Synthesis and characterization of bis(4-amino-2-bromo-6-methoxy)azobenzene derivatives

Article abstract of DOI:10.3762/bjoc.15.296

Aminoazobenzene derivatives with four ortho substituents with respect to the N-N double bond are a relatively unexplored class of azo compounds that show promise for use as photoswitches in biology. Tetra-ortho-methoxy-substituted aminoazobenzene compounds in particular can form azonium ions under physiological conditions and exhibit red-light photoswitching. Here, we report the synthesis and characterization of two bis(4-amino-2-bromo-6-methoxy)azobenzene derivatives. These compounds form red-light-absorbing azonium ions, but only under very acidic conditions (pH < 1). While the low pK_a makes the azonium form unsuitable, the neutral versions of these compounds undergo trans-to-cis photoisomerization with blue-green light and exhibit slow (τ_1/2 ≈ 10 min) thermal reversion and so may find applications under physiological conditions.

Full text of DOI:10.3762/bjoc.15.296

Synthesis and characterization of
bis(4-amino-2-bromo-6-methoxy)azobenzene derivatives
David Martínez-López1, Amirhossein Babalhavaeji2, Diego Sampedro1
and G. Andrew Woolley*2
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2019, 15, 3000–3008.
1Departamento de Química, Universidad de La Rioja, Centro de
Investigación en Síntesis Química (CISQ), Madre de Dios, 53, 26006
Logroño, Spain and 2Department of Chemistry, University of Toronto,
80 St. George St., Toronto, M5S 3H6, Canada
doi:10.3762/bjoc.15.296
Received: 19 August 2019
Accepted: 04 December 2019
Published: 30 December 2019
Email:
G. Andrew Woolley* - andrew.woolley@utoronto.ca
This article is part of the thematic issue "Molecular switches".
Guest Editor: W. Szymanski
* Corresponding author
Keywords:
© 2019 Martínez-López et al.; licensee Beilstein-Institut.
License and terms: see end of document.
azobenzene; azonium; molecular switches; ortho substitution;
photoisomerization; photoswitch; visible light
Abstract
Aminoazobenzene derivatives with four ortho substituents with respect to the N–N double bond are a relatively unexplored class of
azo compounds that show promise for use as photoswitches in biology. Tetra-ortho-methoxy-substituted aminoazobenzene com-
pounds in particular can form azonium ions under physiological conditions and exhibit red-light photoswitching. Here, we report
the synthesis and characterization of two bis(4-amino-2-bromo-6-methoxy)azobenzene derivatives. These compounds form red-
light-absorbing azonium ions, but only under very acidic conditions (pH < 1). While the low pKa makes the azonium form unsuit-
able, the neutral versions of these compounds undergo trans-to-cis photoisomerization with blue-green light and exhibit slow
(τ1/2 ≈ 10 min) thermal reversion and so may find applications under physiological conditions.
Introduction
The application of photoswitches to control biological targets able for in vivo use [8-10]. Typically, the formation of azonium
has been a driving force for the development of photoswitches ions from aminoazobenzenes occurs at pH < 3.5 [11,12], how-
that operate at wavelengths that are compatible with cells and ever, the pKa of the trans-azonium ion 1 is ca. 7.5 in aqueous
tissues. While many classes of photoswitches are known [1,2], solution (Figure 1) [9,10]. The elevated pKa of 1 has been attri-
few of these are easily adaptable to controlling targets, such as buted to resonance stabilization of the azonium cation together
proteins [3], while simultaneously exhibiting robust photochem- with intramolecular H-bonding between the azonium proton and
istry in the red or near-infrared (NIR) regions of the spectrum methoxy groups in ortho-position to the azo double bond [10].
[4-7]. Azonium ions – protonated forms of azobenzenes – have Since the azonium ion 1 forms under physiological conditions,
recently been found to exhibit photoswitching properties suit- i.e., at neutral pH value in an aqueous solution, it is useful as a
3000

Beilstein J. Org. Chem. 2019, 15, 3000–3008.
photoswitch for the photocontrol of biomolecules [6]. It absorbs constructed by adding a photoswitchable unit to a pharma-
red light, undergoes trans-to-cis photoisomerization, and cophore [14,15], thereby significantly increasing the size of the
relaxes to the trans isomer in the dark on the timescale of compound and decreasing its potential as a drug. Therefore, we
seconds so that pulses of red light can be used to drive multiple were interested in exploring other substitution patterns for these
isomerization cycles [10]. Efforts to apply 1 to photocontrol aminoazobenzene derivatives. To allow for the possibility of
protein–protein interactions have recently been reported [13]. intramolecular H-bond formation, we wished to retain at least
one methoxy group. To reduce the rate of thermal reversion,
Despite the usefulness of 1 as a photoswitch, compounds under- only derivatives with substituents in all four ortho-positions
going photoisomerization at longer wavelengths (>700 nm) were considered. Time-dependent density functional theory
would be valuable since the penetration of light through tissue (TD-DFT) calculations were used to predict the absorption
is enhanced in the NIR window [7]. Longer-wavelength absorp- wavelengths of possible derivatives. Based on these considera-
tion is achieved by compounds 2, 3, and related derivatives tions, we carried out the synthesis and photochemical character-
(Figure 1) [9]. However, the lifetime of thermal reversion of the ization of compounds 4 and 5.
cis isomer of 2 is only ≈1 ms at a neutral pH value, and the pKa
for trans-azonium ion formation is ca. 2.6. The low pKa was at-
tributed to a steric clash between the methoxy groups in meta- Computational chemistry
Results and Discussion
position to the azo double bond and the six-membered Calculations were performed using density functional theory
morpholino ring [9]. The rapid thermal reversion was attributed (DFT) methods (B3LYP/6-31+G**) to optimize geometry, and
to the removal of two ortho-methoxy groups, leading to dimin- TD-DFT with a Solvation Model based on Density (SMD) to
ished steric strain in the transition state for reversion [9]. In calculate absorption wavelength maxima. These computational
compound 3, all four ortho-positions are substituted, and the methods have been used successfully with related compounds
meta-oxygen substituents are part of dioxane rings so that the [16,17]. The relative stability of different conformations of the
steric clash with the para-amino substituents is reduced. Com- molecule was calculated, i.e., with the methoxy substituents on
pound 3 and derivatives with pyrrolidino groups in the para-po- the same or on the opposite side of the N–N double bond. The
sitions were shown to be effective NIR switches, undergoing conformation where both methoxy groups were on the opposite
isomerization with 720 nm light under physiological conditions side was found to be the most stable one, although the confor-
[8]. While the photoswitching properties of 3 are suitable for mation where both methoxy groups were on the same side was
the use in biological systems, the overall size of 3 may limit the also predicted to be significantly populated at 20 °C (see Sup-
possibilities for the use as a component of photopharmaceutical porting Information File 1). Calculating the effect of the substi-
agents. Currently, most photopharmaceutical agents are tution pattern on the pKa value is problematic [18] and was not
Figure 1: Structures of azonium ions studied.
3001

Beilstein J. Org. Chem. 2019, 15, 3000–3008.
attempted here. Figure 2 shows calculated structures and spec- Photochemical characterization
tra of the neutral forms of simplified models of 4 and 5, with Despite the morpholino substituent, compound 4 was found to
either a pyrollidino or piperidino substituent in para-position, as be insoluble in water. We therefore dissolved 4 in DCM to
well as the corresponding azonium ions. Calculations indicated obtain the UV–vis spectrum of the neutral form. Addition of
that the nature of the respective amino substituent did not have TFA to this solution produced the corresponding azonium ion,
a large effect on the positions of wavelength maxima. These and the spectra of the neutral and azonium forms of 4 are shown
models predicted that compounds 4 and 5 should absorb at as solid lines in Figure 2. Observed absorption maxima were at
longer wavelengths than 1 [10]. Figure 2 also shows experimen- 426 (neutral form) and 640 nm (azonium form). While the ob-
tal spectra, which are discussed below.
served absorption maximum wavelength was close to that pre-
dicted for the neutral form, the observed absorption maximum
wavelength of the azonium ion was significantly higher than
Synthesis
The overall synthetic route that was taken is shown in predicted, although the signal’s tail was less pronounced. The
Scheme 1. The azo compound 8, carrying two ortho-methoxy absorption maximum wavelength of the azonium ion was also
groups, was prepared from 7 using an oxidative coupling ap- higher than that observed for compound 1 [10], as predicted.
proach [19]. The para-chloro substituents were then replaced by
amino substituents using a Buchwald–Hartwig coupling [20]. We confirmed that the neutral form of 4 underwent photoisom-
Since calculations predicted that 5- and 6-membered rings erization. Exposure of a solution of 4 in DCM to 440 nm light
would have similar effects on the positions of the absorption led to a photostationary state (PSS) in which the absorbance at
maxima, we opted to use 6-membered rings, specifically a 440 nm was diminished and that at 330 nm slightly enhanced
morpholino substituent and a piperazino substituent in an (Figure 3). An estimated PSS of ca. 60% Z-isomer was calcu-
attempt to enhance water solubility. Late-stage functionaliza- lated as described in the Experimental section. Thermal relaxa-
tion of the ortho-position was carried out through a tion from the PSS was monitored by UV–vis spectroscopy,
palladium(II)-catalyzed C–H activation, resulting in ortho-bro- recording a spectrum every minute. As shown in Figure 3, a
minated azobenzenes [21].
half-life of 6.5 minutes was obtained at room temperature.
Figure 2: a) Structures of model compounds used for computations (see Experimental section; in calculations, water was set as the solvent).
b) Calculated spectra of neutral forms i and ii (black spectra) and azonium forms iii and iv (red spectra), carrying either para-piperidino (i and iii,
dashed lines) or para-pyrrolidino (ii and iv, dash-dotted lines) substituents. Solid lines show experimental spectra of 4 in DCM (c ≈ 15 µM) without
(black) or with TFA added (red).
3002

Beilstein J. Org. Chem. 2019, 15, 3000–3008.
Scheme 1: Synthesis of bis(4-amino-2-bromo-6-methoxy)azobenzene compounds.
PSS (dotted line). Thermal reversion from the cis isomer was
monitored by UV–vis spectroscopy, recording a UV–vis spec-
trum every 3 minutes, and a half-life of 12.6 minutes was ob-
tained.
The formation of the azonium ion of 5 in aqueous solution was
explored thereafter. Addition of hydrochloric acid to a neutral
solution of 5 was carried out. Spectra at different pH values are
shown in Figure 4c. As can be seen, the formation of the
azonium ion of compound 5 required strongly acidic conditions.
Even at pH ≈ 0.1, a substantial fraction of the neutral species
was still present, implying that the pKa value of the azonium ion
was below ca. 0.2. This pKa value was at least 7 pH units lower
than for the tetra-ortho-methoxy compound 1. As noted above,
the presence of the piperazino amino groups made compound 5
Figure 3: a) UV–vis spectra of 4 in DCM (ca. 15 µM) at the PSS and
440 nm irradiation (thick dotted line; ca. 60% Z-isomer), and during
thermal reversion to the dark-adapted state (solid black line). b) Time
course of thermal reversion at 22 °C. Data were fitted to a single expo-
nential decay equation (solid line).
To enhance the compounds’ water solubility, the morpholino doubly positively charged at a neutral pH value. This feature
substituents were replaced by piperazino groups because the was expected to reduce the pKa value of the azonium ion
secondary amino groups on the piperazino units were expected through electrostatic effects [23]. The first pKa value of substi-
to have pKa values near 10 [22], and so should be protonated at tuted piperazines falls in the range ca. 4–6 (vs 9–10 for the
neutral pH, creating a doubly charged species. As anticipated, second pKa value) [22]. In addition, the electron-withdrawing
compound 5 was found to be much more water-soluble than 4. bromine atoms were also expected to lower the azonium ion’s
pKa value (the pKa values of 2-bromobenzoic acid and unsubsti-
The UV–vis spectrum of 5 at a neutral pH value is shown in tuted benzoic acid are 2.85 and 4.2, respectively [24]). While
Figure 4. Irradiation with blue light at 440 nm produced the electrostatic effects on the pKa value could be ameliorated
3003

Beilstein J. Org. Chem. 2019, 15, 3000–3008.
Figure 4: a) UV–vis spectra of 5 in aqueous solution (c ≈ 15 µM, 5% methanol, pH 7) at the PSS and 440 nm irradiation (thick dotted line; ca. 60%
Z-isomer) and during thermal reversion to the dark-adapted state (solid black line). b) Time course of thermal reversion at 22 °C. Data were fitted to a
single exponential decay equation (solid line). c) Spectra of an aqueous solution of 5 upon addition of hydrochloric acid, as indicated by pH values.
The wavelength of the absorption maximum of the azonium ion was 575 nm.
Experimental
(i.e., by adding negatively charged groups), this would place ad-
ditional constraints on the general applicability of the com- General
pounds. In addition, unlike compound 4, the wavelength of the All commercial materials (solvents, reagents, and substrates)
absorption maximum of the azonium ion of 5 was not red- were used as received. SilicaFlash silica gel, P60, 40–63 µm
shifted relative to that of 1. Conceivably, the wavelength of particle size (SiliCycle) was used for column chromatography.
maximum absorbance was affected by the charged piperazino High-performance liquid chromatography was performed on a
groups; calculations were done with uncharged piperidino sub- PerkinElmer Series 200 pump with a Waters 2487 Dual λ Ab-
stituents, as shown in Figure 2. Nevertheless, the lack of a red- sorbance Detector connected to an eDAQ PowerChrom 280
shift, combined with the significantly lowered pKa value, made recorder. One-dimensional 1H and 13C NMR spectra were re-
5 unsuitable as an azonium photoswitch under physiological corded on a Varian UnityPlus 500 MHz or Varian Mercury
conditions.
400 MHz spectrometer. Chemical shifts are reported in ppm,
and the signals were referenced to residual undeuterated sol-
Despite this undesired effect on the pKa value, the steric bulk vent signals. Mass spectra were recorded using an Agilent 6538
introduced by the bromine substituents did appear to slow ther- mass spectrometer with a Q-TOF ionization source or a JEOL
mal relaxation of the neutral (unprotonated) azo forms of these AccuTOF mass spectrometer with a DART ionization source.
compounds. Species 5 could be switched with blue and green
light under physiological conditions and be thermally relaxed Synthesis and characterization
with a half-life of 12 minutes. This relaxation rate was substan- 4-Chloro-2-methoxyanilinium chloride (7): (1) 2-Amino-5-
tially lower than other blue-green-absorbing azo compounds chlorophenol (6, 3.5 mmol, 500 mg) was dissolved in THF at
without substituents in all four ortho-positions relative to the room temperature. Then, Boc2O (7 mmol, 1.52 g) was added
azo unit, which showed half-lives ranging from 50 ms [25] to a and the resulting mixture was stirred for 18 h at room tempera-
few seconds [26,27]. Instead, compounds 4 and 5 exhibited ture. The reaction progress was monitored by TLC. When the
photoswitching properties similar to those reported for tetra- reaction was complete, the solvent was removed under reduced
ortho-thiol-substituted azobenzenes [17].
pressure, resulting in a yellowish oil. This was purified by
column chromatography using hexane/ethyl acetate, 4:1, v/v as
eluent. The product, phenol tert-butyl (4-chloro-2-hydroxy-
Conclusion
Substitution of p-aminoazobenzene with ortho-bromo and phenyl)carbamate, was used directly in the next step. (2) tert-
ortho-methoxy groups (i.e., a 2-bromo-6-methoxy substitution Butyl (4-chloro-2-hydroxyphenyl)carbamate (2 mmol, 500 mg)
pattern) was found to lower the pKa of the azonium ion such was dissolved in DMF. Then, an aqueous solution of K2CO3
that it fell outside the normal physiological range. The neutral (6 mmol, 828 mg) was added, together with methyl iodide
version of this compound nevertheless underwent trans-to-cis (Caution: toxic, potential carcinogen; 6 mmol, 851 mg). The
photoisomerization in the presence of blue-green light and mixture was stirred for 18 h until the starting reagent was con-
exhibited slow thermal relaxation (τ1/2 ca. 10 min).
sumed. Then, the solvent was removed under reduced pressure,
3004

Beilstein J. Org. Chem. 2019, 15, 3000–3008.
resulting in a brown oil. This was poured into water and 118.5, 107.7, 98.9, 66.9, 56.7, 48.6; HRMS (m/z): [M + H]+
extracted three times with DCM. The organic fractions were calcd for C22H29N4O4, 413.2189; found, 413.2186.
collected and evaporated under reduced pressure. The product,
tert-butyl (4-chloro-2-methoxyphenyl)carbamate, was used (E)-1,2-Bis(2-bromo-6-methoxy-4-morpholinophenyl)-
directly in the next step. (3) tert-Butyl (4-chloro-2-methoxy- diazene (4): (E)-1,2-Bis(2-methoxy-4-morpholino-
phenyl)carbamate (400 mg, 1.16 mmol) was dissolved in 3 mL phenyl)diazene (9, 30 mg, 0.07 mmol) was dissolved in DCM.
of ethyl acetate. Then, 9 mL of fuming hydrochloric acid were To this solution, palladium acetate (1.6 mg, 0.007 mmol) was
added dropwise to the reaction mixture with vigorous stirring. added, and the resulting mixture was stirred for 15 minutes.
The resulting mixture was stirred for 1 hour. Then, the solvent Then, N-bromosuccinimide (28.6 mg, 0.16 mmol) was added to
was evaporated under reduced pressure, giving a brown oil. the reaction. The reaction mixture was stirred for additional
This was added to 30 mL of hexane in an ice bath to precipitate 30 minutes until completed. The solvent was evaporated under
the title compound. Finally, the product 7 (420 mg, 77% over reduced pressure, and the resulting oil was purified by column
three steps) was isolated by filtration. 1H NMR (400 MHz, chromatography using hexane/ethyl acetate, 1:2, v/v as eluent to
methanol-d4) δ 7.36 (d, J = 8.4 Hz, 1H), 7.30 (d, J = 2.1 Hz, obtain 29 mg (70%) of 4. 1H NMR (300 MHz, chloroform-d)
1H), 7.11 (dd, J = 8.4, 2.1 Hz, 1H), 3.99 (s, 3H).
δ 7.91 (s, 2H), 6.67 (s, 2H), 4.03 (s, 6H), 3.89 (m, 8H), 3.17 (s,
8H); 13C NMR (101 MHz, chloroform-d) δ 157.3, 153.7, 138.8,
(E)-1,2-Bis(4-chloro-2-methoxyphenyl)diazene (8): 4-Chloro- 122.6, 110.8, 104.9, 67.0, 58.6, 51.9; HRMS (m/z): [M + H]+
2-methoxyanilinium chloride (7, 2.5 mmol, 400 mg) was dis- calcd for C22H27Br2N4O4, 569.0393; found, 569.0393.
solved in acetic acid (2 mL), and boric acid (2.11 mmol,
135 mg) was added to the mixture, followed by sodium perbo- Di-tert-butyl 4,4'-(diazene-1,2-diyl)bis(3-methoxy-4,1-
rate (2.5 mmol, 680 mg) in three portions over 15 minutes. phenylene)-(E)-bis(1λ4-piperazine-1-carboxylate) (10):
Then, the reaction mixture was heated at 70 ºC for 18 hours. (E)-1,2-Bis(4-chloro-2-methoxyphenyl)diazene (8, 100 mg,
The reaction progress was monitored by TLC, and when the 0.32 mmol) was dissolved in toluene in a pressure tube. Then,
reaction was complete, the solvent was removed under reduced 1-Boc-piperazine (180 mg, 0.96 mmol), tris(dibenzylideneace-
pressure. The resulting oil was poured into water and extracted tone)dipalladium(0) (29.3 mg, 0.032 mmol), RuPhos (29.8 mg,
with DCM three times. The organic fractions were collected, 0.064 mmol), and cesium carbonate (302.4 mg, 0.96 mmol)
and the solvent was removed. The resulting oil was purified by were added. The mixture was heated at 100 ºC in a pressure
column chromatography using hexane/ethyl acetate, 4:1, v/v as tube for 36 hours. The reaction progress was monitored by TLC
eluent. This way, 250 mg (32%) of 8 could be obtained. until the starting reagent was consumed. The solvent was evap-
1H NMR (300 MHz, chloroform-d) δ 7.59 (d, J = 8.6 Hz, 2H), orated under reduced pressure, and the resulting oil was
7.07 (d, J = 2.1 Hz, 2H), 6.98 (dd, J = 8.6, 2.1 Hz, 2H), 4.01 (s, extracted with DCM three times. The final mixture was puri-
6H); 13C NMR (75 MHz, chloroform-d) δ 157.4, 141.4, 138.3, fied by column chromatography using hexane/ethyl acetate, 1:3,
121.2, 118.5, 113.3, 56.7; HRMS (m/z): [M + H]+ calcd for v/v as eluent to obtain 130 mg (65%) of the product 10.
C14H15N2O2Cl2, 311.0349; found, 311.0341.
1H NMR (400 MHz, DCM-d2) δ 7.57 (m, 2H), 6.50 (m, 4H),
3.99 (s, 6H), 3.58 (m, 8H), 3.29 (m, 8H), 1.47 (s, 18H).
(E)-1,2-Bis(2-methoxy-4-morpholinophenyl)diazene (9):
(E)-1,2-Bis(4-chloro-2-methoxyphenyl)diazene (8, 100 mg, Di-tert-butyl 4,4'-(diazene-1,2-diyl)bis(3-bromo-5-methoxy-
0.32 mmol) was dissolved in toluene in an Ace pressure tube 4,1-phenylene)-(E)-bis(1λ4-piperazine-1-carboxylate) (11):
(10.2 cm × 8 mm). Then, morpholine (84 mg, 0.96 mmol), Di-tert-butyl 4,4'-(diazene-1,2-diyl)bis(3-methoxy-4,1-
tris(dibenzylideneacetone)dipalladium(0) (29.3 mg, phenylene)-(E)-bis(1λ4-piperazine-1-carboxylate) (10, 30 mg,
0.032 mmol), RuPhos (29.8 mg, 0.064 mmol), and cesium 0.04 mmol) was dissolved in DCM, palladium acetate (1 mg,
carbonate (302.4 mg, 0.96 mmol) were added. The mixture was 0.004 mmol) was added, and the resulting mixture was stirred
heated at 100 ºC in the pressure tube for 24 hours. The reaction for 15 minutes. Then, N-bromosuccinimide (22 mg, 0.1 mmol)
progress was monitored by TLC until the starting reagent was was added to the reaction. The reaction mixture was stirred for
consumed. The solvent was evaporated under reduced pressure, additional 30 minutes until completion. The solvent was evapo-
and the resulting oil was extracted with DCM three times. The rated under reduced pressure and the resulting oil was purified
resulting crude product was purified by column chromatogra- by column chromatography using hexane/ethyl acetate, 1:1, v/v
phy using hexane/ethyl acetate, 1:4, v/v as eluent to obtain as eluent to obtain 19 mg (50%) of the product 11. 1H NMR
95 mg (72%) of 9. 1H NMR (400 MHz, chloroform-d) δ 7.71 (s, (400 MHz, DCM-d2) δ 7.74 (s, 2H), 6.61 (s, 2H), 3.93 (s, 6H),
2H), 6.48 (m, 4H), 4.01 (s, 6H), 3.87 (m, 8H), 3.28 (m, 8H); 3.53 (m, 8H), 3.00 (m, 8H), 1.39 (s, 18H); 13C NMR
13C NMR (101 MHz, chloroform-d) δ 158.1, 154.0, 136.7, (126 MHz, DCM-d2) δ 152.9, 150.2, 149.4, 134.1, 117.5, 106.2,
3005

Beilstein J. Org. Chem. 2019, 15, 3000–3008.
75.1, 52.1, 47.0, 23.7; HRMS (m/z): [M + H]+ calcd for this approach, estimated PSS values of 60% ( 10%) Z-isomer
C32H45Br2N6O6, 767.1767; found, 767.1762.
were obtained.
(E)-1,2-Bis(2-bromo-6-methoxy-4-(piperazin-1-yl)- Thermal relaxation rates
phenyl)diazene (5): Di-tert-butyl 4,4'-(diazene-1,2-diyl)bis(3- The cuvette was irradiated with blue light for 1 minute, gently
bromo-5-methoxy-4,1-phenylene)-(E)-bis(1λ4-piperazine-1- mixed by pipetting, then immediately capped with Parafilm to
carboxylate) (11, 15 mg, 0.02 mmol) was dissolved in 5 mL of prevent evaporation, and placed inside the spectrophotometer.
DCM, and 0.25 mL of TFA was added. The resulting mixture The sample was periodically scanned (250 to 600 nm for each
was stirred for 16 hours. 1 mL of a 10% aqueous solution of scan; integration time: 2s; scan speed: 480 nm/min) in 2-minute
sodium bicarbonate was added to neutralize the compound. intervals. Absorbance data vs time were then fitted to a single
Then, the solvent was evaporated to produce the product 5 exponential equation to obtain thermal relaxation half-lives
(90%) without further purification needed. 1H NMR (400 MHz, using Equation 1.
DCM-d2) δ 7.92 (s, 2H), 6.97 (s, 2H), 4.11 (s, 6H), 3.45 (m,
8H), 3.00 (m, 8H); 13C NMR (126 MHz, DCM-d2) δ 158.9,
154.0, 140.2, 123.0, 111.5, 107.2, 66.9, 57.3, 45.1; HRMS
(1)
(m/z): [M + H]+ calcd for C22H29Br2N6O2, 567.0719; found,
567.0713.
Computational methods
DFT calculations were performed using the Gaussian 09 suite
of programs at the B3LYP level of theory using a 6-31+G(d,p)
UV–vis spectroscopy
UV–vis spectra were recorded on a PerkinElmer LAMBDA 35, basis set [28]. Optimizations were followed by harmonic oscil-
Shimadzu UV-2401PC, or an Ocean Optics USB4000 diode lator frequency calculations at the same level of theory to verify
array spectrophotometer. The temperature was maintained at the absence of imaginary frequencies. Though an exhaustive
22 °C for all measurements (Quantum Northwest temperature conformational search was not performed for any of the species,
controller), and 10 mm or 1.5 mm quartz cuvettes (Hellma the following calculations were performed: To ensure that the
Analytics) were used. Samples were prepared in sodium phos- conformation shown in Figure 2 was the thermodynamically
phate buffer, pH 7.0, or in DCM, as described, at nominal con- most stable arrangement around the azo moiety, free energies
centrations of 15 µM. The pH value was adjusted by adding for two alternative arrangements – conformer 2: with both
microliter volumes of aq HCl or NaOH (as to not change the bromine atoms on the same side of the N–N double bond and
total volume of the sample solutions significantly). The pH conformer 3: with a bromine atom in the position of the me-
values were measured directly in the samples using a combina- thoxy group that H-bonded with the azonium proton (see Sup-
tion pH electrode.
porting Information File 1) – were calculated for each com-
pound at 298.15 K. Both alternate conformers were predicted to
have higher energies in vacuo (conformer 2: 1.1 kJ/mol for the
Photoisomerization
UV irradiation was performed by placing a 365 nm LED (897- 6-membered ring and 2.2 kJ/mol for the 5-membered ring;
LZ440U610; LED Engin) operating at 68 mW/cm2 above the conformer 3: 20 kJ/mol for the 6-membered ring and 19 kJ/mol
sample tube for 1 minute. For blue light irradiation, a 440 nm for the 5-membered ring). Thus, conformer 2 was slightly
LED (Luxeon III Star LED Royal Blue Lambertian; Luxeon higher in energy than conformer 1 and should be populated at
Star LEDs) operating at 40 mW/cm2 at 700 mA was used in the room temperature. However, we did not carry out TD-DFT
experiments.
calculations for conformer 2. Second, for the neutral species,
the N=N–C–C dihedral angles in the optimized structures were
manually set to 20° to generate new input files. Reoptimization
yielded the same structures and free energies. The optimized
Estimates of the percentage of cis/trans
isomers in PSSs
Based on NMR spectra, we assumed that the fraction of geometries of all structures were subjected to TD-SCF calcula-
Z-isomer present at equilibrium in the dark was negligible. tions using the same functionals and basis set, assuming the first
Since UV–vis spectra obtained during thermal reversion exhib- 15 singlet excitations, and by applying a SMD (assuming water
ited an isosbestic point, we assumed that only E- and Z-isomers as the solvent) to implicitly approximate the effect of the sol-
contributed to the spectrum at any time. An estimate for the vent [29]. TD-SCF data were used to generate the simulated
spectrum of pure Z-isomer could therefore be obtained simply UV–vis spectra by applying a Gaussian function with 0.333 eV
by subtracting the contribution of the E-isomer from the spec- peak half-width at half-height to each transition. For the lowest-
trum of the PSS, with the restriction that the absorbance of the energy conformer of both protonated species, single point
Z-isomer could not be less than zero at any wavelength. With calculations were performed at B3LYP/6-31+G(d,p) with a
3006

Beilstein J. Org. Chem. 2019, 15, 3000–3008.
10.Samanta, S.; Babalhavaeji, A.; Dong, M.-x.; Woolley, G. A.
SMD (assuming water as the solvent), and molecular orbitals
corresponding to HOMO and LUMO were visualized using
GaussView, with the isovalue set to 0.02. HOMO-to-LUMO
transitions correspond to the longest wavelength with high
oscillator strength. In general, HOMOs were found to be more
delocalized than LUMOs (see Supporting Information File 1).
Angew. Chem., Int. Ed. 2013, 52, 14127–14130.
doi:10.1002/anie.201306352
11.Stoyanov, S.; Antonov, L.; Stoyanova, T.; Petrova, V. Dyes Pigm.
1996, 32, 171–185. doi:10.1016/0143-7208(96)00016-2
12.Stoyanova, T.; Stoyanov, S.; Antonov, L.; Petrova, V. Dyes Pigm.
1996, 31, 1–12. doi:10.1016/0143-7208(95)00095-x
13.Yasuike, N.; Blacklock, K. M.; Lu, H.; Jaikaran, A. S. I.; McDonald, S.;
Uppalapati, M.; Khare, S. D.; Woolley, G. A. ChemPhotoChem 2019, 3,
431–440. doi:10.1002/cptc.201900016
Supporting Information
14.Hüll, K.; Morstein, J.; Trauner, D. Chem. Rev. 2018, 118,
10710–10747. doi:10.1021/acs.chemrev.8b00037
Supporting Information File 1
15.Velema, W. A.; Hansen, M. J.; Lerch, M. M.; Driessen, A. J. M.;
Szymanski, W.; Feringa, B. L. Bioconjugate Chem. 2015, 26,
2592–2597. doi:10.1021/acs.bioconjchem.5b00591
16.Bléger, D.; Schwarz, J.; Brouwer, A. M.; Hecht, S. J. Am. Chem. Soc.
2012, 134, 20597–20600. doi:10.1021/ja310323y
NMR spectra and further computational data.
[https://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-15-296-S1.pdf]
17.Samanta, S.; McCormick, T. M.; Schmidt, S. K.; Seferos, D. S.;
Woolley, G. A. Chem. Commun. 2013, 49, 10314–10316.
doi:10.1039/c3cc46045b
18.Alongi, K. S.; Shields, G. C. Annu. Rep. Comput. Chem. 2010, 6,
113–118. doi:10.1016/s1574-1400(10)06008-1
Funding
We are grateful to the Natural Sciences and Engineering
Research Council of Canada and MINECO/FEDER (CTQ2017-
87372-P) for financial support. D. M.-L. thanks the U. de La
Rioja for a fellowship.
19.Merino, E. Chem. Soc. Rev. 2011, 40, 3835–3853.
doi:10.1039/c0cs00183j
20.Maiti, D.; Fors, B. P.; Henderson, J. L.; Nakamura, Y.; Buchwald, S. L.
Chem. Sci. 2011, 2, 57–68. doi:10.1039/c0sc00330a
21.Konrad, D. B.; Frank, J. A.; Trauner, D. Chem. – Eur. J. 2016, 22,
4364–4368. doi:10.1002/chem.201505061
ORCID® iDs
Diego Sampedro - https://orcid.org/0000-0003-2772-6453
G. Andrew Woolley - https://orcid.org/0000-0002-3446-2639
22.Khalili, F.; Henni, A.; East, A. L. L. J. Chem. Eng. Data 2009, 54,
2914–2917. doi:10.1021/je900005c
23.Borisenko, V.; Sansom, M. S. P.; Woolley, G. A. Biophys. J. 2000, 78,
1335–1348. doi:10.1016/s0006-3495(00)76688-4
24.Woolley, E. M.; Tomkins, J.; Hepler, L. G. J. Solution Chem. 1972, 1,
341–351. doi:10.1007/bf00715992
Preprint
A non-peer-reviewed version of this article has been previously published
as a preprint doi:10.3762/bxiv.2019.88.v1
25.Chi, L.; Sadovski, O.; Woolley, G. A. Bioconjugate Chem. 2006, 17,
670–676. doi:10.1021/bc050363u
26.Beharry, A. A.; Sadovski, O.; Woolley, G. A. Org. Biomol. Chem. 2008,
6, 4323–4332. doi:10.1039/b810533b
27.Sadovski, O.; Beharry, A. A.; Zhang, F.; Woolley, G. A.
Angew. Chem., Int. Ed. 2009, 48, 1484–1486.
References
1. García-Iriepa, C.; Marazzi, M.; Frutos, L. M.; Sampedro, D. RSC Adv.
2013, 3, 6241–6266. doi:10.1039/c2ra22363e
2. Szymański, W.; Beierle, J. M.; Kistemaker, H. A. V.; Velema, W. A.;
Feringa, B. L. Chem. Rev. 2013, 113, 6114–6178.
doi:10.1021/cr300179f
3. Kumita, J. R.; Flint, D. G.; Smart, O. S.; Woolley, G. A.
Protein Eng., Des. Sel. 2002, 15, 561–569.
doi:10.1093/protein/15.7.561
4. Bléger, D.; Hecht, S. Angew. Chem., Int. Ed. 2015, 54, 11338–11349.
doi:10.1002/anie.201500628
doi:10.1002/anie.200805013
28.Gaussion 09, Revision B.01; Gaussian, Inc.: Wallingford, CT, 2009.
29.Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B 2009,
113, 6378–6396. doi:10.1021/jp810292n
5. Samanta, S.; Qin, C.; Lough, A. J.; Woolley, G. A.
Angew. Chem., Int. Ed. 2012, 51, 6452–6455.
doi:10.1002/anie.201202383
6. Dong, M.; Babalhavaeji, A.; Samanta, S.; Beharry, A. A.;
Woolley, G. A. Acc. Chem. Res. 2015, 48, 2662–2670.
doi:10.1021/acs.accounts.5b00270
7. Lentes, P.; Stadler, E.; Röhricht, F.; Brahms, A.; Gröbner, J.;
Sönnichsen, F. D.; Gescheidt, G.; Herges, R. J. Am. Chem. Soc. 2019,
141, 13592–13600. doi:10.1021/jacs.9b06104
8. Dong, M.; Babalhavaeji, A.; Collins, C. V.; Jarrah, K.; Sadovski, O.;
Dai, Q.; Woolley, G. A. J. Am. Chem. Soc. 2017, 139, 13483–13486.
doi:10.1021/jacs.7b06471
9. Dong, M.; Babalhavaeji, A.; Hansen, M. J.; Kálmán, L.; Woolley, G. A.
Chem. Commun. 2015, 51, 12981–12984. doi:10.1039/c5cc02804c
3007

Beilstein J. Org. Chem. 2019, 15, 3000–3008.
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(https://creativecommons.org/licenses/by/4.0). Please note
that the reuse, redistribution and reproduction in particular
requires that the authors and source are credited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(https://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.15.296
3008