Comparative Study of Novel Phosphordiamidite and Phosphite Ligands Used in Alkene Hydroformylation; Synthesis, Characterization, Metalation, and Catalytic Evaluation

Article abstract of DOI:10.1002/ejic.201800724

Three novel ligands have been prepared by the reactions of 1,1′-(chlorophosphinediyl)bis(1H-pyrrole) with either 2-pyridinemethanol or 2,6-pyridinedimethanol or the reaction of 1,1′-biphenyl-2,2′-diyl phosphorochloridite with 2-pyridinemethanol. Measurement of |¹J_P–Se| values demonstrate that the phosphite donor is less basic than the phosphoramidite donors. Coordination preferences of the ligands in octahedral cis-tetracarbonylmolybdenum(0) and square planar cis-dichloropalladium(II) complexes have been evaluated using multinuclear NMR and X-ray crystallography. Rhodium(I) complexes of the bidentate ligands have been evaluated as catalysts for styrene hydroformylation, and their activities are nearly double those of catalysts containing traditional phosphites. The regioselectivities of the Rh^I complexes of the bidentate ligands are not significantly different at 80 °C and 20 atm and are not affected by the addition of a lithium salt. In contrast, changing reaction conditions causes the % n-aldehyde to vary from 15 to 50 %. The regioselectivity of the catalyst containing the bidentate phosphoramidite/pyridine ligand was more sensitive to pressure than temperature with both effects being significant.

Full text of DOI:10.1002/ejic.201800724

DOI: 10.1002/ejic.201800724
Full Paper
Hydroformylation Catalysts
Comparative Study of Novel Phosphordiamidite and Phosphite
Ligands Used in Alkene Hydroformylation; Synthesis,
Characterization, Metalation, and Catalytic Evaluation
Samantha D. Hastings,[a] Ethan C. Cagle,[a] Timothy R. Totsch,[a] Shakyra D. Tyus,[a] and
Gary M. Gray*[a]
Abstract: Three novel ligands have been prepared by the reac- bidentate ligands have been evaluated as catalysts for styrene
tions of 1,1′-(chlorophosphinediyl)bis(1H-pyrrole) with either 2- hydroformylation, and their activities are nearly double those
pyridinemethanol or 2,6-pyridinedimethanol or the reaction of of catalysts containing traditional phosphites. The regioselec-
1,1′-biphenyl-2,2′-diyl phosphorochloridite with 2-pyridine- tivities of the RhI complexes of the bidentate ligands are not
methanol. Measurement of |1JP–Se| values demonstrate that the significantly different at 80 °C and 20 atm and are not affected
phosphite donor is less basic than the phosphoramidite donors. by the addition of a lithium salt. In contrast, changing reaction
Coordination preferences of the ligands in octahedral cis-tetra- conditions causes the % n-aldehyde to vary from 15 to 50 %.
carbonylmolybdenum(0) and square planar cis-dichloropalla- The regioselectivity of the catalyst containing the bidentate
dium(II) complexes have been evaluated using multinuclear phosphoramidite/pyridine ligand was more sensitive to pres-
NMR and X-ray crystallography. Rhodium(I) complexes of the sure than temperature with both effects being significant.
Introduction
of chelating phosphite or phospholane ligands, or hybrids
thereof.[1,25–27] For n-selectivity, some of the most active and
Transition metal catalyzed alkene hydroformylation is an indus- selective catalysts are RhI complexes of chelating phosphites or
trially important, atom economical method of producing “oxo” phosphorodiamidites.[28]
compounds from alkenes and syngas (Figure 1). The regioselec-
tivity of the reaction (major isomer formed iso- or n-) is depend-
ent on reaction conditions, substrate, and catalyst.[1]
Our research has involved synthesis of P-donor ligands that
also contain hard donor sites. Until now, we have focused on
ether donor groups that have the potential to bind alkali metal
Rhodium(I) complexes of P-donor ligands are the best cata- cations, allowing for in situ formation of heterobimetallic RhI
lysts for this reaction. In pursuit of more active and regioselec- complexes. To expand upon this research, it is of interest to
tive catalysts, the development of structurally and electronically incorporate other donating groups that can act as hemilabile
diverse ligands has been emphasized. Initially successful phos- ligands such as pyridyl groups.[14] Similar hetero-
phines have given way to less electron rich P-donor ligands (i.e. donor P,N ligand have been used successfully in Pd-catalyzd
phosphites,[2,3] phosphonites,[4–9] phosphinites,[10–14] and phos- Suzuiki–Miyayra coupling,[29] asymmetric Pd-catalyzed allylic
phoramidites[15–24]) due to increased activities and higher resist- alkylation reactions,[30–32] Cu-catalyzed asymmetric conjugate
ances towards oxidation. To date, the most active and iso-selec- 1,4-additions, and Ir- or Rh-catalyzed alkene hydrogen-
tive catalysts for styrene hydroformylation are RhI complexes ations,[31,33,34] but very few have been studied in Rh-catalyzed
Figure 1. Transition metal catalyzed hydroformylation of styrene.
hydroformylation reactions.[35–37] These catalysts are more
complicated than the mixed P-donor/ether ligands that we
[a] Department of Chemistry, University of Alabama at Birmingham,
901 14th Street South, Birmingham, Alabama 35294-1240, USA
E-mail: gmgray@uab.edu
http://www.uab.edu/chemistry/
Supporting information and ORCID(s) from the author(s) for this article are
previously studied because the pyridyl group is more likely to
competitively bind to the catalytically active RhI metal cen-
available on the WWW under https://doi.org/10.1002/ejic.201800724.
ters.[38–42] Therefore, to gain insight into how these ligands be-
Eur. J. Inorg. Chem. 0000, 0–0
1
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
have in catalytic systems, it is important to study the coordina- NMR spectra). The 31P NMR resonances of L1–L3 are singlets,
tion chemistry in a variety of metal systems in addition to carry- and their chemical shifts are consistent with those reported for
ing out catalytic studies.
other phosphites[2,3] and phosphordiamidites.[28,43,44] Broaden-
ing of the 31P NMR chemical shifts of the pyrrolyl-containing
ligands is typical for these compounds.[45]
In this manuscript, we report the syntheses of three novel
ligands: two potentially bidentate ligands containing either a
phosphite or phosphordiamidite P-donor group and a pyridyl
The |1JP–Se| values of selenium complexes of P-donor ligands
N-donor group, both derived from 2-(pyridine)methanol; and a can be used to estimate their σ-donor capacity (Lewis basicity)
potentially tridentate ligand containing two phosphordiamidite because the magnitude of |1JP–Se| depends on both the electro-
P-donor groups and a pyridyl N-donor group, derived from 2,6- negativity and steric congestion of the substituents of the phos-
pyridinedimethanol. Model complexes of the ligands have been phorus.[1,46] There is an inverse relationship between the
prepared with octahedral cis-tetracarbonylmolybdenum(0) and σ-basicity and the |1JP–Se| of the phosphorus as illustrated by
square planar cis-dichloropalladium(II) centers and have been the following series of |1JP–Se| values: P(OPh)3 (1027 Hz)[47]
<
characterized by multinuclear NMR spectroscopy and X-ray P(pyrrolyl)3 (970 Hz)[48] < P(OMe)3 (954 Hz)[48] < P(Ph)2(pyrrolyl)
crystallography. Additionally, styrene hydroformylation studies (812 Hz)[49] < P(Ph)3 (736 Hz)[49] < P(Me)3 (684 Hz).[49] The
using RhI complexes of these ligands as catalysts have been |1JP–Se| values of the selenium complexes of L1–L3 (Table 1)
carried out in both the presence and absence of an alkali metal indicate that L3 (phosphite) is less basic than are L1 and L2
salt, and the results from these studies have been compared to (phosphordiamidites) and is quite similar to P(OPh3)3, while L1
those of previously reported RhI cataylsts containing phosphite and L2 are quite similar to P(pyrrolyl)3. It is important to note
or bis(phosphite) ligands.
that IR studies of Rh(CO)(acac)(P-donor ligand) complexes
indicate that the π-acidities of the P-donor ligands decrease in
the order P(pyrrolyl)3 (υCO = 2012 cm–1) > P(OPh)3 (υCO
=
2008 cm–1) > P(Ph)3 (υCO = 1978 cm–1),[48] suggesting that
31P–77Se coupling constants alone may not sufficiently describe
the electronic nature of P-donor ligands.
Results and Discussion
Synthesis and Characterization of Phosphordiamidite and
Phosphite Ligands
Table 1. 31P NMR chemical shifts for the ligands and their complexes.
The P-donor ligands L1–L3 were synthesized by the reactions
of either pyridine-2,6-diyldimethanol or 2-pyridinemethanol
with either 1,1′-(chlorophosphinediyl)bis(1H-pyrrole) or 1,1′-bi-
phenyl-2,2′-diyl phosphorochloridite, in the presence of Et3N
with THF as the solvent (Scheme 1). After flash chromatography,
the reactions yielded colorless oily phosphordiamidite or phos-
phite ligands in 65–97 % yields. The ligands are sensitive to
heat and light, therefore were used immediately in subsequent
reactions when deemed pure through spectroscopic methods
(having no unexplained resonances in their 31P, 1H, and 13C
Ligand[a]
Free ligand Se-L
(δ)
[Mo(CO)4(L)n][
[PdCl2Ln]
(δ, |1JP–Se|)[b]
(δ)[a]
(δ)[a]
[c]
ClP(pyrr)2
L1
L2
105.67, bs
114.77, bs
114.56, bs
–
140.93
155.22
–
59.86, d, 967.0
61.04, d, 973.8
85.40[d]
87.43[d]
146.58 (for 2a)
153.98[e] (for 2b)
L3
137.78
80.34, d, 1028.0 187.19
99.78[d]
[a] L, n = 2; L1–L3, n = 1. [b] Chemical shift reported only for 31P–77Se.
[c] Ref.[47] [d] [D6]DMSO. [e] CO elimination generated a second complex,
κ3-P,N,P-[Mo(CO)3(L2)].
Scheme 1. P-donor ligands from this work and previous synthesized by Gray.
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
2
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Like |1JP–Se|, |1JP–W| coupling constants are often used to eval- display three doublet carbonyl resonances characteristic of P,N
uate the Lewis basicity of P-donor ligands. It was of interest to coordination.[34] The magnitudes of |2JP, C| coupling constants
determine if |1JP–Se| and |1JP–W| correlate well with one another are consistent with whether the carbonyl is cis or trans to the
because tungsten analogs of P-donor ligands are easier to ob- phosphorus (cis, approximately 11 Hz; trans, approximately
1
tain than are selenium analogs, and in some cases either selen- 56 Hz). Additionally, both the H and 13C{1H} chemical shifts of
ium or tungsten values are not available in the literature to the α-proton and α-carbon of the pyridine rings are shifted
be used for comparative purposes. The values of |1JP–Se| and downfield in the metal complexes compared to the free ligands.
|1JP–W| that were available in the literature for a variety of li-
The 1:1 reaction of [Mo(CO)4(nbd)] and L3 generated approx-
gands were plotted against one another (Figure 2).[48,50] A linear
imately 10 % of a second product. This minor product was not
least-squares analysis demonstrates a good correlation between
the values (R2 = 0.961), even though the steric and electronic
parameters are changing simultaneously. However, the stan-
dard error is 16 Hz, indicating that using the relationship to
predict the small changes in electronic nature of P-donor li-
gands should be done with caution.
isolated but presumably is the disubstituted P-donor product
based on its smaller 31P{1H} coordination chemical shift com-
pared to that of 3 (minor product: δ = 169.4 ppm; Δδ 31.4 ppm,
major product: δ = 187.2 ppm; Δδ 49.4 ppm). This also is con-
sistent with the Δδ for L3 in 3 being 10 ppm greater than that
of the closely related L5, (2,2′-C12H8O2)POCH2CH2OCH3 in cis-
[Mo(CO)4(L5)2],[52] suggesting the increased Δδ arises from ring
strain. This phenomenon is observed in [RhCOD(L)]BF4 com-
plexes where
a
κ2-P,N chelated [(2,2′-binaphthoxy)P(oxy-
methyl)]pyridine ligand has a 31P NMR chemical shift of δ =
134.0 ppm, whereas the complex in with a κ2-P,N chelated
[(2,2′-binaphthoxy)P(oxyethyl)]pyridine has a 31P NMR chemical
shift of 124.2 ppm due to the less strained seven-membered
chelate ring in the latter complex.[34]
As with L1 and L3, the reaction of L2 with [Mo(CO)4(nbd)]
could potentially give multiple products, however in hexanes,
the reaction yielded cis-[Mo(CO)4(κ2-P,P-L2)] 2a, (31P{1H} NMR
resonance at δ = 146.6 ppm) as a pale yellow powder. This
complex is stable in the solid state; however, if left in a chloro-
form solution at room temperature, the solution turns yellow
and produces a brown precipitate within hours. When this proc-
ess is followed by 31P NMR spectroscopy (Figure 3a), the obser-
vation of a new phosphorus species is indicated by a 31P{1H}
resonance at δ = 146.9 ppm that grows over time. This com-
pound was isolated and characterized by multinuclear NMR
spectroscopy and X-ray crystallography and was identified as
fac-[Mo(CO)3{κ3-P,N,P-(L2)}]. The 31P NMR coordination chemical
shift of L2 in 2b is 6.5 ppm larger than in 2a. It is not clear
whether this is due to additional ring strain or to changes in
the ligands that are coordinated to the molybdenum. As is the
case in 1 and 3, the pyridine coordination in 2b is indicated by
a downfield shift of the α-protons of the pyridine.
Figure 2. Correlation of the literature values of |1JP–Se| to |1JP–W|.
Synthesis and Multinuclear NMR Characterizations of
Model Metal Complexes of L1–L3
To study the preferred coordination geometries of L1–L3, metal
complexes were prepared. Because the Rh complex catalysts
are difficult to isolate and characterize, the coordination behav-
ior of the ligands in octahedral coordination geometry was
studied in Mo(CO)4 complexes, and the coordination behavior
in square-planar complexes was studied in PdCl2 complexes.
In 2a, all eight pyrrolyl-α protons are equivalent, all eight
Each complex was isolated and fully characterized by multi- pyrrolyl-ꢀ protons are equivalent and the CH2 protons are
nuclear NMR spectroscopy, elemental analyses, and, if possible, equivalent. However, when the pyridine displaces a carbonyl to
X-ray crystallography. The 31P{1H} NMR chemical shifts are re- form 2b, the rings are fixed there are now two resonances for
ported in Table 1.
each of the α protons and for each of the ꢀ protons, and the
CH2 protons become diastereotopic (Figure 3b). Through space
interactions [CH2b···pyridine-ꢀ (1), CH2a···pyrr-α1 (2), pyrr-
The cis-[Mo(CO)4Ln] complexes were prepared by the reac-
tion of [Mo(CO)4(nbd)] (nbd = norbornadiene) with the appro-
priate ligand in hexanes (Scheme 2 and Scheme 3). Because
ligands similar to L1 and L3 exhibit multiple coordination
modes: (1) bidentate κ2-P,N chelation, (2) monodentate P-dona-
tion or (3) bidentate P,N bridging,[34,51] the coordination reac-
tions were carried out at multiple stoichiometries. As expected,
ligands L1 and L3 predominantly formed cis disubstituted
1
α1···pyrr-α2 (3)] observed in a H-1H NOESY NMR spectra and
the COSY NMR spectrum (Figure 3b) allow the various resonan-
ces in the 1H NMR spectrum to be assigned. The solution
structure of 2b appears to be quite similar to the solid-state
1
structure as indicated by the fact that the H-1H through space
NOESY interactions are also observed in the solid-state struc-
tetracarbonylmolybdenum(0) complexes with a bidentate κ2- ture of 2b as shown in Figure 3c. The N-pyridine coordination
P,N chelating ligand when a 1:1 stoichiometry was used. The also affects the carbonyl resonances, in the 13C NMR spectra of
coordination mode is confirmed by the 13C NMR spectra that 2a and 2b. While both complexes have two inequivalent carb-
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
3
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 2. Synthesis of chelating molybdenum complexes 1–3.
Scheme 3. Synthesis of palladium chelating complexes 4, 5 and 6.
Eur. J. Inorg. Chem. 0000, 0–0 www.eurjic.org
4
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ample of CO elimination from a group VI tetracarbonyl complex
is with the multi/hetero-donor ligand, Ph2P(o-C6H4)CH=
N(CH2)2(o-C5H4N), (PNN).[54] Peng found that upon heating
cis-[W(CO)4{κ2-P,N-(PNN)}], fac-[W(CO)3{κ3-P,N,N-(PNN)}] was
formed, presumably to satisfy the electron deficient metal cen-
ter (the final N in PNN is a stronger σ-donor and weaker
π-acceptor than is CO).[54]
IR spectroscopy is commonly used to gain insight into the
electron donating/accepting abilities of ligands in transition
metal carbonyl complexes. The stretching frequencies of the
symmetrical stretching band of the carbonyl of the complexes
of the L1 (2029 cm–1) and L3 (2024 cm–1) are quite similar,
which is interesting because one might expect the π-accepting
abilities of these ligands to be different.[45] The values of these
stretching frequencies are similar to those of phosphines {for
example the comparative value for cis-[Mo(CO)4(PPh3)2] is
2023 cm–1}[59] vs. those of phosphites cis-[Mo(CO)4{P(OPh3)}2] is
2046 cm–1.[60]
The PdCl2 complexes were prepared by reactions of the li-
gands with [PdCl2(ACN)2] (Scheme 3). As a whole, these reac-
tions have substantially lower yields compared to the molyb-
denum complexes (43–83 %). Like the molybdenum complexes,
monomer/dimer/oligomer and multiple coordination modes
(cis/trans) and sites (P/N) are available for the labile metal cen-
ters.[29,34,39,40,51,55,61–64] However, due to the relatively limited
bite angles of L1 and L3, in combination with the chelate effect,
the primary isolable products were the mononuclear cis-
[PdCl2(κ2-P,N-L)] complexes (4 and 6). In several cases, addi-
tional broad 31P NMR resonances were detected in the 31P NMR
spectra of the crude Pd complexes. Although these were easily
separated from the predominant product, they could not be
identified. Similar broad peaks in 31P NMR spectra of bidentate
P,N complexes of Pt have been reported as the result of equilib-
rium between cis-[PtCl{Ph2P(CH2)2NMe2-P,N}{Ph2P(CH2)2NMe2-
[65]
P}]Cl and cis-[Pt{Ph2P(CH2)2NMe2-P,N}2]Cl2
as well as cis-
[PtI2(Ph2Ppy-P)2] and the chelated form cis-[PtI{Ph2Ppy-P}-
(Ph2Ppy-P,N)]I.[66]
In addition to the coordination possibilities already men-
tioned for L1 and L3, L2 has the potential for κ3 coordination
similar to that which has been observed with other pyridine
based PNP and PNN ligands.[64,67–69] However the pyridine
nitrogen typically does not participate in metal bonding when
the arms to the central pyridine unit are longer than one
atom.[61,70,71] This is the case for L2, and only cis- [PdCl2{κ2-P,P-
(L2)}], 5, was isolated. This result is consistent with those from
recent 31P NMR spectroscopic studies of the reaction of [Rh-
(μ-Cl)(CO)2]2 with 2,6-bis[(2,2′-binaphthoxy)P(oxymethyl)]pyr-
idine (the binaphthoxy analog of L2). In this case a mixture
of cis and trans monomeric complexes are formed, and DFT
calculations predict that the cis isomer is the energetically fa-
vorable isomer.[71]
Figure 3. a) 31P{1H} NMR spectra of the conversion of 2a to 2b in [D]chloro-
form solution. b) ORTEP diagram of through space interactions in 2b. c) Over-
laid 1H-1H NOESY NMR (blue) and COSY (pink) for 2b.
onyl resonances with AX2 and AXX′ spin systems, the resonance
with the AXX′ spin system is further downfield in 2a (δ =
210.38 ppm for AXX′ vs. 206.84 ppm for AX2) while that for the
AX2 spin system is further downfield in 2b (δ =222.94 ppm for
AX2 vs. 215.19 ppm for AXX′).
It is interesting that cis-[Mo(CO)4{κ2-P,P-(L2)}], 2a, undergoes
CO elimination at ambient temperature because (1) typically
In contrast, the reaction of either cis- or trans-Pt(PhCN)2Cl2
[Mo(CO)3L3] complexes are formed via a trisubstituted interme- with 2,6-bis[(diphenylphosphinyl)oxymethyl]pyridine, a phos-
diate {[Mo(CO)3DMF3][53] or [Mo(CO)3(NCMe)3][54,55]} or (2) can phinite analog of L2, which yields only the dimeric species
be synthesized from [Mo(CO)6] through photolysis. Either [Pt2Cl4(μ-PONOP)2] as the thermodynamic product.[61] This
method requires high temperatures, and/or extended reaction clearly demonstrates that the P-donors preferentially bind to
times.[54,56–58] To the best of our knowledge the only other ex- soft PdII, PtII, and RhI centers, but the type of complex that is
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
5
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
formed is very sensitive to the ligand backbone and the phos-
phorus substituents.
Comparison of the 1H and 13C{1H} NMR spectra of 5,
[PdCl2(L2)], to those of 2a, cis-[Mo(CO)4(L2)], suggest that the
chelate ring conformation of L2 is quite different in the two
complexes. The 1H NMR spectrum of 5 exhibits two singlet reso-
nances for the pyrrolyl α protons and one singlet resonance for
the pyrrolyl ꢀ protons while the 1H NMR spectrum of 2a exhib-
its a singlet resonance for both the pyrrolyl α and ꢀ protons.
Similarly, the 13C{1H} NMR spectrum of 5 has two resonances
for both the pyrrolyl α carbons and the pyrrolyl ꢀ carbons, while
the 13C{1H} NMR spectrum of 2a has one for each. This suggests
that the ring is more rigid/less symmetric in 5 than it is in 2a
although the reason for this is unclear. It does not appear to be
due to pyridine coordination to the Pd in 5 because there are
no shifts in the H and 13C{1H} NMR resonances of the pyridine
protons and carbons relative to those of L2. This is consistent
with the X-ray crystal structure of 5.
1
Figure 5. A stacked plot of the 31P{1H} NMR signals from pure [PdCl2(P,N)L1]
to the disubstituted species and eventual transformation to inequivalent
phosphorus atoms.
Another palladium complex derived from L3 was isolated
when PdCl2 and L3 were stirred in a wet ACN/THF solution for
24 hours. The 31P{1H} NMR spectrum of the product, 7, had
chemical shifts consistent with two inequivalent phosphorus
(δ = 123.51 and δ = 65.32, coupled to one another |2JP, P| =
the parent ligands even though these complexes are structur-
ally diverse (monodentate vs. bidentate) (rigid vs. flexible linkers
in the backbone).[72–75]
1
99.5 Hz) while integration of the H NMR spectrum indicated
that only one pyridyl group was present. On this basis, the
structure of 7 with one chelating [κ2-P,N-(L3)] ligand and one
anionic P(2,2′-C12H8O2)(O–) ligand, shown in Figure 4a was pro-
posed, and this was later confirmed by X-ray crystallography.
Kirchner has found that the reaction of PdCl2 with 2,6-[(2,2′-
C12H8O2)P(amino)]pyridine under similar conditions also results
in selective cleavage of a pyridine arm, giving an analogous κ2-
P,N,P complex, (Figure 4b).[68] Isolation of this product supports
an equilibrium between cis-[Pd(L3-P)2Cl2] and cis-[Pd{κ2-P,N-
(L3)}(L3-P)Cl]Cl suggested by Stelzer.[66] Similarly, when an ex-
cess of L1 is added to an NMR tube charged with [PdCl2{κ2-P,N
(L1)}], a new singlet resonance as well as two doublets appear
in the 31P{1H} NMR (Figure 5). While this demonstrates that the
disubstituted palladium product can be formed, this product
easily converts to a species with inequivalent phosphorus at-
oms. This supports the likelihood of an L1 analog of 7, however
this was not confirmed with X-ray crystallography.
X-ray Crystallography
X-ray quality crystals were obtained for most of the metal com-
plexes synthesized in this research. Details of individual data
collections are in Tables S1–S4. In each, the X-ray data either
confirmed, or established the coordination geometry, and
agreed with NMR spectroscopic analyses. Select bonds and an-
gles are given in Tables S5–S11, and ORTEP[76] diagrams are
given in Figure 6, Figure 7, Figure 8, Figure 9, Figure 10, Fig-
Figure 4. Pd hydrolysis products (a) 7 and (b) Ref.[68]
The palladium complexes of L1 and L3, (4 and 6), have sin-
glet 31P{1H} NMR resonances that are 30–40 ppm upfield of
those of the ligands. These are consistent with the typical Δδ
values for other phosphite-substituted cis-[PdCl2(L)n] com-
plexes, which are between 30 and 34 ppm upfield of those of
Figure 6. Solid-state structure of 1. Thermal ellipsoids are drawn at the 30 %
probability level, and hydrogen atoms are omitted for clarity.
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
6
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ure 11, Figure 12, and Figure 13. All of the structures were
solved without issue except 1, where a disordered pyrrolyl
group was solved in two parts and 3, which was found to be
an inversion twin (major hand: S, 95 %). Also, two molecules
make up the asymmetric unit of 4.
Figure 9. Solid-state structure of 2b. Thermal ellipsoids are drawn at the 30 %
probability level, and hydrogen atoms are omitted for clarity.
Figure 7. Solid-state structure of 3. Thermal ellipsoids are drawn at the 30 %
probability level, and hydrogen atoms are omitted for clarity.
Figure 10. Solid-state structure of 4. Thermal ellipsoids are drawn at the 30 %
probability level, and hydrogen atoms are omitted for clarity.
Figure 8. Solid-state structure of 2a. Thermal ellipsoids are drawn at the 30 %
Figure 11. Solid-state structure of 5. Thermal ellipsoids are drawn at the 50 %
probability level, and hydrogen atoms are omitted for clarity.
probability level, and hydrogen atoms are omitted for clarity.
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
7
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
gands.[77,78] The Mo–P and Mo–N bond lengths are also in line
with the average determined by Mogul 2.473(30) Å and
2.307(39) Å, respectively.[79]
The palladium complexes (4, 5, 6, and 7) have square-planar
coordination geometries as demonstrated by the angles about
the metal center and planes drawn through the metal and at-
oms immediately attached to it (Figure 10–Figure 13, Tables S7–
S9). The atom-to-plane distances are minimal (< 0.08 Å) except
in 5 indicating relatively planar molecules. The Cl–M–Cl angle
is conserved in the most of the complexes, while the bite angles
(P–Pd–P or P–Pd–N), have values > 90° in most cases and P–M–
Cl angles are < 90°. The P–Cl bond distances trans to P atoms
are longer than those trans to N due to the trans effect of the
phosphorus ligand (Table S7).[49,80] The bond lengths of P–Cl,
Pd–N, and Pd–P, and P–O– bonds are comparable to those of
similar compounds.[68,79]
Figure 12. Solid-state structure of 6. Thermal ellipsoids are drawn at the 50 %
probability level, and hydrogen atoms are omitted for clarity.
Chelate Ring Conformations
A six-membered chelate ring consisting of M–N–Cpy–CH2–O–P,
is present in 1, 2b–4, 6 and 7. Table S10 gives the torsion angles
about each of the bonds in this ring in each of the six com-
plexes. The O–P–M–N torsion angles show the largest differen-
ces among the structures; however the ring conformations are
generally conserved and can best be described as twist boats
(1, 2b–4, 6 and 7) or boat (4′). An overlay of the chelate rings
1 and 3 (Figure 14a) shows that the rings align relatively well.
The two molecules in the asymmetric unit of 4 are quite differ-
ent (Figure 14b), possibly due to two different intermolecular
slipped π–π interactions 1) between the pyridyl of 4 and a
pyrrolyl in 4′ [centroid–centroid distance = 3.630(8) Å] and 2)
between the pyridine groups of adjacent 4′ molecules [cen-
troid–centroid distance = 3.696(7) Å].[81] The fact that the
chelate ring conformation is conserved in complexes with dif-
ferent metal centers and different substituents on the phos-
phorus suggests that this is a low energy conformation.
Figure 13. Solid-state structure of 7. Thermal ellipsoids are drawn at the 50 %
probability level, and hydrogen atoms are omitted for clarity.
Metal Center Coordination Geometries
All of the molybdenum carbonyl complexes have distorted oc-
tahedral metal centers. (Figure 6–Figure 10, Table S5 and Table
S6) The biggest distortions from ideal angles about the metal
centers are observed either in the P–Mo–P angles, all greater
than 90°, the P–Mo–N angles, all significantly less than 90°, or
both in the case of 2a (Table S6).
Figure 14. Capped sticks overlays of (a) 1(blue) and 3(red); and (b) overlay of
the two unique molecules in 4.
Steric Evaluation. To gain insight into the steric properties of
ligands, it is important to compare them in different complexes.
The steric properties of monodentate ligands are often defined
Additionally, the carbonyl ligands adjacent to P/N-ligands by cone angles, θ,[82,83] while those of chelating, bidentate li-
bend away from the ligands resulting in angles > 90°. The bond gands are defined by L–M–L bite angles, ꢀn.[84–87] The values
lengths of Mo–C and C–O are within normal values, and none for the ligands in this study are given in Table 2. In complexes
of the structures exhibit shortened C–O bond lengths of the 1, 3, 4, and 6, which have chelating P,N ligands, the bite angles
carbonyl ligands trans to P or N as might be expected due to are relatively similar indicating that neither the phosphorus
the different σ-donating and π-accepting abilities of the li- substituents, (pyrrolyl)2 vs. biphenoxy, nor the coordination ge-
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
8
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ometry of the metal center significantly affect this angle. The [Mo(CO)(6–n){P(X)(2,2′-O2C12H8)}n] complexes (O–P–O–C
=
bite angles of the chelating P,N ligands are smaller than are |46.5 7.1|°, P–O–C–C = |75.1 3.9|°[77,78]). Because the bi-
those of the P,P coordinated ligand L2 in complexes 2a and 5. phenyl groups of the [1,3,2]-dioxaphosphepin rings racemize in
The change from P,P coordination in 2a to P,N,P in 2b does have solution, their complexes often crystallize in centrosymmetric
a significant effect on the bite angles of L2 as shown by the space groups with both enantiomers in the same crystal as do
data in Table S2 and the overlay of 2a (blue) with 2b (red) in 6 (P21/c) and 7 (P21/n).[50,77,88,89] However, enantiomers can also
Figure 15. The P–M–P angle is much larger in 2b than in 2a, segregate upon crystallization into a noncentrosymmetric space
and the P–M–N angles are somewhat smaller than those in 1, groups as does 3.[50,77]
3, 4, and 6. The P,N,P coordination of L2 in 2b does allow an
intramolecular slipped π–π interaction between pyrrolyl groups
[the distance between the centroids drawn at the center of the
pyrrolyl rings is 3.662(12) Å].
Ligand conformation in 7. Complex 7 has an additional inter-
esting conformational feature. The negatively charged oxo
group of the (O–)P(2,2′-C12H8O2) ligand is in a hydrophobic
pocket of biaryl groups and is stabilized by interactions with
the CH2 protons of a neighboring molecule [2.782(19) Å and
2.684(19) Å]. The related Pt complex containing a 2,6-bis-
[(2,2′-C12H8O2)P(amino)]pyridine crystallizes in a similar manner
where the oxo substituent is situated near several aromatic
groups and has a short contact of 2.20(2) Å to an NH proton in
a neighboring molecule.[68] Complex 7 is also interesting
because it has intramolecular π–π interactions between the
two biphenoxy substituents [centroid–centroid distance =
3.813(4) Å].
Table 2. Steric parameters.
Complex
Bn [°]
1
2a
2b
86.91(4)
93.357(18)
P1–P1a:
101.339(15)
P1–N3:
83.95(4)
P1a–N3:
83.32(4)
85.11(6)
92.90(5)
89.68(5)
95.13(9)
90.52
3
4
Styrene Hydroformylation Catalysis
5
6
Activities. The data in Table 3 demonstrate that catalysts con-
taining L1 and L3 display relatively good activities over the
range of conditions studied, except in some low-pressure reac-
tions where reaction completion was not achieved. The hydro-
formylation rates and regioselectivities for catalysts containing
ligands L4–L6 from previous work has been provided to dem-
onstrate that there are clear differences in activities as a result
of structural changes in the ligands. In the absence of salt, com-
parison of the activity of catalysts containing ligands with phos-
phepin-donor groups (L3, L4 and L5) demonstrates that the
presence of the pyridyl group has a positive impact on hydro-
formylation activity (the pseudo first-order rate constant for the
catalyst containing L3 is approximate twice as large as those
for the catalysts containing L4, L5 or L6) (Figure 16). In contrast,
the electronic nature of the ligand has little effect on the activi-
ties (the pseudo first-order rate constants for the catalyst con-
taining L1 and L3 are approximately the same). This is some-
what surprising because typically pyrrolyl based ligands have
higher activities due to increased π acidities (the resulting Rh
complex has a higher affinity for CO elimination and alkene
coordination).[36,44,90] The similarity of the activity of the catalyst
containing L1 to that containing L3 is in agreement with the
similarity between the carbonyl stretching frequencies of the
[Mo(CO)4L] complexes of these ligands.
Figure 15. Capped sticks overlays of 2a(blue) and 2b(red). Full pyrrolyl groups
are omitted for clarity.
Conformations of [1,3,2]Dioxaphosphepin Rings
A number of the complexes in this manuscript contain [1,3,2]di-
In contrast to the RhI complexes of L1, L3 or L4, the RhI
oxaphosphepin rings that have interesting conformations as de- complex of L2 is not a very active catalyst for styrene hydro-
fined by their torsion angles. Two of the torsion angles (C7–C6– formylation. This may be explained by the coordination studies
C1–O1 and O2–C12–C7–C6) are constrained to be nearly zero with molybdenum carbonyl complexes that demonstrated the
due to the planar aromatic rings, |5.2 2.4|°. The other torsion propensity for L2 to adopt a κ3 coordination that may inhibit
angles are highly conserved in all of the structures and differ the catalytic process. It is interesting, however, that the very
by no more than 13° with C1–C6–C7–C12 torsion angles exhib- closely related styrene hydroformylation catalyst 2,6-bis[(2,2′-
iting the smallest deviations, |40.4 2.2|° (Table S11). These binaphthoxy)P(oxymethyl)]pyridine was not shown to coordi-
torsion angles are within the typical ranges observed for nate to Rh(CO)Cl in this manner in NMR experiments; however,
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
9
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 3. Variable temperature, pressure, and salt content in the hydroformyl-
these studies were not carried out under catalytic conditions.[71]
Due to this low activity, the catalyst containing L2 was only
screened under conditions that were expected to give the high-
est and lowest % n-selectivity.
ation of styrene catalyzed by RhI complexes of L1–L6.[a]
For all of these catalysts, best activities for styrene hydro-
formylation are achieved at high temperatures and high syngas
pressures (Table 3), with the latter being a bit unusual. Most
hydroformylation catalysts studied are zero order in H2(g) and
negative order in CO(g),[27,91–96] although there are a few excep-
tions.[97,98]
Because it is well established that the effects of temperature,
ligand, and pressure are not always independent of one an-
other, the activities of RhI complexes containing L1 or L3 in
styrene hydroformylation were evaluated at multiple tempera-
tures and pressures (Table 3). The studies demonstrate that the
activities are sensitive to temperature and pressure to different
extents.[71,100] Increasing pressure from 5 to 20 atm is accompa-
nied by a large increase in activity (catalyst containing
L1 ≈ 230 %; catalyst containing L3 ≈ 290 %). Similar activity
dependences on pressure have recently been demonstrated in
a number of studies.[101] Increasing the temperature from 65
to 80 °C also resulted in increases of activities of the catalysts
(L1 ≈ 95 %, L3 ≈ 150 %). When the reaction catalyzed by the
RhI complex of L1 was run at 100 °C, higher rates were initially
observed but the reaction stopped before all the styrene was
consumed, presumably due to decomposition of the catalyst.
The effect of an alkali metal salt on the activities of styrene
hydroformylation reactions catalyzed by RhI complexes contain-
ing L1–L6 is less pronounced than are those of temperature
and pressure (Table 3). The amount of change in activity that is
observed upon addition of salt is ligand dependent. The cata-
lysts containing L4, L5 or L6 were significantly improved upon
addition of LiBPh4·3dme (57, 91, and 38 % increases, respec-
tively), whereas those of L1 and L3 only have increases of
< 30 %. Unfortunately, although salt effects on activities have
previously been reported, there is not enough data to deter-
mine structure/activity relationships.[71,72,74,101,102]
Regioselectivity Comparison at 80 °C and 20 atm. As Figure 17
illustrates, the regioselectivities of catalysts containing L1 and
L3 are similar when compared at the same temperature and
pressure (80 °C, 20 atm, no salt: 23 1.4 % n-aldehyde). This
is somewhat atypical because the regioselectivities are usually
dependent on the electronic and steric properties of the li-
gand.[1,28,103–105] Specifically, there are literature reports that the
n-selectivites increase in catalysts containing better π-acceptor
ligands.[28,41,44,106,107,109]
[a] All experiments were performed in THF with 1000:1 substrate:Rh. The
pseudo first-order rate constant, k, was obtained from a first-order fit of pres-
sure drop vs. time, k(s–1), where A = Aoe–kt. Error for L4 and L5 was reported
as maximum error, standard deviations were not calculated from data. When
pressure change was no longer observed % n was determined using 1H NMR
spectroscopy. [b] 30 % completion after 2 h. [c] 60 % completion after 12 h.
[d] Experiments were not repeated due to low activity. [e] 2 % aldol conden-
sation. [f] 80 % complete after 13 h. [g] Ref.[110] [h] Refs.[73,74] [i] Ref.[101]
Figure 16. Comparison of activites of rhodium complexes of L1–L6.
Figure 17. Comparison of regioselectivities of rhodium complexes of L1–L6.
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
10
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Our group, followed by several others, has demonstrated phosphites increased n-selectivity from 49 % to 87 %.[28] Like-
that addition of an alkali metal salt to styrene hydroformylation wise increasing the temperature from 50 °C to 80 °C in catalyst
reactions catalyzed by RhI complexes with P-donor ligands may systems containing bis[(2,2′-binaphthoxy)]P[oxymethyl]pyridine
have significant effects on regioselectivity and that this does increased n-selectivity from 10.4 % to 50.8 %.[71] The effect of
not appear to require cation binding.[71,72,74,102,111] Therefore, it temperature on the regioselectivities of RhI catalysts containing
was of interest to determine if alkali metal salts had an effect on P-donor ligands has been attributed to the selectivity-fixing
the regioselectivities of RhI complexes of (oxymethyl)pyridine step (insertion into the Rh–H bond) becoming reversible at high
functionalized L1 and L3 in styrene hydroformylation. The cata- temperatures. The rate of ꢀ-elimination is much greater for the
lysts containing L1 or L3 exhibited less than 3 % decreases in iso isomer than for the n isomer, therefore, increasing tempera-
n regioselectivity upon addition of lithium tetraphenylborate ture results in increased n-selectivities.[28,113] As expected, in-
(65 °C and 20 atm), This is somewhat surprising because similar creasing temperature from 65 to 80 °C with the catalysts con-
RhI complexes of ligands containing phosphite/(oxymethyl)- taining L1 and L3 increased the n-selectivity from 15.9 % to
pyridine donor groups studied by Pereira exhibited large 24.8 % and from 11.9 % to 20.5 %, respectively.
changes in regioselectivities upon addition of an alkali metal
salt [from 51 % n (no salt) to 26 % n (with salt), 80 °C and
14.8 atm].
To further evaluate the importance of the functional substit-
Factorial Design and Response Surfaces
Factorial design has been shown to be a useful method in cata-
lyst optimization allowing the effects of various factors that af-
fect catalyst performance to be quantified and determining any
interactions between these factors.[90,117] Because temperature
and pressure greatly affect the regioselectivities of the RhI com-
plexes of L1 and L3, a 22 factorial design should allow those
parameters to be quantified. Only the catalyst containing L1
was studied because the catalysts containing L1 or L3, exhib-
ited similar activities and regioselectivities in the preliminary
studies.[71,100,116,117] The two parameters (P and T) were studied
at two different levels, resulting in four unique experiments
(Table 4). The catalysts were performed with syngas pressure of
either 5 or 20 atm, and temperatures of either 65 or 80 °C.
uent of the phosphite ligands on regioselectivity, the results
with the RhI complex of L3 were compared to those with other
biphenoxy phosphite ligands L4–L6. At 80 °C and 20 atm of
syn gas, the addition of salt decreased the % n. Decreases were
greatest for the RhI complexes of the mono phosphites, L4
(9 %) and L5, (8 %), and smaller for the RhI complexes of L3
(2.3 %) and the bis(phosphite) L6 (3 %). Although it is clear that
ligand properties influence the effect of the alkali metal salts
on regioselectivities, it is not clear what criteria are necessary
for significant changes to be observed.
Effect of Temperature and Pressure on the
Regioselectivities of Catalysis Containing L1 and L3
Table 4. Factorial design and regioselectivity results of the hydroformylation
of styrene using L1 in THF with variable pressure and temperature.[a]
Catalysts containing L1 or L3 have similar ranges of regioselec-
tivities as the temperatures are varied from 65 to 80 °C and the
pressures are varied from 5 to 20 atm yielding between 15 and
50 % n-aldehyde. The best n-selectivity is observed at high
temperature (80 °C) and low pressure (5 atm) (L1, 51.3 % n-
aldehyde; and L3, 47.9 % n-aldehyde), which is consistent with
general trends seen in the literature.
The absolute change in n-selectivity is larger over the range
of pressures studied than is the change over the range of
temperatures studied. Decreasing pressure of the hydroformyl-
ation reactions catalyzed by RhI complexes of L1 or L3, from
20 atm to 5 atm, at 80 °C increased the n-selectivity from
24.8 % to 51.3 % and from 20.55 % to 47.9 %, respectively. In-
creased n-selectivity with reduced pressure in hydroformylation
reactions has previously been reported in the litera-
ture,[27,94,95,99,100,107,112–115] and has been attributed to different
rate laws at different pressures. As Landis has demonstrated,
decreasing pressures to very low values can increase the
amount of n-aldehyde produced.[27] However because the rates
of the catalysts containing L1 or L3 decrease as the pressure is
decreased, it is not practical to carry out the reactions at low
pressures.
Run
Pressure (P)
Temperature (T)
% n-aldehyde
[atm]
code
[°C]
code
1
2
3
4
5
5
20
20
–1
–1
+1
+1
65
80
65
80
–1
+1
–1
+1
31.9
51.3
15.9
24.8
[a] Conditions (all ratios are molar): CO/H2 = 1:1, substrate:Rh 1000, k (s–1) at
ligand/Rh = 2.4:1. No hydrogenation was observed. When pressure change
was no longer observed % n was determined using 1H NMR spectroscopy.
To determine the significance of the parameters of the facto-
rial design (P and T), a student's t-test was used. The absolute
difference of the response between the high and low levels,
Dex, was calculated using Equation (1), and this value was com-
pared with that of the calculated difference at a given confi-
dence level, Dcalc using Equation (2). If Dex is greater than Dcalc
the factor is significant.
,
(1)
(2)
Increases in n-selectivity with increasing temperature are
well documented in the literature. For example, increasing the
The effects of temperature and pressure are significant with
temperature from 80 °C to 140 °C in styrene hydroformylation a 99.95 % certainty (the value of Dex for pressure is 22.25 and
reactions catalyzed by RhI complexes of a binol-bridged bis- the value for temperature is 14.2, both greater than Dcalc at
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
11
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
99.95 %, 5.58). The cross term between these variables is also the x axis (pressure) than on the y axis (temperature). To deter-
significant (Dex of 5.54, Dcalc 5.22 for p < 0.01), as illustrated by mine the precision of the model, a data point that was not used
the cross-plot in Figure 18. However, because the trend lines to generate the model was compared to the model's output for
do not intersect, the interaction term of pressure and tempera- the specified conditions. We find that the predicted regioselec-
ture is not very important over the range of reaction conditions tivity at 100 °C and 5 atm (64.9 % n) is in very good agreement
studied. For the above error analysis, the variance between two with the experimental value (61.8 % n).
runs was used as an estimate of standard deviation in spooled
(3)
calculations. To validate this assumption, we carried out an
ANOVA using Minitab. As with the t-test analysis, each of the
parameters were significant, although the p-value was a bit
lower, p < 0.0001.
Equation (3) can be used to predict the conditions necessary
to optimize the catalyst for either iso or n-selectivity. The model
predicts that 100 % n-aldehyde selectivity could be obtained at
160 °C and 5 atm, or at 118 °C and 1 atm. Unfortunately, the
catalyst containing L1 decomposed when the reaction was car-
ried out at 100 °C, so neither of these options are possible.
Better n-aldehyde selectivity can be obtained by lowering the
pressure. If pressure is lowered to 1 atm at 80 °C, the model
predicts that the n-aldehyde selectivity would be 62.0 %, unfor-
tunately this would greatly reduce the rate of the reaction. Ac-
cording to the model in order to maximize the % iso-aldehyde
at a temperature of 80 °C, the syngas pressure should be in-
creased to 27.5 atm.
Conclusions
Three novel ligands have been synthesized in good yields and
have been characterized by multinuclear NMR spectroscopy.
Evaluations of the σ-donating abilities of the ligands compared
via |1JP–Se| indicate that the phosphordiamidite ligands, L1 and
L2, are more basic than the phosphite analog, L3. However, this
does not take into account the π-acidities of the ligands.
Model complexes of the ligands with cis-tetracarbonylmolyb-
denum(0) or cis-dichloropalladium(II) centers indicate that L1
and L3 prefer to coordinate in a κ2-P,N fashion with the chelate
effect overcoming the preference of PdII to bind to phosphorus
rather than nitrogen. On the other hand, only cis-[PdCl2(κ2-P,P-
L2)] was isolated, suggesting that if a chelating coordination
mode is possible, the PdII will coordinate P,P rather than P,N.
Because the yields of the PdII complexes were low and not all
of the palladium complexes were identified, it is not possible
to determine what the preferred coordination mode for L2 is
in PdII complexes. Ligand L2 did exhibit multiple coordination
modes when coordinated to tetracarbonylmolybdenum(0) cen-
ters in which the initially formed cis-[Mo(CO)4(κ2-P,P-L2)] com-
plex spontaneously eliminated a carbonyl to form the fac-
[Mo(CO)4(κ3-P,N,P-L2)] complex. IR studies of molybdenum com-
plexes demonstrated that complexes of L1 and L3 have similar
stretching frequencies and their π-accepting character is more
similar to that of phosphines than of phosphites.
Figure 18. Duel plot of 22 factorial design of catalysts containing L1.
The experimental data can be collapsed into one function
defining the system, %regio = a0 + aPP + aTT + aPTPT;
where %regio denotes the percentage of n-aldehyde. The val-
ues of the constant (a0) the main affects (aP and aT) and the
interaction term (aPT) allow for predictions of conditions neces-
sary for optimal regioselectivies. Mathematica was used to solve
the system of equations to generate the resultant function
[Equation (3)] and the response surfaces (Figure 19).[118]
X-ray crystallographic studies were used to determine the ꢀn
and θ steric parameters of the ligands. Ligands that coordinated
in a κ2-P,N manner have constrained ꢀn values between 85 and
89°, whereas complexes in which bidentate ligands are coordi-
nated in κ2-P,P mode or that have monodentate ligands have
ꢀn values over 90°. One exception to this observation is found
Figure 19. Response surface of regioselectivities (% n-aldehyde) of catalysts
containing L1 under variable reaction conditions.
The values of the coefficients in Equation (3) agree with the in [PdCl2(κ2-P,P-L2)] where the square-planar complex can flex
student's t-test as they have the correct sign and relative magni- (become less planar) to accommodate the ligand's steric re-
tude. This is reflected in the larger gradation in Figure 19 on quirements.
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
12
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
analysis was carried out by Atlantic Microlabs, and all samples were
vacuum dried for 24 hours prior to analysis. Abbreviations: py =
pyridine, pyrr = pyrrole or pyrrollyl.
Studies of the transition metal catalyzed hydroformylation
of styrene using ligands L1–L3 were carried out at multiple
temperatures and pressures, with and without alkali metal salts.
The most interesting result is that P-donor/pyridine hybrid li- Synthesis of Ligands (Scheme 1)
gands give catalysts that have nearly double the activities of
catalysts containing traditional phosphites. Unfortunately, these
ligands are thermally unstable.
2-({[Di(1H-pyrrol-1-yl)phosphanyl]oxy}methyl)pyridine (L1):
A
solution of 1.47 mL (15.2 mmol) of 2-pyridinemethanol in 10 mL of
THF was added dropwise to a solution of 3.16 g (16.0 mmol) of 1,1′-
(chlorophosphinediyl)bis(1H-pyrrole) and 2.70 mL (19.0 mmol) of
triethylamine in 40 mL of THF over 30 min. This reaction mixture
was stirred for one hour. An aliquot of the reaction mixture was
Catalysts containing L1 or L3 have very similar regioselec-
tivities over a range of reaction conditions despite the phos-
phorus donors having different σ-donor properties. The regio-
selectivites were not significantly affected by addition of an
alkali metal salt unlike some closely related catalysts reported
in the literature. It is still unclear what criteria are necessary for
alkali metal salt to be an effective tool in tuning regioselectivity.
The factorial design of hydroformylation experiments con-
taining L1, and the resultant response surfaces demonstrate
that the effects of both pressure and temperature on regiose-
lectivities are statistically significant, and that by varying these
parameters over the range studied one can vary the % n-alde-
hyde from 15 to 50 %. The cross-plot indicates that the interac-
tion term between pressure and temperature is small; however,
it does increase the accuracy of the model as demonstrated by
the excellent agreement between the predicted regioselectivity
at 100 °C and 5 atm (64.9 % n) and the experimental value
(61.8 % n). According to the model, the ligand is maxed out in
terms of its n-selectivity. The model does predict that 99 % iso-
selectivity may be obtained at 80 °C with syngas pressures
above 27.5 atm.
1
evaporated to dryness under high vacuum. A H NMR spectrum of
the residue indicated that no unreacted alcohol remained. The reac-
tion mixture was filtered through a 60 mL sintered Schlenk funnel
containing a 1 cm pad of Celite, and the filtrate was first taken
to dryness on a rotary evaporation and then under high vacuum
overnight. Dry toluene (30 mL) was added to the residue, and the
resulting mixture was filtered through a sintered Schlenk funnel
containing 2 cm of activated neutral alumina. The alumina was then
washed with two, 10 mL portions of toluene. The filtrate and the
washes were combined, and the solvent was removed en vacuo to
yield 3.27 g (79.2 %) of colorless oil. The apparent density was found
to be 1.17 0.015 g/mL. 31P{1H} NMR (162.00 MHz): δ = 114.77 (bs)
ppm. 1H NMR, (700 MHz. CDCl3): δ = 8.55 (dd, |3JHH| 4.8, |4JHH| =
2.0 Hz, 1 H, Hpy–α), 7.68 (ddd, |3JHH| ≈ |3JHH| = 7.7, |4JHH| = 2.0 Hz, 1
H, Hpy–γ), 7.31 (dd, |3JHH| = 7.8, |4JHH| = 1.5 Hz, 1 H, Hpy–δ), 7.21 (ddd,
|3JHH| = 7.6, |3JHH| = 4.8, |4JHH| = 1.5 Hz, 1 H, Hpy–ꢀ), 7.02 (aq, |3JPH
|
≈ |3JHH| ≈ |4JHH| = 2.0 Hz, 4 H, Hpyrr–α), 6.34 (at |3JHH| ≈ |3JHH| =
2.0 Hz, 4 H, Hpyrr–ꢀ), 5.06 (d, |3JPH| = 10.0 Hz, 2 H, CH2) ppm. 13C{1H}
NMR (176 MHz. CDCl3): δ = 156.61 (d, |3JPC| = 7.7 Hz, Cpy–ipso) 149.26
(s, Cpy–δ), 137.00 (s, Cpy–γ), 122.94 (s, Cpy–ꢀ), 121.49 (d, |2JPC| =
15.4 Hz, Cpyrr–α), 121.15 (s, Cpy–α), 112.35 (d, |3JPC| = 4.6 Hz, Cpyrr–ꢀ),
69.61 (d, |2JPC| = 18.9 Hz, CH2) ppm.
2,6-Bis({[di(1H-pyrrol-1-yl)phosphanyl]oxy}methyl)pyridine
(L2): A solution of 3.41 g (17.2 mmol) of 1,1′-(chlorophosphine-
diyl)bis(1H-pyrrole) and 2.70 mL (19.0 mmol) of triethylamine in
40 mL of THF was added to a solution of 1.05 g (7.58 mmol) of
pyridine-2,6-diyldimethanol in 20 mL of THF. After thirty minutes,
the solution was filtered through a Schlenk funnel containing a
1 cm pad of Celite. The filtrate was evaporated to dryness first on
a rotary evaporator and then under high vacuum overnight. The
product was then dissolved in toluene (40 mL) and filtered through
activated neutral alumina. The alumina was washed twice with,
10 mL portions of toluene, and the washes and filtrate were com-
bined. After the solvent was removed en vacuo 2.14 g (65.0 % yield)
of colorless oily ligand. 31P{1H} NMR (162 MHz. CDCl3): δ = 116.30
(bs) ppm. 1H (700 MHz. CDCl3): δ = 7.68 (t, |3JHH| = 7.8 Hz, 1 H,
Hpy–ꢀ), 7.24 (d, |3JHH| = 7.7 Hz, 2 H, Hpy–α), 7.05 (aq, |3JPH| ≈ |3JHH| ≈
|4JHH| = 2.0 Hz, 8 H, Hpyrr–α), 6.37 (at,|3JHH| ≈ |3JHH| = 2.1, 8 H,
Hpyrr–ꢀ), 5.02 (d, |3JPH| = 10.3, 4 H, CH2) ppm. 13C{1H} NMR (176 MHz.
CDCl3): δ = 156.29 (d, |3JPC| = 7.3 Hz, Cpy–ipso)137.85 (s, Cpy–ꢀ), 121.53
(d, |2JPC| = 15.3 Hz, Cpyrr–α), 120.21 (s, C Cpy–α), 112.40 (d, |3JPC| =
4.6 Hz, Cpyrr–ꢀ), 69.46 (d, |2JPC| = 18.8 Hz, CH2) ppm.
Experimental Section
Materials and Methods: Each reaction was carried out under an
atmosphere of high purity nitrogen by using standard Schlenk-line
and glove-box techniques. All of the glassware was flame dried and
placed on the vacuum pump for at least 20 minutes. THF was pre-
dried with magnesium sulfate and then distilled twice, first from
CaH2 then from sodium/benzophenone. Triethylamine was predried
with KOH, and was then distilled from of sodium/benzophenone.
Toluene was dried by distillation from sodium. After distillation, all
solvents were stored over molecular sieves for a minimum of 24
hours prior to use (3 Å; 8–12 mesh). Pyrrole was stored over CaH2
at 4 °C and was subsequently fractionally distilled under aspirator
pressure (the fraction used was collected at 48 °C). The 1,1′-bi-
phenyl-2,2′-diol was purchased from Aldrich, and was dried by
azeotropic distillation with toluene. The 1,1′-biphenyl-2,2′-diyl
phosphorochloridite[119] and 1,1′-(chlorophosphinediyl)bis(1H-pyrr-
ole),[108] were prepared according to literature methods.
All NMR spectra were recorded on a Bruker DPX-300, DRX-400, or
Avance-700 MHz spectrometer. The 31P NMR spectra were refer-
enced to external 85 % phosphoric acid, and both the 13C{1H} and
1H NMR spectra were referenced to internal SiMe4 (TMS). Some as-
signments were based on 2D NMR spectra (1H-13C HSQC, 1H-1H
COSY, and 1H-13C HMBC, NOESY, and J-resolved spectroscopy).
2-[(Dibenzo[d,f][1,3,2]dioxaphosphepin-6-yloxy)methyl]pyr-
idine (L3): A solution of 0.60 mL (6.22 mmol) of 2-pyridinemethanol
in 10 mL of THF was added dropwise over 15 min to a solution of
1.54 g (6.14 mmol) of 6–1,1′-biphenyl-2,2′-diyl phosphorochloridite
Chemical shifts (δ) are in parts per million (ppm). The digital resolu- and 2.0 mL (14.3 mmol) of triethylamine in 50 mL of THF. After 1
tions are as follows (Hz): 700 MHz spectra, 1H and 13C{1H} are 0.22
and 0.34 respectively, 400 MHz, 1H, 13C{1H}, and 31P{1H}
are 0.22, 0.60, and 0.99 Hz, and 300 MHz spectra 1H and
hour, the reaction mixture was filtered through a 1 cm pad of Celite
in a Schlenk funnel, and the filtrate was evaporated to dryness on
a rotary evaporation to yield a rust-colored oil. The residue was
13C{1H} are 0.22 and 0.30 Hz. All deuterated solvents were ob- dissolved in 50 mL of dry toluene and purified via flash chromatog-
tained from Cambridge Isotope Laboratories, Inc. The elemental raphy through 2 cm of activated neutral alumina. After removal of
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
13
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
the solvent in vacuo, 1.95 g (97.5 % yield) of colorless oil was ob-
tained. 31P{1H} NMR (162 MHz. CDCl3): δ = 137.78 (s) ppm. 1H
(700 MHz. CDCl3): δ = 8.53 (d, |3JHH| = 4.2 Hz, 1 H, Hpy–α), 7.71 (t,
|3JHH| = 7.0 Hz, 1 H, Hpy–ꢀ), 7.49 (d, |3JHH| = 9.8 Hz, 1 H, Hpy–δ), 7.48
(d, |3JHH| = 8.4 Hz, 2 H, H2, H1), 7.34 (t, |3JHH| = 7.7 Hz, 2 H, H3,H10),
7.28 (t, |3JHH| = 7.0 Hz, 2 H, H4, H9), 7.20 (d, |3JHH| = 7.0 Hz, 1 H,
Hpy–γ), 7.18 (d, |3JHH| = 7.7 Hz, 2 H, H5, H8), 5.11 (d, |3JPH| = 7 Hz, 2
H, CH2) ppm. 13C{1H} NMR (176 MHz. CDCl3): δ = 157.41 (d, |3JCP| =
4.0 Hz, Cpy–ipso), 149.82 (d, |2JCP| = 5.5 Hz, C1, C12), 149.10 (s, Cpy–α),
136.83 (s, C17), 130.96 (d, |3JCP| = 3.0 Hz, C6, C7), 130.08 (s, C2, C11),
129.43 (s, C3, C10), 125.31 (s, C4, C9), 122.61 (s, Cpy–γ), 121.94 (s, C5,
C8), 121.09 (s, Cpy– δ), 66.5 (d, |2JCP| = 3.52 Hz, CH2) ppm.
|2JPC
+
4JP′C| = 3.6 Hz, Cpyrr–α), 118.80 (s, Cpy–α), 111.99 (aq, |3JPC
+
4
5JP′C| = 2.3 Hz, Cpyrr–ꢀ), 69.76 (aq, |2JPC + JP′C| = 4.1 Hz, CH2) ppm.
cis-[Mo(CO)3(L2)] (2b): Approximately 100 mg of 2a (31P{1H} NMR;
146.58 ppm) was dissolved in chloroform and allowed to stand at
room temperature in ambient light in a quartz NMR tube. Within
24 h the solution turned bright yellow, and a new 31P{1H} NMR
resonance for 2b was observed at δ = 154.17 ppm in a 0.18 to 1.0
ratio to that of 2a. After another16 h, the ratio of the two resonan-
ces were 0.28:1 (2b:2a) and several new resonances accounting for
11 % of the phosphorus resonances in the sample were also ob-
served. After two additional days, the solution became dark brown,
and 2a:2b ratio reached 0.44:1. Crystals precipitated from solution,
and 7 mg were collected and were used for X-ray diffraction and
NMR characterization. Attempts to obtain additional 2b from the
filtrate by evaporation to dryness and separation on a prep silica
plate resulted in decomposition of both complexes. 31P{1H} NMR
(162 MHz; [D]chloroform): δ = 154.17 (s) ppm. 1H (700 MHz; [D]-
chloroform): δ = 7.93 (t, |3JHH| = 7.6 Hz, 1 H, Hpy–ꢀ), 7.53 (d, |3JHH| =
7.7 Hz, 2 H, Hpy–α), 7.33 (s, 4 H, Hpyrr–α1), 6.42 (br. s, 4 H, Hpyrr–ꢀ1),
6.15 (br. s, 4 H, Hpyrr–α2), 5.88 (s, 4 H, Hpyrr–ꢀ2), 5.75–5.79 (m, 2 H,
CH2a), 5.22 (m, 2 H, CH2b) ppm. 13C{1H} NMR (176 MHz; [D]chloro-
form): δ = 222.94 (t, |2JPC| = 12.0 Hz, C15), 215.19 (aq, |2JPC + 2JP′C| =
Synthesis of Metal Complexes (Scheme 2 and Scheme 3)
NMR Scale Preparation of Selenium Complexes of L1–L3: An
excess of selenium powder was added to an NMR tube containing
the ligand (L1, L2, or L3) in chloroform, and the tube was shaken
vigorously before being allowed to stand at ambient temperature
for several hours. Then a 31P{1H} NMR spectrum was obtained. In
each case, only a resonance due to the selenium complex was ob-
served. Chemical shifts and coupling constants for these complexes
are given in Table 1.
33.6 Hz, C17, C16), 160.69 (s, C10, C14), 138.97 (s, C12), 126.10 (s, C11
,
cis-[Mo(CO)4(L1)] (1): A mixture of 0.18 mL (0.777 mmol) of L1 and
0.321 g (0.388 mmol) of [Mo(CO)4nbd] in 15 mL of hexanes was
stirred for one hr at room temperature. During this time, the solu-
tion turned bright yellow, and a pale-yellow solid precipitate
formed. After one hour, the solid was collected by filtration, rinsed
with two, 5 mL portions of hexanes, and dried under high vacuum.
This yielded 0.351 g (94.2 %) of the yellow powder. X-ray quality
crystals were obtained through thermal recrystallization from di-
chloromethane/hexanes mixture after flash chromatography
through silica gel. Anal. C18H14MoN3O5P: C, 45.11; H, 2.94; found C,
45.09; H, 2.91. 31P{1H} NMR (162 MHz; [D]chloroform): δ = 155.21 (s)
C
13), 122.50 (aq, |2JPC + 4JP′C| = 3.9 Hz, 4 C, Cpyrr–1α), 120.18 (aq, |2JPC
4
5
+ JP′C| = 4.3 Hz, 4 C, Cpyrr–2α), 112.50 (aq, |3JPC + JP′C| = 2.6 Hz, 4
C, Cpyrr–2ꢀ), 112.32 (aq, |3JPC
+
5JP′C| = 2.9 Hz, 4 C, Cpyrr–1ꢀ), 71.21
(aq,|2JPC + JP′C| = 3.3 Hz, CH2) ppm.
4
cis-[Mo(CO)4(L3)] (3): A mixture of 0.226 g (0.699 mmol) of L3 and
0.210 g (0.699 mmol) [Mo(CO)4nbd] in 10 mL of hexanes was stirred
for one hour at ambient temperature. A 31P{1H} NMR spectrum of
an aliquot of this mixture indicated that the reaction had reached
completion and that both the major product 3 (δ = 187.23 ppm,
93.5 %) and an unidentified minor product (δ = 169.35 ppm, 6.5 %)
were present. The precipitate was collected by filtration and dried
under vacuum yield 0.317 g (85.1 % yield) of crude 3. Thermal re-
crystallization from a 50:50 dichloromethane/hexanes solution re-
sulted in yellow blocky X-ray quality crystals. Anal. C22H14MoNO7P-
(0.2 CH3CN): C, 49.87: H, 2.73; found C, 49.82: H, 2.81. 31P{1H} NMR
(162 MHz; [D]chloroform): δ = 187.23 (s) ppm. 1H (700 MHz;
[D]chloroform): δ = 9.01 (d, |3JHH| = 5.0 Hz, 1 H, Hpy–α), 7.79 (td,
|3JHH| = 7.7 Hz, 1 H, Hpy–γ), 7.53 (dd, |3JHH| = 7.6 Hz, |4JHH| = 1.9 Hz,
2 H, H2, H11), 7.43 (td, |3JHH| = 7.9 Hz, |4JHH| = 1.6 Hz, 2 H, H3, H10),
7.39 (d, |3JHH| = 7.6 Hz, 1 H, Hpy–δ), 7.34 (t, |3JHH| = 7.5 Hz, 2 H, H4,
H9), 7.30 (d, |3JHH| = 22.5 Hz, 2 H, H5, H8), 7.23 (td, |3JHH| = 6.4 Hz,
|4JHH| = 1.5 Hz, 1 H, Hpy–ꢀ), 5.19 (d, |3JHH| = 22.5 Hz, 2 H, CH2) ppm.
13C{1H} NMR (176 MHz; [D]chloroform): δ = 216.93 (d, |2JPC| =
12.9 Hz, 1 C, CO trans to N), 212.32 (d, |2JPC| = 59.5 Hz, 1 C, CO trans
to P), 206.91 (d, |2JPC| = 13.6 Hz, 2 C, CO cis to N and P), 158.54 (d,
|4JPC| = 3.9 Hz, C15), 157.26 (s, Cpy–ipso), 148.56 (d, |2JPC| = 8.9 Hz, C1,
C12), 138.56 (s, Cpy–α), 130.31 (s, Cpy–γ), 130.03 (s, C2, C11), 129.50 (s,
C3, C10), 126.52 (s, Cpy–δ), 125.94 (s, C4, C9), 123.96 (s, Cpy–ꢀ), 122.29
(s, C5, C8), 68.98 (d, |2JPC| = 8.5 Hz, CH2) ppm.
1
ppm. H(700 MHz; [D]chloroform): δ = 8.98 (d, |3JHH| = 5.4 Hz, 1 H,
Hpy–α), 7.84 (ddd, |3JHH| = 7.6, |4JHH| = 1.1 Hz, 1 H, Hpy–γ), 7.46 (d,
|3JHH| = 7.6 Hz, 1 H, Hpy–δ), 7.26–7.28 (m, 1 H, Hpy–ꢀ), 7.01 (br. s, 4
H, Hpyrr–α), 6.34 (br. s, 4 H, Hpyrr–ꢀ), 5.12 (d, |3JPH| = 23.7 Hz, 2 H, CH2)
ppm. 13C{1H} NMR (176 MHz; [D]chloroform): δ = 217.88 (d, |2JPC| =
10.8 Hz, carbonyl trans to N), 211.71 (d, |2JPC| = 52.4 Hz, carbonyl
trans to P), 206.12 [d, |2JPC| = 11.9 Hz, carbonyl trans to carbonyl
(2C)], 158.40 (d, |3JPC| = 4.1 Hz, Cpy–α), 157.29 (s, Cpy–ipso), 138.90 (s,
Cpy–γ), 126.61 (s, C11), 124.33 (s, Cpy–ꢀ), 122.05 (d, |2JPC| = 9.0 Hz,
Cpyrr–α), 112.83 (d, |3JPC| = 6.6 Hz, Cpyrr–ꢀ), 70.06 (d, |2JPC| = 8.9 Hz,
CH2) ppm.
cis-[Mo(CO)4(L2)] (2a): A mixture of 0.240 g (0.552 mmol) of L2
and 0.451 g (0.545 mmol) of [Mo(CO)4nbd] in 15 mL of hexanes
was stirred at room temperature. After 10 min, the contents had
dissolved, and the solution had turned pale yellow. Immediately, a
white solid began to precipitate. After one hour, the off-white solid
was collected by filtration, rinsed twice with hexanes then dried
under high vacuum to yield 0.312 g (83.9 %) of the crude product.
The mother liquor was reduce in volume and filtered through Celite
to remove a brown precipitate. When the filtrate was cooled to 0 °C,
0.040 g (11 %) of white, X-ray quality crystals were obtained. Anal.
C27H23MoN5O6P2: C, 48.30; H, 3.45; found C, 48.36; H, 3.47. 31P{1H}
cis-[PdCl2(L1)] (4): A mixture of 0.26 mL (1.13 mmol) of L1 and
0.200 g (1.13 mmol) PdCl2 in 9 mL of a 1:2 mixture of THF and
acetonitrile was stirred vigorously for 24 h to yield a yellowish green
NMR (162 MHz; [D]chloroform): δ = 146.58 (s) ppm. 1H NMR solution and a pale yellow precipitate. The precipitate was collected
(700 MHz; [D]chloroform): δ = 7.50 (t, |3JHH| = 7.7 Hz, 1 H, Hpy–ꢀ),
6.94 (br. s, 8 H, Hpyrr–α), 6.88 (d, |3JHH| = 7.7 Hz, 2 H, Hpy–α), 6.24 (br.
s, 8 H, Hpyrr–ꢀ), 5.06 (at |3JPH| = 7.4, 4 H, CH2) ppm. 13C{1H}
by filtration, washed twice with 2 mL aliquots of acetonitrile and
dried under reduced pressure to yield 0.220 g (43.5 % yield) of 4 as
a very pale yellow powder. X-ray quality crystals were obtained by
slow diffusion of hexanes into a dichloromethane solution of 4.
Anal. C14H14Cl2N3OPPd: C, 37.48; H, 3.15; found C, 37.68; H, 3.32.
31P{1H} NMR (162 MHz; [D6]DMSO): δ = 85.40 (s) ppm. 1H (300 MHz;
NMR (176 MHz; [D]chloroform): δ = 210.38 (aq,|2JPC
+
2JP′C| =
32.7 Hz, carbonyl trans to P), 206.84 (at,|2JPC| = 12.4 Hz, carbonyl
trans to carbonyl), 153.63 (s, Cpy–ipso), 137.17 (s, Cpy–ꢀ), 122.91 (aq,
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
14
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[D6]DMSO): δ = 9.40 (d, |3JHH| = 5.8 Hz, 1 H, Hpy–α), 8.20 (ddd, |3JHH| =
H2, H11), 7.4 (d, |3JHH| = 7.0 Hz, H5, H8), 7.3 (t, |3JHH| = 4.9 Hz, H3, H4,
7.6, |4JHH| = 1.5 Hz, 1 H, Hpy–γ), 7.85 (d, |3JHH| = 7.6 Hz, 1 H, Hpy–δ), H9, H10), 7.2 (t, |3JHH| = 7.7 Hz, H3a, H10a), 7.1 (t, |3JHH| = 7.0 Hz, H4a
,
7.74–7.66 (m, 1 H, Hpy–ꢀ), 7.18 (br. s, 4 H, Hpyrr–α), 6.42 (br. s, 4 H, H9a), 7.0 (br. s, H2a, H11a), 6.9 (br. s, H5a, H8a), 5.6 (d, |3JHH| = 23.1 Hz,
Hpyrr–ꢀ), 5.79 (d, |3JPH| = 23.0 Hz, 2 H, CH2) ppm. 13C{1H} NMR
H13a, H13b) ppm. 13C{1H} NMR (176 MHz; [D6]DMSO): δ = 153.3 (d,
(176 MHz; [D6]DMSO): δ = 155.29 (s, Cpy–α), 151.57 (d, |3JPC| = 5.8 Hz,
|3JPC| = 4.8 Hz, Cpy–α), 152.7 (s, Cpy–ipso), 148.9 (s, C1, C12, C1a, C12a),
C
py–ipso), 141.67 (s, Cpy– γ), 126.89 (s, Cpy–δ), 126.22 (s, Cpy–ꢀ), 141.8 (s, Cpy–γ), 131.1 (s, C8, C15), 130.5 (s, C2a, C11a), 130.4 (s, C3a,
123.80 (d, |2JPC| = 8.1 Hz, 4 C, Cpyrr–α), 114.02 (d, |3JPC| = 9.0 Hz, 4 C, C10a), 129.8 (s, C2, C11), 129.4 (s, C3, C10), 128.1 (s, C6a, C7a), 126.8 (s,
Cpyrr–ꢀ), 72.58 (d, |2JPC| = 6.7 Hz, CH2) ppm.
C4a, C5a, C8a, C9a), 126.5 (s, Cpy–ꢀ), 126.2 (s, Cpy–δ), 125.6 (s, C4, C9),
122.5 (d, |3JPC| = 9.5 Hz, C7), 121.7 (d, |3JPC| = 9.5 Hz, C6), 68.1 (d,
|3JPC| = 5.1 Hz, CH2) ppm.
cis-[PdCl2(L2)] (5): A mixture of 83.8 mg (0.470 mmol) of PdCl2,
6 mL of acetonitrile and 4 mL of THF was stirred at room tempera-
ture overnight. Then, a solution of 0.200 g (0.460 mmol) of L2 in
5 mL of THF was added, and the resultant solution was stirred over-
night. The yellow precipitate that formed was collected by filtration,
washed with acetonitrile and placed under high vacuum to yield
0.208 g (73.7 %) of the crude product. Diffusion of hexanes into a
dichloromethane solution of the crude product yielded pale-yellow
X-ray quality crystals. Anal. Calc for (C23H23Cl2N5O2P2Pd + 0.3H2O):
C, 42.75; H, 3.68; found C, 42.02; H, 3.59. 31P{1H} NMR (162 MHz;
Procedure for the Rhodium Catalyzed Hydroformylation of
Styrene: The catalysis reactions were carried out using a Parr Series
4560 minireactor connected to a high-pressure gas burette that
introduced gas to the reactor at a constant pressure. The digitized
reactor temperature, burette temperature, and burette pressure
were monitored using Agilent Benchlink Data Logger software on
a PC connected to an Agilent data acquisition switch unit. In a
typical run, a solution of ligand, Rh(CO)2acac, and alkali metal salt,
if used, in 22 mL of dry THF were added to the reactor via needle
transfer under nitrogen pressure through the substrate inlet valve.
1
[D]chloroform): δ = 87.4 (s) ppm. H (700 MHz; [D]chloroform): δ =
7.86 (t, |3JHH| = 7.7 Hz, 1 H, Hpy–α), 7.16 (d, |3JHH| = 7.7 Hz, 2 H,
Hpy–ꢀ), 6.99 (s, 4 H, Hpyrr–α), 6.76 (s, 4 H, Hpyrr–α2), 6.24 (s, 8 H, The molar ratio of rhodium to phosphorus was varied for experi-
Hpyrr–ꢀ), 5.25 (m, |3JHH| = 14.7 Hz, 4 H, CH2) ppm. 13C{1H} NMR
ments to determine its effects on rate and selectivity. Next, the
(176 MHz; [D]chloroform): δ = 152.3 (s, C9, C14), 139.0 (s, Cpy–α), reactor was purged three times with a 1:1 H2/CO gas (syngas) mix-
124.2 (s, 4 C, Cpyrr–α), 123.3 (s, 4 C, Cpyrr–α2), 119.7 (s, Cpy–ꢀ), 113.3
(s, 4 C, Cpyrr–ꢀ), 112.9 (s, 4 C, Cpyrr–ꢀ2), 71.1 (d, |2JCP| = 2.5 Hz, CH2)
ppm.
ture and then pressurized to 20 atm with syngas and heated to
80 °C with mechanical stirring. The reactor was maintained under
these conditions for 45 min to allow for pre-catalyst equilibration
and then was cooled to 30 °C before the pressure was vented. Styr-
ene was then injected into the reactor through the substrate inlet
valve, after which the reactor was first pressurized to the desired
pressure and then heated to the desired pressure. The reaction
progress was monitored using the Agilent software. The data ob-
tained from pressure drop vs. time were fitted to an exponential
function using Graphical Analysis version 3.8.4 to determine the
pseudo-first-order rate constant.[120] The regioselectivities and %
cis-[PdCl2(L3)] (6): A mixture of 99.4 mg (0.561 mmol) PdCl2, 3 mL
of THF and 6 mL of acetonitrile was stirred at room temperature
for 24 h and then the resultant solution was added to 0.181 g
(0.560 mmol) of L3 together with 10 mL of THF. After stirring for
7 min, a white precipitate began to form. After 24 h, the precipitate
was collected, rinsed twice with dichloromethane and dried under
high vacuum for 12 h to yield 0.175 g (62.7 %) of the crude product
as a white powder. Slow diffusion of hexanes into a dichloro-
methane solution of the crude product yielded pale-yellow crystals,
X-ray quality crystals. Anal. C18H14Cl2NO3PPd: C, 43.19; H, 2.82;
found C, 42.33; H, 2.85. 31P{1H} NMR (162 MHz; [D6]DMSO): δ = 99.78
1
conversions of the catalytic reactions were determined by H NMR
after the reaction pressure had ceased changing. In the presence
of large amounts of Li+ and long reaction times (generally over-
night), an alkali metal catalyzed aldol condensation reaction of the
linear aldehyde was sometimes observed. If the proton resonance
of the aldol product was observed at δ = 9.57 ppm, its integrated
intensity was doubled and added to that of the resonance for the
linear aldehyde before the iso/n ratio was calculated.
1
(s) ppm. H (700 MHz; [D6]DMSO): δ = 9.39 (d, |3JHH| = 5.6 Hz, 1 H,
Hpy–α), 8.25 (t, |3JHH| = 7.4 Hz, 1 H, Hpy–γ), 7.86 (d, |3JHH| = 7 Hz, 1 H,
Hpy–δ), 7.76 (t, |3JHH| = 6.3 Hz, 1 H, Hpy–ꢀ), 7.66 (d, |3JHH| = 7.7 Hz,
H5, H8), 7.54 (t, |3JHH| = 7.7 Hz, H2, H11), 7.46 (d, |3JHH| = 9.1 Hz, H3, H4,
H9, H10), 5.74 (d, |3JHH| = 23.8 Hz, CH2) ppm. 13C{1H} NMR (176 MHz;
[D6]DMSO): δ = 155.82 (s, Cpy–α), 152.48 (d, |3JPC| = 4.7 Hz, Cpy–ipso),
148.53 (d, |2JPC| = 11.4 Hz, C1, C12), 142.23 (s, Cpy–γ), 131.04 (s,
C5, C8), 130.92 (s, C3, C10), 128.31 (s, C6, C7), 127.75 (s, C4, C9), 127.39
(s, Cpy–δ), 126.83 (s, Cpy–ꢀ), 122.68 (d, |3JPC| = 3.3 Hz, C2, C11), 70.10
(d, |2JPC| = 7.9 Hz, CH2) ppm.
Experimental Design: The effect of temperature(T) and pressure(P)
on the regioselectivities of RhI complexes of L1 in styrene hydro-
formylation was performed by using a full 22 factorial design follow-
ing standard methodology.[55,56] Two levels were studied for each
variable (T = 65 or 80 °C; P = 5 or 20 atm), resulting in four experi-
ments. The function defining the system is, % regio = a0 + aPP +
aTT + aPTPT; where % regio denotes the percentage of the n-alde-
hyde. From this, the main effects of each variable (aP and aT) and
an interaction term (aPT) are obtained. Mathematica was used to
solve the system of equations and generate response surfaces.[57]
[PdCl(L3){(C12H8O2)PO}] (7): A mixture of 0.200 g (0.614 mmol) of
L3, and 0.110 g (6.19 mmol) PdCl2 in 3 mL of THF and 6 mL of
acetonitrile was stirred for 24 h producing a yellowish green solu-
tion and a pale yellow precipitate. The solid precipitate was col-
lected by filtration, washed twice with 5 mL portions of acetonitrile
and dried under high vacuum to yield 0.179 g (83.6 %) of mixture
of pink and white powders. This mixture was washed with dichloro-
methane, filtered and the filtrate was evaporated to dryness. Recrys-
X-ray Data Collection and Solution: A suitable single crystal of 1–
7 was mounted on a MicroLoop (MiTeGen, Inc.) after coating in
Paratone-N oil (Hampton Research Corp.). Data were collected at
tallization via slow diffusion of hexanes into a dichloromethane so- the specified temperature with a Bruker CCD diffractometer (SMART
lution of the white residue yielded X-ray quality colorless crystals. APEX2) fitted with a low temperature device (Oxford Cryostream).
Anal. Calc. for (C33H22ClNO6P2Pd)·0.8H2O: C, 50.70; H, 3.35; found C,
50.77; H, 3.24. 31P{1H} NMR (162 MHz; [D6]DMSO): δ = 123.51 (d,
|2JPP| = 99.5 Hz), 65.32 (d, |2JPP| = 99.5 Hz) ppm. 1H (300 MHz;
[D6]DMSO): δ = 9.3 (t, |3JHH| = 3.5 Hz, 1 H, Hpy–α), 8.2 (t, |3JHH| =
7.7 Hz, 1 H, Hpy–γ), 7.8 (m, 2 H, Hpy–ꢀ, Hpy–δ), 7.5 (at |3JHH| = 4.2 Hz,
The diffractometer used graphite monochromated Cu-Kα (λ =
1.54178 Å) radiation with a detector distance of 5 cm. The program
SAINT was used to collect and reduce the data.[121] Unit cell con-
stants are based upon the refinement of the XYZ-centroids found
using a standard indexing routine (APEX2).[122] The details of the
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
15
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[20]
Y. Liu, C. A. Sandoval, Y. Yamaguchi, X. Zhang, Z. Wang, K. Kato, K. Ding,
J. Am. Chem. Soc. 2006, 128, 14212–14213.
S. P. Flanagan, P. J. Guiry, J. Organomet. Chem. 2006, 691, 2125–2154.
L. Lefort, J. A. Boogers, A. H. de Vries, J. G. de Vries, Org. Lett. 2004, 6,
1733–1735.
G. Morales Torres, S. Behrens, D. Michalik, D. Selent, A. Spannenberg,
S. Luehr, K. M. Dyballa, R. Franke, A. Börner, ChemistryOpen 2017, 6, 247–
253.
E. Benetskiy, S. Lühr, M. Vilches-Herrera, D. Selent, H. Jiao, L. Domke, K.
Dyballa, R. Franke, A. Börner, ACS Catal. 2014, 4, 2130–2136.
X. Zhang, B. Cao, Y. Yan, S. Yu, B. Ji, X. Zhang, Chem. Eur. J. 2010, 16,
871–877.
A. Gual, C. Godard, S. Castillón, C. Claver, Tetrahedron: Asymmetry 2010,
21, 1135–1146.
I. A. Tonks, R. D. Froese, C. R. Landis, ACS Catal. 2013, 3, 2905–2909.
E. Boymans, M. Janssen, C. Müller, M. Lutz, D. Vogt, Dalton Trans. 2013,
42, 137–142.
E. Cook, J. D. Masuda, A. Xia, Can. J. Chem. 2010, 88, 99–103.
C. G. Arena, D. Drommi, F. Faraone, Tetrahedron: Asymmetry 2000, 11,
2765–2779.
J. Mazuela, O. Pàmies, M. Diéguez, Adv. Synth. Catal. 2013, 355, 2569–
2583.
data collection are given in Table 1– Table 4. All data were corrected
for Lorentz and polarization effects, and were scaled using the nu-
merical method SADABS.[123] The structure was solved and refined
using the Bruker SHELXTL Software Package.[124] Heavy atoms were
found using either Direct Methods or Patterson, and the remainder
of the non-hydrogen atoms were located in difference Fourier maps.
Full-matrix least-squares refinement on F2 was performed on posi-
tional and anisotropic parameters for these atoms. Hydrogen atoms
were placed in calculated positions with the appropriate molecular
geometry δ (C–H = 0.96 Å). The isotropic thermal parameter associ-
ated with each hydrogen atom was fixed equal to 1.2 times the Ueq
of the atom to which it is bound. DELU and SIMU were used to help
model the disorder in 1.[125] The BASF TWIN command was used in
the case of the inversion twin 3 (BASF = 0.05148).
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
CCDC 1844555 (for 1), 1844556 (for 2a), 1844557 (for 2b), 1844558
(for 3), 1844559 (for 4), 1844560 (for 5), 1844561 (for 6), and
1844562 (for 7) contain the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre.
[29]
[30]
[31]
[32]
J. Mazuela, O. Pàmies, M. Diéguez, Chem. Eur. J. 2013, 19, 2416–2432.
Acknowledgments
[33] M. Lega, J. Margalef, F. Ruffo, O. Pàmies, M. Diéguez, Tetrahedron: Asym-
metry 2013, 24, 995–1000.
The authors would like to acknowledge National Science Foun-
dation (NSF) for support under grant EPS-0447675. S. D. H.
thanks GRSP for funding a graduate fellowship. The present
work benefitted from S. D. H. attending the American Crystallo-
graphic Association 2010 Small Molecule X-ray Crystallography
Summer Course, especially for modelling the disorder in 4.
[34]
K. A. Vallianatou, D. J. Frank, G. Antonopoulou, S. Georgakopoulos, E.
Siapi, M. Zervou, I. D. Kostas, Tetrahedron Lett. 2013, 54, 397–401.
C. Saluzzo, J. Breuzard, S. Pellet-Rostaing, M. Vallet, F. Le Guyader, M.
Lemaire, J. Organomet. Chem. 2002, 643, 98–104.
[35]
[36]
[37]
[38]
I. D. Kostas, Curr. Org. Synth. 2008, 5, 227–249.
G. Chelucci, G. Orrù, G. A. Pinna, Tetrahedron 2003, 59, 9471–9515.
K. Tani, M. Yabuta, S. Nakamura, T. Yamagata, J. Chem. Soc., Dalton Trans.
1993, 2781–2789.
[39]
[40]
H. Yang, N. Lugan, R. Mathieu, Organometallics 1997, 16, 2089–2095.
S. M. Aucott, A. M. Slawin, J. D. Woollins, J. Chem. Soc., Dalton Trans.
2000, 2559–2575.
Keywords: P ligands · Hydroformylation · Structure
elucidation · Electronic parameters · Rhodium catalysis
[41]
[42]
[43]
[44]
G. A. Ardizzoia, S. Brenna, S. Durini, B. Therrien, Organometallics 2012,
31, 5427–5437.
M. Rubio, A. Suárez, E. Vega, E. Álvarez, J. Díez, M. P. Gamasa, A. Pizzano,
Eur. J. Inorg. Chem. 2012, 2012, 655–663.
W. Liu, M. Yuan, H. Fu, H. Chen, R. Li, X. Li, Chem. Lett. 2009, 38, 596–
597.
S. C. van der Slot, J. Duran, J. Luten, P. C. J. Kamer, P. W. N. M.
van Leeuwen, Organometallics 2002, 21, 3873–3883.
K. G. Moloy, J. L. Petersen, J. Am. Chem. Soc. 1995, 117, 7696–7710.
a) R. L. Keiter, J. G. Verkade, Inorg. Chem. 1969, 8, 2115–2120; b) R. D.
Kroshefsky, R. Weiss, J. G. Verkade, Inorg. Chem. 1979, 18, 469–472; c)
S. O. Grim, P. R. McAllister, R. M. Singer, J. Chem. Soc., D 1969, 38b–39.
P. W. Dyer, J. Fawcett, M. J. Hanton, R. D. W. Kemmitt, R. Padda, N. Singh,
J. Chem. Soc., Dalton Trans. 2002, 104–113.
T. S. Barnard, M. R. Mason, Organometallics 2001, 20, 206–214.
C. E. Anderson, A. S. Batsanov, P. W. Dyer, J. Fawcett, J. A. Howard, Dalton
Trans. 2006, 5362–5378.
S. D. Hastings, H. Byrd, L. N. Gray, M. J. Jablonsky, J. L. Freeman, G. M.
Gray, Eur. J. Inorg. Chem. 2013, 2013, 2900–2911 and references cited
therein.
M. P. Anderson, A. L. Casalnuovo, B. J. Johnson, B. M. Mattson, A. Mueting,
L. H. Pignolet, Inorg. Chem. 1988, 27, 1649–1658.
P. E. Garrou, Inorg. Chem. 1975, 14, 1435–1439.
P. Maji, W. Wang, A. E. Greene, Y. Gimbert, J. Organomet. Chem. 2008,
693, 1841–1849.
C. C. Yang, W. Y. Yeh, G. H. Lee, S. M. Peng, J. Organomet. Chem. 2000,
598, 353–358.
D. Benito-Garagorri, E. Becker, J. Wiedermann, W. Lackner, M. Pollak, K.
Mereiter, J. Kisala, K. Kirchner, Organometallics 2006, 25, 1900–1913.
T. H. Coffield, R. P. Werner, (Ethyl Co., USA), US patent 3322799 A, May
30, 1967.
S. Song, E. C. Alyea, Can. J. Chem. 1996, 74, 2304–2320.
J. R. Graham, R. J. Angelici, Inorg. Chem. 1967, 6, 2082–2085.
[1] R. Franke, D. Selent, A. Börner, Chem. Rev. 2012, 112, 5675–5732.
[2] E. Rafter, D. G. Gilheany, J. N. H. Reek, P. W. N. M. van Leeuwen, Chem-
CatChem 2010, 2, 387–391.
[3] P. W. N. M. van Leeuwen, P. C. Kamer, C. Claver, O. Pàmies, M. Diéguez,
Chem. Rev. 2011, 111, 2077–118.
[4] A. Ficks, R. M. Hiney, R. W. Harrington, D. G. Gilheany, L. J. Higham, Dalton
Trans. 2012, 41, 3515–22.
[5] B. Zhao, X. Peng, Z. Wang, C. Xia, K. Ding, Chem. Eur. J. 2008, 14, 7847–
57.
[6] D. W. Norman, C. A. Carraz, D. J. Hyett, P. G. Pringle, J. B. Sweeney, A. G.
Orpen, H. Phetmung, R. L. Wingad, J. Am. Chem. Soc. 2008, 130, 6840–
6847.
[7] I. D. Gridnev, C. Fan, P. G. Pringle, Chem. Commun. 2007, 1319–1321.
[8] M. T. Reetz, Russ. J. Org. Chem. 2003, 39, 392–396.
[9] C. Claver, E. Fernandez, A. Gillon, K. Heslop, D. J. Hyett, A. Martorell, A. G.
Orpen, P. G. Pringle, Chem. Commun. 2000, 961–962.
[10] M. Gouygou, J. C. Daran, E. Robé, C. Ortéga, C. R. Chim. 2010, 13, 1054–
1062.
[11] W. Peng, J. M. Shreeve, J. Fluorine Chem. 2005, 126, 1054–1056.
[12] B. Goldfuss, T. Löschmann, F. Rominger, Chem. Eur. J. 2004, 10, 5422–
5431.
[13] P. Bergamini, V. Bertolasi, M. Cattabriga, V. Ferretti, U. Loprieno, N. Manto-
vani, L. Marvelli, Eur. J. Inorg. Chem. 2003, 2003, 918–925.
[14] F. Rahm, A. Fischer, C. Moberg, Eur. J. Org. Chem. 2003, 2003, 4205–4215.
[15] J. F. Teichert, B. L. Feringa, Angew. Chem. Int. Ed. 2010, 49, 2486–528;
Angew. Chem. 2010, 122, 2538.
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[16] P. Shejwalkar, N. P. Rath, E. B. Bauer, Molecules 2010, 15, 2631–2650.
[17] L. Eberhardt, D. Armspach, J. Harrowfield, D. Matt, Chem. Soc. Rev. 2008,
37, 839–64.
[18] B. H. Swennenhuis, R. Chen, P. W. N. M. van Leeuwen, J. G. de Vries, P. C.
Kamer, Org. Lett. 2008, 10, 989–992.
[57]
[58]
[19] M. T. Reetz, G. Mehler, O. Bondarev, Chem. Commun. 2006, 2292–2294.
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
16
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[59]
[92] I. del Río, O. Pàmies, P. W. van Leeuwen, C. J. Claver, J. Org. Chem. 2000,
608, 115–121.
O. Herd, D. Hoff, K. W. Kottsieper, C. Liek, K. Wenz, O. Stelzer, W. S. Shel-
drick, Inorg. Chem. 2002, 41, 5034–5042.
[93] R. Abdallah, J. A. Breuzard, M. C. Bonnet, M. Lemaire, J. Mol. Catal. A
2006, 249, 218–222.
[94] R. M. Carrilho, C. B. Neves, M. A. Lourenço, A. R. Abreu, M. T. Rosado, P. E.
Abreu, M. E. S. Eusébio, L. Kollár, J. C. Bayón, M. M. Pereira, J. Organomet.
Chem. 2012, 698, 28–34.
[95] A. L. Watkins, C. R. Landis, J. Am. Chem. Soc. 2010, 132, 10306–10317.
[96] E. R. Nelsen, C. R. Landis, J. Am. Chem. Soc. 2013, 135, 9636–9639.
[97] M. Beller, S. Gladiali, K. Junge, S. Enthaler, G. Erre, J. Mol. Catal. A 2008,
280, 148–155.
[98] P. E. Goudriaan, M. Kuil, X. B. Jiang, P. W. van Leeuwen, J. N. Reek, Dalton
Trans. 2009, 1801–1805.
[60]
[61]
K. Fukumoto, H. Nakazawa, J. Organomet. Chem. 2008, 693, 1968–1974.
L. Barloy, G. Malaisé, S. Ramdeehul, C. Newton, J. A. Osborn, N. Kyritsakas,
Inorg. Chem. 2003, 42, 2902–2907.
W. J. Fleming, H. Müller-Bunz, V. Lillo, E. Fernández, P. J. Guiry, Org.
Biomol. Chem. 2009, 7, 2520–2524.
[62]
[63]
A. Friedrich, M. Drees, M. Käss, R. Herdtweck, S. Schneider, Inorg. Chem.
2010, 49, 5482–5494.
[64]
[65]
[66]
[67]
C. Gunanathan, D. Milstein, Acc. Chem. Res. 2011, 44, 588–602.
J. P. Farr, F. E. Wood, A. L. Balch, Inorg. Chem. 1983, 22, 3387–3393.
A. Heßler, J. Fischer, S. Kucken, O. Stelzer, Chem. Ber. 1994, 127, 481–488.
D. Vuzman, E. Poverenov, L. J. Shimon, Y. Diskin-Posner, D. Milstein, Orga-
nometallics 2008, 27, 2627–2634.
[99] S. Ley, S. Kasinathan, S. Bourne, P. Tolstoy, P. Koos, M. O'Brien, R. Bates, I.
Baxendale, Synlett 2011, 2011, 2648–2651.
[68]
[69]
[70]
[71]
D. Benito-Garagorri, K. Mereiter, K. Kirchner, Eur. J. Inorg. Chem. 2006,
2006, 4374–4379.
J. I. van der Vlugt, J. N. Reek, Angew. Chem. Int. Ed. 2009, 48, 8832–8846;
Angew. Chem. 2009, 121, 8990.
C. Müller, E. A. Pidko, M. Lutz, A. L. Spek, D. Vogt, Chem. Eur. J. 2008, 14,
8803–8807.
A. F. Peixoto, A. R. Abreu, A. R. Almeida, A. Neves, P. E. Abreu, M. J. S. M.
Moreno, J. C. Bayon, A. A. C. C. Pais, M. M. Pereira, Curr. Org. Synth. 2014,
11, 301–309.
A. A. Kaisare, S. B. Owens, E. J. Valente, G. M. Gray, J. Organomet. Chem.
2010, 695, 1472–1479.
S. B. Owens Jr., G. M. Gray, J. Organomet. Chem. 2008, 693, 3907–3914.
A. A. Kaisare, S. B. Owens, E. J. Valente, G. M. Gray, J. Organomet. Chem.
2010, 695, 2658–2666.
[100] A. F. Peixoto, M. M. Pereira, A. A. C. C. Pais, J. Mol. Catal. A 2007, 267,
234–240.
[101] J. T. Sheff, PhD thesis, University of Alabama at Birmingham (USA),
2012.
[102] S. B. Owens, G. M. Gray, Organometallics 2008, 27, 4282–4287.
[103] P. C. Kamer, J. N. Reek, P. W. van Leeuwen, Mech. Homogeneous Catal.
2005, 231–269.
[104] P. E. Goudriaan, P. W. N. M. van Leeuwen, M. Birkholz, J. N. H. Reek, Eur.
J. Inorg. Chem. 2008, 2008, 2939–2958.
[105] S. R. Khan, B. M. Bhanage, Appl. Organomet. Chem. 2013, 27, 313–317.
[106] C. Chen, Y. Qiao, H. Geng, X. Zhang, Org. Lett. 2013, 15, 1048–1051.
[107] S. Yu, Y. M. Chie, Z. H. Guan, Y. Zou, W. Li, X. Zhang, Org. Lett. 2009, 11,
241–244.
[108] R. Jackstell, H. Klein, M. Beller, K. D. Wiese, D. Roettger, Eur. J. Org. Chem.
2001, 2001, 3871–3877.
[109] B. Breit, J. Mol. Catal. A 1998, 143, 143–154.
[110] A. A. Kaisare, PhD thesis, University of at Birmingham (USA), 2010.
[111] S. B. Owens Jr., PhD thesis, University of Alabama at Birmingham (USA),
2009.
[112] M. Kranenburg, Y. E. M van der Burgt, P. C. J. Kamer, P. W. N. M.
van Leeuwen, K. Goubitz, J. Fraanje, Organometallics 1995, 14, 3081–
3089.
[113] T. Horiuchi, E. Shirakawa, K. Nozaki, H. Takaya, Organometallics 1997,
16, 2981–2986.
[114] C. Y. Zheng, M. Mo, H. R. Liang, X. L. Zheng, H. Y. Fu, M. L. Yuan, R. X.
Li, H. Chen, Appl. Organomet. Chem. 2013, 27, 474–478.
[115] R. Lazzaroni, A. Raffaelli, R. Settambolo, S. Bertozzi, G. Vitulli, J. Mol.
Catal. 1989, 50, 1–9.
[72]
[73]
[74]
[75] J. T. Sheff, A. L. Lucius, S. B. Owens, G. M. Gray, Organometallics 2011, 30,
5695–5709.
[76] M. N. Burnett, C. K. Johnson, Report ORNL-6895, Oak Ridge National Labo-
ratory, Oak Ridge, TN, USA. 1996.
[77] H. Byrd, J. D. Harden, J. M. Butler, M. J. Jablonsky, G. M. Gray, Organo-
metallics 2003, 22, 4198–4205.
[78] M. B. Murphy-Jolly, S. B. Owens Jr., J. L. Freeman, G. M. Gray, C. M. Law-
son, D. P. Shelton, Eur. J. Inorg. Chem. 2010, 2010, 5263–5271.
[79] I. J. Bruno, J. C. Cole, M. Kessler, J. Luo, W. D. Motherwell, L. H. Purkis,
B. R. Smith, R. Taylor, R. I. Cooper, S. E. Harris, A. G. Orpen, J. Chem. Inf.
Comput. Sci. 2004, 44, 2133–2144.
[80] A. Pidcock, R. E. Richards, L. M. Venanzi, J. Chem. Soc. A 1966, 1707–
1710.
[81] P. N. Horton, S. K. Mohamed, A. M. Soliman, E. M. Abdel-Raheem, M.
Akkurt, Acta Crystallogr., Sect. E: Struct. Rep. Online 2012, 68, o938–o938.
[82] C. A. Tolman, Chem. Rev. 1977, 77, 313–348.
[83] T. E. Müller, D. M. P. Mingos, Trans. Met. Chem. 1995, 20, 533–539.
[84] C. P. Casey, G. T. Whiteker, M. G. Melville, L. M. Petrovich, J. A. Gavney Jr.,
D. R. Powell, J. Am. Chem. Soc. 1992, 114, 5535–5543.
[85] L. A. van der Veen, M. D. Boele, F. R. Bregman, P. C. Kamer, P. W. van Lee-
uwen, K. Goubitz, J. Fraanje, H. Schenk, C. J. Bo, J. Am. Chem. Soc. 1998,
120, 11616–11626.
[116] R. G. Brereton, Chemometrics: Data Analysis for the Laboratory and
Chemical Plant, John Wiley & Sons; 2003.
[117] J. N. Miller, J. C. Miller, Statistics and Chemometrics for Analytical Chemis-
try, 5, illustrated; Pearson; 2005.
[118] Mathematica 8. 2010.
[119] L. V. Verizhnikov, P. A. Kirpichnikov, Zh. Obshch. Khim. 1967, 37, 1355–
1358.
[120] Graphical Analysis 3. 2014.
[121] Bruker SAINT+, Version V8.34A. 2013.
[86] P. W. van Leeuwen, P. C. Kamer, J. N. Reek, P. Dierkes, Chem. Rev. 2000,
100, 2741–2770.
[122] Bruker APEX2, Version 2013.10–0. 2013.
[123] Bruker SADABS, Version 2012/1. 2012.
[124] G. M. Sheldrick, SHELXTL, V. Bruker Analytical X-ray Systems Inc., Madi-
[87] Z. Freixa, P. W. N. M. Van Leeuwen, Dalton Trans. 2003, 1890–1901.
[88] M. Hariharasarma, C. H. Lake, C. L. Watkins, G. M. Gray, Organometallics
1999, 18, 2593–2600.
[89] M. Hariharasarma, C. L. Watkins, G. M. Gray, Organometallics 2000, 19,
1232–1238.
son. 2001.
[125] P. Müller, R. Herbst-Irmer, A. L. Spek, T. R. Schneider, M. R. Sawaya (Ed.:
P. Müller), Crystal Structure Refinement: A Crystallographer's Guide to
SHELXL, illustrated, reprint; Oxford University Press: Oxford; New York;
2006.
[90] I. D. Kostas, K. A. Vallianatou, J. Holz, A. Börner, Appl. Organomet. Chem.
2005, 19, 1090–1095.
[91] Y. Yan, X. Zhang, J. Am. Chem. Soc. 2006, 128, 7198–7202.
Received: June 8, 2018
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
17
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Hydroformylation Catalysts
S. D. Hastings, E. C. Cagle,
T. R. Totsch, S. D. Tyus,
G. M. Gray* ....................................... 1–18
Comparative Study of Novel Phos-
phordiamidite and Phosphite Li-
gands Used in Alkene Hydroformyl-
ation; Synthesis, Characterization,
Metalation, and Catalytic Evaluation
Three novel phosphordiamidite and gands have been prepared to study
phosphite ligands have been synthe- coordination preferences with octa-
sized and characterized. Evaluations of hedral and square planar centers and
the σ-donating abilities of the ligands these have been characterized by
indicate that the phosphite ligand is multinuclear NMR and X-ray crystallog-
more basic than the phosphoramidite raphy.
analogs. Model complexes of the li-
DOI: 10.1002/ejic.201800724
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
18
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim