Jayna J. Shah

Microwave dielectric heating of fluids in an integrated microfluidic device

The ability to selectively and precisely control the temperature of fluid volumes ranging from a few microliters to sub-nanoliters in microfluidic networks is vital for a wide range of applications in micro total analysis systems (μTAS). In this work, we characterize and model the performance of a thin film microwave transmission line integrated with a microfluidic channel to heat fluids with relevant buffer salt concentrations over a wide range of frequencies. A microchannel fabricated in poly(dimethylsiloxane) (PDMS) is aligned with a thin film microwave transmission line in a coplanar waveguide (CPW) configuration. The electromagnetic fields localized in the gap between the signal and ground lines of the transmission line dielectrically heat the fluid in the selected region of the microchannel. Microwave S-parameter measurements and optical fluorescence-based temperature measurements are used with a theoretical model developed based on classical microwave absorption theory to fully characterize the temperature rise of the fluid. We observe a 0.95 °C mW−1 temperature rise at 15 GHz and confirm that the temperature rise of the fluid is predominantly due to microwave dielectric heating.

Generalized temperature measurement equations for Rhodamine B dye solution and its application to microfluidics.

Temperature mapping based on fluorescent signal intensity ratios is a widely used noncontact approach for investigating temperature distributions in various systems. This noninvasive method is especially useful for applications, such as microfluidics, where accurate temperature measurements are difficult with conventional physical probes. However, the application of a calibration equation to relate fluorescence intensity ratio to temperature is not straightforward when the reference temperature in a given application is different than the one used to derive the calibration equation. In this report, we develop and validate generalized calibration equations that can be applied for any value of reference temperature. Our analysis shows that a simple linear correction for a 40 degrees C reference temperature produces errors in measured temperatures between -3 to 8 degrees C for three previously published sets of cubic calibration equations. On the other hand, corrections based on an exact solution of these equations restrict the errors to those inherent in the calibration equations. The methods described here are demonstrated for cubic calibration equations derived by three different groups, but the general method can be applied to other dyes and calibration equations.

Microwave-induced adjustable nonlinear temperature gradients in microfluidic devices

We describe on-chip microwave generation of spatial temperature gradients in a polymeric microfluidic device that includes an integrated microstrip transmission line. The transmission line was fabricated photolithographically on commercially available adhesive copper tape. The fluid temperature during microwave heating was measured by observing the temperature-dependent fluorescence intensity of a dye solution in the microchannel. Large interference effects, which were produced by superposition of a sinusoidal and two exponential temperature distributions, were measured at 12 GHz and 19 GHz. Temperature extremes of 31 °C and 53 °C at the minimum and maximum of the sinusoid were established within 1 s. The sinusoid also produced a quasilinear temperature gradient along a 2 mm distance with a slope of 7.3 °C mm−1. This technique has the potential to benefit many biological, chemical and physical applications requiring rapid temperature gradients.

Microwave Power Absorption in Low-Reflectance, Complex, Lossy Transmission Lines

Simple sets of equations have been derived to describe the absorption of microwave power in three-region, lossy transmission lines in terms of S-parameter reflection and transmission amplitudes. Each region was assumed to be homogeneous with discontinuities at the region boundaries. Different sets of equations were derived to describe different assumptions about the amplitudes of the reflection coefficients at the different boundaries. These equations, which are useful when interference effects due to multiple reflections are small, were used to analyze S-parameter measurements on a transmission line that had a microfluidic channel in its middle region. The channel was empty for one set of measurements and filled with water for a second set of measurements. Most of the reflection assumptions considered here produced similar results for the fraction of the applied microwave power that was absorbed by a water-filled microchannel. This shows that the absorbed power is relatively insensitive to the reflection details as long as energy is conserved in the analysis. Another important result of this work is that the difference between the power absorbed in a water-filled channel and the power absorbed in the same empty channel can be a poor predictor of the power absorbed in the water in the presence of competing absorption processes such as absorption by the transmission-line metal.

Microwave Dielectric Heating of Fluids in Microfluidic Devices

The science of designing, manufacturing and formulating processes involving fluidic devices having dimensions down to micrometers is known as microfluidics. In the last two decades, in the areas of chemical and biochemical sciences, there has been a great interest towards using microfluidic systems, which are popularly known as micro-total analysis systems (μ-TAS) (Manz et al., 2010) or lab-on-a-chip systems. These systems improve analytical performance and also facilitate incorporating various functions of distributed systems on a single-chip instead of having a separate device for each function (Reyes et al., 2002). Temperature control inside microfluidic cells is often required in a variety of on-chip applications for enhanced results, without significantly affecting the temperatures of other building blocks of the μ-TAS. This is a big challenge because the microfluidic devices on the chip need to be selectively heated.