University of Groningen Mechanisms in Ruthenium(II) … · 2017. 2. 10. · Mechanism of an diiron(III)bis‐phenolato complex catalyzed oxidation 67 O Fe Fe N N O N N N O N N N 2+

University of Groningen

Mechanisms in Ruthenium(II) photochemistry and Iron(III) catalyzed oxidationsUnjaroen, Duenpen

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:2017

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Unjaroen, D. (2017). Mechanisms in Ruthenium(II) photochemistry and Iron(III) catalyzed oxidations:Photochemical, Electrochemical and Spectroscopic studies. University of Groningen.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 20-04-2021

https://research.rug.nl/en/publications/mechanisms-in-rutheniumii-photochemistry-and-ironiii-catalyzed-oxidations(0567e5c0-4564-4f00-a7f9-b0c21459b98e).html

CHAPTER 6

Mechanism of the oxidation of benzyl alcohols catalyzed by a µ-oxo-diiron(III)bis-phenolato polypyridyl complex The µ‐oxo‐diiron(III) polypyridyl complex [(L)Fe(µ‐O)Fe(L)](ClO4)2 (1) (where HL is 2‐(((di(pyridin‐2‐

yl)methyl)(pyridin‐2‐ylmethyl)amino)methyl)phenol), in which each of the Fe(III) centers is

coordinated to a phenolato ligand, was applied over a decade ago to the selective oxidation of

benzyl alcohols to aldehydes with H2O2. However, in this chapter shows clearly that under catalytic

conditions with excess H2O2, rapid oxidation (< 5 s) and loss of the phenolato moiety occurs

resulting the formation of an N4 ligated Fe(III) complex. This complex forms a Fe(III)‐OOH species

initially before conversion to a relatively stable oxo‐bridged dinuclear Fe(III) complex that does not

bear a phenol moiety. The data show that although the phenolato moiety imparts interesting redox

properties on complex 1 it plays no apparent role in oxidation catalysis with H2O2.

D. Unjaroen, A. Draksharapu, S. K. Padamati, M. Swart, W. R. Browne, manuscript in preparation

Chapter 6

64

6.1 Introduction The redox chemistry of transition metal complexes is a central aspect in mechanisms for many

homogeneous and enzymatic catalytic reactions, with progress in both areas accelerated by the

focus on understanding and modelling the active site of metalloenzymes.1 In addition to the redox

chemistry of the metal centers in metalloenzymes, redox non‐innocent active organic moieties play

an important role, not least the phenolato moiety, such as in the enzyme galactose oxidase (GO).2

GO is a mononuclear copper enzyme that catalyzes the two‐electron oxidation of primary alcohols

to their corresponding aldehydes, with subsequent reduction of dioxygen to H2O2. The structure of

the catalytic site in the inactive form of GO shows that it contains two imidazole rings (histidine

residues) and two phenol moieties provided by tyrosine residues, as well as an acetato ligand bound

to the copper(II) ion.5a,3

The formal oxidation state of the active form of GO has been assigned as copper(III), however,

it is better described as a copper(II)‐phenoxyl radical considering its spectroscopic properties.6

Phenolato containing enzymes have received, as a consequence, considerable attention in recent

years in the context of non‐redox‐innocent ligands,4 as one‐electron oxidation of metal‐phenolato

complexes can result in the formation of either a higher valent metal species (Mn+1L‐) or ligand

radicals (MnL), which can be viewed as electronic isomers.5 Wieghardt et al.6 reported a stable

Fe(III)‐phenoxyl radical complex as early as 1993 and the number of small molecule metal‐phenoxyl

complexes have grown to include a variety of metals and ligand frameworks.7 Indeed, in 1998,

Wieghardt and Stack8 reported a functional model for GO that could oxidize alcohols to aldehydes

with high yield and selectivity.

Diiron complexes have been studied extensively towards modeling the activation of oxygen by

nonheme iron enzymes, such as methane monooxygenase (MMO), ribonucleotide reductase, and

fatty acid desaturases.9 In addition, several µ‐oxo‐diiron(III) complexes have also been reported as

catalysts for the oxidation of organic substrates. Palaniandavar et al.,10 reported the oxidation of

alkanes to alcohols with m‐chloroperbenzoic acid (m‐CPBA) catalyzed by a µ‐oxo‐bridged non‐

heme bis‐phenolate‐diiron(III) complex and Ligtenbarg et al.11 reported the synthesis and structural

characterization of a µ‐oxo‐diiron(III)bis‐phenolato polypyridyl complex (1), as well as

demonstrating its activity in the oxidation of benzylic alcohols to aldehydes with H2O2 as terminal

oxidant.

This latter system (1) showed a notable time dependence for the oxidation of benzyl alcohol and

in particular, the duration of an initial lag period was found to correspond to the loss of visible

absorption (assigned to a phenolate to iron(III) LMCT‐ligand to metal charge transfer band).

Furthermore, addition of a strong acid, i.e. CF3SO3H, resulted in a red shift in visible absorption and

a complete elimination of the lag period ascribed to breaking up of the dinuclear complex to a

catalytically active mononuclear species.11 These data it was proposed that break‐up of the

dinuclear complex to a mono‐nuclear complex occurred upon addition of acid with dissociation of

the phenolato moiety occurring prior to the onset of catalytic activity.

Although phenolato based ligands impart additional flexibility to metal complexes in terms of

redox chemistry (i.e. non‐innocent redox active ligands), these data raise pertinent questions as to

the role and indeed usefulness of such ligand fragments in oxidation catalysis.

In this chapter, we explore the time dependent spectroscopies of complex 1 under the reaction

conditions employed in benzyl alcohol oxidation. We demonstrate that the addition of acid does

not, in fact, lead to break‐up of the dinuclear structure but instead facilitates the reversible

oxidation of 1 by H2O2 to form formally Fe(III)Fe(IV) and Fe(IV)Fe(IV) species in which the phenolato

Mechanism of an diiron(III)bis‐phenolato complex catalyzed oxidation

65

moieties are better described as phenoxyl radicals. A side reaction is observed in which phenol‐

phenol C−C coupling (at the para‐positions) occurs leading to the formation of Fe(III)‐µ‐(4,4’‐

biphenol)‐Fe(III) species, manifested in the transient appearance of intense near‐UV and vis

absorption bands. UV‐vis and resonance Raman confirm that these species are identical to those

observed to form upon electrochemical oxidation, which was reported recently by our group.12

Figure 1. Structures of complexes 1 and 2.

Furthermore, we show that the lag phase observed earlier during the oxidation of alcohol is due to

the time is taken for partial degradation of the ligand to occur, specifically the removal of the

phenolato moiety following its oxidation to a phenoxyl radical, in the presence of excess H2O2 and

water. Furthermore, phenol dimerization through radical C−C coupling is blocked by tertbutyl

groups in the analogous complex 2, addition of excess H2O2 and water results in essentially the

same spectroscopic changes. These data lead to the conclusion that in fact, the catalytically active

species is generated as a result of ligand degradation to yield an N4 ligand based complex similar

to ([Fe(MeN3Py)(CH3CN)]2+ (where MeN3Py = 1,1‐di(pyridin‐2‐yl)‐N‐methyl‐N‐(pyridin‐2‐

ylmethyl)methanamine).13 Indeed, time resolved spectroscopy shows that such species13b form

rapidly under reaction conditions.

The data hold considerable implications for the use of phenoxyl moieties in ligand design

strategies in homogeneous oxidation catalysis and highlight the challenge in disentangling the

complex spectroscopy of such systems to enable elucidation of the precise nature of the species

responsible for catalytic oxidation.

6.2 Results The dinuclear phenolato complexes 1 and 2 were available from chapter 5,12 and the single crystal

X‐ray structure of 1 was reported earlier by Ligtenbarg et al.11 The complexes were characterized

earlier in the solid state by FTIR, Raman spectroscopy and in solution by 1H NMR, and UV‐vis

absorption spectroscopy, ESI mass, cyclic voltammetry, UV‐vis absorption, Raman, and

spectroelectrochemistry.12 Briefly, the dinuclear structures of complexes 1 and 2 in the solid state

are retained upon dissolution in dichloromethane, acetone, and acetonitrile. The µ‐oxo bridge

between the Fe(III) centers is stabilized by strong antiferromagnetic coupling, which is manifested

in broadened 1H NMR signals in the range of ‐5 to 40 ppm in CD3CN and in EPR silence (X‐band 77

K).12 Both complexes show strong visible absorption (nm4.1×103 M‐1 cm‐1) and more intense

near‐UV absorption bands assigned to ligand to metal charge transfer (LMCT) transitions involving

both phenolato to Fe(III)14 as well as pyridyl to Fe(III) charge transfer, based on the resonance

Raman spectra at exc 532 and 355 nm, and TDDFT.12 The absorption band at 343 nm is assigned to

a mixture of phenolato‐to‐Fe(III) and oxo‐to‐Fe(III) charge‐transfer transitions.15 Notably, the

absorption spectra in acetonitrile are not concentration dependent (between 3.9 µM and 1 mM),

Chapter 6

66

in agreement with ESI mass data, which indicates that dissociation to a mononuclear species does

not occur to a significant extent.12

6.2.1 Protonation of 1 and 2.

The UV‐vis absorption spectra of 1 and 2, both in acetonitrile and in acetone, undergo a substantial

bathochromic shift upon addition of TfOH (Figure 2). Earlier these changes were interpreted as

evidence for the formation of mononuclear complexes through protonation of the ‐oxo bridge.11 However, several observations confirm that this is not the case. The absorption bands of 1 shift to

347 nm, 410 nm, 605 nm (shoulder) and 804 nm (Figure 2a, red line) upon addition of TfOH,

however, the extent of the shift is less pronounced with weaker acids such as acetic and

trichloroacetic acid (Figure 2c).

Figure 2. UV‐vis absorption spectrum of (a) 1 in acetonitrile (black line), after addition of 1 equiv. of

TfOH, (red line) and subsequent addition of H2O (0.36.6 vol%). (b) 2 in acetonitrile (0.2 mM),

addition of 1 equiv. of TfOH, and subsequently addition of water. (c) 1 in acetonitrile 0.25 mM with

various acids (45 equiv. of triflic acid, 40 equiv. of trichloroacetic acid and 70 equiv. of acetic acid).

Furthermore, successive addition of acid equivalents leads to changes in the absorption spectrum

without maintenance of isosbestic points, indicative of multiple protonation steps. Indeed the

resonance Raman spectrum of 1 under acidic conditions at 1064 nm is different to that at 785 nm

in regard to the Raman shift of the phenol C−O stretch band, confirming the presence of several

distinct but structurally similar species which exhibit different degrees of red shift of the lowest

absorption bands.


67

OFeFe

N

NO

N

NNO

N

NN

2+

1

OFeFe

N

NOH

N

NNO

N

NN

3+

H1+

+H+O

FeFe

N

NOH

N

NNOH

N

NN

4+

H212+

+H+

Scheme 1. Propose the protonation forms of 1.

DFT calculationsi indicate that sequential protonation of phenolato oxygens is thermodynamically

more favorable than protonation of the µ‐oxo bridge as well as the dissociation and protonation of

a pyridyl moiety. Comparison of resonantly enhanced Raman spectra (at exc 785 nm and 1064 nm)

of 1 with TfOH with the Raman spectrum of 1 at 1064 nm shows that the major differences are to

the Fe‐O(phenol) and C−O(phenol) stretching modes in the 1400 cm‐1 and 600 cm‐1 regions, with

most other (i.e. pyridyl) bands remaining relatively unchanged by addition of acid (Figure 3b).

Furthermore, the resonance Raman spectra (exc 785 nm, Figure 3a) of Hn1n+ are essentially

independent of the acid used and the solvent (acetone vs. acetonitrile) with minor differences

centered around the phenolato C−O stretch and in the low (ca. 600 cm‐1) wavenumber region (785

nm is resonant with the absorption spectra of several of the species formed).

These data indicate that neither the conjugate base of the acids nor solvent, coordinate to the

Fe(III) centers under acidic conditions. Analysis by Evan’s NMR method indicated that solutions of

1 (µeff = 2.7 B.M.) undergo an increase in paramagnetic character in the presence of TfOH (µeff = 6.3

B.M). However, the formation of mononuclear species is unlikely since, as for 1, the UV‐vis

absorption spectra of Hn1n+ show negligible dependence on concentration between 4 µM and 1

mM, and although addition of acid results in broadening of the signals in the 1H NMR spectrum of

1 in acetonitrile‐d3 and in acetone‐d6 between ‐10 to 120 ppm (Appendix A 3.4), only a weak signal

at g = 4.3 is observed by X‐band EPR spectroscopy at 77 K. These data indicate that the

antiferromagnetic coupling between the Fe(III) centers, and especially the spin state, is affected by

protonation, but that dissociation to mononuclear iron(III) complexes to any significant extent can

be excluded.

i The DFT calculation was carried out by Prof. Marcel Swart, Universitat de Girona

Chapter 6

68

Figure 3. Raman spectra of 1 in acetonitrile with 1 equiv. of TfOH (a) exc 785 nm; 1 (9 mM, gray

line) and with TfOH (blue line). (b) exc 1064 nm; saturated solution (black line) and 1 (1 mM) with

TfOH (red line). *distortion due to imperfect subtraction of solvent spectral feature

Although, addition of 1 equiv. of TfOH to 1 in acetonitrile results in a large red‐shift of the longest

wavelength absorption band from 540 nm to ca. 808 nm, subsequent addition of increasing

amounts of water results in a progressive blue shift of the visible absorption bands of 1 (Figure 2a).

The absence of an isosbestic point as the spectrum returns towards its original state confirms that

several species are formed upon protonation and is consistent with the difference observed

between the resonance Raman spectra recorded at 785 and 1064 nm (vide supra). The recovery of

1 upon addition of water (vide supra), due to the levelling effect, is confirmed by the recovery of

the resonantly enhanced Raman (exc 355 nm) bands of 1 at 411 cm‐1 and 431 cm‐1 which

disappeared upon addition of acid, and reappeared upon addition of (6.6 vol%) water (Figure 4).

The loss of these bands upon addition of acid despite the minimal change in absorbance at 355 nm

indicates that protonation affects the moieties associated with these transitions, i.e. the phenolato

groups.

The base peak in ESI mass spectrum of 1 in CH3CN is at m/z 445 ([(L1)Fe(µ‐O)Fe(L1)]2+‐(ClO4)2).12

Addition of TfOH results in the observation of only mononuclear species, with a base peak at m/z

454 ([L1FeIII‐OH]+) and additional signals at m/z 227, 218, 586 assigned to [LFe(III)‐OH2]2+, [LFe]2+,

and [LFe‐OTf]+, respectively. The observation of [LFe]2+ is consistent with the presence of an

unstable dinuclear complex in solution and hence the observation of ions consistent with

mononuclear complexes does not, a priori, confirm their presence in solution as the major species.

Notably a significant signal consistent with either low or high spin mononuclear Fe(III) complex is

absent in the EPR spectrum of 1 upon addition of acid and/or water. As with rR and UV‐vis


69

absorption spectroscopy, subsequent addition of 3 vol% water results in the reappearance of the

signal at m/z 445.

Figure 4. rRaman spectra of 1 in acetonitrile (0.3 mM, black line) at exc 355 nm, with 1 equiv. of

TfOH (red line) and after subsequent addition of water (6.6 vol%, gray line).

6.2.2 Reaction of 1 with near stoichiometric H2O2 in the absence and presence of H2O and acid.

Addition of between 0.5‐3.0 equiv. H2O2 (0.66 vol% in water) to 1 in CH3CN does not affect its UV‐

vis absorption spectrum, however, in the presence of TfOH (2 equiv.) the formation and subsequent

decay of the intense absorption bands at 460 nm and ca. 850 nm is observed (Figure 5).

Figure 5. (a) UV‐vis absorption spectra of 1 in acetonitrile (black line) after addition of 2 equiv. of

TfOH (red line) and 2 equiv. of H2O2 (in H2O, blue line), (b) time dependence of absorbance at 460

nm and 850 nm.

Addition of 0.5‐3.0 equiv. H2O2 (in CH3CN), resulted in similar changes, however, the band at 460

nm appeared immediately and with a much higher maximum absorbance, followed by its decay

concomitant with the appearance of new bands at 780 nm and 850 nm (which decayed more

slowly). Addition of water together with H2O2 is necessary to observe the intense absorption band

at 460 nm. However, if water is added in sufficient excess the equilibrium shifts from H1+ to 1 (c.f.

leveling effect, Figure 6) and hence the rapid oxidation of 1 does not proceed.

Chapter 6

70

Figure 6. UV‐vis absorption spectra of 1 in acetonitrile (a) after addition of 2 equiv. of TfOH and 2

equiv. of H2O2 (in CH3CN), (c) addition of water 10 µL after TfOH, (b) and (d) time dependence of

absorption at selected wavelengths at 460 nm and 850 nm.

The transient concentration of the intermediate species responsible for the bands at 460 nm and

850 nm was maximized with 2 equiv. of TfOH, 0.3 vol% of water and 2 equiv. of H2O2 (Figure 7a).

The decay of the band at 460 nm proceeds concomitant with the increase in absorbance at 780 nm

and 850 nm (Figure 7b) with an isosbestic point maintained at 585 nm. The molar absorptivity of

the band at 460 nm, assuming 40‐50% conversion (vide infra), is comparable to that expected for

a phenoxyl radical. Importantly, however, the initial spectrum of 1 recovers subsequently upon

addition of 6‐10 vol% water together with the reappearance of a base peak at m/z 445 in the ESI

mass spectrum of the reaction mixture. The band at 460 nm reappears upon subsequent addition

of a 2nd and 3rd equivalent of H2O2, although the maximum absorbance reached is less each time.

Furthermore, the recovery of the absorption spectrum of 1 upon subsequent addition of water is

incomplete.


71

Figure 7. (a) UV‐vis absorption spectra of 1 in acetonitrile (0.25 mM, black line) after addition of 2

equiv. of TfOH (red line), 5 µl of H2O (blue line) and 2 equiv. of H2O2. (b) Time dependence of

absorbance at 460 nm and 850 nm following addition of 2 equiv. of H2O2 (at 2 s) to 1 in CH3CN (0.25

mM), (b) Time dependence of Raman band intensities at 1617 cm‐1 (at exc 473 nm) and 1608 cm‐1

(at exc 785 nm).

Despite the appearance of absorption bands typical of phenoxyl radicals, the only signal observed

by X‐band EPR spectroscopy (77 K) at any point (i.e. with samples flash frozen at specific time

points) was a weak band at g = 4.3, which had appeared already upon addition of TfOH to 1 and

was essentially unaffected by addition of water and H2O2 in CH3CN. Hence, all major species present

are EPR silent at 77 K indicating that the phenoxyl radicals are spin coupled with the 3d‐electrons

of the Fe(III) center (vide infra).

The resonance Raman spectra of the phenoxyl radicals reported earlier show characteristic bands

at 1500 cm‐1 and 1600 cm‐1, assigned to the 7a (C−O stretching) and 8a (C−C stretching) mode

of the phenoxyl radical, respectively.16 The species formed upon reaction of 1 with H2O2 were

characterized by time‐dependent resonance Raman spectroscopy at 473 and 785 nm. At exc 473

nm, the Raman spectra show the appearance and subsequent disappearance of strongly resonance

enhanced bands at 1481 cm‐1, 1555 cm‐1, and 1617 cm‐1 and at exc 785 nm, at 1264 cm‐1, 1344 cm‐

1, 1396 cm‐1, 1522 cm‐1 and 1608 cm‐1 (Figure 9).

Chapter 6

72

Figure 8. Raman spectra of 1 in acetonitrile (0.25 mM) at exc 1064 nm with 1 equiv. of TfOH (dark

blue) and after 1 equiv. of H2O2. * Solvent bands.

The time dependence of the change in intensity of the Raman bands at exc 473 nm and exc 785

nm (w.r.t. solvent bands) tracks well the increase and decrease in absorbance at 460 nm and 850

nm, respectively (Figure 7c). Notable, however, the time‐dependent Raman spectra show that the

species absorbing at 460 nm reaches a maximum concentration immediately followed by a slow

decay, whereas the species absorbing at 850 nm reaches a maximum more slowly and then decays.

The time dependence of the formation and decay of this latter species is less apparent from the

absorbance at 850 nm due to overlap with the absorbance of H1+ which decreases concomitant

with the increase in absorbance at 850 nm, and vice versa.

Excitation at λexc 1064 nm is preresonant with both H1+ and the species generated by addition

of H2O2 which absorbs at 850 nm (Figure 8). Bands characteristic of phenoxyl radical species are

not observed but the band at 1330 cm‐1 corresponds to the addition and loss of H2O2. The changes

observed in the Raman spectrum of 1 with TfOH and H2O2 indicate that at most 50% conversion is

observed and that 1 recovers partially after the H2O2 is consumed.


73

OFeFe

N

NO

N

NNO

N

NN

2+

1

OFeFe

N

NOH

N

NNO

N

NN

3+

H1+

+H+

O

Fe

Fe

N

NO

NN N

O

N NN

O

C-C coupled bis phenol complex

OFeFe

N

NOH

N

NNOH

N

NN

4+

H212+

+H+

+ H2O2

+ H2O

OFeFe

N

NO

N

NNO

N

NNH

3+

OOH

H

Scheme 2. Protonation of 1 and subsequent oxidation and C−C coupling of the phenolato moieties.

6.2.3 Identification of species formed upon reaction of 1 with stoichiometric H2O2.

Comparison of the Raman spectra at exc 473 nm and exc 785 nm during the reaction of 1 with a

few equiv. of H2O2 with the products of electrochemical and chemical {[(NH4)2Ce(NO3)6] (CAN)}

oxidation of the same complex, reported earlier,12 allow for definitive assignment of the species

formed. Oxidation of 1 with CAN, i.e. without TfOH or water present, results in the immediate

appearance of an absorption band at 465 nm, which decays over time concomitant with the

appearance of bands at 780 nm and 850 nm. The changes in the absorption spectrum of 1 are

essentially the same as those observed upon addition of near stoichiometric H2O2 under acidic

conditions and, conclusively, the Raman spectra recorded at exc 473 nm and 785 nm are identical

to those obtained with TfOH/H2O/H2O2 (Figure 9).

Chapter 6

74

Figure 9. Comparison the Raman spectra at (a) exc 437 nm and (b) exc 785 nm of 1 in acetonitrile

after addition of 1 equiv. of TfOH and 1 equiv. of H2O2 (black line) to the spectra obtained12 upon

addition of 1 equiv. of CAN to 1 in acetonitrile (red line).

6.2.4 Reaction of 1 with excess H2O2 in the absence and presence of water and acid.

The reaction of 1 in acetonitrile, with and without H2O, with 100 equiv. of H2O2 was monitored over

time by UV‐vis absorption spectroscopy. A decrease in the absorbance of 1 at 535 nm was observed

only after a lag phase of ca. 60 min, which is consistent with the observation made earlier that the

onset of catalytic activity occurred concomitant with the change in color of the reaction mixture

(after ca. 1.5 h) from purple to yellow. The disappearance of the visible absorption bands indicates

loss of the phenolato moiety from the complex. In the present study, the lag phase was found to

eliminated by addition of water (Figure 10), however, the rate of loss of absorbance was reduced

also.

Figure 10. Time dependence of absorption at selected wavelengths at 535 nm after addition of 100

equiv. of H2O2 into 1 in acetonitrile with (red line, 5 vol%) and without H2O (black line).


75

Closer examination of the initial changes that occur upon addition of 100 equiv. of H2O2 to 1 in

acetonitrile with 1 equiv. of TfOH shows a transient increase in absorbance at 460 nm and 730 nm,

both appearing and decaying within 4 s. Subsequently, a band at 556 nm appears reaching a

maximum at 6 s and decaying again over 1.5 min to give a yellow solution (Figure 11a/b). A band at

420 nm appeared over 1 to 2 h (vide infra). The intermediate species formed, which absorbs at 556

nm, was characterized by Raman spectroscopy at λexc 532 nm. Addition of 100 equiv. of H2O2 to 1

in acetonitrile with TfOH (1 equiv.) resulted in the appearance of bands at 801, and 624 cm−1 (Figure

11d). The band at 802 cm−1 was assigned, tentatively, to an O−O stretch and the band at 624 cm−1

to the Fe−O stretch of a [(R‐N3Py)Fe(III)(X)(OOH)]2+ species (where X = solvent, R = H).13b

Figure 11. UV‐vis absorption spectra of 1 in acetonitrile (0.25 mM) with 3 equiv. of TfOH (red line)

(a) upon addition of 100 equiv. of H2O2, (b) before (black) and after (red) addition of TfOH and after

addition of 100 equiv. of H2O2; 2 min (gray), 1 h (green) and 2 h (yellow). (c) Time dependence of

absorption at selected wavelengths at 420 and 556 nm after addition of 100 equiv. of H2O2. (d)

rRaman spectra of 1 in acetonitrile at exc 532 nm with TfOH (black), after addition of 100 equiv. of

H2O2; 4 s (blue), 8s (red) and 12s (yellow).

Additionally, the yellow species that forms after decay of the band at 556 nm showed resonance

enhanced Raman bands at exc 355 nm showed bands at 1612, 1569, 1275, 1158, 1057 and 1032

cm‐1 that are typical of pyridyl modes (Figure 12). The bands at 1291, 1275 and 1210 cm‐1 are

assigned to the alkyl amine backbone of the ligand and the band at 1004 cm‐1 is assigned to a phenyl

ring mode by comparison to the resonance Raman spectrum of [FeII(N4Py)(CH3CN)](ClO4)2.17

Chapter 6

76

Figure 12. Raman spectrum of the yellow solution at exc 355 nm obtained 5 min after addition of

100 equiv. of H2O2 to 1 with 3 equiv. TfOH in acetonitrile. *Distortion due to imperfect solvent

subtraction.

Raman spectra recorded with a time resolution of 200 ms (Figure 13) enabled identification of the

species responsible for the transient absorption at 730 nm. The spectrum of the intermediate is

similar to those obtained with 1‐3 equiv. H2O2, i.e. consistent with a phenoxyl radical species, but

there are clear differences that confirm that it is not the same species (vide supra).

Figure 13. (a) rRaman spectra of 1 in acetonitrile at exc 785 nm with 3 equiv. TfOH (a) before, and

0.5 s (blue) and 2 s (red) after addition of 1000 equiv. of H2O2 and (b) time dependence of signal at

1598 cm‐1 (intermediate, red) and at 870 cm‐1 (H2O2, black).

6.2.5 Spectroscopy of 1 during the catalyzed oxidation of 1‐phenylethanol.

Although the formation of phenxoyl radicals from 1 upon reaction with near stoichiometric H2O2 is

evident, the relevance of such species under conditions used,11 i.e. with up to 100 equiv. of H2O2,

in the oxidation of benzyl alcohols catalyzed by 1 is, a priori, uncertain. The catalytic oxidation of 1‐

phenylethanol in acetonitrile under the conditions reported earlier by Ligtenbarg et al.11 was

monitored by UV‐vis absorption and Raman spectroscopy (Figure 14). Addition of 100 equiv. of

H2O2 to 1 (0.88 mM) and 1‐phenylethanol (1000 equiv.) results in a decay of the absorption band

at 540 nm within 1 h following a lag phase, the duration of which was increased compared to that

observed in the absence of substrate (vide supra). Raman spectroscopy shows that the

consumption of substrate and H2O2 commences only after the end of the lag period. 1H NMR

spectra of the crude reaction in acetonitrile‐d3 showed acetophenone in ~4% yield (40% efficiency

in oxidant).


77

Figure 14. (a) UV‐vis absorption of 1 in acetonitrile (0.88 mM) during catalyzed the oxidation of 1‐

phenylethanol (880 mM) with H2O2 (88 mM). (b) Time dependence of the absorbance at 540 nm.

(c) Raman spectra (exc 785 nm) before and after 160 min and (d) intensity of the bands due to

phenyl‐1‐ethanol, acetophenone and H2O2 over time (with 1,2‐dichlorobenzene as internal

reference).

The presence of 1 equiv of TfOH, prior to addition of H2O2 resulted in dramatic differences in the

time dependence of the changes in both UV‐vis absorption and Raman spectra with rapid loss of

the absorbance of H1+ at ca. 800 nm and the immediate appearance over 10 s of an intermediate

(i.e. FeIII−OOH), which absorbs at 560 nm, followed by its subsequent decay more slowly that in the

absence of substrate (Figure 15a/b). The X‐band EPR spectra (at 77 K) of 1 in acetonitrile with TfOH

and 1‐phenylethanol after addition of 100 equiv. of H2O2 (i.e. at the point a maximum absorbance

at 560 nm was reached) showed a relatively strong signal at g = 4.25, characteristic for a high‐spin

iron(III) species together with a signals at g = 2.12 and 1.96 that are characteristic of a low‐spin (S

= ½) FeIII−OOH species (Figure 15c). After warming the intermediate FeIII−OOH species to room

temperature and subsequently freezing to 77 K again, the signals of the S = ½ species disappeared

together with the decrease in the intensity of the signal of high‐spin iron(III) species. The X‐band

EPR spectrum of the yellow species that formed after the absorption band at 560 nm had

disappeared exhibited only a weak signal at g = 4.25, indicating that mononuclear high‐spin iron(III)

species are intermediates also leading ultimately to formation of anti‐ferromagnetically coupled

dinuclear complexes.13b

Chapter 6

78

Figure 15. (a) UV‐vis absorption of 1 in acetonitrile (0.88 mM) with 1000 equiv. 1‐phenylethanol

and 1 equiv. of TfOH before (thick red line) and after addition of 100 equiv. H2O2. (b) Time

dependence of absorbance at 560 nm. (c) X‐band EPR spectra at 77 K of 1 in acetonitrile with TfOH

and 1‐phenylethanol; 14 s after addition of 100 equiv. of H2O2 (blue), after warm sample to room

temperature (black), 5 min after addition of 100 equiv. of H2O2 (gray).

6.2.6 Reaction of 2 with H2O2 and catalytic oxidation of 1‐phenylethanol.

The structure of 2, and specifically the tert‐butyl groups, precludes oxidative C−C coupling of the

phenols. Addition of 2 equiv. H2O2 has little effect on the UV‐vis absorption spectrum of 2, even

with TfOH, when excess water (0.33 vol%) was present (vide supra), however, with 100 equiv. H2O2

a steady decrease in visible absorbance is observed. Under more acidic conditions (i.e. with less

water present) addition of 100 equiv. of H2O2 to 2 with 1 equiv of TfOH resulted in a rapid decrease

in NIR absorbance within 8 s with a weaker absorption at ca. 550 nm appearing and persisting for

ca. 20 s (Figure 16a/b).


79

Figure 16. (a) UV‐vis absorption spectrum of 2 (0.88 mM) in acetonitrile with 1 equiv. of TfOH (black)

followed by addition of 100 equiv. of H2O2 (spectra are at 2 s intervals). (b) Absorbance at 750 and

550 nm over time. (c) EPR spectra (77 K, X‐band) flash frozen at 0 s (black), 10 s (red), 40 s (blue)

and 74 s (green). (d) rRaman spectra at 0 s (red), 10 s (orange), 20 s (black), 30 s (blue) and 40 s

(green) after addition of H2O2 (100 equiv.).

EPR and Raman spectroscopy confirms that H2+ undergoes rapid oxidation and loss of the

phenolato moiety (the bands at 837, 602 and 568 cm‐1 disappear within 10 s) followed by the

transient formation of an Fe(III)‐OOH species, manifested in the characteristic bands at 801 and

622 cm‐1 and a low spin (S = ½) EPR signal (Figure 16c/d). These changes coincide with the loss of

H2O2 (867 cm‐1) over 1 min. The presence of substrate (1‐phenylethanol) in excess results in an

increase in the maximum concentration of the Fe(III)‐OOH species (max 550 nm) and its persistence

over 15 min.

The loss of the phenolato moiety in 2 is consistent with spectral changes observed for 1 under

catalytic conditions. Resonance Raman spectroscopy at 785 nm confirms that in the presence of

excess H2O2 and water, oxidation phenoxyl radical is followed by a rapid loss, presumably by a

nucleophillic attack of water and subsequent disociation of the phenolato unit, to leave an N3Py

type ligand. Importantly, the C‐C coupling reaction observed with 1 upon (electro)chemical

oxidation and upon addition of stoichiometric amounts of H2O2, is not observed with excess H2O2.

The complexes formed by loss of the phenoxyl units from 1 and 2 are therefore identical and, given

that it forms within a few seconds of addition of H2O2 addition, the catalytic activity of both

complexes should be, and indeed appears to be, the same.

Chapter 6

80

Scheme 3. Summary of processes observed for 1 in solution upon oxidation with stoichiometric

H2O2 and with excess H2O2.


81

6.3 Summary The oxidative chemistry of 1 and 2 is summarized in Scheme 3. Initially, oxiation either

electrochemical or chemical (e.g., Ce(IV), R‐OOH, H2O2 etc.) results in the generation of phenoxyl

radicals which undergo essentially instantaneous C−C coupling in the case of 1 and, in the case of

both 1 and 2, nucelophillic attack by water. The C−C coupled product (3) undergoes oxidation even

more rapidly that 1, due to its lower oxidation potential to form 32+, and equivalent oligomeric

species. In the presence of excess H2O2, however, rapid cleavage of the phenoxyl unit from the

Fe(III) bond L1 takes place to form an N3Py‐Fe(III) complex, analogous to catalytically active

complexes reported earlier. Indeed under catalytic conditions, this series of reactions occurs so

rapidly (with the mixing time) that for 1, C−C coupling is not observed. The role of acid in the

chemistry of 1 was suggested earlier11 to break up the dinuclear structure of the complex, however,

both spectroscopic and DFT studies indicate that this is not the case and instead its role is to

facilitate oxidation with H2O2. It should be noted, however, that acid is not essential and only

accelerates an already spontaneous reaction, allowing intermediates to be observed. Complex 2

bears tert‐butyl groups that are effective in preventing C−C coupling and reducing the rate of

oxidation of the phenolato moieties by H2O2, however, ultimately, phenolato centred oxidation

occurs followed by loss of the phenolato moiety.

6.4 Conclusions The design and synthesis of bio‐inspired homogeneous catalysts benefit from the breadth of

reactivity presented by nature’s enzymes and its remarkable ability to couple the redox chemistry

of metals with that of organic ligands – so called redox non‐innocent ligands. The porphyrin and

phenolato based enzymes stand out in this regard, however, efforts to mimic such redox versatility

are hampered by the lack of the protective protein environment of the enzyme in synthetic

systems. In the present contribution, we have focused on the redox chemistry and spectroscopy of

a phenolato based catalyst applied earlier by Feringa and coworkers in benzyl‐alcohol oxidation

with H2O2. We have shown that although the complexes generate spectroscopically interesting

species upon oxidation, and that the initial oxidation is essentially phenolato centered and only

formally an Fe(III)/Fe(IV) redox couple. These species are at most side reactions or intermediates

on the way towards loss of the phenolato unit and formation of the catalytically active species. In

a broader context, the present study highlights the challenge faced in studying bioinspired catalysts

bearing phenolato moieties and especially where these moieties are not only ‘redox non‐innocent’

in the classic sense but in fact the primary redox active groups. Furthermore, the use of tert‐butyl

groups, although effective in preventing C−C coupling reactions, does not prevent reaction of the

phenolato unit with small nucleophiles and oxidatively induced cleavage of the benzylic C−N bond.

Hence, future efforts towards ligand design should take these aspects into account in order to

reliably harness the potential the redox active phenol units present in oxidation catalysis.

6.5 Acknowledgements Prof. M. Swart (Universitat de Girona) is acknowledged for provision of DFT data.

Chapter 6

82

6.6 References and notes

(1) Holm, R. H.; Kennepohl, P.; Solomon, E. I. Chem. Rev. 1996, 96, 2239−2314.

(2) (a) Whittaker, J. W. Chemical reviews 2003, 103, 2347−2364; (b) Stubbe, J.; van der Donk, W. A. Chem.

Rev. 1998, 98, 705−762.

(3) Halfen, J. A.; Jazdzewski, B. A.; Mahapatra, S.; Berreau, L. M.; Wilkinson, E. C.; Que, L.; Tolman, W. B. J.

Am. Chem. Soc. 1997, 119, 8217−8227.

(4) Butin, K. P.; Beloglazkina, E. K.; Zyk, N. V. Russ. Chem. Rev. 2005, 74, 531−553.

(5) (a) Shimazaki, Y. Adv. Mater. Phys. and Chem. 2013, 3, 60−71; (b) Lyons, C. T.; Stack, T. D., Coord. Chem.

Rev. 2013, 257, 528−540; (c) Shimazaki, Y.; Yamauchi, O., Inorg. Phys. Theor. and Anal. 2011, 50,

383−394.

(6) Hockertz, J.; Steenken, S.; Wieghardt, K.; Hildebrandt, P. J. Am. Chem. Soc. 1993, 115, 11222−11230.

(7) Chaudhuri, P.; Wieghardt, K., Phenoxyl radical complexes. Progress in Inorganic Chemistry 2001, 50,

151−216.

(8) (a) Wang, Y.; DuBois, J. L.; Hedman, B.; Hodgson, K. O.; Stack, T. Science 1998, 279, 537−540; (b)

Chaudhuri, P.; Hess, M.; Flörke, U.; Wieghardt, K. Angew. Chem. Int. Ed. 1998, 37, 2217−2220.

(9) (a) Stoian, S. A.; Xue, G.; Bominaar, E. L.; Que, L., Jr.; Munck, E. J. Am. Chem. Soc. 2014, 136, 1545−1558;

(b) Xue, G.; Geng, C.; Ye, S.; Fiedler, A. T.; Neese, F.; Que, L., Jr. Inorg. Chem. 2013, 52, 3976‐3984; (c)

Do, L. H.; Xue, G.; Que, L., Jr.; Lippard, S. J. Inorg. Chem. 2012, 51, 2393−2402; (d) Cranswick, M. A.;

Meier, K. K.; Shan, X.; Stubna, A.; Kaizer, J.; Mehn, M. P.; Munck, E.; Que, L., Jr. Inorg. Chem. 2012, 51,

10417−10426; (e) De Hont, R. F.; Xue, G.; Hendrich, M. P.; Que, L., Jr.; Bominaar, E. L.; Munck, E. Inorg.

Chem. 2010, 49, 8310−8322; (f) Stubna, A.; Jo, D.‐H.; Costas, M.; Brenessel, W. W.; Andres, H.; Bominaar,

E. L.; Münck, E.; Que, L. Inorg. chem. 2004, 43, 3067−3079.

(10) Mayilmurugan, R.; Stoeckli‐Evans, H.; Suresh, E.; Palaniandavar, M., Dalton trans. 2009, 26, 5101−5114.

(11) Ligtenbarg, A. G. J.; Oosting, P.; Roelfes, G.; Crois, R. M. L.; Hage, R.; Feringa, B. L.; Lutz, M.; Spek, A. L.,

Chem. Commum. 2001, 4, 385−386.

(12) Unjaroen, D.; Swart, M.; Browne. W. R. Inorg. Chem. 2017, 56, 470−479. (13) (a) Klopstra, M.; Roelfes, G.; Hage, R.; Kellogg, R. M.; Feringa, B. L. Eur J Inorg Chem. 2004, 846−856. (b)

Padamati, S. K.; Draksharapu, A.; Unjaroen, D.; Browne, W. R., Inorg. Chem. 2016, 55, 4211−4222.

(14) Ito, S.; Suzuki, M.; Kobayashi, T.; Itoh, H.; Harada, A.; Ohba, S.; Nishida, Y. J. Chem. Soc., Dalton Trans.

1996, 12, 2579−2580.

(15) Yan, S.; Que, L.; Taylor, L. F.; Anderson, O. P. J. Am. Chem. Soc. 1988, 110, 5222−5224.

(16) (a) Schnepf, R.; Sokolowski, A.; Müller, J.; Bachler, V.; Wieghardt, K.; Hildebrandt, P. J. Am. Chem. Soc.

1998, 120, 2352−2364. (b) Itoh, S.; Kumei, H.; Nagatomo, S.; Kitagawa, T.; Fukuzumi S. J. Am. Chem. Soc.

2001, 123, 2165−2175. (c) Lyons, C. T.; Stack, T. D. Coord. Chem. Rev. 2013, 257, 528−540.

(17) (a) Draksharapu, A.; Li, Q.; Logtenberg, H.; van den Berg, T. A.; Meetsma, A.; Killeen, J. S.; Feringa, B. L.;

Hage, R.; Roelfes, G.; Browne, W. R. Inorg. Chem. 2012, 51, 900–913.

Appendix A

Synthesis and Characterization of Ligands and Complexes

Appendix A

84

Commercially available chemicals were used as received without further purification, unless stated

otherwise.

Caution. Perchlorate salts of metal complexes incorporating organic ligands are potentially

explosive. These compounds should be prepared in small quantities and handled with suitable

protective safeguards.

A. 1 Synthesis and characterization of Fe(II) complexes (chapter 2)

Scheme A1. Synthesis of dppz ligand.

1,10‐phenanthroline‐5,6‐dione (4). Ice‐cold mixtures of a concentrate of H2SO4 (30 mL) and HNO3

(15 mL) were added to 1,10‐phenantroline (3.07 g, 17.05 mmol) and KBr (3.04 g, 25.59 mmol). The

mixture was heated at reflux for 3 h. The hot yellow solution was poured over 500 mL of ice water

and neutralized carefully with NaOH solution. The separated precipitated of 1,10‐phenanthroline‐

5,6‐dione was filtrated off, washed with distilled water and dried. The rest of the product was

extracted with CH2Cl2 (3×10 mL) from the water phase, and then the solvent was evaporated to

give the yellow solid (1.50 g, 42%). 1H NMR (400 MHz, CDCl3): 9.07 (dd, J = 4.7 Hz, 1.8 Hz, 2H), 8.46 (dd, J = 7.9 Hz, 1.8 Hz, 2H), 7.56 (dd, J = 7.8 Hz, 4.7Hz, 2H). 13C NMR (125 MHz, CDCl3): 178.7, 156.4, 152.9, 137.3, 128.1, 125.6.

Dipyrido[3,2‐a:2',3'‐c]phenazine (5). A mixture of 1,10‐phenanthroline‐5,6‐dione (1.32 g, 6.27

mmol) and 1,2‐phenylenediamine (0.81 g, 7.53 mmol) in ethanol (250 mL) were stirred at 50 °C for

2 h and then at room temperature overnight. The solvent was evaporated to obtain the cream

solid. The crude product was left to stand for 8 h, methanol‐water (10:90) was then added, and the

product was filtered and recrystallized from methanol to give a cream solid (1.27 g, 72%). 1H NMR

(400 MHz, CDCl3): 9.64 (dd, J = 8.1 Hz, 1.6 Hz, 2H), 9.27 (d, J = 3.2 Hz, 2H), 8.34 (dd, J = 6.5 Hz, 3.4 Hz, 2H) 7.92 (dd, J = 6.6 Hz, 3.4 Hz, 2H), 7.79 (q, J = 8.0 Hz, 4.4 Hz, 2H). 13C NMR (125 MHz, CDCl3):

152.5, 148.3, 142.5, 141.1, 133.8, 130.7, 129.5, 127.6, 124.2.

[Fe(py)4(NCS)2]. Pyridine (10 mL), a solution of hydrated ferrous perchlorate (9.0 g, 35.33 mmol) in

water (250 mL) and a pinch of ascorbic acid were added to a solution of potassium thiocyanate (4.0

g, 41.16 mmol) in water (400 mL) to precipitate the yellow form. The yellow precipitate was filtered

and washes with a mixture consisting of 65% ethanol, 25% water, and 10% pyridine and dried in

vacuo for 30 min.

[Fe(L)2(NCS)2]. The solution of Ligand (L) (0.5 mmol) in hot pyridine (20 mL) was added to a solution

of [Fe(py)4(NCS)2] (0.5 mmol) in hot pyridine (80 mL). After filtration, the filtrate was allowed to

Synthesis and characterization of ligands and complexes

85

stand for one month at room temperature to produce the complex as a dark violet solid. All

operations were conducted under a nitrogen atmosphere.

[Fe(phen)2(NCS)2] (1). Anal. Calc. for C26H16FeN6S2: C, 58.65; H, 3.03; N, 15.78% Found: C, 57.71; H,

2.98; N, 15.51.

[Fe(dppz)2(NCS)2]∙py (2). Anal. Calc. for C43H25FeN11S2: C, 63.31; H, 3.09; N, 18.98% Found: C, 63.01;

H, 3.02; N, 18.76.

A. 2 Synthesis and characterization of Ru(II) complexes (chapter 3 and 4)

Scheme A2. Synthesis of [Ru(CH3CN)(MeN4Py)]2+ and [Ru(CH3CN)(N4Py)]2+

[Ru(Cl)(MeN4Py)]+. A mixture of MeN4Py; 1,1‐di(pyridin‐2‐yl)‐N,N‐bis(pyridin‐2‐ylmethyl)ethan‐1‐

amine (100.1 mg, 0.26 mmol), ruthenium(III) chloride hydrate (RuCl3xH2O) (122.4 mg, 0.59 mmol),

and L‐ascorbic acid (92.4 mg, 0.52 mmol) were heated at reflux overnight in EtOH/H2O (10/15 mL)

and then cooled to room temperature. The solvent was removed in vacuum and the crude product

purified by the column chromatography on neutral alumina, eluting with CH3CN to yield

[Ru(Cl)(MeN4Py)](Cl). Saturated aqueous sodium tetrafluoroborate (NaBF4) was added to

[Ru(Cl)(MeN4Py)](Cl) in water resulting in precipitation of [Ru(Cl)(MeN4Py)](BF4) as a red solid

(107.5 mg, 74%).1H NMR (400 MHz, CD3CN)9.35 (d, J = 5.4 Hz, 2H), 9.04 (d, J = 5.4 Hz, 2H), 7.88 (t, J = 7.8 Hz, 2H), 7.64 (d, J = 8.0 Hz, 2H), 7.54 (t, J = 7.7 Hz, 2H), 7.29 (t, J = 6.4 Hz, 2H), 7.24 (t, J =

6.4 Hz, 2H), 6.98 (d, J = 8.0 Hz, 2H), 4.15 (qAB, J = 18.6 Hz, 4H), 2.25 (s, 3H). ESI mass: m/z 518 [M‐

BF4]+.

[Ru(Cl)(N4Py)]+. Following the procedure as for [Ru(Cl)(MeN4Py)]+ complex, N4Py; 1,1‐di(pyridin‐2‐

yl)‐N,N‐bis(pyridin‐2‐ylmethyl)methanamine (296 mg, 0.80 mmol), RuCl3xH2O (376 mg, 1.81

mmol), and L‐ascorbic acid (284 mg, 1.61 mmol) in EtOH/H2O (20/30 mL) yield [Ru(Cl)(N4Py)](Cl)

which was saturated aqueous potassium hexafluorophosphate (KPF6) resulting in precipitation of

[Ru(Cl)(N4Py)](PF6) as a red solid (164 mg, 31%). 1H NMR (400 MHz, CD3CN):9.38 (d, J = 1.4 Hz, 2H), 9.00 (d, J = 1.4 Hz, 2H), 7.82‐7.89 (m, J = 7.7 Hz, 4H), 7.56 (d, J = 7.8 Hz, 1.6 Hz, 2H), 7.32 (t, J =

Appendix A

86

6.3 Hz, 2H), 7.26 (t, J = 6.4 Hz, 2H,), 7.01 (t, J = 7.9 Hz, 2H), 6.39 (s, 1H), 4.30 (qAB, JAB = 17.6 Hz, 4H),

(*4% of [Ru(CH3CN)(N4Py)](PF6) is observed by 1H NMR spectroscopy).

[Ru(CH3CN)(MeN4Py)](BF4)2. Addition of 1 mL of water into the solution of [Ru(Cl)(MeN4Py)](BF4)

(65 mg, 0.11 mmol) in CH3CN (9 mL) and the mixture was stirred overnight at 55 °C. The reaction

was cooled to room temperature and the solvent reduced in vacuum. A few drops of saturated

aqueous NaBF4 was subsequently added to a product in water yielded [Ru(CH3CN)(MeN4Py)](BF4)2

as a yellow solid (69.0 mg, 92%). 1H NMR (400 MHz, CD3CN) 8.94 (d, J = 5.4 Hz, 2H), 8.84 (d, J = 5.4 Hz, 2H), 7.97 (t, J = 7.7 Hz, 2H), 7.74 (d, J = 8.0 Hz, 2H), 7.64 (t, J = 7.8 Hz, 2H), 7.37 (t, J = 6.6 Hz,

2H), 7.29 (t, J = 6.6 Hz, 2H), 7.07 (d, J = 8.0 Hz, 2H), 4.36 (s, 4H), 2.73 (s, 3H), 2.32(s, 3H). ESI mass:

m/z 262 [M‐(BF4)2]2+. Anal. Calc. for RuC26H26B2F8N6: C, 44.79; N, 12.06; H, 3.75. Found: C, 43.31; N,

11.75; H, 3.76%.

[Ru(CH3CN)(N4Py)](PF6)2. Following the procedure as for [Ru(CH3CN)(MeN4Py)]2+. [RuCl(N4Py)](PF6)

(152 mg, 0.23 mmol) in CH3CN (30 mL) and water (3 mL) were stirred overnight at 60 °C. A few

drops of saturated aqueous KPF6 was subsequently added to a product in water yielded

[Ru(CH3CN)(N4Py)](PF6)2 as a yellow solid (138 mg, 74%). 1H NMR (400 MHz, CD3CN): 8.95 (d, J = 5.6 Hz, 2H), 8.82 (d, J = 5.4 Hz, 2H), 7.89‐7.97 (m, 2H), 7.64 (td, J = 7.8 Hz, 1.5Hz, 2H), 7.38 (t, J = 6.4

Hz, 2H), 7.29 (t, J = 6.4 Hz, 2H), 7.09 (d, J = 7.9 Hz, 2H), 6.49 (s, 1H), 4.49 (s, 4H), 2.72(s, 3H). 13C

NMR (101 MHz, CD3CN,): 163.3, 159.6, 156.2, 152.6, 138.5, 137.39, 125.4, 124.9, 124.5, 121.6,

78.1, 64.8, 4.2. ESI mass: m/z 255 [M‐(PF6)2]2+. Anal. Calc. for C25H24F12N6P2Ru: C, 37.55; N, 10.51;

H, 3.03. Found: C, 36.87; N, 10.67; H, 2.98%.

A. 2.1 Characterization of photoproduct [Ru(CD3CN)(CH3CN)(MeN4Py)]2+

Scheme A3. 2D NMR Characterization of [Ru(CD3CN)(CH3CN)(MeN4Py)]2+

[Ru(CD3CN)(CH3CN)(MeN4Py)]2+. 1H NMR (400 MHz, CD3CN) 9.02 (d, J = 5.8 Hz, 1H), 8.83 (d, J = 5.6 Hz, 1H), 8.80 (d, J = 5.4 Hz, 1H), 8.75 (d, J = 4.4 Hz, 1H), 8.01 (t, J = 7.8 Hz, 1H), 7.88 (t, J = 7.8 Hz,

1H), 7.74, (t, J = 7.7 Hz, 1H), 7.64 (d, J = 8.0 Hz, 1H), 7.59 (m, 2H), 7.53 (d, J = 7.8 Hz, 1H), 7.44 (t, J

= 6.8 Hz, 1H), 7.34 (t, J = 6.9 Hz, 1H), 7.26 (t, J = 6.6 Hz, 1H), 7.14 (d, J = 8.2 Hz, 1H), 6.79 (d, J = 7.6

Hz, 1H), 5.42 (d, J = 14.7 Hz, 1H), 5.21 (d, J = 15.0 Hz, 1H), 3.71 (d, J = 17.5 Hz, 1H), 3.31 (d, J = 17.7

Hz, 1H), 2.78 (s, 3H), 2.36 (s, 3H). 13C NMR (101 MHz, CD3CN): 170.5, 163.2, 161.1, 156.6, 153.9,


87

152.3, 151.8, 149.8, 138.0, 137.9, 137.4, 136.7, 126.5, 124.51, 124.9, 124.8, 124.3, 124.3, 124.0,

123.5, 120.5, 117.3, 81.8, 64.7, 63.0, 28.6, 2.8.

Figure A1. DQF‐COSY NMR spectrum of [Ru(CD3CN)(CH3CN)(MeN4Py)]2+ complex in

acetonitrile‐d3.

A. 2.2 Characterization of photoproduct [Ru(CD3CN)(CH3CN)(N4Py)]2+ A. 2.2.1 2D NMR Spectroscopy

Scheme A4. 2D NMR characterization of [Ru(CD3CN)(CH3CN)(N4Py)]2+ 2a and 2b.

2a, 1H NMR (400 MHz, CD3CN) 9.02 (d, J = 5.5 Hz, 1H), 8.85 (d, J = 4.7 Hz, 1H), 8.69 (d, J = 5.5, 1H), 8.13 (td, J = 7.7 Hz, 1.8 Hz, 1H), 7.93 (d, J = 7.2 Hz, 1H), 7.77‐7.74 (m, 2H), 7.67 (m, 1H), 7.58

(t, J = 7.8 Hz, 1H), 7.39 (t, J = 6.6 Hz, 1H), 7.35 (d, J = 7.8 Hz, 1H), 7.33‐7.27 (m, 1H), 7.25 (t, J = 6.7

Hz, 1H), 7.02 (d, J = 8.4 Hz, 1H), 7.00 (d, J = 7.8 Hz, 1H), 6.19 (s, 1H), 5.56 (d, J =18.8 Hz, 1H), 5.24

(d, J = 15.3 Hz, 1H), 4.54 (d, J = 15.3 Hz, 1H), 3.89 (d, J = 18.8 Hz, 1H), 2.77 (s, 3H). 13C NMR (101

MHz, CD3CN) 165.6, 165.4, 164.7, 156.5, 155.7, 154.6, 153.5, 153.0, 140.8, 140.5, 140.0, 139.9, 130.8, 128.7, 127.9, 127.8, 127.5, 126.7, 125.6, 122.8, 82.2, 71.3, 71.3, 65.4, 65.4, 6.8.

Appendix A

88

2b, 1H NMR (400 MHz, CD3CN) 9.13 (d, J = 5.6 Hz, 1H), 8.80, (d, J = 5.5 Hz, 1H), 8.69 (d, J = 5.5 Hz, 1H), 8.62 (d, br, 1H), 7.96 (m, 1H), 7.91 (m, 1H), 7.72 (m, 1H), 7.67 (m, 1H), 7.55 (m, 1H), 7.52 (m),

7.44 (t, J = 6.6 Hz, 1H), 7.33‐7.27 (m, 2H), 7.08 (d, J = 12.0 Hz, 1H), 6.50 (d, br, 1H), 5.68 (s, 1H), 5.18

(d, J = 15.4 Hz, 1H), 4.95 (d, J = 15.4 Hz, 1H), 4.64, (d, J = 16.0 Hz), 4.18 (d, J = 16.0 Hz, 1H), 2.76 (s,

3H). 13C NMR (101 MHz, CD3CN) 166.5, 166.0, 165.3, 156.5, 155.0, 154.6, 153.3, 140.8, 140.7, 140.2, 139.4, 129.3, 128.0, 127.9, 127.9, 127.2, 127.0,126.3, 126.2, 125.1, 81.9, 72.1, 72.1, 66.6,

66.6, 6.8.

The 1H COSY NMR spectra indicate a correlation between coupled protons of the photoproducts,

with blue lines referring to 2a (major product) and green lines to 2b (minor product).

Figure A2. 1H COSY NMR spectra of 2 in acetonitrile‐d3 after irradiation at 457 nm (region between

3.3‐6.1 ppm).


89

Figure A3. 1H COSY NMR spectra of 2 in acetonitrile‐d3 after irradiation at 457 nm (region between

6.8‐9.3 ppm).

In the 1H NOESY spectrum shows cross peaks with a yellow line indicating correlation of protons of

the photoproduct 2a. Due to the low concentration of the minor product 2b cross peaks are weak.

Figure A4. 1H‐1H NOESY NMR spectra of 2 in acetonitrile‐d3 after irradiation at 457 nm.

Appendix A

90

Figure A5. HSQC NMR spectra of 2 in acetonitrile‐d3 after irradiation at 457 nm.

Figure A6. HMBC NMR spectra of 2 in acetonitrile‐d3 after irradiation at 457 nm


91

A. 2.2.2 X−ray Crystallography of Photoproducts

Figure A7. Molecular structure of compound 2a, showing 50% probability ellipsoids. Hydrogen atoms are omitted for clarity.

A single crystal of compound 2a was mounted on top of a cryoloop and transferred into the

cold nitrogen stream (150 K) of a Bruker‐AXS D8 Venture diffractometer. Data collection and

reduction was done using the Bruker software suite APEX2.1 The final unit cell was obtained from

the xyz centroids of 9479 reflections after integration. A multiscan absorption correction was

applied, based on the intensities of symmetry‐related reflections measured at different angular

settings (SADABS). The structures were solved by direct methods using SHELXT2 and refinement of

the structure was performed using SHELXL.3 Refinement of the initial solution indicated that one of

the PF6 counter ions was disordered, with the electron density of four of the six F atoms smeared

out. This disorder was initially described by two components, but subsequent refinement of that

model indicated that the two components were each also disordered (and might be further split).

To constrain the atomic displacement parameters to reasonable values, ultimately DELU/SIMU

instructions were applied to these F atoms. The sof of the major fraction of the two‐site occupancy

model was refined to 0.85. The hydrogen atoms were generated by geometrical considerations,

constrained to idealized geometries and allowed to ride on their carrier atoms with an isotropic

displacement parameter related to the equivalent displacement parameter of their carrier atoms.

Crystal data and details on data collection and refinement are presented in Table A1.

Table A1. Crystallographic data for 2a.

chem formula C27 H27 F12 N7 P2 Ru

Mr 840.56

cryst syst orthorhombic

color, habit yellow, platelet

size (mm) 0.33 x 0.27 x 0.04

space group Pbca

a (Å) 15.206(4)

b (Å) 17.787(4)

c (Å) 24.112(7)

V (Å3) 6522(3)

Z 8

Appendix A

92

calc, g.cm‐3 1.712

µ(Mo K ), cm‐1 0.680

F(000) 3360

temp (K) 150(2)

range (deg) 2.785 – 27.174

data collected (h,k,l) ‐19:19, ‐22:19, ‐30:30

no. of rflns collected 86605

no. of indpndt reflns 7202

observed reflns 6027 (Fo 2 (Fo)) R(F) (%) 3.99

wR(F2) (%) 9.62

GooF 1.82

Weighting a,b 0.0239, 13.7413

params refined 481

restraints 54

min, max resid dens ‐0.756, 1.025

Figure A8. Molecular structure of compound 2b, showing 50% probability ellipsoids. Hydrogen atoms are omitted for clarity.

A single crystal of compound 2b was mounted on top of a cryoloop and transferred into the

cold nitrogen stream (150 K) of a Bruker‐AXS D8 Venture diffractometer. Data collection and

reduction was done using the Bruker software suite APEX2.1 It was clear from the initial unit cell

determination of several crystals that the quality of the batch of crystals was poor, and all crystals

were rather small. Ultimately, the ‘best’ crystal was measured up to a resolution of 0.95 Å using Cu

K radiation. A dataset was measured to a completeness of ca. 90%. Due to the poor crystal quality

the structure determination only serves the purpose to establish atom connectivity. The final unit

cell was obtained from the xyz centroids of 3789 reflections after integration. A multiscan

absorption correction was applied, based on the intensities of symmetry‐related reflections

measured at different angular settings (SADABS). The structures were solved by direct methods

using SHELXT2 and refinement of the structure was performed using SHELXL.3 The hydrogen atoms

were generated by geometrical considerations, constrained to idealized geometries and allowed to

ride on their carrier atoms with an isotropic displacement parameter related to the equivalent

displacement parameter of their carrier atoms. Crystal data and details on data collection and

refinement are presented in Table A2.


93

Table A 2. Crystallographic data for 2b.

chem formula C27 H27 F12 N7 P2 Ru

Mr 840.56

cryst syst triclinic

color, habit yellow, platelet

size (mm) 0.13 x 0.17 x 0.04

space group P‐1

a (Å) 10.1549(4)

b (Å) 10.3638(4)

c (Å) 16.5618(8)

, deg 76.291(3)

, deg 80.761(3)

, deg 75.251(3)

V (Å3) 1628.23(12)

Z 2

calc, g.cm‐3 1.714

µ(Cu K ), cm‐1 5.760

F(000) 840

temp (K) 150(2)

range (deg) 4.508 – 54.223

data collected (h,k,l) ‐10:9, ‐10:10, ‐16:16

no. of rflns collected 7502

no. of indpndt reflns 3533

observed reflns 2492(Fo 2 (Fo)) R(F) (%) 6.33

wR(F2) (%) 17.83

GooF 1.12

Weighting a,b 0.0454, 18.6676

params refined 444

restraints 287

min, max resid dens ‐1.097, 1.233

Appendix A

94

A. 3 Synthesis and characterization of dimeric Fe(III) complexes (chapter 5 and 6) A. 3.1 Synthesis of ligand HL1

Scheme A5. Synthesis of ligand HL1.

Di(pyridin‐2‐yl)methanone oxime (4). Di(pyridin‐2‐yl)methanone (4.19 g, 22.80 mmol) was

combined with hydroxylamine hydrochloride (3.16 g, 45.60 mmol) and pyridine (25 mL). The

reaction mixture was heated at reflux for 2 h. After stirring overnight at room temperature the

mixture was poured onto ice water (100 mL) yielding a precipitate, which was recovered by vacuum

filtration and washed with cold water to yield 4 (4.24 g, 93%) as a white solid which was used

without further purification. 1H NMR (400 MHz, CDCl3) 8.63 (m, 2H), 7.85 (m, 3H), 7.67 (d, J = 8.0

Hz, 1H), 7.45 (m, 1H), 7.35 (td, J = 6.7 Hz, 2.2 Hz, 1H).

Di(pyridin‐2‐yl)methanamine (5). Zinc powder (3.12 g, 47.76 mmol) was added slowly at room

temperature to a mixture of ammonium acetate (1.02 g, 13.26 mmol) and di(pyridin‐2‐

yl)methanone oxime (2.64 g, 13.26 mmol) in 25% NH3/H2O (20 mL) and ethanol (15 mL). The

mixture was heated at reflux for 2.5 h. After stirring at room temperature overnight, the solution

was filtered over Celite, which was subsequently washed with ethanol. After evaporation of the

solvent in vacuo, 2 M NaOH (30 mL) was added to the residue and the product was extracted with

CH2Cl2 (3×30 mL). The combined CH2Cl2 layers were washed with brine (20 mL), dried (over

anhydrous Na2SO4) and the solvent removed in vacuum to yield 5 (1.93 g, 79%) as a yellow oil. 1H

NMR (400 MHz, CDCl3) 8.54 (d, J = 4.8 Hz, 2H), 7.60 (t, J = 7.6 Hz, 2H), 7.37 (d, J = 8.0 Hz, 1H), 7.11 (t, J = 6.4 Hz, 2H).


95

1,1‐Bis(pyridin‐2‐yl)‐N‐(pyridin‐2‐ylmethyl)methanamine (7). Pyridine‐2‐carboxaldehyde (1.12 g,

10.44 mmol) was added to di(pyridin‐2‐yl)methanamine (1.93 g, 10.44 mmol) was added. A cream

solid was obtained after stirring for 10 min, which was washed with cyclohexane to remove traces

of unreacted starting material, yielding 6. Subsequently, 6 was dissolved in MeOH (30 mL) and

NaBH4 (0.79 g, 20.88 mmol) added in small portions. After stirring at room temperature for 2 h, HCl

(aq) was added to bring the solution to pH < 2. After stirring for 30 min 5 M NaOH (aq) was added

to raise the pH to > 9. The methanol was removed in vacuum and the aqueous layer extracted with

ethyl acetate (3×30 mL). The combined organic layers were washed with brine (30 mL) and dried

over anhydrous Na2SO4. Removal of solvent in vacuum provided 7 (2.63 g, 92% over 2 steps) as a

yellow oil. 1H NMR (400 MHz, CDCl3) 8.54 (m, 3H), 7.60 (m, 3H), 7.46 (d, J = 8.0 Hz, 2H), 7.34 (d, J

= 8.0 Hz, 1H), 7.12 (m, 3H), 5.16 (s, 1H), 3.93, (s, 2H).

o‐Tolyl acetate (9). Acetyl chloride (4.20 g, 30.10 mmol) was added to a solution of o‐cresol (1.93 g,

17.89 mmol) in 1% TfOH/CH3CN (65 mL). After stirring overnight at room temperature the solution

was poured onto cold water and extracted with ethyl acetate (3×20 mL). The combined organic

layers were washed with 1 M HCl (30 mL), sat. NaHCO3 (30 mL), brine (30 mL), and dried over

anhydrous Na2SO4. Removal of solvent in vacuum yielded 9 (2.59 g, 96%) as a clear oil. 1H NMR (400

MHz, CDCl3) 7.24 (m, 2H), 7.16, (td, J = 7.2 Hz, 1.2 Hz, 1H), 7.03 (d, J = 8.0 Hz, 1H), 2.33 (s, 3H),

2.21 (s, 1H).

2‐(Bromomethyl)phenyl acetate (10). Azobisisobutyronitrile (1.36 g, 8.32 mmol) and N‐

bromosuccinimide (3.00 g, 16.65 mmol) were added to a solution of o‐tolyl acetate (2.50 g, 16.65

mmol) in benzene (40 mL). After heating at reflux for 6 h the solution mixture was cooled to room

temperature and extracted with CH2Cl2 (2×30 mL). The combined organic layers were washed with

sat. Na2CO3 (30 mL), brine (30 mL) and dried over anhydrous Na2SO4. Evaporation of solvent in

vacuum followed by column chromatography on silica (10% EtOAc/Pentane) yielded 10 (3.12 g,

79%) as a yellow oil. 1H NMR (200 MHz, CDCl3) 7.55 (m, 2H), 7.36 (m, 2H), 4.61 (s, 2H), 2.55 (s,

3H).

2‐(((Di(pyridin‐2‐yl)methyl)(pyridin‐2‐ylmethyl)amino)methyl)phenol (HL1). Diisopropylethylamine

(0.86 g, 6.62 mmol) was added to a mixture of amine 6 (1.20 g, 4.42 mmol) and 2‐

(bromomethyl)phenyl acetate 9 (1.31g, 5.17 mmol) in ethyl acetate (30 mL). After heating at reflux

overnight, the solvent was removed under vacuum and the residue purified by column

chromatography on silica (EtOAc/Pentane /Et3N = 7:2:1). A mixture of acylated 10 and deacylated

amine HL1 was obtained (1.03 g). The mixture (1.03 g) was dissolved in MeOH/ H2O (35:5 mL) and

sat NaHCO3 (20 mL) added. After stirring overnight, water was added (20 mL) and the mixture

extracted with CH2Cl2 (2×20 mL). The combined organic layers were washed with brined (30 mL)

and dried over anhydrous Na2SO4. Evaporation of the solvent in vacuum yielded HL1 (0.81 g, 48%

over 2 steps) as a yellow oil. 1H NMR (400 MHz, CDCl3) 8.65 (d, J = 5.6 Hz, 2H), 8.49 (d, J = 5.6 Hz, 1.2 Hz, 1H), 7.63 (td, J = 7.6 Hz, 1.6 Hz, 2H), 7.55 (td, J = 7.6 Hz, 1.6 Hz, 1H), 7.44 (d, J = 8.0 Hz, 1H),

7.22 (m, 4H), 7.14 (td, J = 7.6 Hz, 1.6 Hz, 1H), 7.08 (m, 1H), 6.94 (m, 2H), 6.70 (td, J = 7.2 Hz, 1.2 Hz,

1H), 5.27 (s, 1H), 3.92 (s, 2H), 3.77 (s, 2H).

Appendix A

96

A. 3.2 Synthesis of ligand HL2

Scheme A6. Synthesis of Ligand HL2.

2,4‐Di‐tert‐butyl‐6‐(hydroxymethyl)phenol (12). Addition of suspensions of paraformaldehyde (1.50

g, 149 mmol) in MeOH (40 mL) to the solution mixture of 2,4‐di‐tert‐butylphenol (30 g, 145 mmol)

and LiOH (0.28 g, 11.63 mmol) in MeOH (40 mL). The reaction mixture was heated at reflux for 16

h, the solvent then was removed under reduced pressure to give a cream solid residue. Dissolve

the residue with diethyl ether and then washed organic solvent with water and dried over

anhydrous Mg2SO4. Removing the solvent to obtained a white powder which purified by

recrystallization from hexane to yield 12 (17.05 g, 50%) as white solid. 1H NMR (400 MHz, CDCl3) δ

7.55 (s, 1H), 7.30 (s, 1H), 6.91 (s, 1H), 4.84 (s, 2H), 2.13 (br, 1H), 1.45 (s, 9H), 1.30 (s, 9H). 13C NMR

(101 MHz, CDCl3) δ 155.8, 144.3, 139.2, 126.7, 126.6, 125.3, 68.5, 37.6, 36.9, 34.3, 32.4.

2‐(Bromomethyl)‐4,6‐di‐tert‐butylphenol (13). A solution of PBr3 (5.77 g, 21.32 mmol) in CHCl3 (60

mL) was added drop‐wise to a solution of 12 (10.08 g, 42.65 mmol) in CHCl3. The reaction mixture

was stirred at room temperature for overnight. Addition of water drop‐wise to quench the reaction

results a white solution. Organic layer was separated and aqueous was extracted with

dichloromethane (2×30 mL). The organic layer was combined, dried over anhydrous Mg2SO4, and

solvent were removed under pressure to give 13 (12.12 g, 95%) as white solid. 1H NMR (400 MHz,

CDCl3) δ 7.33 (d, J = 2.5 Hz, 1H), 7.10 (d, J = 2.4 Hz, 1H), 4.59 (s, 2H), 2.18 (s, 1H), 1.43 (s, 9H), 1.30

(s, 9H). 13C NMR (101 MHz, CDCl3) δ 154.3, 145.6, 139.8, 128.3, 127.3, 125.9, 37.6, 37.0, 35.3, 34.2,

32.5.

3,5‐di‐tert‐butyl‐2‐(((di(pyridin‐2‐yl)methyl)(pyridin‐2‐ylmethyl)amino)methyl)phenol (HL2).

Addition of 2‐(bromomethyl)‐4,6‐di‐tert‐butylphenol (13) (2.32 g, 7.76 mmol) into the solution

mixture of 1,1‐bis(pyridin‐2‐yl)‐N‐(pyridin‐2‐ylmethyl)methanamine (7) (2.12 g, 7.76 mmol) and

KOH (0.7 g, 12.41 mmol) in dry toluene (60 mL). The reaction mixture was heat at 70 °C under N2

for overnight. Filtered cool solution and solvent was then removed under reduce pressure. Addition

of ethyl acetate (50 mL) to crude product then washed with sat. NaHCO3, brine and dried over

anhydrous Na2SO4. Evaporation of solvent in vacuum followed by column chromatography over

silica (Pentane/EtOAc/Et3N = 2:1:0.5) yielded HL2 (2.18 g, 57 %) as a yellow oil. 1H NMR (400 MHz,


97

CDCl3) δ 11.23 (s, 1H), 8.66 (d, J = 4.6 Hz, 2H), 8.48 (d, J = 4.7 Hz, 1H), 7.62 (t, J = 7.6 Hz, 2H), 7.53

(t, J = 7.6, Hz, 1H), 7.46 (d, J = 7.6 Hz, 1H), 7.31 (d, J = 7.8 Hz, 2H), 7.15–7.23 (m, 3H), 7.06 (t, J = 6.9

Hz, 1H), 6.77 (d, J = 2.4 Hz, 1H), 5.29 (s, 1H), 3.95 (s, 2H), 3.80 (s, 2H), 1.49 (s, 9H), 1.23 (s, 9H). 13C

NMR (101 MHz, CDCl3) δ 162.5, 161.6, 156.8, 151.6, 151.1, 142.3, 139.1, 139.0, 138.1, 128.3, 127.2,

126.2, 125.7, 124.9, 124.9, 124.4, 73.0, 58.9, 57.9, 37.7, 36.7, 34.3, 33.6, 32.3.

A. 3.3 Synthesis of ligand HL2

Scheme A7. Synthesis of Complex 1 and 2.

[(L1)Fe(µ‐O)Fe(L1)](ClO4)2 (1). Fe(ClO4)2∙6H2O (0.16 g, 0.44 mmol) in MeOH (1 mL) was added to a

solution of 2‐(((di(pyridin‐2‐yl)methyl)(pyridin‐2‐ylmethyl)amino)methyl)phenol (HL1) (0.14 g, 0.37

mmol) in MeOH (5 mL). Subsequently, Et3N (52 µL, 0.37 mmol) was added dropwise. After stirring

for 15 min the reaction mixture was filtered over cotton wool and placed in an EtOAc bath. Dark

micro‐crystalline solid formed over several days, which were collected and washed with EtOAc and

recrystallized from MeOH, yielding 1 as a purple crystalline solid. (70 mg, 17 %). 1H NMR (400 MHz,

acetonitrile‐d3) 40 (br), 34 (br), 26 (br), 20.3, 18.6, 17.7, 17.1, 16.0, 14.7, 13.7, 10.9, 8.0, 6.7, 3.1, 1.2, ‐2.8. ESI mass: m/z 445 {M‐2ClO4‐}; Anal. Calc. For C48H42N8O11Fe2Cl2∙H2O: C 52.1; H 4.01; N

10.13%, found: C 52.1; H 3.98; N 9.97%.

[(L2)Fe(µ‐O)Fe(L2)](ClO4)2 (2). Fe(ClO4)2∙6H2O (0.12 g, 0.48 mmol) in MeOH (1 mL) was added to the

solution of 3,5‐di‐tert‐butyl‐2‐(((di(pyridin‐2‐yl)methyl)(pyridin‐2‐ylmethyl)amino)methyl)phenol

(HL2) (0.22 g, 0.44 mmol) in methanol (3 mL) Subsequently, Et3N (62 µL, 0.44 mmol) was added

dropwise. After stirring for 30 min the reaction mixture was filtered over cotton wool and placed in

an EtOAc bath. Dark micro‐crystalline solid formed over several days, which were collected and

washed with EtOAc and recrystallized from MeOH, yielding 2 as a purple crystalline solid. (145 mg,

25 %). 1H NMR (400 MHz, acetonitrile‐d3) δ 29.9 (br), 21.9 (br), 18.1, 17.4, 15.35, 14.27, 12.11, 8.14,

8.13, 7.88, 7.20, 4.06, 3.42, 1.76, 1.55, 1.20. (ESI mass: m/z 557 {M‐2ClO4‐}; Anal. Calc. For

C64H72Cl2Fe2N8O11H2O: C 57.71; H 5.75; N 8.41%, found: C 57.69; H 5.73; N 8.27%.

Appendix A

98

A. 3.4 Extra information

Figure A7. 1H NMR spectra of 1 with 1 equiv. of TfOH in acetone‐d6. The sharp spikes at ca. 23 and

42 ppm are spectral artefacts.

Figure A8. 1H NMR spectra of oxidation reaction of 1‐phenylethanol catalyzed by 1 in acetonitrile

after 2h (a) without acid, (b) with 1 equiv. of TfOH.


99

Figure A9. 1H NMR spectra of oxidation reaction of 1‐phenylethanol catalyzed by 2 in acetonitrile

(a) with 1 equiv. of TfOH (b) without acid.

A. 3.5 References

(1) Bruker, (2012). APEX2 (v2012.4‐3), SAINT (Version 8.18C) and SADABS (Version 2012/1). Bruker AXS Inc.,

Madison, Wisconsin, USA.

(2) Sheldrick, G. M. SHELXT – Integrated Space‐Group and Crystal‐Structure Determination. Acta Cryst.

2015, A71, 3−8.

(3) Sheldrick, G. M. A Short History of SHELX. Acta Cryst. 2008, A64, 112−122.

Appendix B

General Description of Techniques and Measurements

Appendix B

102

Solvents for electrochemical and spectroscopic measurements were of UVASOL (Merck) grade or

better.

B. 1 Physical methods UV‐vis absorption spectroscopy. UV‐vis absorption spectrum is recorded with a Specord600

(AnalytikJena) spectrophotometer in 10 mm path length quartz cuvettes.

Nuclear magnetic resonance spectroscopy. 1H NMR spectra (400, 500 and 600 MHz) and 13C NMR

spectra were recorded on a Varian Mercury Plus and Varian Inova, respectively. Chemical shifts are

denoted relative to the residual solvent peak (1H NMR spectra CD3CN, 1.94 ppm; CDCl3, 7.26 ppm).

Dichloromethane was used as an internal reference with Evan’s method for determination of

magnetic susceptibility. 2D NMR; COSY, NOESY, HSQC and HMBC spectra (400 MHz) were recorded

on a Varian Mercury Plus. 1H NMR spectroscopy with in situ irradiation at 420 nm (LED, M420F2,

10 mW at source, Thorlabs) and 365 nm (LED, M365FP1, 10 mW at source, Thorlabs) with light

delivered via a 5 m (400 micron diameter) optical fiber in which the last 3 cm of cladding was

removed and the bare fiber was lightly sanded to give an approximately uniform emission. The bare

fiber end was inserted in the inner tube of a 5 mm Evan’s NMR tube with the sample solution held

in the outer compartment of the tube. NMR spectra were recorded during irradiation.

Elemental analysis. CHN analysis were performed with a Foss‐Heraeus CHN Rapid or a EuroVector

Euro EA elemental analyzer.

Electrospray ionization analysis. ESI mass spectra of complexes were recorded on a Triple

Quadrupole LC/MS/MS mass spectrometer (API 3000, Perkin‐Elmer Sciex Instruments). High

resolution mass spectra (HRMS) were recorded on Bruker MicroOTOF‐Q Instrument at Serveis

Tecnic of University of Girona.

Electron paramagnetic resonance spectroscopy. EPR spectra (X‐band, 9.46 GHz) were recorded on

a Bruker ECS106 spectrometer, equipped with a Bruker ECS 041 XK microwave bridge and a Bruker

ECS 080 magnet in liquid nitrogen (77 K) and Bruker EMXnano spectrometer version 001DRAFT

(110 K).

Fourier transform infrared spectroscopy. FTIR spectra were recorded using a UATR (ZnSe) with a

Perkin Elmer Spectrum400, equipped with a liquid N2 cooled MCT detector.

Raman spectroscopy. Raman spectra were recorded at 785 nm using a Perkin Elmer Raman Station

at room temperature. Raman spectra were recorded at 1064 nm in ordinary glass 1 cm cuvettes,

using a idus‐InGaAs‐512 diode array coupled to a Shamrock 163 spectrograph with a 600 l/mm

grating blazed at 1200 nm and a 500 mW (200 at sample) 1064 nm laser (Rumba, Cobalt) combined

using an Inphotonics Raman probe. Raman spectra recorded at 355 (10 mW Cobalt lasers), 473

(100 mW Cobolt Lasers) and 561 nm (100 mW, Cobolt Lasers) used a home built system in which

the laser was focused on the sample in a 180° backscattering arrangement and Raman scattering

was collected collimated and subsequently refocused via a pair of 2.5 cm diameter plano‐convex

lens (f = 10 cm). The collected light was filtered by an appropriate long pass edge filter (Semrock)

and dispersed by a Shamrock300i spectrograph (slit width 80 micron, Andor Technology) with a

General description of techniques and measurements

103

1200 L/mm grating blazed at 500 nm. Raman spectra at 532 nm (200 mW, Cobolt Lasers) was

recorded in a 90° backscattering arrangement. Raman spectra recorded at 488 nm used a Nikon

TE‐eclipse inverted microscope and 60x confocal objective. Excitation was provided by a Laser

combiner (Andor technology) equipped with an AOTF to control laser power. Raman scattering was

collected and feed into a Shamrock 300 spectrograph using a 1200 l/mm grating blazed at 500 nm

and a iVac CCD detector (Andor Technology). Data were recorded and processed using Andor Solis

(Andor Technology) with spectral calibration performed using the Raman spectrum of

acetonitrile/toluene 50:50 (v:v).1 A multipoint baseline correction was performed for all spectra.

The concentrations used for resonance Raman studies were 0.25‐0.30 mM in acetonitrile and

dichloromethane and samples were held in quartz 1 cm path length cuvettes (3 mL volume).

Irradiation with UV‐vis absorption detection was carried out at 355 nm (7 mW, Cobolt lasers) 405

nm (ONDAX lasers) and 457 nm (50 mW, Cobolt lasers).

Electrochemistry. Electrochemical measurements were carried out on a model CHI760B, CHI600C

or CHI1200C Electrochemical Workstation (CH Instruments). A Teflon‐shrouded glassy carbon (3

mm diameter), indium tin oxide (ITO) on glass slides (1 cm × 1.4 cm), gold and platinum macro

electrode were employed as working electrode (CH Instruments). A Pt wire was used as an auxiliary

electrode, and Ag/AgCl reference electrode or a saturated calomel electrode (SCE) was used as

reference electrode. Cyclic voltammograms were obtained at a sweep rate of 100 mV s‐1 in

anhydrous dichloromethane or acetonitrile containing 0.1 M tetrabutylammonium

hexafluorophosphate (TBAPF6) with analyte concentrations of 0.5 to 1 mM unless stated

otherwise. Potentials are quoted with respect to SCE unless stated otherwise. Redox potential

(Ep,a, anodic peak potential; Ep,c, cathodic peak potential; E1/2 = (Ep,a + Ep,c)/2) values are ± 10

mV.

Spectroelectrochemistry. Spectroelectrochemical experiments in solution were carried out using an

OTTLE cell (a liquid IR cell modified with Infrasil windows and a platinum mesh working and counter

electrode and a Ag/AgCl reference electrode) mounted in a Specord600 UV‐vis spectrometer with

potential controlled by a CHI600C potentiostat. In situ UV‐vis absorption spectroelectrochemistry

of poly‐1 was carried out by initial modification of an ITO electrode by cyclic voltammetry follow by

transfer to 2 mm quartz cuvette as an electrochemical cell.

B. 2 Quantum yield determinations Absolute quantum yields were determined with reference to the actinometer potassium

ferrioxalate.2 Laser flux was determined by the method of total absorption using 2 mL of 0.15M

potassium ferrioxalate in a 1 cm pathlength cuvette with stirring. The actinometer was irradiated

for 40 s (at 457 nm) and a reference cuvette was held apart from the excitation source. After

irradiation, 1 mL of each solution was added to buffered aqueous phenanthroline (0.1 M, 2 mL),

and diluted 10 fold with water and left to stand in the dark for at least 30 min and the absorbance

determined at 510 nm. The photon flux was calculated using equation (1).

∆Ф

20 1

Appendix B

104

Where L is the pathlength of the cuvette, ε is the molar absorptivity of iron(II) tris‐phenanthroline

(11100 L mol‐1 cm‐1 at λmax 510 nm), Ф is quantum yield of the actinometer at 457 nm, i.e. 0.85,3 t

is the irradiation time (40 s), F is the fraction of the light the actinometer absorbed; as the

absorbance of the ferrioxalate solution was above 2, F was taken to be 100%.

2 mL of a solution of Ru(II) complexes 1 (or 2) (concentration was determined by the absorbance

at 457 nm, which should be is close to 2) was irradiated under identical conditions with the

absorbance at 457 nm monitored by UV‐vis absorption spectroscopy. The change in absorbance

with time was used to determine the photochemical quantum yield using equation (2).

Ф /

2

∆ / /

/ ∗

Where ∆ is the change in absorbance at 457 nm over the irradiation time, is the molar

absorbtivity of complexes at 457 nm, the irradiation time was dependent on the change in

absorbance and was less than 0.2, allowing for the approximation that the number of moles of

photons absorbed per unit time is essentially unchanged, i.e. ca. 0% transmittance and hence equal

to Nhv/t.

B. 3 Spin crossover occurrence The spin crossover effect is observed for complexes of the transition metal ions particularly those

of the first transition series with configurations d4 to d7, which splitting of the energy of the d

orbitals into the t2g and eg sets. The strength of ligand filed about the metal ion can exist in either

the high‐spin (HS) or low‐spin (LS) state that a weak field and strong field stabilized the HS and LS

state, respectively (Figure B1). The spin transition (ST) or spin crossover (SCO) exhibit a switching

phenomenon by external perturbation upon change of temperature, pressure, irradiation with light

or in a magnetic field.4 The population of LSHS transition involves of anti‐bonding orbital which

accompanies a lengthening and weakening of Fe‐L bond lengths (vide infra).5 The change in spin

state is involvement the change in electronic structure of central ion which make the physical and

chemical properties of the whole molecule change.

Figure B1. Electronic distribution between HS and LS states of an octahedral for a d6 Fe(II)

coordination compound exhibited spin crossover (top). Potential energy curve for spin crossover

system, r = change in bond length (bottom).

General description of techniques and measurements

105

B. 4 Light-induced spin transition Light‐induced spin transition in solution is first observed by McGarvey et al.6 who reported using

laser excitation within an MLCT absorption band of Fe(II) spin crossover complexes to increase the

population of the high spin state. In particular, inducing a spin transition of [Fe(ptz)6](BF4)2 complex

in a solid state at 10 K from LS to HS by irradiation with light, know as light induced excited spin

state trapping (LIESST) effect, has attracted much attention since Decurtins and co‐workers7

discovered in 1984.

Figure B2. Schematic showing LIESST and reverse‐LIESST. Light induced excitation is shown by full

arrows, dotted arrows denote non‐radiative internal conversion or formally spin‐forbidden

intersystem crossing.

The photophysical process involved in LIESST phenomenon are represented in the Jablonski

diagram (Figure B2). Upon irradiation of the LS isomer with green light results in a spin‐allowed

transition to a singlet state followed by rapid internal conversion to the lowest 1T state (1T1).

Intersystem crossing to a low‐lying 3T state (weakly allowed through spin‐orbit coupling) which can

decay back to the LS state or to the HS state where it is “frozen” (vide supra). Light‐induced back

conversion (reverse‐LIESST) of metastable HS state is effected by irradiation with red light,

whereby the HS in 5T2 state is excited to 5E state and relaxes to the low spin 1A1 configuration.

B. 5 Resonance Raman scattering Raman spectroscopy involves excitation of the molecule with monochromatic light from initial to

virtual state and then molecule relaxes to the same or different vibrational level of incident light.

When the wavelength of incident light coincides with an electronic transition of the molecule thus

the Raman scattering observed might be enhanced by 104‐106 times under dilute solution.

However, the enhancement is not in general observed for all vibrational modes, rather certain

modes of nature electronic transitions become a selective enhancement.8

Appendix B

106

Figure B3. Schematic diagram depicting resonance Raman scattering.

B. 6 References

(1) Calibration reference for Raman measurement

(2) Hatchard, C.; Parker, C. A., A new sensitive chemical actinometer. II. Potassium Ferrioxalate as a

Standard Chemical Actinometer. Proceedings of the Royal Society of London. Series A. Mathematical and

Physical Sciences 1956, 235, 518−536.

(3) Demas, J.; Bowman, W.; Zalewski, E.; Velapoldi, R., Determination of the Quantum Yield of the

Ferrioxalate Actinometer with Electrically Calibrated Radiometers. J. Phys Chem. 1981, 85, 2766−2771.

(4) (a) Gütlich, P.; Hauser, A. Spiering. H. Thermal and Optical Switching of Iron(II) Complexes. Angew.

Chem. Int. Ed. Engl. 1994, 33, 2024−2054. (b) P. Gütlich, Spin Crossover – Quo Vadis?, Eur. J. Inorg.

Chem. 2013, 5, 518−591.

(5) Browne, R.W.; McGarvey J. J.; Raman Scattering and Photophysics in Spin‐State‐Labile d6 Metal

Complexes Coord. Chem. Rev. 2006, 250, 1696−1709.

(6) McGravey, J. J.; Lawthers, I. Photochemically‐induced Perturbation of the 1A⇌5T Equilibrium in

FeII Complexes by Pulsed Laser Irradiation in the Metal‐to‐Ligand Charge‐Transfer Absorption Band. J.

Chem. Soc., Chem. Commun. 1982, 906−907.

(7) Decurtins, S.; Gütlich, P.; Köhler, C.P.; Spiering, H. Light‐Induced Excited Spin State Trapping in a

Transition‐Metal Complex: The Hexa‐1‐propyltetrazole‐Iron(II) Tetrafluoroborate Spin‐Crossover

System. Chem. Phy. Lett. 1984, 105,1−4.

(8) Browne, W. R.; McGarvey,. J. J. The Raman Effect and its Application to Electronic Spectroscopies in

Metal‐Centered Species: Techniques and Investigations in Ground and Excited States. Coord. Chem. Rev.

2007, 251, 454−473.

Documents

University of Groningen Mechanisms in Ruthenium(II) … · 2017. 2. 10. · Mechanism of an diiron(III)bis‐phenolato complex catalyzed oxidation 67 O Fe Fe N N O N N N O N N N 2+