Thermal stability improvement of sprayable fast-responding pressure-sensitive paint for measurement above 100°C

Di PENG, Feng GU, Zhe ZHONG, Yingzheng LIU,*

a Key Lab of Education Ministry for Power Machinery and Engineering, School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China

b Gas Turbine Research Institute, Shanghai Jiao Tong University, Shanghai 200240, China

KEYWORDS Dynamic response;Polymer-luminophore interaction;Pressure sensitivity;Pressure-sensitive paint;Thermal degradation

Abstract The thermal stability of sprayable fast-responding Pressure-Sensitive Paint (fast PSP)was investigated to explore the possibility for application in turbomachinery and hypersonic research with temperature above 100°C.The first part of the study focused on a widely-used Polymer Ceramic PSP (PC-PSP). The effects of thermal degradation on its key sensing properties,including luminescent intensity,pressure sensitivity and response time,were examined for a temperature range from 60 to 100°C.Severe degradation in intensity and pressure sensitivity was found as temperature reached 70°C or higher,which would cause failure of PSP application in these conditions. Subsequently, a fast-responding Mesoporous-Particle PSP (MP-PSP) was developed which did not show degradation effects until 140°C. The greatly improved thermal stability of MP-PSP was attributed to: selection of polymer with higher glass transition temperature (polystyrene) to delay the saturation effect of oxygen quenching as temperature increased;porous and hollow structure of particles for luminophore deposition that minimizes polymer-luminophore interaction.This new paint formulation has significantly raised the upper temperature limit of fast PSP and offers more opportunities for applications in harsh environment.

1. Introduction

Fast-responding Pressure-Sensitive Paint(fast PSP)is an optical measurement technique capable of providing surface pressure field with high spatial and temporal resolution. Fast PSP typically shows frequency response of several kHz or higher,and has been widely used in unsteady flow diagnostics.1,2In recent years, the application of fast PSP has been extended from conventional wind tunnels to facilities with rather harsh environments,3including hypersonic tunnels4,5and turbomachines.6,7A major challenge of those test conditions was the significant temperature rise on the coated model, where the surface temperature could easily exceed 50°C (i.e. the upper limit of common commercial PSPs).8This would lead to: (A)large temperature-induced error in PSP measurements; (B)potential degradation in the sensing properties of PSP. The former issue could be resolved through adoption of in-situ calibration,9or by temperature correction based on temperature data measured by other techniques, such as infrared imaging6,10or temperature-sensitive paint.7,11The later issue,which has not received sufficient attention so far,is the primary focus of the current work.

There have been only a few studies about the thermal stability of PSP, and most of them focused on conventional polymer-based PSP with limited dynamic response. Mebarki reported the degradation of pyrene-based PSP at 50°C and above,12and Basu et al. found that this thermal aging phenomenon was mostly due to diffusion and evaporation of pyrene.13The degradation rate in intensity was 30-50 %/h at 60°C. A different paint formulation was later developed with clearly improved thermal stability, which was achieved by restricting the diffusion of pyrene within the binder.14Gouin and Gouterman studied the effect of heat treatment on FIBbased PSP and found that the PSP’s temperature-dependency could be reduced by annealing the coating at 75°C.15Higher annealing temperature(150°C)would cause destructive effects to PSP’s sensing property. Compared to conventional PSP,fast PSP has different formulation that features a porous binder to enhance oxygen diffusion and achieve fast response.The binder usually consists of ceramic particles and only small amount of polymer.Recently,Li et al reported a fast PSP with an upper temperature limit of 80°C.Severe loss in luminescent intensity was observed as the temperature reached the limit.16However, the mechanisms behind thermal aging/degradation are not well understood, which impedes the development of fast PSP with a broad temperature range.

In the current study,we have firstly focused on the thermal stability of Polymer Ceramic PSP(PC-PSP),which is a widelyused sprayable fast PSP.The effects of thermal degradation on its key sensing properties, including luminescent intensity,pressure sensitivity, and response time, were examined for a temperature range from 60 to 100°C. Severe degradation in PSP’s sensing performance was found as temperature reached 80°C or higher,which was related to the polymer glass transition. Subsequently, a novel sprayable fast PSP was developed with significantly improved thermal stability(compared to PCPSP), which did not show severe degradation until 140°C.This was achieved by: (A) using polymer with higher glass transition temperature (polystyrene) to delay the saturation effect of oxygen quenching as temperature increased;(B)using mesoporous particles as host for luminophores to minimize the polymer-luminophore interaction. The findings have revealed the thermal degradation mechanism of fast PSP.They can also serve as reference for experimental design and error analysis of PSP measurement in test conditions involving elevated temperature.

2. Materials and fabrication

The PC-PSP used in this study consists of polymer ceramic binder and pressure/oxygen sensor PtTFPP (Frontier Scientific). The binder is basically a mixture of high concentration of ceramic particles(TiO2)with a small amount of acrylic polymer(Rhoplex HA-8)to physically hold the ceramic particles to the model surface. During paint preparation, the binder was air-sprayed on aluminum coupons (1 cm in diameter) and dried for 3-4 h before the PtTFPP solution was directly sprayed on top of the binder. This overspray preparation method resulted in luminophore deposition near the binder surface (see Fig. 1(a)), providing favorable conditions for the interaction between oxygen and luminophore moledules to achieve a response time below 100 μs.17

The Mesoporous-Particle PSP (MP-PSP) was prepared by first dissolving the luminophore (PtTFPP), mesoporous SiO2particles and polymer (polystyrene) into solvent (dichloromethane). Then 1%-2 % dispersant (Tween 80) was added to the mixture to form slurry. The slurry was sonicated for 10 min before it was air-sprayed onto the aluminum coupons.As shown in Fig. 1(b), the luminophores were distributed evenly within the binder (both inside and on surface of the mesoporous particles).The particles’highly porous and hollow structures greatly facilitated oxygen diffusion inside the PSP binder which led to fast response (response time around 100 μs).18Compared to PC-PSP, this paint formulation effectively reduced the negative effect due to the interaction between luminophore and polymer, and therefore showed improved thermal stability as discussed later in Section 4.

Fig. 1 Microstructures of fast PSPs.

3. Experimental setup

3.1. Thermal stability experiment: static calibration

Thermal stability experiments regarding the PSP’s intensity and pressure sensitivity were conducted in a customized calibration device (Instec Inc.) with a controllable temperature range from 20°C to 600°C and a controllable pressure range from 20 to 200 kPa. As shown in Fig. 2, the PSP sample was excited by a 385 nm LED(Prizmatix,UHP-T-LED-385)modulated by a signal generator(Tektronix,AFG1022).The luminescent signal was collected by a CCD camera(pco.1600,pco.imaging) through a 50 mm lens (50 mm/f1.2, Nikon). A 650±25 nm band-pass filter was installed before the lens to remove the excitation light. The PSP’s thermal stability was evaluated by performing static pressure calibration at different temperatures (from 20 to 100°C for PC-PSP, and from 20 to 140°C for MP-PSP). The calibration started fromT=20°C, and then the PSP sample was heated up to a certain temperature and kept steady for 10 min prior to the calibration.Each set of calibration was completed with 5 min,and the reference pressure was 101 kPa. In addition, separate experiments were conducted to obtain the intensity variation of PSP during the heat treatment for 30 min at selected temperatures.

Fig. 2 Experimental setup of static calibration.

Fig. 3 Experimental setup of dynamic calibration.

3.2. Thermal stability experiment: dynamic response

Additional experiments were conducted to evaluate the heating effects on PSP’s dynamic response using a shock tube facility as shown in Fig. 3.The shock tube is 1.6 m long with a driver section of 0.4 m and a low pressure section of 1.2 m which are separated by an aluminum diaphragm. The last 15 cm of the low pressure section is an optical glass tube,providing an optical access to the PSP sample which is located at the end of the tube. During the experiments, the low pressure section was kept at 1 atm while the driver section was pressurized by a compressor.The diaphragm broke at a pressure ratio of about 3.5 and a step change of pressure was created at the end of the tube from the impingement of a travelling shock wave. The PSP sample was continuously illuminated by the LED and the luminescent signal was captured by a photomultiplier tube(PMT,Hamamatsu h9305-03)and then recorded by an oscilloscope at a sampling frequency of 250 MHz. The data acquisition system was triggered by a pressure transducer installed on the side wall of the shock tube.Each signal was averaged over 64 measurements. For each sample, the experiments were performed three times and the averaged data were smoothed with a 20 μs window to filter out the noise. The final result of response time (90% rise time) was the average of 3 measurements. It should be noted that this facility only allows measurement at room temperature. Each PSP sample was heated up to a temperature TH and kept steady for 1 h.The response time was measured before and after the heat treatment.

4. Results and discussion

4.1. Thermal stability of PC-PSP

The static calibration curves obtained at different temperatures are compared in Fig. 4. The pressure values (P) were normalized by the reference pressure (Pref=101 kPa), and the intensity at the reference pressure (Iref)was divided by the intensity values(I)at different pressures.Linear relations between intensity and pressure were found for temperatures below 60°C.The pressure sensitivity atT=20°C was 0.72 %/kPa. Only a slight decrease in pressure sensitivity was observed as the sample was heated to 60°C. As the temperature reached 80°C or higher, the calibration curve exhibited severe nonlinearity with greatly reduced sensitivity in high pressure range(P/Pref≥1). To further investigate the mechanism behind this thermal degradation behavior, the intensity variations during the heat treatment (at constant temperature) were recorded forT=60,70 and 80°C,as presented in Fig.5.The intensity values(I)were normalized by the intensity at the beginning of the heat treatment(I0).At lower temperature(T=60°C),the intensity was mostly stable with only slight decrease over time.In contrast, large intensity drop (near 50% drop in 30 min)was found forT=70°C, which suggested significant change of paint property during the heat treatment. ForT=80°C,the intensity was deceasing faster thanT=60°C but without any sudden drop as shown in the case ofT=70°C. It was assumed that the change of paint property already occurred before temperature reached 80°C, so the large intensity drop was not shown in the result.

Fig. 4 Static calibration results of PC-PSP.

Fig. 5 Intensity variations of PC-PSP during heat treatment.

Based on the above results, it was speculated that the acrylic polymer binder was undergoing glass transition aroundT=70°C that caused the significant change in both PSP intensity and pressure sensitivity. In general,the oxygen diffusivity within glassy polymers was enhanced after glass transition due to the increase of free space between the polymer chains and the decrease of intermolecular forces.19,20The increased oxygen concentration within the binder would lead to stronger oxygen quenching and intensity reduction as observed in Fig. 5. Meanwhile, the strong oxygen quenching would lead to early saturation,meaning that the luminophores are fully quenched at relatively low pressure (P=130 kPa in this case) and therefore do not show pressure sensitivity forP≥130 kPa.Similar saturation effect was previously reported by Mori et al. for a PSP formulation designed for application in low pressure/density environment.21Their Paint Used Poly(TMSP) as binder, which had high oxygen diffusivity. This improved the pressure/oxygen sensitivity at low pressure condition,but also led to poor sensitivity near atmospheric conditions. In addition, a polymer sample was prepared and tested using a Differential Scanning Calorimetry (DSC) device (PerkinElmer DSC8500) for validation purpose. DSC is a thermal analysis technique that measures the amount of heat required to increase the temperature of a sample for identifying heat absorption/release events such as phase transition. During DSC test, the PSP sample was heated up in the device and the device recorded the amount of heat provided to the sample during temperature rise. A sudden change in the DSC curve typically indicates phase transition of the materials which requires additional heat.As shown in Fig.6,a peak was clearly visible around 90°C as a sign of possible phase transition. It should be noted that the onset temperature from DSC measurement was higher than that observed in thermal stability test (between 70 and 80°C). This bias was related to the delayed response of DSC measurement during the heating process.

From the practical point of view,the PC-PSP remains functional as temperature reaches 60°C considering that the temperature effects and slight sensitivity variation can be compensated using in-situ calibration. As temperature reaches 80°C, the highly nonlinear relation between pressure and intensity will cause failure of PSP application in most scenarios. One exception is the application in hypersonic wind tunnels with low static pressure during operation,where the paint can still be functional since the oxygen saturation effect does not occur in low pressure condition.

Fig. 6 DSC results of polymer sample in PC-PSP.

Fig. 7 Static calibration results of MP-PSP.

Fig. 8 Pressure sensitivity of PC-PSP and MP-PSP at different temperatures.

Fig. 9 Temperature calibration results of MP-PSP at different pressure conditions.

4.2. Thermal stability of MP-PSP

The results of PC-PSP indicate that the key to improve the thermal stability of PSP is to avoid polymer glass transition as much as possible. Accordingly, another polymer polystyrene was chosen as binder material for the new paint formulation since it has significantly higher glass transition temperature (around 107°C according to the work by Rieger22).The improved thermal stability of MP-PSP was confirmed by the calibration results shown in Fig. 7. No clear change was observed for the intensity calibration curves with temperature up to 120°C. ForT=140°C, the calibration curve remained linear but the pressure sensitivity was increased.The pressure sensitivity values of MP-PSP are compared with PC-PSP for different temperatures as shown in Fig. 8. The pressure sensitivity of MP-PSP atT=20°C was 0.67%/kPa, which was close but slightly less than PC-PSP under the same condition. For MP-PSP, the pressure sensitivity was mostly stable betweenT=20 and 120°C, until a significant change occurred atT=140°C. For PC-PSP, the pressure sensitivity was unstable forT≥60°C, and the calibration curve became nonlinear atT=80 and 100°C (see Fig. 4). The temperature range with poor thermal stability was indicated by the dashed line. In addition, the temperature calibration curves of MP-PSP at different pressure conditions were presented in Fig.9.The temperature sensitivity was quite stable betweenT=20 and 100°C, which was around 1%/°C.The sensitivity showed significant increase at higher temperatures with some pressure dependence. To further examine the possible thermal degradation effect, intensity variations of MP-PSP during the heat treatment (at constant temperature)were recorded forT=60, 100, 120 and 140°C, as presented in Fig. 10. The luminescent intensity was highly stable forT=60°C. Slight intensity degradation over time was observed as the temperature reached 100°C,and the degradation rate increased with temperature.A large drop in intensity was found forT=140°C, which indicated significant change of paint property during the heat treatment.

The above results show that MP-PSP is applicable for test conditions with temperature up to 140°C before severe intensity degradation occurs.The reason that this paint can remain functional beyond the glass transition temperature of polystyrene is related to the porous and hollow structure of ceramic particles. The mesoporous particles allowed luminophore deposition within the particles to prevent the interaction between luminophore and polymer, so that the negative effect of polymer glass transition can be greatly suppressed. The intensity degradation atT=140°C should be related to either the thermal decomposition of luminophore or chemical reaction between luminophore and polymer. The exact reason is yet to be found in future work.

Fig. 10 Intensity variations of MP-PSP during heat treatment.

4.3. Effect of heating on dynamic response

The response times of PC-PSP and MP-PSP were measured before and after heat treatment (τbeforeand τafter), and the results are presented in Table 1. The original PMT signals are shown in Fig. 11 for samples after heat treatment. In general, PC-PSP showed faster response than MP-PSP since the luminophores were distributed near the surface in PC-PSP.No clear change in dynamic response was observed for PCPSP, while the response time was found to increase after heat treatment for MP-PSP asTreached 100°C(near the polymer’s glass transition temperature). It should be noted that the dynamic response at high temperature could not be exactly simulated due to the lack of heated air in our facility. Therefore, we mainly focused on the effect of heating on the PSP binder structure and its potential impact on dynamic response.Considering that the glass transition was indeed a gradual process occurring within a range of temperature, it could cause irreversible changes to the binder structure during heat treatment above 100°C that led to reduction in oxygen diffusivity.This was actually not in conflict with the argument that the polymer’s oxygen diffusivity was increased after glass transition,because the measurement was made at room temperature after the heat treatment. Due to the reversible nature of glass transition, the oxygen diffusion rate should not change after the heat treatment. But the polymer distribution within PSP binder was likely to change considering the increased mobility of polymer at high temperatures. For MP-PSP, the luminophore solution was pre-mixed with binder material and the luminescent molecules were evenly distributed within the binder. Therefore, its dynamic response was closely related to the oxygen diffusion rate in the whole binder, which would become slow if the channels within mesoporous structure were blocked by polymer.As the response time was increased above 100°C for MP-PSP,the frequency response was reduced from over 10 kHz to below 10 kHz. This would affect the measurement accuracy of the high frequency content in pressure signal(close to or above 10 kHz).

Table 1 Dynamic calibration results of PSP samples before and after heat treatment.

Fig. 11 Dynamic calibration results after heat treatment.

5. Conclusions

The thermal stability of two sprayable fast PSPs (the conventional PC-PSP and a new MP-PSP) was studied in the range from room temperature to 140°C. For PC-PSP, significant intensity degradation was found as temperature reached 70°C. The static calibration curve showed sever nonlinearity beyond this temperature, which would restrict the application of PC-PSP in conditions with elevated temperature.The failure of PSP was related to the polymer glass transition which led to increased oxygen concentration within the binder and early saturation of oxygen quenching. In comparison, the MP-PSP used the polymer polystyrene with a higher glass transition temperature.It showed greatly improved thermal stability even beyond the corresponding glass transition temperature(around 107°C), until severe intensity degradation occurred atT=140°C. The pressure sensitivity values of PC-PSP and MP-PSP were 0.72%/kPa and 0.67%/kPa under room temperature,respectively,indicating that they would have similar performance under regular wind tunnel conditions. As for the dynamic response, the response time of PC-PSP was around 30 μs, which was not affected by the heating process.The response time of MP-PSP was around 90 μs, which was increased after the sample experienced heat treatment at 100°C or higher.

To further expand the temperature range of fast PSP (to higher temperatures), it is important to avoid polymer glass transition and polymer-luminophore interaction as much as possible. The thermal decomposition of luminophore molecules should also be investigated in detail. In future work,the thermal degradation mechanism of PSP will be further uncovered, and validation experiments will be performed to evaluate performance of MP-PSP at high temperature.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Nos.: 11872038 and 11725209) and funding from Gas Turbine Research Institute of Shanghai Jiao Tong University.