Improving the selectivity to C4 products in the aldol condensation of acetaldehyde in ethanol over faujasite zeolites

Article abstract of DOI:10.1016/j.apcata.2014.11.018

The selective conversion of acetaldehyde to C4 products, minimizing the production of secondary (C6, C8) condensation products, could be a potential path in the production of butadiene from ethanol, a process of commercial interest. Therefore, we have investigated the selective aldol condensation of acetaldehyde in liquid phase over faujasite zeolites, NaX and NaY. Specifically, we have examined the influence of the number and location of the exchangeable cations, type of cations, and post-synthesis treatments on product selectivity. At 230 °C, NaY results in higher C4/(C6 + C8) product ratio than NaX, which can be explained in terms of the strength, density, and accessibility of basic sites, which are less favorable in NaX than NaY. In fact, the CO₂ TPD measurements indicate the presence of three types of basic sites of varying strength, of which those with weak and medium strength are most important for the selective condensation. A confinement effect is observed when adding K to the NaY zeolite. The observed selectivity changes suggest that when larger cations partially occupy the supercages, the production of C8 products decreases, while C6 products increase. Also, post-synthesis washing treatments show significant variations in selectivity, which demonstrate the effects of partial occupation of the zeolite pores in the reaction. It is also shown that at a given conversion, the C4/(C6 + C8) ratio can be adjusted by modifying the micro/mesoporosity balance in the zeolite.

Full text of DOI:10.1016/j.apcata.2014.11.018

Accepted Manuscript
Title: Improving the Selectivity to C4 Products in the Aldol
Condensation of Acetaldehyde in Ethanol over Faujasite
Zeolites
Author: Lu Zhang Tu N. Pham Jimmy Faria Daniel E.
Resasco
PII:
S0926-860X(14)00724-8
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2014.11.018
APCATA 15111
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
10-9-2014
12-11-2014
16-11-2014
Please cite this article as: L. Zhang, T.N. Pham, J. Faria, D.E. Resasco, Improving
the Selectivity to C4 Products in the Aldol Condensation of Acetaldehyde
in Ethanol over Faujasite Zeolites, Applied Catalysis A, General (2014),
http://dx.doi.org/10.1016/j.apcata.2014.11.018
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Improving the Selectivity to C4 Products in the Aldol
Condensation of Acetaldehyde in Ethanol over Faujasite Zeolites
1
1
2
1,*
Lu Zhang, Tu N. Pham, Jimmy Faria, and Daniel E. Resasco
1
School of Chemical, Biological, and Materials Engineering, University of Oklahoma,
Norman, OK 73019, USA
2
Abengoa Research, C/ Energía Solar nº 1, Palmas Atlas, Seville 41014, Spain
_
__________________________
*
Corresponding author. Email address: resasco@ou.edu
1
Page 1 of 39

ABSTRACT
The selective conversion of acetaldehyde to C4 products, minimizing the production of
secondary (C6, C8) condensation products could be a potential path in the production of butadiene from
ethanol, a process of commercial interest. Therefore, we have investigated the selective aldol
condensation of acetaldehyde in liquid phase over faujasite zeolites, NaX and NaY. Specifically, we
have examined the influence of the number and location of the exchangeable cations, type of cations,
and post-synthesis treatments on product selectivity. At 230°C, NaY results in higher C4/(C6+C8)
product ratio than NaX, which can be explained in terms of the strength, density, and accessibility of
basic sites, which are less favorable in NaX than NaY. In fact, the CO TPD measurements indicate the
2
presence of three types of basic sites of varying strength, of which those with weak and medium
strength are most important for the selective condensation. A confinement effect is observed when
adding K to the NaY zeolite. The observed selectivity changes suggest that when larger cations partially
occupy the supercages, the production of C8 products decrease while C6 products increase. Also, post-
synthesis washing treatments show significant variations in selectivity, which demonstrate the effects of
partial occupation of the zeolite pores in the reaction. It is also shown that at a given conversion, the
C4/(C6+C8) ratio can be adjusted by modifying the micro/meso porosity balance in the zeolite.
KEYWORDS
Aldol condensation, acetaldehyde, crotonaldehyde, ethanol, butadiene, faujasite, biomass
2
Page 2 of 39

1.
INTRODUCTION
,3-butadiene (BD) is a building block in the chemical industry, used as a monomer or co-
1
monomer in the production of polybutadiene, styrene−butadiene rubber and latex,
acrylonitrile−butadiene−styrene resin, nitrile rubber, and styrene−butadiene block copolymers [1-
3]._ENREF_1 It is also used in the synthesis of monomer precursors, such as adiponitrile and
chloropropene. The global demand for BD is on an annual scale of around 9 million tons [1]. Among the
different approaches for the preparation of BD, steam cracking of paraffinic hydrocarbons accounts for
over 91% of the world’s butadiene supply. However, in this process, BD is a by-product of the
manufacturing of ethylene [2]. Direct production of BD can be accomplished by catalytic
dehydrogenation of n-butane and n-butene, as well as oxidative dehydrogenation of n-butene [1,3,4].
With the increasing interest in utilizing renewable natural resources instead of fossil resources [5], it
becomes important to investigate BD production routes from biomass. The abundant supply of ethanol
from a variety of possible renewable sources makes it an attractive feedstock for the production of
commodity chemicals, such as BD [6-13].
A possible strategy for the conversion of ethanol to BD includes the following steps: (i)
dehydrogenation of ethanol to acetaldehyde; (ii) aldol condensation of acetaldehyde to crotonadehyde;
(
iii) Meerwein−Pondorf−Verley (MPV) hydrogen transfer between ethanol and crotonaldeyde to
produce crotyl alcohol; and (iv) dehydration of crotyl alcohol to BD [14-18]. Aldol condensation is a
well-known reaction for C-C bond formation that can be catalyzed by either acid or basic catalysts [19-
2
7]. Specifically, the aldol condensation of acetaldehyde has been investigated in several previous
studies as an important intermediate reaction in the ethanol to BD process. In these studies [28-32]
metal oxides, hydrotalcite, and amino acid have been used as catalysts, which exhibit varying extents of
selectivity to C4 products. In this contribution, we have investigated the use of faujasite zeolites for the
selective aldol condensation of acetaldehyde to C4 products, trying to minimize the excessive
3
Page 3 of 39

condensation to >C6 products._ENREF_28
Faujasite zeolites are commonly used in base-catalyzed reactions since the strength of their basic
sites (framework O ions) can be adjusted by varying the Si/Al ratio and the electronegativity of the
exchangeable alkali metal cations used for compensating the negative framework charges [33-36]. At
the same time, zeolites offer the possibility of controlling reaction selectivity by adjusting the extent of
molecular confinement [37-39]. In consecutive reactions like the acetaldehyde self-condensation of
interest in this work, the formation of the larger products (>C6) may be inhibited by confinement. This
characteristic distinguishes zeolites from other types of basic catalysts and makes them a better
candidate for obtaining high selectivities to C4 products since further condensation to heavier C6 and
C8 hydrocarbons may be inhibited. While the aldol condensation of acetaldehyde in the vapor phase has
been previously investigated on zeolite X [26], no reports have been found on the aldol condensation of
acetaldehyde in liquid phase over faujasites, which is the focus of the present contribution. Moreover, in
this work, we have investigated the aldol condensation in the presence of an alcohol, which can undergo
hydrogen transfer via Meerwein−Pondorf−Verley (MPV) reaction and significantly affect the product
distribution.
Specifically, in this study, we have examined NaX and NaY zeolites. While both zeolites have
the FAU framework [40], the Si/Al atomic ratio for zeolite X is in the 1.0-1.5 range while for zeolite Y
is above 1.5 [41]. The effect of partially exchanging Na by K has been studied with zeolites KNaX and
KNaY, with different post-synthesis washing procedures.
2.
EXPERIMENTAL SECTION
2
.1
Zeolite synthesis and treatment.
The reagents used in the synthesis of all samples were purchased from Sigma-Aldrich and used
as received. They included sodium silicate (ca. 10.6% Na O and 26.5% SiO ), anhydrous sodium
2
2
aluminate (NaAlO
2
), sodium hydroxide (NaOH, ≥98%), potassium hydroxide (KOH, ≥85%), aluminate
4
Page 4 of 39

sulfate octadecahydrate, and potassium nitrate (≥99.0%).
The synthesis of zeolites NaX and NaY followed conventional methods reported in the literature.
2 2 3 2 2
For zeolite NaX [42], a seed solution with a gel composition of 19 Na O : 1.0 Al O : 19 SiO : 370 H O
was prepared by dissolving 2.1 g of NaOH in 12.36 g of distilled water; then, 0.51 g of NaAlO and
2
1
3.44 g of sodium silicate were added to the mixture while stirring for 30 min in order to obtain a
homogeneous clear solution. After standing at room temperature for 24 h it was ready for use. To
synthesize the zeolite, 2.05 g of NaAlO was added to 20.98 g of distilled water and stirred for 10 min.
2
Then, 1.68 g of NaOH, 8.49 g of sodium silicate, and 3.3 mL of the seed solution were added in
sequence. The final solution was stirred at room temperature for 30 min before it was placed into Teflon
bottles, where crystallization was allowed to occur at 100°C for 4 h. The zeolite sample was obtained by
separating the solid phase from the liquid phase and washed with distilled water until the pH of the
mother liquid reached neutral. The resulting sample was calcined at 400°C for 2 h and stored.
Zeolite NaY was synthesized by following the procedure described in US Patent No. 3,639,099.
In this case, the seed solution was prepared to get a gel composition of 15.6 Na
12 H O. For this, 2.1 g of NaOH were dissolved in 10.43 g of distilled water, and subsequently 0.51 g
of NaAlO and 11.32 g of sodium silicate were added. The mixture was stirred for 10 min until a
2 2 3 2
O : 1.0 Al O : 16 SiO :
3
2
2
homogeneous clear solution was obtained. It was allowed to stand at room temperature for 48 h before
use in the next step. To synthesize the NaY zeolite, 1.67 g of aluminum sulfate, 28.30 g of sodium
silicate, and 11.99 g of distilled water were mixed and stirred for 10 min. Then, 5 g of the as-prepared
2
seed solution was added together with 1.64 g of NaAlO . The final solution was stirred at room
temperature for 1 h. Once a homogeneous mixture was obtained, the solution was subdivided in smaller
aliquots and placed into several Teflon bottles. Crystallization was carried out at 100°C for 10 h. The
zeolite sample was obtained by separating the solid phase from the liquid phase and washed with
distilled water until the pH of the mother liquid was neutral. The washed solid was calcined at 400°C for
2
h and stored.
5
Page 5 of 39

The zeolite KNaX was synthesized by following the method reported by Kühl [43]. Specifically,
the initial gel composition was 5.5 Na O : 1.65 K O : 1.0 Al : 2.2 SiO : 122 H O. A typical
2
2
2
O
3
2
2
procedure was as follows: 7.4 g of KOH, 11.67 g of NaOH, and 35.3 g of distilled water were mixed
and subjected to vigorous stirring for around 5 min. At the same time, 6.56 g of NaAlO and 20 g of
2
distilled water were mixed and added to the former solution. Then, a mixture containing 19.9 g of
sodium silicate and 20 g of water was added. The resultant solution was stirred at room temperature for
1
0 min and then transferred to Teflon bottles, which were capped and sealed with paraffin films.
Crystallization was carried out first at 70°C for 3 h and then at 100°C for 2 h. Upon completion of the
crystallization, the product was filtered and washed with distilled water. The as-synthesized sample was
calcined at 400°C for 2 h before the catalytic test.
The zeolite KNaY was obtained by ion-exchange on a pre-synthesized NaY, using 50 mL of the
KNO solution per g of NaY zeolite, based on a K/Na molar ratio of 2. After stirring for 6 h at room
3
temperature, the exchanged zeolite was recovered by filtration and thoroughly washed with distilled
water. In this case, as with all the other samples, the zeolite was calcined at 400°C for 2 h before use.
2
.2
Zeolite Characterization.
Powder diffraction (XRD) patterns were recorded in reflection geometry on a D8 Series II X-ray
diffractometer (BRUKER AXS) that uses Cu Kα radiation generated at 40 kV and 35 mA. The
elemental compositions of all zeolite samples were determined by ICP at Galbraith Laboratories. N
2
physisorption was performed on all samples by a Micromeritics ASAP 2010 unit. Prior to analysis, the
zeolite samples were degassed in situ at 230°C for 24 h. The micropore volume was derived from the t-
0
plot method (relative pressure range: 0.2−0.6) and the total pore volume was determined at p/p = 0.99.
The mesopore size distribution was obtained by applying the BJH method to the data of the adsorption
branch of the isotherm.
2
7
Al NMR experiments were performed at Florida State University on a Bruker AVIII HD NMR
spectrometer operating at a magnetic field strength of 11.74 T, equipped with a 4 mm Bruker MAS
6
Page 6 of 39

2
7
probe. For the MAS experiments of Al (130.3754 MHz), a single pulse acquisition was applied with a
spinning speed of 14 KHz and a short RF pulse (less than 15°) with a recycle delay of 0.5-1 s. Spectra
were collected after 10240 scans and referenced to AlCl (aq. 1 M) at 0 ppm.
3
The basicity of the catalysts was quantified by temperature-programmed desorption (TPD) of
2
adsorbed CO [44]. Before chemisorption, a 200-mg sample was heated in-situ under a He flow of 30
mL/min with a heating rate of 10°C /min up to 200°C, holding at this temperature for 3 h. After cooling
the sample to room temperature, it was exposed to a 30 mL/min flow of CO for 30 min and then it was
2
2
purged with He for 2 h to remove the physisorbed CO . The TPD was performed under the same He
flow rate by heating from 0 to 600°C with a heating rate of 10°C /min.
2
.3
Catalytic reaction measurements.
The faujasite-catalyzed aldol condensation of acetaldehyde was carried out in the liquid-phase
using a high-pressure 50-mL stainless steel autoclave batch reactor (Parr Corporation) equipped with
impeller, temperature and pressure controllers, and sampling port. Ethanol was selected as the solvent to
simulate conditions that might be encountered in the ethanol-to-BD process. In such a process, only a
fraction of ethanol is dehydrogenated to acetaldehyde, which undergoes the aldol condensation; the
unconverted ethanol can be used as a hydrogen source in the MPV reaction that converts
crotonaldehyde to crotyl alcohol. Reaction temperature range was 130-230°C while the total pressure
was kept at 300 psi. In a typical experiment, a measured amount of catalyst was mixed with
acetaldehyde and ethanol in the reactor vessel. After the reactor was sealed, it was purged with N
2
to
3
00 psi, and heated up to the desired reaction temperature while stirring at 300−400 rpm. In all runs, the
initial ethanol:acetaldehyde volume ratio was kept at 10:1 (2 mL of acetaldehyde). After a given period
of time (6 or 12 h), the reaction was quenched by rapid cooling. Liquid products were filtered and
analyzed using gas chromatography, GC–MS for product identification and GC–FID for quantification.
Chemical standards were used to obtain the retention times and response factors for all reactants and
products.
7
Page 7 of 39

Determining the acetaldehyde conversion carries a large error because it is highly volatile at
room temperature and during reaction can be form from ethanol via dehydrogenation or MPV reaction.
Therefore, the evaluation of the catalytic activity and selectivity of the different zeolite catalysts in the
batch was made on the basis of the total moles of carbon in the products obtained during the given time
period. At the beginning of the reaction, there are 72 mmoles of carbon from 2 mL of
acetaldehyde. However, since ethanol could partially convert to acetaldehyde the carbon yield could in
principle exceed 72 mmoles. The selectivity is defined as the moles of carbon of a specific product to
the total moles of carbon produced. The definitions are as follows:
C yield (mmol) = moles of carbon produced
moles of C4 product
×
4
C selectivity (%) =
×
100
4
Total moles of C in products
moles of C6 product×6
C selectivity (%) =
×100
6
Total moles of C in products
moles of C8 product×8
C selectivity (%) =
×100
8
Total moles of C in products
3.
RESULTS AND DISCUSSION
3
.1
Characterization of zeolites NaX and NaY
XRD patterns were obtained to examine the crystallinity of the NaX and NaY zeolites
Supplementary Material, Figure S1A). Each of the observed diffraction peaks for the two synthesized
(
samples can be assigned to hkl reflections of FAU-type zeolites [45], without any additional peaks,
which indicates that the two samples are pure faujasite zeolites. As expected, the calculated d-spacing
for the NaY sample is smaller than the one for NaX, due to the higher Si/Al ratio in NaY than in NaX
8
Page 8 of 39

(
see Table 1 for more information). This agrees with the relationship between the Si/Al ratio and unit
cell sizes reported previously [46].
2
7
Figure S1B (Supplementary Material) shows the Al MAS NMR spectra of zeolites NaX and
NaY. A single peak is seen for both samples, with a chemical shift in the range of 55-68 ppm, which
4
corresponds to the characteristic tetrahedral Al(OSi) site of a perfectly crystalline faujasite [47], and is
consistent with previously reported spectra [48]. No peak associated to octahedral extra-framework Al
was found in the spectra, demonstrating the high crystallinity of the two catalysts. The slight difference
2
7
in the chemical shift of the Al signal in NaX (61 ppm) and NaY (59 ppm) may be ascribed to the
different mean Al-O-Si bond angle in the framework [47].
The N adsorption-desorption isotherms are included in the Supplementary Material, Figure S2.
2
While zeolite NaY exhibited the characteristic isotherm of microporous materials [49], zeolite NaX
displayed a clear hysteresis, indicating the presence of some mesoporosity and a lower microporosity
than NaY. The use of NaOH during synthesis of NaX may cause partial desilication, which would
account for the observed mesoporosity. All porosity values are summarized in Table 1, along with the
elemental analysis of the zeolites investigated.
The TPD of adsorbed CO
2
provided a semi-quantitative evaluation of the strength and number of
started at about 50°C, reached a
basic sites. Figure 1 shows that, on NaX, the desorption of CO
2
maximum at about 100°C, and was essentially completed at about 250°C. Very small amounts desorbed
at higher temperatures, showing broad peaks in the regions 300-350°C and 400-450°C. An approximate
deconvolution of the profile (Supplementary Material, Figure S3A) suggests the presence of four
desorption peaks, centered at about 100°C, 150°C, 325°C, and 450°C. Previous studies have assigned
2 2
the different TPD-CO peaks to different surface carbonate species, formed via the interaction of CO
with basic sites of varying strength [50-54]. Based on these previous studies, the first two peaks could
be ascribed to bicarbonate and bidentate carbonates species on basic sites of weak to medium strength.
The two peaks at higher temperatures can be ascribed to unidentate carbonates on strong basic sites (see
9
Page 9 of 39

Scheme S1 in Supplementary Material for an illustration of the different carbonate species).
In marked contrast, the CO TPD profile for NaY (Figure 1) shows that not only the total
2
desorbed amount was four-fold lower, but also the distribution of basicity strengths was very different.
Assuming that adsorption of CO was in a 1:1 stoichiometry [53], a comparison of total basic sites in
2
NaX and NaY is presented in Table 2, which includes the TPD results for all the zeolites investigated. It
is observed that the density of total basic sites is approximately four times higher in NaX than in NaY,
which correlates with the lower Si/Al ratio of NaX and therefore its higher density of basic cations
(
Table 1). However, the most interesting observation is that the greatest difference is observed in the
density of weak basic sites, that is, desorption below 250°C. By contrast, the density of strong basic
sites desorbing above 250°C is not significantly different between the two catalysts.
3
.2
Catalytic activity of zeolites NaX and NaY
As shown in Scheme 1, self-aldol condensation of acetaldehyde produces 3-hydroxybutanal,
which typically dehydrates readily to form crotonaldehyde. Another self-condensation pathway
catalyzed by basic sites is the Tishchenko reaction [55] produces ethyl acetate. In addition, acetaldehyde
can also oligomerize forming the trimer paraldehyde [56]. As shown in the scheme, 1,1-diethoxyethane
is another possible product from the acid-catalyzed acetalization [57], in which one acetaldehyde
molecule combines with two of ethanol. Fortunately, these two reactions are reversible and only
important at low conversions. At longer reaction times, these products reverse to acetaldehyde, which
undergoes aldol condensation reaction that becomes dominant. Therefore, these reactions do not inhibit
the high selectivity to the desired C4 products.
However, crotonaldehyde can self-condense to C8 products (2,4,6-octatrienal and meta- or
ortho-tolualdehyde) or can react with acetaldehyde to form C6 products (2,4-hexadienal and 2-vinyl-
crotonaldehyde). 2,4-hexadienal reacts with acetaldehyde to form 2,4,6-octatrienal (C8 product, by aldol
condensation) or form ethyl 3-hexenoate and ethyl hexanoate (C8 products, by Tishchenko).
Finally, another crucial reaction in determining the overall selectivity is the MPV reaction,
1
0
Page 10 of 39

which is catalyzed by Lewis acids or bases. In the presence of ethanol, crotonaldehyde is reduced to
form crotyl alcohol while ethanol is oxidized to form acetaldehyde [58].
The activity and selectivity of the zeolites NaX and NaY for the aldol condensation of
acetaldehyde in ethanol were evaluated in the batch reactor. As mentioned above ethanol undergoes
dehydrogenation and H transfer producing acetaldehyde [59]. Therefore, the catalytic performance was
evaluated in terms of the amount of products obtained as a function of time. In the first set of
experiments, the reaction was carried out at 130°C for 12 h (Table 3). At this relatively low temperature,
the carbon product yield from NaY was significantly higher than from NaX. However, over NaY only
1
,1-diethoxyethane was produced, but no condensation products were obtained with this catalyst. By
contrast, a small crotonaldehyde yield, but clearly higher than on NaY was observed over NaX, which is
in line with the higher density of basic sites in NaX. While the selectivity to the undesired acetalization
product (1,1-diethoxyethane) is very high on both catalysts at 130°C, this reaction is reversible and, as
seen in Table 3, the selectivity to this product greatly decreases as the reaction temperature increases.
For instance, at 180°C, the selectivity to condensation products (crotonaldehyde and 3-hydroxybutanal)
after 12 h reaction over NaX increased to 27%. While the major product was still 1,1-diethoxyethane
(
about 66%) a clear trend was observed. Therefore, to further enhance the selectivity to condensation
products, the reaction temperature was increased to 230°C. To eliminate the catalyst deactivation issues
for the comparison, reactions were performed over zeolite NaY for different reaction times. Since the
acid-catalyzed acetalization is a reversible reaction and the product 1,1-diethoxyethane is converted
back to the reactant (ethanol and acetaldehyde), only condensation product yields are compared here. As
shown in Figure S4 (Supplementary Materials), the condensation product yield increased almost linearly
from 0, to 3 h, to 6 h. Moreover, it is observed that the C4/(C6+C8) ratio is essentially unchanged (at
about 5) for the first 6 h. The results indicate that there is no deactivation during the first 6 h (that is
total condensation yield after 6 h is twice as that of 3 h). Based on the results, 6 h of reaction time was
chosen for the comparison discussed below.
1
1
Page 11 of 39

Table 3 shows that similar carbon product yields were obtained over the two catalysts; that is,
9.0 mmol on NaX and 33.2 mmol on NaY. The decrease in acetalization selectivity was sharper on
2
NaX, but the selectivity to C4 products (crotonaldehyde, 3-hydroxybutanal, and crotyl alcohol) greatly
increased on both catalysts. Accompanying the increase in desirable products, heavier condensation
products such as C6 and C8 began to appear in significant yields on both catalysts. However, NaY
seemed to exhibit a much lower selectivity to C6-C8 products than NaX at comparable C4 yields.
A comparison between desirable and undesirable products on the two zeolites is made in Figure
2
. Clearly, with a higher density of basic sites, NaX exhibits a higher condensation/acetalization product
molar ratio than NaY (Figure 2A). However, excess density of basic sites may be adverse to maximize
C4 condensation products (crotonaldehyde, 3-hydroxybutanal, and crotyl alcohol). As seen in Figure
2
B, with essentially the same overall yield, the C4/(C6+C8) ratio is nearly twice on the less basic NaY
than on the more highly basic NaX.
The location and proximity of basic sites in the faujasite zeolites may significantly vary from
+
NaX to NaY. It is well-known that Na ions may occupy different in the faujasite structure, as illustrated
in Figure 3 [60-62]. One can see that the so-called site I is in the hexagonal prism while site I’ is in the
sodalite cage, facing the position of site I. Similarly, site II is at the center of the hexagonal window
inside the supercage, while site II’ is in the sodalite cage, facing position II. Finally, site III is located
above a square window between two other square windows and site III’ is at the edge of the square
window.
+
The location of Na cations at sites I, I’, II, and III’ in NaX were demonstrated by Olson [63],
+
using single-crystal X-ray diffraction. By contrast, in _ENREF_57zeolite NaY, the Na cations only
occupy sites I, I’, and II. It is well known that the basicity of a zeolite can be ascribed to negatively
charged framework O atoms close to the cations [51]. That is, lower Si/Al ratios require more
exchangeable cations and therefore results in higher basicity. In this case, NaX has a larger number of
cations than NaY and consequently a higher density of basic framework O atoms, connected to the
1
2
Page 12 of 39

cations located at the specific sites described above [64]. Therefore, these basic sites are located inside
and outside the supercages.
One can explain the observed differences in C4/(C6+C8) ratios in terms of the density and
location of the basic sites. A larger number of exchangeable cations occupy the supercage in NaX than
in NaY. This could result in possible confinement effects. Pidko and van Santen [65] have emphasized
that in low-silica zeolites, where the density of exchangeable cations is high, the intrazeolite
arrangement of cations plays a crucial role in activity and selectivity since these ions can undergo
multiple noncovalent interactions with the reactants, resembling the active sites encountered in
enzymes. Such interactions are present even with small molecules such as N O [66].
2
4
We should expect a larger concentration of cations inside the supercages in NaX, which may
hinder the base-catalyzed condensation, but also acetalization, which is catalyzed by Lewis acid sites
inside the cages. Thus, only the basic sites outside the supercages, active for condensation are left
available. These sites, without much steric constrains, may favor both condensation to C4 and
overcondensation to C6 and C8 products.
By contrast, NaY has a lower density of basic sites either inside or outside the supercage, but,
the accessibility to the internal cavities is higher in this zeolite than in NaX; thus, when acetaldehyde is
converted to C4 inside the zeolite, the possibility of further condensation to larger products such as C6
or C8 is more inhibited than in the case of NaX. This confinement effect should result in a higher
selectivity to the C4 product that can leave the zeolite without further reaction. At the same time, the
lower density of basic sites outside the cage further decreases the rate of over-condensation of C4
products outside the supercages. We should also point out that the lower density of basic sites in NaY,
fails to prevent the acid-catalyzed acetalization, which is greatly reduced in NaX (Figure 2A).
3
.3
Catalytic activity of zeolites NaX and NaY after pre-adsorption of CO
2
Another important point to interpret the changes in selectivity among different faujasite zeolites
1
3
Page 13 of 39

is the role played by the different distribution of basicity strengths of the sites in NaX and NaY. The
following experiment was conducted to determine whether only the stronger basic sites (i.e., those
2
responsible for CO desorption above 300°C) are active for acetaldehyde condensation. The NaX or
NaY catalyst was first exposed to CO and then tested under the same 230°C conditions. At 230°C, the
2
2
CO adsorbed on the weak basic sites would have left the surface but that one adsorbed on the strong
basic sites should still remain on the surface blocking those sites.
For NaX, the majority of the CO desorption occurs below 250°C (Figure 1). Then, with the
2
2
reaction temperature of 230°C, only a small fraction of basic sites is blocked by CO . Table 4 shows the
total carbon product yields and the selectivities to products over the CO -loaded zeolites. For NaX, only
2
a slight decrease in total product yield (22.4 mmol vs. 29.0 mmol) was observed. This result indicates
that only a small fraction of the activity can be ascribed to the strong basic sites. As shown in Figure 4,
2
a small loss in yield was observed for all products after exposure to CO , with a moderately higher loss
in yield for the condensation C4, C6, and C8 products than for those catalyzed by acid sites. For NaY
the situation is somewhat different. While the overall loss in total product yield was relatively small
2
after exposure to CO (i.e., 29.4 mmol from 32.2 mmol), the change in product distribution was much
more pronounced than that on NaX. As seen in Figure 4, the yield of acid-catalyzed reaction products
increased significantly, but the condensation C4, C6, and C8 products greatly decreased, which is
consistent with the CO
2
TPD measurements that show that a large fraction of sites was able to hold CO
2
up to temperatures significantly higher than the reaction temperature.
Figure 5A compares the condensation/acetalization product ratios before and after CO
2
loading
on each catalyst. The ratio only decreased by less than 15 % over CO -loaded NaX, but by about 50%
2
over CO -loaded NaY. These results indicate that in NaY, with a small number of basic sites present
2
and a significant fraction being covered by CO
2
, the carbon yield remained high due to the acetalization
+
reaction, which can be catalyzed by Lewis acid sites (Na ions); the yield loss was specifically on the
base-catalyzed condensation products.
1
4
Page 14 of 39

Interestingly, as shown in Figure 5B, the C4/(C6+C8) ratio was practically unchanged for either
catalyst after exposure to CO . This may suggest that all basic sites catalyze both the formation of C4
and its subsequent condensation, simultaneously.
2
3
.4.
Catalytic activity of zeolites KNaX and KNaY
As mentioned above, a high basic site density can only be accomplished at low Si/Al ratios,
which require a large number of exchangeable cations. Low-silica zeolites with Si/Al ratios close to
unity can be achieved with zeolites KNaX, which can result in the maximum possible number of cations
per unit cell [67]. Therefore, to investigate the influence of the number of cations in the zeolite, a KNaX
with low Si/Al ratio was synthesized and tested for the aldol condensation reaction. Similar to NaX, the
+
+
cations in KNaX are stabilized at sites I, I’, II, and III’, with K ions replacing part of the Na ions at
sites I’[68,69]. The KNaX zeolite has more exchangeable cations per supercage and therefore one can
expect a higher number of basic sites on the external surface than NaX [68,69].
2
7
The synthesized KNaX zeolite was thoroughly characterized by XRD, Al NMR, atomic
composition, and N adsorption. The characterization results of the KNaX material are included in the
2
Supplementary Material. They clearly indicate that the synthesized zeolite is a highly pure and
crystalline FAU-type material (Supplementary Material, Figure S5). It has a Si/Al ratio of around one
and K/Na ratio of 0.2 (Table 1). The N
characteristics and also the presence of small mesopore volume (Supplementary Material, Figure S6).
The results from the CO TPD (Table 2) would imply that KNaX has a slightly lower basic site
2
adsorption-desorption isotherms demonstrate its microporous
2
density than NaX. However, this result must be analyzed with some detail, due to its lower Si/Al ratio,
the zeolite KNaX has a larger number of exchangeable cations than NaX and, consequently, should
have a higher density of O atoms with basic characteristics. However, the TPD results do not reflect
this higher density. This discrepancy could be due to the higher density of cations in the supercages and
2
sodalities cages, which may limit the number of CO molecules that can be accommodated in these
1
5
Page 15 of 39

2
cavities. In addition, the CO TPD profile of KNaX (Figure 1) shows a clear difference in the shape of
the low-temperature peak compared to that of NaX. The narrower peak of the former might indicate that
some of the medium-strength basic sites present in the latter are eliminated. It is reasonable to ascribe
+
+
this change to the larger ionic radius of K compared to Na , which could result in weaker interactions
with CO [50].
2
The catalytic results of the acetaldehyde condensation over KNaX tested at 230°C are presented
in Table 4 and should be compared to those obtained on NaX under the same conditions. While both
catalysts exhibited about the same total product yield, KNaX gave a lower selectivity to the acetalization
reaction but at the same time it favored C8 product over the C4 condensation products.
These selectivity differences can be interpreted in terms of the confinement effects mentioned
above. That is, when the cation concentration inside the supercages is too high, not even the initial
condensation of C2 to C4 takes place inside these cavities. Clearly, with even less accessible space in
the supercage and more basic sites on the external sites, condensation to C4 and over-condensation to
C8 products occur at comparable rates, thus lowering the C4/(C6+C8) ratio.
+
To examine the effect of incorporating K ions into the NaY zeolite, a KNaY zeolite was
prepared by ion-exchange. The resulting K/Na ratio determined by ICP was found to be about 2.7. The
2
7
XRD pattern and Al NMR spectrum show the high crystallinity and structural integrity of KNaY, as
compared to NaY (Supplementary Material, Figure S7). The catalytic results of KNaY, obtained at
2
30°C under the same conditions as those used for the other zeolites are included in Table 4.
Compared to the NaY, the KNaY shows a lower carbon yield with a decrease of about 14% in
the desirable C4 product selectivity and an increase of about 11% in selectivity to the less desirable 1,1-
diethoxyethane. Furthermore, a significant change in the C6/C8 ratio was observed. While on NaY, the
2
C8 was much larger than C6, the opposite is true on KNaY. The CO TPD plot indicates that KNaY
contains less (accessible) basic sites than NaY (Table 2, see Figure S8A in Supplementary Material for
the deconvoluted TPD profile). As discussed in the case of KNaX, this decrease results from two
1
6
Page 16 of 39

+
+
factors: the incorporation of K ions inside the cages of the zeolite, which are larger than Na and may
+
block the access of molecules, and the weaker interaction between the basic sites induced by K and
2
CO .
In contrast to NaY, which has low density of basic sites but the majority are strong, when K is
incorporated, the KNaY shows a much lower density of strong basic sites. Again, the disappearance of
+
+
the strong sites from NaY can be ascribed to the weaker polarizing strength of K compared to Na
50,70].
The observed changes in the C6/C8 ratio may be due to the increased confinement caused by the
[
+
+
replacement of smaller Na by larger K ions ions, which leave less space available for condensation of
two C4 molecules that form C8, compared to one C4 and one C2 that form C6 products. The increase in
the acetalization reaction might result from some degree structural collapse that may occur during ion
exchange [71]. With some cations in the sodalite cages exposed to acetaldehyde and ethanol, the Lewis
acid-catalyzed acetalization may be accelerated.
3
.5. Catalytic activity of zeolites NaX and NaY with different extent of post-synthesis washing
The synthesis of zeolites proceeds in a highly alkaline environment. After crystallization, zeolites
are usually separated from the mother liquid and completely washed using large amounts of distilled
water. For example, the zeolites NaX and NaY discussed above were washed until pH was close to
neutral. For KNaX, it was washed until pH reached 9 to avoid hydrolysis [43]. If the zeolites are washed
using lower amounts of water occluded cations may remain in the zeolite and the properties and
catalytic activity may vary significantly. We have examined the effect of different extents of post-
washing on the activity and selectivity for acetaldehyde reaction. In this case, with zeolites NaX and
NaY, the post-washing was stopped when pH=9 was reached instead of 7, as in the standard case.
Likewise, for the KNaX, the washing step was stopped at pH=11 instead of 9, as above. These zeolites,
with lower extents of washing are identified here as NaX (lower-water), NaY (lower-water), and KNaX
(
lower-water).
1
7
Page 17 of 39

The XRD patterns did not differ from those with the more extensive washing. They had only the
2
7
peaks expected for the corresponding faujasite. Similarly, the Al NMR spectra only displayed the
signal of the tetrahedrally coordinated framework Al (Supplementary Material, Figure S6),
demonstrating that the degree of crystallinity remained high. As shown in Table 1, the elemental
analysis shows that the Si/Al ratio of the zeolites washed with low amount of water kept the same Si/Al
ratio as the the corresponding clean ones. However, these zeolites (lower-water) resulted in an excess of
alkali cations compared to Al; that is, the (K+Na)/Al ratio for KNaX (lower-water) was 1.13, compared
to 1.03 for the clean KNaX. Further, as also shown in Table 1, the zeolites (lower-water) had smaller
Vmicro and Vmeso than the corresponding clean zeolites. It is clear that the lower-water zeolites contained
2
extra alkali cations that block some of the internal cavities of the zeolites. Consistently, the CO TPD
data indicate that these impurities block some of the pores, resulting in a lower density of accessible
basic sites (Table 2, see Figure S8B-D in Supplementary Material for the deconvoluted TPD profiles).
Table 5 shows the catalytic performances of the zeolites washed with lower amounts of water
and thus containing excess alkali cations. The three types of zeolites (lower-water) gave similar carbon
yields to the corresponding clean ones (Tables 3 and 4). However, the product selectivity showed a
significant variation. A detailed comparison of the product distribution obtained over the lower-water
and clean zeolites is made in Figure 6. Clearly, over the lower-water zeolites the
condensation/acetalization product ratio was higher, but the C4/(C6+C8) ratio was lower. That is, with
the alkali impurities occluded in the zeolite cavities, the reactions tended to proceed outside the
supercages. Therefore, these lower-water zeolites were less shape selective. Likewise, the
condensation/acetalization product ratio was much lower over lower-water KNaX than over the clean
KNaX (Figure 6), which can be explained with the same arguments mentioned for the other zeolites.
Interestingly, the same selectivity and yields as those obtained on the clean KNaX can be reproduced
(
Supplementary Material, Figure S9) when the lower-water KNaX was rewashed in warm water (around
0°C). These results demonstrate that the observed differences are simply due to the presence of
5
1
8
Page 18 of 39

occluded alkali cations, which can be easily removed by more extensive washing.
Figure 6 compares the condensation/acetalization and the C4/(C6+C8) ratios. Both are desirable
product ratios in this study, but they vary in opposite directions. In principle, one would conclude that
the overall basicity is the main effect responsible for this trend; that is, a higher basicity should result in
higher condensation/acetalization ratio and lower C4/(C6+C8). However, as shown in Table 4, the
condensation yield does not increase with the density of basic sites. In fact the apparent yield/site rates
are much lower in NaX than NaY zeolites. The latter produces about the same amount of condensation
products as the former, but with a much lower density of basic sites. At the same time, the C4/(C6+C8)
is much higher on the NaY catalysts. These results are in good agreement with the concepts mentioned
above regarding the effect of accessibility and confinement. That is, the supercages in NaY zeolites are
more accessible than those of NaX and, particularly KNaX. In those cases, the acid sites responsible for
acetalization are more available, lowering the condensation/acetalization ratio. At the same time, the
greater accessibility plays a role in the C4/(C6+C8) ratio, enhancing the production of C4 and reducing
the over-condensation. By contrast, on the NaX, KNaX, and even more on the lower-water zeolites, the
majority of the cavities are not accessible for condensation reactions. Thus, they occur on the external
basic sites, which cannot benefit by shape selective effects, which lowers the C4/(C6+C8) ratio.
To further evaluate the effect of confinement and mobility inside the zeolite, a NaY containing
enhanced mesoporosity was prepared by dispersing 1 g of NaY sample in 50 mL of 0.1 M NaOH
solution. Upon stirring at room temperature for 30 min, the solid sample was obtained by filtration and
completely washed using distilled water.
The immersion in NaOH results in partial desilication and generation of mesopores. In fact, the
N adsorption-desorption isotherms indicate that the V
meso
/Vmicro ratio of the resulting NaOH-NaY
2
sample is 0.145, which is about 34 % higher than that of NaY. The mesopore size distribution graph
Supplementary Material, Figure S10) shows that NaOH-NaY contains an additional peak at around 12
nm, which reveals the presence of some additional mesoporosity in the NaOH-NaY material. The
(
1
9
Page 19 of 39

catalytic comparison between NaY and NaOH-NaY tested under the 230°C conditions is shown in the
Supplementary Material (Figure S11). Clearly, a lower condensation/acetalization product ratio and a
higher C4/(C6+C8) ratio were obtained over the NaOH-NaY catalyst, which supports the concept that
reduced confinement and enhanced mobility inside the zeolite has a positive impact on the intermediate
condensation toward C4 products, but at the same time sacrifices selectivity relative to the acid-
catalyzed acetalization.
3
.6. Role of ethanol as a solvent
To examine whether the ethanol solvent plays a direct role in the reaction, a blank experiment
was performed using only ethanol as a feed while keeping the same catalyst (400 mg NaY) and reaction
conditions unchanged (230 ℃ and 6 hrs). The result of the blank test shows that the ethanol conversion
was about 5.4% and the products included acetaldehyde, 1,1-diethoxyethane, and small amounts of
crotonaldehyde, which derive from the acetaldehyde produced. That is, the contribution of ethanol to the
condensation reaction is minimal. However, ethanol as a solvent plays an important role in preserving
the catalyst activity. In fact, when ethanol was replaced by n-hexane as the solvent, the initially white
NaY catalyst became black after the reaction, a color change that was not observed when ethanol was
the solvent and reflects the ability of ethanol to prevent coke formation by limiting the formation of
heavy condensation products via the MPV H-transfer reaction described above.
4
CONCLUSIONS
A series of NaX and NaY zeolites have been compared as catalysts for the aldol condensation of
2
0
Page 20 of 39

acetaldehyde in the liquid phase, using ethanol as a solvent and co-reactant. The target desirable product
is the first dimer condensation product (C4). However, at moderately low temperatures (130−180°C),
the dominant reaction is acetalization, producing 1,1-diethoxyethane. The preferred condensation C4
products, including crotonaldehyde, 3-hydroxybutanal, and crotyl alcohol only become major products
at about 230°C, but the C4/(C6+C8) selectivity is a strong function of the type of catalyst investigated.
It was found that, when compared at similar carbon product yields, the NaX results in higher
condensation/acetalization ratio but lower C4/(C6+C8) ratio than NaY. These differences can be
ascribed not only to the higher basicity of NaX, but also to the larger number of exchangeable cations
per supercage. With a more highly occupied cage and higher density of basic sites on the external
surface, NaX has higher activity for the condensation reactions but lower selectivity to the desirable C4
products. By contrast, on the NaY zeolite, the acid-catalyzed acetalization lowers the
condensation/acetalization ratio, but enhances the C4/(C6+C8) ratio due to a more effective shape
selectivity. This conclusion is further supported by the comparison between KNaY and NaY. The
+
enhanced shape selectivity of NaY becomes more significant upon introducing K ions in the supercage,
as reflected by the more pronounced decrease in C8 condensation product compared to C6.
Plugging sites inside the supercages while leaving external sites accessible by either increasing
the number of alkali cations or reducing the extent of post-washing treatment all have the same general
effect, that is, to reduce the importance of acetalization products but also the C4/(C6+C8) product ratio.
The incorporation of mesoporosity improves the mobility of intermediate products and increases the
C4/(C6+C8) ratio.
5
SUPPLEMENTARY MATERIAL
As mentioned in the text, some of the data have been included as supplementary material. They
2
include N adsorption/desorption isotherms, TPD profiles of adsorbed CO , illustration of carbonate
2
2
7
species formed due to CO
2
sorption, mesopore size distributions, XRD patterns, and Al NMR spectra,
as well as comparisons of carbon yields and selectivities.
2
1
Page 21 of 39

6
ACKNOWLEDGEMENTS
This research has been funded by Abengoa Research.
REFERENCES
[
1]
2]
H.-J. Chae, T.-W. Kim, Y.-K. Moon, H.-K. Kim, K.-E. Jeong, C.-U. Kim, S.-Y. Jeong, Appl.
Catal., B 150−151 (2013) 596−604.
[
Butadiene Product Stewardship Guidance Manual, American Chemistry Council Olefins Panel,
American Chemistry Council, 2001, Revised 2010.
[3]
[4]
[5]
J. G. Aston, G. Szasz, H. W. Woolley, F. G. Brickwedde, J. Chem. Phys. 14 (1946) 67−79.
F. J. Dumez, G. F. Froment, Ind. Eng. Chem. Process Des. Dev. 15 (1976) 291−301.
M. D. Jones, C. G. Keir, C. Di Iulio, R. A. M. Robertson, C. V. Williams, D. C. I. Apperley,
Catal. Sci. Tech. 1 (2011) 267−272.
[
[
[
[
[
[
[
[
6]
7]
8]
9]
L. R. Lynd, R. T. Elamder, C. E. Wyman, Appl. Biochem. Biotechnol. 57 (1996) 741−761.
C. E. Wyman, Annu. Rev. Energy Environ. 24 (1999) 189−226.
J. R. Mielenz, Curr. Opin. Microbiol. 4 (2001) 324−329.
C. N. Hamelinck, G. V. Hooijdonk, A. P. Faaij, Biomass Bioenergy 28 (2005) 384−410.
10] S. Phillips, Ind. Eng. Chem. Res. 46 (2007) 8887−8897.
11] J. J. Spivey, A. Egbebi, Chem. Soc. Rev. 36 (2007) 1514−1528.
12] H. van der Heijden, K. J. Ptasinski, Energy 46 (2012) 200−210.
13] A. Dutta, M. Talmadge, J. Hensley, M. Worley, D. Dudgeon, D. Barton, P. Groenendijk, D.
Ferrari, B. Stears, E. Searcy, Environ. Prog. Sustain. Energy 31 (2012) 182−190.
14] S. Kvisle, A. Aguero, R. Sneeden, Appl. Catal. 43 (1988) 117−131.
15] S. Bhattacharyya, S. Sanyal, J. Catal. 7 (1967) 152−158.
[
[
[
[
16] E. Makshina, W. Janssens, B. Sels, P. Jacobs, Catal. Today 198 (2012) 338−344.
17] H. Jones, E. Stahly, B. Corson, J. Am. Chem. Soc. 71 (1949) 1822−1828.
2
2
Page 22 of 39

[
[
[
18] M. León, E. Díaz, S. Ordóñez, Catal. Today 164 (2011) 436−442.
19] W. Ji, Y. Chen, H. H. Kung, Appl. Catal., A 161 (1997) 93−104.
20] L. M. Baigrie, R. A. Cox, H. Slebocka-Tilk, M. Tencer, T. T. Tidwell, J. Am. Chem. Soc. 107
(
1985) 3640−3645.
[
[
21] R. N. Hayes, R. P. Grese, M. L. Gross, J. Am. Chem. Soc. 111 (1989) 8336−8341.
22] O. Kikhtyanin, V. Kelbichová, D. Vitvarová, M. Kubů, D. Kubička, Catal. Today 227 (2014)
1
54−162.
[
[
[
[
[
23] E. Dumitriu, V. Hulea, N. Bilba, G. Carja, A. Azzouz, J. Mol. Catal. 79 (1993) 175−185.
24] A. Biaglow, J. Sepa, R. Gorte, D. White, J. Catal. 151 (1995) 373−384.
25] E. Dumitriu, A. Azzouz, V. Hulea, D. Lutic, H. Kessler, Microporous Mater. 10 (1997)1−17.
26] Y.-C. Chang, A.-N. Ko, Appl. Catal., A 190 (2000) 149−155.
27] E. Dumitriu, V. Hulea, I. Fechete, A. Auroux, J.-F. Lacaze, C. Guimon, Microporous
Mesoporous Mater. 43 (2001) 341−359.
[
[
[
[
[
[
[
[
[
[
[
[
28] S. Luo, J. L. Falconer, Catal. Lett. 57 (1999) 89−93.
29] B. Noziere, A. Cordova, J. Phys. Chem. A 112 (2008) 2827−2837.
30] D. Tichit, D. Lutic, B. Coq, R. Durand, R. Teissier, J. Catal. 219 (2003) 167−175.
31] M. Singh, N. Zhou, D. K. Paul, K. Klabunde, J. Catal. 260 (2008) 371−379.
32] J. E. Rekoske, M. A. Barteau, Ind. Eng. Chem. Res. 50 (2011) 41−51.
33] D. Barthomeuf, J. Phys. Chem. 88 (1984) 42−45.
34] D. Barthomeuf, J. Phys. Chem. 97 (1993) 10092−10096.
35] D. Barthomeuf, Microporous Mesoporous Mater. 66 (2003) 1−14.
36] D. Vinek, H. Noller, M. Ebel, K. Schwarz, J. Chem. Soc., Faraday Trans. 73 (1977) 734−746.
37] S. M. Csicsery, Zeolites 4 (1984) 202−213.
38] A. Corma, J. Catal. 216 (2003) 298−312.
39] J. Jae, G. A. Tompsett, A. J. Foster, K. D. Hammond, S. M. Auerbach, R. F. Lobo, G. W. Huber,
2
3
Page 23 of 39

J. Catal. 279 (2011) 257−268.
[40] Y. Lee, S. W. Carr, J. B. Parise, Chem. Mater. 10 (1998) 2561−2570.
[41] T. Frising, P. Leflaive, Microporous Mesoporous Mater. 114 (2008) 27−63.
[42] G. Xiong, Y. Yu, Z.-C. Feng, Q. Xin, F.-S. Xiao, C. Li, Microporous Mesoporous Mater. 42
(
2001) 317−323.
[
[
[
43] G. H. Kuhl, Zeolites 7 (1987) 451−457.
44] J. Chang Kim, H.-X. Li, C.-Y. Chen, M. Davis, Microporous Mater. 2 (1994) 413−423.
45] M. M. J. Treacy, J. B. Higgins, Collection of Simulated XRD Powder Patterns for Zeolites, 5th
ed.; Elsevier: Amsterdam, The Netherlands, 2007.
[
[
[
[
[
46] J. Turkevich, S. Ciborowski, J. Phys. Chem. 71 (1967) 3208−3217.
47] E. Lippmaa, A. Samoson, M. Magi, 108 (1986) 1730−1735.
48] J. Klinowski, J. Thomas, C. Fyfe, G. Gobbi, J. Hartman, Inorg. Chem. 22 (1983) 63−66.
49] R. Pierotti, J. Rouquerol, Pure Appl. Chem. 57 (1985) 603−619.
50] D. Bonenfant, M. Kharoune, P. Niquette, M. Mimeault, R. Hausler, Sci. Technol. Adv. Mater. 9
(
2008) 013007.
[
[
[
51] D. Barthomeuf, Catal. Rev. 38 (1996) 521−612.
52] J. Di Cosimo, V. Díez, M. Xu, E. Iglesia, C. Apesteguıa, J. Catal. 178 (1998) 499−510.
53] M. A. Aramendı ́a , V. Borau, C. Jiménez, J. M. Marinas, J. R. Ruiz, F. J. Urbano, J. Appl. Catal.,
A 244 (2003) 207−215.
[
54] M. A. Aramendía, V. Borau, C. Jiménez, J. M. Marinas, J. R. Ruiz, F. J. Urbano, J. Colloid
Interface Sci. 238 (2001) 385−389.
[
[
[
55] T. Seki, H. Kabashima, K. Akutsu, H. Tachikawa, H. Hattori, J. Catal. 204 (2001) 393−401.
56] W. K. Busfield, R. M. Lee, D. Merigold, J. Chem. Soc., Faraday Trans. 69 (1973) 936−940.
57] M. R. Capeletti, L. Balzano, G. de la Puente, M. Laborde, U. Sedran, Appl. Catal., A 198 (2000)
L1−L4.
2
4
Page 24 of 39

[
[
[
58] Y. Zhu,; G.-K. Chuah,; S. Jaenicke, J. Catal. 241 (2006) 25−33.
59] Y.-J. Tu, Y.-W. Chen, C. Li, J. Mol. Catal. 89 (1994) 179−189.
60] K. S. Walton, M. B. Abney, M. Douglas LeVan, Microporous Mesoporous Mater. 91 (2006)
7
8−84.
[
[
[
[
61] K. H. Lim, C. P. Grey, J. Am. Chem. Soc. 122 (2000) 9768−9780.
62] E. Jaramillo, S. M. Auerbach, J. Phys. Chem. B 103 (1999) 9589−9594.
63] D. H. Olson, Zeolites 15 (1995) 439−443.
65] E.A. Pidko, R.A. van Santen, “Theoretical chemistry of zeolite reactivity” in: J. Cejka, A. Corma,
S. Zones (eds.), Zeolites and Catalysis: Synthesis, Reactions and Applications, Chapter 13,
Volume 1 Wiley-VCH, Weinheim, 2010.
[
66] E. A. Pidko, P. Mignon, P. Geerlings, R. A. Schoonheydt, R. A. van Santen, J. Phys. Chem. C
1
12 (2008) 5510−5519.
[67] Y. Li, R. T. Yang, J. Phys. Chem. B 110 (2006) 17175−17181.
[
68] M. Iwama, Y. Suzuki, J. Plévert, K. Itabashi, M. Ogura, T. Okubo, Cryst. Growth Des. 10
(
2010) 3471−3479.
[
[
[
69] S. Caldarelli, A. Buchholz, M. Hunger, J. Am. Chem. Soc. 123 (2001) 7118−7123.
70] E. Gallei, G. Stumpf, J. Colloid Interface Sci. 55 (1976) 415−420.
71] K. Sato, Y. Nishimura, N. Matsubayashi, M. Imamura, H. Shimada, Microporous Mesoporous
Mater. 59 (2003) 133−146.
2
5
Page 25 of 39

Highlights
•
•
•
•
•
NaX and NaY can be selective for aldol condensation of acetaldehyde to C4 products
At 230°C, higher C4/(C6+C8) product ratio is produced over NaY than over NaX
A confinement effect observed when adding K ions to N
aY and NaX zeolites
Cage accessibility varied by alkali addition greatly influences C4 selectivity
2
6
Page 26 of 39

ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Tableꢀ1.ꢀMolarꢀraꢁosꢀ(fromꢀICPꢀchemicalꢀanalysis)ꢀandꢀporosityꢀcharacterisꢁcsꢀofꢀtheꢀzeoliteꢀsamples.ꢀ
3
3
Catalystꢀ
NaXꢀ
Si/Alꢀ
ꢀ1.28ꢀ
(K+Na)/Alꢀ
0.99ꢀ
K/Naꢀ
0ꢀ
Vmicroꢀ(cm /g)ꢀ
Vmesoꢀ(cm /g)ꢀ Vmesoꢀ/Vmicroꢀ
0.267ꢀ
0.144ꢀ
0.039ꢀ
0.049ꢀ
NAꢀ
0.539ꢀ
0.108ꢀ
0.170ꢀ
NAꢀ
NaYꢀ
KNaXꢀ
ꢀ2.14ꢀꢀ
ꢀ1.03ꢀ
2.08ꢀ
ꢀ1.01ꢀ
ꢀ1.25ꢀ
ꢀ2.11ꢀ
2.10ꢀ
1.03ꢀ
0.99ꢀ
0.92ꢀ
1.13ꢀ
1.07ꢀ
1.08ꢀ
0.99ꢀ
0ꢀ
0.20ꢀ
2.65ꢀ
0.23ꢀ
0ꢀ
0.362ꢀ
0.288ꢀ
NAꢀ
KNaYꢀ
KNaXꢀ(lowerꢀwater)ꢀ
NaXꢀ(lowerꢀwater)ꢀ
NaYꢀ(lowerꢀwater)ꢀ
NaOH-NaYꢀꢀ
0.248ꢀ
0.225ꢀ
0.331ꢀ
0.324ꢀ
0.037ꢀ
0.085ꢀ
0.027ꢀ
0.047ꢀ
0.149ꢀ
0.378ꢀ
0.082ꢀ
0.145ꢀ
0ꢀ
0ꢀ
Page 27 of 39

ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Tableꢀ2.ꢀDensityꢀofꢀbasicꢀsitesꢀofꢀtheꢀzeoliteꢀsamplesꢀasꢀdeterminedꢀfromꢀCO ꢀchemisorpꢁon.ꢀꢀ
2
Numberꢀofꢀbasicꢀsitesꢀforꢀdifferentꢀpeaksꢀ(μmolꢀCO /g)ꢀ
Totalꢀbasicꢀsitesꢀꢀ
2
Catalystꢀ
(
μmolꢀCO /g)ꢀ
2
DesorbedꢀbelowꢀT=250°Cꢀ
DesorbedꢀaboveꢀT=250°Cꢀ
NaXꢀ
NaYꢀ
ꢀ253.9ꢀ
ꢀ59.2ꢀ
ꢀ224.7ꢀ
ꢀ35.7ꢀ
ꢀ93.2ꢀ
ꢀ236.8ꢀ
ꢀ56.4ꢀ
186.6ꢀ
17.0ꢀ
67.3ꢀ
42.2ꢀ
32.4ꢀ
5.3ꢀ
KNaXꢀ
192.3ꢀ
30.4ꢀ
KNaYꢀ
KNaXꢀ(lowerꢀwater)ꢀ
NaXꢀ(lowerꢀwater)ꢀ
NaYꢀ(lowerꢀwater)ꢀ
87.9ꢀ
4.3ꢀ
193.7ꢀ
29.9ꢀ
43.1ꢀ
26.5ꢀ
Page 28 of 39

ꢀ
Tableꢀ3.ꢀCatalyꢁcꢀperformanceꢀofꢀzeolitesꢀNaXꢀandꢀNaYꢀ(massꢀ400ꢀmg)ꢀforꢀaldolꢀcondensaꢁonꢀofꢀacetaldehyde.ꢀReportedꢀasꢀoverallꢀ
productꢀCꢀyieldꢀandꢀproductꢀdistribuꢁon.ꢀBatchꢀreactorꢀatꢀvaryingꢀtemperatures,ꢀN ꢀpressureꢀ300ꢀpsi,ꢀsꢁrringꢀrateꢀ300−400ꢀrpm,ꢀ
2
solventꢀ(ethanol)ꢀ/ꢀreactantꢀ(acetaldehyde)ꢀvolumeꢀraꢁoꢀ10ꢀwithꢀacetaldehydeꢀ2ꢀmL.ꢀ
Catalystꢀ
NaXꢀ
NaYꢀ
Reacꢁonꢀtemperatureꢀ(℃)ꢀ
Reacꢁonꢀꢁmeꢀ(h)ꢀ
130ꢀ
12ꢀ
180ꢀ
12ꢀ
230ꢀ
6ꢀ
130ꢀ
12ꢀ
ꢀ180ꢀ
ꢀ12ꢀ
ꢀ230ꢀ
ꢀ6ꢀ
ꢀ
ProductꢀCꢀyieldꢀ(mmol)ꢀ
24.8ꢀ
25.8ꢀ
29.0ꢀ
42.2ꢀ
ꢀ21.3ꢀ
ꢀ33.2ꢀ
CarbonꢀSelecꢁvityꢀ(%)ꢀ
1
,1-diethoxyethaneꢀ
Crotonaldehydeꢀꢀ
94.2ꢀ
5.8ꢀ
0.0ꢀ
0.0ꢀ
0.0ꢀ
0.0ꢀ
0.0ꢀ
0.0ꢀ
65.7ꢀ
23.5ꢀ
3.5ꢀ
0.0ꢀ
6.1ꢀ
0.0ꢀ
0.0ꢀ
1.2ꢀ
10.7ꢀ
48.7ꢀ
4.6ꢀ
100.0ꢀ
ꢀ0.0ꢀ
ꢀ0.0ꢀ
ꢀ0.0ꢀ
ꢀ0.0ꢀ
ꢀ0.0ꢀ
ꢀ0.0ꢀ
ꢀ0.0ꢀ
ꢀ74.4ꢀ
ꢀ18.7ꢀ
ꢀ2.5ꢀ
ꢀ0.0ꢀ
ꢀ4.4ꢀ
ꢀ0.0ꢀ
ꢀ0.0ꢀ
ꢀ0.0ꢀ
ꢀ15.7ꢀ
ꢀ51.3ꢀ
ꢀ4.4ꢀ
3
-hydroxybutanalꢀ
Crotylꢀalcoholꢀꢀ
Ethylꢀacetateꢀ
Paraldehydeꢀꢀ
4.1ꢀ
ꢀ4.8ꢀ
11.2ꢀ
0.0ꢀ
ꢀ8.6ꢀ
ꢀ3.3ꢀ
MixedꢀcondensaꢁonꢀC6ꢀ
MixedꢀcondensaꢁonꢀC8ꢀ
3.2ꢀ
ꢀ1.8ꢀ
17.5ꢀ
ꢀ10.0ꢀ
Page 29 of 39