Micromechanical structure

Details Micromechanical structure Micromechanical structure

2001-09-28

2003-05-06

Moller, Richard A. (Department: 2856)

Measuring and testing

Speed, velocity, or acceleration

C310S311000

Reexamination Certificate

active

06557413

ABSTRACT:

TECHNICAL FIELD OF THE INVENTION
The invention relates to a micromechanical structures that include movable elements. In particular the invention relates to an arrangement for coupling such movable elements to other structures of a microelectromechanical system (MEMS).
BACKGROUND ART OF THE INVENTION
In microelectronics the trend has been towards a higher level of integration. The same applies to micromechanics as well. Consequently, micromechanical elements designated especially for microelectronic purposes need to be more highly integrated because of the requirement for smaller and smaller components for electrical applications.
Prior art micromechanical components have been optimized for low frequency (<1 MHz) applications and used mainly for inertial and pressure sensors. The design of micromechanical RF components for 1 to 5 GHz applications used in mobile terminals sets demands on micromachined structures. These demands are partly different from the problems in the low frequency Micro Electromechanical Systems (MEMS) applications.
The optimization of the capacitive micromechanical structures is subject to several parameters:
Sensitivity to the measured value or control force (e.g., acceleration to capacitance transfer function, control voltage to capacitance transfer function),
Signal to noise ratio that depends on the several other device parameters,
Zero point stability of the device with respect to long time periods and temperature.
These optimization criteria convert into more specific device requirements when the application and especially the measurement or operation frequency is taken into account. This invention is related to the use of the micromechanical structure as a part of the high frequency application. Two different examples of such an application are:
MEMS rf components: tunable capacitors and micromechanical microrelays;
Micromechanical low noise, high sensitivity accelerometer using LC resonance as a basis of the measurement electronics;
For both these applications, there are several common requirements for the device:
Series resistance of the device must be minimized;
Series (stray) inductance of the device must minimized and repeatable;
Temperature dependence of the structure must be as small as possible; and
Parasitic capacitance must be minimized.
The prior art micromechanical structures are mostly based on silicon and polysilicon structures. The polysilicon has good mechanical properties and technology to build suspended structures from it is well researched. However, the main disadvantage of these structures is the high series resistance. The series resistance reduces the Q value of the component at high frequencies.
Many devices like the low-noise rf voltage controlled oscillators (VCO) require a resonant device with high Q-factor, since the phase noise of an oscillator is proportional to 1Q
T
2
, where Q
T
is the overall Q-factor of the resonator. High dynamic range filters also require a high Q-resonator, since the dynamic range of the filter is proportional to Q
T
2
. The quality factor within the frequency range 1 to 2 GHz is dominated by the series resistance. Previously, for instance the MEMS tunable capacitors were fabricated from the polysilicon, but the requirement for the low series resistance has forced to consider metal as the material of the structure. Metal can be for instance gold, copper, silver, nickel, aluminum, chromium, refractory metal or alloy of several metals.
In capacitive sensors the ultimate resolution of the capacitance measurement is limited by the series and/or parallel resistances of the sensing capacitance. Most of the prior art capacitive inertial sensors are made of doped monocrystalline or polycrystalline silicon, and the conductivity is limitted to relatively modest values. Furthermore, the additional series resistance due to metal/silicon interfaces increases the series resistance. The inertial sensors based on metal structures have been studied, [1] and [2], because of two clear advantages: 1) metals have higher material density that increases the mass and thus the sensitivity of the capacitive sensor, and 2) metals have higher electrical conductivity that reduces the electrical noise of the capacitive sensor. One of the key problem in using metallic materials for inertial sensors has been the elimination of thermal stress caused by the mismatch of the thermal expansion coefficients between the substrate and the structure.
Metal thus has some disadvantageous characteristics like the built-in stress that can cause warping of the suspended structures. In addition, most metals that are available in the MEMS processes have the thermal expansion coefficient that is very different compared to the thermal expansion coefficient of most substrate materials such as silicon, quartz or borosilicate glass. Thermal stress of the suspended structure, due to the thermal expansion mismatch, can cause severe thermal dependence in the device.
FIG. 1
shows a typical micromechanical bridge. The requirement is to make a mechanically ideal anchor using a minimum of process steps. A simple process is advantageous in the method shown in FIG.
1
. One disadvantage of such metal structure is that the built-in stress and any temperature dependent stress tend to bend the suspended structure.
FIG. 1
illustrates the situation when the micromechanical metal structure with a movable element
110
and anchors
130
,
132
is deposited on top of the silicon substrate
150
. The
FIG. 1
also shows the insulating layer
160
and the fixed electrodes
140
,
142
on the substrate. The change of the internal stress of the metal
110
due to the change in temperature can be calculated as
&Dgr;&sgr;=
E
·(&agr;
2
−&agr;
1
)·&Dgr;
T
  (1)
where E is the Young's modulus, &agr;
1
, and &agr;
2
are the thermal expansion coefficients of the metal film and the silicon substrate, respectively, and &Dgr;T is the temperature change.
For the copper film on top of the silicon substrate,
∂
σ
∂
T
=
2
⁢
 
⁢
MPa
°
⁢
 
⁢
C
.
.
(
2
)
The stress in the metal causes a force F
eff
to the anchoring structures
130
and
132
.
FIG. 2
shows the moment effect at the step-up anchor structure. We suppose that the suspended structure is connected to the substrate from several points and that the thermal expansion mismatch between the substrate and the suspended structure causes strain in the suspended structure. Effect of the strain is shown as two arrows in FIG.
2
. Figure shows how the moment caused by the step-up anchor bends (exaggerated) the suspended structure. Normal dimensions for the suspended structure might be for instance that the suspended structure is 500 &mgr;m long, 1 &mgr;m thick and it is 1 &mgr;m above the substrate. Even a very small bending moment would be catastrophic, since the structure would touch the surface.
FIG. 3
shows how the control voltage is dependent on the residual stress of a copper film double supported beam. The capacitance is kept constant, in this case 0.9 pF. Length of the beam is 0,5 mm, width is 0,2 mm, and thickness is 0.5 &mgr;m. The gap between the control electrode and the beam is 1 &mgr;m. The Figure shows how sensitive the control voltage is to low level residual film stress.
The temperature dependence of the capacitance can be calculated as
∂
C
∂
T
=
∂
C
∂
σ
=
∂
σ
∂
T
(
3
)
The temperature dependence increases with the control voltage. For example, for 5 MPa residual stress, the temperature dependence of the capacitance can be 1%/° C. at 1 V control voltage, and 24%/° C. at 3 V control voltage. If the device is operated at low control voltages, the residual stress of the film must be minimized. At this range, the temperature dependence must be minimized by some structural modifications.
The temperature dependence has been reduced by using flexible spring support for the structure. Such prior art solutions for implementing micromechanical components are d

Affiliated with

Also associated with

Rating

Micromechanical structure does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Micromechanical structure, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Micromechanical structure will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-3002738