A simple and efficient way to synthesize unsolvated sodium octahydrotriborate

Article abstract of DOI:10.1021/ic101543v

A simple and efficient way to synthesize unsolvated sodium octahydrotriborate has been developed. This method avoids the use of dangerous starting materials and significantly simplifies the reaction setup, thus enabling convenient large-scale synthesis. The structure of the unsolvated compound has been determined through powder X-ray diffraction.

Full text of DOI:10.1021/ic101543v

Inorg. Chem. 2010, 49, 8185–8187 8185
DOI: 10.1021/ic101543v
A Simple and Efficient Way to Synthesize Unsolvated Sodium Octahydrotriborate
Zhenguo Huang,† Graham King,‡ Xuenian Chen,‡ Jason Hoy,‡ Teshome Yisgedu,† Hima K. Lingam,†
Sheldon G. Shore,‡ Patrick M. Woodward,‡ and Ji-Cheng Zhao*,†
†Department of Materials Science and Engineering, ‡Department of Chemistry, The Ohio State University,
Columbus, Ohio 43210
Received June 6, 2010
A simple and efficient way to synthesize unsolvated sodium octahydro-
triborate has been developed. This method avoids the use of dangerous
starting materials and significantly simplifies the reaction setup, thus ena-
bling convenient large-scale synthesis. The structure of the unsolvated
compound has been determined through powder X-ray diffraction.
Unsolvated NaB3H8 was conventionally synthesized by the
direct reaction between diborane and sodium metal.7 Because
of the extreme volitality of diborane, its on-site generation and
consumption have recently been developed for such a synthe-
sis.8 However, a high-quality airtight apparatus is mandatory
for the reaction, and the disposal of residual diborane needs
to be carried out properly and carefully. A relatively safe way
to synthesize unsolvated NaB3H8 was reported, but it involves
three steps and requires expensive intermediate compounds
such as Bu4NBr and NaBPh4.9 It has been demonstrated that
the synthesis of NaB3H8 in THF,10 glyme, and tetraglyme11
only led to oily solvates, from which crystalline dioxanate
The octahydrotriborate anion, B3H8-, is of interest because
of its high hydrogen content (NH4B3H8,1 20.5 wt % H) and its
utility as chemical vapor deposition (CVD) precursors [Mg-
3
(B3H8)22 and Cr(B3H8)2 ] and as a starting material for other
inorganic species (NH3B3H7,4 17.7 wt % H, a potential hydro-
gen storage material). Most MB3H8 (M = K, Rb, Cs) salts are
insoluble in ethyl ether and soluble in solvents such as tetra-
hydrofuran (THF).5 The introduction of such solvents can
change the nature of compounds from covalent to ionic [e.g.,
Mg(B3H8)2 being covalent2 and [Mg(THF)6][B3H8]2 being
ionic6]. As a result, these two compounds show very different
properties, with Mg(B3H8)2 being more reactive than [Mg-
(THF)6][B3H8]2.2
NaB3H8 3C4H8O2 can be isolated. Vacuum-thermal desolva-
3
tion of NaB3H8 3C4H8O2 resulted in decomposition of
3
NaB3H8.12
Shore and co-workers have found that the B3H8 anion can
be obtained by reacting certain metal amalgams with
THF-borane (THF BH3).5 Here, we adopted this method
3
to obtain the B3H8 anion and, more importantly, developed
an efficient way to isolate the desired unsolvated NaB3H8
from the oily THF solvate. This synthesis avoids the use of
diborane, largely simplifies the reaction setup, and signifi-
cantly reduces the cost. Although the unsolvated NaB3H8
was first prepared almost 7 decades ago, its structure has
never been reported. Attempts to obtain unsolvated NaB3H8
single crystals have been unsuccessful. In this work, we solved
the structure based on powder X-ray diffraction (XRD) data
using Topas-Academic.13
NaB3H8 has certain advantages over the above-mentioned
MB3H8 (M = K, Rb, Cs) salts. One of them is that NaB3H8
has good solubility in ethyl ether. Because ethyl ether usually
has a weak coordination ability, it can be easily pumped
away. The unsolvated NaB3H8 is also essential for the further
synthesis of other octahydrotriborate compounds such as
unsovlated Mg(B3H8)22 and Cr(B3H8)2.3 In addition, sodium
is cheaper than all of the other alkali metals and thus more
cost-effective for a large-scale synthesis.
Na/Hg amalgam was reacted with THF BH3 in a flask at
3
room temperature for 3 days, during which a fine white precipi-
tate appeared. After removal of the THF solvent and the resi-
*To whom correspondence should be addressed. E-mail: zhao.199@osu.edu.
(1) Antonov, P. S.; Pchelkina, M. A.; Nikitin, V. S.; Egorenko, G. A.;
Vinogradova, Z. F.; Kurekova, A. T. Zh. Neorg. Khim. 1973, 18, 613.
(2) Kim, D. Y.; Yang, Y.; Abelson, J. Y.; Girolami, G. S. Inorg. Chem.
2007, 46, 9060.
dual THF BH3 under dynamic vacuum, a white powder in an
3
oil suspension remained. The white powder is NaBH4, and the
oily product is the THF-coordinated NaB3H8. Dry ethyl ether
(3) Goedde, D. M.; Girolami, G. S. J. Am. Chem. Soc. 2004, 126, 12230.
(4) Yoon, C. W.; Carroll, P. J.; Sneddon, L. G. J. Am. Chem. Soc. 2009,
131, 855.
(5) Hill, T. G.; Godfroid, R. A.; White, J. P.; Shore, S. G. Inorg. Chem.
1991, 30, 2952.
(6) Titov, L. V.; Levicheva, M. D.; Psikha, S. B. Zh. Neorg. Khim. 1984,
29, 668.
(7) (a) Stock, A. Hydrides of Boron and Silicon; Cornell University Press:
Ithaca, NY, 1933; pp 58 and 138. (b) Hough, W. V.; Edwards, L. J.; McElroy, A. D.
J. Am. Chem. Soc. 1956, 78, 689. (c) Hough, W. V.; Edwards, L. J.; McElroy,
A. D. J. Am. Chem. Soc. 1958, 80, 1828.
(8) Goedde, D. M.; Windler, G. K.; Girolami, G. S. Inorg. Chem. 2007,
46, 2814.
(9) Titov, L. V.; Levicheva, M. D.; Rosolovskii, V. Ya. Russ. J. Inorg.
Chem. (Engl. Transl.) 1980, 25, 1625.
(10) Gaines, D.; Schaeffer, R.; Tebbe, F. Inorg. Chem. 1963, 2, 526.
(11) Miller, H. C.; Miller, N. E.; Muetterties, E. L. Inorg. Chem. 1964, 3, 1456.
(12) Titov, L. V.; Eremin, E. R.; Rosolovskii, V. Ya. Russ. J. Inorg. Chem.
(Engl. Transl.) 1982, 27, 500.
(13) Topas-Academic. General Profile and Structure Analysis Software for
Powder Diffraction Data; Bruker AXS: Karlsruhe, Germany, 2004.
r
2010 American Chemical Society
Published on Web 08/17/2010
pubs.acs.org/IC

8186 Inorganic Chemistry, Vol. 49, No. 18, 2010
Huang et al.
Figure 1. 11B (a) and 1H NMR (b) spectra of unsolvated NaB3H8 in
CD3CN at room temperature.
was distilled onto the crude product, and then the solution was
transferred to a vacuum extractor. After NaBH4 was filtered
away, the clear ether solution was pumped via a dynamic
vacuum, leaving an oily product. Dry dichloromethane was
then distilled onto the oily product, resulting in the formation of
a white powder suspended in dichloromethane. The powder
was collected through filtration and pumped under dynamic
vacuum to remove any residual solvents. Experimental details
are included in the Supporting Information.
Figure 2. Powder XRD data (dots) of unsolvated NaB3H8, Rietveld
refinement results (line), difference trace (observed - calculated), and
allowed Bragg reflections (tick marks). The inset shows the expanded
scale from 2θ = 35 to 50°.
Isolation of the unsolvated NaB3H8 from the oily product
is a critical step in the procedure. Dichloromethane can help
to break the coordination between THF and the Na cation,
leaving a solvent-free NaB3H8 powder suspended in dichlor-
omethane. Heating the oily solvate at around 50 °C without
dichloromethane helped (yet incompletely) to remove the
coordinated THF, but the desired NaB3H8 decomposed
slowly, and the crude product became yellowish overnight.
NMR spectra indicate the formation of undesired boron
compounds. Only up to now, a simple and efficient method
has been successfully developed to synthesize NaB3H8 and
isolate it from the oily solvate.
Pure NaB3H8 exhibits a single nonet 11B NMR spectrum
corresponding to the octahydrotriborate anion,5 centered at
δ -30.4 ppm (Figure 1a). The 1H NMR spectrum includes a
decet centered at δ 0.2 ppm (Figure 1b). Proton peaks
associated with THF and Et2O are absent from the 1H
NMR spectrum, indicating the success of the procedure.
The quintet peak at δ 1.94 ppm is from CD3CN.
Figure 3. Crystal structure of unsolvated NaB3H8. Color code: Na,
violet; B, pink; H, gray.
The spent mercury can be regenerated by washing with
cold water, diluted HCl, and acetone to remove the residual
Na, NaBH4, and any solvent and impurities. Prior to reuse,
the mercury was dried under vacuum with stirring.
Compared with all of the reported methods,7-9 this meth-
od avoids the use of hazardous starting materials such as
diborane and boron trifluororide, thus significantly simplify-
ing the reaction apparatus. Moreover, this method gives a
yield higher than 70%. Commercially available THF BH3
3
enables a safe synthsis in the laboratory. Large-scale synth-
esis can be realized simply by increasing the quantity of the
starting materials. About 4.5 g of unsolvated NaB3H8 can be
Figure 4. Powder XRD patterns: (a) obtained Mg(B3H8)2 2Et2O pow-
3
der; (b) calculation based on single-crystal XRD data2 and NaBr (inset).
˚
obtained using 400 mL of 1.0 M THF BH3 in a single
laboratory reaction.
fixed at 1.08 A, with the H(1) atom still allowed an angle of
3
rotation in the xz plane.2,14
The structure of unsolvated NaB3H8 was solved from
powder XRD data (Figure 2). The structure solution is
detailed in the Supporting Information. It is worth noting
that the elimination of H atoms increases Rwp from 4.26 to
7.74, showing a significant contribution from H atoms to
XRD intensities. The terminal H atoms were refined to
sensible positions without any constraints. The bridging
H atom was refined to have an unreasonably short bond
The compound crystallizes in space group Pmmn. The
compound has double-stranded chains that are linked
through a H bridge between Na and B (Figure 3). The three
B atoms form a triangle, with the two B-B distances being
˚
1.794(4) and 1.820(5) A, similar to the single-crystal XRD
data of other B3H8 compounds.2,14 Because the B3H8 anion is
(14) Calabrese, J. C.; Gaines, D. F.; Hildebrandt, S. J.; Morris, J. H. J.
Am. Chem. Soc. 1976, 98, 5489.
˚
distance with B(2) (0.76 A), so the B(2)-H(1) distance was

Communication
Inorganic Chemistry, Vol. 49, No. 18, 2010 8187
sterically larger than the BH4 anion, the observed Na-B
synthesis is close to 100%, much higher than that for the
reported synthesis (41% yield).2
˚
distances [3.1302(8) and 3.370(4) A] are longer than those
˚
observed in NaBH4 (2.80-3.00 A, depending on the
In summary, we have reported a new route to unsolvated
NaB3H8. This route is safer than previously reported meth-
ods and can be easily scaled up. The structure of NaB3H8 has
been solved from powder XRD data and is reported for
the first time. Because of its high hydrogen content, the un-
solvated NaB3H8 could help to develop novel hydrogen
storage materials. The solvent-free NaB3H8 will facilitate
the exploration of various volatile metal-boron-hydrogen-
containing precursors for metal diboride thin film deposition.
phases15,16). All of the terminal H atoms on the boron
triangle unit form short contacts with the Na cations. There
˚
are four short Na-H contacts at 2.38(1) A and two longer
˚
contacts at 2.48(2) A.
The good solubility of NaB3H8 in ethyl ether enables a
high-yield synthesis of Mg(B3H8)2 2Et2O, which is considered
3
to be an interesting precursor for the MgB2 thin film.2
A metathesis reaction between NaB3H8 and MgBr2 Et2O
3
(2:1 molar ratio) in ethyl ether proceeds rapidly because
NaBr is insoluble in ether. After removal of NaBr through
filtering and slow pumping off of the ether, a colorless
Acknowledgment. This work was funded by the U.S.
Department of Energy, Office of Energy Efficiency and
Renewable Energy, under Contract DE-FC3605GO15062
as part of the DOE Metal Hydride Center of Excellence.
Mg(B3H8)2 2Et2O powder was obtained. A powder XRD pattern
3
of obtained Mg(B3H8)2 2Et2O matches the simulated pattern
3
based on single-crystal XRD data (Figure 4).2 The yield of this
Supporting Information Available: Experimental details, re-
finement details, and structural information. This material is
available free of charge via the Internet at http://pubs.acs.org.
(15) Filinchuk, Y.; Hagemann, H. Eur. J. Inorg. Chem. 2008, 20, 3127.
(16) Filinchuk, Y.; Talyzin, A. V.; Chernyshov, D.; Dmitriev, V. Phys.
Rev. B 2007, 76, 092104.