The complex reactivity of β-diketiminato magnesium(i) dimers towards pinacolborane: Implications for catalysis

Article abstract of DOI:10.1039/c9dt01085h

Reactions between the borane, HBpin (pin = pinacolato), and three magnesium(i) dimers, [{(^ArNacnac)Mg-}₂] (^ArNacnac = [(ArNCMe)₂CH]^-; Ar = xylyl (Xyl), mesityl (Mes) or 2,6-diethylphenyl (Dep)), have been carried out in 2:1, 5:1 and 20:1 ratios. In all cases, NMR spectroscopic studies have revealed complex mixtures of many known and novel products from these reactions. From extracts of the crude reaction mixtures, low yields of ten compounds, representing six different types of complexes between β-diketiminato magnesium fragments and boron containing ligands, have been isolated and crystallographically characterised. These include unprecedented examples of compounds in which the γ-carbon of the β-diketiminate ligand has been activated by boron hydride fragments. In addition, boryloxide (OBpin), borate ([B(pin)₂]^- or [(pin)BH₂]^-), B-O bond ruptured [pinBH₂]^-, a diborane(5) dianion, or BH₃ have been shown to be incorporated into the isolated complexes. The complexity of the products of the reported reactions are discussed in light of recently published patents and papers which report that magnesium(i) dimers act as efficient catalysts, or as pre-catalysts to well defined catalysts, for the hydroboration of a variety of unsaturated substrates by HBpin. Our results strongly suggest that magnesium(i) dimers are not catalysts in these reactions, and that there are many more potential (pre-)catalysts that are generated in these reactions, than have previously been reported.

Full text of DOI:10.1039/c9dt01085h

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: C. Jones, D.
Jones and A. J. R. Matthews, Dalton Trans., 2019, DOI: 10.1039/C9DT01085H.
Volume 45 Number
1
7
January 2016 Pages 1–398
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Dalton
Transactions
An international journal of inorganic chemistry
www.rsc.org/dalton
Accepted Manuscripts are published online shortly after
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 Terms & 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

Page 1 of 19
Dalton Transactions
1
View Article Online
DOI: 10.1039/C9DT01085H
The Complex Reactivity of β-Diketiminato Magnesium(I) Dimers Towards Pinacolborane:
Implications for Catalysis
a
a
,a
Dafydd D. L. Jones, Aidan J. R. Matthews and Cameron Jones*
Correspondence:
Prof. Cameron Jones
School of Chemistry
PO Box 23
Monash University
VIC, 3800
Australia
a
School of Chemistry, PO Box 23, Monash University, VIC, 3800, Australia.
E-mail: cameron.jones@monash.edu
Web: http://www.monash.edu/science/research-groups/chemistry/jonesgroup

Dalton Transactions
Page 2 of 19
2
View Article Online
DOI: 10.1039/C9DT01085H
Abstract
Reactions between the borane, HBpin (pin = pinacolato), and three magnesium(I) dimers,
Ar
Ar
-
[
{( Nacnac)Mg} ] ( Nacnac = [(ArNCMe) CH] ; Ar = xylyl (Xyl), mesityl (Mes) or 2,6-
2
2
diethylphenyl (Dep)), have been carried out in 2:1, 5:1 and 20:1 ratios. In all cases, NMR
spectroscopic studies have revealed complex mixtures of many known and novel products from
these reactions. From extracts of the crude reaction mixtures, low yields of ten compounds,
representing six different types of complexes between -diketiminato magnesium fragments and
boron containing ligands, have been isolated and crystallographically characterised. These include
unprecedented examples of compounds in which the -carbon of the -diketiminate ligand has been
-
activated by boron hydride fragments. In addition, boryloxide (OBpin), borate ([B(pin) ] or
2
-
-
[
(pin) BH ] ), B−O bond ruptured [pinBH ] , a diborane(5) dianion, or BH have been shown to be
2 2 2 3
incorporated into the isolated complexes. The complexity of the products of the reported reactions
are discussed in light of recently published patents and papers which report that magnesium(I)
dimers act as efficient catalysts, or as pre-catalysts to well defined catalysts, for the hydroboration
of a variety of unsaturated substrates by HBpin. Our results strongly suggest that magnesium(I)
dimers are not catalysts in these reactions, and that there are many more potential (pre-)catalysts
that are generated in these reactions, than have previously been reported.
Introduction
Since the turn of the millennium considerable progress been made in the field of organic
1
transformations catalysed by alkaline-earth metal complexes. This has been driven in large part by
the high natural abundance of the lighter group 2 metals (magnesium and calcium), their low
toxicity, and their cheap cost compared to the expensive and toxic noble metals conventionally
2
employed in similar catalytic protocols. Many of the recent advances in this area have come from

Page 3 of 19
Dalton Transactions
3
the groups of Hill, Harder and Okuda, who have utilised isolated, or in situ generated, molecular
View Article Online
DOI: 10.1039/C9DT01085H
calcium and magnesium hydride compounds, e.g. 1 and 2 (Figure 1), as, for example, highly
3
efficient catalysts for hydroelementations of a variety of unsaturated substrates. Compounds 1 and
2
are typically thermally stable, resistant to redistribution reactions, and can be readily prepared by
-bond metathesis reactions between -diketiminato metal alkyls or amides, and hydride sources,
e.g. PhSiH3.4,5
Our own interests in group 2 metal chemistry lie largely with magnesium(I), e.g. 3, and
magnesium(II) hydride complexes, e.g. 1 (M = Mg), both of which we have extensively studied as
6
reagents for catalytically relevant small molecule activations and hydromagnesiations. With
respect to catalysis itself, we have begun to explore the potential of magnesium(I) compounds in
oxidative addition/reductive elimination based catalytic processes. Although this may seem a
formidable task, we have recently shown that compounds such as 3 can oxidatively add across
7
alkenes, and these reactions are readily reversible at room temperature. The resulting 1,2-
dimagnesioethanes have been shown to be powerful reagents for small molecule activations and
organometallic transformations in their own right.
Ar
N
Ar
N
Ar
N
Ar
N
THF
M
H
H
H
M
M
M
N
N
H
N
N
Ar
Ar THF
Ar
Ar
1
2
M = Mg or Ca
Ar = bulky aryl
Ar
Ar
Bpin
O
Dip
N
Xyl
N
N
N
Mg Mg
Mg
N
Mg
N
N
O
N
Ar
Ar
Xyl
Dip
BPin
3
Ar = Xyl, Mes or Dep
4
Figure 1. Catalytically active magnesium and calcium hydride complexes, 1 and 2; examples of
previously reported magnesium(I) dimers, 3; and an unsymmetrical magnesium boryloxide
compound, 4 (xylyl (Xyl), mesityl (Mes), 2,6-diethylphenyl (Dep), or 2,6-diisopropylphenyl
(Dip)).

Dalton Transactions
Page 4 of 19
4
In addition to this work, in a series of patents Ma and co-workers have reported that one of
View Article Online
DOI: 10.1039/C9DT01085H
our magnesium(I) compounds (3 Ar = Xyl), and unsymmetrical analogues developed by
themselves, act as efficient catalysts for the hydroboration of, for example, aldehydes, ketones,
alkynes and isocyanates, using pinacolborane (HBpin).8-10 The implication here is that the
magnesium(I) compounds are returned after these catalysed reactions are complete, and therefore
they display no net reaction with the substrate or pinacolborane. Intriguingly, in a subsequent
publication Ma reports that, at least in alkyne hydroborations, an unsymmetrical magnesium(I)
compound, [{(Dip/XylNacnac)Mg} ], is in fact a pre-catalyst, initially yielding the dimeric
2
boryloxide bridged complex, 4, via decomposition of pinacolborane.11 Compound 4 itself is
identified as a pre-catalyst, and is said to react further with HBpin to yield a magnesium hydride, cf.
1
(M = Mg), which is the true catalyst in the reaction. This result is akin to previous work on
magnesium hydride catalysed transformations of isocyanates involving HBpin, described by Hill
1
2
and co-workers. Surprisingly, however, Ma reports that 4 can be formed in moderate isolated
yield (36%) by reaction of the unsymmetrical magnesium(I) dimer with a ca. 10 fold excess of both
HBpin and phenylacetylene after heating at 90 °C for 12 hours, i.e. essentially under the catalytic
conditions they employed for alkyne hydroboration reactions. The fact that 4 is apparently not
converted to the corresponding magnesium hydride under these forcing conditions made us wonder
if 4 does in fact act as a pre-catalyst in their hydroboration reactions.
Coincidentally to Ma's patents and report on the magnesium(I) catalysed and pre-catalysed
hydroborations of unsaturates, we were exploring reactions of a series of magnesium(I) dimers with
HBpin under various stoichiometries. We found that all reactions led to complex mixtures of many
products, a number of which we were able to isolate in low yields, and characterise
crystallographically. Although limited spectroscopic data could be obtained on these, we believed it
worthwhile to report our results here, as these suggest that magnesium(I) dimers are not catalysts in
the hydroboration of unsaturates, and that anyone of a number of products of the reactions of
magnesium(I) dimers with HBpin could be acting as (pre-)catalysts or stoichiometric reactants in
those hydroboration reactions.

Page 5 of 19
Dalton Transactions
5
Results and Discussion
View Article Online
DOI: 10.1039/C9DT01085H
So as to ascertain the stability of magnesium(I) dimers in the presence of various excess
stoichiometries of HBpin, and thus the likelihood of such dimers potentially acting as catalysts for
the hydroboration of unsaturates, toluene solutions of three magnesium(I) dimers, 3 (Ar = Xyl, Mes
or Dep), were treated with either 2, 5 or 20 equivalents of HBpin at room temperature. After stirring
1
3
1
11
1
the reaction mixtures overnight, volatiles were removed in vacuo, and H and B{ H} NMR
1
1
1
spectra of C D solutions of the residues obtained (see Figure 2 for a representative B{ H} NMR
6
6
spectrum and the ESI for all other spectra). In all cases, complex mixtures of many products were
obtained, which seemingly did not include the magnesium(I) dimer starting material, or the -
Ar
14
diketiminato magnesium hydride complexes, [{( Nacnac)Mg(-H)} ], which are akin to the
2
magnesium hydride proposed by Ma and co-workers as the active catalyst in their later paper on the
1
1
hydroboration of alkynes. That said, it is not without the realms of possibility that magnesium
Ar
hydrides [{( Nacnac)Mg(-H)} ] are initially formed in the reactions described here, but these
2
subsequently react with other components of the reaction mixture (see below). Of note is the fact
Ar
that when the above reactions between [{( Nacnac)Mg−} ] and HBpin were carried out in 1:1
2
stoichiometries, similar product mixtures were obtained, but some unreacted magnesium(I) dimer
Ar
4a
was returned, in addition to small quantities of the homoleptic complexes, [Mg( Nacnac) ].
2
1
1
1
Figure 2. B{ H} NMR spectrum of the total non-volatile residue remaining after preliminary
work-up of the reaction between 3 (Ar = Mes) and 20 equivalents of HBpin.

Dalton Transactions
Page 6 of 19
6
1
1
1
Examination of the B{ H} NMR spectra of all the reaction mixtures (see ESI and Figure 2)
View Article Online
DOI: 10.1039/C9DT01085H
allowed the assignment of several known boron containing species. Firstly, and not surprisingly,
small amounts of residual HBpin were present, having not been completely removed during work-
up. In most of the reactions the major product was B pin , identified by a strong broad resonance
2
2
centred at ca.  31.8 ppm. Indeed, from several reactions, this diborane crystallised from the
reaction solution, and was structurally authenticated by X-ray crystallography. It can be speculated
that B pin is generated in these reactions by reduction of HBpin by the magnesium(I) dimer,
2
2
Ar
generating transient [{( Nacnac)Mg(-H)} ], which reacts further with other components of the
2
1
1
1
mixture. The only other signals in the B{ H} NMR spectra of the reaction mixtures that can be
assigned with any confidence are resonances at quite high field (ca.  -41 to -42 ppm) which likely
-
15
correspond to species incorporating the [BH ] anion. These signals are typically more intense in
4
the reactions involving larger excesses of HBpin. Although it cannot be sure what the mechanism of
-
formation of [BH ] is in these reactions, it is of note that the anion has previously been detected as
4
a product of reactions between a -diketiminato calcium hydride, 2 (M = Ca, Ar = Dip), and the
1
6
-
borane HBcat (cat = catecholato), which is closely related to HBpin. Moreover, [BH ] salts are
4
1
7
well known to reduce reactive unsaturates such as carbonyls, in the absence of catalysts. This
raises the question as to whether tetrahydridoborate is acting, at least to some extent, as an in situ
generated stoichiometric reductant in the supposedly magnesium(I) catalysed reductions reported by
Ma (e.g. hydroborations of aldehydes and ketones catalysed by 3 Ar = Xyl8).
As mentioned above, the complexity of reactions between magnesium(I) dimers and HBpin
makes it difficult, if not impossible, to determine what the catalytically active product(s) of these
reactions is(are), if there are any at all. Notwithstanding this complexity, we have isolated a number
of interesting, mechanistically informative, but low yielding, boron containing complexes from the
reactions, the X-ray crystal structures and limited spectroscopic data of which we describe below.
We believe these results can act as an aid to researchers who might wish to carry out similar
chemistry.

Page 7 of 19
Dalton Transactions
7
View Article Online
DOI: 10.1039/C9DT01085H
(i) Isolated products from the reactions of magnesium(I) dimers with 2 equivalents of HBpin
The non-volatile residues from the 2:1 reactions of HBpin with 3 (Ar = Xyl or Mes) were
extracted with hexane and the extracts stored at -30 °C overnight, to yield a few crystals of
compounds 5 and 6 respectively (Scheme 1). It is interesting to note that the complexes are
analogues of complex 7, which was previously reported by Hill and co-workers to be generated
Dip
18
from the reaction of [{( Nacnac)Mg(-H)} ] with B pin . The similarities here provide good
2
2
2
Ar
circumstantial evidence that 5 and 6 are formed by reaction of B pin with [{( Nacnac)Mg(-H)} ]
2
2
2
(Ar = Xyl or Mes), which are confirmed products and postulated transients, respectively, in the
reactions between 3 (Ar = Xyl or Mes) and HBpin. If this is the case the B-B bond of the diborane
is retained, whilst rupture of one of the Bpin fragments occurs upon transfer of two hydrides to the
diborane by the magnesium hydride species.
Ar
Ar
N
Ar
N
Dip
N
Dip
N
Ar
N
O
H
H
N
H
H
2
HBpin
O
B pin
2 2
B
B
Mg
Mg Mg
Mg
Mg
N
Mg
N
N
N
N
N
Ar
O
O
Ar
Ar
Dip
Dip
Ar
3
Ar = Xyl or Mes
5
6
7
Ar = Xyl
Ar = Mes
Ar = Dip
Scheme 1. Synthesis of compounds 5 and 6, and previously reported synthesis of 7.
Although no spectroscopic data could be obtained for 5 and 6 due to their very low yields,
X-ray crystal structure determinations of both were carried out. The quality of the diffraction data
for 6 was too low to include details of its crystal structure here, but its gross molecular framework is
unambiguous, and the compound is isostructural to 5 (and 7). The molecular structure of 5 is
depicted in Figure 3. The hydride ligands of the compound were located from difference maps and
refined isotropically, which allows formulation of the compound as incorporating a dihydro-
Xyl
+
diborane(5) dianion which bridges two cationic ( Nacnac)Mg fragments. The only specific

Dalton Transactions
Page 8 of 19
8
precedent for compounds 5 and 6 lies with 7, the metrical parameters of the central core of which
View Article Online
DOI: 10.1039/C9DT01085H
1
8
are similar to those of 5 (e.g. the B-B distance in 7 = 1.713(2) Å ). The Mg···B distances in 5 are
not consistent with bonding interactions, as they are significantly outside normal Mg−B distances
in, for example, magnesium boryl complexes.19
Figure 3. Molecular structure of 5 (25% thermal ellipsoids are shown; hydrogen atoms, except
hydrides, omitted). Selected bond lengths (Å) and angles (°): Mg(1)-O(3) 1.955(4), Mg(1)-O(1)
2
.049(4), B(1)-B(2) 1.725(9), Mg(2)-O(4) 1.860(4), Mg(2)-H(3) 1.97(5), Mg(2)-H(4) 2.09(5), B(2)-
O(3) 1.521(6), B(2)-H(3) 1.20(5), B(2)-H(4) 1.10(5), N(2)-Mg(1)-N(1) 94.80(16), O(3)-Mg(1)-O(1)
1.91(14), N(4)-Mg(2)-N(3) 90.85(17), H(3)-Mg(2)-H(4) 53.5(19), O(3)-B(2)-B(1) 108.2(4), H(3)-
B(2)-H(4) 105(3).
9
(ii) Isolated products from the reactions of magnesium(I) dimers with 5 equivalents of HBpin
The residues from the 5:1 reactions of HBpin with 3 (Ar = Xyl or Mes) were again extracted
with hexane and the extracts stored at -30 °C overnight, to yield the colourless crystalline
compounds 8-11 (Scheme 2). The first crop of crystals from each reaction (viz. 8 and 9) were
isolated in very low yields (5% and 3% respectively), while the second crops largely comprised 10
and 11, which were obtained in markedly higher isolated yields (27% and 32% respectively). In line

Page 9 of 19
Dalton Transactions
9
1
1
1
with these yields is the fact that signals for both compounds were observed in the B{ H} NMR
View Article Online
DOI: 10.1039/C9DT01085H
spectra of the crude reaction mixtures that yielded them.
The low yielding boryloxide complexes, 8 and 9, are direct analogues of Ma's
1
1
Dip
unsymmetrical analogue, 4, and an example, [{( Nacnac)Mg(-OBpin)} ], reported by Hill and
2
1
2
co-workers. It cannot be sure how 8 and 9 were formed, but it is worth noting that Ma suggests the
bridging oxide in 4 results from decomposition of pinacolborane, which seems a reasonable
proposition. It is also noteworthy that we found 8 and 9 to be unreactive towards excess HBpin, at
least at room temperature. To the best of our knowledge, the higher yielding products, 10 and 11,
have no precedent in the literature. Their formation can be speculated as occurring via reaction of in
Ar
situ generated [{( Nacnac)Mg(-H)} ] with two equivalents of HBpin, with subsequent B−O bond
2
rupture of one of the resultant H Bpin units, leading to a "dangling" Lewis acidic OBH moiety,
2
2
which is attacked by the nucleophilic -carbon of the Nacnac backbone. Although such Nacnac -
2
0
carbon functionalisations are relatively common, this is the first example involving boron.
O
Bpin
O
Ar
N
Ar
N
Ar
N
Ar
N
Ar
N
B
H
Ar O
H
5
HBpin
N
Mg
Mg Mg
Mg
Mg
+
Mg
N
N
N
N
N
O
N
O
Ar
Ar
Ar
Ar
Ar
Bpin
Ar
O
BH2
3
Ar = Xyl or Mes
8
9
Ar = Xyl
Ar = Mes
10 Ar = Xyl
11 Ar = Mes
Scheme 2. Synthesis of compounds 8-11.
Limited spectroscopic data could be obtained for the boryloxide compounds 8 and 9, but
1
their H NMR spectra are consistent with their solid state structures. No signals could be observed
in their 11B{ H} NMR spectra, presumably due to the low mass of sample available for these
1
1
experiments, and their relatively poor solubility in non-coordinating deuterated solvents. The H
1
3
1
and C{ H} NMR spectra of 10 and 11 are complicated due to the unsymmetrical nature of these
compounds, but again, are reminiscent of them retaining their solid state structures in solution. Of

Dalton Transactions
Page 10 of 19
1
0
particular note are signals at δ = 3.27 ppm for 10 and δ= 3.30 ppm for 11, which were assigned to
View Article Online
DOI: 10.1039/C9DT01085H
1
1
1
the borylated -CH units in the compounds. Moreover, The B{ H} NMR spectra of each show
two signals (10:  -0.32 ppm and -9.32 ppm; 11:  -0.22 ppm and -9.34 ppm) which lie in the
2
1
normal region for 4-coordinate boron. Two broad bands in the FTIR spectrum of each compound

1
1
1
1
(
10:  = 2312 cm and 2116 cm ; 11:  = 2293 cm and 2120 cm ) were assigned to B−H
stretching modes. It is presumed that the higher wavenumber bands in each spectrum are associated
with the terminal BH fragments.
2
The X-ray crystal structures of all compounds 8-11 were determined, and as it was found
that the pairs 8 and 9, and 10 and 11, are essentially isostructural, only the structures of 8 and 10 are
depicted in Figure 4. Full crystallographic details of the other two compounds can be found in the
ESI. The structures of 8 and 9 show them to be boryloxide bridged dimers which are very similar to
1
1
Dip
12
previously reported 4 and [{( Nacnac)Mg(-OBpin)} ]. The structures of 10 and 11 are more
2
novel, and reveal them to be unsymmetrical dimetallic systems in which the two magnesium centres
-
are bridged by a [H Bpin] anion which coordinates one Mg centre through a dative O−Mg bond,
2
and the other centre through two crystallographically refined bridging hydrides. The other ligand
that bridges the Mg centres is essentially a B−O bond cleaved version of H Bpin, in which the
2
cleaved O-centre bridges the Mg atoms, while the OBH fragment has been attacked by the
2
backbone -C centre of one Nacnac ligand. This leads to a loss of electronic delocalisation within
that ligand, which is reflected in the bond lengths within its N C backbone. Both of the B centres
2
3
in the complexes have distorted tetrahedral geometries with B-C and/or B-O bond lengths in the
normal ranges.22

Page 11 of 19
Dalton Transactions
1
1
View Article Online
DOI: 10.1039/C9DT01085H
(a)
(b)
Figure 4. Molecular structures of (a) 8 and (b) 10 (25% thermal ellipsoids are shown; hydrogen
atoms, except hydrides, omitted). Selected bond lengths (Å) and angles (°) for 8: Mg(1)-O(1)
1
1
.9605(13), Mg(1)-O(1)' 1.9886(14), Mg(1)-N(1) 2.0234(15), Mg(1)-N(2) 2.0307(15), O(1)-B(1)
.318(2), O(1)-Mg(1)-O(1)' 82.51(6), N(1)-Mg(1)-N(2) 94.49(6), Mg(1)-O(1)-Mg(1)' 97.49(6).
Selected bond lengths (Å) and angles (°) for 10: Mg(1)-O(1) 2.0095(10), Mg(1)-O(2) 2.0175(11),
Mg(1)-H(1) 2.020(17), Mg(1)-H(2) 2.013(19), O(1)-Mg(2) 1.9496(11), B(1)-H(1) 1.182(17), B(1)-
H(2) 1.199(18), Mg(2)-O(3) 2.0086(11), O(2)-B(2) 1.502(2), B(2)-C(28) 1.697(3), Mg(1)-O(1)
2
1
1
.0095(10), Mg(1)-O(2) 2.0175(11), Mg(1)-H(1) 2.020(17), Mg(1)-H(2) 2.013(19), O(1)-Mg(2)
.9496(11), B(1)-H(1) 1.182(17), B(1)-H(2) 1.199(18), Mg(2)-O(3) 2.0086(11), O(2)-B(2)
.502(2), B(2)-C(28) 1.697(3), O(1)-Mg(1)-O(2) 80.67(4), H(1)-Mg(1)-H(2) 56.2(7), Mg(2)-O(1)-
Mg(1) 124.16(5), H(1)-B(1)-H(2) 105.9(12), O(1)-Mg(2)-O(3) 100.93(4), O(2)-B(2)-C(28)
06.20(12).
1
(iii) Isolated products from the reactions of magnesium(I) dimers with 20 equivalents of HBpin
In order to more closely mimic the ratio of HBpin to magnesium(I) dimer that has been
utilised previously in catalytic studies, reactions between the three magnesium(I) compounds and
1
1
1
2
0 fold excesses of the borane were carried out at room temperature. In all cases the B{ H} NMR
spectra of the crude product mixtures were complex, but roughly resembled those for the 5:1

Dalton Transactions
Page 12 of 19
1
2
reactions. Despite this, compounds isolated from those previous reactions could not be isolated
View Article Online
DOI: 10.1039/C9DT01085H
here. Instead, three different compound types (compounds 12-15) were isolated as colourless
crystals in very low yields (Scheme 3).
Ar
Xyl
H
H2B
N
O
O
O
O
N
O
O
O
O
Mg
B
Mg
B
N
N
Ar
N
Ar
N
Ar
Xyl
+
20 HBpin
12 Ar = Mes
3 Ar = Dep
14
Mg Mg
1
N
N
O
Ar
O
B
Ar
Xyl
N
Xyl
3
Ar = Xyl, Mes or Dep
O
N
Mg
Mg
N
N
O
Xyl
Xyl
O
BH2
15
Scheme 3. Synthesis of compounds 12-15.
The very low yields of 12-15 precluded the acquisition of meaningful spectroscopic data,
but all compounds were crystallographically characterised. Compounds 12 and 13 are isostructural,
so only the molecular structure of 12 is depicted in Figure 5, in addition to the structures of 14 and
+
1
5. Complexes 12 and 13 can be formulated as contact ion pairs between [(Nacnac)Mg] cations
-
and the borate anion [B(pin) ] . To the best of our knowledge these are the first examples of
2
-
complexes between this borate and group 2 metals, though a potassium salt of [B(pin) ] has been
2
2
3
reported. It cannot be sure how 12 and 13 are formed, but it is pertinent that Harder and co-
-
workers have suggested the closely related borate [B(cat) ] can be generated via decomposition of
2
Dip
16
the borane HBcat with the calcium hydride complex, [{( Nacnac)Ca(THF)(-H)} ]. A similar
2
process could be in operation here, namely decomposition of HBpin by the likely transiently
Ar
generated complexes [{( Nacnac)Mg(-H)} ] (Ar = Mes or Dep). The metrical parameters within
2
the crystal structures of 12 and 13 are unexceptional and do not warrant further comment.
Compound 14 is somewhat similar to 12 and 13, except that its backbone -carbon centre
has been activated, presumably by nucleophilic attack of that carbon on an in situ generated BH3

Page 13 of 19
Dalton Transactions
1
3
-
complex or B H . The observation of [BH ] in the spectrum of the reaction mixture that gave 14
2
6
4
View Article Online
DOI: 10.1039/C9DT01085H
makes the transient generation of BH seem reasonable. If that is the case, BH is a potent
3
3
hydroboration reagent, and its participation in the previously reported magnesium(I) (pre-)catalysed
hydroborations of substrates cannot be ruled out. The formulation of 14 as an alkyl/trihydrido-
borate complex is confirmed by the location of all hydrides from difference maps, and their
unrestrained isotropic refinement. Both boron centres of the complex have distorted tetrahedral
geometries, as does the now saturated backbone -carbon of the XylNacnac ligand. In line with this
is electronic localisation of the two C=N double bonds of the ligand, as indicated by the lengths of
those bonds.
The molecular structure of compound 15 reveals it to be structurally similar to 10 and 11, in
that a ruptured Bpin unit coordinates both Mg centres through a bridging oxygen, whilst its OBH2
fragment has been attacked by the backbone -carbon of one XylNacnac ligand, leading to loss of
electronic delocalisation within the backbone of that ligand. The difference with 10 and 11 is that in
-
those compounds the other bridging ligand is the [pinBH ] anion, while in 15 it is a boryloxide,
2
similar to that found in 8 and 9.

Dalton Transactions
Page 14 of 19
1
4
View Article Online
DOI: 10.1039/C9DT01085H
(a)
(b)
(c)
Figure 5. Molecular structures of (a) 12, (b) 14 and (c) 15 (25% thermal ellipsoids are shown;
hydrogen atoms, except hydrides, omitted). Selected bond lengths (Å) and angles (°) for 12: Mg(1)-
O(3) 2.0126(15), Mg(1)-O(1) 2.0145(15), O(1)-B(1) 1.528(2), B(1)-O(2) 1.433(2), B(1)-O(4)
1
.443(2), B(1)-O(3) 1.526(2), O(3)-Mg(1)-O(1) 70.89(6), O(2)-B(1)-O(4) 111.18(16), O(3)-B(1)-
O(1) 99.79(14). Selected bond lengths (Å) and angles (°) for 14: Mg(1)-O(1) 1.9941(13), Mg(1)-
O(2) 2.0737(12), Mg(1)-H(9) 1.99(3), B(2)-C(23) 1.680(2), B(2)-H(9) 1.20(3), B(2)-H(6) 1.14(3),
B(2)-H(8) 1.17(3), Mg(1)-O(1) 1.9941(13), Mg(1)-O(2) 2.0737(12), Mg(1)-H(9) 1.99(3), B(2)-
C(23) 1.680(2), B(2)-H(9) 1.20(3), B(2)-H(6) 1.14(3), B(2)-H(8) 1.17(3), O(1)-Mg(1)-O(2)
6
9.23(5), N(1)-Mg(1)-N(2) 89.24(6), C(23)-B(2)-H(9) 109.6(11). Selected bond lengths (Å) and
angles (°) for 15: Mg(1)-O(4) 1.993(2), Mg(1)-O(1) 2.004(2), Mg(1)-O(5) 2.065(2), O(1)-Mg(2)
.012(2), Mg(2)-O(4) 1.926(2), B(2)-O(5) 1.496(4), B(2)-C(46) 1.690(5), Mg(1)-O(4) 1.993(2),
2

Page 15 of 19
Dalton Transactions
1
5
Mg(1)-O(1) 2.004(2), Mg(1)-O(5) 2.065(2), O(1)-Mg(2) 2.012(2), Mg(2)-O(4) 1.926(2), B(2)-O(5)
View Article Online
DOI: 10.1039/C9DT01085H
1
.496(4), B(2)-C(46) 1.690(5), O(4)-Mg(1)-O(1) 80.95(9), O(4)-Mg(1)-O(5) 77.79(9), O(4)-Mg(2)-
O(1) 82.37(9), O(5)-B(2)-C(46) 106.0(2).
Conclusion
In summary, reactions between the borane, HBpin, and three magnesium(I) dimers have
been carried out in 2:1, 5:1 and 20:1 ratios. In all cases, the NMR spectroscopic studies have
revealed complex mixtures of many products from these reactions. Several known compounds have
1
1
1
-
been identified by B{ H} NMR spectroscopy in these mixtures, e.g. B pin and [BH ] , while low
2
2
4
yields of ten compounds, representing six different compound types, have been crystallised from
extracts of the reaction mixtures. The crystallographic characterisation of all of these further
highlights the complexity of the product mixtures derived from reactions between magnesium(I)
dimers and HBpin. These results are important in light of recently published patents and papers
which report that magnesium(I) dimers act as efficient catalysts in their own right, or as pre-
catalysts on the way to well defined catalysts, for the hydroboration of a variety of unsaturated
substrates by HBpin. Although we did not attempt magnesium catalysed unsaturated substrate
hydroborations in this study, or did we examine reactions between magnesium(I) dimers and
excesses of relevant unsaturates,24 our results clearly show that caution must be used when
assigning the nature of the catalyst in reactions involving magnesium(I) dimers and excess HBpin.
Indeed, it is certain that magnesium(I) compounds do not act as catalysts in the previously reported
hydroborations. Moreover, while compounds such as 4, that arise from treating magnesium(I)
dimers with HBpin, might play some part as pre-catalysts in hydroboration, we have shown that
there are many other potential (pre-)catalysts that are generated in these reactions.

Dalton Transactions
Page 16 of 19
1
6
Electronic supplementary information (ESI) available: Experimental procedures and
View Article Online
DOI: 10.1039/C9DT01085H
characterisation data for all new compounds. Crystal data, details of data collections and
refinements. CCDC numbers: 1902608-1902616. For ESI and crystallographic data in CIF format
see DOI: 10.xxxxxxx
Acknowledgement
CJ thanks the Australian Research Council and the US Air Force Asian Office of Aerospace
Research and Development (grant FA2386-18-1-0125) for funding.
References
1
.
Selected reviews: (a) M. S. Hill, D. J. Liptrot, C. Weetman, Chem. Soc. Rev., 2016, 45, 972;
b) M. R. Crimmin, M. S. Hill in "Alkaline Earth Metal Compounds: Oddities and
Applications", Top. Organomet. Chem., vol. 45 (S. Harder ed.), Springer, Berlin, 2012, pp.
91-242; (c) S. Kobayashi, Y. Yamashita, Acc. Chem. Res., 2011, 44, 58; (d) S. Harder,
(
1
Chem. Rev., 2010, 110, 3852; (e) D. Mukherjee, D. Schuhknecht, J. Okuda, Angew. Chem.
Int. Ed., 2018, 57, 9590; (f) K. Revunova, G. I. Nikonov, Dalton Trans., 2015, 44, 840; (g)
C. C. Chong, R. Kinjo, ACS Catal., 2015, 5, 3238.
2
3
.
.
A. Yaroshevsky, Geochem. Int., 2006, 44, 48.
Selected recent references: (a) A. S. S. Wilson, C. Dinoi, M. S. Hill, M. F. Mahon, L.
Maron, Angew. Chem. Int., 2018, 57, 15500; (b) H. Bauer, K. Thum, M. Alonso, C. Fischer,
S. Harder, Angew. Chem. Int. Ed., 2019, 58, 4248; (c) H. Bauer, M. Alonso, C. Fischer, B.
Rösch, H. Elsen, S. Harder, Angew. Chem. Int. Ed., 2018, 57, 15177; (d) D. Schuhknecht, C.
Lhotzky, T. P. Spaniol, L. Maron, J. Okuda, Angew. Chem. Int. Ed., 2017, 56, 12367.
(a) S. J. Bonyhady, C. Jones, S. Nembenna, A. Stasch, A. J. Edwards, G. J. McIntyre, Chem.
Eur. J. 2010, 16, 938; (b) S. P. Green, C. Jones, A. Stasch, Angew. Chem. Int. Ed. 2008, 47,
4
.

Page 17 of 19
Dalton Transactions
1
7
9
079; (c) S. Harder, J. Brettar, Angew. Chem. Int. Ed. 2006, 45, 3474; (d) A. S. S. Wilson,
View Article Online
DOI: 10.1039/C9DT01085H
M. S. Hill, M. F. Mahon, C. Dinoi, L. Maron, Science, 2017, 358, 1168.
5
.
For an excellent review on magnesium hydride complexes and their synthesis, see: D.
Mukherjee, J. Okuda, Angew. Chem. Int. Ed., 2018, 56, 1458.
6
.
(a) C. Jones, Nat. Rev. Chem., 2017, 1, 0059; (b) A. Stasch, C. Jones, Dalton Trans., 2011,
40, 5659.
7.
(a) A. J. Boutland, A. Carroll, C. A. Lamsfus, A. Stasch, L. Maron, C. Jones, J. Am. Chem.
Soc., 2017, 139, 18190; (b) D. Dange, A. R. Gair, D. D. L. Jones, M. Juckel, S. Aldridge, C.
Jones, Chem. Sci., 2019, 10, 3208.
8.
(a) M. Ma, W. Wang, K. Lu, W. Yao, CN Pat., 107602595, 2018; (b) W. Yao, M. Ma, X. Shen
J. Li, CN Pat., 107556196, 2018; (c) M. Ma, H. Hua, X. Shen, J. Li, CN Pat., 108191891, 2018;
(d) M. Ma, W. Wang, M. Luo, Q. Xiao, L. Xu, CN Pat., 108358959, 2018;
9.
The same group has also reported the related magnesium(I) catalysed silylcyanation (using
Me SiCN) of carbonyls using 3 (Ar = Xyl). (a) M. Ma, W. Wang, W. Yao, CN Pat., 107556195,
3
2018; (b) W. Wang, M. Luo, S. A. Pullarkat, M. Ma, Chem. Commun., 2018, 54, 3042.
10.
With regard to the reports detailed in ref. 9, we have previously shown that magnesium(I) dimers
rapidly, and irreversibly, reductively couple or cleave Me SiCN. M. Ma, A. Stasch, C. Jones,
3
Chem. Eur. J., 2012, 18, 10669.
11.
12.
13.
J. Li, M. Luo, X. Sheng, H. Hua, W. Yao, S. A. Pullarkat, L. Xu, M. Ma, Org. Chem. Front.,
2
018, 5, 3538.
Y. Tang, M. D. Anker, J. Fang, M. F. Mahon, L. Maron, C. Weetman, M. S. Hill, Chem. Sci.,
017, 8, 3529.
2
Although stirred overnight, initial monitoring of the reactions by NMR spectroscopy showed
complete consumption of the magnesium(I) starting material within minutes, and the formation
of complex product mixtures.

Dalton Transactions
Page 18 of 19
1
8
1
4.
5.
(a) R. Lalrempuia, C. E. Kefalidis, S. J. Bonyhady, B. Schwarze, L. Maron, A. Stasch, C. Jones,
View Article Online
DOI: 10.1039/C9DT01085H
J. Am. Chem. Soc., 2015, 137, 8944; (b) S. J. Bonyhady, D. Collis, G. Frenking, N. Holzmann, C.
Jones, A. Stasch, Nat. Chem., 2010, 2, 865.
11
1
These signals were significantly broadened in the B NMR NMR spectra, relative to those in the
11
1
B{ H} NMR NMR spectra, but resolution into a quintet was not observed. We presume this is
-
due to the [BH ] anion being associated with one or more magnesium containing cations in non-
4
coordinating C D solutions.
6
6
1
6.
7.
S. Harder, J. Spielmann, J. Organomet. Chem., 2012, 698, 7.
1
L. Banfi, E. Narisano, R. Riva, E. W. Baxter, "Lithium Borohydride" e-EROS Encyclopedia of
Reagents for Organic Synthesis, Wiley, Hoboken, 2001.
1
8.
9.
A.-F. Pécharman , M. S. Hill, M. F. Mahon, Dalton Trans., 2018, 47, 7300.
(a) A.-F. Pécharman, A. L. Colebatch, M. S. Hill, C. L. McMullin, M. F. Mahon, C.
Weetman, Nat. Chem., 2017, 8, 15022; (b) M. Yamashita, Y. Suzuki, Y. Segawa, K. Nozaki,
J. Am. Chem. Soc., 2007, 129, 9570.
1
2
0.
1.
See for example: A. J. Boutland, I. Pernik, A. Stasch, C. Jones, Chem. Eur. J., 2015, 21,
1
5749.
2
H. Nöth, B. Wrackmeyer, "Nuclear Magnetic Resonance Spectroscopy of Boron
Compounds", No. 14 in the series " NMR-Basic Principles and Progress", P. Diehl, E. Fluck,
R. Kosfeld (eds.), Springer Verlag, Berlin, 1978.
2
2
2.
3.
As determined from a survey of the Cambridge Crystallographic Database, March, 2019.
S. Pietsch, E. C. Neeve, D. C. Apperley, R. Bertermann, F. Mo, D. Qiu, M. S. Cheung, L.
Dang, J. Wang, U. Radius, C. Kleeberg, Z. Lin, T. B. Marder, Chem. Eur. J., 2015, 21,
7
082.
2
4.
It is important to note that we have previously shown that magnesium(I) dimers rapidly
reduce stoichiometric amounts of ketones, alkynes and isocyanates, and these reactions are
irreversible. See refs 7 and 10, and C. Jones, L. McDyre, D. M. Murphy, A. Stasch, Chem.
Commun., 2010, 46, 1151.

Page 19 of 19
Dalton Transactions
1
9
Graphical Abstract
View Article Online
DOI: 10.1039/C9DT01085H
The Complex Reactivity of β-Diketiminato Magnesium(I) Dimers Towards Pinacolborane:
Implications for Catalysis
Dafydd D. L. Jones, Aidan Matthews and Cameron Jones*
Reactions of three β-diketiminato magnesium(I) dimers with various excesses of
pinacolborane are complex, yielding mixtures of many known and new products, including ten
structurally characterised magnesium-boron containing complexes (e.g. see picture). These results
bring into question mechanisms recently proposed for the hydroboration of unsaturated substrates,
involving magnesium(I) dimers as catalysts or pre-catalysts.
O
Ar
N
Ar
N
B
H
Ar
N
O
O
O
O
O
H
Mg
Mg
B
Mg
N
N
N
O
Ar
Ar
Ar
O
BH2
O
O
O
Ar
Ar
B
Ar
H
N
H2B
N
MgI dimers
N
O
O
O
O
Mg
Mg
+
Mg
B
N
excess HBpin
N
N
O
Ar
Ar
Ar
O
BH2
Ar = bulky aryl
Ar
N
Bpin
O
Ar
Ar
Ar
N
O
H
H
O
N
N
B
B
Mg
Mg
Mg
Mg
N
N
N
N
O
O
O
Ar
Ar
Ar
Ar
Bpin