An Experimental and Theoretical Study into the Facile, Homogenous (n-heterocyclic Carbene)2-pd(0) Catalyzed Diboration of Internal and Terminal Alkynes

(2016) An experimental and theoretical study into the facile, homogenous (N-Heterocyclic Carbene)2-Pd(0) catalyzed diboration of internal and terminal alkynes. This document is made available in accordance with publisher policies and may differ from the published version or from the version of record. If you wish to cite this item you are advised to consult the publisher's version. Please see the URL above for details on accessing the published version. Copyright and all moral rights to the version of the paper presented here belong to the individual author(s) and/or other copyright owners. To the extent reasonable and practicable, the material made available in SRO has been checked for eligibility before being made available. Copies of full text items generally can be reproduced, displayed or performed and given to third parties in any format or medium for personal research or study, educational, or not-for-profit purposes without prior permission or charge, provided that the authors, title and full bibliographic details are credited, a hyperlink and/or URL is given for the original metadata page and the content is not changed in any way. An experimental and theoretical study into the facile, homogenous (N-heterocyclic carbene) 2-Pd(0) catalyzed diboration of internal and terminal alkynes † a PdĲITMe) 2 (PhCCPh) acts as a highly reactive pre-catalyst in the unprecedented homogenous catalyzed diboration of terminal and internal alkynes, yielding a number of novel and known syn-1,2-diborylalkenes in a 100% stereoselective manner. DFT calculations suggest that a similar reaction pathway to that proposed for platinum phosphine analogues is followed, and that destabilization of key intermediates by the NHCs is vital to the overall success for the palladium-catalyzed B–B addition to alkynes.


Introduction
The transition metal catalysed π-insertion of homo and hetero element-element (E-E′) bonds into alkynes provides the most atom economical route for the stereoselective synthesis of triand tetrasubstituted alkenes. 1 Among the various E-E′ reagents used in such transformations, B-B bonds (diborons) in particular have attracted substantial interest. 2The resulting 1,2-diboryl alkenes, due to their participation in Suzuki-Miyaura cross-coupling, 3 are recognized as important building blocks in, for example, the synthesis of pharmaceuticals, 4 chirotopical devices 5 and optically/electronically active polymeric materials. 6A number of transition metals have been utilized in both homogenous and heterogenous catalytic addition of B-B bonds (diboration) to alkynes including cobalt, 7 copper, 8 iridium, 9 rhodium, 9 iron, 10 platinum, 11 and palladium. 12To date, platinum is by far the most effective and widely studied; 13 this is attributed to the facile cleavage of the B-B bond and the lability of the corresponding bisĲboryl)platinum complexes. 14As a result, even easily handled and often air stable tetraalkoxy-and tetraaryloxydiboron reagents can be utilized, regardless of their relatively high B-B bond strength. 15However, despite the extensive studies, a number of general limitations remain: the use of elevated temperatures, high catalyst loadings and long reaction times.
Only two examples of palladium catalysed alkyne diborations have been described in the literature, both by Braunschweig and co-workers and involving the heterogenous catalysed diboration of alkynes using [2]borametalloarenophanes. 12 The reactions required 6 mol% of Pd/C and proceeded over a period of 5-16 days at temperatures of 95-100 °C (Scheme 1).The dearth of reported palladium examples is attributed to the energetics of the B-B oxidative addition.The process is endothermic with a very low reverse activation barrier 16 and therefore kinetically and thermodynamically unfavourable.
We recently reported the synthesis of the N-heterocyclic carbene bearing 17 complex PdĲITMe) 2 (PhCCPh) (ITMe = 1,3,4,5-tetramethylimidazol-2-ylidene) (1) and its high catalytic reactivity in bis-silylation 18 and silaboration of internal and terminal alkynes. 19This prompted us to investigate its potential in the diboration of alkynes.Herein, we report the use of 1 in the unprecedented palladium catalysed diboration of sterically demanding internal and terminal alkynes, employing low catalytic loadings and mild reaction temperatures.In addition, a thorough density functional theory (DFT) study was conducted in order to establish a likely mechanistic pathway explaining this reactivity.

Results and discussion
The reaction parameters were optimized using diphenylacetylene and commercially available bisĲpinacalato)diboron (B 2 pin 2 ) as the model substrates, with C 6 D 6 as solvent in order to monitor the progression by 1 H NMR. To our delight, 100% stereoselective conversion to (Z)-1,2-diphenyl-1,2-bis(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)ethane (2) was observed using 0.5 mol% of 1 at room temperature in 21 h.Unfortunately, initial work-up procedures proved troublesome, with the use of either silica and alumina columns resulting in very low isolated yields presumably due to reactivity with, or strong binding to the stationary phase.Kugelrohr distillation is an alternative methodology reported in the literature, 20 but is generally applicable to small quantities of material and therefore unviable as a scalable procedure.In our case, the more noticeable impurity was unreacted B 2 pin 2 .To remove it, the crude dry material was stirred in deionized H 2 O at room temperature over 24 h. 21Subsequent filtration and drying resulted in the clean isolation of 2 in a 99% yield.While there are several protocols in the literature for the synthesis of 2, the highest yield was reported by Jin and co-workers 22 who obtained a comparable yield to ours using 2 mol% of nanoporous gold at 100 °C over 12 h.To test the potential of this protocol for scaling-up, the synthesis of 2 was also carried out in non-deuterated benzene and toluene on a larger practical scale, resulting in comparable isolated yields (Table 1).The potential of 1 to catalyse this reaction using other diboron reagents was also investigated, but unfortunately neither bis(catecolato)diboron nor bis(neopentylglycolato)diboron afforded any of the desired product.
With this information in hand, a series of sterically and electronically demanding alkynes were reacted with B 2 pin 2 (Table 1).The diboration of alkyl and aryl terminal alkynes proceeded using 0.5 mol% of 1 at room temperature over 1-48 h with 100% stereoselectivity.A wide range of functionalities on the aryl moiety was tolerated including fluoro, trifluoromethyl, methoxy and alkyl groups in the ortho, meta and para positions.Compounds 3, 4, 5 and 6 were synthesized using lower catalyst loadings, milder temperatures and in higher or comparable yields to the highest yielding proto-col in the literature (2 mol% nanoporous gold, 100 °C), 22 and 5 and 6 were synthesized with comparatively higher stereoselectivities.Low reaction temperatures have been reported for the synthesis of these compounds using both homo-and heterogenous platinum complexes, although at the expense of lower yields and in many cases higher catalyst loadings. 13,23,24ompound 7 was synthesized in a higher yield than the highest yielding protocol in the literature (0.2 mol% Pt/TiO 2 , 70 °C, 16 h). 13The highest yielding synthesis for compound 8 was reported by Miyaura and co-workers (94% yield) 13 using 3 mol% Pt(CO) 2 (PPh 3 ) 2 at 80 °C in DMF over 24 h.
The novel compounds 9, 10 and 11 were synthesized with 100% syn-stereoselectivity as established by NOESY NMR spectroscopy.In the case of 11 chemoselectivity is achieved since the olefin remains unreacted.Unsymmetrical internal alkynes also reacted well under these conditions, albeit at higher-but still mild-temperatures (50 °C).The novel compounds 12 and 13 were synthesized with 100% synstereoselectivities.The diboration of 1-phenyl-2-trimethylsilane, resulting in the formation of 14, required an increased catalyst loading of 2 mol% and a much higher temperature (100 °C).The best procedure for the synthesis of 14 was detailed by Nishihara, obtaining a comparable yield using 5 mol% of PtĲPPh 3 ) 4 at 80 °C. 13Finally, the diboration of 4-octyne resulted in a maximum conversion to 15 of 39%.Even lower conversions and the formation of palladium black were observed when we carried out the reaction at higher temperatures.We presume that the electron-rich nature of the alkyne results in a low binding affinity to the very electron-rich, active catalyst and therefore discourages diboration.
We decided then to investigate the reasons behind this unprecedented activity.The accepted experimental and theoretical mechanism for platinum group transition metal catalysed diboration of alkynes involves: (i) oxidative addition of the B-B bond to a MĲ0)L 2 centre forming L 2 M(II)(B) 2 , (ii) dissociation of an L ligand (a phosphine) and coordination of the alkyne in its place, (iii) insertion of the alkyne into the M-B bond, (iv) isomerization of the resulting complex, followed by re-coordination of the L ligand, and (v) stereoselective reductive elimination. 25This mechanism is general and applies to other E-E′ bond addition to alkynes. 26We recently proposed that the use of NHCs as a ligand set results in a different mechanism, in which both NHCs remained coordinated throughout.This alternative pathway was used as an explanation for the observed increase in reactivity of 1 compared to their phosphine and isocyanide analogues in alkyne bis-silylations 18 and silaborations. 19To gain further insight into the mechanism and role of 1 in the diboration of alkynes, computational studies were carried out on the optimized model substrates (see ESI †).Additionally, a simultaneous study of Pd(0)(PMe 3 ) 2 (PhCCPh) (1-PMe 3 ) was performed to establish a direct comparison with the NHC ligand set.
Initially, the geometry of 1 was optimized at M06-L/BSI level of theory and compared to X-ray diffraction data. 18The optimized Pd-alkyne bond lengths Pd-C1 and Pd-C6 are longer, around 0.01 Å, than the results obtained by X-ray data.
The oxidative addition of bisĲpinacolato)diboron to 1 begins with the dissociation of the alkyne from the η 2 -complex, resulting in the formation of the 14 electron complex I1 (Fig. 2).This dissociation is favourable at 6.4 and 3.5 kcal mol −1 for the NHC and PMe 3 complexes, respectively.The reaction continues through the incorporation of bisĲpinacolato)diboron in the coordination sphere of I1, achieving the intermediate I2.The transition state TSA01 represents the step where the B-B bond is cleaved with concomitant formation of two σ Pd-B bonds.This process has a free energy activation barrier relative to the separated reactants at ΔG ‡ = 9.7 kcal mol −1 for the NHC and 11.1 kcal mol −1 for the PMe 3 bis-ligand complexes.These reaction barrier heights suggest that the oxidative addition for the phosphines is kinetically less favoured than the NHC ligands.BisĲboryl)palladiumĲII) complex (I3) is the product of the oxidative reaction step for both systems, and, energetically, is 8.7 kcal mol −1 and 6.5 kcal mol −1 above the reactants with NHC and PMe 3 ligands, respectively.Alternatively, the oxidative addition step could proceed through the mono-ligand complexes.Scheme 2 depicts two possibilities (pathway A and pathway B) related to dissociative reaction routes.Dissociation of alkynes from 1 to form the 14electron complex I1 is an exergonic process for the NHC (−6.4 kcal mol −1 ) and the PMe3 (−3.5 kcal mol −1 ) ligands (pathway A).However, in pathway B, the dissociation of ligand (L) from 1 to form the mono-ligand alkyne palladiumĲ0) complex 1_alkyne is an endergonic process at 5.9 kcal mol −1 for L = NHC and 8.9 kcal mol −1 for L = PMe 3 .The thermodynamic driving  force for the dissociation of L is greater going through pathway A over pathway B. Furthermore, the breaking of the second Pd-L bond to form the reactive mono-ligand PdĲ0)-L (I4) is easier for L = PMe 3 , than for L = NHC systems.Indeed, the Pd-NHC bond typically is stronger than the Pd-phosphine bond.The bis-ligand complex is energetically preferred for both NHC and phosphine systems, in the oxidative addition of bisĲpinacolato)diboron to the Pd centre (see ESI † for more details on the free energy profile of reaction pathways).
The next step is the alkyne insertion on the metal centre.The first proposal is that the insertion of alkyne occurs directly to the bisĲboryl)palladiumĲII) complex I3.The product of the alkyne insertion can be obtained through the pentacoordinate transition state TSIA1 (Fig. 3 see ESI † for further details and Scheme 3 for the proposed mechanism).
Another possibility for the insertion of the alkyne proceeds via a dissociate pathway (TSIA2).Kinetic results on PtĲ0)catalyzed diboration reactions with phosphine ligands suggested that the alkyne insertion occurs from the three-coordinate species in the oxidative addition. 27Following this, it is a reasonable assumption that the dissociation of one L takes place from complex I3 to generate the monoĲboryl)palladiumĲII) complex I6 (Scheme 3).The transition state TSIA2 is associated to the migratory insertion of the alkyne with one ligand attached on the metal centre.The alkyne triple bond and Pd-B bonds are broken forming a new C-B.
The dissociative pathway (TSIA2) has a lower relative reaction free energy barrier than on the associative pathway (TSIA1) at ΔΔG ‡ = 4.0 kcal mol −1 for L = PMe 3 and at ΔΔG ‡ = 15.4 kcal mol −1 for L = NHC system.Based on these results, This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.

View Article Online
the reductive elimination should occur by the cis-complex I12 from the insertion of the alkyne via the dissociative mechanism for both ligands (further details in ESI †).The recoordination of other ligand is reasonable, since the 16electron configuration is achieved forming the palladiumĲII) complex I13 with a square planar geometry (Scheme 3).
In order to obtain the anti-adduct is necessary to proceed with a consecutive isomerization processes involving the CC bond and C-Bpin moiety in the allyl ligand after the alkyne insertion is accomplished.Cui and co-workers suggested these isomerization pathways are energetically prohibitive because of the rigidity of the CC bond. 25Therefore, the substrate controls the stereoselectivity in the PtĲ0)catalyzed diboration reaction towards the syn-1,2-diborated product.Analogously, this mechanism for selectively could be expanded for the present reaction, since the same substrate was used (alkyne).In this case, if these isomerization processes take place very quickly, the selectivity would be defined solely by the relative energies of the transition states associated with the reductive elimination steps.Fig. 4 shows  the optimized geometry of the transition states TSRE1 and TSRE2 reacted with the syn and anti-adducts, respectively.For the NHC complex, the relative free energy activation of ΔΔG ‡ = 7.4 kcal mol −1 , favouring the transitions state TSRE1, is in perfect agreement with the product detected experimentally.
DFT calculation suggest the PdĲ0)-catalysed alkyne diboration supported by NHC ligands proceeds through the same mechanism as the phosphine ligands (the free energy profile of the overall catalytic cycle is presented in the ESI †).This mechanism (Scheme 3) can therefore be summarized as: (i) the activation of the catalyst by alkyne dissociation from 1, (ii) oxidative addition of the B-B to Pd(0), (iii) ligand dissociation from bisĲboryl)palladiumĲII) complex I3, (iv) insertion of the alkyne into a Pd-B bond via migratory insertion, (v) cistrans isomerization involving the C-Bpin and the allyl ligands, and (vi) reduction of PdĲII) to Pd(0) with the elimination of the syn-1,2-diborylated product.
Cui and co-workers proposed that a reversible oxidative addition step is the reason of the null reactivity of a PdĲ0)L 2 catalyst (L = phosphine) in alkyne diboration reactions. 16The oxidative addition step was predicted to have an activation barrier of 8.6 kcal mol −1 .However, the B-B oxidative addition to Pd(0) was characterized as an endothermic process with a reverse barrier of only 0.1 kcal mol −1 .The cause of this low reverse barrier is attributed to the promotion energy from d 10 Pd(0)L 2 with linear geometry (singletground state) to d 9 s 1 Pd(0)L 2 with bent geometry (tripletexcited state).The energy between these two electronic configurations is larger for Pd(0)L 2 than for Pt(0)L 2 with phosphines.Sasaki and coworkers, 27 studying the activity of Pd(0)L 2 and Pt(0)L 2 catalyst (L = phosphine) in the C-H activation of methane by oxidative addition, reported the destabilization of the M(0)L 2 complexes as an important factor in smoother oxidative additions.Chelating phosphines were used to destabilize the M(0)L 2 complexes by bringing the reactants closer in order to promote the oxidative addition transition state.(NHC)-Pd(0) catalysts were also investigated in the activation of methane by oxidative addition, 28 and considered better candidates as catalysts than the analogous phosphine-based Pd(0) complexes.Based on these results, we propose that the considerably increased reactivity of NHC-bearing complex 1 in the alkyne diboration is a consequence of the oxidative addition step; more specifically, on the destabilization of the (diboron)Pd(0)L 2 complex I2 by the NHC ligands resulting in a lower activation free energy for the oxidative addition (3.9 kcal mol −1 ) compared to PMe 3 (13.3kcal mol −1 ).

Conclusions
We have shown that complex 1 acts as highly active catalyst in the diboration of sterically and electronically demanding alkynes.For terminal alkynes, low catalyst loadings and temperatures were used for the 100% stereoselective synthesis of syn-1,2-diborylalkenes.Internal alkynes can react using this protocol, albeit requiring elevated temperatures.This represents the first example of homogenous palladium catalysed diboration of alkynes.DFT calculations, based on M06 suite density functionals, were performed to understand the activity of the NHC-bearing catalyst 1.The results suggest that 1 proceeds through the same mechanistic pathway as the corresponding phosphine analogues.The dissociation of one NHC is crucial part of the mechanism, unaccounted for in our proposed pathways for the other E-E′ bond additions to alkynes. 18,19Despite their strong coordination to metal centres, it has been previously shown that the reversible dissociation of an NHC from an oxidative addition products is a mechanistic possibility. 29The DFT study also showed that the destabilization of the (diboron)PdĲ0)L 2 adduct by the NHCs was key to the successful oxidative addition of the B-B bond.Further investigations into the scope and limitation of 1 in the B-B bond additions to other unsaturated organic substrates are currently ongoing in our laboratories.

Fig. 2
Fig.2Free energy profile (in kcal mol −1 ) for the oxidative addition pathways with bis-ligand complexes.

Fig. 3
Fig. 3 Associative (TSIA1) and dissociative (TSIA2) transition states for the alkyne insertion, respectively.Distances for selected bonds are given in angstrom units (Å).Relative free energies are given at 298.15 K.

Fig. 4
Fig. 4 Calculated transition states for the reductive elimination step associated with synand anti-1,2-diborated adduct, respectively.Distances for selected bonds are given in angstrom units (Å).Relative free energies are given at 298.15 K.

Table 1
Diboration of terminal and internal alkynes