Bio-nanohybrid catalysts based on l-leucine immobilized in hydrotalcite and their activity in aldol reaction

Article abstract of DOI:10.1016/j.apcata.2016.03.018

Nanohybrid materials based on l-leucine (l-Leu) and hydrotalcites (HT) were prepared by the ion-exchange and reconstruction method, under mild synthesis conditions. The location, amount and the form of the immobilized l-Leu are affected not only by the ti

Full text of DOI:10.1016/j.apcata.2016.03.018

Accepted Manuscript
Title: Bio-nanohybrid catalysts based on L-leucine
immobilized in hydrotalcite and their activity in aldol reaction
Author: Dana-Georgiana Crivoi Ronald-Alexander Miranda
´
Elisabetta Finocchio Jordi Llorca Gianguido Ramis Jesus E.
Sueiras Anna M. Segarra Francisco Medinaa
PII:
S0926-860X(16)30140-5
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2016.03.018
APCATA 15810
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
29-1-2016
11-3-2016
17-3-2016
Please cite this article as: Dana-Georgiana Crivoi, Ronald-Alexander Miranda,
´
Elisabetta Finocchio, Jordi Llorca, Gianguido Ramis, Jesus E.Sueiras, Anna M.Segarra,
Francisco Medinaa, Bio-nanohybrid catalysts based on L-leucine immobilized in
hydrotalcite and their activity in aldol reaction, Applied Catalysis A, General
http://dx.doi.org/10.1016/j.apcata.2016.03.018
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.

Bio-nanohybrid catalysts based on L-leucine immobilized in hydrotalcite and their activity
in aldol reaction
Dana-Georgiana Crivoi,a,b Ronald-Alexander Miranda,a,b* Elisabetta Finocchio,c Jordi Llorca,d,e Gianguido Ramis,c
Jesús E. Sueiras,a,b Anna M. Segarra,a,b and Francisco Medinaa,b*
aDepartament d’Enginyeria Química, Universitat Rovira i Virgili, Av. Països Catalans, 26, Campus Sescelades,
Tarragona 43007, Spain
bEMaS-Centro de Investigación en Ingeniería de Materiales y Micro/nanoSistemas, Campus Sescelades, Tarragona
4300, Spain
cDipartamento di Ingegneria Civile, Chimica e Ambientale , Università di Genova, P.le J.F.Kennedy 1, Genoa
16129, Italy
dInstitut de Tècniques Energètique, Universitat Politècnica de Catalunya, Diagonal 647, ed. ETSEIB, Barcelona
08028, Spain
eCenter for Research in NanoEngineering (CRnE-UPC). Universitat Politècnica de Catalunya, Pasqual i Vila 1-15,
Barcelona 08028, Spain.
*E-mail: ronaldalexandermiranda@hotmail.com , francesc.medina@urv.cat
1

Graphical abstract

Highlights:




The organic/inorganic interactions between L-leucine immobilized into hydrotalcites were
studied
By reconstruction method, the anionic form of L-leucine is immobilized in the interlayer space
while by anionic exchange, both zwitterionic and anionic forms are immobilized
These nanohybrid materials are more active in the asymmetric aldol reaction than the mere L-
leucine
Modulating reaction conditions (solvent, type of nanohybrid used, time) can lead to different
reaction outcome

Nanohybrid materials based on L-Leucine (L-Leu) and hydrotalcites (HT) were prepared by the ion-exchange and
reconstruction method, under mild synthesis conditions. The location, amount and the form of the immobilized L-
Leu are affected not only by the time of synthesis, but also by temperature and ultrasound treatment. The XRD
results demonstrate that the immobilization occurs in either a vertical or oblique orientation with respect to the HT
layers. The catalytic activity of these materials was tested in the aldol addition reaction of cyclohexanone with
different aromatic aldehydes, affording mainly the syn-diastereomer. Furthermore, the present study demonstrates
that both diastereo- and enantioselectivity can be easily modulated by the appropriate combination of nanohybrid
catalyst, solvent and reaction time.
Keywords: Asymmetric catalysis, Hydrotalcite, Immobilization, L-Leucine, Nanohybrid materials

1. Introduction
Nowadays, considerable attention is focused on the synthesis of nanohybrid materials, which exhibit new and
better properties than the corresponding constituent materials. Research into and understanding the
organic/inorganic interaction between bioactive molecules and the surface of inorganic materials have led to their
use in many biochemical applications[1] or as catalysts [2] and drug delivery carriers [3], although these
interactions remain only partially understood. Many bioactive molecules (e.g. peptides, amino acids, proteins, etc.)
are anions under neutral and basic pH, so they can easily be immobilized in positively charged solids. Examples of
such solids are the hydrotalcites (HTs), a family of naturally occurring layered clays with low or null toxicity, good
biocompatibility and a high anion swelling capacity. These properties make HTs interesting materials for
applications in the pharmaceutical field [4], cosmetics [5], catalysis [6] or even medical field [7].
The anionic exchange properties of HTs transform these materials in excellent candidates for the
immobilization of amino acids (AAs). Starting with 1997 when Whilton et al. studied the immobilization of aspartic
and glutamic acid in the interlayer space of layered double hydroxides through coprecipitation method [8], a series
AAs have been immobilized into HT materials by means of three general methods: coprecipitation [9-12], anionic
exchange [12-15] and reconstruction method [16, 17]. Regarding the nature of the interaction, it is now clear that the
immobilization of AAs in HTs structures is pH-dependent, although other factors should be taken into account: the
kind of HT, the synthesis and the physical and chemical properties of the AAs.
AAs are organocatalysts which display several advantages such as non-toxicity, easy manipulation and
stability and have been widely used in the asymmetric aldol reaction [18]. For example, Córdova et al. obtained anti
diastereoselectivity in a direct asymmetric aldol reaction between cyclohexanone and p-NO2-benzaldehyde using a
series of AAs [19]. Wu et al. used a threonine derivative in the aldol addition reaction, producing the anti-
diastereomer when cyclohexanone was the substrate and syn-isomer in the case of hydroxyacetone [20]. Similar
results were obtained by Barbas [21], Lu [22] and Gong [23]. Itoh et al. observed that in the presence of L-t-Leu
compounds with syn diastereoselectivity were obtained in the case of cyclopentanone, cycloheptanone and
cyclooctanone and with anti diastereoselectivity when cyclohexanone was used [24]. Thus, obtaining the syn-
product when cyclohexanone is used as substrate seems to be a challenge. Moreover, AAs used in their natural form
can react with aromatic aldehydes to form the corresponding immonium salts which, by decarboxylation, will form a
stable side-product [24, 25].
In this study, we present the synthesis of nanohybrid materials based on L-Leu immobilized in HTs. We
studied different synthetic procedures and their effects on the nature of the inorganic/organic interaction in order to

evaluate: i) the role of the immobilization time on one hand, and the relationship between the immobilization speed
and the strength and kind of basic centres in the HT on the other hand; ii) the role of the HT precursor in the
immobilization process and iii) the nature of the AA structure in the immobilization process. To achieve this goal,
the nanohybrid materials were synthesized using the anion-exchange and reconstruction method and the
organic/inorganic interactions were investigated by EA, ICP, XRD, FT-IR, Raman, 13C, 27Al MAS NMR and
thermal evolution using TG/DTA analyses.
The new nanohybrid materials were tested in the aldol addition reaction of cyclohexanone with different
aromatic aldehydes. The synergistic effect between the bio-organic guest (L-Leu) and the inorganic host (HTs)
proved to play an important role in modulating both the diastereoselectivity and enantioselectivity of the final
product. Herein, we report that the anti-syn selectivity depends not only on the nature of the catalyst but also on the
solvent used.
2. Experimental
2.1 General
All chemicals and solvents were commercially available (Aldrich Chemical, Fluka) and used without
further purification/drying unless otherwise mentioned.
Molecular formulae were calculated from the results of elemental analyses (EA) and inductively coupled
plasma analyses (ICP). EA were performed using an elemental analyser EA-1108 C.E. instrument from
Thermo Fisher Scientific with a Mattler Toledo MX5 microbalance. The analyses were carried out using
atropine as a standard and Vanadium as an additive to facilitate combustion. ICP analyses were performed
in an ICP-OES Spectro Arcos FHS16 Instrument.
The N2-physisorption analysis of BET surface areas and average pore diameters were performed in a
QuadStar Quantachrome surface analyser at 77 K. Before analysis all the samples were degassed in vacuum
at 393 K for 12 h.
Powder X-ray diffraction (XRD) patterns of the samples were performed on a Bruker-AXS D8-Discover
diffractometer with a 2θ angle ranging from 3º to 70º. The samples were dispersed onto a low background
Si (510) sampler holder. The data were collected with an angular step of 0.03º at 5s per step and sample
rotation. CuKα radiation (λ = 1.54056 Å) was obtained from a copper X-ray tube operated at 40 kV and 40
mA. The crystalline phases were identified using JCPDS files. The interlayered spaces were analysed with
the reflection bands of the (003) and (006) and calculated using the Bragg law.

Fourier transform infrared spectra (FT-IR) were recorded with Nicolet Nexus Fourier Transform
instrument equipped with a DTGS KBr detector. Each analysis was performed using 100 scans in the range
4000–400 cm-1. Pure powders diluted in KBr pressed disks (about 1% w/w) were used for the analysis of
skeletal vibrations. OMNIC software provided by ThermoElectron Corporation was used for spectra
analysis.
Raman spectra were obtained using a Renishaw Raman via reflex instrument. The polarized radiation
(λ=785 nm) of a Ranishaw diode laser of 500 mW was used. A Laica DM2500 optical microscope was used
to determine the part of the sample analysed. The RamaScope was calibrated using a silicon wafer. The
focus (maximum opening 100%) and power (50%) were carefully optimized in order not to alter the sample
during measurement. The spectral resolution was 2 cm-1 with an exposure time of 10 s and 5 accumulations
for each run.
13C and 27Al Magic Angle Spinning-Nuclear Magnetic Resonance (MAS-NMR) spectra were obtained
on a Varian Mercury VXR-400S spectrometer operating at 104.2 MHz with a pulse width of 1 ms. A total
of 4,000 scans were collected with a sweep width of 100 kHz and an acquisition time of 0.2 s. An
acquisition delay of 1s between successive accumulations was selected to avoid saturation effects. 13C MAS
NMR spectra were recollected using tetramethylsilane (TMS) as reference.
High-resolution transmission electron microscopy (HRTEM) was performed with a JEOL 2010F
instrument equipped with a field emission source, working at an acceleration voltage of 200 kV. The point-
to-point resolution of the microscope was 0.19 nm, and the resolution between lines was 0.14 nm.
Thermogravimetric Analyses and Differential Thermal Analyses (TGA/DTA) were measured on a
TGA7 instrument from Perkin Elmer. The analyses were carried out using a sample amount of 10 mg in an
N2 atmosphere. The heating rate was 10 ºC.min-1 within the range 30-900 ºC.
1
The products were characterized by H-NMR using a Varian NMR System 400 MHz and HPLC-DAD
(Diode Array Detector G1315D) Agilent Technologies and HPLC-RID 10A (Refractive Index Detector)
Shimadzu using CHIRALPACK IA column (250*4.6 mm ID).
2.2 Synthesis of hydrotalcite materials (HTs)
Mg-Al HTs (Mg/Al molar ratio 2) containing nitrates and chloride anions were synthesized by the
coprecipitation method at room temperature and pH=10. The materials obtained were named HTNO3 and
HTCl respectively. After the drying process, HTCl was sonicated for 1 hour, while HTNO3 was decomposed
by calcination at 450 ºC overnight in air. The calcined HT (HTcc) was rehydrated in an inert atmosphere

using decarbonated water and ultrasound treatment for 1 hour. The materials obtained were named HTClus
and HTrus respectively.
2.3 Synthesis of LL/HT materials
2.3.1 Anionic-exchange method (method A)
Two procedures were used to synthesize LL/HTx-Ay materials, where x is the type of HT used (HTrus or
HTClus) and y indicates the procedure followed. In the first case, 500 mg of HT was added to a solution
containing 320 or 160 mg (2.4 or 1.2 mmol, respectively) of L-Leu. The mixture was stirred for 30 minutes
at room temperature (method A1). In the second case, 500 mg of HT was added to a solution containing
840.4 mg (6.4 mmol) of L-Leu. The mixture was stirred for 3 hours at 80 ºC (method A2). An Ar
atmosphere and deionized-decarbonated water were used in all cases. Materials synthesized by the anionic
exchange method were named LL/HTrus-A1, LL/HTClus-A1, LL/HTrus-A2 and LL/HTClus-A2.
2.3.2 Reconstruction method (method R)
Two procedures were used to synthesize LL/HTx-Ry materials (where x is the type of HT and y
indicates the procedure followed), in both cases using 250 mg of HTcc added to a solution containing 736.9
mg (5.6 mmol) of L-Leu. In the first procedure, L-Leu-HTcc mixture was first treated by sonication for 1
hour and then the slurry was stirred for another 3 hours at 80 ºC. The material obtained was named
LL/HTrus-R1 (method R1). Alternatively, HTcc-L-Leu mixture was stirred for 3 hours at 80 ºC (method R2).
The material obtained was named LL/HTr-R2. Table 1 summarizes the synthetic methods used in the
present article.
2.4 Typical procedure for the aldol reaction of cyclohexanone
H2O (25 mmol) and cyclohexanone (1.5 mmol) were added to a mixture containing aldehyde (0.5
mmol), catalyst (2.5 mol % L-Leucine with respect to ketone) in DMSO (2ml) or Toluene (2 ml) and the
resulting mixture was stirred at room temperature. After a defined period of time, brine was added and the
reaction mixture was extracted several times with CH2Cl2. Due to its high boiling point, DMSO cannot be
removed by vacuum distillation. In this context, DMSO was diluted by adding brine solution (NaCl
solution) to: (i) facilitate the separation of aqueous and organic phases after adding CH2Cl2 and (ii) destroy
any emulsion that might be formed during the extraction process The organic layer was dried with MgSO4
and concentrated under vacuum. The reaction mixture was purified by column chromatography using silica

1
gel and hexane/ethyl acetate 3/1 as eluent. The conversion was determined by either isolation or H-NMR
analysis and the enantiomeric excess by HPLC analysis.
3. Results and discussion
3.1 Catalyst characterization
3.1.1 Textural properties of HT and LL/HT materials
Nanohybrid materials based on L-Leu and HTs were synthesized by two methods (anion-exchange and
reconstruction) to understand the nature of immobilized L-Leu in the HT structure: the location and ionic
state of this AA. Table 2 shows the L-Leu/Al3+ molar ratio, BET surface area and interlayer space of the
different HTs and nanohybrid materials synthesized.
Cavani et al. explained that the introduction of large anionic species in the interlayer space of a
hydrotalcite material, either by anionic exchange method or reconstruction method, will be reflected in an
increase in the interlayer space of the HT [26]. In accordance with Cavani’s findings, we have observed an
increase in the HT’s gallery height in all the materials where the L-Leu was immobilized inside the
hydrotalcite (entries 7, 9 and 10, Table 2). When the anionic exchange method was used at room
temperature for 30 minutes (LL/HTrus-A1 material) no increase in the gallery height of the support was
observed, even though a ratio of 0.44 L-Leu/Al3+ was obtained (entry 5, Table 2). This demonstrates that
the AA was preferably immobilized on the edges of the HTrus layers. An increase in the amount of
immobilized AA will produce a loss of crystallinity, observed in the XRD patterns presented in the
Supporting Information (Figure I).
Increasing the time and temperature of synthesis to obtain LL/HTrus-A2 increased the amount of
immobilized L-Leu to 1.09 mol L-Leu/mol Al3+, producing an interlayer space to 19.5 Å (entry 7, Table 2).
This finding shows that both time and temperature have an effect on the swelling of the HT structure,
allowing the immobilization of L-Leu molecules in its interlayer space. The loss of crystallinity in all
nanohybrid materials was due to the decrease of the layers’ lengths by the ultrasound effect.
The same conclusions can be drawn from the HRTEM images of the precursor and the nanohybrid
materials. In the case of HTrus, the image shows aggregated layers up to about 100 nm in length, with an
interlayer space around 7.6 Å as deduced by FT analysis (Figure 1a). The LL/HTrus-A1 material (Figure 1
b) has a similar morphology as the HTrus, demonstrating once again that using method A1 the L-Leu is not
found inside the HT layers, but on the edges. The distinctive morphology observed in the case of LL/HTrus-

A2 (Figure 1 c) compared to that of HTrus proves the intercalation of the LL inside the layers and not on the
edges of the material.
To sustain our previous findings, we have conducted several control experiments using hydrotalcites
containing Cl-; these HTs have a lower basicity than the HTrus [27] and do not favour the immobilization of
hydrophobic amino acids through anionic-exchange method [17]. Indeed, in both cases (LL/HTClus-A1 and
LL/HTClus-A2) no increase in the interlayer space was observed, demonstrating that the AA did not
exchange the chloride ions (entries 6 and 8, Table 2). Moreover, the very low amount of L-Leu computed
by EA and ICP analysis suggest that the HTClus surface is not basic enough to immobilize a larger amount of
amino acid.
Materials synthesized by reconstruction method (LL/HTrus-R1 and LL/HTr-R2) exhibited a significant
shift in the d003 peak position compared with the HTrus XRD pattern (Figure I in Supporting Information).
Furthermore, the interlayer space of these nanohybrid materials increased to around 20 Å (entries 9 and 10,
table 2). The HRTEM images (Figure 2) do not conserve the layer morphology of their HT precursor and
revealed interlayer spacing at 22.5-22.6 Å, indicating that the immobilization of L-Leu molecules occurs in
the interlayer space of the HT. In addition to the basal planes at 22.5-22.6 Å, both LL/HTrus-R1 and
LL/HTr-R2 materials present lattice fringes at 14.8-14.9 Å.
Nakayama et al.[17] immobilized L-leucine into HT using a reconstruction method and observed an
increase in the gallery height of 9.8 Å corresponding to the thickness of two L-leucine molecules arranged
in a bilayer structure. The material LL/HTrus-A1 presented a similar gallery height as the normal HT which
demonstrate that the immobilized L-leucine is found at the edges of the HT in a horizontal position with
respect the HT layer. On the other hand, the material LL/HTrus-A2 has a gallery height of 14.3 Å. The
approximate length of a L-leucine molecule is of 7.109 Å, thus the interlayer distance observed for
LL/HTrus-A2 can be explained by a bilayer structure of L-leucine oriented in a vertical position with respect
to the HT layer.
The high gallery heights observed in the case of nanohybrids synthesized using reconstruction method
suggest that the bilayer structure of L-leucine contain the amino acids in both vertical and oblique
orientation with respect to the HT layer.
BET surface area results (found in Table 1) are in agreement with the previous discussion: when L-Leu
is immobilized inside the HT-structure the surface area decreases considerably compared to the parent
material. The LL/HTrus-R1 and LL/HTr-R2 materials presented a more significant decrease in surface area

loss than the ones by A1 and A2 method. This suggests that the AA is found within the aggregates formed
by the disordering of the layers (during ultrasound treatment) and their position hinders the adsorption of
N2, reducing the available surface area of the materials (see the N2 adsorption-desorption isotherms –Figure
II and III in Supporting Information).
3.1.2 FT-IR and Raman spectroscopy
Skeletal FT-IR and Raman spectra of L-Leu, HTClus and HTrus can be found in Supporting Information (Figure
IV) along with the detailed explanations of the specific bands of each parent material. Moreover, the thermal
decomposition curves of the nanohybrid precursors can be found in Supporting Information (Figures VII and VIII).
3.1.2.1 Nanohybrids synthesized by the anionic exchange method
To better understand and evaluate the interaction of L-Leu with the HT surface, we used the control
materials based on HTClus. The FT-IR (up) and Raman (bottom) spectra of LL/HTClus-A1 and LL/HTClus-A2
do not present strong evidence of immobilized L-Leu (Figure 3a and 3b, respectively). In the case of the
LL/HTClus-A1, it is difficult to draw conclusions because of the very low content of L-Leu present in the
sample (entry 6, Table 2). The FT-IR spectrum of LL/HTClus-A2 material showed some differences
compared to its HT precursor. At high frequency, two bands at 2957 and 1581 cm-1 can be observed due to
the small amount of L-Leu molecules interacting with HT material. At low frequency, the differences in the
bands at 970, 790 and 553 cm-1 compared to the spectrum of the HT precursor suggest changes in the Al-
OH species. These changes could result from the presence of a higher amount of OH- anions and to some L-
Leu molecules immobilized in the HT structure. However, the Raman spectrum of LL/HTClus-A2 does not
present any evidence of immobilized L-Leu (Figure 3b, bottom).
The FT-IR spectra of nanohybrid materials (Figure 4 up) synthesized using HTrus material showed,
besides the vibration bands due to the HT precursor, some sharp components attributable to the L-Leu,
demonstrating that the AA was successfully immobilized. LL/HTrus-A1 and LL/HTrus-A2 spectra exhibit the
bands corresponding to the C-H stretching mode at 2975 (νCH) and 2870 cm-1 (νCH3), slightly shifted with
respect to the bands of pure Leu, confirming the immobilization of the L-Leu. The FT-IR spectrum of
LL/HTrus-A1 material exhibits the νa(COO-) and νs(COO-) stretching at 1560 and 1407 cm-1 respectively, while
+
bands attributable to the NH3 group were not clearly detected (Figure 4a, up). These findings, together with
the XRD and HRTEM prove that the immobilization process was mainly caused by an anionic exchange
between anionic L-Leu and the OH- groups located on the HT edges. Similarly, in the FT-IR spectrum of

LL/HTrus-A2 material, the νa(COO-) stretching at 1560 cm-1 indicates the presence of the anionic L-Leu
(Figure 4b up). An exhaustive study of this spectrum also shows small bands at 1581, 1610 and 1516 cm-1
+
due to COO- and NH3 groups of the zwitterionic L-Leu. The decrease in intensity of the broad band
2-
specific to water OH stretching mode and of the band ascribed to CO3 species at 1384 cm-1 in both
LL/HTrus-A1 and LL/HTrus-A2 materials suggests that the incorporation of the L-Leu in the materials causes
the displacement of physisorbed water and the competition with the adsorption of CO32- anions [28, 29].
The Raman spectra of LL/HTrus-A1 and LL/HTrus-A2 materials present bands at 2886, 1227 and 835 cm-
1 due to CH stretching modes (Figure 4a and 4b bottom, respectively). The form in which L-Leu is found in
the LL/HTrus-A1 is difficult to be identified due to the low intensities of the RAMAN bands (Figure 4a,
bottom). In the LL/HTrus-A2 Raman spectrum, the bands at 1471 and 1455 cm-1 attributable to the COO-
group in pure L-Leu shifted to 1464 and 1449 cm-1 respectively (Figure 4b, bottom). In addition, the
relative intensity of both bands changes after immobilization, suggesting the presence of two kinds of COO-
groups in L-Leu: anionic and zwitterionic. Changes in their relative intensity also indicate that both COO-
are interacting with the HT structure, anions located in the interlayer space and/or other L-Leu molecules.
2-
Moreover, the relative intensity of the band at 1061 cm-1 due to CO3 species decreased after L-Leu
immobilization, confirming the reduced incorporation of atmospheric CO2 in the HT structure.
In general, the FT-IR and Raman spectra of nanohybrid materials synthesized by anion exchange
method show that the immobilization process cannot occur on the surface of the HT material. In addition,
the high basicity of the HTrus can favour the immobilization of L-Leu in its anionic form until the accessible
2-
OH- groups in the HT edges are compensated. This interaction protects the material from CO3
incorporation. Nevertheless, the decrease in the strong basic centres in the material causes the formation of
+
zwitterionic AA (L-Leu detected through diagnostic bands ascribed to NH3 ) which could interact with Al-
OH species, structural water, OH- groups still available in the material or other zwitterionic L-Leu
molecules.
3.1.2.2 Nanohybrids synthesized by the reconstruction method
The FT-IR (up) and Raman (bottom) spectra of nanohybrid materials synthesized by the reconstruction
method are presented in Figure 5. The FT-IR spectrum of LL/HTr-R2 (Figure 5b up) presents broad bands
+
at 3064 cm-1 and around 2700 cm-1 assigned to vibration modes of the NH3 group. In addition, the band at
+
2130 cm-1 is also due to a combination of asymmetric deformation and hindered rotation of NH3 groups
[30]. The relative intensity of all the bands assigned to the NH3+ group decreased following immobilization,

+
in comparison to the spectrum of the pure Leu, probably due to H-bonds between the NH3 group and the
oxygen atoms of the HT layers. Bands due to νa(COO-) and νs(COO-) stretching were also detected as in
pure L-Leu. Although the intercalation of some anionic L-Leu cannot be completely ruled out, immobilized
L-Leu using HTcc without ultrasound treatment clearly occurs mostly in its zwitterionic form.
The Raman spectrum of LL/HTr-R2 material presents bands at 2886, 1227 and 835 cm-1 attributable to
CH and CC stretching modes (see Figure 5b, bottom). The band at 1455 cm-1 due to the COO- group in pure
L-Leu spectrum remains unchanged after the immobilization process, while the band at 1471 cm-1 shifted to
1464 cm-1 after immobilization. Moreover, the detection of weak and broad band at 1186 cm-1 and shifted
+
band at 1125 cm-1 due to NH3 group demonstrated the immobilization of both zwitterionic and anionic L-
Leu molecules.
FT-IR and Raman spectra of LL/HTrus-R1 presented no significant differences compared to LL/HTrus-A2
(see Figure 5a and Figure 4b, respectively). The band due to νa(COO-) stretching shifted to lower
+
frequencies from 1581 to 1560 cm-1, while the bands corresponding to NH3 stretching vibrations were not
detected. This suggests that the immobilization of the L-Leu occurs in its anionic form, in agreement with
Aisawa et al. findings [9]. Only two differences in the Raman spectrum of LL/HTrus-R1 were detected in
comparison with the LL/HTrus-A2 spectrum: the band at 1297 cm-1 due to the deformation mode of the -OH
+
in plane and the band at 1187 cm-1 due to the NH3 group were not detected. This indicated that the
immobilization mainly occurred in an anionic form.
In conclusion, the FT-IR and Raman spectra of the nanohybrid materials synthesized using the
reconstruction method show that using ultrasound treatment during the synthesis the L-Leu molecules are
immobilized in their anionic form.
All the information extracted from the Raman and FT-IR analysis of the nanohybrid materials is
summarized in Table 3.
3.1.3 MAS NMR Spectroscopy
To investigate the nature of the interaction between L-Leu and the HTs using MAS NMR spectroscopy,
we have chosen the nanohybrid materials that contained the highest L-Leu/Al3+ molar ratio: LL/HTrus-A2,
LL/HTrus-R1 and LL/HTr-R2.
The typical 27Al MAS-NMR spectrum of HTrus presents an important signal at 9 ppm due to the
octahedral coordinated Al and two small signals at 105 and 81 ppm attributable to extra framework
tetrahedral Al atoms still present after rehydration process (see figure V in Supporting Information) [31-33].

Because none of the nanohybrid materials 27Al MAS-NMR spectra had the signal at 9 pm sifted,
demonstrates that the interaction of the immobilized L-Leu with the HT does not involve the Al atoms
(entries 2, 3 and 4, Table 4).
The 13C MAS-NMR spectrum of the pure L-Leu exhibits the typical signals of the zwitterionic form,
mainly: a signal at 176 ppm due to the carboxylate group, signals at 55 and 44 pm corresponding to Cα and
Cβ respectively and a broad band at 26 ppm assigned to the Cγ and Cδ (entry 1, Table 4). The shifting of the
carboxylate group signal from 176 ppm to 186 ppm in the case of LL/HTrus-A2 and LL/HTrus-R1 spectra
(entries 2 and 3, Table 4) confirm that L-Leu was immobilized in the anionic form [16]. The 13C MAS-
NMR spectrum of the LL/HTrus-A2 material presented also a less intense signal at 176 ppm, suggesting that
some of the L-Leu is found in a zwitterion form (Supporting Information figure VI).
The same two signals were also observed in the case of LL/HTr-R2 material (entry 4, Table 4), but the
intensities were inversed, demonstrating that the L-Leu was immobilized mainly under the zwitterionic
form.
3.2 Nature of the organic/inorganic interaction
The extensive and detailed characterization presented in the above section revealed that variations in the
immobilization methods lead to materials with different characteristics, as it is summarized in Scheme 1.
Briefly, when the starting material was HTcc, the L-Leu immobilization took place in the same time as
the rearrangement of the HT. Ultrasound treatment (in the case of LL/HTrus-R1 material) breaks the HT
layers creating more basic sites, which, in the presence of water and L-Leu facilitate the immobilization
process. The L-Leu is found in the anionic form between the HT layers. When the hybrid was synthesised
without ultrasounds (LL/HTr-R2 material), the L-Leu was immobilized in the interlayer space in both
anionic and zwitterionic form, producing a catalyst with a higher crystallinity. Our results indicate that
immobilization of the anionic L-Leu occurs by compensating the OH- groups found in the HT interlayer
+
space. The immobilization of zwitterionic L-Leu occurs by H-bonding between the -NH3 group of the L-
Leu with water and/or OH- groups in the HT layers.
When the synthesis started with HTrus, the temperature used was the crucial parameter. Thus, the
synthesis carried out at 80 °C (material LL/HTrus-A2) favoured the immobilization of L-Leu in the
interlayer space by ionic exchange until all accessible centres were compensated, then the immobilization
occurred by interactions between the zwitterionic L-Leu with water or OH- anions located in the interlayer

space. When the HTrus was simply stirred in the presence of L-Leu (LL/HTrus-A1), immobilization occurred
by anionic exchange of the OH- anions located on the edge of the HT with the anionic L-Leu.
3.3 Asymmetric aldol reaction of cyclohexanone
T. Itoh et al.[24] reported the low activity of L-Leu in the aldol reaction of p-nitrobenzaldehyde with
cyclohexanone in the presence of DMSO and water for seven days. Under these conditions, the reaction gave the
corresponding -ketoalcohol with a very low yield, moderate anti diastereoselectivity and high enantioselectivity
towards the (2S,1’R)-aldol compound (entry 1, Table 5). The low activity is attributable to the cyclization of the
imine intermediate (formed in the reaction between the amino acid and the aldehyde) followed by a decarboxylation,
resulting in a stable 1,3-oxazolidine compound [25].
Rehydrated hydrotalcites have proven to possess high activity in several reactions, due to the presence of a large
number of –OH groups which act as Brönsted basic sites and are presumed to be responsible for the catalytic
activity of these materials. On this basis, the HTrus was tested in the aldol reaction of different benzaldehyde-like
compounds with cyclohexanone using the conditions mentioned by T. Itoh et al. [24] (Table 5). The HTrus afforded
high yields and efficiently controlled syn-selectivity when substituted aromatic aldehydes were used (entries 2, 5 and
8, Table 5), while the anti-selectivity was preferred in the case of benzaldehyde (entry 11, Table 5).
Using the same conditions, we tested L-Leu/HTrus materials in the aldol reaction of the four chosen aromatic
aldehydes. We used LL/HTrus-R1 and LL/HTrus-A2 as model catalyst due to the nature of AA immobilized (as
mentioned in the previous section, in the R1 the L-Leu is present in the anionic form while in A2 in both zwitterion
and anionic form). The manner in which the catalyst was prepared turned out to be crucially important in the
reaction efficiency. Both catalysts yield good conversions in the reaction of 4-nitrobenzaldehyde (entries 3 and 4,
Table 5) compared to the case when free L-Leu was used (entry 1, Table 5). Even though sin diastereoselectivity
was preferred, the form in which L-Leu was immobilized greatly influenced the enantioselectivity. Thus, when L-
Leu was present in the anionic form - LL/HTrus-R1 – 90% ee for the syn isomer was obtain and only 60% ee for the
anti. When L-Leu was present in both anionic and zwitterion form, the ee % of the syn diastereoisomer decreased to
71% while the ee% for anti-increased to 74%. In the case of 2-nitrobenzaldehyde the reaction done in the presence
of LL/HTrus-R1 afforded also the syn-diastereoisomer, but enantioselectivity was higher towards the anti-isomer
(entry 6, Table 5). This shows that the steric hindrance of the nitric group found in ortho position and the anionic
form of L-Leu play an important part in determining the enantioselectivity. A similar trend was observed when
LL/HTrus-A2 was used, exhibiting a lower stereoselectivity for the syn isomer. The second para-substituted aldehyde
used in this study showed similar trends in both cases, but with a slighter increase in ee% for the syn isomer (entries

9 and 10, Table 5). A totally different behaviour was observed in the case of benzaldehyde: LL/HTrus-R1 favoured
the formation of the syn diastereoisomer, with good to moderate enantioselectivity for both diastereoisomers, while
the LL/HTrus-A2 favoured the anti-isomer but with practical no enantioselectivity and reverse ee % for the syn
(entries 12 and 13, Table 5). All these results clearly show that both LL/HTrus-R1 and LL/HTrus-A2 are far better
catalysts than pure L-Leu, employing the same conditions as T. Itoh et al. [24].
As control experiments, we have studied the catalytic activity of the physical mixture between L-Leu and the
corresponding HT (calcined or rehydrated) for the aldol reaction between benzaldehyde and cyclohexanone (entries
14 and 15, Table 5). High conversions were obtained in both cases and the diastereoselectivity is similar to the one
obtained with LL/HTrus-A2. This catalytic behaviour indicates that part of the HTcc is rehydrated in the reaction
medium favouring the intercalation of the L-Leu to give the LL/HTrus-A2 material. The XRD diffractograms of the
physical mixture of HTcc and L-Leu before and after the reaction, along with the XRD pattern of the LL/HTrus-A2
can be found in Supporting Information. Interestingly, the trends of the ee% for syn and anti are similar to that of the
corresponding catalysts, slightly higher in the case of HTrus + LL and lower for the other physical mixture.
To evaluate the reusability and stability of the catalysts, we chose as substrate benzaldehyde, due to its unique
behaviour in terms of diastereoselectivity and enantioselectivity. When LL/HTrus-A2 was used and under the
conditions proposed by T. Itoh et al. [24] the catalyst lost its activity after first run (entry 1, Table 6). In the
LL/HTrus-A2 material, L-Leu is found in both zwitterion and anionic form and due to the exothermic reaction that
takes place during work-up (adding brine over reaction mixture) the amino acid is removed from the HT, decreasing
thus the activity of the catalyst (figure XI in Supporting Information). Moreover, the Cl- anions found in the brine
solution are able to replace the vacant zones of the HT interlamellar space, leading to the formation of an inactive
catalyst. The amino acid which is now free in solution will react with benzaldehyde leading to the formation of the
corresponding 1,3-oxazolidine (figure XII in Supporting Information). The catalyst LL/HTrus-R1 presented slightly
higher activity in the second run compared to the other case, suggesting a stronger interaction between the anionic
form of the amino acid and the HT interlamellar space than in the case of LL/HTrus-A2 (entry 2, Table 6).
To avoid the use of NaCl, we repeated the reactions using toluene as solvent (entry 3 and 4, Table 6). In this
case, two different behaviours were observed depending on the nature of the catalyst: when LL/HTrus-A2 was used,
an increase towards the anti-diastereoisomer can be observed over the three runs while, in the other case (LL/HTrus-
R1), an increase of the syn-diastereoisomer is observed. In terms of enantioselectivity, a slightly increase in ee% is
detected for both isomers over the three runs, in the case of LL/HTrus-A2, while when LL/HTrus-R1 is used, a change

in enantioselectivity is observed for the syn diastereomer. These preliminary results suggest that the solvent plays an
important part in the reaction pathway and might have an effect on the reaction enantioselectivity.
For an in-depth study of this process and to understand the reaction profile, we performed a 4h kinetic study of
the aldol reaction between cyclohexanone and benzaldehyde under all the conditions mentioned above (catalysed by
HTrus, mechanical mixture between L-Leu and HT, LL/HTrus-A2 and LL/HTrus-R1 and the conditions mentioned by
T. Itoh et al. [24]).
The first study was done using HTrus as catalyst. (Figure 6). It is intriguing to note that, although the reaction
was complete after only 1h, there is a continuous interchanging between the syn and the anti diastereoisomers.
Miller et al. discovered a stereoisomeric system in which spontaneous enantiomeric enrichment occurred in a
homogeneous mixture. They observed an amide isomerization proportional to the fluctuation of each diastereomer’s
enantiomeric ratio as the cis-trans equilibration occurred [34].
To explain the observed behaviour, we have to take into account the strong basic sites present in the HTrus
which can promote the reaction by subtracting an H atom in the α-position of the ketone compound [35]. Scheme 2
presents a possible mechanism of the syn-anti isomerization, where the cyclohexanone (1) will be in equilibrium
with the enol form (2) in aqueous solvent and will react further with the benzaldehyde to give the corresponding
aldol product (4). Due to the presence of the interlayer –OH groups and the water medium, H-bonds play an
important role in stereoselectivity, favouring the syn-anti inversion [36].
A different behaviour is observed when the reaction is carried out using the corresponding HT mixed with L-
Leu. In the first case (Figure 7-A), two different catalysts are present in the reaction mixture: HTcc and L-Leu. As
HTcc has lower basic sites than HTrus, the aldol reaction will require longer time for completion. Additionally, the
quantity of water present in the medium is able to rehydrate the HT material and thus, proceeding in the formation
of LL/HTrus-A2. The free amino acid is involved in three competitive processes: catalysing the formation of 1,3-
oxazolidine compound, catalysing the aldol reaction and being involved in the immobilization process. At this point
is difficult to know which form of L-Leucine is active in this experiment. Surprising, along the 4 h a slow decrease
in the enantioselectivity of the reverse anti isomer is observed.
Furthermore, the system HTrus + L-Leu is much more active giving a higher selectivity towards the aldol
products (figure 7-B). In this case, three processes may occur simultaneously: formation of the 1,3 – oxazolidine,
aldol reaction and immobilization of the AA. From all of them, the reaction system favours the immobilization of L-
Leu on the HTrus and thus, only a small portion of the L-leucine will be involved in the formation of 1,3-oxazolidine

compound and of the anti –aldol product (Figure 7-B – values in brackets). After 2 hours of reaction, the catalyst
LL/HTrus-A2 is already formed.
The degree of interchanging between syn and anti-isomers depends on the nature of the catalyst: in the case of
LL/HTrus-A2 during the first hour the syn diastereoisomer is present in a higher concentration than the anti, but in
time its concentration decrease as it can be seen in Figure 8-A; on the other hand, when LL/HTrus-R1 was used, the
anti diastereoisomer is in excess from the beginning, and after 3h the conversion towards the syn diastereoisomer
increases, as observed in Figure 8-B. Regarding the enantioselectivity, during the first 3 hours of reaction in the
presence of LL/HTrus-R1 the ee% remains constant for the anti-isomer (Figure 8-B – values in brackets) while there
is no enantioselectivity for the syn diastereoisomer. In the case of LL//HTrus-A2 a change in stereoselectivity is
observed for the syn diastereomer and a slightly increase in enantioselectivity for the anti (Figure 8-A – values in
brackets).
Figure 9 illustrates that the reaction proceeds slower in toluene than in aqueous DMSO and total conversion is
achieved after 3h. The catalyst obtained through the anionic exchange method (Figure 9-A) behaves totally different
when toluene is used and in the first 4h of reaction there is no interchanging between the diastereoisomers. In the
same time, the enantioselectivity towards the other syn isomer remains constant, while the enantioselectivity of the
anti-product is much lower than in the case were DMSO was used as solvent. In addition, when LL/HTrus-R1 is
used, after 2 h of reaction an inversion between the diastereoisomers is observed, while the ee% is increasing for the
syn isomer and slightly decreases for the anti (Figure 9-B). As mentioned in the “Nature of the organic/inorganic
interaction” section, when the catalyst is prepared through the A2 method, the active basic sites (interlamellar –OHs)
are interacting with the L-Leu found in both anionic and zwitterion form. In a nonpolar solvent and with no water
present, the LL/HTrus-A2 catalyst is incapable of catalysing the interchanging between the diastereoisomers.
Moreover, when only the anionic form of L-Leu is immobilized, as in the case of LL/HTrus-R1, there are more –OH
available, thus promoting the interchanging as can be seen in figure 9 B.
Literature presents several reactions done “in water” or “in the presence of water”, the last being reported to
increase the reactivity and stereoselectivity [37, 38]. In this context, we have performed the 4h kinetic study of aldol
reaction in toluene and “in the presence of water”, results being shown in Figure 10. Adding only a small amount of
water in toluene completely changed the course of reaction. In both cases there was a high selectivity towards the
anti-diastereoisomer, a very good enantioselectivity of this isomer and no enantioselectivity for the sin isomer
(Figure 10 A and B). Our results show that the solute-solvent interactions are able to affect the stereochemical
outcome of the aldol reaction even when a non-polar solvent is used.

Cordová et al. have shown that there are three possible mechanisms by which a free AA can catalyse the aldol
reaction: the carboxylic acid catalysed enamine mechanism, the amino catalysed enamine mechanism and the
enaminium catalysed mechanism, where the amino acid is found in the zwitterion form [39]. These mechanisms
cannot be applied when the amino acid is immobilized between the layers of hydrotalcites, due to the presence of the
secondary material which is also active in aldol reaction. So far, our results show that when the reaction carried out
in polar solvent (DMSO) the conversion and enantioselectivity are low, but when non-polar solvents are used with a
very small quantity of water, the catalytic activity is improved. This might be explained as follows: in the non-polar
medium, the substrate has to penetrate into the layers of the HT and, as Vijaikumar et al.[2] shown, the polar nature
of the interlayer will help the catalysis as well as the chiral induction of the AA to occur. Scheme 3 presents a
possible mechanism of the aldol reaction catalysed by the LL/HT catalysts. Firstly, the amino group of L-Leu
activates the ketone by a nucleophilic attack at the carbonyl carbon, generating an enaminium intermediate. The syn
diastereoselectivity is influenced by the highly basic character of the hydrotalcite. The –OH groups of the HT
interact with the aldehyde, polarizing it and facilitating the nucleophilic attack of the enamine, which, in presence of
water will give the corresponding syn-adduct. Amedjkouh showed that bases can be used as co-catalysts in the adol
reaction to enforce syn-selectivity retaining a high level of enantioselectivity, and unbounded amino acids usually
favour the formation of anti-aldol product [40]. Next, the syn isomer can undergo isomerization, as presented in
Scheme 2.
The HT interlayer space provides the perfect environment for the L-leucine, increasing its activity by
promoting enantioselectivity through an oriental attack. The activity of this bifunctional catalyst is enhanced by the
polar medium found in HT galleries and by the presence of water. As Gong et al. showed, the double hydrogen
bonding activation of carbonyl functionality influence the transition state, exhibiting high stereoselectivity towards
the syn-aldol product if the reaction is not quenched in the first 4h [41].
The behaviour of these new bio-hybrid catalysts is extremely complicated and certainly require further
investigation for a complete comprehension on how the solvent, time of reaction and the way the catalyst is prepared
affect the activity and enantioselectivity.
4. Conclusions
The controllable basicity, the “memory effect” and the ion exchange property make HTs excellent
materials to immobilize amino acids and study their organic/inorganic interactions. In this context, we have
obtained four new materials based on L-leucine and Mg-Al HT by means of two procedures: reconstruction
method and anionic exchange method.

In the anionic exchange method, the immobilization involves mainly the anionic form of L-Leu and an
increase in the synthesis time and temperature permits the swelling of the HT material, allowing the
immobilization to occur inside the HT layers. An excess of L-Leu in the reaction media will favour the
immobilization on the edges of the HT structure. The use of ultrasound treatment during reconstruction
method not only increased the basic sites of the HT, but also favoured the immobilization of L-Leu inside
its layers. The absence of this treatment will favour the AA intercalation by means of H-bonding between
its structure and the inorganic anions found in the interlayer space.
Regardless of the intercalation method used, we have shown that the immobilization cannot occur
through the interaction between the Al3+ atoms from the HTs structure and the anionic L-Leu. Moreover,
according to the XRD results, the immobilization could occur in a vertical or oblique orientation with
respect to the HT layers (Scheme 1). When the AA was immobilized in the zwitterionic form, the
orientation depended on the chemical environment in the interlayer space from the HT.
These new materials were more active in the asymmetric aldol reaction of cyclohexanone with different
aromatic aldehydes (up to 99% conversion and moderate to good enantioselectivity) than the corresponding
physical mixture and the mere amino acid. When the reaction was carried out in aqueous DMSO, the
catalysts lost their activity after the first run, due to the exothermic reaction that takes place during work-up
(adding brine over reaction mixture) when L-Leu is removed from the HT. Just by using toluene, both
conversion and selectivity were highly improved.
The kinetic studies we performed have shown that the aldol reaction between cyclohexanone and
benzaldehyde carried out using these new catalysts, toluene and “in the presence of water” is completed in
less than 4 h, achieving 99% yield and maintaining the enantioselectivity in the case of the anti-
diastereomer.
The mechanism of reaction is very complex and involves simultaneous activity of the two parts of the
catalyst: the hydrotalcite and the immobilized L-Leu. If in general the free amino acid favours the formation
of the anti-isomer, the presence of the basic HT leads to the formation of the syn-isomer. Moreover,
depending on the conditions of reaction, the syn and anti-diastereoisomer can undergo isomerization.
Therefore, the different synthesis protocols and different reaction conditions can be employed to obtain a
specific aldol-product. Further studies will be carried out to exploit the potential advantages and properties
of these LL/HTrus catalysts in the asymmetric aldol reaction of cycloketones. This future work will focus on

understanding how the solvent, time and the way the catalyst is prepared affect the activity and
enantioselectivity, making use of theoretical computations and possible transition states.
Acknowledgements
Dana-Georgiana Crivoi is grateful to the Spanish Government’s MECD for the FPU predoctoral grant.
Notes and references
[1] J.-H. Choy, S.-J. Choi, J.-M. Oh, T. Park, Clay minerals and layered double hydroxides for novel
biological applications, Appl. Clay. Sci., 36 (2007) 122–132.
[2] S. Vijaikumar, A. Dhakshinamoorthy, K. Pitchumani, L-Proline anchored hydrotalcite clays: An
efficient catalyst for asymmetric Michael addition, Appl. Catal., A, 340 (2008) 25–32.
[3] J.-H. Yang, Y.-S. Han, M. Park, T. Park, S.-J. Hwang, J.-H. Choy, New inorganic-bsed drug
delivery system of indole-3-acetic acid-layered metal hydroxide nanohybrids with controlled
release rate, Chem. Mater., 19 (2007) 2679-2685.
[4] M.D. Arco, E. Cebadera, S. Gutiérrez, C. Martín, M. Montero, V. Rives, J. Rocha, M. Sevilla,
Mg,Al layered double hydroxides with intercalated indomethacin: synthesis, characterization,
and pharmacological study, J. Pharm. Sci., 93 (2004) 1649-1658.
[5] B. Ballarina, A. Mignani, F. Mogaveroa, S. Gabbaninic, M. Morig, Hybrid material based on
ZnAl hydrotalcite and silver nanoparticles for deodorant formulation, Appl. Clay. Sci., 114 (2015)
303-308.
[6] R.J. Chimentão, S. Abelló, F. Medina, J. Llorca, J.E. Sueiras, Y. Cesteros, P. Salagre, Defect-
induced strategies for the creation of highly active hydrotalcites in base-catalyzed reactions, J.
Catal., 252 (2007) 249–257.
[7] D.-H. Park, G. Choi, J.-H. Choy, Bio-layered double hydroxides nanohybrids for theranostics
applications, Photofunctional Layered Materials2015, pp. 137-175.
[8] N.T. Whilton, P.J. Vickers, S. Mann, Bioinorganic clays: synthesis and characterization of
amino- and polyamino acid intercalated layered double hydroxides, J. Mater. Chem., 7 (1997)
1623–1629.
[9] S. Aisawa, S. Takahashi, W. Ogasawara, Y. Umetsu, E. Narita, Direct intercalation of amino
acids into layered double hydroxides by coprecipitation, J. Solid State Chem., 162 (2001) 52-62
[10] B.M. Choudary, B. Kavita, N.S. Chowdari, B. Sreedhar, M.L. Kantam, Layered double
hydroxides containing chiral organic guests: Synthesis, characterization and application for
symmetric C-C bond-formin reactions, Catal. Lett., 78 (2002) 373-377.
[11] Q. Yuan, M. Wei, Z. Wang, G. Wang, X. Duan, Preparation and characterization of L-aspartic
acid-intercalated layered double hydroxide, Clays Clay Miner., 52 (2004) 40–46.
[12] M. Wei, J. Guo, Z. Shi, Q. Yuan, M. Pu, G. Rao, X. Duan, Preparation and characterization of
L-cystine and L-cysteine intercalated layered double hydroxides, Journal of Materials Science 42
(2007) 2684-2689.
[13] Å. Fudala, I. Pálinkó, B. Hrivnák, I. Kiricsi, Amino acid-pillared layered double hydroxide and
montmorillonite thermal characteristics, J. Therm. Anal. Calorim., 56 (1999) 317-322.
[14] Á. Fudala, I. Pálinkó, I. Kiricsi, Amino acids, precursors for cationic and anionic intercalation
synthesis and characterization of amino acid pillared materials, J. Mol. Struct., 482-483 (1999)
33-37.
[15] Á. Fudala, I. Pálinkó, I. Kiricsi, Preparation and characterization of hybrid organic-inorganic
composite materials using the amphoteric property of amino acids: amino acid intercalated
layered double hydroxide and montmorillonite, Inorg. Chem., 38 (1999) 4653-4658.

[16] S. Aisawa, H. Kudo, T. Hoshi, S. Takahashi, H. Hirahara, Y. Umetsu, E. Narita, Intercalation
behavior of amino acids into Zn–Al-layered double hydroxide by calcination–rehydration
reaction, J. Solid State Chem., 177 (2004) 3987–3994.
[17] H. Nakayama, N. Wada, M. Tsuhako, Intercalation of amino acids and peptides into Mg–Al
layered double hydroxide by reconstruction method, Int. J. Pharm., 269 (2004) 469–478.
[18] B. List, R.A. Lerner, C.F.B. III, Proline-catalyzed direct asymmetric aldol reactions, J. Am.
Chem. Soc., 122 (2000) 2395-2396.
[19] A. Córdova, W. Zou, I. Ibrahem, E. Reyes, M. Engqvist, W.-W. Liao, Acyclic amino acid-
catalyzed direct asymmetric aldol reactions: alanine, the simplest stereoselective organocatalyst,
Chem. Commun., (2005) 3586–3588.
[20] X. Wu, Z. Jiang, H.-M. Shen, Y. Lu, Highly efficient threonine-derived organocatalysts for
direct asymmetric aldol reactions in water, Adv. Synth. Catal., 349 (2007) 812 – 816.
[21] S.S.V. Ramasastry, K. Albertshofer, N. Utsumi, F. Tanaka, I. Carlos F. Barbas, Mimicking
fructose and rhamnulose aldolases: organocatalytic syn-aldol reactions with unprotected
dihydroxyacetone, Angew. Chem. Int. Ed., 46 (2007) 5572–5575.
[22] Z. Jiang, Z. Liang, X. Wu, Y. Lu, Asymmetric aldol reactions catalyzed by tryptophan in water,
Chem. Commun., (2006) 2801–2803.
[23] X.-Y. Xu, Y.-Z. Wang, L.-Z. Gong, Design of organocatalysts for asymmetric direct syn-aldol
reactions, Org. Lett., 9 (2007) 4247-4249.
[24] T. Kanemitsu, A. Umehara, M. Miyazaki, Kazuhiro Nagata, T. Itoh, L-t-leucine-catalyzed
direct asymmetric aldol reaction of cyclic ketones, Eur. J. Org. Chem., (2011) 993–997.
[25] F. Orsini, F. Pelizzoni, M. Forte, R. Destro, P. Gariboldi, 1,3 dipolar cycloadditions of
azomethine ylides with aromatic aldehydes. Syntheses of 1-oxopyrrolizidines and 1,3-
oxazolidines, Tetrahedron, 44 (1988) 519-541.
[26] F. Cavani, F. Triffrò, A. Vaccari, Hydrotalcite-type anionic clays: preparation, properties and
applications., Catal. Today, 11 (1991) 173-301.
[27] M.G. Álvarez, R.J. Chimentão, F. Figueras, F. Medina, Tunable basic and textural properties
of hydrotalcite derived materials for transesterification of glycerol, Appl. Clay. Sci., 58 (2012) 16-
24.
[28] S. Aisawa, S. Sasaki, S. Takahashi, H. Hirahara, H. Nakayama, E. Narita, Intercalation of
amino acids and oligopeptides into Zn–Al layered double hydroxide by coprecipitation reaction,
J. Phys. Chem. Solids, 67 (2006) 920–925.
[29] A.R. Garcia, R.B.d. Barros, A. Fidalgo, L.M. Ilharco, Interactions of L-alanine with alumina as
studied by vibrational spectroscopy, Langmuir 23 (2007) 10164-10175.
[30] N.B. Colthup, L.H. Daly, S.E. Wiberley, Introduction to infrared and raman spectroscopy, 3rd
ed.1990.
[31] A. Béres, I. Pálinkó, J.-C. Bertrand, J.B. Nagy, I. Kiricsi, Dehydration-rehydration behaviour of
layered double hydroxides: a study by X-ray diffractometry and MAS NMR spectroscopy J. Mol.
Struct., 410-411 (1997) 13-16.
[32] J. Rocha, M.d. Arco, V. Rives, M.A. Ulibarri, Reconstruction of layered double hydroxides
from calcined precursors: a powder XRD and 27Al MAS NMR study, J. Mater. Chem., 9 (1999)
2499-2503.
[33] F. Rey, V. Fornéa, Thermal decomposition of hydrotalcites an infrared and nuclear magnetic
resonance spectroscopic study, J. Chem. Soc., Faraday Trans., 88 (1992) 2233-2238.
[34] K.T. Barrett, A.J. Metrano, P.R. Rablen, S.J. Miller, Spontaneous transfer of chirality in an
atropisomerically enriched two-axis system, Nature, 509 (2014) 71-75.
[35] E. Dumitriu, V. Hulea, C. Chelaru, C. Catrinescu, D. Tichit, R. Durand, Influence of the acid–
base properties of solid catalysts derived from hydrotalcite-like compounds on the condensation
of formaldehyde and acetaldehyde, Appl. Catal., A, 178 (1999) 145-157.
[36] S. Paladhi, A. Chauhan, K. Dhara, A.K. Tiwari, J. Dash, An uncatalyzed aldol reaction of
thiazolidinedione, Green. Chem., 14 (2012) 2990–2995.

[37] Y. Hayashi, In water or in the presence of water?, Angew. Chem. Int. Ed., 45 (2006) 8103 –
8104.
[38] A.P. Brogan, T.J. Dickerson, K.D. Janda, Enamine-based aldol organocatalysis in water: Are
they really “all wet”?, Angew. Chem. Int. Ed., 48 (2006) 8100 – 8102.
[39] A. Bassan, W. Zou, E. Reyes, F. Himo, A. Córdova, The origin of stereoselectivity in primary
amino acid catalyzed intermolecular aldol reactions, Angew. Chem. Int. Ed., 44 (2005) 7028 –
7032.
[40] M. Amedjkouh, Aqua-organocatalyzed direct asymmetric aldol reaction with acyclic amino
acids and organic bases with control of diastereo- and enantioselectivity, Tetrahedron:
Asymmetry, 18 (2007) 390–395.
[41] X.-H. Chen, J. Yu, L.-Z. Gong, The role of double hydrogen bonds in asymmetric direct aldol
reactions catalyzed by amino amide derivatives, Chem. Commun., 46 (2010) 6437–6448.
[42] M. Kakihana, T. Nagumo, M. Okamoto, H. Kakihana, Coordlnatlon structures for uranyl
carboxylate complexes in aqueous solution studied by IR and 13C NMR Spectra J. Phys. Chem. B,
91 (1987) 6128-6136.
[43] P. Zhou, S. Luo, J.-P. Cheng, Highly enantioselective synthesis of syn-aldols of
cyclohexanones via chiral primary amine catalyzed asymmetric transfer aldol reactions in ionic
liquid, Org. Biomol. Chem., 9 (2011) 1784–1790.
[44] B. Rodríguez, A. Bruchmann, C. Bolm, A highly eficient asymmetric organocatalytic aldol
reaction in a ball Chem. Eur. J., 13 (2007) 4710 – 4722.
[45] L. Li, S. Gou, F. Liu, Highly stereoselective anti-aldol reactions catalyzed by simple chiral
diamines and their unique application in configuration switch of aldol products, Tetrahedron
Lett., 54 (2013) 6358–6362.

Scheme 1. Schematic representation of the LL/HT materials obtained by the different protocols. L-Leu is
immobilized in both zwitterion and anionic form in method A2 and only in anionic form in method R1.

Scheme 2. Possible mechanism for the syn-anti isomerization

Scheme 3. Proposed enamine-type mechanism for the aldol reaction catalysed by LL/HTrus (both R1 and A2)
catalysts.

Figure 1. HRTEM images of the nanohybrids synthesized by anionic-exchange method: a) HTrus, b) LL/HTrus-A1
and c) LL/HTrus-A2.

Figure 2 HRTEM images of the nanohybrids synthesized by the reconstruction method: a) LL/HTrus-R1 and b)
LL/HTr-R2.

Figure 3. Skeletal FT-IR (up) and Raman (bottom) spectra of a) LL/HTClus-A1, b) LL/HTClus-A2.