Photo-Electroswitchable Arylaminoazobenzenes

Article abstract of DOI:10.1021/acs.joc.1c00763

Azobenzenes appended with a redox-active arylamino group (redox auxiliary, RA) are prepared and shown to undergo fast, complete, and catalytic Z→E azo isomerization upon electron loss from the RA unit of the azobenzene. The RA-azo structures can be reversibly (E→Z→E)n cycled by sequential photo- and electrostimulation. Due to the robust nature of the RA?+-azo radical cation chain carrying species, initiation of electron transfer (ET) catalysis occurs at low levels (1.0-0.04 mol %) of catalytic loading and is effective even at Z-RA-azo concentrations of 10-4-10-5 M, yielding TONs (turnover numbers) of 100-2300 under such dilute conditions. The RA-azo Z→E conversion is demonstrated using chemical oxidation (redox switching), electrochemical oxidation (electro switching), and photochemical oxidation (photoredox switching). The Z→E acceleration is shown to be at least 2 × 109-fold for RA-azo 5. DFT calculations on methyl yellow suggest that a N-centered radical cation of the RA group stabilizes the Z→E N-N twist transition state of the RA?+-azo, yielding a large reduction in the barrier for RA?+-azo compared to neutral RA-azo. The RA-azo structure class has nanomechanical features that can be toggled with photo- and electrostimulation, the latter offering a quick switch for complete Z→E conversion.

Full text of DOI:10.1021/acs.joc.1c00763

pubs.acs.org/joc
Article
Photo-Electroswitchable Arylaminoazobenzenes
Carl Jacky Saint-Louis,§ David J. Warner,§ Katie S. Keane, Melody D. Kelley, Connor M. Meyers,
and Silas C. Blackstock*
Cite This: https://doi.org/10.1021/acs.joc.1c00763
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Azobenzenes appended with a redox-active arylamino
group (redox auxiliary, RA) are prepared and shown to undergo fast,
complete, and catalytic Z→E azo isomerization upon electron loss from
the RA unit of the azobenzene. The RA-azo structures can be reversibly
(E→Z→E)n cycled by sequential photo- and electrostimulation. Due to
the robust nature of the RA•+-azo radical cation chain carrying species,
initiation of electron transfer (ET) catalysis occurs at low levels (1.0−
0.04 mol %) of catalytic loading and is effective even at Z-RA-azo
concentrations of 10−4−10−5 M, yielding TONs (turnover numbers) of
100−2300 under such dilute conditions. The RA-azo Z→E conversion is
demonstrated using chemical oxidation (redox switching), electro-
chemical oxidation (electro switching), and photochemical oxidation
(photoredox switching). The Z→E acceleration is shown to be at least 2
× 109-fold for RA-azo 5. DFT calculations on methyl yellow suggest that a N-centered radical cation of the RA group stabilizes the
Z→E N−N twist transition state of the RA•+-azo, yielding a large reduction in the barrier for RA•+-azo compared to neutral RA-azo.
The RA-azo structure class has nanomechanical features that can be toggled with photo- and electrostimulation, the latter offering a
quick switch for complete Z→E conversion.
E form to a dipole moment of ∼3 D in the Z form for the parent
system.16
INTRODUCTION
■
Given the high spatial and temporal resolution of light as an
input to control the properties of chemical systems, photo-
switchable molecules are commonly utilized to create chemical
systems exhibiting switchable characteristics. Ever since Hartley
first reported on the E/Z photoisomerism of azobenzene (1),1
chemists have implemented azobenzenes in chemical systems as
a means to impart switchability of the chemical properties of
such systems.2−12 As demonstrated in Figure 1, photoswitching
from the E isomer to the Z isomer results in a conformational
change such that the distance between the C4 and C4′ carbons is
reduced from ∼9.08 to ∼6.09 Å.13 Additionally, switching from
the planar13,14 E form to the twisted13,15 Z form is accompanied
by an increase in dipole moment from no dipole moment in the
Though irradiating a given azobenzene with an appropriate
radiation source will typically induce rapid E→Z conversion
such that a photostationary state (PSS) enriched in Z isomer is
achieved, photoswitching in the opposite direction is slower,
seldom complete, and often leaves a substantial percentage in
the Z conformation, especially for electron-donor substituted
structures.2 For the parent 1 and most of its derivatives, the E
isomer is more thermally stable than the Z isomer, and as such,
the Z isomer will convert back thermally to the E isomer in the
absence of light.2,17−20 For 1, thermal Z-to-E conversion takes
days, and while this rate can vary widely with substitution, most
systems typically thermally isomerize on the time scale of hours
to days.21−24
We hypothesized that covalently linking a readily oxidized
redox auxiliary (RA) to the azobenzene parent structure would
facilitate Z→E conversion by electron transfer (ET) catalysis via
electron−hole stabilization of the nonbonded electrons of the
azobenzene isomerization transition state, allowing for rapid and
Received: March 31, 2021
Figure 1. (a) Reversible photoswitching of 1 that results in length
contraction upon E→Z conversion or length expansion upon Z→E
conversion. (b) Converting from one isomer to the other moves the
para hydrogen ∼8.5 Å through space.
https://doi.org/10.1021/acs.joc.1c00763
© XXXX American Chemical Society
J. Org. Chem. XXXX, XXX, XXX−XXX
A

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
complete Z→E switching without the use of light or heat.
Oxidation of a catalytic amount of a given RA-appended
azobenzene will initiate the ET catalytic cycle shown in Scheme
1.25,26 Upon one-electron oxidation of the covalently linked aryl
Scheme 1. Postulated ET Catalytic Cycle for a RA-
azobenzene
Figure 2. Structures of RA-azobenzenes 2−5.
Scheme 2. Synthetic Route Used to Obtain RA-azobenzene 2
amino RA, the newly formed (Z)-azo-RA•+ species will rapidly
isomerize to the more thermally stable (E)-azo-RA•+ conformer.
Facile electron exchange between oxidized and unoxidized RA
groups will allow (E)-azo-RA•+ reduction and further (Z)-azo-
RA oxidation, thus perpetuating the ET catalytic cycle. While
our group first demonstrated ET catalytic Z→E switching of
azobenzenes with a covalently attached redox active moiety,25,26
the Hecht group has subsequently reported Z→E switching via
hole catalysis for azobenzenes without a covalently attached
redox active moiety.27 The more general azobenzene Z→E
switching by hole catalysis demonstrated by the Hecht group
relies on transitory oxidation at the Z-azo linkage followed by
rapid isomerization/reduction to avoid long resident lifetimes of
the azo radical cations, which have limited stability. The first
demonstration of ET catalyzed Z→E azo isomerization was
reported by Gescheidt and co-workers for azonorbornane.28
Our RA approach to the Z→E azobenzene ET catalysis allows
residence of the radical cation at the RA center on a longer time
scale without concern for chemical degradation. Thus, the two
scenarios involve somewhat different conditional requirements
and can find accordingly different applications.
display π→π* absorptions that are red-shifted relative to the π→
π* absorption of the parent 1 (∼320 nm), as expected based on
the strong π-electron donor nature of the RA substituent.21,25
For RA-azobenzenes 2−5, these red-shifted π→π* absorptions
obscure the weaker n→π* absorption such that it is not easily
observed.
RESULTS AND DISCUSSION
■
Azobenzenes 2−5 (Figure 2) were synthesized to test the RA
mediated ET-catalysis approach to Z→E azobenzene switching.
The synthetic route to obtain 2 is shown in Scheme 2. Reacting
the aniline 6 with Oxone yielded the nitrosobenzene 7 that was
then reacted with aniline (8) via a Mills condensation to give the
azobenzene 9.25,29,30 Separately, Hartwig−Buchwald amina-
tion31−33 was used to join 10 and 11 to form dianisylamine
(12).34 The bromide 9 and amine 12 were then coupled via
Hartwig−Buchwald amination to yield the target RA-
azobenzene 2. The general synthetic methodology of oxidation
with Oxone followed by Mills condensation followed by
palladium-catalyzed cross-coupling reaction(s) was used to
prepare all RA-azobenzenes 2−5.
E→Z switching of RA-azobenzenes 2−5 was observable by
UV−vis spectroscopy, with the Z isomer of an azobenzene
generally displaying a much less intense π→π* absorption than
its corresponding E isomer. Irradiating a solution of 2 with 430
nm light caused a decrease in the intensity of the π→π*
absorption, indicating E→Z conversion. The PSS produced by
430 nm irradiation comprised 74% Z-2 and 26% E-2, as
determined by NMR analysis of the sample (see the Supporting
Information). All RA-azobenzenes 2-5 underwent E→Z
conversion in response to irradiation to produce Z-rich
photostationary states. Table 1 shows the PSS compositions
for 2−5, as determined by 1H NMR spectroscopy.
Figure 3 shows an overlay of the optical spectra of RA-
azobenzenes 2−5 and that of the parent 1. All RA-azobenzenes
UV−vis spectroscopy was also used to track the rate of Z→E
thermal isomerization of azobenzenes 2−5. Table 2 shows the
B
https://doi.org/10.1021/acs.joc.1c00763
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
Figure 4. Cyclic voltammograms of 2−5 in DCM (0.10 M TBABF4).
[2] = 1.0 mM, [3] = 1.0 mM, [4] = 0.75 mM, and [5] = 0.99 mM. Scan
rates were 20 mV/s for 2 and 3, 50 mV/s for 4, and 100 mV/s for 5.
Figure 3. Overlay of optical spectra of azobenzenes 1−5 in benzene at
22 °C. [1] = 8.67 × 10−5 M, [2] = 6.35 × 10−4 M, [3] = 6.37 × 10−4 M,
[4] = 3.14 × 10−5 M, and [5] = 5.1 × 10−4 M. A 1.0 cm quartz cuvette
was used for 1 and 4, while a 0.1 cm quartz cuvette was used for 2, 3, and
5.
8.2 and a(2N) = 5.7 G, respectively. These N splittings
correspond closely to those of tri-p-anisylamine radical cation
(8.9 G)42,43 and tetra-p-anisyl-p-phenylenediamine radical
cation (5.75 G),39 indicating that the electron hole in RA-azos
3 and 4 is almost entirely centered on the RA moiety in these
structures. In the case of RA-azo 4, the radical cation PF6 salt is
isolably stable.
Table 1. PSS Compositions and the Respective Irradiation
Sources Used to Reach the PSS in C6D6 for RA-azobenzenes
2−5
compd
Z/E at PSS
excitation source (nm)
To test the hypothesis of a catalytic ET-mediated Z→E
conversion of the RA-azobenzenes, treatment of PSS mixtures
with 5−10 mol % oxidant 13 (aka CRET44,45 with E′° = 1.11 V
vs SCE, shown in Figure 5) was carried out and monitored by
2
3
4
5
74:26
74:26
60:40
75:25
430
430
430
395−410
Table 2. Half-Life, Rate Constant and Gibbs Free Energy of
Activation Associated with the Z→E Thermal Isomerization
for RA-azobenzenes in Deaerated Benzene at 22 °C
compd
t1/2
k (s−1
)
ΔG‡ (kcal/mol)
2
3
4
5
2 h 18 min
40 min
30 min
8.40 0.07 × 10−5
2.88 0.02 × 10−4
3.48 0.05 × 10−4
5.58 18.1 × 10−7
22.8 0.1
22.0 0.1
21.9 0.1
25.7 1.9
Figure 5. Structure and oxidation potential (vs SCE in DCM (0.1 M
TBAPF6) of CRET45 (13).
14 days
half-life, rate constant and Gibbs free energy of activation for
each azobenzene studied. The Z-isomer half-lives of 2−4 are
minutes to hours, substantially shorter than that of parent
azobenzene 1, whose Z isomer half-life is ∼2 days in most
organic solvents. This is consistent with the reported effects of p-
substituted azobenzenes with π electron-donating
groups21−25,35 In contrast, the ortho fluorinated azobenzene 5
displays a long 40 day half-life, as attributed to the special Z-
isomer stabilizing effect of the ortho fluoro groups as previously
reported by Hecht and co-workers.36
To evaluate the redox properties of the RA-azobenzenes,
cyclic voltammetry (CV) analysis was conducted. The oxidation
potentials and CV traces of RA-azobenzenes 2−5 are shown in
Figure 4. The chemically reversible oxidation behavior of 2−5
resemble that of substituted triarylamines in general,37−40
indicating that oxidation occurs at the aryl amine redox auxiliary
and the radical cations are stable on the time scale of the scan
(ms to s). In contrast, parent 1 gives a chemically irreversible
cyclic voltammogram with a peak potential of ∼1.72 V vs SCE.41
Additionally, one-electron chemical oxidation of 3 and 4 gives
solution stable radical cations whose EPR spectra show a(N) =
UV−vis spectroscopy. An optical spectrum of an all-E solution
of each RA-azo was first recorded. Next, a Z-rich PSS was
reached by irradiating with light of the appropriate wavelength
(Table 1), and an optical spectrum of the E/Z mixture at the PSS
was then recorded. The oxidant 13 was then added to the
cuvette, which was shaken for 2 s before recording another
optical spectrum. Figure 6 shows the overlay of optical spectra of
2 recorded prior to irradiation, after irradiation at the PSS, and
immediately after the addition of 13. As Figure 6 shows,
complete Z-2→E-2 conversion occurs within the 2 s mixing time
of addition of 0.10 mol % 13.
Azobenzenes 2−5 all undergo ET catalytic Z→E switching,
and the catalytic oxidant loading (∼10−6−10−8 M) needed to
trigger Z→E switching is shown in Table 3 for each RA-
azobenzene. Additionally, the minimum Z→E rate acceleration
is shown for each RA-azo as well as the extrapolated rate
acceleration (again minimum value) assuming 100 mol %
catalytic oxidant loading. These minimum rate accelerations are
based on the assumption of a maximum 0.5 s half-life as observed
for the 2 s mixing period in these experiments, after which
C
https://doi.org/10.1021/acs.joc.1c00763
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
content responses for 3 and 4 during the photo/redox cycling
experiments.
Figure 6. Optical spectra of 2 (6.3 × 10−5 M) in benzene at 22 °C
recorded prior to 430 nm irradiation (solid black), after irradiation at
the PSS (solid blue), and after the addition of 0.10 mol % 13 (dashed
yellow) in a 1.0 cm quartz cuvette.
Table 3. Catalytic Loadings of Oxidant 13 Required to
Trigger Complete Z→E Switching of Azobenzenes 2−5 as
Well as Minimum Rate Accelerations, Extrapolated
Minimum Rate Accelerations Assuming 100 mol % 13, and
Catalytic Turnover Numbers
Figure 7. (Top) Absorbance vs time profile following the irradiation of
3 (1.9 × 10−5 M) at 436 nm after addition of 0.91 mol % 13 as the 430
nm LED source is turned on for 15.5 s and off for 15.5 s repeatedly for
3000 s. (Bottom) Absorbance vs time profile following the irradiation of
4 (4.27 × 10−5 M) at 454 nm after addition of 0.077 mol % 13 as the 430
nm LED source is turned on for 15.5 s and off for 15.5 s repeatedly for
3600 s.
mol %
13
rate
acceleration
extrapolated rate acceleration
assuming 100 mol % 13
compd
TON
2
3
4
5
0.10
1.0
0.044
0.11
18000×
4800×
4000×
18000000×
480000×
9100000×
2300000000×
1000
100
2300
910
2500000×
Complete Z→E (redox-mediated) conversion of 3 is
observed at the end of each E→Z→E cycle through 46 cycles.
After the 46th cycle, Z→E conversion is no longer complete in
the 15.5 s time domain, and the extent of Z→E conversion at the
end of each cycle decreases with increasing number of cycles
performed, as seen in Figure 7, consistent with loss of catalytic
efficacy with increasing photocycling. In contrast, 4 is observed
to undergo 115 E→Z→E cycles with complete Z→E conversion
observed even at the end of the 115th E→Z→E switching cycle.
Thus, 4 proved more robust than 3 when performing repeated
E→Z→E switching cycles in the presence of a low concentration
of radical cation catalyst. Additionally, for the photocycling
experiments of Figure 7, only 0.077 mol % catalytic oxidant
loading is used for photoredox cycling of 4, whereas 0.91 mol %
catalytic oxidant loading is used for the less effective photoredox
cycling of 3. We attribute the lower catalytic oxidant loading
needed and greater number of complete E→Z→E cycles
performed for 4 compared to 3 as a consequence of the very
robust nature of the p-phenylenediamine radical cation RA•+
feature of 4 compared to the somewhat more sensitive
triarylamine radical cation RA•+ of 3. Both oxidized RA units
work remarkable well at these very low concentrations in the
10−7 to 10−8 M range where adventitious impurities can
potentially scavenge the catalytically active azo-RA•+ species. It
is important to note that for both 3 and 4 the E→Z→E cycling
robustness is improved significantly by using purer solvent.
Having shown that addition of a chemical oxidant to a
solution of a given RA-azobenzene enriched in Z isomer can
trigger rapid and complete Z→E conversion, we next sought to
trigger Z→E isomerization via the direct application of voltage.
Figure 8 shows a diagram of the photoelectrochemical device
(PED) utilized for voltage-induced switching. The device is
constructed by dissolving the RA-azobenzene in a (tol)3TBABF4
complete conversion to all-Z had occurred, with no observation
of any unreacted Z isomer.
Table 3 shows that a lower catalytic oxidant loading is needed
to trigger complete Z→E switching of 4 compared to the
amount of oxidant needed to trigger complete Z→E switching of
2, 3, and 5. This result is consistent with the employment of the
p-phenylenediamine RA unit for 4, which yields an especially
robust radical cation, compared to the more localized and
somewhat less robust nature of the triarylamine RA units of 2, 3,
and 5. Thus, of the series, the catalytic turnover number (TON)
observed for 4 is the greatest at 2300, which is achieved even at
the very low catalyst concentration of 1.4 × 10−8 M. The highest
minimum Z→E rate acceleration is observed for 5, which
incorporates a dianisylamino RA like 2 and 3 but whose neutral
thermal Z isomer half-life is 14 days under ambient conditions.
With sufficient catalytic oxidant loading, none of the RA-
azobenzenes undergo measurable E→Z photoconversion once
the oxidant had been added because the rate of Z→E
electrocatalytic switching is greater than the rate of E→Z
photoconversion such that no accumulation of Z isomer is
observed by UV−vis spectroscopy. However, we found that fine-
tuning the amount of oxidized RA-azo employed in the reaction
mixture allowed for observation of E→Z photoconversion even
in the presence of the added oxidant. For 3 and 4, adding a
catalytic amount of the oxidant 13 (CRET) and then repeatedly
toggling the irradiation source on and off allowed multiple E→
Z→E switching cycles to be recorded. Switching between E and
Z isomers was confirmed by following the absorbance at a given
wavelength throughout multiple photoswitching cycles, and the
time absorption profiles shown in Figure 7 depict the Z/E
D
https://doi.org/10.1021/acs.joc.1c00763
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
Figure 8. Diagram of the photoelectrochemical device (PED) utilized
for voltage-induced switching.
electrolyte gel46 that is then deposited onto the indium tin oxide
(ITO) face of an ITO-coated glass slide. A thin (88.9 μm)
polyethylene spacer with a window cut out of it is placed on the
slide such that the RA-azo-containing gel is contained within the
window. A second ITO-coated glass slide is placed on top of this
such that its ITO face is in contact with the gel and the spacer.
The device is held together with two closed jaw compressor
clamps, and a strip of aluminum tape is applied to the ITO face
of each slide. A DC voltage can be applied across the ITO faces
of the PED and to the thin-layer gel containing the RA-azo.
As the ITO-coated plates of the PED are transparent, Z→E
isomerization of the RA-azo dissolved in the gel may be
monitored by UV−vis spectroscopy. The transparency of the
ITO plates also means that Z→E photoisomerization can be
induced by irradiating the device. By irradiating and then turning
off the light source while simultaneously turning on the voltage
supply, multiple E→Z→E switching cycles could be performed.
Figure 9 shows a time radiation profile that follows the
Figure 10. Radiation time profile of 3 (5.38 × 10−5 M) in a deaerated
2.1% MeOH/97.9% toluene solution at 22 °C in a 1.0 cm quartz
cuvette. Absorbance at 440 nm was recorded every 15 s. The green
curve represents cycles of light off/430 nm light on/light off/660 nm
light on when 14 (2.7 × 10−6 M) is present. The black curve represents
the absorbance at 440 nm when the same cycles are performed in the
absence of 14.
excited state is 0.97 V vs SCE.48 Thus, upon photoexcitation, 14
is capable of oxidizing 3 (0.85 V vs SCE; Figure 4).
In the photooxidant cycling experiments, both the RA-azo 3
and photooxidant 14 were present in the same cuvette.
Absorbance at 440 nm, attributed to RA-azo 3, was tracked
every 15 s throughout the experiment. Initially measurements
were taken in the absence of light for 1 min, and then the 430 nm
irradiation source was turned on for 1 min. Next, the 430 nm
source was turned off for 1 min, and then a 660 nm source was
turned on for 1 min. This process of light off/430 nm light on/
light off/660 nm light on was repeated for a total of 10 cycles,
and the radiation time profile tracking absorbance at 440 nm
over the course of the experiment is displayed in Figure 10 as the
green curve. After the last 430 nm irradiation period, the light
was turned off for ∼18 min, during which time absorbance at
440 nm was still being measured. This result showed that in the
absence of 660 nm irradiation to excite photooxidant 14 and
trigger ET catalytic Z→E switching of 3, thermal Z→E switching
proceeds at a much slower rate. Thus, neither ground state 14
nor the methanol needed to dissolve 14 were responsible for the
rapid Z→E conversion observed upon 660 nm irradiation. As an
additional control, cycles of light off/430 nm light on/light off/
660 nm light on were performed in the presence of methanol but
with 14 absent from the cuvette. Though some Z→E conversion
can be seen in Figure 10 (black curve), conversion does not
occur to near the extent seen when the photooxidant 14 is
present. We attribute this significantly slower and less extensive
Z→E conversion to a slight end-on absorption by Z-3 azo of the
660 nm irradiation.
Figure 9. Radiation time profile of 4 (7.5 × 10−3 M) in (tol)3TBABF4
gel at 22 °C in a PED. The 430 nm LED source was turned on for ∼31 s
and off for ∼31 s for 25 cycles. When LED was off, 0.3 V was applied via
DC power supply. Absorbance at 454 nm was recorded every second.
absorbance at 454 nm of 4 incorporated in a PED over the
course of 25 E→Z→E switching cycles. As Figure 9
demonstrates, the RA-azobenzene 4 could be toggled between
E and Z conformations using light and voltage in tandem. We
have also demonstrated this PED switching for 2.25,47
Having successfully performed E→Z→E photoredox switch-
ing cycles by fine-tuning the amount of catalytic oxidant (i.e.,
CRET cycling) and also by incorporating an RA-azo in a PED
and applying light and voltage in tandem, we next sought to use a
photooxidant such that E→Z→E switching cycles could be
performed by employing two different irradiation sources in
tandem. We chose methylene blue (14), shown in Figure 10, as
the photooxidant given that the oxidation potential of its triplet
DFT calculations (B3LYP/6-31G*) shed light on the nature
of the ET catalytic Z→E switching rate acceleration observed for
RA-azobenzenes 2−5. Figure 11 depicts the results of DFT
calculations performed for 15 (p-dimethylaminoazobenzene or
methyl yellow, chosen as a model of the RA azobenzenes 2−5),
which suggest that the energetic barrier for Z-15→E-15
isomerization would be reduced from 23.79 kcal/mol to only
1.83 kcal/mol upon one-electron oxidation of the para-
positioned dimethylamino group. The experimentally deter-
E
https://doi.org/10.1021/acs.joc.1c00763
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
CONCLUSIONS
■
In summary, we have demonstrated that for an azobenzene
substituted with a readily oxidized aryl amine redox auxiliary
(RA), oxidizing a catalytic amount of the RA-azobenzene will
trigger rapid and complete Z→E switching by electron transfer
(ET) catalysis. For RA-azobenzene 5, which displays excep-
tional Z isomer thermal stability (t1/2 = 14 days), adding just
0.11 mol % oxidant results in Z→E switching that is accelerated
at least 2500000-fold (Table 3). Extrapolating to 100 mol %
oxidant loading, we conservatively estimate at least a
2300000000-fold rate acceleration. Fine tuning the catalytic
oxidant loading allows for multiple rapid E→Z→E photo-
electroswitching cycles as demonstrated for 3 and 4. Addition-
ally, it is shown that the p-phenylenediamine moiety of 4 proved
more effective than the dianisylamine RA (incorporated in 2, 3,
and 5) in terms of lower catalytic oxidant loading needed to
trigger rapid and complete Z→E switching (Table 3) and in
terms of robustness for use in CRET-initiated cycling experi-
ments (Figure 7). Z→E switching may also be achieved via the
direct application of voltage in a photoelectrochemical device
(PED). Multiple rapid E→Z→E switching cycles were achieved
by applying light and voltage in tandem (Figure 9). Lastly, we
have demonstrated that using a catalytic amount of photo-
oxidant 14 enables rapid and complete Z→E conversion by
irradiating at a wavelength absorbed by 14 (Figure 10). Under
these conditions, multiple E→Z→E switching cycles could be
performed by irradiating with one wavelength to trigger E→Z
photoisomerization and then irradiating with a different
wavelength to excite the photooxidant 14 and trigger rapid
and complete Z→E switching via electron transfer (ET)
catalysis. These findings represent multiple means of utilizing
electron removal to rapidly and completely induce Z→E
switching of a group of RA-azobenzenes, with little to no
degradation of RA-azobenzene structure.
Figure 11. DFT calculations (B3LYP/6-31G*) indicating a large
reduction in the Z→E barrier for 15 upon one-electron oxidation of the
dimethylamino group.
mined barrier for neutral Z-15→E-15 isomerization (23.7 kcal/
mol)49 agrees well with the barrier predicted by DFT
calculations. The calculated 21.96 kcal/mol decrease in the
Z→E barrier upon electron loss would correspond to an
accelerated rate that is 1.3 × 1016 times faster than the rate for
thermal Z→E conversion in the neutral case. While the
magnitude of electron-loss acceleration in RA azobenzenes 2-5
is likely to be less than the value predicted theoretically for Z-15
because of the more delocalized nature of the RA•+ group
compared to -NMe2•+, this model calculation for Z-15→E-15
redox acceleration predicts the viability of a substantial Z→E
rate enhancement upon RA electron loss in these RA-
azobenzene systems.
We hypothesize that the accelerated Z→E isomerization that
occurs for a given RA•+-azo arises from a stabilization of the azo
N,N twist transition state for this process upon formation of the
RA•+ moiety. Upon electron loss from the RA nitrogen lone pair
orbital, a new stabilizing interaction is introduced between the
unfilled RA•+ N-orbital and the proximal azo N p-orbital at the
azo N,N twist transition state, as illustrated for compd 2•+ in
Figure 12. At the N,N twist transition state for Z→E
EXPERIMENTAL SECTION
■
General Information. Starting materials and solvents were
purchased from commercial sources and used without further
purification unless otherwise noted. Reactions were performed at rt
(295 K) or with heating using a silicone oil bath equipped with a heating
coil, and reaction condition temperatures were those of the oil bath.
Melting points were recorded using either a Mel Temp apparatus from
Laboratory Devices or a Thomas-Hoover 6406 Uni-Melt Capillary
Melting Point Apparatus and are uncorrected. Palladium cross-coupling
reactions were performed in glassware that had been either flame-dried
or oven-dried. Dichloromethane for use in UV−vis spectroscopy, EPR
spectroscopy, and cyclic voltammetry was purified by stirring over
H2SO4, washing with DI H2O, washing with 5% aq KOH, drying with
CaCl2, and distilling from CaH2 under N2 atmosphere.50 Benzene used
in electrocatalytic cycling experiments was purified by stirring over
H2SO4, washing with DI H2O, washing with 5% aq KOH, drying with
anhydr CaCl2, and refluxing over CaH2 for 2 days before distilling from
CaH2 using oven-dried short path distillation glassware.51 Column
chromatography was performed using 60 Å, 32−63 μm standard grade
silica gel from Sorbent Technologies. Thin-layer chromatography was
performed using TLC silica gel 60 F254 aluminum sheets from EMD
Millipore. Manual flash column chromatography was performed using
silica gel (Silicycle, 60 Å, 230−400 mesh). Celite 545 (0.02−0.1 mm
particle size) from EMD Millipore was used in filtrations, along with
activity III alumina, which was prepared by DI H2O addition (6 wt %)
to basic alumina (Aldrich, activated Brockmann I, ∼150 mesh) to
remove fine Pd catalyst particles from crude product mixtures.
Figure 12. Half-filled orbital of the nitrogen atom of RA•+ forms a π-
bond with the singly occupied azo N p-orbital at the azo N,N twist
transition state.
isomerization, the single-occupied diazenyl nitrogen p-orbital
overlaps with the partially filled p-orbital of the RA•+, creating
some pi-bonding between these moieties that substantially
stabilizes the azo N,N twist transition structure.
As shown in Figures 11 and 12, it is noteworthy that the
predicted transition state for Z→E thermal isomerization of
neutral 15 involves N-inversion at the diazenyl group. However,
upon dimethylamino group one-electron oxidation, the
predicted isomerization mechanism changes to an azo N,N
twist pathway.
TBABF4 (TCI America) was purified by recrystallization 3× from a
1:1 Millipore H2O/EtOH and then drying in a vacuum oven (110 °C,
0.1 Torr) for 48 h.52 NBS was recrystallized from THF (stored over
F
https://doi.org/10.1021/acs.joc.1c00763
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
NaOH(s)) prior to use. The chemical oxidant CRET45 (13) was
organic layers were dried with anhydr MgSO4. DCM was removed in
vacuo to yield an amorphous gray-green solid (3.24 g, 17.42 mmol, 96%
crude yield) that was later used without further purification. Rueck-
Braun et al. previously prepared 7 in 70% yield using sublimation to give
≥95% purity level (mp = 96−99 °C).29 Mp = 94−99 °C. 1H NMR (360
MHz, CDCl3): δ 7.78 (s, 4H).
prepared by the procedure of Rathore and co-workers.44 Commercially
̈
available methylene blue (14) was obtained from Riedel-de Haen AG
(now Honeywell Fine Chemicals) and used without further
purification. Regarding Pd-catalyzed N-arylation reaction reagents:
Pd(dba)2 and Pd2(dba)3 were purchased from Ark Pharm Inc. and
P(tBu)3 and NaOtBu were purchased from Alfa Aesar. All were stored
and dispensed in an N2 drybox. P(tBu)3 was deployed as a stock
solution in anhydr toluene via microliter syringe.
4-Bromoazobenzene (9).53,54
1H NMR and 13C NMR spectra were recorded on Bruker AM-360
and AM-500 spectrometers using deuterated solvents (Cambridge
Isotopes) with chemical shifts referenced to the residual nondeuterated
solvent signal or TMS. CW EPR measurements for 3•+PF6− were made
at 293 K using an ELEXSYS E680 EPR spectrometer (Bruker-Biospin,
Billerica, MA) equipped with a Bruker Flexline ER 4118 CF cryostat
and an ER 4118X-MD4 ENDOR resonator. Spectra were recorded
under nonsaturating conditions with a microwave power of 0.79 mW.
Measurements used a microwave frequency of 9.69 GHz, a modulation
amplitude of 1.0 G, and a modulation frequency of 100 kHz. The CW
EPR spectrum was baseline corrected in Xepr (Bruker-Biospin,
Billerica, MA) and plotted in Microsoft Excel. CW EPR spectroscopic
measurements for 4•+PF6− were made at 298 K using a Bruker ER082 x-
band spectrometer at 9.45 GHz. High resolution mass spectrometry
(HRMS) data was acquired using a Waters Autospec NT magnetic
sector mass spectrometer with an electron impact (EI) ionization
source and an electric-magnetic-electric (EBE) sector mass analyzer or
using a Waters Xevo G2-XS Quadrupole Time-of-Flight (QTof) mass
spectrometer with an electrospray ionization (ESI) source and a QTof
mass analyzer. FT-IR spectra were recorded on a Bruker Alpha FTIR
instrument (Bruker Optics, Inc., Billerica, MA) featuring an attenuated
total reflectance (ATR) sampler equipped with a diamond crystal. All
solid samples were placed over the detector and crushed using the anvil.
UV−vis optical spectra were recorded on a HP8452A diode array
spectrophotometer using a 0.1 or 1.0 cm quartz cuvette. Cyclic
Voltammetry was performed on a PAR 273 electrochemical
potentiostat (EG&G Instruments) using a platinum disc working
electrode, platinum wire counter electrode, and SCE reference
electrode. For the photoelectrochemical device (PED) experiments,
voltage was applied to the device using either a model 1666 BK
Precision DC power supply or the PAR 273 electrochemical
potentiostat.
4-Bromonitrosobenzene 7 (1.00 g, 5.38 mmol), aniline (Fisher
Scientific) (490 μL, 0.500 g, 5.37 mmol), and 10 mL of acetic acid
were added to a 25 mL RB flask. The solution appeared red 1 min after
initiating vigorous stirring. The mixture was stirred 24 h at rt to give a
dark brown solution and a yellow-orange precipitate. The solid was
collected by vacuum filtration and rinsed on the filter with 3 × 30 mL
saturated aq NaHCO3 solution. The crude solid was recrystallized from
EtOH to yield 9 (0.80 g, 3.1 mmol, 58%) as an orange crystalline
product. Kaiser and co-workers previously prepared 9 in 71% yield.53
Mp: 83−85 °C (lit.54 mp 88−89 °C). 1H NMR (360 MHz, CDCl3): δ
7.92 (d, J = 8.1 Hz, 2H), 7.81 (d, J = 8.7 Hz, 2H), 7.66 (d, J = 8.7 Hz,
2H), 7.51 (m, 3H). 13C{1H} NMR (125 MHz, CDCl3): δ 152.7, 151.6,
132.5, 131.5, 129.3, 125.5, 124.5, 123.1.
Di-p-anisylamine (12).
A 100 mL RB flask was charged with p-anisidine (Acros Organics) (4.93
g, 40.0 mmol) and 4-bromoanisole (Alpha Aesar) (2.70 mL, 7.39 g,
39.5 mmol) and transferred to a N2-purged drybox, where Pd(dba)2
(0.115 g, 0.200 mmol), a 0.100 M anhydr toluene solution of P(tBu)3
(2.00 mL, 0.0405 g, 0.200 mmol), and NaOtBu (5.77 g, 60.0 mmol)
were added to the 100 mL RB flask. After addition of 60 mL of anhydr
toluene, the mixture was stirred for ∼1 min. The flask was capped with a
rubber septum, removed from the drybox, and stirred under N2 for 1 h
at 90 °C. The reaction mixture was then allowed to cool before being
vacuum filtered through 5 g of basic alumina III. The filter cake was
washed with ∼200 mL ethyl acetate, the solvents were removed in
vacuo, and the resulting crude product was recrystallized from hexanes
to give a pale pink crystalline solid, 12 (8.71g, 38.0 mmol, 95.0%). Endo
and colleagues previously prepared 12 in 67% via Hartwig−Buchwald
amination employing BINAP as the ligand.34 Mp: 98.5−100 °C (lit.34
mp 101−103 °C). 1H NMR (360 MHz, CDCl3): δ 6.94 (d, J = 8.9 Hz,
4H), 6.82 (d, J = 8.9 Hz, 4H), 5.28 (s, 1H), 3.78 (s, 6H). 13C{1H} NMR
(125 MHz, CDCl3): δ 154.5, 138.2, 119.7, 114.9, 55.82.
ITO coated glass (Delta Technologies) was cleaned using a sonicator
(solid state/ultrasonic FS-14, Fischer Scientific). Elemental analyses
were performed by Atlantic Microlab, Inc., Norcross, GA. The 430 nm
light-emitting diode light source (Zhongshan LED Technology Co.,
Ltd.) contains 7 × 1 W Diamond Grow LEDs (input voltage of AC 90−
264 V) housed in a 3.70 in. × 3.86 in. Fins Slotted Aluminum casing.
The 395−410 nm LED source was a single 7 W BLB7W bulb (ADJ
Products, LLC). Excitation spectra for the 395−410 and 430 nm LED
sources are shown in the Supporting Information (Figure S31).
Synthesis of 4-Diarylaminoazobenzenes 2−5. 4-Bromonitro-
sobenzene (7).9
4-(p-Dianisylamino)azobenzene (2).25,55
The synthesis of 2 from dianisylamine 12 (0.1881 g, 0.7962 mmol) and
4-bromoazobenzene 9 (0.2079 g, 0.7962 mmol) in a 2 dram screw-cap
vial equipped with septum was performed analogously to the synthesis
of 12 from p-anisidine and 4-bromoanisole. The vial was charged with
12 and 9, and the mixture was transferred to a N2 purged drybox, where
Pd(dba)2 (0.0046 g, 0.0080 mmol), P(tBu)3 (119 μL of 0.100 M
anhydr toluene solution, 0.0119 mmol), NaOtBu (0.115 g, 1.20 mmol),
and anhydr toluene (2.5 mL) were added. The vial was capped and
removed from the drybox. The reaction mixture was stirred at 80 °C
until TLC monitoring indicated consumption of starting materials and
formation of new product. Purification was achieved via column
chromatography (silica gel, 1:1 benzene/hexanes) to yield 2 (0.31 g,
4-Bromoaniline (Oakwood Chemical) (3.113 g, 18.10 mmol) was
dissolved in 60 mL of DCM with stirring in a 250 mL round bottom
(RB) flask. Separately, Oxone (Alpha Aesar) (22.60 g, 36.76 mmol) was
dissolved in 100 mL of DI H2O in a 125 mL separatory funnel with
vigorous shaking and then added to the stirring solution of 4-
bromoaniline in DCM over the course of 1 min. The organic layer
turned green during this time. TLC monitoring on silica using 1:4
DCM/hexanes indicated that the reaction was essentially complete
(disappearance of starting material) after ∼24 h stirring. The mixture
was stirred at rt for a total of 49 h. After collecting the organic layer, the
aqueous layer was extracted 3× with 20 mL DCM. The combined
G
https://doi.org/10.1021/acs.joc.1c00763
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
0.76 mmol, 95%) as an amorphous red-orange solid. Mp: 157−159 °C.
1H NMR (500 MHz, CDCl3): δ 7.84 (d, J = 7.4 Hz, 2H), 7.77 (d, J = 8.6
N-p-Anisyl-4-amino-4′-methoxyazobenzene (17).
Hz, 2H), 7.48 (t, J = 7.5 Hz, 2H), 7.40 (t, J = 7.5 Hz, 1H), 7.13 (d, J =
8.6 Hz, 4H), 6.95 (d, J = 8.6 Hz, 2H), 6.88 (d, J = 8.6 Hz, 4H), 3.82 (s,
6H). 13C{1H} NMR (125 MHz, CDCl3): δ 156.9, 153.3, 151.6, 146.2,
140.0, 130.0, 129.1, 127.6, 124.5, 122.5, 118.8, 115.1, 55.7. HRMS (EI/
EBE) m/z: [M + H]+ Calcd for C26H23N3O2 409.1790; Found
409.1798. Anal. Calcd for C26H23N3O2: C, 76.26; H, 5.66; N, 10.26.
Found: C, 75.98; H, 5.81; N, 10.11. UV−vis (benzene, λmax, ε,
concentration, path length): 436 nm, 27,200 M−1cm−1, 6.35 × 10−4 M,
0.1 cm. IR (cm−1): 2833, 1593, 1500, 1476.
A 50 mL RB flask was charged with 4-bromo-4′-methoxyazobenzene 16
(1.3050 g, 4.482 mmol) and p-anisidine (Acros Organics) (0.5520 g,
4.482 mmol) and transferred to a N2-purged drybox, where Pd(dba)2
(0.0260 g, 0.0452 mmol), a 0.359 M anhydr toluene solution of P(tBu)3
(140 μL, 0.0128 g, 0.0631 mmol) in anhydr toluene, and NaOtBu
(0.6711 g, 6.983 mmol) were added. Lastly, 14 mL of anhydr toluene
was added, and the mixture was stirred ∼1 min before being capped
with a rubber septum and removed from the drybox. The reaction
mixture was further stirred under N2 for 26 h 15 min at 82 °C, cooled,
and vacuum filtered through 6 g of Celite on top of 15 g basic alumina
III. The filter cake was washed with ∼350 mL ethyl acetate, the solvents
were removed in vacuo, and the resulting crude product was
recrystallized from ethanol to give a red-orange solid, 17 (1.2393 g,
3.717 mmol, 83%). Mp: 135−136.5. °C 1H NMR (360 MHz, acetone-
d6): δ 7.83 (d, J = 9.0 Hz, 2H), 7.78 (d, J = 8.8 Hz, 2H), 7.69 (s, 1H),
7.21 (d, J = 8.8 Hz, 2H), 7.06 (m, 4H), 6.95 (d, J = 8.9 Hz, 2H), 3.89 (s,
3H), 3.80 (s, 3H). 13C{1H} NMR (125 MHz, acetone-d6): δ (ppm)
161.4, 155.9, 148.7, 147.2, 145.2, 134.6, 124.5, 123.8, 123.0, 114.6,
114.2, 113.9, 55.0, 54.8. HRMS (ESI/Q-TOF) m/z: [M + H]+ Calcd
for C20H20N3O2 334.1556; Found: 334.1543. IR (cm−1): 2956, 1598,
1580, 1495.
4-Bromo-4′-methoxyazobenzene (16).56,57
4-Bromonitrosobenzene 7 (2.64 g, 14.2 mmol), p-anisidine (Acros
Organics) (1.75 g, 14.2 mmol) and 36 mL acetic acid were added to a
100 mL RB flask, and the mixture was stirred 15 h 55 min at rt. The
solution turned dark brown, and a yellow-orange precipitate formed
that was collected by vacuum filtration, rinsed on the filter with ∼20 mL
cold saturated aq NaHCO3, and purified by column chromatography
(silica gel, 1:3 toluene/hexanes) to yield 16 (2.01 g, 6.90 mmol, 49%) as
an amorphous orange solid. Feringa and co-workers prepared 16 via
diazonium coupling followed by methylation in 64% yield.56 Mp:
144.5−147.5 °C (lit.57 mp 146.0−146.5 °C). H NMR (360 MHz,
1
CDCl3): δ 7.92 (d, J = 9.0 Hz, 2H), 7.76 (d, J = 8.4 Hz, 2H), 7.62 (d, J =
8.4 Hz, 2H), 7.02 (d, J = 9.0 Hz, 2H), 3.90 (s, 3H). 13C{1H} NMR (125
MHz, CDCl3): δ 162.5, 151.6, 147.0, 132.4, 125.0, 124.7, 124.2, 114.4,
55.7.
N,N-(Di-p-anisyl)aniline (18).59,60
4-(Di-p-anisylamino)-4′-methoxyazobenzene (3).25,26,58
A 100 mL RB flask was transferred to a N2-purged drybox and charged
with Pd(dba)2 (0.0798 g, 0.139 mmol), a 0.100 M anhydr toluene
solution of P(tBu)3 (1520 μL, 0.0308 g, 0.152 mmol), NaOtBu (1.9773
g, 20.575 mmol), and 35 mL anhydr toluene. The mixture was stirred
∼2 min before capping with a rubber septum and removing from the
drybox. While the mixture was stirred under N2, 4-bromoanisole (Alpha
Aesar) (1,750 μL, 2.60 g, 13.9 mmol) and aniline (Fisher Scientific)
(635 μL, 0.647 g, 6.95 mmol), which had both been sparged with N2 for
2 min, were added to the RB flask through the septum via syringe. The
reaction mixture was stirred under N2 for 24 h at 78 °C, cooled to rt, and
vacuum filtered through 6 g of Celite on top of 15 g basic alumina III.
The filter cake was washed with ∼200 mL ethyl acetate, the solvents
were removed in vacuo, and the resulting crude product was
recrystallized from ∼65 mL methanol to give a yellow crystalline
solid, 18 (1.7981 g, 5.8883 mmol, 85%). Liou and co-workers
synthesized 18 via Ullmann coupling in 72% yield.59 Rizzo et al.
prepared 18 in 95% yield via Hartwig−Buchwald amination and
purification by column chromatography using 1:10 EA/cyclohexane
eluent.60 Mp: 107.5−108.5 °C (lit.59 mp 107−109 °C). 1H NMR (360
MHz, acetone-d6): δ 7.17 (t, J = 7.7 Hz, 2H), 7.02 (d, J = 9.0 Hz, 4H),
6.89 (d, J = 9.0 Hz, 4H), 6.84 (m, 3H), 3.78 (s, 6H). 13C{1H} NMR
(125 MHz, C6D6): δ 156.4, 149.5, 141.8, 129.4, 126.9, 121.6, 121.1,
115.2, 55.0
The synthesis of 3 from dianisylamine 12 (0.550 g, 2.41 mmol) and 16
(0.701 g, 2.41 mmol) was performed analogously to the synthesis of 2
from 9 and 12. In a N2-purged drybox, Pd(dba)2 (0.042 g, 0.072 mmol),
P(tBu)3 (0.015 g, 0.072 mmol), NaOtBu (0.231 g, 2.40 mmol), and 7.5
mL anhydr toluene were added to a flame-dried 25 mL RB flask. The
reaction mixture was removed from the drybox and stirred at 90 °C
under N2 until TLC monitoring indicated consumption of starting
materials and formation of a new product. Purification was achieved by
recrystallizing from isopropanol to yield 3 (0.94 g, 2.14 mmol, 89%) as
1
an amorphous orange solid. Mp: 100−104 °C. H NMR (500 MHz,
CDCl3): δ 7.85 (d, J = 9.0 Hz, 2H), 7.72 (d, J = 9.0 Hz, 2H), 7.12 (d, J =
9.0 Hz, 4H), 6.99 (d, J = 9.0 Hz, 2H), 6.96 (d, J = 9.0 Hz, 2H), 6.87 (d, J
= 9.0 Hz, 4H), 3.88 (s, 3H), 3.82 (s, 6H). 13C{1H} NMR (125 MHz,
CDCl3): δ 161.4, 156.7, 151.0, 147.6, 146.4, 140.2, 127.5, 124.3, 124.0,
119.1, 115.0, 114.3, 114.3, 55.7, 55.6. HRMS (EI/EBE) m/z: [M + H]+
Calcd for C27H25N3O3 439.1896; Found 439.1892. Anal. Calcd for
C27H25N3O3: C, 73.79; H, 5.73; N, 9.56. Found: C, 73.56; H, 5.83; N,
9.47. UV−vis (benzene, λmax, ε, concentration, path length): 438 nm,
28800 M−1 cm−1, 6.37 × 10−4 M, 0.1 cm. IR (cm−1): 2833, 1593, 1500,
1476.
H
https://doi.org/10.1021/acs.joc.1c00763
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
4-Bromo-N,N-(di-p-anisyl)aniline (19).59,60
20,000 M−1cm−1, 3.14 × 10−4 M, 0.1 cm. IR (cm−1): 2832, 1594, 1496,
1462, 1440.
4-Bromo-2,6-difluoronitrosobenzene (20).61
A 50 mL RB flask was charged with NBS (Fisher Scientific) (0.3270 g,
1.837 mmol) and 10 mL of THF with stirring while cooling to 0 °C in
an ice bath for 15 min. The resulting solution was a pale yellow color. A
solution of N,N-dianisylaniline 18 (0.5611 g, 1.637 mmol) in 9 mL of
THF was added to the 50 mL RB flask with stirring over the course of
∼1 min with continued cooling in an ice bath. Upon addition of the first
drops of 18 in THF, the solution in the RB flask quickly changed from
pale yellow to brown with a green tint. After the reaction mixture was
stirred for 45 min, the ice bath was removed. After being stirred an
additional 25 h 30 min at rt, the reaction mixture was dark brown but
appeared to have lost its green tint. THF was removed in vacuo, and the
crude product was dissolved in 30 mL DCM and washed with 30 mL DI
H2O in a separatory funnel. After the organic layer was collected, the
aqueous layer was extracted with 20 mL of DCM. The organic layers
were combined and concentrated in vacuo to give a brown oil, which
was recrystallized from ∼24 mL methanol to yield a pale pink crystalline
solid, 19 (0.5276 g, 1.373 mmol, 75%). Liou et al. prepared 19
analogously in 74% yield.59 Rizzo and co-workers prepared 19 in 90%
yield using a similar procedure and purifying by column chromatog-
raphy using 1:4 DCM/pentane eluent.60 Mp 98.5−100 °C (lit.59 mp
91−92 °C). 1H NMR (360 MHz, CDCl3): δ 7.23 (d, J = 8.6 Hz, 2H),
7.02 (d, J = 8.8 Hz, 4H), 6.82 (d, J = 8.8 Hz, 4H), 6.79 (d, J = 8.6 Hz,
2H), 3.79 (s, 6H). 13C{1H} NMR (125 MHz, C6D6): δ (ppm) 156.7,
148.5, 141.0, 132.3, 126.9, 122.6, 115.2, 113.0, 55.0
A 250 mL RB flask was charged with 4-bromo- 2,6-difluoroaniline
(Matrix Scientific) (2.00 g, 9.61 mmol) and 25 mL of DCM to give a
colorless solution. Separately, Oxone (8.88 g, 14.4 mmol) was dissolved
in 95 mL DI H2O in a 125 mL separatory funnel with vigorous shaking.
The solution of Oxone in DI H2O was added all at once to the stirring
mixture in the 250 mL RB flask. The organic layer quickly turned green.
1
It was determined by H NMR spectroscopy that the reaction was
∼95% complete after ∼24 h stirring. The mixture was stirred for a total
of 72 h at rt. After the organic layer was collected, the aqueous layer was
extracted 2× with 20 mL DCM. The organic layers were combined and
dried with anhydr Na2SO4. DCM was removed in vacuo to give a tan-
brown amorphous solid, 20 (2.08 g, 9.37 mmol, 97.5%), that was later
used without further purification. Hecht et al. previously prepared 20
and used it without purification in a Mills condensation.61 1H NMR
(360 MHz, CDCl3, 22.5 °C): δ 7.34 (d, J = 8.0 Hz, 2H).
4-Bromo-2,2′,6,6′-tetrafluoroazobenzene (21).61
4-bromo-2,6-difluoro-nitrosobenzene 20 (2.08 g, 9.37 mmol), 2,6-
difluoroaniline (Oakwood Chemical) (1.010 μL, 1.21 g, 9.37 mmol),
TFA (349 μL, 0.534 g, 4.69 mmol), and 30 mL of DCM were added to a
100 mL RB flask. The mixture was stirred in the dark for 72 h at rt. In the
dark, anhydr K2CO3(s) was added to the reaction mixture, which was
then stirred ∼10 min before the K2CO3(s) was removed by gravity
filtration. DCM was removed in vacuo. In the dark, purification was
achieved by adhering the crude product to ∼1 g silica gel, loading it
onto a ∼1.5 in. plug of silica gel (50 g packed into a 150 mL fritted glass
funnel) and rinsing the filter cake with ∼200 mL 1:1 DCM/hexanes
eluent. Solvents were removed in vacuo to give a red crystalline solid, 21
(2.56 g, 7.89 mmol, 82%). Hecht and co-workers have previously
4-Methoxy-4′-(N,N,N′-(tris(p-anisyl))-1,4-phenylenediamino)-
azobenzene (4).
prepared 21 in 93% yield.61 Mp: 74−76.5 °C. H NMR (360 MHz,
1
CDCl3): δ 7.39 (m, 1H), 7.28 (d, J = 8.0 Hz, 2H), 7.07 (t, J = 8.7 Hz,
2H). 13C{1H} NMR (125 MHz, C6D6): δ (ppm) 156.8 (dd, J = 36.86
Hz, J = 4.55 Hz), 154.7 (dd, J = 40.65 Hz, J = 4.57 Hz), 132.3 (t, J = 9.85
Hz), 131.7 (t, J = 10.35 Hz), 131.2 (t, J = 9.64 Hz), 124.1 (t, J = 11.90
Hz), 116.7 (dd, J = 23.86 Hz, J = 3.42 Hz), 112.6 (dd, J = 19.85 Hz, J =
3.89 Hz).
A 35 mL RB flask was charged with 4-bromo-N,N-(di-p-anisyl)aniline
19 (0.1350 g, 0.3512 mmol) and N-anisyl-4-amino-4′-methoxyazo-
benzene 17 (0.1171 g, 0.351 mmol), and transferred to a N2 purged
drybox, where Pd(dba)2 (0.0072 g, 0.0125 mmol), a 0.100 M anhydr
toluene solution of P(tBu)3 (126 μL, 0.00255 g, 0.0126 mmol), and
NaOtBu (0.0710 g, 0.739 mmol) were added, followed by 8 mL of
anhydrous toluene. The contents were stirred ∼1 min before capping
with a rubber septum, removing from the drybox and stirring under N2
in the dark for 24 h at 80 °C. The reaction mixture was allowed to cool
before vacuum filtering through 2 g of Celite on top of 5 g basic alumina
III. The filter cake was washed with ∼175 mL ethyl acetate, the solvents
were removed in vacuo, and the resulting crude product was
recrystallized from ethanol to give a red amorphous solid, 4 (0.1441
4-(p-Dianisylamino)-2,2′,6,6′-tetrafluoroazobenzene (5).
1
g, 0.2263 mmol, 64%). Mp: 135−136.5 °C. H NMR (360 MHz,
acetone-d6): δ 7.84 (d, J = 9.0 Hz, 2H), 7.74 (d, J = 9.0 Hz, 2H), 7.19 (d,
J = 9.0 Hz, 2H), 7.07−7.05 (m, 8H), 6.99 (d, J = 9.0 Hz, 2H), 6.93−
6.88 (m, 8H), 3.89 (s, 3H), 3.82 (s, 3H), 3.78 (s, 6H). 13C{1H} NMR
(125 MHz, acetone-d6): δ 162.6, 158.3, 157.2, 152.2, 148.2, 146.9,
146.9, 141.9, 140.6, 140.5, 128.9, 127.8, 127.5, 125.0, 124.8, 122.4,
119.3, 116.0, 115.8, 115.3, 56.1, 55.9, 55.9. HRMS (EI/EBE) m/z: [M
+ H]+ Calcd for C40H36N4O4 636.2737; Found 636.2755. Anal. Calcd
for C40H36N4O4: C, 75.45; H, 5.70; N, 8.80. Found: C, 75.15; H, 5.71;
N, 8.78. UV−vis (benzene, λmax, ε, concentration, path length): 454 nm,
A 2 dram screw-cap vial was charged with 21 (0.4117 g, 1.236 mmol)
and dianisylamine 12 (0.2834 g, 1.236 mmol) and was transferred to a
N2-purged drybox, where Pd(dba)2 (0.0385 g, 0.0670 mmol), dppf
(0.0390 g, 0.0703 mmol), and Cs2CO3 (0.6445 g, 1.978 mmol) were
added to the vial followed by ∼3 1/2 mL anhydrous toluene. The
contents were stirred ∼1.5 min before capping with a screw-cap fitted
with a septum, removing from the drybox and stirring under N2 in the
dark for 64 1/2 h at 93 °C. The reaction mixture was allowed to cool
I
https://doi.org/10.1021/acs.joc.1c00763
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
before vacuum filtering through 4 g of Celite on top of 10 g basic
alumina III. The filter cake was washed with ∼130 mL ethyl acetate and
removal of solvents in vacuo gave a dark red oil, which was purified by
column chromatography on silica using 1:1 THF/hexanes eluent to
give a red amorphous solid, 5 (0.53 g, 1.1 mmol, 89%). Mp: 137−139
°C. 1H NMR (360 MHz, C6D6): δ 6.83 (d, J = 9.0 Hz, 4H), 6.62 (d, J =
9.0 Hz, 4H), 6.45 (m, 5H), 3.25 (s, 6H). 13C{1H} NMR (125 MHz,
C6D6): δ 158.6 (dd, J = 260.77 Hz, J = 7.33 Hz), 158.2 (s), 156.0 (dd, J
= 258.36 Hz, J = 4.54 Hz), 152.5 (t, J = 13.27 Hz), 138.2 (s), 133.4 (t, J
= 10.32 Hz), 129.4 (t, J = 10.02 Hz), 128.4 (s)*, 125.2 (t, J = 9.41 Hz),
115.4 (s), 112.4 (dd, J = 19.29 Hz, J = 4.57 Hz), 100.5 (d, J = 25.42 Hz),
55.0 (s). *The missing singlet at 128.4 ppm is revealed in the DEPT 135
spectrum and corroborated by the HSQC spectrum. HRMS (EI/EBE)
m/z: [M + H]+ Calcd for C26H19N3O2F4 481.1413; Found: 481.1418.
Anal. Calcd for C26H19N3O2F4: C, 64.86; H, 3.98; N, 8.73, F, 15.79.
Found: C, 64.82; H, 3.85; N, 8.73, F 15.59. UV−vis (benzene, λmax, ε,
concentration, path length): 406 nm, 30000 M−1cm−1, 5.1 × 10−4 M,
0.1 cm. IR (cm−1): 2844, 1624, 1506, 1464.
tometer without using the commercial cell holder, which was removed
to increase exposure to the irradiation source. Unless otherwise stated,
photochemical experiments were performed at rt (295 K).
E→Z Photoisomerization of 2−5. Samples of RA-azo (2, 3, 4, or 5)
were prepared in deaerated solvents (benzene or C6D6) in 1 mL
volumetric flasks (closed to air and flushed with N2). Using a gastight
syringe, ∼300 μL of solution was transferred to fill a 0.1 cm quartz
cuvette, which was immediately capped with a very lightly greased
Teflon stopper held in place with Teflon tape further secured by an
outer layer of parafilm. The cuvette was placed in the spectropho-
tometer and the entire cell solution was irradiated with a 430 nm or
395−410 nm LED source at a distance of 4−5 in. for several seconds to
minutes until there was no change in the optical spectrum, indicating
PSS had been reached. Absorbance from 200 to 800 nm was measured
every second as soon as the irradiation source was turned on.
Z and E Composition Determination at PSS for 2−5. Samples of
known concentration in deaerated C6D6 were prepared and placed,
under N2, in either a 5 mm NMR tube or a custom-made dual NMR-
UV−vis cell, which was closed with a tightly fitting plastic cap sealed
with Teflon tape or parafilm. The dual cell consists of a rectangular
UV−vis cell (∼1.7 mm path length) fused to the top of a 5 mm NMR
tube. The dual cell allows convenient optical and NMR analysis of the
same sample without the need for any sample transfer. Samples were
irradiated in a mirrored Dewar flask with a 430 nm or 395−410 nm
LED source at a distance of about 4−5 in. for several minutes as needed
until the sample composition remained unchanged as determined by
1H NMR analysis. The resulting composition by 1H NMR gave the PSS
Z/E composition for the sample under these conditions.
Z→E Thermal Isomerization of 2−5. Samples in deaerated solvent
at PSS, prepared as described above, were monitored by measuring
absorbance from 200 to 800 nm every 5−10 min in the dark. UV−vis
spectra were recorded until consecutive spectra were identical to those
observed for the all E composition of the sample, indicating complete
Z→E thermal conversion. Similarly, the absorbance of 5 at PSS was
monitored from 200 to 800 nm every 30 min in the dark, but the
experiment was performed five separate times at five different
temperatures (32, 40, 49, 61, and 69 °C) using a heated cuvette
holder. The Z-5→E-5 thermal rate constants measured at the five
different temperatures were utilized to form an Eyring plot that allowed
for extrapolation of the Z-5→E-5 rate constant at rt (Figures S53−S58).
Temperatures were assumed to have uncertainty of 0.5 °C, and
standard errors were derived from least-squares regression analysis and
error propagation treatments.
UV−vis Analysis of Catalytic ET Z→E Switching. To a freshly
prepared PSS mixture of RA-azo (2, 3, 4, or 5), prepared as described
previously, in a UV−vis cuvette (1.0 or 0.1 cm) was added a microliter
amount of 13 in DCM ([13] ∼ 10−4−10−5 M prepared from purified
and deaerated DCM). Next, the cuvette was quickly opened, and the
oxidant solution was rapidly injected into the center of the cell. The
cuvette was recapped, quickly shaken, and returned to the spectrometer
for immediate UV−vis analysis (spectra taken every 1 s). The process of
injecting 13, shaking the cuvette, and returning it to the
spectrophotometer occurred over the course of 2 s. (Note that it is
helpful to have two chemists to perform the experiment.)
−
−
3•+PF6 and 4•+PF6 Preparation for EPR Analysis. Azoben-
zenes 3 and 4 were oxidized to the radical cation state by NOPF6 in
freshly purified DCM. Samples were deaerated by sparging with N2.
−
EPR samples were ∼1 mM concentration. In the case of 3•+PF6 , the
sample was freshly prepared in solution prior to EPR measurement. In
−
the case of 4•+PF6 , the sample was isolated as a highly colored solid
and redissolved to take the EPR spectrum.
Preparation of Photoelectrochemical Device (PED). Prepara-
tion of Tol3TBABF4 gel.46 An oven-dried screw-cap vial was charged
with purified52 TBABF4 (0.995 g, 3.02 mmol) and 2.8 mL toluene. The
mixture was stirred ∼10 min with gentle warming before removing from
heat and stirring an additional ∼5 min. Upon ceasing stirring, the phases
were separated. The top phase was toluene, and the bottom phase was
Tol3TBABF4 gel (1.9 mL, 1.6 M), as determined by 1H NMR
spectroscopy and reported by Pickett.46 The bottom gel was used for
PED preparations. The vial was capped and the gel could be stored for
up to 2 days at rt.
ITO Glass Cleaning Procedure. The ITO-coated side of the glass
slide was determined using a multimeter resistance measurement. Two
ITO-coated glass slides were rinsed with and soaked in acetone for at
least 5 min prior to sonication in a dilute aqueous soap solution for 10
min. The slides were then rinsed thoroughly with DI H2O, acetone, and
purified DCM, respectively, prior to drying the surface with a stream of
dry N2 gas.
Preparation of Plastic Spacer. Polyethylene plastic sheeting (Husky
Brand; 88.9 μm thickness) was used as a spacer in the PED to prevent
the ITO faces of the two slides from contacting one another and
shorting the circuit. A 1.5 cm × 1.5 cm window was cut out of the
sheeting as the sample compartment.
Preparation of Photoelectrochemical Device (PED). The plastic
spacer with a cut out window was placed on the ITO face of an ITO-
coated glass slide. A 1 dram vial was charged with 4 (0.00095 g, 0.0015
mmol) and 0.20 mL of Tol3TBABF4 gel, and the contents were quickly
mixed via pipet such that 4 was fully dissolved in and evenly distributed
throughout the gel. Immediately, a portion of the gel solution was
deposited on the ITO face of the slide within the plastic spacer window.
A second ITO coated glass slide was quickly placed on top such that its
ITO face contacted the plastic spacer and gel contained within the
spacer window. Closed jaw compressor clamps were used to hold the
PED together such that the gel containing 4 was sandwiched between
the ITO coated slides and contained within the plastic spacer window.
A strip of aluminum tape was applied to the ITO face of each slide, as
shown in Figure 8. Alligator clips placed on the aluminum tape were
used to connect the PED to the DC power supply or PAR potentiostat
during experiments.
Photochemical Experiments. In order to prevent photoisomeri-
zation due to ambient light, the lights in the lab were dimmed during
sample preparation and throughout each UV−vis experiment, and a
black sheet covered the apparatus. Irradiation sources (placed ∼4−5 in.
from the sample) were underneath this sheet so that light could reach
the sample. When samples were irradiated while optical spectra were
being recorded, the cuvette was clamped in place in the spectropho-
CRET-Initiated Photoredox E−Z Cycling of 3. A solution of 1.9 ×
10−5 M 3 in purified, deaerated (with N2 stream) C6H6 was prepared,
and 3.0 mL was transferred via syringe to a 1.0 cm quartz cuvette which
was tightly closed under N2 with a Teflon stopper. The solution was
irradiated with a 430 nm LED source for 1 min, and the optical
spectrum confirmed that the PSS had been reached. Quickly, the cell
was uncapped, 20 μL of a solution of 2.6 × 10−5 M 13 in freshly distilled,
deaerated DCM was injected into the center of the solution of 3 in
C6H6, the cap was quickly replaced and secured by Teflon tape, and the
cell was shaken and returned to the spectrophotometer. Absorbance at
432 nm was measured every 1 s for 3000 s while the 430 nm 7 W LED
irradiation source was repeatedly turned on and off at 15.5 s intervals.
CRET-Initiated Photoredox E−Z Cycling of 4. This experiment was
performed as given for azo 3 except that a solution of 4.27 × 10−5 M 4 in
purified deaerated C6H6 was used. After addition of 3.2 μL of 13 (2.6 ×
10−5 M in DCM), the absorbance at 454 nm was measured every 1 s for
J
https://doi.org/10.1021/acs.joc.1c00763
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
3600 s while the 430 nm 7 W LED irradiation source was repeatedly
turned on and off at 15.5 s intervals.
Authors
Carl Jacky Saint-Louis − Department of Chemistry &
Biochemistry, Kennesaw State University, Kennesaw, Georgia
30144, United States
David J. Warner − Department of Chemistry & Biochemistry,
The University of Alabama, Tuscaloosa, Alabama 35487,
United States
Katie S. Keane − Department of Chemistry & Biochemistry, The
University of Alabama, Tuscaloosa, Alabama 35487, United
States
Melody D. Kelley − Department of Physical Science, Perimeter
College at Georgia State University, Atlanta, Georgia 30303,
United States; orcid.org/0000-0002-7852-2736
Connor M. Meyers − Department of Chemistry &
Biochemistry, The University of Alabama, Tuscaloosa,
Alabama 35487, United States
Photoelectro E−Z Cycling of 4 in Thin-Cell PED. A solution of 7.5 ×
10−3 M 4 in (tol)3TBABF4 gel was prepared and incorporated in the
PED as described above. The thin cell PED was secured in the UV−vis
spectrophotometer beam such that the absorbance of the gel in the PED
window could be measured and the electrodes of the PED were
connected with alligator clips to a DC power supply or the PAR
potentiostat for voltage application. The 430 nm LED light source was
also secured to irradiate the PED cell at a distance of 4−5 in. Initial
optical measurements were performed in the dark, and after 5 s the 430
nm irradiation source was turned on for ∼31 s. As soon as the
irradiation source was turned off, 0.3 V was applied to the PED
electrodes for ∼31 s. This process was repeated 24 more times for a total
of 25 cycles, and absorbance at 454 nm was measured every 1 s
throughout the experiment.
Photophotoredox E−Z Cycling of 3 Using 14. A N2-deaerated 5.51
× 10−5 M solution of 3 in toluene was prepared. To a 1.0 cm quartz
cuvette was added 3.0 mL of 3 in toluene solution along with 65 μL of a
N2-deaerated 1.28 × 10−4 M solution of 14 in methanol (5.0 mol %
based on 3). The resulting concentrations of 3 and 14 in the cuvette
were 5.38 × 10−5 and 2.7 × 10−6 M, respectively. The cuvette was
closed under N2 with a Teflon stopper secured and sealed with Teflon
tape to exclude air as much as possible. Two LED light sources (430 and
660 nm) were positioned to irradiate the cuvette at a distance of 4−5 in.
The lights were toggled on and off, alternating in sequence, at 1 min
intervals while the optical spectrum of the sample was recorded (200−
800 nm) every 15 s. A total of 10 430/660 nm switching cycles were
performed, after which sample monitoring by UV−vis was continued
for an additional 20 min with no irradiation. A control experiment
(eight 430/660 nm switching cycles) was performed analogously but in
the absence of methylene blue (14) photooxidant. The results from the
control experiment (black curve) are overlaid with the results from the
photooxidant cycling experiment (green curve) (Figure 10).
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.joc.1c00763
Author Contributions
§C.J.S.-L. and D.J.W. contributed equally.
Funding
We thank SREB (Southern Regional Education Board) for
fellowship support of C.J.S. and M.D.K. and the Department of
Education (GAANN fellowship) for support of C.J.S. (DoEd
P200A100190) and K.S.K. (DoEd P200A150329). We thank
the NSF-GRF (1058257) for fellowship support for M.D.K. We
also acknowledge undergraduate research (UCRA) funding
from UA for C.M.
Notes
DFT Calculations (Figure 11). The Spartan 2010 calculation
program (Wavefunction Corp.)62 was used to perform DFT
calculations using the B3LYP functional and a 6-31G* basis set.
Methyl yellow (15) was geometrically optimized in Z and E
configurations at both the neutral and radical cation oxidation states
(UB3LYP employed for the radical cations) with a check to ensure the
resulting structures were energy minima (no negative frequencies). The
transition state for Z/E isomerization in each case was located using the
transition state search feature of Spartan, and the resulting transition
state structures were confirmed to have a single negative eigenvalue
with a frequency calculation in each case. The imaginary frequency for
the transition states of 15 and 15•+ Z/E isomerization involved CNN
bending consistent with the N-inversion pathway for neutral 15 and
CNNC twisting consistent with the N,N twist reaction pathway for
15•+. The ground-state and transition-state energies are depicted in
Figure 11. Calculation energies of minima and transition states were
used uncorrected for zero-point energies.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Ben Fowler, Dr. Molly Lockart, and the UA EPR
facility for EPR analysis and Dr. Qiaoli Liang for MS analysis.
REFERENCES
■
(1) Hartley, G. S. Cis form of azobenzene. Nature (London, U. K.)
1937, 140, 281.
(2) Warner, D. J.; Keane, K. S.; Blackstock, S. C., Bridged azobenzenes
and their chemical applications. In Patai’s Chemistry of Functional
Groups Series; Liebman, J. F., Greer, A., Eds.; John Wiley & Sons, 2019;
pp 65−113.
(3) Garcia-Amoros, J.; Velasco, D. In Light-triggered azobenzenes: from
molecular architecture to functional materials; John Wiley & Sons, Inc.,
2014; pp 27−58.
(4) Leonard, E.; Mangin, F.; Villette, C.; Billamboz, M.; Len, C.
Azobenzenes and catalysis. Catal. Sci. Technol. 2016, 6, 379−398.
(5) Mart, R. J.; Allemann, R. K. Azobenzene photocontrol of peptides
and proteins. Chem. Commun. (Cambridge, U. K.) 2016, 52, 12262−
12277.
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.joc.1c00763.
(6) Mahimwalla, Z.; Yager, K. G.; Mamiya, J.-i.; Shishido, A.; Priimagi,
A.; Barrett, C. J. Azobenzene photomechanics: prospects and potential
applications. Polym. Bull. (Heidelberg, Ger.) 2012, 69, 967−1006.
(7) Osorio-Planes, L.; Rodriguez-Escrich, C.; Pericas, M. A.
Photoswitchable Thioureas for the External Manipulation of Catalytic
Activity. Org. Lett. 2014, 16, 1704−1707.
NMR spectra, UV−vis spectra, determination of rate
constants, EPR spectra (PDF)
FAIR data, including the primary NMR FID files, for
compounds 2−5, 7, 9, 12, 16−21 (ZIP)
(8) Peters, M. V.; Stoll, R. S.; Kuehn, A.; Hecht, S. Photoswitching of
basicity. Angew. Chem., Int. Ed. 2008, 47, 5968−5972.
(9) Drillaud, N.; Banaszak-Leonard, E.; Pezron, I.; Len, C. Synthesis
and Evaluation of a Photochromic Surfactant for Organic Reactions in
Aqueous Media. J. Org. Chem. 2012, 77, 9553−9561.
(10) Yu, Y.; Nakano, M.; Ikeda, T. Photomechanics: Directed bending
of a polymer film by light. Nature (London, U. K.) 2003, 425, 145.
AUTHOR INFORMATION
Corresponding Author
Silas C. Blackstock − Department of Chemistry &
Biochemistry, The University of Alabama, Tuscaloosa,
Alabama 35487, United States; orcid.org/0000-0001-
6733-9074; Email: blackstock@ua.edu
■
K
https://doi.org/10.1021/acs.joc.1c00763
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
(11) Szymanski, W.; Beierle, J. M.; Kistemaker, H. A. V.; Velema, W.
A.; Feringa, B. L. Reversible photocontrol of biological systems by the
incorporation of molecular photoswitches. Chem. Rev. 2013, 113,
6114−78.
(12) Warner, D. J.; Keane, K. S.; Saint-Louis, C. J.; Blackstock, S. C.,
Bridged Azobenzenes and Their Biological Applications. In Patai’s
Chemistry of Functional Groups Series, Liebman, J. F., Greer, A., Eds.;
John Wiley & Sons: 2019; pp 251−285.
(13) Robertson, J. M. Crystal structure and configuration of the
isomeric azobenzenes. J. Chem. Soc. 1939, 232−6.
(14) Bouwstra, J. A.; Schouten, A.; Kroon, J. Structural studies of the
system trans-azobenzene/trans-stilbene. I. A reinvestigation of the
disorder in the crystal structure of trans-azobenzene, C12H10N2. Acta
Crystallogr., Sect. C: Cryst. Struct. Commun. 1983, C39, 1121−3.
(15) Mostad, A.; Roemming, C. Refinement of the crystal structure of
cis-azobenzene. Acta Chem. Scand. 1971, 25, 3561−8.
(16) Merino, E.; Ribagorda, M. Control over molecular motion using
the cis-trans photoisomerization of the azo group. Beilstein J. Org. Chem.
2012, 8, 1071−1090.
(32) Guram, A. S.; Rennels, R. A.; Buchwald, S. L. A simple catalytic
method for the conversion of aryl bromides to arylamines. Angew.
Chem., Int. Ed. Engl. 1995, 34, 1348−50.
(33) Hartwig, J. F.; Kawatsura, M.; Hauck, S. I.; Shaughnessy, K. H.;
Alcazar-Roman, L. M. Room-Temperature Palladium-Catalyzed
Amination of Aryl Bromides and Chlorides and Extended Scope of
Aromatic C-N Bond Formation with a Commercial Ligand. J. Org.
Chem. 1999, 64, 5575−5580.
(34) Ohta, K.; Chiba, Y.; Kaise, A.; Endo, Y. Structure-activity
relationship study of diphenylamine-based estrogen receptor (ER)
antagonists. Bioorg. Med. Chem. 2015, 23, 861−867.
(35) Nishimura, N.; Sueyoshi, T.; Yamanaka, H.; Imai, E.; Yamamoto,
S.; Hasegawa, S. Thermal cis-to-trans isomerization of substituted
azobenzenes. II. Substituent and solvent effects. Bull. Chem. Soc. Jpn.
1976, 49, 1381−7.
(36) Bleger, D.; Schwarz, J.; Brouwer, A. M.; Hecht, S. o-
Fluoroazobenzenes as Readily Synthesized Photoswitches Offering
Nearly Quantitative Two-Way Isomerization with Visible Light. J. Am.
Chem. Soc. 2012, 134, 20597−20600.
(17) Cole, L. G.; Gilbert, E. C. J. Am. Chem. Soc. 1951, 73, 5423.
(18) Dias, A. R.; Minas da Piedade, M. E.; Simoes, J. A. M.; Simoni, J.
A.; Teixeira, C.; Diogo, H. P.; Yang, M.; Pilcher, G. Enthalpies of
formation of cis-azobenzene and trans-azobenzene. J. Chem.
Thermodyn. 1992, 24, 439−47.
(19) Schulze, F. W.; Petrick, H. J.; Cammenga, H. K.; Klinge, H.
Thermodynamic properties of the structural analogs benzo[c]-
cinnoline, trans-azobenzene, and cis-azobenzene. Z. Phys. Chem.
(Muenchen, Ger.) 1977, 107, 1−19.
(20) Adamson, A. W.; Vogler, A.; Kunkely, H.; Wachter, R.
Photocalorimetry. Enthalpies of photolysis of trans-azobenzene,
ferrioxalate and cobaltioxalate ions, chromium hexacarbonyl, and
dirhenium decarbonyl. J. Am. Chem. Soc. 1978, 100, 1298−300.
(21) Bandara, H. M. D.; Burdette, S. C. Photoisomerization in
different classes of azobenzene. Chem. Soc. Rev. 2012, 41, 1809−1825.
(22) Blevins, A. A.; Blanchard, G. J. Effect of Positional Substitution
on the Optical Response of Symmetrically Disubstituted Azobenzene
Derivatives. J. Phys. Chem. B 2004, 108, 4962−4968.
(23) Dokic, J.; Gothe, M.; Wirth, J.; Peters, M. V.; Schwarz, J.; Hecht,
S.; Saalfrank, P. Quantum Chemical Investigation of Thermal Cis-to-
Trans Isomerization of Azobenzene Derivatives: Substituent Effects,
Solvent Effects, and Comparison to Experimental Data. J. Phys. Chem. A
2009, 113, 6763−6773.
(24) Crecca, C. R.; Roitberg, A. E. Theoretical Study of the
Isomerization Mechanism of Azobenzene and Disubstituted Azoben-
zene Derivatives. J. Phys. Chem. A 2006, 110, 8188−8203.
(25) Saint-Louis, C. J. Control of Molecular Geometries Using New
Photo-Electro-Switchable Azobenzenes. Ph.D. dissertation. The Uni-
versity of Alabama, Tuscaloosa, AL, 2015.
(26) Blackstock, S. C.; Gray, L.; Kelley, M.; Saint-Louis, C. J. Redox
Auxiliary Catalysis. U.S. Patent 9469602 B2, 2016.
(27) Goulet-Hanssens, A.; Rietze, C.; Titov, E.; Abdullahu, L.;
Grubert, L.; Saalfrank, P.; Hecht, S. Hole Catalysis as a General
Mechanism for Efficient and Wavelength-Independent Z→E Azoben-
zene Isomerization. Chem. 2018, 4, 1740−1755.
(37) Bender, T. P.; Graham, J. F.; Duff, J. M. Effect of Substitution on
the Electrochemical and Xerographic Properties of Triarylamines:
Correlation to the Hammett Parameter of the Substituent and
Calculated HOMO Energy Level. Chem. Mater. 2001, 13, 4105−4111.
(38) Kim, K.-Y. Polyarylamino Redox Arrays for Control of Charge
Transport and Charge Storage. Dissertation, The University of Alabama,
Tuscaloosa, AL, 2003.
(39) Selby, T. D.; Kim, K.-Y.; Blackstock, S. C. Patterned redox arrays
of polyarylamines I. Synthesis and electrochemistry of a p-phenyl-
enediamine and arylamino-appended p-phenylenediamine arrays.
Chem. Mater. 2002, 14, 1685−1690.
(40) Forester, A.; Hay, J.; Thomson, R. Organic Chemistry of Free
Radicals; Academic Press: London, 1968.
(41) Kuder, J. E.; Gibson, H. W.; Wychick, D. Linear free energy
relations. III. Electrochemical characterization of salicylaldehyde anils.
J. Org. Chem. 1975, 40, 875−9.
(42) Walter, R. I. Substituent effects on the properties of stable
aromatic free radicals. The criterion for non-Hammett behavior. J. Am.
Chem. Soc. 1966, 88, 1923−30.
(43) Pearson, G. A.; Rocek, M.; Walter, R. I. Cepstral and direct
analysis of electron spin resonance spectra of substituted triarylami-
nium cation radicals. Correlation of spin distribution with substituent
constants. J. Phys. Chem. 1978, 82, 1185−92.
(44) Rathore, R.; Burns, C. L.; Deselnicu, M. I. Preparation of 1,4:5,8-
dimethano-1,2,3,4,5,6,7,8-octahydro-9,10-dimethoxyanthracenium
−
hexachloroantimonate (4+.SbCl6 ): A highly robust radical-cation salt.
Org. Synth. 2005, 82, 1−9.
(45) Rathore, R.; Kochi, J. K. Isolation of Novel Radical Cations from
Hydroquinone Ethers. Conformational Transition of the Methoxy
Group upon Electron Transfer. J. Org. Chem. 1995, 60, 4399−411.
(46) Pickett, C. J. A simple hydrocarbon electrolyte: completing the
electron-transfer series [Fe4S4(SPh)4]1‑/2‑/3‑/4. J. Chem. Soc., Chem.
Commun. 1985, 323−6.
(47) Kelley, M. D. Electrochromic and Photoelectrochromic
Switching Devices Based On Aryl Amine Redox Chemistry. Ph.D.
dissertation, The University of Alabama, Tuscaloosa, AL, 2014.
(48) Pitre, S. P.; McTiernan, C. D.; Scaiano, J. C. Library of Cationic
Organic Dyes for Visible-Light-Driven Photoredox Transformations.
ACS Omega 2016, 1, 66−76.
(49) Kusumoto, M.; Moriwaki, K.; Nakano, H.; Shirota, Y.
Photochromism of dimethylaminoazobenzene in a molecular glass of
4,4’,4”-tri(N-carbazolyl)triphenylamine. J. Photopolym. Sci. Technol.
1998, 11, 29−32.
(50) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals. Third ed.; Pergamon Press Inc.: Elmsford, New York 10523,
1988; p 391.
(28) Gescheidt, G.; Lamprecht, A.; Heinze, J.; Schuler, B.; Schmittel,
M.; Kiau, S.; Ruechardt, C. Electron-transfer-catalyzed reactions. Part
II. Electron-transfer-catalyzed cis→trans isomerization of 1,1’-azonor-
bornane. Prototype of a reversible two-stage storage system. Helv. Chim.
Acta 1992, 75, 1607−12.
(29) Priewisch, B.; Rueck-Braun, K. Efficient Preparation of
Nitrosoarenes for the Synthesis of Azobenzenes. J. Org. Chem. 2005,
70, 2350−2352.
(30) Merino, E. Synthesis of azobenzenes: the coloured pieces of
molecular materials. Chem. Soc. Rev. 2011, 40, 3835−53.
(31) Louie, J.; Hartwig, J. F. Palladium-catalyzed synthesis of
arylamines from aryl halides. Mechanistic studies lead to coupling in
the absence of tin reagents. Tetrahedron Lett. 1995, 36, 3609−12.
(51) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals. 3rd ed.; Pergamin Press, Inc.: Elmsford, NY, 1988.
L
https://doi.org/10.1021/acs.joc.1c00763
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
pubs.acs.org/joc
Article
(52) Pekmez, N. O.; Can, M.; Yildiz, A. Spectroscopic and
electrochemical observation of hydrogen-bonded imidazole and 2-
aminoimidazole clusters. Acta Chim. Slov. 2007, 54, 131−139.
(53) Kaiser, M.; Leitner, S. P.; Hirtenlehner, C.; List, M.; Gerisch, A.;
Monkowius, U. Azobenzene-functionalized N-heterocyclic carbenes as
photochromic ligands in silver(I) and gold(I) complexes. Dalton Trans.
2013, 42, 14749−14756.
(54) Fraleoni-Morgera, A.; Giorgini, L.; Zanirato, P. [Arylazobenzene-
BF3] dyes: Electronic absorption and NMR spectroscopic evidence for
a novel class of dyes stable in aprotic solvents. Dyes Pigm. 2008, 76,
394−399.
(55) Chiu, K. Y.; Tran, T. T. H.; Wu, C.-G.; Chang, S.-H.; Yang, T.-F.;
Su, Y. O. Electrochemical studies on triarylamines featuring an
azobenzene substituent and new application for small-molecule organic
photovoltaics. J. Electroanal. Chem. 2017, 787, 118−124.
(56) Wegener, M.; Hansen, M. J.; Driessen, A. J. M.; Szymanski, W.;
Feringa, B. L. Photocontrol of Antibacterial Activity: Shifting from UV
to Red Light Activation. J. Am. Chem. Soc. 2017, 139, 17979−17986.
(57) Janus, K.; Matczyszyn, K.; Sworakowski, J.; Biernat, J. F.;
Galewski, Z. Kinetics of Photochromic Reactions of Substituted
Azobenzenes in Solutions, and in Liquid Crystalline and Polymer
Matrices. Mol. Cryst. Liq. Cryst. Sci. Technol., Sect. A 2001, 361, 143−
148.
(58) Hirose, H.; Sawada, K.; Kinoshita, A.; Kawakami, S. Electro-
photographic photoreceptors. JP Patent 61080159 A, 1986.
(59) Lin, H.-T.; Huang, C.-L.; Liou, G.-S. Design, Synthesis, and
Electrofluorochromism of New Triphenylamine Derivatives with AIE-
Active Pendent Groups. ACS Appl. Mater. Interfaces 2019, 11, 11684−
11690.
(60) Schlueter, F.; Riehemann, K.; Kehr, N. S.; Quici, S.; Daniliuc, C.
G.; Rizzo, F. A highly fluorescent water soluble spirobifluorene dye with
a large Stokes shift: synthesis, characterization and bio-applications.
Chem. Commun. (Cambridge, U. K.) 2018, 54, 642−645.
(61) Moreno, J.; Gerecke, M.; Grubert, L.; Kovalenko, S. A.; Hecht, S.
Sensitized Two-NIR-Photon Z→E Isomerization of a Visible-Light-
Addressable Bistable Azobenzene Derivative. Angew. Chem., Int. Ed.
2016, 55, 1544−1547.
(62) Spartan 10, Version 1.1.0; Wavefunction, Inc., Irvine, CA, 2011.
M
https://doi.org/10.1021/acs.joc.1c00763
J. Org. Chem. XXXX, XXX, XXX−XXX