Lab on a Chip Paper
and height of 230 μm and was operated at 3.19 MHz. The
rectangular-cross-section channel had a width of 230 μm and
a height of 150 μm, and was operated at 3.24 MHz and 5.09 MHz,
respectively. The piezo-ceramic ultrasound transducers (PZ26;
Ferroperm piezoceramics, Kvistgaard, Denmark) actuating
the chips were glued to the chips with cyanoacrylate glue
(Loctite Super Glue, Henkel Norden AB, Stockholm, Sweden).
The 3 MHz ultrasound transducer was glued to the silicon and
the 5 MHz transducer was glued to the glass lid, both at the
middle of the chip. To control the temperature, a Peltier element
(Farnell, London, UK) was glued underneath the 3 MHz
ultrasound transducer and a Pt100 or Pt1000 resistance temperature
detector (Farnell, London, UK) was glued to the glass lid.
Instrument set-up
The transducers were actuated using a dual-channel function
generator (AFG 3022B; Tektronix UK Ltd., Bracknell, UK), the
signals were amplified using in-house built power amplifiers
based on an LT1012 power amplifier (Linear Technology Corp.,
Milpitas, CA, USA) and the applied voltage was monitored
using an oscilloscope (TDS 2120; Tektronix). The temperature
was controlled using a Peltier-controller (TC2812; Cooltronic
GmbH, Beinwil am See, Switzerland) and the temperature
was set to 37 °C throughout all experiments. Fluorescent
microscopy images were obtained using a Hamamatsu camera
(Hamamatsu Photonics KK, Hamamatsu, Japan) installed on
an Olympus microscope (BX51WI; Olympus Corporation,
Tokyo, Japan).
Experimental set-up
The flow rates were controlled using syringe pumps
(neMESYS; Cetoni GmbH, Korbussen, Germany) mounted with
glass syringes (Hamilton Bonaduz AG, Bonaduz, Switzerland)
connected to the inlet and the outlet of the side channels.
The center outlet was kept open and sample was collected
from a short piece of tubing directly into an Eppendorf
tube. While the inlet and outlet flow rates were varied, the
outlet flow rates were kept at a split ratio of 40 : 60 at the
center outlet and the outlet connected to the side channels.
To minimize errors caused by sedimentation in the syringes
and tubing, which would vary with the flow rate, sample collection
with the ultrasound either on or off (for each flow
rate) was compared. Particles and bacteria were quantified
using a Coulter counter (Multiziser III; Beckman Coulter,
Brea, CA, USA). Flow rates and voltage settings are given in
Table 1.
Microparticles
Polystyrene microparticles of various sizes were used to
characterize the system: 7.11 μm, 4.99 μm, and 3.17 μm
diameter particles were obtained from Sigma-Aldrich (Buchs,
Switzerland), and 0.992 μm and 0.591 μm particles and 0.49 μm
and 0.24 μm fluorescent particles were obtained from Kisker
(Kisker Biotech GmbH & Co. KG, Steinfurt, Germany). Fluorescent
particles 0.78 μm in diameter were obtained from
Bangs Laboratories (Bangs Laboratories, Fishers, IN, USA).
Particle concentrations were kept below 109 mL−1, to minimize
the effect of acoustic and hydrodynamic interaction
forces between particles.
Bacteria
For biological evaluation of the system, Escherichia coli
(E. coli) DH5-α (containing a plasmid that carries the
ampicillin-resistance gene), a kind gift from Åsa Janfalk
Carlsson, was used. E. coli was cultured in liquid LB medium
or LB plates containing 10 g L−1 tryptone (T1332; Saveen &
Werner, Limhamn, Sweden), 5 g L−1 yeast extract (Hy-Yeast 412;
Sigma-Aldrich), 10 g L−1 NaCl (Sigma-Aldrich) and 100 mg L−1
ampicillin (A9518-5G; Sigma-Aldrich) or agar (bacteriologygrade,
A0949; Saveen & Werner).
Experimental results and discussion
In what follows, a system is presented that reduces the lower
particle size focusing limit for acoustophoresis to the submicrometer
range, thus enabling applications in research
fields such as microbiology. The experiments were carried
out on two variants of an acoustophoresis microfluidic chip,
which had a straight square or rectangular channel with a single
inlet for particle suspensions and a trifurcation outlet split
(Fig. 3). Ideally, with the onset of continuous ultrasonic actuation,
particles are focused in the center of the channel and
exit through the central outlet—to an extent that depends on
the acoustic energy density, the flow rate of the suspension,
and the size and material properties of the particles relative
to the suspending liquid. In the experiments particles with
diameters ranging from 0.6 μm to 7 μm were used and for
Table 1 Nominal flow rates Q as set on the syringe pumps and voltage settings for the different experiments
Particle Rectangular chip 1D Rectangular chip 2D Square chip 2D
Diameter (μm) Manufacturer
Voltage
Upp (V)
Flow rate
Q (μL min−1)
Voltage
Upp,2 (V)
Flow rate
Q (μL min−1)
Voltage
Upp (V)
Flow rate
Q (μL min−1)
7 Sigma-Aldrich 2.5 50, 70, 90, 110, 130 3.16 50, 70, 90, 110, 130
5 Sigma-Aldrich 3.52 50, 70, 90, 110, 130 4.26 50, 60, 70, 80 90
3 Sigma-Aldrich 5.72 70, 80, 90, 100, 150, 200 5.73 50, 60, 70, 80, 90
1 Kisker 10.4 10, 20, 30, 40, 50, 60 0–4 10 10.6 15, 25, 35, 45, 55
0.6 Kisker 11 3, 5, 10, 15, 20 0–4 3 10.6 5, 10, 15, 20
0.5 Kisker 11 0.5, 0.8, 1.2, 2 10.6 0.5, 0.8, 1.2, 2
Lab Chip, 2 This journal is © The Royal Society of Chemistry 2014 014, 14, 2791–2799 | 2795