N-substituted diphosphinoamines: Toward rational ligand design for the efficient tetramerization of ethylene

Article abstract of DOI:10.1016/j.jcat.2006.10.020

Bis(diphenylphosphino)amine (PNP) ligands with different alkyl and cycloalkyl substituents attached to the N atom of the ligand backbone were synthesised and tested together with chromium as ethylene tetramerization catalysts. On activation with a methylaluminoxane-based activator, the catalysts displayed good activity and selectivity toward 1-octene and 1-hexene, with the best ligand systems containing cyclopentyl or cyclohexyl moieties. In addition, it was established that substitution at the 2 position of the cyclohexyl skeleton and, more importantly, an increase in steric bulk at that point, led to a drastic reduction of side product formation (i.e., methyl- and methylenecyclopentane). Interestingly, additional methyl substitution in the 6 position of the cyclohexyl ring changed the selectivity of the catalyst from predominantly tetramerization to a 1:1 mixture of 1-hexene and 1-octene. Structurally similar ligands, such as cyclohexylmethyl and cyclohexylethyl PNP, were also tested and were also found to yield efficient tetramerization catalysts. It was concluded that structural fine tuning of the N-alkyl moiety of the PNP ligand is essential for obtaining efficient tetramerization catalysts, with the best systems achieving combined selectivities as high as 88% (1-octene and 1-hexene) with exceptionally high activities exceeding 2,000,000 g/(g-Cr h).

Full text of DOI:10.1016/j.jcat.2006.10.020

Journal of Catalysis 245 (2007) 279–284
www.elsevier.com/locate/jcat
N-substituted diphosphinoamines:
Toward rational ligand design for the efficient tetramerization of ethylene
Sven Kuhlmann a, Kevin Blann b, Annette Bollmann b, John T. Dixon b, Esna Killian b,
Munaka C. Maumela b, Hulisani Maumela b, David H. Morgan b, Marié Prétorius b,
Nicola Taccardi a, Peter Wasserscheid a,∗
a
Lehrstuhl für Chemische Reaktionstechnik der Universität Erlangen-Nürnberg, Egerlandstraße 3, 91058 Erlangen, Germany
b
Sasol Technology (Pty) Ltd, R&D Division, 1 Klasie Havenga Road, Sasolburg, 1947, South Africa
Received 18 August 2006; revised 12 October 2006; accepted 13 October 2006
Available online 28 November 2006
Abstract
Bis(diphenylphosphino)amine (PNP) ligands with different alkyl and cycloalkyl substituents attached to the N atom of the ligand backbone were
synthesised and tested together with chromium as ethylene tetramerization catalysts. On activation with a methylaluminoxane-based activator, the
catalysts displayed good activity and selectivity toward 1-octene and 1-hexene, with the best ligand systems containing cyclopentyl or cyclohexyl
moieties. In addition, it was established that substitution at the 2 position of the cyclohexyl skeleton and, more importantly, an increase in steric
bulk at that point, led to a drastic reduction of side product formation (i.e., methyl- and methylenecyclopentane). Interestingly, additional methyl
substitution in the 6 position of the cyclohexyl ring changed the selectivity of the catalyst from predominantly tetramerization to a 1:1 mixture
of 1-hexene and 1-octene. Structurally similar ligands, such as cyclohexylmethyl and cyclohexylethyl PNP, were also tested and were also found
to yield efficient tetramerization catalysts. It was concluded that structural fine tuning of the N-alkyl moiety of the PNP ligand is essential for
obtaining efficient tetramerization catalysts, with the best systems achieving combined selectivities as high as 88% (1-octene and 1-hexene) with
exceptionally high activities exceeding 2,000,000 g/(g-Cr h).
© 2006 Elsevier Inc. All rights reserved.
Keywords: Linear α-olefins; Tetramerization; Ethylene; 1-Hexene; 1-Octene; Chromium; Homogeneous catalysis; Bis(diphenylphosphino)amine ligands
1. Introduction
valuable intermediate [2]. Although a number of selective ethyl-
ene trimerization catalysts based on chromium [3–10] and other
transition metals [11–15] have been reported, the selective for-
mation of higher olefins such as 1-octene was discovered only
recently [16–18].
The selective tetramerization of ethylene to 1-octene was
demonstrated successfully using a chromium source [i.e.,
CrIII(acac3)], a bis(diphenylphosphino)amine ligand of the gen-
eral structure (Ph2P)2N–R and a methylaluminoxane (MAO)-
based activator [16–18]. Extensive deuterium labeling studies
have indicated that the olefins produced are indeed formed
by means of an extended metallacycle mechanism that is de-
picted in Scheme 1. According to this mechanism, the activated
chromium catalyst 1 coordinates two ethylene molecules (in-
termediate 2) to form the chromacyclopentane 3. A further
ethylene unit is inserted into this species via intermediate 4
Linear alpha olefins (LAOs) are of vital importance for the
chemical industry, particularly as intermediates for the synthe-
sis of polymers; for example, 1-hexene and 1-octene are the
most important co-monomers for the production of LLDPE [1].
Although the majority of LAOs are obtained through processes
that yield a Schulz–Flory distribution (e.g., the Shell higher-
olefin process and Ziegler-type processes), an interesting al-
ternative has evolved over the last 2 decades. The selective
trimerization of ethylene allows access to co-monomer grade
1-hexene and thus is an economically attractive route to this
*
Corresponding author. Fax: +49 9131 8527521.
E-mail addresses: hulisani.maumela@sasol.com (H. Maumela),
wasserscheid@crt.cbi.uni-erlangen.de (P. Wasserscheid).
0021-9517/$ – see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2006.10.020

280
S. Kuhlmann et al. / Journal of Catalysis 245 (2007) 279–284
Scheme 1. Metallacycle mechanism for the selective tetramerization of ethylene as postulated by Overett et al. (the most prominent products are boxed).
yielding the chromacycloheptane 5. This species can then un-
dergo several possible reaction pathways which will lead to the
3 main products. The most prevalent route is the insertion of
a fourth ethylene unit into the metallacycle ring, thus forming
the chromacyclononane 8 which undergoes rapid elimination
to 1-octene. The second possible pathway is the reductive elim-
ination of 1-hexene from the chromacycloheptane 5 and the
third potential pathway is the suggested rearrangement of 5 to
a cyclopentyl hydride species 6. The formation of methyl- and
methylenecyclopentane is proposed to occur via a subsequent
disproportionation step [19]. It was also shown that by vary-
ing the steric bulk of the ligand via substitution at the P–Ar
moiety, the selectivity could be switched from tetra- to trimer-
ization [20]. Furthermore, kinetic studies on the catalyst system
Hewlett-Packard 5890 chromatograph using a J&W Scientific
50 m × 0.2 mm PONA column.
2.2. Hydrogenation of aniline derivatives: general procedure
To a solution of 200 ml of substituted aniline, 10 g of Ru/C
(supplied by Degussa AG, Germany) were added. The slurry
was stirred briefly to ensure fine dispersion before it was trans-
ferred to a 600-ml Parr autoclave equipped with gas-intake
stirrer to ensure sufficient gas entrainment and internal cool-
ing coil to prevent thermal runaways. The reaction vessel was
pressurized and depressurized with 5 barg argon three times to
ensure inert conditions. After heating to reaction temperature,
hydrogen was fed into the reactor on demand for 24 h. Com-
plete conversion was confirmed via GC-MS. The reaction was
terminated by cooling to ambient temperature and depressuriza-
tion. The catalyst was filtered off and the product was obtained
as colourless liquid by vacuum distillation.
i
CrIII(acac)3, Pr PNP, MAO have revealed that an increase in
pressure from 20 to 45 bar led to an increased formation of
1-octene [21]. However, studies at even higher pressures (up
to 100 bar) have shown that using the (Ph2P)2N–R ligand sys-
tem, the maximum yield of 1-octene cannot be boosted beyond
the 75% limit by increasing the ethylene concentration [22].
More detailed kinetic studies based on the catalyst system
generated from CrIII(acac)3; 1,2,3,4-tetrahydronaphthyl PNP
and MMAO-3A have suggested that the formation of 1-octene
follows a second-order dependence in ethylene concentration,
whereas 1-hexene formation is first order [23].
2.3. Ligand synthesis: general procedure
The alkylcyclohexyl PNP ligands were synthesised by re-
acting the amine and diphenylphosphine chloride Ph2PCl, as
described in the literature [24–26]. The 31P NMR shifts (ppm)
of all of the ligands are given in the supporting information.
2.4. Tetramerization runs: general procedure
2. Experimental
Before each run, the respective autoclave was heated under
vacuum at elevated temperature (120 ◦C) overnight and set un-
der inert gas atmosphere. The reaction vessel was then charged
with solvent and heated to the desired reaction temperature.
Stock solutions of the catalyst components (with chromium-
to-ligand-to-MMAO ratios as indicated in Tables 1, 2, and 3)
were transferred into a Schlenk vessel. The catalyst mixture was
stirred for 1 min and transferred to the autoclave, and the reac-
tion was started by pressurization with ethylene to the desired
pressure. Isothermal process conditions were ensured through-
out the duration of the run, with the ethylene fed on demand.
2.1. General comments
All synthetic work was carried out under argon using stan-
dard Schlenk technique. Solvents were purchased from Aldrich
and dried over Na/benzophenone or CaH2. Ethylene 3.5 was
supplied by Linde and used as received. NMR spectra were
recorded on a Bruker DPX-300 FT spectrometer. GC-MS spec-
tra were recorded on a Varian Saturn 2100T. GC/FID analy-
ses were carried out on either a Varian 3900 chromatograph
using a CP Sil PONA CB 50 m × 0.21 mm column or a

S. Kuhlmann et al. / Journal of Catalysis 245 (2007) 279–284
281
Table 1
Ethylene tetramerization with N-cycloalkyl PNP ligands
a
No
Ligand
Time
(min)
Productivity
(g/(g-Cr h))
S
S
S
S
S
S
S
C
S
Solids
(wt%)
α
C
cy-C in C
1-C in C
C
1-C in C
C
10–C
6
6
6
6
6
8
8
8
14
(wt%)
(%)
(%)
(wt%)
(%)
(%)
(%1)6+
(wt%)
1
2
3
4
5
6
1
2
3
4
5
6
35
41
28
25
22
25
464,396
390,650
544,629
725,280
737,536
757,720
19.5
19.2
18.6
18.7
19.8
22.6
51.9
48.9
34.2
25.2
24.9
15.0
44.3
47.8
63.9
73.9
74.3
84.6
62.5
61.2
63.7
67.2
68.1
66.4
96.5
97.0
98.1
99.0
99.0
99.4
8.9
10.4
9.3
8.2
7.5
4.8
5.6
6.4
4.3
2.6
1.6
68.9
68.6
74.4
80.4
82.1
84.9
1.8
1.3
2.0
0.9
1.2
0.9
8.0
a
◦
All reactions were carried out in 300 ml Parr reactor using 100 ml methylcyclohexane at 60 C, 45 barg, 5 µmol Cr(acac) , 7.5 µmol ligand, 270 eq. MMAO-3A.
3
Table 2
a
Ethylene tetramerization with substituted N-cyclohexyl PNP ligands
No
Ligand
Time
(min)
Productivity
(g/(g-Cr h))
Efficiency
(g/(g-Cr))
S
S
S
S
S
S
Solids
(wt%)
α
C
cy-C in C
1-C in C
C
1-C in C
6
6
6
6
6
8
8 8
(wt%)
(%)
(%)
(wt%)
(%)
(wt%)
7
8
9
10
11
12
4
7
8
9
10
11
16.5
15
13.5
18
13.0
30
2,146,787
2,279,200
2,191,555
1,878,413
2,134,983
666,800
591,082
569,800
493,100
563,524
461,183
333,400
19.4
26.6
26.5
29.6
43.0
31.3
24.0
14.3
11.4
6.9
4.4
11.4
75.0
85.4
88.3
92.9
95.4
88.3
68.3
63.7
64.5
59.8
48.7
60.8
99.0
99.5
99.6
99.6
99.7
99.4
82.2
86.0
87.6
87.1
89.5
88.0
0.8
0.7
0.3
1.3
0.3
1.2
a
◦
Reaction conditions: 60 C, 45 barg, 2.5 µmol Cr(acac) , 2.5 µmol ligand, 270 eq. MMAO-3A (540 for entries 7 and 10), 100 ml cyclohexane.
3
Table 3
a
Ethylene tetramerization with various N-substituted PNP ligands
No
Ligand
Time
(min)
Productivity
(g/(g-Cr h))
S
S
S
S
S
S
Solids
(wt%)
α
C
cy-C in C
1-C in C
C
1-C in C
6
6
6
6
6
8
8 8
(wt%)
(%)
(%)
(wt%)
(%)
(wt%)
13
14
15
16
12
13
14
15
33
92
26
20
479,818
167,386
614,662
845,001
19.8
25.4
27.1
22.8
51.3
13.2
10.7
28.7
45.8
86.5
89.1
70.5
63.9
63.3
62.8
65.3
97.3
99.4
99.5
98.8
71.6
84.9
86.7
80.5
1.6
1.9
0.8
3.0
a
◦
All reactions carried out in 300 ml Parr reactor using 100 ml methylcyclohexane at 60 C, 45 barg, 5 µmol Cr(acac) , 7.5 µmol ligand, 270 eq. MMAO-3A.
3
After the indicated reaction time, the autoclave was cooled to
0 ◦C in an ice bath and slowly depressurized. A sample of the
liquid reaction mixture was filtered and taken for GC analysis.
Solid byproducts were collected by filtration, dried in an oven
overnight, and weighed.
larly promising; initial experiments revealed that exchanging a
methyl-substituent for an isopropyl or cyclohexyl group led to
a significant increase in 1-hexene selectivity within the C6 frac-
tion, whereas the selectivity toward 1-octene remained fairly
unchanged [16].
These encouraging results prompted us to synthesise a
broader range of N-cycloalkyl PNP ligands with varying ring
sizes to gain information on the substitution pattern that would
be necessary for improved selectivity. The ligands were synthe-
sised according to standard methodology [24–26] and tested in
a pressure autoclave. The results of the ethylene tetramerization
experiments with ligands 1–6 (see Fig. 1) are given in Table 1.
As is evident from Table 1, ligands with small ring size
substituents on the nitrogen (i.e., cyclopropyl and cyclobutyl)
proved to form comparatively unselective catalyst systems, be-
cause the combined yield of 1-octene and 1-hexene was rather
low. This is due to the increased formation of all of the above-
mentioned side products. However, a systematic increase in the
ring size from cyclobutyl to cyclododecyl led to an incremental
improvement in desired selectivity. A striking feature of Ta-
ble 1, is that a 16% improvement in the total alpha selectivity
could be achieved by increasing the ring size from 3 to 12 car-
bon atoms. Although this can be attributed to lower levels of
all the side products, it is mainly due to an increased 1-hexene
selectivity within the C6 cut (from 44 to 85%). A decrease in
3. Results and discussion
3.1. Optimization of cycloalkyl PNP ring size
The aim of this study was to optimize the ligand struc-
ture to achieve higher combined selectivity toward the two
most desired products (1-octene and 1-hexene) at the expense
of side product formation. The main side products produced
by these catalyst systems are cyclic products (namely methyl-
and methylenecyclopentane), higher 1-alkenes generated by en-
larged metallacycle growth, and C10–C14 branched olefins from
the secondary incorporation of the main products, as well as
small amounts of polyethylene (PE), into the catalytic cycle.
Although the latter reaction depends mainly on product con-
centration and catalyst time on stream [21], the formation of
cyclic side products potentially can be limited by appropriate
ligand modification.
In an effort to maximize overall selectivity to desired prod-
ucts, we decided to study the influence of substituents at-
tached to the N atom of the PNP moiety. This seemed particu-

282
S. Kuhlmann et al. / Journal of Catalysis 245 (2007) 279–284
Fig. 1. N-cycloalkyl PNP ligands.
the C16+ fraction also contributes to the improved total alpha
selectivity, although to a lesser extent. It is also evident from
Table 1 that an increase in ring size from 4 to 7 carbon atoms,
led to a 6.9% increase in 1-octene selectivity, whereas a further
increase to 12 carbons caused a slight decrease. Interestingly,
the catalyst activity also increases with increasing ring size of
the cycloalkyl moiety, probably due to an increase in catalyst
solubility.
Fig. 2. 2- and 2,6-alkylcyclohexyl PNP ligands.
3.2. Alkyl-substituted cyclohexyl PNP ligands
Based on the above results, a cyclohexyl moiety seemed to
be a suitable skeleton to probe the effect of substituents on the
cycloalkyl group that are in close vicinity to the N atom. This
would allow tailoring of the steric environment of the ligand
with the aim of further enhancing catalyst selectivities. Initially,
it was decided to explore the effect of alkyl substituents on
the 2 position of the cyclohexyl moiety. Because the variety
of commercially available 2-substituted cyclohexylamines is
limited (only 2-methylcyclohexylamine is a commercial com-
pound), different synthetic approaches to prepare these com-
pounds were evaluated. It was found that the most straightfor-
ward route was the hydrogenation of the corresponding aniline
derivatives, using Ru on carbon as the catalyst. This approach
has several major advantages over other pathways, including
ease of reaction, high yields, and facile scale-up.
After preliminary optimization of the hydrogenation con-
ditions for the specific substrate 2,6-dimethylaniline, the re-
action conditions selected were 135 barg hydrogen pressure
and a temperature of 120 ◦C. Using this protocol, all tar-
geted saturated amines could be obtained in good yields by
vacuum distillation. Side reactions like dealkylation, hydrox-
ylation, and condensation—leading to aniline, alcohols and
secondary/tertiary amines, respectively—were negligible under
these conditions. The further synthesis of the equivalent PNP
ligands was carried out according to the literature [24–26] and
was reasonably straightforward for all substrates except 2,6-
diethylcyclohexyl PNP, which could not be synthesised via this
route, probably due to its steric encumbrance. These ligands
were again evaluated for efficacy as selective oligomerisation
catalysts, and the data is reported in Table 2 and Fig. 2.
Fig. 3. Proposed steric interactions between ligands 7 and 10 and metallacycle.
to 2-isopropylcyclohexyl PNP). The activities achieved with
these ligands were around 2,000,000 g/(g-Cr h) and are to our
knowledge among the highest activities reported thus far for a
selective ethylene tri- or tetramerization catalyst.
In an effort to maximize 1-hexene selectivity we decided to
study the influence of further bulk at the cyclohexyl moiety by
introducing two methyl groups (i.e., in the 2 and 6 positions).
Consistent with the aforementioned trend (increased bulk leads
to increased 1-hexene formation), an even higher 1-hexene se-
lectivitiy of 95.4% in the C6 fraction could be achieved with
this ligand (entry 11, Table 2). Most surprisingly, however,
this went hand in hand with a significant swing from pre-
dominantly tetramerization to a 1:1 1-hexene/1-octene ratio.
This unexpected behaviour could be explained if one consid-
ers the steric environment around the catalyst centre. Fig. 3
depicts the proposed steric interactions between ligands 7 and
10 and the growing chromacycle. When using 2-substituted cy-
clohexyl PNP ligands the sterically demanding substituent in
all probabilities does not point toward the metal centre due
to geometric constraints. Thus, increasing the bulk from 2-
methyl to 2-isopropyl only allows rather subtle selectivity tun-
ing due to rather long-range interactions. Introducing a second
methyl group, however, leads to more pronounced interactions
with the catalyst centre, because it is obviously impossible for
both substituents to rotate in a way that would leave the cen-
tre unaffected. If we further assume that chromacyclononane is
sterically more demanding than chromacycloheptane, then the
swing from tetramerization to 1:1 1-hexene/1-octene could be
explained by the preferential elimination of 1-hexene by the
steric interactions between ligand and metal centre. A similar
trend (increased steric bulk leads to more 1-hexene) has been
As was observed with increasing ring size of the ligands re-
ported in Table 1, the general trend was that substitution with
increasingly bulkier alkyl groups at the 2-position of the cy-
clohexyl moiety (ligands 4–9) led to an increase in total alpha
selectivity. The more sterically encumbered ligands favoured
the formation of 1-hexene over 1-octene (selectivity toward 1-
octene drops from 63.7% to 59.8% when going from 2-methyl

S. Kuhlmann et al. / Journal of Catalysis 245 (2007) 279–284
283
and combined alpha selectivity. Further optimization of com-
bined alpha-selectivity could be effected by substitution of alkyl
groups at the 2-position of the cyclohexyl moiety. Increasing
the steric encumbrance of these substituents from methyl to
ethyl to isopropyl suppressed the formation of the side prod-
ucts methyl- and methylenecyclopentane. Activities in excess
of 2,000,0000 g/(g-Cr h) could be achieved for selected lig-
ands. Increasing the steric encumbrance even further by hav-
ing methyl substituents on both the 2- and 6-positions led to a
significant swing away from 1-octene formation toward a 1:1
1-hexene/1-octene mixture.
Fig. 4. Various N-substituted PNP ligands.
reported in the literature for modifications at the P atom of
the ligand. Increasing the number of ortho-alkyl-aryl groups in
Ar2PN(CH3)PAr2 from 1 to 4 led from predominantly tetramer-
ization to trimerization [20].
Supporting information
We also tested 1,2,3,4-tetrahydronaphthyl PNP 11 (entry 12,
Table 2) with chromium as a tetramerization catalyst, due to its
structural similarity to the ligands reported herein. Again, the
improved 1-hexene selectivity and thus higher overall yield was
observed with this ligand (88.3% in the C6 cut). Due to the ster-
ically demanding aromatic moiety attached to the cyclohexyl
skeleton, the ratio of 1-octene to 1-hexene was 2:1 and thus
comparable with that of 2-isopropylcyclohexyl PNP 9. How-
ever, the activity was significantly lower than for the previous
ligands; rates around 650,000 g/(g-Cr h) could be achieved.
The online vision on this article contains additional supple-
mentary information.
Please visit doi: 10.1016/j.jcat.2006.10.020.
References
[1] An excellent general overview is provided by: D. Vogt, in: B. Cornils,
W.A. Herrmann (Eds.), Applied Homogeneous Catalysis with Organo-
metallic Compounds, vol. 1, Wiley–VCH, New York, 2002, p. 240.
[2] For a more recent review on selective trimerization see: J.T. Dixon, M.J.
Green, F.M. Hess, D.H. Morgan, J. Organomet. Chem. 689 (2004) 3641–
3668.
3.3. Other N-substituted PNP ligands
[3] R.M. Manyik, W.E. Walker, T.P. Wilson, J. Catal. 47 (1977) 197–209.
[4] J.R. Briggs, Chem. Commun. 11 (1989) 674–675.
[5] A. Carter, S.A. Cohen, N.A. Cooley, A. Murphy, J. Scutt, D.F. Wass,
Chem. Commun. (2002) 858–859.
[6] D.H. Morgan, S.L. Schwikkard, J.T. Dixon, J.J. Nair, R. Hunter, Adv.
Synth. Catal. 345 (2003) 939–942.
[7] D.S. McGuinness, P. Wasserscheid, W. Keim, D. Morgan, J.T. Dixon,
A. Bollmann, H. Maumela, F. Hess, U. Englert, J. Am. Chem. Soc. 125
(2003) 5272–5273.
[8] D.S. McGuinness, P. Wasserscheid, W. Keim, J.T. Dixon, J.J.C. Grove,
C. Hu, U. Englert, Chem. Commun. (2003) 334–335.
[9] K. Blann, A. Bollmann, J.T. Dixon, F.M. Hess, E. Killian, H. Maumela,
D.H. Morgan, A. Neveling, S. Otto, M.J. Overett, Chem. Commun. (2005)
620–621.
[10] D.S. McGuinness, P. Wasserscheid, D.H. Morgan, J.T. Dixon, Organo-
metallics 24 (2005) 552–556.
[11] C. Andes, S.B. Harkins, S. Murtuza, K. Oyler, A. Sen, J. Am. Chem. Soc.
123 (2001) 7423–7424.
[12] P.J.W. Deckers, B. Hessen, J.H. Teuben, Organometallics 21 (2002) 5122–
5135.
[13] J. Huang, T. Wu, Y. Qian, Chem. Commun. (2003) 2816–2817.
[14] C. Bianchini, G. Mantovani, A. Meli, F. Migliacci, Organometallics 23
(2003) 2545–2547.
After the encouraging results achieved with the substituted
N-cyclohexyl PNP ligands, it was decided to broaden the re-
search and explore ligands with structurally related cycloalkyl
patterns (see Fig. 4). Although ligands 12 and 13 are very close
variations of the cyclohexyl-alkyl motif, ligands 14 and 15 con-
strain the flexibility of the cyclohexyl ring by methylene bridges
and introduce more bulk in the 4 position of the cyclohexyl
moiety in the case of 15. The results of the tetramerization ex-
periments are given in Table 3.
Comparing the data obtained for ligand 12 with that of the
cyclohexyl-PNP derivative 4 (entry 4, Table 1), it is clear that
introduction of a methylene spacer between the N atom and
cyclic moiety, and thus subsequent relief of steric strain, led to
a ca. 9% drop in total alpha selectivity, due mainly to increased
cyclic C6 formation. Substitution of an H atom for a methyl
group on the methylene spacer and consequent reintroduction
of steric strain (ligand 13), resulted in a ca. 13% gain in over-
all alpha selectivity. The selectivity obtained with this ligand
compares favourably with the structurally similar ligand 7. The
slightly increased bulk of ligand 14 produced marginal selectiv-
ity gains relative to ligand 7. As might be expected, removal of
the 2-methyl substituent (ligand 15) gave comparable selectivi-
ties to the cyclohexyl-PNP (ligand 4).
[15] B. Hessen, J. Mol. Catal. A 213 (2004) 129–135.
[16] A. Bollmann, K. Blann, J.T. Dixon, F.M. Hess, E. Killian, H. Maumela,
D.S. McGuinness, D.H. Morgan, A. Nevelling, S. Otto, M. Overett,
A.M.Z. Slawin, P. Wasserscheid, S. Kuhlmann, J. Am. Chem. Soc. 126
(2004) 14712–14713.
[17] WO 2004056478.
[18] WO 2004056479.
4. Conclusions
[19] M.J. Overett, K. Blann, A. Bollmann, J.T. Dixon, D. Haasbroek, E. Killian,
H. Maumela, D.S. McGuinness, D.H. Morgan, J. Am. Chem. Soc. 127
(2005) 10723–10730.
[20] K. Blann, A. Bollmann, J.T. Dixon, F. Hess, E. Killian, H. Maumela, D.H.
Morgan, A. Neveling, S. Otto, M.J. Overett, Chem. Commun. (2005) 620–
621.
In conclusion, we were able to show that excellent fine-
tuning of the selectivity is possible for the Cr-catalyzed tetra-
merization of ethylene by using a number of N-substituted
bis(diphenylphosphino)amine PNP ligands. An increase in ring
size from 3 to 12 carbon atoms of the cycloalkyl moiety
leads to drastic improvements in both catalyst performance
[21] R. Walsh, D.H. Morgan, A. Bollmann, J.T. Dixon, Appl. Catal. A Gen. 306
(2006) 184–191.

284
S. Kuhlmann et al. / Journal of Catalysis 245 (2007) 279–284
[22] S. Kuhlmann, J.T. Dixon, M. Haumann, D.H. Morgan, J. Ofili, O. Spuhl,
N. Taccardi, P. Wasserscheid, Adv. Synth. Catal. 348 (2006) 1200–
1206.
[23] S. Kuhlmann, K. Blann, M. Ehrig, J.T. Dixon, M. Haumann, D.H. Mor-
gan, K. Obert, N. Taccardi, P. Wasserscheid, J. Am. Chem. Soc. (2006),
submitted for publication.
[24] S.J. Dossett, A. Gillon, A.G. Orpen, J.S. Fleming, P.G. Pringle, D.F. Wass,
M.D. Jones, Chem. Commun. (2001) 699.
[25] M.S. Balakrishna, T.K. Prakasha, S.S. Krishnamurthy, J. Organomet.
Chem. 390 (2) (1990) 203.
[26] N.A. Cooley, S.M. Green, D.F. Wass, K. Heslop, A.G. Orpen, P. Pringle,
Organometallics 20 (2001) 4769.