molecules Review
Synergy between Experimental and Theoretical Results of Some Reactions of Annelated 1,3-Azaphospholes Raj K. Bansal *
ID
, Raakhi Gupta and Manjinder Kour
Department of Chemistry, The IIS University, Jaipur 302020, India;
[email protected] (R.G.);
[email protected] (M.K.) * Correspondence:
[email protected]; Tel.: +91-9057-242266 Received: 5 May 2018; Accepted: 24 May 2018; Published: 27 May 2018
Abstract: Computational calculations have been used successfully to explain the reactivity of the >C=P- functionality in pyrido-annelated 1,3-azaphospholes. Theoretical investigation at the Density Functional Theory (DFT) level shows that the lone pair of the bridgehead nitrogen atoms is involved in extended conjugation, due to which electron density increases considerably in the five-membered azaphosphole ring. The electron density in the azaphosphole is further enhanced by the presence of an ester group at the 3-position making the >C=P- functionality electron-rich. Thus, 1,3-azaphospholo[5,1-a]pyridine, i.e., 2-phosphaindolizine having ester group at the 3-position only does not undergo Diels-Alder (DA) reaction with an electron rich diene, such as 2,3-dimethyl-1,3-butadiene (DMB). However, an ester group at 1-position acts as electron-sink, due to which transfer of the electron density to the >C=P- moiety is checked and DA reaction occurs across the >C=P- functionality. The coordination of the Lewis acid to the carbonyl group at the 3-position raises the activation barrier, while it is lowered remarkably when it is coordinated to the P atom. Furthermore, the attack of 1,3-butadiene on the Si face of the P-coordinated (o-menthoxy)aluminum dichloride complex is a lower activation energy path. Fukui functions could be used to explain relative reactivities of indolizine and 2-phosphaindolizine towards electrophilic substitution reactions. Keywords: synergy; DFT calculations
1,3-azaphospholes;
Diels-Alder reaction;
electrophilic substitution;
1. Introduction Almost a century ago, Paul Dirac [1] could foresee the possibility of applying quantum mechanical concepts to solving the experimental problems in chemistry when he remarked, “The underlying physical laws necessary for the mathematical theory of a large part of physics and the whole of chemistry are thus completely known, and the difficulty is only that the exact application of these laws leads to equations much too complicated to be soluble. It therefore becomes desirable that approximate practical methods of applying quantum mechanics should be developed, which can lead to an explanation of the main features of complex atomic systems without too much computation”. Dirac’s dream inched towards fruition when Kohn, Hohenberg, and Sham [2,3] developed the density functional theory (DFT). It was followed by dramatic progress when many DFT codes were made available commercially, which enabled even less experienced persons with modest computational facilities to do calculations with much accuracy and look into the electronic structures and many other properties of the substances of interest. During the last two decades, the availability of a variety of new and more accurate DFT functionals incorporating specific features has brought DFT computations to the stage where they are used advantageously as a tool not only to explain the experimental
Molecules 2018, 23, 1283; doi:10.3390/molecules23061283
www.mdpi.com/journal/molecules
Molecules 2018, 23, 1283 Molecules 2018, 23, x FOR PEER REVIEW Molecules 2018, 23, x FOR PEER REVIEW
2 of 13 2 of 13 2 of 13
explainbut thealso experimental but also to planconditions and modify for achieving results to plan andresults modify experimental forexperimental achieving theconditions desired objectives [4–9]. explain theobjectives experimental results but years, also toincrease plan has and modify experimental conditions for achieving the desired [4–9]. In arecent there a perceptible increase in the number of In recent years, there has been perceptible inbeen the number of research papers in chemistry the desired objectives [4–9]. In recent years, there has been a perceptible increase in the number research papers in chemistry reporting both experimental and theoretical results together. Fristrup reporting both experimental and theoretical results together. Fristrup and co-workers in a recentof research papersininthe reporting both experimental and theoretical results together. Fristrup and co-workers achemistry recent article emphasized the synergy between experimental and theoretical article emphasized synergy between experimental and theoretical methods in the exploration of and co-workers in a recent article emphasized the synergy between experimental and theoretical methods in thetransitional explorationmetal of homogeneous catalysis [10].article In another homogeneous catalysis [10].transitional In anothermetal interesting review titled,interesting “Towards methods in the exploration of homogeneous transitional metal catalysis [10]. In another review article titled, “Towards the computational design of commented, solid catalysts”, Norskov etinteresting al.can [11] the computational design of solid catalysts”, Norskov et al. [11] “Theoretical methods be review article titled, “Towards the computational design of solid catalysts”, Norskov et al. [11] commented, methods can in be detail used and to describe surfacevariations chemicalinreactions detailfrom andone to used to describe“Theoretical surface chemical reactions to understand catalytic in activity commented, “Theoretical methods can be used to describe surface chemical reactions in detail and understand variations in catalytic activity from one catalyst to another.” In spite of many challenges in the catalyst to another.” In spite of many challenges in the field, they added “Computational approaches forto understand variations in catalytic of activity fromhold one to another.” spite of many in the field, they added approaches forcatalyst the discovery andIn development ofchallenges catalysts hold the discovery and“Computational development catalysts great promise for the future.” Similar views have field, they added “Computational approaches for the discovery and development of catalysts hold great expressed promise for future.” views have been expressed in other reports [12–14]. been in the other reportsSimilar [12–14]. great for the future.” have in other reports [12–14]. of In promise this perspective, perspective, we shall presentviews mainly ourbeen ownexpressed results the application In this we shallSimilar present mainly our own results highlighting highlighting the application of DFT DFT In this perspective, we shall present mainly our own results highlighting the application DFT calculations to toplan planand andaccomplish accomplish successfully some reactions of annelated 1,3-azaphospholes. It calculations successfully some reactions of annelated 1,3-azaphospholes. Itofwill calculations to plan and accomplish successfully some reactions of annelated 1,3-azaphospholes. will also be illustrated computational calculations helped resolvean anexperimental experimentallog log jam jam and and It also be illustrated howhow computational calculations helped toto resolve will also be illustrated how calculations helped to resolve experimental log jam and paved the way way for the the use use of computational catalysts, leading leading to aa successful successful reaction andan asymmetric synthesis. paved the for of catalysts, to reaction and asymmetric synthesis. paved the way for the use of catalysts, leading to a successful reaction and asymmetric synthesis. 2. Synthesis of Annelated 1,3-Azaphospholes 2. Synthesis of Annelated 1,3-Azaphospholes Before coming to the reactions, it will be appropriate to describe briefly two general methods of Before coming to1,3-azaphospholes the reactions, it will bewe appropriate to describe briefly two general methods of synthesis of annelated that developed. that we developed. synthesis of annelated 1,3-azaphospholes that we developed. 2.1. [4+1] Cyclocondensation Cyclocondensation Method Method 2.1. [4+1] 2.1. [4+1] Cyclocondensation Method The The reaction reaction of of 2-ethyl-1-phenacylpyridinium 2-ethyl-1-phenacylpyridinium bromide bromide (1) (1) with with PCl PCl33 in in the the presence presence of of Et Et33N N at at The reaction of 2-ethyl-1-phenacylpyridinium bromide (1) with PCl 3 in the presence of Et 3 N room temperature (r.t.) afforded 3-benzoyl-1-methyl-1,3-azaphospholo[5,1-a]pyridine (2) in very good room temperature (r.t.) afforded 3-benzoyl-1-methyl-1,3-azaphospholo[5,1-a]pyridine (2) in veryat room temperature 3-benzoyl-1-methyl-1,3-azaphospholo[5,1-a]pyridine (2)to in very yield (Scheme 1). The general skeleton was named as 2-phosphaindolizine perceiving it to result from good yield (Scheme 1).(r.t.) The afforded general skeleton was named as 2-phosphaindolizine perceiving it result good (Scheme 1).atThe wasnucleus named as 2-phosphaindolizine perceiving it to result afrom CH/P exchange at the 2-position ofskeleton indolizine [15]. a yield CH/P exchange thegeneral 2-position of indolizine nucleus [15]. from a CH/P exchange at the 2-position of indolizine nucleus [15]. Et3N, MeCN + + PCl3 N+ N Et3r.t. N, _ MeCN _ Ph + + N H2C CH3 + PCl3 r.t. H2C N _ _ -3 Et3NHCl, -Et3NHBr Me Ph + P + Ph H2C H2CBrCH 3 -3 Et3NHCl, -Et3NHBr Me O P Ph O 1 Br 2 O O 1 2 Scheme 1. Synthesis of 2-phosphaindolizine derivative. Scheme 1. Synthesis of 2-phosphaindolizine derivative. Scheme 1. Synthesis of 2-phosphaindolizine derivative.
Later, this synthetic strategy made accessible not only variously substituted 2Later, strategy made variously Later, this this synthetic synthetic made accessible accessible not not only only 3-aza-2-phosphaindolizines variously substituted substituted 2-2phosphaindolizines [15–17], butstrategy also 1-aza-2-phosphaindolizines [18], phosphaindolizines [15–17], butbut alsoalso 1-aza-2-phosphaindolizines [18], 3-aza-2-phosphaindolizines [19], phosphaindolizines [15–17], 1-aza-2-phosphaindolizines [18],be3-aza-2-phosphaindolizines [19], and 1,3-diaza-2-phosphaindolizines [20]. Besides, the method could extended to the synthesis and 1,3-diaza-2-phosphaindolizines [20]. Besides, the method could be extended to the synthesis and 1,3-diaza-2-phosphaindolizines [20]. Besides, the[21], method could be extended the synthesis of[19], 1,3-azaphospholes annelated to pyrazine [15], thiazole benzothiazole [21], andtooxazole [22] ofof1,3-azaphospholes annelated totopyrazine [15], thiazole [21], benzothiazole [21], and oxazole [22] 1,3-azaphospholes annelated pyrazine [15], thiazole [21], benzothiazole [21], and oxazole [22] skeletons also (Scheme 2). These reactions have been compiled in a mini-review [23]. skeletons also (Scheme 2). These reactions have been compiled in a mini-review [23]. skeletons also (Scheme 2). These reactions have been compiled in a mini-review [23]. + PX3 N+ X PX3 -4HX N YH2 X H2Z -4HX YH2 H2Z 3 Y,Z =3RC,N Y,Z = RC,N
N Z N Y ZP Y 4P 4
Scheme 2. Synthesis of annelated 1,3-azaphospholes. Scheme Synthesis annelated 1,3-azaphospholes. Scheme 2. 2. Synthesis ofof annelated 1,3-azaphospholes.
2.2. Synthesis Involving Tandem Pyridinium Ylide generation, Disproportionation, and 1,52.2. Synthesis Involving Tandem Pyridinium Ylide generation, Disproportionation, and 1,5Electrocyclization 2.2. Synthesis Involving Tandem Pyridinium Ylide generation, Disproportionation, and 1,5-Electrocyclization Electrocyclization We obtained obtained 1,3-bis(alkoxycarbonyl)-1,3-azaphospholo[5,1-a]pyridines 1,3-bis(alkoxycarbonyl)-1,3-azaphospholo[5,1-a]pyridines (9) serendipitously on on We (9) serendipitously We(alkoxycarbonyl)methylpyridinium obtained 1,3-bis(alkoxycarbonyl)-1,3-azaphospholo[5,1-a]pyridines (9) reaction serendipitously on reacting bromide with Et 3 N and PCl 3 . The involved reacting (alkoxycarbonyl)methylpyridinium bromide with Et3 N and PCl3 . The reaction involved reacting (alkoxycarbonyl)methylpyridinium bromide with Et 3 N and PCl 3 . The reaction involved tandem pyridinium alkoxycarbonyl-dichlorophosphinomethylide (6) generation, disproportionation, tandem pyridinium and alkoxycarbonyl-dichlorophosphinomethylide (6) (Scheme generation, disproportionation, 1,5-electrocyclization, 1,2-elimination to furnish the final product 3) [24,25]. 1,5-electrocyclization, and 1,2-elimination to furnish the final product (Scheme 3) [24,25].
Molecules 2018, 23, 1283
3 of 13
tandem pyridinium alkoxycarbonyl-dichlorophosphinomethylide (6) generation, disproportionation, 1,5-electrocyclization, and 1,2-elimination to furnish the final product (Scheme 3) [24,25]. 3 of 13 Molecules 2018, 23, x FOR PEER REVIEW R2
(i) Et3N - + + (ii) PCl3 Br N N r.t., CH Cl 1 2 2 R1O C C RMolecules O2C 2018, CH223, x FOR PEER REVIEW 2 PCl2 2 R2 5 6R (i) Et N 3
- + Br N CH2
R1O2C
(ii) PCl3 r.t., CH2Cl2
R1O2C
R2
+ N C
1 PR2 CO2R
9
C P R1O2R C2
R1O C
CO R1
R2
R1O2C
-NC5H4
R2.HCl
-NC5H4
R2.HCl
+ N C-
+ N
-PCl3
C P R2 CO2R1 7
N R1O2C R2 P
R1O C
3 of 13
CO2R1
7
Disprop.
2 = H, Me, Et, Bu R1 = Me, Et; R1,2-Elim.
N
+ N C-
+ N
-PCl3
1,2-Elim.
N R1O2C
Disprop.
PCl2 6
5
R2
R2
R2
N
H+ NC5H4R2 CO2R1
H+ 8 NC5H4R2
2 2 P 1 P Scheme 3. Synthesis of2 1,3-bis(alkoxycarbonyl)-1,3-azaphospholo[5,1-a]pyridines.
CO2R Scheme 3. Synthesis of 1,3-bis(alkoxycarbonyl)-1,3-azaphospholo[5,1-a]pyridines. 9
8
Subsequently, isoquinoline [26], and phenanthridine analogues of 9 could also be prepared. R1 = Me, Et; R2 = [27] H, Me, Et, Bu
Subsequently, isoquinoline [26],ofand phenanthridine [27] analogues of 9 could also be prepared. Scheme 3. Synthesis 1,3-bis(alkoxycarbonyl)-1,3-azaphospholo[5,1-a]pyridines. 3. Reactions Response
Subsequently, isoquinoline [26], and phenanthridine [27] analogues of 9 could also be prepared. 3. Reactions Response
3.1. Diels-Alder Reaction 3. Reactions Response 3.1. Diels-Alder Reaction
There have been earlier reports about the Diels-Alder (DA) reaction across the >C=Pfunctionality as dienophile [28,29]. We investigated theoretically the model DAthe reactions ethene Diels-Alder Reactionreports There3.1. have been earlier about the Diels-Alder (DA) reaction across >C=P-offunctionality and phosphaethene with 1,3-butadiene at the DFT(B3LYP/6-31+G**) level [30]. The Frontier as dienophileThere [28,29]. We earlier investigated theoretically the model DA reactions ethene and have been reports about the Diels-Alder (DA) reaction across the of >C=PMolecular Orbitals of the reactants are showntheoretically in Figure 1.the model DA reactions of ethene functionality as(FMO) dienophile [28,29]. We investigated
phosphaethene with 1,3-butadiene at the DFT(B3LYP/6-31+G**) level [30]. The Frontier Molecular and phosphaethene with 1,3-butadiene at the DFT(B3LYP/6-31+G**) level [30]. The Frontier Orbitals (FMO) of the reactants are shown in Figure 1. Molecular Orbitals (FMO) of the reactants are shown in Figure 1.
Figure 1. The HOMO-LUMO energy differences between FMOs of phosphaethene, 1,3-butadiene,
Figure 1. HOMO-LUMO The HOMO-LUMO energy differences between FMOsofofphosphaethene, phosphaethene, 1,3-butadiene, 1,3-butadiene, and Figure 1. The energy differences between FMOs and ethene calculated at the B3LYP/6-31+G** level. and ethene calculated at the B3LYP/6-31+G** ethene calculated at the B3LYP/6-31+G** level.level. It may be noted that the HOMOdiene–LUMOphosphaethene gap (4.55 eV) is much lower than the gap
ItHOMO may be noted that the HOMOdiene–LUMOphosphaethene gap (4.55 eV) is much lower than the gap diene–LUMOethene (6.32 eV). Furthermore, the activation energy barriers (ΔEact.) for the two DA It may be noted that the HOMO gap barriers (4.55 eV) is much lower diene –LUMO HOMO diene–LUMO ethene (6.32 eV). Furthermore, thephosphaethene activation energy (ΔEact.) for the two than DA the reactions are shown in Scheme 4 [30]. gap HOMO –LUMO (6.32 eV). Furthermore, the activation energy barriers (∆Eact.) for the reactions are shown inethene Scheme 4 [30]. diene
two DA reactions are shown in Scheme 4 [30].
Molecules 2018, 23, 1283
4 of 13
Molecules 2018, 23, x FOR PEER REVIEW
4 of 13
Eact Erxn H
H
H
H
kcal/mol
+
TS
kcal/mol
28.00 -29.19
Molecules 2018, 23, x FOR PEER REVIEW
H P H
4 of 13
H H
+
TS
H +
H
P
TS
Eact 12.41 Erxn -29.71 endo: kcal/mol exo: 14.48 -32.15
kcal/mol
28.00 -29.19
H
H
Scheme 4. Model DA reactions of ethene and phosphaethene with 1,3-butadiene computed at Scheme 4. Model HDA reactions of ethene and phosphaethene with 1,3-butadiene computed at H B3LYP/6-31+G** in the gas phase. endo: 12.41 -29.71 B3LYP/6-31+G** in the Pgas phase. + TS H
exo: 14.48 -32.15
P
Encouraged by these theoretical results, we attemptedHDA reactions of 1,3-azaphospholo[5,1Encouraged by these theoretical results, wewith However, attempted DAgreat of Scheme 4. Model DA reactions of ethene and 1,3-butadieneto computed atreactions a]pyridines, i.e., 2-phosphaindolizines prepared byphosphaethene two methods. our surprise, the B3LYP/6-31+G** in the gas phase. 1,3-azaphospholo[5,1-a]pyridines, i.e., 2-phosphaindolizines prepared by two methods. However, compounds 3-ethoxycarbonyl-1-methyl-1,3-azaphospholo[5,1-a]pyridine 10a (10, Z=Me, to our great surprise, the compounds 3-ethoxycarbonyl-1-methyl-1,3-azaphospholo[5,1-a]pyridine 1 2 3 R =R =R =H) Encouraged obtained by through [4+1] cyclocondensation and of1,3-bis(ethoxylcarbonyl)-1,3these theoretical results, we attempted DA reactions 1,3-azaphospholo[5,12 =R3 =H) 10a (10,a]pyridines, Z=Me,i.e., 2-phosphaindolizines R1 =R10b obtained through [4+1]remarkable and 3=H) showed prepared by1=R two2=R methods. However, to ourcyclocondensation great surprise, the azaphospholo[5,1-a]pyridine (10, Z=CO 2Et, R difference in their 1 =R2 =R3 =H) showed 1,3-bis(ethoxylcarbonyl)-1,3-azaphospholo[5,1-a]pyridine 10b (10, Z=CO Et, R compounds 3-ethoxycarbonyl-1-methyl-1,3-azaphospholo[5,1-a]pyridine 10a (10, Z=Me, 2 reactivity with 2,3-dimethylbuta-1,3-diene (DMB). It was found that 3-(ethoxycarbonyl)-1-methyl2=R3=H) R1=R obtained through [4+1] cyclocondensation and 1,3-bis(ethoxylcarbonyl)-1,3remarkable difference in their reactivity with 2,3-dimethylbuta-1,3-diene (DMB). It was found that 1,3-azaphospholo[5,1-a]pyridine did not2Et, undergo reaction with DMB, even azaphospholo[5,1-a]pyridine (10a) 10b (10, Z=CO R1=R2=R3DA =H) showed remarkable difference inupon their boiling in 3-(ethoxycarbonyl)-1-methyl-1,3-azaphospholo[5,1-a]pyridine (10a) did not undergo DA reaction with toluene inreactivity the presence of sulfur (used to oxidize phosphorus atom the initially formed cycloadduct with 2,3-dimethylbuta-1,3-diene (DMB). It was found that of 3-(ethoxycarbonyl)-1-methylDMB, even upon boiling in toluene in the presence of sulfur (used to oxidize phosphorus atom of 1,3-azaphospholo[5,1-a]pyridine (10a)direction, did not undergo reaction with DMB, even upon boiling in to push the reaction in the forward see DA later) [31]. But, 1,3-bis(ethoxylcarbonyl)-1,3the initially formed to (used pushto the reaction in the forward direction, see later) [31]. (r.t.). But, toluene in thecycloadduct presence of sulfur oxidize phosphorus atom of the initially azaphospholo[5,1-a]pyridine (10b) underwent DA reaction with DMB atformed roomcycloadduct temperature to push the reaction in the forward direction, see later)(10b) [31]. underwent But, 1,3-bis(ethoxylcarbonyl)-1,31,3-bis(ethoxylcarbonyl)-1,3-azaphospholo[5,1-a]pyridine DA reaction with DMB However,azaphospholo[5,1-a]pyridine the reaction at room temperature ( ca DA 25 °C) waswith veryDMB slow could be completed (δ 31at P (10b) underwent reaction at and room (r.t.). ◦ C) temperature room temperature (r.t.). However, the reaction at room temperature (ca 25 was very slow and could 31 NMR = 14.1 ppm)the after heating at temperature reflux temperature in chloroform four days. On out However, reaction at room ( ca 25 °C) was very slow and for could be completed (δ carrying P 31 P NMR = 14.1 ppm) after heating at reflux temperature in chloroform for four days. be NMR 14.1 ppm) afterofheating reflux temperature for fouratdays. On4carrying out Thus, the thecompleted reaction in(δ=the presence sulfurator methyl iodide,initchloroform was complete r.t. in days [31]. On out the in of thesulfur presence of iodide, sulfurit or was[31]. complete the reaction the presence or methyl wasmethyl completeiodide, at r.t. in it 4 days Thus, theat r.t. in 4 DA carrying reactions of inreaction 1,3-bis(ethoxycarbonyl)-1,3-azaphospholo[5,1-a]pyridines and also of an DA reactions of 1,3-bis(ethoxycarbonyl)-1,3-azaphospholo[5,1-a]pyridines and also of anand also of days [31]. Thus, the DA reactions of 1,3-bis(ethoxycarbonyl)-1,3-azaphospholo[5,1-a]pyridines isoquinoline analogue with with DMB and couldbebe accomplished successfully (Scheme 5) isoquinoline analogue DMB andisoprene isoprene could accomplished successfully (Scheme 5) an isoquinoline analogue with DMB and isoprene could be accomplished successfully (Scheme 5) [31,32]. [31,32]. [31,32]. 3
2
R3R
RR2
R1
R4
Me
R1 + Me +
N
EtO2CN P Z 10 Z EtO2C P R1, R2 = H, H; 10 3 R = H, Et R1, R2 = H, H; S8, CHCl3, r.t. R3 = H, Et
S8, R3 CHCl R23,
R4
11 Z = CO2Et 11 R4 = Me, H
Z = CO2EtMeI R4 = Me, HCHCl3, r.t. MeI R3
r.t.
R2CHCl , 3
R1 3 EtOR2C
N
R2 P
R4
NMe
S
R EtO2C
CO2Et
R1
r.t.
1
R4
N 3 R+ P
R2 CO2Et Me I R1
N
EtO2CMe 12 CO Et 13+ CO2Et 2 P P Me I S Scheme 5. DA reactions of 1,3-bis-(ethoxycarbonyl)-1,3-azaphospholo[5,1-a]pyridines with 2,35. DA reactions of 1,3-bis-(ethoxycarbonyl)-1,3-azaphospholo[5,1-a]pyridines dimethyl-1,3-butadiene 4 R4 (DMB ) and isoprene. EtO2C
Scheme with R 2,3-dimethyl-1,3-butadiene (DMB ) and isoprene. Me with complete stereo- and regioselectivity, Me All the reactions occurred except the reaction of 1,312 bis(ethoxycarbonyl)-1,3-azaphospholo[5,1-a]isoquinoline (14) with13 isoprene in the presence of
All the reactions occurred with complete stereothe reaction of methyl which afforded two regioisomers, the majorand (62%)regioselectivity, being 15 (Scheme 6)except [32]. Scheme 5. iodide DA reactions of 1,3-bis-(ethoxycarbonyl)-1,3-azaphospholo[5,1-a]pyridines with 2,31,3-bis(ethoxycarbonyl)-1,3-azaphospholo[5,1-a]isoquinoline (14) with isoprene in the presence of dimethyl-1,3-butadiene (DMB ) and isoprene. methyl iodide which afforded two regioisomers, the major (62%) being 15 (Scheme 6) [32]. All the reactions occurred with complete stereo- and regioselectivity, except the reaction of 1,3bis(ethoxycarbonyl)-1,3-azaphospholo[5,1-a]isoquinoline (14) with isoprene in the presence of methyl iodide which afforded two regioisomers, the major (62%) being 15 (Scheme 6) [32].
Molecules 2018, 23, 1283 Molecules Molecules 2018, 2018, 23, 23, xx FOR FOR PEER PEER REVIEW REVIEW
+ +
N N EtO EtO22C C
Me Me
5 of 13 55 of of 13 13
N N
EtO EtO22C C
MeI MeI CHCl CHCl3,, r.t. r.t.
+ + CO CO22Et Et P P Me Me II--
3
CO CO22Et Et
P P
Me Me 15 15
Molecules 2018, 23, x FOR PEER REVIEW
14 14
EtO EtO22C C
+ +
N N + + CO CO22Et Et P P Me Me II--
Me Me
62% 62%
38% 38%5 of 13 16 16
Scheme 6. Regioselectivity in DA of Scheme Regioselectivity in the the DA reaction reaction of 1,3-bis(ethoxycarbonyl)-1,3-azaphospholo[5,11,3-bis(ethoxycarbonyl)-1,3-azaphospholo[5,1Scheme 6.6.Regioselectivity in the DA reaction of 1,3-bis(ethoxycarbonyl)-1,3-azaphospholo[5,1-a]isoquinoline. a]isoquinoline. a]isoquinoline. N
N EtO2C EtO2C + + CO2Et 3 + CO2Et MeI 1 2 4 P P confirmed + The structure of the cycloadduct 12 (R CH=CH-CH=CH-; R =R44=H) =H)was was by single 11R 22= 33=R Me N The structure of the cycloadduct 12 (R R = CH=CH-CH=CH-; R by Me CHCl r.t. = CH=CH-CH=CH-; R =R =H) was confirmed The structure of the cycloadduct 12 (R3, R confirmed by single single I crystal X-rayEtO diffraction studies [32]. That the [2+4]cycloaddition of Me annelated I1,3-azaphosphole CO 2C 2Et crystal studies [32]. 1,3-azaphosphole P crystal X-ray X-ray diffraction diffraction studies [32]. That That the the [2+4]cycloaddition [2+4]cycloaddition of of annelated annelated 1,3-azaphosphole with with Me atom 62%of the initially 38% formed cycloadduct with DMB precedes the oxidation of the phosphorus DMB precedes the oxidation of the phosphorus atom of the initially formed cycloadduct by sulfur DMB precedes the 14 oxidation of the phosphorus atom of by sulfur or or 15the initially formed cycloadduct 16 Me
by sulfur or methyl iodide could be proved unambiguously by carrying out the DA of reaction methyl could be by methyl iodide iodideScheme could be proved proved inunambiguously unambiguously by carrying carrying out out the the DA DA reaction reaction of 1,4,21,4,26. Regioselectivity the DA reaction of 1,3-bis(ethoxycarbonyl)-1,3-azaphospholo[5,1of 1,4,2-diazaphospholo[4-a]pyridine derivative 17 in the presence of methyl iodide, when diazaphospholo[4-a]pyridine derivative diazaphospholo[4-a]pyridine derivative 17 17 in in the the presence presence of of methyl methyl iodide, iodide, when when N-methylated N-methylated a]isoquinoline. N-methylated product 18 (R=Me) was formed (Scheme 7) [33]. It has been reported that [1,4,2] product product 18 18 (R=Me) (R=Me) was was formed formed (Scheme (Scheme 7) 7) [33]. [33]. It It has has been been reported reported that that [1,4,2]diazaphospholo[4[1,4,2]diazaphospholo[41 2 3 4 The structure of the cycloadduct 12 (R R = CH=CH-CH=CH-; R =R =H) was confirmed by single diazaphospholo[4-a]pyridines did not react with methyl iodide [34]. a]pyridines a]pyridines did did not not react react with with methyl methyl iodide iodide [34]. [34]. crystal X-ray diffraction studies [32]. That the [2+4]cycloaddition of annelated 1,3-azaphosphole with DMB precedes the oxidation of the phosphorus atom of the initially formed cycloadduct by sulfur or Me Meout the DA reactionMe Meof 1,4,2methyl iodide could be proved unambiguously by carrying diazaphospholo[4-a]pyridine derivative 17 in the presence of methyl iodide, when N-methylated product 18 (R=Me) Me Me was formed (Scheme 7) [33]. It has been reported that [1,4,2]diazaphospholo[4N N a]pyridines did not react with methyl iodide [34]. H N N + +
+ +
N N H H
N N
P P
R R
Me Me
R R
Me R
17 17
Me
+
N
H MeI CHCl3, r.t.
N
+ H H N N Me Me + + P Me P I-I
H
MeI MeI CHCl CHCl33,, r.t. r.t.
N
Me
Me MeN
+ N Me + H
Me MeP 18 18 I
+ N N Me Me P P I-I
+
P
R N Me R 19 I19
R Me P of 7-methyl-1,4,2-diazaphospholo[4,5-a]pyridine Scheme reaction in presence of Scheme 7. 7. DADA reaction of 7-methyl-1,4,2-diazaphospholo[4,5-a]pyridine in the thein presence of methyl methyl 7. DA reaction of 7-methyl-1,4,2-diazaphospholo[4,5-a]pyridine the presence of Me R iodide. iodide. methyl iodide. 17 18 19 H
Scheme 7. DA reaction of 7-methyl-1,4,2-diazaphospholo[4,5-a]pyridine in the presence of methyl
The observed The regioselectivity regioselectivity observed in in the the reaction reaction of of 1,3-bis(ethoxycarbonyl)1,3-bis(ethoxycarbonyl)- 1,3-azaphospholo[5,11,3-azaphospholo[5,1iodide. The regioselectivity observed in the reaction of 1,3-bis(ethoxycarbonyl)- 1,3-azaphospholo[5,1-a] a]pyridine with isoprene was investigated theoretically at the DFT (B3LYP/6-311++G**//B3LYP/6a]pyridine with with isoprene was investigated theoretically at the pyridine was investigated theoretically at DFT the (B3LYP/6-311++G**//B3LYP/6DFT (B3LYP/6-311++G**// The isoprene regioselectivity observed in the reaction of 1,3-bis(ethoxycarbonyl)1,3-azaphospholo[5,131G**) 31G**) level. level. Following Following model model reactions reactions were were computed computed (Figure (Figure 2). 2). For For studying studying the the solvent solvent effect, effect, B3LYP/6-31G**) Following model reactions were computed (Figure 2). For studying the a]pyridinelevel. with isoprene was investigated theoretically at the DFT (B3LYP/6-311++G**//B3LYP/6single point energy was computed using gas phase-optimized geometry at the B3LYP/6-311++G** single energy computed using gas B3LYP/6-311++G** 31G**) level. was Following model reactions were phase-optimized computed (Figure 2). geometry For studyingatthethe solvent effect, solventpoint effect, single point energy was computed using gas phase-optimized geometry at the level polarizable continuum model (PCM). point energy was computed gas phase-optimized geometry at the B3LYP/6-311++G** level with with single polarizable continuum modelusing (PCM). B3LYP/6-311++G** level with polarizable continuum model (PCM). level with polarizable continuum model (PCM).
+ N+
N N MeO MeO22C C
MeO P 2C
P
+
P P
N N N MeO 2C MeO MeO 2C 2C P
Me 21 Me Me
20
+ +P
Me
+ CO2Me
CO CO22Me Me 20
Eact (kcal/mol)
Eact (kcal/mol) Eact (kcal/mol)
22.8 22.8 22.8 (in gas phase) CO Me (in gas phase) P CO Me 2 (in gas phase) CO Me 2 P 2 22.9 22.9 22.9 (in CHCl3(in ) (in CHCl CHCl33))
21 21
N
MeO MeO22C C
TS A TS TS AA
CO MeCO Me CO P2 2Me 2
20 20
MeO2C N N
Me Me Me
Me Me
TSB
TS TSB B
MeO2C
25.2 (in gas phase) 25.2 25.2 25.3 (in CHCl3(in ) gas (in gas phase) phase)
N
PN N CO2Me MeO MeO22C C CO Me P CO22Me Me P 22
25.3 25.3 (in (in CHCl CHCl33))
Figure 2. Model DA reactions computed at B3LYP/6-311++G**//B3LYP/6-31G**.
Me Me Figure 2. Model DA reactions computed at B3LYP/6-311++G**//B3LYP/6-31G**. 20 20
22 22
Figure Figure 2. 2. Model Model DA DA reactions reactions computed computed at at B3LYP/6-311++G**//B3LYP/6-31G**. B3LYP/6-311++G**//B3LYP/6-31G**.
Molecules 2018, 23, 1283
6 of 13
Molecules 2018, 23, x FOR PEER REVIEW
6 of 13
−∆E/RT The A difference of The regioselectivity regioselectivity was was calculated calculated by by using using the the formula, formula, kk11/k /k22==ee−∆E/RT [35].[35]. A difference of 2.4 2.4 kcal/mol in the activation energies corresponds to only 2% regioselectivity. It is obvious that the kcal/mol in the activation energies corresponds to only 2% regioselectivity. It is obvious that the observed observed regioselectivity regioselectivity of of 100% 100% cannot cannot be be accounted accounted for for at at this this theoretical theoreticallevel. level. We investigated the reason for the difference in the dienophilic reactivities of 3-ethoxycarbonyl3-ethoxycarbonyl We investigated the reason for the difference in the dienophilic reactivities of -1-methyl-1,3-azaphospholo[5,1-a]pyridine (10a) and 1,3-bis(ethoxycarbonyl)-1,3-azaphospholo[5,1-a] 1-methyl-1,3-azaphospholo[5,1-a]pyridine (10a) and 1,3-bis(ethoxycarbonyl)-1,3-azaphospholo[5,1pyridine (10b), theoretically. ForFor thisthis purpose, following fourfour model reactions were computed at the a]pyridine (10b), theoretically. purpose, following model reactions were computed at DFT (B3LYP/6-31G**) level (Scheme 8) [36]. the DFT (B3LYP/6-31G**) level (Scheme 8) [36].
Relative Energies
Model React. No. +
N
1. H
P
TS
H
H
P
23
N
2. MeO2C
TS
MeO2C
H
H
P
TS
H
P
27
N
4. MeO2C
P 29
+0.76
H
29.49
+8.17
24.55
+5.05
22.43
+0.92
N
CO2Me
P
27.52
26
+
N
H
N
25
3.
Erxn (kcal/mol)
24
+
P
Eact (kcal/mol)
N
CO2Me
28
+ CO2Me
TS
MeO2C
N P
CO2Me
30
Scheme 8. 8. Model Model DA DA reactions reactions computed computed at atthe theB3LYP/6-31G** B3LYP/6-31G** level. level. Scheme
As expected, the activation energy barrier for the DA reaction with 1,3-bis(methoxycarbonyl)As expected, the activation energy barrier for the DA reaction with 1,3-bis(methoxycarbonyl)-1, 1,3-azaphospholo[5,1-a]pyridine (29) is lowest. However, it is interesting to find that the activation 3-azaphospholo[5,1-a]pyridine (29) is lowest. However, it is interesting to find that the activation barrier for barrier for the DA reaction with 3-methoxycarbonyl-1,3-azaphospholo[5,1-a]pyridine (25) is highest the DA reaction with 3-methoxycarbonyl-1,3-azaphospholo[5,1-a]pyridine (25) is highest (29.49 kcal/mol), (29.49 kcal/mol), higher than even for the DA reaction with the unsubstituted 1,3-azaphospholo[5,1higher than even for the DA reaction with the unsubstituted 1,3-azaphospholo[5,1-a]pyridine (23). It is a]pyridine (23). It is rather an unexpected effect of an electron-withdrawing group (EWG). To look rather an unexpected effect of an electron-withdrawing group (EWG). To look into the real cause of this into the real cause of this unusual effect, we carried out the natural bond orbital (NBO) analysis unusual effect, we carried out the natural bond orbital (NBO) analysis calculations, which revealed the calculations, which revealed the existence of an interesting phenomenon: due to conjugation of the existence of an interesting phenomenon: due to conjugation of the lone pair of the bridgehead nitrogen lone pair of the bridgehead nitrogen in 23,negative charge density is effectively transferred to the in 23, negative charge density is effectively transferred to the five-membered azaphosphole ring, which five-membered azaphosphole ring, which is further enhanced by the presence of an EWG at the 3is further enhanced by the presence of an EWG at the 3-position. It is shown in Figure 3A. It makes the position. It is shown in Figure 3A. It makes the >C=P- functionality electron-rich, which fails to react >C=P- functionality electron-rich, which fails to react with electron-rich diene, DMB. However, a CO2 Me with electron-rich diene, DMB. However, a CO2Me group at the 1-position acts as a trap for the group at the 1-position acts as a trap for the negative charge, the latter being delocalized over to the negative charge, the latter being delocalized over to the carbonyl group. It is shown in Figure 3B. carbonyl group. It is shown in Figure 3B. Thus in this case, the >C=P- functionality remains unaffected Thus in this case, the >C=P- functionality remains unaffected and is able to undergo the DA reaction and is able to undergo the DA reaction with DMB. with DMB.
Molecules 2018, 23, 1283 Molecules 2018, 23, x FOR PEER REVIEW Molecules 2018, 23, x FOR PEER REVIEW
7 of 13 7 of 13 7 of 13
Figure 3.3.Natural Natural Bond Orbital (NBO) interactions in 3-methoxycarbonyl-1,3-azaphospholo[5,1Figure Bond Orbital (NBO) interactions in 3-methoxycarbonyl-1,3-azaphospholo[5,1-a]pyridine Figure 3. Natural Bond Orbital (NBO) interactions in 3-methoxycarbonyl-1,3-azaphospholo[5,1a]pyridine (A) and 1,3-bis(methoxycarbonyl)1,3-azaphospholo[5,1-a]pyridine (B1 and B2) order with (A) and 1,3-bis(methoxycarbonyl)- 1,3-azaphospholo[5,1-a]pyridine (B1 and B2 ) with second a]pyridine (A) and 1,3-bis(methoxycarbonyl)1,3-azaphospholo[5,1-a]pyridine (B1 and B2) with second order perturbation energies (in parenthesis). perturbation energies (in parenthesis). second order perturbation energies (in parenthesis).
3.2. Lewis Acid Catalyzed Diels-Alder Reaction 3.2. Acid Catalyzed 3.2. Lewis Lewis Acid Catalyzed Diels-Alder Diels-Alder Reaction Reaction Then, we explored the possibility of using a Lewis acid as catalyst to accomplish DA reaction Then, we using a Lewis acidacid as catalyst to accomplish DA reaction with weexplored exploredthe thepossibility possibilityofof using a Lewis as catalyst accomplish DAfollowing reaction with Then, 3-(alkoxycarbonyl)-1-methyl-1,3-azaphospholo[5,1-a]pyridines andtocomputed the 3-(alkoxycarbonyl)-1-methyl-1,3-azaphospholo[5,1-a]pyridines and computed the following model with andstudying computed following model3-(alkoxycarbonyl)-1-methyl-1,3-azaphospholo[5,1-a]pyridines reactions at the DFT (B3LYP/6-31+G**) level (Scheme 9) [37]. For the the solvent effect, reactions at the DFT (B3LYP/6-31+G**) level (Scheme 9) [37].9)For studying the solvent effect, single model reactions at the DFT (B3LYP/6-31+G**) level (Scheme [37]. For studying the solvent effect, single point energy was computed using gas phase-optimized geometry with PCM solvation model. point energy was computed using gas phase-optimized geometry with PCM solvation model. single point energy was computed using gas phase-optimized geometry with PCM solvation model.
X O X O MeO MeO X X Nil Nil Nil AlCl3 AlCl33 AlCl3 O O MeO MeO
N N Z Z
Z Z CH CH P P CH CH P P
31 31
+ + Me Me
Eact (kcal/mol) (kcal/mol) EactCH 2Cl2
TS TS
X O X O MeO MeO
N N Z Z
Me Me
32 32
CH 45.68 2Cl2 45.68 30.80 30.80 46.77 46.77 33.43 33.43
O N N Eact (kcal/mol) = 22.57 + TS O N N E CH + TS (kcal/mol) = 22.57 Me 2Cl2 MeO Me act P P CH Cl Me 2 2 MeO P P AlClMe AlCl3 3 AlCl3 AlCl3 33 34 33 34 Scheme 9. Model DA reactions computed at the B3LYP/6-31+G** level. Scheme 9. 9. Model Model DA DA reactions reactions computed computed at at the the B3LYP/6-31+G** B3LYP/6-31+G** level. Scheme level.
There was another surprise in store for us. When the catalyst, AlCl 3, is coordinated to the Theregroup was another surprise in store for us. When the catalyst, 3, is coordinated to the carbonyl of indolizine or 1,3-azaphospholo[5,1-a]pyridine, the AlCl activation energy barrier is carbonyl group of indolizine or 1,3-azaphospholo[5,1-a]pyridine, the activation energyreveal barrier is increased further as compared with that in the absence of the catalyst. The NBO studies that increased further as compared with that in the absence of the catalyst. The NBO studies reveal that
Molecules 2018, 23, 1283
8 of 13
There was another surprise in store for us. When the catalyst, AlCl3 , is coordinated to the carbonyl group of indolizine or 1,3-azaphospholo[5,1-a]pyridine, the activation energy barrier is Molecules 88 of Molecules 2018, 2018, 23, 23, xx FOR FOR PEER PEER REVIEW REVIEW of 13 13 increased further as compared with that in the absence of the catalyst. The NBO studies reveal that carbonyl group enhances transfer of charge in coordination ofAlCl AlCl33 3with withthe the carbonyl group enhances transfer ofnegative the negative charge in cases, these coordination of of AlCl with the carbonyl group enhances transfer of the the negative charge in these these cases, making the >C=C< and >C=Pfunctionality in indolizine and 1,3-azaphospholo[5,1-a]pyridine, cases, making the >C=C< and >C=Pfunctionality in indolizine and 1,3-azaphospholo[5,1-a]pyridine, making the >C=C< and >C=P- functionality in indolizine and 1,3-azaphospholo[5,1-a]pyridine, respectively, electron-rich. This appears to be the reason why no DA reaction across the >C=C< respectively, moiety respectively,electron-rich. electron-rich.This Thisappears appearsto tobe bethe thereason reasonwhy whyno no DA DA reaction reactionacross acrossthe the>C=C< >C=C< moiety moiety of a catalyst. of indolizine has been reported even in the presence of a catalyst. of indolizine has been reported even in the presence of a catalyst. In the P atom. In 2-phosphaindolizine, 2-phosphaindolizine, however, however, there there is is another another site site for for the the coordination coordination of of AlCl AlCl33:: the the PP atom. atom. of to energy It may be noted that on coordination the catalyst the P atom, the activation barrier It may be noted that on coordination of the catalyst to the P atom, the activation energy barrier is is to 22.57 kcal/mol lowered Cl222))) only. only. These These results results indicated indicated that that in lowered remarkably, remarkably, to to 22.57 22.57 kcal/mol kcal/mol (in (in CH CH222Cl Cl only. These results indicated that in contrast contrast to to indolizine, >C=Pindolizine, DA DA reactions reactions across across the the >C=P>C=P- moiety moiety of of 2-phosphaindolizine 2-phosphaindolizine having having EWG EWG at at 3-position 3-position be possible in the presence of However, in this a only should a Lewis acid catalyst. context only should be possible in the presence of a Lewis acid catalyst. However, in this context a question question arises: arises: if if coordination coordination of of AlCl AlCl33 to to the the carbonyl carbonyl group group is is thermodynamically thermodynamically more more favorable favorable than than its its 3.43 kcal/mol, coordination why should should it coordinate to to the the P coordination to to the the P P atom atom by by 3.43 3.43 kcal/mol, kcal/mol, why why should itit coordinate coordinate to the P atom atom at at all? all? In In our our solution, an equilibrium may exist between the C=O→ AlCl → opinion, opinion, in in solution, solution, an an equilibrium equilibriummay mayexist existbetween betweenthe theC=OC=O-→ → AlCl AlCl333 and and P P→ → AlCl AlCl33 coordinated coordinated species, making a small concentration of the latter available that undergoes DA with the of of thethe latter available thatthat undergoes DA reaction with the diene species, making makingaasmall smallconcentration concentration latter available undergoes DA reaction reaction with the diene to produce irreversibly the cycloadduct, thereby shifting the equilibrium to the right (Scheme 10). to produce irreversibly the cycloadduct, thereby shifting the equilibrium to the right (Scheme 10). diene to produce irreversibly the cycloadduct, thereby shifting the equilibrium to the right (Scheme 10).
RO RO
N N
O O AlCl AlCl33
CH Cl Me Me CH22Cl22
P P
RO RO
P P
O O
35 35
O O
N N Me Me
CH CH22Cl Cl22
N N
RO RO
P P
AlCl AlCl33 36 36
Me Me AlCl AlCl33
37 37
Scheme 10. Equilibrium between C=O and P coordinated species, and DA with the Scheme 10. 10. Equilibrium Equilibriumbetween betweenC=O C=O→→ → and P→ → coordinated species, DA reaction reaction with the Scheme and P→ coordinated species, and and DA reaction with the latter. latter. latter.
Guided by the above above results, we we accomplishedsuccessfully successfully the DA reaction of Guided Guided by by the the above results, results, we accomplished accomplished successfully the the DA DA reaction reaction of of 333-(alkoxycarbonyl/acyl)-1-methyl-2-phosphaindolizines (38) with DMB in the presence of (alkoxycarbonyl/acyl)-1-methyl-2-phosphaindolizines (38) with DMB in the presence of (alkoxycarbonyl/acyl)-1-methyl-2-phosphaindolizines (38) with DMB in the presence of ethylaluminium dichloride as aa promoter promoter (i.e., using its 11 equiv.) in dichloromethane at r.t. under ethylaluminium under ethylaluminium dichloride dichloride as as a promoter (i.e., (i.e., using using its its 1 equiv.) equiv.) in in dichloromethane dichloromethane at at r.t. r.t. under 31 31 nitrogen atmosphere (Scheme 11) [38]. The occurrence of a clean reaction was revealed by P NMR nitrogen reaction was revealed by NMR 31P nitrogen atmosphere atmosphere (Scheme (Scheme 11) 11) [38]. [38]. The The occurrence occurrence of of aa clean clean reaction was revealed by P NMR 31 P NMR chemical shifts reported for 31 signal at δ = − 16.6 to − 5.3 ppm, which is in accordance with the signal at δ = −16.6 to −5.3 ppm, which is in accordance with the P NMR chemical shifts reported 31 signal at δ = −16.6 to −5.3 ppm, which is in accordance with the P NMR chemical3 shifts reported for for the [2+4] cycloadducts obtained obtained from the the DA reaction reaction of P-W(Co) P-W(Co) complexes of of λ33-phosphinines -phosphinines [39]. the the [2+4] [2+4] cycloadducts cycloadducts obtained from from the DA DA reaction of of P-W(Co)55 complexes complexes of λ λ -phosphinines [39]. [39].
O O N N
R R
P P
O O
EtAlCl EtAlCl22,, Me Me CH CH22Cl Cl22,, r.t. r.t.
R R
38 38 38, 38, 39 39 R R
N N P P
Me Me AlEtCl AlEtCl22
39 39 a a OMe OMe
31 31P P (39) (39) -5.8 -5.8
b b
c c
d d
OEt OEt CMe CMe33 Ph Ph -5.3 -16.6 -10.1 -5.3 -16.6 -10.1
Scheme 11. in the presence of DA reaction reaction of 3-(alkoxycarbonyl/acyl)-1-methyl-2-phosphaindolizines in in thethe presence of Scheme 11. DA reactionof of3-(alkoxycarbonyl/acyl)-1-methyl-2-phosphaindolizines 3-(alkoxycarbonyl/acyl)-1-methyl-2-phosphaindolizines presence EtAlCl 2. EtAlCl 2 . of EtAlCl2 .
3.3. 3.3. Asymmetric Asymmetric Diels-Alder Diels-Alder Reaction Reaction with with 2-Phosphaindolizines 2-Phosphaindolizines The The successful successful results results described described above above motivated motivated us us to to investigate investigate asymmetric asymmetric DA DA across the >C=Pmoiety of 2-phosphaindolizines by using a chiral organoaluminium across the >C=P- moiety of 2-phosphaindolizines by using a chiral organoaluminium namely, namely, o-menthoxyaluminium o-menthoxyaluminium dichloride dichloride [40]. [40]. We computed the following model reactions We computed the following model reactions at at the the B3LYP/6-31+G* B3LYP/6-31+G* level level (Scheme (Scheme 12). 12).
reaction reaction catalyst, catalyst,
Molecules 2018, 23, 1283
9 of 13
3.3. Asymmetric Diels-Alder Reaction with 2-Phosphaindolizines The successful results described above motivated us to investigate asymmetric DA reaction across the >C=P- moiety of 2-phosphaindolizines by using a chiral organoaluminium catalyst, namely, o-menthoxyaluminium dichloride [40]. We computed the following model reactions at the B3LYP/6-31+G* level (Scheme 12). 9 of 13 Molecules 2018, 23, x FOR PEER REVIEW Molecules 2018, 23, x FOR PEER REVIEW
attack on Si face (endo) attack on Si face (endo)
Si
9 of 13
Me TS1 N TS1
Si
MeO2CN MeO2C
42 Me
Me Me P AlCl 2
41
OMenth* N C P AlCl 2 O OMe OMenth* C O OMe 40
41
N
40
attack on Si face (exo) attack on Si face (exo)
TS2
N
TS2
MeO2CN MeO2C
attack on Re face (endo) attack on Re face (endo)
Re Re
OMenth* P Me AlCl2 OMenth* P 42 AlCl2
OMenth* P Me AlCl2 OMenth* P AlCl2 43 43Me
N
TS3
MeO2CN
TS3
OMenth* P Me AlCl2 OMenth* P AlCl2
MeO2C 44 44 Scheme 12. Attack of 1,3-butadiene on Si and Re faces of the >C=P- functionality of 2-
Scheme 12. Attack of 1,3-butadiene on Si and Re faces of the >C=P- functionality of 2-phosphaindolizine phosphaindolizine complex computed at on the B3LYP/6-31+G* level. Scheme 12. Attack 1,3-butadiene Si and Re faces of the >C=P- functionality of 2complex computed at theofB3LYP/6-31+G* level. phosphaindolizine complex computed at the B3LYP/6-31+G* level.
Computational calculations reveal that attack of the diene occurs preferentially from the Si face. Furthermore, during attack on reveal the Si face, energy barrier for the exo approach is smaller Computational calculations that attack ofthe the diene occurs preferentially Si face. Computational calculations reveal thatactivation attack of diene occurs preferentially fromfrom the Sithe face. than for the endo approach by 0.3 kcal/mol. Presence ofbarrier the bulky o-menthoxy groupis Furthermore, during attack on ca. the Si face, activation energy barrier forthe theexo exoapproach approach ispossibly smaller than Furthermore, during attack on the Si face, activation energy for smaller makes the endo approach comparatively less accessible. the reactions are moderately exothermic, than for the endo approach by ca. 0.3 kcal/mol. Presence of the bulky o-menthoxy possibly for the endo approach by ca. 0.3 kcal/mol. Presence of theAll bulky o-menthoxy group group possibly makes the but duethe to decrease in entropy, they are endergonic, Gibbs free energies values being +11.65 to +13.08 makes endo approach comparatively less accessible. All the reactions are moderately exothermic, endo approach comparatively less accessible. All the reactions are moderately exothermic, but due to kcal/mol. but due to decrease in entropy, they are endergonic, freevalues energies values being to +11.65 to +13.08 decrease in entropy, they are endergonic, Gibbs free Gibbs energies being +11.65 +13.08 kcal/mol. In accordance with the theoretical results, DA reaction of 2-phosphaindolizines (47) could be kcal/mol. In accordancewith with the theoretical results, DA reaction of 2-phosphaindolizines could be 31P NMR, with a single(47) accomplished complete diastereoselectivity, indicated product In accordance with the theoretical results, DAasreaction of by 2-phosphaindolizines (47) could be 31 P NMR, with a single accomplished with complete diastereoselectivity, as indicated by product being being formed in each case (Scheme 13) [40]. 31 accomplished with complete diastereoselectivity, as indicated by P NMR, with a single product
formed in each case (Scheme 13) [40].13) [40]. being formed in each case (Scheme
Me N 47
HO
Me
HO
Me 45 45 47-49 R 47-49
P(48) 31R 31 31P(49) P(48) 27 31Al(49) P(49)
27Al(49)
EtAlCl2 -50 oC EtAlCl 2 CH2Cl 2 -50 oC CH2Cl2
a
b
Cl2AlO Cl2AlO c
OMe a 196.0 OMe -10.2 196.0 62.3 -10.2
OEt CMe b c 3 217.4 211.3 OEt CMe3 -9.4 217.4 -14.3 211.3 72.3 -14.3 100.9 -9.4 62.3 72.3 100.9
OMenth =
HO
Me
OMenth =
HO
Me
46 46
P Me
N ROC P Me 47 o -50 C ROC Me CH2Cl 2 -50 oC CH2Cl2
Me P Me AlCl2 OMenth N P AlCl2 ROC 48 OMenth ROC 48 N
Me N ROCN ROC
P MeAlCl2 OMenth P AlCl2 OMenth 49
49 dichloride Scheme 13. Diels-Alder reaction of 2-phosphaindolizine-n1-P-aluminium(o-menthoxy) with DMB. 1 Scheme 13. Diels-Alder reaction of 2-phosphaindolizine-n 1-P-aluminium(o-menthoxy) dichloride Scheme 13. Diels-Alder reaction of 2-phosphaindolizine-n -P-aluminium(o-menthoxy) dichloride with DMB.
with DMB. 3.4. Electrophilic Substitution
3.4. Electrophilic In contrastSubstitution to indolizines [41], 1-unsubstituted 2-phosphaindolizines failed to react with MeCOCl, PhCOCl, Me3SiCl, even on 1-unsubstituted prolonged heating in the presence of Et3N [42]. to However, like In contrast toorindolizines [41], 2-phosphaindolizines failed react with indolizines [43], 1-unsubstituted 2-phosphaindolizines reacted with PCl 3 or PhPCl2 (but not with MeCOCl, PhCOCl, or Me3SiCl, even on prolonged heating in the presence of Et3N [42]. However, like Ph 2PCl) in the presence of Et3N to afford the phosphynated products (Scheme 14) [44]. It is indolizines [43], 1-unsubstituted 2-phosphaindolizines reacted with PCl3 or PhPCl2 (but not with Ph2PCl) in the presence of Et3N to afford the phosphynated products (Scheme 14) [44]. It is
Molecules 2018, 23, 1283
10 of 13
3.4. Electrophilic Substitution In contrast to indolizines [41], 1-unsubstituted 2-phosphaindolizines failed to react with MeCOCl, PhCOCl, or Me3 SiCl, even on prolonged heating in the presence of Et3 N [42]. However, like indolizines [43], 1-unsubstituted 2-phosphaindolizines reacted with PCl3 or PhPCl2 (but not10with Molecules 2018, 23, x FOR PEER REVIEW of 13 Molecules 23,presence x FOR PEER 10 of 13 Ph in the of REVIEW Et3 N to afford the phosphynated products (Scheme 14) [44]. It is noteworthy 2 PCl) 2018, that on reacting PCl3 , the initially formed dichlorophosphino product (51,product R=Cl) undergoes noteworthy that with on reacting with PCl3, the initially formed dichlorophosphino (51, R=Cl) noteworthy that on reacting with PCl 3, the initially formed dichlorophosphino product (51, R=Cl) dismutation accompanied accompanied by the loss of PCl PClPCl bridged as undergoes dismutation by3 to theafford lossa of 3 to bis(2-phosphaindolizine) afford a PCl bridged (52) bis(2undergoes dismutation accompanied by the loss of PCl 3 to afford a PCl bridged bis(2the final product. phosphaindolizine) (52) as the final product. phosphaindolizine) (52) as the final product.
R2 R2
R11
R
R3 R3
N N P P 50 50
R2 R2 RPCl2 RPCl2 Et3N Et3N
a a R11 COPh R2 COPh R H R23 H R3 H R H
R2 R2
R3 R3
N N R1 P R1 P 51 51 b c b c COPh COPh COPh COPh Me H Me H n H n Bu H Bu
R = Cl R = Cl -PCl3 -PCl3 PRCl PRCl d d Et CO 2 CO2Et H H H H
e e Et CO 2 CO2Et Me Me H H
R1 R1
R3 R3
N N P P 52 52
PCl PCl 2 2
f f p-NO 2C6H4 p-NO2C6H4 H H H H
Scheme 14. Reaction of 1-unsubstituted 2-phosphaindolizines with chlorophosphines. Scheme 14. Reaction of of 1-unsubstituted 1-unsubstituted 2-phosphaindolizines 2-phosphaindolizines with Scheme 14. Reaction with chlorophosphines. chlorophosphines.
The concept of hard-soft acid-base (HSAB) theory was extended to explain the global and local The concept of hard-soft acid-base (HSAB) theory was extended to explain the global and local reactivity of organic molecules towards nucleophilic and electrophilic reagents [45]. The chemical molecules towards towards nucleophilic and electrophilic electrophilic reagents [45]. [45]. The chemical chemical reactivity molecules reactivity of at organic a particular molecular site nucleophilic could be rationalized using a reagents quantitative descriptor, the reactivity at ataaparticular particularmolecular molecular site could be rationalized using a quantitative descriptor, the reactivity site could be rationalized using a quantitative descriptor, the Fukui Fukui function (f(r)). We recently used Fukui function to rationalize 1,2- versus 1,4-addition of amines Fukui function (f(r)). We recently used Fukui function to rationalize 1,2versus1,4-addition 1,4-additionof of amines function We recently used Fukui function to rationalize 1,2versus to maleic(f(r)). anhydride [46]. The difference in the behavior of indolizine and 2-phosphaindolizine anhydride [46]. [46]. The difference in the behavior of indolizine indolizine and and 2-phosphaindolizine 2-phosphaindolizine to maleicelectrophilic anhydride Thecan difference towards reagents be rationalized on the basis of the Fukui functions calculated at electrophilicreagents reagentscan canbeberationalized rationalizedonon the basis of the Fukui functions calculated at towards electrophilic the basis of the Fukui functions calculated at the the B3LYP/6-31+G* level (Figure 4) [47]. the B3LYP/6-31+G* level (Figure 4) [47]. B3LYP/6-31+G* level (Figure 4) [47]. Fukui functions for electrophilic attack f - x Fukui functions for electrophilic attack f x
5 5
5 6 5 6
6 6
4 7 4 7 4 7 4N 7 N 1 1 3 N 1 0.051 3N -1.109 1 3 0.006 0.051 P -0.892 3 -1.109 0.006 2 P2 -0.892 2 2 1.110 1.110 Indolizine 2-Phosphaindolizine Indolizine 2-Phosphaindolizine +x Fukui functions for nucleophilic attack f + Fukui functions for nucleophilic attack f x O O O O H3C C Cl C6H5 C Cl Me3-Si-Cl PCl3 H3C C Cl C6H5 C Cl Me3-Si-Cl PCl3 -0.083 -0.083
-1.261 -1.261
-0.233 -0.233
0.076 0.076
Figure 4. Fukui functions calculated at the B3LYP/6-31+G* level. Figure level. Figure 4. 4. Fukui Fukui functions functions calculated calculated at at the the B3LYP/6-31+G* B3LYP/6-31+G* level.
Fukui functions reveal that C1 and C3 in indolizine are much harder than C1 and C3 in 2Fukui functions reveal that C1 and C3 in indolizine are much harder than C1 and C3 in 2phosphaindolizine. Thus, hard electrophilic reagents such as acyl chlorides and chlorotrimethylsilane phosphaindolizine. Thus, hard electrophilic reagents such as acyl chlorides and chlorotrimethylsilane attack at C1 and C3 positions of indolizine, but they fail to react with 2-phosphaindolizine. On the attack at C1 and C3 positions of indolizine, but they fail to react with 2-phosphaindolizine. On the other hand, phosphorus trichloride is comparatively softer and hence reacts with 2other hand, phosphorus trichloride is comparatively softer and hence reacts with 2phosphaindolizine at C1 (C3 being substituted in our compounds). Being at the borderline, PCl 3 phosphaindolizine at C1 (C3 being substituted in our compounds). Being at the borderline, PCl 3 reacts with indolizine also. reacts with indolizine also.
Molecules 2018, 23, 1283
11 of 13
Fukui functions reveal that C1 and C3 in indolizine are much harder than C1 and C3 in 2-phosphaindolizine. Thus, hard electrophilic reagents such as acyl chlorides and chlorotrimethylsilane attack at C1 and C3 positions of indolizine, but they fail to react with 2-phosphaindolizine. On the other hand, phosphorus trichloride is comparatively softer and hence reacts with 2-phosphaindolizine at C1 (C3 being substituted in our compounds). Being at the borderline, PCl3 reacts with indolizine also. 4. Conclusions The computational calculations could be used as a useful tool to understand and solve problems encountered during reactions of annelated 1,3-azaphospholers. Furthermore, this strategy worked well while planning the reactions such as Lewis acid catalyzed and asymmetric DA reactions across the >C=P- functionality of 2-phosphaindolizines. Author Contributions: All authors contributed equally in the compilation of this perspective. Funding: This research received no external funding. APC was sponsored by MDPI. Acknowledgments: We are thankful to the authorities of The IIS University, Jaipur (India) for the facilities. Conflicts of Interest: The authors declare no conflict of interest.
References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
13. 14. 15. 16. 17.
Dirac, P.A.M. Quantum mechanics of many-electrons systems. Proc. R. Soc. Lond. Ser. A. 1929, 123, 714–733. [CrossRef] Hohenberg, P.; Kohn, W. Inhomogenous electron gas. Phys. Rev. 1964, 136, B867–B871. [CrossRef] Kohn, W.; Sham, L.J. Self-consistent equations including exchange and correlation effects. Phys. Rev. 1965, 140, A1133–A1138. [CrossRef] Zangwill, A. A half century of density functional theory. Phys. Today 2015, 68, 34–39. [CrossRef] Jain, A.; Shin, Y.; Persson, K.A. Computational predictions of energy materials using density functional theory. Nat. Rev. Mater. 2016, 1, 1–13. [CrossRef] Cohen, A.J.; Mori-Sánchez, P.; Yang, W. Challenges for density functional theory. Chem. Rev. 2012, 112, 289–320. [CrossRef] [PubMed] Jones, R.O. Density functional theory: Its origins, rise to prominence, and future. Rev. Mod. Phys. 2015, 87, 897–923. [CrossRef] Tsipis, A.C. DFT flavor of coordination chemistry. Coord. Chem. Rev. 2014, 272, 1–29. [CrossRef] Cole, D.J.; Hine, N.D.M. Applications of large-scale density functional theory in biology. J. Phys. Condens. Matter 2016, 28, 393001. [CrossRef] [PubMed] Lupp, D.; Christensen, N.J.; Fristrup, P. Synergy between experimental and theoretical methods in the exploration of homogeneous transition metal catalysis. Dalton Trans. 2014, 43, 11093–11105. [CrossRef] [PubMed] Nørskov, J.K.; Bligaard, T.; Rossmeisl, J.; Christensen, C.H. Towards the computational design of solid catalysts. Nat. Chem. 2009, 1, 37–46. [CrossRef] [PubMed] Lineberger, W.C.; Borden, W.T. The synergy between qualitative theory, quantitative calculations, and direct experiments in understanding, calculating, and measuring the energy differences between the lowest singlet and triplet states of organic diradicals. Phys. Chem. Chem. Phys. 2011, 13, 11792–11813. [CrossRef] [PubMed] Sierka, M. Synergy between theory and experiment in structure resolution of low-dimensional oxides. Prog. Surf. Sci. 2010, 85, 398–434. [CrossRef] Mcgrady, J. Guest Editor The themed collection, “The synergy between theory and experiment. Dalton Trans. 2009, 5805–6064. [CrossRef] Bansal, R.K.; Gupta, N.; Karaghiosoff, K.; Schmidpeter, A.; Spindler, C. 2-Phosphaindolizines. Chem. Ber. 1991, 124, 475–480. [CrossRef] Bansal, R.K.; Kabra, V.; Gupta, N.; Karaghiosoff, K. Synthesis of some new 2-phosphaindolizines. Indian J. Chem. 1992, 31, 254–256. Gupta, N.; Jain, C.B.; Heinicke, J.; Bharatiya, N.; Bansal, R.K.; Jones, P.G. 2-Phosphaindolizines. Heteroat. Chem. 1998, 9, 333–339. [CrossRef]
Molecules 2018, 23, 1283
18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.
30. 31. 32.
33. 34.
35.
36.
37.
38.
39.
12 of 13
Karaghioshoff, K.; Bansal, R.K.; Gupta, N. 1,4,2-Diazaphospholo[4,5-a]pyridines. Z. Naturforsch. B 1992, 47b, 373–378. Bansal, R.K.; Pandey, G.; Karaghiosoff, K.; Schmidpeter, A. 1,2,3-Diazaphospholo[1,5-a]pyridines. Synthesis 1995, 173–175. [CrossRef] Bansal, R.K.; Gandhi, N.; Karaghiosoff, K.; Schmidpeter, A. Synthesis of [1,2,4,3]triazaphospholo[1,5-a]pyridines. Z. Naturforsch. 1995, 50b, 558–562. [CrossRef] Bansal, R.K.; Mahnot, R.; Sharma, D.C.; Karaghiosoff, K.; Schmidpeter, A. 1,3-Azaphospholo[5,1-b]thiazolines and benzothiazoles. Heteroat. Chem. 1992, 3, 351–357. [CrossRef] Bansal, R.K.; Jain, C.B.; Gupta, N.; Kabra, V.; Karaghiosoff, K.; Schmidpeter, A. Synthesis and properties of a 5,6-dihydro-1,3-azaphospholo[5,1-b]oxazole. Phosphorus Sulfur Silicon 1994, 86, 139–143. [CrossRef] Bansal, R.K.; Karaghiosoff, K.; Gandhi, N.; Schmidpeter, A. 2-Substituted cycloiminium salts in azaphosphole synthesis. Synthesis 1995, 361–369. [CrossRef] Bansal, R.K.; Surana, A.; Gupta, N. 2-Phosphaindolizines via 1,5-electrocyclization. Tetrahedron Lett. 1999, 40, 1565–1568. [CrossRef] Bansal, R.K.; Gupta, N.; Baweja, M.; Hemrajani, L.; Jain, V.K. Pyridinium dichlorophosphinomethylides. Heteroat. Chem. 2001, 12, 602–609. [CrossRef] Bansal, R.K.; Hemrajani, L.; Gupta, N. 1,3-Azaphospholo[5,1-a]isoquinolines. Heteroat. Chem. 1999, 10, 598–604. [CrossRef] Singh, D.; Sinha, P.; Gupta, N.; Bansal, R.K. Synthesis of 1,3-azaphospholo[1,5-f ]phenanthridines through 1,5-electrocyclization. Phosphorus Sulfur Silicon Relat. Elem. 2016, 191, 488–492. [CrossRef] Appel, R.; Knoll, F.; Ruppert, I. Phospha-alkenes and phospha-alkynes, genesis and properties of the (p-p)π-multiple bond. Angew. Chem. Int. Ed. 1981, 20, 731–744. [CrossRef] Mathey, F.; Mercier, F.; Charrier, C.; Fischer, J.; Mitschler, A. Dicoordinated 2H-phospholes as transient intermediates in the reactions of tervalent phospholes at high temperature. One-step synthesis of 1-phosphabornadienes and phosphorins from phospholes. J. Am. Chem. Soc. 1981, 103, 4595–4597. [CrossRef] Wannere, C.S.; Bansal, R.K.; Schleyer, P.v.R. Diels-Alder reaction of phosphaethene with 1,3-dienes: An ab-initio study. J. Org. Chem. 2002, 67, 9162–9174. [CrossRef] [PubMed] Bansal, R.K.; Gupta, N.; Kumawat, S.K.; Dixit, G. Diastereo- and regioselective Diels-Alder reactions of 2-phosphaindolizines. Tetrahedron 2008, 64, 6395–6401. [CrossRef] Bansal, R.K.; Jain, V.K.; Gupta, N.; Gupta, N.; Hemrajani, L.; Baweja, M.; Jones, P.G. Stereo- and regioselectivity in Diels-Alder reactions of 1,3-azaphospholo[5,1-a]isoquinoline and –[5,1-a]pyridine. Tetrahedron 2002, 58, 1573–1579. [CrossRef] Bansal, R.K.; Karaghiosoff, K.; Gupta, N.; Gandhi, N.; Kumawat, S.K. Diastereo- and regioselectivity in Diels-Alder reaction of [1,4,2]diazaphospholo[4,5-a]pyridines. Tetrahedron 2005, 61, 10521–10528. [CrossRef] Karaghiosoff, K.; Cleve, C.; Schmidpeter, A. Chloromethyl-dichlorophosphane: A useful reagent for the synthesis of new heterocycles with dicoordinated phosphorus. Phosphorus Sulfur Silicon Relat. Elem. 1986, 28, 289–296. [CrossRef] Pellegriner, S.C.; Silva, M.A.; Goodman, J.M. Theoretical evaluation of the origin of the regio- and stereoselectivity in the Diels-Alder reactions of dialkylvinylboranes: Studies on the reactions of vinylborane, dimethylvinylborane, and vinyl-9-BBN with trans-piperylene and isoprene. J. Am. Chem. Soc. 2001, 123, 8832–8837. [CrossRef] Bansal, R.K.; Gupta, N.; Dixit, G.; Kumawat, S.K. Theoretical investigation of an unusual substituent effect on the dienophilicity of >C=P– Functionality in 2-phosphaindolizines. J. Phys. Org. Chem. 2009, 22, 125–129. [CrossRef] Gupta, N.; Jangid, R.K.; Bansal, R.K.; Hopffgarten, M.V. Catalytic effect of organoaluminium chloride reagents on the dienophilic reactivity of indolizine and 2-phosphaindolizine towards [2+4] cycloaddition: A DFT investigation. Curr. Cat. 2012, 1, 93–99. [CrossRef] Jangid, R.K.; Gupta, N.; Bansal, R.K.; Hopffgarten, M.V.; Frenking, G. Diels-Alder reaction of 2-phosphaindolizines catalyzed by organoaluminium reagent: Theoretical and experimental results. Tetrahedron Lett. 2011, 52, 1721–1724. [CrossRef] Alcaraz, J.P.; Mathey, F. Accroissement de la reactivite des phosphorines en tant que dienes et philodienes par complexation du phosphore. Tetrahedron Lett. 1984, 25, 207–210. [CrossRef]
Molecules 2018, 23, 1283
40.
41.
42. 43. 44. 45. 46. 47.
13 of 13
Jangid, R.K.; Sogani, N.; Gupta, N.; Bansal, R.K.; Hopffgarten, M.V.; Frenking, G. Asymmetric Diels-Alder reaction with >C=P– functionality of the 2-phosphaindolizine-η1 -P-aluminium(o-menthoxy) dichloride complex: Experimental and theoretical results. Beilstein J. Org. Chem. 2013, 9, 392–400. [CrossRef] [PubMed] Swinbourne, F.J.; Hunt, J.H.; Klinkert, G. Advances in Indolizine Chemistry in the Series: Advances in Heterocyclic Chemistry; Katritzky, A.R., Boulton, A.J., Eds.; Academic Press: New York, NY, USA, 1979; Volume 23, pp. 103–170. Bansal, R.K.; Heinicke, J. Anellated heterophospholes and phospholides and analogies with related non-phosphorus systems. Chem. Rev. 2001, 101, 3549–3578. [CrossRef] [PubMed] Tolmachev, A.A.; Yurchenko, A.A.; Kozlov, E.S.; Shulezhko, V.A.; Pinchuk, A.M. Phosphorylated indolizines. Heteroat. Chem. 1993, 4, 343–360. [CrossRef] Bansal, R.K.; Gupta, N.; Kabra, V.; Spindler, C.; Karaghiosoff, K.; Schmidpeter, A. Substitution of 2-phosphaindolizines by bromine and by chlorophosphines. Heteroat. Chem. 1992, 3, 359–366. [CrossRef] Mendez, F.; Gázquez, J.L. Chemical reactivity of enolate ions: The local hard and soft-acids and bases principle viewpoint. J. Am. Chem. Soc. 1994, 116, 9298–9301. [CrossRef] Kour, M.; Gupta, R.; Bansal, R.K. Experimental and theoretical investigation of the reaction of secondary amines to maleic anhydride. Aust. J. Chem. 2017, 70, 1247–1253. [CrossRef] Gupta, R.; Paul, B.; Bansal, R.K. Application of Fukui functions for comparing reactivities of indolizine and 2-phosphaindolizine towards electrophilic substitution. 2018, unpublished work. © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).