First silver complexes of tetrahydropyrimid-2-ylidenes

Article abstract of DOI:10.1016/j.jorganchem.2004.04.032

The syntheses and characterizations of homoleptic silver(I) carbene complexes of the type LAgCl (3) and [L₂Ag]⁺ ₂[Pd₂Cl₆]^2- (4) (L=dimesityltetrahydropyrimid-2-ylidene) are described. 3 was obtained by reaction of the corresponding tetrahydropyrimidinium salt with silver(I) oxide. Subsequent reaction of 3 with Pd(CH₃CN) ₂Cl₂ afforded complex 4. The crystal structure of 3 has been determined to be a monomer in the solid state. The ¹³C NMR spectra of both 3 and 4 exhibit ¹³ C-^107,109Ag coupling in solution. Furthermore, both compounds show a downfield shift of the carbene carbon of about 20 ppm to about 205 ppm in the ¹³C NMR compared to five-membered ring silver(I) carbene complexes. Unlike other silver-carbene complexes, 3 does not undergo ligand exchange and can thus not be used for carbene transfer to other metals.

Full text of DOI:10.1016/j.jorganchem.2004.04.032

Journal of Organometallic Chemistry 689 (2004) 2441–2449
www.elsevier.com/locate/jorganchem
q,qq
First silver complexes of tetrahydropyrimid-2-ylidenes
a,
Wolfgang A. Herrmann , Sabine K. Schneider , Karl Ofele , Masato Sakamoto ,
a
a
b
*
€
a
Eberhardt Herdtweck
a
Anorganisch-chemisches Institut der Technischen Universit €a t M €u nchen, D-85747 Garching bei M €u nchen, Germany
Zeon Corporation R&D Center, 1-2-1 Yako Kawasaki-ku, 210-0863 Kawasaki, Japan
b
Received 15 March 2004; accepted 25 April 2004
Abstract
þ
2ꢀ
The syntheses and characterizations of homoleptic silver(I) carbene complexes of the type LAgCl (3) and [L2Ag] [Pd2Cl6] (4)
2
(
L ¼ dimesityltetrahydropyrimid-2-ylidene) are described. 3 was obtained by reaction of the corresponding tetrahydropyrimidinium
salt with silver(I) oxide. Subsequent reaction of 3 with Pd(CH3CN)2Cl2 afforded complex 4. The crystal structure of 3 has been
determined to be a monomer in the solid state. The 13C NMR spectra of both 3 and 4 exhibit 13C–107;109Ag coupling in solution.
Furthermore, both compounds show a downfield shift of the carbene carbon of about 20 ppm to about 205 ppm in the 13C NMR
compared to five-membered ring silver(I) carbene complexes. Unlike other silver–carbene complexes, 3 does not undergo ligand
exchange and can thus not be used for carbene transfer to other metals.
ꢀ
2004 Elsevier B.V. All rights reserved.
Keywords: Silver; Palladium; Tetrahydropyrimid-2-ylidenes
1
. Introduction
substituted tetrahydropyrimid-2-ylidene ligands can be
successfully employed as catalysts in olefin metathesis
reactions [5].
N-Heterocyclic carbenes have become universal li-
gands in organometallic and inorganic coordination
chemistry [1,2]. There is current interest in novel,
stronger donating N-heterocyclic carbene ligands. The
acyclic bis(dialkylamino)carbenes published by Alder et
al. [3] turned out to be the strongest r-donating carbene
ligands so far [4]. Due to their six-membered ring ge-
ometry tetrahydropyrimid-2-ylidenes are supposed to be
comparable to those systems [4].
A very efficient way to obtain NHC metal complexes
has been published by Wang and Lin [6]. They described
the use of silver(I) complexes which are formed easily by
treatment of imidazolium salts with Ag2O. These com-
pounds are able to transfer their NHC ligand to other
metals for e.g. gold or palladium [7].
Silver(I) carbene complexes display a wide variety of
structural motives depending on the size of the coordi-
nated carbene [8]. These species tend to oligomerize via
silver–silver interactions, bridging halide ligands, or
even Ag–(l2-Ccarbene)–Ag interactions [9]. Monomeric
examples of such complexes are rare. In fact, only two
monomeric carbene complexes of AgCl have been
characterized to date [10].
Recently our group has shown, that Grubbs type
ruthenium complexes with isopropyl and mesityl
q
Supplementary data associated with this article can be found, in the
online version, at doi:10.1016/j.jorganchem.2004.04.032.
In recent publications, Buchmeiser and co-workers
11] address the synthesis and catalytic properties of
tetrahydropyrimid-2-ylidene complexes of palladium.
Significant differences between their report and our on-
going studies prompted us to publish our results at this
time.
[
qq
N-heterocyclische carbene, Part 37. – Part 36, see [7b].
Corresponding author. Tel.: +49-89-289-13080; fax: +49-89-289-
*
1
3473.
E-mail address: lit@arthur.anorg.chemie.tu-muenchen.de (W.A.
Herrmann).
0
022-328X/$ - see front matter ꢀ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.04.032

2442
W.A. Herrmann et al. / Journal of Organometallic Chemistry 689 (2004) 2441–2449
2
. Results and discussion
(2, 3) or reddish brown (4), crystalline materials, which
are readily soluble in most common polar organic sol-
vents (DCM, THF), but insoluble in hexane or diethyl
ether. The carbene compounds 3 and 4 are stable against
daylight and thermally stable at room temperature. The
molecular structures of 2, 3 and 4 were determined by
X-ray diffraction.
Additionally, the compounds were characterized by
1H and 13C NMR spectroscopy, elemental analysis and
FAB-MS. The spectroscopic data obtained for 2, 3 and
4 are in accord with the crystallographically determined
structures.
In the case of 3 and 4, direct information concerning
the Ag–C interaction were obtained from the 13C NMR
spectra, since 107Ag and 13C–109Ag coupling is observed
for the carbene carbon atom (3: 228.4 and 260.4 Hz, 4:
of 174.3 and 201.3 Hz) (Fig. 1).
This information directly confirms that neither 3 nor
4 show a lability of the Ag–carbene bond. This is un-
expected, as most of the up to now published com-
pounds of the kind (carbene)AgX show a fluxional
behaviour described by Wang and Lin [6], that results in
a lack of C–Ag coupling. This aspect and the downfield
shift of the carbene carbon atom in the 13C NMR
spectra of about 20 ppm from 180 to 200 ppm (205.9
ppm (3), 205.8 ppm (4)) with respect to literature data
[6–10] is probably a result of the stronger metal–carbene
bond in tetrahydropyrimid-2-ylidene complexes com-
pared to imidazol-2-ylidene complexes. Therefore the
transfer of the carbene ligand to other metals has failed.
The work of Buchmeiser and co-workers [11a] misses
the 13C NMR signal for the carbon atom in 8, making it
impossible to distinguish between an Ag–carbene and a
Pd–carbene bond.
2.1. Synthesis and characterization of the ligand and it’s
silver complexes
The synthesis of the tetrahydropyrimidinium salt 2 is
achieved in two steps (Scheme 1). Starting from mesityl
bromide and 1,3-diaminopropane, the corresponding
mesityl substituted diamine 1 is prepared by a palladium
catalyzed arylamination. Subsequent reaction of the
diamine 1 with triethyl orthoformiate and ammonium
tetrafluoroborate according to Saba and Kaloustian [12]
yielded the tetrahydropyrimidinium salt 2. The silver(I)
carbene complex 3 was obtained after stirring tetrahy-
dropyrimidinium salt 2 in CH Cl with Ag O for 4 h
2
2
2
and workup of the filtered solution in 78% yield. This
compound was expected to serve as a carbene transfer
agent to obtain the corresponding Pd(II) compounds.
Therefore 3 was treated with Pd(CH3CN)2Cl2 according
to a literature procedure for comparable imidazol-2-
ylidene compounds [7c] to obtain a biscarbene Pd(II)
compound according to Scheme 2. Yet, we obtained the
biscarbene Ag(I) compound 4 (Scheme 2) which was
identified unambitiously by thorough analysis (see be-
low). Thus the Ag(I) carbene 3 did not transfer it’s
carbene ligand to Pd but it’s chloride ligand and a sec-
ond equivalent of 3 then obviously transferred it’s car-
bene ligand to the silver monocarbene unit to build a
biscarbene Ag(I) cation. The reactivity is therefore re-
versed to known examples with imidazol-2-ylidene
compounds. In contrast to our results, Buchmeiser and
co-workers [11a], who carried out the same reaction,
claimed to obtain the biscarbene palladium(II) complex
2
þ
4ꢀ
[
L2Pd] [Ag2Br2Cl6] (L ¼ dimesityltetrahydropyrimid-
2
-ylidene) 8. We will demonstrate below, that this claim
2
is unsound.
As expected, the mass spectrum of compound 2
contains only one peak, the tetrahydropyrimidinium
cation. For the Ag(I) carbene complexes 3 and 4, the
most prominent peak in the mass spectra is a [(car-
2
, 3 and 4 can be obtained after recrystallization from
acetone/toluene (2) or DCM/pentane (3, 4) as colorless
Pd dba *CHCl
2
3
3
t
BINAP / NaO Bu
H N
NH2
2
Br
N
H
N
H
2
1
40˚C/100 bar N2
toluene
1
BF4
N
N
N
N
HC(OEt)3 / NH BF
Ag O / NMe Cl
2 4
4
4
H
Ag
CH2Cl2
Cl
2
3
Scheme 1. Synthesis of 1, 2 and 3.

W.A. Herrmann et al. / Journal of Organometallic Chemistry 689 (2004) 2441–2449
2443
N
N
Cl
N
2
N
N
2
AgCl
Pd(CH CN) Cl
3
2
2
Ag
Cl
Pd
Cl
2
3
CH CN
N
3
2
CH CN
3
AgCl
2
N
N
N
Cl
Cl
Cl
Cl
1
/2
+ 1/2
Pd
Pd
Ag
Cl
Cl
N
2
4
Scheme 2. Synthesis of the homoleptic tetrahydropyrimid-2-ylidene silver(I) complex 4.
Fig. 1. 13C–107;109Ag coupling observed for 4 (solvent: CD2Cl2).
þ
bene)2Ag] fragment, which in the case of 3 is obviously
formed under FAB-MS conditions. A second peak that
also part in the FAB-MS of 4. In both spectra (for 3 and
4) the isotope patterns are in accordance with silver
carbene fragments (Fig. 2). In contrast, palladium car-
bene fragments as derived from compound 8 would
exhibit a quite different isotope pattern.
þ
belongs to the [(carbene)Ag] cation, indicates the mo-
lecular structure of the type (carbene)AgX already de-
termined by X-ray diffraction analysis for 3. This peak is

2444
W.A. Herrmann et al. / Journal of Organometallic Chemistry 689 (2004) 2441–2449
þ
þ
Fig. 2. Details of the FAB-MS spectrum of 4. Measured and calculated isotope patterns for the [(carbene)2Ag] fragment in 4 versus [(carbene)2Pd] .
2
.2. Molecular structures of 2, 3 and 4
examples of such silver(I) carbene complexes are simi-
lar to [Ag(Et2-bimy)2][AgBr2] (bimy ¼ 1,3-diethylbenz-
imidazol-2-ylidene), the first Ag(I) carbene complex
synthesized via the Ag2O way [6]. As mentioned above,
monomeric examples of such complexes are rare. One of
them is the corresponding 1,3-bismesityl-imidazol-2-
ylidene silver(I) chloride 6 reported by Abernethy and
co-workers [10b] (Fig. 3) which is compared with 3 in
Table 1.
The complexes 3 and 6 (Fig. 6) enable us to compare
the structural differences between imidazolin-2-ylidene
complexes and tetrahydropyrimid-2-ylidene complexes.
The transition from the five-membered to the six-mem-
bered ring geometry causes a widening of the N–
Ccarbene–N–bond angle from about 105ꢁ in imidazole
complexes to 124.72(15)ꢁ in 3. This results in a com-
pression of the a-angle Cmesityl–N–Ccarbene from 121.8(3)ꢁ
The related compounds 5, 6 and 7 have been previ-
ously reported from different synthetic routes and serve
as interesting comparisons [13,10b,14].
The molecular structure of the tetrahydropyrimidi-
nium salt 2 in the solid state is depicted in Fig. 4 being
the first structure of a tetrahydropyrimidinium salt with
aromatic substituents on both nitrogen atoms. It is very
similar to that of the corresponding diisopropyltetra-
hydropyrimidinium salt reported by Alder et al. (Table
1
) [13]. The detected N–Ccarbene–N-angles and Ccarbene–N
distances of 2 are nearly identical to the one published
by Alder [13].
The molecular structure of 3 in the solid state is
shown in Fig. 5. It is up to now the first structure of a
silver(I) tetrahydropyrimid-2-ylidene complex. Most
BF4
N
N
N
N
N
N
N
N
CF3SO3-
Ag
H
Ag
Cl
5
6
7
Fig. 3. Related compounds for comparison.

W.A. Herrmann et al. / Journal of Organometallic Chemistry 689 (2004) 2441–2449
2445
compressed from average 125.4(4)ꢁ [14] to average
1
18.4(2)ꢁ (Table 1). Like in the monomeric Ag(I) car-
bene complex 3 discussed above, the Ccarbene–Ag dis-
ꢀ
tances of 2.101(3) and 2.097(3) A are longer than those
reported for the corresponding imidazol-2-ylidene com-
ꢀ
pound 2.067(4) and 2.078(4) A [14] (Table 1), indicating
once more the higher steric bulk of the six-membered
ligand.
The silver atom essentially is linear coordinated
(
178.33(10)ꢁ) as it is known from the previously reported
Fig. 4. ORTEP representation of the mono-cationic part of compound
in the solid state. Thermal ellipsoids are drawn at the 50% probability
level.
compound 7 (176.3(2)ꢁ) [14]. The planes of the two
tetrahydropyrimid-2-ylidene rings are twisted by
8
2
6.1(4)ꢁ from coplanarity (39.5(7)ꢁ in 7). This orienta-
tion of the two tetrahydropyrimid-2-ylidene rings is
likely the result of steric repulsion between the mesityl
substituents on the nitrogens.
In summary, our analytical data prove that the re-
action of the tetrahydropyrimid-2-ylidene silver(I)
complex 3 with Pd(CH3CN)2Cl2 yields the biscarbene
silver(I) complex 4 as depicted in Scheme 2 and not the
palladium(II) complex 8 published by Buchmeiser and
co-workers [11a]. Our assignment is based on the fol-
lowing facts:
Fig. 5. ORTEP representation of compound 3 in the solid state.
Thermal ellipsoids are drawn at the 50% probability level. The hy-
drogen atoms are omitted for clarity.
•
The structural data of compound 8 are very similar to
the data for the silver complex 4. It is difficult to as-
sign silver and palladium atoms positions only based
upon X-ray data without confirmation by further an-
alytical methods. However, calculations with silver in
the cationic and palladium in the anionic part of
structure 4 result in significantly better R-values com-
pared to the inverse option proposed by Buchmeiser
in the imidazole complex from Abernethy and co-
workers [10b] to 119.4(2)ꢁ and 118.5(2)ꢁ in 3 (Table 1,
ꢀ
–Ag distance of 2.095(3) A in 3 is
Fig. 6). The C
slightly longer than the one reported for the corre-
carbene
ꢀ
sponding imidazol-2-ylidene compound (2.056(7) A)
et al.
• The carbene carbon atom in the 3C NMR spectrum
1
(
Table 1) [10b]. This reflects the steric congestion around
of complex 4 exhibits C– Ag and 13C–109Ag cou-
pling. This coupling weakens the signal of the carbene
carbon atom compared to a palladium compound of
type 8. Because of this reason Buchmeiser and co-
workers may not have detected a carbene signal in
the 13C NMR.
1
3
107
the metal center for the six-membered ring ligand.
The molecular structure of 4 in the solid state is de-
picted in Fig. 7. The comparison between compound 4
and the corresponding imidazolin-2-ylidene complex 7
previously reported by Arduengo et al. [14] also reveals
the differences, caused by the transition from five-
membered to six-membered ring geometry. The N–
Ccarbene–N-bond angle is widened from 103.6(4)ꢁ and
• In the FAB-MS of 4, all fragments that include the
metal center, show an isotope pattern which is consis-
tent with a Ag containing moiety. The isotope pattern
of the postulated complex 8 is not represented in the
1
04.8(4)ꢁ [14] to 118.7(2)ꢁ and 118.0(2)ꢁ (Table 1).
Therefore again the a-angles Cmesityl–N–Ccarbene are
Table 1
ꢀ
Selected bond angles (ꢁ) and bond lengths (A) in 2, 3, 4 and related structures (for a1 and a2, see Fig. 6)
a
a
2
5 [13]
3
6 [10b]
4
7 [14]
N–Ccarbene–N
124.72(15)
120.38(11)
120.38(11)
–
125.0(2)
119.6(2)
119.6(2)
–
118.3(3)
119.4(2)
118.5(2)
–
104.4(5)
121.8(3)
121.8(3)
–
118.7(2) 118.0(2)
117.3(2) 117.9(2)
118.6(2) 119.7(2)
178.33(10)
103.6(4) 104.8(4)
125.4(4) 124.3(4)
126.3(4) 124.6(4)
176.3(2)
C
C
C
C
C
C
ipr=mes–N–Ccarbene (a
ipr=mes–N–Ccarbene (a
carbene–Ag–Ccarbene
1
)
)
2
carbene–N (a
carbene–N (a
carbene–Ag
1
2
-site)
-site)
1.3147(14)
1.3147(14)
–
1.311(3)
1.311(3)
–
1.338(4)
1.329(4)
2.095(3)
2.3213(10)
1.357(5)
1.357(5)
2.056(7)
2.314(2)
1.337(4) 1.343(4)
1.332(3) 1.332(4)
2.101(3) 2.097(3)
–
1.356(6) 1.360(6)
1.352(6) 1.3556(0)
2.067(4) 2.078(4)
–
Ag–Cl
–
–
a
In italic the corresponding values for the second carbene coordinated to the metal center.

2446
W.A. Herrmann et al. / Journal of Organometallic Chemistry 689 (2004) 2441–2449
N
N
N
N
N
N
α2
α1
α2
α1
α1
Ag
Cl
α2 Ag
Cl
Ag
6
3
N
N
4
Fig. 6. a-Angle in 3, 4 and 6.
complex as the active catalyst species. Heck coupling
reactions have been reported to be catalyzed by sim-
ple palladium(II) salts [15].
3
. Conclusion
We have synthesized and characterized first homo-
leptic tetrahydropyrimidine ylidene silver(I) complexes.
The X-ray crystallographic structures of 3 and 4 show a
nearly linear geometry for the silver center. The crystal
structure of 3 has been determined to be a monomer in
the solid state. The NMR spectra of this carbene com-
plexes exhibit coupling of the silver nucleus with the
carbene carbon atoms. Unlike many silver complexes of
this type, the silver(I) compound 3 does not undergo a
rapid ligand exchange. The signals of the carbene car-
bon atoms of 3 and 4 in the 13C NMR spectra are
downfield shifted of about 20 ppm to about 205 ppm
compared to silver(I) carbene complexes of five-mem-
bered ring systems (about 180 ppm). This aspect and the
observation that the transfer of a carbene ligand to
palladium(II) failed, indicate that the Ag–carbene bond
is much stronger than in comparable five-membered ring
systems. Therefore this new class of carbene ligands
provide stronger r-donating ligands compared to imi-
dazol-2-ylidene systems. Correspondingly, the reported
results of Buchmeiser and co-workers concerning the
palladium biscarbene 8 are wrong as we have mentioned
above. Further utilization of these compounds for the
preparation of transition metal carbene complexes and
their use in catalytic reactions is under investigation.
Fig. 7. ORTEP representation of the mono-cationic part of compound
4 ꢂ 5(CH2Cl2) in the solid state. Thermal ellipsoids are at the 50%
probability level. The hydrogen atoms are omitted for clarity.
publication of Buchmeiser and co-workers [11a].
Nevertheless, as a mono charged cation it would
show a very different isotope pattern. A double
charged [(carbene) Pd]2
þ
should be found at
2
m=z ꢁ 374.
•
•
The structural elements in compound 8, both the
[
Ag2X6]4 anion and the linear coordinated ligand
ꢀ
Pd(II) dication are without precedence in literature.
The brownish colour of complex 8 indicates rather a
4. Experimental
2
ꢀ
Pd X ] anion instead of a [Ag X ] anion. Biscar-
2 6 2
4ꢀ
[
6
bene palladium(II) compounds are colourless (cf.
compound 16 in [11a]).
The Heck coupling results obtained using 8 as a cat-
alyst don’t necessarily indicate a palladium carbene
4.1. General comments
1
•
NMR spectra ( H, 13C) were recorded either on a Jeol
JMX-GX 400, on a Jeol JMX-GX 270 or on a Bruker

W.A. Herrmann et al. / Journal of Organometallic Chemistry 689 (2004) 2441–2449
2447
DPX 400 instrument. Chemical shifts are given in ppm.
The spectra are calibrated to the residual protons of the
2.53 (q, J ¼ 5:9 Hz, 2H, CH2), 2.29 (s, 12H, MesCH3),
2.27 (s, 6H, MesCH3).
1
solvent ( H) or to the 13-carbon signals of the solvent
(13
13C{1H} (100 MHz, CDCl3): d [ppm] ¼ 150.0
(NCN), 140.6, 136.5, 134.4, 130.2 (MesC), 46.5 (NCH2),
21.0 (MesCH3), 19.4 (CH2), 17.5 (MesCH3).
Anal. Found: C, 64.45; H, 7.06; N, 6.91. Calc. for
C H BF N (408.28): C, 64.72; H, 7.16; N, 6.86.
C). NMR multiplicities are abbreviated as follows:
s ¼ singlet, d ¼ doublet, q ¼ quintet, m ¼ multiplet.
Coupling constants J are given in Hz. FAB-MS spectra
were measured at the TU M u€ nchen Mass Spectrometry
Laboratory on a Finnigan MAT 90 mass spectrometer
2
2
29
4
2
þ
MS (FAB): 322 ([M+1-BF ] ).
4
(
carried out by the Microanalytical Laboratory at the
TU M u€ nchen.
xenon/p-nitrobenzyl alcohol). Elemental analyses were
4.4. (Dimesityltetrahydropyrimid-2-ylidene) silver(I)
chloride (3)
Unless otherwise stated, all manipulations were
carried out using standard Schlenk techniques. All
solvents for use in an inert atmosphere were purified by
standard procedures and distilled under nitrogen im-
mediately prior to use. Other chemicals were obtained
from commercial sources and used without further
purification.
A mixture of 2 (350 mg, 0,86 mmol), silver(I) oxide
(109 mg, 0.47 mmol, 0.55 eq.) and NMe Cl (94 mg,
4
0.86 mol, 1 eq.) in 30 ml of DCM was stirred for 4 h.
The solution was filtered through Celite, the filtrate
evaporated to a volume of ca. 5 ml; after addition of
hexane complex 3 was obtained as a white solid (310
mg, 78%).
0
1
4
.2. N,N -Dimesitylpropanediamine (1)
H (270 MHz, CD2Cl2): d [ppm] ¼ 6.98 (s, 4H,
MesH), 3.38 (t, J ¼ 5:9 Hz, 4H, NCH2), 2.30 (s, 6H,
A 100 ml round-bottom flask was charged Pd (dba)
2
MesCH ), 2.27 (s, 12H, MesCH ), (CH not detected).
3
3
3
2
(
91.6 mg, 0.10 mmol) and (rac)-BINAP (124.6 mg, 0.30
13C{1H} (100 MHz, CD Cl ): d [ppm] ¼ 205.9 (dd,
2
2
1 107 13 1 109 13
mmol). The mixture was dissolved in toluene (60 ml) at
room temperature. Mesityl bromide (15.93 g, 80.0
mmol) and 1,3-diaminopropane (2.56 g, 40.0 mmol)
were added to the solution and stirred for 10 min. The
solution was transferred into a glass tube containing
NaOt-Bu (8.96 g, 93.2 mmol) in an autoclave and was
stirred under nitrogen pressure for 70 h at 140 ꢁC in a
J ( Ag, C) ¼ 228.4 Hz, J ( Ag, C) ¼ 260.4 Hz,
NCN), 143.1, 138.3, 134.7, 129.7 (MesC), 42.0 (NCH2),
20.8 (MesCH3), 17.7 (MesCH3), 17.5 (CH2).
Anal. Found: C, 57.23; H, 6.05; N, 6.00. Calc. for
C22H28AgClN2 (463.78): C, 56.97; H, 6.08; N, 6.04.
þ
MS(FAB): 750 ([(carbene)2Ag] ), 429 ([(car-
þ
bene)Ag] ).
2
50 ml autoclave. After the reaction finished, the solu-
tion was washed with toluene. After thorough washing,
the water layer containing the salt was treated with
NaOH to give N; N -dimesitylpropanediamine as a yel-
lowish brown oil, which was separated and dried in
4.5. Di[bis(dimesityltetrahydropyrimid-2-ylidene)silver(I)]
hexachlorodipalladate (4)
0
A solution of Pd(CH3CN)2Cl2 (50 mg, 0.19 mmol)
in 20 ml of DCM was added to a solution of 3 (179
mg, 0.39 mmol, 2 eq.) in 10 ml of DCM. After stir-
ring for 2 h, the solution turned from yellow to dark
red. The solution was filtered off the precipitated silver
chloride and the solvent was removed in vacuo until
ca. 5 ml remained, and 20 ml of hexane was
added, causing a red precipitate to form (152 mg,
vacuo (yield, 9.68 g, 80%).
1
H (400 MHz, CDCl3): d [ppm] ¼ 6.81 (s, 4H, MesH),
3
.51 (br, 4H, NCH ), 2.95 (br, 2H, NH), 2.25 (s, 12H,
2
MesCH ), 2.22 (s, 6H, MesCH ), 1.87 (q, J ¼ 7:2 Hz,
3
3
2
H, CH ). 13C{1H} (100 MHz, CDCl ):
2
d
3
[
3
ppm] ¼ 143.5, 131.4, 129.7, 129.4 (MesC), 47.1 (NCH2),
2.6 (CH2), 20.5 (MesCH3), 18.3 (MesCH3).
8
2%).
1
0
4.3. N,N -Dimesityl-tetrahydropyrimidinium tetrafluoro-
borate (2)
H (270 MHz, CD2Cl2): d [ppm] ¼ 6.90 (s, 4H,
MesH), 3.14 (t, J ¼ 5:9 Hz, 4H, NCH2), 2.14 (q, J ¼ 5:9
Hz, 2H, CH ), 2.33 (s, 6H, MesCH ), 1.76 (s, 12H,
2
3
A mixture of dimesitylpropanediamine 1 (3.22 g, 10.4
mmol, 1.0 eq.), NH BF (1.09 g, 10.4 mmol, 1.0 eq.),
MesCH ).
3
13C{1H} (100 MHz, CD Cl ): d [ppm] ¼ 205.8 (dd,
4
4
2
1
2
1
107
13
109
13
and HC(OEt)3 (1.54 g, 10.4 mmol, 1.0 eq.) was stirred at
35 ꢁC for 17 h. The crude product, after washing with
J ( Ag, C) ¼ 174.3 Hz, J ( Ag, C) ¼ 201.3 Hz,
1
NCN), 142.8, 138.4, 134.8, 129.9 (MesC), 43.9 (NCH2),
21.2 (MesCH3), 20.8 (CH2), 17.8 (MesCH3).
toluene, was extracted with acetone. Removal of the
solvent gave a pinkish powder, that was recrystallized
from hot toluene/acetone to give colorless crystals 2 (958
Anal. Found: C, 54.81; H, 6.08; N, 5.49 Calc. for
C88H112Cl6Ag2N8Pd2 (1923.20): C, 54.96; H, 5.87; N,
5.83.
mg, 23%).
1
þ
H (400 MHz, CDCl ): d [ppm] ¼ 7.46 (s, 1H, NCN),
MS(FAB): 750 ([(carbene) Ag+1] ), 429 ([(car-
2
3
þ
6
.94 (s, 4H, MesH), 3.91 (t, J ¼ 5:9 Hz, 4H, NCH ),
bene)Ag] ).
2

2448
W.A. Herrmann et al. / Journal of Organometallic Chemistry 689 (2004) 2441–2449
4
4
.6. Crystallography
were used to refine 241 parameters. All non-hydrogen
atoms were refined with anisotropic displacement pa-
rameters. All hydrogen atoms were calculated in ideal
positions (riding model). The refinements converged
with R1 ¼ 0.0333 [Io > 2rðIoÞ], wR2 ¼ 0.0908 [all data],
and GOF ¼ 1.053. The final difference-Fourier map
.6.1. General remarks
The preparation of the single crystals is described
above. Selected crystals were coated with perfluorinated
ether, fixed in a capillary, transferred to the diffrac-
tometer, and cooled in a nitrogen stream (Oxford
Cryosystems). Preliminary examination and data col-
lection were carried out with a j-CCD device (NONIUS
MACH3) at the window of a rotating anode (Nonius
FR591) with graphite-monochromated Mo Ka radia-
showed no striking feature (De
¼ þ0:56= ꢀ 0:58
min = max
ꢀ
ꢀ3
e A ). One solvent molecule could not be modelled
proper. The problem was solved by using the PLATON
Calc Squeeze procedure.
ꢀ
tion (k ¼ 0:71073 A). [16] Data were corrected for Lo-
rentz and polarization effects. Potential absorption
effects and/or decay of the crystal were corrected during
the scaling procedure [17]. All structures were solved by
a combination of direct methods and difference Fourier
4.6.4. Crystal data, data collection and refinement for
4 ꢂ 5(CH Cl )
2
2
þ
2ꢀ
2[(C H AgN ) ], [(Cl Pd ) ], 10(CH Cl ), M ¼
4
4
56
4
6
2
2
2
r
3
2772:40, reddish brown needle (0.08 ꢃ 0.10 ꢃ 0.38 mm ),
syntheses. Full-matrix, least-squares refinements were
2 2
monoclinic,P2 =n(No. 14), a ¼ 1232:41ð1Þ; b ¼ 27:1528ð2Þ,
1
ꢀ
P
2
o
ꢀ3
carried out by minimization of
SHELXL-97 weighting scheme and stopped at shift/err
wðF ꢀ F Þ with
c ¼ 18:0106ð1Þ A, b ¼ 97:0294ð3Þꢁ, V ¼ 5981:65ð7Þ A , Z ¼
3
calc 000
c
ꢀ
ꢀ1
2, d ¼ 1:539 gcm , F ¼ 2808, l ¼ 1:245 mm . Data
>
0.001. Neutral atom scattering factors for all atoms
collection were performed atT ¼ 173 K within the H-range of
1:36ꢁ < H < 25:38ꢁ. A total of 139,313 reflections were inte-
grated.Aftermerging(Rint ¼ 0:061), 10962[8842:I > 2rðI Þ]
and anomalous dispersion corrections for the non-hy-
drogen atoms were taken from International Tables for
Crystallography [18]. All calculations were performed
with the STRUX-V [19] system including the programs
PLATON [20], SIR92 [21], and SHELXL-97 [22].
o
o
independent reflections remained and all were used to refine
857 parameters. All non-hydrogen atoms were refined with
anisotropic displacement parameters. All hydrogen atoms of
the cation were found in the final difference-Fourier map and
were allowed to refine. All hydrogen atoms of the five solvent
molecules were calculated in ideal positions (riding model). A
disorder (73:27 for both) observed for two solvent molecules
could be resolved well. The refinements converged with
R ¼ 0:0345 [I > 2rðI Þ], wR ¼0.0737 [all data], and
4
.6.2. Crystal data, data collection and refinement for 2
þ
ꢀ
[
(C22H29N2) ], [(BF4) ], Mr ¼ 408:28, colorless frag-
3
ment (0.30 ꢃ 0.33 ꢃ 0.53 mm ), orthorhombic, Pnma
(
No. 62), a ¼ 9:0427ð1Þ; b ¼ 15:5513ð2Þ, c ¼ 15:0906ð2Þ
ꢀ3
calc
ꢀ
ꢀ3
,
A, V ¼ 2122:13ð5Þ A , Z ¼ 4, d ¼ 1:278 g cm
1
o
o
2
ꢀ
1
F
¼ 864, l ¼ 0:099 mm . Data collection were per-
GOF¼1.029. The final difference-Fourier map showed no
ꢀ
0
00
ꢀ3
formed at T ¼ 133
K
within the H-range of
striking feature (De
¼þ0:61=ꢀ0:59 eA ).
min=max
1
:88ꢁ < H < 25:34ꢁ. A total of 46,510 reflections were
integrated. After merging (Rint ¼ 0:041), 2020 [1801:
Io > 2rðIoÞ] independent reflections remained and all
were used to refine 209 parameters. All non-hydrogen
atoms were refined with anisotropic displacement pa-
rameters. All hydrogen atoms were found in the final
difference-Fourier map and were allowed to refine. The
refinements converged with R1 ¼ 0:0413 [Io > 2rðIoÞ],
wR2 ¼ 0:0991 [all data], and GOF ¼ 1.064. The final
difference-Fourier map showed no striking feature
5. Supplementary data
Crystallographic data (excluding structure factors)
for the structures reported in this paper have been de-
posited with the Cambridge Crystallographic Data
Centre as supplementary publication No. CCDC-
234466 (2), No. CCDC-234465 (3) and No. CCDC-
234467 (4 ꢂ 5(CH Cl )). Copies of the data can be
2
2
ꢀ
ꢀ3
(
pears to be disordered over two positions (50:50).
Demin = max ¼ þ0:34= ꢀ 030 e A ). The BF4 anion ap-
obtained free of charge on application to CCDC, 12 Un-
ion Road, Cambridge CB2 1EZ, UK (fax: (+44)1223-
3
36-033; e-mail: deposit@ccdc.cam.ac.uk).
4
.6.3. Crystal data, data collection and refinement for 3
C H AgClN ,
M ¼ 463.78,
colorless
plate
2
2
28
2
r
3
(
0.08 ꢃ 0.36 ꢃ 0.66 mm ), orthorhombic, Pbcn (No. 60),
Acknowledgements
ꢀ
ꢀ3
a ¼ 30:2113ð3Þ; b ¼ 8:9217ð1Þ,
c ¼ 16:0333ð2Þ
A,
,
ꢀ
3
V ¼ 4321:55ð8Þ A , Z ¼ 8, dcalc ¼ 1:426 g cm
This work was generously supported by the Fonds
der Chemischen Industrie (studentship for S.K.S.). The
Karl Wamsler-Foundation is thanked for a travel fel-
lowship. The authors thank Dr. J o€ rg Eppinger for
helpful discussions. E.H. thanks Cornell Hess for help-
ful inspirations.
ꢀ
1
F000 ¼ 1904, l ¼ 1:064 mm . Data collection were
performed at T ¼ 173 K within the H-range of
2
:38ꢁ < H < 25:37ꢁ. A total of 54,612 reflections were
integrated. After merging (Rint ¼ 0:048), 3965 [3218:
I > 2rðI Þ] independent reflections remained and all
o
o

W.A. Herrmann et al. / Journal of Organometallic Chemistry 689 (2004) 2441–2449
2449
References
[9] J.C. Garrison, R.S. Simons, J.M. Talley, C.A. Tessier, W.J.
Youngs, Organometallics 20 (2001) 1276.
[
[
10] (a) A.A.D. Tulloch, A.A. Danopoulos, S. Winston, S. Kleinhenz,
G.J. Eastham, J. Chem. Soc., Dalton Trans. (2000) 4499;
[
1] (a) W.A. Herrmann, M. Elison, J. Fischer, C. K o€ cher, G.R.J.
Artus, Angew. Chem. Int. Ed. Engl. 34 (1995) 2371;
(
b) T. Raminal, C.D. Abernethy, M.D. Spicer, I.D. McKenzie,
(
b) V.P.W. B o€ hm, T. Weskamp, C.W.K. Gst o€ ttmayr, W.A.
I.D. Gay, J.A.C. Clyburne, Inorg. Chem. 42 (2003) 1391.
11] (a) M. Mayr, K. Wurst, K.-H. Ongania, M.R. Buchmeiser,
Chem. Eur. J. 10 (2004) 1256;
Herrmann, Angew. Chem. 112 (2000) 1672;
Angew. Chem. Int. Ed. Engl. 39 (2000) 1602;
(
c) T. Weskamp, W.C. Schattenmann, M. Spiegler, Angew.
Chem. Int. Ed. Engl. 37 (1998) 2490;
d) T. Weskamp, F.J. Kohl, W. Hieringer, D. Gleich, W.A.
Herrmann, Angew. Chem. Int. Ed. Engl. 38 (1999) 2416;
e) W.A. Herrmann, C.W. Kohlpaintner, Angew. Chem. Int. Ed.
Engl. 32 (1993) 1524;
f) J. Schwarz, E. Herdtweck, W.A. Herrmann, M.G. Gardiner,
Organometallics 19 (2000) 3154;
g) M. Muehlhofer, T. Strassner, W.A. Herrmann, Angew. Chem.
(
(
b) M. Mayr, M.R. Buchmeiser, Macromol. Rapid Commun. 25
2004) 231.
(
[
[
12] S. Saba, A.-M. Brescia, M.K. Kaloustian, Tetrahedron Lett. 32
1991) 5031.
(
(
13] R.W. Alder, M.E. Blake, C. Borlotti, S. Bufali, C.P. Butts, E.
Linehan, J.M. Oliva, A.G. Orpen, M.J. Quale, Chem. Commun.
(
(
1999) 241.
[
[
14] A.J. Arduengo III, H.V.R. Dias, J.C. Calabrese, F. Davidson,
Organometallics 12 (1993) 3405.
15] (a) M.T. Reetz, E. Westermann, Angew. Chem. Int. Ed. Engl. 39
(
Int. Ed. Engl. 41 (2002) 1745.
2] (a) W.A. Herrmann, Angew. Chem. Int. Ed. Engl. 41 (2002) 1290;
[
(
(
2000) 165;
b) Q. Yao, E.P. Kinney, Z. Yang, J. Org. Chem. 68 (2003) 7528,
(
(
(
b) W.A. Herrmann, C. K o€ cher, Angew. Chem. Int. Ed. Engl. 36
1997) 2162;
c) D. Bourissou, O. Guerret, F.P. Gabbai, G. Bertrand, Chem.
€
and references therein.
16] Data Collection Software for Nonius kappa-CCD devices, Delft
[
[
Rev. 100 (2000) 39.
3] R.W. Alder, P.R. Allen, M. Murray, A.G. Orpen, Angew. Chem.
Int. Ed. Engl. 35 (1996) 1121.
(
17] Z. Otwinowski,W. Minor, MethodsinEnzymology1997, 276,307ff.
The Netherlands), 2001.
[
[
[
€
4] W.A. Herrmann, K. Ofele, D.v. Preysing, E. Herdtweck, J.
[18] A.J.C. Wilson (Ed.), International Tables for Crystallography,
vol. C, Tables 6.1.1.4, 4.2.6.8, and 4.2.4.2, Kluwer Academic
Publishers, Dordrecht, The Netherlands, 1992.
Organomet. Chem. 684 (2003) 235.
5] (a) M. Sakamoto, S. Okada, Y. Tsunokai, S. Ikeda, W.A.
€
Herrmann, K. Ofele; Nippon Zeon Co Ltd; JP 2003089689;
[19] G. Artus, W. Scherer, T. Priermeier, E. Herdtweck, STRUX-V, A
Program System to Handle X-Ray Data, TU M u€ nchen, Germany,
€
b) W.A. Herrmann, K. Ofele, S. Okada, S. Ikeda, M. Sakamoto,
(
Y. Tsunogae, WO 03027079.
1
997.
[
[
20] A.L. Spek, PLATON, Utrecht University, Utrecht, The Nether-
lands, 2001.
[
[
6] H.M.J. Wang, I.J.B. Lin, Organometallics 17 (1998) 972.
7] (a) H.M.J. Wang, C.Y.L. Chen, I.J.B. Lin, Organometallics 18
21] A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, M.C.
Burla, G. Polidori, M. Camalli, SIR92, J. Appl. Crystallogr. 27
(
(
1999) 1216;
b) S.K. Schneider, W.A. Herrmann, E. Herdtweck, Z. Anorg.
(
1994) 435.
Allg. Chem. 629 (2003) 2363;
c) D.S. McGuinness, K.J. Cavell, Organometallics 19 (2000) 741.
8] P.L. Arnold, Heteroat. Chem. 13 (2002) 534.
[
22] G.M. Sheldrick, SHELXL-97, Universit €a t G o€ ttingen, G o€ ttingen,
(
Germany, 1998.
[