(177b) Bio-Butadiene Production Via 2,3?Butanediol Dehydration: Mechanism Elucidation and Kinetic Model Development

The production of key monomers such as 1,3-butadiene (BD) continues to depend largely on fossil resources. Given the increasing need to reduce carbon emissions and promote sustainability, the shift towards bio-based production of chemicals, including monomers, is only becoming more essential [1]. In this regard, 2,3-butanediol (2,3-BDO) stands out as a promising precursor for bio-based BD production, as it can be efficiently produced from sugar through fermentation with high yields [1].

The conversion of 2,3-BDO to BD comprises two consecutive dehydrations, with 3-buten-2-ol (3B2OL) as the crucial intermediate obtained after the first dehydration. Optimizing this first step is essential to subsequently achieve a high BD yield. Experimental investigations over acidic catalysts, including Al₂O₃, zeolites with various modifications demonstrated that these catalysts do not selectively produce the desired intermediate [2,3], but rather lead to methyl ethyl ketone (MEK), 2-methylpropanal (2MPAL) and 3-hydroxy-2-butanone (3H2B) [4]. However, catalysts containing weak basicity, such as ThO₂ and Sc₂O_3, performed better [1]. In this work, catalysts with reduced acidity and limited basicity, ZrO₂ and Sc₂O₃, have been characterized, tested and compared at iso-conversion of 2,3-BDO to confirm the nature of the active sites and elucidate corresponding reaction mechanisms of the selective dehydration as well as the competitive side reactions. Furthermore, the dehydration reaction kinetics have been modeled over both catalysts to investigate the impact of adsorption and reaction steps on the selectivity of the products and obtain insights on maximization of the selectivity towards the desired product.

Materials and methods

A monoclinic zirconium oxide catalyst was synthesized via a hydrothermal synthesis method and calcined at 800°C (ZrO₂). Sc₂O₃ was purchased from Sigma-Aldrich and calcined at 800°C (Sc₂O₃). The surface area and porosity, acid-base properties and stability and activity of available species of the materials were characterized by BET, NH₃-TPD, CO₂-TPD and In-situ DRIFT respectively. A Berty (CSTR type) reactor was employed to investigate the gas-solid 2,3-BDO dehydration kinetics. Intrinsic kinetics experiments were performed in a wide range of operating conditions covering the entire 2,3-BDO conversion range: temperature (300-400°C), space-time (56-1130 kg s mol^-1), partial pressure of inlet BDO (0.1-0.5 bar) at a total pressure of 10 bar.

A Langmuir-Hinshelwood Hougen-Watson (LHHW) model was constructed based on various proposed mechanisms and isothermal regressions at all investigated temperatures per catalyst are performed in addition to overall non-isothermal regressions. The model and its parameters were subject to statistical tests such as the F test for model significance and t tests for the individual significance of each of the parameters.

Results and discussion

Experimental investigations reveal a stark contrast in catalytic behavior between Sc₂O₃ and ZrO₂ catalysts, driven by the distinct nature of their active sites. ZrO₂ exhibits a more complex behavior, with more pronounced MEK formation at low 2,3-BDO conversion and competition between MEK and 3B2OL at higher 2,3-BDO conversion, depending on the operating conditions. In contrast, the product distribution over Sc₂O₃ primarily comprises 3B2OL, particularly at lower 2,3-BDO conversions. Moreover, the Sc₂O₃ catalyst demonstrated superior selectivity for BD, achieving a selectivity of 40% (mol mol^-1) at complete 2,3-BDO conversion, while less than 3% BD was observed in ZrO_2, although significant 3B2OL was observed.

Operating conditions also exhibit a significant impact on 2,3-BDO conversion and selectivity. Temperature enhances the 2,3-BDO conversion and, consequently, changing the product distribution for both catalysts. For Sc₂O₃, the lowest 2,3-BDO conversion of 8% (mol mol^-1) at 325°C resulted in 80% (mol mol^-1) selectivity towards 3B2OL, with MEK as the primary byproduct. At 385°C, the BD selectivity was maximal at 40% (mol mol^-1), with MEK accounting for 30% (mol mol^-1) at full 2,3-BDO conversion while the maximum selectivity towards 3B2OL was observed at lower temperature (300°C) and highest space time (1130 kg s mol^-1) in the case of ZrO₂ catalysts. Higher space times lead to more pronounced formation of secondary products, including the desired BD for both catalysts. Controlling the 2,3-BDO inlet partial pressure proved to be essential to limit undesired products formation and maximize BD formation for both catalysts.

This performance is attributed to the mainly medium strength basic active sites of Sc₂O₃. Conversely, ZrO₂ is characterized by a balanced distribution of acidic and basic sites, as evidenced by NH₃ and CO₂-TPD experiments. The predominant formation of 3B2OL during the dehydration of 2,3-BDO over Sc₂O₃ catalyst at low 2,3-BDO conversion, in addition to primarily basic sites, suggests that the reaction follows an E1cb mechanism. The ZrO₂ catalyst contains acid-base concerted sites, which are also responsible for the formation of 3B2OL but via a different mechanism, E2 mechanism. On top of that, unpaired acidic sites on the surface of ZrO₂ open up an additional mechanism for 2,3-BDO conversion, i.e., E1, leading to the formation of mainly MEK. This all together explains the complex behavior of the product distribution over this catalyst. Apart from these mechanistic differences, thermodynamic equilibrium limitations between 3H2B and 2,3-BDO result in an additional complexity to the product distribution on both catalysts, particularly at higher 2,3-BDO conversion.

The mechanistic hypotheses have been further assessed through a kinetic analysis. The activity of the two catalysts was compared, the 2,3-BDO conversion as a function of the space time indicated faster reaction rate for ZrO₂. In terms of turnover frequency (TOF) calculated via average reaction rate at 350°C and determined acidic and base sites for the respective catalyst, ZrO₂ has a rate of 10 kg h^-1 mol^-1 while Sc₂O₃ has a rate of 6E-6 kg h^-1 mol^-1 indicating the extreme difference in mechanism between the two catalysts. In addition, apparent activation energies of the two most important reactions, 3B2OL and MEK formation, over the two catalysts have been determined from the Arrhenius plot. For Sc₂O₃, 3B2OL formation requires an activation energy of 80 kJ mol^-1, which is significantly higher than the 40 kJ mol^-1 observed for ZrO₂. However, MEK formation is governed by an activation energy of 63 kJ mol^-1 across both catalysts, underscoring its thermodynamic favorability in the absence of strict mechanistic control.

The kinetic behavior of the associated reaction mechanisms and appropriate operating parameters for the intended product maximization are determined by further kinetic analysis of the data collected for both catalysts. The model is used to quantitatively capture the intrinsic reaction kinetics and the adsorption phenomena driving selectivity for each catalyst. The estimated parameters for each catalyst provide valuable insights into the interaction between reaction and adsorption processes specific to each catalyst. The activation energy profiles offer critical insights into the energy barriers associated with distinct mechanistic pathways, shedding light on the interplay between catalyst surface properties and reaction kinetics. For ZrO₂, the model confirmed that the adsorption of 2,3-BDO on acid-base concerted sites is pivotal for 3B2OL formation. In contrast, Sc₂O₃ catalyst performance is dictated by the ratio of basic sites to impurities (acidic) interference suppressing undesired byproduct formation.

The adsorption behavior of 2,3-BDO over ZrO₂ catalyst reveals the importance of site-specific interactions in determining selectivity over the catalyst. On ZrO₂, acid-dominated adsorption pathways were linked to MEK production, while base-dominated adsorption favored 3B2OL. The adsorption enthalpy for basic sites was calculated to be significantly more negative than that for acidic sites (-50 kJ mol^-1), highlighting the energetic preference for the base-driven pathway. In the case of Sc₂O₃, the abundance of basic sites promoted 3B2OL formation, with minimal competition from acidic sites, reinforcing the dominance of the E1cb mechanism.

Conclusion

The dehydration of 2,3-butanediol (2,3-BDO) over Sc₂O₃ towards 1,3-butadiene (BD) was found to proceed through an E1cb mechanism catalyzed by basic sites while the side products are formed via E1 reaction catalyzed by residual acid sites on the surface. The maximum BD yield was observed at 385°C, 1130 kg s mol^-1, full 2,3-BDO conversion, and an inlet 2,3-BDO partial pressure of 0.2 bar, equivalent with a selectivity towards BD of 40% (mol mol^-1). The optimal operating conditions for bio-butadiene production in a single step over Sc₂O₃ are a temperature of 385°C, inlet 2,3-BDO partial pressure of 0.2 bar and a space time amounting to 1130 kg s mol^-1. The reaction mechanisms determined allowed for the development of a kinetic model for both catalysts. The kinetic model confirms that the adsorption of 2,3-BDO on the acid-base concerted sites is crucial for BD formation over ZrO₂ catalyst, while the ratio of basic sites to impurities in the catalyst is crucial for BD formation over Sc₂O₃.

References

1. D. Sun, Y. Li, C.Yang, Y. Su, Y. Yamada, S. Sato, Fuel process. Technol. 197, 106193 (2020).
2. F. Zeng, W.J. Tenn, S.N.V.K. Aki, J. Xu, B. Liu, K.L. Hohn, J. Catal. 344,77–89 (2016).
3. Q. Zheng, J. Xu, B. Liu, K.L. Hohn, J. Catal. 360 221–239 (2018).
4. H. Duan, D. Sun, Y. Yamada, S. Sato, Catal. Commun. 48 1–4 (2014).