Versatile Catalytic Hydrogenation Using A Simple Tin(IV) Lewis Acid

Abstract Despite the rapid development of frustrated Lewis pair (FLP) chemistry over the last ten years, its application in catalytic hydrogenations remains dependent on a narrow family of structurally similar early main‐group Lewis acids (LAs), inevitably placing limitations on reactivity, sensitivity and substrate scope. Herein we describe the FLP‐mediated H2 activation and catalytic hydrogenation activity of the alternative LA iPr3SnOTf, which acts as a surrogate for the trialkylstannylium ion iPr3Sn+, and is rapidly and easily prepared from simple, inexpensive starting materials. This highly thermally robust LA is found to be competent in the hydrogenation of a number of different unsaturated functional groups (which is unique to date for main‐group FLP LAs not based on boron), and also displays a remarkable tolerance to moisture.

Abstract: Despite the rapid development of frustrated Lewis pair (FLP) chemistry over the last ten years,its application in catalytic hydrogenations remains dependent on an arrow family of structurally similar early main-group Lewis acids (LAs), inevitably placing limitations on reactivity,s ensitivity and substrate scope.Herein we describe the FLP-mediated H 2 activation and catalytic hydrogenation activity of the alternative LA iPr 3 SnOTf,w hich acts as as urrogate for the trialkylstannylium ion iPr 3 Sn + ,a nd is rapidly and easily prepared from simple,i nexpensive starting materials.T his highly thermally robust LA is found to be competent in the hydrogenation of anumber of different unsaturated functional groups (which is unique to date for main-group FLP LAs not based on boron), and also displays ar emarkable tolerance to moisture.
Since the formalization of the concept within the last decade,great attention has been focused on the development and study of frustrated Lewis pairs (FLPs): Lewis acid (LA) and base (LB) combinations that fail to form the classically expected strong adduct, typically because it is sterically precluded. [1] Ther esulting combined reactivity has been found to lead to ar ange of novel bond activation reactions that do not require the involvement of at ransition metal (TM). [2] Of particular interest has been the activation and cleavage of H 2 ,which has allowed the development of the first general methodology for TM-free catalytic hydrogenation. [3] Computational investigations have suggested that the primary requirements for successful activation of H 2 by an FLP are as ufficient cumulative LA/LB strength, and as uitable steric profile. [4] One appealing aspect of FLP chemistry is therefore the generality of the concept;i ndeed, FLP-type reactions have been identified for abroad spectrum of LAs and LBs. [2,5] Nevertheless inspection of the literature reveals that, despite the apparent breadth of the field, investigations into TM-free FLP-catalyzed hydrogenation have focused overwhelmingly on av ery narrow range of LAs;t hus far this has exclusively been achieved using Bbased acceptors [6] [predominantly (fluoroaryl)borane derivatives,ofwhich B(C 6 F 5 ) 3 is prevalent], [7] with the exception of as ingle report using P-based LAs (for al imited range of activated olefins). [8] This constrained focus is far from ideal, as examining and developing aw ider variety of LAs can be expected to produce novel FLP-catalyzed protocols that display different substrate scope and/or more favorable functional group tolerance. [9] Fore xample,t he application of highly Lewis acidic boranes to the FLP-catalyzed hydrogenation of organic carbonyls has been notably challenging: whilst stoichiometric reductions were reported as early as 2007, [9] it took until 2014 until catalytic protocols were developed. [10] This difficulty can be attributed to the strength of the interaction between the alcohol (ROH) products and the LAs,which renders the LA·ROHadducts strongly acidic [cf.H 2 O·B(C 6 F 5 ) 3 ;p K a = 8.4 (MeCN), < 1( H 2 O, est.)]; [11] consequently,t hese adducts are fundamentally incompatible with the moderately strong N/P-centered LBs typically incorporated into active FLP catalysts.U ltimately,t urnover can only be achieved when such LBs are strictly excluded, due to the necessarily highly Brønsted acidic media [for example, protonated ethers,pK a (H 2 O) ! 0]. [10,12] Based on the above,w ew ere motivated to investigate FLPs based on heavier p-block LAs,w hich have thus far attracted scant attention for use in FLP applications. [13] Specifically,o ur interest was drawn to stannylium ion "R 3 Sn + "( R = alkyl) LAs; [14] these are isolobal with Ar 3 B species commonly employed in FLP chemistry,and have been calculated to possess similar hydride ion affinities (DG H À = 65.83 and 64.95 kcal mol À1 for nBu 3 Sn-H and [H-B(C 6 F 5 ) 3 ] À respectively), [15] suggesting that they ought to demonstrate comparable reactivity in FLP H 2 activation and hydrogenation reactions.F urthermore,C = Or eductions by R 3 SnH in protic media are well known to occur via ionic hydride transfer. [16] Crucially,however, these LAs interact only much more weakly with hydroxylic species [for example, (nBu 3 Sn·x H 2 O) + ;pK a (H 2 O) = 6.25]. [17] Manners et al. have previously investigated the use of nBu 3 SnOTf (an nBu 3 Sn + equivalent;T f= CF 3 SO 2 )a saL A partner in FLP chemistry, [13a] but reported that it was not capable of activating H 2 when combined with the strong amine base TMP (2,2,6,6-tetramethylpiperidine) at 50 8 8C, whereas the B(C 6 F 5 ) 3 /TMP FLP readily cleaves H 2 ,e ven at room temperature; [18] this result was attributed to the poorer electrophilicity of the Sn compound, and it is evident that the Sn-OTf interaction is strong enough to substantially reduce the Lewis acidity of the nBu 3 Sn + fragment.
We envisioned that it should be possible to increase the Lewis acidity,t ot he threshold necessary for favorable H 2 heterolysis,b ys imply increasing the size of the alkyl groups on Sn, thereby increasing the degree of "internal frustration" [19] between the R 3 Sn + and TfO À moieties.T ot his end, we targeted the bulkier trialkylstannyl compound iPr 3 SnOTf ([1]OTf), which was readily prepared via reaction of excess iPrMgCl and SnCl 4 to generate iPr 4 Sn, followed by facile protodealkylation with HOTf (Scheme 1). This straightforward and inexpensive two-step procedure furnishes pure [1]OTfi ng ood yield (42 %, 2s teps), and can easily be performed on amulti-gram scale. [1]OTfisawhite solid that shows moderate solubility in polar halogenated solvents and its 119 Sn{ 1 H} spectrum shows as ingle broad resonance at d = 156 ppm (Dv 1/2 = 130 Hz, CDCl 3 ). Theh igh chemical shift is consistent with significant stannylium ion character,although it is considerably upfield of the value reported for [nBu 3 Sn]-[CB 11 Me 12 ]( d = 454 ppm), which displays the least coordinated trialkylstannylium core to date. [20] Gutmann-Beckett Lewis acidity measurements support this conclusion, [21] indicating increased electrophilicity in comparison with nBu 3 SnOTf,a lthough still lower than B(C 6 F 5 ) 3 [22]  Having demonstrated H 2 activation, our focus shifted to achieving catalytic hydrogenation using [1]OTf.G ratifyingly, heating the archetypal FLP substrates PhCH=NtBu (2a)and PhC(Me) = NtBu (2b)w ith 10 mol % [ 1]OTft o1 20 8 8Cu nder H 2 (10 bar) led to conversion to the respective amines (3aand 3b;T able 1, entries 1a nd 2). Conversely,t he N-phenyl analogue PhCH=NPh (2c)i sr educed far less effectively ( Table 1, entry 3), which is attributed to the reduced basicity of both the imine and amine product, which makes H 2 activation less favorable.C onsistent with this interpretation, addition of 2,4,6-collidine [Col;pK a (MeCN) = 14.98] [23] as an auxiliary base leads to ad ramatic improvement in performance (Table 1, entry 4), and also allows for reduction of the related ketimine PhC(Me) = NPh (2d;T able 1, entry 5), and even PhCH=NTs (2e;T s= O 2 SC 6 H 4 Me,4 -toluenesulfonyl), although the latter reaction is appreciably slower,presumably as the substrate is less basic still (  [24] supporting the idea that radical Sn species do not appear to be involved in this reaction. Accordingly,wepropose that hydrogenation occurs via ap olar mechanism analogous to that for related boranecatalyzed systems: [1d,e,25] H 2 activation by an FLP consisting of [1]OTf/imine is followed by hydride transfer and release of amine at elevated temperature ( Figure S15). This is further supported by the observation that pre-formed 2a·HOTf is rapidly reduced by [1]H even at RT, [26] whereas the equivalent reactions with unprotonated 2a,e ither alone or in the presence of [1]OTf,d on ot lead to significant reduction at 120 8 8C(see SI). Interestingly,there is evidence for autocatalysis during the course of the reaction (16 %c onversion observed after 3h,6 0% after 6h); comparable observations have been made by Paradies et al. for imine hydrogenations catalyzed by B(2,6-F 2 C 6 H 3 ) 3 ,a nd are attributed to the increased basicity of the product amines,r elative to the imine substrate,r endering H 2 activation more favorable as more product is formed. [25] Following success in the hydrogenation of imines,wewere interested to see whether [1]OTfm ight also be capable of mediating the hydrogenation of closely related carbonyl compounds.S atisfyingly,w hen acetone (4a)i se xposed to reaction conditions similar to those used to hydrogenate 2c catalytic conversion to 2-propanol (5a)i so bserved ( Table 2, entry 1). Whilst the reaction at 120 8 8Ci ss omewhat slow,a t 180 8 8Cn ear-quantitative conversion can be observed within 32 h( Table 2, entry 2). Significantly,n oe vidence of catalyst decomposition is observed in this homogeneous reaction, Scheme 1. Synthesis of [1]OTf. either by 1 Hor 119 Sn{ 1 H} NMR spectroscopy, [27] in comparison with analogous FLP protocols mediated by B(C 6 F 5 ) 3 . [1f, 28] To the best of our knowledge this is the first example of ac atalytically active FLP system capable of tolerating such conditions without degradation, and illustrates the impressive thermal stability of [1]OTf, which enables the use of more forcing conditions in order to achieve an improved rate of turnover. As well as 4a,other aliphatic and aromatic ketones and aldehydes (4b-d)can be reduced under these conditions ( Table 2, entries 3-5). In the case of acetophenone (4b), 1 HNMR spectroscopic analysis indicates formation of the expected alcohol 5b,i na ddition to smaller quantities of styrene (6)a nd a-methylbenzyl ether (7). Similar sidereactions were observed in our previous attempts to reduce 4b using B(C 6 F 5 ) 3 in 1,4-dioxane, [10b,12c] but in those cases this led to severe reductions in conversion and rate of turnover. Theease and speed with which it was possible to apply this system to carbonyl hydrogenation stands in contrast to the extended period of development required before more conventional B-based FLPs were successfully used in this transformation. [10a,b] It is also noteworthy that the [1]OTfcatalyzed reaction can proceed using ar ather conventional, moderately-strong,N -centered LB,w hich again contrasts with B-based systems and is consistent with less acidic adducts forming between the product alcohols (e.g. 5a)a nd [1]OTf. Thec hoice of LB is important to the outcome of the hydrogenation of 4a (Table 2, entries 6-8), with inferior results obtained using either aw eaker or stronger base [2,6lutidine (Lut), DABCO;p K a (MeCN) = 14.13, 18.29]. [29] Given the low Brønsted basicity of 4a we propose as lightly different mechanism for its reduction than for 2a, [30] with the substrate activated by [1] + rather than via Hbonding to [Col-H] + (Scheme 2a). [16b] Evidence for this comes from the significantly upfield-shifted 119 Sn{ 1 H} NMR resonance (d = 92 ppm) observed upon addition of 4a (10 equiv.) to [1]OTf,i ndicative of SnÀOb inding. [31] Ap roposed subsequent H À transfer from [1]H to adduct { [1]·4a}OTf,t o form [1]OiPr and regenerate [1]OTf,i ss upported by the observation that [1]H is capable of reducing 4a in the presence of [1]OTf even at RT,w hereas no appreciable conversion is observed in its absence either at RT or 120 8 8C.
Conversely,i f [ 1]OTfi sr eplaced by Col·HOTf,o nly slow release of H 2 is observed at RT. [32] In order for the final H + transfer step to occur efficiently it should be recognized that Col and [1]OiPr must be comparable in base strength and, therefore,itmay be envisaged that once [1]OiPr is formed in the reaction mixture,i tcould also activate H 2 in conjunction with [1]OTf (Scheme 2b). In fact, catalytic hydrogenation can be observed by substituting Col with [1]OiPr (generated in situ from [1]H and 4a;T able 2, entry 9), thus demonstrating its competence in this role.E ven so,t he reduced rate of turnover in this reaction indicates that the auxiliary base does play abeneficial role beyond simply facilitating formation of some initial [1]OiPr, presumably by rendering H 2 activation more favorable. [33] Clear tolerance of alcohol products suggested that these reactions might also demonstrate appreciable moisture tolerance. [10,12] Remarkably,w hen the hydrogenation of model substrate 4a (chosen over an imine to avoid hydrolysis) was prepared on the open bench, with non-anhydrous reagents and solvent, and using [1]OTf that had been exposed to air for 1week, the reaction was observed to proceed without any noticeable reduction in rate ( Table 2, entry 10;details in SI). This is unprecedented in FLP catalysis,w here even the most tolerant of previously reported reactions have been dramatically slowed by adventitious H 2 O, [12] and suggests am ajor advantage of using Sn-based LAs.
Finally,w ei nvestigated the use of [1]OTfi nt he catalytic hydrogenation of compounds containing other unsaturated functionalities;t he heteroaromatic ring of acridine,a nd the C = Cb onds in n-butyl acrylate and 1-piperidino-1-cyclohexene could all be effectively reduced (yields 83-99 %), further demonstrating the versatility of this Sn IV compound (Figure S33). In summary,w eh ave demonstrated the use of readily accessible and inexpensive iPr 3 SnOTf as am ain-group LA catalyst for the hydrogenation of C = C, C = Nand C = Obonds; this constitutes only the second example of an FLP hydrogenation protocol utilizing ap -block LA not incorporating boron, and the first such example shown to be applicable to the reduction of ar ange of different functional groups. Despite the ubiquity of Sn in industrial catalysis this also represents,tothe best of our knowledge,the first example of homogeneous catalytic hydrogenation using aS n-based system of any kind. [34] Of particular interest is the ready applicability of this protocol to C=Ob ond hydrogenation, in ar eaction that displays an unparalleled level of H 2 O tolerance.T his neatly demonstrates the value of pursuing alternative FLP LAs,a nd can be jointly attributed to the formation of weakly acidic LA·ROHa dducts;athermally robust [iPr 3 Sn] + core,a llowing access to high reaction temperatures;a nd the stability of the SnÀCb onds towards protolytic cleavage for example,b yH 2 O. Clearly there is significant scope for variation of the triorganotin(IV) framework in "R 3 Sn + "s pecies;i nvestigations into how this affects their reactivity,f unctional group tolerance,a nd substrate scope are currently underway.