A Photobasic Functional Group

Article abstract of DOI:10.1021/jacs.5b04367

Controlling chemical reactivity using light is a longstanding practice within organic chemistry, yet little has been done to modulate the basicity of compounds. Reported herein is a triazabutadiene that is rendered basic upon photoisomerization. The pH of an aqueous solution containing the water-soluble triazabutadiene can be adjusted with 350 nm light. Upon synthesizing a triazabutadiene that is soluble in aprotic organic solvents, we noted a similar light-induced change in basicity. As a proof of concept we took this photobase and used it to catalyze a condensation reaction.

Full text of DOI:10.1021/jacs.5b04367

Page 1 of 5
Journal of the American Chemical Society
1
2
3
4
5
6
A photobasic functional group
Jie He, Flora W. Kimani & John C. Jewett*
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Department of Chemistry and Biochemistry, 1306 E University Blvd, University of Arizona, Tucson, AZ 85721
Supporting Information Placeholder
ABSTRACT: Controlling chemical reactivity using light is a
longstanding practice within organic chemistry, yet little has
been done to modulate the basicity of compounds. Reported
herein is a triazabutadiene that is rendered basic upon photo-
isomerization. The pH of an aqueous solution containing the
water-soluble triazabutadiene can be adjusted with 350 nm
light. Upon synthesizing a triazabutadiene that is soluble in
aprotic organic solvents we noted a similar light-induced
change in basicity. As a proof of concept we took this photo-
base and used it to catalyze condensation reaction.
a.Chemically blocked precursor
N
N
N
b.Sterically blocked isomer
N
base-blocking
steric wall
non-blocking
steric wall
c.This work. Functionally distinct isomers
The ability to control chemical reactions and reactivity us-
ing light has been of interest to organic chemists since
Klinger’s first forays into photochemical reactions in cloud-
N
X
1
bedeviled Germany. Since its inception, the synthetic power
H
H
of photochemistry has been used to elegantly solve numer-
H
2
,3
ous synthetic riddles, but much less has been done in an
area of chemistry so intrinsic to its core, namely basicity.
Whereas Brønsted photoacids that dissociate more rapidly in
N
X
4
the singlet excited state than the ground state are known,
Figure 1. Photochemically generated bases. a) Upon exposure
to light the protecting group of a masked base decomposes to
reveal a basic nitrogen atom. b) The basic nitrogen atom of a
molecule is obscured by a steric wall that is reversibly swung
away in a photochemically triggered fashion. c) The intrinsic
basicity of a nitrogen-containing functional group is altered by a
photochemical event.
5
there are fewer examples of Lewis photoacids, and an analo-
gous photobase has thus far escaped exposure. Our work fills
in this last remaining gap. To be sure, bases protected with
photo-labile groups have been reported, but generally the
photochemistry is being used to break apart a mask, not
6
,7
change the inherent property of the molecule (Figure 1a).
Following with a more recent trend to utilize the photo-
induced E/Z isomerization of azobenzene compounds,
Hecht cleverly employed a steric wall to block access to a
We recently reported a triazabutadiene scaffold, 1 (Figure
a), which was water-soluble and reacted with hydronium
ions to liberate a protected aryl diazonium species under
physiologically relevant, mild conditions. Inspired by a re-
port by Fanghänel establishing that this general class of com-
2
8
,9
basic nitrogen atom’s lone pair of electrons (Figure 1b).
1
0
To the best of our knowledge, the Lewis basicity of a func-
tional group has yet to be altered with photochemistry (Fig-
ure 1c). Herein we report a water-soluble compound that,
upon photo-irradiation, is rendered significantly more basic
in a reversible fashion. Furthermore, an organic-soluble de-
rivative is used to catalyze a Henry reaction in a light de-
pendent manner.
1
1
pounds readily photoisomerize to a more reactive Z-form,
we irradiated a pH 9 buffered solution of 1. Indeed, with a
simple hand-held UV lamp (“365 nm,” measured at 350 nm)
we observed complete consumption of 1 after only a few
hours. The non-irradiated reaction under similar condi-
tions was stable for days. We hypothesized that if a two-
electron process (illustrated in Scheme S1) is occurring, then
1
2
1
-Z must be more basic than 1-E. Initial NMR analysis of
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 5
samples post-irradiation clearly showed cyclic guanidine 2
Figure 2a), but we did not observe evidence of a benzene
a.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0.8
Me
Me
(
0.0
diazonium species or phenol/azobenzene products derived
therefrom. We conducted a trapping experiment with resor-
cinol because it can serve a dual role as a radical scavenger,
and a trap for the benzene diazonium species that could be
formed. In the event we added an excess of resorcinol to a
pH 9 borate-buffered solution of 1 and irradiated the mix-
ture with light (Scheme S2). Gratifyingly, the known azo-
benzene, Sudan Orange G, was formed in a 65% yield (ver-
H
0.2
3
50 nm
Me
1.0
Me
Me
Me
DMSO
N
N
N
N
N
N
N
N
H
N
N1
N
#
= NOE signal
(normalized)
N3
3
3
KO3S
1-E
KO3S
1-Z
AcOH (via N1 protonation)
AcOH (via N3 protonation)
b.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
E
1
3
sus 4% for the non-irradiated reaction).
E
Mes
We observed that non-buffered solutions became basic
upon exposure to light in a time-dependent manner. This
increase in pH correlated with the appearance of 2, which is
mildly basic. To assess the basicity of 1-Z, a non-buffered
solution of water was irradiated with light while monitoring
the pH in real-time. Compound 1 was dissolved in water and
the pH was adjusted to be ~ 9 to prevent the formation of 2
in the absence of light. Upon exposing the solution to 350
nm light, the solution rapidly spiked up to a pH of 9.8 over
the course of several minutes and upon longer exposure
slowly became more basic (black line, Figure 2). At a low
concentration of 1 these spikes were followed by a decrease
in pH (green line). These data suggest that 1-Z is more basic
than 2, a fact that is corroborated by the finding that 1 is able
N
H
N
E
N
3
2
KO3S
1 (no light) in DMSO
after 1 hour 350 nm light
After addition of AcOH
2 minutes)
c.
E
1
E
(
Z
E
2
E
Z
2
2
E
Z
1
4
to deprotonate water absent of hydronium ions. Once the
more basic Z isomer is formed it can go on to form 2 or re-
vert to 1. Assuming that 10% of 1 is isomerized in the first 30
E
1
5
4.0
3.8
3.6
2.6
2.4
2.2
2.0
1.8
seconds of irradiation and complete consumption of 1 at
5 minutes the relative pK values of 1-Z and 2 in water were
2
b
ppm
found to be 5.1 and 6.0, respectively (see the supporting in-
formation for the calculations).
Figure 3. a) Compound 1 in DMSO was irradiated with UV
light for 1.75 hours and analyzed by NOESY. Signals were nor-
malized to the strongest NOE observed (numbers in circles)
taking into consideration relative concentrations of E and Z. b)
& c) A solution of 1 in DMSO (black spectrum) was treated
with UV light for 1 hour and analyzed by NMR (blue spec-
trum). The solution was treated with 10 equivalents AcOH and
an NMR spectrum was collected after 2 minutes (red spec-
trum). The Z isomer was consumed within this time.
a.
Mes
Ph
Mes
Mes
N
N
N
N
N
N N
N
N
H
N
Ph
N
N
N
3
3
KO3S
KO3S
3
KO3S
2
1
(more basic)
b.
10.
4
mM
mM
In spite of the jump in pH suggesting its existence, 1-Z
remained uncharacterized by NMR in water, even at low
temperatures. Attributing this to a low steady-state concen-
tration of 1-Z that both decomposes and thermally reverts
the inability to directly observe the elusive 1-Z in water
prompted a swapping of solvents to DMSO. As noted previ-
9.8
9.6
9.4
1
9.2
0.5 mM
1
0
ously, 1 is quite stable to an excess of acetic acid in DMSO,
9
.0
.8
showing only 12% degradation to 2 over 14 hours at room
temperature. Upon irradiation with light 1 was completely
converted to 2 over the same timeframe. Gratifyingly, irradi-
ation of 1 in acid-free DMSO for ~2 hours provided a mix-
ture of two compounds, starting material and what appeared
to be its isomer (1-E and 1-Z, respectively). Moreover, un-
like what we assumed was occurring in water, the thermal
reversion from Z to E in pure DMSO is slow, with a half-life
8
5
10
15
20
25
time (minutes)
Figure 2. a) Compound 1 is rendered more basic upon expo-
sure to light. b) Constant UV irradiation of dilute solutions
shows a quick spike in pH (presumably from 1-Z) followed by a
leveling off resulting from complete conversion of 1to 2.
ACS Paragon Plus Environment

Page 3 of 5
Journal of the American Chemical Society
on the order of days. This long half-life enabled NOESY ex-
and slow thermal isomerization. As such we used what we
had learned from 1 to select a proof of concept reaction.
Based on the apparent pK of 3 we matched pK to con-
b a
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
periments that delivered a clear picture of the identity and
conformation of the two isomers (Figure 3a & Figure S1).
To bolster our contention that the new species in DMSO
was the phantom Z-isomer, AcOH was added to a 70:30 (E :
Z) mixture in the absence of light and within 2 minutes the Z
isomer was gone (Figure 3b & c). The speed of its disappear-
ance was remarkable, all that remained was a mixture of the E
isomer and newly formed 2 in a 92:8 ratio. We draw from
this two conclusions: 1) 1-Z is significantly more basic than
1-E; 2) the differentially basic N1 and N3 nitrogen atoms
provide two pathways of reaction for 1-Z, and proton-
catalyzed reversion to 1-E (likely from N1-protonation) is
26
densation substrates. A Henry reaction between nitroethane
(4) and p-nitrobenzaldehyde (5) was chosen to demon-
2
7
strate the virtues of 3 (Figure 4b). Indeed, the reaction
between 4 and 5 occurred rapidly at room temperature in a
light and catalyst dependent manner (Figure 4c). The reac-
tion with 25 mole % 3 in the absence of light was exceedingly
slow. Likewise, the reaction with light but no catalyst also
failed to proceed (Figure S5). The cyclic guanidine was not
observed during a post-reaction analysis of the components
from a 25 mole % 3 run; indicating that the Z-isomer of 3 is
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
16
28
dominant. Indeed Fanghänel had previously reported that
likely to be the catalytically active species in solution. Wor-
an N1 alkylated triazabutadiene was able to isomerize ther-
ried that a trace of cyclic guanidine could catalyze the reac-
tion we treated a mixture of 4 and 5 with 1 % guanidine and
found that it did indeed catalyze the reaction to near comple-
tion with 3 hours (Figure S5). In order to discriminate be-
tween 3-Z and guanidine as the relevant catalyst that we
were observing, we ran a reaction with light and turned it off
once the slope matched that of the guanidine catalyzed reac-
tion. True to our model, the reaction slowed and stopped,
presumably as 3-Z became protonated and thermally isom-
erized (Figure S5). Interestingly, the reaction photo-
catalyzed with 3 was significantly faster than the same reac-
17
mally.
To the best of our knowledge this phenomenon of an
isomerization-induced pK change has escaped the eyes of
b
watchful chemists, as it almost did ours. Unlike the case
where Hecht’s compound is rendered basic upon irradiation
8
by way of moving a steric wall (Figure 1b) , it is unlikely that
steric factors play a significant role in our chemistry, especial-
ly in water. Our current rationale for this effect is that the E
isomer has a stable conjugated π-system with alternating
18,19
non-π involved lone pairs of electrons
whereas the Z
8
tion reported by Hecht.
isomer lacks these features and has a higher energy HOMO.
This higher energy HOMO stems from the a break in conju-
gation between the phenyl ring and the nitrogen π-system of
the Z isomer (consistent with NOE data, Figure 3a, and dim-
inution of UV-vis spectrum compared with the E isomer,
Figure S3), and repulsion between the two adjacent lone
pairs of electrons on the N1-N2 nitrogen atoms. These fac-
tors render the N1-nitrogen atom more electron rich than in
the E isomer, and is thus more basic. This model is bolstered
Mes
Mes
a.
b.
Ph
Ph
water
soluble
N
N
N N
N
N
N
N
N
N
organic
soluble
Mes
3
KO3S
1
3
3
NO2
NO2
(mol %)
NO2
NO2
hv
THF
rt
15
O
by N-NMR data collected on the same general class of
OH
90%
NMR)
4
(excess)
5
20-22
compounds in the early 1980s.
These studies observed a
(
significant upfield shift in the N1 nitrogen signal with a min-
imal effect on the N3 nitrogen upon examining the E vs. Z
isomer. Data from the same study suggests that a similar ef-
fect is true of simple azo-benzenes, but these compounds are
c.
120
80
2
3
generally so much less basic than 1 that the switching oc-
curs in a less synthetically useful or observable range. This
hypothesis is corroborated by derivatives of 1 with electron-
withdrawing groups that were synthesized and irradiated
with light in a buffered solution (Figure S4). Whereas these
compounds decomposed to form 2, when tested in pure
water an initial rapid spike in pH was not observed, simply a
slow increase due to 2 forming.
3
hv
-
+
+
+
25 mol %
mol %
0 mol %
5 mol %
40
5
1
2
0
20
40
60
80
100
time (minutes)
2
4
Seeing the potential utility of 1 as a photo-catalytic base
in the context of organic reactions we desired to test its abil-
ity to catalyze a simple condensation reaction. Due to the
limited organic solubility of 1, we synthesized 3, a com-
Figure 4. Using the photobase as a catalyst. a) Structures of
water-soluble 1 versus organic soluble 3. b) The Henry reaction
between 4 and 5 was carried out at room temperature and vary-
ing amounts of catalyst. c) The reactions were monitored by
TM
2
5
ReactIR , following consumption of aldehyde 5. Red line = 25
pound previously reported by Bielawski (Figure 4a). With
3 we noted similar light-induced acid sensitivity in DMSO
mole % 3 with no light. Orange line = 5 mole % 3 + light. Green
line = 10 mole % 3 + light. Blue line = 25 mole % 3 + light.
ACS Paragon Plus Environment

Journal of the American Chemical Society
In closing, we have reported a remarkable transformation
REFERENCES
Page 4 of 5
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
whereby a molecule is rendered basic upon photo-
isomerization. Steric arguments cannot be employed to ex-
plain the dramatic change, but instead loss of conjugation in
a π-system and co-planar lone pairs of electrons on adjacent
nitrogen atoms appear to be the key difference. These find-
ings are guiding ongoing divergent efforts in our research to:
(
1) Klinger, H.; Standke, O. Ber. Dtsch. Chem. Ges. 1891, 24, 1340.
(2) Bochet, C. G. J. Chem. Soc., Perkin Trans. 1 2002, 125.
3) Bach, T.; Hehn, J. P. Angew. Chem. Int. Ed. Engl. 2011, 50, 1000.
(
(4) Solntsev, K. M.; Huppert, D.; Agmon, N.; Tolbert, L. M. J. Phys. Chem. A
2000, 104, 4658.
(5) Lemieux, V.; Spantulescu, M. D.; Baldridge, K. K.; Branda, N. R. Angew.
Chem. Int. Ed. 2008, 47, 5034.
(
(
6) Edson, J. B.; Spencer, L. P.; Boncella, J. M. Org. Lett. 2011, 13, 6156.
7) Cameron, J. F.; Frechet, J. M. J. J. Org. Chem. 1990, 55, 5919.
1
) better control the release of diazonium species; and, 2)
design a photobase with long-term stability. The operational
ease by which these compounds are rendered basic should
allow them to find their way into other applications. Fur-
thermore, we hope that these findings will promote a better
understanding of factors that should be assessed when exam-
ining Lewis basicity and will inspire researchers in the area of
catalysis, both organic and organometallic.
(8) Peters, M. V.; Stoll, R. S.; Kuhn, A.; Hecht, S. Angew. Chem. Int. Ed. Engl.
2008, 47, 5968.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
(
(
9) Gostl, R.; Senf, A.; Hecht, S. Chem. Soc. Rev. 2014, 43, 1982.
10) Kimani, F. W.; Jewett, J. C. Angew. Chem. Int. Ed. 2015, 54, 4051.
11) Fanghänel, E.; Hohlfeld, J. J. Prakt. Chem. 1981, 323, 253.
(12) We have since made a home-built UV-LED system for more controlled
irradiation. The device is described in the supporting information.
(
13) This yield is similar to that for the reaction non-irradiated reaction carried
out in a pH 6 buffer.
14) Indeed, compound 1 decomposed in the presence of 1.0 N NaOH upon
(
ASSOCIATED CONTENT
exposure to light.
(15) A conservative estimate based on 8% isomer observed after irradiating a
solution in DMSO for 1 minute and analyzing by NMR.
Supporting information including synthesis and character-
ization of new compounds is available free of charge via the
Internet at http://pubs.acs.org.”
(
16) In aqueous solutions we observed a drop in pH upon turning off the UV
light. This effect was amplified with the addition copper sulfate, presumably
acting as a Lewis acid (Figure S2).
(
17) Fanghänel, E.; Poleschner, H.; Radeglia, R.; Hänsel, R. J. Prakt. Chem.
1977, 319, 813.
18) Dorsch, H.-T.; Hoffmann, H.; Hansel, R.; Rasch, G.; Fanghänel, E. J. Prakt.
Chem. 1976, 318, 671.
AUTHOR INFORMATION
Corresponding Author
(
(
(
19) Khramov, D. M.; Bielawski, C. W. J. Org. Chem. 2007, 72, 9407.
20) Radeglia, R.; Stinger, T. H.; Fanghanel, E. J. Prakt. Chem. 1984, 3, 511.
*
jjewett@email.arizona.edu
Notes
(21) Fanghänel, E.; Simova, S.; Radeglia, R. J. Prakt. Chem. 1981, 1981, 239.
(22) Simova, S.; Fanghänel, E.; Radeglia, R. Org. Mag. Res. 1983, 21, 163.
(
1
(
(
The authors declare no competing financial interests.
23) Tian, Y.; Isono, N.; Kawai, T.; Umemura, J.; Takenaka, T. Langmuir
988, 4, 693.
24) Nielson, B. M.; Bielawski, C. W. ACS Catal. 2013, 3, 1874.
ACKNOWLEDGMENTS
25) Tennyson, A. G.; Moorhead, E. J.; Madison, B. L.; Er, J. A. V.; Lynch, V.
We thank Profs. Richard Glass, Jon Njardarson, Elisa To-
mat, Michael Heien & Jeffrey Pyun for fruitful discussions
regarding this work. We thank Drs. Neil Jacobsen & Jixun
Dai for assistance with analysis of NMR data, and to the NSF
for a generous departmental instrumentation grant for the
NMR facility (CHE-0840336). We also thank Dr. Kevin Bao
for assistance with UV-LED array design and development.
This work was funded in part by an ACS PRF grant to JCJ &
start-up funds provided by the AZ Board of Regents to JCJ.
M.; Bielawski, C. W. Eur. J. Org. Chem. 2010, 6277.
26) Z-1 is able to deprotonate AcOH in DMSO, so it can remove protons from
(
substrates of at least pKa of 12.6.
(27) Henry, A. C. R. Acad. Sci. Ser. C. 1895, 120, 1265.
(
28) If exposed to light for prolonged times (>4 hours) degradation of 3 is
observed.
ACS Paragon Plus Environment

Page 5 of 5
SYNOPSIS TOC
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
R
N
Z
N
Ph
N
N
N
350 nm
R'
R
N
proton
source
N
N N
N
Ph
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
R'
E
5
ACS Paragon Plus Environment