| dc.description.abstract | The assignment of structures to two forms of
?,? di tert butylthioacetic esters, detected by NMR, is the
subject matter of this thesis entitled “Factors Influencing
the Chemical Shift - A Case Study with ?,? Di tert butylthioacetic
Esters.”
Rotation about formal single bonds can never be
considered as absolutely “free” since the presence of any
barrier, however low, has the effect of perturbing the
populations of all of the rotational states?¹?. The value of
a barrier separating two rotamers is the resultant of the
action of a number of factors: electron delocalisations,
classical steric effects, dipolar electric influences,
internal hydrogen bonding, solvent effects, etc., and as a
result, the barriers may vary over a wide range. Yet,
stabilisation of particular forms (rotational isomers), to a
degree that enables their detection at ordinary temperature
by a suitable technique, is encountered comparatively rarely.
An instance of this rare type is found with ?,? di tert
butylthioacetic esters (E). Esters E exhibit
doubling into components of unequal intensity of salient
resonances, notably the resonances of ? protons, in their
NMR spectra. The work of Elam² can be taken as
having established conclusively that this doubling is not
due to factors (presence of enolic form, thio ester form,
dimer, etc.) other than the stabilisation of two rotational
isomers about the C–?–C (ester) bond. Elam et al. did
not specifically assign conformations to the detected
rotational isomers.
The present thesis is concerned with an assessment
of influences that may determine the intensity ratio and the
relative chemical shift of the H ? lines in the methyl
ester E.
After describing in detail the researches of
Elam et al. in Chapter I, the possibility of one or both of
the detected forms being a chiral entity is considered. An
experiment, the results of which could be construed as
establishing that neither of the detected stabilisations
was chiral in the case of the ethyl ester E, carried out
earlier in these laboratories, is described.
This is followed by an account of a preliminary, but
inconclusive, attempt to assign the doubled lines to
specific rotational isomers, by relating the intensity of
the components with the enthalpies of particular forms,
calculated by employing a semi empirical approach. The main
reason for the inconclusive nature of the results lay in
not allowing for the effect of the rotation of the tert butyl
groups of E in the calculations.
In the next Chapter (Chapter III) is described an
attempt that takes into account the effect of the rotation
of tert butyl groups. After a brief discussion of the basic
principles of quantum chemical approaches and a short
discussion of the reliability of various such approaches in
the calculation of enthalpies as reported by others with
different systems, the reasons for the choice of the CNDO/2
method are given. The results of the application of this
method to the case of methyl ester E, assuming various
? = 5 dihedrals and combinations of conformations
(internal and external) of the tert butyl groups, are detailed.
A salient item immediately emerging from data thus gathered
has been that over widely different tert butyl conformations, the
forms in which H ? is syn and anti periplanar to C=S are
stabilised over other forms. The barriers separating these
two forms, while high, are highly sensitive to the tert butyl
conformations.
Three possible procedures for weighted averaging
were examined in order to take account of the rotation of
tert butyl groups. Following the presentation of the results
of applying these procedures, their relative merits are
discussed. Overall, it has emerged that the difference in
enthalpies of the syn and anti forms (shown as stabilised
in the CNDO/2 approach after applying any of the averaging
procedures) is too small for reliable assignment, since
approximations implicit in the methods could have had a
profound effect on the sign of the difference. It has,
nevertheless, appeared from these calculations that a
higher degree of librational freedom would be associated
with the syn form than with the anti form. The entropy
factor could, therefore, exert an effect on the outcome in
that the syn form is rendered stabilised despite the fact
that the anti form emerges as the somewhat more stabilised
when the entropy factor is not considered.
A second attempt at assignment was based on the
possibility of ?s density at H ? changing in some specific
manner on going over to the anti form from the syn form.
Calculations, using again the CNDO/2 approach, showed that
?s density at H ? would be lower in the syn form than in the
anti form. A parallel difference was found for analogously
structured systems (e.g., ?,? disubstituted acetaldehydes,
etc.), both by the INDO method (for amenable systems)
and by the CNDO/2 method, thus increasing the confidence in
the results with the esters E.
Insofar as the assignment of the H ? line that
occurs at lower field (and has the higher intensity) to the
syn form (on the basis that ?s density at H ? in this form
is lower) accorded with that form being the more stabilised
if the entropy factor becomes decisive, the CNDO formalism
could be taken as having output internally consistent
results. This is the subject of Chapter IV wherein also are
discussed the factors affecting the chemical shift and
the scaling of electron density changes into chemical shift
differences before presenting the results concerned with
electron density calculations. The Chapter ends with a
cautionary note on relying solely on electron density
changes for assignments of the type attempted.
The anisotropic magnetic properties of the thioester
function could be an important factor in the change of
chemical shift of H ? on proceeding from the syn to the
anti form. Using the working hypothesis that the anisotropy
of the thione function may resemble that of the
carbonyl function, the possible ways in which the properties
of the former are altered when it forms part of the thioester
function are examined. This has required a fairly
extensive review of what is known about the anisotropic
magnetic properties of the carbonyl. The conclusions have
been reached that the anisotropic effect of the thiocarbonyl,
as forming part of a thioester function, is not likely to
be large and that the assignment of the H ? lines on the
basis of electron density change alone, qualitatively
considered, can be relied upon.
The work on the anisotropic magnetic properties of
the carbonyl, reviewed in a section of the Chapter
wherein the matters dealt with in the previous paragraph are
discussed (Chapter V), has shown that its anisotropic
properties arrived at on a semi empirical (statistical)
basis? differ from those arrived at starting from quantum
chemical principles?. A number of instances (example -
ref. 5) were found in the literature where, in extending the
anisotropic properties of the carbonyl to the thiocarbonyl,
the latterly accepted anisotropic properties? have not been
taken into proper account. These instances are re examined
and a new mnemonic has been suggested whereby the effect of
the thiocarbonyl can be explained even while considering
that the statistically arrived at anisotropic properties of
the carbonyl are validly extendible to the thione.
In the final section, note is taken of the fact that
rather poor agreement is found between values of chemical
shift change calculated on the basis of the new geometry of
the carbonyl anisotropy and observed changes in the chemical
shifts of H ? in certain substituted aldehydes. An
explanation based on the dependence of H ? shifts on the
degree of branching of the ? substituents as well as on the
dependence of ?s density at H ? on the H–C–C=O
dihedral is suggested.
Three appendices are included.
The first deals with a study of the potential
surface generated when simultaneous rotation about the
C–?–C (ester) and C–?–C–t Bu bonds occurs in
methyl tert butylthioacetate. The nature of this surface has
been found to be indicative of an interesting phenomenon
whereby these types of rotation become coordinated (gear
effect).
The results of a truncated Fourier analysis? of
various ?,? disubstituted carbonyl systems and their thio
analogues are presented in Appendix II. The analysis was
undertaken in an attempt at an examination of the relative
importances of electronic, steric, etc. factors in
determining the ?,? enthalpies in the systems.
In Appendix III possible methods of experimental
verification of the assignment of structures to the
rotational isomers of the esters E, suggested in the main
body of the thesis, are set forth. | |
