2012-杂质对于氢化物反应器工作的影响

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 37(2012) 13843e 13848

Available online at

www.sciencedirect.com

journal homepage:www.elsevier.com/locate/he

Influenceof impurities on hydrogen absorption in a metal hydride reactor

Dmitry Dunikov *, Vasily Borzenko, Stanislav Malyshenko

Joint Institute for High Temperatures of Russian Academy Sciences, Izhorskaya 13Bld. 2, 125412Moscow, Russian Federation

a r t i c l e i n f o

Article history:

Received 10December 2011Received in revised form 9April 2012

Accepted 12April 2012Available online 12May 2012Keywords:Hydrogen Storage Hydride Heat transfer Impurities

Critical phenomenon

a b s t r a c t

Absorption of pure and impure hydrogen in AB 5-type metal hydride reactors is experi-mentally investigated. The process can be divided into three phases:“adiabaticheating”phase, “heattransfer”phase and “endof reaction”phase. Critical phenomenon is observed between “adiabaticheating”and “heattransfer”phases. The crisis occurs when temper-ature of metal hydride bed reaches maximum, which is close to equilibrium temperature for inlet pressure, and is followed by significantslowdown of the reaction rate. Presence and accumulation of impurities in the voids of metal hydride bed precipitates crisis due to decreasing of hydrogen partial pressure. Two strategies of hydrogen purificationwith the aid of metal hydrides are discussed. Mixture filtrationthrough the metal hydride bed is recommended for high concentration of impurities and PSA (orTSA) suits for nearly pure hydrogen.

Copyright ª2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

reserved.

1. Introduction

Integration of renewable energy (RE)sources into electric power systems requires technologies for energy storage. Energy storage in hydrogen may be an option for isolated electricity networks, small-scale self-sufficientpower supply systems [1], hydrogen can provide longer-term seasonal storage on electricity grids relying exclusively on variable RE inputs [2]. About 2/3of the territory of Russian Federation with population about 20millions are not covered by centralized electric grids and rely on autonomous power systems based on imported fuels or local resources of fossil energy sources causing substantial environmental damage [3]. It is also esti-mated that in European Union around 300,000households do not have an access to reliable electricity grids [4], thus there is a great potential for the renewable hydrogen market.

Hydrogen production from wind and solar energy is considered as the hydrogen production by water electrolysis. Up-to-date electrolyzers produce hydrogen with the purity at almost 100%,which is sufficientfor PEM fuel cells.

In contrast hydrogen produced from biomass may contain a lot of impurities. Two major routes for hydrogen production are exist:thermochemical (gasification,pyrolysis, reforming, hydrothermal liquefaction) and biochemical (photosynthetic,fermentation and esterification)[5]. Thermochemical hydrogen production from biomass is similar to very well developed technologies of hydrogen production from fossil fuels where mature technologies of pressure swing adsorption (PSA)and membrane separation are used [6].

In the recent years high attention is paid to biological production of hydrogen. Biohydrogen usually contains more than 50%of CO 2and is not suitable without purificationfor

*Corresponding author .Tel.:+[1**********]; fax:þ[1**********]. E-mail address:[email protected](D.

Dunikov).

0360-3199/$e see front matter Copyright ª2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijhydene.2012.04.078

13844

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 37(2012) 13843e 13848

fuel cells, especially PEM FC [7]. Industrial technologies of hydrogen purificationdo not exactly meet the needs of kW-scale power units. PSA process produces hydrogen at 99þ%but requires the feed mixture containing 75e 90%H 2at pres-sure 15e 30bar and economically competitive for product flowhigher than 1000m 3/h,membrane systems have lowest capital costs for small flowrates and can operate with feed flowcontaining 30e 90%H 2at pressure about 10bar [8], but polymeric membrane systems such as Polysep by UOP and PRISM by Monsanto (AirProducts and Chemicals Inc.) produce hydrogen at purities ranging only from 70to 99%[6]. Thus there is a need in developing of alternative technology suit-able for small-scale hydrogen production from biological sources. Palladium alloy membranes may be the choice due to their highest selectivity but they operate at high temperatures (350e 500 C) and can be damaged by impurities [9].

Metal hydride systems can be used both for hydrogen storage for stationary and portable fuel cell power supply and hydrogen purification[10]. Solid state hydrogen storage offers higher volumetric energy density than pressurized vessels and can operate at near ambient working temperature and pressure [11]. Selective hydrogen sorption by intermetallic alloys can be used for hydrogen purification[10]. Pilot plants were made for recovery of hydrogen from industrial off-gas streams [12]in ammonia production [13], and hydride tech-nologies can be more efficientthan conventional ones [14]. In our opinion integrated metal hydride purificationand storage systems can be used for purificationand utilization of biohydrogen.

There are numerous alloys and intermetallic compounds that have properties of real commercial interest and value for applications [15]. AB 5-type alloys (LaNi5family) show fast hydrogenation kinetics, good cyclic life and tolerance to impurities [16,17]and can be used for off-grid power supply and be competitive to the batteries, even Li-ion [18]. Low gravimetric hydrogen content (1.37%for LaNi 5H 6) is the drawback of low temperature metal hydrides, but it is not a critical option for stationary applications.

Thermal management in metal hydride bed and purity of hydrogen are the main limiting factors in operation of metal hydride devices. Hydrogenation produces a lot of heat due to high caloric effect of the reaction, which is about 20e 50kJ/molH 2for different low temperature alloys [15]. Activated metal hydrides are finepowders with low effective thermal conductivity (about1W/(mK)) [19e 21], and poor internal heat

transfer obstructs an evacuation of reaction heat from a metal hydride bed during sorption and heating the bed during desorption. A lot of efforts are made to create optimal design of metal hydride reactor [22e 32]and solve the problem of heat management.

Impurities have negative influenceon the performance of metal hydride devices, some admixtures such as H 2S or CO have poisonous effect [33e 35], even CO 2can contaminate metal hydrides [34,35], and inert admixtures such as N 2slow down the reaction of hydrogen sorption [36].

This paper represents the summary of experimental investigations of heat transfer in metal hydride reactors during sorption of pure and impure hydrogen in order to demonstrate possibility of hydrogen purificationby metal hydrides.

2. Experimental setup

Four types of reactors filledwith AB 5-type metal hydride were used in experiments. Schematic drawings of the reactors are presented in Fig. 1.

Type A reactor consists of four metal hydride modules (onlyone is presented in Fig. 1A) with permeable walls. The modules are placed into a robust casing hermetically sealed with a cover. The casing is cooled by cold water flowingthrough heat exchanger (notshown in Fig. 1A). Alloy:Mm 0.8La 0.2Ni 4.1Fe 0.8Al 0.1, 4.69kg; capacity 690st. L.

Type B reactor consists of one metal hydride cartridge (seeFig. 1B) with corrugated outer surface. Inner surface of reac-tion chamber is cooled by water and outer surface can be cooled by water or air. Alloy:Mm 0.8La 0.2Ni 4.1Fe 0.8Al 0.1, 0.7kg; capacity 100st. L.

Type C reactor consists of seven metal hydride cartridges (Fig. 1C) with common collectors for hydrogen and cooling water. Inner and outer surfaces of the reaction chamber of each cartridge are cooled by water. Alloy LaFe 0.1Mn 0.3Ni 4.8, 5kg; capacity 810st. L.

Type D reactor consists of a robust casing with two hydrogen connections at bottom and cover, thus hydrogen can be filteredthrough the metal hydride bed. Alloy LaFe 0.1Mn 0.3Ni 4.8, 1.6kg; capacity 260st. L.

During experiments reactors were fed by pure hydrogen or hydrogen e nitrogen mixture from standard gas cylinders through reducing valves. Hydrogen flowwas controlled

at

Fig. 1e Schematic drawings of the metal hydride reactors (seetext for details).

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 37(2012) 13843e 13848

13845

inlet (andoutlet for reactor D) by Bronkhorst EL-FLOW Select series mass flowcontrollers, temperature is measured by platinum resistance temperature detectors Heraeus M422and pressure is measured by Aplisens PC28pressure transmitters.

3.

Results and discussion

3.1.

Heat and mass transfer crisis

Typical results of experiment on charging of the A-type reactor with pure hydrogen at constant hydrogen loading rate are shown in Fig. 2. In more details the experiments on constant charging rate of the metal hydride reactor are pre-sented in our previous work [37].

Following the work [38]we can divide the process of hydrogen sorption in a reactor into three phases:at the beginning the reaction is limited by hydrogen flowin a supply line (Q ¼const in Fig. 2), then the reaction is limited by a rate of heat evacuation from hydride bed and the process comes to an end with gradual cooling of the bed. The transition between the firstand the second phases is connected with a critical phenomenon:reaction rate sharply drops down, pressure (P ) curve has visible kink and temperature (T ) rea-ches maximum. Such a behavior is observed in the most of experiments with constant hydrogen loading rates (seefor example Ref. [39]). Hydrogen pressure inside reactor is the main controlling parameter of sorption. With pressure as an independent variable the results for different initial condi-tions become uniform for the second and the third phases and differ only during heating in the beginning.

In our previous works we propose to call this phenomenon heat and mass transfer crisis [31,37]. Indeed, transition between the firstand the second reaction phases is connected with a change in heat and mass transfer law. Convective and heat transfer terms of energy conservation equation for a porous metal hydride bed can be compared with the aid of Peclet number, and in general it is supposed that Pe is much smaller than unity and convection term can be neglected [40].

Fig. 2e Hydrogen flowat inlet, temperature in the center of metal hydride bed inside the hydride module of A-type reactor during sorption of pure

hydrogen.

Our experiments show that for the firstphase of the reaction convective term should be taken into account. Fig. 3shows a difference between temperatures in the center and at the water cooled edge of the of metal hydride bed in the module of A-type reactor for different hydrogen loading rates. During the firstphase of sorption temperature at the edge of metal hydride bed is higher than temperature in the center. This can be explained only by domination of convective heat transfer mechanism. Temperature gradient typical for heat conduc-tance is formed during the crisis and heat conductance is dominative mechanism of heat transfer during the second and the third phases.

Experiments on hydrogen sorption in B-type reactor were made to findan influenceof cooling regime on duration of the firstphase. Metal hydride bed was cooled by water, by air convection or by natural convection, all other conditions were identical, hydrogen loading rate was set to 19st. L/minat pressure 0.9MPa. Description and results are given in Table 1. Type of cooling has great influenceon charging rate, the sorption time for cold water cooling was only 6.5min (I)whereas for cooling by natural convection the sorption time was 66min (VI).If cooling rate is insufficient,the crisis occurs when a maximum temperature is reached and amount of reaction heat generated by absorbed hydrogen corresponds to the sensible heat needed to heat up the metal hydride bed to maximum temperature. Heat transfer to coolant can be neglected in this case and heating at the firstphase can be treated as adiabatic [12]. Thus the heat and mass transfer crisis is connected with transition between the adiabatic heating and the heat transfer phases of the reaction.

3.2. Hydrogen purification

Impurities have significantinfluenceon kinetics of hydrogen sorption in metal hydrides. Even nonpoisonous impurities such as N 2in small amounts slow down the reaction

rate.

Fig. 3e Temperature difference between center and water cooled edge of metal hydride bed inside the hydride module of A-type reactor for different hydrogen loading rates at inlet (from240to 24st. L/min).

13846

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 37(2012) 13843e 13848

Fig. 4e Gas flowat inlet and pressure inside the C-type reactor during charging with impure (left)and pure hydrogen (right).

Comparison of experimental results on hydrogen sorption of pure hydrogen and hydrogen from mixture with 3.25%of nitrogen are presented in Fig. 4. Before crisis the C-type reactor can be charged with 400st. L of pure hydrogen but only 140st. L can be charged from mixture. Pressure raises much faster in the case of presence of impurities. Due to accumu-lation of nitrogen in the voids of metal hydride bed hydrogen partial pressure is lowering during the process. Residual partial pressure of nitrogen is about 0.6MPa (seeFig. 4).

To increase sorption rate one can blow off residual impu-rities from the free volume of reactor. This technique is similar to the pressure swing adsorption widely used in industry [41]. Results of experimental hydrogen purificationin a PSA-like mode are presented in Fig. 5. Hydrogen containing

Fig. 5e Amount of charged hydrogen (st.L) and share of hydrogen losses (%)at PSA-like hydrogen purification(6.6%N 2) in the D-type reactor. Fig. 6e Amount of mixture at inlet (27%N 2), at outlet and hydrogen concentration in outlet stream for flow-throughpurificationin the D-type reactor.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 37(2012) 13843e 13848

13847

6.6%of nitrogen is fed to inlet of the C-type reactor at 0.45MPa. When pressure inside the reactor reaches 0.4MPa the mixture is blown off into atmosphere and the process is repeated. After 55min pure hydrogen is discharged from the reactor.

During purge some hydrogen is lost and share of losses increases during the process reaching 30%(seeFig. 5). Thus optimization of automatic process is needed. With manual regulation of purge 8%of losses is reached.

To be competitive with other hydrogen purificationtechnol-ogies 90%hydrogen recovery is needed especially for purificationof hydrogen-poor gas (e.g.biohydrogen) and PSA-like mode doesn’tsuit well this purpose. Introduction of flow-throughreactors [42]can improve situation with hydrogen losses.

Results of experiments on purificationof hydro-gen e nitrogen mixture (27%N 2) are shown in Fig. 6. Mixture is fed to the D-type reactor from the inlet in the reactor cover and depleted mixture leaves the reactor from the outlet in the bottom. Initially hydrogen is absorbed in metal hydride and H 2concentration in outlet flowis small with total losses lower than 5%.When the reactor is charged about 66%hydrogen breaks through the metal hydride bed, hydrogen concentra-tion at outlet rises and losses increase. Experiments show that flow-throughtechnique can be feasible and high hydrogen recovery rates could be reached.

4. Conclusion

Absorption of hydrogen in a metal hydride reactor be divides into three phases:

1. Adiabatic heating phase. Heat is generated due to rapid absorption by cold particles of alloy and heat transfer is limited by hydrogen convection within the metal hydride bed. Heat of reaction is compensated by sensible heat of the metal hydride bed. Temperature rise is limited by equilib-rium temperature corresponding to inlet pressure. This phase ends with heat and mass transfer crisis when maximum temperature is reached.

2. Heat transfer phase. Heat transfer is limited by heat conduction within the metal hydride bed, and heat of reaction is compensated by heat transfer to reactor walls. Sorption rate is much slower.

3. End of reaction phase. Reaction rate is too slow to compensate cooling. Hydride bed is gradually cooled down. Presence and accumulation of impurities in the voids of metal hydride bed results in the decreasing of hydrogen partial pressure. Hydrogen partial pressure falls below the inlet pressure and the maximum equilibrium temperature decreases, duration of the firstphase of reaction becomes shorter and the crisis occurs earlier. Two strategies of hydrogen purificationwith the aid of metal hydrides are possible. Usage of pressure swing absorption (orthermal swing absorption) is connected with high hydrogen losses and thus can be used preferably at low concentration of impuri-ties. Flow-through technique based on filtrationof mixture through the metal hydride bed is better for high concentration of impurities and hydrogen recovery higher than 95%can be achieved.

Acknowledgments

The authors would thank D. Blinov and V. Zhemerikin for aid in experimental work and Russian Ministry for Education and Sciences for the financialsupport (statecontracts 16.516.11.6052, 16.516.11.6103, SC-3717.2010.8).

r e f e r e n c e s

[1]Beaudin M, Zareipour H, Schellenberglabe A, Rosehart W.

Energy storage for mitigating the variability of renewable electricity sources:an updated review. Energy for Sustainable Development 2010;14:302e 14.

[2]Andrews J, Shabani B. Re-envisioning the role of hydrogen in

a sustainable energy economy. International Journal of Hydrogen Energy 2012;37:1184e 203.

[3]Fortov VE, Popel OS. Power engineering in the modern world.

Dolgoprudnyi:Intellect Publishing House; 2011[inRussian].[4]Zoulias EI, Lymberopoulos N. Hydrogen-based autonomous

power systems:techno-economic analysis of the

integration of hydrogen in autonomous power systems. Springer; 2008.

[5]Muradov NZ, Veziro g

lu TN. “Green”path from fossil-based to hydrogen economy:an overview of carbon-neutral

technologies. International Journal of Hydrogen Energy 2008; 33:6804e 39.

[6]Ritter JA, Ebner AD. State-of-the-art adsorption and

membrane separation processes for hydrogen production in the chemical and petrochemical industries. Separation Science and Technology 2007;42:1123e 93.

[7]Levin DB, Pitt L, Love M. Biohydrogen production:prospects

and limitations to practical application. International Journal of Hydrogen Energy 2004;29:173e 85.

[8]Freeman B, Yampolskii Y. Membrane gas separation. John

Wiley &Sons; 2011.

[9]Yun S, Ted Oyama S. Correlations in palladium membranes

for hydrogen separation:a review. Journal of Membrane Science 2011;375:28e 45.

[10]Sandrock G, Bowman RC. Gas-based hydride applications:

recent progress and future needs. Journal of Alloys and Compounds 2003;356e 357:794e 9.

[11]Schlapbach L, Zuttel A. Hydrogen-storage materials for

mobile applications. Nature 2001;414:353e 8.

[12]Sheridan III JJ, Eisenberg FG, Greskovich EJ, Sandrock GD,

Huston EL. Hydrogen separation from mixed gas streams using reversible metal hydrides. Journal of the Less Common Metals 1983;89:447e 55.

[13]Au M, Chen C, Ye Z, Fang T, Wu J, Wang O. The recovery,

purification,storage and transport of hydrogen separated from industrial purge gas by means of mobile hydride

containers. International Journal of Hydrogen Energy 1996; 21:33e 7.

[14]Mordkovich VZ, Baichtock YK, Dudakova NV,

Korostyshevsky NN, Sosna MH. Comparative efficiencyof using hydrides in industrial processes of hydrogen recovery and compression. International Journal of Hydrogen Energy 1993;18:839e 42.

[15]Sandrock G. A panoramic overview of hydrogen storage

alloys from a gas reaction point of view. Journal of Alloys and Compounds 1999;293e 295:877e 88.

[16]Kolachev BA, Shalin RE, Ilyin AA. Hydrogen storage alloys.

Metallurgy, Moscow; 1995[inRussian].

[17]Hirscher M, Hirose K. Handbook of hydrogen storage:new

materials for future energy storage. Wiley-VCH; 2010.

13848

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 37(2012) 13843e 13848

[18]Gray EM, Webb CJ, Andrews J, Shabani B, Tsai PJ, Chan SLI.

Hydrogen storage for off-grid power supply. International Journal of Hydrogen Energy 2011;36:654e 63.

[19]Sun D-W, Deng S-J. A theoretical model predicting the

effective thermal conductivity in powdered metal hydride beds. International Journal of Hydrogen Energy 1990;15:331e 6. [20]Hahne E, Kallweit J. Thermal conductivity of metal hydride

materials for storage of hydrogen:experimental

investigation. International Journal of Hydrogen Energy 1998; 23:107e 14.

[21]Pons M, Dantzer P. Determination of thermal conductivity

and wall heat transfer coefficientof hydrogen storage

materials. International Journal of Hydrogen Energy 1994;19:611e 6.

[22]Sun D-W. Designs of metal hydride reactors. International

Journal of Hydrogen Energy 1992;17:945e 9.

[23]Oi T, Maki K, Sakaki Y. Heat transfer characteristics of the

metal hydride vessel based on the plate-fintype heat exchanger. Journal of Power Sources 2004;125:52e 61.

[24]MacDonald BD, Rowe AM. Impacts of external heat transfer

enhancements on metal hydride storage tanks. International Journal of Hydrogen Energy 2006;31:1721e 31.

[25]Askri F, Ben Salah M, Jemni A, Ben Nasrallah S. Optimization

of hydrogen storage in metal-hydride tanks. International Journal of Hydrogen Energy 2009;34:897e 905.

[26]FørdeT, NæssE, Yartys VA. Modelling and experimental

results of heat transfer in a metal hydride store during hydrogen charge and discharge. International Journal of Hydrogen Energy 2009;34:5121e 30.

[27]Kaplan Y. Effect of design parameters on enhancement of

hydrogen charging in metal hydride reactors. International Journal of Hydrogen Energy 2009;34:2288e 94.

[28]Melnichuk M, Silin N, Peretti HA. Optimized heat transfer fin

design for a metal-hydride hydrogen storage container. International Journal of Hydrogen Energy 2009;34:3417e 24. [29]Visaria M, Mudawar I, Pourpoint T, Kumar S. Study of heat

transfer and kinetics parameters influencingthe design of heat exchangers for hydrogen storage in high-pressure metal hydrides. International Journal of Heat and Mass Transfer 2010;53:2229e 39.

[30]Yang FS, Wang GX, Zhang ZX, Meng XY, Rudolph V. Design of

the metal hydride reactors e a review on the key technical issues. International Journal of Hydrogen Energy 2010;35:3832e 40. [31]Borzenko V, Dunikov D, Malyshenko S. Crisis phenomena in

metal hydride hydrogen storage facilities. High Temperature 2011;49:249e 56.

[32]Garrison SL, Hardy BJ, Gorbounov MB, Tamburello DA,

Corgnale C, vanHassel BA, et al. Optimization of internal heat exchangers for hydrogen storage tanks utilizing metal hydrides. International Journal of Hydrogen Energy 2012;37:2850e 61.

[33]Sandrock GD, Goodell PD. Surface poisoning of LaNi 5, FeTi

and (Fe,Mn)Ti by O 2, Co and H 2O. Journal of the Less Common Metals 1980;73:161e 8.

[34]Sandrock GD, Goodell PD. Cyclic life of metal hydrides with

impure hydrogen:overview and engineering considerations. Journal of the Less Common Metals 1984;104:159e 73.

[35]Schweppe F, Martin M, Fromm E. Hydrogen absorption of

LaNi 5powders precovered with O 2, CO, H 2S, CO 2or N 2. Journal of Alloys and Compounds 1997;253-254:511e 4. [36]Artemov VI, Lazarev DO, Yan’kovGG, Borzenko VI,

Dunikov DO, Malyshenko SP. The effect of non-absorbable gas impurities on heat and mass transfer in metal-hydride devices for storage and purificationof hydrogen. High Temperature 2004;42:987e 95.

[37]Dunikov DO, Borzenko VI, Malyshenko SP. Optimization of

heat transfer in metal hydride reactor. In:Stolten D, Grube T, editors. 18th World hydrogen energy conference 2010Essen, Germany, vol. 4; 2010. p. 83e 8.

[38]Turillon PP. Design of hydride containers for hydrogen

storage. In:4th World hydrogen energy conference. Pasadena, CA, USA, vol. 3; 1982. p. 1289e 305.

[39]Botzung M, Chaudourne S, Gillia O, Perret C, Latroche M,

Percheron-Guegan A, et al. Simulation and experimental validation of a hydrogen storage tank with metal

hydrides. International Journal of Hydrogen Energy 2008; 33:98e 104.

[40]Chaise A, Marty P, Rango Pd, Fruchart D. A simple criterion

for estimating the effect of pressure gradients during hydrogen absorption in a hydride reactor. International Journal of Heat and Mass Transfer 2009;52:4564e 72.

[41]Saitou T, Sugiyama K. Hydrogen purificationwith metal

hydride sintered pellets using pressure swing adsorption method. Journal of Alloys and Compounds 1995;231:865e 70. [42]Rudman PS, Sandrock GD, Goodell PD. Hydrogen separation

from gas mixtures using LaNi 5pellets. Journal of the Less Common Metals 1983;89:437e 46.