See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/3330333

Thermal Isolation of Encapsulated MEMS Resonators

Article  in  Journal of Microelectromechanical Systems · March 2008

DOI: 10.1109/JMEMS.2007.904332 · Source: IEEE Xplore

CITATIONS READS
48 245

9 authors, including:

Matthew A Hopcroft Saurabh Chandorkar

Red Dog Research Consulting Stanford University
91 PUBLICATIONS   3,054 CITATIONS    68 PUBLICATIONS   1,623 CITATIONS   

SEE PROFILE SEE PROFILE

Rob Candler
University of California, Los Angeles
94 PUBLICATIONS   2,154 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

DARPA MTO HERMIT Program View project

Haptic Feedback for Robotic Surgery View project

All content following this page was uploaded by Rob Candler on 17 March 2014.

The user has requested enhancement of the downloaded file.

JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 17, NO. 1, FEBRUARY 2008 175

Thermal Isolation of Encapsulated

MEMS Resonators
Chandra Mohan Jha, Student Member, ASME, Matthew A. Hopcroft, Saurabh A. Chandorkar,
James C. Salvia, Student Member, IEEE, Manu Agarwal, Member, IEEE, Rob N. Candler, Member, IEEE,
Renata Melamud, Bongsang Kim, and Thomas W. Kenny

Abstract—This paper presents an in-chip thermal-isolation Re Electrical resistance, Ω.

technique for a micro-ovenized microelectromechanical-system Rth Thermal resistance, K W−1 .
resonator. Resonators with a microoven can be used for high- C Specific heat per unit volume at constant volume,
precision feedback control of temperature to compensate for
the temperature dependence of resonator frequency over a wide J m−3 K−1 .
temperature range. However, ovenization requires power con- ν Velocity of the energy carrier, m s−1 .
sumption for heating, and the thermal time constant must be kB Boltzmann constant, 1.381 × 10−23 J K−1 .
minimized for effective temperature control. This paper demon- nm Molecular number density, m−3 .
strates an efficient local-thermal-isolation mechanism, which can m Molecular mass of the energy carrier, kg.
reduce the power requirement to a few milliwatts and the ther-
mal time constant to a few milliseconds. In this method, the T Temperature, K.
mechanical suspension of the resonator is modified to provide V Volume, m3 .
thermal isolation and include an integrated resistive heater. This V Voltage, V.
combination provides mechanical suspension, electrical heating, Cchip Thermal capacitance of chip, J K−1 .
and thermal isolation in a compact structure that requires low Cresonator Thermal capacitance of resonator, J K−1 .
heating power and has a small thermal time constant. A power
consumption of approximately 12 mW for a 125 ◦ C tempera- Ac Cross-sectional area, m2 .
ture rise and a thermal time constant ranging from 7 to 10 ms As Surface area, m2 .
is reported in this paper, which is orders of magnitude lower q Rate of joule heat per unit volume, W m−3 .
than that of commercially available ovenized quartz resonators. q Rate of total heat generated, W.
A CMOS-compatible wafer-scale encapsulation process is used to l Length, m.
fabricate this device, and the thermal-isolation design is achieved
without any modification to the existing resonator fabrication E Modulus of elasticity, N m−2 .
process. [2007-0042] I Area moment of inertia, m4 .
P Axial force, N.
Index Terms—Encapsulation, microelectromechanical devices,
microresonators, thermal isolation. ml Mass per unit length, kg m−1 .
f Frequency, Hz.

N OMENCLATURE
I. I NTRODUCTION
k Thermal conductivity, W m−1 K−1 .
Surface emissivity, 0 ≤ ε ≤ 1.
ε
σ
ρ
Stefan–Boltzmann constant,5.67×10−8W m−2 K−4.
Density, kg m−3 .
T HE CMOS-compatible microelectromechanical-system
(MEMS) resonators are becoming an interesting and vi-
able technology [1]–[7] as a replacement for quartz crystals
cp Specific heat capacity at constant pressure, [8]–[13] for timing- and frequency-reference applications. Cur-
J kg−1 K−1 . rently, oven-controlled quartz resonators are used to generate
high-precision frequency references suitable for high-end in-
Manuscript received February 21, 2007; revised June 10, 2007. This work
dustry and military standards [10]–[13]. In this method, the
was supported in part by DARPA HERMIT under Grant ONRN66001-03- resonator is held at a fixed temperature to compensate for the
1-8942, in part by the Robert Bosch Corporation, Palo Alto Research and temperature dependence of the resonator frequency. The extent
Technology Center, in part by a CIS Seed Grant, in part by The National
Nanofabrication Users Network facilities funded by the National Science
to which the resonator is heated depends on the difference
Foundation under Award ECS-9731294, and in part by The National Science between the set point and the ambient temperature. For an
Foundation Instrumentation for Materials Research Program under Grant DMR ovenized resonator that is required to operate within a tem-
9504099. Subject Editor D. DeVoe.
C. M. Jha, M. A. Hopcroft, S. A. Chandorkar, J. C. Salvia, M. Agarwal, perature range of −40 ◦ C to 85 ◦ C, the heating has to cover
R. Melamud, B. Kim, and T. W. Kenny are with the Department of Mechanical a range of 125 ◦ C. Due to the large volume of a conventional
and Electrical Engineering, Stanford University, Stanford, CA 94305 USA quartz-crystal oscillators, which can be up to 1000 mm3 [14],
(e-mail: [email protected]).
R. N. Candler was with Stanford University, Stanford, CA 94305 USA. He is the power consumption for heating can be as much as 10 W
now with the Robert Bosch Corporation Research and Technology Center, Palo with a warm-up time of approximately 30 min [13]–[15].
Alto, CA 94304 USA. MEMS technology offers miniaturization to submillime-
Color versions of one or more of the figures in this paper are available online
at http://ieeexplore.ieee.org. ter scales, which can provide substantial power reduction
Digital Object Identifier 10.1109/JMEMS.2007.904332 [1], [16]–[18].

1057-7157/$25.00 © 2008 IEEE

176 JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 17, NO. 1, FEBRUARY 2008

A microoven, in general, constitutes a heater for joule heating

and thermal isolation for reducing heat loss. Microoven designs
for MEMS devices have been reported with few milliwatts
of power consumption and a time constant in milliseconds
[16]–[18]. In these designs, a MEMS structure is suspended on
a microplatform. The microplatform is thermally isolated from
the substrate and contains separate heater and sense resistors.
Although promising, these designs have limitations in terms of
lack of mechanical stiffness, large thermal mass, and complex
fabrication processes.
This paper describes a miniature thermal-isolation design,
which achieves a small thermal time constant with low power
consumption. The designs illustrated here are compatible with
our previously published “epi-seal” wafer-scale encapsulation
process [19], [20]; and all its advantages, such as low leak
rate, no requirement for a getter, long-term stability [21], and
low-cost manufacturing, are maintained. Designs for thermal
isolation with various heating methods will be discussed and
compared, before presenting the experimental results of the
miniature local-thermal-isolation design.

II. D ESIGNS FOR T HERMAL I SOLATION

A. Heating Entire Chip for Temperature Control
Thermal isolation of the ovenized device from the surround-
ings is required to prevent heat loss during temperature control.
It is therefore essential to thermally isolate the heater from the
ambient but, at the same time, have minimum heat loss between
the heater and the resonator. Hence, it is highly desirable to
place the heater as close to resonator as possible. An added
constraint is to achieve this thermal isolation without modifying Fig. 1. (a) Schematic of a typical MEMS resonator chip attached to a package
the existing fabrication process of the resonator. with adhesive. (b) Isometric view of the device layer of the chip which contains
A double-ended tuning-fork (DETF)-type resonator, encap- resonant structure with input–output electrodes and a resistive heater for joule
heating. The thickness of the device layer is 20 µm.
sulated within a silicon die (chip), is used (Fig. 1). The chip
is attached to the package using an adhesive and wire-bonded
to make electrical contacts [Fig. 1(a)]. Heating resistors are
placed in the device layer [Fig. 1(b)] in the vicinity of the
resonant structure. The DETF is designed for flexural-mode
actuation, as shown in Fig. 2. The biased resonator beams are
electrostatically actuated by providing an alternating stimulus
signal to input electrode. The capacitive transduction between
the beams and the input electrodes cause the resonator to
vibrate. A resonance occurs when the frequency of the input
stimulus signal becomes equal to the natural frequency of the
flexural mode of the beam. The output signal is then amplified
to measure the resonant frequency.
The power consumption and the thermal time constant can
be estimated by using a simple 1-D lumped-parameter thermal
model [22], [23]. The expression for conductive thermal resis-
tance is given by

l
Rth = . (1)
kAc

Thermal capacitance can be evaluated as

Fig. 2. (a) Top view of DETF-type resonant structure. (b) FEM simulation of
Cth = ρV  cp . (2) flexural-vibration mode of a DETF (exaggerated view).
JHA et al.: THERMAL ISOLATION OF ENCAPSULATED MEMS RESONATORS 177

Fig. 3. Thermal equivalent circuit. Package is assumed to be at ambient Fig. 4. Thermal equivalent circuit when there is radiative heat loss from the
temperature. Unit of thermal resistances shown above is in Kelvin per watt. bottom of the chip Rradbot in the absence of the adhesive. Unit of thermal
resistances shown above is in Kelvin per watt.
Radiative and convective thermal loss can be found by
1 is shown in Fig. 4, and the total effective thermal isolation is
Rrad =
hrad As estimated to be approximately 3000 K/W—an improvement by
1 a factor of five.
Rconv = (3) This method of improving thermal resistance is effective but
hconv As
not robust and may lead to packaging problems. Furthermore,
where hrad = εσ(Ts + Tsurr )(Ts2 + Tsurr 2
) is the linearized since the heat transfer from the heater to the resonator takes
radiation-heat-transfer coefficient [22]. Ts is the surface tem- place in the device layer, the device layer gets heated which
perature, and Tsurr is the surrounding temperature. The hconv is leads to unwanted heating up of the entire chip because of
the convective-heat-transfer coefficient. Fig. 3 shows the equiv- the lack of thermal isolation between the device layer and
alent thermal-resistance circuit, where P is the input power. the substrate. The thermal mass of a typical resonator chip is
Approximate values for the material constants are taken from approximately 1200 times larger than that of a single DETF,
the literature [24]–[27]. and hence, heating of the entire chip leads to longer thermal
In this design configuration of the MEMS chip, the ther- time constant and more heat loss resulting in increased input
mal resistance calculated for the device layer and the chip power.
(∼20 K/W) is very small as compared to that of wire bond and
adhesive. It is therefore assumed that the entire chip, including
the resonator, is approximately at a constant temperature. It is B. Heating Resonator Alone With Local Thermal Isolation
also assumed that the package acts as a heat sink, and hence, its
Therefore, it is desired to have an alternative technique to
temperature is the same as the ambient temperature. The total
increase the heating efficiency by heating only the resonator
effective thermal isolation Reff is estimated to be approximately
and simultaneously providing large thermal isolation between
600 K/W.
the resonator and its immediate vicinity in the device layer. This
The power required to achieve a ∆T rise in temperature of
calls for a local heat delivery and thermal-isolation mechanism.
the resonator can be found by
This can be achieved by designing a resonator coupled with
V2 ∆T an in-built heater and restricting the heat loss to the ends of the
P = = . (4) heaters. The in-built heater, in this configuration, serves the dual
Re Reff
purpose of heating, as well as thermally isolating, the DETF.
To obtain ∆T of 125 ◦ C, the input power is approximately To design a resonator for good thermal isolation, it is nec-
200 mW. The thermal time constant can be estimated by (5) essary to study various heat-loss mechanisms in the structure.
and is approximately 500 ms. The resonator is encapsulated, and the atmosphere inside the
encapsulation consists mainly of hydrogen gas at a low pressure
τ = Reff Cth . (5) of approximately 1 Pa [19]. The three modes of heat transfer
considered here are as follows:
One method to improve thermal isolation is to release the chip
from the package by removing the adhesive and keep it floating 1) convection due to hydrogen molecules in the cavity;
in air, thus reducing the heat loss from substrate to the package. 2) radiation from the resonator;
The chip, in this case, is supported from the top by six wire 3) conduction through the silicon beams of the in-built
bonds. The equivalent thermal circuit for the released device heater.
178 JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 17, NO. 1, FEBRUARY 2008

Fig. 5. Temperature profile along the length of a current-carrying resistive heater having thermal resistance of Rth and electrical resistance of Re and its
equivalent lumped model.

Fig. 6. Resonator design with local thermal isolation. The heater is in-built to the DETF, such that, the resonant structure is attached at the center of the heater.
The entire structure is released except at the four anchors.

Convection due to hydrogen molecules can be analyzed by simplification, it is also assumed that the heat loss due to
using microscopic-particle-based kinetic theory of heat diffu- radiation is relatively small and can be neglected.
sion [28]. Thermal conductivity of molecular-energy carriers It is, therefore, the conductive heat transfer through the sili-
can be evaluated as con beams which is the dominant heat-loss mechanism in this
encapsulated MEMS resonator. If a current-carrying resistive
1 heater of constant cross section is analyzed, the temperature
CνΛ
kh = (6)
3 profile along the length of the heater, in the absence of con-

vection and radiation heat loss, is given by (7) [22], [23] and
where C = (3/2)kB nM , and ν = 8kB T /πm. shown in Fig. 5
It is assumed that the mean free path Λ is equal to the
  2 
gap width of the cavity (1.5 µm) because of low pressure. ql 4x2
The thermal conductivity due to the hydrogen molecules kh is Tx = 1− 2 + Ts (7)
8k l
estimated to be around 1 × 10−6 W/m/K at room temperature,
resulting in an effective thermal resistance of the order of where x varies from −l/2 to +l/2, and Ts is the end surface
1 × 107 K/W and, hence, can be assumed that the heat loss temperature of the heater (Fig. 5).
due to molecular conduction is negligible as compared to other The maximum temperature occurs at the center of the heater
modes. (at x = 0) and is given by
The thermal resistance due to radiation for a DETF structure,   2
assumed to be at a maximum heating temperature of 425 K, ql
Tc = + Ts . (8)
is estimated to be approximately 1 × 106 K/W. Hence, for 8k
JHA et al.: THERMAL ISOLATION OF ENCAPSULATED MEMS RESONATORS 179

Fig. 8. Equivalent thermal circuit schematic for the resonator with in-built
heater.

Fig. 7. One-dimensional resistor network to estimate the total effective ther-

mal resistance of the in-built heater. Equation (10) is used for the estimation of
effective thermal resistance Reff .

The expression for the rise in temperature at the center of the

heater, with respect to its end surface, can be written as
 2 
q  l2 V Rth
∆T = Tc − Ts = = = qR (9)
8k Re 8

where q  = q/lA = ((V 2 /Re )/lA).

From (9), we can establish an equivalence between a current-
carrying beam having thermal resistance of Rth and heat gen-
eration of q with a lumped model having thermal resistance of
R and heat flow of q for the same rise in temperature, where R
is eight times smaller than Rth . This suggests that the heating Fig. 9. Finite-element simulation of the thermally isolated DETF resonator
efficiency can be maximized if the resonator is attached at the showing the temperature distribution in Kelvin for a heating voltage of 6 V,
which corresponds to 14 mW of heating power.
center of the heater and that, in order to achieve the maximum
temperature rise at the center of the heater for a given input TABLE I
power, the thermal resistance of the heater should be as large as POWER CONSUMPTION AND TIME-CONSTANT COMPARISON
possible.
However, the micro-ovenized resonator design is suspended
from both ends by in-built resistors (Fig. 6), so the expression
for the temperature of the resonator due to heating power
applied to one end of the resonator must take into account
the heat loss through the other end of the resonator. Using the
equivalent thermal resistance from (9), we can estimate the
resonator temperature by considering the second resistor in
Finite-element simulations (Fig. 9) and measurements indicate
parallel with the heater resistor (Fig. 7). In this case, an ap-
that (10) overestimates the resonator temperature by about 20%.
proximate expression for the rise in temperature at the center of
The layout of the thermally isolated DETF with two in-built
the heater, with respect to its end surface, can be written as
heaters is shown in Figs. 6 and 7. The cross section of each
 2     heater beam is 5 by 20 µm, and its total length is approximately
V 1 V2 Rth
∆T ∼ 1 1 = 2300 µm. The resonator is attached at the center of the heater
Re (Rth /8) + (Rth /4)
Re 12 for maximum heating. The entire structure is released except
= qReff . (10) at the four anchors which act as mechanical supports at the
bottom and provide electrical contacts at the top. The anchors
From (10), for a resonator with a heating resistor with inter- are electrically insulated by silicon dioxide [see Fig. 1(a)]. The
nal heat generation of q at one end and an insulating resistor thermal resistance of the in-built heater Rth is approximately
with no heat generation at the other end, both having thermal 150 000 K/W. Fig. 7 shows an equivalent lumped-capacitance
resistance of Rth , we can establish an equivalent lumped- model of the DETF with in-built heater. The effective thermal
element model with a thermal resistance of Reff and heat flow resistance Reff of the micro-ovenized structure is calculated
of q for the same rise in temperature, where Reff is 12 times using (10) and is evaluated to be approximately 12 500 K/W,
smaller than Rth . This relationship allows us to easily predict which is significantly larger than nonisolated designs.
the resonator temperature for a given heating power, although The thermal isolation of the resonator also leads to a reduced
the prediction is not exact because the temperature distribution thermal time constant, because the effective thermal mass com-
on the heater is modified by the heat flow across the resonator. prises only the mass of the DETF and the in-built resistors,
180 JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 17, NO. 1, FEBRUARY 2008

Fig. 10. (a) Optical image of the top view of the fabricated device before the deposition of the encapsulation layer. (b) SEM cross section of a resonator beam
after the deposition of the encapsulation layer.

Fig. 11. Isometric view of device layer schematic showing the DETF with Fig. 12. Schematic of the test setup for frequency measurement.
the in-built heater. A stimulus signal is applied to the input electrode. Heating
voltages V1 and V2 are controlled using a feedback control loop to maintain a
constant bias for the resonator. A finite-element simulation was done to examine the tem-
perature distribution along the length of the in-built heater
as opposed to the entire silicon die described in Section II-A. and the DETF, as shown in Fig. 9. From the simulation, a
This means that there will be rapid heating and cooling of the power consumption of approximately 12 mW was required
resonator over and above the slower thermal response of the for a temperature increase of 125 ◦ C. The time constant and
chip. A 1-D equivalent thermal-circuit model of the in-built heating power consumption of the three designs are compared
heater is shown in Fig. 8. As with temperature, the thermal in Table I. Clearly, the in-built heater can be very effective, both
time constant of the resonator cannot be calculated precisely in terms of reducing power consumption and dynamic thermal
by a simple lumped-element model because of the complex response.
temperature profile through the device, but we can make an
estimate using the effective resistance of the heaters and the
total thermal capacitance of the heaters and the resonator, as III. F ABRICATION
shown in
One of the biggest advantages of this technique is with
respect to fabrication. A CMOS compatible “epi-seal” encap-
τ = Reff (Cresonator + Cheaters ) (11)
sulation process [19], [20] is used to fabricate this device. The
τchip ≈ Reff _ext Cchip . (12) structure of the fabricated device is shown in Fig. 1(a). The
device layer is insulated from the encapsulation layer by a
For the thermally isolated DETF, τ is estimated to be ap- thin sacrificial oxide, and openings for electrical contacts are
proximately 15 ms. The actual dynamic behavior of the device made through the encapsulation layer over the anchors. Since
is characterized by multiple time constants from the nonheated the in-built heaters are in the device layer with the resonator,
resistor, the silicon die, and the resonator. Finite-element sim- no changes to the fabrication process are required to create
ulations indicate that the temperature change of the resonator the thermally isolated DETF. This process ensures a vacuum
itself may be up to 50% slower than the heaters. inside the encapsulation with a pressure of < 1 Pa. Long term
JHA et al.: THERMAL ISOLATION OF ENCAPSULATED MEMS RESONATORS 181

Fig. 13. Variation of resonator frequency due to joule heating of the in-built heater. The decrease in frequency (right y-axis) corresponds to a temperature rise
(left y-axis) with increasing input power. Experimental results are compared with theoretical estimates. The analytical expression (10) estimates the temperature
at the center of the in-built heater, while the FEM results are for the temperature at the center of the resonator.

(∼1 year) stability of this vacuum condition has also been

verified [21]. An optical image and a SEM cross section of
encapsulated MEMS resonator with in-built heater is shown
in Fig. 10.

IV. E XPERIMENTAL R ESULTS

Resonators with 1.3-MHz frequency and a mechanical Q of
approximately 104 were used for the experiment. Since silicon
resonator frequency varies nearly linearly with temperature
[24], [29], frequency was used as a measure of temperature for
this work. However, resonator frequency is also sensitive to bias
voltage and axial strain.
The bias voltage induces capacitive nonlinearity, which
causes the effective stiffness of the resonator beam to decrease
[29], reducing the resonator frequency. It is necessary to ensure
that the bias voltage is not affected by the heating current. Fig. 14. Dynamic thermal response of the micro-ovenized resonator. (Inset)
Fig. 11 shows the device layer schematic with circuit diagram The in-built heater formed one leg of a wheatstone bridge. The measured
voltage output from the bridge represents the change in heater resistance as
of the resonator having in-built heater. Voltages V1 and V2 are the heater cools down following a heating pulse.
applied, such that the potential difference between them acts as
a joule heating voltage Vh across the in-built heater strain on the DETF, leading to a shift in frequency. The natural
Vh frequency of the resonator beam in the presence of compressive
V1 = Vb + axial force can be found using the following expression [30]:
2
Vh   12   12
V2 = Vb − (13) (4.73)2 P l2 EI
2 f= 1+ . (14)
2πl2 EIπ 2 m
where Vb is the effective bias voltage. Ideally, the resonator
should see a constant bias voltage, because the portions of the The stiffness of the in-built heater is approximately 10 000×
in-built heaters, before and after the resonator, are nominally smaller than the resonator beams, and the resulting frequency
identical, but fabrication uncertainties result in asymmetry of error due to differential thermal expansion of the heater is esti-
the heaters that cause the bias voltage to change with the mated to be within 0.5% of the total change in frequency due to
heating voltage. To remove the effect of this change in bias, a a temperature increase of 150 ◦ C. Since the error is not large and
feedback-control loop was implemented to maintain a constant we are interested in the approximate estimation of the power
bias voltage on the resonator irrespective of the variable heating consumption, we ignore this error. The resonator was calibrated
voltage. The bias-control circuit had a compliance of 1.0 mV, in a thermal chamber by measuring its frequency at different
which is equivalent to a < 0.1 ppm change in frequency. A ambient temperatures with no power applied to the in-built
schematic of the experimental test setup is shown in Fig. 12, heaters. The measured temperature coefficient of frequency was
and the results of the measurement are shown in Fig. 13. found to be nearly linear and equal to −29 ppm/◦ C, which is
Stress due to differential thermal expansion of the device approximately the same as that of a stress-free single-anchor
layer, the chip, and the package can create compressive axial silicon resonator [24], [31].
182 JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 17, NO. 1, FEBRUARY 2008

Fig. 15. Drop test resulted in a temporary change in frequency at the time of drop.

It is also necessary to check the influence of the in-built heating pulse of 4.5 V was used, and the voltage was maintained
heater suspension on the mechanical quality factor (Q) of the at 0.5 V during the cooling period in order to observe the change
resonator. Since the heater suspension is on both sides of the in voltage during cooling.
DETF, there is a linear temperature gradient across the length The measured time constants of several resonators varied
of the resonator beam. This gradient in temperature affects the between 7 and 10 ms. A typical measurement is shown in
Q. It has been shown [32] that the Q for a resonator with a Fig. 14. We expect the measurement to understate the time
linear temperature gradient and average temperature TA will be constant, because the measurement indicates the average tem-
slightly higher than the Q of a resonator with no temperature perature of the heater, not the temperature of the resonator itself.
gradient and uniform temperature TA . However, the tempera- It is to be noted that the thermal capacitance of the DETF is
ture gradient has a minimal effect on the frequency–temperature approximately equal to that of an in-built heater, and so, there is
calibration of the resonator. potential in reducing the thermal time constant in future designs
by reducing the thermal mass of the resonator itself.
A. Power Consumption
C. Impact Resistance of Mechanical Suspension
The DETF was heated using the in-built heater, and its
frequency was measured as a function of input power. As shown The thermal resistance of the resistive heater directly depends
in Fig. 13, the frequency decreases with the increase in input on the length of the beam and is inversely proportional to its
power as the temperature rises. The temperature rise was evalu- cross-sectional area. Design of a large thermal resistance is
ated using the calibration data of frequency versus oven temper- limited by the reduction of mechanical stiffness of the structure.
ature. It has been observed that there is a rise in temperature of However, miniaturization allows a stiffer design of the heater
approximately 125 ◦ C with total power consumption of around having relatively higher thermal resistance. To investigate the
12 mW. The experimental output is observed to be slightly stiffness of the thermally isolated DETF structure, a drop test
nonlinear as compared to the simulation results. This can be was carried out. The chip was soldered to an oscillator circuit
attributed to the fact that the material properties, including ther- board, which was rigidly bolted to a heating chuck (Fig. 15)
mal conductivity and electrical resistivity of silicon, vary with maintained at a constant temperature of 70 ◦ C. The chuck was
temperature but were considered constant for the calculation. dropped from a height of 1.0 cm onto a rigid platform, and the
response of the resonator frequency was measured, as shown in
Fig. 15. At the time of impact, the oscillator frequency changed
B. Thermal Time Constant
by approximately −45 000 ppm. The resonator immediately
The thermal time constant of the micro-ovenized resonator returned to normal operation. The change in frequency before
is an important parameter for temperature control. The thermal and after the impact is within the noise of the frequency fluc-
response was evaluated using a transient electrical-resistance tuation of the uncompensated resonator due to small variation
measurement. A voltage pulse was applied to the heater, which in the chuck temperature, as shown in Fig. 15. Furthermore,
caused its resistance to increase as it heated up. After the the resonators survived the 5000-rpm rotation (up to ∼1400 g
pulse ended, the heater resistance decreased as it cooled. A of acceleration) during the photoresist spin-coating steps of the
wheatstone bridge was used to measure the change in heater fabrication process. The spin duration was approximately 1 min
resistance during this cycle (Fig. 14). For the measurement, a and was repeated six to eight times during the fabrication.
JHA et al.: THERMAL ISOLATION OF ENCAPSULATED MEMS RESONATORS 183

V. C ONCLUSION [16] C. T.-C. Nguyen and R. Howe, “Microresonator frequency control and
stabilization using an integrated micro oven,” in Proc. Transducers,
An efficient heat delivery and thermal-isolation mechanism Yokohama, Japan, 1993, pp. 1040–1043.
for a MEMS resonator has been demonstrated. The in-built [17] C. T.-C. Nguyen, “Micromechanical resonators for oscillators and filters,”
in Proc. IEEE Ultrason. Symp., 1995, vol. 1, pp. 489–499.
heater-based thermal-isolation technique serves a dual purpose [18] M. A. Huff, S. D. Senturia, and R. T. Howe, “Thermally isolated
of localized heating and local thermal isolation, thereby provid- microstructure suitable for gas sensing applications,” in Tech. Dig. IEEE
ing maximum heating with minimized input power. At the same Solid-State Sens., Actuator Workshop, Hilton Head, SC, Jun. 6–9, 1988,
pp. 47–50.
time, the device has a small thermal time constant and high [19] R. N. Candler, M. A. Hopcroft, B. Kim, W. T. Park, R. Melamud,
impact resistance because of its miniature design. Compared M. Agarwal, G. Yama, A. Partridge, M. Lutz, and T. W. Kenny, “Long-
to the commercially used quartz crystals (1–10 W and around term and accelerated life testing of a novel single-wafer vacuum encapsu-
lation for MEMS resonators,” J. Microelectromech. Syst., vol. 15, no. 6,
30 min warm-up time), this paper has demonstrated orders- pp. 1446–1456, Dec. 2006.
of-magnitude improvement in power dissipation and dynamic [20] R. N. Candler, W.-T. Park, H. Li, G. Yama, A. Partridge, M. Lutz, and
thermal response with a potential for further improvement. T. W. Kenny, “Single wafer encapsulation of MEMS devices,” IEEE
Trans. Adv. Packag., vol. 26, no. 3, pp. 227–232, Aug. 2003.
Furthermore, this method is simple enough to implement it [21] B. Kim, R. N. Candler, M. A. Hofcroft, M. Agarwal, W.-T. Park, and
into any existing MEMS fabrication process. The described T. W. Kenny, “Frequency stability of wafer-scale film encapsulated sili-
design of microoven is highly suitable for temperature sta- con based MEMS resonators,” Sens. Actuators A, Phys., vol. 136, no. 1,
pp. 125–131, May 2007.
bilization of microresonators and for very precise control of [22] F. P. Incropera and D. P. Dewitt, Fundamentals of Heat and Mass Transfer,
frequency (< 1.0 ppm) over a large temperature span. Future 5th ed. New York: Wiley, 2002.
work will focus on designing thermal-isolation structures with a [23] M. N. Ozisik, Heat Conduction, 2nd ed. New York: Wiley, 1993.
[24] C. Bourgeois, E. Steinsland, N. Blanc, and N. F. de Rooij, “Design of
single point of attachment to the resonator, which will improve resonators for the determination of the temperature coefficients of elastic
performance by eliminating temperature gradients across the constants of monocrystalline silicon,” in Proc. IEEE Int. Freq. Control
resonator and by reducing power consumption and thermal Symp., 1997, pp. 791–799.
[25] J. D. Plummer, M. Deal, and P. B. Griffin, Silicon VLSI Technology:
capacitance. Fundamentals, Practice, and Modeling. Englewood Cliffs, NJ: Prentice-
Hall, 2000.
[26] A. D. McConnell, U. Srinivasan, and K. E. Goodson, “Thermal conductiv-
ity of doped polysilicon layers,” J. Microelectromech. Syst., vol. 10, no. 3,
R EFERENCES pp. 360–369, Sep. 2001.
[1] C. T.-C. Nguyen, “MEMS technology for timing and frequency control,” [27] M. Asheghi, K. Kurabayashi, R. Kasnavi, and K. E. Goodson, “Thermal
IEEE Trans. Ultrason., Ferroelectr., Freq. Control, vol. 54, no. 2, pp. 251– conduction in doped single crystal silicon films,” J. Appl. Phys., vol. 91,
270, Feb. 2007. no. 8, pp. 5079–5088, Apr. 2002.
[2] V. Kaajakari, J. Kiihamaki, A. Oja, S. Pietikainen, V. Kokkala, and [28] W. G. Vincenti and C. H. Kruger, Jr., Introduction to Physical Gas
H. Kuisma, “Stability of wafer level vacuum encapsulated single-crystal Dynamics. New York: Wiley, 1975.
silicon resonators,” Sens. Actuators A, Phys., vol. 130/131, pp. 42–47, [29] V. Kaajakari, T. Mattila, A. Oja, and H. Seppa, “Nonlinear limits
Aug. 2006. for single-crystal silicon microresonators,” J. Microelectromech. Syst.,
[3] C. T.-C. Nguyen and R. T. Howe, “An integrated CMOS micromechanical vol. 13, no. 5, pp. 715–724, Oct. 2004.
resonator high-Q oscillator,” IEEE J. Solid-State Circuits, vol. 34, no. 4, [30] R. D. Blevins, Formulas for Natural Frequency and Mode Shape.
pp. 440–455, Apr. 1999. Malabar, FL: Krieger, 2000.
[4] D. Sparks, S. Massoud-Ansari, and N. Najafi, “Chip-level vacuum pack- [31] R. Melamud, M. A. Hopcroft, C. M. Jha, B. Kim, S. A. Chandorkar,
aging of micromachines using nanogetters,” IEEE Trans. Adv. Packag., R. N. Candler, and T. W. Kenny, “Effects of stress on the temperature
vol. 26, no. 3, pp. 277–282, Aug. 2003. coefficient of frequency in double clamped resonators,” in Proc. Solid-
[5] W.-T. Hsu and C. T.-C. Nguyen, “Stiffness-compensated temperature- State Sens., Actuators, Microsyst., 2005, vol. 1, pp. 392–395.
insensitive micromechanical resonators,” in Proc. IEEE MEMS, 2002, [32] S. A. Chandorkar, H. Mehta, M. Agarwal, M. A. Hopcroft, C. M. Jha,
pp. 731–734. R. N. Candler, G. Yama, G. Bahl, B. Kim, R. Melamud, K. E. Goodson,
[6] E. P. Quevy and R. T. Howe, “Redundant MEMS resonators for precise and T. W. Kenny, “Non-isothermal micromechanical resonators,” in Proc.
reference oscillators,” in Proc. IEEE Radio Freq. Integr. Circuits Symp., MEMS, Kobe, Japan, 2007.
2005, pp. 113–116.
[7] M. A. Hopcroft, M. Agarwal, K. K. Park, B. Kim, C. M. Jha,
R. N. Candler, G. Yama, B. Murmann, and T. W. Kenny, “Tem-
perature compensation of a MEMS resonator using quality factor
as a thermometer,” in Proc. 19th IEEE Int. Conf. MEMS, Jan. 2006,
pp. 222–225.
[8] J. R. Vig, “Temperature-insensitive dual-mode resonant sensors—A
review,” IEEE Sensors J., vol. 1, no. 1, pp. 62–68, Jun. 2001.
[9] R. L. Filler and J. R. Vig, “Resonators for the microcomputer com- Chandra Mohan Jha received the B.E. degree in
pensated crystal oscillator,” in Proc. 43rd Annu. Freq. Control Symp., mechanical engineering from the Birla Institute of
1989, pp. 8–15. Technology, Mesra, India, in 1996 and the M.S.
[10] J. R. Vig, “Temperature stable crystal oscillator,” IEEE Trans. Ultrason., degree in mechanical engineering from Stanford
Ferroelectr., Freq. Control, vol. 42, no. 4, pp. 797–799, Jul. 1995. University, Stanford, CA, in 2006. He is currently
[11] A. Ballato and J. R. Vig, “Static and dynamic frequency-temperature working toward the Ph.D. degree in the Department
behavior of singly and doubly rotated, oven-controlled quartz resonators,” of Mechanical Engineering and the Department of
in Proc. 32nd FCS, 1978, pp. 180–188. AD-A955 718. Electrical Engineering, Stanford University.
[12] I. Balaz and M. Minarik, “Towards an OCXO with infrared heater,” in From 1996 to 1997, he was with LML Ltd.,
Proc. IEEE Int. FCS, 1996, pp. 674–680. India, as an Engineer. From 1997 to 2004, he was
[13] M. Vaish, “A high precision quartz oscillator with performance compa- with Bhabha Atomic Research Centre as a Scientific
rable to rubidium oscillators in many respects,” in Proc. IEEE Int. FCS, Officer, working on design and development of mechanical systems for harsh
1996, pp. 752–760. environments. His current research is focused on the design and optimization
[14] Vectron International, C4550 OCXO. [Online]. Available: http://www. of micro-oven-based MEMS resonators for low power and high-temperature
vectron.com stability.
[15] M. E. Frerking, Crystal Oscillator Design and Temperature Compensa- Mr. Jha is a Student Member of the American Society of Mechanical
tion. New York: Van Nostrand, 1978. Engineers (ASME).
184 JOURNAL OF MICROELECTROMECHANICAL SYSTEMS, VOL. 17, NO. 1, FEBRUARY 2008

Matthew A. Hopcroft received the B.Sc. degree in Rob N. Candler (S’98–M’00) received the B.S.
computer engineering from the George Washington degree in electrical engineering from Auburn
University, Washington, D.C., in 1998, the M.Phil. University, Auburn, AL, in 2000 and the M.S.
degree from Cambridge University, Cambridge, and Ph.D. degrees in electrical engineering from
England, in 2002, and the Ph.D. degree in mechan- Stanford University, as a National Defense Science
ical engineering from Stanford University, Stanford, and Engineering Graduate Fellow, in 2002 and 2006,
CA, in 2007. respectively.
He is currently a Research Specialist with the He was a Consulting Assistant Professor with the
Berkeley Micromechanical Analysis and Design Department of Electrical Engineering and the De-
Group and the Berkeley Sensor and Actuator Cen- partment of Mechanical Engineering, Stanford Uni-
ter, University of California, Berkeley. His research versity. He is currently a Senior Research Engineer
interests include MEMS material property measurements, microscale and with Bosch Research and Technology Center, Palo Alto, CA, working on
portable power systems, and micromechanical resonators. MEMS-based sensors for automotive and commercial markets. He also directs
Bosch’s involvement in the DARPA Science and Technology Fundamentals
program. His research interests include RF MEMS, wafer-scale sensor pack-
aging, and sensor reliability.

Saurabh A. Chandorkar received the B.Tech. Renata Melamud received the B.S. degree in me-
degree in mechanical engineering from the Indian chanical engineering from Carnegie Mellon Univer-
Institute of Technology, Mumbai, India, in 2003 sity, Pittsburgh, PA, in 2003 and the M.S. degree
and the M.S. degree in mechanical engineering in mechanical engineering from Stanford University,
from Stanford University, Stanford, CA, in 2005. Stanford, CA, in 2006. She is currently working
He is currently working toward the Ph.D. degree at toward the Ph.D. degree in the Department of Me-
Stanford University. chanical Engineering and the Department of Elec-
His current research interests include the study trical Engineering, Stanford University, wherein she
of energy loss mechanisms, namely, thermoelastic is supported by the Gabilan Stanford Graduate Fel-
dissipation, Akhieser effect loss and clamping loss, lowship for her research on passive compensation
and micromechanical resonators. techniques to reduce the temperature dependence of
frequency in silicon resonators.

Bongsang Kim received the B.S. degree in mechan-

ical design and production engineering from Seoul
National University, Seoul, Korea, and the M.S.
James C. Salvia (S’03) received the B.S. degree and Ph.D. degrees in mechanical engineering from
in electrical and computer engineering from Stanford University, Stanford, CA, in 2004 and 2007,
Carnegie Mellon University (CMU), Pittsburgh, PA, respectively.
in 2005. He is currently working toward the M.S. and From 1998 to 2001, he was with Hyundai Mobis
Ph.D. degrees in electrical engineering at Stanford as a Mechanical Engineer. He was an Intern with
University, Stanford, CA. Agilent Labs, Palo Alto, CA, in 2005, working on the
He was a Research Assistant with CMU, where development of nanostepper for AFM application.
he developed on-chip magnetic inductors for high- He is currently a Postdoctoral Researcher with the
frequency applications. He is currently with the University of California, Berkeley. His research interests include RF MEMS,
Microstructures and Sensors Laboratory, Department reliability of MEMS structures, MEMS packaging, material diffusion, and
of Mechanical Engineering and the Department of energy loss and stability of micromechanical resonators.
Electrical Engineering, Stanford University, working on circuit development Dr. Kim received the Best Student Poster Presentation Award at the IEEE
for the stabilization and control of microresonator-based oscillators. Frequency Control Symposium 2007.

Thomas W. Kenny received the B.S. degree

in physics from the University of Minnesota,
Minneapolis, in 1983 and the M.S. and Ph.D. de-
Manu Agarwal (S’01–M’06) received the B.Tech. grees in physics from the University of California,
degree in electrical engineering from the Indian In- Berkeley, in 1987 and 1989, respectively.
stitute of Technology, Kanpur, India, in 2003 and From 1989 to 1993, he was with NASA Jet
the M.S. and Ph.D. degrees in electrical engineer- Propulsion Laboratory, working on the development
ing, with a doctoral thesis on modeling nonlinear- of electron-tunneling high-resolution microsensors.
ities in electrostatically actuated MEMS resonators Since 1994, he has been with the Department of Me-
and study of their impact on phase noise, from chanical Engineering and the Department of Electri-
Stanford University, Stanford, CA, in 2005 and 2007, cal Engineering, Stanford University, Stanford, CA,
respectively. where he directs MEMS-based research in a variety of areas including res-
He was an Intern with the Technical University of onators, wafer-scale packaging, cantilever beam force sensors, microfluidics,
Ilmenau, Ilmenau, Germany, and the Ecole Polytech- and novel fabrication techniques for micromechanical structures. He is a
nique Fédérale de Lausanne, Lausanne, Switzerland. He was an Intern with Founder and Chief Technical Officer of Cooligy, a microfluidics chip cooling
Agilent Labs, Palo Alto, CA, working on characterizing dipole surface drive components manufacturer and Founder and Board Member of SiTime, a
MEMS actuators. He is currently with Netlogic Microsystems, Mountain View, developer of CMOS timing references using MEMS resonators. He is currently
CA, working as a Mixed-Signal Circuit Design Engineer. He is the author or a Stanford Bosch Faculty Development Scholar and is serving as a Program
coauthor of more than 30 technical publications. He is the holder of one U.S. Manager with the Microsystems Technology Office, DARPA. He is the author
patent. or coauthor of more than 200 scientific papers. He is the holder of 40 patents.

View publication stats