Monolithic Silica Support for Immobilized Catalysis in Continuous Flow

Article abstract of DOI:10.1002/adsc.201901185

Monolithic and packed-bed reactors featuring immobilized catalysts are well-precedented in continuous flow synthesis but can suffer from adverse pressure drops during use due to their small pore sizes and/or structural changes. Herein, we overcome this challenge with the synthesis of a structurally robust silica-based monolith featuring pore sizes on the millimeter scale. The 3-dimensional solid support structure is constructed from a polystyrene foam-based template and features a functional group handle that can be modified to display a reactive catalyst. Here we functionalize the support with palladium(0) for hydrogenation reactions and a modified proline catalyst for the alpha functionalization of aldehydes. Both reactors showed good activity and excellent catalytic longevity when utilized under continuous flow conditions. (Figure presented.).

Full text of DOI:10.1002/adsc.201901185

Title: Monolithic Silica Support for Immobilized Catalysis in Continuous
Flow
Authors: M. Grace Russell, Cedrick Veryser, James Hunter, Rachel
Beingessner, and Timothy Jamison
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Adv. Synth. Catal. 10.1002/adsc.201901185
Link to VoR: http://dx.doi.org/10.1002/adsc.201901185

10.1002/adsc.201901185
Advanced Synthesis & Catalysis
COMMUNICATION
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Monolithic Silica Support for Immobilized Catalysis in
Continuous Flow
M. Grace Russell,a Cedrick Veryser,b James F. Hunter,c Rachel L. Beingessner,a and
Timothy F. Jamisona*
a
Department of Chemistry, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA, 02139,
USA
Email: tfj@mit.edu
b
Molecular Design and Synthesis, Department of Chemistry, KU Leuven, Celestijnenlaan 200F, Box 2404, 3001
Leuven, Belgium
c
Department of Material Science and Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue,
Cambridge, MA, 02139, USA
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.
of such monoliths offers many advantages[1]
including
Abstract. Monolithic and packed-bed reactors featuring
immobilized catalysts are well-precedented in continuous
flow synthesis but can suffer from adverse pressure drops
during use due to their small pore sizes and/or structural
changes. Herein, we overcome this challenge with the
synthesis of a structurally robust silica-based monolith
featuring pore sizes on the millimeter scale. The 3-
dimensional solid support structure is constructed from a
polystyrene foam-based template and features a functional
group handle that can be modified to display a reactive
catalyst. Here we functionalize the support with
palladium(0) for hydrogenation reactions and a modified
proline catalyst for the alpha functionalization of
aldehydes. Both reactors showed good activity and
excellent catalytic longevity when utilized under
continuous flow conditions.
Keywords: continuous flow; heterogeneous catalysis;
hydrogenation; immobilization; organocatalysis
The use of immobilized catalysts[1-5] in continuous
flow synthesis is well-precedented in the literature,
but not without its challenges. For example,
functionalized beads within a traditional packed-bed
reactor (Figure 1A) are randomly packed and there
can be movement and changes to these beads during
use. This can reduce the catalytic surface area in
contact with the substrate solution, lead to poor mass
transfer properties, restrict solution flow and cause an
adverse pressure drop throughout the reactor.
Channels may form, leading to rate differences across
the bed.[1]
As an alternative to packed-bed reactors,
researchers have prepared monolithic solids via
polymerization processes to support various catalysts
(Figure 1B).[1,2,6-9] In many cases, these materials
have pore sizes in the range of 1 µm - 100 µm and are
comprised of inorganic polymers such as silica, or
organic polymers such as polystyrene. While the use
Figure 1. (A) Packed-bed comprised of functionalized
beads. (B) Monolith support resulting from
a
polymerization reaction with the catalytically active
[2,6–9]
material inside a reactor.
(C) Monolith support
derived from a 3-dimensional printed structure plated with
metal zero catalysts.[10,11] (D) In this work, a silica-based
support that can be functionalized post synthesis is
constructed from a removable 3-dimensional scaffold.
a large surface area of catalyst in contact with the
substrate fluid, they can be characterized with having
high flow resistance due to polymer swelling and/or
small pore sizes. Recently, a 3-dimensional-printed
monolith was reported that is coated with ground
state metal (Figure 1C).[10,11] Although a minimal
pressure drop was observed while using this monolith,
the method is only applicable for ground state metal
catalysis.
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901185
Advanced Synthesis & Catalysis
(A)
X
X
OH Me
O Si O Si O Si
OEt
+
Me
HCl (aq)
+
O
EtO Si OEt
OEt
EtO Si OEt
OEt
EtO Si OEt
OEt
then curing (temperature, time)
O
O
O
1, 1 equiv
2, 2.4 equiv
3a, X = I
4a, X = I
3b, X = NH2
0.3 equiv
4b, X = NH2
O
(B)
HN
H2N
O
5
OH Me
OH Me
O Si O Si O Si
OH
O Si O Si O Si
O
O
EDC, DMAP
DCM, RT
O
O
O
O
O
O
6
4b
NH2
NH2
O
I
HN
HN
Pd0
(C)
PdCl2, DMF
H2N
NH2
OH Me
O Si O Si O Si
OH Me
O Si O Si O Si
OH Me
80 °C
7
O
O Si O Si O Si
O
Xylene, K2CO3
O
O
O
O
O
O
O
O
O
80 °C
4a
8
Catalyst I
Ar
Ar
(D)
Ar
Ar
N
O
Boc
H2N
H2N
N
O
OH
10
H
O
H
1. EDC, DMAP
DCM, RT
N
OH Me
O Si O Si O Si
Me
TMSO
O Si O Si O Si
TMSOTf, Bu3N
O
O
O
DCM, 0 °C
Me
TMSO
O Si O Si O Si
O
O
O
2. 5:1 DCM:TFA
CF3
O
O
O
O
O
O
O
4b
9
Ar =
CF3
Catalyst II
Figure 2. (A) Synthesis of 4a and 4b using tetraethyl orthosilicate (TEOS, 1), methyl triethoxysilane (MTES, 2) and 3a or
3b. (B) Synthesis of 6 from the silica-based support 4b. (C) Synthesis of Catalyst I from the silica-based support 4a. (D)
Synthesis of Catalyst II reactor from 4b. In this case, support 4b was constructed using polystyrene foam beads 0.4 - 0.8
mm in diameter and modified equivalents of 2 and 3b (see SI for details).
Herein, we envisioned the design of a silica-based
monolithic support that features pore sizes on the
millimeter scale (Figure 1D). By constructing the
solid
support to have larger pore sizes than mentioned
above,[2,6–9] we hypothesized that the reactor would
have a limited pressure drop during operation. The
use of silica for the support structure offers the
benefit of resistance to swelling in many organic
solvents and tolerance to acidic reaction conditions
and oxidative environments. A silica surface can also
be readily functionalized using well-precedented
chemistry.
feature a handle for further functionalization.
Following curing and removal of the beads, the 3-
dimensional monolithic support would be provided
that could be modified on the surface to generate the
desired final catalyst system.
Since known procedures to synthesize silicas
generally focus on making finely controlled
mesoporous particles or thin films,[12-15] we focused
our initial efforts on developing reaction conditions to
prepare a simple solid structure. We began by
investigating the copolymerization of tetraethyl
orthosilicate (TEOS (1), Figure 2A) and methyl
triethoxysilane (MTES (2), Figure 2A) at varying
molar ratios in an attempt to form a solid material
with minimal fractures through visual inspection.
Such flaws formed during the curing process would
render the resulting structure more prone to collapse
under continuous flow conditions. In general,
Our approach to the design of the silica-based
monolithic support was to fill an empty stainless steel
reactor with commerically available polymer beads
(diameter of ~ 1 mm) prior to the addition of
hydrolyzed silane monomers, including those that
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901185
Advanced Synthesis & Catalysis
conducting the reaction under acidic aqueous
conditions (HCl) led to the formation of a solid
material, whereas utilization of basic aqueous
conditions (NH4OH) resulted in flaky particles.
Notably, the use of 2 - 5 equivalents of MTES (2)
relative to TEOS (1) slowed the rate of
polymerization to 24 hours or more, which greatly
reduced visible fractures within the resulting solid.[16]
As shown in Figure 2A, we next investigated the
copolymerization of 1 and 2 in the presence of silane
3 in order to enable surface functionalization after the
curing process. Both (3-iodopropyl) triethoxysilane
(3a) and (3-aminopropyl) triethoxysilane (APTES)
(3b) were investigated to allow for functionalization
via nucleophilic substitution and amidation reactions,
respectively. The resulting solids 4a and 4b that were
synthesized using the optimized equivalents of the
respective monomers (Figure 2A) displayed minimal
fractures and appeared similar to the solid produced
in the absence of 3 under otherwise identical reaction
conditions.
We subsequently constructed the 3-dimensional
structure of the monolithic supports 4a and 4b by
filling a stainless steel reactor (2 inches length, ¼
inch diameter) with polystyrene foam beads (1.0 - 1.5
mm diameter) prior to the addition of the aqueous
acidic solution of 1, 2 and 3a or 3b (Figure 2A). This
solution was added from the bottom of the reactor
until all of the air was displaced. Silica curing next
proceeded for 24 hours at room temperature to
prevent the beads from dissolving. After this initial
24 hour period, we observed greater bead stability
and thus the curing temperature was raised to 60 °C
for an additional 24 hours. The final phase of the
curing process was performed at 160 C for 2 hours
to ensure the removal of water and ethanol. Finally,
removal of the beads to provide the silica supports 4a
and 4b was accomplished using a simple acetone
wash. It is important to note that when hard
polyethylene beads were used instead of the soft
foam beads, numerous stress fractures were produced
during the curing process. Additionally, a high
temperature (500 °C overnight) was required to
remove the hard beads, which would have limited the
types of functional groups on 3 that could be used.
Overall, 4a and 4b had similar architectures from
batch to batch due to the fairly consistent packing
patterns of the relatively large foam beads. The
general structure of the support can be seen in the
optical image of 6 (Figure 2B) viewed end-on under
both light (Figure 3A) and UV-light (Figure 3B) as
well as along the length under UV-light (Figure 3C).
Monolith 6 was prepared from 4b via a 1-ethyl-3-(3-
dimethylaminopropyl)carbodiimide (EDC) mediated
coupling reaction with cyclopropane carboxylic acid
(5) (Figure 2B). Importantly, assessment of 4a and 4b
under continuous flow conditions revealed the
stability of the 3-dimensional structure and no
detectable pressure drop at flow rates as high as 1
mL/min.
Having successfully constructed the silica support,
we next prepared Catalyst I featuring Pd(0) for
hydrogenation reactions by first functionalizing 4a
with ethylenediamine (7) (Figure 2C). A solution of
PdCl2 in DMF was then passed over 8 at 80 °C for 48
hours, followed by washing with DMF (Figure 2B).
The surface of Catalyst I was visualized using SEM
imaging (Figure 3D-F) and revealed the deposition of
well-dispersed Pd(0) on the surface of the solid
support (Figure 3E). Elemental analysis was also
performed and indicated the presence of 0.045 mmol
of palladium within the 2-inch reactor.
The catalytic activity of Catalyst I was next
screened via
a
series of continuous flow
hydrogenation reactions (Figure 4). A 1:1 mixture of
ethanol:ethyl acetate was used to ensure hydrogen gas
and substrate
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901185
Advanced Synthesis & Catalysis
Figure 3. (A) Optical image of 6 (end view). Optical image of 6 under UV light (298 nm) viewed from the (B) end and (C)
along the length of the reactor. (D) SEM image of Catalyst I. (E and F) SEM image of Catalyst I with energy dispersive
x-ray analysis showing Pd(0) (green) and the silica-based monolithic support (blue), respectively.
Catalyst I
25 µL/min
Substrate (0.2 M)
in EtOH/EtOAc (1:1)
Product
11 - 20
BPR
120 psi
60 °C
Residence time: 1.77 min
Nitro Reduction
Imine Reduction
F
NH2
NH2
NH2
H2N
NH2
O
O
O
N
N
N
H
NC
Cl
O
11
12
13
98%a
14
74%a
15
93%a
16
83%a
88%
96%
Alkene Reduction
Alkyne Reduction
N
O
HO
O
OEt
O
O
HO
17
O
20
99%
18
19
O
N
97%
93%b
92%a,c
Figure 4. Scope of hydrogenation reactions using Catalyst I functionalized with Pd(0). Yields presented are isolated
yields. Collection time = 100 minutes or 0.5 mmol. aReaction performed at 100 °C. bReaction performed at 120 °C. c0.1 M
in EtOH/EtOAc (1:1). BPR = back pressure regulator.
solubility. Temperature was varied to optimize
selectivity and product yield. As summarized in
Figure 4, aryl nitro groups, alkenes, alkynes and
imines were successfully reduced to the
corresponding products 11 - 20 in 74% - 99% isolated
yield within a residence time of 1.77 min. [5,10,11,17,18]
Challenging nitro groups[19] which normally require
long reaction times and often result in low yields of
the amine product, were successfully reduced as
demonstrated with the formation of 14 in 74% yield.
While the reduction of compound 11 required a
temperature of 60 °C to suppress dehalogenation, a
higher temperature of 120 °C was necessary to
achieve full conversion of quinine to compound 18.
A longevity study was performed using the
Catalyst I reactor by monitoring the reduction of
eugenol (21) to 17 over a period of 113 hours (Figure
5).[20] In general, excellent and consistent yields of 17
As shown in Figure 2D, we also prepared Catalyst
II featuring a modified Jørgensen catalyst 10.[21,22] In
this case, support 4b was constructed to feature a
larger surface area via the use of smaller polystyrene
foam beads (0.4 - 0.8 mm in diameter) so that
sufficient catalyst would be covalently attached on
the surface of the reactor.[20] When an EDC mediated
amide coupling between 10 and 4b was initially
attempted, we observed rapid consumption of the
activated carboxylic acid by FlowIR and only 10% of
the expected conversion to the amide product. This
suggested the participation of OH groups and
possibly water molecules present on the surface of
the support.
1
were observed by H NMR spectroscopy with the
exception of four data points shown in grey during
hour ten (69% yield) to hour fifteen (90% yield).
During that time period, there was a small leak in the
system at the point of the back pressure regulator
(BPR, 120 psi) and thus not related to the silica-based
monolithic support. Notably, a single reactor was
utilized for this longevity study as well as for the
substrate scope shown in Figure 4. Between
substrates, Catalyst I was washed with 5 mL of the
reaction solvent (1:1 mixture of ethanol:ethyl acetate)
to prevent cross-contamination.
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901185
Advanced Synthesis & Catalysis
Figure 5. Continuous flow reduction of eugenol (21) to 17
using Catalyst I over 113 hours. Yields were measured
using 1H NMR spectroscopy with 1,3,5-trimethoxybenzene
as an external standard. The four data points showed in
grey were due to a leak in the system at the point of the
BPR (back pressure regulator).
isolated yields (40% - 63%) and diastereoselectivity
with excellent enantioselectivity as observed using
chiral HPLC. These results are generally comparable
to the yields (63% - 83%) and enantioselectivites
(94% - 98% ee) obtained of similar or identical
products under batch conditions as described in the
literature.[22] Notably, Catalyst II performed reliably
without any observable decrease in activity or
selectivity over more than 110 experiments or 170
mmol of substrate passed over the reactor.
In conclusion, a new and practical monolithic
silica-based support for the immobilization of
catalysts has been developed. We have demonstrated
the functionalization of the support with Pd(0) for
hydrogenation reactions as well as a modified proline
catalyst for the alpha functionalization of aldehydes.
Both reactors displayed good activity and longevity
under continuous flow conditions. Overall, these
preliminary results allude to the potential for
functionalizing the silica-based support with a broad
range of catalysts and this will be the focus of future
investigations within our laboratory.
In order to remedy this undesired reactivity, the
surface was treated with TMSOTf and tributylamine
at 0 °C prior to performing the amide coupling
reaction (Figure 2D). Finally, the active Catalyst II
was generated by passing a 5:1 mixture of
DCM:trifluoroacetic acid over the reactor for 2 hours
at room temperature followed by rinsing with DCM
and a 5:1 mixture of DCM and triethylamine.
Elemental analysis was performed and revealed the
presence of 0.0055 mmol of the modified Jørgensen
catalyst within the 2-inch reactor.
The catalytic activity of Catalyst II was examined
by performing a Mannich reaction using a preformed
imine and an aldehyde partner (Figure 6).[22,23] DMF
was determined to be the most suitable solvent in
combination with 20% N’N-diphenylthiourea as an
additive to activate the imine electrophile.[23] The
desired products 22 - 25 were obtained in modest
PMP
N
EtO
(1 M in DMFa)
O
O
Catalyst II
R
O
H
EtO
PMP
O
N
H
(3 equiv neat)
50 °C
R
H
O
Residence time: 26 min
22 - 25
O
O
O
H
H
H
EtO
PMP
EtO
O
PMP
H
N
H
EtO
PMP
N
H
N
EtO
PMP
O
H
N
H
O
24b
40%
15:1 dr, 98% ee
25
57%
4.6:1 dr, 98% ee
22
58%
3.5:1 dr, 98% ee
23
63%
O
2.3:1 dr, 98% ee
Figure 6. Scope of the alpha functionalization of aldehydes using Catalyst II. Collection = 0.5 mmol. Flow rate modified
a
to achieve residence time and stoichiometry. PMP = p-methoxyphenyl.
20% of N,N-diphenylthiourea added to this
stream. bC3 stereocenter is a racemic mixture.
foam beads (1.0 - 1.5 mm) loaded into folded weighing
paper with static removed were then tapped into the
stainless steel tube. The homogeneous silane solution was
added to the reactor from the bottom to push air bubbles
out. The open end of the reactor was then plugged with
folded parafilm, inverted and the fitting where the solution
was added was removed. After curing for several hours at
room temperature, the volume of material decreased and
thus additional silane solution was added to the top of the
reactor. After a total of 24 hours at room temperature, the
reactor was subjected to a temperature of 60 °C for 24
hours, followed by 160 °C for 2-3 hours. The beads were
Experimental Section
General Method for Reactor Synthesis
TEOS (1) (1 equiv), MTES (2) (2.4 equiv) and 3a or 3b
(0.3 equiv) were premixed and treated with 1M aqueous
HCl (0.25 mL) at room temperature. The biphasic solution
was stirred for approximately 5 minutes until clear and
homogeneous.
A fitting was placed on the bottom end of a stainless steel
tube (2 inches in length, ¼ inch in diameter). Polystyrene
5
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901185
Advanced Synthesis & Catalysis
then removed upon cooling to room temperature via
rinsing with 10 mL of acetone to provide 4a or 4b.
[14] K. S. Rao, K. El-Hami, T. Kodaki, K. Matsushige, K.
Makino, J. Colloid Interface Sci. 2005, 289, 125–131.
[15] I. A. Rahman, V. Padavettan, J. Nanomater. 2012,
2012, 1–15.
Acknowledgements
We thank Dr. Bruce Adams (MIT) for his assistance with NMR
characterization. This work was supported by the DARPA Make-
It program under contract ARO W911NF-16-2-0023. C.V. thanks
FWO (PhD Fellowship of the Research Foundation – Flanders)
for the fellowship received.
[16] E. J. Kappert, D. Pavlenko, J. Malzbender, A.
Nijmeijer, N. E. Benes, P. A. Tsai, Soft Matter 2015, 11,
882–888.
[17] A. K. Rathi, M. B. Gawande, V. Ranc, J. Pechousek,
M. Petr, K. Cepe, R. S. Varma, R. Zboril, Catal. Sci.
Technol. 2016, 6, 152–160.
References
[1] G. Jas, A. Kirschning, Chem. - A Eur. J. 2003, 9, 5708–
[18] C. J. Mallia, I. R. Baxendale, Org. Process Res. Dev.
2016, 20, 327–360.
5723.
[2] A. Tanimu, S. Jaenicke, K. Alhooshani, Chem. Eng. J.
2017, 327, 792–821.
[19] A. Padwa, M. Dimitroff, A. G. Waterson, T. Wu, J.
Org. Chem. 1997, 62, 4088–4096.
[3] R. Porta, M. Benaglia, A. Puglisi, Org. Process Res.
Dev. 2016, 20, 2–25.
[20] See Supporting Information for further details.
[21] A. B. Xia, C. Zhang, Y. P. Zhang, Y. J. Guo, X. L.
Zhang, Z. B. Li, D. Q. Xu, Org. Biomol. Chem. 2015,
13, 9593–9599.
[4] D. Cantillo, C. O. Kappe, ChemCatChem 2015, 6,
3286–3305.
[5] M. Irfan, T. N. Glasnov, C. O. Kappe, ChemSusChem
2011, 4, 300–316.
[22] J. Franzén, M. Marigo, D. Fielenbach, T. C. Wabnitz,
A. Kjærsgaard, K. A. Jørgensen, J. Am. Chem. Soc.
2005, 127, 18296–18304.
[6] M. I. Burguete, A. Cornejo, E. García-Verdugo, M. J.
Gil, S. V Luis, J. A. Mayoral, V. Martínez-Merino, M.
Sokolova, J. Org. Chem. 2007, 72, 4344–4350.
[23] S. Perera, D. Sinha, N. K. Rana, V. Trieu-Do, J. C.-G.
Zhao, J. Org. Chem. 2013, 78, 10947-10953.
[7] N. Hird, I. Hughes, D. Hunter, M. G. J. T. Morrison, D.
C. Sherrington, L. Stevenson, Tetrahedron 1999, 55,
9575–9584.
[8] E. C. Peters, F. Svec, J. M. J. Fréchet, Adv. Mater. 1999,
11, 1169–1181.
[9] A. Gömann, J. A. Deverell, K. F. Munting, R. C. Jones,
T. Rodemann, A. J. Canty, J. A. Smith, R. M. Guijt,
Tetrahedron 2009, 65, 1450–1454.
[10] C. Genet, X. Nguyen, B. Bayatsarmadi, M. D. Horne,
J. Gardiner, C. H. Hornung, J. Flow Chem. 2018, 8,
81–88.
[11] C. H. Hornung, X. Nguyen, A. Carafa, J. Gardiner, A.
Urban, D. Fraser, M. D. Horne, D. R. Gunasegaram, J.
Tsanaktsidis, Org. Process Res. Dev. 2017, 21, 1311–
1319.
[12] T. S. Glazneva, E. V. Rebrov, J. C. Schouten, E. A.
Paukshtis, Z. R. Ismagilov, Thin Solid Films 2007, 515,
6391–6394.
[13] J. E. Martin, M. T. Anderson, J. Odinek, P. Newcomer,
Langmuir 1997, 13, 4133–4141.
6
This article is protected by copyright. All rights reserved.

10.1002/adsc.201901185
Advanced Synthesis & Catalysis
COMMUNICATION
Silica-based monolith
functionalized
H2 (g)
with palladium (0)
Monolithic Silica Support for Immobilized
Catalysis in Continuous Flow
product
10 examples
(74% - 99%)
alkene or alkyne
or imine or nitro
back-pressure
regulator
in EtOH/EtOAc
Hydrogenation
Silica-based monolith
functionalized with a
Adv. Synth. Catal. Year, Volume, Page – Page
PMP
N
O
modified Jørgensen catalyst
EtO
R
O
H
in DMF
O
EtO
PMP
N
O
H
M. Grace Russell, Cedrick Veryser, James F.
Hunter, Rachel L. Beingessner and Timothy F.
Jamison*
H
Mannich Reaction
R
4 examples
(40% - 63%, 98% ee)
R = butyl, ethyl, benzyl,
Silica-based monolith
• Facile Synthesis
• Structurally Robust
• Millimeter-Scale Pore Size
• Limited Pressure Drop in Flow
• Readily Functionalized
• Long Catalytic Lifetime
7
This article is protected by copyright. All rights reserved.