WO2013102352A1

WO2013102352A1 - Online test structure for residual stress of polycrystalline silicon material

Info

Publication number: WO2013102352A1
Application number: PCT/CN2012/079133
Authority: WO
Inventors: 李伟华; 张卫青; 蒋明霞; 周再发; 刘海韵
Original assignee: 东南大学
Priority date: 2012-01-06
Filing date: 2012-07-25
Publication date: 2013-07-11
Also published as: CN102565143A; CN102565143B

Abstract

An online test structure for the residual stress of a polycrystalline silicon material, comprising three deflecting pointers of polycrystalline silicon having substantially the same structure, the three deflecting pointers of polycrystalline silicon being arranged in a delta shape with all pointers pointing to the centre. Each deflecting pointer comprises a horizontal driving beam (101, 103, 105), a pointer (102, 104, 106) perpendicular to the driving beam (101, 103, 105), and two anchoring regions fixed on a substrate, the two anchoring regions (107, 108, 109, 110, 111, 112) fixing one end of the driving beam (101, 103, 105) and one end of the pointer (102, 104, 106) respectively. By controlling the initial direction of deflection of the pointer under the action of the residual stress, the maintenance and change in distance can effectively reflect the amount and properties of the residual stress. During testing, heat driving is utilised, and the measured parameters are the electrical resistance of the front and rear driving beams for heat driving. The measurement and calculation do not require a thermal expansion co-efficient, avoiding the effect of errors in the measurement results during online testing of the thermal expansion co-efficient.

Description

Polysilicon material residual stress online test structure

TECHNICAL FIELD The present invention relates to an on-line test structure for residual stress of polysilicon materials, and belongs to the field of on-line testing of microelectromechanical systems (MEMS) material parameters. Background technique

The performance of MEMS devices is closely related to the material parameters. Due to the influence of the processing process, some material parameters will change. These uncertainties caused by the processing technology will make the device design and performance prediction uncertain and unstable. Case. The material parameter online test aims to measure the material parameters of the MEMS device manufactured by the specific process in real time, monitor the stability of the process, and feed back the parameters to the designer to correct the design. Therefore, online testing without the processing environment and using common equipment becomes a necessary means of process monitoring. The online test structure usually uses electrical excitation and electrical measurement methods to obtain physical parameters of the material by electrical quantity values and targeted calculation methods.

Polysilicon is an important and fundamental material in the fabrication of MEMS devices and is typically fabricated using chemical vapor deposition (CVD) methods. The polysilicon material will generate internal stress during the manufacturing process, that is, residual stress. The residual stress is divided into compressive stress and tensile stress. When the MEMS structure is released, the residual stress will cause initial deformation of the structure or an effect on other material parameters, resulting in a deviation of the actual performance from the design performance.

Summary of the invention

OBJECT OF THE INVENTION: To overcome the deficiencies in the prior art, the present invention provides an in-line test structure for residual stress of polysilicon materials.

Technical Solution: In order to achieve the above object, a polysilicon material residual stress on-line test structure includes three polysilicon deflection pointers, and three polysilicon deflection pointers respectively include a polysilicon drive beam, a polysilicon pointer and an anchor region; and three polysilicon deflection pointers In the "product" type placement, the polysilicon pointer points to the center; the lower left polysilicon deflection pointer and the lower right polysilicon deflection pointer structure are identical to test the left and right mirror directions of the vertical center line of the structure, the upper polysilicon deflection pointer is located at the center, the pointer direction is The lower left and right polysilicon deflection pointers are opposite; the entire test structure is fabricated on the insulating substrate, except for the anchor region and the metal electrodes thereon, after the structure is released, the driving beam and the pointer are in suspension to facilitate the release of residual stress and Free expansion and deflection.

The spacing of the left, middle and right pointers is affected by the residual stress of polysilicon. The spacing between the left and middle pointers is not affected by the residual stress of polysilicon. The spacing of the end points of the middle-right pointer varies with the size and properties of the residual stress of polysilicon. . In the lower left polysilicon deflection pointer, a metal electrode is respectively formed on an anchor region at one end of the driving beam and an anchor region at one end of the pointer; in the right lower polysilicon deflection pointer, an anchor region at one end of the driving beam and an anchor region at one end of the pointer are Metal electrodes are respectively fabricated; in the upper polysilicon deflection pointer, metal electrodes are formed only on the anchor region at one end of the pointer.

The polysilicon drive beams of the three polysilicon deflection fingers are of equal length.

[Advantageous Effects] The in-line test structure of the residual stress of the polysilicon material of the present invention is arranged in a "product" shape by three polysilicon deflection pointers having the same basic structure, and the residual stress of the polysilicon residual stress is affected by the same characteristics, so that the residual stress is obtained. The size and nature of the measurement can be effectively measured by heating the deflection beam of the drive beam to expand it, and in turn pushing the pointer to deflect, measuring the effect of residual stress on the amount of deflection. The invention uses the invention to test the residual stress of polysilicon, the method is simple, the test equipment requirements are low, the processing process is synchronized with the micro-electromechanical device, and there is no special processing requirement, which fully meets the requirements of the online test. The calculation method in the present invention is limited to a simple mathematical formula. Although the thermal expansion principle is adopted, the measurement calculation does not require a thermal expansion coefficient, and the influence of the error when the thermal expansion coefficient is tested on the measurement result is avoided, and the test structure is simple, the electrical signal is loaded and The measurement is simple and the calculation method is stable. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic view showing an on-line test structure of residual stress of polysilicon material according to the present invention; FIG.

Figure 2 is a cross-sectional view taken along line A-A of Figure 1.

DETAILED DESCRIPTION OF THE INVENTION The present invention will be further described below in conjunction with the accompanying drawings.

As shown in FIG. 1 and FIG. 2, the polysilicon material residual stress in-line test structure of the present invention comprises three deflection pointers having the same basic structure, each deflection pointer comprising a horizontal drive beam 101, 103, 105, a drive beam 101, 103, 105 vertical pointers 102, 104, 106 and two anchor regions fixed to the substrate, the two anchor regions respectively fixing one end of the drive beams 101, 103, 105 and the pointers 102, 104, 106 One end. The body of the test structure is made of polysilicon material.

The three polysilicon deflection pointers are placed in a "pin" shape, and the pointers 102, 104, 106 are all pointing toward the center; the upper polysilicon deflection pointer includes the drive beam 101, the pointer 102, the anchor regions 107, 108, and the metal electrodes above the anchor region 108. 118; The lower left polysilicon deflection pointer includes a drive beam 103, a pointer 104, anchor regions 109, 110 and metal electrodes 113, 114 on the anchor region; the lower right polysilicon deflection pointer includes a drive beam 105, a pointer 106, an anchor region 111, 112 and metal electrodes 115, 116 on the anchor region. The polysilicon deflection pointers of the lower left and lower right portions are identical, placed symmetrically with respect to the centerline of the pointer 102, and the centerline of the pointer 102 is the vertical centerline of the entire test structure. Entire test structure Fabricated on the insulating substrate 117, in addition to the anchor region and the metal electrodes thereon, after the structure is released, the drive beams 101, 103, 105 and the pointers 102, 104, 106 are in a suspended state, so that these portions are released. Stress and freedom to stretch and deflect.

The lengths of the drive beams 101, 103, 105 are both L ₂ , and after the structure is released, the beam will initially expand and contract due to residual stress and thereby push the pointers 102, 104 and 106 to deflect about their rotational axes, because the basic structure is the same, The deflection angle produced by the residual stress is the same. Drive beam 101, 103, 105 in the vertical direction of the center line to respective anchor region 108, the distance 110, 111 is L _5.

The pointers 104, 106 are all L _{1 in} length _{; the} length of the pointer 102 is equal to L _{1 +} L ₄ , and L ₄ is the overlap length of the pointer 102 and the pointers 104, 106 in the vertical direction, which is much smaller. The head of the pointer 102 is a rectangle having a large width, and the purpose of increasing the width is to maintain a large distance between the anchor regions 110 and 111, which facilitates the use of the test probe.

The design spacing of the pointers 102 and 104 is _gi and the design spacing of the pointers 102 and 106 is g ₂ .

Metal electrodes 118, 113, 114, 115, and 116 are fabricated on the anchor regions 108, 109, 110, 111, and 112, respectively, wherein the metal electrode 114 extends all the way to the drive beam 103, and the metal electrode 115 extends straight to the drive beam On the 105, when the heating drive is operated, the effective heat generating region length is L _{2 -} L ₃ .

The polysilicon material residual stress online test structure of the present invention has the following principle:

Because the residual stresses present in the polysilicon will cause the drive beams 101, 103 to undergo an initial length change after the structure is released, thereby causing the pointers 102, 104 to initially deflect. However, since the pointers 102 and 104 are deflected in opposite directions of rotation and the deflection angle values are the same under the residual stress of the same nature and size, the pitch of the end of the pointer can be kept constant, still g-like, The residual stress will cause the drive beam 105 to produce an initial length change after the structure is released, thereby causing the pointer 106 to deflect initially, the direction of deflection being opposite the pointer 104. Since the direction of rotation of the pointer 102 and the pointer 106 is the same, the actual spacing is deviated from the design value g ₂ due to residual stress. If the residual stress is compressive stress, the actual spacing is less than g ₂ , and if the residual stress is tensile stress, the actual The spacing is greater than g ₂ .

The invention employs a thermally driven deflection pointer rotation mode of operation. Current is applied between the metal electrode 113 and metal electrode 114, the drive beam 103 in section L ₂ -L ₃ thermally expand, clockwise deflection of the push pointer 104, the pointer 104 when the pointer 102 comes into contact with the tip of the pointer tip 104 cis The hour hand is deflected by the distance _gi . Is applied between the metal electrode 115 and metal electrode 116 current, the drive beam 105 in the portion L ₂ -L ₃ thermal expansion of the push pointer counter 106 is deflected assumed that when the tip of the pointer and the pointer 102 106 comes into contact, the pointer tip 106 The actual distance deflected counterclockwise is g. Obviously, the distance g of the tip deflection of 106 will not be equal to g ₂ due to the residual stress. If g is equal to g ₂ , it means that the residual stress is zero. The test structure of the present invention is completed using a basic microelectromechanical processing process. The fabrication process of the test structure is illustrated below by a typical two-layer polysilicon microelectromechanical surface processing process.

An N-type semiconductor wafer is selected, a silicon dioxide layer having a thickness of 100 nm is thermally grown, and a 500 nm-thick silicon nitride is deposited by a low pressure chemical vapor deposition process to form an insulating substrate 117. A layer of 300 nm polysilicon is deposited by a low pressure chemical vapor deposition process and N-type heavily doped to make the layer of polysilicon a conductor, which is etched by photolithography to form part of the anchor region. A 2000 nm thick phosphosilicate glass (PSG) was deposited using a low pressure chemical vapor deposition process to form a pattern of anchor regions by photolithography. A layer of 2000 nm thick polysilicon is deposited by a low pressure chemical vapor deposition process, and the polysilicon is doped with N-type heavily. The photolithography process forms a polysilicon material residual stress on-line test structure pattern, and the thickness of the anchor region is the sum of the thicknesses of the two polysilicon layers. . A metal electrode pattern is formed on the anchor region by a lift-off process. Finally, the structure is released by etching the phosphosilicate glass.

The test method of the invention is simple, using a simple variable current source as an excitation source, and an ordinary multimeter is used to monitor whether two hands are in contact, and the resistance is measured by an electric resistance meter. The specific process is as follows:

1), the testing process

The testing process is carried out in several stages:

1 The electrical resistance between the metal electrode 113 and the metal electrode 114 (or the metal electrode 115, the metal electrode 116) is measured at room temperature, and is denoted as W _{∞ ;}

2 A slowly increasing current is applied between the metal electrode 113 and the metal electrode 114, and the resistance between the metal electrode 118 and the metal electrode 114 is monitored. When the resistance changes from infinity to a finite value, it indicates that the pointer 104 and the pointer 102 have occurred. Contact, stop the increase of heating current;

3 measuring the resistance between the metal electrode 113 and the metal electrode 114 when the pointer 102 and the pointer 104 are in contact, denoted as R _TL , closing the current between the metal electrode 113 and the metal electrode 114, causing the pointer 104 to rotate away from the pointer 102;

4 A slowly increasing current is applied between the metal electrode 115 and the metal electrode 116, and the resistance between the metal electrode 108 and the metal electrode 115 is monitored. When the resistance changes from infinity to a finite value, it indicates that the pointer 106 and the pointer 102 have occurred. Contact, stop the increase of heating current;

5 measuring the resistance between the metal electrode 115 and the metal electrode 116 when the pointer 106 and the pointer 102 contact, denoted as R _TR , closing the current between the metal electrode 113 and the metal electrode 114, causing the pointer 106 to rotate away from the pointer 102;

2) Calculate the thermal expansion coefficient of polysilicon material

The relationship between the resistance of the length of the L ₂ -L ₃ portion of the polysilicon drive beam 103 and the average temperature change amount ΔΓ _£ thereon is as follows: R _TL = R _∞ {l + a ₁ AT _L + a ₂ AT^)

Where α _Ρ a ₂ is the temperature coefficient of the polysilicon resistor, and the average temperature change amount ΔΓ _£ is the difference between the average temperature of the L ₂ -L ₃ portion of the heat-driven beam 103 and the room temperature when the hands 102 and 104 are in contact.

From the basic thermal expansion relationship, the length of the heat-driven beam 103 varies by AL _£ = (L ₂ - )·α·ΔΓ _£ , where α is the coefficient of thermal expansion of the polysilicon material, so:

_ AL _L

a~ (L ₂ -L ₃ )-AT _L

Similarly, the relationship between the resistance of the length of the L ₂ -L ₃ portion of the polysilicon drive beam 105 and the average temperature change ΔΓ thereof is as follows:

R _TR ^R _∞ (l+ _ai AT _R +a ₂ AT^)

Where ΔΓ is the difference between the average temperature of the L ₂ -L ₃ portion of the heat-driven beam 105 and the room temperature when the pointer 102 and the pointer 106 come into contact. And have:

a =

(L ₂ -L ₃ )-AT _R solution:

It has been found that the temperature coefficient α _Ρ α _{2 of the} polysilicon resistor can be obtained by measurement, and therefore, _fll and ^ are treated as known amounts.

The root formula of the test is obtained:

AT _r =

When the polysilicon resistor is a negative temperature coefficient, the root number is preceded by a "-" sign; when the polysilicon resistor is a positive temperature coefficient, the root number is preceded by a "+" sign;

The change in length of the drive beam 103, AL _£, is obtained from the geometric relationship:

AL _L =^ - Similarly, the measured R _∞ and RTM are substituted into the resistance formula, which is obtained from the root formula of the quadratic equation:

One (^士^^ + 4a ₂ k, RT.- R~

ΔΓ = where k _K =

2α ₂ R _∞

The length of the drive beam 105 varies from the geometric relationship:

AL _R = -^, where g is the distance that the tip of the pointer 106 actually deflects counterclockwise. From the relationship of thermal expansion coefficient, we get:

So there are:

L, - ΔΓ„ · AL,

s =— L

L ₅ - AT _L

The variable in the formula is either a geometric dimension or a calculated value that can be obtained. Therefore, the distance g of the tip of the pointer 106 that is actually deflected counterclockwise can be calculated by the above equation.

From the value of g, AL _s can be obtained. Therefore, the strain f generated by the residual stress in the drive beam is:

AL, - AL _R

2. L ₂

The residual stress σ is:

σ = Ε ε

Tantalum is the Young's modulus of polysilicon material.

If s = 0, it means there is no residual stress in polysilicon; if s > 0, it means that the residual stress of polysilicon is compressive stress; if s < 0, it means that the residual stress of polysilicon is tensile stress.

The invention can effectively reflect the magnitude and nature of the residual stress by controlling the initial deflection direction under the residual stress by the pointer. The test structure is simple in manufacturing process and has no special processing requirements; The measurement parameter is the resistance of the drive beam before and after the heat drive. In the process of using the invention, although the principle of thermal expansion is adopted, the thermal expansion coefficient is not required for the measurement calculation, and the influence of the error when the thermal expansion coefficient is tested on the measurement result is avoided.

The above description is only a preferred embodiment of the present invention, and it should be noted that those skilled in the art can also make several improvements and retouchings without departing from the principles of the present invention. It should be considered as the scope of protection of the present invention.

Claims

Claim

1. A polysilicon material residual stress on-line test structure, comprising: three polysilicon deflection pointers, three polysilicon deflection pointers respectively comprising polysilicon drive beams (101, 103, 105) and polysilicon hands (102, 104, 106) And the anchor zone; the three polysilicon deflection pointers are placed in a "good" shape, the pointers (102, 104, 106) are all pointing to the center; the lower left polysilicon deflection pointer and the lower right polysilicon deflection pointer are identical in structure to test the vertical center of the structure The left and right mirror directions, the upper polysilicon deflection pointer is located at the center, and the pointer (102) direction is opposite to the direction of the lower left and right polysilicon deflection pointers (104, 106); the entire test structure is fabricated on the insulating substrate (117), in the structure After being released, the drive beam (101, 103, 105) and the pointers (102, 104, 106) are all in suspension to facilitate release of residual stress and free expansion and contraction and deflection.

2. The polysilicon material residual stress in-line test structure according to claim 1, wherein: in the left lower polysilicon deflection pointer, the anchor beam (103) at the end of the driving beam (103) and the anchor (104) are anchored at the end Area

(110) is respectively fabricated with metal electrodes (113, 114), and the metal electrode (114) extends all the way to the driving beam (103); in the right lower polysilicon deflection pointer, the anchor beam (105) is anchored at the end ( 112) A metal electrode (116, 115) is formed on the anchor region (111) at one end of the pointer (106), and the metal electrode (115) extends all the way to the driving beam (105).

3. The polysilicon material residual stress in-line test structure according to claim 1, wherein: the driving beams (101, 103, 105) of the three polysilicon deflection pointers are of equal length.