Facile Access to Challenging ortho-Terphenyls via Merging Two Multi-Step Domino Reactions in One-Pot: A Joint Experimental/Theoretical Study

Article abstract of DOI:10.1002/cctc.201900746

ortho-Terphenyls are of high interest for medicinal chemistry and materials science, but they are difficult to access. Herein, we demonstrate a straightforward and sustainable synthesis of highly functionalized ortho-terphenyls via joining an organocatalyzed two-step domino reaction (Knoevenagel/vinylogous Michael) with a DABCO/CuBr₂ co-catalyzed three-step domino reaction (cyclization/tautomerization/aromatization) in a one-pot process. Overcoming necessity to isolate intermediate products leads to a reduction of energy, costs and waste for a broad scope of reactions. DFT calculations have been performed to investigate the thermodynamics of this one-pot process towards ortho-terphenyls and to study the reaction profile of the vinylogous Michael reaction under inclusion of solvent effects. Role of London dispersion forces in this transformation has been elucidated. It is shown that reaction kinetics and thermodynamics are slightly influenced by dispersion interactions. Furthermore, the addition of dispersion energy donors leads to small changes of reaction energies in some cases.

Full text of DOI:10.1002/cctc.201900746

Accepted Article
Title: Facile Access to Challenging ortho-Terphenyls via Merging Two
Multi-Step Domino Reactions in One-Pot: A Joint Experimental/
Theoretical Study
Authors: Svetlana B. Tsogoeva, Benedikt W. Grau, Simon Bönisch,
Alexander Neuhauser, Frank Hampel, and Andreas Görling
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: ChemCatChem 10.1002/cctc.201900746
Link to VoR: http://dx.doi.org/10.1002/cctc.201900746
A Journal of
www.chemcatchem.org

10.1002/cctc.201900746
ChemCatChem
FULL PAPER
Facile Access to Challenging ortho-Terphenyls via Merging Two
Multi-Step Domino Reactions in One-Pot: A Joint Experimental/
Theoretical Study
Benedikt W. Grau,+[a] Simon Bönisch,+[b] Alexander Neuhauser,[a] Frank Hampel,[a] Andreas Görling,*[b]
and Svetlana B. Tsogoeva*[a]
Abstract: ortho-Terphenyls are of high interest for medicinal
difficult precursor synthesis, or low functionalization of
resulting products.
chemistry and materials science, but they are difficult to access.
Herein, we demonstrate
a
straightforward and sustainable
One of the most elegant synthetic approaches providing a
facile access to highly functionalized molecules is a domino
process in which two or more bond-forming transformations
occur, based on functionalities formed in the previous step.[12]
Domino reactions are a most economical, sustainable and
straightforward type of synthesis, due to decrease of waste
and time compared to a traditional stop-and-go procedure.[13]
In particular, organocatalyzed domino reactions provide
additional advantages, using easily available starting materials
and non-expensive, environmentally friendly catalysts.[14]
Additionally, combination of an organocatalyst with a metal-
based catalyst is also an already established approach and
enables synthesis of novel complex molecules.[15] Combination
of an organocatalyzed domino reaction with a metal-catalyzed
domino reaction in a sequential one-pot procedure may even
further reduce waste, costs and time compared to a classical
route of synthesis.[16]
synthesis of highly functionalized ortho-terphenyls via joining an
organocatalyzed two-step domino reaction (Knoevenagel/
vinylogous Michael) with a DABCO/CuBr2 co-catalysed three-step
domino reaction (cyclization/ tautomerization/ aromatization) in a
one-pot process. Overcoming necessity to isolate intermediate
products leads to a reduction of energy, costs and waste for a
broad scope of reactions. DFT calculations have been performed
to investigate the thermodynamics of this one-pot process towards
ortho-terphenyls and to study the reaction profile of the vinylogous
Michael reaction under inclusion of solvent effects. Role of London
dispersion forces in this transformation has been elucidated. It is
shown that reaction kinetics and thermodynamics are slightly
influenced by dispersion interactions. Furthermore, the addition of
dispersion energy donors leads to small changes of reaction
energies in some cases.
Introduction
Terphenyls, among them ortho-, meta- and para-isomers, are
aromatic hydrocarbons ubiquitous in synthetic bioactive
molecules, natural products and polyaromatic hydrocarbons
(PAHs).[1],[2] In particular, applications of sterically demanding
ortho-terphenyls span from synthetic pharmaceuticals[3] over
key structures in organic light-emitting diodes (OLEDs)[4] to
matrixes for photoinduced formation of optical absorption
anisotropy.[5] Nonetheless, synthesis of ortho-terphenyls
remains challenging due to steric hindrance of two ortho-
standing aromatic rings. Existing synthetic routes towards
ortho-terphenyls include Diels-Alder reactions followed by
aromatization[6] or Ullman-coupling.[7] More recent reports use
Hiyama,[8] Kumada,[9] Suzuki,[10] and Negishi[11] cross-coupling
reactions. Some practical shortcomings of existing synthetic
methods include necessity of expensive metal-catalysts,
[a]
[b]
[+]
B. W. Grau,+ A. Neuhauser, Dr. F. Hampel, Prof. Dr. S. B. Tsogoeva
Institute of Organic Chemistry and Interdisciplinary Center for
Molecular Materials (ICMM), Friedrich-Alexander-Universität
Erlangen-Nürnberg, Nikolaus-Friebiger-Straße 10, 91058 Erlangen,
Germany e-mail: svetlana.tsogoeva@fau.de
S. Bönisch,+ Prof. Dr. A. Görling
Scheme 1. Selected examples of reactions leading to ortho-terphenyl core
structures.
Chair of Theoretical Chemistry and Interdisciplinary Center for
Molecular Materials (ICMM), Friedrich-Alexander-Universität
Erlangen-Nürnberg, Egerlandstraße 3, 91058 Erlangen, Germany
e-mail: andreas.goerling@fau.de
While a few organocatalytic or one-pot approaches towards
meta-terphenyls
have
been
reported,[17]
only
one
These authors contributed equally to this work.
organocatalytic method[3b] and one base catalyzed route[18]
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900746
ChemCatChem
FULL PAPER
towards ortho-terphenyls have been established, leading to
substituted cyano-aniline derivatives (Scheme 1a,b). Even
though these methods provide highly functionalized ortho-
terphenyls, major drawbacks are restricted versatility, small
scope of substrates, poor yields and requirement of complex
starting materials (Scheme 1a,b). For this reason,
development of new sustainable, maximum efficient synthetic
methods towards ortho-terphenyls, starting from simple
precursors, is in high current demand.
Table 1. Screening of reaction conditions.
Entry
1
Base, mol %
TEA, 100
Co-catalyst, mol %
Temperature
50°C
50°C
50°C
R.T.
Yield [%][a]
CuBr2, 15
FeCl3, 15
53
Recently we reported an organocatalyzed three-component
domino Knoevenagel/vinylogous Michael reaction and
2
TEA, 100
traces
traces
18
demonstrated
a potential application of isolated domino
products ,-dicyanoolefins for the synthesis of ortho-
terphenyl derivatives via a base/Cu-co-catalyzed three-step
domino (cyclization/ tautomerization/ aromatization) reaction
(Scheme 1c).[19] Based on these preliminary results, we
envisioned to combine these two domino reactions in a one-
pot process to avoid time-consuming and waste-producing
isolation and purification of product intermediates (Scheme 1c).
Herein we report development of such an efficient and
straightforward one-pot five-step synthetic strategy to access a
broad scope of highly functionalized ortho-terphenyls starting
from easily available compounds.
3
TEA, 100
organocat.[b], 15
CuBr2, 15
CuBr2, 15
CuBr2, 15[c]
CuBr2, 15[d]
CuBr2, 15
CuBr2, 15
CuBr2, 15
CuBr2, 15
CuBr2, 15
CuBr2, 5
4
TEA, 100
5
TEA, 100
70°C
50°C
50°C
50°C
50°C
50°C
50°C
50°C
50°C
50°C
43
6
TEA, 100
14
7
TEA, 100
22
8
TEA, 15
n.c.
n.c.
traces
33
9
DEPA, 100
DMEA, 100
DIPEA, 100
DABCO, 100
DABCO, 100
DABCO, 100
10
11
12
13
14
Results and Discussion
70
Optimization of reaction and substrate scope screening
59
We started our investigation by exploring a selected three-step
domino reaction of ,-dicyanoolefin 1a towards ortho-
terphenyl 2a (Table 1). Initially we applied triethylamine, which
we reported for cyclisation step in our previous work,[19] and
CuBr2 co-catalyst as it was recently demonstrated for efficient
aromatisation of cyclohexene derivatives.[20] To our delight the
domino product was obtained in 53 % yield (Table 1, entry 1).
To better understand the role of the additive, CuBr2 was
exchanged for FeCl3 acting as a Lewis acid (Table 1, entry 2),
Schreiners’ catalyst as H-bond donor (Table 1, entry 3), which
yielded the product only in traces. Further experiments with
CuBr2 also indicate, that the in situ formed copper-amine
complex might be the active catalytic species in the reaction,
as mixing the two co-catalysts (amine and CuBr2) results in a
drastic colour change with formation of a precipitate (Figure 1).
CuBr2, 100
41
Conditions: 0.15 mmol of 1a, CH3CN (0.075M) [a] Isolated yield. [b] Schreiner’s
thiourea-catalyst was used. [c] 4 h reaction time, 10 ml of solvent was used.
[d] under O2 atmosphere, n.c. = no conversion.
Variations of the temperature and reaction time did not result
in an improvement of yield (Table 1, entries 4-6). To our
surprise the reaction under oxygen atmosphere resulted in
further decrease of the yield to 22% (Table 1, entry 7). Another
interesting result was the use of a 1:1 ratio of copper(II)
bromide (15 mol%) and trimethylamine (15 mol%), which
resulted in no conversion to desired product (Table 1, entry 8).
Scheme 2: Further experiments for optimisation of conditions. Conditions:
nitrostyrene (1.00 equiv.), phenylacetaldehyde (1.00 equiv.), malononitrile
(1.00 equiv.), base (0.15 equiv.), solvent (0.3M). [a] Determined by NMR
without isolation. [b] Yield of isolated product.
Figure 1: Preparation of the copper-amine complex.
This may additionally give a hint on formation of a copper-
amine complex: If the free amine was completely bound to the
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900746
ChemCatChem
FULL PAPER
metal, lack of base catalyst (required for cyclization step) may
be responsible for the suppressed reaction. After exchanging
triethylamine by different related tertiary amine bases, DABCO
was found to perform best in the reaction (Table 1, entries 9-
11 vs. entry 12). Variation of CuBr2 loading (Table 1, entries 13
and 14) proved that 15 mol% of CuBr2 worked best for the
reaction.
Ar1
Ar1
NC
NC
NO2
NC
NO2
NO2
Ar
Imidazole
DABCO, CuBr2
NH2
CN
O
Ar2
NO2
CN
1a - RS
Ar2
CN
1a - SS
+
+
Ar
NC
CN
CH2Cl2, r.t., 6 h,
CH3CN, 50°C, 18 h
Ar1
Ar1
3a-w
5a
2a-w
4a
NO2
NC
NO2
CN
1a - SR
Ar2
CN
1a - RR
Ar2
without isolation
F
NO2
NO2
NO2
NO2
NO2
NH2
NO2
NH2
NH2
NH2
CN
NH2
NH2
CN
CN
CN
CN
CN
2a
2b
2c
2d
2e
2f
42% yield[a]
28% yield[c]
32% yield[a]
23% yield[d]
61% yield[a]
31% yield[a]
37% yield[a]
14% yield[a]
Cl
NO2
NO2
Cl
Cl
Br
F3C
NO2
NO2
NO2
NO2
NO2
NH2
NH2
NH2
NH2
NH2
NH2
CN
CN
CN
CN
CN
CN
2g
2h
2i
2j
2k
2l
38% yield[a]
31% yield[a]
30% yield[a]
13% yield[a]
51% yield[a]
22% yield[a]
O
O
NC
O
O
N
NO2
O
NO2
NO2
NO2
NO2
NH2
NO2
NH2
NH2
NH2
NH2
NH2
CN
CN
CN
CN
CN
CN
2m
20% yield[a]
2n
18% yield[a]
2o
2p
2q
2r
32% yield[a]
39% yield[a]
45% yield[a]
21% yield[a]
N
N
NO2
NO2
NO2
NH2
NO2
NH2
NO2
O
S
NH2
NH2
CN
N
NH2
N
CN
CN
CN
CN
2s
18% yield[a]
2t
29% yield[b]
2u
22% yield[b]
2v
24% yield[a]
2w
24% yield[b]
X-ray
structure of 2d
X-ray
structure of 2k
X-ray
structure of 2u
Scheme 3: Substrate scope including variations of the nitrostyrene 3a-w. Conditions of the five-step one-pot process: nitrostyrene (1.00 equiv.),
phenylacetaldehyde (1.00 equiv.), malononitrile (1.00 equiv.), imidazole (0.15 equiv.), DABCO (1.00 equiv.), CuBr2 (0.15 equiv.). Substrates 3-5: [a] 1.00 mmol,
[b] 0.50 mmol, [c] 12.5 mmol, [d] 50.0 mmol.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900746
ChemCatChem
FULL PAPER
To check, if the one-pot procedure can be simplified further by
using either same solvent or same catalyst in both reactions,
we performed the first domino reaction separately with altered
the five-step one-pot process showed again the high functional
group tolerance of this procedure, even though the reaction
seems to be more sensitive to functionalization of the
phenylacetaldehyde, especially for derivatives bearing an
EWG or EDG functional group (e.g., compounds 7d and 7e).
To expand the scope even further, we used cyanoacetic ester
5b for Knoevenagel condensation to obtain corresponding
product 7f (bearing an ester functional group) in 30% yield
(Scheme 4).
conditions
(Scheme
2).
However,
exchange
of
dichloromethane by acetonitrile resulted in a decreased yield
of 1a (<60%) (in contrast to previously reported 82% yield for
1a[19]) and formation of a complex mixture of byproducts, which
could not be separated from the desired dicyanoolefin. By
employing DABCO instead of imidazole, larger amounts of
Michaeladdition product of malononitrile and nitrostyrene is
formed compared to the reaction performed with imidazole as
catalyst. Additionally, a decomposition of malononitrile takes
place, which makes DABCO an unsuitable catalyst for the first
domino reaction.
Having now the optimized conditions in hand, we studied the
scope of reaction (Scheme 3). As the nitrostyrene building
block is an easily available substrate, we demonstrated the
broad functional group tolerance, including groups with +I (2b
and 2c), -I (2f-2k), -M (2l-2n) and +M (2o-2r) effect, expanded
π-systems (2d and 2e), heterocycles (2u-2w) and sterically
demanding carbazole moieties (2s and 2t). Only aliphatic
nitroolefins, like tertbutyl nitrostyrene, did not yield the desired
product. All functional groups were tolerated in the reaction,
even though no clear trend depending on the substitution
pattern can be found. For example, highest yield was obtained
for 2c, the nitrostyrene derivative with two tert-butyl groups,
which should be deactivated as Michael-acceptor due to the +I
effect of the group. Another surprising result was the low yield
of the bromo-derivative 2j and the fluoro-derivative 2f
compared to the trifluoro- and the p-chloro derivative, as all of
these halogen derivatives should give corresponding products
in comparable yields. To investigate influence of the
substitution patterns on the phenylacetaldehydes to the
reaction, derivatives 4a-4e have been synthesized and applied
(Scheme 4).
Post-modifications of ortho-terphenyls
As synthesized ortho-terphenyl derivatives bear three
functional groups, we performed postmodifications on selected
derivative 2a to demonstrate versatility of the obtained domino
products (Scheme 5). By reducing the nitro group to the
corresponding amine selectively, employing iron(III)chloride
and zinc, absorbance of the molecule changed drastically.
Resulting diamine 11 even exhibits solid state fluorescence at
366 nm wavelength and a weak solvatochromic effect as
depicted in Scheme 5d. The amino and cyano function in ortho
position additionally allow transformation to
a
novel
quinazoline 12. Quinazoline derivatives possess a broad
variety of biological activities and are therefore used in drugs,
featuring
antimicrobial,
anti-inflammatory,
diuretic,
anticonvulsant, antiallergic, antihypertensive, antiparkinsonian,
and anticancer activities.[17d, 21] Novel synthesized quinazoline
derivative 12 will be investigated towards its potential
biological activities and the results will be reported elsewhere.
Scheme 5: Postmodification of the unsubstituted ortho-terphenyl 2a, [a]
FeCl3/Zn, DMF/water, r.t.; [b] formamide, 220°C, [c] demonstration of the
solid state fluorescence and [d] solvatochromic properties of 11 at 366 nm
(solvents: Hexane, Toluene, DCM, CHCl3, CH3CN, THF, Et2O, Acetone,
MeOH).
DFT Study on the thermodynamics of the one-pot process
and the reaction profile of the vinylogous Michael reaction
We investigated thermodynamics of the one-pot synthesis and
kinetics of the vinylogous Michael reaction by DFT calculations
using the PBE functional.[22] and the def2-TZVP basis set.[23]
The effects of the solvents (DCM, ε=8.93[24] and acetonitrile
ε=36.64[24]) on both energies and geometries were included
using the conductor-like screening (COSMO) method.[25]
Furthermore we investigated the influence of dispersion
interactions on this one-pot synthesis by applying or not
applying the semi-empirical D3 dispersion interaction
correction developed by Grimme and coworkers[26] including
Scheme 4: Substrate scope including variations of phenylacetaldehyde 4a-
e
and malononitrile 5a-b. Conditions: nitrostyrene (1.00 equiv.),
phenylacetaldehyde (1.00 equiv.), malononitrile (1.00 equiv.), imidazole
(0.15 equiv.), DABCO (1.00 equiv.), CuBr2 (0.15 equiv.). Substrates 3-5: [a]
1.00 mmol.
All these starting materials are easily available, as they can be
synthesized by oxidizing the corresponding alcohol with Dess-
Martin periodinane. Employing these phenylacetaldehydes for
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900746
ChemCatChem
FULL PAPER
Becke-Johnson dampening.[27] All calculations were performed
with the TURBOMOLE program package.[28] See Supporting
Information for additional computational details.
We first studied imidazole catalyzed reactions leading to 1a, in
DCM (see Scheme 6 for nomenclature). During this catalytic
cycle, first phenylacetaldehyde 4a and malononitrile 5a form
the Knoevenagel intermediate 8 in a reaction with a Gibbs
reaction energy of -3.5 kcal mol-1. Intermediate 8 reacts with
trans-β-nitrostyrene 3a in a vinylogous Michael reaction to
intermediate 1a. The intermediate 1a has two different
diastereomers, each having two enantiomers. For the following
investigations, we chose the 1a-SR and 1a-RR diastereomer.
Scheme 6: Proposed mechanistic cycles for the synthesis of ortho-terphenyls via merging two domino reactions in one-pot.
All results and conclusions for the 1a-SR apply equally to the
1a-RS diastereomer and all results and conclusions for the 1a-
RR diastereomer apply equally to the 1a-SS diastereomer. 1a-
SR and 1a-RR are formed with a Gibbs energy of reaction of -
1.9 and -0.9 kcal mol-1, respectively (see Figures 2b, 3b, 4a).
1a-SR is therefore slightly more stable by ~1 kcal mol-1 than
1a-RR in agreement with the experimentally found ratio of
83:17.[19]
In order to study the role of the kinetics on the observed ratio
we studied the vinylogous Michael reaction (8+3a to 1a), which
is the only possible diastereoselective step in imidazole
catalyzed formation of 1a. We calculated for both the formation
of the SR (see Figure 2a for optimized structures and 2b for
the reaction profile) as well as the RR diastereomer (see
Figure 3a for structures and 3b for the reaction profile) low
lying transition states (TS) (see supporting information for
details) for the deprotonation of 8, C-C-bond formation yielding
deprotonated 1a and the protonation step to 1a.
The deprotonation starts with the formation of a reactant
complex (rc1) consisting of imidazole and 8. This formation is
endergonic due to entropic effects (reduction of particle
number). There exist two enantiomeric possibilities for this
complex (attack from above and below the plane formed by 8).
One leads to 1a-SR, the other to 1a-RR. Alternative TSs,
where also the nitrostyrene is present in the TS as a van der
Waals complex, were found to be less stable due to entropic
effects (see supporting information for pictures of these TSs).
Since the activation barrier is rather modest with only 3.2 kcal
mol-1, these two enantiomers, present as a racemic mixture,
can easily interconvert. Furthermore, a shift of the imidazolium
cation from below to above the plane formed by deprotonated
8 can also interconvert these to structures. Therefore, this step
does not play a role for the observed diastereomeric ratio of
1a-SR and 1a-RR. The resulting intermediate of this reaction
(int1), consisting of deprotonated 8 and the imidazolium cation,
forms together with nitrostyrene 3a the reactant complex of the
C-C bond formation (rc2). Complexes for SR (rc2-SR, see
Figure 2) and RR (rc2-RR, see Figure 3) are diastereomeric,
preforming diastereomers 1a-SR and 1a-RR, and have
therefore slightly different energies. Reaction of rc2 to the
corresponding complex consisting of deprotonated 1a and
imidazolium cation (int2) via TS2 is endergonic for both
diastereomers, although the complex formed by RR-
diastereomer (int2-RR) is 2.55 kcal mol-1 more stable than that
of the SR-diastereomer. Furthermore, the activation barrier is
1.4 kcal mol-1 lower and the TS 0.7 kcal mol-1 more stable in
case of the 1a-RR diastereomer. 1a-RR is therefore both
thermodynamically as well as kinetically favored at this stage.
The reason for this might be, that in the lowest lying TS,
phenyl substituents are in syn position towards each other.
This conformation is the most stable for 1a-RR (see Figure 4b).
The lowest lying conformer of 1a-SR, on the other hand, has
the phenyl substituents in anti position towards each other and
the syn position is disfavored for this diastereomer. The
favorable motif found in the TS2-SR, with imidazole bridging
the reactants (see Figure 2a), is not possible in the anti-
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900746
ChemCatChem
FULL PAPER
conformer, due to the nitro- and dicyanoethenyl group pointing
in opposite directions, resulting in the lowest lying TS being
TS2-SR, despite its less favorable positioning of the phenyl
groups (See supporting information for details).
thermodynamically, due to the more stable pc-RR. The lowest
lying conformer of 1a-SR is more stable than the most stable
conformer of 1a-RR, though.
The use of an augmented basis set, the aug-cc-pVTZ basis
set,[29] which could potentially describe the anionic species
involved in the reaction better, has only a minor effect on the
energetics (See SI for additional details).
In case of protonation of deprotonated 1a towards the complex
between 1a and imidazole, the product complex (pc) of the
reaction, TSs (“TS3”) for both diastereomers have been
obtained (see Figure 2 and 3). The corresponding reactant
complex (rc3) has a lower SCF-energy (1.2/0.8 kcal mol-1 for
SR/RR, respectively) than the corresponding TS, but a higher
Gibbs energy. It is unclear whether this is due to inaccuracies
of the calculation of the chemical potential and the protonation
has a small barrier in reality, or, if this transition state does not
play a role in this reaction and rearrangement of int2 to rc3
leads directly to pc without an activation barrier. Compared to
int2, rc3 has a 7.7/9.1 kcal mol-1 (for SR/RR, respectively)
higher Gibbs energy and the barrier is therefore lower in case
of the SR diastereomer. On the other hand, rc3-RR is 1.2 kcal
mol-1 more stable than rc3-SR. Furthermore, the product
complex (pc) is also more stable in case of RR diastereomer
by 2.0 kcal mol-1
The influence of the functional on the results presented was
tested by recalculating the reaction profiles using the B3LYP
functional[30] (see SI for computational details). The Gibbs
energies of all species were shifted by several kcal mol-1
compared to the reactants. This shift was stronger in case of
the zwitterionic species and even stronger in case of the TSs
(the largest shift occurred in case of TS3-SR with 6.9 kcal mol-
1). Interestingly for the B3LYP functional the protonation (TS3)
has a small activation barrier (~1 kcal mol-1) compared to the
PBE functional, where rc3 is slightly higher in energy than TS3
and the protonation therefore had no activation barrier.
The overall barrier compared to 8 and 3a is 22.5 kcal mol-1 in
case of the SR and 20.6 in case of the RR diastereomer
(compared to 17.0/15.8 kcal mol-1 in case of PBE). While the
overall barrier is higher, the difference between SR and RR
(1.9 kcal mol-1 for B3LYP; 1.2 kcal mol-1 for PBE) is very
similar. The same is true for other differences between SR and
RR. (See Supporting information for reaction profiles using the
B3LYP functional). Altogether, the reaction profiles are
qualitatively very similar and more importantly, the differences
between the SR and RR diastereomer, and therefore the main
result of these calculations remain quantitatively the same
(difference smaller than 1 kcal mol-1)
Dispersion interactions strongly lower energies of reactant
complexes, intermediates and TSs, compared to the reactants.
Furthermore, all intermediates and TSs starting from rc2 are
stabilized by dispersion interactions, compared to species
involved in the initial deprotonation (compare red and blue
profiles in Figure
2 and 3). Activation barrier of the
deprotonation is lowered by ~1 kcal mol-1. In case of the C-C
bond formation (rc2 to int2), RR-diastereomer is stabilized ~1
kcal mol-1 more than the SR-diastereomer by dispersion
interactions. Furthermore, TS2 is slightly more stable in case
of the SR-diastereomer without dispersion interactions, and
therefore, formation of 1a-SR is slightly faster, compared to the
opposite case with dispersion interactions.
Encouraged by these findings, we performed a time resolved
NMR-experiment to investigate if an initial formation of the RR
diastereomer is indeed observable. As the two diastereotopic
protons (Figure 5, 1a-SR/RS highlighted in red, 1a-RR/SS
highlighted in blue), which give signals in the range of 4.9 –
4.3 ppm, are easily distinguishable, diastereomeric ratio can
be determined by integration. The data obtained is depicted in
Figure 5 displaying an excess of the kinetically favored RR-
Altogether, it is not clear, what the rate determining step is as
the barrier between rc2 and TS2 and int2 and rc3 is of
comparable height. The highest point of the profile is rc3 with
an overall barrier compared to 8+3a of 17.0 kcal mol-1 in case
of the SR diastereomer and 15.8 kcal mol-1 in case of the RR
diastereomer Altogether it seems, that RR diastereomer is
favored over SR diastereomer kinetically and – initially - also
diastereomer
in
the
beginning,
turning
into
the
thermodynamically favored SR-diastereomer (Figure 5).
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900746
ChemCatChem
FULL PAPER
Figure 2: a) Optimized structures involved in and b) Reaction profile (ΔG; in kcal mol-1) of the vinylogous Michael reaction yielding 1a-SR with (blue) and without
(red) dispersion interactions; solvent: DCM (ε=8.93); energies relative to 3a+4a+5a+imidazole; energies relative to previous species in parenthesis, in case of the
RS diastereomer all energies would be identical to and all structures would be exact mirror images structures of the SR diastereomer. rc=reactant complex;
TS=transitions state; int=intermediate complex; pc=product complex.
Overall, it seems, that 1a-RR is initially kinetically favored over
the SR-diastereomer but is converted to thermodynamically
more stable 1a-SR over time.
This conversion can occur via two possible pathways. One
possibility is a back reaction using the same path as formation
of 1a (Figure 2 and 3). Alternatively, proton adjacent to the
dicyanoethenyl group can be deprotonated by imidazole,
because it is an allylic proton, and therefore converts 1a-RR to
1a-SR via the anion. We so far investigated formation of 1a in
DCM in detail. This formation is the first cycle of the one-pot
procedure (see Scheme 6 and Figure 4a). At this point solvent
for this intermediate is switched from DCM (1a-XR, X=S,R) for
acetonitrile (1a’-XR, X=S,R ), the energy difference between
1a-XR and 1a’-XR is the difference of the solvation energy
between DCM and acetonitrile of these intermediates (see
Figure 4a).
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900746
ChemCatChem
FULL PAPER
Figure 3: a) Optimized structures involved in and b) Reaction profile (ΔG; in kcal mol-1) of the vinylogous Michael reaction yielding 1a-RR with (blue) and without
(red) dispersion interactions; solvent: DCM (ε=8.93); energies relative to 3a+4a+5a+imidazole, energies relative to previous species in parenthesis, in case of the
SS diastereomer all energies would be identical and all structures would be exact mirror images of the RR diastereomer. rc=reactant complex, TS=transitions
state, int= intermediate complex, pc= product complex.
Under the reaction conditions of the second domino reaction
(CuBr2, DABCO) intermediate 1a’ undergoes a cyclization to 9
with a reaction energy of -4.7 kcal mol-1 for the SR and -7.3
kcal mol-1 for RR diastereomer. Unlike to 1a, in case of 9, RR
diastereomer is 1.9 kcal mol-1 more stable than the SR-
diastereomer. It should be noted that 9 has an additional
stereocenter. Structures shown in Figure 4b and the
corresponding energies refer to the most stable conformer of
the more stable diastereomer (the SRR and RRS diastereomer,
see Figure 4b) obtained during the conformer search (See SI
for details). Imine-enamine tautomerization of 9 leads to the
15.3/14.1 kcal mol-1 (SR/RR diastereomer, respectively)
enamine 10, losing the previously formed stereocenter in the
process. Here again 10-RR is slightly (0.6 kcal mol-1) more
stable than 10-SR.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900746
ChemCatChem
FULL PAPER
Figure 4: a) Gibbs energies of reaction (ΔG; in kcal mol-1) with (blue) and without (red) dispersion interactions relative to 3a+4a+5a+O2 in DCM; energies relative
to previous species in parenthesis. For 8+3a and 1a-XR (X=S, R) the solvent is DCM, for 1a’-XR and compounds 9,10 and 2a the solvent is acetonitrile; b)
structures of the intermediates; in case of the RS and SS diastereomer all energies would be identical to and all structures would be exact mirror images of the
SR and RR diastereomer, respectively.
Higher stability of the RR-diastereomers for 9 and 10 likely
derives again from the finding, that the lowest lying conformer
of 1a-RR has the two phenyl substituents already in syn
position towards each other while most stable conformer of 1a-
SR has the phenyl groups in anti position towards each other
(see Figure 4b). As last step, both diastereomers yield same
product 2a via aromatization with oxygen as oxidation agent,
leading to formation of water. This step is the most exergonic
step, with a Gibbs energy of reaction of -57.9/-57.3 kcal mol-1
(SR/RR diastereomer, respectively). Without dispersion
interactions, formation of 1a turns endergonic. Furthermore,
reaction 1a to 9 is more exergonic, especially for reaction 1a-
SR to 9-SR (ΔΔG=2.2 kcal mol-1), while the aromatization (10
to 2a) is less exergonic with dispersion interactions, since 2a is
less stabilized by dispersion interactions than 10 (stabilization
of 9.6 kcal mol-1 for 2a and ~11 kcal mol-1 for 10) (See Figure
4a).
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900746
ChemCatChem
FULL PAPER
Ha + Hb
Hc
Hd
115 min
105 min
95 min
85 min
75 min
65 min
55 min
45 min
35 min
25 min
15 min
5 min
ppm
4.86 4.84 4.82 4.80 4.78 4.76 4.74 4.72 4.70 4.68 4.66 4.64
4.48 4.46 4.44 4.42
Figure 5: Time-resolved NMR investigation of the two-step domino reaction. Conditions: nitrostyrene (1.00 equiv.), phenylacetaldehyde (1.00 equiv.),
malononitrile (1.00 equiv.), imidazole (0.15 equiv), CD2Cl2 (0.3M).
Influence of dispersion energy donors (DEDs[31]) on the
overall reaction energies of the one-pot reaction.
of 2a is more exergonic in all cases but derivative 17. Only in
case of 13, 14, 16, and, to a lesser degree- 20 addition of
DEDs showed a noticeable stabilizing effect, though.
DEDs also influence energetics of reaction 1(R1) and reaction
2(R2) (see Table 2). Most notable is the significantly less
exergonic R1 (in fact R1 is slightly endergonic for 14 and 16)
and the significantly more exergonic R2 in case of 14 and 16
and the more exergonic R1 and R2 in case of 13.
Without taking dispersion into account (see Table S1 in the
supporting information), only 14 is slightly (0.1/0.3 kcal mol−1
for R1+R2 of SR/RR, respectively) more stable than 2a while
formation of derivatives of 2a is less exergonic in all other
cases. Most prominent changes in case of R1 and R2 when
neglecting dispersion interactions are a less favorable R1 and
R2, compared to 2a for 13 and to a lesser degree for 20, and
an even more disfavored R1 compared to 2a in case of 14 and
16.
ortho-Terphenlys have two aryl groups adjacent to each other.
They therefore could experience a thermodynamic stabilization
and lead to a potentially higher yield of the one-pot procedure
by addition of DEDs to the phenyl groups. We therefore
investigated eight potential derivatives, featuring bulkier
substituents (13-20; see Figure 6b for structure formulas and
the supporting information for pictures of optimized structures).
We calculated reaction energies of the formation of the
derivative of 1a (reaction 1 (R1); also see Figure 6a) and the
overall reaction energy of the cyclization/aromatization (1a to
2a) for the derivatives 13-20 (reaction 2 (R2); also see Figure
6a) again using the PBE functional with dispersion interaction
correction. Solvent effects are included via the COSMO model
((R1): DCM, ε=8.93; (R2): acetonitrile ε=36.64)
Gibbs energies of reaction for all derivatives compared to the
respective Gibbs energies of reaction of 2a (including
dispersion interactions), are shown in Table 2, for results
without dispersion interactions see Table S1 of the supporting
information.
We first consider the overall Gibbs energy of reaction (R1+R2)
(see Table 2). In case of 15 and 17-19, the addition of DEDs
has only a minor effect, although formation of the derivatives
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900746
ChemCatChem
FULL PAPER
which is less the case in the product 14 and 16. Overall, the
most promising candidates are 13 and 14.
Conclusions
In summary, we demonstrated that synthetically challenging
ortho-terphenyls can easily be prepared under mild conditions
via merging an organocatalyzed two-step domino reaction with
a DABCO/CuBr2 co-catalysed three-step domino reaction in a
one-pot process. Tolerance of this new environmentally
compatible protocol to various substrates (nitrostyrenes and
arylacetaldehydes) was demonstrated. The DFT study on this
one-pot procedure revealed, that the C-C-bond formation
(activation energy of 7.0/5.8 kcal mol−1 for SR/RR,
respectively) and the subsequent protonation (activation
energy of 7.7/9.1 kcal mol−1 for SR/RR, respectively) have
similar barriers. The overall barrier is 15.8 kcal mol−1 in case of
the RR diastereomer and 17.0 kcal mol−1 in case of the SR-
diastereomer. Therefore, overall, the RR diastereomer is
kinetically favored over the SR diastereomer, which is the
thermodynamically favored diastereomer though. In the three-
step domino reaction the RR-diastereomers were found to be
thermodynamically favored. It was revealed, that dispersion
interactions slightly influence reaction kinetics and
thermodynamics of the one-pot procedure. Addition of DEDs
leaded to small changes in reaction energies in some cases.
The obvious advantages of the developed process are easily
available and inexpensive reagents and high atom efficiency.
Figure 6: a) Definition of reaction 1 (R1) and reaction 2 (R2) b) Studied
potential ortho-terphenyls featuring DEDs. DED = Dispersion Energy Donor.
Table 2. Gibbs energies of reaction (ΔG) relative to the corresponding
Gibbs energies of reaction (ΔG(2a)) of ortho-terphenyl product 2a in kcal
mol-1, using PBE with dispersion interactions; Species 1 corresponds to
derivative of 1a leading to respective products 13-20).
Experimental Section
R1+
R2
(RR)
ortho-ter-
phenyls
R1 to
1-SR
R1 to
1-RR
R2 from
1-SR
R2 from
1-RR
R1+R2
(SR)
Materials
2a
13
14
15
16
17
18
19
20
0.00
-1.29
6.46
-0.61
6.25
0.39
0.15
-0.19
-0.54
0.00
-1.00
7.09
0.54
6.47
0.24
0.80
0.49
-0.84
0.00
0.00
0.00
0.00
For detailed synthetic procedures and 1H, 13C NMR, infrared- and
mass spectroscopic data of all synthesized compounds, see the
Supporting Information.
-1.78
-11.90
-0.04
-8.10
0.16
-2.62
-12.56
-0.83
-8.79
-0.13
-1.25
-0.67
-0.06
-3.07
-5.43
-0.65
-1.85
0.56
-3.62
-5.47
-0.29
-2.32
0.11
Supplementary crystallographic data
CCDC 1909148 (2d), CCDC 1909149 (2k) and CCDC 1909147 (2u)
contain the supplementary crystallographic data for this paper. These
data are provided free of charge by The Cambridge Crystallographic
Data Centre via http://www.ccdc.ac.uk/data request/cif.
-0.35
-0.15
-0.57
-0.20
-0.34
-1.11
-0.45
-0.18
-0.90
Acknowledgements
S.B.T. and A.G. gratefully acknowledge the financial support
by grants from Deutsche Forschungsgemeinschaft TS 87/15-1
and Priority Programme „Control of London Dispersion
Interactions in Molecular Chemistry“ (SPP 1807): TS 87/17-1
and GO 523/16-1. S.B.T. also thanks the Interdisciplinary
Center for Molecular Materials (ICMM) and the Research
Network “Solar Technologies go Hybrid” (SolTech) - an
Solvent effects included via COSMO (R1: DCM, ε=8.93; R2: acetonitrile
ε=36.64); energies for the RS and SS diastereomer would be identical to the
SR and RR diastereomer, respectively.
“Rn (n=1,2) to/from 1-XR (X=S,R)” corresponds to Rn to/from the derivative
of 1a-XR. “R1+R2(XR)” is the sum of “R1 to 1-XR” and “R2 from 1-XR”
In case of 16, also R2 is less favored compared to 2a when
neglecting dispersion. In case of 14 and 16, repulsion seems
to dominate interactions in the corresponding derivative of 1a,
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900746
ChemCatChem
FULL PAPER
Frohlich, F. Hampel, A. Kahnt, C. Hutterer, M. Steingruber, H. Bahsi,
C. von Bojnicic-Kninski, D. S. Mattes, T. C. Foertsch, A. Nesterov-
Mueller, M. Marschall, S. B. Tsogoeva, Nat. Commun. 2017, 8,
15071.
initiative of the Bavarian State Ministry for Science, Research
and Art for generous research support and funding.
Keywords: domino reactions • DFT • multicomponent
[18] F. V. Singh, A. Parihar, S. Chaurasia, A. B. Singh, S. P. Singh, A. K.
Tamrakar, A. K. Srivastava, A. Goel, Bioorg. Med. Chem. Lett. 2009,
19, 2158-2161.
reactions • one-pot reactions • organocatalysis • o-terphenyls
[19] D. Grau, B. W. Grau, F. Hampel, S. B. Tsogoeva, Chem. Eur. J. 2018,
24, 6551-6556.
[1]
[2]
[3]
G. R. Ames, Chem. Rev. 1958, 58, 895-923.
J. K. Liu, Chem. Rev. 2006, 106, 2209-2223.
[20] V. Shirinian, A. Kavun, A. Lvov, I. Zavarzin, M. Krayushkin, Synthesis
2016, 49, 1255-1263.
a) K. Affleck, P. J. Sidebottomc, P. J. Sidebottom, N. L. Taylor, C. S.
Drake, M. Todd, A. Jowett, G. Webb, J. Antibiot. 1999, 52, 89-95; b)
C. G. Yan, Q. F. Wang, X. K. Song, J. Sun, J. Org. Chem. 2009, 74,
710-718.
[21] R. Ovadekova, S. Jantova, M. Theiszova, J. Labuda, Biomed. Pap.
2005, 149, 455-459.
[22] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77,
3865-3868.
[4]
[5]
a) A. V. Mishra, D. C. Hasija, V. R. Patil, ChemistrySelect 2018, 3,
13665-13669; b) C. W. Lee, J. Y. Lee, Org. Electron. 2013, 14, 1602-
1607.
[23] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297-
3305.
S. Y. Grebenkin, B. V. Bol'shakov, J. Photochem. Photobiol. 2006,
184, 155-162.
[24] J. R. Rumble, Permittivity (Dielectric Constant) of Liquids" in CRC
Handbook of Chemistry and Physics, 99th Edition (Internet Version),
ed., CRC Press/Taylor & Francis, Boca Raton, FL., 2018.
[25] A. Klamt, G. Schüürmann, J. Chem. Soc., Perkin Trans. 2 1993, 799-
805.
[6]
[7]
K. Alder, J. Haydn, Liebigs Ann. Chem. 1950, 570, 201-213.
a) C. H. R. J. Bushby, J. Chem. Soc. Perkin Trans. I 1984, 4; b) W. E.
Bachmann, H. T. Clarke, J. Am. Chem. Soc. 1927, 49, 2089-2098.
a) D. H. Lee, J. Y. Jung, M. J. Jin, Chem. Commun. 2010, 46, 9046-
9048; b) H. Yorimitsu, Y. Yamamoto, A. Baralle, A. Godefroy, K.
Murakami, A. Osuka, Heterocycles 2017, 95, 568.
Z. Xiong, X. Li, S. Zhang, Y. Shi, H. Sun, Organometallics 2016, 35,
357-363.
[8]
[9]
[26] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010, 132,
154104.
[27] S. Grimme, S. Ehrlich, L. Goerigk, J. Comput. Chem. 2011, 32, 1456-
1465.
[28] Turbomole 6.6 and 7.3, a development of University of Karlsruhe and
Forschungszentrum Karlsruhe GmbH, 1989, Turbomole GmbH, since
2007; available from www.turbomole.com.
[10] a) T. Tu, Z. Sun, W. Fang, M. Xu, Y. Zhou, Org. Lett. 2012, 14, 4250-
4253; b) P. K. Mandali, D. K. Chand, Catal. Commun. 2013, 31, 16-
20.
[29] a) T. H. Dunning, J. Chem. Phys. 1989, 90, 1007-1023; b) R. A.
Kendall, T. H. Dunning, R. J. Harrison, J. Chem. Phys. 1992, 96,
6796-6806.
[11] R. C. Borner, R. F. W. Jackson, J. Chem. Soc., Chem. Commun.
1994, 845.
[12] a) L. F. Tietze, Chem. Rev. 1996, 96, 115-136; b) L. F. Tietze,
Domino Reactions: Concepts for Efficient Organic Synthesis, Vol.
Wiley-VCH, Weinheim, 2014.
[30] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648-5652; b) P. J.
Stephens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J. Phys.
Chem. 1994, 98, 11623-11627.
[13] a) H. C. Guo, J. A. Ma, Angew. Chem. Int. Ed. 2006, 45, 354-366; b)
H. Pellissier, Tetrahedron 2006, 62, 1619-1665; c) H. Pellissier,
Tetrahedron 2006, 62, 2143-2173; d) K. C. Nicolaou, D. J. Edmonds,
P. G. Bulger, Angew. Chem. Int. Ed. 2006, 45, 7134-7186; e) C.
Frost, C. Chapman, Synthesis 2007, 2007, 1-21; f) C. Vaxelaire, P.
Winter, M. Christmann, Angew. Chem. Int. Ed. 2011, 50, 3605-3607;
g) C. M. Bock, G. Parameshwarappa, S. Bonisch, C. Neiss, W. Bauer,
F. Hampel, A. Gorling, S. B. Tsogoeva, Chem. Eu. J. 2016, 22, 5189-
5197; h) C. M. Bock, G. Parameshwarappa, S. Bonisch, W. Bauer, C.
Hutterer, M. Leidenberger, O. Friedrich, M. Marschall, B. Kappes, A.
Gorling, S. B. Tsogoeva, ChemistryOpen 2017, 6, 364-374.
[14] a) T. Chanda, J. C. G. Zhao, Adv. Synth. Catal. 2018, 360, 2-79; b) M.
Rueping, I. Atodiresei, Comprehensive Chirality 2012, 6, 345-373; c)
R. Marcia de Figueiredo, M. Christmann, Eur. J. Org. Chem. 2007,
2007, 2575-2600; d) C. Grondal, M. Jeanty, D. Enders, Nat. Chem.
2010, 2, 167-178; e) E. Marques-Lopez, R. P. Herrera, M.
Christmann, Nat. Prod. Rep. 2010, 27, 1138-1167; f) M. Rueping, J.
Dufour, F. R. Schoepke, Green Chem. 2011, 13, 1084; g) D. Enders,
C. Grondal, M. R. Huttl, Angew. Chem. Int. Ed. 2007, 46, 1570-1581.
[15] a) N. T. Patil, Angew. Chem. Int. Ed. 2011, 50, 1759-1761; b) A. S.
Hashmi, C. Hubbert, Angew. Chem. Int. Ed. 2010, 49, 1010-1012; c)
L. Albrecht, H. Jiang, K. A. Jorgensen, Angew. Chem. Int. Ed. 2011,
50, 8492-8509.
[31] a) S. Grimme, R. Huenerbein, S. Ehrlich, ChemPhysChem 2011, 12,
1258-1261; b) J. P. Wagner, P. R. Schreiner, Angew Chem Int Ed
Engl 2015, 54, 12274-12296.
[16] a) N. T. Patil, V. S. Shinde, B. Gajula, Org. Biomol. Chem. 2012, 10,
211-224; b) T. Yang, A. Ferrali, L. Campbell, D. J. Dixon, Chem.
Commun. 2008, 2923-2925.
[17] a) D. Xue, J. Li, Z. T. Zhang, J. G. Deng, J. Org. Chem. 2007, 72,
5443-5445; b) S. Bhattacharjee, A. T. Khan, Tetrahedron Lett. 2016,
57, 2994-2997; c) A. Zonouzi, Z. Izakian, K. Abdi, S. W. Ng, Helv.
Chim. Acta 2016, 99, 355-360; d) F. E. Held, A. A. Guryev, T.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201900746
ChemCatChem
FULL PAPER
Entry for the Table of Contents
FULL PAPER
B. W. Grau, S. Bönisch, A. Neuhauser,
F. Hampel, A. Görling,* S. B. Tsogoeva*
Page No. – Page No.
Facile Access to Challenging ortho-
Terphenyls via Merging Two Multi-
Step Domino Reactions in One-Pot: A
Joint Experimental/ Theoretical Study
A straightforward synthesis of difficult to access ortho-terphenyls via joining two-
step and three-step domino reactions in a one-pot process, is presented.
Overcoming necessity to isolate intermediate products leads to a reduction of
energy, costs and waste for a broad scope of reactions. DFT calculations have
been performed to investigate this facile five-step one-pot process.
This article is protected by copyright. All rights reserved.