Abstract:Research highlights ► We investigated four measures to enhance the electro-osmotic regeneration system. ► The use of platinum-plated titanium anode effectively improved the work lifetime. ► Application of filter cloth could significantly enhance the regeneration rate. ► The interrupted power was found to enhance the regeneration rate up to 1.5 times. ► Optimal electric field strength ranges from 8.5 to 13 Vcm −1 in regeneration system. Abstract Electro-osmotic regeneration for the solid desiccant has been proved to have many merits such as regeneration without the heat source; energy-saving and simple structure. However; the previous work has revealed that its performance is seriously limited by the severe Joule heating effect and electrode corrosion; which demands further improvement to meet the practical requirement. In this paper; four possible improvement methods are investigated experimentally; including changing the material of anode; changing layout of cathode; applying the interrupted power and optimizing the electrical field strength. Through detailed experiments and analysis; we found that applying the platinum-plated titanium mesh as anode could improve the working lifetime from 6 h to over 120 h and effectively reduce Joule heating effect simultaneously; laying a piece of filter cloth under the cathode could enhance the EO regeneration rate up to 0.0021 g s −1 ; the application of interrupted power could increase the regeneration rate up to 1.5 times; the optimal on–off-time was found at 30 s:1.3 s with 17 V cm −1 electric field strength and 30 s:0.8 s with 11 V cm −1 ; and the most suitable value of electric field strength was observed as ranging from 8.5 to 13 V cm −1 in our EO regeneration system. Keywords Electro-osmotic Regeneration Solid desiccant Improvement Nomenclature E 0 applied electric field (V) EO electro-osmotic (–) m ˙ water removal rate (kg s −1 ) Subscripts weg by electro-osmotic and gravitational force wg by gravitational force we by electro-osmotic effect 1 Introduction Since the outbreak of SARS in 2003; and the recent outbreak of H1N1 these years; people pay more attentions to the indoor air quality [1] . Because of the separate control of latent load to make a more comfortable indoor environment; the desiccant dehumidification air-conditioning system has drawn increasing attentions [2,3] . And it also has other strong points such as the energy-saving; the reduction of the use of chlorofluorocarbons and so on [4–6] . To make the desiccant work repeatedly; regeneration components are required in the dehumidification systems; which are encouraged to be driven by the waste or renewable energy. [7,8] . The electro-osmotic (EO) regeneration; which has been concerned by many researchers these years; was first proposed in 2004. This novel regeneration method is proposed to regenerate the solid desiccant in a humidity independent control system; with the application of EO force generated by the external electric field. In 2004; Mina and Newell [9] conducted a primary theoretical research about the EO regeneration and found it feasible in the theoretical analysis. In 2007; Zhang and Yu [10] put a zeolite desiccant with an electric field into a high humidity box. However; no water was removed from the desiccant in their experiment. In 2010; Qi et al. [11] validated the possibility of the EO regeneration with the artificial zeolite desiccant experimentally; and the maximum regeneration rate observed was 1.1 × 10 −7 kg m −1 s −1 V −1 . In their investigation; the whole process of EO regeneration was found to be divided into three phases. However; in the preliminary work; the performance of the EO regeneration system was found to be seriously impaired because of the Joule heating effect and electrode corrosion. The Joule heating effect; which was caused by the high initial current; led to a considerable reduction of the water removal rate in the first 2–3 h. The electrode corrosion; which is generated by the electrolytic reactions; not only substantially limited the lifetime of the system; but also reduced the water removal rate. Moreover; the layout of the electrode was considered to play a significant role in the EO regeneration; which required further experiments and investigations to determine the details. And the effect of electric field strength also required further research to find out the optimum parameters. In conclusion; in order to improve the potential of the EO regeneration system for practical application; a system with better performance; including longer work life; higher regeneration rate and stability; is expected. In this essay; we developed and verified four methods to enhance the performance of the EO regeneration system by detailed experiments; including changing the material of anode; changing the layout of cathode; introducing the interrupted power and changing the electrical field strength. During the tests; the optimal control parameters of the improvements were also investigated. 2 Test bench As shown in the Fig. 1 ; the brief principle of the EO regeneration method is because of the rearrangement of the ions; the adsorbed water in the internal channel of the solid desiccant is driven by the electro-osmotic force from the anode to the cathode [11] . According to the previous research; the mean benefits of the EO regeneration are: (1) no heat source required; (2) simple structure (only a DC power supply needed); (3) suitable for renewable energy application. Fig. 2 shows a schematic of the EO regeneration system. The sample box is filled with the artificial zeolite powder (Radius: 200–250 μm). A DC power supply is used to provide specified electric field across the electrodes; which is located on the top and bottom surface of the sample. The relative humidity of the test chamber is maintained at 95 ± 3%; with the continuous inputting moist air generated by a supersonic oscillation humidifier. The measurement devices applied in this experiment are listed in Table 1 . During the experiment; all the required data were collected by a data logger and then delivered to a computer in every five minutes. It is noted that both EO and gravitational forces could remove water from the solid desiccant. So the amount of the removed water should be measured in the condition with and without the electric field to calculate the water removal rate by only electro-osmosis; which is the most significant assessment parameter of the performance. The first step of the experiment is: According to the different methods; adjust the experimental system; including changing the applied electric field strength; the structure of the sample box; the material of anode and introducing the interrupted power. The detailed adjustments were listed in the following parts of this essay. The following steps of the test are similar to our previous experiments; which are shown in the literature [11] . In all the sub-tests listed in part 3–6; the thickness of the sample between two electrodes was 15 mm; and the desiccant powder about 5 mm thick was covered above the anode. The porosity of the desiccant was kept around 80% after the adsorption period (Step (1) in [11] ). For further discussion; the water removal rate by electro-osmosis ( m ˙ we ; g s −1 ) is defined as the difference between the water quantity removed by gravity and that removed by both EO and gravitational forces per unit time ( m ˙ weg ; g s −1 ). 3 Anode material 3.1 Methodology In 2007; Sivapullaiah found that the electrode corrosion caused rapid increase of resistance in the EO soil concretion; which could be avoided by anti-corrosion electrodes [12] . However; whether the result can be applied in EO regeneration for solid desiccant needs further verification. Two types of anode; the conventional stainless steel mesh and the platinum-plated titanium mesh; were tested. Platinum was chosen for its excellent anti-corrosion performance in electro-osmotic pump applications [13] . However; the high cost of pure platinum anode prohibits the widespread use of it; especially in practical projects. So we selected titanium; which is considered as an economic-friendly material with high strength and anti-corrosion characteristics; as the base of the anode. The thickness of the coating was 1 μm; and both sides of the base were plated with the platinum. Sample with two different materials of anode was referred as A1 (stainless steel mesh) and A2 (platinum-plated titanium mesh); respectively. The applied exterior power was 17 V cm −1 in both conditions. And the layout of cathode inside the sample box was set as L1; which was to be described in the next chapter. 3.2 Results As shown in Fig. 3 a; it was noticed that after first 3 h of the test; serious electrode corrosion was observed on the stainless steel mesh electrode. The corrosion is caused by water electrolytic reaction; which means Fe (the main element of the anode) was transformed into Fe 3+ constantly and dissolved into water during the EO regeneration. However; the platinum-plated titanium mesh was insoluble and steady in the electrolytic reaction throughout the test with a total time beyond 120 h; as shown in Fig. 3 b. And no visible change was found on the surface of anode; and a longer operation-time could be predicted. The experimental results of sample A1 and A2; including the time of Phase 1 and 2; peak and initial current; were shown in Table 2 . Phase 1 refers to the process of the EO regeneration impacted by the serious Joule effect; which should be shortened. And Phase 2 is proofed to be the most effective period in the EO regeneration and should be lengthened as long as possible. The detailed description of Phase 1 and 2 are referred in literature [11] . As shown in the Fig. 4 ; the m weg and current of the first 4 h in the experiment presented large difference between sample A1 and A2. From Table 2 and Fig. 4 ; it is noticed that the initial current of A2 was evidently lower than that of A1; especially in the first half hour. Although the m weg of A2 was not as high as that of A1 in first few minutes; the shorter time period of Phase 1 of A2 could make up the difference. And then the m weg of A2 kept ahead during the experiment. It is clear that the performance of EO regeneration was highly improved by the application of the platinum-plated titanium anode. First; because of the reversible reaction of this anode (Pt + 2OH − ⇔ Pt(OH) 2 + 2e − ) in the EO regeneration; the anode could not be corroded. So the work lifetime of the system (the time of Phase 2) was considerably increased from around 6 h to over 120 h. And no visible changes were found on the surface of anode. The corrosion was considered to have a serious negative effect to the performance of the EO regeneration for three reasons. (a) The anode was corroded rapidly during the regeneration process and the area was reduced; which would seriously reduce the water removal rate. (b) It could cause the insoluble substance deposit into the material; which led to the increase of resistance and the decrease of current. (c) When the anode was wholly corroded; the EO regeneration would stop. So the platinum-plated titanium anode significantly enhanced the practical application potential of the new regeneration method. Moreover; the use of titanium anode; which was made of inert metal with low electric conductivity; effectively reduced the initial current and the Joule heating effect. (The serious Joule heat effect caused by high initial current; which happened in the application of ordinary anode; not only led to the stagnation of EO regeneration of 2–3 h (Phase 1); but also added extra cooling load to the air-conditioning system.) As shown in Table 2 ; at the beginning process of the regeneration; compared to the corroded anode; the peak and mean initial current of using platinum-plated titanium anode reduced around 80%. 4 Layout of cathode 4.1 Methodology In the previous experiment; it was found that a thin drenched layer (filter cloth) under the cathode had a significant influence on the water removal rate m we ∗ ) of the EO regeneration [11] . And different layouts between the layer and cathode may have different effects to the regeneration rate. So we designed four kinds of layout; with the different relative position between the cathode and filter cloth inside the sample box; to investigate the detail effect of the drenched layer. As shown in the Table 3 ; the layout was named L1; L2; L3 and L4; respectively. During the test; the applied electric strength was 17 V cm −1 in all conditions. And the anode was the platinum-plated titanium mesh. 4.2 Results The EO regeneration rates with four different layouts in Phase 2 were shown in Table 4 . To make the result more accurate; we repeated the experiment four times with every layout inside the sample box. From the results; it is clear that when the sample was fitted with the cathode only (L3); the EO regeneration rate was quite low. If a layer of filter cloth was added under the cathode; the regeneration rate increased; even without any measures (L2). As the trestle was used to keep the contact area between the cathode and the filter cloth as large as possible (L1); the average regeneration rate conspicuously increased. However; if the layer of filter cloth was placed upon the cathode (L4; the filter cloth was fitted between the desiccant and cathode); the water removal quantity by gravity only was superior than the water removal quantity by both gravity and EO forces; which means the EO regeneration rate became negative in this condition. The possible reason of the results might be that the influences on the flow pattern inside the micro-channels of desiccant varied from these four kinds of layout. In 2006; Buhai et al. [14] published that different flow patterns between open and close electrodes in the micro-channels and their influences. As shown in Fig. 5 a; the simulation results show that the impermeable electrodes lead to a cyclic flow pattern inside the pore of the desiccant; which makes the whole flow rate weaken significantly. However; it is contrast to the system with permeable electrodes represented by Fig. 5 b. In our experiment; the cathode with a layer of filter cloth above was similar to the first one; in which the cyclic flow pattern caused the reduction of water removal quantity by both forces. In contrast; when the filter cloth was placed under the cathode with a trestle; the sample box system was more similar to the second one because of the water absorption effect of cloth. So the flow direction was almost from the anode to cathode; which means that the EO regeneration could have a better performance in this condition. And the level of the flow pattern approaching to the ideal one increases as the increase of the contact area between the cathode and the filter cloth. So the use of trestle could result in higher performance of the regeneration system. 5 Interrupted power 5.1 Methodology In EO soil concretion; it has been proved that interrupted power can remove more water than continuous power in electro-osmotic dewatering by Robie in 1994 [15] ; and then Gopalakrishnan et al. found that there was an optimal off-time which yields maximum water removal in 1996 [16] . However; there is no experimental investigation to validate whether the measure could enhance the EO regeneration rate or not; and to find out the optimal on/off-time in the specified electric field strength. Based on this consideration; we tested the EO regeneration with the interrupted power supply; by the time-delay relay which could periodically interrupt the power and simultaneously short-circuited the electrodes. To find out the optimal parameter of the method; we fixed the on-time (time of connection) at 30 s; and change the off-time (time of interruption) from 0.1 s to 2 s. Besides; according to the results above; we applied the platinum-plated titanium anode and laid the sample box as layout L1. 5.2 Results The enhanced percentage of EO regeneration rate; which means the ratio between m we ∗ gained with and without the time-delay relay; was shown in Fig. 6 . As shown in the results; the interrupted power generated by the relay significantly enhanced the regeneration rate up to 1.5 times; comparing with the condition without it. The energy-saving potential could also be improved; consequently. The main reasons were: (a) Because of the electrolytic reaction; H + was generated on the anode and OH − was generated on the cathode persistently. Then these ions assembled at the opposite electrode with the water flow and caused the decrease of the effective electric field. The interrupted power resists the reduction by neutralizing the positive and negative ions in electrodes during the off-time; (b) In the conventional EO regeneration; the Zeta potential near the anode decreases because of the pH decrease [17] caused by the electrolytic reaction; which reduces the EO regeneration rate. The interrupted power reverses the electrode reaction on the anode and thus the pH near the anode increases during the off-time. However; during the off-time; the electric field inside the desiccant was opposite; which caused the reversed water flow. As a result; if the off-time was set too long; the EO regeneration rate would decrease. So an optimal ratio between the on and off-time should be investigated to present the better enhancing effect. As shown in the Fig. 6 ; the optimal on/off-time was found at 30 s:1.3 s with the applied electric field strength at 17 V cm −1 ; and the optimal on/off-time was 30 s:0.8 s with the applied electric field strength at 11 V cm −1 . Based on our previous research [1] ; to compare with other regeneration methods; the energy consumption is represented by the energy consumption rate. The rate is defined as the ratio of the energy consumed and the moisture mass removed. As shown in Table 5 ; the energy consumption rate of the EO regeneration was relative lower than several other methods. If the improvements (changing layout of cathode and introducing in the interrupted power) are applied properly in the EO regeneration system; the regeneration rate will increase and the consumption rate will be further reduced to 1.0 × 10 3 kJ kg −1 . It is proved that the improved EO regeneration system has higher energy-saving potential. 6 Electric field strength 6.1 Methodology In the conventional electro-osmosis theory; the EO flow rate was verified to vary directly with the electric field strength. For instance; Yao et al. found that increasing the electric field strength would linearly raise the maximum flow rate of EO pump in 2003 [18] . However; whether the relationship is correct in the EO regeneration system needs further verification. In this test; we changed the applied electric strength from 0 to 20 V cm −1 by a tunable DC power. For the better performance of the system; we applied the platinum-plated titanium anode and layout L1 in the test. The time-delay relay was not included in the circle because of the changing of electric field strength. 6.2 Results As shown in the Fig. 7 ; it is clear that the EO regeneration rate was not proportional to the electric field strength. The main reason was the mismatch of the adsorption and regeneration (desorption) of the desiccant. As shown in the Fig. 8 ; when the electric field strength was raised; the temperature of desiccant increased accordingly because of the more serious Joule heat effect. Higher temperature not only significantly impacted the adsorption ability of the artificial zeolite [18] ; but also improved the evaporation of the water inside. Because of the evaporation and reduction of adsorption; the porosity of the desiccant decreased rapidly; which caused the serious decrease of EO regeneration rate [19] . As shown in Fig. 7 ; when the applied field strength was adjusted at 8.5–13 V cm −1 ; the m we ∗ was superior compared to other conditions; which means that the most suitable value of electric field strength ranges from 8.5 to 13 V cm −1 for our EO regeneration system. It was verified that it is possible to obtain higher regeneration rate in EO regeneration system by applying the optimal electric field strength. However; it should be noticed that the optimal value varies with the type of desiccant; the structure of system; the operation parameters and environment and so on. 7 Summary In this essay; we experimented and analyzed four measures to enhance the performance of the EO regeneration system. From the results; several conclusions are listed as follow: (a) The use of platinum-plated titanium mesh anode improved the work lifetime of the system from 6 h to over 120 h; and effectively reduced the initial current and the Joule heating effect. (b) Application of a layer of filter cloth closely under the cathode with a trestle could significantly enhance the EO regeneration rate up to 0.0021 g s −1 . Besides; putting the filter cloth between the desiccant and cathode made the EO regeneration rate negative because of the cyclic flow inside the desiccant. (c) The interrupted power enhanced the regeneration rate up to 1.5 times compared with the condition without it. And the optimal on–off-time was found at 30 s:1.3 s with 17 V cm −1 and 30 s:0.8 s with 11 V cm −1 . (d) Because of the mismatch of adsorption and regeneration characteristics of desiccant; the most suitable value of electric field strength ranges from 8.5 to 13 V cm −1 in our EO regeneration system. The improvement of the EO regeneration performance improves the potential of practical use and energy-saving. The further work includes investigating whether the improved EO regeneration system could work under the ordinary summer outdoor environment and finding the influencing factors of it. References [1] C.H. Liang L.Z. Zhang L.X. Pei Independent air dehumidification with membrane-based total heat recovery: modeling and experimental validation Int J Refrig 33 2 2010 398 408 [2] X.Q. Zhai R.Z. Wang J.Y. Wu Design and performance of a solar-powered air-conditioning system in a green building Appl Energy 85 297–311 2008 [3] Y.T. Li H.X. Yang Investigation on solar desiccant dehumidification process for energy conservation of central air-conditioning systems Appl Therm Eng 28 10 2008 1118 1126 [4] L.Z. Zhang Energy performance of independent air dehumidification systems with energy recovery measures Energy 31 8–9 2006 1228 1242 [5] L.Z. Zhang Total heat recovery: heat and moisture recovery from ventilation air 2008 Nova Science Publishing Co. New York [6] Y. Yao W.J. Zhang S.Q. Liu Feasibility study on power ultrasound for regeneration of silica gel – a potential desiccant used in air-conditioning system Appl Energy 86 11 2009 2394 2400 [7] U. Eicker D. Schneider J. Schumacher Operational experiences with solar air collector driven previous termdesiccantnext term cooling systems Appl Energy 87 12 2010 3735 3747 [8] L.Z. Zhang D.S. Zhu X.H. Deng Thermodynamic modeling of a novel air dehumidification system Energy Build 37 3 2005 279 286 [9] Mina E, Newell TA. Electroosmotic dehumidification. In: Proceedings of the 21st IIR. Arlington (Virginia, USA): International Congress of Refrigeration; 2004. [10] Zhang YQ; Yu J. The research of a refrigeration system based on electroosmosis for dehumidification. Master Thesis; Beijing University of Technology; 2007 [in Chinese]. [11] R.H. Qi C.Q. Tian S.Q. Shao Experimental investigation on possibility of electro-osmotic regeneration for solid desiccant Appl Energy 87 7 2010 2266 2272 [12] P.V. Sivapullaiah B.S. Nagendra Prakash Electroosmotic flow behaviour of metal contaminated expansive soil J Hazard Mater 143 3 2007 682 689 [13] S.H. Yao D.E. Hertzog S.L. Zeng Porous glass electroosmotic pumps: design and experiments J Colloid Interface Sci 268 1 2003 143 153 [14] B. Buhai T. Binser R. Kimmich Electroosmotic flow; ionic currents; and pressure-induced flow in microsystem channel networks: NMR mapping and computational fluid dynamics simulations Appl Magn Reson 32 1–2 2006 25 49 [15] H.R. Rabie A.S. Mujumdar Interrupted electroosmotic dewatering of clay suspensions Separ Technol 4 1 1994 38 46 [16] S. Gopalakrishnan A.S. Mujumdar Optimal off-time in interrupted electroosmotic dewatering Separ Technol 6 3 1996 197 200 [17] W.T. Tsai K.J. Hsien H.C. Hsu Adsorption of organic compounds from aqueous solution onto the synthesized zeolite J Hazard Mater 166 2–3 2009 635 641 [18] S. Yao J.G. Santiago Porous glass electroosmotic pumps: theory J Colloid Interface Sci 268 1 2003 133 142 [19] F. Xiao G.M. Ge Control performance of a dedicated outdoor air system adopting liquid desiccant dehumidification Appl Energy 88 1 2011 143 149

Electro-osmotic Regeneration for Solid Desiccant

Experimental Investigation on Possibility of Electro-Osmotic Regeneration for Solid Desiccant

Development of Solid Desiccant Dehumidification Using Electro-Osmosis Regeneration Method for HVAC Application

Investigation on Electro-osmotic Regeneration of Solid Desiccants

Experimental investigation on performance improvement of electro-osmotic regeneration for solid desiccant

Design And Study Of A Novel System For Solid Desiccant Dehumidification

Experimental Investigation on Adsorption and Electro-Osmosis Regeneration of Macroporous Silica Gel Desiccant

Study on the performance of solar interfacial evaporation for high-efficiency liquid desiccant regeneration

An ion mass transfer model of electrodialysis regenerator for inorganic salt liquid desiccants

Performance Investigation of a Heat Pump Driven, Vacuum Liquid Desiccant Regeneration System

Heat and mass transfer model for desiccant solution regeneration process in liquid desiccant dehumidification system

Experimental study of a liquid desiccant regeneration system: performance analysis for high feed concentrations

Investigation on liquid desiccant regeneration for advanced air-conditioning: Aerodynamic thermal method versus mechanical vapor recompression method

An exergy analysis and parameter optimization of solid desiccant heat pumps recovering the condensation heat for desiccant regeneration and heat transfer enhancement

Regeneration behavior of solid desiccants with microwave drying

Performance investigation of a liquid desiccant regenerator with CFD technology

Performance Exploration on a Small Sorption Device Extracting Water from Air for Museum Display Cases

Numerical Simulation On The Performance Of Regenerator In Liquid Desiccant Air Conditioning Systems

Heat and mass transfer performance analysis of cross-flow regenerators in a liquid desiccant air conditioning system

Performance evaluation of a rotary dehumidifier with molecular sieve desiccant using coupled regeneration mode: Experimental Investigation

Effects of temperature and humidity ratio on the performance of desiccant dehumidification system under low-temperature regeneration