English

• 1. Introduction

  In the field of chemical engineering, efficient, safe, and environmentally friendly process routes have always been the goals that people strive for. For the synthetic industry, process routes can be designed under the direction of the Retrosynthetic Analysis Method (by E.J. Corey) [1] and the Atom Economy Theory (by B.M. Trost) [2]. However, the aforementioned two theories fall short in conducting a comprehensive assessment of the industrial application of the reaction. This deficiency implies that they fail to undertake a thorough and all-encompassing examination and analysis of how the reaction can be effectively and viably utilized within the industrial setting. Such an oversight hinders a complete understanding of the reaction's potential, challenges, and implications in real-world industrial scenarios.

  As depicted in Fig. 1, the Retrosynthetic Analysis Method deduces synthetic equivalents through disconnection (dis), functional group interconversion (FGI) and functional group addition (FGA) analysis from the target product, and then reagents are selected based on this outcome. This method can significantly enhance the efficiency of synthetic route design, but it is prone to being restricted by existing thinking paradigms and choosing traditional routes and reagents. Atom economy can be reflected by the atom utilization rate, which is a crucial index for evaluating the efficiency of chemical reactions (Fig. 1). Taking atom economy as the consideration objective, the most effective route can be selected from multiple synthetic routes derived from retrosynthetic analysis results.

  Figure 1

  

  Figure 1.  Comparison of the theories.

  DownLoad: Full-Size Img PowerPoint

  However, in chemical industrial process, the influencing factors are constantly changing. Particularly in the production of low value-added products, relying solely on the aforementioned two theories is insufficient to select the most optimal solution. Besides reaction steps and atom economy, significant indicators involved in choosing a chemical process route also encompass raw material cost, the availability and substitutability of raw materials, the stability of the industrial chain, reaction energy consumption, environmental matters, etc. The competition in the field of modern industrial synthesis is intense, and the requirements are becoming increasingly elevated. These demands breaking with conventions, acquiring distinctive techniques, and achieving technological overtaking on curves. Hence, a new systematic theory is requisite to guide the design of chemical processes. In 2020, we proposed the Element Transfer Reaction (ETR) theory, which is an important fundamental theory guiding chemical process design [3, 4]. This theory approaches problems from the perspective of element circulation. It simplifies the complexity and decomposes the core issues of chemical processes into three indexes: element source (S), driving force (D), and output (O) (Fig. 1). Under the guidance of this theory, we have achieved a series of accomplishments in the industry. This paper aims to further elucidate the scientific connotation of the theory and discuss its application prospects.

  2. Element source

  For synthetic process design, when analysing the source of elements, not only should we consider whether the raw materials are low-cost, but also it is necessary to ensure that the raw materials are not easily restricted by external conditions. Additionally, the properties of the raw materials should be stable, safe, and environmentally friendly. Regarding environmental protection issues, analysing the source of elements can identify the source of pollutants for treatment. Selecting a reasonable process route and avoiding unnecessary elements in raw materials is the most effective approach to eliminating the residue of that element in products.

  The synthesis of adiponitrile is a typical case for clarifying the strategy of choosing element sources. Adiponitrile is used to synthesize hexamethylene diamine (HMDA), a key material in the Nylon-66 industry. Conventionally, it is synthesized through three methods: the direct cyanation of butadiene [5, 6], the electrolytic dimerization of acrylonitrile [7, 8], and the catalytic amination of adipic acid [9]. Due to the high cost of raw materials, high energy consumption issues and long process routes, the electrolytic dimerization of acrylonitrile and the catalytic amination of adipic acid have been phased out. Currently, the direct cyanation of butadiene is the mainstream process route applied in the industrial production of adiponitrile. In this method, the reaction of butadiene with HCN initially yields the (E)-pent-3-enenitrile and pent-4-enenitrile mixtures, which are then transformed into adiponitrile through isomerization and cyanation processes. Despite the toxicity of HCN, using butadiene as the carbon and hydrogen sources might cause the process route to be limited by the raw materials. Butadiene originates from petroleum, and its supply is readily constrained by international geopolitical factors.

  Thus, as analysed by the ETR theory, it is indispensable to develop a new synthetic route for adiponitrile production using non-petroleum starting materials. In 2022, we proposed "How to Synthesize Adiponitrile from Non-Petroleum Starting Materials", and it was selected as one of the top ten industrial technology issues by the China Association for Science and Technology [10, 11]. We contend that synthesizing adiponitrile from cyclohexene is a pragmatic strategy to tackle this issue because cyclohexene is produced through the partial hydrogenation of benzene [12], which can be derived from both petroleum and coal. Given that our country is abundant in coal, the route of synthesizing adiponitrile from cyclohexene has the resource guarantee. Switching the source of the upstream raw material benzene will not impact the downstream process routes, and this characteristic is conducive to ensuring the stability of the industrial chain.

  There are recent examples of synthesizing adiponitrile from cyclohexene. For instance, the reaction of cyclohexene with H₂O₂ can readily generate cyclohexane-1, 2-diol [13], which can be transformed into adiponitrile through the ammoxidation reaction [14]. Adiponitrile was also synthesized through the ammoxidation reaction of cyclohexanol, which was the hydration product of cyclohexene [15]. In our case, we have declared a patent on the selenium-catalysed ammoxidation reaction of cyclohexene to produce adiponitrile directly [16]. Indeed, as depicted in Scheme 1, the reaction was constituted by several steps of unit reactions, such as the selenium-catalysed oxidative cracking reaction of alkene [17-21], the amination of carbonyls, the oxidation of imine to aldoxime, and the selenium-catalysed dehydration of aldoxime to nitrile [22, 23]. How to inhibit side reactions, such as the over oxidation of aldehydes into carboxylic acids as well as the deoximation reaction, is the key to this technique (Scheme 1). The inhibition of the deoximation reaction confronts formidable challenges. As reported in the existing research, this reaction is prone to occur under oxidative conditions. Chalcogen elements (including both Se and Te), metals, free radical initiators, and even its carbonyl products can all act as catalysts for this transformation [24-30].

  Scheme 1

  

  Scheme 1.  Selenium-catalysed ammoxidation reaction of cyclohexene to produce adiponitrile.

  DownLoad: Full-Size Img PowerPoint

  3. Driving force

  Reaction driving force reflects the analysis of reaction energy consumption and can be utilized to determine whether a reaction is likely to occur, thereby ascertaining whether to attempt certain specific synthetic routes. For instance, we consider that generating stable molecules that can be promptly removed from the reaction system constitutes a driving force for the reaction; the formation of stable chemical bonds serves as a driving force for the reaction; the generation of products with lattice energy significantly exceeding that of the reactants is also a driving force for the reaction.

  The synthesis of fluorochemicals using CaF₂ as the direct fluorinating reagent is a typical example guided by the ETR theory through the analysis of the driving force. CaF₂ can be abundantly obtained from fluorspar ore and is the main fluorine source in the fluorochemical industry. However, due to its high lattice energy, CaF₂ cannot be directly utilized as the fluorinating reagent and is usually transformed into HF by acidification with H₂SO₄. HF is highly toxic and highly corrosive, posing significant harm to operators and the environment. Therefore, the direct synthesis of fluorochemicals using CaF₂ is a long-awaited technology and a historic breakthrough in this field.

  LiPF₆ is currently the main commercial electrolyte salt for lithium-ion battery electrolyte. In the conventional approach, the reaction of PCl₅ with HF initially yields PF₅, and the subsequent reaction of PF₅ with LiF leads to LiPF₆ (Scheme 2, method a). Guided by the ETR theory, we contemplated that the fluorination reaction of PCl₅ could directly employ CaF₂ as the fluorinating reagent since there are two types of driving forces in the reaction: Firstly, the generated P-F in the product PF₅ is relatively more stable than P-Cl in the PCl₅ reactant; Secondly, the generated PF₅ is in a gaseous state and can be removed from the reaction system, which may continuously drive the reaction in the direction of generating PF₅. This new concept of non-HF production of LiPF₆ has been successfully industrialized in 2019 (Scheme 2, method b) [31].

  Scheme 2

  

  Scheme 2.  Methods for the synthesis of LiPF₆.

  DownLoad: Full-Size Img PowerPoint

  The next challenge lies in synthesizing LiPF₆ by utilizing non-chloride starting materials, as chloride residues in the product might be corrosive to equipment in application scenarios. We hypothesized that P₂O₅ could be employed as the phosphorus source instead of PCl₅. Nevertheless, the high P-O bond energy within P₂O₅ poses a significant obstacle to realizing this concept. As depicted in Scheme 2, method c, the new method might generate Ca(PO₃)₂ as a by-product, which can be converted into Ca₃(PO₄)₂. These calcium salts possess much larger lattice energy than that of the reactant CaF₂. According to the ETR theory, this could provide sufficient driving forces for the conversions. Thus, we boldly adopted the process route presented in Scheme 2, method c and achieved success, producing LiPF₆ with chloride residue less than 0.2 ppm [32]. Later, Patel et al. also reported in Science a process for preparing fluorochemicals using CaF₂ as the fluorine source, and the substantial lattice energy difference between the produced Ca₃(PO₄)₂ and the starting CaF₂ was also supposed to be the driving force of the reaction [33].

  4. Output

  The output of the reaction reflects the destination of the elements and serves as a crucial indicator for assessing the efficiency of element utilization in chemical routes. Unlike the definition of the atom utilization rate in atom economy indicators, we take into account not only the atoms in the expected products but also those in the by-products. The generation of by-products may not lead to pollution if they can be transformed into readily consumed materials. The by-products can be consumed either by being converted into marketable products or through recycling and reuse within the industrial chain.

  The non-HF production of LiPF₆ can be utilized to elucidate the concept [29]. In this process, CaCl₂ was produced equivalently, but it did not cause pollution. As the factory was located in a cold zone at high altitude, the highways nearby were covered with snow for several months each year. Thus, CaCl₂ could be consumed as the snowmelt reagent and used on the adjacent highways. It should be noted that the location of the factory might have an impact on the analysis of by-product treatment, and this should be carefully contemplated in the design of the industrial process.

  Another example pertains to the treatment of fluorinated organosilicon by-products in the production of LiPO₂F₂. LiPO₂F₂ is a crucial lithium-ion battery additive that can enhance the battery's high and low-temperature resistance performances. It can be industrially synthesized through the F-O exchange of LiPF₆ with Li₂CO₃, but this process has a high equipment requirement due to the generation of CO₂, along with less soluble reactants resulting in a bubbled porridge-like reaction liquid [34]. The organic siloxane method is capable of circumventing the aforementioned disadvantage. It involves the reaction of organic siloxane with LiPF₆, and the formation of a high energy Si-F bond can provide the driving force for the partial F-O exchange of LiPF₆ to produce LiPO₂F₂ [35]. However, this method inevitably leads to the generation of trimethylfluorosilane (Me₃SiF), a flammable gas hazardous to the environment. Although Me₃SiF is a usable chemical, its market demand is very limited. Therefore, from the perspective of output analysis, it is necessary to convert this chemical into more consumable and high-value-added products.

  The hydration of Me₃SiF with aqueous NaOH occurs under mild conditions, yielding 1, 1, 1, 3, 3, 3-hexamethyldisiloxane (Me₆Si₂O) and NaF (Eq. 1) [36]. NaF, as a fluorine source, can be recycled and reused in the industry and is deemed easily consumable based on the output analysis of the ETR theory. Me₆Si₂O is a useful chemical employed as sealing agents, cleaning agents, defoaming agents, and organic synthesis intermediates, etc. Its market demand is substantial, and the current price is approximately $11,000/t, which can counterbalance the processing cost. Furthermore, the boiling point of Me₆Si₂O is 101 ℃, signifying that it can be effortlessly purified through distillation with low energy consumption. Burning purified Me₆Si₂O in O₂ can generate high-purity silica (Eq. 2), which is a bottleneck material in the chip industry and can be sold at a favorable price (approximately $69,000/t) [36].

  (1)

  (2)

  The conversion of glycerol into 1, 3-propanediol (1, 3-PDO) could be a fine example to elucidate the disparity between the ETR theory and the Atom Economy Theory. This process is currently accomplished via the fermentation method in the industry [37]. The catalytic hydrogenation method is appealing due to its low cost and straightforward process [38]. Nevertheless, aside from the problem of low catalyst durability, low reaction selectivity is also one of the major hurdles impeding the industrial application of this process. Pessimistically speaking, this issue might be insoluble since the middle hydroxyl in glycerol is less reactive than the terminal ones, and 1, 2-propanediol (1, 2-PDO) is likely to be generated inevitably during the process (Eq. 3). From the perspective of the Atom Economy Theory, this reaction with low atom economy is of little significance in the industry. By contrast, from the viewpoint of the ETR theory, if a method for the consumption of 1, 2-PDO is discovered, the catalytic hydrogenation method might also be applicable in the industry in the future. This might inspire a new direction for research, that is, not to attempt to prevent the generation of 1, 2-PDO, but to develop its consumption approach. Indeed, 1, 2-PDO is also a useful intermediate for the preparation of unsaturated polyester resin and can be utilized as a wetting agent in the cosmetic industry, but the market demand is much lower than that of 1, 3-PDO [39]. Therefore, developing products that can consume a large quantity of 1, 2-PDO is the key to resolving this problem.

  (3)

  5. Conclusions and perspectives

  In conclusion, ETR theory emerges as a valuable supplement to the Retrosynthetic Analysis Method and the Atom Economy Theory in the field of synthetic chemistry. The fundamental difference between the ETR theory and the preceding two theories lies in its pronounced concentration on the analysis of industrial synthesis routes. These routes are often subject to a multiplicity of variables and complex influences. Specifically, the ETR theory exhibits several distinct and significant characteristics.

  Firstly, it places considerable emphasis on the selection of element sources. This is not merely restricted to achieving low prices but encompasses a comprehensive consideration of various factors such as stability, accessibility, and critical safety concerns. A meticulous assessment of these aspects ensures the robustness and feasibility of the synthetic process.

  Secondly, the analysis of the reaction driving force within the ETR theory holds the potential to transcend conventional and outdated perspectives. By doing so, it paves the way for the conception and design of novel and unprecedented synthetic routes. This innovative approach can potentially revolutionize the field by opening up new avenues and possibilities.

  Finally, in contrast to the Atom Economy Theory, the ETR theory adopts a more holistic and dynamic stance. It does not merely rely on the mechanical calculation of the atom utilization rate to indiscriminately reject a synthetic route. Instead, it places significant emphasis on the consumption of by-products, aiming to strike a balance with market requirements. This perspective recognizes the complexity and interrelated nature of industrial processes and strives to create a more sustainable and economically viable synthetic framework.

  In essence, the ETR theory provides a more comprehensive, flexible, and forward-looking approach to synthetic chemistry, offering valuable insights and strategies for the development of efficient and sustainable synthetic routes in the industrial context.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Hongen Cao: Writing – original draft. Xinrui Xiao: Writing – original draft. Xu Zhang: Writing – review & editing. Yiyang Zhang: Writing – review & editing. Lei Yu: Writing – review & editing, Writing – original draft, Funding acquisition, Conceptualization.

  Acknowledgments

  We thank the Yangzhou Key Research and Development Program: Industry Foresight and Key Core Technology (No. YZ2023019), Cooperation Project of Yangzhou City with Yangzhou University (No. YZ2023209), Sichuan Tianfu Talent Programme (No. A.2200732), Chengdu Rongpiao Talent Programme (No. 1043), SeleValley Scholars Basic Research Project (No. 2301) and Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD) for financial support.

  
  
• 1. [1]

     E.J. Corey, Angew. Chem. Int. Ed. 30 (1991) 455–465. doi: 10.1002/anie.199104553

  2. [2]

     B.M. Trost, Science 254 (1991) 1471–1477. doi: 10.1126/science.1962206

  3. [3]

     C. Chen, Y. Cao, X. Wu, et al., Chin. Chem. Lett. 31 (2020) 1078–1082. doi: 10.1016/j.cclet.2019.12.019

  4. [4]

     L. Yu, Element Transfer Reactions, Guangming Daily Press, Beijing, 2021.

  5. [5]

     L. Bini, C. Müller, D. Vogt, Chem. Cat. Chem. 2 (2010) 590–608. doi: 10.1002/cctc.201000034

  6. [6]

     L. Bini, C. Müller, D. Vogt, Chem. Commun. 46 (2010) 8325–8334. doi: 10.1039/c0cc01452d

  7. [7]

     D.E. Blanco, A.Z. Dookhith, M.A. Modestino, React. Chem. Eng. 4 (2019) 8–16. doi: 10.1039/c8re00262b

  8. [8]

     L. Yu, J. Wang, H. Cao, K. Ding, Q. Xu, Chin. J. Org. Chem. 34 (2014) 1986–1991. doi: 10.6023/cjoc201405004

  9. [9]

     J.C.J. Bart, S. Cavallaro, Ind. Eng. Chem. Res. 54(2015) 1–46. doi: 10.1021/ie5020734

  10. [10]

      L. Yu, Y. Chen, H. Ju, Focusing On the Future of Science and Technology, China Science and Technology Press, Beijing, 2023, pp. 305–312.

  11. [11]

      Y. Chen, H. Zhou, L. Yu, Chin. Sci. Bull. 69 (2024) 370–380. doi: 10.1360/tb-2023-0607

  12. [12]

      T. Wu, P. Zhang, T. Jiang, D. Yang, B. Han, Sci. China Chem. 58 (2015) 93–100. doi: 10.1007/s11426-014-5249-x

  13. [13]

      J. Huang, R. Qian, S. Wang, H. Cao, Chin. J. Org. Chem. 41 (2021) 1639–1645. doi: 10.6023/cjoc202008032

  14. [14]

      J. Xu, X. Jia, J. M, et al., Patent, CN201410164307.5, 2014.

  15. [15]

      J. Xu, X. Jia, J. Ma, et al., Patent, CN201711336931.9, 2017.

  16. [16]

      L. Yu, K. Cao, X. Deng, Y. Yang, Patent, CN108530318, 2018.

  17. [17]

      T. Wang, X. Jing, C. Chen, L. Yu, J. Org. Chem. 82 (2017) 9342–9349. doi: 10.1021/acs.joc.7b01245

  18. [18]

      X. Chen, J. Mao, C. Liu, et al. Chin. Chem. Lett. 31 (2020) 3205–3208. doi: 10.1016/j.cclet.2020.07.031

  19. [19]

      W. Zhou, X. Xiao, Y. Liu, X. Zhang, Chin. J. Org. Chem. 42 (2022) 1849–1855. doi: 10.6023/cjoc202201023

  20. [20]

      X. Li, H. Hua, Y. Liu, L. Yu, Org. Lett. 25 (2023) 6720–6724. doi: 10.1021/acs.orglett.3c02569

  21. [21]

      X. Li, H. Zhou, R. Qian, X. Zhang, L. Yu, Chin. Chem. Lett. 36 (2025) 110036. doi: 10.1016/j.cclet.2024.110036

  22. [22]

      X. Zhang, J. Sun, Y. Ding, L. Yu, Org. Lett. 17 (2015) 5840–5842. doi: 10.1021/acs.orglett.5b03011

  23. [23]

      X. Jing, T. Wang, Y. Ding, L. Yu, Appl. Catal. A: Gen. 541 (2017) 107–111. doi: 10.1016/j.apcata.2017.05.007

  24. [24]

      X. Deng, R. Qian, H. Zhou, L. Yu, Chin. Chem. Lett. 32 (2021) 1029–1032. doi: 10.1016/j.cclet.2020.09.012

  25. [25]

      F. Wang, T. Chen, Y. Shi, L. Yu, Asian J. Org. Chem. 10 (2021) 614–618. doi: 10.1002/ajoc.202000675

  26. [26]

      H. Li, X. Jing, Y. Shi, L. Yu, React. Chem. Eng. 6 (2021) 119–124. doi: 10.1039/d0re00333f

  27. [27]

      F. Wang, C. Yang, Y. Shi, L. Yu, Mol. Catal. 514 (2021) 111849.

  28. [28]

      C. Chen, X. Zhang, H. Cao, et al. Adv. Synth. Catal. 361 (2019) 603–610. doi: 10.1002/adsc.201801163

  29. [29]

      Y. Zheng, A. Wu, Y. Ke, H. Cao, L. Yu, Chin. Chem. Lett. 30 (2019) 937–941. doi: 10.1016/j.cclet.2019.01.012

  30. [30]

      X. Jing, D. Yuan, L. Yu, Adv. Synth. Catal. 359 (2017) 1194–1201. doi: 10.1002/adsc.201601353

  31. [31]

      J. Liu, Y. Cai, C. Xiao, et al., Ind. Eng. Chem. Res. 58 (2019) 20491–20494. doi: 10.1021/acs.iecr.9b04958

  32. [32]

      J. Liu, Y. Cai, H. Pang, et al., Chin. Chem. Lett. 33 (2022) 4061–4063. doi: 10.1016/j.cclet.2021.12.068

  33. [33]

      C. Patel, E. André-Joyaux, J.A. Leitch, et al., Science 381 (6655) (2023) 302–306. doi: 10.1126/science.adi1557

  34. [34]

      Y. Wang, J. Ye, J. Wang, Chem. Ind. 51 (2019) 51–54.

  35. [35]

      M. Takehara, Patent, CN101507041, 2009.

  36. [36]

      L. Yu, N. Liu, X. Wu, et al., Patent, CN116715244, 2023.

  37. [37]

      C.S. Lee, MK. Aroua, W.M.A.W. Daud, et al., Sust. Energ. Rev. 42 (2015) 963–972. doi: 10.1016/j.rser.2014.10.033

  38. [38]

      Q. Tong, Q. Gao, B. Xu, L. Yu, Y. Fan, Chin. J. Org. Chem. 37 (2017) 753–758. doi: 10.6023/cjoc201610002

  39. [39]

      F. Coccia, N. d'Alessandro, A. Mascitti, E. Colacino, L. Tonucci, Chem. Cat. Chem. 16 (2024) e202301672. doi: 10.1002/cctc.202301672

• 

  Figure 1  Comparison of the theories.

  下载: 全尺寸图片 幻灯片
  

  Scheme 1  Selenium-catalysed ammoxidation reaction of cyclohexene to produce adiponitrile.

  下载: 全尺寸图片 幻灯片
  

  Scheme 2  Methods for the synthesis of LiPF₆.

  下载: 全尺寸图片 幻灯片
•