Nucleophilic aromatic substitution of bis(pentafluorophenyl)mercury with various bulky nucleophiles and the structures of [Hg(C6F 4X-4)2] (X≤cyclo-C5H10N, OCH(CH3)2, OC(CH3)3)

Article abstract of DOI:10.1071/CH13227

Reactions of bis(pentafluorophenyl)mercury with piperidine, sodium iso-propoxide or sodium tert-butoxide have yielded the corresponding 4-substituted tetrafluorophenylmercurials, [Hg(C6F4X-4)2] (X≤cyclo-C5H10N (1), OCH(CH3)2 (2), OC(CH3)3 (3)), in reasonable yields but the bulkier nucleophiles, cis-2,6-dimethylpiperidine and 2,6-di-iso-propylphenolate (from sodium 2,6-di-iso-propylphenolate) decomposed the mercurial into pentafluorobenzene. Treatment of bis(pentafluorophenyl)mercury with another bulky nucleophile, 2,6-diphenylphenolate (from sodium 2,6-diphenylphenolate), in methanol, resulted in the unexpected formation of [Hg(C6F4(OMe)-4)2] (4). The structures of all the mercurials have linear C-Hg-C stereochemistry with two coplanar aryl rings. Amongst a complex series of supramolecular interactions, HgO bonding is observed for the alkoxy substituted mercurials but there are no HgN interactions in the structure of bis(2,3,5,6-tetrafluoro-4-piperidinophenyl) mercury.

Full text of DOI:10.1071/CH13227

RESEARCH FRONT
CSIRO PUBLISHING
Aust. J. Chem. 2013, 66, 1253–1259
Full Paper
http://dx.doi.org/10.1071/CH13227
Nucleophilic Aromatic Substitution of
Bis(pentafluorophenyl)mercury with Various Bulky
Nucleophiles and the Structures of [Hg(C6F4X-4)2]
(X 5 cyclo-C5H10N, OCH(CH3)2, OC(CH3)3)
A C
,
B C
,
A
Glen B. Deacon,
Peter C. Junk,
and Jenny Luu
A
School of Chemistry, Monash University, Clayton, Vic. 3800, Australia.
School of Pharmacy and Molecular Sciences, James Cook University, Townsville, Qld
4811, Australia.
B
C
Corresponding authors. Email: glen.deacon@monash.edu; peter.junk@jcu.edu.au
Reactions of bis(pentafluorophenyl)mercury with piperidine, sodium iso-propoxide or sodium tert-butoxide have yielded
the corresponding 4-substituted tetrafluorophenylmercurials, [Hg(C6F4X-4)2] (X ¼ cyclo-C5H10N (1), OCH(CH3)2 (2),
OC(CH3)3 (3)), in reasonable yields but the bulkier nucleophiles, cis-2,6-dimethylpiperidine and 2,6-di-iso-propyl-
phenolate (from sodium 2,6-di-iso-propylphenolate) decomposed the mercurial into pentafluorobenzene. Treatment of
bis(pentafluorophenyl)mercury with another bulky nucleophile, 2,6-diphenylphenolate (from sodium 2,6-diphenyl-
phenolate), in methanol, resulted in the unexpected formation of [Hg(C6F4(OMe)-4)2] (4). The structures of all the
mercurials have linear C–Hg–C stereochemistry with two coplanar aryl rings. Amongst a complex series of supramolec-
ular interactions, Hg?O bonding is observed for the alkoxy substituted mercurials but there are no Hg?N interactions in
the structure of bis(2,3,5,6-tetrafluoro-4-piperidinophenyl)mercury.
Manuscript received: 1 May 2013.
Manuscript accepted: 18 June 2013.
Published online: 30 July 2013.
Introduction
different rotations of the aryl rings (08–908) in structures of such
compounds.[9–18] Simple explanations in terms of electronic or
steric effects are ruled out on considering the sequence [Hg
(C6F5)2] (dihedral angle: 58.48),[17,18] [Hg(C6H2F3-2,4,6)2]
(coplanar rings),[11] [Hg(C6H3F2-2,6)2] (dihedral angle:
888),[11] and [Hg(C6F4(NO2)-2)2] (coplanar).[16] Accordingly,
it has been proposed that the rotation angles may be influenced
by supramolecular interactions, which lead to different crystal
packing assemblies.[16] As two rotamers of [Hg(C6F4(NO2)-3)2],
with twist angles of 878 and 598, can be crystallised from
the same solvent at room temperature and ꢀ108C, respectively,
the interconversion energy must be low and this is borne out by
calculations.[16] The supramolecular interactions and crystal
packing of the two rotamers are different.[16] Thus, the structures
of the new mercurials, [Hg(C6F4X-4)2] (X ¼ cyclo-C5H10N,
(OiPr), (OtBu), prepared in this study have been determined
and are discussed together with their supramolecular
interactions.
Nucleophilic aromatic substitution reactions of poly-
fluoroarenes, with the displacement of a fluorine substituent,
have been an extensively studied topic.[1–3] Much less studied
are these types of reactions with polyfluoroarylorganometallics
(and organometalloids) and only [Hg(C6F5)2], [Hg(C6F5)Cl],
[Cr(CO)3C5H4F], [Si(C6F5)(CH3)3], [Li(C6F5)], and [S(C6F5)2O2]
appear to have been investigated.[4–8] In an early study, the
mercury compound [Hg(C6F5)2] was shown to give good yields
of [Hg(C6F4X-4)2] (X ¼ OH, OMe, NHNH2) on reaction with
KOH, NaOMe, or hydrazine, respectively, although the reaction
with hydrazine also yielded a considerable amount of Hg metal.
By contrast, methyllithium gave (pentafluorophenyl)lithium and
[Hg(CH3)2]. Reactions of [Hg(C6F5)Cl] were more complex.
Thus, hydrazine caused symmetrisation into Hg metal and [Hg
(C6F5)2], and sodium methoxide produced [Hg(C6F4(OMe)-4)2]
from a combination of nucleophilic substitution and symme-
trisation.[4] The nucleophiles used for these previous studies are
relativelysmalland notlikely toinhibitsubstitution at an aromatic
carbon.[4] We now report the effect of steric bulk on reactions of
[Hg(C6F5)2] from investigations of treatment with various
bulky nucleophiles (piperidine, cis-2,6-dimethylpiperidine,
iso-propoxide (ꢀOiPr), tert-butoxide (ꢀOtBu), 2,6-di-iso-
propylphenolate (ꢀOdip), and 2,6-diphenylphenolate (ꢀOdpp).
The last four nucleophiles are from their sodium derivatives.).
In addition, the structural aspects of new bis(polyfluoro-
phenyl)mercury compounds are of interest, in view of reports of
Results and Discussion
The mercurials [Hg(C6F4X-4)2] (X ¼ cyclo-C5H10N (1), (OiPr)
(2), (OtBu) (3)) were synthesised by treating [Hg(C6F5)2] with
piperidine, sodium iso-propoxide, and sodium tert-butoxide,
respectively (Scheme 1). In order to determine the reaction
times required, the reaction mixtures were tracked at 1 h inter-
vals using 19F{1H} NMR spectroscopy. Complete reaction was
indicated when the three resonances of [Hg(C6F5)2] were
Journal compilation Ó CSIRO 2013
www.publish.csiro.au/journals/ajc

1254
G. B. Deacon, P. C. Junk, and J. Luu
F
F
F
F
F
F
F
F
N
N
Hg
1
H
i. Piperidine
reflux
N
4h
ii. 10% HCl
F
F
F
F
F
F
F
Hg
F
F
F
i. tert-butanol
i. iso-propyl alcohol
reflux
6.5h
reflux
5h
ii. 10% HCl
ii. 10% HCl
Na
O
Na
O
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
O
O
Hg
O
O
Hg
2
3
Scheme 1. Reaction scheme for the synthesis of 1–3.
Table 1. Reaction conditions used to synthesise 1 3
]
Product
Amount of
Nucleophile (amount
used [mmol])
Solvent (amount
used [mL])
Reaction time [h]
Yield [%] (recrystallisation
solvent)
[Hg(C6F5)2] used [mmol]
1
2
3
0.75
0.93
0.47
PiperidineA (40)
ꢀOiPrB (3.8)
ꢀOtBuB (5.4)
–
4
36 (chloroform)
67 (ethanol)C
32 (acetone)
iso-propyl alcohol (15)
tert-butanol (25)
5
6.5
APiperidine used as both the nucleophile and solvent.
BFrom sodium derivative.
CCrude product was recrystallised twice.
completely replaced by the two signals of the desired product,
with one exhibiting mercury satellites.
A representative 199Hg NMR spectrum was obtained for 1 which
exhibited a pentet at ꢀ774.34 ppm with a JHg–F coupling
3
The synthesis of 1 was initially attempted using ethanol as the
solvent, following a report of the nucleophilic aromatic substi-
tution of bromopentafluorobenzene with piperidine in etha-
nol.[19] However, this proved ineffective with [Hg(C6F5)2], as
only a small signal corresponding to the desired product was
present in the 19F{1H} NMR spectrum despite a long reaction
time (Fig. S1A in the Supplementary Material). As [Hg(C6F5)2]
dissolves readily in piperidine, which has a higher boiling point
than ethanol, the synthesis was then tried with piperidine as the
solvent as well as a reactant. After four hours the 19F{1H} NMR
signals of 1 replaced those of the mercurial reactant. The
19F{1H} NMR spectrum of 1 is shown in Fig. S1B in the
Supplementary Material.
constant (410 Hz) which corresponds well to that observed in
the 19F{1H} NMR spectrum (406 Hz). The chemical shift of 1 is
higher than for [HgPh2] (ꢀ799 ppm),[20] in contrast to other bis-
(polyfluoroaryl)mercurials ([Hg(C6F5)2] (ꢀ953 ppm)[21] and
[Hg(C6F4H-4)2] (ꢀ946 ppm))[20] with the exception of [Hg
(C6H3F2-2,6)2] (ꢀ780 ppm).[20] This may be attributed to the
electron donating properties of the cyclo-C5H10N substituent.
Bulkier nucleophiles (cis-2,6-dimethylpiperidine, ꢀOdip and
ꢀOdpp) were also examined to determine the limits of steric
bulk on these type of nucleophilic aromatic substitutions. The
reactions involving the nucleophiles cis-2,6-dimethylpiperidine
(reaction 1, Table 2) and ꢀOdip (reaction 2, Table 2) gave
pentafluorobenzene[22] as the fluorocarbon product and possibly
a trace amount of [Hg(C6F4(Odip)-4)2]. Although [Hg(C6F5)2]
has been recrystallised unchanged from concentrated sulfuric
acid,[23] it is susceptible to nucleophilic attack at the mercury
The synthesis of complexes 2 and 3 required longer reactions
times than for 1. The amounts of reactants, reaction times,
and solvents used for these reactions are listed in Table 1.

Nucleophilic Aromatic Substitution of Bis(pentafluorophenyl)mercury
1255
Table 2. Reaction conditions for the attempted substitution reactions
Reaction Amount of [Hg(C6F5)2] Nucleophile (amount
Solvent (amount Reaction time Products
used [mL])
used [mmol]
used [mmol])
1A
2
0.47
0.47
cis-2,6-dimethylpiperidine (300)
ꢀOdipB (3.7)
–
10.5 h
10 h
C6F5H
dipOH (20)
C6F5H [Hg(Odip)2]C
[Hg(C6H2(CH(CH3)2
(ONa))(Odip)]C
4 and HOdpp
3
0.47
ꢀOdppB (0.92)
methanol (25)
3 weeks
Acis-2,6-dimethylpiperidine is used as both the nucleophile and solvent.
BFrom the sodium derivative.
COn the basis of mass spectra.
centre giving C6Fꢀ5 and then C6F5H, e.g. by iodide ions or
(a)
F(3)
N(1)
F(4)
(Ph2P)2NH.[24,25] The reaction mixture from an attempted
synthesis (reaction 2, Table 2) also gave products with mercury
isotope patterns in the mass spectrum. One peak at m/z 663
may correspond to {[Hg(Odip)-4)2] þ OHꢀ þ 5H2O}ꢀ, while
the peak at m/z 713 could be assigned to the mercuration
product {[Hg(C6H2(CH(CH3)2ONa)(Odip)] þ OHꢀ þ 2H2O þ
2MeCN}ꢀ. Sodium 2,6-di-iso-propylphenolate is activated with
regards to electrophilic mercuration. The formation of C6F5H in
attempted syntheses (reactions 1 and 2) suggests that it is also
inert to cis-2,6-dimethylpiperidine and ꢀOdip. In order to con-
firm this, C6F5H was treated with cis-2,6-dimethylpiperidine
on an NMR scale. After prolonged sonication, no reaction
was observed. The solution was then left standing for a
week at room temperature and the 19F{1H} NMR spectrum
showed mainly C6F5H but also trace amounts of the potential
product by resonances at ꢀ141.9 and ꢀ152.0 ppm (1 : 1). In
addition, a new peak at ꢀ139.9 ppm may correspond to
1,2,4,5-tetrafluorobenzene.[22] Thus, C6F5H is also resistant
to this bulky nucleophile.
Hg(1)
F(1)
F(4)
F(2)
F(3)
(b)
Hg(1)
O(1)
F(1)
F(4)
F(2)
As HOdpp is a solid, it was not used as a reaction medium for
the reaction of [Hg(C6F5)2] with NaOdpp (reaction 3, Table 2).
The reaction was attempted in methanol. Unexpectedly,
the known[4,9] compounds [Hg(C6F4(OMe)-4)2] (4) and 2,6-
diphenylphenol (HOdpp) were isolated from hand-picked
crystals. Despite the relative acidities of methanol and HOdpp,
the excess methanol and the thermodynamic sink provided by
the nucleophilic substitution of [Hg(C6F5)2] by methoxide ions
have driven this outcome. Overall, cis-2,6-dimethylpiperidine,
ꢀOdip, and ꢀOdpp are too bulky to displace a fluorine atom by
nucleophilic substitution.
F(3)
(c)
Hg(1)
O(1)
F(2)
F(1)
Fig. 1. X-ray crystal structures of (a) 1, (b) 2 and (c) 3 showing 50 %
probability thermal ellipsoids. Hydrogen atoms have been omitted for
clarity.
Structural Determinations
The molecular structures were determined for three complexes
(1–3, Fig. 1), with crystals prepared by slow evaporation from
chloroform, ethanol, and acetone, respectively. The crystal data
of these compounds are given in Table 3 and selected bond
lengths and angles are given in Table 4. These compounds are
monomeric at the molecular level with the mercury metal cen-
tres having a characteristic coordination number of two.[26]
Complex 1 has the piperidinyl substituent in a chair configura-
tion. Interestingly, the aromatic rings of all these compounds
have a coplanar arrangement (Fig. 1), with a C–Hg–C angle of
1808 (Table 2). The aryl twist angle of 1–3 (08) contrasts those of
[Hg(C6F4(NH2)-4)2] (62.28) and [Hg(C6F4(OMe)-4)2] (52.98),
which contain less bulky 4-substituents.[9] There have been
several reports of diarylmercury compounds with a planar ori-
entation,[10,11] notably diphenylmercury,[12–14] even though
calculations favour 908.[14] Although the contrast between the
present results and the structures of [Hg(C6F4X-4)2] (X ¼ NH2
and OMe)[9] might suggest that coplanar rings are associated
with bulky para-substituents, [Hg(C6F4H-4)2] also has coplanar
rings.[11]
The Hg–C distances of 1–3 are comparable to those of other
similar reported structures.[9–11,15–18] The N–C(Ar) and O–C(Ar)
bond lengths lie between those of N(O)–C and N(O)¼C[27]
indicating partial double bond character. This is attributed to the
partial conjugation of the lone pairs of the nitrogen or oxygen
atoms of the NC5H10, OiPr, and OtBu substituents into the aryl
ring. Other related complexes, such as [Hg(C6F4(NH2)-4)2]
and [Hg(C6F4(OMe)-4)2] show comparable N–C(OAr) and
O–C(Ar) bond lengths.[9]
Different supramolecular assemblies are observed in these
three structures and may contribute to their coplanar

1256
G. B. Deacon, P. C. Junk, and J. Luu
Table 3. Crystallographic data for 1–3
Parameter
1
2
3
Empirical formula C22H20F8HgN2 C18H14F8HgO2 C20H18F8HgO2
Molecular weight 664.99
614.88
Triclinic
P-1
642.93
Crystal system
Space group
Monoclinic
Monoclinic
P21/c
P21/c
Crystal colour
Colourless
Colourless
Colourless
Crystal size [mm3] 0.12 ꢁ 0.1 ꢁ 0.1 0.2 ꢁ 0.1 ꢁ 0.1 0.12 ꢁ 0.05 ꢁ 0.04
˚
a [A]
˚
b [A]
˚
c [A]
12.3696(17)
10.2968(15)
8.2449(12)
90
6.7648(4)
7.1018(5)
10.6119(6)
71.292(4)
85.555(3)
77.176(3)
470.83(5)
1
10.4453(6)
10.8951(6)
9.2388(5)
90
Fig. 2. A simplified representation showing the rotation of the p-stacked
pairs of 1.
a [deg.]
b [deg.]
g [deg.]
98.324(5)
90
93.230(3)
90
3
˚
Hg(1)
V [A ]
1039.1(3)
2
1049.73(10)
2
Z
O(1)
Hg(1)
Density (calc.)
[g cmꢀ3
2.125
2.169
2.034
]
R1 (I . 2s(I))
F000
0.0584 (0.0328) 0.0255 (0.0255) 0.0258 (0.0152)
O(1)
636
290
612
O(1)
O(1)
˚
l [A]
0.71073
55.0
0.71073
55.0
0.71073
55.0
Hg(1)
2ymax [deg.]
Final GoF
O(1)
Hg(1)
1.018
1.124
1.015
Hg(1)
O(1)
wR2 (I . 2s(I))
0.0797 (0.0701) 0.0476 (0.0476) 0.0426 (0.0381)
7.491 8.260 7.416
m [mmꢀ1
]
O(1)
Hg(1)
O(1)
O(1)
Hg(1)
Table 4. Selected bond lengths and angles for 1–3
Bonds/atoms
1
2
3
Fig. 3. The crystal packing diagram of 2 highlighting the Hg?O
interactions.
˚
Bond distances [A]
2.069(4)
Hg–C
2.064(5)
1.390(7)
4.559(5)
2.057(2)
1.364(3)
2.981(2)
C(4)–N(O)
Hg?N(O)
1.367(4)
another feature that differs from 1 which may also contribute to
the different ring rotation angle of 62.28. An intermolecular
hydrogen bonding interaction between the NH2 groups and a
F substituent is present in this particular structure, and may
contribute to the twist of the aryl rings despite them being locked
in a p–p interaction.[9]
3.3147(2)
Bond angles [deg.]
180.0(3)
˚
Deviations from the ring plane [A]
C–Hg–C
180.0(1)
180.0(2)
Hg
0.076(8)
0.100(9)
0.104(6)
0.125(6)
0.077(5)
0.109(5)
N(O)
While these hydrogen bonds (NH2?F) and Hg?N interac-
tions are absent in 1, other possible interactions are observed.
These include a weak intermolecular F(2)?Hg contact at a
arrangement. For 1, each of the aromatic rings exhibits a p–p
interaction with one other ring (Fig. 2 and Fig. S2 in the
Supplementary Material). The p–p stacking of these rings is
in a parallel offset face-to-face arrangement with an interplanar
˚
distance of 3.451(3) A, a value which lies just outside the sum of
[36]
and mercury
˚
the van der Waals radii of fluorine (1.47 A)
˚
(1.73 A minimum value) but is within the limit using higher
proposedvalues formercury.[29–33] Interestingly, a short halogen–
halogen contact can be observed between the same F(2) of one
molecule and F(1) of another. The distance between these two
˚
˚
distance of 3.286(8) A, much shorter than the sum (3.46 A) of
the van der Waals radii of two aryl rings.[28] A minimum C–C
0
˚
contact of 3.314(8) A between pairs of carbons (C(2)?C(3 ),
C(20)?C(3)) of these rings is observed. The packing of this
molecule can be described as a herringbone structure, and the
offset pairs lie near perpendicular (85.2(3)8) to their neighbour-
ing pairs (Fig. 2).
˚
atoms (2.785(5) A) is within the sum of two van der Waals radii
[36]
It appears that this interaction may
˚
of fluorine (2.94 A).
contribute to the offset pairs lying near perpendicular to each
other, as the aromatic ring of F(1) is rotated almost 908 relative to
the aryl ring containing F(2) of this interaction.
Although 2 showed a less complex assembly, major interac-
tions can still be observed such as p–p interactions and F?H
non-classical hydrogen bonding, the latter not a feature of 1. The
In contrast to its less sterically hindered derivative [Hg
(C6F4(NH2)-4)2],[9] 1 exhibits no Hg?N interactions as the
˚
separation (4.559(5) A) is well outside the van der Waals radii
[9,34,35]
[29–33]
˚
for mercury (1.73 A
˚
and can be up to 2.1–2.2 A,
˚
particularly when perpendicular to a strong bonding axis) and
[36]
A possible explanation for this could be
interplanar distance between the aromatic rings is 3.273(4) A
0
˚
nitrogen (1.55 A).
˚
with a C(1)–C(5 ) contact of 3.281(6) A. This p–p stacking is
also arranged in an offset face-to-face manner and represents a
sandwich herringbone structure. In contrast to the assembly of 1,
the p–p interactions do not exist as perpendicular pairs of pairs
but instead as sets of parallel pairs. Moreover, there is also
plausible evidence of a Hg?O interaction (Fig. 3) at a distance
˚
the deviation of the nitrogen (0.100(9) A) from the plane of the
aromatic rings away from the mercury atom, whereas in [Hg
(C6F4(NH2)-4)2] the nitrogen is bent towards the mercury atom
and is therefore in a better position to donate electron density to
mercury. Analysis of the [Hg(C6F4(NH2)-4)2] structure revealed

Nucleophilic Aromatic Substitution of Bis(pentafluorophenyl)mercury
1257
(a)
Table 5. Hydrogen bonding data for 2, 3, and 4
˚
dD–H [A] dH–A [A] dD–A [A]
˚
˚
Complex
D
H
A
,DHA [deg.]
O(1)
2
3
4
C9 H9B F(1)
C8 H8A F(1)
C5 H1 F(1)
0.98
0.96
0.88
2.55
2.58
2.63
3.273(6)
3.219(4)
3.3272(9)
130.25
124.72
136.12
Hg(1)
O(1)
bonding to occur. In addition, the methyl groups in 2 and 3 have
more flexibility than in the OMe substituent of 4. Hydrogen
bonding for these complexes is summarised in Table 5.
(b)
H(8A)
Conclusions
F(1)
Three new organomercury compounds 1–3 have been synthe-
sised by nucleophilic aromatic substitution reactions through
treatment of [Hg(C6F5)2] with piperidine, NaOiPr or NaOtBu.
The bulkier nucleophiles used in attempted reactions 1–3
(Table 2) proved unsuitable due to their steric bulk preventing
access to the ring carbons. The mercurials from the successful
reactions have crystal structures with co-planar ring arrange-
ments. In the crystal packing different assemblies and supra-
molecular interactions were observed. The major features were
p–p stacking, and Hg?O and F?H interactions, which may
contribute to the planar configurations of these structures.
F(1)
H(8A)
Fig. 4. The (a) Hg?O and (b) F?H interactions of 3.
˚
of 3.315(3) A, which is near the sum of the van der Waals radii of
mercury (minimum value) and oxygen (3.25 A).[29–33,36] How-
˚
ever, there have been credible reports of stacking bonding
˚
Experimental
distances of 3.30–3.40 A for Hg?O, implying a mercury van
[29,34,35]
General
˚
der Waals radius of 1.9–2.0 A.
Thus, phenylmercuric
All reactions, except for the synthesis of 1, were performed
under a nitrogen atmosphere using standard Schlenk procedures
for air-sensitive compounds but the isolated mercurials are air-
stable. Methanol, isopropyl alcohol, and tert-butanol were dried
over calcium oxide and then distilled. 2,6-Dimethylpiperidine
quinolin-8-olate and 2-methylquinolin-8-olate have intermolec-
ular association in the liquid and gas phase[37] while the
intermolecular Hg?O distances in the solid state are in the
[35]
˚
˚
range 3.3–3.4 A. In the structures of 4 (Hg?O: 3.313 A) and
˚
[Hg(C6H4(OMe)-4)(C6F5)] (Hg?O: 3.411 A), Hg?O interac-
tions are proposed.[9,38] Consequently a Hg?O interaction is
plausible. Furthermore, the oxygen atom is displaced out of the
˚
and 2,6-di-iso-propylphenol were dried over KOH and 4 A
molecular sieves, respectively, and then distilled. [Hg(C6F5)2],
[NaOiPr], [NaOtBu], and [NaOdpp] were synthesised by liter-
ature methods.[39–42] All other reagents were used without fur-
ther purification. All NMR spectra were recorded using a Bruker
DPX300 spectrometer and the Top Spin NMR software on a
Windows NT workstation. The solvents used were either deu-
terated chloroform or deuterated acetone. The proton chemical
shifts were calibrated using the residual proton peak from the
deuterated solvents and the fluorine chemical shifts were cali-
brated using trichlorofluoromethane as an external reference.
The mercury chemical shift was calibrated using 1 M HgCl2 in
DMSO-d6 at ꢀ1501.6 ppm.[43]
˚
plane of the aromatic ring (0.126(7) A) towards the mercury
atom, consistent with an interaction.
Such an interaction is more apparent with 3 as the Hg?O
˚
distance is 2.981(2) A (Fig. 4a), which is well within the sum of
van der Waals radii with the minimum value for mercu-
ry.[29–33,36] Similar to 2, the oxygen atoms of 3 are also displaced
out of the plane of the aryl ring towards the mercury atom by
´
˚
0.109(5) A.
Apart from the Hg?O interaction, the assembly of this
structure contains similar interactions to 1, such as p–p
˚
stacking interactions (interplanar distance: 3.072(4) A)
˚
An electrothermal IA6304 apparatus was used to measure the
melting points of all the complexes and was calibrated against
benzoic acid (mp 1228C). The infrared spectra of the compounds
were recorded with a Cary 630 FTIR spectrometer using ATR
measurements in a range of 4000–600 cmꢀ1. All the electro-
spray ionisation mass spectrometry (ESI MS) measurements
were performed using a Micromass Platform II ESI-MS linked
with a Waters 2690 HPLC system. Each of the reported peaks of
Hg-containing ions correspond to the main peak (containing
202Hg) of a cluster with the expected isotopic pattern, as
determined by the program Molecular Weight Calculator.
The microanalysis for 1 was conducted by the Campbell
Microanalytical Laboratories, University of Otago, Dunedin,
New Zealand, while the other two complexes were analysed by
the ElementalAnalysisService, LondonMetropolitan University.
and F?F interactions (2.830(3) A). However, an intermolecu-
lar H?F interaction can also be observed in this assembly
˚
(Fig. 4b) at a distance of 2.576 A, fitting within the sum of van
der Waals radii of 2.67 A.[36] These interactions may also play a
˚
role in the coplanarity of the structure.
Comparison of 2 and 3 with 4, which contains the less bulky
4-OMe substituent, revealed some similar interactions such as
p–p stacking, Hg?O interactions, and F?H non-classical
hydrogen bonding.[9] However, unlike the coplanar complexes
2 and 3, the dihedral angle of 4 is 52.98.[9] One contributing
factor to this difference in angles may be the F?H(CH3) non-
classical hydrogen bonding. Complexes 2 and 3 contain more
than one methyl substituent at different angles compared with 4,
enabling hydrogen bonding without twisting of the rings,
whereas rotation may be required in 4 in order for the F?H

1258
G. B. Deacon, P. C. Junk, and J. Luu
X-Ray Crystallography
(vs), 842 (vs), 807 (s), 765 (w), 717 (m), 665 (w). dH (300 MHz,
(CD3)2CO) 1.18 (s, CH3). dF{1H} (282.4 MHz, (CD3)2CO)
Single crystals were loaded onto a fine glass fibre or cryoloop
using viscous hydrocarbon oil with the collections kept at 123K
using an open-flow N2 Oxford Cryosystem. A Bruker X8 Apex II
diffractometerwasused tocollect the data, whichwere processed
using the SAINT [44] program. Each dataset was empirically
corrected for absorption (SADABS)[45] and then merged. The
structures were solved using direct methods and refined by full-
matrix least-squares on all F2 data using SHELX-97[46] for 3 and
SHELX-2013[46] for 1 and 2 with the X-Seed graphical inter-
face.[47] All the hydrogen atoms attached to carbon were placed
in idealised positions and allowed to ride on the atom to which
they were attached. Crystal and refinement data are in Table 3.
3
ꢀ122.5 (m, with 199Hg satellites J(Hg–F) 430, 4F, F(2,6)),
ꢀ151.3 (m, 4F, F(3,5)). m/z (ESIꢀ): 221 (100 %, C6F4OtBuꢀ),
645 (2, [M þ H]ꢀ), 679 (8, [M þ Cl]ꢀ). Anal. Calc. for
C20H18F8O2Hg: C 37.36, H 2.82. Found: C 37.37, H 2.74 %.
Attempted Synthesis of [Hg(C6F4(Odip)-4)2]
A solution of NaOdip was made by treating sodium metal
˚
(0.090 g, 3.7 mmol) with dried (4 A molecular sieves) 2,6-di-
iso-propylphenol (15 mL). The 19F{1H} NMR spectrum of the
reaction mixture showed signals for C6F5H;[22] no Hg satellites
were observed. The solid product was recovered by evaporation
under vacuum.
m/z (ESIꢀ) mercury containing cluster 663 (10 %, [([Hg
(Odip)-4)2] þ OHꢀ þ 5H2O)]ꢀ, 713 (100, [([Hg(Odip)
(C6H2(CH(CH3)2)ONa)] þ OHꢀ þ 2H2O þ 2MeCN)]ꢀ.
General Procedure
[Hg(C6F5)2] was added to a three necked round bottom flask
containing the appropriate nucleophile and solvent. The mixture
was then heated at reflux under N2. Once cooled, the reaction
mixture was poured into 10 mL of HCl (10 %) and then extracted
with diethyl ether (30 mL) three times. The ether layers were
then combined, dried with anhydrous magnesium sulfate, and
then evaporated to dryness. The amounts of reagents and-
reaction times used for each reaction are listed in Table 1
and Table 2.
Attempted Synthesis of [Hg(C6F4(Odpp)-4)2]
Once the reaction had been performed (reaction 3, Table 2),
the resulting crude products were recrystallised from acetone
and the crystals were handpicked to obtain compounds 4 and
2,6-diphenylphenol. The 1H NMR signals matched the spectrum
of the starting material 2,6-diphenylphenol[48] and the 19F{1H}
NMR spectrum corresponded well to the literature data for 4.[49]
Hand-picked single crystals were identified by unit cell
measurements.
[Hg(C6F4(cyclo-NC5H10)-4)2] 1
This synthesis was performed without a N2 atmosphere.
Colourless crystals (0.18 g, 36 %), mp 167–1698C. nmax
(ATR)/cmꢀ1 2943 (m), 2852 (m), 2814 (m), 2707 (w), 1633
(m), 1451 (vs), 1380 (s), 1354 (s), 1319 (s), 1276 (m), 1224 (s),
1152 (m), 1098 (s), 1069 (s), 1032 (m), 1000 (s), 950 (vs), 904
(s), 873 (m), 850 (m), 828 (m), 777 (m), 717 (m). dH (300 MHz,
CDCl3) 1.67 (m, 12H, 3 ꢁ CH2 overlapping), 3.23 (m, 8H,
N–CH2). dF{1H} (282.4 MHz, CDCl3) ꢀ122.3 (m, with 199Hg
˚
3
˚
2,6-Diphenylphenol unit cell: a 11.65 A, b 18.399 A,
˚
˚
c 6.368 A, a ¼ b ¼ g ¼ 908, V 1308.145 A ; 4 unit cell:
˚
˚
˚
a 14.2692 A, b 7.0972 A, c 14.7601 A, a ¼ g ¼ 908, b 105.918,
3
V 1437.52 A . Both are in agreement with reported data.
[9,50]
˚
Attempted Synthesis of HC6F4(NC5H8Me2-2,6)-4
C6F5H (0.1 mL) and 1 mL of cis-2,6-dimethylpiperidine
were placed in an NMR tube and sonicated. A 19F{1H} NMR
spectrum of the reaction mixture was recorded at 1 h intervals
for 3 h and then the solution was left to sonicate overnight and
only C6F5H was detected. The NMR tube was then taken out and
left standing at room temperature for one week and was heated
for 3 h at 808C.
satellites 3J(Hg–F) 406, 4F, F(2,6)), ꢀ149.3 (m, 4F, F(3,5)). d
199Hg
3
(71.6 MHz, CDCl3) ꢀ774.3 (p, J(Hg–F) 410). m/z (ESIþ) 667
(100 %, [M þ H]þ). Anal. Calc. for C22H20F8N2Hg: C 39.74, H
3.03, N 4.21. Found: C 40.00, H 2.95, N 4.12 %.
Attempted Synthesis of [Hg(C6F4(NC5H8Me2-2,6)-4)2]
The same spectra were observed after one week of standing
and then after 3 h of heating:
A
19F{1H} NMR spectrum of the reaction mixture was
recorded and the signals observed indicated the formation of
C6F5H;[22] no Hg satellites were observed.
dF{1H} (282.4 MHz, (CD3)2CO): ꢀ139.0 (m, 2F, F(2,6)
(C6F5H)), ꢀ139.9 (m, 1F, (C6F4H2-p)), ꢀ141.9 (m, 0.016F,
F(2,6) (HC6F4(NC5H8Me2-2,6)-4)), ꢀ152.0 (m, 0.018F, F(3,5)
ꢀ152.0 (HC6F4(NC5H8Me2-2,6)-4)), ꢀ154.0 (m, 1F, F(4)
(C6F5H)), ꢀ162.5 (m, 2F, F(3,5) (C6F5H)).
[Hg(C6F4(OiPr)-4)2] 2
Colourless crystals (0.38 g, 67 %), mp 121–1228C. nmax
(ATR)/cmꢀ1 2986 (m), 2926 (m), 2883 (m), 1633 (m), 1470
(vs), 1367 (s), 1337 (s), 1268 (w), 1177 (m), 1145 (m), 1080 (vs),
956 (vs), 901 (s), 865 (w), 824 (s), 797 (s), 733 (m), 693 (w), 659
(w). dH (300 MHz, CDCl3) 1.39 (d, 3J(H–H) 6, 12H, CH3), 4.60 (s,
3J(H–H) 6, 2H, CH). dF{1H} (282.4 MHz, CDCl3) ꢀ122.0 (dm,
3J(F–F) 15, with 199Hg satellites 3J(Hg–F) 413, 4F, F(2,6)), ꢀ153.7
(dm, 3J(F–F) 16, 4F, F(3,5)). m/z (ESIꢀ) 207 (100 %, C6F4OiPrꢀ),
651 (28, [Mþ Cl]ꢀ). Anal. Calc. for C18H14F8O2Hg: C 35.16,
H 2.29. Found: C 35.17, H 2.02 %.
Crystallographic Data
X-Ray data have been deposited with the Cambridge Crystal-
lographic Data Centre, CCDC Nos. 1 936417, 2 936418 and 3
936416. Copies of this information may be obtained free of
charge from The Director, CCDC, 12 Union Road, Cambridge
CB2 1EZ, UK (fax þ44-1223-336-033; email deposit@ccdc.
cam.ac.uk or http://www.ccdc.cam.ac.uk.)
[Hg(C6F4(OtBu)-4)2] 3
Supplementary Material
The 19F{1H} NMR spectra obtained for both 1 and the initial
attempt to syntheses 1 in ethanol, as well as the crystal packing
diagram for 1 are available on the Journal’s website.
Colourless crystals (0.096 g, 32 %), mp 210–2128C. nmax
(ATR)/cmꢀ1 2982 (m), 2839 (w), 2185 (w), 1631 (m), 1463
(vs), 1396 (m), 1368 (s), 1262 (m), 1161 (m), 1078 (vs), 952

Nucleophilic Aromatic Substitution of Bis(pentafluorophenyl)mercury
1259
Acknowledgements
[25] M. L. Cole, G. B. Deacon, P. C. Junk, K. Konstas, P. W. Roesky, Eur.
J. Inorg. Chem. 2005, 1090. doi:10.1002/EJIC.200400932
[26] D. Grdenic´, Q. Rev. Chem. Soc. 1965, 19, 303. doi:10.1039/
QR9651900303
[27] L. E. Sutton, Tables of Interatomic Distances and Configuration in
Molecules and Ions 1958, Spec. Publ. No. 11 (The Chemical Society:
London).
The authors thank the Faculty of Science for provision of a Faculty of Sci-
ence Dean’s scholarship (J.L.) and also Dr Chris S. Hawes and Dr Craig
M. Forsyth for their assistance with X-ray crystallography. In addition, they
are grateful to Rory P. Kelly for all of his assistance.
References
[28] L. Pauling, The Nature of the Chemical Bond and the Structure of
Molecules and Crystals 1960 (Cornell University Press: Ithaca, NY).
[29] A. J. Canty, G. B. Deacon, Inorg. Chim. Acta 1980, 45, L225.
doi:10.1016/S0020-1693(00)80155-3
[30] P. Pyykko¨, M. Straka, Phys. Chem. Chem. Phys. 2000, 2, 2489.
doi:10.1039/B001171L
[1] M. Hudlicky, Chemistry of Organic Fluorine Compounds: A Labora-
tory Manual with Comprehensive Literature Coverage 1970 (Ellis
Horwood: Chichester).
[2] R. D. Chambers, Fluorine in Organic Chemistry 2004 (Wiley-
Blackwell: Oxford).
[3] R. E. Banks, B. E. Smart, J. C. Tatlow, Organofluorine Chemistry,
Principals and Commercial Application 1994 (Plenum Press:
New York, NY).
[31] M. Tsunoda, F. P. Gabbai, J. Am. Chem. Soc. 2000, 122, 8335.
doi:10.1021/JA0016598
[32] M. R. Haneline, M. Tsunoda, F. P. Gabbai, J. Am. Chem. Soc. 2002,
[4] J. Burdon, P. L. Coe, M. Fulton, J. C. Tatlow, J. Chem. Soc. 1964, 2673.
doi:10.1039/JR9640002673
[5] C. Baldoli, P. DelButtero, E. Licandro, S. Maiorana, Gazz. Chim. Ital.
124, 3737. doi:10.1021/JA0166794
[33] M. R. Haneline, R. E. Taylor, F. P. Gabbai, Chem. – Eur. J. 2003,
9, 5188. doi:10.1002/CHEM.200305109
[34] K. R. Flower, V. J. Howard, S. Naguthney, R. G. Pritchard,
J. E. Warren, A. T. McGown, Inorg. Chem. 2002, 41, 1907.
doi:10.1021/IC010881Q
[35] C. L. Raston, B. W. Skelton, A. H. White, Aust. J. Chem. 1978, 31, 537.
doi:10.1071/CH9780537
[36] A. Bondi, J. Phys. Chem. 1964, 68, 441. doi:10.1021/J100785A001
[37] R. J. Bertino, G. B. Deacon, J. M. Miller, Aust. J. Chem. 1978, 31, 527.
doi:10.1071/CH9780527
[38] G. B. Deacon, P. C. Junk, J. Chem. Crystallogr. 2003, 33, 605.
doi:10.1023/A:1024945826698
[39] G. B. Deacon, J. E. Cosgriff, E. T. Lawrenz, C. M. Forsyth,
D. L. Wilkinson, in Hermann-Brauer Synthetic Methods of Organo-
metallic and Inorganic Chemistry (Ed. W. A. Herrmann) 1997, Vol. 6
(Ed. F. T. Edelmann), pp. 48–49 (G. Thieme: Stuttgart).
[40] A. Agarwal, K. Srivastava, S. K. Puri, P. M. Chauhan, Bioorg. Med.
Chem. Lett. 2005, 15, 3133. doi:10.1016/J.BMCL.2005.04.011
[41] F. Montanari, P. Tundo, J. Org. Chem. 1982, 47, 1298. doi:10.1021/
JO00346A027
1988, 118, 409.
[6] F. W. G. Fearon, H. Gilman, J. Organomet. Chem. 1967, 10, 535.
doi:10.1016/S0022-328X(00)83184-8
[7] R. D. Chambers, J. A. Cunningham, D. A. Pyke, Tetrahedron 1968, 24,
2783. doi:10.1016/S0040-4020(01)82550-3
[8] R. D. Chambers, J. A. Cunningham, D. J. Spring, J. Chem. Soc. C 1968,
1560. doi:10.1039/J39680001560
[9] G. B. Deacon, C. M. Forsyth, D. M. M. Freckmann, G. Meyer,
D. Stellfeldt, Z. Anorg. Allg. Chem. 2000, 626, 540. doi:10.1002/(SICI)
1521-3749(200002)626:2,540::AID-ZAAC540.3.0.CO;2-M
[10] D. S. Brown, A. G. Massey, D. A. Wickens, J. Organomet. Chem.
1980, 194, 131. doi:10.1016/S0022-328X(00)86660-7
[11] D. Naumann, F. Schulz, Z. Anorg. Allg. Chem. 2005, 631, 122.
doi:10.1002/ZAAC.200400244
[12] B. Ziolkovska, R. M. Myasnikova, A. I. Kitaigorodskii, J. Struct.
Chem. 1965, 5, 678. doi:10.1007/BF00744055
[13] D. Grdenic´, B. Kamenar, A. Nagl, Acta Crystallogr. B 1977, 33, 587.
doi:10.1107/S056774087700418X
[14] C. Glidewell, J. N. Low, J. L. Wardell, Acta Crystallogr. C 2005, 61,
[42] D. J. Darensbourg, J. C. Yoder, G. E. Struck, M. W. Holtcamp,
J. D. Draper, J. H. Reibenspies, Inorg. Chim. Acta 1998, 274, 115.
doi:10.1016/S0020-1693(97)05962-8
m107. doi:10.1107/S0108270104034134
[15] G. B. Deacon, P. W. Felder, P. C. Junk, K. Mu¨ller-Buschbaum,
T. J. Ness, C. C. Quitmann, Inorg. Chim. Acta 2005, 358, 4389.
doi:10.1016/J.ICA.2005.05.014
[43] M. A. Sens, N. K. Wilson, P. D. Ellis, J. D. Odom, J. Magn. Reson.
1975, 19, 323.
[16] G. B. Deacon, C. M. Forsyth, P. C. Junk, E. I. Izgorodina, T. J. Ness,
G. Meyer, I. Pantenburg, CrystEngComm 2011, 13, 88. doi:10.1039/
C0CE00258E
[17] N. R. Kunchur, M. Mathew, Chem. Commun. 1966, 71.
[18] D. L. Wilkinson, J. Riede, G. Mu¨ller, Z. Naturforsch B 1991, 46, 285.
[19] V. V. Bardin, L. S. Pressman, L. N. Rogoza, G. G. Furin, J. Fluor.
Chem. 1991, 53, 213. doi:10.1016/S0022-1139(00)82342-8
[20] N. J. Barassi, G. B. Deacon, J. A. Weigold, Z. Anorg. Allg. Chem. 1994,
620, 993. doi:10.1002/ZAAC.19946200607
[44] Bruker AXS, Apex2 v 2.0-2005 2005 (Bruker AXS: Madison, WI).
[45] G. M. Sheldrick, SADABS: Program for Scaling and Absorption Correc-
tion of Area Detector Data 1997 (Universita¨t Go¨ttingen: Go¨ttingen).
[46] G. M. Sheldrick, Acta Crystallogr. A 2008, 64, 112. doi:10.1107/
S0108767307043930
[47] L. J. Barbour, J. Supramol. Chem. 2001, 1, 189. doi:10.1016/S1472-
7862(02)00030-8
[48] D.-H. Lee, J.-Y. Jung, M.-J. Jin, Chem. Commun. 2010, 46, 9046.
doi:10.1039/C0CC03535A
[21] W. McFarlane, J. Chem. Soc.
A 1968, 2280. doi:10.1039/
[49] G. B. Deacon, D. Tunaley, J. Organomet. Chem. 1978, 156, 403.
doi:10.1016/S0022-328X(00)93546-0
[50] K. Nakatsu, H. Yoshioka, K. Kunimoto, T. Kinugasa, S. Ueji, Acta
J19680002280
[22] M. I. Bruce, J. Chem. Soc. A 1968, 1459. doi:10.1039/J19680001459
[23] R. D. Chambers, G. E. Coates, J. G. Livingstone, W. K. R. Musgrave,
J. Chem. Soc. 1962, 4367. doi:10.1039/JR9620004367
Crystallogr. B 1978, 34, 2357. doi:10.1107/S0567740878008183
[24] H. B. Albrecht, G. B. Deacon, M. J. Tailby, J. Organomet. Chem. 1974,
70, 313. doi:10.1016/S0022-328X(00)83362-8