.
Angewandte
Communications
Table 2: Continuous gas-phase hydroaminomethylation of ethylene and
diethylamine using SILP catalysts (Rh-Xantphos in [MMMIM][NTf2])—
influence of different support materials.[a]
tandem reaction of hydroformylation, amination, and enam-
ine hydrogenation using Rh-Xantphos complexes as the
immobilized homogeneous catalyst. Interestingly, the main
effect of the ionic liquid solvent (compared to reaction of the
Rh-complex dissolved in supported aldol products in the
absence of the IL; Table 1) is on the competition between the
desired amination/hydrogenation and the undesired aldol
condensation. While the IL-free system and the SILP systems
with a basic acetate coating form almost exclusively aldol
condensation products from the initial hydroformylation
product propanal, SILP catalysts based on neutral oxide and
porous carbon supports and ILs of low basicity and lip-
ophilicity result in very selective hydroaminomethylation
catalysis. The long-term stability of the developed systems is
remarkably high and this highlights the potential of these
SILP catalysts for future applications in industrial amine
production.
Support (PZC)[b]
X
[%]
TOF
Yield [%][c]
propanal
[hÀ1
]
aldol prod.
DEPA
Silica 100 (6.38)
Pural TH 60 (7.95)
AlOx neutral (7.69)
PBSAC[d] (11.3)
51
32
22
72
351
219
144
520
13
13
5
2
8
1
0
35
10
15
70
2
[a] Reaction conditions: 1208C, 10 bar, L=Xantphos, L/Rh 5:1, 0.2 wt%
Rh, IL=[MMMIM][NTf2], aIL =0.1, 2.5 g SILP catalyst, 200 NmLminÀ1
H2, 100 NmLminÀ1 CO, 10 NmLminÀ1 ethylene, 1 ghÀ1 diethylamine,
residence time=8.4 s. Values after 20 h time-on-stream. [b] PZC=point
of zero charge; PZC values were measured to compare the acidity of the
support (see Supporting Information for details). [c] determined by GC
(enamine, alcohols, and higher boiling aldol condensation products were
less than 1%). [d] 2 ghÀ1 diethylamine.
Experimental Section
SILP catalysts were prepared by impregnation under argon atmos-
phere using standard Schlenk techniques.
A solution of Rh-
(acac)(CO)2 (0.0144 g, 0.056 mmol) and Xantphos (4,5-bis(diphenyl-
phosphino)-9,9-dimethylxanthene; 0.162 g, 0.28 mmol) in dichloro-
methane was stirred for 20 min and the respective amount of ionic
liquid (0.36–0.42 g) equal to aIL = 0.1 (aIL = VIL/Vpore) was added.
Finally, 2.86 g of the support calcinated Silica 100 (pore volume
V
pore = 0.98 mLgÀ1, BET surface area = 358.9 m2gÀ1), Pural TH 60
(Vpore = 0.60 mLgÀ1, BET surface area = 325.63 m2 gÀ1), Alox neutral
(Vpore = 0.28 mLgÀ1, BET surface area = 1358 m2 gÀ1), or activated
carbon (PBSAC from Blꢀcher; Vpore = 1.18 mLgÀ1, BET surface
area = 2005 m2 gÀ1) was added and the suspension was stirred for
another 10 min before the solvent was slowly removed under reduced
pressure at 408C. The continuous gas-phase hydroaminomethylation
experiments were conducted at 1208C and 10 bar pressure using
Figure 3. Hydroaminomethylation of ethylene and diethylamine with
SILP catalyst (Rh-Xantphos in [MMMIM][NTf2] on PBSAC support).
^
Conversion of ethylene ( ), yield of aldol adduct (&), yield of aldol
a
fixed bed reactor set-up. The gaseous feedstocks (H2
condensation product (*), yield of propionaldehyde (*), yield of
DEPA (&)—Reaction conditions: 1208C, 10 bar, L=Xantphos, L/Rh
5:1, 0.2 wt% Rh, PBSAC, aIL =0.1, 2.5 g SILP catalyst, 200 NmLminÀ1
H2, 100 NmLminÀ1 CO, 10 NmLminÀ1 ethylene, 0–17 h; 1 ghÀ1 diethyl-
amine, 17–40 h; 2 ghÀ1 diethylamine, residence time=8.4 s.
200 NmLminÀ1, CO 100 NmLminÀ1) were fed by mass flow con-
trollers (MFC) and ethylene (10 NmLminÀ1) through a mini coriolis
flow (mini CORI) from Bronkhorst. Liquid diethylamine (1–2 ghÀ1
)
was pumped through an HPLC pump (Techlab) and fed into
a controlled evaporator and mixer (CEM) from Bronkhorst, where
it was evaporated. The gaseous feed stream led to a reactor of 30 cm3
volume where a fixed bed of SILP material (2.5 g) catalyzed the
reaction with a residence time of 8.4 seconds. The product feed stream
was continuously analyzed by a DANI Master GC equipped with an
FID detector. All gas flow meters and the GC were calibrated prior to
use. Additional experimental details can be found in the Supporting
Information.
PBSAC support for more than 18 days time-on-stream reach-
ing a total turnover number of 115000. [NTf2]À based ionic
liquids are particularly suitable for such long-term testing,
because these hydrophobic ILs are characterized by high
stability against thermal decomposition and hydrolysis.[21]
After a slight initial deactivation (by approximately 20%),
the catalyst activity remained almost stable during this
extended period at a level of TOF = 450 hÀ1 with an average
DEPA selectivity of more than 99% (for details see the
Supporting Information). No formation of aldol side products
and enamines was observed during the entire time of
operation. During the long-term experiment, the product
condensate was analyzed for traces of Rh, phosphine, and
ionic liquid by inductively coupled plasma atomic emission
spectroscopy (ICP-AES). Gas-phase leaching was found to be
negligible, with all relevant elements remaining below the
detection limit of 2 ppm.
Received: February 15, 2013
Revised: March 29, 2013
Published online: May 16, 2013
Keywords: amine formation · hydroaminomethylation ·
.
hydroformylation · ionic liquids · supported catalysts
[1] W. Reppe, H. Vetter, Liebigs Ann. Chem. 1953, 582, 133.
[2] W. Reppe, Experientia 1949, 5, 93.
[4] P. Eilbracht, L. Bꢁrfacker, C. Buss, C. Hollmann, B. Kitsos-
Rzychon, C. Kranemann, T. Rische, R. Roggenbuck, A.
In conclusion, we have demonstrated that SILP materials
are suitable catalyst materials for the continuous gas-phase
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 6996 –6999