The Anomeric Effect and Associated Stereoelectronic Effects

Chapter 13

Do Stereoelectronic Effects Control the Structure and Reactivity of TrigonalBipyramidal Phosphoesters?

Downloaded by STANFORD UNIV on June 5, 2013 | http://pubs.acs.org Publication Date: November 23, 1993 | doi: 10.1021/bk-1993-0539.ch013

Philip Tole and Carmay Lim Departments of Molecular and Medical Genetics, Chemistry, and Biochemistry, University of Toronto, 1 King's College Circle, Toronto, Ontario M5S 1A8, Canada

The stationary points for trigonal-bipyramidal transition states and intermediates of phosphate and phosphoamidate esters have been calculated using ab initio methods. The fully optimized ge ometries have been examined to establish whether stereoelectronic control through antiperiplanar lone pair effects can influence the structure and reactivity of these phospho-esters. The results show that although stereoelectronic effects cannot completely be ruled out, a more appropriate concept that is consistent with the results is conformational rearrangement to optimize electrostatic interac tions. This is especially important in the long-range transition states corresponding to the rate-limiting hydroxyl ion attack of methyl ethylene phosphate and its acyclic analog, trimethyl

phosphate. The results also show that stereoelectronic effects do not appear to play a significant role in enhancing the reactivity of five-membered cyclic esters relative to their six-membered or acyclic analogs.

It is very useful to identify general principles that govern chemical structure and reactivity which would allow us to construct predictive models. One such principle is stereoelectronic control of X—C—Y systems, where overlap between the antiperiplanar (app) lone pair on Y (ny) and the antibonding orbital of the C-X bond (&c-x) thought to strengthen the C-Y bond and weaken the C-X bond; for reviews see (1-4)- This concept has been extended to X-P-Y systems, originally by Lehn and Wipff (5) to account for the preferred conformations of phosphoric acid, and subsequently by Gorenstein and co-workers (1,6) to explain part of the enhanced reactivity of five-membered cyclic phospho-esters relative to their acyclic counterparts. 1S

0097-6156/93/0539-0240$06.00/0 © 1993 American Chemical Society

In The Anomeric Effect and Associated Stereoelectronic Effects; Thatcher, G.; ACS Symposium Series; American Chemical Society: Washington, DC, 1993.


13. TOLE AND LIM

Trigonal-Bipyramidal Phosphoesters

241

Based on semi-empirical CNDO/2 and Hartree-Fock/STO-3G calculations on acyclic trigonal-bipyramidal (TBP) neutral and dianionic dimethoxy phosphoranes, Gorenstein and co-workers (6) proposed that the lowest energy reaction path for acyclic and cyclic phospho-ester hydrolyses should have a TBP tran sition state with the equatorial oxygen lone pair(s) app (or anticlinal) to the apical leaving group. This η—σ* interaction would then cause the axial bond to weaken and facilitate its cleavage. It was stated that the activation entropy for acyclic phospho-ester hydrolysis, which required two rotational degrees of freedom to befixedabout the acyclic P-O(Me) bonds, should be more negative than that for hydrolysis of its cyclic counterpart, where the equatorial oxygen lone pairs were already partially app to the axial P-0 ring bond due to the constraint of the ring. The authors (6) suggested that such stereoelectronic effects in the TBP transition states were consistent with the large entropie con tribution (18 cal/mol/K) to the rate acceleration in ethyl propyl phosphonate observed by Aksnes and Bergesen (7). However, Kluger and Taylor (8) measured an activation entropy for ethyl propyl phosphonate of -27 cal/mol/K. They found that the activation entropy differ ences between methyl esters of ethylene phosphate or propyl phosphonate and their acyclic analogs was less than 1 cal/mol/K. On the other hand the differ ences for the ethyl esters were about 8 cal/mol/K. The conclusion was that ac tivation entropy differences stemming from stereoelectronic effects were not the source of the rate enhancement. Furthermore, the rate constant ratio of methyl ethylene phosphate (MEP)/phosphonate compared with the acyclic analogs is similar to the rate constant ratio of methyl propyl phosphonate/phosphinate to their acyclic counterparts (8). This is inconsistent with stereoelectronic theory which would predict that methyl ethylene phosphonate should have a larger cyclic/acyclic rate constant ratio than methyl propyl phosphinate whose TBP transition state has no equatorial app lone pairs to stereoelectronically aid in ring opening. Doubts about the stereoelectronic control of TBP phosphorane transition states have also been raised due to the fact that CNDO/2 calculations do not repro duce the anomeric effect at the carbon centre (9) and the ST0-3G basis does not predict the relative stability of TBP phosphorane structures (10,11). When Gorenstein and co-workers (6) suggested that stereoelectronic effects were re sponsible for the extra stabilization of five-membered cyclic versus acyclic phos phate diester transition states,ringopening was implicitly assumed to be the rate-determining step. However, calculations on the alkaline hydrolysis of ethy lene phosphate (10), MEP (12, 13), dimethyl phosphate (11) and trimethyl phosphate (TMP) (Tole, P.; Lim, C. manuscript in prep.) established that hydroxyl ion attack is rate-limiting. Thus, stereoelectronic effects originating from one app lone pair or two anticlinal lone pairs on the equatorial ring atom in the TBP transition state cannot control the observed rate acceleration. In previous work we have employed quantum mechanical and continuum dielec tric calculations to map out the detailed gas-phase and solution reaction profiles


242

THE ANOMERIC EFFECT AND ASSOCIATED STEREOELECTRONIC EFFECTS

for the alkaline hydrolysis of MEP (12, 18) and its acyclic analog, TMP (Tole, P.; Lim, C. manuscript in prep.). The geometries of TBP transition states and/or intermediates formed from hydrolysis offive-memberedcyclic MEP, methyl aminoethylene phosphonate MNP (MEP with one of the ring oxygens replaced by NH) and acyclic TMP have been fully optimized using ab initio methods. The next section briefly outlines these methods. In the results and discussion section the TBP structures are examined to determine the factors, in particular stereoelectronic effects arising from η—σ* interactions, that affect their conformation and stability.


Method

Details of the method have been described in (14) and only a brief summary is given here. The ab initio calculations were performed using the program Gaussian 90 (15). All geometries were fully optimized at the Hartree-Fock (HF) level using the 3-21+G* basis set, which has Is orbitals on hydrogens, Is, 2s, 2p , 2p , 2p orbitals on carbon and oxygen atoms, Is, 2s, 2ρ , 2p , 2p , 3s, 3p , 3p , 3p , 3d , 3d , 3d , 3d , 3d , 3d orbitals on phosphorus as well as diffuse s, ρ , p , p functions on all atoms except hydrogens. The geometries of cyclic and acyclic tetrahedral phospho-esters have been found to be in good agreement with experiment (13). The electronic correlation energy was estimated by second-order M0ller-Plesset perturbation theory (MP2) with the 6-31+G * basis for single points corresponding to the 3-21+G* optimized geometries. The electrostatic potential was calculated from the HF/3-21-I-G* wavefunctions and fitted to an atomic point charge model using the CHELP program (16). The stereoelectronic effect was analyzed in terms of η—σ* inter actions where the lone pairs are either sp or sp hybrids, as indicated by the X-O-P (X= C, H) angle. z

y

x

y

z

z

ζ

xx χ

yy

y

zz

xy

yz

y

z

xz

z

3

2

Results and Discussion

The following nomenclature is adopted in this paper. The first letter of each structure defines the starting material with M, A and Τ representing MEP, methyl aminoethylene phosphonate and acyclic TMP respectively. The second letter denotes either a TBP intermediate (I) or transition state (TS) with a su perscript a/e indicating an apical/equatorial hydroxyl group. The third letter describes the nature of the complex: L denoting a long-range ion-dipole inter action where the P-O(H) bond is not fully formed (a superscript Ν denotes an apical ring nitrogen), Ρ for pseudorotation, H and Me for hydroxyl and methoxy group rotation about P-0(H/Me), Ν and X for endo- and exo-cyclic cleavage. Table I contains the HF and MP2 energies and the thermodynamic parameters. The structural information is listed in Tables II-IV where the 2' and 3' denote an oxygen or nitrogen in the axial 2' or equatorial 3' position. The Mulliken and CHELP atomic charges are given in Tables V and VI, respectively. The results will be discussed in the following order: (i) the long-range tran sition state complexes between the reactants that determine the rate-limiting


TOLE AND LIM

243

Trigonal-Bipyramidal Phosphoesters


Table I. Energies and Thermodynamic Parameters E

ΕΛ/Ρ2

A H

F

6

ETRV°

STRV*

atomic units (a.u.) kcal/mol cal/mol/K 92.391 MEP -754.13769 -759.15445 82.196 90.957 MNP -734.40294 -739.30189 90.499 101.807 96.903 TMP -755.31038 -760.32676 MTS L -829.15439 -834.76724 89.774 97.420 95.836 ATS L -809.44313 -814.94414 98.392 98.653 ATS L -809.40901 -814.90689 97.935 108.273 TTS L -830.33481 -835.94541 104.632 91.328 MTS P -829.20178 -834.81786 90.763 92.857 MPN -829.21132 -834.82843 91.835 93.157 MTS H -829.19343 -834.81163 90.389 92.429 MPX -829.21107 -834.82866 91.846 AP -809.46421 -814.96539 99.882 93.687 92.900 -809.47454 -814.97331 100.107 APN 93.060 ATS H -809.45551 -814.95770 98.751 93.922 APX -809.46705 -814.97009 99.927 a) HF energies for fully optimized 3-21+G* geometries. b) Single point MP2/6-31+G* energy calculation at 3-21+G* geometry. c) E RV = Etrans + E + E in kcal/mole. d) STRV = Strans + S + S& in cal/mol/K. a

a

a

N

a

a

e

e

T

rot

roi

vib

m



244


MEP MNP TMP MTS L ATS L ATS L TTS L MTS P MPN MTS H MPX AI APN ATS H APX a

a

a

a

a

e

a

e

N

Table Ρ 01 1.46 1.47 1.47 1.46 1.48 1.47 1.48 1.51 1.50 1.51 1.50 1.52 1.51 1.51 1.51

I L H F / 3 - 2 1 + G * B o n d lengths (À) Ρ Ρ Ρ Ρ C2' C3' C(Me) 02 0 2' 2' 3' 3' 05' 1.57 1.60 1.60 1.47 1.47 1.47 1.47 1.58 1.64 1.60 1.48 1.47 1.47 1.57 1.59 1.57 1.46 1.47 1.48 1.56 1.65 1.61 2.92 1.45 1.45 1.46 1.59 1.65 1.65 2.53 1.45 1.46 1.48 1.57 1.69 1.61 2.90 1.47 1.46 1.46 1.59 1.58 1.62 2.59 1.44 1.46 1.63 1.79 1.66 1.69 1.42 1.44 1.44 1.44 1.63 1.82 1.67 1.67 1.42 1.44 1.44 1.65 1.76 1.68 1.69 1.43 1.44 1.64 1.73 1.68 1.73 1.43 1.44 1.43 1.63 1.74 1.69 1.78 1.43 1.46 1.44 1.43 1.65 1.81 1.69 1.70 1.42 1.46 1.43 1.65 1.75 1.68 1.76 1.43 1.46 1.43 1.63 1.73 1.70 1.80 1.43 1.46


C2' C3' 1.55 1.55 1.55 1.54 1.54 1.54 1.54 1.54 1.54 1.54 1.54 1.54 1.54


e

a

a

e

a

e

a

a

N

02 Ρ 01 01 2' 02 01 02 3' 3' A 01 Ρ 0 Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Ρ Β C(Me) 02 3' 3' 05' 05' 2' 05' 05' 2' 2' C 114.2 118.1 104.6 129.0 MEP 116.9 105.5 95.14 111.9 117.9 104.6 126.7 118.8 107.4 94.29 MNP 116.6 113.1 102.5 126.5 114.8 102.5 105.8 TMP MTS L 111.6 97.87 91.98 89.70 73.07 158.4 73.84 120.2 118.5 110.8 130.9 109.7 97.03 91.58 89.96 82.37 158.3 69.16 114.5 123.8 113.4 126.7 ATS L ATS L 111.5 100.6 91.22 87.84 74.88 159.6 72.49 116.0 120.8 110.7 129.9 TTS L 106.3 95.72 95.54 82.32 79.09 171.4 79.56 120.8 121.8 109.3 134.1 MTS P 96.20 85.98 84.32 98.38 88.59 165.3 86.68 120.0 119.3 120.6 128.4 96.82 82.26 83.83 102.2 87.15 160.9 88.77 122.5 113.9 123.0 121.5 MI N MTS H 90.90 84.82 83.70 102.5 86.78 161.6 85.34 117.3 114.1 128.5 122.0 99.41 85.22 86.10 99.17 84.31 161.4 86.65 123.2 114.3 122.5 122.1 MPX AP 98.73 86.66 85.59 93.82 89.75 167.2 85.10 118.1 124.9 117.0 125.3 98.41 86.99 83.95 97.76 89.66 163.3 83.44 115.3 125.7 119.0 121.5 APN 98.27 87.42 85.00 96.66 90.37 164.5 82.69 113.4 127.6 119.0 122.5 ATS H APX 99.58 87.29 86.18 96.24 84.14 164.2 86.18 120.8 120.7 118.4 122.0 The rows Α, Β, C give the atom connectivities for the respective bond angles in degrees.

Table I I I . H F / 3 - 2 1 + G * B o n d Angles (Degrees)

Ρ 2' C2' 113.6 114.1 128.0 115.1 114.9 113.5 124.3 111.4 111.0 111.8 111.3 114.1 112.6 114.0 112.6


Ρ 3' C3' 114.7 114.9 125.7 112.2 115.9 116.4 135.5 119.9 119.8 119.0 117.9 120.3 121.0 121.1 119.3

-

104.4 105.5 101.2 66.66 101.9 102.6 102.6 102.6 104.2 103.7 104.0 104.2

110.0 115.2 105.1 101.4 110.6 110.8 119.1 110.4 109.5 117.6 114.6 110.2

103.5 102.4 105.9 15.31 104.5 105.2 104.5 104.4 101.6 102.0 102.0 101.8

-

2' 3' C2' C3' C3' C2' 103.6 104.0 102.7 105.3 -

Ρ 0 Η

246


Q

CU

Q0 10 os os es M H r-l IO · oô ^ H H OO H 00 H g

O W

°î 00

The Anomeric Effect and Associated Stereoelectronic Effects

Recommend Documents