Synthesis and low temperature spectroscopic observation of 1,3,5-trioxane-2,4,6-trione: The cyclic trimer of carbon dioxide

Article abstract of DOI:10.1021/acs.joc.6b00647

1,3,5-Trioxane-2,4,6-trione (cyclic trimer of CO₂) is the product of a four-step synthesis: chlorination of isobutyraldehyde; cyclotrimerization of 2-chloro-2-methylpropanal; dehydochlorination of 2,4,6-tris(2-chloropropan)-2-yl-1,3,5-trioxane; ozonolysis at -80 °C of 2,4,6-tri(propan-2-ylidene)-1,3,5-trioxane. This trioxane-trione is detected in solution at temperatures between -80 to -40 °C, and its conversion to CO₂ is monitored by ¹³C NMR and FTIR. The CO₂ trimer has a half-life of approximately 40 min at -40 °C.

Full text of DOI:10.1021/acs.joc.6b00647

Subscriber access provided by University of Sussex Library
Article
Synthesis and Low Temperature Spectroscopic Observation of
1,3,5-Trioxane-2,4,6-Trione: The Cyclic Trimer of Carbon Dioxide.
Michael J. Rodig, Arthur W. Snow, Paul Scholl, and Simon Rea
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.6b00647 • Publication Date (Web): 16 May 2016
Downloaded from http://pubs.acs.org on May 18, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Organic Chemistry is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
7
Synthesis and Low Temperature Spectroscopic Observation of 1,3,5-Trioxane-2,4,6-
Trione: The Cyclic Trimer of Carbon Dioxide.
8
9
Michael J. Rodig,†,‡ Arthur W. Snow,*,† Paul Scholl§, Simon Rea§
†Chemistry Division, Naval Research Laboratory, Washington, D. C. 20375
‡Current address: Eastman Chemical Company, Kingsport, TN 37662
§MettlerꢀToledo AutoChem, Inc., Columbia, MD 21046
arthur.snow@nrl.navy.mil
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[TOC Abstract Graphic]
[Abstract]
ABSTRACT: 1,3,5ꢀTrioxaneꢀ2,4,6ꢀtrione (cyclic trimer of CO2 ) is the product of a fourꢀstep
synthesis: chlorination of isobutyraldehyde; cyclotrimerization of 2ꢀchloroꢀ2ꢀmethylpropanal;
dehydochlorination of 2,4,6ꢀtris(2ꢀchloropropan)ꢀ2ꢀylꢀ1,3,5ꢀtrioxane; ozonolysis at ꢀ80°C of
2,4,6ꢀtri(propanꢀ2ꢀylidene)ꢀ1,3,5ꢀtrioxane. This trioxaneꢀtrione is detected in solution at
temperatures between ꢀ80 to ꢀ40°C, and its conversion to CO2 is monitored by 13C NMR and
FTIR. The CO2 trimer has a halfꢀlife of approximately 40 min at ꢀ40°C.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 26
1
2
3
4
Introduction
5
6
7
8
As a product of combustion and respiration whose accumulation in the atmosphere has
become a cause for significant concern, carbon dioxide has been the subject of much research
directed at its reutilization. Various approaches toward this CO2 reutilization goal have been
described in excellent reviews over the past two decades.1ꢀ4 Important processes involve
reduction with hydrogen,5 coupling with other small molecules,6 incorporation into polymers7
and artificial photosynthesis.8 The main products include fuels, solvents, chemical intermediates
and polymers. The efficiency of these commercial processes in terms of reagent usage is
relatively low with respect to the fraction of CO2 incorporated into the product; the highest being
for urea (57%), and decreasing for salicylic acid (36%) and methanol (10%).3 This could be
raised to 100% if a CO2 selfꢀfixation chemistry could be developed. Ideally with a sufficient
input of energy, CO2 would react with itself to yield a liquid or solid product from which this
energy could be extracted when needed for useful work. Such chemistry has been the subject of
theoretical calculation for structures representing the linear polymer and cyclic oligomers of
CO2.9ꢀ14 With respect to thermodynamic stability, the cyclic trimer has been described as
“feasible” although energetically less stable than three CO2 molecules by 27 kJ/mol per CO2
unit.10 Regarding kinetic stability of the cyclic trimer toward fragmentation to CO2, calculated
barriers for this decomposition have ranged from activation energies of 61 to 172 kJ/mol
depending on the computational method with calculated halfꢀlives days to milliseconds at
ambient conditions and substantially longer at lower temperatures.9,12,14 The cyclic trimer of
CO2 has also been proposed as a lowꢀenergy intermediate in the transformation of CO2 to an
extended solid.14 The formation of an orthocarbonate extended covalent structure of
interconnected sixꢀmembered rings was predicted by model calculation with the finding of a
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 3 of 26
The Journal of Organic Chemistry
1
2
stabilization energy that increased with molecular size.10,11 Later experimental work found under
extreme pressure/temperature (40 GPa/1800°K), CO2 will transform to a metastable extended
solid which has been characterized as a Phase V form of CO2 with a sigma bonded quartzꢀlike
structure.15 It has also been proposed that sorption of CO2 into the isolated nanoscale confined
spaces of sulfurꢀ or nitrogenꢀtreated porous carbon at 30 bar pressure can produce a polymeric
structure of carbon dioxide16 as has been reported for other molecules in nanoꢀconfined spaces.17
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The 1,3,5ꢀtrioxaneꢀ2,4,6ꢀtrione structure of the CO2 cyclic trimer, 1, may represent an
important intermediate or product in the selfꢀfixation of this gas. Theoretical studies on this
molecule have indicated a possibility of kinetic stability at room temperature9 and as well as a
possibility for it to be thermodynamically feasible.10 To date, no experimental evidence has been
reported for its existence. The objective of this work is to synthesize compound 1 and to make
an assessment of its stability. The approach is that of a model compound synthesis where the
trioxane ring is first generated from substituted aldehydes and then the peripheral carbonyl
structures are incorporated at low temperature in the final step. As will be shown, compound 1
does not possess the stability for facile isolation and storage.
Results
A retrosynthetic route for the synthesis of compound 1 is illustrated in Scheme 1. The
three steps are an oxidative cleavage of a triolefinꢀtrioxane, 2, a dehydrochlorination of
αꢀchloroaldehyde trimer, 3, and a trimerization of two candidate αꢀchloroaldehydes, 4 and 5.
The chemical properties of triolefinꢀtrioxane compounds are not wellꢀknown, nor has the
susceptibility of this type of olefin toward oxidative cleavage been reported. This synthesis
requires that all three olefin bonds of 2 be cleaved thereby adding great importance to the yield
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 4 of 26
1
2
3
4
5
6
7
8
9
of this reaction. Thus, both the unsubstituted 2,4,6ꢀtrisꢀmethyleneꢀ1,3,5ꢀtrioxane, 2b, and the
hexamethylꢀsubstituted analog 2,4,6ꢀtrisꢀisopropylidineꢀ1,3,5ꢀtrioxane, 2a, were both targeted
for comparative purposes as well as a strategy of having an alternative if difficulties are
encountered en route to 1. In Scheme 1 the absence or presence of methyl substituents on the
exocyclic olefin structure 2 may affect the oxidative cleavage to compound 1 and will determine
detectable byproducts, both of which are issues of interest.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 1. Retrosynthetic route toward the CO2 trimer.
The synthesis of the hexamethyl triolefinꢀtrioxane 2a is depicted in Scheme 2.
Isobutyraldehyde, 6, was chlorinated by a procedure using sulfuryl chloride18 to afford the
corresponding αꢀchloroaldehyde, 4. This aldehyde was then trimerized to 2,4,6ꢀtris(1ꢀchloroꢀ1ꢀ
methylethyl)ꢀ1,3,5ꢀtrioxane, 3a, in good yield by a procedure using concentrated H2SO419,20 and
isolated as a crystalline solid (see Supporting Information for crystal structure). The third step
of dehydrohalogenating 3a to the hexamethyl triolefinꢀtrioxane 2a was problematic. In our
hands a patented procedure21 for preparation of compound 2a (Route A, reaction of 3a with
sodium hydroxide dispersed in dimethylheptanol at 185°C) produced no yield of product. An
alternate approach of using potassium tertꢀbutoxide in THF (Route B) afforded 2a in good yield.
Compound 2a prepared via Route B is a liquid while that reported from Route A was a solid with
a very broad melting point (54ꢀ62°C).21 A thorough analysis of this product was conducted (1Hꢀ
NMR, 13CꢀNMR, IR, UVꢀVis, HRꢀMS, elemental analysis and differential scanning calorimetry;
ACS Paragon Plus Environment

Page 5 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
see Experimental and Supporting Information for data and discussion), and these results are in
good agreement with the structure assigned to compound 2a.
7
8
9
Scheme 2. Synthesis of substituted triolefinꢀtrioxane.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Attempts to synthesize the unsubstituted triolefinꢀtrioxane 2b met with considerable
challenges (Scheme 3). Trimerization of the commercially available aqueous solution of 5 with
concentrated H2SO4 produce low yields (26%) of 3b and often inconsistent results. Multiple
attempts to dehydrohalogenate trioxane 3b, using either the previously reported conditions21
(Route A) or our modified procedure (Route B) failed to yield the desired unsubstituted triolefinꢀ
trioxane 2b. Confident in the structure of the methyl substituted triolefinꢀtrioxane and the
reproducibility of our procedure; our focus was placed on the oxidation chemistry of 2a.
Scheme 3. Attempted synthesis of unsubstituted triolefinꢀtrioxane.
The initial approach to oxidative transformation of the 2a olefin bonds to carbonyl bonds
involved preliminary experiments to determine if the three CꢀC double bonds could be
oxidatively cleaved in a stepwise manner. Cleavage using oxo metals such as OsO422,23 or
combinations of RuCl324,25 and coꢀoxidants was unsuccessful as was using singlet oxygen
generated from ozone and phosphine.26 Byproducts interfered with analyses and obscured
isolation of anticipated products of unknown stability.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 6 of 26
1
2
A complete oxidation approach using ozonolysis was then explored.27ꢀ29 The Criegee
mechanism30 for this chemistry as applied to compound 2a is depicted in Scheme 4. Ozone and
the olefin bonds of 2a initially at low temperature may or may not form a π or chargeꢀtransfer
complex, 7,31ꢀ33 followed by formation of the primary ozonide, 8, which breaks down to the
carbonyl oxide “Criegee intermediate”, 9, and the complementary carbonyl compound. In an
inert solvent these moieties recombine to form the secondary ozonide, 10, which is sometimes
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4. Criegee Ozonolysis Mechanism applied to Compound 2a (for clarity,
chemistry is represented on one of the three carbonꢀcarbon double bonds).
7
8
2a
CH3OH
9
+
10
S(CH3)2
11
1
SO(CH )
+
+
3 2
sufficiently stable for isolation but which is usually reacted with a reductant to yield the desired
pair of carbonyl compounds as indicated by the left route in Scheme 4. Initially, our selected
reaction conditions (ꢀ78 °C, saturated solution of 2a in CH2Cl2 with a flowing 5% O3/O2 mixture
until the blue coloration of excess ozone persists followed by purging the O3 then by the addition
of (CH3)2S to quench the reaction and warming to ambient temperature) did not result in the
detection of the desired trioneꢀtrioxane product 1. However, the resultant oxidation byꢀproduct,
ACS Paragon Plus Environment

Page 7 of 26
The Journal of Organic Chemistry
1
2
(CH3)2SO, was observed consistent with the action of some ozonide intermediate on Me2S.34 As
some secondary ozonides are known to be sufficiently stable for ambient isolation,35 limited
attempts were made to isolate an ozonide intermediate by withholding the addition of the
reducing agent, but a stable and structurally characterized product was not obtained. The
intramolecular constraint of three olefin bonds initially held in close proximity when the primary
ozonide and the reactive carbonyl oxide are generated from one of them is conjectured to be a
route to a complex product mixture.
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The reaction path on the right side of Scheme 4 was then pursued. Methanol is employed
as a wellꢀknown and quantitative trapping agent for the carbonyl oxide intermediate and converts
it into the relatively stable 2ꢀmethoxypropꢀ2ꢀyl hydroperoxide, 11.36 The hypothesis for using
methanol in this manner is that, when the primary ozonide 8 fragments, the carbonyl oxide
intermediate will incorporate the exocyclic methyl groups of 2a (structure 9) and not incorporate
the trioxane ring structure of 2a. In favor of this hypothesis is the reported ozonolysis of 1,1ꢀ
dimethoxyethylene preferentially yielding dimethyl carbonate and the carbonate functional
group’s resistance to further reaction with carbonyloxides.37 The presence of methanol is also
useful in preventing the selfꢀcoupling of the acetone carbonyl oxide 9 to form dimeric and
oligoperoxidic compounds.38 As a proof of concept, the ozonolysis reaction of 2a was carried
out in a similar manner as before in a chilled (1:1) mixture of CH2Cl2 and CH3OH in the absence
of any reducing agent. After the reaction warmed to ambient temperature, NMR analysis of the
evaporated mixture revealed hydroperoxide 11 (average of 62% over two reactions) as the
dominant product. This observation indicates that the oxidative cleavage is, in fact, occurring as
planned since the trapped carbonyl oxide is concomitant with the generation of the new
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 8 of 26
1
2
3
4
5
6
7
8
9
carbonate functionality. Due to the lack of any other observable product, it seems possible that
the trioneꢀtrioxane 1 is formed but subsequently degrades. If this is the case, further support for
its existence was needed.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To obtain evidence for the existence of the trioneꢀtrioxane 1, a variable temperature 13Cꢀ
NMR experiment was conducted. A solution of 2a dissolved in a 1:1 CD2Cl2:CD3OD mixture
was prepared in an NMR cell, and an initial spectrum is obtained. This solution is then cooled to
ꢀ78 °C, and a 5% O3/O2 stream was passed into this solution until the blue coloration of excess
ozone was observed. The cell was then purged with oxygen and quickly transferred into a preꢀ
cooled NMR probe at ꢀ80 °C. Spectra were acquired at progressive temperatures
of ꢀ80, ꢀ60, ꢀ40, ꢀ20 and 0 °C. These spectra are presented in Fig. 1 with expansions of important
resonances depicted in inserts. The initial 20°C spectrum displayed the three resonances of
triolefinꢀtrione 2a at chemical shifts of 15.0 (ꢀCH3), 87.5 (=C(CH3)2) and 145.2 (=C(O)2) ppm.
After cooling and completion of the ozone treatment, these resonances disappeared, and the three
resonances attributable to the hydroperoxide compound 11 appeared at 22.0 (ꢀCH3), 49.1
(ꢀOCH3, obscured by the CD3OD) and 104.1 (CꢀOOH) ppm in excellent agreement with those
previously reported for this compound.39 At this ꢀ80 °C temperature an additional new
resonance appeared at 156 ppm which we assign to the carbon in the structure of 1. This
chemical shift is consistent with that of other cyclic carbonates.40 This resonance persisted on
warming to ꢀ60 and ꢀ40 °C, but on further temperature increase, it was rapidly displaced by
another resonance at 125 ppm. This new signal is consistent with dissolved CO2 in chlorinated
solvents and was subsequently verified at the end of this experiment by introduction of CO2 and
observed growth of this signal. The observation of these coupled signals indicates that a
ACS Paragon Plus Environment

Page 9 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fig. 1. 13C NMR spectra from the triolefinꢀtrione – ozone reaction. The lowest spectrum is the
substrate 2a CD2Cl2/CD3OD solution prior to ozone treatment. The series spectra above were
recorded after the ꢀ78°C ozone saturation and purging from this solution at successive 20°C
temperature increments from ꢀ80 to 0°C. The compound structures for which resonances are
assigned are depicted at the left of each spectra and the chemical shifts discussed in the text are
indicated by the dashed lines and values at the top of this figure. Those resonances of particular
importance have a magnified insert.
carbonate type structure, presumably the trioneꢀtrioxane 1, was present at low temperatures then
decomposes to its monomeric form as CO2 gas.41,42 As further support for assignment of the
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 10 of 26
1
2
3
4
5
6
7
8
9
156 ppm resonance to the carbon nucleus in compound 1, SCF GIAO magnetic shielding tensor
calculations were carried out on this compound, CO2 and TMS. The calculated chemical shifts
relative to TMS are 150.2 and 123.9 ppm respectively which are in reasonably good agreement
with the 156 and 125 experimental values.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To complement this NMR evidence for the existence and stability of this trioneꢀtrioxane,
compound 1, an in situ FTIR experiment was conducted under essentially the same conditions.
Instead of an NMR cell, the reaction was conducted in a 25 mL flask which was fitted with an
ozone inlet and the ATR probe of a ReactIR™ reaction monitoring system. With the capability
of generating several spectra per minute, this method is better suited for quantitative kinetic
analysis, but it does require sophisticated software for spectral deconvolution and tracking of the
time dependent profiles of species in the reaction medium. Control spectra of the CH2Cl2 and
CH3OH solvents, triolefinꢀtrioxane 2a and O3 reagents and hydroperoxide 11 and CO2
byproducts were input to the software, and the reaction is monitored by repetitive recording of
FTIR spectra in real time. The composite spectra are deconvoluted for individual species, and
the corresponding trend lines depict the quantities of reactants, intermediates and products as a
function of time. The upper half of Fig. 2 presents trend lines for the in situ ꢀ80 °C ozonolysis of
2a, depicting this reagent, intermediates and the trioneꢀtrioxane product 1. Three distinct
intermediates appear to form and are consumed during the course of this reaction. The product
once formed has an indefinite stability at ꢀ80 °C. In the lower half of Fig. 2 this product is
monitored as a function of time for 1.5 hr intervals at ꢀ80, ꢀ60 and ꢀ40 °C. Once the temperature
and baseline have stabilized, the decomposition of 1 in the ꢀ40 °C interval is readily apparent and
is accompanied by the generation of CO2. This is in good agreement with the NMR experiment
ACS Paragon Plus Environment

Page 11 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fig. 2. FTIR trend lines for in situ monitoring of (above) ozonolysis conversion of compound 2a
to compound 1 at ꢀ80 °C in 1:1 CH2Cl2:CH3OH and of (below) stability of compound 1 during
20 °C stepped interval warming from ꢀ80 °C to 0 °C.
Fig. 3. Infrared spectrum of trioneꢀtrioxane 1 derived from in situ ozonolysis reaction spectra
and from thermal stability – variable temperature spectra.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 12 of 26
1
2
3
4
5
6
7
8
of Fig. 1. From the slope of the trend line in this region (indicated by the dashed window in Fig.
2), a halfꢀlife of 40 min is estimated at this temperature which is in reasonable agreement with a
range of halfꢀlives calculated by various theoretical methods.9
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The infrared spectrum of the trioneꢀtrioxane 1 was deconvoluted from the accumulated
spectra during the ozonolysis reaction mixture and from the postreaction spectra as the
temperature was increased to assess the thermal stability of this compound. As presented in Fig.
3, these spectra duplicate each other closely. Assignment of the major bands may be correlated
with the anhydride and carbonate functional groups experimentally observed for molecules of
closely related structure and with calculated from the MP2 and DFT(B3LYP) levels of theory.
Experimentally, infrared bands characteristic of the cyclic anhydride functional group as part of a
six membered ring (a doublet of comparable intensity derived from the symmetric and
asymmetric C=O stretching modes in the 1750 to 1850 cmꢀ1 range and a single CꢀO band in the
1160ꢀ1260 cmꢀ1 range)43 are consistent with those observed. As a closely related structure,
pyranꢀ2,4,6ꢀtrione (acetonedicarboxylic anhydride, 3ꢀoxoglutaric anhydride) displays such bands
at 1810, 1740 and 1245 cmꢀ1 44 which are in relatively close conformity with the 1821, 1733 and
1272 cmꢀ1 bands in Fig. 3. The carbonate group displays C=O and CꢀO bands in a similar range
although less sensitive to ring strain and with a different intensity ratio between the symmetric
and antisymmetric C=O bands.43 The broad bands in the 1380ꢀ450 cmꢀ1 range are difficult to
assign as most model compounds have CꢀH bands in that range. However, in ring systems
containing carboxyl groups and an absence of CꢀH bonds, such bands are observed and
correlated with outꢀofꢀplane deformations.43 While the functional group character of compound
ACS Paragon Plus Environment

Page 13 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
1 combines structural elements of both anhydride and carbonate, its infrared spectrum more
resembles the anhydride. Vibrational calculations for compound 1 have been reported at the
MP2(FC)/6ꢀ31G* level with a scaling factor of 0.94.9 Infrared frequencies calculated for the
C=O symmetric and asymmetric stretching modes are 1991 cmꢀ1 (corrected to 1872 cmꢀ1; relative
intensity 0) and 1918 cmꢀ1 (corrected to 1803 cmꢀ1; relative intensity 89) respectively, and those
for the CꢀO symmetric and asymmetric ring stretching modes are 1302 cmꢀ1 (corrected to 1224
cmꢀ1; relative intensity 100) and 1043 cmꢀ1 (corrected to 980 cmꢀ1; relative intensity 3)
respectively. Comparable infrared frequency calculation at the BL3YP/6ꢀ31G* level without a
scaling factor have been reported as 2015 cmꢀ1 (relative intensity 26), 1931 cmꢀ1 (relative
intensity 764) and 1286 cmꢀ1 (relative intensity 742) for the respective C=O symmetric and C=O
asymmetric and CꢀO ring stretching modes.13 These calculated band frequencies, when
corrected by a scaling factor, correlate with those of the experimentally observed bands centered
at 1821, 1733 and 1272 cmꢀ1 in Fig. 3. The 1821 and 1733 cmꢀ1 bands display weak unresolved
satellite bands. This feature has been observed in a study of 5ꢀ and 6ꢀmembered lactone
compounds comparing experimental and MP2 and DFT calculated spectra and was attributed to
a Fermiꢀresonance interaction between the C=O vibration and overtone or combination bands.45
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
As final evidence in this study for the formation of 1 from 2a and its subsequent
decomposition into CO2, quantitative yield measurements of the hydroperoxide byproduct 11
from the ozonolysis and of the CO2 evolved during the postreaction warming were conducted.
The yield measurement for compound 11 was conducted on a 200 mg scale following the
ozonolysis procedure of the NMR and IR experiments and produced a 67% yield. A small
quantity of residual mixed impurities was also observed which may be due an incomplete
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 14 of 26
1
2
3
4
5
6
7
8
9
trapping of the carbonyl oxide 9 in Scheme 4 or to some decomposition or side reaction of this
hydroperoxide byproduct 11. Measurement of the yield of CO2 evolved from 1 after warming to
room temperature was conducted by two procedures utilizing infrared spectroscopy and the
reaction conditions employed above in the measurement of compound 11. The first procedure
utilized an infrared solution cell with CaF2 windows to quantify the intensity CO2 band at 2339
cmꢀ1. Immediately after the ozonolysis of 1 was completed and warmed to near room
temperature, a measured aliquot of the reaction mixture was transferred to fill the IR solution cell
and the spectrum recorded. An analytical solution of CO2 in the 1:1 CH2Cl2:CH3OH solvent was
then prepared with a CO2 concentration having a comparable 2339 cmꢀ1 band intensity in its IR
spectrum. The CO2 yield derived from 1 was then determined from a proportionality between
the integrated intensities of this band in the ozonolysis reaction and analytical spectra. This
method resulted in a 37% yield based on quantity of 2a and on assumptions of quantitative
conversions of the ozonolysis and of the CO2 evolution reactions. There was also the possibility
of CO2 escape from the solution IR cell during transfer and measurement of the condensed phase
reaction mixture. A second procedure was conducted using a gas IR cell with an attached
cylindrical well in which the ozonolysis reaction and subsequent warming of the product were
conducted. After completion of the ozonolysis, the reaction solution was cooled from ꢀ78 °C
to ꢀ195 °C, the cell was evacuated to remove the O3/O2 from the headspace and then slowly
warmed to 20 °C while monitoring gas phase species in the isolated cell. This procedure was
repeated with a quantitatively prepared CO2/CH2Cl2:CH3OH solution (volume and CO2
concentration calculated for a 100% yield from 2a) placed in the cylindrical well. The yield of
CO2 measured in the gas phase for the ozonolysis reaction was 61% relative to that measured
from the CO2/CH2Cl2:CH3OH analytical solution. This 61% yield of CO2 appears to
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 15 of 26
The Journal of Organic Chemistry
1
2
3
4
complement the above 67% yield of the hydroperoxide byproduct 11. Based on these
5
6
measurements, a 64% yield is supported for the formation of the trioneꢀtrioxane 1 in Scheme 4.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Discussion
While the selfꢀcondensation of CO2 into larger molecules is attractive for the perspective
of a high efficiency in CO2 utilization and perhaps energy storage, the kinetic stability of such
compounds is a key issue. For the case of the 2,4,6ꢀtrioneꢀ1,3,6ꢀtrioxane 1, the result of this
work indicates that its preparation from CO2 would be very tedious and would require
maintenance of low temperatures, high pressures and a yet to be defined catalyst. This molecule
has received significant theoretical analyses, and estimates of its kinetic stability or halfꢀlife are
useful information even if variable. Older calculation methods appear to significantly over
estimate these paremeters,9 while those based on density functional theory are low (41 msec/298
°K) and to be regarded as lower limits.12 Halfꢀlife calculation based on CCSD coupled cluster
theory (~100 sec/298 °K)14 is in reasonable good agreement with the 40 min/233 °K
experimental value.
A comparison of this kinetic stability of 1 with known compounds of related structure is
worthy of comment. Molecules composed of multiple CO2 units are known for the oligoꢀ
carbonate series (CH3)3COꢀ[C(O)O]nꢀC(CH3)3, n = 1, 2, 3. All are stable at ambient conditions
and have progressively decreasing decomposition temperatures: n = 1 (>200 °C)46; n = 2 (100
°C)47; n = 3 (65 °C)48. For polymer analogs of undefined end groups claims are made of subꢀ
ambient stability both theoretically9,13 and experimentally16. While a kinetic instability increases
with an increasing ꢀ[C(O)O]nꢀ chain length, a chain length of three units can be considered stable
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 16 of 26
1
2
3
4
5
6
7
8
9
at ambient conditions. However, at such a short chain length end group effects play a role, and it
has been reported that changing the tertꢀbutyl to isoꢀpropyl end group diminished stability and
precluded purification.48 Ring strain may also play a role in the stability of 1. The calculation
optimized the conformation of 1 as planar with small departures (4°) of the OꢀCꢀO and CꢀOꢀC
bond angles from 120°14 which would indicate the ring strain is relatively modest as expected.
By comparison, the CO2 cyclic dimer, 1,3ꢀdioxetanedione has a halfꢀlife that is shorter than that
for 1 by seven orders of magnitude.14 Another structural feature of interest with respect to
stability is the orientation of the bonding between CO2 units. If one CO2 unit is reversed as is the
case in 1,2ꢀdioxetanedione where the connecting linkages of the CO2 units are CꢀC and OꢀO
bonds, the halfꢀlife is reported to increase by many orders in magnitude compared with the 1,3ꢀ
dioxetanedione structure.14 In a linear structure the CO2 unit orientation reversal has been
realized in diꢀtertꢀbutylperoxyoxalate, (CH3)3CꢀOꢀOC(O)ꢀC(O)OꢀOꢀC(CH3)3, where CꢀC and
OꢀO bonds are important determinants of compound stability. This compound, while not as
stable as the linear tricarbonate analog above, has a halfꢀlife of over 400 sec at 60 °C49 which is
significantly greater than that of 1. The stability of reverse CO2 orientation (headꢀtoꢀhead and
tailꢀtoꢀtail placements) in poly(carbon dioxide) has been investigated for thermodynamic stability
relative to the headꢀtoꢀtail placement with the latter more stable by 160 kJ/mol.50 Finally, the
kinetic stability gains by successively replacing a C=O moiety by a CH2 group in the 1,3,5ꢀ
trioneꢀtrioxane structure of 1 have been calculated with the result of a halfꢀlife increase of 104
with each substitution.12 The trends described in this paragraph, from the perspective of
increasing kinetic stability of molecular adducts derived from CO2 additions, appear to indicate
that CꢀC bond formation between a CO2 unit and a small nonꢀpolar coꢀreactant would be
advantageous.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment

Page 17 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
Conclusion
7
8
9
The work reported here provides substantial evidence for the existence and kinetic
stability of 2,4,6ꢀtrioneꢀ1,3,5ꢀtrioxane 1 prepared in a model compound synthesis for the cyclic
trimer of carbon dioxide. This compound is synthesized in four steps: chlorination of
isobutyraldehyde 6; trimerization of chloroꢀisobutyraldehyde 4; dehydrochlorination of chloroꢀ
isobutylꢀtrioxane 3a; and ozonolysis of the hexamethyl triolefinꢀtrioxane 2a to the trioneꢀ
trioxane 1. During the ozonolysis reaction, it is critical that methanol be present to trap the
carbonyl oxide intermediate 9 and to prevent side reactions. Analysis of 1 during its formation
and decomposition to CO2 by both 13C NMR and in situ midꢀFTIR spectroscopy is consistent
with assigned spectral features. The FTIR spectral features are interpreted to indicate that 1 is
more related to a cyclic anhydride than to a cyclic carbonate. Its yield from the ozonolysis of 2a
is 64% as derived from yield measurements of the hydroperoxide byproduct 11 and of the CO2
decomposition product. Compound 1 has a halfꢀlife of 40 minutes at ꢀ40 °C in 1:1
CH3OH:CH2Cl2 solution is stable at ambient pressures and decomposes quantitatively to carbon
dioxide above ꢀ40 °C.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental Section
General Information. All reactions were conducted in ovenꢀdried glassware under an inert
atmosphere of dry argon unless otherwise specified. Solvents were refluxed and distilled over
Na/benzophenone ketyl or CaH2 prior to use. All commercial reagents were used as received
without further purification. Ozonolysis was performed using an LGꢀ7 CD Laboratory Ozone
Generator by Ozone Engineering. Analytical TLC was done using preꢀcoated silica gel 60 F254
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 18 of 26
1
2
3
4
5
6
7
8
9
plates and visualized with either UV light (254 nm), KMnO4 or Cerium Molybdate staining
solution. NMR spectra were recorded at 300 MHz for 1H NMR and at 75.4 MHz for 13C NMR.
All chemical shifts for 1H and 13C NMR spectroscopy were referenced to residual signals from
CDCl3 (1H) 7.26 ppm and (13C) 77.2 ppm or from CD2Cl2 (13C) 53.8 ppm. FTIR spectra were
obtained in transmission with sample supported on a NaCl plate under a nitrogen purge of 40
cm3minꢀ1 and in ATR reflection with sample supported on a diamond window. The ozonolysis
reaction was monitored by real time in situ FTIR spectra were obtained with a MettlerꢀToledo
ReactIR™ 15 using a silver halide fiber optic probe assembly containing a silicon or diamond
ATR sensor at the tip of the probe. Mass spectral analyses of parent ion were conducted with an
ESIꢀTOF instrument. Elemental analysis was performed by a commercial contractor.
Compounds 3a, 3b, 4 and 11 were prepared using known procedures. Compound 2a was
prepared using a modified procedure.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthetic Procedures
2,4,6-tris(chloromethyl)-1,3,5-trioxane (3b).51 In an atmosphere of air, chloroacetaldehyde
(≥45% in H2O) (1.95g, 25.0 mmol) was added to a round bottom flask equipped with a stir bar
and cooled to 0°C via ice bath. Conc. H2SO4 was added drop wise with vigorous stirring until a
precipitate occurs. The suspension was then poured over ice (ca. 50 mL). The suspension was
then vacuum filtered and washed liberally with water. The material was then reꢀsuspended in a
2:1 mixture of ethanol and ice cold water followed by vacuum filtration to remove any colored
material. The material was then dried on high vacuum to give a white powder (0.23 g, 26%). Mp
85ꢀ86 °C (Lit.51 86 °C); 1H NMR (CDCl3, 300 MHz) δ = 3.61 (d, J = 4.71 Hz, 6 H), 5.16 (t, J =
ACS Paragon Plus Environment

Page 19 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
4.71 Hz, 3 H) ppm; 13C NMR (CDCl3, 75 MHz) δ = 43.2, 99.8 ppm; IR νmax (solid) 1127, 1024,
760 cmꢀ1. Xꢀray crystal structure (Tables S1ꢀS6, Supporting Information).
7
2-chloro-2-methylpropanal (4).18 Isobutyraldehyde (23.2g, 0.32 mmol) was added to a twoꢀ
neck round bottom flask equipped with a magnetic stir bar, pressure equalizing funnel and reflux
condenser. Sulfuryl chloride (33.8g, 0.32 mmol) was then added to the addition funnel and added
dropwise at a rate to maintain an external oil bath temperature below 40°C. The evolved gases
were vented through a trap containing water (ca. 50 mL). Once the sulfuryl chloride was added
the reaction was heated to 45°C with stirring for 1.5 h then allowed to stir at RT for an additional
2.5 h. The addition funnel and reflux condenser were replaced with a stopper and a distillation
head fitted with a receiving flask. The crude material was distilled under vacuum (95 °C, 115
mm Hg). A second distillation was necessary to achieve the pure material as a colorless liquid
(12.48 g, 37%). 1H NMR (CDCl3, 300 MHz) δ = 1.56ꢀ1.87 (m, 6 H), 9.42 (s, 1 H) ppm; 13C
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NMR (CDCl3, 75 MHz) δ = 25.8, 69.3, 194.9 ppm.
2,4,6-tris(2-chloropropan-2-yl)-1,3,5-trioxane (3a).18 In an atmosphere of air, aldehyde 4
(20.0g, 0.18 mol) was added to a round bottom flask equipped with a stir bar. Conc. H2SO4 was
added drop wise with vigorous stirring until a precipitate occurs. The suspension was then
poured into ice cold water (ca. 50 mL). Large chunks of material were finely pulverized with a
large glass stir rod. The suspension was then vacuum filtered and washed liberally with water.
The material was then reꢀsuspended in a 2:1 mixture of ethanol and ice cold water followed by
vacuum filtration to remove any colored material. The material was then dried on high vacuum
to give a white powder (12.2 g, 61%). Mp 106ꢀ108 °C (Lit.18 106ꢀ107 °C); 1H NMR (CDCl3, 300
MHz) δ = 1.54ꢀ1.63 (m, 18 H), 4.86ꢀ4.90 (m, 3 H) ppm; 13C NMR (CDCl3, 75 MHz) δ = 26.7,
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 20 of 26
1
2
67.2, 103.0 ppm; IR νmax (solid) 1154, 1122, 1107 cmꢀ1. Xꢀray crystal structures (Tables S7ꢀS12,
3
4
5
6
Supporting Information).
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2,4,6-tri(propan-2-ylidene)-1,3,5-trioxane (2a). A 100ꢀmL Schlenk flask equipped with a
magnetic stir bar was charged with KOtꢀBu (10.48 g, 93.8 mmol) in a glove box under an
atmosphere of dry argon. The flask was then removed from the glove box and connected to a
Schlenk line under argon. Dry THF (35 mL) was then added via syringe, this mixture was then
cooled to 0°C via ice bath and a solution of trioxane 3a (4.97 g, 15.6 mmol) dissolved in dry
THF (10 mL) was added slowly to the cooled reaction mixture. After stirring 15 min at 0°C, the
ice bath was removed and the mixture warmed to room temperature. The flask was then fitted
with a reflux condenser and heated to 75°C for 18 h. The mixture was then cooled to room
temperature, and the THF removed under vacuum. Et2O (20 mL) was then added and the
resultant suspension was vacuum filtered over a bed of celite, washing with Et2O (20 mL). The
solvent was then removed under vacuum, and the crude product was purified by vacuum
distillation (10 mm Hg at 185°C) to yield the title compound as a colorless liquid (2.32 g, 71%).
1H NMR (CDCl3, 300 MHz) δ = 1.63 (s, 18 H,) ppm; 13C NMR (CDCl3, 75 MHz) δ = 15.0, 86.9,
144.7 ppm; IR νmax (liquid) 2991, 2919, 2863, 1726, 1284, 1212 cmꢀ1; UV (CH3CN) λmax = 210
nm (ε = 1.57x104 L/moleꢁcm); HRMS (ESI) m/z calcd for C12H18O3 [M+H]+: 211.1334; found:
211.1342; Anal. Calcd. for C12H18O3: C, 68.54; H, 8.68; O, 22.83. Found: C, 68.48; H; 8.76.
Note: in an earlier published procedure21 for compounds 2a and 2b difficulties were encountered
in both obtaining the products and reproducing characterization data (see Supporting Information
for detailed discussion and data).
ACS Paragon Plus Environment

Page 21 of 26
The Journal of Organic Chemistry
1
2
2-hydroperoxy-2-methoxypropane (11).39 Compound 11 was synthesized by ozonolysis of
3
4
5
6
7
2,3ꢀdimethylbutene according to literature procedure 39 1H NMR (CDCl3, 300 MHz) δ = ppm
.
1.35 (s, 6 H), 3.27 (s, 3 H), 8.83 (s, 1 H); 13C NMR (CDCl3, 75 MHz) δ = ppm 22.0, 49.1, 105.2.
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Variable Temperature 13C-NMR of 2a-Ozonolysis Reaction. A solution of triolefinꢀtrioxane
2a (30.0 mg, 0.143 mmol) in 0.5 ml 1:1 CD2Cl2:andCD3OD solvent mixture was prepared in a 5
mm NMR cell and cooled to ꢀ78 °C with a dry ice – acetone bath in a tall Dewar flask with a
viewing window feature. A 5% O3/O2 mixture from the ozone generator (10V coronal discharge)
was delivered to the cold solution through a long syringe needle such that the bubbling rate was
rapid but not excessive. The ozone was delivered until the solution just developed a pale blue
color. The NMR cell was then purged of any excess O3 with pure O2 until the solution was again
colorless. The cell was then rapidly transferred to the NMR preꢀcooled to ꢀ80 °C. Two
acquisitions were taken at 20 deg. temperature intervals starting at ꢀ80 °C and warming to ꢀ20 °C
with a scan count of 64 and a delay time of 30 sec.
Variable Temperature IR of in situ 2a-Ozonolysis Reaction. Control experiments for
individual spectra of solvents, reagents, and byproducts were first conducted using a Mettlerꢀ
Toledo ReactIR™ 15 system with the ATR silicon probe to record spectra of CH2Cl2; CH3OH,
1:1 CH2Cl2:CH3OH mixture, 2a/1:1 CH2Cl2:CH3OH; O3/1:1 CH2Cl2:CH3OH, 11/1:1
CH2Cl2:CH3OH. The ozonolysis reaction was conducted in a 25 ml round bottom flask with
necks to accommodate the ReactIR fiber optic probe and the ozone dispersion tube. A solution
of the triolefinꢀtrioxane 2a (599.7 mg, 2.80 mmol) dissolved in 10 ml (1:1 CH2Cl2:CH3OH) was
introduced to the reaction flask and cooled under Ar atmosphere to ꢀ80 °C by an acetone bath
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 22 of 26
1
2
3
4
5
6
7
8
9
regulated by the feedback control of a refrigerated cooling coil. The ozone generator was then
set to deliver 5% O3/O2 at a modest bubbling rate. The experiment was initiated with IR spectral
reaction monitoring of the ozonolysis reaction. Within the first hour the triolefinꢀtrioxane
reagent 2a was consumed, and the bath temperature was successively raised in 20 °C intervals
for 1.5 hr periods to generate the data displayed in Fig. 2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Byproduct 11 Yield from 2a-Ozonolysis Reaction. A solution of 2a (210 mg, 1.01mmol) in
20 ml 1:1 CH2Cl2:CH3OH solvent mixture was cooled to ꢀ78 °C and treated with 5% O3/O2 via
dispersion tube until a persistent blue coloration was observed (15min). After warming to 20 °C,
solvents were removed by rotary evaporation to yield 215 mg (2.03 mmol, 67%).
Byproduct CO2 Yield from 2a-Ozonolysis Reaction. Two IR transmission spectroscopy
based techniques were used: a solution measurement and a gas phase measurement. The first
procedure utilized an infrared solution cell with CaF2 windows to quantify the intensity CO2
band at 2339 cmꢀ1. Immediately after the ozonolysis of 2a as described above was completed
and warmed to near room temperature, a measured aliquot of the reaction mixture was
transferred by syringe to fill the IR solution cell and the spectrum was recorded. An analytical
solution of CO2 in the 1:1 CH2Cl2:CH3OH solvent was then prepared via transfer of CO2 from a
calibrated bulb to provide a solution spectrum with a comparable 2339 cmꢀ1 band intensity. This
method resulted in a 37% CO2 yield based on quantity of 2a and on assumptions of quantitative
conversions of the ozonolysis and of the CO2 evolution reactions. There was also the possibility
of CO2 escape from the solution IR cell during transfer and measurement of the condensed phase
reaction mixture. A second procedure was conducted using a gas IR cell with an attached
cylindrical well in which the ozonolysis reaction and subsequent warming of the product were
conducted. After completion of the ozonolysis, the reaction solution was cooled from ꢀ78 °C
ACS Paragon Plus Environment

Page 23 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
to ꢀ195 °C, the cell was evacuated to remove the O3/O2 from the headspace and slowly warmed
to 20 °C while monitoring gas phase species. This procedure was repeated with an analytically
prepared CO2/1:1 CH2Cl2:CH3OH solution (volume and CO2 concentration calculated for a
100% yield from 2a) placed in the cylindrical well, frozen, evacuated and subsequently warmed
to 20 °C in the isolated cell. The yield of CO2 measured in the gas phase for the ozonolysis
reaction was 61% relative to that measured from the CO2/CH2Cl2:CH3OH analytical solution.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Acknowledgments. American Society of Engineering Education, ASEE, is gratefully
acknowledged for a Resident Research Associateship at NRL for M. J. R. Drs. Andrew P. Purdy
and James P. Yesinowski are acknowledged with special appreciation for assistance in
establishing conditions for 13C NMR data acquisition. Drs. Brett I. Dunlap and Igor V.
Schweigert are gratefully acknowledged for SCF GIAO calculation of 13C NMR chemical shifts.
Dr. Damon A. Parrish appreciatively acknowledged for the crystal structures of compounds 3a
and 3b. This work was supported by the Office of Naval Research. Funding for this work was
provided to M. J. R. and A. W. S. by the Office of Naval Research through the Naval Research
Laboratory Base Program.
Supporting Information Available: Discussion of synthesis of compounds 2a and 2b; NMR,
IR and UV spectra of compound 2a; DSC thermogram for compound 2a; and crystal structure
data for compounds 3a and 3b. This material is available free of charge via the Internet at
http://pubs.acs.org.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 24 of 26
1
2
3
4
References
5
6
7
8
(1) Halmann M. M. Chemical Fixation of Carbon Dioxide: Methods for Recycling CO2 into
Useful Products; CRC Press: Boca Raton, 1993.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(2) Song, C.; Gaffney, A. M.; Fujimoto, K., Eds. CO2 Conversion and Utilization: ACS Symp.
Ser. 809; American Chemical Society, Washington, 2002.
(3) Aresta, M., Ed. Carbon Dioxide Recovery and Utilization; Kluwer Academic Publishers:
Dordrecht, 2003.
(4) Peters, M.; Köhler, B.; Kuckshinrichs, W.; Leitner, W.; Markewitz, P. Müller, T. E.
ChemSusChem 2011, 4, 1216ꢀ1240.
(5) Benson, E. E.; Kubiak, C. P.; Sathrum, A. J.; Smieja, J. M. Chem. Soc. Rev. 2009, 38, 89ꢀ99.
(6) Omae, I. Catalysts Today 2006, 115, 33ꢀ52.
(7) Sugimoto, H.; Shohei, I. Pure Appl. Chem. 2006, 78, 1823ꢀ1834.
(8) Collings, A. F.; Critchley, C., Eds. Artificial Photosynthesis: From Basic Biology to
Industrial Application; WileyꢀVCH Verlag: Weinheim, 2005.
(9) Lewars, E. J. J. Mol. Struct.: THEOCHEM 1996, 363, 1ꢀ15.
(10) Bodor, N.; Huang, M.ꢀJ. Chem. Phys. Lett. 1992, 192, 386ꢀ389.
(11) Bodor, N.; Buchwald, P. Chem. Phys. Lett. 2000, 319, 645ꢀ649.
(12) Shirel, M. L.; Pulay, P. J. Am. Chem. Soc. 1999, 121, 8544ꢀ8548.
(13) Frapper, G.; Saillard, J.ꢀY. J. Am. Chem. Soc. 2000, 122, 5367ꢀ5370.
(14) Dunlap, B. I.; Schweigert, I. V.; Purdy, A. P.; Snow, A. W.; Hu, A. J. Chem. Phys. 2013,
138, 134304.
(15) Iota, V.; Yoo, C. S.; Cynn, H. Science 1999, 283, 1510ꢀ1513.
ACS Paragon Plus Environment

Page 25 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
(16) Hwang, C.ꢀC.; Tour, J .J.; Kittrell, C.; Espinal, L.; Alemany, L. B.; Tour, J. M. Nat.
Commum. 2014, 5. DOI: 10.1038/ncomms4961.
7
8
9
(17) Rousseau, D. L.; Proto, S. P. S.; Science 1970, 167, 1715ꢀ1719.
(18) Stevens, C. L.; Gillis, B. T. J. Am. Chem. Soc., 1957, 79, 3448–3451
(19) Wakasugi, T.; Tonouchi, N.; Miyakawa, T.; Ishizuka, M.; Yamauchi, T.; Itsuno, S.; Ito, K.
Chem. Lett. 1992, 171ꢀ172.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(20) Starr, J.; Vogl, O. J. Macromol. Sci.ꢀChem. 1978, A12, 1017ꢀ1039.
(21) Boudakian, M. M.; Lapkin, M. Substituted Trioxanes. US Patent 3,560,526, 1971.
(22) Kim, B. M.; Chung, J.; Kim, S. Tetrahedron Lett. 2001, 52, 1363–1367.
(23) Borhan, B.; Narayan, R. S.; Travis, B. R. J. Am. Chem. Soc., 2002, 124, 3824–3825.
(24) Oget, N.; Lecuire, J.; Mieloszynski, J. Meux, E.; Zimmermann, F. Tetrahedron Lett. 2005,
46, 3201–3203.
(25) Zhang, C.; Yang, D. J. Org. Chem. 2001, 66, 4814–4818.
(26) Li, C.; Lu, H. Org. Lett. 2006, 8, 5365–5367.
(27) Bailey, P. S. Ozonation in Organic Chemistry; Academic Press: New York, 1978; Vols. I
and II.
(28) Van Ornum, S. G.; Champeau, R. M.; Pariza, R. Chem. Rev. 2006, 106, 2990–3001.
(29) Caron, S.; Dugger, R. W.; Ruggeri, S. G.; Ragan, J. A.; Ripin, D. H. B. Chem. Rev. 2006,
106, 2943–2989.
(30) Criegee, R. Angew. Chem. Int. Ed. 1975, 14, 745–752.
(31) Hull, L. A.; Hisatsune, I. C.; Heicklen, J J. Am. Chem. Soc., 1972, 94, 4856ꢀ4864.
(32) Nelander, B.; Nord, L J. Am. Chem. Soc., 1979, 101, 3769ꢀ3770.
(33) Singmaster, K. A.; Pimentel, G. C. J. Phys. Chem., 1990, 94, 5226ꢀ5229.
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 26 of 26
1
2
3
4
5
6
7
8
9
(34) Stotter, P. L.; Eppner, J. B. Tetrahedron Lett. 1973, 26, 2417ꢀ2420.
(35) Murray, R. W. Tech. Methods Org. Organomet. Chem. 1969, 1.
(36) Bunnelle, W. H. Chem. Rev. 1991, 91, 335 – 362.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(37) Wojciechowski, B. J.; Chiang, CꢀY.; Kuczkowski, R. L. J. Org. Chem., 1990, 55, 1120–
1122.
(38) Barton, M.; Ebdon, J. R.; Foster, A. B.; Rimmer, S. J. Org. Chem. 2004, 69, 6967 – 6973.
(39) Dussault, P.; Sahli, A. J. Org. Chem. 1992, 57, 1009ꢀ1012.
(40) Wang, JꢀL.; He, LꢀN.; Dou, XꢀY.; Wu, F. Aust. J. Chem. 2009, 62, 917ꢀ920.
(41) Kosugi, Y.; Imaoka, Y.; Gotoh, F.; Rahim, M. A.; Matsui, Y.; Sakanishi, K. Org. Biomol.
Chem. 2003, 1, 817ꢀ821.
(42) The chemical shift of CO2 dissolved in a (1:1) mixture of CD2Cl2 and CD3OD was
confirmed independently in our laboratory and was found to match that of the experimentally
derived value.
(43) Avram, M. and Mateescu, GH. Infrared Spectroscopy: Applications in Organic Chemistry;
WileyꢀInterscience: New York, 1972, pp. 434ꢀ437, 464ꢀ467.
(44) Kiang, A. K.; Tan, S. F.; Wong, W. S. J. Chem. Soc. C, 1971, 2721ꢀ2726.
(45) Breda, S.; Reva, I.; Fausto, R. J. Mol. Struct. 2008, 887, 75ꢀ86.
(46) Choppin, A. R.; Rogers, J. W. J. Am. Chem. Soc. 1948, 70, 2967ꢀ2967.
(47) Howe, J. H.; Morris, L. R. J. Org. Chem. 1962, 27, 1901ꢀ1902.
(48) Dean, C. S.; Tarbell, D. S. Chem. Comm. 1969, 728ꢀ729.
(49) Bartlett, P. D.; Benzing, E. P.; Pincock, R. E. J. Am. Chem. Soc. 1960, 82, 1762ꢀ1768.
(50) Baughman, R. H.; Iqbal, Z.; Eckhardt, H. Macromolecules 1988, 21, 1832ꢀ1838.
(51) Iwata, T.; Wasai, G.; Saegusa, T.; Furukawa, J. Makromol. Chem. 1964, 77, 229–236.
ACS Paragon Plus Environment