Design of a heterogeneous catalytic process for the continuous and direct synthesis of lactide from lactic acid

Article abstract of DOI:10.1039/c6gc02405j

We present a continuous one-step reaction pathway for optically pure lactide under atmospheric conditions based on a novel SnO₂-SiO₂ nanocomposite catalyst. The new heterogeneous catalytic system gave a record high lactide yield of 94% with almost 100% enantioselectivity and long-term stability (>2500 h) from l-lactic acid.

Full text of DOI:10.1039/c6gc02405j

View Article Online
View Journal
Green
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: P. Upare, J. W.
Yoon, D. Hwang, U. Lee, Y. K. Hwang, D. Hong, J. C. Kim, J. H. Lee, S. K. Kwak, H. Shin, H. Kim and J.
Chang, Green Chem., 2016, DOI: 10.1039/C6GC02405J.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/greenchem

Page 1 of 8
Green Chemistry
View Article Online
DOI: 10.1039/C6GC02405J
Green Chemistry
RSCPublishing
COMMUNICATION
Cite this: DOI: 10.1039/x0xx00000x
Design of a heterogeneous catalytic process for
continuous and direct synthesis of lactide from lactic
acid
Received 00th August 2016,
Accepted 00th September 2016
DOI: 10.1039/x0xx00000x
www.rsc.org/
Pravin P. Upare[a]‡, Ji Woong Yoon[a]‡, Dong Won Hwang[a,b]*, UꢀHwang
Lee[a,b], Young Kyu Hwang[a,b], DoꢀYoung Hong[a,b], Jin Chul Kim[c], Jeong
Hyeon Lee[c], Sang Kyu Kwak[c], Hyeyoung Shin[d], Hyungjun Kim[d], Jongꢀ
San Chang[a,e]
*
We present a continuous one-step reaction pathway for optically
Recently, Dusselier et al. reported a new process in which LT
pure lactide under atmospheric condition based on a novel could be synthesized with 85% yield from aqueous LA in a batch
SnO2–SiO2 nanocomposite catalyst. The new heterogeneous reactor using Hꢀbeta zeolite and toluene15,16. They also proposed that
catalytic system gave a record high lactide yield of 94% with LT (purity > 98%) could be separated from a mixture of toluene,
almost 100% enantioselectivity and long-term stability (>2,500 LA, LT, oligomer, and zeolite catalyst by successive filtration,
h) from
L
-lactic acid.
extraction with water, and distillation. However, considering that the
LT yield in the commercialized prepolymer process is higher than
Today, owing to concerns about the shortage of fossil fuel resources 90%10, the LT yield obtained using this catalytic reaction is not
and climate change, much attention has been paid to finding satisfactory, despite using the toxic solvent toluene17. In addition, the
substitutes for petroleumꢀbased plastics1ꢀ3 and fuels4,5. Polylactic zeolite catalyst must be reactivated after the reaction to remove LA
acid (PLA) is regarded as a representative bioꢀbased biodegradable and the oligomer before reuse.
polymer that has many potential applications in packaging, fibres,
In the early 1990s, DuPont and Biopak patented a oneꢀstep
coatings, and biomedical fields6,7. The global PLA market is continuous process for LT synthesis from LA. In this process,
expected to grow to one million tonnes per year by 20208.
aqueous LA is first converted into vapour, which is passed through a
Industrially, PLA with a molecular weight (Mw) greater than 105 fixedꢀbed reactor with an Al2O3 or SiO2/Al2O3 catalyst to produce
g molꢀ1 is made from lactide (LT), a sixꢀmembered dimeric cyclic LT18,19. Compared with the prepolymer route, this simple reaction
ester of (usually
polymerization technique6,9. In this method, LT is produced using a without formation of the oligomer intermediate, and the entire
soꢀcalled a prepolymer process. In this process, ꢀLA is first process can be operated at atmospheric pressure. Unfortunately, this
polymerized at ~180 °C to form ꢀLA oligomer (Mw < 3000), and technology has not been commercialized as it suffers from low LT
then depolymerized at ~210 °C to give ꢀLT via a backꢀbiting yield (<40%) owing to poor activity of the conventional acidꢀtype
mechanism using a Snꢀbased catalyst, such as SnCl210ꢀ12 (red path in catalysts. Hence, it is necessary to develop a more efficient catalytic
Fig. 1). However, it is difficult to produce optically pure ꢀLT with system for this technology to be comparable with the conventional
high yields using this prepolymer route because it requires a long prepolymer process.
It should be noted that LA selfꢀoligomerizes in the absence of an
Lꢀ) lactic acid (LA), using the ringꢀopening pathway appears favourable because LA is converted directly to LT
L
L
L
L
residence time (>10 h) at a high reaction temperature and the
chemical stability of
ꢀLT is very low13. In addition, the whole
process should be conducted under high vacuum to separate ꢀLT
rapidly from the reaction system; otherwise, the produced ꢀLT is
degraded easily to ꢀLA oligomer or mesoꢀLT (MꢀLT) through
reactions with impurities, such as water, LA and catalyst10,14
Another drawback of this process is the production of heavy
oligomeric waste (Mw > 5000) that contains Sn metal, which is
toxic. Sn is very difficult to recover after the reaction because it is
dissolved completely in the oligomer.
acid catalyst because a condensation reaction between the carboxyl
and alcohol groups can proceed with ease20ꢀ23. LA is first condensed
to LA dimer (L2A) by dehydration (Fig. 2a). L2A can then be
converted to LT by dehydration; however, most L2A gives LA trimer
(L3A) via a second condensation reaction, especially at long
residence times. Thus, to increase LT selectivity, a short residence
time is desirable, but a loss of LA conversion is inevitable at this
reaction condition. In addition, because LT is transformed easily to
L3A when exposed to LA or MꢀLT at high temperatures, LT should
be desorbed rapidly from the reaction system to prevent its
degradation. Considering this reaction mechanism for LA
dehydration, it would not be easy to find a reaction system that
L
L
L
L
.
This journal is © The Royal Society of Chemistry 2013
J. Name., 2013, 00, 1-3 | 1

Green Chemistry
Page 2 of 8
COMMUNICATION
VJieowuArrtnicalelONnlainme e
DOI: 10.1039/C6GC02405J
91%. The lower surface area of SnO2(90)ꢀSiO2 (134 m2/g) might be
a main reason of the lower catalytic activity than SnO2(80)ꢀSiO2
(180 m2/g). Interestingly, compared with ZSMꢀ5, crystalline SnO2
satisfies both high LA conversion and high LT selectivity without an
effective catalyst. Here, we report novel heterogeneous catalysis for
continuous and direct synthesis of optically pure LT using a SnO2–
SiO2 nanocomposite catalyst under atmospheric conditions (blue
path in Fig. 1).
First, we investigated the catalytic activities of conventional
acidic metal oxides as heterogeneous catalysts for oneꢀstep
continuous LT synthesis. The LA dehydration reaction was
also gave a good LA conversion of 80% and
LꢀLT yield of 69%.
The product distribution determined
using liquid
chromatography clearly illustrates that the SnO2(80)–SiO2
nanocomposite (SSOꢀ80) produces LT selectively, even at high LA
conversion, whereas the production of LT by ZSMꢀ5 (Si/Al = 23) is
not selective, with a huge amount of oligomers formed (Fig. 4a). The
selective formation of LT on SSOꢀ80 was also confirmed by
OperandoꢀIR spectroscopy (Fig. 4b). The LT concentration on the
SSOꢀ80 surface was highest when the cell temperature was
maintained at 250 °C (Fig. S6). However, SiO2, which had very low
LA conversion, only showed a peak corresponding to LA. ZSMꢀ5,
which had high LA conversion and low LT yield, showed peaks
corresponding to LA oligomer and a negligible LA peak.
performed in a fixedꢀbed reactor with continuous flow of aqueous Lꢀ
LA (75 wt%). Typically, the reaction temperature was maintained at
240 °C with a weight hourly space velocity (WHSV) of 1.0 hꢀ1 under
N2 carrier gas (Fig. S1). The catalytic results for various catalysts are
summarized in Fig. 2b (for analysis, see Fig.S2, S3, Tables S1 and
S2). SiO2 gave a very low LꢀLT yield (13%) at 27% LA conversion,
which was almost the same as the result of the blank test (no
catalyst). ZSMꢀ5 (Si/Al = 23) zeolite gave a much higher LA
conversion (67%), but a
LꢀLT yield of only 25% was obtained
To determine the origin of the high catalytic activity of SSOꢀ80,
we investigated its physicochemical properties using various
techniques. Most Sn was converted to a tetragonal SnO2 phase after
calcination in all SnO2–SiO2 composites with different Sn loadings,
as confirmed by Xꢀray diffraction (XRD) patterns (Fig. 4c). The Sn
oxidation state of SSOꢀ80 was identified as Sn4+ by Sn 3d Xꢀray
photoelectron spectroscopy (XPS) (Fig. 4d). From the XRD and
XPS results, it is clear that tetragonal SnO2 species are highly
dispersed on the SiO2 surface, even for loadings as high as 80
wt%26,27. The IR spectra of pyridine chemisorption (Fig. 4e) and
NH3 desorption experiments (Fig. 4f) confirmed that weak Lewis
acid sites are dominant on SSOꢀ80, while Brønsted acid sites and
strong Lewis acid sites are dominant on ZSMꢀ5.
because of the high oligomer yield (39%). When the Si/Al ratio of
ZSMꢀ5 was increased to 50, the LA conversion was not much
changed, but the
when the Si/Al ratio of ZSMꢀ5 was further increased to 280, the
LT yield decreased considerably to 29%, although the ꢀLT
LꢀLT yield was greatly improved to 59%. However,
Lꢀ
L
selectivity was not much affected. These results indicate that the
acidity of the catalyst has a positive effect on LA conversion owing
to dehydration of LA, but a negative effect on
owing to successive oligomerization of L2A and
L
ꢀLT selectivity
ꢀLT. We also
L
investigated beta zeolite (Si/Al = 25), a shape selective zeolite
catalyst15, in a liquidꢀphase reaction, which showed similar catalytic
activity to ZSMꢀ5 (Si/Al = 280). Therefore, there is a limitation on
the yield of
LꢀLT obtained from LꢀLA using Brønsted acidic
The morphology and particle size distribution of SSOꢀ80 was
analysed using scanning electron microscopy (SEM) (Fig. S7), and
uniform particles with sizes smaller than 100 nm were observed. The
distribution and size of the SnO2 particles were measured using
transmission electron microscopy (TEM) (Fig. S8). Most of the
SnO2 nanoparticles were spherical in shape without mutual
aggregation. The nanoparticles were monodisperse and the average
particle size was 2 nm, which corresponds to the crystallite size
calculated from the XRD patterns. In highꢀresolution TEM images,
uniform fringes with an interval of 3.34 Å, corresponding to the (110)
lattice spacing of the cassiterite phase, were observed over the entire
region of the SnO2 nanoparticle.
The BET surface areas of the SnO2–SiO2 nanocomposites were
much higher than that of the crystalline SnO2 (Table S5). It is
thought that the good dispersion of SnO2 nanoparticles on the SiO2
surface is responsible for the high surface area, even at high SnO2
loading. The nitrogen physisorption isotherms of SSOꢀ80 showed
characteristic type IV shapes, indicating that SSOꢀ80 has
mesoporosity (Fig. S9).
Based on this structural information, we performed density
functional theory (DFT) calculations to investigate possible reaction
paths converting LA to L2A or LT over SnO2 surface. First,
consistent with the GCMC results, the DFT results further support
the strong interaction between LA and the SnO2 surface (ꢁGB.LA = ꢀ
0.67 eV) owing to both the hydrogen bond between the surface
oxygen of SnO2 and the hydrogen of the LA carboxylic group, and
the Lewis acidꢀbase interaction between surface Sn4+ of SnO2 and
O=C of the LA carboxylic group. For intermolecular dehydration,
the OH group of another LA molecule is required to be adjacent to
catalysts due to unselective acidꢀcatalysed condensation of LA to
oligomer.
To overcome the limitations of conventional Brønsted acidicꢀ
type materials, we designed novel SnO2ꢀbased nanocomposites as
heterogeneous catalysts because LA adsorption and LT desorption
by SnO2 is more favourable than that by TiO2 and Al2O3, according
to our thermodynamic calculation (Fig. 3). Using grand canonical
Monte Carlo (GCMC) calculations, we found that hydrogen bonding
of LA to SnO2 was more abundant than that to TiO2 and Al2O3, while
hydrogen bonding of LT to SnO2 was not observed (Fig. S4 and Fig.
S5). The higher adsorption energy of LA for SnO2 resulted from a
strong Coulombic interaction between LA and the bridging oxygen.
This particular electrostatic interaction caused adsorption of LA on
SnO2 to be favoured, regardless of the weak van der Waals
interaction. However, when LA was transformed into LT, this
characteristic interaction disappeared and a slightly smaller total
adsorption energy was observed for SnO2 than for the other
substrates (Table S3). We prepared SnO2–SiO2 nanocomposites by
simple precipitation–deposition of aqueous SnCl4 on colloidal SiO2
particles in NaOH, similar to our previously developed method24,25
.
The Sn content, confirmed by inductively coupled plasma
spectrometry, was almost identical to the theoretical loading (Table
S4).
Increasing the SnO2 loading from 40 to 80 wt% in the SnO2–
SiO2 nanocomposite resulted in an increase of the LA conversion
from 79% to 94%, and the
LꢀLT yield was also increased from 63%
to 93%. However, further increasing the SnO2 loading to 90%
resulted in decrease of the LA conversion to 93% and LꢀLT yield to
2 | J. Name., 2013, 00, 1-3
This journal is © The Royal Society of Chemistry 2013

Page 3 of 8
Journal Name
Green Chemistry
View Article Online
COMMUNICATION
DOI: 10.1039/C6GC02405J
dGraduate School of EEWS, Korea Advanced Institute of Science and
Technology (KAIST), 291 Daehakꢀro, Yuseong, Daejeon 305ꢀ701, Republic
of Korea
the COOH group of the first LA molecule adsorbed on the SnO2
surface. To retain this local atomistic arrangement, two distinct
reaction paths were proposed based on DFT energetics, as shown in
Fig. 5. The L2A formation path (blue in Fig. 5) proceeds via a linear
eDepartment of Chemistry, Sungkyunkwan University, Suwon 440ꢀ476,
Republic of Korea
*
configuration of (LA)2 , which is thermodynamically more
Author Contributions
favourable by 0.26 eV, but the second dehydration required for L2A
formation proceeds via a highly unstable conformation of L2A on the
SnO2 surface (by losing a key COOH–SnO2 interaction), and is
thereby kinetically unfavourable. In contract, the LT formation path
‡These authors contributed equally.
Electronic Supplementary Information (ESI) available: [details of any
supplementary information available should be included here]. See
DOI: 10.1039/c000000x/
*
(red in Fig. 5) proceeds via a cyclic configuration of (LA)2 , where
the COOH and OH groups of both LA molecules are positioned in
opposite directions, which is thermodynamically less preferred
during the initial stage, but involves no significantly unstable states.
For continuous heterogeneous catalysis, the catalyst stability is
vital, together with high activity. It is notable that high LA
1.C. O. Tuck, E. Perez, I. T. Horvath, R. A. Sheldon and M. Poliakoff,
Science, 2012, 337, 695ꢀ699.
2.D. Saygin, D. J. Gielen, M. Draeck, E. Worrell and M. K. Patel, Renew.
Sustain. Energy Rev. 2014, 40, 1153ꢀ1167.
3.A. Corma, S. Iborra and A. Velty, Chemical Rev., 2007, 107, 2411ꢀ2502.
4.Y. RománꢀLeshkov, C. J. Barrett, Z. Y. Liu and J. A. Dumesic, Nature,
2007, 447, 982ꢀ985.
5.T. P. Vispute, H. Zhang, A. Sanna, R. Xiao and G. W. Huber, Science,
2010, 330, 1222ꢀ1227.
6.S. Inkinen, M. Hakkarainen, A.ꢀC. Albertsson and A. Södergård,
Biomacromol., 2011, 12, 523ꢀ532.
7.N. Saito, T. Okada, H. Horiuchi, N. Murakami, J. Takahashi, M. Nawata,
H. Ota, K. Nozaki and K. Takaoka, Nature Biotechnol., 2001, 19, 332ꢀ335.
8.L. Shen, E. Worrell and M. Patel, Biofuel, Bioprod. & Bioref., 2010, 4, 25ꢀ
40.
9.E. T. Vink, S. Davies and J. J. Kolstad, Ind. Biotechnol., 2010, 6, 212ꢀ224.
10. P. R. Gruber, E. S. Hall, J. J. Kolstad, M. L. Iwen, R. D. Benson and R.
L. Borchardt, US Patent, 1993, 5247059.
11. D. K. Yoo, D. Kim and D. S. Lee, Macromolecular research, 2006, 14,
510ꢀ516.
conversion, as well as high LꢀLT selectivity, was maintained over
SSOꢀ80 at the optimized reaction conditions for more than 2,500 h
without a notable loss in activity (Fig. S10). The XRD pattern of the
used catalyst was very similar to that of the fresh catalyst, with a
tetragonal SnO2 phase present in both XRD patterns. The TEM
analysis confirmed that the used catalyst had wellꢀdispersed SnO2
nanoparticles with sizes smaller than 5 nm, similar to the fresh
catalyst (Fig. S11). The TGA analysis showed almost identical
profiles for the fresh and used catalyst, except below 100 °C, which
indicates that coke deposition is negligible over the SSOꢀ80 catalyst
(Fig. S12). In addition, no decomposition products, such as CO and
CO2, were detected, as confirmed by analysis of the gaseous
products using a TCD detector.
After the catalytic reaction using SSOꢀ80, the crude LT crystal
12. M. Dusselier, P. Van Wouwe, A. Dewaele, E. Makshina and B. F. Sels,
Energy & Environ. Sci., 2013, 6, 1415ꢀ1442.
1
was separated from the aqueous solution by filtration. The H NMR
13. T. Tsukegi, T. Motoyama, Y. Shirai, H. Nishida and T. Endo, Polymer
Degrad. Stabil., 2007, 92, 552ꢀ559.
14. P. P. Upare, Y. K. Hwang, J.ꢀS. Chang and D. W. Hwang, Ind. & Eng.
Chem. Res., 2012, 51, 4837ꢀ 4842.
15. M. Dusselier, P. Van Wouwe, A. Dewaele, P. A. Jacobs and B. F. Sels,
Science, 2015, 349, 78ꢀ80.
spectrum of the crude LT sample had two strong characteristic peaks
at 1.68 and 5.05 ppm, corresponding to the methyl (–CH3) and
methine (–CH) groups, respectively, together with weak LA and
H2Oꢀrelated peaks (Fig. S13).
16. P. Van Wouwe, M. Dusselier, E. Vanleeuw and B. Sels, ChemSusChem,
2016, 9, 907ꢀ921.
Conclusions
17. K. Alfonsi, J. Colberg, P. J. Dunn, T. Fevig, S. Jennings, T. A. Johnson,
H. P. Kleine, C. Knight, M. A. Nagy, D. A. Perry and M. Stefaniak, Green
Chemistry, 2008, 10, 31ꢀ36.
We designed a novel heterogeneous catalyst for continuous and
direct synthesis of optically pure LꢀLT from LꢀLA under atmospheric
18. H. E. Bellis and K. K. Bhatia, US Patent, 1992, 5138074.
19. H. P. Benecke, R. A. Markle and R. G. Sinclair, US Patent, 1994,
5332839
20. S. I. Moon and Y. Kimura, Polymer Int., 2003, 52, 299ꢀ303.
21. Y. M. Harshe, G. Storti, M. Morbidelli, S. Gelosa and D. Moscatelli,
Macromolecular Reaction Engineering, 2007, 1, 611ꢀ621.
22. T. Trimaille, M. Möller and R. Gurny, J. Polym. Sci. A: Polym. Chem.,
2004, 42, 4379ꢀ4391.
conditions based on SnO2–SnO2 nanocomposites with the average
crystallite size of 2 nm, based on GCMC and DFT calculations. The
new catalytic system gave a record high LT yield of 94% with
almost 100% enantioselectivity and excellent longꢀterm stability
(>2,500 h) at a WHSV of 1.0 hꢀ1. By adopting this green process,
optically pure lactide can be produced easily from aqueous lactic
acid with a high yield and no waste.
23. A. P. Dove, V. C. Gibson, E. L. Marshall, A. J. White and D. J. Williams,
Chem. Commun., 2001, 283ꢀ284.
24. D. W. Hwang, P. Kashinathan, J. M. Lee, J. H. Lee, U.ꢀh. Lee, J.ꢀS.
Hwang, Y. K. Hwang and J.ꢀS. Chang, Green Chem., 2011, 13, 1672ꢀ1675.
25. P. P. Upare, J. M. Lee, Y. K. Hwang, D. W. Hwang, J. H. Lee, S. B.
Halligudi, J. S. Hwang and J. S. Chang, ChemSusChem, 2011, 4, 1749ꢀ
1752.
Acknowledgements
This work was supported by the R&D Program of Institutional Research
Program of KRICT [KKꢀ1601ꢀC00].
Notes and references
26. H. Deng, F. J. Lamelas and J. M. Hossenlopp, Chem. Mat., 2003, 2429ꢀ
2436.
aGreen Carbon Catalysis Research Group, Korea Research Institute of
Chemical Technology (KRICT), 141 Gajeongro, Yuseoung, Daejeon 305– 27. H. Ma, K. Teng, Y. Fu, Y. Song, Y. Wang and X. Dong, Energy &
Environ. Sci., 2011, 4, 3067ꢀ3074.
bDepartment of Green Chemistry & Biotechnology, University of Science 28. C. D. A. C. Erbetta, R. J. Alves, J. M. Resende, R. F. d. S. Freitas and R.
600, Republic of Korea; Email: dwhwang@krict.re.kr, jschang@krict.re.kr
and Technology (UST), 113 Gwahangno, Yuseong, Daejeon 305–333,
Republic of Korea
G. d. Sousa, J. Biomat. Nanobiotech., 2012, 3, 208ꢀ225.
cSchool of Energy and Chemical Engineering, Ulsan National Institute of
Science and Technology (UNIST), 50 UNISTꢀgil, Uljuꢀgun, Ulsan 689ꢀ798,
Republic of Korea
This journal is © The Royal Society of Chemistry 2013
J. Name., 2016, 00, 1-3 | 3

Green Chemistry
Page 4 of 8
View Article Online
DOI: 10.1039/C6GC02405J
Figure 1. Lactide production routes from biomass-derived lactic acid. Conventional two-step
batch homogeneous catalysis under vacuum (red) vs. one-step continuous heterogeneous
catalysis under atmospheric pressure (blue).

Page 5 of 8
Green Chemistry
View Article Online
DOI: 10.1039/C6GC02405J
Figure 2. Reaction pathways and catalyst performance. a) Reaction pathways for lactic acid
dehydration to lactide via lactic acid dimer. b) Comparison of catalytic activity over various
metal oxide catalysts (reaction conditions: 101.3 kPa, 240 °C, WHSV: 1.0 h-1, N2 flow rate:
250 mL min-1, time-on-stream: 50 h).

Green Chemistry
Page 6 of 8
View Article Online
DOI: 10.1039/C6GC02405J
Figure 3. Adsorption isotherms with respect to fugacity at 544 K for L-lactic acid (a) and L-
lactide (b). The confidence limits of the data are smaller than the symbols. The number of
adsorbed molecules was divided by the surface area of each catalyst, and then multiplied by a
factor, i.e. 1000, which is similar to the area of the catalysts, for the clear view

Page 7 of 8
Green Chemistry
View Article Online
DOI: 10.1039/C6GC02405J
Figure 4. Product distributions with catalyst and characterization of SnO2–SiO2
nanocomposites. a) HPLC product profiles. b) Operando-IR spectra. c) XRD patterns. d) Sn
3d XPS patterns. e) FTIR spectra of pyridine chemisorption (pretreatment at 400 °C for 2 h;
adsorption at 20 °C; desorption at 100 °C for 1 h). f) NH3 desorption spectra (pretreatment at
600 °C for 2 h; adsorption at 100 °C).

Green Chemistry
Page 8 of 8
View Article Online
DOI: 10.1039/C6GC02405J
Figure 5. DFT calculated free energy changes of the reaction paths for L2A formation (blue)
and LT formation (red), where green, red, grey, and white balls represent Sn, O, C, and H
atoms, respectively, and black dashed lines represent hydrogen bonds