Relative Reactivities of Three Isomeric Aromatic Biradicals with a 1,4-Biradical Topology Are Controlled by Polar Effects

Article abstract of DOI:10.1002/chem.201806106

Unexpectedly, the 5-dehydroquinoline radical cation was formed in the gas phase from the 5-iodo-8-nitroquinolinium cation upon ion-trap collision-activated dissociation. This reaction involves the cleavage of a nitro group to generate an intermediate monoradical, namely, the 8-dehydro-5-iodoquinolinium cation, followed by rearrangement through abstraction of a hydrogen atom from the protonated nitrogen atom by the radical site. Dissociation of the rearranged radical cation through elimination of an iodine atom generates the 5-dehydroquinoline radical cation. The mechanism was probed by studying isomeric biradicals and performing quantum chemical calculations. The 5-dehydroquinoline radical cation showed greater gas-phase reactivity toward dimethyl disulfide, cyclohexane, and allyl iodide than the isomeric 5,8-didehydroquinolinium cation, which is more reactive than the isomeric 5,8-didehydroisoquinolinium cation studied previously. All three isomers have a 1,4-biradical topology. The order of reactivity is rationalized by the vertical electron affinities of the radical sites of these biradicals instead of their widely differing singlet–triplet splittings.

Full text of DOI:10.1002/chem.201806106

A Journal of
Accepted Article
Title: Relative Reactivities of Three Isomeric Aromatic Biradicals with a
1,4-Biradical Topology Are Controlled by Polar Effects
Authors: Xin Ma, Chunfen Jin, Duanda Wang, John J. Nash, and
Hilkka Kenttamaa
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201806106
Link to VoR: http://dx.doi.org/10.1002/chem.201806106
Supported by

Chemistry - A European Journal
10.1002/chem.201806106
Relative Reactivities of Three Isomeric Aromatic
Biradicals with a 1,4-Biradical Topology Are
Controlled by Polar Effects
Xin Ma, Chunfen Jin, Duanda Wang, John J. Nash, and
Hilkka I. Kenttämaa*
Department of Chemistry, Purdue University, 560 Oval Drive,
West Lafayette, Indiana, 47907
Submitted to Chemistry – A European Journal
*
Corresponding author: Professor Hilkka I. Kenttämaa, Email: hilkka@purdue.edu,
Telephone: +1 (765)-494-0882
Supporting information for this article is given via a link at the end of the document.
1
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
Abstract
Unexpectedly, the 5-dehydroquinoline radical cation was formed in the gas phase from
the 5-iodo-8-nitroquinolinium cation upon ion-trap collision-activated dissociation.
This reaction involves the cleavage of a nitro group to generate an intermediate
monoradical, the 8-dehydro-5-iodoquinolinium cation, followed by rearrangement via
abstraction of a hydrogen atom from the protonated nitrogen atom by the radical site.
Dissociation of the rearranged radical cation via elimination of an iodine atom generates
the 5-dehydroquinoline radical cation. The mechanism was probed by studying
isomeric biradicals and via quantum chemical calculations. The 5-dehydroquinoline
radical cation showed greater gas-phase reactivity toward dimethyl disulfide,
cyclohexane and allyl iodide than the isomeric 5,8-didehydroquinolinium cation, which
is more reactive than the isomeric 5,8-didehydroisoquinolinium cation studied
previously, all with a 1,4-biradical topology. The reactivity ordering is rationalized by
the vertical electron affinities of the radical sites of these biradicals instead of their
widely different singlet-triplet splittings.
2
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
Introduction
Aromatic carbon-centered σ-type biradicals are reaction intermediates that are involved
1
–3
4
in various biochemical and chemical processes. para-Benzynes, in particular, have
drawn much attention due to their roles as the biologically-active intermediates of some
5
–9
enediyne anticancer drugs. However, understanding of the fundamental chemical
properties of para-benzynes and the factors that control their reactivity is limited.
n
Therefore, tandem mass spectrometry (MS ) has been used in the past to generate such
biradicals and related species with a chemically-inert charge site for mass spectrometric
1
0–13
manipulation.
Previously studied para-benzynes, including the 2,5-
1
2,13
13
didehydropyridinium
cation and the 5,8-didehydroisoquinolinium cation, showed
surprisingly low reactivity. In order to explore this issue further, generation of the 5,8-
didehydroquinolinium cation from the 5-iodo-8-nitroquinolinium cation precursor via
two collision-activated dissociation (CAD) steps was attempted in this study (Scheme
1
). However, an entirely unexpected isomer was generated. The identity of this isomer,
the rearrangement mechanism leading to its formation, and a comparison of its
reactivity to that of the 5,8-didehydroisoquinolinium cation (generated from a different
precursor) and the 5,8-didehydroquinolinium cation is described here.
Scheme 1. Attempted generation of the 5,8-didehydroquinolinium cation.
Results and Discussion
Biradical Reactivity
The charged biradical generated upon two consecutive CAD events in the linear
quadrupole ion trap (i.e., upon ion-trap CAD or ITCAD) from 1 (Scheme 1) yielded
diagnostic product ions that demonstrated the existence of two radical sites.14,15 For
example, it reacted with dimethyl disulfide predominantly via abstraction of two SCH
3
groups and an HSCH
molecule, which is typical for related biradicals.1 Similarly, it
6,17
3
generated the expected reaction products upon interactions with allyl iodide and
cyclohexane. However, it displayed unusually high reactivity toward all the reagents
studied compared to the reactivity previously reported for the 5,8-
1
3
didehydroisoquinolinium cation. As the previous study used in-source CAD (ISCAD)
3
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
instead of ITCAD to generate the 5,8-didehydroisoquinolinium cation, this biradical
was regenerated here from the 5-iodo-8-nitroprecursor 4 (Figure 1) by using ITCAD to
make comparison with the unknown biradical more reliable (6; Table 1). Indeed, the
unknown biradical has reaction efficiencies of 19%, 4% and 8% for dimethyl disulfide,
cyclohexane, and allyl iodide, while the 5,8-didehydroisoquinolinium cation (6) shows
much lower efficiencies (1%, 0.1% and 0.2%, respectively; Table 1). This finding
suggests that the unknown biradical does not have the expected structure.
In order to explore this result further, the authentic 5,8-didehydroquinolinium cation
was generated from two different precursors, 5-nitro-8-iodo- and 5,8-
diiodoquinolinium cations (2 and 3, Figure 1). Similar reactivity observed for the
product ion generated from these two different precursors supports the expectation that
authentic 5,8-didehydroquinolinum ion (5) had been generated in both cases (Table 1).
The 5,8-didehydroquinolinium cation showed lower reactivity than the unknown
biradical toward dimethyl disulfide (Figure 2) but greater reactivity than the 5,8-
didehydroisoquinolinium cation (6, Table 1, for the reaction mass spectra of the radicals
with cyclohexane and allyl iodide, see Supporting Information). These reactivity
differences suggest that the unknown biradical does not have the structure 5.
Figure 1. Four protonated biradical precursors.
HSCH abs
3
1
abs
75
76
biradical
(
a)
1
00
2
x SCH abs
3
SCH
3
1
28
5
0
222
1
1
00
(
(
b)
c)
5
0
1
76
222
222
100
5
0
1
76
100
(
d)
50
0
100
120
140
160
180
m/z
200
220
240
Figure 2. Mass spectra measured after 1000 ms reaction with dimethyl disulfide for the
biradicals generated from precursors 1 – 4 (a – d, respectively).
4
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
Table 1. Reactions, Reaction Efficiencies and Primary Product Branching Ratios for Gas-phase Reactions of Dimethyl Disulfide,
Cyclohexane and Allyl Iodide With the Biradicals Generated from 1 – 4 Upon ITCAD and their Calculated Vertical Electron Affinities
(
EA
v
) and Singlet-Triplet Splittings (ΔES-T
)
EA
v
(eV) a
5.59
-3.8
5.59
-3.8
5.15
-4.8
-1 b
ΔES-T (kcal mol )
SCH
3
abs.c 68%
SCH
2° SCH
HSCH abs. 33%
3
abs.
67%
SCH
2° SCH
HSCH abs.
3
abs.
70%
30%
SCH
2° SCH
HSCH abs.
3
abs.
58%
42%
2
°SCH
3
abs.d
3
abs
3
abs.
3
abs.
HSCH abs. 32%
3
3
3
3
1
7% URI e
Efficiency = 19%
92% URI
Efficiency = 4%
69% URI
Efficiency = 8%
70% URI
Efficiency = 1%
5
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
2
x H abs.
60%
24%
2 x H abs.
Addition
69%
21%
2 x H abs.
Addition
61%
23%
2 x H abs.
Addition
55%
33%
Addition
-
-
-
-
H abs.
16%
H abs.
10%
H abs.
16%
H abs.
12%
1
6% URI
97% URI
47% URI
95% URI
Efficiency = 4%
Efficiency = 2%
Efficiency = 1%
Efficiency = 0.1%
Allyl abs.
° I abs.
Allyl-H abs. 26%
I abs. 1%
73%
Allyl abs.
2° I abs.
Allyl-H abs. 15%
I abs.
2° I abs.
80%
Allyl abs.
2° I abs.
Allyl-H abs.
I abs.
81%
Allyl abs.
2° I abs.
I abs.
77%
23%
2
13%
6%
5%
2° I abs.
2
° I abs.
2° I abs.
2
8% URI
92% URI
Efficiency = 4%
93% URI
Efficiency = 5%
96% URI
Efficiency = 0.2%
Efficiency = 8%
a
b
All EA
v
values were calculated at the CASPT2/CASSCF(13,12)/cc-pVTZ//CASSCF(12,12)/cc-pVTZ level of theory. All S-T splittings were calculated at the
CASPT2/CASSCF(12,12)/cc-pVTZ//CASSCF(12,12)/cc-pVTZ level of theory. Corrected for zero-point vibrational energy differences at 298K by using the
c
d
º
e
UB3LYP/cc-pVTZ frequencies. abs. = abstraction. 2 indicates a secondary product and it is listed after the primary product that generated it. URI = unreactive
isomer; the CAD processes used to generate these biradicals also yield isomeric enediynes that are unreactive (ref. 13b).
6
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
In order to conclusively demonstrate that the unknown biradical does not have structure
, the product formed upon abstraction of two SCH groups from dimethyl disulfide
5
3
was isolated and subjected to ITCAD. The fragmentation behavior was compared to
that of the authentic 5,8-dithiomethylquinolinium ion (m/z 222) generated from a
synthesized precursor. The fragmentation patterns were found to be distinctly different
(
Figure 3). This finding unambiguously demonstrates that the unknown biradical does
not have structure 5.
1
28
100
(
(
a)
b)
8
0
0
0
0
2
22
6
4
2
-
2 x SCH
3
1
62
175
207
100
8
0
0
0
0
-
CH
3
6
4
2
2
22
175 -SCH3
0
100
120
140
160
180
m/z
200
220
240
Figure 3. ITCAD mass spectra obtained using the same collision energy (20, arbitrary
units) for (a) the product ion of m/z 222 formed upon abstraction of two SCH groups
3
from dimethyl disulfide by the biradical generated from 1 and (b) the authentic 5,8-
dithiomethylquinolinium cation.
Besides ITCAD, ISCAD was also used to generate the unknown biradical of interest.
ISCAD takes place in the medium pressure free-jet expansion region immediately after
the MS orifice. In ISCAD, ions are accelerated by an electric field, collide with neutral
1
8
molecules present (mostly from air), and fragment. This method differs from ITCAD
in three important ways. First, the collision gas molecules are larger (mostly molecules
from air) for ISCAD than the helium atoms in ITCAD, which results in greater internal
energies deposited into the ions upon ISCAD. Second, the ion kinetic energies are
greater upon ISCAD than in ITCAD, which again results in greater excitation. For
example, earlier studies on para-benzynes and enediynes in the gas phase suggested
that ISCAD can be used to homolytically cleave both C-I and C-NO bonds nearly
2
1
9
simultaneously, which is not the case for ITCAD. Third, ion activation in ISCAD only
involves a few activating collisions while ions undergo a very large number of
collisions upon CAD ion traps, each collision depositing only a small amount of internal
energy into the ions, until the activation threshold has been reached. This process often
1
3b
leads to rearrangement reactions. Therefore, ISCAD was used in this study to test
whether the formation of the highly reactive, unknown biradical requires several
activation steps as in ITCAD or whether it can be generated in just a few steps by using
7
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
ISCAD. The biradical generated by ISCAD was found to have low reactivity toward
dimethyl disulfide, as expected if it had a para-benzyne structure (Figure 4).
Furthermore, its reaction efficiency was similar to that of 5 (7%; 6% measured for the
authentic biradical; Table 1). Similar observations were made for cyclohexane and allyl
iodide reagents (see Supporting Information). These findings suggest that a nearly
simultaneous cleavage of both C-I and C-NO
,8-didehydroquinolinium cation (5) while the several consecutive activation steps
utilized in ITCAD yield a rearranged product.
2
bonds upon ISCAD of 1 generates the
5
HSCH abs
3
1
76
1
8
00
0
(
a)
biradical
1
28
6
4
2
0
0
0
2
x SCH abs
3
SCH
3
abs
75
2
22
1
1
8
00
0
(
b)
6
4
2
0
0
0
2
22
1
180
m/z
76
0
1
00
120
140
160
200
220
240
Figure 4. Reaction product mass spectra measured for the unknown biradical generated
from 1 by using (a) ITCAD and (b) ISCAD after 1000 ms reaction with dimethyl
disulfide.
These results suggest that the formation of the unknown biradical takes place after the
first ITCAD event that generates the 8-dehydro-5-iodoquinolinium cation (9) from 1
(
Scheme 1). To explore whether 9 is the only precursor which generates this reactive
biradical, another monoradical, the 5-dehydro-8-nitroquinolinium cation (10) that was
also generated upon the first ITCAD event on precursor 1 (Figure 5a), was isolated and
subjected to a second ITCAD to cleave off the nitro group. The reactivity of the so
generated biradical toward dimethyl disulfide was found to be similar to that measured
for 5 (efficiencies 5% and 6%, respectively) and much lower than that of the unknown
biradical (efficiency 19%) generated from 9 (upon elimination of an iodine atom;
Figure 6), suggesting that it is the 5,8-didehydroquinolinium cation (5). Interestingly,
when precursor 2 (8-iodo-4-nitroquinolinium cation) was employed to generate the
biradical in yet a different way, by cleavage of a nitro group from 5-nitro-8-
dehydroquinolinium cation, no I atom cleavage was observed upon the first ITCAD
event on 2 (Figure 5b). Therefore, the 5-nitro-8-dehydroquinolinium cation could not
be generated.
8
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
2
55
1
00
(
(
a)
b)
8
0
0
0
0
-
NO
2
6
2
71
-
4
2
•NO
1
74
- I•
301
100
3
01
8
0
0
0
0
-
NO
2
6
4
2
-
•NO
0
160 180 200 220 240 260 280 300 320 340 360 380 400
m/z
Figure 5. ITCAD mass spectra obtained using the same collision energy (40 arbitrary
units) for (a) 1 and (b) 2 (ions of m/z 301). Note that 9 has a higher relative abundance
than 10 in (a) because the C-NO bond is easier to break than the C-I bond in a cation.
2
HSCH abs
3
1
76
1
00
biradical
(
(
a)
b)
8
0
1
28
6
4
0
0
0
SCH
3
abs
75
2 x SCH abs
3
2
22
1
2
100
8
0
0
0
0
6
4
2
1
76
222
220
0
100
120
140
160
180
m/z
200
240
Figure 6. The product ion mass spectra measured for (a) the unknown reactive biradical
generated from 9 upon ITCAD and (b) the biradical generated from 10 upon ITCAD,
both after reactions with dimethyl disulfide for 1000 ms.
Based on these findings, monoradical cation 9 is the only precursor for the highly
reactive, unknown biradical. This may be rationalized by the vicinity of the radical site
at C-8 and the protonated nitrogen atom. The reaction mechanism proposed for the
formation of the unknown biradical involves a hydrogen atom shift from the protonated
nitrogen atom to C-8 in 9 to form the 5-iodoquinoline radical cation upon the second
ITCAD event (Scheme 2). After this rearrangement, the iodine atom is lost to generate
biradical 7 (Scheme 2).
9
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
Scheme 2. Proposed mechanism for the formation of 7 from 1 upon ITCAD.
The proposed isomerization pathway was supported by quantum chemical calculations
at the M06-2X/6-311G(d,p) level of theory. The activation enthalpy for the
rearrangement via hydrogen atom transfer (Scheme 2) was calculated to be 54.4 kcal
-
1
-1
mol , 7.4 kcal mol lower than the calculated C-I bond dissociation enthalpy. Asimilar
rearrangement cannot occur for 2 because the iodine atom is not cleaved upon the first
ITCAD event (Figure 5). On the other hand, the calculated enthalpy changes for
homolytic C-I bond cleavages in 3 (Scheme 3) suggest that the first bond cleavage
occurs at C-5, not at C-8. Therefore, 9 is not generated and rearrangement does not take
place.
Scheme 3. Calculated (M06-2X/6-311G(d,p)) enthalpy changes for two possible iodine
-1
atom loss reactions from 3 to generate 5. All values are in kcal mol and are relative to
the protonated radical precursor.
In order to provide more support for the above reaction mechanism, authentic 5-
dehydroquinoline radical cation (7) was generated (Scheme 4) and its reactivity
explored. Comparison of its reactivity to that of the biradical generated from 1 reveals
Scheme 4. Generation of authentic biradical 7.
10
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
great similarities (Table 2). In fact, the only significant difference between the reactivity
of the unknown biradical and 7 is the observation of 16 – 28% of an unreactive isomer
for the unknown biradical but not for 7. The unreactive isomer is likely formed upon
ring-opening followed by iodine atom loss upon ITCAD of 9, as shown in Scheme 5
2
0
for precursor 11 (and previously reported for the isoquinolinium analog ). This is not
-1
surprising since there is only a relatively small energy difference (7.4 kcal mol )
between the activation enthalpy for rearrangement (Scheme 2) and the C-I bond
dissociation enthalpy for 9, and the energy difference between the activation enthalpy
for the ring-opening and the rearrangement should be even smaller (for the
isoquinolinium system, the activation enthalpy for ring opening and the C-I bond
2
0
dissociation enthalpy have been calculated to be 56.8 and 62.2 kcal/mol, respectively).
Scheme 5. Two competing reaction pathways after the first ITCAD event of 3. Top:
ring-opening followed by iodine atom loss; bottom: para-benzyne (5) formation. Note
that the same applies to generation of biradical 6 (ref. 20).
1
28
100
(
(
a)
b)
8
0
0
0
0
2
22
6
-
2 x SCH
3
4
2
1
62
1
75
207
207
100
8
0
0
0
0
2
22
6
4
2
1
75
1
62
0
1
00
120
140
160
180
m/z
200
220
240
Figure 7. ITCAD mass spectra for the product ion of m/z of 222 generated upon
abstraction of two SCH groups from dimethyl disulfide by (a) the unknown biradical
generated from 1 and (b) biradical 7.
3
11
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
Table 2. Reactions, Reaction Efficiencies, and Primary Product Branching Ratios
for Gas-Phase Reactions With Dimethyl Disulfide, Cyclohexane and Allyl Iodide
for the Unknown Biradical Generated from 1 and for Biradicals 7 and 8 and Their
Calculated Vertical Electron Affinities (EA
v
) and Singlet-Triplet Splittings (ΔES-T
)
7.40a
-20.3c
8.40b
–
EA
v
(eV)
-1
ΔES-T (kcal mol )
SCH
3
abs 68% SCH
3
abs 66% e abs
53%
30%
2
°SCH
3
abs
2°SCH
3
abs
H abs
HSCH abs 32% HSCH
3
3
abs 34% SCH
3
abs 17%
1
7% URI
Efficiency = 19% Efficiency = 21% Efficiency = 28%
2
x H abs
60% 2 x H abs 63% H abs
100%
Addition
H abs
24% Addition
16% H abs
26%
11%
-
-
1
6% URI
Efficiency = 4%
Efficiency = 4%
Efficiency = 3%
Allyl abs
° I abs
Allyl-H abs 26% Allyl-H abs 19%
I abs 1% I abs 1%
2° I abs
73% Allyl abs
80% I abs
97%
3%
2
2° I abs
Allyl abs
2
° I abs
2
8% URI
Efficiency = 8%
Efficiency = 7%
Efficiency = 19%
a
b
c
Calculated at the CASPT2/CASSCF(13,12)/cc-pVTZ//CASSCF(12,12)/cc-pVTZ level of theory.
Calculated at the CASPT2/CASSCF(12,11)/cc-pVTZ//CASSCF(11,11)/cc-pVTZ level of theory.
Calculated at the CASPT2/CASSCF(12,12)/cc-pVTZ//CASSCF(12,12)/cc-pVTZ level of theory.
Corrected for zero-point vibrational energy differences at 298K by using the UB3LYP/cc-pVTZ
frequencies.
12
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
Further, the product ion of m/z 222 formed upon abstraction of two SCH
3
groups from
dimethyl disulfide by 7 was subjected to ITCAD and the fragmentation behavior was
compared to that of the ion of m/z 222 produced from the unknown biradical generated
from 1. The fragmentation patterns are similar enough to conclude that the same ion,
biradical 7, produced both of them. This finding conclusively identifies the unknown
biradical as 7.
The reactivity of the quinoline radical cation (8) was also studied (Table 2). Its reactivity
is distinctly different from that of 7. For example, while 7 predominantly abstracts an
allyl group from allyl iodide, 8 predominantly abstracts an iodine atom. This finding,
as well as the similarity of the reactivity of 7 to that of the two para-benzynes discussed
above, suggests that the reactions of 7 are initiated by the radical site at C-5 rather than
the nitrogen-radical cation site.
The different reactivities of 5, 6 and 7 are next considered. Based on previous studies,
the extent of spin-spin coupling of the unpaired electrons either through-bond, through-
space, or a combination of the two, is a major factor that controls the reactivity of most
singlet biradicals, such as the ones described here. The extent of spin-spin coupling is
reflected by the magnitude of the singlet-triplet (S-T) splitting of the biradical, which
is defined as the energy difference between the lowest-energy triplet (excited) electronic
state and the singlet (ground) electronic state.10,21–23 The greater the (calculated) S-T
splitting, the lower the reactivity of the singlet biradical. Another important reactivity-
controlling factor is the ability of the biradical to polarize the transition state of its
reactions, or polar effects, quantified here using the calculated vertical electron affinity
2
4–26
(
EA
v
) of the radical sites.
The greater the EA , the more electrophilic the biradical
v
1
3
site, and the more reactive the biradical.
All of the EA
v
s and S-T splittings calculated for the biradicals studied here are listed in
-1
Tables 1 and 2. While biradicals 5 and 6 have similar S-T splittings (-3.8 kcal mol vs.
-
1
–
4.8 kcal mol ), the greater EA of 5 (5.59 eV compared to 5.15 eV for 6) is mainly
v
responsible for the greater reactivity of this biradical. As for biradical 7, relatively
strong spin-spin coupling between the unpaired electrons on the nitrogen atom and on
-1
C-5 contributes to the large S-T splitting of this biradical (-20.3 kcal mol ), which
hinders radical reactivity. However, this biradical has a very high EA (7.40 eV), which
v
makes it more reactive than the para-benzynes 5 and 6. This is a particularly remarkable
example demonstrating that the more important reactivity-controlling factor for
biradicals with a 1,4-topology is the vertical electron affinity rather than the S-T
splitting, in agreement with previous studies.13
Conclusions
An unexpected, highly reactive biradical, the 5-dehydroquinoline radical cation, was
generated in the gas phase via hydrogen atom transfer in the 8-dehydro-5-
13
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
iodoquinolinium cation upon ITCAD followed by a cleavage of the iodine atom in a
linear quadrupole ion trap mass spectrometer. The hydrogen atom transfer was found
to have a lower activation enthalpy than the homolytic bond dissociation enthalpy of
the iodine atom, which explains the formation of the 5-dehydroquinoline radical cation
instead of the 5,8-didehydroquinolinium cation. The 5-dehydroquinoline radical cation
showed greater reactivity toward the organic reagents studied than two isomeric para-
benzynes, the 5,8-didehydroquinolinium and –isoquinolinium cations, of which the
quinoline-based para-benzyne is more reactive. This reactivity ordering is rationalized
solely by the vertical electron affinities of the radical sites in these biradicals (EA ; 7.40
v
eV, 5.59, and 5.15 eV, respectively). This finding supports the previous proposal that
polar effects predominantly control the reactivities of biradicals with a 1,4-topology,1
3
-1
even when they have substantially different singlet-triplet splittings (-20.3 kcal mol
vs. -3.8 and -4.8 kcal mol ).
-1
Experimental section
Materials
The biradical precursors, 5-iodo-8-nitroquinoline, 8-iodo-5-nitroquinoline, 5,8-
diiodoquinoline, and 5-iodo-8-nitroisoquinoline, as well as 5-iodoquinoline and 8-
iodoquinoline used to generate monoradicals were synthesized via a Sandmeyer
2
7
reaction followed by a nitration reaction from 5-aminoquinoline, 8-aminoquinoline
and 5-aminoisoquinoline, respectively. The starting materials were purchased from
Sigma-Aldrich (St. Louis, MO) and used as received. The model compound 5,8-
dithiomethylquinoline was synthesized from 2,5-dibromoaniline via Skraup quinoline
synthesis followed by an aromatic nucleophilic substitution reaction28,29 (see
Supporting Information for detailed synthetic procedures). The reagents, dimethyl
disulfide, allyl iodide and cyclohexane, were purchased from Sigma-Aldrich and used
as received.
Experiments
All of the gas-phase reactions were studied in a Thermo LTQ linear quadrupole ion trap
mass spectrometer (LQIT-MS) coupled with a home-built reagent mixing manifold for
reagent introduction. All biradical precursors were dissolved in methanol, injected into
the mass spectrometer, and protonated via atmospheric pressure chemical ionization
(
APCI). Protonated biradicals were generated either by two consecutive collision-
activated dissociation (CAD) events in the ion trap (ITCAD, collision energy ranging
from 30 to 50, arbitrary units; the monoradical formed in the first ITCAD event was
isolated before activating it in the second ITCAD event) and/or in-source CAD (ISCAD,
collision energy 60 V), which resulted in homolytic cleavages of the C-I and C-NO
bonds. The protonated biradicals were then isolated and allowed to react with dimethyl
2
14
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
1
3
disulfide, allyl iodide or cyclohexane in the ion trap for variable reaction times up to
0000 ms. The monoradical precursors 5-iodoquinoline and quinoline were dissolved
in CS and ionized via APCI to generate their molecular ions. The 5-iodoquinoline
1
2
molecular ion was subjected to ITCAD to generate the 5-dehydroquinoline radical
cation. The reactivities of the radical cations were studied toward dimethyl disulfide,
allyl iodide, and cyclohexane as described above.
Pseudo-first order reaction rate plots were used to determine the reaction efficiencies
(
defined as the fraction of collisions that lead to reaction) given as kexp/kcoll, where kexp
is the experimentally measured rate constant and kcoll is the collision rate constant
calculated using a parameterized trajectory theory.30 The rate plots were examined
carefully, including the fit of the decay of the reactant ion to a single exponential, to
rule out the presence of mixtures of isomeric reactive ions. Branching ratios for primary
products were also determined. Many of the observed ionic reaction products were
isolated and subjected to further ITCAD to obtain structural information.
Computational Methods
Molecular geometries for all species were optimized at the multiconfigurational self-
consistent field (MCSCF) and density functional (DFT) levels of theory by using either
3
1
the correlation-consistent polarized valence-triple- (cc-pVTZ) basis set or the 6-
3
2
3
11G(d,p) basis set. The latter basis set was used for calculations in which the
chemical species contained at least one iodine atom. The MCSCF calculations were of
3
3
the complete active space (CASSCF) variety and included (in the active space) the
full -space for each molecule and, for each of the mono- and biradicals, the
nonbonding  orbital(s). The DFT calculations used either (1) the gradient-corrected
3
4
exchange functional of Becke, which was combined with the gradient-corrected
3
5
36
correlation functional of Lee, Yang and Parr (B3LYP), or (2) the M06-2X functional.
All DFT geometries were verified to be local minima by computation of analytic
vibrational frequencies, and these (unscaled) frequencies were used to compute zero-
point vibrational energies (ZPVE) and 298 K thermal contributions (H298 – E ) for all
0
species. DFT calculations for the mono- and biradicals employed an unrestricted
formalism.
To improve the molecular orbital calculations, dynamic electron correlation was also
accounted for by using multi-reference second-order perturbation theory37,38 (CASPT2)
for multi-configurational self-consistent field (MCSCF) reference wave functions;
these calculations were carried out for the MCSCF optimized geometries. Some caution
must be applied in interpreting the CASPT2 results since this level of theory is known
to suffer from a systematic error proportional to the number of unpaired electrons.39
Thus, the electronic energies are of the CASPT2/CASSCF(m,n)/cc-
pVTZ//CASSCF(m,n)/cc-pVTZ variety (where m is the number of active electrons and
n is the number of active orbitals), and estimates of the thermodynamic quantities, E
0
and H298, are derived by adding to these electronic energies ZPVE and the sum of ZPVE
15
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
and (H298 – E
0
), respectively, where the latter are derived from the DFT (B3LYP)
calculations.
In order to compute vertical electron affinities (EA
single-point calculations (CASPT2/CASSCF(m,n)/cc-pVTZ),
v
) for the mono- and biradicals,
using the
CASSCF(m,n)/cc-pVTZ optimized geometry for each radical, were also carried out for
the states that are produced when a single electron is added to the nonbonding  orbital
(
(
or one of the two such orbitals) of each molecule.40 Thus, for the monoradicals,
doublet ground states) these calculations were carried out for (zwitterionic) singlet
states, whereas for the biradicals (singlet ground states) calculations were carried out
4
1
for (zwitterionic) doublet states.
4
2
All CASPT2/MCSCF and DFT calculations were carried out with the MOLCAS 8.0
and Gaussian 09 electronic structure program suites, respectively.
4
3
Acknowledgement
The authors thank the National Science Foundation (NSF, grant number CHE-1464712)
for financial support.
Keywords
1
,4-biradicals, para-benzynes, aromatic, gas-phase reactions, quantum chemical
calculations, reactivity
(
(
1) Wenk, H. H.; Winkler, M.; Sander, W. Angew. Chem. Int. Ed. 2003, 42, 502–528.
2) Biggins, J. B.; Prudent, J. R.; Marshall, D. J.; Ruppen, M.; Thorson, J. S. Proc.
Natl. Acad. Sci. 2000, 97, 13537–13542.
(
(
3) Plourde, II, G.; El-Shafey, A.; Fouad, F. S.; Purohit, A. S.; Jones, G. B. Bioorg.
Med. Chem. Lett. 2002, 12, 2985–2988.
4) Jones, G. B.; Wright, J. M.; Plourde, G.; Purohit, A. D.; Wyatt, J. K.; Hynd, G.;
Fouad, F. J. Am. Chem. Soc. 2000, 122, 9872–9873.
(
(
(
(
5) Nicolaou, K. C.; Dai, W.-M. Angew. Chem. Int. Ed. Engl. 1991, 30, 1387–1416.
6) Nicolaou, K. C.; Smith, A. L. Acc. Chem. Res. 1992, 25, 497–503.
7) Gredičak, M.; Jerić, I. Acta Pharm. 2007, 57, 133-150.
8) Pratviel, G.; Bernadou, J.; Meunier, B. Angew. Chem. Int. Ed. Engl. 1995, 34,
7
46–769.
(
(
9) Thoen, K. K.; Kenttämaa, H. I. J. Am. Chem. Soc. 1997, 119, 3832–3833.
10) Williams, P. E.; Jankiewicz, B. J.; Yang, L.; Kenttämaa, H. I. Chem. Rev. 2013,
113, 6949–6985.
16
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
(
(
(
11) Gao, J.; Jankiewicz, B. J.; Reece, J.; Sheng, H.; Cramer, C. J.; Nash, J. J.;
Kenttämaa, H. I. Chem. Sci. 2014, 5, 2205–2215.
12) Hoffner, J.; Schottelius, M. J.; Feichtinger, D.; Chen, P. J. Am. Chem. Soc. 1998,
20, 376–385.
1
13) (a) Sheng, H.; Ma, X.; Lei, H.-R.; Milton, J.; Tang, W.; Jin, C.; Gao, J.; Wittrig,
A. M.; Archibold, E. F.; Nash, J. J.; Kenttämaa, H. I. ChemPhysChem 2018, 19,
2
839–2842. (b) Wittrig, A. M.; Archibold, E. F.; Sheng, H.; Nash, J. J.;
Kenttämaa, H. I. Int. J. Mass Spectrom. 2015, 377, 39–43.
14) Jankiewicz, B. J.; Reece, J. N.; Vinueza, N. R.; Nash, J. J.; Kenttämaa, H. I.
Angew. Chem. Int. Ed. 2008, 47, 9860–9865.
(
(
15) Stirk, K. M.; Orlowski, J. C.; Leeck, D. T.; Kenttamaa, H. I. J. Am. Chem. Soc.
1
992, 114, 8604–8606.
(
(
16) Price, J. M.; Kenttämaa, H. I. J. Phys. Chem. A 2003, 107, 8985–8995.
17) Jankiewicz, B. J.; Adeuya, A.; Yurkovich, M. J.; Vinueza, N. R.; Gardner, S. J.;
Zhou, M.; Nash, J. J.; Kenttämaa, H. I. Angew. Chem. Int. Ed. 2007, 46, 9198–
9
201.
(
(
(
18) Pramanik, B. Applied Electrospray Mass Spectrometry; CRC Press: New York,
N.Y., 2002.
19) Jankiewicz, B. J.; Vinueza, N. R.; Kirkpatrick, L. M.; Gallardo, V. A.; Li, G.;
Nash, J. J.; Kenttämaa, H. I. J. Phys. Org. Chem. 2013, 26, 707–714.
20) Kirkpatrick, L. M.; Vinueza, N. R.; Jankiewicz, B. J.; Gallardo, V. A.; Archibold,
E. F.; Nash, J. J.; Kenttämaa, H. I. Chem. - Eur. J. 2013, 19, 9022–9033.
21) Schottelius, M. J.; Chen, P. J. Am. Chem. Soc. 1996, 118, 4896–4903.
22) Wenthold, P.; Hu, J.; Squires, R. J. Am. Chem. Soc. 1996, 118, 11865–11871.
23) Logan, C. F.; Chen, P. J. Am. Chem. Soc. 1996, 118, 2113–2114.
24) Nash, J. J.; Kenttämaa, H.; Cramer, C. J. J. Phys. Chem. A 2006, 110, 10309–
(
(
(
(
1
0315.
25) Donahue, N. M.; Clarke, J. S.;Anderson, J. G. J. Phys. Chem. A 1998, 102, 3923-
933.
26) Clarke, J. S.; Kroll, J. H.; Donahue, N. M.; Anderson, J. G. J. Phys. Chem. A
998, 102, 9847–9857.
27) Son, J.-H.; Pudenz, M. A.; Hoefelmeyer, J. D. Dalton Trans. 2010, 39, 11081-
1090.
(
(
(
3
1
1
(
(
28) Palmer, M. H. J. Chem. Soc. 1962, 0, 3645-3652.
29) Maślankiewicz, A.; Marciniec, K.; Pawlowski, M.; Zajdel, P. Heterocycles 2007,
7
1, 1975.
(
(
(
30) Su, T.; Chesnavich, W. J. J. Chem. Phys. 1982, 76, 5183–5185.
31) Dunning, T. H. J. Chem. Phys. 1989, 90, 1007-1023.
32) (a) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys. 1980, 72,
6
50-654. (b) Glukhovstev, M. N.; Pross, A.; McGrath, M. P.; Radom, L. J. Chem.
Phys. 1995, 103, 1878-1885.
(
(
(
33) Roos, B. O.; Taylor, P. R.; Siegbahn, P. E. M. Chem. Phys. 1980, 48, 157-173.
34) Becke, A. D. Phys. Rev. A 1988, 38, 3098-3100.
35) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B, 1988, 37, 785-789.
17
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
(
(
36) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215-241.
37) Andersson, K.; Malmqvist, P.-Å.; Roos, B. O.; Sadlej, A. J.; Wolinski, K. J. Phys.
Chem. 1990, 94, 5483-5488.
(
(
(
38) Andersson, K. Theor. Chim. Acta 1995, 91, 31-46.
39) Andersson, K.; Roos, B. O. Int. J. Quantum Chem. 1993, 45, 591-607.
40) Note that, for these calculations, we are computing the electron affinity of the
radical site not the electron affinity of the molecule.
(
41) Because the mono- and biradicals studied here contain a formal positive charge
on the nitrogen atom, the state that is produced when an electron is added to the
nonbonding orbital of any one of these species is formally zwitterionic; that is, it
contains localized positive () and negative () charges.
(
42) (a) MOLCAS 7.4: Aquilante, F.; De Vico, L.; Ferré, N.; Ghigo, G.; Malmqvist,
P.-°A.; Neogrády; Pedersen, T. B.; Pitonak, M.; Reiher, M.; Roos, B. O.; Serrano-
Andrés, L.; Urban, M.; Veryazov, V.; Lindh, R. J. Comp. Chem. 2010, 31, 224-
2
47. (b) Code Development: Veryazov, V.; Widmark, P.-O.; Serrano- Andrés, L.;
Lindh, R. Roos, B. O. Int. J. Quant. Chem. 2004, 100, 626-635. (c) MOLCAS
: Karlstrőm, G.; Lindh, R.; Malmqvist, P.-°A.; Roos, B. O.; Ryde, U.; Veryazov,
7
V.; Widmark, P.-O.; Cossi, M.; Schimmelpfennig, B.; Neogrády, P.; Seijo, L.
Comp. Mat. Sci. 2003, 28, 222-239.
(
43) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.;
Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.;
Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven,
T.; Montgomery, Jr., J. A.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.;
Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand,
J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.;Cossi,
M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.;Yazyev, O.; Austin, A.
J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.;
Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.;
Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D.
J. Gaussian, Inc., Wallingford CT, 2013.
18
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201806106
Table of contents
Text for
Table of
Contents
19
This article is protected by copyright. All rights reserved.