JSSSJournal of Sensors and Sensor SystemsJSSSJ. Sens. Sens. Syst.2194-878XCopernicus PublicationsGöttingen, Germany10.5194/jsss-6-121-2017Transferable micromachined piezoresistive force sensor with integrated
double-meander-spring systemHamdanaGerryg.hamdana@tu-bs.deBertkeMaikDoeringLutzFrankThomasBrandUweWasistoHutomo SuryoPeinerErwinInstitute of Semiconductor Technology (IHT), Technische
Universität Braunschweig, Hans-Sommer-Straße 66, 38106
Braunschweig, GermanyLaboratory for Emerging Nanometrology (LENA), Langer Kamp 6a, 38106
Braunschweig, GermanyDepartment 5.1 Surface Metrology, Physikalisch-Technische
Bundesanstalt (PTB), Bundesallee 100, 38116 Braunschweig, GermanyCiS Forschungsinstitut für Mikrosensorik GmbH,
Konrad-Zuse-Straße 14, 99099 Erfurt, GermanyGerry Hamdana (g.hamdana@tu-bs.de)2March20176112113330September20169February20179February2017This work is licensed under a Creative Commons Attribution 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by/3.0/This article is available from https://jsss.copernicus.org/articles/6/121/2017/jsss-6-121-2017.htmlThe full text article is available as a PDF file from https://jsss.copernicus.org/articles/6/121/2017/jsss-6-121-2017.pdf
A developed transferable micro force sensor was evaluated
by comparing its response with an industrially manufactured device. In order
to pre-identify sensor properties, three-dimensional (3-D) sensor models
were simulated with a vertically applied force up to 1000 µN. Then, controllable batch fabrication was performed by alternately
utilizing inductively coupled plasma (ICP) reactive ion etching (RIE) and
photolithography. The assessments of sensor performance were based on sensor
linearity, stiffness and sensitivity. Analysis of the device properties
revealed that combination of a modest stiffness value (i.e., (8.19 ± 0.07) N m-1) and high
sensitivity (i.e., (15.34 ± 0.14) V N-1) at different
probing position can be realized using a meander-spring configuration.
Furthermore, lower noise voltage is obtained using a double-layer silicon on
insulator (DL-SOI) as basic material to ensure high reliability and
an excellent performance of the sensor.
Introduction
Rapid development of micro-/nano-fabrication techniques in the past years has
had a significant impact not only on higher integration density of the
micro-/nano-components leading to enhanced device performance but also on
the need of better and more precise quality control of the individual
component itself. Consequently, more precise characterizations of physical
properties in smaller scale (e.g., micro- or nanoscale) and their know-how
are in demand.
Three-dimensional (3-D) model of proposed SOI-based micro force sensor (a). Magnified
top view of different spring designs (b, c). Cross sectional view of the
micro force sensor based on design 1 showing the point loading on the
probing area (boss) and the fixing of the sensor to the device holder (d).
Simulated 3-D model of both device configurations (i.e., designs 1
and 2). The color legend shows deflection state on the boss structure in
µm (a). The sensor displacement in z direction was computed
with a maximum force of 1000 µN (b).
As one of the fundamental techniques in material characterization,
nano-indentation has already been able to perform measurements at a
resolution down to 1 nN with a minimum displacement of < 1 nm.
During measurement, this method applies certain loads to create indentation
patterns on the sample, which can be used to determine hardness
(Fu et
al., 2015; Li and Brand, 2013; Nili et al., 2013; Yetna Njock et al., 2016).
However, to date there is still no transferable standard available, which
can be used for calibrating nano-indentation instruments in the micro- and
nano-range. For this purpose, a system for calibrating very small forces is
therefore needed. Although different approaches (e.g., piezoelectric:
Choi et al., 2016; Lee
et al., 2014; Mohan et al., 2015; capacitive:
Brand et al., 2016;
Viry et al., 2014; Woo et al., 2014; and magnetic flux change:
Mehrtash et al., 2015) have been used to realize small
force scales. Measuring forces in small scales is also very useful for
robotics and medical engineering (e.g., a silicon on insulator (SOI)-based piezoresistive sensor
was attached on the tip of the guidewire to detect motions in
z axis during the catheterization process) while connecting the
micro- and macro-environments is possible by closed-loop disturbance
compensation of the system (Mizutani, 2013, 2014; Ousaid
et al., 2015).
Micro-fabricated silicon-based force standards for calibrating hardness
testing instruments were described in detail elsewhere and are commercially
available from Simetrics GmbH, Germany (Frühauf et
al., 2007a, b). However, these artifacts do not offer a direct
self-sensing output. Furthermore, they have rather large stiffness values
(i.e., ∼ 20 mN µm-1), which are not adequate for the
calibration procedure of indentation instruments in the µN range. In micro-electromechanical system
(MEMS)
technology, several transduction mechanisms are available, which can be taken
into consideration for the design of a self-sensing force artifact. Among
them, capacitive, optical, piezoelectric and piezoresistive approaches are
most popular. With capacitive sensors, the available measurement range could
be limited by the small gap (representing the capacitance) between the
probing area of the force sensor and the bottom plate. Measuring small
deflections optically is only possible on a reflective surface. Fabrication of
an additional reference structure on the probing area and specialized
instrumentation are needed to perform reliable deflection measurements. The
use of piezoelectric transduction requires deposition of a piezoelectric
layer on the spring beam, which may induce residual stress. In contrast,
piezoresistive strain gauges can be directly realized using doped silicon
resistors; i.e., different robust designs of force sensors are possible by
combining standard silicon processing steps.
Recently, a force artifact for the micro-Newton range based on double-sided
clamped probing body via combined meander–membrane springs (Doering et al.,
2013) were proposed. Meander-type springs in combination with bending
springs are well suited to take up the lateral strain, which increases with
the deflection of the probing body and are thus able to provide a large
linear range of the micro force artifact. Samples of this artifact, which
were realized using reactive ion etching (RIE) at cryogenic temperature in
silicon, confirm the load-deflection behavior derived from finite element
modeling (FEM; Hamdana et al., 2016; Wasisto et
al., 2015a). For an application under industrial conditions the micro force
artifact comprises a strain-sensing piezoresistive Wheatstone bridge (WB) of
very low cross-sensitivity to non-constant ambient conditions such as
temperature, humidity and light. Furthermore, noise of the WB limiting the
minimally detectable force should be as low as possible. For both low
cross-sensitivity and noise, electrical and thermal decouplings of the WB
from the membrane spring using a buried SiO2 layer were proven to be
effective (Kähler et al., 2012, 2013). Utilizing this approach, we aim to fulfill the need of small
force standards; e.g., for nano-indenters following specifications for small force calibration procedure are required (Table 1).
Required specifications of small force calibration standards for nano-indenters.
ForceMeasurementResolutionResonanceBandwidthChip sizeProbing areacomponentrangefrequencyz0–1 mN0.1 µN> 100 Hz10 Hz< 20 mm × 20 mm> 1 mm × 1 mm
In this paper, we assess the performance of our fabricated force sensors
that have double-meander-spring structures and compare them to industrially
manufactured sensor devices. First, mechanical and electrical properties of
both designs were numerically analyzed. Afterwards, sensor fabrication was
carried out using inductively coupled plasma (ICP) deep reactive ion etching (DRIE) at cryogenic temperature as the key process
(Wasisto et al., 2014; Merzsch et al., 2014).
Direct-sensing capability of different types of piezoresistive strain gauges
(i.e.,
etched and implanted WBs) was carried out. Furthermore, point-force
applications at different positions along the x axis of the boss
structure were carried out to evaluate the sensor performance under
conditions corresponding to the typical nano-indenter force calibration
procedure.
Design and simulation
In this paper, we aim to evaluate the performance of our developed micro force sensor
(i.e., design 1) in comparison with the fabricated device from the
industry (i.e., design 2). The first step was to model accurate
three-dimensional (3-D) representations of both micro force sensors and
determine their physical properties (i.e., stiffness and sensitivity) in a
finite element simulation using COMSOL Multiphysics 4.3b
(COMSOL Multiphysics, 2013). In general, the model geometry was
adapted to the actual chip size of the device, i.e., 20 mm × 2.2 mm.
The sensor geometry consists of a probing area (i.e., boss structure),
two meander-beam spring structures and two piezoresistive strain gauge
elements (WBs) located at both clamped ends of the device
(Fig. 1a). Two designs of the sensor were investigated by
employing different spring structure configurations. Design 1 applies deep
meander structures perpendicular to the x-y plane with small regular
gaps between the structures (Fig. 1b), while design 2 adopts a
micro-structure with various lateral dimensions and gaps between the
structures (Fig. 1c). When a vertical force is applied on the
center of the boss structure, a smooth bending of the beams followed by
elastic material deformation and changes of resistance values on both WB
structures occurred (Fig. 1d). During the simulation, several
mesh sizes were used in order to reduce computational time without
sacrificing results reliability. In order to replicate the real device
conditions, anisotropic single crystal silicon was selected and the 3-D
models were rotated by 45∘ in the x-y plane, allowing for
simulations along < 110 > crystal orientation
(Bonev and Zlatanov, 2002; COMSOL
Multiphysics, 2013; Hopcroft et al., 2010).
Numerical simulation of both spring configurations was performed by
utilizing several predefined physics interfaces, i.e., solid mechanics and
piezoresistivity, domain currents (pzrd). For the mechanical boundary
conditions, the bottom part of the SOI-based micro force sensor was given a
fixed motion constraint. In contrast, other parts of the models were set to
be free. Moreover, the electrical potential boundary condition of 1 V was
applied to the input electrode of WB, whereas diagonally opposite output
electrode was grounded thereby creating a DC input voltage
(Vin). As a result, changes of the output voltage
(Voff) can be determined through the two remaining WB
contacts.
Simulated 3-D model of device sensitivity under an applied load of
1000 µN (a). Sensor output voltage vs. vertical loads up to 1000 µN (b).
The mechanical and electrical properties of both sensor configurations
(design 1 and design 2) were investigated. To capture the behavior of the
device in the envisaged application range, vertical forces between 0 and
1000 µN with a step size of 100 µN were
used. Owing to symmetry considerations, in the following we consider only
one of the two meander-beam springs of the sensor. We define the coordinate
system origin of the x axis to be located within the clamped area of
the beam. The origin of the z axis, where the deflection
z direction as a function of an applied force were given,
corresponds to a neutral layer of the vertical beam deflection. Along this
layer, the longitudinal stress equals to zero. FEM results of different
spring configurations under vertical loading up to 1000 µN are
presented in Fig. 2a. Theoretically, the longitudinal stress
σL in the [110] direction can be calculated using the
following expression (Park et al., 2010):
σL=12(l-x)zwt3F,
where x and z represent the positions on x and
z axes from the system origin located at the clamping positions and
the neutral layer of the membrane springs, respectively. In Eq. (1)
the geometrical factors of the beam are given by the beam length l,
width w and thickness t (Park
et al., 2010). Subject to material properties, the stress σL
can be described as a product of the elastic modulus E and
the longitudinal strain εL along the
x axis.
σL=EεL
At the same deflection state, the mechanical deformation on the WB
structures depends on the membrane bending radius r, which can be
determined by
r=zεL.
Taking geometrical, material and bending properties into account, the force
applied on the device can be derived from Eqs. (1)–(3) as
F=wt3E12r(l-x).
The relationship between the applied vertical load ΔF and
its resulting deflection of the probing body (boss) Δz is
referred as device stiffness k:
k=ΔFΔz.
Figure 2(b) shows the deflections obtained using FEM with different
sensor designs. The vertically applied load on design 1 was incrementally
increased and leading to a linear rise of the boss deflection. As a result,
design 1 exhibits a stiffness of (23.39 ± 9 × 10-9) N m-1.
In comparison, design 2 indicates a stiffer structure response than design
1. In this case, a stiffness of (100.18 ± 5 × 10-13) N m-1
was obtained, which is a factor of ∼ 4.3 higher than that of
design 1. A higher stiffness value implicates the requirement of higher
applied forces to deflect the structure, which in practical applications can
be limited by the instrument or probe to be calibrated. Fine tuning during
measurement of very small forces may be problematic since the output signal
(i.e., resistance changes of the WB) depends strongly on bending mechanism
of the membrane part of the spring. Hence, sensor sensitivity needs to be
taken into consideration, which was numerically analyzed by FEM based on
the resistance changes (ΔR/R) of the WB
(Wasisto et al., 2015b):
ΔRR=σLΠL+σTΠT,
where σL, σT, ΠL and ΠT are mechanical stresses along
longitudinal and transverse directions and piezoresistive coefficients for
each direction, respectively. For the WB structures and their electrical
contact regions, doping concentrations of 5 × 1018
and 1 × 1019 cm-3 are assumed, respectively. To
calculate resistance changes of the WB, a supply voltage (Vin)
of 1 V was applied diagonally
(Wasisto et al., 2015c) and the
output voltage (Voff) was determined on the opposite diagonal
of the resulting full bridge configuration consisting of four resistors
(i.e., R1=R2=R3=R4) along < 110> and its transverse
direction leads to following relationship between input and output:
Voff=Vin×R2R1+R2+R3R3+R4.
Figure 3a shows the distributed voltage value within WB under
vertical applied force of 1000 µN obtained by FEM. The
relation between load F and output voltage Voff on
different meander form is linear for both designs; i.e., we can define a
sensitivity:
S=ΔVoffΔF.
Comparing the different spring configurations, we find by a factor of
∼ 4 higher sensitivity value (i.e., (8.07 ± 4.07 × 10-6) V N-1) for design 1 than design 2
(i.e., (2.13 ± 2.98 × 10-7) V N-1). This result is correlated directly with the
simulated results for stiffness. Considering both device properties (i.e.,
stiffness and sensitivity), design 1 provides a more flexible spring
structure and higher sensitivity than design 2, which should be better
suited for calibrating nano-indenters in the range of small forces.
Averaged measured sensor calibration values of designs 1
and 2 at different positions on the boss along the longitudinal
(x axis, long.) and transverse
(y axis, trv.) directions.
Schematic representation of fabrication process of proposed
device: dicing the SOI wafer into smaller pieces and cleaning (a), growing
of thin oxide layer for boron diffusion (b), etching process of WB
structures (c), metal deposition on contact area (d), dry etching of device
front (e) and back sides (f).
Fabricated micro force sensor (a). Magnified view of boss and
meander-spring structures (b) and WB structures clamped on both device
ends (c).
Optical micrograph of WB of design 1 with “half metal” contact.
Micro-fabrication
Micro force sensors (design 1) were fabricated using a p-type
double-layer silicon on insulator (DL-SOI) wafer with < 100 >
orientation and a resistivity of 0.01–0.02 Ω cm
(Active Business Company GmbH, Germany). One major advantage of using DL-SOI
wafers is that a defined layer thicknesses can be selected. In this case, we
have utilized a DL-SOI wafer with a top layer (device layer), middle layer and
bottom layer (handle layer) of (3 ± 1), (25 ± 0.5) and (350 ± 15) µm,
respectively. Furthermore, an oxide layer was buried between device and
middle layers (BOX1), as well as between middle and handle layers (BOX2).
The thicknesses of these layers were (0.2 ± 0.01) µm (BOX1) and (0.5 ± 0.025) µm (BOX2). The use of DL-SOI
for design 1 provides better control of the spring and WB structures with
respect to the geometrical and doping uniformity. Using this novel approach, the etched
WB structures were realized, which are electrically insulated from the
underlying spring by the upper buried oxide layer. Thereby, current leakage
can be avoided, which occurs by using implanted p-type resistor in
n-type bulk silicon as in case of design 2. However, higher wafer
cost compared to standard bulk silicon material may be set against the use
of DL-SOI.
To begin the sensor fabrication, a 4 ′′ DL-SOI wafer was cut into 26 mm × 26 mm pieces.
Afterwards, the wafer was put into piranha solution
(H2O2 : H2SO4=1:1) and boiled at 90 ∘C
within 5 min to remove organic contamination on the surface
(Fig. 4a). This step was performed prior to and after oxidation and
before photolithography followed by producing ∼ 300 nm
thermal oxide on the device layer. Subsequently, a oxide film on a particular
contact position was etched and p+ diffusion (boron) was
performed to obtain a high-quality contact formation (Fig. 4b).
In contract to the WB of design 2, which was prepared by using of implanted
piezoresistors, WB structures of design 1 were anisotropically created using
ICP DRIE process
utilizing O2 and SF6 as etch gases at a cryogenic temperature
(Sentech Instruments GmbH, Germany; Wasisto et al., 2012, 2013,
2014). In this case, an etching parameter set of an ICP power of 500 W, a
high-frequency power of 6 W, an O2 flux of 7 sccm (sccm is standard cubic centimeter per minute) and an SF6 flux
of 129 sccm at -80 ∘C was utilized (Fig. 4c). Once the
WB structures had been fabricated, the oxide layer (BOX1) on the middle of
the sensor was removed using a buffered hydrofluoric acid (HF; 6–7 %).
Following this treatment, a “half-metal” contact (i.e., metal layer was not
deposited on the membrane) was realized on the device with 30 nm chromium
and 300 nm gold by means of a lift-off process (Figs. 4d, 6).
In the next step, photolithography was performed prior to etching of the
middle layer. An etching duration of 12–15 min was needed for structuring the middle
layer. Thereby, the boss structure was created and the front side of meander
structures was defined. The removal of the second oxide layer (BOX2) was
carried out by buffered HF (Fig. 4e). The most crucial step
during the whole fabrication process was the back-side structuring of the
meander spring. A 300 µm deep etching of micro-structures
with small gaps (i.e., 50 µm) was performed within
∼ 1.5 h using an ICP power of 500 W, a high-frequency power of
7 W, and a mixture of etch gases of O2 (6 sccm) and SF6 (129 sccm)
(Fig. 4f).
After completing the fabrication steps, a total of six micro force sensors
with chip dimensions of 20 mm × 2 mm were obtained from a
26 mm × 26 mm DL-SOI sample (Fig. 5a). Evaluation in scanning
electron microscope (SEM) was performed to analyze the dimensional stability
of the realized WB and meander structures. The realized gaps between the
meander-spring structures were larger than the designed ones (i.e., around
∼ 60 µm) due to overetching (Fig. 5b). However, this deviation was not crucial to the sensor performance in
general (Hamdana et al., 2016). Moreover, this
irregularity can be eliminated with fine tuning of the etch recipe,
especially regarding the distribution of the process temperature.
Furthermore, WB structures show only a small deviation of (7.20 ± 0.10) µm (i.e., below 3 %) from each other (Fig. 5c).
Hence, a low offset voltage Voff (i.e., below 10 mV V-1) was
expected, which is very acceptable for the signal processing of the WB.
Measured resistance values of WBs comparing designs 1 and 2.
Measurement setup for calibrating micro force sensors (a).
Different loading points were selected along two perpendicular lines (along
x axis: long.; along y axis: trv.) across the center of the
boss (b). Starting and end points of scanning procedure in longitudinal and
transverse directions are shown in (c) and (d).
Non-linearity curves of design 1 with meander structure (a) and
design 2 with combined micro-structures as spring (b).
Direct comparison sensor properties of designs 1 and 2.
ParameterRequiredDesign 1 (DL-SOI)Design 2 (p-doped WB)Chip area< 20 mm × 20 mm20 mm × 2 mm20 mm × 2 mmProbing area> 1 mm × 1 mm2 mm × 2 mm2 mm × 2 mmResonance frequency> 100 Hz805 Hz805 HzNoise at bandwidth of 10 Hz< 0.1 µN0.5 µV/0.03 µN*0.33 µV/0.07 µNMeasurement range1 mN50 µN50 µNTypical wafer price per pcs (4′′,10 pcs)–around EUR 400around EUR 25
* Measured at WB structures.
Device characterizations
To determine electrical properties of the sensor, measurements of WB
resistances were carried out. The WB resistor arrangement of the sensor
(design 1) was shown in Fig. 6. The bridge resistances of
both designs were measured under zero applied load to the probing. As shown
in Fig. 7, design 1 shows a value of (2979 ± 297) Ω
compared to design 2 with a higher resistance but lower standard deviation
of (5456 ± 146) Ω. Furthermore, offset voltage values of both
devices were also determined. In this case, design 1 has an offset value of
(0.05 ± 0.01) mV V-1, whereas design 2 exhibits a higher offset value of
(23.15 ± 0.008) mV V-1. The differences of resistor and offset voltage
values may be due to material properties and fabrication processes of those
two-types of WB structures. In addition, cross-sensitivities of the fabricated
WBs against light, temperature and moisture were determined at a WB supply
voltage of 1 V. For measurements under direct illumination with a specific
wavelength of 635 nm, a high power cold-light source (KL 1500, Schott AG,
Germany) and an optical power meter (Thorlabs GmbH, Germany) were used.
Measurements under controlled temperature and moisture were performed in
a sealed chamber. Assuming controlled measuring room conditions (i.e.,
illuminance level < 104 lumen m-2, temperature drift
< 1 K and relative humidity change < 6 %), we observed
changes of offset voltage less than 1 µV corresponding to force
errors below 0.1 µN.
Results of sensor calibration procedure (maximum applied force:
50 µN) at different loading points on the boss structure in
longitudinal direction: stiffness (a, b), force sensitivity (c, d) and bending
sensitivity (e, f).
Results of sensor calibration procedure (maximum applied force:
50 µN) at different loading points on the boss structure in
transverse direction: stiffness (a, b), force sensitivity (c, d) and bending
sensitivity (e, f).
Electrical noise performance of micro force sensors applied
without (DC) or with (AC) carrier frequency of an SOI-based WB (a) compared
to implanted (b) WB.
During the force application procedure, the sensor was mounted on an
aluminum holder of a three axis nano-positioning system. Precise sensor movement
with a reproducibility of 5 nm and at resolution of 1 nm was performed
against a probing body with a 300 µm ruby sphere, which was mounted
on the pan of a compensation balance (Fig. 8a). Thus, the
resulting force on the probing area was controlled at a resolution of 1 nN
and a reproducibility of 2.5 nN (Peiner and Doering,
2005). Sensor performance was assessed based on calibration on different
probing positions on boss structure, i.e., along the x axis
(longitudinal) and the y axis (transverse). Total length and
increment of 500 and 50 µm were selected during
measurements in both longitudinal (long) and transverse (trv) directions,
respectively (Fig. 8b–d).
Turning now to the force application procedure, the sensor was fixed on an
aluminum holder, which was mounted on a nano-positioning unit. This part can
be precisely moved in 3-D directions with a maximum displacement of 100 µm
and a resolution of 1 nm. The contact to the sensor was
then realized using a stylus with a glued ruby sphere with diameter of 300 µm (Fig. 8a). For both designs, a maximum force
of 50 µN was applied and controlled by a compensation
balance. Measured values were collected in steps of 400 nm movement along
the z axis. In this work, we assessed the sensor responses on force
applications at different positions on the probing body (boss). Therefore,
the load was applied along the longitudinal axis (i.e., x axis) and
the transverse axis (i.e., y axis) through the center of the boss.
Between the measurement points (i.e., position 1 and position 2), increment
and maximum longitudinal position changes of 50 and 500 µm were used, respectively (Fig. 8b–d).
As shown in Fig. 9a–b, both sensor designs demonstrate good
linearity up to 50 µN. Regarding their structural flexibility,
design 1 indicates a more compliant behavior with a greater deflection value of
boss in the z axis than design 2. Consequently, the first and second
derivatives of the force-deflection curve of design 2 exhibit greater
values, which indicate the need of higher force to achieve the same
deflection. This attribute could be unfavorable for material stability and
signal processing. Nevertheless, mechanical properties and sensor stability
were investigated by monitoring sensor responses during increasing and
decreasing loading. The term increasing loading (i.e., loading) referred to
measurement state, when the stylus first contacted the boss until its final
lower position. On the contrary, decreasing loading (i.e., unloading)
referred to boss movement from a lower position to its initial position under
a zero applied load. From repeated measurements of both motions, three
important device properties (i.e., stiffness, force sensitivity and bending
sensitivity) can be extracted, and the performances of different sensor
designs can be evaluated.
Mean device stiffness derived from loading and unloading of different sensor
designs are illustrated in Fig. 10a and b. As
predicted in the FEM simulations, design 1 exhibits more compliant behavior
than design 2. With a mean stiffness value in longitudinal (8.15 ± 0.08) N m-1 and transverse
(8.21 ± 0.06) N m-1 directions, the maximum
stiffness deviations along the probing area (boss) are below 2 % from the
mean value (Figs. 10a, 11a). On the other side,
design 2 yields a higher mean stiffness (70.57 ± 2.58)
(longitudinal) and (71.66 ± 1.26) N m-1 (transverse) with a stiffness
deviation up to ∼ 8% from the mean value (Figs. 10b, 11b). This means that design 1 can be deflected with
lower force and smaller increments between forces and being less affected by
the loading position than design 2. Combining this advantage with sensitivity its high measured mean (15.45 ± 0.17) and (15.34 ± 0.25) V N-1 in
longitudinal and transverse directions, respectively
and a maximum deviation value of ∼ 3 %, the design 1 shows better
performance to measure small forces then the design 2 with a maximum
deviation in sensitivity of around ∼ 39 % (Figs. 10c–d, 11c–d). In terms of bending sensitivity, design
1 yielded mean values in both longitudinal and transverse directions, i.e.,
(126.59 ± 2.10) and (126.29 ± 2.11) V m-1, with a maximum
deviation of ∼ 4 %. In contrast, design 2 exhibited bending
sensitivity values of (304.47 ± 38.60) and (310.66 ± 33.67) V m-1
with deviations of ∼ 33 % (Figs. 10e–f, 11e–f). Tables 2 and 3 summarize the measurement
results on both micro force sensor designs together with the requirements.
Therefore, specified parameters of both designs were confirmed
experimentally. However, the measurement range is limited between 0 and 50 µN, which has to be extended to 1 mN.
The minimal detectable force of the sensor is limited by the effective noise
voltage of the WB. In comparison to a WB with implanted resistances, the
SOI-based bridges are electrically isolated to the surrounded substrate by a
BOX layer. This allows us to operate it at an AC voltage with no current leakage. For the
measurements, we used the MGC plus measuring amplifier system of Hottinger Baldwin Messtechnik (HBM) GmbH
with the insert modules ML10B for DC supply and the ML30B for AC supply. The
ML30B uses a carrier frequency of 600 Hz. Figure 12a shows the
measured noise voltage of the SOI-based micro force sensor (i.e., design 1)
with DC and AC supplies. In this case, using an AC carrier frequency,
the noise voltage of the micro force sensor in the frequency ranging from
0.1 to 1 kHz could be reduced by a factor of 3. In contrast, the
p–n junctions of the implanted WBs are massively affected
by the potential change between p-type WB and the n-doped substrate in case of AC supply to the WB. Figure 12b
depicts the noise voltage of an implanted bridge (i.e., design 2) with and
without carrier frequency. Applying DC voltage during the measurements,
noise voltages varied from 0.1 to 50 Hz have shown more or less the same values.
Meanwhile, when the AC voltage was supplied to the bridges, the noise values are > 20 times higher than those at DC supply.
Conclusions
This study has set out to measure and assess a developed meander-type sensor
for calibration of nano-indenters in the micro-Newton force range compared to
an industrially manufactured device. Finite element modeling (FEM) was
performed to predict and to provide better understanding of sensor response.
The impact on different spring structures (i.e., meander structure for
design 1 and combined micro-structures of design 2) was investigated through
the evaluation of the device properties using FEM and fabricated sensor.
Furthermore, repeated measurements by applying vertical loads at different
contact points along the x and y axes across the probing
body (boss) were performed. While both designs show good measurement
linearity, the new one (design 1) has a more flexible structure with higher
sensitivity than the industrially manufactured device (design 2). Moreover,
design 1 shows an improved behavior concerning the dependence of measured
force and deflection on the loading position of the sensor. The loading
position can be changed within a range of ±250 µm and force
and deflection sensitivity do not change by more than 3 %. Furthermore, it
has been shown that double-layer silicon on insulator (DL-SOI) material
provides better sensor performance in terms of electrical noise. These
characteristics will have implications for providing transferable force
standards with excellent functionality and high reliability for the
micro-Newton scale.
Biographies
Gerry Hamdana received his Bachelor of Engineering degree in
mechanical engineering from the Esslingen University of Applied Science in
2012 and Master of Science degree in mechanical engineering specializing in
mechatronic/microsystems technology from the Braunschweig University of
Technology (TU Braunschweig), Germany, in 2015. Currently, he is pursuing his
PhD at the Institute of Semiconductor Technology (IHT), TU
Braunschweig, Germany. His research interests include semiconductor
processing, microsystems technology and micro-/nano-electromechanical systems
(MEMSs/NEMSs).
Maik Bertke received the Bachelor of Science degree and the Master
of Science degree in electrical engineering from the Technische
Universität Braunschweig, Germany, in 2013 and 2016, respectively.
Currently, he is working towards the PhD at the Institute of
Semiconductor Technology (IHT), TU Braunschweig and Laboratory for Emerging
Nano-metrology, Germany, where his main interests are in the fields of
micro-/nano-electromechanical system (MEMS/NEMS)-based sensors.
Lutz Doering received a Dr.-Ing. degree in micro-electronics
technology, precision instrument technology and computer science from the
Dresden University of Technology, Dresden, Germany, in 1990. He joined the
Department 5.1 Nano- and Micro-metrology, Physikalisch-Technische
Bundesanstalt, Braunschweig, Germany, in 2001. His current research
interests include characterization and optimization of micromachined
cantilever sensors designed for measuring coordinate and roughness
parameters and for the transfer of the micro- and nano-Newton force standard
to tactile probing tools.
Thomas Frank received a Dr.-Ing. degree in mechanical engineering,
microtechnology at the Technical University of Ilmenau, Ilmenau, Germany, in 2002.
Until 2008, he was Managing Director of Little Things Faktory GmbH in Ilmenau.
Since 2009 he has been manager of the business unit MEMSs of
the CiS Research Institute for Mikrosensorik GmbH in Erfurt, Germany.
The range of his tasks includes silicon-based piezoresistive sensors for pressure,
force, hardness, geometry and roughness.
Uwe Brand received a PhD in physics from the TU Braunschweig, Germany, in 1991.
Since 2000, he has been the head of the working group 5.11 Hardness and Tactile Probing Methods
in German National Metrology Institute, Physikalisch-Technische Bundestanstalt (PTB), Braunschweig, Germany.
His main research interests include the further development of the nano-indentation
technique and standard devices. His research group creates the basis for the traceability
of hardness measurements in industry, research and calibration laboratories to the national standards.
Hutomo Suryo Wasisto
received the Bachelor of Engineering degree in electrical engineering (Cum
Laude) from the Gadjah Mada University, Indonesia, the Master of Engineering
degree in semiconductor engineering (Cum Laude) from the Asia University,
Taiwan, and the Doktor-Ingenieur (Dr.-Ing.) degree in electrical engineering
(Summa Cum Laude) from the Technische Universität Braunschweig (TU
Braunschweig), Germany in 2008, 2010 and 2014, respectively. He was a
postdoctoral research fellow at the School of Electrical and Computer
Engineering (ECE), Georgia Institute of Technology, Atlanta, USA, in
2015–2016. Since 2016, he has been Head of Optoelectromechanical Integrated
Nano-systems for Sensing (OptoSense) Group in the Laboratory for Emerging
Nano-metrology (LENA), Braunschweig, Germany. His main research interests
include nano-opto-electro-mechanical systems (NOEMS), nano-sensors,
nano-electronics, nano-LEDs, nano-generators and nano-metrology. He has
published more than 60 papers in international scientific peer-reviewed
journals and conference proceedings as well as two European and German
patents with more than 200 citations. Hutomo Suryo Wasisto has also been the
recipient of the best-paper award and the best young scientist award at the
8th IEEE International Conference on Nano-/Micro-Engineered and Molecular
Systems (IEEE NEMS 2013) in Suzhou, China, and the 26th European Conference
on Solid-State Transducers (Eurosensors 2012) in Krakow, Poland. In 2014, he received the Walter-Kertz-Studienpreis (Walter
Kertz Study Award) for his excellent doctoral dissertation and achievements
of scientific studies at the interface between physics, electrical
engineering and information technology from the TU Braunschweig, Germany.
In 2015, he has been awarded with the Transducers 2015 travel grant award
from Transducer Research Foundation (TRF), USA, in the 18th International
Conference on Solid-State Sensors, Actuators and Microsystems (Transducers
2015), Anchorage, Alaska USA. Since 2013, he has been a reviewer for more
than 20 scientific journals and international research organizations (e.g.,
IEEE Journal of Micro-electromechanical Systems, IEEE Sensors Journal, IEEE Transactions on Industrial Electronics, IOP Journal of Micro-mechanics and
Micro-engineering, Sensors and Actuators A: Physical, International Journal of Electronics, Sensors, Applied Surface Science, and Journal of Hazardous Materials).
Erwin Peiner received PhD degrees in metastable binary metal
compounds by ion beam mixing from the University of Bonn, Bonn, Germany, in
1988, and the Venia Legendi degree in semiconductor technology from the
Faculty of Mechanical and Electrical Engineering, TU Braunschweig,
Braunschweig, Germany, in 2000. Currently, he is the leader and professor of
the Semiconductor Sensors and Metrology Group at IHT, TU Braunschweig. He
has published more than 250 papers in international journals and conference
proceedings. He is the project coordinator of the collaborative project
“HmtS” funded by the BMBF.
Data availability
Research data are available upon request to the authors.
The authors declare that they have no conflict of interest.
Acknowledgements
The authors thank Juliane Breitfelder and Doris Rümmler for
their technical support. The first author gratefully acknowledge support by
Braunschweig International Graduate School of Metrology (B-IGSM). This work
was performed in the collaborative project “Hochgeschwindigkeitsmikrotaster
für die Messung an Oberflächen von Strukturen mit großem
Aspektverhältnis (HmtS)” funded by German Federal Ministry of Education
and Research (BMBF) under no. 03V0409.
Edited by: G. Gerlach
Reviewed by: two anonymous referees
References
Bonev, I. and Zlatanov, D.: Advantages of the modified Euler angles in the
design and control of PKMs, Proceeding Parallel Kinematic Mach. Int. Conf.,
171–1880, 2002.Brand, U., Li, Z., Gao, S., Hahn, S., and Hiller, K.: Silicon double spring
for the simultaneous calibration of probing forces and deflections in the
micro range, Meas. Sci. Technol., 27, 15601,
10.1088/0957-0233/27/1/015601, 2016.Choi, W., Lee, J., Yoo, Y. K., Kang, S., Kim, J., and Lee, J. H.: Enhanced
sensitivity of piezoelectric pressure sensor with microstructured
polydimethylsiloxane layer Enhanced sensitivity of piezoelectric pressure
sensor with microstructured polydimethylsiloxane layer, Appl. Phys. Lett.,
104, 123701, 10.1063/1.4869816, 2016.COMSOL Multyphisics: COMSOL Multiphysics Release Notes: Version 4.3b, COMSOL AB,
Stockholm, Sweden, available at: https://www.comsol.de/shared/downloads/4.3b/ReleaseNotes.pdf, 2013.
Doering, L., Nesterov, V., and Peiner, E.: Mikrokraftsensor, German Patent No. 10 2010 012 701, 2013.
Frühauf, J., Gärtner, E., Herrmann, K., Menelao, F., Schwenk, D.,
Chudoba, T., and Vollmar, H.: Calibration of Instruments for Hardness
Testing, Recent Adv. Theory Pract. Hardness Meas. HARDMEKO, 141, 2007a.Frühauf, J., Gärtner, E., and Herrmann, K.: Standard for Calibration
of Instruments of Hardness Testing, Tech. Mess., 74, 7–8,
10.1524/teme.2007.74.7-8.385, 2007b.Fu, K., Chang, L., Zheng, B., Tang, Y., and Wang, H.: On the determination of
representative stress – strain relation of metallic materials using
instrumented indentation, J. Mater., 65, 989–994,
10.1016/j.matdes.2014.10.018, 2015.Hamdana, G., Wasisto, H. S., Doering, L., Yan, C., Zhou, L., Brand, U., and
Peiner, E.: Double-meander spring silicon piezoresistive sensors as
microforce calibration standards, Opt. Eng., 55, 91409,
10.1117/1.OE.55.9.091409, 2016.Hopcroft, M. A., Nix, W. D., and Kenny, T. W.: What is the Young's modulus of
silicon?, J. Microelectromech. S., 19, 229–238,
10.1109/JMEMS.2009.2039697, 2010.Kähler, J., Stranz, A., Doering, L., Merzsch, S., Heuck, N., Waag, A.,
and Peiner, E.: Fabrication, packaging, and characterization of p-SOI
Wheatstone bridges for harsh environments, Microsyst. Technol., 18,
869–878, 10.1007/s00542-011-1396-6, 2012.
Kähler, J., Peiner, E., Stranz, A., and Waag, A.: High temperature
piezoresisitive strain gauges made of Silicon-On-Insulator, United States
Patent application, US 2013/0068008 A1, WO 2012/141843 A3, 2013.Lee, J., Choi, W., Yong, Y. Kyoung, Hwang, K. Seon, Lee, S. M., Kang, S.,
Kim, J., and Lee, J. Hoon: A micro-fabricated force sensor using an all thin
film piezoelectric active sensor, Sensors, 14, 22199–22207,
10.3390/s141222199, 2014.
Li, Z. and Brand, U.: Towards Quantitative Characterisation of the Small
Force Transducer Used in Nanoindentation Instruments, Modern
Instrumentation, 61–67, 2013.Mehrtash, M., Zhang, X., and Khamesee, M. B.: Bilateral Magnetic
Micromanipulation Using Off-Board Force Sensor, IEEE/ASME Trans.
Mechatronics, 20, 3223–3231, 10.1109/TMECH.2015.2417116, 2015.Merzsch, S., Steib, F., Wasisto, H.S., Stranz, A., Hinze, P., Weimann, T., Peiner, E., and Waag, A.:
Production of vertical nanowire resonators by cryogenic-ICP-DRIE, Microsystem technologies, 20, 759–767, 10.1007/s00542-013-2032-4, 2014.
Mizutani, Y.: Disturbance Compensation based on Data Memory for Macro –
Micro Bilateral Control, IEEE Int. Symp. Ind. Electron., 2013.
Mizutani, Y.: Analysis, Modeling, and Compensation of Friction for Scaled
Bilateral Control, IEEJ J. Ind. Appl., 3, 344–349, 2014.Mohan, N., Luo, Y., and Shen, Y.: Developing a Self-Powered and Directly
Digitized Piezoelectric Micro Sensor for Monitoring Blood Pressure Change
inside Brain Aneurysm after Endovascular Treatment?: A Feasibility Study,
5th Annu. IEEE Int. Conf. Cyber Technol. Autom. Control Intell. Syst.,
2115–2120, 10.1109/CYBER.2015.7288275, 2015.Nili, H., Kalantar-zadeh, K., Bhaskaran, M., and Sriram, S.: Progress in
Materials Science In situ nanoindentation?: Probing nanoscale
multifunctionality, Prog. Mater. Sci., 58, 1–29,
10.1016/j.pmatsci.2012.08.001, 2013.Ousaid, A. M., Haliyo, D. S., Regnier, S., and Hayward, V.: A Stable and
Transparent Microscale Force Feedback Teleoperation System, IEEE/ASME Trans.
Mechatronics, 20, 2593–2603, 10.1109/TMECH.2015.2423092, 2015.Park, S. J., Doll, J. C., and Pruitt, B. L.: Piezoresistive cantilever
performance part I: Analytical model for sensitivity, J. Microelectromech. S., 19, 137–148,
10.1109/JMEMS.2009.2036581, 2010.Peiner, E. and Doering, L.: Force calibration of stylus instruments using
silicon microcantilevers, Sensor. Actuat.-A Phys., 123–124, 137–145,
10.1016/j.sna.2005.02.031, 2005.Viry, L., Levi, A., Totaro, M., Mondini, A., Mattoli, V., Mazzolai, B., and
Beccai, L.: Flexible three-axial force sensor for soft and highly sensitive
artificial touch, Adv. Mater., 26, 2659–2664,
10.1002/adma.201305064, 2014.Wasisto, H. S., Merzsch, S., Waag, A., Uhde, E., Salthammer, T., and Peiner,
E.: Airborne engineered nanoparticle mass sensor based on a silicon resonant
cantilever, Sensor. Actuat. B-Chem., 180, 77–89,
10.1016/j.snb.2012.04.003, 2012.Wasisto, H. S., Merzsch, S., Stranz, A., Waag, A., Uhde, E., Salthammer, T.,
and Peiner, E.: Femtogram aerosol nanoparticle mass sensing utilising
vertical silicon nanowire resonators, Micro Nano Lett., 8, 554–558,
10.1049/mnl.2013.0208, 2013.Wasisto, H. S., Huang, K., Merzsch, S., Stranz, A., Waag, A., and Peiner, E.:
Finite element modeling and experimental proof of NEMS-based silicon pillar
resonators for nanoparticle mass sensing applications, Microsyst. Technol.,
20, 571–584, 10.1007/s00542-013-1992-8, 2014.Wasisto, H. S., Merzsch, S., Uhde, E., Waag, A., and Peiner, E.: Handheld
personal airborne nanoparticle detector based on microelectromechanical
silicon resonant cantilever, Microelectron. Eng., 145, 96–103,
10.1016/j.mee.2015.03.037, 2015a.
Wasisto, H. S., Merzsch, S., Uhde, E., Waag, A., and Peiner, E.: Partially integrated cantilever-based airborne nanoparticle
detector for continuous carbon aerosol mass concentration monitoring, J. Sens. Sens. Syst., 4, 111–123, 10.5194/jsss-4-111-2015, 2015. 2015b.Wasisto, H. S., Doering, L., Daus, A., Brand, U., Frank, T., and
Peiner, E.:
Development of silicon microforce sensors integrated with double meander springs for standard hardness test instruments, Proc. SPIE 9517, Smart Sensors,
Actuators, and MEMS VII, and Cyber Physical Systems, 95171X (21 May 2015),
10.1117/12.2180117, 2015c.Woo, S.-J., Kong, J.-H., Kim, D.-G., and Kim, J.-M.: A thin all-elastomeric
capacitive pressure sensor array based on micro-contact printed elastic
conductors, J. Mater. Chem., 2, 4415, 10.1039/c4tc00392f, 2014.Yetna Njock, M., Roudet, F., Idriss, M., Bartier, O., and Chicot, D.:
Work-of-indentation coupled to contact stiffness for calc. elastic modulus
by instrumented indentation, Mech. Mater., 94, 170–179,
10.1016/j.mechmat.2015.12.003, 2016.