertia moment-mayzC around the x axis. The stress condition is symmetrical to the y axis, and the stress of the two side beams on the x axis is equal in magnitude and opposite in direction. Therefore, only one fourth of the structure, namely the stress of the beams 1-2, 3-4 and the stress of the beams 3-4, need to be analyzed: shear force at 3, 4 parts of beam $Q_{3\mathrm{ym}}=Q_{4\mathrm{ym}}=\frac{{\mathrm{Ewh}}^3{a}_2}{{2l}^3}{\mathrm{\&theta;}}_y$ Bending moment at beam 3 $M_{3\mathrm{ym}}=\frac{{\mathrm{Ewh}}^3}{{4l}^2}{\mathrm{\&theta;}}_x$ Bending moment at beam 4 $M_{4\mathrm{ym}}=\frac{{\mathrm{Ewh}}^3}{{4l}^2}{\mathrm{\&theta;}}_x$ Maximum tensile stress of the upper surface of the beam 3 ${\mathrm{\&sigma;}}_{3\mathrm{ym}}=\frac{M_{3\mathrm{ym}}}{\frac{{\mathrm{wh}}^2}{6}}$ Maximum compressive stress of upper surface of beam 4 ${\mathrm{\&sigma;}}_{4\mathrm{ym}}=-\frac{M_{4\mathrm{ym}}}{\frac{{\mathrm{wh}}^2}{6}}$ The beams 3-4 are subjected to maximum stress ${\mathrm{\&sigma;}}_{3-4y\max }={\mathrm{\&sigma;}}_{3\mathrm{ym}}={\mathrm{\&sigma;}}_{4\mathrm{ym}}$ The beams 3-4 being subjected to a torque $M_{\mathrm{Ty}34}=\frac{{\mathrm{Gwh}}^3}{3l}{\mathrm{\&theta;}}_x$ The shear stress caused by the torque is the same on each section of the beam 3-4 $t_{x3}=\frac{M_{\mathrm{Ty}3-4}}{\frac{1}{3}{\mathrm{wh}}^2}=t_{x4}$ The maximum shear force occurs at the beam top midpoint $t_{\max }=\frac{M_{T34}}{0.25{\mathrm{wh}}^2}$ Stress of the beam 1-2: 1, 2 shearing force $Q_{1\mathrm{ym}}=Q_{2\mathrm{ym}}=\frac{{\mathrm{Ewh}}^3}{{2l}^3}(a1+l){\mathrm{\&theta;}}_y$ Bending moment at beam 1 $M_{1\mathrm{ym}}=-\frac{{\mathrm{Ewh}}^3}{{12l}^2}({3a}_1+2l){\mathrm{\&theta;}}_x$ Bending moment at beam 2 $M_{2\mathrm{ym}}=-\frac{{\mathrm{Ewh}}^3}{{12l}^2}({3a}_1+4l){\mathrm{\&theta;}}_x$ Maximum tensile stress of upper surface of beam 1 ${\mathrm{\&sigma;}}_{1\mathrm{ym}}=\frac{M_{1\mathrm{ym}}}{\frac{{\mathrm{wh}}^2}{6}}$ Where thetax is the corner of the mass, ${\mathrm{\&theta;}}_x=\frac{3m{a}_y{z}_c{l}^3}{{\mathrm{wh}}^3[E(6a_{1}l+3a_1^2+{4l}^2)+3a_2^{2}E+4G{l}^2]}$ C. the structure is subjected to a z-direction acceleration az and is deformed under a force as shown in fig. 19. The stress condition of each beam is the same, only one of the beams is analyzed, and the shearing force of the beams 1-2 and the beams 1 and 2 is analyzed $Q_{1z}=Q_{2z}=\frac{{\mathrm{<figure-callout\; id="8"\; label="ma\; z"\; filenames="0012617200231.png,0012617200272.png"\; state="\{\{state\}\}">ma</figure-callout>}}_z}{8}$ Bending moment at the beam 1 $M_{1z}=\frac{{\mathrm{<figure-callout\; id="16"\; label="ma\; z\; l"\; filenames="0012617200272.png,0012617200273.png"\; state="\{\{state\}\}">ma</figure-callout>}}_{z}l}{16}$ Bending moment at the beam 2 $M_{2z}=\frac{{\mathrm{<figure-callout\; id="16"\; label="ma\; z\; l"\; filenames="0012617200272.png,0012617200273.png"\; state="\{\{state\}\}">ma</figure-callout>}}_{z}l}{16}$

The beam 1-2 is subjected to the maximum stress ${\mathrm{\&sigma;}}_{1-2\mathrm{zma}}={\mathrm{\&sigma;}}_{1z}={\mathrm{\&sigma;}}_{2z}$

Fig. 6, the piezoresistors form a bridge circuit for measuring X, Y, Z in three directions, wherein R1= R2= R3= R4= R5= R6= R7= R8= R12= R13= R, R9= R10= R11=4R and R9, R10, R11, R12 and R13 are fixed resistors on the outer edge of the chip. The bridge is excited with a constant voltage source Ui, Ui =5V, as shown in fig. 7, 8 and 9, and has acceleration bridge circuits measuring X, Y, Z, in three directions, respectively, as follows:

(1) the bridge circuit in the X direction comprises a series loop formed by R7, R8, R12 and R13, wherein a tap E is arranged between R7 and R8, a tap D is arranged between R12 and R13, and taps are arranged between R7 and R12 and between R8 and R13 to serve as a power supply end Ui of an excitation power supply; and the two ends of E and D are X-direction acceleration measurement output ends.

(2) The Y-direction bridge circuit consists of R5, R6, R12 and R13 which form a series loop, wherein a tap C is arranged between R5 and R6, a tap D is arranged between R12 and R13, and taps are arranged between R5 and R12 and between R6 and R13 to serve as a power supply end Ui of an excitation power supply; and the two ends of C and D are Y-direction acceleration measurement output ends.

(3) The bridge circuit in the Z direction is formed by connecting R1, R2, R3 and R4 in series, and then forming a series loop with R11, R10 and R9; wherein, taps A between R4 and R11, taps B between R10 and R9, and taps between R11 and R10 and between R9 and R1 are power supply terminals Ui of the excitation power supply; and the two ends of A and B are Z-direction acceleration measurement output ends. The invention relates to a method for manufacturing and processing a chip of an integrated silicon micro-resistance type acceleration sensor, which comprisesthe following specific processing technologies and steps for the acceleration sensor chip with eight deformation beam structures:

(1) preparing a wafer, namely selecting a special double-parabolic wafer for a P-type (100) micro machine;

(2) first oxidation to form SiO2 protective layer

(3) Photoetching and low-resistance wire patterns;

(4) phosphorus diffusion, low resistance wire;

(5) carrying out secondary oxidation to generate a SiO2 protective layer;

(6) photoetching the front side to form a resistance pattern;

(7) injecting phosphorus ions to form an N-type piezoelectric resistor;

(8) annealing to eliminate the defect and make the crystal obtain conductivity;

(9) oxidizing for the third time to generate a SiO2 protective layer;

(10) back photoetching to form a back mass block moving allowance graph;

(11) EPW anisotropic etching to obtain quality block;

(12) photoetching the front side to form a pattern of the beam;

(13) etching by positive RIE, etching out a beam by a dry method, and etching by reactive ions with vertical incidence of free radicals;

(14) oxidizing for four times to generate a SiO2 protective layer;

(15) photoetching the front side to form a connection window of the low-resistance silicon and the metal layer;

(16) photoetching the front side to form a metal wire layer pattern;

(17) evaporating TiPtAu to form a metal wire layer which can resist EPW corrosive;

(18) back photoetching to etch the figure of the mass block, including convex angle compensation;

(19) EPW anisotropic etching, etching out the shape of the mass. The key process in the specific processing process of the acceleration sensor chip with the eight deformation beam structures is described in detail as follows: EPW anisotropic etching

Single crystal silicon is an anisotropic material and has different physicochemical properties in different directions due to different atomic line densities, surface densities, and bond densities in different crystal directions and planes.

The bulk silicon micromachining technology has two types, isotropic etching and anisotropic etching. The anisotropic etching of silicon means that different crystal planes of silicon have different etching rates. Anisotropic etching has a smaller lateral etching rate than isotropic etching, so that it is more frequently used in processing a fine structure. The anisotropic etch rate is related to the crystal orientation. Common anisotropic etching solutions are the KOH system and the EPW system. However, the KOH system is not microelectronic compatible. The accelerometer mass is obtained from EPW anisotropic etching liquid. EPW is a mixture of ethylenediamine, catechol, and water. EPW the etching rate of the etching solution is related to the proportion of the components, the etching temperature and the doping concentration of the silicon substrate. The anisotropic etch can limit the etch facets, with the 111 facets etching much slower than the 100 facets in the EPW system. The etched V-shaped cavity on the (100) substrate has the {111} crystal plane as the side surface of the mask window along the<110>direction, and the included angle theta between {111} and {100} is = 54.7 deg. Etching of the<100>substrate with EPW resulted in a<110>oriented cavity with<111>walls for any closed pattern window over time. Along the<110>square aperture at the mask window. The side surface of the etched cavity when the mask window on the (100) substrate is along the<110>direction is the {111} crystal plane, and the included angle theta between {111} and {100} is = 54.7 deg. As shown in fig. 20 and 21.

EPW the reaction equation for etching silicon can be expressed as: wb is the width of the top of the cavity, ws is the width of the bottom of the cavity, and h is the depth of the etched cavity. $W_s=W_b-\sqrt{2}h$

(II) lobe Compensation

Anisotropic etching does not etch the reentrant corners, so the cavity boundaries are very clean. While anisotropic etching tends to etch the lobes, which have an etch rate that is related to the type of etchant and the exposed area. When an orderly mesa structure is desired in the manufacture of an acceleration sensor, in practice, when a (100) substrate is etched by using an EPW etching solution, a mask is rectangular, and is dodecagonal after short-time etching and octagonal after long-time etching. The lateral corrosion of the convex angle of the mesa is related to factors such as corrosion depth, proportion type of the corrosive agent and the like. The corrosion experiment and measurement show that the inclined plane is the {212} plane and the intersection line of the {212} plane and the {100} plane is<121>for EPW 'B' etchant. Therefore, a method of lobe compensation must be used in the fabrication of such devices. There are two methods of compensation: 1. adding a compensation strip at the top angle of the mask; 2. the mask is offset at an angle relative to the<110>direction. The first method is used in the design. There are several methods for adding the compensation strips, and the design adopts a scheme of adding squares, as shown in fig. 22, 23 and 24. EPW 'B' corrosive agent (ethylene diamine, pyrocatechol and water in the molar ratio of 43.8 to 4.2 to 52%) is used for corrosion, the corrosion depth is H at 115 ℃, the vertical distance from the intersection point of two inclined planes on a (100) plane to the top angle of the mask at a convex angle is dc, and the distance from the edge of the mask to the top end of a contraction point is d. The geometry of d, dc and w conforms to the following equation:

dc=dsinα+dcosα(d≤0．5w)

dc = 0.5 wcos α + dsin α (d ≧ 0.5 w) dc, w satisfying the following relationship with etch depth H:

H=(0．5wsinα+dcosα)／UC(d≤0．5w)

h = (0.5 cos α + dsin α)/Uc (d ≧ 0.5 w) where Uc is the normalized value in the experiment, Uc = dc/H = 0.42, which is related to the type of etchant, composition ratio and temperature α is the angle between the intersection of the {212} plane and the {100} plane and the<110>direction, α = 18.5 o when d = 0.5 w, H =340 μm, w = 225.6 μm, d = 112.8 μm, fig. 22, fig. 23 and fig. 24 of the mask can be found.

(III) RIE Dry etching

Although widely used in micromachining, wet etching has many disadvantages, such as the use of large amounts of toxic chemicals, low efficiency; the lateral corrosion effect exists, and the high-precision size is difficult to obtain; it is difficult to corrode some materials (such as Si3N4, SiO2, etc.) and refractory metals. Some dry etching techniques have recently been developed. RIE is one of them. Rie (reactive Ion etching), also known as Reactive Sputter Etching (RSE), is a type of plasma etching that is commonly referred to as radical normal incidence reactive Ion etching. Plasma etching has three distinct advantages: the shape of the etched surface is easy to control, and no residual etched surface exists; the back of the second etching does not need to be protected; the three etching processes are compatible with the CMOS process; and fourthly, a larger depth-to-width ratio can be obtained. When the RIE method is used for dry etching, a bias voltage is applied to the anode, so that ions for conveying active particles are accelerated to vertically irradiate the surface of the wafer, and the active particles and an etching material perform chemical reaction to form etching. Due to the directional incidence, the etching is isotropic, which is very advantageous for the fabrication of high resolution patterns.

(IV) annealing

Accompanying the ion implantation process, incident ions have a series of defects around their passing, even forming damaged regions, and implanted impurity atoms tend to be in interstitial positions, generally not providing conductivity. Semiconductor annealing technology is to utilize the thermal effect produced by various energy forms to eliminate various defects and internal stress produced in the course of ion implantation or to make dopant atoms to be electrically activated effectively. A common annealing method is a thermal annealing technique.

(V) layout design

A P-type wafer is used as a substrate, and phosphorus is implanted to serve as an N-type voltage resistor.

The resistance layer is obtained by ion implantation, and the square resistance is about 400 omega/port. The voltage resistor and the metal layer are connected by low-resistance silicon which is obtained by a diffusion method. The low resistance silicon has a sheet resistance of about 2 omega/port and a resistance of about 30 omega per beam. The low resistance silicon to metal connections are windowed. The bulk silicon corrosion of the proof mass is compensated by a convex angle.

The accelerometer needs 7 photo-etching masks. The mass block and the convex angle compensation plate are respectively one plate, wherein the resistance layer plate, the diffusion layer plate, the metal layer plate, the beam plate and the plate are provided with a movable space between the mass block and the base on the back surface. Five pieces on the front surface and two pieces on the back surface.

Fig. 25 is a photograph showing a real object of the result of processing the three-dimensional acceleration sensor chip.

The miniature three-dimensional acceleration integrated silicon micro-resistance type acceleration sensor provided by the invention is reliably used for the collision experiment of the automobile.

Claims (7)

1. An integrated silicon micro-resistance type acceleration sensor comprises a base (1), a shell (2) which is arranged on the base and plays a role in protection, and a detection chip (3) which is composed of a stress sensitive part, a mass block part, a piezoresistor and a bridge circuit; the detection chip is characterized in that the center of the detection chip is provided with a micro-motion mass block (4), two, four or eight beams are symmetrically distributed around the mass block and used for supporting the mass block, and the sensitive part is a beam (5).

2. The integrated silicon micro-piezoresistive acceleration sensor according to claim 1, wherein the mass and the strain beam are formed by a single-crystal silicon wafer formed by micromachining.

3. The integrated silicon micro-resistor type acceleration transducer of claim 2, wherein the mass is a regular diamond platform with four side faces and a bottom face having an included angle of α, the outer peripheral edge of the supporting strain beam is trapezoidal, and the inner side face and the bottom face have an included angle of β.

4. An integrated silicon micro-resistor type acceleration sensor according to claim 3, characterized by that the piezoresistors (6) are resistor structures implanted with phosphorus ions onto the silicon wafer, at the maximum stress point on each beam.

5. An integrated silicon micro-resistor type acceleration sensor according to claims 1 to 4, characterized by that the piezoresistors constitute respectively a bridge circuit for measuring X, Y, Z acceleration in three directions: r1= R2= R3= R4= R5= R6= R7= R8= R12= R13= R, R9= R10= R11=4R and R9, R10, R11, R12, R13 are fixed resistances on the outer edge of the chip.

6. The manufacturing and processing method of the chip of the integrated silicon micro-resistance type acceleration sensor according to claim 1, characterized in that: (1) selecting materials, namely taking a double-polished silicon wafer special for a micro machine; (2) adopting a SiO2 protective layer; (3) photoetching, namely, conducting wires, resistors, mass blocks and diamond-shaped tables; (4) manufacturing a low-resistance wire by a phosphorus diffusion method; (5) injecting phosphorus ions to form an N-type piezoresistor; (6) annealing (silicon wafer); (7) EPW anisotropically etching to obtain quality blocks and diamond-shaped platforms; (8) RIE etching, dry etching out the strain beam.

7. The method for manufacturing and processing the chip of the integrated silicon micro-resistance type acceleration sensor according to claim 6, characterized in that the specific process steps are as follows: (1) preparing a wafer, namely selecting a special double-parabolic wafer for a P-type (100) micro machine; (2) carrying out primary oxidation to generate a SiO2 protective layer (3), photoetching and low-resistance wire patterns; (4) phosphorus diffusion, low resistance wire; (5) carrying out secondary oxidation to generate a SiO2 protective layer; (6) photoetching the front side to form a resistance pattern; (7) injecting phosphorus ions to form an N-type piezoresistor; (8) annealing to eliminate the defect and make the crystal obtain conductivity; (9) oxidizing for the third time to generate a SiO2 protective layer; (10) back photoetching to form a back mass block moving allowance graph; (11) EPW anisotropic etching to obtain quality block; (12) photoetching the front side to form a pattern of the beam; (13) etching by positive RIE, etching out a beam by a dry method, and etching by reactive ions with vertical incidence of free radicals; (14) oxidizing for four times to generate a SiO2 protective layer; (15) photoetching the front side to form a connection window of the low-resistance silicon and the metal layer; (16) photoetching the front side to form a metal wire layer pattern; (17) evaporating TiPtAu to form a metal wire layer which can resist EPW corrosive; (18) back photoetching to etch the figure of the mass block, including convex angle compensation; (19) EPW anisotropic etching, etching out the shape of the proof mass.

Images