Thermolysis of 3-alkyl-3-methyl-1,2-dioxetanes: Activation parameters and chemiexcitation yields

Article abstract of DOI:10.1002/hc.1069

3-Methyl-3-(3-pentyl)-1,2-dioxetane 1 and 3-methyl-3-(2,2-dimethyl-1-propyl)-1,2-dioxetane 2 were synthesized in low yield by the α-bromohydroperoxide method. The activation parameters were determined by the chemiluminescence method (for 1 ΔH? = 25.0 ± 0.3 kcal/mol, ΔS? = -1.0 entropy unit (e.u.), ΔG? = 25.3 kcal/mol, k₁ (60°C) = 4.6 × 10^-4s^-1. Thermolysis of 1-2 produced excited carbonyl fragments (direct production of high yields of triplets relative to excited singlets) (chemiexcitation yields) for 1: φ^T = 0.2, φ ≤ 0.0005; for 2: φ^T = 0.02 φ^S ≤ 0.0004). The results are discussed in relation to a radical-like mechanism.

Full text of DOI:10.1002/hc.1069

Heteroatom Chemistry
Volume 12, Number 6, 2001
Thermolysis of 3-Alkyl-3-Methyl-1,2-
Dioxetanes: Activation Parameters and
Chemiexcitation Yields
Alfons L. Baumstark, Sean L. Anderson, Chariety J. Sapp,
and Pedro C. Vasquez
Department of Chemistry, Center for Biotechnology and Drug Design, Georgia State University,
Atlanta, GA 30303
Received 02 November 2000; revised 22 January 2001
in an excited state (direct production of high yields
ABSTRACT: 3-Methyl-3-(3-pentyl)-1,2-dioxetane 1
of excited triplets relative to excited singlets) [1].
and 3-methyl-3-(2,2-dimethyl-1-propyl)-1,2-dioxetane
Historically, two mechanistic extremes have been
2 were synthesized in low yield by the -bromohydro-
proposed [1] to describe the thermal decomposition
peroxide method. The activation parameters were de-
of simply substituted dioxetanes: (1) concerted and
termined by the chemiluminescence method (for 1
(2) diradical (Scheme 1). The electron-transfer type
1H‡ = 25.0 0.3 kcal/mol, 1S‡ = 1.0 entropy unit
processes [1,2] that occur for certain peroxides do
(e.u.), 1G‡ = 25.3 kcal/mol, k1 (60 C) = 4.6
not occur readily with simply substituted diox-
4
1
10 s ; for 2 1H‡ = 24.2 0.2 kcal/mol, 1S‡ =
2.0 e.u., 1G‡ = 24.9 kcal/mol, k1 (60 C) = 9.2
etanes. Most mechanistic studies have been inter-
preted to support a diradical-type two-step mech-
anism [3]. A merged mechanism has also been
proposed [4] based on effect of the degree and pat-
tern of methyl substitution. Previous results on alkyl
dioxetanes have shown that relative stability de-
pended on the steric interactions of the substituents
[3]. We report here the synthesis and characteriza-
tion of two 3,3-disubstituted dioxetanes in which the
effective steric size of one substituent is varied.
4
1
10 s .Thermolysisof1–2 produced excited carbonyl
fragments (direct production of high yields of triplets
relative to excited singlets) (chemiexcitation yields for
T
T
S
1:
= 0.02,
0.0005; for 2:
= 0.02,
0.0004). The results are discussed in relation to a dir-
adical-like mechanism. 2001 John Wiley & Sons, Inc.
Heteroatom Chem 12:459–462, 2001
C
INTRODUCTION
The thermolysis of phenyl-substituted and/or alkyl-
substituted dioxetanes has been shown to produce
carbonyl fragments, one of which may be produced
Correspondence to: Alfons L. Baumstark.
Contract Grant Sponsor: Ronald E. McNair Post-baccalaureate
Program (S.A., C.S.).
Contract Grant Sponsor: Alliance for Minority Participation
Program.
Contract Grant Number: OSP-98-01-661-001.
2001 John Wiley & Sons, Inc.
c
SCHEME 1
459

460 Baumstark et al.
TABLE 1 Activation Parameters for the Thermolysis of 3-
Alkyl-3-Methyl-1,2-dioxetanes 1 and 2 in Xylenes
RESULTS
3-Methyl-3-(3-pentyl)-1,2-dioxetane 1 and 3-methyl-
(2,2-dimethyl-1-propyl)-1,2-dioxetane 2 were syn-
thesized in low yield by the Kopecky method [1],
with closure of the corresponding -bromohydro-
peroxides with a base at low temperature (Reaction
1).
1
Diox-
etane 3-Alkyl
1G‡
k1s
a,b
1H‡
1S‡e.u.b kcal/mol b (60 C)
4
4
1
2
3-pentyl 25.0 0.3
neopentyl 24.2 0.3
1.0
2.0
25.3
24.9
4.6 10
9.2 10
aAll errors reported at 95% confidence limits.
bCalculated at 60 C.
(1)
TABLE 2 Chemiexcitation Yields for the Thermolysis of
Dioxetanes 1 and 2 in Xylenesa
T
S
Dioxetane
1
2
0.02
0.02
0.0005
0.0004
The -bromohydroperoxides were synthesized by
the Kopecky procedure [1], treatment of the corre-
sponding alkenes with an electrophilic bromine
source in the presence of concentrated hydrogen
peroxide (caution!) at low temperature in fair yield.
The dioxetanes were purified by low-temperature
aInstrument calibrated with tetramethyl-1,2-dioxetane:
= 0.30;
T
S
= 0.002 (DBA/DPA method at 60 C); error limits 50% of
observed values [5].
1
cence without affecting the kinetics. The yields of
chemiexcitation generated during dioxetane ther-
molysis were determined by the DBA/DPA (chemi-
luminescence) method [1,5]. For both dioxetanes,
thermolysis directly produced relatively high yields
of excited triplets ( T) and low yields of excited sin-
column chromatography and characterized by H
NMR and 13C NMR spectroscopy. Dioxetanes 1–2
were further characterized by analysis of their ther-
molysis products; in all cases, only the expected
cleavage products were produced (Reaction 2).
glets ( S). The T values for both compounds at 60 C
were 2%, while the
S
(2)
values were less than or equal
to 0.05%. The results are summarized in Table 2.
DISCUSSION
The rates of thermolysis of dioxetanes 1–2 were
monitored by the decay of chemiluminescence in-
tensity in aerated xylenes with added fluorescers at
constant temperature ( 0.2 ). The rates of thermal
decomposition were clearly first order for at least
three half-lives and showed no dependence on the
type or amount of added fluorescer. At 60 C, the
Conformational and steric substituent effects on
dioxetane stability have been extensively investi-
gated [1b]. Extensive studies of the relative stabil-
ity of unsymmetric cis/trans pairs of dioxetanes [3a]
and 3,3-cyclic substituted dioxetanes [6] had shown
that steric considerations were important to diox-
etane properties. A key study [3b] on 3-alkyl-3-meth-
yl-1,2-dioxetanes showed that the relative stability
series was: Et 3 < i-Pr 4 < t-Bu 5 (see Table 3). The
major effect showed up in the 1H‡ terms while little
or no observable trends were noted in the 1S‡ terms.
The results showed that the increased stability of the
dioxetanes correlated with increased branching of
the 3-alkyl substituent in the series. Furthermore,
dioxetanes in this series in which the ethyl group was
formally replaced by either n-propyl or n-butyl [3b]
were found to be of the same relative stability (ki-
netically indistinguishable). These were interpreted
to be consistent with steric interactions between
the 3-alkyl group(s) and oxygen-2 in a diradical
process.
4
1
value of k1 for dioxetane 1 was 4.6 0.1 10 s ,
4
1
while that for 2 was 9.2 0.2 10 s . The first-
order rate constants (k1) were determined over a
50 C temperature range. Correlation coefficients
were 0.995 or greater for all cases. The activation
parameters were determined by the Arrhenius
method. Dioxetane 1 with the 3-pentyl substituent
was found to be more stable than 2 with the neopen-
tyl group. The activation parameter data with 95%
confidence limits on errors are shown in Table 1.
Without the presence of added fluorescers, the
thermolyses of dioxetanes 1–2 exhibited only weak
chemiluminescence. Addition of 9,10-dibromoan-
thracene (DBA) or 9,10-diphenylanthracene (DPA)
greatly increased the intensity of chemilumines-

Thermolysis of 3-Alkyl-3-Methyl-1,2-Dioxetanes: Activation Parameters and Chemiexcitation Yields 461
TABLE 3 Activation Parameters for the Thermolysis of 3-
Alkyl-3-Methyl-1,2-dioxetanes 3–6 in Xylenesa
Dioxetane Synthesis
The following two-step procedure for the synthesis
of 3-methyl-3-(3-pentyl)-1,2-dioxetane 1 was em-
ployed for the preparation of the two dioxetanes. A
50 mmol sample of 3-ethyl-2-methyl-1-pentene was
converted to the -bromohydroperoxide by the Ko-
pecky procedure [1]. The -bromohydroperoxide, 1-
1
Diox-
etane 3-Alkyl
1G‡
k1s
b
1H‡
1S‡e.u.b kcal/mol b (60 C)
3
4
4
3
4
5
Ethyl
23.9 0.2
2.7
2.6
0.2
24.8
25.2
25.8
1.0 10
5.8 10
2.4 10
Isopropyl 24.3 0.3
t-Butyl 25.7 0.3
bromo-3-ethyl-2-hydroperoxy-2-methylpentane,
a
aRecalculated parameter data from reference 3b.
bAll errors reported at 95% confidence limits; calculated at 60 C.
clear, viscous oil (caution!) was purified by low tem-
perature ( 78 C) column chromatography (silica
gel, pentane/dichloromethane) (yield 60%): 1H
NMR (CDCl3) 0.98 (t, 6H); 1.19 (s, 3H); 1.22 (m,
4H); 1.66 (m, 1H); 3.63-3.73 (AB, 2H); 7.82 (br
s, 1H); 13C NMR (CDCl3) 13.31, 13.53, 16.80, 22.06,
23.18, 38.69, 45.20, 85.65; for 1-bromo-2-hydrope-
The effective size of substituents in the previous
study was limited [3b]. In the present study, the 3-
alkyl substituents are larger both in total number of
atoms and effective size. Surprisingly, the larger al-
kyl groups in dioxetanes 1 and 2 have little or no
effect over those of the appropriate model com-
pounds. Thus, dioxetane 1 with a 3-pentyl substitu-
ent has activation parameter data essentially iden-
tical to that of 4. Similarly, dioxetane 2 with a
neopentyl group has essentially the same stability as
that of 3 with an ethyl substituent. Clearly, the larger
3-alkyl groups in 1 and 2 do not lead to increased
stabilities. This indicates that buttressing (steric) ef-
fects between the large substituents and position-4
groups are not a factor. Furthermore, the correlation
of the relative stabilities of 1 and 2 with the degree
of branching of the 3-alkyl substituent implies that
the conformation of the substituents is such that the
extra atoms do not interact sterically with oxygen-2.
Molecular mechanics calculations [7] have been
shown to be of value in interpreting dioxetane prop-
erties [1b,3a,f]. No torsion angle changes seem to be
involved in the present cases. The results appear to
be consistent with 3,3-interactions.
1
roxy-2,3,3-trimethylbutane (yield 65%): H NMR
(CDCl3) 1.01 (s, 9H), 1.36 (s, 3H), 1.50 (d, 1H), 1.809
(d, 1H), 3.52 (AB, 2H), 7.68 (s, 1H); 13C NMR (CDCl3)
; 21.32, 31.12, 31.34, 40.57, 46.84, 83.88. Active ox-
bygen content was determined to be 87 6%.
The purified -bromohydroperoxide (14 mmol)
(caution!) was placed in 20 mL of methylene chlo-
ride with rapid magnetic stirring and cooled by an
ice bath. A solution of 5.0 g of potassium hydroxide
in 10 mL of cold, deionized water was added drop-
wise (5 minutes) to the -bromohydroperoxide so-
lution to yield a two-phase mixture. This mixture
was stirred in an ice bath ( 0 C) for 2 hr. The pro-
gress of the reaction was monitored by the amount
of light produced by an aliquot of the organic layer.
After the organic phase was separated, additional ex-
tractions of the aqueous layer with methylene chlo-
ride were required to increase the yields of the diox-
etanes. The pale yellow organic layers were
combined, dried over magnesium sulfate, and fil-
tered. The solvent was removed under reduced pres-
sure, and the dioxetane was purified by column chro-
The chemiexcitation yields ( 2% T) are as ex-
pected for disubstituted dioxetanes [1]. The data
suggest that dioxetanes 1 and 2 are undergoing ther-
molysis by the standard diradical-like mechanism.
Work is in progress on dioxetanes with two large
substituents.
matography (metal-ion free) at
78 C using a
jacketed 1 cm i.d. column packed with 20 g of silica
gel containing 1% Na2 EDTA (pentane). The impure
dioxetane in approximately 1 mL CCl4 was added to
the column and washed with 50 mL of pentane fol-
lowed by successive 50 mL additions of a 5% meth-
ylene chloride/pentane step gradient. Fractions were
assayed for dioxetane content relative to light inten-
sity by placing a small aliquot of each fraction into
a DBA solution in the chemiluminescence apparatus
at 60 C. Fractions containing the most dioxetane
were combined and the solvent removed under re-
duced pressure. The purified dioxetanes 1–2 were
EXPERIMENTAL
All solvents were of reagent grade. 1H and 13C NMR
spectra were recorded on a Varian 300 MHz spec-
trometer. 9,10-Diphenylanthracene (Aldrich) and
9,10-dibromoanthracene (Aldrich) were recrystalli-
zed from xylenes (Aldrich) before use. 3-Ethyl-2-
methyl-1-pentene (Wiley Organics) and 2,4,4-tri-
methyl-1-pentene (Aldrich) were commercially
available and were used without further purification.
yellow oils. The purity was checked by H and 13C
NMR spectroscopy. Dioxetane samples that were less
than 95% pure were passed through the column
a second time. The overall yield of dioxetane was
1
CH analyses were carried out, in house, on the
bromohydroperoxides.
-

462 Baumstark et al.
1–2% in both cases. The dioxetanes were stored in
CCl4 at 30 C or lower. Little decomposition was
noted even after several months of storage. The H
NMR data (CDCl3) are: 1 0.83 (m, 6H); 1.17 (m,
4H), 1.52 (s, 3H), 2.00 (m, 1H), 4.71 (d, 1H),
5.06 (d, 1H); 13C NMR (CDCl3 ) 12.26, 12.56, 20.02,
20.73, 26.66, 49.81, 82.40, 90.22; for 2, 0.99 (s, 9H);
1.84 (s, 3H); 1.89 (d, 1H), 1.97 (d, 1H), 4.80 (d,
1H), 5.22 (d, 1H); 13C NMR (CDCl3 ) 26.19, 30.55,
30.57, 52.32, 83.40, 88.49.
mined by the DBA method as 0.30 at 60 C. All mea-
surements were carried out at 60 C with a constant
concentration of dioxetane. The T and S yields were
calculated by a method that has been discussed in
detail [1]. The concentration of dioxetane was deter-
1
1
mined by H NMR spectroscopy vs. concentration
of added standard. The experimental error by the
DBA/DPA method is estimated to be 50% of ob-
served value.
Product Studies
REFERENCES
The following general procedure was employed for
the thermolysis of dioxetanes 1–2. A solution of diox-
etane (ca. 0.2 M) in CCl4 was heated at 60 C in an
NMR sample tube until the yellow color disap-
peared. In all cases, the expected carbonyl fragments
were the sole products detected by NMR spectros-
copy. The formaldehyde generated from the cleavage
of 1 and 2 was not observed. The carbonyl products
were identified by comparison with authentic
samples.
[1] (a) For reviews, see: Baumstark, A. L.; Rodriguez, A.
In Organic Photochemistry and Photobiology; Hor-
spool, W. M., Song, P.-S., Eds.; CRC Press: New York,
1995; Chapter 27; (b) Baumstark, A. L. In Advances
in Oxygenated Processes; Baumstark, A. L., Ed.; JAI
Press: Greenwich, CT, 1988, Vol. 1, Chapter 2; (c)
Baumstark, A. L. In Singlet Oxygen; Frimer, A., Ed.;
Uniscience CRC: Boca Raton, FL, Vol. II, Chapter 1;
(d) Adam, W. In The Chemistry of Peroxides; Patai,
S., Ed.; Wiley: New York, 1982; Chapter 24; (e) Adam,
W.; Zinner, K.; In Chemical and Biological Genera-
tion of Electronically Excited States; Cilento, G.,
Adam, W., Eds.; Academic Press: New York, 1982;
Chapter 5; (e) Adam, W. In Chemical and Biological
Generation of Electronically Excited States; Cilento,
G., Adams, W., Eds.; Academic Press: New York,
1982; Chapter 4; (f) Kopecky, K. R. In Chemical and
Biological Generation of Electronically Excited
States; Cilento, G., Adams, W. Eds.; Academic Press:
New York, 1982; Chapter 3.
[2] (a) Schuster, G. B.; Schmidt, S. P. Adv Phys Org Chem
1982, 18, 187–238; (b) For recent example, see: Adam,
W.; Trofimov, A. V. J Org Chem 2000, 65, 6474–6478.
[3] (a) Baustark, A. L.; Retter, C. A.; Tehrani, K.; Kellog,
C. J Org Chem 1987, 52, 3308–3311; (b) Bausmtark,
A. L.; Dunams, T.; Catalani, L. H.; Bechara, E. J. H. J
Org Chem 1983, 48, 3713–3716; (c) Baumstark, A. L.;
Vasquez, P. C. J Org Chem 1984, 49, 2640–2642; (d)
Schaap, A. P.; Gagnon, S. D.; Zalika, K. A. Tetra-
hedron Lett 1982, 23, 2943–2946; (e) Richardson, W.
H.; Thomson, S. A. J Org Chem 1985, 50, 1803–1810;
(f) Baumstark, A. L.; Parker, A. N. J Heterocycl Chem
1992, 29, 791–794.
Kinetic Studies
The chemiluminescence monitoring system is essen-
tially identical with that previously described [1].
The reaction cell was jacketed, and the temperature
was maintained by using a constant temperature
bath. The temperature in the cell ( 0.2 C) was
monitored by use of a YSI Model 425C apparatus
with a series 400 probe. The cell was pretreated with
a conc. aq. Na2 EDTA solution and washed with sol-
vent before use. Kinetic experiments were carried
out employing xylenes (mixture of isomers) as sol-
vent. The initial dioxetane concentrations were ca.
4
10 M to avoid induced decomposition. Experi-
ments carried out without added fluorescer and with
3
low concentration ( 10 ) of DBA or DPA were of
the first order for at least three half-lives and showed
no measurable dependence on the type or amount
of added fluorescer. Reproducibility of k1 values was
excellent (better than 5% of value). All k1 determi-
nations had correlation coefficients of greater than
0.999.
[4] Adam, W.; Baader, W. J. J Am Chem Soc 1985, 107,
410–416.
[5] Wilson, T.; Golan, D. E.; Harris, M. S.; Baumstark,
A. L. J Am Chem Soc 1976, 98, 1086–1091.
[6] Baumstark, A. L.; Niroomand, F.; Vasquez, P. C. J Org
Chem 1984, 49, 4497–4500.
[7] Burkert, U.; Allinger, N. L. In Molecular Mechanics,
ACS Monograph 177; American Chemical Society:
Washington, DC, 1982.
Chemiexcitation Yields
The instrument was calibrated with tetramethyl-1,2-
dioxetane [5] by taking the triplet yield ( T) deter-