Isomerization of the constituents of ion/neutral complexes during the fragmentation of protonated dialkyl-substituted 1,3-diphenylpropanes

Article abstract of DOI:10.1016/j.ijms.2010.10.007

The fragmentation of gaseous ion/neutral complexes [R⁺? C₆H₅CH₂CH₂CH₂C ₆H₄-R′] with (i) R = R′ = C₄H ₉, (ii) R = C₄H₉ and R

Full text of DOI:10.1016/j.ijms.2010.10.007

International Journal of Mass Spectrometry 306 (2011) 167–174
Contents lists available at ScienceDirect
International Journal of Mass Spectrometry
journal homepage: www.elsevier.com/locate/ijms
Isomerization of the constituents of ion/neutral complexes during the
fragmentation of protonated dialkyl-substituted 1,3-diphenylpropanes
Dietmar Kuck∗, Carsten Matthias, Dieter Barth, Matthias C. Letzel
Department of Chemistry, Bielefeld University, Universitätsstraße 25, 33615 Bielefeld, Germany
a r t i c l e i n f o
a b s t r a c t
The fragmentation of gaseous ion/neutral complexes [R+· · ·C6H5CH2CH2CH2C6H4–Rꢀ] with (i)
R = Rꢀ = C4H9, (ii) R = C4H9 and Rꢀ = CH3 and (iii) R = C6H11 and Rꢀ = H has been studied by CI(CH4)-MIKE spec-
trometry of the corresponding alkyl-substituted 1,3-diphenylpropanes. Different from all other isomers
containing two para-alkyl substituents, the [M+H]+ ion generated from the symmetrical ion [(4-tert-C4H9-
C6H4)CH2CH2CH2(C6H4-4-tert-C4H9) + H]+ shows the characteristic fragmentation pattern of ion–neutral
complexes containing a meta-alkyl-substituted 1,3-diphenylpropane. This indicates a proton-induced
1,2-shift of one or even both of the tert-C4H9 groups and requires the presence of the meta-(tert-C4H9)-
substituted diphenylpropane as the neutral constituent of the eventually fragmenting I/N complex. As
Article history:
Received 31 August 2010
Received in revised form 6 October 2010
Accepted 7 October 2010
Available online 14 October 2010
Dedicated to Professor Tino Gäumann on
the occasion of his 85th birthday.
+
a consequence, it appears that the reactive complex [C4H9 · · ·C6H5CH2CH2CH2(C6H4-3-tert-C4H9)] is
Keywords:
Ion/neutral complexes
Alkyl cation
+
formed prior to the generation of the expected “para-isomer”, [C4H9 · · ·C6H5CH2CH2CH2(C6H4-4-tert-
C4H9)]. Isobutyl analogues, such as [(4-iso-C4H9-C6H4)CH2CH2CH2(C6H4-4-iso-C4H9) + H]+, do not show
+
Isomerization
evidence for the intermediacy of “isomerized” I/N complexes containing a tert-C4H9 ion. The fragmen-
tation of ion–neutral complexes containing C6H11 ions, formed from the [M+H]+ ions of (4-cyclohexyl)-
Proton transfer
Hydride ion transfer
Metastable ion
+
and of 4-(1-methylcyclopentyl)-substituted 1,3-diphenylpropane, indicate that the C6H11+ ions only par-
tially retain their structural identity: while the secondary isomer, (CH2)5 > CH+, predominantly transfers
a proton in competition to hydride abstraction, indicating its stronger Bronsted acidity, the tertiary iso-
mer, (CH2)4 > C+CH3, mainly reacts by hydride abstraction. In spite of the partial isomerization, deuterium
labelling experiments corroborate the usual regioselectivity of the hydride abstraction from the benzylic
methylene groups in both cases.
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
“designed” gaseous ions that are prone to form “tailored” gaseous
ion/neutral complexes may have its justification even nowadays.
For an organic chemist who is fascinated by the beauties
of organic mass spectrometry, the unimolecular formation of
ion/neutral complexes as reactive intermediates [1–9] during the
fragmentation of synthetically designable precursor molecules is a
long-lasting motif of research. And since the ionic and the neu-
tral component of such I/N complexes behave, at least in part,
physical in nature, physical chemists may keep being fascinated
as well. While the physical basis of I/N complexes has been well
established [10,11], the chemical diversity of mass spectrometric
fragmentation reactions that necessarily require the intermediacy
of I/N complexes is still growing [12,13]. More recent examples
have not fallen short of the fascination on the phenomenon, the
more so as understanding of the chemistry in the highly diluted
gas phase of a mass spectrometer is often ignored. Therefore, as
mentioned in previous reports in this series [14–23], the study of
The “organic-synthesis” approach to study the reactivity
of small carbenium ions solvated within I/N complexes ␣,␻-
diphenylalkanes and related alkylbenzenes [7,8,14–21,24–27] as
well as functionalized neutrals [23,28–31] has brought many
detailed insights into the “intra-complex” chemistry (Scheme 1).
In particular, tertiary cations such as tert-C4H9+, released by
protonolysis into complexes with ␣,␻-diphenylalkanes of differ-
ent Bronsted basicity and hydride donor reactivity, react by proton
transfer to the “large” neutral constituent, or/and by highly regios-
elective hydride abstraction from the benzylic positions of the
propylene (or oligomethylene [15]) chain. It has become obvi-
ous that the ease of proton transfer reflects the basicity of the
entire diphenylalkane molecule, rather than only of the ben-
zene ring originally bearing the tert-butyl group. In turn, the
competition between the two fragmentation channels of the
[M+H]+ ions also reflects the structure of the alkyl cation released
into the complex. For example, the complex {2.2.2[bicyclooct-
1-yl]+· · ·1,3-diphenylpropane}, generated from protonolysis of
the corresponding para-(2.2.2-bicyclooct-1-yl)-substituted 1,3-
∗
Corresponding author. Tel.: +49 521 106 2060; fax: +49 521 106 6417.
E-mail address: dietmar.kuck@uni-bielefeld.de (D. Kuck).
1387-3806/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.ijms.2010.10.007

168
D. Kuck et al. / International Journal of Mass Spectrometry 306 (2011) 167–174
H+
R2
R2
+ H+
CI(CH4)
R1
R1
[M + H]+
R2
R2
+
H
[R1]+
R1
ipso-[M + H]+
Ion/neutral complex
R1: Substituents released
into the I/N complexes:
tert-butyl (ref. [14-21]),
[2.2.2]bicyclooct-1-yl [20]
sec-butyl [21]; isobutyl,
(1-methylcyclopent)-1-yl
and cyclohexyl (this work)
The alkyl cation [R1]+
reacts as a Lewis acid
The alkyl cation [R1]+
reacts as a Bronsted acid
− (R1 − H)
− (R1 + H)
R2:
Electronically active
substituents besides H:
methyl ([16,17], this work],
methoxy, fluorine [19],
tert-butyl, isobutyl, n-butyl
(this work)]
Alkane loss
Alkene loss
by intra-complex
hydride transfer
by intra-complex
proton transfer
Scheme 1. Formation of ion/neutral complexes as intermediates of the fragmentation of metastable protonated alkyl-substituted 1,3-diphenylalkanes. A selection of sub-
stituents (R1), from which cationic constituent of the I/N complexes may be generated, and of substituents (R2) which modify the reactivity of the neutral constituent, are
mentioned.
diphenylpropane, exclusively undergoes the hydride transfer
process, as expected for this tertiary carbocation, thus excluding
the isomerization to the {2.2.2[bicyclooct-2-yl] ion within the I/N
complex [20]. Again in turn, secondary carbocations, such as sec-
and scanning the field of the second electrostatic analyzer. The
MIKE spectra are representative examples for several independent
measurements and averaged from at least ten consecutive scans.
+
C4H9 ions released from protonated para-(sec-butyl)-substituted
2.2. Synthesis
1,3-diphenylpropane and even iso-C3H7+ ions released from proto-
nated para-(isopropyl)-substituted 1,3-diphenylpropane, show far
predominant proton transfer to the neutral constituent but also
minor but significant hydride abstraction [21].
General: 1H NMR spectra (300 MHz or 500 MHz, CDCl3/TMS)
were measured on a Bruker AM 300 instrument and a Bruker DRX
500 instrument, respectively. Mass spectra were obtained with a
VG Autospec double-focusing instrument by electron ionization
(EI, 70 eV). Deuterium contents were evaluated from the EI mass
spectrometric data after correction for naturally occurring 13C. IR
spectra were recorded on a Nicolet-380 FT-IR spectrometer; solids
were measured in KBr pellets and liquid as films. Melting points
(uncorrected): Electrothermal melting point apparatus. Combus-
tional analyses: Leco CHNS-932. All distillations were performed
using a Büchi GKR 50 kugelrohr apparatus. TLC: Silica (Kieselgel
60) on aluminum foil with fluorescence indicator F254, thickness
0.2 mm (Merck).
In this contribution, we demonstrate that the fragmentation
behavior of the metastable [M+H]+ ions of diphenylalkanes may
reflect the isomerization of either the neutral constituent or the
ionic partner of the intermediate I/N complexes. Thus, not only
may the I/N complexes play a crucial role for the overall frag-
mentation, as it is well accepted in the literature [10,11]; the I/N
complex may also act as a probe for a preceding structural rear-
rangement of the initial ionic structure [8]. These insights were
obtained from CI(CH4)-MIKE spectrometric measurements per-
formed with the previously unknown hydrocarbons 1 and 6–12
depicted in Scheme 2.
Compounds: Details of the syntheses (Scheme 3) and physical
and spectroscopic properties of the compounds are collected as
Supplementary data.
2. Experimental
2.1. Mass spectrometry
3. Results and discussion
All measurements were carried out on a double-focusing instru-
ment, AutoSpec (Fisons, Manchester/UK) with a three-sector, EBE
geometry. The compounds were introduced into the CI source via
the heatable inlet rod. Methane was used as the reactant gas at
a (nominal) pressure of 4 × 10−5 ≤ p ≤ 1 × 10−4 mbar. The electron
energy was set at 70 eV, the trap current at 200 ␮A, the acceler-
ating voltage at 8000 V, and the source temperature at ca. 170 ◦C.
Fragmentation of the metastable ions in the third field-free region
was registered by selecting the precursor ion by the magnetic field
3.1. Proton-induced 1,2-tert-butyl shift
A single experimental result provided the initiative for this
study. When 1,3-bis-(4-tert-butylphenyl)propane 1 was subjected
to chemical ionization with methane as the reactant gas in a
double-focusing sector-field mass spectrometer, the metastable
ions [1+H]+ were found to eliminate isobutene through the pre-
ponderant fragmentation channel (m/z 253, 72%) and isobutane
though the minor one only (m/z 251, 28%). This is shown in the

D. Kuck et al. / International Journal of Mass Spectrometry 306 (2011) 167–174
169
1
7
9
6
8
10
CH3
11
12
D
D
D
D
CH3
D
H
11a
12a
Scheme 2. Alkyl-substituted 1,3-diphenylpropanes synthesized and studied in this work.
O
O
OHC
(a)
(b) or (c)
R'
R'
+
R'
R
R
R
15a−15d
16a−16g
17−24
(d)
1, 6−12
Comp'd
R
Comp'd R'
Compounds
R
R'
17
18
19
20
21
22
23
24
15a
15b
15c
15d
16a
1
6
7
8
tert-butyl
isobutyl
cyclohexyl
1-methyl-
tert-butyl
tert-butyl
isobutyl
isobutyl
isobutyl
4-tert-butyl
4-n-butyl
4-isobutyl
4-methyl
3-methyl
2-methyl
H
4-tert-butyl
O
16b
16c
16d
16e
16f
4-n-butyl
4-isobutyl
4-methyl
3-methyl
2-methyl
H
cyclopentyl
9
isobutyl
10
11
12
R
cyclohexyl
1-methyl-
cyclopentyl
16g
25 (R = cyclohexyl)
26 (R = 1-methylcyclopentyl)
H
(e), (f)
(g)
O
D
D
D
D
OH
Me
Ph
Me
Me
(h)
(i)
D
H
27
15d
28
11a
12a
Scheme 3. Syntheses of compounds hydrocarbons 1, 6–12 and the isogtopomers 11a and 12a. (a) KOH, MeOH/H2O; (b) H2, Pd/C, HOAc, 5 bar, 25 ◦C (see Supplementary data);
(c) H2, Pd/BaSO4, EtOH, 5 bar, 25 ◦C (see Supplementary data); (d) H2, PtO2, EtOAc, 25 ◦C; (e) LiAlD4/AlCl3, Et2O, 0 → 36 ◦C, 24 h (see Supplementary data); (f) D2, RhCl(PPh3)3,
benzene, 25 ◦C; (g) LiAlD4/AlCl3, Et2O, 0 → 36 ◦C, 70 h; (h) benzene, AlCl3, 25 → 80 ◦C; (i) AcCl, AlCl3, 1,2-dichloroethane, 0 → 25 ◦C.

170
D. Kuck et al. / International Journal of Mass Spectrometry 306 (2011) 167–174
Fig. 1. MIKE spectrum of protonated 1,3-di-(4-tert-butylphenyl)propane, [1+H]+,
generated by CI(CH4). The peak at m/z 293 is due to the methyl loss from the isobaric
ions [13C1]-1+•
.
Fig. 2. MIKE spectrum of protonated 1-(4-cyclohexylphenyl)-3-phenylpropane,
[11+H]+, generated by CI(CH4). The minor peaks at m/z 117, 160 and 175 are due to
the fragmentation of the isobaric ions [13C1]-11+•. The latter ions also contribute in
minor amounts (ca. 4%) to the peak at m/z 197.
mass-analyzed ion kinetic energy (MIKE) spectrum (Fig. 1). The
+
direct release of C4H9 ions was minute only. The branching ratio
[M+H–C4H10]+:[M+H–C4H8]+ = 0.39 was far different from what
should be the expected value for an I/N complex consisting of a
tautomeric ions [13+H]+ from ions [1+H]+ should be exothermic,
as should be the subsequent isomerization of ions [13+H]+ to the
even more stable tautomers of the protonated meta,meta-di-tert-
butyl isomer [14+H]+, such as [14+H]m+. In particular, owing to the
+
tert-C4H9 ion and a para-alkyl-substituted 1,3-diphenylpropane.
The reasons are given in the following.
+
In an early study the role of ion/neutral complexes as
reactive intermediates in the fragmentation of protonated tert-
butylbenzenes bearing an ␻-phenylalkyl group, we found that
the MIKE spectra of the three isomeric protonated 1-para-
tert-butylphenyl)-3-tolylpropanes reflect the different gas-phase
basicities of the neutral 1-phenyl-3-tolylpropane constituent of
the I/N complexes (Scheme 4) [16,17]. Whereas the para-tolyl
isomer, [2+H]+, undergoes exclusive elimination of isobutane
by hydride abstraction from both of the benzylic methylene
groups, the meta-tolyl isomer, [3+H]+, exhibits predominant loss
presence of a meta-dialkylbenzene unit, both ions [13+H]p and
+
+
[14+H]m should be more stable than the isomeric ion [13+H]m
.
Thus, the formation of the reactive complex IN(m-tBu) and its frag-
ments should be favourable due to the intermediacy of relatively
stable meta-dialkylbenzenium ions.
To account for the increased size of the C4-residue, when com-
paring the fragmentation of the twofold para-alkyl-substituted ions
[1+H]+ and [2+H]+, we also synthesized the isomer 6 contain-
ing an n-butyl substituent in the para position. As expected, the
metastable ions [6+H]+ suffer far predominant loss of isobutene,
whereas isobutene loss is of minor importance only (Table 1). It is
obvious that the n-butyl group is retained in the para-position of
the “remote” benzene ring, probably giving rise to a small increase
of the overall gas-phase basicity of the neutral constituent of the
I/N complex, as may be inferred from the proton affinity (PA) and
gas-phase basicity scale of the simple alkylbenzene homologues
[34,35]. In this particular case, it is noted that GB(n-butylbenzene)
exceeds GB(toluene) by ca. 8 kJ mol−1 [34,35].
+
of isobutene by proton transfer from the tert-C4H9 ion to the
neutral hydrocarbon. The ortho-tolyl isomer, [4+H]+, represents
an intermediate case with predominant hydride abstraction and
minor proton transfer. Clearly, this reflects the well-known order
gas-phase basicity of underlying dimethylbenzenes: GB(para-
xylene) < GB(ortho-xylene) < GB(meta-xylene) [32–37]. When the
ortho-methyl substituent was introduced – by synthesis design
(vide supra) – at the same benzene ring as the tert-butyl group,
cf. [5+H]+, the fragmentation was found to be the same as with
its “ring isomer” [4+H]+. This and other findings revealed that the
basicity of the whole 1,3-diarylalkane neutral, rather than that of
the aromatic ring originally bearing the tert-butyl group, governs
the propensity of the I/N complex to transfer a proton. In addition,
the electronic influence of the methyl (or other) electron-releasing
substituents on the hydride abstraction channel amplify the overall
effect [17,19].
On the basis of these arguments, the predominant isobutene
loss from metastable ions [1+H]+ points to (“hidden”) formation
of the isomerized I/N complex IN(m-tBu), rather than the putative
complex IN(p-tBu), as the reactive intermediate from which the
loss proper of the C4H8 and C4H10 fragments occurs (Scheme 5).
The meta-tert-butyl substituent in IN(m-tBu) should increase the
basicity of the 1,3-diphenylpropane constituent with the complex
to a similar extent as does the meta-methyl group in the com-
plex formed upon fragmentation of ions [3+H]+. In the same time,
the hydride donor ability of the 3-CH2 group should be decreased
as compared to the para-tert-butyl isomer. It appears that the
+
release of the tert-C4H9 ion from the two (equivalent) para-
Fig. 3. MIKE spectrum of protonated 1-[1-methylcyclopentyl)phenyl]-3-
phenylpropane, [12+H]+, generated by CI(CH4). The peaks at m/z 117, 160,
175 and 264 are due to the fragmentation of the isobaric ions [13C1]-12+•. The latter
ions also contribute (ca. 20%) to the peak at m/z 197.
+
protonated tautomers of ions [1+H]p into the complex IN(p-tBu)
requires more energy than the 1,2-C shift leading to the protonated
meta,para-di-tert-butyl isomer [13+H]m+. In fact, the formation of

D. Kuck et al. / International Journal of Mass Spectrometry 306 (2011) 167–174
171
+ H+
+ H+
CH3
CH3
[2 + H]+
[3 + H]+
isobutane loss 100%
isobutene loss 0%
isobutane loss 35%
isobutene loss 65%
+ H+
+ H+
CH3
CH3
[5 + H]+
[4 + H]+
isobutane loss 87%
isobutene loss 13%
isobutane loss 87%
isobutene loss 13%
Scheme 4. Loss of isobutane vs. loss of isobutene from the metastable protonated 1-(4-tert-butylphenyl)-3-tolylpropanes, [2+H]+, [3+H]+ and [4+H]+, and the “reversed”
ortho-methyl isomer, [5+H]+ (from Ref. [16]).
3.2. Protonated isobutyl isomers
neutral. Whereas the loss of n-alkyl groups as the correspond-
ing alkanes can be excluded (as assumed for the case of [6+H]+,
for example), ␤-branched primary alkyl groups could undergo
1,2-hydride shifts under the protonolysis conditions of CI(CH4)-
MIKE spectrometry applied here. The typical loss of isobutane from
metastable protonated bis-1,3-(4-isobutylphenyl)propane, [7+H]+,
would be a reliable probe for the presence of tert-C4H9+ as the ionic
component of an I/N-complex intermediate. Besides the diisobutyl
isomer 7 of the di-tert-butyl compound 1, we studied the three
isobutyl- and methyl-substituted hydrocarbons 8, 9 and 10 as the
respective isomers of 2, 3 and 4.
Incipient isobutyl cations formed upon protonolysis of isobutyl-
arenes in the gas phase can easily isomerize to tert-butyl and even
sec-butyl cations [25,26]. Therefore, we envisioned the possibility
that the suspected rearrangement would become evident through
the elimination of isobutane along with, or even instead of, the
loss of isobutene from the [M+H]+ ions of isobutyl-substituted 1,3-
diphenylpropanes. In a previous study [21], we found that not only
tertiary but also secondary alkyl cations, such as sec-C4H9+ and sec-
C3H7+, undergo hydride abstraction from the 1,3-diphenylalkane
(b)
(a)
H
+
+
+
H
H
+
+
+
[1 + H]p
[13 + H]p
[13 + H]m
(a)
H
+
+
+
[14 + H]m
+
putative I/N complex IN(p-tBu)
reactive ("isomerized")
I/N complex IN(m-tBu)
− C6H5-(CH2)3-C6H4-C4H9
(> 1%)
> 95 : 5
28 : 72
(expected)
(observed)
+
H
H
m/z 57
Scheme 5. Isomerization of ions [1+H]+ via the para-protonated tautomer, [1+H]p+, to the meta,para- and meta,meta-isomers [13+H]+ and [14+H]m+, respectively, from which
the reactive “isomerized” ion/neutral complex IN(m-tBu) is formed.

172
D. Kuck et al. / International Journal of Mass Spectrometry 306 (2011) 167–174
Table 1
Relative abundancesa of the fragment ions generated by hydride abstraction (alkane loss) and by proton transfer (alkene loss) from the metastable [M+H]+ ions of diphenyl-
propanes 1–4 and 6–12 (MIKE spectra).
[M+H]+ m/z
Substituents R and Rꢀ in the [M+H]+ ions
Alkane loss m/z (%)c
Alkene loss m/z (%)c
Other fragment ions, m/z (%)
+
[1+H]+ 309
[2+H]+b 267
[3+H]+b 267
[4+H]+b 267
[6+H]+ 309
[7+H]+ 309
[8+H]+ 267
[9+H]+ 267
[10+H]+ 267
[11+H]+ 279
[12+H]+ 279
4-tert-C4H9, 4ꢀ-tert-C4H9
4-tert-C4H9, 4ꢀ-CH3
4-tert-C4H9, 3ꢀ-CH3
4-tert-C4H9, 2ꢀ-CH3
4-tert-C4H9, 4ꢀ-n-C4H9
4-iso-C4H9, 4ꢀ-iso-C4H9
4-iso-C4H9, 4ꢀ-CH3
28
>99.5
35
87
92
2.5
2.5
1.5
1.5
20
72.
<0.5
65
13
8
97.5
97.5
98.5
98.5
80
C4H9
C4H9
C4H9
C4H9
C4H9
C4H9
C4H9
C4H9
C4H9
≈0.5d
≈0.5d
≈0.5d
≈0.5d
≤0.5d
3.5e
+
+
+
+
+
+
+
+
8.0 e
8.0e
4-iso-C4H9, 3ꢀ-CH3
4-iso-C4H9, 2ꢀ-CH3
4-cyclo-C6H11, H
4-(1-CH3-cyclo-C5H8), H
8.0e
+
+
C6H11
C6H11
2.0d
72
28
35d,f
a
Data as obtained after correction for the fragmentation of naturally occurring ions [13C1]-M+•
Data taken from Ref. [16].
Data given as % of the sum of alkane and alkene losses.
Abundances given as relative to the sum of alkane and alkene losses.
Relative abundances given as relative to those of alkane loss only.
The predominating fragment ions are due to the losses of C3H6, C4H8 and C5H10 (see Fig. 3).
.
b
c
d
e
f
The results are straightforward. In all cases, the metastable
the diphenylpropanes 11 and 12 bearing a cyclohexyl or a tert-
methylcyclopentyl group, respectively, in one of the para-positions.
Parts of the synthesis of these hydrocarbons as well as of their
deuterium-labelled analogues 11a and 12a were not straightfor-
ward (see Supplementary data).
[M+H]+ ions, generated under the same conditions as in the
previous measurements, undergo almost exclusive loss of C4H8
(presumably isobutene, Table 1). Only about 2% of the fragmen-
tation occurs through elimination of isobutane. Hence, there is no
hint to a regiospecific effect of the methyl groups on the basic-
ity of the 1-phenyl-3-tolylpropane neutral in the putative I/N
complex (Scheme 1), if formed at all. Rather, it can be assumed
that, in analogy to the fragmentation of simple long-lived alkyl-
benzenium ions [25,26], the protonolysis of isobutyl-substituted
1,3-diphenylpropanes like 7–10 occurs via a considerable activa-
tion barrier, with the proton being transferred to the benzene ring
originally bearing the isobutyl group. Nevertheless, a minor frac-
The MIKE spectrum of ions [11+H]+ (Fig. 2) was found to be
dominated by the loss of C6H10 (most probably cyclohexene) but
it also exhibited the elimination of C6H12 (cyclohexane) as well as
+
the formation of free C6H11 ions (m/z 83, ca. 2% only). By con-
trast, the MIKE spectrum of the isomeric ions [12+H]+ (Fig. 3) was
much more complex: these metastable ions eliminate several small
neutrals, viz. C3H6, C4H8 and C5H10, in large relative abundances,
showing that, in fact, protonolysis gives rise to deep-seated skele-
tal rearrangements of the alicyclic residue. Interestingly, however,
tion of the [M+H]+ ions does form C4H9 cations which, however,
+
+
+
appear as free (probably tertiary) butyl cations (m/z 57, Table 1). In
minor fractions of [M+H–C6H10] (m/z 197) and [M+H–C6H12]
the case of ions [7+H]+, a ratio {C4H9 }:{[M+H–C4H10]+} ≈ 1.3 was
(m/z 195) ions are also formed, with the latter ones predominat-
ing. This points to the expected losses of methylcyclopentene (or
methylenecyclopentane) and methylcyclopentane by proton trans-
+
found, and for the methyl-substituted isomers [8+H]+, [9+H]+ and
[10+H]+, free C4H9 was generated in even larger relative abun-
+
dances, {C4H9 }:{[M+H–C4H10]+} = 4–5. By contrast, the release
fer and hydride abstraction, respectively. In addition, free C6H11
+
+
+
of free C4H9 ions is negligible in the case of the tert-butyl ana-
ions (m/z 83) are formed in noticeable amounts. Thus, in both cases,
the cycloalkyl cations do not react as specific as expected for sec-
ondary (cyclohexyl) and tertiary (methylcyclopentyl) carbocations.
logues (Table 1 and Scheme 5). Therefore, it appears that any
I/N complexes formed after the proton-induced (isobutyl → tert-
butyl) rearrangement are too-highly activated and short-lived
to undergo the intra-complex hydride abstraction; instead, they
Rather, the putative cyclo-C6H11+ ions appear to undergo ring con-
+
traction to CH3-cyclo-C5H8 ions, as indicated by the loss of C6H12
.
+
directly release the tert-C4H9 ions.
In turn, the latter carbenium ions, which should be released into
the I/N complex with 1,3-diphenylpropane at least as easily as tert-
+
C4H9+, seemingly isomerize to secondary C6H11 structures, the
+
3.3. Intra-complex C6H11 ions from cyclohexyl and
+
cyclo-C6H11 ions being only one of the possible isomers. On the
methylcyclopentyl derivatives
whole, the situation is considerably more complicated than with
the [M+H]+ containing (acyclic) alkyl groups and the propensity of
cyclo- or bicycloalkyl cations to undergo skeletal rearrangement
is confirmed. Nevertheless, the fact that hydride abstraction does
occur to a minor extent with [11+H]+ ions and to a (relatively) large
extent with [12+H]+ ions clearly reflects the participation of tertiary
A previous study within this series indicated that more com-
plex carbenium ions released by protonolysis into an I/N complex
with 1,3-diphenylpropane may, at least in part, undergo skeletal
rearrangement prior fragmentation by alkane and/or alkene loss.
In particular, the [2.2.2]bicyclooct-1-yl group was found to lose
its symmetrical bicyclic framework under the CI(CH4) conditions,
although intra-complex bicyclo[2.2.2]octyl ions, CH(CH2CH2)3C+,
released by protonolysis, react exclusively by hydride abstrac-
tion, as expected for such tertiary, bridgehead cations [20]. On the
other hand, as mentioned above, secondary alkyl cations, such as
sec-C4H9+ and sec-C3H7+, react almost exclusively by proton trans-
fer but with a minor contribution of the regioselective hydride
abstraction from the 1,3-diphenylpropane constituent of the I/N
complexes [21].
+
C6H11 ions, possibly in an incomplete equilibrium within the I/N
complexes with the 1,3-diphenylpropane neutral (Scheme 6).
In this view, it is interesting to note that the deu-
terium labelling experiments again confirm the common behav-
ior of the hydride-abstracting ions in the I/N complexes
[R+· · ·diphenylalkane], found already in many previous cases:
the MIKE spectrum of the methylcyclopentyl-substituted iso-
topomer [12a+H]+ bearing an ␣,␣-dideuterated methylene group
exhibits the typical ratio of H− and D− abstraction, viz.
[12a+H–C6H12]:[12a+H–C6H11D] = 1.7 ( 0.1) (Fig. 4b). Although
the respective complementary labelling has not been performed
here, it can be safely assumed that the methylcyclopentyl ions
Gaseous cyclohexyl cations are known to undergo skele-
tal rearrangement to the (thermochemically more stable) 1-
methylcyclopentyl cations [38–44]. Therefore, we synthesized

D. Kuck et al. / International Journal of Mass Spectrometry 306 (2011) 167–174
173
[11 + H]+
[12 + H]+
+
+
IN(Chx)
IN(MCp)
− C6H10
− C6H12
+ H+
α (γ)
+
m/z 195
m/z 197
Scheme 6. Isomerizing ion/neutral complexes IN(Chx) and IN(MCp) generated from the protonated precursors 11 and 12. In both cases, the hydride abstraction occurs from
both the ␣-CH2 and the ␥-CH2 groups of the diphenylpropane neutral.
Fig. 4. Partial MIKE spectra of protonated isotopomers [11a+H]+ (a) and [12a+H]+ (b). To minor extents, the peaks at m/z 200 and at m/z 199, respectively, indicating the loss
of C6H10 (H+ transfer channel), are due to the fragmentation of the corresponding isobaric ions [13C1]-11a+• and [13C1]-12a+• (see text).
within the I/N complex [C6H11 · · ·C6H5C␣H2CH2C␥H2C6H5] react
with the usual and characteristic regioselectivity, k␥/k␣ = 1, at both
of the benzylic methylene groups, and with the typical kinetic
isotope effect observed for these hydrocarbon ions, kH/kD ≈ 1.6
cases, the characteristic fragmentation behavior of the intermedi-
ate I/N complexes can also reflect the isomerization of the ionic
constituent of an intermediate I/N complex.
+
+
[14–21]. This holds true even for the C6H11 ions undergoing
4. Conclusion
the hydride abstraction in the I/N complex formed from ions
[11+H]+. Although the fractions of cycloalkane loss are really
minor here, the MIKE spectrum of the corresponding protonated
(␣,␣,␤-trideuterated) isotopomer, [11a+H]+, reflects about the
same ratio: [11a+H–C6H12]:[11a+H–C6H11D] = 1.6 ( 0.15) (Fig. 4a).
Thus, it is very likely that, in fact, tert-methylcyclopentyl ions are
formed upon gas-phase protonolysis of the cyclohexyl derivative 11
(Scheme 6). These results demonstrate that, at least in favourable
Long-lived, metastable [M+H]+ ions of 1,3-diphenylpropanes
bearing branched alkyl groups (R, Rꢀ) at one or both of the aromatic
rings eliminate these groups mostly as alkenes (by proton trans-
fer) and alkanes (by hydride transfer). These processes involve the
ion/neutral complexes [R+· · ·C6H5CH2CH2CH2C6H4–Rꢀ] as reactive
intermediates. In some of the examples presented in this work, the
proton-induced isomerization of the constituents of these I/N com-

174
D. Kuck et al. / International Journal of Mass Spectrometry 306 (2011) 167–174
plexes is reflected by the relative rates of the alkene and alkane
losses. Thus, the 1,2-shift of a tert-butyl group from the para- to a
meta-position of the protonated benzene ring gives rise to a char-
acteristic increase of the alkene loss due to the increase of the
gas-phase basicity of the neutral component of the I/N complex,
together with a decreased alkane loss. Isobutyl-substituted iso-
mers behave different from the tert-butyl-substituted [M+H]+ ions;
[10] T.H. Morton, in: P.B. Armentrout (Ed.), Encyclopedia of Mass Spectrometry, vol.
1, Elsevier, Amsterdam, 2003, pp. 467–479 (Chapter 6).
[11] T.H. Morton, in: N.M.M. Nibbering (Ed.), Encyclopedia of Mass Spectrometry,
vol. 4, Elsevier, Amsterdam, 2005, pp. 165–173 (Chapter 2).
[12] J. Bialecki, J. Ruzicka, A.B. Attygalle, J. Mass Spectrom. 41 (2006) 1195.
[13] Y. Chai, K. Jiang, Y. Pan, J. Mass Spectrom. 45 (2010) 496.
[14] D. Kuck, C. Matthias, J. Am. Chem. Soc. 114 (1992) 1901.
[15] C. Matthias, D. Kuck, Org. Mass Spectrom. 28 (1993) 1073.
[16] C. Matthias, K. Weniger, D. Kuck, Eur. Mass Spectrom. 1 (1995) 445.
[17] C. Matthias, S. Anlauf, K. Weniger, D. Kuck, Int. J. Mass Spectrom. 199 (2000)
155.
[18] C. Matthias, D. Kuck, Int. J. Mass Spectrom. 217 (2002) 131.
[19] C. Matthias, A. Cartoni, D. Kuck, Int. J. Mass Spectrom. 255–256 (2006) 195.
[20] C. Matthias, B. Bredenkötter, D. Kuck, Int. J. Mass Spectrom. 228 (2003) 321.
[21] C. Matthias, D. Kuck, Croat. Chem. Acta 82 (2009) 7.
[22] D. Kuck, U. Filges, Org. Mass Spectrom. 23 (1988) 643.
[23] H.E. Audier, F. Dahhani, A. Milliet, D. Kuck, J. Chem. Soc., Chem. Commun. (1997)
429.
[24] H.E. Audier, C. Monteiro, P. Mourgues, D. Berthomieu, Org. Mass Spectrom. 25
(1990) 245.
[25] J.P. Denhez, H.E. Audier, D. Berthomieu, Rapid Commun. Mass Spectrom. 9
(1995) 1210.
+
they do not form I/N complexes containing tert-C4H9 ions which
would reflect the basicity of the neutral constituent. Rather, iso-
+
merization to the more acidic sec-C4H9 ions should occur. The
competing losses of cycloalkene(s) and cycloalkane(s) from the
metastable [M+H]+ ions of cyclohexyl- and (methylcyclopentyl)-
substituted 1,3-diphenylpropanes reflects the at least partial
isomerization of the ionic constituents of the correspond-
ing I/N complexes, [(CH2)5 > CH+· · ·C6H5CH2CH2CH2C6H5] and
[(CH2)4 > C+CH3· · ·C6H5CH2CH2CH2C6H5]. Thus, in suitably tailored
cases, the fragmentation of organic ions via reactive ion/neutral
complexes can reflect the structure–reactivity relation of the ionic
and/or the neutral constituents of these intermediates.
[26] J.P. Denhez, H.E. Audier, D. Berthomieu, Org. Mass Spectrom. 28 (1993) 704.
[27] D. Berthomieu, V. Brenner, G. Ohanessian, J.P. Denhez, P. Millie, H.E. Audier, J.
Am. Chem. Soc. 115 (1993) 2505.
[28] D.J. McAdoo, C.E. Hudson, Int. J. Mass Spectrom. Ion Processes 88 (1989) 133.
[29] E.L. Chronister, T.H. Morton, J. Am. Chem. Soc. 112 (1990) 133.
[30] T.A. Molenaar-Langeveld, C. Gremmen, S. Ingemann, N.M.M. Nibbering, Int. J.
Mass Spectrom. 199 (2000) 1.
[31] C. Denekamp, A. Stanger, J. Chem. Soc., Chem. Commun. (2002) 236.
[32] J.L. Devlin III, J.F. Wolf, R.W. Taft, W.J. Hehre, J. Am. Chem. Soc. 98 (1976) 1990.
[33] D.A. Aue, M.T. Bowers, in: M.T. Bowers (Ed.), Gas Phase Ion Chemistry, vol. 2,
Academic Press, New York, San Francisco, 1979, pp. 1–51.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.ijms.2010.10.007.
References
[34] E.P.L. Hunter, S.G. Lias, J. Phys. Chem. Ref. Data 27 (1998) 413.
[35] NIST Standard Reference Database Number 69, March 2005 Release,
National Institute of Standards and Technology, Gaithersburg, MD 20899
(http://webbook.nist.gov/chemistry).
[1] P. Longevialle, Mass Spectrom. Rev. 11 (1992) 157.
[2] T.H. Morton, Tetrahedron 38 (1982) 3195.
[36] M. Eckert-Maksic´, M. Klessinger, Z.B. Maksic´, Chem. Eur. J. 2 (1996) 1251.
[37] Z.B. Maksic´, B. Kovacˇevic´, D. Kovacˇek, J. Phys. Chem. A 101 (1997) 7446.
[38] C. Wesdemiotis, R. Wolfschütz, H. Schwarz, Tetrahedron 36 (1980) 275.
[39] M. Attina, F. Cacace, P. Giacomello, J. Am. Chem. Soc. 103 (1981) 4711.
[40] M. Attina, F. Cacace, R. Cipollini, M. Speranza, J. Am. Chem. Soc. 107 (1985) 4824.
[41] M. Speranza, Int. J. Mass Spectrom. Ion Processes 118–119 (1992) 395.
[42] I.D. Mackie, J. Govindhakannan, G.A. DiLabio, J. Phys. Chem. A 112 (2008) 4004.
[43] I.D. Mackie, J. Govindhakannan, J. Mol. Struct., THEOCHEM 939 (2010) 53.
[44] K. Shimizu, T. Sunagawa, K. Arimura, H. Hattori, Catal. Lett. 63 (1999) 185.
[3] T.H. Morton, Org. Mass Spectrom. 27 (1992) 353.
[4] R.D. Bowen, Acc. Chem. Res. 24 (1991) 364.
[5] D.J. McAdoo, Mass Spectrom. Rev. 7 (1988) 363.
[6] D.J. McAdoo, T.H. Morton, Acc. Chem. Res. 26 (1993) 295.
[7] G. Thielking, U. Filges, H.F. Grützmacher, J. Am. Soc. Mass Spectrom. 3 (1992)
417.
[8] D. Berthomieu, V. Brenner, G. Ohanessian, J.P. Denhez, P. Millié, H.E. Audier, J.
Phys. Chem. 99 (1995) 712.
[9] A.G. Harrison, J.Y. Wang, Int. J. Mass Spectrom. Ion Processes 160 (1997) 157.