Collision-induced loss of AgH from Ag+ adducts of alkylamines, aminocarboxylic acids and alkyl benzyl ethers leads exclusively to thermodynamically favored product ions

Article abstract of DOI:10.1002/jms.1507

The loss of AgH from [M + Ag]⁺ precursor ions of tertiary amines, aminocarboxylic acids and aryl alkyl ethers is examined by deuterium labeling combined with collision activation (CA) dissociation experiments. It was possible to demonstrate tha

Full text of DOI:10.1002/jms.1507

Research Article
Received: 16 May 2008
Accepted: 28 August 2008
Published online in Wiley Interscience: 28 November 2008
(www.interscience.com) DOI 10.1002/jms.1507
Collision-induced loss of AgH from Ag+
adducts of alkylamines, aminocarboxylic acids
and alkyl benzyl ethers leads exclusively to
thermodynamically favored product ions
∗
¨
Mathias Schafer, Frank Dreiocker and Herbert Budzikiewicz
The loss of AgH from [M + Ag]+ precursor ions of tertiary amines, aminocarboxylic acids and aryl alkyl ethers is examined by
deuterium labeling combined with collision activation (CA) dissociation experiments. It was possible to demonstrate that the
AgH loss process is highly selective toward the hydride abstraction. For tertiary amines and aminocarboxylic acids, hydrogen
originates from the α-methylene group carrying the nitrogen function (formation of an immonium ion). In all cases examined,
the most stable, i.e. the thermodynamically favored product ion is formed. In the AgH loss process, a large isotope effect
operates discriminating against the loss of D. The [M + Ag]+ ion of benzyl methyl ether loses a hydride ion exclusively from the
benzylic methylene group supporting the experimental finding that the AgH loss reaction selectively cleaves the weakest C–H
c
bond available. Copyright ꢀ 2008 John Wiley & Sons, Ltd.
Keywords: Ag+ adducts; tertiary amines; MS/MS-fragmentation; primary isotope effect; AgH loss
Introduction
During our investigations on the gas-phase structure of solution-
phase zwitterions[1,2] we also tested the influence of Ag+ on
the fragmentation behavior of derivatized amino acids. In this
context, we observed the loss of AgH from [M + Ag]+ precursor
ions upon collision activation (CA). The fragmentation reactions
of Ag+ molecular adduct ions of primary amines, α-amino
acids and peptides have been studied extensively inter alia
by Shoeib, Siu and Hopkinson.[3–10] Additional to the tandem
MS experiments, the studies included computational modeling
calculations to elucidate the energetics of the AgH loss reaction
mechanism.[3–10] It turned out that for all [M+Ag]+ precursor ions
possessing a hydrogen atom in the α-position [M − H]+ product
ions are obtained upon CA by loss of AgH.[3,11] Additionally,
O’Hair et al. demonstrated that electrospray-MSn (ESI-MSn) of
other silver amino acid clusters [(M + Ag − H)x + Ag]+ wit+h
M = glycine or N,N-dimethylglycine, deliver Ag3+, Ag5+, Ag7
and silver hydride cluster ions Ag2H+, Ag4H+ and Ag6H+ in the
gas phase.[12]
Scheme 1. Structures of [M + Ag]+ molecular ions of α-amino acids:
charge-solvation 1, salt bridge structure 2 of proline.
structure of proline, the Ag+ ion is bound to the negatively
charged carboxylate group, which in turn is stabilized via
a salt bridge to the protonated nitrogen (structure 2 in
Scheme 1). For amino acids with functionalized side chains,
theory predicts the formation of tridentate complexes where a
side chain heteroatom or an aromatic residue provides a third
binding site for the Ag+ ion (e.g. lysine, histidine, phenylalanine
etc.).[5,8,9,15,16] The rather limited D-labeling studies reported in
literature show that the hydrogen lost in AgH neither originates
from the exchangeable hydrogen atoms bound to nitrogen
or oxygen[3,8,9,17] nor from the benzylic ones in the case of
phenylalanine.[15]
For [M + Ag]+ precursor ions of α,ω-diamino alkanes, a
competition occurs between the loss of AgH and H2.[4] However,
similar reaction schemes were observed for a number of [M + X]+
molecular ions of alkyl amines with singly charged group 8–10
transition metal ions X+ as well as for Cu+ a member of transition
metal group I (electron configuration s0d10).[13,14]
For α-amino acids with alkyl side chains, the most stable
conformation was calculated to be a bidentate complex with
Ag+ coordinated to the nitrogen of the NH2-group and the
carbonyl oxygen (charge-solvation structure 1 in Scheme 1).
Owing to the higher gas-phase basicity of the NH group in
proline, a zwitterionic structure is found to be more stable than
an alternative charge-solvation structure.[3,10] In that zwitterionic
∗
¨
Correspondenceto:Mathias Schafer,InstituteforOrganicChemistry,University
Cologne; Greinstr. 4, 50939 Cologne, Germany.
E-mail: mathias.schaefer@uni-koeln.de
Institute for Organic Chemistry, University Cologne; Greinstr.4, 50939 Cologne,
Germany
c
J. Mass. Spectrom. 2009, 44, 278–284
Copyright ꢀ 2008 John Wiley & Sons, Ltd.

Collision-induced loss of AgH from Ag+ adducts
Experimental
organic layer was washed with saturated NaHCO3-solution and
dried over magnesium sulfate. The solvent was evaporated to give
[α,α-D2]-benzyl bromide as yellowish oil in 80% yield. This oil was
reacted with CH3ONa in methanol for 3 h at 0 ◦C and for 16 h at
20 ◦C. The resulting mixture was taken up in water and extracted
four times with tert-butyl methyl ether. The organic layers were
combined and dried over magnesium sulfate. After removal of the
solvent under reduced pressure [α,α-D2]-benzyl methyl ether was
obtained as yellowish oil in 77% yield.
All chemicals were used as purchased without further pu-
rification (Acros Organics, Geel, Belgium and ABCR, Karlsruhe,
Germany). Toluene and diethyl ether were dried by distillation
from benzophenone and sodium, dichloromethane by distillation
from calcium hydride. Methanol was distilled from magnesia and
all other solvents were distilled over a column prior to use.
Chemicals
6-Dimethylaminohexanoicacid[18] wasderivedfromtherespective
ε-amino acid in an Eschweiler-Clarke reaction. 6-Aminohexanoic
acid was therefore refluxed in HCHO/HCOOH/H2O for 16 h. After
evaporation of the reaction mixture, the resulting white solid
was suspended in toluene. The solvent was removed under
reduced pressure and the crystalline crude product was purified
by crystallization from acetonitrile (yield 74%).
N-Methylproline[19] was produced by methylation of proline by
treatment with an aqueous formaldehyde solution (37%) for 16 h
at 20 ◦C under reductive conditions (Pd/C + H2 at 1 bar). After
filtration over celite, the yellow liquid was concentrated under
reduced pressure. The resulting oil was taken up in methanol.
After removal of the solvent under reduced pressure, the desired
product was obtained as a yellow solid in 98% yield.
[1,1-D2]-1-Dimethylaminohexane[20] was synthesized from hex-
anoicaciddimethylamidebyreductionwithLiAlD4 for20 hat20 ◦C
in diethyl ether. After quenching with saturated Na/K-tartrate so-
lution, the resulting mixture was extracted with tert-butyl methyl
ether four times. The combined organic layers were dried over
magnesium sulfate and concentrated under reduced pressure to
give the product in 89% yield as slightly yellowish oil.
[1-D]-1-Dimethylaminohexane[20–22] was produced in three
steps, starting with the reduction of hexanal with LiAlD4 for 92 h
at 20 ◦C. After the addition of saturated Na/K-tartrate solution,
the reaction mixture was extracted three times with ethyl acetate.
The combined organic layers were dried over magnesium sulfate
and the solvent was removed under reduced pressure. [1-D]-
hexanol was isolated as colorless oil in 86% yield. [1-D]-Hexanol
was dissolved in dichloromethane and subsequently treated with
0.5 equ. PBr3 for 2 h at 0 ◦C and for 63 h at 20 ◦C. After the
addition of saturated NaHCO3-solution, the organic layer was
separated and the aqueous layer was extracted three times with
tert-butyl methyl ether. The organic layers were combined, dried
over magnesium sulfate and the solvent was removed under
reduced pressure to give [1-D]-1-bromohexane as colorless oil in
65% yield. [1-D]-1-bromohexane was taken up in methanol and
treated with (CH3)2NH (40% in H2O) for 1 h at 0 ◦C and for 16 h
at 20 ◦C. After addition of sodium hydroxide (10% in H2O), the
mixture was extracted with tert-butyl methyl ether four times
and the combined organic layers were dried over magnesium
sulfate. After removal of the solvent under reduced pressure [1-
D]-1-dimethylaminohexane was isolated as yellowish oil in 64%
yield.
Benzyl trideuteromethyl ether[23] was synthesized by re-
acting benzyl bromide for 1 h at 20 ◦C with CD3ONa in
2-methyltetrahydrofuran. After the addition of water, the aqueous
layer was extracted with ethyl acetate two times. The combined
organic layers were dried over magnesium sulfate and the sol-
vent was removed under reduced pressure to give the product as
yellowish liquid in 80% yield.
[α,α-D2]-Benzylmethylether[20,21,23] wasobtainedinathreestep
synthesis, beginning with the reduction of benzoic acid ethyl ester
with LiAlD4 in diethyl ether for 1 h at 0 ◦C and for 2 h at 35 ◦C. After
acidification with sulfuric acid (25%) the mixture was extracted
with ethyl acetate four times. The combined organic layers were
dried over magnesium sulfate and concentrated under reduced
pressure to give the product as liquid in 91% yield. The resulting
[α,α-D2]-benzyl alcohol was treated with 0.5 equ. PBr3 in CH2Cl2
for 1 h at 0 ◦C and for 15 h at 20 ◦C. After the addition of water, the
Mass spectrometry
Mass spectral data were obtained with a MAT 900 ST instrument
providinganEB-quadrupoleiontrap(QIT)geometryandequipped
withanESIIIionsource(ThermoFisherScientific,formerlyFinnigan
MAT, Bremen, Germany). Analyte solutions were pumped through
a fused silica capillary (inner diameter 100 µm; Mascom, Bremen,
Germany) by a syringe pump (Harvard Apparatus, Holliston, MA,
USA) safeguarding for a constant flow rate of 3 µl/min. The fused
silica capillary was fed through a stainless steel capillary to which
the ESI spray voltage of 3.6 kV was applied. No additional sheath
gas was used to support the spraying process. The front end of
the ESI II ion source was a heated transfer capillary which was
held at 230 ◦C to provide complete desolvation. ESI-ion source
conditions (adjustable voltages on the tube, ESI-capillary and the
skimmer) were optimized with regard to maximum abundance of
the desired [M + Ag]+ ion species selected as precursor ions for
the CA-MS/MS experiments. The resolution of the double focusing
two sector field analyzer system was adjusted to be >5000 (10%
valley). The samples were dissolved in methanol or water to
give stock solutions with approximately 0.01 mM concentration.
Ag+ adduct ions of the compounds of interest were generated
by adding 10 µl of a 0.01 mM solution of CF3COOAg in CH3CN
to the respective analyte solution prior to ESI-MS experiments.
The ESI-MS spectra of the analytes containing nitrogen exhibited
predominantly protonated molecular ions [M + H]+ along with
[M + Ag]+ adduct ions of moderate abundance. In the ESI-MS
spectrum of benzyl methyl ether the sodium adduct ion [M+Na]+
exceeded the respective [M + Ag]+ ion in abundance. The ESI-MS
spectrumofhexylmethyletherexclusivelyexhibitedthe[M+Na]+
molecular ion. All ion source optimization efforts failed for that
analyte to generate [M + Ag]+ species of notable abundance.
Fragmentation reactions of the [M + Ag]+ precursor ions were
induced by low energy CA either in an octapole located in front of
the QIT or in the QIT itself (helium was used in the QIT as bath gas
with ∼2 × 10−3 Pa; helium diffusing in the octapole was utilized
as collision gas therein, collision gas pressure in the octapole was
not measured separately). For all product ion experiments mono-
isotopic precursor ion selection was adjusted. The center of mass
collision energy Ecm and the laboratory collision energy Elab of the
tandem MS measurements in either the QIT or the octapole were
notexplicitlydetermined.[24] Appropriateexperimentalconditions
were adjusted independently for each individual experiment to
yield significant analytical information.
The AgH loss reaction was investigated toward a primary kinetic
isotope effect (KIE) by tandem MS with the [M+Ag]+ precursor ion
c
J. Mass. Spectrom. 2009, 44, 278–284
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/jms

M. Scha¨fer, F. Dreiocker and H. Budzikiewicz
of [1-D]-1-dimethylaminohexane. General understanding of KIEs
distinguishes certain cases.[25] For the examined fragmentation
reaction, the precursor ion decomposes to give identical but
isotopically distinct product ions, i.e. [M + Ag − AgH]+ or
[M + Ag − AgD]+. In this case, it is justified to approximate a
primary KIE (i.e. the ratio of kinetic constants of the competing
fragmentation channels as functions of internal energy of the
precursor species) by the ratio of the signal intensities of
the respective product ions. Furthermore, it is reasonable to
assume that a given KIE of that kind might not vary much
over a narrow range of energy.[25] The experiments with [1-
D]-1-dimethylaminohexane were executed in the QIT and the
octapole of the Finnigan MAT 900 instrument under respective CA
conditions. Both experiments yielded a ratio of the product ion
abundances related to the AgH-to-AgD loss of about 2 : 1. Hence,
the distribution of internal energy of the precursor ions after CA in
the QIT or in the octapole seems to be at least comparable.
The measurement of the exact ion masses of the [M + 107Ag]+
and [M+109Ag]+ precursor ions of 6-dimethylaminohexanoic acid
and of the shared [M − H]+ product ion, i.e. generated after the
107/109AgH loss, were conducted on LTQ-Orbitrap XL instrument
(Thermo Fisher Scientific, Bremen, Germany) at a resolution of
100 000 fwhh by Dr E. Damoc. The measurement of the exact
ion masses were first carried out with e+xternal calibration and
repeated referring to 107Ag+ and 109Ag as internal standard
ions. The absolute experimental error of the exact ion mass
measurements ranged between 0.3 and 0.4 mmu, the relative
error was 1.1–1.9 ppm. The contribution of Dr E. Damoc (Thermo
Fisher, Bremen Germany) is explicitly acknowledged.
amines with an unsymmetrical substitution pattern to enable a
differentiation of the [M + Ag − AgH]+ immonium ions observed
as product ions of the AgH loss.
Hence,
the
[M + Ag]+
molecular
ion
of
6-
dimethylaminohexanoic acid was selected for CA experiments
(in all investigations, parallel experiments were performed
with ions containing the isotopes 107Ag and 109Ag). CA of the
[M + Ag]+ molecular ion of 6-dimethylaminohexanoic acid
delivered exclusively the respective [M + Ag − AgH]+ product
ion (Fig. 1). A control experiment with the COOD analog of
6-dimethylaminohexanoic acid showed that the deuterium atom
is completely retained in the [M + Ag − AgH]+ product ion,
i.e. an exclusive AgH loss was observed. With that we assume
that the tertiary nitrogen plays a more important role for the
complexation of the silver ion in the [M + Ag]+ ion than the
oxygens of the acid moiety. However, the experimental findings
are in good agreement with the published data on the AgH loss
reactions of amines.[3,8,9,17] Nevertheless, the exact structure of
the [M + Ag − AgH]+ product ion cannot be deduced.
To yield structural information on the particular immonium
ion and to study the selectivity of this reaction [1,1-D2]-1-
dimethylaminohexane was chosen as a model compound for
the analog amino acid. The [M + Ag]+ precursor ion of [1,1-
D2]-1-dimethylaminohexane exhibits an exclusive AgD loss upon
CA in the QIT (Fig. 2), thus proving evidence for the exclusive
formation of the thermodynamically favored immonium ion
(structure 3 in Fig. 2). To see to what extent an isotope effect[25]
may operate, [1-D]-1-dimethylaminohexane was synthesized and
investigated. CA experiments of the respective mono-deuterated
[M + Ag]+ precursor ion in the octapole and in the QIT exhibited
a ratio of the AgH-to-AgD loss of about 2 : 1 (Fig. 3). It is worth
noting that notwithstanding the substantial isotope effect and
despite of the statistical ratio of 1 : 3 for CD2/2CH3 for [1,1-D2]-
1-dimethylaminohexane elimination occurs only from the CD2
methylene group. In this context, it is interesting to note that an
isotope effect of similar extend (1.6 and 1.7) discriminating against
deuterium was found for the abstraction of benzylic hydride ions
Results and Discussion
For our systematic study of the AgH loss reaction, we selected
tertiary amino acid compounds with the intention to block the
competing H2 loss fragmentation channel that is observed for
[M + Ag]+ molecular ions of primary and secondary amines upon
CA.[4,10–13] Furthermore, we decided to study deuterated tertiary
Figure 1. QIT-MS2 product ion spectrum of the [M + 107Ag]+ ion of 6-dimethylaminohexanoic acid at m/z 266. The neutral loss of ꢀm = 108 u was
determined by exact mass measurements of the precursor and product ion to be the loss of 107AgH. Analogous results were obtained for the loss of 109Ag
from [M + 109Ag]+.
c
www.interscience.wiley.com/journal/jms
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
J. Mass. Spectrom. 2009, 44, 278–284

Collision-induced loss of AgH from Ag+ adducts
Figure 2. QIT-MS2 product ion spectrum of the 107Ag+ molecular adduct ion of [1,1-D2]-1-dimethylaminohexane at m/z 238. The exclusive neutral loss
of ꢀm = 109 u shows the selective loss of 107AgD and the formation of the respective [M + Ag − AgD]+ immonium ion 3 at m/z 129.
Figure 3. Octapole-MS/MS product ion spectrum of the [M+107Ag]+ ion of [1-D]-1-dimethylaminohexaneat m/z 237. An isotope effect was documented
by the ratio of the signal intensities of the product ions at m/z 128: 129 (i.e. the loss of AgD vs loss of AgH) which was found to be ≈1:2.
by tert-butyl cations (CH2 vs CD2).[26,27] Analog KIEs were found for
the highly regioselective CA decomposition FeO+ adduct ions of
methoxybenzenes and related analytes, in which the C–H bond of
the methoxy group was cleaved.[28] Finally, the gas-phase reaction
of N,N-dimethylanilines with the oxo iron(IV) porphyrin radical
cation exhibits a formal hydride transfer which is consistent with
the operation of a substantial KIE.[29]
However, the results presented above show with convincing
clearness that the AgH loss strictly accounts for the stability of the
product ion as it exclusively leads to the formation of the higher
substituted immonium ion regardless to the hydrogen isotope
present.
an adequate candidate for this purpose as the AgH loss behavior
of the natural amino acid analog was closely examined before.
In N-methylproline, the proton affinity[30] of the nitrogen atom
is further increased as a result of the additional alkyl substituent
making a ‘‘salt bridge’’ structure of the [M + Ag]+ ion in the
gas phase energetically even more favored than in the case
of proline (structure 2 in Scheme 1). The loss of AgH is the
main fragmentation process upon CA of the [M + Ag]+ ion
of N-methylproline. The hydrogen atom of the acid moiety is
retained in the [M + Ag − AgH]+ product ion of N-methylproline
as a D2O exchange experiment shows. In that experiment, an
interesting loss of 3 u is observed with minimal abundance (data
not shown). In analogy to published data of proline, we conclude
that a HD elimination is detected, which requires an internal
protonation of the nitrogen atom prior to the fragmentation
After the examination of the tertiary amine with different n-alkyl
substituents and analogous tert-amino acid compounds we chose
a cyclic analog for further testing. Obviously, N-methylproline is
c
J. Mass. Spectrom. 2009, 44, 278–284
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/jms

M. Scha¨fer, F. Dreiocker and H. Budzikiewicz
reaction.[10] Accordingly, the presence of a gas-phase zwitterion
of N-methylproline in the [M + Ag]+ ion can be proposed.
These considerations offer another aspect to the discussion of
the experimental results achieved with 6-dimethylaminohexanoic
acid. As presented above in the product ion experiment, the
respective [M + Ag]+ ion loses exclusively AgH. Consequently,
the absence of H2 loss in the case of 6-dimethylaminohexanoic
acid points toward the presence of a charge solvation structure in
the [M + Ag]+ ion. With that, the criteria that are determining the
formationofgas-phasezwitterionscanbeevaluatedanew. Clearly,
the amphoteric character of the analyte has to be sufficiently
pronounced (i.e. a high gas-phase basicity and acidity), but,
the molecule structure is crucial as the formation of efficient
salt bridge interactions has to be arranged to substantially
stabilize a zwitterion in the gas phase. According to this
interpretation, the latter criterion permits the formation of a
zwitterion structure of N-methylproline but prevents the same
for 6-dimethylaminohexanoic acid in the respective [M + Ag]+
ions. However, it must be emphasized, that the nature and the
physicochemical character of the counter ion, i.e. the ion radius,
charge state and polarizability play a major role in the formation
and stabilization of zwitterions as well.[31,32]
According to Pearson,[35] Ag+ is a soft Lewis acid and should
thereforebindpreferentiallytosoftbaseslikeN-orS-functionalities
rather to the harder rated oxygen if available at the same time.
This assumption is convincingly supported by the results of the
series of CA experiments with the [M + Ag]+ ions of the amino
acid compounds as the exchangeable hydrogens are retained in
the [M + Ag − AgH]+ product ions.
In that context it seemed of interest to test the behavior of
Ag+ toward isolated oxygen functionalities. The only pertinent
observation in this respect reported in literature is that several
methyl glycosides[36] form [M + Ag]+ ions under Fast Atom
Bombardment (FAB) conditions. The silver adduct ions of these
molecules lose AgH when the methoxy group and the C-2 bound
hydroxyl group are in trans position (observed in mass-analyzed
ion kinetic energy, MIKE, experiments). Specific D-labeling at C-1,
C-2 and C-6 and exchange of the four OH-groups to OD showed
that only the anomeric D (C-1) is lost as AgD. Coordination of
Ag+ with the hydroxyl group at C-2 and elimination of AgH in a
five-membered transition state was suggested.
To examine [M + Ag]+ of compounds containing an oxygen
we chose compounds with an ether functionality, instead of
the tertiary amino nitrogen. In a first attempt n-hexyl methyl
ether was tested. However, we failed to generate [M + Ag]+
ions of notable abundance of that dialkyl ether. The obviously
lacking stability of the [M + Ag]+ ions of this compound is
very likely due to the absence of a second coordination site
for the Ag+ ion. This prerequisite is however fulfilled in the
case of benzyl methyl ether where the twofold coordination of
Ag+ to the oxygen and to the π-system of the phenyl ring
is responsible for the observation of an abundant [M + Ag]+
ion of obviously sufficient stability (cf phenylalanine).[5,15,16,30]
Specific D-labeling of the methyl and the benzylic methylene
group showed exclusive AgH loss from the former and AgD
loss from the latter (Fig. 4(a) and (b)). These results are in
complete accordance to the experimental findings of the tertiary
amines and respective amino acids. In all the cases investigated
exclusively the most stable fragment ion is formed by loss
of AgH.
For the silver and the alkali metal adductions of proline [M+X]+
with X = Li, Na, K, Ag theory predicts a gas-phase structure
including an internal salt bridge, i.e. a zwitterionic structure in
which the negatively charged carboxylate group coordinates the
metal ion and the nitrogen function is protonated (salt bridge
structure 2 in Scheme 1).[5,10,33,34] This structural assignment is
supported by the results of the CA experiments: a loss of H2
(2u) is observed for the [M + Ag]+ precursor ion of proline and
a loss of HD (3u) is found in an analogous CA experiment with
D2O exchange.[10] Although this unique fragmentation channel
is of low abundance, it is a strong hint toward the presence of
zwitterionic structure of proline and its N-methylated analog in
their gas-phase [M + Ag]+ ions.
The structure of the [M + Ag − AgH]+ immonium ion of proline
was elucidated by CA experiments. Such an experiment with
the [M + Ag]+ ion of [2-D]-proline showed retention of the D-
atom, i.e. a loss of AgH and of H2 was detected.[10] Additionally, the
exchangeableH-atomsatthecarboxylandtheaminefunctionality
are retained during the loss of AgH. In agreement with the
structures for the respective immonium ions discussed above
the [M + Ag − AgH]+ ion of proline was assigned to be structure
4a (Scheme 2). Owing to the very similar fragmentation behavior
of N-methylproline and proline regarding the loss of AgH and H2
we assume that both immonium ions formed are closely related
(structure 4a and 4b in Scheme 2).
In addition, to further the ESI-tandem-MS investigations of silver
adductionsofhydrocarbons,[37] theexaminationofrespectivegas-
phase [M + Ag]+ species by extended density functional theory
calculations are projected.
Conclusions
The study of the AgH loss reaction observed upon CA dissociation
from [M + Ag]+ precursor ions yielded detailed information
on the structure of the product ions and indirect evidence
on the structure of the precursor ions as well. By selective
labeling experiments it was possible to demonstrate that the AgH
loss process is highly selective toward the hydride abstraction.
Tertiary amines and aminocarboxylic acids lose a hydride
from the α-methylene group carrying the nitrogen function
(formation of an immonium ion). In all cases examined, the
thermodynamically favored product ion, i.e. the most effectively
stabilized immonium ion, is formed. The [M + Ag]+ ion of benzyl
methyl ether loses a hydride H- exclusively from the benzylic
position, again forming the most stable oxonium product ion
[M + Ag − AgH]+. The formation of the most stable product
ion seems to be a universal feature of the AgH loss reaction
independent from the nature of the binding partners that form
Scheme 2. Structures of [M+Ag−AgH]+ immonium product ions formed
by the AgH loss from [M + Ag]+ precursor ions of proline 4a and its
N-methyl derivative 4b.
c
www.interscience.wiley.com/journal/jms
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
J. Mass. Spectrom. 2009, 44, 278–284

Collision-induced loss of AgH from Ag+ adducts
(a)
(b)
Figure 4. (a) QIT-MS2-product ion spectrum of the [M+107Ag]+ ion of benzyl trideuteromethyl ether at m/z 232, (b) QIT-MS2-product ion spectrum of the
[M + 107Ag]+ ion of [α,α-D2]-benzyl methyl ether at m/z 231. The formation of the low abundant product ions at m/z 191 in (a) and 190 in (b) cannot be
reasonably explained by CA of the benzyl methyl ether. We assume that the detected neutral loss of 41 u points toward the unintended CA dissociation
of an isobaric species at m/z 232 and 231, respectively.
the complex with the silver ion (e.g. nitrogen, oxygen, aromatic
π-system).
examined by Quadrupole Ion Trap (QIT) MS2 and infrared
multiphoton dissociation fouriertransform ion cyclotron resonance
mass spectrometry IRMPD-FT-ICR MS: Evidence for Salt-bridge
structures in the Gas Phase. International Journal of Mass
Spectrometry 2004, 237/1, 33.
The examination of the [M + Ag]+ ion of proline and its N-
methylated derivative gave rise to an additional product ion
species formed by the loss of H2. Labeling experiments clearly hint
toward the presence of zwitterionic structures for those amino
acid compounds that show this characteristic neutral loss in the
gas-phase. However, the hydrogen loss reaction was only found
with minimal abundance and is therefore error-prone. In that
respect more evidence has to be collected.
[3] R. N. Grewal, C. F. Rodriquez, T. Shoeib, I. K. Chu, Y. P. Tu,
A. C. Hopkinson, K. W. M. Siu. Elimination of AgR (R = H, CH3, C6H5)
from Collisionally-activated argentinated amines. European Journal
of Mass Spectrometry 2000, 6, 187.
[4] T. Shi, J. Zhao, T. Shoeib, K. W. M. Siu, A. C. Hopkinson. Fragmenta-
tionofsinglychargedsilver/α,ω-diaminoalkanecomplexes:compe-
tition between the loss of H2 and AgH molecules. European Journal
of Mass Spectrometry 2004, 10, 931.
[5] T. Shoeib, K. W. M. Siu, A. C. Hopkinson. Silver ion binding energies
of amino acids: use of theory to assess the validity of experimental
silver ion basicities obtained from the kinetic method. Journal of
Physical Chemistry A 2002, 106, 6121.
References
[1] C. Schmuck, T. Rehm, L. Geiger, M. Scha¨fer. Synthesis and Self-
association properties of flexible guanidiniocarbonyl Pyrrole-
carboxylate-zwitterions in DMSO: Intra- versus intermolecular
charge interactions. Journal of Organic Chemistry 2007, 72, 6162.
[2] M. Scha¨fer, C. Schmuck, L. Geiger, M. J. Chalmers, C. L. Hendrickson,
A. G. Marshall. Structurally related Non-covalent complexes
[6] T. Shoeib, C. F. Rodriquez, K. W. M. Siu, A. C. Hopkinson.
A
comparisonofcopper(I)andsilver(I) complexesofglycine,diglycine
and triglycine. Physical Chemistry Chemical Physics 2001, 3, 853.
[7] I. K. Chu, T. Shoeib, X. Guo, C. F. Rodriquez, A. C. Hopkinson,
K. W. M. Siu, T. C. Lau. Characterization of the product ions from the
c
J. Mass. Spectrom. 2009, 44, 278–284
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/jms

M. Scha¨fer, F. Dreiocker and H. Budzikiewicz
Collision-induced dissociation of argentinated peptides. Journal of
the American Society for Mass Spectrometry 2001, 12, 163.
[8] T. Shoeib, A. Cunje, A. C. Hopkinson, K. W. M. Siu. Gas-phase
fragmentation of the Ag+-phenylalanine complex: Cation-π
interactions and radical cation formation. Journal of the American
Society for Mass Spectrometry 2002, 13, 408.
[9] H. Grade, N. Winograd, R. G. Cooks. Cationization of organic
molecules in secondary ion mass spectrometry. Journal of the
American Chemical Society 1977, 99, 7725.
Marked contrast between kinetic and thermodynamic control of
product composition. Journal of Organic Chemistry 1991, 56, 2011.
[23] A. Williamson. Ueber die theorie der aetherbildung. Justus Liebigs
Annalen der Chemie 1851, 77, 37.
[24] J. M. Wells, S. A. McLuckey. Methodology of collisional activation
and dissociation. In Encyclopedia of Mass Spectrometry, vol. 1,
M. L. Gross, R. Caprioli (eds). Elsevier: Amsterdam, 2003, 441.
[25] P. J. Derrick. Isotope effects in fragmentation. Mass Spectrometry
Reviews 1983, 2, 285.
[10] T. Shoeib, A. C. Hopkinson, K. W. M. Siu. Collision-induced dissoci-
ation of the Ag+-proline complex: fragmentation pathways and
reaction mechanisms – a synergy between experiment and theory.
Journal of Physical Chemistry B 2001, 105, 12399.
[11] S. W. Sigsworth, A. W. Castleman Jr. Rates of hydride abstraction
from amines via reactions with ground-state Ag+ and Cu+. Journal
of the American Chemical Society 1989, 111, 3566.
[26] H. E. Audier, C. Monteiro, DCMR Mourgues, D. Berthomieu.
Protonated alkylbenzenes: Hydride abstraction in π complexes.
Organic Mass Spectrometry 1990, 25, 245.
[27] D. Kuck, C. Matthias. Formation of gaseous π and Ion-neutral
complexes as probed by interannular tert-Butyl cation transfer
in protonated tert-Butyl-substituted diphenylalkanes. Journal of the
American Chemical Society 1992, 114, 1901.
[12] G. N. Khairallah, R. A. J. O’Hair. Gas phase synthesis and reactivity
[28] M. Bro¨nstrup, D. Schro¨der, H. Schwarz. A mechanistic study of the
FeO+-mediated decomposition pathways of phenol, anisol, and
their thio analogues. Chemistry-A European Journal 2000, 6, 91.
[29] B. Chiavarino, R. Cipollini, M. E. Crestoni, S. Fornarini, F. Lanucara,
A. Lapi. Probing the compound I-like reactivity of a bare High-
valentoxoironprophyrincomplex:Theoxidationoftertiaryamines.
Journal of the American Chemical Society 2008, 130, 3208.
[30] D. H. Aue, H. M. Webb, M. T. Bowers. Quantitative proton affinities,
ionizationpotentials,andhydrogenaffinitiesofalkylamines.Journal
of the American Chemical Society 1976, 98, 311.
of Agn and Agn−1H+ cluster cations. Journal of the Chemical
+
Society-Dalton Transactions 2005: 2702.
[13] K. Eller, H. Schwarz. Organometallic chemistry in the gas phase.
Chemical Reviews 1991, 91, 1121.
[14] M. A. Tolbert, J. L. Beauchamp. Homolytic and heterolytic Bond-
dissociation energies of the 2nd-row Group-8, Group-9, and Group-
10 diatomic Transition-metal hydrides – correlation with electronic
structure. Journal of Physical Chemistry 1986, 90, 5015.
[15] B. A. Perera, M. P. Ince, E. R. Talaty, M. J. Van Stipdonk. Gas phase
attachmentofwaterandmethanoltoAg(I)complexeswithα-amino
acids in an ion trap mass spectrometer. Rapid Communications in
Mass Spectrometry 2001, 15, 615.
[31] N. C. Polfer, J. Oomens, R. C. Dunbar. IRMPD spectroscopy of Metal-
Ion/Tryptophan complexes. Physical Chemistry Chemical Physics
2006, 8, 2744.
[16] B. A. Perera, A. L. Gallardo, J. M. Barr, S. M. Tekarli, V. Anbalagan,
E. R. Talaty, M. J. Van Stipdonk. Influence of a ring substituent
on the tendency to form H2O adducts to Ag+ complexes with
phenylalanine analogues in an ion trap mass spectrometer. Journal
of Mass Spectrometry 2002, 37, 401.
[17] E. R. Talaty, B. A. Perera, A. L. Gallardo, J. M. Barr, M. J. Van Stipdonk.
Elucidation of fragmentation pathways for the Collision-induced
dissociation of the binary Ag(I) complex with phenylalanine. Journal
of Physical Chemistry A 2001, 105, 8059.
[32] N. C. Polfer, J. Oomens, D. T. Moore, G. von Helden, G. Meijer,
R. C. Dunbar.InfraredspectroscopyofphenylalanineAg(I)andZn(II)
complexesinthegasphase.JournaloftheAmericanChemicalSociety
2006, 128, 517.
[33] P. B. Armentrout, R. M. Moison. The special Five-membered ring of
proline: An experimental and theoretical investigation of alkali
metal cation interactions with proline and its four- and six-
membered ring analogues. Journal of Physical Chemistry A 2006,
110, 3933.
[18] S. Rahal, L. Badache. An anomalous Eschweiler-clarke reaction.
Tetrahedron Letters 1991, 32, 3847.
[34] C. Kapota,J. Lemaire,P. Maitre,G. Ohanessian.Vibrationalsignature
of charge solvation vs salt bridge isomers of sodiated amino acids
in the gas phase. Journal of the American Chemical Society 2004,
126, 1836.
[19] L. Aurelio, J. S. Box, R. T. C. Brownlee, A. B. Huges, M. M. Sleebs. An
efficient synthesis of N- methylamino acids by way of intermediate
5-oxazolidinones. Journal of Organic Chemistry 2003, 68, 2652.
[20] R. F. Nystrom, W. G. Brown. Reduction of organic compounds by
lithium aluminum hydride. I. Aldehydes, ketones, esters, acid
chlorides, and acid anhydrides. Journal of the American Chemical
Society 1947, 69, 1197.
[35] R. G. Pearson. Hard and soft acids and bases. JournaloftheAmerican
Chemical Society 1963, 85, 3533.
[36] J. M. Berjeaud, F. Couderc, J. C. Prome. Stereochemically controlled
decompositionofSilver-cationizedmethylglycosides.OrganicMass
Spectrometry 1993, 28, 455.
[21] C. S. Hornberger Jr,
R. F. Heitmiller,
I. C. Gunsalus,
[37] E. Bayer, P. Gfro¨rer, C. Rentel. Coordination-ionspray-MS (CIS-MS), a
universaldetectionandcharacterizationmethodfordirectcoupling
with separation techniques. Angewandte Chemie-International
Edition 1999, 38, 992.
G. H. F. Schnakenberg, L. J. Reed. Synthesis of DL-α-lipoic acid.
Journal of the American Chemical Society 1953, 75, 1273.
[22] Y. Naruta, Y. Nishigaichi, K. Maruyama. Highly selective trapping of
unsymmetrical pentadienyl anions by silicon and tin electrophiles.
c
www.interscience.wiley.com/journal/jms
Copyright ꢀ 2008 John Wiley & Sons, Ltd.
J. Mass. Spectrom. 2009, 44, 278–284