Nickel(0) and palladium(0) complexes with 1,3,5-triaza-7-phosphaadamantane. Catalysis of buta-1,3-diene oligomerization or telomerization in an aqueous biphasic system

Article abstract of DOI:10.1135/cccc19970355

New homoleptic nickel(0) and palladium(0) complexes with a water-soluble ligand, 1,3,5-triaza-7-phosphaadamantane, were prepared and characterized by ¹H, ¹³C, and ³¹P NMR spectra. The complexes, together with the known ana

Full text of DOI:10.1135/cccc19970355

Nickel(0) and Palladium(0) Complexes
355
NICKEL(0) AND PALLADIUM(0) COMPLEXES WITH 1,3,5-TRIAZA-
7
-PHOSPHAADAMANTANE. CATALYSIS OF BUTA-1,3-DIENE
OLIGOMERIZATION OR TELOMERIZATION IN AN AQUEOUS
BIPHASIC SYSTEM
1
2
3
Jan CERMAK , Magdalena KVICALOVA and Vratislav BLECHTA
Institute of Chemical Process Fundamentals, Academy of Sciences of the Czech Republic,
1
1
65 02 Prague 6-Suchdol, Czech Republic; e-mail: cermak@icpf.cas.cz,
kvicalova@icpf.cas.cz, blechta@icpf.cas.cz
2
3
Received December 20, 1996
Accepted January 22, 1997
Dedicated to Dr Karel Mach on the occasion of his 60th birthday.
New homoleptic nickel(0) and palladium(0) complexes with a water-soluble ligand, 1,3,5-triaza-7-
1
13
31
phosphaadamantane, were prepared and characterized by H, C, and P NMR spectra. The com-
plexes, together with the known analogous Ni(0) and Pd(0) complexes with
tris(hydroxymethyl)phosphine, were found to be catalysts for buta-1,3-diene oligomerization or telo-
merization with water in an aqueous biphasic system without a cosolvent or a modifier. Te-
trakis[tris(hydroxymethyl)phosphine]nickel (7) preferentially catalyzes oligomerization (both linear
and cyclic) in the first example of a nickel-catalyzed buta-1,3-diene oligomerization in an aqueous
biphasic system. Palladium complexes give telomers or linear oligomers in quantitative yields. In the
case of the triazaphosphaadamantane complex 4, high selectivity to octadienyl ethers (87%) was ob-
served. High values of metal leaching into the product phase in these reactions suggest an easy ex-
traction of starting or intermediate metal complexes caused by the fact that both monomer and
products are good ligands for the metal complexes in this particular case.
Key words: Buta-1,3-diene, oligomerization, telomerization; Catalysis in water; Nickel(0); Palla-
dium(0).
Homogeneous catalysis with organometallic complexes in systems containing water has
1
–3
attracted much attention in recent years and the field continues to develop at a steady
pace. The principal advantage of carrying out the process in a two-phase system with
the catalyst in an aqueous phase is the facility with which the products are separated
from the catalyst.
Since phosphine ligands are probably the most frequently used ancillary ligands in
catalytic systems with transition metals, focus lies on transferring known catalytic pro-
cesses with phosphine ligands from the organic to the biphasic aqueous medium. Sulfo-
nation of the aromatic rings of triphenylphosphine and other ligands containing phenyl
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

3
56
Cermak, Kvicalova, Blechta:
groups, as a straightforward strategy to make the catalytic complex water-soluble, was
4
first used by Chatt many years ago.
While an industrial process based on trisulfonated triphenylphosphine has been
5
known for more than ten years and new sophisticated ligands capable of, e.g., high
6
degree of asymmetric induction have been designed , the purification of such ligands
7
for model studies is difficult and tedious . Furthermore, phosphine ligands with sulfo-
nated phenyl rings become even bulkier than the parent phosphines, which may be
advantageous in a particular catalytic reaction and disadvantageous in another one.
Clearly, a broader spectrum of water soluble ligands is needed and some were indeed
prepared and used for catalysis by several groups in the past.
We concentrated our attention on two similar phosphines, viz. tris(hydroxy-
8
methyl)phosphine (THP) (1) first prepared by Chatt et al. , and 1,3,5-triaza-7-phos-
9
phaadamantane (PTA, 2) first synthesized by Daigle and co-workers .
1
0,11
11
12
13–15
13,16
Molybdenum
,3,5-triaza-7-phosphaadamantane were synthesized but nickel and palladium com-
plexes are unknown*. Although Ni(0) and Pd(0) complexes of P(CH OH) were pre-
, tungsten , gold , rhodium
, and ruthenium
complexes of
1
2
3
1
7
pared by Pringle et al. , they were not used for catalytic purposes.
P
P(CH2OH)3
N
N
N
1
2
Complexes of low-valent nickel and palladium are well known as oligomerization
1
8
and telomerization catalysts for conjugated dienes . The presence of carbon dioxide is
reported to be a prerequisite for successful palladium-catalyzed telomerization of buta-
1
9
1
,3-diene with water . In a recent paper, a strong dependence of conversion and selec-
tivity of the reaction in an aqueous biphasic system on the structure of an additional
2
0
modifier, trialkylamine, was observed .
The present work deals with the synthesis of new Ni(0) and Pd(0) complexes of
1
,3,5-triaza-7-phosphaadamantane and their use, together with the analogous THP com-
plexes, in the catalysis of butadiene oligomerization or telomerization in an aqueous
biphasic system.
*
In the course of the preparation of this paper the authors were informed (A. Kathó, private communication)
that such complexes were prepared independently by Darensbourg and co-workers.
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Nickel(0) and Palladium(0) Complexes
357
EXPERIMENTAL
The syntheses of metal complexes were carried out under an argon atmosphere with standard Schlenk
2
1
and cannula techniques . Toluene was dried by refluxing with sodium bis(2-methoxyethoxy)alu-
minum hydride, methanol and ethanol by refluxing with magnesium alkoxide. Tris(hydroxymethyl)-
1
7
9
22
phosphine , 1,3,5-triaza-7-phosphaadamantane , bis(cycloocta-1,5-diene)nickel , tris(dibenzylidene-
2
3
3
24
acetone)dipalladium , bis[(η -2-methylpropen-2-yl)chloropalladium] , and tetrakis[tris(hydroxy-
methyl)phosphine]nickel were prepared according to reported procedures as indicated. Buta-1,3-
diene (Fluka, 98%) was used as received.
1
7
NMR spectra were taken on a Varian UNITY 200 spectrometer at 200.1 MHz, 50.3 MHz, and
1
13
31
8
1
1.0 MHz for H, C and P, respectively. Hexamethyldisilane was used as an internal standard for
13
31
H and C, external H PO was used for P. Mass spectra were measured by GC/MS method on a
3
4
capillary gas chromatograph (Varian, model 3500) equipped with a mass detector (Finnigan MAT,
model ITD 800). Nickel and palladium contents in complexes and organic layers after oligomeriza-
tion/telomerization were determined by atomic absorption spectroscopy on an AAS 3 instrument
(
Zeiss, Jena, Germany).
Preparation of Complexes
Tetrakis(1,3,5-triaza-7-phosphaadamantane)nickel (3). Bis(cycloocta-1,5-diene)nickel (0.43 g,
1
6
.58 mmol) was dissolved in toluene (35 ml) and a solution of 1,3,5-triaza-7-phosphaadamantane (1.0 g,
.4 mmol) in methanol (3 ml) slowly added with stirring. The off-white precipitate started to form
almost immediately. The mixture was further stirred for 1 h and then left to stand overnight. The
product was filtered off with a cannula and dried in vacuum. Yield 1.0 g (92%). For C H N NiP
4
2
4
48 12
(
687.3) calculated: 8.54% Ni; found: 8.10% Ni.
Tetrakis(1,3,5-triaza-7-phosphaadamantane)palladium (4). Tris(dibenzylideneacetone)dipalladium
(
1.62 g, 1.77 mmol) was dissolved in toluene (250 ml) and a solution of 1,3,5-triaza-7-phosphaada-
mantane (2.25 g, 14.3 mmol) in methanol (10 ml) was added with stirring which was continued over-
night. Yellow-orange compound started to precipitate after 3 h. The precipitate was filtered off with
a cannula and dried in vacuum. Yellow-orange powder 5 was obtained (0.6 g, 26%).
The filtrate was concentrated to about half of its original volume and left to stand for 2 days.
Another portion of the yellow-orange compound precipitated but since the compound in contact with
the mother liquor slowly changed colour to grey, it was left to stand further until all the precipitate
turned grey (about 1 week). The precipitate was then filtered off with a cannula and dried in vacuum.
Yield 1.9 g (73%) of a grey powder. For C H N P Pd (735.0) calculated: 14.48% Pd; found:
2
4
48 12 4
1
4.33% Pd.
Tetrakis[tris(hydroxymethyl)phosphine]palladium (6). Solid tris(hydroxymethyl)phosphine (3.25 g,
3
2
6.3 mmol) was added with stirring to a hot (about 60 °C) solution of bis[(η -2-methylprop-2-
enyl)chloropalladium] (1.0 g, 2.54 mmol) in 50 ml of ethanol and dissolved rapidly. The yellow sol-
ution was left to cool down to laboratory temperature while stirred for about 1 h during which time
it slowly turned red. Orange-red precipitate which formed on cooling was filtered off with a cannula
1
13
31
and dried in vacuum. Yield 2.0 g (65%). The H, C, and P NMR spectra of the product are in
accordance with the published data .
1
7
Oligomerization/Telomerization Experiments – General Procedure
A steel rocking autoclave cooled to about –20 °C was charged under argon with butadiene (10.0 g,
0
.185 mol) and then with metal complex (0.73 mmol) dissolved in water (25 ml) deaerated by re-
peated vacuum/argon cycles. The autoclave was placed in a preheated mantle and rocked at 80 ± 3 °C
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

3
58
Cermak, Kvicalova, Blechta:
for 20 h. After cooling with water the contents were removed and the layers separated. A sample for
metal content determination was taken from the organic layer which was then subjected to analysis
by gas chromatography (HP 5890 gas chromatograph equipped with a 50 m × 0.2 mm i.d. quartz
capillary column, HP-5 stationary phase) and GC/MS after filtration of a sample through a Cellite
layer.
Comparison of retention times and mass spectra of products with those of commercial samples
was used for identification of 4-vinylcyclohexene, cycloocta-1,5-diene, (E,E,Z)- cyclododeca-1,5,9-
triene, and (E,E,E)- cyclododeca-1,5,9-triene. The other products were isolated from the higher-scale
1
13
run (four times) with [Pd(PTA) ] catalyst after vacuum rectification. The H and C NMR spectra of
4
2
5–29
those products are in good agreement with published data
.
RESULTS AND DISCUSSION
New 1,3,5-triaza-7-phosphaadamantane complexes of Ni(0) and Pd(0) were prepared
by the reaction of PTA with the appropriate zero-valent complexes with easily displa-
ceable ligands: bis(cycloocta-1,5-diene)nickel ([Ni(COD) ]) and tris(dibenzylidenea-
2
cetone)dipalladium ([Pd (dba) ]), respectively (Scheme 1).
2
3
Ni(COD)2
4
PTA
N
N
N
N
3, M = Ni
4, M = Pd
P
P
P
P
N
N
N
M
N
N
N
N
PTA
N
2
[
Pd(dba)(PTA) ]
5
2
2
PTA
1/2 Pd2(dba)3
SCHEME 1
While the precipitation of the nickel complex from the methanol–toluene solvent
mixture occurs almost immediately, the synthesis of the palladium complex is not so
straightforward. Yellow-orange compound 5 precipitates from the same solvent mixture
first and is then slowly converted to the desired product. It is insoluble in many com-
mon organic solvents both polar and nonpolar, and in water. Fortunately, it is soluble in
chloroform, though with slow decomposition (days), and hence NMR spectra could be
measured.
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Nickel(0) and Palladium(0) Complexes
359
A single line in 31P NMR spectrum at δ = –59.8 ppm is the evidence of a phoshorus
atom coordinated to palladium with a chemical shift close to that of [Pd(PTA) ] (see
4
1
below). Two lines at 4.35 ppm and 4.49 ppm in H NMR spectrum and two lines at
0.46 ppm and 73.19 ppm in 1 C NMR spectrum confirm the presence of the PTA
3
5
ligand. Both proton and carbon-13 NMR spectra show that dibenzylideneacetone ligand
is coordinated to palladium; from the proton integrals the ratio PTA : dba = 2 : 1 was
calculated. We therefore formulate the yellow-orange compound 5 as [Pd(dba)(PTA)2].
Phosphorus NMR chemical shifts of [Ni(PTA) ] and [Pd(PTA) ] in D O were found
4
4
2
to be –46.5 ppm and –59.2 ppm, respectively, with the corresponding coordination
shifts (relative to the value of –97.9 ppm for free PTA in the same solvent) being 51.4 ppm
and 38.7 ppm. Both complexes are air-sensitive even in the solid state, as expected.
However, they are stable in water solution for periods of time long enough to measure
NMR spectra. In this solution, [Pd(PTA) ] decomposes to ligand and metallic palla-
4
dium whereas its nickel analogue decomposes with the formation of 1,3,5-triaza-7-
phosphaadamantane P-oxide (δ = –2.2 ppm).
1
H NMR spectrum of [Pd(PTA) ] in D O consists of a singlet at 3.79 ppm assigned
4
2
to protons of the methylene carbons next to phosphorus and an AB system of nonequi-
valent protons of methylene carbons between nitrogen atoms with chemical shifts cal-
2
culated at δ 4.41 ppm and 4.49 ppm and J(H-H) = 13.2 Hz. The upfield shift of the
1
1
former signal relative to the signal of free PTA was not expected since Darensbourg
observed a downfield shift of those protons in M(CO) (PTA) (M = Mo, Cr, W) com-
5
plexes. The upfield shift in Ni and Pd complexes is probably due to the high electron
2
density at a low-valent metal centre. Coupling constant J(P-H) approaching zero was
observed in accord with the results found with the above complexes. Proton spectrum
of [Ni(PTA) ] gave only very broad lines without any useful information, probably
4
owing to the presence of a paramagnetic impurity.
1
3
C NMR spectra of [Ni(PTA) ] and [Pd(PTA) ] in D O showed signals at 59.52 and
4
4
2
5
9.24 ppm, assigned to α carbons to phosphorus, and signals at 73.17 and 73.23 ppm
assigned to γ carbons. α Carbon signals were broad but no coupling to phosphorus
could be resolved.
Tetrakis[tris(hydroxymethyl)phosphine]nickel, 7 was prepared according to the re-
1
7
ported procedure ; unsatisfactory results were, however, obtained when the synthesis
of [Pd{P(CH OH) } ] was attempted. New route to this complex was developed in
2
3 4
3
which bis[(η -2-methylprop-2-enyl)chloropalladium] was reduced by excess P(CH OH) (Eq. (1)).
2
3
3
1
/2 [(η -C H )PdCl] + 5 P(CH OH)
[Pd{P(CH OH) } ] +
4
7
2
2
3
2
3 4
+
–
+
CH =C(CH )CH P (CH OH) Cl
(1)
2
3
2
2
3
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

3
60
Cermak, Kvicalova, Blechta:
For the butadiene reactions catalyzed by the above Ni(0) and Pd(0) complexes, a
simple experimental setup was chosen with the aim of excluding the influence of cosol-
vents or other agents facilitating phase transfer so that the only variable was the inher-
ent ability of the complex to catalyze the reaction. It is important to note here that THP
and PTA themselves are not surfactants in contrast to sulfonated phosphines with ion
pair structures; hence, the phase separation is facile.
The ligands are very similar. They are moderately electron donating and their Tolman
3
0
10,17
cone angles were estimated
to be about 118° (THP) and 102° (PTA); these values
lie close to those of trimethylphosphine or triethylphosphine. The main difference between
the two phosphines is the ability of the former to create a shell-like network of intra-
1
7
molecular hydrogen bonds in complexes like [Pd{P(CH OH) } ]. We were interested
2
3 4
in the possibility that this intramolecular network could govern the catalytic activity or
selectivity of the complex.
8
9
10
11
12
13
OH
OH
14
15
O
O
2
16
17
The results of catalytic experiments are summarized in Table I. Linear (8, 9) as well
as cyclic (10–13) oligomers and linear telomers (14–17) were identified among the
products. No reaction was observed with 3 as the catalyst, this complex being the most
air-sensitive of 3–7. Either a small amount of adventitious oxygen during the filling of
the autoclave or low thermal stability of the compound is believed to cause the absence
of catalytic activity. In all other cases, high yields of products in the absence of carbon
dioxide or other modifiers were obtained. The experiments were not optimized and we
believe that both reaction time and the amount of catalyst can be decreased without
losing much yield.
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Nickel(0) and Palladium(0) Complexes
361
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

3
62
Cermak, Kvicalova, Blechta:
Nickel complex 7 catalyzes oligomerization, mainly linear, whereas with palladium
complexes telomers of butadiene and water form a substantial part of the reaction mix-
ture. To our best knowledge, the first example of nickel-catalyzed oligomerization in
aqueous biphasic medium is presented here. The comparison of selectivity of PTA and
THP complexes is possible only in the case of palladium complexes. With [Pd(PTA)4]
high selectivity to octadienyl ethers was observed (87% ), the other products being also
telomers with the exception of 4% of octa-1,3,7-triene. The observed selectivity re-
3
1
sembles that found in telomerization of buta-1,3-diene in water emulsion although the
ratio between bis(octa-2,7-dienyl) ether and its isomer (octa-2,7-dienyl)(octa-1,7-die-
nyl) ether is 7 : 1, i.e.. much higher than that observed in water emulsion. However,
reaction catalyzed by [Pd(THP) ] gives strikingly different results. The main course of
4
the reaction is linear dimerization and the product distribution is similar to that in
3
2
single-phase reaction in THF-water mixture catalyzed by [Pd(PPh ) ] (ref. ). Interes-
3
4
tingly, octadienols were only minor products in our experiments.
As mentioned above, the principal advantage of catalysis in aqueous biphasic sys-
tems should be the easy separation of catalyst from the products. In our experiments in
these model systems without an organic solvent, leaching of metal into organic phase
as high as 63% in one run was observed, which prevents the use of the reaction in a
continual process. Although leaching of nickel is considerably lower than that of palla-
dium, it is still too high. Our results point to the general problem of the extraction of
the catalyst into the organic phase in aqueous biphasic systems in the cases when pro-
ducts are good ligands for the transition metal used.
The work was supported by the Grant Agency of the Academy of Sciences of the Czech Republic
(
Grant No. A4072610). The authors thank Miss A. Surova for experimental assistance and Dr R. Rericha
for GC/MS analyses.
REFERENCES
1. Herrmann W. A., Kohlpaintner C. W.: Angew. Chem., Int. Ed. Engl. 32, 1524 (1993).
2. Kalck P., Monteil F.: Adv. Organomet. Chem. 34, 219 (1992).
3. Horvath I. T., Joo F. (Eds): Aqueous Organometallic Chemistry and Catalysis, NATO ASI Series
Vol. 3/5. Kluwer, Dordrecht 1995.
4
5
6
7
8
9
. Ahrland S., Chatt J., Davies N. R., Williams A. A.: J. Chem. Soc. 1958, 276.
. Kuntz E.: CHEMTECH 17, 570 (1987).
. Herrmann W. A., Albanese G., Manetsberger R., Schmid R., Schwer C.: Ref. , p. 127.
3
. Herrmann W. A., Kulpe J. A., Konkol W., Bahrmann H.: J. Organomet. Chem. 389, 85 (1990).
. Chatt J., Leigh G. J., Slade R. M.: J. Chem. Soc., Dalton Trans. 1973, 2021.
. Daigle D. J., Pepperman A. B., Jr., Vail S. L.: J. Heterocycl. Chem. 11, 407 (1974).
1
1
1
0. DeLerno J. R., Trefonas L. M., Darensbourg M. Y., Majeste R. J.: Inorg. Chem. 15, 816 (1976).
1. Darensbourg M. Y., Daigle D.: Inorg. Chem. 14, 1217 (1975).
2. Fackler J. P., Jr., Staples R. J., Assefa Z.: J. Chem. Soc., Chem. Commun. 1994, 431.
Collect. Czech. Chem. Commun. (Vol. 62) (1997)

Nickel(0) and Palladium(0) Complexes
363
1
3. Darensbourg D. J., Joo F., Kannisto M., Katho A., Reibenspies J. H.: Organometallics 11, 1990
1992).
4. Darensbourg D. J., Stafford N. W., Joo F., Reibenspies J. H.: J. Organomet. Chem. 488, 99
1995).
(
1
(
1
1
5. Joo F., Nadasdi L., Benyei A., Darensbourg D. J.: J. Organomet. Chem. 512, 45 (1996).
6. Darensbourg D. J., Joo F., Kannisto M., Katho A., Reibenspies J. H., Daigle D. J.: Inorg. Chem.
3
3, 200 (1994).
1
1
7. Ellis J. W., Harrison K. N., Hoye P. A. T, Orpen A. G., Pringle P. G., Smith M. B.: Inorg.
Chem. 31, 3026 (1992).
8. Keim W., Behr A., Röper M. in: Comprehensive Organometallic Chemistry (G. Wilkinson, F. G.
A. Stone and E. W. Abel, Eds), Vol. 8, p. 371. Pergamon Press, Oxford 1982.
9. Atkins K. E., Wolker W. F., Manyik R. M.: Chem. Commun. 1971, 330.
0. Monflier E., Bourdauducq P., Couturier J.-L., Kervennal J., Mortreux A.: J. Mol. Catal., A 97, 29
1
2
(
1995).
2
1. Shriver D. F., Drezdzon M. A.: The Manipulation of Air-Sensitive Compounds, 2nd ed. Wiley,
New York 1986.
2
2
2
2
2
2
2. Yamazaki N., Ohta T.: Polym. J. 4, 616 (1973).
3. Ukai T., Kawazura H., Ishii Y., Bonnet J. J., Ibers J. A.: J. Organomet. Chem. 65, 253 1974).
4. Dent W. T., Long R., Wilkinson A. J.: J. Chem. Soc. 1964, 1585.
5. Denis P., Mortreux A., Petit F., Buono G., Peiffer G.: J. Org. Chem. 49, 5274 (1984).
6. Kasatkin A. N., Kulak A. N., Tolstikov G. A., Lomakina S. I.: Zh. Org. Khim. 24, 2080 (1988).
7. Dzhemilev U. M., Sidorova V. V., Kunakova R. V.: Izv. Akad. Nauk SSSR, Ser. Khim. 1983,
5
84.
2
2
3
3
8. Oppolzer W., Bättig K., Hudlicky T.: Tetrahedron 37, 4359 (1981).
9. Page P. C. B., Rayner C. M., Sutherland I. O.: J. Chem. Soc., Perkin Trans. 1 1990, 1375.
0. Tolman C. A.: Chem. Rev. 77, 313 (1977).
1. Kanaka M., Mukaida Y., Wada H., Takita H., Enomoto S. (Kureha Chemical Industry Co.):
Japan 7 626 809; Chem. Abstr. 85, 77643 (1976).
3
2. Romanelli M. G., Kelly R. J.: Ger. 2 011 163; Chem. Abstr. 74, 53040 (1971).
Collect. Czech. Chem. Commun. (Vol. 62) (1997)