Chapter 1: Oxygen-Directed

1.1 The Versatility of Organoboranes

Organoboron species are among the most versatile functionalities in synthetic chemistry. Aliphatic and alkenyl can be oxidized to and carbonyl groups, respectively.1 Aliphatic boranes can also be converted to amines2 and are well known for undergoing one-carbon homologation chemistry, allowing for installation of formyl groups, esters, and nitriles.3-5 Transition metal catalysis greatly expands the utiltity of . Palladium catalysis enables Suzuki cross-coupling reactions with both aryl/vinyl6 and alkyl7,8 coupling partners as well as with carbon monoxide,9,10 while rhodium catalyzes addition of vinyl boranes to .11-13

Trifuoroborate salts are more robust than other organoboron species, as they resist oxidation upon exposure to air and even by dimethyldioxirane (DMDO), enabling

Scheme 1A: The Synthetic Transformations of Organoborons OH O BX NHBn n R 1A1

B R R 1A8 C 1A2 l 3 , NaOOH O B D n M N D 3

H BX LiCCl OMe N ClCH2CN n 2 O R R 1A7 R 1A3 O 0 H 0 Pd , CO, 'C Pd , R 0 , ROH h X R R' CO R OH 2 R R 1A6 1A4

R 1A5

1 oxidation of olefins in the presence of boron.14 What makes all of these boron species even more attractive as synthetic intermediates is that they are all conveniently accessible by hydroboration.

1.2 The Limitation of Steric and Electronic Influence on Intermolecular Hydroboration Regioselectivity

Hydroboration is crucial for the synthesis of organoboranes, and involves the syn-addition of a boron-hydrogen bond across a carbon-carbon multiple bond in a four- membered transition state such as 1B2b.15 Intermolecular hydroboration of simple olefins is controlled by the steric environment of the olefin in concert with its electronic properties to provide anti-Markovnikov selectivity. The steric component of regioselectivity reflects the smaller bulk of hydride compared to a BR2 moiety (R= alkyl or H). Therefore, a (HBR2) preferentially approaches an olefin with boron at the

2 less hindered carbon. Electronic effects supplement the steric preference. The dipole of a borane B-H bond provides the hydrogen with negative character and the boron with positive character. The four-membered transition state features the hydride of the borane forming a bond with the more substituted carbon, which favors having partial positive character relative to the less substituted carbon.16 (Scheme 1B). The combination of steric and electronic effects leads to excellent selectivity (19:1) with terminal olefins even when R=H. Good selectivity is also achieved with trisubstituted olefins (98:2) and

2,2-disubstituted olefins (99:1). However, achieving regioselective hydroboration of a

1,2-disubstituted olefin remains an unresolved issue in the hydroboration literature. The following chapter presents the reports in the literature that discuss the possibility of affecting regio- and by directing hydroboration with oxygen-containing functionalities.

1.3 Mechanistic Proposals of Brown (Dissociative) and Pasto (Associative)

It has been proposed that intermolecular hydroboration can occur via either a dissociative pathway or an associative pathway. The work of H. C. Brown et al. supports a mechanism requiring that an uncomplexed trivalent borane be generated via dissociation from either its dimer 1C1 or a Lewis-base complex 1C2, in order to react with an olefin (Scheme 1C).17-20 There are several kinetic studies that report the observation of first order (in dimer) and three-halves order (1/2 order in dimer) kinetics for reactions of 9-BBN dimer with fast reacting and slow reacting olefins, respectively.17-19 Another study using provides similar kinetic evidence that dialkylboranes participate in via a dissociative pathway.20

3

Scheme 1C: Brown's Dissociation Pathway for Hydroboration17-20 R H R BR2 B B R H R 1C5 1C1 R H BR BR2 LB H B 2 R 1C3 1C4 1C5 H R BR2 B THF LB R 1C5 1C2

Compared to monoborane (BH3), dialkylboranes are much easier reagents with which to conduct kinetic studies. This is due to the fact that BH3 reacts with to form multiple alkylborane species RXBH(3-X), rendering kinetic data unclear. Despite this setback, Brown initiated Lewis base concentration studies involving dimethylsulfide-

21 borane (Me2S·BH3; BMS) and triethylamine-borane (Et3N·BH3). A qualitative rate reduction was observed upon increasing the concentration of excess Me2S and Et3N in the reactions of their corresponding borane complex with 1-. This led Brown to propose a dissociative mechanism for BH3, as a rate reduction under such conditions would not be expected if an associative mechanism were in effect.

Pasto has published an alternative associative hydroboration mechanism, proposing that THF remains complexed to boron while an olefin complexes to the boron

22, 23 atom of THF·BH3 1D1 and undergoes hydroboration (Scheme 1D). This mechanism is supported by kinetic studies conducted with 2-methyl-2- and 1D1. In addition to the observation of second order kinetics, first order in both olefin and borane, Pasto also reports an entropy value of -27 ± 1 eu.22 The kinetic data support a mechanistic pathway in which 1D1 is directly attacked by an olefin and the entropy value indicates that this attack does not result in the displacement of THF. The mechanism has been supported by

4 several theoretical studies,24,25 including work by Schleyer, which supports

26 complexation to R2O·BH3 occurring in a SN2-like fashion.

Scheme 1D: Pasto's Associative Mechanism for Hydroboration22,23

H H H H H H O B H O B O B O B H H H H 1D1 1D2 1D3 1D4

1.4 Defining Oxygen-Directed Hydroboration In order to discuss oxygen-directed hydroboration (ODHB), one must first establish what is meant by „directed reaction‟. Whenever the incorporation of an oxygen atom into a substrate affects the reactivity or selectivity of a transformation, one can argue that the change is due to some form of oxygen direction. This is regardless of whether the effect is due to the steric environment of oxygen or to an inductive or resonance polarization effect the oxygen might impose upon the substrate.

For example, Brown refers to “directive effects in the hydroboration of substituted styrenes” in a study reporting the effects of a methoxy group on regioselectivity

(Equation 1).27 Due to the planar structure of the fully conjugated system that separates oxygen from the -borane complex in the transition state with five bonds, these

5 results are due to the electronic perturbation placed on the substrate by remote oxygen.

“Directive Effects” is also the terminology used by Brown to describe the inductive effects that chloride and tosylate groups have on the regioselectivity of a proximal terminal alkene (Equation 2).28

For the purposes of providing a perspective on literature accounts of ODHB in the following chapter, the definition of a directed reaction to be used will be that of Hoveyda,

Evans, and Fu:29 “Preassociation of the reacting partners either through hydrogen bonding, covalent, or Lewis acid-base union is followed by the maintenance of this interaction during the ensuing chemical transformation.” This is not to indicate that all reports of ODHB to be discussed share(d) the same definition, but to establish a point of reference for the discussion herein.

1.5 Oxygen Directed Hydroboration in the Literature

1.5i Ether Directed Hydroboration

Narutis et al. have recognized contrasting results in hydroboration studies conducted by Gassman and Brown.30 Brown has reported that 1E5 provides

>99:1 exo-selectivity upon hydroboration/oxidation, while increasing steric bulk at C-7 via incorporation of a syn to the alkene (1E6) leads to 22:78 exo:endo selectivity (Scheme 1E).31 However, Gassman has reported that 7,7-dimethoxy- norbornene 1E1 undergoes hydroboration with THF·BH3 to provide a 78:22 exo/endo mixture of 1E2 upon oxidative workup.32 This selectivity does not correspond with Brown‟s steric argument. Narutis attributes the contrast of the 7,7-dimethoxy- norbornene result to the formation of ether-borane complex 1E3 but a proposed mechanistic pathway is not specified.30

6

Methyl ethers have also been reported as affecting the regioselectivity of the hydroboration of E- and Z-methoxy-ene-ynes.33 Zweifel et al. have reported that treating

Z-methoxy-ene-ynes 1F1 and 1F5 with dicyclohexylborane (chex2BH) in THF leads to preferential delivery of the boron to the carbon proximal to the methoxy group.

On the other hand, treating the corresponding E-substrates 1F4 and 1F8 under the same conditions results in an increased preference for boron delivery to the distal alkyne carbon. The authors rationalize the change in regioselectivities by invoking a transition state 1F9, which illustrates the Z-methyl ether maintaining an interaction with chex2BH while the B-H bond is added across the carbon-carbon triple bond (Scheme 1F). It should be noted that the olefin geometry equilibrates to trans under the oxidation conditions, after the hydroboration events.

7

Scheme 1F: Anomalous Hydroboration Results on Methoxy Ene-Ynes H H H O MeO O MeO H 1.chex2BH

OMe THF H H 2. NaOOH 1F1 1F2 1F3 96 : 4 H H O H 1.chex2BH MeO O MeO MeO THF H H H 2. NaOOH 1F4 1F2 1F3 65 : 35

H H H O H 1.chex BH MeO O MeO 2 Si OMe Si THF H H 2. NaOOH Si 1F5 1F6 1F7 56 : 44 H H H O MeO 1.chex2BH MeO O MeO Si H Si THF H 2. NaOOH Si H 1F8 1F6 1F7 1 : 99 H H

MeO B R H R' R' 1F9 Results reported by Suzuki serve as evidence against ether complexation to a hindered dialkylborane in an intramolecular hydroboration. In pursuit of avencolide,

Suzuki observed unexpectedly low anti-Markovnikov regioselectivity upon hydroboration (THF·BH3)/oxidation (NaOOH) of a monosubstituted olefin 1G1. The expected terminal alcohol 1G3 was isolated in 39% yield while the regioisomeric secondary alcohol 1G4 was isolated in 31% yield. The authors attributed this unusual result to a six-membered transition state 1G2. Interestingly, using a hindered dialkylborane (chex)2BH exclusively provides the expected terminal alcohol in 86% yield

8 after oxidative workup. The steric bulk of the dialkylborane is believed to be responsible for making the corresponding 1G2-like complex unfavorable, thus providing the regioselectivity one would expect from intermolecular hydroboration. Jung et al. have also discussed the possibility of benzyl ether-directed hydroboration with 5-benzyloxy-2- 1G5, which provides a 65:35 regioselectivity in favor of the 3,5-disubstituted product 1G6 upon hydroboration with THF·BH3 followed by oxidative workup (Equation

3). ODHB was dismissed because no diastereoselectivity was observed.34

OBn THF·BH OH OBn OBn 3 (3) NaOOH 65 : 35 OH 1G5 1G6 1G7

9

1.5ii Acetate Directed Hydroboration

There is one example of acetate-directed hydroboration proposed in the literature.

In the pursuit of epiallogibberic acid, House and Melillo treated the tetracyclic acetoxy olefin 1H1 with disiamylborane, generating cis-diol 1H3 after oxidative workup.35 The authors propose that the disiamylborane is “solvated” by the acetoxy group of the

substrate, which provides the diastereo- and regioselectivity (Scheme 1H). Jung et. al have reported that hydroboration of 5-acetoxy-2-heptene 1H4 provides 3,5-diol 1H5 with

75:25 regioselectivity after oxidative workup (Equation 4), but ODHB was dismissed in their report because no diastereoselectivity was observed.34

OAc THF·BH OH OH OH 3 (4) NaOOH 75 : 25 OH 1H4 1H5 1H6

1.5iii Alcohol Directed Hydroboration

Several diastereoselective hydroboration results have been attributed to alcohol direction. Results reported by Bryson in the pursuit of helenalin are the most intriguing.36

Unsaturated alcohol 1I1 was treated with either THF·BH3 or thexylborane (ThxBH2), generating respective 1:7 and 1:11 ratios of diastereomeric diols 1I3 and 1I4. On the other hand, treating unsaturated iodide 1I5 with THF·BH3 generates a 4:1 ratio of

10 diastereomers 1I6 and 1I7. This reversal of diastereoselectivity was attributed to an intermediate monoalkoxyborane that reacts via the bridged bicyclic transition state 1I2.

The proposal of an intramolecular hydroboration via a thexylalkoxyborane is supported by the work of Cha et al., who have reported that such species are viable reagents for intermolecular hydroboration, even of substrates even as hindered as 2,4,4-trimethyl-2- (Scheme 1I).37

Ohloff et al. have reported an outstanding level of diastereoselectivity achieved by the hydroboration of isopulegol 1J1 (Scheme 1J).38 The authors attribute a 19:1 ratio of

1,4-diols 1J3 and 1J4 to an intramolecular transition state 1J2 involving a dialkoxyborane intermediate. However, the dialkoxyborane intermediate was not observed and the authors did not comment on the low reactivity of the known dialkoxyboranes 1J5 and 1J6. They did show that the same product ratio was obtained

11 starting from the corresponding isopulegyl borate [(RO)3B] and , evidence that a disproportionation product is the species responsible for hydroboration.

A variant of the dialkoxyborane species invoked by Ohloff for intramolecular hydroboration was also proposed by Panek et al. in an unconventional approach to

ODHB.39 They reported a unique result involving diastereoselective hydroboration of

-methoxy--unsaturated esters with dimethylsulfide borane (BMS). The hydroboration is accompanied by reduction of the ester functionality. Low temperature studies revealed that ester reduction uncharacteristically precedes hydroboration, which the authors attribute to an activating effect of the -alkoxy group.40 It was proposed that the dialkoxyborane 1K2, generated from ester reduction, is the species that undergoes subsequent intramolecular hydroboration to provide diastereoselectivity opposite to that predicted by the Kishi model for intermolecular hydroboration.41 Treating the alcohol analog of 1K1 (1K4) with BMS resulted in neither regio- nor diastereoselectivity. Thus, the entire -alkoxy-homoallylic ester is apparently the requisite moiety for achieving selectivity even though the hydroboration itself is proposed to be alkoxy-directed.

In pursuit of (+)-mikrolin 1L3, Smith et al. demonstrated that diastereoselective hydroxyl-influenced hydroboration is not a consistent phenomenon. In an attempt to

12 synthesize diol 1L2, the homoallylic alcohol 1L1 was treated with an unspecified hydroborating agent, but 1L2 was not formed (Scheme 1J).42 This demonstrates that the results reported by Bryson and Ohloff do not represent a general trend. Jung et al. points out that an alcohol group influences the hydroboration of acyclic 5-hydroxy-2-heptene

1L4, which provides 3,5-diol 1L5 with 73:27 regioselectivity. However, no diastereoselectivity was observed, which is consistent with Jung‟s ether and acetoxy results in Equations 3 and 4, so ODHB was dismissed as before.34

OH THF·BH OH OH OH 3 (5) NaOOH 73 : 27 OH 1L4 1L5 1L6

13

1.6 Discussion

Theoretically, ODHB can be broken down into three simple processes: 1) a Lewis acid-base interaction forms an oxygen-borane complex 1M2, 2) an olefin tethered to the boron-complexed oxygen forms a complex to the borane 1M3, and 3) a B-H bond adds to the C=C bond. There are multiple reactions with which it remains unclear how an oxygen can affect the regio- and/or stereocontrol of the transformation without an interaction with borane (Scheme 1M).

There are two important issues to consider in order to relate section 1.5 to Scheme

1M: 1) how feasible is the formations of each proposed borane species? and 2) how viable an agent is each intermediate for its corresponding transformation to the next stage? For each case, these issues revolve around the directing group in question and how it relates to the mechanistic proposals of Brown and Pasto.

1.6i Intramolecular Hydroboration with Alkoxyboranes

Mechanistically, the proposed transitions states of Bryson (1I2) and and Ohloff

(1J2) are distinct from other oxygen-directed hydroborations. Each alcohol has reacted with borane to generate an alkoxyborane that resembles Panek‟s transition state structure

1K2. Boranes are well known for reacting with alcohols to form alkoxyboranes, as illustrated by the Cha results (Scheme 1L), so the formation of alkoxyboranes is not controversial.37 The important question in this context is whether the resulting species

14 have sufficient reactivity for intramolecular hydroboration, despite the covalent interaction between boron and oxygen in these structures. Since a Lewis base-borane complex is no longer the proposed hydroborating species, the mechanistic proposals of

Brown and Pasto no longer apply and there is no problem regarding transition state bonding. On the other hand, the issue of boron electrophilicity and reactivity remains critical.

Bryson‟s proposed intramolecular reaction of a monoalkoxy-alkylborane 1I2 is the most convincing example of an oxygen-directed internal hydroboration. One would expect the intermediate monoalkoxyborane to have reactivity between that of BH3 and a dialkoxyborane, and Cha‟s demonstration that alkenes are hydroborated by thexylalkoxyboranes provides evidence that one alkoxy group does not reduce the electrophilicity of boron sufficiently to prevent hydroboration.37 Therefore, Bryson‟s proposal that an oxygen-directed hydroboration occurs with thexylborane is consistent with the literature precedent.

Another study by Cha et al. investigated the reactivity of thexylalkoxyboranes

1I10 with simple alcohols.43 Hydrogen evolution studies demonstrated that a stoichiometric amount of various alcohols does not react quantitatively, and that conversion correlates to the steric bulk of the alcohol used. While they state, “it is best to defer for the present consideration of the reason why the hydrogen evolution stops beneath the stoichiometric point,” there appears to be something unique involved with the incorporation of the thexyl group on boron, considering that it is well known that treating boranes with alcohols easily forms borates,44 boronates,45 and borinates.46 If the thexyl group is the key to forming monoalkoxyborane species, Cha‟s work suggests that

15 diastereoselective reaction using THF·BH3 may reflect additional unknown factors that influence the product ratio. However, this does not rule out participation by an alkoxyborane 1I2 (R=H).

The case for internal hydroboration via dialkoxyboranes is less strong. As already mentioned in connection with Scheme 1J, pinacolborane 1J5 and catecholborane 1J6, are relatively unreactive, respectively requiring heating to 40 °C and 80 °C in order to hydroborate olefins without metal catalysis.47,48 This is due to two covalently bound oxygens that donate electron density to the empty p-orbital of boron, rendering it pseudo- tetravalent and less electrophilic. The behavior of 1J5 and 1J6 raises questions about the feasibility of an intramolecular hydroboration via dialkoxyborane.

A plausible argument that one might be tempted to make in defense of a room temperature intramolecular hydroboration with a dialkoxyborane is that the intramolecular nature of the transformation could provide enough of an entropic advantage to overcome the diminished electrophilicity of the dialkoxyborane. However, if this were the governing principle behind such results, ODHB would be a more common occurrence in the literature. If the entropic advantage of intramolecularity is large enough, then the mono- and/or dialkoxyboranes generated upon treating unsaturated alcohols such as Panek‟s 1K4 with BMS would result in regioselective hydroboration.

No such selectivity was observed (Scheme 1K). Additionally, work by Heathcock indicates that not only is selectivity not achieved upon treating an unsaturated alcohol with an equivalent of borane, but no reaction is even observed.49 Evidently, other unknown factors also play a role, culminating in results that are highly substrate dependant.

16

1.6ii Ether and Acetoxy Direction

Ether/acetoxy directed hydroboration is a more difficult phenomenon to account for compared with alkoxy direction. Compared with alkoxyboranes, ether-borane complexes (1M2, R=alkyl) are more transient species in oxygenated solvents due to the facile equilibrium with the solvent-borane complex. This disadvantage for an ensuing intramolecular process is supplemented by the fact that there is no clear rationalization that explains how an unsaturated ether-borane 1M2 can both 1) remain intact and 2) react with alkene to provide selectivity in the product mixture 1M4. Despite this lack of understanding, ODHB remains a tempting explanation for the results of Gassman,

Zweifel, and Suzuki, due to the absence of obvious controlling steric and/or electronic factors.

Ether-borane complexes are well precedented species so there is no doubt that an ether-containing substrate could form a complex to BH3. According to calculations published by Rauk et al., dimethylether has a greater affinity than methyl acetate for

50 (BF3) (71 kJ/mol vs. 58 kJ/mol). This suggests that an acetoxy group would form a weaker complex to borane than would an acyclic ether. The same study reports an 82 kJ/mol affinity between THF and BF3. These affinity values raise a question that challenges the proposals of Narutis, Zweifel, Suzuki, and House: how can the complexes 1E3, 1F9, 1G2, and (the precursor to) 1H2 form when THF or diglyme is being used as the reaction solvent?

The Brown or Pasto mechanisms are not consistent with an ensuing ODHB once the ether-borane complex 1M2 is formed. Brown‟s dissociative mechanism (Scheme 1C) can account for the formation of -complex 1M3, but not for any selectivity observed in

17 the subsequent hydroboration, as oxygen does not interact with boron in the transition state. Pasto‟s associative mechanism (Scheme 1D) cannot be applied because it is constrained by the geometric requirements imposed upon intramolecular SN2-like transformations, as reported by Beak51 and Baldwin (Scheme 1N).52

Scheme 1N: Intramolecular variant of Pasto's Mechanism Conflicts with Geometric Restrictions

R H disfavored H H H H R O Me B H O B O B O BH2 Me Me Me H H R H '5.5'-Endo-Tet 1N1 1N2 1N3 1N4

The reports by Zweifel (Scheme 1F) and Suzuki (Scheme 1G) include particularly intriguing proposals of ODHB and steric effects cannot account for the surprising regioselectivities. However, oxygen has been proposed to influence the selectivity of hydroboration with a variety of effects, and alternative explanations are worth evaluating, since ODHB has proven so difficult to rationalize in the context of ether-direction.

Inductive effects have been investigated in depth by Brown with acyclic allyl, homoallyl, and crotyl derivatives (Table 1O).28,53,54 Alcohols, ethers, chorides, triflates, and trifluormethyl groups clearly reduce the anti-Markovnikov regioselectivity achieved with , and increase delivery of borane to C-2 in the crotyl environment (R1=

CH3). The same substituents also work against the normal anti-Markovnikov regioselectivity of the allylic examples (terminal alkene, R1= H). The extent of the effect is related to the electron withdrawing capability of the X group, as the trifluoromethyl group (m = 0.46) effects selectivity more than chloride (m = 0.37). In both the crotyl and allyl systems, the allylic heteroatom promotes hydroboration at the nearest alkene carbon relative to the all-carbon analogies. A similar trend is also reported with substituted styrenes.27 The same trends in the context of 3- and 4- methoxy-cyclohexanes

18

Table 1O:Inductive Effects of Allylic, Crotylic, and Homoallylic Groups on Hydroboration Regioselectivity OH BH3·THF R1 X R1 X R1 X NaOOH OH 1O1 1O2 1O3

Entry R1 X 1O2 : 1O3 Entry R1 X 1O2 : 1O3 1 CH3 CH3 50 : 50 8 CH3 OPh 86 : 14 2 H CH3 6 : 94 9 H OPh 14 :86 3 H CH2CH3 7 : 93 10 H CH2OPh 13 : 87 4 CH3 OH 90 : 10 11 CH3 Cl 100 : 1 5 H OH 22 : 72 12 H Cl 40 : 60 6 H CH2OH 14 : 86 13 H CH2Cl 18 : 81 7 H OTf 45 : 55 14 H CF3 74 : 26 are shown in Table 1P, accompanied by anti-diastereoselectivity, which indicates no complexation between borane and the heteroatom substituted styrenes.27 The same trends in the context of 3- and 4- methoxy-cyclohexanes are shown in Table 1P, accompanied by anti-diastereoselectivity, which indicates no complexation between borane and the heteroatom.

Table 1P: Allic vs. Homoallyic Effects in Cyclohexene Systems

R2 R2 R2 R1 R3 BH3·THF R1 R3 R1 R3 NaOOH OH OH 1P1 1P2 1P3

Entry R1 R2 R3 1P2 : 1P3 trans- : cis- (1P2) 155 H H Cl 96 : 4 89 : 11 255 H H OMe 91 : 9 89 : 11 356 -C(O)N(Me)OMe H OMe “95 : 5” 95 : 5 457 -CH2SO2Ph H OMe 88 : 12 99 : 1 555 H OMe H 56 : 44 61 : 39

Steric interactions provide negligible hindrance to the hydroboration directing effect of allylic electron-withdrawing groups. Brown has reported that crotyl alcohols, ethers, and chlorides direct boron delivery to the proximal alkene carbon with

19 dialkylborane reagents disiamylborane and 9BBN (Table 1Q, entries 1-5).58 Gung et al. have shown that the added bulk at the carbinol carbon of 1-tBu-crotyl alcohol does nothing to prevent boron incorporation of either BH3 or thexylborane at the proximal carbon with complete regioselection (Table 1Q, entries 6&7).59 Gung invokes transition state 1Q4 to propose that hyperconjugation contributes to the excellent regioselectivity reported, and this is consistent with the data. However, the consistent trend in allylic examples is that regioselectivity is directly related to the electron withdrawing capability of the substituent, which polarizes the double bond such that it resists steric preferences.

The same effect is seen in the homoallylic examples, although it is smaller. Cyclic carbamates are also capable of inductively influencing hydroboration, according to a report by Sibi, in which the regioselectivity is attributed to the electron withdrawing capability of the allylic nitrogen (Equation 6).60 There could also be a small contribution from the homoallylic oxygen in Sibi‟s system.

20

Table 1Q: Hydroboration of Allylic Derivatives vs. Reagent Steric Demand

R1 R1 OH R1 1.R BH 2 t-Bu  Me X Me X Me X OH 2. NaOOH R OH H B H R  1Q1 1Q2 1Q3 1Q4

Entry X R1 HB Reagent 1Q2 : 1Q3 58 1 OH H Sia2BH 87 : 13 258 OH H 9BBN 100 : 0 358 OMe H 9BBN 92 : 8 58 4 Cl H Sia2BH 100 : 0 558 Cl H 9BBN 100 : 0 59 6 OH tBu BH3 100 : 0 59 7 OH tBu ThxBH2 100 : 0

O O O

THF·BH3 O NX C14H29 O NX O NX OH (6) NaOOH C14H29 C14H29 OH 1R4 X= H 1R6 X=Boc 20:1 regioselectivity, 16:1 anti-: syn- 1R5 X= Boc 1R7 X= H 15:1 regioselectivity, 3.4:1 anti-: syn-

Inductive effects are evident in acyclic substrates (Tables 1O & 1Q; Equation 3), and cyclic systems (Table 1P), indicating that the conformational mobility is not crucial for regioselective hydroboration. Keese et al. provide reinforcing evidence in a report on the hydroboration of 1-substituted norborn-2-enes (Table 1S).61 In this conformationally locked system, allylic electron withdrawing groups, including methoxy (entry 5), maintain a significant effect on hydroboration regioselectivity, and the authors point out that these results indicate that conformational mobility does not play a dominant role in achieving regioselectivity.

21

Entry X 1R2 : 1R3 1 Cl 74 : 26 2 Br 68 : 32 3 I 65 : 35

4 CH3 52 : 48 5 OMe 75 : 25

6 CH2OMe 54 : 46

Taking these reports into consideration, the Zweifel results remain fascinating because two similar enyne substrates provide such contrasting results (Scheme 1F).

Inductive effect arguments do not apply because there is no difference in oxygen-olefin tether length. Switching enol ether geometry could perturb the resonance interaction between alkyne and methoxy, but one would expect effective electron donation from the methoxy group in both geometries due to the planar, unhindered nature of the molecules.

Through-space oxygen-olefin orbital interactions have been invoked as a rationalization for stereoselectivity of hydroboration,34,62 but no report has been found in which it has been invoked to explain regioselectivity. Brown‟s work with ortho-methoxy-styrene provides indirect evidence that a through space effect does not account for Zweifel‟s regioselectivity, because the o-methoxy group leads to increased delivery of boron to the distal carbon in a „Z-methoxy-ene‟ system (Equation 7). This is attributed to the resonance electron donating effect of oxygen.

22

X X OH X THF· BH3 OH (7) NaOOH

X=H 18 : 82 X=OMe 12 : 88

Inductive effects must contribute to the reduced anti-Markovnikov selectivity reported by Suzuki with terminal alkene G1. Allylic and homoallylic electron withdrawing functionalities consistently provide a preference for delivery of boron to the proximal alkene carbon. While the influence of a homoallylic substituent is weaker, it can still be significant. Table 1O, entry 14 shows that three homoallylic fluorines cause a reversal in selectivity from 1:19 to 3:1 relative to 3 hydrogen atoms. Approximation using the Boltzmann equation indicates that this is a ca. 2.4 kcal/mol perturbation in G≠, meaning that each fluorine contributes a noteworthy ~0.8 kcal/mol. Homoallylic ethers have been shown to influence regioselectivity by ca. 0.1-0.4 kcal/mol (Equation 3; Table

1O, entry 10; Table 1P, entry 5; Table 1S, entry 6) and bishomoallylic ethers, such as the one in of G1, can also be expected to provide an influence. A Serratosa report also indicates that a homoallylic acetonide, analogous to the one incorporated into G1, provides an inductive effect (Equation 8),63 which correlates with Jung‟s homoallylic ether results (Equation 3). If excess borane is used, one could argue complexation of the alkoxy groups may lead to greater inductive effects, but this cannot be corroborated by the experimental details provided. Furthermore, complexation would be reversible, so at least some of the uncomplexed alkoxy substrate would be present and react faster.

Regardless of these details, questions remain regarding the role of inductive effects, and whether they alone can account for the reported product mixture.

23

R HO R HO R R R R1 THF·BH3 (8) NaOOH 1 : 1 R=R1 =H 1 2 : 1 R,R = -OCH2C(Me)CH2O-

Whatever the phenomena responsible for ether direction, they are not well understood and are most likely unique in each case. Through-space orbital interactions cannot be ruled out completely, but their effect(s) on regioselectivity is as poorly established as that of OHDB. Thus, there is no conspicuous conclusive evidence to oppose the ODHB proposals of Zweifel and Suzuki, despite the fact that the mechanistic pathway(s) remain a mystery. On the other hand, inductive effects cannot be dismissed as a contributing factor, Zweifel‟s methoxy-ene-ynes not-withstanding.

1.7 Summary of Uncatalyzed Oxygen-directed Hydroboration

Oxygen-directed hydroboration (ODHB) has been a casually invoked concept in synthetic organic chemistry literature for rationalizing a variety of anomalous regio- and stereochemical results. The reality is that, beyond some intriguing empirical indications, there is underwhelming evidence to support a conclusion that ODHB has ever occurred in any general sense. This does not eliminate the possibility that ODHB is, in fact, responsible for some of the results presented, nor can it be said that the mechanistic proposals of either Brown or Pasto provide conclusive insight into how such transformations transpire.

24

1.8 Metal-Catalyzed Oxygen-Directed Hydroboration

Transition metal-catalyzed hydroboration is a relatively recent development in hydroboration methodology that provides a mechanistically more straightforward approach to achieving a directed hydroboration. The inability of the Brown and Pasto mechanisms to account for oxygen-borane complexes undergoing intramolecular hydroboration is not a concern in the context of metal catalysis. Transition metals can insert into the B-H bonds of relatively unreactive dialkoxyboranes such as pinacolborane and catecholborane.64,65 An olefin can then complex to the metal rather than to boron.

The metal-hydride bond is added across the olefin, followed by reductive elimination to form the C-B bond, thus achieving a net hydroboration (Scheme 1O). Several studies have taken advantage of an oxygen atom tethered to the olefin in order to direct the M-H addition across the olefin by complexing to the metal (Scheme 1T).

25

1.8i Transition Metal Catalysis of Directed Hydroboration

In 1988, Evans specifically targeted alcohol-directed hydroboration in the context of transition metal catalysis.66 While rhodium(I) successfully promotes the hydroboration of olefins by catecholborane, alcohols are not viable substrates under such conditions because they form borates with catecholborane, precluding the desired transformation. A “net hydroxyl-directed”67 variant was pursued in which the hydroxyl groups were protected as diphenylphosphites and submitted to the reaction conditions, resulting in excellent diastereoselectivity. The phosphites were cleaved upon oxidative workup to generate diol products, which were assayed after acylation (Scheme 1U). The major limitation of this work was that it required stoichiometric amounts of Wilkinson‟s catalyst.

Evans et al. improved upon these results several years later by introducing the first catalytic directed hydroboration using secondary amides as directing groups under conditions requiring only 5 mol % of iridium catalyst.67 Good diastereoselectivity (91:9) was achieved on cyclic systems and excellent regioselectivity (99:1) was achieved on homoallylic amides (Scheme 1V).

26

Fu developed a method in which catalytic rhodium is viable for benzyl ether-directed hydroboration of a cyclic olefin. The key to this work is the exploitation of ring slippage in the indenyl ligands on rhodium.68 Ring slippage is a phenomenon observed with cyclopentadienyl-type ligands in which there is interconversion between 5- and 3- complexation to a metal. This conversion to 3-complexation creates an additional coordination site on (Ind)Rh(C2H4)2, allowing the coordination of both benzyl ether and olefin components of 4-benzyloxy-1-cyclohexene 1W1 to rhodium. Hydroboration ensues and oxidative workup provides a mixture of 1W2 and 1W3 isomers, as illustrated in Table 1W. A solvent study demonstrated that a less coordinating solvent improves diastereoselectivity, as hexane and CH2Cl2 provide respective diastereoselectivities of

82:18 and 79:21 (Entries 1 & 2) while THF provides only 62:38 diastereoselectivity

(Entry 3). The >10:1 ratios of cis-1,3- : cis-1,4-monoprotected diols in both hexane and

CH2Cl2 are indicative of an ether-directed reaction. A screen of several cyclopentadienyl

27 ligands was also reported. Based on the idea that selectivity is directly proportional to a ligand‟s propensity for ring slippage, entries 4-7 illustrate that the indenyl ligand has the best combination of 5- vs 3- binding and compatibility with other variables.

Entry CpX (mol %) Solvent cis-1R2 cis-1R3 trans-1R2 trans-1R3 1 Indenyl (10) hexane 75 7 7 11 2 Indenyl (10) CH2Cl2 74 5 8 13 3 Indenyl (10) THF 47 15 23 15 4 Indenyl (2.5) 75 7 8 11 5 1,2,3-Me3-indenyl (2.5) 65 10 12 13 6 Cp (2.5) 30 28 30 12 7 Cp* (2.5) 26 29 33 12

Recognizing that three electron donating methyl groups in the indenyl ligand caused a decrease in selectivity in Fu‟s work, Sowa Jr. et al. pursued increased selectivity by investigating indenyl ligands incorporating electron withdrawing groups.69 Entries

1-4 of Table 1X illustrate that both diastereo- and regioselectivity (of cis-isomers) are improved upon using 1-trifluormethylindenyl, 2-trifluormethylindenyl, or

1,3-(bis)trifluoromethylindenyl ligands on rhodium. Corresponding iridium catalysts were also screened with the same ligands, demonstrating superior selectivities. The best selectivity was achieved using the most electron deficient indenyl ligand on Ir(COD)

(Entry 8). Coupled with the reports by the Evans group, these results show that iridium is superior to rhodium in terms of reactivity and selectivity.

28

Entry M CpX (mol %) cis-1R2 cis-1R3 trans-1R2 trans-1R3 1 Rh Indenyl 74 11 9 6 2 Rh 1-CF3-Indenyl 81 6 10 4 3 Rh 2-CF3-Indenyl 81 6 9 4 4 Rh 1,3-(CF3)2-Indenyl 84 3 8 5 5 Ir Indenyl 93 <1 5 2 6 Ir 1-CF3-Indenyl 96 <1 2 2 7 Ir 2-CF3-Indenyl 96 2 2 <1 8 Ir 1,3-(CF3)2-Indenyl 98 2 <1 <1

A report by Gevorgyan et al. adds esters to the list of oxygen-containing functionalities that direct transition metal-catalyzed hydroboration, which already includes phosphites, amides, and ethers. Using pinacolborane 1J5 and [Rh(COD)Cl]2,

>99:1 cis diastereoselectivity was achieved with 3,3-disubstituted cyclopropenes.70

Adding (R)-BINAP to the reaction mixture provided excellent enantioselectivity (Table

1Y) .

Entry R R1 cis/trans ee (%) abs. config Yield (%) 1 Me Me >99 : 1 94 1S,2R 94 2 Et TMS >99 : 1 97 1R,2R 99 3 Me Ph >99 : 1 92 1S, 2R 99

29

1.8ii Lanthanide Catalysis in Alcohol-Directed Hydroboration

The previous section has illustrated that a variety of oxygen-containing groups can direct transition metal-catalyzed hydroboration. However, the only report in the literature in which conditions compatible with free hydroxyl groups are reported requires the use of a lanthanide. Evans et al. reported that samarium triiodide (SmI3) can catalyze the hydroboration of olefins by catecholborane and this methodology was applied to a homoallylic alcohol to test whether or not oxygen direction could be achieved.71 This method generated 1,3-pentane-diol with 11:1 regioselectivity from 3-penten-1-ol after oxidative workup (Scheme 1Z) but the mechanism remains unclear. The authors acknowledge that ODHB is not confirmed by this result and, as an alternative to the mechanism one depicted in scheme 1T, they present activated catecholborane 1Z4 as a potentially relevant species in the hydroboration event(s). No other substrates were investigated.

30

1.8iii Copper Catalyzed Hydroboration

Hoveyda has reported interesting results with copper-catalyzed hydroboration of oxygenated alkenes using N-heterocyclic carbene (NHC) complexes.72 Using copper complex 1AA2, delivery of pinacolatoboron to the -carbon of a variety of oxygenated styrenic olefins is achieved with >98:<2 regioselectivity (Table V). Using the chiral

NHC 1AA3 provides identical regioselectivity with enantioselectivities as high as 96% ee. Applying this methodology to non-oxygenated substrates provides identical regioselectivity, as do other metal-catalyzed hydroborations of styrenes.73 Therefore, these transformations need not be oxygen-directed and the authors do not propose that they are. However, the compatibility with the alcohol substituent makes them unique

(entries 4 & 7) in that they might succeed where rhodium has failed66 in providing a method for directed hydroboration of aliphatic unsaturated alcohols

Entry Catalyst R1 R2 X Solvent Regioselectivity Yield (%) ee (%)

1 1AA2 H CO2Me Na THF >98 : <2 76 - 2 1AA2 H OAc Na THF >98 : <2 82 - 3 1AA2 Me Me Na THF >98 : <2 96 - 4 1AA2 H H Na toluene >98 : <2 80 - 5 1AA3 H Me K THF >98 : <2 75 96-R 6 1AA3 Me Me K THF >98 : <2 51 89-R 7 1AA3 H H K THF >98 : <2 74 96-R

31

1.9 Oxygen-Directed Hydroboration: A Mechanistic Enigma

Several examples of transition metal-catalyzed hydroboration have been presented above. The mechanisms of these reactions are easily understood compared to the mechanism(s) of the proposed uncatalyzed ODHB reactions discussed in sections 1.3-1.5.

On the other hand, while the net result of these catalyzed reactions is indeed hydroboration, the directed aspect of the reaction is the hydrometalation step. Therefore, it remains debatable whether or not any oxygen-directed hydroboration is well understood.

32

References for Chapter 1

1. Johnson, J. R.; Van Campen, M. G. Organoboron compounds IV Reaction of tri- n-butylborine with peroxides and with oxygen - Mechanism autooxidation. J. Am. Chem. Soc. 1938, 60, 121.

2. Hupe, E.; Denisenko, D.; Knochel, P. Stereoselective Migration of Sterically Hindered Organoboranes in Cyclic and Acyclic systems. A Stereoselective Allylic C-H Activation Reaction. Tetrahedron 2003, 59, 9187.

3. Brown, H. C.; Bhat, N. G.; Campbell, J. B. Base-induced Alpha-Alkylation of Ethyl Bromoacetate, Phenacyl Bromide, and Chloroacetonitrile via B-trans-1- alkenyl-9-borabicyclo[3.3.1]nonanes. J. Org. Chem. 1986, 51, 3398.

4. Rathke, M. W.; Chao, E.; Wu, G. Preparation and Reactions of Esters of Dichloromethaneboronic Acid. J. Organomet. Chem. 1976, 122, 145.

5. Matteson, D. S. Halo Boronic Esters - Intermediates for Stereodirected Synthesis. Chem. Rev. 1989, 89, 1535.

6. Miyaura, N.; Suzuki, A. Palladium-Catalyzed Cross-Coupling Reactions of Organoboron Compounds. Chemical Reviews 1995, 95, 2457-2483.

7. Kirchhoff, J. H.; Netherton, M. R.; Hills, I. D.; Fu, G. C. Boronic Acids: New Coupling Partners in Room Temperature Suzuki Reactions of Alkyl Bromides. Crystallographic Characterization of an Oxidative Addition Adduct Generated Under Remarkably Mild Conditions. J. Am. Chem. Soc. 2002, 124, 13662.

8. Netherton, M. R.; Dai, C. Y.; Neuschutz, K.; Fu, G. C. Room-temperature Alkyl- alkyl Suzuki Cross-coupling of Alkyl Bromides that Possess Beta Hydrogens. J. Am. Chem. Soc. 2001, 123, 10099.

9. Suzuki, A. Recent Advances in the Cross-Coupling Reactions of Organoboron Derivatives with Organic Electrophiles, 1995-1998. J. Organomet.Chem. 1999, 576, 147.

10. Chemler, S. R.; Trauner, D.; Danishefsky, S. J. The B-alkyl Suzuki-Miyaura cross-coupling reaction: Development, mechanistic study, and applications in natural product synthesis. Angew. Chem. Int. Ed. 2001, 40, 4544.

11. Batey, R. A.; Thadani, A. N.; Smil, D. V. Potassium Alkenyl- and Aryltrifluoroborates: Stable and Efficient Agents for Rhodium-Catalyzed Addition to Aldehydes and Enones. Organic Lett. 1999, 1, 1683.

12. Sakai, M.; Ueda, M.; Miyaura, N. Rhodium-Catalyzed Addition of Organoboronic Acids to Aldehydes. Angew. Chem. Int. Ed. 1998, 37, 3279.

33

13. Takaya, Y.; Ogasawara, M.; Hayashi, T.; Sakai, M.; Miyaura, N. Rhodium- Catalyzed Asymmetric 1,4-Addition of Aryl- and Alkenylboronic Acids to Enones. J. Am. Chem. Soc. 1998, 120, 5579.

14. Molander, G. A.; Ribagorda, M. Expanding Organoboron Chemistry: Epoxidation of Potassium Organotrifluoroborates. J. Am. Chem. Soc. 2003, 125, 11148.

15. Brown, H. C.; Zweifel, G. A Stereospecific Cis Hydration of the Double Bond in Cyclic Derivatives. J. Am. Chem. Soc. 1959, 81, 247.

16. Brown, H. C.; Zweifel, G. Hydroboration. VII. Directive Effects in the Hydroboration of Olefins. J. Am. Chem. Soc. 1960, 82, 4708.

17. Brown, H. C.; Scouten, C. G.; Wang, K. K. Unusual Kinetics for the Hydroboration of Alkenes with -Borabicyclo[3.3.1]nonane. J. Org. Chem. 1979, 44, 2589.

18. Wang, K. K.; Brown, H. C. Hydroboration Kinetics. 2. Improved Procedure for Following the Kinetics for the Reaction of Alkenes with 9- Borabicyclo[3.3.1]nonane - Further Evidence for the Dissociation Mechanism. J. Org. Chem. 1980, 45, 5303.

19. Wang, K. K.; Brown, H. C. Hydroboration Kinetics. 6. Hydroboration of Alkenes with 9-borabicyclo[3.3.1]nonane Dimer and 9-borabicyclo[3.3.1]nonane Lewis Base Complexes in Various Solvents- an Interpretation of the Catalytic Effect of Ether Solvents on the Hydroboration Reaction. J. Am. Chem. Soc. 1982, 104, 7148.

20. Chandrasekharan, J.; Brown, H. C. Hydroboration Kinetics.11. A Reinvestigation of the Kinetics of Hydroboration of Representative Alkenes with Disiamylborane Dimer - Conclusive Evidence for the Dissociative Mechanism in the Hydroboration of Alkenes with Dialkylborane Dimers J. Org. Chem. 1985, 50, 518.

21. Brown, H. C.; Chandrasekharan, J. Mechanism of Hydroboration of Alkenes with Borane-Lewis Base Complexes- Evidence That the Mechanism of the Hydroboration Reaction Proceeds Through a Prior Dissociation of Such Complexes. J. Am. Chem. Soc. 1984, 106, 1863-1865.

22. Pasto, D. J.; Cheng, T. C.; Lepeska, B. Measurement of Kinetics and Activation Parameters for Hydroboration of Tetramethylethylene and Measurement of Isotope-Effects in Hydroboration of Alkenes. J. Am. Chem. Soc. 1972, 94, 6083.

23. Pasto, D. J.; Lepeska, B.; Balasubr.V Measurement of Relative Rate Ratios of First and Second Steps of Hydroboration Reaction and Rates of Alkylborane Redistribution Reactions. Discussion of the Overall Mechanism of the Hydroboration Reaction. J. Am. Chem. Soc. 1972, 94, 6090.

34

24. Van Eikema Hommes, N. J. R.; Schleyer, P. V. 3-Center Transition Structures for Alkene Hydroboration and Alkylborane Rearrangement. J. Org. Chem. 1991, 56, 4074.

25. DiMare, M. Ab Initio Computational Examination of Carbonyl Reductions by Borane: The Importance of Lewis Acid-Base Interactions. J. Org. Chem. 1996, 61, 8378.

26. Clark, T.; Wilhelm, D.; Schleyer, P. V. Mechanism of Hydroboration in Ether Solvents. A Model ab initio Study. J. Chem. Soc. Chem. Commun. 1983, (11), 606-608.

27. Brown, H. C.; Sharp, R. L. Hydroboration. 24. Directive Effects in Hydroboration of some Substituted Styrenes. J. Am. Chem. Soc. 1966, 88, 5851.

28. Brown, H. C.; Cope, O. J. Hydroboration. 23. Directive Effects in Hydroboration of Representative Allyl Derivatives. The Elimination of -Substituted Organoboranes. J. Am. Chem. Soc. 1964, 86, 1801.

29. Hoveyda, A. H.; Evans, D. A.; Fu, G. C. Substrate-Directable Chemical Reactions. Chem. Rev. 1993, 93, 1307.

30. Wilt, J. W.; Narutis, V. P. Competitive Exo Hydroboration of Syn-7- Arylnorbornenes. J. Org. Chem. 1979, 44, 4899.

31. Brown, H. C.; Kawakami, J. H. Additions to Bicyclic Olefins. 1. Stereochemistry of Hydroboration of Norbornene, 7,7-Dimethylnorbornene, and Related Bicyclic Olefins - Steric Effects in 7,7-Dimethylnorbornyl System. J. Am. Chem. Soc. 1970, 92, 1990.

32. Gassman, P. G.; Marshall, J. L. Synthesis and Solvolysis of 7-Ketonorbornyl Tosylates. J. Am. Chem. Soc. 1966, 88, 2822.

33. Zweifel, G.; Najafi, M. R.; Rajagopalan, S. Hydroboration of Methoxyenynes - A Novel Synthesis of (E)-Methoxyenones. Tetrahedron Lett. 1988, 29, 1895.

34. Jung, M. E.; Karama, U. Highly diastereoselective Markovnikov hydration of 3,4- dialkoxy-1-alkenes and 4,5-dialkoxy-2-alkenes via a hydroboration-oxidation process. Tetrahedron Lett. 1999, 40, 7907.

35. House, H. O.; Melillo, D. G. Perhydroindan Derivatives. 16. Sythesis of Racemic Epiallogibberic Acid. J. Org. Chem. 1973, 38, 1398.

36. Welch, M. C.; Bryson, T. A. Boron Annulation in Organic Synthesis. 3. Stereoselectivity and the Formal Synthesis of (+/-) Helenalin. Tetrahedron Lett. 1989, 30, 523.

35

37. Cha, J. S.; Seo, W. W.; Kim, J. M.; Kwon, O. O. Thexylalkoxyborane as Hydroborating Agent for Alkenes and . Bull. Kor. Chem. Soc. 1996, 17, 892.

38. Schulte-Elte, K. H.; Ohloff, G. Uber Eine Aussergewohnliche Stereospezifitat Bei der Diastereomeren (1R)-Isopulegole mit Diboran. Helv. Chim. Acta 1967, 50, 153.

39. Panek, J. S.; Xu, F. Diastereoselectivity in the Borane Methyl Sulfide Promoted Hydroboration of -Alkoxy--unsaturated Esters - Documentation of an Alkoxy- Directed Hydroboration Reaction. J. Org. Chem. 1992, 57, 5288.

40. Panek, J. S.; Xu, F.; Rondon, A. C. Chiral crotylsilane-based approach to benzoquinoid ansamycins: Total synthesis of (+)-macbecin I. J. Am. Chem. Soc. 1998, 120, 4113.

41. Schmid, G.; Fukuyama, T.; Akasaka, K.; Kishi, Y. Synthetic Studies on Polyether Antibiotics. 4. Total Synthesis of Menensin. 1. Stereocontrolled Synthesis of the Left Half of Monensin. J. Am. Chem. Soc. 1979, 101, 259.

42. Smith, A. B.; Yokoyama, Y.; Huryn, D. M.; Dunlap, N. K. Total Synthesis of (+)- Mikrolin. Tetrahedron Lett. 1987, 28, 3659.

43. Cha, J. S.; Chang, S. W.; Kim, J. M.; Kwon, O. O.; Chun, J. H.; Cho, S. D. Reaction of Thexylalkoxyboranes with Selected Organic Compounds Containing Representative Functional Groups Comparison of Reducing Characteristics of the Alkoxy Derivatives. Bull. Kor. Chem. Soc. 1998, 19, 243.

44. Brown, C. A.; Krishnamurthy, S. Facile Reaction of Alcohols and Phenols with Borane-Methyl Sulfide. A New, General, and Convenient Synthesis of Borate Esters. J. Org. Chem. 1978, 43, 2731.

45. Brown, H. C.; Park, W. S.; Cha, J. S.; Cho, B. T.; Brown, C. A. Addition Compounds of Alkali Metal Hydrides. 28. Preparation of Potassium Dialkoxyalkylborohydrides from Cyclic Boronic Esters. A New Class of Reducing Agents. J. Org. Chem. 1986, 51, 337.

46. Brown, H. C.; Kulkarni, S. U. Organoborane.24. Facile Substitution and Exchange Reactions of 9-Borabiycyclo[3.3.1]nonane (9-BBN) and its B- Substituted derivatives. Simple Convenient Preparations of B-Halo Derivatives. J. Organomet. Chem. 1979, 168, 281.

47. Brown, H. C.; Gupta, S. K. Hydroboration. 39. 1,3,2-Benzodioxaborole (Catecholborane) as a New Hydroboration Reagents for Alkenes and Alkynes - General Synthesis of Alkaneboronic and alkeneboronic Acids and Esters via Hydroboration - Directive Effects in Hydroboration of Alkenes and Alkynes with Catecholborane. J. Am. Chem. Soc. 1975, 97, 5249-5255.

36

48. Tucker, C. E.; Davidson, J.; Knochel, P. Mild and Selective Hydroborations of Functionalized Alkynes and Alkenes Using Pinacolborane. J. Org. Chem. 1992, 57, 3482.

49. Heathcock, C. H.; Jarvi, E. T.; Rosen, T. Acyclic Stereoselection. 21. Synthesis of an Ionophore Synthon Having 4 Asymmetric Carbons by Sequential Aldol Addition, Claisen Rearrangement and Hydoboration. Tetrahedron Lett. 1984, 25, 243.

50. Rauk, A.; Hunt, I. R.; Keay, B. A. Lewis Acidity and Basicity: An Ab Initio Study of Proton and BF3 Affinities of Oxygen-Containing Organic Compounds. J. Org. Chem. 1994, 59, 6808.

51. Beak, P. Determinations of Transition-State Geometries by the Endocyclic Restriction Test: Mechanisms of Substitution at Nonstereogenic Atoms. Acc. Chem. Res. 1992, 25, 215.

52. Baldwin, J. E. Rules for Ring-Closure. J. Chem. Soc., Chem. Commun. 1976, (18), 734-736.

53. Brown, H. C.; Unni, M. K. Hydroboration. 25. Hydroboration of 3-Butenyl Derivatives Containing Representative Substituents. J. Am. Chem. Soc. 1968, 90, 2902.

54. Brown, H. C.; Gallivan, R. M. Hydroboration. 26. Hydroboration of 2-Butenyl (Crotyl) and Related Derivatives Containing Representative Substituents. Control of Elimination Reaction of -Substituted Organoboranes. J. Am. Chem. Soc. 1968, 90, 2906.

55. Pasto, D. J.; Hickman, J. Transfer Reactions Involving Boron. 16. The Hydroboration of 3- and 4-Heterosubstituted Cyclohexenes. J. Am. Chem. Soc. 1968, 90, 4445.

56. Kocienski, P.; Stocks, M.; Donald, D.; Perry, M. A Synthesis of the C24-C34 Segment of FK 506. Synlett 1990, (1), 38.

57. Linde, R. G.; Egbertson, M.; Coleman, R. S.; Jones, A. B.; Danishefsky, S. J. Efficient Preparation of Intermediates Corresponding to C22-C27 and C28-C34 of FK-506. J. Org. Chem. 1990, 55, 2771.

58. Brown, H. C.; Chen, J. C. Hydroboration. 57. Hydroboration with 9- Borabicyclo[3.3.1]nonane of Alkenes Containing Representative Functional Groups. J. Org. Chem. 1981, 46, 3978.

59. Gung, B. W.; Ohm, K. W.; Smith, D. T. Regiofacial and Diastereofacial selective Hydroboration of Chiral Allylic Stannanes, Silanes, and Germanes. Synth. Commun. 1994, 24, 167.

37

60. Sibi, M. P.; Li, B. Q. Regioselective and Stereoselective Hydroborations of Chiral Allyl Amines- Synthesis of Amino Alcohols Tetrahedron Lett. 1992, 33, 4115.

61. Luef, W.; Vogeli, U. C.; Keese, R. Hydroboration and Oxymercuration of Some 1-Substitued Norborn-2-enes. Helv. Chim. Acta 1983, 66, 2729.

62. Houk, K. N.; Rondan, N. G.; Wu, Y. D.; Metz, J. T.; Paddon-Row, M. N. Theoretical Studies of Stereoselective Hydroborations. Tetrahedron 1984, 40, 2257.

63. Carceller, E.; Castello, A.; Garcia, M. L.; Moyano, A.; Serratosa, F. Regioselective Functionalization of Cis-Bicyclo[3.3.0]octenone Derivatives. Oxymercuration/Reduction versus Hydroboration/Oxidation. Acetal Groups as Regioselective and Stereoselective Control Elements. Chem. Lett. 1984, (5), 775.

64. Kono, H.; Ito, K.; Nagai, Y. Oxidative Addition of 4,4,6-Trimethyl-1,3,2- dioxaborinane and Benzo[1,3,2]dioxaborole to Tris(triphenylphosphine)halogenorhodium. Chem. Lett. 1975, (10), 1095.

65. Crudden, C. M.; Hleba, Y. B.; Chen, A. C. Regio- and Enantiocontrol in the Room-temperature Hydroboration of Vinyl Arenes with Pinacol Borane. J. Am. Chem. Soc. 2004, 126, 9200.

66. Evans, D. A.; Fu, G. C.; Hoveyda, A. H. Rhodium(I)-Catalyzed Hydroboration of Olefins- The Documentation of Regiochemical and Stereochemical Control in Cyclic and Acyclic Systems. J. Am. Chem. Soc. 1988, 110, 6917.

67. Evans, D. A.; Fu, G. C.; Hoveyda, A. H. Rhodium(I)-Catalyzed and Iridium(I)- Catalyzed Hydroboration Reactions - Scope and Synthetic Applications. J. Am. Chem. Soc. 1992, 114, 6671.

68. Garrett, C. E.; Fu, G. C. Exploiting (5)- to (3)-indenyl ring slippage to access a directed reaction: Ether-directed, rhodium-catalyzed olefin hydroboration. J. Org. Chem. 1998, 63, 1370.

69. Brinkman, J. A.; Nguyen, T. T.; Sowa, J. R. Trifluoromethyl-substituted indenyl rhodium and iridium complexes are highly selective catalysts for directed hydroboration reactions. Org. Lett. 2000, 2, 981-983.

70. Rubina, M.; Rubin, M.; Gevorgyan, V. Catalytic enantioselective hydroboration of cyclopropenes. J. Am. Chem. Soc. 2003, 125, 7198.

71. Evans, D. A.; Muci, A. R.; Sturmer, R. Samarium(III)-Catalyzed Hydroboration of Olefins with Catecholborane- A General Approach to the Synthesis of Boronate Esters. J. Org. Chem. 1993, 58, 5307.

38

72. Lee, Y. M.; Hoveyda, A. H. Efficient Boron-Copper Additions to Aryl-Substituted Alkenes Promoted by NHC-Based Catalysts. Enantioselective Cu-Catalyzed Hydroboration Reactions. J. Am. Chem. Soc. 2009, 131, 3160.

73. Burgess, K.; Ohlmeyer, M. J. Transition-Metal Promoted Hydroborations of Alkenes, Emerging Methodology for Organic Transformations. Chem. Rev. 1991, 91, 1179.

39

Chapter 2: Metal-Free Oxygen-Directed Hydroboration

2.1 A New Approach to Heteroatom-Directed Hydroboration

Despite repeated efforts over many years and several tantalizing empirical results that suggest oxygen-directed hydroboration (ODHB), definitive examples of this process remain elusive.1-13 Evans‟ metal catalyzed reactions of catecholborane with several unsaturated alcohols, phosphinites, and carboxamides are the only methods known to date with established synthetic potential for a range of substrates.2-5 Another case of

ODHB involving an -methoxy--unsaturated ester was encountered by Panek et al.6,13 using Me2S·BH3 (BMS). This example approaches the regioselectivity of the Evans result with a homoallylic alcohol (8:1 vs. 11:1), but appears to be a special case reflecting unusual reactivity due to the combined presence of an ester and an alkoxy group in the starting material. The other historical examples reveal interesting perturbations of hydroboration stereoselectivity or regioselectivity by oxygen substituents,7-12 but these reactions generally do not give useful product ratios. The purpose of the work described in this chapter is to demonstrate a mechanistically distinct version of ODHB using metal- free conditions that provide regiocontrol in the hydroboration of generic homoallylic alcohols.

The dissociative mechanism of hydroboration presented by Brown and widely accepted by the chemical community suggests that heteroatom-directed hydroboration is not possible, in that it requires a trivalent borane species to form via dissociation from a

Lewis base complex before it can react with an olefin (Scheme 2A).14-19 On the other hand, Pasto has reported an associative mechanism that involves an olefin complexing to a borane and undergoing hydroboration while the borane maintains an interaction with a

40

Lewis base (Scheme 2B).20,21 This lesser-known mechanistic proposal has received support in the literature22-24 and is one source of inspiration for the pursuit of intramolecular hydroboration via activation of heteroatom-borane complexes.

Another source of inspiration is a study by Schleyer et al., who have supported an

SN2-like complexation of an olefin to a borane-Lewis base complex as the mechanism for intermolecular hydroboration.22 The simplest version of this reaction cannot occur intramolecularly in a 5- or 6-membered cyclic transition state, based on the work of

Beak25 and Baldwin,26 because the 180° orbital geometry required to form the trigonal bipyramidal SN2 transition state via 5-endo-tet and 6-endo-tet transition states is disfavored. Therefore, the „5.5-endo-tet’ transition state 2C2 necessary to invoke an

41 intramolecular variant of Pasto‟s mechanism is also disfavored. On the other hand,

5-exo-tet and 6-exo-tet processes are highly favored in contrast to the 5-endo-tet and 6- endo-tet. Thus, if one could introduce an exo-leaving group on boron, then the concepts put forth by Pasto and Baldwin would no longer be dissonant in the context of an intramolecular hydroboration pathway (Scheme 2C). Previous work in the Vedejs group investigated the effects of introducing such a leaving group into homoallylic amine- borane systems with good results, as described below.27,28

Homoallylic amine-iodoboranes 2D3 were generated by activating purified

29 amine-boranes 2D1 with iodine (I2) according to the work of Ryschkewitsch. These amine-iodoboranes, while designed to allow intramolecular olefin-borane complexation via an SN2-like associative pathway, might also achieve the desired complex 2D5 via a dissociative SN1 pathway (Scheme 2D), although no evidence for this was found. Once the intramolecular -complex 2D5 is formed, hydroboration can occur by either a fused-

(2D8) or bridged (2D7) bicyclic transition state to generate 1,3- and 1,4-aminoalcohols

2D11 and 2D10, respectively, upon oxidation. The kinetic advantage of 5-membered ring formation, and the greater thermodynamic stability of the fused bicyclic transition state 2D8, led to the prediction that 1,3-aminoalcohols 2D11 would predominate in the product mixture. The expected selectivity was observed: up to >20:1 selectivity was achieved with both E- and Z-1,2-disubstituted olefin substrates.27,28 In the case of a terminal olefin, anti-Markovnikov selectivity was drastically reduced from 1:19

30 (expected with THF·BH3) to 1:3. Furthermore, -substituted styrene (R=Ph) provided a

2:1 preference for the 1,3-amino-alcohol 2D11 after oxidative workup, in contrast to the typical 1:5 ratio expected for -substituted styrene.30

42

2.2 Regioselectivity via Metal-Free Alcohol-Directed Hydroboration

The goal of the work reported in the remainder of this chapter was to develop a metal-free method for oxygen-directed hydroboration through application of the amine- direction precedents to the hydroboration of homoallylic alcohols, alkoxides, and ethers.

The investigation began with a focus on alcohol-directed hydroboration.

While an analogous activation approach for alcohol-directed hydroboration using a potentially exocyclic leaving group might appear to be facile (Scheme 2E), amine-

43 borane and alcohol-borane complexes are very different in terms of stability. This leads to restrictions on the alcohol system that were of no concern in the amine system. While many amine-boranes can be generated by stirring the amine with THF·BH3 and then purified by chromatography or crystallization, these are not options for alcohol-boranes in the context of ODHB. Alcohol-boranes spontaneously evolve hydrogen gas at or below room temperature to generate the presumably undesired alkoxy boranes

RO(3-n)BH(n), thereby eliminating the option of purification by chromatography.

Furthermore, generating an alcohol-borane in ethereal solvents commonly associated with hydroboration is not practical for ODHB since the solvent would compete for

44 borane complexation. Thus, if alcohol-directed hydroboration is to occur in an analogous fashion compared to the amine case, the alcohol-boranes must be generated and activated in situ at low temperatures with no ethereal solvents present.

Dichloromethane (DCM) was a straightforward solvent choice based on work from both the Brown and Vedejs groups.27,31 Choosing a borane carrier was a more delicate matter in that the ideal carrier must find a balance between the contrasting properties of 1) forming a sufficiently strong complex with borane to minimize the concentration of undesired Lewis base in solution and 2) forming a weak enough borane complex to allow the substrate to complex to borane. Thioanisole was chosen based on a study by Brown, which reported that borane dissolves in thioanisole with 3 M

32 concentration despite a relatively weak complex between PhSMe and BH3. Such solutions are not commercially available due to their instability over time, but are relatively easy to generate in 10 mL batches by bubbling excess diborane gas through neat PhSMe. This procedure consistently provided borane concentrations of 2.5 M.

The first investigations into the activated borane approach to ODHB were carried out with 3-penten-1-ol in DCM and thioanisole-borane [Ph(Me)S·BH3] activated by iodine (Table 2F; Entries1-7). A favorable regioselectivity of 7:1 for 1,3-pentanediol was achieved in the initial experiment in which the substrate/borane solution was activated at

-40 °C and was then warmed to between -20 °C and 0 °C (Entry 1). Subsequent control experiments demonstrated that unactivated Ph(Me)S·BH3 was capable of hydroborating substrate at -40 °C, resulting in a reduced 3:1 regioselectivity for 1,3-diol 2F2 (Entry 3) due to undesired background reaction. Regioselectivities were dramatically improved by introducing both Ph(Me)S·BH3 and iodine to the substrate solution at lower temperatures.

45

Entry R1 R2 Activator X (°C) T1 (h) T2 (h) 1,3-: 1,4- Conversion

1 Ph Me I2 -40 1 5 7 : 1 total 2 Ph Me - -20 6 n/a 3 : 1 total 3 Ph Me - -40 6 n/a 3 : 1 partial 4 Ph Me - -78 6 n/a - trace

5 Ph Me I2 -60 1 5 13 : 1 total

6 Ph Me I2 -78 1 5 18 : 1 total

7 Ph Me I2 -78 5 n/a 18 : 1 trace

8 Me Me I2 -78 1 10 4.4 : 1 partial

9 Me Me I2 -78 1 19 4.4 : 1 total

Entries 5 and 6 display regioselectivities of 13:1 and 18:1 from activation temperatures

of -60 °C and -78 °C, respectively. Entry 7 shows that the reaction had to be warmed

from -78 °C to proceed at an acceptable rate.

These results with Ph(Me)S·BH3 were an encouraging indication that ODHB

could be achieved in analogous fashion to the amine-directed methodology. However,

the short-lived nature of Ph(Me)S·BH3 solutions, coupled with safety concerns for its

generation, and its unenviable odor led to the pursuit of alternative borane sources for a

method viable on scale. Dimethylsulfide borane (BMS) was investigated because it is

commercially available. Its scent would be forgivable if it could provide comparable

regioselectivity to Ph(Me)S·BH3. Not only did BMS fail to provide comparable

46 regioselectivity (4.4:1 vs. 18:1), it required longer reaction times to achieve total conversion (Table 2F; Entries 8 & 9).

Previous statements pertaining to the characteristics of an ideal borane carrier are perhaps misleading in that the most ideal borane carrier in the pursuit of ODHB is no carrier at all. Therefore efforts were made to generate solutions of (di)borane in DCM.

A DCM·BH3 stock solution approach was studied by bubbling excess diborane gas through DCM cooled to -78 °C. Borane concentrations of 0.44M were generated in

DCM, but loss of borane was observed upon warming. Table 2G illustrates that stock solutions of 1:1 PhSMe:BH3 generated in situ from a variety of borohydrides were also not viable as storable reagents. Sodium borohydride (NaBH4) failed to generate any detectable borane due to the insolubility of NaBH4 in DCM, but tetrabutylammonium borohydride (nBu4NBH4) did give excellent initial concentrations of borane upon treatment with iodine. It was hypothesized that replacing the sulfide carrier with an alcohol substrate would allow borane retention in solution to achieve ODHB upon activation with I2, but uncertainty would remain regarding the timescale for activation vs. the timescale for reaction of the alcohol with borane to generate borates and H2 gas.

% BH3 in solution R After 3 h After 15 h After 7 d Na 0 0 0

n-Bu4N 90 66 50 Bn(Et)3N - 54 47

47

Conversion of the unsaturated alcohol substrate was achieved using in situ borane generation from nBu4NBH4 followed by iodine activation, but regioselectivity dropped from 18:1 to ~ 2:1. To ensure that this decrease in selectivity was not due to the absence of sulfide, PhSMe was added to the reaction solution. As no change in regioselectivity was observed, it was concluded that presence of sulfide has no effect on the reaction

(Scheme 2H). On the other hand, replacing I2 with triflic acid (TfOH) as the borane- generating and activating reagent provided excellent >20:1 regioselectivity with 3-nonen-

1-ol under sulfide-free conditions (Scheme 2I). By itself, this result is significant in that it confirms that sulfide is not necessary for achieving good regioselectivity. Considering the impact that TfOH activation had on the eventual development of an ODHB method for alcohols, this result proved to be absolutely critical.

48

The I2-activated Ph(Me)S·BH3 method that provided excellent selectivity on 3- penten-1-ol (Table 2F; Entry 6) failed to provide comparable results on more complex substrates. Even 3-hexen-1-ol failed to give good regioselectivities. Yields were also too low to be synthetically practical. Fortunately, it was discovered that TfOH-activated

BMS achieved moderate to good yields on a variety of substrates with superior regioselectivity than that of the I2-Ph(Me)S·BH3conditions. Best results were achieved when BMS was preactivated with TfOH before addition of substrate (Table 2J).33

All alkyl-substituted substrates provided 1,3-diols with excellent regioselectivities. Selectivity is not highly dependent on olefin geometry, although trans- olefins provide a somewhat improved result, as 28:1 and 37:1 regioselectivities were achieved on cis- and trans- 3-hexen-1-ol, respectively (Table 2J; Entries 2&3).

Branching at C-5 is directly related to regioselectivities as demonstrated by the secondary cyclohexyl group and the tertiary tbutyl group, respectively, providing 56:1 and 82:1 regioselectivities (Entries 5&6).

Aromatic substituents have a strange and unexpected effect on substrate reactivity that seemingly relates to their proximity to the olefin. The styrenic olefin in entry 7 underwent only trace conversion even after extended reaction times. While the styrene system was of particular interest due to the electronic perturbation of the olefin by the phenyl ring, it had been expected that this perturbation would be a challenge in terms of regioselectivity rather than reactivity. In general, -substituted styrenes undergo intermolecular hydroboration to generate benzylic boranes with a 5:1 regioselectivity preference due to the conjugative electronic effect of the phenyl ring. However, the observed drop in reactivity in this case cannot be attributed only to conjugation, as

49

Table 2J: Regioselectivities Achieved Under Optimized Conditions Me S BH3 OH 1. Me , TfOH OH OH OH R R R CH2Cl2, -78°C OH 2. -78°C to -20°C, time 2J1 3. MeOH, NaOOH 2J2 2J3

a R= CH3 d R= nCH5H11 g R= Ph b R= CH2CH3 e R= cC6H11 h R= Bn c R= Z-C2H5 f R= tC4H9 i R= (CH2)2Ph

Entry Stg Time (h) Yield 2J2 : 2J3 1 2J1a 10 41% >20 : 1 2 2J1b 10 51% 37 : 1 3 2J1c 5 51% 28 : 1 4 2J1d 10 69% > 20 : 1 5 2J1e 5 80% 56 : 1 6 2J1f 5 56% 82 : 1 7 2J1g 20 < 3% n/d 8 2J1h 10 22% n/d 9 2J1i 5 59 % >20 : 1

inserting a methylene tether between the phenyl ring and olefin does not restore reactivity comparable to that of the alkyl substrates (Entry 8). Not until a second methylene linker is added to the phenyl-olefin tether does reactivity return to expected levels (Entry 9).

The styrenic system behaves well in the amine directed hydroboration. Therefore, a 1:1 substrate-borane complex would be expected to behave well in the alcohol system.

However, a 1:1 complex is not a realistic assumption to make with a single equivalent of activated borane present, let alone the two equivalents present in the reaction mixture

(standard conditions). It is tentatively proposed that activated borane forms a -complex

50 to the aromatic ring that inductively deactivates the olefin. Adding a methylene tether reduces the inductive effect and a second methylene group renders it insignificant.

Having achieved excellent regioselectivity on a variety of homoallylic alcohols the next goal for this work was to confirm that the regioselectivity was due to ODHB.

This was investigated by running competition experiments in the presence of excess cyclohexene. Not only was no cyclohexanol observed upon oxidative workup of these reactions, but the yield of the desired diols actually increased by 10 % (Table 2K). This is believed to be due to the cyclohexene behaving as a scavenger of acidic species in solution.

Entry R Regioselectivity Yield (without additive) Cyclohexanol Observed 1 Me >20 : 1 51% (41%) none 2 Et >20 : 1 66% (51%) none

These acidic species are responsible for catalyzing a cyclization side reaction that went unobserved until special care was taken with the reaction mixture of 6-phenyl-4- hexen-1-ol 2L1. 2-Phenethyltetrahydrofuran 2L3 was recovered in low yield due to its surprising volatility (Scheme 2L), but its formation suggests that a similar side reaction accounts for the moderate yields in Table 2J. Despite good conversion of the aliphatic substrates in Table 2J, the moderate yields were initially baffling because the volatile tetrahydrofurans have the same mass as the corresponding starting materials, rendering

51 assay of the crude reaction mixtures by nominal mass spectrometry ineffective. Thus the missing mass balance is attributed to the volatile tetrahydrofuran side products.

The successful ODHB of homoallylic alcohols led to the investigation of other unsaturated alcohol substrates. 4-Hexen-1-ol 2M1 was used to study a bis-homoallylic alcohol with unexpected results. It was believed that good regioselectivity for 1,4-diols would be achieved in such systems for the same kinetic and thermodynamic reasons that one would expect to account for the formation of 1,3 diols from homoallylic systems.

However, 1,4-hexane-diol 2M2 was generated with a modest 4.4:1 regioselectivity compared to 1.6: 1 using THF·BH3 (Scheme 2M).

Scheme 2M: Investigating Bishomoallylic Substrates OH 1. THF·BH3, THF, 0 °C OH OH OH 2. NaOOH 2M1 OH 2M2 2M3 1.6 : 1

1. BMS, TfOH, DCM OH -78 to -20 °C OH OH OH 2. NaOOH OH 2M1 2M2 2M3 4.4 : 1

52

In 1972 Brown reported that intramolecular hydroboration of thexyl-4- pentenylborane 2N1 favors the five membered ring 2N2. This must be due to a kinetic preference for five-membered ring formation, considering the well known preference of dialkylboranes to react with anti-Markovnikov selectivity.34 On the other hand, in the case of thexyl-5-hexenylborane 2N4, the kinetic preference for six-membered ring formation does not begin to compete with the inherent preference of the dialkylborane for anti-Markovnikov regioselectivity (Scheme 2N). Negligible regioselectivity was also

28 achieved upon I2 activation of bishomoallylic amine-boranes 2O1 (Scheme 2O). These two reports, in addition to the results with 2M1 support the conclusion that regioselective delivery of borane to an olefin from the -position of a simple acyclic substrate is inherently difficult, regardless of the element at the -position, attached to boron.

Entry Olefin Geometry R Yield (%) 1,4- : 1,5- 1 E H 95 2 : 1 2 Z H 91 2 : 1 3 Z Bn 63 1.5 : 1

53

The results with the bishomoallylic alcohol precluded the study of longer olefin- borane tethers so a shorter (allylic) tether was briefly evaluated using 2-hexen-1-ol.

However, the conditions used on the homoallylic substrates did not consume olefin in the allylic system. The reduced reactivity is attributed to 1) the alcohol-borane complex inductively deactivating the olefin and 2) the thermodynamically disfavored [2.2.0] fused- 2P2 and [2.1.1]-bridged 2P5 bicyclic transition states through which an intramolecular transformation would have to proceed (Scheme 2P).

Scheme 2P: Allylic Alcohols are Not Viable Substrates n-Pr H disfavored O H B O H B H 2P1 2P2 H OTf OTf

BMS,TfOH

1. BMS, TfOH, DCM -78 to -20 °C OH OH OH 2P3 2. NaOOH 2P4

BMS,TfOH

n-Pr disfavored H H B O O H H B OTf 2P1 H OTf 2P5

2.3 Pursuing Diastereoselectivity via Metal-Free Alcohol-Directed Hydroboration

Having developed a method capable of regiocontrol in the hydroboration of homoallylic alcohols, the new procedure was applied to a variety of secondary alcohols to investigate its utility in the context of diastereocontrol. Isopulegol 2Q1 was an ideal initial substrate as it is commercially available and had been previously studied by Ohloff

54

(Scheme 2Q).7 However, a background reaction run to confirm Ohloff‟s 19:1 diastereoselectivity did quite the opposite. Treating isopulegol with 1M THF·BH3 provided only 5:1 diastereoselectivity favoring 2Q2. The (unknown) reason for this discrepancy compared to the literature result is of little importance because treating

TfOH-activated BMS with isopulegol followed by subsequent oxidative workup provided superior diastereocontrol (>30 : 1) in the generation of 1,4-diols 2Q2 and 2Q3 (Scheme

2Q). This increase is tentatively attributed to ODHB, but the reaction proved surprisingly complex.

Scheme 2Q: Diasteroselectivity Achieved on Isopulegol

1. THF·BH3, THF

OH 2. MeOH, NaOOH OH OH OH OH H H 2Q1 2Q2 95 % 2Q3 5 : 1

Me 1. S BH3 Me , TfOH OH OH OH OH OH CH2Cl2, -78°C 2. -78°C to -20°C, 5h OH OH 3. MeOH, NaOOH H H OH 2Q1 2Q2 2Q3 2Q4 2Q5

18% 11% 32% >30 : 1 H2O BH2OTf

OH OH O H OTf OTf OTf 2Q6 2Q7 2Q8

Two side products were produced under ODHB hydroboration conditions.

Isolating the isomeric 1,3-diol 2Q4 in 18% yield presented the possibility that the hydroxyl directing effect might overcome the strong preference for anti-Markovnikov

55 hydroboration. However, the identification of citronellol 2Q5 as the second byproduct indicated that acidic side reactions were interfering. Since isopulegol can be prepared by acid catalyzed cyclization of citronellal, the reverse transformation should also be feasible under these conditions.35 Protonating the olefin of isopulegol would lead to a tertiary triflate 2Q6 that can undergo ring opening via carbocation 2Q7, to form protonated citronellal 2Q8, which is envisioned as being reduced by an activated borane species in solution to provide citronellol. Generation of 1,3-diol 2Q4 might come from either 2Q6 or 2Q8 upon aqueous workup. Cyclohexene did not suppress these side reactions.

Secondary homoallylic alcohols were submitted to TfOH-activated BMS conditions to investigate diastereoselectivity. Although no mechanistic evidence for

ODHB had been obtained at this stage of the investigation, several speculative rationales had emerged, one of which assumes the simplest version of alcohol-directed

56 hydroboration, shown in Scheme 2R. A priori, diastereoselectivity on E-substrates was expected based on chair-like transition states 2R1 and 2R3. Syn-1,3-diols were expected to be favored, as they would be derived from the transitions state 2R1, in which the olefin is in the extended conformation. The alternative conformer 2R3 should be disfavored due to the A-1,3 strain generated between the olefin and the pseudoaxial proton next to oxygen. Contrary to these predictions, only negligible diastereoselectivity was observed with 4-decen-2-ol 2R5 and 2,2-dimethyl-5-docecen-3-ol 2R6 (Scheme 2R).

It was hypothesized that using a Z-olefin would lead to increased diastereoselectivity due to the drastic steric interaction between the alkyl substituent and the psuedoaxial carbinol proton in the disfavored transition state 2S3. However, similarly poor results were observed on Z-substrates despite incorporating t-butyl groups at the

57 carbinol and distal olefinic carbons (Scheme 2S). It is important to note that almost no regioselectivity was observed in either the E- or Z- systems. Apparently, the steric environment of the secondary alcohol had prevented a useful alcohol-borane interaction.

Since excellent regioselectivity had only been achieved with primary alcohols, an experiment with 2-isopropyl-3-penten-1-ol 2T1 was designed. Moving the chiral center from the carbinol carbon to C-2 would allow a better alcohol-borane interaction while bulkiness of the isopropyl group should provide a stereochemical bias. Excellent regioselectivity was restored by using the primary alcohol. However, the C-2 isopropyl group provided no diastereoselectivity. This result indicates that the A-1,3 interaction of both psuedoaxial olefin and carbinol proton in transition state 2T4 is insufficient to provide a stereochemical preference over the psuedoequatorial transition state 2T3, assuming that the cyclic mechanism has been presented correctly.

Another approach to achieving diastereoselectivity was taken in which a bulkier hydroborating agent was investigated. TfOH-activated phenylborane was chosen because the introduction of steric bulk is accompanied by delocalization of electron density

58 involving boron. Generation of the desired borane species was confirmed by forming its

N,N-dimethyl-4-amino-pyridine (DMAP) complex in two different routes. Treating lithium phenylborohydride36 with an equivalent of TfOH to generate phenylborane followed by a second (activating) equivalent of TfOH and subsequent addition of an equivalent of DMAP generated a species with a 11B NMR signal at = +5ppm. An identical signal was observed upon treating DMAP-phenylborane complex with an

Scheme 2U: Pursuing Diastereoselectivity with Phenyl(triflate)borane Li H H H H DMAP B TMSCl B DMAP B H H H

2U1 2U2 2U3 11B : -4ppm

LiOTf H2 H2 TfOH TfOH

H H H DMAP B TfOH B DMAP B H OTf OTf

2U4 H2 2U5 2U6 11B : 5ppm Li H H B H 1. , TfOH OH OH

CH2Cl2, -78°C OH 2. -78°C to -20°C 2J1e 3. MeOH, NaOOH 2J2e 22 : 1 regioselectivity Li H H B H 1. , TfOH n-pentyl n-pentyl OH CH2Cl2, -78°C OH OH 2. -78°C to -20°C 2S7 3. MeOH, NaOOH 2U9 8 : 1 regioselectivity 1 : 1 diastereoselectivity

59 equivalent of TfOH. Thus it was concluded that the signal at = +5ppm represents the

DMAP complex of Ph(OTf)BH, which confirms generation of Ph(OTf)BH in the first route (Scheme 2U).

Treating a solution of Ph(OTf)BH with 4-cyclohexyl-3-buten-1-ol 2J1e provided

1,3-diol 2J2e with 22:1 regioselectivity upon oxidative workup, demonstrating that the new reagent was capable of ODHB. However, submitting both the E- and Z- isomers of

2,2-dimethyl-5-dodecen-3-ol 2S7 to the Ph(OTf)BH conditions provided moderate regioselectivity and no diastereoselectivity. The resistance of acyclic homoallylic alcohols towards diastereo-induction necessitates further consideration of the feasibility of the proposed mechanism.

Intramolecular hydroboration in analogous hydrocarbon systems is known to provide diastereoselectivity with three-to-five atom tethers between boron and an olefin.37-42 Still has reported diastereocontrol in the internal hydroboration step starting

60 from of diene and thexylborane. Dienes 2,5-dimethyl-1,4-hexadiene 2V1 and

2,7-dimethyl-2,7-octadiene 2V7 provide diastereo-enriched diol mixtures as expected from sequential inter- and intramolecular transformations. Diols 2V5 and 2V6 were generated in a 15:1 ratio upon treating E-2,6-dimethyl-1,4-heptadiene 2V4 under the same conditions. The authors state “only in the case of [2V4] is some ambiguity involved, and this is presumably due to a steric interaction between the isopropyl and thexyl substituents.” The issue of intramolecular hydroboration is of particularl interest in the case of 2V4, as the intermediate generated by monohydroboration is a hydrocarbon analog for 2R5-borane and 2R6-borane complexes. A specific transition state was not proposed for these substrates.

Chair-like transition states have been suggested, involving boranes 2W1, generated by hydroboration of a diene precursor. Yokoyama proposes that intramolecular hydroboration occurs via cyclic transitions states 2W2a and 2W2b.41,42 The lack of A-

1,3 strain between R3 and RZ in 2W2b is responsible for the selectivity. The steric bulk of thexylborane is key, as THF·BH3 provides poor selectivity. Acyclic diastereocontrol is clearly achievable via intramolecular hydroboration in the all-carbon substrates, based on the results of Still and Yokoyama. The contrasting lack of diastereoselectivity with the oxygen analogs (homoallylic alcohols of section 2.3) cannot be accounted for by inserting oxygen into Yokoyama‟s proposed transition states. The poor selectivity with the original activated borane conditions (Schemes 2R, 2S, & 2T) is perhaps less confusing considering the result in entry 1 of Table 2V. On the other hand, the steric bulk of the PhB(H)OTf reagent should provide improved selectivity, acting as a steric

61

entry R1 R2 R3 RE RZ anti- : syn- reference

1 H H Me -(CH2)5- 76 : 24 41 2 H Thx Me -(CH2)5- 96 : 4 41 3 H Thx Me CH2OBn Me 96 : 4 42 4 H Thx Me Me CH2OBn >98 : <2 42

replacement for a thexyl group, but no improvement was observed. Thus, the mechanism of the alcohol directed hydroboration is clearly more complicated than initially proposed, based on these contrasts with the hydrocarbon analogy.

2.4 Mechanistic Investigations

Mechanistic studies were done with the intent of supporting an intramolecular pathway. The first experiment investigated the activated borane reagent using 11B NMR spectroscopy. A -20 °C solution of BMS 2X1 (11B: = -20.7 ppm, q, J= 104 Hz) in

11 CH2Cl2 was treated with slightly less than one equiv of TfOH to provide a B triplet at

= -1.6 ppm (J= 128 Hz), which is believed to represent the expected dimethyl sulfide complex of triflate borane 2X2.

62

To investigate the reactivity activated borane 2X2 with alcohols without the complication of olefin reactivity interfering, 2X2 was treated with ethanol at -78 °C and warmed to -20 °C while monitoring by 1H NMR spectroscopy. Interestingly, the proton

NMR spectra taken at -78 ºC and -20 °C showed a temperature dependant signal shift. At

-78 °C a single peak was observed for both the methylene (= 3.76 ppm) and methyl (=

1.2 ppm) protons of ethanol. As the sample warms, the methylene (= 3.76; 3.94 ppm) and methyl (= 1.2; 1.29 ppm) protons appear as two signals. After 10 min at -20 °C the more downfield signal is dominant. This observation suggests that either 1) the alcohol does not complex to the sulfide-complexed BH2OTf or 2) an ethoxyborane is forming at the elevated temperature. The latter was addressed with hydrogen evolution studies.

Quantitative hydrogen evolution upon treating 2X2 with an alcohol would prove that complex 2X3 is transitory under reaction conditions, which would eliminate the possibility of the envisioned mechanistic pathway. Hydrogen evolution studies began with a control experiment in which a -20 °C DCM solution of Me2S·BH3 was treated with ethanol, generating 9% of the theoretical amount of H2 over 90 minutes (Equation

2). Boron NMR spectroscopy of the resulting mixture was complex but no monoalkoxyborane was observed at the expected value (= ~50 ppm). These results indicate that ethoxyboranes are not formed under such conditions.

63

H EtOH (0.51 mmol) H CH2Cl2 Me B H (2) S H 2 -20 °C, 90 min Me 9% 2X1 .6 mmol .05 mmol

A more intricate experiment was required to investigate the real reaction conditions. The first stage of the experiment involved measuring the amount of hydrogen evolved from the activation of 2X1 with TfOH. Treating a -78 °C solution of 2X1 with

1 equivalent of TfOH led to the collection of >90% of the theoretical amount of H2.

Substrate was then added via addition funnel to the reaction mixture at -78 °C. No gas evolution was observed at -78°C. Upon warming to -20 °C, ca. 10% of the theoretical volume of H2 was collected (Scheme 2X). Assuming that complexes of the alcohol and

BH2OTf are the species observed at low temperature, then they appear not to decompose to monoalkoxyboranes at -20°C: 11B NMR spectroscopy does not reveal monoalkoxyborane peaks in the expected range (= 50 to 55 ppm) amid complex signals including maxima at = -20.7 ppm, -8.43 ppm, -1.7 ppm, and +35.2 ppm. Broad signals obscure the spectrum in the range of -18 ppm < < +24 ppm.

Scheme 2X: Hydrogen Evolution Studies OH H H H TfOH OTf Et Me B Me B S H H2 S H CH Cl , -78 °C Me 2 2 Me CH2Cl2, -78 °C >90% 30 min 2X1 2X2

H " O T f " H B O H -78 to -20 °C Et H2 45 min 10% 2X3

64

The requisite species for achieving intramolecular hydroboration, according to the original mechanistic proposal, is the substrate-BH2OTf complex 2X3 (Scheme 2X). The proton remaining on oxygen, along with the resulting positive charge, serves to activate the tetravalent boron by making it more electrophilic. Hydrogen evolution studies indicate that a small (ca. 10%) amount of 2X3 undergoes hydrogen evolution which could form an alkoxyborane 2Y5, (Scheme 2Y) by reacting with an external hydride source, or perhaps more plausibly, the corresponding trifluorosulfonyloxy analogue 2Y1 if hydrogen evolution involves the internal hydride of 2X3. While monoalkoxyboranes are not confirmed as hydroborating species, formation of alkoxyborane 2Y5 should not be taken to preclude ODHB. Similarly, formation of 2Y1 has not been detected nor does it have any precedent in the literature, but that does not mean that 2Y1 can be ruled out as one of the species capable of undergoing internal hydroboration.

The reaction conditions include two equivalents of activated borane. This enables potentially unreactive intermediate 2Y5 (Scheme Y) to complex to another molecule of activated borane, forming the bis-borane species 2Y6, which is analogous to 2Y3 where the activating proton has been replaced by a Lewis acid. Thus, the transition state 2Y2 can still be achieved by all substrates regardless of any possible monoalkoxyborane formation. Monitoring the reaction of trans-3-hexen-1-ol by 11B NMR spectroscopy provided very complex spectra, but no downfield species (= 70-80 ppm) indicative of trivalent 2Y1 or either of the borenium ions 2Y6 or 2Y7 was observed. The spectra do not define a clear mechanistic picture, but the 11B signal at = -8.43 ppm (tetravalent boron; buried triplet, JBH= ~128 Hz) lends support for the formation of complex 2X3.

11 Brown has reported the B signal of a similar species BH2Cl·OEt2 as a triplet at = -5

65

Scheme 2Y: Loss of Hydrogen Does Not Preclude ODHB

OTf H OTf TfO H2 OTf R B H B X BH2 R H B O H - H2 O H O O R R 2Y1 2X3 2Y2 2Y3

Me2S·BH2OTf H H H TfO B B B BH R O H H O H X O 2 OH OH R R TfO R 2Y5 2Y6 2Y7 2Y4

43 ppm with a JBH= 136 Hz. Therefore, the originally envisioned mechanism, in which

2X3 forms complex 2Y3 via 2Y2, remains feasible.

2.5 Alkoxide-Directed Hydroboration

Acid-catalyzed side reactions are the main inconvenience of using alcohols to direct hydroboration with triflate-activated borane as demonstrated by formation of tetrahydrofuran 2L3 from 2L1, and 2Q4 and 2Q5 from isopulegol 2Q1. The alcohol- borane complex 2X3 that is believed to be responsible for intramolecular delivery of borane is also a species that could possibly generate TfOH, leading to catalysis of undesired reaction pathways. Thus, undesired reactions might be unavoidable with alcohols. Lithium alkoxide substrates were investigated for this reason.

The best substrate in the alcohol series was 4-cyclohexyl-3-buten-1-ol 2J1e; it provided an 80% yield of diol with 9% of recovered starting material (RSM), while the rest of the mass was presumably lost due to acid-catalyzed formation of a volatile tetrahydrofuran. The corresponding lithium alkoxide 2AA1e was studied to see if better mass balance could be obtained without losing regioselectivity. Almost quantitative

(98%) mass balance was achieved upon treating TfOH-activated BMS with 2AA1e,

66 followed by oxidation with NaOOH. Regioselectivity favoring 1,3-diol 2U8 remained excellent (>20 : 1), however, a decrease in yield was observed (Scheme 2Z).

A detailed investigation of the alkoxide-directed hydroboration was initiated to draw comparison to the alcohol series. Comparable regioselectivity was achieved on all aliphatic substrates. The reactivity of phenyl-substituted substrates continued to suffer, presumably due to the same inductive deactivation effects of arene-borane complexes discussed in relation to Table 2J. Comparing diol yields from alcohols with those from alkoxides reveals no consistent advantage. Substrates with little steric bulk at C-4 provide greater yields in the alkoxide experiments (Table 2AA, entries 1 & 2) while substrates with bulkier C-4 substituents provide greater yields in the alcohol experiments.

The improved results with unhindered substrates are reminiscent of the results with

Ph(Me)S·BH3/iodine conditions studied initially, which were only effective on

3-penten-1-ol (Table 2F). It appears that only substrates unhindered at C-5 benefit from the alkoxide reaction conditions. The bis-homoallylic alkoxide of 2M1 (2X2) was also investigated, and the selectivity was comparable to the selectivity for 2M1 (Equation 3).

67

Entry Stg Time (h) Yield Regioselectivity 1 2AA1a 5 70% >20 : 1 2 2AA1b 5 90% >20 : 1 3 2AA1d 5 38% > 20 : 1 4 2AA1e 5 60% >20 : 1 5 2AA1f 5 44% >20 : 1 6 2AA1g 20 < 3% n/d 7 2AA1h 20 28% n/d 8 2AA1i 5 42 % 16 : 1

Neither the alcohol nor alkoxide method is perfect. The alcohol method generally provides greater yields but acid catalyzed side reactions destroy some starting material. The alkoxide method preserves material but, in addition to generally low diol yields, the procedure is relatively inconvenient as it requires a system that incorporates a cold-jacketed addition funnel (Figure 2-1). However, the preservation of starting material in the alkoxide case

68 presented the possibility that conversion could be improved. The effect of “TfOBH2” stoichiometry on conversion was investigated in an effort to achieve higher conversion to diol. Entries 1-4 of Table 2BB illustrate that diol yield does increase with additional reagent, but starting material starts decomposing by an unknown pathway(s) with 4 equivalents of activated borane added (Entry 4). Yields never surpassed that which was obtained in the alcohols series (Entry 7).

Table 2BB: Alkoxide Conditions vs Alcohol Experiments Me S BH3 OH OR 1. Me , TfOH OH CH2Cl2, -78°C 2J1e R =H 2. -78°C to -20°C, 5h 2J2e 2AA1e R =Li 3. MeOH, NaOOH

Entry R Equiv „BMSOTf‟ Additive Regioselectivity Yield RSM 1 Li 1.1 - >20 : 1 16% 76% 2 Li 2 - >20 : 1 60% 38% 3 Li 3 - >20 : 1 65% 29% 4 Li 4 - >20 : 1 70% 7% 5 H 1.1 - >20 : 1 55% 23% 7 H 2 - >20 : 1 80% 9%

A search for a viable acid scavenger was undertaken in an attempt to suppress acid catalyzed side reactions without the use of nbutyl lithium (nBuLi). With acidic conditions providing the best reactivity, and alkoxide conditions preventing undesired side reactions, it was hypothesized that using hindered amines as acid scavengers would provide the benefits of both to provide optimal yield. Cyclohexene was included in the study because of the results in Table 2K. This was an opportunity to confirm the

69 hypothesis that acid catalysis is responsible for the generation of 2Q4 and 2Q5 from isopulegol 2Q1 under alcohol direction conditions.

Table 2CC illustrates that suppression of 2Q4 and 2Q5, accompanied by increased yields and good diastereoselectivity, was achieved with diisopropylethylamine

(Hunig‟s base) or 2,6-di-tert-butyl-4-methyl-pyridine. However, the 1,3-diol was still observed in the hindered amine experiments (Entries 3 & 4), while the lithium alkoxide approach essentially eliminated this product (Entry 5). The total suppression of 1,3-diol

2Q4 and citronellol 2Q5 in the alkoxide experiment supports the hypothesis that both

2Q4 and 2Q5 are generated via acid-catalyzed side reactions that are not completely avoidable with an amine acid scavenger. This does not discount the possibility that the amine proton scavengers could be effective on other substrates, as 2,2-disubstituted olefins are more inclined to undergo protonation than 1,2-disubstituted olefins like 2J1.

As mass balance was best with the hindered pyridine in Table 2BB, this additive was also tested with 2J1e. However, oxidative workup revealed poor conversion to the 1,3-diol with good regioselectivity (Equation 4).

70

Table 2CC: Investigating Alkoxide Conditions on Isopulegol Me 1. S BH3 , TfOH additive Me CH Cl , -78°C OH OH OH OR 2 2 OH 2. -78°C to -20°C, 5h OH OH 3. MeOH, NaOOH H H OH 2Q1 2Q2 2Q3 2Q4 2Q5

Entry R additive Q2+Q3 Q2 : Q3 Q4 Q5 RSM 1 H - 32% >30 : 1 18% 11% 20%

2 H 30% >30 : 1 22% 4% 10% (1 equiv)

3 H N 52% >20 : 1 2% 0% 18% (1 equiv)

4 H 50% >20 : 1 9% 0% 28%

5 Li - 70% ND trace 0% 21%

Me But N tBu S BH3 OH 1. Me , TfOH OH 2J1e (4) CH2Cl2, -78°C OH 2J1e Me 2. -78°C to -20°C, 5h 2J2e 3. MeOH, NaOOH 22% yield 63% >20 : 1

Lithium alkoxide experiments provided no discernable advantage in the context of diastereoselectivity on acyclic substrates. Homoallylic secondary alkoxides generated from trans- 5-decen-2-ol and trans-2,2-dimethyl-dodecen-2-ol provided nearly identically poor regio- and stereoselectivity in comparison to the corresponding alcohols

(Equation 5). Because of these results and the more difficult alkoxide vs. alcohol procedure, further investigation of secondary alkoxides was not pursued.

71

The one constant in the substrates used to probe acyclic diastereoselection is that the functionality at the chiral center can easily occupy a pseudo-equatorial position of a chair-like transition state. To ensure that a substituent occupies the psudeo-axial position at the carbinol carbon of a chair-like transition state one must use a tertiary alcohol. Thus lithium 1-allyl-cyclohexanoxide 2DD1 was treated with TfOH-activated BMS conditions.

Surprisingly, 1,3-diol 2DD2 was recovered in 33% yield with 10:1 Markovnikov

Scheme 2DD: Effect of Psuedo-Axial Substituent at Carbinol Position OLi OH 1. BMS + TfOH OH CH2Cl2 2DD1 -78 to -20ºC 2DD2 2. MeOH, NaOOH 10:1 regioselectivity 33% yield Li Li

O H O B H B H H vs. OTf OTf 2DD3 3DD4

OLi OH H 1. BMS + TfOH H

H CH2Cl2 H OH 2DD5 -78 to -20ºC 2DD6 2. MeOH, NaOOH 3:1 regioselectivity 12% yiled Li Li

H O H H O B vs. B H H H H H OTf OTf 2DD7 2DD8

72 regioselectivity. This was unexpected because of the 1:19 anti-Markovnikov selectivity reported for terminal olefins under normal conditions. The selectivity is believed to be due to destabilization of the [3.1.1] bridged bicylic transition 2DD4 required for anti-

Markovnikov hydroboration by a steric interaction between bridging hydride and the pseudo-axial component of the cyclohexane ring. This conclusion is supported by the fact that lithium 3-buten-1-oxide 2DD5 provides only 3:1 Markovnikov regioselectivity

(Scheme 2DD). Tertiary alkoxides were not investigated further because the reaction of

Me2S·BH2OTf with 2DD1 led to a complex mixture of products. This is presumably due to the tertiary center participating in transformation involving easily displaced species upon interaction between alkoxide and excess activated borane

2.6 Ether-Directed Hydroboration

Ether-directed hydroboration was investigated in an effort to differentiate the oxygens in the final products. The methyl ether of 4-cyclohexyl-3-butanol 2EE1 was treated with standard preactivation conditions to provide 2EE2 in 20 : 1 regioselectivity.

In contrast to the alcohol substrates, an in situ method was better than the preactivation procedure, providing monoprotected 1,3-diol 2EE2 with improved selectivity of 50 : 1 in

60% yield (Scheme 2DD).

Scheme 2EE: Effect of Order of Addition on Homoallylic Ethers Preactivation Method: 1. BMS, TfOH OH OMe CH2Cl2, -78 to -20°C OMe OMe

2. NaOOH, MeOH OH 2EE1 2EE2 2EE3 20: 1 regioselectivity

In situ Method: 1. BMS, TfOH OH OMe CH2Cl2, -78 to -20°C OMe OMe

2. NaOOH, MeOH OH 2EE1 2EE2 2EE3 61% yield 50: 1 regioselectivity

73

This successful ether direction is mechanistically relevant (vide infra), but a methyl group is not a synthetically convenient protecting group. Collaboration with Dr.

Guoqiang Wang and Ms. Sarah Breed led to the discovery that an allyl ether can effectively direct hydroboration (Equation 6), providing a conveniently mono-protected

1,3-diol that can be manipulated without interference from the primary oxygen. In situ activation provided mono-allyl protected 1,3-diol with 27 : 1 regioselectivity in 55% yield while minimizing diol formation (<5%) and monohydroboration of the allyl group

(<5%) after oxidation. The poor reactivity of allylic substrates in the alcohol series

(Scheme 2P) sheds light on why the chemoselectivity is so good in this reaction.

Intramolecular hydroboration of the proximal (allylic) alkene is disfavored due to strained transition states and inductive deactivation of the olefin upon oxygen-borane interaction. Thus, the chemoselectivity supplements the regioselectivity of 27:1 as support for an intramolecular reaction pathway.

2.7 Mechanistic Insights

This work provides the first transition metal free method for a substrate-directed hydroboration using generic homoallylic alcohols, alkoxides, and ethers. However, the mechanistic picture is not as clear as was hoped. Scheme 2FF illustrates that all three categories investigated (alcohols; alkoxides; ethers) could follow simple and analogous mechanistic pathways via 2Y2 to provide regioselectivity. As previously discussed,

74 alcohol-borane complexes 2X3 evolve substoichiometric quantities of H2 gas, presumably to form an alkoxyborane 2Y1 or derived species, but this is a minor pathway below -20 °C. Alkoxide-borane complexes 2FF1 could release lithium triflate in a somewhat analogous fashion, providing the same alkoxyborane 2Y1. However, the methyl ether-borane complex 2FF2 cannot evolve methane or methyl triflate and still provide monoprotected diol upon oxidative workup. This demonstrates that the alkoxyborane 2Y1 is not an obligatory intermediate for the ODHB to occur, which supports the possibility that all three species 2X3, 2FF1, and 2FF2 follow the envisioned pathway to form olefin boron complexes 2Y3 via an associative mechanism akin to the

75 pathway proposed by Pasto for intermolecular hydroboration. However, diastereoselectivity has not been achieved on acyclic alcohol or alkoxide substrates, demonstrating that the reaction mechanism(s) is not as straightforward as envisioned.

The striking contrast with the all-carbon intramolecular hydroboration is not understood, although nearly all of our data were obtained using the unhindered reagents derived from

Me2S·BH2OTf. The steric bulk of PhB(H)OTf did not provide a result analogous to the thexylborane reactions previously reported with the all-carbon carbon substrates and no diol products were observed attempts to generate the more closely analogous activated reagent ThxB(H)OTf as described in the next section.

2.8 Testing Intramolecular Hydroboration of Thexylalkoxyboranes

Cha has reported thexylalkoxyboranes as viable intermolecular hydroboration reagents44 and Bryson has proposed a similar intermediate undergoing intramolecular hydroboration to account for unexpected diastereoselectivity (Scheme 2GG).45

Therefore, the hypothesis that thexylalkoxyboranes might undergo unactivated oxygen- directed intramolecular hydroboration was investigated in the context of homoallylic

76 alcohols. A 1:1 solution of 3-penten-1-ol and ThxBH2 was generated at -50 °C to prevent intermolecular hydroboration. Warming to 0 °C led to the observation of bubbling and a

11B NMR signal at = +51 ppm, which correlates to Cha‟s value for thexylethoxyborane

(= +50.4 ppm). This indicates that the desired intermediate was formed. The thexylalkoxyborane solution was stirred at rt for 12h and oxidized with NaOOH to provide 1,3-diol 2F2 with only 2:1 regioselectivity and 40% yield. This demonstrates that the unactivated thexylalkoxyborane approach to intramolecular hydroboration is not viable for regioselective hydroboration of acyclic homoallylic alcohols (Scheme 2HH).

Triflic acid activated thexylborane ThxB(H)OTf was also investigated with the notion that it could improve diastereoselectivity of ODHB. Treating ThxBH2 with triflic acid followed by 4-nonen-2-ol 2R5 at -78 °C did not produce diol upon warming to -20

°C and oxidation with NaOOH. Submitting the simpler substrate 3-octen-1-ol 2HH2 to identical conditions did not produce diols 2HH3 (Scheme 2HH). This ended investigation into ODHB with thexylborane.

Scheme 2HH: Investigating Oxygen-Directed Hydroboration with Thexylborane

unactivated:

H B 2 1. 0 °C to rt, 12h OH O BH OH 2. NaOOH THF, -50 to 0°C OH

2F1 2HH1 2F2 11B = +51 ppm 2 : 1 40% yield

activated: BH 1. 2 , TfOH OH OH OH CH2Cl2 R -78 to -20 °C R 2HH2 R= H 2. NaOOH 2HH3 R=H 2R5 R= Me 2R8 R=Me

77

2.9 Summary

Described above is the investigation of an alternative mechanistic proposal for

ODHB, which has been derived from the work of Pasto,20, 21 Schleyer,22 Beak,25 and

Ryschkewitsch.29 This borane activation approach has proven applicable to heteroatom- directed hydroboration in the context of homoallylic amines,27, 28 and it was hoped that it would provide some semblance of clarity to the topic of ODHB, discussed in the previous chapter. Homoallylic alcohols, alkoxides, and ethers have now been shown to undergo highly regioselective intramolecular hydroborations upon application of the borane activation approach using triflic acid to generate TfOBH2 or equivalent species.

However, a clear mechanistic picture remains elusive. Poor diastereoselectivity on chiral, branched acyclic substrates demonstrates that the mechanism(s) of these transformations is not straightforward, and argues against the simplest version of the mechanism that had been our working model. Mechanistic studies have not provided significant clarity, and have not resulted in a more convincing mechanistic proposal. Therefore, while generic

ODHB has now been achieved without the use of metal catalysts, the chemical community still awaits mechanistic understanding of oxygen-directed hydroboration.

78

Experimental

Substrates 5b and 5c are commercially available, while 5a,46 5d,47 5e,48 5f,47 5g,49 5h,50 and 5i51 have been reported in the literature. Alcohols 2J1d, 2J1e, and 2J1f were prepared using methods reported by Kocienski et al.,47 2J1g, 2J1h, and 2R5 were prepared by the method of Maryanoff et al.,49 and 2J1i was prepared using the method of

Negishi et al.52 The following diols have been reported previously: 2J2a,5 2J3a,53 2J2b,54

2J3b,54 2J2d,55 2J3d,56 2J2e,57 2J3e,58 2J2f,58 2J3f,59 2J2g,60 2J3g,53 2J2h,61 2J3h,57

2J3i,59 and 2DD2.62

Preparation of borane-thioanisole complex (BH3·SMePh)

32,63 The procedure combines features reported in prior work as follows: NaBH4 (4.56 g,

0.120 mol) was suspended in 60 mL of diglyme using a 250 mL rb flask fitted with an addition funnel (nitrogen atmosphere throughout). To a separate flask at 0 °C, connected to the first using a gas dispersion tube, was added thioanisole (5.08 g, 0.040 mol). Iodine

(15.1 g, 0.060 mol) in 60 mL of diglyme was then added dropwise to the NaBH4 suspension causing bubbling that indicated generation of diborane (B2H6). The gas was passed into neat thioanisole via a gas dispersion tube and any diborane that was not reacted was passed through a second outlet and quenched with acetone. Upon completion of the iodine addition, a gentle nitrogen flow was used to push any remaining diborane into the acetone-containing vessel. The BH3·SMePh was then capped and stored at -20 °C. The determination of molarity was performed by measuring out 0.50 mL of BH3·SMePh into a flask with 2.0 mL of CDCl3. The solution was cooled to 0 °C and an excess amount of dimethylbenzylamine (0.3-0.4 mL) was added dropwise. The

79 mixture was stirred for 30 min and analyzed by 1H NMR. The ratio between the downfield shifts of 4.0 ppm for the complexed amine and 3.4 ppm for uncomplexed amine was used to calculate the concentration of BH3·SMePh. Typically a range of 3-4

M was obtained.

ODHB of homoallylic alcohols; reaction of 2J1e with Me2S·BH3/TfOH.

CH2Cl2 (16 mL) in a 50 mL rb flask was cooled to -78 °C. Neat Me2S·BH3 (BMS; 300

L, 3.03 mmol) was added followed by TfOH (270 L, 3.04 mmol) dropwise. Each drop of TfOH initially froze on the surface, forming a white solid that dissipated after a few seconds of stirring. Gas evolution was observed. This solution was stirred for 35 min before dropwise addition of a solution of 2J1e (0.214 g, 1.39 mmol) in CH2Cl2 (11 mL).

The clear solution was stirrred at -20 °C for 10 h and was then treated slowly with a solution of 20% NaOH (4.8 mL) in MeOH (5.2 mL). The mixture was stirred 30 min at -

20 ºC before being stirred vigorously at 0 °C for slow dropwise addition of 35% H2O2

(2.4 mL ) in MeOH (2.6 mL). This mixture was warmed to rt and stirred for 10 h before transferring to a separatory funnel with 75 mL of Et2O. Brine (10 mL) was then added, and the aqueous layer was extracted with Et2O (3x, 75 mL). The combined organic layers were dried (MgSO4), concentrated (aspirator), and purified by flash chromatography (silica gel) using 30% EtOAc in hexanes to give 0.194 g of 2J2e (80%) as a 56:1 mixture with 2J3e according to the NMR assay described below. All other substrates 2J1 reported in Table 2 were reacted under analogous conditions.

80

Derivatization with 2-(trifluoromethyl)benzoyl chloride for NMR assay of regioselectivity.

A sample of 2J2e/2J3e ( 0.024 g, 0.136 mmol) was taken up in CH2Cl2 (1 mL) and cooled to 0 °C in a 10 mL rb flask. Neat 2-(trifluoromethyl)benzoyl chloride (0.1 mL,

0.68 mmol) was added dropwise to the magnetically stirred solution followed by DMAP

(0.085 g, 0.69 mmol) in CH2Cl2 (1 mL). The clear solution was treated with Et3N (0.08 mL, 0.57 mmol) and removed from the ice bath. The solution gradually became yellow- orange over 6h at rt. The reaction was loaded directly onto a preparatory TLC plate and was developed twice with 10% EtOAc/hexane. The UV active band with Rf of 0.25 was extracted with EtOAc. Concentration (aspirator) gave pure diaroylated product (0.069 g,

0.13 mmol) in 97% yield. Regioisomer ratios were established by comparing integrals for carefully phased, expanded 1H NMR spectra as follows:

2-(trifluoromethyl)benzoate from 2J2a/2J3a: The methyl triplet (0.90 ppm) from 2J3a ester was compared to the methyl doublet of 2J3a ester (1.2 ppm).

2-(trifluoromethyl)benzoate from 2J2/2J3 b,c,d,e,i: The 3-CH(O) methine proton was compared to the 4-CH(O) methine proton.

2-(Trifluoromethyl)benzoate from 2J2f/2J3f: The t-butyl singlet (0.90 ppm) from 2J3f ester was compared to the 13C satellite peaks of the t-butyl singlet (0.97 ppm by 1H

NMR) from 2J2f ester.

81

Table 2II NMR Data and Ratios from Controls Using Excess BH3·THF and ODHB precursor ester from alcohol 2J2 ester from alcohol 2J3 2J3 : 2J3 2J3 : 2J3 a alcohols  C(3)H (ppm)  C(4)H (ppm) (excess BH3) (ODHB) 2J2a/2J3a m, 5.19-5.33 m, 5.19-5.33 4.3 : 1b >20 : 1

2J2b/2J3b m, 5.23-5.31 m, 5.04-5.11 2 : 1 37 : 1

2J2c/2J3c m, 5.23-5.31 m, 5.04-5.11 2 : 1 28 : 1

2J2d/2J3d m, 5.29-5.35 m, 5.17-5.23 2 : 1 >20 :1

2J2e/2J3e m, 5.42-5.49 m, 5.05-5.11 2.7 : 1 56 : 1

2J2f/2J3f N/A N/A 2 : 1 82 : 1

2J2i/2J3i m, 5.34-5.41 m, 5.25-5.29 2 : 1 >20 : 1

(a) Control experiments: see procedure from 2J1i to 2J2i/2J3i, below, THF·BH3. (b) Me2S·BH3 in THF instead of THF·BH3

Control experiments; hydroboration of 2J1i with THF·BH3 and characterization of 2J2i.

To a 0 °C solution of 2J1i (0.059 g, 0.33 mmol) in THF (2 mL) was added excess

THF·BH3 (1 mL, 1 mmol). The solution was stirred for 4h before treating with premixed

20% NaOH (0.8 mL) and 35% H2O2 (0.4 mL) dropwise. The reaction mixture was transferred to a separatory funnel containing brine and extracted with Et2O. The organic layers were combined, dried over MgSO4, and concentrated (aspirator). According to

NMR assay after derivatization as described above, the crude mixture gave a 2:1 ratio of

2J2i:2J3i. Purification via silica gel chromatography in 70% EtOAc in hexanes eluted

+ + 1 pure 2J2i: HRMS-ES (m/z): [M + Na] calcd for C12H18O2, 217.120; found, 217.121. H

NMR (400 MHz, CDCl3, ): 1.47-1.59 (m, 2H). 1.62-1.82 (m, 4H), 2.33 (br s, 1H), 2.41

(br, s, 1H). 2.64 (t, J= 8 Hz, 2H), 3.77-3.84 (m, 1H), 3.84-3.91 (m, 2H), 7.16-7.20 (m,

13 3H), 7.25-7.30 (m, 2H) C NMR (100 MHz, CDCl3, ): 27.30, 35.80, 37.35, 38.28,

82

61.88, 72.162, 125.79, 128.32, 128.40, 142.24. IR (neat, cm-1): 3350 (br), 2950(s), 2860

(m). Regioisomer 2J3i was eluted in later fractions, and was identified by comparison with literature data.59

ODHB of 2J1b in the presence of cyclohexene.

CH2Cl2 (60 mL) in a 250 mL rb flask was cooled to -78 °C under nitrogen. Neat BMS

(1.1 mL, 11.11 mmol) was added followed by TfOH (1 mL, 11.26 mmol) dropwise.

Each drop of TfOH initially froze on the surface, forming a white solid that dissipated after a few seconds of stirring. Gas evolution was observed but no temperature change was detected by internal temperature monitoring. This stirred for 30 min before addition of a solution of 2J1b (0.68 mL, 5.55 mmol) and cyclohexene (3.0 mL, 29.6 mmol) in

CH2Cl2 (30 mL) over 45 min. The internal temperature rose 2-3 degrees during addition.

The clear solution was stirrred at -20 °C for 10 h and was then treated slowly with a solution of 20% NaOH (6.0 mL) in MeOH (6.0 mL). The mixture stirred 30 min at -20

ºC and was then stirred vigorously at 0 °C for slow dropwise addition of 35% H2O2 (3.0 mL ) in MeOH (3.0 mL). This mixture was warmed to rt, stirred for 20 h, and transferred to a separatory funnel using 75 mL of Et2O and 8 mL of brine. The aqueous layer was extracted with Et2O (4x, 75 mL) and the organic layers were combined, dried over

MgSO4, filtered, concentrated (aspirator), and purified via silica gel chromatography using 50% EtOAc in hexanes to give 0.456 g of 2J2b as a >20:1 mixture with 2J3b. The product was contaminated with 13% dimethylsulfone. A 62% yield of diols was calculated based on the NMR ratio of sulfone and diol signals.

83

Hydroboration of isopulegol 2Q1 with TfOH-activated BMS.

A -78 °C solution of BMS (490 L, 4.9 mmol) in CH2Cl2 (16 mL) was treated with

TfOH (450 L, 4.9 mmol) under N2. The mixture became a solution as gas evolution was observed. A solution of isopulegol 2Q17 (0.25 g, 1.6 mmol) was added over 30 min to the -78 ºC solution. The reaction was stirred at -20 ºC for 10 h before adding a mixture of 5 N NaOH (5.4 mL) and MeOH (6 mL), followed by 30% H2O2 diluted with H2O (3 mL) over 30 min. The reaction stirred for 12 h at rt before transferring to a separatory funnel containing satd K2CO3 (8 mL) and Et2O (100 mL). The aqueous layer was extracted with Et2O (4X 50 mL). The organic layers were combined, dried over Na2SO4, filtered, concentrated (aspirator), and purified via silica gel chromatography using 10%

EtOAc in hexanes (500 mL), 30% EtOAc in hexanes (600 mL), and 35% EtOAc in hexanes (400 mL) to give 0.051 g 2Q17 (20%), 0.028 g citronellol 2Q564 (11%), 0.0483 g

1,3-diol 2Q465 (17%), and 0.0923 g of 1,4-diols 2Q27 and 2Q37 (32%, >30 : 1 diastereoselectivity). The same conditions were used for entries 3 & 4 of Table 2CC, except 1.12 equiv of the appropriate amine was introduced as part of the substrate solution in CH2Cl2.

Monitoring the ODHB of 2J1b using 11B and 1H NMR spectroscopy.

CD2Cl2 (3 mL) in a 10 mL rb flask was cooled to -78 °C. Neat BMS (60 L, 0.61 mmol) was added followed by TfOH (55 L, 0.62 mmol). Each drop of TfOH initially froze on the surface, forming a white solid that dissipated after a few seconds of stirring (gas evolution). This stirred for 30 min and was then treated with a solution of 5b (0.030 g,

0.3 mmol) in CD2Cl2 (2 mL) dropwise. The solution was cannulated into an oven-dried,

11 1 N2-flushed, septum-capped NMR tube submersed in a -78 °C bath and B and H spectra

84 were taken at -78 °C. The sample was kept in a -78 °C bath while the probe was warmed to -20 °C and 1H spectra were then taken over 2.5 h – once every 5 min for the first 45 min followed by one every 15 min. The first 3 spectra contained a signal at 12.5 ppm that disappeared by the 20th min at -20 °C. Olefin signals remained, but were nearly gone after 2.5 h. Using the same sample, 11B spectra were taken at 2 min, 5 min, 15 min, 45 min, 2 h, and 2.5 h after warming to -20 °C. At 5 min signals appeared at -20.6 ppm

(residual BMS), -2.5 ppm (tentatively, TfOBH2), 7.5 ppm (broad) and -8 ppm (broad).

At 45 min all of the aforementioned signals were present but a new broad signal appeared at 34 ppm. At 2.5 h the signals at 34 and 7.5 ppm had broadened to the point of almost disappearing while the signal at -8 had sharpened slightly, appearing as a broadened triplet. The signal at -2.5 ppm dominated all 11B spectra and the signal and -20.6 diminished over time.

Attempt at non-catalyzed ODHB with thexyl-3-pentenoxyborane 2HH1

O B H

2HH1

A sample of 2,3-dimethyl-2-butene (0.25 mL, 2.1 mmol) was added to a 0 °C solution of

0.95 M THF·BH3 (2.1 mL, 2 mmol) under N2. The reaction stirred for 3 h while warming to rt. The solution was then cooled to -30 °C and 3-pentenol (0.26 mL, 2.1 mmol) was added. After 75 min at -30 °C 11B NMR spectroscopy revealed a dominant signal at = +51 ppm, which is assigned as 2HH1. The reaction was then stirred for 12 h at rt before adding premixed 20% NaOOH (1 mL) and 35% H2O2 (0.4 mL) and stirring for 7 h. The reaction mixture was transferred to a seperatory funnel containing brine (3

85 mL) and Et2O (40 mL). The aqueous layer was extracted Et2O (4X 40 mL). The organic layers were combined, dried over MgSO4, filtered, concentrated (aspirator), and purified via silica gel chromatography using 50% EtOAc in hexanes (300 mL), 60% EtOAc in hexanes (200 mL), and EtOAc (200 mL) to give 0.037 g of 3-pentenol (18%) and 0.1 g of a 2:1 mixture of 1,3- and 1,4-diols (40%).

Hydroboration of 2J1e using PhB(OTf)H 2U5

Lithium phenylborohydride 2U1 was prepared according to the report by Graham et al.66

A -78 °C suspension of 2U1 (0.10 g, 1 mmol) in CH2CL2 (4 mL) was treated with TfOH

(160 L, 1.8 mmol) dropwise under N2. A solution of 2J1e (0.07 g, 0.46 mmol) in

CH2Cl2 (2 mL) was added to the activated phenylborane solution, which was then warmed to -20 °C and stirred for 5 h. The reaction was treated with premixed 20%

NaOH (0.8 mL) and MeOH (1.5 mL) followed by 35% H2O2 (0.4 mL) in MeOH (1.5 mL). The reaction was stirred for 10 h, then transferred to a seperatory funnel containing brine (3 mL) and Et2O (40 mL). The aqueous layer was extracted Et2O (4X 40 mL). The organic layers were combined, dried over MgSO4, filtered, concentrated (aspirator), and purified via silica gel chromatography using 50% Et2O in hexanes (150 mL) and 70%

EtOAc in hexanes (300 mL), to give 0.016 g of diols 2J2e & 2J3e (20%). Derivatization and NMR assay as described above showed 22:1 regioselectivity. Hydroboration of 2S7 using 2U5 was done in analogous fashion.

ODHB of 2J1e via the lithium alkoxide 2AA1e.

BMS (60 L, 0.61 mmol) was taken up in 3 mL of CH2Cl2 placed in a 2-neck 25 mL rb flask fitted with a cold-jacketed addition funnel containing a solution of 2J1e (0.045 g,

0.29 mmol) in 2 mL of CH2Cl2. Both flask and funnel were cooled to -78 °C. TfOH (55

86

L, 0.62 mmol) was added dropwise to the rb flask. Each drop of TfOH initially froze on the surface, forming a white solid that dissipated after a few seconds of stirring (gas evolution observed). After stirring for 30 min, n-BuLi (2.11M, 150 L, 0.33 mmol) was added dropwise to the addition funnel, swirling the apparatus after addition was complete. The resulting alkoxide solution was then added dropwise into the activated borane solution at -78 °C, and the reaction was then warmed to -20 °C and stirred for 5 h before MeOH (1 mL at -20°C) was added dropwise. This stirred vigorously at 0 °C for

30 min followed by addition of premixed 20% NaOH (0.6 mL) and 35% H2O2 (0.3 mL) dropwise. This stirred for 12 h and was then transferred to a separatory funnel containing

1 mL of saturated K2CO3 solution using 20 mL Et2O. The aqueous layer was extracted with Et2O (4x, 20 mL). The organic layers were combined, dried over MgSO4, and concentrated (aspirator). The crude product was diaroylated as described above to give

0.097 g of the 2-(trifluoromethyl)benzoate of 2J2e for NMR assay (64% over 2 steps;

63:1 mixture with the regioisomer from 2J3e).

Preparation of E-4-cyclohexyl-1-methoxy-3-butene (2EE1).

OMe

2EE1

Sodium hydride (0.151 g, 3.77 mmol) was supended in THF (4 mL) and cooled to 0 °C in a 25 mL rb flask. A solution of 5i (0.269 g, 1.75 mmol) in THF (2 mL) was added dropwise and the reaction was allowed to warm to rt for 1 h. The suspension was cooled to 0 °C and neat MeI (0.23 mL, 3.71 mmol) was added dropwise. The reaction and warmed to rt for 2 h, quenched with H2O, and stirred for 15 min. The mixture was

87 transferred to a separatory funnel with 20 mL of Et2O and extracted with Et2O (3x, 20 mL). The organic layers were combined, dried over MgSO4, concentrated (aspirator), and purified via silica gel chromatography using 2% Et2O in hexanes to give 0.149 g of

+ + 1 13 (50%). HRMS-CI (m/z): [M + H] calcd for C11H20O, 169.159; found, 169.159. H

NMR (400 MHz, CDCl3, ): 0.99-1.31 (m, 5H), 1.57-1.74 (m, 5H), 1.85-1.96 (m, 1H),

2.26 (q, J= 6.8 Hz, 2H), 3.34 (s, 3H), 3.38 (t, J= 7.2 Hz, 2H) 5.23-5.49 (m, 2H). 13C

NMR (100 MHz, CDCl3, ): 26.04, 26.16, 33.00, 33.03, 40.65, 58.52, 72.77, 123.48,

138.57. The sample contained 12% Z-isomer (quartet at 2.34 ppm and singlet at 3.35 ppm by 1H NMR). IR (neat, cm-1): 2925(s), 2850 (m).

Control experiment hydroboration of 13; preparation of 4-cyclohexyl-3-hydroxy-1- methoxy-butane (2EE2) and 4-cyclohexyl-4-hydroxy-1-methoxy-butane (2EE3). OH OMe OMe

OH 2EE2 2EE3

To a 0 °C solution of 13 (0.040 g, 0.26 mmol) in THF (1 mL) was added excess

THF·BH3 (1 mL, 1 mmol). The solution stirred for 2h before treating with premixed

20% NaOH (0.8 mL) and 35% H2O2 (0.4 mL) dropwise. The reaction mixture was transferred to a separatory funnel containing brine and extracted with Et2O. The organic

1 layers were combined, dried over MgSO4, and concentrated (aspirator). Crude H NMR showed a 1.4:1 ratio of 2EE2:3EE3. Silica gel chromatography with 15% EtOAc in

+ + hexanes eluted 2EE2; HRMS-CI (m/z): [M + H] calcd for C11H22O2, 187.170; found,

1 187.169. H NMR (400 MHz, CDCl3, ): 0.80-1.00 (m, 2H), 1.08-1.31 (m, 4H), 1.37-

1.52 (m, 2H), 1.60-1.73 (m, 6H), 1.77-1.85 (m, 1H), 2.89 (br s, 1H), 3.35 (s, 3H), 3.51-

13 3.59 (m, 1H), 3.6-3.67 (m, 1H), 3.86-3.94 (br m, 1H). C NMR (100 MHz, CDCl3, ):

88

26.16, 26.31, 26.56, 32.91, 33.87, 34.06, 36.80, 45.28, 58.82, 68.60, 71.69. IR (neat, cm-

1): 3420 (br), 2920(s), 2850 (m). Pure 15 was eluted in later fractions; HRMS-CI+ (m/z):

+ 1 [M + H] calcd for C11H22O2, 187.170; found, 187.170. H NMR (400 MHz, CDCl3, ):

0.95-1.48 (m, 8H). 1.60-1.87 (m, 7H), 2.32 (br s, 1H), 3.35 (s, 4H, -OCH3 + -CH-O).

13 3.42 (dt, J= 1.2, 4.8 Hz, 2H). C NMR (100 MHz, CDCl3, ): 26.19, 26.33, 26.48, 26.53,

27.95, 29.2, 31.49, 43.70, 58.56, 73.08, 75.85. IR (neat, cm-1): 3420 (br), 2920(s), 2850

(m).

In Situ activation method for ODHB of 2EE1.

13 (0.10 g, 0.59 mmol) was taken up in CH2Cl2 (10 mL) and cooled to -78 °C in a 25 mL rb flask. Neat BMS (120 L, 1.21 mmol) was added dropwise and the resulting solution was stirred for 30 min before being treated slowly with neat TfOH (105 L, 1.18 mmol) dropwise. Each drop of TfOH acid initially froze on the surface, forming a white solid that dissipated after a few seconds of stirring. Gas evolution was observed. The clear solution was stirred at -20 °C for 10 h and treated slowly with a solution of 20% NaOH

(0.6 mL) in MeOH (1.0 mL). The mixture stirred 10 min before being stirred vigorously at 0 °C for slow dropwise addition of 35% H2O2 (0.3 mL ) in MeOH (0.7 mL). This mixture was warmed to rt and stirred for 10 h before transferring it to a separatory funnel containing 3 mL of brine using 30 mL of Et2O. The aqueous layer was extracted with

Et2O (3x, 25 mL) and the organic layers were combined, dried over MgSO4, concentrated

(aspirator), and purified via silica gel chromatography using 20% EtOAc in hexanes to give 0.067 g of 2EE2 (61%) as 50:1 mixture with 2EE3 along with 16% of recovered 13.

A trace (<1%) of demethylated alcohol (2J1e) was observed, but demethylated products

2J2e or 2J3e were not detected.

89

Synthesis of trans-2,2-dimethyl-5-dodecen-3-ol 2R6

OH 2R6

2R6 was synthesized using the method of Negishi et al.49 A 0 °C solution of 1-

(2.36 mL, 18 mmol) in hexanes (40 mL) was treated with a solution of DIBAL-H (3.2 mL, 18 mmol) in hexanes (16 mL) under N2. The transfer of DIBAL-H was completed with hexanes (2 mL). The reaction was stirred for 30 min at rt before warming to 55 °C for 4 h. The reaction was then cooled to rt and 1.42 M nBuLi (12.67 mL, 18 mmol) was added dropwise. The reaction became a white sludge while stirring for 20 min before a solution of tbutyl-oxirane (2.4 mL, 19.7 mmol) in hexanes (20 mL) was added via cannulation under N2. The reaction was stirred for 24 h at rt before cooling to 0° C and adding 3N HCl (10 mL), which stirred for 1 hwhile warming to rt. Satd Rochelle‟s salt

(30 mL) was added to the reaction, which was then transferred to a separatory funnel.

The aqueous layer was removed and rinsed with hexanes (2X, 80 mL). The organic layer were combined, dried over MgSO4, filtered, and concentrated (aspirator), and purified via silica gel chromatography in 15% EtOAc in hexanes followed by kugelrohor

1 distillation to provide 2R6. Yield was not recorded. ( H NMR (500 MHz, CDCl3, ):

0.88 (t, J= 7.5 Hz, 3H). 0.91 (s, 9H), 1.22-1.40 (m, 6H), 1.65 (d, J= 3 Hz, 1H). 1.85-

1.94 (m, 1H), 2.02 (q, J= 7.5 Hz ,2H), 2.27-2.34 (m, 1H), 3.19 (dt, J= 11 Hz, 2.5 Hz,

2H), 5.39-5.47 (m, 1H), 5.52-5.59 (m, 1H).

90

Synthesis of cis-2,2,7,7-tetramethyl-5-octen-3-ol 2S6 and cis-2,2-dimethyl-5-dodecen- 3-ol 2S7.

OH OH 2S6 2S7

To a -78 °C solution of 3,3-dimethyl-1- (3.7 mL, 30 mmol) in THF (30 mL) was added 1.38 M nBuLi (21 mL, 29 mmol), under N2. The reaction stirred for 20 min at -78

°C before neat tbutyl-oxirane (4 mL, 33 mmol) was added dropwise, followed by

BF3·OEt2 (3 mL, 24.3 mmol). The reaction stirred for 2.5 h at -78 °C before 15 mL of satd NaHCO3 was added. The reaction warmed to 0 °C over 10 h. The THF layer was separated from the aqueous layer, which was then rinsed with Et2O (2X 40 mL). The organic layers were combined, dried over MgSO4, filtered, and concentrated to provide crude 2,2,7,7-tetramethyl-5-octyn-3-ol. To a 0 °C solution of cyclohexene (9.8 mL, 97 mmol) in Et2O (30 mL) was added 10 M BMS (4.6 mL, 46 mmol) dropwise under N2.

The reaction stirred at 0 °C for 3 h before the solvent was blown down with N2 through the septum followed by placing the white powder under high vac.67 The white powder was suspended in THF (20 mL) and cannulated into a 0 °C solution of 2,2,7,7- tetramethyl-5-octyn-3-ol (19.56 mmol) in THF (10 mL). The transfer was completed with THF (2X 10 mL) and the reaction stirred for 14h at rt. The reaction was cooled to 0

°C and AcOH (15 mL, 250 mmol) was added. The reaction was heated to 60 °C for 90 min. Solvent was removed via rotovap and the residue was treated with 1:1 CH2Cl2: 4M

KOH (60 mL) and stirred for 40 h. The mixture was transferred to a separatory funnel and the organic layer was removed from the aqueous layer, which was rinsed with

91

CH2Cl2 (3X 30 mL). The organic layer were combined and rinsed with satd NH4Cl (3X

25 mL) and Brine (40 mL), dried over MgSO4, filtered, concentrated (aspirator), and purified via silica gel chromatography using 2% Et2O in hexanes to give 1.5 g of cis-

1 2,2,7,7-tetramethyl-5-octen-3-ol 2S6 (42%): ( H NMR (400 MHz, CDCl3, ): 0.93 (s,

9H). 1.12 (s, 9H), 1.27 (br s, 1H), 2.32 (m, 2H). 3.23 (dt, J= 9.2 Hz, 4 Hz, 1H), 5.26 (m,

1H), 5.54 (dt, J= 12 Hz, 2 Hz, 1H). Analogous conditions were used with 1-heptyne or

68 1 methyloxirane to generate 2S5 and 2S7. ( H NMR (400 MHz, CDCl3, ): 0.89 (t, J=

6.8 Hz, 3H). 0.93 (s, 9H), 1.22-1.40 (m, 6H), 2.0-2.26 (m, 4H). 3.09 (dt, J= 10 Hz, 3.2

Hz, 1H), 5.4-5.49 (m, 1H), 5.54-5.63 (dt, J= 12 Hz, 2 Hz, 1H).

Synthesis of trans-2-isopropyl-3-pentenol 2T1

OH

2T1

To a -78 °C solution of 3-pentenoic acid (0.2 mL, 1.93 mmol) in HMPA (2 mL) and THF

(8 mL) was added 1.28 M (3.92 mL, 5 mmol) under N2. The reaction stirred for 30 min before addition of 2-iodopropane (0.24 mL, 2.4 mmol) in THF (2 mL). The reaction stirred for 3 h, warming to 0 °C. 1 N HCl (4 mL) was added to quench the reaction, which was transferred to a separatory funnel containing Et2O (40 mL). The aqueous layer was extracted with Et2O (4X 40 mL). The organic layers were combined, dried over MgSO4, filtered, and concentrated. Reduction of the crude mixture with LAH (0.31 g, 11.2 mmol) followed by silica gel chromatography in 10% EtOAc in hexanes provided

0.072 g of synthesis of trans-2-isopropyl-3-pentenol 2T1 (29%). 1H NMR (500 MHz,

92

CDCl3, ): 0.86 (d, J= 7 Hz, 3H). 0.89 (d, J= 7 Hz, 3H), 1.52 (br s, 1H), 1.62 (m, 1H).

1.72 (dd, J= 6.5 Hz, 1.5 Hz, 3H), 1.95 (m, 1H), 3.38 (t, J= 10 Hz, 1H), 3.63 (dd, J= 10.5

Hz, 5 Hz, 1H), 5.22 (m, 1H), 5.56 (dq, J= 15 Hz, 6.5 Hz, 1H). 13C NMR (125 MHz,

CDCl3, ): 18.13, 19.58, 20.77, 28.89, 52.53, 64.09, 129.25, 130.53.

ODHB with 2T1

To a -78 °C solution of 10.1 M BMS (60 L, 0.61 mmol) in CH2Cl2 (3 mL) was added

TfOH (50 L, 0.55 mmol) dropwise under N2. This stirred for 30 min at -78 °C before adding a solution of 2T1 (0.141 g, 0.32 mmol) and cyclohexene (0.16 mL, 1.58 mmol) in

CH2Cl2 (2 mL). The reaction mixture was then stirred for 10 h at -20 °C before adding

20% NaOH (0.4 mL) in MeOH (0.6 mL) followed by 35% H2O2 (0.2 mL). The reaction mixture stirred for 12 h at rt before transferring it to a separatory funnel containing brine

(2 mL) and Et2O (25 mL). The aqueous layer was extracted with Et2O (4X, 25 mL). The organic layers were combined, dried over Na2SO4, filtered, concentrated (aspirator), and purified via silica gel chromatography in 20% EtOAc in hexanes to collect 0.016 g of one

1,3-diol diastereomer (35%) and 0.015 g of the other (32%). No 1,4-diol regioisomer

1 was observed. Top 1,3-diol diastereomer: H NMR (400 MHz, CDCl3, ): 0.92 (d, J= 7

Hz, 3H). 0.96 (t, J= 7.6 Hz, 3H), 1.01 (d, J= 7 Hz, 3H), 1.18 (m, 1H). 1.56-1.65 (m, 2H),

1.97 (oct, J= 6.4 Hz, 1H), 2.66 (br s, 1H), 2.91 (br s, 1H), 3.81 (m, 1H), 3.88-3.98 (m,

1 1H). Bottom 1,3-diol diastereomer: H NMR (400 MHz, CDCl3, ): 0.95 (d, J= 6.4 Hz,

6H). 1.03 (t, J= 7.2 Hz, 3H), 1.5-1.68 (m, 4H), 2.74 (br s, 2H). 13.74-3.89 (m, 3H).

93

References for Chapter 2

1. Hoveyda, A. H.; Evans, D. A.; Fu, G. C. Substrate-Directable Chemical Reactions. Chem. Rev. 1993, 93 (4), 1307-1370.

2. Evans, D. A.; Fu, G. C. Amide-directed Iridium-Catalyzed Hydroboration of Olefins- Documentation of Regiochemical and Stereochemical Control in Cyclic and Acyclic Systems. J. Am. Chem. Soc. 1991, 113, 4042.

3. Evans, D. A.; Fu, G. C.; Hoveyda, A. H. Rhodium(I)-Catalyzed and Iridium(I)- Catalyzed Hydroboration Reactions - Scope and Synthetic Applications. J. Am. Chem. Soc. 1992, 114, 6671.

4. Evans, D. A.; Fu, G. C.; Hoveyda, A. H. Rhodium(I)-Catalyzed Hydroboration of Olefins- The Documentation of Regiochemical and Stereochemical Control in Cyclic and Acyclic Systems. J. Am. Chem. Soc. 1988, 110, 6917.

5. Evans, D. A.; Muci, A. R.; Sturmer, R. Samarium(III)-Catalyzed Hydroboration of Olefins with Catecholborane- A General Approach to the Synthesis of Boronate Esters. J. Org. Chem. 1993, 58, 5307.

6. Panek, J. S.; Xu, F.; Rondon, A. C. Chiral crotylsilane-based approach to benzoquinoid ansamycins: Total synthesis of (+)-macbecin I. J. Am. Chem. Soc. 1998, 120, 4113.

7. Schulte-Elte, K. H.; Ohloff, G. Uber Eine Aussergewohnliche Stereospezifitat Bei der Diastereomeren (1R)-Isopulegole mit Diboran. Helv. Chim. Acta 1967, 50, 153.

8. Heathcock, C. H.; Jarvi, E. T.; Rosen, T. Acyclic Stereoselection. 21. Synthesis of an Ionophore Synthon Having 4 Asymmetric Carbons by Sequential Aldol Addition, Claisen Rearrangement and Hydoboration. Tetrahedron Lett. 1984, 25, 243.

9. Smith, A. B.; Yokoyama, Y.; Huryn, D. M.; Dunlap, N. K. Total Synthesis of (+)- Mikrolin. Tetrahedron Lett. 1987, 28, 3659.

10. Zweifel, G.; Najafi, M. R.; Rajagopalan, S. Hydroboration of Methoxyenynes - A Novel Synthesis of (E)-Methoxyenones. Tetrahedron Lett. 1988, 29, 1895.

11. Suzuki, K.; Miyazawa, M.; Shimazaki, M.; Tsuchihashi, G. Silicon-Directed Stereoselective Synthesis of 2-Vinyl-1,3-diols - Stereo-divergence with aand without the Silyl Group Related to the Synthesis of Avenaciolide and Isoavenaciolide. Tetrahedron 1988, 44, 4061.

94

12. Jung, M. E.; Karama, U. Highly diastereoselective Markovnikov hydration of 3,4- dialkoxy-1-alkenes and 4,5-dialkoxy-2-alkenes via a hydroboration-oxidation process. Tetrahedron Lett. 1999, 40, 7907.

13. Panek, J. S.; Xu, F. Diastereoselectivity in the Borane Methyl Sulfide Promoted Hydroboration of -Alkoxy--unsaturated Esters - Documentation of an Alkoxy- Directed Hydroboration Reaction. J. Org. Chem. 1992, 57, 5288.

14. Brown, H. C.; Chandrasekharan, J. Mechanism of Hydroboration of Alkenes with Borane-Lewis Base Complexes- Evidence That the Mechanism of the Hydroboration Reaction Proceeds Through a Prior Dissociation of Such Complexes. J. Am. Chem. Soc. 1984, 106, 1863-1865.

15. Brown, H. C.; Moerikofer, A. W. Hydroboration .10. Rates of Reaction of Bis-3- methyl-2-butylborane with Representative Cycloalkenes and Isomeric cis-trans Alkenes. J. Am. Chem. Soc. 1961, 83, 3417.

16. Brown, H. C.; Yoon, N. M. Monoisopinocampheylborane - New Chiral Hydroborating Agent for Relatively Hindered (Trisubstituted) Olefins. J. Am. Chem. Soc. 1977, 99, 5514.

17. Chandrasekharan, J.; Brown, H. C. Hydroboration Kinetics.11. A Reinvestigation of the Kinetics of Hydroboration of Representative Alkenes with Disiamylborane Dimer - Conclusive Evidence for the Dissociative Mechanism in the Hydroboration of Alkenes with Dialkylborane Dimers J. Org. Chem. 1985, 50, 518.

18. Wang, K. K.; Brown, H. C. Hydroboration Kinetics. 2. Improved Procedure for Following the Kinetics for the Reaction of Alkenes with 9- Borabicyclo[3.3.1]nonane - Further Evidence for the Dissociation Mechanism. J. Org. Chem. 1980, 45, 5303.

19. Wang, K. K.; Brown, H. C. Hydroboration Kinetics. 6. Hydroboration of Alkenes with 9-borabicyclo[3.3.1]nonane Dimer and 9-borabicyclo[3.3.1]nonane Lewis Base Complexes in Various Solvents- an Interpretation of the Catalytic Effect of Ether Solvents on the Hydroboration Reaction. J. Am. Chem. Soc. 1982, 104, 7148.

20. Pasto, D. J.; Cheng, T. C.; Lepeska, B. Measurement of Kinetics and Activation Parameters for Hydroboration of Tetramethylethylene and Measurement of Isotope-Effects in Hydroboration of Alkenes. J. Am. Chem. Soc. 1972, 94, 6083.

21. Pasto, D. J.; Lepeska, B.; Balasubr.V Measurement of Relative Rate Ratios of First and Second Steps of Hydroboration Reaction and Rates of Alkylborane Redistribution Reactions. Discussion of the Overall Mechanism of the Hydroboration Reaction. J. Am. Chem. Soc. 1972, 94, 6090.

95

22. Clark, T.; Wilhelm, D.; Schleyer, P. V. Mechanism of Hydroboration in Ether Solvents. A Model ab initio Study. J. Chem. Soc. Chem. Commun. 1983, (11), 606-608.

23. DiMare, M. Ab Initio Computational Examination of Carbonyl Reductions by Borane: The Importance of Lewis Acid-Base Interactions. J. Org. Chem. 1996, 61, 8378.

24. Van Eikema Hommes, N. J. R.; Schleyer, P. V. 3-Center Transition Structures for Alkene Hydroboration and Alkylborane Rearrangement. J. Org. Chem. 1991, 56, 4074.

25. Beak, P. Determinations of Transition-State Geometries by the Endocyclic Restriction Test: Mechanisms of Substitution at Nonstereogenic Atoms. Acc. Chem. Res. 1992, 25, 215.

26. Baldwin, J. E. Rules for Ring-Closure. J. Chem. Soc., Chem. Commun. 1976, (18), 734-736.

27. Scheideman, M.; Shapland, P.; Vedejs, E. A Mechanistic Alternative for the Intramolecular Hydroboration of Homoallylic Amine and Phosphine Borane Complexes. J. Am. Chem. Soc. 2003, 125, 10502.

28. Scheideman, M.; Wang, G. Q.; Vedejs, E. Amine-directed hydroboration: Scope and limitations. J. Am. Chem. Soc. 2008, 130, 8669.

+ 29. Nainan, K. C. R., G. E. Cations Derived from BH2 . Unsymmetrical Bis-Amine Substitution on Boron. J. Am. Chem. Soc. 1969, 91, 330.

30. Brown, H. C.; Zweifel, G. Hydroboration. VII. Directive Effects in the Hydroboration of Olefins. J. Am. Chem. Soc. 1960, 82, 4708.

31. Kanth, J. V. B.; Brown, H. C. Unusual rapid hydroboration of alkenes using diborane in chlorohydrocarbon solvents. Tetrahedron Lett. 2000, 41, 9361.

32. Zaidlewicz, M.; Kanth, J. V. B.; Brown, H. C. Molecular Addition Compounds. 17. Borane and Chloroborane Adducts with Organic Sulfides for Hydroboration. J. Org. Chem. 2000, 65, 6697.

33. Rarig, R.-A. F.; Scheideman, M.; Vedejs, E. Oxygen-Directed Intramolecular Hydroboration. J. Am. Chem. Soc. 2008, 130, 9182.

34. Brown, H. C.; Negishi, E. I. Hydroboration. 32. Cyclic Hydroboration of Dienes with Thexylborane. J. Am. Chem. Soc. 1972, 94, 3567.

96

35. Kropp, P. J.; Breton, G. W.; Craig, S. L.; Crawford, S. D.; Durland, W. F.; Jones, J. E.; Raleigh, J. S. Surface-mediated Reactions. 6. Efects of Silica-Gel and Alumina on Acid-Catalyzed Reactions. J. Org. Chem. 1995, 60, 4146.

36. Cole, T. E.; Bakshi, R. K.; Srebnik, M.; Singaram, B.; Brown, H. C. Organoboranes. 49. An Examination of Convenient Procedures for the Generation of Borane and Monoalkylboranes and Dialkylboranes from Lithium Borohydride and Monoalkylborohydrides and Dialkylborohydrides. Organometallics 1986, 5, 2303.

37. Harada, T.; Matsuda, Y.; Uchimura, J.; Oku, A. Highly Stereoselective Synthesis pf 1,3-Diols Utilizing Intramolecular Hydroboration of Allyl Vinyl Ethers. J. Chem. Soc. Chem. Commun. 1989, (19), 1429.

38. Harada, T.; Matsuda, Y.; Wada, I.; Uchimura, J.; Oku, A. Stereochemical Control of Consecutive Stereogenic Centers by Intramolecular Hydroboration of Dialkenyl Carbinol Derivatives. J. Chem. Soc. Chem. Commun. 1990, (1), 21-22.

39. Still, W. C.; Darst, K. P. Remote Asymmetric Induction. A stereoselective Approach to Acyclic Diols via Cyclic Hydroboration. J. Am. Chem. Soc. 1980, 102, 7385.

40. Still, W. C.; Shaw, K. R. Acyclic Stereoselection va Cyclic Hydroboration. Synthesis of the Prelog-Djerassi Lactonic Acid. Tetrahedron Lett. 1981, 22, 3725.

41. Yokoyama, Y.; Kawashima, H.; Kohno, M.; Ogawa, Y.; Uchida, S. Stereospecific Construction of 3 Contiguous Asymmetric Centers via Cyclic Hydroboration. Tetrahedron Lett. 1991, 32, 1479.

42. Yokoyama, Y.; Kawashima, H.; Masaki, H. A(1,3) Strain-Controlled Cyclic Hydroboration of 1,4- and 1,-5 Dienes. Chem. Lett. 1989, (3), 453.

43. Brown, H. C.; Ravindran, N. Hydroboration. 40. Hydroboration of Alkenes and Alkynes with Monochloroborane Etherates. Convenient Procedures for Preparation of Dialkylchloroboranes, Monoalkylchloroboranes and Dialkenylchloroboranes and Their Derivatives J. Am. Chem. Soc. 1976, 98, 1785.

44. Cha, J. S.; Seo, W. W.; Kim, J. M.; Kwon, O. O. Thexylalkoxyborane as Hydroborating Agent for Alkenes and Alkynes. Bull. Kor. Chem. Soc. 1996, 17, 892.

45. Welch, M. C.; Bryson, T. A. Boron Annulation in Organic Synthesis. 3. Stereoselectivity and the Formal Synthesis of (+/-) Helenalin. Tetrahedron Lett. 1989, 30, 523.

97

46. Crombie, L.; Rainbow, L. J. Stereoselective Synthesis of Alcohols Containing (Z)- Olefins and (E)-Olefins, Dienes, Enynes, and Styrenes. Cyclic -Halogeno Ether Scissions Using Samarium Diiodide as the Electron-Transfer Agent. J. Chem. Soc.-Perkin Trans 1 1994, (6), 673.

47. Kocienski, P.; Wadman, S.; Cooper, K. A Stereoselective Synthesis of Functionalized Alkenyllithiums and Alkenyl Cyanocuprates by the Cu(I)- Catalyzed Coupling of Organo-Lithium Reagents with a-Lithiated Cyclic Enol Ethers. J. Am. Chem. Soc. 1989, 111, 2363.

48. Ryu, I. H.; Hirai, A.; Suzuki, H.; Sonoda, N.; Murai, S. One-Pot Conversions of (Silylmethyl)cyclopropanes to Homoallylic Alcohols and 1,4-Diols Based on Haloborane-Induced Ring-Opening. J. Org. Chem. 1990, 55, 1409.

49. Maryanoff, B. E.; Reitz, A. B.; Duhlemswiler, B. A. Stereochemistry of the . Effect of Nucleophillic Groups in the Phosphonium Ylide. J. Am. Chem. Soc. 1985, 107, 217.

50. Doyle, M. P.; Austin, R. E.; Bailey, A. S.; Dwyer, M. P.; Dyatkin, A. B.; Kalinin, A. V.; Kwan, M. M. Y.; Liras, S.; Oalmann, C. J.; Pieters, R. J.; Protopopova, M. N.; Raab, C. E.; Roos, G. H. P.; Zhou, Q. L.; Martin, S. F. Enantioselective Intramolecular of Allylic and Homoallylic Diazoacetates and Diazoacetates and Diazoacetamides Using Chiral Dirhodium(II) Carboxamide Catalysts. J. Am. Chem. Soc. 1995, 117, 5763.

51. Perlman, N.; Livneh, M.; Albeck, A. Epoxidation of Peptidyl Olefin Isosteres. Stereochemical Induction Effect of Chiral Centers at Four Adjacent C- Positions. Tetrahedron 2000, 56, 1505.

52. Negishi, E. I.; Baba, S.; King, A. O. Stereoselective Synthesis of b-Hydroxy Substituted Trans-Alkenes by Reaction of Trans-Alkenyltrialkylaluminates with . J. Chem. Soc.-Chem. Commun. 1976, (1), 17-18.

53. Nishiyama, H.; Kitajima, T.; Matsumoto, M.; Itoh, K. Silylmethyl Radical Cyclization. New Stereoselective Method for 1,3-Diol Synthesis form Allylic Alcohols. J. Org. Chem. 1984, 49, 2298.

54. Tanner, D.; Groth, T. Regioselective Nucleophilic Ring Opening of Epoxides and Aziridines Derived from Homoallylic Alcohols. Tetrahedron 1997, 53, 16139.

55. Cohen, T.; Jeong, I. H.; Mudryk, B.; Bhupathy, M.; Awad, M. M. A. Synthetically Useful b-Lithioalkoxides from Reductive Lithiation of Epoxides by Aromatic Radical-Anions. J. Org. Chem. 1990, 55, 1528.

56. Ito, M.; Osaku, A.; Shiibashi, A.; Ikariya, T. An Efficient Oxidative Lactonization of 1,4-Diols Catalyzed by Cp*Ru(PN) Complexes. Org. Lett. 2007, 9, 1821.

98

57. Batey, R. A.; Smil, D. V. The First Boron-tethered Radical Cyclizations and Intramolecular Homolytic Substitutions at Boron. Angew.Chem. Int. Ed. 1999, 38, 1798.

58. Harada, T.; Kurokawa, H.; Kagamihara, Y.; Tanaka, S.; Inoue, A.; Oku, A. Stereoselective Acetaliaztion of 1,3-Alkanediols by 1-Menthone. Application to the Resolution of Racemic 1,3-Alkanediols and to the Determination of the Determination of the Absolute Configuration of Enantiomeric 1,3-Alkanediols J. Org. Chem. 1992, 57, 1412.

59. Foubelo, F.; Saleh, S. A.; Yus, M. 3-Chloropropyl and 4-Chlorobutyl Phenyl Ethers as Sources of 1,3-Dilithiopropane and 1,4-Dilithiobutane: Sequential Reaction with Carbonyl Compounds. J. Org. Chem. 2000, 65, 3478.

60. Klos, A. M.; Heintzelman, G. R.; Weinreb, S. M. A -Hydroxyethyl Carbanion Equivalent. J. Org. Chem. 1997, 62, 3758.

61. Fujioka, H.; Ohba, Y.; Hirose, H.; Murai, K.; Kita, Y. Mild and Efficient Removal of Hydroxyethyl Unit from 2-Hydroxyethyl Ether Derivatives Leading to Alcohols. Org. Lett. 2005, 7, 3303.

62. Moon, S.; Waxman, B. H. Lead Tetraacetate. 6. Stereochemical Studies on Formation of Bicyclic Ethers from Alicyclic Primary Alcohols. J. Org. Chem. 1969, 34, 288.

63. Narayana, C.; Periasamy, M. A Simple Convenient Method for the Generation of Diborane from NaBH4 and I2. J. Organomet. Chem. 1987, 323, 145.

64. Kropp, P. J.; Breton, G. W.; Craig, S. L.; Crawford, S. D.; Durland, W. F.; Jones, J. E.; Raleigh, J. S. Surface-Mediated Reactions. 6. Effects of Silica Gell and Alumina on Acid-Catalyzed Reactions. J. Org. Chem. 1995, 60 (13), 4146.

65. Hong, B. C.; Wu, M. F.; Tseng, H. C.; Liao, J. H. Enantioselective Organocatalytic formal [3+3]-Cycloaddition of ,-Unsaturated Aldehydes and Application to the Asymmetric Synthesis of (-)-Isopulegol Hydrate and (-)- Cubebaol. Org. Lett. 2006, 8 (11), 2217.

66. Graham, L. A.; Fout, A. R.; Kuehne, K. R.; White, J. L.; Mookherji, B.; Marks, F. M.; Yap, G. P. A.; Zakharov, L. N.; Rheingold, A. L.; Rabinovich, D. Manganese(I) Poly(mercaptoimidazolyl)borate Complexes: Spectroscopic and Structural Characterization of Mn···H-B Interactions in Solution and in the Solid. Dalton Transactions 2005, 171.

99

67. Abiko, A. Dicyclohexylboron trifluoromethanesulfonate. Org. Synth. 2003, 1, 103-108.

68. Blum, A.; Hess, W.; Studer, A. Stereocontrolled Formation of Vinylsilanes via Homolytic Substitution at Silicon. Synthesis 2004, 2226.

100

Chapter 3: The Impact of Ipc2BH on Synthetic Organic Chemistry

The creation of chiral centers in asymmetric fashion is a cornerstone of organic synthesis and a frontier whose limits will be pushed for generations to come. Almost fifty years ago, hydroboration earned the distinction of being one of the first synthetically practical, non-enzymatic methods for generating an asymmetric center. This was achieved by H. C. Brown through the generation of a C2-symmetric boron environment in diisopinocampheylborane (Ipc2BH). While Brown immediately recognized the ground-breaking nature and potential of the asymmetric induction, the C2-symmetric nature of the reagent was initially unheralded. The following chapter will illustrate the impact of Brown‟s discovery over the past half century in the context of 1) the evolution of asymmetric hydroboration and 2) the development of C2-symmetric boron species as enantioselective reagents for organic synthesis.

3.1: The Evolution of Asymmetric Hydroboration

3.1i: Diisopinocampheylborane

In 1961, just five years after hydroboration in ethereal solvents was first reported,1

Brown and Zweifel discovered that hydroboration of the naturally abundant chiral terpene -pinene 3A1 stops at the dialkylborane stage to asymmetrically form the

C2-symmetric dialkylborane diisopinocampheylborane (Ipc2BH) (3A2). This is attributed to the steric bulk of both 3A1 and 3A2 preventing further reaction.

Fortunately, further reaction of 3A2 occurs with less hindered substrates. Olefins such as cis-2-butene, cis-3-, and norbornene react with 3A2 to provide (R)-2-butanol (R)-

3A3, (R)-3-hexanol (R)-3A4, and (1R,2S)-exo-norborneol 3A5 with respective enantiomeric purities of 87%, 91%, and 83% after oxidative workup (Scheme 3A).2

101

Scheme 3A: The First Reported Asymmetric Hydroborations with Ipc2BH H NaBH , BF ·OEt 1. cis-2-butene 4 3 2 B OH Diglyme 2. NaOOH 3A1 3A2 3A3 1. cis-3-hexene 90 % yield 87% ee 1. 2. NaOOH

2. NaOOH

OH OH

3A5 H 3A4 62% yield 81% yield 83% ee 91% ee Obtaining (R)-3A4 with an enantiomeric purity of 91% is of particular importance: since

-pinene was available in 90% purity, this indicates that the formation of 3A2 occurs with complete enantioselectivity.

High enantioselectivity is elusive with a broader range of olefins. While cis-1,2-disubstituted (Type II) olefins are excellent substrates for asymmetric hydroboration with Ipc2BH (3A2), the more hindered trans-1,2-disubstituted (Type III) and trisubstituted (Type IV) olefins lead to lower enantioselectivities and loss of pinene from 3A2 via retrohydroboration during the reaction. In the case of trans-2-butene two equivalents of substrate are consumed for every equivalent of pinene released in the reaction, while only one equivalent of the more hindered 1-methyl-cyclopentene is consumed for each equivalent of pinene released. Brown reports that the loss of pinene is the result of the equilibrium between the Ipc2BH dimer 3B1 and triisopinocampheyldiborane 3B2 (Ipc3B2H3), which then reacts with the substrate at the

102 less substituted boron. 3 In the case of 3-methyl-cyclopentene, the initial hydroboration must lead to a species (presumably 3B5) with sufficient steric bulk to prevent a second hydroboration. Conversely, the less hindered trans-2-butene can react a second time, accounting for the different ratios of consumed substrate : recovered pinene (Scheme

3B). Also worthy of note is the fact that the low enantiomeric purities observed in these cases were opposite to those predicted by Brown‟s model for Ipc2BH.

Enantiomeric purities of alcohols derived from the hydroboration of 1,1- disubstituted (Type I) olefins with Ipc2BH (A2) are also low. In the case of 2-methyl-1- butene, Ipc2BH generates (R)-2-methyl-1-butanol (R)-C1 with 21% ee (Scheme 3C).

Brown noted that reduced selectivity is to be expected in cases where the boron does not become directly attached to the stereogenic carbon,4 and differentiating a methyl group

103 from an ethyl group presents a challenge for any asymmetric transformation. Increasing the steric differentiation of the olefin substituents using 2,3-dimethyl-1-butene leads to

(R)-2,3-dimethyl-1-butanol (R)-3C2 with 30% ee after oxidative workup (Scheme 3C).

Partial loss of pinene (9 % and 7 %, respectively) is observed in these reactions, suggesting, though not explicitly concluded by Brown, that Ipc3B2H3 3B2 is contributing to the results.

Almost twenty years after the discovery of 3A2, Brown published an improved synthesis of the reagent, making it readily available in 99% enantiomeric purity from

92% enantio-enriched pinene.5 This enables greater selectivity with type II olefins in tetrahydrofuran (THF) at -25 °C. The increased purity of the reagent and lower reaction temperatures are necessitated by the switching of solvent from diglyme to THF leading to increased formation of monoisopinocampheylborane 3F2 (IpcBH2), which interferes with selectivity at room temperature. More hindered substrates require higher reaction temperatures, which lead to the aforementioned loss of pinene, rendering the improved

3A2 less effective. A summary of the selectivities achieved with 3A2 is provided in

Table 3D.

104

Table 3D: Asymmetric Hydroboration with Diisopinocampheylborane

Type of Olefin Substrate ee of Alcohol (%) II 2-butene 98 (R)a II 3-hexene 93 (R)a II norbornene 83 (1S,2S)a III 2-butene 13 (S)b IV 1-Me-cyclopentene 22 (S,S) b I 2-Me-1-butene 21 (R)b I 2,3-dimethyl-1-butene 30 (R) b a) Achieved with the 99% enantiopure 3 in THF @-25 °C5 b) Achieved with 92% enantiopure 3 in diglyme @ rt.3, 4

3.1ii: Monoisopinocampheylborane

Ten years passed between the report that monoisopinocampheylborane 3F2

(IpcBH2) can be generated in situ from triisopinocampheyldiborane 3B2 and the report

6 introducing pure IpcBH2 as a hydroborating reagent. Synthesis of 3F2 cannot be achieved by simply treating a single equivalent of pinene with a stoichiometric amount of

7 BH3·THF because the reaction cannot be stopped at the monohydroboration stage.

Brown et al. circumvented this complication by generating IpcBH2·triethylamine complex 3E3 from thexylborane-triethylamine complex (ThxBH2·NEt3) 3E1 and pinene

3A1 (Scheme 3E). The thexyl group serves as a protecting group for the borane: initially preventing the formation of Ipc2BH·NEt3 before undergoing retro-hydroboration to generate tetramethylethylene (TME) 3E2 and the desired complex 3E3. Hydroboration studies were conducted by treating 3E3 with an equivalent of BH3·THF to free the

IpcBH2, followed by 1-methyl-cyclopentene, 2-methyl-2-butene, or 1-methyl- cyclohexene. Subsequent oxidation led to (1S,2S)-2-methyl-cyclopentanol (1S,2S)-3E4,

(S)-3-methyl-2-butanol (S)-3E6, and (1S,2S)-2-methyl-cyclohexanol (1S,2S)-3E5 with

55% ee, 53% ee, and 72% ee, respectively.

105

A revised synthetic route was developed before IpcBH2 was presented as a reagent for hydroborating other types of olefins. The convenience and optical purity provided by the route displayed in Scheme 3E was surpassed by an approach involving treatment of Ipc2BH 3A2 with N,N,N,N-tetramethylethylene-diamine (TMEDA) followed by equilibration at 34 °C to form the TMEDA·2BH2Ipc complex 3F1 with loss of pinene

(Scheme 3F). TMEDA·2BH2Ipc complex 3F1 can be formed with enantiomeric purity approaching 100% from 94% ee pinene, and 3F1 releases IpcBH2 (3F2) upon treatment with BF3·OEt2. The byproduct TMEDA·2BF3 (3F3) precipitates from THF but 3F3 is inert so it‟s removal is not necessary for hydroboration to occur smoothly.8

106

Scheme 3F: Improved Synthesis of IpcBH2 H H H N B BH3·SMe2 B TMEDA B N H H 3F1 3F1 3A2 3A1

H B N BF3 BF3·OEt2 H 3F1 F3B N

3F2 3F3

Table 3G summarizes the results of a series of publications following this revised

9-11 synthesis of IpcBH2 3F2, illustrating its synthetic utility. It is important to note that

Ipc2BH (A2) and IpcBH2 3F2 are complimentary in terms of substrate scope: while 3A2 only works well on Type II olefins, 3F2 works well on Types III and IV but poorly on

Type II olefins. It is also convenient that both of these reagents can generate either desired enantiomer due to the commercial availability of both (+) and (-) -pinene. Table

3P is provided for a direct comparison of asymmetric hydroboration reagents. Despite all the advantages of Brown‟s Ipc2BH (3A2) and IpcBH2 (3F2), the only substrates that are beyond optimization are 1-phenyl-cyclopentene and, arguably, Type II olefins. Every other substrate provides opportunity for improvement, and Type I olefins in particular.

107

Table 3G: Enantioselective Hydroboration with Monoisopinocampheylborane

H 1. Substrate, THF B OH H Alcohol + 2. NaOOH

3F2

Substrate Product Yield ee Substrate Product Yield ee (%) (%) (%) (%) Ph Ph OH 92 100 NRa 23.6

HO S 1S, 2R Ph Ph OH 79 88 NRa 19.7 Et Et HO Et Et S 1S, 2R

OH OH Ph Ph 89 81 73 73 S 2S, 3R OH OH 91 82 Et Et 83 75 Et Et S Ph Ph 2S, 3S OH OH Ph Ph 87 85 72 75 Ph Ph R Et Et 2S, 3R OH Et Et 95 85 Ph Ph 2S, 3S a) Result published using method from Scheme D 12

108

3.1iii: Dilongifolylborane

Despite the complementary nature of Ipc2BH (3A2) and IpcBH2 (3F2), Brown pursued a reagent that could stereoselectively hydroborate both hindered and unhindered olefins, proposing that the defining characteristic of such a reagent would be a steric environment simultaneously more hindered than 3F2 and less hindered than 3A2. These requirements were met by the hydroboration of the sesquiterpene (+)-longifolene 3H1 to form dilongifolylborane 3H2.13 The reagent is convenient to synthesize because the hydroboration of 3H1 stops at the diakylborane stage, at which point 3H2 precipitates from ether. While the bulkiness of 3H2 makes its steric requirement greater than that of

3F2, the fact that the C-B carbons are primary enables the steric requirement to be less than that of 3A2. This does allow the use of 3H2 for hydroboration of a broader range of olefins with good selectivity (Table 3H). However, Type I and Type II olefins are not

Table 3H: Enantioselective Hydroboration with Dilongifolylborane

BH ·SMe 1. Substrate 3 2 Alcohol Et2O 2. NaOOH HO B H H2O2 H H H 3H3 3H1 3H2

Substrate Alcohol Yield (%) ee (%) Substrate Alcohol Yield (%) ee % OH OH 76 75 71 78 Et Et R R OH 83 63 81 71 Et Et Et Et HO R 1R, 2R OH Et Et 81 59 79 70

R HO 1R, 2R

109 discussed in this report and the selectivities reported for type II and type IV olefins are not as synthetically useful as those for 3A2 and 3F2.

3.1iv: Limonylborane

Jadhav synthesized limonylborane 3I3 from another naturally abundant terpene: limonene 3I1. Treating 3I1 with monochloroborane etherate (BH2Cl·OEt2) forms the bicyclic chloroborane 3I2, which is converted to 3I3 by LiAlH4 in the presence of the alkene substrate. While 3I3 does provide reasonable enantioselectivities for all but type I olefin substrates, the selectivities, once again, do not approach those achieved with

Ipc2BH 3A2 and IpcBH2 2F2 (Table 3I).

Table 3I: Enantioselective Hydroboration with Limonylborane

Cl H HO BH2Cl·OEt2 B LiAlH4 B 1. Substrate Alcohol + 2. NaOOH HO 3I1 3I2 3I3 3I4

Substrate Alcohol Yield (%) ee (%) Substrate Alcohol Yield (%) ee (%) OH 78 55 HO 75 5.2

R R OH 75 58.6 77 46

R HO 1R, 2R OH 70 66.5 R

110

3.1v: (R,R) and (S,S)-2,5-Dimethyl-borolane and Related Reports

Not until 1985 was Brown‟s dominance in the field of asymmetric hydroboration faced with strong competition. Foregoing the Brown approach of forming chiral (di)alkyl boranes by hydroborating terpenes, but returning to a C2-symmetric species, Masamune targeted (R,R)-2,5-dimethylborolane (R,R)-3J7 using an approach involving chiral resolution via amino alcohol complexes.14 The synthetic route, illustrated in Scheme 3J, is quite arduous. The borolane ring is assembled as an aminodialkylborane 3J2 by treating (diethylamino)dichloroborane with bis-Grignard reagent 3J1. After methanolysis to methoxyborolane 3J3, multiple resolutions with aminoalcohols allow isolation of enantio-enriched material. N,N-dimethylethanolamine preferentially complexes to cis-3J3 to form 3J4, which precipitates from Et2O, allowing separation of

111 racemic trans-3J4 from the solution. (S)-Prolinol selectively forms complex 3J5 from

(R,R)-3J3, allowing for separation from its enantiomer (S,S)-3J3, which can be recovered as an enriched complex to (S)-valinol (not pictured). Complex 3J5 undergoes methanolysis to form (R,R)-3J3, which is reduced by LiAlH4 to form the lithium borohydride-etherate 3J6. Treating 3J6 with excess methyl iodide (MeI) abstracts a hydride, allowing borolane dimer 3J7 to form.15 One of the most impressive aspects of the synthesis is that it was undertaken with the knowledge that 3J7 might dimerize and then isomerize into 2,5,5,9-tetramethyl-1,6-diborocyclodecane, 3J8. The analogous conversion occurs with the parent borolane, rendering it incapable of hydroboration.16, 17

Fortunately, the isomerization is slow enough that 3J7 can be used as a reagent for exceptionally enantioselective hydroboration with any generic prochiral olefin except

Type I (Table 3K). This work could have rendered Brown‟s Ipc2BH 3A2 and IpcBH2

3F2 obsolete if not for the incredibly challenging synthesis of 3J7. Masamune‟s work has yet to be duplicated in any context reported in the literature but the proof of concept provides a challenge to the chemical community to find a more synthetically viable method of generating enantio-pure C2-symmetric hydroboration agents.

112

Table 3K: Enantioselective Hydroboration with (R,R)-2,5-Dimethylborolane

OH H 1. Substrate, MeI, Et2O B Li·OEt2 Alcohol + H 2. NaOOH OH 3J9 4K1

Yield ee Yield ee Substrate Alcohol Substrate Alcohol (%) (%)a (%) (%)a

OH 75 97.6 89 100 HO (S,S) S

71 99.5 60 95.6 HO (S,S)

OH OH 83 99.9 97 99.3 Et Et Et Et S S

Et 83 99.5 Et

OH H 90 94.2 HO 85 1.5

S S a) adjusted to compensate for the slight cis-borolane impurity present during the reaction.

Hodgetts and coworkers tried to answer the challenge by targeting the more hindered borolane, 2,5-diisopropyl-borolane 3L5.18 Optimum conditions were developed for diastereoselective hydroboration of 2,7-dimethyl-2,6-octadiene 3L1 to favor trans-

2,5-diisopropyl-borolane. Treating 3L1 with dimethylsulfide monobromoborane

(BH2Br·SMe2) in carbon tetrachloride at 76 °C provides 4 : 1 trans-selectivity in 70 % overall yield. The undesired cis-borolane forms complex 3L3 upon treatment with pyrrolidinoethanol, allowing purification of racemic trans-methoxyborolane via cannula

113 filtration and distillation to provide 3L4 in 63% yield. However, no further progress towards an enantiopure hydroborating reagent was reported, leaving Masamune‟s legacy unfulfilled (Scheme 3L). Despite taking Masamune‟s arduous aminoalcohol resolution approach, the improved borolane ring synthesis via diastereoselective hydroboration rather than bis-Grignard addition makes Hodgetts' work noteworthy, as it doubles the yield of racemic-trans-borolane after separation from the undesired cis-isomer.

Scheme 3L: Hodgett's Progress Toward Enantiopure 2,5-Diisopropylborolane

N 1.BH2Br·SMe2 OH B hexanes, -78 °C B B CCl , 76 °C 4 OMe H 2. MeOH OMe 3L1 3L2 3L4 3L5 70% racemic (S,S) or (R,R) 4 : 1 d.r. O B N 3L3

Knochel recognized the significance of C2-symmetry in asymmetric hydroborating reagents and pursued what he refers to as pseudo-C2-symmetric monoalkylboranes 3M1-3, in which C2-symmetric alkyl appendages are incorporated on boron.19 Considering the general difficulty in achieving chiral induction with Type I olefins due to the spatial separation of the chiral boron environment from the prochiral center of the substrate,4 it is understandable that the enantioselectivity achieved with 3M2 is lower than that achieved with borolane 3J7 (Table 3M). Not only is the selectivity low, but the syntheses of boranes 3M1 and 3M2 involve 9-10 steps and will not be elaborated in this text. Psuedo-C2-symmetric boranes are clearly not worthy successors to Masamune‟s borolane, as their syntheses are longer and their selectivities are inferior.

114

Table 3M: Enantioselective Hydroboration with Knochel’s boranes Borane Substrate Yield ee Borane Substrate Yield ee (%) (%) (%) (%)

BH2 Ph Ph 59 38 68 38

3M1 BH2 54 55 67 64

3M2 BH2 Ph Ph 57 29 66 52

3M3

3.1vi: B-H-9-Boracyclo[3.3.2]decanes

Less than two years ago, came a report of a reagent that improved upon the 30% ee hydroboration with Type I olefins achieved by Brown in 1964. The Soderquist group combined a version of Masamune‟s resolution technique with an ingenious homologation approach to access chiral bicyclic boranes reminiscent of Jadhav‟s limonylborane 3I3.20

Exploiting single carbon homologation capabilities of boron, B-methoxy-9- borabicylco[2.2.1]nonane (B-MeO-9-BBN) 3N1 was converted into B-methoxy-10- trimethylsilyl-9-borabicylco[3.3.2]decane 3N2a and B-methoxy-10-phenyl-9- borabicylco[3.3.2]decane 3N2b using the appropriately substituted diazomethane reagents. Each enantiomer is isolable by crystalization via sequential treatment with the two enantiomers of pseudoephedrine (Scheme 3N).

115

Scheme 3N: Generation and Resolution of Soderquist's 9-Boracyclo[3.3.2]decanes Ph Me OMe OMe OMe HO NHMe B RCHN B B Me Ph 2 TMS TMS MeCN hexanes MeHN OH reflux, 10 h 3N2 3N1 3N2a R= TMS: 97 % Ph Ph 3N2b R= Ph: 90 % Me Me racemic O O MeHN B NHMe TMS B TMS

3N3a(S): R= TMS: 38 % 3N3a(R): R= TMS: 28 % 3N3b(S): R= Ph: 39.5 % 3N3b(R): R= Ph: 28 %

The enantiopure complexes 3N3a and 3N3b are reduced to their respective borohydrides 3O1 and 3O2, followed by hydride abstraction with TMSCl to generate the chiral reagent for hydroboration, similar to the approach of Masamune (Table 3O).

Enantioselectivities from 28 to 98% ee are achieved with Type I olefins, which sets this work apart from its predecessors.21 The 10-TMS- and the 10-Ph-9-borabicylco-

[3.3.2]decane reagents 3O1 and 3O2 are similar to Brown‟s Ipc2BH 3A2 and IpcBH2

3F2 in that neither is ideal for all four olefin types, although they complement each other in terms of substrate scope. Table 3P has been compiled for convenient comparison of asymmetric hydroborating agents. Soderquist‟s reagents are clearly best for Type I olefins and hold their own with Masamune‟s borolane with Type III olefins. Synthetic route aside, Masamune‟s borolane is the best reagent for Type IV olefins and Type II olefins, which also provide comparable enantioselectivities upon treatment with Brown‟s

Ipc2BH 3A2.

116

Borane Substrate Alcohol Yield ee Borane Substrate Alcohol Yield ee Source (%) (%) Source (%) (%) OH 3O1-R 98 84 3O2-S 90 32 S OH OH 3O1-R 95 95 3O2-S 90 96 R S H OH

3O1-R 3O2-S 79 74 HO 87 40 S S H H

3O1-R HO 3O2-S HO 84 92 S 60 56 R

D H D H

3O1-R Ph HO Ph 3O2-S 97 92 S 83 66 Ph HO R Ph

H H

3O1-S HO 3O2-S HO R Ph 95 78 82 52 R

3O1-S 3O2-R 83 28 86 98

H 3O2-R HO 97 38 S

117

3.1vii: Summary of Asymmetric Hydroboration

In summary, the major breakthroughs in asymmetric hydroboration include

Brown‟s terpene-hydroboration approach, Masamune‟s C2-symmetric borolane approach, and Soderquist‟s bicyclicborane approach. All three incorporate a resolution of some kind: Brown‟s IpcBH2 from Ipc2BH are resolvable using TMEDA and both

Masamune and Soderquist use chiral aminoalcohols for resolving borane enantiomers.

Both the Brown ipc-derived reagents and the two Soderquist reagents 3O1 and 3O2 complement each other to allow good enantioselectivity with all four types of prochiral alkene substrates. However, neither of those four reagents approach the versatility of

Masamune‟s C2-symmetric borolane, which provides both 1) a standard by which other asymmetric hydroboration reagents are judged, and 2) an obvious indication as to how said standard could be surpassed.

118

119

3.2 C2-Symmetric Boron Reagents in Organic Synthesis

The synthetic application of Brown‟s C2-symmetric diisopinocampheylboron

(Ipc2B-) moiety is not limited to hydroboration. It has been used in a wide variety of asymmetric transformations including ketone reductions, aldol reactions, and several different variations of allylation. It has also provided a model for implementing new C2-symmetric boron species in the context of the aforementioned transformations and others.

3.2i Asymmetric Aldol Reactions Using C2-Symmetric Borons as Lewis Acids

In 1984, Meyers used diisopinocampheylborane triflate (Ipc2BOTf) to investigate an alternative to Evans‟ chiral auxiliary approach to asymmetric aldol reactions.

Whereas Evans has achieved stereoselective aldol reactions by forming a boron enolate from an enantio-enriched substrate 3Q1 with a racemic boron Lewis acid (Scheme 3Q),22

Meyers proposed using a racemic substrate with an enantio-enriched boron Lewis acid.23

Treating 2-ethyl-4,4-dimethyl-2-oxazoline 3R1 with Ipc2BOTf provides azaenolate 3R2, which provides the threo-addition product 3R4 with ≥9 : 1 diastereoselectivity and 77% ee after treatment with an aldehyde followed by subsequent oxidative workup with H2O2

(Table 3R). While these selectivities do not eclipse those achieved by Evans it is clear that the C2-symmetric boron environment generates significant enantioinduction.

Scheme 3Q: Evans' Chiral Auxiliary Approach to Stereoselective Aldol Chemistry

O O O B O OB(nBu)2 O O OH SMe SMe H O N OTf O N O N THF SMe -78 °C

3Q1 3Q2 3Q3 >99 : 1 Erythro 96.8% ee

120

Table 3R: Stereoselectivity of Ipc2B Azaenolates in Aldol Reactions O Ipc H O OH Ipc BOTf O 2 O H R O N Et O, -78 °C N B R R N iPr NEt N 2 O H O 2 Ipc 2 2 Me B(Ipc)2 3R1 3R2 3R3 threo-3R4

Entry R Threo : Erythro ee (%) 1 Et 92 : 8 77 2 nPr 91 : 9 77 3 nPent 90 : 10 77 4 iPr 91 : 9 85 5 chex 95 : 5 84 6 tBu 94 : 6 79

Meyers‟ success inspired Paterson to investigate a more traditional chiral enolate

24,25 equivalent using Ipc2BX species with ethyl and methyl ketones. It was found that treating an ethyl ketone 3S1 with Ipc2BOTf and Hunig‟s base in CH2Cl2 at -78 °C provides a >97:3 preference for the Z-enolates 3S2. Treating the Z-enolate with an aldehyde followed by oxidative workup provides excellent syn-diastereoselectivity and enantioselectivity (Table 3S). Enolization of diethylketone with Ipc2BCl and Et3N occurs with modest selectivity to generate a 4:1 E- : Z- mixture of enol borinates. Treating this mixture with methacrolein followed by H2O2 generates a 4:1 anti:syn aldol mixture, which is to be expected. However, the corresponding anti-aldol product is generated in

<20% ee and the syn-aldol product is generated in 80% ee, which is 11% lower than the result reported in entry 1 of Table 3S. Paterson et al. point out that “enol diisopinocampheylborinates with E configuration are unlikely to be useful for asymmetric anti aldol reactions.” Investigation of methylketone-derived enol borinates led to the observation of good enantio-induction using either Ipc2BCl or Ipc2BOTf.

121

Interestingly, the enantiofacial selectivity of aldehydes with methyl ketone-derived enol borinates (re-face) is the opposite of that for the ethyl ketone analogs (si- face).

Table 3S: Stereoselectivity of Enol Borinates in Aldol Reactions O Ipc2BOTf O OH M Ipc O iPr NEt OB(Ipc)2 2 H R2 B O R1 R2 R1 R1 S CH Cl L O H2O2 2 2 H 2 Me -78 °C R 3S1 3S2 3S3 3S4 > 97 :3 Z- Major Isomer

Entry R1 R2 syn : anti ee (%) yield (%)

1 Et H2C=C(Me) 98 : 2 91 78 2 Et nPr 97 : 3 80 92 3 Et iPr 96 : 4 86 75 4 Et 2-furyl 96 : 4 80 84

5 Ph H2C=C(Me) 98 : 2 91 97 6 iPr H2C=C(Me) 95 : 5 88 99 7 iBu H2C=C(Me) 97 : 3 86 79

Masamune26,27 and Reetz28,29 have investigated asymmetric aldol reactions with enol borinates generated from their respective

C2-symmetric borolanes. The Masamune reagent 3T5 was generated from the borolane triflate 3T3 while the Reetz reagents

3T6 was done using borolane chloride 3T4. Table 3T illustrates that enolates generated from either borolane consistently provide excellent enantioselectivity. Both groups rationalize their enantioselectivities using a Zimmerman-Traxler model 3S5. While the selectivities from Tables 3S and 3T cannot be directly compared, other work by Reetz30 and Paterson25 indicates that these 2,5-trans-disubstituted borolanes provide superior enantio-induction compared to the Ipc2B moiety.

122

Table 3T: Comparing the Masamune and Reetz Borolanes in Aldol Reactions

2 2 R2 R B R X 2 B O O 4T3 R = Me, X= OTf O OH O 2 2 3 1 4T4 R = Ph, X= Cl R H R R 3 SC(Et)3 SC(Et)3 R SC(Et)3 R1 R1 3T1 R1= H 3T5 R2= Me 3T7 1 2 3T2 R = Me 3T6 R = Ph

Entry R1 R2 R3 anti- : syn- Yield of 3T7 (%) ee (%)a 1 H Me nPr - 82 91.2 2 H Ph nHex - 78 95 3 H Me iPr - 81 91.5 4 H Ph iPr - 72 92 5 H Me cHex - 95 90.1 6 H Ph cHex - 87 95 7 Me Me nPr 33 : 1 91 97.9 8 Me Ph nHex >155 : 1 36 100 9 Me Me iPr 30 : 1 85 99.5 10 Me Ph iPr >155 : 1 82 99 11 Me Me cHex 32 : 1 82 98 12 Me Ph cHex >155 : 1 58 98.7 13 Me Me tBu 30 : 1 95 99.9 14 Me Ph tBu >155 : 1 68 94.3

Corey has also approached the problem of asymmetric aldol transformations by application of a C2-symmetric five-membered boracycle to the formation of chiral enolates.31 N,N’-disulfonyl-1,4-trans-diphenyl-2,5-diaza-borolanes 3U2 and 3U3 were studied for convenient comparison to the work of Paterson, Reetz, and Masamune. The enol borinate generated from diethylketone and 3U2 reacts with various aldehydes to provide >94 : 6 syn-diasteroselectivity and excellent enantioselectivity (Table 3U, Entries

1-3). The diastereoselectivity indicates the formation of a Z-enol borinate and the enantioselectivities surpass those achieved by Paterson et al. The enol borinate generated from phenyl thioacetate and 3U2 reacts with aldehydes to provide enantioselectivities similar to those reported by Reetz and Masamune (Table 3T, entries 4&5).

123

Bromoboracyle 3U2 is incapable of enolizing the phenylthio ester of propionic acid but

3U3 can. The resulting Z-enol borinate is an excellent complement to the E-enolates reported by Reetz and Masamune (Table 3T), as it provides >94 : 6 syn- diastereoselectivity and excellent enantioselectivities upon treatment with aldehydes

(entries 6 & 7).

Table 3U: Aldol Reactions with B-Bromo-N,N'-disulfonyl-1,4-trans-diphenyl-2,5-diaza-borolanes O 2 Ar R 3 R SO2 O H R3 O O OH H N Ph O 1 3 R1 B R R 1 H 2 Ph Ph R N 2 R Ph R O O O2S S N N S Ar 3U1 Ar B Ar 3U4 3U5 O O Br 3U2: Ar= tol 3U3: Ar= pNO2-C6H4

Entry R1 R2 R3 Ar syn- : anti- yield (%) ee (%)

1 Et Me Et p-CH3-C6H4 >98 : 2 91 >98 2 Et Me iPr p-CH3-C6H4 98 : 2 85 95 3 Et Me Ph p-CH3-C6H4 94.3 : 5.7 95 97 a 4 PhS H iPr p-CH3-C6H4 - 82 83 a 5 PhS H Ph p-CH3-C6H4 - 84 91 6 PhS Me iPr p-NO2-C6H4 94.5 : 5.5 72 97 7 PhS Me Ph p-NO2-C6H4 98.3 : 1.7 70 95 a) The isolated enantiomer is the opposite of the one illustrated

This section serves as a demonstration of both the versatility and influence of

Brown‟s C2-symmetric Ipc2B moiety. Paterson and Meyers have made it clear that ipc borane reagents have synthetic applications beyond hydroboration, as do the

C2-symmetric Masamune and Reetz borolanes. These borolanes participate in aldol

124 reactions to provide excellent diastereo- and enantioselectivities that surpass and/or complement those achieved by Corey et al. with C2-symmetric 2,5-diazaborolanes. The versatility of several of these chiral boron environments will be demonstrated further in the following sections.

3.2ii Asymmetric Allylations Using C2-Symmetric Allyl-Boron Species

Brown et al. have extensively studied a variety of asymmetric allylations with

B-allyl-diisopinocampheylboranes (Table 3V).32-36 Allylation of several -branched alkyl aldehydes provides secondary homoallylic alcohols with good enantioselectivity

(83-90% ee; entries 1-3).32 Analogous methallylation34 (entries 4-6) and isoprenylation33

(entries 7 & 8) have also been achieved, (89-96% ee). Similarly, crotylation using either

E- or Z-crotyldiisopinocampheylborane generates highly diastereo-enriched alcohols with excellent enantioselectivity (entries 9-14).35,36 These high selectivities have been attributed to six-membered transition states resembling erythro-3V3 and threo-3V3.

Corey has also applied his C2-symmetric B-allyl-N,N’-disulfonyl-1,4-trans- diphenyl-2,5-diaza-borolane 3W1 to the allylation of aldehydes. The enantioselectivities

(≥95% ee) meet if not exceed those previously reported (Table 3W).37

125

Table 3V: Enantioselectivity with Diisopinocampheylborane Allylation Reagents

M O M S L 1 1 S L R 4 R OH H 2 H R H B S Ipc2B R B S H R4 H C O M 3 O R L R3 R2 R3 R L M CH3 H erythro-3V3 3V1 3V2 threo- 3V3

Entry R1 R2 R3 R4 Yield (%) threo- : erythro- ee (%) Configuration 1a H H H nPr 72 - 87 R 2a H H H iPr 86 - 90 S 3a H H H tBu 88 - 83 S 4b Me H H nPr 56 - 91 S 5b Me H H iPr 57 - 96 R 6b Me H H tBu 55 - 90 R 7b H Me Me nPr 79 - 92 S 8b H Me Me iPr 73 - 89 S 9a H Me H Et 70 >99 : 1 90 - a 10 H Me H CH2=CH 65 >99 : 1 90 - 11a H Me H Ph 79 >99 : 1 88 - 12a H H Me Et 70 1 : 99 90 - a 13 H H Me CH2=CH 63 1 : 99 90 - 14a H H Me Ph 72 1 : 99 88 - (a) (+)--pinene was used. (b) (-)--pinene was used

Entry R ee (%) configuration 1 nPent 95 S 2 cHex 97 R

3 PhCH2=CH 97 R 4 Ph 95 R

126

The Roush group has also reported good enantioselectivities for the allylation of simple aldehydes using 2-allyl-1,3,2-dioxaborolane-4,5-dicarboxylic esters derived from tartrate esters 3X1.38,39 These C2-symmetric reagents and their E-crotyl analogs have also been investigated in the context of double asymmetric synthesis with the

-dialkoxyaldehydes 3X2 and 3Y1. Roush et al. discovered that, when using aldehyde

3X2, the diastereoselectivity of alcohol formation can be reversed by simply switching the enantiomer of diisopropyltartrate (DIPT) from which the reagent is generated. The diastereoselectivies are not identical, which indicates that there is a matched vs. mismatched effect, albeit a minor one. The anti-diastereoselectivity is good with either enantiomer of the crotylating reagent (Table 3X).

Table 3X: Stereoselectivity with 2-Allyl-1,3,2-dioxaborolane-4,5-dicarboxylic Esters

O CO2R H O R1 O R1 O O 3X2 1 O O O R B CO2R O OH OH 3X1 3X3 3X4

Entry R1 tartrate anti- : syn- 3X3 : 3X4 Yield (%) 1 H (+)-DIPT - 96 : 4 91 2 H (˗)-DIPT - 8 : 92 - 3 Me (+)-DIPT 96 : 4 87 : 9 87 4 Me (-)-DIPT 98 : 2 2 : 96 85

The aldehyde 3Y1 apparently creates a greater matched vs. mismatched effect in the context of simple allylation. While the (-)-DIPT-derived reagent provides excellent

96% dr, the (+)-DIPT-derived reagent provides only 18% dr of the opposite anti-enantiomer. Other (+)-tartrate esters were screened to find that the highest selectivity that could be achieved was 24% dr, with (+)-diethyltartrate [(+)-DET] (Table

127

3Y, Entries 1-3). Interestingly, the corresponding crotylation provides comparable dr and good anti-diastereoselectivities with both DIPT-derived boronate enantiomers (Entries 4-

5). While there are clearly mismatched scenarios that lead to poor selectivity in certain cases, this methodology provides convenient access to a variety of acyclic polyoxygenated moieties in a stereocontrolled fashion.

Table 3Y: Stereoselectivity with 2-Allyl-1,3,2-dioxaborolane-4,5-dicarboxylic Esters

O H O CO2R 1 R1 O R O O O 3Y1 O 1 O R B CO2R O OH OH 3X1 3Y2 3Y3

Entry R1 tartrate anti- : syn- 3Y2 : 3Y3 Yield (%) 1 H (+)-DET - 68 : 32 63 2 H (+)-DIPT - 64 : 36 - 3 H (˗)-DIPT - 2 : 98 94 4 Me (+)-DIPT 97 : 3 93 : 4 88 5 Me (-)-DIPT 92 : 8 4 : 88 80

As with hydroboration and aldol chemistry, C2-symmetric boron species allow significant progress in the field of asymmetric allylation of aldehydes. Again, Brown‟s

Ipc2B moiety has provided a foundation upon which others have built: Corey‟s diazaborolane provides improved selectivity for allylation while Roush‟s tartrate-derived boronates have demonstrated significant allylation and crotylation applications in double asymmetric synthesis.

128

3.2iii Asymmetric Reduction of Ketones with C2-Symmetric Boron Species

The final common application for Brown‟s Ipc2B moiety is the reduction of ketones by B-chloro-diisopinocampheylborane (Ipc2BCl or DIPCl). Aryl-alkyl ketones are the best substrates for DIP-Cl, providing almost perfect enantioselectivity, while the enantioinduction achieved on dialkyl ketones depends on -substitution of one of the alkyl functionalities. These trends are presented in Table 3Z: only 4% ee is achieved in

DIPCl reduction of 2-butanone (entry 1) but adding a 3-methyl group provides 32% ee

(entry 2) and adding another provides 93% ee (entry 3), which is almost as high as the selectivity achieved on acetophenone and indanone (entries 5&6).40, 41

Table 3Z: Reduction of Prochiral Ketones with (-)-Ipc2BCl Entry Ketone ee (%) configuration 1 2-butanone 4 S 2 3-methyl-2-butanone 32 S 3 3,3-dimethyl-2-butanone 93 S 4 2,2-dimethylcyclopentanone 71 S 5 acetophenone 98 S 6 1-indanone 97.4 S

Masamune‟s borolane 3J9 is a complementary to DIP-Cl, in that it reduces dialkyl ketones with excellent enantioselectivity.42 Borolane 3AA1 by itself is not a selective reducing reagent. However, in the presence of its corresponding mesylate

3AA2, which serves as a chiral Lewis acid,15 superb enantioselectivity is achieved on a variety of dialkylketones (Table 3AA). No aryl-alkyl ketones were studied. However, the commercially available DIP-Cl is a much more convenient reagent for the asymmetric reduction of aryl-alkyl ketones reduction due to the difficult synthesis of 3J9.

129

Table 3AA: Reduction of Ketones Using the Masamune Borolane O

H MsOH R1 R2 OH B H B OMs B Li·OEt2 1 2 H R R

3J9 3AA1 3AA2 3AA3

Entry R1 R2 yield (%) ee (%) configuration 1 Me Et 75 80.3 R 2 Me Bn 69 98.9 R 3 Me iPr 69 100 R 4 Me cHex 83 99.5 R 5 Me tBu 72 99.3 R 6 nBu iBu 72 96.8 R

Kagan has pursued asymmetric reduction of ketones using C2-symmetric oxazaborolidine catalysts.43 Treating acetophenone with 10 mol% of either 3BB2 or

3BB4 and stoichiometric THF·BH3 provides 1-phenyl-ethanols S-3BB3 and R-3BB3 in

65% ee and 38% ee, respectively (Scheme 3BB). These results are significant because of the substoichiometric amounts of chiral species used. All of the other applications discussed previously require stoichiometric amounts of asymmetric reagent(s).

130

3.2iv: C2-Symmetric Boron Species as Ligands

Kagan has also investigated asymmetric using C2-symmetric phenylboronate 3CC5 derived from his (R,R)-DIOP ligand.44 This chiral ligand was designed to have a dual role. The phosphines are ligands for rhodium while the boronate can participate in a Lewis-acid/base interaction with the substrate to be hydrogenated.

However, ligand 3CC5 provides reasonable selectivities on several substrates: N-acetyl dehydrophenylalanine 3CC1 and ketopantolactone 3CC3 are both hydrogenated with

3CC5/(RhClCOD)2 to yield 3CC2 and 3CC4 in 73% ee and 54% ee, respectively. The boronate does not provide an advantage over the dimethyl acetal of DIOP, which provides selectivities of 81% ee & 52% ee with 3CC1 and 3CC3, respectively.

Scheme 3CC: Asymmetric Hydrogenation with Boron Ananlog of DIOP Ligand Ph H Ph H O HO H (RhClCOD)2 H (RhClCOD)2 O O O N CO H O N CO H 2 3CC5, H2 2 3CC5, H2 H H O O 3CC1 3CC2 3CC3 3CC4 100% yield 100% yield 73% ee 54% yield

Ph H Ph H O HO H (RhClCOD)2 H (RhClCOD)2 O O O N CO H O N CO H 2 DIOP, H2 2 DIOP, H2 H H O O 3CC1 3CC2 3CC3 3CC4 100% yield 100% yield 81% ee 52% ee

H PPh H PPh O 2 O 2 Ph B O PPh O H 2 H PPh2 3CC5 DIOP

131

3.3: Summary

The C2-symmetric diisopinocampheylborane discovered by Brown in 1962 has led to a multitude of breakthroughs in asymmetric synthesis, not only as a practical asymmetric hydroboration reagent, but also as the starting point for enantioselective allylation, reduction, and aldol reactions. In the realms of hydroboration, ketone reduction, and aldol reactions, the standards for enantioselectivity set using the Ipc2B- moiety have been surpassed by the Masamune trans-2,5-dimethyl-borolane.

Unfortunately, said borolane is inconvenient to synthesize. Corey‟s diazaborolane moiety has proven to be a superior reagent for allylation and aldol reactions and is conveniently produced. These results, in addition to the allylation studies published by

Roush, demonstrate that C2-symmetric (hetero)borolanes are effective boron species for asymmetric synthesis.

132

References for Chapter 3

1. Brown, H. C.; Rao, B. C. S. A New Technique for the Conversion of Olefins Into Organoboranes and Related Alcohols. J. Am. Chem. Soc. 1956, 78, 5694.

2. Brown, H. C.; Zweifel, G. Hydroboration as a Convenient Procedure for Asymmetric Synthesis of Alcohols of high Optical Purity. J. Am. Chem. Soc. 1961, 83, 486.

3. Brown, H. C.; Ayyangar, N. R.; Zweifel, G. Hydroboration.19. Reaction of Diisopinocampheylborane with Representative Trans + Hindered Olefins (Triisopinocampheyldiborane as Reagent for Configurational Assignment of Alcohols + Olefins via Hydroboration. J. Am. Chem. Soc. 1964, 86, 1071.

4. Zweifel, G.; Munekata, T.; Brown, H. C.; Ayyangar, N. R. Hydroboration. 20. Reaction of Diisopinocampheylborane with Representaive 2-Methyl-1-alkenes - Convenient Synthesis of Optically Active 2-Methyl-1-Alkanols. J. Am. Chem. Soc. 1964, 86, 1076.

5. Brown, H. C.; Desai, M. C.; Jadhav, P. K. Hydroboration. 61. Diisopinocampheylborane of High Optical Purity - Improved Preparation and Asymmetric Hydroboration of Representative Cis-Disubstituted Alkenes. J. Org. Chem. 1982, 47, 5065.

6. Brown, H. C.; Yoon, N. M. Monoisopinocampheylborane - New Chiral Hydroborating Agent for Relatively Hindered (Trisubstituted) Olefins. J. Am. Chem. Soc. 1977, 99, 5514.

7. Brown, H. C.; Klender, G. J. Organoboranes. 2. Preparation and Properties of Alkyldiborane Containing Bulky Alkyl Substituents. Inorg. Chem. 1962, 1, 204.

8. Brown, H. C.; Schwier, J. R.; Singaram, B. Simple Synthesis of Monoisopinocampheylborane of High Optical Purity. J. Org. Chem. 1978, 43, 4395.

9. Mandal, A. K.; Jadhav, P. K.; Brown, H. C. Monoisopinocampheylborane - And Excellent Chiral Hydroborating agent for Phenyl-Substituted Tertiary Olefins - Synthesis of Alcohols Approaching 100-Percent Enantiomeric Excess. J. Org. Chem. 1980, 45, 3543.

10. Brown, H. C.; Jadhav, P. K. High Asymmeric Induction in the Chiral Hydroboration of Trans- Alkenes with isopinocampheylborane - evidence for a Strong Steric Dependence in Such Asymmetric Hydroborations. J. Org. Chem. 1981, 46, 5047.

133

11. Brown, H. C.; Jadhav, P. K.; Mandal, A. K. Hydroboration. 62. Monoisopinocampheylborane, an Excellent Chiral Hydroboration Agent for Trans-disubstituted and Trisubstituted Alkenes- Evidence for a Strong Steric Dependence in Such Asymmetric Hydroborations. J. Org. Chem. 1982, 47, 5074.

12. Mandal, A. K.; Yoon, N. M. Anomalies in Asymmetric Hydroboration of Olefins with 1-1 Adduct of (+)-Alpha-Pinene and BH3·THF. J. Organomet. Chem. 1978, 156, 183.

13. Jadhav, P. K.; Brown, H. C. Dilongifoylborane- A New Effective Chiral Hydroborating Agent with Intermediate Steric Requirements. J. Org. Chem. 1981, 46, 2988.

14. Masamune, S.; Kim, B. M.; Petersen, J. S.; Sato, T.; Veenstra, S. J.; Imai, T. Organoboron Compounds in Organic Synthesis. 1. Asymmetric Hydroboration J. Am. Chem. Soc. 1985, 107, 4549.

15. Masamune, S.; Kennedy, R. M.; Petersen, J. S.; Houk, K. N.; Wu, Y. Organoboron Compounds in Organic-Synthesis. 3. Mechanism of Asymmetric Reduction of Dialkyl Ketones with (R,R)-2-5-Dimethylborolane. J. Am. Chem. Soc. 1986, 108, 7404.

16. Brown, H. C.; Negishi, E. Bisborolane- Highly Elusive Bisboracyclane. J. Am. Chem. Soc. 1971, 93, 6682.

17. Brown, H. C.; Negishi, E. Boraheterocycles via Cyclic Hydroboration. Tetrahedron 1977, 33, 2331.

18. Laschober, G.; Zorzi, M.; Hodgetts, K. J. Synthesis of (+/-)-trans-2,5- diisopropylborolane. Molecules 2001, 6, 244.

19. Graf, C. D.; Knochel, P. Asymmetric hydroboration with new chiral monoalkylboranes bearing a non-stereogenic, chirotopic center. Tetrahedron 1999, 55, 8801.

20. Gonzalez, A. Z.; Roman, J. G.; Gonzalez, E.; Martinez, J.; Medina, J. R.; Matos, K.; Soderquist, J. A. 9-Borabicyclo[3.3.2]decanes and the Asymmetric Hydroboration of 1,1-Disubstituted Alkenes. J. Am. Chem. Soc. 2008, 130, 9218.

21. Thomas, S. P.; Aggarwal, V. K. Asymmetric Hydroboration of 1,1-Disubstituted Alkenes. Angew. Chem. Int. Ed. 2009, 48, 1896.

22. Evans, D. A.; Bartroli, J.; Shih, T. L. Enantioselective Aldol Condensations. 2. Erythro-Selective Chiral Aldol Condensations via Boron Enolates. J. Am. Chem. Soc. 1981, 103, 2127.

134

23. Meyers, A. I.; Yamamoto, Y. Stereoselectivity in the Aldol Reaction:The Use of Chiral and Achiral Oxazolines as Their Boron Azaenolates. Tetrahedron 1984, 40, 2309.

24. Paterson, I.; Lister, M. A.; McClure, C. K. Enantioselective Aldol Condensations - The Use of Ketone Boron Enolates with Chiral Ligans Attached to Boron. Tetrahedron Lett. 1986, 27, 4787.

25. Paterson, I.; Goodman, J. M.; Lister, M. A.; Schumann, R. C.; McClure, C. K.; Norcross, R. D. Enantio- and Diasteroselective Aldol Reactions of Achiral Ethyl and Methyl Ketones with Aldehydes: The Use of Enol Diisopinocampheylborinates. Tetrahedron 1990, 46, 4663.

26. Blanchette, M. A.; Malamas, M. S.; Nantz, M. H.; Roberts, J. C.; Somfai, P.; Whritenour, D. C.; Masamune, S.; Kageyama, M.; Tamura, T. Synthesis of Bryostatins. 1. Construction of the C(1)-C(16) Fragment. J. Org. Chem. 1989, 54, 2817.

27. Masamune, S.; Sato, T.; Kim, B. M.; Wollmann, T. A. Organoboron Compounds in Organic Synthesis. 4. Asymmetric Aldol Reactions. J. Am. Chem. Soc. 1986, 108, 8279.

28. Reetz, M. T. Asymmetric C-C Bond Formation Using Organometallic Chemistry. Pure Appl. Chem. 1988, 60, 1607.

29. Reetz, M. T.; Rivadeneira, E.; Niemeyer, C. Reagent Control in the Aldol Addition of Chiral Boron Enolates Based of the 2,5-Diphenylborolane Ligand System. Tetrahedron Lett. 1990, 31, 3863.

30. Reetz, M. T.; Kunisch, F.; Heitmann, P. Chiral Lewis-Acids for Enantioselective C-C Bond Formation. Tetrahedron Lett. 1986, 27, 4721.

31. Corey, E. J.; Imwinkelried, R.; Pikul, S.; Xiang, Y. B. Practical Enantioselective Diels-Alder and Aldol Reactions Using a New Chiral Controller System. J. Am. Chem. Soc. 1989, 111, 5493.

32. Brown, H. C.; Jadhav, P. K. Asymmetric Carbon-Carbon Bond Formation via B- Allyldiisopinocampheylborane. Simple Synthesis of Secondary Homoallylic Alcohols with Excellent Enantiomeric Purities. J. Am. Chem. Soc. 1983, 105, 2092.

33. Brown, H. C.; Jadhav, P. K. 3,3-Dimethylallyldiisopinocampheylborane: A NOvel Reagent for Chiral Isoprenylation of Aldehydes. Synthesis of (+)- and (-)- Artemisia Alcohol in Exceptionally High Enantiomeric Purity. Tetrahedron Lett. 1984, 25, 1215.

135

34. Brown, H. C.; Jadhav, P. K.; Perumal, P. T. Asymmetric Methallylboration of Prochiral Aldehydes with Methallyldiisopinocamphenylborane - Synthesis of 2- Methyl-1-alken-4-ols in Greater-Than 90% Enantiomeric Purities. Tetrahedron Lett. 1984, 25, 5111.

35. Brown, H. C.; Bhat, K. S. Chiral Synthesis via Organoboranes. 7. Diastereoselective and Enantioselective Synthesis or erythro- and threo-- Methylhomoallyl Alcohols via Enantiomeric (Z)- and (E)-Crotylboranes. J. Am. Chem. Soc. 1986, 108, 5919.

36. Brown, H. C.; Bhat, K. S. Enantiomeric (Z)- and (E)- Crotyldiisopinocampheylboranes. Synthesis in High Optical Purity of all Four Possible Stereoisomers of -Methylhomoallyl Alcohols. J. Am. Chem. Soc. 1986, 108, 293.

37. Corey, E. J.; Yu, C. M.; Kim, S. S. A Practical and Efficient Method for Enantioselective Allylation of Aldehydes. J. Am. Chem. Soc. 1989, 111, 5495.

38. Roush, W. R.; Walts, A. E.; Hoong, L. K. Diastereo- and Enantioselective Aldehyde Addition Reactions of 2-Allyl-1,3,2-dioxaborolane-4,5-dicarboxylic Esters, a Useful Class of Tartrate Ester Modified Allylboronates. J. Am. Chem. Soc. 1985, 107, 8186.

39. Roush, W. R.; Halterman, R. L. Diisopropyl Tartrate Modified (E)- Crotylboronates: Highly Enantioselective Propionate (E)-Enolate Equivalents. J. Am. Chem. Soc. 1986, 108, 294.

40. Brown, H. C.; Chandrasekharan, J.; Ramachandran, P. V. Highly Efficient Asymmetric Reduction od -Tertiary Alkyl Ketones with Diisopinocampheylchloroborane. J. Org. Chem. 1986, 51, 3394.

41. Brown, H. C.; Chandrasekharan, J.; Ramachandran, P. V. Chiral Synthesis via Organoboranes. 14. Delective Reductions. 41. Diisopinocampheylchloroborane, an exceptionally Efficient Chiral Reducing Agent. J. Am. Chem. Soc. 1988, 110, 1539.

42. Imai, T.; Tamura, T.; Yamamuro, A.; Sato, T.; Wollmann, T. A.; Kennedy, R. M.; Masamune, S. Organoboron Compounds in Organic Synthesis. 2. Asymmetric Reduction of Dialkyl Ketones with (R,R)-2,5-Dimethylborolane or (R,R)-2,5- Dimethylborolane. J. Am. Chem. Soc. 1986, 108, 7402.

43. Dubois, L.; Fiaud, J. C.; Kagan, H. B. Enantioselective Borane Reduction of Acetophenone Catalyzed by Oxaborolidines Derived from Chiral Diethanolamines. Tetrahedron: Asymmetry 1995, 6, 1097.

44. Borner, A.; Ward, J.; Kortus, K.; Kagan, H. B. A Boron Analog of DIOP: Synthesis and Properties. Tetrahedron: Asymmetry 1993, 4, 2219.

136

Chapter 4: Asymmetric Hydroboration with N-Tosyl-(R,R)-2,6-diisopropyl-1,4- borazinane

4.1 C2-Symmetric Boranes as Ideal Hydroborating Agents

Hydroboration holds the distinction of being the first practical, non-enzymatic, enantioselective reaction in the annals of synthetic organic chemistry. Despite this head start, the search continues for an ideal asymmetric hydroborating agent. Brown‟s seminal works with diisopinocampheylborane (Ipc2BH) and monoisopinocampheylborane

(IpcBH2) have stood the test of time, providing complementary species for the asymmetric hydroboration of a variety of alkenes (Types II, III, & IV – Equation 1).1-8

Much more recently, Soderquist‟s work with 9-boracyclo[3.3.2]decanes represents dramatic progress in addressing the longstanding problem of asymmetrically hydroborating Type I alkenes.9 Despite the undeniable importance of these developments it is Masamune‟s trans-2,5-dimethylborolane [(R,R)-4A1]10,11 that serves as the closest example of an ideal asymmetric hydroborating reagent in the literature.

Masamune‟s borolane (R,R)-4A1 could have rendered Brown‟s IpcXBH(3-X) reagents synthetically inconsequential if not for the staggering effort involved in its synthesis. Type II, III, and IV alkenes undergo hydroboration with (R,R)-4A1 to provide alcohols with >95% ee after oxidation.11 No other single reagent can claim such versatility, which falls short of perfection only in that <5% ee was observed with Type I alkenes. Borolane (R,R)-4A1 is generated in seven air-sensitive steps from bis-Grignard reagent 4A4, including a prolinol resolution that provides diastereo-enriched 4A3

137

(Equation 2) from the mixture of trans-2,5-dimethylborolane enantiomers. Perhaps more courageous than the synthesis itself is that it was undertaken knowing that (R,R)-4A1 might irreversibly form an unreactive species 4A6 (Equation 2), as borolane 4A7 forms

1,6-diboradecane 4A9 at room temperature (Equation 4).12,13

In pursuing an ideal hydroborating agent, it cannot be ignored that two of the most significant asymmetric hydroborating reagents known, Masamune's (R,R)-4A1 and

Brown's Ipc2BH, both incorporate a C2-symmetric boron environment. While the significance of (R,R)-4A1 has not gone unnoticed,14-16 its legacy remains unfulfilled. In an effort to expand upon what is known about cyclic C2-symmetric boranes, hopefully to develop easier synthetic access, and perhaps also to improve upon (R,R)-4A1, we targeted C2-symmetric borinanes and their symmetrical heteroatom-containing analogs as described below.

Three proposed improvements to the Masamune protocol were investigated in the pursuit of a six-membered trans-2,6-disubstituted-boracycle: 1) assemble the boracycle with trans-diastereoselectivity, 2) increase the steric bulk of the chiral substituents by

138 incorporating isopropyl groups, and 3) use an amino acid as a complexing agent for resolution of boracycle mixture. Diastereoselective ring assembly would conserve material in comparison to racemic boracycle generation from 4A4. Increased steric bulk of the isopropyl substituents vs. Masamune's methyl groups should help to maintain enantioselectivity in case the more flexible borinane environment promotes a looser hydroboration transition state compared to the borolane system. It is conceivable that the isopropyl groups might even provide greater stereoinduction than methyl groups, depending on conformational preferences of the borinane compared to the borolane.

Resolution of diastereomers as well as enantiomers with a single complexing reagent would streamline the overall process and an amino acid complex should help suppress the air-sensitivity issues that contribute to the inconvenience of the Masamune protocol by forming a more stable complex with the cyclic borane.

4.2 Diastereoselective Assembly of trans-2,6-Diisopropyl-(4-hetero)borinanes

Diastereoselective assembly of C2-symmetric trans-2,6-diisopropyl-borinanes and their 4-hetero-analogs was initially pursued by Mr. John Nelson. 2,8-Dimethyl-2,7- nonadiene 4B1a was treated with I2-activated DMAP·BH3 (DMAP·BH2I) or THF·BH3 in an effort to achieve diastereoselective generation of trans-2,6-diisopropyl-borinane-

DMAP complex 4B4a. In the case of THF·BH3, it was envisioned that the trisubstituted alkenes would provide excellent anti-Markovnikov regioselectivity in an intermolecular hydroboration step to provide intermediate 4B2a, which would undergo subsequent intramolecular, anti-Markovnikov hydroboration to provide borinane 4B3a. Addition of

DMAP should then produce the complex 4B4a for convenient assay of diastereoselectivity in the crude product by 1H NMR spectroscopy. Alternatively, 4B4a

139 should be formed directly from 4B1a and DMAP·BH2I, generated in situ from

17-20 DMAP·BH3 and iodine. Given the differences in boron substituents at the intramolecular hydroboration stage from 4B2a or from 4B7a, we expected to have several options for fine-tuning hydroboration diastereoselectivity to favor the desired trans-4B4a. However, the simpler reagent THF·BH3 proved to react with the best trans- selectivity with all of the substrates.

Preliminary NMR assay of the product mixture left no doubt that the minor product from 4B1a and THF·BH3 is the symmetrical DMAP complex cis-4B4a, and by implication, supported the unsymmetrical structure trans-4B4a for the major complex.

However, both cis- and trans-4B4a were somewhat unstable upon silica gel chromatography and could not be purified. To corroborate the NMR assays, the mixture of hydroboration products was therefore converted into diols 4B5a and 4B6a in excellent

140 yield via standard oxidation conditions (Table 4C, entries 1&2). Both diols were symmetrical as expected and could be assayed by HPLC. The DMAP·BH2I conditions afforded 4B5a and 4B6a in a ratio of 7.8:1, reflecting predominant cis- diastereoselectivity in the internal hydroboration step from 4B7. On the other hand, the

THF·BH3 conditions provided a 1:1.9 ratio of 4B5a:4B6a, corresponding to modest trans-diastereoselectivity from 4B2a.

Entry X Conditionsa Time (h) Yield (%) 4B5:4B6 b 1 CH2 A 0.5 99 7.8 : 1 b 2 CH2 B 0.5 92 1 : 1.9 b 3 SO2 A 18 88 4.0 : 1 b 4 SO2 B 0.5 99 1 : 1.3 5 NTs A 14 90 5.8 : 1 6 NTs B 0.5 98 1 : 3.6 7c NTs B 24 NR 1 : 5.6d

(a) Conditions A: DMAP·BH3 (1.2 equiv) was activated with I2 (0.6 equiv) in DCM at 0 °C. After warming to rt, 10 was added, the solution was stirred (time), and was quenched with NaOOH/MeOH. Conditions B:

A rt solution of 10 in THF was treated with 1.0 M THF·BH3 (1.2 equiv) and stirred (time) prior to quenching with NaOOH/MeOH. (b) Assayed as the dibenzoyl derivative. (c) Performed at -30 °C. (d) Ratio determined by NMR assay of crude diol.

It was hypothesized that the inductive effect of incorporating an electron withdrawing group (EWG) at the 4-position of the borinane might sufficiently stabilize the DMAP-boron complex, allowing for isolation. The same inductive effect might also help to improve diastereoselectivity in the 4-hetero-borinane hydroborations. Studies with diprenyl sulfone 4B1b confirmed the former hypothesis, as both cis- and trans-

141

DMAP complexes 4B4b were stable to silica gel chromatography despite being difficult to separate from one another. However, improved diastereoselectivity was not observed; lower cis diastereoselectivity (4:1) was found under DMAP·BH2I conditions and marginal trans diastereoselectivity (1:1.3) was observed using THF·BH3 (entries 3 & 4). Thus the primary concern became achieving better diastereoselectivity without sacrificing stability of the DMAP complex.

N,N-Diprenyl-4-toluenesufonamide 4B1c was investigated because the N-tosyl group should stabilize the DMAP complex 4B4c by electron withdrawal and its tendency to improve crystalinity might facilitate diastereomer separation and resolution of trans-

4B3c. As in the prior examples 4B1a and 4B1b, the DMAP·BH2I conditions favored cis diastereoselectivity with 4B1c (5.8:1), but THF·BH3 provided improved 1:3.6 trans diastereoselectivity at rt. Selectivity was increased to 1:5.6 upon conducting the hydroboration at -30 °C (Table 4C, entries 5-7). The diastereomeric mixture 4B4c was separable by silica gel chromatography, and both cis and trans diastereomers were obtained in crystalline form. The structure of trans-4B4c was established by X-ray crystallography, which also confirmed that the enantiomers were not resolvable by recrystalization, as both enantiomers were contained in the unit cell of the crystals.

It is proposed that the diastereoselectivity of 4B3c is derived from boat-like transition state 4D1eq with a pseudo-equatorial isopropyl group being favored over the boat-like transition state 4D1ax, which is thermodynamically less stable due to the pseudo-axial isopropyl group (Scheme 4D). A chair-like transition state is unlikely, as one would expect a diequatorial conformation leading to cis-4B3c to dominate such a

142 transformation. Thus, a chair-like transition state correlates well with the DMAP·BH3 results.

Scheme 4D: Diastereoselective Formation of N-Tosyl-trans-2,6-diisopropyl-1,4-borazinane Ts H N H i-Pr favored B N B Ts H H H 4D1 trans-4B3c Ts eq 5.6:1 diasteroselectivity N THF·BH3 -30ºC H H i-Pr Ts 4B1c B H N N disfavored Ts H B H 4D1ax cis-4B3c

4.3 Resolution of (R,R)-2,6-Diisopropyl-1,4-borazinane Ring with Alanine

No method exists for predicting an ideal amino alcohol for crystallographic resolution of a given racemic borane species, and chromatographic separation of diastereomers is precluded by the instability of amino alcohol-borane complexes. Thus, it came as little surprise that alaninol, phenylalaninol, valinol, and ephedrine complexes of 4B4c all failed to provide convenient crystallization conditions and decomposed on silica gel. It was hypothesized that using an amino acid would provide a more stable complex from 4B4c due to the greater electron demand of the carboxyl group, a factor that might allow for chromatographic separation of the resulting diastereomeric mixture.

143

The separation of borazinane 4B3c diastereomers and resolution of racemic trans-

4B3c was investigated by Mr. John Nelson. Generating alanine complex 4E2 was complicated by the poor solubility of alanine in THF. Heating borazinane 4B3c in the presence of alanine was out of the question for fear of ruining the trans-diastereomeric excess via retro-hydroboration/hydroboration pathways. To avoid this potential complication, B-methoxy-1,4-borazinane 4E1 was generated by methanolysis of 4B3c at

-30 °C. Heating the more stable 4E1 with l-alanine at 45 °C generated complex 4E2

(Scheme 4E). Not only was chromatography viable for separating the diastereomers, but the trans-4E2 diastereomers could be separated by crystallization. The (R,R)-

Scheme 4E: Separation of Diastereomers and Resolution of 1,4-Borazinane Enantiomers Ts Ts Ts Ts 1. BH3·THF N N N N -30 °C, 20 h

2. MeOH, i-Pr B i-Pr i-Pr B i-Pr i-Pr B i-Pr -30 °C to rt OMe OMe OMe 4B1c (R,R)-4E1 (S,S)-4E1 cis-4E1 2.8 : 2.8 : 1 Ts Ts N N 1. d-Alanine, 45 ºC 1. l-Alanine, 45 ºC i-Pr B i-Pr 4E1 i-Pr B i-Pr O O H2N 2. recrystalization 2. recrystalization H2N

O O d-(S,S)-4E2 1. l-Alanine, 45 ºC l-(R,R)-4E2 2. chromatography

Ts Ts Ts N N N

i-Pr B i-Pr i-Pr B i-Pr i-Pr B i-Pr O O O H2N H2N H2N

O O O l-(R,R)-4E2 l-(S,S)-4E2 l-cis-4E2

63%

144 diastereomer (98:2 dr) was particularly easy to isolate, as it precipitated almost immediately upon dissolving the crude product in methylene chloride. Two crystalizations provided l-(R,R)-4E2 in 27% yield. Treating 4E1 with d-alanine provided the d-(S,S)-4E2 enantiomer in similar fashion. On the other hand, crystallization conditions for facile separation of the remaining mixtures have not been elucidated. Silica gel chromatography can separate the three diastereomers (63% yield) but yield must be sacrificed for separation or vice versa because 4E2 is not completely stable on silica gel. While 4E2 is stable to air upon isolation it is not as robust in solution. After several crystallizations the mixture undergoes Ts degradation. Oxidative pathways are at least partially responsible as N evidenced by the isolation of side product 4E3 and by the fact that B O MeO using deoxygenated solvents for crystallization suppresses 4E3 degradation.

4.4 Transforming l-(R,R)-4E2 into a Viable Asymmetric Hydroborating Agent

Having resolved the trans-diastereomers of alanine complex 4E2, conversion to the corresponding lithium dialkyl borohydride 4F1 was investigated with a screen of lithium hydride reagents: lithium aluminum hydride (LAH), lithium

21 trimethoxyaluminum hydride [Li(MeO)3AlH], lithium monoethoxy-aluminum hydride

9 [Li(EtO)AlH3], super hydride (LiEt3BH), and lithium tri-tert-butoxy-aluminum hydride

11 [Li(tBuO)3AlH]. The first two reagents produced a proton-coupled triplet by B NMR at

= -20 ppm, which is assigned as 4F1. However, these reductions also produced undesired boron species [= -35 ppm, -4.2 ppm (LAH); 2 to 10 ppm (Li(MeO)3AlH)].

11 Li(EtO)AlH3 and super hydride generated multiple boron species, according to B NMR

145 spectroscopy. The only reagent found to cleanly generate the desired boron species was

Li(tBuO)3AlH.

The single drawback of using Li(tBuO)3AlH for the reduction of 4E2 was that the resulting borohydride 4F1 could not be isolated from the non-boron-containing side products. Therefore, an in situ approach was taken to investigate hydroboration. Treating crude 4F1 with the hydride abstracting agents TMSCl or TMSOTf generated an unknown tetravalent species 4F2 with a 11B NMR signal at = -10 ppm. This signal is too far upfield to represent either the trivalent borazinane 4B3c or its dimer 4F3 (Scheme 4F) according to a comparison with di-t-butyl-borane (= 47 ppm), which is reported in a mixture also containing 11B signals at = -20.1, -17.2, -13.6, -.02, 24, 51.7, 83 ppm,22 and

10 23 the borolane dimer 4A5 (= 31.5 ppm) or diborane (B2H6; = 18 ppm). Isolation of complex 4B4c upon treatment of 4F2 with DMAP confirmed that species 4F2 is a

146 relatively weak Lewis base complex of the borazinane 4B3c. The Lewis base remains undefined but is probably a reduced alanine residue. Considering the tetravalent nature of 4F2, it was no surprise that hydroboration was not observed upon treating 4F2 with 1- octene.

It was proposed that potassium salts of the reduction residues would be more convenient to remove due to their poorer solubilities, thus enabling generation of free borazinane 4B3c. Soderquist‟s activated potassium hydride (KH*) was used initially to investigate the degree of substitution at boron in the reaction intermediates, but turned out to be critical to the generation of borohydride 4G1 (Scheme 4G).24,25 While precipitated

147 solids were observed throughout the KH* reduction, cannulation of the borohydride solution away from the residues and subsequent hydride abstraction with TMSCl resulted in a product that remained soluble in THF. However, the solution did not contain borazinane 4B3c , the unknown complex 4F2, or dimer 4F3 according to 11B NMR monitoring. Instead, a more downfield signal (= +11.5 ppm) of a new unknown species

4G2 was generated. Treating 4G2 with 1-octene followed by reductive workup with

KH* led to the observation of a proton-coupled doublet by 11B NMR at = -18.3 ppm (J=

79.1 Hz), which is assigned as the potassium trialkylborohydride 4G3 (Scheme 4G).

Subsequent oxidation with NaOOH and benzoylation provided octyl benzoate 4G4 in

22% yield, suggesting marginal hydroboration reactivity for the unknown species 4G2.

This marginal reactivity caused concern that the hydroboration observed might possibly be due to formation of BH3 from decomposition of 4G2. Therefore, the Type III alkene ethylidene-cyclohexane was treated using identical hydroboration conditions to probe for any enantioselectivity, which would confirm a chiral boron environment being involved in the transformation (Scheme 4G). Reductive workup with KH* provided a proton-coupled 11B doublet at = -18 ppm (J= 81.3 Hz), which is assigned as potassium trialkylborohydride 4G5. Subsequent oxidation with NaOOH followed by acylation gave the acetate 4G6 in low yield, but 19% ee was observed by GC assay on a chiral support. In addition to the observation of trialkylborohydrides 4G3 and 4G5 from the reductive workup of the hydroboration of 1-octene and ethylidenecyclohexane, respectively, this enantioinduction provided further evidence that the borazinane ring remains intact and participates in hydroboration. However, the tetravalent 11B chemical

148 shift and marginal reactivity of 4G2 indicate that an unidentified Lewis base „Y‟ interferes with reactivity by inhibiting dissociation to 4B3c.

The Lewis base reduction byproducts from the generation of lithium borohydride

4F1 or potassium borohydride 4G1 had to be addressed in order to accurately evaluate the true potential of borazinane 4B3c. Solvent rinsing and crystallization attempts were made to remove the undesired species but all were unsuccessful. A Lewis acid screen including BF3·OEt2, TMSOTf, Zn(OTf)2, Cu(OTf)2, and LiOTf was done expecting to achieve competitive complexation to the undesired Lewis base contaminants, thus freeing trivalent 4B3c and/or dimer 4F3. However, no dominant boron species resembling 4B3c or dimer 4F3 was observed by NMR in any case. A Lewis acid screen including

BF3·OEt2, TMSOTf, Cu(OTf)2, Sc(OTf)3, AgOTf, (C6F5)3B, and MgBr2 was undertaken to investigate the possibility of competitive complexation of DMAP from complex 3B4c, but this was also unsuccessful. It became evident that removing alanine from diastereo- enriched alanine complex l-(R,R)-4E2 prior to reduction to the borohydride would be the most efficient way of generating pure potassium borohydride (R,R)-4G1.

Masamune generated pure borohydride 4A2 from prolinol complex 4A3

(Equation 2) by methanolysis followed by distillation and reduction of the resulting methoxyborolane.11 However, a variety of methanolysis conditions were unsuccessful in cleanly converting alanine complex 4E2 to B-methoxy-borazinane4 E1. On the other hand, hydrolysis conditions reported by Burke et al. for unmasking boronic acids from

N-methyliminodiacetic acid complexes26 were used with 4E2 to generate a single 11B

NMR signal at = +53 ppm (Scheme 4H). This shift corresponds to methyl borinates

(45 ppm <  < 54 ppm),27 indicating the formation of the desired borinic acid (R,R)-4H1.

149

Scheme 4H: Alanine-Free Hydroboration Ts Ts N Ts 1. aq. NaOH N N 2. Na3PO4 KH* i-Pr B i-Pr H N O i-Pr B i-Pr i-Pr B i-Pr 2 H H OH (2) O alanine K (R,R)-4E2 (R,R)-4H1 (R,R)-4G1 11B  = +53 ppm Ts Ts N TMSCl N

THF i-Pr B i-Pr i-Pr B i-Pr H H O H K (R,R)-4G1 (R,R)-4H2 11B = +17.9 ppm TMSCl, Et2O Ts Ts N N heterogeneous mixture; 1-octene 1-octene no 11B NMR signals i-Pr B i-Pr i-Pr B i-Pr

nC8H17 H 4H3 (R,R)-4B3c 11 B = +87 ppm (in Et2O)

NaOOH Ts Ts N N H + nC8H18OH + i-Pr i-Pr i-Pr OH OH OH 4B6c 4H4 4H5

The corresponding borinic anhydride was not observed by mass spectrometry, but the possibility that it is present in the crude product has not been eliminated. No distillation was needed before reduction because 1H NMR spectroscopy showed that no trace of alanine remained after aqueous extraction. Soderquist‟s KH* reduced (R,R)-4H1 into

(R,R)-4G1 very cleanly in THF according to 11B NMR spectroscopy. Addition of distilled hexanes to a concentrated THF solution of (R,R)-4G1 led to its precipitation as a

150 white sludge. Placing the ether-rinsed sludge under high vacuum formed a bubble of amorphous solid, which disintegrated into a white powder upon agitation.

Treating a mixture of the isolated borohydride powder 4G1 in Et2O with TMSCl resulted in a heterogeneous mixture with no soluble boron species observed by NMR.

However, upon addition of 1-octene, a trivalent 11B signal was observed at = +87 ppm, which is assigned as B-octyl-borazinane 4H3 (Scheme 4H). Treating a THF solution of

4G1 with TMSCl provided a boron species with a clean 11B signal at = +17.9 ppm,

23 which is similar to the chemical shift of the bridged dimer B2H6 (= +18 ppm), but is too far upfield relative to the borolane dimer 4A5 (= 31.5 ppm)10 and other dialkylborane dimers (= 23-27 ppm)27 to be assigned as a H-bridged dimer.

Calculations done by Mr. Aleksanders Prokofjevs predict a 11B NMR shift of 27 ± 3 ppm for dimer 4F3. The possibility of a hydride-bridged [R2(H)B-H-

B(H)R2] species was eliminated based on the upfield shifts of NaB2H7 K H B H B H (= -25.3 ppm),27 the intramolecularly bridged species 4H6 (-4.9

28 29 ppm), and Li(Et3B)2H (= +8.7 ppm). 4H6

Treating potassium borohydride (R,R)-4G1 with a solution of TMSCl in CH2Cl2 provided shifts of = +25 and +71 ppm by 11B NMR spectroscopy, which are respectively assigned as the dimer 4F3 and the monomer (R,R)-4B3c. The observation of two species in CH2Cl2 and a single (third) species in THF lead to the conclusion that the

11B signal at +17.9 ppm in THF is the THF·borazinane complex (R,R)-4H2. This chemical shift value is downfield compared to the THF complex of 9-BBN (= +14 ppm),30 but 4H2 may be somewhat deshielded by the N-Ts subunit compared to the

9-BBN environment.

151

Addition of 1-octene to a solution of 4H2 resulted in the observation of a 11B

NMR signal at = +86.5 ppm (Scheme 4H). Oxidative workup of 1-octene

13 hydroborations in both Et2O and THF reactions provided 1-octanol, diol (S,S)-4B6c (85-

91%), and the alcohol byproduct S-4H5 (<5%).14 For unknown reasons, complex 4H2 was not observed by 11B NMR upon treating the crude KH* reduction reaction mixture with TMSCl. Thus, precipitating (R,R)-4G1 from THF-hexane in its powdered form is absolutely crucial for achieving hydroboration in THF.

Developing conditions for hydroboration enabled comparisons between the presumed source of 4B3c and Masamune‟s borolane 4A1 in terms of substrate scope and enantioselectivity. Initial screening of the four alkene types was done specifically using substrates that had been tested by Masamune et al.3 Ether was used as the solvent due to the volatility and low molecular weight of some of the alcohol products. At room temperature enantioselectivity with 4B3c was good for the Type II alkene cis-3-hexene

(86.4% ee - Table 4I; entry 2), and moderate for Type IV alkene ethylidenecyclohexane

(44.6% ee – entry 9). By comparison, the Masasmune borolane 4A1 provides 99.9% ee

3 with cis-3-hexene and 99.3% ee with ethylidenecyclohexane while Brown‟s ipc2BH provides 93% ee with cis-3-hexene5 and ca. 20% ee with Type IV alkenes.3 However, enantioselectivity with 4B3c was poor for the type I and type III alkenes 2-methyl-1- butene (<5% ee – entry 1) and trans-3-hexene (3.8% ee – entry 7), respectively. By comparison, Masamune borolane 4A1 provides 1.5% ee and 99.5% ee for the respective

3 4 substrates while Brown‟s Ipc2BH provides 21% ee with 2-methyl-1-butene and ca. 13% ee for Type III alkenes.3

152

Ensuing optimization studies used the type II alkene cis-1,4-diphenyl-2-butene due to advantages in assay and product recovery, and demonstrated that Et2O promotes marginally better enantioselectivity (86.4% ee) than THF (85.2 % ee), toluene (84% ee), and CH2Cl2 (79.2% ee) in the generation of R-1,4-diphenyl-2-butanol R-4I2 (Table 4I; entries 2-3, 5-6). In an attempt to improve enantioselectivity, the reaction was conducted at -20 °C in THF, a solvent that ensures reagent solubility at the lower temperature.

Instead of the expected outcome, a slight decrease in enantioselectivity was observed

(80.2% ee; entry 4). With solvent and temperature optimizations having no substantial effect, the enantioselectivity trends of borazinane 4B3c remain more reminiscent of

Brown‟s Ipc2BH than of the Masamune borolane 4A1.

Like ipc2BH, borazinane 4B3c works well specifically with type II alkenes, which is attributed to uncanny structural similarity for the transition states. Brown proposed transition state conformation 4J1/4J2 for ipc2BH and Houk provided support for this geometry with theoretical studies.3,31 Borazinane (R,R)-4B3c fits into this model beautifully, assuming a chair conformation 4J3. According to Brown‟s model, transition state 4J4 would form upon approach of cis-1,4-diphenyl-2-butene (Scheme 4J).

Subsequent hydroboration and oxidation would provide (R)-4I2, which is the enantiomer obtained in 86% ee. On the other hand, there is crystallographic evidence that both the alanine complex l-(R,R)-4E2 and the DMAP complex 4B4c exist in a twist-boat boracycle conformation in the solid state. Thus it is possible that non-bonded interactions in (R,R)-4B3c may cause it to adopt more of a twist boat conformation 4J7. The reduced versatility and enantioselectivity of 4B3c compared to 4A1 is consistent with

Masamune's proposal that monomeric borolane 4A1 prefers a trans-diaxial envelope

153

Entry Alkene Type Solvent Product ee (%) b 1 I ether 4I1 < 5 c f,g 2 Ph Ph II ether 4I2 86.4 c g,h 3 II THF 4I2 85.2 c g,i 4 II THF 4I2 80.2 c g 5 II CH2Cl2 4I2 79.2 d g 6 II Toluene 4I2 84 d 7 II ether 4I3 84.6 d 8 III ether 4I3 3.8 9e IV ether 4I4 44.6j

(a) Procedure: To a stirred mixture of 23 at rt in solvent was added alkene (4 equiv) followed by TMSCl (1 equiv). After stirring 20 h, oxidation with NaOOH at 0 °C gave alcohol products, which were purified by chromatography prior to derivatization and/or assay. (b) Assayed by GC. (c) Assayed by HPLC. (d) Assay after conversion to TMS ether. (e) Assayed after acetylation. (f) R-4I2 recovered in 72% yield. (g) R-enantio-selectivity. (h) R-4I2 recovered in 61% yield. (i) - 20 °C. (j) S-enantioselectivity

conformation in solution,11 a conformation that cannot be adopted by the isopropyl groups of 4B3c. Transition state 4J8 illustrates that a (pseudo)diaxial relationship can be achieve by the protons on C-2 and C-6, presenting the possibility that the reduced enantioselectivity of 4B3c is due to less bulky diaxial substituents adjacent to boron compared with 4A1. 1H NMR decoupling experiments with DMAP complex (R,R)-4B4c in THF have revealed two coupling constants of J=10.8 Hz and 11.2 Hz between the

154 methine and methylene protons of the ring, providing evidence against borazinane

(R,R)-4B3c existing in chair conformation 4J3 in solution.

The overall generation of borazinane 4B3c, already simplified by diastereoselective ring assembly and stability of its alanine complex, was streamlined further upon discovery that a one pot in situ reduction-hydroboration sequence was viable in Et2O. A heterogeneous mixture of l-alanine complex l-(R,R)-4E2 and KH* (4 equiv) was stirred in Et2O before addition of cis-1,4-diphenyl-2-butene and TMSCl (3 equiv).

Subsequent oxidative workup provided (R)-4I2 in 37% yield and 86% ee (Equation 5).

The low yield is attributed to the heterogeneity of the reduction step causing poor conversion of (R,R)-4E2 to (R,R)-4G1 although this cannot be confirmed due to neither species being soluble in Et2O. Diol (S,S)-4B6c was recovered in 95% yield, which, in addition to the enantioselective formation of (R)-4I2, indicates that the borazinane ring is not compromised throughout the reaction. Soderquist‟s KH* proved essential once again,16 despite its insolubility, as attempted borane generation via a one-pot procedure

155 with ether-soluble Li(tBuO)3AlH failed to give alcohol (R)-4I2 after oxidative workup.

The yield provides room for improvement, but this development demonstrates that a one pot procedure can replace three sensitive steps in the Masamune protocol (Equation 1) if one uses an amino acid rather than an amino alcohol for chiral borane resolution and diastereomer separation.

4.5 Other Applications for the N-Tosyl-(R,R)-2,6-diisopropyl-1,4-borazacycle

Borazinane (R,R)-4B3c was also tested to see if it could achieve asymmetric reduction of ketones, as the Masamune borolane

4A1 is also an effective reagent for the asymmetric reduction of ketones in the presence of its corresponding mesylateborane (R,R)-4J9.

Masamune has reported that treating a 4:1 mixture of borolane 4A1 and mesylate (R,R)-

4J9 in pentane with 1-phenyl-2-propanone provided (R)-1-phenyl-2-propanol (R)-4K2 in

98.9% ee.10,32 The Lewis-acid 4J9 is necessary for stereoinduction, as <5% ee is achieved without it. No reduction product 4K2 was observed upon treating a pentane suspension of borohydride (R,R)-23 with 1.2 equiv of MsOH followed by 1-phenyl-2- propanone, but switching the solvent to toluene provided (R)-4K2 in 35% ee (Scheme

4K). This result suggests that (R,R)-4K1 was generated, which raised interest in utilizing

156 the chiral boracycle of (R,R)-4B3c as a Lewis acid. This led to an investigation of generating a chiral borenium species for Diels-Alder catalysis.

Scheme 4K: Asymmetric Reduction with 4B3c and Corresponding Mesylate Ts Ts Ts MsOH 1. N N N (1.2 equiv) O -20 °C OH i-Pr B i-Pr solvent, rt i-Pr B i-Pr i-Pr B i-Pr H H 2. DMAP H OMs K (R,R)-4G1 (R,R)-4B3c (R,R)-4K1 4K2 (0.8 equiv) (0.2 equiv) Pentane: Not observed Toluene: 35% ee

There are several reports of borenium species acting as chiral catalysts for Diels-

Alder chemistry33-38 but none of them are derived from DMAP-borane complexes.

DMAP complexes of several boranes of varying degrees of substitution were treated with

MsOH and monitored by 11B NMR in order to investigate if the species would be tri- or tetravalent in DCM. Treating DMAP·BH3 4L1, DMAP·BH2Ipc 4L3, DMAP·BH(chex)2

4L5, and borazinane complex 4B4c with MsOH generated 11B shifts of = +0.6 ppm,

= +6.6 ppm, = +11.7 ppm, = +9.7 ppm, respectively. These shifts are indicative of a tetravalent species believed to be 4L2, 4L4, 4L6,and 4L7 (Scheme 4L). Increasing the steric bulk at boron generates a more downfield 11B signal, indicating a more trivalent nature, which correlates to the mesylate interaction with boron being weakened by the alkyl groups. In order to pursue a more reactive species, complex 4B4c was treated with triflic acid to introduce a weaker complexing anion. This generated a 11B NMR signal at

+15.8 ppm, which has been assigned as 4L8. Treating a -78 °C solution of 4L8 (7 mol%) in DCM with methacrolein (1 equiv) followed by cyclopentadiene (5 equiv) generated cycloadduct 4L9 with 97:3 endo:exo diastereoselectivity and 41% ee in 64% yield. It is interesting that 4L8 catalyzes the reaction at all considering its tetravalent nature. The

157 enantioselectivity and yield suggests that the borazinane not only remains intact under the reaction conditions but turns over as a catalyst.

158

4.6 Summary

The synthesis of N-tosyl-(R,R)-2,6-diisopropyl-1,4-borazinane 4B3c incorporates diastereoselective assembly of the boracycle via cyclic of diene 4B1c, resolution of dialkylborane diastereomers via alanine complexation rather than amino alcohol complexation, and the ability to achieve hydroboration from the amino acid complex in a single pot process. Despite its synthetic advantages, the hydroboration and reduction results with 4B3c stand as the latest affirmation of how impressive the Masamune borolane 4A1 is in terms of its synthetic versatility. These studies have also revealed that

DMAP complexes of chiral dialkylboranes, while not useful for resolution/separation of diastereomeric borane mixtures, can be stable precursors to borenium catalysts such as

4L8 for asymmetric Diels-Alder cycloadditions.

This work represents the first report of a six-membered boracycle with a

C2-symmetric 2,6-dialkylborane environment and it shows promise in multiple applications. However, if the asymmetric induction of these borazinane species is representative for six-membered boracycles, then future efforts toward C2-symmetric asymmetric boracycles should focus on 5-membered rings rather than 6-membered rings.

159

Experimental

Compounds 4B1, 4B4, 4B5, 4B6, 4B1c, 4E1, 4E2, and 4E3 were prepared and characterized in the unpublished work of Mr. John Nelson.

In Situ Generation of borohydride 4F1.

Ts N

i-Pr B i-Pr H H Li 4F1

A sample of trans-4E2 (0.0963 g, 0.24 mmol) was taken up in distilled, degassed THF (2 mL) under nitrogen and cooled to 0 °C. A solution of Li(tBuO)3AlH (0.246 g, 0.97 mmol) in THF (2 mL) was cannulated dropwise into the cooled solution of 4E2. THF

(0.5 mL) was used to complete the transfer of Li(tBuO)3AlH. The reaction was allowed to stir for 1 h at rt before an NMR sample was cannulated into an NMR tube. 11B NMR spectroscopy displayed a signal at = -20 ppm (t, JBH= 78.9 Hz) which is assigned as

4F1.

Generating the unknown borazinane complex 4F2 and DMAP complex 4B4c

Ts Ts N N

i-Pr B i-Pr i-Pr i-Pr X H B H DMAP Li 4F2 trans-4B4c

A sample of trans-4E2 (0.1025 g, 0.25 mmol) was taken up in distilled, degassed THF (2 mL) under nitrogen and cooled to 0 °C. A solution of Li(tBuO)3AlH (0.256 g, 1.0 mmol) in THF (2 mL) was cannulated dropwise into the cooled solution of 4E2. THF (0.5 mL)

160 was used to complete the transfer of Li(tBuO)3AlH. The reaction was allowed to stir for

1 h at rt before distilled, degassed TMSCl (115 L, 0.9 mmol) was added dropwise to the reaction. The reaction stirred for 15 minutes before an NMR sample was cannulated into an NMR tube. 11B NMR spectroscopy displayed a broad signal at = -10 ppm with no apparent B-H coupling, which is assigned as 4F2. DMAP (0.22 g, 1.87 mmol) was dissolved in THF (4 mL). The resulting solution was added to the NMR sample (0.5 mL) and the original reaction mixture (3.5 mL). The resulting mixtures were combined and concentrated. Purification by column chromatography in 2:3 EtOAc:hexanes eluted

0.086g of 4B4c (77%), identified by comparing 1H NMR spectra with authentic material.

In Situ hydroboration of 1-octene via borohydride trans-4G1

Ts Ts Ts Ts N N N N

i-Pr B i-Pr i-Pr B i-Pr i-Pr B i-Pr i-Pr B i-Pr H H Y H H H Me nC8H17 c-hex K K K K trans-4G1 4G2 4G3 4G5

KH in mineral oil *(0.21 g) was activated as described by Soderquist.25 A sample of trans-4E2 (0.10, 0.26 mmol) was taken up in distilled THF (3mL). The resulting solution was cannulated into the KH* and stirred for 1 h. The resulting mixture was cannulated into a culture tube under nitrogen and centrifuged. The supernatant solution was cannulated into a re-sealable flask and degassed via freeze-pump-thaw. In a glove box, an aliquot was transferred to an NMR tube and treated with TMSOTf. Monitoring by 11B

NMR spectroscopy lead to the observation of a dominant signal at = +11.5 ppm, which is an unknown species 4G2. 1-octene (70 L, 0.44 mmol) was added to the reaction, followed by TMSCl (40 L, 0.31 mmol). A second batch of KH* was generated from

161

KH in mineral oil (0.20 g). The hydroboration mixture was cannulated onto the second batch of KH* under nitrogen and stirred for 30 min. An aliquot was cannulated into an

11 oven dried, N2-flushed, septumed NMR tube. B NMR spectroscopy displayed a signal at = -18.3 ppm (d, JBH= 81.3 Hz), which is assigned as 4G3. The aliquot was combined with the reaction mixture, then cannulated into a culture tube and centrifuged. The supernantant solution was cannulated into a clean flask and treated with TMSCl (40 mL,

0.31 mmol) followed by premixed satd NaOH in MeOH (0.9 mL) and 35% H2O2 (0.9 mL). The reaction mixture was transferred to a separatory funnel containing brine and extracted with Et2O. The organic layers were combined, dried over MgSO4, and concentrated (aspirator). The crude reaction mixture was taken up in CH2Cl2 (3 mL) and treated with Bz2O (0.25 g, 1.11 mmol), Et3N (0.3 mL, 2.15 mmol), and DMAP (0.07 g,

0.58 mmol). The benzoylation was quenched with water and transferred to a separatory funnel. Ether (10 mL) was added and the organic layer was rinsed with 1 N HCl and satd

NaHCO3. Each water layer was back extracted with hexanes. The organic layers were combined, dried over MgSO4, and concentrated (aspirator), and purified via silica gel

39 chromatography using 10% Et2O in hexanes to give 0.0131g of 4G4 (22%). The same

40 procedure was used to generate 4G5 [= -18 ppm (JBH= 81.3 Hz)] and 4G6 was generated using analogous hydroboration / reduction / oxidation conditions followed by acylation. The enantioselectivity of 4G6 was assayed by GC on a Chrompack Chirasil

Dex (25m x 0.32 mm x 0.25 mm) column (90 °C, 1 mL/min)40: (S)-11.3 min , (R)-15.4 min.

162

One-pot procedure for hydroboration with l-(R,R)-4E2 Ts N In a glove box, a sample of l-(R,R)-4E2 (0.10 g, 0.25 mmol) was

i-Pr B i-Pr added to a flask containing KH*(0.04 g, 1.05 mmol) prepared as H2N O before. This mixture was suspended in Et2O (10 mL) and stirred O (R,R)-4E2 vigorously for 20 h at rt before addition of cis-1,4-diphenyl-2- butene (100 L, 0.48 mmol) followed by TMSCl (100 L, 0.78 mmol). The resulting mixture was stirred for 20 h before cooling to -78 °C for addition of premixed 20%

NaOH (2 mL) and 35% H2O2 (1 mL). The mixture was warmed to rt and stirred for 8 h before transferring it to a separatory funnel containing brine (5 mL). The aqueous layer was extracted with Et2O (4 x 20 mL). The organic layers were combined, dried over

MgSO4, and concentrated (aspirator), and purified via silica gel chromatography using

10%, 15%, and 30% EtOAc : hexanes to give 0.022 g of alcohol 4I2 (37%),40 0.082 g of diol 4B6c (95%), and 0.003 g of the elimination side product 4H5 (3%). The enantiomeric excess of 4I2 was assayed as described later in data for Table 4I.

1 Compound 4H5: ESMS m/z: 258.4 (M+1). H NMR (400 MHz, CDCl3): 0.87 (d, J=

6.8 Hz, 3H, CH3), 0.91(d, J= 6.8 Hz, 3H, CH3), 1.65 (m, 1H, CH), 1.92 (br d, J= 4.4 Hz,

1H, OH), 2.43 (s, 3H, CH3), 2.81 (ddd, J= 4.4 Hz, 8.3 Hz, 12 Hz, 1H, CH2N), 3.13 (ddd,

J= 3.2 Hz, 8 Hz, 12 Hz, 1H, CH2N), 3.4 (m, 1H, CH-O), 4.88 (br s, 1H, NH), 7.32 (dd,

J= 8.8 Hz, 0.8 Hz, 2H, Ar H), 7.75 (d, J= 8.8 Hz, 2H, Ar H). 13C NMR

(CDCl3z,) 143.5, 136.7, 128.8, 75.4, 46.6, 31.9, 21.5, 18.5, 17.8.

163

Generating alanine free (R,R)-4G1

A sample of l-(R,R)-4E2 (1.02 g, 2.5 mmol) was converted to (R,R)-4H1 using the method of Burke et al.26 Thus, a solution of 4E2 in THF (3 mL) was treated with 1M

NaOH (0.8 mL) and stirred vigorously open to the air for 10 min. The reaction was then treated with 0.5 M aq Na3PO4 (3 mL) and Et2O (3 mL) and stirred for 30 min, adding distilled H2O (~10 mL) to dissolve the white precipitate that formed. The reaction mixture was poured into a separatory funnel and the organic layer collected. The water layer was extracted with Et2O (1x 20 mL). Some extra water had to be added once more to dissolve white precipitates upon addition of Et2O. The organic layers were combined, dried over MgSO4, filtered, and concentrated (aspirator followed by High-Vac for 30 min). 11B NMR spectrospocopy indicated a single boron species at = +53 ppm, which is tentatively assigned as borinic acid 4H1. Based on ESMS m/z = 259 amu for M-1+Na

(Calc‟d 260 amu). No m/z for the corresponding borinic anhydride or borinic acid was observed but this does not eliminate the possiblility that borinic anhydride was present and was hydrolytically cleaved under the ESMS conditions (H2O, MeOH, NaCl). The crude product was taken up in distilled THF (20 mL) and cannulated into a flask containing KH* (0.51 g, 12 mmol) under nitrogen at rt. The transfer was completed with

THF (5mL) and the reaction stirred for one hour at rt. The resulting mixture was transferred to a culture tube and centrifuged. The supernatant liquid was cannulated into a RB flask containing a magnetic stir bar and was blown down with N2 until ~3 mL of solution remained. This solution was stirred vigorously while hexane (2 mL) was added dropwise. A white powder-like solid started to precipitate. As more hexanes (13 mL) was added the powder congealed into a gum-like mass as the mixture was stirred. The

164 solvent was decanted via syringe, and the residual gum was then rinsed with hexanes (20 mL) and Et2O (20 mL). Residual solvent was blown off with N2 and high vacuum was applied, causing the gel to form an amorphous bubble, which disintegrated into 0.53 g of powder (59%) upon agitation. This (R,R)-4G1 powder was stored and dispensed in a glove box. (R,R)-4G1: 1H NMR (500 MHz, THF-D8): 0.31 (br s, 2H, B-CH), 0.71(q,

JBH= 77 Hz, 2H, BH2), 1.00 (d, J= 6.5 Hz, 6H), 1.02 (d, J= 6.5 Hz, 6H), 1.59 (m, 2H, -

CHMe2), 2.49 (s, 3H, CH3), 2.92 (br s, 2H, -CH2N), 3.1 (br s, 2H, -CH2N), 7.37 (d, J= 8

Hz, 2H, Ar H), 7.69 (d, J= 8 Hz, 2H, Ar H).

Establishing Hydroboration yield by NMR internal standard

In the glove box, a sample of (R,R)-4G1 (0.1007 g, 0.28 mmol) was suspeneded in Et2O

(8 mL) under N2 and stirred vigorously at rt as cis-1,4-diphenyl-2-butene (100 L, 0.48 mmol) was added followed by TMSCl (35 L, 0.27 mmol). The mixture was stirred for

20 h, removed from the glovebox, and cannulated into a 0 °C mixture of 20% NaOH (2 mL), 35% H2O2 (1.1 mL), and 1:1 MeOH:Et2O (3 mL). The mixture was warmed to rt and stirred for 8 h before transferring it to a separatory funnel containing brine (5 mL).

The aqueous layer was extracted with Et2O (4 x 20 mL) and the organic layers were combined, dried over MgSO4, and concentrated (aspirator). Ph3CH (0.205 g, 0.840 mmol) was added to the crude product mixture and an NMR spectrum was taken in

CDCl3. The methine proton of Ph3CH (= 5.72 ppm) integrated to 3.91, the methyl group of diol 4B6c (= 2.54 ppm) integrated to 3.00, and the methylene group of alcohol

4I2 (= 2.75 ppm) integrated to 1.88. This indicates that 0.215 mmol of 4B6c and 0.202 mmol of 4I2 (94% relative to 4B6c) were generated. Purification via silica gel

165 chromatography using 10%, 15%, and 30% EtOAc : hexanes gave 0.077 g 4B6c (0.223 mmol), and 0.046 g of 4I2 (0.204 mmol, 91.5% relative to 4B6c).

Generation of 4H2 and 4H3

Ts Ts N N

i-Pr B i-Pr i-Pr B i-Pr O H nC8H17 d8 (R,R)-4H2 4H3

In a glovebox, a solution of (R,R)-4G1 (0.057 g, mmol) in THF (3 mL) was treated with

TMSCl (20 L, 0.15 mmol). An aliquot was transferred to a dry NMR tube and 11B spectroscopy revealed a single signal at = +17.9 ppm. The species assigned as 4H2 also appears to be present by 1H NMR spectroscopy (500 MHz, THF-D8): 0.97-1.03 (m,

2H, B-CH), 1.03-1.09 (m, 12H, CH3), 2.02-2.13 (m, 2H, -CHMe2), 2.52 (s, 3H, CH3)

3.10-3.29 (m, 4H, -CH2N), 7.45-7.49 (m, 2H, Ar H), 7.75-7.79 (m, 2H, Ar H). Treating the solution with 1-octene led to observation of a signal at +86.5 ppm, which has been assigned as 4H3.

166

Hydroboration with 4F3 and 4B3 in Methylene Chloride

Ts N Ts N i-Pr B i-Pr H H i-Pr B i-Pr i-Pr B i-Pr H N (R,R)-4B3c Ts 4F3

In the glove box, a sample of (R,R)-4G1 (0.072 g , 0.2 mmol) in a dry culture tube was treated with a solution of TMSCl (26 L, 0.2 mmol) in CD2Cl2 (5 mL). The resulting mixture was centrifuged, a precipitate-free aliquot was transferred to a dry NMR tube, and 11B spectroscopy revealed a signals at = +71 ppm, +52 ppm, and +25 ppm in a ratio of ca. 3:1:2. The peak at +52 ppm is attributed to partial oxidation of the sample, the peak at +71 ppm is assigned as the monomer 4B3c and the peak at +25 ppm is assigned

1 as the dimer 4F3. The H NMR spectrum in of this mixture in CD2Cl2 is clean, but a

1 mixture is evident. H NMR (400 MHz, CD2Cl2): 0.64 (qd, JBH= 34.8 Hz, J= 6.8Hz,

1H, B-H), 0.72-.96 (m, 12H, CH3), 1.72-1.78 (m, 2H, -CHMe2), 2.37 (s, 3H, CH3) 2.78-

3.03 (m, 2H, -CH2N), 3.03-3.16 (m, 2H, -CH2N), 3.63(m, 2H, B-CH), 7.20-7.32 (m, 2H,

Ar H), 7.48-7.61 (m, 2H, Ar H). The NMR sample was transferred back to the original mixture, which was then treated with cis-1,4-diphenyl-2-butene (100 L, 0.48 mmol) and stirred for 20 h, removed from the glovebox, and cannulated into a 0 °C mixture of 20%

NaOH (2 mL), 35% H2O2 (1.1 mL), and THF (3 mL). The mixture was warmed to rt and stirred vigorously for 16 h before transferring it to a separatory funnel containing brine (5 mL). The aqueous layer was extracted with Et2O (3 x 20 mL). The organic layers were

167 combined, dried over MgSO4, concentrated (aspirator), and purified via silica gel chromatography using 10%, 15%, and 30% EtOAc : hexanes to give 0.021 g of alcohol

4I2 (49%), 0.049 g of diol 4B6c (72%), and 0.003 g of the fragmentation side product

4H5 (4%).

Hydroboration procedure starting with (R,R)-4G1 powder (Table 4I)

In the glove box, a sample of (R,R)-4G1 (0.103 g, 0.28 mmol) was suspeneded in Et2O (8 mL) at rt and stirred vigorously as 1-octene (100 L, 0.62 mmol) was added followed by

TMSCl (35 L, 0.27 mmol) under nitrogen. The mixture was stirred for 20 h, and an aliquot was removed. 11B NMR spectroscopy showed a signal at = +87 ppm (br s), which is assigned as 4H3. The reaction mixture was removed from the glovebox and cannulated into a 0 °C mixture of 20% NaOH (2 mL), 35% H2O2 (1.1 mL), and 1:1

MeOH:Et2O (3 mL). The mixture was warmed to room temp and stirred for 8 h before transferring it to a separatory funnel containing brine (5 mL). The aqueous layer was extracted with Et2O (4 x 20 mL) and the organic layers were combined, dried over

MgSO4, concentrated (aspirator), and purified via silica gel chromatography using 40%

Et2O : pentane to give 0.021 g of octanol (57%), 0.06 g of diol 4B6c (62%), and 0.003 g of the fragmentation side product 4H5 (4%).

Table 4I Data

Each substrate in Table 4I was hydroborated using analogous hydroboration / and oxidation conditions. The enantioselectivity of terminal alcohol 4I1 was assayed by GC on a Chrompack Chirasil Dex (25m x 0.32 mm x 0.25 mm) column (90 °C, 1 mL/min):

9.5 min, 9.7 min; as was the trimethylsilylether of 4I341(90 °C, 1 mL/min): (R)-13.8 min

(minor), (S)-14.45 min (major); and the acetate of 4I4 (90 °C, 1 mL/min)40: (S)-11.3 min

168

(major) , (R)-15.4 min (minor). The ee of 4I2 was assayed by HPLC on a Chiralcel-OD column (5% IPA: Hexanes, 1 mL/min): (S)-13 min (minor), (R)-19.6 min (major).42 The assignment of stereochemistry in table 4I is confirmed by the work of Blakemore42 and

Stampfer.40

Entry Alkene Type Solvent Product ee (%) b 1 I ether 4I1 < 5 c f,g 2 Ph Ph II ether 4I2 86.4 c g,h 3 II THF 4I2 85.2 c g,i 4 II THF 4I2 80.2 c g 5 II CH2Cl2 4I2 79.2 d g 6 II Toluene 4I2 84 d 7 II ether 4I3 84.6 d 8 III ether 4I3 3.8 9e IV ether 4I4 44.6j

(a) Procedure: To a stirred mixture of 23 at rt in solvent was added alkene (4 equiv) followed by TMSCl (1 equiv). After stirring 20 h, oxidation with NaOOH at 0 °C gave alcohol products, which were purified by chromatography prior to derivatization and/or assay. (b) Assayed by GC. (c) Assayed by HPLC. (d) Assay after conversion to TMS ether. (e) Assayed after acetylation. (f) R-4I2 recovered in 72% yield. (g) R-enantio-selectivity. (h) R-4I2 recovered in 61% yield. (i) - 20 °C. (j) S-enantioselectivity

169

Asymmetric reduction of 1-phenyl-2-propanone

A sample of (R,R)-4G1 (0.099 g, 0.27 mmol) was suspended in toluene (3 mL) under nitrogen, treated with MsOH (22 L, 0.34 mmol), and stirred vigorously for 2 h at rt.

The reaction was cooled to -40 °C before 1-phenyl-2-propanone was added. The reaction was stirred at -20 °C for 48 h before quenching with a solution of DMAP (0.0361g, 0.3 mmol) in toluene (2 mL). The reaction was concentrated and purified by silica gel chromatography using 40% Et2O in pentane to give 1-phenyl-2-propanol 0.002 g 4K2

(< 10%). The ee of 4K2 was assayed by HPLC with a Chiralcel OD-H column (1.5%

IPA: hexane, 0.5 mL/min): (S)-enantiomer:16.6 min (major); (R)-enantiomer:17.8 min

(minor)43 to provide 35% ee.

Generating Mesylate Complexes of DMAP Borenium Cations

Ts N H H B MsO MsO DMAP B B DMAP H MsO DMAP i-Pr B i-Pr MsO DMAP 4L2 4L4 4L6 4L7

Complexes 4L2, 4L4, 4L6, and 4L7 were generated in situ by treating DCM solutions of their DMAP-borane precursors with methanesulfonic acid in a glovebox. 4B4c (0.02 g,

0.2 mmol) in 1 mL of DCM was treated with MsOH (7 L, 0.1 mmol) to generate a new species observed by 11B NMR spectroscopy at = +9.7 ppm, which is assigned as tetravalent 4L7. Starting material was also evident by 11B NMR spectroscopy (~1:1).

4L318 ( 0.04 g, 0.15 mmol) in 1 mL DCM was treated with MsOH (7 L, 0.15 mmol) to generate a new species observed by 11B NMR spectroscopy at = +6.6 ppm, which is assigned as tetravalent 4L4. 4L118 ( 0.02 g, 0.16 mmol) in 1 mL DCM was treated with

170

MsOH (10 L, 0.15 mmol) to generate a new species observed by 11B NMR spectroscopy at = +0.6 ppm, which is assigned as tetravalent 4L2. Dicylcohexylborane was generated as previously reported,44 and was quenched with an equivalent of DMAP and concentrated (aspirator, then high vacuum) to give crude 4L5 (11B = +2.5 ppm).

Treating a solution of crude 4L5 (0.046 g, 0.015 mmol) with MsOH (10 L, 0.15 mmol) generated a new species observed by 11B NMR spectroscopy at = +11.6 ppm, which is assigned as tetravalent 4L6.

Asymmetric Diels-Alder catalysis with 4L8

A sample of (R,R)-4B4c (0.084 g, 0.19 mmol) was taken up in Ts CH2Cl2 (8 mL). TfOH (10L, mmol) was added, causing gas N

11 evolution. B NMR spectroscopy enabled observation of a boron i-Pr B i-Pr TfO N species at = +15.8, which is assigned as 4L8, and an unknown species at = -14 ppm. The reaction was stirred for 10 minutes N 4L8 before cooling to -95 °C and adding methacrolein (0.25 mL, 3.03 mmol) followed by cyclopentadiene (1.22 mL, 14.9 mmol).33 The reaction stirred at -78

°C for 10 h before quenching with a solution of DMAP (0.10 g, 0.83 mmol). The solution was concentrated and purified by silica gel chromatography using 2% Et2O in hexanes to give 0.3 g of 4L9 (64%). Enantioselectivity was assayed with by chiral GC using Chrompack Chirasil Dex (25 m x 0.32 mm x 0.25 mm) column (1.3 mL/min, 100

°C) after NaBH4 reduction to the corresponding alcohol 2-methyl-bicyclo[2.2.1]hept-2- ene-2-methanol.45 Minor diastereomer: 33.6 min (minor) and 35.3 min (major): 25% ee.

Major diastereomer: 37.4 min (minor) and 38.7 min (major): 41% ee.

171

References for Chapter 4

1. Brown, H. C.; Ramachandran, P. V. Versatile Alpha-Pinene-Based Borane Reagents for Asymmetric Synthesis. J. Organomet. Chem. 1995, 500, 1.

2. Brown, H. C.; Zweifel, G. Hydroboration as a Convenient Procedure for Asymmetric Synthesis of Alcohols of high Optical Purity. J. Am. Chem. Soc. 1961, 83, 486.

3. Brown, H. C.; Ayyangar, N. R.; Zweifel, G. Hydroboration.19. Reaction of Diisopinocampheylborane with Representative Trans + Hindered Olefins (Triisopinocampheyldiborane as Reagent for Configurational Assignment of Alcohols + Olefins via Hydroboration. J. Am. Chem. Soc. 1964, 86, 1071.

4. Zweifel, G.; Munekata, T.; Brown, H. C.; Ayyangar, N. R. Hydroboration. 20. Reaction of Diisopinocampheylborane with Representaive 2-Methyl-1-alkenes - Convenient Synthesis of Optically Active 2-Methyl-1-Alkanols. J. Am. Chem. Soc. 1964, 86, 1076.

5. Brown, H. C.; Yoon, N. M. Monoisopinocampheylborane - New Chiral Hydroborating Agent for Relatively Hindered (Trisubstituted) Olefins. J. Am. Chem. Soc. 1977, 99, 5514.

6. Mandal, A. K.; Jadhav, P. K.; Brown, H. C. Monoisopinocampheylborane - And Excellent Chiral Hydroborating agent for Phenyl-Substituted Tertiary Olefins - Synthesis of Alcohols Approaching 100-Percent Enantiomeric Excess. J. Org. Chem. 1980, 45, 3543.

7. Brown, H. C.; Desai, M. C.; Jadhav, P. K. Hydroboration. 61. Diisopinocampheylborane of High Optical Purity - Improved Preparation and Asymmetric Hydroboration of Representative Cis-Disubstituted Alkenes. J. Org. Chem. 1982, 47, 5065.

8. Brown, H. C.; Jadhav, P. K.; Mandal, A. K. Hydroboration. 62. Monoisopinocampheylborane, an Excellent Chiral Hydroboration Agent for Trans-disubstituted and Trisubstituted Alkenes- Evidence for a Strong Steric Dependence in Such Asymmetric Hydroborations. J. Org. Chem. 1982, 47, 5074.

9. Gonzalez, A. Z.; Roman, J. G.; Gonzalez, E.; Martinez, J.; Medina, J. R.; Matos, K.; Soderquist, J. A. 9-Borabicyclo[3.3.2]decanes and the Asymmetric Hydroboration of 1,1-Disubstituted Alkenes. J. Am. Chem. Soc. 2008, 130, 9218.

10. Masamune, S.; Kennedy, R. M.; Petersen, J. S.; Houk, K. N.; Wu, Y. Organoboron Compounds in Organic-Synthesis. 3. Mechanism of Asymmetric

172

Reduction of Dialkyl Ketones with (R,R)-2-5-Dimethylborolane. J. Am. Chem. Soc. 1986, 108, 7404.

11. Masamune, S.; Kim, B. M.; Petersen, J. S.; Sato, T.; Veenstra, S. J.; Imai, T. Organoboron Compounds in Organic Synthesis. 1. Asymmetric Hydroboration J. Am. Chem. Soc. 1985, 107, 4549.

12. Brown, H. C.; Negishi, E. Bisborolane- Highly Elusive Bisboracyclane. J. Am. Chem. Soc. 1971, 93, 6682.

13. Brown, H. C.; Negishi, E. Boraheterocycles via Cyclic Hydroboration. Tetrahedron 1977, 33, 2331.

14. Graf, C. D.; Knochel, P. Asymmetric hydroboration with new chiral monoalkylboranes bearing a non-stereogenic, chirotopic center. Tetrahedron 1999, 55, 8801.

15. Laschober, G.; Zorzi, M.; Hodgetts, K. J. Synthesis of (+/-)-trans-2,5- diisopropylborolane. Molecules 2001, 6, 244.

16. Thomas, S. P.; Aggarwal, V. K. Asymmetric Hydroboration of 1,1-Disubstituted Alkenes. Angew. Chem. Int. Ed. 2009, 48, 1896.

17. Clay, J. M.; Vedejs, E. Hydroboration with Pyridine Borane at Room Temperature. J. Am. Chem. Soc. 2005, 127, 5766.

18. Shapland, P.; Vedejs, E. Isopinocampheylborane Derivatives with > 99% ee via the DMAP complex. J. Org. Chem. 2006, 71, 6666.

19. Clay, J. M.; Karatjas, A. G.; Vedejs, E. Hydroboration with pyridine borane at room temperature J. Am. Chem. Soc. 2008, 130, 10828.

20. Karatjas, A. G.; Vedejs, E. Formation of Pinacol Boronate Esters via Pyridine Iodoborane Hydroboration. J. Org. Chem. 2008, 73, 9508.

21. Brown, H. C.; Krishnamurthy, S.; Hubbard, J. L. Addition Compoundspf Alkali- Metal Hydrides. 16. Facile Reaction of Trialkylboranes with Lithium Trimethoxyaluminohydride to Form Lithium Trialkylborohydrides: Rapid and Quantitative Transfer of Hydride from Aluminum to Boron. J. Organomet. Chem. 1979, 166, 271.

22. Biffar, W.; Noth, H.; Sedlak, D. Chemistry of Boron. 130. The reaction of Organo-Lithium Compounds with Borane Donors. Preparation and Isolation of Lithium Monoorganotrihydroborates. Organometallics 1983, 2, 579.

23. Good, C. D.; Ritter, D. M. Alkenylboranes. 2. Improved Preparative Methods and New Observations on Methylvinylboranes. J. Am. Chem. Soc. 1962, 84, 1162.

173

24. Soderquist, J. A.; Rivera, I. A Simple, Efficient Method for the Purification of Potassium Hydride and its Role in New Borohydride Chemistry. Tetrahedron Lett. 1988, 29, 3195.

25. Medina, J. R.; Cruz, G.; Cabrera, C. R.; Soderquist, J. A. New Direct B-11 NMR- based Analysis of Organoboranes Through Their Potassium Borohydrides. J. Org. Chem. 2003, 68, 4631.

26. Gillis, E. P.; Burke, M. D. A Simple and Modular Strategy for Small Molecule Synthesis: Iterative Suzuki-Miyaura Coupling of B-Protected Haloboronic Acid Building Blocks. J. Am. Chem. Soc. 2007, 129, 6716.

27. Noth, H.; Wrackmeyer, B., Nuclear Magnetic Resonance Spectroscopy of Boron Compounds. Springer-Verlag: Berlin, 1978; Vol. 14, p 460.

28. Saturnino, D. J.; Yamauchi, M.; Clayton, W. R.; Nelson, R. W.; Shore, S. G. Singly Hydrogen-Bridged Borane Anions Derived from Hydroboration products of 1,3-butadiene. X-Ray Structure Determination of [N(n-C4H9)4+][B2(C4H8)2H3- ]. J. Am. Chem. Soc. 1975, 97, 6063.

29. DuBois, D. L.; Blake, D. M.; Miedaner, A.; Curtis, C. J.; DuBois, M. R.; Franz, J. A.; Linehan, J. C. Hydride Transfer from Rhodium Complexes to Triethylborane. Organometallics 2006, 25, 4414.

30. Wang, K. K.; Brown, H. C. Hydroboration Kinetics. 6. Hydroboration of Alkenes with 9-borabicyclo[3.3.1]nonane Dimer and 9-borabicyclo[3.3.1]nonane Lewis Base Complexes in Various Solvents- an Interpretation of the Catalytic Effect of Ether Solvents on the Hydroboration Reaction. J. Am. Chem. Soc. 1982, 104, 7148.

31. Houk, K. N.; Rondan, N. G.; Wu, Y. D.; Metz, J. T.; Paddonrow, M. N. Theoretical Studies of Stereoselective Hydroborations Tetrahedron 1984, 40, 2257.

32. Imai, T.; Tamura, T.; Yamamuro, A.; Sato, T.; Wollmann, T. A.; Kennedy, R. M.; Masamune, S. Organoboron Compounds in Organic Synthesis. 2. Asymmetric Reduction of Dialkyl Ketones with (R,R)-2,5-Dimethylborolane or (R,R)-2,5- Dimethylborolane. J. Am. Chem. Soc. 1986, 108, 7402.

33. Corey, E. J.; Shibata, T.; Lee, T. W. Asymmetric Diels-Alder Reactions Catalyzed by a Triflic Acid Activated Chiral Oxazaborolidine. J. Am. Chem. Soc. 2002, 124, 3808.

34. Ryu, D. H.; Lee, T. W.; Corey, E. J. Broad Spectrum Enantioselective Diels-Alder Catalysis by Chiral, Cationic Oxazaborolidines. J. Am. Chem. Soc. 2002, 124, 9992.

174

35. Ryu, D. H.; Zhou, G.; Corey, E. J. Enantioselective and Structure Selective Diels- Alder Reactions of Unsymmetrical Quinones Catalyzed by a Chiral Oxazaborolidinium Cation. Predictive Selection Rules. J. Am. Chem. Soc. 2004, 126, 4800.

36. Hu, Q. Y.; Zhou, G.; Corey, E. J. Application of Chiral Cationic Catalysts to Several Classical Syntheses of Racemic Natural Products Transforms them into Highly Enantioselective Pathways. J. Am. Chem. Soc. 2004, 126, 13708.

37. Ryu, D. H.; Zhou, G.; Corey, E. J. Nonparallelism Between Reaction Rate and Dienophile Catalyst Affinity in Catalytic Enantioselective Diels-Alder Reactions. Org. Lett. 2005, 7, 1633.

38. Liu, D.; Canales, E.; Corey, E. J. Chiral Oxazaborolidine Aluminum Bromide Complexes are Unusually Powerful and Effective Catalysts for Enantioselective Diels-Alder Reactions. J. Am. Chem. Soc. 2007, 129, 1498.

39. Murahashi, S. I.; Naota, T.; Ito, K.; Maeda, Y.; Taki, H. Ruthenium-Catalyzed Oxidative Transformation of Alcohols and Aldehydes to Esters and Lactones. J. Org. Chem. 1987, 52 (19), 4319.

40. Stampfer, W.; Kosjek, B.; Faber, K.; Kroutil, W. Biocatalytic Asymmetric Hydrogen Transfer Employing Rhodococcus Ruber DSM 44541. J. Org. Chem. 2003, 68, 402.

41. Szollosi, G.; Torok, B.; Bartok, M. Monitoring of Optical Isomers of Chiral Alcohols and Derivatives by Chiral Gas Chromatography. Effect of Derivatization on the Enantio-Differentiation. Chromatographia 1998, (1-2), 81.

42. Blakemore, P. R.; Burge, M. S. Iterative Stereospecific Reagent-Controlled Homologation of Pinacol Boronates by Enantioenriched -Chloroalkyllithium Reagents. J. Am. Chem. Soc. 2007, 129, 3068.

43. Birman, V. B.; Li, X. M. Homobenzotetramisole: An Effective Catalyst for Kinetic Resolution of Aryl-Cycloalkanols. Org. Lett. 2008, 10, 1115.

44. Abiko, A. Dicyclohexylboron trifluoromethanesulfonate. Org. Synth. 2003, 1, 103-108.

45. Boeckman, R. K.; Nelson, S. G.; Gaul, M. D. Toward the Development of a General CHiral Auxiliary. 1. Preparation of a New Class of Camphor Lactam Imides and Their Application to the Construction of Quaternary Centers via Diels-Alder Cycloaddition. J. Am. Chem. Soc. 1992, 114, 2258.

175

Appendix

NMR Spectra

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198