Santa Clara University

mechanical banner

Temperature Characterization in Micro Channels

This research is a partnership of:
Santa Clara University (SCU)
Hewlett Packard-Aguadilla (HP-Aguadilla)
University of Puerto Rico at Mayaguez (UPRM)
with support from the National Science Foundation (NSF)


Determination of temperature variation in space and time in micro-channels after boiling events.


Micro-chambers are subjected to a boiling process, after several boiling events there could be an energy accumulation in the chamber fluid and along the micro-channels feeding the micro-chamber. This energy accumulation could be high enough to initiate unexpected boiling events in the micro-chamber or micro-channels. Since this is an event of the order of micro-seconds, Laser Induced Fluorescence Thermometry (LIFT) will be used to follow those bulk fluid temperature changes. However, since every fluid has different physical properties, LIFT has to be calibrated for each fluid.

LIFT, a non-intrusive high spatial resolution temperature measurement:

The need of development of non-contact sensors to measure temperature of in-flight droplets is evident for droplets in the order of hundreds of micro meters or less and for low emissivity fluids. A few studies have been reported in the literature addressing experimental studies of temperature measurements of in-flight droplets. However, these studies were for droplets with diameters of millimeters. Over the last 10 years a few researchers envisioned the importance of remote temperature sensors for small droplets. Although infrared thermometry is the most common way to make remote temperature measurements, this method is subject to surfaces with relative high emissivity and slow movement. Rainbow thermometry is another technique based on the refractivity of the light due to temperature, here light should pass through a transparent medium and then the angle of refractivity is measured as function of temperature. Meanwhile, laser induced fluorescence (LIF) has been used mainly in combustion systems. LIF has the advantage that it reacts quickly and temperature can be measured for very dynamic systems and it is a technique based on imaging.

At UPRM, LIF thermometry has been used to measure temperature of 50 to 200 mm droplets formed with ethanol alcohol and water. These droplets were formed with a monodispersed atomizer using the Rayleigh instability. The first sensor developed by UPRM was made with the purpose of measuring the temperature of falling ethanol droplets in a nitrogen atmosphere. A calibration curve was obtained for droplets formed with ethanol doped with Pyrene in a 0.1% mass based concentration. Figure 1 shows the existent answer of the fluorescence emission to temperature changes. Final calibration curve is based in the emission of two wavelength ranges named monomer (IM) from 360nm to 408nm and excimer (IE) from 408nm to 600nm. These emissions are related to the temperature of the solution as shown in Figure 2. With this calibration curve it was possible the measurement of temperature of falling droplets in a quiescent atmosphere as shown in Figure 3. In the same figure, observations are compared with the solution from a numerical model for an evaporating falling ethanol droplet.


Figure 2: Calibration curve or Fluorescence Remote Thermometer for the Ethanol based solution



Figure 3: Comparison of experimental temperature measurements against theoretical temperatures at several distances

A second remote sensor was developed for aqueous solutions using initially pyrene as the fluorescence dye. Pyrene is an effective fluorescence probe due to its long lifetime for formation of pyrene monomers (t = 450 nanoseconds and quantum yield of 0.60 in cyclohexane) and efficient of excited dimers, “excimers” for short. However, pyrene has a poor solubility in water so the addition of a surfactant for the formation of micelles to solubilize the pyrene is necessary. Surfactant molecules are characterized by possessing both hydrophobic and hydrophilic moieties, and are often called amphiphiles. Usually the hydrocarbon part is a long-chain hydrocarbon (eight C atoms or more), while the hydrophobic portion of the molecule may be a group of ethylene oxide units, as in the Triton series of surfactants, an anionic group, as in sodium lauryl sulfate, NaLS, or a cationic group, as in cetyltrimethylammonium bromide, CTAB. Above a certain concentration, called the critical micelle concentration, the surfactant molecules group together forming micelles. In this work the surfactant used was cetyldimethylbenzylammonium chloride “CDBAC” which is a cationic surfactant. Nevertheless, the pyrene shows enhanced solubility in the surfactant molecule due to affinity to the non-polar alkyl groups present in the surfactant molecules. In addition, the proximity introduced by micelles is well illustrated by efficient protection from the oxygen molecule and formation of excimer of arenes in micellar solution. Micelles provided by the CDBAC could improve fluorescence emission avoiding natural fluorescence quenching due to oxygen in the solutions.

The in-situ relationship between fluorescence emissions and temperature for a pyrene/CDBAC/Water solution at a concentration of 5 mM of pyrene and 5mM of surfactant is shown in Figure 4. This in-situ calibration curve shows two well defined regions, one region is the 23°C until approximately 50°C where the Ln (IE/IM) ratio increase when the temperature increase, and the other region from 45 until 60°C where the Ln (IE/IM) changes the slope and the emission ratio decreases. The maximum temperature in the calibration curve was limited by the boiling point of the solution and the maximum temperature that could be obtained from droplets ejected from the atomizer. With the in-situ calibration curve at hand, reliable temperature measurements for in-flight droplets can be conducted. In this case the solution was preheated in the atomizer to set the initial thermal state of the droplets. Three different exit temperatures were used 30, 40 and 50°C. For all three temperatures the remaining exit conditions were the same including pressure equal to 34.473 kPa (5 psi), frequency of 7.0 KHz, and the velocity which was about 6.67m/s. The ambient conditions for each run were: Tdb= 24°C, Twb= 19°C, and the atmospheric pressure of 101325 Pa. The droplets were falling in the downward vertical direction, and a reference distance was taken at the nozzle tip. The temperature at the exit in the nozzle was controlled by a heat tape and it was stabilized for approximately 20 minutes for each temperature assuring in this manner the steady state and that there are not fluctuations in the exit temperature. The temperature measurements are along the jet and it allowed to track the thermal history of a droplet along the traveling path.

Figure 5 shows the resulting curves and in general it shows a decreasing temperature pattern as the droplets fly away from the nozzle. The total decrease in temperature along the 50 mm of traveling distance for an initial temperature of 30, 40 and 50°C is of 2.1, 4.24 and 5.01°C respectively.

The validity of the present non-invasive LIF thermometry technique was verified by comparing the experimental results with a numerical simulation for an-flight stream of droplets. The mathematical model for in-flight droplets was based on a one-dimensional Lagrangian viewpoint as presented by Gonzalez and Black which considers the droplets to be spherical in shape, neglects collision or coalescence between droplets, and the properties in the moist air are considered as those of dry air. It can be observed from the figure that the numerical simulation shows a good agreement with the experimental data for all three cases with a minor exception for the injection at a temperature of 50°C. Since the model considers that the droplets are completely spherical when they leave the nozzle, the validation exercise was limited to a starting point 10mm from the nozzle tip when the droplets were observed to be completely spherical.



Figure 6. Laser Induced Fluorescence Thermal sensor for aqueous solutions.

Currently, the Laboratory is being set to use some of the developed calibrated curves in micro-channels. As progress is made, it will be posted.