Elucidation of the Formation Mechanisms of the Octahydrotriborate Anion (B3H8-) through the Nucleophilicity of the B-H Bond

Article abstract of DOI:10.1021/jacs.8b03785

Boron compounds are well-known electrophiles. Much less known are their nucleophilic properties. By recognition of the nucleophilicity of the B-H bond, the formation mechanism of octahydrotriborate (B₃H₈^-) was elucidated on the bases of both experimental and computational investigations. Two possible routes from the reaction of BH₄^- and THF·BH₃ to B₃H₈^- were proposed, both involving the B₂H₆ and BH₄^- intermediates. The two pathways consist of a set of complicated intermediates, which can convert to each other reversibly at room temperature and can be represented by a reaction circle. Only under reflux can the B₂H₆ and BH₄^- intermediates be converted to B₂H₅^- and BH₃(H₂) via a high energy barrier, from which H₂ elimination occurs to yield the B₃H₈^- final product. The formation of B₂H₆ from THF·BH₃ by nucleophilic substitution of the B-H bond was captured and identified, and the reaction of B₂H₆ with BH₄^- to produce B₃H₈^- was confirmed experimentally. On the bases of the formation mechanisms of B₃H₈^-, we have developed a facile synthetic method for MB₃H₈ (M = Li and Na) in high yields by directly reacting the corresponding MBH₄ salts with THF·BH₃. In the new synthetic method for MB₃H₈, no electron carriers are needed, allowing convenient preparation of MB₃H₈ in large scales and paving the way for their wide applications.

Full text of DOI:10.1021/jacs.8b03785

Subscriber access provided by Kaohsiung Medical University
Article
Elucidation of the Formation Mechanisms of the Octahydrotriborate
Anion (B3H8-) Through the Nucleophilicity of the B-H Bond
Xi-Meng Chen, Nana Ma, Qian-Fan Zhang, Jin Wang, Xiaoge Feng,
Changgeng Wei, Lai-Sheng Wang, Jie Zhang, and Xuenian Chen
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b03785 • Publication Date (Web): 07 May 2018
Downloaded from http://pubs.acs.org on May 7, 2018
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 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 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.
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
Elucidation of the Formation Mechanisms of the Octahydrotriborate
Anion (B3H8 ) Through the Nucleophilicity of the B−H Bond
−
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
Xi-Meng Chen,†,⊥ Nana Ma,†,⊥ Qian-Fan Zhang,‡ Jin Wang,† Xiaoge Feng,† Changgeng Wei,†
Lai-Sheng Wang,*,‡ Jie Zhang,*, † and Xuenian Chen*,†
†School of Chemistry and Chemical Engineering, Henan Key Laboratory of Boron Chemistry and Advanced Energy Materi-
als, Henan Normal University, Xinxiang, Henan 453007, China
‡Department of Chemistry, Brown University, Providence, Rhode Island 02912, USA
Supporting Information
ABSTRACT: Boron compounds are well-known electrophiles. Much less known are their nucleophilic properties. By recognizing
–
the nucleophilicity of the B−H bond, the formation mechanism of octahydrotriborate (B3H8 ) was elucidated on the bases of both
–
–
experimental and computational investigations. Two possible routes from the reaction of BH4 and THF⋅BH3 to B3H8 were pro-
−
posed, both involving the B2H6 and BH4 intermediates. The two pathways consist of a set of compicated intermediates, which can
convert to each other reversibly at room temperature and can be represented by a reaction circle. Only under reflux can the B2H6
and BH4− intermediates be converted to B2H5− and BH3(H2) via a high energy barrier, from which H2 elimination occurs to yield the
−
B3H8 final product. The formation of B2H6 from THF⋅BH3 by nucleophilic substitution of the B−H bond was captured and identi-
–
−
fied, and the reaction of B2H6 with BH4 to produce B3H8 was confirmed experimentally. On the bases of the formation mecha-
–
nisms of B3H8 , we have developed a facile synthetic method for MB3H8 (M = Li and Na) in high yields by directly reacting the
corresponding MBH4 salts with THF·BH3. In the new synthetic method for MB3H8, no electron carriers are needed, allowing con-
venient preparation of MB3H8 in large scales and paving the way for their wide applications.
Potassium, rubidium, and cesium salts of B3H8− were also pre-
INTRODUCTION
pared by the reactions of THF·BH3 with the corresponding
alkali metal amalgams.2d Very recently, K/Na alloy was suc-
The octahydrotriborate anion, as a basic hydroborate unit
–
−
− 2i
following BH4 and B2H7 , plays important roles in boron
chemistry.1,2 The preparation of the alkali salts of octahydro-
triborate dated back to Stock’s work in the 1920s.2a By shak-
ing potassium or sodium amalgam in the presence of diborane
gas, a new compound with an empirical formula of K(B2H6) or
Na(B2H6) was obtained.2a-c In 1949, Newkirk found that the
compound with the empirical composition of Na(B2H6) had an
identical X-ray powder diffraction pattern to sodium borohy-
dride (NaBH4),3 the structure of which was first characterized
by X-ray powder diffraction in 1947.4 The products of the
reaction between sodium amalgam and diborane was finally
determined to be a mixture of NaB3H8 and NaBH4 in 1:1 ratio
based on the results of X-ray, infrared and elemental chemical
analyses in the late 1950s.5 Since then, the preparation meth-
ods for octahydrotriborate salts have been continually im-
proved,2a-j but its formation mechanisms have not been fully
cessfully used to prepare the potassium salt of B3H8 . Here
we report an investigation on the formation mechanisms of
−
B3H8 upon recognition of the importance of the B−H bond
nucleophilic properties, leading to a more convenient prepara-
tive method for LiB3H8 and NaB3H8 by simply using the more
convinient MBH4 compounds (Eq. 2):
MBH4 + 2 THF·BH3 = MB3H8 + H2 + THF
(2)
M = Li and Na
Boron and boron-containing compounds are characterized
by their electron deficiency, so that BH3 or BR3 groups usually
Scheme 1. Hydroboration reactions of borane complexes (A);
Nucleophilic substitution reactions of borane complexes (B).
understood.1a,2k-o Recently, the bonding in the B3H8 anion has
–
also been analyzed, highlighting its high stability.6
The incremental improvement of the synthetic methods for
octahydrotriborate2a-j has been mainly focused on using other
boranes to replace the flammable and toxic diborane2d-i or
dispersing sodium on other media (Eq. 1), instead of using
sodium amalgam, which causes health and safety concerns.2e,f
2 M/(Media) + 4 THF·BH3 = MB3H8 + MBH4 + 4 THF
(1)
M = alkili metals
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 10
act as Lewis acids in reactions with Lewis bases to form Lewis
Scheme 2. Reactions of NaNH2BH3 with AB (A) or THF·BH3
(B)
acid-base adducts.7 Hydroborations are well known for borane
complexes.8 In this class of reactions, it is generally under-
stood that the electron-deficient boron uses its empty 2p or-
bital to interact with the electron-rich C=C, C=O, C=N double
bonds, or C≡C triple bond of substrates, leading to an addition
reaction of the B−H bond (Scheme 1A).9 On the other hand,
the slightly lower electronegativity of boron (2.0 in the
Pauling scale) compared to that of hydrogen (2.1) makes the
hydrogen atoms on boron slightly negative in boron hydrides
1
2
3
4
5
6
7
8
9
+
−
(Bδ −Hδ ). Thus, the partially negatively-charged hydrogen
atoms in boron hydrides possess Lewis base properties. This
Lewis base property of the B−H bond enables RʺBH3 (Rʺ = H−,
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
–
first and then convert to (NH2BH2)n polymer and BH4 with
the bridge hydrogen moving to the BH3 moiety in the unsup-
ported B−H−B bond (Scheme 2B).
−
NH2 , THF, and NH3) to be a nucleophilic reagent in nucleo-
philic substitution reactions as shown in Scheme 1B. Even
though the Lewis acid property of boron complexes has been
well recognized,7f,10 relatively little is known about the Lewis
base property of the B−H bond.
In a few known examples of nucleophilic substitution reac-
tions involving the B−H bond as a nucleophilic reagent (red in
Stability of the unsupported B−H−B bond. The newly
formed unsupported B−H−B bond by the B−H bond nucleo-
philic substitution is unstable and can easily cleave. The un-
supported B − H − B bond only occurs in weak or non-
coordinating solvents, or when it is stabilized by additional
chemical bonding via a subsequent reaction. The B−H−B bond
has three possibilities in subsequent reaction steps.
Scheme 1B), the hydride on the B−H bond usually attacks
another boron atom (blue) to form an unsupported B−H−B 3c-
2e bond. In fact, the dimerization of two BH3 groups to form
diborane belongs to this type of reactions (Scheme S1a). With
this important property of the B−H bond, new boron chemistry
has been developed recently.11a-d For example, the hydride of
the B−H bond in ammonia borane (NH3BH3, AB) has been
used to substitute THF in THF·BH3 to form ammonia dibo-
rane (NH3BH2(µ-H)BH3, AaDB), which is unstable but can be
observed in 11B NMR.11a The observation of this intermediate
involving the unsupported B−H−B bond is critical for under-
standing the formation mechanism of the diammoniate of di-
borane ([H2B(NH3)2][BH4], DADB), which has puzzled chem-
ists for many years.11a,d-i Furthermore, using the same nucleo-
philic substitution of NH3BH3 and THF·BH3, a facile prepara-
tive method for aminodiborane (NH2B2H5, ADB) has been
(1) The unsupported B−H−B bond is cleaved by a subse-
–
quent reaction. As described above, when NH2BH3 reacted
–
with THF·BH3 to form NH2BH2(µ-H)BH3 , the unsupported B
−H−B bond broke, resulting in the formation of the final prod-
ucts of the (NH2BH2)n polymer and BH4– anions.14 In this reac-
tion, the formation of the (NH2BH2)n precipitate and the stable
–
BH4 anion provides the reaction driving force. Another ex-
–
–
ample is the B2H7 anion formed by the reaction of BH4 with
–
THF·BH3: it can reversibly convert to THF·BH3 and BH4 by
cleaving the unsupported B−H−B bond (Figure S2). It is belei-
ved that, compared with a bridge hydride, the coordinating
solvents are better eletron donors and the solvent molecules
can readily break the weak B−H−B bond to replace the bridge
hydride from the B atom in the B−H−B bond.
(2) The unsupported B − H − B bond survives in non-
coordinating solvents. The stability of the B−H−B bond de-
pends on the counter cation and the coordination ability of the
solvents. Although LiB2H7 cannot be isolated from the reacti-
on of LiBH4 and THF·BH3 in THF solution (Figure S2), the
reaction of NR4BH4 (R = C2H5, C4H9) with B2H6 in CH2Cl2
solution afforded NR4B2H7 which was isolated (Eq. 3):15
–
developed.11b,c From this point of view, the borohydrate (BH4 )
–
and its organic borohydrate derivatives (BH2R2 ) can also be
considered as a nucleophilic reagent and their reactions with
transition metal or lanthanide cations to form single, double,
or triple M−H−B bonds should belong to the nucleophilic sub-
stitution reactions.12
RESULTS AND DISCUSSION
NR4BH4 + 1/2 B2H6 → NR4B2H7 (R = C2H5, C4H9)
(3)
Nucleophilic behaviors of the B−H bond. According to
the Lewis base property of the B−H bond, four-coordinated
–
This result indicates that B2H7 is only stable in non-
coordinating solvents with an organic counter cations.
–
RʺBH3 complexes (Rʺ = H–, NH2 , THF, and NH3) can act as a
nucleophile using its B−H bond to substitute a weaker base
from a Lewis acid-base adduct. The role of the bonding pair
electrons in the B−H bond in this reaction is similar to that of
the lone pair electrons in classical nucleophiles. For example,
(3) The unsupported B−H−B bond survives in the final
products with the cooperation of other chemical bonds. In the
reaction of AB with THF·BH3 in THF solution, the hydride in
AB attacked the B in THF·BH3 to substitute THF forming the
AaDB intermediate, which is unstable and can only be ob-
served in its 11B NMR in the reaction solution.11a However,
−
both the lone pair electrons and the B−H bond in the NH2BH3
anion can act as a nucleophile leading to nucleophilic substitu-
tion reactions, but the products are different. When
NaNH2BH3 reacted with AB, the lone pair electrons on N at-
tacked the B atom in AB to form Na[NH2(BH3)2] with one
NH3 leaving (Figure S1 and Scheme 2A).13 In contrast, when
NaNH2BH3 reacted with THF·BH3, the final products are
(NH2BH2)n polymer and BH4–.14 It is believed that in this reac-
tion the B−H bond in the NH2BH3– anion acts as a nucleophile
Scheme 3. Formation of ADB by reaction of NH3BH3 and
THF·BH3 in which the formation of the B-NH2-B bond stabi-
lizes the formed B-H-B bond
–
to attack the B atom in THF·BH3 to form NH2BH2(µ-H)BH3
2
ACS Paragon Plus Environment

Page 3 of 10
Journal of the American Chemical Society
the subsequent molecular hydrogen elimination resulted in the
computed to be 15.8 kcal/mol in THF solvent (Table 1). Then,
a terminal hydride of B2H7– further attacks the B in the second
THF·BH3 to replace THF to form B3H10– (Figure 2). This reac-
tion requires a free energy of 10.3 kcal/mol, but the reverse
energy barrier is much lower (2.4 kcal/mol, Figure 2).
1
2
3
4
5
6
7
formation of a stable four-membered ring compound (ADB),
in which the formation of the B−NH2−B bond stabilizes the B−
H−B bond,11b,c as depicted in Scheme 3 and Scheme S1b. An-
other example is the dimerization of two BH3 groups to form
diborane in which one unsupported B−H−B bond formed first
and the second B−H−B bond formed subsequently, resulting in
a relatively stable B2H6 (Scheme S1a).
These reactions demonstrate that the nucleophilic substitu-
tion of the B−H bond can lead to a new B−H−B bond. How-
ever, whether the resulting B−H−B bond can survive in the
final product depends on the reaction conditions15 or if a sub-
sequent reaction can take place to form a stable final prod-
uct.11b,c This is probably why the nucleophilic substitution of
the B−H bond has not attracted much attention, even though it
has been known to occur for a long time.
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
The formation mechanisms of octahydrotriborate anion.
The formation mechanism of octahydrotriborate has been dis-
cussed previously but it has not been fully understood.1a,2k-o On
the basis of the nucleophilic properties of the B−H bond dis-
cussed above, we have elucidated the formation mechanisms
of octahydrotriborate. Due to the better solubility of lithium
borohydride in THF, we have chosen the reaction of LiBH4
with THF·BH3 for the mechanistic study. In this reaction, the
BH4– anion acts as a nucleophilic reagent, using the B−H bond
–
Figure 2. Energy profiles for the reactions from BH4 and
−
THF·BH3 to B3H10
.
Table 1. Calculated electronic energy (at 0 K), enthalpy (at
298 K), and free energy (at 298 K) in kcal/mol.
a
∆E
(0K)
0
∆H
(298K)
0
∆G/∆Gcorr
(298K)
0
–
to attack the B in THF·BH3 and substitute THF to give B2H7
by forming an unsupported B−H−B bond (Scheme 1B). The
B2H7– product was clearly observed in solution on the basis of
11B NMR, even though the product could not be isolated be-
cause of a reversible reaction (Figure S2). Envisioning that the
resulting B2H7– anion may undergo a similar substitution reac-
tion with a second THF·BH3 to form a H3B(µ-H)BH2(µ-
H)BH3 anion (B3H10–), we reacted LiBH4 with 2 equiv of
–
THF⋅BH3+ BH4
TS1
THF+ B2H7
THF⋅BH3+ B2H7
20.6
2.5
0
18.9
2.4
0
18.6/15.8
0.7
0
–
–
TS2
THF + B3H10
15.4
9.9
0
14.3
9.0
0
13.1/10.3
7.9
0
–
–
B3H10
–
THF·BH3. However, only the starting materials and the B2H7
TS3
B2H6 + BH4
5.9
1.3
34.3
33.5
42.4
-4.9
0
13.9
5.0
0
4.8
1.2
31.4
31.8
38.1
-6.1
0
14.7
8.9
0
6.0
0.9
signals were observed in the 11B NMR at room temperature
(Figure S3) or even at 50 °C (Figure 1a). But when the reac-
–
TS4
B2H5+BH3(H2)
TS5
32.0
31.4
36.6
-11.2
0
15.5/12.7
10.1
0
–
tion mixture was refluxed, the B3H8 signal appeared (Figure
1b) and THF·BH3 could be completely consumed after reflux
for about 36 h (Figure 1e).
B3H8– + H2
2THF⋅BH3
TS1-1
THF⋅B2H6 + THF
THF⋅B2H6
TS2-1
B2H6 + THF
3.1
-6.7
4.8
-0.1
5.1
-2.0
aΔGcorr = ΔG ‒ 2.8 kcal/mol, based on free volume theory for
bimolecular reactions in solution.
The DFT calculation further provides significant insight
–
–
about how B3H10 converts to B3H8 by eliminating molecular
hydrogen, which is important because no intermediates could
be captured experimentally. First, we calculated the NPA
charges and found that the charges of all the hydrogen atoms
are negative (Figure 3a), implying that direct hydrogen release
from B3H10 would be difficult. Interestingly, a transition state
TS3 for the formation of B2H6 and BH4 was located with an
energy barrier of only 6.0 kcal/mol (Figure 4). The natural
population analysis (NPA) charges of the B2H6···BH4 com-
Figure 1. The 11B NMR spectra of LiBH4 and THF·BH3 in THF
a) at 50 °C for 12 hours, and under reflux condition for b) 6 hours,
c) 15 hours, d) 24 hours and e) 36 hours.
–
–
In order to gain more insights into the formation mechanism
–
–
of B3H8 , DFT calculations were performed using the Gaussi-
an 09 program.16 The theoretical results are consistent with the
plex showed that the bridge hydrogens are positively charged
(0.11 and 0.19) and the hydrogens of BH4 are negatively
experimental observations: BH4– reacts with the first THF·BH3
–
–
to form B2H7 by attacking the B of THF·BH3 with the B−H
charged (Figure 3b), suggesting the possibility of hydrogen
elimination by the Hδ+···Hδ– interaction, a dihydrogen bond.11d
bond. The free energy barrier of the transition state TS1 is
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 10
high reverse-reaction energy barrier (47.8 kcal/mol). The es-
cape of the resulting H2 gas from the system would also make
the reverse reaction impossible.
1
2
3
4
5
6
7
8
9
The experimental and computational results are fundamen-
tally consistent with the proposed formation mechanism of
–
B3H8 . The low energy barriers of the two nucleophilic substi-
–
–
tution reactions of BH4 and B2H7 with THF·BH3 molecules
to yield B3H10– (15.8 and 10.3 kcal/mol, respectively, in Table
1 and Figure 2) suggested that both reactions could occur at
–
Figure 3. NPA charges of (a) B3H10– and (b) B2H6…BH4 .
–
room temperature (Figure S2). Only the signal of B2H7 could
be observed in the 11B NMR due to the fact that the resulting
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
–
–
B3H10 woukd be rapidly converted back to B2H7 with an en-
ergy barrier of only 2.4 kcal/mol (TS2, reverse reaction) or
–
further converted to B2H6 and BH4 via TS3 with an energy
barrier of only 6.0 kcal/mol. The calculated energy barrier of
–
31.1 kcal/mol (TS4) for the reaction of B2H6 and BH4 to pro-
duce B3H8– in THF supports the experimental observation that
this reaction only occurred under reflux condition. Further-
–
–
more, the transformation from B3H10 to B3H8 is exothermic
–
by 11.2 kcal/mol, so B3H8 is expected to be very stable once
formed.
Observation of the B2H6 intermediate and its roles in the
–
reactions to form B3H8 . The B2H6 formed via TS3 seems to
be a key intermediate in the proposed mechanism. Hence, it
would be important to find out how B2H6 is formed and the
roles it plays in the reactions. However, it is difficult to detect
the formed B2H6 in THF solution because B2H6 would rapidly
react with THF to produce the starting material THF⋅BH3 (re-
action III). On the other hand, it is known that the BH3 group
exists as a B2H6 dimer rather than L⋅BH3 in diethyl ether
(Et2O).17 We also did experiment and found that the BH3
group exists as the B2H6 dimer in dimethoxyethane (DME),
albeit with a limited solubility (Figure S4).11j So we chose
Et2O to trap the resulting gases in the reaction of LiBH4 with
THF·BH3. We indeed captured B2H6 and identified it by 11B
NMR (Figure S5).
Figure 4. Energy profiles for the reaction from B3H10– to B3H8 .
–
−
The B2H5 and BH3(H2) intermediates were formed through
TS4 with a calculated energy barrier of 31.1 kcal/mol in THF
solvent. Then molecular hydrogen elimination (TS5) resulted
–
in the formation of B3H8 , which is exothermic by 11.2
kcal/mol (Figure 4).
According to the energy profiles of Figures 2 and 4, the
B2H6 intermediate has two possible reactions: reacting with
BH4− to reversibly reform B3H10− and further to B2H7− (reaction
In order to determine the roles of the B2H6 intermediate in
the reactions, we used Et2O and DME as solvents to investi-
gate the reaction of B2H6 with LiBH4 to gain further insight
into the proposed mechanisms. In both Et2O and DME sol-
vents, we expect that the hydride of BH4− attacks the B in B2H6,
−
−
I) or reacting with BH4 to convert to B3H8 by releasing one
H2 (reaction II). Furthermore, B2H6 could also react with the
THF solvent to form THF·BH3 as the reverse reactions shown
in Figure 5 (reaction III). Though the possible reactions I and
III have relatively low energy barriers of 5.1 and 2.4 kcal/mol
(TS3 and TS2, reverse reactions) and 7.1 and 2.6 kcal/mol
(TS2-1 and TS1-1, reverse reactions) (Figure 5), respectively,
−
breaking one of its H-bridge bond to form B3H10 by one step
−
(Scheme 4). Even though the nascent B3H10 cannot be ob-
served experimentally, its lower-barrier reverse reactions via
TS2 (Figure 2) should be able to be observed.
Indeed, when an excess amount of B2H6 was bubbled into a
–
a higher temperature would be required to form B3H8 via re-
–
DME solution of LiBH4 at low temperature (-78 °C), the sig-
action II. Once formed, the final product B3H8 is stable and
cannot be reversibly converted to B2H6 and BH4 due to the
̶
–
nals corresponding to the BH4 starting material and the ex-
̶
pected B2H7 product were clearly observed in 11B NMR (Fig-
ure S6a). This observation confirmed that B2H6 reacted with
̶
−
BH4 at low temperature to form B3H10 via TS3 (the energy
barrier of 5.1 kcal/mol, reverse reaction), which then form
−
B2H7 facilely via TS2 (barrier of only 2.4 kcal/mol). The ob-
servation of the BH4− signal implies that the reaction could not
be completed at low temperature. No B2H6 signal was ob-
served, probably due to the limited solubility of B2H6 in DME
at the low temperature. When the experiment was carried out
−
at room temperature, only B2H7 signal was observed (Figure.
S6b), suggesting that BH4 can completely react with exces-
−
sive B2H6 in DME at room temperature. Furthermore, when
the B2H6 bubbling experiment was carried out under reflux
Figure 5. The formation mechanism of B2H6 from two THF⋅BH3.
4
ACS Paragon Plus Environment

Page 5 of 10
Journal of the American Chemical Society
−
condition, the B3H8 signal appeared (Figure. S6c) and simul-
H)BH3 attacked another B in the same intermediate to form
B2H6 with THF leaving and ring closing (TS2-1, 5.1 kcal/mol)
(Figure 5, Scheme S1). The resulting B2H6 can react with BH4
−
1
2
taneously a large amount of BH4 formed. This observation
indicated that the reflux temperature of DME provides enough
reaction driving force for the formation of B3H8 by crossing
−
̶
3
4
5
6
7
8
9
through reactions I or II as described above.
the high energy barrier of TS4 (31.1 kcal/mol). The large
Isotopic experiment indicated that the deuteride/hydride ex-
change between THF⋅BD3 and LiBH4 occured rapidly, thus the
rate of the formation of B2H7 did not change obviously (Fig-
ure S11, Scheme S3). However, the rate of the formation of
LiB3H8 is much slower, which revealed that the cleavage of
the B−H bond to form H2 is the rate-limiting step (Figure S12).
This result is consistent with the proposed mechanisms.
−
amount of BH4 is expected, due to the low energy barrier for
the formation of BH4− via TS1 (reverse reaction).
−
Similar B2H6 bubbling experiment was also carried out in
−
−
Et2O with similar observations. Both B2H7 and BH4 signals
were observed at low and room temperatures (Figures S7, S8a).
However, when the B2H6 bubbling experiment was carried out
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
under reflux, only a BH4 signal was observed in 11B NMR
The combination of experimental and computational studies
indicated that B2H6 can be formed via both proposed pathways.
The formation of B2H6 by the second pathway need to cross
the highest energy of 12.7 kcal/mol (TS1-1) but in the first one
−
−
without the B3H8 signal (Figure S8b). We think the low boil-
ing point of Et2O cannot drive the reaction to cross the high
energy barrier of TS4 (31.1 kcal/mol). On the other hand, the
−
−
−
formed B2H7 anion could convert to BH4 after cleaving BH3
under reflux by crossing TS1 (reverse reaction, Figure 2). The
cleaved BH3 dimerized to B2H6, which escaped from the reac-
is 15.8 kcal/mol (TS1). On considering the fact that B2H7 can
be formed at both room and low (−78 °C) temperatures (Fig-
ure S13), the second proposed pathway seems to be more fa-
vorable. A previous report of the reaction between B2H6 and
alkali metal borohydrides in ethers at 100 ˚C readily produced
octahydrotriborate salts.2k Recent studies indicated that B2H6
produced in situ reacted with sodium borohydride to produce
NaB3H8.2l-o A gas-solid reaction of B2H6 with MgNiH4 was
found to produce Mg(B3H8)2. It is believed that the Mg(BH4)2
intermediate continually reacted with B2H6 to produce the final
product.1a All of these experimental results seem to support the
second pathway, but the first one cannot be ruled out.
−
tion mixture. Thus, only BH4 signal was observed under re-
flux (Scheme S2). These results confirmed the two possible
−
reactions I and II) of B2H6 and BH4 in the proposed mecha-
nism.
Scheme 4. The proposed mechanisms of the reaction between
B2H6 with LiBH4 in DME to produce LiB2H7 and LiBH4 or
produce LiB3H8 under reflux.
The experimental and computational results discussed
above suggested that the two proposed pathways from the
−
−
starting reagents of THF⋅BH3 and BH4 to the B2H6 + BH4
intermediates consist of a complicagted set of reactions and
reverse reactions, that can be represented by a reaction circle
at room temperature (Scheme S4). All the species in the circle,
THF⋅BH3 + BH4 − , B2H7 − , B3H10 − , B2H6 + BH4 − , and
THF⋅BH2(µ-H)BH3, could be reversibly converted to its
neighboring species at room temperature. The elimination of
−
−
H2 could not happen directly from B3H10 to produce B3H8 .
−
Only at reflux condition, could the reaction of B2H6 with BH4
−
afford the B2H5 and BH3(H2) intermediates, from which a H2
elimination led to the formation of B3H8 (Scheme S4). This
−
Trapping B2H6 formed from THF⋅BH3. After finding the
level of complexity underlies the reason why it has been diffi-
−
crucial roles of the B2H6 intermediate in the reactions to form
cult to understand the formation mechanisms of B3H8 .
−
B3H8 , we further explored whether B2H6 could be formed by
New methods for the syntheses of lithium and sodium
octahydrotriborate. The previously reported syntheses of the
octahydrotriborate anion were all carried out by reacting bora-
nes (diborane or THF·BH3) with alkali metal amalgam or alka-
li metal dispersing in other media such as silicagel or innert
salt, or in the presence of other electron carriers (naphthaleni-
de and tetraphenylborate).2a-j Moreover, in these reactions the
final products are B3H8– and BH4– salts in 1:1 ratio (Eq.1), i.e.,
at least a quarter of boranes did not convert to the expected
product. There has been no report for preparation of octahy-
drotriborates by straightforwardly reacting borohydride with
THF·BH3 . Based on the understanding of the nucleophilicity
other pathways, other than from B3H10−. The BH3 group was
known to exist as a B2H6 dimer in Et2O, so it was used to trap
the resulting gases when a THF⋅BH3 solution (1M) was re-
fluxed without addition of LiBH4 for 36 hours. Indeed, we
found that B2H6 and other unidentified boranes were formed
(Figure S9). We were surprised that B2H6 could also be
trapped in Et2O when N2 was simply bubbled through a
THF⋅BH3 solution (1M) at room temperature (Figure S10).18
Thus, the formed B2H6 in the proposed mechanism might be
from the THF⋅BH3 solution. With this consideration, the sec-
ond possible pathway was proposed (Scheme 4) and the calcu-
lation results support the new proposed pathway for the forma-
tion of B2H6 (Figure 4).
–
of the B−H bond and the formation mechanism of B3H8 , we
have developed a new and facile synthetic method for lithium
and sodium salts of octahydrotriborate (Eq. 2). A mixture of
LiBH4 and THF·BH3 in 1:2 ratio in THF solution was refluxed
for 36 h. We observed that most of the starting materials were
In THF⋅BH3 solution, the B−H bond in one THF⋅BH3 mole-
cule acts as a nucleophile to attack the B in another THF⋅BH3
molecule to form THF⋅B2H6 (TS1-1, 12.7 kcal/mol) (Figure 5).
Then, the terminal hydride of the BH3 moiety in THFBH2(µ-
–
converted to B3H8 , according to Eq. 2.
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 10
autoclave. When a 1 M THF·BH3 solution reacted with LiBH4
1
2
3
4
5
6
7
8
in the autoclave at 65 ℃ for 15 h, a NMR yield of 78.5%
(Figure S24) and an isolated yield of 68% were achieved.
Hence, we conclude that the reaction in autoclave is an effi-
cient way for the preparation of lithium octahydrotriborate.
Slight decomposition of LiB3H8 to LiBH4 was also observed
in refluxing THF after 36 hours (Figure S25). The unreacted
or decomposed LiBH4 could be easily removed by the addition
of a small amount of water.
Sodium octahydrotriborate can be similarly prepared by us-
ing NaBH4 and THF·BH3, according to Eq. 2. A longer reac-
tion time (66 h) was required because of poor solubility of
NaBH4 in THF. The purification procedure is similar to that of
LiB3H8 (Figure S26). It is worthy to note that, in this newly
developed procedure, unsolvated NaB3H8 can be obtained
simply by addition of only CH2Cl2 without further additoin of
Et2O as described in the literature.2g The purity of the final
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
Figure 6. The 11B NMR spectra of the reaction solution of LiBH4
and THF·BH3 in THF refluxed for 36 hours (a) and after addition
of water (0.5 mL) into the reaction solution (b).
Only a small amount of LiBH4 was left in the solution on
the basis of the 11B NMR (Figure 6a). After a small amount of
water was added to consume the unreacted LiBH4 (Figure 6b),
the reaction mixture was filtered and the filtrate was then
pumped under kinetic vacuum to produce an oily solid product,
1
product is more than 95% on the basis of 11B and H NMR
(Figures S27, S28) with a 43% isolated yield.
The isolated yield of NaB3H8 was incresed to 51% by using
an autoclave at 65 °C for 24 h via the reaction of NaBH4 with
1 M THF⋅BH3, which is not a huge increase in comparison
with the 43% yield for the reaction under reflux. The increase
of the isolated yield is not as much as that in the LiBH4 reac-
tion. This could be attributed to the longer reaction time be-
cause of the limited solubility of NaBH4 in THF, leading to
more THF⋅BH3 decomposition.
1
which was confirmed by 11B and H NMR (Figures S14 and
S15) to be the desired Li(THF)1.5B3H8 product with a purity of
95% in 41% isolated yield.
A small amount of the LiBH4 reactant was always observed
in the reaction solution, even if an excess amount of THF·BH3
was used. This observation was mainly due to the fact that part
of THF⋅BH3 reactant was converted to B2H6 and other borane
complexes (unidentified currently) when the THF⋅BH3 solu-
tion was refluxed. The formed B2H6 was evaporated from the
reacting system, as mentioned above (Figure S9), resulting in
a net loss of the THF⋅BH3 reactant. We found that the loss of
THF⋅BH3 via the release of B2H6 depended on the THF⋅BH3
concentration. When a 1 M THF·BH3 solution was refluxed
for 36 hours, about 32% of BH3 was lost on the basis of the
integrated values in 11B NMR using NMe3BH3 as an internal
standard (Figure S16). When a 0.5 M THF·BH3 solution was
refluxed for 40 hours, about 25% of BH3 was lost (Figure S17).
When a 0.25 M THF·BH3 solution was refluxed for 40 hours,
about 14% of BH3 was lost (Figure S18). The inevitable loss
of THF⋅BH3 via B2H6 release under reflux was the reason why
a small amount of unreacted LiBH4 was always observed,
even when an excess amount of THF⋅BH3 was used.
The reactions of LiBH4 with different concentrations of
THF·BH3 were carried out in order to examine the influence
of the THF·BH3 concentration on the yields and to optimize
the reaction condition. As expected, a higher yield was ob-
tained with a lower concentration of THF·BH3. When a 0.5 M
and 0.25 M THF·BH3 solution reacted with LiBH4, the yields
were 65% and 73%, respectively, based on the integrated val-
ues in 11B NMR (NMR yield,19 Figures S19 and S20); and the
isolated yields were 54% and 66%, respectively. These are
much higher than 55% (NMR yield) and 41% (isolated yield)
of the reaction using 1M THF·BH3 solution (Figure S21). Fur-
thermore, when a 20% excess amount of 0.5 M or 0.25 M
THF·BH3 solution was used in the reaction, the yield was fur-
ther increased to 79% and 81% (NMR yield) (Figures S22 and
S23), and 69% and 73% (isolated yield) calculated on the ba-
sis of LiBH4.
We should point out that the similar procedure cannot be
used to prepare potassium octahydrotriborate probably due to
the very low solubility of KBH4 in THF (Figure S29).
CONCLUSIONS
–
In summary, the formation mechanisms of B3H8 from the
−
direct reaction of BH4 and THF·BH3 has been elucidated by
the combination of experimental and computational studies
based on the understanding of the nucleophilicity of the B−H
bond. An intuitive mechanism for the formation of LiB3H8
was proposed. The nucleophilic substitution of the B−H bond
–
in BH4– to THF in THF·BH3 produced B2H7 . The terminal B−
H bond in B2H7– further substitutes THF in a second THF·BH3
to produce B3H10–. Theoretical calculations indicated that a H2
–
molecule cannot be directly eliminated from B3H10 to form
–
−
−
B3H8 , but B3H10 could convert to B2H6 and BH4 and then to
−
B2H5 and BH3(H2) intermediates, from which the elimination
of H2 resulted in the final B3H8 product. The high calculated
energy barrier of the B2H6 + BH4 ! B2H5 + BH3(H2) step
−
−
−
(TS4 in Figure 4) is consistent with the experimental observa-
–
tion that this step can only take place under reflux. The B2H7
intermediate was observed in 11B NMR and the resulting B2H6
intermediate was captured. In another proposed mechanism,
an initial reaction of two THF⋅BH3 molecules produced B2H6,
which further reacted with BH4 to form B2H5 + BH3(H2) and
then to B3H8 . The conversion from B2H6 to B3H8 was con-
−
−
–
–
−
firmed experimentally. The fact that B2H7 cannot only be
formed at room temperature but also at low temperature (−78
°C) suggests that the second proposed pathway seems to be
−
more favorable for the formation of B3H8 , but the first one
cannot be ruled out. The two proposed pathways consist of a
−
−
reaction circle in which all species of THF⋅BH3 + BH4 , B2H7 ,
−
B3H10−, B2H6 + BH4 , and THF⋅BH2(µ-H)BH3 could be re-
In order to prevent the resulting B2H6 from escaping from
the reaction system, we further carried out the reaction in an
versibly converted to its neighboring species at room tempera-
6
ACS Paragon Plus Environment

Page 7 of 10
Journal of the American Chemical Society
−
Reaction of LiBH4 with THF⋅BH3 at low temperature. In
a drybox, LiBH4 (0.11 g, 5 mmol) was put into a 25 mL flask
which was connected to a Schlenk line and then 10 mL
THF·BH3 (10 mL, 10 mmol) was condensed into the flask at -
78 °C. The reaction mixture was stirred for 2 hours at -78 °C
and then the mixture solution was transferred into a NMR tube
which was kept in a -78 °C bath. The mixture was monitored
by 11B NMR at -88 °C (Figure S13).
ture. Only under reflux, can the B3H10 intermediate be con-
−
1
2
3
4
5
6
7
8
verted to the B3H8 via H2 elimination. On the basis of the
understanding of the formation mechanisms of B3H8 , a facile
−
method for the synthesis of MB3H8 (M = Li and Na) with high
yields has been developed by the reactions of MBH4 and
THF·BH3 under reflux. The synthetic method significantly
simplifies the reaction procedures without using the highly
toxic and reactive diborane, electron carriers, or other reaction
media and allows convenient preparation of MB3H8 in large
scales.
Reaction of NaNH2BH3 with NH3BH3. NaNH2BH3 (0.53 g,
10 mmol) and NH3BH3 (0.31 g, 10 mmol) were placed in a
100 mL flask. About 30 mL THF was condensed into the flask
and the reaction mixture was refluxed for 9 hours. Then 30 mL
n-hexane was added into the solution and a white precipitate
was formed. After filtration, THF and n-hexane were removed
from the filtrate under dynamic vacuum to leave a white pow-
der product Na[BH3NH2BH3] (0.25 g, 38%, Figure S1). 11B
NMR (128 MHZ, THF-d8): δ -19.9 (d, J = 90 Hz) ppm.
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
EXPERIMENTAL SECTION
General remarks. All manipulations were carried out on a
Schlenk line or in a glovebox filled with high-purity nitrogen.
The 11B NMR spectra were obtained at 128 MHz or 193 MHz
and externally referenced to BF3·OEt2 in C6D6 (δ = 0.00 ppm).
The 1H NMR and 1H{11B} spectra were obtained at 400 MHz.
Reaction of B2H6 with LiBH4 in DME. Lithium borohy-
dride (0.11 g, 5 mmol) was added to a 25 mL flask and then 10
mL DME was condensed into the flask at -78 ℃. The prepared
B2H6 by the reaction of NaBH4 and I2 (excess, about 10-15
mmol) with N2 was bubbled into the reaction solution under
stirring and then the solution was monitored by 11B NMR
(Figure S6a). Similar precedures were carried out at both room
temperature and under reflux. The resulting solutions were
monitored by 11B NMR (Figures S6b and S6c).
Chemicals of LiBH4, NaBH4, KBH4, NaH, NaBD4 and I2
were purchased from Sigma-Aldrich and the NaH was washed
with tetrahydrofuran (THF) and n-hexane. THF·BH3,
Me3NBH3, and NH3BH3 were purchased from United Boron
(Zhengzhou) Energy Materials S&T LLC and used as received.
NaNH2BH3 was prepared according to the literature.20 B2H6
was prepared by the reaction of NaBH4 with I2 in DME (see
SI).21 THF, dimethoxyethane (DME), diethyl ether (Et2O), n-
hexane and toluene were dried over sodium/benzophenone and
freshly distilled prior to use. Dichloromethane was dried over
calcium hydride.
The reaction of B2H6 with LiBH4 in Et2O was similar to the
reaction in DME. Detailed procedures were describe in SI.
Stability of THF⋅BH3. A sample of Me3NBH3 (0.73 g, 10
mmol) was added to a 50 mL flask and then about 20 ml of 1
M THF·BH3 was condensed into the flask. Then the solution
was refluxed for 36 hours and the evaporated gases were trap-
ped by Et2O (Figure S9) through the ice-water condenser. The
mixture solution of Me3NBH3 and THF·BH3 was determined
by 11B NMR before and after reflux (Figure S16). Similar pro-
cedures to examine the stability of 0.5 M and 0.25 M
THF·BH3 under reflux were described in SI.
Computational Details. All DFT calculations were per-
formed using the Gaussian 09 program.16 The geometries were
optimized at M062X/6-311++G(d,p) level of theory, and sin-
gle-point energy calculations in the THF solvent with the
SMD approach22 were carried out using the fully optimized
geometries in vacuum. M062X functional was proved to be
very accurate for reproducing thermodynamic data, barrier
heights, and isomerization energies.23 Vibrational frequencies
were calculated at the same level to determine whether a spe-
cies was an equilibrium structure or a transition state (TS).
Occurence of imaginary vibrational frequencies would suggest
the related species to be a transition state (TS). Natural popu-
lation analysis (NPA) was used to evaluate atomic charges.
Gibbs free energy (ΔG, kcal/mol at 298 K) barriers were com-
puted to understand the reactions. For bimolecular reactions, it
was difficult to accurately estimate activation entropies for the
solution-based reactions. We corrected the activation Gibbs
free energy ΔG (in kcal/mol) at 298 K using the free volume
theory.24 This providesd a rate for a solution-based bimolecu-
lar reaction that was about 80 times faster than the correspon-
ding reaction in the gas phase. This result suggested that the
activation free energy for a solution reaction should be about
2.8 kcal/mol smaller than the value calculated in the gas phase.
The Cartesian coordinates and vibrational frequencies of the
studied model species are listed in the Supporting Information.
A sample of 1 M THF·BH3 (20 mL, 20 mmol) solution was
condensed into a 50 mL flask. At room temperature, N2 was
bubbled through the reaction solution at a flow rate of 5 mL
min–1 for 6 hours under constant stirring. And then the gases
were trapped by Et2O through ice-water condenser. The trap-
ped Et2O solution was examined by 11B NMR (Figure S10).
Synthesis of Li(THF)1.5B3H8. Lithium borohydride (0.44 g,
20 mmol) was added to a 100 mL flask and then 1 M
THF·BH3 (40 mL, 40 mmol) was condensed into the flask.
The reaction mixture was refluxed under stirring for 36 hours.
The formed gases were passed through tol-d8, in which the
formed H2 was identifed (Figure S30). After reaction, a small
amount of water (0.5 mL) was injected into the solution to
destroy the residue LiBH4. After filtration, THF was removed
from the filtrate under dynamic vacuum to leave
Li(THF)1.5B3H8 as a clear oily product (NMR yield: about 55%,
isolated yield: 1.27 g, 41% (Figures S14 and S15). 11B NMR
(128 MHZ, D2O): δ -31.2 (nonet, J = 33 Hz) ppm. 11B NMR
(128 MHZ, CD3CN): δ -30.8 (nonet, J = 33 Hz) ppm. 1H NMR
(400 MHZ, D2O): δ 3.77 (m, 4α-H of THF), 1.91 (m, 4β-H of
THF), 0.30 (decet, J = 33 Hz, 8H of B3H8) ppm. 1H{11B} (400
MHZ, D2O): δ 3.78 (m, 4α-H of THF), 1.92 (m, 4β-H of THF),
0.3 (s, 8H of B3H8) ppm. 1H NMR (400 MHZ, CD3CN): δ 3.65
(m, 4α-H of THF), 1.80 (m, 4β-H of THF), 0.13 (decet, J = 33
Reaction of LiBH4 with THF⋅BH3 at room temperature.
In a drbox, LiBH4 (0.11 g, 5 mmol) was put into a 50 mL flask
which was connected to a Schlenk line and then 20 mL
THF·BH3 (20 mmol) was condensed in to the flask at -78 °C.
The reaction mixture was stirred for 20 hours at room tempe-
rature. After the LiBH4 reagent was completely consumed
(Figure S2a), THF and excess THF·BH3 were removed from
the reaction solution under dynamic vacuum. The final residue
sticky solid was identified as LiBH4 by 11B NMR (Figure S2b).
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 10
Hz, 8H of B3H8) ppm. 1H{11B} (400 MHZ, CD3CN): δ 3.65 (m,
Corresponding Author
1
2
4α-H of THF), 1.81 (m, 4β-H of THF), 0.13 (s, 8H of B3H8)
ppm.
xnchen@htu.edu.cn; jie.zhang@htu.edu.cn;
lai-sheng wang@brown.edu
3
4
5
Similar procedures for synthesis of Li(THF)1.5B3H8 by
LiBH4 with different concentrations of THF⋅BH3 solutions
were described in SI.
Author Contributions
⊥These authors contributed equally to this work.
6
7
8
9
Synthesis of Li(THF)1.5B3H8 in autoclave. Lithium boro-
hydride (0.22 g, 10 mmol) was added to a 50 mL autoclave
and then 1 M THF·BH3 (20 mL, 20 mmol) was condensed into
the autoclave. The reaction mixture was stirred for 15 hours at
65 ℃. Then THF and residue THF·BH3 were removed from
the reaction solution to leave a oily product. 20 mL THF and
water (0.1 mL) were injected into the oily product to remove
the residue LiBH4. After the formed precipitate was filtered,
THF was removed from the filtrate under dynamic vacuum to
leave Li(THF)1.5B3H8 as a clear oily product (NMR yield:
about 78.5%, isolated yield: 1.06 g, 68%, Figure S24).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
The authors are grateful for the financial support from the Na-
tional Natural Science Foundation of China (Grant Nos.
21771057, 21571052, and 21603062) and the U.S. National Sci-
ence Foundation (DMR-1655066 to L.S.W.). The authors would
also like to thank Prof. Jun Li from Tsinghua University, Prof.
Todd Marder from Universität Würzburg, Prof. Yanhui Guo from
Fudan University, and Prof. Thomas Evans for valuable discus-
sions.
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
Synthesis of unsolvated NaB3H8. Sodium borohydride
(0.76 g, 20 mmol) was added to a 100 mL flask and then
THF·BH3 (40 mL, 40 mmol) was condensed into the flask.
The reaction mixture was refluxed for 66 hours under stirring.
After filtration, water (0.05 mL) was injected into the solution
to consume the soluble residue NaBH4 in solution. The formed
precipitate was filtered and then THF was removed from the
filtrate under dynamic vacuum to leave an oily product. The
oily product was dissolved in CH2Cl2 and converted into sus-
pension under stirring, after filtration, unsolvated NaB3H8 was
obtained and dried under dynamic vacuum. Approximately
0.55 g NaB3H8 was obtained with the 43% isolated yield (Fi-
gures S27 and S28, no NMR yield is available because NaBH4
cannot be completely disolved in THF). 11B NMR (128 MHZ,
CD3CN): δ -30.0 (nonet, J = 33 Hz) ppm. 1H NMR (400 MHZ,
CD3CN): δ 0.14 (decet, J = 33 Hz) ppm.
REFERENCES
(1) (a) Huang, J.; Yan, Y.; Remhof, A.; Zhang, Y.; Rentsch, D.; Au,
Y. S.; Jongh, P. E.; Cuevas, F.; Ouyang, L.; Zhu, M.; Züttel, A.
Dalton Trans. 2016, 45, 3687-3690. (b) Goedde, D. M.; Windler, G.
K.; Girolami, G. S. Inorg. Chem. 2007, 46, 2814-2823. (c) Bykov, A.
Yu.; Zhizhin, K. Yu.; Kuznetsov, N. T. Russ. J. Inorg. Chem. 2014,
59, 1539-1555.
(2) (a) Stock, A.; Poland, E. Ber. Dtsch. Chem. Ges. 1926, 59,
2210-2215. (b) Stock, A.; Laudenklos, H. B. Z. Anorg. Allgem. Chem.
1936, 228, 178-192. (c) Stock, A. Hydrides of Boron and Silicon,
Cornell University, 1933, pp. 58-58. (d) Hill, T. G.; Godfroid, R. A.;
White, J. P.; Shore, S. G. Inorg. Chem. 1991, 30, 2952-2954. (e) Chen,
W.; Wu, G.; He, T. Li, Z.; Guo, Z.; Liu, H.; Huang, Z.; Chen, P. Int. J.
Hydrogen Energy 2016, 41, 15471-15746. (f) Dunbar, A. C.; Macor, J.
A.; Girolami, G. S. Inorg. Chem. 2014, 53, 822-826. (g) Huang, Z.;
King, G.; Chen, X. Hoy, Yisgedu, J. T. H.; Lingam, K.; Shore, S. G.;
Woodward, P. M.; Zhao, J.-C. Inorg. Chem. 2010, 49, 8185-8187. (h)
Fu, H.; Wang, X.; Shao,Y.; Chen, J.; Zhang, X.; Fu, H.; Zheng, J.; Li,
X. Int. J. Hydrogen Energy 2016, 41, 384-391. (i) Zheng, X.; Yang,
Y.; Zhao, F.; Fang, F.; Guo, Y. Chem. Commun. 2017, 53, 11083-
11086. (j) Moury, R.; Gigante, A.; Hagemann, H.; Int. J. Hydrogen
Energy 2017, 42, 22417-22421. (k) Gaines, D. F.; Schaeffer, R.;
Tebbe, F. Inorg. Chem. 1963, 2, 526-528. (l) Bykov, A. Yu.; Razgon-
yaeva, G. A.; Mal’tseva, N. N.; Zhizhin, K. Yu.; Kuznetsov, N. T.
Russ. J. Inorg. Chem. 2012, 57, 471-473. (m) Myakishev, K. G.;
Volkov, V. V. Sibirskii Khim. Zh. 1991, 7, 144. (n) Volkov, V. V.;
Myakishev, K. G.; Gorbacheva, I. I. Zh. Neorg. Khim. 1989, 34,
1665-1668. (o) Volkov, V. V.; Gorbacheva, I. I.; Myakishev, K. G. Zh.
Neorg. Khim. 1985, 30, 593-597.
(3) Kasper, J. S.; McCarty, L. V.; Newkirk, A. E. J. Am. Chem. Soc.
1949, 71, 2583-2583.
(4) Soldate, A. M. J. Am. Chem. Soc. 1947, 69, 987-988.
(5) (a) Buchi, G. D.; Ayer, E. J. Am. Chem. Soc. 1956, 78, 689-690.
(b) Hough, W. V.; Edwards, L. J.; McElroy, A. D. J. Am. Chem. Soc.
1958, 80, 1828-1829.
(6) Olson, J. K.; Boldyrev, A. I. Comput. Theor. Chem. 2011, 967,
1-4.
(7) (a) Tanaka, N.; Shoji, Y.; Hashizume, D.; Sugimoto, M.; Fuku-
shima, T. Angew. Chem. Int. Ed. 2017, 56, 5312-5316. (b)
Braunschweig, H.; Chiu, C.-W.; Gamon, D.; Gruß, K.; Hörl, C.;
Kupfer, T.; Radacki, K.; Wahler, J. Eur. J. Inorg. Chem. 2013, 9,
1525-1530. (c) Romanescu, C.; Galeev, T. R.; Li, W. L.; Boldyrev, A.
I.; Wang, L.-S. Acc. Chem. Res. 2013, 46, 350-358. (d) Zhai, H. J.;
Chen, Q.; Bai, H.; Li, S. D.; Wang, L.-S. Acc. Chem. Res. 2014, 47,
2435-2445. (e) Kawai, S.; Saito, S.; Osumi, S.; Yamaguchi, S.; Foster,
A. S.; Spijker, P.; Meyer, E. Nat. Commun. 2015, 6, 8098. (f) Kaiser,
R. I.; Balucani, N. Acc. Chem. Res. 2017, 50, 1154−1162.
Synthesis of unsolvated NaB3H8 in autoclave. Sodium bo-
rohydride (0.38 g, 10 mmol) was added to a 50 mL autoclave
and then 1 M THF·BH3 (20 mL, 20 mmol) was injected into
the autoclave. The reaction mixture was stirred for 24 hours at
65 ℃. After the insoluble NaBH4 was filtered away, the THF
and residue THF·BH3 was removed from the reaction solution
to leave a oily product. 20 mL THF and water (0.05 mL) were
injected into the oily product to destroy the soluble residue
NaBH4. After filtration, THF was removed from the filtrate
under dynamic vacuum to leave a clear oily product. The oily
product was dissolved in CH2Cl2 and converted into suspensi-
on under stirring, after filtration, unsolvated NaB3H8 was ob-
tained and dried under dynamic vacuum (0.32 g, 51%).
Reaction of KBH4 with THF·BH3. Potassium borohydride
(0.54 g, 10 mmol) was added to a flask and then THF·BH3 (20
mL, 20 mmol) was condensed into the flask. The reaction mix-
ture was refluxed for 80 hours under stirring, very small
amount of B3H8– product was observed (Figure S29).
ASSOCIATED CONTENT
Supporting Information
Supporting Information is available free of charge on the ACS
Publications website.
Details for experiments, characterizations, and computa-
tions (PDF).
AUTHOR INFORMATION
8
ACS Paragon Plus Environment

Page 9 of 10
Journal of the American Chemical Society
mand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.;
(8) (a) Brown, H. C.; Suzuki, A.; Miyaura, N.; Itoh, M.; Holland, G.
Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.;
Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A.
D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J.
Gaussian 09, Revision D.01; Gaussian, Inc.: Wallingford, CT, 2009.
(17) Gaines, D. F. Inorg. Chem. 1963, 2, 523-526.
W.; Negishi, E. J. Am. Chem. Soc. 1971, 93, 2792-2793. b) Brown, H.
C.; Gupta, S. K. J. Am. Chem. Soc. 1971, 93, 1816-1818. (c) Brown,
H. C.; Gupta, S. K. J. Am. Chem. Soc. 1970, 92, 6983-6984. (d)
Brown, H. C.; Kabalka, G. W. J. Am. Chem. Soc. 1970, 92, 714-716.
(e) Xu, S.; Zhang, Y.; Li, B.; Liu, S.-Y. J. Am. Chem. Soc. 2016, 138,
14566-14569. f) Yang, D.-T.; Mellerup, S. K.; Wang, X.; Lu, J.-S.;
Wang, S. Angew. Chem. Int. Ed. 2015, 54, 5498–5501.
(9) Geier, S. J.; Vogels, C. M.; Westcott, S. A.; Boron Reagents in
Synthesis, Vol.1236, ACS Symposium Series, 2016, pp. 209-225.
(10) (a) Heuclin, H.; S. Ho, Y.-F.; Goff, X. F. L.; So, C.-W.;
Mézailles, N. J. Am. Chem. Soc. 2013, 135, 8774-8777. ( b) Borger, J.
E.; Ehlers, A. W.; Lutz, M.; Slootweg, J. C.; Lammertsma, K. Angew.
Chem. Int. Ed. 2016, 55, 613-617. (c) Wrackmeyer, B.; Klimkina, E.
V.; Milius, W. Eur. J. Inorg. Chem. 2014, 28, 4865-4876. (d) Chen,
W.-C.; Lee, C.-Y.; Lin, B.-C.; Hsu, Y.-C.; Shen, J.-S.; Hsu, C.-P.;
Yap, G. P. A.; Ong, T.-G. J. Am. Chem. Soc. 2014, 136, 914-917. (e)
Celik, M. A.; Sure, R.; Klein, S.; Kinjo, R.; Bertrand, G.; Frenking, G.
Chem. Eur. J. 2012, 18, 5676 – 5692. (f) Jonas, V.; Frenking, G.;
Reetz, M. T. J. Am. Chem. Soc. 1994, 116, 8141-8753.
(11) (a) Chen, X.; Bao, X.; Zhao, J.-C.; Shore, S. G. J. Am. Chem.
Soc. 2011, 133, 14172-14175. (b) Chen, X.; Zhao, J.-C.; Shore, S. G.
J. Am. Chem. Soc. 2010, 132, 10658-10659. (c) Li, H.; Ma, N.; Meng,
W.; Gallucci, J.; Qiu, Y.; Li, S.; Zhao, Q.; Zhang, J.; Zhao, J.-C.;
Chen, X. J. Am. Chem. Soc. 2015, 137, 12406-12414. (d) Chen, X.;
Zhao, J.-C.; Shore, S. G. Acc. Chem. Res. 2013, 46, 2666-2675. (e)
Zhao, Q.; Li, J.; Hamilton, E. J. M.; Chen X., J. Organomet. Chem.
2015, 798, 24-29. (f) Li, H.; Yang, Q.; Chen, X.; Shore, S. G. J. Or-
ganomet. Chem. 2014,751, 60-66. (g) Shore, S. G.; Hall, C. L. J. Am.
Chem. Soc. 1966, 88, 5346-5347. (h) Eastham, J. F.; J. Am. Chem.
Soc. 1967, 89, 2237-2238. (i) Shore, S. G.; Hall, C. L. J. Am. Chem.
Soc. 1967, 89, 3947-3948. (j) Chen, X.; Bao, X.; Billet, B.; Shore, S.
G.; Zhao, J.-C. Chem. Eur. J. 2012, 18, 11994-11999.
(12) (a) Xu, Z.; Lin, Z. Coordin. Chem. Rev. 1996, 156, 139-162.
(b) Maseras, F.; Lledo´s, A. Chem. Rev. 2000, 100, 601-636. (c) Es-
teruelas, M. A.; López, A. M.; Oliván, M. Chem. Rev. 2016, 116,
8770−8847. (d) Koehne, I.; Schmeier, T. J.; Bielinski, E. A.; Pan, C.
J.; Lagaditis, P. O.; Bernskoetter, W. H.; Takase, M. K.; Würtele, C.;
Hazari, N.; Schneider, S. Inorg. Chem. 2014, 53, 2133−2143. (e)
Desrochers, P. J.; Sutton, C. A.; Abrams, M. L.; Ye, S.; Neese, F.;
Telser, J.; Ozarowski, A.; Krzystek, J.; Inorg. Chem. 2012, 51,
2793−2805. (f) Arliguie, T.; Belkhiri, L.; Bouaoud, S.-E.; Thue´ry, P.;
Villiers, C.; Boucekkine, A.; Ephritikhine, M. Inorg. Chem. 2009, 48,
221-230. (g) Demachy, I.; Esteruelas, M. A.; Jean, Y.; Lledo´s, A.;
Maseras, F.; Oro, L. A.; Valero, C.; Volatron, F. J. Am. Chem. Soc.
1996, 118, 8388-8394. (h) Marks, S.; Heck, J. G.; Habicht, M. H.;
Oña-Burgos, P.; Feldmann, C.; Roesky, P. W.; J. Am. Chem. Soc.
2012, 134, 16983−16986. (i) Meyer, N.; Jenter, J.; Roesky, P. W.;
Eickerlingc, G.; Scherer, W. Chem. Commun. 2009, 4693–4695.
(13) Daly,S. R.; Bellott, B. J.; Kim, D. Y.; Girolami, G. S. J. Am.
Chem. Soc. 2010, 132, 7254-7255.
1
2
3
4
5
6
7
8
(18) Although attempt to trap and detect the formed B2H6 by the
similar experiment at low temperature is unsuccessful, it dose not
imply that B2H6 cannot be formed in the THF solution.
9
(19) The NMR yield was obtained from the integrated values in 11B
NMR. The calculation was based on the LiB3H8 product and the
LiBH4 starting material since part of THF⋅BH3 would convert to B2H6
and other unidentified boranes, which escaped from the system. For
example, in Figure S19, the NMR yield of LiB3H8 was:
(1/3)/[(1/3)+0.18] = 65%. The NMR yield might not be very accurate
because the integral values were from different types of boron atoms,
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
−
B3H8− or BH4 .
(20) Xiong, Z.; Yong, C. K.; Wu, G.; Chen, P.; Shaw, W.;
Karkamkar, A.; Autrey, T.; Jones, M. O.; Johnson, S. R.; Edwards, P.
P.; David, W. I. F. Nat. Mater. 2008, 7, 138–141.
(21) Narayana, C.; Periasamy, M. J. Organomet. Chem. 1987, 323,
145-147.
(22) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem.
B 2009, 113, 6378−6396.
(23)(a) Zhao, Y.; Truhlar, D. G. J. Phys. Chem. A 2006, 110,
5121−5129. (b) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120,
215−241.
(24) (a) Liu, Q.; Lan, Y.; Liu, J.; Li, G.; W, Y.-D.; Lei, A. J. Am.
Chem. Soc. 2009, 131, 10201-10210. (b) Li, Z.; Fu, Y.; Guo, Q.-X.;
Liu, L. Organometallics 2008, 27, 4043-4049.
.
(14) Chen, X.-M.; Li, H.; Yang, Q.-Y.; Wang, R.-R.; Hamilton, E.
J. M.; Zhang, J.; Chen, X. Eur. J. Inorg. Chem. 2017, 38-39, 4541–
4545.
(15) (a) Shore, S. G.; Lawrence, S. H.; Watkins, R. Bau, M. I. J.
Am. Chem. Soc. 1982, 104, 7669-7670. (b) Hertz, R. K.; Johnson, H.
D.; Shore, S. G.; Inorg. Chem. 1973, 12, 1875-1877. (c) Evans, W. G.;
Holloway, C. E.; Sukumarabandhu, K.; McDaniel, D. H. Inorg. Chem.
1968, 7, 1746-1748. (d) Brown, H. C.; Stehle, P. F.; Tierney, P. A. J.
Am. Chem. Soc. 1957, 79, 2020-2021. (e) Brown, H. C.; Tierney, P.
A.; J. Am. Chem. Soc. 1958, 80, 1552-1558.
(16) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.;Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada,
M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.;
Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Mont-
gomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.;
Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Nor-
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 10
–
–
The formation mechanisms of B3H8 through direct reactions between BH4 and THF·BH3 were elucidated by the combination of experimental
1
2
3
4
–
and computational studies in recognizing the nucleophilic properties of the B−H bond in borane complexes. The B2H7 intermediate was observed
in 11B NMR and B2H6 was captured and detected. The conversion from B2H6 to B3H8 was confirmed experimentally. The discovery of the new
–
mechanisms has led to a facile synthetic method for LiB3H8 and NaB3H8 in high yield.
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
For Table of Contents Only
10
ACS Paragon Plus Environment