Synthesis, characterization of zinc complexes with neutral α-diimine ligands and application in ring-opening polymerization of σ-caprolactone

Article abstract of DOI:10.2533/chimia.2017.773

A series of neutral α-diimine ligands with diacetyl and acenaphthenequinone skeletons were prepared by the reaction between diacetyl and the corresponding aromatic amine. These ligands reacted with ZnCl2 to generate symmetric α-diimine zinc complexes C1-C

Full text of DOI:10.2533/chimia.2017.773

Note
CHIMIA 2017, 71, No. 11 773
doi:10.2533/chimia.2017.773
Chimia 71 (2017) 773–776 © Swiss Chemical Society
Synthesis, Characterization of Zinc
Complexes with Neutral α-Diimine
Ligands and Application in Ring-opening
Polymerization of ε-Caprolactone
Xiaodan Wang*, Xuehong Liu, and Ju Huang
Abstract: A series of neutral α-diimine ligands with diacetyl and acenaphthenequinone skeletons were prepared
by the reaction between diacetyl and the corresponding aromatic amine. These ligands reacted with ZnCl2 to
generate symmetric α-diimine zinc complexes C1–C10. Experimental results indicated that the α-diimine zinc
complexes with a diacetyl skeleton (C1–C4) were active in ring-opening polymerization (ROP) of ε-caprolactone
(ε-CL). The complexes with an acenaphthenequinone skeleton showed a small steric effect (C5, C8 and C9) but
the complex substituted with an electron-withdrawing group (C10) showed high activity in the monomer conver-
sion rate during ROP of ε-CL. The ROP catalysts of ε-CL demonstrated the mechanism of monomer activation
in the presence of benzyl alcohol.
Keywords: ε-Caprolactone · Catalyst · Neutral ligands · Ring-opening polymerization · Zinc complex
In recent years, only a few neutral li- low solid complex (C1) with a yield of
Introduction
gands[14–17] have been developed to stabi- 72.5%.[21] By using the same method, we
lize divalent zinc. The research group of
Börner[18,19] has designed and synthesized yield of 66.9% and 80.2%, respectively.
a series of guanidine imine zinc complexes
synthesized complexes C3 and C4 with a
Due to specific biodegradability and
biocompatibility, aliphatic polyesters, such
as polycaprolactone (PCL),[1,2] polylac-
tide[3,4] and their copolymers, are frequent-
ly used as surgical sutures, drug delivery
devices, and gene carriers in medical set-
tings.[5–7] Given these applications, reduc-
ing the toxicity and adverse events induced
by the catalysts used in polymerization is
Theα-diiminezinccomplexeswithace-
with neutral ligands. These guanidine im-
naphthenequinone skeletons were obtained
ine zinc catalysts possess satisfactory sta- directly via one-pot synthesis (Scheme 1).
bility toward air and moisture at room tem- With formic acid as the catalyst, acenaph-
perature and show moderate activity in the thenequinone and aniline were refluxed for
polymerization of lactide. To synthesize a
simple catalyst with low toxicity and high ture refluxed for another 2 h to obtain red-
stability to be applied in the medicine field, brown crystals (C5) with 63.1% yield.[22]
2 h. Zinc chloride was added and the mix-
of great importance. This objective has ac-
we designed and synthesized a novel series Complexes C7 and C8 were produced with
of neutral α-diimine ligands with diacetyl a yield of 67.5% and 56.7% following the
and acenaphthenequinone skeletons along reaction of acenaphthenequinone with
with the corresponding zinc complexes, 2,6-diisopropylaniline or 3,5-dimethylani-
which were commercially available, easy line. Complexes C2, C6, C9 and C10 were
to handle and could tolerate air and mois- synthesized according to the approaches
ture. The catalytic activity of these Zn-
catalysts was analyzed with respect to the All the obtained complexes were
electronic and steric influence of the imino separated, purified, and characterized by
celerated the development, separation, and
purification of efficient catalytic systems
with low or zero toxicity (e.g. zinc, magne-
sium, calcium, and iron complexes)[2,8–10]
to catalyze the ring-opening polymeriza-
tion (ROP) of ε-CL or the polymerization
of lactide. Zinc catalysts with a phenylic
amino,[11] or a β-imino group,[12] or with a
Schiff base moiety[13] were proven to have
high activity and excellent capability of
dimensional control. However, these cata-
lysts are unstable, sensitive to the impact
of air and moisture, and difficult to store.
previously described.[23–25]
ligands on the catalyzed reaction.
1H-NMR. The molecular structures of
complexes C1, C2, C5, C7, and C8 were
confirmed by X-ray structural analysis.
Single crystals of the complexes C1
and C8 suitable for X-ray structural analy-
Methods and Results
The α-diimine zinc complexes with
diacetyl skeletons were obtained via a ylene chloride and n-hexane solutions. The
two-step reaction. The ligands were first Oak Ridge Thermal Ellipsoid Plot draw-
sis were obtained from concentrated meth-
synthesized and subsequently purified to ings that depicted the selected bond dis-
produce zinc complexes with zinc chlo- tance and angle of the molecular structures
ride (Scheme 1). Under acid catalysis, di- of C1, C2, and C8 are illustrated in Figs 1,
acetyl reacted with 2,6-dimethylaniline,[20] 2, and 3, respectively.
3,5-dimethylaniline, or 4-methoxy aniline
The molecular structures show that the
*Correspondence: MS. X. Wang
School of Pharmaceutical Chemistry and Chemical
Engineering
Guangdong Pharmaceutical University
13 Changmingshui Road, Wuguishan Town
Zhongshan 528458, China
to produce ligands L1, L3 and L4, respec- central metal zinc is tetrahedrally coordi-
tively. L1 was dissolved in toluene. Zinc nated and the ligand angle is between 77°
chloride and a catalytic amount of glacial and 80°, close to that reported in ref. [26].
acetic acid were added to the solution at
The size and structure of the aromatic
E-mail: junhaotian62@sina.com
room temperature to produce a pale yel- amine ligands influence zinc coordination.

774 CHIMIA 2017, 71, No. 11
Note
Fig. 1. Molecular structure of C1·3CH2Cl2 de-
picted with 30% thermal ellipsoids. Hydrogen
atoms and three uncoordinated dichloro-
methane molecules were omitted for clar-
ity. Selected bond distances(Å) and angles
(°): Zn(1)–N(1) 2.078(3), Zn(1)–N(2) 2.073(3),
Zn(1)–Cl(1) 2.204(1), Zn(1)–Cl(2) 2.208(1), N(1)-
Zn(1)-N(2) 78.83(11), N(2)-Zn(1)-Cl(1) 116.68(8),
N(1)-Zn(1)-Cl(1) 113.67(8), N(2)-Zn(1)-Cl(2)
114.91(8), N(1)-Zn(1)-Cl(2) 115.96(8), Cl(1)-
Zn(1)-Cl(2) 115.96(8).
Scheme 1. The synthetic route and chemical structures of the zinc complexes of C1–C10.
Substituents on the aromatic amines were hedral angle between the ligands and the
located above the plane of the metal coor- metal center when acenaphthenequinone
dination center. Steric hindrance of these served as the skeleton structure. In this
substituents has been proven to affect the study, the dihedral angles for C7, C5, and
Fig. 2. Molecular structure of C2 depicted with
30% thermal ellipsoids. Hydrogen atoms were
omitted for clarity. Selected bond distances (Å)
and angles (°): Zn(1)–N(1) 2.090(4), Zn(1)–N(1A)
2.090(4), Zn(1)–Cl(1) 2.189(2), Zn(1)–Cl(2)
2.201(2), N(1)-Zn(1)-N(1A) 77.7(2), N(1A)-Zn(1)-
Cl(1) 117.18(11), N(1)-Zn(1)-Cl(1) 117.18(11),
catalytic properties of complexes. The ste-
C8 were 88.41°, 60.65°, and 58.78°, re-
ric hindrance of aromatic amines in C1 spectively. The reduction in the dihedral
was more significant than in C2 when bu- angle decreased the shielding effect of
tanedione served as the skeleton structure. the aromatic ring on the metal center. The
N(2)-Zn(1)-Cl(2) 111.68(11), N(1)-Zn(1)-Cl(2)
The bond length of Zn–N(1) and Zn–N(2)
X-ray crystal structural analysis demon-
was 2.078(3) and 2.073(3) Å in C1 and strated that the aromatic ring on the ligand
2.090(4) Å (both Zn–N bonds) in C2. The plane approached the axial position with
111.68(11), Cl(1)-Zn(1)-Cl(2) 115.79(9).
results indicated that the metal complex significant hindrance of aromatic amines.
capacity of ligands was reduced when the Consequently, the open nature of the metal
quantity of hindered aromatic amines was center and the ability to coordinate mono-
increased, whichisconsistentwiththefind- mers were thereby reduced. The catalytic
ings in this study. In addition, substituents activity of the catalyst also declined.
on the aryl amine were observed above the
In this investigation, the hindrance and
complex plane of the metal center. Steric electronic effects on catalytic activity were
hindrances have been known to affect the confirmed when the α-diimine zinc com-
catalytic properties of complexes. The di-
hedral angles between the ligand plane of As reported in Table 1, the catalytic activ-
the metal center and the aromatic amine ity of diacetyl skeletons was significantly
plexes catalyzed ROP of ε-CL[27] (Table 1).
ring of C1 attached to the amido nitro- higher than that of acenaphthenequinone
gen atoms N(1) and N(2) were 87.05° and skeletons with the same substituent on the
Fig. 3. Molecular structure of C8·CH Cl de-
aromatic ring. Therefore, strengthening
85.10°, and 86.62° and 86.65° for C2. The
2
2
picted with 30% thermal ellipsoids. Hydrogen
atoms and one uncoordinated dichlorometh-
ane molecule were omitted for clarity. Selected
bond distances(Å) and angles (°): Zn(1)–N(1)
2.104(2), Zn(1)–N(2) 2.095(3), Zn(1)–Cl(1)
2.211(1), Zn(1)–Cl(2) 2.170(1), N(1)-Zn(1)-N(2)
80.60(9), N(2)-Zn(1)-Cl(1) 108.69(8), N(1)-Zn(1)-
Cl(1) 110.69(8), N(2)-Zn(1)-Cl(2) 114.42(8),
N(1)-Zn(1)-Cl(2) 114.43(8), Cl(1)-Zn(1)-Cl(2)
introduction of a large, hindered aromatic the skeleton structure decreased the cata-
amine caused the aromatic ring to become lytic activity of the catalysts. Moreover,
approximately perpendicular to the com- increasing the steric hindrance of the aryl
plex plane of the metal center and reduced imines also decreased the catalytic activity
the coordination ability of the monomer of the catalysts. For C1 and C2, the ste-
with the metal center, thereby decreasing ric hindrance of the ortho-substituent on
the catalytic activity of the catalysts.
Similarly, a larger steric hindrance turn reduced the monomer conversion rate
of the aromatic amines decreased the di- from 86.1% to 78.1% (entry 1 and entry 120.94(5).
the aromatic ring was increased, which in

Note
CHIMIA 2017, 71, No. 11 775
2), indicating the reduced catalytic activ-
both cases (entry 10 and entry 12). Such
results may be attributed to the polymer-
ity of the catalysts. The same conclusions
could be drawn with acenaphthenequinone
as the skeleton structure of the catalysts. In
addition, C6 and C7 showed no catalytic
activity (entry 6 and entry 7).
ization mechanism.
Discussion
Scheme 2. The synthesis of PCL.
The highest catalytic activity was ob-
served for C10 when the monomer conver- In this experiment, the zinc complex-
sion rate reached 95.2% (entry 10). This es did not change structurally when they
were subjected to 24-h reflux in benzyl
metal center. These substituents were con-
metal Lewis acidity and the electropositive
catalyst comprised electron-withdrawing
substituents on the aromatic ring. Even
with a large steric acenaphthenequinone
skeletal structure, the catalysts achieved
the maximal catalytic activity. Meanwhile,
the monomer conversion rate for C9 with
an electron-donating substituent was only
alcohol solution, suggesting that the tradi-
tional form of zinc metal alkoxide cannot
ducive to the formation of bonds between
the monomers and the zinc.
ROP of ε-CL with and without the ad-
dition of benzyl alcohol was performed.
With C1 as the catalyst, the catalytic activ-
be obtained (Scheme 2).[2] These findings
probably result from the characteristics
of the Zn-Cl bond, which is significantly
stable and unlikely to be cleaved.
Given that α-diimine zinc complexes
are extremely stable, their role in the po-
lymerization process was found to be simi-
lar to that of Lewis acids in monomer acti-
vation. Benzyl alcohol was added as an ini-
ity in the presence of benzyl alcohol was
48.5% (entry 9). The electron-withdrawing
significantly higher than without benzyl
alcohol (entry 1 and entry 11). With C10
as the catalyst under the same reaction con-
ditions, the reaction rate was the same in
group clearly increased the catalytic activ-
ity of the catalysts. A potential reason was
the introduction of electron-withdrawing
substituents, which tended to increase zinc
tiator. The catalytic polymerization exerted
both of these combined effects. Therefore,
Table 1. Polymerization of ε-caprolactone with α-diimine zinc catalysts
unlike the polymerization mechanism pro-
posed by Herri–Pawlis for biguanide zinc
complexes,[28] these catalysts demonstrat-
Zna
Znb
Entry Catalyst [I]:[CL]:BnOH
Conv (%)
PDIc
ed that the monomer mechanisms were
(GC)
9916
8024
13622
9074
13479
–
(Theory)
activated (Scheme 3).
1
C1
C2
C3
C4
C5
C6
C7
C8
C9
C10
C1
C10
1:200:1
1:200:1
1:200:1
1:200:1
1:200:1
1:200:1
1:200:1
1:200:1
1:200:1
1:200:1
1:200:0
1:200:0
86.1
78.1
71.4
83.4
81.2
0
19739
17915
16387
19123
18622
–
1.99
1.95
1.75
1.75
2.01
–
The ratio of ε-CL, the catalyst, and the
initiator (benzyl alcohol) was 50:1:1. Thus,
ε-CL could be completely converted to a
polymer after a 24-h reaction at 100 °C.
The resulting PCL was characterized by
1H-NMR. The results are illustrated in
Fig. 4. The chain ends of the PCL were
benzyl alcohol, which validated the acti-
vated monomer mechanism. When benzyl
alcohol was added to the catalytic system,
it served as the initiator and catalyzed the
polymerization with zinc complexes.
In the activated monomer mechanism,
the ability of ligands to bind to the met-
al center was affected by the large steric
hindrance of the complex. The catalytic
activity of the catalyst was consequently
reduced. If the ligand contained electron
withdrawing substituents, the electroposi-
tivity of the metal center increased, the
ability of the ligand to enhance the activa-
2
3
4
5
6
7
0
–
–
–
8
40.3
48.5
95.2
55.9
83.9
5313
3621
14209
8477
19640
9296
11166
21814
12853
19238
1.31
1.42
1.73
1.52
2.20
9
10
11
12
tion of the monomers increased, and the
polymerization reaction proceeded favor-
aObtained from GC analysis and calibrated by a polystyrene standard; bCalculated from the
molecular weight of ε-caprolactone×the conversion yield; cPolymer dispersity index obtained from
GC analysis.
ably. These outcomes were shown in the
crystal structure analysis and the catalytic
ROP.
Conclusion
In conclusion, a novel series of neutral
α-diimine zinc complexes with diacetyl
and acenaphthenequinone skeletons were
synthesized and fully characterized. All
these complexes were found to be very
stable in air and moisture. The complexes
were also determined to be capable of cata-
lyzing the ROP of ε-CL. Compared with
that of guanidine zinc complexes,[18,19] the
Scheme 3. Monomer activation mechanism.

776 CHIMIA 2017, 71, No. 11
Note
Fig. 4. 1H-NMR spectrum of polycaprolactone.
[24] T. Krachko, M. Bispinghoff, A. M. Tondreau, D.
[12] C. C. Roberts, B. R. Barnett, D. B. Green, J. M.
Fritsch, Organometallics 2012, 31, 4133.
[13] Q. Wang, H. Mao, W. Wang, H. Zhu, L. Dai, Y.
catalytic activity of the neutral α-diimine
zinccomplexesincreasedandthesamemo-
lecular weight distribution of the polymer
Stein, M. Baker, A. W. Ehlers, J. C. Slootweg,
H. Grützmacher, Angew. Chem. Int. Ed. 2017,
56, 7948.
Chen, X. Tang, Biometals 2017, 30, 575.
˙
was obtained. These catalysts demonstrat-
[25] E. Kober, Z. Janas, T.
Nerkowski, L. B.
[14] K. Zółtowska, U. Piotrowska, E. Oledzka, M.
Sobczak, Molecules 2015, 20, 21909.
Jerzykiewicz, Dalton Trans. 2013, 42, 10847.
[26] A. L. Sarria, A. F. Vilela, B. M. Frugeri, J. B.
Fernandes, R. M. Carlos, M. F. da Silva, Q. B.
ed activated monomer mechanisms in the
presence of benzyl alcohol. Furthermore,
the steric and electronic effects of ligands
exerted an effect upon the catalytic perfor-
[15] J. Börner, U. Flörke, A. Döring, S. Herres-
Pawlis, Sustainability 2009, 1, 1226.
Cass, C. L. Cardoso, J. Inorg. Biochem. 2016,
[16] J. Börner, U. Flörke, T. Glöge, J. Mol. Catal. A:
164, 141.
Chem. 2010, 316, 139.
[27] Ring opening polymerization of ε-CL: catalyst
mance, and catalytic activities were en- [17] S. Nayab, H. Lee, H. Jeong, Polyhedron 2011,
(0.047 mmol) with benzyl alcohol (0.047 mmol)
30, 405.
hanced in the presence of benzyl alcohol.
or without benzyl alcohol added into a 25 ml
[18] R. Petrus, P. Sobot, Dalton Trans. 2013, 42,
flask under N atmosphere. ε-CL (9.4 mmol,1.0
13838.
ml) was injec2ted quickly to the flask. The flask
Acknowledgment
[19] J. Börner, U. Flörke, K. Huber, A. Döring, D.
We thank for the financial support from the
was then immersed in an oil bath at 140 °C.After
14 h, the polymerization was terminated by ad-
Kuckling, S. Herres-Pawlis, Chem. Eur. J. 2009,
15, 2362.
[20] Preparation of ligand L1: 2,6-dimethylaniline
Synthesis and Characterization of DPP Units
Based Small Molecular Materials for Solar
Cells (Grant No. 2012YM-0080).
dition of a minimum amount of alcohol. The ob-
tained polymers were dissolved in 2 ml CH2Cl2
and then poured into alcohol with stirring. The
sticky polymers were filtered and collected,
(84 mmol, 2.1 equiv) was added to a solution
of butanedione (40 mmol) in 50 ml ethanol. A
catalytic amount of formic acid was dropped in.
Conflict of Interest
then redissolved in CH Cl and precipitated by
The reaction mixture was stirred for 8 h at 40
2
2
The authors declare that they have no con-
H2O, and dried in a vacuum oven at 45 °C for
°C. After cooling to room temperature and fil-
tered, the crude product was recrystallized with
ethanol, affording yellow crystals at a yield of
77.9%.
flict of interests.
24 h.
[28] J. Börner, I. S. Vieira, A. Pawlis, Chem. Eur. J.
2011, 17, 4507.
Received: June 7, 2017
[21] Preparation of complex C1: ZnCl2 (2 mmol)
was added to a solution of L1 (2 mmol) in 20 ml
toluene. A catalytic amount of glacial acetic acid
[1] S. Sorenson, M. Popova, A. M. Arif, L. M.
Berreau, Acta Crystallogr. C Struct. Chem.
2017, 73, 703.
was dropped in. The reaction mixture was stirred
[2] X. Zhang, Y. Zhu, H. Gao, C. Zhao, Inorg.
Chem. 2014, 53, 11903.
[3] Z. Setifi, D. Zambon, F. Setifi, M. El-Ghozzi,
R. Mahiou, C. Glidewell, Acta Crystallogr. C
Struct. Chem. 2017, 73, 674.
[4] A. Baishya, L. Kumar, M. K. Barman, H. S.
Biswal, S. Nembenna, Inorg. Chem. 2017, 56,
9535.
for 24 h at room temperature and filtration. The
crude product was recrystallized with CH Cl2
and hexane, affording yellow crystals at a y2ield
of 73.2%. 1H NMR (300 MHz, DMSO-d6): δ
(ppm) 7.02–6.83 (m, 6H, Ar-H), 1.91 (s, 12H,
CH3), 1.88 (s, 6H, N=C-CH3). 13C NMR (75
MHz, DMSO-d6): δ (ppm) 167.27, 147.70,
127.54, 123.79, 122.82, 17.29, 15.56.
[5] X. Q. Zhou, L. B. Meng, Q. Huang, J. Li, K. [22] Preparation of complex C5: Aniline (0.6 ml,
Zheng, F. L. Zhang, J. Y. Liu, J. P. Xue, Chem.
Med. Chem. 2015, 10, 304.
6.3 mmol) was added to a solution of ace-
naphthenequinone (0.5 g, 2.74 mmol) in 50
[6] Y. Liu, O. Samsonova, B. Sproat, O. Merkel, T.
Kissel, J. Control. Rel. 2011, 53, 262.
was dropped in. The reaction mixture was re-
[7] T. K. Endres, B. M. Broichsitter, O. Samsonova,
fluxed for 2 h. ZnCl (4.4 mmol) was added and
T. Renette, T. H. Kissel, Biomater. 2011, 32,
reflux continued fo2r another 2 h. After filtra-
tion the crude product was recrystallized with
CH2Cl2, affording red-brown crystals at a yield
of 63.1%. 1H NMR (300 MHz, DMSO-d6):
δ (ppm) 8.10 (d, J=7.2 Hz, 2H, Ar-H), 7.54 (m,
6H, Ar-H), 7.31 (m, 2H, Ar-H), 7.10 (m, 4H,
ml ethanol. A catalytic amount of formic acid
7721.
[8] E. Pahontu, V. Fala, A. Gulea, D. Poirier, V.
Tapcov, T. Rosu, Molecules 2013, 18, 8812.
[9] M. G. Cushion, P. Mountford, Chem. Commun.
2011, 47, 2276.
[10] M. A. Telpoukhovskaia,
C. Rodríguez-
Rodríguez, L. E. Scott, B. D. Page, B. O.
Patrick, C. Orvig, J. Inorg. Biochem. Chem.
2014, 132, 59.
Ar-H), 6.73 (d, J=7.2 Hz, 2H, Ar-H). 13C NMR
(75 MHz, DMSO-d6): δ (ppm) 163.42, 151.49,
149.87, 129.46, 127.75, 127.13, 123.23, 120.84,
117.41, 116.34.
[11] Y. Huang, W. Wang, C. C. Lin, M. P. Blake, L. [23] C. S. K. Mak, H. L. Wong, Q. Y. Leung, W. Y.
Clark, A. D. Schwarz, P. Mountford, Dalton
Trans. 2013, 42, 9313.
Tam, W. K. Chan, A. B. Djurišic´, J. Organomet.
Chem. 2009, 694, 2770.