Effect of Surface Thermal Variations during Cryogen Spray Cooling in Dermatologic Laser Therapy

Cryogen spray cooling (CSC) is a spatially selective heat transfer technique that provides epidermal protection during laser treatment of selected dermatoses, such as port wine stain (PWS) birthmarks. Most numerical studies of CSC-assisted PWS therapies to date assume constant cooling conditions at the skin surface. In the present study, however, we show that cooling conditions at the skin surface vary significantly both in time and space. The objective of this paper is to assess the effect of thermal variations at the skin surface on the heat extraction process during PWS laser therapy. First, a single temperature sensor systematically recorded temperature changes along the sprayed area of a skin model. Next, a multiple temperature sensor acquired temperature data at four strategic radial locations namely, at the center, middle, perimeter and outside the sprayed area. Finally, recorded temperatures along with an inverse heat conduction problem (IHCP) algorithm were used to study the heat extraction process at the surface. Spatial and dynamic profiles of surface temperatures, heat fluxes, heat transfer coefficients and heat removal are presented. Results show that local and temporal variations of the boundary conditions may have a strong influence on CSC cooling efficiency during dermatologic laser therapy. The study shows that external conditions must be considered and ideally controlled to optimize current laser therapies of selected dermatoses, such as PWS. Corresponding author Introduction Cooling and heating processes are essential to many medical treatments, such as laser therapy of dermatologic vascular lesions, basal skin carcinomas and cartilage reshaping for reconstructive surgery. Port wine stain (PWS) is a congenital and progressive vascular malformation of the dermis. Laser irradiation at appropriate wavelengths induces permanent thermal damage to PWS blood vessels. However, laser energy is also absorbed by epidermal melanin causing localized heating therein. As a consequence, complications such as hypertrophic scarring and skin dyspigmentation may occur. Cryogen spray cooling (CSC) is a spatially selective heat transfer technique that spurts liquid cryogen onto the skin surface. The cryogen evaporates extracting heat from the epidermis thereby increasing the threshold for epidermal damage [1]. Heat transferred through the skin surface during CSC is a function of many fundamental spray parameters [2, 3] that vary in time and space within the spray cone (average droplet diameter and velocity, mass flow rate, temperature and spray density). Therefore, a non uniform surface heat flux at the skin surface occurs during CSC-assisted dermatologic laser treatment. During CSC, uniform cooling within the sprayed area is highly desirable but not practical. Despite surface variations, most numerical studies on CSC-assisted PWS therapy to date assumed constant heat transfer conditions at the skin surface. In the present study, the effect of surface thermal variations during CSC is experimentally and numerically addressed. First, a single temperature sensor was used to measure temporal and radial temperature values along the radius of the sprayed surface of a skin model. Next, a multiple temperature sensor was used to measure temperatures values at four specific radial locations on the skin model namely, three within and one outside the sprayed surface. Finally, an inverse heat conduction problem (IHCP) algorithm was applied to determine the temperature, heat flux, heat transfer coefficient and total heat removal at the surface of the skin model. Experimental and Numerical Methods In this section, only a brief description of the materials and experimental and numerical methods is presented. A more detailed description of the temperature sensors and computer algorithm can be found in the referenced literature. Cryogen spray system Cryogen R-134a (1, 1, 1, 2-tetrafluoroethane) was delivered through a high pressure hose to a fuel injector attached to a straight-tube-nozzle. R-134a, (with a boiling temperature at atmospheric pressure of Tb≈26 °C) was kept at saturation pressure (600 kPa at 25 °C). A nozzle with an inner diameter d1=0.7 mm and length l1=63.6 mm was used for the single temperature sensor. A nozzle with d2=0.57 mm and l2=8 mm was used for the multiple temperature sensor. The spray characteristics produced by these and other nozzles have previously been reported [4, 5]. Nozzle length has little influence on spray characteristics. However, nozzle diameter choice, on the other hand, has a dramatic effect on spray characteristics [4, 6]. Nozzles can be classified as narrow (0.5≤d≤0.8 mm) and wide (d≈1.4 mm) based on spray characteristics [6]. Therefore, the two nozzles employed in this study are classified as narrow with equivalent cryogen spray characteristics. Temperature sensors The first measuring device employed a single temperature sensor as schematically shown in Fig. 1(a). A miniature type-K thermocouple was placed on top of a 12.5 mm square bar of polymethyl methacrylate (Plexiglass®). Cellulose tape (Scotch tape®), 50 μm thick, was placed next to the thermocouple bead. A piece of aluminum foil (15 x 10 mm) covers the sensor. Thermal paste around the thermocouple bead is used to ensure good thermal contact. The second measuring device employed multiple temperature sensors as schematically shown in Fig. 1(b). The device has four silver disks of 3.18 mm diameter and 0.17 mm thickness. Disks are placed 1 mm apart from each other. The top surface of each disk is exposed to the spray while the lateral and bottom surfaces are embedded in epoxy. Type-K thermocouples are welded to the bottom of each disk. These temperature sensors provide a substrate that has thermal properties similar to skin such that total heat removal Q, surface heat flux q and heat transfer coefficient h are qualitatively similar to those expected Plexiglass® and epoxy can be found in the literature [7, 8, 9]. The single sensor was used to obtain temperature measurements every 0.5 mm on the area where the cryogen was deposited. The multiple sensors were used to measure temperatures namely, at the center, middle, perimeter and outside the sprayed area. for human skin. Thermal properties of human skin, Hea ansfer calculations t conduction algorithm was used Experiments st set of experiments, the nozzle located 40 m esults and Discussion f experimental results noise is filte ure sens Ther al dynamics and radial variations orded by the sing t tr A one-dimensional hea to compute the surface heat flux q from temperature measurements. This approximation relies on the fact that the width of the sprayed area (≈ 16 mm) is much larger than the depth (≈ 0.17 mm) of temperature measurements. Indeed, the time scale and depth of relevance in PWS treatment are milliseconds and 0.5 mm, respectively. The applied inverse heat conduction problem (IHCP) algorithm is based on Beck's sequential function specification method [10], where q is estimated as a piecewise constant function of time. A discussion of this algorithm as applied to CSC can be found elsewhere [11]. In the fir m from the surface of the single sensor delivered a cryogen spurt of 60 ms. The thermocouple was initially placed at the center of the spray cone (r=0 mm), subsequently, the thermocouple was displaced 0.5 mm from the center. Temperature measurements were recorded every 0.5 mm from r=0 to 7 mm. In the subsequent experiments, the nozzle located 40 mm away from the surface of the multiple temperature sensor delivered a cryogen spurt of 100 ms. The initial skin model temperature in both cases was 22.5 °C. R In the presentation o red out by taking a running average using the subsequent and previous five measurements in time. Four independent runs with the single temperat or were performed at every radial location. Figure 2 shows the average values of four independent temperature measurements (TA) at r=0 mm. The error bars denote standard deviations. Similar curves were obtained for the remaining locations. Three independent runs were performed with the multiple sensor. The standard deviations obtained were similar to those for the single sensor. Plexiglass Scotch tape Aluminum foil