Reaction pathways at the initial steps of trioxane polymerisation

Article abstract of DOI:10.1039/c8cy01691g

Cleavage and (re)formation of oxymethylene moieties are essential steps during the build-up of polyoxymethylene (POM), a technically relevant high-performance polymer. To reveal how the catalyst accomplishes the cleavage of oxymethylene moieties, the kinetics of the ring-opening of trioxane was studied. Thereby insights into the initial phase of the growth of oxymethylene chains were obtained. The chain length of short oligomers was controlled with acetic anhydride as transfer agent. With a high ratio of acetic anhydride to trioxane, the first homologues of the series, trioxymethylene diacetate, dioxymethylene diacetate and monooxymethylene diacetate were obtained in excellent yield. The homologues show distinct features in the NMR and IR spectra that were related to their reactivity during chain propagation. Formation of intermediate hemi-acetal oxonium moieties is suggested to be a key step in the reaction pathway. Such molecular level insight may be a crucial factor for further tailoring production and properties of polyoxymethylene as well as introducing oligomeric oxymethylene diacetates to new application fields.

Full text of DOI:10.1039/c8cy01691g

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: M. Hoffmann, C.
Bizzarri, W. Leitner and T. E. Müller, Catal. Sci. Technol., 2018, DOI: 10.1039/C8CY01691G.
This is an Accepted Manuscript, which has been through the
Volume 6 Number 1 7 January 2016 Pages 1–308
Royal Society of Chemistry peer review process and has been
accepted for publication.
Catalysis
Science &
Technology
Accepted Manuscripts are published online shortly after
www.rsc.org/catalysis
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
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
ISSN 2044-4753
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
PAPER
Qingzhu Zhang et al.
Catalytic mechanism of C–F bond cleavage: insights from QM/MM
analysis of fluoroacetate dehalogenase
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
rsc.li/catalysis

Please do not adjust margins
Catalysis Science & Technology
Page 1 of 10
View Article Online
DOI: 10.1039/C8CY01691G
Catal. Sci. Technol.
ARTICLE
Reaction Pathways at the Initial Steps of Trioxane Polymerisation
Matthias Hoffmann,a Claudia Bizzarri,a,b Walter Leitnerc and Thomas E. Müllerc,d,*
Received 00th January 20xx,
Accepted 00th January 20xx
Cleavage and (re)formation of oxymethylene moieties are essential steps during the built-up of polyoxymethylene (POM),
a technically relevant high-performance polymer. To reveal how the catalyst accomplishes the cleavage of oxymethylene
moieties, the kinetics of the ring-opening of trioxane was studied. Thereby insights into the initial phase of the growth of
oxymethylene chains were obtained. The chain length of short oligomers was controlled with acetic anhydride as transfer
agent. With a high ratio of acetic anhydride to trioxane, the first homologues of the series, trioxymethylene diacetate,
dioxymethylene diacetate and monooxymethylene diacetate were obtained in excellent yield. The homologues show
distinct features in the NMR and IR spectra that were related to their reactivity during chain propagation. Formation of
intermediate hemi-acetal oxonium moieties is suggested to be a key step in the reaction pathway. Such molecular level
insight may be a crucial factor for further tailoring production and properties of polyoxymethylene as well as introducing
oligomeric oxymethylene diacetates to new application fields.
DOI: 10.1039/x0xx00000x
www.rsc.org/
Gaseous formaldehyde is used in producing homopolymers,
such as DelrinꢀPOM, which is acetateꢀcapped.12 Hydroxyꢀ
terminated POM is produced by copolymerising trioxane with
small amounts of a comonomer such as ethylene oxide11,13 or
dioxolane.11 The comonomer introduces oxyethylene moieties
into the polymer chain stabilising the chain against
depolymerisation.14
POM finds manifold technical applications as engineering
plastics in the automotive, electronics and consumer goods
sectors. Its characteristics include outstanding chemical
robustness, compatibility with most solvents, low friction and
adequate lifetime.11 The highly crystalline POM is readily
moulded and extruded. Outstanding mechanical and physical
properties are tailored to the requirements of the application by
adjusting the fine structure of the polymer scaffold and the type
and quantity of comonomers.11,12,15,16
Introduction
Formaldehyde, a base chemical,1 is produced commercially on
a 30 mega ton/a scale2 largely by partial oxidation3 (Scheme 1)
or dehydrogenation of methanol.4 Conventionally, methanol is
produced mostly from syngas.5 Bioꢀmethanol is accessible
from renewable raw materials,7 e.g. by fermentation of sugar
and cellulose based biomass, from syngas obtained by
gasification of biomass or by reducing carbon dioxide with
hydrogen produced in a renewable manner with electricity from
sustainable energy sources.8
The option to switch the carbon source to a renewable
feedstock renders formaldehyde a stimulating intermediate for
the future production of polymers, fuels and chemicals. The
trimer of formaldehyde, trioxane, is a likewise interesting
intermediate. Simple and energyꢀefficient processes for the
industrial production of trioxane are under development,9 as
the current production process, dating from the 1960s, is energy
intensive. Downstream polyoxymethylene (POM) is produced
either directly from gaseous formaldehyde or trioxane. More
than 800 kt/a POM resins are produced worldwide,11 whereby
copolymers represented more than half of the market share.
,6
,
10
Biomass
1
/
2 O2
O
H
H
3x
R
R'
Syngas
CH3OH
O
O
O
O
O
O
H2O
n
CO2 / 4 H2
Formaldehyde
Trioxane
Polyoxymethylene(POM)
Scheme 1. Formaldehyde, readily accessible from renewable raw materials, is a
stimulating C1 polymer building block for the production of polyoxymethylene
(POM,5 9 R = Initiator, R’ = end group) and oligomeric oxymethylene derivatives
,
(this paper) for innovative application fields
Although POM has been produced commercially for decades,
the elementary steps during initiation of the trioxane
polymerisation are not yet fully understood.17ꢀ20 For solidꢀacid
catalysts, such as AmberlystTM 46, ringꢀopening of trioxane,
followed by the release of molecular formaldehyde, which is
consecutively built into the polyoxymethylene chain has been
proposed.21 As alternative pathway, the direct insertion of
trioxane as a trimer into the growing polymer chain has been
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 2 of 10
View Article Online
DOI: 10.1039/C8CY01691G
ARTICLE
Journal Name
suggested.26 At present, scarce information is available on the oxymethylene diacetates with higher chain length (32 molꢀ%,
initial steps of trioxane polymerisation. Initiation involves average length of 9.2 units). Overall, the average chain length
several chemical steps, whereby catalyst, a chainꢀtransfer agent, n(avg) was 5.2, which corresponds to an average molecular
like acetic anhydride, interact with the formaldehyde source.
weight of 258.8 g/mol. The precipitate that had formed
Consequently, this study is aimed at shedding light on how the consisted mostly of OHꢀterminated oligomeric oxymethylene
catalyst impacts the ringꢀopening of trioxane. Focusing on moieties. More detailed analysis of aliquots taken during the
acetic anhydride as chainꢀtransfer agent, the initial growth of course of the reaction revealed that TOD, DOD and MOD were
oligomers was followed by in situ IR spectroscopy. Kinetic formed only during the initial phase of the reaction as long as
profiles were recorded and the initial steps described with a acetic anhydride was present. In contrast, oligomeric
model involving
a direct and an indirect pathway. By oxymethylene diacetates were formed over the entire course of
determining the corresponding rate constants, the significance reaction.
of the two pathways was evaluated. Finally, we discuss the
plausible mechanism of the trioxane ringꢀopening reaction in
the context of the current findings and suggest to consider the
use of oligomeric oxymethylene acetates as fuels with
22,23
properties similar to dimethylether
fuels.24,25
and other oxygenated
Results and discussion
Openꢀchain oxymethylene diacetates, the shortest chain
analogues of POM, were obtained by the Brønsted acidꢀ
catalysed reaction of trioxane with acetic anhydride (Eq. 1).26
While at low temperatures (ꢀ60 °C) trioxymethylene diacetate
(TOD, n = 3) was obtained as the main product, an equimolar
mixture of dioxymethylene diacetate (DOD, n = 2) and
monooxymethylene diacetate (MOD, n = 1) was formed at
higher temperatures (65 °C). These oxymethylene diacetates
were isolated and further on served as reference compounds.
Figure 1. Timeꢀconcentration profile of the reaction of trioxane with acetic
anhydride in the presence of catalytic amounts of trifluoromethane sulfonic acid
(n(trioxane)/n(Ac2O) 3/1, dichloromethane,
0
°C). Concentrations were
determined by convoluting the timeꢀresolved IR spectra (insert, IR spectra taken
every 30 s). The concentrations of TOD, DOD, MOD and oligomeric diacetates
are omitted for clarity and can be found as supplementary information.
(1)
Reaction pathway and kinetics
To obtain more detailed insight into the initial course of the
reaction, the CF3SO3Hꢀcatalysed reaction was repeated with a
lower molar ratio of trioxane to acetic anhydride (1:4.5). The
profile recorded with in situ IR spectroscopy (Figure 2)
In an analogous manner, we targeted at synthesising oligomeric
oxymethylene diacetates with n > 3 by reacting trioxane with
acetic anhydride in a molar ratio of 3:1. The reaction was
performed at 0 °C in the presence of catalytic amounts of
trifluoromethane sulfonic acid. Unexpectedly, the amount of
trioxane built into the diacetates was much lower than
anticipated. To obtain insight into the underlying reasons, the
course of the reaction was monitored by means of in situ IR
spectroscopy. Inspection of the IR spectra recorded during the
reaction revealed fast conversion of acetic anhydride and
trioxane during the initial phase of the reaction (Figure 1).
Trioxane conversion slowed down, once acetic anhydride had
been fully consumed. This coincided with the formation of a
precipitate. After further 10 min, the consumption of trioxane
revealed
a fast initial decrease in the intensity of the
characteristic bands at 1162 and 1124 cmꢀ1 assigned to the
ν
as(COC) ether and ester stretch vibration of trioxane and acetic
1
accelerated. With H NMR spectroscopy the chain lengths of
the diacetates in the soluble fraction of the product mixture
obtained at the end of the reaction (105 min) was determined.
The mixture contained MOD (3 molꢀ%), DOD (13 molꢀ%),
TOD (21 molꢀ%), tetraoxymethylene diacetate (17 molꢀ%),
pentaoxymethylene diacetate (14 molꢀ%) and oligomeric
2 | Catal. Sci. Techn., 2018, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 3 of 10
View Article Online
DOI: 10.1039/C8CY01691G
Journal Name
ARTICLE
Figure 2. Timeꢀresolved IR profile for the reaction of trioxane with acetic
anhydride catalysed by trifluoromethane sulfonic acid (n(trioxane)/n(Ac2O) 1/4.5,
dichloromethane, 0 °C). The characteristic bands of trioxane, acetic anhydride,
TOD, DOD and MOD are highlighted.
Analysis of the rate constants revealed that the reaction
proceeded predominantly via the reaction of trioxane with
acetic anhydride to TOD (k1 = 0.021 (mol/L)ꢀ1sꢀ1). To a smaller
extent, DOD and MOD were formed directly from trioxane (k2
= 0.002 (mol/L)ꢀ1/s). The consecutive reaction that converted
TOD to DOD and MOD had a much lower rate constant
(k3 < 0.001 (mol/L)ꢀ1/s).
anhydride, respectively. In parallel, the intensity of the
characteristic band at 932 cmꢀ1, assigned to the
νsym(COC) ether
stretch vibration of TOD, increased rapidly at the initial time of
the reaction. At a later stage the intensity decreased. The
intensity of characteristic bands at 950 and 1199 cmꢀ1 assigned
Consecutive pathway
Cleavage of DOD
not observed
to the
νsym(COC) ether stretch vibration of DOD and the
O
O
O
νas(COC) ester stretch vibration of MOD, respectively,
[Cat.]
Ac2O
[Cat.]
O
O
O
O
O
O
[Cat.]
Ac2O
O
O
increased steadily. Once the trioxane band had vanished (after
190 s), the intensity of the band assigned to acetic anhydride
decreased more slowly. The decreasing intensity of the band
assigned to TOD was accompanied by the steady increase in the
intensity of the bands assigned to DOD and MOD.
MOD
Ac2O
+
k1
k3
O
O
O
O
(absence
confirmed
in reference
reaction)
(confirmed
in reference
reaction)
3
TOD
2
DOD
[Cat.]
2 Ac2O
Direct pathway
To quantify the concentrations, the timeꢀresolved IR spectra
were deconvoluted and referenced against calibration spectra of
the pure components. The timeꢀconcentration profile (Figure 3)
revealed that the concentration of trioxane and acetic anhydride
decreased quickly during the initial phase of the reaction
(rini/c(Cat.) ꢀ0.482 and ꢀ0.531 sꢀ1, respectively). In parallel, the
concentration of TOD increased rapidly (0.433 sꢀ1) and that of
DOD and MOD more slowly (both 0.049 sꢀ1). After trioxane
had been fully converted (after 190 s) the concentration of TOD
decreased at a relatively low rate (rmax/c(Cat.) ꢀ0.009 sꢀ1). In
parallel, the concentrations of DOD and MOD continued to
increase. It is noteworthy that the concentrations of DOD and
MOD remained equal. This suggests that DOD and MOD are
invariably parallel products.
k2
Kinetic model
Formation of TOD: r1 = k'1 [Trioxane] [Ac2O]
Direct pathway to DOD and MOD: r2 = k'2 [Trioxane] [Ac2O]
Cleavage of TOD to DOD and MOD: r3 = k'3 [TOD] [Ac2O]
[Cat.]
O
O
O
O
O
O
[Cat.]
Ac2O
O
+
+
O
O
O
O
O
O
H
H
2
3
(not observed)
Scheme 2. Kinetic model for describing the concentration profiles of the reaction
of trioxane with acetic anhydride under the conditions used in this study;
k’i = ki [Cat.].
To confirm the consecutive nature of the reaction, TOD was
reacted with acetic anhydride (ratio 1:4.9) under the same
reaction conditions. The concentration of TOD and acetic
anhydride decreased, while DOD and MOD were formed in
parallel in equimolar ratio. This proves that TOD is able to
react with acetic anhydride to DOD and MOD. To exclude the
possibility of an additional consecutive reaction step from DOD
to MOD, also DOD was reacted with excess acetic anhydride
(ratio 1:4.6) in the presence of catalytic amounts of CF3SO3H.
No conversion was detected. Furthermore, upon closer
inspection of the in-situ IR spectra, the absence of bands
assigned to monomeric formaldehyde was established.
Physicochemical characterisation of oxymethylene diacetates
To understand their reactivity, the isolated oxymethylene
diacetates (TOD, DOD and MOD) were fully characterised (see
supplementary information).
A downfield shift of the 1H NMR signal (in CDCl3) was
observed for oxymethylene moieties neighbouring the acetate
groups in the sequence TOD (5.30 ppm), DOD (5.32 ppm) and
MOD (5.62 ppm). The signal of the internal oxymethylene
group in TOD was observed at 4.90 ppm (Table 1). This
corresponds to a fairly high electron density for the inner
oxymethylene protons of TOD. In comparison, the 1H NMR
signal of the trioxane protons was at 5.15 ppm. Thus, the
central oxymethylene moiety in TOD is much less positively
polarised than the oxymethylene moieties in trioxane. This
implies that likewise to trioxane the inner oxymethylene moiety
Figure 3. Timeꢀconcentration profile of the reaction of trioxane with acetic
anhydride catalysed by trifluoromethane sulfonic acid (n(trioxane)/n(Ac2O) 1/4.5,
CH2Cl2, 0 °C). The continuous lines provide a fit of the data with the model
according to Scheme 2.
To obtain further insight into the reaction pathways, the
concentration profile was fitted with kinetic models. The profile
was described well, when the formation of TOD (r1) and
consecutive cleavage of TOD to DOD and MOD (r3) as well as
the direct cleavage of trioxane to DOD and MOD (r2) were
taken into account (Scheme 2). This kinetic model provided
excellent fit with the observed concentrations (Figure 3).
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 4 of 10
View Article Online
DOI: 10.1039/C8CY01691G
ARTICLE
Journal Name
of TOD ought to react readily with electrophiles. In contrast, influenced strongly by the nucleophilicity of the anion. The
the oxymethylene moieties neighbouring an acetate groups are reactivity is reflected also in the ratio of the rate constants k2/k3
positively polarised attenuating the tendency to react with (3.5 for CF3SO3H and 0.16 for HClO4) suggesting stronger
electrophiles.
stabilisation of a cationic intermediate by the more polarisable
ꢀ
This is confirmed by close inspection of the IR data. The ClO4 ion (vide infra).
similar ether stretch vibration (ν
sym) of trioxane (930 cmꢀ1) and
Table 2. Initial rate and product distribution of the reaction of trioxane with
acetic anhydride in the presence of Brønsted acid catalysts.
TOD (932 cmꢀ1) in IR suggest similar reactivity of the
oxymethylene moieties. The higher frequency of the ether
stretch vibration in DOD (950 cmꢀ1) suggests a lower reactivity.
Thus, assuming that the rate determining step involves
breakage of the oxymethylene ether bond, the highest reactivity
with acetic anhydride ought to be expected for TOD and
trioxane. Note that the frequency of the ester stretch vibrations
Catalyst
CF3SO3H HClO4
pKa value
ꢀ14
ꢀ0.48
8.9
31
3.5
ꢀ10
ꢀ0.27
97
16
0.16
100
(rini)/c(Cat.) for trioxane conversion a
[sꢀ1]
[ꢀ]
[ꢀ]
[ꢀ]
[%]
b
k1/k2
k1/k3
k2/k3
b
b
c
Conversion Trioxane
100
sym) are similar for TOD (1002 cmꢀ1), DOD (1007 cmꢀ1) , and
MOD (1016 cmꢀ1). The most distinct bands in IR for trioxane
(1162 cmꢀ1, as(COC)), acetic anhydride (1124 cmꢀ1,
as(COC)),
TOD (932 cmꢀ1, sym(COC)), DOD (950 cmꢀ1,
sym(COC)) and
MOD (1199 cmꢀ1,
as(COC)) are separated sufficiently to allow
Molar ratio TOD / DOD / MOD d
[molꢀ%] 3 / 51 / 46 1 / 51 / 47
(
ν
a Initial rate in molTrioxane.(molCat..s)ꢀ1
;
b ratio of rate constants determined from
ν
ν
the respective timeꢀconcentration profile (based on in situ IR data).
c conversion after 2 h; d molar ratio in isolated product mixtures.
ν
ν
ν
Lewis acid catalysts
for deconvoluting the IR spectra of reaction mixtures. This
makes in situ IR spectroscopy an excellent tool for monitoring
the course of this reaction.
Then the performance of metal triflates as catalysts was
explored as their Lewis acidity can be adjusted in a wide range
by choosing an appropriate metal +III centre.29,30 The influence
of the cation radius (hardness)31 and the influence of the
electronic configuration were studied for the main group metal
triflates SbIII(OTf)3 (ionic radius r = 76 pm) and BiIII(OTf)3 (r =
103 pm) and the transition metal triflates ScIII(OTf)3 (r = 75
pm), YIII(OTf)3 (r = 103 pm), and LaIII(OTf)3 (r = 103 pm).
With SbIII(OTf)3 trioxane was converted fast at an initial rate
(rini/c(Cat.)) of ꢀ0.61.sꢀ1, and full conversion was achieved
within 100 s. Similarly, with BiIII(OTf)3 trioxane was converted
fast (full conversion within of 300 s) at an initial rate
(rini/c(Cat.)) of ꢀ0.35 sꢀ1. In contrast, with ScIII(OTf)3 trioxane
was converted more slowly at an initial rate of ꢀ0.14 sꢀ1 and full
conversion was obtained within 3000 s. With YIII(OTf)3
trioxane was converted very slowly (rini/c(Cat.) = ꢀ0.0012.sꢀ1)
and conversion was only 37 % after 2 h of reaction time. With
LaIII(OTf)3 there was no detectable reaction within of 2 h. Thus,
a higher activity was found for main group elements compared
to their early transition metal Lewis acid counterparts with
similar radius. Thus, the main group Lewis acids are more
active compared to the closedꢀshell early transition metals,
most likely due to their open shell configuration.
The activity of the Lewis acid catalysts increased with the
decreasing radius of the metal cation within the chosen main
group and transition metal series. For instance, SbIII(OTf)3 gave
a 1.7ꢀfold higher rate compared to BiIII(OTf)3. Similarly,
ScIII(OTf)3 gave a 120 times higher rate compared to YIII(OTf)3.
Hence, small cations, corresponding to hard Lewis acids,
resulted in higher rates. The formation of DOD and MOD was
significantly less pronounced for ScIII(OTf)3, or did not occur at
all for YIII(OTf)3, compared to the main group counterparts.
This is also reflected by the ratios of the determined rate
constants (Table 3). The ratio of the rate constant k1/k2 and
k1/k3 were similar for SbIII(OTf)3 (3.9 and 13, respectively),
BiIII(OTf)3 (6.3 and 32.7, respectively). In contrast, ScIII(OTf)3
(22 and 1400, respectively) and YIII(OTf)3 (no secondary
reactions) were highly selective to TOD formation. Moreover,
Table 1 Frequency of the symmetric ether and ester stretch vibrations in
oxymethylene diacetates in comparison to trioxane and calculated force
constants.
δ(OꢀCH2ꢀO)a
[ppm]
vsym (Ether)
vsym (Ester)
[cmꢀ1]
1002
[cmꢀ1]
932
950
ꢀ
TOD
DOD
MOD
Trioxane
5.30/4.90 b
5.32
5.62
1007
1016
ꢀ
5.15
930
a Chemical shift of the oxymethylene groups in 1H NMR spectroscopy; b outer
and inner oxymethylene group, respectively.
Brønsted acid catalysts
To explore the effect of the acid strength, the activity of
trifluoromethane sulfonic acid (CF3SO3H) was compared with
other Brønsted acids (Table 2). CF3SO3H (pKa = ꢀ14)27
provided an initial rate of trioxane conversion of rini/c(Cat.) ꢀ
0.48.sꢀ1 and full conversion within of 200 s. With perchloric
acid (HClO4, pKa = ꢀ10),26,27 there was fast conversion of
trioxane, albeit at a lower rate (rini/c(Cat.) ꢀ0.27.sꢀ1; full
conversion within of 300 s). In contrast, there was no
observable reaction with methane sulfonic acid (CH3SO3H, pKa
= ꢀ2.6),27
pꢀtoluene sulfonic acid (pꢀCH3ꢀC6H4ꢀSO3H, pKa =
0.7)27 and acetic acid (AcOH, pKa = 4.6).27 This clearly shows
that only very strong Brønsted acids catalyse the reaction.
Likewise, it was reported that H2SO4 (pKa = ꢀ3.0)27 catalyses
the reaction.28 The need to employ strong acids is consistent
with the low basicity of the substrate trioxane. Apparently, only
very strong Brønsted acids can sufficiently activate these
substrates.
The higher initial rate with CF3SO3H compared to HClO4
suggests that the activity increases with increasing acidity of
the Brønsted acid. The inverse ratio of the rate constant k1/k2
and k1/k3 for CF3SO3H (8.9 and 31, respectively) and HClO4
(97 and 16, respectively) reveals that the pathway via TOD is
more pronounced for HClO4. Thus, the reaction pathway is
4 | Catal. Sci. Techn., 2018, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 5 of 10
View Article Online
DOI: 10.1039/C8CY01691G
Journal Name
ARTICLE
the ratios of the rate constants k2/k3 of SbIII(OTf)3 (3.5) and intermediate I1. Opening of the trioxane ring provides the
BiIII(OTf)3 (5.2) compared to ScIII(OTf)3 (67.2) underline a hemiꢀacetal oxonium intermediate I2. Reaction with acetic
different selectivity of main group and early transition metal anhydride or acetic acid provides TOD. There are two pathways
triflates. In a reference experiment, trioxane was reacted in for formation of DOD and MOD. (i) In the direct pathway to
dichloromethane solution in the presence of catalytic amounts DOD and MOD the hemiꢀacetal oxonium intermediate I2 reacts
of ScIII(OTf)3 (1 molꢀ%) at room temperature. Even after three with acetic anhydride, whereby a single oxymethylene unit is
days no reaction was observed. This shows that the presence of transferred in a concerted manner to form MOD. Thereby, the
acetic anhydride is required for ScIII(OTf)3 to be able to convert hemiꢀacetal oxonium intermediate is shortened by one
trioxane.
oxymethylene unit to intermediate I4. Reaction with acetic
anhydride or acetic acid provides DOD. (ii) In the consecutive
pathway to DOD and MOD an acylium ion reacts with TOD to
form intermediate I5. Later is cleaved to MOD and to
intermediate I4. Reaction with acetic anhydride or acetic acid
provides DOD. Both pathways provide an equimolar ratio of
DOD and MOD.
Table 3. Initial rate and product distribution of the reaction of trioxane with
acetic anhydride in the presence of Lewis acid catalysts.
Catalyst
r
SbIII(OTf)3 BiIII(OTf)3 ScIII(OTf)3 YIII(OTf)3
[pm]
[sꢀ1]
76
103
75
103
rini/c(Cat.) for
ꢀ0.61
ꢀ0.35
ꢀ0.14
ꢀ0.0012
trioxane conversion a
b
k1/k2
k1/k3
k2/k3
[ꢀ]
[ꢀ]
[ꢀ]
3.9
13
3.5
100
6.3
32.7
5.2
22
ꢀ
ꢀ
ꢀ
37
b
1.4 × 103
67
b
Conv. of trioxane c [%]
Molar ratio
100
100
[molꢀ%]15/44/41 15/45/40 71/15/14 37/0/0
TOD/DOD/MOD d
a Initial rate in molTrioxane.(molCat..s)ꢀ1; b ratio of rate constants determined from
the respective timeꢀconcentration profile (based on in situ IR data).
c conversion after 2 h; d molar ratio in isolated product mixtures.
As with Brønsted acid catalysts, the extent of consecutive
reactions was explored for Lewis acid catalysts. Thus, TOD
was reacted with an excess of acetic anhydride in the presence
of either BiIII(OTf)3 or ScIII(OTf)3. As with Brønsted acid
catalysts, DOD and MOD formed in equimolar ratios. The use
of the harder Lewis acid BiIII(OTf)3 resulted in higher Scheme 3 Potential activation modes for acetic anhydride (top) and trioxane
conversion compared to ScIII(OTf)3, in agreement with the
(bottom) by Brønsted acid (HX) and Lewis acid (MX3) catalysts leading to
formation of an acylium cation or hemiꢀacetal oxonium, respectively.
aboveꢀdescribed trends in catalyst activity.
Accordingly, trioxane reacts either with one acetic anhydride
Role of the catalyst
molecule to TOD or with two molecules of acetic anhydride to
Analysis of the potential role of the catalyst in the reaction of
DOD and MOD. A relationship between the anion Xꢀ of the
trioxane with acetic anhydride revealed two modes of how the
catalyst and the selectivity becomes obvious upon comparing
substrates can be activated (Scheme 3). The carboxyl oxygen
the ratio k2/k3 for CF3SO3H (3.48) and HClO4 (0.16). Both
atom of acetic anhydride is susceptible to protonation by a
pathways are feasible in the presence of CF3SO3H, while the
Brønsted acid or coordination by a Lewis acid. This activates
consecutive pathway is much more strongly pronounced in the
acetic anhydride to be cleaved into an acylium cation and acetic
presence of HClO4. The CF3SO3ꢀ anion is a weaker nucleophile
acid or a coordinated acetate ion, respectively (Scheme 3, top).
ꢀ
compared to ClO4 , reflected in the higher acidity of the
Likewise, the oxygen atom of trioxane is susceptible to
corresponding trifluoromethane sulfonic acid. Most likely, the
better separation of the ion pair I2 in case of CF3SO3H
facilitates the concerted transfer of the single oxymethylene
unit via I3. In contrast, in the presence of the stronger
protonation by a strong Brønsted acid or coordination to a
Lewis acid, whereby ringꢀopening and formation of a hemiꢀ
acetal oxonium ion may be triggered (Scheme 3, bottom).
Activation of acetic anhydride seems more likely than
ꢀ
nucleophile ClO4 intermediate I2 is more stabilised.
activation of trioxane. This claim is supported by the reduced
Consequently, the consecutive pathway is favoured.
rate of trioxane conversion, which was observed once all acetic
Noteworthy, in the presence of excess acetic anhydride, TOD
anhydride had been consumed. Considering the initial rate of
was cleaved in the presence of catalyst. In contrast, DOD was
trioxane conversion, the catalyst activity decreased in the
stable under our reaction conditions. Closer inspection of the
sequence CF3SO3H > BiIII(OTf)3 > ScIII(OTf)3. A similar trend
corresponding NMR data of TOD reveals a significant highꢀ
in catalyst activity had been reported for the related Friedel
field shift signal of the inner oxymethylene groups (∂H = 4.90
Crafts acylation of toluene with benzoic anhydride.32 Activation
ppm,
∂
= 92.3 ppm) compared to the corresponding
oxymethylene groups that have neighbouring acetate units (∂H
= 5.30 ppm, ∂C = 85.4 ppm). The oxymethylene units of DOD
C
of the anhydride to the corresponding acylium ion as key
intermediate had been proposed.
In the next step (Scheme 4), the acylium cation is proposed to
react with trioxane resulting in the cationic acylium trioxane
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 6 of 10
View Article Online
DOI: 10.1039/C8CY01691G
Catal. Sci. Technol.
ARTICLE
O
3
Initiation
O
X
O
O
O
O
O
O
O
O
O
O
+ HX/MX3
3 TfO
O
O
M
O
O
O
O
O
- HAc/MAcX2
X
O
I1
Acylium
O
O
O
O
O
O
O
Ac2O or AcOH
- Ac+ X- or - HX
Ac
2O
O
O
O
O
O
O
O
H2C
O
X
H2C
TOD
O
X
I2
O
I3
X
Direct
O
pathway
O
O
O
O
Consecutive
pathway
via TOD
O
O
O
O
O
O
CH2
X
O
O
O
O
X
I4
I5
MOD
Ac2O or AcOH
- Ac+ or - HX
O
O
O
O
O
Ac2
O or Ac
OH
H2C
O
O
O
O
O
O
O
Ac+ X- o
-
r - HX
I4
X
MOD
DOD
MOD and DOD formed in parallel
either by consecutive or direct pathway
Scheme 4 Proposed mechanism of ringꢀopening and cleavage of trioxane in the presence of acetic anhydride as chain transfer agent and Br nsted acid or late transition
ø
metal triflate catalyst (M = Sb, Bi)) and transition state of the concerted pathway with early transition metal triflate catalysts (M = Sc, Y, insert).
(∂H = 5.32 ppm, ∂C = 86.7 ppm) are very similar to the acetate ways: First, the concerted transition state requires a particular
neighboured oxymethylene units of TOD. Thus, the high orientation of the molecules. The low probability of adopting
electron density at the inner oxymethylene groups that is the appropriate orientation lowers the overall rate of reaction.
reflected in the highꢀfield shift facilitates the reaction of the Second, the effective concentration of acylium cations may be
acylium ion with TOD to form intermediate I5. Apparently, for lower, since in the concerted first reaction step the substrates
DOD the electron density at the oxymethylene unit is compete for ScIII(OTf)3 as the reaction partner. Interestingly,
insufficient to allow for reaction with the acylium ion.
YIII(OTf)3 catalysed only the formation of TOD albeit at a low
In contrast, the ScIII(OTf)3ꢀcatalysed reaction may follow an rate that is reflected in the low conversion of trioxane of 37 %.
alternative mechanism, as considerably higher ratios of k1/k2 The formation of DOD and MOD was not observed. Being a
(21.5) and k1/k3 (1441) were determined even though, a similar soft Lewis acid, the yttrium catalysed reaction most likely
ratio was expected due to the triflate ligands. Also, the reaction proceeds only via the concerted transition state.
was secondꢀorder with respect to trioxane. We propose that a
Mechanism of the cationic oligomerisation of trioxane
second trioxane molecule is required as ligand at the scandium
With respect to the formation of oligomeric oxymethylene
diacetates with n > 3 we propose that a similar mechanism is
metal centre to achieve the right electronic density to catalyse
the reaction between acetic anhydride and an incoming trioxane
molecule. We suggest the reaction to proceed predominantly
via an innerꢀsphere concerted transition state (Scheme 4,
insert). The much lower overall rate can be explained in two
adopted. The low reaction temperature well below the ceiling
33
temperature
does not favour the formation of monomeric
formaldehyde. We propose that chain growth occurs in a
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 6
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 7 of 10
View Article Online
DOI: 10.1039/C8CY01691G
Journal Name
ARTICLE
Synthesis of oxymethylene diacetates26
similar manner as described above by nucleophilic attack of a
hemiꢀacetal oxonium intermediate on trioxane (Scheme 6).
Acetic anhydride functions as chainꢀtransfer agent, and an
acylium ion is released, when a growing chain is terminated.
Such chainꢀtransfer is known for many polar compounds, such
as water, methanal, dimethoxymethane, esters and acid
anhydrides as well as some olefins and aromatics.34ꢀ38 Thereby
an end group is introduced, while the released cationic
component initiates a new growing chain. Compared to the
initiation reaction with acetic anhydride reꢀinitiation of an
oxymethylene diacetate moiety is less favoured. Consequently,
chain growth commences once acetic anhydride had been fully
consumed. This readily explains the preferential formation of
oligomeric oxymethylene diacetates. Later are formed by
several consecutive chain growth steps before the terminal
hemiꢀacetal oxonium reacts with another nucleophile. The
precipitate is undoubtedly formed, when the chain growth
reaction dominates.
A mixture of acetic anhydride (94 mL, 1 mol) and perchloric
acid (70 %, 0.2 mL, 0.0023 mol) was cooled to ꢀ60 °C.
Trioxane (19.831 g, 0.22 mol) was added over 10 minutes and
the reaction mixture stirred for one hour. Then the reaction
mixture was poured into a saturated aqueous solution of sodium
carbonate (200 mL). The mixture was extracted four times with
diethyl ether (50 mL). The combined organic phases,
containing a high percentage of TOD (Table 4), were dried over
Na2SO4, the volatiles removed under partial vacuum and TOD
isolated by distillation of the crude product mixture at reduced
pressure (1 mbar).
Table 4. Composition of the crude reaction mixtures
Entry a Temp. [°C] TOD [molꢀ%] DOD [molꢀ%] MOD [molꢀ%]
1
2
ꢀ 60 °C
+ 65 °C
78
0
13
39
9
61
a The yields of TOD, DOD and MOD together are 78 wtꢀ% and 80 wtꢀ% for
entries 1 and 2, respectively.
A mixture of acetic anhydride (47 mL, 0.5 mol) and perchloric
acid (70 %, 0.1 mL, 0.0012 mol) was heated to 63 °C. Trioxane
(9.926 g, 0.11 mol) was added over 10 minutes and the reaction
mixture stirred for another 10 minutes. Then, the reaction
mixture was poured into a saturated aqueous solution of sodium
carbonate (200 mL). The mixture was extracted four times with
diethyl ether (50 mL). The combined organic phases,
containing a high percentage of DOD and MOD (Table 4),
were dried over Na2SO4. The volatiles were removed in a
partial vacuum and DOD and MOD isolated by fractionated
distillation of the crude mixture at reduced pressure (1 mbar).
Trioxymethylene diacetate (TOD)
Scheme 5. Mechanism proposed for chain growth during the ringꢀopening
oligomerisation of trioxane (St = initiator).39 The low reaction temperature does
not favour the formation of monomeric formaldehyde.33,39
∂
(400 MHz, CDCl3) = 1.99 (s, 6 H, C(O)CH3), 5.30 (s, 4 H,
AcOCH2OCH2OCH2OAc), 4.90 (s, 4 H, OCH2OCH2OCH2O).
H
∂
(100
MHz,
CDCl3)
=
20.7
(H3CC(O)ꢀ),
85.4
C
(AcOCH2OCH2OCH2OAc), 92.3 (AcOCH2OCH2OCH2OAc ) 170.1
Considering the mass balance, the extent of termination
reactions must be very low. In case of Brønsted acid catalysts,
reaction of the hemiꢀacetal oxonium intermediate with acetic
acid formed during initiation regenerates the catalyst.
(H3CC(O)ꢀ).
υmax/cmꢀ1: 2983 (w, CH3), 2921 (w, CH2), 1742 (s, C=O), 1465 (w), 1428
(w), 1368 (w, CH3), 1222 (s, υas ester), 1204 (w, υas ester), 1165 (w,
υas ether), 1127 (s, υas ether), 1049 (w), 999 (s, υsym ester), 955 (w,
υsym ester), 930 (vs, υsym ether), 833 (w), 607 (w), 534 (w).
m/z (CI): 326.3 (<1 %), 325.3 (4), 297.3 (3), 296.3 (12) 295.3 (100, M+ +
Ac2O + H), 294.7 (2, M+ + Ac2O), 267.3 (1), 266.2 (6), 265.3 (63, M+
+ Ac2O + H – CH2=O), 205.3 (<1, 2 × Ac2O + 1), 103.2 (5).
Elemental analysis
Termination occurs in the presence of excess protic species as it
41,42
is known for instance with alcohols.40,41 Also occlusion
of
the chain end in the bulk of the precipitating crystalline
polymer may terminate the cationic oligomerisation of trioxane.
This may explain the aboveꢀmentioned gradual decrease in the
rate of trioxane consumption. Workꢀup involved quenching the
reaction mixture with a saturated aqueous solution of sodium
carbonate. Alternatively, chain growth can be terminated by
addition of basic compounds such as sodium alkoxides.43
Another termination method employs alkyl and aryl
phosphines, which results in trapping of growing chains as
quaternary phosphonium salts.44
Found: C, 43.6; H, 6.45. Calc. for C7H12O6 C, 43.75; H, 6.45 wt.ꢀ%
Dioxymethylene diacetate (DOD)
∂
(400 MHz, CDCl3) = 2.00 (s, 6 H, C(O)CH3), 5.32 (s, 4 H,
OCH2OCH2O).
(100 MHz, CDCl3) = 20.7 (H3CC(O)ꢀ), 86.7 (AcOCH2OCH2OAc),
170.0 (H3CC(O)ꢀ).
H
∂
C
υmax/cmꢀ1 = 2983 (w, CH3), 2927 (w, CH2), 1744 (s, C=O), 1461 (w),
1426 (w), 1368 (w, CH3), 1219 (s, υas ester), 1187 (w, υas ester), 1142
(s, υas ether), 1087 (s, υas ether), 1029 (w, υsym ester), 1004 (s, υsym
ester), 937 (vs, υsym ether), 832 (w), 606 (w), 550 (w).
m/z (EI) = 267.2 (< 1 %), 266.1 (2), 265.1 (20, M+ + Ac2O + H), 205.1 (4,
2 × Ac2O + 1), 146.1 (<1), 145.1 (8), 104.1 (5), 103.2 (100, Ac2O +
H), 102.2 (<1), 73.2 (10).
Experimental
Elemental analysis
Found: C, 44.25; H, 6.1. Calc. for C6H10O5 C, 44.45; H, 6.2 wt.ꢀ%
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 8 of 10
View Article Online
DOI: 10.1039/C8CY01691G
ARTICLE
Journal Name
Monooxymethylene diacetate (MOD)
Table 7. Quantities employed in the Lewis acid catalysed reaction of trioxane
with acetic anhydride in comparison to the corresponding Brønsted acid
catalysed reaction with CF3SO3H and selected experimental data.
∂
(400 MHz, CDCl3) = 2.00 (s, 6 H, C(O)CH3), 5.62 (s, 2 H,
H
AcOCH2OAc).
∂
(100 MHz, CDCl3) = 20.5 (H3CC(O)ꢀ), 79.0 (AcOCH2OAc), 169.5
(H3CC(O)ꢀ).
C
Catalyst
Trioxane
Acetic anhydride
Catalyst
CF3SO3H BiIIIOTf3 ScIIIOTf3
υmax/cmꢀ1 = 2994 (w, CH3), 2945 (w, CH2), 1757 (s, C=O), 1453 (w),
1431 (w), 1368 (w, CH3), 1239 (w, υas ester), 1189 (vs, υas ester),
1084 (w), 1065 (w), 1047 (w), 1004 (vs, υsym ester), 977 (s, υsym
ether), 820 (w), 666 (w), 605 (w), 518 (w), 456 (w).
m/z (ESI) = 167.3 (12 %), 157.2 (22), 155.1 (11), 150.0 (63.2), 143.1
(16), 133.9 (4, M+ + H), 132.9 (100, M+), 103.1 (26, Ac2O + H),
102.2 (85, Ac2O), 69.1 (48).
[mmol] 5.00
[mmol] 24.4
[mmol] 0.227
5.00
24.5
0.243
73
6.44
26.3
0.221
11
a
Conv.Trioxane
[%]
73
Ratio TOD/DOD/MOD a
Yield b
[molꢀ%] 16/44/40 17/46/36 82/10/8
[g] 1.84 1.79 2.04
a Molar ratio in isolated product mixtures and conversion of TOD. b Yields
are reported for product mixture isolated after aqueous work up.
Elemental analysis
Found: C, 45.2; H, 6.25. Calc. for C5H8O4 C, 45.5; H, 6.1 wt.ꢀ%
General procedure of kinetic experiments
Table 8. Quantities employed in the Brønsted acid catalysed reaction of DOD
with acetic anhydride, Lewis acid catalysed reaction of trioxane with acetic
anhydride and selected experimental data.
A background measurement with dichloromethane (30 mL) was
performed at 0 °C with an in-situ IR probe. Trioxane and acetic
anhydride were added, and the in-situ IR measurement was
started. The catalyst (Table 5, Table 6, Table 7, and Table 8)
was added to the reaction mixture at 0 °C, which was then
b
Catalyst
DOD
CF3SO3Ha
5.75
ScIIIOTf3
ꢀ
[mmol]
[mmol]
Acetic anhydride
26.3
ꢀ
stirred for 2 h at 0 °C. The reaction mixture was poured into a Trioxane
[mmol]
[mmol]
[%]
ꢀ
11.5
saturated aqueous solution of sodium carbonate (50 mL). The
organic layer was separated and washed with water (50 mL)
and, after another separation, was washed with brine (50 mL).
The organic layer was then dried with sodium sulphate and the
Catalyst
0.227
0.110
Conversion
No reaction
No reaction
a 30 mL DCM, 0 °C, 2h b 10 mL DCM, 25 °C, 3 days
solvent removed in vacuo
.
Determination of time-concentration profiles
Table 5. Quantities employed in the Brønsted acid catalysed reaction of
trioxane with acetic anhydride and selected experimental data.
Timeꢀresolved IR spectra were deconvoluted using PEAXACT
software,45 whereby the IR bands were described by Lorentz
functions accounting for changes in the position of the peak
with concentration. Initial rates were calculated by fitting the
timeꢀconcentration profile with a polynomial function and
subsequent extrapolation of the first derivative to zero time.
Catalyst
CF3SO3H HClO4 CH3SO3H
Trioxane
Acetic anhydride
Catalyst
[mmol] 6.01
[mmol] 26.9
[mmol] 0.227
5.96
26.4
0.255
100
1/51/47
2.55
6.06
25.66
0.274
No reaction
ꢀ
a
Conv.Trioxane
[%]
[molꢀ%] 3/51/46
[g] 2.70
100
Ratio TOD/DOD/MOD a
Yield b
0.637
Conclusions
a Molar ratio in isolated product mixtures and conversion of trioxane. b Yields
are reported for product mixture isolated after aqueous work up.
To obtain insight in how the catalyst impacts the initial phase of
the growth of the oxymethylene polymer chain, the ringꢀ
opening reaction of trioxane with acetic anhydride was studied
in detail. The first homologues of the series, trioxymethylene
diacetate (TOD), dioxymethylene diacetate (DOD), and
monooxymethylene diacetate (MOD), were isolated and fully
characterised. Noteworthy is the upfield shift of the signal of
the inner oxymethylene protons in TOD in 1H NMR
spectroscopy. Accordingly, TOD showed high reactivity
comparable to that of trioxane. Kinetic profiles recorded for the
reaction of trioxane with acetic anhydride in the presence of a
Brønsted or Lewis acidic catalyst revealed TOD to be a key
intermediate. TOD was found to be a primary reaction product
formed by the direct incorporation of trioxane into acetic
anhydride. To a lower extend, also MOD and DOD were
formed directly by reaction of one trioxane with two acetic
anhydride molecules. This direct formation was ascribed to the
cleavage of a hemiꢀacetal oxonium intermediate (I2) with acetic
anhydride. Noteworthy DOD was no further cleaved with acetic
anhydride under our reaction conditions to MOD moieties. In
the contrary, TOD was cleaved readily with acetic anhydride to
equimolar amounts of MOD and DOD. Thus, two pathways
were present accounting for the formation of DOD and MOD.
Table 6. Quantities employed in the Lewis acid catalysed reaction of trioxane
with acetic anhydride and selected experimental data.
Catalyst
SbIIIOTf3 BiIIIOTf3 ScIIIOTf3 YIIIOTf3
Trioxane
Acetic anhydride
Catalyst
[mmol] 5.89
[mmol] 25.3
[mmol] 0.260
6.03
26.5
0.259
100
6.08
24.9
0.483
100
6.08
26.4
0.257
37
a
Conv.Trioxane
[%]
100
Ratio TOD/DOD/MODa [molꢀ%] 15/44/41 15/45/40 71/15/14 37/0/0
Yield b
[g]
2.3
2.32
2.75
3.28
a Molar ratios in isolated product mixtures and conversions relative to the
amount of trioxane employed. b Yields are reported for the amount of product
mixture isolated after aqueous work up.
8 | Catal. Sci. Techn., 2018, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 9 of 10
View Article Online
DOI: 10.1039/C8CY01691G
Journal Name
ARTICLE
The extent to which the two reaction pathways are followed
depends critically on the choice of the catalyst. By fitting a
kinetic model of the reaction sequence to the recorded reaction
profiles, the corresponding rate constants for the individual
reaction steps were determined for different Brønsted acid and
Lewis acid catalysts. For Brønsted acid catalysts, the rate of
trioxane conversion increased with increasing acidity of the
catalyst. Noteworthy, the reaction pathway with Brønsted acid
catalysts showed a strong influence on the corresponding anion.
Acknowledgements
We acknowledge scientific discussions with Dr. Henning Vogt
and Dr. Burkhard Köhler. We thank Volker Marker, Alexandra
Keldenich and Mario Krautschick for experimental support.
Dirk Engels (S-PACT) is acknowledged for support in
deconvoluting the IR spectra.
Notes and references
Trifluoromethane sulfonic acid (pKa
=
14) catalysed
predominantly the direct pathway for DOD and MOD
formation. In contrast, perchloric acid (pKa = ꢀ10) catalysed
mostly the formation of DOD and MOD via the consecutive
pathway. We believe that the higher nucleophilic character of
the perchlorate stabilises the cationic intermediates by more
pronounced cation/anion interactions.
For Lewis acid catalysts, the activity was found to increase with
the hardness of the cation. Also the electron configuration had a
strong influence on the activity of the catalyst. The main group
Lewis acids SbIII(OTf)3 and BiIII(OTf)3 with their open shell
configuration provided much higher activity compared to the
transition metal Lewis acids ScIII(OTf)3 and YIII(OTf)3. Also
the reaction order with respect to trioxane differed for Brønsted
and main group Lewis acid catalysts (first order) and transition
metal Lewis acid catalysts (second order). While relative rate
constants were similar for Brønsted and main group Lewis acid
catalysts, transition metal Lewis acid catalysts had very
different rate constants. Accordingly, we propose two
alternative reaction mechanisms for the ringꢀopening reaction
1 G. Reuss, W. Disteldorf, A. O. Gamer, A. Hilt, Ullmann’s Encyclopaedia of
Industrial Chemistry, 2012, Vol. 15, WileyꢀVCH, Weinheim, p. 735ꢀ768.
L. E. Heim, H. Konnerth and M. H. G. Prechtl, Green Chemistry, 2017, 19,
2
2347.
3 M. D. Thomas, Journal of the American Chemical Society, 1920, 42, 873.
4 S. Su, P. Zaza and A. Renken, Chemical Engineering Technology, 1994, 17,
34.
5
S. Wesselbaum, T. Vom Stein, J. Klankermayer, W. Leitner, Angewandte
Chemie International Edition, 2012, 51, 7499.
6 J. Klankermayer, S. Wesselbaum, K. Beydoun and W. Leitner, Angewandte
between trioxane and acetic anhydride. Brønsted acid and main Chemie International Edition, 2016, 55, 7296.
7
A. S. Suntana, K. A. Vogt, E. C. Turnblom, R. Upadhye, Applied Energy,
group Lewis acid catalysts most likely follow a cationic
reaction mechanism, while the transition metal Lewis acids
follow a neutral concerted reaction mechanism. The formation
of hemiꢀacetal oxonium intermediates with Brønsted acid and
main group Lewis acid catalysts likewise accounts for the
cationic oligomerisation in the presence of excess trioxane.
Acetic anhydride thereby functions as chainꢀtransfer agent,
whereby a new acylium ion is released, when a growing chain
is terminated.
In summary, trifluoromethane sulfonic acid and bismuth (III)
triflate were identified as promising catalysts for the ringꢀ
opening oligomerisation of trioxane and cyclic anhydrides.
Oxymethylene diacetates with low molecular weight and
similar products obtained by ringꢀopening of trioxane with
other chainꢀtransfer agents, like methanol or dimethoxymethane
(DMME), have an interesting property profile. Application
fields of these highly oxygen containing oxymethylene
derivatives include the use as sustainable fuels.
2009, 86, S215.
8 J. Artz, T. E. Müller, K. Thenert, J. Kleinekorte, R. Meys, A. Sternberg, A.
Bardow and W. Leitner, Chemical Reviews, 2018, 118, 434.
9
T. Grützner, H. Hasse, N. Lang, M. Siegert and E. Ströfer, Chemical
Engineering, 2007, 62, 5613.
10
M. Siegert, N. Lang, E. Ströfer, A. Strammer, T. Friese, H. Hasse,
Deutsches Patent, DE10361516.
11
S. Lüftl, P.M. Visakh, S. Chandran, Polyoxymethylene Handbook, Wiley
VCH, 1st Edition, 2014.
12 a) N. R. MacDonald, US Pat., 2 768 994, 1956; b) C. E. Schweitzer, R. N.
MacDonald and J. O. Punderson, Journal of Applied Polymer Science, 1959,
2, 158.
13 J. Majer, O. Hainova, Kolloid-Zeitschrift, 1965, 1, 23.
14 C. Bizzarri, H. Vogt, G. Baráth, W. Leitner, T. E. Müller, Green Chemistry
2016, 18, 5160ꢀ5168.
15
C. F. Hammer, T. A. Koch and J. F. Whitney, Journal of Applied Polymer
Science, 1959, 1, 169.
16
W. H. Linton and H. H. Goodman, Journal of Applied Polymer Science,
1959, 1, 179.
17 V. Jaacks and W. Kern, Die Makromolekulare Chemie, 1963, 62, 1.
18
Y.ꢀT. Shieh, M.ꢀJ. Yeh and S.ꢀA. Chen, Journal of Polymer Science, A:
Polymer Chemistry, 1999, 37, 4198.
19
Y.ꢀT. Shieh and S.ꢀA. Chen, Journal of Polymer Science, A: Polymer
Chemistry, 1999, 37, 483.
20
J. Masamoto, K. Matsuzaki, T. Iwaisako, K. Yoshida, K. Kagawa and H.
Nagahara, Journal of Applied Polymer Science, 1993, 50, 1317.
J. Burger, E. Ströfer and H. Hasse, Industrial & Engineering Chemistry
Conflicts of interest
21
There are no conflicts to declare.
Research, 2012, 51, 12751.
22
E. Catizzone, G. Bonura, M. Migliori, F. Frusteri, G. Giordano, Molecules
201), 23(1), 31/1ꢀ31/28.
23
S. Roy, A. Cherevotan, S. C. Peter, ACS Energy Letters 2018, 3(8), 1938ꢀ
1966.
24 S. E. Iannuzzi, C. Barro, K. Boulouchos, J. Burger, Jakob, Fuel 2016, 167,
49ꢀ59.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Please do not adjust margins
Catalysis Science & Technology
Page 10 of 10
View Article Online
DOI: 10.1039/C8CY01691G
ARTICLE
Journal Name
25 E. Ranzi, A. Frassoldati, R. Grana, A. Cuoci, T. Faravelli, A. P. Kelley, C.
K. Law, Progress in Energy and Combustion Science 201), 38(4), 468ꢀ501.
26 J. Tomiška and E. Spousta, Angewandte Chemie, 1962, 74, 248.
27 Evans pKa table: http://evans.rc.fas.harvard.edu/pdf/evans_pKa_table.pdf
28 W. D. King and D. J. Stanonis, Journal of Applied Polymer Science, 1974,
18, 547.
29
S. Kobayashi, M. Sugiura, H. Kitagawa, W. W.ꢀL. Lam, Chemical
Reviews, 2002, 102, 2227.
30
J.‐F. Gal, C. Iacobucci, I. Monfardini, L. Massi, E. Duñach, S. Olivero,
Journal of Physical Organic Chemistry, 2013, 26, 2, 87ꢀ97.
31 Y. Yao, K. Nie, Rare Earth Elements, ed. D. A. Atwood, 2012, 459ꢀ474.
32 C. Le Roux and J. Dubac, Synlett, 2002, 2, 181.
33 W. Kern and V. Jaacks, Journal of Polymer Science, 1960, 48, 399.
34 M. Kučera and E. Spousta, Makromolekulare Chemie, 1965, 82, 60.
35
H. Baader, V. Jaacks and W. Kern, Makromolekulare Chemie, 1965, 82,
213.
36
H. D. Hermann, E. Fischer, K. Weissermel, Makromolekulare Chemie,
1966, 90, 1.
37
V. Jaacks, H. Baader and W. Kern, Makromolekulare Chemie, 1965, 83,
56.
38 H. D. Hermann, Makromolekulare Chemie, 1966, 99, 35.
39 K. Weissermel, E. Fischer, K. Gutweiler, H. D. Hermann and H. Cherdron,
Angewandte Chemie International Edition, 1967, 6, 526.
40 N. Brown, Journal of Macromolecular Science: Part A, 1967, 2, 209.
41 O. Vogl, Journal of Macromolecular Science: Part C, 1975, 1, 109.
42
F. Morelli, R. Tartarelli and G. Masetti, European Polymer Journal, 1968,
4, 555.
43
V. Jaacks, K. Boehlke and E. Eberius, Die Makromolekulare Chemie,
1968, 118, 354.
44
K. Brzezińska, W. Chwiałkowska, P. Kubisa, K. Matyjaszewski and S.
Penczek, Makromolecular Chemistry, 1977, 178, 2491.
45
D. Engel, PEAXACT 3.0.7-Software for Quantitative Spectroscopy and
Chromatography, S•PACT GmbH, Aachen, Germany, 2013.
10 | Catal. Sci. Techn., 2018, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins