Synthesis and structure of titanium complexes with phosphonium-bisphenolate ligands “{P⁺[O₂]}H₂” and their catalytic trimerization of 1-octene

Abstract

A methylaluminoxane-activated titanium complex catalyzed the selective trimerization of 1-octene in bulk or benzene, exhibiting a trimer selectivity that exceeded 80%. The selective trimerization of 1-octene catalyzed by this complex catalyst with a monoanionic phosphonium-bisphenolate ligand demonstrated catalytic activity comparable to that of previously reported complexes with monoanionic tridentate ligands.

---

Lubricants are crucial for reducing friction, lowering energy consumption, and extending the lifespan of equipment.¹ Almost every lubricant used in plants today started off as just a base oil, and their performance depends on their base oils, among which group IV polyalphaolefins (PAOs) exhibit superior properties.^2–9 PAOs, which are mainly derived from the cationic oligomerization of α-olefins, are key components of high-performance engine oils.^10–19 However, this cationic oligomerization process often leads to carbocation rearrangements, which result in skeletal modifications, hydrogenation-resistant olefins, and excessive methyl branching, all of which degrade lubricant quality.²⁰ Developing efficient catalysts to suppress these side reactions and achieve the selective trimerization of α-olefins remains a major challenge.

In 2001, a chromium-based tridentate ligand catalytic system for α-olefin trimerization was reported, and it exhibited high selectivity (>80%) and reactivity (TON ∼ 1000) for the trimerization of 1-dodecene.²¹ Subsequently, in 2010, a titanium complex with tridentate phenoxyimine ligands was discovered,²² and in 2013, this titanium catalyst system was improved so that it would exhibit catalytic activity for the selective trimerization of α-olefin, thereby achieving even higher selectivity (>95%) with a TON of ∼350.²³ The recent study of selective trimerization, published in 2022, reported a chromium-based catalytic system with significantly enhanced activity (TON ∼ 11 000) for the trimerization of 1-octene.¹ However, despite these advancements, the catalytic activities achieved thus far remain inadequate for commercialization.

Efficient and cost-effective catalysts for α-olefin trimerization are still highly sought after. Apart from the aforementioned systems, very few catalytic systems are capable of selectively promoting α-olefin trimerization. We synthesized a phosphonium-bisphenolato titanium(IV) complex and evaluated its catalytic performance during the trimerization of 1-octene. In this study, we report that this system exhibits high selectivity (>80%) and reactivity (TON ∼ 300). Notably, metal complexes bearing monophenolate ligands have been reported to catalyze the cyclotrimerization of terminal alkynes and to promote the polymerization of 1-hexene with high stereoregularity.²⁴ This suggests that the cationic-bisphenolate ligand framework could play a crucial role in controlling the trimerization of the resulting products (Scheme 1).

Scheme 1 Cr- and Ti-based catalyst for selective α-olefin trimerization.

The route for the synthesis of the phosphonium-bisphenolate ligand “{P⁺[O₂]}H₂” (1) and titanium complex (2) bearing 1 is depicted in Scheme 2. Ligand 1 was obtained as a white solid in 41% yield by refluxing 5H-benzo[b]phosphindole and 2,4-di-tert-butyl-6-chloromethylphenol for 19 h, which was followed by solvent washing with hexane and toluene. The ligand exhibited high thermal stability and remained stable in air. The structure of ligand 1 was elucidated through elemental analysis and nuclear magnetic resonance (NMR) spectroscopy. Subsequently, a solution of ligand 1 was added dropwise to a solution of TiCl₄(thf)₂ at room temperature in a glove box and allowed to react under stirring for 24 h. The resulting mixture was washed with hexane to yield a red solid in 74% yield. The chemical composition of titanium complex 2 was confirmed by elemental analysis, and its molecular structure was determined by NMR spectroscopy and single-crystal X-ray diffraction. Single crystals of the titanium complex that were suitable for X-ray structural analysis were obtained through recrystallization in a mixture of toluene and pentane. Their molecular structures are presented in Fig. 1. In the solid state, the titanium atoms are penta-coordinated in the zwitterionic complexes. The coordination environment around titanium displayed a distorted trigonal bipyramidal (TBP) geometry; the τ value for the five-coordinated metal complex was 0.84, approaching 1 as is expected for TBP geometry.

Scheme 2 Synthesis of “{P⁺[O₂]}H₂” 1 and 2.

Fig. 1 Molecular structure of 2.

To evaluate the catalytic activity of the titanium catalyst, the reaction of 1-octene with MAO as a cocatalyst was investigated (Table 1). At 18 °C, 5 μmol of the titanium catalyst and 5 mmol of MAO as a cocatalyst ([Al]/[Ti] = 1000) were reacted with 1 g of 1-octene. Notably, reaction in the polar solvent dichloromethane and reaction in bulk or in the non-polar solvent benzene yielded different results. In the case of reaction in dichloromethane, Size Exclusion Chromatography (SEC) measurements revealed that polymers were obtained at a yield of 90% (M_n = 2000, M_w/M_n = 2.4) (Run 3). Conversely, in the case of bulk reaction or reaction in nonpolar solvents, it was observed that neither 1-octene dimers nor any other oligomer were formed; only trimers of 1-octene and a small amount of poly(octene) were produced (Runs 1 and 2). The total ion chromatogram of GC-MS(EI) is depicted in Fig. 2. It is evident that no dimer was formed. Furthermore, two molecular ion peaks with the same molecular weight were observed, suggesting the formation of different isomers of the trimer. Additionally, the SEC chromatogram demonstrates the presence of the polymer component. The separation of the polymer from the trimer was attempted, and it was determined to be feasible owing to their differential solubility in acetone. The trimer was obtained as the acetone soluble component, and the trimerization selectivity of the titanium catalyst for the polymer was 80% under these bulk conditions.

Fig. 2 Chromatogram of GC-MS.

Table 1 Effect of reaction solvent on the reaction of 1-octene using 2/MAO

Run	Solvent	Trimer		Polymer
		Yield (mg)	Yield (tri%)	Yield (mg)	Yield (poly%)	M n kg mol^−1	M w/Mn
Ti = 5.0 μmol, MAO = 5.0 mmol, 1-octene = ∼1.0 g, solvent = 2 mL, reaction time = 24 h, reaction temperature = 18 °C.a Bimodal distribution.
1	(bulk)	170	14	41	3.6	69	5.4
2	Benzene	49	4.1	11	1.0	40	7.5
3	CH2Cl2	—	—	930	89	2^a	2.4^a

---

The ¹H NMR spectra of the obtained trimer isomer mixture are presented in Fig. 3. The structure of the trimer isomer was elucidated through an analysis of its terminal olefin, terminal methyl group, and allyl protons. The assignment of each proton signal was determined through analyses of the ¹³C NMR, DEPT135, HSQC, and HMBC spectra.

Fig. 3 ¹H NMR of trimer isomers.

NMR measurements indicated that the product of the trimerization reaction corresponds to the trimer structures shown on the right side of Fig. 3. Based on the trimerization products and on previous studies,^22,23,25,26 it was hypothesized that a reaction mechanism involving Ti(II) active species had occurred. The plausible reaction mechanism is shown in Scheme 3. Initially, the tail-to-tail oxidative-cyclization of two olefins into the divalent titanium active species transpired, and this was followed by oxidative coupling, whereby the titanium active species reverted to tetravalent titanium to form titanacyclopentane.

Scheme 3 Plausible mechanism for the trimerization of 1-octene.

Furthermore, it was postulated that the 1,2 insertion or 2,1 insertion of the third olefin would result in titanium becoming titanacycloheptane, and subsequently, β-hydrogen elimination and reductive elimination would occur to produce the four trimer isomers previously described, after which the divalent titanium active species would be regenerated. The mechanism of formation of the Ti(II) active species is shown in Scheme 4. The titanium complex initially reacted with MAO, which resulted in methylation and was followed by the removal of one methyl group to form a tetravalent titanium active species with an open coordination site. This tetravalent titanium active species was presumed to be involved in the polymerization reaction. Subsequently, the coordination of a single molecule of olefin into the open coordination site occurred. The generation of Ti(II) chemical species by two distinct pathways were anticipated, as depicted in the Scheme 4. In this context, when a titanium complex with a bisphenol-type ligand is combined with MAOs as a catalyst, the polymerization of α-olefins typically proceeds; for instance, it was reported the isotactic-specific polymerization of 1-hexene using {l,l′-(2,2′,3,3′-OC₁₀H₅SiMe₃)}₂Ti(CH₂Ph)₂ and MAOs.²⁴ The ligand employed in this study is a bisphenol ligand; however, the overall charge of the ligand is monoanionic, similar to Bercaw's monoanionic tridentate ligand. This similarity might have facilitated the formation of the titanium divalent complex, which might have functioned as a catalyst for the progressive trimerization of 1-octene.

Scheme 4 Plausible mechanism of change in the catalytically active species.

The limited production of poly(octene) is hypothesized to be associated with the Ti(IV) species depicted in Scheme 4. To elucidate the transformations that pertain to this catalytically active species, an experiment was conducted to monitor the reaction progression from 2 to 168 h. Within the initial 24 h of the reaction, a minimal quantity of polymer was generated, and no substantial increase was observed thereafter. Conversely, the trimerization reaction exhibited an increased rate over time and maintained activity for an extended period, which suggested that the titanium(II) active species formed a relatively stable complex. The absence of increased polymer yield after 24 h is postulated to have resulted from the decomposition of Ti(IV) involved in the polymerization, as the Ti(IV) catalytically active species transitioned to Ti(II) over time. When the reaction duration exceeded one week, a small quantity of the dimer was detected in the reaction product, analogous to the observations at elevated reaction temperatures. This phenomenon is attributed to the potential further decomposition of the tetravalent or divalent titanium active species.

The titanium catalyst exhibited high catalytic activity without trimer formation when the reaction was conducted in dichloromethane. However, in the case of bulk reaction with benzene, the same selective trimerization reaction occurred, and the catalytic activity was also low. This difference may be attributed to the polarity of the solvent.

When the reaction was carried out in polar solvents, the cationic titanium active species and anionic MAO formed solvent-separated ion pairs, and the olefin was more readily coordinated into the titanium active species, suggesting that the polymerization progressed rapidly. Conversely, when the reaction was conducted in a non-polar solvent such as bulk or benzene, the titanium active species and MAO formed a contact ion pair, and olefin coordination was less likely to occur. After a small amount of polymer was formed, the tetravalent titanium active species decomposed to divalent species, which subsequently participated in the trimerization reaction. It was hypothesized that this process had been catalyzed by the divalent active species.

Finally, we compared the catalytic activity with other titanium catalyst system for the trimerization of 1-octene. The turnover number (TON) of the tridentate FI catalyst²³ was 350, which is comparable to that of our phosphonium-bisphenolate-Ti catalyst (300). However, the turnover frequency (TOF) of the tridentate FI catalyst²³ was 88 h⁻¹, indicating greater activity than the 13 h⁻¹ observed for our catalyst. Although the TOF of our catalyst is lower than that of the reported tridentate FI catalyst, the present system demonstrates a distinct structural platform based on a phosphonium-bridged bisphenolate ligand, which enables regioselective trimerization with controlled isomer distribution. This unique feature, together with the observed ligand-dependent activity, offers a valuable basis for rational catalyst design. Ongoing efforts aim to exploit this tunability to improve both activity and selectivity.

In this study, the catalytic activity of a novel titanium complex activated by MAO was evaluated for octene reaction. The catalyst exhibited notable activity in dichloromethane, producing polyoctene in 90% yield, while in bulk or nonpolar solvents such as benzene, selective trimerization occurred, yielding four trimer isomers with 80% selectivity. One major isomer accounted for approximately 60% of the trimers. The catalytic activity increased over time, possibly indicating a Ti(IV) to Ti(II) transformation during activation. Although the exact nature of the active species remains unclear, this behavior warrants further mechanistic investigation. This is the first report of a titanium complex with a monoanionic bisphenolate ligand enabling selective trimerization. Further studies on related ligand frameworks are underway.

We thank Dr H. Sato (Rigaku Corporation) for assistance with single-crystal X-ray diffraction analyses.

Conflicts of interest

There are no conflicts to declare.

Data availability

Data supporting the findings of this study are available within the article and its SI. Detailed experimental procedures for the synthesis of the compounds, with their characterization by NMR spectroscopy, crystallographic data and parameters obtained from single-crystal X-ray diffraction analyses, are provided in the SI. The NMR data and assignments of the trimer obtained in the catalytic reactions are also included. See DOI: https://doi.org/10.1039/d5cc04541j.

CCDC 2413503 contains the supplementary crystallographic data for this paper.²⁷

Notes and references

1. J. W. Baek, J. H. Ko, J. H. Park, J. Y. Park, H. J. Lee, Y. H. Seo, J. Lee and B. Y. Lee, Organometallics, 2022, 41, 2455–2465.
2. J. W. Baek, Y. B. Hyun, H. J. Lee, J. C. Lee, S. M. Bae, Y. H. Seo, D. G. Lee and B. Y. Lee, Catalysts, 2020, 10, 990.
3. R. Benda, J. Bullen and A. Plomer, J. Synth. Lubr., 1996, 13, 41–57.
4. S. Ray, P. V. C. Rao and N. V. Choudary, Lubr. Sci., 2012, 24, 23–44.
5. I. Nifant’ev and P. Ivchenko, Polymers, 2020, 12, 1082.
6. M. Z. Saidi, A. Pasc, C. El Moujahid, N. Canilho, M. Badawi, C. Delgado-Sanchez, A. Celzard, V. Fierro, R. Peignier, R. Kouitat-Njiwa, H. Akram and T. Chafik, J. Colloid Interface Sci., 2020, 562, 91–101.
7. S. Soltanahmadi, T. Charpentier, I. Nedelcu, V. Khetan, A. Morina, H. M. Freeman, A. P. Brown, R. Brydson, M. C. P. Van Eijk and A. Neville, ACS Appl. Mater. Interfaces, 2019, 11, 41676–41687.
8. J. M. Liñeira del Río, E. R. López, J. Fernández and F. García, J. Mol. Liq., 2019, 274, 568–576.
9. K. Narita, Lubricants, 2014, 2, 11–20.
10. A. Rahbar, N. Bahri-Laleh and M. Nekoomanesh-Haghighi, Fuel, 2021, 302, 121111.
11. S. Echaroj, C. Asavatesanupap, S. Chavadej and M. Santikunaporn, Catalysts, 2021, 11, 1105.
12. M. Mashayekhi, S. Talebi, S. Sadjadi and N. Bahri-Laleh, Appl. Catal., A, 2021, 623, 118274.
13. C. Zhou, W. Li, M. Qiu, W. Li, H. Liu, H. Liu, K. Zhang and X. Chen, New J. Chem., 2021, 45, 9109–9117.
14. J. M. Hogg, A. Ferrer-Ugalde, F. Coleman and M. Swadźba-Kwasńy, ACS Sustainable Chem. Eng., 2019, 7, 15044–15052.
15. H. Sun, B. Shen, D. Wu, X. Guo and D. Li, J. Catal., 2016, 339, 84–92.
16. W. Wang, S. Jiang, Y. Shen, S. Xia, J. Xu and I. Lubr, Ind. Lubr. Tribol., 2016, 68, 52–56.
17. J. M. Hogg, F. Coleman, A. Ferrer-Ugalde, M. P. Atkins and M. Swadźba-Kwaśny, Green Chem., 2015, 17, 1831–1841.
18. J. M. Hogg, A. Ferrer-Ugalde, F. Coleman and M. Swadźba-Kwaśny, ACS Sustainable Chem. Eng., 2019, 7, 15044–15052.
19. S. Echaroj, M. Santikunaporn and S. Chavadej, Energy Fuels, 2017, 31, 9465–9476.
20. J. C. Gee, B. L. Small and K. D. Hope, J. Phys. Org. Chem., 2012, M25, 1409–1417.
21. P. Wasserscheid, S. Grimm, R. D. Köhn and M. Haufe, Adv. Synth. Catal., 2001, 343, 814–818.
22. Y. Suzuki, S. Kinoshita, A. Shibahara, S. Ishii, K. Kawamura, Y. Inoue and T. Fujita, Organometallics, 2010, 29, 2394–2396.
23. A. Sattler, J. A. Labinger and J. E. Bercaw, Organometallics, 2013, 32, 6899–6902.
24. A. Linden, C. J. Schaverien, N. Meijboom, C. Ganter and A. G. Orpen, J. Am. Chem. Soc., 1995, 117, 3008–3021.
25. D. K. Steelman, D. C. Aluthge, M. C. Lehman, J. A. Labinger and J. E. Bercaw, ACS Catal., 2017, 7, 4922–4926.
26. A. Sattler, D. C. Aluthge, J. R. Winkler, J. A. Labinger and J. E. Bercaw, ACS Catal., 2016, 6, 19–22.
27. T. Toda, Y. Cheng and K. Takenaka, CCDC 2413503: Experimental Crystal Structure Determination, 2025 DOI:10.5517/ccdc.csd.cc2m0fy1.

---