Modeling and Numerical Simulation of Material Science, 2013, 3, 4-8
Published Online January 2013 (http://www.SciRP.org/journal/mnsms)
Controllable Growth of Ni Nanocrystals Embedded in
BaTiO3/SrTiO3 Superlattices
Zhengwei Xiong, Weidong Wu
Science and Technology on Plasma Physics Laboratory Research Center of Laser Fusion, China Academy of Engineering Physics
Mianyan, China
Received 2012
ABSTRACT
BaT i O3/SrTiO3 superlattices with embedded Ni nanocrystals (NCs) have been grown on SrTiO3 (001) substrate using
laser molecular beam epitaxy (L-MBE). In situ reflection high-energy electron diffraction (RHEED) was employed to
investigate the process of lattice strain in the self-organization of Ni NCs and the epitaxial growth of BaTiO3/SrTiO3
superlattices. The results indicated that the strain from large lattice mismatch drove the self-organization of Ni NCs.
Als o , the layer-by-layer growth of BaTiO3 /SrTiO3 superlattices and the island growth of Ni NCs were controllable ac-
curately. The fine alternation of the two processes would provide a possible route to engineer controllably the nano-
composite microstructure.
Keywords: Nanocrystal; Superlattices; Self-organization
1. Introduction
Oxide artificial superlattices, especially (001) oriented
BaT i O3/SrTiO3 superlattices (BTO/STO SLs), have been
attracted more attention because their dielectricand fer-
roelectric properties can be enhanced by controlling the
lattice strain along the polarized direction [1-3]. The di-
electric matrix with the embedded metal NCs has widely
potential application in nonlinear optical and electronic
device [4, 5]. Therefore, several methods have been con-
sidered to fabricate the BTO and STO-based material,
including sol-gel [6], reactive evaporation [7],
rf-sputtering [8] and laser ablation methods [9]. However,
the controllable fabrication of nanostructure remains the
daunting challenge for many deposition methods. There
is an excellent method referred to as self-or ga nized
growt h, in which the strain would drive the
three -dimensional (3D) island to form in the lattice mis-
matched growth process [10, 11]. Especially for the fa-
brication of quantum dot (QD) structures, the
self-or ga nized g rowth is greatly succe ss ful in semicon-
ductor devices, such as InGaAs on GaAs [10], AlN on
GaN [12]. Wu et al. have successfully fabricated
Co:BTO and Ni:BTO composite film using
self-organized method [13-15].
In this paper, the laser molecular beam epitaxy was
used to embed the Ni NCs in the epitaxial films of Ba-
TiO3/SrTiO3 superlattices (Ni NCs:BTO/STO SLs). The
fabrication of the nanocomposite system is interesting
and significant for both fundamental and application as-
pects. Such composite films offer the combination of
ferroelectric and ferromagnetic characteristics. This
magnetoelectric interaction will be particularly strong in
systems which simultaneously exhibit both ferromagnet-
ism and ferroelectricity, i.e., in magnetoelectric multi-
ferroics [16-18].
2. Experimental
The Ni NCs:BTO/STO SLs films were grown on STO
(001) substrate with STO buffer layer by L-MBE. The
acceleration voltage of RHEED electron gun was 25kev
and the grazing incidence angle was about 1-3°. The
growth process was performed in an ultrahigh -vac uu m
(UHV) system and the background vacuum was ～2
×10-6 Pa. The schematic diagram of Ni NCs:BTO/STO
SLs film was shown in Fig. 1. The experimental parame-
ters were listed in detail in Table 1. Before the deposition,
the substrate STO (001) wa s in situ annealed about 2
hours in order to remove surface contamination. At 10 Pa
oxygen pressure, the sample was annealed about 20 mi-
nutes after accomplishing the BTO layer. The samples
wer e characterized by XRD with Cu ka radiation and
high -resolution transmission electron microscopy
(HRTEM).
3. Results and discussion
Fig.2 sho ws the oscillation of the RHEED intensity and

Z. XIONG, W. WU
Figure 1. A schematic diagram of Ni NCs:BTO/STO SLs
film. One deposition cycle involves a number of pulses on
the Ni target in ultra-high vacuum, followed by the epitaxial
growth of BTO/STO SLs.
Table 1. Experimental parameters for Ni NCs:BTO/STO
SLs films fabrication
Background vacuum ~2×10-6 Pa
Working vacuum ~2×10-5 Pa
Target
BaTiO3 purity >99.99%
SrTiO3 purity >99.99%
Ni purity >99.99%
Substrate SrTiO3(001)
Laser pulse
frequency
1Hz for SrTiO3 dep os i tion
1Hz for BaTiO3 dep os ition
2Hz for Ni deposition
The distance between
the target and
substrate
5 cm
Annealing condition 650 ℃ 20 min
10Pa O2 press u re
diffraction patterns of BTO deposited on STO surface.
The diffraction sharp streaks indicate a smooth surface
and good crystalline quality [19]. Compared with the
unannealed STO substrate [Fig. 2(a)], the diffraction
streaks become more obvious after in situ anneal at
650℃ [Fig. 2(b)], indicated that the crystallization of
STO surface is improved. During the growth processes,
the characteristic streaks of BTO became more striking
shown in Fig.1(c) and Fig.1 (d). The prefect epitaxial
layers of BTO were formed on the STO s urface. In this
case, the intensity oscillations are obviously observed. It
is concluded that the deposited BTO layers are grown
with la yer -by-layer growth mode.
Figure 2. Variation of RHEED intensity oscillation and
patterns along the [100] azimuth of BTO on STO layer: (a)
unannealed STO substrate; (b) annealed STO surface; (c)
deposited 60s BTO; (d) deposited 120s BTO
Figure 3. Transition of RHEED patterns during the self-organized
embe dded Ni process: (a) BTO(001) surface; (b) 100 pulses Ni; (c)
500 pulses Ni; (d) 1000 pulses Ni
As indicated in Fig. 3, the self-o rga nize d Ni NC s along
the [100] azimuth were deposited the new surface of
BTO (001). It is obvious that the streaky patterns of BTO
disappeared gradual ly wit h the increasing Ni pulses,
while the spots of Ni NCs become more striking. Addi-
tionally, due to the large lattice mismatch between BTO
and Ni (>10%), that results in an overall decrease in
RHEED intensity. Thus, this case is considered as the
formation of 3D islands. The surface-lattice parameter,
which is proportional to the inverse of the distance be-
tween the different diffraction spots or streaks, can be
directly measured from the RHEED patterns [19]. By
measuring the vertical spacing and horizontal spacing for
the dominant Ni diffraction spots, the lattice constant of
Ni in cubic structure was c onf ir med to 0.354nm in real
space. This result approaches to the lattice constant of
bulk metal Ni (0.352 nm) in cubic structure [20]. There-
for e, it is demonstrated that the deposited Ni maintains
metallic property.
5

Z. XIONG, W. WU
Figure 4. Variation of RHEED intensity oscillation during STO
layer deposited on the Ni NCs. The insets are the RHEED patterns
recorded at 10s, 40s and 120s, respective ly.
After the completion of Ni NCs, the next layer is STO.
As shown in Fig. 4, at the initial stage of STO deposition,
the RHEED intensity decreases instantaneously because
of the increasing random distribution of the incoming
STO atoms on the surface. Then, the deposited STO
atoms and the formed clusters are rearranged on the sur-
face. That induces the increasing RHEED oscillated in-
tensity. For the insets Fig. 4(a)-(d), the dominant Ni spots
change to the characteristic streaks of STO with the in-
creasing STO pulses. Meanwhile, the RHEED intensity
oscillations are observed again. Those results demon-
strate that the STO layer covers the geometrically irregu-
lar surface provided self-orga nize d Ni NCs, and is gro w n
wit h layer-by-layer growth mode. The accurate thickness
of every individual layer is controlled by RHEED inten-
sity oscillation which is extremely sensitive to the pres-
ence of surface atoms in the top crystalline layer [21, 22].
After the laser ended for some time, the oscillated inten-
sity increased slowly, the lattice relaxation tended to
smooth for the surface roughness. Therefore, the Ni NCs
are embedded successfully in BTO/STO SLs by the al-
ternation of 3D island Ni NCs and perfect epitaxial
BTO/STO SLs. Importantly, the embedded Ni NCs can
not disturb the epitaxial growth of BTO/STO SLs.
In order to explain the formation mecha nis m of 3D
island Ni NCs, the changes in the (aNi-aBTO)/aBTO as a
function of the pulses, using the spot spacing for Ni NCs
and the streak spacing for BTO, were calculated (Fig. 5).
Initially, the Ni NCs retain an in-plane lattice parameter
identical to that of the BTO layer, and then recover to the
value of bulk Ni rapidly. Owing to the large lattice mis-
match between Ni and BTO (>10%), Ni NCs were sub-
jected to considerable in-plane tensile strain. Simulta-
neously, this in-plane straining would shorten the
out-of-plane lattice parameter of Ni NCs, leading to the
structural transition from cubic to tetragonal. With the
deposition being continued, the increasing strain energy
was relieved by the transformation from 2D to 3D islands
as the thickness exceeded the critical thickness hc, in
which the lattice was relaxed laterally to the unstrained
value, and the critical thickness is in inverse proportional
to the lattice mismatch, according to the Mat-
the ws -Blakeslee formula. The strain from lattice mis-
match acts as a source of driving force for the
self-organization of Ni NCs. Therefore such metal NCs
growth route is feasible especially in the large lattice
mismatch system.
Figure 5. The in-plane lattice relaxation during the
self-organization of Ni NCs. Inset an expanded view of the first 90
pulses
Figure 6. The cross-sectional HRTEM image of Ni NCs:BTO/STO
SLs nanocomposition film
Fig. 6 is the cross-section HRTEM image of Ni
NCs:BTO/STO SLs composition films. The irregular
interface of strained Ni NC layers is alternated with 7
times. We estimate the sizes of Ni NCs about 5nm. As
shown in Fig. 6, the perfect laye r-by-laye r epitaxia
growth of STO and BTO layers is performed. The sepa-
rate BTO layer and STO layer have the uniform thick-
ness of 8 nm and 10 nm accordingly. The thickness of
6

Z. XIONG, W. WU
every separate layer is accurately controlled by RHEED
intensity oscillation which is extremely sensitive to the
presence of surface atoms in the top crystalline layer [3].
The individual BTO layer and STO layer have the uni-
form thickness of 14nm and 21nm, respectively. Ac-
cording to the results of RHEED monitored, the growth
rate of STO and BTO are estimated to be 0.037 ML/s and
0.043 ML/s, where 1 monolayer (ML) corresponds to a
layer thickness of 0.781 nm (STO) and 0.806 nm (BTO),
the double value of the c-axis lattice constant of bulk
STO (0.3905 nm) and BTO (0.4029 nm) [5], respectively.
So it is concluded that one period of RHEED intensity
oscillation corresponds to the growth of two STO unit
cells and two BTO unit cells, respectively. We achieve
atomic-level control for the growth of the nanocomposite
fil m.
Figure 7. XRD θ-2θ scans of the composite films on STO (001)
substrate; the inset is an expand view around the BTO (002) peak
Fig. 7 shows the X-ray diffraction θ-2θ scans of the
composite films grown on STO (001) substrate. The
characteristic peaks of BTO and STO show that the epi-
taxial growth is satisfying along basically c-axis with the
lower diffraction angles. Using STO single-crystal sub-
strate, we have obtained very high quality epitaxial films.
Magnifying the BTO (002) peak for the inset, a weak
peak is observed beside the leading peak. We suppose
that the weak peak is Ni (111).The crystallographic
orientation with Ni (111) takes priority of the fcc metal
structure. The appearance of the Ni (111) is supposed to
utmost compressed plane in order to get minimum sur-
face energy and remain metallic property, so as to give
the flat surface to the vicinal separation layer. This way
reduces extremely the effect of the strained distortion by
the large lattice mismatch between the different interfac-
es. The results of the XRD characterization are in good
agree with RHEED analysis.
4. Conclusion
In summary, Ni NCs:BaTiO3/SrTiO3 superlattices were
fabricated by L-MBE and controlled accurately by in situ
monitoring RHEED. With the self-or ga nized method, the
alternate growth of Ni NCs was embedded successfully
in BaTiO3/SrTiO3 superlattices and could not disturb the
epitaxial growth of BaTiO3/SrTiO3 superlattices, in good
agree with the anal ysi s of XRD and HRTEM. The results
wer e concluded that the layer-by-layer growth of Ba-
TiO3/SrTiO3 superlattices and the t hree -dimensional isl-
and growth of Ni NCs. The strain in the composite films
from lattice mismatch acts as a source of driving force
for the self-organization of Ni NCs. Therefore such
self-or ga nized metal NCs growth route is propitious to
fabricate three -di mensio nal quantum dot devices.
REFERENCES
[1] H. N. Lee, H. M. Christen, M. F. Chisholm, C. M. Rou-
leau & D. H. Lownders, “Strong polarization enhance-
ment in asymmetric three-component ferroelectric super-
lattices”, Nature, Vol. 433, 2005, pp.
395-399.
[2] O. Nakagawara, T. Shimuta, T. Makino, S. Arai, H. Ta-
bata and T. Kawai, “Epitaxial growth and dielectric prop-
erties of (111) oriented BaTiO3/SrTiO3 superlattices by
pulsed-laser deposition”, Appl. Phys. Lett, Vol. 77, No. 20,
2000, pp. 3257-3260. doi:
doi:10.1038/nature03261
10.1063/1.1324985
[3] H. Tabata, H. Tanaka, and T. Kawai, “Formation of ar-
tificial BaTiO3/SrTiO3 superlattices using pulsed laser
deposition and their dielectric properties”, Appl. Phys.
Lett, Vol. 65, No. 15, 1994, pp.
1970-1973.
[4] A. Nahata, R. A. Linke, T. Ishi, and K. Ohashi, “En-
hanced nonlinear optical conversion from a periodically
nanostructure metal film”, Opt. Lett, Vol. 28, No. 6, 2003,
pp. 423-425.
doi:10.1063/1.112837
[5] T. Shimuta, O. Nakagawara, T. Makino, and S. Arai,
“Enhancement of remanent polarization in epitaxial Ba-
TiO3/SrTiO3 superlattices with “asymmetric” structure”, J.
Appl. Phys, Vol. 91, No. 4, 2002, pp. 2290-2294.
doi: 10.1364/OL.28.000423
[6] T. Hayashi and T. Tanaka, “Preparation and Dielectric
Properties of SrTiO3/BaTiO3 Multilayer Thin Films by
Sol-Gel Method”, Jpn. J. Appl. Phys, Vol. 34, 1995, pp.
5100-5104.
doi:
10.1063/1.1434547
[7] T. Kuroiwa, Y. Tsunemine, T. Horikawa, T. Makita, J.
Tanimura, N. Mikami and K. Sato, “Dielectric Properties
of (BaxSr1-x)TiO3 Thin Films Prepared by RF Sputtering
for Dynamic Random Access Memory Application”, Jpn.
J. Appl. Phys, Vol. 33, No. 9, 1994, pp.
5187-5191.
doi:10.1143/JJAP.34.5100
[8] K. Ijima, T. Terashima, Y. Bando, K. Kamigaki and
H.Terauch i, “Atomic layer growth of oxide thin films
with perovskite-type structure by reactive evaporation”, J.
Appl. Phys, Vol. 72, No. 7, 1992, pp.
2840-2845.
doi:10.1143/JJAP.33.5187
[9] B. D. Qu, M. Evstigneev, D. J. Johnson and R. H. Prince,
“Dielectric properties of BaTi O 3/SrTiO3 multilayered thin
doi:10.1063/1.351536
7

Z. XIONG, W. WU
films prepared by pulsed laser deposition”, Appl. Phys.
Lett, Vol. 72, No. 11, 1998, pp. 1394-1397.
[10] H. Drexler, D. Leonard, W. Hansen, J. P. Kotthaus, and P.
M.Petroff, “Spectroscopy of Quantum Levels in
Charge-Tunable InGaAs Quantum Dots”, Phys. Rev. Lett,
Vol. 73, No. 16, 1994, pp.
2252-2255.
doi:1063/1.121066
[11] J. Coraux, V. Favre-Nicolin, H. Renevier, M. G. Proietti,
B. Amstatt, E. Bellet-Amalric and B. Daudin, “Quantita-
tive structural characterization of GaN quantum dot ripe-
ningbusing reflection high-energy electron diffraction”,
Appl. Phys. Lett, Vol. 101, No. 5, 2007, pp. 1-3.
doi:10.1103/PhysRevLett.73.2252
[12] J. Coraux, H. Renevier, V. Favre-Nicolin, G. Renaud, and
B. Daudin, “In situ resonant x-ray study of vertical corre-
lation and capping effects during GaN/AlN quantum dot
growth”, J. Appl. Phys, Vol. 88, No. 15, 2007, pp.1-3.
doi:
10.1063/1.2422902
[13] W. Wu, Y. He, F. Wang, Z. Chen, Y. Tang and W. Sun,
“Preparation and characterization of Co-BaTiO3 na-
no-composite films by the pulsed laser deposition”, J.
Crystal Growth, Vol. 289, No. 1, 2006, pp. 408-413.
doi:
10.1063/1.2192572
[14] F. F. Ge, X. M. Wang, L. H. Cao, et al., “Self-Organized
Ni Nanocrystal Embedded in BaTiO3 Epitaxial Film”,
Nano Res. Lett, Vol. 5, No. 5, 2010, pp. 834-838.
doi:
10.1016/j.jcrysgro.2005.11.041
[15] Z. Xiong, W. Sun, X. Wang, F. Jiang, W.Wu, “Dielectric
enhancement of BaTiO 3/SrTiO3 superlattices with em-
bedded Ni nanocrystals”, J. Alloys Compd, Vol. 513,
2012, pp. 300-303.
doi:
10.1007/s11671-010-9570-9
[16] Y. H. Lin, S. Zhang, C. Deng, Y. Zhang, X. Wang, C. W.
Nan, “Magnetic behavior and thickness dependence in
Co-doped BaTiO3 thin films”, Appl. Phys. Lett, Vol. 92,
No. 11, 2008, pp. 1-3.
doi: 10.1016/j.jallcom.2011.09.103
[17] Y. H. Lin, J. Yuan, S. Zhang, et al., “Multiferroic beha-
vior observed in highly orientated Mn-doped BaTi O 3 thin
films”, Appl. Phys. Lett, Vol. 96, No. 3, 2009, pp. 1-3.
doi:10.1063/1.2898525
[18] S. Geprägs, A. Brandlmaier, M. Opel, R. Gross, and S. T.
B. Goennenwein, “Electric field controlled manipulation
of the magnetization in Ni/BaTiO3 hybrid structures”,
Appl. Phys. Lett, Vol. 96, No. 14, 2010, pp. 1-3.
doi:
10.1063/1.3182793
[19] K. Godo, J. H. Chang, H. Makino, T. Takai, T. Hanada
and T. Yao, T, Sascao, and T. Goto, “Formation
processes of CdTe quantum dots on ZnTe substrates stu-
died by reflection high-energy electron diffraction and
photoluminescence”, J. Appl. Phys, Vol. 92, No. 9, 2002,
pp. 5490-5493.
doi:
10.1063/1.3377923
[20] S. M. Foiles, M. I. Baskes, and M. S. Daw, “Embed-
ded-atom-method functions for the fcc metals Cu, Ag, Au,
Ni, Pd, Pt, and their alloys”, Phys. Rev. B, Vol. 33, No. 12,
1986, pp. 7983-7991.
doi: 10.1063/1.1513888
[21] B. Shin and M. J. Aziz, “Modeling RHEED intensity
oscillations in multilayer epitaxy: Determination of the
Ehrlich-Schwoebel barrier in Ge(001) homoepitaxy”,
Phys. Rev. B, Vol. 76, No. 16, 2007, pp. 1-12.
doi: 10.1103/PhysRevB.33.7983
[22] M. Dabrowska-Szata, “Analysis of RHEED pattern from
semiconductor surfaces”, Mater. Chem Phys, Vol. 81,
No. 2, 2003, pp. 257-259.
doi:
10.1103/PhysRevB.76.165408
doi:
10.1016/S0254-0584(02)00569-2
8