Experimental comparison of the heat transfer of supercritical R134a in a micro-fin tube and a smooth tube

Highlights

• Heat transfer of supercritical R134a was compared in a ST and a MF tube.

• The $\frac{{\mathit{Gr}}_b}{{\mathit{Re}}_b^{2.0}}\left(\frac{{\rho }_b}{{\rho }_w}\right)\frac{x}{d}$ criterion give the best prediction for buoyancy effect.

• Threshold for the onset of the buoyancy in ST was around 100 and 2000 in MFT.

• α in MFT top and bottom side was 1.68 and 1.59 times that in ST.

Abstract

Measured heat transfer rates of supercritical R134a were compared for flows in a horizontal micro-fin tube and in a smooth tube for mass fluxes from 100 to 700 kg m⁻² s⁻¹, heat fluxes from 10 to 70 kW m⁻², and pressures from 4.26 to 5 MPa. The results showed that the micro-fin tube wall temperatures had smaller changes as the heat flux was increased than on the smooth tube. The heat transfer coefficients along the top of the smooth tube are reduced more by the buoyancy effect than in the micro-fin tube. Then, the results were used to evaluate four buoyancy criteria for horizontal flows for both the smooth tube and micro-fin tube with the $\frac{{\mathit{Gr}}_b}{{\mathit{Re}}_b^{2.0}}\left(\frac{{\rho }_b}{{\rho }_w}\right)\frac{x}{d}$ criterion found to give the best prediction accuracy for both types of tubes. The threshold for the onset of the buoyancy effect in the smooth tube was around 100 but was around 2000 in the micro-fin tube, which indicates that the effect of buoyancy is greatly reduced in the micro-fin tube. Comparisons of the measured heat transfer coefficients for all 5520 experimental data points showed that the heat transfer coefficient in the micro-fin tube was 1.68 times that in the smooth tube on the top and 1.59 times that in the smooth tube on the bottom.

Introduction

The Organic Rankine Cycle (ORC) has been rapidly developed in recent decades to convert low and medium grade waste heat to electricity [1], [2], [3], [4]. The Trans-critical Organic Rankine Cycle is believed to have even higher efficiencies and lower exergy losses than the ORC since the transcritical ORC does not involve isothermal boiling [5], [6], [7], [8], [9], [10], [11], [12], [13]. Chen et al. [5] compared ORC and transcritical ORC systems and showed that transcritical ORC systems are 10.80% more efficient than ORC systems for a maximum cycle temperature of 393 K. Maraver et al. [8] concluded that the transcritical ORC efficiency will always be better than the ORC efficiency when the critical temperature of the working fluid is much lower than the heat source temperature. Furthermore, the elimination of the boiling simplifies the transcritical ORC heating system [7]. Thus, a review of the literature shows that transcritical ORC systems can be more efficient for many working conditions.

However, one important issue that has not been extensively studied is the heat transfer of supercritical organic fluids, especially for horizontal flows in tubes, which is quite important since the heat exchangers in most commercial transcritical ORC systems involve horizontal flow. Although a few heat transfer investigations have been conducted with supercritical R134a [14], [15], [16], [17], [18], [19], most of these studies focused on fluid cooling for supercritical heat pumps or air conditioners. Karellas et al. [20] stated that there have been few studies of the heat transfer mechanisms for supercritical organic fluids directly related to transcritical ORC applications. To the best of the authors’ knowledge, only four studies [17], [18], [21], [22] have investigated the heat transfer of supercritical organic working fluids during heating, three of them are for supercritical R134a [17], [18], [21] and one is for supercritical R22 [22]. However, all these studies were for vertical flows with no research on horizontal flows. There have been many reports of abnormal heat transfer phenomena during the heating of supercritical fluids in various systems. The heat transfer for supercritical fluids can experience normal heat transfer, heat transfer enhancement, or heat transfer deterioration depending on the conditions [23]. And the buoyancy effect, which was caused by density variation, was found to be an important factor that may affect the heat transfer. The buoyancy effect may modify the shear stress distribution across the vertical tube, with consequential change in turbulence production and occurrence of heat transfer deterioration [24]. Many buoyancy criteria have been built to predict the buoyancy based on different working fluids (supercritical water [25], [26], supercritical CO₂ [27], [28]). The accuracy of existing buoyancy criteria should be properly evaluated for supercritical R134a. Since the drastic heat transfer deterioration caused by buoyancy effect may affect stable operation of transcritical ORC systems and the organic fluid may even decompose [29] due to excessive temperatures. Therefore, the safe operation of organic fluid systems requires further investigation of the heat transfer for supercritical organic fluids before transcritical ORC systems can be widely used in real applications.

The research on supercritical fluid cooling shows that micro-fin tubes can significantly improve the heat transfer [30], [31], [32]. The micro fins may delay or restrict the heat transfer deterioration to effectively protect the organic working fluid from decomposition. The present study investigates the heating of a supercritical organic working fluid, R134a, in a horizontal micro-fin tube. The results in the micro-fin tube were then compared with the heat transfer coefficients in a smooth tube. Some previous studies have compared the heat transfer coefficients for supercritical water in a smooth tube and an enhanced (internally ribbed) tube [33], [34], [35], [36]. However, the data could not be effectively compared because the working conditions, such as the pressures, mass fluxes and heat fluxes, were not the same due to the different experimental loops used for the smooth and internally ribbed tubes. In addition, the inside tube diameters were quite different which also affected the comparisons [33] (18 mm in the smooth tube and 26 mm for in the internally ribbed tube). The heat transfer with a supercritical fluid is quite sensitive to the conditions and the heat transfer can be easily affected by the operating parameters, so comparisons are needed for such tubes with the same working conditions.

The present study compares the heat transfer coefficients for smooth and micro-fin tubes for identical working conditions (same pressures, heat fluxes and mass fluxes) in the same experimental loop. The tube diameters were also similar (7.76 mm for the micro-fin tube and 10.3 mm for the smooth tube). The experiments examine the effects of pressure, heat flux, and buoyancy on the heat transfer coefficient. The heat transfer coefficients are compared for 5520 experimental data points in the smooth and micro-fin tubes. The experimental results show the effects of the micro fins on the heat transfer to supercritical fluids as a reference for the design of transcritical ORC systems.