Applied Energy 212 (2018) 57–83

Contents lists available at ScienceDirect

Applied Energy
journal homepage: www.elsevier.com/locate/apenergy

Large-scale liquid hydrogen production methods and approaches: A review T

⁎
Majid Aasadnia, Mehdi Mehrpooya
Renewable Energies and Environment Department, Faculty of New Sciences and Technologies, University of Tehran, Tehran, Iran
Hydrogen and Fuel Cell Laboratory, Faculty of New Sciences and Technologies, University of Tehran, Tehran, Iran

H I G H L I G H T S

• Large-scale hydrogen liquefaction (LHL) methods are chronicled.

• AHybrid
novel classification of hydrogen liquefaction systems is introduced.
• Specificconceptual hydrogen liquefaction plants are reviewed.
• energy consumption (SEC) of hydrogen liquefiers is discussed.

A R T I C L E I N F O A B S T R A C T

Keywords: Large-scale hydrogen liquefaction (LHL) methods and different approaches of the configuration of hydrogen
Liquid hydrogen liquefaction cycles are chronicled. History landmarks of permanent gases liquefaction are quick reviewed and
Hydrogen liquefier the basic hydrogen liquefaction cycles, the existing in-service LHL plants around the world, and LHL conceptual
Large-scale hydrogen liquefaction proposed plants, including the state of the art plants, are recorded and categorized based on the systems’ main
Hybrid production
parameters. In addition, a novel classification of hydrogen liquefaction systems in terms of heat exchange and
Exergy analysis
expansion process method is introduced. As well as, the authors infer that renewable energy technologies section
Renewable energy
should be added to the old sectioning of the hydrogen liquefaction plants. In addition, hybrid conceptual hy-
drogen liquefaction plants, combining with renewable power cycles are reviewed and the increasing contribu-
tion of this new approach is demonstrated. Finally, the operational costs of the plants considering the systems’
efficiency are examined, and a trend in specific energy consumption (SEC) and exergy efficiency of hydrogen
liquefiers is discussed. Accordingly, considering the existing technologies, SEC reduction of hydrogen lique-
faction will not be abrupt in near future and it will remain within the range of 5–8 kWh/kg LH2 . Moreover,
exploiting of isentropic expansion processes instead of isenthalpic one, cascading of refrigerating cycles, using of
new mixed refrigerants as working fluid of refrigeration cycles, and hybridization of renewable energy power
cycles to refrigeration cycles are the main four growing approaches in the hydrogen liquefaction context.

1. Introduction privileged energy sources will mitigate concerns arising from damage to
the environment [12–14].
Hydrogen is an abundant and accessible element, which is also Moreover, hydrogen liquefaction provides higher storage densities.
known as the final optimum fuel [1,2]. In addition, liquid hydrogen has Therefore, it is a suitable solution making possible large-scale hydrogen
unique characteristics such as lower weight and volume and higher storage and long distance transportation [15–20]. Finally, demand for
energy content than the gaseous hydrogen. Therefore, hydrogen is the liquid hydrogen, driven by environmental friendly applications and
most promising energy carrier for storage in chemical form within the numerous other uses [21] will increase in the near future. As the result,
large energy storage systems [3–5]. LHL plants will play an important role within clean energy supply chain
Large energy storage systems can eliminate the problem of energy [22,23].
demand fluctuations of renewable energy grids [6–8] by storing excess Hydrogen liquefaction is an energy-intensive process and low effi-
produced energy and compensating energy demand deficit [9–11]. ciency of equipment and processes are the obstacles to the realization of
Therefore, the development of hydrogen energy storage systems will a hydrogen economy. Patterns and trends of LHL systems’ SEC and
facilitate the evolution of renewable energy sources utilizing. These exergy efficiency changes are investigated. Opportunities for

⁎
Corresponding author at: Renewable Energies Department, Faculty of New Sciences and Technologies, University of Tehran, Tehran, Iran.
E-mail address: [email protected] (M. Mehrpooya).

https://doi.org/10.1016/j.apenergy.2017.12.033
Received 21 September 2017; Received in revised form 22 November 2017; Accepted 3 December 2017
Available online 08 December 2017
0306-2619/ © 2017 Elsevier Ltd. All rights reserved.
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

Nomenclature Max maximum

Config configuration
Symbols LH Linde-Hampson
Sp liquid separator
ex specific exergy, kJ/kg BFD block flow diagram
ṁ mass flow rate, kg/s Eff. efficiency
h specific enthalpy, kJ/kg ARS absorption refrigeration system
P pressure ALSRC auto-cascade solar Rankine cycle
s specific entropy, kJ/kg−K HX heat exchanger
T temperature SEC specific energy consumption
W power, kW Sat saturated
w specific work/energy requirement, kJ/kg LH2 or kWh/kg LH2 n- normal
Q̇ heat flow, kW p- para
Ẇ Work transfer rate, kW o- ortho
PV/T photovoltaic/Thermal
Abbreviations Ref. reference
i or in inlet
MR mixed refrigerant o out or outlet
C compressor US United States
Comp compressors SLH simple Linde-Hampson
JB Joule-Brayton SC simple Claude
LH2 liquid hydrogen PFD process flow diagram
GH2 gaseous hydrogen TPD tons per day
LN2 liquid nitrogen N2 nitrogen molecule
LNG liquid natural gas
NGL natural gas liquids Subscripts
WE-NET world energy network system
EX expander Comp of the compressors
J-B Joule-Brayton in of the inlet stream
Sys system geo of the geothermal system
L-H Linde-Hampson a, b, … to f of a, b, … to f
H2 hydrogen molecule 1, 2, … to n of the numbers: 1, 2, … to n/of stream numbers: 1, 2,
o-p ortho-para … to n/ of equipment numbers: 1, 2, … to n
Min minimum

improvements in the LHL plant's performance are identified through Vapor

this way. As well as, an overview of status of the renewable energy
technologies, as a growing section, in the processes of liquid hydrogen
production is obtained. Finally, review of the historical development of
the large-scale hydrogen liquefaction cycles clarifies the new area of Liquid
investigations for LHL plants improvement. Large-scale hydrogen li-
quefaction methods and different approaches in liquefiers’ configura- Salt and ice
tion are reviewed.

2. Initial preparations prior to hydrogen liquefaction

Fig. 1. Faraday bent glass tube [33]. The compression pressure is generated by the gas
The feeling that some things are warmer than others and some are
production in the closed vessel.
cooler dates back to the dawn of the human history. However, it may be
argued that the thermometers were invented to quantify this difference.
The emerging of the thermometer is reviewed [24]. Results introduce experiments to liquefy gases. In 1823, Faraday presented a method of
Galileo as one of the possible inventors of the thermometer in 1592. gas liquefying using a bent glass tube [32]. Accordingly, he filled the
This invention can be considered as the beginning of the thermo- thick-walled tube shown in Fig. 1 with chlorine gas while heating one
dynamics science and emerging of the question whether there is a lower end and cooling the other end. However, the heated gas is expanded,
limit for the temperature. The existence of the absolute zero is predicted and it pressurized the gas in the cold end [33]. Therefore, chlorine is
in 1704 [25,26]. Gases liquefaction is one of the main and important liquefied combining the pressure and the cooling effect. Faraday li-
sections of the history of the quest to achieve absolute zero tempera- quefied ammonia and carbon dioxide using the same method [34],
ture. however, was unable to liquefy oxygen, hydrogen, and nitrogen in this
The story commences almost from the beginning of the 21 century, way. Moreover, in 1844, the experiments are repeated on a large scale
when experiments with the gases that had been discovered a few dec- by Thilorier [35] using a mechanical compressor [36,37]. As the re-
ades earlier, were great interest to the researchers [27]. Accordingly, sults, compressors are developed to produce greater quantities.
liquefaction of gases by cooling and compressing was a challenge that is In 1845, a hydraulic pump is used to make larger quantities of liquid
attempted by Sir Humphry Davy at the royal institute in London and solid carbon dioxide by Faraday [38]. According to the Faraday,
[28,29]. However, in 1812, he employed Michael Faraday [30,31], the compressing of some gases is lead to liquefaction and six known gases
founder of the development of industrial gases, who carried out many including oxygen, nitrogen, carbon monoxide, nitrous oxide, methane,

58
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

and hydrogen are called permanent gases and supposed to be non- system is likely the first multilevel temperature cascade cryogenic
condensable. Helium that it is discovered later in 1869 is not considered system, which inspired Dewar [56] and Onnes [57–59], the ones who
[39,40]. The normal boiling temperatures of these permanent gases are first liquefied hydrogen and helium respectively.
known as cryogenic temperatures. The term “cryogenic” is first in- However, oxygen in the liquid form and not in the mist form is
vented in 1894 employing it in a paper entitled, “On the cryogenic produced in 1883, nearly 20 years after the emergence of liquid oxygen
laboratory at Leiden and on the production of very low temperatures” mist [60]. Results show that a modified apparatus of Cailletet’s one is
[41]. As well as, cryogenics can be defined as the combining of ther- used adding an inverted U to the glass tube and reducing the ethylene
modynamic science, engineering knowledge, and research and devel- temperature by creating a negative relative pressure above it. Oxygen
opment area that dealing with temperatures less than about 120 K liquefaction, as an exciting scientific achievement, started a new session
[42–44]. in accessing to lower temperature race [53].
However, the critical temperature is introduced in 1863 and so an Before continuing, two epochal equipment should be pointed out
opening is developed [45]. Investigations revealed that gases could not that are important landmarks in the historical review of the pioneers of
be liquefied unless by decreasing the temperature of the gas below the cryogenics. First, Vacuum receiver known as “Dewar’s dewar” [61], an
critical temperature. In addition, the reason for the failure to liquefy the admirable invention, is used extensively in the subsequent competitions
permanent gases is well described [45–48]. Based on the findings, any for the liquefaction of gases [62]. It is a double wall reservoir, skillfully
gas is called “vapor” at a temperature below the critical point and is made from an inner silver-coated chamber surrounded by an external
called “gas” if it is above this temperature. Accordingly, a simple in- one, which the volume of the interstitial space between the two
creasing in any vapor pressure leads to liquefaction and coexistence of chambers is evacuated to improve the thermal insulation [56]. Results
vapor-liquid equilibrium. Therefore, the insufficient pressure is re- reveal that here is the first time that vacuum insulation is applied for
cognized as the only barrier for the liquefaction of oxygen and nitrogen, the storing and experimenting of liquefied gasses. Another pivotal ap-
and it is stipulated in 1870 that hydrogen would be liquefied if it were paratus that influenced rivals, is the recuperative contour flow heat
cooled enough [49]. exchanger that its concept is developed first in 1851 and improved in
Discovering of cooling production from expansion is another mile- 1857 [63].
stone in cryogenic history. Temperature reduction in experiments on The first serious use of the recuperative heat exchangers is com-
sound velocity measuring in different gases passing capillary tube, is menced for air liquefaction in 1895 by Hampson [64]. Research shows
ignored as experimental errors in 1850 [49]. However, Joule-Thomson replacing of cascade bath used previously for pre-cooling oxygen by
effect is presented to the knowledge of thermodynamics very soon in recuperative heat exchangers. As well as, a Joule-Thompson valve
1852 [50]. Shortly, an elementary expansion mechanism is used in the combined with the heat exchanger is used to decrease temperature
refrigerating apparatus of Siemens [51]. As the author descriptions, the through expansion. The liquefaction rate is reported to be 1 L/h . The
expansion of air in a cylinder or a suitable engine leads to lowering of use of the Siemens and Kirk engine is examined by Linde in Germany
its temperature. In 1860, first Stirling cooling system is built by Kirk [65]. However, lubrication problems motivated him to use the Joule-
[52]. Accordingly, an elementary cooling mechanism similar to Sie- Thomson effect. Therefore, a throttle device is substituted instead of
mens’ one is used and interchanging function accomplished by a re- Siemens’ expander similar to the elementary cooling mechanism [66].
generator. Finally, at the same time, air is liquefied in an industrial scale by Linde
In 1877, using the Joule-Thomson effect and considering critical [67]. As the results, the experiment is based on the Joule-Thompson
temperature, oxygen is first liquefied in the form of a mist by [53,54]. effect and it can be considered as the beginning of a new era of modern
As can be seen in Fig. 2, Cailletet pressurized oxygen and carbon liquefaction. Fig. 4 shows a laboratory type of Linde apparatus of air
monoxide to 300 atm, pre-cooled them to −29 °C and finally let the liquefaction [68]. A modification of Linde-Hampson cycle is presented
gases to expand rapidly. In addition, results reveal that Faraday and in 1902 by Claude for air liquefaction [69]. Results show the first
Perkins [55] techniques are used together with a cascade of pre-cooling successful use of an expansion turbine instead of an expansion valve in
baths shown in Fig. 3, by Pictet. This early step by step refrigeration the cryogenics. The so-called Claude cycle had higher efficiency and

Fig. 2. Cailletet large liquefaction apparatus of

oxygen liquefaction [53].

59
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

Fig. 3. Pictet multilevel temperature cascade ap-

paratus of oxygen liquefaction [27].

pressure to atmospheric pressure. However, he published accurate data

about the hydrogen physical properties [75]. In 1895, Olszewski con-
firmed these experiments by repeating them on a larger scale [76]. The
concept of hydrogen liquefaction, using an expansion turbine, is in-
troduced by Onnes in 1896 [70,77]. Results imply that technical
shortages forced him to give up the concept. Notably, even as praise-
worthy a man as James Clerk Maxwell doubted the possibility of hy-
drogen liquefying [78]. However, hydrogen is liquefied for the first
time in 1898 by Sir James Dewar with 4 cc/min flow rate [79]. As can
be seen in Fig. 5, Dewar pressurized gaseous hydrogen to 180 bar ,
precooled it to −250 °C by carbonic acid and liquid air, and finally
expanded it to atmosphere in Dewar’s flask. The system is similar to the
Linde system for the liquefaction of air and the yield of the liquid is
about 1 percent of the gas.
Finally, rivalry among existing competitors for the permanent gases

Fig. 4. A laboratory-type of Linde apparatus of air liquefaction [68].

lower pressure. Linde carried out air separation using cryogenic tech-
nology in the same year [70].
Trying to win another scoring in the permanent gases liquefying
came to fruition by hydrogen liquefaction. In 1877, Cailletet claimed to
see hydrogen mist [71] and Meunier confirmed him [72]. However,
Cailletet admitted that he had not seen anything special during the first
test of hydrogen liquefaction [53]. As well as, in 1878, Pictet claimed
that he had employed his cascade system to liquefy and even solidify
hydrogen. However, other researchers strongly rejected these claims
[73]. In l884, Wroblewski created a transient vapor of hydrogen [74]. Fig. 5. Dewar hydrogen liquefaction apparatus [68]. (A) Compressed hydrogen Cylinder,
As the results, pre-cooled compressed hydrogen expanded into a ca- (B, and C) Liquid carbonic acid and liquid air vacuumed vessels of hydrogen pre-cooling,
pillary glass tube, from liquid oxygen temperature and 100 atm (D) Regenerating coil in the vacuumed vessel.

60
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

liquefaction came to the end by helium liquefaction by Onnes in 1908 the hydrogen liquefaction and guidelines for the design of high-effi-
[80–82]. The last reference points out many preliminary laboratory ciency plants. Ortho-para conversion effect on the system design and
activities around the world that provided the prerequisites for helium rate of production is presented as one of the main challenges. The im-
liquefaction. In addition, it focusses on the effect of the scarcity of portance of the ortho-para conversion is explained in Section 4. Three
natural sources on the development of cryogenics. other challenges are presented as the large variation of specific heat of
Fig. 6 depicts the more important historical events in the terms of hydrogen over cryogenic temperatures, the impact of high sonic velo-
time and temperature. As can be seen, the time interval between the city of hydrogen on turbomachines design, and related cost of safety
carbon dioxide liquefaction by Faraday and oxygen liquefaction by problems. In addition, increasing of refrigeration levels or cascading
Wroblewski and Olszewski is roughly 2.3 times the time interval be- and using of expansion devices are recommended. SEC improvement
tween the oxygen liquefaction and the helium liquefaction by Onnes. trend of hydrogen liquefier systems is illustrated from 1898 to 2009
Therefore, it reveals the accelerated rate of cryogenic knowledge pro- [98]. The significant result reveals that the main method used for in-
gress in the second half of the nineteenth century. dustrial and proposed hydrogen liquefaction plant is the same method
used in the classic cycles, which is still the same as 50 years ago.
3. More historical details about pioneers of hydrogen liquefaction
4. Ortho-para conversion challenge within hydrogen liquefaction
Too many references may be investigated to study the details of
each epoch considered. Methods of refrigeration are classified in the In 1912, a weird phenomenon is discovered during the experiments
temperature range of 4–300 K [83]. According to the results, re- works with hydrogen at low temperatures that knowledge of physics
frigeration systems are categorized into two major groups: steady state did not has an explanation for it at that time [99]. Accordingly, while
and periodic systems, which the latter concept is usually used in min- measuring heat capacities of hydrogen at cryogenic temperatures, a
iature cryocoolers. Miniature cryocoolers survey is outside of the scope distinct hysteresis is observed comparing the cooling curve and the
of this review and may be studied elsewhere [84–87]. In addition, warming curve. In 1927, the existence of two spin isomers of hydrogen
Joule–Thomson (Linde-Hampson cycle for refrigerating and not lique- is postulated [100–102]. In 1929, the truth of the hypothesis is proved
fying), Joule-Brayton (vapor compression cycle with turbo-expansion), by experimental evidence [103]. Accordingly, two isomeric forms of
and Claude recuperative cryocooler cycles are presented as the systems molecular hydrogen exist. As can be seen in Fig. 7, two proton nuclear
that are utilized in the temperature range of 4–120 k. Moreover, a se- spins are in the same direction in ortho-hydrogen and they are in the
lection guidance table for refrigeration methods is concluded and multi- opposite direction in para-hydrogen. These two different directions of
stage Claude cycle with pre-cooling is proposed for hydrogen lique- spins affect magnetic, optical, and thermal properties of the two iso-
faction. Cryocoolers are classified based on their working fluid role mers [104,105].
[88]. Results express that the system is called a Claude cycle liquefier
when the working fluid gas is liquefied after final expansion through a 4.1. Thermal aspects of the ortho-para equilibrium concentration
Joule-Thomson valve. However, it is called a Joule-Bryton cycle re-
frigerator, when the final expansion is through an expansion engine and Total rotational energy of the two spin isomers of molecular hy-
the system is used as a refrigerator rather than a liquefier. Internal drogen are not same and the ortho-hydrogen structure is an excited
cooling and external cooling are introduced as two main types of the state with a higher energy level than para-hydrogen [106]. Therefore,
processes used for supplying the required cooling [89]. Results explain energy is released when ortho-hydrogen is converted to para-hydrogen.
that throttling or expansion are internal cooling occurring without heat In addition, Fig. 8 from Ref. [107] depicts the equilibrium concentra-
removal. Table 1 lists various classification criteria of refrigeration tion of ortho and para hydrogen in the ideal gas state versus tempera-
systems using the mentioned references and Ref. [90]. It may be used ture. The effect of pressure on these equilibrium concentrations is
for initial screening and provides a good overview of the employed considered to be negligible [108]. Para-hydrogen concentration de-
techniques. creases with temperature. Moreover, the equilibrium composition is
History of industrial gases is well explained [27]. Accordingly, almost pure para-hydrogen, near the boiling point. At the ambient
history of exploration, liquefaction, usage and applications, storage, temperature that is called normal hydrogen, it contains 75% ortho-
and transportation of industrial gases are reviewed in a detailed survey. hydrogen and 25% para-hydrogen.
Improvement of cryogenics prior to 1950 is briefly reviewed [91]. Re-
sults show that thermodynamic fundamentals of cryogenic knowledge 4.2. The necessity of o-p conversion in the hydrogen liquefaction processes
are developed from 1850 to 1900 and gas liquefaction technologies are
improved from 1900 to 1950. Development of large-scale production Challenges associated with storage and handling of liquid hydrogen
processes used for oxygen, nitrogen, hydrogen, and natural gas lique- are investigated [109–112]. Since the energy level of the ortho-
faction is reviewed over a fifty years period before 1950 [92]. Results
describe developments that result in improvement in efficiency, the
scale of application, and economics of systems. Cryogenics from 1880
to 1940 is reviewed [93]. The review focus on the first helium lique-
faction by Onnes and long necessary preparations before it. Cryogenics
history is reviewed from beginning to 1911 [94]. The review is started
from the initiating of the new science of low-temperature physics in
Onnes laboratory in Leiden until superconductivity phenomenon dis-
covery in 1911. The importance of achievements of the pioneer of
oxygen liquefaction is explained [95]. Results show similarities of
Cailletet’s method for gas liquefaction with the production of very low
temperatures down to the order of micro-Kelvin by nuclear adiabatic
demagnetization. A selective choice from the one hundred years history
of refrigeration is provided [96]. Results introduce Cailletet and Pictet
as the founders of all subsequent refrigeration methods, even of mag-
netic and nuclear cooling. A rapid review on hydrogen liquefiers is
Fig. 6. Historical development of the gas liquefaction Knowledge.
carried out from 1950 to 1959 [97]. Results consider challenges facing

61
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

Table 1
An overview of refrigeration systems classification criteria.

Classification criteria Remarks

Steady state/periodic cycles [83] Vapor compression, turbine-based compression, vapor absorption, and thermoelectric refrigerators are steady state and
regenerative, magnetic and adsorption cycles are periodic
Open/close cycles [90] In open cycles, the cryogen liquid is expendable and usually is vented to the atmosphere after cold recovery. They are used
extensively in space and military applications
Refrigerator/liquefier cycles (based on working The Joule-Bryton cycle is used as a refrigerator rather than a liquefier and in the Claude cycle liquefier, working fluid gas is
flow role) [88] liquefied
Mechanical/non-mechanical cycles [90] A system with no moving parts is non-mechanical, similar to thermoelectric, magnetic, sorption, and radiant coolers
Recuperative/regenerative cycles [90] In recuperative systems, the refrigerant flows steadily in on direction. Regenerative systems include at least a regenerative
heat exchanger similar to Stirling, Vuilleumier, Gifford-MacMahon, Solvy, Postle, and pulse tube refrigeration systems
Internal/external cooling based cycles [89] Throttling or expansion are internal cooling and in external cooling heat is transferred to an external system whose
temperature is below that of the object to be cooled
Temperature range of the cycles [89] Three ranges may be distinguished: moderately low temperature for air conditioning, food storage, etc. (300 k to 120 k),
cryogenic temperature for air separation, gas liquefaction, etc. (120 k to 4 k), ultralow-temperature for physical
experiments (below 4 k)
Cycles purposes [89] Three purposes may be identified: refrigerating, gas liquefying, and gas separating
The cooling principle used in the cycles [89] Four cooling methods depending on the parameter to be varied are recognized: thermomechanical, magnetocaloric, and
electrocaloric methods and methods that use quantum properties of helium isotopes

hydrogen is higher than para-hydrogen, spontaneous conversion of the 100

Para-hydrogen consentration (%)

ortho-hydrogen to para-hydrogen is always occurring to get the equi- 90
librium concentration. However, such a fast conversion is exothermic 80
and the generated heat is about 0.388 kcal/mol while latent heat of
70
hydrogen evaporation is approximately 0.213 kcal/mol . Therefore, if
stored liquid hydrogen contains ortho-hydrogen, some liquid hydrogen 60
will be evaporated. This phenomenon is called “boil off” [113]. As the 50
result, to minimize the evaporation in the storage tank of liquid hy-
40
drogen, the process of o-p conversion is essential and necessary for long
time storage [114]. So, most liquefaction processes have o-p conversion 30
sections to reduce boil off losses [115,116]. Special catalysts are used in 20
different ways to accelerate the o-p conversion of hydrogen 10 20 30 40 50 60 70 80 90 100 120 150 200 250 300
[106,117,118]. In addition, addressing of safety issues of LH2 supply Temprature (k)
chain, risk assessment in production, storage and distribution stages of Fig. 8. Equilibrium para-hydrogen concentration versus temperature.
LH2 , and analysis of the causes and consequences of accident scenarios
are another important topics which can be investigated [119–121].
liquefaction processes can be classified according to the maximum in-
version temperature of the gas to be liquefied in comparison with
ambient temperature. Moreover, since cooling takes place on throttling
5. Basic cycles of hydrogen liquefaction review
only if the maximum inversion temperature of the gas is higher than the
ambient temperature, the pre-cooling section is a necessary merged
The working principles of the basic cycles used for large hydrogen
sub-cycle of the LHL systems. On the other hand, as the maximum in-
liquefaction and simple configuration of them are described [122–127].
version temperature of the hydrogen is lesser than the ambient tem-
Accordingly, throttling is the common feature of the large gas lique-
perature, therefore pre-cooling processes are used for hydrogen lique-
faction cycles after creating non-ideality in the working fluid of the
faction. Finally, basic recuperative systems that can be used to mass
cycle, by increasing its pressure or decreasing its temperature or uti-
liquefy hydrogen are simple Claude, Kapitza, dual pressure Claude, pre-
lizing of both ways. In addition, considering the J-T effect, the

Fig. 7. Spin isomers of molecular hydrogen. (a)

para-hydrogen: anti-parallel nuclear spins with
lower energy state, and (b) ortho-hydrogen: par-
allel nuclear spins with higher energy state.

62
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

cooled Linde–Hampson, pre-cooled dual- pressure Linde–Hampson, pre- 5.2.1. Energy analysis of the simple ideal Claude cycle
cooled simple Claude, pre-cooled dual pressure Claude, helium-pre- For control volumes at steady state, mass balances in Fig. 10 give:
cooled Claude, and pre-cooled mixed refrigerant cycles [123]. On the
ṁ 2 = ṁ 1 = ṁ (1)
other hand, the basic cycles that cannot be used for hydrogen lique-
faction are Solvay, Simple Linde–Hampson, Dual-Pressure Linde–- ṁ 3 = ṁ e (2)
Hampson, and Heylandt cycles.
ṁ 10 = ṁ 1−ṁ f = ṁ −ṁ f (3)

5.1. Dewar system of hydrogen liquefaction where

A very subtle mist of hydrogen is produced by Wroblewski in l894 ṁ x = Mass rate of stream x (kg/s)
[27,128]. He employed Cailletet’s expansion technique using com-
pressed pre-cooled hydrogen in a capillary glass tube. He then expands Moreover, energy balance between the outlets of the compressor,
it from the pressure of 100 atm and the temperature of the liquid expander, and the separator gives:
oxygen to the ambient pressure. However, an unstable mist in the tube ṁ 10 h10 + ṁ 3 h3 + ṁ f hf = ṁ 2 h2 + ṁ e he (4)
was a slight dispersion of hydrogen. The experiments are repeated on a
larger scale and confirmed by Olszewski in 1895 [74]. As the results, substituting mass rates from Eqs. (1)(3) and considering that h10 = h1
the time for liquid hydrogen to remain in the tube was so high that he gives:
could determine the temperature with a special thermometer. However, (ṁ −ṁ f ) h1 + ṁ e h3 + ṁ f hf = mh
̇ 2 + ṁ e he (5)
the major obstacles to achieving liquid hydrogen were leakage of heat
and boil-off losses [109,114,129,130]. where
Sir James Dewar invented the Dewar flask in 1892, a vacuum in-
sulated vessel that was a milestone in gas liquefaction investigations h x = Mass enthalpy of stream x (kJ/kg)
[61]. Using a cascade process based on Joule-Thomson expansion and
heat exchange, he was the first to liquefy hydrogen for the first time in Let x be the mass fraction of the expander, i.e.:
1898 with 4 cc/min flow rate [79]. As shown in Fig. 5, gaseous hy- ṁ e
drogen is precooled by liquid air at 180 atm in this system to−250 °C. It x=
ṁ (6)
then expanded through a valve in a double-walled thermos vessel with
the silver-coated walls also cooled by liquid air. The fraction of liquefaction or liquid yield is then given as:

ṁ f h1−h2 h −h
y= = + x ⎜⎛ 3 e ⎞⎟
5.2. Simple cycles of hydrogen liquefaction ṁ h1−hf ⎝ h1−hf ⎠ (7)

Details of the equipment and methodology used by researchers for The net work required per unit mass of compressed gas can be ex-
gases liquefaction are described and illustrated using simple graphic pressed as:
shapes [68]. Accordingly, Fig. 9 illustrates the simple methodology of Ẇ
gases liquefaction before 1895. As can be seen, the three stages are w= = [T1 (s1−s2)−(h1−h2)]−x (h3−he )
ṁ (8)
discrete and sequential, and there is neither pre-cooling and nor ex-
ternal work of a working medium. The working medium is either hy- where
drogen itself, and saturated vapor of it is not returned as the coolant
after the internal working through throttling. w = Net energy consumption of the cycle (kJ/kg)
Since 1895, recuperative cooling is practically employed in the gas Ẇ = Net power consumption of the cycle (kW)
liquefaction techniques [131]. Accordingly, it means using the lique- sx = Mass entropy of stream x (kJ/kg K)
fying fluid itself as the coolant in the system. Fig. 10 depicts the con-
ceptual design of a simple Claude cycle, as a combination of the simple Specific energy consumption (SEC) of the cycle is defined as the
methodology of gas liquefaction (Fig. 9), recuperative cooling, and energy consumption divided by the physical production or the energy
expansion engine, which is the basis for all the large-scale hydrogen consumption per unit mass liquefies, as follows:
liquefaction plants in use, across the world, and the most other con- Ẇ Ẇ
ventional liquefaction cycles [126]. In addition, Fig. 11 shows Tem- SEC = wl = =
ṁ f ̇
my (9)
perature-Entropy or T-S diagram of the simple ideal Claude cycle with
isothermal, isobaric, isentropic, and isenthalpic processes. where

Fig. 9. Gases liquefaction process up to 1895.

63
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

Fig. 10. Conceptual design of the simple universal Claude cycle.

– Efficiency of the compressor and expander is 85%.

– Min temperature approach for the heat exchangers is 1.5 °C–2 °C
– no ortho-para conversion system
– no isothermal and isentropic process
– neglecting pressure drops
– mass fraction 0f 65% for expander
– ṁ = 1 kg/s

The following results are obtained from the software test run:

– y= 8.077e−002
– Ẇ = 6426.4 kW
– SEC = wl = 79564.2 kJ/kg LH2 = 22.1 kWh/kg LH2
– COP = 5.5%

Obviously, taking into account the other irreversibility and espe-

cially considering heat generation due to hydrogen ortho-para con-
version, the energy consumed per unit mass of liquefied hydrogen will
be greater.
Fig. 11. T-S diagram of the simple ideal Claude cycle [125].

5.2.2. Exergy analysis of the simple ideal Claude cycle

Thermodynamic second law analysis or exergy analysis is an ef-
wl = Net energy consumption of the cycle per unit mass of liquefied gas fective tool for analyzing and comparing of thermodynamic systems
(kJ/kg LH2) and has great advantages over the energy analysis, [137,138]. Ac-
cordingly, exergy analysis shows useful energy stream or work potential
The coefficient of performance is defined as the ratio of cooling destruction through the process. The exergy method of thermal plant
provided by the working fluid to the net input power of the cycle, as analysis is elaborated [139]. A comprehensive criterion is developed for
follows: large-scale hydrogen liquefaction systems comparison, which compen-
sates for non-uniformity in feed specification [140]. As the results, the
ṁ f × (hFeed−hf ) overall system exergy efficiency is defined as the weighted average of
COP =
Ẇ (10) pre-cooling and cryogenic-liquefying sections efficiencies based on
their respective contributions to the net power consumption. Therefore,
The process is simulated with Aspen HYSYS version 9 developed by considering Fig. 10, the exergy efficiency μex is defined as below:
AspenONE. It has an extensive databank and powerful methods for
calculation of physical properties. Isobaric heat capacities and equation (Eẋ )f −(Eẋ )Feed
μex =
of states of the model in the cryogenic region of interest are verified Ẇ (11)
with derived data from McCarty et al. [108]. Reference fluid thermo-
where
dynamic and transport properties database (REFPROP) Software, de-
veloped by national institute of standards and technology (NIST) can be
μex = Exergy efficiency of liquefaction process
used for data verification. Also, thermo-physical properties of hy-
(Eẋ )f = Product LH 2 exergy flow (kW)
drogen, found in REFPROP version 9.1 can be checked in [132,133].
(Eẋ )Feed = Feed GH 2 exergy flow (kW)
Peng-Robinson equation of state, the most conventional model in the
Aspen HYSYS, is adopted for simulation due to the largest applicability
From the viewpoint of exergy utilization, this rational efficiency
range in terms of temperature and pressure, largest binary interaction
[139] is a very useful definition as it compares the required ideal re-
parameter database, simplicity, and fast computation. However, Eco-
versible work with the actually consumed work in the process.
simPro [134], Pro/II [135,136], HYSIM, and ASPEN PLUS are some
The figure of merit (FOM) for a liquefaction process, however, is
other simulators that may be used for LHL process simulation.
defined as the ratio of the ideal work, −wi to the actual work −wl . This
Considering the following assumptions:
quantity is another expression of μex and considering Fig. 10, may be

64
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

expressed as: the compressor is subcritical and if the exit from expander becomes
−wi saturated vapor, then the third HX may be eliminated [125]. However,
FOM = although the Linde-Hampson liquefaction plants capital cost is lower
−wl (12)
than Claude one, the Claude operational cost is lower than Linde-
wi = T0 (sFeed−sf )−(hFeed−hf ) (13) Hampson one due to the lower power consumption rate in the Claude
cycles.
where Furthermore, The flow rate of expander can be optimized reaching
minimum work required [123]. Expander mass fraction of three basic
wi = Net ideal work for an ideal liquefaction process (kJ/kg LH2) hydrogen liquefaction cycles is optimized [145]. Accordingly, the op-
T0 = Temperature of dead state = 298.15 K timized percentage of expander mass fraction is 55–65% for the Claude
cycle. Effect of expander flow fraction on the exergetic efficiency of a
The following results are obtained from the software test run of helium liquefaction system, based on a modified Claude cycle, is ana-
simple Claude cycle through Aspen HYSYS commercial software by lyzed [146]. Results show that the sensitivity of the cycle performance
authors, considering mentioned assumptions: to variations in compressor pressure ratio increases when the ex-
panders’ flow rate exceeds their optimum values. Moreover, expander
– wi = 14426.4 kJ/kg LH2 = 4.0 kWh/kg LH2 optimum mass fraction is found to be 80% of the cycle total flow. The
– FOM = 18.1% performance of a multistage helium liquefier is evaluated using exergy
analysis [147]. In addition, compressor discharge pressure, heat ex-
5.2.3. The simple Claude cycle variations changer effective UA and expander flow rate are selected as in-
In 1934, Kapitza process that is a variation of simple Claude cycle is dependent variables. As the results, optimum expander mass fraction
described by Piotr Leonidovich Kapitza [141], the Russian cryogenics for a liquefier with 2–5 expanders is 80%. Finally, considering Eq. (7),
pioneer [142]. As the results, the first reciprocating expansion engine as the enthalpy is minimum after isothermal compression (i.e. at point
for making liquid helium is used instead of liquid hydrogen pre-cooling 2); the liquid yield is maximum.
bath of Onnes’s J-T process. Moreover, the Kapitza engine is worth
considering as the first to use a hydrodynamic gas lubricated piston to 5.3. Precooled systems of hydrogen liquefaction
minimize frictional heat generation at low temperature. In addition, the
process also can be used for the liquefaction of hydrogen; however, it Precooling effects on the cryogenic systems’ performance are ex-
seems that its practical advantage is for the large-scale liquid hydrogen amined [87,123,125,126,148]. Accordingly, pre-cooling is an in-
production. evitable necessity in liquefying gases whose inversion temperature is
The minimum work required in the simple Claude cycle is propor- below ambient. In addition, the liquid yield can be improved by low-
tional to the pressure ratio, not the pressure difference in compressor ering the enthalpy of the high-pressure gas at the warm end of the HX-1
[143]. Therefore, the dual pressure Claude process is based on this idea, in Fig. 10 (i.e. h2 in Eq. (7)). Moreover, high pressure required in the
as another variation of Claude cycle. Moreover, the compression energy ideal liquefaction process can be reduced by decreasing the tempera-
varies as the logarithm of the compression ratio [144]. Therefore, as the ture of the working fluid before throttling. As well as, Barron [123]
compression ratio of the simple Claude cycle is 200/1 and its logarithm implies that purification of gases in the heat exchanger can be accom-
is 5.30 and the compression ratio of the main compressor of the dual plished using precooling.
pressure Claude cycle is 200/5 and its logarithm is 1.39, the compres-
sion energy of the dual pressure Claude cycle is 26.2% times of simple 5.3.1. Liquid-nitrogen pre-cooled basic cycles
one. Authors found no literature publication about SEC of dual pressure As the maximum inversion temperature of hydrogen (i.e. 205 K) is
Claude cycle. below ambient [149], pre-cooling is an inevitable necessity of the basic
simple cycles modification. Linde-Hampson system that utilized for air
5.2.4. Special features of the simple ideal Claude cycle liquefying can be improved, combining a liquid nitrogen bath, to be
Minimum work required for ideal simple Claude cycle is examined utilized for hydrogen liquefaction [123]. Fig. 12 depicts the PFD of a
[125]. Accordingly, as the high-pressure increases, the minimum work LN2 pre-cooled Linde-Hampson cycle and Fig. 13 shows Temperature-
required for liquefaction decreases and the expander flow ratio at the Entropy or T-S diagram of the process with isothermal, isobaric, and
minimum work decreases too. Moreover, the simple Claude cycle ca- isenthalpic processes. Unlike the simple Linde-Hampson cycle, the
pital cost is more than simple Linde-Hampson due to the addition of an simple Claude cycle can be used to liquefy hydrogen without any pre-
expander and two additional heat exchangers, however, the exit from cooling [150]. However, the process performance can be improved by

Fig. 12. Pre-cooled Linde-Hampson PFD for hydrogen liquefaction.

65
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

calculated as 12.12 kWh/kgLH2 and 6.66 kWh/kgLH2 [151]. As the results,

an estimate of non-ideality effect of seven factors including of heat
2 Isothermal exchangers, ortho-para conversion, purification of hydrogen gas, com-
1
pression equipment, expansion equipment, nitrogen liquefaction pro-
3 cess, and insulation on SEC of the dual pressure Claude cycle is esti-
evaporation mated. Accordingly, the combined effect of non-idealities increases the
6 recycle flow by approximately 43 per cent. Therefore, the SEC of the
dual pressure Claude cycle considering of non-idealities effect is re-
ric ported as 12.26 kWh/kgLH2 . Liquefaction systems using air as the
ba

working fluid are compared [123]. It is showed that SEC of a liquid-

Iso

ric
nitrogen precooled dual pressure Linde system is 56% of the single

ba
pressure precooled Linde system and the FOM of the dual pressure
Iso
system is 1.73 times of single pressure. SEC of the precooled single
pressure and dual pressure ideal Claude cycles is calculated [150]. As
4 Isenthalpic the results, the energy consumption in the single pressure Claude cycle
is 1.12 times that of the dual pressure Claude cycle. Furthermore, dual
pressure Claude cycle process for air liquefaction is in its optimum
f 5 g performance if the intermediate pressure is adjusted between 50 and
60 atm and the high pressure in 200 atm when the expander mass
fraction is set to be 70% [123]. Finally, the performance of a dual
Fig. 13. T-S diagram of the liquid-nitrogen pre-cooled Linde-Hampson cycle [125].
pressure LN2 pre-cooled Claude cycle for several expander inlet tem-
perature is examined as a function of the intermediate pressure, while
precooling. the inlet pressure is adjusted at 100 atm for all states [152]. Accord-
With ideal conditions such as ideal compression, ideal heat ex- ingly, the minimum SEC all occur in the region of 30 atm of inter-
changers and conversion to equilibrium conditions the SEC of the pre- mediates pressure.
cooled ideal simple Line-Hampson cycle is calculated as
16.24 kWh/kgLH2 [151]. The Average energy consumption per unit mass
of liquefied hydrogen of the precooled Linde-Hampson cycle and pre- 5.3.3. Helium pre-cooled systems
cooled simple Claude cycle is calculated as respectively 68.1 kWh/kgLH2 , Hydrogen liquefaction cost is estimated and categorized to capital,
and 29.9 kWh/kgLH2 [145]; moreover, the medium FOM is determined energy supply, and operating and maintenance costs [153]. As the re-
as 4.75% and 11.1% too. The FOM of a liquid nitrogen precooled sults, the investment cost is more than 50% of the total cost. Therefore,
Claude cycle is computed to be 50–60% higher than the precooled the smaller the size of the compressor, as an expensive equipment, the
Linde-Hampson one [123]. The differences between reported values is lower the cost of liquefaction. The size of the compressor is directly
due to differences in supposed operating conditions. related to its outlet pressure [154], therefore, if the compressor output
pressure is reduced then the capital cost will decrease. Precooling in-
creasing by introducing auxiliary refrigeration systems in the lique-
5.3.2. Liquid-nitrogen pre-cooled basic cycles’ variations faction process in addition to liquid-nitrogen will reduce the required
The energy requirement for hydrogen liquefying can be reduced output pressure of the compressor.
combining of pre-cooling and dual pressure [123]. Considering ideal Helium-refrigerated hydrogen liquefaction system, as another the-
conditions, the SEC of the pre-cooled ideal dual pressure Linde- oretical liquefaction system with a more complicated configuration in
Hampson cycle and the pre-cooled ideal dual pressure Claude cycle is compare to pre-cooled Claude cycle, is analyzed [145]. According to

Fig. 14. Conceptual design of the liquid-nitrogen and Joule-Bryton pre-cooled Claude cycle.

66
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

the results, a helium refrigeration auxiliary system is combined with a bubble temperatures through an isobaric process constant, which also
pre-cooled Claude cycle. Helium is the working flow of the auxiliary called temperatures gliding. Therefore, it is provided a good match
system and itself pre-cooled in the liquid nitrogen bath. Furthermore, between the process and refrigerant temperature profiles; thus, the
the helium is not condensed in the cycle, however, cooled lower than minimum the temperature differences, the minimum exergy losses
the hydrogen temperature. Moreover, average SEC and FOM of the [161]. In other words, a more simple and reliable system in comparison
cycle is calculated as 39.4 kWh/kgLH2 and 8.7% respectively. The helium to pure refrigerant one is provided that has promising potential to be
system is not relatively sensitive to helium pressures and it is claimed more efficient. More investigations on cryo-cooling with mixed re-
that FOM is 11%, for a system with a helium pressure of 1 atm [123]. frigerants and MR properties may be found elsewhere [162–176].
In 1959, klimenko proposed natural gas liquefaction using a mixed
5.3.4. Joule-Brayton pre-cooled systems refrigerant process [177]. Fig. 15 depicts Klimenko refrigerator as a
Considering that, the helium system is used as a refrigerator rather variation of simple Claude cycle [163]. As can be seen, stream divider
than a liquefier, and the final expansion is through an expansion engine and expansion engine are substituted by a phase separator and throt-
not through a Joule-Thomson valve, and the working fluid gas is not tling valve in compare to Fig. 10. Obviously, a liquid separator may be
liquefied after final expansion and just is cooled lower than the hy- replaced the evaporator for gas liquefaction cycles. It is noteworthy that
drogen temperature, the system may be characterized as a pre-cooled more than 95% of the natural gas liquefaction plants around the world
Joule-Brayton cycle [88]. However, the working fluid of the Joule- use mixed refrigerants liquefaction systems, however, the most con-
Brayton auxiliary refrigeration system may not be helium in this cate- ventional cryogenic refrigerators and gas liquefaction systems operate
gory [155]. Fig. 14 illustrates a complete version of Fig. 10 in- with pure fluids [163].
corporating auxiliary refrigeration systems combined with a pre-cooled Fig. 16 shows the general configuration of mixed refrigerant pro-
simple Claude cycle. cesses combined with simple Claude cycle for hydrogen liquefaction
It is worth noting that in the so-called reverse Joule-Bryton pre- [143]. As illustrated, mixed refrigerant cycle acts as an auxiliary pre-
cooler auxiliary systems, a pure or mixed refrigerant is used. However, cooler section of the main liquefaction system. Moreover, MR is re-
the working fluid operates solely in the gas phase and passes through peatedly divided into liquid and vapor phases passing through phase
the expansion engine to provide cooling effect only. Conceptual JB pre- separators, is cooled, partially is condensed, is expanded in forwarding
cooling LHL systems will be reviewed in Section 6 of this paper. streams and finally warmed through return streams. Therefore, the
design methods of these processes are more complex and accurate
knowledge about MR thermodynamic properties is required to predict
5.3.5. Mixed refrigerant pre-cooled systems the behavior of them in expanders and heat exchangers. A major dis-
In 1936, Podbielniak employed an easily accessible mixture of gases advantage of mixed refrigerant systems is the problems caused by the
and vapors as working fluid in a refrigeration system [156]. According two-phase flow through the heat exchangers. Conceptual MR systems
to the author claims, even the component with the lowest boiling point combined with pre-cooling LHL systems will be reviewed soon in Sec-
in the gas mixture becomes liquid and avoiding high pressure, the ex- tion 6 of this paper.
pensive and bulky equipment required for high pressure are avoided.
However, the importance of using of mixture refrigerants is illustrated
and auto-cascade cycles with phase separation against the MR Linde- 5.3.6. Cascade systems
Hampson cycles with no phase separators introduced later by Fuderer Cascade systems are in fact an extension of the pre-cooling systems
[157], are founded [87]. In 1947, Ruhemann used a hydrocarbon where each independent refrigeration cycle pre-cools the other re-
mixture refrigerant for a vapor compression cycle [158]. As the results, frigeration cycle, or more than one auxiliary refrigeration cycle is em-
the advantage of phases separation is described. Accordingly, con- ployed to pre-cool the main gas stream [123]. Accordingly, as a cascade
sidering the mixed refrigerant separation into a liquid phase richer in system has the potential of getting close to ideal reversible liquefaction
the less volatile component, and a vapor phase richer in the more vo- process, it is an attractive option for gas liquefaction. In addition, cas-
latile component than the initial mixture, the lowest available tem- cading is introduced as the most efficient and reversible process that
perature, boiling point, is the lowest available pressure for the mixed can be done by refrigerant baths, usually liquid cryogens, or re-
refrigerant vapor phase, which generally will be lower than initial cuperative heat exchangers [178]. Furthermore, cascade cycles or gas
mixture’s one. Further of Ruhemann’s contribution may be found liquefying systems with multilevel cooling are good solutions when a
studying the separation of gases methodologies and hydrocarbon pro- wide temperature range is required to achieve improved efficiencies
cessing at low temperatures elsewhere [159]. Mixed refrigerant cycles [160]. As well as, cascading advantages are not limited to the liquefying
are analyzed and optimized [160]. As the results, the main advantages gases whose inversion temperature is below ambient but also extend to
of mixed refrigerants are investigated. Accordingly, mixed refrigerants other gases to increase liquid yield [126].
can experience a range of temperature changes including the dew and Historically, the early step by step refrigeration system used for

Fig. 15. Klimenko refrigerator as another variation of the simple Claude cycle.

67
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

Fig. 16. General configuration of the mixed refrigerant pre-cooling simple Claude cycle.

oxygen liquefaction is likely the first multilevel temperature cascade investigated [180]. As the results, the intermediate temperature should
cryogenic liquefaction system [179]. As the results, using piston com- be adjusted so that high, to the low-temperature ratio at each stage
pressors, vapor –compression systems continuously pre-cooled oxygen become same to minimize entropy generation in the system. A general
passing through cascading baths. The performance of a typical cascade mathematical thermodynamics model is developed to optimize a cas-
system for LNG production is analyzed [123]. According to the results, cade refrigeration system [181]. Accordingly, the early mentioned ratio
refrigeration capacity of each stage should be adjusted so that approach should be same at each stage to maximize the coefficient of perfor-
temperature in the evaporators and heat exchangers get a practical mance of the system. A mixed refrigerant auto-cascade system is stu-
minimum, to decrease thermodynamic irreversibility and energy con- died [182]. As the results, the relation between COP, cooling capacity,
sumption of the system. A multi-stage cryogenic refrigeration system is evaporation temperature and refrigerant pressures and temperatures is

Table 2
Technical comparison of the basic hydrogen liquefaction cycles.

Hydrogen liquefaction cycle ya (%) SEC (kWh/kg LH2) FOM (%) Remarks

(1) Wroblewski system Compressing feed hydrogen to 100 atm and pre-cooling to −183 °C and expanding (Fig. 9)
(2) Olszewski system Similar to Wroblewski, but on a large scale (Fig.9)
(3) Dewar system Compressing feed hydrogen to 100 atm and pre-cooling to −183 °C and expanding (Fig. 5). Capacity = 0.24 (l/h)
(4) The thermodynamically ideal liquefaction (Carnot cycle)
[123,150,151]
Feed: 21 bars, 25 °C, n-GH2
Output: 1 bar. −253 °C. ,n−LH2 – 3.27 100 SEC = 2.18 [98]
Output: 1 bar, −253 °C,99.8%p−LH2 – 3.92 100 SEC = 2.89 [98]
(5) Ideal simple Linde-Hampson Cannot liquefy hydrogen
(6) Pre-cooled ideal LH [150] 30 16.27 20 Fig. 12
Output: 1 bar. −253 °C. ,n−LH2 SEC=16.24 [151]b
(7) Pre-cooled ideal dual pressure LH [150] 41 12.14 27 SEC = 12.12 [151]b
Output: 1 bar. −253 °C. ,n−LH2
(8) Pre-cooled Linde-Hampson [145] 12–17 72.8–79.8 4.5–5.0 Fig.12
Output: 1 bar. −253 °C. ,n−LH2
(9) Simple Claude [current paper] 8 22.1 18.1 Fig. 10
Output: 1 bar. −253 °C. ,n−LH2 Capacity = 4.46 (l/hr)
(10) Pre-cooled ideal single pressure Claude [150] 52 10 32.6 –
Output: 1 bar. −253 °C. ,n−LH2
(11) Pre-cooled ideal dual pressure Claude [150] 59 8.68 37.7 SEC = 6.66 [151]b
Output: 1 bar. −253 °C. ,n−LH2
(12) Pre-cooled Claude [145] 16–20 28–39.2 9.2–13 –
Output: 1 bar. −253 °C. ,n−LH2
(13) Pre-cooled dual pressure Claude [151] – 12.26 – –
Output: 1 bar, −253 °C. ,99.8%p−LH2
(14) Helium-refrigrated system [145] 100(n−H2) 33.6–56 6.5–11 Fig.16
Output: 1 bar. −253 °C. ,n−LH2 54(p−H2)

a
Liquid yield ( y= ṁ f /ṁ ) .
b
Output: 1 bar, −253 °C,99.8%p−LH2 .

68
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

depicted for different conditions. A cascade natural gas liquefaction 6. Industrial hydrogen liquefier
system is exergy analyzed. As the results, SEC depends only on the
properties of the incoming and outgoing natural gas and is calculated as Table 3 shows worldwide hydrogen liquefaction commercial plants
456.8 kJ/kgLNG . Moreover, COP and exergy efficiency of the system are [98,150,184,185]. The first hydrogen liquefier in the US is built in 1904
calculated as 1.8 and 38.5% respectively. A new concept called the using Dewar hydrogen liquefier, which is exhibited at the 1904 St.
cascade Rankine cycle to cold energy recovery for power generation is Louis world fair, and bought by US national bureau of standards (NBS)
proposed [183]. Results show, the more number of stages the more net [97]. Considering US atomic energy commission decision to build a
power output and efficiencies. Moreover, the three-stage system with hydrogen liquefaction plant in 1949, a new hydrogen liquefier is de-
propane as the working fluid has highest net power output and effi- signed and built at the NBS facility in Boulder in 1952 [186]. Results
ciencies. Hundreds of investigation about cascade systems performance reveal that the production rate of the plant was 320 l/h of normal liquid
analysis may be reviewed, however, pre-cooling concept has been used hydrogen. An ortho-para convertor is added to the plant to make it
in all pre-cooling cycles discussed above, and related conceptual LHL capable of producing 240 l/h of 90–95% para liquid hydrogen in 1953
cycles will be reviewed in Section 6 of this paper. Various technical [117]. A large-scale hydrogen liquefaction plant is constructed under
parameters of the basic hydrogen liquefaction cycles are listed in direction of the US air force in Ohio in 1957 [187]. Results show that
Table 2. the system was similar to the NBS one. A third hydrogen liquefier plant
is put into full service in 1958 [188]. Results imply that the plant is
built at the US defense department program similar to NBS liquefier and
the production rate is still classified. In addition, the plant hydrogen

Table 3
Hydrogen liquefaction plants of the world in the order of history [98,150,184,185].

Country Location Owner Size (TPDa) Year established In service

USA Colorado NBS 0.5 1952 –

USA Ohio APCIb 1 1956 –
USA Painsville Air Products 3 1957 No
USA West Palm Beach Air Products 3.2 1957 No
USA Florida APCI 3.5 1957 –
USA California Stearns-Roger Mfg. Co. 1.5 1957 –
USA Florida APCI 30 1958 –
USA West Palm Beach Air Products 27 1959 No
USA Mississippi Air Products 32.7 1960 No
USA California Stearns-Roger Mfg. Co. 7 1960 –
USA Ontario, CA Praxair 20 1962 Yes
USA California Stearns-Roger Mfg. Co. 26 1962 –
USA California APCI 32.5 1963 –
USA Sacramento Union Carbide 54 1964 No
USA New Orleans, LA APCI 34 1977 Yes
USA New Orleans, LA APCI 34 1978 Yes
Japan Amagasaki Iwatani 1.2 1978 No
USA Niagara Falls, NY Praxair 18 1981 Yes
Canada Sarnia Ontario, APCI 30 1982 Yes
Japan Tashiro MHI 0.6 1984 No
Japan Akita Prefecture Tashiro 0.7 1985 Yes
USA Sacramento, CA APCI 6 1986 Yes
Japan Tane-Ga-Shima Japan Liquid Hydrogen 1.4 1986 Yes
Japan Oita Pacific Hydrogen 1.9 1986
Canada Monterial Air Liquide 10 1986 Yes
Holland Rosenburg APCI 5 1987 Yes
France Waziers, Lille Air Liquide 10 1987 Yes
Japan Minamitane Japan Liquid Hydrogen 2.2 1987 Yes
Canada Becancour Quebec Air Liquide 12 1988 Yes
USA Niagara Falls, NY Praxair 18 1989 Yes
Canada Magog, Quebec BOC 15 1989 Yes
Guyanac Kouru F. Air Liquide 5 1990 Yes
Canada Monterial BOC 14 1990 Yes
Germany Ingolstadt Linde 4.4 1991 Yes
India Mahendragiri ISRO 0.3 1992 Yes
USA Pace, FL APCI 30 1994 Yes
USA McIntosh, AL Praxair 24 1995 Yes
China Beijing CALT 0.6 1995 Yes
USA East Chicago, IN Praxair 30 1997 Yes
Japan Kimitsu Air Products 0.3 2003 Yes
India Saggonda Andhra Sugars 1.2 2004 Yes
Japan Osaka Iwatani 11.3 2006 Yes
Germany Leuna Linde 5 2008 Yes
Japan Tokyo Iwatani 10 2008 Yes
India India Asiatic Oxygen 1.2 – Yes
USA California Stearns-Roger Mfg. Co. 62.5 – –
USA New Jersey Air reduction Sales Co. 6 – –
USA Ashtabula, OH Praxair – – No

a
Tons per day.
b
Air Products and Chemicals, Inc.
c
French Guyana.

69
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

feed is supplied from refinery waste gas. Flynn implies that there is [201]. Accordingly, the last one has the best SEC of 8.49 kWh/kgLH2 and
more correlation between hydrogen liquefaction development projects helium J-B cycle has the best exergy efficiency of 45.6%. In 2002, a MR
and US defense and aerospace activities than oxygen or helium lique- precooling plant is proposed with helium-neon refrigerant in cryogenic
faction [189]. Since 1957, several large hydrogen plants is built in the section and ethane-propane refrigerant in precooling area [202]. Ac-
United States, however, no new plant construction has been reported cordingly, the SEC of the plant is reported as 6.98 ≅ 7.0 kWh/kgLH2 and
since 1997, and existing liquid hydrogen capacity is still greater than the exergy efficiency of the plant is predicted to be 60.7%. In addition,
demand [184]. Hydrogen liquefaction plants capacities is reported the mixture of MR in cryogenic section comprises helium and neon,
since 1958 to 1998 through investigating of commercial uses of liquid which has better heat transfer properties and compression features than
hydrogen [190]. However, no corresponding figure is reported for pure refrigerant, depend on the percentage of each component in the
power consumption. Four main category of Japan hydrogen energy mixture. Helium improves heat transfer properties, while neon im-
research and development programs are briefly reported [191]. Ac- proves compression features. However, Krasae-in et al. [98] indicated
cordingly, the scientific acceleration of the development of hydrogen that high-efficiency of the plant cannot be determined because the
energy systems is estimated to be slow but steady. In a more recent configuration of the proposed propane refrigeration system is im-
research, the main results achieved during the WE-NET project in Japan possible for low power consumption. In 2004, a conceptual study of a
is reported [192]. As the results, four processes including hydrogen, helium refrigeration cycle, precooled with LN2 is carried out, con-
helium, MR, and neon JB liquid-nitrogen precooled Claude cycle are sidering four versions of hydrogen liquefier [203]. Results show that
examined, considering special condition about production capacity and the minimum SEC of the plant, which is an evolved form of proposed
system efficiency. Furthermore, of the four processes considered, hy- plant of Belyakov [204] is 12.7 kWh/kgLH2 . In addition, it is showed
drogen Claude cycle is evaluated to be suitable, and the SEC of hy- through a sensitivity analysis that using of a helium refrigeration cycle
drogen Claude cycle is calculated as 7.9 kWh/kgLH2 . However, the se- instead of a hydrogen refrigeration one, the thermodynamic efficiency
lected cycle is still a conceptual design. The establishment of three will be improved by 6% and explosion and fire safety is reduced to the
liquid hydrogen production plants with capacities of 100, 400 and minimum. In 2006, a large-scale H2 liquefaction plant integrated with a
750 dm3/h in China in 1970 is mentioned [193]. Accordingly, the LNG precooling plant is analyzed [205], and Kramer imply that in-
construction of these plants was through a national development plan tegration of hydrogen liquefaction with an LNG regasification terminal
for a new rocket vehicle and no other details is reported. Typical SEC of will reduces the energy expenditure considerably from 10 kWh/kgLH2 to
Praxair plants in the US is calculated between 12.5 and 15 kWh/kgLH2 4 kWh/kgLH2 [206]. However, LNG is required and pre-cooled feed
[148]. SEC for Linde large-scale plant in Ingolstadt Germany, is re- hydrogen gas and the plant should be located near a seaport. In 2006, a
ported as 13.58 [194,195], 13 [196] and 15 kWh/kgLH2 [197]. More- conceptual plant based on MR cycle process for precooling hydrogen
over, the Ingolstadt plant is a reference for new conceptual plants SEC from 25 °C down to near −193 °C is conducted [207]. According to the
assessment. There is no literature of SEC of the other worldwide large results, the SEC of the Ingolstadt plant in Germany is simulated to be
existing hydrogen plants, e.g. Air Products, Air Liquide and new Linde 11.54 kWh/kgLH2 , a mixed refrigerant pre-cooling cycle is used in the
large-scale plant system in Leuna. Table 4 summarizes information of precooling section as an alternative to liquid nitrogen, and a liquid
the industrial hydrogen liquefiers that their SEC has been reported in helium bath is used in the cryogenic section. Moreover, the preliminary
published literature. Overall, as can be seen from published researches, design of the proposed plant showed that an exergy efficiency of 60%
the best reported SEC of the in service LHL plants is in the range of and SEC of 7.0 kWh/kgLH2 is feasible. Again in 2006, a process based on
10–15 kWh/kgLH2 , however, achieving of energy consumption of cooling, liquefying, and sub-cooling of hydrogen using helium re-
7.5–9 kWh/kgLH2 is predicted [198]. Finally, hydrogen liquefiers of frigerant cycle is introduced [110]. According to the results, a helium
moderate size from 10 to 500 l/h that usually used for laboratory re- refrigerated cycle is used for precooling and cryocooling of feed hy-
search are concerned [199]. Accordingly, five laboratory liquefiers of drogen, which one of its heat exchangers is located in LH2 tank. Other
NBS cryogenic engineering laboratory at Boulder, Colorado, the Ber- H2 liquefaction process is provided that precooling is carried out using a
keley liquefier (1932), the Pennsylvania state university liquefier helium refrigeration cycle [208]. Accordingly, special characteristic of
(1953), the Clarendon laboratory in the University of Oxford (1953), the process is removal of compression work from hydrogen com-
and the liquefier at royal radar establishment (RRE), Malvern (1952), pressors, and transfer it to helium compressors, that has a significant
are reviewed. Furthermore, the capacity of them is reported as 230, 60, impact on energy consumption (8.7 kWh/kgLH2 ). In 2008, A helium
50, 18, and 16 l/h . refrigerated precooling LHL system is proposed and analyzed [209].
According to the results, exergy efficiency is calculated as 39% for the
7. The proposed theoretical systems for hydrogen liquefaction base configuration of the proposed plant that may be improved to 44%
with increases in heat exchangers area. In addition, it is calculated as
Many types of research are investigated to obtain the optimum SEC 18–25% for pilot plant and there is no report of SEC of proposed plant.
and exergy efficiency. In 1961, a hydrogen liquefaction system is ana- Again in 2008, the SEC of four helium Joule–Brayton cascade cycle by
lyzed as a variation of pre-cooled dual pressure Claude cycle [151]. Valenti is reported as 5.76 kWh/kgLH2 [210]. Results show that the in-
Accordingly, SEC in the ideal state is calculated as 6.66 kWh/kgLH2 and tegrated helium refrigerant cycle carries out both the hydrogen pre-
considering non-idealities effect it is determined to be 12.26 kWh/kgLH2 cooling and cryogenic refrigeration, however, the system is not guar-
[151]. Table 5 demonstrate differences of proposed systems for LHL anteed to have a high-efficiency [98]. The SEC improvement trend of
plants by technical parameters. In 1978, a modified Claude cycle with a
LN2 bath used for precooling is proposed [200]. Results show that the Table 4
core of the system is a dual pressure Claude cycle with normal hydrogen Summary of the published SECs of the industrial hydrogen liquefiers.
feed at 1 atm and 308 K , and LH2 product at 929 kPa and 20.57 K .
Moreover, a sensitivity analysis is done and SEC of the system is re- Hydrogen liquefaction system kWh ⎞ ηx a (%)
SEC ⎛⎜ ⎟

ported in the range of 9.15–11.68 kWh/kgLH2 . In addition, LH2 para ⎝ kgLH2 ⎠

content, compressor efficiency and recycle return pressure are in- (1) Praxair plant [148] 19.3–23.1
12.5–15 kWh/kgLH2
troduced as the major effective parameters. Furthermore, the most ex- (2) Ingolstadt Plant [194,195] 13.58 kWh/kgLH2 21.2
ergy losses are occurred in LN2 recycle compressor and N2 refrigerator (3) Ingolstadt Plant [196] 13 kWh/kgLH2 22.2
as 29.35% and 25.02% respectively. In 1997, four cycles of hydrogen J- (4) Ingolstadt Plant [197] 15 kWh/ kgLH2 19.3
B pre-cooling Claude, helium J-B, basic neon J-B and neon J-B with cold
a
pump are analyzed and compared to meet WE-NET project objectives Exergy efficiency.

70
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

Table 5
Technical comparison of the proposed theoretical systems for hydrogen liquefaction.

Hydrogen liquefaction system ṁ f a (TPD) yb (%) kWh ⎞ ηx c (%) Config. according to the Fig. 21
SEC ⎛ ⎜ ⎟
Feed: P (kPa)/T (K) /condition ⎝ kgLH2 ⎠
Output: P (kPa)/T (K) /condition
Remarks

(1) Baker and Shaner [200] (1987) 250 100 10.85 36.0 LN2 pre-cooled dual pressure Claude
929/308/25%p−LH2 H2 expander isentropic eff. = 0.79, comp. average weighted isentropic eff. = 0.79, liquefier
100/20.57/97%p−LH2 max P = 4137 kPa

(2) Matsuda and Nagamei (WE- J-B pre-cooling Claude

NET) [201] (1997)
5000/300/25%p−LH2 Expanders isentropic efficiency = 0.85 (power recovery efficiency = 0.90), comp. isentropic
100/20.4/>95%p−LH2 eff. = 0.80, liquefier max P ≈ 5000 kPa.
(2-1) Hydrogen J-B pre-cooling 300 100 8.60 46.4 According to the trade-off evaluation of processes:
Claude (a) refrigerant and heat exchangers cost of hydrogen case is minimum and its compressors and
(2-2) Helium J-B pre-cooling 300 100 8.76 45.6 safety cost are maximum.
Claude (b) heat exchangers cost of helium case is minimum.
(2-3) Basic neon J-B pre-cooling 300 100 8.65 46.2 (c) refrigerant, maintenance and plant cost of neon with cold pump case is maximum and its
Claude operation, management of refrigerant and expanders cost is minimum
(2-4) Neon with cold pump J-B 300 100 8.49 47.1
pre-cooling Claude

(3) Quack [202] (2002) J-B pre-cooling Claude with mixed refrigerant
8000/300/25%p−LH2 MR comprises 20% neon and 80% helium, expanders isentropic eff. = 0.85–0.90, comp.
100/20.15/– isentropic eff. = 0.85, liquefier max P ≈ 8000 kPa, heat exchangers minimum temperature
(3-1) Without pressure drop in 173 100 ≈ 7.0 60.7 approach = 2.12 K, increasing the circumferential speed of turbocompressors is reported as
calculation the most urgent field of development
(3-2)With pressure drop in 173 100 7.39 53.8
calculation

(4) Kuzmenko et al. [203] (2004) 54 100 12.7 34.6 Helium pre-cooling Claude intregrated with a N2 J-B cycle
1600/300/25%p−LH2 Hydrogen compressor isothermal eff. = 0.6, helium compressor isothermal eff. = 0.53,
150/19.9/>98%p−LH2 nitrogen compressor isothermal eff. = 0.53, N2 cycle expanders isothermal eff. = 0.73–0.75,
helium cycle expanders isothermal eff. = 0.85, max pressure of N2 , helium, and H2 cycles is
respectively 28, 25, and 50 bar

(5) Kuendig et al. [205] (2006) 50 100 4 [206] – LNG and LN2 bath pre-cooling Claude. In fact, LNG bath pre-cooling integrated with a N2 J-B
cycle
2200/300/25%p−LH2 Commercial machinery is introduced that may be used to realize this concept. LNG is required
200/22.8/99.5%p−LH2 and pre-cooled feed hydrogen gas and the plant should be located near a seaport

(6) Stang et al. [207] (2006) – 100 7.0 60 MR and helium bath pre-cooling
2000/300/25%p−LH2 Helium-neon mixture is used as the mixed refrigerant, The onset of the inevitable
00/20.3/– accompaniment of the mixed refrigerant cycles and helium refrigerating cycles

(7) Notardonato [110] (2006) – 100 – – Helium pre-cooling Claude

2400/300/25%p−LH2 Compressors and expansion turbines isentropic eff. = 0.80, It is shown that the cycle has the
–/–/– maximum efficiency when the hydrogen and helium pressure are set at 2400 kPa and
1200 kPa respectively

(8) Shimko and Gardiner [208] 50 100 8.73 44.6 Helium pre-cooling Claude
(2008)
2100/300/25%p−LH2 Expansion turbines (exclude wet expander) isentropic eff. = 0.83–0.86, wet expander
100/20.0/95%p−LH2 isentropic eff. = 0.90, He and H2 comp. eff. = 0.8, helium cycle max P ≈ 1500 kPa, hydrogen
cycle max P ≈ 2100 kPa, remove of compression load from hydrogen compressors and transfer
to helium compressors is the main feature of the cycle

(9) Staats et al. [209] (2008) 50 100 – 39 Helium pre-cooling Claude

2100/300/25%p−LH2 Expansion turbines (exclude wet expander) isentropic eff. = 0.83–0.86, wet expander
100/20.0/95%p−LH2 isentropic eff. = 0.90, He compressor isothermal eff. = 0.80, H2 compressor isothermal
eff. = 0.60, hydrogen cycle max P ≈ 2100 kPa

(10) Valenti [210] (2008) 864 100 5.04 47.7 J-B pre-cooling Claude with helium as the refrigerant
6000/300/25%p−LH2 Polytrophic efficiency of helium expanders = 0.88–0.93, helium compressors polytrophic
150/20/– eff. = 0.92, hydrogen expander polytrophic eff. = 0.85, J-B cycle max P = 4000 kPa, heat
exchangers minimum temperature approach = 2.30–3.15 K, it is reported that helium
compressors and heat exchangers have the most exergy loss

(11) Krasae-in et al. [211] (2010) 100 100 5.35 54.0 MR and J-B pre-cooling Claude/
2100/298.15/25%p−LH2 Mixed refrigerant with ten components, compressors and expansion turbines isentropic
130/19.45/95%p−LH2 eff. = 0.80, heat exchangers minimum temperature approach = 1–2 °C, J-B cycles max
P = 4000 kPa, MR cycle max P = 1800 kPa

(12) Krasae-in [136] (2014) 100 100 5.91 48.9 MR and J-B pre-cooling Claude
2100/298.15/25%p−LH2 Five-component mixture consists of 4% hydrogen, 18% nitrogen, 24% methane, 28% ethane,
130/19.45/95%p−LH2 and 26% butane (by mole), J-B cycles max P = 4000 kPa, MR cycle max P = 1800 kPa, it is
concluded that pressure drops in the heat exchangers do not have a significant impact on the
overall plant total power consumption

(13) Yuksel et al [214] (2017) 50 [209] 100 – 57.13 Helium pre-cooling Claude
2100/298.15/25%p−LH2 [209] Helium and H2 compressors isentropic eff. = 0.80, helium expanders eff. = 0.85, it is
202/20.55/99%p−LH2 concluded that the largest exergy destruction rates occur in helium turbo expanders
(continued on next page)

71
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

Table 5 (continued)

Hydrogen liquefaction system ṁ f a (TPD) yb (%) kWh ⎞ ηx c (%) Config. according to the Fig. 21
SEC ⎛ ⎜ ⎟
Feed: P (kPa)/T (K) /condition ⎝ kgLH2 ⎠
Output: P (kPa)/T (K) /condition
Remarks

(14) Sadaghiani and Mehrpooya ≈ 300 100 4.36 55.47 MR and J-B pre-cooling Claude
[215]
2100/298.15/25%p−LH2 Mixed refrigerant with nine components for the pre-cooling cycle, a mixed refrigerant with
130/20.15/95%p−LH2 three components for the cryogenic J-B cycle, adiabatic eff. of compressors and expanders is
supposed to be 0.90 and 0.85 respectively, heat exchangers minimum temperature
approach = 1.4–2.2 °C. J-B cycles max P = 1000 kPa. MR cycle max P = 1600 kPa.

(15) Asadnia and Mehrpooya [216] ≈ 100 100 7.69 39.5 MR and J-B pre-cooling Claude
2100/298.15/25%p−LH2 Mixed refrigerant with 11 components for the pre-cooling cycle, a mixed refrigerant with two
130/20.95/95%p−LH2 components for cryogenic J-B cycles, isentropic eff. of compressors and expanders is supposed
to be 0.8, heat exchangers minimum temperature approach = 1–2 °C. J-B cycles max
P = 3000 kPa, MR cycle max P = 1805 kPa, it is reported that heat exchangers and
compressors, the most expensive equipment, including more than half of the exergy losses of
58% of the total

a
Capacity (Tones per day).
b
Liquid yield ( y= ṁ f /ṁ ) .
c
Exergy efficiency.

the hydrogen liquefier systems is recorded from 1898 to 2009 in- heat capacity of hydrogen above the critical point and to make it easier
troducing a literature review [98]. Following this work, a new large- achieving a better temperature fit in the heat exchangers. Feed flow rate
scale MR system is introduced in 2010, which MR system precools of 1.157 kg/s is equivalent to 100 tons per day (TPD) to provide re-
hydrogen and a four H2 Joule-Brayton cascade refrigeration system li- quired LH2 for a large urban area with 100,000–200,000 hydrogen
quefies hydrogen [211]. Result reveals that the overall power con- vehicles [206]. The efficiency of the compressors is the worst case and
sumption of the plant is 5.35 kWh/kgLH2 , however, internal simulation of expanders is the state-of-the-art of turbomachines.
tests run in Aspen HYSYS commercial software by the authors specified
that the system are not guaranteed to have the reported specific energy 8. Hybrid plants for liquid hydrogen production
consumption. In 2010, a laboratory MR hydrogen liquefaction system is
exergy analyzed [212] and a modified version of it is simulated in 2011 Multigenerational systems having improved efficiency are a poten-
[213]. Results reveals that a MR refrigeration system and helium bath is tial route for the green supply energy chain and researchers have also
used in original test rig and a MR precooling system in the modified started studying of using various renewable energies in hydrogen pro-
version. In 2014, another research work is offered [136]. Accordingly, duction and liquefaction systems [218]. Geothermal energy is used for a
the novel configuration of optimized H2 Joule- Brayton cycles of the proposed combined refrigeration system including of a conventional
system carries out the transition of the equilibrium ortho–para hy- vapor-compression cycle cascaded with an absorption refrigeration
drogen gas from temperature −193 °C to the temperature of −253 °C. cycle and is thermodynamically analyzed [219]. Results show that the
Moreover, the overall power consumption of the cycle is calculated as conventional cycle energy consumption is approximately 37% and 54%
5.91 kWh/kgLH2 . In 2017, a super critical LHL system that is a modified more than the combined system under the same operating conditions.
version from reference [209] is proposed [214]. Accordingly, a helium Low enthalpy geothermal heat is used for an ammonia-water ARS
refrigeration system is employed and there is no report about SEC and [220]. Accordingly, cooling capacity and actual COP are reported as
exergy efficiency here. Again in 2017, the minimum SEC among large- 10.5 kW and 0.433. Moreover, it is stated that actual COP is 50–70% of
scale L H2 production conceptual plants is reported as 4.41 kWh/kgLH2 calculated values. Using of geothermal energy is proposed for hydrogen
from Sadaghiani et al. [215]. They utilized a novel and perfect MR in liquefaction through investigation of three scenarios [221]. As the re-
accordance with plant configuration, and redesigned heat exchangers sults, using of output work of geothermal systems as the input for a
and expanders at a wide range of temperature. Moreover, a MR re- liquefaction cycle, output heat of geothermal system in an absorption
frigeration system is used in precooling section and a mixed refrigerant refrigeration system (ARS) to precool feed gas of liquefaction cycle, or
J-B system is used in cryogenic area. They reported the COP and exergy both scenarios improve liquefaction cycle performance. Four cases are
efficiency of the whole liquefaction process of their proposed process as analyzed for the use of geothermal energy for hydrogen production
0.1797 and 55.47%, respectively. A novel mixed refrigerant LHL system
is proposed by authors [216]. Accordingly, two combined auxiliary MR Table 6
precooling systems are used for precooling of hydrogen feed before Typical data for a large-scale conceptual plant for the liquefaction of hydrogen [216].
transient temperature and six MR Joule-Brayton cycles are used in the
Technical parameter Feed hydrogen Produced LH2
cryogenic section of the proposed system. Moreover, the SEC, COP and
Exergy of the plant are calculated as 7.69 kWh/kgLH2 , 17.1% and 39.5% Pressure (bar) 21 1.3
respectively. The novelties of the system are new proposed MR and Temperature (K) ≈ 300 ≈ 21
combined configuration of the system. Finally, a proposed hydrogen Mass flow rate (kg/s) 1.157 1.157
liquefaction system is exergy and exero-economic analyzed [217]. Ac- Percentage p−LH2 (%) 25 95
Lowest pressure (bar) 1.3
cordingly, the total cost rate of the system will be reduced using effi- Highest pressure (bar) 21
cient equipment as the best option. Table 6 listed the key technical H2 Expander exit temperature (K) 21
parameters of a typical LHL conceptual plant from Ref. [216]. The ir- Temperature approach of heat exchangers (K) 1–2
reversible losses take into account for the listed pressures and tem- Isentropic efficiency of the expanders (%) 80
Isentropic efficiency of compressors (%) 80
peratures that are correspond to the minimum SEC of the plant. The
Work of liquefaction (kWh/kgLH2) 7.69
input pressure of 21 bar has a definite minimum; as according to the
Exergy efficiency (%) 39.5
Walnum [161], it should be as high as possible to reduce the peak in the

72
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

[222]. As the results show, hydrogen production amount by one kg of

geothermal water at 200 °C in the reversible condition is in the range of
1.22–1.91 gr, and exergy efficiency of the system is in the range of
16.1–37.2%. Hydrogen liquefaction and production by geothermal en-
ergy is regarded from an economic view [223]. In the aforementioned
hybrid system, hydrogen is liquefied using a Linde-Hampson liquefac-
tion cycle, integrated with an absorption cooling system, which pre-
cools hydrogen before the liquefaction process. Moreover, a geothermal
based power plant supplies required power, and a high-temperature
steam electrolysis process is utilized for hydrogen production. Fur-
thermore, the results reveal that as the geothermal water temperature
increases the cost of the hydrogen liquefaction and production de-
crease. In addition, the cost of hydrogen production and liquefaction is
estimated as 0.979−2.615 "$"/kg LH2 . A triple effect absorption cooling
system integrated with geothermal, solar photo-voltaic/thermal (PV/
T), and Linde-Hampson cycle is investigated [224]. Results of the hy-
brid system investigation show simultaneous changes of liquefied hy-
drogen amount with the mass flow rate of geothermal hot water and
inlet air to PV/T. A pre-cooled Linde –Hampson cycle is combined with
an integrated geothermal-quadruple effect absorption system for hy- Fig. 18. Overview of conventional gas liquefaction methodology.
drogen liquefaction [225]. Accordingly, as the mass flow rate of geo-
thermal water increase, the energetic and exergetic coefficient of per-
analyzed [228]. Accordingly, an absorption system replaces vapor
formances of the system increase. Another integrated geothermal based
compression refrigeration cycle that its required heat duty is supplied
triple effect ARS and a Linde-Hampson cycle for hydrogen gas is pro-
from waste heat of the plant. Moreover, SEC of the plant is reported as
posed [226]. As the findings, as the geothermal source temperature
0.172 kWh/kgLNG . Another hybrid system using nuclear energy for hy-
increase, the mass of hydrogen gas precooled by the absorption system
drogen production and liquefaction is proposed [229]. As the results,
per unit mass flow rate of geothermal water and the amount of hy-
energy and exergy efficiencies of the plant are reported as 18.6%, and
drogen liquefied per unit mass flow rate of geothermal water increase.
31.35%.
A LN2 pre-cooled Claude cycle is combined with an integrated geo-
Low-temperature solar Rankine cycles for power generation have
thermal-ammonia-water ARS for the precooling of feed hydrogen
the potential for utilizing heat from low-temperature sources and fa-
[227]. According to the results, absorption cycle precools feed hy-
vorable characteristics for integration into energy systems [230].
drogen to −26.9 °C at a rate of 29.53 kg/s . Moreover, COP and exergy
Fig. 17 shows a typical PFD of an integrated hybrid system that consists
efficiency of the overall system is calculated as 0.162 and 67.9% re-
of a low-temperature auto-cascade solar Rankine cycle (ALSRC) system
spectively. A hybrid system for LNG production is introduced and
for power generation [230], an ammonia-water ARS for the pre-cooling

Fig. 17. Typical PFD of an integrated hybrid system. ARS/ALSRC/SC combined systems [227,230].

73
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

Fig. 19. The cooling methodology of the liquefying

gas. Interchanging improvement occurs by (a) pre-
cooling with a cryogenic liquid bath or auxiliary
refrigeration systems, (b) parallel isentropic ex-
pansion, and (c) serial isentropic expansion.

of hydrogen [227], and the simple Claude cycle for hydrogen lique- 10. Economics of the hydrogen liquefaction
faction. Geothermal energy supplies heat of generator of the absorption
system, and ALSRC supplies power of the first stage of Claude cycle Three hydrogen liquefaction systems including of LN2 precooled L-H
compressor. The ALSRC system includes two solar collectors, two ex- cycle (A), LN2 precooled simple Claude cycle (B), and conceptual LHL
panders, a regenerator, and an internal heat exchanger. cycle (C) are analyzed economically [239]. In this work electrical or
energy expenditures (EEX), capital expenditures (CAPEX), and opera-
9. H2 production/liquefaction and renewable energy sources tional and maintenance expenditures (OMEX) are considered. As the
result, OMEX is the smallest between the others. However, EEX has
Hybrid plants review in Section 8 reveals that renewable energy strong dependency to liquefier efficiency and less dependent on the
sources may be used in different combined configuration within LHL production rate and OMEX primarily depend on liquefier efficiency and
plants and play a promising role in the cost reduction. Integrated utility production rate. CAPEX primarily depends on the system type and flow
and biogas supply networks for hydrogen production are optimized rate of produced LH2 . In addition, compressors and heat exchangers are
[231]. As a result, the optimized integrated system decreases the total introduced as the vital equipment in hydrogen liquefaction plants.
costs by about 13% compared to the existing situation. A large-scale However, compressors, heat exchangers, and intercoolers installation
wind hydrogen plant integrated with an oil industry is assessed [231]. cost are the highest for A, B, and C cases respectively. Finally, the cost
The optimal plant yields uncompetitive hydrogen production cost of of liquefaction is evaluated "$"0.63/kg LH2 for C case at production flow
"$"8.43/kg H2 relative to conventional methods of hydrogen production rate of 720 TPD, considering $0.04/kWh for the cost of consumed
from fossil fuels. Hydrogen production systems from biomass and nat- electricity. In another research, CAPEX, EEX, and OMEX are reported as
ural gas (NG) are evaluated using industrial emerging evaluation in- 63%, 29%, and 8% of total cost of liquefaction respectively [153]. It is
dices [231]. Results show that efficiency index of hydrogen production reaffirmed that energy costs are strong functions of the liquefaction
from NG is higher than hydrogen production from biomass but sus- process efficiency and less dependent on the production rate [240]. In
tainability index of hydrogen production from biomass is higher. Using addition, EEX and non-EEX are comparable in small plants, however,
of some bacteria for hydrogen production from glycerol is reviewed EEX become more important at LHL plants. Moreover, following
[231]. New technologies, optimal conditions, and some other features equation on CAPEX is derived:
are examined. One-step liquid hydrogen production from methane and
CAPEX(€ 2000/(kg/h)) = 828,313 × (production rate )−0.48 (1)
landfill gas is analyzed [232]. In this study hydrogen feed is directly
supplied for liquefaction process, therefore, there is no CO2 emissions A hybrid system for liquid hydrogen production is investigated for
and H2 losses within the integrated process. However, there are seven models of geothermal energy utilization [223]. The results show
countless studies in the case of using renewable energy sources the that cost of hydrogen production and liquefaction decrease with geo-
production and liquefaction of hydrogen, which could be the subject of thermal water temperature. Moreover, it is evaluated that cost of li-
a review article [222,223,233–237]. However, according to interna- quefaction ranges between "$"0.979/kg LH2 and "$"2.615/kg LH2 depending
tional energy agency [238], in the short term, electrolysis of water and on the model.
small scale natural gas reforming are suitable methods for hydrogen It is investigated that large-scale hydrogen liquefaction is econom-
production, and in the medium to long term, hydrogen production is ically viable [22]. According to the results, total costs of a LHL plant
justified based on the use of fossil fuels. Moreover, other hydrogen with a production rate of 100 TPD can be reduced to nearly one-third of
production methods are far from commercialization and require more a 50 TPD liquefier. Moreover, a large-scale liquefaction process is op-
research and development. In addition, hydrogen production from timized [241,242]. According to the results, liquefaction costs for a 25
biomass requires more focus on feed preparation and transportation, TPD liquefier are about half the liquefaction costs of a 5 TPD.
and it is likely to be economic in large scale. Photo electrolysis is a Production and distribution of hydrogen as an automotive transport
recent developed step and cost of material and practical issues must be fuel is discussed [243,244]. Accordingly, “clean hydrogen” term is de-
solved. Photo biological processes are very novel and low efficiency. fined as centrally produced hydrogen, which its CO2 emissions is cap-
Finally, in high temperature processes, material development need to tured and separated. An interesting finding of these researches is that
be focused more, for example, high temperature membranes and heat well to wheel costs of clean gaseous and liquid hydrogen are similar.
exchangers. The reason for this can be seen in the remarkable savings caused by

74
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

Fig. 20. The overall configuration of LHL systems emphasizing the status of renewable energy technologies in hydrogen liquefaction context.

distribution of liquid hydrogen, which compensates liquefaction costs. with electricity due to low energy conversion efficiency in the hydrogen
Costs of liquid hydrogen distribution are significantly lower than costs economy. The suspicion about the hydrogen economy is criticized
of gaseous hydrogen due to the high energy density of liquid hydrogen. [196]. In [246,247] instead of considering a comprehensive approach
It is possible to deliver more liquid fuel with a trailer (3500 kg) com- including complete energy chains and their variations, some individual
pared to delivery of gaseous fuel with the same trailer (350 kg). It is processes have been picked. They are technical and resource plausi-
showed that using the clean hydrogen, well-to-wheel CO2 emissions is bilities, efficiencies, greenhouse gas and other emissions as well as
brought down from 10 to 15 kg CO2 per kg H2 to about 2.5 kg CO2 per costs. Finally, nowadays there is no consensus on whether or not hy-
kg H2 . Technical and economic aspects of replacing fossil fuels used in drogen economy works. In addition to providing evidence for the
the air and vehicular transport with liquid hydrogen are reviewed and transition to a hydrogen economy in the prospect, there are some ideas
discussed [245]. According to the conclusions, the use of LH2 as a long about its low efficiency [248–253].
term replacement for fossil fueles for land and air transportation is
technically feasible and energy cost of LH2 is "$"0.174/kg LH2 .
It is claimed that hydrogen economy (versus electricity economy), 11. Hydrogen liquefaction systems review analysis
as an industrial system is not viable [246,247]. Accordingly energy
consumption for different hydrogen delivery paths are calculated. The results of this extensive literature survey, in which the histor-
Technical calculations reveal that the electrical energy input exceeds ical, technical and economic aspects of the large-scale hydrogen li-
higher heating value (HHV) of delivered hydrogen to end user. It is quefaction processes were reviewed, can be divided into cognitive and
concluded that the hydrogen as an energy carrier cannot be competed applied divisions.

75
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

Fig. 21. A novel Classification of hydrogen lique-

Hydrogen liquefaction systems in the papers faction systems in terms of heat exchange and ex-
pansion process method.

Without With With

recuperator recuperator regenerator

Gases liquefaction process up to 1895 Outside of the study scope of this article

Other
Throttling Expansion
principles

Outside of the study scope of this article

Pre-cooled Linde-Hampson Simple Claude

Pre-cooled dual pressure L-H Dual pressure Clude

Kapitza

Liquid bath pre-cooling

JB pre-cooling

Helium pre-cooling

MR pre-cooling

Hybrid systems

Table 7
Cited publications of the hydrogen liquefaction systems’ configuration showed in Fig. 21.

Item Identified configurations used in hydrogen Published literature that investigates the configuration Related figure of configuration in this Number of cited
liquefaction cycles paper

A Without recuperator [74,79,128] Figs. 5 and 9 3

B Pre-cooled Linde-Hampson [123,145,151] Figs. 2, 3, 4. and 12 3
C Pre-cooled dual pressure L-H [151] Figs. 2, 3, 4. and 12 1
D Simple Claude (single pressure Claude) [current paper], [123] Fig.10 2
E Dual pressure Claude – Fig.10 –
F Kapitza [142] Fig.15 1
G Liquid bath pre-cooling [145,151,194,200,205,148,203,212] Figs.12 and 14 8
H JB pre-cooling [201,202,205,210,211,136,215–217] Fig.14 9
I Helium pre-cooling [145,201,203,204] [207,110,208,209,214] Fig. 16 9
J MR pre-cooling [202,207,211–213,136,215–217] Fig.16 9
K Hybrid systems [221,223–228] Fig.17 7

11.1. Cognitive results from LHL process review (1-2) Employing of auxiliary systems including helium, Joule-
Bryton, and MR systems.
Fig. 18 is deduced from the historical analysis of conventional gas (2) Work against external forces at the expense of internal energy or
liquefaction cycles, considering the main used methodologies. As can be isentropic expansion (Figure b and c)
seen, after purification that is a main but ignored section, the obligatory (3) Work against internal forces by J-T expansion or isenthalpic ex-
methodology of gas liquefaction is creating of non-ideality for the gas, pansion (Figure d)
as ideal gases temperature do not change during isenthalpic processes
[254]. The next important stage is related to the gases whose inversion Moreover, from another view, interchanging is the main core pro-
temperature is below ambient and the bypass route is for others. Before cess that can be enhanced using three ways of (a) pre-cooling, (b)
1895, all of the methods of liquefying gases pass through the bypass parallel isentropic expansion, and (c) serial isentropic expansion, plus
route. Finally, expansion is a mandatory stage of all of the conventional to isenthalpic expansion.
gas liquefaction methods. According to the methodologies classification shown in
Fig. 19 depicts conventional methodologies of cooling of the gas Figs. 18 and 19, and LHL systems historical review, it can be said that
that is being liquefied in Fig. 18. As can be seen, there are three main the basic classical recuperative cycles for hydrogen liquefaction had
procedures to cool the liquefying gas: four main pillars after gas purifying [161,202,207]. Pre-compression
and pre-cooling sections to make non-ideality for feed gas and getting
(1) Precooling (Figure a) below the inversion temperature, cryogenic cooling section to improve
(1-1) Using a cryogen liquid bath (generally liquid nitrogen bath). liquid yield after expanding, and final expanding-liquefaction section.

76
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

Fig. 22. The frequency of published literature in

each category listed in Table 7.

K- Hybrid systems
J- MR pre-cooling
I- Helium pre-cooling
H- JB pre-cooling
G- Liquid bath pre-cooling
F- kapitza
E- Dual pressure Claude
D- Simple Claude (single pressure Claude)
C- Pre-cooled dual pressure L-H
B- pre-cooled Linde-Hampson
A- Without recuperator

Fig. 23. Published literature of hydrogen liquefac-

The investigating area tion systems through the history since 1961.
11- Hybrid systems
10- MR pre-cooling
9- Helium pre-cooling
8- JB pre-cooling
7- Liquid bath pre-cooling
6- Kapitza
5- Dual pressure Claude
4- Simple Claude
3- Pre-cooled dual pressure L-H
2- Pre-cooled Linde-Hampson
1- Without recuperator

As a novel finding of this paper, considering of LHL systems historical the main different configuration used in the investigated papers.
review, authors infer that renewable energy technologies section should Table 7 presents the published literature that directly investigated a
be added to this old classification as another independent section. hydrogen liquefaction system versus configurations shown in Fig. 21
Fig. 20 illustrates a typical BFD of the overall configuration of LHL and related figures in this paper. Moreover, Fig. 22 shows a pie chart of
systems. As can be seen, the status of renewable energy technologies is the frequency of published literature in each category shown in Table 7.
identified as the supplier of feed gas pre-pressurizing and pre-cooling. It can be seen that the use of auxiliary refrigeration systems, i.e. Joule-
As well as, they may provide auxiliary power for the liquefaction cycle, Brayton, helium and MR pre-cooling systems, has surpassed the use of
and cooling for output streams of the system’s compressors. The main cryogenic liquid baths of pre-cooling with a total frequency of 51%.
compressors provide supplementary pressurizing, and MR, JB, and he- Furthermore, hybrid systems have allocated 14% of liquidation systems
lium cascade systems provide more required pre-cooling. It is supposed in a short time, and therefore they are a growing and important sector,
that ortho-para conversion is done using catalysts embedded in heat considering their optimistic environmental approach. However, using
exchangers. of cryogen liquid bath is of interest to researchers with the frequency of
Fig. 21 illustrates another novel classification of hydrogen lique- 15%.
faction systems in terms of heat exchange and expansion process Fig. 23 depicts the published literature of hydrogen liquefaction
method, considering published literature that directly discusses hy- systems through the history since 1961. As can be seen, the in-
drogen liquefaction systems’ configuration. As can be seen, review of vestigating area includes using MR systems for the pre-cooling section
regenerative methods of refrigeration such as Stirling, Gifford- and Joule-Brayton systems for the cryogenic section of LHL plants. It is
McMahon, pulse tube and so on, are out of the scope of this research supposed that the helium refrigerating systems are a modification of JB
and can be studied in numerous other references [87,255]. As well as, systems using helium as the refrigerant. Moreover, hybrid systems will
special methodologies of refrigeration such as thermoelectric and contribute more in this area in near future. In addition, Table 8 lists all
magnetic refrigeration are not studied here. Therefore, it can be seen of the SECs reported in the published literature, by the cited order in
that hydrogen liquefaction methodologies are in fact categorized into this paper.
two main contexts: Claude liquefaction cycles with the characteristic of The trend of SEC of hydrogen liquefaction plants is illustrated in
using an expander for the expansion process and Linde-Hampson li- Fig. 24. As can be seen, energy consumption of the proposed cycles has
quefaction cycles, which use expansion valves instead of the expanders. no sharp change in recent years. Accordingly, since 2014 the medium
Sub categories of these two main contexts are determined considering SEC of four proposed plants ([136,211,215,and216]) is 5.84 kWh/kgLH2 .

77
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

Table 8 11.2. Applied results for LHL process improvement

Calculated SEC of hydrogen liquefaction cycles in the published literature as outlined in
this paper.
Renewable energy based cooling, heating, and power cycles can be
Item Configuration of the hydrogen liquefaction cycle SEC (kWh/kg LH2)
combined with LHL plants, coupling points of the two systems should
be investigated. Therefore, some area of improvement are as follows:
1 Simple Claude (Current paper) 22.1
2 Pre-cooled ideal Linde-Hampson cycle [151] 16.24 – A high-pressure electrolysis process or a steam reforming plant may
3 Pre-cooled Linde-Hampson cycle [145] 68.1 be utilized to raise feed gas input pressure.
4 Pre-cooled single pressure Claude cycle [145] 29.9
5 Pre-cooled ideal dual pressure Linde-Hampson cycle 12.12
– A solar ARS system may be used to reduce the temperature of the
[151] hydrogen feed gas or compressors output stream.
6 Pre-cooled ideal dual pressure Claude cycle [151] 6.66 – Geothermal CHP power plants can be combined with a LHL process.
7 Pre-cooled dual pressure Claude cycle [151] 12.26 Its output power is used as the input for the LHL process and its
8 Helium-refrigerated hydrogen liquefaction system 44.8
output heat in an ARS combined with the LHL process, or an op-
[145]
9 WE-NET project in Japan [192] 7.9 timum combination of them is utilized.
10 Praxair plants’ medium SEC in the US [148]. 13.75 – Auto-cascade solar Rankine cycles power may be used as the input
11 Linde large-scale plant in Ingolstadt Germany 13.58 for the LHL plant.
[194,195]
12 Linde large-scale plant in Ingolstadt Germany [196] 13
Compressors, expanders, and heat exchangers are the vital equip-
13 Linde large-scale plant in Ingolstadt Germany [197] 15
14 Baker system’s medium SEC [200] 10.42 ment of a LHL plant. Exergy losses in LHL plants may be reduced by
15 Matsuda (WE-NET project) [201] 8.5 improving their performance. As well as, EEX have a strong dependency
16 Large-scale conceptual plant by Quack [202] 7.0 on turbomachines efficiency. Applied practices for EEX and exergy
17 Kuzmenko helium refrigeration cycle, pre-cooled with 12.7
losses reduction may be defined as follows:
LN2 [203]
18 Stang MR cycle [207] 7.0
19 Shimko helium refrigeration cycle [208] 8.7 – Design new high-efficiency compressors and expanders that could
20 Four helium Joule–Brayton cascade cycle by Valenti 5.76 be combined with ARS’ for cooling the compressors’ body.
[210] – Utilizing of expanders instead of Joule-Thompson valves.
21 Large-scale MR system by Krasae-in in 2010 [211] 5.35
– Improving the effectiveness of aluminum plate-fin heat exchangers
22 Large-scale MR system by Krasae-in in 2014 [136] 5.91
23 Large-scale L H2 production conceptual plants by 4.41 to reduce the exergy losses. More examination is necessary to find
Sadaghiani [215] appropriate size of MR heat exchangares.
24 Mixed refrigerant LHL combined system by Asadnia 7.69 – Improving insulation material for reducing the manufacturing cost
[216]
and increasing the efficiency.
25 The thermodynamically ideal liquefaction system 2.89
[98]
26 The thermodynamically ideal liquefaction system 3.92 Layout and scale of LHL plants affect CAPEX and OMEX. Some of
[123,150,and151] considerations that should be regarded include:

– Building the LHL plants on a larger scale or multiple plants of the

The trend can be divided into four areas between four ideal cycles, same design.
indexed by square shapes on the diagram: – Constructing LHL plants near cold recovery facilities for hydrogen
feed gas pre-cooling.
– Area 1: area of basic classical cycles that have been developed long – Optimum cascading of the systems may increase production effi-
time ago. ciency.
– Area 2: area of in-service today large plants around the world. – Integration of hydrogen energy supply chain by the methods such as
– Area 3: area of the proposed plants with the SEC more than co-locating the LHL plants with hydrogen and power production
6.66 kWh/kgLH2 , which is the SEC of precooled ideal dual pressure plants.
Claude cycle [151].
– Area 4: area of the proposed plants with the SEC less than Exergy efficiency of industrial LHL plants is in the range of 20–30%.
6.66 kWh/kgLH2 . Therefore, there is a considerable improvement potential. On the other
hand, exergy efficiency of the proposed LHL plants is 40–50%. Some
The medium SEC of the industrial plants in area 2 is area of improvement in this field may be expressed as follows:
13.83 kWh/kgLH2 . It is completely logical considering the SEC of the
borderline cycle of precooled ideal dual pressure Linde-Hampson of – Using the optimal design methodologies in the new proposed plants
10.95 kWh/kgLH2 . In addition, equipment in a real plant have not ideal and providing novel productive configuration for LHL processes.
performance. The proposed plants in area 3 try to approach the second – Developing novel mixed refrigerants.
ideal borders and average SEC of them is 10.95 kWh/kgLH2 , which has – Developing a new and accurate equation of state for predicting the
meaningful distance with the SEC of the next ideal border line cycle of hydrogen behavior in the range of −200 °C to −250 °C.
6.66 kWh/kgLH2 . However, area 4 has proposed cycles that have special – Utilizing more productive compressors with more stages and inter-
energy consumptions between two ideal states. They are SEC of the coolers aiming to get isothermal compression.
precooled ideal dual pressure Claude cycle of 6.66 kWh/kgLH2 and SEC – Using heavier mixed refrigerants
of the thermodynamically ideal liquefaction system of 2.89 kWh/kgLH2 – Exploiting of non-conventional contexts of refrigeration for hy-
as reported in Ref. [98] and of 3.92 kWh/kgLH2 as reported in Refs. drogen liquefaction such as magnetic or acoustic liquefaction.
[123,150,151]. Therefore, it is deduced that considering the existing
technologies, SEC reduction for hydrogen liquefaction will not be However, it sounds that changing the configuration of the cycles to
abrupt in the near future. reduce energy consumption is not a promising solution. New in-
vestigations usually concentrate on improvement potentials of the
supply chain of hydrogen energy as a whole to optimize exergy effi-
ciency and energy consumption of the processes.

78
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

Fig. 24. The SEC trend of hydrogen liquefaction cycles versus history and methodologies.

12. Summary and conclusion sharp reduction in near future if no technology replaces early conven-
tional methods of hydrogen liquefaction. The SEC of good fabricated
A historical review of hydrogen liquefaction systems is done since industrial plants will be in the range of 13- 15 kWh/kgLH2 , and about
thermometer invention time up to now. After the first hydrogen li- proposed large-scale plants in the range of 5–8 kWh/kg LH2 . It is sound
quefaction in 1898 by Sir James Dewar, a dozen of the methodology of that changing of the configuration of the cycles to reduce energy con-
liquid hydrogen production is reviewed and they are classified in terms sumption is not a promising solution, considering the proximity of the
of heat exchange and expansion process method. Exploiting of isen- energy consumption of proposed cycles to ideal cycles. The main po-
tropic expansion processes instead of isenthalpic one, cascading of re- tential areas of improvement of cycles’ performance include of in-
frigerating cycles, using of newly mixed refrigerant as working fluid of tegration of LHL plants in hydrogen energy supply chain, development
refrigeration cycles and hybridization of renewable energy power cy- of more efficient equipment, hybridization to renewable technologies,
cles to refrigeration cycles are the main four growing approaches in the and using of new methodologies of refrigeration production such as
hydrogen liquefaction context. SEC of proposed plant will not have a magnetic or acoustic liquefaction.

79
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

Acknowledgement (1835). Isis 1952;43(2):109–13.

[36] Natterer J. Darstellung der Kohlensäure und des oxydirten Stickgases,
Stickstoffoxyduls, im starren Zustande. Adv Synth Catal 1844;31(1):375–6.
The first author would like to show his gratitude to Mr. Alireza [37] Brenni P. Johann A. Natterer (1821-1900) and His Pumps for Liquifying Gases.
Kadivar, former executive director of Jonoob-E-Shargh zone of IOPTC [38] Faraday M. On the liquefaction and solidification of bodies generally existing as
for sharing his pearls of wisdom with him during the course of this gases. Philos Trans R Soc London 1845;135.
[39] Tilden WA. Sir William Ramsay: memorials of his life and work. Macmillan and
research. In addition, he thanks Dr. Rabiei, head of the Kar higher Company limited; 1918.
education institute for his support of this research. However, they may [40] Moore RB. Helium: Its history, properties, and commercial development. J
not agree with all of the interpretations of this paper and any errors are Franklin Inst 1921;191(2):145–97.
[41] Kamerlingh Onnes H. On the Cryogenic Laboratory at Leiden and on the produc-
authors’ own and should not tarnish the reputations of these esteemed tion of very low temperatutes. Communications 1894;14.
persons. [42] Scott RB. Cryogenic engineering. Van Nostrand; 1959.
[43] Kurti N. Low temperature terminology. Cryogenics (Guildf) 1970;10(3):183–5.
[44] Scurlock RG. A matter of degrees: a brief history of cryogenics. Cryogenics (Guildf)
References
1990;30(6):483–500.
[45] Andrews T. Bakerian lecture: on the continuity of the gaseous and liquid states of
[1] Dell RM, Bridger NJ. Hydrogen—the ultimate fuel. Appl Energy 1975;1(4):279–92. matter. Proc R Soc London 1869;18(114–122):42–5.
[2] Hoffman P. Hydrogen—the optimum chemical fuel. Appl Energy [46] Andrews T. The Bakerian lecture: on the gaseous state of matter. Philos Trans R
1994;47(2):183–99. Soc London 1876;166:421–49.
[3] Uyar TS, Beşikci D. Integration of hydrogen energy systems into renewable energy [47] Wisniak J. Thomas Andrews. Rev CENIC Ciencias Químicas 2008;39(2).
systems for the better design of 100% renewable energy communities. Int J [48] Rowlinson JS. The work of Thomas Andrews and James Thomson on the lique-
Hydrogen Energy 2017;42(4):2453–6. faction of gases. Notes Rec R Soc 2003;57(2):143–59.
[4] Won W, Kwon H, Han J-H, Kim J. Design and operation of renewable energy [49] V. Regnault, Relation des expériences entreprises par ordre de monsieur le ministre
sources based hydrogen supply system: Technology integration and optimization. des travaux publics, et sur la proposition de la Commission Centrale des Machine a
Renew Energy 2017;103:226–38. Vapeur, pour determiner les principales lois et les données numériques qui entrent
[5] Wang X, Yang J, Chen L, He J. Application of liquid hydrogen with SMES for dans le calcul des, vol. 3. Firmin Didot, 1870.
efficient use of renewable energy in the energy internet. Energies 2017;10(2):185. [50] Joule JP, Thomson W. LXXVI. On the thermal effects experienced by air in rushing
[6] Antonelli M, Barsali S, Desideri U, Giglioli R, Paganucci F, Pasini G. Liquid air through small apertures. Philos Mag Ser 4 1852;4(28):481–92.
energy storage: potential and challenges of hybrid power plants. Appl Energy [51] Siemens CW. Improvements in refrigerating and producing ice, and in apparatus or
2017;194:522–9. machinery for that purpose. Br Pat 1857;2064:29.
[7] Amrollahi MH, Bathaee SMT. Techno-economic optimization of hybrid photo- [52] Weinstock H. A review of cryogenic history and principles of refrigeration. Appl
voltaic/wind generation together with energy storage system in a stand-alone Cryog Technol 1968;1.
micro-grid subjected to demand response. Appl Energy 2017;202:66–77. [53] Papanelopoulou F. Louis Paul Cailletet: The liquefaction of oxygen and the
[8] Aneke M, Wang M. Energy storage technologies and real life applications–a state of emergence of low-temperature research. Notes Rec R Soc 2013.
the art review. Appl Energy 2016;179:350–77. [54] Pictet R. On the liquefaction of hydrogen; 1878.
[9] Al-Hallaj S, Wilke S, Schweitzer B. Energy storage systems for smart grid appli- [55] Holme I. Sir William Henry Perkin: a review of his life, work and legacy. Color
cations. In: Water, energy & food sustainability in the middle east. Springer; 2017. Technol 2006;122(5):235–51.
p. 161–92. [56] Wisniak J. James Dewar-More than a flask. Indian J Chem Technol
[10] Hemmati R, Saboori H. Emergence of hybrid energy storage systems in renewable 2003;10(4):424–34.
energy and transport applications–A review. Renew Sustain Energy Rev [57] Keesom WH. Prof. Dr. H. Kamerlingh Onnes. His life-work, the founding of the
2016;65:11–23. Cryogenic Laboratory. Commun Phys Lab Leiden 1926(57):3–21.
[11] Eriksson ELV, Gray EM. Optimization and integration of hybrid renewable energy [58] Lorentz HA. Heike Kamerlingh Onnes (1853–1926). In: Collected Papers. Springer;
hydrogen fuel cell energy systems–A critical review. Appl Energy 1939. p. 407–410.
2017;202:348–64. [59] Oesper RE. Heike Kamerlingh Onnes. J Chem Educ 1944;21(6):263.
[12] Panwar NL, Kaushik SC, Kothari S. Role of renewable energy sources in environ- [60] Wróblewski SU, Olszewski SK. Liquid state oxygen and nitrogen. Ann Phys
mental protection: A review. Renew Sustain Energy Rev 2011;15(3):1513–24. 1883;20:256–60.
[13] Chevalier J, Geoffron P. The new energy crisis: climate, economics, and geopo- [61] Scurlock RG. 100 years after Dewar’s dewar: cryogenics today. In: Proceedings of
litics. Springer; 2016. the Royal Institution of Great Britain, vol. 65; 1994. p. 145.
[14] Demirbas A. Political, economic and environmental impacts of biofuels: a review. [62] Muir MM. The liquefaction of gases. Nature 1895;51:388–9.
Appl Energy 2009;86:S108–17. [63] Scurlock RG. History and origins of cryogenics. Clarendon Press; 1992.
[15] Gurz M, Baltacioglu E, Hames Y, Kaya K. The meeting of hydrogen and auto- [64] Claude G. Liquid Air, Oxygen, Nitrogen. P. Blakiston’s Son & Company; 1913.
motive: A review. Int J Hydrogen Energy 2017:1–13. [65] Thevenot R. A history of refrigeration throughout the world. International Institute
[16] Stroman RO, Schuette MW, Swider-Lyons K, Rodgers JA, Edwards DJ. Liquid hy- of Refrigeration; 1979.
drogen fuel system design and demonstration in a small long endurance air ve- [66] von Linde C. Process and apparatus for liquefying gases or gaseous mixtures, and
hicle. Int J Hydrogen Energy 2014;39(21):11279–90. for producing cold, more particularly applicable for separating oxygen from at-
[17] Corbo P, Migliardini F, Veneri O. Hydrogen fuel cells for road vehicles; 2011. mospheric air,” Ger. Patent# GB189512528, 1896.
[18] Yang C, Ogden J. Determining the lowest-cost hydrogen delivery mode. Int J [67] von Linde C. Aus meinem Leben und von meiner Arbeit: Aufzeichnungen für meine
Hydrogen Energy 2007;32(2):268–86. Kinder und meine Mitarbeiter; als Manuskript gedruckt. Oldenbourg; 1916.
[19] Sharma S, Ghoshal SK. Hydrogen the future transportation fuel: From production [68] Sloane T. Liquid air and the liquefaction of gases. London: The Norman W. Henley;
to applications. Renew Sustain Energy Rev 2015;43:1151–8. 1920.
[20] El-Osta W, Zeghlam J. Hydrogen as a fuel for the transportation sector: possibilities [69] Claude G. Sur la liquéfaction de l’air détenté avec travail extérieur récupérable’.
and views for future applications in Libya. Appl Energy 2000;65(1):165–71. CR Acad Sci Fr 1902;130:500–2.
[21] Denton WH, Nicholls CM. Technology and uses of liquid hydrogen. Pergamon [70] Linde H. Aus der Geschichte der Tieftemperaturverfahrenstechnik. KI Luft-und
Press; 1964. Kältetechnik 1995;31(9):419–24.
[22] Cardella U, Decker L, Klein H. Economically viable large-scale hydrogen lique- [71] Cailletet LP. Sur la condensation des gaz réputés incoercibles. CR, vol. 85; 1877. p.
faction. In: IOP conference series: materials science and engineering, vol. 171(1); 1270–1271.
2017. p. 12013. [72] Meunier S. Académie des Sciences—Séance de 31 décembre. Liquéfaction de
[23] Berstad D, Stang J, Nekså P. A future energy chain based on liquefied hydrogen; l’oxygene’,” La Nat., no. 240; 1878. p. 95–96.
2009. p. 1–23. [73] Wisniak J. Raoul-Pierre Pictet—The liquefaction of oxygen and achievement of
[24] Taylor FS. The origin of the thermometer. Ann Sci 1942;5(2):129–56. low temperatures; 2003.
[25] Mendelssohn KAG. The quest for absolute zero; 1966. [74] Travers MW. The liquefaction of hydrogen. Proc Phys Soc London 1899;17(1):561.
[26] Wisniak MJ. Guillaume amontons. Rev CENIC Ciencias Químicas [75] Kobe KA, Lynn RE. The critcal properties of elements and compounds. Chem Rev
2005;36(3):187–95. 1953;52(1):117–236.
[27] Almqvist E. History of industrial gases. Springer Science & Business Media; 2003. [76] Rafalska-Łasocha A. Karol Olszewski’s 100th anniversary of death and his con-
[28] Forgan S. Science and the sons of genius: Studies on Humphry Davy. Science tribution to cryogenics (in Polish). Pr Kom Hist Nauk PAU 2015;14(1):335–44.
Reviews Limited; 1980. [77] Gavroglu K, Goudaroulis Y. Heike Kamerlingh Onnes’ researches at Leiden and
[29] Davy H. The Collected Works of Sir Humphry Davy (etc.), vol. 1. Smith, elder; their methodological implications. Stud Hist Philos Sci Part A 1988;19(2):243–74.
1839. [78] Maxwell JC. The scientific papers of James Clerk Maxwell. University Press; 1890.
[30] Thompson SP. Michael Faraday: his life and work. The Macmillan; 1898. [79] Dewar J. Liquid hydrogen. Science (80-.) 1898;8(183):3–6.
[31] Faraday M. “Faraday,” in a history of chemistry, Springer; 1964. p. 99–141. [80] Plank R. Geschichte der Kälteerzeugung und Kälteanwendung. Entwicklung
[32] Faraday M, Davy H. On fluid chlorine. Philos Trans R Soc London 1823;113:160–5. Wirtschaftliche Bedeut. Werkstoffe; 1954. p. 1–160.
[33] Dull CE, Brooks WO, Metcalfe HC. Modern Physics. Henry Holt. [81] van Delft D. The liquefaction of helium. Europhys News 2008;39(6):23–5.
[34] Topham S, Bazzanella A, Schiebahn S, Luhr S, Zhao L, Otto A, Stolten D. Carbon [82] Reif-Acherman S. Several motivations, improved procedures, and different con-
dioxide. Ullmann’s Encycl. Ind. Chem; 1986. texts: The first liquefactions of helium around the world. Int J Refrig
[35] Roller DHD, Thilorier M. Thilorier and the first solidification of a‘ permanent’ gas 2009;32(5):738–62.

80
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

[83] Chakravarthy VS, Shah RK, Venkatarathnam G. A Review of Refrigeration Methods [124] Windmeier C, Barron RF. Cryogenic technology. Ullmann’s Encycl Ind Chem 2000.
in the Temperature Range 4–300 K. J Therm Sci Eng Appl 2011;3(2):20801. [125] Mukhopadhyay M. Fundamentals of cryogenic engineering. PHI Learning Pvt.
[84] Richardson RN, Evans BE. A review of pulse tube refrigeration. Int J Refrig Ltd.; 2010.
1997;20(5):367–73. [126] Walker G. Cryocoolers. Part 1: Fundamentals. New York: Springer Science &
[85] Walker G. Miniature Stirling cryocoolers: trends in development. Cryogenics Business Media; 1983.
(Guildf) 1986;26(7):387–91. [127] Walker G. Cryocoolers: Part 2: applications. Springer Science & Business Media;
[86] Walker G, Weiss M, Reader G, Fauvel OR. Small-scale liquefaction of hydrogen. Int 2012.
J Hydrogen Energy 1992;17(8):593–7. [128] Kubbinga H. A tribute to Wróblewski and Olszewski. Europhys News
[87] Maytal B-Z, Pfotenhauer JM. Miniature Joule-Thomson cryocooling: principles and 2010;41(4):21–4.
practice. Springer Science & Business Media; 2012. [129] Zhou L. Progress and problems in hydrogen storage methods. Renew Sustain
[88] Walker G. “Claude and Joule-Brayton Systems”, in Cryocoolers. Springer; 1983. p. Energy Rev 2005;9(4):395–408.
297–353. [130] Felderhoff M, Weidenthaler C, von Helmolt R, Eberle U. Hydrogen storage: the
[89] Kostionk VV. A text book of cryogenics. Discovery Publishing House; 2003. remaining scientific and technological challenges. PCCP 2007;9(21):2643–53.
[90] Walker G, Bingham ER. Low-capacity cryogenic refrigeration. Oxford University [131] Davies M. William Hampson (1854–1926): A Note. Br J Hist Sci
Press; 1994. 1989;22(1):63–73.
[91] Radebaugh R. Historical summary of cryogenic activity prior to 1950. Cryogenic [132] Mccarty RD, Weber LA. Thermophysical properties of parahydrogen from the
Engineering. Springer; 2007. p. 3–27. freezing liquid line to 5000 R for pressures to 10000 psia; 1972.
[92] Castle WF. Fifty-years’ development of cryogenic liquefaction processes. Cryogenic [133] Leachman JW, Jacobsen RT, Penoncello SG, Lemmon EW. Fundamental equations
engineering. Springer; 2007. p. 146–60. of state for parahydrogen, normal hydrogen, and orthohydrogen. J Phys Chem Ref
[93] de Bruyn Ouboter R. Cryogenics at the end of the 19th and the first half of the 20th Data 2009;38(3):721–48.
century (1880–1940). J Phys: Condens Matter 2009;21(16):164221. [134] Bradu B, Blanco E, Gayet P, Avezuela R, Cobas P, Veleiro A. CRYOLIB. A com-
[94] Foerg W. History of cryogenics: the epoch of the pioneers from the beginning to the mercial library for modelling and simulation of cryogenic processes with
year 1911. Int J Refrig 2002;25(3):283–92. EcosimPro; 2012.
[95] Kurti N. From Cailletet and Pictet to microkelvin. Cryogenics (Guildf) [135] Berstad DO, Stang JH, Nekså P. Large-scale hydrogen liquefier utilising mixed-
1978;18(8):451–8. refrigerant pre-cooling. Int J Hydrogen Energy 2010;35(10):4512–23.
[96] Kurti N. From the first mist of liquid oxygen to nuclear ordering: anecdotes from [136] Krasae-In S. Optimal operation of a large-scale liquid hydrogen plant utilizing
the history of refrigeration. Phys B+ C 1982;109:1737–52. mixed fluid refrigeration system. Int J Hydrogen Energy 2014;39(13):7015–29.
[97] McIntosh GE, Waynert J, Barclay J, Breon S, Daly E, Demko J, DiPirro M, Hull JR, [137] Rosen MA. Does industry embrace exergy? Exergy An Int J 2002;2(4):221–3.
Kelley PJ, Kittel P. Hydrogen liquefiers since 1950. AIP Conf Proc [138] Bejan A. Fundamentals of exergy analysis, entropy generation minimization, and
2004;710(1):9–15. the generation of flow architecture. Int J Energy Res 2002;26(7).
[98] Krasae-in S, Stang JH, Neksa P. Development of large-scale hydrogen liquefaction [139] Kotas TJ. The exergy method of thermal plant analysis. Elsevier; 2013.
processes from 1898 to 2009. Int J Hydrogen Energy 2010;35(10):4524–33. [140] Berstad DO, Stang JH, Nekså P. Comparison criteria for large-scale hydrogen li-
[99] Eucken A. Die Molekularwärme des Wasserstoffs bei tiefen Temperaturen; 1912. quefaction processes. Int J Hydrogen Energy 2009;34(3):1560–8.
[100] Heisenberg W. Multi-body problem and resonance in quantum mechanics II. Z [141] Shoenberg D. Piotr Leonidovich Kapitza. 9 July 1894–8 April 1984. Biogr Mem
Phys 1927;41:239. Fellows R Soc 1985;31:327–74.
[101] Hund F. Zur deutung der molekelspektren. II. Zeitschrift für Phys A Hadron Nucl [142] Kapitza P. The liquefaction of helium by an adiabatic method. Proc R Soc Lond A
1927;42(2):93–120. Math Phys Sci 1934;147(860):189–211.
[102] Dennison D. DENNISON 1927. Proc Roy Soc 1927;115:483. [143] Flynn TM. Crogenic engineering: revised and expanded, second ed.; 2005.
[103] Bonhoeffer KF, Harteck P. Experiments on para-hydrogen and ortho-hydrogen. [144] Al-Otaibi DA, Dincer I, Kalyon M. Thermoeconomic optimization of vapor-com-
Naturwissenschaften 1929;17:182. pression refrigeration systems. Int Commun Heat Mass Transf 2004;31(1):95–107.
[104] Giauque WF, Johnston HL. Symmetrical and antisymmetrical hydrogen and the [145] Nandi TK, Sarangi S. Performance and optimization of hydrogen liquefaction cy-
third law of thermodynamics. Thermal equilibrium and the triple point pressure. J cles. Int J Hydrogen Energy 1993;18(2):131–9.
Am Chem Soc 1928;50(12):3221–8. [146] Thomas RJ, Ghosh P, Chowdhury K. Exergy analysis of helium liquefaction sys-
[105] Brickwedde FG, Scott RB, Taylor HS. The difference in vapor pressures of ortho- tems based on modified Claude cycle with two-expanders. Cryogenics (Guildf)
and paradeuterium. J Res Natl Bur Std 1935;15:463–75. 2011;51(6):287–94.
[106] Schmauch GE, Singleton AH. Technical aspects of ortho-parahydrogen conversion. [147] Kundu A, Chowdhury K. Evaluating performance of mixed mode multistage he-
Ind Eng Chem 1964;56(5):20–31. lium plants for design and off-design conditions by exergy analysis. Int J Refrig
[107] Jensen JE, Stewart RG, Tuttle WA, Brechna H. Brookhaven national laboratory 2014;38(1):46–57.
selected cryogenic data notebook: sections I-IX. Brookhaven National Laboratory; [148] Drnevich R. Hydrogen Delivery, Liquefaction & Compression. strategic initiatives
1980. for hydrogen delivery workshop. New York; 2003.
[108] McCarty R, Hord J, Roder HM. NBS Monograph 168 Selected Properties of [149] Hendricks RC, Peller I, Baron AK. Joule Thomson Inversion Curves and Related
Hydrogen (Engineering Design Data); 1981. Coefficients. NASA Tech. Note, no. July; 1972. p. 1–32.
[109] Sherif SA, Zeytinoglu N, Veziroglu TN. Liquid Hydrogen: Potential, Problems and a [150] Peschka W. Liquid hydrogen: fuel of the future. Springer Science & Business
Proposed Research Program. Int J Hydrogen Energy 1997;22(7):683–8. Media; 2012.
[110] Notardonato WU. Analysis and testing of an integrated refrigeration and storage [151] Vander Arend PC, Chelton DB. The liquefaction of hydrogen 3A. Basic principles.
system for liquid hydrogen zero boil-off, liquefaction, and densification, vol. 68(1) Technol Uses Liq Hydrog 1964. p. 38.
2006. [152] Macinko J, Chelton DB, Dean J. Hydrogen liquefaction cycles. Advances in cryo-
[111] Gursu S, Sherif SA, Veziroglu TN, Sheffield JW. Analysis and optimization of genic engineering. Springer; 1960. p. 1–10.
thermal stratification and self-pressurization effects in liquid hydrogen storage [153] Drnevich R. Hydrogen delivery: liquefaction and compression. In: Strategic in-
systems—Part 1: model development. J Energy Resour Technol itiatives for Hydrogen Delivery Workshop, Tonawanda, NY, 7 May 2003; 2003.
1993;115(3):221–7. [154] Brown RN. Compressors: Selection and sizing. Gulf Professional Publishing; 1997.
[112] Zhang J, Fisher TS, Ramachandran PV, Gore JP, Mudawar I. A review of heat [155] Valenti G. Symbols and constants. Compend Hydrog Energy Hydrog Storage,
transfer issues in hydrogen storage technologies. J Heat Transfer Distrib Infrastruct 2016;27.
2005;127(12):1391–9. [156] Podbielniak W. Art of refrigeration; 26-May-1936.
[113] Tzimas E, Filiou C, Peteves SD, Veyret JB. Hydrogen storage: state-of-the-art and [157] Fuderer M, Andrija A. Verfahren zur Tiefkühling. Ger Pat 1969; 1426956.
future perspective. EU Comm. JRC Petten, EUR 20995EN; 2003. [158] Ruhemann M. Improvements in and relating to cooling processes and machines.
[114] Kalanidhi A. Boil-off in long-term stored liquid hydrogen. Int J Hydrogen Energy Sogutma ve lSI pompas1 Sist. sogutkan kan § lmlan Kullan. E. Erta § 1950; 3: p.
1988;13(5):311–3. 201–212.
[115] Fradkov AB, Troitskii VF. Liquefier with two-stage conversion to obtain 98 per [159] Ruhemann M. The separation of gases. London: Clarendon Press; 1949.
cent parahydrogen. Cryogenics (Guildf) 1965;5(3):136–7. [160] Del Nogal F, Kim J-K, Perry S, Smith R. Optimal design of mixed refrigerant cycles.
[116] Bliesner RM. Parahydrogen-orthohydrogen conversion for boil-off reduction from Ind Eng Chem Res 2008;47(22):8724–40.
space stage. Washington State University; 2013. [161] Walnum HT, Berstad D, Drescher M, Neksa P, Quack H, Haberstroh C. Principles
[117] Barrick PL, Weitzel DH, Connolly TW. Ortho-Para conversion studies. Advances in for the liquefaction of hydrogen with emphasis on precooling processes. In: 12th
Cryogenic Engineering. Cryogenics 2012; 2012. p. 8.
[118] White JF. Development of high-activity para-to ortho-hydrogen conversion cata- [162] Maytal B-Z, Pfotenhauer JM. Mixed coolant cryocooling. Miniature Joule-
lysts. Volume 2,” Air Products and Chemicals Inc Allentown Pa; 1989. Thomson Cryocooling. Springer; 2013. p. 277–334.
[119] Lowesmith BJ, Hankinson G, Chynoweth S. Safety issues of the liquefaction, sto- [163] Venkatarathnam G, Timmerhaus KD. Cryogenic mixed refrigerant processes vol.
rage and transportation of liquid hydrogen: An analysis of incidents and HAZIDS. 100. New York: Springer; 2008.
Int J Hydrogen Energy 2014;39(35):20516–21. [164] Venkatarathnam G, Mokashi G, Murthy SS. Occurrence of pinch points in con-
[120] Mazloomi K, Gomes C. Hydrogen as an energy carrier: prospects and challenges. densers and evaporators for zeotropic refrigerant mixtures Frrquence des points de
Renew Sustain Energy Rev 2012;16(5):3024–33. pincement dans des condenseurs et des 6vaporateurs pour des mrlanges de fri-
[121] Veziroglu TN. The fusion-hydrogen energy system. Appl Energy gorigrnes zrotropes mCp (- ∼ x mCp (∼ x rh (Oh ∼ ArUq i (Oh,” vol. 19, no. 6,
1994;47(2):227–49. pp. 361–368, 1996.
[122] Haselden GG. Cryogenic fundamentals. New York: Academic Press Inc; 1971. [165] Venkatarathnam G. Density marching method for calculating phase envelopes. 2.
[123] Barron RF. Liquefaction cycles for cryogens. Advances in cryogenic engineering. Three-phase envelopes. Ind Eng Chem Res 2014;53(30):12122–8.
Springer; 1972. p. 20–36. [166] Venkatarathnam G, Srinivasa Murthy S. Effect of mixture composition on the

81
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

formation of pinch points in condensers and evaporators for zeotropic refrigerant [206] Kramer GJ, Huijsmans J, Austgen D. Clean and green hydrogen. WHEC
mixtures. Int J Refrig 1999;22(3):205–15. 2006;16:1–9.
[167] Venkatarathnam G, Kumar PS, Murthy SS. Performance of a mixed refrigerant [207] Stang J, Nekså P, Brendeng E. On the design of an efficient hydrogen liquefaction
cascade refrigerator; 2003. p. 478–83. process. In: WHEC; 2006. p. 1–6.
[168] Xu X, Liu J, Cao L. Optimization and analysis of mixed refrigerant composition for [208] Shimko MA. Innovative hydrogen liquefaction cycle; 2008.
the PRICO natural gas liquefaction process. Cryogenics (Guildf) 2014;59:60–9. [209] Staats WL, Smith JL, Brisson JG. Analysis of a supercritical hydrogen liquefaction
[169] Xu X, Liu J, Jiang C, Cao L. The correlation between mixed refrigerant composition cycle. 2008; .p. 721–730.
and ambient conditions in the PRICO LNG process. Appl Energy [210] Valenti G, Macchi E. Proposal of an innovative, high-efficiency, large-scale hy-
2013;102:1127–36. drogen liquefier. Int J Hydrogen Energy 2008;33(12):3116–21.
[170] Narayanan V, Venkatarathnam G. Performance of two mixed refrigerant processes [211] Krasae-In S, Stang JH, Neksa P. Simulation on a proposed large-scale liquid hy-
providing refrigeration at 70 K. Cryogenics (Guildf) 2016;78:66–73. drogen plant using a multi-component refrigerant refrigeration system. Int J
[171] Wang M, Khalilpour R, Abbas A. Thermodynamic and economic optimization of Hydrogen Energy 2010;35(22):12531–44.
LNG mixed refrigerant processes. Energy Convers Manage 2014;88:947–61. [212] Krasae-In S, Stang JH, Neksa P. Exergy analysis on the simulation of a small-scale
[172] Ghorbani B, Mafi M, Salehi GR, Amidpour M, Nayenian SMM. The mathematical hydrogen liquefaction test rig with a multi-component refrigerant refrigeration
method and thermodynamic approaches to design multi-component refrigeration system. Int J Hydrogen Energy 2010;35(15):8030–42.
used in cryogenic process part II : optimal arrangement. 2014; 2(1): p. 13–21. [213] Krasae-In S, Bredesen AM, Stang JH, Neksa P. Simulation and experiment of a
[173] Quack H, Haberstroh C, Seemann I, Klaus M. Nelium, a refrigerant with high hydrogen liquefaction test rig using a multi-component refrigerant refrigeration
potential for the temperature range between 27 and 70 K. Phys Procedia system. Int J Hydrogen Energy 2011;36(1):907–19.
2015;67:176–82. [214] Yuksel YE, Ozturk M, Dincer I. Analysis and assessment of a novel hydrogen li-
[174] Khan MS, Lee S, Rangaiah GP, Lee M. Knowledge based decision making method quefaction process. Int J Hydrogen Energy 2017:1–10.
for the selection of mixed refrigerant systems for energy efficient LNG processes. [215] Sadaghiani MS, Mehrpooya M. Introducing and energy analysis of a novel cryo-
Appl Energy 2013;111:1018–31. genic hydrogen liquefaction process configuration. Int J Hydrogen Energy
[175] Mohanraj M, Jayaraj S, Muraleedharan C. Environment friendly alternatives to 2017;42(9):6033–50.
halogenated refrigerants-A review. Int J Greenh Gas Control 2009;3(1):108–19. [216] Asadnia M, Mehrpooya M. A novel hydrogen liquefaction process configuration
[176] Kochenburger TM, Grohmann S, Oellrich LR. Evaluation of a two-stage mixed with combined mixed refrigerant systems. Int J Hydrogen Energy 2017;42(23).
refrigerant cascade for HTS cooling below 60 K. Phys Procedia 2015;67:227–32. [217] Ansarinasab H, Mehrpooya M, Mohammadi A. Advanced exergy and ex-
[177] Kleemenko AP. One flow cascade cycle. In: Proc. of the 10th International ergoeconomic analyses of a hydrogen liquefaction plant equipped with mixed
Congress of Refrigeration, vol. 1; 1959. p. 34–39. refrigerant system. J Clean Prod 2017;144:248–59.
[178] Kerry F. Industrial gas handbook gas separation. Gas Sep Purif 2006. [218] Muradov NZ, Veziroğlu TN. ‘Green’ path from fossil-based to hydrogen economy:
[179] Pictet R. The liquefaction of oxygen and hydrogen, and the solidifying of hy- an overview of carbon-neutral technologies. Int J Hydrogen Energy
drogen. Sci Am 1878;5:2030–1. 2008;33(23):6804–39.
[180] Jeong S, Smith JL. Optimum temperature staging of cryogenic refrigeration [219] Kairouani L, Nehdi E. Cooling performance and energy saving of a compression-
system. Cryogenics (Guildf) 1994;34(11):929–33. absorption refrigeration system assisted by geothermal energy. Appl Therm Eng
[181] Xuan XC. Optimum staging of multistage exo-reversible refrigeration systems. 2006;26(2–3):288–94.
Cryogenics (Guildf) 2003;43(2):117–24. [220] Best R, Heard CL, Fernandez H, Siqueiros J. Developments in geothermal energy in
[182] Du K, Zhang S, Xu W, Niu X. A study on the cycle characteristics of an auto-cascade Mexico—Part five: The commissioning of an ammonia/water absorption cooler
refrigeration system. Exp Therm Fluid Sci 2009;33(2):240–5. operating on low enthalpy geothermal energy. J Heat Recover Syst
[183] Choi IH, Lee S, Seo Y, Chang D. Analysis and optimization of cascade Rankine 1986;6(3):209–16.
cycle for liquefied natural gas cold energy recovery. Energy 2013;61:179–95. [221] Kanoglu M, Dincer I, Rosen MA. Geothermal energy use in hydrogen liquefaction.
[184] James Evans West. The economics of small to medium liquid hydrogen facilities; Int J Hydrogen Energy 2007;32(17):4250–7.
2003. [222] Kanoglu M, Bolatturk A, Yilmaz C. Thermodynamic analysis of models used in
[185] Arend PCW, Chelton DB. 3B2. Large-Scale Hydrogen Liquefaction Facilities. hydrogen production by geothermal energy. Int J Hydrogen Energy
Technol Uses Liq Hydrog 2013. p. 79. 2010;35(16):8783–91.
[186] Brickwedde FG. A few remarks on the beginnings of the NBS-AEC cryogenic la- [223] Yilmaz C, Kanoglu M, Bolatturk A, Gadalla M. Economics of hydrogen production
boratory. Advances in cryogenic engineering. Springer; 1960. p. 1–4. and liquefaction by geothermal energy. Int J Hydrogen Energy
[187] Vander Arend PC. Large-scale production, handling, and storage of liquid hy- 2012;37(2):2058–69.
drogen. Advances in cryogenic engineering. Springer; 1960. p. 49–54. [224] Ratlamwala TAH, Dincer I, Gadalla MA, Kanoglu M. Thermodynamic analysis of a
[188] Gifford WE, Fuller PD. Liquid hydrogen from refinery waste gases. Advances in new renewable energy based hybrid system for hydrogen liquefaction. Int J
cryogenic engineering. Springer; 1961. p. 177–84. Hydrogen Energy 2012;37(23):18108–17.
[189] Flynn TM, Smith CN. Trends in cryogenic fluid production in the United States. [225] Ratlamwala TAH, Dincer I, Gadalla MA. Thermodynamic analysis of a novel in-
Advances in cryogenic engineering. Springer; 1972. p. 1–7. tegrated geothermal based power generation-quadruple effect absorption cooling-
[190] Kinard GE. The commercial use of liquid hydrogen over the last 40 years. In: hydrogen liquefaction system. Int J Hydrogen Energy 2012;37(7):5840–9.
International cryogenic engineering conference; 1998. p. 39–44. [226] Gadalla MA, Ratlamwala TAH, Dincer I, Kanoglu M. Performance assessment of an
[191] Ohta T, Abe I. Hydrogen energy research and developments in Japan. Int J integrated absorption cooling–hydrogen liquefaction system using geothermal
Hydrogen Energy 1985;10(5):275–9. energy. Int J Exergy 2013;12(2):205–25.
[192] Ohira K. A summary of liquid hydrogen and cryogenic technologies in Japan’s WE- [227] Kanoglu M, Yilmaz C, Abusoglu A. Geothermal energy use in absorption pre-
NET Project. AIP Conf Proc 2004;710(1):27–34. cooling for Claude hydrogen liquefaction cycle. Int J Hydrogen Energy 2016:1–16.
[193] Li SM. A brief overview of cryogenics in China. Cryogenics (Guildf) [228] Mehrpooya M, Omidi M, Vatani A. Novel mixed fluid cascade natural gas lique-
1995;35(5):317–9. faction process configuration using absorption refrigeration system. Appl Therm
[194] Bracha M, Lorenz G, Patzelt A, Wanner M. Large-scale hydrogen liquefaction in Eng 2016;98:591–604.
Germany. Int J Hydrogen Energy 1994;19(1):53–9. [229] Ozcan H, Dince I. Thermodynamic modeling of a nuclear energy based integrated
[195] Gross R, Otto W, Patzelt A, Wanner M. Liquid hydrogen for Europe–the Linde plant system for hydrogen production and liquefaction. Comput Chem Eng
at Ingolstadt. Linde-Reports Sci Technol 1994;54:37–43. 2016;90:234–46.
[196] Weindorf W, Bünger U, Schindler J. Comments on the paper by Baldur Eliasson [230] Bao JJ, Zhao L, Zhang WZ. A novel auto-cascade low-temperature solar Rankine
and Ulf Bossel ‘The future of the hydrogen economy: bright or bleak?’. Ottobrunn: cycle system for power generation. Sol Energy 2011;85(11):2710–9.
LB-Systemtechnik GmbH; 2003. [231] Hwangbo S, Lee S, Yoo C. Optimal network design of hydrogen production by
[197] Quack H. Die Schlusselrolle der Kryotechnik in der Wasserstoff-Energiewirtschaft. integrated utility and biogas supply networks. Appl Energy 2017.
Ki Luft Und Kältetechnik 2002;3:157–61. [232] Huang C, T-Raissi A. Analyses of one-step liquid hydrogen production from me-
[198] Ohlig K, Decker L. The latest developments and outlook for hydrogen liquefaction thane and landfill gas. J Power Sources 2007; 173(2 SPEC. ISS.): p. 950–958.
technology. In: 20th World Hydrog. Energy Conf. WHEC 2014, vol. 1; 2014. p. [233] Holladay JD, Hu J, King DL, Wang Y. An overview of hydrogen production
639–646. technologies. Catal Today 2009;139(4):244–60.
[199] Croft AJ. 3B1. Medium size hydrogen liquefiers. Technol Uses Liq Hydrog [234] Nikolaidis P, Poullikkas A. A comparative overview of hydrogen production pro-
1964:56. cesses. Renew Sustain Energy Rev 2017;67.
[200] Baker CR, Shaner RL. A study of the efficiency of hydrogen liquefaction. Int J [235] Lipman T. An overview of hydrogen production and storage systems with re-
Hydrogen Energy 1978;3(3):321–34. newable hydrogen case studies. Clean Energy States Alliance 2011;32.
[201] Matsuda H, Nagami M. Study of large hydrogen liquefaction process. 수소에너지 [236] Yilmaz C. Thermoeconomic modeling and optimization of a hydrogen production
1997; 8(3): p. 175. system using geothermal energy. Geothermics 2017;65:32–43.
[202] Quack H. Conceptual design of a high efficiency large capacity hydrogen liquefier. [237] Coskun C, Akyuz E, Oktay Z, Dincer I. Energy analysis of hydrogen production
AIP Conf Proc 2002;613(1):255–63. using biogas-based electricity. Int J Hydrogen Energy 2011;36(17):11418–24.
[203] Kuz’menko IF, Morkovkin IM, Gurov EI. Concept of Building Medium-Capacity [238] Riis T, Hagen EF, Vie PJS, Ulleberg Q. Hydrogen production and storage. R&D
Hydrogen Liquefiers With Helium. Chem Pet Eng 2004;40(1–2):94–8. Priorities Gaps 2006.
[204] Belyakov V, Krakovskii B. Low-capacity hydrogen liquefier with a helium cycle. [239] Syed M. An economic analysis of three hydrogen liquefaction systems. Int J
Chem Pet 2002;38(3):26–8. Hydrogen Energy 1998;23(7):565–76.
[205] Kuendig A, Lorhlein K, Kramer GJ, Huijsmans J. Large scale hydrogen liquefaction [240] Krewitt W, Schmid S. Fuel cell technologies and hydrogen production/distribution
in combination with LNG re-gasification. In: Proceedings of the 16th World options. Dlr, no. September; 2005. p. 1–113.
Hydrogen Energy Conference 2006, Lyon, France; 2006. p. 3326–3333. [241] Cardella U, Decker L, Sundberg J, Klein H. Process optimization for large-scale

82
M. Aasadnia, M. Mehrpooya Applied Energy 212 (2018) 57–83

hydrogen liquefaction. Int J Hydrogen Energy 2017;42(17):12339–54. [250] Demirbas A. Future hydrogen economy and policy. Energy Sources Part B Econ
[242] Cardella U, Decker L, Klein H. Roadmap to economically viable hydrogen lique- Plan Policy 2017;12(2):172–81.
faction. Int J Hydrogen Energy 2017;42(19):13329–38. [251] Momirlan M, Veziroglu TN. The properties of hydrogen as fuel tomorrow in sus-
[243] Kramer GJ, Huijsmans J, Austgen D. Clean and green hydrogen. Whec, no. June, tainable energy system for a cleaner planet. Int J Hydrogen Energy
2006. p. 1–9. 2005;30(7):795–802.
[244] Wolf J. Liquid hydrogen technology for vehicles. MRS Bull 2002;27(9):684–7. [252] Barreto L, Makihira A, Riahi K. The hydrogen economy in the 21st century: a
[245] Jones LW. Towards a liquid hydrogen fuel economy. 1970. p. 1–24. sustainable development scenario. Int J Hydrogen Energy 2002;28(43):1–10.
[246] Bossel U, Eliasson B, Taylor G. The future of the hydrogen economy: bright or [253] Berstad DO, Stang HGJ, Nekså P. A future energy chain based on liquefied hy-
bleak? Cogener Distrib Gener J 2003;18(3):29–70. drogen; 2009.
[247] Bossel U. Does a hydrogen economy make sense? Proc IEEE 2006;94(10):1826–36. [254] Cengel YA, Boles MA. Thermodynamics: an engineering approach. Sea
[248] Marchenko OV, Solomin SV. The future energy: Hydrogen versus electricity. Int J 2002;1000:8862.
Hydrogen Energy 2015;40(10):3801–5. [255] Garceau NM, Lim CM, Kim SY, Oh IH, Baik JH, Karng SW. Development of a small
[249] Penner SS. Steps toward the hydrogen economy. Energy 2006; 31(1 SPEC. ISS.): p. hydrogen liquefaction system. In: 20th World Hydrog. Energy Conf. WHEC 2014,
33–43. vol. 1; 2014. p. 631–38.

83