Liquid Organic Hydrogen Carriers as an efficient vector for the transport and storage of renewable energy

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

Available online at www.sciencedirect.com

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

Liquid Organic Hydrogen Carriers as an efficient vector for the transport and storage of renewable energy Daniel Teichmann*, Wolfgang Arlt, Peter Wasserscheid University of Erlangen-Nuremberg, Department for Chemical and Bioengineering, 91058 Erlangen, Germany

article info

abstract

Article history:

This contribution proposes the usage of Liquid Organic Hydrogen Carriers (LOHC) for the

Received 26 June 2012

storage and subsequently the transport of renewable energy. It is expected that a signifi-

Received in revised form

cant share of future energy consumption will be satisfied with the import of energy coming

11 August 2012

from regions with high potential for renewable generation, e.g. the import of solar power

Accepted 13 August 2012

from Northern Africa to Europe. In this context the transport of energy in form of chemical

Available online 27 September 2012

carriers is proposed supplementary to electrical transmission. Because of their high storage density and good manageability under ambient conditions Diesel-like LOHC substances

Keywords:

could be transported within the infrastructure that already exists for the handling of liquid

Hydrogen storage

fossil fuels (e.g. oil tankers, tank trucks, pipelines, etc.). A detailed assessment of energy

renewable energy

consumption as well as of transport costs is conducted that confirms the feasibility of the

Energy transport

concept.

LOHC

Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

Liquid hydrogen carriers

reserved.

Feasibility study

1.

Introduction

In many countries around the world there exist ambitious goals to reduce mankind’s dependency on fossil fuels like coal, oil and gas in favor of an increasing share of renewable energies like solar energy or wind power [1,2]. The hitherto existing vast use of the aforementioned fossil fuels in very different sectors has been a consequence not only of their good availability and comparably low costs, but also of their very beneficial physical handling characteristics. Especially liquid fossil fuels like crude oil have a high energy density and are therefore good energy carriers which can be transported and stored very efficiently e both in regard to technological and economic aspects.

With the transition of the energy system toward a higher share of renewable energy, hydrogen is often considered a very capable future energy vector [3]. It can be produced from renewable wind or solar power via electrolysis and has a wide range of potential applications in all important fields of energy supply. The gravimetric energy storage density of hydrogen is excellent. One kilogram carries about 33 kWh of energy. Being the chemical element with the lowest density, the volumetric storage density of hydrogen is a huge problem though. Under ambient conditions 1 l of gaseous hydrogen stores about 3 Wh of energy only. In existing technical applications hydrogen is therefore either stored in its gaseous state under very high pressures up to 700 bar (called “Compressed Gaseous

* Corresponding author. E-mail address: [email protected] (D. Teichmann). 0360-3199/$ e see front matter Copyright ª 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijhydene.2012.08.066

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

Hydrogen” or CGH2) or in its liquid state which requires temperatures below
253  C (called “Liquid Hydrogen” or LH2). This very low temperature allows for an ambient pressure storage of hydrogen but still with the extremely low liquid density of 71.2 kg/m3. In addition to the technological complexity of these storage concepts, the huge investment costs that are necessary to establish a nation-wide distribution infrastructure for hydrogen are a big challenge. The handling of CGH2 for example is very well known in the chemical industry and a few hydrogen pipelines already exist. But to make hydrogen available in the same way like natural gas or electricity, a complete new hydrogen distribution system would have to be installed in addition to the already existing grids for electricity and natural gas. Due to these limitations, researchers work on alternative concepts for the storage and transport of hydrogen in chemically bound forms [4]. One particularly attractive concept in this context is Energy Carrying Compounds (ETS, from the German name “EnergieTragende Stoffe”) [5,7]. Here the energy is stored in form of chemical compounds with a high energy content, which opens a chance to store big quantities over a longer period of time. The compound under consideration travels from the spot of energy delivery to the spot of energy demand and back. It is being charged with energy if the latter is available and it vice versa releases energy on demand. Like a catalyst the ETS is not consumed but undergoes a cyclic process of hydrogen loading and hydrogen releasing steps. This route of energy storage and distribution is virtually carbon free, so no CO2 is released in the utilization of the stored energy. One example for ETS is “Liquid Organic Hydrogen Carriers” (LOHC) where hydrogen does not exist in its molecular form but is covalently bound to a liquid carrier substance via hydrogenation [6e8]. At the time and place of energy demand, hydrogen can be released via dehydrogenation. The hydrogen carrying liquid itself is not consumed but can be reloaded and used in further cycles. Various substances have been discussed as potential ETS candidates, e.g. methylcyclohexaneetoluene by [6], a variety of cycloalkanes [9] and ammoniaborane-based systems [8,10]. The focus of this contribution lies on heterocyclic aromatic hydrocarbons like N-ethylcarbazole [11], which e due to extensive research e are currently among the best understood LOHC systems with convenient material properties for the application as an energy carrier. As hydrocarbazoles have many physico-chemical similarities to Diesel fuel, the complexity of handling, transporting and storing gaseous hydrogen is basically reduced to the handling of a liquid diesel-like substance. The fundamentals of the hydrogenation and dehydrogenation reactions of Nethylcarbazole are illustrated in Fig. 1. For a detailed description there exist various publications [12e14]. The melting point of fully dehydrogenated N-ethylcarbazole is 69.1  C and it is therefore a solid at ambient temperature. Perhydro-N-ethylcarbazole and the partially hydrogenated intermediates in contrast are liquids. To guarantee full liquid handling of the LOHC substance, the dehydrogenation process can be restricted to around 90% discharging by limiting the residence time within the catalytic

18119

Fig. 1 e Energy storage via hydrogenation and dehydrogenation of N-ethylcarbazole.

reaction system. While the theoretical storage density of Nethylcarbazole e perhydro-N-ethylcarbazole would be 5.8 wt%, only 5.3 wt-% materials energy density can be achieved following this approach. Existing safety data sheets show that the toxicity of Nethylcarbazole is rather uncritical, see Table 1 [15]. Because of its very low vapor pressure there is no detectable vapor phase under ambient conditions which further facilitates handling and safety issues of the carbazole LOHC system. Under the assumption that the return and reloading of the unloaded carrier material is made possible, the existing, very well established infrastructure for the distribution of mineral oil based fuels could be used for LOHC. Because of its high energy storage density and good handling characteristics a variety of applications for mobility, heating, long distance energy transport or long-term energy storage (for example for energy coming from intermittent producing renewable energies) can be envisaged [16]. The local potential for the installation of renewable energies is geographically not evenly distributed and therefore there often is a local mismatch between the level of energy consumption (strongly depending on population and industrialization density) and the potential for renewable energy. Therefore, it is foreseeable that with the transition toward a higher share of renewable energies like wind or solar, more and more power must be transported over long distances in the future. The ‘Desertec Industrial Initiative’ [17] for instance planned to produce electricity on a grand scale in Northern Africa and to import it to Europe either by HVDC-lines under

Table 1 e Material safety information N-ethylcarbazole. CAS number Toxicity data LD50 (oral) Skin Eye LC50 (96 h) Globally Harmonized System of Classification and Labeling of Chemicals Transport information

86-28-2 >5000 mg/kg ¼ non-toxic Not irritating (OECD 404) Not irritating (OECD 405) <10 mg/l (Golden Orfe) Aquatic chronic 2, H411, no signal word UN 3077 environmentally hazardous substance, solid, n.o.s., Class 9, PG III, Label 9

18120

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

the sea or through notably unstable countries on land. Similarly new power lines must be built in Germany to transport electricity from wind plants, which are primarily installed onand off-shore in Northern Germany, to the industrial centers in the South and West. It must be noted that an enhanced grid increases the number of customers but does not store energy at all. In public discussion an electric transmission of energy is often considered as the main option. But due to their attractive handling and storage characteristics (high energy density, negligible losses in long term storage), LOHC systems using the existing infrastructure for liquid energy carriers (tank ships, storage tanks, and filling stations) can be a very promising alternative.

2.

Methodology

It is the purpose of this contribution to evaluate the use of LOHCs for the storage, the transport and consequently the provision of renewable energy. Consequently individual process steps for different means of transportation e e.g. per ship, truck or pipeline e are evaluated with regard to the following two dimensions: - Efficiency dimension: energy consumption during transport and provision - Economic dimension: specific costs per kg hydrogen. This approach follows the concept of a “Life Cycle Assessment”, similar to existing Well-to-Wheel analyses that are carried out to assess energy consumption of mobile fuels provision and usage [18]. It tries to incorporate consumption in all steps of the fuel provision chain. For comparison’s sake the energy transport via LOHC is matched to the more mature hydrogen storage concepts of liquid (cryogenic) hydrogen (LH2) and compressed gaseous hydrogen (CGH2). To assure a good and fair comparability of the technologies, a high degree of homogeneity is strived for in all assumptions and considerations. The analyses do not act on the pre-condition of an allrenewable energy system where all forms of energy production and usage are satisfied by emission-free renewable energies. Instead it is assumed that hydrogen shall be produced sustainably from renewable energies, but its transportation via ship, truck, etc. might at least partly consume fossil energy. As it might take several decades to completely change our energy system from fossil to renewable energies, there will be a long transition phase during which the level of sustainability is continually raised. This fact provokes a new constraint to a new technology: it must be possible to introduce it step-bystep. This is fully reflected by the LOHC-concept. The results of the analysis depend significantly on the input parameters that are used. Some of the concepts that are considered here are in an early stage of development or research. The technology descriptions and assumptions for CGH2 and LH2 are primarily based on literature. Where possible, future cost degression and technology improvements were taken into account but still the matter-of-fact progress could prove otherwise. The data input for the analysis of LOHC is primarily based on own estimations,

technology comparisons and extrapolation of experimental findings.

2.1.

Efficiency dimension e energy consumption

During the processing, transportation and storage of hydrogen, energy is consumed that must be taken into account for the calculation of the energetic feasibility of the fuel provision chain. Although in thermodynamics energy can never be lost but only converted, the term “energy loss” or “energy consumption” is used here to describe for example waste heat that cannot be regained economically. For the considered paths the energy losses (Ex) occurring during a process step are calculated relative to the amount of transported energy (expressed by the lower heating value of the hydrogen, LHV, 33 kWh/kg). The energy content of Diesel fuel or marine fuel is 11.1 kWh/kg.

2.2. Economic dimension e specific costs for hydrogen transport The specific costs of hydrogen transport are very important indicators for the economic feasibility of the concept. They are calculated per kilogram of hydrogen. The “Equivalent Annuity Method” is applied in this contribution to determine the financial impact of the capital. Future cash flows are discounted to their ‘Net Present Value (NPV)’ to account for the differing points in time in which they occur. The ‘Equivalent Annuity Method’ calculates the annualized cash flow of an investment by dividing its net present value by the so-called annuity factor AF. Based on depreciation period and interest rate it expresses the annual cost of owning and using a piece of equipment or a facility. The socalled annuity factor AF is defined as follows (i being the interest rate, n the number of years of usage): n

AFi;n ¼

ð1 þ iÞ $i n ð1 þ iÞ
1

The annualized capital costs are calculated by multiplication of the AF with the investment costs. All costs are expressed in Euro. Whenever a conversion from US-Dollar is necessary, an exchange rate of 1.30 $/V is applied. Other types of costs: - Operating expenses: These on-going costs incur during the operation of a piece of equipment and include costs e.g. for maintenance and service, labor, insurance, etc. - Energy costs: The cost for electric energy is e unless stated otherwise e assumed to be 0.05 V/kWh. This takes into account that the energy-intensive processes will primarily be carried out in times of energy overproduction from renewable energies. It can be assumed that electricity prices are considerably lower than average during these times. The cost for Diesel fuel for road transport is assumed to be 1.40 V/l. Marine fuel oil is usually traded in US-$ per ton of fuel. The prices varied heavily in a range of 200e600 $/ton over the last years [19]. In the following a price of 350 V/ton is assumed.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

- Other costs: This position comprises various other costs which are characteristic for single processes and only occur in their specific context. All means of transport or storage of liquid cryogenic hydrogen come with a certain hydrogen loss rate due to boil-off. The economic value of the lost energy is taken into account with the hydrogen price being assumed to be 5 V/kg.

3. Assessment of different means of transportation for hydrogen and electricity

18121

The hydrogenation of LOHC is exothermic. As a consequence heat with a temperature between 150 and 170  C is produced during the reaction. As some part of this heat might be used in future applications this waste heat possesses a certain economic value. In favor of a conservative evaluation the use of this exothermic heat is not taken into account in this publication. The hydrogenation of the LOHC carrier material comprises the following cost parameters: a) Depreciation of capital costs

In this section different concepts for the transport of hydrogen as well as the electrical transmission of energy are analyzed. As the usage of compressed or liquid hydrogen has been the topic of various publications these technologies are used as reference in order to assess the LOHC concept.

3.1.

Processing of hydrogen

Because of its low volumetric energy density, hydrogen needs to be technically processed in order to improve its transportability and storage stability. In this contribution compression of hydrogen, liquefaction and the storage in liquid organic hydrogen carriers (LOHC) are considered. For LH2 and LOHC there is only one conditioning process as the first stage of the transport process. Subsequently the physical state is kept until the hydrogen is finally used. By contrast a singular exclusive condition step usually doesn’t exist for CGH2 as the pressure level varies over the distribution chain. For example there might be a compression step right after hydrogen production up to 200 bar for truck transport and then a final compression to 750 bar at the pumping station. Even if there might not be a single compression but several consecutive steps a compression from production output level (30 bar) to 350 bar is assumed in the following.

3.1.1.

Compression of hydrogen (CGH2)

Usually multi-stage compressors are used to provide high output pressures. The pressure level depends on the application and means of transport. In tube trailers hydrogen is usually stored at 200 bar. For mobility applications CGH2 is usually delivered at 350 or 700 bar pressure.

3.1.2.

Liquefaction of hydrogen (LH2)

Hydrogen in its liquid state has an energy density of around 2.3 kWh/l. As hydrogen has to be cooled down to
253  C the liquefaction process is very energy intensive. Worldwide there exist several liquefaction plants with some tons of hydrogen capacity per day. In future facilities with higher capacity, a significant reduction of energy consumption and much lower investment costs seem possible.

3.1.3.

Hydrogenation of carrier substance (LOHC)

Facilities for catalytic hydrogenation exist in large scale in refineries and chemical plants. The special application of LOHC hydrogenation does e apart from laboratory scale e not exist today. But very similar molecules are hydrogenated in large scale operations. Therefore important parameters like capital costs or process efficiencies can be derived from these similar cases.

Large-scale hydrogenation of aromatic hydrocarbons is a standard procedure in the chemical industry with millions of tons being processed every year. Because of the physicochemical similarities between LOHC and Diesel fuel, Diesel Hydrodesulfurization (HDS) and Hydrodearomatization (HDA) (both processes are basically hydrogenations) can be used as benchmark to assess the level of investment costs for LOHC hydrogenation [20]. Yamaguchi gives investment costs between 1600 and 2300 USD per barrel per stream day (bpsd) of throughput [21]. Uhde mentions specific costs for a HDS/HDA facility of around 2300 USD/bpsd [22]. Conservatively, 3000 V/ bpsd or 260 V/kW (LHV of stored hydrogen) of investment cost are assumed. This value corresponds pretty well to the assumptions that are proposed by TIAX for a large-scale LOHC hydrogenation device [23] (TIAX estimates the total costs for a hydrogenation device with 250 tons of H2 per day as 526 million $. Following the approach of this publication where carrier material and storage cost is assessed separately this corresponds to 129 million $ or 285 V/kW). Regarding the high throughput of LOHC material there is also a contribution to the capital cost coming from the LOHC material which is permanently tied up in the plant. With the assumption that this amount corresponds to half a day of production output, 6.8 kg of LOHC per kW installed power with an estimated economic value of 27 V (4 V per kg LOHC) are also to be assessed as capital investment. b) Operating expenses In accordance with the assumptions made for LH2 and CGH2, operating expenses are 3% of the investment costs per year. c) Cost of catalyst material The hydrogenation of LOHC is a catalytic reaction. Like for technologically similar hydrogenations of organic compounds (e.g. hydro-desulfurization of Diesel fuel), catalyst systems based on palladium or platinum can be used for the hydrogenation of LOHC (weight-percentage of precious metals is typically below 0.5%). TIAX assumes a catalyst productivity of 500,000 kg of LOHC per one kg of catalyst material and catalyst costs of around 150 V per kg [23]. d) Cost for recycling/replacement of carrier material As the carrier material is not consumed in the energy releasing process, it is being reloaded with hydrogen after

18122

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

every cycle. On a laboratory scale many consecutive cycles of hydrogenation and dehydrogenation have been performed in the past years and showed a very high cycle stability with no perceptible degradation or carrier loss. In order to attain a conservative evaluation of the concept a degradation rate per cycle of 0.1% and LOHC material costs of 4 V/kg are assumed [23,24]. e) Electricity costs Apart from the water electrolysis to produce hydrogen, electric power is needed primarily for the compression of hydrogen from the outlet pressure level of the electrolysis (30 bar) to the reaction pressure of 70 bar. Around 0.011 kWh of electrical energy are necessary per kWh of processed hydrogen [5].

3.1.4.

Results

Table 2 shows assumptions and results regarding costs and energy consumption of hydrogen storage in detail.

3.2.

Hydrogen transport by sea

Regarding a world-wide system of renewable energy production and demand, the long-distance transport of hydrogen by ship might play an important role in the future. While the concept of transporting liquid hydrogen per ship has already been considered, the long-distance transport of compressed hydrogen is not attractive due to its low energy density and not feasible in existing ships. The Diesel-like attributes of LOHC however make it possible to transport this material like crude oil in existing oil tankers. As this approach seems very promising it is evaluated in detail in the following.

3.2.1.

Sea transport of liquid hydrogen

The intercontinental transport of liquid hydrogen has been considered, for example in connection with the ‘Euro Quebec Hydro Hydrogen Pilot Project (EQHHPP)’. The purpose of this project was to evaluate the transport of electricity from hydro power from Canada to Germany via sea transportation of liquid hydrogen. The values used are based on a concept for a ‘LH2-barge carrier’ [26]. This carrier ship would carry 5 LH2 containers, each containing 210 tons of liquid hydrogen.

3.2.2.

Sea transport of LOHC

As mentioned, the transportation of energy via LOHC is very interesting as existing ships and port infrastructures could be used. Tankers are often classified by their size in deadweight tonnage (dwt) ranging from 10,000 dwt for small tankers up to 550,0004 dwt for ‘Ultra-large crude carriers’ [27]. ‘Product tankers’ are used for the transport of chemical substances and would be the first choice for a transportation of LOHC. For the transportation distances that are considered here ‘Handysize’ (10,000e30,000 dwt) or ‘Handymax’ classes (30,000e45,000 dwt) are a feasible option. Currently there are about 2300 ships of these two classes in operation worldwide (total number of tankers >5000 dwt is 5300) [28]. The considerations are based on a 45,000 dwt product tanker (Handymax class) with a drive power of 9000 kW [27] and a speed of 15 knots.

3.2.3.

Results

Table 3 shows assumptions and results of the analysis for the transport of hydrogen by sea in detail. There are two main reasons for the comparably high costs of LH2 transport: - With 1050 tons of hydrogen the maximum cargo is comparably low resulting in high energy consumption and costs per kg.

Table 2 e Cost and energy assessment for the storage of hydrogen. Description Cost assessment

Assumptions

Results

Energy

Capital cost per kW (LHV hydrogen) Consumption of electric energy Depreciation period Interest rate Annuity factor Operating time per year Operating expenses Catalyst productivity Substitution rate of LOHC Depreciation costs per kg hydrogen Electricity costs Operating expenses Catalyst cost Substitution of LOHC material Total costs for conditioning Energy consumption

Unit V/kW kWhel/kWhHyd years

hours % of invest p.a. kg LOHC/kg Cat % per cycle V/kg hydrogen V/kg hydrogen V/kg hydrogen V/kg hydrogen V/kg hydrogen V/kg hydrogen kWhx/kWhHyd

LH2

CGH2

LOHC

a

b

287 0.011 20 6.00% 8.7% 8000 3% 500,000 0.1% 0.103 V 0.018 V 0.036 V 0.006 V 0.075 V 0.238 V 1.1%

797 0.21a 20 6.00% 8.7% 8000 3%

383 0.035b 20 6.00% 8.7% 8000 3%

0.287 V 0.347 V 0.099 V

0.138 V 0.058 V 0.047 V

0.732 V 21.0%

0.243 V 3.5%

a [18] and [25] give an energy consumption of 0.3 kWh of electrical energy per kWh of liquefied hydrogen with a possible improvement to 0.21 until 2030. Investment costs for a 300 MW unit (lower heating value of the hydrogen) are given to be 239 million V. b The energy consumption for a hydrogen compression from 30 to 350 bar mainly depends on the efficiency and size of the compressors. In the following calculations an energy consumption of 0.035 kWh of electrical energy per kWh of compressed hydrogen is used [25,31]. While [31] gives specific investment costs of 12,500 V per kg/h compression capacity, [25] gives around 2300 V and [48] around 9000 V per kg/h capacity. In the following capital costs of 12,600 V per kg/h are assumed.

18123

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

Table 3 e Cost and energy assessment for the sea transport of hydrogen via ship. Description

Unit

1000 km Cost assessment

Assum.

Results

Energy

Cargo (mass of transported hydrogen) Drive power Purchase costs of ship Depreciation period Interest rate Annuity factor Hydrogen boil off Cost of operation Duration of loading / unloading Cruising speed Total time of (un-)loading and travel Fuel consumption per day Depreciation cost for ship per day Amount of consumed fuel Depreciation cost per kg Hydrogen Operating expenses Fuel cost Hydrogen loss Total costs of ship transport Energy consumption

tons kW million V years

% of cargo per day V per day days knots days tons V tons V/kg hydrogen V/kg hydrogen V/kg hydrogen V/kg hydrogen V/kg hydrogen kWhx/kWhHyd

LOHC

LH2 5000 km

1050 10,540 146a 25 6.00% 7.8% 0.1%b 11,500 Vc 2 18 6.5 16.5 65 31,291 V 163 815 0.194 V 0.492 V 0.071 V 0.181 V 0.054 V 0.272 V 0.006 V 0.031 V 0.326 V 0.976 V 5.2% 26.1%

1000 km

5000 km

2400 9000 35d 25 6.00% 7.8% e 5000 Ve 2 15 7

19 56 7501 V 167 835 0.022 V 0.059 V 0.015 V 0.040 V 0.024 V 0.122 V 0.061 V 2.3%

0.221 V 11.7%

a Capital costs were assumed to be 146 million V with the ship having 10 MW drive power at full speed (18 kn ¼ 33 km/h) [26]. b Like for all cryogenic hydrogen storage systems a minimal heat transfer is unavoidable despite complex insulation of the tanks. The boil-off rate e the amount of hydrogen that is lost due to evaporation and consequently boiling off e is expected to be in the order of 0.1% of the cargo per day (not on the return trip as the cargo is empty). Assuming a hydrogen price of 4 V/kg the boil-off contributes to the total cost of transportation. c According to [26] operating expenses are approx. 11,500 V per day. d According to [27] the capital costs for a new ship of this type are approximately 43 million $ or 35 million V. 45,000 deadweight tonnage of LOHC corresponds to about 2400 tons of Hydrogen cargo (storage density of 5.3%). e The operating expenses of a tanker of this class can be assumed to be 5000 V per day [29].

- The capital investment for the LH2 ship is very high due to the technologically complex insulation of the cryogenic LH2 containers. That results in high annual depreciation.

3.3.2. Road transportation of compressed gaseous hydrogen (CGH2)

The usage of existing fleets of product tankers that could be realized via LOHC promises considerably lower investments and the option of an incremental introduction of hydrogen technology.

CGH2 distribution is carried out by trailers which carry a multitude of pressure cylinders (cylinder bundle). The amount of hydrogen that is carried depends on the pressure level. According to the study Roads2Hy [30] 340 kg can be carried while Valentin [31] gives a value of 462 kg. Optimistically 500 kg are assumed in the following.

3.3.

3.3.3.

Hydrogen distribution per truck

Hydrogen delivery via road is comparably expensive. Nevertheless, the ‘last mile’ of transportation to the end user is often conducted by a truck. There already exist cryogenic trailers for LH2 and cylinder bundle trailers for CGH2 transportation. For the delivery of LOHC standard road tankers could be used. In this assessment, distances of 20 km and 50 km were considered for road transportation of hydrogen. These are typical distances for the distribution of fuel from a central storage facility to the consumer (e.g. gasoline station, home heating systems).

3.3.1.

Road transportation of liquid hydrogen (LH2)

LH2 can be transported in cryogenic trailers with a hydrogen capacity of 3500 kg [26] for a 40-ton truck. Hydrogen loss due to boil-off is approximately 0.5% per journey [31].

Road transportation of LOHC

It is assumed that road tankers, which qualify for the transport of liquid, flammable substances (class 3, GHS e ‘Globally Harmonized System of Classification and Labeling of Chemicals’) can also be used for the transport of LOHC [23,32]. These road tankers usually consist of several separated compartments and could therefore even carry loaded and unloaded LOHC material at the same time if necessary. Regarding the maximum cargo the gross vehicle weight is the limiting factor for the transport of LOHC. Subtracting the weight of the truck, the maximum net load is 28.5 tons. Based on a storage density of 5.3 wt-%, a LOHC trailer could carry 1500 kg of hydrogen. While the maximum cargo for liquid hydrogen LH2 is mainly determined by volume constraints, the net load capacity is the limiting factor for LOHC transport. Unlike a gasoline tanker, the LOHC trailer would always be filled as the unloaded carrier has to be returned in exchange to fresh material. It has to be considered that this return journey

18124

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

Table 4 e Cost assessment for the truck and the trailers used for hydrogen transport. Trucka Purchasing costs Depreciation period Interest rate Annuity factor Maintenance per year Operating time Truck/trailer costs

Euro years

% of capital cost hours per year per hour

LH2-trailer

160,000 V 8 6.00% 0.16 12% 3500 12.85 V

450,000 V 12 6.00% 0.12 2% 3500 17.91 V

CGH2-trailer

b

250,000 V 12 6.00% 0.12 2% 3500 9.95 V

c

LOHC-trailer 80,000 Vd 12 6.00% 0.12 2% 3500 3.18 V

a It is assumed that special trailers for CGH2, LH2 and LOHC transportation are towed by a Mercedes Benz Actros truck. According to [26] the average fuel consumption is 35 l/100 km, the capital costs approx. 160,000 V with a depreciation period of 8 years. b Capital costs are assumed to be in the order of 450,000 V (500,000 V in [26], according [31] 770,000 V at the year of the publication and 435,000 V as future potential). The depreciation period is assumed to be 12 years. Hydrogen Boil-off during transport and especially during loading/ unloading operations is assumed to be 0.5% of the transported hydrogen [26]. c The capital costs for the trailer are 190,000 V [30] or according to [31] 360,000 at present and 250,000 V in the future. The latter value is assumed in this assessment. d The capital costs for the described liquid fuel trailer are assumed to be 80,000 V [23] with a depreciation period of 12 years.

e back to the fuel silo e is also necessary for the empty CGH2 or LH2 trailers. The additional energy consumption (as a consequence of the higher vehicle weight) is comparably small.

3.3.4.

Results

As a first step, the costs for the truck and for the different trailers per hour of operation were calculated according to the assumptions (Table 4). Using these results Table 5 shows the total costs and energy consumption for different means of hydrogen transport via truck. The costs and energy consumptions for the road transport of LH2 and LOHC are similar. Apparently the higher cargo of LH2 compensates for the higher capital investment. CGH2 trailers can only carry a small amount of hydrogen which makes this way of transportation rather inefficient and very costly.

3.4.

Release of hydrogen

After transportation hydrogen has to be released for further use. Liquid Hydrogen needs to be vaporized prior to use in a fuel cell. The energy demand depends on the pressure requirements of the application and can amount to up to 1.9% [18]. Only if it is being used in an internal combustion engine liquid cryogenic hydrogen might improve power characteristics by cooling down charge air. Also compressed gaseous hydrogen has to be expanded under heat input for low pressure applications. Due to its endothermicity the release of hydrogen from a LOHC system via catalytic dehydrogenation requires energy input. For the system N-ethylcarbazole/perhydro-N-ethylcarbazole about 20% of the energy content of the hydrogen need to be supplied thermally to enable the reaction (temperature level approx. 230  C). As no buffer storage between release and usage of hydrogen is envisaged, there is

Table 5 e Cost and energy assessment for the land transport of hydrogen via truck. Description

Unit

LH2 20 km

Cost assessment

Assum.

Results

Energy

Cargo (mass of hydrogen) Duration of loading/unloading Average driving speed Journey time (return trip) Duration of (un-)loading and travel Truck costs Trailer costs Labor costsa Fuel costs Hydrogen loss due to boil-off Total costs of truck transport Energy consumption

kg hours km/h hours hours V/kg hyd V/kg hyd V/kg hyd V/kg hyd V/kg hyd V/kg hyd kWhx/kWhHyd

50 km

3500 3 30 45 1.33 2.22 5.2b 4.3b 0.016 0.019 0.022 0.027 0.043 0.052 0.006 0.014 0.025 0.025 0.112 0.137 0.62% 0.80%

CGH2 20 km

30 1.33 3.3 0.086 0.066 0.163 0.039 e 0.355 0.83%

50 km

500 2 45 2.22 4.2 0.108 0.084 0.226 0.098 e 0.516 2.08%

LOHC 20 km

50 km

1500 2 30 45 1.33 2.22 2.8 3.7 0.024 0.032 0.006 0.008 0.066 0.087 0.013 0.033 e e 0.109 0.159 0.28% 0.69%

a Labor costs for the driver are assumed to be 35 V/h. Note that the driver also executes the charging operations so that no further personnel is required. b According to [31] and [26] the LH2 charging operation lasts between 5 and 8 h today with a potential decrease to 1.5 h in the future, when there exist more standardized operation protocols and technologies. In the following, a total time demand of 3 h is assumed for both charging and discharging of the trailer.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

a close temporal link between the energy need for hydrogen release and the typically exothermic hydrogen usage. Downstream consumers for hydrogen like gas turbines or fuel cells have limited efficiency. As long as the heat level of their waste heat is sufficiently high, this waste heat can be used for the dehydrogenation of LOHC. That holds true for gas turbines, combustion engines and solid oxide fuel cells (SOFC). Pairing LOHC dehydrogenation with a PEM fuel cell currently necessitates additional heat supply because of the insufficient operating temperature. As electric heating is usually not reasonable, a possible solution would be to burn some of the produced hydrogen in a catalytic burner in order to support the reaction e this results in an energy penalty. It is important to note however, that fuel cell producers report about small waste flows of hydrogen in the fuel cell that could be also used for this heating purpose. The use of a PEM fuel cell is realistic for mobility applications, for example in a hydrogen car. The energy penalty doesn’t apply though if using solid oxide fuel cells, combustion engines or gas turbines in stationary applications, such as house heating (either combustion of hydrogen or SOFC) or electricity generation (presumably gas turbines or SOFC) which are also the standard scenario for the following considerations. The dehydrogenation of organic substances is e unlike hydrogenation e not a standard procedure in the refinery or chemical industries. TIAX has considered the cost for an automotive on-board dehydrogenation device which is the basis for the values that are considered here [23]. While a car application comes with only some hundred operation hours per year, the focus of this publication is on large-scale offboard dehydrogenation with high capacity factors. Therefore some modifications to TIAX’s data are made. In order to power a 70 kW fuel cell (efficiency 55%), a 127 kW dehydrogenation device is necessary. According to TIAX, investment costs for this device could be 2500 USD or 15 V/kW (LHV hydrogen) including a catalytic burner for heat generation (for the PEM FC case). Regarding the requirements of a long-life stationary application, 40 V/kW are assumed in the following as a conservative value. Indirect costs like land and projects costs etc. are included by assuming a mark up on the reactor costs (Table 6).

Table 6 e Costs for hydrogen release from LOHC. Description Assumptions

Results

Capital costs Depreciation period Interest rate Annuity factor Operating hours per year Indirect costs Depreciation Indirect costs Total costs for H2 release from LOHC

Unit

Stationary HRU

V/kW years

40 20

hours

6.00% 8.72% 8000

% of capital V/kg H2 V/kg H2 V/kg H2

20% 0.014 V 0.003 V 0.017 V

18125

3.5. Electric transmission of energy via High Voltage Direct Current Transmission (HVDC) In comparison to HVAC (alternating current), HVDC technology allows for energy transport over long distances with only little losses. A good overview over this technology is given in the Trans-CSP study by DLR [33]. Table 7 shows costs and energy consumption for the transmission of electric energy via HVDC over a distance of 3100 km. Thereby the costs for electricity lost during transmission is also accounted for.

4. Scenarios for the long-distance transport of energy Having evaluated energy consumption and costs for several means of transportation, this section uses the gathered information to comprehensively evaluate relevant scenarios for the transport of energy.

4.1. Transportation of solar power from Northern Africa to Europe The annual energy demand of the whole world could theoretically be satisfied by the solar irradiation on a comparably small fraction of the Sahara desert in Northern Africa [17]. The ‘Desertec Industrial Initiative’, which was founded in 2009, therefore strives for the installment of huge CSP (Concentrated Solar Power) plants. Part of the produced energy could be exported to Europe via HVDC (high voltage direct current) transmission which allows for lower energy losses than conventional high voltage AC power lines. Several publications have discussed the alternatives of energy transport, namely via electrical transmission versus the transformation into a chemical carrier like hydrogen [34e36]. For the evaluation of the relative competitiveness it is a crucial assumption whether the energy is demanded as electrical energy or as hydrogen at the destination of the transport. In the case that an electricity-to-electricity transmission of energy is realized via hydrogen as energy carrier, the total efficiency is limited to a maximum of about 40% just by the efficiency of the “electricity to hydrogen” and “hydrogen to electricity” processes. In contrast, HVDC transmission allows for total efficiencies of up to 90%. However, it is important to note that this HVDC efficiency value only takes into account the transport losses in the long-distance transfer of electric energy and does not take into account the huge investment into the HVDC infrastructure. The future share of energy that is needed in the world in form of hydrogen depends on many factors and cannot be determined per se. It can be concluded however that an allelectric energy system (consequently with no demand for chemical energy carriers like hydrogen) appears rather unrealistic at this date due to the difficulties in storing large amounts of electric energy. In the contrary, the need for energy storage is increasing as a consequence of the increasing share of the unsteady production of renewable energies from wind or solar. Massive additional capacities in both wind and solar energy production are currently being installed in parts of Europe. Hydrogen in the form of LOHC

18126

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

Table 7 e Costs for transmission of electrical energy via High Voltage Direct Current Transmission (HVDC). Description Cost assessment

Assumptions

Results

Energy

Cable length HVDC overhead line cost HVDC sea cable costs Share of overhead lines Terminal costs Depreciation period Interest rate Annuity factor Capacity factor Energy losses Terminal transformation losses Transported energy Energy losses Total investment Depreciation cost Value of energy losses Total costs Energy consumption

could be stored efficiently in large quantities until it is needed (day to seasonal storage). It could be used not only for electrical applications but could replace fossil fuels in other fields, for example in residential heating, mobility or industry. Assuming a sufficient demand for hydrogen at the destination of the energy transport, there are two options regarding the place of hydrogen production from electricity via electrolysis. This process can either happen in Europe with electricity being transmitted from Northern Africa via power lines. Alternatively, the electrolysis could be carried out at the place of energy production in Northern Africa. In this case, hydrogen would be the energy transport vector. Hydrogen transport is considered in the following via LH2 or LOHC, respectively (Fig. 2). These two aforementioned concepts are compared in the following. The total future hydrogen demand for Europe is very difficult to forecast and depends heavily on the introduction and propagation of hydrogen technologies. In the following an annual European hydrogen demand of 1 million tons (33 TWh LHV) is assumed. This amount of hydrogen would be realistic only for early stages of a hydrogen-based energy system. Assuming an average annual hydrogen consumption per fuel cell car of 140 kg (average driving range

Unit km MV/1000 km MV/1000 km of total length MV Years

% per 1000 km TWh per year TWh per year MV V/kWh/1000 km V/kWh V/kWh kWhx/kWh

HVDC 3100 400 2100 80% 650 30 6.0% 7.3% 50% 2.5% 0.7% 21.9 1.85 2.944 0.0098 V 0.0042 V 0.014 V 8.5%

of 14,000 km, fuel consumption 1 kg H2/100 km), 1 million tons of hydrogen would be sufficient to power about 7 million cars which represent only 3% of the whole fleet of vehicles in the European Union (234 million cars). Consequently overall hydrogen demand not only for mobility but also for stationary applications would be much higher. The amount of hydrogen necessary to supply 10% of the overall primary energy demand of Europe (76 EJ in 2008 [2]) would be 64 million tons per year. As the effects of cost degression play a minor role regarding the considered dimensions, the mentioned earlyphase scenario of 1 million tons of hydrogen demand is capable of providing a realistic estimation of feasibility also for future higher demands. The transmission capacity which is needed for the aforementioned scenario (1 million tons hydrogen) corresponds very well to the 2020 scenario of desertec where 60 TWh of energy are to be exported to Europe via two HVDC lines [17]. One million tons of hydrogen would consequently be equivalent to one of these transmission lines. For future hydrogen production the efficiency and capital costs of electrolysis play an important role, especially PEMelectrolysis (polymer electrolyte membrane) has a high potential [37].

Fig. 2 e Options for energy transfer from Northern Africa to Europe for the purpose of hydrogen production.

18127

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

Table 8 e Cost and energy assessment for the production of hydrogen via electrolysis. Description Cost assessment

Assumptions

Results

Energy

Unit

Capital costs Electricity price Depreciation period Interest rate Annuity factor Operating hours per year Operating expenses Energy consumption Depreciation cost Electricity costs Operating expenses Total costs Energy consumption

Table 8 shows assumption and results for the costs and energy consumption of hydrogen production via electrolysis [37]. Two different prices for electric energy are assumed that are relevant for the scenarios that are under consideration in the following. The transportation route for LOHC and LH2 consists of the following steps (shorter distances would be very well possible depending on the destination):  4000 km ship transport to relevant ports in Northern Europe  100 km truck transport in Northern Africa and Europe combined. For electric transmission a cable length of 3100 km is assumed [33], 20% of the distance are via sea cables.

V/kW V/kWh Years

Hours % of capital cost kWhel/kgHyd V/kg V/kg V/kg V/kg kWhx/kWhHyd

Electrolysis 800 0.02

0.05 20 6.00% 8.72% 6000 3% 47

0.55 0.94 0.19 1.67 V 29.79%

0.55 2.35 0.19 3.08 V 29.79%

For all three routes further transportation of the hydrogen may be necessary depending on the application that it is used for. As this depends on the chosen scenario assumptions in equal measure no further distribution is accounted for (Table 9). The analysis was performed under some simplifications, for example there were no detailed considerations for the exact routes of transport for ships and electric lines. Still the results give a first indication that energy transport of solar power from Northern Africa to Europe via LOHC could be an interesting option to complement the electric transmission of energy. Besides the costs shown here there are more arguments in favor of a chemical energy transport instead of electrical transmission that are given in the conclusion section.

Table 9 e Cost and energy assessment for the transmission of energy from Northern Africa to Europe. Process steps

Location

Form of energy processed El. Power TWh

Electr. Transm. Export power HVDC transm. (3100 km) Electrolysis Final hydr delivery Total energy/costs LH2 Export power Electrolysis LH2 liquefaction LH2 transport Final hydr delivery Total energy/costs LOHC Export power Electrolysis LOHC conditioning LOHC transport Final hydr release Total energy/costs

North Africa Europe Europe

51.3 51.3 47

Hydrogen million tons

1 1

Energy consumption TWh

Costs million V

4.3 14

658 3080

18.3 ¼ 0.55 kWhx/kWh

North Africa North Africa North Africa

48.2 48.2

Europe

North Africa North Africa North Africa Europe

47.14 47

1.03 1.03 1.03 1

1 1 1 1

14.4 7.1 7.6

3738 ¼ 3.74 V/kg

3157 750 1081

29.1 ¼ 0.88 kWhx/kWh

4988 ¼ 4.99 V/kg

14 0.363 3.5 0 17.9 ¼ 0.54 kWhx/kWh

3080 238 495 17 3830 ¼ 3.83 V/kg

18128

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

Table 10 e Cost and energy assessment for the production of hydrogen in Iceland and transport to Germany (2000 km scenario). Process steps

Form of energy processed El. Power TWh

LOHC Export power Electrolysis LOHC conditioning LOHC Transport Final hydr release Energy consumption per kWhHyd Costs per kg Hydrogen

10.00 10.00

The transport of liquid hydrogen via ship is expensive and inefficient mainly due the high energy consumption during liquefaction and the limited cargo capacity of the ship.

4.2.

Import of renewable energy from Iceland

4.2.1.

Evaluation of costs and efficiency

Thanks to its geologically unique position on the tectonically active Mid-Atlantic ridge, Iceland is able to cover almost all of its energy consumption from renewable sources. In 2009 about 81% of primary energy and 100% of electricity demand were satisfied by geothermal and hydroelectric sources. The geothermal energy is mainly used for heating purposes and increasingly for electricity production (27% in 2009) while the remaining 73% come from hydro power, fed mainly from glaciers [38,39]. Due to its comparably low electricity prices Iceland has attracted energy-intensive industries, especially aluminum production which accounts for 74% of Iceland’s total electricity consumption (16.2 TWh in 2009). Furthermore, there exists considerable additional potential for production from renewable energies. According to [38] the potential for annual geothermal electricity production is 20 TWh and for hydro power 30 TWh, respectively. Only projects with limited environmental impact have been considered. Due to the large distance to consumers in Europe or North America the electricity market of Iceland is isolated. There is only a limited number of consumers (318,000 inhabitants) facing a vast potential for renewable energy generation as described. As a consequence, electricity prices for industrial consumers are reported to be far below 3 V-Ct./kWh in Iceland [40]. Market insiders have even reported about prices below 1 V-Ct./kWh for the aluminum industry. Wouldn’t it be for the lack of technical solutions for electricity storage and transport, Iceland could be a successful exporter of energy. The following analysis calculates cost and energy consumption for the transport of energy from Iceland to Europe via LOHC. Analogous to the LOHC scenario in Chapter 4.1 hydrogen is produced in Iceland and stored in the form of LOHC. The shortest possible connection to Germany is approx. 2000 km. The basis for the calculation is a net export of 10 TWhs of energy per year. Electricity price is assumed to be 2 Ct./kWh. The results are shown in Table 10.

Assessment of

Hydrogen million tons

Energy TWh

0.30 0.30 0.30 0.30

14.2 0.1 0.5 0 5.8%

Costs million tons

506 72 32 5 2.03 V

Export of ‘clean’ hydrogen could be an interesting option for Iceland’s economy. As Table 10 shows, total cost for hydrogen production could be around 2 V per kg with energy prices contributing 0.94 V (assuming 2 V-Ct./kWh electricity), depreciation and operation of the electrolysis unit 0.74 V and transport costs of 0.35 V/kg as shown (Fig. 3).

4.2.2. Comparison with hydrogen production from natural gas Nowadays hydrogen is almost exclusively produced by steam reforming of natural gas. In this process 1 kg of hydrogen is produced from about 4.2 kg of natural gas and 8.9 kg of water. Also 9.3 kg of carbon dioxide are released into the atmosphere. While the worldwide availability of natural gas is relatively high in comparison to other fossil resources like crude oil, hydrogen produced from NG is not sustainable for obvious reasons. Therefore plans exist to produce methane from hydrogen and carbon dioxide via the Sabatier reaction and to feed it into the pipeline distribution grid for natural gas. This so-called Renewable Power Methane (RPM) could cost approx. 8 Ct./kWh (electricity prices of 3e5 Ct./kWh assumed) in the future [41]. Table 11 shows the costs of hydrogen production via steam methane reforming based on the actual very low gas price of 3.25 Ct./kWh for natural gas (industry price Germany 2008e2009 [42]) and 8 Ct./kWh for Renewable Power Methane respectively. Data for capital costs and process parameters are derived from Linde Engineering [43].

Fig. 3 e Cost contributions for hydrogen produced in Iceland and imported to Germany via LOHC transport.

18129

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

Table 11 e Cost assessment for the production of hydrogen via steam methane reforming. Description Cost assessment

Assumptions

Results

Hydrogen output Capital costs Depreciation period Interest rate Annuity factor Natural gas price Natural gas consumption CO2 certificate price Operating hours per year Operating expenses Electric energy consumption Depreciation cost Natural gas purchasing costs Electricity costs CO2 emission cost Operating expenses Total costs

The comparison shows that the import of hydrogen produced from renewable energy in Iceland or other countries that offer high renewable potential (e.g. Greece, Spain, etc.) via hydrogen that is stored in LOHC could be an economically feasible option even in times where natural gas is extremely cheap.

4.3.

Unit

Grid connection of off-shore wind farms

Alongside with solar power, wind farms are expected to deliver a significant share of the energy production of the future. While off-shore wind parks are much more expensive in investment and maintenance they deliver more energy more steadily than on-shore parks due to their higher capacity factor. According to the strategy of the German government, 25,000 MW of wind power should be installed until 2030 in the German parts of the North and East Sea [44]. The typical distance for these off-shore wind farms to the shore is around 30e80 km. As a consequence grid connection is a rather complex and capital intensive task. The first German Offshore farm Baltic 1 was connected to the main land via a 150 kV AC line with a total length of 77 km [45]. For the platform BARD 1, a 200 km HVDC connection (150 kV), the first German direct current line for the connection of a wind farm, was built [46]. The installation of off-shore converter stations and underwater sea cables is expensive and technically sophisticated. It is estimated that about 20e30% of the total project costs for a wind farm are related to its grid connection [47]. Having installed the first wind farms some of these companies have given warnings because of the technical and economic challenges. Especially for small-to-medium sized wind farms where the distance to the coast is high, one could also think about an energy transport based on LOHC. The power generated by wind turbines could be used to produce hydrogen that could be stored in LOHC and consequently be collected from ships within a certain time interval. An alternative could be that the loaded LOHC stays in the wind farm and produces electricity in calm times. This would improve the usage of the electrical network. Additionally, the blade heating in winter times can be done via

Steam methane reforming

3

Nm /h million V years

V/kWh kWhNG/kWhHyd V/tons CO2 Hours % of capital cost kWhel/kgHyd V/kg V/kg V/kg V/kg V/kg V/kg

67,000 86 20 6.00% 8.72% 0.0325

0.08 1.35 15 8000 3% 0.005 0.16

1.45

3.56 0.01 0.14 0.05

1.80 V

3.92 V

the exothermal hydrogenation reaction. As Table 12 e based on the existing wind farm Baltic 1 e shows, this facility would produce approx. 200 tons of loaded LOHC material per day. The number and allocation of stations as well as the number and size of ships, etc. would determine the optimal collection strategy. Assuming e.g. a five-day interval, about 1000 tons of LOHC material would have to be stored for example on the platform, which is currently used to house the transformer station. Another option could be to store the LOHC in the foot of the wind wheel. Assuming that the 21 wind mills have an inner diameter of 4 m, a section with the height of 3.8 m would be sufficient for the above mentioned amount of LOHC substance, which could then be collected by much smaller ships than those considered in Section 3.2 of this contribution. Analogous to the argumentation in the considerations about energy transport from Northern Africa, one advantage of this concept is that it would decouple the energy delivery from the intermittency of the energy production of the wind farm. Storage of energy in LOHC could therefore contribute to achieve more stability in the electricity networks.

5.

Discussion of results

Table 13 summarizes the results of the analysis for the different energy transport options regarding cost and energy consumption for CGH2, LH2 and LOHC. These numbers give a first indication that the transport of energy via LOHC could be a promising alternative. Energy

Table 12 e Hydrogenation of LOHC on off-shore platforms. Wind farm Baltic 1 Power output Annually produced energy Efficiency of electrolysis Amount of hydrogen produced per day Amount of LOHC per day

50 185 70% 11 203

MW GWh tons tons

18130

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

Table 13 e Overview over costs and energy consumption of the described storage and transport processes.

Conditioning Costs Energy consumption Transport e Ship Cargo H2 1000 km 5000 km

Transport e Truck Cargo H2 20 km 50 km

CGH2

LH2

LOHC

Compression

Liquefaction

Hydrogenation

[Ct./kg] [kWhx/kWhHyd]

24 3.5%

73 21.0%

24 1.1%

[tons] Costs [Ct./(ton km)] Energy [kWhx/kWhHyd] Costs [Ct./(ton km)] Energy [kWhx/kWhHyd]

Not considered

1050 32.6 5.2% 19.5 26.1%

2400 6.1 2.3% 4.4 11.7%

[kg] Costs [V/(ton km)] Energy [kWhx/kWhHyd] Costs [V/(ton km)] Energy [kWhx/kWhHyd]

500 17.7 0.8% 10.3 2.1%

3500 5.6 0.6% 2.7 0.8%

1500 5.5 0.3% 3.2 0.7%

consumption as well as the costs of transport are rather low in comparison with LH2 and CGH2 delivery. In particular, longdistance transport via ship seems promising due to the low cost level and the possibility to use existing fleets of product tankers. Furthermore these tankers serve as a storage for a large amount of energy. Regarding the costs for the provision of hydrogen Fig. 4 summarizes the scenarios that were considered in this publication. It shows that it could be a technologically and economically feasible option to produce hydrogen in countries which have a high potential for renewable energy and to subsequently transport it to Europe via LOHC. Other than hydrogen production via steam methane reforming no green house gases are emitted into the atmosphere and no limited fossil feed stock is being consumed. Besides these figures there are further arguments that are in favor of transporting at least a share of the produced energy in form of a chemical carrier like LOHC:  Required capital investment Most of the energy transport concepts that are discussed for the transformation of our energy system from fossil fuels to renewable energy require the installation of a completely

new infrastructure. As a consequence the capital investment that is necessary for implementation is enormous. Two examples shall be given: The ‘desertec initiative’ estimates the total investment for the HVDC transmission lines from Northern Africa to Europe with 5 billion V until 2020 (capacity of 60 TWh/year) and even 45 billion V until 2050 (capacity of 700 TWh/year). Our actual experience with the existing pipeline grid for natural gas can give a valuable hint regarding the investment costs for a comprehensive hydrogen pipeline system. The International Energy Agency IEA estimates that the investment cost of the American NG pipeline system (1.74 million km distribution pipeline and 1.05 million km service pipelines, serving 55 million costumers) is 230 billion USD. Assuming transport of the same amount of energy (14 ExaJoule per year) via hydrogen the potential investment would double to 460 billion USD [48]. On a worldwide basis the global gas use of 55 Exajoule would correspond to investment into a hydrogen pipeline network being worth approx. 2:5  1012 USD. It is clear that these huge investments could slow down e if not even rule out e the transition to alternative energy sources. Mutual causation of missing infrastructure and missing customer, often referred to as the ‘heneegg-problem’ further inhibit necessary investments.

Fig. 4 e Overview of hydrogen provision costs for the considered scenarios.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

A technology like LOHC that can build upon the existing infrastructure would allow for a steady transition without exorbitant investments. Regarding for example the transfer of solar power from Northern Africa to Europe, already existing ships e preferably product tankers e could be used for LOHC transport and only the on-shore infrastructure (e.g. for LOHC loading and unloading in ports) as well as hydrogenation equipment would need to be modified/installed. Instead of a very large one-time investment an incremental introduction of the LOHC technology would be possible. The same considerations hold true for other means of transportation e e.g. truck transport, modification of gasoline stations, train carriers, etc., as well.  Vulnerability of the energy system One of the most important aspects of energy supply is the security and reliability of the system. A lot of people see the transition to renewable energy as a possibility to improve the self-supply with energy for some countries which are today heavily dependent on exports of fossil energy carriers. The increase of the share of energy that is electric e and therefore transmitted via electric transmission lines e raises the vulnerability of the system. This is also a common criticism in disfavor of the Desertec concept: with only a few transmission lines covering up to 15% of Europe’s energy demand in 2050, there would be a certain risk of breakdown of supply due to technical problems, political reasons or even criminal activities. Especially with regard to the fact that the political systems of some of the countries concerned for the energy transfer are rather fragile. In contrast, it is very unlikely that the fragmented delivery of a chemical energy transport by dozens of individual ships navigating in the open sea (and thus outside of the influence of individual states) can be completely cut off. The character of segmentation that comes with the LOHC concept therefore increases the security of supply and reduces the system’s vulnerability to technical dysfunctions, sabotage or political arbitrariness.  Storage capability Today’s energy supply system which is mainly based on fossil fuels is intrinsically robust because every step of transportation is time-lagged and therefore acts as an energy storage capacity. As a consequence energy production must not match energy demand perfectly at all times but only in average. Obviously, this important point does not hold for instant electrical transmission where electricity generation and consumption must be equivalent at all times. Consequently, the storage of energy is an important issue for the stability of the global energy systems. Energy vectors that per se allow for a temporary storage of energy can contribute to the stability of energy supply. Solar power from Northern Africa that is transported to Europe via LOHC could even help to compensate the fluctuations generated by the intermittent producing renewable sources that are installed within Europe. It is noteworthy in this context, that a purely electrical transmission of energy from North Africa to Europe cannot fulfill the same task. Due to the fact that North-Africa

18131

has the same dayenight-cycle as Europe it can be expected that most of the electric energy production there will coincide with the peak in solar energy production in Europe.

6.

Conclusion

The transition toward renewable energies that we currently witness is a huge challenge. Renewable sources like wind or sun can only substitute fossil or nuclear fueled electricity generation if their intermittent, fluctuating character can be compensated for example by adequate storage concepts. It consequently seems very likely today that we will see a multitude of intertwined technologies and concepts within our energy system in the future e small and mid-sized domestic energy production as well as energy imports e.g. from Northern Africa or Iceland, electrical transmission as well as chemical storage of energy and hydrogen cars as well as battery-electric vehicles.

references

[1] 2010 renewable energy data book. U.S. Department of Energy; 2011. [2] Birol F. World energy outlook 2010. Paris: IEA (International Energy Agency); 2010. [3] Sartbaeva A, Kuznetsov VL, Wells SA, Edwards PP. Hydrogen Nexus in a sustainable energy future. Energy Environ Sci 2008;1:79e85. [4] Eberle U, Felderhoff M, Schu¨th F. Chemische und physikalische Lo¨sungen fu¨r die Speicherung von Wasserstoff. Angew Chem 2009;121:6732e57. [5] Mu¨ller B, Mu¨ller K, Teichmann D, Arlt W. Energiespeicherung mittels Methan und energietragenden Stoffen e ein thermodynamischer Vergleich. Chem Ing Tech 2011;83:2002. [6] Scherer G, Newson E, Wokaun A. Economic analysis of the seasonal storage of electricity with liquid organic hydrides. Int J Hydrogen Energy 1999;24:1157e69. [7] Teichmann D, Arlt W, Wasserscheid P, Freymann R. A future energy supply based on Liquid Organic Hydrogen Carriers (LOHC). Energy Environ Sci 2011;4:2767e73. [8] Luo W, Campbell PG, Zakharov LN, Liu S. A singlecomponent liquid-phase hydrogen storage material. J Am Chem Soc 2011;133:19326. [9] Pradhan AU, Shukla A, Pande JV, Karmakar S, Biniwale RB. A feasibility analysis of hydrogen delivery system using liquid organic hydrides. Int J Hydrogen Energy 2011;36:680e8. [10] Mu¨ller K, Stark K, Mu¨ller B, Arlt W. Amine borane based hydrogen carriers: An evaluation. Energy Fuels 2012;226(6): 3691e6. [11] Pez G, Scott A, Cooper A, Cheng H, Bagzis L, Appleby J. Patent ‘Hydrogen storage by reversible hydrogenation of piconjugated substrates’. International publication number WO 2005/000457A2; 2005. [12] Sobota M, Nikiforidis I, Amende M, Sanmartin Zanon B, Staudt T, Ho¨fert O, et al. Dehydrogenation of DodecahydroN-ethylcarbazole on Pd/Al2O3 model catalysts. Chem Eur J 2011;17:11542. [13] Sotoodeh F, Huber BJM, Smith KJ. Dehydrogenation kinetics and catalysis of organic heteroaromatics for hydrogen storage. Int J Hydrogen Energy 2012;37:2715e22. [14] Eblagon KM, Rentsch D, Friedrichs O, Remhof A, Zu¨ttel A, Ramirez-Cuesta AJ, et al. Hydrogenation of 9-ethylcarbazole

18132

[15] [16]

[17] [18]

[19] [20] [21]

[22] [23]

[24]

[25] [26]

[27] [28] [29] [30] [31]

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 7 ( 2 0 1 2 ) 1 8 1 1 8 e1 8 1 3 2

as a prototype of a liquid hydrogen carrier. Int J Hydrogen Energy 2010;35:11609e21. Clariant SE. Safety data sheet in accordance with EUregulation Nr. Nr. 453/2010 regarding N-ethylcarbazole; 2011. Teichmann D, Arlt W, Wasserscheid P. Stabile Energieversorgung trotz unsteter Erzeugung. Solarzeitalter I 2012. Desertec Foundation. Clean power from deserts e whitebook; 2009. Concawe, EUCar, JRC. Well-to-wheels-analysis of future automotive fuels and powertrains in the European context. Tank-to-Wheel report; 2007. Bunkerworld. TWI special report: overview; 2002 to 2010. Hydrocarbon processing, refining processes handbook; 2010. Yamaguchi N. Hydrodesulfurization technologies and costs. Presented at ‘The William and Flora Hewlett Foundation Sulfur Workshop’, Mexico City; 2003. Uhde Thyssen-Krupp. Project brochure installation of HDA/ HDS unit in Bourgas, Bulgaria; 2010. Ahluwalia RK, Hua TQ, Peng J-K, Kromer M, Lasher S, McKenney K, et al. TIAX-report ‘Technical assessment of organic liquid carrier hydrogen storage systems for automotive applications’; 2011. Toseland B, Tyndall D, Varma V. DOE hydrogen program FY2007 annual progress report, III.C.1 reversible liquid carriers for an integrated production, storage and delivery of hydrogen; 2007. Simbeck D, Chang E. NREL report ‘Hydrogen supply: estimate for hydrogen pathways e scoping analysis’; 2002. Altmann M, Gaus S, Landinger H, Stiller C, Wurster R. Final report ‘GEO-Studie: Wasserstofferzeugung in OffshoreWindparks’. Ludwig-Bo¨lkow Systemtechnik; 2001. United Nations conference on trade and development, review of maritime transport; 2006. MAN Diesel & Turbo, Propulsion trends in Tankers. Morgan JP. Asia Pacific equity research paper: ‘Asia Pacific Dry Bulk Shipping’; 2008. Prieur A, Favreau D, Vinot S. Roads2Hy, well to tank e technology pathways and carbon balance; 2009. Valentin B. Wirtschaftlichkeitsbetrachtung einer Wasserstoffinfrastruktur fu¨r Kraftfahrzeuge. University of Applied Sciences Munich; 2001.

[32] Nexant Inc.. Interim report ‘H2A hydrogen delivery infrastructure analysis models and conventional pathway options analysis results’; 2008. [33] Final Report. Trans-Mediterranean interconnection for concentrating solar power. German Aerospace Center; 2006. [34] Stiller C, Svensson AM, Moller-Holst S, Bu¨nger U, Espegren KA, Holm OB, et al. Options for CO2-lean hydrogen export from Norway to Germany. Energy 2008;33:1623e33. [35] Wietschel M, Hasenauer U. Feasibility of hydrogen corridors between the EU and neighbouring countries. Renew Energy 2007;32:2129e46. [36] Kaske G, Schmidt P, Kanngießer KW. Comparison between high-voltage direct current transmission and hydrogen transport. Int J Hydrogen Energy 1991;16:105e14. [37] Smolinka T, Gu¨nther M, Garche J. Stand und Entwicklungspotential der Wasserelektrolyse zur Herstellung von Wasserstoff aus regenerativen Quellen. NOW-Studie 2010. [38] Bardadottir H, Sturludottir LK. Energy in Iceland. 2nd ed.; 2006. [39] Eggertson H, Thorsteinsson I, Ketilsson J, Loftsdottir A. Energy statistics in Iceland 2010; 2010. [40] Arnarson H. Iceland 2025: diverse industries, submarine cable and green revenue. Landsvirkjun Annual Meeting; 2011. [41] Sterner Michael. Bioenergy and renewable power methane in integrated 100% renewable energy systems. In: Schmid Ju¨rgen, editor. Renewable energies and energy efficiency; 2009. [42] German Federal Ministry of Economics and Technology. Energiedaten; 2011. [43] Linde Engineering. Product presentation about steam methane reforming. Personal correspondence; 2010. [44] Raumordnungsplan fu¨r die deutsche ausschließliche Wirtschaftszone; 2009. [45] Kamm V. Company information: ‘Erneuerbare Energien nutzbar machen-der Netzanschluss des Windparks EnBW Baltic 1 in der Ostsee’. 50Hertz Offshore GmbH; 2010. [46] Transpower Stromu¨bertragungs GmbH. Company information ‘Chancen der Energiegewinnung aus OffshoreAnlagen sowie deren Bau’; 2009. [47] German Energy Agency. Information ‘Offshore-Netze: Die Netzanbindung der Offshore-Windparks’; 2009. [48] International Energy Agency. Energy technology analysis: prospects for hydrogen and fuel cells; 2005.