Reversible switching between housane and cyclopentanediyl isomers: An isonitrile-catalysed thermal reverse reaction

Article abstract of DOI:10.1039/d0dt02688c

The photo-isomerization of an isolable five-membered singlet biradical based on C, N, and P ([TerNP]2CNDmp, 2a) selectively afforded a closed-shell housane-type isomer (3a) by forming a transannular P-P bond. In the dark, the housane-type species re-isomerized to the biradical, resulting in a fully reversible overall process. In the present study, the influence of tBuNC on the thermal reverse reaction was investigated: The isonitrile acted as a catalyst, thus allowing control over the thermal reaction rate. Moreover, tBuNC also reacted with the biradical to form an adductspecies ([TerNP]2CNDmp·CNtBu, 4a), which can be regarded as the resting state of the system. The reactive species 2a and 3a could be re-generated in situ by irradiation with red light. The results of this study extend our understanding of this new class of molecular switches.

Full text of DOI:10.1039/d0dt02688c

Dalton
Transactions
View Article Online
View Journal
PAPER
Reversible switching between housane and
cyclopentanediyl isomers: an isonitrile-catalysed
Cite this: DOI: 10.1039/d0dt02688c
thermal reverse reaction†
a
a
a,b
a,b,c
Henrik Beer,
Jonas Bresien, * Dirk Michalik, Axel Schulz
*
and
a
Alexander Villinger
The photo-isomerization of an isolable five-membered singlet biradical based on C, N, and P
2
([TerNP] CNDmp, 2a) selectively afforded a closed-shell housane-type isomer (3a) by forming a transan-
nular P–P bond. In the dark, the housane-type species re-isomerized to the biradical, resulting in a fully
reversible overall process. In the present study, the influence of tBuNC on the thermal reverse reaction
was investigated: the isonitrile acted as a catalyst, thus allowing control over the thermal reaction rate.
Received 31st July 2020,
Accepted 24th August 2020
Moreover, tBuNC also reacted with the biradical to form an adduct species ([TerNP] CNDmp⋅CNtBu, 4a),
which can be regarded as the resting state of the system. The reactive species 2a and 3a could be re-gen-
2
DOI: 10.1039/d0dt02688c
erated in situ by irradiation with red light. The results of this study extend our understanding of this new
rsc.li/dalton
class of molecular switches.
The reversibility of the process and stability of the isomers
and 3 are noteworthy considering that analogous carbon-
Introduction
2
Singlet biradicals (also known as biradicaloids) are worthwhile based cyclopentane-1,3-diyls are usually short-lived and their
1
–5
target molecules in synthetic chemistry. Due to their unique isomerization to housane or cyclopentene derivatives is typi-
6
–10
4,35–38
32,33
electronic structure,
interesting properties such as small molecule activation,
they combine high reactivity with cally irreversible,
with few exemptions.
Also, the
1
1–14
photoisomerization of a related heterocyclic housane was
1
5,16
39
non-linear optical properties,
or (potential) applicability as found to be irreversible.
Especially for the latter use, it is desir- Since the housane-type species 3 does not possess signifi-
able to devise systems whose two (meta-)stable states possess cant biradical character, as opposed to the biradical 2 (β =
1
6–23
molecular switches.
2
3,34
different degrees of biradical character, so the biradical 24–27%),
switching between these two isomers also allows
6
,8,24–26
character
stimulus.
can be influenced by an outside reversible manipulation of the biradical character. In particu-
2
7–30
lar, switching the biradical character “on” and “off” made it
As stable photo-switchable biradicals are barely possible to influence a (thermal) equilibrium reaction, which
2
2,31–33
23
known,
we recently started to investigate the photo- involved the activation of tBuNC by biradical 2 (Scheme 2).
chemistry of five-membered heterocycles, namely hetero-cyclo-
Interestingly, it was noted that the concentration of tBuNC
pentanediyls (2) generated from the four-membered biradical could influence the rate of the thermal reverse reaction 3 →
2
0,21
23,34
23
[
P(μ-NR)] (1, Scheme 1).
In two recent publications
we 2, indicating a catalytic effect of the isonitrile. This prompted
2
could show that these five-membered biradicals can be reversi- us to investigate the mechanism of the thermal reverse reac-
bly switched to the corresponding housane-type isomer (3) tion in more detail, so as to understand the reactivity of the
using red light, while the reverse reaction takes place under
thermal regime.
a
Institute of Chemistry, University of Rostock, Albert-Einstein-Straße 3a, 18059
Rostock, Germany. E-mail: jonas.bresien@uni-rostock.de, axel.schulz@uni-rostock.de
b
Leibniz Institute for Catalysis, Albert-Einstein-Straße 29a, 18059 Rostock, Germany
c
Department Life, Light & Matter, University of Rostock, 18051 Rostock, Germany
†
Electronic supplementary information (ESI) available: Syntheses, equipment,
Scheme 1 Synthesis and photochemistry of hetero-cyclopentanediyls
(2). R = Ter (2,6-dimesitylphenyl), Hyp (tris(trimethylsilyl)silyl); R’ = Mes
(mesityl), Dmp (2,6-dimethylphenyl), tBu. Similar compounds can be
obtained by reaction of 1 with CO (not shown).
analytical data, reaction kinetics, computations. CCDC 2013704. For ESI and
crystallographic data in CIF or other electronic format see DOI: 10.1039/
d0dt02688c
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 1 Left: molecular structure of the adduct 4a in the crystal. Right:
2
3
Scheme 2 The molecular switch 2 could be utilized to control a
chemical equilibrium reaction involving the activation of tBuNC by 2.
The equilibrium with 1 is only observed for 2b.
the molecular structure of the biradical 2a is depicted for comparison.
Thermal ellipsoids set at 50% probability (123 K). Selected bond lengths
(Å) and angles (°): 4a: P1–N1 1.779(1), P1–N2 1.734(1), P1–C58 1.888(1),
P2–N1 1.776(1), P2–C49 1.880(1), P2–C58 1.937(1), N2–C49 1.403(2),
N3–C49 1.270(2), N4–C58 1.250(2), N1–P2–P1–N2 116.57(7); 2a: P1–N1
1
1
.729(1), P1–C49 1.792(2), P2–N1 1.649(1), P2–N2 1.680(1), N2–C49
.430(2), N3–C49 1.287(2), N1–P1–P2–N2 178.5(1).
housane 3 towards tBuNC in comparison with the activation
chemistry of biradical 2. Being the most stable
2
1,23,34
derivative,
the Ter and Dmp substituted variant of 2 (i.e.
2
a, R = Ter, R′ = Dmp) was chosen for these investigations.
ture is similar to other derivatives of 4a with different
substituents.
2
1
Photoisomerization and thermal reverse reaction
Results and discussion
When dissolving colourless crystals of adduct 4a in THF or
benzene, a blue solution was obtained. This could be attribu-
ted to partial dissociation of the adduct, which led to partial
formation of the blue biradical 2a and tBuNC in an equili-
brium mixture with adduct 4a (cf. Scheme 3). The mixture was
then irradiated with red light (638 nm) in the NMR spectro-
meter, resulting in partial formation of the photoproduct 3a
Molecular structure
2
3
As previously observed in NMR experiments, the reaction of
a with tBuNC led to formation of adduct 4a (Scheme 3). As it
2
had only been characterized spectroscopically, it was of inter-
est to isolate and fully characterize adduct 4a. This could be
accomplished by using the isonitrile as solvent (i.e. in a large
excess) and slowly evaporating the isonitrile until colourless
crystals of the desired product 4a were obtained.
(
70%, Scheme 4). All these reactions could easily be traced by
P NMR spectroscopy, owing to the distinct AX spin systems
31
Compound 4a crystallized in the monoclinic space group
C2/c with 8 formula units per cell. As expected, the five-mem-
of 2a (221.7, 258.3 ppm; 136 Hz), 3a (−129.1, −63.4 ppm; 65
Hz), and 4a (188.2, 222.7 ppm; 33 Hz). The experimental data
correspond well to calculated values (Table S4†).
bered N
2 2
P C ring system of the former biradical moiety adopts
an envelope conformation, contrary to the planar structure of
the free biradical 2a (Fig. 1). The two P atoms are bridged by
the C atom of the former isonitrile moiety, resulting in a
In contrast to irradiation of the pure biradical 2a or the
2
3
23
equilibrium mixture associated with 2b (cf. Scheme 2), it was
[2.1.1]-bicyclic structure; that is, the adduct can be regarded as
a [2 + 1] cycloaddition product of the biradical 2a and tBuNC.
It is worthy to note that the P1/P2–C58 bond lengths (1.888(1),
1
.937(1) Å) are somewhat longer than typical P–C single bonds
4
0
(
∑rcov = 1.86 Å), indicating a rather weak interaction. This is
in agreement with the fact that the adduct formation is a
reversible equilibrium reaction (vide infra). The overall struc-
Scheme 4 The thermal reverse reaction 3a → 2a was found to be cata-
lysed by tBuNC. The reactions considered for the kinetic model are
highlighted in red.
Scheme 3 tBuNC could be activated using biradical 2a, leading to for-
mation of adduct 4a.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
however not possible to achieve full photoconversion; species
3
1
2
a and 4a were detectible in the P NMR spectrum even under
irradiation (20% and 10%, respectively). Moreover, the thermal
equilibrium mixture was restored almost instantly when the
light source was switched off, indicating a rather fast thermal
reverse reaction. (Note that for the pure system 2a ⇄ 3a, the
2
3
half-life of 3a was about 7 minutes at ambient temperature.)
These findings implied that the presence of tBuNC had an
influence on the rate of the thermal reverse reaction 3a → 2a.
This could be confirmed by low-temperature NMR experiments
using different concentrations of tBuNC, which showed a
direct proportionality between the rate of the thermal reverse
reaction and the tBuNC concentration (Scheme 4 and ESI, p.
Fig. 2 Thermal reverse reaction 3a → 2a in the presence of tBuNC.
Left: −20 °C,
1 equiv. tBuNC. Right: −30 °C, 10 equiv. tBuNC.
Experimental data plotted as circles; fit functions plotted as solid lines.
S20ff†). It follows that besides the intramolecular P–P bond For more temperatures and concentrations of tBuNC, please refer to the
ESI, p. S26ff.†
cleavage of 3a (as observed for the pure system 2a ⇄ 3a), there
must be an alternative pathway for the isomerization of the
housane 3a to the biradical 2a, involving a reaction between
tBuNC and 3a.
Most importantly, the same rate constants were obtained
when the concentration of tBuNC was varied and the tempera-
ture left unchanged (ESI†), in agreement with the consider-
Reaction mechanism
1
ation that k is associated with a second order reaction. The
To further shed light on the reaction mechanism, the reverse temperature dependence of the rate constants was then
3
1
reaction was traced by P NMR spectroscopy at different temp- exploited to estimate the activation barriers of the three reac-
eratures and at different concentrations of tBuNC. The NMR tions associated with k , k , and k using transition state
1
2
3
‡
−1
spectra were used to estimate the concentrations of all species theory (Table 1, Fig. S11†). This gave ΔG = 75.9(6) kJ mol for
in solution, and the time-dependent concentration data were the reaction of tBuNC and 3a at 298.15 K, which is lower than
then used to fit a kinetic model involving a second order reac- the activation barrier of intramolecular P–P bond cleavage
‡
−1 23
tion step between tBuNC and 3a (rate constant k , Scheme 4). (ΔG = 88(4) kJ mol
)
in accordance with experimental
1
As the putative intermediate 5a (or any other intermediates) observation.
could not be observed spectroscopically, it was assumed that
Moreover, it became evident from the concentration plots
the reaction step tBuNC + 3a was the rate determining step, i.e. in Fig. 2 that the adduct 4a was the main product of the
all reaction steps between 5a and 2a were much faster (k ≫ k ) thermal reverse reaction, i.e. it can be regarded as the resting
4
1
and therefore negligible. Also, taking into consideration state of the system. Upon irradiation, the equilibrium was per-
that the thermal reverse reaction was orders of magnitude turbed and the thermal reaction was formally shifted back to
faster in the presence of tBuNC, the thermal equilibration of t=0, yielding the housane 3a and free tBuNC as main pro-
3
a via intramolecular P–P bond cleavage was neglected ducts. In the dark, the system relaxed to t→∞. As indicated,
(Scheme 4).
this process is fully reversible and can be repeated many times
The above considerations led to the following rate (>20) without detectable degradation of the system.
equations
Computational study
d½2aꢀ
¼
þk1½3aꢀ½tBuNCꢀ ꢁ k2½2aꢀ½tBuNCꢀ þ k3½4aꢀ
d½3aꢀ
dt
The experimental data clearly demonstrated that the housane
3a reacted with tBuNC in a catalytic manner. However, up to
¼
ꢁk1½3aꢀ½tBuNCꢀ
this point, it remained unclear what the nature of the inter-
mediate 5a was, as it could not be observed in any of the NMR
spectra. We therefore decided to perform DFT and ab initio cal-
dt
d½4aꢀ
dt
¼
þk2½2aꢀ½tBuNCꢀ ꢁ k3½4aꢀ
d½tBuNCꢀ
dt
¼
ꢁk2½2aꢀ½tBuNCꢀ þ k3½4aꢀ
Table 1 Experimental enthalpies and entropies of activation
where [2a], [3a], [4a], and [tBuNC] are the concentrations of 2a,
a, 4a, and tBuNC, respectively, and t is the reaction time. The
ΔH‡ [kJ mol−1]
ΔS‡ [J (mol K)−1]
‡
2
a
[kJ mol−1
]
Step
ΔG
98K
3
1
2
(k
(k
1
)
)
42.2 ± 0.3
43 ± 3
67 ± 6
−113 ± 1
−106 ± 11
−50 ± 23
75.9 ± 0.6
74 ± 6
82 ± 13
differential equations were used in a non-linear fitting pro-
cedure (ESI, p. S21ff†) to obtain the parameters k , k , and k
2
1
2
3
3 (k3)
as well as the initial concentrations of all species ([2a]
0 0
, [3a] ,
a
Extrapolated values. The rate constants could only be determined at
lower temperatures due to the high reaction rates at ambient
temperature.
[
4a] , [tBuNC] ) for all different temperatures and concen-
0
0
trations of tBuNC (Fig. 2 and ESI, p. S26ff†).
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
culations to investigate the reactivity of housane 3a towards
tBuNC.
Next, it was of interest to determine if the same reaction
pathway could also be found in case of the “real” system 3a +
In a first step, a model system was chosen to obtain an tBuNC → 2a + tBuNC. To keep matters simple and reduce the com-
initial impression of what the reaction between the housane putational effort, we decided only to investigate the shorter
and isonitrile might look like. To this end, all substituents pathway via TS2 (highlighted in red in Fig. 3) and the intra-
(Ter, Dmp, tBu) were replaced by methyl groups. Using this molecular P–P bond cleavage via TS0 (grey). Indeed, the results
truncated model (2c, 3c), loose van der Waals complexes of 3c implied that the same nucleophilic attack of tBuNC is possible in
and MeNC as well as 2c and MeNC were optimized. These case of 3a despite the presence of the bulky substituents, and that
were then used as fixed end points for the calculation of a it has a lower activation barrier than the intramolecular P–P bond
4
1–46
Nudged Elastic Band (NEB)
trajectory, which resulted in cleavage (Fig. 4). The calculated energies of TS0 and TS1 are some-
an approximate Minimum Energy Path (MEP) connecting the what larger than the experimental activation energies, but this is
two end points on the Potential Energy Surface (PES). likely attributed to some multireference character of the species
Optimization of all stationary points as well as additional involved, which is expected to decrease the accuracy of these calcu-
4
7,48
Intrinsic Reaction Coordinate (IRC)
scans resulted in two lations (cf. ESI p. S29f†). Moreover, we did not include any solvent
possible reaction pathways that encompass a nucleophilic effects in our calculations. Nonetheless, the experimental trends
attack of the isonitrile at one of the P atoms of the housane as are well reflected in the computational model. In particular, the
rate determining step (via TS1), leading to open-chain inter- computed mechanism also predicts the addition of tBuNC to
mediate 5c (Fig. 3). Further reaction steps leading to the bi- housane 3a to be the rate determining step, in agreement with the
radical 2c and free isonitrile involved either a nucleophilic kinetic model derived from NMR data.
attack by the N1 atom (via TS2) or an addition of the N4 atom
to the second P atom (TS4), followed by [2 + 2] cycloreversion nitrile and biradical was found in the model system (cf. A3 in
TS5). All these steps exhibited very low activation barriers; Fig. 3), we were also interested in different adduct types of the
Taking into consideration that a side-on adduct of the iso-
(
thus, it would be expected that the concentration of all poss- actual biradical 2a. Given that the end-on adduct 4a was identi-
ible intermediates would be close to zero throughout the reac- fied as the “resting state” of the system, it seemed particularly
tion (Fig. 3).
Additionally, the intramolecular P–P bond cleavage was cal- reactivity of the biradical 2a or the housane 3a.
culated (via TS0), similarly to what was reported before for the Contrary to the model system, though, our computations
interesting to see if a side-on adduct might play a role in the
2
3
reaction of the pure system 2a ⇄ 3a. In comparison with the revealed that the side-on adducts (stationary points A2 and A3
‡
isonitrile-catalysed reaction, the activation energy (ΔG ) of the in Fig. 4) of tBuNC and the biradical 2a are much higher in
−
1
intramolecular pathway is some 17 kJ mol higher, in agree- energy than the end-on adduct 4a. This agrees with the
ment with our experimental observations. absence of other adduct species in the NMR spectra, and
Fig. 3 Thermal reverse reaction of the model housane 3c (UPBE-D3/def2-TZVP, p° = 1 atm). The addition of MeNC is the rate determining step,
leading to formation of metastable adduct 5c (stationary point INT1). The adduct can react to biradical 2c via two pathways, both of which have very
low activation barriers. The shortest path is highlighted in red.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
Fig. 4 Computed mechanism for the thermal reverse reaction of the housane 3a in the presence of tBuNC (DLPNO-CCSD(T)/def2-TZVP//
−1
UPBE-D3/def2-SVP, c° = 1 mol L ). The reaction with tBuNC is energetically favoured in comparison with intramolecular P–P bond cleavage, in
agreement with experimental results.
moreover, it indicates that the second reaction pathway found
All manipulations were carried out under oxygen- and
for the model system is most likely not favoured in case of the moisture-free conditions under an inert atmosphere of argon
actual system, due to the large substituents. Furthermore, it is using standard Schlenk or Glovebox techniques. NMR spectra
worthy to note that adduct 4a is predicted to be the global under irradiation were recorded using our previously pub-
2
3
minimum of the reaction coordinate, in accord with experi- lished setup, which was adopted from a setup published by
4
9
mental observation.
the Gschwind group.
Syntheses
Conclusions
Biradical 2a was synthesized according to a modified literature
2
3
procedure.
To
a
solution of [P(µ-NTer)]
2
(459 mg,
In summary, our investigations of the thermal reverse reaction of
the housane 3a indicate that two distinct isonitrile adducts play an
important role for the reactivity of the molecular switch 2a ⇄ 3a:
firstly, the biradical 2a can activate the isonitrile in a [2 + 1] cyclo-
addition reaction, resulting in a [2.1.1]bicyclic structure, which also
represents the global minimum of the reaction coordinate.
Secondly, the housane 3a can be activated by the isonitrile in a
nucleophilic attack, leading to an unstable open-chain adduct (5a)
that quickly fragments into the biradical 2a and the isonitrile. This
opens up a different pathway for the thermal reverse reaction of
the housane 3a, and allows chemical control over the rate of the
thermal reverse reaction.
Moreover, knowledge of the adduct formation between
housane 3a and isonitrile will help us devise new reactions
that take advantage of the nucleophilic activation of the
housane. Inter alia, this could lead to stable adducts that facili-
tate trapping of the thermally labile housane species, or even
open new avenues in photoswitchable reactions.
0
.640 mmol) in benzene (10 mL), DmpNC (82 mg, 0.64 mmol)
was added. An immediate colour change from red to deep blue
was observed. After two hours the solvent was removed, and
the blue residue was dried in vacuo (1 × 10− mbar). The
product was crystallized from a minimal amount of fresh
benzene at ambient temperature. The supernatant was
removed by syringe and the crystals were dried in vacuo (1 ×
3
−
3
10
mbar). Yield: 480 mg (0.560 mmol, 88%). Mp. 207 °C.
CHN calcd (found) in %: C 80.73 (80.36), H 7.01 (6.54), N 4.95
3
1
1
2
31
(
4.81). P{ H} NMR (C D , 202.5 MHz): δ = 221.7 (d, J( P,
6 6
3
1
2
31
31
P) = 136 Hz, 1P, NPC), 258.3 (d, J( P, P) = 136 Hz, 1P,
NPN). A full set of analytical data can be found in the ESI.†
Adduct 4a: biradical 2a (60 mg, 0.07 mmol) was dissolved in
tBuNC (1.849 mg, 2.5 mL, 22.1 mmol). The light blue solution was
stored at −20 °C. Only a few colorless crystals could be collected.
Yield <20%. Mp. 148 °C. CHN calcd (found) in %: C 79.97 (79.05),
H 7.36 (7.18), N 6.02 (5.87). Deviations due to adhering impurities,
31
1
cf. ESI.† P{ H} NMR (243 K, THF-d , 101.5 MHz): δ = 188.2 (d,
8
2
31 31
2 31 31
J( P, P) = 33 Hz, 1P, NPC), 222.7 (d, J( P, P) = 33 Hz, 1P,
NPN). Additional analytical data can be found in the ESI.†
Experimental
General information
Kinetic studies
For detailed information on syntheses, equipment, analytical To investigate the thermal reverse reaction 3a → 2a in the pres-
data, computational methods etc. please also consult the ESI.†
ence of tBuNC, solutions of the biradical 2a with different con-
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
centrations of tBuNC were irradiated in the NMR spectrometer
6 L. Salem and C. Rowland, Angew. Chem., Int. Ed. Engl.,
1972, 11, 92–111.
(optical power output approx. 200 mW). When a dynamic
steady-state between photoconversion (2a → 3a) and thermal
reverse reaction (3a → 2a) was reached, the laser diode was
turned off and the thermal equilibration (Fig. S5†) was traced
by in situ P NMR spectroscopy. The concentrations of all
species were inferred from the time-resolved NMR spectra.
Using a non-linear fitting procedure, the differential rate
7 Diradicals, ed. W. T. Borden, Wiley-Interscience, New York,
1982.
8 E. Miliordos, K. Ruedenberg and S. S. Xantheas, Angew.
Chem., Int. Ed., 2013, 52, 5736–5739.
9 G. Gryn’ova, M. L. Coote and C. Corminboeuf, Wiley
Interdiscip. Rev.: Comput. Mol. Sci., 2015, 5, 440–459.
3
1
equations were fitted to the experimental data, yielding the 10 T. Stuyver, B. Chen, T. Zeng, P. Geerlings, F. De Proft and
rate constants as fitting parameters. For more details, please
refer to the ESI, p. S20ff.†
R. Hoffmann, Chem. Rev., 2019, 119, 11291–11351.
11 A. Hinz, R. Kuzora, U. Rosenthal, A. Schulz and
A. Villinger, Chem. – Eur. J., 2014, 20, 14659–14673.
Computational methods
12 A. Hinz, R. Kuzora, A.-K. Rölke, A. Schulz, A. Villinger and
4
6
Computations were carried out using ORCA 4.2.1
and
R. Wustrack, Eur. J. Inorg. Chem., 2016, 2016, 3611–3619.
5
0
Gaussian 09. Structure optimizations were performed using 13 J. Bresien, A. Hinz, A. Schulz and A. Villinger, Dalton
5
1,52
the PBE exchange–correlation functional
in conjunction
Trans., 2018, 47, 4433–4436.
and Ahlrichs’s 14 J. Bresien, A. Hinz, A. Schulz and A. Villinger, Eur. J. Inorg.
def2 basis set family. Accurate electronic energies for opti- Chem., 2018, 2018, 1679–1682.
mized structures were computed by single-point DLPNO-CCSD 15 J.-J. Wang, Z.-J. Zhou, H.-M. He, D. Wu, Y. Li, Z.-R. Li and
5
3,54
with Grimme’s dispersion correction D3(BJ)
5
5
5
6–59
55
(
T)
def2-TZVP/C correlation fitting basis.
Transition states were located on the PES using the Nudged
calculations employing the def2-TZVP basis set and
H.-X. Zhang, J. Phys. Chem. C, 2016, 120, 13656–13666.
16 K. Okuno, Y. Shigeta, R. Kishi and M. Nakano, J. Phys.
Chem. Lett., 2013, 4, 2418–2422.
6
0
4
1–45
Elastic Band (NEB) algorithm
implemented in ORCA at 17 H. Li, A. C. Fahrenbach, A. Coskun, Z. Zhu, G. Barin,
the PBE-D3/def2-TZVP level of theory. All transition state (TS)
structures were verified to be connected to the corresponding
Y.-L. Zhao, Y. Y. Botros, J.-P. Sauvage and J. F. Stoddart,
Angew. Chem., Int. Ed., 2011, 50, 6782–6788.
18 A. T. Buck, J. T. Paletta, S. A. Khindurangala, C. L. Beck and
A. H. Winter, J. Am. Chem. Soc., 2013, 135, 10594–
10597.
4
7,48
minima using Intrinsic Reaction Coordinate (IRC)
scans.
For further details (including optimized structures, elec-
tronic and thermal energies, NMR data) please refer to the ESI,
p. S29ff.†
19 J. Sun, Y. Wu, Y. Wang, Z. Liu, C. Cheng, K. J. Hartlieb,
M. R. Wasielewski and J. F. Stoddart, J. Am. Chem. Soc.,
2
015, 137, 13484–13487.
20 A. Hinz, A. Schulz and A. Villinger, Angew. Chem., Int. Ed.,
015, 54, 2776–2779.
1 A. Hinz, A. Schulz and A. Villinger, J. Am. Chem. Soc., 2015,
37, 9953–9962.
2 P. Ravat, T. Šolomek, D. Häussinger, O. Blacque and
M. Juríček, J. Am. Chem. Soc., 2018, 140, 10839–10847.
3 J. Bresien, T. Kröger-Badge, S. Lochbrunner, D. Michalik,
H. Müller, A. Schulz and E. Zander, Chem. Sci., 2019, 10,
Conflicts of interest
2
There are no conflicts to declare.
2
2
2
1
Acknowledgements
We thank the University of Rostock for access to the high-per-
formance computing facilities, and especially Malte Willert for
his assistance with the queueing system and software installa-
tions. Special thanks are due to Peter Kumm and Thomas
Kröger-Badge for their continuous commitment in our insti-
tute’s technical and electrical workshops. Furthermore, we
wish to thank the DFG (SCHU/1170/12-2) for financial support.
3486–3493.
2
2
4 K. Yamaguchi, Chem. Phys. Lett., 1975, 33, 330–335.
5 V. Bachler, G. Olbrich, F. Neese and K. Wieghardt, Inorg.
Chem., 2002, 41, 4179–4193.
2
2
2
6 D. Herebian, K. E. Wieghardt and F. Neese, J. Am. Chem.
Soc., 2003, 125, 10997–11005.
7 M.-M. Russew and S. Hecht, Adv. Mater., 2010, 22, 3348–
3360.
Notes and references
8 J. L. Zhang, J. Q. Zhong, J. D. Lin, W. P. Hu, K. Wu,
G. Q. Xu, A. T. S. Wee and W. Chen, Chem. Soc. Rev., 2015,
44, 2998–3022.
1
H. Grützmacher and F. Breher, Angew. Chem., Int. Ed., 2002,
1, 4006–4011.
4
2
3
F. Breher, Coord. Chem. Rev., 2007, 251, 1007–1043.
M. Abe, J. Ye and M. Mishima, Chem. Soc. Rev., 2012, 41,
29 D. Bléger and S. Hecht, Angew. Chem., Int. Ed., 2015, 54,
11338–11349.
3
808–3820.
30 P. C. Knipe, S. Thompson and A. D. Hamilton, Chem. Sci.,
2015, 6, 1630–1639.
4
5
M. Abe, Chem. Rev., 2013, 113, 7011–7088.
S. González-Gallardo and F. Breher, in Comprehensive 31 E. Niecke, A. Fuchs and M. Nieger, Angew. Chem., Int. Ed.,
Inorganic Chemistry II, Elsevier, 2013, vol. 1, pp. 413–455. 1999, 38, 3028–3031.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
3
3
3
2 M. Abe, E. Kubo, K. Nozaki, T. Matsuo and T. Hayashi, 50 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
Angew. Chem., Int. Ed., 2006, 45, 7828–7831.
3 S. Yoshidomi and M. Abe, J. Am. Chem. Soc., 2019, 141,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Peterson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi,
J. Normand, K. Raghavachari, A. Rendell, J. C. Burant,
S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam,
M. Klene, J. E. Know, J. B. Cross, V. Bakken, C. Adamo,
J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski,
R. L. Martin, K. Morokuma, V. G. Zakrezewski, G. A. Voth,
P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels,
O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and
D. J. Fox, Gaussian 09, Revision E.01, Gaussian, Inc.,
Wallingford CT, 2013.
3
920–3933.
4 H. Beer, J. Bresien, D. Michalik, A.-K. Rölke, A. Schulz,
A. Villinger and R. Wustrack, J. Org. Chem., 2020, acs.
joc.0c00460.
3
5 W. Adam, W. T. Borden, C. Burda, H. Foster,
T. Heidenfelder, M. Heubes, D. A. Hrovat, F. Kita,
S. B. Lewis, D. Scheutzow and J. Wirz, J. Am. Chem. Soc.,
1
998, 120, 593–594.
3
3
3
6 F. Kita, W. Adam, P. Jordan, W. M. Nau and J. Wirz, J. Am.
Chem. Soc., 1999, 121, 9265–9275.
7 D. Y. Zhang, D. A. Hrovat, M. Abe and W. T. Borden, J. Am.
Chem. Soc., 2003, 125, 12823–12828.
8 M. Abe, W. Adam, W. T. Borden, M. Hattori, D. A. Hrovat,
M. Nojima, K. Nozaki and J. Wirz, J. Am. Chem. Soc., 2004,
1
26, 574–582.
3
4
4
4
9 B. J. Guddorf, C. Mück-Lichtenfeld, A. Hepp and F. Lips,
Chem. Commun., 2019, 55, 12896–12899.
0 P. Pyykkö and M. Atsumi, Chem. – Eur. J., 2009, 15, 12770–
2779.
1 G. Mills, H. Jónsson and G. K. Schenter, Surf. Sci., 1995,
24, 305–337.
2 H. Jónsson, G. Mills and K. W. Jacobsen, in Classical and
51 J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett.,
1996, 77, 3865–3868.
1
52 J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett.,
1997, 78, 1396–1396.
3
53 S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem.
Phys., 2010, 132, 154104.
Quantum Dynamics in Condensed Phase Simulations, World 54 S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem.,
Scientific, 1998, pp. 385–404. 2011, 32, 1456–1465.
3 G. Henkelman and H. Jónsson, J. Chem. Phys., 2000, 113, 55 F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005,
978–9985. 7, 3297–3305.
4 G. Henkelman, B. P. Uberuaga and H. Jónsson, J. Chem. 56 C. Riplinger and F. Neese, J. Chem. Phys., 2013, 138,
Phys., 2000, 113, 9901–9904. 034106.
5 E. Maras, O. Trushin, A. Stukowski, T. Ala-Nissila and 57 D. G. Liakos, M. Sparta, M. K. Kesharwani, J. M. L. Martin
4
4
4
4
4
9
H. Jónsson, Comput. Phys. Commun., 2016, 205, 13–21.
6 F. Neese, Wiley Interdiscip. Rev.: Comput. Mol. Sci., 2018, 8,
e1327.
and F. Neese, J. Chem. Theory Comput., 2015, 11, 1525–
1539.
58 C. Riplinger, P. Pinski, U. Becker, E. F. Valeev and F. Neese,
J. Chem. Phys., 2016, 144, 024109.
7 K. Ishida, K. Morokuma and A. Komornicki, J. Chem. Phys.,
1
977, 66, 2153–2156.
59 D. G. Liakos, Y. Guo and F. Neese, J. Phys. Chem. A, 2020,
124, 90–100.
4
4
8 K. Fukui, Acc. Chem. Res., 1981, 14, 363–368.
9 C. Feldmeier, H. Bartling, E. Riedle and R. M. Gschwind, 60 A. Hellweg, C. Hättig, S. Höfener and W. Klopper, Theor.
J. Magn. Reson., 2013, 232, 39–44. Chem. Acc., 2007, 117, 587–597.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.