Ti Group Metallocene-Catalyzed Synthesis of 1-Hexene Dimers and Tetramers

Abstract

1-Hexene transformations in the catalytic systems L₂MCl₂–XAlBuⁱ₂ (L = Cp, M = Ti, Zr, Hf; L = Ind, rac-H₄C₂[THInd]₂, M = Zr; X = H, Bu ⁱ) and [Cp₂ZrH₂]₂-ClAlR₂ activated by MMAO-12, B(C₆F₅)₃, or (Ph₃C)[B(C₆F₅)₄] in chlorinated solvents (CH₂Cl₂, CHCl₃, o-Cl₂C₆H₄, ClCH₂CH₂Cl) were studied. The systems [Cp₂ZrH₂]₂-MMAO-12, [Cp₂ZrH₂]₂-ClAlBuⁱ₂-MMAO-12, or Cp₂ZrCl₂-HAlBuⁱ₂-MMAO-12 (B(C₆F₅)₃) in CH₂Cl₂ showed the highest activity and selectivity towards the formation of vinylidene head-to-tail alkene dimers. The use of chloroform as a solvent provides further in situ dimer dimerization to give a tetramer yield of up to 89%. A study of the reaction of [Cp₂ZrH₂]₂ or Cp₂ZrCl₂ with organoaluminum compounds and MMAO-12 by NMR spectroscopy confirmed the formation of Zr,Zr-hydride clusters as key intermediates of the alkene dimerization. The probable structure of the Zr,Zr-hydride clusters and ways of their generation in the catalytic systems were analyzed using a quantum chemical approach (DFT).

1. Introduction

Alkene dimers and oligomers represent a large class of compounds that are used as comonomers in ethylene polymerization and as raw materials for the production of adhesives, surfactants, fragrances, synthetic lubricating fuel additives, etc. [1,2,3,4,5,6]. The following industrial processes have been successfully developed and implemented for the production of olefin oligomers: (i) oligomerization of ethylene in the presence of triethylaluminum with subsequent oxidation to higher alcohols (Ziegler–Alfol process), (ii) Philips process of ethylene oligo—and polymerization on chromium catalysts, (iii) Ni-catalyzed synthesis of linear α-olefins via ethylene oligomerization (shell higher olefin process (SHOP)), (iv) oligomerization of ethylene to linear C4–C10 olefins by α-select technology (Axens) or α-SABLIN technology (Sabic and Linde) [7]. Currently, some research groups are developing strategies for the production of highly efficient jet and diesel fuels via oligomerization of alkenes (1-butene, 1-hexene) synthesized from renewable plant raw materials [2,8].

Among the catalysts, Ti group metallocene complexes showed high efficiency in alkene di-, oligo-, and polymerization [9,10,11,12]. In this way, the process of zirconocene-catalyzed dimerization of terminal alkenes to give products with a vinylidene moiety >C=CH₂ is well-known (Scheme 1) [5,6,13,14,15,16,17,18,19,20].

Despite significant progress in this area, the mechanism of the dimerization reaction is not fully understood. Earlier, the hypothesis on the key role of the hydride cation [Cp₂ZrH]⁺ was put forward [14]. Subsequently, it was postulated that bimetallic hydride complexes L₂Zr(µ-H)₃AlR₂ can act as catalytically active species in the alkene dimerization [21,22]. Recently, we studied the systems Cp₂ZrCl₂-XAlBuⁱ₂ (X = H, Buⁱ) and [Cp₂ZrH₂]₂-ClAlR₂ (R = Me, Et, Buⁱ) activated by MMAO-12 (modified methylaluminoxane) or organoboron compounds and found new biszirconium hydride intermediates (Cp₂ZrH₂^.Cp₂ZrHCl^.ClAlR₂), which provide selective formation of head-to-tail dimerization products [19,20,23]. However, the questions on the possible structure of the key intermediates and their formation in the catalytic systems remain open.

Thus, our work was aimed at investigating the effect of transition metal η⁵-complexes and solvents (chlorine-containing solvents) on the activity, chemo- and regioselectivity of the systems L₂MCl₂–XAlBuⁱ₂ (L = Cp, M = Ti, Zr, Hf; L = Ind, rac-H₄C₂[THInd]₂, M = Zr; X = H, Buⁱ) and [Cp₂ZrH₂]₂-ClAlR₂ activated by MMAO-12, B(C₆F₅)₃, or (Ph₃C)[B(C₆F₅)₄] in alkene oligomerization. NMR studies and quantum chemical calculations were applied to establish the possible structure of the hydride intermediates formed in these systems. The results showed that the M,M-type bimetallic hydrides and their action mechanisms are unique for the dimerization reaction.

2. Results and Discussion

2.1. Transformations of 1-Hexene with Cp₂MY₂ (M = Ti, Zr, Hf; Y = H, Cl)-XAlBuⁱ₂ (X = H, Buⁱ) Catalytic Systems Activated by MMAO-12, B(C₆F₅)₃, or (Ph₃C)[B(C₆F₅)₄]

The conditions for the selective synthesis of vinylidene dimers in the presence of catalytic systems Cp₂ZrY₂ (Y = H, Cl)-XAlBuⁱ₂ (X = H, Buⁱ)-activator (Scheme 2, catalytic systems A and B), developed in our previous study, were taken as the starting point for the experiment (

Table 1. Catalytic activity and chemoselectivity of systems A and B in the reaction with 1-hexene.

Entry	Catalytic Systems			[Zr]: [Al]: [Activator]:[1-Hexene]	Solvent	T, °C	Time, min	Alkene Conversion, %		Product Composition, % h
	Zr Complex	OAC a	Activator						4	5	6			7
											n = 1	n = 2	n = 3
1 [19]	[Cp2ZrH2]2	ClAlBui2	MMAO-12	1:3:30:100	C6H5CH3	40	15	>99 b		86
2		ClAlBui2	MMAO-12	1:3:30:400	CH2Cl2	40	15	>99		98	1.3
3							180			98.2	1.8
4		ClAlBui2	MMAO-12	1:3:30:400	CHCl3	40	15	92		92
5							60	>99		98
6							105			97	1.3	1.6
7							180		9	12.6	1.4	2		75
8							960		17	-	1.4	2.6		79
9		-	MMAO-12	1:30:400	CHCl3	40	30	96	1	91	1
10					CH2Cl2			>99	<1	89	8
11					o-Cl2C6H4		60	86	1	73	10	<1
12							960	93	<1	74	13	3	1
13					(CH2Cl)2		60	90		83	5
14							960	99		91	4	1
15 [20]		ClAlBui2	B(C6F5)3	4:8:1:400	C6H6	40	90	81		81
16		-	B(C6F5)3	4:1:400	CHCl3	40	960	0
17					CH2Cl2		960	0
18		ClAlBui2		4:16:1:400	CHCl3		180	42		40	2
19							960	75		71	4
20					CH2Cl2		960	0
21 [20]		ClAlEt2	(Ph3C)[B(C6F5)4]	4:8:1:400	C6H6	40	90	91 c		86
22		-	(Ph3C)[B(C6F5)4]	4:1:400	CHCl3	40	960	0
23					CH2Cl2		960	0
24		ClAlBui2		4:16:1:400	CHCl3		180	44		9	20	-		15
25							960	81		17	36	13		15
26					CH2Cl2		180	>99		13	18			69
27 [19]	Cp2ZrCl2	HAlBui2	MMAO-12	1:3:30:100	C6H5CH3	40	15	>99 d		91
28		HAlBui2	MMAO-12	1:3:30:400	CH2Cl2	40	30	98		97	1
29							60	99		98	1
30		-		1:30:400	CH2Cl2	40	180	98		96	-	-
31							960	99		92	4	<1
32		HAlBui2	MMAO-12	1:3:30:1000	CH2Cl2	40	30	82		80	2
33							60	88		77	2			9
34		HAlBui2	MMAO-12	1:3:30:400	CHCl3	40	30	>99		98	2
35		HAlBui2	MMAO-12	1:3:30:1000	CHCl3	40	30	>99		90	7			3
36		HAlBui2	MMAO-12	1:3:30:400	CHCl3	20	180	>99		97	2
37		-	MMAO-12	1:30:400	CHCl3	40	180	>99	5	3	2			89
38		-	MMAO-12	1:10:400	CHCl3	40	960	91		91	<1
39		-	MMAO-12	1:30:400	o-Cl2C6H4	40	30	99		93	4	-
40							960	99		91	5	1
41		-	MMAO-12	1:30:400	(CH2Cl)2	40	30	99		95	2
42							960	>99		96	2
43 [20]		HAlBui2	B(C6F5)3	4:16:1:1000	C6H6	40	60	>99 e		93
44		HAlBui2	B(C6F5)3	4:16:1:1000	CH2Cl2	40	60	>99		99	1
45					CHCl3		60	83		82	1
46		-	B(C6F5)3	4:1:1000	CH2Cl2	40	960	0
47					CHCl3	40	960	0
48 [20]		HAlBui2	(Ph3C)[B(C6F5)4]	4:16:1:1000	C6H6	60	90	97 f		67
49		HAlBui2	(Ph3C)[B(C6F5)4]	4:16:1:1000	CH2Cl2	20	180	>99		92	6			2
50							960		1	55	8	3		33
51		HAlBui2	(Ph3C)[B(C6F5)4]	4:16:1:1000	CH2Cl2	40	30	>99 g	3	60	6	1		27
52							960			1	2	1	24	43
53		HAlBui2	(Ph3C)[B(C6F5)4]	4:16:1:1000	CHCl3	20	180	>99		76	8	2		13
54						40	180			15	8	11		65
55							960			8	8	12		72
56		-	(Ph3C)[B(C6F5)4]	4:1:1000	CH2Cl2	40	960	0
57					CHCl3		960	0

^a OAC—organoaluminum compound; ^b hydrometalation product, 4%; saturated dimer, 2%; unsaturated methylalumination product, 8% [19]; ^c hydrometalation product, 5% [20]; ^d hydrometalation product, 5%; saturated dimer, 1%; unsaturated methylalumination product, 2%; ^e hydrometalation product, 3%; trimers, 3%; ^f oligomers, 23%; ^g 6-mers (29%) are identified; ^h wt % in the reaction mixture, determined by GC–MS of deuterolysis or hydrolysis products (the details of the GC–MS analysis is shown in Supporting Information, Figures S28 and S29).

, entry 1) [19,20].

The reaction proceeds in toluene or benzene at 20–60 °C to give the target products in 81–98% yields within 5 to 150 min, depending on the type of activator (Table 1, entries 1, 15, 21, 27, and 43). The use of chlorine-containing solvents in these systems was found to accelerate the reaction. For example, the system [Cp₂ZrH₂]₂-ClAlBuⁱ₂-MMAO-12 with a molar ratio [Zr]:[Al]:[MMAO‑12]:[1-hexene] = 1:3:30:400 at a temperature of 40 °C in CH₂Cl₂ gives 1-hexene dimer in a yield of more than 98% by the 15th minute of the reaction (Table 1, entry 2). When the reaction is carried out in chloroform, the relative content of dimers decreases to 92% (entry 4). Under the same conditions, the dimers produced in the first minutes of the reaction serve as substrates for the subsequent dimerization to afford dimers of dimer 7 in 75–79% yields in 180 min (entries 7, 8, Figures S26 and S27). According to ¹H and ¹³C NMR data, the structures of these products correspond to those proposed earlier [16,22,24,25,26]. Dimerization of 5 occurs probably via participation of cationic species formed upon the reaction of CHCl₃ with ClAlR₂ or MMAO-12.

The replacement of MMAO-12 by B(C₆F₅)₃ in the system [Cp₂ZrH₂]₂-ClAlBuⁱ₂-activator leads to a significant reduction in substrate conversion. The reaction occurs only in chloroform, and the conversion of 1-hexene in 16 h is 75% with a 71% content of dimers in the product mixture (entries 18, 19). Unlike B(C₆F₅)₃, compound (Ph₃C)[B(C₆F₅)₄] can activate the system [Cp₂ZrH₂]₂-ClAlBuⁱ₂. Under these conditions, the conversion of 1-hexene in 3 h was higher in CH₂Cl₂ (99%, entry 26), and the reaction predominantly gave tetramer 7 in 69% yield. In contrast, in chloroform, the 1-hexene conversion was 81% in 16 h, and oligomeric products 6, including compound 7, prevailed in the product mixture (entries 24, 25).

The use of chlorine-containing solvents CH₂Cl₂ and CHCl₃ in the case of the system [Cp₂ZrH₂]₂-MMAO-12 at the ratio [Zr]:[MMAO-12]:[1-hexene] = 1:30:400 allows the reaction to be carried out in the absence of ClAlR₂ with a >99% conversion of 1-hexene and up to 89–91% yield of dimer 5 (entries 9, 10). In o-dichlorobenzene and 1,2-dichloroethane, the reaction proceeds to a lower substrate conversion and a lower yield of the dimeric product (entries 11–14). The composition of the reaction products formed in the system [Cp₂ZrH₂]₂-MMAO-12-1-hexene for 16 h in any of the solvents (CH₂Cl₂, CHCl₃, o-Cl₂C₆H₄, ClCH₂CH₂Cl) did not change significantly. Attempts to activate [Cp₂ZrH₂]₂ with boron-containing activators in the absence of ClAlR₂ were unsuccessful (entries 16, 17, 22, 23).

Thus, vinylidene dimer 5 is formed in a high yield and with high selectivity under the action of [Cp₂ZrH₂]₂-ClAlBuⁱ₂-MMAO-12 or [Cp₂ZrH₂]₂-MMAO-12 in CH₂Cl₂. Chloroform facilitates the formation of a non-classical tetramer of 1-hexene (7), which could be used as a component of lubricants [16,24,25,26].

The catalytic system based on Cp₂ZrCl₂, HAlBuⁱ_2, and MMAO-12 also provides vinylidene dimers 5 in chlorinated solvents. In CH₂Cl₂ and CHCl₃, the reaction proceeds in 30 min with the conversion of 1-hexene above 99% and the yield of dimers of 98% (entries 28, 34). Increasing 1-hexene initial concentration to 1000 equivalents leads to decreased alkene conversion to 82% and the yield of dimeric product to 80% in CH₂Cl₂. Under these conditions, the conversion and the yield of dimers are higher in CHCl₃ than in CH₂Cl₂ and amount to >99% and 90–98% in 30 min, respectively (entries 32, 35).

The reaction carried out in the presence of Cp₂ZrCl₂ and MMAO-12 at the ratio [Zr]:[MMAO-12]:[1-hexene] = 1:30:400 in chlorinated solvents (CH₂Cl₂, o-Cl₂C₆H₄, ClCH₂CH₂Cl) at 40 °C results in 91–96% yield of dimers (entries 30, 31, 39–42). Carrying out the reaction in chloroform gives dimers of dimer 7 in a yield of up to 89% (entry 37). A decrease in the relative amount of MMAO-12 in the system to [Zr]:[MMAO-12]:[1-hexene] = 1:10:400 provides the selective formation of dimerization products at the level of 91% within 16 h.

The highest yield of dimer 5 at the level of 99% is achieved by using a neutral boron-containing activator B(C₆F₅)₃ in the system Cp₂ZrCl₂-HAlBuⁱ₂ at the ratio [Zr]:[Al]:[B(C₆F₅)₃]:[1-hexene] = 4:16:1:1000 and in dichloromethane as a solvent (entry 44).

Activation of system Cp₂ZrCl₂-HAlBuⁱ₂ by (Ph₃C)[B(C₆F₅)₄] at a ratio [Zr]:[Al]:[(Ph₃C)[B(C₆F₅)₄] ]:[1-hexene] = 4:16:1:1000 and a temperature 40 °C affords alkene oligomers in both CH₂Cl₂ and CHCl₃ (entries 50–52, 54, 55). Selective formation of a dimer (92%, entry 49) occurs in CH₂Cl₂ at 20 °C, with the reaction time being up to 180 min.

Subsequently, the content of dimers decreases to 55%, and the amount of tetramer 7 increases to 33% (entry 50). Using chloroform as a solvent raises the yield of 7 to 65–72% (entries 54, 55).

The activity of Cp₂TiCl₂ (1a) and Cp₂HfCl₂ (1c) in the studied systems was lower than that of Cp₂ZrCl₂ (1b). Indeed, the conversion of 1-hexene in CH₂Cl₂ at 40 °C was 80% in 60 min of the reaction in the case of the Ti complex (

Table 2. Catalytic activity and chemoselectivity of systems Cp₂MCl₂-HAlBuⁱ₂-activator (M = Ti, Hf) in the reaction with 1-hexene.

Entry	Catalytic Systems			[Zr]: [Al]: [Activator]:[1-Alkene]	Solvent	T, °C	Time, min	Alkene Conversion, %		Product Composition, % d
	Zr Complex	OAC	Activator						4	5	6			7
											n = 1	n = 2	n = 3
1	Cp2TiCl2 a	HAlBui2	MMAO-12	1:3:30:400	CH2Cl2	40	60	80		20	34	4		15
2							180	91		17	44	5		20
3					CHCl3		60	93		16	36	6	5	28
4							960	>99	5	4	41	11	7	30
5	Cp2HfCl2	HAlBui2	MMAO-12	1:3:30:400	CH2Cl2	40	120	84	2	53	20	6	2
6							180	94	2	54	21	8	2
7							960	96	-	59	22	8	3	4
8			MMAO-12	1:3:30:400	CHCl3	40	60	60	-	42	15	3		1
9							120	74 b	-	30	16	4		5
10							960	>99 c	6	17	17	15		9
11	Cp2TiCl2	HAlBui2	(Ph3C)[B(C6F5)4]	4:16:1:400	CH2Cl2	40	960	0
12					CHCl3		960	0
13	Cp2HfCl2	HAlBui2	(Ph3C)[B(C6F5)4]		CH2Cl2	40	960	0
14					CHCl3		960	0

^a regioselectivity is significantly reduced due to double bond migration; ^b byproducts of toluene mono- (13%), di- (4%) and tri- (2%) alkylation with 1-hexene are formed; ^c byproducts of toluene mono- (22%), di- (7%), and tri- (4%) alkylation with 1-hexene are formed; ^d wt % in the reaction mixture, determined by GC–MS of deuterolysis or hydrolysis products (the details of the GC–MS analysis is shown in Supporting Information).

, entries 1) and 84% in 120 min for the Hf complex (entries 5). The use of chloroform as a solvent increases substrate conversion to 93% for Cp₂TiCl₂ and decreases the conversion to 60% for Cp₂HfCl₂ (entries 3,8). Moreover, these catalysts are significantly inferior in chemo- and regioselectivity to zirconocene dichloride. Thus, in the presence of Cp₂TiCl₂, the yield of oligomeric products 6 and tetramer 7, which represent a mixture of regioisomers, increases in total to 89%. The use of Cp₂HfCl₂ in CH₂Cl₂ leads to increased the content of vinylidene oligomers up to 28–37%. In chloroform, the yield of regioisomeric oligomers also increases in the first 60 min of the reaction (entry 8). Further, the appearance of products of toluene mono-, di-, and trialkylation with 1-hexene was found (entries 9,10). It should be noted that in the presence of HAlBuⁱ₂, upon replacing MMAO-12 with boron-containing activators, the systems based on Cp₂TiCl₂ and Cp₂HfCl₂ completely lost their activity (Table 2, entries 11–14).

The effect of the ligand structure of zirconocenes on the activity of catalytic systems and the reaction route was also studied. For this purpose, we tested Zr η⁵-complexes (C₅Me₅)₂ZrCl₂ (1d), Ind₂ZrCl₂ (1e), and rac-H₄C₂(THInd)₂ZrCl₂ (1f) (Scheme 2) as components of the systems L₂ZrCl₂-HAlBuⁱ₂-MMAO-12 at molar ratio [Zr]:[Al]:[MMAO-12]:[1-hexene] = 1:3:30:400 and temperature of 40 °C. It turned out that (C₅Me₅)₂ZrCl₂ does not lead to the formation of the target products either in CH₂Cl₂ or in CHCl₃ (

Table 3. Catalytic activity and chemoselectivity of systems L₂ZrCl₂-HAlBuⁱ₂-activator in the reaction with 1-hexene.

Entry	Catalytic Systems			[Zr]: [Al]: [activator]:[1-alkene]	Solvent	T, °C	Time, min	Alkene Conversion, %		Product Composition, % e
	Zr Complex	OAC	Activator						4	5	6
											n = 1	n = 2	n = 3
1	(C5Me5)2ZrCl2	HAlBui2	MMAO-12	1:3:30:400	CHCl3	40	30	>99 a
2					CH2Cl2		30
3	Ind2ZrCl2	HAlBui2	MMAO-12	1:3:30:400	CH2Cl2	40	30	19		11	5	2	1
4							180	48		20	19	6	3
5							960	85 b		9	18	18	16
6					CHCl3	40	180	>99		13	51	24	12
7							960	>99 c		16	21	19	18
8	rac-H4C2[THInd]2ZrCl2	HAlBui2	MMAO-12	1:3:30:400	CHCl3	40	30	>99 d		39	38	16	4
9					CH2Cl2		960	0

^a 1-hexene is completely consumed for toluene alkylation; ^b 6-mers (15%) and 7-mers (9%); ^c 6-mers (17%) and 7-mers (9%); ^d 6-mers (3%); ^e wt % in the reaction mixture, determined by GC–MS of deuterolysis or hydrolysis products (the details of the GC–MS analysis is shown in Supporting Information).

, entries 1,2). In this case, 1-hexene was consumed for toluene alkylation (toluene is present in the system as a solvent for MMAO-12). The complex Ind₂ZrCl₂ showed lower activity than Cp₂ZrCl₂ (Table 3, entries 3–7) and shifted the reaction route toward the oligomerization (up to 87%) both in CH₂Cl₂ and CHCl₃. Ansa-zirconocene rac-H₄C₂(THInd)₂ZrCl₂ was inactive in CHCl₃ (entry 9); however, it catalyzed alkene oligomerization in CH₂Cl₂ (1-hexene conversion of >99%, entry 8). Thus, the zirconocene ligand structure significantly affects the course of the reaction, and some tested complexes mainly provided the formation of oligomeric products.

2.2. NMR Study of Intermediate Structure in MMAO-12-Activated Systems Cp₂ZrY₂ (Y = H, Cl)-OAC in Chlorinated Solvents

The effect of chlorinated solvents (CD₂Cl₂ and CDCl₃) on the structure of the intermediates responsible for the alkene dimerization in the catalytic systems [Cp₂ZrH₂]₂-MMAO-12 was studied by NMR. The addition of MMAO-12 to a solution of [Cp₂ZrH₂]₂ in CD₂Cl₂ gives rise to a triplet at −5.95 ppm (J = 17.0 Hz) in the ¹H NMR spectra (Figure 1c, Figures S21–S23). In the COSY HH spectrum, the triplet correlates with the signal at −0.47 ppm, which is superimposed with the region of resonance lines of methyl group protons of MMAO-12 (−0.82–0.32 ppm). The spectra also exhibited signals for the Cp rings at 6.16 ppm (108.4 ppm in ¹³C NMR spectra), 6.39 ppm (112.8 ppm in ¹³C NMR spectra), and 6.61 ppm (116.2 ppm in ¹³C NMR spectra). The NOESY spectra showed the cross-peaks of the Cp-ring signal at 6.16 ppm with the upfield signals at −5.95 ppm and −0.47 ppm. Relying on these results and the data from previous studies [19,20], these signals were assigned to the biszirconium trihydride complex 9. The signals at 6.61 ppm and 6.39 ppm correspond to the Cp rings of Cp₂ZrCl₂ and Cp₂ZrMeCl (11), respectively [27].

As shown in Scheme 3, the system [Cp₂ZrH₂]₂-MMAO-12 in CD₂Cl₂ produces complex 9 via in situ formation of Cp₂ZrHCl and Cp₂ZrCl₂ by the reaction of zirconocene dihydride with a chlorine-containing solvent [28]. Residual AlMe₃ present in the MMAO-12 solution reacts with Cp₂ZrCl₂ to give methyl chloride complex 11 and ClAlMe₂. The reaction of Cp₂ZrHCl with the starting Cp₂ZrH₂ and ClAlMe₂ makes it possible to selectively obtain complex 9 and then high-molecular-weight associates with MMAO-12 (9·MAO), which are active in the alkene dimerization [19,20].

The ¹H and ¹³C NMR spectra of the system [Cp₂ZrH₂]₂-ClAlBuⁱ₂ (Figure 1a, Figures S14 and S15) in CD₂Cl₂ exhibited signals of previously described intermediates [19,23,28,29]: complex 8 (broadened signals of hydride atoms at δ_H −0.86 and −1.91 ppm), dimeric complex 10 (broadened signals of hydride atoms at δ_H −1.42 and −2.38 ppm), and biszirconium trihydride complex 9 (doublet signal at δ_H −0.70 ppm (²J = 17.0 Hz) and triplet at −5.87 ppm). The addition of MMAO-12 to the equilibrium mixture of the complexes leads to the vanishing of signals of complex 10 from the ¹H NMR spectrum and the appearance of an additional triplet at −6.01 ppm attributable to the adduct 9·MAO (Figure 1b, Figures S16–S18).

A similar picture is observed in the ¹ H and ¹³C NMR spectra of Cp₂ZrCl₂-HAlBuⁱ₂ in CD₂Cl₂ (Figures S5 and S6). When MMAO-12 is added to the system, the signals of the dimeric complex 10 also disappear, the intensity of the signals of the trihydride 8 decreases, and additional triplets of 9·MAO appear in the upfield region at −6.21 ÷ −5.92 ppm (Figures S7 and S8).

The reaction of [Cp₂ZrH₂]₂ with MMAO-12 in either CDCl₃ or CD₂Cl₂ gives complexes Cp₂ZrCl₂ (δ_Cp 6.76 ppm) and Cp₂ZrMeCl (δ_Cp 6.58 ppm) (Figure 2c, Figures S19 and S20). Moreover, the ¹ H NMR spectrum showed a significant broadening of the signals of the hydride atom at −6.12 ppm and Cp rings of the heavy adduct 9·MAO, which disappeared from the spectra due to the precipitation of the heavy fraction to the bottom of the NMR tube.

In the reaction of Cp₂ZrCl₂ with HAlBuⁱ₂ (1:4) in CDCl₃, only the Zr,Al-trihydride complex 8 was identified (Figure 2a, Figures S1 and S2). The signals of complexes 9 and 9·MAO appeared in the ¹ H NMR spectra after the addition of MMAO-12 and 1-hexene to the system (Figure 2b, Figures S3 and S4). The formation of complexes 8–10 was observed in the reaction of zirconocene dihydride with ClAlBuⁱ₂ taken in a 1:2 ratio in CDCl₃ (Figure S9). The addition of MMAO-12 to this system results in the disappearance of signals of complex 10, broadening of complex 8 signals, and the appearance of an additional broadened multiplet at −6.03 ppm, corresponding to the heavy adduct 9·MAO (Figure S11). NMR monitoring of the system’s reaction with 1-hexene showed the consumption of 9·MAO adduct and the accumulation of the vinylidene dimer (Figures S24 and S25).

Thus, biszirconium complex 9 is readily formed in the systems Cp₂ZrCl₂-HAlBuⁱ₂, [Cp₂ZrH₂]₂-MMAO-12, and [Cp₂ZrH₂]₂-ClAlBuⁱ₂ both in CD₂Cl₂ and in CDCl₃. This intermediate reacts with methylaluminoxane to give a heavy adduct, which selectively provides vinylidene dimers of 1-hexene.

2.3. DFT Study of the Structure of Biszirconium Complex 9

To refine the structure of complex 9, probable structures of cyclic isomers 9a–9d were optimized using the PBE/3ζ quantum chemical method [30,31,32]. Their structure is in line with the obtained NMR spectral data on the ratio of the signal intensities of the constituent moieties and the symmetry of the complex (Scheme 4). As follows from Scheme 4, the structures of the complexes significantly differ from one another. For example, in complex 9c, all three hydrides are located between the zirconium atoms and form a trihydride bridge, while in molecule 9a, there is only one bridging H atom. If the optimized structures of the two most energetically favorable hydride complexes 9a and 9c (Figure 3) are considered in detail, the length of the Zr–H bond varies, which may cause differences in the reactivity of the studied complexes. Thus, the lengths of both Zr–H bonds in the Zr–H–Zr moiety of complex 9a is 2.09 Å, while the lengths of two terminal Zr–H bonds are d_Zr1-H = d_Zr2-H = 1.83 Å. It is comparable with the Zr–H bond length in Cp₂ZrHCl calculated by the same method (d_Zr-H = 1.84 Å).

In isomer 9c, all three hydrogen atoms are inside the biszirconium cage. Meanwhile, the lengths of bridging Zr–H bonds are also increased compared to those in Cp₂ZrHCl, but are not equivalent to each other: d_Zr1-H1 = d_Zr2-H1 = 2.01 Å, d_Zr1-H2 = d_Zr2-H3 = 1.97 Å, d_Zr1-H3 = d_Zr2-H2 = 1.99 Å. Thus, one hydrogen atom is equidistant from both zirconium atoms, while the other two H atoms, forming an “inner” bridge, are characterized by some displacement towards one of the zirconium atoms. It is quite obvious that structural differences also determine energy differences (

Table 4. Relative thermodynamic parameters of isomeric complexes 9.

Complex	∆E, Hartree	∆EZPVE, Hartree	∆H, kcal/mol	∆G, kcal/mol	T∆S, cal/mol
9a	0.000000	0.000000	0.0	0.0	2475.2
9b	0.013103	0.012948	8.2	9.6	1139.2
9c	0.003022	0.003398	2.4	3.3	1517.2
9d	0.022146	0.023111	14.4	16.9	0.0

, Table S1).

The most thermodynamically stable complex is 9a, in which the chlorine atoms are part of the Zr–Cl–Al moieties. Isomer 9b is higher in energy by 3.3 kcal/mol. The presence of bulky Cl atoms in the inner bridge of the biszirconium cage of structure 9d makes this complex least thermodynamically stable relative to other compounds.

To identify the structure of the complex observed by NMR spectroscopy, we compared theoretical and experimental NMR chemical shifts of hydride atoms in each of them (

Table 5. Calculated and experimental chemical shifts of isomeric complexes 9a–d (numbering according to Scheme 4).

Complex	δ(H 1), ppm	δ(H 2), ppm	δ(H 3), ppm	δ(Cp), ppm
9a	−4.2	2.9	2.9	5.9
9b	−2.3	1.8	1.8	6.2
9c	−3.6	3.7	3.7	6.2
9d	0.7	0.8	0.8	6.0
9e	−1.6	−0.4	−0.4	6.1
9 (experimental)	−5.9	−0.7	−0.7	6.1

). It was found that in complex 9a, the hydrogen atom of the Zr–H–Zr moiety is significantly shielded (δH¹ = −4.2 ppm), while two other hydride atoms experience a de-shielding effect (δH² = δH³ = 2.9 ppm). Thus, the difference in chemical shifts between the considered hydrogen atoms is ∆δ = 7.1 ppm, which is in good agreement with the experimental data. A similar trend is also observed for complex 9c, in which one of the hydride atoms of the inner Zr–H–Zr bridge should be in the upfield region of the ¹H NMR spectrum. As follows from Table 5, the calculated NMR data for complexes 9b and 9d are in poor agreement with the NMR spectral parameters observed. It should be noted for comparison that the calculated chemical shifts of the hydride atoms of the open structure 9e proposed earlier [23] also do not agree well with the NMR experiment. As a result of a comprehensive analysis of chemical shifts and relative Gibbs energy, the complex 9a was proposed as the most probable structure.

3. Materials and Methods

3.1. General Procedures

All operations for organometallic compounds were performed under argon according to the Schlenk technique. Zirconocene 1 was prepared from ZrCl₄ (99.5%, Merck, Darmstadt, Germany) using the standard procedure [33]. The synthesis of [Cp₂ZrH₂]₂ (2) from (1) was carried out as described previously [28,34]. The solvents (CHCl₃, CH₂Cl₂, ClCH₂CH₂Cl) were distilled from P₂O₅ immediately before use. The solvent o-Cl₂C₆H₄ (anhydrous, 99%, Merck) was used without further purification. Commercially available Cp₂HfCl₂ (98%, Strem, Newburyport, MA, USA), Cp₂TiCl₂ (99%, Strem), (C₅Me₅)₂ZrCl₂ (97%, Acros), rac-C₂H₄(THInd)₂ZrCl₂ (97%, Merck), HAlBuⁱ₂ (99%, Merck), ClAlBuⁱ₂ (97%, Strem), MMAO-12 (7% wt Al in toluene, Merck), (Ph₃C)[B(C₆F₅)₄] (97%, Abcr, Karlsruhe, Germany), B(C₆F₅)₃ (95%, Merck), and 1-hexene (97%, Fisher Scientific, Pittsburgh, Pennsylvania, USA) were used for the reactions.

CAUTION: pyrophoric nature of aluminum alkyl and hydride compounds requires special safety precautions in their handling.

¹H and ¹³C NMR spectra were recorded on a Bruker AVANCE−400 spectrometer (400.13 MHz (¹H), 100.62 MHz (¹³C)) (Bruker, Rheinstetten, Germany). As the solvents and the internal standards, CD₂Cl₂ and CDCl₃ were employed. 1D and 2D NMR spectra (COSY HH, HSQC, HMBC, NOESY) were recorded using standard Bruker pulse sequences.

The products were analyzed using a GC–MS-QP2010 Ultra gas chromatograph–mass spectrometer (Shimadzu, Tokyo, Japan) equipped with the GC−2010 Plus chromatograph (Shimadzu, Tokyo, Japan), TD-20 thermal desorber (Shimadzu, Tokyo, Japan), and an ultrafast quadrupole mass-selective detector (Shimadzu, Tokyo, Japan). Details on the GC–MS analysis of dimers and oligomers are given in the SI.

3.2. Reaction of [Cp₂ZrH₂]₂ with ClAlBuⁱ₂, Activators (MMAO-12, (Ph₃C)[B(C₆F₅)₄] or B(C₆F₅)₃) and 1-Hexene

A flask with a magnetic stirrer was filled under argon with 10 mg (0.045 mmol) [Cp₂ZrH₂]₂ (all ratios are given relative to the monomer), 0.026 mL (0.135 mmol) ClAlBuⁱ₂, 0.58 mL (1.35 mmol) MMAO-12, 2.24 mL or 5,63 mL (17.9 or 45 mmol) 1-hexene, and 2 mL solvent (CHCl₃, CH₂Cl₂, o-Cl₂C₆H₄, ClCH₂CH₂Cl). For organoboron activators (Ph₃C)[B(C₆F₅)₄] or B(C₆F₅)₃ the following amounts were used: 10 mg (0.045 mmol) [Cp₂ZrH₂]₂, 0.018–0.036 mL (0.10–0.18 mmol) ClAlBuⁱ₂, 6 mg (0.011 mmol) B(C₆F₅)₃ or 10 mg (0.011 mmol) (Ph₃C)[B(C₆F₅)₄], 0.6 mL (4.5 mmol) 1-hexene, and 2 mL solvent (CHCl₃, CH₂Cl₂, o-Cl₂C₆H₄, ClCH₂CH₂Cl). The reaction was carried out with stirring at a temperature of 40 °C. After 5, 10, 15, 30, 60, 90, 120, 180, and 960 min, samples (0.1 mL) were syringed into tubes filled with argon and decomposed with 10% HCl at 0 °C. The products were extracted with CH₂Cl₂, and the organic layer was dried over Na₂SO₄. The yield of dimers or oligomers was determined by GC/MS.

3.3. Reaction of L₂MCl₂ (1a–f) with HAlB ⁱ₂, MMAO-12, (Ph₃C)[B(C₆F₅)₄] or B(C₆F₅)₃ and 1-Hexene

A flask with a magnetic stirrer was filled under argon with 0.034 mmol (9–15 mg) L₂MCl₂, 0.019 mL (0.108 mmol) HAlBuⁱ₂, 0.44 mL (1.02 mmol) MMAO-12, 1.7 or 4.2 mL (13.6 or 34 mmol) 1-hexene, and 2 mL solvents (CHCl₃, CH₂Cl₂, o-Cl₂C₆H₄, ClCH₂CH₂Cl). For organoboron activators (Ph₃C)[B(C₆F₅)₄] or B(C₆F₅)₃ the following amounts were used: 0.034 mmol (9–15 mg) L₂MCl₂, 0.018–0.036 mL (0.10–0.18 mmol) HAlBuⁱ₂, 5 mg (0.0085 mmol) B(C₆F₅)₃ or 10–16 mg (0.011–0.017 mmol) (Ph₃C)[B(C₆F₅)₄], 1 or 1.7 mL (8.5 or 13.6 mmol) 1-hexene, and 2 mL solvents (CHCl₃, CH₂Cl₂, o-Cl₂C₆H₄, ClCH₂CH₂Cl). The reaction was carried out with stirring at a temperature of 40 °C. After 15, 20, 30, 60, 90, 120, 180, and 960 min, samples (0.1 mL) were syringed into tubes filled with argon and decomposed with 10% HCl or DCl at 0 °C. The products were extracted with CH₂Cl₂, and the organic layer was dried over Na₂SO4. The yields of the products were determined by GC/MS.

3.4. NMR Study of the Reaction of [Cp₂ZrH₂]₂ with ClAlR₂ and MMAO-12

An NMR tube was charged with 0.045 mmol (10 mg) [Cp₂ZrH₂]₂ in an argon-filled glovebox. The tube was cooled to 0 °C, and 0.045–0.2 mmol (8–35.3 mg) ClAlBuⁱ₂ was added dropwise. Then CDCl₃ or CD₂Cl₂ was added. The mixture was stirred, and the formation of complexes 8–10 was monitored by NMR at room temperature. Then 0.135–0.9 mmol (60–300 mg) MMAO-12 was added, and the formation of intermediates was monitored by NMR at room temperature (Figures S9–S18).

In the case of the system [Cp₂ZrH₂]₂-MMAO-12, the number of components was the same as described above (Figures S19–S23). The deuterated solvent was added to the NMR tube as the last component.

3.5. NMR Study of the Reaction of Cp₂ZrCl₂ with HAlBuⁱ₂ and MMAO-12

An NMR tube was charged with 0.034 mmol (10 mg) Cp₂ZrCl₂ in an argon-filled glovebox. The tube was cooled to 0 °C, and 0.068–0.136 mmol (10–19.3 mg) HAlBuⁱ₂ was added dropwise. Then CDCl₃ or CD₂Cl₂ was added. The mixture was stirred, and the formation of complexes 8–10 was monitored by NMR at room temperature. Then 0.135–0.9 mmol (60–300 mg) MMAO-12 was added, and the formation of intermediates was monitored by NMR at room temperature (Figures S1–S8).

3.6. Computational Details

DFT calculations were carried out in Priroda-06 program [32,35]. Geometry optimization, vibrational frequency analysis, and calculation of absolute chemical shielding, entropy, and thermodynamic corrections to the total energy of the compounds were carried out using the Perdew−Burke−Ernzerhof (PBE) functional [30]. PBE functional was used in combination with a 3ζ basis set [31]. The electronic configurations of the molecular systems were described by the orbital basis sets of contracted Gaussian-type functions of size (5s1p)/[3s1p] for H, (11s6p2d)/[6s3p2d] for C, (15s11p2d)/[10s6p2d] for Al and Cl, and (20s16p11d)/[14s11p7d] for Zr, which were used in combination with the density-fitting basis sets of uncontracted Gaussian-type functions of size (5s2p) for H, (10s3p3d1f) for C, (14s3p3d1f1g) for Al and Cl, and (22s5p5d4f4g) for Zr. No symmetry of internal coordinate constraints was applied during optimizations. Thermodynamic parameters were determined at 298.15. Normal-mode vibrational frequency analysis was performed to confirm minima structures. Computations of absolute chemical shielding, σ, were carried out in the GIAO approach [36,37]. Chemical shifts were calculated as δ = σ_TMS − σ, where σ_TMS was the calculated shielding constant of tetramethylsilane. For comparison, structural and energetic parameters of complexes 9a–e were calculated using Gaussian 09 [38] at the PBE0 level of theory [39] employing the def2-TZVP basis set [40,41] with and without Grimme’s D3(0) empirical dispersion correction (GD3) [42] (Tables S2 and S3). Moreover, the method was successfully employed to calculate chemical shifts of Au hydrides [43]. Calculated chemical shifts obtained for 9a by two methods was comparable (PBE0/def2-TZVP: δH¹ = −3.5 ppm, δH² = 3.1 ppm, δH³ = 3.1 ppm). Therefore, the PBE/3ζ method was used for the calculation of the chemical shifts of other complexes. As follows from Table S3 (Supporting Information), using the GD3 correction does not lead to significant changes in the energy of the studied complexes. Furthermore, we carried out the study on the solvent effect (CH₂Cl₂ and CHCl₃) using the conductor-like polarizable continuum model (CPCM) [44,45]. The data indicated minor energy changes (Supporting Information, Tables S4 and S5).

The program ChemCraft [46] was used to visualize obtained quantum chemical data. The energy at 0 K, the ZPVE correction, the enthalpy, the Gibbs free energy in gas, the temperature multiplied by the entropy, and Cartesian coordinates for all optimized structures are given in the Supporting Information.

4. Conclusions

As a result of studying the catalytic transformations of 1-hexene under the action of Ti group metallocenes, organoaluminum compounds, and activators MMAO-12, B(C₆F₅)₃ or (Ph₃C)[B(C₆F₅)₄] in chlorinated solvents, it was found that systems based on Zr complexes [Cp₂ZrH₂]₂-ClAlBuⁱ₂-MMAO-12, Cp₂ZrCl₂-HAlBuⁱ₂-MMAO-12 in CH₂Cl₂ selectively afford dimeric products in high yields. The use of CHCl₃ as the solvent facilitates the formation of non-classical tetramers of 1-hexene, the products of dimer dimerization.

NMR studies showed that systems Cp₂ZrCl₂-HAlBuⁱ₂-MMAO-12, [Cp₂ZrH₂]₂-MMAO-12, and [Cp₂ZrH₂]₂-ClAlBuⁱ₂-MMAO-12 provide the adduct of biszirconium complex (Cp₂ZrH₂∙Cp₂ZrHCl∙ClAlR₂) with an activator both in CD₂Cl₂ and CDCl₃. The adduct reacts with 1-hexene and produces a vinylidene dimer.

The probable structure of the key biszirconium hydride complex was proposed based on a comparison of experimental and theoretical NMR data and estimation of the thermodynamical stability of the complexes. These studies are necessary to further understand the mechanism of key intermediate activation by methylaluminoxane or organoboron compounds for selective alkene dimerization.