Thermodynamic Evaluation of Potential Organic Hydrogen Carriers

Organic hydrogen carriers show a great potential as future energy storage systems. Limitations of the dehydrogenation step due to thermodynamic limitations are one of the major challenges. Based on the reaction equilibria for some example compounds the feasibility of certain substance classes is discussed. Hydrogen storage is a technology of major interest for the energy systems of the future. Fluctuations in power production by renewable energy technologies necessitate the storage of energy. Because most storage technologies suffer from low storage densities and capacities as well as high costs, chemical storage by using hydrogen is meaningful.1, 2 Hydrogen is currently produced mainly by conversion of fossil fuels (steam reforming). In the future the production of hydrogen through the electrolysis of water, using electricity from wind or solar power, is likely to contribute substantially to the total hydrogen production volume. Furthermore, hydrogen could be produced via other renewable sources of energy, such as biomass3 or photocatalytic water splitting.4, 5 A number of issues have to be addressed in order to store hydrogen. For elemental hydrogen high pressures or very low temperatures must be applied. This causes additional energy demands and consequently a drop in overall efficiency. The energy demand of liquefaction consumes about 30 % of the lower heating value of hydrogen. The energy demand of compressing hydrogen is lower (about 12 % for 350 bar; about 15 % for 700 bar) and the overall efficiency for compressed hydrogen is thus higher.6 Furthermore, for liquid hydrogen significant amounts are lost by evaporation during storage. There again, the storage densities of compressed hydrogen are lower than for liquid hydrogen. To overcome the limitations associated with elemental hydrogen, its conversion into another form is advisable. Three basic types of conversion exist: Irreversible chemical conversion into another substance Reversible storage on the surface of another substance, by means of physical bonds Reversible storage in another substance, by means of chemical bonds For the first case hydrogen can be converted into for example, methane7 or methanol.8 These fuels can be burnt when energy is needed. Chemical conversion is an interesting option but suffers from certain drawbacks, such as the need of a clean and concentrated carbon dioxide stream and the low overall efficiency.9 In reversible storage options hydrogen is bonded to a carrier material. This bonding can be physical or chemical in nature. Physical bonding means that the hydrogen is stored in or on the material without the formation of a new chemical compound. Storage inside a material is limited by the low solubility of hydrogen. Storage on the surface of a material by adsorption can be promising if high-surface-area materials (e.g., zeolites) are used. However, storage by adsorption still requires elevated pressures and the handling of the solid materials is difficult. The loading is always exothermal, which may cause further safety issues. A number of Review articles on hydrogen storage have been published in recent years (e.g., Refs, 6, 10–12). Crabtree22 presented an analysis of nitrogen-containing heterocyclic compounds as hydrogen carriers in 2008. Therein he described how nitrogen atoms can have a positive effect on the thermodynamic properties concerning their use as hydrogen carriers. Especially substitution with nitrogen in five-membered rings can be advantageous. Recently Crabtree et al. presented another review on potential liquid organic hydrogen carriers (LOHCs) that addressed their dehydrogenation potential in direct organic fuel cells,23 with assessments based on DFT-based computational methods. The aim of this contribution is to give a short thermodynamic overview over organic compounds as hydrogen carriers, with a focus on finding carriers that can be dehydrogenated under moderate conditions, and to show further criteria for the assessment of potential organic hydrogen carriers. The catalysis that is necessary for the (de)hydrogenation is not the topic of this work. Organic hydrogen carriers should meet certain requirements to be appropriate for technical use: The substance should be nontoxic, and should not cause other safety issues. Prices should be low. Storage density should be high. The substance should be temperature-stable. The substance should be easy to handle (i.e., preferably liquid). Hydrogenation as well as dehydrogenation can be carried out at reasonable technical conditions. For safety reasons as well as for the ease of handling it is important that the material can be stored at ambient pressure. Gaseous compounds are therefore not preferred. To meet criteria 1 and 5, the vapor pressure should be low (preferably below 100 mbar). Compounds that are absolutely nontoxic would be preferable, but substances with a toxicity lower than or similar to established fuels (gasoline, diesel) would be acceptable (e.g., octane, with LD50,rat,oral=1297 mg kg−1 24). Concerning the price (criterion 2), one has to keep in mind that the carrier material can be recharged and higher prices than for fuels that are combusted are thus acceptable. The aim should be that the overall costs per unit of energy stored should not exceed those of the benchmark technology (at the moment: fossil fuels). A common goal for storage density (criterion 3) is the DoE target of 5.5 wt % hydrogen, which is discussed in the section on manageability.25 The maximum temperature for criterion 4 is linked to the maximum temperature of the respective process. This temperature is strongly influenced by the thermodynamics of the reaction (see below). For criterion 5 a low melting point (below 0 °C) would be desirable (in addition to the low vapor pressure mentioned above). The feasibility of hydrogenation and dehydrogenation (criterion 6) is strongly influenced by the reaction thermodynamics and will be discussed in the following. Another important requirement is that both, hydrogenation as well as dehydrogenation, can be carried out at reasonable technical conditions. A substance, such as ammonia borane21, 26 that can release hydrogen but is difficult to regenerate is only suited for disposable use. Nevertheless, works are progressing to recycle ammonia borane.27 Thermodynamically the dehydrogenation is as complicated as the respective hydrogenation is easy, since it has the same free energy of reaction ΔRg with a different algebraic sign. Thus, none of the reactions should be favored too strongly by thermodynamics (ΔRg≈0). Most hydrogenation reactions are favored by the reaction equilibrium, even though hydrogenation is unfavorable by entropy (ΔRs<0). The change in entropy in an isothermal hydrogenation of an unsaturated compound at 298.15 K and 1 bar is between −130 and −110 J K−1 (mol H2)−1. Consequently the free energy of reaction ΔRg at standard conditions is about 36 kJ (mol H2)−1 higher than the enthalpy of reaction ΔRh. Because hydrogenation reactions are usually highly exothermic ΔRh=−68.73 kJ (mol H2)−1 in a benzene ring, they are thermodynamically favorable at room temperature, even though they are not favored by entropy. Due to the exothermic nature of the reaction the equilibrium can be shifted towards dehydrogenation by increasing the temperature. It is therefore desirable to find chemical compounds that can be hydrogenated with an enthalpy of reaction equal to or slightly lower than −40 kJ (mol H2)−1, to allow for an easy (de)hydrogenation. If the enthalpy of reaction is higher than this value hydrogenation is barely feasible, because temperatures lower than 20 °C would be required to overcome thermodynamic limitation, making reaction kinetics unacceptably slow. On the other hand temperatures for dehydrogenation would be very high if the reaction enthalpy of the hydrogenation is significantly lower than −40 kJ (mol H2)−1. A huge number of potential organic hydrogen carriers exist. In this contribution we selected a group of compounds that represent the major groups of unsaturated compounds that could be used as hydrogen carriers. The dehydrogenated forms of these substances are shown in Figure 1. Compounds under consideration. The basic data for the evaluation of these compounds as organic hydrogen carriers are given in Table 1. The data for the enthalpy and free energy of the compounds in their hydrogenated and dehydrogenated forms were taken from the DIPPR database.28 The data for the enthalpy of compound 17 (N-ethylcarbazole) were taken from Ref. 29. The group contribution method suggested by Benson and Russ30 was used for the hydrogenated forms of compounds 7 and 8 (decalin, tetradecahydroantracene) and the dehydrogenated form of compound 11 (4H-pyran), because no experimental data were available. Group contributions for the enthalpy and free energy of reaction for the compounds 13 and 14 were missing. Thus, DFT calculations on the B3LYP/def2-TZVPP31, 32 level using Turbomole (Version 6.2, COSMOlogic, Leverkusen, Germany) were used in these cases. Entry Hydrogen carrier (dehydrogenated form) CAS registry number Max. hydrogenation per molecule ΔRg kJ (mol H2)−1 ΔRh kJ (mol H2)−1 Gravimetric storage density wt % (carrier only) 1 ethene 74-85-1 1 −100.4 −136.3 6.7 2 1,3-butadiene 106-99-0 2 −83.3 −118.4 6.9 3 benzene 71-43-2 3 −32.6 −68.7 7.1 4 toluene 108-88-3 3 −31.6 −68.3 6.1 5 styrene 100-42-5 4 −43.6 −79.7 7.1 6 phenol 108-95-2 3 −25.6 −63.3 6.0 7 naphthalene 91-20-3 5 −20.5 −66.3 7.2 8 anthracene 120-12-7 7 −19.6 −66.9 7.3 9 1,3,5,7-cyclooctatetraene 629-20-9 4 −69.9 −105.9 7.1 10 furan 110-00-9 2 −40.3 −74.7 5.5 11 4H-pyran 289-65-6 2 −64.9 −104.8 4.6 12 thiophene 110-02-1 2 −40.4 −74.6 4.5 13 pyridine 110-86-1 3 −29.5 −63.1 7.0 14 pyrrole 109-97-7 2 −22.8 −55.9 5.6 15 N-ethylcarbazole 86-28-2 6 −18.2 −53.2 5.8 16 1,2-dihydro-1,2-azaborine 6680-69-9 3 −2.7 −35.9 7.1 17 1-tert-butyl-2-hydro-1,2-azaborine 3 −2.9 −37.9 4.3 It is striking that the hydrogenation of ethene to ethane (compound 1) is highly favorable by thermodynamics and, consequently, dehydrogenation is hardly feasible. The same holds for other systems containing isolated double bonds. In systems with conjugated double bonds (compounds 2, 9) thermodynamics are slightly better for hydrogen storage purposes but the driving force of dehydrogenation is still too low. If the conjugated double bonds are situated within an aromatic ring (compounds 3–8) dehydrogenation is more favorable, due to the very stable aromatic system. However, dehydrogenation of additional double bonds outside the aromatic ring, even if they are conjugated with the ring, is difficult (compound 5). If ethylcyclohexane is dehydrogenated, the formation of ethylbenzene is relatively easy, while the total dehydrogenation to form styrene is highly limited by thermodynamics. Heteroatoms within a ring can change the thermodynamic driving force. Substitution with oxygen (compounds 10, 11) or sulfur (compound 12) has a negative effect on the dehydrogenation characteristics (especially if the aromaticity is disrupted; compare compound 11). Substitution with nitrogen on the other hand can be advantageous (compounds 14, 15). If boron is further added to the nitrogen-containing heterocyclic compounds (compounds 16, 17) the thermodynamics can be shifted further in favor of dehydrogenation. However, synthesis of these azaborines is complicated and their long-term stability is doubtful. The equilibrium yields for the dehydrogenation of some organic hydrogen carriers at 1 bar are shown exemplarily in Figure 2. As expected dehydrogenation of an alkane to form an alkene (here: ethene) necessitates high temperatures and even then is hardly possible. The equilibrium of the dehydrogenation to form an aromatic compound (here: benzene) allows hydrogen production at far lower temperatures, but still a minimum of 260 °C is needed to reach an equilibrium conversion of 50 %. For a nitrogen-containing compound (here: pyrrole) the minimum temperature can be shifted to lower values (less than 200 °C for a conversion of 50 % at 1 bar). When using azaborines (here: 1,2-dihydro-1,2-azaborine) dehydrogenation would be possible in terms of thermodynamics at very mild conditions (less than 50 °C for a conversion of 50 % at 1 bar). However, it has to be doubted that total dehydrogenation of azaborines is possible at such mild conditions or if only the boron and nitrogen atoms are dehydrogenated, while dehydrogenation of the carbon atoms only occurs at higher temperatures. Dehydrogenation equilibrium of selected compounds at 1 bar. —: ethene (1), - - - -: benzene (3), •–•–: pyrrole (14), ••••: azaborine (16). Of course, in all cases conversion and thus temperature could be increased by reducing the partial pressure (e.g., adding an inert gas or lowering the system pressure). However, it is not advisable to do so, because the produced hydrogen is either diluted or additional energy input is necessary. Hydrogenation of all compounds is rather easy regarding the reaction equilibrium. The temperatures for 50 % conversion at 1 bar are the same as for dehydrogenation. In any event, keeping the temperatures below these values is not mandatory for hydrogenation because the equilibrium can be shifted to high conversion at higher temperatures by applying higher pressures. The equilibrium yields shown in the diagram agree with experimental results given in the literature (e.g., Refs. 33, 34). The yields given here are always slightly higher than the yields of the dehydrogenation experiments in the respective publications, because practical conversions can never fully reach the thermodynamic maximum. The manageability of a hydrogen carrier is predominantly determined by the storage density and the temperature range in which it stays liquid. The target of the U.S. Department of Energy for 2015 is a gravimetric storage density of 5.5 wt % of hydrogen relative to the storage system, not solely the carrier (for details on these targets, see Ref. 25). If no solvent is used all mentioned compounds except 11, 12, and 17 would meet this requirement, if the tank and dehydrogenation unit are neglected. The maximum storage density that could be reached by using organic hydrogen carriers is about 7.3 wt % (aromatic hydrocarbons without any further functional groups). However, the systems storage density is lowered by the weight of the tank and the dehydrogenation unit. When taking these units into account it is likely that many potential LOHCs would fail this target, making this approach most suited for stationary energy storage rather than mobile applications. To avoid the use of solvents (and therefore keep high storage densities) it is desirable that the compounds have a low melting point. Many of the mentioned compounds melt at sufficiently low temperatures, but especially larger molecules solidify even at room temperature and above (e.g., compounds 7, 8 or 15). To hinder crystallization alkyl groups could be added, which destroy the symmetry of the molecule. This is illustrated by the hydrogenated forms of 3 and 4: While cyclohexane has a melting point of 6.5 °C, its methylated derivative methylcyclohexane has a melting point of −126.6 °C.28 Adding side chains, however, decreases the gravimetric storage density. Besides the melting point another relevant parameter is the boiling point. The boiling point should be as high as possible to minimize evaporation during dehydrogenation and storage. Larger molecules are advantageous in this regard. Side chains can further decrease the vapor pressure. The boiling point of the rather small molecule toluene (4) is only 110.6 °C,28 which is far below the temperature necessary for dehydrogenation without additional inert gas. The larger naphthalene (7) has a boiling point of 218.0 °C, which would allow at least partial dehydrogenation in the liquid phase, but still evaporation would be significant. For the even larger anthracene (8) the boiling point is 342.0 °C and thus sufficiently high. These criteria can be applied exemplarily to the compounds shown in Figure 2 to allow for a further rating beyond thermodynamic feasibility. It is obvious that the usability of ethene as a hydrogen carrier is not only limited by the equilibrium, because ethene as well as its hydrogenated form (ethane) are gases at ambient conditions, which complicates storage as well as separation of hydrogen and carrier after dehydrogenation. Benzene and its dehydrogenated form (cyclohexane) have melting points slightly above 0 °C, which makes them usable but can cause problems in winter time. Pyrrole and its hydrogenated form (pyrrolidine) melt significantly lower (−23.4 °C and −57.8 °C, respectively). Nevertheless, the low boiling points (130 °C and 86.6 °C, respectively) can cause problems during dehydrogenation.28 The melting point of the azaborine in its dehydrogenated form would be sufficiently low (about −45 °C), but the hydrogenated form solidifies at about +65 °C and is therefore problematic for practical use. In conclusion, hydrogen can be stored with high storage densities by using organic hydrogen carriers, preferably the so-called liquid organic hydrogen carriers (LOHC). In principle each unsaturated compound could be used for this purpose, but the suitability of different substances strongly varies. Regeneration of the spent form (hydrogenation) is not limited thermodynamically, but the release of hydrogen (dehydrogenation) can be an obstacle. It could be demonstrated that an ideal LOHC should have an enthalpy of reaction of about 40 kJ (mol H2)−1. A new carrier material should therefore preferably be hydrogenated at a small exothermicity. Nevertheless, heat tone should not go significantly below 40 kJ (mol H2)−1 to avoid thermodynamic limitations in the regeneration step. Nonaromatic compounds are barely able to meet this requirement. Especially nitrogen-containing aromatic compounds exhibit enthalpy changes during hydrogenation that make them very well suited for hydrogen storage. free energy of reaction enthalpy of reaction entropy of reaction The authors thank Prof. Peter Wasserscheid for valuable discussions.