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 3 9 ( 2 0 1 4 ) 1 2 9 2 4 e1 2 9 3 3

Available online at www.sciencedirect.com

ScienceDirect

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

Experimentally tuned dual stage hydrogen

compressor for improved compression ratio

Md Mainul Hossain Bhuiya a,1, Chi Young Lee a,2, Taeseon Hwang c,
Sirajum Munira a,3, Ryan Hopkins a, Hyungkee Yoon c, Sang Heup Park d,
Kwang J. Kim b,*
a
Mechanical Engineering Department, University of Nevada, Reno (UNR), 1664 N Virginia Street, Reno, NV 89557,
United States
b
Mechanical Engineering Department, University of Nevada, Las Vegas (UNLV), 4505 Maryland Parkway,
Las Vegas, NV 89154-4027, United States
c
Solar/Geo Thermal Research Laboratory, Korea Institute of Energy Research (KIER), 152 Gajeong-ro, Yuseong-gu,
Daejeon 305-343, Republic of Korea
d
Division of Mechanical and Automotive Engineering, College of Engineering, Kongju National University,
275 Budae-dong, Cheonan 330-717, Republic of Korea

article info abstract

Article history: An experiment-driven design procedure for optimizing the combination of stages of a dual
Received 4 February 2014 stage hydrogen compressor with enhanced compression ratio is presented herein. Three
Received in revised form different combinations of reactors were used using LaNi5, Ca0.6Mm0.4Ni5 and Ca0.2Mm0.8Ni5
12 April 2014 as hydrogen storage materials. Compression ratios were found to be similar for low supply
Accepted 31 May 2014 pressure conditions, which improved significantly for high supply pressure conditions in
Available online 9 July 2014 single stage experiments. A dual stage compressor system with LaNi5 in first stage and
Ca0.2Mm0.8Ni5 in the second stage was proposed based on single stage results, which was
Keywords: found to be very effective for enhancing compression ratio. Results show that 53% higher
Metal hydrides compression ratio can be attained by selecting appropriate storage materials for stages,
Hydrogen storage compared to LaNi5 based economic dual stage system.
Thermally driven hydrogen Copyright 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
compressor reserved.
Dual stage hydrogen compressor
Stage optimization

* Corresponding author. Tel.: 1 702 774 1419 (office).

E-mail addresses: bhuiyam@unlv.nevada.edu (M.M.H. Bhuiya), whitelcy270@gmail.coms, chiyounglee@kaeri.re.kr (C.Y. Lee),
Taeseon.hwang@unlv.edu (T. Hwang), sirajum.munira@utexas.edu (S. Munira), hopkinsawai@yahoo.com (R. Hopkins), hgyoon@
kier.re.kr (H. Yoon), weldpark@kongju.ac.kr (S.H. Park), kwang.kim@unlv.edu (K.J. Kim).
1
University of Nevada, Las Vegas (UNLV), 4505 Maryland Parkway, Las Vegas, NV 89154-4027, United States.
2
LWR (Light Water Reactor) Fuel Technology Development Division, KAERI (Korea Atomic Energy Research Institute), Korea.
3
Globalfoundries, 400 Stone Break Road Extension, Malta, NY 12020, United States.
http://dx.doi.org/10.1016/j.ijhydene.2014.05.186
0360-3199/Copyright 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
 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 3 9 ( 2 0 1 4 ) 1 2 9 2 4 e1 2 9 3 3
Introduction
Hydrogen is one of the most anticipated renewable energy
sources for the future. Due to the unique ability of storing
hydrogen at room temperature within the lattice and with
density greater than the liquid hydrogen, metal hydrides (MH)
have long been established as excellent mediums for
hydrogen storage [1e7]. While the kinetics of hydrogen ab-
sorption and desorption remain complex and are greatly
influenced by a number of thermo-physical properties of the
storage material, the processes for hydrogen storage and
compression can be explained by simpler physicochemical
process that takes place during the operation. Hydrogen is
first absorbed into the porous metallic pellets stacked inside
the reactor at low supply pressure and at low temperature. It
remains in the hydride structure until the system is exposed
to an elevated temperature or vacuum, which cause the stored
hydrogen to exit the hydride lattice. If the temperature rise is
sufficient and the system gains the activation energy to
initiate desorption reaction, hydrogen exits the metal hydride
at pressures that range from approximately 3e10 times [8e22]
the original supply pressure of a single stage compressor.
Multistage metal hydride hydrogen compressors use a com-
bination of metal hydrides at different stages to enhance the
final compression ratio by maximizing absorption from the
desorbed hydrogen of prior stages. Recently a dual stage
hydrogen compressor [11,23,24] has been proposed that uti-
lized both AB5 and AB2 metal hydrides to compress hydrogen
from 4 MPa to a final storage pressure of 45 MPa using hot
water as the working fluid. A final compression ratio of
approximately 11 was obtained by this dual stage system.
The working principle of a dual stage compressor can be
described using the simplified illustration of such system
shown in Fig. 1(A). The single stage compressor is also shown
in the inset of the figure. During the operation of a dual stage
compressor, initially all the valves remain closed. When
valve-1 is opened to expose the storage material to low pres-
sure hydrogen at a lower temperature, hydrogen is absorbed
and hydride is formed by an exothermic reaction inside the
Fig. 1 e (A) Schematic diagram and (B) working principle [23,25] based on pressureetemperature data of multi stage metal
hydride hydrogen compression system.
12925
reactor 1. When saturation is reached, valve-1 is closed and
subsequently reactor 1 is heated until the steady state H2
pressure is built inside. In the meantime reactor 2 is being
cooled to prepare it for absorption process. Opening valve-2 at
this point allows the desorbed hydrogen from the first stage,
to be absorbed in the second. Valve-2 is then closed after
saturation is reached and subsequently reactor 2 is being
heated to initiate final desorption process. Finally, valve 3 is
opened to obtain high pressure H2. This working principle
corresponds to the system temperature and desorption pres-
sure as shown in Fig. 1(B) [23]. State A in the figure represents a
lower absorption temperature at which LaNi5 reactor absorbs
and stores hydrogen. As the temperature is increased after
saturation, system pressure gradually increases since
hydrogen desorbs from the storage material. At State B,
desorption pressure reaches the peak and then the high
pressure hydrogen is allowed to flow into the second stage for
absorption representing state C. At this point there is a sig-
nificant decrease in the system pressure since there pres-
surize hydrogen starts to get absorbed in the second stage
storage material and also occupy the plumbing volume be-
tween stages. Finally, the second stage stored hydrogen is
desorbed by application of heat and the system eventually
reaches a final state D where the final pressure is significantly
above the initial absorption pressure at state A. The process as
described can be continued for subsequent cycles as required.
The single stage compressor works in the same principle
except the fact that there is no second reactor and the process
is performed for the first stage only.
The materials selected for storage of hydrogen at different
stages of reactor comprises of three variation of AB5 type
nickel based storage system. They are: LaNi5, Ca0.6Mm0.4Ni5
and Ca0.2Mm0.8Ni5. As shown in Fig. 2, AB5 type (such as Ni
based materials) is comparatively better option because of the
fact that they operate at moderate temperature, which is
within the range of waste heat temperature. Also, from the
economic aspect of the project and availability of the mate-
rials, nickel based AB5 type was the best option that was
considered for the project. Even though AB2 (such as Ti based
materials) types can yield even higher compression ratio, they
12926 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 3 9 ( 2 0 1 4 ) 1 2 9 2 4 e1 2 9 3 3
Fig. 2 e Van't Hoff plots for different metal hydrides [5].
may require extreme low temperatures to operate. It can be
very difficult and extremely expensive to manage such low
temperature operation. Hence, the storage materials of choice
were three different variations of nickel based AB5 type
hydrogen storage system for this investigation.
Heat is the driving force of such hydrogen compressor
system that takes advantage of extreme storage capability of
metal hydrides. Since the process of absorption and
Fig. 3 e Metal hydride powder (A) before, (B) after copper encapsulation process; (C) sample prepared for SEM analysis, and
(D) LaNi5 particle with copper coated layer.
desorption of hydrogen requires removal and supply of heat
required by the reaction mechanisms of the two processes,
thermal conductivity of metal hydride plays a vital role in
enhancing the overall performance of the system. Among
many different procedures reported, two methods may be
mentioned here as they carry close similarity to the procedure
adopted for enhancing thermal conductivity in this study. In
the first method, LaNi5 powder with aluminum as the matrix
material was compacted into cylindrical pellets using high
pressure sintering process. The second method consists of
micro-encapsulation of metallic particle using copper, addi-
tion of a binder and compaction of mixture into pellets using a
high pressure die. Reported thermal conductivity using the
first method ranges from 8 to 23 W/m K [26] while the second
method results in thermal conductivities in the range of
3e6 W/m K [19].
In this study, copper encapsulation technique was
employed, followed by compaction of particles into cylindrical
pellets using a press and die for enhancing thermal conduc-
tivity of the storage material. The manufacturing process of
porous metal hydride compacts (PMHC) are outlined in the
following. Hydrogen compression performances of three types
of metal hydrides, LaNi5, Ca0.6Mm0.4Ni5 and Ca0.2Mm0.8Ni5,
were then investigated using low-grade heat source (T < 125  C)
and different supply pressure conditions for single stage
compressors. Based on single stage experimental results, the
combination of reactors was optimized for maximizing the
compression ratio of the dual stage compressor.
 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 3 9 ( 2 0 1 4 ) 1 2 9 2 4 e1 2 9 3 3
Table 1 e Percentage of copper detected by Energy
Dispersive Spectroscopy (EDS) [27].
Alloy Coating time Cu weight% Cu atomic%
LaNi5 40 s 1.54 6.11
Ca0.6Mm0.4Ni5 40 s 5.03 13.23
Preparation of PMHC
The metal particles selected for the experiment (LaNi5,
Ca0.6Mm0.4Ni5 and Ca0.2Mm0.8Ni5) were obtained from Japan
Metals & Chemical Co. Ltd. The procedure for preparing Cop-
per encapsulated metal particles is essentially mixing of metal
particles in copper enriched chemical solution. First, a solu-
tion of distilled water, copper sulfate and sulfuric acid is
prepared. This mixture is stirred thoroughly at room tem-
perature until the copper sulfate powder completely dissolves
in the solvent. A known amount of metal hydride powder is
then added to this solution and vigorously stirred for a pre-
determined duration of time. Copper coated powder is then
removed and rinsed twice with ethyl alcohol and distilled
water. Finally, the powders are filtered and allowed to dry
overnight at room temperature. At the end of the process, all
the particles would be coated with copper. Fig. 3(A) and (B)
shows metallic particles before and after the process of copper
encapsulation using copper rich chemical solution. To observe
the thickness of the coated later, the coated particles were
attached to a transparent polymeric disc using epoxy as
shown in Fig. 3(C). After curing the particles were fine polished
using alumina particles, which is a standard procedure used
in the metallurgical lab for observing cross-sections of micro
particle. Fig. 3(D) shows the SEM image of the coated particle
depicting a coated copper layer on the LaNi5 particle. The
thickness of the coated layer is approximately 1e2 micron
thick as visualized in the SEM image. Expected outcome of the
coating process is tabulated in Table 1 with process duration
and copper weight percentage as reported in the previous
study from the same group.
Fig. 4 e (A) Die, (B) metal hydride compacts, (C) reactor tube, fittings and compacts with dimensions, and (D) reactor
assembly of a stage.
12927
Metal hydride compacts were prepared using the die as
shown in Fig. 4(A). Each LaNi5 pellet, as shown in Fig. 4(B),
contains 5 g of Cu coated particles and 0.5 g of tin (Sn) binder
which was compressed at a pressure of 17.24 MPa to maximize
thermal conductivity of the metal hydride while maintaining
an optimum porosity as suggested by the experimental results
[27] reported previously. Due to size constraints, pellets of
Ca0.6Mm0.4Ni5 and Ca0.2Mm0.8Ni5 were made using 3 g of Cu
encapsulated metal particles with only 0.3 g of Sn binder. The
mixture was then compressed at 20.68 MPa to form pellets.
The pellets were then stacked inside a stainless steel cylin-
drical tube as shown in Fig. 4(C). Three of such tubes were
connected in parallel to make up a stage of hydrogen storage
for the dual staged compressor as shown in Fig. 4(D). Each of
these tubes can hold 18e20 pellets inside. In the end, all the
pellets went thru an activation process which consists of
exposure to pressurized hydrogen for 30 min and then
extraction of the absorbed gas using vacuum. This process
was repeated until the temperature rise during absorption
process showed consistent behavior.
Experimental details
Schematic diagram of the single stage hydrogen compressor
system is shown in Fig. 5(A). The hydrogen storage reactor
tubes of 0.12 inch thick wall were made from SS-316 which is
rated to withstand an operating pressure up to 55.2 MPa.
Initially the reactors are cooled to absorption favorable tem-
perature. Valve (1) is then opened to let hydrogen flow into the
reactors (9), which exposes the porous metallic storage system
to low pressure hydrogen supply. At this instance the metal
particles start to react with Hydrogen and forms metal hy-
drides through an exothermic reaction. Once the pellets inside
the reactors become saturated with hydrogen, valve (1) is
closed to cut the supply off. Then, Valve (2) is kept open while
the reactor goes through a heating process which results in
desorption of hydrogen. The hydrogen desorbed is then
collected and stored in high pressure storage tanks (7).
12928 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 3 9 ( 2 0 1 4 ) 1 2 9 2 4 e1 2 9 3 3

Fig. 5 e (A) Single stage and (B) Dual stage MH hydrogen compression system.

The design and operation of the dual stage is based on
similar features of a single stage system. For this case, des-
orbed hydrogen of the first stage is the source of supply for the
second, which causes increased supply pressure for the sec-
ond stage. Generally, increase in supply pressure causes to
increase the desorption pressure maintaining a constant
compression ratio. Applying that to a dual stage system sim-
ply implies that the dual stage system will increase the final
pressure of the system by applying a second compression
cycle.
Schematic diagram of a dual stage hydrogen compressor is
shown in Fig. 5(B). Temperature of reactors during absorption
and desorption is maintained using cold and hot liquid cir-
culation, respectively, which is controlled by a high capacity
chiller. First, reactor 1 is cooled to absorption favorable tem-
perature. Hydrogen starts to flow into reactor 1 at a pre-
determined pressure. At this instance hydrogen is being
absorbed in metals forming metal hydride. Absorption in the
first stage ends as soon as the system is saturated with
hydrogen which can be deduced by the flow meter reading
showing zero flow of hydrogen into the reactor. After the end
of absorption in first stage, Reactor 2 is cooled to the same
suitable temperature for absorption to begin. Now the first
reactor is gradually heated to a predetermined temperature to
release all the hydrogen that was being absorbed in the first
place. This high pressure hydrogen is then allowed to flow in
the pre cooled reactor 2 to initiate absorption process at higher
pressure. By the end of absorption in second reactor all the
hydrogen absorbed is driven out again by increasing the
reactor temperature of second stage to a preset level and
subsequently storing the high pressure hydrogen in the stor-
age tank. The actual experimental set-up for the dual stage
metal hydride hydrogen compression system with the reactor
stages immersed in heating/cooling bath is shown in Fig. 6(A).
The components of such system is identified along with the
label descriptions in Fig. 6(B).
Results
Storing hydrogen
Reactor temperature during the charging of hydrogen is a
simple way to identify the end of charging process. The
reactor temperature response during the charging of
hydrogen is shown in Fig. 7. The exothermic hydrogenation

Fig. 6 e (a) Actual dual stage hydrogen compressor system with reactors immersed in temperature controlled pool of liquid;
(b) Components of dual stage system: 1-low pressure inlet valve, 2-hydrogen mass flow meter, 3-vacuum port, 4-
thermocouple, 5-pressure gage, 6-pressure transducer, 7-first stage (reactor 1), 8-valve, 9-high pressure valve separating the
two stages, 10-second stage (reactor 2), 11-inlet/outlet valve, 12-high pressure storage tanks, and 13-storage tank inlet
valve.
 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 3 9 ( 2 0 1 4 ) 1 2 9 2 4 e1 2 9 3 3 12929

Fig. 7 e Temperature response of the reactor during charging of hydrogen at different pressures: (A) 1.38 MPa and (B)
4.14 MPa.

reaction generates heat causing a sharp peak in the reactor
temperature as soon as hydrogen is introduced into the
reactor. Depending on the supply pressure and type of mate-
rial, the amount of heat generated varies which can be real-
ized from the peak temperature reached for each material. At
1.48 MPa supply pressure, LaNi5 reactor reaches about 33  C
whereas about 27  C is reached by the reactors filled with
calcium based storage systems. Initially the reactors remain at
temperature levels similar to the coolant temperature in the
chiller, which is shown as the base temperature in the graphs.
This temperature is hardly affected by the heat of reaction as
there is almost no indication of fluctuations in the chiller fluid
temperature over time. Regardless of the type of material used
for storage, reactions were observed to be reaching saturation
within 500 s. This duration may not represent the charging
duration but most of the material is assumed to become
saturated at that instance for corresponding absorption
pressure. The end of charging was determined when the
reactor temperature and the chiller temperatures appeared
parallel in the real time data logging screen.
Single stage experimental results
Two different supply pressure conditions were applied during
absorption to understand its effect on desorption pressure
and compression ratio. First, absorption was performed at low
supply pressure of 1.38 MPa and at low temperature of
approximately 10  C. Desorption process was performed by
gradually increasing the reactor temperature to a targeted
temperature of 90  C which is within the temperature range of
waste heat. Fig. 8 shows the change in desorption pressure
with temperature for LaNi5 and Ca0.2Mm0.8Ni5 filled reactors.
The curves of both systems are very similar and only above
50  C a subtle difference in the desorption pressures is
observed. Above 50  C, desorption pressure of Ca0.2Mm0.8Ni5 is
slightly higher than that of LaNi5 which might attribute to
higher storage capacity of the latter material. Since the
composition of Ca0.6Mm0.4Ni5 is very similar and the temper-
ature response was observed to be a close match of
Ca0.2Mm0.8Ni5 based system, desorption pressure behavior
was anticipated to be identical. Compression ratios of

7 18
5
6 4
LaNi5 Ca0.2Mm0.8Ni5
4
15 Ca0.6Mm0.2Ni5 108 Co
5 Ca0.2Mm0.8Ni5
3 LaNi5
Compression Ratio

4
Compression Ratio
Pressure (MPa)

Pressure (MPa)

3
12
3
2

2 9
2

1 6
20 40 60 80 100 70 80 90 100 110 120

Temperature (oC) Temperature (oC)

Fig. 8 e Desorption pressure and compression ratio for Fig. 9 e Desorption pressure and compression ratio for
supply pressure of 1.38 MPa. supply pressure of 4.14 MPa.
12930 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 3 9 ( 2 0 1 4 ) 1 2 9 2 4 e1 2 9 3 3
Fig. 10 e (a) Pressure vs. hydrogen to meal atom ratio at 30  C and (b) Pressureetemperature diagram for AB5 type metal
hydrides [21,28].
Ca0.2Mm0.8Ni5 and LaNi5 metal hydride is compared in Fig. 8.
With the increase in desorption temperature, increased
amount of hydrogen is desorbed causing arise in desorption
pressure which directly translate into higher compression
ratio. In other words, compression ratio increases with
increasing desorption temperature. From single stage experi-
ments, compression ratios of Ca0.2Mm0.8Ni5 and LaNi5 at 90  C
were approximately 4.2 and 3.6 respectively.
Fig. 9 illustrates desorption pressure rise with increasing
reactor temperature of LaNi5, Ca0.6Mm0.4Ni5 and Ca0.2Mm0.8Ni5
based storage systems. In this case the absorption process was
performed at 4.14 MPa supply pressure. Higher supply pressure
usually translates to increased amount of hydrogen absorbed
into the storage material. Hence, It is evident that with the
increase in reactor temperature, more hydrogen is desorbed
causing a gradual increase in desorption pressure at levels
higher than the previous case. Ca0.6Mm0.4Ni5 and
Ca0.2Mm0.8Ni5 appear to have higher desorption pressures
compared to LaNi5 over the entire temperature range. It means
that these two systems can store more hydrogen during the
absorption at higher pressures and consequently generate
higher desorption pressure during desorption phase which
results in higher compression ratios compared to LaNi5. Real-
izing that the reactors are now storing increased amount of
hydrogen than the previous case, the desorption temperature
was elevated to the upper range of waste heat source.
Desorption pressures obtained at about 120 C were 10.34 MPa,
13.8 MPa and 15.51 MPa for LaNi5, Ca0.6Mm0.4Ni5 and
Ca0.2Mm0.8Ni5 respectively. Interestingly, Desorption pres-
sures for the latter two calcium based metal hydride systems
are very similar up to about 108 C. Above that temperature,
however, Ca0.6Mm0.8Ni5 shows declining pressure which is an
Table 2 e Probable combinations of dual stage system for enhanced compression ratio.
Cases First stage

MH Operating temperature ( C)
I LaNi5 10e90
II LaNi5 10e90
III LaNi5 10e90
indication of depletion of hydrogen in the reactor. Desorption
pressure of Ca0.2Mm0.4Ni5 keeps increasing gradually even at
temperature above 107  C which can be attributed to higher
absorption or storage capacity of this metal hydride system.
Compression ratios of LaNi5, Ca0.6Mm0.4Ni5 and Ca0.2Mm0.8Ni5
with varying bath temperature are also shown In Fig. 9. At
125  C Compression ratios of LaNi5, Ca0.6Mm0.4Ni5 and
Ca0.2Mm0.8Ni5 are 2.5, 3.4 and 3.9 respectively. Remarkably
Ca0.2Mm0.8Ni5 based system exhibits 56% and 14.7% higher
compression ratio than LaNi5 and Ca0.6Mm0.4Ni5, respectively.
Optimizing dual stage compressor
From the single stage experiments, it is evident that
CayMm1yNi5 hydrides are generally high pressure hydrides as
they absorb and desorb at higher pressure. LaNi5, on the other
hand, is low pressure hydride since it absorbs and desorbs
hydrogen close to atmospheric pressure. Fig. 10(A) shows the
Pressure hydrogen vs. metal to hydrogen atom ratio plot.
CayMm1yNi5 hydrides show high pressure ranges with less
distinct plateau, whereas, LaNi5 has well defined pressure
plateau, which makes it very suitable for integrating in the
design of any engineering system. Fig. 10(B) shows pressure vs
inverse temperature plot in a semi log graph. At room tem-
perature of about 30  C, for LaNi5 system, the pressure is about
2.5 times the atmospheric pressure. CayMm1yNi5 hydrides,
on the other hand, show as high as 30 times the atmospheric
pressure at such temperature. Hence, for Multistage systems,
low pressure hydrides should be considered for the earlier
stages and high pressure hydrides for later stages.
The desorption pressures significantly increases for cases
where higher absorption pressures were implemented. Based
Second stage
MH Operating temperature ( C)
LaNi5 10e125
Ca0.6Mm0.4Ni5 10e125
Ca0.2Mm0.8Ni5 10e125
 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 3 9 ( 2 0 1 4 ) 1 2 9 2 4 e1 2 9 3 3 12931

18 18

Projected
A performance:
Experimental observation:
15 CASE III 15 CASE III
CASE II CASE II
CASE I CASE I

Pressure (MPa)
Pressure (MPa)

12 12

Second Stage
First Stage
9 9

6
Second stage
6
absorption begins

3 3

0 0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
(A) Time (Hours) (B) Time (Hours)

Fig. 11 e (A) System potential, and (B) actual performance of a dual stage system.

on single stage experimental results, at low absorption pres- the first stage, final desorption pressure in 2nd stage may
sure (e.g. 1.38 MPa), desorption pressure of LaNi5 is somewhat reach 10.34 MPa, 13.8 MPa and 15.86 MPa for case I, II and III
similar to the other metal hydride systems. Moreover, pellets respectively for the best system design where there would be
of LaNi5 reactor system are easy to manufacture, economic no drop in pressure during the transition period from first
and has small hysteresis if repeated absorption and desorp- stage to the second. In ideal scenario, there wouldn't be any
tion of hydrogen is performed. For all these reasons LaNi5 was pressure drop due to the connecting/plumbing volume in be-
used in the first stage of the compressor. Since both tween the reactors. It can be achieved by minimizing the
Ca0.6Mm0.4Ni5 and Ca0.2Mm0.8Ni5 showed higher compression connecting volume or integrating a pressure reservoir that can
ratio than LaNi5 for cases when high absorption pressures maintain desorption pressure levels in the plumbing volume
were supplied, they are the best candidates for the second between two stages. Hence, the pressure drop can be dimin-
stage of the dual stage hydrogen compression system. ished. The sudden pressure drop can only be caused by the
Nevertheless, three logical variations of dual stage sudden increase in volume that connects the stages of reactor
compressor with LaNi5 in the first stage are further investi- if there is no leak in the system. Thus, minimizing the
gated for maximizing compression ratio of the system. The plumbing volume or maintaining desorption pressure levels
arrangements are listed in Table 2. in the connecting spaces will enable this system to operate
ideally at its highest potential. In such ideal case overall
Dual stage experimental results compression ratios of 15, 20 and 23 can be attained for case I, II
and III, which would be a staggering 36% and 57% approximate
For dual stage experimentation, absorption pressure was set increase in compression ratios for case II and III, respectively,
at 0.69 MPa for first stage and the temperatures of cooling and compared to case I projection.
heating were set at 10 and 90  C respectively. The results of the Based on the experimental results, however, case III (LaNi5
dual stage compression system are showed in Fig. 11. Both and Ca0.2Mm0.8Ni5) shows the best compression performance,
Ca0.6Mm0.4Ni5 and Ca0.2Mm0.8Ni5 shows similar desorption which is 53% higher in terms of compression ratio compared to
pressure of approximately 7.6 MPa at the end of the second
stage desorption. Thus, overall compression ratio is approxi-
mately 11. Interestingly when the absorption process begins
in second stage, there is a significant drop in supply pressure
for second stage. This is due to the fact that there is a signif-
icant increase in volume as soon as the valve to second reactor
is opened. This volume comprises of the empty volume of the
2nd reactor as well as the plumbing volume that occupies the
path between the two stages. Due to the sudden increase in
volume, as soon as the valve on top of the first reactor opens,
supply pressure to the 2nd stage drops to about 1.72 MPa
which is not sufficient to exceed the dehydriding pressure of
8.27 MPa at the end of 2nd stage desorption. If constant
pressure can be maintained in the system after the end of first
stage desorption and at the beginning of second stage ab-
sorption, compression ratios of 15e23 can be achieved for dual
stage compressor system based on the compression ratio
recorded during the single stage experiments. As illustrated in Fig. 12 e Attainable and experimental compression ratio of
Fig. 11(A), starting with the absorption pressure of 0.7 MPa in the dual stage hydrogen compressor.
12932 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 3 9 ( 2 0 1 4 ) 1 2 9 2 4 e1 2 9 3 3
case I. Case II also showed a 39.7% increase in the experimental
results compared to case I. Fig. 12 shows a comparison chart for
projected and experimentally determined compression ratio
obtained from the dual stage hydrogen compressor system for
three different combinations. Even though the trend is similar,
there is a significant gap in expected outcome and actual per-
formance. As discussed earlier, the cause lies in the sudden
increase in the plumbing volume in between two stages, which
can be easily minimized to improve system performance.
Conclusions
For single stage compression system, at the low supply pres-
sure of 1.38 MPa, compression performance of Ca0.2Mm0.8Ni5
is similar to that of LaNi5. On the other hand, at supply pres-
sure of 4.14 MPa and dehydriding temperature of 125  C,
Ca0.2Mm0.8Ni5 exhibits 56% and 14.7% higher compression
ratios than that of LaNi5 and Ca0.6Mm0.4Ni5 respectively.
Design aspects and selection of metal hydride have a great
influence on the performance of dual stage thermally driven
hydrogen compressor. By minimizing pressure drop between
stages by efficient design and reducing plumbing volume,
compression ratio can be enhanced significantly. Combina-
tion of LaNi5 in the first stage and Ca0.2Mm0.8Ni5 in the second
is the optimum configuration for maximum overall
compression ratio that can be obtained using a dual stage
thermally driven hydrogen compression system.
The procedure demonstrated in this study can be used for a
wide variety of hydrogen storage materials in order to identify
their single stage behavior and consequently optimize the
combination of stages for enhanced overall compression ratio
within the applicable temperature range.
Thus, higher compression ratio can be obtained from dual
stage thermally driven compressors using low-grade waste
heat for various industrial applications. Also, it can be
deduced that proper selection of metal hydride and efficient
design are important aspects in enhancing the performance of
dual stage thermally driven hydrogen compressor system.
Acknowledgments
The authors extend sincerely thanks for the financial support
from the National Renewable Energy Laboratory, Office of En-
ergy Efficiency and Renewable Energy (NREL) (NAX-9-66014-06)
and special thanks to Ms. Lynn Billman of NREL. Also, the
support for Drs. H. Yoon and S. H. Park was provided by Korea
Institute of Energy Research and Kongju National University,
respectively, for their sabbatical at University of Nevada, Reno.
references
[1] Klebanoff L, Keller J. 5 Years of hydrogen storage research in
the US DOE Metal Hydride Center of Excellence (MHCoE). Int J
Hydrogen Energy 2013;38:4533e76.
[2] Schlapbach L, Zuttel A. Hydrogen-storage materials for
mobile applications. Nature 2001;414(6861):353e8.
[3] Folonari C, Iemmi G, Manfredi F. Metal hydride fuel cells: a
feasibility study and perspectives for vehicular
applications,. J Less Common Met 1980;74(2):371e8.
[4] Reilly JJ, Sandrock GD. Hydrogen storage in metal hydrides.
Sci Am 1980;242:118e29.
[5] Schuth F, Bogdanovic B, Felderhoff M. Light metal hydrides
and complex hydrides for hydrogen storage. Chem Commun
2004;0(20):2249e58.
[6] Sakai T, Honda N, Yonezu I. Metalhydride container and
metal hydride heat storage system. J Heat Recover Syst
1986;6(1):vii.
[7] Sandrock GD, Snape E. Am Chem Soc 1980:293e322.
[8] Kelly NA, Girdwood R. Evaluation of a thermally-driven
metal-hydride-based hydrogen compressor. Int J Hydrogen
Energy 2012;37(14):10898e916.
[9] Laurencelle F, Dehouche Z, Morin F. Experimental study on a
metal hydride based hydrogen compressor. J Alloy Compd
2009;475(1e2):810e6.
[10] Muthukumar P, Prakash Maiya M, Srinivasa Murthy S.
Experiments on a metal hydride based hydrogen
compressor. Int J Hydrogen Energy 2005;30(8):879e92.
[11] Wang X, Chen R, Zhang Y. Hydrogen storage alloys for high-
pressure suprapure hydrogen compressor. J Alloy Compd
2006;420(1e2):322e5.
[12] Wang XH, Bei YY, Song XC. Investigation on high-pressure
metal hydride hydrogen compressors. Int J Hydrogen Energy
2007;32(16):4011e5.
[13] Feng Z, Deyou B, Lijun J. Metal hydride Compressor and its
application in cryogenic technology. J Alloy Compd
1995;231(1e2):907e9.
[14] Muthukumar P, Prakash Maiya M, Murthy SS. Parametric
studies on a metal hydride based single stage hydrogen
compressor. Int J Hydrogen Energy 2002;27(10):1083e92.
[15] Kim J, Park I, Kim KJ. A hydrogen-compression system using
porous metal hydride pellets of. Int J Hydrogen Energy
2008;33(2):870e7.
[16] Luo G, Chen JP, Li SL. Properties of La0.2Y0.8Ni5xMnx alloys
for high-pressure hydrogen compressor. Int J Hydrogen
Energy 2010;35(15):8262e7.
[17] Lototskyy M, Klochko Y, Linkov V. Thermally driven metal
hydride hydrogen compressor for medium-scale
applications. Energy Procedia 2012;29(0):347e56.
[18] Muthukumar P, Prakash Maiya M, Srinivasa Murthy S.
Performance tests on a thermally operated hydrogen
compressor. Int J Hydrogen Energy 2008;33(1):463e9.
[19] Kim KJ, Lloyd G, Razani A. Development of LaNi5/Cu/Sn
metal hydride powder composites. Powder Technol
1998;99(1):40e5.
[20] Vanderhoff Alexandra, Kwang JK. Experimental study of a
metal hydride driven braided artificial pneumatic muscle.
Smart Mater Struct 2009;18(12):125014.
[21] Park I, Kim J, Kim KJ. Investigation of coupled AB5 type high-
power metal hydride reactors. Int J Hydrogen Energy
2009;34(14):5770e7.
[22] Lloyd GM, Kim KJ. Smart hydrogen/metal hydride actuator.
Int J Hydrogen Energy 2007;32(2):247e55.
[23] Hopkins RR, Kim KJ. Hydrogen compression characteristics
of a dual stage thermal compressor system utilizing LaNi5
and Ca0.6Mm0.4Ni5 as the working metal hydrides. Int J
Hydrogen Energy 2010;35(11):5693e702.
[24] Talagan ~ is BA, Esquivel MR, Meyer G. A two-stage hydrogen
compressor based on (La,Ce,Nd,Pr)Ni5 intermetallics
obtained by low energy mechanical alloying e low
temperature annealing treatment. Int J Hydrogen Energy
2009;34(4):2062e8.
 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 3 9 ( 2 0 1 4 ) 1 2 9 2 4 e1 2 9 3 3 12933

[25] Bhuiya MMH, Lee CY, Hopkins R. A high-performance dual-
stage hydrogen compressor system using Ca0.2Mm0.8Ni5
metal hydride. In: ASME 2011 Conference on Smart
Materials, Adaptive Structures and Intelligent Systems.
ASME; 2011. pp. 745e51.
[26] Ron M, Bershadsky E, Josephy Y. The thermal conductivity of
porous metal matrix hydride compacts. J Less Common Met
1991;172e174(Part 3(0)):1138e46.
[27] Lee M, Kim KJ, Hopkins RR. Thermal conductivity
measurements of copper-coated metal hydrides (LaNi5,
Ca0.6Mm0.4Ni5, and LaNi4.75Al0.25) for use in metal hydride
hydrogen compression systems. Int J Hydrogen Energy
2009;34(7):3185e90.
[28] Dantzer P, Orgaz E. Thermodynamics of hydride chemical
heat pumpdII. How to select a pair of alloys. Int J Hydrogen
Energy 1986;11(12):797e806.