Heterogeneous Coupling of Phenylethyne over Cu-Mg-Al Mixed Oxides: Influence of Catalyst Composition and Calcination Temperature on Structural and Catalytic Properties

Article abstract of DOI:10.1006/jcat.1997.1670

Cu-Mg-Al mixed oxides in basic media are shown to be efficient catalysts in the oxidative coupling of phenylethyne to 1,4-diphenylbuta-1,3-diyne. Catalysts containing 33 at.% of Cu(II), related to the total metal content, with different ratios of Mg(II) and Al(III) were prepared by precipitation of metal nitrates with sodium carbonate and calcined at different temperatures. The samples were characterized by means of XRD, TEM, TG-MS, XPS, N₂O titration, and N₂ physisorption. The catalytic behavior was tested in n-butanol (0-0.1 M NaOH) in an autoclave under oxygen atmosphere (0.5-4 MPa) at different temperatures (303-353 K). Highest yields of 1,4-diphenylbuta-1,3-diyne (84%) were obtained with catalysts based on Cu-Mg-Al lamellar double hydroxides (LDHs). Typical selectivities to 1,4-diphenylbuta-1,3-diyne were in the range of 85-92%, depending on reaction conditions and conversion. Cu(II) on the catalyst surface was found to be crucial as a redox site in the coupling of phenylethyne. However, the mere presence of Cu(II) on the surface is not sufficient for a highly active catalyst. Cu(II) must be embedded into a specific matrix in order to exhibit high catalytic activity. The hydrotalcite-like phase (LDH) in Cu-Mg-Al catalysts provides such a matrix.

Full text of DOI:10.1006/jcat.1997.1670

JOURNAL OF CATALYSIS 169, 1–12 (1997)
ARTICLE NO. CA971670
Heterogeneous Coupling of Phenylethyne over Cu–Mg–Al Mixed Oxides
Influence of Catalyst Composition and Calcination Temperature on Structural
and Catalytic Properties
1
S. M. Auer, R. Wandeler, U. G o¨ bel, and A. Baiker
Department of Chemical Engineering and Industrial Chemistry, Swiss Federal Institute of Technology, ETH Zentrum, CH-8092 Z u¨ rich, Switzerland
Received June 28, 1996; revised November 18, 1996; accepted January 7, 1997
water. LDHs are also referred to as anionic clays (2–4), dou-
Cu–Mg–Al mixed oxides in basic media are shown to be effi- ble hydroxides (5, 6), hydrotalcite-like compounds (7–9), or
cient catalysts in the oxidative coupling of phenylethyne to 1,4- pillared hydroxycarbonates (10). De Roy et al. (11) have re-
diphenylbuta-1,3-diyne. Catalysts containing 33 at.% of Cu(II),
related to the total metal content, with different ratios of Mg(II)
and Al(III) were prepared by precipitation of metal nitrates with
sodium carbonate and calcined at different temperatures. The sam-
ples were characterized by means of XRD, TEM, TG-MS, XPS,
LDHs are either used as such or, more often, after calcina-
N2O titration, and N2 physisorption. The catalytic behavior was
tested in n-butanol (0–0.1 M NaOH) in an autoclave under oxy-
gen atmosphere (0.5–4 MPa) at different temperatures (303–353
K). Highest yields of 1,4-diphenylbuta-1,3-diyne (84%) were ob-
viewed LDHs concentrating on preparation, structure, ion
exchange, and pillaring.
In a comprehensive review Cavani et al. (1) discuss the
structural and catalytic properties of LDHs. In catalysis,
tion. Calcined LDHs form highly dispersed mixed oxides
stable to thermal treatments. They possess basic and redox
properties, depending on the metal cations built into the
tained with catalysts based on Cu–Mg–Al lamellar double hydrox- layered matrices. In basic catalysis, LDHs are applied in
ides (LDHs). Typical selectivities to 1,4-diphenylbuta-1,3-diyne polymerization of alkene oxides (12, 13), aldol condensa-
were in the range of 85–92%, depending on reaction conditions tions (14), and other reactions. In redox catalysis, they are
and conversion. Cu(II) on the catalyst surface was found to be
crucial as a redox site in the coupling of phenylethyne. How-
ever, the mere presence of Cu(II) on the surface is not suffi-
cient for a highly active catalyst. Cu(II) must be embedded into
a specific matrix in order to exhibit high catalytic activity. The
hydrotalcite-likephase(LDH)in Cu–Mg–Alcatalystsprovidessuch
used as mixed oxide catalysts in gas-phase reactions such as
methanol synthesis (15) and a few fine chemical reactions,
e.g., Baeyer–Villiger oxidation with oxygen and benzalde-
hyde (16).
In order to investigate the catalytic properties of LDHs,
we focused on a catalytic reaction which involves both basic
and redox elementary steps and is catalyzed by uncalcined
as well as calcined LDHs. In a preliminary study (17) we
have shown that the oxidative coupling of phenylethyne
to 1,4-diphenylbuta-1,3-diyne catalyzed by a Cu–Mg–Al
a matrix.
�c 1997 Academic Press
INTRODUCTION
LDH and sodium hydroxide fulfills these requirements
Lamellar double hydroxides (LDHs) consist of pos-
itively charged metal hydroxide layers separated from
each other by anions and water molecules. The layers
contain metal ions of at least two different oxidation
states. The most common case of di- and trivalent cations
leads to the general chemical formula [M(II)1� xM(III)x
(
Scheme 1). Sodium hydroxide must be used as a cocatalyst
since the intrinsic basicity of the LDH does not suffice for
the deprotonation step of the coupling reaction. Modifica-
tion of the reactant leads to substituted 1,4-diphenylbuta-
1
,3-diynes which form an important class of liquid crystals
x+
n�
for use in infrared and long wavelength application (18).
Note that previous to our studies only the homogeneously
catalyzed coupling of phenylethyne has been reported
(
OH)2] (A )x/n � mH2O, where x refers to the n[M(III)]/
n[M(II) + M(III)]ratio rangingbetween ca. 0.25and 0.4(1).
The positive charge of the metal hydroxide layers is com-
n�
(
19, 20).
pensated byinterstitiallayersbuilt ofanionsA and crystal
The aim of this study was to elucidate the relation
between structural and chemical properties of different
1
–
–
To whom correspondence should be addressed. Fax: +41 1 632 11 63. Cu Mg Al LDHs and mixed oxides and their catalytic be-
E-mail: a.baiker@tech.chem.ethz.ch.
havior in the coupling of phenylethyne.
1
0021-9517/97 $25.00
Copyright �c 1997 by Academic Press
All rights of reproduction in any form reserved.

2
AUER ET AL.
SCHEME 1. Reaction scheme of the Cu-catalyzed oxidative coupling of phenylethyne.
�
EXPERIMENTAL
and a heating rate of 10 K min 1 in the temperature range
of 298 to 1323 K. Evolving gases were monitored on-line
using a Balzers QMG 420 quadrupole mass spectrometer
connected to the thermoanalyzer by a heated capillary.
XPS analyses were performed on a Leybold LHS 11 in-
strument with MgK� radiation (240 W). The analyzer was
operated at 150 eV constant pass energy at a scale cali-
brated versus the Au4f7 line at 84.0 eV. Corrections of the
energy shift, due to the steady-state charging effect, were
accomplished taking the C1s line of adsorbed hydrocarbons
at 285.0 eV as internal standard. Samples were degassed at
Preparation
VariousamountsofCu(NO3)2 � 3H2O, Mg(NO3)2 � 6H2O,
and Al(NO3)3 � 9H2O (total metal nitrates, 0.24 mol) were
dissolved in 500 ml distilled water and transferred into a
1
l five-necked flask equipped with propeller mixer, reflux
/2
condenser, pH meter (Metrohm 614 pH meter), thermo-
couple, and feeding unit for liquids. This system was equili-
brated at 333 � 2 K under vigorous stirring. Subsequently,
3
(
00 to 500 ml 0.75 M Na2CO3 solution was slowly added
�
9
2
� 10 mbar for 3 days before analysis. For an estimation
Metrohm 665 Dosimat) to the metal nitrate solution over
of the surface composition in mol% , the intensities of the
elemental peaks Na1s, C1s, O1s, Cu2p3 , Mg2p, and Al2s
were determined by a curve-fitting program (25) and di-
vided by the corresponding atomic sensitivity factors (26).
Cu surface areas were determined by N2O titration us-
ing a pulse technique similar to that reported by Evans
et al. (27). H2 and He were of ultra high purity. He was
further freed from O2 using an Oxisorb system (Messer
Griessheim). N2O (Matheson, 99.998% ) was dried over
a column filled with activated molecular sieve, type 4 A˚ .
a time period of 2 h until a pH of 8 was reached. The light-
blue suspension was stirred for 15 h at room temperature
and filtered. The precipitates were intensively washed with
distilled water and dried at 363 K below 10 kPa for 24 h.
The obtained samples were calcined for 4 h at 15 Pa and
different temperatures.
/2
Preliminary tests using other preparation methods like
precipitation at low or high supersaturation according to
(
1) led to catalysts with lower activity in the phenylethyne
coupling reaction.
Samples were reduced in flowing H2/Ar (5% hydrogen,
3
� 1
� 1
.
7
5 cm min ) by heating from 373 to 548 K at 5 K min
Characterization
The samples were held at this temperature for 15 min and
3
� 1
Catalyst materials were investigated with regard to their then exposed to pure H (75 cm min ) for 30 min at the
2
3
physical and chemical properties by means of powder same temperature. The H was purged with He (50 cm
X-ray diffraction (XRD), transmission electron microscopy min ) at 548 K. After cooling to 363 K under He, pulses
2
�
1
3
(
TEM), thermoanalytical analysis (TG, DTA) in combina- of N O (0.5 cm ) were injected. N and N O were sepa-
2 2 2
tion with mass spectrometry (MS), X-ray photoelectron rated on a Porapak N column and detected by means of a
spectroscopy (XPS), nitrous oxide titration, and nitrogen thermal conductivity detector (Gow Mac Instruments). Cu
1
9
physisorption.
metal surface areas were calculated assuming 1.46 � 10
2
X-ray diffraction patterns were measured on a Siemens Cu atoms/m (27, 28) and an adsorption stoichiometry of
D5000 powder X-ray diffractometer. Diffractograms were Cu /O
(ads)
recorded with detector-sided Ni-filtered CuK� radiation measurements was within 5% .
= 2. The accuracy of the copper surface area
(
s)
�
�
(
40 mA, 40 kV) over a 2�-range of 2 to 70 and a po-
BET surface areas (SBET), mean cylindrical pore diam-
�
sition sensitive detector using a step size of 0.010 and a eters (hd i) and specific adsorption pore volumes (Vp,N2 )
p
step time of 2.5 s. Measured patterns were compared with were obtained by physisorption of nitrogen at 77 K us-
JCPDS data files (21). Mean crystallite sizes were estimated ing a Micromeritics ASAP 2000 instrument. Prior to mea-
using the Scherrer equation (22, 23) and the widths of se- surement, the samples were degassed to 0.1 Pa at 473 K.
lected diffraction lines which were fitted with a Split Pear- SBET were calculated in a relative pressure range be-
son 7 function. Peak broadening due to the instrumental tween 0.05 < p/p < 0.2 assuming a cross-sectional area
0
�
2
line broadening of 2� = 0.08 was taken into account (24). of 0.162 nm for the nitrogen molecule. Pore size dis-
Transmission electron microscopy was carried out using a tributions were calculated applying the Barrett–Joyner–
Philips CM30ST electron microscope with a CeB6-cathode Halenda (BJH) method (29) to the desorption branch of
and an acceleration voltage of 300 kV.
the isotherms (30). The assessments of microporosity were
Thermoanalytical measurements (TG, DTA) were per- made from t-plot constructions (0.3 < t < 0.5 nm) using the
formed on a Netzsch STA 409 using �-Al2O3 as a reference Harkins–Jura correlation (31).

PHENYLETHYNE COUPLING OVER Cu–Mg–Al MIXED OXIDES
3
Catalytic Tests
0.33]T, [Cu-0.29]T, [Cu-0.25]T, [Cu-0.0]T), are depicted in
Fig. 1 as a function of both n[Cu(II)]/n[M(II)] and x value.
Coupling of phenylethyne was carried out in a 100-ml
stainless-steel autoclave (Baskerville R&A) equipped with
a 50-ml glass liner and a PTFE lid. The reaction mixture
was stirred with a magnetic PTFE stirrer (0–1500 rpm). The
reactor was heated by a 10 l heating bath (1T = � 2 K) con-
Characteristic Features of Phenylethyne
Coupling Reaction
The influence of the reaction parameters on the phenyl-
nected to a temperature controlling unit. Preliminary tests ethyne coupling reaction (Scheme 1) was investigated using
showed that a constant temperature of the reaction mixture [Cu-0.33]673. Conversion of phenylethyne and selectivity to
was reached after 30-min equilibration time at 100 rpm. 1,4-diphenylbuta-1,3-diyne depended on sodium hydrox-
Oxygen pressure in the reactor was maintained using a gas ide concentration (0–0.1 M), oxygen pressure (0.5–4 MPa),
dosing unit with reducing valve (Tescom) compensating for and reaction temperature (303–353 K) as shown in Table 1.
O2 consumed during the reaction.
Conversion of phenylethyne increased when these three
The reaction mixture comprised 2 g phenylethyne (Fluka reaction parameters were increased in the given ranges. In
9
8% , 2% styrene), 50 mg catalyst (� 180 �m particle size), contrast, selectivity decreased with increasing sodium hy-
and different amounts of dry NaOH dissolved in 25 ml droxide concentration and reaction temperature. Without
1
-butanol (0–0.1 M NaOH). In order to equilibrate the sys- sodium hydroxide, no conversion was observed. Prelimi-
tem, the reaction mixture was kept under N2 (99.998% ) at nary optimization of the reaction conditions ([Cu-0.33]673,
the desired reaction temperature for 30 min at 100 rpm. 0.1 M NaOH, 2 MPa O2, 353 K) led to 84% yield of 1,4-
The starting procedure, which lasted about 15 s, involved diphenylbuta-1,3-diyne (Table 1).
purging of the autoclave with O2 (99.998% ) and adjusting
the desired oxygen pressure (0.5–4 MPa) and stirring speed (>750 rpm) and catalyst particle size (<315 �m) indicating
1000 rpm). At the end of the reaction, the pressure was that the reaction rate was not influenced by interparticle or
The reaction rate was independent on stirring speed
(
released and the stirrer was stopped. Catalytic tests quoted intraparticle mass transfer limitations in the chosen param-
in this article were performed over a time period of 15 min. eter range. In addition, the reaction rate was linearly depen-
Phenylethyne conversion and selectivity to 1,4-diphenyl- dent on the amount of catalyst below amounts of 100 mg,
buta-1,3-diyne were determined byGC analysis(HP1, FID) implying that transfer of oxygen from gas to liquid phase
using pentadecane (Fluka) as internal standard. Conver- did not limit the observed reaction rate.
sions and selectivities were reproducible within � 2 and
Side products were determined in the reaction using
0.1 M NaOH solution and 2 MPa oxygen pressure at 333 K.
The reaction mixtures were analyzed for dissolved Cu At low conversion, butyl phenylacetate was produced
�
5% , respectively.
species by AAS using a Varian SpectrAA-10BQ atomic in a homogeneous side reaction consuming ca. 5% of
absorption spectrometer. After filtration through a 0.2-�m phenylethyne converted. At higher conversions no side
PTFE filter (Infochroma), the reaction samples were mixed products were detected by GC/MS, possibly due to poly-
with 0.1 M HNO3 in water and washed with diethylether merization or complete oxidation of butyl phenylacetate;
to suppress matrix effects of organics during measurement.
The aqueous phase was concentrated in a rotary evaporator
and analyzed using a VarianAA copper lamp (4 mA, slit 0.5,
TABLE 1
Dependence of Conversion of Phenylethyne and Selectivity to
,4-Diphenylbuta-1,3-diyne on NaOH Concentration, O2 Pressure,
3
24.8 nm).
1
and Reaction Temperature
RESULTS
All catalysts prepared, LDHs and mixed oxides, were
c(NaOH) O2 pressure Reaction temperature Conversion Selectivity
(
M)
(MPa)
(K)
(% )
(% )
based on Cu(II) as well as Mg(II) and/or Al(III). The
samples contained 33 at.% of Cu(II) related to the total
metal content, affording catalysts of type [Cu0.33Mg(0.67� x)–
Alx–CO3] according to the denotation suggested by De
Roy et al. (11). The x value of the samples, i.e., the ra-
tio of Al(III) to total metal content, was varied between
0
.1
2
2
2
4
1
0.5
2
2
2
333
333
333
333
333
333
353
323
313
303
74
47
32
84
26
14
97
16
5
86
88
92
87
85
70
87
94
99
—
0.075
0.05
0
0
0
0
0
.1
.1
.1
.1
.1
0
.67 and 0.0. The composition of our catalysts is fully de-
fined by the x value. We will therefore use the denota-
tion [Cu-x]T for our samples, where T is the calcination
temperature of the compound in K. Uncalcined catalysts
are indicated by the subscripts uc. The compositions of
0.1
.1
0
2
1
Note. Reactions were performed with [Cu-0.33]673 (calcined at 673 K)
for 15 min. Note that no reaction was observed in the absence of the
the investigated Cu–Mg–Al mixed oxides ([Cu-0.67]T, [Cu- catalyst.

4
AUER ET AL.
FIG. 1. Influence of catalyst composition on conversion of phenylethyne. Bulk composition of the Cu–Mg–Al mixed oxides is depicted as a function
of both n[Cu(II)]/n[M(II)] and x value (x = n[M(III)]/n[M(II) + M(III)]). Conversion of phenylethyne is proportional to the bubble diameter at the
corresponding catalyst composition. The dotted line represents catalyst compositions with a Cu(II) content of 33 at.% related to total metal content.
Phenylethyne conversion is additionally shown in a bar graph as a function of x value only. Reactions were performed in 0.1 M NaOH at 2 MPa O2
pressure and 333 K for 15 min; catalysts were calcined at 673 K. [Cu-x], catalyst with a Cu(II) content of 33 at.% related to total metal content
and the indicated x value; n[Cu(II)], molar amount of Cu(II); n[M(II)], molar amount of divalent metal; n[M(III)], molar amount of trivalent metal;
n[M(II) + M(III)], total molar amount of di- and trivalent metal.
however, neither the polymer nor the possible side product
Crystalline phases of the samples after precipitation,
CO2 could be detected. Moreover, about 0.2 mg of yellow calcination, and reaction were determined by XRD
copper(I)phenylethynilide (32) was formed, as evidenced (Fig. 2). After precipitation (Fig. 2a), two crystalline
by XRD. This compound exhibited some catalytic activity phases were detected, a hydrotalcite-like phase (HT-phase;
in the coupling reaction, contributing ca. 1% to the overall JCPDS 37-0630) and a magnesium-copper hydroxide ni-
conversion of phenylethyne.
trate phase (MgCu-phase; JCPDS 26-1221, 15-0014) (21).
The amount of copper dissolved in the reaction medium With increasing x value, the content of crystalline HT-
depended significantly on the sodium hydroxide concentra- phase increased, whereas the content of crystalline MgCu-
tion. Up to ca. 0.15 M sodium hydroxide, the concentration phase decreased. In [Cu-0.25]uc and [Cu-0.29]uc both phases
of dissolved copper remained below the resolution limit coexisted, whereas [Cu-0.0]uc and [Cu-0.33]uc showed
�
4
of AAS (<5 � 10 M Cu); higher sodium hydroxide con- exclusively reflections of one phase, MgCu or HT respec-
centrations led to an increase in the formation of soluble tively. The pattern of [Cu-0.67]uc only exhibited weak reflec-
�
3
copper hydroxide species (1.5 � 10 M dissolved copper tions similar to the HT-phase. After calcination at 673 K
in 0.4 M NaOH). In contrast to copper(I)phenylethynilide, (Fig. 2b), all samples contained crystalline CuO (JCPDS
the amount of copper dissolved at sodium hydroxide con- 41-0254) (21). Crystallinity of this phase decreased with
centrations below 0.15 M did not result in a detectable con- increasing x value (crystallite size data in Table 2). In
tribution to catalytic activity.
addition to CuO, [Cu-0.0]673 contained crystalline MgO
JCPDS 4-0829) (21). After reaction (Fig. 2c), all sam-
ples showed similar patterns to those before reaction. [Cu-
(
Catalyst Properties
Influence of composition. The catalytic activity in the 0.67]673, [Cu-0.33]673, and [Cu-0.29]673 contained crystalline
coupling of phenylethyne in relation to the bulk composi- Cu-phenylethynilide (32) after reaction. In [Cu-0.0]673 crys-
tion of the Cu–Mg–Al mixed oxides is shown in Fig. 1. For talline Mg(OH)2 (JCPDS 44-1482) (21) was detected after
constant Cu content, conversion generally increased with reaction instead of MgO present before reaction.
increasing x value, except for [Cu-0.67]673 which was slightly
Surface areas, pore volumes, and mean pore diame-
less active than [Cu-0.33]673. Interestingly, selectivity to 1,4- ters did not exhibit a singular trend (Table 2). However,
diphenylbuta-1,3-diyne was almost independent on catalyst graphically assessed maxima of the pore size distributions
composition (Table 3).
were similar for all samples, except for [Cu-0.67]673, which

PHENYLETHYNE COUPLING OVER Cu–Mg–Al MIXED OXIDES
5
FIG. 2. X-ray diffraction pattern of [Cu-0.0], [Cu-0.25], [Cu-0.29], [Cu-0.33], and [Cu-0.67]. (᭢) Hydrotalcite-like phase (HT-phase); (� ) Cu–Mg
hydroxide nitrate (MgCu-phase); ( ) CuO; ( ) MgO; (♦) Mg(OH)2; (᭹) Cu(I)phenylethynilide. (a) uncalcined; (b) calcined at 673 K; (c) after
reaction.
showed a bimodal pore size distribution with an additional monitored by MS (Fig. 3). The binary samples [Cu-0.0] and
microporous maximum. All isotherms were of type-IV with [Cu-0.67] exhibited a different thermal behavior than the
a type-H3 hysteresis according to IUPAC (33), which is ternary compounds, as expected from the major differences
generally observed for mesoporous aggregates of plate-like in composition. The decomposition pattern of [Cu-0.29]
particles. Since all samples exhibited similar morphology, contained characteristic features of both [Cu-0.25] and
intraparticle mass transfer was expected to be similar to [Cu-0.33].
that in [Cu-0.33]673 for all catalysts.
Cu surface areas of samples calcined at 673 K, deter-
Gases evolved by thermal decomposition of the uncal- mined by N2O titration and corroborated by XPS are listed
cined samples during thermoanalytical measurements were in Table 3. The ternary samples all exhibited higher Cu sur-
face areas than [Cu-0.0]673 and [Cu-0.67]673. After reaction,
the catalysts showed smaller Cu surface areas in N2O titra-
tion than after calcination, most likely due to deposition of
organic residues on the catalyst surface (during reaction),
which could not be removed completely during pretreat-
TABLE 2
Morphological Properties of Catalysts [Cu-x]673 (Calcined
at 673 K) Determined by N2 Physisorption and XRD
ment prior to N2O titration.
a
b
c
Crystallite sized
CuO (nm)
Bulk composition
SBET
(m g
Vp,N2
(cm g
dp,max (hdpi)
2
� 1
3
� 1
x value
)
)
(nm)
Influence of calcination temperature. The influence of
calcination temperature on structure and catalytic activity
in the coupling of phenylethyne was investigated for the
most active catalyst [Cu-0.33].
Catalytic activity in relation to calcination temperature is
listed in Table 4. Up to a calcination temperature of 773 K,
high conversions were obtained. Calcination at 873 K led
to a prominent loss of catalytic activity. Interestingly, selec-
tivity to 1,4-diphenylbuta-1,3-diyne did not depend signifi-
cantly on calcination temperature.
0
0
0
0
0
.67
.33
.29
.25
.0
131
72
55
73
36
0.13
0.52
0.38
0.41
0.29
3, 20 (5)
30 (29)
35 (28)
30 (22)
35 (30)
<3
3
10
15
20
a
b
c
BET surface area.
BJH cumulative desorption pore volume.
Graphically assessed maximum of pore size distribution (in parentheses:
mean pore diameter, hdpi = 4Vp,N /SBET).
2
d
Determined from XRD line broadening.

6
AUER ET AL.
FIG. 3. Evolving gases during thermal decomposition of [Cu-0.0], [Cu-0.25], [Cu-0.29], [Cu-0.33], and [Cu-0.67] as monitored by MS.
Crystalline phases of the sample after calcination at dif- cination at 673 K. With higher calcination temperatures,
ferent temperatures were determined by XRD (Fig. 4). crystalline CuO (JCPDS 41-0254), MgO (JCPDS 4-0829),
Cu-0.33]uc showed the diffraction pattern of hydrotalcite- and CuAl2O4 (JCPDS 33-0448) phases were formed (21).
[
like phases (JCPDS 37-0630) (21). With increasing cal- This behavior is in accordance with the literature (3). Mean
cination temperature, crystallinity of this phase declined, crystallite sizes of HT and CuO for different calcination
leading to a predominantly amorphous catalyst after cal- temperatures are listed in Table 5.
TABLE 3
TABLE 4
Cu Surface Area Measured by XPS and N2O Titration, Con-
version to Phenylethyne, and Selectivity to 1,4-Diphenylbuta-1,3-
diyne for Catalysts [Cu-x]673 (Calcined at 673 K)
Cu Surface Area Measured by XPS and N2O Titration, Con-
version to Phenylethyne, and Selectivity to 1,4-Diphenylbuta-1,3-
diyne for [Cu-0.33] Calcined at Different Temperatures
Cu surface area
Bulk
Cu surface area
Calcination
composition
x value
N2O titration
XPS
(% )
Conversion
(% )
Selectivity
(% )
temperature
(K)
N2O titration
XPS
(% )
Conversion
(% )
Selectivity
(% )
2
� 1
a
b
2
� 1
a
(m g
)
(m g
)
0
0
0
0
0
.67
.33
.29
.25
.0
3 (2)
—
10 {7}
8 {5}
4 {3}
—
68
74
62
19
1
86
86
87
90
—
Uncalcined
473
4
5
7
8
12
10
6 {4}
7 {5}
78
82
66
74
78
28
89
89
86
86
88
86
8 (4)
9 (6)
6 (5)
2 (1)
573
673
773
10 {7}
10 {7}
10 {7}
8 {6}
873
a
In parentheses: results after reaction.
Values in curled brackets represent Cu surface areas (m g ) esti-
mated by multiplication of XPS-derived Cu surface concentrations (% )
with corresponding BET surface areas (m g ).
b
2
� 1
a
2
� 1
Values in curled brackets represent Cu surface areas (m g ) esti-
mated by multiplication of XPS-derived Cu surface concentrations (% )
2
� 1
2
� 1
with corresponding BET surface areas (m g ).

PHENYLETHYNE COUPLING OVER Cu–Mg–Al MIXED OXIDES
7
FIG. 4. Crystalline phases of [Cu-0.33] after calcination at different temperatures determined by XRD. (᭢) Hydrotalcite-like phase (HT-phase);
) CuO; ( ) MgO; (᭜) CuAl2O4.
(
Surface areas, pore volumes, and mean pore diameters istic for LDHs (34). An exception was a CO2 peak in the
of [Cu-0.33] were hardly affected by calcination at differ- decomposition pattern around 870 K, which to our knowl-
ent temperatures (Table 5). All isotherms were of type-IV edge has not been reported so far. CO2 evolution at this
with a type-H3 hysteresis according to IUPAC (33), which remarkably high temperature was observed in all samples
is generally observed for mesoporous aggregates of plate- containing crystalline HT.
like particles. Because the samples exhibited similar mor-
phology, intraparticle mass transfer was not expected to be ied by TEM (Fig. 5). All samples showed a laminar struc-
controlling, as shown for [Cu-0.33]673. ture, typical for LDHs (11). Laminae of [Cu-0.33]uc and
Morphologicalchangesinduced bycalcination were stud-
Gases evolved during thermal decomposition of [Cu- [Cu-0.33]673 were smooth, while laminae of [Cu-0.33]873
.33] were monitored by MS (Fig. 3). TG/DTA data (not were covered with little holes (white dots in Fig. 5c) (34).
0
shown) as well as evolution of CO2 and H2O are character- In addition to this lamellar structure, [Cu-0.33]673 and es-
pecially [Cu-0.33]873 contained small, spherical particles
(
black spots in Figs. 5b and 5c).
Cu surface areas of samples calcined at different calci-
TABLE 5
Morphological Properties of [Cu-0.33] Calcined at Different
Temperatures Determined by N2 Physisorption and XRD
nation temperature were determined by N O titration and
corroborated by XPS (Table 4). Up to 773 K, Cu surface
2
area generally increased with increasing calcination tem-
perature; [Cu-0.33]873 exhibited a slightly lower Cu surface
area than [Cu-0.33]773
.
Crystallite sized
Calcination
temperature
K)
a
b
c
SBET
(m g
Vp,N2
(cm g
dp,max (hdpi)
HT
CuO
(nm)
2
� 1
3
� 1
(
)
)
(nm)
(nm)
DISCUSSION
Uncalcined
60
66
68
72
66
69
0.50
0.47
0.50
0.52
0.56
0.48
30 (33)
30 (29)
30 (29)
40 (29)
30 (34)
20 (28)
10
—
—
—
—
—
—
—
—
3
15
20
4
5
6
7
8
73
73
73
73
73
Characteristic Features of Phenylethyne Coupling Reaction
Reaction conditions in the heterogeneous coupling of
phenylethyne were chosen so as to eliminate the impact
of mass transfer on reaction rate. Hence, the experimental
results reflect the behavior of the catalytic system in the
kinetically controlled regime.
a
b
c
BET surface area.
BJH cumulative desorption pore volume.
Graphically assessed maximum of pore size distribution (in paren-
In analogy to the homogeneous catalytic route (35),
the heterogeneous coupling of phenylethyne is believed
theses: mean pore diameter, hdpi = 4Vp,N /SBET).
2
d
Determined from XRD line broadening.

8
AUER ET AL.
FIG. 5. Morphological changes occurring during calcination as seen by TEM. (a) [Cu-0.33]uc; (b) [Cu-0.33]673; (c) [Cu-0.33]873.
to proceed in three main steps, namely deprotonation of conversion was observed without addition of NaOH to the
phenylethyne, oxidation of phenylethynilide, and coupling system. This is possibly due to blocking of the basic sites of
to 1,4-diphenylbuta-1,3-diyne (17). As to the deprotona- the mixed oxides by CO2 or H2O (36). The knowledge of the
tion ofphenylethyne, the intrinsic basicity of the Cu–Mg–Al kinetics of the homogeneously catalyzed coupling (37) and
mixed oxides does not suffice to catalyze this step since no the linear dependence of the reaction rate on the amount

PHENYLETHYNE COUPLING OVER Cu–Mg–Al MIXED OXIDES
9
FIG. 5—Continued
of catalyst (38) strongly suggest that the deprotonation of reaction is reoxidized by oxygen closing the catalytic cycle.
phenylethyne is not the rate-limiting step of the reaction, Again, this is consistent with studies of the homogeneous
but occurs in a preceding equilibrium. The subsequent step, coupling route (35, 37). Coupling to 1,4-diphenylbuta-
the oxidation ofphenylethynilide isbelieved to be catalyzed 1,3-diyne occurs by combination of two phenylethynilide
by Cu(II) on the catalyst surface (17). Cu(I) formed in this radicals. Whether the first and third reaction step occur

1
0
AUER ET AL.
FIG. 5—Continued
homogeneously or on the catalyst surface could not be re- global reaction rate under the conditions investigated. It
solved in this study. seems most likely that the surface reaction involving Cu(II)
Our experimental findings indicate that neither mass is the rate-determining step, although this has not been di-
transfer nor deprotonation of phenylethyne determines the rectly proven.

PHENYLETHYNE COUPLING OVER Cu–Mg–Al MIXED OXIDES
11
Catalyst Properties
to the evolution of CO2 at about 870 K (Fig. 3) leading to
small holes in the lamellae and spherical particles of CuO
In the subsequent discussion, we assume that the surface
reaction involving Cu(II) is rate determining. The kinetic
studies indicate that the various Cu–Mg–Al mixed oxides
(
Fig. 5).
In conclusion we can state that the activityofallsamplesis
correlated to the portion of HT in the catalyst, as long as the
phase resulting upon calcination is not strongly altered, e.g.,
by temperatures exceeding 773 K. The HT-phase provides
a suitable matrix for active Cu(II). As expected, the effect
of this matrix on catalyst activity is more pronounced with
well-dispersed Cu(II) than with aggregated Cu(II), where
the active sites are partly detached from the matrix.
Studies of the redox behavior of Cu(II) species embed-
ded in the different oxidic matrices will be necessary to gain
further insight into the functioning of this complex catalytic
system.
show major differences in activity, whereas the selectivity to
1
,4-diphenylbuta-1,3-diyne is only weakly affected by com-
position (Table 3) and calcination temperature (Table 4).
Different activities can originate either from a change in
the number of accessible active Cu(II) sites and/or their
intrinsic activity.
As to the number of active sites, neither for catalysts with
different composition (Table 3) nor for catalyst [Cu-0.33]
calcined at different temperatures (Table 4) the ex situ mea-
sured Cu surface areas can be correlated with conversion
of phenylethyne. This is true for Cu surface areas measured
both before and after reaction. Although the number of ac-
tive Cu(II) sites under reaction conditions cannot be safely
derived from the measured Cu surface areas, this cannot
explain the drastically different catalytic behavior. The fact
that [Cu-0.0]673 is not active at all supports this suggestion.
Similar behavior has been reported for other reactions in-
volving a Cu species as active site, e.g., methanol synthesis
over Cu–ZnO–Al2O3 (39, 40).
CONCLUSIONS
Cu–Mg–Al mixed oxides containing 33 at.% Cu related
to the totalmetalamount prepared byprecipitation ofmetal
nitrates with sodium carbonate are efficient catalysts for the
oxidative coupling of phenylethyne to 1,4-diphenylbuta-
1
,3-diyne in the presence of NaOH as cocatalyst. Highest
yields of 1,4-diphenylbuta-1,3-diyne (84% ) were obtained
with catalysts containing a hydrotalcite-like phase.
Consequently, the catalysts must differ in the intrinsic ac-
tivity of the Cu(II) sites. Changes in the redox activity of
Cu(II) originate from alterations of the matrix in which
Cu(II) is embedded. Catalysts with compositions [Cu-
The heterogeneous coupling of phenylethyne is sug-
gested to proceed in three chemical reaction steps, namely
deprotonation of phenylethyne, oxidation of phenyl-
ethynilide involving Cu(II) on the catalyst surface, and cou-
pling to 1,4-diphenylbuta-1,3-diyne. The activity of the cru-
cial Cu(II) sites is strongly influenced by the structural and
chemical properties of the matrix in which they are em-
bedded. The hydrotalcite-like phase in Cu–Mg–Al catalysts
0
.33]673, [Cu-0.67]673, [Cu-0.29]673, and [Cu-0.25]673 appar-
ently provide a suitable matrix for active Cu(II), whereas
Cu-0.0]673 does not. For [Cu-0.33], calcined at different
[
temperatures, an activating matrix exists up to tempera-
tures of 773 K, whereas [Cu-0.33]873 exhibits a significant
drop in activity indicating major changes in the matrix.
A molecular view on these changes could not be gained in
this study. However, our data indicate that the catalysts con-
tained active and inactive phases in varying ratios. For cata-
lysts with different composition, the MgCu-phase in [Cu-
forms a favorable matrix for Cu(II) sites, whereas other
phases such as magnesium–copper hydroxide nitrate are
less suitable.
ACKNOWLEDGMENTS
0
.0]uc leads to a completely inactive matrix for the Cu(II)
catalyzed coupling of phenylethyne, whereas the HT-phase
in [Cu-0.33]uc provides an active matrix (Fig. 2, Table 3).
Thanks are due to M. Maciejewski for performing thermoanalytical
analyses and to R. Wessiken for taking TEM pictures. Financial support by
XRD patterns (Fig. 2) and thermal decomposition curves the Swiss National Fonds (Project No. 21-41850.94) is gratefully acknowl-
edged. S.M.A. thanks the Fonds der Chemischen Industrie, Germany, for
(
Fig. 3) of the uncalcined catalysts show that [Cu-0.25]uc
granting a stipendium.
and [Cu-0.29]uc contained the MgCu-phase and HT-phase
in varying ratios, leading to catalysts with medium activ-
ity compared to [Cu-0.0]uc and [Cu-0.33]uc. In fact, the HT
portion of the catalysts correlates with the activity of the
calcined samples in the coupling reaction. In [Cu-0.67]uc,
traces of the HT-phase, providing an active matrix, could
be detected by both XRD and thermal decomposition. As
concerns [Cu-0.33], calcined at different temperatures, the
HT-phase leads to an activating matrix up to temperatures
of 773 K. After calcination at 873 K, a major change in the
matrix could be detected (Figs. 4 and 5) causing the signif-
icant loss in activity for [Cu-0.33]873. This change is related
REFERENCES
. Cavani, F., Trifir o` , F., and Vaccari, A., Catal. Today 11, 173 (1991).
. Vaccari, A. (Ed.), Appl. Clay Sci. 10, 1 (1995).
. Reichle, W. T., Chemtech 58 (1986).
1
2
3
4. Reichle, W. T., Solid State Ionics 22, 135 (1986).
5
6
7
. Feitknecht, W., Helv. Chim. Acta 25, 555 (1942).
. Feitknecht, W., Helv. Chim. Acta 25, 131 (1942).
. Putyera, K., Bandosz, T. J., Jagiello, J. and Schwarz, J. A., Appl. Clay
Sci. 10, 177 (1995).
. Kooli, F., Kosuge, K., and Tsunashima, A., J. Solid State Chem. 118,
285 (1995).
8

1
2
AUER ET AL.
9
. Kannan, S., and Swamy, C. S., Appl. Catal. B 3, 109 (1994).
0. Morpurgo, S., Lo Jacono, M., and Porta, P., J. Mater. Chem. 4, 197
1994).
27. Evans, J. W., Wainwright, M. S., Bridgewater, A. J., and Young, D. J.,
Appl. Catal. 7, 75 (1983).
28. Chinchen G. C., Hay, C. M., Vanderwell, H. D., and Waugh, K. C.,
J. Catal. 103, 79 (1987).
29. Barrett, E. P., Joyner, L. G., and Halenda, P. P., J. Am. Chem. Soc. 73,
373 (1951).
1
1
(
1. De Roy, A. D., Forano, C., El Malki, K., and Besse, J. P., in “Synthesis
of Microporous Materials” (M. L. Occelli and H. E. Robson, Eds.),
p. 108. Van Nostrand Reinhold, New York, 1992.
1
1
2. Nakatsuka, T., Kawasaki, H., Yamashita, S., and Kohjiya, S., Bull. 30. Broekhoff, J. C. P., in “Preparation of Heterogeneous Catalysts II”
Chem. Soc. Jpn. 52, 2449 (1979).
3. Kohjiya, S., Sato, T., Nakayama, T., and Yamashita, S., Macromol.
Chem. Rapid Commun. 2, 231 (1981).
(P. G. B. Delmon, P. Jakobs, and G. Poncelet, Eds.), Vol. 3, p. 663.
Elsevier, Amsterdam, 1979.
31. Harkins, W. D., and Jura, G., J. Chem. Phys. 11, 431 (1943).
32. “Gmelin Handbook of Inorganic Chemistry, Copper Organic Com-
pounds, Part 3.” Springer Verlag, Berlin, 1986.
1
1
4. Reichle, W. T., J. Catal. 94, 547 (1985).
5. Busetto, C., Del Piero, D., Manara, G., Trifir o` , F., and Vaccari, A.,
J. Catal. 85, 260 (1984).
33. Sing, K. S. W., et al., Pure Appl. Chem. 57, 603 (1985).
1
1
1
6. Kaneda, K., Ueno, S., and Imanaka, T., J. Chem. Soc., Chem. Commun. 34. Reichle, W. T., Kang, S. Y., and Everhardt, D. S., J. Catal. 101, 352
97 (1994). (1986).
7. Auer, S. M., Schneider, M., and Baiker, A., J. Chem. Soc., Chem. Com- 35. Klebansky, A. L., Grachev, I. V., and Kuznetsova, O. M., Zh. Obshch.
7
mun. 2057 (1995).
Khim. 27, 2977 (1957).
8. Khoo, I.C., and Wu, S. T., in “Optics and Nonlinear Optics of Liquid
Crystals,” p. 42. World Scientific, Singapore, 1993.
9. Hay, A. S., J. Org. Chem. 27, 3320 (1962).
36. Tanabe, K., in “Catalysis by Acids and Bases” (B. Imelik, C. Naccache,
G. Coudurier, Y. Ben Taarit, and J. C. Vedrine, Eds.), Vol. 20, p. 1.
Elsevier, Amsterdam, 1985.
1
2
0. Sonogashira, K., in “Comprehensive Organic Synthesis” (B. M. Trost
and I. Fleming, Eds.), p. 551. Pergamon Press, Oxford, 1991.
37. Bohlmann, F., Sch o¨ nowsky, H., Inhoffen, E., and Grau, G., Chem. Ber.
97, 794 (1964).
2
1. “Mineral Powder Diffraction Data Files.” JCPDS-International Cen- 38. Auer, S. M., Wandeler, R., Schneider, M., and Baiker, A., in “Pro-
ter for Diffraction Data, Swarthmore, PA, 1995.
2. Scherrer, P., G o¨ tt. Nachr. 2, 98 (1918).
3. Klug, H. P., and Alexander, L. E., “X-Ray Diffraction Procedures for
Polycrystalline and Amorphous Materials.” Wiley, New York, 1974.
4. Halder, N. C., and Wagner, C. N. J., Acta Cryst. 20, 312 (1966).
ceedings, Catalysis on Solid Acids and Bases” (J. Weitkamp and
B. L u¨ cke, Eds.), 9601, p. 23. DGMK German Society for Petroleum
and Coal Science and Technology, Berlin, 1996.
2
2
39. Gusi, S., Trifir o` , F., Vaccari, A., and Del Piero, G., J. Catal. 94, 120
(1985).
2
2
5. Pause, H., and Mensing, J., “SpecsLab 1.6.” Specs GmbH, Berlin, 40. Doesburg, E. B. M., H o¨ ppener, R. H., de Koning, B., Xiaoding, X.,
1
996.
and Scholten, J. J. F., in “Preparation of Catalysts IV” (B. Delmon,
P. Grange, P. A. Jacobs, and G. Poncelet, Eds.), p. 767. Elsevier,
Amsterdam, 1987.
2
6. Briggs, D., and Seah, M. P., “Practical Surface Analysis by Auger and
X-Ray Photoelectron Spectroscopy.” Wiley, Chichester, 1983.