Tuneable reversible redox of cobalt(iii) carbazole complexes

View Article Online

View Journal

Dalton

Transactions

Accepted Manuscript

This article can be cited before page numbers have been issued, to do this please use: S. Brooker, M. S.

Bennington, H. L. Feltham, Z. J. Buxton and N. White, Dalton Trans., 2017, DOI: 10.1039/C7DT00295E.

This is an Accepted Manuscript, which has been through the

Volume 45 Number

1 7 January 2016 Pages 1–398

Royal Society of Chemistry peer review process and has been

accepted for publication.

Dalton

Transactions

Accepted Manuscripts are published online shortly after

An international journal of inorganic chemistry

www.rsc.org/dalton

acceptance, before technical editing, formatting and proof reading.

Using this free service, authors can make their results available

to the community, in citable form, before we publish the edited

article. We will replace this Accepted Manuscript with the edited

and formatted Advance Article as soon as it is available.

You can find more information about Accepted Manuscripts in the

author guidelines .

Please note that technical editing may introduce minor changes

to the text and/or graphics, which may alter content. The journal’s

standard T erms & Conditions and the ethical guidelines, outlined

in our author and reviewer resource centre, still apply. In no

event shall the Royal Society of Chemistry be held responsible

for any errors or omissions in this Accepted Manuscript or any

consequences arising from the use of any information it contains.

ISSN 1477-9226

PAPER

Joseph T. Hupp, Omar K. Farha et al.

Eﬃcient extraction of sulfate from water using

framework

a Zr-metal–organic

rsc.li/dalton

Please do not adjust margins

Dalton Transactions

Page 1 of 15

View Article Online

DOI: 10.1039/C7DT00295E

Dalton Transactions

ARTICLE

Tuneable reversible redox of cobalt(III) carbazole complexes

Michael S. Bennington,^aHumphrey L. C. Feltham,^aZoë J. Buxton,^aNicholas G. White,^a,band Sally

Brooker^a*

Received 00th January 20xx,

Accepted 00th January 20xx

Four tridentate carbazole-based ligands, HL^tBu/H(3,6-di(tert-butyl)-1,8-bis[5-(3-benzyl-1,2,3-triazole)]-9H-carbazole),

HL^tBu/tBu(3,6-di(tert-butyl)-1,8-bis[5-(3-(4-tert-butyl)benzyl-1,2,3-triazole)]-9H-carbazole), HL^H/H(1,8-bis[5-(3-benzyl-1,2,3-

triazole)]-9H-carbazole) and HL^H/tBu(1,8-bis[5-(3-(4-tert-butyl)benzyl-1,2,3-triazole)]-9H-carbazole), were prepared and

complexed with cobalt(II) tetrafluoroborate. In situ air oxidation resulted in cobalt(III) complexes 1-4 with the general

formula [Co^III(L)₂]BF₄∙xH₂O (1: L = L^tBu/H, x = 2; 2: L = L^tBu/tBu, x = 1; 3: L = L^H/H, x = 0.5; 4: L = L^H/tBu, x = 2). X-ray structural

characterisation confirmed that the four complexes are isostructural, with two orthogonally coordinated deprotonated

tridentate ligands providing an octahedral N₆-donor set to the cobalt(III) ion. ¹H NMR studies show that this structure is

maintained in CDCl₃and DMSO-d₆solution. Cyclic voltammetry on 1-4 in MeCN showed that all of the complexes exhibit

two reversible, one-electron oxidation processes (probably due to ligand oxidations), and an irreversible or quasi-

reversible reduction process (probably due to reduction of Co(III) to Co(II)). As expected, the oxidations move 120-140 mV

to lower potentials on adding tert-butyl substituents to the 3 and 6 positions of the carbazole rings (and the potentials are

far less sensitive to the nature of the benzyl ring substituents).

DOI: 10.1039/x0xx00000x

www.rsc.org/

twelve nickel(II) complexes, NiL₂, of

a family of acyclic

Introduction

diphenylamine-dipyrazolyl ligands (II, Figure 1) also exhibited

interesting redox properties.⁷These complexes have two

quasi-reversible one-electron processes, ascribed by the

authors to ligand centered oxidations, which were tuned over

Transition metal complexes of carbazole-based acyclic ligands

have many practical applications,¹including as catalysts for

Nozaki-Hiyama reactions,²and as cyanide detection agents

using quenching of fluorescence (I

, Figure 1).³There are also a

a wide range (700 mV) by varying the two substituents

X and Y

para- to the central amine nitrogen on each ligand (Figure 1).

few examples of macrocyclic complexes featuring carbazole

units in porphyrin-like systems,¹with the extended aromatic

system giving potentially interesting photoelectric properties,^4,

5

while others feature alternating carbazole and pyridine units

IV, Figure 2).⁶

(

Several of these complexes have interesting redox

properties: the most relevant of which, to the present study, is

the macrocyclic cobalt(II) carbazole complex (IV, Figure 2). This

complex exhibited two reversible one-electron redox events,

at E_m= 0.78 and 1.05 V (vs. Ag/Ag⁺), which the authors

suggested could be due to Co(II) to Co(III) oxidation and a

ligand centered oxidation, respectively.⁶However, the neutral

metal-free macrocycle showed two irreversible oxidation

processes at significantly higher potentials (first onset at 1.25

V), which were attributed to the two carbazole moieties being

consecutively oxidized to cation radicals. A related series of

Figure 1. Cu(II) and Ni(II) complexes of carbazole and diphenylamine based

acyclic ligands: [Cu^II(L^tBu/H)X] , left;³[Ni^II( )₂] II, right.⁷

I

X,Y

One area of potential improvement for more efficient and

industrially viable dye-sensitized solar cells (DSSCs) is the

-

development of alternatives to the wide-spread use of I₃/I^-

redox electrolyte,⁸with considerable interest being focused on

the use of cobalt based redox electrolytes.^9-11

A key

requirement for an effective redox electrolyte is an

appropriately positioned reversible redox process, with the

ideal positioning of this redox process dependent on the

specific dye used. Reports of effective cobalt based redox

electrolytes have had processes centered near the Fc/Fc⁺

redox couple, with some examples falling slightly below (-90

This journal is © The Royal Society of Chemistry 20xx

Dalton Trans., 2017, 00, 1-6 | 1

Please do not adjust margins

Dalton Transactions

Page 2 of 15

View Article Online

DOI: 10.1039/C7DT00295E

ARTICLE

Dalton Transactions

Scheme 1: Synthetic route to the family of bis(triazolyl)carbazole ligands from 9H-carbazole. Reagents and conditions: (a) ^tBuCl, anhydrous AlCl₃, dry DCM, RT; (b) N-

bromosuccinimide, silica gel, dry DCM, RT, light-free; (c) anhydrous AlCl₃, dry toluene, RT; (d) TMS-acetylene, Pd(PPh₃)₂Cl₂, CuI, dry NEt₃, reflux; (e) K₂CO₃, MeOH, reflux; (f) 4-

substituted benzyl bromide, NaN₃, ascorbic acid, CuSO₄∙5H₂O, Na₂CO₃, DMF:H₂O (4:1).

mV)¹⁰and others slightly above (+220 mV)¹¹this reference 2.35 Å). Interestingly, of these four cobalt(II) compounds, two

couple. We have reported cobalt complexes with reversible were prepared without the use of inert atmospheres (IV and

redox in this region,^12-14but which had poor solubility and

V

),^{6, 15}one under a nitrogen atmosphere (III),¹⁶and one under

were not readily tuneable. Hence we decided to instead target strict Schlenk conditions (VI).¹⁷

cobalt complexes of readily tuneable carbazole-based

tridentate ligands (Scheme 1).

^tBu

A

search

of

the

CSD

using

a

generic

diphenylamine/carbazole N₃-bound cobalt fragment (ESI,

15-17

All

N

Py

2.16

Figure S57) returned just four structures (Figure 2).^6,

four complexes were of cobalt(II), but they differ in

2.09

THF

N

^2.31N

H₂O

2.27

N_2.35

1.93

B(Ph)₄

Co

N _2.37

2.26

N

^2.32N

2.21

Co

1.95

N

2.22

N

Py

coordination numbers, ranging from 7 for III (N₇pentagonal

2.25

THF

N

bipyramidal), to 6 for IV (N₄O₂octahedral), to 5 for

V (N₃Cl₂

trigonal bipyramidal), to 4 for VI (N₃Cl flattened tetrahedral).

Complexes III, IV and VI were red, with the last two stated to

be high-spin Co(II), and III reported to have μ = 4.6 B.M,

IV

^tBu

III

^tBu

consistent with also being high-spin Co(II). In contrast,

V was

blue, with no spin state explicitly stated. Two of the four

NH

ligands were carbazole-based (IV and VI, Figure 2) and two

2.33

2.08

were diphenylamine-based (III and

were macrocyclic ligands (III and IV, Figure 2).^{6, 16}In one of the

diphenylamine complexes, blue complex , the cobalt(II) was

V, Figure 2). Two of them

2.05

N

Co

O

N

O

1.91

2.07

N

2.28

2.32

Co

_2.07N

VI

Cl

V

Cl

2.23

V

ⁱPr ⁱPr

Mes

Cl

bound through a still protonated amine group (Co-N 2.33 Å)

whereas in III this moiety was deprotonated (Co-N 2.09 Å). In

comparison the carbazole complexes had shorter Co-

Figure 2. Structurally characterized cobalt complexes of carbazole and diphenylamine

based ligands. Clockwise from top left: [Co^IIL³(py)₂]BPh₄;¹⁶[Co^IIL(THF)(H₂O)]∙THF;⁶

[(5)Co^IICl];¹⁷[Co^II(Bopa-ip)Cl₂].¹⁵Bond distances for the cobalt(II) coordination spheres

N(central) bond lengths of 1.91-1.95 Å. The Co-N(flanking) are given in Å and rounded to 3 s.f.

bond lengths are shorter in the two acyclic complexes, 2.05-

2.08 Å,¹⁷than in the two macrocyclic complexes, 2.31-2.37 Å.^6,

¹⁶This is likely due in part to the smaller coordination numbers

for the two acyclic complexes (4 and 5, versus 6 and 7 for the

macrocycles). The geometrical constraints imparted by the

macrocycles may also be playing a role. High spin complex IV

also exhibits two significantly shorter ‘axial’ Co-N(carbazole)

bonds (1.93 and 1.95 Å) than the four ‘equatorial’ bonds (2.22-

Given (a) the redox processes of the macrocyclic carbazole

cobalt(II) complex IV (Figure 2) fall close to the range

interesting for DSSCs,⁶(b) the series of nickel complexes II

(Figure 1) show how acyclic carbazole-like ligands can provide

redox tuneability,⁷(c) the paucity of cobalt complexes of

carbazole or carbazole-like ligands and the complete lack, to

date, of bis-ligand CoL₂or cobalt(III) complexes, it was thought

that CoL₂complexes of carbazole ligands such as that reported

2 | Dalton Trans., 2017, 00, 1-3

This journal is © The Royal Society of Chemistry 20xx

Please do not adjust margins

Dalton Transactions

Page 3 of 15

Dalton Transactions

View Article Online

DOI: 10.1039/C7DT00295E

ARTICLE

by Jang and co-workers (for generating a cyanide sensing degradation of the product. Attempts to purify the

copper complex,

, Figure 1),³may prove interesting. The deprotected alkyne formed via this route proved unsuccessful,

synthesis and redox chemistry of the first family of such with a clean sample unable to be obtained, likely due to

I

complexes is reported herein.

further degradation of the product during purification. This

method did provide material suitable for subsequent reaction,

however the obtained HL^tBu/Halso proved difficult to purify,

Results and Discussion

thus

a

cleaner route was desired. To this end,

ethynyltrimethylsilane (TMS-acetylene) was instead employed

as the protected alkyne. The two protected 1,8-dialkyne

intermediates are obtainable in excellent yields (84-89%, step

d), slightly higher than the 81% reported for the 2-methyl-3-

butyn-2-ol protected alkyne.³The TMS protecting group is

Synthesis of ligands

The synthetic route to the four ligands is shown in Scheme 1

and was inspired by the published synthesis³of HL^tBu/H, but

differs in the details (including the protecting group employed,

and use of one-pot CuAAC), all of which are provided herein.

Because the 3,6-positions of 9H-carbazole are the most

reactive sites,¹⁸the first step is protection of these positions

through Friedel-Crafts alkylation (step a).^17,

then removed through

a straightforward methanolysis

reaction, which gives the corresponding pair of deprotected

terminal dialkynes cleanly in quantitative yields. This change to

a TMS protecting group proved a real boon, greatly facilitating

the clean synthesis of this family of ligands, while also

improving yield for the two steps from 71% to 81-88%. For

completeness we note that, concurrent with the present

study, one of these two dialkynes was reported as being

prepared by a similar method, although specific synthetic

details were not provided in that paper.²⁶

19, 20

After

recrystallisation from cyclohexane to remove the other

alkylated products (mostly mono- and tri-alkylated), this gives

clean 3,6-di-tert-butyl-9H-carbazole in moderate yield (37%,

comparable to literature reports of 47%),²⁰and is easily carried

out on a tens of grams scale. It should be noted that,

subsequent to the present study, an improved protocol for this

step was reported by Cole and Kunz in 2016,²¹which allowed

for a similar scale (20 g) whilst vastly improving the yield

(95%), by using a method published by Hou in 2006.²²

Alternative

protecting

groups

such

as

tributyl(trimethylsilylethynyl)tin have been used and can give

very good yields (99% reported for the two steps),²⁷but the

reagent is both more expensive and significantly more toxic.

The TMS-protected alkyne appears stable indefinitely, but

the deprotected alkynes are somewhat prone to degradation,

so they were prepared from the protected alkyne as required

for prompt subsequent reaction, or stored in the freezer for

short periods (stable for a week or two).

In the next step, bromination of the 1- and 8- positions is

carried out using N-bromosuccinimide (NBS) and silica in DCM

(step b); a method adapted from the protocol Smith reported

in 1992 for the bromination of carbazoles.¹⁸The reaction slurry

is stirred in light-free conditions for several days, and affords

microanalytically

clean

1,8-dibromo-3,6-di-tert-butyl-9H-

carbazole in excellent yields (96%). Previously this step has

involved use of Br₂(a volatile liquid and cumulative toxin) and

glacial acetic acid, so avoiding the use of Br₂reduces exposure

to hazardous materials, and facilitates larger scale reactions,

while still giving comparable yields (96% with NBS, cf. 99%

using Br₂and glacial AcOH).²⁰The yield reported here is similar

to that reported for the analogous NBS reaction with 3,6-

dimethylcarbazole (95%).¹⁷

The final step is in situ generation of the benzyl or 4-tert-

butylbenzyl azide of choice from the corresponding organic

bromide and NaN₃,²⁸then, in the same pot, a CuAAC

reaction^29-31with the appropriate 1,8-diethynylcarbazole to

form the four ligands targeted in this work. These ligands were

obtained in moderate to good yields (41-83%), although

H

L^tBu/Hwas obtained in lower yields than previously reported

(60% cf. 81%).³Compared to the reported synthesis, in which

benzylazide was first isolated and then reacted with the

alkyne, we consider that the ease of a one-pot reaction and

the safety benefit of not isolating a potentially explosive

organic azide are worth the slightly reduced yield.

If desired, it is then possible to remove the tert-butyl

groups from the 3,6-positions in a reverse Friedel-Crafts

reaction to give 1,8-dibromocarbazole (step c), using

a

method²³adapted from Tashiro’s protocols reported in 1979

for the trans-alkylation of aromatic compounds.²⁴This method

is a viable alternative to the literature route, which instead

progresses through nitration, bromination, reduction of the

nitro groups to amines, then removal of the amine groups

(15% in 4 steps cf. 18% in 3 steps in the current study).²⁵It also

has the advantage of utilizing a precursor already prepared in

large quantities for synthesis of the other ligands described

herein.

Synthesis of cobalt complexes

Initially attempts were made to isolate the [Co^IIL₂] complexes,

with

a

view towards potential spin crossover activity.³²

However it proved difficult to isolate the cobalt(II) complexes

as whilst the early yellow-orange of the reaction solutions

indicate that [Co^IIL₂] species may be formed initially, they are

very air-sensitive and the color continues to change, first to

yellow-green and then on to the green-blue of the cobalt(III)

complexes. Given the negative potentials of the Co^III/Co^IIredox

couple (vide infra), it is unsurprising that those initial attempts

to isolate the cobalt(II) complexes were unsuccessful. Given

the reversible redox observed, we decided to instead target

The Sonogashira cross-coupling of the 1,8-dibrominated

carbazoles with a protected alkyne was altered from that

reported by Jang and co-workers in 2011 for the synthesis of

H

L^tBu/H, in which a 2-methyl-3-butyn-2-ol group was used.³The

alcohol protected alkyne proved difficult to isolate cleanly, and

the subsequent deprotection step appears to cause some

This journal is © The Royal Society of Chemistry 20xx

Dalton Trans., 2017, 00, 1-3 | 3

Please do not adjust margins

Dalton Transactions

Page 4 of 15

View Article Online

DOI: 10.1039/C7DT00295E

ARTICLE

Dalton Transactions

the isolation of a family of cobalt(III) complexes and to carbazole groups, and between carbazole tert-butyl groups

investigate the tune-ability of the redox potentials. (i.e. H₁₅) and benzyl arms (Figures S48 and S49); importantly

With this revised goal in mind, the four cobalt(III) these interactions are not observed for the free ligand (Figure

complexes 1–4 were synthesized by combining, initially under S56). Taken together, the 1D and 2D NMR data are consistent

an inert atmosphere, Co^II(H₂O)₆(BF₄)₂and two equivalents of with the complexes having a tightly folded structure in both

the desired ligand in either acetone:methanol (

and ) or DMF ( ). For the complexations not performed in

pyridine ( and 3), an excess of triethylamine was added as a is shown in Figures 3-4 (2 and 3) and Figures S41 - S46.

1), pyridine (

2

CDCl₃and DMSO-d₆solution.

4

3

A comparison of the four complexes with their free ligands

1

base. In order to ensure complete oxidation to cobalt(III), after Generally, the complexes show relatively similar ¹H NMR

a minimum of 30 minutes the reaction solutions were opened spectra, although interestingly the triazole resonances in

to air. In the case of

ether vapour into the reaction solution yielded crystals which the carbazole moiety, are significantly downfield relative to

were filtered and then dried to give [Co^III(L^tBu/H)₂]BF₄∙2H₂O (

), the same resonances in complexes and , which do not have

[Co^III(L^H/H)₂]BF₄∙0.5H₂O ( ) and Co^III(L^H/tBu)₂]BF₄∙2H₂O (

), as these tert-butyl groups.

blue-green solids in yields of 64, 61 and 39% respectively. For

, the reaction was instead taken to dryness and the solid

1, 3 and 4, subsequent diffusion of diethyl complexes 1 and 2, which have tert-butyl groups attached to

1

3

4

3

4

2

residue dissolved in acetonitrile before vapor diffusion of

diethyl ether into the complex solution. The resulting blue-

green crystalline solid was filtered and dried in vacuo giving

[Co^III(L^tBu/tBu)₂]BF₄∙H₂O (

2) in 63% yield.

The four cobalt(III) complexes were characterized by

1

microanalysis, ESI-MS, H, ¹³C and 2D NMR spectroscopy and

X-ray crystallography. In all four cases,

1

–

4

, the microanalysis

data matched the calculated values for [Co^III(

L

)₂]BF₄∙xH₂O and

the mass spectra showed a single major peak which matches

the predicted isotope pattern of the expected [Co^III(

L

)₂]⁺

fragment.

NMR spectroscopy

The ¹H NMR spectra of these diamagnetic cobalt(III) complexes

all show the conspicuous absence of the amine proton signal

of the carbazole, signifying deprotonation and subsequent

metal coordination has occurred (Figures 3 and 4). There is no

broadening or splitting of the signals of the complexes

compared to the free ligand, consistent with diamagnetic

cobalt(III), and indicating that the symmetry of the free ligands

seen by NMR is retained in these complexes.

Figure 3. Partial ¹H NMR spectrum of HL^tBu/tBuand the corresponding cobalt(III) complex

2 (298 K, CDCl₃).

Comparison of the ¹H NMR spectra in CDCl₃of the free

ligand HL^tBu/tBuand the corresponding complex

2 reveals

several shifts in signal positions on deprotonation and

subsequent cobalt coordination (Figure 3). Notably, the H₄

(carbazole) and H₁₅(carbazole tert-butyl) peaks have been

shifted downfield relative to the free ligand. In contrast, all

other peaks have been shifted upfield, with the protons on the

benzyl tert-butyl group, benzyl ring, and CH₂linker

experiencing the largest shifts. The downfield shift in the

carbazole proton presumably arises due to the 3⁺cobalt ion

removing electron density from the formally anionic ligand,

but it is interesting to note that the other carbazole proton, H₂,

is unshifted. The significant upfield shift of the resonances of

the benzyl arm is consistent with shielding due to through-

space interactions with the carbazole moiety of the other

coordinated ligand (as is observed in the solid state, vide infra).

The existence of these close contacts was further investigated

by 2D ROESY NMR spectroscopy in DMSO-d₆. Through-space

H–H interactions are observed between the benzyl groups and

Figure 4. Expansion of the key region of the ¹H NMR spectra of the cobalt(III) complexes

2 and 3 and the corresponding free ligands HL^tBu/tBuand HL^H/H(298 K, DMSO-d₆).

Scheme 1 shows the numbering scheme used. Stacked plots of all four complexes and

their corresponding ligands are provided in the SI, S41-44.

4 | Dalton Trans., 2017, 00, 1-3

This journal is © The Royal Society of Chemistry 20xx

Please do not adjust margins

Dalton Transactions

Page 5 of 15

Dalton Transactions

View Article Online

DOI: 10.1039/C7DT00295E

ARTICLE

Table 1 Selected bond lengths (Å) and angles (°) for the structures of the family of four [Co^III(L)₂]BF₄complexes. Where a range is provided *-*, the average is presented in [*].

[Co^III(L^tBu/tBu)₂]BF₄∙1.5Et₂O

[Co^III(L^H/H)₂]BF₄∙0.5DMF

[Co^III(L^H/tBu)₂]BF₄·1.5Et₂O

[Co^III(L^tBu/H)₂]BF₄

∙solvent^a

∙0.5H₂O^b

∙solvent^a

P-1

Space Group

M-N_cbz

P4/n

P2₁/c

P2₁/n

1.928(3)

1.916-1.921(3) [1.919]

1.924-1.929(2), 1.917-1.923(2),

[1.923]

1.937-1.940(5) [1.939]

1.938-1.951(5) [1.946]

M-N_triazole

1.945(2)

1.936-1.949(3) [1.944]

1.950-1.957(2), 1.943-1.947(2),

[1.949]

N_triazole-M-N_triazole

179.52(12)

177.53(13), 176.90(14) [178.72]

176.54(8), 177.17(8),

178.06(8),175.68(8) [176.86]

178.39(8), 179.48(8) [178.94]

176.3(2), 175.35(18) [175.8]

N_cbz-M-N_cbz

Cis-angles^c

N_cbz-M-N_triazole

Σ (O_h) ^c

180

89.76-90.24 [90.00]

90.24(6)

1.92

178.61(14)

87.55-92.12 [90.01]

90.09-91.30 [90.78]

12.07

177.93(19)

84.7-92.66 [90.06]

89.8-91.5[90.7]

18.09

84.44-93.21 [90.01]

90.21-91.55[90.92]

17.95, 15.73

Cbz∠Cbz^d

Trz∠Trz^e

90.00

74.17

82.11, 89.07 [85.59]

17.69, 13.64, 4.67, 17.34 [13.34]

81.74

20.98

16.29, 24.57 [20.43]

8.83, 10.41, 15.41, 16.28 [12.73]

17.41, 17.71 [17.56]

7.96, 9.45, 10.00, 7.94 [8.84]

Cbz∠Trz^f

14.30

8.85, 9.13, 2.99, 14.73,

4.66, 6.95, 5.46, 11.96 [8.09]

^aPlaton-Squeeze was used to account for lattice electron density (presumably disordered solvents of crystallisation) that was unable to be modelled in a sensible

way.^{33 b}The crystal structure contains two full complexes within the asymmetric unit. ^cRange of cis angles for the CoN₆center(s) with the mean value in parentheses. ^d

d

Σ (O_h), the octahedral distortion parameter, is calculated as the sum of the variances of each of the 12 cis angles from the ideal 90°: Σ (O_h) = 0° is a perfect O_h.^34-37

Angle between the mean planes of the carbazole rings of a complex. ^eAngle between the mean planes of each triazole ring within a ligand. ^fAngle between the mean

plane of each triazole ring and the mean plane of the carbazole ring for each ligand.

Structure Descriptions

towards opposite sides of the plane of the carbazole moiety,

creating ‘propeller-like’ motif. The content of non-

Single crystals of all four [Co^III(

L

)₂]BF₄complexes were grown

a

coordinated lattice solvent also varies between the structures.

by diffusion of diethyl vapor into solutions of the complexes

(see experimental section for details) and X-ray structure

determinations carried out. The four complexes are

isostructural (Figure 5, and SI Figure S58). In each complex, the

cobalt(III) ion is coordinated in an N₆octahedral fashion by

-

two, roughly perpendicular, deprotonated tridentate (L)

ligands (Figure 5). The oxidation state of the cobalt ion is clear

from consideration of the sharp NMR spectra (see earlier), the

charge balance, and the relatively short Co-N bond lengths

(1.92-1.97 Å), all of which are consistent with cobalt(III) (Table

1). These are the first examples of cobalt(III) complexes of such

ligands.

Each ligand is coordinated through the N₃donor pocket

created by the nitrogen atom of the carbazole moiety and two

triazole nitrogen atoms. The ligands impose very regular (Σ ≤

18^o, Table 1) octahedral coordination geometry on the

cobalt(III) ion, with all cis-angles close to 90° (84.4-93.2°, Table

1). The planes of the carbazole moieties of each ligand are

roughly perpendicular to one another (74-90°, Table 1), as are

the planes of the triazole rings.

The structures of the complexes do vary in the position of

the benzyl ‘arms’ relative to the plane of the carbazole moiety

(Figure 6). In [Co^III(L^tBu/tBu)₂]⁺, [Co^III(L^H/H)₂]⁺and [Co^III(L^H/tBu)₂]⁺,

the arms of one ligand are orientated towards the same side of

the plane of the carbazole moiety, while in the other ligand the

arms are pointing to opposite sides of the plane. However, in

[Co^III(L^tBu/H)₂]⁺, the benzyl arms of both ligands are orientated

Figure 5. Perspective view of the cation of [Co^III(L^H/H)₂]BF₄∙0.5DMF∙0.5H₂O. For clarity,

solvent molecules, the counter anion and hydrogen atoms have been omitted. See SI,

Figure S58, for figures showing the structures of the other three complexes.

This journal is © The Royal Society of Chemistry 20xx

Dalton Trans., 2017, 00, 1-3 | 5

Please do not adjust margins

Dalton Transactions

Page 6 of 15

View Article Online

DOI: 10.1039/C7DT00295E

ARTICLE

Dalton Transactions

Figure 6. Comparison of benzyl arm positions found in the four cobalt(III) complexes:

trans-trans propeller (left, [Co^III(L^tBu/H)₂]⁺) and cis-trans (right, [Co^III(L^tBu/tBu)₂]⁺,

[Co^III(L^H/H)₂]⁺and [Co^III(L^H/tBu)₂]⁺).

Electrochemistry

Cyclic voltammetry was carried out on the family of four

cobalt(III) complexes in MeCN versus Ag/AgNO₃. The solubility

of the present set of complexes is not optimal, with

concentrations employed in these cyclic voltammograms (CVs)

ranging from 0.3 - 0.8 mmol L^-1, rather lower than the 1 mmol

L^-1usually employed. Nevertheless, multiple processes were

readily observed in the CVs between -1.02 and 0.80 V,

including two reversible redox events at positive potentials for

each complex (Table 2, Figure 7). These are tentatively

assigned to successive single electron oxidations [as ΔE ~

ΔE(Fc/Fc⁺)] of each ligand to form carbazole-based cation

radicals. At negative potentials an irreversible or quasi-

reversible reduction event (Table 2, Figure 8), with weaker

return waves, was also observed and is tentatively assigned to

the Co^III/Co^IIredox couple. Due to the proximity of the two

positive potential redox events, scan rate studies were carried

out on the pair of them (Figure 9 and ESI, Figures S2, S7, S10).

As expected for reversible processes,^38,

Figure 7. Cyclic voltammograms to positive potentials for [Co^III(L^tBu/H)₂]BF₄∙2H₂O (1)

(pink, 0.4 mmol L^-1although not fully dissolved), [Co^III(L^tBu/tBu)₂]BF₄∙H₂O (2) (blue, 0.4

mmol L^-1), [Co^III(L^H/H)₂]BF₄∙0.5H₂O (3) (red, 0.3 mmol L^-1) and [Co^III(L^H/tBu)₂]BF₄∙2H₂O (4)

(black, 0.8 mmol L^-1) in MeCN (0.1 mol L^-1TBAPF versus 0.01 mol L^-1Ag/AgNO₃) at 100

mV s^-1scan rate. The Fc/Fc⁺redox couples occurred at E_m= 0.08±0.01V with ΔE =

0.07±0.01 V.

39

no change was

observed in the observed E_paand E_pcvalues with respect to

scan rate (Figure 9 and ESI, Tables S1-S4).

Table 2: Summary of cyclic voltammetry data for the cobalt(III) complexes, 1-4, of the

four carbazole ligands.^a

The first reversible redox process occurs at E_m(ΔE) of

0.36(0.05), 0.38(0.07), 0.48(0.07), 0.52(0.07) V for complexes

, respectively. The second reversible process occurs close to

the first, at E_m(ΔE) of 0.56(0.07), 0.59(0.07), 0.68(0.11),

0.72(0.08) V for , respectively.

Comparisons of the CVs of complex

complex with indicates that, as expected, the addition of a

1

Complex^b

Reduction, E_pa

(E_pc) /V

Oxidation 1,

E_m/V,

Oxidation 2,

E_m/V,

–

4

R = Cz substituents

R²= Bn substituents

(ΔE /mV)

1 – 4

(1) R = ^tBu R²= H

(2) R = ^tBu R²= ^tBu

(3) R = H R²= H

(4) R = H R²= ^tBu

-1.02 (-0.48)

-1.00 (-0.59)

-0.85 (-0.76)

-0.91 (-0.63)

0.36 (51)

0.38 (65)

0.48 (76)

0.52 (66)

0.56 (72)

0.59 (65)

0.68 (115)

0.72 (84)

1 with 2 and of

3

4

tert-butyl group to the 4-position of the benzyl ‘arm’, a

relatively remote position, has a negligible effect on the

carbazole redox processes: the two reversible redox events

shift by only 0.02-0.04 V, to slightly more positive potentials. In

^aData collected in MeCN (0.1 mol L^-1TBAPF versus 0.01 mol L^-1Ag/AgNO₃) at 100

mV s^-1scan rate. The Fc/Fc⁺redox couple occurred at E_m= 0.08±0.01V with ΔE =

0.07±0.01 V. The processes at positive potentials were examined in separate CVs

to those at negative potentials. ^bSee Scheme 1 for substituent positions of R and

R².

contrast, comparisons of the CVs of

1 with 3 and of 2 with 4

reveal that the addition of the tert-butyl groups to the 3,6-

positions of the carbazole unsurprisingly has a much greater

effect on the carbazole-based redox processes: the two

reversible redox events occur at 0.12-0.14 V lower potentials,

consistent with the expected inductive electron donation of

these tert-butyl groups making the oxidation of the carbazole

moieties easier.

6 | Dalton Trans., 2017, 00, 1-3

This journal is © The Royal Society of Chemistry 20xx

Please do not adjust margins

Dalton Transactions

Page 7 of 15

View Article Online

DOI: 10.1039/C7DT00295E

Dalton Transactions

ARTICLE

Figure 9. Cyclic voltammograms of [Co^III(L^tBu/tBu)₂]BF₄∙H₂O (2) in MeCN from 0 → +1.1 →

0 V versus 0.01 mol L^-1Ag/AgNO₃at scan rates 25, 50, 100, 200 and 400 mV s^-1

.

Figure 8. Cyclic voltammograms to negative potentials for [Co^III(L^tBu/H)₂]BF₄∙2H₂O (1)

(pink, 0.4 mmol L^-1although not fully dissolved), [Co^III(L^tBu/tBu)₂]BF₄∙H₂O (2) (blue, 0.4

mmol L^-1), [Co^III(L^H/H)₂]BF₄∙0.5H₂O (3) (red, 0.3 mmol L^-1) and [Co^III(L^H/tBu)₂]BF₄∙2H₂O (4)

(black, 0.8 mmol L^-1) in MeCN (0.1 mol L^-1TBAPF versus 0.01 mol L^-1Ag/AgNO₃) at 100

mV s^-1scan rate. The Fc/Fc⁺redox couples occurred at E_m= 0.08±0.01V with ΔE =

0.07±0.01 V.

The quasi-/irreversible reduction process observed for each

complex occurred at E_pa[E_pc] = -1.02[-0.48], -1.00[-0.59], -0.85[-

0.76], -0.91 [-0.63] for complexes

1 – 4 respectively (Table 1).

Scan rate studies of these processes show that both E_paand E_pc

values shift significantly as scan rates increase (Figures 11, S4,

S6, S8 and S11, Tables S1-4). The quasi-reversible processes of

the 3,6-unsubstituted carbazole complexes,

3 and to a lesser

Figure 10. Cyclic voltammograms of [Co^III(L^H/H)₂]BF₄∙0.5H₂O (3) in MeCN, from 0 → −1.1

→ 0 V versus 0.01 mol L^-1Ag/AgNO₃for 100, 200 and 400 mV s^-1scan rates, and −0.5 →

-1 → -0.5 V for 25 and 50 mV s^-1scan rates.

extent 4, become more reversible when isolated from the

positive potential redox events (Figure 10 and Figure S1 in ESI,

cf. Figures S8 and S11 in ESI): whilst the E_pcand E_papeaks of

this quasi-reversible process move further apart at higher scan

rates (from 25 to 400 mVs^-1: ΔE = 79 to 148 mV for

to 430 mV for ), the E_mvalues remain relatively stable (-0.81

V for and -0.77 V for ; Tables S3-4). In contrast, the

reduction process for each of the 3,6-di-tert-butyl substituted

3 and 147

Conclusions

4

A detailed multi-step route to families of new tridentate

carbazole-based ligands is reported, with four such ligands

successfully prepared, and the analytically pure cobalt(III)

complexes of them obtained. X-ray structural characterization

of all four complexes confirmed that, as expected, the

complexes are isostructural, with two ligands binding in a

3

4

carbazole complexes,

S3 and S5).

1 and 2, remains irreversible (ESI Figures

1

tridentate fashion to an octahedral cobalt(III) center. H NMR

This journal is © The Royal Society of Chemistry 20xx

Dalton Trans., 2017, 00, 1-3 | 7

Please do not adjust margins

Dalton Transactions

Page 8 of 15

View Article Online

DOI: 10.1039/C7DT00295E

ARTICLE

Dalton Transactions

studies indicate that this motif is maintained in solution. Cyclic X-Ray crystallographic data were in all cases collected on an

voltammetry showed similar redox behaviour for all four Oxford Diffraction SuperNova diffractometer equipped with a

complexes, with two reversible, single-electron oxidation Cryostream N₂open-flow cooling device at 100 K using

events tentatively assigned to radical formation on each of the graphite monochromated Mo or Cu Kα radiation and an Atlas

carbazole ligands. An irreversible or quasi-reversible reduction CCD (Table 3). Scans were performed in such a way as to

event was also observed, which was tentatively assigned as collect a complete set of unique reflections to a maximum

the Co(III)/Co(II) couple. The positions of the reversible redox resolution of 0.80 Å. Raw frame data (including data reduction,

processes were able to be tuned by up to 140 mV (Table 2): inter-frame scaling, unit cell refinement and absorption

introducing electron releasing tert-butyl groups to the 3 and 6 corrections) for all structures were processed using CrysAlis

positions of the carbazole dropped the redox potential down Pro.⁴⁰

to 0.36 V versus 0.01 mol L^-1Ag/AgNO₃.

Structures were solved and refined using SIR92,⁴¹

This family of robust carbazole-based N₃ligands has great SUPERFLIP⁴², SHELXL-97⁴³and the CRYSTALS⁴⁴suite. Hydrogen

potential. The resulting cobalt(III) complexes are the first of atoms were visible in the Fourier difference maps for three

such carbazole ligands, and they exhibit reversible and complexes (not [Co^III(L^H/tBu)₂]BF₄

·1.5Et₂O∙solvent), and were

tuneable redox behavior at potentials close to that of the initially refined using restraints on bond lengths and positions,

Fc/Fc⁺couple. In future studies the tuneability of the redox with these initial positions then used as the basis for the riding

potential will be further probed: the use of 3,6- substituents on model⁴⁵(except for the O-H hydrogen atoms). For

the carbazole ring that have more pronounced electron [Co^III(L^H/tBu)₂]BF₄

·1.5Et₂O∙solvent, hydrogen atoms were

releasing properties is expected to bring the redox potentials inserted at calculated positions and for all complexes the C-H

down into the region of interest for use as redox electrolytes in hydrogen atoms rode on the atoms to which they were bound,

DSSCs. As expected, this study confirms that the benzyl with

U(H)

=

1.2

U(attached

atom).

For

substituents off the triazole rings have little effect on the redox [Co^III(L^H/H)₂]BF₄∙0.5H₂O∙0.5DMF, hydrogen atoms on the

potentials, so these substituents will also be varied more oxygen atom of the water molecule were located from the

widely, but with the aim of enhancing the solubility of the difference map and refined using restraints on bond lengths

resulting complexes.

and angles. PLATON-SQUEEZE³³was used in the case of

[Co^III( ^tBu/tBu)₂]BF₄∙1.5Et₂O∙solvent

and

L

[Co^III(L^H/tBu)₂]BF₄∙1.5Et₂O∙solvent, where the anion and solvent

molecules listed were not excluded prior to treating the

structure with SQUEEZE.

Experimental

General

Crystals of [Co^III(L^tBu/H)₂]BF₄were twinned, but an

appropriate twin law was found using ROTAX.⁴⁶The BF₄anion

was disordered over two sites, both situated on a four-fold

rotation axis. This disorder was modelled using two partial

occupancy sites (0.75:0.25) which required restraints on bond

lengths, angles, and thermal and vibrational ellipsoid

parameters to reach a sensible refinement.

Where applicable, MeCN, toluene and DCM were dried using a

Pure Solv MD-6 Solvent Purification System, other solvents and

reagents were reagent grade, unless otherwise specified, and

used as received from commercial suppliers. Dry triethylamine

was distilled and stored over KOH pellets. The EDTA solution

used was made as follows: EDTA (29 g, 0.10 mol) stirring in

aqueous ammonia solution (37%, 100 mL) was diluted to 1 L

using distilled H₂O.

The structure of [Co^III( ^tBu/tBu)₂]BF₄∙1.5Et₂O∙solvent had one

L

half of an Et₂O sitting about a special position, and one tert-

butyl group twirl-disordered, which was modelled over two

sites (0.65:0.35) with constraints on C-C bond length.

Restraints were also applied to bonds of the solvent

molecules, and the thermal and vibrational ellipsoid

parameters of the solvent molecules, several tert-butyl groups,

and the BF₄anion. There was an additional region of electron

density, appearing to correspond to disordered partial

occupancy Et₂O and MeCN solvent molecules, which was

unable to be sensibly modelled, so SQUEEZE was applied. The

SQUEEZE result gave 26 e^-per complex molecule, which

reasonably matches one MeCN (22 e^-per complex); a formula

of [Co^III(L^tBu/tBu)₂]BF₄∙1.5Et₂O∙MeCN would be consistent with

this.

Elemental analyses (C, H, N and Br) were carried out by

Campbell Microanalytical Laboratory at the University of

Otago, with a standard absolute error of ±0.3%. NMR spectra

were collected on Varian 400-MR or 500 MHz VNMRS

spectrometers at 298 K. ESI-MS spectra were collected on a

Bruker MicrOTOF_Qmass spectrometer at 180°C.

Electrochemistry

Electrochemical measurements were performed using an

Iviumstat.XRe potentostat in MeCN freshly distilled over

calcium hydride. Experiments were carried out on the family of

four cobalt(III) complexes 1-4 (0.3-0.8 mmol L^-1, variation due

to solubility) in a 0.1 mol L^-1NBu₄PF₆MeCN solution versus a

0.01 mol L^-1AgNO₃/Ag reference electrode. The Fc/Fc⁺redox

couple was used as a reference check for all studies, and it

occurred at E_m= 0.08±0.01 V with ΔE = 0.07±0.01 V.

For [Co^III(L^H/H)₂]BF₄∙0.5H₂O∙0.5DMF the thermal ellipsoids

of the benzyl groups are somewhat enlarged and elongated,

likely due to significant motion; unfortunately an alternative

positon was not visible.

The data of [Co^III(L^H/tBu)₂]BF₄

·1.5Et₂O∙solvent are of

X-ray crystallography

relatively low quality, and numerous restraints were required

8 | Dalton Trans., 2017, 00, 1-3

This journal is © The Royal Society of Chemistry 20xx

Please do not adjust margins

Dalton Transactions

Page 9 of 15

Dalton Transactions

View Article Online

DOI: 10.1039/C7DT00295E

ARTICLE

to achieve a sensible result. One of the Et₂O molecules was applied to bond lengths and angles of the solvent, BF₄anion,

located on a symmetry position, causing only half of it to lie in tert-butyl groups, and some of the benzyl rings to maintain a

the asymmetric unit. This Et₂O (0.5:0.5) and the BF₄anion sensible refinement. Thermal and vibrational ellipsoid

(0.6:0.4) were both disordered over two sites. Restraints were

Table 3: Crystal data and structure refinement for the four cobalt(III) complexes. Note that [Co^III(L^tBu/tBu)₂]BF₄· 1.5Et₂O·solvent and [Co^III(L^H/tBu)₂]BF₄·1.5Et₂O·solvent had SQUEEZE

applied, so M_ris inaccurate.

[Co^III(L^tBu/H)₂]BF₄

[Co^III(L^tBu/tBu)₂]BF₄∙

1.5Et₂O∙solvent

C₉₈H₁₂₃BCoF₄N₁₄O_1.5

1666.90

[Co^III(L^H/H)₂]BF₄∙

0.5H₂O∙0.5DMF

C₁₂₃H₉₇B₂Co₂F₈N₂₉O₂

2304.80

[Co^III(L^H/tBu)₂]BF₄·

1.5Et₂O∙solvent

C₈₂H₉₁BCoF₄N₁₄O_1.5

1442.44

Empirical formula

M_r

C₇₆H₇₆BCoF₄N₁₄

1331.27

Cu: 1.54184

Tetragonal

P4/n

Wavelength [Å]

Crystal system

Space group

a [Å]

Mo: 0.71073

Monoclinic

P2₁/c

Mo: 0.71073

Monoclinic

P2₁/n

Cu: 1.54184

Triclinic

P-1

16.32274(18)

13.22212(18)

90

16.6488(4)

20.3208(6)

28.7288(11)

90

17.3437(3)

24.7996(4)

26.5881(5)

90

15.7973(3)

16.5287(4)

18.5432(4)

72.8513(19)

74.9270(19)

67.484(2)

4214.09(9)

2

b [Å]

c [Å]

α

β

γ

[°]

90

94.014(3)

90

107.976(2)

90

Volume [Å³]

3522.79(5)

2

9695.6(4)

4

10877.7(2)

4

Z

Density (calcd) [Mg m^ꢀ3]

Abs. coeff.[mm^-1]

F(000)

1.255

1.142

1.407

1.137

2.426

0.237

0.388

2.076

1396

3556

4760

1522

Crystal size [mm]

θ range for data collection

[°]

0.45 x 0.40 x 0.35

3.342 to 76.224

0.42 x 0.19 x 0.13

2.927 to 29.747

0.51 x 0.30 x 0.21

2.867 to 29.748

0.64 x 0.58 x 0.39

3.071 to 77.217

Reflections collected

Indpt reflections

R(int)

19255

3681

98773

23829

175733

28282

83092

17513

0.052

0.061

0.029

0.064

Max/min transmission

Data/restraints

/parameters

1.00 / 0.63

3664 / 94 / 237

1.00 / 0.59

17264 / 190 / 1114

1.00 / 0.83

27139 / 2 / 1501

1.00 / 0.63

13732 / 1114 / 979

2

Goodness-of-fit on F

1.0276

0.9664

0.9564

0.9790

R₁/wR₂[I>2σ(I)]

R₁/wR₂(all data)

Max/Min e^-density [eÅ^ꢀ3]

0.0619 / 0.1725

0.0673 / 0.1809

1.44 / -1.09

0.0870 / 0.2103

0.1029 / 0.2234

1.74 / -0.79

0.0582 / 0.1409

0.0696 / 0.1517

1.02 / -0.67

0.1173 / 0.3113

0.1361 / 0.3408

1.07 / -0.92

restraints were also required for all atoms. Two SQUEEZE See Scheme

refinements were carried out, with one applying to all solvent assignments.

1

for numbering schemes used for NMR

electron density and the other only the diffuse residual

electron density. The full SQUEEZE did not lead to

3,6-Di(tert-butyl)-9H-carbazole was prepared according to

a literature procedure,²⁰with the substitution of cyclohexane

a

significantly improved refinement, hence is not presented. The for the hexane recrystallisation (see SI).

SQUEEZE results gave 78 e^-per complex, which was difficult to

1,8-dibromo-3,6-di(tert-butyl)-9H-carbazole. To

a

pale

assign, although it was consistent with either three Et₂O (75 e^-orange mixture of 3,6-di(tert-butyl)-9H-carbazole (1.96 g, 7.01

per complex) or an Et₂O, H₂O and pyridine (77 e^-per complex). mmol) and silica (6.0 g) in dry DCM (75 mL) was added N-

Despite the issues, all non-H atoms were able to be refined bromosuccinimide (2.49 g, 14.0 mmol). The resulting dark

anisotropically, and no restraints were applied to the bonds of brown mixture was stirred at RT under light-free conditions for

the carbazole/triazole parts, so the connectivity is still clear 3 days. Filtration of the reaction mixture gave a dark grey

and general conclusions about bonding can be derived.

solution, which was passed through a silica plug. Elution with

Crystallographic data for the structures have been DCM yielded a pink-orange solution, which was dried over

deposited with the Cambridge Crystallographic Data Centre, MgSO₄, filtered and taken to dryness under reduced pressure

CCDC 1510199-1510202.

to provide the title compound as a pale grey solid (2.93 g, 6.70

mmol, 96%). Analysis found: C, 55.16; H, 5.31; Br, 36.54; N,

3.21. Calc. for C₂₀H₂₃Br₂N: C, 54.94; H, 5.30; Br, 36.55; N, 3.20.

Synthesis

ESI-MS (neg., CHCl₃/MeOH) m/z

= 436.008 (calc. for

This journal is © The Royal Society of Chemistry 20xx

Dalton Trans., 2017, 00, 1-3 | 9

Please do not adjust margins

Dalton Transactions

Page 10 of 15

View Article Online

DOI: 10.1039/C7DT00295E

ARTICLE

Dalton Transactions

[C₂₀H₂₂Br₂N]^-436.011). ¹H NMR (400 MHz, CDCl₃) δ (ppm) = 2962, 2899, 2865, 2150, 1589, 1490, 1290, 1273, 1249, 1238,

8.13 (br s, 1H, N ), 7.97 (dd, 2H, J = 1.6, 0.6 Hz, H₄) 7.63 (d, 2H, 1202, 1047, 929, 838, 759, 674, 645, 564, 460.

J = 1.7 Hz, H₂), 1.44 (s, 18H, H₅). IR (ATR), inter alia, ν (cm^-1) =

1,8-bis[(trimethylsilyl)ethynyl]-9H-carbazole.

3451, 2960, 2901, 2863, 1719, 1616, 1562, 1491, 1361, 1282, suspension of 1,8-dibromo-9H-carbazole (1.46 g, 4.49 mmol),

1264, 1232, 1889, 1053, 862, 843, 729, 665, 636. CuI (0.05 g, 0.27 mmol) and Pd(PPh₃)₂Cl₂(0.16 g, 0.23 mmol) in

H

To

a

1,8-dibromo-9H-carbazole. To a pale gold solution of 1,8- dry triethylamine (23 mL) was added ethynyltrimethylsilane

dibromo-3,6-di(tert-butyl)-9H-carbazole (1.09 g, 2.49 mmol) in (1.5 mL, 10.8 mmol). The reaction mixture was deoxygenated

dry toluene (50 mL) was added anhydrous AlCl₃(1.00 g, 7.50 with argon, then stirred at 75 °C under an argon atmosphere

mmol). The resulting dark brown reaction mixture was stirred for 2 days. The resulting dark brown mixture was diluted with

at RT for 8.5 h. The reaction mixture became dark green within petroleum ether (50 mL), then filtered through celite. The

the first 30 min. Complete conversion to the desired product celite plug was washed with petroleum ether (200 mL). The

was indicated by t.l.c. (5:1 petroleum ether:EtOAc). The resulting yellow-orange solution was taken to dryness under

reaction mixture was washed with H₂O (2 × 50 mL) and the reduced pressure to yield dark brown oil, which was subjected

aqueous washes were combined and extracted into DCM. The to column chromatography on silica. Elution (R_f= 0.24) with

combined organic extracts were dried over MgSO₄, filtered and petroleum ether isolated the title compound as a pale yellow

taken to dryness under reduced pressure to yield a brown solid (1.35 g, 84%). Analysis found: C, 73.69; H, 7.24; N, 3.83.

waxy oil (1.32 g). The crude residue was dissolved in diethyl Calc. for C₂₂H₂₅NSi₂: C, 73.48; H, 7.01; N, 3.89. ESI-MS (neg.,

ether (12 mL) to give an orange solution, which was MeOH): m/z = 358.139 (calc. for [C₂₂H₂₄NSi₂]: 358.145). ¹H

concentrated to 3 mL and cooled in an ice bath to yield a pale NMR (400 MHz, CDCl₃) δ (ppm) = 8.57 (br s, 1H, NH), 8.02 (d,

brown precipitate. The title compound was isolated via 2H, J = 7.8 Hz, H₄), 7.56 (dd, 2H, J = 7.5, 1.1 Hz, H₂), 7.20 (dd,

filtration as an off-white solid (0.28 g). Treating the filtrate to 2H, J = 7.7, 7.7 Hz, H₃), 0.37 (s, 18H, H₅). IR (ATR), inter alia, ν

the same concentration/cooling procedure provided a second (cm^-1) 3460, 3060, 2958, 2149, 1598, 1497, 1409, 1580, 1298,

crop of product (0.13 g, combined yield: 0.41 g, 50%). ESI-MS 1260, 1251, 1226, 1162, 1004, 840, 833, 771, 740, 708, 662,

(neg., MeOH): m/z = 323.892 (calc. for [C₁₂H₆Br₂N]^-323.885). 564.

¹H NMR (400 MHz, CDCl₃) δ (ppm) = 8.36 (br s, 1H, N

H

) 7.99 (d,

3,6-di(tert-butyl)-1,8-diethynyl-9H-carbazole.

A

2H, J = 7.8 Hz, H₄), 7.61 (dd, 2H, J = 7.8, 0.9 Hz, H₂), 7.16 (dd, suspension of 3,6-di(tert-butyl)-1,8-bis[(trimethylsilyl)ethynyl]-

2H, J = 7.8, 7.8 Hz, H₃). ¹³C NMR (101 MHz, CDCl₃): δ 137.9, 9H-carbazole (1.42 g, 3.01 mmol) in MeOH (60 mL) was

128.9, 125.0, 121.4, 120.0, 104.6. IR (ATR), inter alia, ν (cm^-1) = deoxygenated with argon. K₂CO₃(2.49 g, 18.0 mmol) was

3454, 3068, 2959, 2864, 1906, 1844, 1785, 1604, 1560, 1495, added and the resulting yellow suspension was stirred at reflux

1477, 1424, 1298, 1261, 1228, 1182, 830, 755, 724, 679. R_f= under an argon atmosphere for 30 min. The reaction mixture

0.73 (5:1 petroleum ether:EtOAc).

was diluted with H₂O (60 mL) then extracted into petroleum

ether (3 × 100 mL). The combined organic extracts were dried

3,6-di(tert-butyl)-1,8-bis[(trimethylsilyl)ethynyl]-9H-

carbazole. 1,8-dibromo-3,6-di(tert-butyl)-9H-carbazole (3.85 g, over MgSO₄, filtered and taken to dryness under in vacuo to

8.80 mmol) in dry triethylamine (25 mL) was deoxygenated give the product as an off-white foam in quantitative yield.

with argon for 20 minutes. Afterwards, [Pd(PPh₃)₂Cl₂] (0.368 g, Analysis found: C, 87.77; H, 7.77; N, 4.18. Calc. for C₂₄H₂₅N: C,

0.524 mmol) and CuI (0.139 g, 0.731 mmol) were added, 88.03; H, 7.70; N, 4.28. ESI-MS (pos., MeOH): m/z = 328.208

followed by

a

further

5

minutes of argon bubbling. (calc. for [C₂₄H₂₆N]⁺328.206). ¹H NMR (400 MHz, CDCl₃) δ

), 8.08 (d, 2H, J = 1.9 Hz, H₄), 7.64 (d,

Ethynyltrimethylsilane (1.99 g, 20.3 mmol) was added, the vial (ppm) = 8.45 (br s, 1H, N

H

capped, and the mixture heated in an 80 °C oil bath for 1 day, 2H, J = 1.8 Hz, H₂), 3.46 (s, 2H, H₅), 1.44 (s, 18H, H₆). ¹³C NMR

then left stirring overnight to cool. The mixture was diluted (126 MHz, CDCl₃) δ (ppm) = 142.9, 139.3, 127.7, 123.3, 118.1,

with petroleum ether (150 mL), filtered through celite using 104.0, 81.5, 80.6, 34.9, 32.0. IR (ATR), inter alia, ν (cm^-1) =

petroleum ether (5 × 40 mL) to rinse, then solvents were 3456, 3287, 2954, 2902, 2864, 2104, 1761, 1601, 1489, 1435,

removed in vacuo to give a caramel coloured solid (4.71 g). The 1396, 1362, 1324, 1289, 1271, 1237, 1202, 919, 838, 749, 644.

solid was dissolved in minimum petroleum ether and filtered

1,8-diethynyl-9H-carbazole.

A

suspension of 1,8-

through silica (50 mL) using petroleum ether to elute. The first bis[(trimethylsilyl)ethynyl]-9H-carbazole (0.43 g, 1.5 mmol) in

pale yellow fraction was collected until further elution was MeOH (25 mL) was deoxygenated with argon. K₂CO₃(1.00 g,

colourless, then the solvents removed in vacuo to give a pale 7.2 mmol) was added and the resulting yellow suspension was

yellow solid (3.68 g, 89%). Analysis found: C, 76.23; H, 8.74; N, stirred at reflux under an argon atmosphere for 30 min. The

2.92. Calc. for C₃₀H₄₁NSi₂(471.84 g mol^-1): C, 76.37; H, 8.76; N, reaction mixture was diluted with H₂O (25 mL) then extracted

2.97. ESI-MS (pos., DCM/MeOH) m/z: 472.282 (calc. for into DCM (3 × 50 mL). The combined organic extracts were

[C₃₀H₄₂NSi₂]⁺= 472.285); 494.266 (calc. for [C₃₀H₄₁NSi₂Na]⁺

494.267). ¹H NMR (400 MHz, CDCl₃) δ (ppm) = 8.34 (s, 1H, N

=

dried over MgSO₄filtered and taken to dryness in vacuo to give

H

), the product as a pale pink-brown solid (0.25 g, 97%). Analysis

8.04 (d, 2H, J = 1.5 Hz, H₄), 7.60 (d, 2H, J = 1.8 Hz, H₂), 1.44 (s, found: C, 88.47; H, 4.32; N, 6.59. Calc. for C₁₆H₉N: C, 89.28; H,

18H, H₆), 0.35 (s, 18H, H₅). ¹³C NMR (126 MHz, CDCl₃) δ (ppm) 4.21; N, 6.51. Calc. for C₁₆H₉N∙0.03DCM: C, 88.40; H, 4.19; N,

= 142.6, 139.0, 126.8 (C₂), 123.1, 117.7 (C₄), 104.9, 101.4, 98.6, 6.43. ¹H NMR (400 MHz, CDCl₃): δ (ppm) = 8.64 (br s, 1H, N

H),

34.7, 31.9 (C₆), 0.21 (C₅). IR (ATR), inter alia, ν (cm^-1) = 3468, 8.07 (d, 2H, J = 7.9 Hz, H₄), 7.60 (dd, 2H, J = 7.6, 1.1 Hz, H₂),

7.22 (dd, 2H, J = 7.7, 7.7 Hz, H₃), 3.50 (s, 2H, H₅). ¹³C NMR (126

10 | Dalton Trans., 2017, 00, 1-3

Please do not adjust margins

Dalton Transactions

Page 11 of 15

Dalton Transactions

View Article Online

DOI: 10.1039/C7DT00295E

ARTICLE

MHz, CDCl₃) δ (ppm) = 140.7, 129.89, 123.3, 121.7, 119.9, Calculated for C₄₆H₅₅N₇: C, 78.26; H, 7.85; N, 13.89. ESI-MS

104.9, 82.3, 79.9. IR (ATR), inter alia, ν (cm^-1) = 3366, 6275, (pos., CHCl₃/MeOH) m/z: 728.4363 (calc. for [C₄₆H₅₅N₇Na]⁺

=

1

3058, 2918, 2100, 1596, 1579, 1495, 1419, 1333, 1297, 1236, 728.4411). H NMR (400 MHz, CDCl₃) δ (ppm) = 11.70 (s, 1H,

1221, 777, 742, 656, 617, 568.

NH), 8.10 (d, 2H, J = 1.3 Hz, H₄), 7.96 (s, 2H, H₆), 7.67 (d, 2H, J =

HL^tBu/H

.

To a suspension of ascorbic acid (0.106 g, 0.6 1.5 Hz, H₂), 7.43 (d, 4H, J = 8.3 Hz, H₁₀), 7.33 (d, 4H, J = 8.3 Hz,

mmol), CuSO₄∙5H₂O (0.075 g, 0.3 mmol), Na₂CO₃(0.212 g, 2.0 H₉), 5.64 (s, 4H, H₇), 1.47 (s, 18H, H₁₅), 1.32 (s, 18H, H₁₃). ¹³C

mmol) and NaN₃(0.150 g, 2.3 mmol) in DMF/H₂O (4:1, 3 mL) NMR (101 MHz, CDCl₃) δ (ppm) = 151.8 (C₁₁), 147.9, 141.7,

was added benzyl bromide (0.27 mL, 2.3 mmol). The resulting 135.7, 131.9, 127.8, 126.0, 123.8, 120.2, 119.1, 116.3, 112.7,

suspension was stirred at RT for 5 min, during which time it 54.0 (C₇), 34.7 (C₁₂), 34.6 (C₁₄), 32.1 (C₁₃), 31.3 (C₁₅).

turned yellow. To this suspension was added 3,6-di(tert-butyl)-

HL^H/H

. To a suspension of ascorbic acid (0.11 g, 0.6 mmol),

1,8-diethynyl-9H-carbazole (0.327 g, 1.0 mmol) in DMF/H₂O CuSO₄∙5H₂O (0.075 g, 0.3 mmol), Na₂CO₃(0.21 g, 2.0 mmol)

(4:1, 1 mL). The resulting red-orange mixture was stirred at RT and NaN₃(0.15 g, 2.3 mmol) in DMF/H₂O (4:1, 3 mL) was

for 16 h, by which time a pale orange precipitate had formed. added benzyl bromide (0.27 mL, 2.3 mmol). The resulting

The reaction mixture was diluted with EDTA solution (100 mL) suspension was stirred at RT for 5 min, during which time it

and stirred at RT for 1 h. An orange-brown solid was collected turned yellow. To this suspension was added 1,8-diethynyl-9H-

via filtration, washed with H₂O (200 mL) then dissolved in DCM carbazole (0.22 g, 1.0 mmol) in DMF/H₂O (4:1, 1 mL). The

(100 mL). The resulting dark orange solution was washed with resulting dark yellow mixture was stirred at RT for 20 h, by

water (100 mL), and the aqueous layer extracted into DCM which time an off-white precipitate had formed with a yellow

(100 mL). The combined organic extracts were taken to supernatant. The reaction mixture was diluted with EDTA

dryness under reduced pressure to yield an orange-brown solution (100 mL) and stirred at RT for 1 h. An off-white solid

solid. Recrystallisation of the crude residue from toluene (35 was collected via filtration, washed with H₂O (200 mL) and

mL) provided HL^tBu/Has a pale brown solid (0.359 g, 60%). dried in vacuo (0.32 g). The crude residue was subjected to

Analysis found: C, 76.90; H, 6.78; N, 16.27. Calc. for C₃₈H₃₉N₇: column chromatography on silica. Elution (R_f= 0.18) with 5%

C, 76.87; H, 6.62; N, 16.51. ESI-MS (pos., DCM/MeOH): m/z = EtOAc in CHCl₃/toluene (1:1) provided HL^H/Has a white solid

594.3336 (calc. for C₃₈H₄₀N₇594.3340). ¹H NMR (500 MHz, (0.20 g, 41%). Analysis found: C, 74.73; H, 4.79; N, 20.20. Calc.

CDCl₃) δ (ppm) = 11.75 (s, 1H, N

7.94 (s, 2H, H₆), 7.66 (d, J = 1.8 Hz, 2H, H₂), 7.45 - 7.36 (m, 10H, m/z: 504.1886 (calc. for [C₃₀H₂₃N₇Na]⁺504.1907). ¹H NMR (500

H_9/10/11), 5.68 (s, 4H, H₇), 1.47 (s, 18H, H₁₅). ¹³C NMR (126 MHz, MHz, CDCl₃): δ 12.21 (br s, 1H, N

) 8.08 (d, J = 7.7 Hz, 2H, H₄),

H) 8.10 (d, J = 1.8 Hz, 2H, H₄), for C₃₀H₂₃N₇: C, 74.83; H, 4.81; N, 20.36. ESI-MS (pos., MeOH)

H

CDCl₃) δ (ppm) = 148.2, 141.9 (C₁₂), 135.9, 135.1 (C₈), 129.3 7.91 (s, 2H, H₆), 7.61 (dd, J = 7.5, 1.1 Hz, H₂), 7.43 - 7.35 (m,

(C_9/10/11), 128.9 (C_9/10/11), 128.3 (C_9/10/11), 124.0, 120.4 (C₂), W_h/2= 8.4 Hz, H_9/10/11), 7.24 (dd, J = 7.7, 7.7 Hz, 2H, H₃), 5.67 (s,

119.3 (C₆), 116.3 (C₄), 112.9 (C_4a), 54.5 (C₇), 34.9 (C₁₄), 32.3 4H, H₇). ¹³C NMR (126 MHz, CDCl₃) = δ 147.8 (C₅), 137.2, 134.9

(C₁₅). IR (ATR), inter alia, ν (cm^-1) = 3397, 3129, 3065, 3032, (C₈), 129.3 (C_9/10/11), 128.9 (C_9/10/11), 128.3 (C_9/10/11), 124.1,

2952, 2901, 2865, 1607, 1537, 1485, 1434, 1362, 1288, 1268, 122.6 (C₂), 120.2 (C₄), 119.4 (C₆), 119.1 (C₃), 113.7, 54.5 (C₇). IR

1193, 1052, 1002, 866, 708, 692, 646.

HL^tBu/tBu

A mixture of ascorbic acid (209 mg, 1.19 mmol), 1496, 1437, 1425, 1312, 1267, 1057, 1046, 774, 710, 635, 618,

Na₂CO₃(186 mg, 1.75 mmol), NaN₃(149 mg, 3.82 mmol) and 591, 554.

4-tert-butylbenzyl bromide (868 mg, 3.82 mmol) in DMF:H₂O

HL^H/tBu

(ATR), inter alia, ν (cm^-1) = 3381, 3136, 3037, 2948, 1611, 1590,

.

A mixture of NaN₃(272 mg, 4.18 mmol), ascorbic

(4:1, 4.5 mL) was deoxygenated with N₂bubbling for 35 acid (338 mg, 1.92 mmol), 4-tert-butylbenzyl bromide (0.77

minutes. 3,6-di(tert-butyl)-1,8-diethynyl-9H-carbazole (568 mg, mL, 4.20 mmol), Na₂CO₃(204 mg, 1.92 mmol) and CuSO₄(160

1.73 mmol) and CuSO₄∙5H₂O (142 mg, 0.57 mmol) were added mg, 0.64 mmol) in DMF:H₂O (4:1, 9 mL) was deoxygenated

with DMF:H₂O (4:1, 3 mL) followed by a further 5 minutes of with N₂bubbling for 35 minutes. 1,8-diethynyl-9H-carbazole

N₂bubbling. The orange mixture was then stirred at RT for 3 (417 mg, 1.94 mmol) was added using DMF:H₂O (4:1, 1 mL) to

days under argon atmosphere. EDTA solution (40 mL) was rinse, followed by a further 10 minutes of N₂bubbling. The N₂

added and the resulting mustard-coloured mixture vigorously flow was replaced with a septa and an argon balloon, and then

stirred before adding H₂O (50 mL). The mixture was cautiously the orange-brown mixture was stirred at RT for 2 days. EDTA

extracted with DCM (100 mL, 50 mL), then the combined DCM solution (80 mL) was added and the mixture vigorously stirred

fractions washed with a 1:1 brine:H₂O solution (125 mL, 100 before adding H₂O (100 mL). The greenish suspension was

mL; brine dilution seems to help the layers to separate). The cautiously extracted with DCM (2 × 100 mL, 40 mL), then the

aqueous washes were back-extracted with DCM (100 mL), and extracts combined and washed with a dilute aqueous NaCl

then the DCM fractions combined as an orange solution. The solution (2 × 120 mL). The washes were back-extracted with

DCM solution was dried over MgSO₄, filtered, and then DCM (40 mL), before combining the DCM fractions, drying

solvents removed in vacuo to give the crude product as a them over MgSO₄, filtering, and drying in vacuo to give the

sandy-coloured solid (1.33 g), which was stirred overnight in crude product as a sandy-white coloured solid. The crude

diethyl ether (200 mL). The suspension was filtered to give an product was stirred in diethyl ether (120 mL) for 20 minutes

orange filtrate and an off-white solid, which was dried further then filtered to give a yellow filtrate and an off-white solid,

in vacuo to give microanalytically clean ligand (1.02 g, 1.44 which was dried further in vacuo (899 mg, 78%). The ligand

mmol, 83%). Analysis Found: C, 77.98; H, 7.92; N, 13.76. was used without further purification. ESI-MS (pos.,

Dalton Trans., 2017, 00, 1-3 | 11

Please do not adjust margins

Dalton Transactions

Page 12 of 15

View Article Online

DOI: 10.1039/C7DT00295E

ARTICLE

Dalton Transactions

DCM/MeOH) m/z: 616.307 (calc. for [C₃₈H₃₉N₇Na]⁺616.316). of blue-green crystals (58.2 mg, 31%; 118.7 mg, 63% total).

¹H NMR (400 MHz, CDCl₃) δ (ppm) = 12.14 (s, 1H, N

), 8.06 (d, Analysis found: C, 69.99; H, 7.25; N, 12.18. Calculated for

H

2H, J = 7.7 Hz, H₄), 7.92 (s, 2H, H₆), 7.62 (d, 2H, J = 7.4 Hz, H₂), C₉₂H₁₀₈N₁₄CoBF₄∙H₂O (1573.73 g mol^-1): C, 70.22; H, 7.05; N,

7.43 (d, 4H, J = 8.4 Hz, H₁₀), 7.32 (d, 4H, J = 8.3 Hz, H₉), 7.23 (t, 12.46. ESI-MS (pos., MeCN): m/z = 1467.8270, 1468.8313 (calc.

2H, J = 7.6 Hz, H₃), 5.64 (s, 4H, H₇), 1.32 (s, 18H, H₁₃). ¹³C NMR for [C₉₂H₁₀₈N₁₄Co]⁺([Co(L^tBu/tBu)₂]⁺) = 1467.8208, 1468.8240).

(101 MHz, CDCl₃) δ (ppm) = 151.9 (C₁₁), 147.5, 137.1, 131.7, ¹H NMR (400 MHz, CDCl₃) δ (ppm) = 8.49 (s, 4H, H₄), 7.94 (br. s,

128.0, 126.1, 123.8, 122.4, 120.0, 119.2, 118.8, 113.6, 54.0 4H, H₆), 7.69 (br. s, 4H, H₂), 7.05 (d, 8H, J = 8.3 Hz, H₁₀), 6.53 (d,

(C₇), 34.6 (C₁₂), 31.2 (C₁₃).

[Co^III(L^tBu/H)₂]BF₄·2H₂O (1).

8H, J = 8.1 Hz, H₉), 4.84 (s, 8H, H₇), 1.59 (s, 36H, H₁₅), 1.21 (s,

L^tBu/H(30 mg, 0.050 mmol) was 36H, H₁₃). H NMR (400 MHz, DMSO-d₆) δ (ppm) = 8.99 (s, 4H,

1

H

dissolved in acetone (5 mL) containing triethylamine (5 drops) H₆), 8.53 (d, 4H, J = 1.8 Hz, H₄), 7.86 (d, 4H, J = 1.8 Hz, H₂), 6.94

and the solution deoxygenated with bubbling argon. (d, 8H, J = 8.4 Hz, H₁₀), 6.31 (d, 8H, J = 8.4 Hz, H₉), 4.82 (s, 4H,

Co(H₂O)₆(BF₄)₂(8.5 mg, 0.025 mmol) was dissolved in MeOH (2 H₇), 1.51 (s, 18H, H₁₅), 1.12 (s, 18H, H₁₃). ¹³C NMR (101 MHz,

mL) and the solution deoxygenated with bubbling argon. The DMSO-d₆) δ (ppm) = 150.5 (C₁₁), 144.2 (C₅), 140.8 (C_1a), 138.6

pale pink metal salt solution was added to the yellow ligand (C₃), 131.3 (C₈), 127.6 (C_4a), 126.8 (C₉), 124.9 (C₁₀), 122.5 (C₆),

solution, causing it to darken slightly. The reaction was stirred 118.8 (C₄), 117.5 (C₂), 108.7 (C₁), 54.4 (C₇), 34.6 (C₁₄), 34.1

under a nitrogen balloon for 30 minutes, then opened to the (C₁₂), 32.1 (C₁₅), 31.0 (C₁₃). ¹³C NMR (126 MHz, CDCl₃), inter

air and the dark green solution subjected to diethyl ether alia, δ (ppm) = 151.8 (C₁₁), 130.2, 127.8 (C₉), 125.6 (C₁₀), 55.5

vapour diffusion, which resulted in dark green crystals of (C₇), 34.5 (C₁₄), 32.4 (C₁₅), 31.2 (C₁₃). IR (ATR), inter alia, ν (cm^-1)

[Co^III( ^tBu/H)₂](BF₄), some of which were twinned crystals = 3132 (w), 2954 (s), 2903 (m), 2866 (m), 1668 (w), 1614 (w),

L

suitable for X-ray crystallography. The crystals were isolated 1548 (m), 1512 (w), 1476 (s), 1441 (s), 1362 (s), 1329 (m), 1274

via filtration, washed with diethyl ether, and dried in vacuo (22 (vs), 1229 (vs), ~1100 – 1000 (vs, broad), 1023 (vs), 861 (m),

mg, 64%). Analysis found: C, 66.7; H, 5.9; N, 14.1. Calculated 784 (s), 731 (w), 712 (s), 686 (m), 666 (m), 647 (m), 629 (w),

for C₇₆H₇₆N₁₄CoBF₄∙2H₂O (1367.31 g mol^-1): C, 66.8; H, 5.9; N, 559 (m), 522 (m). Λ_m(MeCN) = 134 Ω^-1cm²mol^-1.

14.3. ESI-MS (pos., MeOH): m/z

=

1243.567 (calc. for

1

[Co^III(L^H/H)₂]BF₄·0.5H₂O (3). Separately, a yellow solution of

H

L^H/H(96 mg, 0.199 mmol) in DMF (6 mL) and a bright pink

[C₇₆H₇₆N₁₄Co]⁺([Co( ^tBu/H)₂]⁺) = 1243.570). H NMR (500 MHz,

L

DMSO-d₆) δ (ppm) = 8.99 (s, 4H, H₆), 8.54 (d, 4H, J = 2 Hz, H₄), solution of Co(H₂O)₆(BF₄)₂(34 mg, 0.100 mmol) in DMF (2 mL)

7.83 (d, 4H, J = 2 Hz, H₂), 7.12 (dd, 4H, J = 8, 8 Hz, H₁₁), 7.00 were deoxygenated with N₂bubbling for 45 minutes. The

(dd, 8H, J = 8, 8 Hz, H₁₀), 6.44 (d, 8H, J = 8 Hz, H₉), 4.90 (s, 8H, cobalt solution was then added via syringe to the ligand

H₇), 1.52 (s, 36H, H₁₅). ¹³C NMR (126 MHz, DMSO-d₆) δ (ppm) = solution, and the resultant amber solution stirred for a further

144.2 (C₅), 140.8 (C_1a), 138.1 (C₃), 134.1 (C₈), 128.2 (C₁₀), 128.0 5 minutes under N₂. Addition of NEt₃(0.28 mL, 2.01 mmol)

(C₁₁), 127.6 (C_4a), 127.2 (C₉), 122.9 (C₆), 118.8 (C₄), 118.4 (C₂), caused an immediate darkening of the solution, progressing

108.6 (C₁), 55.0 (C₇), 34.6 (C₁₄), 32.1 (C₁₅). IR (ATR), inter alia, ν through dark brown to orange and then brown-green over 40

(cm^-1) = 3130 (w), 2950 (m), 2900 (w), 2865 (w), 1588 (w), minutes whilst stirring under N₂. After opening to air the

1547 (m), 1497 (w), 1471 (m), 1455 (m), 1435 (m), 1362 (m), solution slowly turned more intensely green-blue over several

1329 (w), 1274 (vs), 1229 (s), 1202 (m), 1176 (w), ~1100 – 1000 hours; the solution was left stirring overnight. The resultant

(broad, s), 1068 (s), 1052 (s), 1022 (vs), 869 (m), 860 (m), 710 deep blue-green solution was subjected to diethyl ether

(vs), 693 (s), 581 (w), 546 (w), 526 (w). Λ_m(MeCN, at 0.5 mmol vapour diffusion, which yielded X-ray quality single crystals of

L^-1) = 56 Ω^-1cm²mol^-1.

[Co^III(L^H/H)₂]BF₄∙0.5H₂O∙0.5DMF as dark blue blocks (68.4 mg,

[Co^III(L^tBu/tBu)₂]BF₄H₂O (2). A solution of HL^tBu/tBu(169 mg, 61%). The crystals were isolated via filtration and brief air

∙

0.239 mmol) in pyridine (50 mL) was deoxygenated with N₂drying, followed by drying in vacuo overnight. Analysis Found:

bubbling for 30 minutes. To the stirring solution was added C, 64.31; H, 4.01; N, 17.74 %. Calculated for

Co(H₂O)₆(BF₄)₂(41 mg, 0.120 mmol) under N₂flow, followed C₆₀H₄₄N₁₄CoBF₄∙0.5H₂O (1115.85 g mol^-1): C, 64.58; H, 4.06; N,

by a further 5 minutes of N₂bubbling. The N₂flow was 17.57. ESI-MS (pos., MeCN): m/z = 1019.3284 (calc. for

1

switched for a septum and argon balloon, and the red-orange [C₆₀H₄₄N₁₄Co]⁺([Co(L^H/H)₂]⁺) = 1019.3200). H NMR (400 MHz,

solution was then stirred for 3 hours at room temperature. The DMSO-d₆) δ (ppm) = 8.87 (s, 4H, H₆), 8.49 (d, 4H, J = 8.0 Hz, H₄),

red solution was then opened to air and diethyl ether (40 mL) 7.76 (d, 4H, J = 6.9 Hz, H₂), 7.30 (t, 4H, J = 7.5 Hz, H₃), 7.12 (t,

was added before stirring the solution overnight. The solvents 4H, J = 7.5 Hz, H₁₁), 6.99 (t, 8H, J = 7.8 Hz, H₁₀), 6.40 (d, 8H, J =

were removed in vacuo yielding an oily red solid and some off- 8.0 Hz, H₉), 4.92 (s, 8H, H₇). ¹³C NMR (126 MHz, DMSO-d₆) δ

white solid. The red solid was redissolved in MeCN and the (ppm) = 144.2 (C₅), 142.5 (C_1a), 134.4 (C₈), 128.9 (C₁₀), 128.6

greenish-orange mixture filtered to give an off-white solid and (C₁₁), 128.1 (C_4a), 127.8 (C₉), 123.4 (C₆), 122.6 (C₄), 121.6 (C₂),

a greenish-orange filtrate. The filtrate was then subjected to 116.7 (C₃), 109.6 (C₁), 55.4 (C₇). IR (ATR), inter alia, ν (cm^-1) =

diethyl ether vapour diffusion, which yielded X-ray quality 3129 (w), 3032 (w), 1664 (w), 1610 (w), 1578 (w), 1551 (w),

single crystals of [Co^III(L^tBu/tBu)₂]BF₄∙1.5Et₂O∙solvent as blue- 1496 (w), 1478 (m), 1456 (w), 1420 (s), 1343 (m), 1266 (vs),

green plates (60.5 mg, 32%). The crystals were isolated via 1229 (s), 1167 (w), ~1100 – 950 (vs, broad), 938 (m), 774 (vs),

filtration and brief air drying, followed by drying in vacuo 746 (s), 721 (vs), 701 (vs), 626 (m), 571 (s), 518 (m), 473 (w),

before analysis. The filtrate was reduced in volume, and 461 (w), 408 (w). Λ_m(MeCN) = 132 Ω^-1cm²mol^-1.

resubjected to ether vapour diffusion, leading to a second crop

12 | Dalton Trans., 2017, 00, 1-3

Please do not adjust margins

Dalton Transactions

Page 13 of 15

Dalton Transactions

[Co^III(L^H/tBu)₂]BF₄·2H₂O (4).

View Article Online

DOI: 10.1039/C7DT00295E

ARTICLE

5

6

M. Nath, J. C. Huffman and J. M. Zaleski, Accelerated Bergman

cyclization of porphyrinic-enediynes, Chem. Commun., 2003, 7,

858-859.

H

L^H/tBu(90 mg, 0.1516 mmol)

was dissolved in pyridine (16 mL) with gentle heating, and then

deoxygenated on a Schlenk line with successive vacuum and

nitrogen cycles. Pyridine (2 mL) was degassed as above, and

then Co(H₂O)₆(BF₄)₂(25.3 mg, 0.0743 mmol) added under

positive N₂pressure to give a pale orange-pink solution, which

was degassed with further vacuum/purge cycles; N₂flow was

left on. A transfer needle was added between the metal salt

and ligand flasks, and then the cobalt solution transferred

dropwise to the ligand, causing the pale yellow solution to turn

orange. The orange solution was then degassed with another

vacuum/N₂cycle before being stirred for 1 day under a

nitrogen balloon, before exposing the orange solution to air

and subjecting it to diethyl ether vapour diffusion. This yielded

L. Arnold, H. Norouzi-Arasi, M. Wagner, V. Enkelmann and K.

Mullen, A porphyrin-related macrocycle from carbazole and

pyridine building blocks: synthesis and metal coordination,

Chem. Commun., 2011, 47, 970-972.

J. S. Hewage, S. Wanniarachchi, T. J. Morin, B. J. Liddle, M.

Banaszynski, S. V. Lindeman, B. Bennett and J. R. Gardinier,

Homoleptic Nickel(II) Complexes of Redox-Tunable Pincer-type

Ligands, Inorg. Chem., 2014, 53, 10070 - 10084.

7

8

9

M. Wang, N. Chamberland, L. Breau, J.-E. Moser, R. Humphry-

Baker, B. Marsan, S. M. Zakeeruddin and M. Graetzel, An

organic redox electrolyte to rival triiodide/iodide in dye-

sensitized solar cells, Nat. Chem., 2010, 2, 385-389.

X-ray quality single crystals of [Co^III(L^H/tBu)₂]BF₄

·1.5Et₂O∙solvent

A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K.

Nazeeruddin, E. W.-G. Diau, C.-Y. Yeh, S. M. Zakeeruddin and M.

Grätzel, Porphyrin-Sensitized Solar Cells with Cobalt (II/III)–

Based Redox Electrolyte Exceed 12 Percent Efficiency, Science,

2011, 334, 629.

as blue blocks (39 mg, 0.0287 mmol, 39%). The crystals were

isolated via filtration followed by air drying, and then analysed.

Found: C, 66.67; H, 5.80; N, 14.52. Calculated for

C₇₆H₇₆N₁₄Co∙BF₄∙2H₂O (1367.31 g mol^-1): C, 66.76; H, 5.90; N,

10 M. K. Kashif, M. Nippe, N. W. Duffy, C. M. Forsyth, C. J. Chang, J.

R. Long, L. Spiccia and U. Bach, Stable Dye-Sensitized Solar Cell

14.34 %. . ESI-MS (pos., MeCN): m/z = 1243.5634 (calc. for

[C₇₆H₇₆N₁₄Co]⁺([Co(

1

L

^H/tBu)₂]⁺) = 1243.5704). H NMR (500 MHz,

Electrolytes Based on Cobalt(II)/(III) Complexes of

a

Hexadentate Pyridyl Ligand, Angew. Chem. Int. Ed., 2013, 52

5527-5531.

11 J.-H. Yum, E. Baranoff, F. Kessler, T. Moehl, S. Ahmad, T. Bessho,

A. Marchioro, E. Ghadiri, J.-E. Moser, C. Yi, M. K. Nazeeruddin

and M. Grätzel, A cobalt complex redox shuttle for dye-

sensitized solar cells with high open-circuit potentials, Nature

,

DMSO-d₆) δ (ppm) = 8.84 (s, 4H, H₆), 8.50 (d, 4H, J = 7.6 Hz, H₄),

7.73 (d, 4H, J = 7.4 Hz, H₂), 7.29 (t, 4H, J = 7.5 Hz, H₃), 6.96 (d,

8H, J = 8.4 Hz, H₁₀), 6.32 (d, 8H, J = 8.4 Hz, H₉), 4.85 (s, 8H, H₇),

1.17 (s, 36H, H₁₃). ¹³C NMR (126 MHz, DMSO-d₆) δ (ppm) =

150.5 (C₁₁), 143.6 (C₅), 142.1 (C_1a), 130.9 (C₈), 127.6 (C_4a), 127.0

(C₉), 125.1 (C₁₀), 122.9 (C₆), 122.0 (C₄), 121.1 (C₂), 116.1 (C₃),

109.6 (C₁), 54.6 (C₇), 34.1 (C₁₂), 31.0 (C₁₃). IR (ATR), inter alia, ν

(cm^-1) = 3131 (w), 3058 (w), 2952 (m), 2902 (w), 2866 (w),

1612 (w), 1579 (w), 1552 (w), 1512 (w), 1477 (m), 1420 (s),

1349 (m), 1266 (vs), 1229 (s), 1213(m), 1167 (w), ~1100-1000

(broad, vs), 1051 (vs), 936 (w), 767 (vs), 746 (vs), 712 (m), 691

(m), 625 (w), 564 (m), 575 (s), 519 (m). Λ_m(MeCN) = 123 Ω^-1

cm²mol^-1.

Commun., 2012, 3, 631.

12 M. G. Cowan, R. G. Miller and S. Brooker, Pyrazine-imide

complexes: reversible redox and MOF building blocks, Dalton

Trans., 2015, 44, 2880-2892.

13 M. G. Cowan and S. Brooker, Nine non-symmetric pyrazine-

pyridine imide-based complexes: reversible redox and isolation

of [M^II/III(pypzca)₂]^0/+when M = Co, Fe, Dalton Trans., 2012, 41

,

1465-1474 and front cover feature.

14 M. G. Cowan, J. Olguín, S. Narayanaswamy, J. L. Tallon and S.

Brooker, Reversible switching of a cobalt complex by thermal,

pressure and electrochemical stimuli: abrupt, complete,

hysteretic spin crossover, J. Am. Chem. Soc., 2012, 134, 2892–

2894 and front cover feature.

15 T. Inagaki, L. T. Phong, A. Furuta, J.-i. Ito and H. Nishiyama, Iron-

and Cobalt-Catalyzed Asymmetric Hydrosilylation of Ketones

and Enones with Bis(oxazolinylphenyl)amine Ligands, Chem.

Eur. J., 2010, 16, 3090-3096.

Acknowledgements

This work was supported by the Department of Chemistry,

University of Otago, as well as the MacDiarmid Institute for

Advanced Materials and Nanotechnology, and the Marsden

Fund (postdoctoral fellowship to H.L.C.F.).

16 C. W. G. Ansell, J. Lewis and P. R. Raithby, Synthesis and x-ray

structure of a cobalt(II) complex of a negatively charged

quinquedentate macrocyclic ligand, J. Chem. Soc., Dalton Trans.,

1982, 12, 2557-2559.

17 G. J. P. Britovsek, V. C. Gibson, O. D. Hoarau, S. K. Spitzmesser,

A. J. P. White and D. J. Williams, Iron and cobalt ethylene

polymerization catalysts: variations on the central donor, Inorg.

Chem., 2003, 42, 3454-3465.

18 K. Smith, D. M. James, A. G. Mistry, M. R. Bye and D. J. Faulkner,

A new method for bromination of carbazoles, ß-carbolines, and

imidodibenzyls by use of N-bromosuccinimide and silica gel,

Tetrahedron, 1992, 48, 7479-7488.

19 A. Neugebauer and H. Fischer, tert.-Butyl-substituierte

Carbazole, Chem. Ber., 1972, 105, 2686-2693.

Notes and references

1

S. J. Malthus, S. A. Cameron and S. Brooker, First row transition

metal complexes of di-o-substituted-diarylamine-based ligands

(including carbazoles, acridines and dibenzoazepines), Coord.

Chem. Rev., 2016, 316, 125–161.

2

3

4

M. Inoue and M. Nakada, New preparation of tridentate bis-

oxazoline carbazole ligand effective for enantioselective Nozaki-

Hiyama reaction, Heterocycles, 2007, 72, 133-138.

H.-C. Gee, C.-H. Lee, Y.-H. Jeong and W.-D. Jang, Highly sensitive

and selective cyanide detection via Cu²⁺complex ligand

exchange Chem. Commun., 2011, 47, 11963-11965.

H. Aihara, L. Jaquinod, D. J. Nurco and K. M. Smith,

Multicarbocycle formation mediated by arenoporphyrin 1,4-

diradicals: Synthesis of picenoporphyrins, Angew. Chem. Int.

Ed., 2001, 40, 3439-3441.

20 V. C. Gibson, S. K. Spitzmesser, A. J. P. White and D. J. Williams,

Synthesis and reactivity of 1,8-bis(imino)carbazolide complexes

Dalton Trans., 2017, 00, 1-3 | 13

Please do not adjust margins

Dalton Transactions

Page 14 of 15

View Article Online

DOI: 10.1039/C7DT00295E

ARTICLE

Dalton Transactions

of iron, cobalt and manganese, Dalton Trans., 2003, 13, 2718- 38 R. Greef, R. Peat, L. M. Peter, D. Pletcher and J. Robinson,

2727.

Instrumental Methods in Electrochemistry, Ellis Horwood,

21 E. Jurgens, K. N. Buys, A.-T. Schmidt, S. K. Furfari, M. L. Cole, M.

Woking, 1985.

Moser, F. Rominger and D. Kunz, Optimised synthesis of 39 J. Heinze, Cyclic Voltammetry - "Electrochemical Spectroscopy",

monoanionic bis(NHC)-pincer ligand precursors and their Li-

Angew. Chem. Int. Ed. Engl., 1984, 23, 831-845.

complexes, New J. Chem., 2016, 9160-9169.

40 CrysAlisPro.

22 Y. Liu, M. Nishiura, Y. Wang and Z. Hou, Pi-Conjugated Aromatic 41 A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, M. C.

Enynes as

a

Single-Emitting Component for White

Burla, G. Polidori and M. Camalli, SIR92 - a program for

automatic solution of crystal structures by direct methods, J.

Appl. Cryst., 1994, 27, 435-435.

Electroluminescence, J. Am. Chem. Soc., 2006, 128, 5592-5593.

23 S. A. Cameron, PhD thesis, University of Otago, New Zealand,

2012.

24 M. Tashiro, Selective synthesis of aromatic compounds using

positional protective groups, Synthesis, 1979, 12, 921-936.

42 L. Palatinus and G. Chapuis, SUPERFLIP. A computer program for

the solution of crystal structures by charge flipping in arbitrary

dimensions, J. Appl. Crystallogr., 2007, 40, 786-790.

25 M. S. Mudadu, A. N. Singh and R. P. Thummel, Preparation and 43 G. M. Sheldrick and T. R. Schneider, SHELXL, Methods Enzymol.,

Study of 1,8-Di(pyrid-2'-yl)carbazoles, J. Org. Chem., 2008, 73

,

1997, 277, 319-343.

6513-6520.

44 P. W. Betteridge, J. R. Carruthers, R. I. Cooper, K. Prout and D. J.

26 D. I. Bezuidenhout, G. Kleinhans, G. Guisado-Barrios, D. C. Liles,

G. Ung and G. Bertrand, Isolation of a potassium bis(1,2,3-

Watkin, Crystals suite of programs, J. Appl. Cryst., 2003, 36,

1487.

triazol-5-ylidene)carbazolide:

a stabilizing pincer ligand for 45 R. I. Cooper, A. L. Thompson and D. J. Watkin, CRYSTALS

reactive late transition metal complexes, Chem. Commun.,

2014, 50, 2431-2433.

enhancements: dealing with hydrogen atoms in refinement, J.

Appl. Cryst., 2010, 43, 1100–1107.

27 C. Maeda, T. Yoneda, N. Aratani, M. C. Yoon, J. M. Lim, D. Kim, 46 R. I. Cooper, R. O. Gould, S. Parsons and D. J. Watkin, The

N. Yoshioka and A. Osuka, Synthesis of Carbazole-Containing

Porphyrinoids by Multiple Annulation Strategy: Core-

Modified and π-Expanded Porphyrin, Angew. Chem. Int. Ed.,

2011, 50, 5691-5694.

derivation of non-merohedral twin laws during refinement by

analysis of poorly fitting intensity data and the refinement of

non-merohedrally twinned crystal structures in the program

CRYSTALS, J. Appl. Crystallogr., 2002, 35, 168-174.

a

A

28 J. D. Crowley and P. H. Bandeen, A multicomponent CuAAC

“click” approach to a library of hybrid polydentate 2-pyridyl-

1,2,3-triazole ligands: new building blocks for the generation of

metallosupramolecular architectures, Dalton Trans., 2010, 39

,

612–623.

29 C. Wang, D. Ikhlef, S. Kahlal, J.-Y. Saillard and D. Astruc, Metal-

catalyzed azide-alkyne “click” reactions: Mechanistic overview

and recent trends, Coord. Chem. Rev., 2016, 316, 1-20.

30 L. Liang and D. Astruc, The copper(I)-catalyzed alkyne-azide

cycloaddition (CUAAC) "click" reaction and its applications. An

overview, Coord. Chem. Rev., 2011, 255, 2933-2945.

31 J. E. Hein and V. V. Fokin, Copper-catalyzed azide–alkyne

cycloaddition (CuAAC) and beyond: new reactivity of copper(I)

acetylides, Chem. Soc. Rev., 2010, 39, 1302-1315.

32 R. G. Miller, S. Narayanaswamy, J. L. Tallon and S. Brooker, Spin

crossover with thermal hysteresis in cobalt(II) complexes and

the importance of scan rate, New J. Chem., 2014, 38, 1932-

1941.

33 P. van der Sluis and A. L. Spek, BYPASS: an effective method for

the refinement of crystal structures containing disordered

solvent regions, Acta Crystallogr., Sect. A, 1990, A46, 194-201.

34 P. Guionneau, C. Brigouleix, Y. Barrans, A. E. Goeta, J.-F. Létard,

J. A. K. Howard, J. Gaultier and D. Chasseau, high pressure and

very low temperature effects on the crystal structures of some

iron(II) complexes, Comptes Rendus Acad. Sci. Ser. IIc, 2001, 4,

161-171.

35 M. G. B. Drew, C. J. Harding, V. McKee, G. G. Morgan and J.

Nelson, Geometric Control of Manganese Redox State, J. Chem.

Soc., Chem. Commun., 1995, 1035-1038.

36 F. A. Deeney, J. H. Charles, G. G. Morgan, V. McKee, J. Nelson, S.

J. Teat and W. Clegg, Response to steric constraint in

azacryptate and related complexes of iron-(II) and -(III), J. Chem.

Soc., Dalton Trans., 1998, 1837-1843.

37 D. M. Adams, G. J. Long and A. D. Williams, Spectroscopy at very

high pressures. 36. An infrared study of spin-state equilibriums

in some iron(II) complexes, Inorg. Chem., 1982, 21, 1049-1053.

14 | Dalton Trans., 2017, 00, 1-3

Please do not adjust margins

Dalton Transactions

Page 15 of 15

ARTICLE

Dalton Transactions

Please do not adjust margins

View Article Online

DOI: 10.1039/C7DT00295E

Table of contents text:

Four structurally characterised [Co^IIIL^R/R’₂]BF₄complexes, of four acyclic bis(triazolyl)carbazole ligands L^R/R’, all show a pair of

reversible one-electron redox processes which can be tuned by variation of the carbazole substituents R.

Table of Contents graphic:

15

Please do not adjust margins

Article Doi

Tuneable reversible redox of cobalt(iii) carbazole complexes

DOI: 10.1039/C7DT00295E

Source and publish data:

Authors:

Article abstract of DOI:10.1039/C7DT00295E

Full text of DOI:10.1039/C7DT00295E

Products guided by the article

R&D Labs maybe for 1350542-36-7

Relevant to this article

Hot Product