Continuous Flow Synthesis of 2-Imidazolidinone from Ethylenediamine Carbamate in Ethylenediamine Solvent over the CeO₂ Catalyst: Insights into Catalysis and Deactivation

Abstract

Synthesis of 2-imidazolidinone (EU) from ethylenediamine carbamate (EDA-CA), which was produced via chemical absorption of CO2 in ethylenediamine (EDA), with a fixed-bed flow reactor was performed using CeO2 and EDA as a heterogeneous catalyst and reaction solvent, respectively. In the previously reported batch system for the same reaction, the yield of EU was low (62%) due to its sequential reaction with EDA into N,N′-bis(2-aminoethyl)urea (LU) with 38% yield in the liquid phase under the optimized reaction conditions at 413 K. In stark contrast, the flow reaction enabled the highly selective and high-yielding production of EU (94% yield) under the optimized reaction conditions at 363 K. This difference in the selectivity of EU originated from the lower ratio of the amount of EDA to that of the CeO2 catalyst in the current flow reaction system compared to the previous batch system. The highly crystallized CeO2 surface bearing both acid and base sites was found to be responsible for its high catalytic performance. The combination of the kinetics, density functional theory calculations, and adsorption experiments demonstrated that EDA-CA is preferentially adsorbed onto the surface of CeO2 even in the presence of an excess amount of EDA, which is used as a solvent, to be activated and transformed into EU. The solid-state 13C magic angle spinning/cross polarization nuclear magnetic resonance spectroscopy revealed that the CeO2 catalyst was deactivated gradually in a long-term reaction operation due to the formation and deposition of polyurea-like organic compounds on the catalyst surface.

Improvement of Stability of CeO2–Based Catalysts by Mn Doping for the Synthesis of 2-Imidazolidinone from Ethylenediamine Carbamate

“…A 0.5 g portion of each catalyst (pelletized form with 1.0–1.7 mm diameter) was charged into a SUS304 tubular reactor with a diameter of 5 mm and length of either 50 or 100 mm. It should be noted that the pellet size does not matter in this reaction, as demonstrated previously . The liquid mixture of EDA-CA and EDA was continuously fed with a pump from the bottom of the reactor.…”

“…Although a slight increase in the activity was observed in the case of the Mn content of 0.5 wt % against CeO 2 (HS), the Mn doping showed almost no effect on catalytic activity. The apparent activation energies ( E a ) for the EDA-CA transformation to EU over CeO 2 (HS), CeO 2 -873(Precip), and Mn-1.0-CeO 2 -873(Co-precip) were determined to be 97, 106, and 92 kJ mol –1 (the Arrhenius plots are shown in Figure S7); these comparable E a values also suggested no improvement effect on catalytic activity by the doped Mn species. On the other hand, the catalyst stability represented by the deactivation rate constant was greatly dependent on the Mn content.…”

“…As discussed above, the order of the EU yield was CeO 2 -873(Precip) < Mn-1.0-CeO 2 -873(Co-precip) < CeO 2 (HS) when these catalysts were compared at the same W Cat / F EDA‑CA . The EU yield provided by the CeO 2 (HS) catalyst reached as high as 94% yield of EU at the W Cat / F EDA‑CA of 3.48 g h mmol –1 , as reported in our previous study . Although the slightly longer W Cat / F EDA‑CA of 4.88 g h mmol –1 was required, Mn-1.0-CeO 2 -873(Co-precip) produced EU in up to 95% yield.…”

“…The reaction mixture was also analyzed by 1 H NMR spectroscopy with tert -butyl alcohol and methanol- d 4 as the internal standard and deuterated solvent, respectively. Consistent with our previous works, − the unreacted EDA-CA was quantified via 1 H NMR spectroscopy due to its thermal decomposition into EDA and CO 2 in the vaporization chamber of the GC. In some cases, another urea derivative, LU, was also detected in addition to EDA-CA.…”

“…We previously found with the continuous flow reactor that the catalytic activity of CeO 2 gradually decreased in the long-term reaction of EDA-CA to EU, e.g., from 26% yield of EU at the time on stream (TOS) of 9 h to 17% at 72 h, because of the deposition of polyurea-like organic compounds on the catalyst surface (entry 13 in Table S1). Based on this insight into the deactivation of CeO 2 catalyst, there are several possible approaches to developing stable catalysts, for example, suppression of the formation of polyurea-like compounds and/or their precursors and decrease in the affinity between the catalyst surface and polyurea-like compounds and/or their precursors to avoid the accumulation of poisonous compounds on the catalyst surface. Our previous research has demonstrated that the formation of LU, which is considered as a precursor to polyurea-like compounds, occurs in a liquid phase irrespective of CeO 2 catalyst. , The adsorption of polyurea-like compounds onto the surface of CeO 2 is assumed to be caused by the acid–base interaction.…”