Thermal analysis of nanofluids in microfluidics using an

infrared camera
5

Pyshar Yi*a, Aminuddin Kayania, Adam F. Chrimesa, Kamran Ghorbania, Saeid Nahavandib, Kourosh
Kalantar-zadeh*a, and Khashayar Khoshmanesh*a
Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX
DOI: 10.1039/b000000x

10

In this paper, a microfluidic platform integrated with an infrared camera is used to investigate the heat
transfer of the system by adding 1% and 2.5% w/w of Al2O3 nanoparticles to the base fluid. Temperature
profiles are measured at different flow rates of 10, 40 and 120 lit/min, corresponding to Reynolds
numbers of ~1.5, 6 and 17.5 at the microchannel. Computational fluid dynamics (CFD) simulations are
also carried out to comprehend the obtained results.

Introduction
15

20

25

30

35

40

45

50

Over the past few years, electronic components have experienced significant improvements. The capabilities and speed of data
processing in such devices have been increasing at a dramatic rate. To facilitate this, the electronic components such as central
processing units (CPUs), integrated circuits (ICs) and mobile phones are equipped with more transistors and draw more power. This
leads to greater heat produced by those components. At the same time, the size of such electronic devices has been dramatically reduced,
making the cooling process even more challenging. Heat sinks have been employed to almost all modern electronic components to
transfer some of the heat from the hot spots into the surrounding environment. Most of heat sinks still rely on fans to enable a forced
convection cooling using the surrounding air. But as the fans get larger and louder the liquid cooling has emerged as a solution.
Liquid cooling enables the more efficient cooling of the hot spots due to the higher thermal conductivities of liquids such as water
compared to air (thermal conductivity of water is almost 23 times higher than that of air at room temperature). The other advantage of
liquid cooling is the reduction of noise level within the system.
However, considering the demand for more sophisticated and capable while smaller electronic systems, it seems that the thermal
conductivity of the common liquid coolants such as water ethylene glycol or oil should be improved.
One approach to enhance the thermal conductivity of liquids is to add solid particles into liquids of high thermal conductivity to
them, as theoretically suggested by James Clerk Maxwell in 1870s3. Interestingly, the first experimental measurements on thermal
characteristics of such mixtures were reported only several decades or so ago. Investigations carried out by Ahuja and Lui et al. 6 are
such examples showing the hydrodynamic and heat transfer characterisation of slurries. Such slurries were mixtures of particles of
various dimensions generally in the order of micro and millimeters. However, these mixtures suffer from a number of problems
including: the erosion of components caused by the abrasive action of the large particles and clogging in small flow channels resulting
significant pressure drop. Therefore, although the slurries have superior thermal conductivities, from other technological point of views,
they are rarely useable as liquid coolants.
The abovementioned setbacks which are associated with the dimensions of micro or larger size of particles can be solved by applying
particles of nanometer dimensions7. From a series of research works, this idea was first materialized by Choi8, who named the fluids with
suspended nanosize particles as nano-fluids, which drew more attention until today.
Recently, the addition of a small volume of nanoparticles has become a popular technique in heat transfer enhancement. The
enhancement is due to the fact that small size and volume fraction of particles resolves the clogging and pressure drop issues9. Moreover,
by having a large surface area produced by nanoparticles the heat exchange between the fluid and solid is augmented, while the
sedimentation of particles is reduced due to the floating stability of the particles.
Reports on suspensions including small amounts of metals such as Cu, or non-metallic nanoparticles such as Al2O3 and CuO have
shown substantially higher thermal conductivities than those of the base fluids10-14. The experimental results measured by Li et al.15
detailed the enhancement of thermal conductivity due to addition of 36 nm diameter Al2O3 particles into the water. It showed that
mixtures containing 0.5%, 2% and 4% volume fractions of particles increase the thermal conductivity by 3%, 7.7% and 9.3%,
respectively. Other works proved that even volume fractions less than 1% of Cu nanoparticles or carbon nanotubes dispersed in ethylene
glycol or oil can increase the thermal conductivity by 40% and 161%, respectively16. Despite all the aforementioned studies, none of
these works have ever focused on the study of nanofluids in microfluidics for cooling/heating applications. The microfluidics paradigm is
of special importance as it enables the incorporation of nanofluids for cooling of the next generation electronic components.

10

In this paper, we use a microfluidic system, integrated with a microheater and infrared camera to demonstrate the enhancement of heat
transfer in microfluidic systems by application of nanofluids. The microfluidic system consists of a polydimethylsiloxane (PDMS)
microchannel integrated onto a glass slide. DI water is used as the base fluid while Al2O3 nanoparticles with diameters less than 50 nm
are used to make nanofluid suspensions. The thermal characteristics of the system have been examined at different nanoparticle
concentrations of 0%, 1% and 2.5% w/w. Temperature profiles are measured at different flow rates of 10, 40 and 120 lit/min,
corresponding to Reynolds numbers of ~1.5, 6 and 17.5 at the microchannel. Extensive computational fluid dynamics (CFD) simulations
are applied to comprehend the response of the system and calculate the variations of heat flux at desired locations of the system. Despite
of several reports by various groups, very little attention has been paid to the application of nanofluids in microfluidic system at such low
Reynolds numbers. Moreover, to the best of our knowledge, it is one of the first reports of the accurate measurement of temperature
profiles to investigate the response of the systems before and after application of nanofluids.

Theory

15

Three dimensional numerical simulations were conducted to analyse the variations of velocity, pressure, temperature, and heat loss
across the microfluidic system. The highest Reynolds number, Re=D/ ( is fluid density, is the average fluid velocity, D is the
hydraulic diameter of the rectangular microchannel, and is the fluid viscosity) obtained at the flow rate of 120 lit/min is 17.7,
indicating the laminar characteristics of the flow. The simulation was conducted using Gambit 2.3 software (Fluent, Lebanon, NH, USA)
to create the geometry and mesh generation. Subsequently, the finite-volume based Fluent 6.3 software (Fluent, Lebanon, NH, USA) was
used to solve the associated differential equations governing the balance of mass, momentum, and energy within the microfluidic system,
as given by Khoshmanesh et al37.

U 0

(1)

(U ) U P 2U

c p (U ) T k 2T Q
20

(2)
(3)

where, U , P and T are the velocity, pressure and temperature of the fluid, , , cp and k are the density, dynamic viscosity, heat capacity
and thermal conductivity of the fluid, and Q is the heat source term corresponding to the microheater. The thermophysical properties of
nanofluid are obtained as below:

n f (1 ) f p

n f

(4)

1
f
(1 ) 2.5

(5)

( c p ) n f (1 ) ( c p ) f ( c p ) p
kn f

k p (n 1) k f (n 1) (k f k p )
k p (n 1) k f (k f k p )

(6)
kf

(7)
(8)

25

30

where f, p and nf indices refer to fluid, particle and nanofluid, respectively, is the volume fraction of particles in the suspension, n is the
shape factor, and is the sphericity of particles, which was taken as 0.5 due to the non-spherical shape of the particles. The
thermophysical properties of the base fluid were taken as f=998 kg/m3, f=0.001 Pa.s, cp-f =4178 J/kg.K and kf=0.6 W/m.K while those
of Al2O3 particles were taken as f=3950 kg/m3, cp-f =930 J/kg.K and kf=40 W/m.K.
The Knudsen number, Kn=2/d ( is the mean free path length or the average distance between the molecules of water and equals to 0.31
nm 32, and d is the average diameter of Al2O3 particles and equals to 30 nm) is 0.02 and since it is more than 0.001, the slip boundary
condition is applied at the surfaces of the system. In doing so, the slip velocity was defined as below 33:
uslip Lslip

35

u
n

(9)

where Lslip is the slip length of PDMS surface taken as 57 nm 34 while u n is the gradient of flow velocity along the axis normal to the
surface.

The other boundary conditions consisted of zero pressure at the inlet and flow rates of 10-120 L/min at the outlet (this is because the
suspension is sucked out via the outlet). The inlet temperature of the flow was set to the ambient temperature which was 25.8C. The
value of Q was obtained by dividing the power supplied to the microheater (by reading the voltage and current of the DC power supply)
by the volume of the microheater. The voltage was changed to maintain the temperature of the microheater centre at 54C. The external
surfaces of the system exchanged heat with the ambient air via the free convection mechanism assuming the convection coefficient as
12.5 W/m2.K while the ambient temperature was measured as 25.8C. The heat passed through the walls of the glass slide and PDMS via
the conduction mechanism with kglass=1.05 W/m.K and KPDMS=0.15 W/m.K.

Experimental
Apparatus
10

15

The schematic of the experimental setup is shown in Fig. 1(a). The system consisted of a polydimethylsiloxane (PDMS) block
integrated onto a microscopic glass slide, as shown in Fig.1(b). The PDMS hosted a microchannel and two reservoirs as the inlet/outlet of
the microchannel, as seen in Fig.1(c). The glass slide accommodated a metallic microheater of patterned on its surface as shown in
Fig.1(d). The flow was provided through the microchannel via a syringe pump (Harvard, PHD 2000 Infusion). The syringe pump was
activated in refill mode to supply suction, avoiding the leakage and generation of bubbles within the microchannel. The microheater was
energised via a DC power supply (Gw Instek, GPS-X303 series, Taiwan) and was used to heat up the liquid to the desired temperatures
before entering the microchannel. The temperature measurement along the bottom glass surface that covered the microchannel (Fig. 1 C)
was recorded by an infrared camera (FLIR Systems, ThermoVision A320, Sweden) interfaced with ThermaCAM Researcher software.
The microfluidic device was rotated upside down such that the glass substrate was immediately facing the camera, as shown in Fig.1(a).
20

25

30

35

40

45

50

Fig. 1. Experimental setup of the system: A. Schematic diagram, B. Microfluidic device, C. Microchannel geometrical details, D.
Microheater geometrical details.

Nanofluid preparations

Nanosized Al2O3 powders of the diameters of 50 nm or less were purchased from Sigma-Aldrich, Australia. Al2O3 powder was
suspended in distilled water at room temperature to create a final solution of 1 and 2.5 % w/w concentrations. A liquid surfactant (Triton
X-305) of 0.2% w/w was added to produce a stable suspension; followed by 20 minutes sonication using an ultrasonic bath (Unisonics,
Australia). The size of the nanoparticles was confirmed by scanning electron microscopy (FEI Nova NanoSEM, USA) imaging to be in
the order of less than 50 nm (Fig. 2).

10

15

20

Fig. 2. SEM image of Al2O3 nanoparticles

25

Design Details of Microfluidic system

30

35

40

45

50

The microchannel was fabricated from PDMS using soft photolithography techniques. In doing so, SU8-2050 (Microchem, USA)
layer was spin coated on a 3-inch diameter silicon wafer to produce a 75 m thickness layer. The sample was then exposed to UV light
source using an MA6 mask aligner for 28 seconds, and developed in SU-8 developer to recognize the patterns on the master. A 10 g
mixture of PDMS (SYLGARD 184, Dow Corning, USA) base and curing agent was mixed in 10:1 weight ratio, and degassed to remove
the trapped air bubbles using a vacuum oven. The PDMS mixture was transferred onto the master and cured in an oven with a
temperature of 70C for 20 minutes. The substrate was allowed to cool down for 5 minutes. The PDMS block of 50 mm 10 mm 5
mm (lengthwidthheight) was carefully peeled from the master. The microchannel dimensions were set to 150 m 75 m 40 mm
(widthheightlength). The channels height of 75 m was chosen to allow a free flow of naonoparticles with sizes less than 50 nm and
thus prevent clogging.
The PDMS block was integrated onto a glass slide (Menzel-Glaser, USA) of 60 mm 20 mm 100 m (lenghtwidthheight). The
glass thickness of 100 m was chosen to facilitate the exchange of heat between the liquid and the environment, as well as allowing the
most efficient observation using the infrared camera. The microheater was patterned onto the glass substrate using photolithography
technique. In doing so, thin layers of Au/Cr (150/100 nm) were deposited on the glass surface using electron beam evaporation process.
Next, the substrate was spinned coated with AZ1512 photo-resist using a spinner (Karl Suss RC8). This substrate was backed in an oven
at 90C for 20 minutes, before exposed to UV light through a PDF mask. The exposure left a transparency of the heater design.
Subsequently, the exposed substrate was developed in AZ400K developer and a gold etchant (niro-hydrochloric acid, 1:4 v/v ratio) was
applied to etch away the unwanted gold. The substrate was rinsed and cleaned using acetone, methanol and DI water. A chromium
etchant was then used to remove the excess chromium. The substrate was again cleaned with acetone, methanol, DI water and air dry.
The fabrication process was conducted in a class 1000 clean room.
Microtubes with an internal diameter of ~400 m were placed into the holes punched of the PDMS. These tubes allowed the
interfacing of syringe pumps and samples with the microchannel.

Measurements

10

First, we investigated the thermal performance of our microfluidic system using DI water as the working fluid at a flow rate of 40 l/min.
We placed the IR camera at a distance of 0.3 m from any surface of the microfluidic system which we were interested to measure its
temperature. The voltage of the DC power supply was varied to provide a constant temperature of 54 C at the middle of the inlet
reservoir when seen from the glass side. The temperatures were recorded in 90 s frames using the IR camera and the images were
extracted from the last frame.
Examples of these measurements are shown in Fig. 3(AC). As clearly seen in the images, the IR camera enables us to observe how the
heat is dissipated within the PDMS block, the glass slide, and more importantly throughout the microchannel. We also compared the
measured temperature with results of CFD simulations, as shown in Fig. 3(DF). The colours of the temperature bar are selected to
match with the IR camera settings. However, the exact comparison between the experimental and numerical results is given later.

15

20

Fig. 3. Temperature distribution across the external surfaces of the microfluidic system with water at a flow rate of 40 lit/min. Measured
temperatures using infrared camera: (a) 3D view, (b) side view and (c) bottom view, Calculated temperatures using CFD simulations: (e)
3D view, (f) side view and (g) bottom view.
25

30

35

Next, we moved on to characterise the thermal performance of our microfluidic system at different flow rates and concentrations of
Al2O3 nanoparticles. Fig. 4(AC) show the temperature profiles across the glass slide (the bottom view) after applying only DI water
through the system at different flow rates of 10, 40 and 120 l/min. These figures clearly show the extension of golden (5040 C) and
pink (4030 C) regions throughout the glass surface by increasing the flow rate. Obviously at higher flow rates convection becomes the
dominant mode of heat transfer along the microchannel, as the liquid does not have enough time to lose its internal heat to the
environment via the channel surfaces. Fig. 4(DF) show the temperature profiles across the glass slide (the bottom view) after applying
DI water and water mixed with Al2O3 concentrations of 1% and 2.5% w/w at a constant rate of 40 lit/min.
By careful examination of the temperature contours one can see their changes due to incorporation of nanoparticles. As can be seen, the
increased thermal conductivity of liquid with 2.5% w/w Al2O3 increases the heat conduction along the microchannel, reducing the
temperature gradient along the length. To quantify the thermal impact of the nanoparticles, we investigated the variations of temperature
along the microchannel centreline. In doing so, we defined 38 points along this line and read their temperatures using the images
obtained from the IR camera. Fig. 5(A) shows the variations of temperature along this centreline of the microchannel at different flow
rates of 10, 40 and 120 l/min and different Al2O3 concentrations of 0%, 1% and 2.5% suspended in DI water.

40

45

10

15

20

25

Fig. 4 Temperature distribution across the bottom surface of glass slide: (AC) DI water as the working fluid at different flow rates of 10,
40 and 120 l/min, (DF) DI water with different Al2O3 concentrations of 0%, 1% and 2.5% w/w at a constant flow rate of 40 lit/min.

Results and Discussions

30

35

40

45

50

55

Fig. 5(A) shows the experimental results only. This figure shows that the temperature drop along the microchannel significantly
decreases by increasing the flow rate of the suspension while maintaining constant concentrations of Al2O3, indicating the enhancement
of convection within the microchannel. For example, for the case of DI water, while a temperature drop of 28C was observed along the
abovementioned line at 10 lit/min, it reduced to 25C at 40 lit/min and further reduced to 21C at 120 lit/min. Similar trends can be
seen for other concentrations of Al2O3. Fig. 4(A) also shows that the temperature drop along the microchannel decreases by increasing
the concentration of Al2O3 while maintaining constant flow rates of the suspension, indicating the superior thermal conductivies of
nanofluids. For example at 10 lit/min, while a temperature drop of 28.2C was observed along the abovementioned line for DI water, it
reduced to 27.6C for the 1% Al2O3 suspension and further reduced to 27.4C for the 2.5% Al2O3 suspension. Interestingly, the
effectiveness of nanofluids was more tangible at higher flow rates. For example at 120 lit/min, while a temperature drop of 22.7C was
observed along the abovementioned line for DI water, it reduced to 21.6C for the 1% Al2O3 suspension and further reduced to 19.5C
for the 2.5% Al2O3 suspension.
Fig. 5(B-C) compare the variations of experimental and numerical temperatures along the line which connects the centres of the inlet and
outlet reservoirs. Results show a good agreement between the experimental and numerical results. In general, the numerical temperatures
were lower at the inlet of the microchannel while higher at the outlet of the microchannel.

10

15

20

25

30

35

Fig. 5. Variation of temperature at the bottom surface of the microfluidic system along the line connecting the centres of inlet and outlet
reservoirs. A. Experimental temperatures at the flow rates of 10, 40 and 120 lit/min, and different w/w Al2O3 concentrations of 0%, 1%
and 2.5%. B-D. Experimental versus numerical temperatures at 10, 40 and 120 lit/min for B. water, C. water plus 1% w/w Al2O3, D.
water plus 2.5% w/w Al2O3.
Next, we investigated the variations of heat flux along the line connecting the centres of the inlet and outlet reservoirs using our CFD
model, as shown in Fig.6. As before, we examined different flow rates of 10, 40 and 120 lit/min and different Al2O3 concentrations of
0%, 1% and 2.5% suspended in DI water. According to the results, at a low flow rate of 10 lit/min, the addition of nanoparticles did not
have a considerable effect on heat flux curves and could slightly improve the heat flux at the inlet of the microchannel. Increasing the
flow rate to 40 lit/min, the average heat flux increased by 6.5% for the 1% Al2O3 suspension and by 13.5% for the 2.5% Al2O3
suspension. Similarly, increasing the flow rate to 120 lit/min, the average heat flux increased by 10.8% for the 1% Al2O3 suspension
and by 23.5% for the 2.5% Al2O3 suspension.

40

45

Fig. 4. Heat flux distribution

50

Fig. 6. Variations of heat flux at the bottom surface of the microfluidic system along the line connecting the centres of inlet/outlet
reservoirs

Conclusion

10

The work is a comprehensive study on the thermal characteristics of a micofluidic system with and without Al2O3 nanoparticles.
According to our observations, at low flow rates of 10 lit/min, no significant difference was observed between the base fluid and
nanofluids as confirmed by the numerical simulations. In contrast, at higher flow rates of 120 lit/min, the exchange of heat for the liquid
with nanoparticles was enhanced. As nanoparticles are becoming more readily accessible, their application in microfluidics is becoming
increasingly economical. Further improvement of the system is needed to account for a number of factors, such as the investigation of
different sizes and types of nanoparticles and the modification of the microchannel design in order to obtain higher heat transfer
efficiency.
Notes and references
a RMIT University, School of Electrical and Computer Engineering,
Melbourne, Victoria Australia. Fax: 992 52007. Tel: (03) 992 53254.
E-mail: pyshar.yi@gmail.com, khashayar.khoshmanesh@rmit.edu.au and kourosh.kalantar@rmit.edu.au
b Centre for Intelligent Systems Research, Deakin University, Victoria, Australia. Tel: (03) 5227 330.

15

Electronic Supplementary Information (ESI) available: [Materials and methods and Temperature profiles along the microchannel
vs/flow rates] see DOI: 10.1039/b000000x/

20

25

1.

A. K. Agarwal, L. Dong, D. J. Beebe and H. Jiang, Lab on a Chip, 2007, 7, 310-315.

2.

D. Erickson, D. Sinton and D. Q. Li, Lab on a Chip, 2003, 3, 141-149.

3.

K. V. Wong and a. M. J. Castillo, Advanes in Mechanical Engineering, 2009, 1-9.

4.

W. Yu and S. U. S. Choi, Journal of Nanoparticle Research, 2003, 5, 167-171.

5.

S. M. S. Murshed, K. C. Leong and C. Yang, International Journal of Thermal Sciences, 2005, 44, 367-373.

6.

S. Movahed and D. Li, Electrophoresis, 2011, 32, 1259-1267.

7.

C. L. Altan and S. Bucak, Nanotechnology, 2011, 22, 285713

8.

S. Ozerinc, S. Kakac and A. G. Yazicioglu, Microfluidics and Nanofluidics, 2009, 8, 145-170.

9.

Liu, K. V, Choi, U.S and K. E. Kasza, Argonne National Laboratory Report,, 1988, ANL-88-15.

10. S. Kondaraju, E. K. Jin and J. S. Lee, Microfluidics and Nanofluidics, 2011, 10, 133-144.
11. S. U. S. Choi, Developments and Applications of Non-Newtonian Flows, 1995, 231, ASME, 99-105.
30

12. S. H. Hashemabadi, S. M. Peyghambarzadeh, M. S. Jamnani and S. M. Hoseini, Applied Thermal Engineering, 2011, 31, 1833-1838.
13. O. Manca, Y. Jaluria and a. D. Poulikakos, Advanes in Mechanical Engineering, 2010, 1-2.
14. S. K. Das, N. Putra, P. Thiesen and W. Roetzel, Journal of Heat Transfer-Transactions of the Asme, 2003, 125, 567-574.
15. B. Wang, H. Li and a. X. Peng, Journal of Thermal Science, 2002, 11, 193-288.
16. Y. M. Xuan and Q. Li, International Journal of Heat and Fluid Flow, 2000, 21, 58-64.

35

17. H. Xie, J. Wang, T. Xi, Y. Liu and a. F. Ai, Journal of Materials Science Letters, 2002, 21, 1469-1471.
18. H. Q. Xie, J. C. Wang, T. G. Xi, Y. Liu, F. Ai and Q. R. Wu, Journal of Applied Physics, 2002, 91, 4568-4572.
19. C. H. Chon, K. D. Kihm, S. P. Lee and S. U. S. Choi, Applied Physics Letters, 2005, 87, 153107.
20. S. M. Peyghambarzadeh, S. H. Hashemabadi, M. S. Jamnani and S. M. Hoseini, Applied Thermal Engineering, 2011, 31, 1833 - 1838.
21. S. Sonawane, K. Patankar, A. Fogla, B. Puranik, U. Bhandarkar and S. S. Kumar, Applied Thermal Engineering, 2011, 31, 2841-2849.

40

22. H. A. Mintsa, G. Roy, C. T. Nguyen and D. Doucet, International Journal of Thermal Sciences, 2009, 48, 363-371.
23. T.-P. Teng, Y.-H. Hung, T.-C. Teng, H.-E. Mo and H.-G. Hsu, Applied Thermal Engineering, 2010, 30, 2213-2218.
24. C. T. Nguyen, G. Roy, C. Gauthier and N. Galanis, Applied Thermal Engineering, 2007, 27, 1501-1506.
25. M. P. Beck, T. F. Sun and A. S. Teja, Fluid Phase Equilibria, 2007, 260, 275-278.
26. C. Popa, S. Fohanno, C. T. Nguyen and G. Polidori, International Journal of Thermal Sciences, 2009, 49, 901-908.

45

27. C. H. Li and G. P. Peterson, Journal of Applied Physics, 2007.

28. S. Shaikh, K. Lafdi and R. Ponnappan, Journal of Applied Physics, 2007.
29. J. Atencia and D. J. Beebe, Nature, 2005, 437, 648-655.
30. A. Gunther, S. A. Khan, M. Thalmann, F. Trachsel and a. K. F. Jensen, Lab on a Chip, 2004.
31. P. S. Lee, S. V. Garimella and D. Liu, International Journal of Heat and Mass Transfer, 2005, 48, 1688-1704.

50

32. H. A. Mohammed, P. Gunnasegaran and N. H. Shuaib, International Communications in Heat and Mass Transfer, 2010.
33. H. A. Mohammed, G. Bhaskaran, N. H. Shuaib and R. Saidur, Superlattices and Microstructures, 50, 215-233.
34. C.-S. Jwo, L.-Y. Jeng, T.-P. Teng and C.-C. Chen, Journal of Alloys and Compounds, 2010.
35. W. Y. Lai, P. E. Phelan, S. Vinod and R. Prasher, Convective heat transfer for water-based alumina nanofluids in a single 1.02-mm tube, 2008.
36. R. Chein and J. Chuang, International Journal of Thermal Sciences, 2007, 46, 57-66.

55

37. K. Khoshmanesh, J. Akagi, S. Nahavandi, J. Skommer, S. Baratchi, J. M. Cooper, K. Kalantar-Zadeh, D. E. Williams and D. Wlodkowic, Analytical
Chemistry, 2011, 83, 2133-2144.