Film transfer enabled by nanosheet seed layers on arbitrary sacrificial substrates
A. P. Dral, M. Nijland, G. Koster, and J. E. ten Elshof
Citation: APL Materials 3, 056102 (2015); doi: 10.1063/1.4921070 View online: http://dx.doi.org/10.1063/1.4921070
View Table of Contents: http://scitation.aip.org/content/aip/journal/aplmater/3/5?ver=pdfcov Published by the AIP Publishing
Articles you may be interested in
BiFeO3/La0.7Sr0.3MnO3 heterostructures deposited on spark plasma sintered LaAlO3 substrates Appl. Phys. Lett. 104, 082914 (2014); 10.1063/1.4867021
High-throughput synthesis of thermoelectric Ca3Co4O9 films Appl. Phys. Lett. 103, 143123 (2013); 10.1063/1.4824212
Epitaxial growth of ferroelectric SrBi 2 Ta 2 O 9 thin films of mixed (100) and (116) orientation on SrLaGaO 4 (110)
Appl. Phys. Lett. 79, 2961 (2001); 10.1063/1.1415376
Structural and electrical anisotropy of (001)-, (116)-, and (103)-oriented epitaxial SrBi 2 Ta 2 O 9 thin films on SrTiO 3 substrates grown by pulsed laser deposition
J. Appl. Phys. 88, 6658 (2000); 10.1063/1.1321776
Structural characterization of the low-temperature phase in Sr–Bi–Ta–O films Appl. Phys. Lett. 74, 2690 (1999); 10.1063/1.123938
Film transfer enabled by nanosheet seed layers on arbitrary
sacrificial substrates
A. P. Dral, M. Nijland, G. Koster, and J. E. ten Elshofa
MESA+ Institute for Nanotechnology, University of Twente, 7500 AE Enschede, The Netherlands
(Received 13 February 2015; accepted 28 April 2015; published online 11 May 2015)
An approach for film transfer is demonstrated that makes use of seed layers of nanosheets on arbitrary sacrificial substrates. Epitaxial SrTiO3, SrRuO3, and BiFeO3
films were grown on Ca2Nb3O10nanosheet seed layers on phlogopite mica substrates.
Cleavage of the mica substrates enabled film transfer to flexible polyethylene tere-phthalate substrates. Electron backscatter diffraction, X-ray diffraction, and atomic force microscopy confirmed that crystal orientation and film morphology remained intact during transfer. The generic nature of this approach is illustrated by growing films on zinc oxide substrates with a nanosheet seed layer. Film transfer to a flexible substrate was accomplished via acid etching. C 2015 Author(s). All article content,
except where otherwise noted, is licensed under a Creative Commons Attribution 3.0 Unported License.[http://dx.doi.org/10.1063/1.4921070]
A major challenge in the development of flexible electronics is to combine thin film growth that requires temperatures of several hundreds of degrees, with flexible substrate materials that have an organic nature and therefore limited thermal stability. Perovskite films possess useful electronic properties such as ferroelectricity and ferromagnetism. A common deposition technique for these materials is pulsed laser deposition (PLD), and temperatures as low as 350◦C are reported in very
few cases,1but oriented growth of perovskite films generally requires temperatures of a few hundred degrees higher. Such a high deposition temperature makes it impossible to deposit the films directly onto arbitrary flexible substrates. An approach to overcome this problem is to grow the films on a substrate with high thermal stability and transfer them afterwards to the flexible material. Several available methods for the transfer of perovskite films are the laser lift-off process,2,3ion cutting,4the use of water-soluble substrates,5,6and the epifree process.7,8
In this letter, an approach for film transfer is demonstrated that makes use of seed layers of nanosheets on arbitrary sacrificial substrates. Nanosheets are essentially two-dimensional single crys-tals; they have a constant thickness of a few nanometers at most and their lateral size is mostly in the micrometer range,9while their oriented crystal structure enables them to direct epitaxial growth of thin films. At present, usually thick single crystalline substrates are used as templates to direct the growth of oriented films. Drawbacks of such substrates are not only their relatively high prices and size limitations but also the fact that they can only induce epitaxial growth for films with matching lattice parameters. An arbitrary substrate with a seed layer of nanosheets may present a viable alternative to single crystal substrates; this approach has been reported for the oriented growth of ZnO, TiO2,
SrTiO3, LaNiO3, SrRuO3, and Pb(Zr,Ti)O3.10–13Nanosheets covering a wide range of lattice
parame-ters are known and can, in principle, be derived from any layered crystal. When using a seed layer of nanosheets, the substrate underneath does not need to have lattice parameters that match those of the film. Thus, the choice of substrate becomes a tool to engineer the subsequent process. Film transfer is possible by choosing a sacrificial substrate material that can be removed easily.
Films of epitaxial SrTiO3, SrRuO3, and BiFeO3 were grown on a seed layer of Ca2Nb3O10
nanosheets on phlogopite mica (001) substrates. Mica crystals have a layered structure with
aAuthor to whom correspondence should be addressed. Electronic mail:j.e.tenelshof@utwente.nl.
056102-2 Dral et al. APL Mater. 3, 056102 (2015)
interlayer cations that enable cleavage between the layers.14 Little force is required for cleav-age, and when films are grown on top of the mica substrates, the films can be peeled off easily. To make devices that need only limited flexibility, mica substrates could even serve as flexible substrates by themselves. Ca2Nb3O10 nanosheets were prepared and deposited as a seed layer by
Langmuir-Blodgett (LB) deposition15 following the procedure described elsewhere.13,16 In short, the layered parent compound HCa2Nb3O10 was dispersed in demi-water and tetra-n-butyl
ammo-nium hydroxide was added as an exfoliation agent to separate the crystals into discrete nanosheets. Monolayers of these nanosheets were deposited onto freshly cleaved phlogopite mica substrates by LB deposition. After deposition of the first monolayer, tetra-n-butyl ammonium residues on the substrates were decomposed by heating to 600◦C for 30 min in a microwave oven equipped with silicon carbide elements to convert the microwaves into heat. A second monolayer of nanosheets was deposited to improve the overall coverage. The substrates were annealed inside a PLD chamber at 700◦C in vacuum for 3 h prior to film deposition. This led to small irregularities in the nanosheets
as observed with atomic force microscopy (AFM) but did not significantly destruct their crystal lattice since their film orienting function was maintained. A stack of SrTiO3, SrRuO3, and BiFeO3
was grown by PLD at the same temperature with a fluency of 2.0-2.4 J/cm−2 and spot size of
2.5 mm2in a pure oxygen atmosphere. SrTiO
3was deposited at 1 Hz in 0.01 mbar O2for 15 min,
SrRuO3was deposited at 1 Hz in 0.08 mbar O2for 15 min, and BiFeO3was deposited at 0.5 Hz in
0.30 mbar O2for 60 min. This yielded films of approximately 42 nm SrTiO3, 18 nm SrRuO3, and
20 nm BiFeO3. The lattice parameters of the Ca2Nb3O10 nanosheets induced epitaxial growth of
SrTiO3 in its cubic [100]cdirection, SrRuO3 in its orthorhombic [110]o direction, and BiFeO3 in
its rhombohedral [100]r direction. Figure1 shows high resolution scanning electron microscope
(HR-SEM) and electron backscatter diffraction (EBSD) images that confirm epitaxial growth. The relief in the HR-SEM image in Figure1(a)indicates the localization of individual nanosheets and this corresponds with the crystal contours seen in the EBSD inverse pole figure maps of the same area in Figures 1(b) and1(c). The uniform green color in Figure 1(b)confirms that all BiFeO3
crystals were oriented in the [100]r direction out-of-plane. Figure 1(c) confirms the presence of
a single in-plane orientation on top of each nanosheet; though the two colors shown on each nanosheet represent two crystal directions, namely, [a01] and [0a1], both directions have identical lattice spacings. There may be local differences in symmetry within a single crystal and this may explain the mix of two directions on each nanosheet, but the mix may also result from random denotation by the EBSD software because both directions correspond to the same lattice spacing.
Film transfer to a flexible indium tin oxide coated polyethylene terephthalate (PET) substrate was done as shown schematically in Figure2. The mica substrate with the film was partially glued onto an intermediate silicon support with softened QuickStick TM 135 mounting wax at 120◦C
(Figure 2(a)) and then cooled to room temperature. No wax was applied near the edges of the film. The mica substrate was peeled off (Figure 2(b)), leaving the center part of the film behind
FIG. 1. HR-SEM image (HE-SE2 detector) (a) and EBSD inverse pole figure maps with the out-of-plane (b) and in-plane (c) crystal orientations of the surface of stacked SrTiO3, SrRuO3, and BiFeO3grown on a mica substrate with a seed layer
of Ca2Nb3O10nanosheets. Red striped lines indicate the areas where individual nanosheets are located. The legend indicates
FIG. 2. Schematic representation of the experimental method to transfer films from mica substrates to plastic substrates by mechanical cleavage. (a) The top side of the film was placed on a droplet of soft wax on a rigid support (e.g., silicon) at 120◦C and the wax was hardened by cooling to room temperature; (b) the mica substrate was peeled off, leaving the film behind on the wax; (c) the bottom side of the film was placed on a plastic substrate (e.g., PET); (d) the wax was molten by heating to 150◦C and the support was carefully pushed off the plastic substrate.
on the wax. The bottom side of the film was then placed on a PET substrate of larger size (Figure
2(c)), the wax was molten by heating to 150◦C, and the silicon support was removed by carefully sliding it away over the PET substrate (Figure2(d)). The PET substrate was cooled to room temper-ature and residual wax was removed by rinsing with acetone, followed by rinsing with ethanol and drying with nitrogen gas. Pictures of a grown and transferred film are shown as example in Figure S1 in the supplementary material.17Figure3(a) shows an AFM image of the film surface after transfer, confirming that the morphology was preserved. The individual crystals remained smooth with a roughness of 0.8 nm (rms). The crystal orientation of the film remained intact upon transfer, as shown by the X-ray diffraction (XRD) θ-2θ spectrum in Figure 3(b) and the EBSD inverse pole figure maps in Figure 4. In Figure 3(b), the peak at 46.3◦corresponds to the (200)
planes of SrTiO3(a= b = c = 3.9050 Å) and the shoulder at 46.0◦corresponds to the (220) planes
of SrRuO3 (a= 5.5730 Å, b = 5.5381 Å, c = 7.8560 Å). The broadness of the signals may be
attributed to both lattice strain and grain boundaries present in the film. The expected weak reflec-tion of the (200) planes of BiFeO3 (a= b = c = 3.965 Å, α = β = γ = 89.35◦) around 45.3◦ was
systematically undistinguishable for films grown on mica due to overlap of the strong substrate peak (identical films grown on other substrate materials did show clear BiFeO3 reflections). The
peak around 45.0◦corresponds to the (005) planes of mica (phlogopite, a= 5.295 Å, b = 9.135 Å, c= 10.167 Å), indicating that some mica residues were transferred along with the film. This further
FIG. 3. AFM image (a) and XRD θ-2θ spectrum (b) of stacked SrTiO3, SrRuO3, and BiFeO3after transfer to a flexible PET
substrate. The film was originally grown on a mica (001) substrate with a seed layer of Ca2Nb3O10nanosheets. The weak
056102-4 Dral et al. APL Mater. 3, 056102 (2015)
FIG. 4. EBSD inverse pole figure maps with the out-of-plane (a) and in-plane (b) crystal orientations of the surface of stacked SrTiO3, SrRuO3, and BiFeO3after transfer to a flexible PET substrate. The film was originally grown on a mica substrate
with a seed layer of Ca2Nb3O10nanosheets. The legend indicates the crystal orientations corresponding to each color.
confirms that cleavage took place within the substrate, well underneath the film, yielding transfer of all three film layers. The uniform green color in Figure4(a)confirms that all BiFeO3crystals were
still oriented in the [100]rdirection out-of-plane and Figure4(b)confirms the presence of a single
in-plane orientation on top of each nanosheet. The preserved morphology and film orientation upon transfer indicate preservation of functional properties as well, though no properties were analyzed. Repeated bending of the PET substrate with the transferred film did not cause the film to detach.
The ease with which mica crystals could be cleaved makes it an interesting material for sacri-ficial substrates, but in principle such cleavage can occur between any two crystal layers in the substrate. Any stack of residual layers that remains attached to the film may continue to be a “weak spot” that can cause the transferred film to detach again from its new support. This possi-bility should be taken into account when optimizing the process for functional devices. Alternative sacrificial substrate materials are, for example, materials that can be removed by chemical etching. This option was shown successful by using zinc oxide instead of mica; films grown on zinc oxide substrates were transferred to PET substrates by mild etching of zinc oxide with 0.5 vol. % aqueous hydrochloric acid. Experimental details and results of this transfer procedure are shown in the supplementary material.17 When using a procedure that includes etching, the film material must be compatible with the etchant. As for the concept of using seed layers of nanosheets for oriented film growth, the bottleneck is in the degree of perfection of both the seed layer and subsequent film growth. Depositing nanosheets with perfect coverage and preferably with a full control over their in-plane orientation is currently the main challenge in the field. But improved control over the solid state synthesis of parent compounds and the process of exfoliation is also desirable. However, the conceptual value of nanosheets is clear and they exist in a wide variety of compositions and crystal structures, enabling epitaxial growth for many different film materials. As demonstrated in this letter, subsequent film transfer to arbitrary substrates can be achieved by choosing suitable sacrificial substrates.
Thanks to Mark Smithers of the MESA+ Nanolab for his technical assistance with EBSD analyses. This work was financially supported by the Chemical Sciences division of the Netherlands Organization for Scientific Research (NWO-CW) in the framework of the TOP program.
1Y.-H. Chu, C.-W. Liang, S.-J. Lin, K.-S. Liu, and I.-N. Lin,Jpn. J. Appl. Phys., Part 143, 5409–5413 (2004). 2Y. H. Do, W. S. Jung, M. G. Kang, C. Y. Kang, and S. J. Yoon,Sens. Actuators, A200, 51–55 (2013). 3C. H. Lee, S. J. Kim, Y. Oh, M. Y. Kim, Y.-J. Yoon, and H.-S. Lee,J. Appl. Phys.108, 102814 (2010).
4W. Liu, D. Zhan, X. Ma, Z. Song, and S. Feng,J. Vac. Sci. Technol., B: Microelectron. Nanometer Struct.26, 206–208
(2008).
5S.-W. Yeh, H.-L. Huang, D. Gan, and P. Shen,J. Cryst. Growth289, 690–702 (2006). 6P. Taylor, W. Ma, and Y. Chen,Ferroelectrics195, 191–194 (1997).
7V. Depauw, Y. Qiu, K. Van Nieuwenhuysen, I. Gordon, and J. Poortmans,Prog. Photovoltaics: Res. Appl.19, 844–850
(2011).
8I. Mizushima, T. Sato, S. Taniguchi, and Y. Tsunashima,Appl. Phys. Lett.77, 3290–3292 (2000). 9R. Ma and T. Sasaki,Adv. Mater.22, 5082–5104 (2010).
10T. Shibata, T. Ohnishi, I. Sakaguchi, M. Osada, K. Takada, T. Kogure, and T. Sasaki,J. Phys. Chem. C113, 19096–19101
(2009).
11T. Shibata, K. Fukuda, Y. Ebina, T. Kogure, and T. Sasaki,Adv. Mater.20, 231–235 (2008).
13M. Nijland, S. Kumar, R. Lubbers, D. H. A. Blank, G. Rijnders, G. Koster, and J. E. ten Elshof,ACS Appl. Mater. Interfaces
6, 2777–2785 (2014).
14C. Simbrunner, G. Hernandez-Sosa, M. Oehzelt, T. Djuric, I. Salzmann, M. Brinkmann, G. Schwabegger, I. Watzinger, H.
Sitter, and R. Resel,Phys. Rev. B83, 115443 (2011).
15M. Muramatsu, K. Akatsuka, Y. Ebina, K. Wang, T. Sasaki, T. Ishida, K. Miyake, and M. Haga,Langmuir21, 6590–6595
(2005).
16A. P. Dral, D. Dubbink, M. Nijland, J. E. ten Elshof, G. Rijnders, and G. Koster,J. Vis. Exp.(94), e52209 (2014). 17See supplementary material athttp://dx.doi.org/10.1063/1.4921070for pictures of a grown and transferred film and for
