Organometallic titanocene complex as highly efficient bifunctional catalyst for intramolecular Mannich reaction

Article abstract of DOI:10.1002/aoc.4925

Bifunctional catalysts bearing two catalytic sites, Lewis acidic organometallic titanocene and Br?nsted acidic COOH, have been assembled in situ from Cp₂TiCl₂ with carboxylic acid ligands, showing high catalytic activity over an intramolecular Mannich reaction towards synthesis of 2-aryl-2,3-dihydroquinolin-4(1H)-ones. The determination of the bifunctional catalyst Cp₂Ti(C₈H₄NO₆)₂ was elucidated by single X-ray HR-MS and investigation of catalytic behavior. In particular, masking the Br?nsted acidic COOH catalytic site with dormant COOMe lowered the reaction yield greatly, indicating that two catalytic sites work together to maintain high catalytic efficiency.

Full text of DOI:10.1002/aoc.4925

Received: 22 January 2019
DOI: 10.1002/aoc.4925
Revised: 28 February 2019
Accepted: 7 March 2019
C O M M U N I C A T I O N
Organometallic titanocene complex as highly efficient
bifunctional catalyst for intramolecular Mannich reaction
Yunyun Wang1,2 | Yajun Jian1 | Ya Wu3 | Huaming Sun1 | Guofang Zhang1 |
Weiqiang Zhang1
| Ziwei Gao1
1 Key Laboratory of Applied Surface and
Colloid Chemistry, MOE, School of
Chemistry and Chemical Engineering,
Shaanxi Normal University, Xi'an 710062,
China
2 College of Chemistry and Chemical
Engineering, Ningxia Normal University,
Guyuan 756000, People's Republic of China
Bifunctional catalysts bearing two catalytic sites, Lewis acidic organometallic
titanocene and Brønsted acidic COOH, have been assembled in situ from
Cp2TiCl2 with carboxylic acid ligands, showing high catalytic activity over
an intramolecular Mannich reaction towards synthesis of 2‐aryl‐2,3‐
dihydroquinolin‐4(1H)‐ones. The determination of the bifunctional catalyst
Cp2Ti(C8H4NO6)2 was elucidated by single X‐ray HR‐MS and investigation of
catalytic behavior. In particular, masking the Brønsted acidic COOH catalytic
site with dormant COOMe lowered the reaction yield greatly, indicating that
two catalytic sites work together to maintain high catalytic efficiency.
3 College of Chemistry and Chemical
Engineering, Xi'an Shiyou University,
Xi'an 710065, People's Republic of China
Correspondence
Yajun Jian, Key Laboratory of Applied
Surface and Colloid Chemistry, MOE,
School of Chemistry and Chemical
Engineering, Shaanxi Normal University
Xi'an 710062, People's Republic of China.
Email: yajunjian@snnu.edu.cn
KEYWORDS
bifunctional catalysts, intramolecular Mannich reaction, Titanocene complex
Ziwei Gao, Key Laboratory of Applied
Surface and Colloid Chemistry, MOE,
School of Chemistry and Chemical
Engineering, Shaanxi Normal University
Xi'an 710062 People's Republic of China.
Email: zwgao@snnu.edu.cn
Funding information
National Natural Science Foundation of
China, Grant/Award Number: 21571121,
21771122, 21602129; Natural Science Basic
Research Plan in Shaanxi Province of
China, Grant/Award Number:
2017JM2020; the 111 Project, Grant/
Award Number: B14041; the Program for
Chang jiang Scholars and Innovative
Research Team in University, Grant/
Award Number: IRT_14R33
1 | INTRODUCTION
and powerful strategy for many difficult or unattainable
chemical transformations, owing to its multiple activa-
tion effect on substrates, compared with mono‐catalysis.
In particular, bifunctional catalysis, where two discrete
catalytic site resides in one molecule, can not only
Multi‐catalysis, including bifunctional catalysis,[1] cas-
cade catalysis,[2] double activation catalysis[3] and syner-
gistic (cooperative) catalysis,[4] has become a promising
Appl Organometal Chem. 2019;e4925.
wileyonlinelibrary.com/journal/aoc
© 2019 John Wiley & Sons, Ltd.
1 of 7
https://doi.org/10.1002/aoc.4925

2 of 7
WANG ET AL.
activate both the nucleophile and electrophile simulta-
neously, but also bring two reactants close to each other,
facilitating the creation of new bonds.[5] From the perspec-
tive of acid–base theory, reports of successful chemical
transformation via the bifunctional catalysis mechanism
can be classified into the following modes: Lewis acid–
Lewis base (LA–LB),[6] Lewis acid–Lewis acid (LA–
LA),[7] Lewis acid–Brønsted base (LA–BB),[8] Brønsted
acid–Brønsted base (BA–BB)[9] and Brønsted acid–Lewis
base (BA–LB) (Scheme 1).[10] However, there are rare
homogeneous bifunctional catalysis reports that fall into
the Lewis acid–Brønsted acid (LA–BA) combination.[11]
Therefore, it is of great importance to design a novel
bifunctional catalytic system via the LA–BA combination,
thus consummating bifunctional catalytic combinational
modes, as well as providing practicable strategies to realize
challenging chemical transformations.
However, there are still great challenges in terms of
their efficiency and the adaptability of their application
as catalysts in industry, owing to inherent characteristic
air and water sensitivity. One workable solution is to
employ a readily activated chemically inert Ti (IV) pre-
cursor, and Cp2TiCl2 has proven be an excellent model,
which can be easily activated in situ by ligand substitu-
tion. By using various O‐, N‐donor ligands, Cp2TiCl2 has
been successfully activated and applied in many impor-
tant C–C and C–N bond formations.[13] Nevertheless,
the ordinary ligand activation strategy can only boost
the Lewis acidity of the Ti (IV) centre, limiting catalytic
capability. We hypothesize that the application of func-
tional ligands may solve this dilemma, as the added
functional group on the ligands may serve as a new cat-
alytic site in the in situ assembled titanocene complex.
Herein, we present our further study on activation of
Cp2TiCl2 with a double carboxylic acid‐containing
ligand to generate a bifunctional catalyst in LA–BA
Ti (IV) salts have long been employed as efficient
Lewis acidic catalysts in synthetic organic chemistry.[12]
SCHEME 1 Various combinational
modes of bifunctional catalyst

WANG ET AL.
3 of 7
mode in situ, where the Ti (IV) centre functions as the
Lewis acidic catalytic site and the COOH group on the
ligand functions as the Brønsted acidic catalytic site
(Scheme 1).
was elevated to 45–50% (entries 5–7). Thus, further inves-
tigation was based on L2 owing to its higher catalytic effi-
ciency. Introduction of an electron‐withdrawing group,
including NO2 (L5) and COOH (L6), led to a higher reac-
tion yield, especially for L5, for which the reaction yield
could reach 90%, the optimal ligand we have screened
(entry 8). For the complementary comparison of mono‐
catalysis with multi‐catalysis, L5 alone was used to catal-
yse the reaction and only 40% yield of product was
obtained, showing the necessity of cooperation between
Lewis acid Cp2TiCl2 and Brønsted acid L5.
To demonstrate the utility of this method, substrate
scope was explored extensively via the variation of the
reactant aldehydes. As shown in Scheme 2, aromatic
aldehydes bearing electron‐donating groups (3a1–3a4)
gave 2‐aryl‐2, 3‐dihydroquinolin‐4(1H)‐ones in good to
high yields. The exception was when two ‐OMes were
carried, were the yield decreased to only 69%. This might
be because of the strong electron‐donating ability of
atomic oxygen's lone electron pair to the –CHO group
through p–π conjugation, decreasing the reaction activity
of aldehyde with the amino group to form imine. The
entries of those aldehydes with mild electron‐
withdrawing substituent groups, including Cl (3a9–3a10),
Br (3a11–3a13) and F (3a14), were proved to be well toler-
ated under the optimized conditions, as the reaction yield
was at least 84%. The exception is when Br‐ was at the o‐
position to the ‐CHO group (3a11, 79% yield). We tenta-
tively attributed this poorer performance to steric
2 | RESULTS AND DISCUSSION
To test our model of the establishment of an LA–BA
bifunctional catalyst, intramolecular Mannich reactions
between o‐amino acetophenone (1a) and p‐methoxy‐
benzaldehyde (2a) were chosen as the model reaction,
one of the most efficient approaches to obtain an impor-
tant class of bioactive dihydro‐4‐quinolones.[14] As shown
in Table 1, the catalyst‐free condition did not lead to any
product being produced (entry 1). When Cp2TiCl2 was
employed as the catalyst, 16% yield of product was
detected, indicating that unmodified titanocene also is
not effective (entry 2). As mentioned in the Introduction,
the designed bifunctional catalyst must have two func-
tional groups, one for coordinating with the Ti centre,
the other functioning as a Brønsted acid.
Based on our previous work,13d,15 COOH was chosen
as the ligand to coordinate with Cp2TiCl2 to form a more
active titanocene organometallic complex. Benzoic acid
(L1) was used as the starting point to locate the optimal
bifunctional ligand, with only a slight yield improvement
to 20% yield (entry 4). However, after introducing the
ligand bearing an extra COOH group (L2–4), the yield
TABLE 1 Optimization of the reaction conditions.
Entry
Deviation from standard conditions
Yield (%)
Ligands screened
1
2
3
4
5
6
7
8
9
None
Cp2TiCl2
L5
Cp2TiCl2 + L1
Cp2TiCl2 + L2
Cp2TiCl2 + L3
Cp2TiCl2 + L4
Cp2TiCl2 + L5
Cp2TiCl2 + L6
—
16
40
20
50
48
45
90
60
(a) All reactions were conducted on a 1 mmol scale. (b) Isolated yields.

4 of 7
WANG ET AL.
SCHEME 2 Substrates scope for the synthesis of 2‐aryl‐2,3‐dihydro‐4‐quinolones. [a] Reaction conditions: a mixture of 1a (1 mmol), 2a
(1 mmol), and catalyst (15 mol% Cp2TiCl2, 30 mol% L5) in MeOH (2 mL) were stirred at r.t. for a period time needed; [b] Isolated yields
hindrance caused by bulky Br rather than Cl for imine
formation. However, when strong electron‐withdrawing
groups were attached on the aromatic aldehydes, a rela-
tively low yield of 46% was obtained (3a15). Optimized
conditions were also viable for regular heteroaromatic
aldehydes, where 2‐aryl‐2,3‐dihydro‐4‐quinlones could
be generated in moderate yields (3a17–3a19). Substituted
o‐amino acetophenones in terms of the diversity of the
electronic nature of the substituent were also examined.
Our method worked well for the substrates bearing weak
electrophilic‐withdrawing groups, like Cl‐ (3a21) and Br‐
(3a22), as well as electrophilic‐donating groups like Me‐
(3a23) and MeO‐ (3a24), where the product could be gen-
erated in 52–76% yield.
catalysing the intramolecular Mannich reaction effi-
ciently, how they cooperate during the reaction was
unknown. To gain more information on the true catalytic
species, single crystal analysis is most straightforward and
convincing. Fortunately, the reaction of Cp2TiCl2 with
ligand L5 gave a crystalline complex, Cp2Ti(C8H4NO6)2,
Ι (CCDC1882248; Scheme 3a), which was also corrobo-
rated by ESI‐MS analysis, as a signal with an m/z
value at 599.0410 was detected. Analyzing the
crystallographic data, it can be seen that the bond lengths
of Ti–O (Ti–O002, 2.039 Å and Ti–O00a, 1.969 Å) were
longer than those of any of our previous synthesized
titanium complexes (specific information is available in
the Supplementary Material), which is most likely linked
to the enhanced Lewis acidity of the Ti (IV) centre.
Exploration of the coordination between Cp2TiCl2 and
Although during the screening experiment, it has been
verified that Cp2TiCl2 and L5 worked in concert,

WANG ET AL.
5 of 7
SCHEME 3 [a] Preparation and X‐ray
crystal structure of organometallic
structure I; [b] a mixture of 1a (1 mmol),
2a (1 mmol), single crystal Ι in MeOH
(2 mL) were stirred at r.t. for 32 h
L5 indicated that complex Cp2Ti(C8H4NO6)2 was the
activated version of titanocene complex, although
whether it was the true catalyst for the intramolecular
Mannich reaction was still ambiguous. Therefore, the
model reaction was carried out in the presence of crystal
Cp2Ti(C8H4NO6)2, as shown in Scheme 3b. Gratifyingly,
the title product 3a4 could be produced in 93% yield.
Taking the structural information and catalytic perfor-
mance of single crystal Ι, it is reasonable to draw a con-
clusion that the true catalytic species is complex Ι.
The next task is to confirm whether the COOH on
the spectator ligand functions as the Brønsted acid cata-
lytic site. It is widely acknowledged that the confirma-
tion of a reaction through the bifunctional mechanism
is challenging, mainly owing to the difficulty of the
acquisition of the co‐crystal structure of the substrate
catalyst. However, masking one catalytic site with an
inert group is well accepted and adopted as an alterna-
tive strategy.[16] The methyl group was chosen to mask
the COOH group as it not only can be restrained from
showing its Brønsted acidic catalytic ability, but also
does not change the electronic nature of the spectator
ligand. As shown in Scheme 4, when ligand L7 was
used replacing L5, the reaction yield dropped to 60%,
suggesting that COOH on the spectator ligand L7
worked as the second catalytic site.
Given that the bifunctional catalysis species had been
confirmed, we next focused on mapping the reaction
route for the reaction. There are two possible reaction
routes from starting materials to final product 2‐aryl‐2,3‐
dihydroquinolin‐4(1H)‐ones: (1) initiated by the aldol
condensation between o‐amino acetophenone 1a with
benzaldehyde to form 2′‐aminochalcones 4, followed by
intramolecular aza‐Michael addition; and (2) imine
formed first, followed by intramolecular Michael addi-
tion. Although the imine route was proposed to be dom-
inant during the reaction, the evidence is not solid,
confined to the instability of the in situ formed imine
intermediate.[17] As shown in Scheme 5, there were no
aldol adduct 2′‐aminochalcones, 4, whose synthesis and
spectra can be found in the Supporting Information,
detected during the full reaction process, monitored by
1
comparing the H NMR spectra of the reaction mixture
with synthesized samples. As the imine adduct, 5, is too
unstable to survive flash column chromatography, its
identity and determination of content were conducted
by reduction of the reaction mixture to form more stable
amine, 6. When NaBH4 was adopted as the reductant,
SCHEME 4 The comparison of
synthesis of product 3a4 with the
involvement of ligand L5 and L7
respectively, where in the bottom trial the
3‐COOH on L5 was masked with Me
group to be L7

6 of 7
WANG ET AL.
SCHEME 5 Evaluation on two possible
reaction routes for the intramolecular
Mannich reaction
amine 6 was produced in 33% yield, indicating that the
reaction was carried out through the imine route.
fully lengthened, suggesting that the Lewis acidity of Ti
(IV) centre is greatly enhanced, thus facilitating the for-
mation of Ti (IV) species (II) via breakage of the bonds
of O–Ti for further nucleophilic attack. For the substrate
side, intermediate imine 5 was produced under the catal-
ysis of the acid catalyst. Once the imine 5 was formed, it
functioned as a ligand to coordinate cation species II to
form Ti complex III, the key transition state for the
On the basis of these results, a plausible reaction
mechanism for the intramolecular Mannich reaction is
proposed in Scheme 6. Initially, inert pre‐catalyst
Cp2TiCl2 was activated with 3‐nitrophthalic acid to gen-
erate active intermediate I, verified by single X‐ray spec-
troscopy and ESI–MS, in which the O–Ti bonding is
SCHEME 6 Proposed mechanism for
Intramolecular Mannich reaction
catalysed by Titanocene‐based
bifunctional catalyst

WANG ET AL.
7 of 7
[11] P. Gao, Q. Wang, J. Xu, G. Qi, C. Wang, X. Zhou, X. Zhao, N.
reaction, in which the Ti (IV) centre as the Lewis acidic
catalytic site activated the acetyl group, and the COOH
on the spectator ligand played the Brønsted acidic cata-
lytic role, activating imine 5 via the H‐bonding between
H atom on COOH and the N atom on imine. This was
followed by the nucleophilic attack of enol to activated
imine, and the product 3a5 was formed as a transient
ligand. Finally, 3a5 dissociated from complex III, leading
to free product and regeneration of the catalyst II.
Feng, X. Liu, F. Den, ACS Catal. 2018, 8(1), 69.
[12] a) S. A. Ryken, L. L. Schafer, Acc. Chem. Res. 2015, 48, 2576. b)
Y. Qian, J. Huang, M. D. Bala, B. Lian, H. Zhang, H. Zhang,
Chem. Rev. 2003, 103, 2633.
[13] a) T. Takeda, M. Ando, T. Sugita, A. Tsubouchi, Org. Lett. 2007,
9, 2875. b) T. K. Hollis, N. P. Robinson, B. Bosnich, Organome-
tallics 1992, 11, 2745. c) R. Qiu, X. Xu, L. Peng, Y. Zhao, N. Li,
S. Yin, Chem. Eur. J. 2012, 18, 6172. d) Y. Wu, C. Chen, G. Jia,
X. Zhu, H. Sun, G. Zhang, W. Zhang, Z. Gao, Chem. – Eur. J.
Chem. Eur. J. 2014, 20, 8530. e) X. Zhu, C. Chen, B. Yu, G.
Zhang, W. Zhang, Z. Gao, Chem. Lett. 2014, 43, 1832. f) Y.
Wu, X. Wang, Y. Luo, J. Wang, Y. Jian, H. Sun, G. Zhang, W.
Zhang, Z. Gao, RSC Adv. 2016, 6, 15298. g) X. Wang, Z. Wang,
G. Zhang, W. Zhang, Y. Wu, Z. Gao, Eur. J. Org. Chem. 2016, 3,
502. h) X. Zhang, X. Xu, N. Li, Z. Liang, Z. Tang, Tetrahedron
2018, 74, 1926. i) X. Zhang, X. Zhang, N. Li, X. Xu, R. Qiu, S.
Yin, Tetrahedron Lett. 2014, 55, 120.
3 | CONCLUSION
In conclusion, we have established a novel LA–BA
bifunctional catalytic system, which could catalyse an
intramolecular Mannich reaction with high efficiency,
generating a series of 2‐aryl‐2,3‐dihydroquinolin‐4(1H)‐
ones. In addition, mechanistic study via in situ NMR
and trapping the transient intermediate indicates that
the Mannich reaction undergoes the imine route. Our
work not only provides a new approach to achieve highly
catalytic organometallic titanocene species, but also pre-
sents a paradigm for LA–BA bifunctional catalysis.
[14] a) S. Chandrasekhar, K. Vijeender, C. Sridhar, Tetrahedron
Lett. 2007, 48(28), 4935. b) H. Zheng, Q. Liu, S. Wen, H. Yang,
Y. Luo, Tetrahedron: Asymmetry. 2013, 24(15–16), 875. c) B.
Mondal, S. Pan, Org. Biomol. Chem. 2014, 12, 9789. d) P. Knipe,
M. D. Smith, Org. Biomol. Chem. 2014, 12, 5094. e) R. P. Pandit,
K. Sharma, Y. R. Lee, Synthesis. 2015, 47(24), 3881.
[15] J. Wang, W. Zhang, H. Sun, G. Zhang, Y. Wu, Z. Gao, Transi-
tion Met. Chem. 2016, 41, 731.
[16] a) H. W. Zanthoff, M. Lahmer, M. Baerns, E. Klemm, M. Seitz,
G. Emig, J. Catal. 1997, 172(1), 203. b) H. Shen, M. Schmuck, I.
Pilz, N. R. Gilkes, D. G. Kilburn, R. C. Miller Jr., R. A. Warren,
J. Biol. Chem. 1991, 266(17), 11335. c) K. J. Wieder, F. F. Davis,
J Appl. Biochemistry 1983, 5(4–5), 337.
ORCID
Weiqiang Zhang https://orcid.org/0000-0001-8810-9606
Ziwei Gao https://orcid.org/0000-0002-2547-9375
[17] G. Pan, L. Su, Y. Zhang, S. Guo, Y. Wang, RSC. Adv. 2016, 6,
REFERENCES
25375.
[1] M. Shibasaki, M. Kanai, S. Matsunaga, N. Kumagai, Acc. Chem.
Res. 2009, 42, 1117.
SUPPORTING INFORMATION
[2] Q. Cai, Z. Zhao, S. You, Angew. Chem. Int. Ed. 2009, 48, 7428.
[3] Y. J. Park, J. W. Park, C.‐H. Jun, Acc. Chem.Res. 2008, 41, 222.
[4] A. E. Allen, D. W. C. MacMillan, Chem. Sci. 2012, 3, 633.
[5] S. Han, S. J. Miller, J. Am. Chem. Soc. 2013, 135, 12414.
Additional supporting information may be found online
in the Supporting Information section at the end of the
article.
[6] M. Shibasaki, M. Kanai, K. Funabashi, Chem.Commun. 2002,
18, 1989.
[7] S. F. Liu, A. Motta, M. Delferro, T. J. Marks, J. Am. Chem. Soc.
2013, 135, 8830.
How to cite this article: Wang Y, Jian Y, Wu Y,
et al. Organometallic titanocene complex as highly
efficient bifunctional catalyst for intramolecular
Mannich reaction. Appl Organometal Chem. 2019;
e4925. https://doi.org/10.1002/aoc.4925
[8] Y. Ito, M. Savamra, E. Shirakawa, K. Hayashizaki, T. Haysshi,
Tetrahedron 1988, 44(17), 5253.
[9] M. G. Núñez, A. J. M. Farley, D. J. Dixon, J. Am. Chem. Soc.
2013, 135(44), 16348.
[10] K. Ishihara, Y. Ogura, Org.Lett. 2015, 17(24), 6070.