Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2024 Apr 3;146(15):10367–10380. doi: 10.1021/jacs.3c12721

N-Heterocyclic Carbene to Actinide d-Based π-bonding Correlates with Observed Metal–Carbene Bond Length Shortening Versus Lanthanide Congeners

Conrad A P Goodwin †,‡,§, Ralph W Adams §, Andrew J Gaunt †,*, Susan K Hanson †, Michael T Janicke †, Nikolas Kaltsoyannis ‡,§,*, Stephen T Liddle ‡,§, Iain May †, Jeffrey L Miller †, Brian L Scott ∥, John A Seed ‡,§, George F S Whitehead §
PMCID: PMC11029940  PMID: 38569081

Abstract

graphic file with name ja3c12721_0014.jpg

Comparison of bonding and electronic structural features between trivalent lanthanide (Ln) and actinide (An) complexes across homologous series’ of molecules can provide insights into subtle and overt periodic trends. Of keen interest and debate is the extent to which the valence f- and d-orbitals of trivalent Ln/An ions engage in covalent interactions with different ligand donor functionalities and, crucially, how bonding differences change as both the Ln and An series are traversed. Synthesis and characterization (SC-XRD, NMR, UV–vis–NIR, and computational modeling) of the homologous lanthanide and actinide N-heterocyclic carbene (NHC) complexes [M(C5Me5)2(X)(IMe4)] {X = I, M = La, Ce, Pr, Nd, U, Np, Pu; X = Cl, M = Nd; X = I/Cl, M = Nd, Am; and IMe4 = [C(NMeCMe)2]} reveals consistently shorter An–C vs Ln–C distances that do not substantially converge upon reaching Am3+/Nd3+ comparison. Specifically, the difference of 0.064(6) Å observed in the La/U pair is comparable to the 0.062(4) Å difference observed in the Nd/Am pair. Computational analyses suggest that the cause of this unusual observation is rooted in the presence of π-bonding with the valence d-orbital manifold in actinide complexes that is not present in the lanthanide congeners. This is in contrast to other documented cases of shorter An–ligand vs Ln–ligand distances, which are often attributed to increased 5f vs 4f radial diffusivity leading to differences in 4f and 5f orbital bonding involvement. Moreover, in these traditional observations, as the 5f series is traversed, the 5f manifold contracts such that by americium structural studies often find no statistically significant Am3+vs Nd3+ metal–ligand bond length differences.

Introduction

Our understanding of the chemical bonding and coordination chemistry of the lanthanide (Ln) and actinide (An) elements has evolved substantially over the last century.1−7 Through this period, the postulate that Ln3+ cations operate in a highly ionic bonding regime, comparable to that of alkaline earth metals, has remained essentially unchanged.3 The core-like 4f manifold is too radially contracted to engage in substantial spatial-overlap-driven covalent interactions with bound ligands. In contrast to the lanthanides, many actinides have multiple readily accessible oxidation states, and the greater radial expansion of the 5f manifold (versus 4f) can result in appreciable interaction with ligand orbitals,8 though this decreases across the 5f row as Zeff increases.9−20 Complexes in higher oxidation states (e.g., IV–VI), particularly those of uranium which are more studied than transuranium complexes, frequently show that the more radially diffuse 6d manifold (versus 5f) is capable of accepting ligand density and participates in multiple bonding interactions such as in actinyl or mono-oxo ({AnOx}n+) or bis-({An(NR)2}n+) linkages,21−24 and in other An=E combinations.7,25−33 In some instances, the covalency of An=element multiple bonds, featuring rich 5f, 6d, and 7p components, can approach or surpass that of transition metal complexes.34

Differences between lanthanide and actinide complexes, brought about, at least in part, by variations in metal–ligand interactions, are exploited in separations science applications.35−41 However, crystallographic studies comparing soft and hard σ-donor binding in metal pairs with similar ionic radii (e.g., 6-coordinate Ce3+ = 1.01 Å; U3+ = 1.025 Å) often show that for hard (e.g., N, O) σ donors, the two metals behave similarly and display minimal bond length differences. In studies of soft donors (e.g., Se, Te, and N-heterocycles) with f-block metals, bonding differences are often observed between the lanthanide and actinide series; however, to the best of our knowledge,11,14−18,20,42−47 in every homologous series studied thus far by single-crystal X-ray diffraction (SC-XRD), the magnitude of differences decreases to statistical insignificance as the f-block is traversed. This effect is such that by Am, an element challenging to separate/chemically distinguish from lanthanides, Am–ligand versus lanthanide–ligand bond length differences are usually statistically insignificant, at least with conventional statistical treatments.48 Nevertheless, these differences are exploited in Am3+, Cm3+, and Ln3+ separation schemes.36−41 Turning to π-bonding, where both metals are trivalent, Ln3+ ions are much less likely to have (energetically and spatially) accessible orbitals to engage in such interactions than An3+ ions.49 With π-acids (e.g., CO, {CN}−), actinide–ligand bonds are almost invariably shorter than lanthanide–ligand bonds,50,51 but much more frequently, the corresponding lanthanide complexes are simply not isolable.52−54 The prospect of U3+ 5f → L [or involving ligand symmetry-matched orbitals from the (U3+L3) fragment] π back-bonding has been advanced as a plausible mechanism to explain some of these differences.52,55−61 The opposite case, of L → M π-bonding, is rarely documented outside of multiply bonded species, which are challenging to isolate for trivalent lanthanides, hindering comparative studies. In the case of amido and alkoxide systems which may permit L → M π-donation, reports show that where present this involves donation to the 5f manifold.62,63 In some simple systems such as the [MCl6]3− series, differences attributable to L → An π (L → 6d and 5f) contributions have been described.64

Furthermore, if π back-bonding from metal f orbitals drives an observable effect, then transuranium actinides might be expected to show a weaker effect than uranium due to radial contraction from larger Zeff; thus, a homologous series that spans several trivalent actinides is necessary to establish the nature of π-bonding and how it changes as the 5f row is traversed—few such homologous series are well studied.11,14−18,20,43−47,50,65−71

NHCs (NHC = N-heterocyclic carbene) are classically regarded as strong CNHC → M σ2 donors which means they will bind tightly and form isolable complexes offering a promising avenue to address some of the hindering factors above. Indeed, some transuranium NHC-complexes have recently been reported.25 Where orbital occupancy and overlap allow, synergic M → NHC π back-donation from low-valent transition metals is often observed (Figure 1 left). For Ln3+, the valence electrons, where present, reside in highly spatially contracted 4f-orbitals and so M → NHC back-donation is likely to be weak or absent. The greater radial extent of the 5f valence orbital set on An3+ (relative to 4f) may be significant enough to drive observable physical differences, such as bond length differences, derived from the differing f-orbital properties.51,56 Furthermore, many NHCs show nonzero 2pz density at the carbenic carbon, which in principle can give rise to CNHC → M π-bonding interactions – i.e., σ2πx bonds, where x is small but nonzero (Figure 1 right).72−80 The somewhat (relative to 4f) radially expanded 5f and 6d (vacant) orbital sets on An3+ ions could both provide spatial overlap to form CNHC → M π-bonding interactions. As such, f-block NHC complexes present a potential testbed for the investigation of lanthanide/actinide structural differences derived from both f-based M → L π back-bonding and L → M f- or d-based π-bonding.

Figure 1.

Figure 1

Open in a new tab

Various bonding interactions between an NHC ligand and d-orbitals.

Ephritikhine and co-workers have explored the binding of the simple NHC {C(NMeCMe)2}, denoted IMe4 henceforth, to both Ce3+ and U3+ within the [M(Cpt)3(IMe4)] (Cpt = {C5H4-tBu}) and also [M(Cp*)2(I)(IMe4)] (Cp* = {C5Me5}) frameworks.51 High-quality single-crystal X-ray diffraction data were obtained from reactions using just ∼25-35 mg of starting material, and the data showed that the U–CNHC distance was 0.037 Å shorter than Ce–CNHC; note that (6-coordinate) Ce3+ is ca. 0.02 Å smaller than U3+, which hinders an ideal comparison. Here, we report the preparation of [M(C5Me5)2(X)(IMe4)] (X = I, M = La, Ce, Pr, Nd, U, Np, Pu; X = Cl, M = Nd; X = I/Cl, M = Nd, Am) complexes which show unusually large differences in the M–CNHC bond length between members of the trivalent lanthanide and actinide series with similar ionic radii, and those differences do not significantly reduce as the f-block series is traversed to include Nd3+/Am3+ complexes.

Results and Discussion

Synthesis of [M(Cp*)2(I)(THF)] Complexes

Lanthanide [Ln(Cp*)2(I)(THF)] (1M, M = La, Ce, Pr, and Nd) complexes were synthesized on a ca. 150 μmol scale in an argon-filled inert-atmosphere glovebox at room temperature. THF was added to a solid mixture of binary MI3 and a slight excess (2.2 equiv) of KCp* in a glass scintillation vial with a Teflon-coated stirrer bar (Scheme 1). Workup and crystallization from warm toluene with a drop of THF (12 mg) gave free-flowing plank-shaped crystals of each complex in poor to fair yield (ca. 30–50%), though the exact % yields depend upon the amount of coordinated THF in each batch of “M(Cp*)2(I)(THF)m” (M = La, Ce, Pr, Nd; m = 0 to 1) which remained after drying (see Supporting Information). The corresponding actinide complexes 1Np and 1Pu were synthesized using the previously reported procedure, beginning with [AnI3(THF)4] (An = Np, Pu),68 and were used as crude products without recrystallization for subsequent reaction steps.

Scheme 1. Synthesis of Isolated Crystalline [M(Cp*)2(I)(THF)] (1M, M = La, Ce, Pr, Nd, Np, and Pu) from Trivalent Iodide Precursors.

Scheme 1

Open in a new tab

Synthesis of [M(Cp*)2(I)(IMe4)] Complexes

Due to the poor solubility and stability of 1M (M = La, Ce, Pr, Nd, Np, and Pu) in noncoordinating solvents such as toluene and hexane, we opted to synthesize all [M(Cp*)2(I)(IMe4)] complexes (2M, M = La, Ce, Pr, Nd, Y, Np, Pu, and Am) from 1M prepared in situ/as crude material (Scheme 2). A slight excess of solid IMe4 (1.05 equiv) was added to the stirred solutions of 1M in toluene. Workup and low-temperature (−35 °C) crystallization from the reaction solvent gave 2M in modest yield (30–40%) as well-isolated large plank-shaped crystals in all cases. We have included the yttrium compound, 2Y, here as a smaller member of the series for comparative purposes, but the isolation of both 1Y and 1Am was not attempted (vide infra).

Scheme 2. Synthesis of Isolated Crystalline [M(Cp*)2(I)(IMe4)] (2M, M = La, Ce, Pr, Nd, Y, Np, and Pu) from 1M, [M(Cp*)2(I)(THF)], Prepared in Situ.

Scheme 2

Open in a new tab

IMe4 = {C(NMeCMe)2}. Note that for M = Am, the X-atom depicted is a mixture of iodide and chloride.

Rare earth 2M compounds (M = La, Ce, Pr, Nd, and Y) were assessed by elemental analysis and found to be analytically pure and in agreement with their [M(Cp*)2(I)(IMe4)] formulation. In the case of 2M for Np and Pu, we can only judge their purity by 1H and 13C{1H} NMR spectroscopies (vide infra). For the synthesis of 2Am, we attempted to generate putative “AmI3(DME)m” in analogy to our previous synthesis of the chloride congener (Scheme 2).14,81 The putative “AmI3(DME)m” was used as-synthesized for subsequent reaction steps–see Supporting Information for details. Instead of the anticipated Am3+ analogue of the other 2M complexes, we isolated golden orange plank-shaped crystals of [Am(Cp*)2(InCl1–n)(IMe4)] (4Am), where n is approximately 0.65 as determined independently by both 1H NMR spectroscopy and single-crystal X-ray diffraction (vide infra). This suggests that the treatment of the putative “AmCl3(DME)m” precursor with Me3Si–I did not result in complete substitution of Cl for I. The corresponding Nd3+ complex [Nd(Cp*)2(InCl1–n)(IMe4)] (4Nd, where n is also approximately 0.65) was then synthesized analogously starting from Nd3+ in aqueous 6 M HCl solution. Finally, to aid the interpretation of the bonding data in these mixed I/Cl 4M complexes, [Nd(Cp*)2(Cl)(IMe4)] (5Nd) was synthesized in low yield (24%) over two steps using binary NdCl3 as the metal source.

Molecular Structures

Single-crystal X-ray diffraction studies revealed all four new 1M complexes to be isomorphous to the previously reported 1U, 1Np, and 1Pu (except 1La for which the a axis is doubled in length), and the Sm,82 Dy,83 and Yb analogues.84 All crystallize in the triclinic space group P1̅ with two molecules per asymmetric unit (Z′ = 2, except 1La for which Z′ = 4)—the structure of 1Ce is depicted in Figure 2. Detailed discussion of structural metrics for 1M is provided in the Supporting Information and is summarized in Tables 1 & 2 and Figure 4 to facilitate ready comparison to 2M complexes (vide infra).

Figure 2.

Figure 2

Open in a new tab

Molecular structure of 1Ce. Ellipsoids set at 50% probability and H atoms, along with a second unit containing Ce(2), removed for clarity (operations: X, Y, Z). Ce(1)–I(1) = 3.1027(8) Å; Ce(1)–O(1) = 2.511(5) Å; Ce(1)–Cpcent = 2.497(5) Å; Ce(1)–Cpcent = 2.506(4) Å; Cpcent–Ce–Cpcent = 136.04(11)°.

Table 1. Bond Lengths (Å) and Angles (deg) for M(1) in 1M (M = La, Ce, Pr, Nd, U, Np, and Pu)a.

  1Lab 1U(66) 1Ce 1Np(68) 1Pr 1Pu(68) 1Nd
M–I 3.1400(6) 3.0636(6) 3.1027(8) 3.0560(14) 3.0851(8) 3.0353(7) 3.0618(10)
M–O 2.536(4) 2.496(4) 2.511(5) 2.504(13) 2.491(5) 2.460(5) 2.461(8)
M–Cpcent 2.562(4) 2.476(4) 2.497(5) 2.530(5) 2.473(5) 2.455(6) 2.467(6)
  2.542(16) 2.494(4) 2.521(4) 2.561(5) 2.500(5) 2.463(5) 2.471(7)
Cpcent–M–Cpcent 138.08(17) 135.56(9) 136.04(11) 135.12(11) 135.92(12) 135.49(13) 135.48(16)
Open in a new tab
a

Due to different numbering conventions across some examples, “M(1)” is denoted here as that with the shortest M–I distance, and numbering proceeds from there. Except for 1La, all are isomorphous. However, 1La is isostructural.

b

Only the largest component of the disordered La–Cp rings is listed.

Table 2. Comparison of M–I and M–O Bond Lengths (Å) between 1M (M = La, Ce, Pr, U, Np, and Pu) Complexes.

  M–I
 M–O
  M(1) M(2) M(1) M(2)
1La 3.1400(6) 3.1492(5) 2.536(4) 2.531(4)
1U 3.0636(6) 3.0955(6) 2.496(4) 2.486(4)
Δ(La–U) 0.0764(8) 0.0537(8) 0.040(6) 0.045(6)
1Ce 3.1027(8) 3.1272(7) 2.511(5) 2.507(4)
1Np 3.0560(14) 3.0832(12) 2.504(13) 2.472(12)
Δ(Ce–Np) 0.0467(8) 0.0440(7) 0.007(6) 0.035(6)
1Pr 3.0851(8) 3.1070(8) 2.491(5) 2.499(5)
1Pu 3.0353(7) 3.0594(6) 2.460(5) 2.463(6)
Δ(Pr–Pu) 0.0498(1) 0.0476(1) 0.031(6) 0.036(6)
Δ(Ce–Pu) 0.0674(11) 0.0678(9) 0.051(7) 0.044(7)
Open in a new tab

Figure 4.

Figure 4

Open in a new tab

M3+ ionic radius (calculated 8-coordinate)14 vs the M–E bond length for M–I with both 1M and 2M, and M–CNHC for 2M and 4Am. Open symbols are used for Ln complexes (La, Ce, Pr, Nd, and Y from left to right), and solid symbols are used for An complexes (U, Np, Pu, and Am from left to right). Solid lines denote a linear fit of the data points (see Supporting Information for statistical parameters), and shaded areas show the 50% confidence interval for extrapolated data. For the 1M series, only metal site (1) is shown here, which corresponds in all cases to the shorter M–I length, to minimize the differences and not overstate any conclusions. See Supporting Information for additional data plots.

The NHC-ligated complexes 2M (M = La, Ce, Pr, Nd, Y, Np, and Pu) and [M(Cp*)2(IxCl1–x)(IMe4)] (M = Nd, 4Nd; Am, 4Am) isolated herein are isomorphous with the previously reported 2U and 2Ce analogues,51 crystallizing in the monoclinic space group P21/c with one molecule per asymmetric unit. In the case of 2La and 2Ce an additional polymorph was identified (see Supporting Information Tables S14 and S15 for more details). The structures of 2Pu and 4Am are shown in Figure 3. These complexes display a pseudo-tetrahedral bent-metallocene structure with a staggered Cp*···Cp* arrangement, like 1M, as would be expected for a simple replacement of the THF moiety with the IMe4 NHC. Crystallographically characterized Pu–NHC complexes, and Pu–C σ-bonds in general, remain extremely scarce;50,67,69,85 also, to the best of our knowledge, 4Am represents the first structural authentication of an Am–C σ-bonding interaction to any organic ligand.

Figure 3.

Figure 3

Open in a new tab

Molecular structure of 2Pu and 4Am. Ellipsoids set at 50% probability and H atoms removed for clarity (operations: X, Y, Z). In 4Am, Cl(1) has been removed for clarity and the I(1) occupancy (0.65) has been refined competitively against Cl(1) (0.35). Pu(1)–I(1) = 3.0938(7) Å; Pu(1)–C(1) = 2.637(8) Å; Pu(1)–Cpcent = 2.489(4) Å; Pu(1)–Cpcent = 2.510(4) Å; Cpcent–Pu(1)–Cpcent = 135.47(13)°. Am(1)–I(1) = 3.0666(11) Å; Am(1)–C(1) = 2.631(3) Å; Am(1)–Cpcent = 2.4744(17) Å; Am(1)–Cpcent = 2.4895(15) Å; Am(1)–Cl(1) = 2.682(8) Å; and Cpcent–Am–Cpcent = 135.64(6)°.

To investigate the influence of the mixed Cl/I occupancy in 4Am on the overall structure, the Nd-congener, 4Nd, was structurally characterized as well as the pure chloride Nd-analogue, [Nd(Cp*)2(Cl)(IMe4)] (5Nd). The Cpcent–M–Cpcent angles for all 2M lie over an exceptionally narrow range of ca. 135.20(8)–135.89(6)° (135.61(9)° using the mean and a weighted standard deviation as the error).48 The M–Cpcent ranges in 2M [e.g., 2.5576(17)–2.5708(15) Å for 2La and 2.4941(13)–2.5081(12) Å for 2Nd] are nevertheless very close and on the order of only ca. 0.03 Å longer than 1M, while statistically distinct from those in the corresponding 1M complexes. The Cpcent–M–Cpcent angles for 4Nd (135.18(11)°) and 4Am (135.64(6)°) are very close or actually overlap with the average for all others (including 2Y) at 135.61(9)°.48,51 Similarly, both display M–CCp ranges (4Nd, 2.741(7)–2.816(7) Å; 4Am 2.718(3)–2.802(3) Å), which follow smoothly from the previous members of their respective series, where the small decrease in length is commensurate with the reduction in ionic radius. Tables 3 and 4 summarize pertinent lengths and angles within the 2M series and 4Am/4Nd; see the Supporting Information for a comparison of 2La and 2Ce to their polymorphs (2Laβ and 2Ceβ).

Table 3. Experimental and (in Italics) Computed (Vide Infra) Bond Lengths (Å) and Angles (deg) for 2M (M = La, Ce, Pr, Nd, Y, U, Np, and Pu), and 4M (M = Nd, Am).

  2La 2U(51) 2Cea 2Np 2Pr 2Pu 2Nd 4Nd 4Am 2Y
M–I 3.1769(3) 3.1266(4) 3.1584(6) 3.1003(7) 3.1393(6) 3.0938(7) 3.1209(2) 3.103(2) 3.0666(11) 3.0409(4)
  3.140 3.095 3.118 3.081 3.100 3.074 3.085 –b –b  
M–CNHC 2.751(4) 2.687(5) 2.724(7) 2.670(9) 2.700(7) 2.637(8) 2.693(3) 2.683(7) 2.631(3) 2.583(4)
  2.754 2.667 2.720 2.665 2.698 2.643 2.690 –b –b  
M–Cpcent 2.5590(17) 2.514(3) 2.534(4) 2.495(4) 2.511(3) 2.489(4) 2.4941(13) 2.500(4) 2.4744(17) 2.3967(15)
  2.5723(15) 2.532(2) 2.547(3) 2.518(4) 2.529(3) 2.510(4) 2.5081(12) 2.511(3) 2.4895(15) 2.4126(15)
Cpcent–M–Cpcent 135.45(5) 135.20(8) 135.92(11) 134.82(14) 135.95(11) 135.47(13) 135.30(4) 135.18(11) 135.64(6) 135.89(6)
Open in a new tab
a

2Ce has been reported previously;51 however, we have resynthesized it here and used values from our single-crystal X-ray diffraction study. Mean absolute deviation between the experiment and calculation = 0.032 Å for M–I and 0.006 Å for M–C.

b

Geometry optimization was not performed on the 2Am component of 4Am due to the unavailability of dispersion corrections for Am,86 and the 2Nd component of 4Nd was not optimized as 2Nd has been geometry-optimized separately starting from coordinates from crystals of the pure compound.

Table 4. Comparison of Experimental and Computed M–I and M–CNHC Bond Lengths (Å) between 2M (M = La, Ce, Pr, U, Np, and Pu) and 4M (M = Nd, Am) Complexes.

  M–I
 M–CNHC
  Expt. Comp. Expt. Comp.
2La 3.1769(3) 3.140 2.751(4) 2.754
2U 3.1266(4) 3.095 2.687(5) 2.667
Δ(La–U) 0.0503(5) 0.045 0.064(6) 0.087
2Ce 3.1584(6) 3.118 2.724(7) 2.720
2Np 3.1003(7) 3.081 2.670(9) 2.665
Δ(Ce–Np) 0.0581(9) 0.037 0.054(11) 0.055
2Pr 3.1393(6) 3.100 2.700(7) 2.698
2Pu 3.0938(7) 3.074 2.637(8) 2.643
Δ(Pr–Pu) 0.0455(9) 0.026 0.063(11) 0.055
Δ(Ce–Pu) 0.0646(9) 0.044 0.087(11) 0.077
2Nd 3.1209(2) 3.085 2.693(3) 2.690
4Am 3.0666(11) –b 2.631(3) –b
Δ(Nd–Am)a 0.0543(11) –b 0.062(4) –b
4Nd 3.103(2) –b 2.683(7) –b
Δ(Nd–Am)a 0.0364(20) –b 0.052(8) –b
Open in a new tab
a

Entries compare 4Am to 2Nd, or 4Nd, respectively.

b

Geometry optimization was not performed on the 2Am component of 4Am due to the unavailability of dispersion corrections for Am,86 and the 2Nd component of 4Nd was not optimized as 2Nd has been geometry-optimized separately starting from coordinates from crystals of the pure compound.

Complex 5Nd crystallized in the triclinic space group P1̅ with Z′ = 2, unlike both 2M (M = La, Ce, Pr, Nd, Y, U, Np, and Pu) and 4M (M = Nd, Am), which suggests that beyond a certain chloride occupancy the bulk lattice changes. However, all pertinent bond lengths and angles for 5Nd (e.g., Nd–CCp range = 2.723(3)–2.85(3) Å; Nd–CNHC = 2.708(3) Å; Cpcent–Nd–Cpcent = 134.3(4)° and 136.91(8)°) are very similar to both 2Nd and 4Nd. Between 2Nd, 4Nd, and 5Nd, all metrical parameters overlap within the 3σ criterion with the exception of the Nd–I lengths (3.1209(2) Å in 2Nd, and 3.103(2) Å in 4Nd). This suggested that at least for Nd, within this molecular framework, there is a minimal perturbation to the rest of the molecular structure due to the influence of iodide (2Nd) vs chloride (5Nd)—though packing forces differ. Importantly, across all three structures with Nd, the Nd–CNHC bond lengths are within 3σ of each other except for one of the Nd–CNHC distances in 5Nd (there are two independent molecules in the asymmetric unit); 2Nd, 2.693(3) Å; 4Nd, 2.683(7) Å; and 5Nd, 2.667(3) Å and 2.708(3) Å. Thus, we suggest that it is possible to draw useful comparisons with other mixed-halide structures such as 4Am—noting that M–X (X = I or Cl) bond lengths are likely to be somewhat less reliable than those from a data set which does not contain mixed occupancy atoms.

The M–CNHC bond lengths within the 2M series provide the most striking comparison. For example, with 2La and 2U, the difference in the M–CNHC bond (Δ = 0.064(6) Å) is similar to the difference between 2Pu (M–CNHC = 2.637(8) Å) and either 2Ce (M–CNHC = 2.724(7) Å; Δ = 0.087(11) Å) or 2Pr (M–CNHC = 2.700(7) Å; Δ = 0.063(11) Å). Crucially, the difference in M–CNHC lengths between the lanthanide and actinide series does not diminish significantly from U to Pu (La to Pr). By comparison, as Tables 2 and 3 show, the magnitude of the difference between lanthanide and actinide M–I bond lengths broadly decreases along both 1M and 2M series, which is the normal trend.18,20,47,65,87−90 In 4Am, we see that the M–CNHC length (2.631(3) Å) is shorter than all of the 2M (M = La, Ce, Pr, Nd, U, Np, and Pu) complexes except that of 2Y. When comparing 4Am to all of 4Nd, 2Nd, and 5Nd (8-coordinate ionic radii: Am3+ = 1.09 Å, Nd3+ = 1.109 Å; Δ = 0.019 Å), we find that none of the Nd-complexes has any overlap of their Nd–CNHC lengths with the Am–CNHC bond in 4Am, even to a 5σ level of significance. For example, the difference in the M–CNHC bond between 4Am and 2Nd is 0.062(3) Å (with 2Nd bearing the longer bond length), and for 4Nd, it is 0.082(8) Å, again where Nd bears the longer distance. The smallest difference between 4Am and any of the Nd3+ complexes we observe is when comparing Nd(1) of 5Nd with 4Am (Δ = 0.036(4) Å)—again this difference is larger than the difference in their ionic radii even when accounting for the standard uncertainty in the measurements. To the best of our knowledge, this is unprecedented in studies of molecular systems with trivalent f-elements that span across La/U, Ce/Np, Pr/Pu, and Nd/Am comparisons. Figure 4 shows a plot of the calculated 8-coordinate ionic radius vs the M–E bond length for M–I in 1M, and both M–I and M–CNHC in 2M.14,91 See Figure S44 in the Supporting Information for the same data plotted against the experimentally derived 6-coordinate ionic radii, which shows the same trend.

In assigning the coordination numbers of complexes herein, we have counted Cp as three sites;92 therefore, the metals are formally 8-coordinate, and so values for 8-coordinate ionic radii are used.14,91 This is to avoid overstating the significance of bond length differences, which can manifest when 6-coordinate radii are used with complexes which are not formally 6-coordinate. When comparing isomorphous Am and Nd complexes with formal coordination numbers (CN) greater than 6–which is every molecular Am complex to have been structurally characterized except [Am{N(O=PPh2)2}3] (6 CN),14 [AmBr3(OPCy3)3] (6 CN),89 and [AmCl6][PPh4]3 (6 CN)64 – one should account for the 0.019 Å difference between 8-coordinate Am3+ and Nd3+. When Am-ligand bond lengths are shorter than Nd-ligand bond lengths by >0.019 Å, they are more likely to be of significance beyond simple well-established ionic bonding trends as opposed to shortening, which lies between 0.008 and 0.019 Å.

It is important to place the metrical data in context with previous Am/Nd comparisons in the literature, which reflects the scarcity of clear, unambiguous cases where Am3+-ligand bond lengths are shorter than Nd3+-ligand bond lengths. By way of example, the M–Se bond lengths in a series of 9-coordinate complexes, [M{N(E=PPh2)2}3] (M = La, Ce, Nd, U, Np, Pu, and Am; E = Se),14,68 show that the An–Se length is shorter than the Ln–Se length for U/La and Pu/Ce pairs by 0.0360(5) and 0.0303(4) Å, but upon reaching Am/Nd, the difference dropped to 0.0176(9) Å which is on the edge of the difference in the 8-coordinate ionic radii of Am3+/Nd3+ (Δ = 0.019 Å), while much larger than the difference in 6-coordinate ionic radii for these metals (Δ = 0.008 Å)—noting of course that these complexes are 9-coordinate.91 In [M(Cptet)3] [M = Am, Nd; Cptet = (C5Me4H)], the Am–Cpcentroid distance is 2.517(8) Å and 2.518(1) Å for the Nd–Cpcentroid distance, meaning that they are statistically identical.81 In [{M(Cp′)3}2(μ-4,4′-bipy)] (M = Am, Nd; Cp′ = {C5H4(SiMe3)}; bipy = bipyridine),18 the average Am and Nd M–Cpcentroid distances to the Cp′ ligand are 2.524(3) and 2.543(2) Å, respectively, (Δ = 0.019(4) Å)—again on the edge of significance when the differences in 8-coordinate ionic radius is considered.92 The difference in the metal–nitrogen distances is slightly more pronounced in the [{M(Cp′)3}2(μ-4,4′-bipy)] complexes, where the Am–N bond (2.618(3) Å) is shorter than the Nd–N bond (2.6482(16) Å) by 0.030(3) Å.18 In a study of f-block dithiocarbamate complexes, [M(S2CNEt2)3(N2C12H8)] (M = Nd, Sm, Eu, Gd, Dy, Am, Cm, and Cf), the average Am–N and Am–S distances were found to overlap within 3σ of each other once ionic radius differences were considered.46 Similarly, inconclusive signs of Am vs Nd bonding differences were found in a pair of dithiophosphinate complexes, [M{S2P(tBu2C12H4)}4]− (M = Am, Nd), which feature ligands directly relevant to S-donor extractant molecules that show selectivity for the minor actinide ions Am3+ and Cm3+, over Ln3+, in biphasic solvent extractions.14,16,68,88 Thus, the substantially shorter Am–CNHC length in 4Am compared with the isomorphous 4Nd congener is unusual and warrants further spectroscopic/computational investigation to provide insight into the origin of the observed metrical differences.

Quantum Chemical Calculations

To help elucidate the nature of the bonding and electronic structure differences in complexes 2M, we turned to computational quantum chemistry in the form of scalar relativistic, hybrid density functional theory at the PBE0 level. Full details of the calculations are given in the Supporting Information. We began by optimizing the geometries of 2M (M = La, Ce, Pr, Nd, U, Np, and Pu); M–I and M–CNHC distances are collected in Tables 3 and 4. There is very good agreement in these metrics between experiment and calculation, especially for M–CNHC lengths, for which the mean absolute deviation is only 0.006 Å. As with the single-crystal X-ray diffraction data, the difference between the M–I distances in corresponding pairs of lanthanides and actinides decreases systematically across the series, while the analogous difference in M–CNHC distances does not, with that of the Pr/Pu pair being the same as that for Ce/Np, Δ = 0.055 Å. Note that due to the unavailability of dispersion corrections for Am,86 the structure of 2Am was not optimized and was taken directly from the single-crystal X-ray diffraction geometry (that is, the iodo-component of 4Am).

We initially anticipated that natural bond orbital (NBO) analysis would allow us to address the nature of the M–CNHC interaction within a localized orbital framework, but NBO did not locate any M–CNHC bonding orbitals. Therefore, we turned to an analysis of the Kohn–Sham orbitals, recognizing that these are typically rather delocalized in large, low-symmetry f-element organometallic compounds. Valence molecular orbital (MO) energy level diagrams are presented for 2An (An = U, Np, Pu, Am) in Figure 5, with analogous diagrams for 2Ln given in the Supporting Information (Figure S96). The energies of the An–Cp* and An–I orbitals change little across the series, in contrast to the metal 5f manifold, which shows the expected significant decrease in energy from U to Am. The NHC-based MOs–distinguished as σ and π according to their principal character about the M···C axis–are the most stable of those shown, except for 2Am, for which two metal 5f-based orbitals lie in between the NHC-based levels (also the case for 2Nd, see Figure S96).

Figure 5.

Figure 5

Open in a new tab

Molecular orbital energy level diagrams for 2An (An = U, Np, Pu, and Am), obtained from the optimized geometries of 2U, 2Np, and 2Pu, and from a single point calculation at the single-crystal X-ray diffraction geometry of the 2Am component of 4Am (due to unavailability of dispersion corrections for Am).86 Principal orbital character is indicated, though sometimes the indicated character is present in other, energetically close, orbitals. HOMO = highest occupied molecular orbital.

Before exploring the composition of the NHC-based σ and π MOs, it is instructive to examine the total energy surface for scanning the M–CNHC distance. Figure 6 shows the change in the total (SCF) energy of 2La and 2U upon performing relaxed scans of this distance. Clearly, these energy surfaces are very flat; compressing or elongating the bonds by 0.05 Å, an amount similar to the Ln vs An bond length differences seen in the 2M series, costs only ca. 0.5 kJ·mol–1, suggesting that quantum chemical differences in M–CNHC bonding between Ln/An pairs are likely small. The energy penalties for the M–CNHC bond length changes are within the range of crystal packing forces, but the consistent shortening of An–CNHC relative to Ln–CNHC within this series, to a degree which is not commensurate with the difference in ionic radius, suggests that the crystal packing forces are not the main cause. It is notable that the energy well for 2U is slightly steeper than that for 2La, suggesting that the U–CNHC bond is the stronger. To further estimate this, we split both 2U and 2La into two fragments, the NHC ligand and the Cp2*MI component at their geometries in 2U and 2La, and calculated the energies of the separated fragments vs those of the full molecules. This yields fragmentation energies of 192.6 and 174.4 kJ·mol–1 for 2U and 2La, respectively, at the SCF level.

Figure 6.

Figure 6

Open in a new tab

Energy relative to that at the fully optimized geometry for changing the M–CNHC distance in 2La and 2U, while allowing the rest of the atomic positions to relax.

Table 5 presents Mulliken population analysis data for the NHC-based σ and π MOs, together with the M–C delocalization indices δ, calculated from the quantum theory of atoms in molecules (QTAIM). δ is often taken as a QTAIM measure of bond order. None of the δ values are large, indicating that the M–CNHC interactions are not strong, in agreement with Figure 6. The δ values display a striking similarity among the lanthanides, and within the actinide series. The latter have slightly higher δ than the former, suggestive of a consistently larger M–CNHC interaction in the 5f series, in agreement with the experimental and computational structural data.

Table 5. Kohn–Sham Molecular Orbital Compositions (Mulliken Analysis, 1% Threshold) and QTAIM Delocalization Indices (δ) for 2M (M = La, Ce, Pr, Nd, U, Np, Pu, and Am)a.

  2La 2U 2Ce 2Np 2Pr 2Pu 2Nd 2Am
metal content of M–NHC σ (%) 2.0s, 1.8p, 6.8d 1.9s, 1.0p, 7.7d, 1.4f (s + p + d = 10.6) 1.9s, 1.9p, 7.2d 1.8s, 1.1p, 7.4d, 2.4f (s + p + d = 10.3) 1.9s, 1.7p, 7.0d, 1.4f (s + p + d = 10.7) 1.8s, 1.2p, 7.4d, 3.3f (s + p + d = 10.6) b 1.1s, 5.4d, 6.7f (s + p + d = 6.5)
total metal content (%) 10.6 12.0 11.0 12.7 12.1 13.8   13.2
metal content of M–NHC π (%)   1.3d   1.2d b 1.2d, 3.4f   1.2d, 1.4f
M–CNHC δ 0.30 0.38 0.31 0.38 0.31 0.38 0.31 0.37
Open in a new tab
a

Data obtained from the optimized geometries of all 2M bar 2Am, for which a single point calculation at the single-crystal X-ray diffraction geometry was performed (due to non-availability of dispersion corrections for Am).

b

Extensive mixing of ligand orbitals with metal 4f levels (see Supporting Information Figure S96) precludes clear population analysis.

Turning to the population analysis of the NHC-based σ and π MOs, we begin with the former. It is noteworthy that, within the first three Ln/An pairs, the sum of the metal s, p, and d contributions to these orbitals is very similar, all lying within the range 10.3–11.0%. For the actinide member of each pair, there is also a 5f contribution, which rises from 1.4% for 2U to 3.3% for 2Pu (noting also the 1.4% 4f contribution for 2Pr). 2Am differs from the other actinides in having reduced s + p + d and larger 5f (6.7%) composition, though it has a total metal contribution very similar to 2Np and 2Pu. It is tempting to ascribe the shorter An–CNHC distances vs their Ln analogues to the extra f content of the σ MO, but while recognizing that we cannot rule this out as a contributory factor, we sound the following cautionary note: Bursten’s FEUDAL (f’s essentially unaffected, d’s accommodate ligands) model of the bonding in early actinide complexes tells us that, in general, it is the metals’ d-orbitals that are primarily responsible for metal–ligand binding, not the 5f. Furthermore, it is well known that periodic increases in metal f contributions to MOs featuring both metal and ligand content typically arise from atomic orbital energy matching, rather than reflecting overlap-driven covalency. We note that the M–CNHC distances in 2Pu and 2Am are almost identical, despite the Am 5f contribution to the σ MO being twice that in 2Pu.

The structural data suggest that we are looking for a consistent difference in the An–CNHC versus Ln–CNHC bonding, a conclusion supported by the δ values in Table 5. Figure 6 tells us that such a difference is likely small, and the FEUDAL approach directs us to the 6d orbitals.93−97 It is therefore noticeable that all of the actinide complexes have a very small but consistent metal d contribution to the NHC-based π MOs (Table 5), which is absent in all of the lanthanide systems. Complexes 2Pu and 2Am also have small 5f contributions to this MO, but note the arguments above about the nature of 5f-based covalency in this part of the actinide series. Figure 7 presents images of the NHC-based π MOs in 2La and 2U; the metal contribution to the latter is clearly visible.

Figure 7.

Figure 7

Open in a new tab

NHC-based π molecular orbital in 2La (left) and 2U (right), isovalue = 0.02. Hydrogen atoms omitted for clarity.

If there is An–CNHC bonding with a π-bonding component as described above, we would expect that rotating the NHC ligand about the M–CNHC axis would be energetically more costly in the An vs Ln systems. We therefore attempted relaxed total energy surface scans for this distortion in 2La and 2U, but these calculations were not well-behaved, and indeed, the NMR spectroscopic data (vide infra) suggest that rotation about the M–NHC bond is a high-energy process, requiring significant structural rearrangement. Instead, Figure 8 presents the energies relative to that of the fully optimized geometry for rotating the NHC ligand while keeping all the other atomic positions fixed. Such an approach yields relative energies larger than in relaxed energy surface scans but typically provides an upper bound to the energetics. Clearly, from the 1H NMR data of all 2M complexes, there is a penalty for rotating the NHC ligand about the metal as evidenced by the inequivalence of the Me-resonances at room temperature. This is likely to be predominantly steric in origin as it is seen even for 2La. However, computationally, the energy penalty for rotating the NHC ligand in 2U is significantly larger than in 2La. This may, in part, reflect the ca. 0.1 Å shorter U–CNHC distance vs the La equivalent but may also result in part from the differing M–NHC orbital character shown in Figure 7.

Figure 8.

Figure 8

Open in a new tab

Energy relative to that at the fully optimized geometry for rotating the NHC ligand about the M–C axis in 2La and 2U, while keeping the rest of the atomic positions fixed.

QTAIM bond critical point (BCP) ellipticities ε can give information on single vs multiple bonding; values close to zero indicate cylindrical symmetry about the BCP (single or triple bonds), while significant deviations from zero suggest (partial) double bonding. In our experience, f-element–ligand BCP electron densities are typically very low, which can lead to highly variable curvatures (and hence curvature ratios, which define ε). The BCP electron densities ρ in our target systems are indeed small; those in 2Ln range from 0.042 electron·bohr–3 (for 2La) to 0.045 for 2Pr and 2Nd, with those for 2An being slightly larger, at 0.051 for 2U, rising to 0.052 for 2Pu and 2Am. Such low values do indeed lead to some scatter in the ε values, but, in general, these are larger for the actinide–NHC bonds than for the lanthanide analogues, the clearest separation being 0.261/0.023 for 2Np vs 2Ce and 0.283/0.091 for 2Pu vs 2Pr. Lastly, we note that during the revision stage of this manuscript, a report was published that documented a plutonium complex with a Pu–CNHC interaction, complementing earlier analogous neptunium chemistry.25,98 That research was focused on the characterization of Pu=C multiply bonded interactions through the coordination of diphosphonioalkylidene ({C(PPh2=NSiMe3)2}, BIPM) ligands to the actinide metal ion. One of those complexes also contained coordinated IMe4 NHC ligands, for which shorter Np–CNHC vs Ce–CNHC and Pu–CNHC vs Pr–CNHC bonds were observed with differences on the order of ∼0.045–0.060 Å. Consistent with this study, the bonding was found to be largely electrostatic in nature, but in those works, no differences in molecular orbital compositions related to the metal–NHC bonding were found to correlate with the bond metrics, nor were La/U and Nd/Am comparisons possible.

NMR Spectroscopy

To further support the characterization of the complexes herein, NMR spectra were collected for 1M (M = La, Ce, Pr, Nd) in d6-benzene with the addition of a weighed amount of H8-THF to prevent precipitation of the insoluble material. We were able to observe the C5Me5 singlet for all four molecules by 1H NMR (see Supporting Information for full details). Room temperature solution magnetic susceptibilities were determined for 1Ce, 1Pr, and 1Nd by the Evans method, and they agreed well with the free-ion values: 1Ce (2F5/2, measured: 2.47 μB vs 2.54 μB expected), 1Pr (3H4, measured: 3.48 μB vs 3.58 μB expected), and 1Nd (4I9/2, measured: 3.58 μB vs 3.62 μB expected).

Well-resolved 1H and 13C{1H} NMR spectra were also collected all 2M complexes (except previously reported 2U), as well as 4Am, in neat d6-benzene (see Supporting Information for 13C{1H} NMR spectroscopy and all data). As previously observed with 2Ce and 2U,51 all 2M complexes showed that rotation about the M–CNHC bond is restricted. A variable temperature NMR (VT-NMR) of 2La showed that even at 100 °C in d8-toluene, the La–CNHC bond is restricted (Figures S57–S59). As La3+ is the largest ion studied herein and 2La possesses the longest M–CNHC bond length, the calculated barrier to rotation in this complex represents a lower bound for this series of complexes. Nevertheless, it is highly disfavored as ΔS⧧ = −11.6 J·K–1 mol–1 (−61.8 J·K–1 mol–1 to 38.5 J·K–1 mol–1) and ΔH⧧ = 92.3 kJ·mol–1 (74.6 kJ·mol–1 to 110.0 kJ·mol–1), where the values in brackets are the 95% confidence intervals. Presumably, the large barrier arises through a combination of predominantly steric effects, though, in the case of the actinide complexes, there may also be a contribution from the M–CNHC π-bonding component which is supported by calculations (Figure 8). Table 6 shows the 1H NMR chemical shifts for the Cp* and IMe4 ligands of all 2M complexes.

Table 6. 1H NMR Chemical Shifts (ppm vs d6-Benzene Residual) for 2M (M = La, Ce, Pr, Nd, Y, U, Np, and Pu). Spectra Were Recorded at Ambient Temperature (295–298 K).

  2La 2Ua(51) 2Ce 2Np 2Pr 2Pu 2Nd 2Y
C5Me5 2.17 0.68 6.10 0.77 13.61 1.65 11.82 2.09
IMe4 C(CH3) 1.21 –53.99 –23.68 0.18 –70.81 1.25 –33.77 1.21
  1.36 –46.82 –19.34 0.54 –47.13 1.25 –25.74 1.35
IMe4 N(CH3) 2.99 –11.19 –3.52 2.64 –13.10 3.94 –5.43 2.91
  3.54 –10.49 –3.23 4.27 –10.59 4.98 –4.70 3.66
Open in a new tab
a

2U was reported previously in d8-THF.

The 1H NMR spectrum of 4Am in d6-benzene shows features similar to all the 2M complexes but with an apparent doubling of every signal (see Figures S64–S67). A variable temperature NMR spectroscopic (VT-NMR) study in d6-benzene (Figure S66), across a small temperature range due to radiological safety considerations, revealed that up to 50 °C, the peaks did not coalesce, nor did their relative ratios change. We attribute the doubling to the presence of a mixed halide species, [Am(Cp*)2(InCl1–n)(IMe4)], which informed and is in excellent agreement with our crystallographic study (where n is ca. 0.65 determined by competitive refinement of the halide sites).

UV–Vis–NIR Spectroscopy

The UV–vis–NIR spectra of 1M (M = La, Ce, Pr, and Nd) complexes were collected in THF at ambient temperature and can be compared with the spectra of their 2M counterparts collected in toluene. Figure 9 shows the spectra of 1Pr, 2Pr, 1Ce, and 2Ce as examples. The influence of THF vs IMe4 bonding on the f → f and f → d transitions in 1M and 2M is instructive toward the potential origin of the structural difference between the 4f and 5f series.

Figure 9.

Figure 9

Open in a new tab

Solution UV–vis–NIR spectra of [M(Cp*)2(I)(THF)] (1M, M = Ce, dotted black; Pr, solid black) (3 mM, THF) and [M(Cp*)2(I)(IMe4)] (2M, M = Ce, dotted red; Pr, solid red) (3 mM, toluene) shown between 7000 and 29,000 cm–1 (1429–345 nm) at ambient temperature.

The absorption spectra of both 1Ce and 2Ce are simple and characteristic of Ce3+ (4f1, 2F5/2) complexes. A broad, somewhat featureless transition tails in from the UV region down to ca. 24,000 cm–1 (417 nm), and a single additional broad peak is observed, which is the 5d ← 4f dipole allowed interconfigurational transition. In 1Ce, this is seen at 20,675 cm–1 (484 nm, ε = 259 M–1 cm–1), and in 2Ce, it lies at 20,308 cm–1 (492 nm, ε = 406 M–1 cm–1), a red shift of 367 cm–1. See the Supporting Information for a computational analysis of this transition, which supports this assignment. For Pr3+ (4f2, 3H4) ions, the 4f → 5d absorption energy is usually sufficiently large that it does not interfere with the vis–NIR absorption spectrum,99−104 and it is the first member of the series for which f → f (Laporté forbidden, intraconfigurational) transitions are observed and thus can directly report on the impact of THF vs IMe4 donor properties on the 4f manifold. The 3P0 ← 3H4 and 3P1 ← 3H4 transitions typically occur around 20,700 cm–1 (483 nm) and 21,400 cm–1 (467 nm), respectively.99−101,103−105 In 1Pr these appear at 20,161 cm–1 (496 nm, ε = 27 M–1 cm–1 for 3P0) and 20,782 cm–1 (481 nm, ε = 38 M–1 cm–1 for 3P1). In 2Pr, these transitions occur at 20,097 cm–1 (498 nm, ε = 19 M–1 cm–1) and 20,747 (482 nm, ε = 16 M–1 cm–1), and so, like with 1Ce and 2Ce above, both exhibit a modest redshift in 2Pr vs 1Pr, though less than the cerium complexes. When comparing 1Nd and 2Nd, we see a much smaller redshift between most of the features than what is seen for the Ce3+ and Pr3+ complexes, though the spectra are much more complex (Figure S82), which precludes assigning a redshift value between any set of peaks. However, when comparing 2Nd [Nd(Cp*)2(I)(IMe4)] to 5Nd [Nd(Cp*)2(Cl)(IMe4)], we see essentially no difference (Figure S93) reflecting that replacing chloride with iodide has little impact upon the observed electronic transitions.

The UV–vis–NIR spectra of 2Np and 2Pu are remarkably similar to the previously reported spectra for 1Np and 1Pu (Figure 10).68 All four feature broad absorptions which tail in from the UV region down to ca. 24,000 cm–1 (417 nm), which lead into a series of poorly resolved features presumably derived from the 6d ← 5f transitions with fine structure arising from splitting of the 5f manifold or vibronic coupling of the excited 6d state to ligand modes. For 1Np, we previously noted a main band which extends from ca. 14,000–22,000 cm–1 (ca. 714–455 nm, εmax = 849 M–1 cm–1), which appears somewhat red-shifted in 2Np and resides from ca. 13,000–20,000 cm–1 (769–500 nm, εmax = 498 M–1 cm–1; Figure 10 top). Complex 1Pu shows a similar broad peak from ca. 18,000–25,000 cm–1 (556–400 nm, εmax = 704 M–1 cm–1), which appears much less red-shifted in 2Pu (Figure 10 bottom) than between the Np3+ examples, such that the feature approximately overlaps the same range as in 1Pu (i.e., 18,000–25,000 cm–1; 556–400 nm; εmax = 831 M–1 cm–1). Fitting the peak groupings with a Gaussian curve (see Figures S94 and S95) shows the center of the main 5f → 6d grouping in 1Np resides at 18,309(6) cm–1, while for 2Np, it is at 16,467(9) cm–1 which is roughly 1800 cm–1 lower in energy. For 1Pu (22,947(7) cm–1) and 2Pu (22,211(13) cm–1), the shift is much smaller at ca. 730 cm–1. This trend of increasing energy in the 6d ← 5f transitions from Np to Pu is consistent with previous works.68−70

Figure 10.

Figure 10

Open in a new tab

Top: solution UV–vis–NIR spectrum of [Np(Cp*)2(I)(THF)] (1Np, black line) and [Np(Cp*)2(I)(IMe4)] (2Np, red line) in toluene. Bottom: solution UV–vis–NIR spectra of [Pu(Cp*)2(I)(THF)] (1Pu, black line) and [Pu(Cp*)2(I)(IMe4)] (2Pu, red line). All spectra were collected in toluene at ambient temperature and are shown between 7000–33,333 cm–1 (1429–333 nm).

At the lower energy region of these spectra, characteristically weak and somewhat sharp Np3+ and Pu3+ f → f transitions can be seen106−110 and are remarkably similar within each of the two pairs. As with 2Pr and 2Nd, there is a small red shift for the IMe4 adduct vs the THF-adduct of ca. 70 to 200 cm–1 for 2Np, depending on the pairs of peaks chosen, and at most ca. 150 cm–1 for 2Pu.

Finally, the UV–vis–NIR spectrum of 4Am (Figure 11) is broadly typical of Am3+ in solution, whereby we can identify features corresponding to the 7F6 ← 7F0 and 5L6 ← 7F0 transitions; however, both appear to be “doubled”. In 4Am, the higher-energy 5L6 ← 7F0 transition appears as two sharp peaks, 18,832 cm–1 (531 nm, ε = 544 M–1 cm–1) and 19,004 cm–1 (526 nm, ε = 507 M–1 cm–1), while the 7F6 ← 7F0 transition appears as two sharp peaks at 11,751 cm–1 (851 nm, ε = 340 M–1 cm–1) and 12,071 cm–1 (828 nm, ε = 348 M–1 cm–1). A similar, but genuine, splitting of the high energy 5L6 ← 7F0 feature was seen in [Am(Cptet)3],81 and also in [{Am(Cp′)3}2(μ-4,4′-bipy)].18 With 4Am, given the NMR and structural data which strongly suggest that the bulk is a mixture of [Am(Cp*)2(I)(IMe4)] and the chloride congener, it is likely that the doubling is in fact due to differences in the f → f transitions of around 172 and 320 cm–1 for the major transitions in the iodide- and chloride-ligated complexes. This contrasts the little to no effect with Nd (vide supra).

Figure 11.

Figure 11

Open in a new tab

Solution UV–vis–NIR spectra of [Am(Cp*)2(IxCl1–x)(IMe4)] (4Am, ca. 1 mM, toluene) shown between 7000–33,000 cm–1 (1429–303 nm) at ambient temperature. Molar absorptivity values were based on using the MW of the mixed I/Cl species, but individual bands are not assigned to the specific I vs Cl species.

The influence of THF vs IMe4 coordination on the f → f transitions in Np3+ and Pu3+ appears to be larger than seen with Pr3+ and Nd3+, which might be expected based upon the better spatial overlap of the 5f orbitals with ligands vs that of the 4f orbitals in the lanthanide counterparts. However, it is smaller than the changes seen in the 5f → 6d region of all of the spectra, which suggests that the key differences in the way THF and IMe4 bind to these ions involve the 5d (lanthanide) and 6d (actinide) orbitals. These data support conclusions derived from the calculations.

Conclusions

Analysis across a series of NHC-ligated bent-metallocene complexes, [M(Cp*)2(X)(IMe4)] (X = I, M = La, Ce, Pr, Nd, U, Np, and Pu; X = Cl, M = Nd; X = I/Cl, M = Nd, and Am), reveals significant shortening of the metal–CNHC ligand bond length with actinide metals vs lanthanide examples with closely matched ionic radii. This homologous series extends from La to Nd and from U to Am, including rare or unique examples of M–C σ-bonding to NHC ligands in the case of Pu and Am. Most remarkably, we observe no significant decrease in the extent of An vs Ln metal–CNHC length shortening as the series are traversed. Structural and quantum chemical analyses between the NHC complexes and their THF-ligated precursors reveal that hard/soft arguments of the primarily electrostatic origin explain the anticipated lanthanide vs actinide differences in M–I and M–O bond lengths. However, the An–NHC π-type Kohn–Sham molecular orbitals consistently feature small (ca. 1–2%) 6d contributions, which are absent for all the lanthanide congeners. Quantum theory of atoms in molecules data suggest a consistently larger M–CNHC interaction in the 5f series, and the computed barrier to rotation of the NHC ligand around the U–CNHC axis is larger than that for the La analogue. Together, these results suggest larger M–CNHC covalency in the actinide series and that ligand-π to vacant 6d interactions can differentiate lanthanide and actinide ions with these neutral donor ligands. This is a key distinction between this molecular framework and others and could more generally inform the design of future ligand systems to differentiate otherwise similar f-block ions. Typically, lanthanide and actinide differences (within homologous donor series) are driven by the modest differences between the 4f and 5f manifold, but for the NHC complexes here, it is 6d orbital participation in the form of a minor π-bonding contribution in these simple Lewis-base adducts which is present in the actinide complexes but absent in the lanthanide congeners which correlates with substantial structural and spectroscopic differences. This bonding mechanism was proposed over 30 years ago with simple π-basic ligands on uranium, and this work presents evidence that extends deeper into the transuranium series.94 In principle, strong σ-donor ligand systems capable of π-donation, such as amides or alkoxides, may show similar effects to those found herein. However, this may be restricted to metals in lower oxidation states (i.e., An3+), as previous works on Np4+ alkoxides show π-donation into the 5f manifold, rather than 6d.62 As organometallic chemistry of the transuranium elements is experiencing a renaissance as far across the actinide series as californium (the highest atomic number element which can be used for synthetic molecular chemistry on a mg scale), the results here suggest future studies into the role of d orbital-derived π-bonding beyond uranium, and indeed beyond americium, will be insightful.71,111

Acknowledgments

Experimental work (except for VT-NMR measurements on 2La) was conducted at the Los Alamos National Laboratory (LANL) for which C.A.P.G., A.J.G., and B.L.S. thank the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, Chemical Sciences, Geosciences, and Biosciences Division, Heavy Element Chemistry Program at LANL (contract DE-AC52-06NA25396). C.A.P.G. also thanks the LANL Laboratory Directed Research and Development (LDRD) program for a Distinguished J. R. Oppenheimer Postdoctoral Fellowship (LANL-LDRD 20180703PRD1) and the Royal Society for a University Research Fellowship (URF\211271) during manuscript preparation, data analysis, and supporting synthetic work. For computational work, N.K. is grateful to the University of Manchester for computing resources via its Computational Shared Facility and associated support services. S.T.L. thanks the Alexander von Humboldt Foundation for a Freidrich Wilhelm Bessel Research Award and the EPSRC (EP/M027015/1). We acknowledge funding from the EPSRC (EP/V007580/1 and EP/K039547/1) for NMR spectroscopy resources. Research data files supporting a portion of this publication are available from Figshare at doi: 10.48420/25037141.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacs.3c12721.

  • Experimental details, photographs taken during syntheses, crystallography details, molecular structure details and metrics, NMR and UV–vis–NIR spectra, and computational details (PDF)

The authors declare no competing financial interest.

Supplementary Material

ja3c12721_si_001.pdf (5.8MB, pdf)

References

  1. Clark D. L.; Geeson D. A.; Hanrahan R. J. Jr.. Plutonium Handbook; American Nuclear Society, 2019. [Google Scholar]
  2. Cotton S.Lanthanide and Actinide Chemistry; John Wiley & Sons, Ltd, 2006. [Google Scholar]
  3. Liddle S. T.; Mills D. P.; Natrajan L. S.. The Lanthanides and Actinides: Synthesis, Reactivity, Properties and Applications. Lanthanides and Actinides; Chemistry LibreTexts, 2022. [Google Scholar]
  4. Morss L. R.; Edelstein N. M.; Fuger J.. The Chemistry of the Actinide and Transactinide Elements; Springer Dordrecht, 2011. [Google Scholar]
  5. Birnbaum E. R.; Fassbender M. E.; Ferrier M. G.; John K. D.; Mastren T.. Actinides in Medicine. In Encyclopedia of Inorganic and Bioinorganic Chemistry; Scott R. A., Ed.; Wiley Online Library, 2020; pp 1–21. [Google Scholar]
  6. Liddle S. T. The Renaissance of Non-Aqueous Uranium Chemistry. Angew. Chem., Int. Ed. 2015, 54 (30), 8604–8641. 10.1002/anie.201412168. [DOI] [PubMed] [Google Scholar]
  7. Hayton T. W. Metal-ligand multiple bonding in uranium: structure and reactivity. Dalton Trans. 2010, 39 (5), 1145–1158. 10.1039/B909238B. [DOI] [PubMed] [Google Scholar]
  8. Kelley M. P.; Su J.; Urban M.; Luckey M.; Batista E. R.; Yang P.; Shafer J. C. On the Origin of Covalent Bonding in Heavy Actinides. J. Am. Chem. Soc. 2017, 139 (29), 9901–9908. 10.1021/jacs.7b03251. [DOI] [PubMed] [Google Scholar]
  9. Kaltsoyannis N. Transuranic Computational Chemistry. Chem.—Eur. J. 2018, 24 (12), 2815–2825. 10.1002/chem.201704445. [DOI] [PubMed] [Google Scholar]
  10. Kaltsoyannis N. Does covalency increase or decrease across the actinide series? Implications for minor actinide partitioning. Inorg. Chem. 2013, 52 (7), 3407–3413. 10.1021/ic3006025. [DOI] [PubMed] [Google Scholar]
  11. Jones M. B.; Gaunt A. J.; Gordon J. C.; Kaltsoyannis N.; Neu M. P.; Scott B. L. Uncovering f-element bonding differences and electronic structure in a series of 1:3 and 1:4 complexes with a diselenophosphinate ligand. Chem. Sci. 2013, 4 (3), 1189–1203. 10.1039/c2sc21806b. [DOI] [Google Scholar]
  12. Kirker I.; Kaltsoyannis N. Does covalency really increase across the 5f series? A comparison of molecular orbital, natural population, spin and electron density analyses of AnCp3 (An = Th-Cm; Cp = η5-C5H5). Dalton Trans. 2011, 40 (1), 124–131. 10.1039/C0DT01018A. [DOI] [PubMed] [Google Scholar]
  13. Tassell M. J.; Kaltsoyannis N. Covalency in AnCp4 (An = Th-Cm): a comparison of molecular orbital, natural population and atoms-in-molecules analyses. Dalton Trans. 2010, 39 (29), 6719–6725. 10.1039/c000704h. [DOI] [PubMed] [Google Scholar]
  14. Goodwin C. A. P.; Schlimgen A. W.; Albrecht-Schönzart T. E.; Batista E. R.; Gaunt A.; Janicke M. T.; Kozimor S. A.; Scott B. L.; Stevens L. M.; White F. D.; Yang P. Structural and spectroscopic comparison of soft-Se vs hard-O donor bonding in trivalent americium/neodymium molecules. Angew. Chem., Int. Ed. 2021, 60 (17), 9459–9466. 10.1002/anie.202017186. [DOI] [PubMed] [Google Scholar]
  15. Ingram K. I.; Tassell M. J.; Gaunt A. J.; Kaltsoyannis N. Covalency in the f Element-Chalcogen Bond. Computational Studies of M[N(EPR2)2]3 (M = La, Ce, Pr, Pm, Eu, U, Np, Pu, Am, Cm; E = O, S, Se, Te; R = H, iPr, Ph). Inorg. Chem. 2008, 47 (17), 7824–7833. 10.1021/ic800835k. [DOI] [PubMed] [Google Scholar]
  16. Gaunt A. J.; Reilly S. D.; Enriquez A. E.; Scott B. L.; Ibers J. A.; Sekar P.; Ingram K. I.; Kaltsoyannis N.; Neu M. P. Experimental and theoretical comparison of actinide and lanthanide bonding in M[N(EPR2)2]3 complexes (M = U, Pu, La, Ce; E = S, Se, Te; R = Ph, iPr, H). Inorg. Chem. 2008, 47 (1), 29–41. 10.1021/ic701618a. [DOI] [PubMed] [Google Scholar]
  17. Gaunt A. J.; Scott B. L.; Neu M. P. A Molecular Actinide-Tellurium Bond and Comparison of Bonding in [MIII{N(TePiPr2)2}3] (M = U, La). Angew. Chem., Int. Ed. 2006, 45 (10), 1638–1641. 10.1002/anie.200503372. [DOI] [PubMed] [Google Scholar]
  18. Long B. N.; Beltrán-Leiva M. J.; Celis-Barros C.; Sperling J. M.; Poe T. N.; Baumbach R. E.; Windorff C. J.; Albrecht-Schönzart T. E. Cyclopentadienyl coordination induces unexpected ionic Am-N bonding in an americium bipyridyl complex. Nat. Commun. 2022, 13 (1), 201. 10.1038/s41467-021-27821-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Sperling J. M.; Warzecha E. J.; Celis-Barros C.; Sergentu D. C.; Wang X.; Klamm B. E.; Windorff C. J.; Gaiser A. N.; White F. D.; Beery D. A.; Chemey A. T.; Whitefoot M. A.; Long B. N.; Hanson K.; Kogerler P.; Speldrich M.; Zurek E.; Autschbach J.; Albrecht-Schönzart T. E. Compression of curium pyrrolidine-dithiocarbamate enhances covalency. Nature 2020, 583 (7816), 396–399. 10.1038/s41586-020-2479-2. [DOI] [PubMed] [Google Scholar]
  20. Greer R. D. M.; Celis-Barros C.; Sperling J. M.; Gaiser A. N.; Windorff C. J.; Albrecht-Schönzart T. E. Structure and Characterization of an Americium Bis(O, O’-diethyl)dithiophosphate Complex. Inorg. Chem. 2020, 59 (22), 16291–16300. 10.1021/acs.inorgchem.0c02085. [DOI] [PubMed] [Google Scholar]
  21. Denning R. G.; Snellgrove T. R.; Woodwark D. R. The electronic structure of the uranyl ion. Mol. Phys. 1979, 37 (4), 1109–1143. 10.1080/00268977900100841. [DOI] [Google Scholar]
  22. Denning R. G.; Green J. C.; Hutchings T. E.; Dallera C.; Tagliaferri A.; Giarda K.; Brookes N. B.; Braicovich L. Covalency in the uranyl ion: A polarized x-ray spectroscopic study. J. Chem. Phys. 2002, 117 (17), 8008–8020. 10.1063/1.1510445. [DOI] [Google Scholar]
  23. Hayton T. W.; Boncella J. M.; Scott B. L.; Palmer P. D.; Batista E. R.; Hay P. J. Synthesis of imido analogs of the uranyl ion. Science 2005, 310 (5756), 1941–1943. 10.1126/science.1120069. [DOI] [PubMed] [Google Scholar]
  24. Anderson N. H.; Odoh S. O.; Yao Y.; Williams U. J.; Schaefer B. A.; Kiernicki J. J.; Lewis A. J.; Goshert M. D.; Fanwick P. E.; Schelter E. J.; Walensky J. R.; Gagliardi L.; Bart S. C. Harnessing redox activity for the formation of uranium tris(imido) compounds. Nat. Chem. 2014, 6 (10), 919–926. 10.1038/nchem.2009. [DOI] [PubMed] [Google Scholar]
  25. Goodwin C. A. P.; Wooles A. J.; Murillo J.; Lu E.; Boronski J. T.; Scott B. L.; Gaunt A. J.; Liddle S. T. Carbene Complexes of Neptunium. J. Am. Chem. Soc. 2022, 144 (22), 9764–9774. 10.1021/jacs.2c02152. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Dutkiewicz M. S.; Goodwin C. A. P.; Perfetti M.; Gaunt A. J.; Griveau J.-C.; Colineau E.; Kovács A.; Wooles A. J.; Caciuffo R.; Walter O.; Liddle S. T. A Terminal Neptunium(V)-Mono(Oxo) Complex. Nat. Chem. 2022, 14, 342–349. 10.1038/s41557-021-00858-0. [DOI] [PubMed] [Google Scholar]
  27. Wooles A. J.; Mills D. P.; Tuna F.; McInnes E. J. L.; Law G. T. W.; Fuller A. J.; Kremer F.; Ridgway M.; Lewis W.; Gagliardi L.; Vlaisavljevich B.; Liddle S. T. Uranium(III)-carbon multiple bonding supported by arene δ-bonding in mixed-valence hexauranium nanometre-scale rings. Nat. Commun. 2018, 9 (1), 2097. 10.1038/s41467-018-04560-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kent G. T.; Yu X.; Wu G.; Autschbach J.; Hayton T. W. Synthesis and electronic structure analysis of the actinide allenylidenes, [{(NR2)3}An(CCCPh2)]−- (An = U, Th; R = SiMe3). Chem. Sci. 2021, 12 (43), 14383–14388. 10.1039/D1SC04666G. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Staun S. L.; Sergentu D. C.; Wu G.; Autschbach J.; Hayton T. W. Use of 15N NMR spectroscopy to probe covalency in a thorium nitride. Chem. Sci. 2019, 10 (26), 6431–6436. 10.1039/C9SC01960J. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Smiles D. E.; Wu G.; Hrobárik P.; Hayton T. W. Use of 77Se and 125Te NMR Spectroscopy to Probe Covalency of the Actinide-Chalcogen Bonding in [Th(En){N(SiMe3)2}3]−- (E = Se, Te; n = 1, 2) and Their Oxo-Uranium(VI) Congeners. J. Am. Chem. Soc. 2016, 138 (3), 814–825. 10.1021/jacs.5b07767. [DOI] [PubMed] [Google Scholar]
  31. Mullane K. C.; Hrobarik P.; Cheisson T.; Manor B. C.; Carroll P. J.; Schelter E. J. 13C NMR Shifts as an Indicator of U-C Bond Covalency in Uranium(VI) Acetylide Complexes: An Experimental and Computational Study. Inorg. Chem. 2019, 58 (7), 4152–4163. 10.1021/acs.inorgchem.8b03175. [DOI] [PubMed] [Google Scholar]
  32. Cheisson T.; Kersey K. D.; Mahieu N.; McSkimming A.; Gau M. R.; Carroll P. J.; Schelter E. J. Multiple Bonding in Lanthanides and Actinides: Direct Comparison of Covalency in Thorium(IV)- and Cerium(IV)-Imido Complexes. J. Am. Chem. Soc. 2019, 141 (23), 9185–9190. 10.1021/jacs.9b04061. [DOI] [PubMed] [Google Scholar]
  33. Lu E.; Boronski J. T.; Gregson M.; Wooles A. J.; Liddle S. T. Silyl-Phosphino-Carbene Complexes of Uranium(IV). Angew. Chem., Int. Ed. 2018, 57 (19), 5506–5511. 10.1002/anie.201802080. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Du J.; Seed J. A.; Berryman V. E. J.; Kaltsoyannis N.; Adams R. W.; Lee D.; Liddle S. T. Exceptional uranium(VI)-nitride triple bond covalency from 15N nuclear magnetic resonance spectroscopy and quantum chemical analysis. Nat. Commun. 2021, 12 (1), 5649. 10.1038/s41467-021-25863-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Adam C.; Kaden P.; Beele B. B.; Müllich U.; Trumm S.; Geist A.; Panak P. J.; Denecke M. A. Evidence for covalence in a N-donor complex of americium(III). Dalton Trans. 2013, 42 (39), 14068–14074. 10.1039/c3dt50953b. [DOI] [PubMed] [Google Scholar]
  36. Law J. D.; Peterman D. R.; Todd T. A.; Tillotson R. D. Separation of trivalent actinides from lanthanides in an acetate buffer solution using Cyanex 301. Radiochim. Acta 2006, 94 (5), 261–266. 10.1524/ract.2006.94.5.261. [DOI] [Google Scholar]
  37. Chen J.; Wang S.; Xu C.; Wang X.; Feng X. Separation of Americium from Lanthanides by Purified Cyanex 301 Countercurrent Extraction in Miniature Centrifugal Contactors. Procedia Chem. 2012, 7, 172–177. 10.1016/j.proche.2012.10.029. [DOI] [Google Scholar]
  38. Panak P. J.; Geist A. Complexation and Extraction of Trivalent Actinides and Lanthanides by Triazinylpyridine N-Donor Ligands. Chem. Rev. 2013, 113 (2), 1199–1236. 10.1021/cr3003399. [DOI] [PubMed] [Google Scholar]
  39. Lewis F. W.; Harwood L. M.; Hudson M. J.; Drew M. G. B.; Desreux J. F.; Vidick G.; Bouslimani N.; Modolo G.; Wilden A.; Sypula M.; Vu T.-H.; Simonin J.-P. Highly Efficient Separation of Actinides from Lanthanides by a Phenanthroline-Derived Bis-triazine Ligand. J. Am. Chem. Soc. 2011, 133 (33), 13093–13102. 10.1021/ja203378m. [DOI] [PubMed] [Google Scholar]
  40. Nilsson M.; Nash K. L. Review Article: A Review of the Development and Operational Characteristics of the TALSPEAK Process. Solvent Extr. Ion Exch. 2007, 25 (6), 665–701. 10.1080/07366290701634636. [DOI] [Google Scholar]
  41. Gelis A. V.; Lumetta G. J. Actinide Lanthanide Separation Process–ALSEP. Ind. Eng. Chem. Res. 2014, 53 (4), 1624–1631. 10.1021/ie403569e. [DOI] [Google Scholar]
  42. Gilson S. E.; Burns P. C. The crystal and coordination chemistry of neptunium in all its oxidation states: An expanded structural hierarchy of neptunium compounds. Coord. Chem. Rev. 2021, 445, 213994. 10.1016/j.ccr.2021.213994. [DOI] [Google Scholar]
  43. Gaunt A. J.; Scott B. L.; Neu M. P. Homoleptic uranium(III) imidodiphosphinochalcogenides including the first structurally characterised molecular trivalent actinide-Se bond. Chem. Commun. 2005, 3215–3217. 10.1039/b503106k. [DOI] [PubMed] [Google Scholar]
  44. Ingram K. I. M.; Kaltsoyannis N.; Gaunt A. J.; Neu M. P. Covalency in the f-element-chalcogen bond: Computational studies of [M(N(EPH2)2)3] (M = La, U, Pu; E = O, S, Se, Te). J. Alloys Compd. 2007, 444–445, 369–375. 10.1016/j.jallcom.2007.03.048. [DOI] [Google Scholar]
  45. Macor J. A.; Brown J. L.; Cross J. N.; Daly S. R.; Gaunt A. J.; Girolami G. S.; Janicke M. T.; Kozimor S. A.; Neu M. P.; Olson A. C.; Reilly S. D.; Scott B. L. Coordination chemistry of 2,2’-biphenylenedithiophosphinate and diphenyldithiophosphinate with U, Np, and Pu. Dalton Trans. 2015, 44 (43), 18923–18936. 10.1039/C5DT02976G. [DOI] [PubMed] [Google Scholar]
  46. Cary S. K.; Su J.; Galley S. S.; Albrecht-Schmitt T. E.; Batista E. R.; Ferrier M. G.; Kozimor S. A.; Mocko V.; Scott B. L.; Van Alstine C. E.; White F. D.; Yang P. A series of dithiocarbamates for americium, curium, and californium. Dalton Trans. 2018, 47 (41), 14452–14461. 10.1039/C8DT02658K. [DOI] [PubMed] [Google Scholar]
  47. Long B. N.; Sperling J. M.; Windorff C. J.; Huffman Z. K.; Albrecht-Schonzart T. E. Expanding Transuranium Organoactinide Chemistry: Synthesis and Characterization of (Cp’3M)2(μ-4,4’-bpy) (M = Ce, Np, Pu). Inorg. Chem. 2023, 62 (16), 6368–6374. 10.1021/acs.inorgchem.3c00217. [DOI] [PubMed] [Google Scholar]
  48. Clegg W.; Blake A. J.; Cole J. M.; Evans J. S. O.; Main P.; Parsons S.; Watkin D. J.. Crystal Structure Analysis; Oxford University Press, 2009. [Google Scholar]
  49. Neidig M. L.; Clark D. L.; Martin R. L. Covalency in f-element complexes. Coord. Chem. Rev. 2013, 257 (2), 394–406. 10.1016/j.ccr.2012.04.029. [DOI] [Google Scholar]
  50. Kovacs A.; Apostolidis C.; Walter O. Competing Metal-Ligand Interactions in Tris(cyclopentadienyl)-cyclohexylisonitrile Complexes of Trivalent Actinides and Lanthanides. Molecules 2022, 27 (12), 3811. 10.3390/molecules27123811. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Mehdoui T.; Berthet J. C.; Thuéry P.; Ephritikhine M. The remarkable efficiency of N-heterocyclic carbenes in lanthanide(III)/actinide(III) differentiation. Chem. Commun. 2005, 2860–2862. 10.1039/b503526k. [DOI] [PubMed] [Google Scholar]
  52. del Mar Conejo M.; Parry J. S.; Carmona E.; Schultz M.; Brennann J. G.; Beshouri S. M.; Andersen R. A.; Rogers R. D.; Coles S.; Hursthouse M. Carbon Monoxide and Isocyanide Complexes of Trivalent Uranium Metallocenes. Chem.—Eur. J. 1999, 5 (10), 3000–3009. . [DOI] [Google Scholar]
  53. Evans W. J.; Kozimor S. A.; Ziller J. W. A monometallic f element complex of dinitrogen: (C5Me5)3U(η1-N2). J. Am. Chem. Soc. 2003, 125 (47), 14264–14265. 10.1021/ja037647e. [DOI] [PubMed] [Google Scholar]
  54. Evans W. J.; Kozimor S. A.; Nyce G. W.; Ziller J. W. Comparative reactivity of sterically crowded nf3 (C5Me5)3Nd and (C5Me5)3U complexes with CO: formation of a nonclassical carbonium ion versus an f element metal carbonyl complex. J. Am. Chem. Soc. 2003, 125 (45), 13831–13835. 10.1021/ja036631l. [DOI] [PubMed] [Google Scholar]
  55. Lukens W. W.; Speldrich M.; Yang P.; Duignan T. J.; Autschbach J.; Kögerler P. The roles of 4f- and 5f-orbitals in bonding: a magnetochemical, crystal field, density functional theory, and multi-reference wavefunction study. Dalton Trans. 2016, 45 (28), 11508–11521. 10.1039/C6DT00634E. [DOI] [PubMed] [Google Scholar]
  56. Nakai H.; Hu X.; Zakharov L. N.; Rheingold A. L.; Meyer K. Synthesis and characterization of N-heterocyclic carbene complexes of uranium(III). Inorg. Chem. 2004, 43 (3), 855–857. 10.1021/ic035142j. [DOI] [PubMed] [Google Scholar]
  57. Maron L.; Eisenstein O.; Andersen R. A. The Bond between CO and Cp′3U in Cp′3U(CO) Involves Back-bonding from the Cp′3U Ligand-Based Orbitals of π-Symmetry, where Cp′ Represents a Substituted Cyclopentadienyl Ligand. Organometallics 2009, 28 (13), 3629–3635. 10.1021/om801098b. [DOI] [Google Scholar]
  58. Mehdoui T.; Berthet J. C.; Thuéry P.; Ephritikhine M. Clear-Cut Lanthanide(III)/Actinide(III) Differentiation in Coordination of Pyrazine to Tris(cyclopentadienyl) Complexes of Cerium and Uranium, Involving Reversible UIII→UIV Oxidation. Eur. J. Inorg. Chem. 2004, 2004 (10), 1996–2000. 10.1002/ejic.200400133. [DOI] [Google Scholar]
  59. Brennan J. G.; Stults S. D.; Andersen R. A.; Zalkin A. Crystal structures of (MeC5H4)3ML [M = uranium or cerium; L = quinuclidine or P(OCH2)3CEt]. Evidence for uranium to phosphorus π-back-bonding. Organometallics 1988, 7 (6), 1329–1334. 10.1021/om00096a016. [DOI] [Google Scholar]
  60. Berthet J.-C.; Miquel Y.; Iveson P. B.; Nierlich M.; Thuéry P.; Madic C.; Ephritikhine M. The affinity and selectivity of terdentate nitrogen ligands towards trivalent lanthanide and uranium ions viewed from the crystal structures of the 1–3 complexes. J. Chem. Soc., Dalton Trans. 2002, (16), 3265–3272. 10.1039/B203725D. [DOI] [Google Scholar]
  61. Seed J. A.; Gregson M.; Tuna F.; Chilton N. F.; Wooles A. J.; McInnes E. J. L.; Liddle S. T. Rare-Earth- and Uranium-Mesoionic Carbenes: A New Class of f-Block Carbene Complex Derived from an N-Heterocyclic Olefin. Angew. Chem., Int. Ed. 2017, 56 (38), 11534–11538. 10.1002/anie.201706546. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Shephard J. J.; Berryman V. E. J.; Ochiai T.; Walter O.; Price A. N.; Warren M. R.; Arnold P. L.; Kaltsoyannis N.; Parsons S. Covalent bond shortening and distortion induced by pressurization of thorium, uranium, and neptunium tetrakis aryloxides. Nat. Commun. 2022, 13 (1), 5923. 10.1038/s41467-022-33459-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Goodwin C. A. P.; Tuna F.; McInnes E. J. L.; Liddle S. T.; McMaster J.; Vitorica-Yrezabal I. J.; Mills D. P. [UIII{N(SiMe2tBu)2}3]: a structurally authenticated trigonal planar actinide complex. Chem.—Eur. J. 2014, 20 (45), 14579–14583. 10.1002/chem.201404864. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Cross J. N.; Su J.; Batista E. R.; Cary S. K.; Evans W. J.; Kozimor S. A.; Mocko V.; Scott B. L.; Stein B. W.; Windorff C. J.; Yang P. Covalency in Americium(III) Hexachloride. J. Am. Chem. Soc. 2017, 139 (25), 8667–8677. 10.1021/jacs.7b03755. [DOI] [PubMed] [Google Scholar]
  65. Galley S. S.; Sperling J. M.; Windorff C. J.; Zeller M.; Albrecht-Schmitt T. E.; Bart S. C. Conversion of Americia to Anhydrous Trivalent Americium Halides. Organometallics 2019, 38 (3), 606–609. 10.1021/acs.organomet.8b00840. [DOI] [Google Scholar]
  66. Wedal J. C.; Windorff C. J.; Huh D. N.; Ryan A. J.; Ziller J. W.; Evans W. J. Structural variations in cyclopentadienyl uranium(III) iodide complexes. J. Coord. Chem. 2021, 74 (1–3), 74–91. 10.1080/00958972.2020.1856824. [DOI] [Google Scholar]
  67. Windorff C. J.; Sperling J. M.; Albrecht-Schönzart T. E.; Bai Z.; Evans W. J.; Gaiser A. N.; Gaunt A. J.; Goodwin C. A. P.; Hobart D. E.; Huffman Z. K.; Huh D. N.; Klamm B. E.; Poe T. N.; Warzecha E. A Single Small-Scale Plutonium Redox Reaction System Yields Three Crystallographically-Characterizable Organoplutonium Complexes. Inorg. Chem. 2020, 59 (18), 13301–13314. 10.1021/acs.inorgchem.0c01671. [DOI] [PubMed] [Google Scholar]
  68. Goodwin C. A. P.; Janicke M. T.; Scott B. L.; Gaunt A. J. [AnI3(THF)4] (An = Np, Pu) preparation bypassing An0 metal precursors: access to Np3+/Pu3+ nonaqueous and organometallic complexes. J. Am. Chem. Soc. 2021, 143 (49), 20680–20696. 10.1021/jacs.1c07967. [DOI] [PubMed] [Google Scholar]
  69. Černá M.; Seed J. A.; Garrido-Fernandez S.; Janicke M. T.; Scott B. L.; Whitehead G. F. S.; Gaunt A. J.; Goodwin C. A. P. Isostructural σ-hydrocarbyl phospholide complexes of uranium, neptunium, and plutonium. Chem. Commun. 2022, 58 (95), 13278–13281. 10.1039/D2CC04803E. [DOI] [PubMed] [Google Scholar]
  70. Goodwin C. A. P.; Ciccone S. R.; Bekoe S.; Majumdar S.; Scott B. L.; Ziller J. W.; Gaunt A. J.; Furche F.; Evans W. J. 2.2.2-Cryptand complexes of neptunium(III) and plutonium(III). Chem. Commun. 2022, 58 (7), 997–1000. 10.1039/D1CC05904A. [DOI] [PubMed] [Google Scholar]
  71. Long B. N.; Beltran-Leiva M. J.; Sperling J. M.; Poe T. N.; Celis-Barros C.; Albrecht-Schonzart T. E. Altering the spectroscopy, electronic structure, and bonding of organometallic curium(III) upon coordination of 4,4’-bipyridine. Nat. Commun. 2023, 14 (1), 3774. 10.1038/s41467-023-39481-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Arnold P. L.; Casely I. J. F-block N-heterocyclic carbene complexes. Chem. Rev. 2009, 109 (8), 3599–3611. 10.1021/cr8005203. [DOI] [PubMed] [Google Scholar]
  73. Hopkinson M. N.; Richter C.; Schedler M.; Glorius F. An overview of N-heterocyclic carbenes. Nature 2014, 510 (7506), 485–496. 10.1038/nature13384. [DOI] [PubMed] [Google Scholar]
  74. Bourissou D.; Guerret O.; Gabbaï F. P.; Bertrand G. Stable Carbenes. Chem. Rev. 2000, 100 (1), 39–92. 10.1021/cr940472u. [DOI] [PubMed] [Google Scholar]
  75. Nelson D. J.; Nolan S. P. Quantifying and understanding the electronic properties of N-heterocyclic carbenes. Chem. Soc. Rev. 2013, 42 (16), 6723–6753. 10.1039/c3cs60146c. [DOI] [PubMed] [Google Scholar]
  76. Huynh H. V. Electronic Properties of N-Heterocyclic Carbenes and Their Experimental Determination. Chem. Rev. 2018, 118 (19), 9457–9492. 10.1021/acs.chemrev.8b00067. [DOI] [PubMed] [Google Scholar]
  77. Huynh H. V.; Frison G. Electronic structural trends in divalent carbon compounds. J. Org. Chem. 2013, 78 (2), 328–338. 10.1021/jo302080c. [DOI] [PubMed] [Google Scholar]
  78. Jacobsen H.; Correa A.; Costabile C.; Cavallo L. π-Acidity and π-basicity of N-heterocyclic carbene ligands. A computational assessment. J. Organomet. Chem. 2006, 691 (21), 4350–4358. 10.1016/j.jorganchem.2006.01.026. [DOI] [Google Scholar]
  79. Scott N. M.; Dorta R.; Stevens E. D.; Correa A.; Cavallo L.; Nolan S. P. Interaction of a bulky N-heterocyclic carbene ligand with Rh(I) and Ir(I). Double C-H activation and isolation of bare 14-electron Rh(III) and Ir(III) complexes. J. Am. Chem. Soc. 2005, 127 (10), 3516–3526. 10.1021/ja043249f. [DOI] [PubMed] [Google Scholar]
  80. Maron L.; Bourissou D. Lanthanide Complexes of Amino-Carbenes: On the Samarium-Carbene Bond from DFT Calculations. Organometallics 2007, 26 (4), 1100–1103. 10.1021/om060924g. [DOI] [Google Scholar]
  81. Goodwin C. A. P.; Su J.; Albrecht-Schmitt T. E.; Blake A. V.; Batista E. R.; Daly S. R.; Dehnen S.; Evans W. J.; Gaunt A. J.; Kozimor S. A.; Lichtenberger N.; Scott B. L.; Yang P. [Am(C5Me4H)3]: An Organometallic Americium Complex. Angew. Chem., Int. Ed. 2019, 58 (34), 11695–11699. 10.1002/anie.201905225. [DOI] [PubMed] [Google Scholar]
  82. Evans W. J.; Grate J. W.; Levan K. R.; Bloom I.; Peterson T. T.; Doedens R. J.; Zhang H.; Atwood J. L. Synthesis and x-ray crystal structure of di(pentamethylcyclopentadienyl)lanthanide and yttrium halide complexes. Inorg. Chem. 1986, 25 (20), 3614–3619. 10.1021/ic00240a017. [DOI] [Google Scholar]
  83. Meng Y. S.; Zhang Y. Q.; Wang Z. M.; Wang B. W.; Gao S. Weak Ligand-Field Effect from Ancillary Ligands on Enhancing Single-Ion Magnet Performance. Chem.—Eur. J. 2016, 22 (36), 12724–12731. 10.1002/chem.201601934. [DOI] [PubMed] [Google Scholar]
  84. CCDC 125077.
  85. Zwick B. D.; Sattelberger A. P.; Avens L. R.. Transuranium Organometallics Elements: The Next Generation. In Transuranium Elements: A Half Century; Morss L. R., Fuger J., Eds.; American Chemical Society, 1992; pp 239–246. [Google Scholar]
  86. Grimme S.; Antony J.; Ehrlich S.; Krieg H. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 2010, 132 (15), 154104. 10.1063/1.3382344. [DOI] [PubMed] [Google Scholar]
  87. Su J.; Batista E. R.; Boland K. S.; Bone S. E.; Bradley J. A.; Cary S. K.; Clark D. L.; Conradson S. D.; Ditter A. S.; Kaltsoyannis N.; Keith J. M.; Kerridge A.; Kozimor S. A.; Loble M. W.; Martin R. L.; Minasian S. G.; Mocko V.; La Pierre H. S.; Seidler G. T.; Shuh D. K.; Wilkerson M. P.; Wolfsberg L. E.; Yang P. Energy-Degeneracy-Driven Covalency in Actinide Bonding. J. Am. Chem. Soc. 2018, 140 (51), 17977–17984. 10.1021/jacs.8b09436. [DOI] [PubMed] [Google Scholar]
  88. Cross J. N.; Macor J. A.; Bertke J. A.; Ferrier M. G.; Girolami G. S.; Kozimor S. A.; Maassen J. R.; Scott B. L.; Shuh D. K.; Stein B. W.; Stieber S. C. Comparing the 2,2’-Biphenylenedithiophosphinate Binding of Americium with Neodymium and Europium. Angew. Chem., Int. Ed. 2016, 55 (41), 12755–12759. 10.1002/anie.201606367. [DOI] [PubMed] [Google Scholar]
  89. Windorff C. J.; Celis-Barros C.; Sperling J. M.; McKinnon N. C.; Albrecht-Schmitt T. E. Probing a variation of the inverse-trans-influence in americium and lanthanide tribromide tris(tricyclohexylphosphine oxide) complexes. Chem. Sci. 2020, 11 (10), 2770–2782. 10.1039/C9SC05268B. [DOI] [PMC free article] [PubMed] [Google Scholar]
  90. Jensen M. P.; Bond A. H. Comparison of Covalency in the Complexes of Trivalent Actinide and Lanthanide Cations. J. Am. Chem. Soc. 2002, 124 (33), 9870–9877. 10.1021/ja0178620. [DOI] [PubMed] [Google Scholar]
  91. Shannon R. D. Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides. Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr. 1976, 32 (5), 751–767. 10.1107/S0567739476001551. [DOI] [Google Scholar]
  92. Crabtree R. H.The Organometallic Chemistry of the Transition Metals; John Wiley & Sons, Inc., 2014. [Google Scholar]
  93. Bursten B. E.; Rhodes L. F.; Strittmatter R. J. The bonding in tris(η5-cyclopentadienyl) actinide complexes IV: Electronic structural effects in AnCl3 and (η5-C5H5)3An (An ≡ Th - Cf) complexes. J. Less-Common Met. 1989, 149, 207–211. 10.1016/0022-5088(89)90487-6. [DOI] [Google Scholar]
  94. Bursten B. E.; Rhodes L. F.; Strittmatter R. J. Bonding of tris(η5-cyclopentadienyl) actinide complexes. 3. Interaction of π-neutral, π-acidic, and π-basic ligands with (η5-C5H5)3U. J. Am. Chem. Soc. 1989, 111 (8), 2758–2766. 10.1021/ja00190a003. [DOI] [Google Scholar]
  95. Bursten B. E.; Rhodes L. F.; Strittmatter R. J. Bonding in tris(η5-cyclopentadienyl) actinide complexes. 2. The ground electronic configurations of ″base-free″ Cp3An complexes (An = thorium, protactinium, uranium, neptunium, plutonium). J. Am. Chem. Soc. 1989, 111 (8), 2756–2758. 10.1021/ja00190a002. [DOI] [Google Scholar]
  96. Strittmatter R. J.; Bursten B. E. Bonding in tris(η5-cyclopentadienyl) actinide complexes. 5. A comparison of the bonding in neptunium, plutonium, and transplutonium compounds with that in lanthanide compounds and a transition-metal analog. J. Am. Chem. Soc. 1991, 113 (2), 552–559. 10.1021/ja00002a024. [DOI] [Google Scholar]
  97. Bursten B. E.; Palmer E. J.; Sonnenberg J. L.. On the Role of f-Orbitals in the Bonding in f-Element Complexes: the “Feudal” Model as Applied to Organoactinide and Actinide Aquo Complexes. In Recent Advances In Actinide Science; Alvares R., Bryan N. D., May I., Eds.; Special Publications, The Royal Society of Chemistry, 2006; pp 157–162. [Google Scholar]
  98. Murillo J.; Seed J. A.; Wooles A. J.; Oakley M. S.; Goodwin C. A. P.; Gregson M.; Dan D.; Chilton N. F.; Gaunt A. J.; Kozimor S. A.; Liddle S. T.; Scott B. L. Carbene Complexes of Plutonium: Structure, Bonding, and Divergent Reactivity to Lanthanide Analogs. J. Am. Chem. Soc. 2024, 146 (6), 4098–4111. 10.1021/jacs.3c12719. [DOI] [PMC free article] [PubMed] [Google Scholar]
  99. Hargreaves W. A. Energy levels of Pr3+ ions in halide crystals. J. Condens Matter. Phys. 1992, 4 (28), 6141–6154. 10.1088/0953-8984/4/28/015. [DOI] [Google Scholar]
  100. Carnall W. T.; Fields P. R.; Rajnak K. Electronic Energy Levels in the Trivalent Lanthanide Aquo Ions. I. Pr3+, Nd3+, Pm3+, Sm3+, Dy3+, Ho3+, Er3+, and Tm3+. J. Chem. Phys. 1968, 49 (10), 4424–4442. 10.1063/1.1669893. [DOI] [Google Scholar]
  101. Sugar J. Energy Levels of Pr3+ in the Vapor State. Phys. Rev. Lett. 1965, 14 (18), 731–732. 10.1103/PhysRevLett.14.731. [DOI] [Google Scholar]
  102. Sarup R.; Crozier M. H. Analysis of the Eigenstates of Pr3+ in LaCl3 Using the Zeeman Effect in High Fields. J. Chem. Phys. 1965, 42 (1), 371–376. 10.1063/1.1695702. [DOI] [Google Scholar]
  103. Crosswhite H. M.; Dieke G. H.; Carter W. J. Free-Ion and Crystalline Spectra of Pr3+ (Pr IV). J. Chem. Phys. 1965, 43 (6), 2047–2054. 10.1063/1.1697073. [DOI] [Google Scholar]
  104. Caspers H. H.; Rast H. E.; Buchanan R. A. Energy Levels of Pr3+ in LaF3. J. Chem. Phys. 1965, 43 (6), 2124–2128. 10.1063/1.1697083. [DOI] [Google Scholar]
  105. Carnall W. T.; Fields P. R.; Wybourne B. G. Spectral Intensities of the Trivalent Lanthanides and Actinides in Solution. I. Pr3+, Nd3+, Er3+, Tm3+, and Yb3+. J. Chem. Phys. 1965, 42 (11), 3797–3806. 10.1063/1.1695840. [DOI] [Google Scholar]
  106. Carnall W. T.; Fields P. R.. Lanthanide and Actinide Absorption Spectra in Solution. Lanthanide/Actinide Chemistry; Advances in Chemistry; American Chemical Society, 1967; Vol. 71, pp 86–101. [Google Scholar]
  107. Carnall W. T. A systematic analysis of the spectra of trivalent actinide chlorides in D3h site symmetry. J. Chem. Phys. 1992, 96 (12), 8713–8726. 10.1063/1.462278. [DOI] [Google Scholar]
  108. Carnall W. T.; Wybourne B. G. Electronic Energy Levels of the Lighter Actinides: U3+, Np3+, Pu3+, Am3+, and Cm3+. J. Chem. Phys. 1964, 40 (11), 3428–3433. 10.1063/1.1725018. [DOI] [Google Scholar]
  109. Carnall W. T.; Crosswhite H.; Crosswhite H. M.; Hessler J. P.; Edelstein N.; Conway J. G.; Shalimoff G. V.; Sarup R. Energy level analysis of Np3+:LaCl3 and Np3+:LaBr3. J. Chem. Phys. 1980, 72 (9), 5089–5102. 10.1063/1.439797. [DOI] [Google Scholar]
  110. Varga L. P.; Baybarz R. D.; Reisfeld M. J.; Asprey L. B. Electronic spectra of the 5f5 and 5f9 actinides: Am4+, Pu3+, Bk2+, Cf3+ and Es4+. J. Inorg. Nucl. Chem. 1973, 35 (8), 2775–2785. 10.1016/0022-1902(73)80508-1. [DOI] [Google Scholar]
  111. Goodwin C. A. P.; Su J.; Stevens L. M.; White F. D.; Anderson N. H.; Auxier J. D.; Albrecht-Schönzart T. E.; Batista E. R.; Briscoe S. F.; Cross J. N.; Evans W. J.; Gaiser A. N.; Gaunt A. J.; James M. R.; Janicke M. T.; Jenkins T. F.; Jones Z. R.; Kozimor S. A.; Scott B. L.; Sperling J. M.; Wedal J. C.; Windorff C. J.; Yang P.; Ziller J. W. Isolation and Characterization of a Californium Metallocene. Nature 2021, 599 (7885), 421–424. 10.1038/s41586-021-04027-8. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ja3c12721_si_001.pdf (5.8MB, pdf)

Articles from Journal of the American Chemical Society are provided here courtesy of American Chemical Society

RESOURCES