Circumambulatory movement of negative charge ("ring walk") during gas-phase dissociation of 2,3,4-trimethoxybenzoate anion

Article abstract of DOI:10.1021/jo500234w

A dramatic "ortho effect" was observed during gas-phase dissociation of ortho-, meta-, and para-methoxybenzoate anions. Upon activation under mass spectrometric collisional activation conditions, anions generated from all three isomers undergo a CO₂ loss. Of the m/z 107 ions generated in this way, only the 1-dehydro-2-methoxybenzene anion from the ortho isomer underwent an exclusive formaldehyde loss. A peak for a formaldehyde loss in the spectra of 2,4-, 2,5-, and 2,6-dimethoxybenzoates and the absence of an analogous peak from 3,4- and 3,5-dimethoxy derivatives confirmed that this is a diagnostically useful ortho-isomer-specific phenomenon. Moreover, the spectrum from 2,3-dimethoxybenzoic acid showed peaks for two consecutive formaldehyde losses. The 1-dehydro-2,3,4-trimethoxybenzene anion (m/z 167) generated from 2,3,4-trimethoxybenzoate in this way endures three consecutive eliminations of formaldehyde units. For this, the negative charge, initially located on position 1, circumambulates to position 2, then to position 3, and finally to position 4 to form the final phenyl anion. The proposed stepwise fragmentation pathway, which resembles the well-known E1cB-elimination mechanism, is supported by tandem mass spectrometric observations made with 2-[¹³C ²H₃]methoxy-3-[¹³C]methoxy-4-methoxybenzoic acid, and ab initio calculations. In addition, the spectra of ions such as 1-dehydro-3,4-dimethoxybenzene anion show peaks for consecutive methyl radical losses, a feature that establishes the 1,2-relationship between the two methoxy groups.

Full text of DOI:10.1021/jo500234w

Article
pubs.acs.org/joc
Circumambulatory Movement of Negative Charge (“Ring Walk”)
during Gas-Phase Dissociation of 2,3,4-Trimethoxybenzoate Anion
Kithsiri B. Herath,†,‡ Carl S. Weisbecker,† Sheo B. Singh,‡ and Athula B. Attygalle*,†
†Center for Mass Spectrometry, Department of Chemistry, Chemical Biology, and Biomedical Engineering, Stevens Institute of
Technology, Hoboken, New Jersey 07030, United States
‡Merck Research Laboratories, Rahway, New Jersey 07065, United States
S
* Supporting Information
ABSTRACT: A dramatic “ortho effect” was observed during gas-phase dissociation of ortho-,
meta-, and para-methoxybenzoate anions. Upon activation under mass spectrometric collisional
activation conditions, anions generated from all three isomers undergo a CO2 loss. Of the m/z 107
ions generated in this way, only the 1-dehydro-2-methoxybenzene anion from the ortho isomer
underwent an exclusive formaldehyde loss. A peak for a formaldehyde loss in the spectra of
2,4-, 2,5-, and 2,6-dimethoxybenzoates and the absence of an analogous peak from 3,4- and
3,5-dimethoxy derivatives confirmed that this is a diagnostically useful ortho-isomer-specific
phenomenon. Moreover, the spectrum from 2,3-dimethoxybenzoic acid showed peaks for two
consecutive formaldehyde losses. The 1-dehydro-2,3,4-trimethoxybenzene anion (m/z 167)
generated from 2,3,4-trimethoxybenzoate in this way endures three consecutive eliminations of
formaldehyde units. For this, the negative charge, initially located on position 1, circumambulates
to position 2, then to position 3, and finally to position 4 to form the final phenyl anion. The
proposed stepwise fragmentation pathway, which resembles the well-known E1cB-elimination
mechanism, is supported by tandem mass spectrometric observations made with 2-[13C2H3]methoxy-3-[13C]methoxy-4-
methoxybenzoic acid, and ab initio calculations. In addition, the spectra of ions such as 1-dehydro-3,4-dimethoxybenzene anion
show peaks for consecutive methyl radical losses, a feature that establishes the 1,2-relationship between the two methoxy groups.
derivatives with multiple methoxy groups are widely encountered
as secondary metabolites. Some examples are narceine from fungi
(1),6 sekikaic acid from lichens (2),7 and the antibiotic Thielavin
B from the marine fungus Thielavia terricola (3).8 Identification
of these compounds is not usually straightforward because their
isolation in pure form is rather difficult. In addition, the deter-
mination of ring-substitution pattern by NMR spectroscopy is
INTRODUCTION
■
Electrospray ionization, one of the most widely used sample
introduction techniques in modern mass spectrometry, generates
predominantly even-electron (EE) gaseous ions.1 The gaseous
ions generated in this way undergo little or no fragmentation
because electrospray imparts only very little internal energy to
ions. To obtain structural information from EE ions, they are
usually mass selected by tandem mass spectrometry and subjected
to collision-induced dissociation (CID). Although CID spectra
often appear simple, their interpretation is not always straightfor-
ward because of many rearrangements that take place during
fragmentation. Many attempts have been made to formulate frag-
mentation rules based on empirical observations and use artificial
intelligence (AI) to interpret CID spectra of even-electron ions.2,3
Similar efforts have been made without much success for decades
to interpret electron ionization (EI) mass spectra. Although AI
approaches are successful for sequence determination of oligo-
meric compounds such as peptides, they offer little help for un-
ambiguous identification of novel small molecules. Undoubtedly,
before attempting to develop meaningful AI programs for the
interpretation of EE spectra,2 we must first thoroughly com-
prehend the underlying fragmentation pathways.
In the course of our explorations on fragmentation mech-
anisms of low-molecular-weight anions,4,5 we noted an ortho-
isomer-specific loss of formaldehyde from methoxybenzoic
acids isomers. Understanding the fragmentation pathways of
this group of compounds is important because benzoic acid
Received: February 4, 2014
Published: April 21, 2014
© 2014 American Chemical Society
4378
dx.doi.org/10.1021/jo500234w | J. Org. Chem. 2014, 79, 4378−4389

The Journal of Organic Chemistry
Article
Figure 1. Unit-resolution CID mass spectra recorded from the m/z 151 anion generated from (a) o-, (c) m-, and (e) p-methoxybenzoic acid, and (b, d,
and f) those of their respective in-source-generated m/z 107 ions at a collision energy setting of 20 eV (note: the smaller m/z 107 ion undergoes more
fragmentation at 20 eV than the larger m/z 151 ion).
cumbersome due to the presence of only a few aromatic protons.
In contrast, mass spectrometric analysis can be performed
directly on crude mixtures.
Scheme 1. Products from Collision-Induced Dissociation of
the ortho-, meta-, and para-Methoxybenzoate Anions
Thus, we undertook a comprehensive investigation on
interpretation of CID mass spectra of congeners of methoxy-
benzoic acids. Our initial results showed that the negative charge
of the anion, generated by the incipient decarboxylation, moves
in a circumambulatory manner when the methoxy group is
adjacent to the ring carbon bearing the charge. By definition,
a circumambulatory rearrangement is a type of unimolecular
chemical reaction in which a chemical entity undergoes a dis-
placement around the periphery of a conjugated ring.9 These
mechanisms are sometimes described more colorfully as a
“merry-go-round” or as “ring walking” rearrangements.10,11 The
mechanism of circumambulation is a well-debated subject in the
study of gas-phase chemistry of aromatic ions. A 1,2-hydrogen
shift mechanism has been proposed as one possible pathway to
rationalize circumambulation in 1-dehydrobenzene anions.12,13
Ben-Ari et al. have reviewed the chemistry of the 1,2-hydrogen
shift isomers of several aryl cations and aryl anions.14 They
reported that most hydrogen shift isomers are kinetically stable
with high energetic barriers to unimolecular isomerization in the
gas phase. In this study, we discuss a more facile “ring walking”
mechanism with a lower energy barrier to rationalize gas phase
dissociation of methoxyphenyl anions. The ion fragmentation
4379
dx.doi.org/10.1021/jo500234w | J. Org. Chem. 2014, 79, 4378−4389

The Journal of Organic Chemistry
Article
Scheme 2. Proposed Mechanism for the Loss of
Formaldehyde from the 1-Dehydro-2-methoxybenzene Anion
(7) Produced by Decarboxylation of ortho-Methoxybenzoate
Anion (4)
RESULTS AND DISCUSSION
■
A comparison of CID spectra recorded from the m/z 151 anions
generated from ortho- (4), meta- (5), and para-methoxybenzoic
acid (6) indicated that all three isomers initially underwent a
facile expulsion of a CO2 molecule to produce ions of m/z 107
(Figure 1a,c,e). However, the spectrum recorded from the ortho
isomer was distinctly different from those of the meta and para
isomers. For example, the product ion spectrum of the ortho-
methoxybenzoate anion showed an additional peak at m/z 77,
which indicated that this compound underwent a fragmentation
mechanism specific to the ortho isomer (Figure 1a). To inves-
tigate this phenomenon further, we mass selected the hydrogen-
shift isomeric anions12 produced by decarboxylation of the ortho,
meta, and para precursors and subjected them to collision-
induced dissociation (Figure 1b,d,f). The results showed that
only the spectrum of the anion 7 generated from the ortho
isomer showed a peak at m/z 77 for a loss of a 30 Da neutral
molecule (Figure 1b). We envisaged that this loss of a 30 Da
neutral molecule represents an elimination of a formaldehyde
molecule by an ortho-isomer-specific fragmentation mechanism.
In contrast, the spectra of meta- and para-isomers did not show a
tenets concluded from this investigation are diagnostically useful
for structure elucidation of methoxybenzoic acid congeners.15
Figure 2. Unit-resolution CID mass spectra recorded from the m/z 181 anion generated from (a) 3,5- and (c) 3,4-dimethoxybenzoic acid, and (e) that
of the m/z 187 anion generated from 3,4-[C2H3]dimethoxybenzoic acid. The product ion spectra (b and d) were recorded from in-source-generated
m/z 137 ions from 3,5-, and 3,4-dimethoxybenzoic acid, respectively (all spectra were recorded at 20 eV, except the one shown in panel e, which was
recorded at a higher collision energy setting).
4380
dx.doi.org/10.1021/jo500234w | J. Org. Chem. 2014, 79, 4378−4389

The Journal of Organic Chemistry
Article
Figure 3. Unit-resolution CID mass spectra, recorded at a collision energy setting of 20 eV, from the m/z 181 anion generated from (a) 2,3-, (c) 2,4-, and
(e) 2,5-dimethoxybenzoic acid, and (b, d, and f) those of their respective in-source-generated m/z 137 ions. Note: the smaller m/z 137 ion undergoes
more fragmentation at 20 eV than the larger m/z 181 ion.
peak at m/z 77 for a formaldehyde loss; instead, peaks were
observed at m/z 92 for methyl radical losses (Scheme 1).
Subsequent experiments confirmed that the positional
integrity of the constituent atoms in the m/z 107 product ions
remain unaffected within the time duration of the tandem mass
spectrometric fragmentations conducted inside a collision cell of
a tandem-quadrupole instrument. In other words, the hydrogen-
shift isomers 7, 8, and 9 do not undergo isomerization to a stable
common structure. This is noteworthy because, for example, iso-
meric C7H7O− ions generated from anisole have been reported
to undergo major interconversions upon collisional activation.16
Hydrogen scrambling in positively charged aromatic systems and
protonated arenes, particularly by 1,2-hydride transfer mecha-
nisms, is a widely studied phenomenon.17 Fragmentation experi-
ments18−21 and quantum mechanical calculations for negative
ions have shown that the energy barriers for proton-shift iso-
merization are exceedingly high.14 Nevertheless, 1,2-proton
transfers have been suggested to sometimes occur in phenyl
anions.13,22
From our results, it was clear that a methoxy group should be
present at the ring position ortho to the negative charge-bearing
carbon atom in order to eliminate a formaldehyde molecule
because neither anion 8 nor 9 underwent a formaldehyde
loss. This is a diagnostically useful generalization because many
methoxy-substituted aromatic carboxylic acids are found as
natural products (Structures 1−3) and synthetic intermediates.
A mechanism analogous to a pathway that had been previously
proposed for the loss of an aldehyde molecule from anions
generated from O-alkyl ethers of ortho-hydroxybenzoic acids may
be postulated to rationalize this ortho-specific elimination of
formaldehyde.5 The carbanion-mediated two-step mechanism
proposed here resembles the E1cB mechanism known in organic
chemistry.23 After the initial decarboxylation, the o-methoxy-
phenyl anion that is formed (7) undergoes an internal proton
transfer from the methyl group to give the intermediate 13. A
charge-mediated β-scission of the PhO bond of anion 13 then
forms a phenyl anion (m/z 77) and eliminates a formaldehyde
molecule (Scheme 2).
To establish a general tenet that can be used for structure
elucidation, we conducted further experimentation with the six
isomeric dimethoxybenzoic acids. The results revealed that only
certain specific anions produced by the initial decarboxylation fail
to undergo the formaldehyde loss. For example, neither the
spectrum of 3,5-dimethoxybenzoate anion nor that of its
decarboxylated product shows a peak for a 30 Da formaldehyde
loss (Figure 2a,b). This was predicted in our generalization
because 3,5-dimethoxybenzoate does not bear a methoxy group
on the ring carbon adjacent to the carboxyl group. Instead, the
3,5-dimethoxyphenyl anion lost a methyl radical to show a
peak m/z 122 by a fragmentation mechanism analogous to
that proposed for meta- and para-methoxyphenyl anions
(Figure 2a,b).
The spectrum 3,4-dimethoxybenzoate anion, on the other
hand, showed a small peak at m/z 107 (Figure 2c). The peak at
4381
dx.doi.org/10.1021/jo500234w | J. Org. Chem. 2014, 79, 4378−4389

The Journal of Organic Chemistry
Article
Figure 4. Unit-resolution CID mass spectra, recorded at a collision energy setting of 20 eV, from the m/z 181 anion generated from (a) 2,6-
dimethoxybenzoic acid and (c) the m/z 184 ion from 2-methoxy-6-[C2H3]methoxy benzoic acid (15), and (b and d) those of their respective in-source-
generated m/z 137 and 140 ions. Note: spectra of 2,6-dimethoxybenzoic acid were recorded at 20 eV, and those of 2-methoxy-6-[C2H3]methoxy
benzoic acid were recorded at a higher collision energy setting.
m/z 107 in the product ion spectrum of the m/z 137 ion, which
was generated in the ion source from 3,4-dimethoxybenzoate
anion, appeared to indicate a loss of a 30 Da unit from the
precursor ion (Figure 2d). At a first look, the 3,4-dimethoxy-
benzoate anion seemed to violate our generalization about
formaldehyde loss because the 1-dehydro-3,4-dimethoxyben-
zene anion does not bear a methoxy group adjacent to the
charge-bearing carbon. The product ion spectrum recorded from
the m/z 137 ion showed two intense peaks at m/z 122 and 107
(Figure 2d). However, we could subsequently demonstrate that
the m/z 107 peak represents two consecutive methyl radical
losses (Figure 2d) by a homolytic fragmentation process that
blatantly violates the so-called EE rule advocated by some
spectrometry textbooks.24 On the other hand, many violations of
this rule have been documented.24−29 The formation of an m/z
107 ion from a 3,4-dimethoxybenzoate anion can be attributed
to a formation of an ortho-quinone anion and not to a
formaldehyde loss from the m/z 137 (Figure 2d). To confirm
that the m/z 107 ion from 3,4-dimethoxybenzoate is not due to a
formaldehyde loss, we synthesized 3,4-di[C2H3]methoxybenzoic
acid (14) and recorded its CID spectrum. In fact, the m/z 187 ion
generated from acid 14 initially lost CO2 to produce an m/z
143 ion, which then underwent two consecutive CD3 radical
losses (and not a [2H2]formaldehyde loss) to produce the ion
for m/z 107 (Figure 2e). Thus, the formation of an m/z 107 ion,
which is not due to a formaldehyde loss, suggested that the
1,2-relationship of the two methoxy groups can be established
by this charge-remote two-step radical loss. For example, the
spectrum of the 3,5-dimethoxybenzoate anion shows a peak only
for one methyl radical loss, although it bears two methoxy groups
because the methoxy groups are not adjacent to each other
(Figure 2a). In contrast, the spectrum of the 3,4-dimethoxy-
benzoate anion shows a peak for two methyl radical losses because
the methoxy groups are adjacent to each other (Figure 2d).
The spectrum of 2,3-dimethoxybenzoic acid, a compound with
two methoxy groups adjacent to the carboxyl group, showed two
peaks at m/z 107 and 77 for two consecutive formaldehyde losses
(Figure 3a,b). In contrast, the spectra of 2,4- and 2,5-dimeth-
oxybenzoic acids showed only one peak for a formaldehyde loss,
confirming that only the methoxy group adjacent to the COOH
group participates specifically in the elimination (Figure 3c−f).
The unanticipated finding was that the m/z 137 ion generated
from 2,6-dimethoxybenzoic acid underwent only one form-
aldehyde loss although the compound bears two methoxy groups
on both sides of the carboxylate group (Figure 4a,b). This obser-
vation provided unequivocal evidence to conclude that the
negative charge located on a phenyl ring does not “ring walk” by a
1,2-proton shift mechanism. Our results are congruent with
Ben-Ari et al.’s observations on hydrogen-shift isomers of 2-, 3-,
and 4-dehydrobenzenesulfonates.14 They reported that the energies
of the transition states involved for the hydrogen migration are
much higher than those of the other competing pathways.14
The peak observed at m/z 107 in the spectrum of the 1-dehydro-
2,6-dimethoxybenzene anion, formed upon activation of the
2,6-dimethoxybenzoate anion, confirmed that it undergoes an
elimination of formaldehyde (Figure 4a; Scheme 3). In the anion
formed in this way (1-dehydro-3-methoxybenzene anion 20), the
charge is specifically located on the carbon atom meta to the
methoxy group, thereby preventing a second formaldehyde
elimination. In other words, the negative charge does not “ring
walk” if the methoxy group is not located at an adjacent carbon.
To confirm the elimination mechanism, we synthesized the
deuteriated analogue, 2-methoxy-6-[C2H3]methoxy benzoic acid
(15) and recorded its mass spectrum. Figure 4c,d illustrates that
4382
dx.doi.org/10.1021/jo500234w | J. Org. Chem. 2014, 79, 4378−4389

The Journal of Organic Chemistry
Article
that of the starting intermediate anion 18. Intermediate 19
formed in this way, then eliminated a formaldeyde molecule via
the transition state TS19−20 to generate the 1-dehydro-3-
methoxybenzene anion (m/z 107; 20). We also evaluated ener-
getics associated with other alternative procedures, including a
possible concerted cyclic E2 mechanism. However, a transition
state to support a concerted mechanism could not be located.
Instead, we consistently found two transition states that were
associated with the two-step process of eliminating form-
aldehyde.
Scheme 3. Gas-Phase Fragmentation of
2,6-Dimethoxybenzoate Anion
In the spectrum of 2-methoxy-6-[C2H3]methoxybenzene anion
(Figure 4c), the intensity of the m/z 108 peak was significantly
lower than that at m/z 110, which indicated a kinetic isotope
effect (KIE) of the magnitude 1.67. Figure 6 and Table 1 show
the optimized structures and KIE values calculated by QM
computations of dissociation products of 18a and 18b, which are
two conformers of 1-dehydro-2-methoxy-6-[C2H3]methoxy-
benzene. The Gibbs free energies for a transition state (G⧧)
and for a reactant (G) were calculated as described in the
Computational Methods section. The ΔG⧧ is equal to the
difference between G⧧ and G converted to kilocalorie per mole
(kcal/mol) units. KIE was calculated as the ratio of two rate
constants kH/kD for protonated and deuteriated reactants by the
procedure described by Zeller and Strassner.30,31 Calculated KIE
values (kH/kD) by QM computations for ion dissociative steps
18 → TS18−19 and 19 → TS19−20 were 6.6 and 1.3, respec-
tively (Figure 6; Table 1). Despite high margins of error, a KIE
value as large as 2.0−4.0 or greater is considered to demonstrate
the participation of a cleavage of a C−H/D bond in the rate-
determining step.32 Our experimental value of 1.67 (Figure 4d)
indicates only a secondary isotope effect, which suggests that in
the anion generated by the initial decarboxylation eliminates
either a H2CO or a D2CO molecule to generate the m/z
110 or 108 ions, respectively.
To determine the relative energies associated with the dis-
sociative steps of 2,6-dimethoxy benzoate (16), a series of QM
computations were carried out using the B3LYP density
functional and the 6-311++G(d,p) basis set. The results are
summarized in Figure 5. In the first step, a molecule of CO2 is
eliminated in a barrierless step to form the 1-dehydro-2,6-
dimethoxybenzene anion (m/z 137; 17) as an intermediate. The
calculated relative energy of intermediate 17 formed by the CO2
elimination is about 35.3 kcal/mol above that of the starting
carboxylate anion 16. Intermediate 17 then twists into a more
stable intermediate (18) via the transition state TS17−18.
Intermediate 18 then transfers a proton to the charge center on
the ring via 5-membered transition state TS18−19. The energy
of the transition state TS18−19 lies about 19.6 kcal/mol above
Figure 5. Relative energies (in kcal/mol) and structures of energy-optimized product ions and transition states associated with the dissociation of
2,6-dimethoxybenzoate anion (16).
4383
dx.doi.org/10.1021/jo500234w | J. Org. Chem. 2014, 79, 4378−4389

The Journal of Organic Chemistry
Article
Figure 6. Energy-optimized species of dissociation products of 18a and 18b, two conformers of 1-dehydro-2-methoxy-6-[C2H3]methoxybenzene anion
(yellow-colored atoms represent deuterium).
Table 1. KIE Values (kH/kD) Calculated by QM Computations for the Two Ion Dissociative Steps 18 → TS18−19 and 19 → TS19−
20 for Two Conformers of 1-Dehydro-2-methoxy-6-[C2H3]methoxybenzene (18a and 18b) and the Two Intermediate Products
(19a and 19b), Respectively
dissociative step
G (hartrees)
G⧧ (hartrees)
ΔG⧧ (kcal/mol)
k (s−1
)
t1/2 (s−1
)
kH/kD
18a → TS18a−19a
18b → TS18b−19b
19a → TS19a−20a
19b → TS19b−20b
−460.702 389
−460.702 409
−460.687 501
−460.686 909
−460.670 482
−460.668 727
−460.649 093
−460.648 256
20.0
21.1
24.1
24.3
1.31 × 10−2
2.00 × 10−3
1.34 × 10−5
1.04 × 10−5
52.8
346.0
6.6
51611.2
66900.5
1.3
the rate-determining step (the H2CO elimination from 19), a
direct cleavage of a C−H bond is not involved. This agrees well
with the computational prediction that the energy barrier
between 18 and 19 (the C−H bond cleavage via TS18−19) is
lower than that between 19 and 20 (formaldehyde elimination
via TS19−20). Presumably, more accurate estimates for the free
energy of activation may be calculated by advanced methods;
however, the procedure we used here is widely accepted as the
general method for interpreting experimentally determined KIEs
and eventually relating the calculated values to proposed reaction
mechanisms.31
Furthermore, our results from the 1-dehydro-2,6-dimethoxy-
benzene 17, and 2-methoxy-6-[C2H3]methoxybenzene anions
showed that the formaldehyde elimination follows a specific
pathway without any indication of a loss of integrity due to scram-
bling. In a previous report, however, Kleingeld and Nibbering33
concluded that there is an equilibrium between the three
deuterium atoms on the methoxy group and the ortho-aryl hydro-
gens prior to the loss of formaldehyde during fragmentation
of the 1-dehydro-2-[C2H3]methoxybenzene anion (m/z 110).
Apparently, the m/z 110 ion, isolated in an FT-ICR cell, showed
peaks for two nearly equal losses of CD2O (32 Da) or CDHO
(31 Da).33 In contrast, as mentioned above, our results were ob-
tained on a tandem quadrupole instrument, and upon activation,
the 2-methoxy-6-[C2H3]methoxybenzene anion showed peaks
only for a loss of CD2O (32 Da) or CH2O (30 Da) (Figure 4d).
Clearly, the overall fragmentation processes that the 1-dehydro-
2-[C2H3]methoxybenzene anions undergo are different in the
two instruments. In an FT-ICR instrument (and other ion-
trapping instruments), mass-isolated ions are activated for longer
periods of time than in tandem quadrupole instruments.
Moreover, isolated ions in ion-trapping instruments, particularly
in quadrupolar ion traps, are prone to additional ion/molecule
Figure 7. Unit-resolution CID mass spectra recorded for 3,4,5-
trimethoxybenzoic acid generated from (a) the m/z 211 anion and
(b) its in-source-generated m/z 167 ion.
reactions with residual molecules such as water.34 Evidently, the
integrity of the 1-dehydromethoxybenzene anions are not lost
4384
dx.doi.org/10.1021/jo500234w | J. Org. Chem. 2014, 79, 4378−4389

The Journal of Organic Chemistry
Article
Figure 8. Unit-resolution CID mass spectra recorded from the m/z 211 anion generated from (a) 2,4,6-, (c) 2,4,5-, and (e) 2,3,6-trimethoxybenzoic
acid, and (b, d, and f) those of their respective in-source-generated m/z 167 ions.
during the brief activation period in tandem quadrupole
instruments. Moreover, at a first look, the calculated rate-limiting
rate constants may appear to be too large for a formaldehyde loss
during the short ion residence time in the collision cell. How-
ever, once the precursor ion is activated in the collision cell, it
undergoes a rapid loss of formaldehyde within the dwell time in
the collision cell (Table 1). If collision energy is not applied, then
the ion does not eliminate formaldehyde during its residence
time in the cell.
Scheme 4. Gas-Phase Fragmentation of
2,3,6-Trimethoxybenzoate Anion
The spectra recorded from trimethoxybenzoic acids provided
further validation that our generalizations are applicable to many
other examples. The spectrum of 3,4,5-trimethoxybenzoate
showed a peak at m/z 167 for a CO2 loss followed by peaks at
m/z 152 and 137 for two consecutive methyl radical losses to
form the predicted orthoquinone anion (Figure 7a). In contrast,
1-dehydro-2,4,6-trimethoxybenzene anion generated from 2,4,6-
trimethoxybenzoate underwent only one formaldehyde loss
(Figure 8a,b). The 1-dehydro-2,4,5-trimethoxybenzene anion
generated from 2,4,5-trimethoxybenzoate, on the other hand,
showed a peak at m/z 137 for a formaldehyde loss, followed by
peaks at m/z 122 and 107 for two consecutive methyl radicals
losses (Figure 8c,d). The spectrum of 1-dehydro-2,3,6-
trimethoxybenzene anion generated from 2,3,6-trimethoxyben-
zoic acid, with two methoxy groups adjacent to the carboxyl
group, showed a peak at m/z 137 for an initial formaldehyde loss.
The ion of nominal mass m/z 107 is a composite because it can
originate from a second formaldehyde loss or from two succes-
sive methyl radical losses (Figure 8e,f; Scheme 4).
The fragmentation of 2,3,4-trimethoxybenzoate ion generated
by deprotonation is the most interesting case. As we envisaged,
the 1-dehydro-2,3,4-trimethoxybenzene anion (m/z 167)
formed by decarboxylation underwent three consecutive form-
aldehyde losses and produced ions for peaks observed at m/z
4385
dx.doi.org/10.1021/jo500234w | J. Org. Chem. 2014, 79, 4378−4389

The Journal of Organic Chemistry
Article
Figure 9. Unit-resolution CID mass spectra recorded from the m/z 211 anion generated from (a) 2,3,4- trimethoxy benzoic acid, (c) m/z 216 anion
generated from 2-[13C2H3]methoxy-3-[13C]methoxy-4-methoxybenzoic acid, (e) and m/z 213 anion generated from 2,3-di[13C]methoxy-4-
methoxybenzoic acid, and (b, d, and f) those of their respective in-source-generated m/z 167, 172, and 169 ions.
137, 107, and 77 (Figure 9a,b; Scheme 5). To confirm the cir-
cumambulatory movement of the ring charge, we synthesized
2-[13C2H3]methoxy-3-[13C]methoxy-4-methoxybenzoic acid
(21). The spectrum shows a peak at m/z 139 for the loss of
the initial 33 Da molecule (2H213CO). The m/z 139 ion then
lost a 31 Da molecule (H213CO), followed by an elimination of
30 Da molecule (H2CO) (Figure 9b,c). Analogously, the
spectrum of 2,3-[13C]dimethoxy-4-methoxybenzoic acid (22)
showed peaks at m/z 138, 107, and 77 for three consecutive
losses of H213C1O, H213CO, and H2CO, respectively
(Figure 9e,f).
which then loses formaldehyde to form a phenyl anion. If addi-
tional methoxy groups were available on adjacent positions, then
the negative charge circumambulates by successive formalde-
hyde eliminations. In contrast, 1-dehydro-3-methoxybenzene
and 1-dehydro-4-methoxybenzene anions do not eliminate form-
aldehyde but lose a methyl radical by homolytic cleavage.
EXPERIMENTAL SECTION
■
Materials and Methods. All chemicals were used as purchased
from commercial suppliers. Reactions were monitored by thin layer
chromatography. Purification of synthetic sample was performed by
flash chromatography on silica gel (particle size 60 μm). Structures of all
synthetic products were confirmed by H NMR, 13C NMR, and mass
1
CONCLUSIONS
■
spectrometry. NMR spectra were recorded on a 500 MHz spectrometer
(500 MHz for 1H and 125 MHz for 13C acquisitions). Chemical shifts
(δ) are reported in parts per million with tetramethylsilane or the
solvent resonance as the internal standard. Coupling constants (J) are
given in hertz. Multiplicities are classified as follows: s, singlet; d,
doublet; t, triplet; q, quartet; sept, septet; combinations thereof; m,
multiplet; or br, broad signal. Two-dimensional NMR methods (HSQC,
HMBC) were used for the assignment. EI mass spectra (70 eV) were
recorded by GC−MS. Collision-induced dissociation (CID) mass
spectra were recorded on a triple-quadrupole tandem mass spec-
trometer equipped with a Z-spray electrospray ion source. Samples were
infused as acetonitrile−water−NH3 (1:1:2 ×10−3) solutions at a flow
Upon activation in the collision cell of a tandem mass spec-
trometer, a negative charge located on a ring carbon atom of
a phenyl group does not easily undergo scrambling by a 1,2-
proton-shift to lose the positional integrity of ring protons.
However, if a methoxy group is attached to the ring carbon atom
adjacent to the charge site, then the charge migrates to the
adjacent position by losing a formaldehyde molecule by a step-
wise process. For example, a proton transfer from the methoxy
group of the 1-dehydro-2-methoxybenzene anion to the charge
site generates a “hydrogen-shift” isomer as an intermediate,
4386
dx.doi.org/10.1021/jo500234w | J. Org. Chem. 2014, 79, 4378−4389

The Journal of Organic Chemistry
Article
Acid (22). Compounds 21 and 22 were synthesized by the procedure
summarized in Scheme 6.
Scheme 5. Consecutive Losses of Three Formaldehyde Units
from the 2,3,4-Trimethoxyphenyl Anion Produced by
Decarboxylation of 2,3,4-Trimethoxybenzoate Anion
Methyl 2,3-Dihydroxy-4-methoxybenzoate (24). An ethereal solu-
tion of diazomethane (3 mL, ∼250 mM) was added to a solution of
2,3-dihydroxy-4-methoxy benzoic acid (23, 95 mg, 0.5 mmol) in THF
(2 mL) at −78 °C. The mixture was stirred for 30 min at −78 °C and
quenched with acetic acid (50 μL). The solvent was evaporated to yield
24 as a white solid (100 mg, 98%). 1H NMR (500 MHz, CD3OD) δH
6.59 (1H, d, J = 9.0, H-5), 7.39 (1H, d, J = 9.0, H-6), 3.89 (3H, s,
4-OCH3), 3.90 (3H, s, COOCH3); 13C NMR (125 MHz, CD3OD) δc
172.0 (COO), 52.5 (COOCH3), 106.0 (C-1), 135.0 (C-2), 135.0 (C-3),
154.0 (C-4), 104.0 (C-5), 122.0 (C-6), 56.6 (4-OCH3); methyl signal at
δc 52.5, δH 3.90 showed three bond HMBC correlations to the carbonyl
group at δc 172; EI m/z (%) 198 (M+•, 58), 167 (24), 166 (77), 148
(67), 137 (32), 120 (100), 109 (10), 95 (37), 66 (8), 53 (29).
Methyl 2-Hydroxy-3-[13C]methoxy-4-methoxybenzoate (25) and
Methyl 2,3-[13C]Dimethoxy-4-methoxybenzoate (26). To a mixture of
24 (60 mg, 0.3 mmol) and Cs2CO3 (2 equiv) in anhydrous THF (2 mL)
at rt under N2 was added 13CH3I (2 equiv), and the mixture was stirred
for 50 h. The reaction was quenched with 1 M HCl (5 mL), and the
product was extracted into ethyl acetate. The separated ethyl acetate
layer was concentrated to afford a solid (95 mg), which was purified by
flash chromatography (silica gel, 10% ethyl acetate/hexane containing
1% acetic acid). Then, 5 mL fractions were collected, factions 33−45 and
50−64 were pooled, and solvent was evaporated to yield a colorless
oily material (26, 6.6 mg, 10%) and a white solid (25, 20 mg, 32%),
respectively.
Compound 25. 1H NMR (500 MHz, CD3OD) δH 6.61 (1H, d, J =
9.0, H-5), 7.60 (1H, d, J = 9.0, H-6), 3.88 (3H, s, 4−OCH3), 3.90
(3H, s, COOCH3), 3.78 (3H, d, J = 144 Hz, 3-O13CH3); 13C NMR
(125 MHz, CD3OD) δc 171.8 (COO), 52.8 (COOCH3), 108.8 (C-1),
156.9 (C-2), 137.7 (C-3), 159.7 (C-4), 104.8 (C-5), 127.2 (C-6), 61.0
(3-O13CH3), 56.7 (4-OCH3). The positions of the methoxy groups in
25 were confirmed by the strong three-bond HMBC correlations
observed from H-5 and H-6 to C-3 and C-2, respectively, and the two-
bond HMBC correlation from 3H (3-O13CH3) to C-3. MS (EI+, m/z
(%) 213 (M+•, 61), 182 (24), 181 (100), 165 (11), 153 (100), 151 (26),
rate of 10 μL/min. Source and the desolvation temperatures were held at
70 and 200 °C, respectively. Argon gas pressure in the collision cell was
set to attenuate the precursor ion transmission by 30−50%.
Synthesis of 2-[13C2H3]Methoxy-3-[13C]methoxy-4-methoxy-
benzoic Acid (21) and 2,3-[13C]Dimethoxy-4-methoxybenzoic
Scheme 6. Synthesis of 2-[13C2H3]Methoxy-3-[13C]methoxy-4-methoxybenzoic Acid (21) and 2,3-[13C]Dimethoxy-4-
methoxybenzoic Acid (22)
4387
dx.doi.org/10.1021/jo500234w | J. Org. Chem. 2014, 79, 4378−4389

The Journal of Organic Chemistry
Article
a
Scheme 7. Synthesis of 2-Methoxy-6-[C2H3]methoxybenzoic Acid
a
Reagents and conditions: (a) CH2N2 (1.5 equiv), THF, −78 °C, 30 min, 98%; (b) C2H3I (2 equiv), NaH (2 equiv), THF, rt, 24 h, 30%; (c) 5 M
NaOH (0.5 mL), rt, 30 min. 98%.
a
Scheme 8. Synthesis of 3,4-di[C2H3]Methoxybenzoic Acid (14)
a
Reagents and conditions: (a) CH2N2 (1.5 equiv), THF, rt, 30 min; (b) C2H3I (2 equiv), Cs2CO3 (2 equiv), THF, rt, 72 h; (c) 5 M NaOH
(0.5 mL), rt, 30 min.
137 (51), 123 (24), 122 (16), 121 (31), 120 (36), 109 (25), 95 (21), 94
(23), 66 (25), 53 (33).
methoxybenzoic acid methyl ester (26; 2 mg in 150 μL of methanol),
and the mixture was stirred for 30 min at rt. The reaction mixture was
acidified with 1 M HCl and extracted into ethyl acetate. Ethyl acetate was
removed under a stream of N2 to yield 1.9 mg (98%) of 22 as a white
solid. 1H NMR (500 MHz, CD3OD) δH 6.86 (1H, d, J = 8.5, H-5), 7.563
(1H, d, J = 8.5, H-6), 3.89 (3H, s, 4-OCH3), 3.90 (3H, d, J = 145 Hz,
2-O13CH3), 3.82 (3H, d, J = 145 Hz, 3-O13CH3); 13C NMR (125 MHz,
CD3OD) δc 168.9 (COO), 118.8 (C-1), 155.6 (C-2), 144.2 (C-3),
158.8 (C-4), 108.6 (C-5), 128.4 (C-6), 62.4 (2-O13CH3), 61.4
(3-O13CH3), 56.6 (4-OCH3); EI m/z (%) 214 (M+•, 100), 198 (30),
180 (42), 167 (40), 152 (20), 137 (66), 123 (11), 109 (37), 94 (27), 77
(29), 53 (45).
Compound 26. 1H NMR (500 MHz, CD3OD) δH 6.85 (1H, d, J =
9.0, H-5), 7.56 (1H, d, J = 9.0, H-6), 3.89 (3H, s, 4-OCH3), 3.84
(3H, s, COOCH3), 3.86 (3H, d, J = 144 Hz, 2-O13CH3), 3.81 (3H, d, J =
144 Hz, 3-O13CH3); 13C NMR (125 MHz, CD3OD)δc 167.7 (COO),
52.4 (COOCH3), 118.8 (C-1), 155.7 (C-2), 144.2 (C-3), 158.8 (C-4),
108.5 (C-5), 128.0 (C-6), 62.3 (2-O13CH3), 61.3 (3-O13CH3), 56.6 (4-
OCH3); EI m/z (%) 228 (M+•, 91), 212 (10), 197 (100), 196 (72), 181
(13), 180 (39), 166(15), 153 (33), 137 (20), 109 (12), 94 (17), 66 (21).
Methyl 2-[13C2H3]Methoxy-3-[13C]methoxy-4-methoxybenzoate
(27). To a stirred solution of methyl 2-hydroxy-3-[13C] methoxy-4-
methoxybenzoate (25; 18 mg, 0.1 mmol) and 2 equiv of Cs2CO3 in
anhydrous THF (2 mL) at rt under N2, was added 2 equiv of 13C2H3I
and the mixture was stirred at rt under N2 for 72 h. The reaction was
quenched with 1 M HCl (5 mL), and the product was extracted into
ethyl acetate. The organic layer was separated and concentrated. The
reaction product was purified by flash-silica chromatography (10%
EtOAc/hexane with 1% AcOH) to give a colorless oil (17, 12 mg, 61%).
The structure of compound 27 was confirmed by the strong three-
bond HMBC correlations from H-5 and H-6 to C-3 and C-2, respec-
tively, and the two-bond HMBC correlation from 3H (3-O13CH3) to
C-3. 1H NMR (500 MHz, CD3OD) δH 6.83 (1H, d, J = 8.5, H-5), 7.55
(1H, d, J = 8.5, H-6), 3.88 (3H, s, 4-OCH3), 3.83 (3H, s, COOCH3),
3.80 (3H, d, J = 145 Hz, 3-O13CH3); 13C NMR (125 MHz, CD3OD) δc
167.6 (COO), 52.3 (COOCH3), 118.8 (C-1), 155.7 (C-2), 144.5 (C-3),
158.8 (C-4), 108.5 (C-5), 128.0 (C-6), 61.5 (septet, J = 87.5,
2-O13C2H3), 61.4 (3-O13CH3), 56.6 (4-OCH3); EI m/z (%) 231
(M+•, 100), 215(12), 200 (100), 182 (20), 153(28), 137 (20), 109 (12),
53 (10).
Methyl 2-Methoxy-6-hydroxybenzoate (29). A freshly prepared
solution of diazomethane (3 mL, ∼250 mM) in ether was added drop-
wise to a stirred solution of 2-methoxy-6-hydroxybenzoic acid (28;
84 mg, 0.5 mmol) in THF (2 mL) at −78 °C. The mixture was stirred for
30 min at −78 °C and quenched with acetic acid (50 μL). The solvent
1
was evaporated to yield a white solid (29, 88 mg, 97%). H NMR
(500 MHz, CD3OD) δH 6.48 (2H, d J = 8.5, H-3,5), 7.24 (1H, t, J = 8.5,
H-4), 3.78 (3H, s, 2-OCH3), 3.85 (3H, s, COOCH3), and 13C NMR
(125 MHz, CD3OD) δc 171.3 (COO), 52.6 (COOCH3), 108.1 (C-1),
160.9 (C-2), 103.4 (C-3), 134.4 (C-4), 103.4 (C-5), 110.1 (C-6), 56.4
(2-OCH3); EI m/z (%) 182 (M+•, 34), 150 (100), 136 (8), 122 (45),
107 (85), 93 (7), 79 (10).
Methyl 2-Methoxy-6-[C2H3]methoxybenzoate (30). To a stirred so-
lution of methyl 2-methoxy-6-hydroxybenzoate (29; 40 mg, 0.22 mmol)
and 2 equiv of sodium hydride in 2 mL of anhydrous THF was added
2 equiv of C2H3I, and the mixture was stirred at rt under N2 for 24 h. The
reaction was quenched with 1 M HCl (5 mL), and the product was
extracted into ethyl acetate and concentrated under vacuum. The
reaction product was purified by flash silica chromatography (10%
EtOAc/hexane with 1% AcOH) to yield a colorless oily material (30)
(11 mg, 30%). 1H NMR (500 MHz, CD3OD) δH 6.65 (2H, d,d J = 8.5,
1.5, H-3,5), 7.32 (1H, t, J = 8.5, H-4), 3.78 (3H, s, 2-OCH3), 3.81
(3H, s, COOCH3); 13C NMR (125 MHz, CD3OD) δc 169.1 (COO),
52.7 (COOCH3), 114.2 (C-1), 158.7 (C-2), 105.0 (C-3), 132.5 (C-4),
105.0 (C-5), 158.7 (C-6), 56.5 (2-OCH3); EI m/z (%) 199 (M+•, 35),
168 (100), 153 (12), 107 (22).
2-[13C2H3]Methoxy-3-[13C]methoxy-4-methoxybenzoic Acid (21).
A solution of 5 M NaOH (0.5 mL) was added to a solution of methyl
2-[13C2H3]methoxy-3-[13C]methoxy-4-methoxybenzoate (27; 0.2 mg
in 100 μl of methanol), and the mixture was stirred for 30 min at rt.
The product was acidified with 1 M HCl and extracted into ethyl acetate.
Ethyl acetate was removed under a stream of N2 to yield 21 as a white
solid (0.18 mg, 98%). 1H NMR (500 MHz, CD3OD) δH 6.86 (1H, d, J =
9, H-5), 7.63 (1H, d, J = 9, H-6), 3.89 (3H, s, 4-OCH3), 3.82 (3H, d,
J = 145 Hz, 3-O13CH3), and 13C NMR (125 MHz, CD3OD) δC 168.9
(COO), 118.7 (C-1), 155.6 (C-2), 144.2 (C-3), 158.8 (C-4), 108.5
2-Methoxy-6-[C2H3]methoxybenzoic Acid (31). A solution of 5 M
NaOH (0.5 mL) was added to a solution of methyl 2-methoxy-6-
[C2H3]methoxy benzoate (30, 2.1 mg) in 300 μL of methanol and was
stirred for 30 min at rt. The product was acidified with 1 M HCl and
extracted into ethyl acetate. It was concentrated to yield 31 (Scheme 7)
as a white solid (1.9 mg, 98%). 1H NMR (500 MHz, CD3OD) δH 6.65
(2H, d J = 8.5, H-3,5), 7.31 (1H, t, J = 8.5, H-4), 3.81 (3H, s, 2-OCH3);
13C NMR (125 MHz, CD3OD) δc (125 MHz) 170.1 (COO), 116.5
2
(C-5), 128.3 (C-6), 61.8 (septet, J = 87.5, 2-O13C1 H3), 61.3 (3-
O13CH3), 56.6 (4-OCH3); EI m/z (%) 217 (M+•, 100), 210 (25), 182
(21), 168 (25), 137 (60), 123 (10), 109 (20), 95 (11), 78 (18), 66 (25),
53 (35).
2,3-[13C]Dimethoxy-4-methoxybenzoic Acid (22). A solution of 5 M
NaOH (0.5 mL) was added to a solution of 2,3-[13C]dimethoxy-4-
4388
dx.doi.org/10.1021/jo500234w | J. Org. Chem. 2014, 79, 4378−4389

The Journal of Organic Chemistry
Article
(C-1), 158.4 (C-2), 105.1 (C-3), 132.2 (C-4), 105.1 (C-5), 158.4 (C-6),
56.4 (2-OCH3); EI m/z (%) 185 (M+•, 100), 168 (49), 153 (11),
139 (34), 138 (40), 136 (28), 122 (18), 121 (30), 120 (21), 107 (100),
89 (30), 78 (30), 63 (30).
(4) Attygalle, A. B.; García-Rubio, S.; Ta, J.; Meinwald, J. J. Chem.
Society, Perkin Trans. 2 2001, 498−506.
(5) Attygalle, A. B.; Bialecki, J. B.; Nishshanka, U.; Weisbecker, C. S.;
Ruzicka, J. J. Mass Spectrom. 2008, 43, 1224−1234.
(6) Chowdhury, D.; Maurya, S.; Pandey, M. B.; Sarma, B. K.; Singh, U.
P. Mycobiology 2005, 33, 206−209.
(7) Majmudar, C. Y.; Højfeldt, J. W.; Arevang, C. J.; Pomerantz, W. C.;
Gagnon, J. K.; Schultz, P. J.; Cesa, L. C.; Doss, C. H.; Rowe, S. P.;
Vsquez, V.; Tamayo-Castillo, G.; Cierpicki, T.; Charles, L.; Brooks, C.
L.; Sherman, D. H.; Mapp, A. K. Angew. Chem., Int. Ed. 2012, 51, 11258−
11262.
3,4-Di[C2H3]methoxybenzoic Acid (14) . A solution of freshly pre-
pared diazomethane (1.5 equiv) in ether was added dropwise to a stirred
solution of 3,4-dihydroxybenzoic acid (32) in 200 μL of THF at −78 °C.
The mixture was stirred for 30 min at rt, then solvent was evaporated to
yield a white solid residue (33). To a solution of methyl 3,4-dihydroxy-
benzoate (33) in anhydrous THF (0.5 mL) was added 2 equiv of C2H3I
and Cs2CO3, and the mixture was stirred for 72 h. The reaction was
quenched with 1 M HCl (0.2 mL), and the product was extracted into
ethyl acetate and concentrated under vacuum to yield methyl 3,4-
di[C2H3]methoxybenzoate (34) as a colorless oil, which was dissolved
in methanol (300 μL) and mixed with aqueous NaOH (5 M, 0.5 mL).
After stirring for 30 min at rt, the reaction was quenched with 1 M HCl,
and the product was extracted into ethyl acetate. The separated organic
extract was evaporated to yield 3,4-di[C2H3]methoxybenzoic acid (14)
as a white solid (Scheme 8).
(8) Kitahara, N.; Endo, A.; Furuya, K.; Takahashi, S. J. Antibiot. (Tokyo)
1981, 34, 1562−1568.
(9) Minkin, V. I.; Minyaev, R. M.; Dorogan, I. V. J. Mol. Struct.:
THEOCHEM 1997, 398/398, 237−253.
(10) Bulo, R. E.; Jansen, H.; Ehlers, A. W.; Kanter, F. J. J.; Schakel, M.;
Lutz, M.; Spek, A. L.; Lammertsma, K. Angew. Chem., Int. Ed. 2004, 43,
714−714.
(11) Dorogan, I. V.; Minkin, V. I.; Novikova, L.; Mendeleev, M. Chem.
Commun. 2003, 13, 205−207.
Computational Methods. All QM calculations were performed
using the Gaussian 03W program package. The predicted structures of
reactants, dissociation products, and transition states were optimized
using the B3LYP exchange-correlation function and the 6-311++G(d,p)
basis set. Calculated stationary states were confirmed to be either ground
states or transition states by vibrational frequency analysis. Calculated
electronic energies of stationary states were corrected for the zero-point
vibrational energy (ZPE), and the ZPE corrections were scaled by a
factor of 0.9877.35 Gibbs free energy values were calculated assuming a
temperature of 298 K and a pressure of 1 Pa in the collision cell. Uni-
molecular rate constants (k) were calculated from the Gibbs free energy
differences between transition states and the reactants (ΔG⧧) using the
following equation.36
(12) Lavorato, D. J.; Terlouw, J. K.; McGibbon, G. A.; Dargel, T. K.;
Koch, W.; Schwarz, H. Int. J. Mass Spectrom. 1998, 179/180, 7−14.
́
(13) Bienkowski, T.; Danikiewicz, W. Rapid Commun. Mass Spectrom.
2003, 17, 697−705.
(14) Ben-Ari, J.; Etinger, A.; Weisz, A.; Mandelbaum, A. J. Mass
Spectrom. 2005, 40, 1064−1071.
(15) Dickstein, J. S.; Curto, J. M.; Gutierrez, O.; Mulrooney, C. A.;
Kozlowski, M. C. J. Org. Chem. 2013, 78, 4744−4761.
(16) Eichinger, P. C. H.; Bowie, J. H.; Hayes, R. N. Aust. J. Chem. 1989,
42, 865−874.
(17) Kuck, D. Int. J. Mass Spectrom. 2002, 213, 101−144.
(18) Wenthold, G.; Paulino, J. A.; Squires, R. R. J. Am. Chem. Soc. 1991,
113, 7414−1715.
−ΔG⧧
RT
kBT
h
(19) Wenthold, P. G.; Squires, R. R. J. Am. Chem. Soc. 1994, 116,
6401−6412.
k =
exp
(20) Wenthold, P. G.; Hu, J.; Squires, R. R. J. Am. Chem. Soc. 1996, 118,
11865−11871.
where kB is the Boltzmann constant; h is Planck’s constant; R is the gas
constant; T is the temperature, and ΔG⧧ is the Gibbs free energy
difference between a transition state and a reactant
(21) Wenthold, P. G.; Hu, J.; Squires, R. R. J. Mass Spectrom. 1998, 33,
796−802.
(22) McAnoy, A. M.; Dua, S.; Blanksby, S. J.; Bowie, J. H. J. Chem. Soc.
Perk. Trans 2 2000, 1665−1673.
ASSOCIATED CONTENT
■
S
* Supporting Information
(23) Koch, H. F. Acc. Chem. Res. 1984, 17, 137−144.
(24) Crews, P.; Rodrigues, J.; Jaspars, M. Organic Structure Analysis;
Oxford University Press: New York, 1998.
Atom coordinates and absolute energies of structures in Figures 5
1
and 6, and copies of H and 13C NMR spectra for new com-
pounds. This material is available free of charge via the Internet at
http://pubs.acs.org.
(25) Karni, M.; Mandelbaum, A. Org. Mass Spectrom. 1980, 15, 53−64.
(26) Krauss, D.; Maninx, H. G.; Tauscher, B.; Bischof, P. Org. Mass
Spectrom. 1985, 20, 614−618.
(27) Attygalle, A. B.; Ruzicka, J.; Varughese, D.; Bialecki, J. B.; Jafri, S. J.
Mass Spectrom. 2007, 42, 1207−1217.
AUTHOR INFORMATION
■
Corresponding Author
(28) Hu, N.; Tu, Y.-P.; Jiang, K.; Pan, Y. J. Org. Chem. 2010, 75, 4244−
4250.
*E-mail: athula.attygalle@stevens.edu
(29) Chai, Y.; Hu, N.; Pan, Y. J. Am. Soc. Mass Spectrom. 2013, 24,
1097−1101.
Notes
The authors declare no competing financial interest.
(30) Zeller, A.; Strassner, Th. Organometallics 2002, 21, 4950−4954.
(31) Khairallaha, G. N.; O’Hair, R. A. J. Int. J. Mass Spectrom. 2006, 254,
145−151.
ACKNOWLEDGMENTS
■
This research was supported by funds provided by Stevens
Institute of Technology (Hoboken, NJ) and Merck Research
Laboratories (Rahway, NJ). We thank Bristol-Myers Squibb
(New Brunswick, NJ) for the donation of the Quattro Ultima
mass spectrometer.
(32) Schroder, D.; Semialjac, M.; Schwarz, H. Int. J. Mass Spectrom.
̈
2004, 233, 103−109.
(33) Kleingeld, J. C.; Nibbering, N. M. M. Tetrahedron 1983, 39,
4193−4196.
(34) Chang, C.-C.; Axe, F.; Bolgar, M.; Attygalle, A. B. J. Mass Spectrom.
2010, 45, 1130−1138.
(35) Mercero, J. M.; Matxain, J. M.; Lopez, X.; York, D. M.; Largo, A.;
Eriksson, L. A.; Ugalde, J. M. Int. J. Mass Spectrom. 2005, 240, 37−99.
(36) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971, 54,
724−728.
REFERENCES
■
(1) Cole, R. B. Electrospray and MALDI Mass Spectrometry:
Fundamentals, Instrumentation, Practicalities, and Biological Applications;
John Wiley & Sons: Hoboken, NJ, 2010.
(2) Hill, A. W.; Mortishire-Smith, R. J. Rapid Commun. Mass Spectrom.
2005, 19, 3111−3118.
(3) HighChem, Ltd. http://www.highchem.com (accessed 2013).
4389
dx.doi.org/10.1021/jo500234w | J. Org. Chem. 2014, 79, 4378−4389