Studies in sulfur-nitrogen nucleophilicity

Article abstract of DOI:10.1002/poc.3841

As part of a study of nitrogen vs sulfur nucleophiles, the behavior of methylation products from dimethyl-(2-methylthioethyl)amine CH₃SCH₂CH₂N(CH₃)₂ 1 is described. Of the 2 potential products (a sulf

Full text of DOI:10.1002/poc.3841

Received: 9 January 2018
DOI: 10.1002/poc.3841
Revised: 6 February 2018
Accepted: 9 February 2018
S P E C I A L I S S U E A R T I C L E
Studies in sulfur‐nitrogen nucleophilicity
Marjorie C. Caserio
| Jhong K. Kim
Department of Chemistry, University of
California Irvine, Irvine, CA, USA
Abstract
As part of a study of nitrogen vs sulfur nucleophiles, the behavior of methylation
products from dimethyl‐(2‐methylthioethyl)amine CH3SCH2CH2N(CH3)2 1 is
described. Of the 2 potential products (a sulfonium salt or an ammonium salt),
the ammonium salt from N‐methylation 2 dominated. The isomeric sulfonium
salt 3 prepared by an independent route was found to be unstable and rearranged
to the isomeric ammonium salt. The rearrangement pathway was investigated
using deuterium‐labeled reactants. The sulfonium salt 3 also produced a
piperazinium double salt 9 on heating. The reaction pathway was also followed
by deuterium labeling. The results support the conclusion that production of the
double salt 9 involves intermediate formation of N,N‐dimethylaziridium ion 4.
Correspondence
Marjorie C. Caserio, Department of
Chemistry, University of California Irvine,
Irvine, CA, USA.
Email: mcaserio@ucsd.edu
KEYWORDS
Nucleophilicity, rearrangement, sulfonium salt
1 | INTRODUCTION
Our initial interest was in structural effects of bifunc-
tional molecules such as when molecules with both
sulfide and amine functions compete for a common elec-
trophile. The molecular framework chosen for study was
a RS(CH2)nNR2 system with n = 0, 1, or 2, the
objective being to determine how the proximity of sulfide
and amine functions affect their nucleophilic behavior to
different electrophiles.
This paper is a modest contribution to the field of physical
organic chemistry, but it comes with deep appreciation
and recognition of the profoundly important contribu-
tions made by John D. Roberts to the field throughout
his long career. It is a privilege to offer this paper for inclu-
sion in the special issue of the journal in recognition of
Robert's lifetime scientific achievements and the impact
he has had on generations of students and colleagues.
The work described here is a continuation of a study
reported some years ago,[1] but the science remains relevant
and, we hope, of continuing interest. It concerns the compar-
ative strengths of sulfur and nitrogen nucleophiles. Multiple
factors contribute to nucleophilicity arising from the dona-
tion of a lone electron pair on the donor atom to an acceptor
electrophile. In qualitative terms, they include size and
structure of the nucleophile, its electronegativity, polariz-
ability, oxidation potential, basicity, and external influences
such as solvent, molecular environment, and, importantly,
sensitivity to the approaching electrophile. These factors
are embodied in the prevailing theory of hard vs soft acids
and bases, nucleophiles, and electrophiles,[2] and the related
theory of control by interacting frontier molecular orbitals.[3]
Theresultsofourearlierstudy[1] confirmedthat,regard-
less of the S/N proximity, nitrogen is the preferred site of
protonation and methylation. This is consistent with much
related evidence that tricoordinate nitrogen is more basic
than dicoordinate sulfur and more nucleophilic than
sulfur to hard electophiles (eg, CH3+). However, sulfur
rather than nitrogen is the preferred site of attack by soft
electrophiles (eg, CH3S+)—consistent with the general
observation that soft nucleophiles prefer soft electrophiles
and hard nucleophiles prefer hard electrophiles.[2,3]
J Phys Org Chem. 2018;e3841.
wileyonlinelibrary.com/journal/poc
Copyright © 2018 John Wiley & Sons, Ltd.
1 of 8
https://doi.org/10.1002/poc.3841

2 of 8
CASERIO AND KIM
of 2 to 3 (Scheme 1). Experimental details are provided as
Whereas proximity of the sulfur and nitrogen atoms does
Supporting Information accompanying this article.
not appear to invert their relative basicity, or nucleophilicity,
it does influence the product distribution. For n = 0 and
n = 1, reactions with electrophiles resulted in complex
products of S─C and N─C cleavage.[1] However, with
n = 2, the initial products formed with the same electrophiles
did not result in immediate cleavage of the S or N
heteroatoms, implying that the hetero atoms are sufficiently
distant for their interaction to be minimal. However, several
features of the chemistry observed for the n = 2 system
proved to be interesting and worthy of further investigation.
This is the focus of the current study described here.
3 | DIMERIZATION AND ELIMINATION
On heating, a second notable reaction of 3‐BF4 occurred.
At 100°C in nitromethane or acetonitrile, 3‐BF4 rapidly
eliminated dimethylsufide and formed a high‐melting
white crystalline solid identified as N,N,N,N‐tetramethy
lpiperazinium fluoroborate 9‐2BF4 (Scheme 1).
The formation of 9 as the chloride double salt was pre-
viously reported by Bartlett et al from the structurally
related dimethyl‐(2‐chloroethyl)amine 5.[4] Kinetic stud-
ies led these workers to conclude that the transformation
involves the intermediacy of an aziridium ion 4 which
undergoes further reaction with 5 to form the correspond-
ing piperazinium double salt 9‐2Cl (Equation 1).
2 | REARRANGEMENT REACTIONS
AND ALKYL TRANSFER FROM
SULFUR TO NITROGEN
In earlier work,[1] we found that methylation of dimethyl‐
(2‐methylthioethyl)amine 1 with 1 equivalent of Me3O+
−
BF4 gave the ammonium salt 2‐BF4. There was no evi-
dence of S‐methylation of 1 to produce the isomeric sulfo-
nium salt 3‐BF4 although 1 reacted with 2 equivalents of
Me3O+ BF4 to give the ammonium sulfonium double
(1)
−
salt 6‐2BF4. The isomeric sulfonium salt 3‐BF4 was suc-
cessfully prepared by an independent route involving pro-
tection of the amine function of 1 by N‐protonation
followed by S‐methylation to the sulfonium‐ammonium
double salt 6‐HBF4.BF4 and deprotonation to the sulfo-
nium salt 3‐BF4 (Scheme 1). However, it was not possible
to obtain a pure sample of 3‐BF4 because it slowly
rearranged at ambient temperature to the isomeric
ammonium salt 2 by methyl transfer from sulfur to nitro-
gen. After 10 days, the ratio of 2 to 3 reached 3:2, but we
cannot be sure this represents the equilibrium composi-
tion as we have no evidence of the reverse rearrangement
Synthesis and isolation of stable aziridium salts have
been reported,[5] and mechanistic studies of piperazine
formation from aziridines have also appeared in the liter-
ature.[6] Thus, it seems entirely plausible that aziridium
intermediates may also be involved in the transformation
of 3‐BF4 to 9‐2BF4. The only difference between the Bart-
lett system and ours is the leaving group (Me2S vs Cl−).
We therefore decided to investigate the formation of 9
by deuterium labeling in the parent 3 to probe for the evi-
dence of an aziridium intermediate. We also used labeled
3 to investigate its rearrangement to 2.
SCHEME 1 Methylation Reactions of
Dimethyl‐(2‐methylthioethyl)amine 1

CASERIO AND KIM
3 of 8
transfer from 3‐d2‐BF4 to another 3‐d2‐BF4 to give 1‐d2
and a double salt 6‐d2‐2BF4 (Table 1). And, importantly,
after most of the 3‐d2 had rearranged to the isomeric
ammonium salt 2, there was no evidence of rearrange-
ment of the deuterium‐labeled methylene in either the
remaining reactant 3‐d2 or in the product 2‐d2.
The absence of crossover products and label shuf-
fling support our initial, seemingly obvious, conclusion
that rearrangement of 3 to 2 is intramolecular by way
of a 5‐membered transition state TS‐1, as represented
in Equation 2.
4 | SYNTHESIS OF DEUTERIUM‐
LABELED SULFONIUM SALT 3‐d‐2BF4
FROM DIMETHYL‐(2‐METHYLTHIOETHYL)
AMINE 1‐d2
Deuterium labeled 1‐d2 was prepared by the sequence
shown in Scheme 2. Dimethylamine with chloroacetyl
chloride gave the corresponding chloroacetamide. Dis-
placement of chloride with methanethiolate followed by
reduction with lithium aluminum deuteride (LAD) gave
1‐d2. It was necessary to protect the amine nitrogen of
1‐d2 by protonation with HBF4 to methylate 1‐d2 at sul-
fur to produce the sulfonium salt 3‐d2‐BF4.
5 | REARRANGEMENT OF 3‐d2
The methyl migration process of salt 3‐d2 was followed by
1H and 13C NMR using an equimolar mixture of 3‐d2 as
the BF4 salt with triethylamine in nitromethane. The pur-
pose of adding triethylamine was to determine whether,
under the conditions of rearrangement, 3 also transfers
an S‐methyl group to the triethylamine nitrogen. If so,
this would be evidence for an intermolecular methyl
transfer process. During 4 weeks at room temperature in
a sealed tube, the 1H signal intensities of 3‐d2‐BF4
(ppm; 2.29 (CH3)2N, 3.52 (CH2+S), and 2.91 (CH3)2S+)
slowly diminished as the intensities of those of the iso-
meric ammonium salt 2‐2d‐BF4 (ppm: 2.18 (CH3S), 2.96
(CH2S), and 3.22 (CH3)3N+) increased proportionally.
There was no evidence of methyl(triethyl)ammonium salt
formation. Also, there was no 1H NMR evidence of prod-
ucts that might be expected from intermolecular methyl
(2)
However, we were reminded by perceptive reviewers
that the SN2 transition state depicted in Equation 2 is a
5‐endo‐tet cyclic process that is highly unlikely for geo-
metric reasons. An SN2 transition state requires a large
bond angle between the nucleophile, the electrophile,
and the leaving group. That geometry is necessary for
the HOMO‐LUMO interacting orbitals to achieve bonding
through a 180° backside approach of the nucleophile,
which is not possible within the 5‐membered endocyclic
molecular structure of TS‐1.[3,7] We have to conclude that
the evidence in support of an intramolecular rearrange-
ment is inconclusive. To distinguish between intramolec-
ular vs intermolecular rearrangement would require
application of the endocyclic restriction test.[8] This test
SCHEME 2 Preparation and Methylation of Deuterium‐labeled Dimethyl‐(2‐methylthioethyl)amine 1‐d2

4 of 8
CASERIO AND KIM
TABLE 1 Proton chemical shifts (ppm) of ammonium and sulfonium salts
Structure
MeN+
MeN
MeS+
MeS
CH2S+
CH2S
CH2CO
CH2N
ClCH2CON(me)2
2.99(s)
2.94(s)
2.83(s)
2‐25(s)
4.07(s)
3.15(s)
MeSCH2CON(me)2
2.06(s)
2.12(s)
2.18(s)
MeSCH2CD2N(me)2
(1‐d2)
2.59(s)
2.95
MeSCH2CD2N(me)2HCl
(1‐D2‐HCl)
2.80(d)
3.16(d)
Me2SCH2CD2N(me)2H 2BF4
(6‐d2‐HBF4.BF4)
Me2SCH2CD2N(me)2 BF4
(3‐d2‐BF4)
3.12(s)
2.91(s)
3.82(S)
3.52(s)
2.29(s)
MeSCH2CD2N(me)3 BF4
(2‐d2‐BF4)
3.22(s)
2.18(s)
2.96(s)
8a‐d4
2.29(s)
2.29(s)
2.45(s)
2.45(s)
8b‐d4
9‐d4‐2BF4a
3.51(s)
4.04 (CH2N+)
Spectra were recorded at 300 MHz in CDCl3 for neutral compounds and in CD3NO2 solution for salts.
aProduced from 3‐d2‐BF4 as a mixture of 9a‐d4‐2BF4 and 9b‐d4‐2BF4.
has been used in related 5, 6, and 7‐endo cyclizations and
shown to support an intermolecular process.
6 | DIMERIZATION OF 3‐d2 WITH
ELIMINATION OF Me2S
Larger ring endocyclizations could accommodate the
large bond angle required for
a
typical SN2
The labeled fluoroborate sulfonium salt 3‐d2‐BF4 was
subjected to the same reaction conditions known to lead
to dimethylsufide elimination and dimerization of unla-
beled 3 to give the piperazinium double salt 9‐2BF4.
Determination of the label distribution in the labeled
product 9‐d4‐2BF4 by NMR was not possible, partly
because of low solubility but primarily because the spectra
of 9a and 9b are indistinguishable. The 1H NMR of a mix-
ture of 9a and 9b produced on heating 3‐d2‐BF4 showed
only 2 resonances corresponding to the (Me)3N+ reso-
nance at 3.51 ppm and the CH2N+ resonance at
4.05 ppm. Instead, we chose to use IR spectroscopy to ana-
lyze for the label distribution. This proved to be
displacement.[8b] A conceivable pathway for the conver-
sion of 3 to 2 as a double SN2 intermolecular rearrange-
ment by way of the transition state shown
diagrammatically as TS‐2 would be consistent with the
experimental data thus far. Synchronous methyl transfers
would amount to an allowed 10‐endo‐tet configuration for
TS‐2. However, double labeling experiments are required
for an endocyclic restriction test to definitively test for
crossover products anticipated from TS‐2. Rearrangement
by way of a methyl radical (a single electron transfer
(SET) mechanism) is also conceivable, but, as far as we
know, no positive evidence for SET cyclization has been
reported.[8b] But, in the absence of further experiments,
the mechanistic details remain unresolved.
SCHEME 3 Preparation of Deuterium‐labeled Piperazinium
Salts 9‐d4‐2BF4

CASERIO AND KIM
5 of 8
TABLE 2 Carbon‐13 chemical shifts (ppm) of ammonium and sulfonium salts
Structure
MeN+
MeN
MeS+
MeS
CH2S+
CH2S
CH2CO
CH2N
ClCH2CON(me)2
MeSCH2CD2N(me)2
(1‐d2)
38.25
45.97
16.11
16.40
35.65
32.5
169.34
MeSCH2CD2N(me)2HX
(1‐d2‐HCl)
43.18
45.49
15.60
28.44
Me2SCH2CD2N(me)2H 2X
(6‐d2‐HBF4.BF4)
Me2SCH2CD2N(me)2 X
(3‐d2‐BF4)
26.59
25.8
38.65
44.35
45.20
MeSCH2CD2N(me)3 X
(2‐d2‐BF4)
53.23
14.38
25.88
8a‐d4
54.95
54.87
45.95
45.98
8b‐d4
Spectra were recorded at 300 MHz in CDCl3 for neutral compounds and in CD3NO2 solution for salts
definitive, particularly in the fingerprint region of the IR
spectrum that is diagnostic of the vibrational modes of
the molecular structure as a whole. Aided by comparison
of the product mixture 9a‐ and 9b‐d4‐2BF4 with authen-
tic labeled samples prepared by independent methods,
we were able to determine the label distribution in 9‐d4.
Thus, lithium aluminum deuteride (LAD) reduction of
1,4‐N,N‐dimethyl‐2,6‐piperazinedione 7a to give 8a
salt product 9‐d4‐2Cl (formed from the dimerization of
3‐d2‐BF4 and conversion to the corresponding chloride
salt) was a mixture of 9a‐d4‐2Cl and 9b‐d4‐2Cl in equal
amounts. That is so say, the IR absorption bands of the
dimerization product could be closely reproduced from a
50:50 mixture of authentic samples of 9a‐d4‐2Cl and 9b‐
d4‐2Cl (see Figure 1a and 1b). Furthermore, label shuf-
fling in the reactant 3‐d2 can be ruled out as a source of
label rearrangement observed in the 9‐d4 product mixture
because no label shuffling in 3‐d2 was observed under the
reaction conditions. We conclude therefore that forma-
tion of mixtures of 9a‐d4 and 9b‐d4 as fluoroborates is
plausible evidence of a reaction pathway involving the
aziridium ion intermediate 4‐d2‐BF4 as summarized in
Scheme 4 with the caveat that 4 must be formed irrevers-
ibly from 3.
The mechanism in Scheme 4 appears to parallel that
suggested by Bartlett et al in their related studies of
piperazinium formation from dimethyl‐(2‐chloroethyl)
amine 5.[4] Indeed, we prepared the deuterium‐labeled
dimethyl‐(2‐chloroethyl)amine salt 5a‐d2‐HCl by LAD
reduction of ethyl N,N‐dimethylaminoacetate to give N,
N‐dimethylaminoethanol‐d2 followed by conversion to
protonated N,N‐dimethylamino‐2‐chloroethane‐d2 5a‐
d2‐HCl with thionyl chloride (see Supporting Informa-
tion). When deprotonated in nitromethane at room tem-
perature overnight, the incipient product 5 dimerized to
the dichloride salt 9‐d4‐2Cl. We identified the dichloride
salt to be a 50:50 mixture of 9a‐d4‐2Cl and 9b‐d4‐2Cl by
IR analysis and by comparison with the spectra of authen-
tic samples (Figures 1a and 1b). The result is consistent
followed by methylation with
2
equivalents of
trimethyloxonium fluroroborate gave 9a‐d4‐2BF4, while
the isomeric 2,5‐piperazinedione 7b on LAD reduction
to 8b and methylation gave 9b‐d4‐2BF4 (Scheme 3).
(See Tables 1, 2, and 3 for chemical shift data and the
prominent IR absorption frequencies). Note that the
NMR spectra of 8a and 8b are indistinguishable.
Figure 1 shows the IR spectra of 9a and 9b in the finger-
print region).
Because strong and broad IR absorption bands of the
–
BF4 anion interfered with analysis of the IR region of
interest in the labeled salt products, it was necessary to
convert the fluoroborate double salts 9‐2BF4 to the corre-
sponding chloride salts by metathesis with KCl. IR analy-
sis of the chloride double salts in KBr pellets revealed a
fine pattern of distinctive absorption bands for each deu-
terated isomer in the fingerprint region (800‐900 cm−1).
Careful comparison of related key absorption bands (fre-
quencies and intensities) evident in the IRs of authenti-
cally labeled samples of 9a‐d4‐2Cl and 9b‐d4‐2Cl
(Figure 1) with the labeled dimerization products 9‐d4‐
2Cl (Figure 1b) allowed for a reliable estimate of the label
distribution. The results of this analysis revealed that the

6 of 8
CASERIO AND KIM
TABLE 3 IR absorbances of tetramethylpiperazinium salts (KBr
pellet)
3010, 2970, 2236 (C‐D stretching),
1484, 1411, 1305
1038(BF4), 958, 954, 942, 885, 858, 840,
772 cm−1
3011, 2970, 2238 (C‐D stretching),
1618, 1478, 1326,
1300, 1038(BF4), 987, 942, 863, 807,
771 cm−1
Dimer salt
9‐d4‐2BF4
3011, 2969, 2237 (C‐D stretching),
1483, 1410, 1326,
1305, 1032 (BF4), 985, 9,54, 942, 884,
863, 840,807, 771 cm−1
3011, 2967, 1618, 1483, 1467, 1439,
1410, 1390,
1318, 1231, 1200, 1132, 1136, 1061, 1022,
972, 921, 872, 718, 634 cm−1
3010, 2970, 2236 (C‐D stretching),
1483, 1411, 1364, 1304, 1207, 1175,
1159, 1042, 1006, 984, 955, 924, 885,
858, 840, 814, 746 cm−1
3011, 2970, 2237(C‐D), 1478, 1410,
1326, 1221,
1197, 1049, 1011, 987, 944, 863, 807 cm
−1
Dimer salt cl
9‐d4‐2Cl
3011, 2969, 2236(C‐D), 1481, 1407,
1364, 1326,
(1303, 1256, 1221, 1197, 1175, 1158,
1042, 1006, 986, 954, 943, 885, 863,
839, 807 cm−1
FIGURE 1 (A) TOP: IR absorbances 800 to 900 cm−1 of A (9a‐d4‐
2Cl) 885, 858, 840; B (9b‐d4‐2Cl) 877, 863, 807; (B) BOTTOM: IR
absorbances 800 to 900 cm−1 of product mixture (as dichloride
double salts from heating 3‐d2‐BF4 (885, 864, 840, 806); 5‐d2 (885,
863, 840. 807); and a 50:50 mixture of authentic 9a and 9b‐d4‐2Cl
(885, 863, 840, 807)
Bold numbers indicate the differences in absorbances as a result of deuterium
labeling. Dimer salt data represent absorbances of product derived from the
labeled precursors, 3‐d2‐BF34 fluoroborate salt or 5‐d2.
Evidence from the deuterium labeling experiments
support the intermediacy of dimethylaziridinium ion 4
in the formation of 9.
The displacement step to form the aziridium interme-
diate 4 is a 3‐exo‐tet cyclization of the homoanomeric sys-
tem n(N) → sigma*(C ‐ X) known to be important for
favorable cyclization kinetics.[9]
with the intervention of the symmetrical intermediate
aziridium ion 4. However, NMR analysis of recovered
starting material 5 revealed that label rearrangement
had also occurred, which suggests that, in this case, the
aziridium intermediate 4 is formed reversibly from 5.
There is, however, a basic difference in the behavior of
4 depending on the nature of X. Apparently 4 is formed
reversibly from 5 (X = Cl) but irreversibly from 3
(X = Me2S). A likely explanation is that under aprotic
solvent conditions, the aziridium intermediate from 5 is
actually a tight ion pair that can readily return internally
7 | CONCLUSION
Heating (2‐substituted)‐ethylamines of the type
Me2NCH2CH2X where the X substituent is a good leaving
group (Cl− in 5, and SMe2 in 3) lead rapidly to elimination
of X and formation of tetramethylpiperizinium salt 9.

CASERIO AND KIM
7 of 8
SCHEME 4 Aziridium Ion
Intermediates in the Formation of
Piperazinium Salts
before capture to give the dimeric product 9. This would
certainly explain label shuffling observed in both reactant
and product. It is also consistent with the expectation that
the nucleophilicity of the chloride anion is enhanced in
aprotic solvents such as nitromethane or acetonitrile and
would compete favorably for capture of 4 over an
approaching amine nitrogen of 5 that would lead to the
product 9.
pathways, one via a 3‐exo‐tet path, the other by a 5‐endo‐
tet path.
In contrast, in the case of 3, the counter ion is non‐
nucleophilic (BF4−), and a comparable ion‐pair‐internal
return sequence involving 4 is not relevant. Because the
deuterium label is shuffled only in the product, recapture
of the aziridium intermediate 4 by the sulfide nucleophile
Me2S does not compete kinetically with capture of 4 by
the amine nitrogen of 3 to form the product. These results
are summarized in Scheme 4.
Finally, the behavior of the dimethylsulfonium salt 3
is worthy of further comment. There are 2 reactions of
3, one leading to the double salt 9 via the aziridium
ion 4 and the other to the isomeric ammonium salt 2.
In principle, both reactions could involve cyclization
As noted, the results reported here are consistent with
the known favorable kinetics of the 3‐exo‐tet process to
give 4. The alternative 5‐endo‐tet path to 2 is non‐compet-
itive kinetically and is disfavored for stereoelectronic rea-
sons.[3,7] Nevertheless, 3 does rearrange to 2, but not by
the 5‐endo‐tet route. The actual pathway is not entirely
clear and is slow compared with the kinetically favored
transformation of 3 to 4, but it demonstrates that 2 is ther-
modynamically more stable than 3.
In summary, the behavior of structure 3 reveals reac-
tions that illustrate both kinetic and thermodynamic
nucleophilicity of nitrogen compared with sulfur. Thus,

8 of 8
CASERIO AND KIM
nitrogen competes favorably with sulfur for the carbon
electrophile in the thermodynamically controlled rear-
rangement of 3 to 2. Also, nitrogen is apparently kineti-
cally dominant over sulfur in a 3‐exo‐tet cyclization
process, as evidenced by the facile elimination of Me2S
and formation of aziridium ion 4 on heating 3. A compa-
rable cyclization of the ammonium isomer 2 to eliminate
Me3N via an episulfonium intermediate was not observed.
This is consistent with the reported favorable kinetics of
homoanomeric n(X) → sigma*(C ‐ Y) systems in a 3‐exo‐
tet closure when the donor nucleophile X is nitrogen.[9]
REFERENCES
[1] J. K. Kim, Y. Souma, N. Beutow, C. Ibbeson, M. C. Caserio,
J. Org. Chem. 1989, 54, 1714.
[2] a.) R. G. Pearson, Chattara, Pratim, Chemtracts 2008, 21(1), 1;
b) R. G. Pearson, Coord. Chem. Rev. 1990, 100, 403.
[3] K. Fukui, Acc. Chem. Res. 1971, 4, 57.
[4] a) P. D. Bartlett, S. D. Ross, C. G. Swain, J. Am. Chem. Soc. 1947,
69, 2971; b) P. D. Bartlett, J. W. Davis, D. Ross, C. G. Swain, ibid
1949, 69, 2977; c) P. D. Bartlett, S. D. Ross, C. G. Swain, ibid 1947,
71, 1415.
[5] a) N. J. Leonard, J. V. Paukstelis, J. Org. Chem. 1965, 30, 821;
b) N. J. Leonard, J. V. Paukstelis, Tetrahedron 1969, 25, 1651.
[6] C. R. Dick, J. Org. Chem. 1967, 32, 72.
ACKNOWLEDGEMENT
[7] a) J. E. Baldwin, J. Chem. Soc. Chem. Commun. 1976, 18, 734;
b) J. E. Baldwin, J. Org. Chem. 1977, 42, 3846.
Legacy has many dimensions—tangible and intangible.
The seminal contributions to physical organic chemistry
from J. D. Roberts will remain his tangible legacy to sci-
ence. No less significant is the less tangible but equally
important legacy he leaves as an educator, mentor, advi-
sor, role model, and leader. As one of many beneficiaries,
I take this opportunity to express deep admiration and
gratitude for the opportunity he gave me to participate
as a member of his research group for 8 incredible years
(Marjorie C. Caserio) eventually leading to a faculty
appointment at the University of California Irvine, and
many productive years of research with colleague and
coauthor Jhong K. Kim.
[8] a) P. Beak, Acc. Chem. Res. 1992, 25, 215; b) P. Beak, K. C. Basu,
J. S. Li, J. Org. Chem. 1999, 64, 5218.
[9] I. V. Alabugin, M. Manoharan, T. A. Zeidan, J. Am. Chem. Soc.
2003, 125, 14014.
SUPPORTING INFORMATION
Additional Supporting Information may be found online
in the supporting information tab for this article.
How to cite this article: Caserio MC, Kim JK.
Studies in sulfur‐nitrogen nucleophilicity. J Phys
Org Chem. 2018;e3841. https://doi.org/10.1002/
poc.3841
ORCID
Marjorie C. Caserio http://orcid.org/0000-0002-2096-2237