Half-Sandwich Guanidinate-Osmium(II) Complexes: Synthesis and Application in the Selective Dehydration of Aldoximes

Article abstract of DOI:10.1002/ejic.201501221

The novel guanidinate-osmium(II) complexes [OsCl{κ²-(N,N′)-C(NR)(NiPr)NHiPr}(η⁶-p-cymene)] [R = Ph (3a), 4-C₆H₄F (3b), 4-C₆H₄Cl (3c), 4-C₆H₄CF₃ (3d), 3-C₆H₄CF₃ (3e), 3,5-C₆H₃(CF₃)₂ (3f), 4-C₆H₄CN (3g), 4-C₆H₄Me (3h), 3-C₆H₄Me (3i), 2-C₆H₄Me (3j), 4-C₆H₄tBu (3k), 2,6-C₆H₃iPr₂ (3l), 2,4,6-C₆H₂Me₃ (3m)] have been synthesized in high yields (70-88 %) by treatment of THF solutions of the dimeric precursor [{OsCl(μ-Cl)(η⁶-p-cymene)}₂] (1) with 4 equivalents of the corresponding guanidine (iPrHN)₂C=NR (2a-m) at room temperature. The easily separable guanidinium chloride salts [(iPrHN)₂C(NHR)]Cl (4a-m) were also formed in these reactions. The structures of 3a, 3d, and 3h were unequivocally confirmed by X-ray diffraction methods. Complexes 3a-m proved to be active in the catalytic dehydration of aldoximes. The best results were obtained with [OsCl{κ²-(N,N′)-C(N-4-C₆H₄CF₃)(NiPr)NHiPr}(η⁶-p-cymene)] (3d; 5 mol-%), which, in acetonitrile at 80C, was able to convert selectively a large variety of aromatic, heteroaromatic, α,β-unsaturated, and aliphatic aldoximes into the corresponding nitriles in high yields and short reaction times. Novel osmium(II)-guanidinate complexes of general composition [OsCl{κ²-(N,N′)-C(NR)(NiPr)NHiPr}(η⁶-p-cymene)] (R = Ar) have been synthesized and successfully employed as catalysts for the selective conversion of aldoximes into nitriles.

Full text of DOI:10.1002/ejic.201501221

DOI: 10.1002/ejic.201501221
Full Paper
Half-Sandwich Guanidinates
Half-Sandwich Guanidinate–Osmium(II) Complexes: Synthesis
and Application in the Selective Dehydration of Aldoximes
Javier Francos,[a] Pedro J. González-Liste,[a] Lucía Menéndez-Rodríguez,[a] Pascale Crochet,[a]
Victorio Cadierno,*[a] Javier Borge,[b] Antonio Antiñolo,*[c] Rafael Fernández-Galán,[c] and
Fernando Carrillo-Hermosilla[c]
Abstract: The novel guanidinate–osmium(II) complexes were also formed in these reactions. The structures of 3a, 3d,
[OsCl{κ2-(N,N′)-C(NR)(NiPr)NHiPr}(η6-p-cymene)] [R = Ph (3a), 4- and 3h were unequivocally confirmed by X-ray diffraction
C6H4F (3b), 4-C6H4Cl (3c), 4-C6H4CF3 (3d), 3-C6H4CF3 (3e), 3,5- methods. Complexes 3a–m proved to be active in the catalytic
C6H3(CF3)2 (3f), 4-C6H4CN (3g), 4-C6H4Me (3h), 3-C6H4Me (3i), dehydration of aldoximes. The best results were obtained with
2-C6H4Me (3j), 4-C6H4tBu (3k), 2,6-C6H3iPr2 (3l), 2,4,6-C6H2Me3 [OsCl{κ2-(N,N′)-C(N-4-C6H4CF3)(NiPr)NHiPr}(η6-p-cymene)] (3d;
(3m)] have been synthesized in high yields (70–88 %) by treat- 5 mol-%), which, in acetonitrile at 80 °C, was able to convert
ment of THF solutions of the dimeric precursor [{OsCl(μ-Cl)(η6- selectively a large variety of aromatic, heteroaromatic, α,ꢀ-un-
p-cymene)}2] (1) with 4 equivalents of the corresponding guan- saturated, and aliphatic aldoximes into the corresponding
idine (iPrHN)2C=NR (2a–m) at room temperature. The easily nitriles in high yields and short reaction times.
separable guanidinium chloride salts [(iPrHN)2C(NHR)]Cl (4a–m)
Introduction
to be very efficient catalysts for the base-free redox isomeriza-
tion of allylic alcohols [turnover frequency (TOF) up to
1200 h–1].[4] These complexes were easily obtained from the
reactions of the dimeric precursors [{RuCl(μ-Cl)(η6-p-cymene)}2]
and [{RuCl(μ-Cl)(η3:η3-C10H16)}2] (C10H16 = 2,7-dimethylocta-2,6-
diene-1,8-diyl), respectively, with an excess of the correspond-
ing guanidine (iPrHN)2C=NR.[4]
Since the seminal work by Lappert and co-workers in 1970,[1]
a
large number of metal complexes containing guanidinate li-
gands have been described in the literature.[2] Notably, some of
them have found utility in catalysis (e.g., polymerization of ole-
fins, hydroamination of alkynes), as well as in materials science
as precursors for chemical vapor deposition (CVD) and atom
layer deposition (ALD) applications.[2] The easy generation of
guanidinate mono- and dianions from readily available guan-
idines,[3] along with their high steric and electronic modulabil-
ity, have allowed the coordination of these ligands to virtually
all transition metals.[2] In this context, we recently reported the
preparation of a series of ruthenium(II)– and ruthenium(IV)–
guanidinate derivatives (see A and B in Figure 1), which proved
Figure 1. Structures of the ruthenium–guanidinate complexes A and B.
[a] Laboratorio de Compuestos Organometálicos y Catálisis (Unidad Asociada
al CSIC), Centro de Innovación en Química Avanzada (ORFEO-CINQA),
Departamento de Química Orgánica e Inorgánica, Instituto Universitario
de Química Organometálica “Enrique Moles“, Universidad de Oviedo,
Julián Clavería 8, 33006 Oviedo, Spain
In addition to A and B, the synthesis of a significant number
of other mono- and dinuclear ruthenium–guanidinate com-
plexes have been reported,[5] and several iron representatives
are known.[6] In marked contrast, within this group of the peri-
odic table, little attention has been devoted to the chemistry
of osmium compounds with this class of ligands. In fact, to the
best of our knowledge, only four examples have been de-
scribed so far in the literature (see Figure 2): 1) The mononu-
clear complexes C and D, containing a mono- and dianionic
guanidinate ligand, respectively,[5c,5f] and 2) the dinuclear pad-
dlewheel-type species E and F, in which anions of the bicyclic
guanidine 1,3,4,6,7,8-hexahydro-2H-pyrimido[1,2-a]pyrimidine
E-mail: vcm@uniovi.es
http://www.unioviedo.es/comorca
[b] Departamento de Química Física y Analítica, Centro de Innovación en
Química Avanzada (ORFEO-CINQA), Universidad de Oviedo,
Julián Clavería 8, 33006 Oviedo, Spain
[c] Centro de Innovación en Química Avanzada (ORFEO-CINQA), Depar-
tamento de Química Inorgánica, Orgánica y Bioquímica, Facultad de
Ciencias y Tecnologías Químicas - Campus de Ciudad Real, Universidad de
Castilla La Mancha,
Campus Universitario, 13071 Ciudad Real, Spain
E-mail: Antonio.Antinolo@uclm.es
https://www.uclm.es/profesorado/afantinolo/
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/ejic.201501221.
n+
act as bridges in Os2 cores (n = 6, 7).[7] Worthy of note, none
of them has found applications in homogeneous catalysis.
Eur. J. Inorg. Chem. 2016, 393–402
393
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 2. Structures of the known osmium–guanidinate complexes C–F.
Scheme 1. Synthesis of the novel osmium–guanidinate complexes 3a–m.
With these precedents in mind, and as a continuation of our
previous studies with ruthenium, we considered it of interest to
prepare new examples of osmium–guanidinate complexes and
explore their catalytic potential. As a result of this, we present
herein the high-yielding synthesis of a family of half-sandwich
osmium(II)–guanidinate complexes structurally related to A,
that is, compounds [OsCl{κ2-(N,N′)-C(NR)(NiPr)NHiPr}(η6-p-cym-
ene)] (R = Ar), and their successful application in the catalytic
dehydration of aldoximes to generate nitriles.
1
δH = 2.92–3.27 ppm for the NH proton in the H NMR spectra,
and 3) a downfield singlet for the central CN3 carbon in the
13C{1H} NMR spectra at δC = 164.5–170.2 ppm. In accord with
the stereogenic character of the osmium atom in 3a–m, the
NMR spectra also show four differentiated signals for the aro-
matic CH protons and carbon atoms of the η6-coordinated cy-
mene ring, along with two separated resonances for the methyl
groups of its iPr unit. In addition, as previously observed in the
analogous ruthenium(II) complexes [RuCl{κ2-(N,N′)-C(NR)(NiPr)-
NHiPr}(η6-p-cymene)] (R = 2,6-C6H3iPr2, 2,4,6-C6H2Me3),[4a] for
complexes 3l,m, the rotation of the N-aryl units is restricted in
solution, as clearly evidenced by the chemical inequivalence in
Results and Discussion
Following the same synthetic procedure employed in the prep- the NMR spectra of the Me and iPr substituents located at ortho
aration of the ruthenium compounds A and B (Figure 1),[4] the positions of the aromatic rings of the guanidinate ligands.
novel osmium(II)–guanidinate complexes [OsCl{κ2-(N,N′)-
C(NR)(NiPr)NHiPr}(η6-p-cymene)] [R = Ph (3a), 4-C6H4F (3b), 4-
C6H4Cl (3c), 4-C6H4CF3 (3d), 3-C6H4CF3 (3e), 3,5-C6H3(CF3)2 (3f),
4-C6H4CN (3g), 4-C6H4Me (3h), 3-C6H4Me (3i), 2-C6H4Me (3j), 4-
C6H4tBu (3k), 2,6-C6H3iPr2 (3l), 2,4,6-C6H2Me3 (3m)] were easily
generated by treatment, at room temperature, of THF solutions
of the dimeric precursor [{OsCl(μ-Cl)(η6-p-cymene)}2] with
4 equivalents of the appropriate guanidine (iPrHN)2C=NR (2a–
m; Scheme 1). In these reactions, the corresponding guan-
idinium chloride salts [(iPrHN)2C(NHR)][Cl] (4a–m) were also
formed. Extraction of the crude reaction mixtures with pentane
allowed the separation of complexes 3a–m from these salts and
their isolation in pure form (70–88 % yield) after crystallization
(see details in the Exp. Sect.).
Compounds [OsCl{κ2-(N,N′)-C(NR)(NiPr)NHiPr}(η6-p-cymene)]
[R = Ph (3a), 4-C6H4CF3 (3d), 4-C6H4Me (3h)] were further char-
acterized by single-crystal X-ray diffraction analysis. Diffraction-
quality crystals were obtained in all cases by cooling a saturated
solution of the complex in benzene. ORTEP-type views of the
three structures are shown in Figure 3, and selected bond pa-
rameters are collected in Table 1.[8] We would like to note at
this point that these are the first solid-state structures of mono-
nuclear osmium–guanidinate complexes. Only the dinuclear
complexes E and F (Figure 2), with which few comparisons can
be made, have been previously studied by X-ray diffraction.[7]
As shown in Figure 3, the three molecules exhibit the expected
pseudo-octahedral three-legged piano-stool geometry with the
p-cymene ligand in the usual η6 coordination mode. Concern-
The air-stable complexes 3a–m were characterized by ele- ing the coordination of the guanidinate anions to osmium, de-
mental analysis, IR, and multinuclear NMR (1H, 13C{1H} and spite the different electronic nature of the N-aryl groups present
19F{1H}) spectroscopy, the data obtained being fully consistent in these complexes, no significant structural differences are
with the structural proposals (see the Exp. Sect. for details). The found (see Table 1). The Os–N1 and Os–N2 bond lengths, in the
key spectroscopic features are as follows: 1) A characteristic range 2.097(4)–2.138(4) Å, are longer than those in the crystal
ν(N–H) absorption band in the IR spectra in the 3315– structures of the dinuclear species E and F (Os–N distances of
3358 cm–1 region, 2) a doublet signal (3JH,H = 10.2–11.1 Hz) at ca. 2.040 Å).[7,9] The sum of the angles around the central car-
Eur. J. Inorg. Chem. 2016, 393–402
www.eurjic.org
394
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 3. ORTEP-type views of the structures of the osmium(II)–guanidinate complexes 3a (left), 3d (center), and 3h (right) showing the crystallographic
labeling scheme. Hydrogen atoms, except those on the N3 atoms, have been omitted for clarity. Thermal ellipsoids are drawn at the 30 % probability level.
bon atom of the CN3 skeletons (360°) indicates that the guanidi-
nate anions are planar, with the resonance form G making an
important contribution to their bonding (see Figure 4). This
bonding description is supported by the shorter C11–N2 bond
lengths [1.31(1)–1.317(6) Å] in comparison with those of C11–
N1 and C11–N3 [1.345(6)–1.37(1) Å and 1.36(1)–1.376(6) Å, re-
Figure 4. Limiting resonance forms of the guanidinate ligands coordinated to
spectively].
the [OsCl(η6-p-cymene)]+ fragment.
Table 1. Selected bond lengths and angles for complexes 3a, 3d, and 3h.
3a
3d
3h
tially focused on this catalytic transformation. It should be men-
tioned at this point that, although the osmium complexes have
been much less studied than the ruthenium complexes,[10] the
recent work of Esteruelas[11] and Baratta[12] and their co-workers
has demonstrated that efficient catalysts can also be obtained
from this metal (TOF up to 460 h–1). However, the catalytic ac-
tivities observed for 3a–m in the isomerization of the model
substrate 1-octen-3-ol were very disappointing. Thus, perform-
ing the reactions in THF at 80 °C with 1 mol-% of these com-
plexes, a maximum conversion of 56 % was achieved with
[OsCl{κ2-(N,N′)-C(N-4-C6H4tBu)(NiPr)NHiPr}(η6-p-cymene)] (3k)
after 24 h. The addition of a base to the medium (2 mol-% of
Cs2CO3) led to a slight improvement, but an incomplete reac-
tion was again observed after 24 h (78 % GC yield of octan-3-
one; TOF = 3 h–1).
Bond lengths [Å]
Os1–Cl1
Os1–N1
Os1–N2
Os1–C*[a]
C11–N1
C11–N2
C11–N3
N1–C18
N2–C12
N3–C15
2.407(2)
2.108(4)
2.138(4)
1.668(1)
1.345(6)
1.316(7)
1.376(6)
1.393(6)
1.448(7)
1.486(6)
2.416(2)
2.109(7)
2.119(7)
1.665(1)
1.37(1)
1.31(1)
1.36(1)
1.38(1)
1.46(1)
1.47(1)
2.411(1)
2.097(4)
2.130(4)
1.669(1)
1.353(6)
1.317(6)
1.364(6)
1.404(6)
1.459(6)
1.459(6)
Bond angles [°]
C*–Os1–Cl1[a]
C*–Os1–N1[a]
C*–Os1–N2[a]
Cl1–Os1–N1
Cl1–Os1–N2
N1–Os1–N2
Os1–N1–C11
Os1–N1–C18
Os1–N2–C11
Os1–N2–C12
N1–C11–N2
N1–C11–N3
N2–C11–N3
129.2(1)
134.6(1)
136.2(1)
83.5(1)
129.9(1)
137.1(1)
135.7(1)
84.0(2)
128.9(1)
134.7(1)
136.4(1)
85.2(2)
The low activities shown by 3a–m in the isomerization of 1-
octen-3-ol led us to explore other catalytic reactions, and, fortu-
nately, we found that these complexes are useful for promoting
the selective dehydration of aldoximes. This dehydration proc-
ess represents a convenient and benign method for the synthe-
sis of nitriles, because it avoids the use of the toxic cyanide
sources commonly employed in the preparation of this impor-
tant class of compounds (note that aldoximes are readily acces-
sible from inexpensive aldehydes by condensation with hydrox-
ylamine). Although the dehydration of aldoximes has been ex-
86.1(1)
61.3(2)
94.7(3)
134.6(3)
94.3(3)
137.1(3)
109.0(4)
125.5(5)
125.5(5)
83.3(2)
61.7(3)
93.9(5)
129.5(6)
95.5(5)
137.3(6)
108.1(7)
124.8(7)
127.1(7)
86.1(2)
61.2(2)
95.7(3)
131.6(3)
95.4(3)
137.4(3)
107.5(4)
125.3(4)
127.2(4)
[a] C* denotes the centroid of the p-cymene ring (C1–C2–C3–C4–C5–C6).
Having characterized the complexes 3a–m, we next explored tensively studied and applied in synthetic organic chemistry for
their catalytic potential. In particular, given the good results a long time, conventional protocols involve the use of stoichio-
obtained with the ruthenium–guanidinate derivatives A and B metric amounts of reagents and suffer from limitations arising
(Figure 1) in the redox isomerization of allylic alcohols, we ini- from the sensitivity of some functional groups to the reaction
Eur. J. Inorg. Chem. 2016, 393–402
www.eurjic.org
395
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
conditions.[13] More appealing protocols have emerged in re- gardless of their substitution pattern and electronic nature. As
cent years employing metal catalysts, which usually operate un- in the precedent case, the formation of only trace amounts of
der milder reaction conditions and with higher functional group the respective benzamides (<2 %) was observed by GC in these
compatibility. In this context, efficient catalytic systems involv- reactions. Heteroaromatic (entries 14–15), aliphatic (entries 16–
ing Re,[14] Fe,[15] Ru,[16] Co,[17] Rh,[18] Ni,[19] Pd,[20] Cu,[21] Zn,[22] 19), and α,ꢀ-unsaturated (entry 20) aldoximes were also suc-
Ga,[23] and In[24] have been reported. To the best of our knowl- cessfully dehydrated, thus confirming the generality of this cat-
edge, complexes 3a–m are the first examples of osmium com- alytic transformation. Again, they delivered the desired nitriles
pounds able to catalyze the dehydration of aldoximes.
in high yields (99 % by GC; ≥81 % after chromatographic purifi-
cation) and short reaction times (1–5 h). Notably, the chiral cen-
ter of (S)-citronellaldoxime remained unaffected by the dehy-
dration reaction (entry 19).
As shown in Table 2, by employing 5 mol-% of these com-
plexes and performing the reactions in acetonitrile, commer-
cially available (E)-benzaldoxime could be transformed into
benzonitrile in ≥87 % GC yield after 2–5 h of heating at 80 °C.
Minor amounts of benzamide (1–3 %), a byproduct resulting
from the formal rearrangement of the substrate,[25] were also
formed in these reactions. Interestingly, the electronic nature
of the N-aryl substituents of the guanidinate ligands plays a
significant role in both the efficiency and selectivity of the pro-
cess. Thus, the best results were obtained with complexes 3d–
g bearing the strong electron-withdrawing CF3 and CN groups,
which led to the almost quantitative consumption (≥96 % by
GC) of the starting (E)-benzaldoxime after only 2 h (entries 4–
7). By employing these complexes, the quantity of benzamide
formed never exceeded 1 %. We would like to stress here that,
contrary to other catalytic systems previously described in the
literature, no additives or co-catalysts were needed.[26]
Table 3. Catalytic dehydration of aldoximes using the guanidinate–osmium(II)
complex [OsCl{κ2-(N,N′)-C(N-4-C6H4CF3)(NiPr)NHiPr}(η6-p-cymene)] (3d).[a]
Entry
R
Time [h]
Conv.[b] [%]
Yield[b] [%]
1
2
3
4
5
6
7
8
Ph
2
1
2
2
1
3
1
1
2
1
1
1
1
1
1
3
5
2
2
2
1.5
6
99
99
>99
>99
99
98 (85)
99 (86)
98 (88)
98 (87)
98 (84)
96 (85)
99 (86)
99 (90)
98 (88)
99 (86)
99 (85)
99 (82)
98 (86)
99 (89)
99 (85)
99 (83)
99 (81)
99 (88)
99 (86)
99 (84)
99
2-C6H4Me
3-C6H4Me
4-C6H4Me
2-C6H4OMe
4-C6H4OMe
4-C6H4SMe
2-C6H4Cl
96
99
>99
99
>99
>99
99
>99
>99
>99
>99
>99
>99
>99
>99
>99
90
Table 2. Catalytic dehydration of (E)-benzaldoxime using the guanidinate–
osmium(II) complexes [OsCl{κ2-(N,N′)-C(NR)(NiPr)NHiPr}(η6-p-cymene)] (3a–
m).[a]
9
4-C6H4F
10
11
12
13
14
15
16
17
18
19
20
21[c]
22[d]
2,4-C6H3Cl2
2,6-C6H3Cl2
2-Cl-6-C6H3F
2-C6H4NO2
3-furyl
2-thienyl
n-C5H11
n-C6H13
CH2CH2Ph
(S)-citronellyl
(E)-CH=CHPh
2,6-C6H3Cl2
2,6-C6H3Cl2
Entry
Catalyst
Time [h]
Conv.[b] [%]
Yield of benzonitrile[b]
[%]
1
2
3
4
5
6
7
8
3a
3b
3c
3d
3e
3f
3g
3h
3i
4
3
3
2
2
2
2
5
5
5
5
5
5
95
98
98
99
98
99
96
95
96
93
95
90
92
92
97
96
98
97
98
95
94
94
91
94
87
90
90
[a] Reactions were performed under Ar with 0.5 mmol of the corresponding
aldoxime (0.33 in acetonitrile). [b] Determined by GC (uncorrected GC ar-
M
eas); isolated yields after chromatographic work-up are given in parentheses.
The difference between the GC conversion and yield corresponds to the pri-
mary amide present in the reaction mixture. [c] Reaction performed with
3 mol-% of complex 3d. [d] Reaction performed with 1 mol-% of complex
3d.
9
10
11
12
13
3j
3k
3l
3m
On the other hand, it is worth noting that the use of a lower
catalyst loading was tolerated without a drastic increase in the
reaction times. For example, by using 3 mol-% of 3d, 2,6-dichlo-
robenzaldoxime was completely converted into 2,6-dichloro-
benzonitrile after 1.5 h of heating (entry 21; to be compared
[a] Reactions were performed under Ar with 0.5 mmol of (E)-benzaldoxime
(0.33 in acetonitrile). [b] Yield of benzonitrile determined by GC (uncor-
M
rected GC areas). The difference between GC conversion and yield corre-
sponds to the amount of benzamide generated in the reactions.
By using one of the most active complexes, that is, [OsCl{κ2- with entry 11).[28] However, a further reduction of the osmium
(N,N′)-C(N-4-C6H4CF3)(NiPr)NHiPr}(η6-p-cymene)] (3d), the scope loading to 1 mol-% resulted in an incomplete transformation
of the process was next explored by carrying out the dehydra- (entry 22). Also of note is the fact that the H and 13C{1H} NMR
1
tion of differently substituted benzaldoximes (Table 3).[27] Thus, spectra obtained from the crudes show the formation of an
as observed for (E)-benzaldoxime (entry 1), all the substrates equimolar amount of acetamide with respect to the generated
tested could be selectively transformed into the corresponding nitrile in all the reactions presented in Table 3. This fact, along
benzonitriles in high yields (≥96 % by GC; ≥82 % after chromat- with the low conversions obtained when using toluene as the
ographic purification; entries 2–13) after 1–3 h of heating, re- reaction medium,[29] strongly suggests the active participation
Eur. J. Inorg. Chem. 2016, 393–402
www.eurjic.org
396
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
of the acetonitrile solvent in the dehydration process. In this Experimental Section
sense, a catalytic cycle similar to the one proposed by Tambara
General: The synthetic procedures were performed under dry ar-
gon by using vacuum-line and standard Schlenk or sealed-tube
techniques. The solvents were dried by standard methods and dis-
tilled under argon before use. The complex [{OsCl(μ-Cl)(η6-p-
cymene)}2] (1),[33] the guanidines (iPrHN)2C=NR (2a–m),[34] and all
the aldoximes employed in the catalytic experiments,[35] except for
(E)-benzaldoxime, were prepared by following the methods re-
ported in the literature. IR spectra were recorded with a Perkin–
Elmer 1720-XFT spectrometer. GC measurements were performed
by using a Hewlett–Packard HP6890 apparatus (Supelco Beta-
DexTM 120 column, 30 m length, 250 μm diameter). Elemental anal-
yses were performed by the Analytical Service of the Instituto de
Investigaciones Químicas (IIQ-CSIC) of Seville. The NMR spectra
were recorded with a Bruker DPX300 or AV400 spectrometer. The
chemical shifts are given in parts per million and are referenced to
the residual peak of the deuteriated solvent employed (1H and 13C)
or the CFCl3 standard (19F). DEPT experiments were carried out for
all the compounds reported in this paper.
and Pantoş can be evoked to explain these observations
(Scheme 2).[20b] In the first step, a molecule of the acetonitrile
solvent coordinates to the metal with the dissociation of the
chloride ligand to generate the catalytically active species K.[30]
Subsequent coordination of the aldoxime to K generates the
intermediate species L.[31] The coordinated aldoxime is dehy-
drated and the acetonitrile molecule converted into acetamide,
via the five membered metallacyclic intermediate M, to give
the nitrile-containing intermediate N. Final displacement of the
nitrile by the acetonitrile solvent closes the catalytic cycle.[32]
[OsCl{κ2-(N,N′)-C(NR)(NiPr)NHiPr}(η6-p-cymene)] [R = Ph (3a), 4-
C6H4F (3b), 4-C6H4Cl (3c), 4-C6H4CF3 (3d), 3-C6H4CF3 (3e), 3,5-
C6H3(CF3)2 (3f), 4-C6H4CN (3g), 4-C6H4Me (3h), 3-C6H4Me (3i),
2-C6H4Me (3j), 4-C6H4tBu (3k), 2,6-C6H3iPr2 (3l), 2,4,6-C6H2Me3
(3m)]:
A
solution of [{OsCl(μ-Cl)(η6-p-cymene)}2] (1; 0.200 g,
0.253 mmol) in tetrahydrofuran (15 mL) was treated with the appro-
priate guanidine 2a–m (1.012 mmol) at room temperature for 1 h.
A rapid color change from orange to lemon-yellow was observed.
After this time, the solvent was evaporated to dryness and pentane
(20 mL) was added to the resulting oily residue, leading to the
appearance of a white solid precipitate of the corresponding guan-
idinium chloride salt [(iPrHN)2C(NHR)]Cl (4a–m).[4] The suspension
was then filtered by using a cannula and the filtrate stored in a
freezer at –10 °C for 48 h. Cooling led to the precipitation of
[OsCl{κ2-(N,N′)-C(NR)(NiPr)NHiPr}(η6-p-cymene)] (3a–m) as yellow
solids, which were separated, washed with cold pentane (3 mL),
and dried under vacuum.
Scheme 2. Proposed mechanism for the catalytic dehydration reactions.
1
3a: Yield 0.231 g (79 %). IR (KBr): ν = 3317 (m, N–H) cm–1. H NMR
˜
3
(C6D6): δ= 7.45–7.31 (m, 4 H, CHarom), 6.99 (t, JH,H = 6.9 Hz, 1 H,
CHarom), 5.46 and 5.41 (d, JH,H = 5.4 Hz, 1 H each, CH of cym), 5.17
and 5.13 (d, JH,H = 5.1 Hz, 1 H each, CH of cym), 3.53 and 3.37 (m,
1 H each, NCHMe2), 3.04 (d, JH,H = 10.8 Hz, 1 H, NH), 2.47 (m, 1 H,
3
3
Conclusions
3
3
A series of guanidinate–osmium(II) complexes 3a–m of general
composition
CHMe2 of cym), 2.31 (s, 3 H, Me of cym), 1.47 and 0.78 (d, JH,H
=
=
[OsCl{κ2-(N,N′)-C(NR)(NiPr)NHiPr}(η6-p-cymene)]
6.3 Hz, 3 H each, NCHMe2 or CHMe2 of cym), 1.19 and 1.17 (d, 3JH,H
3
(R = Ar) have been synthesized by the reaction of the dimer 6.6 Hz, 3 H each, NCHMe2 or CHMe2 of cym), 1.11 (d, JH,H = 6.0 Hz,
3 H, NCHMe2 or CHMe2 of cym), 1.09 (d, 3JH,H = 6.9 Hz, 3 H, NCHMe2
or CHMe2 of cym) ppm. 13C{1H} NMR (C6D6): δ= 164.7 (s, CN3), 149.5
(s, Carom), 128.6, 122.0, and 120.5 (s, CHarom), 89.9 and 89.1 (s, C of
cym), 70.5, 69.8, 68.0, and 67.6 (s, CH of cym), 45.4 and 45.0 (s,
NCHMe2), 32.0 (s, CHMe2 of cym), 25.6, 25.0, 24.3, 22.9, 22.5, and
22.4 (s, NCHMe2 and CHMe2 of cym), 18.9 (s, Me of cym) ppm.
C23H34ClN3Os (578.22): calcd. C 47.78, H 5.93, N 7.27; found C 47.90,
H 6.02, N 7.33.
[{OsCl(μ-Cl)(η6-p-cymene)}2] (1) with an excess of the corre-
sponding guanidine (iPrHN)2C=NR (2a–m). These compounds
represent the first examples of half-sandwich osmium com-
plexes containing heteroallyl guanidinate monoanions as li-
gands reported to date, and the structures of three of these
complexes have been unequivocally established by means of
single-crystal diffraction techniques. In addition, their synthetic
utility has been demonstrated through their ability to catalyze
the selective dehydration of aldoximes. In particular, by using
[OsCl{κ2-(N,N′)-C(N-4-C6H4CF3)(NiPr)NHiPr}(η6-p-cymene)] (3d),
a large variety of aromatic, heteroaromatic, α,ꢀ-unsaturated,
and aliphatic aldoximes have been cleanly converted into the
corresponding nitriles in high yields and short reaction times.
Notably, to the best of our knowledge, no previous examples
of osmium-based catalytic systems for this important transfor-
mation have been reported in the literature.
3b: Yield 0.247 g (82 %). IR (KBr): ν = 3319 (m, N–H) cm–1 19F{1H}
.
˜
NMR (C6D6): δ= –123.1 (s) ppm. 1H NMR (C6D6): δ= 7.25 and 7.00
(m, 2 H each, CHarom), 5.42 and 5.15 (d, 3JH,H = 5.1 Hz, 1 H each, CH
of cym), 5.33 and 5.08 (d, 3JH,H = 5.4 Hz, 1 H each, CH of cym), 3.45–
3
3.30 (m, 2 H, NCHMe2), 2.97 (d, JH,H = 10.5 Hz, 1 H, NH), 2.46 (m, 1
H, CHMe2 of cym), 2.29 (s, 3 H, Me of cym), 1.44, 1.17, and 0.77 (d,
3JH,H = 6.3 Hz, 3 H each, NCHMe2 or CHMe2 of cym), 1.16, 1.11, and
3
1.08 (d, JH,H = 7.2 Hz, 3 H each, NCHMe2 or CHMe2 of cym) ppm.
1
13C{1H} NMR (C6D6): δ= 164.8 (s, CN3), 157.8 (d, JC,F = 238.8 Hz,
Eur. J. Inorg. Chem. 2016, 393–402
www.eurjic.org
397
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Carom), 145.7 (d, 4JC,F = 2.2 Hz, Carom), 123.1 (d, 3JC,F = 7.3 Hz, CHarom), 6.3 Hz, 3 H each, NCHMe2 or CHMe2 of cym), 1.04 and 0.95 (d, 3JH,H
=
=
2
3
115.0 (d, JC,F = 22.0 Hz, CHarom), 89.8 and 89.2 (s, C of cym), 70.5,
6.9 Hz, 3 H, NCHMe2 or CHMe2 of cym), 0.94 and 0.81 (d, JH,H
69.7, 68.1, and 67.6 (s, CH of cym), 45.4 and 44.9 (s, NCHMe2), 32.1 6.0 Hz, 3 H each, NCHMe2 or CHMe2 of cym) ppm. 13C{1H} NMR
2
(s, CHMe2 of cym), 25.6, 25.0, 24.2, 22.8, 22.6, and 22.5 (s, NCHMe2
(C6D6): δ= 164.7 (s, CN3), 151.5 (s, Carom), 131.8 (q, JC,F = 32.5 Hz,
1
and CHMe2 of cym), 18.9 (s, Me of cym) ppm. C23H33ClFN3Os
Carom), 124.2 (q, JC,F = 272.6 Hz, CF3), 120.7 and 111.9 (s, CHarom),
(596.21): calcd. C 46.33, H 5.58, N 7.05; found C 46.25, H 5.66, N
7.09.
90.4 and 89.9 (s, C of cym), 70.1, 69.9, 68.6, and 67.3 (s, CH of cym),
46.0 and 45.7 (s, NCHMe2), 32.0 (s, CHMe2 of cym), 25.1, 24.4, 24.0,
22.5, 22.2, and 22.0 (s, NCHMe2 and CHMe2 of cym), 18.8 (s, Me of
cym) ppm. 19F{1H} NMR (C6D6): δ= –62.6 (s) ppm. C25H32ClF6N3Os
(714.21): calcd. C 42.04, H 4.52, N 5.88; found C 41.98, H 4.61, N
6.09.
1
3c: Yield 0.270 g (87 %). IR (KBr): ν = 3313 (m, N–H) cm–1. H NMR
˜
3
(C6D6): δ= 7.31 and 7.20 (d, JH,H = 8.8 Hz, 2 H each, CHarom), 5.41
3
and 5.32 (d, JH,H = 5.1 Hz, 1 H each, CH of cym), 5.12 and 5.07 (d,
3JH,H = 5.4 Hz, 1 H each, CH of cym), 3.43 and 3.36 (m, 1 H each,
3
NCHMe2), 2.96 (d, JH,H = 10.8 Hz, 1 H, NH), 2.43 (m, 1 H, CHMe2 of
cym), 2.27 (s, 3 H, Me of cym), 1.42 and 1.15 (d, JH,H = 6.3 Hz, 3 H
each, NCHMe2 or CHMe2 of cym), 1.14 (d, JH,H = 6.9 Hz, 3 H,
3g: Yield 0.220 g (72 %). IR (KBr): ν = 3332 (m, N–H), 2212 (s,
˜
3
1
3
C≡N) cm–1. H NMR (C6D6): δ= 7.30 and 7.07 (d, JH,H = 8.7 Hz, 2 H
each, CHarom), 5.40 and 5.29 (d, 3JH,H = 5.1 Hz, 1 H each, CH of cym),
5.08 and 5.06 (d, 3JH,H = 4.8 Hz, 1 H each, CH of cym), 3.34–3.23 (m,
3
3
NCHMe2 or CHMe2 of cym), 1.09 (d, JH,H = 6.0 Hz, 3 H, NCHMe2 or
CHMe2 of cym), 1.07 and 0.75 (d, JH,H = 6.6 Hz, 3 H each, NCHMe2 2 H, NCHMe2), 3.00 (d, 3JH,H = 10.8 Hz, 1 H, NH), 2.31 (m, 1 H, CHMe2
3
or CHMe2 of cym) ppm. 13C{1H} NMR (C6D6): δ= 164.7 (s, CN3), 148.3
and 125.0 (s, Carom), 128.6 and 123.0 (s, CHarom), 90.0 and 89.3 (s, C
of cym), 70.4, 69.8, 68.0, and 67.4 (s, CH of cym), 45.4 and 45.1 (s,
NCHMe2), 32.0 (s, CHMe2 of cym), 25.5, 24.9, 24.2, 22.8, 22.6, and
22.4 (s, NCHMe2 and CHMe2 of cym), 18.9 (s, Me of cym) ppm.
C23H33Cl2N3Os (612.66): calcd. C 45.09, H 5.43, N 6.86; found C 45.15,
H 5.39, N 6.98.
of cym), 2.23 (s, 3 H, Me of cym), 1.37 and 1.07 (d, JH,H = 6.3 Hz, 3
H each, NCHMe2 or CHMe2 of cym), 1.10 and 0.75 (d, 3JH,H = 6.6 Hz,
3 H each, NCHMe2 or CHMe2 of cym), 1.06 and 1.00 (d, 3JH,H = 6.9 Hz,
3 H each, NCHMe2 or CHMe2 of cym) ppm. 13C{1H} NMR (C6D6): δ=
164.5 (s, CN3), 153.4 and 102.2 (s, Carom), 132.6 and 121.1 (s, CHarom),
120.5 (s, C≡N), 90.7 and 89.7 (s, C of cym), 70.4, 70.1, 67.8, and 67.1
(s, CH of cym), 45.7 and 45.6 (s, NCHMe2), 31.9 (s, CHMe2 of cym),
25.2, 24.6, 24.3, 22.6, 22.5, and 22.1 (s, NCHMe2 and CHMe2 of cym),
18.8 (s, Me of cym) ppm. C24H33ClN4Os (603.23): calcd. C 47.79, H
5.51, N 9.29; found C 47.92, H 5.45, N 9.45.
3
1
3d: Yield 0.278 g (85 %). IR (KBr): ν = 3358 (m, N–H) cm–1. H NMR
˜
3
(C6D6): δ= 7.60 and 7.28 (d, JH,H = 8.4 Hz, 2 H each, CHarom), 5.41
3
3
and 5.32 (d, JH,H = 5.4 Hz, 1 H each, CH of cym), 5.11 (d, JH,H
=
=
5.1 Hz, 2 H, CH of cym), 3.41–3.27 (m, 2 H, NCHMe2), 3.00 (d, 3JH,H
3h: Yield 0.243 g (81 %). IR (KBr): ν = 3315 (m, N–H) cm–1. H NMR
1
˜
3
10.8 Hz, 1 H, NH), 2.36 (m, 1 H, CHMe2 of cym), 2.26 (s, 3 H, Me of
(C6D6): δ= 7.38 and 7.17 (d, JH,H = 8.1 Hz, 2 H each, CHarom), 5.46
3
3
cym), 1.40, 1.81, and 0.74 (d, JH,H = 6.3 Hz, 3 H each, NCHMe2 or
and 5.43 (d, JH,H = 5.4 Hz, 1 H each, CH of cym), 5.20 and 5.15 (d,
3
CHMe2 of cym), 1.13 (d, JH,H = 6.0 Hz, 3 H, NCHMe2 or CHMe2 of
3JH,H = 5.1 Hz, 1 H each, CH of cym), 3.56 and 3.38 (m, 1 H each,
3
3
cym), 1.11 (d, JH,H = 6.6 Hz, 3 H, NCHMe2 or CHMe2 of cym), 1.02
NCHMe2), 3.05 (d, JH,H = 10.8 Hz, 1 H, NH), 2.51 (m, 1 H, CHMe2 of
(d, 3JH,H = 6.9 Hz, 3 H, NCHMe2 or CHMe2 of cym) ppm. 13C{1H} NMR
cym), 2.32 (s, 6 H, ArMe and Me of cym), 1.48 and 1.21 (d, JH,H
=
=
3
3
(C6D6): δ= 164.6 (s, CN3), 152.8 (s, Carom), 125.8 (q, JC,F = 3.3 Hz,
6.3 Hz, 3 H each, NCHMe2 or CHMe2 of cym), 1.19 and 1.12 (d, 3JH,H
1
2
3
CHarom), 125.6 (q, JC,F = 270.7 Hz, CF3), 121.5 (q, JC,F = 32.2 Hz, 6.9 Hz, 3 H each, NCHMe2 or CHMe2 of cym), 1.13 (d, JH,H = 5.7 Hz,
Carom), 121.1 (s, CHarom), 90.4 and 89.6 (s, C of cym), 70.3, 70.1, 68.0,
and 67.3 (s, CH of cym), 45.6 and 45.5 (s, NCHMe2), 32.1 (s, CHMe2
of cym), 25.3, 24.7, 24.3, 22.7, 22.6, and 22.2 (s, NCHMe2 and CHMe2
of cym), 18.9 (s, Me of cym) ppm. 19F{1H} NMR (C6D6): δ= –60.5
(s) ppm. C24H33ClF3N3Os (646.21): calcd. C 44.61, H 5.15, N 6.50;
found C 44.72, H 5.09, N 6.58.
3 H, NCHMe2 or CHMe2 of cym), 0.81 (d, 3JH,H = 6.6 Hz, 3 H, NCHMe2
or CHMe2 of cym) ppm. 13C{1H} NMR (C6D6): δ= 164.8 (s, CN3), 147.1
and 129.6 (s, Carom), 129.2 and 122.1 (s, CHarom), 89.8 and 89.0 (s, C
of cym), 70.6, 69.7, 68.1, and 67.7 (s, CH of cym), 45.4 and 44.8 (s,
NCHMe2), 32.0 (s, CHMe2 of cym), 25.7, 25.1, 24.3, 23.0, and 22.5 (s,
2 C, NCHMe2 and CHMe2 of cym), 20.7 and 19.0 (s, ArMe and Me of
cym) ppm. C24H36ClN3Os (592.25): calcd. C 48.67, H 6.13, N 7.10;
found C 48.55, H 6.15, N 7.23.
1
3e: Yield 0.248 g (76 %). IR (KBr): ν = 3324 (m, N–H) cm–1. H NMR
˜
(C6D6): δ= 7.81 (s, 1 H, CHarom), 7.40 (m, 1 H, CHarom), 7.17 (m, 2 H,
3
1
CHarom), 5.43 and 5.35 (d, JH,H = 5.1 Hz, 1 H each, CH of cym), 5.14
3i: Yield 0.234 g (78 %). IR (KBr): ν = 3323 (m, N–H) cm–1. H NMR
˜
and 5.12 (d, 3JH,H = 6.1 Hz, 1 H each, CH of cym), 3.45–3.27 (m, 2 H,
(C6D6): δ= 7.32–7.28 (m, 3 H, CHarom), 6.84 (d, JH,H = 6.0 Hz, 1 H,
CHarom), 5.47 and 5.46 (d, JH,H = 4.8 Hz, 1 H each, CH of cym), 5.20
and 5.18 (d, JH,H = 4.5 Hz, 1 H each, CH of cym), 3.58 and 3.37 (m,
1 H each, NCHMe2), 3.06 (d, JH,H = 10.8 Hz, 1 H, NH), 2.51 (m, 1 H,
3
3
3
NCHMe2), 3.02 (d, JH,H = 10.5 Hz, 1 H, NH), 2.38 (m, 1 H, CHMe2 of
cym), 2.24 (s, 3 H, Me of cym), 1.39 and 0.81 (d, JH,H = 6.3 Hz, 3 H
each, NCHMe2 or CHMe2 of cym), 1.14 and 1.05 (d, JH,H = 6.6 Hz, 3
3
3
3
3
3
H, NCHMe2 or CHMe2 of cym), 1.11 and 1.03 (d, JH,H = 6.6 Hz, 3 H CHMe2 of cym), 2.36 and 2.32 (s, 3 H each, ArMe and Me of cym),
each, NCHMe2 or CHMe2 of cym) ppm. 13C{1H} NMR (C6D6): δ= 164.7
1.48, 1.20 and 1.11 (d, JH,H = 6.3 Hz, 3 H each, NCHMe2 or CHMe2
3
2
3
(s, CN3), 150.3 (s, Carom), 130.8 (q, JC,F = 31.4 Hz, Carom), 129.2 and
125.1 (s, CHarom), 125.3 (q, JC,F = 272.3 Hz, CF3), 117.5 (q, JC,F
of cym), 1.18 and 1.12 (d, JH,H = 6.9 Hz, 3 H each, NCHMe2 or
1
3
3
=
CHMe2 of cym), 0.81 (d, JH,H = 6.6 Hz, 3 H, NCHMe2 or CHMe2 of
3
3.3 Hz, CHarom), 116.2 (q, JC,F = 3.6 Hz, CHarom), 89.9 and 89.7 (s, C cym) ppm. 13C{1H} NMR (C6D6): δ= 164.9 (s, CN3), 149.5 and 137.9
of cym), 70.2, 70.0, 68.3, and 67.4 (s, CH of cym), 45.5 and 45.4 (s,
NCHMe2), 32.0 (s, CHMe2 of cym), 25.4, 24.7, 24.2, 22.6, 22.5, and
22.2 (s, NCHMe2 and CHMe2 of cym), 18.9 (s, Me of cym) ppm.
(s, Carom), 128.5, 122.7, 121.5, and 119.3 (s, CHarom), 89.7 and 89.2 (s,
C of cym), 70.5, 69.8, 68.2, and 67.7 (s, CH of cym), 45.4 and 45.0 (s,
NCHMe2), 32.0 (s, CHMe2 of cym), 25.7, 25.1, 24.4, 22.9, 22.5, and
19F{1H} NMR (C6D6): δ= –62.0 (s) ppm. C24H33ClF3N3Os (646.21): 22.4 (s, NCHMe2 and CHMe2 of cym), 21.4 and 19.0 (s, ArMe and Me
calcd. C 44.61, H 5.15, N 6.50; found C 44.68, H 5.30, N 6.61.
of cym) ppm. C24H36ClN3Os (592.25): calcd. C 48.67, H 6.13, N 7.10;
found C 48.78, H 6.09, N 6.98.
1
3f: Yield 0.318 g (88 %). IR (KBr): ν = 3324 (m, N–H) cm–1. H NMR
˜
1
(C6D6): δ= 7.79 (s, 2 H, CHarom), 7.47 (m, 1 H, CHarom), 5.41, 5.34,
5.16, and 5.04 (d, JH,H = 5.1 Hz, 1 H each, CH of cym), 3.31–3.16
(m, 2 H, NCHMe2), 3.11 (d, JH,H = 11.1 Hz, 1 H, NH), 2.32 (m, 1 H,
CHMe2 of cym), 2.13 (s, 3 H, Me of cym), 1.29 and 1.07 (d, JH,H
3j: Yield 0.216 g (72 %). IR (KBr): ν = 3332 (m, N–H) cm–1. H NMR
˜
(C6D6): δ= 7.36 (d, 3JH,H = 7.5 Hz, 1 H, CHarom), 7.21 (m, 2 H, CHarom),
3
3
3
7.07 (m, 1 H, CHarom), 5.37 (d, JH,H = 5.1 Hz, 1 H, CH of cym), 5.05
(m, 3 H, CH of cym), 3.37 and 3.26 (m, 1 H each, NCHMe2), 3.05 (d,
3
=
Eur. J. Inorg. Chem. 2016, 393–402
www.eurjic.org
398
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
3JH,H = 10.2 Hz, 1 H, NH), 2.69 and 2.23 (s, 3 H each, ArMe and Me
The characterization data for the novel guanidinium chloride salts
[(iPrHN)2C(NHR)]Cl [R = 4-C6H4CF3 (4d), 3-C6H4CF3 (4e), 3,5-
3
of cym), 2.52 (m, 1 H, CHMe2 of cym), 1.49 and 1.20 (d, JH,H
=
6.3 Hz, 3 H each, NCHMe2 or CHMe2 of cym), 1.26, 1.17 and 1.14 (d,
C6H3(CF3)2 (4f), 2-C6H4Me (4j)] are as follows:
3
3JH,H = 6.9 Hz, 3 H each, NCHMe2 or CHMe2 of cym), 0.65 (d, JH,H
=
4d: Yield 0.116 g (71 %). IR (KBr): ν = 3240 (s, N–H), 3178 (s, N–
˜
6.6 Hz, 3 H, NCHMe2 or CHMe2 of cym) ppm. 13C{1H} NMR (C6D6):
δ= 166.6 (s, CN3), 149.7 and 132.6 (s, Carom), 131.1, 125.5, 122.0, and
121.9 (s, CHarom), 90.6 and 87.3 (s, C of cym), 69.7, 69.4, 68.8, and
67.7 (s, CH of cym), 45.2 and 44.3 (s, NCHMe2), 31.4 (s, CHMe2 of
cym), 25.7, 24.9, 24.0, 23.7, 22.5, and 22.2 (s, NCHMe2 and CHMe2 of
cym), 19.7 and 18.7 (s, ArMe and Me of cym) ppm. C24H36ClN3Os
(592.25): calcd. C 48.67, H 6.13, N 7.10; found C 48.51, H 6.11, N
7.23.
1
H) cm–1. H NMR (CDCl3): δ= 10.31 (br. s, 1 H, NH), 7.99 (br. s, 2 H,
3
NH), 7.56 and 7.39 (d, JH,H = 8.2 Hz, 2 H each, CHarom), 3.97 (br. s,
2 H, CHMe2), 1.20 (d, 3JH,H = 6.3 Hz, 12 H, CHMe2) ppm. 13C{1H} NMR
2
(CDCl3): δ= 154.6 (s, CN3), 141.0 (s, Carom), 126.8 (q, JC,F = 34.0 Hz,
1
Carom), 126.7 and 122.0 (s, CHarom), 123.8 (q, JC,F = 271.6 Hz, CF3),
46.4 (s, CHMe2), 22.5 (s, CHMe2) ppm. 19F{1H} NMR (CDCl3): δ= –62.3
(s) ppm. C14H21ClF3N3 (323.78): calcd. C 51.93, H 6.54, N 12.98; found
C 52.06, H 6.47, N 13.11.
1
3k: Yield 0.247 g (77 %). IR (KBr): ν = 3339 (m, N–H) cm–1. H NMR
˜
4e: Yield 0.126 g (77 %). IR (KBr): ν = 3198 (s, N–H), 3180 (s, N–
˜
3
(C6D6): δ= 7.35 (br., 4 H, CHarom), 5.37 and 5.35 (d, JH,H = 5.0 Hz, 1
1
H) cm–1. H NMR (CDCl3): δ= 10.19 (br. s, 1 H, NH), 7.86 (br. s, 2 H,
H each, CH of cym), 5.10 and 5.09 (d, 3JH,H = 4.2 Hz, 1 H each, CH of
cym), 3.47 and 3.30 (m, 1 H each, NCHMe2), 2.95 (d, 3JH,H = 10.5 Hz,
1 H, NH), 2.42 (m, 1 H, CHMe2 of cym), 2.22 (s, 3 H, Me of cym), 1.37
NH), 7.41–7.29 (m, 4 H, CHarom), 3.86 (br. s, 2 H, CHMe2), 1.09 (d,
3JH,H = 6.1 Hz, 12 H, CHMe2) ppm. 13C{1H} NMR (CDCl3): δ= 154.5 (s,
2
CN3), 138.3 (s, Carom), 131.7 (q, JC,F = 32.4 Hz, Carom), 130.0, 124.9,
3
(d, JH,H = 6.4 Hz, 3 H, NCHMe2 or CHMe2 of cym), 1.31 (s, 9 H,
CMe3), 1.10 (d, JH,H = 6.1 Hz, 3 H, NCHMe2 or CHMe2 of cym), 1.08
(d, JH,H = 6.9 Hz, 3 H, NCHMe2 or CHMe2 of cym), 1.00 (d, JH,H
6.8 Hz, 3 H, NCHMe2 or CHMe2 of cym), 0.98 (d, JH,H = 6.2 Hz, 3 H,
1
121.6, and 118.2 (s, CHarom), 123.4 (q, JC,F = 272.5 Hz, CF3), 46.2 (s,
3
CHMe2), 22.3 (s, CHMe2) ppm. 19F{1H} NMR (CDCl3): δ= –63.0
(s) ppm. C14H21ClF3N3 (323.78): calcd. C 51.93, H 6.54, N 12.98; found
C 51.80, H 6.63, N 13.09.
3
3
=
3
3
NCHMe2 or CHMe2 of cym), 0.70 (d, JH,H = 6.6 Hz, 3 H, NCHMe2 or
CHMe2 of cym) ppm. 13C{1H} NMR (C6D6): δ= 164.8 (s, CN3), 146.7
and 143.0 (s, Carom), 125.3 and 121.9 (s, CHarom), 89.9 and 89.2 (s, C
of cym), 70.3, 69.9, 68.1, and 67.6 (s, CH of cym), 45.5 and 44.9 (s,
NCHMe2), 34.0 (s, CMe3), 32.1 (s, CHMe2 of cym), 31.5 (s, CMe3), 25.7,
25.1, 24.3, 22.8, and 22.6 (s, 2 C, NCHMe2 and CHMe2 of cym), 19.0
(s, Me of cym) ppm. C27H42ClN3Os (634.33): calcd. C 51.12, H 6.67,
N 6.62; found C 51.20, H 6.81, N 6.79.
4f: Yield 0.158 g (80 %). IR (KBr): ν = 3195 (s, N–H), 3066 (s, N–
˜
1
H) cm–1. H NMR (CDCl3): δ= 10.59 (br. s, 1 H, NH), 8.14 (br. s, 2 H,
NH), 7.79 (s, 2 H, CHarom), 7.60 (s, 1 H, CHarom), 3.95 (br. s, 2 H,
3
CHMe2), 1.24 (d, JH,H = 6.2 Hz, 12 H, CHMe2) ppm. 13C{1H} NMR
2
(CDCl3): δ= 154.4 (s, CN3), 139.7 (s, Carom), 132.8 (q, JC,F = 33.9 Hz,
C
arom), 122.7 (q, 1JC,F = 272.8 Hz, CF3), 120.6 and 118.0 (br. s, CHarom),
46.9 (s, CHMe2), 22.4 (s, CHMe2) ppm. 19F{1H} NMR (CDCl3): δ= –63.2
(s) ppm. C15H20ClF6N3 (391.78): calcd. C 45.98, H 5.15, N 10.73; found
C 46.12, H 5.18, N 10.85.
1
3l: Yield 0.234 g (70 %). IR (KBr): ν = 3329 (m, N–H) cm–1. H NMR
˜
(C6D6): δ= 7.36–7.22 (m, 3 H, CHarom), 5.41 and 5.22 (d, 3JH,H = 5.4 Hz,
1 H each, CH of cym), 5.36 and 5.27 (d, JH,H = 5.1 Hz, 1 H each, CH
4j: Yield 0.101 g (74 %). IR (KBr): ν = 3215 (s, N–H), 3199 (s, N–
3
˜
H) cm–1
.
1H NMR (CDCl3): δ= 9.48 (br. s, 1 H, NH), 7.34 (br. s, 2 H,
of cym), 4.46, 3.52, 3.33, and 3.10 (m, 1 H each, NCHMe2 and CHMe2
NH), 7.14–5.98 (m, 4 H, CHarom), 3.89 (br. s, 2 H, CHMe2), 2.21 (s, 3
3
of Ar), 2.92 (d, JH,H = 10.8 Hz, 1 H, NH), 2.83 (m, 1 H, CHMe2 of
3
H, ArMe), 1.05 (d, JH,H = 6.3 Hz, 12 H, CHMe2) ppm. 13C{1H} NMR
cym), 2.15 (s, 3 H, Me of cym), 1.58, 1.55, 1.50, 1.47, and 1.41 (d,
(CDCl3): δ= 154.5 (s, CN3), 135.3 and 133.7 (s, Carom), 131.2, 127.0,
126.9, and 125.3 (s, CHarom), 45.5 (s, CHMe2), 22.5 (s, CHMe2), 18.1
(s, ArMe) ppm. C14H24ClN3 (269.81): calcd. C 62.32, H 8.97, N 15.57;
found C 62.46, H 9.03, N 15.69.
3JH,H = 6.6 Hz, 3 H each, NCHMe2, CHMe2 of cym or CHMe2 of Ar),
3
1.36 and 1.33 (d, JH,H = 7.2 Hz, 3 H each, NCHMe2, CHMe2 of cym
3
or CHMe2 of Ar), 1.24, 1.21, and 0.56 (d, JH,H = 6.3 Hz, 3 H each,
NCHMe2, CHMe2 of cym or CHMe2 of Ar) ppm. 13C{1H} NMR (C6D6):
δ= 170.2 (s, CN3), 146.1, 144.2, and 143.5 (s, Carom), 124.8, 124.4, and
124.0 (s, CHarom), 92.2 and 83.9 (s, C of cym), 70.2, 69.1, 68.9, and
67.0 (s, CH of cym), 45.2 and 44.4 (s, NCHMe2), 31.9 (s, CHMe2 of
cym), 27.6 and 27.2 (s, CHMe2 of Ar), 26.2, 25.5, 25.4, 24.5, 24.4, 23.6,
23.5, 23.2, 22.6, and 22.5 (s, NCHMe2, CHMe2 of cym and CHMe2 of
Ar), 18.7 (s, Me of cym) ppm. C29H46ClN3Os (662.38): calcd. C 52.58,
H 7.00, N 6.34; found C 52.71, H 7.11, N 6.53.
General Procedure for the Catalytic Dehydration of Aldoximes
Employing the Osmium–Guanidinate Complex [OsCl{κ2-(N,N′)-
C(N-4-C6H4CF3)(NiPr)NHiPr}(η6-p-cymene)] (3d): The aldoxime
(0.5 mmol), acetonitrile (1.5 mL), and osmium(II) complex 3d
(0.016 g, 0.025 mmol) were introduced into a Teflon-capped sealed
tube, and the reaction mixture stirred at 80 °C for the indicated
time (see Table 3). The course of the reaction was monitored by
regularly taking samples of around 20 μL, which, after extraction
with CH2Cl2 (3 mL), were analyzed by GC. To isolate the nitrile pro-
ducts, the identities of which were assessed by comparison of their
NMR data with those reported in the literature (copies of the NMR
spectra are given in the Supporting Information), the solvent was
eliminated under reduced pressure and the crude reaction mixture
purified by column chromatography on silica gel using an ethyl
acetate/hexane mixture (40:60, v/v) as eluent.
1
3m: Yield 0.235 g (75 %). IR (KBr): ν = 3324 (m, N–H) cm–1. H NMR
˜
(C6D6): δ= 7.06 and 6.98 (s, 1 H each, CHarom), 5.32 and 5.27 (d,
3JH,H = 5.4 Hz, 1 H each, CH of cym), 5.19 and 5.09 (d, 3JH,H = 5.1 Hz,
1 H each, CH of cym), 3.45 and 3.09 (m, 1 H each, NCHMe2), 3.27
(d, 3JH,H = 10.2 Hz, 1 H, NH), 2.79 (m, 1 H, CHMe2 of cym), 2.75, 2.42,
and 2.35 (s, 3 H each, ArMe), 2.10 (s, 3 H, Me of cym), 1.54 and 1.24
3
(d, JH,H = 6.3 Hz, 3 H, NCHMe2 or CHMe2 of cym), 1.31 and 1.27 (d,
3JH,H = 6.9 Hz, 3 H, NCHMe2 or CHMe2 of cym), 1.15 (d, 3JH,H = 6.2 Hz,
3 H, NCHMe2 or CHMe2 of cym), 0.59 (d, 3JH,H = 6.6 Hz, 3 H, NCHMe2
or CHMe2 of cym) ppm. 13C{1H} NMR (C6D6): δ= 167.7 (s, CN3), 143.9,
134.3, 132.2, and 131.8 (s, Carom), 129.7 and 128.7 (s, CHarom), 92.0
and 84.0 (s, C of cym), 69.6, 69.4, and 68.5 (s, 2 C, CH of cym), 45.0
and 44.1 (s, NCHMe2), 31.8 (s, CHMe2 of cym), 25.8, 25.4, 24.5, 23.7,
22.8, and 22.5 (s, NCHMe2 and CHMe2 of cym), 20.9, 20.7, 19.2, and
18.8 (s, ArMe and Me of cym) ppm. C26H40ClN3Os (620.30): calcd. C
50.34, H 6.50, N 6.77; found C 50.45, H 6.43, N 6.90.
X-ray Crystal Structure Determinations of Complexes 3a, 3d,
and 3h: Crystals of 3a, 3d, and 3h suitable for X-ray diffraction
analysis were obtained by cooling a saturated solution of the corre-
sponding complex in benzene. The most relevant crystal and refine-
ment data are collected in Table 4. In all the cases, data collection
was performed with a Rigaku-Oxford Diffraction Xcalibur Onyx
Nova single-crystal diffractometer using Cu-Kα radiation (λ =
1.5418 Å). Images were collected at a fixed crystal-detector distance
Eur. J. Inorg. Chem. 2016, 393–402
www.eurjic.org
399
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
of 62 mm by using the oscillation method with 1.20° oscillation for
3a, 1.00° for 3d, and 1.20° for 3h, and 1.25–2.50 s variable exposure
time per image for 3a, 10.00–60.03 s for 3d, and 3.50–11.00 s for 3h.
The data collection strategy was determined by using the program
of the refinement. The PLATON[39] TwinRotMat algorithm proposed
a twin law [a rotation axis (0 0 1) in reciprocal space] for the crystals
of this complex, most likely because the b and c axes are almost
equal. Although the fractional contribution of one of the domains
is small (2.4 %), the use of the twin law significantly improved the
results, especially in the analysis of the variance. K, defined as
CrysAlisPro [36]
Data reduction and cell refinement were also per-
.
formed with the program CrysAlisPro[36] and an empirical absorption
correction was applied by using the SCALE3 ABSPACK algorithm as
2
mean(Fo2)/mean(Fc ), differs markedly from unity for the weak re-
implemented in the program CrysAlisPro [36]
.
The presence of os-
flections if the twin law is suppressed. In all cases, the maximum
residual electron density is located near to osmium atoms. Complex
3d deserves special attention. Four peaks greater than 1 e Å–3 were
mium atoms remarkably increases the absorption coefficients
(11.138, 9.357, and 10.668 mm–1 for 3a, 3d, and 3h, respectively).
To prevent errors due to different path lengths through the crystal found (3.62, 3.41, 2.57, and 1.48 e Å–3). These values are certainly
for different reflections, an additional numerical absorption correc- not too high for a heavy atom structure. The last two peaks are
tion based on gaussian integration over a multifaceted crystal
model was applied also by using the SCALE3 ABSPACK algorithm.
All the structures were solved by Patterson interpretation and phase
very close to osmium and can be discarded without further consid-
eration, but the first two are quite distant. A careful analysis showed
that the four peaks are aligned (parallel to the a axis) between
expansion using DIRDIF2008.[37] Isotropic least-squares refinements two osmium atoms of different unit cells. Inspection of the R value
were performed on F2 by using SHELXL2014[38]. During the final
stages of the refinements, all the positional and anisotropic dis-
statistics as a function of resolution confirmed that, in this crystal
structure, the strong reflections are much more affected by errors
placement parameters of all non-hydrogen atoms were refined (ex- than the rest (this fact does not occur in the other two structures).
cept fluorine atoms of a highly disordered trifluoromethyl group Both observations, and the lack of a reasonable chemical interpreta-
found in complex 3d; these fluorine atoms were isotropically re- tion, led to the conclusion that these are spurious peaks (ripples),
fined, with the help of appropriate restraints, by using the two pos- caused by the inaccuracy of the strong reflections when introduced
sible sites suggested by SHELXL2014). The hydrogen atoms were
into the Fourier synthesis to obtain the electron density function.
geometrically located and their coordinates were refined riding on The function minimized was [Σw(Fo – Fc ) /Σw(Fo2)2]1/2, in which
2
2 2
their parent atoms. Four molecules per unit cell of benzene were
found in the crystals of complex 3d. Restraints on distances and
displacement parameters were used to improve the convergence
w = 1/[σ2(Fo2) + (aP)2 + bP] (a and b values are given in Table 4)
2
with σ(Fo2) from counting statistics and P = [max(Fo2,0) + 2Fc ]/3.
Atomic scattering factors were taken from International Tables for
Table 4. Crystal data and structure refinement details for 3a, 3d, and 3h.
3a
3d
3h
Chemical formula
Mr
OsC23H34N3Cl
578.18
OsC24H33F3N3Cl·1/2C6H6
685.24
OsC24H36N3Cl
592.21
T [K]
293(2)
293(2)
293(2)
Wavelength [Å]
1.5418
1.5418
1.5418
Crystal system
Space group
Crystal size [mm]
a [Å]
b [Å]
c [Å]
α[°]
ꢀ[°]
γ[°]
triclinic
monoclinic
I2/a
triclinic
P1
P1
¯
¯
0.35 × 0.24 × 0.18
12.0265(4)
12.4358(5)
17.2701(6)
70.253(3)
87.880(3)
81.011(3)
4
0.20 × 0.06 × 0.02
11.5298(3)
22.9365(5)
22.5089(4)
90
99.301(2)
90
8
0.22 × 0.12 × 0.04
11.9464(5)
12.8170(5)
17.6562(6)
70.005(4)
87.315(3)
81.120(3)
4
Z
V [Å3]
2400.8(2)
1.600
11.138
5874.3(2)
1.550
9.357
2510.0(2)
1.567
10.668
ρcalcd. [g cm–3
]
μ[mm–1
F(000)
]
1144
2712
1176
θrange [°]
Index ranges
3.72 to 69.56
–14 ≤ h ≤ 13
–14 ≤ k ≤ 15
–19 ≤ l ≤ 20
98.1
20881
8866 (Rint = 0.0454)
541/0
3.85 to 69.67
–13 ≤ h ≤ 13
–19 ≤ k ≤ 27
–27 ≤ l ≤ 26
96.2
3.71 to 69.67
–14 ≤ h ≤ 12
–14 ≤ k ≤ 15
–18 ≤ l ≤ 21
98.1
22231
9301 (Rint = 0.0363)
539/0
Completeness to θmax [%]
Number of data collected
Number of unique data
Number parameters/restraints
Refinement method
15656
5448 (Rint = 0.0363)
322/38
Full-matrix least-squares on F2
1.099
Goodness of fit on F2
Weight function (a, b)
R [F2 > 2σ(F2)][a]
1.062
0.0788, 0.0000
0.0448
0.1122
0.0475
1.017
0.0483, 0.0000
0.0321
0.0809
0.0390
0.1453, 20.7181
0.0635
0.2038
0.0680
0.2133
wR(F2) [F2 > 2σ(F2)][a]
R (all data)
wR(F2) (all data)
0.1156
2.130 and –2.483
0.0866
1.374 and –1.237
Largest diff peak and hole [e Å–3
]
3.624 and –0.987
[a] R = Σ|Fo – Fc|/Σ|Fo|; wR(F2) = {Σ[w(Fo – Fc2)2]/Σ[w(Fo2)2]}1/2
.
2
Eur. J. Inorg. Chem. 2016, 393–402 www.eurjic.org
400
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Crystallography, Volume C.[40] Geometrical calculations were carried
out by using PLATON.[39] The crystallographic plots were drawn by
using ORTEP3.[41]
[6] See, for example: a) N. J. Bremer, A. B. Cutcliffe, M. F. Farona, J. Chem.
Soc. D 1970, 932–933; b) S. R. Foley, G. P. A. Yap, D. S. Richeson, Chem.
Commun. 2000, 1515–1516; c) S. R. Foley, G. P. A. Yap, D. S. Richeson,
Inorg. Chem. 2002, 41, 4149–4157; d) S. D. Bunge, J. A. Bertke, T. L. Cle-
land, Inorg. Chem. 2009, 48, 8037–8043; e) L. Fohlmeister, S. Liu, C. Schul-
ten, B. Moubaraki, A. Stasch, J. D. Cashion, K. S. Murray, L. Gagliardi, C.
Jones, Angew. Chem. Int. Ed. 2012, 51, 8294–8298; Angew. Chem. 2012,
124, 8419–8423; f) L. Fohlmeister, C. Jones, Aust. J. Chem. 2014, 67, 1011–
1016; g) A. K. Maity, A. J. Metta-Magaña, S. Fortier, Inorg. Chem. 2015,
54, 10030–10041.
CCDC 1432851 (for 3a), 1432852 (for 3d), and 1432853 (for 3h)
contain the supplementary crystallographic data for this paper.
These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre.
[7] a) R. Clérac, F. A. Cotton, L. M. Daniels, J. P. Donahue, C. A. Murillo, D. J.
Timmons, Inorg. Chem. 2000, 39, 2581–2584; b) F. A. Cotton, N. S. Dalal,
P. Huang, C. A. Murillo, A. C. Stowe, X. Wang, Inorg. Chem. 2003, 42, 670–
672; c) F. A. Cotton, G. M. Chiarella, N. S. Dalal, C. A. Murillo, Z. Wang,
M. D. Young, Inorg. Chem. 2010, 49, 319–324.
Acknowledgments
This work was financially supported by the Spanish Ministerio
de Economía y Competitividad (MINECO) (projects CTQ2013-
40591-P and CTQ2014-51912-REDC), the Gobierno del Princi-
pado de Asturias (project GRUPIN14-006) and the Junta de Co-
munidades de Castilla-La Mancha (project PEII-2014-041-P).
L. M.-R. thanks the European Social Fund (ESF) for the award of
an FPI fellowship.
[8]
For complexes 3a and 3h, two independent molecules were found in
the asymmetric unit featuring very similar structural parameters. For
brevity, only one of them in each case is presented in Figure 3 and only
their bond lengths and angles are presented in Table 1.
Shorter Os–N distances (average 2.042 Å) were also found in the struc-
ture of trans-[OsCl2{κ2-(N,N′)-CEt(NPh)2}2], which is the only amidinate–
osmium complex characterized so far by single-crystal X-ray diffraction
techniques, see: T. Clark, S. D. Robinson, D. A. Tocher, J. Chem. Soc., Dal-
ton Trans. 1992, 3199–3202.
[9]
Keywords: Homogeneous catalysis · Osmium · N ligands ·
Aldoximes · Cyanides
[10]
For reviews on the redox isomerization of allylic alcohols, see: a) R. Uma,
C. Crévisy, R. Grée, Chem. Rev. 2003, 103, 27–51; b) R. C. van der Drift, E.
Bouwman, E. Drent, J. Organomet. Chem. 2002, 650, 1–24; c) V. Cadierno,
P. Crochet, J. Gimeno, Synlett 2008, 1105–1124; d) P. Lorenzo-Luis, A.
Romerosa, M. Serrano-Ruiz, ACS Catal. 2012, 2, 1079–1086; e) J. García-
Álvarez, S. E. García-Garrido, P. Crochet, V. Cadierno, Curr. Top. Catal.
2012, 10, 35–56.
[1] G. Chandra, A. D. Jenkins, M. F. Lappert, R. C. Srivastava, J. Chem. Soc. A
1970, 2550–2558.
[2] For review articles on this field, see: a) P. J. Bailey, S. Pace, Coord. Chem.
Rev. 2001, 214, 91–141; b) W. E. Piers, D. J. H. Emslie, Coord. Chem. Rev.
2002, 233–234, 131–155; c) F. T. Edelmann, Adv. Organomet. Chem. 2008,
57, 183–352; d) F. T. Edelmann, Chem. Soc. Rev. 2009, 38, 2253–2268; e)
C. Jones, Coord. Chem. Rev. 2010, 254, 1273–1289; f) A. A. Trifonov, Coord.
Chem. Rev. 2010, 254, 1327–1347; g) S. Collins, Coord. Chem. Rev. 2011,
255, 118–138; h) F. T. Edelmann, Chem. Soc. Rev. 2012, 41, 7657–7672; i)
F. T. Edelmann, Adv. Organomet. Chem. 2013, 61, 55–374; j) S. T. Barry,
Coord. Chem. Rev. 2013, 257, 3192–3201; k) A. Kurec, P. G. Gordon, S.
Karle, A. Devi, S. T. Barry, Aust. J. Chem. 2014, 67, 989–996; l) T. Chlupatý,
A. Růžička, Coord. Chem. Rev. 2015, in press, DOI: http://dx.doi.org/
10.1016/j.ccr.2015.07.012.
[3] For recent reviews on the synthesis of guanidines, see: a) C. Alonso-
Moreno, A. Antiñolo, F. Carrillo-Hermosilla, A. Otero, Chem. Soc. Rev.
2014, 43, 3406–3425; b) B. Lemrová, M. Soural, Eur. J. Org. Chem. 2015,
1869–1886; c) W.-X. Zhang, L. Xu, Z. Xi, Chem. Commun. 2015, 51, 254–
265; d) S. Tahir, A. Badshah, R. A. Hussain, Bioorg. Chem. 2015, 59, 39–
79.
[11]
[12]
[13]
[14]
M. Batuecas, M. A. Esteruelas, C. García-Yebra, E. Oñate, Organometallics
2010, 29, 2166–2175.
E. Putignano, G. Bossi, P. Rigo, W. Baratta, Organometallics 2012, 31,
1133–1142.
See, for example: Y. Song, D. Shen, Q. Zhang, B. Chen, G. Xu, Tetrahedron
Lett. 2014, 55, 639–641, and references cited therein.
a) K. Ishihara, Y. Furuya, H. Yamamoto, Angew. Chem. Int. Ed. 2002, 41,
2983–2986; Angew. Chem. 2002, 114, 3109–3112; b) Y. Furuya, K. Ishihara,
H. Yamamoto, Bull. Chem. Soc. Jpn. 2007, 80, 400–406.
[15]
a) J. Hart-Davis, P. Battioni, J.-L. Boucher, D. Mansuy, J. Am. Chem. Soc.
1998, 120, 12524–12530; b) A. Rapeyko, M. J. Climent, A. Corma, P. Con-
cepción, S. Iborra, ChemSusChem 2015, 8, 3270–3282.
[16]
[17]
a) S. H. Yang, S. Chang, Org. Lett. 2001, 3, 4209–4211; b) E. Choi, C. Lee,
Y. Na, S. Chang, Org. Lett. 2002, 4, 2369–2371.
P. Tamilselvan, Y. B. Basavaraju, E. Sampathkumar, R. Murugesan, Catal.
Commun. 2009, 10, 716–719.
[4] a) R. García-Álvarez, F. J. Suárez, J. Díez, P. Crochet, V. Cadierno, A. Anti-
ñolo, R. Fernández-Galán, F. Carrillo-Hermosilla, Organometallics 2012,
31, 8301–8311; b) L. Menéndez-Rodríguez, E. Tomás-Mendivil, J. Francos,
P. Crochet, V. Cadierno, A. Antiñolo, R. Fernández-Galán, F. Carrillo-Her-
mosilla, Organometallics 2015, 34, 2796–2809.
[18]
[19]
K. Kaneda, K. Doken, T. Imanaka, Chem. Lett. 1988, 285–286.
a) D. R. Zuidema, A. L. Dennison, E. Y. Park, R. C. Mebane, Synth. Commun.
2008, 38, 3810–3815; b) Y.-T. Li, B.-S. Liao, H.-P. Chen, S.-T. Liu, Synthesis
2011, 2639–2643.
[20]
[21]
a) H. S. Kim, S. H. Kim, J. N. Kim, Tetrahedron Lett. 2009, 50, 1717–1719;
b) K. Tambara, G. D. Pantoş, Org. Biomol. Chem. 2013, 11, 2466–2472; c)
L. Menéndez-Rodríguez, E. Tomás-Mendivil, J. Francos, C. Nájera, P. Cro-
chet, V. Cadierno, Catal. Sci. Technol. 2015, 5, 3754–3761.
a) O. Attanasi, P. Palma, F. Serra-Zanetti, Synthesis 1983, 741–742; b) N.
Jiang, A. J. Ragauskas, Tetrahedron Lett. 2010, 51, 4479–4481; c) X.-Y. Ma,
Y. He, T.-T. Lu, M. Lu, Tetrahedron 2013, 69, 2560–2564.
a) M. H. Sarvari, Synthesis 2005, 787–790; b) S. Enthaler, M. Weidauer, F.
Schröder, Tetrahedron Lett. 2012, 53, 882–885.
P. Yan, P. Batamack, G. K. S. Prakash, G. A. Olah, Catal. Lett. 2005, 101,
141–143.
a) D. C. Barman, A. J. Thakur, D. Prajapati, J. S. Sandhu, Chem. Lett. 2000,
1196–1197; b) C. L. Allen, C. Burel, J. M. J. Williams, Tetrahedron Lett.
2010, 51, 2724–2726.
Transition-metal complexes are known to promote the Beckmann-type
rearrangement of aldoximes to primary amides. For a recent review, see:
P. Crochet, V. Cadierno, Chem. Commun. 2015, 51, 2495–2505.
For example, by using the ruthenium(II) dimer [{RuCl(μ-Cl)(η6-p-
cymene)}2] as catalyst, the addition of molecular sieves to the reaction
[5] See, for example: a) P. J. Bailey, L. A. Mitchell, S. Parsons, J. Chem. Soc.,
Dalton Trans. 1996, 2839–2841; b) J. L. Bear, Y. Li, B. Han, K. M. Kadish,
Inorg. Chem. 1996, 35, 1395–1398; c) M. B. Dinger, W. Henderson, B. K.
Nicholson, J. Organomet. Chem. 1998, 556, 75–88; d) K. T. Holman, S. D.
Robinson, A. Sahajpal, J. W. Steed, J. Chem. Soc., Dalton Trans. 1999, 15–
18; e) P. J. Bailey, K. J. Grant, L. A. Mitchell, S. Pace, A. Parkin, S. Parsons,
J. Chem. Soc., Dalton Trans. 2000, 1887–1891; f) S. D. Robinson, A. Sahaj-
pal, J. Steed, Inorg. Chim. Acta 2000, 303, 265–270; g) F. A. Cotton, C. A.
Murillo, X. Wang, C. C. Wilkinson, Inorg. Chim. Acta 2003, 351, 191–200;
h) F. A. Cotton, C. A. Murillo, J. H. Reibenspies, D. Villagrán, X. Wang, C. C.
Wilkinson, Inorg. Chem. 2004, 43, 8373–8378; i) J. S. Pap, J. L. Snyder,
P. M. B. Piccoli, J. F. Berry, Inorg. Chem. 2009, 48, 9846–9852; j) G. M.
Chiarella, F. A. Cotton, C. A. Murillo, M. D. Young, Q. Zhao, Inorg. Chem.
2010, 49, 3051–3056; k) R. Lee, Y. Yang, G. K. Tan, C.-H. Tan, K.-W. Huang,
Dalton Trans. 2010, 39, 723–725; l) T. Singh, R. Kishan, M. Nethaji, N.
Thirupathi, Inorg. Chem. 2012, 51, 157–169; m) R. Kishan, R. Kumar, S.
Baskaran, C. Sivasankar, N. Thirupathi, Eur. J. Inorg. Chem. 2015, 3182–
3194; n) G. M. Chiarella, C. A. Murillo, M. D. Young, Polyhedron 2016, 103,
15–20.
[22]
[23]
[24]
[25]
[26]
Eur. J. Inorg. Chem. 2016, 393–402
www.eurjic.org
401
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
media, to trap the water released in the process, was needed to achieve
good conversions and selectivity (see ref.[16]).
mosilla, I. López-Solera, A. Otero, A. Serrano-Laguna, E. Villaseñor, J. Orga-
nomet. Chem. 2012, 711, 35–42.
[32] Coordination of the aldoxime substrate to N would explain the forma-
tion of the minor amounts of the rearrangement-derived amide pro-
ducts. Due to the large excess of solvent over the substrate, nitrile ex-
change is expected to occur very quickly, thereby disfavoring this com-
peting reaction pathway.
[33] R. Castarlenas, M. A. Esteruelas, E. Oñate, Organometallics 2005, 24,
4343–4346.
[34] C. Alonso-Moreno, F. Carrillo-Hermosilla, A. Garcés, A. Otero, I. López-
Solera, A. M. Rodríguez, A. Antiñolo, Organometallics 2010, 29, 2789–
2795.
[35] a) R. García-Álvarez, A. E. Díaz-Álvarez, J. Borge, P. Crochet, V. Cadierno,
Organometallics 2012, 31, 6482–6490; b) P. J. González-Liste, V. Cadierno,
S. E. García-Garrido, ACS Sustainable Chem. Eng. 2015, 3, 3004–3011.
[36] CrysAlisPro: Data Collection and Processing Software for Agilent X-ray Dif-
fractometers, XRD Products, Agilent Technologies, Oxford, UK, 2013.
[37] P. T. Beurskens, G. Beurskens, R. de Gelder, S. García-Granda, R. O. Gould,
J. M. M. Smits, The DIRDIF2008 Program System, Crystallography Labora-
tory, University of Nijmegen, The Netherlands, 2008.
[38] G. M. Sheldrick, Acta Crystallogr., Sect. C 2015, 71, 3–8.
[39] A. L. Spek, Acta Crystallogr., Sect. D 2009, 65, 148–155.
[40] A. J. C. Wilson (Ed.), International Tables for X-ray Crystallography, volume
C, Kluwer Academic Publishers, Dordrecht, The Netherlands, 1992.
[41] L. J. Farrugia, J. Appl. Crystallogr. 2012, 45, 849–854.
[27]
[28]
Most of the aldoximes included in Table 3 were synthesized as mixtures
of the corresponding E and Z isomers in ratios ranging from 95:5 to
30:70. Differences in the reactivity of the two stereoisomers were not
observed. Monitoring of the reactions by GC showed that E and Z iso-
mers are consumed at similar rates.
Although complex 3d is not the most efficient catalyst described to date
for this catalytic transformation, its activity is comparable or surpasses
that of the known Ru-, Pd-, Cu-, Zn-, or Ga-based systems, which require
metal loadings of 4–10 mol-% to deliver the nitriles in high yields.
[29] As a representative example, when the model (E)-benzaldoxime was de-
hydrated in toluene with 5 mol-% of 3d, only 36 % of benzonitrile was
formed after 24 h of heating at 80 °C.
[30] The NMR spectra of 3a–m recorded in [D3]acetonitrile indicate that an
equilibrium between these complexes and the corresponding solvates
K is established in solution.
[31] The vacant coordination site for the coordination of the aldoxime is most
probably generated by a change in the coordination mode of the guan-
idinate ligand [from κ2-(N,N′) to κ1-(N)]. Dissociation of the p-cymene
ligand seems unlikely because the catalytic reactions are not affected
when carried out in the presence of an excess of free p-cymene
(100 equiv. per Os). The Hg0 test also ruled out the formation of catalyti-
cally active nanoparticles. Examples of changes in the hapticity of coor-
dinated guanidinate ligands can be found, among others, in ref.[5l,5m]
and: a) S. M. Mullins, A. P. Duncan, R. G. Bergman, J. Arnold, Inorg. Chem.
2001, 40, 6952–6963; b) R. Fernández-Galán, A. Antiñolo, F. Carrillo-Her-
Received: October 23, 2015
Published Online: December 21, 2015
Eur. J. Inorg. Chem. 2016, 393–402
www.eurjic.org
402
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim