Low-energy collision-induced fragmentation of negative ions derived from ortho-, meta-, and para-hydroxyphenyl carbaldehydes, ketones, and related compounds

Article abstract of DOI:10.1002/jms.1252

Collision-induced dissociation (CID) mass spectra of anions derived from several hydroxyphenyl carbaldehydes and ketones were recorded and mechanistically rationalized. For example, the spectrum of m/z 121 ion of deprotonated ortho-hydroxybenzaldehyde shows an intense peak at m/z 93 for a loss of carbon monoxide attributable to an ortho-effect mediated by a charge-directed heterolytic fragmentation mechanism. In contrast, the m/z 121 ion derived from meta and para isomers undergoes a charge-remote homolytic cleavage to eliminate an ?H and form a distonic anion radical, which eventually loses CO to produce a peak at m/z 92. In fact, for the para isomer, this two-step homolytic mechanism is the most dominant fragmentation pathway. The spectrum of the meta isomer on the other hand, shows two predominant peaks at m/z 92 and 93 representing both homolytic and heterolytic fragmentations, respectively. ¹⁸O-isotope-labeling studies confirmed that the oxygen in the CO molecule that is eliminated from the anion of meta-hydroxybenzaldehyde originates from either the aldehydic or the phenolic group. In contrast, anions of ortho-hydroxybenzaldehyde and 2-hydroxy-1-naphthaldehyde, both of which show two consecutive CO eliminations, specifically lose the carbonyl oxygen first, followed by that of the phenolic group. Anions from 2-hydroxyphenyl alkyl ketones lose a ketene by a hydrogen transfer predominantly from the α position. Interestingly, a very significant charge-remote 1,4-elimination of a H₂ molecule was observed from the anion derived from 2,4-dihydroxybenzaldehyde. For this mechanism to operate, a labile hydrogen atom should be available on the hydroxyl group adjacent to the carbaldehyde functionality. Copyright

Full text of DOI:10.1002/jms.1252

JOURNAL OF MASS SPECTROMETRY
J. Mass Spectrom. 2007; 42: 1207–1217
Published online 10 August 2007 in Wiley InterScience
(
www.interscience.wiley.com) DOI: 10.1002/jms.1252
Low-energy collision-induced fragmentation of
negative ions derived from ortho-, meta-, and
para-hydroxyphenyl carbaldehydes, ketones, and
related compounds
∗
Athula B. Attygalle, Josef Ruzicka, Deepu Varughese, Jason B. Bialecki and Sayed Jafri
Center for Mass Spectrometry, Department of Chemistry and Chemical Biology, Stevens Institute of Technology, Hoboken, New Jersey 07030, USA
Received 26 February 2007; Accepted 9 May 2007
Collision-induced dissociation (CID) mass spectra of anions derived from several hydroxyphenyl
carbaldehydes and ketones were recorded and mechanistically rationalized. For example, the spectrum of
m/z 121 ion of deprotonated ortho-hydroxybenzaldehyde shows an intense peak at m/z 93 for a loss of
carbon monoxide attributable to an ortho-effect mediated by a charge-directed heterolytic fragmentation
mechanism. In contrast, the m/z 121 ion derived from meta and para isomers undergoes a charge-remote
homolytic cleavage to eliminate an žH and form a distonic anion radical, which eventually loses CO
to produce a peak at m/z 92. In fact, for the para isomer, this two-step homolytic mechanism is the
most dominant fragmentation pathway. The spectrum of the meta isomer on the other hand, shows
two predominant peaks at m/z 92 and 93 representing both homolytic and heterolytic fragmentations,
1
8
respectively. O-isotope-labeling studies confirmed that the oxygen in the CO molecule that is eliminated
from the anion of meta-hydroxybenzaldehyde originates from either the aldehydic or the phenolic
group. In contrast, anions of ortho-hydroxybenzaldehyde and 2-hydroxy-1-naphthaldehyde, both of which
show two consecutive CO eliminations, specifically lose the carbonyl oxygen first, followed by that of
the phenolic group. Anions from 2-hydroxyphenyl alkyl ketones lose a ketene by a hydrogen transfer
predominantly from the a position. Interestingly, a very significant charge-remote 1,4-elimination of a H
2
molecule was observed from the anion derived from 2,4-dihydroxybenzaldehyde. For this mechanism to
operate, a labile hydrogen atom should be available on the hydroxyl group adjacent to the carbaldehyde
functionality. Copyright  2007 John Wiley & Sons, Ltd.
KEYWORDS: CID; negative ions; hydroxybenzaldehyde; hydroxynaphthaldehyde; hydroxyacetophenone; even-electron
ions; fragmentation
INTRODUCTION
molecules including many natural products, pharmaceuti-
cal and combinatorial compounds and their metabolites,
and environmental pollutants is important, the current AI
Many computer programs have been developed to assist
mass spectral interpretations of electron ionization (EI)
2
programs for CID spectra are totally inadequate, particularly
when peaks representing both radical and neutral molecule
1
spectra. Attempts have been made to extend EI fragmen-
tation rules to interpret collision-induced dissociation (CID)
mass spectra of even-electron (EE) ions. Though sometimes
helpful, this approach is basically flawed since the fragmen-
tation principles of the two processes are different. Several
programs are available to assign structures to ions derived
5
losses are observed simultaneously.
In our pursuit of deeper understanding of collision-
6
induced fragmentation of low-molecular-weight anions, we
noted that even the predication of spectra of simple com-
pounds such as hydroxyphenyl carbaldehydes and ketones
2
–4
by collision-induced decomposition.
In order to develop
7
,8
2
is not straightforward. Understanding the fragmentation
pathways of this group of compounds is important since
acylphenols are often encountered as secondary metabo-
lites. For example, polyketide-derived secondary metabolites
such as phlorocaprophenone (1), and related compounds are
biosynthetic precursors for differentiating-inducing factors
that modulate unicellular amoebas of Dityostelium discoideum
to aggregate and form a ‘multicellur’ slug that can migrate
successful programs for interpretation of EE spectra, the
underlying fragmentation pathways must not only be ratio-
nalized, but also be scrutinized thoroughly by detailed
experimentation. Although interpretation of spectra of small
Ł
Correspondence to: Athula B. Attygalle, Center for Mass
Spectrometry, Department of Chemistry and Chemical Biology,
Stevens Institute of Technology, Hoboken, New Jersey 07030, USA.
E-mail: athula.attygalle@stevens.edu
9
toward light and heat.
Copyright  2007 John Wiley & Sons, Ltd.

1208
A. B. Attygalle et al.
OH
O
General procedure for synthesis of hydroxyphenyl
alkyl ketones
An alkyl bromide (0.9 mmol) was added dropwise to a
cooled suspension of magnesium turnings (50 mg, 2 mmol),
and a few crystals of iodine in diethyl ether (5 ml). The
mixture was stirred for 10 min, and an ethereal solution of
HO
OH
(
1)
2
-, 3-, or 4-hydroxybenzaldehyde (50 mg, 0.4 mmol) was
added slowly. After adding water (10 ml), the mixture
was extracted with diethyl ether (10 ml ð 3). The ether
In a previous study, we reported the formation of an
unprecedented carboxylate anion during the CID of negative
ions derived from ortho isomers of hydroxyphenyl carbonyl
extract was dried over MgSO and concentrated to yield the
4
corresponding (1-hydroxyalkyl)-phenols. The solid material
obtained in this way was characterized by GC-MS and ESI
mass spectrometry. Yields were 30–40%.
7
compounds. For example, the CID spectra of 2-hydroxy
derivatives of benzaldehyde, acetophenone, propiophenone,
isobutyrophenone, and pivalophenone show distinct peaks
at m/z 45, 59, 73, 87, and 101 for the formate, acetate,
Each (1-hydroxyalkyl)-phenol (28 mg, 0.2 mmol)
obtained from the above step was dissolved in CH
(
(
2
Cl
2 ml) and stirred with pyridine-chromium trioxide complex
0.4 mmol) at room temperature. The mixture was filtered
2
7
propionate, isobutyrate, and pivalate anions, respectively.
In fact, there are many more additional interesting features in
the fragmentation mechanisms of hydroxyphenyl carbalde-
hydes and ketones that enable unequivocal characterization
of their positional isomers. Here, we present our results.
through a bed of silica gel. The filtrate was washed with dil.
HCl (1 M) and water, and extracted with dichloromethane.
The organic layer was separated, dried over MgSO4 and
evaporated. All products were isolated by preparative TLC
(
1 : 1 ethyl acetate-hexane), and confirmed by GC- and ESI-
MS. Yields, 30–40%.
EXPERIMENTAL
Chemicals
Synthesis of hydroxyacetonaphthones
2
4
2
-Hydroxybenzaldehyde,
-hydroxybenzaldehyde, 1-hydroxy-2-naphthaldehyde,
-hydroxy-1-naphthaldehyde, pyridine, ethyl bromide,
3-hydroxybenzaldehyde,
Hydroxyacetonaphthones were synthesized from corre-
sponding hydroxynaphthaldehydes by a procedure similar
to that described above for hydroxyacetophenones.
0
0
isopropyl bromide, t-butyl bromide, 6 -methoxy-2 -
1
8
acetonaphthone, D
O; Isotec) were obtained from Sigma-Aldrich Chemical
2
O (99.9 atom% D), and H
2
O (95 atom%
ꢀ
ꢀ
1
8
Synthesis of 6 -hydroxy-2 -acetonaphthone
0
0
6
-Methoxy-2 -acetonaphthone (100 mg, 0.5 mmol) was dis-
Company (St Louis, MO).
solved in 33% HBr in acetic acid (50 mmol, 10 ml). The
solution was irradiated for 10 min in a microwave oven
Mass spectrometry
°
(
200 W, 110 C). The reaction mixture was cooled and neutral-
ized with aqueous Na CO . The mixture was extracted with
CH Cl
GC-MS analysis on an HP 5989B quadrupole mass spectrom-
eter linked to a Hewlett Packard 5890 Series II gas chromato-
graph showed the purity of each sample to be >99.9%. The
CID mass spectra were recorded on a Micromass Quattro I,
or a Thermo TSQ Quantum Ultra spectrometer. The Quattro
I mass spectrometer was equipped with an electrospray ion
source. Samples were infused as acetonitrile-water-NaOH
2
3
2
2
ꢀ25 ml ð 3ꢁ. The combined extract was dried and con-
0
0
centrated to yield 6 -hydroxy-2 -acetonaphthone. Yield, 50%.
2
Preparation of hydroxy-[2,2,2- H ]acetophenones
and hydroxy-[2,2,2- H ]acetonaphthones
Eachhydroxy-acetophenone(25 mg, 0.18 mmol), orhydroxy-
3
2
3
ꢀ
5
ꢀ1
(
9 : 1 : 10 ) solutions at a flow rate of 6 µl min . The source
acetonaphthone, was dissolved in NaOD in D
prepared by dissolving metallic sodium (5 mg) in D
5 ml)], and stirred overnight at room temperature. The reac-
2
O [freshly
temperature was held at 80 °C. The argon gas pressure in the
2
O
collision cell was set to attenuate precursor ion transmission
by 30–50%. The TSQ Quantum Ultra mass spectrometer was
equipped also with an electrospray ion source and samples
were infused in 90% ACN at a flow rate of 10 µl min . The
capillary temperature was held at 220 °C and the collision
(
tion mixture was extracted with dichloromethane (5 ml ð 3).
ꢀ
1
The products were determined by GC-MS. Yield, 85%.
1
8
gas (argon) pressure at 1.5 m Torr. All spectra were recorded
at unit resolution (Q1 D Q2 D 0.7 Da FWHM).
Preparation of 2-hydroxy[ O]benzaldehyde, and
related compounds
To a solution of 2-hydroxybenzaldehyde (20 mg, 0.16 mmol)
High resolution (R ³ 15 000) and accurate mass spectra
were measured on a Waters-Micromass Q-ToF API-US
spectrometer equipped with a nanoelectrospray ion source.
Signals were acquired in the W-mode operation. Ion series
generated by water clusters charged by bicarbonate anion
1
8
in THF (3 ml), water (95% H
concentrated H SO ꢀ5 µlꢁ were added, and the mixture
was stirred at 50 °C for 18 h. THF was removed and the
2
O) (32 µl, 6.2 mmol) and
2
4
1
8
residue was dried to give 2-hydroxy[ O]benzaldehyde
ꢀ
was used as reference mass peaks. The source temperature
[ESI-MS: m/z
D
123 ꢀM ꢀ Hꢁ ]. Similarly, 3- and
1
8
was held at 80 °C. Pressure of argon gas in the collision cell
was held at 5.1 ð 10ꢀ5 Pa.
4-hydroxy[ O]benzaldehyde were prepared from the cor-
responding hydroxybenzaldehydes.
Copyright  2007 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2007; 42: 1207–1217
DOI: 10.1002/jms

CID of hydroxyphenyl carbaldehydes and ketones 1209
ꢀ
18
2
Preparation of 3 -hydroxy[ O]acetophenone
Synthesis of 1-bromo[1,1- H ]hexane and
2
0
ꢀ
2
3
-Hydroxyacetophenone (5 mg, 0.02 mmol) in THF (1 ml)
was stirred with H
2 -hydroxy-[2,2- H2]heptanophenone
Phosphorous tribromide (216 mg, 0.8 mmol) was slowly
1
8
2
O ꢀ10 µlꢁ in the presence of con.
2
H
2
SO
4
ꢀ1 µlꢁ for 35 h at 50 °C. The mixture was extracted
added to 1-[1,1- H
2
]hexanol (205 mg, 2.00 mmol) and stirred
with diethyl ether and product was characterized by GC-MS
for 2 h. The mixture was dissolved in diethyl ether (2 ml) and
washed with water (2 ml) and 1 M NaOH (2 ml). The ethereal
ꢀ
(
10% exchange), [ESI-MS: m/z D 137 ꢀM ꢀ Hꢁ ].
layer was dried over MgSO , filtered through a bed of silica
4
gel, then through anhydrous aluminum oxide and evapo-
rated to yield the product as a clear liquid (135 mg, 41%).
1
8
Preparation of 2-hydroxy-1-[ O]naphthaldehyde
2
2
2
-Hydroxy-1-naphthaldehyde (20 mg, 0.16 mmol) in THF
1-Bromo[1,1- H ]hexane obtained in this way was con-
1
8
(
1 ml) was heated with H
2
O (95%, 32 mg) and concentrated
(3 mg) at 50 °C for 18 h., and the mixture was evapo-
rated. After two repetitions of this process, the residue was
dissolved in Et O and washed with water. Electrospray ion-
verted to Grignard reagent and employed in the general
ketone synthesis procedure described above to give first
H SO
2 4
0
0
0
2
2
2-(1 -hydroxy[2 ,2 - H ]heptyl)phenol, which was oxidized
0
2
2
to yield 2 -hydroxy-[2,2- H ]heptanophenone ( 2).
2
ization mass spectrometric (ESI-MS) analysis of the product
showed that approximately 90% of the oxygen atoms at the
C-1 position had been replaced selectively with 18O atoms.
O
CD2
OH
2
Synthesis of 1-[1,1- H ]hexanol
2
Hexanoic acid (280 mg, 2.4 mmol) in anhydrous diethyl ether
(
2)
(
(
2 ml) was added to a mixture of lithium aluminum deuteride
121 mg, 2.9 mmol) in anhydrous diethyl ether (2 ml). The
RESULTS AND DISCUSSION
reaction mixture was stirred for 2 h at 35 °C, cooled to 0 °C
and quenched with water (0.2 ml). The mixture was shaken
with 2 M HCl (10 ml) to dissolve the inorganic solids and
the aqueous layer was discarded. The ethereal evaporated to
yield the product as a clear liquid (205 mg, 83%).
CID spectra of EE negative ions derived from hydroxyben-
zaldehydes appear simple. However, there are distinct differ-
ences in the fragmentation pathways that the deprotonated
ortho, meta, and para isomers undergo upon interaction with
100
A
100
D
93
93
CHO
CH18
O
O
O
O
O
65
65
1
21
123
0
0
40
60
80
100
3
120
40
60
80
100
120
CHO
21
CH18
123
92
O
100
B
100
E
CH18
O
92
1
9
95
O
O
O
93
65
65
120
122
O
0
0
40
60
80
100
120
40
60
80
100
120
92
CH18
O
100
C
CHO
100
F
92
123
121
O
O
O
O
1
20
122
65
0
0
40
60
80
100
120
40
60
80
100
120
m/z
m/z
ꢀ
Figure 1. Unit–resolution product-ion spectrum of m/z 121 [ꢀM ꢀ Hꢁ ] from ortho- (A), meta- (B) and para-hydroxybenzaldehyde (C),
1
8
and that of m/z 123 from ortho- (D), meta- (E), and para-hydroxy[ O]benzaldehyde (F) at a collision energy settings of 25 eV
recorded on a TSQ Quantum Ultra mass spectrometer (all parameters including collision gas pressure and skimmer voltage were
identical).
Copyright  2007 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2007; 42: 1207–1217
DOI: 10.1002/jms

1210
A. B. Attygalle et al.
O
C
O
C
H
b
-
CO
OH
O
OH
m/z 93 (5)
m/z 121
m/z 121
a
-
CO
O
O
-CO
m/z 65 (7)
H
m/z 93 (6)
H
O
O
O
(
3)
(4)
Scheme 1. Proposed pathways for fragmentation of m/z 121 ion derived from o-hydroxybenzaldehyde.
the collision gas. Consequently, the spectra of hydroxyben-
m/z 121
100
A
zaldehyde positional isomers show characteristic differences
Fig. 1). To start with, the ortho isomer requires very little
CHO
(
collisional activation to undergo fragmentation. Its spectrum
shows an intense peak at m/z 93, attributable to a loss of CO,
which is visible even at a laboratory-frame collision energy
value as low as 10 eV (Fig. 2(A)). The spectrum recorded
O
5
0
m/z 93
0
1
8
from m/z 123 of 2-hydroxy[ O]benzaldehyde confirmed
that the CO is lost exclusively from the carbonyl group of
the ortho isomer (Fig. 1(D)). Since under similar low-energy
collision conditions, the m/z 121 ion from meta- and para-
hydroxybenzaldehydes does not undergo any observable
decomposition (Fig. 2(B) and (C)), the facile loss of carbon
monoxide from the anion of ortho-hydroxybenzaldehyde can
be attributed to a charge-mediated ortho effect as depicted in
Scheme 1. The oxytane heterocyclic intermediate (3), which
we proposed previously to rationalize the specific formation
0
10
20
30
40
Collision Energy (eV)
CHO
100
B
m/z 121
O
50
m/z 92
m/z 93
0
7
of a carboxylate anion from the ortho isomers, could rear-
0
10
20
30
40
range to a formate intermediate (4) and lose CO as shown
in ‘Pathway a in Scheme 1. Alternatively, a migration of
Collision Energy (eV)
ꢀ
CHO
the aldehydic hydrogen to the charge site followed by a
1
00
50
0
C
CO loss can be proposed (‘Pathway b,’ Scheme 1). Even if
ꢀ
m/z 121
‘
Pathway b operates, the initial m/z 93 ion (5) that is pro-
m/z 92
posed is expected to rapidly undergo a transition to the more
1
0
stable phenoxide anion (6) as reported by Kumar et al.
O
A product-ion spectrum recorded from m/z 93 ion showed
that it subsequently loses CO to produce the ion observed at
m/z 65 for the cyclopentadienyl anion (7) (Fig. 1(A)).
Support for the presumption that the m/z 93 ion is
the more stable phenoxide anion (6), originating either
from ‘Pathway a,’ or by a transformation of the less stable
initial m/z 93 ion 5, comes from tandem mass spectrometric
m/z 93
0
10
20
Collision Energy (eV)
30
40
Figure 2. Plot of relative abundance of ions versus
laboratory-frame collision energy for m/z 121 (- -), 93 (-ꢀ-),
and 92 (-ꢁ-) ions derived from ortho- (A), meta- (B), and
ž
experiments conducted with CO
the m/z 93 anion formed by in-source fragmentation of
-hydroxybenzaldehyde. Undoubtedly, the anion 5 given
2
as the collision gas and
para-hydroxybenzaldehyde (C).
2
acid. They reported that the m/z 93 ion derived from meta-
and para-hydroxybenzoic acid readily adds CO , in contrast
2
to that from salicylic acid, which does not react because it
undergoes a rapid transformation to the phenoxide ion.
Quantum mechanical calculations show that the proton
transfer to form the phenoxide anion occurs in a virtually
in Scheme 1 is a strong nucleophile. In contrast to weakly
nucleophilic anion 6, the anion 5 is expected to react with
1
0
CO
our hands a peak at m/z 137 was not observed from the
m/z 93 ion derived from 2-hydroxybenzaldehyde when CO
2
to form a salicylate anion (m/z 137). However, in
2
1
0
was used as the collision gas. Kumar et al. has observed
a similar situation with m/z 93 ion derived from salicylic
1
0
barrierless fashion.
Copyright  2007 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2007; 42: 1207–1217
DOI: 10.1002/jms

CID of hydroxyphenyl carbaldehydes and ketones 1211
Compared to the amount of collision energy required
for the decomposition of the ortho-hydroxybenzaldehyde,
both meta and para isomers require more energy to initiate
fragmentation (Fig. 2). For example, about 20–25 eV of
laboratory-frame collision energy is necessary to observe a
the formation of an m/z 92 ion represents a formal loss
of elements of the žCHO radical, which violates the
so-called EE rule advocated by some mass spectrometry
1
1
textbooks. Then again, many violations of this rule
1
2,13
have been documented.
A tandem mass spectrometric
5
0% reduction of the m/z 121 signal from the meta- and para-
experiment for a ‘neutral loss’ confirmed that the anion
derived from p-hydroxybenzaldehyde is able to lose 29 Da.
Since a peak was not observed at m/z 94 in the product-ion
hydroxybenzaldehyde, whereas the ortho isomer requires
only 10 eV (Fig. 2). Moreover, the spectra of the three isomers
are markedly different from each other; the most intense peak
for the ortho isomer is at m/z 93, whereas that for the para
isomer is at m/z 92 (Fig. 1(C)). The spectrum of the meta
isomer, on the other hand, shows two peaks of more or less
equal intensity for both m/z 92 and 93 ions (Fig. 1(B)).
The m/z 92 ion represents a loss of a 29 Da fragment
from the m/z 121 ion (Fig. 1(B) and (C)). In other words,
1
8
spectra recorded from 3- and 4-hydroxy[ O]benzaldehyde,
it was evident that the žCHO radical is lost from the
carbaldehyde group. A close scrutiny of the fragmentation
of meta and para isomers under high-resolution tandem mass
spectrometric conditions on a Q-ToF instrument revealed
that the žCHO loss is a two-step process. High-resolution
product ions spectra of both meta and para isomers clearly
show a peak at m/z 120 indicating a loss of a hydrogen atom
(
Fig. 3(B) and (C)). A peak at m/z 120 was not observed in
1
1
1
00
A
B
C
9
3
the spectra recorded from the ortho isomer at any collision
energy setting (Fig. 3(A)) (note: the low-mass ions at m/z
4
mass spectrometric experiments conducted using a Quattro
instrument are not observed here, since the high-resolution
Q-Tof instrument has a low-mass cutoff, which precludes
low-mass ion detection). The predominant fragmentation
pathways for the meta and para isomers appear to initiate
first by a charge-remote homolytic fission of the aldehydic
C–H bond to produce a gaseous species bearing a spatially
separated charge and an odd spin site (Scheme 2). Such
entities are called as distonic radical ions. Although charge-
remote loss of neutral molecules from activated gaseous
ions is well demonstrated,
in an analogous manner has not been widely reported. On
the other hand, the chemistry of distonic radical ions is
now well established.
tandem mass spectrometric experiment performed on the
m/z 120 ion (produced by in-source fragmentation) from
meta and para-hydroxybenzaldehydes showed a loss of CO
to form the ion observed at m/z 92. Thus, the loss of elements
of žCHO radical appears to be a two-step process, while
some contribution from single-step mechanisms may also be
expected (Scheme 2). Although a similar mechanism could be
proposed for the ortho isomer as well, peaks appear neither at
m/z 120, nor 92, in its mass spectrum; therefore, the homolytic
mechanism appears to play little or no significant role in the
fragmentation of the ortho isomer. Unlike the m/z 93 anion
CHO
5 and 65, which we reported previously from tandem
O
7
1
21
120
CHO
0
60
60
60
80
80
80
100
00
121
O
1
4
9
2
9
3
120
15–18
the elimination of radicals
0
100
120
00
14,19–22
CHO
A high-resolution product-ion
121
9
2
O
120
9
3
0
100
120
m/z
ꢀ
Figure 3. Product-ion spectra of m/z 121 ([M ꢀ H] ) from
ortho- (A), meta- (B) and para-hydroxybenzaldehyde
C) recorded under high-resolution conditions on a Q-Tof
(
instrument.
(
(
6), which readily loses CO to give the cyclopentadienyl anion
7), tandem mass spectrometric experiments show that the
O
H
O
m/z 92 radical anion is very defiant to further fragmentation.
-
H
-CO
A
18
O
B
1
8
O
1
8
C
H
-C18
O
O
m/z 121
m/z 120
m/z 92
H
O
O
O
-
CO
C
H
O
O
O
m/z 95
m/z 123
m/z 93
m/z 123
-
CHO
Scheme 2. Fragmentation mechanism of m/z 121 ion derived
from p-hydroxybenzaldehyde.
Scheme 3. Loss of C18O (A) and C16O (B) from m/z 123 ion
derived from m-hydroxy-[18O]benzaldehyde.
Copyright  2007 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2007; 42: 1207–1217
DOI: 10.1002/jms

1212
A. B. Attygalle et al.
-
CO
143
100
A
O
O
O
O
CHO
71
m/z 143 (9)
CO
OH
O
O
H
H
H
O
m/z 171 (8)
28
-
O
28
1
45
0
H
O
50
100
150
+
O
143
m/z 45
m/z 115 (10)
100
B
O
Scheme 4. Fragmentation pathways of m/z 171 ion derived
CH18
O
from 2-hydroxy-1-naphthaldehyde.
30
H
O
Although not as prominent as that in the spectrum of the
ortho isomer, a peak is observed at m/z 93 in the spectra of
meta and para-hydroxybenzadehyde as well (Figs 1 and 2).
In fact, the peak at m/z 93 is much more intense for the meta
isomer than that for the para isomer. A ‘neutral-loss’ tandem
mass spectrometric experiment, in which fragment ions
from p-hydroxylbenzaldehyde were produced in the source,
established that a 28 Da molecule can be also eliminated
directly from the m/z 121 ion. The relatively low intensity
of the m/z 93 ion in the spectra of meta and para isomers,
compared to that of the ortho isomer, indicated that the m/z
173
18O
28
1
15
47
0
50
100
150
ꢀ
Figure 4. Product-ion spectrum of m/z 171 ([M ꢀ H] ) of
2
2
-hydroxy-1-naphthaldehyde (A), and
-hydroxy-1-[18O]naphthaldehyde (B) recorded on a Quattro
mass spectrometer (collision energy settings of 35 eV).
To demonstrate more general occurrence, and possi-
ble analytical applications of the proposed mechanisms,
we recorded a product-ion spectrum of deprotonated 2-
hydroxy-1-naphthaldehyde (8) (Fig. 4(A)). Since this is a
hydroxy carbaldehyde with two functional groups attached
to adjacent carbon atoms of an aromatic ring, no peak
was observed for a hydrogen radical loss (compare with
Fig. 3(A)). Moreover, the m/z 171 ion (obs. mass 171.0450;
9
3 ion could originate from pathways other than the less
energetic ortho-specific elimination mechanism observed for
1
8
the ortho isomer. Product-ion spectra recorded from O-
labeled hydroxybenzaldehydes supported this hypothesis.
For example, the alternative pathways by which the meta
isomer can lose CO are illustrated in Scheme 3. Of course,
calcd. for C11
CO molecule to produce a peak at m/z 143 (9)(obs. mass
43.0498; calcd. for C10 O, 143.0497) (Fig. 4) by a mech-
7 2
H O 171.0446) showed a facile loss of a
1
5–18
these charge-remote mechanisms
call for more energy
than that required for the facile charge-mediated CO loss
observed for the ortho isomer (Fig. 2). From the relative
intensity of the m/z 93 peak of the product-ion spectrum of
m/z 121 of p-hydroxybenzaldehyde, and that of the m/z 123
1
H
7
anism similar to that illustrated in Scheme 1 for the loss
of CO from ortho-hydoxybenzaldehyde. Subsequently, the
m/z 143 ion loses a molecule of CO to produce a peak at
m/z 115, which presumably represents the H-indene-1-yl
1
8
ion of p-hydroxy-[ O]benzaldehyde which showed a very
1
8
weak peak at m/z 93 for a C O loss from the parent anion,
it is evident that this charge-remote mechanism is not very
significant for the para isomer.
anion (10)(obs. mass 115.0546; calcd. for C
a manner similar to that illustrated in Scheme 1 for the m/z
3 anion from o-hydroxybenzaldehyde. Consecutive losses
9 7
H 115.0548) in
9
In contrast, the meta-isomer appears to undergo both
charge-remote heterolytic and homolytic fragmentations
with similar ease since significant peaks are observed at m/z
of CO from the carbonyl group and the phenolic group
was proven by recoding a product-ion spectrum from the
m/z 173 ion of 2-hydroxy-1-[ O]naphthaldehyde. The m/z
18
9
3 and 92 (Fig. 1(B)). Accurate mass CID spectra recorded
from the m/z 121 ion (obs. mass 121.0291; calcd. for C
21.0290) confirmed the validity of proposed elemental
compositions for the m/z 93 (obs. 93.0337; calcd. for C
3.0340), and m/z 92 (obs. 92.0266; calcd. for C O 92.0262)
18
1
73 ion first loses 30 Da for a C O followed by a 28 Da
H
5
O
2
16
7
elimination for a C O (Fig. 4(B), Scheme 4). Of course, the
first CO loss is much more favorable than the second CO
loss. The small peak at the m/z 45 (Fig. 4(A)) represents the
unprecedented formate anion reported by us previously. In
the spectrum of 2-hydroxy-1-[ O]naphthaldehyde, this peak
1
6
5
H O
9
6
H
4
7
ions. Although one may envisage that the CO loss originates
exclusively from the carbonyl group, a product-ion spectrum
18
becomes m/z 47 (Fig. 4(B)) as predicted by the ortho-specific
mechanism. Interestingly, the fragmentation spectrum of
1
8
of m/z 123 ion derived from m-hydroxy-[ O]benzaldehyde
demonstrated the existence of two separate mechanisms.
The spectrum showed peaks at m/z 93 and 95 for the loss of
7
the anion derived from 1-hydroxy-2-naphthaldehyde anion
is nearly identical to that of 2-hydroxy-1-naphthaldehyde.
Although a peak for the loss of aldehydic hydrogen atom
was not observed from the anion of o-hydroxybenzaldehyde,
an analogous loss of a methyl radical was observed from
1
8
16
C O and C O from the parent anion (Fig. 1(E)). Thus, two
mechanisms are in operation by which either the aldehydic
oxygen (Scheme 3(A)), or the phenolic oxygen (Scheme 3(B))
is lost when CO is eliminated.
0
that of 2 -hydroxyacetophenone (Fig. 5(A)). In fact, a loss
Copyright  2007 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2007; 42: 1207–1217
DOI: 10.1002/jms

CID of hydroxyphenyl carbaldehydes and ketones 1213
O
O
100
A
B
C
D
9
3
9
2
O
O
1
1
1
00
A
9
3
9
2
O
O
O
O
1
35
1
35
O
59
1
20
59
1
20
O
6
5
0
5
0
100
150
0
2
5
5
5
50
75
100
125
1
1
1
00
149
O
00
B
O
92
O
O
9
3
O
135
73
92
O
4
1
93
120
65
120
0
0
50
100
150
2
50
75
100
125
00
O
163
00
C
O
9
2
9
2
O
O
9
3
1
35
87
1
20
1
20
93
0
0
50
100
150
2
50
75
m/z
100
125
00
1
77
O
Figure 5. Product-ion spectrum recorded from m/z 135
ꢀ
0
[
3
M ꢀ H] of 2 -hydroxyacetophenone (A),
92
0
0
-hydroxyacetophenone (B), and 4 -hydroxyacetophenone
O
O
(C) on a Quattro mass spectrometer (all parameters including
120
collision gas pressure and collision energy were kept identical).
101
0
of the alkyl side chain by a charge-remote homolytic fission
mechanism appears to be common to all anions derived from
hydroxylphenyl alkyl ketones (Fig. 6). The m/z 120 radical
anion formed in this way, subsequently loses CO to generate
the radical anion observed at m/z 92.
5
0
100
m/z
150
Figure 6. Product-ion spectrum recorded from m/z 135
ꢀ
0
ꢀ
[
M ꢀ H] of 2 -hydroxyacetophenone (A), m/z 149 [M ꢀ H] of
0
0
0
ꢀ
2
2
2
-hydroxypropiophenone (B), m/z 163 [M ꢀ H] of
Similar to the facile CO loss seen from orthohydrox-
yphenyl carbaldehydes, a somewhat corresponding ketene
ꢀ
-hydroxyisobutyrophenone (C), and m/z 177 [M ꢀ H] of
-hydroxypivalophenone (D) on a Quattro mass spectrometer
0
loss is observed in the spectra of 2 -hydroxyacetophenone
(
all parameters including collision gas pressure and collision
energy were kept identical).
and other related ketones (Scheme 5). For example, the
0
product-ion spectra of m/z 135 ion from 2 -hydroxyaceto-
0
0
phenone (11), m/z 149 from 2 -hydroxypropiophenone (12),
Moreover, the spectrum of m/z 138 anion derived from 2 -
hydroxy-[2,2,2- H
peak at m/z 94 for a loss of D
cific transfer of ˛-hydrogen or deuteron (Fig. 8). On the other
hand, the m/z 93 peak is negligibly weak in the spectrum of
0
2
and m/z 163 from 2 -hydroxyisobutyrophenone (13) all
3
]acetophenone showed its most intense
O confirming the spe-
show a peak at m/z 93 for the loss of ketene (42 Da, 14),
methylketene (56 Da, 15), and dimethylketene (70 Da, 16),
respectively, by a specific transfer of an ˛ hydrogen (Fig. 6).
2
C
C
O
O
R1
R2
R1
R2
R1
+
O
C
C
R2
H
O
OH
OH
m/z = 93
1
2
1
2
1
1
1
1, R = H; R = H; m/z = 135
14, R = H; R = H; 42 Da
1
2
1
2
2, R = H; R = CH ; m/z = 149
15, R = H; R = CH ; 56 Da
3
3
1
2
1
2
3, R = CH ; R = CH ; m/z = 163
16, R = CH ; R = CH ; 70 Da
3 3
3
3
Scheme 5. Loss of ketene from anions derived from 2-hydroxylphenyl alkyl ketones.
Copyright  2007 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2007; 42: 1207–1217
DOI: 10.1002/jms

1214
A. B. Attygalle et al.
OH
O
O
OH
O
C
+
H
HO
HO
OH
m/z 223
OH
m/z 125 (17)
HO
O
Scheme 6. Loss of a ketene from m/z 179 anion derived from phlorocaprophenone.
0
2
-hydroxypivalophenone indicating the probable role of a ꢂ
1
00
A
B
C
OH
O
1
85
[
M-H]
hydrogen atom for the ketene loss (Fig. 6(D)).
Interestingly, the formation of an m/z 125 ion has been
noted from the anion derived from phlorocaprophenone
1), although no rational explanation was provided. We
propose that a ketene is eliminated by mechanism similar to
that given in Scheme 5 to form the ion observed at m/z 125
Scheme 6), which would transform rapidly to a phenoxide
anion (17)(Scheme 1).
Evidently, a negative charge on the phenolic oxygen at
the ortho position and the presence of an ˛ hydrogen are the
prerequisites for this rearrangement. In an indirect way this
is an ortho effect, similar to that well-established for radical
cations.
even for EE cations.
A small peak observed at m/z 93 in the spectra of meta and
para isomers of hydroxyacetophenone (Fig. 5) indicates that
ketene can be lost also by a less significant alternate mecha-
nism (Scheme 7).This minor pathway, somewhat resembles
the CO loss shown in Scheme 3(A). Although, there could be
a small contribution from this charge-remote mechanism for
the ortho isomers as well, it is not large enough to confound
the diagnostic value of the more facile charge-mediated
mechanism specific for the 1,2-relationship. For example, of
the four hydroxyacetonaphthone isomers we examined, only
those bearing a hydroxyl group adjacent to the acyl group
143
42
O
2
3
(
5
9
142
0
(
5
5
5
0
0
0
100
100
100
100
150
100
O
1
85
1
43
[
M-H]
OH
O
2
4,25
42
A few examples of ortho effects have been noted
2
6–28
59
0
150
100
O
143
1
85
[M-H]
OH
4
2
59
O
0
150
showed a facile 42 Da loss of ketene (Fig. 7(A), (B), and
O
0
1
00
D
(
C)), whereas for 6 -hydroxy-2-acetonaphthone the 42 Da
185
1
42
[
M-H]
0
loss is insignificant (Fig. 7(D)). The anion of 6 -hydroxy-
2
-acetonaphthone loses 43 Da to form a radical anion as
expected.
In fact, 18O labeling studies indicate that the mechanism
HO
43
shown in Scheme 8 also operates to a small extent in the
143
O
0
ortho isomers. Even in the spectrum of m/z 177 of 2 -
0
hydroxypivalophenone a tiny peak is observed at m/z 93
50
150
(
Fig. 6(D)). A ˇ-hydrogen transfer (or a transfer from some
m/z
ꢀ
Figure 7. Product-ion spectrum recorded from [M ꢀ H] from
O
0
0
0
1
-hydroxy-2 -acetonaphthone (A),
R1
R2
R1
0
2 -hydroxy-1 -acetonaphthone (B),
+
O C C
R2
0
0
H
3 -hydroxy-2 -acetonaphthone (C), and
O
O
0
0
-hydroxy-2 -acetonaphthone (D) on a Quattro mass
6
m/z = 93
spectrometer (all parameters including collision gas pressure
and collision energy were kept identical).
Scheme 7. Loss of ketene from anions derived from 4-hydrox-
ylphenyl alkyl ketones.
Copyright  2007 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2007; 42: 1207–1217
DOI: 10.1002/jms

CID of hydroxyphenyl carbaldehydes and ketones 1215
O
O
Clearly, the 1,2-relationship of the hydroxyl and car-
baldehyde group is an absolute prerequisite for the proposed
H
2
elimination mechanism. However, the spectrum of 2,3-
dihydoxybenzaldehyde does not show a peak for a H loss
Fig. 9(A)). Although, this is not surprising because the 2-
+
+
CO
R
R
2
OH
H
H
(
hydroxy group is more acidic than the 3-hydroxy group
due to the electron withdrawing effects of the carbaldehyde
group. Consequently, the 2-hydroxy group preferentially
gets deprotonated to the phenoxide anion rendering the
Scheme 8. An alternative way of forming the m/z 93 ion.
other position) should also be taking place since the spectrum
0
2
of 2 -hydroxy[2,2- H
3 (Fig. 8(B)) (the peak observed at m/z 131 is the carboxylate
anion highly specific for 2-hydroxyphenyl alkyl ketones we
2
]heptanophenone shows a peak at m/z
2
labile hydrogen unavailable for the H loss (Scheme 10(A)).
9
Albeit the ionization could have taken place in some
molecules at the 3-hydroxy group (Scheme 10(B)), the anion
formed in this way does not benefit from the electron-
donating assistance of the phenoxide group to enforce a
7
,8
have reported previously ).
The spectra recorded from dihydroxybenzaldehydes
showed further interesting fragmentations. For example, the
spectrum of the anion derived from 2,4-dihydroxybenzal-
dehyde shows a dramatic loss of a 2 Da unit, which can
be attributed also to an ortho effect (Fig. 9(C)). In fact,
this 2 Da loss take precedence over the 28 Da loss char-
acteristic for 2-hydroxyphenyl carbaldehydes. Presumably,
this loss dihydrogen follows either a charge-remote 1,4-
elimination (Scheme 9(A)), or a charge-mediated mechanism
2
H elimination (Fig. 9(A)).
CONCLUSIONS
In addition to the unprecedented formation of a carboxylate
anion noted from the ortho isomers of hydroxyphenyl and
7
hydroxynaphthyl carbaldehydes and ketones, the spectra
of the carbonyl compounds present here demonstrate many
other interesting features. The anions derived from com-
pounds bearing a 1,2 relationship between the hydroxyl and
carbaldehyde groups show a prominent 28 Da loss for the
elimination of a CO molecule. Hydroxyphenyl and hydrox-
ynaphthyl carbaldehydes without the 1,2 relationship show
a 1 Da loss followed by a 28 Da loss for an overall loss of ele-
ments of a žCHO radical. Anions of hydroxyacetophenones
lose the alkyl group by a homolytic fission followed by an
elimination of a CO molecule. Hydroxynaphthyl ketones
(
2
Scheme 9(B)). Since the peak for the H loss is very promi-
nent, presumably the charge-mediated mechanism is pre-
ferred. A spectrum taken from 2,4-dihydroxybenzaldehyde
dissolved in D
group adjacent to the carbaldehyde group should remain
unionized for the H elimination to take place (Scheme 9,
2
O showed a 3 Da loss proving that the OH
2
Fig. 9(D)). Accurate mass measurements made under high-
resolution conditions (R D 15 000) supported the elemental
compositions of the proposed product ions.
A
B
O
O
O
O
-
H2
-H2
O
O
H
O
O
H
C
C
O
O
H
O
H
O
m/z 137
m/z 135
m/z 137
m/z 135
Scheme 9. Proposed mechanisms for the loss of H2 from 2,4-dihydroxybenzaldehyde anion.
100
94
D
O
O
138
CD3
O
O
CD3 62
O
O
92
0
50
100
150
200
100
O
O
O
O
207
9
4
CD
2
O
9
3
O
131
9
2
120
O
CD2
0
50
100
150
200
m/z
ꢀ
0
2
Figure 8. Product-ion spectrum recorded from [M ꢀ H] from 2 -hydroxy[2,2,2- H3]acetophenone (A), and
0
2
2
-hydroxy[2,2- H2]heptanophenone (B) on a Quattro mass spectrometer.
Copyright  2007 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2007; 42: 1207–1217
DOI: 10.1002/jms

1216
A. B. Attygalle et al.
A
B
OH
O
OH
OH
O
O
O
O
OH
O
OH
O
H
H
H
H
H
Scheme 10. Ionization of 2,3-dihydoxybenzaldehyde.
1
1
1
00
A
B
C
REFERENCES
HO
109
-
[
M-H]
137
1
. McClure TD, Schram KH. Computerized interpretation of the
electron ionization mass spectra of nucleoside TMS derivatives.
Nucleosides and Nucleotides 1990; 9: 459.
HO
CHO
108
2. Hill AW, Mortishire-Smith RJ. Automated assignment of high-
resolution collisionally activated dissociation mass spectra
5
3
3
8
1
91
using
a systematic bond disconnection approach. Rapid
0
4
0
60
80
100
120
140
Communications in Mass Spectrometry 2005; 19: 3111.
. Advanced Chemistry Development. Ontario, Canada.
http://www.acdlabs.com [accessed 2007].
3
OH
00
HO
137
136
4. HighChem, Slovak Republic. http://www.highchem.com
accessed 2007].
1
08
[
5
5
. Klagkou K, Pullen F, Harrison M, Organ A, Firth A, Langley GJ.
Approaches towards the automated interpretation and
prediction of electrospray tandem mass spectra of non-
peptidic combinatorial compounds. Rapid Communications in
Mass Spectrometry 2003; 17: 1163.
CHO
92
1
4
1
9
81
119
0
4
0
60
80
100
120
140
37
6. Attygalle AB, Garc ´ı a-Rubio S, Ta J, Meinwald J. Collisionally-
induced dissociation mass spectra of organic sulfate anions.
Journal of the Chemical Society. Perkin Transactions 2 2001; 498.
7. Attygalle AB, Ruzicka J, Varughese D, Sayed J. An unprece-
dented ortho effect in mass spectrometric fragmentation of
even-electron negative ions from hydroxyphenyl carbaldehydes
and ketones. Tetrahedron Letters 2006; 47: 4601.
OH
00
9
1
4
1
1
HO
1
35
CHO
8
. Attygalle AB, Ruzicka J, Varughese D, Sayed J. Corrigendum
to ‘‘An unprecedented ortho effect in mass spectrometric
fragmentation of even-electron negative ions from
hydroxyphenyl carbaldehydes and ketones.’’ Tetrahedron Letters
9
3
65
1
09
0
4
0
60
80
100
120
140
2007; 48: 3485.
9
. Austin MB, Saito T, Bowman ME, Haydock S, Kato A, Moore BS,
Kay RR, Noel JP. Biosynthesis of Dictyostelium discoideum
differentiation-inducing factor by a hybrid type I fatty acid–type
III polyketide synthase. Nature Chemistry and Biology 2006; 2: 494.
OD
-
1
00
D
[M-D]
1
38
9
1
135
DO
60
10. Kumar MK, Sateesh B, Prabhakar S, Sastry GN, Vairamani M.
Generation of regiospecific carbanions under electrospray
ionization conditions and their selectivity in ion-molecule
reactions with CO2. Rapid Communications in Mass Spectrometry
CHO
9
4
4
1
66
110
0
2006; 20: 987.
40
80
m/z
100
120
140
1
1
1
1. Crews P, Rodriguez J, Jaspars M. Organic Structure Analysis,
Oxford University Press: Oxford, 1998; 246.
2. Karni M, Mandelbaum A. The ‘Even-Electron Rule’. Organic
Mass Spectrometry 1980; 15: 53.
3. Krauss D, Mainx HG, Tauscher B, Bischof P. Fragmentation of
trimethylsilyl derivatives of 2-alkoxyphenols: a further violation
of the ‘even-electron rule’. Organic Mass Spectrometry 1985; 20:
614.
Figure 9. (A) Product-ion spectra of m/z 137 derived from
,3-dihydroxybenzaldehyde (A), 3,4-dihydroxybenzaldehyde
B) and 2,4-dihydroxybenzaldehyde (C), and m/z 138
2
(
[
ꢀ
ꢀM ꢀ Dꢁ ] of 2,4-dihydroxybenzaldehyde sprayed as a D2O
solution, recorded on a Quattro mass spectrometer.
1
1
1
1
4. Stirk KM, Kiminkinen LKM, Kentt a¨ maa HI. Ion-molecule
reactions of distonic radical cations. Chemical Reviews 1992; 92:
1649.
5. Gross ML. Charge-remote fragmentation: an account of research
on mechanisms and applications. International Journal of Mass
Spectrometry 2000; 200: 611.
6. Tomer KB, Crow FW, Gross ML. Location of double bond
position in unsaturated fatty acids by negative ion MS/MS.
Journal of the American Chemical Society 1983; 105: 5487.
7. Jensen NJ, Tomer KB, Gross ML. Gas-phase ion decompositions
occurring remote to a charge site. Journal of the American Chemical
Society 1985; 107: 1863.
that show a 1,2 relationship and bear at least one ˛ hydrogen
atom on the alkyl substituent loses a ketene (or a substituted
ketene). In contrast, hydroxynaphthyl ketones without a 1,2
relationship loses the acyl group as a radical. Anions of poly-
hydroxy aldehydes, loses H if the charge is not located on
2
the OH group adjacent to the carbaldehyde group.
Acknowledgements
DV thanks Prof. Ajay K. Bose for laboratory facilities. This
research was supported by funds provided by Stevens Institute
of Technology, NJ.
18. Chen G, Pramanik BN, Bartner PL, Saksena AK, Gross ML.
Multiple-stage mass spectrometric analysis of complex
Copyright  2007 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2007; 42: 1207–1217
DOI: 10.1002/jms

CID of hydroxyphenyl carbaldehydes and ketones 1217
oligosaccharide antibiotics (Everninomicins) in a quadrupole
ion trap. Journal of the American Society for Mass Spectrometry
polyketide synthase in Streptomyces griseus. The Journal of
Biological Chemistry 2002; 277: 4628.
2
002; 13: 1313.
24. McLafferty FW, Turecek F. Interpretation of Mass Spectra.
University Science Books: Mill Valley, CA, 1993.
25. Schwarz H. Some newer aspects of mass spectrometric ortho
effects. Topics in Current Chemistry 1978; 73: 231.
1
9. Yates BF, Bouma WJ, Radom L. Detection of the prototype
phosphonium (CH2PH3), sulfonium (CH2SH2) and chloronium
(CH2ClH) ylides by neutralization-reionization mass
spectrometry: a theoretical prediction. Journal of the American
Chemical Society 1984; 106: 5805.
0. Yates BF, Bouma WJ, Radom L. Distonic radical cations.
26. Reddy PN, Srikanth R, Venkateswarlu N, Rao NR, Srinivas R.
Electrospray ionization tandem mass spectrometric study
of three isomeric substituted aromatic sulfonic acids;
differentiation via ortho effects. Rapid Communications in Mass
Spectrometry 2005; 19: 72.
2
2
2
2
Guidelines for the assessment of their stability. Tetrahedron 1986;
42: 6225.
1. Hill BT, Poutsma JC, Chyall LJ, Hu J, Squires RR. Distonic ions
of the ‘‘Ate’’ Class. Journal of the American Society for Mass
Spectrometry 1999; 10: 896.
2. Thoen KK, Smith RL, Nousiainen JJ, Nelson ED, Kentt a¨ maa HI.
Charged phenyl radicals. Journal of the American Chemical Society
27. Moraes LAB, Sabino AA, Meurer EC, Eberlin MN. Absolute
configuration assignment of ortho, meta, or para isomers by mass
spectrometry. Journal of the American Society for Mass Spectrometry
2005; 16: 431.
28. Xiaolan C, Lingbo Q, Jiansha L, Lei G, Youzhu Y, Yufen Z.
Ortho effects in bis-phosphoric esters of the dihydroxy- and
aminohydroxybenzenes observed by ESI-MS/MS. Journal of
Mass Spectrometry Society of Japan 2004; 2: 63.
1996; 118: 8669.
3. Funa N, Ohnishi Y, Ebizuka Y, Horinouchi S. Properties and
substrate specificity of RppA, a chalcone synthase-related
Copyright  2007 John Wiley & Sons, Ltd.
J. Mass Spectrom. 2007; 42: 1207–1217
DOI: 10.1002/jms