Absorption refrigeration systems

In the last years there is renewed interest of sorption systems due to the increasing price of the primary energy, which leads to the more efficient distributed model of energy production. In this distributed model, sorption systems could play an important role. Small capacity systems (less than 15 kW) could be an interesting option in the present situation. There have been many industrial developments in the last decade mainly in Europe, USA and China. However, up to now its implementation has been limited due to the initial high investment necessary. One the main reasons could be the lack of standardization both in the components and systems.

Fig. 1:Solar-cooling installation
By one hand, in the South of Europe there is a large amount of solar energy available that can be used for air-conditioning in summer and for heating in winter. In this line, two types of solar-driven system can be developed (Figure 1): i) high performance, flat plate solar collectors with highly insulated cover driving an absorption chiller; ii) flat plate solar collectors driving an adsorption chiller. On the other hand, waste heat is other of the possible uses of the sorption systems: by using such devices, the overall energy exploitation in a cogeneration installation can be up to 75-80% (see Figure 2).
Fig. 2: Cogeneration installation

In the Heat Transfer Laboratory (CTTC) of the Universitat Politècnica of Catalunya (UPC) there is a research line based on the development of small capacity sorption systems. The research approach is based on the systematic application of numerical simulation modeling adequately validated with experimental data.
The methodology adopted in the development of the sorption chillers is general for any thermal system:

  1. Exhaustive analysis of the physical phenomena and geometry.
  2. Proposal of possible mathematical formulations. Study of the diverse possibilities of use of levels of modelling of different degree of detail.
  3. Numerical resolution of the governing equations. In this point all the numerical solutions must be checked in order to achieve results with no programming errors and not dependent of the calculation mesh used (verification of numerical errors).
  4. Empirical validation. Comparison of the numerical predictions with the empirical results obtained of an experimental facility.
  5. Use of the simulations as virtual experimental units. Once the mathematical models have been verified and experimentally validated, it is possible to use the codes as virtual experimental units for design or prediction purposes.

A modular object-oriented simulation platform is being developed, which allows the linking between the different components (solar collectors, pump, valves, heat exchangers, etc.) of each system. In this numerical platform each component is an object, which can be either an empirical-based model (e. g. heat exchangers, solar collector) or a more detailed CFD calculation if necessary. With this platform parallel computing is allowed. By this simulation strategy several levels of simulation are allowed:

  • System level: in this level, each component (solar collector, chiller, tank, etc.) is modeled by means an overall mass and energy balance [1].
  • Cycle level at chiller inside. In this level each component o the chiller (pump, valves, heat exchangers, etc.) is modeled by means an overall mass and energy balance, together with equilibrium relation equations [2, 3. 4]. Transient modeling is preferred.
  • Component level.  Special effort is being devoted to the design of the heat exchangers [5, 6]. Finite volume techniques are used in order to simulate such elements. Two main types of heat exchangers are considered: i) serpentine tube bundle; ii) batteries of finned tubes (see Figure 3). In this level of simulation empirical information is necessary in terms of heat & mass transfer coefficients for calculating the overall heat and mass exchanged and friction factors for pressure drop calculations.
  • Detailed level. In case that there is not available empirical information detailed CFD simulations are required for specific situations [7]. For instance, in absorption processes special effort has to be devoted in case of techniques of enhancement are used: i) use of additives; ii) use of advanced surfaces.

Fig. 3: Heat exchangers simulation. Up: tube bundle; down: batteries of finned tubes

Focusing on basic research, the combined heat and mass transfer in the absorption process represents the major critical issue in absorption systems and one of the key issues. For this reason the development of mathematical models for the simulation and experimental data for the validation are always useful tools for the design and improvement of vertical falling film. Vertical falling film absorption are close related with compact air-cooled absorption machines. The absorber usually is the largest element of absorption machines due to its low heat and mass transfer coefficients, and this fact determines the total design of the whole system.

Moreover, one of the main reasons of the discrepancies between theoretical predictions made by models of absorbers of H2O-LiBr absorption chillers when they are compared with experimental results under real conditions is the presence of non-condensables gases. These non-condensables gases are inside the shell of the absorption chiller mainly for two reasons: i) air leakages (Oxygen-Nitrogen); ii) gases produced by corrosion (Hydrogen).
A mathematical model of falling film absorption of H2O by LiBr aqueous solutions which considers the influence of non-condensable gases has been implemented. The model is semi-empirical, based on Navier Stokes equations together with energy and mass species simplified under the boundary layer hypotheses. Under such conditions, the differential system of equations in partial derivatives, becomes parabolic and could be solved by means of finite difference method in a step by step procedure. Detailed heat and mass transfer balances are applied at the interface to specify the boundary conditions between liquid and gas phases. In order to calculate gradient of air at the interface, the penetration theory is applied in order to avoid a detailed calculation of the gas phase. Numerically the presence of air in the interface results in a pressure drop and consequently in a reduction in heat and mass transfer rates.

On the other hand, there is a parallel development of three types of experimental of sorption systems. From one hand there is a facility to validate all basic research that is being carried regarding falling film phenomena. On the other applied research competence, two chillers will be small sized, therefore avoiding cooling tower could be an interesting issue:

  1. Regarding heat and mass transfer phenomena analysis, a testing device has been designed and built for reproducing absorption phenomena in vertical tubes with the primary objective to obtain experimental data in LiBr -H2O vertical absorbers; LiBr – H2O is used as working fluid. The versatility of the experiment allows to obtain a wide range of data. Falling film configuration is typical in finned heat exchangers (air-cooled). Absorption process is controlled adjusting the following independent variables: solution concentration, solution temperature, cooling water temperature, absorber pressure and solution mass flow.
  2. A 7 kW air-cooled absorption chiller (already constructed). This research is done taking advantage of a previous European funded project [12]. In such project a laboratory prototype of the same characteristics was developed and tested [13]. With the tests performed the mathematical models employed in the present design were experimentally validated (see Figure 4). According to the steps in the developments, we are in 4 and 5 points (2nd generation prototype). In this stage of development, the spin-off of the Group Termo Fluids S. L., is boosting the reaching to the market of the chiller.
  3. An small capacity adsorption chiller (to be constructed). In this case the experience acquired in the development of the absorption chiller will be employed for the new adsorption chiller. New concepts in the reactor(s) will be developed, in order to avoid cooling tower. According to the steps in the developments, we are in 1 to 3 points.

Fig. 4: Single-effect, air-cooled absorption machine


  1. Castro, J., Oliva, A., Oliet, C. and Pérez-Segarra, C. D. “Construction of a pre-industrial prototype of an air-cooled H2O-LiBr absorption chiller for solar cooling applications”, Proceedings of the EUROSUN 2006 Conference.
  2. Wang X. and Chua H.T., Two bed silica gel-water adsorption chillers: An effectual lumped parameter model. Int. Journal of Refrigeration, Vol. 30, pp. 1417-1426, 2007.
  3. Chua, H.T., Ng K.C., Wang W., Yap C. and Wang X.L. Transient modeling of a two-bed silica gel-water adsorption chiller, Int. Journal of Heat and Mass Transfer, Vol. 47, pp. 659-669, 2004.
  4. Evola G., Le Pierrès N., Boudehenn F. and Papillon P., Proposal and validation of a model for dynamic simulation of a solar-assited single-stage LiBr/water absorption chiller, Int. Journal of Refrigeration, Vol. 36, pp. 1025-1028, 2013.
  5. Castro, J., Oliva, A., Pérez-Segarra, C. D. and Oliet, C. “Modelling of the heat exchangers of a small capacity, hot water driven, air-cooled H2O-LiBr absorption cooling machine”, International Journal of Refrigeration, Vol. 31, No. 1,  pp. 75-86.
  6. Niazmand H. and Dabzadeh I., Numerical simulation of heat and mass transfer in adsorbent beds with annular fins, Int. Journal of Refrigeration, Vol. 35, pp. 581-593, 2012.
  7. Castro, J., Leal, L., Pérez-Segarra, C. D. and Pozo, P. “Numerical study of the enhancement produced in absorption processes using surfactants”, International Journal of Heat and Mass Transfer, Vol. 47, No. 14-16,  pp. 3463-3476.
  8. E. García-Rivera, J. Castro, J. Farnós, A. Oliva. Two mathematical models of the falling film in vertical tubes. 8th IIF/IIR Gustav Lorentzen Conference on Natural Working Fluids. 7th – 10th of September 2008. Copenhagen, Denmark.
  9. E. García-Rivera, J. Castro, J. Farnós, A. Oliva. Simulation of absorption of H2O in falling of LiBr aqueous in vertical tubes in wavy regime. Sources/Sinks alternative to outside Air for Heat Pump & AC Techniques + International Sorption Heat Pump Conference (ISHPC11) . 6th – 8th of April 2011. Padua, Italy.
  10. E. García-Rivera, J. Castro, J. Farnós, A. Oliva. Modelling of absorption of H2O in falling film of LiBr aqueous in vertical tubes with presence of non-absorbables and wavy regime. 10th IIF/IIR Gustav Lorentzen Conference on Natural Working Fluids.  25-27th june 2012. Delft, Netherlands.
  11. E. García-Rivera, J. Castro, J. Farnós, A. Oliva. Experimental and numerical investigation of H2O vapor absorption processes in falling film of LiBr aqueous solution in vertical tubes. 8th World Conference on Experimental Heat Transfer, Fluid Mechanics, and Thermodynamics June 16-20, 2013, Lisbon, Portugal.
  12. Oliva, A., Pérez-Segarra, C.D., Castro, J. and Quispe, M.. "ACABMA project. Final report -Publishable-", 2002.
  13. Castro, J., Oliva, A., Pérez-Segarra, C.D. and Cadafalch, J, “Evaluation of a small capacity, hot water driven, air-cooled H2O-LiBr absorption machine”, HVAC Research, Vol.13, No 1, pp. 59-75, 2007.