Combining Sanford Arylations on Benzodiazepines with the Nuisance Effect

Article abstract of DOI:10.1002/adsc.201700626

5-Phenyl-1,3-dihydro-2H-1,4-benzodiazepin-2-ones react under palladium- and visible light photoredox catalysis, in refluxing methanol, with aryldiazonium salts to afford the respective 5-(2-arylphenyl) analogues. With 2- or 4-fluorobenzenediazonium derivatives, both fluoroaryl- and methoxyaryl- products were obtained, the latter resulting from a S_NAr on the fluorobenzenediazonium salt (“nuisance effect”). A computational DFT analysis of the palladium-catalysed and the palladium/ruthenium-photocalysed mechanism for the functionalization of benzodiazepines indicated that, in the presence of the photocatalyst, the reaction proceeds via a low-energy SET pathway avoiding the high-energy oxidative addition step in the palladium-only catalysed reaction pathway. (Figure presented.).

Full text of DOI:10.1002/adsc.201700626

Title: Combining Sanford Arylations on Benzodiazepines with the
Nuisance Effect
Authors: Raysa Khan, Sarote Boonseng, Paul Kemmitt, Robert Felix,
Hazel Cox, Simon Coles, Graham Tizzard, Gareth Williams,
Olivia Simmonds, Jessica Harvey-Cox, John Atack, and John
Spencer
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Adv. Synth. Catal. 10.1002/adsc.201700626
Link to VoR: http://dx.doi.org/10.1002/adsc.201700626

10.1002/adsc.201700626
Advanced Synthesis & Catalysis
UPDATE
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Combining Sanford Arylations on Benzodiazepines with the
Nuisance Effect
Raysa Khan,a Sarote Boonseng,a Paul D. Kemmitt,b Robert Felix,c Simon J. Coles,d Graham J.
Tizzard,d Gareth Williams,e Olivia Simmonds,e Jessica-Lily Harvey,e John Atack,e Hazel Cox a and
John Spencera,*
a
Department of Chemistry, School of Life Sciences, University of Sussex, Falmer, BN1 9QJ, UK.
Email: j.spencer@sussex.ac.uk
b
AstraZeneca, Mereside Alderley Park, Macclesfield, SK10 4TG, UK.
Tocris Bioscience, the Watkins Building, Atlantic Road, Avonmouth, Bristol, BS11 9QD, UK.
c
d UK National Crystallography Service, School of Chemistry, University of Southampton, Highfield, Southampton, SO17
1BJ, U.K.
e Sussex Drug Discovery Centre, School of Life Sciences, University of Sussex, Falmer, BN1 9QJ, UK.
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.((Please
delete if not appropriate))
Abstract. 5-Phenyl-1,3-dihydro-2H-1,4-benzodiazepin-2-ones
react under palladium- and visible light photoredox catalysis, in
refluxing methanol, with aryldiazonium salts to afford the
respective 5-(2-arylphenyl) analogues. With 2- or 4-
fluorobenzenediazonium derivatives, both fluoroaryl- and
methoxyaryl- products were obtained, the latter resulting from a
SNAr on the fluorobenzenediazonium salt (“nuisance effect”).
A computational DFT analysis of the palladium-catalysed and
the palladium/ruthenium-photocalysed mechanism for the
functionalization of benzodiazepines indicated that in the
presence of the photocatalyst the reaction proceeds via a low-
energy SET pathway avoiding the high-energy oxidative
addition step in the palladium-only catalysed reaction pathway.
Keywords: C-H activation; benzodiazepine;
photocatalysis; palladacycle, DFT.
conditions, relatively high commercial cost, and
multistep synthesis of iodonium salts[4] (ArIAr’+),
coupled with a poor atom economy (Ar-I is a
byproduct) prompted us to consider a visible-light
photocatalyzed Pd-mediated protocol involving
diazonium salts[5].
Introduction
There is a growing impetus for atom economical
routes to high value end products employing late
stage functionalization (LSF) processes[1]. These are
particularly desirable in medicinal chemistry since
they increase diversity and chemical space and enable
rapid SAR (structure activity relationship) and
ADME-Tox (Absorption, distribution, metabolism,
elimination-toxicity) feedback that is key to costly,
high attrition, drug development. Late stage C-H
activation is a powerful tool in generating novel
compounds for biological evaluation[2]. We recently
described a palladium-catalyzed ortho-arylation of
benzodiazepines employing iodonium salts in acetic
acid under microwave irradiation[3]. The harsh
Results and Discussion
Our initial reaction trials were performed on the
benzodiazepine
1a,
using
the
2-fluoro-
benzenediazonium salt 2a under reflux (external oil
bath temperature set at 700 C). To our surprise, in
addition to the expected product 3a, we were able to
isolate the ether product 4a. However, reaction of the
3-isomer 2b led exclusively to the fluorobiaryl
derivative 3b, whereas the 4-isomer 2c
1
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700626
Advanced Synthesis & Catalysis
afforded a mixture of fluorobiaryl 3c and methoxy
product 4c (Scheme 1). Repeating the reaction in
ethanol led to the ethyl ether 4d.
Entry
Lamps
26W
Pd(OAc)2
(mol %)
Ru(bpy)3Cl2.
6H2O
Temp.
(oC)
Conv.
LC/MS (%)
(mol %)
1
2
3
4
5
6
Yes
Yes
No
10
10
10
0
2.5
2.5
2.5
2.5
0
rt
52
61
35
0
Refluxa
Refluxa
Refluxa
Yes
Yes
No
10
10
Refluxa
57
20
b
0
Scheme 1. Benzodiazepine library synthesis.
aExternal oil bath temperature; 700C, b microwave (MW), 125 oC, 1h.
Characterization of 4c was enabled by determination
of its solid state x-ray structure[6] (Scheme 1) and by
its unequivocal synthesis starting from 4-
To explain the formation of the ether products we
propose a competing SNAr, termed “nuisance effect,”
which has historically been observed for halogen-
substituted benzenediazonium salts, given the strong
electron withdrawing effects of the diazo group,
notably operating on the 2- and 4-substituted
isomers[7]. Indeed, simple alcoholysis of compound 2c
was achieved in the appropriate alcohol solvent at
700C (Scheme 2).
methoxybenzenediazonium tetrafluoroborate
2d
(Table 1) where we found slightly better yields under
reflux (Entry 1 vs. 2) compared to either ambient
temperature or to the absence of photocatalyst (Entry
5). Moreover, a palladium catalyst was essential
(Entry 4) for achieving a good yield. Microwave-
mediated chemistry, in the absence of light and
photocatalyst, gave little conversion of product.
Scheme 2. “Nuisance effect” on diazonium salts.
The C-H activation reaction was also applied to
aryldiazoniums incapable of undergoing such a F-
substitution and, hence derivatives 4e-4i‘ were
synthesized in good to excellent yields (Scheme 3).
Indeed, yields tend to be either similar or higher than
thnose reported for the corresponding reactions
Table 1. Synthesis of an anisole derivative.
2
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700626
Advanced Synthesis & Catalysis
4
5
421.54 ± 86.1
303.25 ± 60.7
involving iodonium salts, e.g. 4e (60% vs. 56%), 4f
(54% vs. 35%), 4g (71% vs. 55%) and 4i (64% vs.
63%).
6
689.56 ± 480.3
Sanford et al. proposed a possible mechanism to
[5a]
explain their Pd/Ru photocatalysed C-H arylation
.
Here we present a computational study of a Pd-
catalysed and a Sanford-derived Pd/Ru photocalysed
mechanism for the functionalization of 1a to 4g
(Scheme 4) to rationalise the increased yield in the
presence of light and a Ru photocatalyst.
Scheme 3. Other arylated benzodiazepines.
In the synthesis of 4i, relatively large amounts of the
diarylated adduct 4i‘ were also observed. Such di-
arylations were previously reported by us.3b
The current and previous library of benzodiazepines
(Scheme 1) was tested for GABA binding[8]. None of
the current benzodiazepines displayed any
appreciable biological activity although 7-chloro-
benzodiazepines, as expected, had reasonable activity,
although were ca. 7-10 fold less active than
nordazepam and diazepam controls (Entries 1 and 2
respectively, Table 2) and were not pursued any
further.
Scheme 4. The formation reaction of 4g with (i) and
without (ii) the Ru photocatalyst, investigated using
DFT.
The detailed mechanism is shown in Scheme 5 and
the reaction profile (relative to the reactants) in Figure
1. The reaction mechanism, with and without the
Ru(II)-photocatalyst, essentially follows the same
path except that the oxidative addition step in the
presence of just the Pd(II)-catalyst (path shown in
green, Scheme 5 and Figure 1), is replaced by a
single-electron-transfer (SET) process when the
Ru(II)-photocatalyst is added (shown in red, Scheme
5 and Figure 1).
Table 2. GABA activity of library.
Entry Compound
mean Ki (nM)/
SEM (nM) vs.
GABA.
1
51.62 ± 2.0
2
3
41.41 ± 4.9
373.45 ± 110.5
3
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700626
Advanced Synthesis & Catalysis
Pd and the agostic H in Int2 is 1.903 Å, which is in
good agreement with similar agostic interactions in
the literature: Pd---H = 1.91 Å[9] and Rh---H (1.95
Å)[10]. The barrier to C-H bond activation is 41.4 kJ
mol-1, and involves H migration from C to O via a
six-membered ring (TS2-3). Prior to coordination
+
with the p-nitrobenzenediazonium (Ar-N2 ) the
complex undergoes an isomerisation step (TS3-4),
which involves a change in the C1-Pd-O3 angle from
132.0 to 172.0 degrees with an energy barrier of 27.8
kJ mol-1 to form Int4.
+
In the absence of the photocatalyst, Ar-N2 interacts
with the Pd(II) complex and follows an oxidation
addition (OA) pathway, (highlighted in green,
Scheme 5 and Figure 1). The oxidative addition via
TS5-6(OA) has an energy barrier of 127.1 kJ mol-1
and involves the formation of an Ar-Pd(IV) complex.
The N2 is then eliminated leading to Int7.
When the Ru(II)-photocatalyst is present, the
+
nitrobenzene radical (Ar*) is generated from Ar-N2
+
(via oxidative quenching of Ar-N2 by the photo-
excited [Ru(bpy)3]2+ complex to form [Ru(bpy)3]3+[11]
and follows a single-electron-transfer (SET) pathway,
(in red, Scheme 5 and Figure 1). The square planar
geometry of the Pd(II) complex Int4 becomes a
Pd(III) distorted-octahedral structure when the Ar
binds to the Pd centre in Int5(SET); this is consistent
with the crystal structure of other Pd(III)-complexes
Scheme 5. The reaction mechanism for the
functionalization of benzodiazepine. From Int4 to
Int7 the transformation follows the green path in the
presence of the Pd catalyst and the red path in the
presence of the Pd/Ru catalysts. Both paths were
considered.
although we did not consider bimetallic species[12]
.
Int7 is formed directly from Int5(SET) by the
transfer of an electron to the [Ru(bpy)3]3+ complex to
recover the photocatalyst. The Gibbs free energy
barrier for single electron transfer (SET) resulting in
the formation of the Pd(IV) complex Int7 was
calculated to be 2.5 kJ mol-1 using Marcus and
Savéant theory[13]. The details of this calculation are
provided in the Computational Method section. This
The initial step of the catalysed mechanism involves
the coordination to Pd(OAc)2 by a N atom on the un-
functionalised benzodiazepine to provide Int1,
followed by the formation of an agostic complex Int2
prior to C-H activation. The atomic distance between
4
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700626
Advanced Synthesis & Catalysis
barrier is very small but similar to literature values calculations and in a palladium complex crystal
that range from 0.4 – 15.1 kJ mol-1. [14] structure. [17]
Both mechanisms (OA and SET) result in the same It is clear from Figure 1 that in the presence of the Pd-
Pd(IV) structure for Int7. At this stage reductive catalyst, the oxidative addition step is rate
elimination occurs via TS7-8 with a barrier of 43.2 kJ determining with a considerable energy barrier.
mol-1. This step involves the formation of a C-C bond However, in the presence of both the Pd(II)-catalyst
Figure 1. The reaction energy profile for the and the Ru(II)-photocatalyst this OA step, and hence
formation of 4g from 1a, with (red path) and without
(green path) a photocatalyst. Steps common to both
mechanisms are shown in blue. [Ru2+] and [Ru3+]
represent [Ru(bpy)3]2+ and [Ru(bpy)3]3+, respectively.
to facilitate the functionalization of the large energy barrier, is avoided as the reaction
benzodiazepine and the oxidation state of the Pd- proceeds via a very low-energy single-electron-
transfer process. This provides a rationale for the
increased yield in the presence of a photocatalyst.
center changes from Pd(IV) to Pd(II) (Int7  Int8).
The geometry Int8, involves an 2(C=C) interaction
with Pd. A similar interaction was observed by
Ariafard et al.[15] and Canty et al.[16] in their DFT
Conclusion
5
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700626
Advanced Synthesis & Catalysis
The C-H activation of benzodiazepines with 2- or 4-
fluorobenzene diazonium salts under Pd catalysis with All reactions were conducted under an inert
a Ru photocatalyst, in alcohol solvent, under reflux, atmosphere unless specified otherwise. All
leads to a mixture of both fluoroaryl and alkoxyaryl commercially purchased materials and solvents were
products. Reaction temperature is a key factor in used without further purification unless specified
determining the ratio of expected vs. “nuisance effect” otherwise.
(SNAr) products. At ambient temperature trace NMR spectra were recorded on a Varian VNMRS 500
amounts of the SNAr product are detected whereas (1H: 500 MHz, 13C: 126 MHz) spectrometer and
significant amounts can be obtained after prolonged prepared in deuterated solvents such as CDCl3 and
1
13
heating under reflux. This process can also be DMSO-d6. H and C chemical shifts were recorded
1
extended to other aryl diazonium salts affording in parts per million (ppm). Multiplicity of H-NMR
ortho-arylated benzodiazepines. These were tested for peaks are indicated by s – singlet, d – doublet, dd –
biological activity but were found to be significantly doublets of doublets, t – triplet, pt – pseudo triplet, q
less active than e.g. nordazepam and diazepam – quartet, m – multiplet and coupling constants are
controls. Density functional theory (DFT) has been given in Hertz (Hz).
used to provide a detailed mechanistic understanding Electrospray ionisation – high resolution mass spectra
of the functionalization of the benzodiazepines and to (ESI-HRMS) were obtained using a Bruker Daltonics
offer an explanation for the increased yield in the Apex III where Apollo ESI was used as the ESI
presence of a Ru(II)-photocatalyst. The Pd/Ru source. All analyses were conducted by Dr A. K.
catalytic cycle follows the mechanism proposed by Abdul-Sada at Sussex. The molecular ion peaks [M]+
Sandford et al. [5a]. The increased yield in the visible- were recorded as mass to charge (m/z) ratio.
light photocatalysed Pd-mediated protocol is LC-MS spectra were acquired using a Shimadzu LC-
attributed to the transformation step leading to the MS 2020, on a Gemini 5 µm C18 110 Å column and
formation of the Pd(IV) complex. In the presence of percentage purities were run over 30 minutes in
the photocatalyst the reaction proceeds via a low- water/acetonitrile with 0.1% formic acid (5 min at 5%,
energy SET pathway and avoids the high-energy 5%-95% over 20 min, 5 min at 95%) with the UV
oxidative addition step in the Pd-only catalysed detector at 254 nm. Purifications were performed by
reaction pathway.
flash chromatography on silica gel columns or C18
Current studies are aiming to extend the columns using a Combi flash RF 75 PSI, ISCO unit.
arylation/nuisance effect chemistry to a wider scope The following CCDC deposition numbers have been
of privileged structures with different nucleophiles for obtained, in parentheses; for 4c (1518056), 4d
application in medicinal chemistry library generation (1551609) and 4h (1551610).
and will be reported in due course.
4-Methoxybenzenediazonium
(2d)
tetrafluoroborate
Experimental Section
A stirred suspension of 4-fluorobenzenediazonium
tetrafluoroborate (0.10 g, 0.48 mmol) in methanol (2
General Information
6
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700626
Advanced Synthesis & Catalysis
mL) was heated at 700C by using an external oil bath and the combined aqueous layers were extracted with
for 1 hour. The reaction was allowed to cool to ethyl acetate (50 mL). Thereafter the combined
ambient temperature and concentrated under reduced organic layer was washed with brine (50 mL), dried
pressure. The residue was precipitated by the addition (MgSO4) and concentrated under reduced pressure.
of diethyl ether and collected by filtration, affording The resulting crude material was purified by reversed
2d as a white solid (0.090 g, 85%). The spectral data phase chromatography (water/acetonitrile with 0.1%
were concurrent with those reported.[18]
formic acid, 5 min at 0%, 30%-90%). Starting
material 1a was recovered (0.014 g, 0.06 mmol). Two
products were generated; 3a was obtained as a white
4-Ethoxybenzenediazonium tetrafluoroborate (2e)
The reaction was conducted by the same procedure as solid (0.022 g, 28%) and 4a was obtained as a white
for 2d but ethanol (2 mL) was used instead of solid (0.030 g, 37%). 3a: The spectral data were
1
methanol and heated at 70oC for 1 hour. 2e was concurrent with those reported.3 4a: H-NMR (500
3
obtained as a white solid (0.071 g, 63%). The spectral MHz) CDCl3: δ = 7.98 (s, NH, 1H), 7.68 (d, JHH
=
data were concurrent with those reported.[19]
7.0 Hz, ArH, 1H), 7.52 – 7.42 (m, ArH, 2H), 7.28 (d,
3JHH = 8.0 Hz, ArH, 1H), 7.19 – 7.12 (m, ArH, 1H),
7.06 – 6.98 (m, ArH, 2H), 6.90 – 6.83 (m, ArH, 1H),
2-Methoxybenzenediazonium
(2f)
tetrafluoroborate
3
6.80 (d, JHH = 7.5 Hz, ArH, 1H), 6.69 – 6.60 (m,
3
The reaction was conducted by the same procedure as ArH, 2H), 6.52 (d, JHH = 8.0 Hz, ArH, 1H), 4.22 (s,
for
2d
but
2-fluorobenzenediazonium COCH2, 2H), 3.51 (s, O-CH3, 3H). 13C-NMR (126
tetrafluoroborate (0.10, 0.48 mmol) was used instead. MHz) CDCl3: δ = 173.1(C=O), 171.1 (C=N), 156.1
2f was obtained as a white solid (0.073 g, 72%). The (ArC), 140.8 (ArC), 139.0 (ArC), 137.4 (ArC), 131.3
spectral data were concurrent with those reported.
(ArC), 131.5 (ArC), 131.4 (ArC), 131.1 (ArC), 130.3
(ArC), 129.8 (ArC), 129.6 (ArC), 129.3 (ArC), 128.9
(ArC), 127.7 (ArC), 123.3 (ArC), 120.3 (ArC), 120.2
5-(2’-Fluorobiphenyl-2-yl)-1,3-dihydro-2H-1,4-
benzodiazepin-2-one (3a); 5-(2’-methoxybiphenyl- (ArC), 110.0 (ArC), 56.7 (COCH2), 55.3 (O-CH3).
HRMS-ESI (m/z) calculated for C22H18FN2O2 [+H] +:
2-yl)-1,3-dihydro-2H-1,4-benzodiazepin-2-one (4a)
5-Phenyl-1,3-dihydro-2H-1,4-benzodiazepin-2-one
343.1441, found: 343.1446. LCMS purity (UV) =
(0.070 g, 0.3 mmol), 2-fluorobenzenediazonium 96 %, tR 10.63 min.
tetrafluoroborate (0.25 g, 1.20 mmol) and palladium
(II) acetate (0.0067 g, 0.03 mmol) were suspended in
degassed, anhydrous methanol (5 mL). Two
5-(3’-Fluorobiphenyl-2-yl)-1,3-dihydro-2H-1,4-
benzodiazepin-2-one (3b)
fluorescent light bulbs (26 W) were placed on either The reaction was conducted on a 0.20 mmol scale by
side of the reaction vessel and the reaction mixture the same procedure as for 3a/4a but 3-
was heated at 700C by using an external oil bath for 4 fluorobenzenediazonium tetrafluoroborate (0.17 g, 0.8
hours. The reaction was allowed to cool to ambient mmol) was used instead of 2-fluorobenzenediazonium
temperature, diluted with ethyl acetate (50 mL), tetrafluoroborate. Starting material, 1a was recovered
washed with water (20 mL) and aqueous sodium (0.010 g, 0.042 mmol) and 3b was obtained as a white
sulphite (10%, 35 mL x 2). The layers were separated
7
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700626
Advanced Synthesis & Catalysis
solid (0.040 g, 77%). The spectral data were g, 0.085 mmol). Two products were generated,
concurrent with those reported.
product 3c was obtained as a white solid (0.043 g,
39%) and Product 4d was obtained as a white solid
1
(0.026 g, 22%). 4d: H-NMR (500 MHz) CDCl3: δ =
5-(4’-Fluorobiphenyl-2-yl)-1,3-dihydro-2H-1,4-
3
benzodiazepin-2-one (3c); 5-(4’-Methoxybiphenyl- 8.20 (s, NH, 1H), 7.68 (d, JHH = 7.5 Hz, ArH, 1H),
3
7.57 – 7.38 (m, ArH, 2H), 7.28 (d, JHH = 7.5 Hz,
2-yl)-1,3-dihydro-2H-1,4-benzodiazepin-2-one (4c)
3
This reaction was conducted on a 0.42 mmol scale by ArH, 1H), 7.15 (pt, JHH = 7.5 Hz, ArH, 1H), 6.91 –
the
same
procedure
as
3a/4a and
4- 6.81 (m, ArH, 4H), 6.69 (d, 3JHH = 8.0 Hz, ArH, 1H),
3
fluorobenzenediazonium tetrafluoroborate (0.35 g, 6.60 (d, JHH = 8.0 Hz, ArH, 2H), 4.29 (s, COCH2,
3
1.67
mmol)
was
used
instead
of
2- 2H), 3.94 (q, JHH = 7.0 Hz, O-CH2CH3, 2H), 1.36 (t,
fluorobenzenediazonium tetrafluoroborate. Starting 3JHH = 7.0 Hz, O-CH2CH3, 3H). 13C-NMR (126 MHz)
material, 1a was recovered (0.015 g, 0.06 mmol) and CDCl3: δ = 173.2 (C=O), 170.7(C=N), 157.8 (ArC),
the reaction generated two products; 3c was obtained 141.7 (ArC), 139.5 (ArC), 137.3 (ArC), 133.2 (ArC),
as a white solid (0.053 g, 45%) and 4c was obtained 131.1 (ArC), 130.1 (ArC), 129.9 (ArC), 129.7 (ArC),
1
as a white solid (0.038 g, 32%). 3c: H-NMR (500 129.8 (2 x ArC), 129.5 (ArC), 129.1 (ArC), 128.1
MHz) DMSO-d6: δ = 10.39 (s, ArH, 1H), 7.60 – 7.55 (ArC), 126.9 (ArC), 123.1 (ArC), 113.8 (2 x ArC),
(m, 1H), 7.55 – 7.52 (m, ArH, 1H), 7.50 (d, 3JHH = 7.5, 63.5 (O-CH2CH3), 56.5 (COCH2), 14.8 (O-CH2CH3).
Hz, ArH, 1H), 7.33 – 7.30 (m, ArH, 1H), 7.21 – 7.17 HRMS-ESI (m/z) calculated for C23H20N2O2 [+Na] +:
(m, ArH, 1H), 6.92 – 6.86 (m, ArH, 4H), 6.83 – 6.77 379.1417, found: 379.1419. LCMS purity (UV) =
3
(m, ArH, 2H), 6.69 (d, JHH = 8.0 Hz, 1H), 4.03 (s, 87 %, tR 10.89 min.
COCH2, 2H).
13C-NMR (126 MHz) DMSO-D6: δ = 172.1 (C=O),
5-Phenyl-2-yl)-1,3-dihydro-2H-1,4-benzodiazepin-
169.7 (C=N), 161.5 (d, 1JFC = 244.0 Hz, ArC,), 140.4
2-one (4e)
(ArC), 139.8 (ArC), 139.2 (ArC), 136.9 (ArC), 131.5 The reaction was conducted by the same procedure as
3
(ArC), 130.4 (A rC), 130.5 (d, JFC = 7.5 Hz, 2 x for 3a/4a but benzenediazonium tetrafluoroborate
ArC), 130.2 (ArC), 130.1 (ArC), 129.3 (ArC), 128.3 (0.23 g, 1.20 mmol) was used instead of 2-
(ArC), 127.8 (ArC), 122.7 (ArC), 120.7 (ArC), 114.9 fluorobenzenediazonium tetrafluoroborate. Starting
(d, 2JFC = 22.0 Hz, 2 x ArC), 57.3 (COCH2). HRMS- material 1a was recovered (0.016 g, 0.067 mmol) and
+
ESI (m/z) calculated for C21H15FN2O [+H]
:
4e was obtained as a white solid (0.043 g, 60%). All
331.1241, found: 331.1244. LCMS purity (UV) = Spectral data were concurrent with those reported.
92%, tR 11.16 min. 4c: The spectral data were
concurrent with those reported.
5-(4’-Methoxybiphenyl-2-yl)-1,3-dihydro-2H-1,4-
benzodiazepin-2-one (4f)
The reaction was conducted on a 0.32 mmol scale by
the same procedure as for 3a/4a but 4-
5-(4’-Ethoxybiphenyl-2-yl)-1,3-dihydro-2H-1,4-
benzodiazepin-2-one (4d)
The same method as that of 3a/4a was used but methoxybenzenediazonium tetrafluoroborate (0.28 g,
ethanol (5 mL) was used as the solvent instead of 1.28 mmol) was used instead of 2-
methanol. Starting material, 1a, was recovered (0.020 fluorobenzenediazonium tetrafluoroborate. Starting
8
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700626
Advanced Synthesis & Catalysis
3
material, 1a was recovered (0.015 g, 0.063 mmol) and 1H), 6.87 (dd, J = 8.0, 5.9 Hz, 3H), 6.82 (pt, JHH
=
3
4f was obtained as a white solid (0.048 g, 54%). All 7.5 Hz, 1H), 6.73 (d, JHH = 8.0 Hz, 1H), 4.05 (s,
Spectral data were concurrent with those reported.
COCH2, 2H). 13C-NMR (126 MHz) DMSO-d6: δ =
171.9 (C=O), 169.8 (C=N), 140.2 (ArC), 139.8 (ArC),
139.6 (ArC), 139.3 (ArC), 131.5 (ArC), 131.1 (ArC),
131.0 (ArC x 2), 130.6 (ArC x 2), 130.3 (ArC), 130.2
5-(4’-Nitrobiphenyl-2-yl)-1,3-dihydro-2H-1,4-
benzodiazepin-2-one (4g)
The reaction was conducted on a 0.45 mmol scale by (ArC), 130.1 (ArC), 129.5 (ArC), 128.2 (ArC), 128.1
the same procedure as for 3a/4a but 4- (ArC), 122.7 (ArC), 120.7 (ArC), 56.7 (COCH2).
nitrobenzenediazonium tetrafluoroborate (0.43 g, 1.80 HRMS-ESI (m/z) calculated for C21H15BrN2O [+H] +:
mmol) was used instead. Starting material, 1a was 391.0441, found: 391.0449. LCMS purity (UV) =
recovered (0.020 g, 0.085 mmol) and 4g was obtained 95 %, tR 14.56 min.
1
as a white solid (0.093 g, 71%). H-NMR (500 MHz)
3
CDCl3: δ = 8.78 (s, 1H), 7.94 (dd, JHH = 8.5, 1.5 Hz,
2H), 7.78 – 7.72 (m, 1H), 7.60 – 7.52 (m, 2H), 7.34 –
7.28 (m, 1H), 7.22 – 7.13 (m, 3H), 6.90 – 6.83 (m,
5-(3’-Trifluoromethylbiphenyl-2-yl)-1,3-dihydro-
2H-1,4-benzodiazepin-2-one (4i); 5-(3,3’-
bistrifluoromethylbiphenyl-2,6-yl)-1,3-dihydro-
2H-1,4-benzodiazepin-2-one (4i‘)
3
2H), 6.72 (d, JHH = 8.0 Hz, 1H), 4.31 (s, COCH2,
2H). ). 13C-NMR (126 MHz) CDCl3: δ = 172.1 (C=O), The reaction was conducted on a 0.39 mmol scale by
170.6 (C=N), 147.4 (ArC), 146.6 (ArC), 139.7 (ArC), the same procedure as for 3a/4a but 3-
139.6 (ArC), 137.5 (ArC), 131.7 (ArC), 130.4 (ArC), trifluoromethylbenzenediazonium tetrafluoroborate
129.9 (ArC), 129.8 (ArC), 129.6 (ArC), 129.5 (ArC x (0.41 g, 1.56 mmol) was used instead. 4i was obtained
2), 128.7 (ArC), 128.6 (ArC), 123.3 (ArC), 122.7 as a brown solid (0.094 g, 64%) and the bisarylated
(ArC x 2), 120.1 (ArC), 56.5 (COCH2). HRMS-ESI product, 4i‘, was obtained as a brown solid (0.061 g,
+
(m/z) calculated for C21H15N3O3 [+H] : 358.1186, 30%). 4i: All spectral data were concurrent with those
1
found: 358.1191. Elemental Analysis: Calculated for reported. 4i‘: H-NMR (500 MHz) DMSO-d6: δ =
3
C21H15N3O3 (%): C, 70.58, H, 4.23, N, 11.76, found: 10.01 (s, 1H), 7.66 (pt, JHH = 7.5 Hz, 1H), 7.55 (d,
3
C, 70.41, H, 4.23, N, 11.60.
3JHH = 8.0 Hz, 2H), 7.50 (d, JHH = 7.5 Hz, 2H), 7.45
3
(pt, JHH = 7.5 Hz, 2H), 7.41 – 7.36 (m, 4H), 7.22 –
7.18 (m, 1H), 6.96 – 6.89 (m, 2H), 6.75 (d, J = 8.0 Hz,
1H), 3.65 (s, COCH2, 2H). 13C-NMR (126 MHz)
5-(4’-Bromobiphenyl-2-yl)-1,3-dihydro-2H-1,4-
benzodiazepin-2-one (4h)
The reaction was conducted on a 0.25 mmol scale by CDCl3: δ = 170.4 (C=O), 169.6 (C=N), 141.4 (ArC),
the same procedure as for 3a/4a but 4- 141.0 (ArC), 138.1 (ArC), 137.4 (ArC), 132.4 (ArC x
2
bromobenzenediazonium tetrafluoroborate (0.27 g, 2), 131.7 (ArC), 130.1 (q, JFC, 33 Hz, ArC x 2),
1.0 mmol) was used instead. Starting material, 1a was 129.8 (ArC x 2), 129.4 (ArC x 2), 129.3 (ArC x 2),
3
recovered (0.012 g, 0.051 mmol) and 4h was obtained 128.7 (ArC), 128.2 (ArC x 2), 125.8 (q, JFC, 3.5 Hz,
1
as a white solid (0.051 g, 65%). H-NMR (500 MHz) ArC x 2), 123.9 (q, 3JFC, 272.0 Hz ArC x 2), 123.7 (q,
3
DMSO-d6: δ = 10.44 (s, 1H), 7.56 (pt, JHH = 8.0 Hz, 3JFC, 3.5 Hz, ArC x 2), 123.4 (ArC), 120.2 (ArC),
2H), 7.54 – 7.48 (m, 1H), 7.34 (d, 3JHH = 8.0 Hz, 1H), 55.7 (COCH2). C29H18F6NO2 [+H] +: 525.1396, found:
3
3
7.28 (d, JHH = 8.0 Hz, 2H), 7.23 (pt, JHH = 7.5 Hz, 525.1402. LCMS purity (UV) = 98 %, tR 22.50 min.
9
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700626
Advanced Synthesis & Catalysis
Here Δ퐺푟 is the reaction energy for the electron
transfer step and Δ퐺0≠is the intrinsic barrier, which
can be calculated as:
Computational Details
휆
Δ퐺0≠ =
Density functional theory (DFT) calculations were
4
2)
performed
at
the
ωB97XD/6-
In Eq. (2), 휆 is the reorganisation energy and consists
of the inner reorganisation energy of the reactants, 휆푖,
and the solvent reorganisation energy, 휆표. For outer-
sphere electron transfer as in the present case, 휆푖 is
assumed to be zero (following literature
precedents[27]) thus 휆 is equal to 휆표.
311++G(2df,2p)[SDD]//PBE/6-31+G(d,p)[SDD]
level of theory, using the Gaussian09 program[20]. The
Pople basis sets were used on all atoms except Pd and
Ru for which the SDD relativistic effective core
potentials were used[21]. The PBE functional[22] was
used for the geometry optimisation and frequency
analysis as it combines good accuracy for Pd
complexes with computational speed[23]. The long-
range corrected hybrid functional ωB97XD[24], which
includes empirical dispersion corrections, was used
The reaction energy for the electron transfer step Δ퐺푟
is calculated as the energy of the reaction: Pd(III)-
complex + [Ru(bpy3)]3+  Pd(IV)-complex +
[Ru(bpy3)]2+ (i.e. Int5(SET) to Int7, Scheme 5). The
energy for this step is -83.4 kJ mol-1.
for energies to ensure accurate energetics[25]
.
Methanol solvent energy corrections were applied
using the conductor-like polarisable continuum model
(CPCM)[26]. Accordingly, the Gibbs free energies
presented in Figure 1 were obtained by adding the
thermal free energy corrections obtained at the
PBE/6-31+G(d,p)[SDD] level of theory to the
solvent-corrected electronic energies obtained at the
ωB97XD/6-311++G(2df,2p)[SDD] level of theory.
All stationary states were verified as minima or
transition states by the absence or presence,
respectively, of a single imaginary vibrational
frequency. Eigenvector following was used to ensure
transition states connected the desired minima.
The reorganisation energy 휆 = 휆표 is calculated using
the following equation[27-28]
:
푁 푒2
1
1
1
1
1
− )
푅
퐴
(3)
휆표 =
(
−
) (
+
4휋휀0 휀표푝 휀푠 2푟
2푟2
1
where 푁 is the Avogadro constant (6.0221023 mol-1),
퐴
e is the electronic charge (1.60210-19 C), 휀0 is the
vacuum permittivity (8.85410-12 J-1C2m-1) and,
휀표푝 and 휀푠 are the optical and static dielectric constant
for solvent, respectively. For methanol, 휀표푝 is 1.76
and 휀푠 is 32.613. 푟 , 푟2 and R are the hard sphere
1
radii of the donor, the acceptor, and their sum. In this
work, the hard sphere radii approximation of
[Ru(bpy)3]3+ and the Pd(III)-complex (Int5(SET))
were calculated using the VOLUME keyword in
Gaussian09. The calculated [Ru(bpy)3]3+ radius is
6.18 Å and the calculated Pd(III)-complex radius is
6.47 Å. Using these values in Eq. (3) gives 휆표= 59.1
kJ mol-1, and hence Δ퐺0≠= 14.8 kJ mol-1. Substituting
The Gibbs free energy barrier for single electron
transfer (SET), Δ퐺퐸≠푇, was calculated using the
following equation from Marcus and Savéant
theory[13b-d]
:
2
Δ퐺푟
4Δ퐺0≠
Δ퐺퐸≠푇 = Δ퐺0≠ [1 +
]
1)
10
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700626
Advanced Synthesis & Catalysis
these values for Δ퐺0≠and Δ퐺푟 in Eq. (1), provides a
SET barrier, Δ퐺퐸≠푇 = 2.5 kJ mol-1.
2016, 138, 15482-15487; m) J. He, L. G.
Hamann, H. M. Davies, R. E. Beckwith,
Nat. Commun. 2015, 6, 5943; n) Q.
Michaudel, G. Journot, A. Regueiro-Ren,
A. Goswami, Z. Guo, T. P. Tully, L. Zou,
R. O. Ramabhadran, K. N. Houk, P. S.
Baran, Angew. Chem. Int. Ed. Engl. 2014,
53, 12091-12096; o) D. A. Nagib, D. W.
MacMillan, Nature 2011, 480, 224-228;
p) Y. Zhu, M. Bauer, L. Ackermann,
Chem. Eur. J. 2015, 21, 9980-9983. q) F.
Yang, J Koeller, L. Ackermann, Angew.
Chem. Int. Ed. 2016, 55, 4759-4762.
a) J. Spencer, B. Z. Chowdhry, A. I.
Mallet, R. P. Rathnam, T. Adatia, A.
Bashall, F. Rominger, Tetrahedron 2008,
64, 6082-6089; b) R. Khan, R. Felix, P. D.
Kemmitt, S. J. Coles, I. J. Day, G. J.
Tizzard, J. Spencer, Adv. Synth. Catal.
2016, 358, 98-109.
a) E. A. Merritt, B. Olofsson, Angew.
Chem. Int. Ed. Engl. 2009, 48, 9052-
9070; b) M. S. Yusubov, A. V. Maskaev,
V. V. Zhdankin, Arkivoc 2011, 1, 370-
409; c) S. G. Modha, M. F. Greaney, J.
Am. Chem. Soc. 2015, 137, 1416-1419.
a) D. Kalyani, K. B. McMurtrey, S. R.
Neufeldt, M. S. Sanford, J. Am. Chem.
Soc. 2011, 133, 18566–18569; b) J. Jiang,
W. M. Zhang, J. J. Dai, J. Xu, H. J. Xu, J.
Org. Chem. 2017, 82, 3622-3630; c) M.
Majek, A. Jacobi von Wangelin, Acc.
Chem. Res. 2016, 49, 2316-2327; d) L.
Marzo, I. Ghosh, F. Esteban, B. König,
ACS Catal. 2016, 6780-6784; e) D. P.
Hari, P. Schroll, B. Konig, J. Am. Chem.
Soc. 2012, 134, 2958-2961.
Acknowledgements
EPSRC, AstraZeneca and Tocris are thanked for
financial support (grant # EP/M507568/1). We would
also like to thank the Royal Thai Government for the
fully funded PhD scholarship awarded to SB and we
acknowledge the use of the EPSRC UK National [3]
Service for Computational Chemistry Software
(NSCCS) at Imperial College London in performing
part of this work. We also thank the EPSRC UK
National Mass Spectrometry Facility at Swansea
University.
[4]
References.
[1]
[2]
T. Cernak, K. D. Dykstra, S. Tyagarajan,
P. Vachal, S. W. Krska, Chem. Soc. Rev.
2016, 45, 546-576.
a) M. A. J. Duncton, MedChemComm
2011, 2, 1135; b) T. Gensch, M. N.
[5]
Hopkinson, F. Glorius, J. Wencel-Delord,
Chem. Soc. Rev. 2016, 45, 2900-2936; c)
T. A. Bedell, G. A. Hone, D. Valette, J. Q.
Yu, H. M. Davies, E. J. Sorensen, Angew.
Chem. Int. Ed. Engl. 2016, 55, 8270-
8274; d) A. Sharma, J. F. Hartwig, Nature
2015, 517, 600-604; e) J. J. Topczewski,
P. J. Cabrera, N. I. Saper, M. S. Sanford,
Nature 2016, 531, 220-224; f) L.
McMurray, F. O'Hara, M. J. Gaunt, Chem.
Soc. Rev. 2011, 40, 1885-1898; g) A.
McNally, B. Haffemayer, B. S. Collins,
M. J. Gaunt, Nature 2014, 510, 129-133;
h) F.-L. Zhang, K. Hong, Tuan-Jie Li, H.
Park, J.-Q. Yu, Science 2016, 351, 252-
256; i) F. O'Hara, D. G. Blackmond, P. S.
Baran, J. Am. Chem. Soc. 2013, 135,
12122-12134; j) O. Abdulla, A. D.
Clayton, R. A. Faulkner, D. M. Gill, C. R.
Rice, S. M. Walton, J. B. Sweeney,
Chem., Eur. J. 2017, 23, 1494-1497; k) A.
F. Noisier, M. A. Brimble, Chem. Rev.
2014, 114, 8775-8806; l) J. P. Barham, M.
P. John, J. A. Murphy, J. Am. Chem. Soc.
[6]
[7]
S. J. Coles, P. A. Gale, Chem. Sci. 2012, 3,
683-689.
a) J. F. Bunnett, R. E. Zahler, Chem. Rev.
1951, 273- 412; b) I. K. Barbem, H.
Suschitz, J. Chem. Soc. 1960, 2735-2739.
J. R. Atack, K. A. Wafford, S. J. Tye, S.
M. Cook, B. Sohal, A. Pike, C. Sur, D.
Melillo, L. Bristow, F. Bromidge, I.
Ragan, J. Kerby, L. Street, R. Carling, J.
L. Castro, P. Whiting, G. R. Dawson, R.
M. McKernan, J. Pharmacol. Exp. Ther.
2006, 316, 410-422.
[8]
[9]
D. L. Davies, S. M. A. Donald, S. A.
Macgregor, J. Am. Chem. Soc. 2005, 127,
13754-11375.
11
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700626
Advanced Synthesis & Catalysis
[10] A. Vigalok, O. Uzan, L. J. W. Shimon, Y. [25] a) Y. Minenkov, G. Occhipinti, V. R.
Ben-David, J. M. L. Martin, D. Milstein,
J. Am. Chem. Soc. 1998, 120, 12539-
12544.
Jensen, J. Phys. Chem. A 2009, 113,
11833–11844; b) N. Sieffert, M. Buhl,
Inorg. Chem. 2009, 48, 4622-4624; c) M.
L. Laury, A. K. Wilson, J. Chem. Theory
Comput. 2013, 9, 3939-3946; d) Y. Zhao,
D. G. Truhlar, J. Chem. Theory Comput.
2011, 7, 669-676.
[11] F. Teplý, Collection Czech Chem. Comm.
2011, 76, 859-917.
[12] a) J. R. Khusnutdinova, N. P. Rath, L. M.
Mirica, J. Am. Chem. Soc. 2010, 132,
7303–7305. b) David C. Powers and
Tobias Ritter, Nat. Chem. 2009, 1, 302–
309
[13] a) R. A. Marcus, Ann. Rev. Phys. Chem.
1964, 15, 155-196; b) R. A. Marcus, J.
Chem. Phys. 1956, 24, 966-978; c) R. A.
Marcus, J. Chem. Phys. 1956, 24, 979-
989; d) J.-M. Saveant, J. Am. Chem. Soc.
1987, 109, 6788-6795.
[26] a) V. Barone, M. Cossi, J. Phys. Chem. A
1998, 102, 1995-2001; b) M. Cossi, N.
Rega, G. Scalmani, V. Barone, J. Comp.
Chem. 2003, 24, 669-681.
[27] a) G. O. Jones, P. Liu, K. N. Houk, S. L.
Buchwald, J. Am. Chem. Soc. 2010, 132,
6205–6213; b) C. Y. Lin, M. L. Coote, A.
Gennaro, K. Matyjaszewski, J. Am. Chem.
Soc., 130, 12762–12774.
[14] Q. Zhang, Z.-Q. Zhang, Y. Fu, H.-Z. Yu,
ACS Catal. 2016, 6, 798-808.
[28] R. A. Marcus, N. Sutin, Biochimica et
Biophysica Acta 1985, 811, 265-322.
[15] A. Ariafard, C. J. T. Hyland, A. J. Canty,
M. Sharma, B. F. Yates, Inorg. Chem.
2011, 50, 6449–6457.
[16] A. J. Canty, A. Ariafard, M. S. Sanford, B.
F. Yates, Organometallics 2013, 32,
544−555.
[17] H. Ossor, M. Pfeffer, J. T. B. H.
Jastrzebski, C. H. Stam, Inorg. Chem.
1987, 26, 1169-1117 1161.
[18] a) P. Hanson, J. R. Jones, A. B. Taylor, P.
H. Walton, A. W. Timms, J. Chem. Soc.
Perkin Trans. 2 2002, 1135-1150; b) B.
Schmidt, R. Berger, F. Holter, Org.
Biomol. Chem. 2010, 8, 1406-1414.
[19] S. H. Korzeniowski, A. Leopold, J. R.
Beadle, M. F. Ahern, W. A. Sheppard, R.
K. Khanna, G. W. Gokel, J. Org. Chem.
1981, 46, 2153-2159.
[20] M. J. Frisch, G. W. Trucks, H. B.
Schlegel, G. E. Scuseria, M. A. Robb, et.
al., Gaussian Inc., Wallingford, CT, 2009.
[21] D. Andrae, U. HuBermann, M. Dolg, H.
Stoll, H. PreuB, Theor. Chim. Acta 1990,
77, 123-141.
[22] J. P. Perdew, K. Burke, M. Ernzerhof,
Phys. Rev. Lett. 1997, 77, 1396.
[23] a) S. Boonseng, G. W. Roffe, J. Spencer,
H. Cox, Dalton Trans. 2015, 44, 7570-
7577; b) P. Surawatanawong, M. B. Hall,
Organometallics 2008, 27, 6222-6232.
[24] J. D. Chai, M. Head-Gordon, Phys. Chem.
Chem. Phys. 2008, 10, 6615-6620.
12
This article is protected by copyright. All rights reserved.

10.1002/adsc.201700626
Advanced Synthesis & Catalysis
13
This article is protected by copyright. All rights reserved.