Naphthalenes and Quinolines by Domino Reactions of Morita–Baylis–Hillman Acetates

An efficient synthetic route to highly functionalized naphthalenes and quinolines has been developed using domino reactions between Morita–Baylis–Hillman (MBH) acetates and active methylene compounds (AMCs) promoted by anhydrous K2CO3 in dry N,N-dimethylformamide (DMF) at 23 °C. The substrates incorporate allylic acetates positioned adjacent to a Michael acceptor as well as an aromatic ring activated toward a SNAr ring closure. A control experiment indicated that the initial reaction was an SN2’-type displacement of a side chain acetoxy by the AMC anion to afford the alkene product bearing the added nucleophile trans to the SNAr aromatic ring acceptor. Thus, equilibration of the alkene geometry of the initial product was required prior to cyclization. Products were isolated in good to excellent yields. Numerous cases (24) are reported, and several mechanistic possibilities are discussed.


Introduction
One facet of our research program has focused on domino reactions for the rapid assembly of molecules with potential use in drug synthesis. This requires highly functionalized compounds with reactive sites strategically positioned to capture intermediates from an initial reaction in one or more subsequent reactions to produce targets of high value in an efficient, eco-friendly manner. Among the compounds meeting these requirements are the products of the Morita-Baylis-Hillman (MBH) reaction. These highly functionalized adducts have proven quite useful in drug synthesis [1,2]. Additionally, the naphthalenes [3] and quinolines [4,5] targeted in this work could have immense value in medicinal chemistry.
Previous work by others has appeared in this area, but details were lacking, and the diversity of examples reported was limited. One article outlined the formation of a number of naphthalene derivatives [6] from MBH acetates and active methylene compounds (AMCs) ( Figure 1A), but the initial adduct was assumed to have the correct geometry for ring closure, and the only molecule eliminated during the final aromatization was benzenesulfinic acid (PhSO 2 H). A second report described a synthesis of 3-quinolinecarboxylic esters [7] involving the reaction of ethyl 2-((2-chlorophenyl)((tosylsulfonamido)methyl)acrylate acrylate with tosylsulfonamide ( Figure 1B). A third study [8] advanced a synthesis of dihydroacridines from MBH acetates derived from 2-chloroquinoline-3-carboxaldehyde and AMCs to fuse a substituted benzene ring to the heterocycle ( Figure 1C). The final account detailed the AMC-dependent annulation of substituted benzene rings by reaction of this same MBH precursor to give acridines and phenanthridines [9] ( Figure 1D). The current work describes the synthesis of naphthalenes and quinolines with a broader range of functionality, evaluates a selection of different leaving groups, and discusses several mechanistic possibilities.

Results and Discussion
The MBH reaction was first reported in 1968 [10] and has undergone many modifications [11][12][13] to improve its outcome. The generally accepted procedure calls for 1 equiv. of the aldehyde and 1.2 equiv. of the electron-poor alkene in the presence of 1.5 equiv. of 1,4-diazabicyclo [2.2.2]octane (DABCO) in acetonitrile (ACN) at room temperature (23 °C) [14]. For the current project, we have generated these adducts by the reaction of polarized alkenes with aromatic aldehydes incorporating functionality that further activates the aromatic ring toward a SNAr reaction using a 2:1 stoichiometry of alkene:aldehyde with 1.2 equiv. of DABCO (Scheme 1). Thus, 2-fluoro-5nitrobenzaldehyde (1a), 2-fluoro-5-cyanobenzaldehyde (1b), and 2-fluoronicotinaldehyde (1c) were each treated with ethyl acrylate (2a) and acrylonitrile (2b) in ACN at 23 °C for 2 days to give the MBH alcohols (3a-e) in 86-98% yields. Attempts to acylate the alcohol adduct 3b with acetic anhydride at reflux led to acylation of the alcohol and allylic inversion of the acetate to afford (E)-2cyano-3-(2-fluoro-5-nitrophenyl)allyl acetate (4). Interestingly, an X-ray structure of this product showed the acetoxymethyl group of the rearranged acetate to be trans to the substrate aromatic ring (see Scheme 1 and the Supplementary Materials). Acylation of the MBH adducts without rearrangement was achieved using trimethylsilyl trifluoromethanesulfonate (TMSOTf) as an acylation catalyst [15] in dichloromethane (DCM). This catalyst permitted the acylation of 3a-e at 0 °C in 30 min and delivered the unrearranged acetates 5-9, respectively, in 95-98% yields.
The results of our study are summarized in Tables 1 and 2. The reaction of substrates 5-8 (1 equiv.) with AMCs 10-14 (1.5 equiv.) yielded 1,3,6-trisubstituted naphthalenes 15-18, while precursor 9 afforded 6,8-disubstituted quinolines 19 with the same panel of nucleophiles. Naphthalene formation was promoted by K2CO3 (1.5 equiv.) in N,N-dimethylformamide (DMF) at 23 °C in 1 h, while quinolines were similarly prepared but required heating at 90 °C for 6 h. Yields were uniformly good to excellent, with the 2-fluoro-5-nitrophenyl SNAr acceptor giving the most efficient ring closure, followed by the 2-fluoro-5-cyano compound and finally 2-fluoropyridine. The overall process appears to involve diastereoselective addition of the nucleophile to the double bond methylene of the MBH substrate with loss of the acetoxy group. Following this addition,

Results and Discussion
The MBH reaction was first reported in 1968 [10] and has undergone many modifications [11][12][13] to improve its outcome. The generally accepted procedure calls for 1 equiv. of the aldehyde and 1.2 equiv. of the electron-poor alkene in the presence of 1.5 equiv. of 1,4-diazabicyclo [2.2.2]octane (DABCO) in acetonitrile (ACN) at room temperature (23 • C) [14]. For the current project, we have generated these adducts by the reaction of polarized alkenes with aromatic aldehydes incorporating functionality that further activates the aromatic ring toward a S N Ar reaction using a 2:1 stoichiometry of alkene:aldehyde with 1.2 equiv. of DABCO (Scheme 1). Thus, 2-fluoro-5-nitrobenzaldehyde (1a), 2-fluoro-5-cyanobenzaldehyde (1b), and 2-fluoronicotinaldehyde (1c) were each treated with ethyl acrylate (2a) and acrylonitrile (2b) in ACN at 23 • C for 2 days to give the MBH alcohols (3a-e) in 86-98% yields. Attempts to acylate the alcohol adduct 3b with acetic anhydride at reflux led to acylation of the alcohol and allylic inversion of the acetate to afford (E)-2-cyano-3-(2-fluoro-5-nitrophenyl)allyl acetate (4). Interestingly, an X-ray structure of this product showed the acetoxymethyl group of the rearranged acetate to be trans to the substrate aromatic ring (see Scheme 1 and the Supplementary Materials). Acylation of the MBH adducts without rearrangement was achieved using trimethylsilyl trifluoromethanesulfonate (TMSOTf) as an acylation catalyst [15] in dichloromethane (DCM). This catalyst permitted the acylation of 3a-e at 0 • C in 30 min and delivered the unrearranged acetates 5-9, respectively, in 95-98% yields.
The results of our study are summarized in Tables 1 and 2. The reaction of substrates 5-8 (1 equiv.) with AMCs 10-14 (1.5 equiv.) yielded 1,3,6-trisubstituted naphthalenes 15-18, while precursor 9 afforded 6,8-disubstituted quinolines 19 with the same panel of nucleophiles. Naphthalene formation was promoted by K 2 CO 3 (1.5 equiv.) in N,N-dimethylformamide (DMF) at 23 • C in 1 h, while quinolines were similarly prepared but required heating at 90 • C for 6 h. Yields were uniformly good to excellent, with the 2-fluoro-5-nitrophenyl S N Ar acceptor giving the most efficient ring closure, followed by the 2-fluoro-5-cyano compound and finally 2-fluoropyridine. The overall process appears to involve diastereoselective addition of the nucleophile to the double bond methylene of the MBH substrate with loss of the acetoxy group. Following this addition, deprotonation of the methine hydrogen from the added nucleophile, S N Ar cyclization, and elimination fuses a disubstituted benzene ring to the original aromatic nucleus. The exact sequence of events for this process is unclear and is discussed in more detail below. In the current reactions, the final aromatization occurred by the elimination of benzenesulfinic acid (PhSO 2 H) [6,16], nitrous acid (HNO 2 ) [9,17], or ethoxycarbonyl (CO 2 Et, presumably by hydrolysis and decarboxylation) [9,18], all of which have precedent in the literature, although the loss of CO 2 Et in preference to the CN group was not expected. Aromatization by the elimination of CN is known [19,20], but it appears to be quite rare.
Molecules 2020, 25, x 3 of 17 deprotonation of the methine hydrogen from the added nucleophile, SNAr cyclization, and elimination fuses a disubstituted benzene ring to the original aromatic nucleus. The exact sequence of events for this process is unclear and is discussed in more detail below. In the current reactions, the final aromatization occurred by the elimination of benzenesulfinic acid (PhSO2H) [6,16], nitrous acid (HNO2) [9,17], or ethoxycarbonyl (CO2Et, presumably by hydrolysis and decarboxylation) [9,18], all of which have precedent in the literature, although the loss of CO2Et in preference to the CN group was not expected. Aromatization by the elimination of CN is known [19,20], but it appears to be quite rare.  The syntheses are quite efficient, requiring only three steps and one chromatographic purification for the final products 15-19. The products were isolated in good to excellent yields (75-96%) and were fully characterized by spectroscopic methods. The IR spectra indicated the presence of the expected functional groups (conjugated CO 2 Et or CN, polarized conjugated double bonds, and NO 2 ). The naphthalenes showed three one-proton doublets (J 1,3 < 3 Hz or apparent singlets) for the isolated protons at C2, C4, and C5, a doublet of doublets (J 1,2 and 1,3 = large and small) for the proton at C7, and a doublet (J 1,2 > 8 Hz) for the proton at C8. The chemical shifts were in accordance with those expected for an electron-deficient aromatic system. The quinolines exhibited two doublets (J 1,3 < 3 Hz) for the protons at C2 and C4 as well as characteristic chemical shifts and couplings for the protons on C5-C7 of the system. The 13 C-NMR were appropriate with respect to the number of carbonyl, aromatic, and aliphatic carbons. The mass spectra showed an ion corresponding to the molecular weight of the compound, and elemental analyses confirmed the formulas and purity.
Two basic mechanisms are possible for this domino transformation: (1) initial S N Ar attack by the AMC anion on the aromatic ring, followed by conjugate addition to the side chain double bond with an elimination of acetoxy or (2) initial S N 2 -type displacement of acetoxy from the side chain, followed by a S N Ar cyclization. Option 1 would unquestionably predict the final product of the reaction and would not depend on the selective formation of a single alkene from the S N 2 process. Option 2 would require a diastereoselective addition to give the double bond geometry that positions the active methylene fragment cis to the S N Ar acceptor ring. To probe this aspect of the transformation, a control experiment was performed wherein the anion of methyl phenylsulfonylacetate was generated in the presence of a mixture of 2-fluoro-5-nitrotoluene (20) and 2-cyano-1-(2-fluorophenyl)allyl acetate (21) to determine the relative reactivity of the side chain versus the aromatic ring. If the side chain reacted preferentially, we were also interested in assessing any E-Z diastereoselectivity in the allylically rearranged double bond. Under the standard conditions, a reaction occurred exclusively at the side chain of the MBH acetate and also afforded only methyl (Z)-4-cyano-5-(2-fluorophenyl)-2-(phenylsulfonyl)-4-pentenoate (22, 70%) having the added active methylene nucleophile positioned trans to the aromatic S N Ar acceptor ring. This was confirmed by an X-ray structure of adduct 22 (see Scheme 2 and the Supplementary Materials). Compound 20 (93%) was recovered unchanged from this experiment. Thus, the overall sequence must proceed via Option 2 above, followed by double bond equilibration prior to S N Ar ring closure.   The syntheses are quite efficient, requiring only three steps and one chromatographic purification for the final products 15-19. The products were isolated in good to excellent yields (75-96%) and were fully characterized by spectroscopic methods. The IR spectra indicated the presence Molecules 2020, 25, x 4 of 17 Table 1. Formation of naphthalenes. CO2Et The syntheses are quite efficient, requiring only three steps and one chromatographic purification for the final products 15-19. The products were isolated in good to excellent yields (75-96%) and were fully characterized by spectroscopic methods. The IR spectra indicated the presence of the expected functional groups (conjugated CO2Et or CN, polarized conjugated double bonds, CO2Et The syntheses are quite efficient, requiring only three steps and one chromatographic purification for the final products 15-19. The products were isolated in good to excellent yields (75-96%) and were fully characterized by spectroscopic methods. The IR spectra indicated the presence of the expected functional groups (conjugated CO2Et or CN, polarized conjugated double bonds, CO2Et a AMC = active methylene compound. b L = leaving group. The syntheses are quite efficient, requiring only three steps and one chromatographic purification for the final products 15-19. The products were isolated in good to excellent yields (75-96%) and were fully characterized by spectroscopic methods. The IR spectra indicated the presence of the expected functional groups (conjugated CO2Et or CN, polarized conjugated double bonds, and NO2). The naphthalenes showed three one-proton doublets (J1,3 < 3 Hz or apparent singlets) for the isolated protons at C2, C4, and C5, a doublet of doublets (J1,2 and 1,3 = large and small) for the CO2Et CO2Et a AMC = active methylene compound. b L = leaving group.
The syntheses are quite efficient, requiring only three steps and one chromatographic purification for the final products 15-19. The products were isolated in good to excellent yields (75-96%) and were fully characterized by spectroscopic methods. The IR spectra indicated the presence of the expected functional groups (conjugated CO2Et or CN, polarized conjugated double bonds, and NO2). The naphthalenes showed three one-proton doublets (J1,3 < 3 Hz or apparent singlets) for the isolated protons at C2, C4, and C5, a doublet of doublets (J1,2 and 1,3 = large and small) for the proton at C7, and a doublet (J1,2 > 8 Hz) for the proton at C8. The chemical shifts were in accordance   couplings for the protons on C5-C7 of the system. The 13 C-NMR were appropriate with respect to the number of carbonyl, aromatic, and aliphatic carbons. The mass spectra showed an ion corresponding to the molecular weight of the compound, and elemental analyses confirmed the formulas and purity.
Two basic mechanisms are possible for this domino transformation: (1) initial SNAr attack by the AMC anion on the aromatic ring, followed by conjugate addition to the side chain double bond with an elimination of acetoxy or (2) initial SN2′-type displacement of acetoxy from the side chain, followed by a SNAr cyclization. Option 1 would unquestionably predict the final product of the reaction and would not depend on the selective formation of a single alkene from the SN2′ process. Option 2 would require a diastereoselective addition to give the double bond geometry that positions the active methylene fragment cis to the SNAr acceptor ring. To probe this aspect of the transformation, a control experiment was performed wherein the anion of methyl phenylsulfonylacetate was generated in the presence of a mixture of 2-fluoro-5-nitrotoluene (20) and 2-cyano-1-(2-fluorophenyl)allyl acetate (21) to determine the relative reactivity of the side chain versus the aromatic ring. If the side chain reacted preferentially, we were also interested in assessing any E-Z diastereoselectivity in the allylically rearranged double bond. Under the standard conditions, a reaction occurred exclusively at the side chain of the MBH acetate and also afforded only methyl (Z)-4-cyano-5-(2-fluorophenyl)-2-(phenylsulfonyl)-4-pentenoate (22, 70%) having the added active methylene nucleophile positioned trans to the aromatic SNAr acceptor ring. This was confirmed by an X-ray structure of adduct 22 (see Scheme 2 and the Supplementary Materials). Compound 20 (93%) was recovered unchanged from this experiment. Thus, the overall sequence must proceed via Option 2 above, followed by double bond equilibration prior to SNAr ring closure.
couplings for the protons on C5-C7 of the system. The 13 C-NMR were appropriate with respect to the number of carbonyl, aromatic, and aliphatic carbons. The mass spectra showed an ion corresponding to the molecular weight of the compound, and elemental analyses confirmed the formulas and purity. Two basic mechanisms are possible for this domino transformation: (1) initial SNAr attack by the AMC anion on the aromatic ring, followed by conjugate addition to the side chain double bond with an elimination of acetoxy or (2) initial SN2′-type displacement of acetoxy from the side chain, followed by a SNAr cyclization. Option 1 would unquestionably predict the final product of the reaction and would not depend on the selective formation of a single alkene from the SN2′ process. Option 2 would require a diastereoselective addition to give the double bond geometry that positions the active methylene fragment cis to the SNAr acceptor ring. To probe this aspect of the transformation, a control experiment was performed wherein the anion of methyl phenylsulfonylacetate was generated in the presence of a mixture of 2-fluoro-5-nitrotoluene (20) and 2-cyano-1-(2-fluorophenyl)allyl acetate (21) to determine the relative reactivity of the side chain versus the aromatic ring. If the side chain reacted preferentially, we were also interested in assessing any E-Z diastereoselectivity in the allylically rearranged double bond. Under the standard conditions, a reaction occurred exclusively at the side chain of the MBH acetate and also afforded only methyl (Z)-4-cyano-5-(2-fluorophenyl)-2-(phenylsulfonyl)-4-pentenoate (22, 70%) having the added active methylene nucleophile positioned trans to the aromatic SNAr acceptor ring. This was confirmed by an X-ray structure of adduct 22 (see Scheme 2 and the Supplementary Materials). Compound 20 (93%) was recovered unchanged from this experiment. Thus, the overall sequence must proceed via Option 2 above, followed by double bond equilibration prior to SNAr ring closure.  Analysis of the reactant in the SN2′ reaction (Option 2 above) demands that the starting conformation minimizes the steric interaction between the aryl, the acetoxy, and the electronwithdrawing group (EWG = CO2Et or CN) (Scheme 2). An earlier study describing the reaction of an acrylate-derived MBH alcohol with HBr proposed conformation 23 (Figure 2), which has an additional stabilizing H-bond between the OH and the EWG, to rationalize the products [20]. This model would favor formation of the alkene that places the incoming active methylene fragment cis to the 2-fluoro-5-nitrophenyl moiety in the product. With the current substrates, H-bonding is not possible, and so the large acetoxy group was expected to orient away from the EWG (perhaps more so for CO2Et than CN) to minimize steric crowding (Scheme 3). Then, an initial SN2′ reaction involving a syn orientation of the nucleophile and leaving group invoked in some earlier work [21,22] would give an adduct with the AMC fragment trans to the SNAr acceptor ring as in the aforementioned competitive reaction. Thus, in order for the cyclization to occur, the initial Michael adduct 22 must equilibrate to bring the nucleophilic center cis to the activated aromatic ring to allow cyclization by an intramolecular SNAr reaction. Possible mechanisms for this isomerization are discussed below. Analysis of the reactant in the S N 2 reaction (Option 2 above) demands that the starting conformation minimizes the steric interaction between the aryl, the acetoxy, and the electron-withdrawing group (EWG = CO 2 Et or CN) (Scheme 2). An earlier study describing the reaction of an acrylate-derived MBH alcohol with HBr proposed conformation 23 (Figure 2), which has an additional stabilizing H-bond between the OH and the EWG, to rationalize the products [20]. This model would favor formation of the alkene that places the incoming active methylene fragment cis to the 2-fluoro-5-nitrophenyl moiety in the product. With the current substrates, H-bonding is not possible, and so the large acetoxy group was expected to orient away from the EWG (perhaps more so for CO 2 Et than CN) to minimize steric crowding (Scheme 3). Then, an initial S N 2 reaction involving a syn orientation of the nucleophile and leaving group invoked in some earlier work [21,22] would give an adduct with the AMC fragment trans to the S N Ar acceptor ring as in the aforementioned competitive reaction. Thus, in order for the cyclization to occur, the initial Michael adduct 22 must equilibrate to bring the nucleophilic center cis to the activated aromatic ring to allow cyclization by an intramolecular S N Ar reaction. Possible mechanisms for this isomerization are discussed below.
involving a syn orientation of the nucleophile and leaving group invoked in some earlier work [21,22] would give an adduct with the AMC fragment trans to the SNAr acceptor ring as in the aforementioned competitive reaction. Thus, in order for the cyclization to occur, the initial Michael adduct 22 must equilibrate to bring the nucleophilic center cis to the activated aromatic ring to allow cyclization by an intramolecular SNAr reaction. Possible mechanisms for this isomerization are discussed below. In reactions at the side chain allylic moiety, there are two possible scenarios to explain the addition to the MBH acetate. The first [22] invokes an SN2′ process where a nucleophile attacks C1 (unsubstituted) of the allylic moiety, resulting in an accumulation of electron density on the opposite face of the molecule at C2. Then, this electron density is responsible for assisting in the departure of the acetoxy leaving group at C3. This process has been used to rationalize the reported preference for a syn orientation between the incoming nucleophile and the leaving group often seen in the reaction [23][24][25], although exceptions are known [26] with certain Nu -/Lcombinations. The second scenario [27] claims that there is no justification for a syn orientation between the nucleophile and the leaving group. In fact, in highly substituted derivatives, such as 5-9, it has been suggested that the formation of a stabilized carbocation is more likely, in which case the avoidance of steric interactions guides the transformation to give the final product stereochemistry.
The SN2′ initiated mechanism is depicted in Scheme 3 for the reaction of 6 with 11 to give 16b. The initial conformation of 6 would allow minimal steric interaction between the groups on the  The second mechanistic scenario, illustrated in Scheme 4, begins with loss of the acetoxy group to form stabilized 3-arylallylic cations E and F [27]. At this point, it may appear that addition to carbocation F should be easier, since the alkene terminus is less hindered. Furthermore, since the EWG should not be coplanar with the allylic cation, it should have less steric influence on the reaction, and the larger bond angles around the carbocation should permit the formation of both E and F. Addition of the active methylene nucleophile would then lead to both 22 and 24, which could cyclize as above. In both examples cited in this paper, the alkene with the nucleophile trans to the SNAr aromatic system was isolated with no contamination by the cis. This observation could disqualify this mechanism, but if 24 cyclizes very rapidly, it may not be possible to detect or isolate this short-lived intermediate.  In reactions at the side chain allylic moiety, there are two possible scenarios to explain the addition to the MBH acetate. The first [22] invokes an S N 2 process where a nucleophile attacks C1 (unsubstituted) of the allylic moiety, resulting in an accumulation of electron density on the opposite face of the molecule at C2. Then, this electron density is responsible for assisting in the departure of the acetoxy leaving group at C3. This process has been used to rationalize the reported preference for a syn orientation between the incoming nucleophile and the leaving group often seen in the reaction [23][24][25], although exceptions are known [26] with certain Nu -/Lcombinations. The second scenario [27] claims that there is no justification for a syn orientation between the nucleophile and the leaving group. In fact, in highly substituted derivatives, such as 5-9, it has been suggested that the formation of a stabilized carbocation is more likely, in which case the avoidance of steric interactions guides the transformation to give the final product stereochemistry.
The S N 2 initiated mechanism is depicted in Scheme 3 for the reaction of 6 with 11 to give 16b. The initial conformation of 6 would allow minimal steric interaction between the groups on the allylic side chain. In this conformation, the acetoxy substituent should hinder approach to the top face while the planar aromatic ring does not significantly block the bottom face. Introduction of the nucleophile to the bottom face would result in an accumulation of negative charge on the top face of the allylic system, as shown in A. Rotation of the acetoxy group 120 • would position it anti to the accumulated negative charge and move the substrate 2-fluoro-5-nitrophenyl (Ar) moiety away from the newly added nucleophile to give B. Then, the elimination of acetoxy would deliver the Z alkene 22. Subsequent equilibration of the double bond geometry to generate 24 and deprotonation of the residual active methine proton to give C would be followed by ipso attack at the fluorine-bearing aromatic carbon to afford Meisenheimer intermediate D.
Rearomatization by loss of fluoride should then lead to 25.
After ring closure, a conformation aligned for the elimination of benzenesulfinic acid by an E2 process would afford the fused aromatic compound 16b.
The second mechanistic scenario, illustrated in Scheme 4, begins with loss of the acetoxy group to form stabilized 3-arylallylic cations E and F [27]. At this point, it may appear that addition to carbocation F should be easier, since the alkene terminus is less hindered. Furthermore, since the EWG should not be coplanar with the allylic cation, it should have less steric influence on the reaction, and the larger bond angles around the carbocation should permit the formation of both E and F. Addition of the active methylene nucleophile would then lead to both 22 and 24, which could cyclize as above.
In both examples cited in this paper, the alkene with the nucleophile trans to the S N Ar aromatic system was isolated with no contamination by the cis. This observation could disqualify this mechanism, but if 24 cyclizes very rapidly, it may not be possible to detect or isolate this short-lived intermediate. The second mechanistic scenario, illustrated in Scheme 4, begins with loss of the acetoxy group to form stabilized 3-arylallylic cations E and F [27]. At this point, it may appear that addition to carbocation F should be easier, since the alkene terminus is less hindered. Furthermore, since the EWG should not be coplanar with the allylic cation, it should have less steric influence on the reaction, and the larger bond angles around the carbocation should permit the formation of both E and F. Addition of the active methylene nucleophile would then lead to both 22 and 24, which could cyclize as above. In both examples cited in this paper, the alkene with the nucleophile trans to the SNAr aromatic system was isolated with no contamination by the cis. This observation could disqualify this mechanism, but if 24 cyclizes very rapidly, it may not be possible to detect or isolate this short-lived intermediate.  Two pathways can be envisioned for the equilibration of 22 to 24 that must occur before S N Ar cyclization: (1) a well-precedented Michael-reverse Michael process [28][29][30][31][32] involving the excess AMC or perhaps (2) an intramolecular addition-elimination reaction (Scheme 5). This alternative process would involve loss of the acidic methine proton from 22 and addition of the resulting anion G to the benzylic double bond to give H. This may be possible, since the electron-deficient aromatic ring should stabilize the benzylic anion. Then, bond rotation to give I and ring opening would deliver the anion needed for cyclization. Once equilibration occurs, ring closure to 25 and elimination to generate 16b should be facile. Even if the equilibrium does not strongly favor the required alkene geometry, product formation should gradually siphon the initial adduct over to the fused aromatic target. Two pathways can be envisioned for the equilibration of 22 to 24 that must occur before SNAr cyclization: (1) a well-precedented Michael-reverse Michael process [28][29][30][31][32] involving the excess AMC or perhaps (2) an intramolecular addition-elimination reaction (Scheme 5). This alternative process would involve loss of the acidic methine proton from 22 and addition of the resulting anion G to the benzylic double bond to give H. This may be possible, since the electron-deficient aromatic ring should stabilize the benzylic anion. Then, bond rotation to give I and ring opening would deliver the anion needed for cyclization. Once equilibration occurs, ring closure to 25 and elimination to generate 16b should be facile. Even if the equilibrium does not strongly favor the required alkene geometry, product formation should gradually siphon the initial adduct over to the fused aromatic target. SO  Addition to the substrates with formation of the Z alkene (EWG and added AMC trans) was confirmed by both a side reaction observed during the synthesis to form acetate 4 and a control experiment to form 15 (Scheme 2). The formation of a single product has significant precedent in the case of EWG = CO2Et, but it is less probable for the sterically smaller EWG = CN group [33]. Addition to the substrates with formation of the Z alkene (EWG and added AMC trans) was confirmed by both a side reaction observed during the synthesis to form acetate 4 and a control experiment to form 15 (Scheme 2). The formation of a single product has significant precedent in the case of EWG = CO 2 Et, but it is less probable for the sterically smaller EWG = CN group [33]. Unfortunately, the ester intermediate, while a single isomer, was not a solid and could not be subjected to X-ray structure analysis. Additionally, Nuclear Overhauser enhancement (NOE) measurements were inconclusive. Nevertheless, products incorporating both EWGs formed in high yields. In cases where the initial addition yields the Z alkene, a reversible Michael reaction [28][29][30][31][32] or intramolecular isomerization could establish an E-Z equilibrium that would eventually cyclize and eliminate H-L to afford the aromatic products. A similar rationale can be applied to the pyridine-containing substrates that lead to quinolines. The stereoselective formation of the alkenes with the added nucleophile cis to the aromatic ring has been observed in the past with small nucleophiles [21,[34][35][36][37][38][39][40]. This has been confirmed by spectral characterization, and in one instance, X-ray analysis [41]. However, our results, using large stabilized nucleophiles, appear to differ from these earlier findings.

General Methods
Unless otherwise indicated, all reactions were carried out under dry N 2 in oven-dried glassware. All reagents and solvents were used as received. Reactions were monitored by thin layer chromatography on silica gel GF plates (Analtech No 21521

Representative Procedure for the Synthesis of MBH Alcohols (3a-e)
To a stirred solution of the aldehyde 1a-c (1 equiv.) and DABCO (1.2 equiv.) in ACN (8 mL) under N 2 was added ethyl acrylate (2a) or acrylonitrile (2b) (2 equiv.) at 23 • C. After 24-48 h, TLC analysis (10% EtOAc/hexane) indicated the reaction was complete. The solution was added to water, and the mixture was extracted with EtOAc (3 × 25 mL). The combined organic layers were washed with 0.5 M HCl (2 × 10 mL), saturated NaHCO 3 , and saturated NaCl and then dried (Na 2 SO 4 ). Removal of the solvent under vacuum resulted in pure MBH alcohols 3a-e, which were used without further purification.

Synthesis of (E)-2-cyano-3-(2-fluoro-5-nitrophenyl)allyl acetate (4)
A solution of the MBH alcohol 3b (1.11 g, 5 mmol) was dissolved in acetic anhydride (5.0 mL) and boiled for 6 h. The mixture was cooled, concentrated under vacuum, and the residue was dissolved in DCM (25 mL). The solution was washed with saturated NaHCO 3 (3 × 50 mL) and water, dried (Na 2 SO 4 ), and concentrated under vacuum to afford 4 (1.  A solution of the MBH alcohol 3a-d (1 equiv.) in DCM (2 mL) was treated with acetic anhydride (1.5 equiv.) at 0 • C, followed by a 1 M solution of TMSOTf in DCM (20 µL/mmol substrate) [15]. The MBH alcohol 3e (1 equiv.) required acetic anhydride (3 equiv.) and 1 M TMSOTf in DCM (60 µL/mmol substrate). After 30 min, TLC analysis (10% ether in hexane) showed the reaction was complete. The reaction mixture was added to saturated NaHCO 3 , and the mixture was extracted with DCM (3 × 5 mL). The organic extracts were washed with water, dried (Na 2 SO 4 ), and evaporated under vacuum to afford clean MBH acetates 5-9, which did not require further purification. After optimization, this procedure was scaled up by a factor of five with no significant decrease in yield. 3.5. Representative Procedure for the Synthesis of Naphthalene and Quinoline Analogs Using MBH Acetates and Active Methylene Compounds A 50 mL, round-bottomed flask equipped with a condenser, stir bar, and N 2 inlet, was charged with an MBH acetate (1 mmol) in DMF (2 mL). The corresponding active methylene compound (1.5 mmol) in DMF (1 mL) and K 2 CO 3 (207 mg, 1.5 mmol) were added at room temperature with continued stirring. For naphthalenes 15-18, the reaction was complete in 1 h. For quinolines 19, the reaction was stirred at room temperature for 1 h and gradually heated to 90 • C with stirring for 6 h. In each case, TLC analysis (30% EtOAc in hexane) indicated the reaction was complete. The solution was poured into de-ionized water (15 mL), and the mixture was extracted with EtOAc (3 × 25 mL). The combined organic layers were washed with saturated NaCl and dried (Na 2 SO 4 ). Removal of the solvent under vacuum gave the crude product, which was purified by silica gel column chromatography to afford the pure naphthalene/quinoline derivatives.