2.1. Synthesis routes
Usually, the synthesis of heterometallic transition metal-rare-earth complexes is performed by dissolution of RE(NO)
3 and transition metal reagents (chlorides, nitrates, acetates) in the water-alcohol mixture as solvent, using amines to deprotonate the phenolic ligands. These reactions usually lead to thermodynamically stable crystalline products with molecular structure mostly depending on the reaction conditions, used reagents, or crystallization procedure. The more controlled synthesis method was used to prepare group 4 – rare earth oxo alkoxides by substituting the halogen atoms in RECl
3 by K[Zr
2(O
iPr)
9] in THF solution, leading to the formation of [REZr
2(O
iPr)
9Cl
2]
x (RE
III = Y
III, Nd
III, Ho
III, Ce
III for
x = 2; Er
III, Yb
III for
x = 1) [
27]. Alternative method claims the synthesis of titanium-lanthanide oxo clusters pictured in
Figure 2, i.e., [RETi
4(O)
3(O
iPr)
2(OMc)
11] (RE = La, Ce; OMc = methacrylate), [RE
2Ti
6O
6(OMc)
18(HO
iPr)] (RE = La, Ce, Nd, Sm) and [RE
2Ti
4O
4(OMc)
14(HOMc)
2] (RE = Sm, Eu, Gd, Ho) by the reaction of titanium isopropoxide, lanthanide acetate and methacrylic acid [
28].
The direct route for the rare-earth complexes from the pure metals has not been considered as attractive because it requires activation by Hg, HgCl
2, or I
2 and prolonged heating [
29].
Recently Sobota and coworkers reported the synthesis of heterometallic oxo-alkoxides by the reaction of metallic lanthanides with divalent transition metal chlorides MCl
2 (M
II = Mn
II, Ni
II or Co
II) or group 4 metallocene dichlorides Cp
2MCl
2 (where M
IV = Ti
IV, Zr
IV, Hf
IV), using 2-methoxy ethanol or ethanol as solvents and reagents [
30,
31]. The reaction of RECl
3 with the molar excess of alkaline or alkaline earth ligand precursors is used as the main method for synthesizing heterometallic rare earth – group 1/group 2 complexes [
32,
33].
Employing precursors with well-defined structures in synthesizing heterometallic complexes minimizes the risk of side reactions or the formation of multiple reaction products, ensuring a more controlled and selective outcome that enhances the synthesis efficiency and selectivity. The structurally authenticated precursors act as templates, guiding the coordination of specific ligands to metal sites to yield a predetermined complex. The advantages of this methodology are particularly evident in the enhanced reproducibility of heterometallic complexes and the ability to scale up the synthesis for practical applications. Moreover, the structurally authenticated precursors enable a deeper understanding of the reaction mechanism, facilitating the design of more sophisticated and tailored heterometallic rare-earth complexes with desired properties for diverse catalytic and material applications. For example, trinuclear ionic zinc compound based on [Zn
3(OAc)
2(L)
2(H
2O)
2]
2+ cation (where LH = 2-ethoxy-6-((pyridin-2-ylmethylimino)methyl)phenol)) [
34] could be used as attractive molecular platform for the synthesis of a wide range heterometallic Zn
IIYb
III clusters by the reaction with various ytterbium salts as shown in
Figure 3.
The use of Yb(ClO
4)
3 led to the [Zn
2Yb(L)
2(OAc)
4]ClO
4, and Yb(NO
3)
3 led to [ZnYb(L)
2(NO
3)
3]. The treatment of trinuclear zinc cation with Yb(OAc)
3 allows the formation of [ZnYb(OAc)
4(L)], but the reaction with YbCl
3 led to the [Yb
2(L)
2(OAc)(H
2O)Cl]
2Cl
4. The reversed sequence was applied in the reaction of [RE(L)(THF)
n] (where RE
III = Y
III, Sm
III, Nd
III, La
III;
n == 1, 2) with Co(OAc)
2 that as the results of ligand redistribution lead to the formation of [Co
2RE
2(OAc)
4(L)
2(THF)
n] containing acetato bridged RE
III dimers capped by Co
II trisphenolato units, as seen in
Figure 4 below [
35].
2.2. Magnetism
Coordination compounds that behave as single-molecule magnets (SMMs), which exhibit slow magnetization reversal and distinct hysteresis in the absence of long-range magnetic order, are of particular interest because of their potential applications in high-density information storage [
36]. Lanthanide complexes show great potential for SMMs because of their strong magnetic anisotropy and large-spin ground states, of which the best candidates are Tb
III, Dy
III, and Ho
III. However, their combination with 3d metal ions is essential to suppress quantum tunneling effectively and establish a magnetic exchange between metal ions (3d–4f) [
37,
38,
39]. The combination of different paramagnetic metal ions within the same molecular entity leads to a wide variety of magnetic properties of the heterometallic compounds. 3d–4f complexes have been intensively studied in order to reveal the factors governing the nature and magnitude of the exchange interaction between a 3d metal ion and various 4f metal ions. The comparison of magnetic properties of the series of [M
3Ln
2(opba)
3] (for M
II = Cu
II, Ni
II; opba
4- =
ortho-phenylenebis(oxamato)) based on a ladder-type structure with isostructural zinc compounds allow to determine the nature of interactions in Ln
III−M
II pairs. This led to the conclusion that the interaction is ferromagnetic for Gd
III, Tb
III, and Dy
III and seems to be antiferromagnetic for Ce
III, Pr
III, Nd
III, and Er
III [
40,
41]. For the series of trinuclear complexes of general formula [Cu
2Ln(L)
2(NO
3)
2]
+ for (LH
2 = 2,6-di(acetoacetyl)pyridine), the presence of an antiferromagnetic interaction for Ln
III = Ce
III, Pr
III, Nd
III and Sm
III, and ferromagnetic for Ln
III = Gd
III, Tb
III, Dy
III, Ho
III, and Er
III was observed [
42]. The structures of those complexes can be found in
Figure 5.
The nature of the Ni
II−Ln
III exchange interaction within binuclear [NiLn(valpn)(CH
3CN)
2(NO
3)
3] (for Ln
III = Pr
III, Nd
III, Sm
III, Eu
III, Gd
III, Er
III) found in
Figure 6, [NiLn(valpn)(CH
3CN)(H
2O)
4(NO
3)](NO
3)
2 (for Ln
III = Tb
III, Dy
III) and [NiCe(valpn)(CH
3CN)
2(H
2O)(NO
3)
3]·[NiCe(valpn)(CH
3CN)(H
2O)
2(NO
3)
2](NO
3) (for valpn
2- = 1,3-propanediylbis(2-iminomethylene-6-methoxy-phenolato)) has been emphasized by comparison with the cryomagnetic behavior of the related [Zn
IILn
III] derivatives. This route allowed for establishing that the interaction within these compounds is antiferromagnetic with the 4f ions of the beginning of the Ln series (Ce
III – Eu
III) and turns ferromagnetic from Gd
III to Er
III. The [Ni
IIDy
III] complex shows slow relaxation processes of the magnetization close to 2 K [
43].
For the heterodinuclear [CuLn(L)(NO
3)
3] (
Figure 7) complexes (Ln
III = Ce
III−Yb
III; LH
2 =
N,
N‘-ethylenebis(3-ethoxysalicylaldiimine)), ferromagnetic interactions seem to be exhibited for the Gd
III, Tb
III, Dy
III, Ho
III, Tm
III, and Yb
III, while for Ce
III, Nd
III, and Sm
III the interaction was antiferromagnetic. Pr
III and Eu
III analogs behaved as spin-uncorrelated systems, and no definite conclusions were reached for the Er
III complex [
44]. Within the series of [CuLn(L)(Me
2CO)(NO
3)
3] complexes pictured in
Figure 8, the nature of the coupling between the Cu
II–Ln
III ions was antiferromagnetic for Ce
III, Nd
III, Sm
III, Tm
III, and Yb
III, and ferromagnetic for Gd
III, Tb
III, Dy
III, Ho
III, and Er
III, with Cu
II–Pr/Eu
III pairs devoid of any significant interaction [
45].
The first example of a 3d–4f SMM was [CuTb(hfac)
2(L)]
2 (LH
3 = 1-(2-hydroxybenzamido)-2-(2-hydroxy-3-methoxy-benzylideneamino)-ethane; hfacH = hexafluoroacetylacetone), which has an effective energy barrier of 21 K, a relaxation time of 2.7 ×10
–8 s, and an estimated blocking temperature of 1.2 K [
46]. Since this discovery, combinations of Tb
III , Dy
III, Er
III, Sm
III, Yb
III, Gd
III, or Ho
III with transition metal ions such as Co
II, Mn
II/III, Fe
II/III, Ni
II, and Cu
II have been extensively studied. SMM behavior has also been observed for Tb
III and Dy
III complexes within the series of trimetallic [CuLn(L)(C
3H
6O)(NO
3)
3] (where Ln = Gd
III, Tb
III, Dy
III, Ho
III, Er
III; L
2- = 2,2'-((2,2-dimethylpropane-1,3-diyl)bis((nitrilo)methylylidene))bis(6-methoxyphenolato))) with activation energies of magnetization reversal equal to 42.3(4) and 11.5(10) K, respectively. The magnetic exchange couplings in Cu
II−Ln
III display a monotonic decrease of ferromagnetic J
Ln–Cu in the order of the atomic number, from
64Gd to
68Er (
JLn–Cuk
B−1/K = 6.9 (Gd), ≥ 3.3 (Tb), 1.63(1) (Dy), 1.09(2) (Ho), 0.24(1) (Er)) [
47].
The family of hetero-tri-metallic complexes [{CuTb(L)(H
2O)
4}{M(CN)
6}]
n (for M
III = Co
III, Cr
III, Fe
III; L
2- = 2,2'-(propane-1,3-diylbis(nitrilomethylylidene))bis(6-methoxyphenolato), 2,2'-(ethane-1,2-diylbis(nitrilomethylylidene))bis(6-methoxyphenolato); n = 1, 2, ∞) are interesting examples that illustrate how SMM behavior of the [CuTb(L)]
3+ moiety can be modulated via the control of intermolecular interactions with [M(CN)
6]
3− species [
48].
When [Cr/Fe(CN)
6]
3− are used, weak antiferromagnetic interactions are responsible for the decrease of the SMM efficiency. The combination of the diamagnetic [Co(CN)
6]
3− with a [CuTb(L)]
3+ moiety resulted in an improvement of the SMM properties compared to the reference [CuTb] complex, with a significantly longer relaxation time. The increase of the effective anisotropic barriers (U
eff) from 5–7 cm
−1 present for the [{CuTb}Cr/Fe] compounds to 15–18 cm
−1 for [{CuTb}Co] was also observed [
49].
The heterometallic [MnLn
2(QCl)
8] (Ln = Dy
III, Tb
III; QCl
- = 5-chloro-8-quinolinolate) display clearly resolved out-of-phase susceptibility maxima below 10 K originating from SMM behavior. Both of them are interesting examples demonstrating that the co-complexation of [Bu
4N][Ln(QCl)
4] with Mn(NO
3)
2 led to a structural transformation from mononuclear to trinuclear compounds enhancing single molecule magnetic (SMM) behavior. Molecular structure of [MnLn
2(QCl)
8] is presented in
Figure 9 [
50].
Heterometallic Ni
IIDy
III complexes generally exhibit field-induced single-molecule magnet behavior with an effective energy barrier of 6.6 - 16.9 K for reversal of the magnetization, i.e., [NiDy(valpn)(hfac)
2(N
3)(H
2O)
2] [
51], [NiDy(L)(ArCOO)(NO
3)
2] (LH
2 = N,
N′,
N″-trimethyl-
N,
N″-bis(2-hydroxy-3-methoxy-5-methylbenzyl)diethylenetriamine); Ar = benzyl, 9-anthracenyl) [
52], [Ni
2Dy
2(CH
3COO)
3(HL)
4(H
2O)
2](NO
3)
3 (LH
-=2-methoxy-6-[(E)-2′-hydroxymethyl-phenyliminomethyl]-phenolate) [
53], [Ni
2Dy
2(μ
4-CO
3)
2(3-MeOsaltn)
2(H
2O)
2(NO
3)
2] [
54] (3-MeOsaltn
- =
N,
N′-bis(3-methoxy-2-oxybenzylidene)-1,3-propanediaminato), [Ni
2Dy
2(L)
4(NO
3)
2(MeOH)
2] [
55] (L
2- = 2-(((2-hydroxyphenyl)imino)methyl)-6-methoxyphenolato), [R
3NH]
2[Ni
2Dy
2(μ
3-OH)
2(
tBuCOO)
10] [
56] and many others [
57,
58,
59]. Chosen Ni
IIDy
III complexes are presented in
Figure 10.
The compounds [Zn
2Ln(L)
2(NO
3)(SCN)
2] (where Ln
III = Ce
III, Nd
III; L
2- = 2,2'-(ethane-1,2-diylbis(nitrilomethylylidene))bis(6-methoxyphenolato)) based on a linear Zn
II–Ln
III–Zn
II motif with an axially stressed ligand field demonstrated the appearance of field-induced SMM behavior, which was correlated with the even-numbered
Jz sublevels of Ce(III) and Nd
III ions known as the Kramers system [
60]. Field-induced SMM with an estimated
Ueff barrier of 59K was [Zn
2Dy(L)(NO
3)
3(OH)], despite the fact that mononuclear [Dy(H
3L)(H
2O)(NO
3)](NO
3)
2 do not show slow relaxation of the magnetization [
61]. An uncommon example of heterometallic clusters is [Zn
2Dy
4(HL)
4(
o-vanillin)
2(OH)
4(CH
3OH)
2](NO
3)
2] that shows typical ferromagnetic single molecule magnetic behavior with a slow zero-field relaxation [
62].
In the first bifunctional SMM [ZnDy(NO
3)
3(L)(H
2O)] (where LH
2 =
N,
N’-bis(3-methoxysalicylidene)-1,2-diaminoethane), pictured in the
Figure 11, diamagnetic zinc cation provides the compound with two significant profits: it increases the negative charge of the phenolates and it elevates rare-earth cations’ crystal field splitting without the suppression of the Dy
III emission in the visible spectral window [
63,
64].
2.3. Catalysis, ring-opening polymerization and copolymerization of cyclic esters
Rare earth complexes offer the promise of catalytic performance of many organic reactions due to their high coordination numbers as well as their ability to change the coordination environment quickly. The catalytic applications of heterometallic rare earth complexes are mainly related to Shibasaki catalysts of the formula [M
3RE(binol)
3] (where M
I is alkali metal ion, binol
2- - binaphtholato as shown in
Figure 12). Since the beginning of the 1990s, Shibasaki catalysts have been mainly used in asymmetric synthesis, and in particular in the Michael addition, [
65] aldol [
66]/nitro-aldol [
67] condensation, cyclopropanation of enons, [
68] aldol–Tishchenko reaction, [
69] cyanation of aldehydes, [
70] reduction of 1,4-benzoquinones, [
71] Mannich reaction, [
72] protonation, [
73] Diels-Alder reaction, [
74] hydrophosphonation of imines and aldehydes [
75].
In asymmetric synthesis, the increase in enantioselectivity was mainly dependent on the type of MI, when the influence of the REIII was much lower. The effectiveness of these catalysts is due to their ability to behave as both a Lewis acid via REIII ions to activate electrophiles and Brönsted base via Mbinol to activate pronucleophiles, which is analogous to metalloenzyme reactions. The synergic action of the two metal centers allows for transformations that were not possible with conventional catalysts.
Schelter published the ethylzinc Shibasaki catalyst analogs [(EtZn)
3RE(binol)
3 (THF)
3][
76] (where RE = La
III, Pr
III, and Eu
III) presented in
Figure 13, which were used as catalysts for the enantioselective addition of ethyl groups to benzaldehyde in toluene at room temperature (yields 95% (La
III), 89% (Pr
III), 99% (Eu
III). The greatest control over the process (
ee = 70%) was achieved using an equimolar mixture of Zn
II3Eu
III complex, ZnEt
2, and triphenylphosphine oxide.
Dual activity of [NiRE(L)
3] [
77] (where RE = Lu
III, Y
III, La
III, L
- = (
iPr
2PCH
2NPh)
-) was discovered in the hydrogenation of diphenylacetylene (DPA) to (E)-stilbene (4.6 atm H
2 at 70 °C, 2.5 mol % of cat.). In the presence of DPA, [NiRE(L)
3] compounds catalyze the production of (Z)-stilbene (99-98% conversion after 24 h). In the absence of DPA they are responsible for the cis to trans isomerization of stilbene, with trans:cis ratio: >99:1 (for Lu
III and Y
III) or 20:80 (for La
III)), as presented in
Figure 14. In performed reactions, electron-rich Ni
II centers were engaged in alkyne binding, while RE
III ions provide an open, intramolecular coordination site to favor the hemilability of the phosphine ligand.
Another example was bimetallic rare-earth coordination polymers of formula {[Fe
7RE(Hpmida)
6]·2H
2O}
n [
78] (where RE
III = Eu
III, Dy
III, Ho
III, Y
III; H
4pmida = N-(phosphonomethyl)iminodiacetic acid), which were useful as highly selective (95-98%) heterogenous catalysts for Knöevenagel condensation, an important C-C coupling reaction widely used for the synthesis of fine chemicals and pharmaceuticals. Depending on the reactants chosen, the conversion reached 20% to 57% (27 h, toluene, 60 °C), with the most active europium and dysprosium. In the experiments performed, it turned out that the presence of RE
III improves the course of the reaction and significantly increases the conversion (from non-catalyzed 7-8%).
Heterometallic rare-earth complexes may also be used as initiators in the ring-opening polymerization (ROP) of cyclic esters or their copolymerization (ROCOP) with cyclic epoxides and anhydrides. ROP of cyclic esters initiated by metal catalysts permits the preparation of well-defined polyesters with strictly controlled molecular weight, dispersity, chain microstructure, and tacticity [
79]. ROP and ROCOP initiators featuring f-block metals have been developed to take advantage of their Lewis acidity and large coordination spheres. The impact of RE
III in monomer activation usually was observed for the alkali metal-rare earth initiators.
In the ε-caprolactone (ε-CL) polymerization studies initiated by [Na
8Sm
2(OCH
2CH
2NMe
2)
12(OH)
2], [K
12Na
4Sm
2(OCH
2CH
2NMe
2)
18(OH)
4] and [K
20RE
4(OCH
2CH
2NMe
2)
26(OH)
6] (where RE
III = Sm
III, Nd
III, Pr
III, Yb
III) it has been shown that the interaction of heterobimetallic centers (
Figure 15) allows for achieving within 1 min from 76 to 100% conversion of monomer at stoichiometry of ε-CL/I = 6.000 - 15.000, while homometallic alkoxides were inactive in the study process [
80].
Another example is [LiY(C
5Me
4SiMe
2NCH
2CH
2X)
2] [
81] (where X = OMe, NMe
2), which in the ROP of ε-CL in toluene at room temperature reached monomer conversion of 75-94% in just 90 minutes. Poly(ε-caprolactone)s obtained were of high molecular weight (M
n > 30 kDa) with the highest values over 170 kDa (for ratio [ε-CL]/[initiator] = 188/1) and dispersity (Đ) < 2.0. Reported complexes provided higher conversion and lower Đ than [Y(N(SiMe
3)
2)
3] for identical reaction conditions. Other examples of heterometallic initiators in ROP of ε-CL are mixed-metal allyl complexes of formula [RE(
η3-C
3H
5)
3(
μ-C
4H
8O
2)·Mg(
η1-C
3H
5)
2(
μ-C
4H
8O
2)
1.5]
n[
82] (where RE
III = La
III, Y
III). Both compounds are highly effective under mild conditions and reach monomer conversion of 99% (t = 0.3 min) and 61% (t = 1.3 min) with M
n values of 42.1 and 29.2 kDa with Đ ~1.4, respectively. The catalytic activity of both initiators was better than monometallic [Mg(
η1-C
3H
5)
2(BDI)(THF)] (BDIH = (2-(2,6-diisopropylphenyl)amino-4-(2,6-diisopropylphenyl)imino-2-pentene; [ε-CL]/[initiator] = 200/1, conversion 92% in 6 min, M
n = 13.6 kDa, Đ = 1.4). The interesting group of heterometallic complexes was obtained using Ferrocene-based tetradentate Schiff base ligands.[
83] The key finding of this study was the use of [Ce(phosphen)(THF)
2] (where phosphen = 1,10-di(2-tert-butyl-6-diphenylphosphiniminophenoxy)ferrocene) as redox control initiator for ROP of
l-lactide (
l-LA), which course was dependent on the oxidation state of the Ce
III/
IV ion [
84]. The Ce
III complex presented in
Figure 16 was active in
l-LA polymerization at 0 °C with a reagents stoichiometry of
l-LA/Ce = 100/1, and it allowed to achieve 96% monomer conversion over 0.5 h, whereas under these conditions Ce
IV compound was inactive. In turn, the isostructural Y
III compound was used as a model system to determine the influence of the oxidation state of Fe atoms in the ferrocene backbone on the Y
III–OR activity in the
l-LA polymerization. An initiator containing Fe
II atoms allowed to achieve 24% conversion of monomer in polymerization performed at a ratio of
l-LA/Y = 100 through 1h at 25 ° C, while the complex containing Fe
III centers was completely inactive in the studied reaction [
85].
Series of heterometallic Ni
II-RE
III complexes (where RE
III = Ce
III, Nd
III, Sm
III, Eu
III, Tb
III, Ho
III, Tm
III) with the acyclic Salen-type ligand (LH
2 = N,N’-bis(3-methoxysalicylidene)ethylene-1,2-diamine), with the structure shown in
Figure 17, were used as initiators for ROP of
l-LA in bulk at 160 °C for 12 h (for ratio [
l-LA][catalyst] = 1000/1). The presence of rare-earth ions was an influential factor because it effectively passivated the catalytic behaviors on the ROP, leading to increased molecular weights of obtained polymers (M
n = 28 961 – 31 555), and improved the polymerization control (Đ = 1.12 – 1.19). For the mentioned complexes, it was discovered that the catalytic activity is relative to the intramolecular Ni
II-RE
III separations without the lanthanide contraction sequence [
86].
Two-dimensional coordination polymer (
Figure 18) [CuEr(pdc)
2(Hpdc)(H
2O)
4]
n [
87] (where pdcH
3 = 3,5-pyrazole dicarboxylic acid) was used as a heterogeneous catalyst in the cyclopropanation of styrene with ethyldiazoacetate at room temperature in dichloromethane, leading to achieving high diastereoselectivity (84%) at the 12% conversion rate after 24 hours.
A series of heterometallic {[Cu
3Ln
2(L)
6(H
2O)
6]·10H
2O}
n [
88] (where Ln
III = La
III, Gd
III, Yb
III, Lu
III) coordination polymers were used for catalytic oxidation of olefins and aromatic benzylic substrates using
tBuOOH or O
2 as oxidants. The series of used compounds showed that their activity increases with the increase in the atomic number within the lanthanide group. The most active in the cyclohexene oxidation reaction (1,2-dichloroethane,
tBuOOH (70%), 75 °C, 0.001 mol% [Cu]) was the Cu
II-Lu
III complex (
Figure 19), which after 24 h led to the achievement of 60% substrate conversion, while the remaining compounds allowed to achieve conversions of 48% (Cu
II-La
III), 52% (Cu
II-Gd
III), and 57% (Cu
II-Yb
III). An identical trend in the reactivity of compounds was also maintained in the styrene oxidation reaction, in which the substrate conversion values were 75% for Cu
II-Lu
III, 64% for Cu
II-La
III, 69% for Cu
II-Gd
III, and 73% for Cu
II-Yb
III for the same reaction conditions.
2.4. CO2 conversion
In recent decades, rapid increases in atmospheric carbon dioxide concentrations, primarily driven by population growth, economic development, and energy consumption, have become a global concern. The elevated CO
2 levels, mainly from fossil fuel use, contribute to ecological imbalances, including temperature growth, melting snow cover, permafrost thaw, and rising sea levels [
89]. Consequently, mitigating CO
2 emissions and reducing atmospheric levels have become crucial global objectives to combat climate change. Efforts have been directed toward capturing, utilizing, and storing CO
2 to achieve decarbonization and emissions reduction goals. Carbon dioxide offers a broad spectrum of potential applications, ranging from direct uses in oil recovery, food processing, water treatment, fire retardants, coolants, and cleaning agents to chemical conversions into value-added products. CO
2 offers an accessible, non-toxic, low-cost, renewable carbon feedstock for producing chemicals, fuels, plastics, and raw materials. However, due to its low reactivity, carbon dioxide conversion needs suitable reagents, catalysts, or high-energy sources.
Most published studies use highly reactive substrates and harsh reaction conditions to overcome the high thermodynamic stability and chemical inertness of CO
2. Therefore, molecular catalysis has recently focused on developing more efficient systems that promote CO
2 transformation, especially under mild conditions, in order to reduce production costs and energy consumption. Numerous metal-based catalysts, including main group elements, transition elements, and rare-earth elements, have been used for the chemical fixation of CO
2 into value-added products. Both homogenous and heterogenous metal catalysts play a crucial role in CO
2 conversion reactions, examples of which are shown in
Figure 20. The synthesis of cyclic carbonates by cycloaddition of CO
2 and epoxides is one of the most studied and significant reactions in green and sustainable chemistry. In this reaction, the Lewis acidic RE
III site activates the epoxide molecule towards the nucleophilic attack of the Lewis base (X
- = Cl
-, Br
-, I
-), leading to the epoxide’s ring opening.
Then CO
2 insertion occurs, forming a carbonate intermediate that undergoes intramolecular ring closure to release the cyclic carbonate. Lanthanum complex [La(L)(THF)] stabilized by tris(phenolato) ligand (L
3- = 2,2’-[({2-[{[3,5-di-t-butyl-2-(hydroxy)phenyl]methyl}(methyl)amino]ethyl}azanediyl)bis(methylene)]bis(4,6-di-t-butylphenolato)}) is a rare example of catalyst that enables the cycloaddition of terminal epoxides with CO
2 under mild conditions (i.e., 25 °C, 1 bar CO
2, 0.3 mol% cat., TBAI 0.6 mol%), leading to cyclic carbonates with a yield of 49–99%, as seen in
Figure 21 [
90]. Several other studies have shown that synergistic interactions between different metal centers can improve the catalytic activity of the catalyst and the selectivity of this reaction [
91]. The commonly accepted mechanism assumes simultaneous activation of epoxide and CO
2 on adjacent metal centers.
For example, [ZnLa
2(OBn)
2(L)
2] for L
3- = 2,2’-[{[2-(2-oxidoethoxy)ethyl]azanediyl}bis(methylene)]bis(4,6-di-t-butylphenolato), a catalyst presented in
Figure 22, showed 2 to 8 times better catalytic activity in the reaction of CO
2 and 1,2-epoxyhexane (0.5 mol% cat., 1 mol% TBAB, 25 °C, 24 h, 1 atm CO
2) than [La
2(L)
2(THF)
2] or zinc aryloxide [Zn(OBn)
2] [
92].
Observed enhancement of the catalytic activity of heterometallic [EtZnY(L)(THF)] (L
4- = N,N,N’,N'-tetrakis(3,5-di-t-butyl-2-oxybenzyl)ethane-1,2-diamine)), occurs as a result of the presence of Zn
II centers, which allows for better delocalization of the electron density in the complex and drastically changes the energy barrier of the ring opening step by decreasing the electrostatic repulsion between yttrium center and bromine anion from the cocatalyst. Therefore, the epoxide conversion (0.2 mol% cat., 0.8 mol% TBAB, 40 °C, 18 h, 1 atm CO
2) of 81% for heterometallic catalyst was much better than for [Y(HL)(THF)] amounting to 41% or Zn(OAc)
2 16% [
93]. Another example of catalysts with a synergistic interaction between RE
III and Zn
II Lewis acidic sites were [Zn
4RE
2(
μ3-OH)
2L
4(OAc)
2(NO
3)
2(DMF)
2] (for RE
III = Dy
III, Nd
III, Tb
III; L
2- = N-[(3-methoxy-2-oxidophenyl)methylidene]pyridine-3-carbohydrazonato), which showed higher catalytic activity than equivalent amounts of a Zn
II salt, RE
III salts, and a ligand mixture of each. The structure of the mentioned complexes can be found in
Figure 23 [
94].
Excellent catalytic activity during cycloaddition between CO
2 and styrene oxide (0.01 mol% cat., 0.8 mol% TBAB, 80 °C, 1 bar CO
2) was also reported for a series of heterometallic clusters [Zn
2RE
2(
μ3-OH)
2L
4(NO
3)
4] (for RE
III = Eu
III, Tb
III, Er
III, Yb
III, Nd
III; L
- = 2-methoxy-6-(methoxycarbonyl)phenolato), which convert from 88 to 93% of epoxide within 14h.[
95] There are also several examples of heterometallic lanthanide−zinc clusters that are less efficient than their homometallic counterparts due to the steric effect of the ligands and crowded coordination environments [
96].
A reaction of great interest in the synthesis of cyclic carbonates is the oxidative carboxylation of olefins. However, so far, only a few lanthanide-based MOFs have been investigated as heterogeneous catalysts for this purpose. The application of [Nd
2(BIPA-TC)
1.5]
n, [Eu(H
2BIPA-TC)(BIPA-TC)
0.5]
n or [Tb(H
2BIPA-TC)(BIPA-TC)
0.5]
n as catalysts in the reaction of styrene, tert-butyl hydroperoxide, and CO
2, gave 80 – 87% cyclic carbonate after 10 h (0.18 mol % of MOF cat., 1 bar of CO
2, 80 °C). The proposed mechanism for this reaction can be found in
Figure 24 below [
97].
Much attention has been paid to the construction of C–N bonds through CO
2 fixation, but these reactions require high temperatures and pressures and the use of equivalent amounts of base. The bis(amidate) lanthanide complex (
Figure 25) [Ln
2L
2(N(SiMe
3)
2)
2(THF)
2] (Ln
III = Eu
III, Yb
III; L
- = N-(2,6-diisopropylphenyl)benzenecarboximidato) showed good catalytic performance in the synthesis of 2,4-quinazolidinones from CO
2 and 2-aminobenzonitriles (5 mol % cat., 5mol% DBU, 1 bar of CO
2, 100 °C, 24h) leading to final product with a yield of 61 or 91% [
98].
Other rare earth amides [RE
2L
2(N(SiMe
3)
2)
2(THF)
2] (RE
III = La
III, Nd
III, Y
III; LH
2 = N, N’-(cyclohexane-1,2-diyl)bis(4-tert-butylbenzamide) turned out to be effective catalysts for the direct carboxylation of terminal alkynes at ambient pressure, leading to the formation of the C–C bond and the synthesis of acetylenic carboxylic acids with a yield of 80-89% (phenylacetylene (1.0 mmol), Cs
2CO
3 (3.0 mmol), cat. (0.04 mmol), 1 atm of CO
2, 60 °C, DMSO (10 mL)) [
99].
The transformation of CO2 into macromolecular compounds can be carried out by direct copolymerization of CO2 with epoxides/aziridines, polycondensation with amines, alcohols, and amino alcohols, or by the synthesis of CO2-based monomers that will be used in polymerization reactions dependent on the nature and functionality of the appropriate monomer. These synthetic routes enable the production of a wide range of polycarbonates (PC), polyurethanes (PU), polyureas (PUA), and polyesters using versatile polymerization techniques, including ring-opening polymerization (ROP), ring-opening copolymerization (ROCOP), polycondensation and terpolymerization. However, for synthesizing CO2-based polymers, the low reactivity of CO2 or its derivatives requires elevated temperatures, removal of by-products, and activation of the monomers by multiple metal centers to complete the reaction.
Direct polymerization routes to CO
2-based polymers are mainly limited to the ROCOP of CO
2 with epoxides or aziridines. CO
2 is also used as a comonomer in the synthesis of multiblock copolymers by sequential monomer addition and tandem approach [
100].
In the indirect approach, CO
2 is first converted into linear or cyclic building blocks, which are then used to synthesize polymeric materials. Cyclic monomers (carbonates, carbamates) can be ring-opening polymerized to form the corresponding homopolymers (PCs, PUs). Cyclic 5-membered carbonates can also copolymerize with diols, diamines, diamines, and diols or cyclic ureas to form PCs, PUAs, or PUs [
101]. Copolymerization of cyclic carbonates with lactones is also widely studied to obtain copolyesters.
Copolymerization of CO
2 and epoxides is the most investigated method of synthesis of PCs [
102]. High molecular weight polycarbonates exhibit properties suitable for replacing petrochemical polymers in sectors including packaging, coatings, rigid plastics, and medical materials. Recently, developing and understanding the catalytic performance of heterometallic catalysts that exhibit metal synergy are among the most frequently studied aspects of the synthesis of biodegradable polymers. It was shown that some heterometallic complexes having two different metal centers showed much higher activity than their homometallic counterparts. It is assumed that different metals play different roles in copolymerization; the stronger Lewis acid activates the epoxide, and the softer Lewis acid forms a labile bond with the carbonate of the propagating chain, making the carbonate more nucleophilic. Several heterometallic zinc-lanthanide complexes have been shown to initiate the copolymerization of CO
2 and cyclohexene oxide (CHO).
For example, the heterometallic cluster pictured in
Figure 26, [Zn
2Nd
2(
μ-OBn)
2(L)
2] (for L
2- = (((2-(bis(3,5-di-t-butyl-2-oxidobenzyl)amino)phenyl)amino)methyl)-4,6-di-t-butylphenolato)) gave polycarbonates with high molecular weight (Mn up to 295.8 kDa) with narrow dispersity (Đ = 1.65) and high selectivity (99%) at [CHO]:[cat.] = 2000:1 ratio, 25 °C, 12h and 7 bar CO
2, and its reactivity was nine times greater than that of the isostructural yttrium catalyst [
103]. Using [ZnLn(L)(L
’)(NSiHMe
2)
2] (for L
- = (cyclohexane-1,2-diylbis(azanylylidenemethylylidene))bis(6-methylphenolato); L’
- = 2,2’-{[(2-methoxyethyl)azanediyl]bis(methylene)}bis(4,6-di-t-butylphenolato) as catalysts for CO
2/CHO copolymerization, it was shown that the radius of the RE
III ion significantly influences their activity. The most effective catalysts turned out to be Zn
IIDy
III and Zn
IISm
III complexes containing lanthanide ions with a moderate ionic radius, leading to polycarbonates with M
n = 148 kDa, Đ = 1.52–1.62, and selectivities of 99% (carbonate bonds). Another example was heterometallic Zn
3RE clusters (RE
III = La
III, Ce
III, Pr
III, Nd
III, Sm
III, Eu
III, Gd
III, and Dy
III) based on macrocyclic tri(salen) ligands, which showed a unique and rapid exchange of intra- and intermolecular acetate ligands. Lanthanide complexes with larger ionic radii (La
III, Ce
III, Pr
III, Nd
III) showed higher catalytic activity than those based on smaller lanthanides (Sm
III, Eu
III, Gd
III, Dy
III). The most active of them was Zn
3Ce, which enables the synthesis of polycarbonates with M
n = 14 kDa, Đ = 1.3 at 100 °C within 3 h [
104].
All the above-mentioned heterometallic catalysts for CO
2 copolymerization operate via the chain-shuttling mechanism in which the Lewis acidic RE
III enhances CHO coordination, and the unstable Zn-carbonate bond enhances nucleophilic attack to open the epoxide ring. The resulting RE-alkoxide and CO
2 coordinated to Zn
II form a carbonate complex, leading to the growth of the polycarbonate chain [
105]. Introducing Co
II instead of Zn
II into heterometallic complexes led to the discovery of one of the most efficient multimetallic catalysts, Co
3Nd, which had a turnover number (TON) of 13,000. This gave a polymer with >99% selectivity of carbonate bonds, M
n = 114 kDa and Đ = 1.05 at 2 MPa CO
2, 130 °C after 8 h. Pathway of that reaction may be found in
Figure 27 [
106].
2.5. Catalysts for energy conversion processes
Recently, particular attention was placed on using 3d-4f clusters as promising catalysts for energy conversion processes like hydrogen evolution reaction (HER), oxygen evolution reaction (OER), overall water splitting, and CO
2 reduction. The synergic cooperation between both RE
III/M
II centers led to their enhanced activity. For example, [Ni
36Gd
102(
μ3-OH)
132(L)
18(L’)
18(H
2L’)
24(OAc)
84(SO
4)
18(NO
3)
18(H
2O)
30]Br
6,(NO
3)
6 (Ni
36Gd
102, LH = 2-mercapto-5-methyl-1,3,4-thiadiazide; L’H
3 = 2,2-dimethylol propionic acid) found in
Figure 28 shows remarkable activity in photocatalytic CO
2 reduction, providing a TON of 29700 and a turnover frequency (TOF) of 1.2 s
−1 over 10 h with a selectivity of 90.2% for CO formation. This performance is much better than those of most homogeneous CO
2-reduction catalysts because the Lewis-acidic Gd
III modulates the electronic structure of the catalytic Ni
II centers, enhancing photocatalytic activity [
107].
[Co
12Eu
36(
μ4-O)
6(
μ3-OH)
84(OAc)
18(Cl)
2(NO
3)]
33+ shows effective water oxidation activity under acidic conditions (TOF of 1.5 s-1 at 1.8 V) owing to the synergistic effect of Eu
III and Co
II ions on O–O bond formation [
108]. Heterometallic cooperativity in water oxidation has also been reported for [Mn
2RE
2(O
2CMe)
6(pdmH)
2(L)](NO
3) (RE
III = Dy
III and Gd
III; pdmH
2 = 2,6-pyridine dimethanol; LH
2 = (6-hydroxymethylpyridin-2-yl)-(6-hydroxymethylpyridin-2-ylmethoxy-methanol)) which structure can be found in
Figure 29 [
109].
The series [Co
3RE(hmp)
4(OAc)
5(H
2O)], where RE = Ho
III, Er
III, Tm
III, or Yb
III (hmpH = 2-(hydroxymethyl)pyridine)), [
110] and [NdCo
3(btp)
2(OAc)
2(NO
3)
2](NO
3) (btp = 2,6-bis(1,2,3-triazol-4-yl)pyridine) anchored in phospho-doped graphitic carbon nitride, [
111] are further examples of water-oxidation catalysts, being mimetics of the {CaMn
4O
5} oxygen-evolution complex in photosystem II. [RE
52Ni
56(IDA)
48(OH)
154(H
2O)
38]
18+ clusters (RE = Pr
III, Eu
III, Gd
III; IDA = iminodiacetate) supported on CdS forms lanthanide–transition metal catalysts RE
52Ni
56-xCd
x/CdS that show activity for photocatalytic hydrogen evolution. The high photocatalytic efficiency of 25353 μmol h
−1 g
−1 was achieved usingNi
56-xCd
xEu
52/CdS, which enhances H
2 production under visible-light irradiation (≥ 420 nm) owing to the formation of catalytic Eu
II centers by the transfer of photoexcited electrons from CdS to the LUMO of Eu
III [
112]. Furthermore, the iodide-templated 3D-coordination polymer {[Cu
5Eu
2(OH)
2(pydc)
6(H2O)
8]·I
8}
n (pydc = pyridine-2,6-dicarboxylate) has been demonstrated to be an efficient photocatalyst for H
2 production under UV irradiation, providing H
2 evolution at a rate of 2262 μmol h
−1 g
−1 [
113].
2.6. Molecular precursors of functional inorganic materials.
The use of heterometallic 3d−4f compounds as single-source molecular precursors for the synthesis of functional inorganic materials is also very limited. One of the few examples is the series of isostructural compounds [Fe
2Ln
2((OCH
2)
3CR)
2(O
2C
tBu)
6(H
2O)
4] (Ln = La, Gd and R = Me, Et), which were used to prepare lanthanide orthoferrite perovskites LnFeO
3 [
114]. Another example is the use of an equimolar mixture of (NH
4)[Ln(EDTA)] (Ln
III = Pr
III, Sm
III, Eu
III, Gd
III, Dy
III, Er
III) and (NH
4)
3[V(O)
2(EDTA)] at 800 °C for the preparation of lanthanide vanadates (LnVO
4) [
115]. Furthermore, the solid-phase thermal decomposition of [Ln(Mn(CO)
3Cp
COOH)
2(OAc)(MeOH)]
n (Ln
III = Nd
III, Gd
III, Dy
III), [Ln
2(Mn(CO)
3Cp
COOH)
4(OAc)
2(H
2O)
4] (Ln
III = Ho
III, Er
III, Tm
III), and [Ln
2(Mn(CO)
3Cp
COOH)
4(NO
3)
2(DME)
2] (Ln
III = Eu
III, Tb
III) at 670–900 °C has been investigated for the synthesis of metamagnetic LnMn
2O
5 [
116,
117].
Various synthesis methods have been explored in the literature to synthesize La
2CuO
4. One of them involves the thermal decomposition of an amorphous precursor [La
2Cu(DTPA)
1.6]·6H
2O[
118] (H
5DTPA = diethylenetriaminepentaacetic acid) initially at 450 °C to eliminate volatile organic constituents, and then by heat treatment at 650 °C yielded La
2CuO
4 with an average crystallite size of around 29 nm. An alternate approach to synthesize La
2CuO
4 included hydrolyzing a mixture of [La
4(CO
3)(O
2CNBu
2)
10] and [Cu(O
2CNBu
2)(py)
2] (py = 4-dimethylaminopyridine) in a toluene solution, resulting in the isolation of the compound [La
2Cu(CO
3)
4]·5H
2O [
119]. The subsequent procedure involved heating the isolated precursor at 600°C, yielding a tetragonal polymorph with an average crystallite size of 15 nm. Above 850°C, the tetragonal form transformed into a rhombohedral phase with crystallites of sizes of 50 nm.
Recently, we have developed a simple and efficient synthetic strategy for the preparation of industrially important heterometallic perovskite-type materials of LaMnO
3, GdMnO
3, NdMnO
3, Pr
0.9MnO
3 (
Figure 30)
, and PrCoO
3 by thermal decomposition of heterometallic 3d–4f alkoxides [Mn
2Ln
4(µ
6-O)(µ
3-OR)
8(HOR)
4Cl
6] (Ln = La
III, Nd
III, Gd
III); [Co
2Pr
4(µ
6-O)(µ
3-OR)
8(HOR)
2Cl
6]; and [Mn
2Pr
4(µ
3-OH)
2(µ
3-OR)
4(µ-OR)
4(µ-Cl)
2(HOR)
4Cl
6], which molecular structures can be found in
Figure 31, and external [MnCl
2(HOR)]
n or [Co
4(µ
3-OR)
4(HOR)
4Cl
4] at 1100 °C [
30]. When we performed the thermolysis of [Ni
2Pr
4(µ
6-O)(µ
3-OR)
8(HOR)
4Cl
6] and [NiCl
2(HOR)
2] the formation of a mixture of the homo- and heterometallic oxides PrOCl, PrO
2, NiO, PrNiO
3, and Pr
2NiO
4 was observed. The representative PXRD pattern of Pr
0.9MnO
3 is presented in
Figure 32. The synthesis of 3d–4f precursors was performed using an uncommon synthetic method involving the reaction of metallic lanthanides (Ln
III = La
III, Pr
III, Nd
III, Gd
III) with divalent transition metal chlorides (MCl
2, where M = Mn
II, Ni
II, or Co
II) using 2-methoxyethanol (ROH) as the solvent and ligand precursor.
Another work reported that group 4–lanthanide ethoxides are attractive starting materials for the production of pyrochlore type phases Ln
2M
2O
7 of considerable interest because of their use as materials for thermal coatings of turbine components for protection against hot and corrosive gas streams [
120], high-temperature electrolytes in solid-oxide fuel cells [
121], or radiation-resistant materials [
122]. Thermal decomposition of the isostructural compounds [Ln
2Ti
4(µ
4-O)
2(µ
3-OEt)
2(µ-OEt)
8(OEt)
6(HOEt)
2Cl
2] (Ln
III = La
III, Nd
III) at 950 °C gave La
0.66TiO
3 or a mixture of Nd
4Ti
9O
24 and Nd
0.66TiO
3. Calcination of [M
2La
2(µ
3-O)(µ-OEt)
5(µ-Cl)(OEt)
2(HOEt)
4Cl
4]
n and [M
4Nd
4(µ
3-O)
2(µ-OEt)
10(µ-Cl)
4(OEt)
8(HOEt)
10Cl
2] (M
IV = Zr
IV, Hf
IV) at 950–1500 °C led to the selective formation of heterometallic La
2Zr
2O
7, La
2Hf
2O
7, Nd
2Zr
2O
7, and Nd
2Hf
2O
7 phases, respectively [
31].