научная статья по теме GAS-FLUSHING MECHANISM EFFECT ON THE ADSORPTION CAPACITY OF CRYOGENIC PUMPS IN LARGE-SCALE HYDROGEN SYSTEMS Комплексное изучение отдельных стран и регионов

Текст научной статьи на тему «GAS-FLUSHING MECHANISM EFFECT ON THE ADSORPTION CAPACITY OF CRYOGENIC PUMPS IN LARGE-SCALE HYDROGEN SYSTEMS»

Hydrogen storage

GAS-FLUSHING MECHANISM EFFECT ON THE ADSORPTION CAPACITY OF CRYOGENIC PUMPS IN LARGE-SCALE HYDROGEN SYSTEMS

Gusev A. L.

«TATA» Science and Engineering Cente P.B.O. 787, Sarov, Nizhny Novgorod Region, 607183, Russia Phone: 8-83130-63107, Fax:8-83130-63107, e-mail: gusev@sar.ru,

Complications resulting occurring when heating cryogenic vacuum systems of hydrogen power facilities encouraged researchers to seek new way for gas release improvement. One of the potential techniques involves the so-called flushing (blasting) or «rinsing» of the vacuum system with a compound that can be then easily removed by pumping.

The gas flushing (blasting) mechanism efficiency of vacuum systems to remove undesired vapors or gases was indicated by M. G. Kaganer as early as in 1966: «Good results are also obtained from freon-12 blasting». One of the flushing efficiency explanation is that adsorbed molecules are assumed to be replaced with flushing gas molecules. Vacuum system flushing should use gases with low adsorption binding energy so that they can be easily removed from the surface. The paper presents the tests of cryogenic adsorption systems functioning at 90 K, 77 K, 20 K temperatures. Each series of cryogenic adsorption system regeneration used the adequacy principle for the basic processing operations: adsorbent warm-up, coil inlet air temperature of 400 K; air pressure of 3 atm, void cavity pressure of 1x10-1 mm mercury, regeneration time — 48 hours, pumpdown was accomplished with HB3-500 pump using a nitrogen trap.

The experiment was conducted for the following cases: a) standard regeneration, cooling, evaluation of pump characteristics in «self-consistent» mode; b) regeneration, adsorbent cooling, nitrogen rinsing of six cycles: nitrogen feeding from Dewar vessel through the pipe to the operational pump chamber up to 10 Pa — warm-up - pumpdown through mechanical pump - cooling - nitrogen feeding; c) regeneration - adsorbent cooling - freon rinsing of three cycles: freon feeding from the vessel through the pipe to the operational pump chamber up to 10 Pa - warm-up - pumpdown through mechanical pump - cooling - freon feeding (freon-114).

For the experiment credibility, after each series of cycles a), b), c) air loading was applied to the device till complete adsorbent saturation and P pressure stabilization at 10 Pa with cooled adsorbent. The absorption capacity at 10 Pa was determined by passing gas load through the gas counter.

Adsorption pumps are generally regenerated twice a year. The regeneration quality determines the limit operational pressure in thermal insulation cavity of the cryogenic item and hence reservoir evaporability. The oper-

ation revealed poor adsorbent regeneration instances along with the cases requiring the operational cycle to run while the adsorption device no onger allowed the operation because of accumulated damage.

The adsorption capacity is the basic adsorption device parameter — this characteristic is frequently the only efficiency criterion for cryogenic adsorption pumps. How efficiently the absorption devices are regenerated determines functioning requirements of the devices. Field testing of multiple large-scale adsorption devices allowed to conclude that the regeneration quality can be significantly enhanced by using additional regeneration methods. Those methods were developed during pilot cryogenic system operation and successfully implemented. The highest performance of the cryogenic pump is achieved by preparing the adsorbent surface with simple regeneration method using additional nitrogen rinsing.

If the adsorbent is prepared with simple regeneration then the cryogenic pump entering the standard operation regime demonstrates instable adsorbent pump-down capacity. The same is observed after freon rinsing (though to a less extent as compared to standard regeneration). After adsorbent regeneration and nitrogen rinsing the adsorbent works with a high time stability. Additional nitrogen rinsing (6 times) allows to increase significantly the absorbing capacity of the cryogenic pump and to improve dramatically the inter-schedule time under adequate gas loadings. Low efficiency of flushing gas rinsing (like freon vapors) can be explained by that among multiple factors governing the sorption rates on ceolites (polarization capability of cations, polarizabili-ty of cations and polarity of sorbed molecules, evacuation conditions etc.) the most important is the molecule size regarding the inlet diameter; note that the molecule shape is critical rather than the volume. Freon molecules allow to distinguish between the molecular sieve 5A and chabazite belonging to the same class (type 3 in Barrer classification). CF3Cl and CHF2Cl molecules sorb on either sorbent. The sorption rate for molecules with the same critical size decreases with increasing the molecular length. Fig. 1 shows the dependence of the effective rate of cryogenic pump on cryogenic pump in let pressure under different conditions of «gas flushing».

Gas and vapor sorption kinetics on ceolite demonstrates some specific features as compared to the sorption kinetics on activated carbon, silica gels and other sorbents. The basic feature is that the sorption is highly

ISJAEE Special issue (2003)

Second International Symposium «Safety and Economy of Hydrogen Transport»

IFSSEHT-2003

100

ss

1 ^ ^

&>

#

91

Cryogenic pump inlet pressure, (Pa)

Fig. 1. Effective speed of cryogenic pump as a function ofpressure for various surface treatment conditions (curve 1 — standard regeneration, curve 2 — regeneration and nitrogen flushing, curve 3 — regeneration and Freon-114 rinsing)

sensitive to the size of sorbed molecules. This is caused by that the void sizes (more precisely the size of inlets to the ceolite sorption cavities) are so small that are nearly identical to (and sometimes are even smaller than) gas molecule sizes. Therefore the diffusion coefficients and activation energy of the diffusion process vary within a wide range for different materials.

Consider the case of microporous zones of the same size. If a. is the concentration of the I mixture component in the adsorption phase then a,=a+a (i=1,2,..., n).

Here a and ai2 are the concentrations of the I component in adsorbed state in the transport voids and in microporous zones.

The averaged mass balance equation for the I component can be written as

dai dci ■ + ■

dt dt

-yJi, (i=l, 2..., n),

where c is the local component concentration in the gas phase inside the grain (averaged over the volume AQ); ji is the averaged mass flux of the i component.

The actual implementation of the research results allowed to increase the inter-schedule time of built-in cryogenic sorption pumps whose regeneration necessitates emptying the cryogenic vessel.

The cost-efficiency obtained from the technology application is determined by the volume of the cryogenic hydrogen vessel since each hydrogen re-pumping to the empty vessel results in great losses (up to 10% of total mass). Technology adopting would allow to reduce hydrogen evaporability in storage phase.

The author is grateful to Kuprianov V. I. for useful recommendations and to Yuriev G. V. for the assistance in testing.

Для дальнейшего прочтения статьи необходимо приобрести полный текст. Статьи высылаются в формате PDF на указанную при оплате почту. Время доставки составляет менее 10 минут. Стоимость одной статьи — 150 рублей.

Показать целиком