Optimization of Synthetically Versatile Pyridylidene Amide Ligands for Efficient Iridium-Catalyzed Water Oxidation

A Journal of

Accepted Article

Title: Optimization of synthetically versatile pyridylideneamide ligands

for efficient iridium-catalyzed water oxidation

Authors: Miquel Navarro, Christene Smith, Mo Li, Stefan Bernhard, and

Martin Albrecht

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: Chem. Eur. J. 10.1002/chem.201705619

Link to VoR: http://dx.doi.org/10.1002/chem.201705619

Supported by

10.1002/chem.201705619

Chemistry - A European Journal

Optimization of synthetically versatile pyridylideneamide ligands for

efficient iridium-catalyzed water oxidation

Miquel Navarro,^aChristene A. Smith,^a,bMo Li,^cStefan Bernhard,*^,cMartin Albrecht*^a

^aDepartment of Chemistry & Biochemistry, University of Bern, Freiestrasse 3, CH–3012

Bern, Switzerland.

^bpermanent address: Department of Chemistry, Queen’s University, 90 Bader Lane,

Kingston, Ontario, Canada.

^cDepartment of Chemistry, Carnegie Mellon University, 4400 Fifth Avenue, Pittsburgh, PA

15213, USA

E-mail: martin.albrecht@dcb.unibe.ch; bern@cmu.edu

1

10.1002/chem.201705619

Chemistry - A European Journal

Abstract

The synthetic versatility of PYA ligands has been exploited to prepare and evaluate a diverging

series of iridium complexes containing C,N-bidentate chelating aryl-PYA ligands for water

oxidation catalysis. The phenyl-PYA lead structure 1 was modified (i) electronically through

introduction of one, two, or three electron-donating methoxy substituents on the aryl ring, (ii)

by incorporating long aliphatic chains to the pyridyl fragment of the PYA unit, and (iii) by

altering the PYA positions from para-PYA to its ortho- and meta-isomers. Electrochemistry

indicated no substantial electronic effect of the aliphatic chains, and only minor changes of the

electron density at iridium when modifying the aryl ligand site, yet substantial alteration if the

PYA ligand is the ortho- (E_1/2= +0.72 V), para- (E_1/2= +0.64 V) or meta-isomer (E_1/2= +0.56

V vs SCE). In water oxidation catalysis, the long alkyl chains did not induce any rate

enhancement compared to the phenyl-PYA lead compound, while MeO groups incorporated in

the aryl group enhanced catalytic activity from a TOF_max=1600 h^–1in the original Ph-PYA

system gradually as more MeO groups were introduced up to a TOF_max= 3300 h^–1for a

tris(MeO)-substituted aryl-PYA system. The variation of the PYA substitution had only a minor

impact on catalytic activity and revealed only a weak trend in the sequence ortho > meta >

para. The high activity of the tris(MeO) system and the ortho-PYA isomer were attributed to

efficient hydrogen bonding that assists O–H bond activation and proton transfer. Remarkably

merging of the two optimized motifs, i.e. an aryl unit with three OMe substituents and the PYA

as ortho isomer, into a single new aryl-PYA ligand system failed to improve catalytic activity.

Computational analysis suggests too much congestion at the active site, which hinders catalytic

turnover. These results illustrate the complexity of ligand design and the subtle effects at play

in water oxidation catalysis.

Keywords: ligand design --- N-donor ligands --- mesoionic ligands --- iridium --- water

oxidation catalysis

2

10.1002/chem.201705619

Chemistry - A European Journal

Introduction

Artificial photosynthesis entailing catalytic water splitting is one of the most promising

processes to harness and store solar and other renewable energy.^1,2In pursuit of developing

efficient water splitting systems, the water oxidation half-cycle has been demonstrated to be a

severe limitation. Hurdles include, inter alia, the highly uphill thermodynamics of water

oxidation, and as well as the complexity and kinetic barriers associated with the transfer of four

protons and four electrons that is required for the formation of O₂.³In order to mediate the

multiple transfer of electrons and to overcome the energetic barrier of O–H bond cleavage and

O–O bond formation, a high redox-flexibility of the catalytic metal center is required.⁴Initial

approaches have been inspired by the natural oxygen evolving complex,⁵focusing on di- and

tetrametallic systems for chemically driven molecular water oxidation.^6,7More recently, one of

us demonstrated that a single iridium center is active in water oxidation when bound to two

phenylpyridyl ligands (A, Figure 1).⁸While this work triggered investigations towards single

site metal catalysts with other transition metals,⁹including first row transition metal systems,¹⁰

the activity of iridium complexes remained a benchmark.^11,12The use of stronger donor ligands

–

like Cp* (C₅Me₅) further enhanced catalytic activity (B),¹³and even though strong support is

available for a gradual oxidation of the Cp* unit under catalytic conditions,¹⁴the [Ir(Cp*)] has

emerged as a privileged synthon for designing water oxidation catalysts.¹¹In particular the

introduction of chelating spectator ligands based on strong σ-donors like N-heterocyclic

carbenes (NHCs) led to robust molecular catalysts with high activity (C).^15,16Recently, we have

introduced pyridylideneamides (PYAs) as strong donor ligands¹⁷to the [Ir(Cp*)] iridium

complex 1 (Figure 1).¹⁸This complex shows excellent performance in water oxidation catalysis,

providing maximum turnover frequencies (TOF_max) of 1,600 h^–1and up to 86,000 turnovers.

This high activity might be attributed to the considerable donor flexibility of the PYA ligand,

and the ensuing stabilization of different metal oxidation states during the catalytic process (cf

limiting resonance structures A and B for 1 in Figure 1).

Here we have capitalized on the synthetic versatility of the phenyl-PYA ligand to optimize the

lead iridium catalyst further. Based on recent studies, which showed a substantial rate

enhancement by simple alterations on key positions of the ligands,^19,20we have exploited such

modifications for enhancing the catalytic activity of complex 1 by modulating specifically i)

the donor properties of the Ph ligand, ii) the N-substituent on the PYA ligand, and iii) the donor

properties of the PYA ligand.

3

10.1002/chem.201705619

Chemistry - A European Journal

N

OH₂

Ir

N

Cl

N

2

A

B

C

Ir

Cl

Ir

Cl

N

O

1

Figure 1. Iridium catalysts for water oxidation catalysis and limiting resonance structures of iridium PYA

complex 1.

Results and discussion

Synthesis of complexes. i) Variation of donor properties of the Ph unit: We have

recently described synthetic protocols for the introduction of methoxy groups on the phenyl

ring of the parent complex 1.¹⁸The preparation of these ligands was conveniently achieved due

to the availability of the corresponding methoxy-substituted benzoic acids, which were readily

transformed to the PYA ligands by reaction with 4-aminopyridine, pyridine alkylation and

amide deprotonation. Metalation with [Ir(Cp*)Cl₂]₂afforded the cyclometalated complexes 2–

5 (Figure 2), which exhibited enhanced catalytic activity in transfer hydrogenation of ketones

and imines, as well as in alkyne hydrosilylation.¹⁹The better catalytic performance relative to

complex 1 was attributed to the higher electron density at the active site and the ensuing

stabilization of higher oxidation states during the catalytic cycle. In an effort to quantify

electronic impact of the different aryl-PYA ligands, we have now performed electrochemical

analysis. Cyclic voltammetry (CV) in CH₂Cl₂solution revealed a fully reversible oxidation

wave that was assigned to an Ir^III/IVoxidation process (Figure S1). For complex 1 without any

substituent at the phenyl unit, this oxidation is centered at E_1/2= +0.64 V vs SCE (SCE =

saturated calomel electrode), while methoxy-substituents shift the oxidation to lower potential,

as expected for electron-donating groups. The half-wave potentials were shifted non-linearly to

E_1/2= +0.61 V and +0.62 V for 2 and 3, respectively (both with one MeO group), and to lower

potential, E_1/2= +0.58 V for complex 4 with two MeO substituents. However, complex 5 with

4

10.1002/chem.201705619

Chemistry - A European Journal

three MeO units was oxidized at higher potential, E_1/2= +0.61 V. While this data helps to

rationalize the lower performance of complex 1 in transfer hydrogenation, it is remarkable that

the substituent effect is not additive. The lowest oxidation potential was measured for complex

4 with two MeO groups, which was also the most active transfer hydrogenation catalyst. In

contrast, a third MeO group in complex 5 increases the potential again slightly, yet the catalytic

activity is radically changed and the complex is essentially inactive. The small differences

between the oxidation potentials of all complexes 1–5 suggests another limiting effect to reach

the +IV oxidation state of the iridium center, especially once the first methoxy substituent is

introduced, such as enhanced inductive effects of the oxo substituents, or steric factors imparted

by the MeO substituents.

MeO

Ir

Cl

Ir

Cl

N

MeO

N

MeO

O

MeO

O

OMe

2

3

4

5

Figure 2. Iridium complexes 2–5 bearing methoxy-substituted PYA ligands.

ii) Modulating the N-substituent on the PYA ligand: Variation of the N-substituent on

the PYA unit was motivated by work using triazolylidene iridium complexes for water

oxidation.²⁰In that work, a ten-fold rate enhancement was observed when substituting a methyl

substituent at the triazolylidene heterocycle by an octyl group, which was tentatively attributed

to a specific aggregation processes induced by the long aliphatic chain. Based on these data,

the pyridyl-bound methyl group of the PYA ligand in complex 1 was substituted with alkyl

groups of different length. To this end, N-(pyridin-4-yl)benzamide was alkylated with 2-

iodopropane, 1-iodobutane and 1-iodooctane to yield the corresponding pyridinium salts 6a–c

in excellent yields (91–95%, Scheme 1). Selective alkylation was unambiguously confirmed by

HR-MS analysis, which displayed the expected [M–I]⁺mass ion for the isopropyl, butyl and

octyl salts at 241.1329, 255.1489 and 311.2111 amu, respectively (theoretical values 241.1341,

1

255.1492 and 311.2118 amu, respectively) and by H NMR spectroscopy, which showed the

appearance of a new set of resonances corresponding to the new NCH at d_H= 4.88 (6a), and

for the NCH₂unit at d_H= 4.47 (6b) and 4.44 (6c), together with the expected resonances in the

aliphatic region of the spectrum. The reaction of 6a–c with aqueous NaOH afforded the free

1

ligands 7a–c in good yields (73–84%). The H NMR spectra of the free pyridylidene amides

5

10.1002/chem.201705619

Chemistry - A European Journal

indicated the disappearance of the amide proton around 8 ppm, as well as a significant upfield

shift of almost a full ppm for the heterocyclic proton resonances, which is in agreement with

previous studies and the relevance of the diene-type resonance structure.²¹In contrast, the aryl

proton signals were barely affected, indicating that only the pyridylidene system is involved in

stabilizing the deprotonated amide, but not the aryl unit.

I

R

O

N

O

N

O

N

[IrCp*Cl₂]₂

NaOAc

Ir

RI

NaOH

Cl

N

H

N

H

N

R

O

6

7

8 R = ⁱPr (a), nBu (b); Oct (c)

Scheme 1. Synthesis of IrCp*PYA-type complexes 8a–c with different alkyl chains.

Metalation was performed by reacting the free ligand with [Ir(Cp*)Cl₂]₂in the presence of

NaOAc as base,²²which induced cyclometalation and afforded the iridium complexes 8a–c in

good yields (54–75%; Scheme 1). These complexes are air-stable and were purified by standard

silica column chromatography without any precaution. Cyclometalation was demonstrated for

1

all the complexes by the loss of one proton signal in the aromatic region of the H NMR

spectrum. The structures of complex 8b and 8c were unequivocally confirmed by single crystal

X-ray diffraction analysis (Figure 3). The molecular structures are unsurprising and bond

lengths and angles around the iridium center are identical within standard deviations for both

complexes and similar to related complexes (Table 1).¹⁹The pyridyl C_α–C_βbonds are

consistently shorter (average 1.35 Å) compared to the C_β–C_γbonds (1.41 Å), confirming

substantial contribution of the neutral diene-type resonance structure in the solid state, in

agreement with previous studies (see Table S1).^19,20,23

6

10.1002/chem.201705619

Chemistry - A European Journal

Figure 3. ORTEP diagram of 8b and 8c. (50% probability; hydrogen atoms and co-crystallised solvent molecules

omitted for clarity, only one of the two independent molecules of 8b,c in the unit cell shown). Selected bond

lengths (Å) and angles (°) for 8b: Ir–C_Ph2.008(11), Ir– N_PYA2.127(9), Ir–Cl 2.409(3), Ir–Cp(centroid) 1.823(5),

C_Ph–Ir–N_PYA77.4(4); dihedral angle C(O)–N_PYA–C_g–C_b24.5(11); for 8c: Ir–C_Ph2.040(5), Ir–N_PYA2.121(4), Ir–Cl

2.4161(14), Ir–Cp(centroid) 1.820(2), C_Ph–Ir–N_PYA77.5(2); dihedral angle C(O)–N_PYA–C_g–C_b29.1(8).

Electrochemical analysis of complexes 8a–c by CV in CH₂Cl₂revealed a fully reversible

oxidation at E_1/2= +0.64 V vs SCE for all complexes 8a–c, which is identical to the potential

measured for complex 1. This identical oxidation potential indicates that modulation of the N-

bound alkyl group has no significant influence on the donor properties of the PYA ligand, nor

on the electronic configuration at the iridium center.

iii) Modulation of the PYA donor properties: Modification of the donor properties of the PYA

unit was anticipated to be effective by changing the position of the amide unit. While complex

1 may be considered as a para-PYA unit with the amide in para position of the pyridyl nitrogen,

derivatives with the amide in meta or ortho substitution will have considerably different

electronic properties. An ortho-PYA system will feature a neutral limiting resonance

structure,²⁴though the proximity of the N-methyl group and the amide C=O unit is expected to

induce a dihedral strain (Scheme 2A). In contrast, the meta-PYA unit lacks a neutral resonance

form and therefore represents a unique mesoionic N-donor system (Scheme 2B), as detailed in

a previous communication.²⁵Based on these considerations, variation of the substitution pattern

was expected to have a pronounced effect on the donor properties.

7

10.1002/chem.201705619

Chemistry - A European Journal

M

N

A

N

O

M

N

B

O

Scheme 2. Limiting resonance structures of a generic ortho PYA complex (A, note also the steric constraint of the

N–CH₃group with either the C=O or the ligands at the metal); and some of the relevant resonance structures of a

generic meta PYA complex (B).

The synthesis of the iridium complexes analogous to 1 but with meta- and ortho-PYA units was

straightforward and started from benzoyl chloride, which was coupled with 2- or 3-

aminopyridine to form the amides 9a and 9b, respectively (Scheme 3).²⁶Subsequent

methylation using MeI formed the meta pyridinium 10b in excellent yield (95%), while 10a

was obtained only in a moderate 39% yield. This fact might be due to the lower nucleophilicity

of the pyridine nitrogen or because of entropic effects associated with steric rigidity. Both salts

showed the characteristic [M–I]⁺ion in the HR-MS at 539.1684 and 539.1665 (theoretical value

539.1669 amu). Alkylation was also supported by ¹H NMR spectroscopy, in particular by the

appearance of a new singlet for the N–CH₃group (4.29 and 4.41 ppm for 10a and 10b,

respectively) as well as a downfield shift of the pyridinium proton signals when compared to

the resonances of the pyridine precursors. Due to the lack of symmetry in the heterocycle, both

pyridinium systems feature four distinct CH groups both in the ¹H and ¹³C NMR spectra. Amide

deprotonation was accomplished with NaOH and afforded the free ligands 11a,b as yellowish

1

oils in good yields (~70%). The H NMR spectra of the free bases revealed the same

characteristic signals as those described for the para-PYA 7, viz. disappearance of the amide

proton as well as a 0.3–1.1 ppm upfield shift of the resonance frequencies of the heterocyclic

protons.²¹

The reaction of the free ligands with [Ir(Cp*)Cl₂]₂in the presence of NaOAc induced

cyclometalation and yielded, after purification by column chromatography, the iridium

complexes 12a and 12b in good yields (~70%) as orange air-stable solids (Scheme 2). ¹H NMR

spectroscopy confirmed cyclometalation for both complexes by the loss of one proton signal in

the aromatic region. Four distinct signals were present each for the pyridylidene unit as well as

for the cyclometalated aryl ring.

8

10.1002/chem.201705619

Chemistry - A European Journal

Scheme 3. Synthesis of iridium complexes 12a and 12b. i) PhCOCl, NEt₃; ii) MeI; iii) NaOH; iv) [Ir(Cp*)Cl₂]₂,

NaOAc.

The molecular structure of complex 12b with a mesoionic PYA ligand was further

investigated by single crystal X-ray diffraction analysis. The complex reveals the classical

three-legged piano-stool geometry with a pseudo-tetrahedral iridium center (Figure 4). Bond

lengths and angles around the iridium center are similar to those of the related iridium

complexes 1–5,^18,19and 8 (vide supra). However, the C–C and N–C bonds of the pyridylidene

unit do not show any bond length alteration pattern, supporting the low relevance of a diene-

type structure with localized double bonds as noted for the para-PYA free ligands and

complexes 1–5. All C–C bond length distances are equivalent within standard deviation and

indicate high double bond conjugation as expected for an aromatic pyridyl ring.

Figure 4. ORTEP diagram of iridium complex 12b (thermal ellipsoids at the 50% probability level; hydrogen

atoms, co-crystallized solvent molecules, and second independent molecule of the unit cell omitted for clarity).

Selected bond lengths (Å): Ir–C1 2.035(3), Ir–N1 2.103(2), Ir–Cl 2.4219(8), Ir–Cp(centroid) 1.819(2), C8–C9

1.399(4), C8–C12 1.401(4), C9–C10 1.385(4), C10–C11 1.381(5). Selected bond angles (deg): C1–Ir–N1

77.83(11), C1–Ir–Cl 86.42(8), N1–Ir–Cl 87.89(7); dihedral angle C(O)–N1–C8–C9 39.0(3).

9

10.1002/chem.201705619

Chemistry - A European Journal

The donor properties were evaluated by electrochemical analysis. CV in CH₂Cl₂

revealed a fully reversible oxidation process at E_1/2= +0.72 (vs SCE) for the ortho-PYA

complex 12a and at +0.56 V (vs SCE) for complex 12b with a meta-PYA ligand (Figure 5).

These potentials diverge considerably from the +0.64 V measured for the parent para-PYA

complex 1 and indicate that the mesoionic PYA ligand 11b is a substantially stronger donor

than para-PYA. Such stronger donor ability is expected when considering that p-acidic imine-

type bonding of the PYA ligand is disfavored in the mesoionic meta-PYA ligand, but much less

in the para-PYA isomer. The significantly higher oxidation potential of complex 12a is

remarkable and suggests that the ortho-PYA ligand is much less suitable for stabilizing higher

metal oxidation states. The 80 mV potential difference between the para- and ortho-PYA is

substantial, and presumably does not originate from simple resonance structure effects, but may

relate to out of plane distortion of the PYA ligand instead.

Figure 5. Superimposed CV plots of complexes 1 (black), 12a (blue) and 12b (red) in CH₂Cl₂solution (0.1 M

Bu₄NPF₆as supporting electrolyte, 100 mV s^–1scan rate, potential vs SCE.

Catalytic water oxidation. The structurally modified iridium complexes 2–5, 8a,b,c, and

12a,b as well as the non-substituted parent complex 1 were investigated for catalytic water

oxidation. The processes were chemically driven by (NH₄)₂[Ce(NO₃)₆] (CAN), a one-electron

oxidant unable to oxidize water without a catalyst (Figure 6). All Ir(III) complexes were soluble

in water (solubility at least 1 mg mL^–1), and in the presence of CAN dissolved with 1M HNO₃,

immediate gas formation was observed and quantified dynamically with the homebuilt pressure

10

10.1002/chem.201705619

Chemistry - A European Journal

transducer setup.⁸Complexes 8b and 8c have limited solubility in water due to the attached

long alkyl chains (nBu and Oct respectively), and thus their concentrations used in the water

oxidation experiments did not exceed 20 µM, while all the other complexes were tested at

concentrations up to 50 µM. Stability tests in 0.1 M HNO₃(pH 2), and in the presence of 4

molequiv. CAN did not indicate any complex degradation, nor loss of the MeO substituents

(Fig. S2).²⁷Quantitative analysis of the oxygen evolution revealed higher or equivalent activity

of complexes 2–5, 8, and 12 in comparison to the parent complex 1 (Table 2, entries 1–5).

Methoxy substituents accelerated the catalytic rates: mono-substituted complexes 2 and 3

displayed TOFs of 2400 h^–1and 1800 h^–1respectively (entries 6–11), while di-substituted

complex 4 (entries 12–14) and tri-substituted complex 5 (entries 15–17) proved to be the most

active catalysts exhibiting TOFs that exceeded 3000 h^–1. As for the transfer hydrogenation

catalysis, the incorporation of electron-donating groups enhances the catalytic activity in the

water oxidation reaction, an effect that culminated with complexes 4 and 5, two of the most

active iridium catalyst reported to date.²⁸These rate differences are considerable, and

interestingly they are not correlated with the electrochemical measurements. Obviously higher

oxidation states than the +IV species will need to be reached in a catalytic cycle and the

electrochemically measured Ir^III/IVredox potentials are therefore not correlated to turnover-

limiting processes. Though it is worth noting that there is no extrapolation from these potentials

to the critical step of the catalytic cycle.²⁹

Figure 6. Oxygen evolution traces of the initial 11 h of water oxidation catalyzed by complex 1 at different

concentrations (4.5-49.6 µM; for accurate concentrations of the complex, see Table 1). Solid lines represent the

average of multiple measurements. Faded areas illustrate error bands at 95 % confidence interval.

11

10.1002/chem.201705619

Chemistry - A European Journal

Complexes 8a–c with different alkyl groups attached the pyridyl nitrogen all display about the

same TOF as observed with complex 1 (~1500 h^-1at ~50 µM and ~1100 h^-1at ~20 µM; entries

18–25). These results suggest that self-aggregation—as hypothesized in previous studies²⁰—is

not a key issue in this catalytic process, and that the presence of long alkyl chains in the

triazolylidene system may therefore have different effects.³⁰

Table 1. Catalytic water oxidation with complex 1–5, 8a–c and 12a,b.^a

Entry

1

Complex

conc. [µM] CAN/complex

TON

TOF_max(h^–1)

1,600 ± 500

1,200 ± 200

840 ± 20

1

49.6

19.9

9.9

7380

20200

40380

80580

406510

7300

1,880 ± 30

2

4,670 ± 90

3

1

9,700 ± 200

19,000 ± 1000

86,000 ± 1000

1,920 ± 30

4

1

4.5

560 ± 60

5

1

0.9

160 ± 10

6

2

50.1

19.9

4.5

2,400 ± 500

1,600 ± 200

600 ± 100

1,800 ± 400

1,400 ± 200

700 ± 100

3,300 ± 700

1,500 ± 200

700 ± 100

3,200 ± 700

1,600 ± 100

650 ± 90

7

2

18450

81180

7450

4,730 ± 90

8

2

19,000 ± 400

1,920 ± 30

9

3

49.1

19.6

4.5

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

3

18730

81180

7370

4,900 ± 100

20,000 ± 400

1,870 ± 20

3

4

49.7

19.8

4.6

4

18520

79870

7300

4,770 ± 50

4

19,000 ± 300

1,900 ± 30

5

50.1

20.3

4.6

5

18090

79910

7240

4,680 ± 50

5

19,000 ± 300

1,840 ± 20

8a

8b

8c

50.4

0.9

1,500 ± 300

130 ± 10

406510

20250

40500

80990

19820

39640

79,000 ± 1000

4,910 ± 80

19.8

9.9

1,100 ± 100

900 ± 200

500 ± 100

1,100 ± 100

800 ± 100

10,000 ± 300

17,800 ± 400

5000 ± 90

4.9

20.2

10.1

9,600 ± 200

12

10.1002/chem.201705619

Chemistry - A European Journal

25

26

27

28

29

30

31

32

33

8c

5.1

49.9

19.6

9.8

79270

7360

18,200 ± 400

1,850 ± 30

540 ± 80

1,800 ± 300

1,400 ± 300

1,000 ± 200

700 ± 200

12a

12b

18790

37380

74750

7360

4,700 ± 100

9,300 ± 100

17,500 ± 100

1,860 ± 20

4.9

49.9

19.6

9.8

1,700 ± 400

1,300 ± 200

700 ± 300

18790

37380

74750

4,640 ± 90

9,100 ± 200

17,400 ± 400

4.9

500 ± 300

^aMeasurements were performed in a sealed 40 mL EPA vial containing 10 mL 0.4M CAN solution buffered in

1M HNO₃and the appropriate complex. O₂evolution was dynamically monitored with digital manometry and end

points were calibrated by gas chromatography; TOF_maxvalues were determined by calculating the rate of change

of the generated oxygen over time from the kinetic trace.

The water oxidation catalytic activity of the regioisomers 12a and 12b of the parent

para-PYA complex 1 was also evaluated (Table 2). Complex 12b with a mesoionic PYA ligand

displayed water oxidation activity similar to that of the 4-substituted complex 1, reaching TOF

of ~1700 h^–1(entries 30–33). This activity enhancement is marginal at best, despite the presence

of a more electron-donating mesoionic ligand, which was expected to stabilize higher oxidation

states better than the ortho- and para-PYA ligands (cf electrochemistry above). Surprisingly,

complex 12a bearing a lower donor strength ligand than analogues 1 and 12b shows about the

same performance as the other two regioisomers and achieves TOFs up to 1800 h^–1(entries 26–

29). Obviously, catalytic activity is not controlled solely by the donor strength of the ligand,

but also by other factors such as steric influences that affect the (de)stabilization of the metal-

bound water or hydroxide intermediates during the catalytic reaction.

DFT modeling was used to rationalize the structure-activity relationship observed for

these catalysts. A geometry optimization (B3LYP/LANL2DZ) indicated why compounds 4 and

5 were outperforming the other catalysts: the oxygen lone-pair of the methoxy group is in close

proximity to the aqua ligand and forms a strong hydrogen bond (Figure 7a). This arrangement

facilitates the deprotonation of this aqua site, especially as oxidation of the Ir(III) center

increases the acidity of the linked proton dramatically.^16c,31Such hydrogen bonding and transfer

can be repeated for the second proton after the iridium center is further oxidized to the Ir^IV–OH

species, a process which constitutes a pathway for efficient proton-coupled electron transfer

(PCET) and therefore results in a low-energy pathway towards a very reactive oxo-complex.

13

10.1002/chem.201705619

Chemistry - A European Journal

The PCET processes have been found to accelerate water oxidation in biological systems as

well as in artificial electrochemically or chemically driven processes.³²

The smaller rate increase observed with complexes 12a and 12b is less obvious from

the DFT structures. A hydrogen bond between the carbonyl group and the methyl group forming

the pyridinium moiety stabilizes the ligand and induces a significant dihedral angle between

the pyridylidene heterocycle and the phenylamide portion of the ligand (q = 53°), which

weakens the electronic conjugation and lowers the relevance of the diene-type resonance

structure (Figure 7b). It is important to note that the increase in reactivity imparted by this twist

due to steric crowding is less notable than that caused by the direct link formed by the hydrogen

bond between the methoxy group and the aqua ligand in 4 and 5.

A)

B)

5

12a

12b

Figure 7. Hydrogen bonds lead to faster water oxidation: a) The methoxy group attached to the carbon adjacent

to the cyclometalating site in complexes 4 and 5 is optimally positioned for accepting the proton from the aquo

(and likewise hydroxyl) ligand upon oxidation of the iridium center; b) A weak hydrogen bond between the PYA

N–CH₃and the amide CO group stabilizes the ortho-PYA ligand in 12a at a bigger dihedral angle (53°), which

increases its donor strength compared to the considerably flatter para-substituted parent compound 1.

Catalyst optimization. While the introduction of long aliphatic chains had no effect on catalyst

activity, modulation of the donor properties of both the phenyl group (by introducing electron-

donating methoxy substituents) and the PYA residues (by introducing mesoionic meta-PYA or

sterically constrained ortho-PYA motifs) led to more active water oxidation catalysts. Merging

these two leading catalyst design strategies was therefore a clear path to further improve the

performance of the catalyst. The synthesis and characterization of the mesoionic target complex

16b has been communicated recently.²⁵The analogue featuring an ortho-PYA ligand was

prepared along the same route starting from the trimethoxy-substituted benzoyl chloride and 2-

aminopyridine to form the amide 13, followed by selective pyridine methylation with MeI to

14

10.1002/chem.201705619

Chemistry - A European Journal

form the corresponding pyridinium salt 14 (Scheme 4). Deprotonated of the amide with aqueous

NaOH afford the free PYA ligand as air-stable solid. The structure of the mesoionic PYA 15b

has been reported as the first example of a mesoionic N-donor ligand.²⁵Complexation to

iridium and cyclometalation was performed as described above in the presence of NaOAc and

yielded the iridium complexes 16a,b in good yields (69–77%). Both complexes were fully

characterized by NMR spectroscopy and HR-MS. In addition, single crystals of iridium

complex 16a were obtained and analyzed by X-ray diffraction (Figure 8). The overall molecular

geometry as well as bond lengths and angles around the iridium center are similar to the other

structures reported here. Specifically, contribution of the diene-type resonance form of the

ortho-PYA unit was noted by the bond length alteration in the heterocycle. The C8–C9 and

C10–C11 bond lengths are 1.391(5) Å and 1.397(6) Å, respectively, and slightly longer than

the C9–C10 and C11–C12 bonds (1.374(5) and 1.353(6) Å, respectively). These differences are

smaller than in the para-PYA ligand of complexes 1–5, and 8a–c, but distinct from the

mesoionic and fully conjugated meta-PYA system in complex 12b (cf Figure 4). Moreover, the

pyridylidene heterocycle is significantly twisted out of the phenylamide plane, as indicated by

the C7–N2–C8–N1 torsion angle q = –60.40(4)°. This value is close to the 53° predicted by

DFT for these ortho-PYA iridium complexes (cf Figure 7b) and hence validate the steric

argument suggested for the rate enhancement observed for complex 12a. Moreover, the

distance between the N–CH₃proton and the amide oxygen atom is 2.337(2) Å, very similar to

the 2.22 Å predicted by DFT calculations.

MeO

Ir

Cl

MeO

N

I

O

N

O

MeO

O

MeO

ii)

iii)

i)

MeO

Cl

N

MeO

N

16a

N

H

N

OMe

MeO

Ir

Cl

14

a = ortho

b = meta

15

N

MeO

O

16b

Scheme 4. Synthesis of iridium complexes 16a and 16b. i) 2- or 3-aminopyridine, NEt₃, MeI. ii) NaOH. iii)

[Ir(Cp*)Cl₂]₂, NaOAc.

15

10.1002/chem.201705619

Chemistry - A European Journal

Figure 8. ORTEP diagram of iridium complex 16a. Thermal ellipsoids are given at the 50% probability level;

hydrogen atoms and co-crystallized solvent molecules are omitted for clarity. Selected bond lengths (Å): Ir–C1

2.051(3), Ir–N2 2.116(3), Ir–Cl 2.4274(9), Ir–Cp(centroid) 1.8189(16), C8–C9 1.391(5), C9–C10 1.374(5), C10–

C11 1.397(6), C11–C11 1.353(6), N2–C8 1.370(4), C8–N1 1.370(4), C12–N1 1.363(4). Selected bond angles

(deg): C1–Ir–N1 77.40(12), C1–Ir–Cl 85.67(8), N2–Ir–Cl 87.33(8). The twisted out of plane distortion of the PYA

ligand is clearly visible (dihedral angle C7–N2–C8–N1 60.40(4)°).

Electrochemical analysis (CV) revealed a reversible oxidation potential at E_1/2= +0.69

and +0.55 V vs SCE for complexes 16a and 16b, respectively. The potentials are moderately

lower by 30 and 10 mV when compared to the potentials of the corresponding complexes

without methoxy substituents (E_1/2= +0.72 for complex 12a and +0.56 V for 12b). A similar

lowering of the oxidation potential by 30 mV was observed upon introduction of three MeO

groups in the para-PYA complex 1 (E_1/2= +0.64 V, vs +0.61 V for 5).³³

Table 3. Catalytic water oxidation with complex 16a and 16b.^a)

Entry

Complex

16a

conc. [µM] CAN/complex

TON

TOF_max(h^–1)

1600±400

1160±80

1

2

3

4

49.9

19.9

49.9

19.9

7360

18440

7360

1810±30

4590±30

1810±40

4600±70

16a

16b

1500±400

16b

18440

1210±60

^a)Measurements were performed in a sealed 40 mL EPA vial containing 10 mL 0.4M CAN solution buffered in

1M HNO₃and the appropriate complex. O₂evolution was dynamically monitored with digital manometry and end

points were calibrated by gas chromatography; TOF_maxvalues were determined by calculating the rate of change

of the generated oxygen over time from the kinetic trace.

Even though each of the two structural features employed in the ligand design of

complexes 16a and 16b, viz the three methoxy substituents on the cyclometalating phenyl group

(cf complex 5) and the variation of the PYA ligands (cf complexes 12a and 12b), individually

improved water oxidation, the water oxidation activity of complexes 16a and 16b is much more

aligned with the performance of the parent complex 1, and no improved performance was

16

10.1002/chem.201705619

Chemistry - A European Journal

observed (Table 3). The activity of both 16a and 16b is less than that of the methoxy-free

homologues 12a and 12b (e.g. for complex 16a, TOF_maxat 50 µM catalyst concentration is

1600 h^–1vs 1800 h^–1for 12a) and substantially lower than that of the (MeO)₃-substituted para-

PYA analogue 5 (TOF_max= 3200 h^–1under identical conditions). Similarly, complex 16b

reaches a TOF_max= 1500 h^–1at 50 µM iridium concentration, which is also lower than complex

12b without methoxy groups (TOF_max= 1700 h^–1).

Structural analysis of the aqua intermediate by DFT (B3LYP/LANL2DZ) calculations

offered two possible explanations for the underperformance of these two complexes, namely

lower stability of the complex and lower accessibility of the catalytically active site. First,

moving the pyridinium site from the isolated 4-position to the more crowded parts of the

molecule increased the steric energy of the complexes significantly. The calculated heat of

formation for the meta-PYA complex 16b is 6.2 kcal mol^–1higher, while the ortho- regioisomer

16a has a 7.6 kcal mol^–1higher energy than the most stable complex 5. Second, the steric effects

observed in the more congested isomers cause substantial twisting of the PYA ligand that is

amplified by the three added methoxy groups. While the increase in steric energy might lead to

an easier degradation of the two more congested complexes during catalysis, it is also evident

that access to the active site is hindered by the twisted pyridinium ring. Figure 9 highlights this

effect with a space-filling model of the two regioisomers 5 and 16a containing a substrate water

molecule bound to iridium rather than a chloride. At a certain point on the catalytic cycle, water

has to attack the oxygen of the active site to form an O–O bond.^3bThis attack is partially

hindered by the congestion and the ensuing twisting of the ligand in complex 16a (dashed green

arrow), and this reduced access will decrease water oxidation rates of this catalyst. In contrast,

access to the iridium-bound ligand is much less restricted in complex 5 (solid green arrow).³⁴

16a

16b

5

Figure 9. DFT calculations (B3LYP/LANL2DZ) were used to produce optimized structures of the para-PYA

complex 5 and the regioisomeric ortho-PYA complex 16a. The access to the active site required to form a O–O

17

10.1002/chem.201705619

Chemistry - A European Journal

bond is hindered by the twisted pyridinium ring (dashed arrow, arrow head pointing to coordinated water

molecule), which leads to a decrease in reaction rates in 16a and 16b when compared with the isomeric champion

catalyst 5 (compare accessibility of iridium (light blue) and coordinated water molecule as indicated by solid

arrow). The increased steric energy of 16a and 16b might also lead to a more rapid degradation of these two

catalysts.

Conclusions

We have successfully exploited the synthetic versatility of PYA ligands to prepare and evaluate

a diverging series of iridium complexes containing chelating aryl-PYA ligands for water

oxidation catalysis. Simple modifications, either by incorporating long aliphatic chains, or

through electronic modulation by introducing electron-donating methoxy substituents on the

aryl ring, or by altering the PYA positions (ortho-, meta-, para-PYA), were evaluated to

optimize iridium-catalyzed water oxidation. While the incorporation of long alkyl chains had

no significant effect, electronic modulation was effective and led to a substantial enhancement

of catalytic activity and doubled the catalytic activity of the original lead compound to a TOF_max

= 3300 h^–1. This rate is substantially faster than most of the iridium-based water oxidation

catalysts (typically 1000–1800 h^–1), but lower than the fastest triazolylidene or pyridyl-

carboxylate catalysts (TOF_max7,000–15,000 h^–1). Even though optimization of the aryl and the

PYA domains of the aryl-PYA chelate each led to improved catalysis, merging of the two

optimized motifs in a single new aryl-PYA ligand failed to improve catalytic activity.

Computational insights suggest an over-congestion at the active site, which hinders catalytic

turnover. Moreover, these results indicate that optimization of the second coordination sphere

provides an effective approach for further enhancing the catalytic activity of the metal center,

particularly through introduction of advantageous hydrogen bonding sites. These results

illustrate the complexity of ligand design and the subtle effects at play in catalytic reactions,

and they emphasize the high versatility of the PYA platform for optimizing catalytic activity.

Finally, it is worth noting that the electrochemical analysis of the catalyst precursors showed

no correlation with catalytic activity. Since the stringent structure-activity relationship observed

in this series of complexes is consistent with a molecularly defined catalytically active site,

different mechanisms such as redox potentials of the pre-activated species appear to be relevant

for facilitating catalytic turnover. Such mechanisms are not necessarily correlated to properties

of the catalyst precursor, as demonstrated here with the electrochemical data.

18

10.1002/chem.201705619

Chemistry - A European Journal

Experimental

35

General. The metal precursor salt [Ir(Cp*)Cl₂]₂and complexes 1,¹⁸2–5,¹⁹and compounds

9,²⁶14b–16b²⁵were all synthesized as reported in the literature. All other reagents were

commercially available and used as received. Unless specified otherwise, NMR spectra were

recorded at 25 °C on Bruker spectrometers operating at 300 or 400 MHz (¹H NMR), and 100

MHz (¹³C NMR), respectively. Chemical shifts (δ in ppm, coupling constants J in Hz) were

referenced to residual solvent signals (¹H, ¹³C). Assignments are based on homo- and

heteronuclear shift correlation spectroscopy. Purity of bulk samples of the complexes has been

established by NMR spectroscopy, and by elemental analysis, which were performed at the

University of Bern Microanalytic Laboratory using a Thermo Scientific Flash 2000 CHNS-O

elemental analyser. Residual solvent was confirmed by NMR spectroscopy and also by X–ray

structure determinations. High-resolution mass spectrometry was carried out with a Thermo

Scientific LTQ Orbitrap XL (ESI-TOF).

Electrochemistry. Electrochemical measurements were carried out using an EG&G Princeton

Applied Research potentiostat model 273A typically at a 100 mV s^–1sweep rate employing a

gastight three-electrode cell under an argon atmosphere. A Pt disk with a 3.80 mm²surface area

was used as the working electrode and was polished before each measurement. The reference

electrode was a Ag/AgCl electrode; the counter electrode was a Pt wire. Bu₄NPF₆(0.1 M) in

dry CH₂Cl₂was used as a base electrolyte with analyte concentrations of approximately 10^–3

M. The ferrocenium/ferrocene redox couple was used as an internal reference (E_1/2= +0.46 V

vs SCE in CH₂Cl₂).³⁶

General procedure for amide formation (9a, 9b, and 13b). A solution of the acyl chloride

(11 mmol) in THF (10 mL) was added dropwise to a solution of aminopyridine (0.940 g, 10

mmol) and NEt₃(1.50 mL, 11 mmol) in THF (30 mL), which resulted in the formation of a

white suspension. The reaction was refluxed for 48 h. After cooling to rt, the solids were

removed by filtration and washed with THF (50 mL). The filtrates were combined and all

volatiles were removed in vacuo to leave a white solid, which was purified by column

chromatography (SiO₂; CH₂Cl₂/MeOH 95:5). All solvents were removed to yield the

19

10.1002/chem.201705619

Chemistry - A European Journal

corresponding amide as a white solid. Analytical data of these compounds are in agreement

with those reported.^37,38

I. General procedure for the formation of pyridinium salts (6a–c, 10a,b and 14a). N-

(pyridin-4-yl)benzamide, 9a,b or 13a (2.0 mmol) was dissolved in MeCN (15 mL) in a pressure

tube. The alkylating agent RI (R =ⁱPr, ⁿBu or Oct, 3.0 mmol) was added and the reaction was

stirred for 18 h at 80 °C. The solution was cooled to rt and concentrated to 5 mL. Et₂O (50 mL)

was added affording compound as 6a–c, 10a,b and 14a a yellow solid, which was collected by

filtration.

II. General procedure for synthesis of the pyridylidene amides (7a–c, 11a,b and 15a). A

suspension of the pyridinium salt 6a–c, 10a,b or 14a (1.0 mmol) in CH₂Cl₂(10 mL) was placed

into a separating funnel. Aqueous NaOH (15 mL, 2M) was added. After vigorous mixing, the

organic phase was collected. The aqueous layer was extracted with CH₂Cl₂(3 x 15 mL), the

combined organic layers were dried over anhydrous Na₂SO₄and filtered. All volatiles were

removed, thus giving free ligand 7a–c, and 15a as white-yellow solids, which were purified by

recrystallization from pentane and 11a,b as yellow oils.

III. General procedure for the synthesis of iridium complexes (8a–c, 12a,b and 16a).

Compound 7a–c, 11a,b or 15a (0.4 mmol), NaOAc (33 mg, 0.4 mmol) and [Ir(Cp*)Cl₂]₂(120

mg, 0.15 mmol) were dissolved in CH₂Cl₂and stirred at rt for 18h. All volatiles were removed

under reduced pressure and the crude solid was purified by column chromatography (SiO₂;

CH₂Cl₂/MeOH 98:2), thus yielding 8a–c, 12a.b and 16a as orange solids.

Compound 6a

Compound 6a was prepared according to the general procedure I from N-(pyridin-4-

yl)benzamide, (396 mg, 2.0 mmol) and ⁱPrI (294 µL, 3.0 mmol). Yield: 690 mg, 94%.

¹H NMR (300 MHz, d₆-DMSO): δ 11.54 (s, 1H, NH), 8.98 (d, ³J_HH= 6.4 Hz, 2H, CH_pyr), 8.31

(d, ³J_HH= 6.4 Hz, 2H, CH_pyr), 8.03 (d, 2H, ³J_HH= 7.6, CH_Ph), 7.72 (t, 1H, ³J_HH= 7.6 Hz, ⁴J_HH

1.3 Hz, CH_Ph), 7.63 (t, 2H, ³J_HH= 7.6 Hz, CH_Ph), 4.88 (septet, 1H, ³J_HH= 6.8 Hz, CHMe₂), 1.57

=

3

13

(d, 6H, J_HH= 6.8 Hz, C–CH₃). C{¹H} NMR (100 MHz, d₆-DMSO): δ 167.7 (CO), 152.4

(C_pyr), 143.3 (CH_pyr), 133.3 (CH_Ph), 132.9 (C_Ph), 128.8 (CH_Ph), 128.3 (CH_Ph), 115.9 (CH_pyr),

20

10.1002/chem.201705619

Chemistry - A European Journal

62.2 (CHMe₂), 22.3 (C–CH₃). HR-MS: m/z calculated for C₁₅H₁₇N₂O [M–I]⁺= 241.1341;

found, 241.1329.

Compound 6b

Compound 6b was prepared according to the general procedure I from N-(pyridin-4-

yl)benzamide, (396 mg, 2.0 mmol) and nBuI (341 µL, 3.0 mmol). Yield: 728 mg, 95%.

¹H NMR (400 MHz, d₆-DMSO): δ 11.55 (s, 1H, NH), 8.88 (d, ³J_HH= 7.3 Hz, 2H, CH_pyr), 8.31

(d, ³J_HH= 7.3 Hz, 2H, CH_pyr), 8.11–7.96 (m, 2H, CH_Ph), 7.75–7.68 (m, 1H, CH_Ph), 7.64–7.60

(m, 2H, CH_Ph), 4.47 (t, ³J_HH= 7.4 Hz, 2H, NCH₂), 1.86 (quint, ³J_HH= 7.4 Hz, 2H, NCH₂CH₂),

1.29 (sext, ³J_HH= 7.4 Hz, 2H, CH₂CH₃), 0.92 (t, ³J_HH= 7.4 Hz, 3H, CH₃). ¹³C{¹H} NMR (100

MHz, d₆-DMSO): δ 167.2 (CO), 152.0 (C_pyr), 145.0 (CH_pyr), 133.2 (CH_Ph), 132.9 (C_Ph), 128.8

(CH_Ph), 128.3 (CH_Ph), 115.7 (CH_pyr), 58.9 (NCH₂), 32.3 (NCH₂CH₂), 18.7 (CH₂CH₃), 13.3

(CH₃). HR-MS: m/z calculated for C₁₆H₁₉N₂O [M–I]⁺= 255.1492; found 255.1489. Elem.

Anal. Calcd. for C₁₅H₁₉IN₂O: C, 50.28; H, 5.01; N, 7.33; Found: C, 50.31; H, 5.29; N, 7.53%.

Compound 6c

Compound 6c was prepared according to the general procedure I from N-(pyridin-4-

yl)benzamide, (396 mg, 2.0 mmol) and nOctI (541 µL, 3.0 mmol). Yield: 799 mg, 91%.

¹H NMR (400 MHz, d₆-DMSO): δ 11.55 (s, 1H, NH), 8.85 (d, ³J_HH= 7.4 Hz, 2H, CH_pyr), 8.28

(d, ³J_HH= 7.4 Hz, 2H, CH_pyr), 8.11–8.00 (m, 2H, CH_Ph), 7.78–7.67 (m, 1H, CH_Ph), 7.64–7.59

(m, 2H, CH_Ph), 4.44 (t, ³J_HH= 7.2 Hz, 2H, NCH₂), 1.86 (quint, ³J_HH= 7.2 Hz, 2H, NCH₂CH₂),

3

13

1.31–1.21 (m, 10H, CH₂), 0.86 (t, J_HH= 7.0 Hz, 3H, CH₃). C{¹H} NMR (100 MHz, d₆-

DMSO): δ 167.4 (CO), 152.5 (C_pyr), 144.9 (CH_pyr), 133.1 (CH_Ph), 133.1 (C_Ph), 128.7 (CH_Ph),

128.3 (CH_Ph), 115.8 (CH_pyr), 59.1 (NCH₂), 31.1 (NCH₂CH₂), 30.3, 28.4, 28.3, 25.3, 22.0 (5 ´

CH₂), 13.9 (CH₃). HR-MS: m/z calculated for C₂₀H₂₇N₂O [M–I]⁺= 311.2118; found; 311.2111.

Elem. Anal. Calcd. For C₂₀H₂₇IN₂O C, 54.80; H, 6.21; N, 6.39; Found: C, 54.83; H, 6.97; N,

6.37%.

Compound 7a

Compound 7a was obtained as a white solid following the general procedure II from compound

6a (368 mg, 1.0 mmol). Yield: 175 mg, 73%.

¹H NMR (300 MHz, d₆-DMSO): δ 8.13 (d, ³J_HH= 6.0 Hz, 2H, CH_pyr), 8.12 (d, ³J_HH= 7.3 Hz,

2H, CH_Ph), 7.47 (d, ³J_HH= 6.0 Hz, 2H, CH_pyr), 7.42 (t, 1H, ³J_HH= 7.3, CH_Ph), 7.38 (t, 2H, ³J_HH

3

= 7.3, CH_Ph4.45 (septet, 1H, J_HH= 6.8, CHMe₂), 1.44 (d, 6H, J_HH= 6.8, C–CH₃). ¹³C{¹H}

21

10.1002/chem.201705619

Chemistry - A European Journal

NMR (100 MHz, d₆-DMSO): δ 173.9 (CO), 164.6 (C_pyr), 139.8 (C_Ph), 138.9 (CH_pyr), 130.1

(CH_Ph), 128.7 (CH_Ph), 127.6 (CH_Ph), 117.9 (CH_pyr), 58.7 (CHMe₂), 22.2 (C–CH₃). HRMS: m/z

calculated for C₁₅H₁₇N₂O [M+H]⁺= 241.1335; found, 241.1328. Elem. Anal. Calcd. for

C₁₅H₁₆N₂O: C, 74.97; H, 6.71; N, 11.66; Found: C, 74.29; H, 6.22; N, 11.85%.

Compound 7b

Compound 7b was obtained as a white solid following the general procedure II from compound

6b (382 mg, 1.0 mmol). Yield: 219 mg, 84%.

¹H NMR (400 MHz, d₆-DMSO): δ 8.17–8.08 (m, 2H, CH_Ph), 8.02 (d, 2H, ³J_HH= 7.4, CH_pyr),

7.46 (d, ³J_HH= 7.4 Hz, 2H, CH_pyr), 7.43–7.34 (m, 3H, CH_Ph) 4.05 (t, ³J_HH= 7.4 Hz, 2H, NCH₂),

3

1.73 (quint, J_HH= 7.4 Hz, 2H, NCH₂CH₂), 1.26 (sext, J_HH= 7.4 Hz, 2H CH₂CH₃), 0.90 (d,

13

³J_HH= 7.4 Hz, 3H, CH₃). C{¹H} NMR (100 MHz, d₆-DMSO): δ 173.8 (CO), 164.2 (C_pyr),

140.9 (CH_pyr), 139.7 (C_Ph), 130.1 (CH_Ph), 128.7 (CH_Ph), 127.5 (CH_Ph), 117.7 (CH_pyr), 56.2

(NCH₂), 32.3 (NCH₂CH₂), 18.8 (CH₂CH₃), 13.4 (CH₃). HR-MS: m/z calculated for C₁₆H₁₉N₂O

[M+H]⁺= 255.1492; found; 255.1485. Elem. Anal. Calcd. for C₁₆H₁₈N₂O: C, 75.56; H, 7.13;

N, 11.01; Found: C, 75.34; H, 7.02; N, 11.01%.

Compound 7c

Compound 7c was obtained as white solid following the general procedure II from compound

6c (438 mg, 1.0 mmol). Yield: 239 mg, 77%.

¹H NMR (400 MHz, d₆-DMSO): δ 8.23–8.08 (m, 2H, CH_Ph), 8.02 (d, 2H, ³J_HH= 7.4, CH_pyr),

7.45 (d, ³J_HH= 7.4 Hz, 2H, CH_pyr), 7.43–7.34 (m, 3H, CH_Ph) 4.04 (t, ³J_HH= 7.2 Hz, 2H, NCH₂),

1.74 (quint, ³J_HH= 7.2 Hz, 2H, NCH₂CH₂), 1.32–1.18 (m, 10H, CH₂), 0.85 (d, ³J_HH= 6.8 Hz,

3H, CH₃). ¹³C{¹H} NMR (100 MHz, d₆-DMSO): δ 173.9 (CO), 164.3 (C_pyr), 140.9 (CH_pyr),

139.8 (C_Ph), 130.1 (CH_Ph), 128.7 (CH_Ph), 127.5 (CH_Ph), 117.7 (CH_pyr), 56.5 (NCH₂), 31.1

(NCH₂CH₂), 30.3, 28.5, 28.4, 25.5, 22.0 (5 ´ CH₂), 13.9 (CH₃). HR-MS: m/z calculated for

C₁₅H₁₈N₂O [M+H]⁺= 311.2118; found; 311.2114. Elem. Anal. Calcd. for C₁₆H₁₈N₂O: C, 77.38;

H, 8.44; N, 9.02; Found: C, 76.96; H, 8.40; N, 9.14%.

Compound 8a

Compound 8a was prepared according to the general procedure III from 7a (96 mg, 0.40 mmol)

to give an orange solid. Yield: 98 mg, 54%.

¹H NMR (500 MHz, CDCl₃): δ 8.54 (d, ³J_HH= 6.9 Hz, 2H, CH_pyr), 7.69 (d, ³J_HH= 7.4 Hz, 1H,

CH_Ph), 7.67 (d, ³J_HH= 6.9 Hz, 2H, CH_pyr), 7.57 (d, ³J_HH= 7.4 Hz, H, CH_Ph) 7.21 (t, ³J_HH= 7.4

22

10.1002/chem.201705619

Chemistry - A European Journal

3

Hz, 1H, CH_Ph), 6.93 (t, J_HH= 7.4 Hz, 1H, CH_Ph), 4.27 (septett, J_HH= 7.0 Hz, 1H, CHMe₂),

1.54 (s, 15H, Cp–CH₃), 1.43 (d, ³J_HH= 6.3 Hz, CH–CH₃). ¹³C{¹H} NMR (125 MHz, CDCl₃):

δ 181.1 (CO), 162.4 (C_pyr), 159.4 (C_Ph–Ir), 140.9 (C_Ph), 137.4 (CH_pyr), 136.2 (CH_Ph), 132.3

(CH_Ph), 127.9 (CH_Ph), 122.1 (CH_Ph), 121.1 (CH_pyr), 87.3 (C_Cp), 60.0 (NCHMe₂), 23.0 (CH–

CH₃), 9.2 (Cp–CH₃). HR-MS: m/z calculated for C₂₅H₃₀IrN₂O [M–Cl]⁺= 567.1988; found,

567.1976. Elem Anal. Calcd. for C₂₅H₃₀ClIrN₂O x 0.5 CH₂Cl₂: C, 47.51; H, 4.85; N, 4.35;

Found C, 47.99; H, 4.76; N, 4.31.

Compound 8b

Compound 8b was prepared according to the general procedure III from 7b (102 mg, 0.40

mmol) to give an orange solid. Yield: 110 mg, 76%.

¹H NMR (400 MHz, CDCl₃): δ 8.31 (d, ³J_HH= 7.0 Hz, 2H, CH_pyr), 7.72 (dd, ³J_HH= 7.6 Hz, ⁴J_HH

= 1.1 Hz 1H, CH_Ph), 7.61–7.52 (m, 3H, CH_pyr+ CH_Ph), 7.22 (td, ³J_HH= 7.3 Hz, ⁴J_HH= 1.5 Hz,

1H, CH_Ph), 6.95 (td, ³J_HH= 7.3 Hz, ⁴J_HH= 1.1 Hz, 1H, CH_Ph), 3.81–3.52 (m, 2H, NCH₂), 1.51–

1.41 (m, 2H, NCH₂CH₂), 1.46 (s, 15H, Cp–CH₃), 1.11–0.93 (m, 2H, CH₂CH₃), 0.75 (t, ³J_HH

=

7.3 Hz, 2H, CH₂CH₃). ¹³C{¹H} NMR (100 MHz, CDCl₃): δ 183.3 (CO), 162.4 (C_pyr), 159.6

(C_Ph–Ir), 141.2 (C_Ph), 140.3 (CH_pyr), 136.3 (CH_Ph), 132.2 (CH_Ph), 127.8 (CH_Ph), 122.1 (CH_Ph),

121.2 (CH_pyr), 87.2 (C_Cp), 58.3 (NCH₂), 32.2 (NCH₂CH₂), 19.2 (CH₂CH₃), 13.5 (CH₂CH₃), 9.1

(Cp–CH₃). HR-MS: m/z calculated for C₂₆H₃₂IrN₂O [M–Cl]⁺= 581.2144; found, 581.2125.

Elem Anal. Calcd. for C₂₆H₃₂ClIrN₂O: C, 50.68; H, 5.23; N, 4.55; Found: C, 50.12; H, 4.98; N,

4.53%.

Compound 8c

Compound 8c was prepared according to the general procedure III from 7c (124 mg, 0.40

mmol) to give an orange solid. Yield: 120 mg, 75%.

¹H NMR (400 MHz, CDCl₃): δ 8.30 (d, ³J_HH= 7.1 Hz, 2H, CH_pyr), 7.71 (dd, ³J_HH= 7.5 Hz, ⁴J_HH

= 1.0 Hz 1H, CH_Ph), 7.62–7.56 (m, 1H, CH_Ph), 7.57 (d, ³J_HH= 7.1 Hz, 2H, CH_pyr), 7.21 (td, ³J_HH

= 7.3 Hz, ⁴J_HH= 1.5 Hz, 1H, CH_Ph), 6.94 (td, ³J_HH= 7.3 Hz, ⁴J_HH= 1.3 Hz, 1H, CH_Ph), 3.72–

3.53 (m, 2H, NCH₂), 1.45 (s, 15H, Cp–CH₃),1.44–1.39 (m, 2H, NCH₂CH₂), 1.30–1.21 (m, 2H,

CH₂), 1.19–1.08 (m, 6H, CH₂), 0.96–0.86 (m, 2H, CH₂), 0.86 (t, ³J_HH= 7.1 Hz, 3H, CH₂CH₃).

¹³C{¹H} NMR (100 MHz, CDCl₃): δ 183.2 (CO), 162.5 (C_pyr), 159.6 (C_Ph–Ir), 141.2 (C_Ph),

140.4 (CH_pyr), 136.3 (CH_Ph), 132.2 (CH_Ph), 127.7 (CH_Ph), 122.1 (CH_Ph), 121.2 (CH_pyr), 87.2

(C_Cp), 58.6 (NCH₂), 31.7 (NCH₂CH₂), 31.2, 29.1, 28.9, 25.9, 22.7 (5 ´ CH₂), 14.2 (CH₂CH₃),

9.1 (Cp–CH₃). HR-MS: m/z calculated for C₃₀H₄₀IrN₂O [M–Cl]⁺= 637.2770; found, 637.2739.

23

10.1002/chem.201705619

Chemistry - A European Journal

Elem Anal. Calcd. for C₃₀H₄₀ClIrN₂O ´ 0.5 CH₂Cl₂: C, 51.15; H, 5.43; N, 3.92; Found: C,

50.58; H, 5.43; N, 3.84%.

Compound 10a

Compound 10a was prepared according to the general procedure I from 9a (396 mg, 2.0 mmol)

and MeI (187 µL, 3.0 mmol): Yield 251 mg, 39%.

¹H NMR (400 MHz, d₆-DMSO): δ 11.40 (s, 1H, NH), 8.95 (d, ³J_HH= 6.3, Hz, 1H, CH_pyr), 8.58

(t, ³J_HH= 8.1, ⁴J_HH= 1.4 Hz, 1H, CH_pyr), 8.22 (dd, ³J_HH= 8.1, ⁴J_HH= 1.4 Hz, 1H, CH_pyr), 8.11–

8.01 (m, 2H, CH_Ph), 7.89 (t, ³J_HH= 6.3 Hz, 1H, CH_pyr), 7.75–7.68 (m, 1H, CH_Ph), 7.65–7.59 (m,

2H, CH_Ph), 4.29 (s, 3H, NCH₃). ¹³C{¹H} NMR (100 MHz, d₆-DMSO): δ 166.6 (CO), 148.1

(C_pyr), 146.0 (CH_pyr), 145.6 (CH_pyr), 133.2 (CH_Ph), 132.5 (C_Ph), 128.7 (CH_Ph), 128.6 (CH_Ph),

124.1 (CH_pyr), 123.2 (CH_pyr), 44.3 (NCH₃). HR-MS: m/z calculated for C₁₃H₁₃N₂O [M–I]⁺=

213.1022; found, 213.1025. Elem. Anal. Calcd. For for C₁₃H₁₃IN₂O ´ 0.5 H₂O: C, 44.72; H,

4.04; N, 8.02; Found: C, 44.43; H, 3.90; N, 7.98%.

Compound 10b

Compound 10b was prepared according to the general procedure I from 9b (396 mg, 2.0 mmol)

and MeI (187 µL, 3.0 mmol): Yield 646 mg, 95%.

¹H NMR (400 MHz, d₆-DMSO): δ 11.17 (s, 1H, NH), 9.52 (s, 1H, CH_pyr), 8.76 (d, ³J_HH= 5.9,

3

Hz 1H, CH_pyr), 8.65 (d, J_HH= 8.8 Hz, 1H, CH_pyr), 8.14 (dd, J_HH= 8.8, 5.9 Hz, 1H, CH_pyr),

8.05–7.99 (m, 2H, CH_Ph), 7.73–7.66 (m, 1H, CH_Ph), 7.64–7.57 (m, 2H, CH_Ph), 4.41 (s, 3H,

NCH₃). ¹³C{¹H} NMR (100 MHz, d₆-DMSO): δ 166.3 (CO), 140.3 (C_pyr), 139.0 (CH_pyr), 136.3

(CH_pyr), 134.6 (CH_pyr), 133.0 (C_Ph), 132.8 (CH_Ph), 128.5 (CH_Ph), 127.9 (CH_Ph), 127.8 (CH_pyr),

48.6 (NCH₃). HR-MS: m/z calculated for C₁₃H₁₃N₂O [M–I]⁺= 213.1022; found, 213.1032.

Compound 11a

Compound 11a was obtained as pale yellow oil following the general procedure II from

compound 10a (320 mg, 1.0 mmol). Yield: 174 mg, 82%.

¹H NMR (400 MHz, d₆-DMSO): δ 8.31 (d, ³J_HH= 8.9, Hz, 1H, CH_pyr), 8.21–8.17 (m, 2H, CH_Ph),

8.17–8.13 (m, 1H, CH_pyr), 7.74 (ddd, ³J_HH= 8.1, 6.7, ⁴J_HH= 1.5 Hz 1H, CH_pyr), 7.49–7.36 (m,

3H, CH_Ph), 6.73 (td, ³J_HH= 6.7, ⁴J_HH= 1.5 Hz, 1H, CH_pyr), 3.87 (s, 3H, NCH₃). ¹³C{¹H} NMR

(100 MHz, d₆-DMSO): δ 172.0 (CO), 158.4 (C_pyr), 141.0 (CH_pyr), 140.1 (CH_pyr), 139.4 (C_Ph),

130.5 (CH_Ph), 128.8 (CH_Ph), 127.7 (CH_Ph), 119.2 (CH_pyr), 111.2 (CH_pyr), 41.0 (NCH₃). HR-MS:

m/z calculated for C₁₃H₁₃N₂O [M+H]⁺= 213.1022; found, 213.1022.

24

10.1002/chem.201705619

Chemistry - A European Journal

Compound 11b

Compound 11b was obtained as pale yellow oil following the general procedure II from

compound 10b (320 mg, 1.0 mmol). Yield: 145 mg, 68%.

¹H NMR (400 MHz, d₆-DMSO): δ 9.22 (s, 1H, CH_pyr), 8.32 (d, ³J_HH= 9.0 Hz, 1H, CH_pyr), 8.20–

8.09 (m, 2H, CH_Ph), 8.02 (d, ³J_HH= 5.8 Hz, 1H, CH_pyr), 7.66 (dd, ³J_HH= 9.0, 5.8 Hz, 1H, CH_pyr),

7.30–7.26 (m, 3H, CH_Ph), 4.20 (s, 3H, NCH₃). ¹³C{¹H} NMR (100 MHz, d₆-DMSO): δ 170.5

(CO), 153.8 (C_pyr), 141.2 (C_Ph), 138.6 (CH_pyr), 136.4 (CH_pyr), 131.6 (CH_pyr), 128.9 (CH_Ph),

128.2 (CH_Ph), 127.2 (CH_Ph), 126.3 (CH_pyr), 47.3 (NCH₃). HR-MS: m/z calculated for

C₁₃H₁₃N₂O [M+H]⁺= 213.1022; found, 213.1030.

Compound 12a

Compound 12a was prepared according to the general procedure III from 11a (85 mg, 0.40

mmol) to give an orange solid. Yield: 124 mg, 72%.

¹H NMR (400 MHz, CD₂Cl₂): δ 8.74 (bs, 1H, CH_pyr), 8.08–7.94 (m, 2H, CH_pyr), 7.72 (d, ³J_HH

= 7.5 Hz, 1H, CH_Ph), 7.54 (dd, ³J_HH=7.5, ⁴J_HH= 1.6 Hz, 1H, CH_Ph), 7.25–7.15 (m, 1H, CH_pyr),

7.25–7.15 (m, 1H, CH_Ph), 6.96 (td, ³J_HH= 7.3, ⁴J_HH= 1.2 Hz, 1H, CH_Ph), 3.96 (s, 3H, NCH₃),

1.46 (s, 15H, Cp–CH₃). ¹³C{¹H} NMR (100 MHz, CD₂Cl₂): δ 179.6 (CO), 162.2 (C_pyr), 158.6

(C_Ph–Ir), 142.8 (CH_pyr), 141.1 (CH_pyr), 140.4 (C_Ph), 136.3 (CH_Ph), 131.8 (CH_Ph), 129.1 (CH_pyr),

128.6 (CH_Ph), 121.1 (CH_Ph), 119.6 (CH_pyr), 87.5 (C_Cp), 44.9 (NCH₃), 9.4 (Cp–CH₃). HR-MS:

m/z calculated for C₂₃H₂₆IrN₂O [M–Cl]⁺= 539.1669; found, 539.1665. Elem. Anal. Calcd. For

C₂₃H₂₆ClIrN₂O: C, 48.12; H, 4.56; N, 4.88; Found: C, 47.37; H, 4.30; N, 4.69.

Compound 12b

Compound 12b was prepared according to the general procedure III from 11b (85 mg, 0.40

mmol) to give an orange solid. Yield: 117 mg, 68%.

¹H NMR (400 MHz, CD₂Cl₂): δ 9.65 (s, 1H, CH_pyr), 9.00 (d, ³J_HH= 8.7 Hz, 1H, CH_pyr), 7.70

(d, ³J_HH= 7.6 Hz, 1H, CH_Ph), 7.67 (d, ³J_HH= 5.9 Hz, 1H, CH_pyr), 7.55–7.45 (m, 2H, CH_Ph+

CH_pyr), 7.17 (td, ³J_HH= 7.6, ⁴J_HH= 1.5 Hz, 1H, CH_Ph), 6.93 (t, ³J_HH= 7.6 Hz, 1H, CH_Ph), 4.12

(s, 3H, NCH₃), 1.47 (s, 15H, Cp–CH₃). ¹³C{¹H} NMR (100 MHz, CD₂Cl₂): δ 181.1 (CO),

158.7 (C_Ph–Ir), 152.9 (C_pyr), 141.8 (C_Ph), 141.6 (CH_pyr), 141.0 (CH_pyr), 136.3 (CH_Ph), 133.3

(CH_pyr), 131.6 (CH_Ph), 127.9 (CH_pyr), 126.9 (CH_Ph), 121.9 (CH_Ph), 87.3 (C_Cp), 48.8 (NCH₃),

9.4 (Cp–CH₃). HR-MS: m/z calculated for C₂₃H₂₆IrN₂O [M–Cl]⁺= 539.1669; found,

25

10.1002/chem.201705619

Chemistry - A European Journal

539.1684. Elem. Anal. Calcd. For C₂₃H₂₆ClIrN₂O x 0.2 CH₂Cl₂: C, 47.14; H, 4.50; N, 4.74;

Found: C, 46.96; H, 4.31; N, 4.57%.

Compound 14a

Compound 14a was prepared according to the general procedure I from 13a (576 mg, 2.0 mmol)

and MeI (187 µL, 3.0 mmol): Yield 500 mg, 58%.

¹H NMR (300 MHz, CD₃CN): δ 10.32 (s, 1H, NH), 8.63 (dd, ³J_HH= 6.4, ⁴J_HH= 1.7 Hz, CH_pyr

1H, CH_pyr), 8.46 (td, ³J_HH= 8.1, ⁴J_HH= 1.7, 1H, CH_pyr), 8.26 (dd, ³J_HH= 7.7, ⁴J_HH= 1.4 Hz, 1H,

CH_pyr), 7.77 (ddd, ³J_HH= 8.1, 6.4, ⁴J_HH= 1.4 Hz, 1H, CH_pyr), 7.42 (s, 2H, CH_Ph), 4.30 (s, 3H,

NCH₃), 3.94 (s, 6H, OCH₃), 3.79 (s, 3H, OCH₃). ¹³C{¹H} NMR (100 MHz, d₆-DMSO): δ 166.3

(CO), 153.9 (C_Ph), 148.8 (C_pyr), 146.9 (CH_pyr), 146.1 (CH_pyr), 142.6 (C_Ph), 127.5 (C_Ph), 125.9

(CH_pyr), 124.6 (CH_pyr), 106.8 (CH_Ph), 60.7 (OCH₃), 57.0 (OCH₃), 46.2 (NCH₃). Elem. Anal.

Calcd. For C₁₆H₁₉IN₂O₄: C, 44.67; H, 4.45; N, 6.51; Found: C 44.29; H, 4.58; N, 6.30%.

Compound 15a

Compound 15a was obtained as pale yellow solid following the general procedure II from

compound 14a (320 mg, 1.0 mmol). Yield: 241 mg, 80%

¹H NMR (400 MHz, d₆-DMSO): δ 8.26 (d, ³J_HH= 8.7 Hz, 1H, CH_pyr), 8.15 (dd, ³J_HH= 6.6,

⁴J_HH= 1.8 Hz, 1H, CH_pyr), 7.73 (ddd, ³J_HH= 8.7, 6.6, ⁴J_HH= 1.8 Hz, 1H, CH_pyr), 7.51 (s, 2H,

CH_Ph), 6.72 (td, ³J_HH= 6.6, ⁴J_HH= 1.8 Hz, 1H, CH_pyr), 3.87 (s, 3H, NCH₃), 3.83 (s, 6H,

OCH₃), 3.71 (s, 3H, OCH₃). ¹³C{¹H} NMR (100 MHz, d₆-DMSO): δ 171.4 (CO), 158.2

(C_pyr), 152.1 (C_Ph), 141.0 (CH_pyr), 140.0 (CH_pyr), 139.8 (C_Ph), 134.9 (C_Ph), 119.0 (CH_pyr),

111.0 (CH_pyr), 106.1 (CH_Ph), 60.0 (OCH₃), 55.7 (OCH₃), 41.0 (NCH₃). HR-MS: m/z

calculated for C₁₆H₁₉N₂O₄[M+H]⁺= 303.1339; found, 303.1338.

Compound 16a

Compound 16a was prepared according to the general procedure III from 15a (120 mg, 0.40

mmol) to give an orange solid. Yield: 165 mg, 69%.

¹H NMR (400 MHz, CD₂Cl₂): δ 8.92 (bs, 1H, CH_pyr), 8.01–7.90 (m, 2H, CH_pyr), 7.20 (td, ³J_HH

= 6.9, ⁴J_HH= 1.5 Hz, 1H, CH_pyr), 7.08 (s, 1H, CH_Ph), 3.97 (s, 3H, NCH₃), 3.93 (s, 3H, OCH₃),

3.83 (s, 3H, OCH₃), 3.71 (s, 3H, OCH₃), 1.46 (s, 15H, Cp–CH₃). ¹³C{¹H} NMR (100 MHz, d₆-

DMSO): δ 178.8 (CO), 161.5 (C_pyr), 156.3 (C_Ph–Ir), 149.8 (C_Ph), 146.1 (C_Ph), 143.4 (C_Ph), 142.6

(CH_pyr), 140.5 (CH_pyr), 134.6 (C_Ph), 128.6 (CH_pyr), 119.5 (CH_pyr), 108.4 (CH_Ph), 87.5 (C_Cp), 61.5

(OCH₃), 60.9 (OCH₃), 56.1 (OCH₃), 44.7 (NCH₃), 9.5 (Cp–CH₃). HR-MS: m/z calculated for

26

10.1002/chem.201705619

Chemistry - A European Journal

C₂₆H₃₂IrN₂O₄[M–Cl]⁺= 629.1991; found, 629.1974. Elem. Anal. Calcd. For C₂₆H₃₂ClIrN₂O₄:

C, 47.02; H, 4.86; N, 4.22; Found: C, 46.94; H, 4.86; N, 4.13.

Dynamic Oxygen Evolution Measurements. For oxygen evolution measurements, a solution

of the iridium complex (1 mL) was injected into a sealed 40 mL EPA vial containing 10 mL

CAN solution (~ 0.4 M) buffered in 1M HNO₃. The generated oxygen lead to an increase in

gas phase pressure, which was dynamically monitored via differential digital manometry as

described previously.⁸End points of the reactions were verified by gas chromatography and

corrected for nitrogen contamination.

DFT Calculations. All DFT calculations were performed with Gaussian 09 using B3LYP

functional and LANL2DZ basis set. Default parameters and thresholds were used for gradient

convergence. Geometry optimizations were carried out without additional constraints of solvent

environment. The optimized geometries of the complexes were visualized with Avogadro.³⁹

Crystal structure determinations. Suitable crystals of 8b, 8c, 12b and 16a were mounted in

air at ambient conditions and measured on an Oxford Diffraction SuperNova area-detector

diffractometer at T = 173(2) K using mirror optics monochromated Mo Kα radiation (λ =

0.71073 Å) and Al filtered.⁴⁰Data reduction was performed using the CrysAlisPro program.⁴¹

The intensities were corrected for Lorentz and polarization effects, and an absorption correction

based on the multi-scan method using SCALE3 ABSPACK in CrysAlisPro was applied. The

structures were solved by direct methods using SHELXT,⁴²and all non-hydrogen atoms were

refined anisotropically. All H-atoms were placed in geometrically calculated positions and

refined using a riding model with each H-atom assigned a fixed isotropic displacement

parameter (1.2Ueq of its parent atom, 1.5Ueq for the methyl groups). Structures were refined

on F²using full-matrix least-squares procedures. The weighting schemes were based on

counting statistics and included a factor to downweight the intense reflections. All calculations

were performed using the SHELXL-2014 program.⁴²

The crystal of 8b contains two independent molecules of the metal complex in the asymmetric

unit and one co-crystallized CH₂Cl₂molecule. The liberation of the butyl chain of one of the

two independent molecules was found to be quite large (probably affected by some dynamic

disorder that could not be modelled properly). A restraint on the most external C–C bond

27

10.1002/chem.201705619

Chemistry - A European Journal

distance was applied as well as a limitation of the anisotropic displacement parameter for the

two C atoms involved. The crystal of 8c contains two metal complex molecules and one co-

crystallized dichloromethane molecule in the asymmetric unit. Part of one octyl group is

disordered over two conformations. The co-crystallized dichloromethane molecule is

disordered over two sites. The geometries of the disordered moieties were restrained to be

equal. Its ADP’s were restrained by the SHELXL SIMU and DELU instructions. The crystal

12b contains two independent molecules of the complex in the asymmetric unit and co-

crystallized CH₂Cl₂(one molecule in general position, one disordered about an inversion

center). The crystal possesses pseudo-translational elements that would produce a C-1 type

structure (equivalent to a P-1 with halved volume). This would imply that the two complexes

are equivalent but their conformation clearly differ, especially in the Cp* group. Moreover, the

position of the clathrated solvent about the inversion center makes the double unit cell the best

suggestion (as also proved by the accurate analysis of the weaker reflections). A structure

solved in C-1 gave clearly worse refinement indices (R1 = 3.37%, instead of 2.18% of the P-1

solution) and larger atomic displacement parameters, especially for the Cp* atoms. Moreover,

the pseudo C translation would be violated by half of the measured reflections. For compound

16a, there are two solvent molecules in a crystal structure. Position of water molecule was

refined without doubts, but the second one was assumed a CH₂Cl₂molecule as it was used for

crystallization. The molecule is highly disordered and no hydrogen atoms were assigned.

Chlorine and carbon atoms were assigned according to the donor-acceptor adjacent of the

molecule. Further details are given in Tables S2–S5. Crystallographic data for all four structures

have been deposited with the Cambridge Crystallographic Data Centre (CCDC) as

supplementary publication numbers 1585829 (8b), 1585830 (8c), 1585831 (12b), and 1585832

(16a).

Acknowledgements

We acknowledge generous financial support from the European Research Council (CoG

615653) and from the Swiss National Science Foundation (200021_162868, R'equip projects

206021_128724 and 206021_170755) and the U.S. National Science Foundation (CHE-

1362629). C.A.S. thanks NSERC (Natural Science and Engineering Research Council of

Canada), the Water Sumner Foundation, and Queen’s University. We thank Albert Farre for

technical assistance and the group of Chemical Crystallography of the University of Bern for

the X-ray analysis of all reported structures.

28

10.1002/chem.201705619

Chemistry - A European Journal

References

[1] a) EIA, 2012 International Energy Outlook, 2012. b) J. T. Kiehl, K. E. Trenberth, Bull. Am.

Meteorol. Soc. 1997, 78, 197–208.

[2] a) R.E. Blankenship, Molecular Mechanisms of Photosynthesis. Blackwell Science

(Oxford), 2002; b) T. Gust, D. A. Moore, A. L. Moore, Acc. Chem. Res. 2009, 42, 1890–

1898; c) S. Berardi, S. Drouet, L. Francas, C. Gimbert-Surinach, M. Guttentag, C.

Richmond, T. Stoll, A. Llobet, Chem. Soc. Rev. 2014, 43, 7501–7519; for a recent

example, see: d) C. Liu, B. C. Colon, M. Ziesack, P. A. Silver, D. G. Nocera, Science

2016, 352, 1210–1213.

[3] a) N. D. McDaniel, S. Bernhard, Dalton Trans. 2010, 39, 10021–10030; b) S. Romain, L.

Vigara, A. Llobet, Acc. Chem. Res. 2009, 42, 1944–1953.

[4] A. W. Rutherford, A.Boussac, Science 2004, 303, 1782–1784.

[5] a) K. N. Ferreira, T. M. Iversion, K. Maghlaoui, J. Barber, S.Iwata, Science 2004, 303,

1831–1838; b) J. P. McEvoy, G. W. Brudvig, Chem. Rev. 2006, 106, 4455–4484; c) Y.

Umena, K. Kawakami, Y.-R. Shen, N. Kamiya, Nature 2011, 473, 55–60.

[6] For pioneering work using bimetallic catalysts: a) S. W. Gersten, J. G. Samuels, T. J. Meyer,

J. Am. Chem. Soc. 1982, 104, 3855–3864; b) J. A. Gilbert, D. S. Eggleston, W. R.

Murphy, D. A. Geselowitz, S. W. Gersten, D. J. Hodgson, T. J. Meyer, J. Am. Chem.

Soc. 1985, 107, 3855–3864; c) C. Sens, I. Romero, M. Rodríguez, A. Llobet, T. Parella,

J. Benet-Buchholz, J. Am. Chem. Soc. 2004, 126, 7798–7799; d) J. Limburg, J. S.

Vrettos, H. Chen, J. C. de Paula, R. H. Crabtree, G. W. Brudvig, J. Am. Chem. Soc.

2001, 123, 423–430.

[7] For pioneering work using tetrametallic catalytic systems, see: a) A. Sartorel, M. Carraro,

G. Scorrano, R. De Zorzi, S. Ceremia, N. D. McDaniel, S. Bernhard, M. Bonchio, J. Am.

Chem. Soc. 2005, 127, 12802–12803; b) A. Sartorel, P. Miró, E. Salvadori, S. Romain,

M. Carraro, G. Scorrano, M. Di Valentin, A. Llobet, C. Bo, M.Bonchio, J. Am. Chem.

Soc. 2009, 131, 16051–16053; c) Y. V. Geletii, B. Botar, P. Kögerler, D. A. Hillesheim,

D. G. Musaev, C. L. Hill, Angew. Chem. Int. Ed. 2008, 47, 3896–3899; d) F. Song, R.

More, M. Schilling, G. Smolentsev, N. Azzaroli, T. Fox, S. Luber, G. R. Patzke, J. Am.

Chem. Soc. 2017, 139, 14198–14208.

29

Article Doi

Optimization of Synthetically Versatile Pyridylidene Amide Ligands for Efficient Iridium-Catalyzed Water Oxidation

DOI: 10.1002/chem.201705619

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.201705619

Full text of DOI:10.1002/chem.201705619

Products guided by the article

R&D Labs maybe for 4589-12-2

Relevant to this article

Hot Product