Benchmark properties of diphenyl oxide as a potential liquid organic hydrogen carrier: Evaluation of thermochemical data with complementary experimental and computational methods

Highlights

• New enthalpy of vaporization of diphenyl oxide measured by transpiration.

• New enthalpy of formation was measured by combustion calorimetry.

• Available thermochemical data for diphenyl oxide have been evaluated.

• G4 gas phase enthalpy of formation was in agreement with experiment.

• Hydrogenation enthalpy was derived and compared with other LOHC.

Abstract

The standard molar enthalpy of vaporisation of diphenyl oxide was derived from the vapour pressure temperature dependences measured by the transpiration method. Thermodynamic data on vaporisation processes available in the literature were collected. They were evaluated and combined with our own experimental results. Additional combustion experiment on the highly pure diphenyl oxide helped to resolve ambiguity in the enthalpy of formation for this compound. We have evaluated and recommended the set of vaporisation and formation enthalpies for the diphenyl oxide at 298.15 K (in kJ·mol⁻¹): ${\mathrm{\Delta }}_{\text{l}}^{\text{g}}{H}_{\text{m}}^{\text{o}}$ = (66.7 ± 0.2), ${\mathrm{\Delta }}_{\text{f}}{H}_{\text{m}}^{\text{o}}$ (liq) = −(15.8 ± 1.4), and ${\mathrm{\Delta }}_{\text{f}}{H}_{\text{m}}^{\text{o}}$ (g) = (50.9 ± 1.4), as the reliable benchmark properties for further thermochemical calculations. Gas phase molar enthalpy of formation of diphenyl oxide calculated by the high-level quantum-chemical method G4 was in an excellent agreement with the recommended experimental data. The standard molar Gibbs function of formation and the standard molar entropy of formation of diphenyl oxide were estimated. The hydrogenation/dehydrogenation reaction enthalpy of diphenyl oxide was calculated and compared with the data for other potential liquid organic hydrogen carriers.

Introduction

The concept of the liquid organic hydrogen carriers (LOHCs), in which hydrogen is covalently bonded to an organic molecule of a liquid substance, was proposed in the 1980s and a variety of organic molecules has been discussed to date [1]. Our recent experimental and computational studies on heteroaromatic compounds, such as carbazole, N-ethylcarbazole, as well as on the benzyl and dibenzyltoluenes have established the thermodynamic background for applications of this fully reversible hydrogen storage [2], [3], [4], [5]. Our current activities have been focused on a search for new low-cost LOHC compounds for the effective hydrogen storage. Diphenyl oxide [CAS 101-84-8, see Fig. 1] could be considered as one of the possible LOHC candidates for the hydrogen storage mostly due to its commercial availability on the large scale. Diphenyl oxide as an ingredient of organic heat transfer fluids Therminol® VP-1, Dowtherm^TM and Diphyl®, is used in concentrated solar power technology to transfer the heat from the solar collectors to the power cycle. The accurate and reliable thermodynamic properties of diphenyl oxide are required for analysis of the effectiveness of the hydrogen storage. In this context, we have collected in the literature thermodynamic data for diphenyl oxide, but the evaluation of the available results has been thwarted with complications, because the melting temperature of this compound (T_fus = 300.0 K) is slightly above the reference temperature. This fact is especially important for the combustion calorimetry, where the initial state must be defined clearly. For this reason, Furukawa et al. [6] have deliberately changed the conventional combustion procedure and measured combustion energies at 303.15 K in order to keep sample definitely in the liquid state. The published later combustion result by Cass et al. [7] was derived from only three experiments but on the solid sample. The consistency of two data sets is rather poor. Another example, the available enthalpies of vaporisation of diphenyl oxide spread from 60 kJ mol⁻¹ [8] to 70 kJ mol⁻¹ [9]. New additional thermochemical experiments with diphenyl oxide have been conducted to help with establishing of consistency in the available thermodynamic data. The aim of this study was to evaluate the thermochemical data available for diphenyl oxide with the complementary experimental and computational methods in order to recommend benchmark thermochemical properties for this compound.