Cooperative Al-H Bond Activation in DIBAL-H: Catalytic Generation of an Alumenium-Ion-Like Lewis Acid for Hydrodefluorinative Friedel-Crafts Alkylation

Article abstract of DOI:10.1021/jacs.7b09444

The Ru-S bond in Ohki-Tatsumi complexes breaks oligomeric DIBAL-H structures into their more reactive monomer. That deaggregation is coupled to heterolytic Al-H bond activation at the Ru-S bond, formally splitting the Al-H linkage into hydride and an alumenium ion. The molecular structure of these Lewis pairs was established crystallographically, revealing an additional Ru-Al interaction next to the Ru-H and Al-S bonds. That bonding situation was further analyzed by quantum-chemical calculations and is best described as a three-center-two-electron (3c2e) donor-acceptor σ(Ru-H) → Al interaction. Despite the extra stabilization of the aluminum center by the interaction with both the sulfur atom and the Ru-H bond, the hydroalane adducts are found to be stronger Lewis acids and electrophiles than the free ruthenium catalyst and DIBAL-H in its different aggregation states. Hence, the DIBAL-H molecule and its Al-H bond are activated by the Ru-S bond, but these hydroalane adducts are not to be mistaken as sulfur-stabilized alumenium ions in a strict sense. The Ohki-Tatsumi complexes catalyze C(sp³)-F bond cleavage with DIBAL-H, and the catalytic setup is applied to hydrodefluorinative Friedel-Crafts alkylations. A broad range of CF₃-substituted arenes is efficiently converted into unsymmetrical diarylmethanes with various arenes as nucleophiles. Computed fluoride-ion affinities (FIAs) of the hydroalane adducts as well as DIBAL-H in its aggregation states support this experimental finding.

Full text of DOI:10.1021/jacs.7b09444

Subscriber access provided by AUSTRALIAN NATIONAL UNIV
Article
Cooperative Al-H Bond Activation in DIBAL-H: Catalytic
Generation of an Alumenium-Ion-Like Lewis Acid
for Hydrodefluorinative Friedel-Crafts Alkylation
Francis Forster, Toni T. Metsanen, Elisabeth Irran, Peter Hrobarik, and Martin Oestreich
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b09444 • Publication Date (Web): 06 Oct 2017
Downloaded from http://pubs.acs.org on October 7, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 10
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Cooperative Al–H Bond Activation in DIBAL–H: Catalytic Generation
of an Alumenium‐Ion‐Like Lewis Acid for Hydrodefluorinative
Friedel–Crafts Alkylation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Francis Forster,† Toni T. Metsänen,†,‡ Elisabeth Irran,† Peter Hrobárik,*,†,§ and Martin
Oestreich*,†
† Institut für Chemie, Technische Universität Berlin, Strasse des 17. Juni 115, 10623 Berlin, Germany
§ Department of Inorganic Chemistry, Faculty of Natural Sciences, Comenius University, 84215 Bratislava, Slovakia
Supporting Information Placeholder
ABSTRACT: The Ru–S bond in Ohki–Tatsumi complexes break oligomeric DIBAL–H structures into their more reactive
monomer. That deaggreation is coupled to heterolytic Al–H bond activation at the Ru–S bond, formally splitting the Al–H
bond into hydride and an alumenium ion. The molecular structure of these Lewis pairs was established
crystallographically, revealing an additional Ru–Al interaction next to the Ru–H and Al–S bonds. That bonding situation
was further analyzed by quantum‐chemical calculations and is best described as a three‐center‐two‐electron (3c2e)
donor‐acceptor σ(Ru–H)→Al interaction. Despite its extra stabilization by the interaction with both the sulfur atom and
the Ru–H bond, the hydroalane adducts are found to be stronger Lewis acids and electrophiles than the free ruthenium
catalyst and DIBAL–H in its different aggregation states. Hence, the DIBAL–H molecule and its Al–H bond are activated
by the Ru–S bond but these hydroalane adducts are not to be mistaken as sulfur‐stabilized alumenium ions in a strict
sense. The Ohki–Tatsumi complexes catalyze C(sp3)–F bond cleavage with DIBAL–H, and the catalytic setup is applied to
hydrodefluorinative Friedel–Crafts alkylations. A broad range of CF3‐substituted arenes is efficiently converted into
unsymmetrical diarylmethanes with various arenes as nucleophiles. Computed fluoride‐ion affinities (FIAs) of the
hydroalane adducts as well as DIBAL–H in its aggregation states support this experimental finding.
combination of DIBAL–H as the hydride source and
various trityl salts [Ph3C]+[X]– {X = B(C6F5)4, Al(C6F5)4, or
Al[OC(CF3)3]4} as initiator or catalyst.7 However, Terao,
Kambe, and co‐workers showed in the same year that
stoichiometric DIBAL–H alone is able to convert these
monofluorides into alkanes.8 By replacing the
aforementioned counteranions [X]‒ in [Ph3C]+[X]– with
Reed’s [CHB11H5Br6]–, Ozerov and co‐workers then
achieved the hydrodefluorination of CF3 groups;
competing transfer of hydride or an alkyl substituent
from DIBAL–H was observed.9
INTRODUCTION
Examples of [R2Al–LB] (R = alkyl/aryl; LB = Lewis base)
are scarce while heteroatom‐substituted alumenium ions
as well as cyclopentadienyl complexes are well
established.1 The earliest example of an [R2Al]+‐type
cation is the ion‐like complex [Et2Al]+[CB11H6X6]– (X = Cl
or Br) disclosed by Kim and Reed.2 Weak bidentate Al…X
interactions stabilize its highly electron‐deficient
aluminum center. Wehmschulte and co‐workers later
accomplished the structural characterization of the meta‐
We together with Ohki and Tatsumi introduced the
coordinatively unsaturated ruthenium(II) complexes
[1]+[X]–10 for cooperative Si–H11,12 and B–H13 bond
activation (Scheme 1, top). The polar Ru–S bond in
[1]+[X]– was shown to split H–H,10 Si–H,11,12 and B–H13
bonds into hydride and the corresponding proton or
main‐group cation (Scheme 1, middle).14 We anticipated
that Al–H bonds could also undergo this bond activation,
thereby forming sulfur‐stabilized alumenium ions for
catalysis (Scheme 1, bottom).
terphenyl‐substituted
complex
[(2,6‐
Mes2C6H3)2Al]+[B(C6F5)4]– with the mesityl groups
sterically shielding and stabilizing the electron gap at the
aluminum center.3 Those examples were followed by
similar complexes such as [Dipp*AlEt]+[CB11H6Cl6]–,4a
([Et2Al]+)2[B12Cl12]2–,4b or {[Me2Al]+[1‐Me‐CB11F11]–}2.4c As to
that, Krossing’s recent review of reactive main‐group
cations provides an excellent summary of alumenium
ions.5
A
prominent application of strongly electrophilic
aluminum compounds is the hydrodefluorination of
C(sp3)–F bonds.6 Rosenthal and co‐workers reported the
hydrodefluorination of an alkyl fluoride by the
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 10
Scheme 1. Tethered Ru–S complexes [1]+[X]– and
Cooperative Bond Activation of E–H Bonds [E = H, Si,
B, and Al]
2
[1b∙iBu2AlH]+[B(C6F5)4]–
[1c∙iBu2AlH]+[B(C6F5)4]–
[1d∙iBu2AlH]+[B(C6F5)4]–
([1b∙iBu2AlH]+)2[B12Cl12]2–
–13.0
–12.7
–11.7
–13.0
–12.1
–11.9
–10.1
–9.0
–8.1
26.3
26.4
25.3
25.9
18.9
17.7
1
2
3
4
5
6
7
8
3
4
5
613
713
813
913
1011f
[1a∙Cy2BH]+[BArF ]–
4
[1a∙9‐BBN]+[BArF ]–
4
[1a∙catBH]+[BArF ]–
28.4
45.1
49.9
4
[1a∙pinBH]+[BArF ]–
4
9
[1a∙EtMe2SiH]+[BArF ]–
4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aExperiments were performed in a J. Young NMR tube
using complexes [1]+[X]– (1.0 equiv) and DIBAL–H (1.2 equiv).
Structural Characterization of the Hydroalane
Adduct. Single crystals suitable for X‐ray diffraction were
obtained at –35 °C from a solution of [1a]+[B(C6F5)4]– or
[1c]+[B(C6F5)4]– and DIBAL–H in fluorobenzene layered
with n‐pentane (Figures
1 and 2). The molecular
structures of the corresponding hydroalane adducts of
[1a]+[B(C6F5)4]– and [1c]+[B(C6F5)4]– prove that the
aluminum atom is bound to the sulfur atom with Al–S
bond lengths of 2.31 and 2.32 Å, respectively (Table 2).
Compared to [1a∙9‐BBN]+[BArF ]‒ (B–S bond length = 1.94
4
Å13) and [1a∙EtMe2SiH]+[BArF ]‒ (Si–S bond length = 2.24
4
Å11f), the Al–S bond is notably longer. That can be
attributed to a larger covalent/ionic radius of aluminum
as compared to boron and silicon as well as to a still
existing bonding interaction between aluminum and
hydrogen (see bonding analysis below). The observed
bond length is in accordance with those of aluminum
sulfides (Al–S, bond lengths = 2.2720a–2.34 Å20b) while
complexes with coordinating thioethers display longer
distances (Al–S, bond length = 2.51 Å21a). The hydride is
located at the ruthenium center with Ru–H bond lengths
RESULTS AND DISCUSSION
NMR Spectroscopic Analysis of the Adduct
Formation.
Initial
experiments
showed
4
rapid
degradation of the routinely used [BArF ]– counteranion
in presence of alumenium ions. We therefore turned
toward more robust counteranions such as [B(C6F5)4]– and
[B12Cl12]2– devoid of C(sp3)–F bonds.15 Mixing complexes
[1]+[X]– and DIBAL–H in o‐C6D4Cl2 at room temperature
instantly
delivered
the
hydroalane
adducts
[1∙iBu2AlH]+[X]– (Table 1). The adduct formation is
accompanied by a color change from green to yellow. The
observed hydride shifts are comparable to the earlier
example of the B–H bond activation in Cy2BH ((1H) –12.1
of 1.65 and 1.53
Å
in [1a∙iBu2AlH]+[B(C6F5)4]– and
[1c∙iBu2AlH]+[B(C6F5)4]–, respectively. This bond is
comparably shorter than in the aforementioned
ppm with [1a]+[BArF ]– and (1H) –11.4 ppm with
hydroborane
and
hydrosilane
adducts
([1a∙9‐
4
BBN]+[BArF ]‒: Ru–H bond length
=
1.80 Å13 and
[1d]+[BArF ]–).13 The JH,P coupling constants are in the
2
4
4
[1a∙EtMe2SiH]+[BArF ]‒: Ru–H bond length = 1.61 Å11f). The
4
range of 26 Hz and as such larger compared to those of
hydroborane adducts (2JH,P ≈ 18 Hz) but smaller than in
the hydrosilane adducts (2JH,P ≈ 50 Hz).16 Observation of
the 27Al nucleus (S = 5/2) by NMR spectroscopy failed
because of the high linewidth.17 However, our relativistic
DFT calculations for [1a∙iBu2AlH]+ predict the 27Al
resonance signal to appear at (27Al) 178 ppm, thus only
slightly downfield from that in parent DIBAL–H with
(27Al) 158 ppm in toluene‐d8.18,19
Al–H bond in [1a∙iBu2AlH]+[B(C6F5)4]– (bond length = 1.98
Å) is markedly elongated compared to known
hydroalanes [(2,6‐Mes2C6H3)2Al–H with coordination
number 3, Al–H bond length = 1.4322a and (tBuCH2)2Al–H
with coordination number 4, Al–H bond length = 1.62–
1.76 Å22b].
Table 2. Selected experimental bond lengths of E–H
Adducts [E = Si, B, and Al]
Table 1. 1H NMR Shifts and Coupling Constants of the
Ruthenium(II) Hydrides in the iBu2AlH Adductsa
entry
adduct
Ru–H
[Å]
E–S
[Å]
C–E–C
[°]
1
[1a∙iBu2AlH]+[B(C6F5)4]–
[1c iBu2AlH]+[B(C6F5)4]–
1.65
1.53
2.31
2.32
1.94
2.24
119
115
109
110
2
313
411f
[1a∙9‐BBN]+[BArF ]–
1.80
4
[1a∙EtMe2SiH]+[BArF ]–
1.58
4
Interestingly,
the
molecular
and
structures
of
[1a∙iBu2AlH]+[B(C6F5)4]–
[1c∙iBu2AlH]+[B(C6F5)4]–
entry
1
adduct
[1a∙iBu2AlH]+[B(C6F5)4]–
1H NMR [ppm]
–12.4
2JH,P [Hz]
reveal a close contact between the aluminum atom and
the ruthenium center with Ru–Al interatomic distances of
26.1
ACS Paragon Plus Environment

Page 3 of 10
Journal of the American Chemical Society
2.81 and 2.83 Å, respectively. The experimentally observed Quantum‐Chemical Analysis of the Hydroalane
1
2
3
4
5
6
7
8
bond angle for the C–Al–C fragments are 119° and 115°;
other known sulfur‐coordinated aluminum compounds
possess slightly wider angles (C–Al–C bond angles = 119–
125°).20,21
Adducts and Comparison with Their Group 13 and
Group 14 Analogs. To ascertain that the close contacts of
the aluminum atom with sulfur, hydrogen, and the
ruthenium center in [1∙iBu2AlH]+[B(C6F5)4]– complexes are
not an artifact of crystal packing forces and to resolve
uncertainties with hydride atom positions determined by
X‐ray diffraction, we carried out DFT calculations for a
series of complexes [1]+, their hydroborane adducts, their
hydrosilane adducts, and their heavier group 13 and group
14 analogs (Tables S3 and S4 in the Supporting
Information). The molecular structures (without
counteranions) were fully optimized at the B3LYP‐
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
D3(BJ)/ECP/6‐31+G** level, using
a quasi‐relativistic
small‐core pseudopotential for ruthenium along with
atom‐pairwise corrections for dispersion forces. In
addition, bulk solvent effects were simulated using an
SMD solvation model (Computational Details in the
Supporting Information). We note an excellent match
between experimental and DFT optimized structures,
with differences in bond lengths of less than 0.03 Å
(Figures 1 and 2). Only the precise detection of the Ru–H
bond lengths and the Al–H bond lengths suffered from
weak scattering of X‐rays by light hydrogen atoms and
thus the inaccurate hydride positions.
Figure 1. Molecular structure of [1a∙iBu2AlH]+[B(C6F5)4]–.
Hydrogen atoms (except for Ru–H) and counteranion are
omitted for the sake of clarity. Selected experimental bond
lengths (Å): Ru–S, 2.41; Ru–Al, 2.81; Ru–H, 1.65; Al–S, 2.31; Al–
H, 1.98. Selected experimental bond angle: C–Al–C, 119°
(average of 110° and 127°). For comparison, the DFT
optimized distances (Å) and angles are as follows: Ru–S, 2.42;
Ru–Al, 2.78; Ru–H, 1.65; Al–S, 2.33; Al–H, 1.85; C–Al–C, 119.5°.
The reliability of DFT optimized structures, including
hydride positions, is further demonstrated by the
1
excellent agreement between computed H NMR hydride
2
shifts (Ru–H) and coupling constants JH,P with
experimental data (Table S5 and Figures S3 and S4 in the
Supporting Information).
To quantify the degree of E–H bond dissociation and
bond covalencies for pertinent atom pairs, we evaluated
Mayer bond orders (MBOs) for a wide series of [1]+
adducts with various group 13 and group 14 hydrides RnEH
(Tables S3 and S4 in the Supporting Information). Hence,
MBOs of Ru–H bonds and E–H bonds show a superb
correlation: the stronger and thereby shorter the Ru–H
bond is, the weaker the E…H bonding interaction (Figure
3). The electron‐density depletion in the E‒H bond is thus
connected with electron‐density accumulation in the Ru‒
H bond, which can be monitored spectroscopically by
measuring 2JH,P coupling constants.23
MBO(E–H) = –1.32 MBO(Ru–H) + 1.12
R2 = 0.98
Figure 2. Molecular structure of [1c∙iBu2AlH]+[B(C6F5)4]–.
Hydrogen atoms (except for Ru–H) and counteranion are
omitted for the sake of clarity. Selected experimental bond
lengths (Å): Ru–S, 2.40; Ru–Al, 2.83; Ru–H, 1.53; Al–S, 2.32,
Al–H, 2.16. Selected experimental bond angle: C–Al–C, 115°.
For comparison, the DFT optimized distances (Å) and angles
are as follows: Ru–S, 2.42; Ru–Al, 2.80; Ru–H, 1.65; Al–S, 2.33;
Al–H, 1.85; C–Al–C, 115.8°.
Figure 3. Correlation between Mayer bond orders (MBO) of
Ru–H bonds and E–H bonds within the same adduct of [1]+
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 10
with various group 13 and group 14 element hydrides (Tables
S3 and S4 in the Supporting Information for numerical data).
1
2
3
4
5
6
7
8
In this regard, the adducts with completely broken E–H
bonds and well‐formed Ru–H bonds display 2JH,P coupling
constants larger than 42 Hz (this is particularly the case
for group 14 hydrides)11f while systems with weakly
activated E
–H bonds and partially formed Ru–H bonds
possess much smaller coupling constants. On this basis,
the E–
H bond in dialkylborane adducts of [1]+ with 9‐BBN
9
or Cy2BH (2JH,P ≈ 18 Hz) is less activated, which means less
heterolytically cleaved, than that of [1∙iBu2AlH]+ (2JH,P ≈ 26
Hz), that is also in accordance with the MBO values
computed for the E…H atom pairs (Table S3 in the
Supporting Information).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The B…H interaction in these adducts preserves about
40–50% bonding character as evident from MBO(E–H)
values for free hydroboranes and their adducts with [1]+,
while the Al…H interaction preserves (only) 25–33%
bonding character, even though the E–H bond elongation
upon coordination to [1]+ is of the same magnitude (~0.25
Å) for both hydroboranes and hydroalanes.
Figure 4. Natural localized molecular orbital (NLMO)
analysis corresponding to 3c2e bond between ruthenium,
hydrogen, and aluminum atoms (isosurface plot ±0.06 a.u.;
hydrogen atoms except Ru–H are omitted for clarity).
Note that apart from [1∙pinBH]+, the donor‐acceptor
interaction
between
low‐valent,
electron‐rich
ruthenium(II) d6 center and electrophilic, coordinatively
unsaturated main‐group atom, is common feature for all
adducts with group 13 hydrides, as also experimentally
found in this work for both [1∙iBu2AlH]+ complexes
(Figures
1
and 2) and previously for [1a∙(9‐BBN)]+
(however, without noting a short Ru…B bond interaction =
2.52 Å).13
We note further that the Ru…Al distances in
[1∙iBu2AlH]+ adducts are notably above the sum of single‐
bond covalent radii (2.51 Å)24 or those found in
[Ru(AlCp*)5] (with the average Ru–Al bond‐length of 2.38
Å)25, but are comparable with that computed for
hypothetical [RcAlMe2] complex (Rc = ruthenocene;
d(Ru…Al) = 2.81 Å) with a large dip angle (38°) due to the
Ru…Al bonding (Figure S5 in the Supporting Information
and Ref. 26 for analogs, synthetically available [FcSiR2]+
systems with Fe…Si bonding).
The unusual bonding situation in [1a∙iBu2AlH]+ is
already apparent from the Mayer bond orders between
aluminum and its neighbors (Table S3 in the Supporting
Information). While a considerably polarized single bond
is clearly present between aluminum and sulfur,27 both,
the Ru–H bond as well as the Al–H bond possess partial
covalent character, indicative of delocalized bonding.
This is examined in more detail by the natural localized
molecular orbital (NLMO) analysis and the electron
localization function (ELF), both revealing a three‐center‐
two‐electron (3c2e) bond between ruthenium, hydrogen
and aluminum (Figures 4 and 5), which can be best
Figure 5. Cut‐plane plots from electron localization function
(ELF) analysis of bonding attractors in [1a
∙
Me2AlH]+,
characterized as
a
σ(Ru–H)→Al donor‐acceptor
showing Ru…Al, Al
–
S, and Ru S bonding interactions. Gray‐
–
white regions represent ELF maxima (bonding attractors).
interaction with an appreciable electron‐density
accumulation/depletion in the Ru…Al/Ru–H interatomic
region.
According to the NLMO analysis of [1a∙iBu2AlH]+, the
population of the sp3‐hybridized aluminum atomic
orbitals in this 3c2e bond is at 10% while the population of
ruthenium is at 30% with predominant d‐character. The
ACS Paragon Plus Environment

Page 5 of 10
Journal of the American Chemical Society
electron density depletion in the Ru–H bond is accompa‐ different aggregation states [(iBu2AlH)n with n ≥ 2, Table
1
2
3
4
5
6
7
8
accompanied by its elongation and more upfield hydride
shifts in the H NMR spectra. In addition, the σ(Ru–
3; see also Table S8 in the Supporting Information for
more data).29
1
H)→Al interaction is responsible for (only) partial
heterolytic splitting (activation) of the Al–H bond and is
predicted to appear also in heavier group 13 analogs (the
highest covalency is predicted for the gallium analog and
markedly weaker interaction in indium and thallium
congeners).28
Table 3. Computed FIAs (in kJmol–1) for the Reaction
of Lewis Acids with F3CO– as a Fluoride Donor (LAq +
F3CO– → LA–Fq–1 + F2CO)30 and CM5 Partial Atomic
Charges at the Ruthenium and Aluminum Centera
G0FIA [kJmol–1]
Lewis acid (LA)
[1a
iBu2AlH]+
q(Ru)
0.202
q(Al)
0.309
We note that the Ru–S bond in [1]+ becomes longer
9
∙
–
–
–
151
114
99
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
upon the hydride adduct formation (by about 0.13–0.19 Å)
but still preserves an appreciable covalent bonding
character (Figure 6).
[(Et3P)RuSDmp]+ (1a+)
0.281
—
—
iBu2AlH
0.352
(iBu2AlH)2
(iBu2AlH)3
–97
–82
—
—
0.223
0.222
aB3LYP‐D3(BJ)/ECP/6‐31+G** results using an SMD
solvation model (Computational Details in the Supporting
Information).
According to q(Al) charges,
a
slightly higher
electrophilicity is expected for monomeric DIBAL–H,
which is absent in solution and exists in non‐polar
organic solvents only in the form of di‐/trimeric
structures.31 As expected, oligomeric DIBAL–H structures
are much weaker electrophiles than both monomeric
iBu2AlH and its adduct [1a∙iBu2AlH] (the molecular
electrostatic potential (MEP) is displayed for the
calculated hydroalane adduct [1a∙Me2AlH]+ in Figure 6).
The activation of DIBAL–H can be additionally viewed as
a dissociation of the oligomeric DIBAL–H structures
initiated by the Lewis acidic ruthenium center in [1]+ and
accompanied by partial heterolytic cleavage of the Al–H
bond.
Figure 6. Molecular electrostatic potentials (MEP) of
[1a∙Me2AlH]+. Isovalue surfaces are displayed at an electron
density of 0.04 a.u. (B3LYP/ECP/6‐31+G** results). Blue
region corresponds to the extreme positive potentials,
indicating the electrophilic center for a fluoride abstraction.
Application to C(sp3)–F Activation. As shown earlier
by us, [1d]+[BArF ]– catalyzes the hydrodefluorination of
4
CF3‐substituted anilines with hydrosilanes.12 These
reactions require the substoichiometric addition of an
alkoxide base, and role of that additive is understood on
the basis of an unusual dual role of the catalyst (not
shown). The catalyst loading is also high (10 mol %), and
the catalyst system is limited to highly activated CF3
groups (reaction temperature 20 °C for para Me2N group
but 100 °C for meta).32 Trying to enhance the performance
of this setup, we decided to replace the hydrosilane by the
more reactive DIBAL–H. Orientation experiments with
the far less activated 1,1’‐biphenyl‐4‐yl‐substituted CF3
group (as in 3a) quickly revealed that the alkoxide‐free
catalysis with the hydroalane works at room
temperature33 while completely failing with Me2PhSiH
(not shown). The dominant reaction pathway was
however the hydrodefluorination coupled with Friedel–
Crafts benzylation of the arene solvent, e.g., toluene (4a).
Joining benzylic C–F bond activation with Friedel–Crafts
alkylations using BF3 as the Lewis acid catalyst was
Apart from hydrocarbons, the overall E–H bond
activation process for MenEH hydrides is exergonic with
formation energies (G0 ) ranging from –38 kJmol–1 to –
f
146 kJmol–1 for corresponding hydrosilanes and
hydroalanes, respectively (Table S4 in the Supporting
Information). Our previous computational study for
hydrosilane adducts showed that the heterolytic cleavage
of the Si–H bond proceeds via a concerted four‐
membered transition state with relatively small reaction
energy barriers of ca. 10–40 kJmol–1 depending on the
hydrosilane.11f DFT simulations of the cooperative bond
activation of DIBAL–H by [1a]+ show that the Al–H bond
cleavage is practically barrierless on the electronic energy
surface.
Interestingly, while group 14 hydrides show a steady
decrease in G0 as the main‐group atom becomes
f
heavier, non‐monotonous behavior is observed for the
group 13 congeners, presumably due to the
abovementioned σ(Ru–H)→E interaction.
Comparing computed CM5 partial atomic charges at
the aluminum center reveals that in coherence with
computed fluoride‐ion affinities (G0FIA), complex
[1a∙iBu2AlH]+ is a stronger electrophile and a stronger
Lewis acid than both free [1a]+ and DIBAL–H in its
reported by Olah and co‐workers three decades ago.34,35
A
and
few
additional
examples
of
inter‐36a,37
intramolecular36b Friedel–Crafts alkylations subsequent to
C(sp3)–F36a,37 or even C(sp2)–F36b bond activation involving
silicon cations were also described. Stephan and co‐
workers recently demonstrated that the electrophilic
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 10
phosphonium salt [(C6F5)3PF]+[B(C6F5)4]– promotes broad‐
broadly applicable Friedel–Crafts alkylations and
benzylations initiated by fluoride abstraction.38 Both
benzylic and aliphatic CF3 groups (with further
hydrodefluorination)38a as well as benzylic fluorides38b
participate well in this transformation with hydrosilanes
as the hydride source.
10
11
12d
[1b]+[B(C6F5)4]–
[1b]+[B(C6F5)4]–
[1a]+[B(C6F5)4]–
iPr3P
iPr3P
Et3P
PhCl
78:22
—
>99
1
2
3
4
5
6
7
8
n‐C5H12
PhF
<6
77:23
<33
aAll reactions were performed according to General
Procedure 3. bProduct ratios and conversions were
determined by GLC analysis using tetracosane as internal
standard. Catalyst was formed in situ from corresponding
neutral chloride complex. Polymeric AlH3 was used instead
of DIBAL–H.
c
We then embarked on the systematic investigation of
the hydrodefluorinative Friedel–Crafts alkylation of 4‐
trifluoromethyl‐1,1’‐biphenyl with toluene as the arene
nucleophile (3a+4a→5aa, Table 4). The reactions were
performed in fluorobenzene at room temperature.
DIBAL–H alone did not affect the fluoride abstraction
(entry 1). Full conversion was obtained with catalysts
[1]+[B(C6F5)4]– independent of the phosphine ligand
(entries 2–5). This distinguishes the present system from
d
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
With an optimal setup established (Table 4, entry 3),
we gauged the functional‐group tolerance in the CF3‐
containing precursor (Table 5). The model reaction
afforded the coupled compound in 60% yield (3a→5aa,
entry 1). No para‐substitution (as in 3b) as well as a
methyl and a bromo substituent (as in 3c and 3d) were
well tolerated (3b–d→5ab–ad, entries 2–4). Conversely,
the
hydrodefluorination
with
hydrosilanes.12
([1b]+)2[B12Cl12]2– as catalyst led to inconsistent results,
likely due to solubility issues. Conversion was poor with
preformed ([1b]+)2[B12Cl12]2– (entry 6) whereas full
diaryl
ether
3e
reacted
under
exclusive
hydrodefluorination (entry 5). Precursors with ortho‐
substitution also participated but required higher
reaction temperature (3f→5af, entry 6) or longer reaction
times (3g→5ag, entry 7). Regioselectivities were generally
moderate (rs ≈ 75:25) and as such usually better than
conversion
was
achieved
with
in‐situ‐formed
([1b]+)2[B12Cl12]2– (from the corresponding neutral chloride
complex and Na2B12Cl12, entry 7). The [BArF ]–
4
those seen with Stephan’s protocol (rs
<
67:33).38
counteranion with its CF3 groups slowly degraded but
45% conversion were nevertheless reached with
Intramolecular Friedel–Crafts reactions essentially failed,
e.g. 3h→5h (Scheme 2); traces of 5h were detected by GC‐
MS analysis along with hydrodefluorination (7) and
anthracene (8; dehydrogenated 5h). Stephan had
achieved excellent 68% yield of 5h for the same
transformation.
Table 5. Scope I: Variation of the CF3‐Containing
Substrate in the Hydrodefluorinative Friedel–Crafts
Alkylationa
[1b]+[BArF ]– (entry 8); higher conversions were not
4
observed at prolonged reaction times. Arene solvents
other than fluorobenzene worked as well with slightly
diminished regioselectivities (entries 9 and 10). Hardly
any conversion was seen in n‐pentane (entry 11). Using
polymeric AlH3 prepared from LiAlH4 and AlCl339 as the
hydride source gave only low conversion (entry 12), and
increased amounts of AlH3 as well as longer reaction
times still did not let the reaction go to completion (not
shown).
Table
4.
Catalyst
Screening
for
the
Hydrodefluorinative Friedel–Crafts Alkylationa
entry CF3‐containing substrate
product
5aa
p:o (%)b
79:21
75:25
79:21
70:30
—
yield (%)c
1
p‐Ph(C6H4)CF3 (3a)
C6H5CF3 (3b)
60
75
73
82
2
5ab
5ac
3
p‐Me(C6H4)CF3 (3c)
p‐Br(C6H4)CF3 (3d)
p‐PhO(C6H4)CF3 (3e)
o‐Ph(C6H4)CF3 (3f)
o‐I(C6H4)CF3 (3g)
4
5ad
5ae
d
5
—
6e
7f
5af
79:21
77:23
54
79
5ag
entry
catalyst
phosphine
solvent p:o
(%)b
conv.
(%)b
aAll reactions were performed according to General
Procedure 3. bProduct ratios and conversions were
determined by GLC analysis using tetracosane as internal
standard. The reactions were run until complete conversion
1
—
—
PhF
PhF
PhF
PhF
PhF
PhF
PhF
PhF
PhH
—
—
2
3
[1a]+[B(C6F5)4]–
[1b]+[B(C6F5)4]–
[1c]+[B(C6F5)4]–
[1d]+[B(C6F5)4]–
([1b]+)2[B12Cl12]2–
([1b]+)2[B12Cl12]2–
Et3P
78:22
79:21
79:21
78:22
79:21
79:21
70:30
76:24
>99
>99
>99
>99
<14
c
iPr3P
Cy3P
of the trifluoromethyl compound. Percentages are displaying
isolated yields after column chromatography. dOnly
4
5
6
7c
8
9
e
p‐(FC6H4)3P
iPr3P
iPr3P
iPr3P
iPr3P
hydrodefluorination and alkylation were observed. Reaction
was performed at 80 °C. fReaction was quenched after 5 d.
Scheme 2. Intramolecular Friedel–Crafts Alkylationa,b
>99
<45
>99
[1b]+[BArF ]–
[1b]+[B(C6F5)4]–
4
ACS Paragon Plus Environment

Page 7 of 10
Journal of the American Chemical Society
aAll reactions were performed according to General
Procedure 3. bProduct ratios and conversions were
determined by GLC analysis using tetracosane as internal
1
2
3
c
standard. The reactions were run until complete conversion
of the trifluoromethyl compound. Percentages are displaying
4
5
6
7
d
isolated yields after column chromatography. Reaction was
e
f
performed at 80 °C. Reaction was quenched after 5 d. Only
hydrodefluorination and alkylation were observed. Np =
naphthyl, Th = thienyl, Fc = ferrocenyl.
8
9
10
11
For dialkyl‐substituted arenes such as xylenes and
cymenes moderate to good yields were generally observed
(Scheme 3). m‐Xylene (4m) reacted to the single
regioisomer 5ma whereas o‐cymene (4o) gave a mixture
of three regioisomers (5oa). Trisubstituted mesitylene
(4q) afforded 5qa in near‐quantitative yield. Testing the
limits of the system, 1,4‐di‐tert‐butylbenzene did not give
the desired product, but instead only hydrodefluorination
was observed (not shown).―Bis(4‐tolyl) ether (4r) led to
the expected diarylmethane 5ra in a 95:5 mixture with
xanthene 6 (3a→5ra, Scheme 4).
aThe reaction was performed according to General
Procedure 3. bo‐C6H4F2 was used instead of PhF to avoid side
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
c
reactions with the solvent. Product ratios and conversions
were determined by GLC analysis using tetracosane as
internal standard.
We then probed the scope of the arene nucleophile
(Table 6). The screening of benzene (4b) and several
donor‐substituted (4c–e) as well as monohalogenated
arenes (4f–h) revealed that only benzene and electron‐
rich arenes do react (3a→5ba–ea, entries 1–4). For
unreactive nucleophiles, we have always obtained
hydrodefluorination and alkyl transfer. However, neither
aryl alkyl ethers (alkyl = methyl and isopropyl) nor
dimethylamino groups were stable under the reaction
conditions (dealkylation, not shown). For this reason,
diaryl (thio)ethers 4c and 4d and diisopropylamino‐
substituted 4e were employed. These furnished the
corresponding diarylmethanes 5ca–ea in moderate to
good yields, requiring elevated temperature (80 °C) in the
case of 4d and 4e. The higher reaction temperature was
Scheme 3. Scope IIb: Variation of the Arene
Nucleophile in the Hydrodefluorinative Friedel–
Crafts Alkylationa–c
[1b]+[B(C6F5)4]– (1.0 mol %)
4l 4q
, 5.0 equiv)
DIBAL–H (4.0 equiv)
Nu–H (
–
CF3
Nu
PhF, r.t., 24 h
– H2
Ph
Ph
– DIBAL–F
3a
5la–5qa
detrimental to regiocontrol (3a→5da, entry
3 and
3a→5ia, entry 8) but steric hindrance was able to
compensate this trend (3a→5ea, entry 4). Thiophene (4j)
and ferrocene (4k) led to the corresponding products 5ja
and 5ka, respectively, only in poor yields.
Table 6. Scope IIa: Variation of the Arene
Nucleophile in the Hydrodefluorinative Friedel–
Crafts Alkylationa
Ph
Ph
Ph
Ph
5la
5ma
89:11, 60%
>99:1, 71%
iPr
5na
5oa
55:36:9, 46%
46%
Ph
Ph
entry Arene Nu–H
product rs (%)b
yield (%)c
iPr
1
PhH (4b)
Ph2O (4c)
Ph2S (4d)
Ph2S (4d)
PhNiPr2 (4e)
PhF (4f)
5ba
5ca
5da
5da
5ea
5fa
—
56
85
72
43
49
5pa
5qa
2
p:o 98:2
p:o 85:15
p:o >99:1
p:o >99:1
—
85:15, 43%
96%
3d
4e
4d
5
aAll reactions were performed according to General
Procedure 3. bProduct ratios and conversions were
determined by GLC analysis using tetracosane as internal
c
standard. The reactions were run until complete conversion
f
—
of the trifluoromethyl compound. Percentages are displaying
isolated yields after column chromatography.
f
6
PhCl (4g)
PhBr (4h)
NpH (4i)
NpH (4i)
ThH (4j)
FcH (4k)
5ga
5ha
5ia
—
—
f
7
—
—
Scheme 4. Scope IIc: Variation of the Arene
Nucleophile in the Hydrodefluorinative Friedel–
Crafts Alkylationa–c
8d
9e
1od
11d
α:β 75:25
α:β 81:19
α:β 83:17
—
52
46
35
23
5ia
5ja
5ka
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 10
Experimental details, characterization and crystallographic
data as well as 1H, 13C, 11B, 19F, and 31P NMR spectra.
Computational details, the results of DFT calculations and
more detailed analysis. This material is available free of
charge via the Internet at http://pubs.acs.org.
1
2
3
4
5
6
7
8
AUTHOR INFORMATION
Corresponding Authors
peter.hrobarik@tu‐berlin.de;
martin.oestreich@tu‐berlin.de
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Present Address
‡Department of Chemistry, University of Jyväskylä, FI‐40014
Jyväskylän yliopisto, Finland.
ORCID
Francis Forster: 0000‐0002‐0364‐5491
Toni T. Metsänen: 0000‐0002‐3966‐6067
Elisabeth Irran: 0000‐0001‐6098‐1996
Peter Hrobárik: 0000‐0002‐6444‐8555
Martin Oestreich: 0000‐0002‐1487‐9218
aThe reaction was performed according to General
Procedure 3. bProduct ratios and conversions were
determined by GLC analysis using tetracosane as internal
standard. cThe reaction was run until complete conversion of
the trifluoromethyl compound. The percentage is displaying
isolated yield after column chromatography.
Notes
The authors declare no competing financial interests.
CONCLUSION
ACKNOWLEDGMENT
The present work discloses
a new example of
This research was supported by the Deutsche
Forschungsgemeinschaft (Oe 249/8‐1). T.T.M. and P.H. were
funded by the Cluster of Excellence Unifying Concepts in
Catalysis of the Deutsche Forschungsgemeinschaft (EXC
314/2). M.O. is indebted to the Einstein Foundation (Berlin)
for an endowed professorship.
cooperative catalysis at the Ru–S bond in Ohki–Tatsumi
complexes.14 Al–H bond activation is now added to the
activation of hydrosilanes, hydroboranes, and dihydrogen.
The Ru–S functional group assists in deaggregation of
oligomeric DIBAL–H structures and partial heterolytic
splitting of the Al–H bond accompanied by the formation
of Ru–H and Al–S bonds.
Our initial hypothesis that the Al–H bond is completely
cleaved heterolytically did not prove to be true and,
hence, the assumed sulfur‐stabilized alumenium ion (next
to the ruthenium(II) hydride) is not an accurate
description of these hydroalane adducts. Structural
characterization by X‐ray diffraction revealed a Ru–Al
interaction in addition to the Ru–H and Al–S bonds. Our
quantum‐chemical calculations explain this bonding
motif as an σ(Ru–H)→Al donor‐acceptor interaction.
DIBAL–H is still cooperatively activated in these Lewis
adducts but not to the extreme of the hydrosilane
adduct.11f
REFERENCES
(1) (a) Atwood, D. A. Coord. Chem. Rev. 1998, 176, 407–430. (b)
Dagorne, S.; Atwood, D. A. Chem. Rev. 2008, 108, 4037–4071.
(2) Kim, K.‐C.; Reed, C. A.; Long, G. S.; Ayusman, S. J. Am.
Chem. Soc. 2002, 124, 7662–7663.
(3) Young, J. D.; Khan, M. A.; Wehmschulte, R. J.
Organometallics 2004, 23, 1965–1967.
(4) (a) Klis, T.; Powell, D. R.; Wojtas, L.; Wehmschulte, R. J.
Organometallics 2011, 30, 2563–2570. (b) Kessler, M.; Knapp, C.;
Zogaj, A. Organometallics 2011, 30, 3786–3792. (c) Ivanov, S. V.;
Peryshkov, D. V.; Miller, S. M.; Anderson, O. P.; Rappé, A. K.;
Strauss, S. H. J. Fluorine Chem. 2012, 143, 99–102.
(5) Engesser, T. A.; Lichtenthaler, M. R.; Schleep, M.; Krossing,
I. Chem. Soc. Rev. 2016, 45, 789–899.
(6) Stahl, T.; Klare, H. F. T.; Oestreich, M. ACS Catal. 2013, 3,
1578–1587.
(7) Klahn, M.; Fischer, C.; Spannenberg, A.; Rosenthal, U.;
Krossing, I. Tetrahedron Lett. 2007, 48, 8900–8903.
(8) Terao, J.; Begum, S. A.; Shinohara, Y.; Tomita, M.; Naitoh,
Y.; Kambe, N. Chem. Commun. 2007, 855–857.
(9) Gu, W.; Haneline, M. R.; Douvris, C.; Ozerov, O. V. J. Am.
Chem. Soc. 2009, 131, 11203–11212.
(10) Ohki, Y.; Takikawa, Y.; Sadohara, H.; Kesenheimer, C.;
Engendahl, B.; Kapatina, E.; Tatsumi, K. Chem. Asian J. 2008, 3,
1625–1635.
Despite this only partial Al–H bond separation, the
hydroalane adducts are found to be stronger electrophiles
than the free ruthenium catalyst, DIBAL–H in its different
aggregation states and hydrosilane adducts12 which is
consistent with our experimental findings. The Ru–S‐
activated DIBAL–H molecule was shown to engage in
C(sp3)–F bond cleavage, and the Ohki–Tatsumi complexes
turned
hydrodefluorinative Friedel–Crafts alkylations of CF3‐
substituted arenes to afford unsymmetrical
out
to
be
excellent
catalysts
for
diarylmethanes with various arene nucleophiles.
(11) (a) Klare, H. F. T.; Oestreich, M.; Ito, J.‐i.; Nishiyama, H.;
Ohki, Y.; Tatsumi, K. J. Am. Chem. Soc. 2011, 133, 3312–3315. (b)
Königs, C. D. F.; Klare, H. F. T.; Ohki, Y.; Tatsumi, K.; Oestreich,
M. Org. Lett. 2012, 14, 2842–2845. (c) Königs, C. D. F.; Müller, M.
F.; Aiguabella, N.; Klare, H. F. T.; Oestreich, M. Chem. Commun.
ASSOCIATED CONTENT
Supporting Information
ACS Paragon Plus Environment

Page 9 of 10
Journal of the American Chemical Society
2013, 49, 1506–1508. (d) Königs, C. D. F.; Klare, H. F. T.;
(27) Note that the E–S bonding interactions in hydroborane
1
2
3
4
5
6
7
8
Oestreich, M. Angew. Chem., Int. Ed. 2013, 52, 10076–10079. (e)
Hermeke, J.; Klare, H. F. T.; Oestreich, M. Chem. – Eur. J. 2014,
20, 9250–9254. (f) Stahl, T.; Hrobárik, P.; Königs, C. D. F.; Ohki,
Y.; Tatsumi, K.; Kemper, S.; Kaupp, M.; Klare, H. F. T.; Oestreich,
M. Chem. Sci. 2015, 6, 4324–4334. (g) Omann, L.; Oestreich, M.
Angew. Chem., Int. Ed. 2015, 54, 10276–10279. (h) Metsänen, T. T.;
Oestreich, M. Organometallics 2015, 34, 543–546. (i) Bähr, S.;
Simonneau, A.; Irran, E.; Oestreich, M. Organometallics 2016, 35,
925–928.
(12) Stahl, T.; Klare, H. F. T.; Oestreich, M. J. Am. Chem. Soc.
2013, 135, 1248–1251.
(13) Stahl, T.; Müther, K.; Ohki, Y.; Tatsumi, K.; Oestreich, M.
J. Am. Chem. Soc. 2013, 135, 10978–10981.
and hydrosilane adducts are more covalent and thereby in line
with the electronegativities of corresponding main‐group atoms.
(28) Note that the partial atomic charges may be viewed as an
approximate measure of the electrophilicity only when steric
effects are absent or constant across the investigated series.
(29) However, saturating the coordination sphere of
aluminum in parent hydride such as in [(tBuO)3AlH]– hinders
the σ(Ru–H)→Al donation and is computed to result in an
adduct with the complete Al–H bond separation (Table S3 in the
Supporting Information).
9
(30) Christe, K. O.; Dixon, D. A.; McLemore, D.; Wilson, W.
W.; Sheehy, J. A.; Boatz, J. A. J. Fluorine Chem. 2000, 101, 151–153.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
(31) As also confirmed by H NMR measurements and NMR
(14) Omann, L.; Königs, C. D. F.; Klare, H. F. T.; Oestreich, M.
Acc. Chem. Res. 2017, 50, 1258–1269.
shift calculations, the hydride shift for monomeric DIBAL–H is
predicted to appear at (1H) 4.9 ppm while only two hydride
signals at (1H) 3.3 and 3.1 ppm are observed in C6D6; similarly,
the computed 27Al NMR resonance for monomeric DIBAL–H is
about 130 ppm downfield as compared to that found
experimentally for DIBAL–H in toluene‐d8 (Table S9 in the
Supporting Information).
(15) Complex ([1b]+)2[B12Cl12]2– could not be isolated in pure
form (see the Supporting Information).
2
(16) Coupling constant of [1a∙Cy2BH]+[BArF ]– is JH,P = 19 Hz.
4
See Ref. 13 for further details. Coupling constant of
2
[1a∙EtMe2SiH]+[BArF ]– is JH,P = 50 Hz. See Ref. 11f for further
4
(32) Hydrodefluorination is also observed with [1a]+[BArF ]– as
details.
4
(17) Akitt, J. W. Prog. Nucl. Magn. Reson. Spectrosc. 1989, 21, 1–
catalyst in the absence of alkoxide, further limiting the process
to just the para‐CF3‐substituted aniline.
149.
(18) Performance of two‐component ZORA‐SO method in 27Al
NMR shift calculations for a broad range of Al(III) compounds is
demonstrated in Table S7 and Figure S7 in the Supporting
Information. The experimental 27Al NMR data are collected in:
Benn, R.; Janssen, E.; Lehmkuhl, H.; Rufínska, A. J. Organomet.
Chem. 1987, 333, 155–168.
(33) The formation of DIBAL–F was detected by 19F NMR
spectroscopy: (19F) –148.9, –148.1, –146.1 ppm in C6D6; cf. Ref. 9
in CDCl3.
(34) Olah, G. A.; Olah, J. A.; Ohyama, T. J. Am. Chem. Soc.
1984, 106, 5284–5290.
(35) (a) Tsuchimoto, T.; Tobita, K.; Hiyama, T.; Fukuzawa, S.‐i.
J. Org. Chem. 1997, 62, 6997–7005. (b) Mertins, K.; Iovel, I.;
Kischel, J.; Zapf, A.; Beller, M. Angew. Chem., Int. Ed. 2005, 44,
238–242. (c) Iovel, I.; Mertins, K.; Kischel, J.; Zapf, A.; Beller, M.
Angew. Chem. Int. Ed. 2005, 44, 3913–3917. (d) Rueping, M.;
Nachtsheim, B. J.; Ieawsuwan, W. Adv. Synth. Catal. 2006, 348,
1033–1037. (e) Schäfer, G.; Bode, J. W. Angew. Chem., Int. Ed.
2011, 50, 10913–10916. (f) Champagne, P. A.; Benhassine, Y.;
Desroches, J. Paquin, J.‐F. Angew. Chem., Int. Ed. 2014, 53, 13835–
13839. (g) Mo, X.; Yakiwchuk, J.; Dansereau, J.; McCubbin, J. A.;
Hall, D. G. J. Am. Chem. Soc. 2015, 137, 9694–9703. (h) Ricardo,
C. L.; Mo, X.; McCubbin, J. A.; Hall, D. G. Chem. – Eur. J. 2015, 21,
4218–4223. (i) Vuković, V. D.; Richmond, E.; Wolf, E.; Moran, J.
Angew. Chem., Int. Ed. 2017, 56, 3085–3089.
(36) (a) For an intermolecular version see: Douvris, C.;
Ozerov, O. V. Science 2008, 321, 1188–1190. (b) For an
intramolecular version see: Allemann, O.; Duttwyler, S.;
Romanato, P.; Baldridge, K. K.; Siegel, J. S. Science 2011, 332, 574–
577.
(37) Lühmann, N.; Panisch, R.; Müller, T. Appl. Organomet.
Chem. 2010, 24, 533–537.
(19) The surprisingly small deshielding in [1a∙iBu2AlH]+ can be
rationalized by an extra stabilization of the alumenium ion by
electron donation from multiple ligand atoms (see the structural
characterization and discussion below).
(20) (a) Postigo, L.; Maestre, M. d. C.; Mosquera, M. E. G.;
Cuenca, T.; Jiménez, G. Organometallics 2013, 32, 2618–2624. (b)
Kischel, M.; Dornberg, G.; Krautscheid, H. Inorg. Chem. 2014, 53,
1614–1623.
(21) (a) Lamberti, M.; D’Auria, I.; Mazzeo, M.; Milione, S.;
Bertolasi, V.; Pappalardo, D. Organometallics 2012, 31, 5551–5560.
(b) For further crystal structures containing Al–S bonds see: Uhl,
W.; Vester, A.; Hiller, W. J. Organomet. Chem. 1993, 443, 9–17.
(22) (a) Young, J. D.; Khan, M. A.; Powell, D. R.; Wehmschulte,
R. J. Eur. J. Inorg. Chem 2007, 1671–1681. (b) Uhl, W.; Appelt, C.;
Backs, J.; Klöcker, H.; Vinogradov, A.; Westenberg, H. Anorg.
Allg. Chem. 2014, 640, 106–109.
(23) A modest correlation (R2=0.91) with DFT optimized Ru‒H
bond lengths was observed (Figure S5 in the Supporting
Information). A similar correlation but of less quality (R2=0.84)
is also seen for 1H hydride shifts, which tend to be more
deshielded for adducts with completely broken E‒H bond
(Figure S6 in the Supporting Information).
(24) Pyykkö, P. J. Phys. Chem. A 2015, 119, 2326–2337.
(25) Steinke, T.; Cokoja, M.; Gemel, C.; Kempter, A.; Krapp, A.;
Frenking, G.; Zenneck, U.; Fischer, R. A. Angew. Chem., Int. Ed.
2005, 44, 2943–2946.
(38) (a) Zhu, J.; Pérez, M.; Caputo, B. C.; Stephan, D. W.
Angew. Chem., Int. Ed. 2016, 55, 1417–1421. (b) Zhu, J.; Pérez, M.;
Stephan, D. W. Angew. Chem., Int. Ed. 2016, 55, 8448–8451.
(39) Yartys, V. A.; Denys, R. V.; Maehlen, J. P.; Frommen, C.;
Fichtner, M.; Bulychev, B. M.; Emerich, H. Inorg. Chem. 2007,
46, 1051–1055.
(26) Müther, K.; Hrobárik, P.; Hrobáriková, V.; Kaupp, M.;
Oestreich, M. Chem. – Eur. J. 2013, 19, 16579–16594.
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 10
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
10
ACS Paragon Plus Environment