A comparison between conventional and ultrasound-mediated heterogeneous catalysis: Hydrogenation of 3-buten-1-ol aqueous solutions

Article abstract of DOI:10.1016/j.jcat.2003.08.008

A power flow scheme applicable to probe-type ultrasound reactors is presented, that has been deduced from both theoretical estimates and experimental measurements employing a thermal insulated vessel. Under typical conditions for water at 1 atm pressure, 77% of the electrical power is converted into mechanical motion of the probe, that in turn is dissipated to both acoustic power (～12%) and cavitational heating (～88%). Approximately 92% of the mechanical power of the probe was converted into heat, with the remaining power presumably converted into audible acoustic and/or mechanical motion. In a second type of experiment performed here, heterogeneous catalysis experiments have been performed at 298 K in an isothermal (i.e., jacketed) reaction vessel comparing chemistry in conventional (e.g., thermal) versus ultrasound-assisted systems. Both product state distribution and reaction rate measurements have been performed for the hydrogenation (using hydrogen gas) of aqueous 3-buten-1-ol solutions employing Pd-black powder. Products from the heterogeneous catalysis include isomerization to cis- and trans-2-buten-1-ol, as well as hydrogenation to 1-butanol. A reaction scheme involving surface-bound alkyl-radical species, consistent with previous published work, is proposed to explain product formation. Based on the observed differences in cis- to trans-2-buten-1-ol ratios in conventional versus ultrasound experiments, employing untreated and prereduced catalysts, it has been determined that ultrasound creates catalyst site(s) enhancing the cis-to-trans 2-buten-1-ol ratio from 0.25 to 0.55. In addition, comparing the total isomerization to hydrogenation ratio (cis- plus trans-2-buten-1-ol to 1-butanol ratio), for ultrasound-assisted and conventional catalysis, reveal a ～5-fold enhancement in isomerization relative to the more energetically favored hydrogenation due to the application of ultrasound. Finally, the product formation rates for 1-butanol, as well as isomerization plus hydrogenation, revealed that conventional and ultrasound experiments showed both a nonlinear dependence with applied ultrasound power and no differences between untreated and prereduced catalysts. The observed reaction rate enhancements were 1:36:183 for the conventional, 90 W ultrasound, and 190 W ultrasound experiments, respectively.

Full text of DOI:10.1016/j.jcat.2003.08.008

Journal of Catalysis 221 (2004) 347–353
www.elsevier.com/locate/jcat
A comparison between conventional and ultrasound-mediated
heterogeneous catalysis: hydrogenation of 3-buten-1-ol aqueous solutions
∗
R.S. Disselkamp, K.M. Judd, T.R. Hart, C.H.F. Peden, G.J. Posakony, and L.J. Bond
Pacific Northwest National Laboratory, Environmental Molecular Sciences Laboratory, Richland, WA 99352, USA
Received 25 June 2003; revised 4 August 2003; accepted 4 August 2003
Abstract
A power flow scheme applicable to probe-type ultrasound reactors is presented, that has been deduced from both theoretical estimates
and experimental measurements employing a thermal insulated vessel. Under typical conditions for water at 1 atm pressure, 77% of the
electrical power is converted into mechanical motion of the probe, that in turn is dissipated to both acoustic power (∼ 12%) and cavitational
heating (∼ 88%). Approximately 92% of the mechanical power of the probe was converted into heat, with the remaining power presumably
converted into audible acoustic and/or mechanical motion. In a second type of experiment performed here, heterogeneous catalysis exper-
iments have been performed at 298 K in an isothermal (i.e., jacketed) reaction vessel comparing chemistry in conventional (e.g., thermal)
versus ultrasound-assisted systems. Both product state distribution and reaction rate measurements have been performed for the hydrogena-
tion (using hydrogen gas) of aqueous 3-buten-1-ol solutions employing Pd-black powder. Products from the heterogeneous catalysis include
isomerization to cis- and trans-2-buten-1-ol, as well as hydrogenation to 1-butanol. A reaction scheme involving surface-bound alkyl-radical
species, consistent with previous published work, is proposed to explain product formation. Based on the observed differences in cis- to
trans-2-buten-1-ol ratios in conventional versus ultrasound experiments, employing untreated and prereduced catalysts, it has been deter-
mined that ultrasound creates catalyst site(s) enhancing the cis-to-trans 2-buten-1-ol ratio from 0.25 to 0.55. In addition, comparing the
total isomerization to hydrogenation ratio (cis- plus trans-2-buten-1-ol to 1-butanol ratio), for ultrasound-assisted and conventional catalysis,
reveal a ∼ 5-fold enhancement in isomerization relative to the more energetically favored hydrogenation due to the application of ultrasound.
Finally, the product formation rates for 1-butanol, as well as isomerization plus hydrogenation, revealed that conventional and ultrasound
experiments showed both a nonlinear dependence with applied ultrasound power and no differences between untreated and prereduced cata-
lysts. The observed reaction rate enhancements were 1:36:183 for the conventional, 90 W ultrasound, and 190 W ultrasound experiments,
respectively.

2003 Elsevier Inc. All rights reserved.
Keywords: Sonochemistry; Ultrasound; Heterogeneous catalysis; Product state distribution; Kinetics
1
. Introduction
the vicinity of a solid–liquid interface. These liquid jets are
thought to possess speeds of > 100 m/s that, in principle,
can yield heterogeneous chemistries different from conven-
tional (bulk thermal) techniques. Furthermore, SEM/TEM
imaging studies of powders dispersed in liquids before and
after sonication suggest that during cavitation high-speed
interparticle collisions occur, that can lead to catalyst activa-
tion and eventually sintering into larger structures [1]. These
combined phenomena, in part, serve as motivation for our
work here.
Sonochemically activated heterogeneous catalysis is in-
creasing in popularity [14], unfortunately much of the re-
search activity is not easily accessible in the West [15].
Catalysts employed in previous studies have been primarily
nickel (either activated powder or Raney type) [16,17] with
In recent years, the chemical effects of ultrasound have
been applied to heterogeneous catalysis [1–5], as well as
related phenomena of crystallization [6], nanoparticle for-
mation [7], catalyst production [8], and reagent (e.g., pollu-
tant) remediation [9,10]. Pioneering work by Lauterborn and
co-workers [11,12] and Blake and Gibson [13] has shown
that during cavitation near an extended interface, impinge-
ment of liquid “jets” into the surface can occur. This result
was obtained from high-speed microcinemagraphic observa-
tion that during cavitation bubble implosion often occurs in
*
Corresponding author.
E-mail address: robert.disselkamp@pnl.gov (R.S. Disselkamp).
0021-9517/$ – see front matter  2003 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2003.08.008

348
R.S. Disselkamp et al. / Journal of Catalysis 221 (2004) 347–353
some investigations using Pd/C [18]. Of particular interest in
the Pd/C investigations is the ability of both methanol [14]
and formic acid [19] to act as hydrogenation species to
alkene or ester compounds. Extending the scope of this
prior work, here we examine the hydrogenation of aque-
ous 3-buten-1-ol solutions employing Pd-black powder. In
particular, the branching ratio of isomerization relative to hy-
drogenation is examined.
probe type) reactor or the isothermal heterogeneous catalysis
studies at 298 ± 3 K, respectively.
A brief summary of the experimental procedure for ex-
periments employing untreated Pd catalyst material is as fol-
lows. First, the catalyst and aqueous reagents were weighed
and a (prerun) sampling vial was filled for subsequent analy-
sis. Second, the catalyst/aqueous solution was introduced
into the jacketed reaction vessel, manually stirred, and at-
tached to the ultrasonic probe assembly. Third, a NesLab
RTE-140 bath circulation unit used to maintain constant
temperature throughout an experiment. Temperature was
maintained at 298 ± 3 K by setting the bath circulation tem-
perature to account for the temperature difference between
the inside of the reactor and the bath temperature during son-
ication. For example, a linear relationship was observed for
this temperature differential as a function of applied electri-
cal power such that a 4.0 K differential existed when 100 W
of electrical power was supplied to the system. Fourth, ∼ 10
pressure/vent cycles of the closed reactor cell with hydrogen
gas was performed prior to filling with a static pressure of
In the present work two types of studies are discussed.
First, the power flow scheme in probe-type ultrasonic reac-
tors is mapped out relevant to our investigations. The pur-
pose of this measurement is to address issues pertaining to
the required power necessary to achieve a given product for-
mation rate. Second, hydrogenation experiments employing
a simple water-soluble alkene compound (3-buten-1-ol) are
performed so as to examine reaction pathways (i.e., product
state distribution) leading to isomerization and hydrogena-
tion. Furthermore, rates of reaction differences in conven-
tional versus ultrasound-assisted catalysis are determined.
Finally, one criteria often utilized by industry for evaluating
a new synthetic approach is based on the atom economy and
the E-factor [20]. The atom economy means the number of
atoms in the starting materials that are found in the product.
The E-factor is based on the amount of waste (in kg per kg)
of product. This factor can also be correlated with the toxic-
ity of waste. Since the atom economy of ultrasound-assisted
heterogeneous catalysis in prior studies has often not been
identified, in the work here an effort is made to report these
values.
6
.8 atm of hydrogen gas. Fifth, ultrasound was applied for
a specified time and electrical power supplied by the soni-
fier was noted. No mechanical stirring of the solution was
performed during sonication. Sixth, the hydrogen gas was
vented and the reactor assembly was disassembled. Seventh,
the solution was again mechanically stirred and a sample
was collected for subsequent analysis by filtering (0.45-µm
hydrophilic Millipore filter). Eighth, the reaction vessel was
reassembled and the process repeated.
The experimental procedure for experiments employing
prereduced Pd catalyst material was identical to that given
above except for the following. First Pd catalyst powder,
water, and hydrogen gas (6.8 atm) were introduced as de-
scribed above to the reaction vessel and sonicated for 5 min
at 200 W (electrical) power prior to each experiment. Next,
the 3BEN reagent was added to the prereduced catalyst solu-
tion and mechanically stirred, and the first sample collected
nominally 1 min after 3BEN addition. The catalyst prere-
duction experiments yielded product formation immediately
after 3BEN addition, thus the first sample collected at 1 min
had measurable products present.
2
. Experimental
2
.1. Materials
Palladium black of 99.9% purity (metals basis; Alfa Ae-
2
sar), having a nominal BET specific surface area of 20 m /g,
was used either without prior reduction or prereduced by
sonication in a procedure discussed below. Aqueous so-
lutions were prepared using deionized water (18 Mꢀ-cm)
using 3-buten-1-ol (98+% purity; Aldrich Chemical Com-
pany) as a reactant. All reagent solutions contained 1.0 wt%
2
.3. Reactant and product sample analyses
3
-buten-1-ol and 50 mg Pd-black powder. The hydrogena-
tions employed hydrogen gas from A&L specialty gas
99.99% purity) at a pressure of 6.8 atm.
Qualitative identification and quantitative analyses were
performed using an Hewlett-Packard GC/MS (5890 Se-
ries II GC and 5972 MSD) equipped with a DB-5 column
(
2
.2. Reaction apparatus and procedure
(
30 m, 0.2 mm i.d., 1.00-µm film). Manual sample injec-
tions (1.0 µL) in triplicate were performed as part of the
routine data reduction. A statistical analysis revealed that
relative concentrations were accurate to within ± 7.0% of
the stated percentage concentration, whereas the accuracy
among manual injections was ± 15%. Analyses of the liquid
samples both prior to sonication and after reactions allowed
an estimate of the mass conservation to be performed. Of
particular interest is whether the application of ultrasound
Commercially available sonication equipment was used
to perform the studies described here. A Branson Ultrasonics
Model 450 Sonifier II unit, capable of delivering up to 400 W
20 kHz), was used. In addition, both unjacketed (Branson
Model 101-021-003)and jacketed (Branson Model 101-021-
06) reaction cells were used for studies of either the power
flow scheme in our piezoelectric stack on resonant horn (i.e.,
(
0

R.S. Disselkamp et al. / Journal of Catalysis 221 (2004) 347–353
349
results in degradation of the reactant 3-buten-1-ol into (un-
desirable) products such as carbon dioxide, hydrogen, and
water, as has been observed for other aqueous organic solu-
tions [21,22]. Such analyses allow the atom economy to be
determined [20].
at 100% amplitude under atmospheric conditions (e.g., only
1.0 atm pressure). Therefore, pressurization with hydrogen
is seen to significantly enhance coupling between the probe
and the fluid.
Scheme 1 can be understood by considering the follow-
ing. From the discussion above, electrical power is sent to
a convertor that generates mechanical motion from electri-
cal energy, the degree of total power dissipation set by the
coupling between the solution and the titanium probe, as
well as the percentage full-scale amplitude setting of the
sonifier unit. By operating the probe in both water and air,
where the latter yields only dielectric losses in the convertor,
the percentage power transfer to the 1.27-cm-diameter probe
can be determined. Hence, the 71 and 83% conversions of
electrical power into mechanical power should be applied to
the 112 and 42 W electrical powers, respectively. Next, me-
chanical motion of the probe can generate the following ef-
fects: (1) acoustic power; (2) nonlinear cavitational heating;
and (3) noncavitating mechanical mixing such as acoustic
streaming and radiation pressure [24]. The former has been
estimated from the power transmission coefficient from tita-
nium to water using appropriate acoustic impedances [25].
Finally, a calorimetric measurement of the temperature rise
of water (employing the unjacketed insulated reactor) was
used to compute the pathway percentage into heat with the
remainder assigned to audible acoustic/mechanical power
dissipated to the surroundings. As the complete conversion
of acoustic power into heat cannot account for the observed
3
. Results
3
.1. Power dissipation scheme
An important aspect of understanding ultrasound-assisted
heterogeneous catalysis is to map out the ultrasound power
flow scheme. The nature of this investigation is to understand
power flow in a probe-type ultrasonic reactor. The pertinent
question that arises is: For a given electrical power output
(
W) from the power supply of the sonifier unit, what power
flow pathways exist and what are their relative importance
leading to the eventual conversion into heat? Our analysis of
the power dissipation scheme closely follows the approach
taken by Loning et al. [23].
The ultrasonic unit is composed of three primary parts: 1,
power supply; 2, convertor consisting of piezo-ceramic ele-
ments; and 3, probe assembly. Put simply, electrical power
from the power supply is sent to the convertor where a mi-
nor fraction is converted into heat and a major fraction is
converted into mechanical motion of the probe. The power
flow scheme as outlined above for deionized, air-saturated,
water at 1 atm pressure, at both 100 and 50% amplitude set-
tings, is illustrated in Scheme 1. For a particular amplitude,
the electrical power delivered by the power supply depends
on the nature of the coupling between the titanium probe
and the fluid being sonicated. It is interesting to note that
for the hydrogenation experiments employing 6.8 atm hy-
drogen gas, 190 W electrical power at 60% amplitude was
delivered. This value can be compared to 112 W of power
(
calorimetric) heating of water, a combination of cavitational
heating and/or fluid mixing effects must also be important.
The ultrasound-assisted and conventional heterogeneous
catalysis experiments were all performed at 298 ± 3 K us-
ing a jacketed (water-cooled) reactor. It is unclear to what
degree both purely acoustic power and cavitational heating
power gives rise to enhancements in heterogeneous reaction
rates and selectivity (vide infra). An attempt was made here
utilizing the isothermal reactor to exclude bulk-heating ef-
fects as the principle cause of differences in these types of
chemistry. In principle, transient effects such as cavitational
heating include effects such as bubble formation and col-
lapse near an interface, perhaps causing liquid jets incident
upon the catalyst surface and inducing chemistry [11–14].
3
.2. Hydrogenation and isomerization
A primary objective of this study is to perform a di-
rect comparison between conventional versus ultrasound-
mediated heterogeneous catalysis. We have chosen 3-buten-
1
-ol (3BEN) for this comparison for the following reasons:
(
(
1) It is a relatively simple, water-soluble, alkene compound.
2) Hydrogenation employing a standard catalyst and hy-
drogen gas can be investigated. (3) The relative conversion
to 1-butanol (hydrogenation) versus cis- and trans-2-buten-
Scheme 1. A schematic illustration of the power dissipation scheme for our
probe-type ultrasonic reactor is presented. Percentages have been arrived at
using both experimental and theoretical estimates (see text) using deionized
1
-ol (C2BEN and T2BEN) (isomerization) can be studied.
(4) A comparison between product formation rates yielding
enhancements arising from the application of ultrasound can
(air saturated) water as the fluid at 1 atm pressure.

350
R.S. Disselkamp et al. / Journal of Catalysis 221 (2004) 347–353
Table 1
Product distribution as a function of reaction time for ultrasound and con-
ventional experiments
Time (min)
3BEN
1BUT
T2BEN
C2BEN
Ultrasound UT1: (190 W)
0
0
1
2
3
4
5
100
75.43
66.31
0
0
8.85
12.71
15.94
29.38
44.34
10.07
13.47
17.15
28.98
40.72
5.65
7.51
9.06
12.51
10.13
57.85
29.13
4.81
Ultrasound UT2: (92 W)
0
4
8
2
6
0
100
89.7
67.85
53.38
37.91
20.25
0
3.88
10.39
14.77
22.01
28.79
0
4.1
13.5
20.63
27.47
36.77
0
2.32
8.26
11.22
12.61
14.19
1
1
2
Fig. 1. A plot of the cis- to trans-2-buten-1-ol ratio (C2BEN/T2BEN) is
given for the 190 W ultrasound, 92 W ultrasound, and conventional ex-
periments, as a function of percentage 1-butanol product formed. Also
presented is the total isomerization to hydrogenation ratio, (T2BEN +
C2BEN)/1BUT. No prior reduction of the catalyst was performed.
Conventional UT3 (no ultrasound)
0
0
100
95.36
0
0
0
0
0.41
0.69
1.23
1.73
5
3.66
7.67
10.98
15.37
18.81
0.98
2.09
3.1
1
1
2
2
00
50
00
50
89.83
85.23
79.03
73.6
4.37
5.86
Conventional PR1 (prereduced; 90 W ultrasound)
a
1
4
8
2
6
86.11
74.97
65.13
33.07
4.33
5.36
9.04
13.22
24.99
41.85
5.94
10.64
14.51
29.62
44.5
2.58
5.35
7.14
12.32
9.32
1
1
Conventional PR2 (prereduced; no ultrasound)
a
1
0
00
50
00
50
87.1
5
8.2
10.52
13.01
15.55
17.6
5.38
7.48
8.63
9.74
11.27
11.86
2.52
3.56
3.74
4.11
4.47
4.89
5
80.76
77.11
73.14
68.71
65.65
1
1
2
2
a
After Pd prereduction 3BEN was added to solution, the solution was
mechanically mixed, and the first sample was collected. Approximate mix-
ing time for this first sample was 1 min.
Fig. 2. A plot of the cis- to trans-2-buten-1-ol ratio (C2BEN/T2BEN) is
given for the 92 W ultrasound and conventional experiments, as a function
of percentage 1-butanol product formed. Also presented is the total isomer-
ization to hydrogenation ratio, (T2BEN+C2BEN)/1BUT, similar to Fig. 2.
Prereduction of the catalyst was performed using hydrogen gas and the ap-
plication of ultrasound (see text).
be examined. The third item above could have commercial
significance as ultrasound may provide an effective means
of converting a terminal alkene to internal cis- and trans-
alkenes.
Experimental data for untreated catalyst experiments
(
UT1, UT2, UT3 runs) and two prereduced catalyst experi-
Fig. 1 presents a reduced form of the data presented in Ta-
ble 1 for the untreated catalyst experiments (UT1, UT2, and
UT3). In particular, the C2BEN/T2BEN isomer ratio is il-
lustrated, as is the total isomerization to hydrogenation ratio,
given as (C2BEN + T2BEN)/1BUT, as a function of per-
centage 1BUT product formed. In a similar fashion, Fig. 2
presents the same product ratios except that the prereduced
(PR1 and PR2) experimental data have been used. It is im-
portant to note that additional experiments revealed that no
product formation was observed in the absence of catalyst
ments (PR1, PR2 runs) are presented in Table 1. Listed are
the mass percentages for the 3BEN, 1BUT, T2BEN, and
C2BEN species as a function of time. The average mass
conservation for the three ultrasound experiments was com-
puted to be 104 ± 10%, indicating that cavitational effects,
should they be present to an appreciable extent, do not ef-
fectively degrade the reagent 3BEN into products other that
those desired. Thus a very favorable atom economy exists
(
∼ 100%) for these experiments.

R.S. Disselkamp et al. / Journal of Catalysis 221 (2004) 347–353
351
Fig. 3. Product formation rates are presented for 1BUT and 1BUT +
T2BEN + C2BEN (hydrogenation plus isomerization), for both untreated
Pd catalysts and prereduced Pd catalysts. Note the nonlinear dependence in
product formation rate with applied ultrasonic power, and the nearly identi-
cal product formation rates between untreated and prereduced catalysts.
Scheme 2. A reaction scheme illustrating a proposed mechanism for isomer-
ization and hydrogenation is given. Rate coefficients for the various forward
reaction pathways are given as A, B, C, and D. Important reaction inter-
mediates proposed to exist, from which products are formed, are labeled Ia
and Ib.
and/or hydrogen gas. Thus 3BEN plus catalyst and hydro-
gen are needed for chemistry to proceed.
3
.3. Rate of reaction
hydrogenation (using Pd) of the aqueous 3-buten-1-ol and
gaseous 1-butene systems, where the latter has been ex-
tensively studied previously. Second, the type of catalyst
active site(s) responsible for chemistry in the ultrasound-
irradiated and no ultrasound (blank) experiments will be
approximately accounted for, despite there being insufficient
knowledge for an absolute determination of their properties.
A reaction scheme that accounts for product formation
in our experiments, and more importantly is consistent with
previously published studies, is presented in Scheme 2. This
scheme can be interpreted as follows. Hydrogen atom addi-
tion, with rate coefficient A, generates both Ia (C4 H-atom
addition) and Ib (C3 H-atom addition). The addition of a
second surface-adsorbed H atom to Ia or Ib, with rate coef-
ficient B, generates the hydrogenated product 1BUT. Owing
to the fact that the isopropyl radical is more stable than the
ethyl radical by 2.5 kJ/mol [26], and the bonding of each
to the Pd surface is assumed identical, the Boltzman distri-
bution of states expression suggests that the Ia/Ib ratio at
298 K is 2.7. This two-step alkene hydrogenation scheme
where the intermediate is a relatively stable surface bound
alkyl-radical species was originally proposed by Horuiti and
Polayni nearly 70 years ago [27]. Prior literature support
for this mechanism (process A and B) comes from identical
products being obtained in both equilibrated and nonequili-
brated hydrogen/deuterium mixtures [28–30]. Furthermore,
recent experimental work that determined hydrogen gas re-
action orders (∼ 0.5) suggests that either process A or B
can be rate limiting [31,32]. Also taken from prior published
Another important parameter to characterize is the rates
of product formation in conventionaland ultrasound-assisted
systems, for both untreated and prereduced Pd catalyst ex-
periments. Fig. 3 presents the percentage mass of product
formed per minute, for 1BUT, as well as 1BUT +T2BEN+
C2BEN (total product), as a function of applied ultrasound
electrical powers. Here zero ultrasound power corresponds
to the conventional catalysis results. Several points are worth
noting about Fig. 3. First, the 1BUT+T2BEN+C2BEN
and 1BUT formation rates are strongly correlated with one
another. The relation between these two rates are such
that for the ultrasound experiments only the 1BUT pro-
duction rate is lower by a factor of 2.5 compared to the
1
BUT+T2BEN+C2BEN product formation rate. Second,
a nonlinear relationship exists between product formation
rates and applied ultrasound power. In particular, the 1BUT
product formation rates for the conventional, 90/92 W ul-
trasound, and 190 W ultrasound experiments were observed
to have a ratio of 1:36:183, respectively. Third, and perhaps
most importantly, within experimental error, the product for-
mation rates are equal for the untreated and prereduced Pd
catalyst experiments.
4
. Discussion
In this section two main issues are discussed. First, a reac-
tion scheme is presented that draws on similarities between

352
R.S. Disselkamp et al. / Journal of Catalysis 221 (2004) 347–353
mechanisms [30] is the C2 hydrogen atom elimination re-
actions yielding T2BEN and C2BEN, each proceeding with
rate coefficients C and D, respectively. It is clear that only
species Ia can yield T2BEN and C2BEN, as H elimination
from C3 of species Ib would yield the reagent 3BEN. This
mechanism is consistent with our data as in principle ef-
ficient rotation about the C2–C3 σ-bond prior to H-atom
elimination can occur for the surface-bound alkyl radical
species yielding the isomerized product.
An examination of Figs. 1 and 2 can yield insight into the
type of catalyst active site(s) responsible for the observed
chemistry. For example, Fig. 1 (non prereduced catalyst)
shows that the C2BEN/T2BEN ratios for the ultrasound and
no ultrasound (blank) experiments are 0.55 and 0.25, re-
spectively. However, for the prereduced catalyst experiments
of Fig. 2 the ultrasound and no ultrasound ratios are both
other at 185 W (electrical) ultrasound powers. These exper-
iments showed an identical cis/trans ratio with percentage
1BUT formation and comparable rates of reaction compared
to the aqueous experiments. Thus OH group adsorption to
the surface (from methanol, 3BEN, T2BEN, or C2BEN)
does not appear to control product state distribution. Fur-
thermore, as our mass conservation was ∼ 100% and little
butanal was detected (less than 2%), 3BEN apparently did
not act as a hydrogenation species. This result differs some-
what from prior published work [14].
5. Conclusions
For the 3BEN molecule studied here, the conversion of
a terminal alkene into an internal one forming both cis and
trans isomers with an enrichment ∼ 5-fold compared to con-
ventional catalysis was noted. In addition, a comparison be-
tween untreated and prereduced catalysts for conventional
experiments has shown that intrinsically different catalyst
site(s) exist, as reflected in the product state distributions.
The ability of ultrasound to enhance the production of less
energetically favorable products has the potential of being
useful in commercial synthesis. Mechanistically, based on
prior work, the C3 surface-adsorbed alkyl radical interme-
diate (Ia) bound to the Pd surface can undergo C3 H-atom
addition to 1-butanol, or C2 H-atom elimination to cis- and
trans-2-buten-1-ol. One goal of future work proposed here
will be to find experimental conditions (catalyst type, ul-
trasound power, hydrogen gas pressure, etc.) that maximize
differences between ultrasound and conventional catalytic
processing methods. Consistent with this goal is to moni-
tor energy expenditure resulting in product formation. The
power flow scheme presented here is the first task undertaken
to accomplish this goal. This information is likely crucial in
assessing energy savings for a process. For example, if en-
richment in, say, cis-to-trans isomers for a synthetic process
can be obtained with ultrasound, a cost analysis must con-
sider ultrasound-expended energy compared to conventional
(thermally) generated products plus the cost associated with
sample purification (e.g., distillation).
∼
0.50. It can be proposed therefore that ultrasound either
applied as part of the catalyst prereduction process, or dur-
ing the course of an experiment, acts to create catalyst active
site(s) that increase the C2BEN/T2BEN ratio, and that even
after ultrasound is not applied to the system a persistent high
C2BEN/T2BEN ratio exists.
A somewhat more interesting situation arises for the
isomerization relative to hydrogenation ratio, (C2BEN+
T2BEN)/1BUT (which we will denote I/H). For this ratio,
Fig. 2 shows that the ultrasound and blank values of ∼ 1.7
and 0.3, respectively, differ by a large amount. It can be
proposed that ultrasound prepares actives site(s) greatly fa-
voring a large I/H ratio. Furthermore, from the pre-reduced
catalyst results of Fig. 2 it is seen that the blank experiment
has the I/H ratio monotonically decreasing with increasing
percentage 1BUT formation. This suggests that the catalyst
site(s) responsible for the I/H ratio are less stable during the
course of an experiment and hence change over time, unlike
the site(s) controlling the C2BEN/T2BEN ratio. Examining
the proposed mechanism of Scheme 2, since Ia is a precur-
sor to product formation, it can be concluded that the ratio of
rate coefficients (C + D)/B is enhanced by the application
of ultrasound.
As a point to note, despite there being differences in prod-
uct state distribution according to Figs. 1 and 2, the overall
rate of product formation for the no ultrasound and 92 W
ultrasound experiment of Fig. 3 shows that there is little
difference in the rate of product formation between the pre-
reduced and reduced experiments. Also, since the rate of
product formation is nonlinear in applied ultrasound power,
it is most cost effective (% product/W is greater) to operate
at the higher ultrasound powers.
The experimental data presented here have shown that
differences in product state distribution and reaction rate
exist between conventional and ultrasound-assisted hetero-
geneous catalytic systems. Much work is needed to map out
these differences as a function of temperature, reagent con-
centration (hydrogen gas pressure and liquid phase concen-
trations), reaction time (extent of reaction), catalyst type, and
ultrasound/static reactor operating conditions (i.e., power).
The role that the OH group in 3BEN plays in the isomer-
ization and hydrogenation reactions cannot be deduced from
our data employing water as a solvent. As no isomerization
was observed without the presence of hydrogen gas, the OH
group likely does not bond to the surface and hence play a
role in isomerization. To test this hypothesis, we employed
methanol as a solvent and performed two experiments for the
non-prereduced 3BEN hydrogenation—one at 40 W and the
Acknowledgments
We note many fruitful discussions of this work with
James White and John Holladay of Pacific Northwest Na-
tional Laboratory (PNNL). The encouragement, and yet cri-

R.S. Disselkamp et al. / Journal of Catalysis 221 (2004) 347–353
353
tique, of this work by them is particularly noted. In addition,
[12] W. Lauterborn, H. Bolle, J. Fluid Mech. 72 (1975) 391.
[13] J.R. Blake, D.C. Gibson, Annu. Rev. Fluid Mech. 19 (1987) 99.
the advice of Dr. Ajit Mal of UCLA pertaining to the ultra-
sonic power dissipation scheme is greatly appreciated. Fi-
nancial support was obtained through a Laboratory Directed
Research and Development (LDRD) grant administrated by
PNNL. PNNL is operated by Battelle for the US Department
of Energy.
[
14] R. van Eldik, C.D. Hubbard, Chemistry Under Extreme or Non-Clas-
sical Conditions, Ultrasound as a New Tool for Synthetic Chemists,
Wiley, New York, 1997, Chap. 8.
[15] A.N. Maltsev, Russ. J. Phys. Chem. 50 (1976) 995.
[16] G. Saracco, F. Arzano, Chimica Industria (Milano) 50 (1968) 314.
[
[
17] A. Alexakis, N. Lensen, P. Mangeney, Synlett (1991) 625.
18] T.J. Mason, J.P. Lorimer, L. Paniwnyk, P.W. Wright, A.R. Harris,
J. Catal. 147 (1994) 1.
[
[
[
19] P. Boudjouk, B.H. Han, J. Catal. 79 (1983) 489.
20] W. Bonrath, Ultrasonics Sonochem. 10 (2003) 55.
21] A. Tauber, G. Mark, H.P. Schuchmann, C. von Sonntag, J. Chem. Soc.,
Perkin Trans. 2 (1999) 1129.
References
[
[
[
1] K.S. Suslick, D.J. Casadonte, J. Am. Chem. Soc. 109 (1987) 3459.
2] K.S. Suslick, Science 247 (1990) 1439.
3] B. Torok, K. Balazsik, M. Torok, K. Felfoldi, M. Bartok, Catal. Lett. 81
[
[
[
22] J. Lifka, J. Hoffmann, B. Ondruschka, Water Sci. Technol. 44 (2001)
1
39.
23] J.-M. Loning, C. Horst, U. Hoffmann, Ultrasonics Sonochem. 9 (2002)
69.
24] A. Mal, University of California Los Angeles, personnel communica-
tion, March 2003.
(
2002) 55.
4] G. Szollosi, I. Kun, B. Torok, M. Bartok, Ultrasonics Sonochem. 7
2000) 173.
5] C.L. Bianchi, F. Martini, V. Ragaini, Ultrasonics Sonochem. 8 (2001)
31.
[
[
1
(
1
[
6] S.L. Hem, Ultrasonics 5 (1967) 202.
[25] L.E. Kinsler, A.R. Frey, A.B. Coppens, J.V. Sanders, Fundamentals of
Acoustics, 4th ed., Wiley, New York, 2000, and references therein.
[26] M. Jones Jr., Organic Chemistry, Norton, New York, 1997.
[27] J. Horuiti, M. Polayni, Trans. Faraday Soc. 30 (1934) 1164.
[28] J.E. Douglas, B.S. Rabinovitch, J. Am. Chem. Soc. 74 (1952) 2486.
[29] G.C. Bond, J. Turkevich, Trans. Faraday Soc. 49 (1953) 281.
[30] G. Etrl, H. Knozinger, J. Weitkamp (Eds.), Handbook of Heteroge-
neous Catalysis, Vol. 5, Wiley–VCH, Weinheim, 1997.
[
7] K.S. Suslick, in: G. Ertl, H. Knozinger, J. Weitkamp (Eds.), Handbook
of Heterogeneous Catalysis, Vol. 3, Wiley–VCH, Weinheim, 1997,
Chap. 8.6.
[
8] K.S. Suslick, in: Kirk–Othmer Encyclopedia of Chemical Technology,
Vol. 26, 4th ed., Wiley, New York, 1998, pp. 517–541.
9] Y. Jiang, C. Petrier, T.D. Waite, Ultrasonics Sonochem. 9 (2002) 317.
[
[
10] M.H. Entezari, C. Petrier, P. Devidal, Ultrasonics Sonochem. 10
2003) 103.
11] W. Lauterborn, A. Vogel, Annu. Rev. Fluid Mech. 16 (1984) 223.
(
[31] E.E. Gonzo, M. Boudart, J. Catal. 52 (1978) 462.
[32] D. Reinig, PhD thesis, TH Darmstadt, Germany, 1992.
[