Unsteady Evaporation of Water from Wire Mesh Structures at Sub-Atmospheric Pressures

Abstract

Adsorption heat pumps and chillers can provide thermal energy with a low carbon footprint, therefore, this technology could contribute to a sustainable future heat and cold supply system. Most adsorption modules work with the refrigerant water at sub-atmospheric pressures, which poses a challenge for effective evaporation and requires a customized design concept for the evaporator heat exchanger. The favorable nucleate boiling regime can hardly be reached under the given boundary conditions. Instead, evaporation from extensive thin refrigerant films on capillary structures represents a promising approach. \nThis work refers to a heat exchanger concept for compact one-chamber adsorption modules, which employs porous capillary structures for cyclic condensation and evaporation without the need of continuous refrigerant supply. Such a concept necessarily involves unsteady evaporation conditions due to a continuous reduction of the refrigerant charge. Since scientific publications in this field are scarce, this work focuses on investigating unsteady evaporation of water from porous structures at sub-atmospheric pressures on the example of copper wire mesh structures. Special attention is paid to the question, how the transforming refrigerant distribution and heat transfer conditions interact, and how this interaction affects the dynamic evaporation performance. In this context, the impact of different structure geometry parameters (i.e. porosity, pore size, structure height, wire orientation angle) and thermodynamic conditions (vapor saturation pressure, heat flux) on the evaporation mechanisms and performance is addressed. Further questions are, if the evaporation dynamics can be reproduced with a simple mathematical model, and if wire mesh evaporators can finally be considered as a promising approach for the envisaged application. \nInvestigations were pursued via two methodical approaches: Firstly, unsteady evaporation measurements with wire mesh structure samples of different geometry specifications were conducted. Secondly, a simple resistance-capacitance model was developed which includes time-dependent resistance and capacitance definitions representing the presumed heat transfer components. In both measurements and simulations, the overall heat transfer coefficient of the structure (from structure base temperature to vapor saturation temperature) was used as the main evaluation quantity. Additionally, the refrigerant storage capacities of the capillary structures were evaluated. \nMeasurements and simulations revealed that the pore size of the mesh structure crucially affects the dynamic refrigerant distribution, refrigerant storage capacity, and evaporation performance for the investigated conditions: Large and medium pore sizes (0.8 mm clear mesh width and larger) involve a predominance of gravitational forces as against capillary forces (large Bond number) which leads to the dewetting pattern of a receding evaporation front. The thermal conduction resistance of the refrigerant-filled section represents the performance-limiting factor in a broad refrigerant charge interval for this dewetting type. A small pore size (0.375 mm clear mesh width), in contrast, implicates distinctly different dewetting and evaporation characteristics which presumably originate from a combination of a receding front and a pattern of wet and dry clusters and which can be ascribed to the increasing impact of capillary forces. Besides a potentially higher refrigerant storage capacity, the investigated structure with smallest pore size also reached the highest overall heat transfer coefficients of up to 23…28 kW/(m2K). \nFurther analyses of geometry impacts indicated that a low porosity and low structure height are beneficial by trend, however, the optimal choice for these geometry factors depends on the envisaged application case. \nIn the standard version of the resistance-capacitance model the conception of a receding refrigerant front was implemented. Respective simulations show a fairly good qualitative agreement with the measured evaporation dynamics of structures with large and medium pore size (≥ 0.8 mm). The prediction quality for the dynamics of small pore sizes is poor since for these structures the receding front dewetting characteristics do not apply. An alternative model conception (“receding front + static front”) implies possible dewetting mechanisms of fine pored structures and yields a better agreement with the respective measurements. Quantitative simulation results from the standard “receding front” approach match the measurement results quite well in several cases, however, the outcomes adumbrate that certain model parameters are imprecise. Despite the necessity for a revision of these definitions, the developed evaporation model is considered as a valuable tool for the prediction of unsteady evaporation processes. Integrated into a model on heat exchanger level it could potentially serve a basis for dimensioning methods. \nIn order to assess the tested mesh structures with regard to the envisaged application, thermal transmittance (UA) and refrigerant storage capacity were referred to the structure volume which usually represents a critical design factor. On structure level the volume-specific thermal transmittance equals the structure-height-specific overall heat transfer coefficient, which is used as the assessment quantity for heat transfer on structure level. Here, the heat transfer coefficient refers to the temperature difference between structure base and saturation temperature of the vapor atmosphere. Considering the diverging requirements of a power-focused versus efficiency-(COP-)focused adsorption module design, the mesh structure with smallest pore size (0.375 mm clear mesh width) and medium structure height (10 mm) showed the best suitability for both cases (with a structure-volume-specific refrigerant storage capacity of about 850 kg/m3 and a structure-height-specific optimal mean heat transfer coefficient of 1350… 2500 kW/(m3K), depending on the required refrigerant turnover). A structure type with medium pore size (0.9 mm clear mesh width) and low structure height (5 mm) proved to be the second-best variant. \nThese two most promising structure types were used for a potential assessment of a hypothetical wire mesh evaporator heat exchanger. A round tube heat exchanger design with external porous structure was assumed. Its geometry was adapted to a specific finned tube heat exchanger for partially flooded continuous operation which is one of the best-performing evaporators among current research activities and which was employed as an ambitious reference. The calculated absolute thermal transmittance values (UA values) reveal that the potential of advanced mesh structures can only be exploited if a sufficiently high fluid-side heat transfer is ensured. As an assessment quantity on heat exchanger level the construction-volume-specific UA value was used, which refers to the temperature difference between heat transfer fluid inside the tube and saturation temperature of the vapor atmosphere, and to the construction volume of the entire heat exchanger. From the estimation results it can be deduced that – depending on the considered conditions – the hypothetical mesh evaporator could reach similar construction-volume-specific UA values (ranging up to 1000 kW/(m3K)) as the highly efficient reference evaporator. An optimization of the structure geometry – such as a reduction of pore size and porosity and modification of the structure height – is expected to allow for further improvements. For a cyclic operation in one-chamber adsorption modules a mesh evaporator could prove particularly advantageous due to its low required refrigerant mass. Furthermore, it involves a high constructional flexibility. These outcomes suggest that the integration of wire mesh structures in an evaporator in cyclic operation is generally a promising approach for the application in adsorption heat pumps and chillers, and that further investigations are justified.