T.R. Tumeli and J.L. van Wyk
Inorganica Chimica Acta 527 (2021) 120563
Table 1
NMR spectroscopy data for salicylaldimine ligands HL1 – HL3.
Ligand structure
NMR spectra recorded in CDCl
3
1
H NMR: δ (ppm) 5: 6.87 (td, 3H, JHH = 7.4); 4: 6.95 (dd, 1H, JHH = 8.2); 3&6:7.25 – 7.31 (m, 2H,);
7
:8.30 (s, 1H); 9:13.13 (br s, 1H) 8: 3.53 (t, 2H, JHH = 6.4); 10: 2.15 (q, 2H, JHH = 6.8); 11: 4.05 (t, 2H,
HH = 6.9); 13: 6.91 (br s, 1H); 14: 7.06 (br s, 1H);12: 7.45 (s, 1H)
3
J
1
C NMR: δ (ppm) 1: 118.7; 2: 160.9; 3: 131.4; 4: 117.0; 5: 118.6; 6: 132.6; 7:166.0; 8: 44.3; 10: 31.8;
1
1: 55.9; 12: 137.1; 13: 118.8; 14: 129.8,
1
H NMR: δ (ppm) 4 & 5: 6.84 (m, 2H); 3 & 6: 7.23–7.33 (m, 2H)9:12.07 (br, s, 1H) 7:8.54 (s, 1H) 8: 4.56
t, 2H, JHH = 6.4); 10: 2.39 (q, 2H, JHH = 6.8); 11: 3.81 (t, 2H, JHH = 6.9); 12: 10.04 (s, 1H), 13: 7.29 (s,
(
1
H); 14: 7.53 (s, 1H); 15: 4.11 (t, 2H, JHH = 7.02); 16: 1.78 (q, 2H, JHH = 7.7); 17: 1.32 (septet, 2H, JHH
=
8.2); 18: 0.90 (t, 2H, JHH = 7.4)
1
3
C NMR: δ (ppm) 1:118.2; 2:160.5; 3:131.6; 4:116.4; 5:118.6; 6:132.3; 7:166.4 8: 48.2; 10:30.8;
1
1:55.2; 12:135.7; 13:122.1; 14:122.4; 15: 49.6; 16:131.5; 17:19.2; 18:13.2
1
2
1
8
H NMR: δ (ppm) 4 & 5: 7.45 (m, 2H); 3 & 6: 7.35 (m, 2H)9:12.91(br, s, 1H) 7:8.55 (s, 1H)8: 4.55 (t,
H, JHH = 6.4); 10: 2.40 (q, 2H, JHH = 6.8); 11: 3.82 (t, 2H, JHH = 6.9); 12: 9.95 (s, 1H), 13: 7.24 (s, 1H);
4: 7.46 (s, 1H); 15: 4.10 (t, 2H, JHH = 7.02); 16: 1.78 (q, 2H, JHH = 7.7); 17: 1.33 (septet, 2H, JHH
.2) 18: 0.91 (t, 2H, JHH = 7.02)
=
1
3
C NMR: δ (ppm) 1:118.0; 2:160.3; 3: 131.4; 4: 116.1; 5:118.3; 6:132.1; 7:166.1; 8:47.9; 10: 30.5; 11:
5
5.6; 12:135.0; 13:123.0; 14:122.2; 15:49.3; 16:31.2; 17:18.9; 18: 13.0
br: broad, s: singlet, d: doublet, dd: doublet of doublets, t: triplet, td: triplet of doublets, q quintet, JHH: coupling constant in Hz
2
.6. Magnetic susceptibility of complexes 1–6
closes and producing the cyclic carbonate product as shown in Scheme 3
with the catalyst being regenerated. In the case of multifunctional cat-
alysts, the nucleophile is part of the catalyst structure and thus a co-
catalyst that would function a nucleophile is no longer required. The
slight increase in activity when [PPN]Cl was used as co-catalyst in-
dicates that the nucleophile required for ring opening of the epoxide
(Scheme 3) is not as readily available at room temperature when com-
plex 3 is used alone and may be due to the poor solubility of the complex
in PO at room temperature. The slight increase in activity is most likely
The Evans’ method was used to determine the magnetic suscepti-
bility of the metal complexes. The magnetic moment for the Cr(II)
complexes 1 and 2 were found to be 4.81 and 4.94B.M. respectively.
These values are close to the expected spin only values of 4.90 B.M for a
high spin tetrahedral complex [30]. Complexes 3 and 4 have magnetic
moment of 4.45 and 4.59 B.M respectively. These values are comparable
to magnetic moments reported for other Cr(III) complexes with a five
coordinated metal center [31]. The respective magnetic moments of Co
as result of [PPN]Cl which alone can also catalyze the coupling of CO
2
(
II) complexes 5 and 6 are 1.84 and 2.17 B.M. This indicates that in
and propylene oxide. The activity of complex 3 improves significantly
◦
solution these complexes are most likely in a low spin square planar
geometry with one unpaired electron [32].
when the catalytic reactions are performed at 80 C producing a PO
ꢀ
1
conversion of 19.4% resulting in a TOF of 16.2 h which is a significant
ꢀ
1
◦
improvement form the TOF of 0.3 h
observed at 25 C. The
improvement in activity is most likely due to the improved solubility of
2
.7. Evaluation of complexes as catalysts for the coupling CO
2
and PO
◦
the metal complex at 80 C. However when comparing the catalytic
activity of the complex 3 to Cr(III) complexes based on salen ligands the
performance is inferior, however the results presented here is a pre-
liminary screening of the viability of the complexes presented in this
paper for this coupling process [7]. Optimization of the catalytic ex-
The chromium and cobalt complexes were evaluated as catalyst for
the coupling of CO and propylene oxide (PO). These reactions were
performed solventless in a bottom stirred stainless steel reactor using
.0286 mmol complex at 1 MPa CO pressure. All reactions were per-
formed for 12 h using a 1:1000 metal to PO ratio. The crude reaction
2
0
2
1
periments should be done for a true comparison. In the H NMR spectra
of experiments performed with complex 3 only propylene carbonate
1
3
residues were analyzed with H NMR in CDCl in order to determine if
(
PC) was observed as coupling product.
any coupling of the CO with PO was observed. The data obtained for the
2
catalytic experiments performed with the complexes is summarized in
Table 5. Since majority of the metal complexes based on tetracoordinate
2.8. Effect of metal oxidation state on catalytic performance
2
Schiff base ligands typically used for the coupling of CO with epoxides
are Cr(III) or Co(III) complexes [33,34] the performance of complex 3 as
When comparing the catalytic performance of the Cr(II) analogues 1
and 2 to their Cr(III) counter parts it is evident that the Cr(III) complexes
3 and 4 are slightly more active than the Cr(II) complexes. For instance
Cr(III) complex 3 performed better than its corresponding Cr(II) com-
◦
catalyst was evaluated first at 25 C. For this experiment 0.4% PO
conversion was obtained which increased to 2.4% PO conversion in the
presence of [PPN]Cl which was used as co-catalyst in a 1:1 ratio with the
metal complex. Carbon dioxide and epoxide coupling catalyzed by metal
complexes is proposed to occur through a coordination-insertion
mechanism as shown in Scheme 3 [35,36]. The pathway is initiated
by the coordination of the epoxide to the metal center, followed by
ꢀ
1
ꢀ 1
plex 1 resulting in a TOF of 16.2 h compared to the TOF of 11.0 h
observed with complex 1. This difference in activity is most likely due to
the fact that PO will bind more readily to the Cr(III) metal center which
is slightly more electron deficient than the Cr(II) metal center. However,
the difference in activity may also be due to the difference in stability of
the complexes. It is possible that the Cr(III) analogues are slightly more
stable than the Cr(II) complexes under the experimental condition used
attack by a nucleophile which results in ring opening of the epoxide. CO
2
insertion between the metal and the ring opened epoxide results in the
formation of a metal carbonate. The formed metal carbonate can ring
3