Photolytic, autocatalyzed decomposition of benzylic dialkoxy disulfides

Article abstract of DOI:10.1016/j.tetlet.2009.01.115

The dialkoxy disulfide moiety has been shown to go through an intramolecular fragmentation to liberate trappable S₂, and can yield an alkoxy radical under photolytic conditions. We have examined a family of benzylic dialkoxy disulfides (X-Ph-CH₂-O-S-S-O-CH₂-Ph-X) under photolytic conditions and observed a correlation of decomposition based upon the substituent. We have been able to show that the decomposition is autocatalyzed and has a parabolic correlation with Swain and Lupton's field constant, F.

Full text of DOI:10.1016/j.tetlet.2009.01.115

Tetrahedron Letters 50 (2009) 1629–1632
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Photolytic, autocatalyzed decomposition of benzylic dialkoxy disulfides
*
DiAndra M. Rudzinski, Ronny Priefer
Department of Chemistry, Biochemistry, and Physics, Niagara University, Niagara University, NY 14109, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The dialkoxy disulfide moiety has been shown to go through an intramolecular fragmentation to liberate
trappable S2, and can yield an alkoxy radical under photolytic conditions. We have examined a family of
benzylic dialkoxy disulfides (X–Ph–CH2–O–S–S–O–CH2–Ph–X) under photolytic conditions and observed
a correlation of decomposition based upon the substituent. We have been able to show that the decom-
position is autocatalyzed and has a parabolic correlation with Swain and Lupton’s field constant, F.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 13 January 2009
Revised 19 January 2009
Accepted 21 January 2009
Available online 25 January 2009
Since the initial synthesis of the dialkoxy disulfides in 18951, lit-
tle has been done on this functionality until recently.2–11 In fact, it
was not until 1997 when the actual linear arrangement, –OSSO–,
was unambiguously confirmed.3 Harpp and co-workers have re-
ported the isomeric form, thionosulfite (–OS(@S)O–), however only
in cyclic structures.4 Structural aspects of dialkoxy disulfides that
are unusual are its short S–S bond (ca. 1.95 Å)5 coupled with its
greatly enhanced barrier of rotation of ca. 18 kcal/mol.6 The nov-
elty of this functionality has been illustrated both thermolytically
and photolytically. When benzylic7 or cubylcarbinolic8 dialkoxy
disulfides were heated in the presence of dienes, a pseudo-Diels–
Alder reaction was observed yielding cyclic di- and tetra-sulfides;
presumably from the liberation of S2 (Fig. 1). Under photolytic con-
ditions, Lunazzi and Placucci reported that the alkoxy radical was
produced which reacted with fullerenes.9 In addition, it was re-
ported that bis-cubylmethyl-dialkoxy disulfide was converted to
bis-cubylmethyl sulfite by ambient light while open to the air.8
Very recently, dialkoxy disulfides have been incorporated into
polymers as a new high energy-storage material.10 Due to the nov-
elty of this group, we report here the photolytic fragmentation of
para-substituted benzylic dialkoxy disulfides and examine the ef-
fect of electron-withdrawing and electron-donating groups of this
functionality.
We initially synthesized seven bis(phenylmethoxy) disulfides
with various electronegative groups in the para position based
upon past procedures.11 All were obtained in good to excellent
yields, {1: bis(benzyloxy) disulfide, BBD; 2: bis(p-methylbenzyl-
oxy) disulfide, BMBD; 3: bis(p-methoxybenzyloxy) disulfide,
BMxBD; 4: bis(p-nitrobenzyloxy) disulfide, BNBD; 5: bis(p-chlo-
robenzyloxy) disulfide, BCBD; 6: bis(p-t-butylbenzyloxy) disulfide,
BtBBD; 7: bis(p-phenylbenzyloxy) disulfide, BPBD}, with BPBD (7)
being newly synthesized (Scheme 1).12 All were successfully puri-
fied via column chromatography and stored in the freezer in the
dark until use.
R
R
R
H
O
O
S
H
O
O
R
R
R
S
S
S
R
R
R
R
S
S
S
S
S
S
+
Figure 1. Intramolecular fragmentation and pseudo-Diels–Alder reaction of dialkoxy disulfides.
* Corresponding author. Tel.: +1 716 286 8261; fax: +1 716 286 8254.
E-mail address: rpriefer@niagara.edu (R. Priefer).
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.01.115

1630
DiAndra M. Rudzinski, R. Priefer / Tetrahedron Letters 50 (2009) 1629–1632
X
X
S2Cl2, Et3N
CH2Cl2, 0oC
X
O
O
OH
S
S
X = H,
Me,
1: BBD
2: BMBD
3: BMxBD
4: BNBD
5: BCBD
6: BtBBD
7: BPBD
MeO,
NO2,
Cl,
t-Bu,
Ph,
Scheme 1. Synthesis of library of dialkoxy disulfides.
Each dialkoxy disulfide was dissolved in toluene (0.1 g/10 mL),
wrapped in aluminum foil (to prevent light from entering) and
argon was bubbled through for 30 min. After this time, a sample
was removed and concentrated in vacuo and a 1H NMR performed.
This was classified as our starting point.13 The mixture was then
placed in a Srinivasan–Griffin–Rayonet Photochemical Reactor,
the wrap was removed, then continually irradiated (350 nm)
and maintained at 35 °C while still under an argon atmosphere.
At various time intervals, 0.5 ml of the mixture was removed,
and 1H NMR was performed. All seven dialkoxy disulfides eventu-
ally completely converted to the corresponding alcohol, however,
at different rates and all were autocatalyzed. Figure 2 illustrates
the photolytic decomposition of both BBD (1) and BMxBD (3). It
is clear that there is an initial lag, presumably due to the stability
of the dialkoxy disulfide. However, as time proceeds there is a
drastic loss of the dialkoxy disulfide. Lunazzi and Placucci postu-
lated that under photolytic conditions, initially a homolytic cleav-
age occurs between the disulfide bond and the subsequent R–O–SÅ
would rapidly convert to the alkoxy radical at temperatures above
ꢀ100 °C.9 As this in a non-linear rate (hence non-zero order
decomposition), the curve suggests that this process is in some
way autocatalyzed.14 Based upon the fact that in all cases a small
amount of aldehyde was formed along with quantitative amounts
of elemental sulfur, two possible competing mechanisms are
proposed here.
100
BBD (1)
90
BMxBD (3)
One possible mechanism (Scheme 2) involves the initial homo-
lytic cleavage of the dialkoxy disulfide (8) providing 9 which would
further cleave to form an alkoxy radical, 10, as the temperature at
which the experiments are done is at 35 °C.9 This reactive
intermediate could abstract a benzylic proton from another start-
ing dialkoxy disulfide, 8, forming the aldehyde, 11, and ultimately
S2 and another alkoxy radical. This pathway is reasonable; how-
ever, it would indicate that a 1:1 ratio should form between alco-
hol and aldehyde. As at most there was a 9:1 ratio of alcohol to
aldehyde an alternate autocatalyzed pathway is also proposed.
In addition to the alkoxy radical (10) that would be formed
initially, reactive sulfur (12) will also be produced. It is theorized
(Scheme 3) that this species could interact with another dialkoxy
disulfide (8) forming two more alkoxy radicals and S3 (13). This
allotrope of sulfur would continue to react until S8 is formed.
Although, both mechanisms are reasonable and possibly compet-
80
70
60
50
40
30
20
10
0
0
2
4
6
8
10
Time (hrs)
Figure 2. Photolytic decomposition of BBD (1) and BMxBD (3).
X
X
X
X
.
2X
O
O
O
O
.
S
S
S
10
9
8
H
S
S
O
O
X
X
.
S
X
S
.
O
O
O
S
S
+
+
+
O
H
X
X
X
11
Scheme 2. Possible mechanism I for the photolytic, autocatalyzed decomposition of dialkoxy disulfides.

DiAndra M. Rudzinski, R. Priefer / Tetrahedron Letters 50 (2009) 1629–1632
1631
X
X
X
2X
O
O
O
.
S
S
S
9
8
. .
S
12
.
S
S
S
O
O
O
S
S
8
X
13
X
10
X
.
.
S
S
S
S7
S9
2S8
+
Scheme 3. Possible mechanism II for the photolytic, autocatalyzed decomposition of dialkoxy disulfides.
30
ing, the large excess of alcohol formed compared to that of alde-
hyde suggests that the mechanism II is the predominant pathway.
To compare the rates of autocatalyzed reactions, they are
reported as the time to ½ life of the starting dialkoxy disulfides.
It is believed that at this time, the rate would be at a maximum
that Figure 2 visually illustrates. Table 1 lists the rates for each of
the dialkoxy disulfides tested.
It can be seen that there is a great difference in rate based upon
which substituent is present. The shortest ½ life for photolytic
decay was compound 3 at 2.6 h, whereas the lengthiest was 4,
requiring 28.3 h. This almost 1000% difference in rate based upon
substituent present, led us to initially examine Hammett’s con-
stant.15 It was not too surprising there were no direct correlations
NO2
100
80
60
40
20
0
NO2
2
25
20
15
10
5
tBu
Cl
OMe
-0.2
0
0.2
Ph
0.4
0.6
0.8
1
1.2
-20
-40
Me
tBu
H
-60
-value
-80
Me
Cl
H
Ph
OMe
0
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
-value
with
r-values, as there is no conjugation of the substituent and the
dialkoxy disulfide. However, we discovered that there was a para-
Figure 3. Half-life of photolytic decomposition of the library of dialkoxy disulfides
versus Swain and Lupton’s field constant, F.
bolic correlation with Swain and Lupton’s field constant, F.16 The
parabolic relationship gave
a curve with an equation of
t1=2 ¼ 59:5F2 ꢀ 49:9F þ 11:1 with a R2 value = 0.868. By examin-
ing the slope of the curve versus the F-value, we do obtain a linear
correlation of dt1=2=dF ¼ 119F ꢀ 49:9. What this suggests is that
at high and low field constants the rate of decomposition is slowed.
With the methoxy substituent having the fastest rate of decompo-
sition, it is proposed that the strong electron-donating potential of
this group possibly stabilizes the radical intermediate through
space and bond transmissions. As the electron-donating potential
decreases, the stability of the radical intermediate is hindered, thus
a slower rate (i.e., 4, BNBD). In addition, with the tert-butyl substi-
tuent (6, BtBBD), the electron density of this group is remote from
the ring, destabilizing any possible radical intermediate. It has
have shown that substituent effects have been related almost so-
lely to the field constant, F17,18 (Fig. 3).
The novelty of dialkoxy disulfides has been illustrated by its
thermolytic and photolytic behavior. This moiety can serve as an
S2 as well as an alkoxy donor, respectively. We have examined
the photolytic decay of a range of para-substituted bis(benzyloxy)
disulfides each of which decomposes at different rates but all auto-
catalytically. In addition, their rate of decomposition is paraboli-
cally correlated to Swain and Lupton’s field constant, F.
Examination of the currently unknown ortho- and meta-substi-
tuted derivatives of dialkoxy disulfides may provide additional
information concerning this pathway.
Å
been reported that formation of X–C6H4–CH2 has no correlation
to Swain and Lupton’s resonance constant, R.16 In addition, studies
Acknowledgment
Table 1
We thank Niagara University and the Niagara University Aca-
demic Center for Integrated Science for financial support.
Half-life for library of dialkoxy disulfides.
Compound #
Substituent
½ life (h)
References and notes
1
2
3
4
5
6
7
–H
6.3
11.5
2.6
28.3
6.5
–Me
–OMe
–NO2
–Cl
–t-Butyl
–Ph
1. Lengfeld, F. Chem. Ber. 1895, 28, 449.
2. (a) Tardif, S. L.; Harpp, D. N. J. Org. Chem. 2000, 65, 4791; (b) Cerioni, G.;
Plumitallo, A. Magn. Reson. Chem. 1998, 36, 461; (c) Snyder, J. P.; Nevins, N.;
Tardiff, S.; Harpp, D. N. J. Am. Chem. Soc. 1997, 119, 12685.
3. Lunazzi, L.; Mazzanti, A.; Spagnolo, P.; Degl’Innocenti, A. J. Org. Chem. 1997, 62,
2263.
23.6
3.5

1632
DiAndra M. Rudzinski, R. Priefer / Tetrahedron Letters 50 (2009) 1629–1632
4. Zysman-Colman, E.; Abrams, C. B.; Harpp, D. N. J. Org. Chem. 2003, 68,
7059.
5. Koritsanszky, T.; Buschmann, J.; Luger, P.; Schmidt, H.; Steudel, R. J. Phys. Chem.
1994, 98, 5416.
6. Borghi, R.; Lunazzi, L.; Placucci, G.; Cerioni, G.; Foresti, E.; Plumitallo, A. J. Org.
Chem. 1997, 62, 4924.
7. Tardif, S. L.; Williams, C. R.; Harpp, D. N. J. Am. Chem. Soc. 1995, 117, 9067.
8. Priefer, R.; Farrell, P. G.; Harpp, D. N. Tetrahedron Lett. 2002, 43, 8781.
9. Borghi, R.; Lunazzi, L.; Placucci, G.; Cerioni, G.; Plumitallo, A. J. Org. Chem. 1996,
61, 3327.
quenched with water, and the organic layer was separated. The aqueous phase
was extracted with CH2Cl2 (2 ꢁ 10 mL), and the combined organic was washed
with dH2O, dried over MgSO4, filtered, and concentrated. The crude product
was purified by silica-gel column chromatography with EtOAc/hexanes (10:1)
to furnish the pure BPBD, (7: 1.98 g, 84.7%) as an off-white solid. 1H NMR
(400 MHz, CDCl3) d 4.87, 4.98 (ABq, J = 10.8 Hz, 4H), 7.36–7.38 (m, 2H), 7.43 (d,
J = 8.0 Hz, 8H), 7.59 (d, J = 7.2 Hz, 8H). 13C NMR (100 MHz, CDCl3) d 76.5, 127.1,
127.3, 127.5, 128.8, 129.2, 135.5, 140.6, 141.4. Exact mass calculated for
C26H22O2S2 430.1056, found 430.10611.
13. Due to the thermo instability of some of the dialkoxy disulfides, the typical
starting amounts ranged from 92% to 100%.
14. Galajda, M.; Lente, G.; Fabian, I. J. Am. Chem. Soc. 2007, 129, 7738.
15. McDaniel, D. S.; Brown, H. C. J. Org. Chem. 1958, 23, 420.
16. Swain, C. G.; Lupton, E. C., Jr. J. Am. Chem. Soc. 1968, 90, 4328.
17. Nakayama, C.; Wataya, Y.; Meyer, R. B., Jr.; Santi, D. V. J. Med. Chem. 1980, 23,
962.
10. Su, Y.-Z.; Dong, W.; Zhang, J.-H.; Song, J.-H.; Zhang, Y.-H.; Gong, K.-C. Polymer
2007, 48, 165.
11. Zysman-Colman, E.; Harpp, D. N. J. Org. Chem. 2005, 70, 5964.
12. Typical experimental procedure: To a stirred solution of 4-phenylbenzyl alcohol
(2.00 g, 0.0109 mol) in CH2Cl2 (40 mL) was added Et3N (1.52 mL, 0.0109 mol) at
0
oC followed by cannulation of a solution of S2Cl2 (0.44 mL, 0.0054 mol) in
CH2Cl2 (20 mL) over 5 min. After stirring at 0 °C for 2 h, the reaction was
18. Chakraborty, K.; Devakumar, C. J. Agric. Food Chem. 2006, 54, 1868.