Nickel-catalyzed synthesis of 1,1-diborylalkanes from terminal alkenes

Organoboron compounds play an irreplaceable role in synthetic chemistry and the related transformations based on the unique reactivity of C–B bond are potentially the most efficient methods for the synthesis of organic molecules. The synthetic importance of multiboron compounds in C–C bond formation and function transformation reactions is growing and the related borations of activated or nonactivated alkenes have been developed recently. However, introducing directly two boron moieties into the terminal sites of alkenes giving 1,1-diborylalkanes in a catalytic fashion has not been explored yet. Here we describe a synthetic strategy of 1,1-diborylalkanes via a Ni-catalyzed 1,1-diboration of readily available terminal alkenes. This methodology shows high level of chemoselectivity and regioselectivity and can be used to convert a large variety of terminal alkenes, such as vinylarenes, aliphatic alkenes and lower alkenes, to 1,1-diborylalkanes.

Within this manuscript, Xiao and Fu describe the synthesis of gem-diborylalkanes from 1-alkenes using nickel-catalyzed alkene dehydrogenative borylation and hydroboration. This work represents the first example of the preparation of such synthetically valuable compounds with readily accessible terminal alkenes as the starting materials. The method works for both aromatic alkenes and aliphatic alkenes. The reaction with the latter is of particular interest because the dehydrogenative borylation of aliphatic alkenes is often more challenging than that with aromatic alkenes. Although the yields of 1,1-diborylalkane products in general are moderate to good, the present method provides a convenient approach to access gem-diborylalkanes from simple staring materials. With that being said, I support the publication of this manuscript after addressing the following points. 1) It is necessary to add the other synthetic routs to 1,1-diborylalkanes, for examples, the dual hydroboration of alkynes, the carbene transfer reactions by Jianbo Wang, and the direct double borylation by Hartwig. These could be added in Figure 1 and a concise introduction should be included accordingly.
2) Did the authors observe other boron-containing products, such as the monoborylalkanes or 1,2-diborylalkanes, in table 2 and table 3? 3) I do not know how to address it, but the current description, "modification of important molecules" seems to be odd. The definition of "important molecules" could vary. To some readers, some molecules shown in tables 2 and 3 can be more important than those in Figure 2. 4) A quantitative 13C NMR analysis of H/D scrambling is appropriate for the deuterium labeling experiment. 5) Adding a mechanistic cycle will be useful to the readers. 6) In figure 1, one carbon atom is missing in the diborylalkane product.
Reviewer #2 (Remarks to the Author): The paper by Fu and co-workers reports a Ni-catalyzed synthesis 1,1-diborylalkanes from terminal alkenes. The reaction employs 5 mol % of commercially available nickel salt and a phosphine ligand, in combination with an alkoxide base to deliver products efficiently and in a site-selective manner. The authors demonstrate good scope across broad substrate range. In particular, the reaction is tolerant of terminal alkenes (aryl and alkyl) bearing various functional groups such as ethers, amides, amines, and alkenes. Tolerance of the reaction to more complex as well as very simple substrates is highlighted through the catalytic diborylation of terminal alkenes within sugars, a liquid crystal, and ethylene. The authors also provide mechanistic experiments that show that two molecules of B2(pin)2 are required, and that deuterium scrambling takes place. The Supporting information adequately supports the claims in the paper, barring a few errors (see below).
While 1,1-diborylalkanes are emerging useful building blocks for chemical synthesis, many 1,1diborylalkanes can already be synthesized by alternative stoichiometric and catalytic methods. Such methods include: (1) alkylation of diborylmethane (e.g., Matteson, Morken, Meek, Shibata, etc.); (2) hydroboration of 1-alkenyl organoborons (Yun, Angew. Chem. Int. Ed. 2013, 52, 1); (3) isomerization/hydroboration (Chirik, Org. Lett., 2015, 17, 2716, (4) C-H functionalization (Chirik, JACS, 2016, 138, 766); 1,1-dihalide substitution (Morken, JACS, 2014, 136, 10581); (5) Diboration of a terminal alkene (not very efficient, Murakami, Angew. Chem. Int. Ed. 2015, 54, 12659). In the submitted manuscript by Fu and co-workers, the catalytic synthesis of 1,1diborylalkanes from terminal alkenes is clearly a significant advance in this area, however, it will not particularly alter the thinking of how people in the field utilize 1,1-diborylalkanes. For example, the reported method by Fu requires a terminal alkene, and as such cannot be used to synthesize the 1,1-diborylalkanes made by C-H functionalization or alkylation (see refs above). Hence, the methods are complementary, and one is not necessarily superior over the other. The paper also leaves a number of questions unanswered. For example: (1) What does being able to form a 1,1diboron from an alkene of a liquid crystal and sugars enable. Granted, they are more complicated substrates and interesting reactions but fall short of showing why the molecules efficiently made this way are significant. (2) Could the ethylene result be improved? The diboration of ethylene serves as a very nice and useful way to make 1,1-diborylethane, however, the reaction seems quite underwhelming. The borylation of ethylene produces a 42% combined yield (1:2.5 selectivity), which corresponds to a 12% yield of desired product.
In general, the reported catalytic method will impact the efficiency with which synthetic chemists prepare 1,1-diborylalkanes, but falls short in showcasing how 1,1-diborylalkanes synthesized by this approach prove advantageous in chemical synthesis. Overall, this is a nice paper describing practical catalytic method that will be of interest to the synthetic community but not necessarily the wider field, and unfortunately does not meet the level of importance to be published in Nature.
Below is a list of additional comments/suggestions/errors that should be taken into consideration by the authors. 1) SI Missing silica gel chromatography conditions for 1,1-diborylalkane products.
2) SI missing physical appearance of product (I.e., colorless oil, white solid, etc.) 3) SI missing, 1H NMR yields, isolated yields, and mass data for vinyl arenes. Only the spectral data is included. The missing information should be added. 4) Products only have three pieces of characterization data,where relevant IR for key functional groups (e.g., amides carbonyl), should be included. 5) All calculated HRMS values are 0.0004-0.0006 greater than what they should be. This is consistent throughout SI and seems like a simple error, which can easily be corrected. 6) In some instances, characterization data for the alkenes substrates is given but the synthetic procedures to make them is absent, unless a references are missing the experimental procedures used should be included. 7) Optical rotations are missing for chiral compounds 20, 24, 36, and 39. 8) Compound 41 has multiple signals in the 19F NMR likely due to chair conformations or diasteroisomers? A comment as to why there are some many signals should be added. 9) In the manuscript, all alkenes have a wedge, this should be removed. 10) 41 is an achiral compound and its name in the SI should not contain (R) stereochemical descriptor. 11) A proposed mechanism would be very helpful based on the mechanistic data provided. 12) For such a practical process, a gram-scale reaction to highlight practicality is noticeably absent from the paper.

To Reviewer #1:
"Within this manuscript, Xiao and Fu describe the synthesis of gem-diborylalkanes from 1-alkenes using nickel-catalyzed alkene dehydrogenative borylation and hydroboration. This work represents the first example of the preparation of such synthetically valuable compounds with readily accessible terminal alkenes as the starting materials. The method works for both aromatic alkenes and aliphatic alkenes. The reaction with the latter is of particular interest because the dehydrogenative borylation of aliphatic alkenes is often more challenging than that with aromatic alkenes. Although the yields of 1,1-diborylalkane products in general are moderate to good, the present method provides a convenient approach to access gem-diborylalkanes from simple staring materials. With that being said, I support the publication of this manuscript after addressing the following points."

Response:
We thank reviewer #1 for the positive comments and supporting the publication of this manuscript. Figure 1 and a concise introduction should be included accordingly."

"(1) It is necessary to add the other synthetic routes to 1,1-diborylalkanes, for examples, the dual hydroboration of alkynes, the carbene transfer reactions by Jianbo Wang, and the direct double borylation by Hartwig. These could be added in
Response: Synthetic routes to 1,1-diborylalkanes (Fig. 1 f) and related concise introduction were added in revised manuscript. We added references 46-54 in the revised manuscript.

"(2) Did the authors observe other boron-containing products, such as the monoborylalkanes or 1,2diborylalkanes, in table 2 and table 3?"
Response: We observe less than 5% terminal monoborylalkene and trace amount of 1,2diborylalkane. Though some 1,1-diboration reactions showed moderate yields, we are definitely sure that the 1,1-diboration products are absolutely main products in table 2 and table 3. "(3) I do not know how to address it, but the current description, "modification of important molecules" seems to be odd. The definition of "important molecules" could vary. To some readers, some molecules shown in tables 2 and 3 can be more important than those in Figure 2" Response: The description "modification of important molecules" was adjusted to "1,1diboration of sugar derivatives, LCs and lower alkenes" in the revised manuscript.

"(4) A quantitative 13 C NMR analysis of H/D scrambling is appropriate for the deuterium labeling experiment."
Response: A quantitative 13 C NMR analysis of compound 29-d 2 was measured (4s and 8s relaxation delay, 1024 scans). The peaks from benzylic and homobenzylic carbons should be split into triplet. However, we were failed to observe this splitting, because the signals were very weak. And we checked 1 H NMR and HRMS date carefully to make sure the correctness of deuterium atom migration.

"(5) Adding a mechanistic cycle will be useful to the readers."
Response: We conducted a set of mechanism experiments (e.g., crossover experiments and deuterium labelling experiments) and achieved some key mechanistic information (e.g., the origin of two boron motifs in products and the migration hydrogen atom in alkenes) (see SI, P65-P69). A relatively probable catalytic cycle was proposed based on our obtained information and we added the cycle in SI.
"(6) In Figure 1, one carbon atom is missing in the diborylalkane product." Response: We corrected this typo and checked the manuscript carefully to avoid alike typos.

To Reviewer #2:
"The paper by Fu and co-workers reports a Ni-catalyzed synthesis 1,1-diborylalkanes from terminal alkenes. The reaction employs 5 mol % of commercially available nickel salt and a phosphine ligand, in combination with an alkoxide base to deliver products efficiently and in a site-selective manner. The authors demonstrate good scope across broad substrate range. In particular, the reaction is tolerant of terminal alkenes (aryl and alkyl) bearing various functional groups such as ethers, amides, amines, and alkenes. Tolerance of the reaction to more complex as well as very simple substrates is highlighted through the catalytic diborylation of terminal alkenes within sugars, a liquid crystal, and ethylene. The authors also provide mechanistic experiments that show that two molecules of B 2 (pin) 2 are required, and that deuterium scrambling takes place. The Supporting information adequately supports the claims in the paper, barring a few errors (see below). While 1,1diborylalkanes are emerging useful building blocks for chemical synthesis, many 1,1-diborylalkanes can already be synthesized by alternative stoichiometric and catalytic methods. Such methods include: (1) alkylation of diborylmethane (e.g., Matteson, Morken, Meek, Shibata, etc.); (2) Hydroboration of 1-alkenyl organoborons (Yun, Angew. Chem. Int. Ed. 2013, 52, 1); (3) isomerization/hydroboration (Chirik, Org. Lett., 2015, 17, 2716; (4) C-H functionalization (Chirik, JACS, 2016, 138, 766). 1,1-dihalide substitution (Morken, JACS, 2014, 136, 10581); (5) Diboration of a terminal alkene (not very efficient, Murakami, Angew. Chem. Int. Ed. 2015, 54, 12659). In the submitted manuscript by Fu and co-workers, the catalytic synthesis of 1,1-diborylalkanes from terminal alkenes is clearly a significant advance in this area, however, it will not particularly alter the thinking of how people in the field utilize 1,1-diborylalkanes. For example, the reported method by Fu requires a terminal alkene, and as such cannot be used to synthesize the 1,1-diborylalkanes made by C-H functionalization or alkylation (see refs above). Hence, the methods are, and one is not necessarily superior over the other."
"The paper also leaves a number of questions unanswered. For example: (1) What does being able to form a 1,1-diboron from an alkene of a liquid crystal and sugars enable. Granted, they are more complicated substrates and interesting reactions but fall short of showing why the molecules efficiently made this way are significant."

Response:
The two newly formed C-B bonds made 41 possible to transform in a more enriched way to afford other possible LCs molecules via further C-B bond transformations. We noticed that double bond in the side carbon chain of liquid crystals is not a necessary functional group, and derivation on this double bond can derive many other useful LCs molecules. (Crystals, 2013, 3, 443. P467, table 14;ref. 59 Chart 1). The same diverse conversions could also be conducted on 36 and 39. We added above description in revised manuscript. If reviewer #2 thinks these diborylated molecules are not significant enough to be listed as a separate scheme, we are agreed to move these substrates to table 2.
"(2) Could the ethylene result be improved? The diboration of ethylene serves as a very nice and useful way to make 1,1-diborylethane, however, the reaction seems quite underwhelming. The borylation of ethylene produces a 42% combined yield (1:2.5 selectivity), which corresponds to a 12% yield of desired product." Response: Despite our best efforts (fine-tuning reaction parameters, using high pressure of ethylene), the selectivity was slightly improved to 1:2 with 45% combined yield (Details of optimization results were added in SI, P79). Fortunately, propylene, as one of the most important low alkenes, could be 1,1-diborylated to afford desired product in high yield (82%) (SI, P61). Diboration results of propylene and ethylene indicate that steric hindrance in alkenes plays a positive role in this site-selective matter. To better highlight practicability in low alkenes, propylene took the place of ethylene in revised manuscript (Fig. 2) and related work of ethylene was removed to SI.