Published: July 08, 2011
r 2011 American Chemical Society 11458 dx.doi.org/10.1021/ja204179g | J. Am. Chem. Soc. 2011, 133, 11458–11461
COMMUNICATION
pubs.acs.org/JACS
Na2CsBe6B5O15: An Alkaline Beryllium Borate as a Deep-UV Nonlinear
Optical Crystal
Shichao Wang and Ning Ye*
Key Laboratory of Optoelectronic Materials Chemistry and Physics, Fujian Institute of Research on the Structure of Matter,
Chinese Academy of Sciences, Fuzhou, Fujian 350002, P. R. China
bS Supporting Information
ABSTRACT: A new deep-UV nonlinear optical crystal,
Na2CsBe6B5O15, has been grown through spontaneous
crystallization with a molten flux based on Na2O�C-
s2O�B2O3. Na2CsBe6B5O15 contains two-dimensional al-
veolate beryllium borate layers [Be2BO5]∞ that are bridged
via planar [BO3] groups. UV�vis diffuse reflectance analysis
on a powder sample of Na2CsBe6B5O15 indicates that the
short-wavelength absorption edge of Na2CsBe6B5O15 is
below 200 nm. Second-harmonic generation (SHG) on
the powder sample wasmeasured with a 1064 nm laser using
the Kurtz and Perry technique, which showed that Na2Cs-
Be6B5O15 is a phase-matchable material, and its measured
SHG coefficient is∼1.17 times as large as the d36 coefficient
of potassium dihydrogen phosphate. The relatively larger
SHG coefficient of Na2CsBe6B5O15 originates from its
shorter distance between the adjacent layers bridged via
the small [BO3] groups.
Deep-UV nonlinear optical (NLO) crystals that can producedeep-UV coherent light with wavelengths below 200 nm
have become increasingly important and are attracting more
attention because of their promising applications in photonic
technologies over the past decade.1�8 In spite of many reports in
the literature,4,6 it remains challenging to obtain practically useful
materials possessing high NLO coefficients and wide UV trans-
parency, especially those covering the deep-UV region. Cur-
rently, the exploration of such crystals is mainly focused on alkali-
and alkaline-earth-metal beryllium borates that exhibit short
transmission cutoff wavelengths in the UV region, such as
ABe2BO3F2 (A = Na, K, Rb, Cs, Tl)
1�5 and M2Be2B2O7
6�8
(M = Sr, Ba). On the basis of the relationship between the
structure and overall NLO properties, there are two approaches
for producing large NLO effects: (1) choosing favorable struc-
tural units and having them aligned parallel9�15 and (2) increas-
ing the density of theNLO structural units. In our previous study,
we reported the structural design and preparation of a series
of new unitary alkali-metal beryllium borates with stoichio-
metries of β-KBe2B3O7, RbBe2B3O7, and γ-KBe2B3O7
16 that con-
sist of two-dimensional (2D) alveolate beryllium borate layers,
[Be2BO5]∞, bridged via flexible one-dimensional (1D) [BO2]∞
chains or rigid, planar [B3O6] groups. The strong connections
provided by the covalent bonds between layers mitigate draw-
backs such as layer growth habit and cleavage, which have limited
the optical applications of KBe2BO3F2 (KBBF) crystals. Mean-
while, the distances between adjacent layers in β-KBe2B3O7
(8.73 Å), RbBe2B3O7 (8.86 Å), and γ-KBe2B3O7 (8.70 Å) are
longer than that in KBBF (6.25 Å), resulting in a lower [BO3]
density in the former. Therefore, the SHG coefficients of
β-KBe2B3O7, RbBe2B3O7, and γ-KBe2B3O7 were found to
be smaller than that of KBBF. In our continued quest for large
SHG effects in these layered compounds containing alveo-
late networks, we tried to construct their structures through the
introduction of smaller [BO3] groups to serve as bridges in
order to increase their [BO3]
3� densities. One of the resulting
alkaline beryllium borates, Na2CsBe6B5O15, was found in this
work to exhibit a shorter interlayer distance, and consequently, a
larger SHG effect was obtained through proper allocation of binary
alkaline cations to fit the cavities between adjacent layers.
Because of the high toxicity of the beryllium oxide powders
upon inhalation, all of the experiments were performed under
sufficient ventilation. Single crystals of Na2CsBe6B5O15 were
grown from a high-temperature solution using Na2O�Cs2O�
B2O3 as a flux. This solution was prepared in a platinum crucible
after melting of a mixture of Na2CO3, Cs2CO3, BeO, and B2O3
having a Na2O/Cs2O/BeO/B2O3 molar ratio of 2:3:4:9. The
mixture (10 g) was heated in a temperature-programmable
electric furnace at 1000 �C until the melt became transparent
and clear. The homogenized melt solution was then cooled
rapidly (50 �C/h) to the initial crystallization temperature
(800 �C). It was further cooled slowly (3 �C/h) to the final
crystallization temperature (700 �C) and then allowed to cool
to room temperature after the furnace was turned off. The flux
attached to the crystal was readily dissolved in water. Trans-
parent and colorless Na2CsBe6B5O15 crystals (Figure S1 in the
Supporting Information) are found to be moisture-stable. The
powder X-ray diffraction (PXRD) patterns of grown crystals of
Na2CsBe6B5O15 and the theoretical simulations from single-
crystal structures match each other very well. The differences
between the peak intensities for the same crystallographic index
in the two patterns are believed to be caused by the preferential
orientation of the powder samples (see Figure S2 in the
Supporting Information). The inductively coupled plasma
elemental analysis of Na2CsBe6B5O15 (Table S1 in the Sup-
porting Information) is consistent with the compositions
determined by single crystal X-ray analysis.
As shown in Figure S3 in the Supporting Information, the
differential thermal analysis (DTA) curves of Na2CsBe6B5O15
exhibit only one endothermic peak starting at 866 �C upon
heating to 1100 �C. The PXRD pattern of the residues reveals
that Na2CsBe6B5O15 decomposes into BeO, suggesting that it
Received: May 6, 2011
11459 dx.doi.org/10.1021/ja204179g |J. Am. Chem. Soc. 2011, 133, 11458–11461
Journal of the American Chemical Society COMMUNICATION
melts incongruently. Therefore, large crystals of Na2CsBe6B5O15
must be grown with a flux and below the decomposition
temperature.
UV�vis diffuse reflectance spectra were collected for Na2Cs-
Be6B5O15 (Figure S4 in the Supporting Information). Absorp-
tion (K/S) data were calculated from the following Kubelka�
Munk function:17,18 F(R) = (1 � R)2/2R = K/S, where R is the
reflectance, K is the absorption, and S is the scattering. In the
plots ofK/S versus E, extrapolating the linear portion of the rising
curve to zero provides the onset of absorption. No obvious
absorption peak in the range 6.22�1.55 eV (corresponding to
wavelengths of 200�800 nm) is observed for Na2CsBe6B5O15,
indicating a high potential of the crystal for deep-UV NLO
applications.
Na2CsBe6B5O15 crystallizes in a monoclinic crystal system
with a chiral space group of C2. The structure as viewed along
the b axis is illustrated in Figure 1. Each B atom is coordinated to
three O atoms to form a planar [BO3] triangle, with B�O bond
lengths ranging from 1.339(10) to 1.419(3) Å and O�B�O
bond angles ranging from 115.6(2) to 123.6(2)�. Each Be atom
is bound to four O atoms to form a distorted [BeO4] tetra-
hedron, with Be�O bond lengths between 1.519(4) and
1.682(6) Å and O�Be�O angles between 103.3(3) and
122.5(3)�. The 2D alveolate beryllium borate [Be2BO5]∞
layers in this structure are similar to those in β-KBe2B3O7,
RbBe2B3O7, and γ-KBe2B3O7, except that the connections
between the adjacent [Be2BO5]∞ layers in this structure are
bridged via [BO3] planar groups that lay in the ac plane and
point in opposite directions.
Two types of tunnels exist in the 3D framework running along
the b axis. The Na+ and Cs+ cations reside in these smaller and
larger tunnels, respectively, where Na+ cations are located in a
seven-coordinate environment with Na�O bond lengths ran-
ging from 2.373(2) to 2.693(2) Å and Cs+ cations are located in
an eight-coordinate environment with Cs�O bond lengths
ranging from 3.230(6) to 3.453(2) Å.
The structural evolution can be illustrated through a com-
parison of the layer structures of KBBF, β-KBe2B3O7,
γ-KBe2B3O7, RbBe2B3O7, and Na2CsBe6B5O15. Although they
all have similar alveolate beryllium borate layers ([Be2BO3F2]∞
or [Be2BO3O2]∞), the orientation periodicity of the O or F
atoms bound to the Be atoms and protruding out of the layers
are different. Since there is only one unique orientation of
[BO3]
3� groups in the [Be2BO3]∞ layers of the aforementioned
compounds, the structural differences among them are clearly
illustrated when [BO3]
3� groups are oriented in the same direc-
tion. (Figures 2 and 3)
To realize the structural design to increase the SHG effects of
these layered compounds containing alveolate networks, an ideal
strategy is to construct a more compact structure through
bridging the 2D layer via smaller [BO3] groups. From this a
hypothesized formula is readily derived as A3Be6B5O15 (A =
alkali metal) by substituting [BO3] for [B3O6] in γ-KBe2B3O7
when the formula is viewed as a sum of three parts, namely,
cations, NLO-active 2D layers, and connectors:
γ-KBe2B3O7 � 3: 3K þ 3½Be2BO3� þ 2½B3O6�
A3Be6B5O15 : 3A þ 3½Be2BO3� þ 2½BO3�
On this basis, a target compoundmay be constructed by carefully
arranging the framework connection and alkali-metal allocation
as follows. The [Be2BO3F2]∞ layer in KBBF is composed of
repeating unit “A” (Figure 2a). The distance between the
adjacent Be atoms that have the same Be�O bond direction
(4.427 Å) is too long to bond the two O atoms from a [BO3]
group (Figure 3a). In the case of β-KBe2B3O7, meanwhile, the
[Be2BO3O2]∞ layer is composed of “AB” repeating units
(Figure 2b). Although the aforementioned distance in this case
(2.689 Å) is acceptable for binding two O atoms of a [BO3]
group (Figure 3b), it is rather undesired to leave a dangling
Be�O bond in the adjacent layer to which it is connected.
Therefore, a reasonable configuration results from binding twoO
atoms of the [BO3] group to one layer and one to the other. Such
configuration is herein realized in the title compound Na2CsBe6-
B5O15, whose [Be2BO3O2]∞ layer is composed of “ABA”
repeating units (Figure 2c). The resulting orientation of the
[BeO4] groups and the Be�Be distance (2.559 Å) enables the
[Be2BO3O2]∞ layers to be bridged via [BO3] groups (Figure 1).
Consequently, the orientations in the [Be2BO3O2]∞ layers result
in the existence of two types of tunnels, one smaller and the other
bigger, between the adjacent [Be2BO3O2]∞ layers. Two types of
cations with different radii (i.e., Na+ andCs+) are thus required to
reside in these two types tunnels, respectively (Figure 1). This is
Figure 1. Crystal structure of Na2CsBe6B5O15 shown along the b axis.
Figure 2. Comparison of [Be2BO3F2]∞ or [Be2BO3O2]∞ layers in (a)
KBBF, (b) β-KBe2B3O7, and (c) Na2CsBe6B5O15.
11460 dx.doi.org/10.1021/ja204179g |J. Am. Chem. Soc. 2011, 133, 11458–11461
Journal of the American Chemical Society COMMUNICATION
likely the reason that we are able to find alveolate beryllium
borate [Be2BO5]∞ layers bridged via smaller planar [BO3]
groups only in binary alkaline beryllium borates and not in
unitary alkaline beryllium borates.
KBBF, β-KBe2B3O7, γ-KBe2B3O7, RbBe2B3O7, and Na2Cs-
Be6B5O15 all have similar structures, and therefore, their non-
linearity originates from the same structure�property relationship.
Although the layer-connector [BO3] groups in Na2CsBe6B5O15 are
NLO-active building blocks, the sum of their microscopic NLO
susceptibilities is close to zero because they are related by an
approximate center of symmetry in this structure. In con-
trast, as a common character of the coplanar beryllium borate
network, the [BO3]
3� groups in the [Be2BO3]∞ layers of each
compound have the same orientation, which is the optimal
arrangement to yield large overall NLO properties. However,
the distances between adjacent layers in β-KBe2B3O7 (8.73 Å),
γ-KBe2B3O7 (8.70 Å), RbBe2B3O7 (8.86 Å), and KBBF (6.25 Å)
are longer than that in Na2CsBe6B5O15 (6.23 Å), yielding a higher
[BO3]
3� density for the latter. Therefore, the SHG coefficients of
Na2CsBe6B5O15 are predicted to be approximately as large as
those of KBBF and therefore larger than those of β-KBe2B3O7, γ-
KBe2B3O7, and RbBe2B3O7. Furthermore, the strong connections
provided by the covalent bonds between layersmitigate drawbacks
such as layer growth habit and cleavage that otherwise limit the
optical applications of KBBF crystals.
SHG signals as a function of particle size obtained from the
measurements made on ground crystals of Na2CsBe6B5O15 with
a Q-switched Nd:YAG laser of wavelength 1064 nm are shown in
Figure 4. The results are consistent with phase-matching beha-
vior according to the rule proposed by Kurtz and Perry.19 A
LiB3O5 (LBO) sample was selected as a reference to ensure
accurate measurements because it is the same type of optical
biaxial crystal as the sample to be measured. The second-
harmonic signal for Na2CsBe6B5O15l was found to be 0.55 times
that for LBO. Such a value is proportional to the square of the
nonlinear deff coefficient. Since the reported deff coefficient for
LBO is 0.832 pm/V [2.133 times the d36 coefficient of potassium
dihydrogen phosphate (KDP)],20 the derived deff coefficient for
Na2CsBe6B5O15 is 0.46 pm/V, which is ∼1.17 times as large as
that of d36(KDP).
The above argumentation on structure�property correlations
is in good agreement with the SHG measurements. According to
the anionic group theory, the nonlinearity of Na2CsBe6B5O15
relative to that of KBBF [1.12� d36(KDP)] is determined by the
density of [BO3]
3� groups in the [Be2BO3]∞ layers within a
unit cell, based on the same arrangement of the NLO-active
[BO3]
3� groups. The density of the [BO3]
3� groups is 0.00921
per unit volume inNa2CsBe6B5O15, which is comparable to that in
KBBF (0.00946). Therefore, the nonlinearity of Na2CsBe6B5O15
is calculated to be 97.4% of that of KBBF. These results agree
approximately with the experimental value for Na2CsBe6B5O15,
which is 1.05 times that for KBBF. The correlations between the
NLO effects and structural characteristics are summarized in
Table 1. As shown by this table, the nonlinearity of this type of
borate containing 2D alveolate beryllium borate layers is approxi-
mately inversely proportional to the interlayer distance. Among
these compounds, Na2CsBe6B5O15 possesses the largest NLO
coefficient as a result of its shortest interlayer distance, which is due
to the compact layer structure bridged via small [BO3] groups.
In summary, a new binary alkaline beryllium borate, Na2Cs-
Be6B5O15, has been synthesized through spontaneous crystal-
lization with a molten flux based on Na2O�Cs2O�B2O3.
Na2CsBe6B5O15 consists of 2D alveolate beryllium borate
[Be2BO5]∞ layers bridged by planar [BO3] groups. UV�vis
Figure 4. SHGmeasurements on ground Na2CsBe6B5O15 crystals (b)
with LBO (O) as a reference.
Figure 3. [Be2BO3F2]∞ or [Be2BO3O2]∞ layers in the structures of (a)
KBBF, (b) β-KBe2B3O7, and (c, d) Na2CsBe6B5O15.
Table 1. NLO Effects and Structural Characteristics of Alkaline
Beryllium Borates
crystals
space
group
interlayer
bridging
interlayer distances
(Å)
SHG
coefficienta
KBBFb R32 K+�F� 6.25 1.12
γ-KBe2B3O7
c P21 [B3O6] 8.70 0.68
β-KBe2B3O7
c Pmn21 [BO2]∞ 8.73 0.75
RbBe2B3O7
c Pmn21 [BO2]∞ 8.86 0.79
Na2CsBe6B5O15
d C2 [BO3] 6.23 1.17
a In multiples of d36(KDP).
bData from ref 2. cData from ref 16.
dThis work.
11461 dx.doi.org/10.1021/ja204179g |J. Am. Chem. Soc. 2011, 133, 11458–11461
Journal of the American Chemical Society COMMUNICATION
diffuse reflectance analysis of a powder sample of Na2CsBe6-
B5O15 indicates that the short-wavelength absorption edge of
Na2CsBe6B5O15 is below 200 nm. Second-harmonic generation
in the powder sample was measured using the Kurtz and Perry
technique, which suggested that Na2CsBe6B5O15 is phase-
matchable, and its measured SHG coefficient is ∼1.17 times
as large as d36(KDP). The relatively large SHG coefficient of
Na2CsBe6B5O15 likely originates from the short distance be-
tween the adjacent layers bridged by the small [BO3] groups.
’ASSOCIATED CONTENT
bS Supporting Information. DTA traces, crystal pictures,
PXRD patterns, diffuse reflectance absorption curves, and crystal
data (CIF). This material is available free of charge via the
Internet at http://pubs.acs.org.
’AUTHOR INFORMATION
Corresponding Author
nye@fjirsm.ac.cn
’ACKNOWLEDGMENT
This work was supported by the National Natural Science
Foundation of China (50872132 and 90922035). The authors
thank Prof. R. G. Xiong and X. Y. Chen for the measurement of
the temperature-dependence of the SHG.
’REFERENCES
(1) Mei, L. F.; Wang, Y. B.; Chen, C. T.Mater. Res. Bull. 1994, 29, 81.
(2) Mei, L.; Huang, X.; Wang, Y.; Wu, Q.; Wu, B.; Chen, C. Z.
Kristallogr. 1995, 210, 93.
(3) McMillen, C. D.; Kolis, J. W. J. Cryst. Growth 2008, 310, 2033.
(4) Chen, C. T.; Luo, S. Y.; Wang, X. Y.; Wang, G. L.; Wen, X. H.;
Wu, H. X.; Zhang, X.; Xu, Z. Y. J. Opt. Soc. Am. B 2009, 26, 1519.
(5) McMillen, C. D.; Hu, J.; VanDerveer, D.; Kolis, J. W. Acta
Crystallogr., Sect. B 2009, 65, 445.
(6) Chen, C. T.; Wang, Y. B.; Wu, B. C.; Wu, K. C.; Zeng, W. L.; Yu,
L. H. Nature 1995, 373, 322.
(7) Qi, H.; Chen, C. T. Inorg. Chem. Commun. 2001, 4, 565.
(8) Qi, H.; Chen, C. T. Chem. Lett. 2001, 352.
(9) Pan, S.; Smit, J. P.; Watkins, B.; Marvel, M. R.; Stern, C. L.;
Poeppelmeier, K. R. J. Am. Chem. Soc. 2006, 128, 11631.
(10) Chang, H. Y.; Kim, S. H.; Halasyamani, P. S.; Ok, K. M. J. Am.
Chem. Soc. 2009, 131, 2426.
(11) Chang, H. Y.; Kim, S. H.; Ok, K. M.; Halasyamani, P. S. J. Am.
Chem. Soc. 2009, 131, 6865.
(12) Chang, H. Y.; Kim, S. H.; Ok, K. M.; Halasyamani, P. S. Chem.
Mater. 2009, 21, 1654.
(13) Sun, C. F.; Hu, C. L.; Xu, X.; Ling, J. B.; Hu, T.; Kong, F.; Long,
X. F.; Mao, J. G. J. Am. Chem. Soc. 2009, 131, 9486.
(14) Halasyamani, P. S.; Poeppelmeier, K. R. Chem. Mater. 1998,
10, 2753.
(15) Wu, H. P.; Pan, S. L.; Poeppelmeier, K. R.; Li, H. Y.; Jia, D. Z.;
Chen, Z. H.; Fan, X. Y.; Yang, Y.; Rondinelli, J. M.; Luo, H. S. J. Am.
Chem. Soc. 2011, 133, 7786.
(16) Wang, S. C.; Ye, N.; Li, W.; Zhao, D. J. Am. Chem. Soc. 2010,
132, 8779.
(17) Kubelka, P.; Munk, F. Z. Tech. Phys 1931, 12, 593.
(18) Tauc, J. Mater. Res. Bull. 1970, 5, 721.
(19) Kurtz, S. K.; Perry, T. T. J. Appl. Phys. 1968, 39, 3798.
(20) Roberts, D. A. IEEE J. Quantum Electron. 1992, 28, 2057.