Sensing with Phosphor-doped PDMSF. Sabri*, K. Lynch*, R. Wilson*, S. W. Allison†

*U. S. A, Department of Physics, University of Memphis, Memphis, TN., †U. S. A., Emerging Measurements, Collierville, TN.

Keywords: Phosphor Thermometry, flexible electronics, doped PDMS Sensing.

AbstractThermometry with thermographic phosphors usually involvessuperficial application of a thin coating to the surface of interest. In contrast, the subject here concerns a thicker luminescent target consisting of a flexible phosphor-dopedpolymer, from the polydimethylsiloxane (PDMS) family. The purpose is to produce a flexible attachable/detachable material which can be used for temperature measurement of the underlying surface it contacts. PDMS samples loaded with 1, 5, 15, and 50% of the phosphor La2O2S:Eu were fabricated and tested. For all doping levels, detectable and useful fluorescence signals were readily attained for 365 and 405 nm incident excitation. The PDMS host appeared not to affect the spectral properties of the incorporated phosphor. Additional characterization of optical, thermal and mechanical properties (stress/ strain behavior and temperature dependent compression response) as a function of dopant concentration described here validate this as a temperature sensing material.

1 Introduction

Luminescence from many phosphors can be exploited for temperature measurement [1]. Usually this involves incorporating the fluorescent powder into a coating that is applied to the surface of interest. For the case of ceramic phosphors, the coating may be produced using the material itself by plasma spray, electron beam deposition, etc. In other instances, the material may be mixed with a binder and applied as paint. In these examples, the coatings are hard and thin. This technique has received considerable attention particularly for high temperature situations such as encountered in piston and turbine engines [2,3,4].

In contrast, the subject of the present effort concerns a thicker luminescent target consisting of a phosphor impregnated polymer, specifically Sylgard 184 which is from the polydimethylsiloxane (PDMS) family and the characterization of its thermal and mechanical properties. PDMS is a clear and inert material often used for a wide variety of situations including optical, biomedical and aerospace applications [5,6,7,8]. A near term goal is to produce a flexible material which can be used for temperature measurement of an underlying surface to which it is attached. But also, this effort serves the larger goal which is to investigate how the incorporation of phosphor particles may

be used as a tool for characterizing, understanding, diagnosing and improving the host material.

In previous work, Aizawa [9] impregnated silicone with ruby powder. They tested silicone sheets 1 mm thick by 10 mm diameter with dopant levels of 10, 20, 30, 40, and 50% by weight. They utilized the fluorescence phase shift approach, demonstrating a temperature dependent phase shift from -3 to 77 °C. The time response to temperature of the underlying copper block matched that of a Type K thermocouple.

Many luminescent material choices are available for thermometry. Aizawa [9] chose ruby from several other Cr-doped materials. Brübach [4] lists over 100 candidate phosphors that have received attention for phosphor thermometry. The best choice for a given application depends on a wide variety of factors. The present work involves the phosphor La2O2S:Eu. The deep red emission from ruby is easily excited in the visible. This contrasts with La2O2S:Eu which requires deep blue or near ultraviolet for efficient excitation where the sources are less bright and more expensive. However, temperature sensitivity of this Eu-based phosphor is greater. Emission from the 5D2 state produces a number of narrow bands in the blue and green for which the decay time changes about 5% per °C at room temperature. For ruby it is less than 1%. This brief comparison does not exhaust the considerations that are required for choosing a luminescent material for a given application. It is expected that both of these as well as other phosphor materials should be scrutinized and examined for future work in this area.

2 Sample PreparationPhosphor particles (Phosphor Technology SKL63) were combined with the prepolymer and crosslinker of Sylgard 184 (Dow Corning) prior to the curing stage of the two-part elastomer according to the method described previously [8]. Doped samples with weight concentrations of 1, 5, 15, and 50% were prepared in sheets and cut to specific final geometries appropriate for each characterising methoddescribed below.

3 Stress-Strain Response of Phosphor-Doped PDMSStress-strain behaviour of phosphor-doped Sylgard 184 was tested under ambient conditions by means of a Mark-10bench-top tensile tester. Samples were strained at a rate of 50mm/min and results are shown in Figure 1 for the ratios tested. The highest concentration of phosphor particles (50%) leads to a compound material with an increased stiffness

1

compared to lower dopant concentrations where a significant degree of elasticity can be observed. At 50% concentration a somewhat linear behaviour is observed while for lower concentrations classic elastomeric behaviour resembling neat Sylgard 184 is witnessed. The overall rupture load does not appear to be significantly dependent on the concentration amount for the percentages tested here. Stress-strain behaviour for higher concentrations remains to be investigated.

Figure 1: Room temperature stress-strain behaviour of phosphor-doped Sylgard 184.

Temperature-Dependent Compression Behavior of Phosphor-Doped PDMS: The samples were placed individually on an INSTEC mK-1000 temperature control stage, positioned on the base of a Mark-10 ESM301 tensile tester. The base temperature was held at a specific temperature for 10min before compression data was collected at a rate of 1 mm/min and data collection at a rate of one measurement per second.. Thermal equilibrium may not have occurred at the time of data collection. The compression experiment was manually terminated at a preset value of 8 N.Figure 2 shows compression responses at three specific temperatures 100, 25, and -40°C. Overall, the family of curves collected at each temperature suggest that the doped polymer is less stiff at higher temperatures (100°C) and stiffens significantly at lower temperatures (-40 °C) which is the expected temperature response from the encapsulating PDMS. Therefore, it is hypothesized that mechanical response for the concentrations tested here are primarily dominated by the polymer and not as much by the dopants. Individual compression-travel curves for each dopant concentration at each temperature are however distinguishable reflecting a temperature sensitive response in each case.

Figure 2: Compression-travel behaviour of phosphor-doped PDMS at 100°C, 25°C, and -40°C.

4 Sample Temperature Dependence

The goal of a temperature dependence test was to establish that luminescence is detectable and produces the characteristic expected response.

The test arrangement for this is shown in Figure 3 which illustrates a sample compartment with a Peltier heater/cooler into which the sample was placed. It was in intimate contact with a copper plate on the heater/cooler surface and secured with adhesive taped on its edges. A type K thermocouple was situated into a small hole in the copper for establishing the

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.60

1

2

3

4

5

Stre

ss (N

/mm

2 )

Strain

1%5%15%50%

T= 100 C

T=-40 C

T= 25 C

2

temperature. A Thorlab model M365L2 LED which emits 365 nm was directed at the target. A BNC Model 575 pulse/delay generator provided electrical impulses of 5 to 7 volts to drive the LED. A lens gently focussed the LED output onto the sample. The fluorescence thus generated passed through an intervening filter, Andover 510FS10-25. Its peak transmission at 510 nm and 10 nm width (full width half maximum) adequately permits several stark components of the 5D2-7F3 transition to reach the photomultiplier tube (PMT) for detection, Hamamatsu H5783-01.

Figure 4 shows the complete test layout. The PMT output connected to a Tektronix 2012B digital oscilloscope where a trigger synchronous with the LED pulse from the pulse generator triggered the display.

Figure 4: Data acquisition and display block diagram.

The amplitude analog-to-digital conversion resolution of this Oscilloscope is 8 bits. The digitized signal consists of 2500 data pairs. As visualized by Figure 3, a small patch of doped-PDMS was placed in the compartment. For this temperature dependence test, the doping level was 50%.

Figure 5 depicts a typical fluorescence waveform. There is a rise time that characterizes the emission. When the pulsewidth is sufficiently long, the signal reaches a limiting value. When the LED terminates, immediately the fluorescencedecreases. The temporal dependence is commonly expressedas = +sensitive function of temperature. In this case C is a constant signal that as practice teaches may result from constant background light, detector noise, electronic direct currentoffset or possibly other sources.

Figure 5: LED-excited fluorescence rise and decay at 9.2 °C.

Figure 6 shows the decay portion of several signals acquired from 9.5 to 54.2 °C. It is seen that I0 is also sensitive to temperature. This feature is not so pronounced when illumination is with an excitation pulse of width much less than the decay time. Overall, the observed temperature dependence was as expected.

Figure 6: Decay waveforms versus temperature.

5 Phosphor-doped PDMS and Phosphor Coating Comparison

Figure 3: Illustration of Sample Compartment.

3

A careful comparison was made of fluorescence from a phosphor-doped PDMS sample and a thin layer of phosphor coated onto a copper plate. The latter sample was 15% by weight and about 2 mm thick. It was placed in the sample compartment and allowed to equilibrate with ambient temperature. Eight waveforms were acquired in succession which were averages of 128 signals. Then these were subsequently averaged in software, thus improving the signal-to-noise. The temperature was 20.6 °C. The sample was then removed so that the phosphor coated surface in the sample compartment could be illuminated. After about 8 minutes, the temperature was 20.8 °C. The phosphor coating is a mixture of phosphor and VHT Flameproof™ SP115 Satin Clear. This is a high temperature paint, colorless, sold in auto parts stores. The signal from the coating was initially about a factor of two higher. Subsequently the gain on the photomultiplier detector was adjusted to obtain a signal approximately the same amplitude as the doped PDMS sample. This is to eliminate the possibility of a nonlinear response by the detector. The two signals are shown in Figure 7.

Figure 7: Comparison signals from doped PDMS and phosphor coating.

It is seen that the rise times are slightly different. However, in this view, the fluorescence decays appears almost identical. This suggests that the PDMS matrix has little physical or chemical effect on the phosphor. This is all the more striking since the phosphors used in the PDMS and the coating were from different batches. The coating phosphor was Lot # 23015.

6 Absorption Measurements

For all but the thickest sample of 50% doping both 365 nm and 405 nm produced luminescence that emerged from the side opposite of illumination, as seen in Figure 8. Thus temperature may be measured from either side of the material.

Figure 8: 15% sample with light emerging from back of side illuminated at 405 nm.

In order to assess what thicknesses for which wavelengths may be viable, absorption measurements are of value. Absorption behaviour in the ultraviolet and visible ranges were made by means of a scanning Beckman spectrometer such that prepared samples were placed inside a quartz cuvette and scanned at a rate of 500nm/min. The pure phosphor powder was compressed into a quartz cuvette and scanned separately in order to establish the effect of the surrounding encapsulating polymer on the absorption/ emission behaviour of the phosphor particles. The results are shown in Figure 9 and the appearance of bands associated with La2O2S:Eu can be clearly seen. Graphs presented include background subtraction.

Figure 9: Absorption spectrum UV-Vis range.

7 Toroidal Inductor Measurement

An example temperature measurement application was selected for testing the phosphor-doped PDMS as a temperature sensor. It involved the surface of the magnet of a toroidal inductor on a computer motherboard. A thin (1< 1mm) rectangular 5% phosphor-impregnated patch was attached by tape at its edges. It was considered that the signal would have been much brighter using the 50% sample. Since this was already tested earlier, it was thought that using a more dilute material would be instructive.

For this test demonstration, a 405 nm emitting diode laser (Star 405F100 from Roithner) was directed onto the phosphor patch as seen on the bottom frame of Figure 10 bottom shows this in the test setup. With the room lights off, the top frameof Figure 10 shows where the laser strikes the inductor. To

4

the human eye it appears pink since what is seen is both the diffuse blue from the laser and the longer emission of the fluorescence that is comprised of many colors. The bottom is the nearly identical scene with lights on The fiber probe aimed at the target contains a central fiber that convey the excitation and it is surrounded by receiving fibers that transmitted received fluorescence to a PMT detector.. It is situated about 1 ½ inches from the fluorescing target. The fluorescing spot is about 3 mm diameter, at least. The laser was pulsed with a duration of 40μs and repetition rate of 10 pulses per second. The applied voltage was 5.5 volts. Full power, 100 mwatts, is at 6.0 volts according to the specification sheet. Thus the energy per pulse is estimated to be a few microjoules per pulse. It should be noted that the signal is the average of 128 sequential signals, i.e. over about 12.8 seconds.

An ambient reading was acquired one minute before the power to the computer was switched on. Readings were taken every minute from 1 to 8 minutes and then less frequently after that. The temperature rises from about 23 to 37 °C. After 10 minutes, the power was turned off. The table shows temperature versus time. Figure 11 is a plot of T

determination was taken from a different batch of phosphor and might deviate slightly from this material. A careful calibration of this particular sample will be performed later. The correction of the curve slightly up or down will not affect the conclusions here.

Figure 10: Toroidal inductor temperature measurement.

Figure 11: Measured temperature versus time.

8 Concluding Remarks

The phosphor-doped sensing material is chemically stable and shows promise as a detachable and reusable temperature sensor. The issues of contact resistance, thermal conductivity, and thermal equilibrium of course need to be considered in future work. Depending on the application, a thermal gradient across the material may be a problem or a feature to exploit for heat flux determination. Another distinguishing feature of the subject elastomer material is that it is can be easily stressed and/ or strained incrementally.Thus, future research will address ways to exploit the fluorescence to indicate this as well as temperature. The method described here specifically refers to incorporation of phosphor particles in Sylgard 184 host but can be easily applied to other elastomeric and non-elastomeric transparent encapsulating polymers.

Finally, other material candidates that may incorporate phosphor materials are aerogels. The authors have some laboratory tests relating to this to be reported at a future time.

Acknowledgements

S. W. Allison wishes to acknowledge the assistance of his Emerging Measurements colleagues M. A. Akerman and D. L. Beshears with the toroidal inductor measurement.

References

[1] S. W. Allison and G. T. Gillies, “Remote thermometry with thermographic phosphors: Instrumentation and applications,” Review of Scientific Instruments, 68, 2615–2650 (1997).

[2] A. H. Khalid and K. Kontis, “Thermographic Phosphors for High Temperature Measurements: Principles, Current State of the Art and Recent Applications,” Sensors, 8, 5673–5744 (2008).

5

[3] M. Aldén, A. Omrane, M. Richter, and G. Särner, “Thermographic phosphors for thermometry: A survey of combustion applications,” Progress in Energy and Combustion Science, 37, 422–461 (2011).

[4] J. Brübach, C. Pflitsch, A. Dreizler, and B. Atakan, “On surface temperature measurements with thermographic phosphors: A review,” Progress in Energy and Combustion Science, 39, 37–60 (2013).

[5] F. Sabri, M. E. Sebelik, R. Meacham, J. D. Boughter Jr, M. J. Challis, and N. Lenventis, “In Vivo Ultrasonic Detection of Polyurea Crosslinked Silica Aerogel Implants” published 20 Mar 2012 | PLOS ONE10.1371/journal.pone.0033242

[6] F. Sabri, T. Werhner, J. Hoskins, A.C. Schuerger, A.M. Hobbs, J. A. Barreto, D. Britt, and R. A. Duran, “Thin film surface treatments for lowering dust adhesion on Mars Rover calibration targets,” Advances in Space Research, Vol.41, No.1, 2008, pp. 118–128.

[7] F. Sabri, J. Marchetta, and K. Smith, “Thermal Conductivity Studies of a Polyurea Cross-linked Silica Aerogel-RTV 655 Compound for Cryogenic Propellant Tank Applications in Space,” Acta Astronautica, Vol.91, No. October-November, 2013, pp.173-179.

[8] Sabri F, Marchetta JG, Sinden-Redding M, Habenicht JJ, Chung TP, et al.(2012) Correction: Effect of Surface Plasma Treatments on the Adhesion of Mars JSC 1 Simulant Dust to RTV 655, RTV 615, and Sylgard 184. PLoS ONE 7(11): DOI: 10.1371/journal.pone.0045719

[9] H. Aizawa, M. Sekiguchi, T. Katsumata, S. Komuro, and T. Morikawa; “Fabrication of ruby phosphor sheet for the fluorescence thermometer application” Review of Scientific Instruments 77(4) 2006.

6