SEARCH

SEARCH BY CITATION

Keywords:

  • dehydrogenation;
  • energy conversion;
  • hydrocarbons;
  • hydrogen storage;
  • sustainable chemistry

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusions
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

Liquid organic hydrogen carrier (LOHC) systems offer a very attractive method for the decentralized storage of renewable excess energy. In this contribution, industrially well-established heat-transfer oils (typically sold under trade names, e.g., Marlotherm) are proposed as a new class of LOHC systems. It is demonstrated that the liquid mixture of isomeric dibenzyltoluenes (m.p. −39 to −34 °C, b.p. 390 °C) can be readily hydrogenated to the corresponding mixture of perhydrogenated analogues by binding 6.2 wt % of H2. The liquid H2-rich form can be stored and transported similarly to diesel fuel. It readily undergoes catalytic dehydrogenation at temperatures above 260 °C, which proves its applicability as a reversible H2 carrier. The presented LOHC systems are further characterized by their excellent technical availability at comparably low prices, full registration of the H2-lean forms, and excellent thermal stabilities.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusions
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

In recent years the energy systems of many countries have been profoundly changed. Many of these changes have been initiated because of environmental considerations. The introduction of emission trading to reduce greenhouse gas emissions and the nuclear phase-out in Germany after the Fukushima disaster in 2011 are only the most prominent examples. Although political debate has led to quite different concepts for future energy systems in different countries, there is a general consensus that electricity generation from renewable energy sources will become considerably more important. For example, 61 countries have claimed in the Renewable Energy Policy Network for the 21st Century (REN) that they intend to reduce their dependence on imported primary energy sources and their impact on climate change by increasing the share of renewable energy in their electricity production.1

Unfortunately, the production of renewable energy, for example, by photovoltaic or wind energy, is highly intermittent in character and depends largely on unforeseeable and uncontrollable meteorological factors. Therefore, any energy system that is based to a larger extent on such intermittent producers requires energy storage capacities to adapt the stochastic production profile to the specific energy demand. Large-scale energy storage becomes absolutely crucial when it comes to very high shares of intermittent energy sources. For example, the energy plan of the German government aims for an 80 % share of renewable energy in national electricity production in 2050.2 It is clear that such ambitious targets can only be realized if massive capacities of regenerative energy production are installed. These capacities will naturally cause energy overproduction at times and at locations of favorable weather conditions. This excess production must be efficiently stored and transported to compensate for energy shortages under unfavorable weather conditions. The required type of energy storage systems must not only manage the natural day–night cycle for photovoltaic energy production but also seasonal differences both for wind and solar power. Ideally, these energy storage systems are also suitable to transport energy in existing infrastructures (e.g., tanker ships, pipelines) from regions in the world privileged for the production of renewable energy (e.g., hydroenergy from Canada and Iceland or solar power from the African deserts) to highly industrialized regions with a very high energy demand. For Germany, it has been calculated that a massive need for such long-term energy storage systems will arise from renewable energy shares of around 40 % onwards.3 The total storage demand for the year 2020 has been estimated for Germany to be up to 4 TWh, which is a factor of 100 more than the available (and geographically limited) hydroelectric storage facilities in Germany.4

Most chemical energy storage technologies that are actually discussed in the literature consider the conversion of excess electricity into H2 by water electrolysis as the first step. This H2 is subsequently bound to a H2-lean molecule in a catalytic hydrogenation reaction. If this H2-lean molecule is N2, the corresponding H2-rich form is ammonia.5 If the H2-lean molecule is CO2, the corresponding H2-rich forms are either formic acid,6 methane,7 methanol,8 or Fischer–Tropsch products,9 which depends on the applied catalyst and process parameters during the hydrogenation process. If the H2-lean molecule is N-ethylcarbazole (NEC), the resulting H2-rich form is dodecahydro-N-ethylcarbazole (H12-NEC).10 All of these H2-rich molecules can be stored for extended periods of time without energy loss and transported over long distances using established energy transport logistics (e.g., pipelines, ships, trucks). At times and at places of energy demand, H2 can be released in a catalytic dehydrogenation reaction (Scheme 1).

thumbnail image

Scheme 1. Schematic representation of H2-based concepts of chemical energy storage and transport using existing distribution infrastructure.

Download figure to PowerPoint

Among these systems only the NEC/H12-NEC pair is an energy-carrying system (ETS) according to the definition of reference,10a which allows a closed energy cycle without binding or releasing other substances from the atmosphere than water and O2. In contrast, systems that use gaseous substances as energy-lean molecules (e.g., N2 or CO2) release mixtures of H2 and H2-lean gas during the dehydrogenation reaction, whereas in the case of liquid organic hydrogen carrier (LOHC) systems (of which the NEC/H12-NEC pair is a prominent example) pure H2 is obtained (after appropriate condensation of the high-boiling H2-lean carrier molecule). Consequently, LOHC systems do not require the isolation of CO2 or N2 from air or exhaust gas streams in appropriate quality for the H2 storage process, an important aspect for successful application especially in decentralized scenarios. Decentralized energy storage is an intuitive and logical choice given that most installations to produce energy from wind or sun are decentralized and that grid expansion is both expensive and not well accepted by the public, at least in Germany. Another even more relevant argument arises from the fact that decentralized energy storage allows the utilization of unavoidable heat loss during the storage process in an effective storage–heat coupling (typical efficiencies of electrolyzers are 70 %, typical efficiencies of fuel cells are 55 %). Detailed energetic and economic calculations for such a local storage–heat combination that demonstrate the high potential of this approach have been recently published by some of us.11

Despite the growing interest in the application of LOHC systems for decentralized energy storage, all of the major research activities in Northern America12 and Europe13, 14 that have been conducted so far have focused on the development of LOHC systems for automotive applications. In these projects, the LOHC system NEC/H12-NEC has been intensively studied and identified as most promising. The use of NEC/H12-NEC as a reversible H2 carrier was first suggested by the company AirProducts and Chemicals.15, 16 The thermodynamic suitability of N-functionalized alicyclic compounds for efficient dehydrogenation has been recently confirmed by Arlt and coworkers.17

Despite the unquestionable attractiveness of the NEC/H12-NEC pair for dynamic H2 storage, this system is characterized by a number of features that complicate its rapid implementation for real-life storage applications. These drawbacks include:

a) the currently limited technical availability of NEC, as most NEC is obtained from coal tar distillation with an annual production quantity of below 10 000 tons per year;

b) the fact that fully dehydrogenated, pure NEC is a solid at room temperature (m.p: 68 °C), which complicates the technical dehydrogenation process of H12-NEC and limits the useable H2 capacity of the NEC/H12-NEC system under liquid-handling conditions to a maximum of 5.2 wt %;

c) the thermal stability of NEC being limited to some extent by the dealkylation of the compound to carbazole, a reaction that becomes relevant at temperatures above 270 °C in the presence of typical dehydrogenation catalysts (see below for details). However, typical process conditions for H12-NEC dehydrogenation are at 230 °C under which the thermal stability issues of the NEC/H12-NEC system are much less relevant. Nevertheless, the observed upper thermal stability limit restricts to some extent the practical process window for H12-NEC dehydrogenation and prevents higher volumetric productivity by simple temperature increase.

In this contribution we describe alternative LOHC systems that appear to be highly suitable for stationary application in decentralized energy storage as they avoid the drawbacks mentioned above. Our paper focuses on the remarkable finding that mixtures of isomeric benzyltoluenes and dibenzyltoluenes, which are industrially widely used as heat-transfer fluids (e.g., under the trade names Marlotherm, Farolin, and Diphyl), can be readily applied as LOHC systems that show excellent technical availability, high H2 capacities without solidification, very high thermal stability, and full (eco)toxicological characterization of the liquid H2-lean form of the storage system.

Results and Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusions
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

Table 1 shows a detailed comparison of the relevant properties of the technical benzyltoluene mixture (e.g., Marlotherm LH; MLH), the technical dibenzyltoluene mixture (e.g., Marlotherm SH; MSH), and NEC/H12-NEC with respect to their potential application as LOHC systems.

Table 1. Relevant properties of the heat-transfer-oil-derived LOHC systems compared to the standard NEC/H12-NEC system.[18–20]
LOHC systemMSH/H18-MSHMLH/H12-MLHNEC/H12-NEC
structure of H2-lean form
Thumbnail image of
Thumbnail image of
Thumbnail image of
structure of H2-rich form
Thumbnail image of
Thumbnail image of
Thumbnail image of
m.p. of H2-lean form [°C]−39 to −34−3068
normal b.p. of H2-lean form [°C]390280270
H2 capacity [wt %]6.26.25.8
heat of (de)hydrogenation [kJ molinline image−1]717155
hazard symbols of H2-lean formNXi, NXi

It is clear from these data that the proposed LOHC systems offer significant advantages compared to the NEC/H12-NEC system with respect to the melting point and H2 capacity criteria. Regarding toxicity and boiling point, MLH/H12-MLH is comparable to NEC/H12-NEC, whereas MSH/H18-MSH has a significantly higher boiling point of the H2-lean form and a better toxicology profile, in particular with respect to its ecotoxicological impact on aqueous environments.1820 The difference in the heat of reaction between the hydrogenation/dehydrogenation reaction of NEC/H12-NEC versus MLH/H12-MLH and MSH/H18-MSH of 16 kJ molinline image−1 indicates a less favorable thermodynamic driving force for the dehydrogenation of the N-free, alicyclic compounds. Consequently, slightly harsher conditions are expected for the dehydrogenation of H12-MLH and H18-MSH to compensate for this important difference.

Owing to the need to operate the dehydrogenation step at elevated temperatures, we firstly investigated the thermal stability of the H2-lean representatives of the different LOHC systems. This was examined by heating MLH, MSH, and NEC in the presence of a typical dehydrogenation catalyst (5 wt % Pt/Al2O3) for 72 h at 270 °C. A temperature of 270 °C is well above the typical dehydrogenation temperature of 230 °C recommended for H12-NEC dehydrogenation.10a The respective results are displayed in Table 2.

Table 2. Chemical and thermal stability of MSH, MLH, and NEC (represented by the amount of decomposition products formed from the decomposition of MSH and MLH and dealkylation of NEC) under prolonged heating at 270 °C in the presence and absence of a typical dehydrogenation catalyst (10 g LOHC, 0.2 mol % catalyst if applied).
HeatingCatalystAmount of decomposition products [%]
time [h] MSHMLHNEC
24<0.01<0.01<0.01
48<0.01<0.01<0.01
72<0.01<0.01<0.01
24Pt/Al2O3<0.01<0.01<1
48Pt/Al2O3<0.01<0.01<1
72Pt/Al2O3<0.01<0.01<2

None of the tested LOHC systems shows a significant formation of decomposition products in the absence of catalyst. In presence of the catalyst, however, NEC undergoes slow but clearly detectable formation of the dealkylation product carbazole, whereas MLH and MSH remain completely stable within the detection limits of the applied GC–MS analysis. Notably, in practice a typical H2 release step would result in thermal stress times between 0.2 and 4 h. Furthermore, carbazole has been found in our laboratories to be fully hydrogenable and dehydrogenable under conditions similar to NEC/H12-NEC. However, the formation of larger amounts of dealkylated LOHC species is still unfavorable as the latter increase the melting point of the LOHC mixture, which further limits the H2 capacity of the fully liquid LOHC system.

With the thermal stability of MLH and MSH confirmed, even in the presence of a typical dehydrogenation catalyst, we further investigated the catalytic hydrogenation of MLH and MSH. Again, the results were compared to the hydrogenation of the well-established LOHC compound NEC (Table 3).

Table 3. Catalytic hydrogenation of the H2-lean compounds MLH, MSH, and NEC under identical reaction conditions (150 °C, 50 bar H2, 1 h, 0.25 mol % Ru/Al2O3) to the H2-rich compounds H12-MLH, H18-MSH, and H12-NEC, respectively.
LOHC systemDegree of H2 loading[a] [%]Time to full H2 loading [h]
  1. [a] Amount of H2 bound to the H2-lean form divided by the maximum H2 loading after 1 h reaction time.

MSH/H18-MSH454
MLH/H12-MLH>991.5
NEC/H12-NEC703

Under identical hydrogenation conditions, the loading times for all three compounds are quite similar, and MLH is fully converted to the corresponding perhydro compound fastest, followed by NEC and MSH. It is assumed that pore-diffusion effects within the heterogeneous catalyst could account for the observed differences in the hydrogenation kinetics of MLH and MSH. A detailed study that highlights these aspects in more detail will be published elsewhere.

With the next set of experiments, we aimed to demonstrate H2 release from H18-MSH and H12-MLH, respectively. We used the same catalyst system as previously optimized for the dehydrogenation of H12-NEC and we first pursued a direct comparison by applying identical reaction conditions (Table 4 and Figure 1).

Table 4. Catalytic H2 release from H12-MLH, H18-MSH, and H12-NEC under identical reaction conditions in norm liter (NL; 0.1 mol LOHC, 270 °C, 2 h, 0.1 mol % Pt applied as 0.5 wt % Pt on Al2O3, dehydrogenation protocol B).
LOHC systemH2 releaseTotal amount of H2
 after 2 h [NL]degree [%]in 100 % release [N]
MSH/H18-MSH84020.2
MLH/H12-MLH10.17513.5
NEC/H12-NEC13.510013.5
thumbnail image

Figure 1. Catalytic H2 release from H12-MLH, H18-MSH, and H12-NEC under identical reaction conditions (270 °C, 2 h reaction time, 0.1 mol % Pt applied as 0.5 wt % Pt on Al2O3, dehydrogenation protocol B).

Download figure to PowerPoint

From this comparison, it is clear that the dehydrogenation kinetics of H12-MLH and H18-MSH are slower compared to that of H12-NEC under identical conditions. However, it can also be seen that H2 is released from the purely hydrocarbon compounds at 270 °C. Owing to the different chemical nature of the different hydrogenated LOHC systems under investigation, we were interested to see if a different catalyst would be more suitable to promote the dehydrogenation of H12-MLH and H18-MSH. Consequently, catalyst screening experiments were performed that used H18-MSH as the representative H2-rich compound. The experiments were performed with a constant molar amount of precious metal in the experiments (0.15 mol % with respect to MSH). As a result of the different metal loadings in the applied commercial catalysts, the total mass of catalyst was adapted for each experiment. The results are presented in Table 5.

Table 5. Screening of commercial catalysts for the dehydrogenation of H18-MSH (0.1 mol LOHC, 0.15 mol % catalyst, 270 °C, 3.5 h, dehydrogenation protocol A) 20.2 L of liberated H2 corresponds to full H2 release.
EntryCatalystVolume of H2 [NL]Degree of dehydrogenation [%]
1Pt/Al2O3 (0.5 wt %)10.451
2Pt/Al2O3 (5 wt %)8.240
3Pt/C (5 wt %)11.155
4Pt/C (1 wt %)14.471
5Pt/SiO2 (1 wt %)2.010
6Pd/C (5 wt %)3.316
7Pd/Al2O3 (5 wt %)1.68

Pt on C catalysts are more suitable for the dehydrogenation of H18-MSH than Pt on alumina or silica (Table 5, entries 3 and 4 versus 1, 2, and 5). Lower metal loadings resulted in a better H2 release (comparing identical overall amounts of the metal; Table 5, entries 1 versus 2 and 3 versus 4). Pd on different supports seems to be a much less suitable catalyst for the dehydrogenation of H18-MSH (Table 5, entries 6 and 7) compared to Pt-based systems.

In the next set of experiments, we aimed to elucidate the temperature dependence of the H2 release rate from H12-MLH and H18-MSH with the two most promising catalyst systems, namely, Pt on C with 5 and 1 wt % metal loading. The results are shown in Table 6.

Table 6. Variation of the dehydrogenation temperature in the catalytic H2 release from H18-MSH and H12-MLH using Pt on C catalysts with different metal loadings at identical amounts of precious metal.
EntryLOHCCatalystVolume of H2 [NL] (degree of dehydrogenation [%])
   230 °C250 °C270 °C290 °C
  1. [a] Theoretical maximum volume of H2=20.2 NL. [b]  Theoretical maximum volume of H2=13.5 NL. Reaction conditions: 0.1 mol LOHC, 0.15 mol % catalyst, 230–290 °C, 3.5 h, dehydrogenation protocol A.

1H18-MSH[a]Pt/C (5 wt %)2.9 (14)6.7 (33)11.1 (55)16.6 (82)
2H18-MSH[a]Pt/C (1 wt %)3.3 (16)7.3 (36)14.4 (71)19.8 (98)
3H12-MLH[b]Pt/C (5 wt %)0.5 (4)8.4 (62)11.8 (87)
4H12-MLH[b]Pt/C (1 wt %)5.0 (37)10.7 (79)12.9 (96)

These experiments confirm the higher activity of catalysts with a lower degree of metal loading based on a large number of tests. Moreover, the dehydrogenation of H12-MLH and H18-MSH is strongly temperature dependent. To reach a degree of H2 release of more than 95 % in 210 min reaction time, a reaction temperature of 290 °C was required for H18-MSH, whereas 270 °C was sufficient to reach a similar level of H12-MLH dehydrogenation. The dehydrogenation of H12-MLH was not investigated at temperatures above 270 °C as the boiling point of the hydrogenated form was reached at this temperature.

Finally, we were interested in the H2 release kinetics of H18-MSH in direct comparison with H2 production from H12-NEC. In detail, we wanted to identify the temperature at which the rate of H2 release from H18-MSH with the best identified catalyst would match the rate of H2 release from H12-NEC at 230 °C (also with the best catalyst for H12-NEC dehydrogenation). The applied volume of hydrogenated LOHC was adjusted to result in comparable maximum volumes of produced H2, and the same molar ratio of precious metal catalyst were used in this set of experiments. This resulted in a higher mass of applied precious metal in the case of H12-NEC dehydrogenation because of the higher molar weight of H18-MSH than H12-NEC. The results are displayed in Table 7 and Figure 2.

Table 7. Adjustment of comparable H2-release rates from H12-NEC and H18-MSH by temperature variation. The most suitable catalyst systems were applied for each substrate.
EntryLOHCCatalystT [°C]Maximum H2 volume flow [mL min−1]Total H2 release [NL]Degree of H2 release [%]
  1. [a] Theoretical volume of H2 release=6.5 NL. [b] Theoretical volume of H2 release=6.7 NL. Reaction conditions: 10.7 mL LOHC, 0.1 mol % catalyst, dehydrogenation protocol B.

1H12-NEC[a]Pd/Al2O3 (0.5 wt %)2303966.295
2H18-MSH[b]Pt/C (1 wt %)2902256.089
3H18-MSH[b]Pt/C (1 wt %)3103736.697
thumbnail image

Figure 2. Identification of conditions for similar H2 release from H12-NEC and H18-MSH by temperature variation. Substrate volumes have been adjusted to result in similar maximum volumes of H2 release. For each substrate, the most suitable catalyst system was applied according to dehydrogenation protocol B.

Download figure to PowerPoint

The H2 release curves in Figure 2 demonstrate that it is possible to reach the same H2 production rates from H18-MSH as from H12-NEC at 230 °C by using the optimized catalyst systems for the respective reaction. However, a pronounced difference in the dehydrogenation kinetics between H12-NEC and H18-MSH is still apparent, which results mainly from the different thermodynamic driving forces for the dehydrogenation of the different substrates. Thus an increase in the reaction temperature of 80 °C is required in the case of the catalysts applied here to reach a reaction rate identical to that of H12-NEC dehydrogenation. Even at 310 °C, no decomposition products were detected by using GC–MS analysis. With the higher thermal stability of the industrially approved heat-transfer oils MLH and MSH, such high reaction temperatures are feasible and reasonable as long as the efficient heat integration between the energetic H2 utilization at high temperatures (e.g., a H2 combustion engine or solid oxide fuel cell (SOFC) units) and the H2 release unit can be realized.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusions
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

The use of industrially well-established heat-transfer oils (mixtures of aromatic isomers typically sold under trade names, e.g., Marlotherm) as a new class of liquid organic H2 carrier (LOHC) systems has been evaluated. Compared to the most established LOHC system, dodecahydro-N-ethylcarbazole/N-ethylcarbazole (H12-NEC/NEC), the new mixtures provide a 19 % higher H2 capacity (6.2 versus 5.2 wt %) in the liquid state at room temperature. The technical challenge of the solidification of the pure unloaded H2 carrier, which has often been seen as a major drawback of the H12-NEC/NEC system, is effectively solved for the aromatic mixtures discussed here [m.p. NEC: 68 °C versus m.p. Marlotherm SH (MSH): −39 to −34 °C]. It has been demonstrated that MSH and Marlotherm LH (MLH) can be readily and fully hydrogenated using a commercial Ru on alumina catalyst. The hydrogenation rates are faster for MLH compared to the hydrogenation of NEC under comparable reaction conditions. The hydrogenation of MSH is slightly slower than the NEC reference system.

The dehydrogenation of H12-MLH and H18-MSH requires the use of catalysts different than that of H12-NEC. Although for the first two reactions Pt on C catalysts proved to be most efficient, the dehydrogenation of H12-NEC is typically best catalyzed by Pd or Pt on alumina. For thermodynamic reasons, the complete dehydrogenation of H12-MLH and H18-MSH requires a higher temperature and shows a higher reaction enthalpy compared to that of activated H12-NEC, which leads to a poorer energy and exergy balance of the process. However, it is possible to reach the same rate of H2 production from H18-MSH as from H12-NEC if the reaction temperature is 80 °C higher in the first case. H2 release from H12-MLH is consistently faster than that from H18-MSH, and high H2 production dynamics are observed from H12-MLH at 270 °C.

With respect to other physicochemical and practically relevant properties, H12-MLH/MLH and H18-MSH/MSH are closer to diesel than H12-NEC/NEC. This is confirmed by the densities, viscosities, surface tensions, boiling points, flammability, and material compatibility aspects of MLH19 and MSH20 that have been very well documented because of the wide industrial use of these liquids as heat-transfer oils. Interestingly, on comparing (eco)toxicology, MSH is not only more favorable than NEC but clearly less problematic than common diesel, for example, with respect to its toxicity against aqueous environments.

It is fair to state that the new LOHC systems presented here, MLH and MSH, have some structural and mechanistic similarities to the H2 storage systems benzene/cyclohexane,21 toluene/methylcyclohexane,22 and naphthalin/decalin23 presented previously. However, these systems are much less attractive for practical applications as they are characterized by a significantly higher vapor pressure (which increases the purification effort required to obtain fuel-cell-grade H2 significantly) and by problematic (eco)toxicological aspects.

As a result of the very similar physicochemical properties of MLH and MSH in comparison to typical hydrocarbon fuels, it can be expected that today’s infrastructure for fuel distribution (tanker ships, pipelines, storage tanks) can be made fully available for these LOHC systems.24 It is thus anticipated that MLH and MSH could make excellent storage materials for H2 produced from excess renewable energy production in stationary applications. The storage of large amounts of energy over longer times (e.g., to compensate for day–night cycles or to shift an energy surplus from production over the weekend into the working week at industrial production sites) represents a very attractive application scenario. If the heat for dehydrogenation can be provided at the appropriate level from energetic H2 use, the electricity-to-electricity efficiencies of such a storage system can be up to 38 %, in-line with earlier calculations for other LOHC materials. This efficiency can be further increased by using the heat losses of the electricity storage process to heat or cool buildings, for example.

Experimental Section

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusions
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

Materials: MLH and MSH of appropriate quality for catalytic hydrogenation were purchased from Hydrogenious Technologies GmbH, Nürnberg (www.hydrogenious.net). The commercial catalysts used are listed in Table 8.

Table 8. List of commercial catalyst materials applied in this study.
EntryCatalystSupplierLOT No.
1Ru/Al2O3 (5 wt %)Hydrogenious TechnologiesBN22001
2Pd/C (5 wt %)Sigma–AldrichBCBF8206V
3Pt/C (5 wt %)Hydrogenious TechnologiesBM17001
4Pt/C (1 wt %)Hydrogenious TechnologiesBM14001
5Pt/SiO2 (1 wt %)Sigma–AldrichMKBH0371V
6Pt/Al2O3 (0.5 wt %)Hydrogenious TechnologiesBM23001
7Pt/Al2O3 (5 wt %)Sigma–AldrichMKBH3784
8Pd/Al2O3 (5 wt %)Sigma–AldrichMKBH9857

Stability tests: The stability tests were performed in a 100 mL glass flask at 270 °C and 101 kPa. For each experiment, the dehydrogenated LOHC compound (10 g) was added to the reactor and mixed with a precalculated amount of a reduced 5 wt % Pt/Al2O3 catalyst at a ratio of 0.2 mol %, which refers to the metal content of the catalyst. The reactor was purged with N2, and the mixture was stirred and heated to the desired temperature. Liquid samples were collected after 24, 48, and 72 h from the reactor and were analyzed by using a Bruker GC–MS (Scion SQ/451-GC) equipped with a Restek Rxi-17Sil MS 30 m×0.25 mm capillary column.

MLH, MSH, and NEC hydrogenation: All hydrogenation experiments were performed by using a 300 mL stainless-steel Parr batch autoclave equipped with a four-blade gas-entrainment stirrer (n=1200 rpm). Unloaded LOHC (150 mL) was placed in the pressure vessel, and a constant molar ratio of 400:1 (LOHC/catalyst metal) of the catalyst was added. To ensure an inert atmosphere, the gas volume of the reactor was replaced three times with Ar 4.6 (Rießner Gase) by flushing the reactor. After heating the LOHC to the desired reaction temperature by using an external electrical heating jacket, the autoclave was pressurized with 50 bar of H2 5.0 (Linde), and this pressure was kept constant during the experiment by continuous dosing of H2. Samples of the liquid mixture were withdrawn during the reaction to analyze the progress of the hydrogenation. The samples were analyzed by using a Bruker GC–MS (Scion SQ/451-GC) equipped with a Restek Rxi-17Sil MS 30 m×0.25 mm capillary column.

H12-MLH, H18-MSH, and H12-NEC dehydrogenation: The dehydrogenation experiments in this study were performed by using two different protocols.

Dehydrogenation protocol A: A 100 mL three-neck glass flask equipped with an intensive condenser was loaded with fully hydrogenated LOHC (0.1 mol) and catalyst (0.15 mol %). The set-up was purged with N2 4.0 (Linde) to remove air from the reactor. After calibration of the analyzer at RT, the mixture was heated to the desired reaction temperature by using an external heating jacket. The released H2 was stripped by a constant N2 flow and monitored continuously for 3.5 h by using a fast thermal conductivity analyzer (FTC200, version 1.05, Wagner).

Dehydrogenation protocol B: A 100 mL three-neck glass flask equipped with an intensive condenser was loaded with fully hydrogenated LOHC (0.1 mol). The setup was purged with N2 4.0 (Linde) under stirring to remove air from the reactor. After calibration of the analyzer at RT, the LOHC was heated to the desired reaction temperature by using an external heating jacket. On reaching the preset target temperature, the selected catalyst (0.1 mol %) was added through an integrated loading device. The H2 release was monitored continuously for 2 h by using a fast thermal conductivity analyzer (FTC200, version 1.05, Wagner), and no stripping gas was applied.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusions
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

All authors thank BMW AG, Munich, for permission to publish this work. The German Science Foundation (DFG) has provided additional support for this study through its Erlanger Cluster of Excellence “Engineering of Advanced Materials”. In addition, P.W. and K.O. thank the European Research Council for dedicated resources to study the fundamental aspects of LOHC dehydrogenation catalysis under the ERC Advanced Investigator Grant no. 267376.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusions
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
cssc_201300426_sm_miscellaneous_information.pdf135Kmiscellaneous_information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.