On the dielectric and optical properties of surface-anchored metal-organic frameworks: A study on epitaxially grown thin films Engelbert Redel, Zhengbang Wang, Stefan Walheim, Jinxuan Liu, Hartmut Gliemann et al. Citation: Appl. Phys. Lett. 103, 091903 (2013); doi: 10.1063/1.4819836 View online: http://dx.doi.org/10.1063/1.4819836 View Table of Contents: http://apl.aip.org/resource/1/APPLAB/v103/i9 Published by the AIP Publishing LLC.
Additional information on Appl. Phys. Lett. Journal Homepage: http://apl.aip.org/ Journal Information: http://apl.aip.org/about/about_the_journal Top downloads: http://apl.aip.org/features/most_downloaded Information for Authors: http://apl.aip.org/authors
APPLIED PHYSICS LETTERS 103, 091903 (2013)
On the dielectric and optical properties of surface-anchored metal-organic frameworks: A study on epitaxially grown thin films Engelbert Redel,1,2,a) Zhengbang Wang,1 Stefan Walheim,2 Jinxuan Liu,1 €ll1,a) Hartmut Gliemann,1 and Christof Wo
1 Institute of Functional Interfaces (IFG), Karlsruhe Institute of Technology (KIT), Hermann-von-Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany 2 Institute for Nanotechnology (INT), Karlsruhe Institute of Technology (KIT), Hermann-von-Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany
(Received 9 May 2013; accepted 12 August 2013; published online 30 August 2013) We determine the optical constants of two highly porous, crystalline metal-organic frameworks (MOFs). Since it is problematic to determine the optical constants for the standard powder modification of these porous solids, we instead use surface-anchored metal-organic frameworks (SURMOFs). These MOF thin films are grown using liquid phase epitaxy (LPE) on modified silicon substrates. The produced SURMOF thin films exhibit good optical properties; these porous coatings are smooth as well as crack-free, they do not scatter visible light, and they have a homogenous interference color over the entire sample. Therefore, spectroscopic ellipsometry (SE) can be used in a straightforward fashion to determine the corresponding SURMOF optical properties. After careful removal of the solvent molecules used in the fabrication process as well as the residual water adsorbed in the voids of this highly porous solid, we determine an optical constant of n ¼ 1.39 at a wavelength of 750 nm for HKUST-1 (stands for Hong Kong University of Science and Technology-1; and was first discovered there) or [Cu3(BTC)2]. After exposing these SURMOF thin films to moisture/EtOH atmosphere, the refractive index (n) increases to n ¼ 1.55–1.6. This dependence of the optical properties on water/EtOH adsorption demonstrates C 2013 AIP Publishing LLC. the potential of such SURMOF materials for optical sensing. V [http://dx.doi.org/10.1063/1.4819836] Materials with good mechanical properties and low k dielectric constants are of paramount interest for the next generation of electronics, since low k materials are required for the needed increases in clock frequency. To date, special types of organic polymers have been investigated with regard to these applications.1 However, it is very difficult to achieve dielectric constants below k ¼ 2 with conventional polymers.2 Here, we focus on another highly tunable class of materials, metal-organic frameworks (MOFs), which are also referred to as porous coordination polymers (PCPs). Metalorganic frameworks are highly porous hybrid materials consisting of organic linkers connected to inorganic metal (or metal/oxo) clusters.3–5 Due to their crystalline, highly ordered, and porous structures, this class of solids exhibits a number of highly interesting properties. The size of the pores within MOFs has been shown to be highly adjustable6 and pore diameters up to 10 nm7 have recently been realized, yielding exceptionally low densities. Because of this very low mass density, the static dielectric constants k should be very small and drop to values far below 2.8 In previous theoretical work, the static dielectric properties of MOFs have been determined on the basis of electron structure calculations. In the case of HKUST-1 or [Cu3(BTC)2],8 which is one of the most investigated MOF materials and consists of Cu ions and 1,3,5-benzenetricarboxylate (BTC), a value of k ¼ 1.7 has been obtained which corresponds to a refractive index of n ¼ 1.3.8 Unfortunately, only a few experimental a)
Authors to whom correspondence should be addressed. Electronic addresses:
[email protected] and
[email protected]
0003-6951/2013/103(9)/091903/5/$30.00
reports on the dielectric values for MOF materials have been published. For example, the dielectric constant for polycrystalline ZIF 8 films deposited on Si wafers has been determined using impedance analysis, and a value of k ¼ 2.33 was found at frequencies of 100 kHz.9 Powders are the most common modification of MOFs, and standard solvothermal bulk synthesis yields powders having a wide size distribution in the lm range. The fabrication of millimeter-sized crystals has been reported only in exceptional cases. Determining the solid state properties of this exciting class of porous solids by conventional methods is greatly simplified if high quality thin films of MOFs grown by liquid phase epitaxy (LPE)10 can be used instead of powders; such epitaxially grown MOF layers are referred to here as surface-anchored metal-organic frameworks (SURMOFs). The suitability of SURMOFs for investigating solid state elastic and mechanical properties has been recently demonstrated. The Young’s modulus of HKUST-1 SURMOF (9.3 GPa) has been determined directly using indentation studies of lm thick, epitaxially grown SURMOFs.11 Homogenous thin films are much better suited than powders for determining the refractive index (n), since spectroscopic ellipsometry (SE) can be applied to directly measure the frequency-dependent optical constant. While in principle it is also possible to determine the refractive index (n) for powders samples,12,13 corresponding results for MOF powders have not yet been reported. Determining the optical properties of SURMOFs is not only interesting with regard to the dielectric constant but may also open the path for upcoming optical applications of SURMOFs.
103, 091903-1
C 2013 AIP Publishing LLC V
091903-2
Redel et al.
It is quite obvious that the optical properties of MOFs change when guest molecules are loaded into the pores of this material; this has been demonstrated in an earlier study in which the uptake of water was monitored through changes in the interference colors of thin film MOF coatings.14 Optical measurements for epitaxially grown, high quality, and orientated SURMOF films have not yet been reported; the measurement and determination of the refractive index (n) of nano- and micro-sized powder samples12,13 is a general problem. For example, these problems include the handling of powder compounds with different liquid immersion oils which are possibly toxic. The process uses a timeconsuming measurement series of different refractive index mixtures for which the refractive index (n) can be determined only at one specific wavelength; this occurs during a sedimentation process resulting in a significant loss of powder material.12 Therefore, SURMOFs offer a clear advantage in this context. A number of different techniques and methods are currently known for preparing polycrystalline MOF coatings;15 in addition, the deposition of powder MOF16 material has been described for the fabrication of optical sensors.14 Here, we determine the optical properties of two particular SURMOF types, HKUST-1 and Cu-BDC (benzenedicarboxylate). For both types, the material is grown on pretreated Si substrates using liquid phase epitaxy (LPE)10 in conjunction with a spray process and yields thicknesses in the range of 80–120 nm (see supplementary material23(a)).17 Prior to measuring the refractive index (n) with SE, the SURMOF films were characterized using X-ray diffraction and IRRAS spectroscopy. Morphological studies through scanning electron microscope (SEM) cross-sectional measurements have been performed to check the continuity, compactness, and homogeneity of the SURMOF thin films, as well as to detect possible meso/macro porosity within the epitaxially grown SURMOF coatings. In recent years, a number of other methods have been developed for MOF thin film fabrication,
Appl. Phys. Lett. 103, 091903 (2013)
including painting, dipping into suspensions, and electrochemical methods.5,16,17 However, it has been difficult to obtain high quality MOF thin films using these methods. In particular, determining the refractive index (n) using spectroscopic ellipsometry (SE) is not possible for these polycrystalline films because of their large roughness. However, with liquid phase epitaxy (LPE), it is possible to obtain SURMOF thin films of high morphological quality and low surface roughness by optimizing the parameters of the LPE spray process (see supplementary material23(b)). The high quality of thin films on modified silicon substrates is the essential prerequisite for optical studies and enables reproducible results. SURMOFs made from other metal-organic framework systems have been studied in previous work with regard to their mechanical,11 optical,14 magnetic,18 or electrical19 behavior, which are key properties for the desired functionality.20 The present study examines two different types of MOF materials. First, HKUST-1, is a typical metal-organic framework composed of Cuþþ dimers connected by benzenetricarboxylate (BTC) units (see Fig. 1(a)). This MOF forms a crystalline, 3D pore structure with cubic symmetry and a lat˚ . The second MOF type, tice constant of a ¼ 26.343 A referred to as SURMOF-2, is a surface-anchored variant of the 2D MOF-2 metal-organic framework. SURMOF-2 exhibits P4 symmetry instead of the P2 symmetry present in the Cu-DBC MOF-2. The SURMOF-2 class of metalorganic frameworks is a reticular series of frameworks and can be realized with pore sizes ranging between Ø 0.8 and 2.8 nm.6 In the present study, we investigate one element of these isoreticular series, namely Cu-BDC, which consists of paddlewheel-shaped Cu dimers and benzenedicarboxylate (BDC) as an organic linker.6,20 The structure of this framework is shown in Fig. 1(b). Basically, the MOF consists of planes in which Cu paddlewheel units are arranged in a quadratic fashion. The MOF thus corresponds to a “sheetlike” or “laminar” material (see Fig. 1(b)). The crystal
˚ . (b) Cu-BDC SURMOF-2 with a ¼ b ¼ 10.803 A ˚; FIG. 1. Structure/architecture of the porous solids used in the present study. (a) HKUST-1 with a ¼ 26.343 A ˚. c ¼ 5.60 A
091903-3
Redel et al.
structure is triclinic and the lattice constants are ˚ , c ¼ 5.60 A ˚. a ¼ b ¼ 10.803 A The chemicals used in this work are all commercially available from different sources and suppliers. SURMOFs were grown on pretreated/modified Si wafers (Si wafers (100) orientation, polished, thickness: 525 6 20 lm, specific resistivity: 8–12 X/cm; from Silchem Handelsgesellschaft GmbH, DE). The Si wafers were treated by oxygen plasma (Diener Plasma; airflow: 50 sccm, mixture: pure O2) for 10 min to remove impurities as well as increase the number of OH functional groups and the hydrophilicy on the Si surface before growing the SURMOFs. The Cu-BDC and HKUST-1 SURMOFs were epitaxially grown on the pretreated Si substrate using the LPE spray method described elsewhere,5,16,17 in which the metalcontaining solution [1 mmol Cu(OAc)2] and the linker solution [0.1 mmol BTC and 0.2 mmol BDC] are sprayed subsequently on the substrate (for further details, see the supplementary material23(a)). The desired thickness can be adjust using a number of distinct LPE spray cycles; e.g., 40 spray cycles for HKUST-1 results in a total thickness of 100–120 nm; and 20 spray cycles for Cu-BDC results in a total thickness of 70–85 nm. After the epitaxially growth of the SURMOFs, all samples were characterized using X-ray diffraction (XRD) measured using a D8-Advance Bruker AXS diffractometer with Cu Ka radiation (k ¼ 1.5418 A ) in h/2h geometry equipped with a position sensitive detector. XRD data have been recorded for all samples used in the experiments, before and after the heating process; the measured and simulated XRD data are provided in the supplementary material.23(c) Spectroscopic ellipsometry (SE) analyses were performed with a Woollam M-44 ellipsometer at a fixed incidence angle of 70.0 for the silicon wafers, in the range 400–800 nm. Modeling, fitting, and regression of the ellipsometric data were performed using the VASE software provided by the manufacturer. Morphology studies were performed using cross-sectional images with a Leo-ZEISS SEM operated at 5 kV for SURMOF films on silicon substrates on an Al-SEM cross-section holder. After synthesis, MOFs typically contain residual solvent molecules (EtOH) from LPE synthesis or adsorbed water (H2O) molecules from the environment; these occur through exposure to the ambient air/ environment. Thus, it is crucial to remove any residual molecules from the pores prior to measurements determining the intrinsic SURMOF optical data. In the present study, the MOF materials were heated in a vacuum to elevated temperatures to remove guest molecules hosted in the pores. This is a prerequisite for determining the intrinsic properties of the empty MOF framework.20 The SURMOF coated substrates were then transferred into a glove box (with N2 as the inert atmosphere; H2O ¼ 0.1 ppm; O2 < 10 ppm) and heated over time periods of 1 h, 3 h, and 18 h under vacuum in a roundbottomed Al vessel at 100 C, 150 C, and 200–240 C, respectively (see supplementary material23(d)). After heating the samples were cooled to room temperature and were then transferred into a Teflon cell (see above) for further SE measurements and characterizations. The change in refractive index (n) was measured in a sealed Teflon cell as a function of the heating temperature.
Appl. Phys. Lett. 103, 091903 (2013)
Figures 2(a) and 2(b) show the refractive index (n) obtained for the SURMOFs HKUST-1 (Fig. 2(a)) and CuBDC (Fig. 2(b)) as a function of wavelength after heating at different temperatures. Note that after heating to different activation temperatures, the X-ray diffraction data still indicate the presence of a crystalline material (for details; see supplementary material23(c)). For HKUST-1, the ellipsometric data obtained for a SURMOF heated at 200 C yield a refractive index n ¼ 1.39 at a wavelength of 750 nm, which is in good agreement with the value of n 1.30 calculated from the static dielectric constant reported by Seifert and co-workers.8 Interestingly, freshly prepared MOF thin films in an ambient environment have an optical constant of n ¼ 1.52 at 750 nm. This increase in the optical constant is not consistent with the simple inclusion of water or solvent (EtOH) molecules, since the optical constants of water and EtOH are n 1.33 and n 1.36, respectively. Studies using infrared reflection absorption spectroscopy (IRRAS) (see supplementary material23(e)) for SURMOFs exposed to water vapor reveal the presence of coordinated/bound water molecules, which themselves change the electronic structure and thus the refractive index of the films. In Figs. 2(a) and 2(b), the difference in the refractive index (n) change in the samples measured (i) as prepared, (ii) at 100 C, and (iii) at 150 C was quite small; however, changes at (iv) 200 C and above were significantly larger. A detailed study on the crystalline SURMOF XRD pattern showed that the crystallinity strongly depends on the heating time. For the SURMOFs annealed at 150 C and 225–240 C for 1 h, a crystalline XRD pattern for epitaxially grown CuBDC (001) and (002), as well as for HKUST-1 (222) SURMOF thin films, was observed with good intensity. For longer heating times, e.g., for 3 h and 18 h, the intensities strongly decreased, but the crystalline patterns of the HKUST-1 and Cu-BDC SURMOF thin films still appeared (for details see supplementary material23(c)). Cycling studies and refractive index (n) measurements were performed on HKUST-1 SURMOFs with activated films by exposing them overnight (18 h) to a solvent/EtOH atmosphere. It was shown that the loading and unloading of water/solvent molecules within the SURMOF thin film could be monitored through the change in the refractive index (n) and furthermore was reproducible. A detailed study of the SE and XRD data can be found in the supplementary material.23(c),23(f) For the second type of SURMOF presented, Cu-BDC, similar changes were found upon annealing the SURMOFs. The refractive index (n) value measured for samples annealed, e.g., at 200 C for 18 h was n 1.34 at a wavelength of 750 nm, see also Fig. 2(b). In principle, one would expect at slightly higher value than that for the HKUST-1 SURMOF, since the corresponding density is higher (HKUST-1 SURMOF: 0.984 g cm 3 and Cu-BDC SURMOF-2: 1.157 g cm3). We believe that the lower value is caused by the rough nature of the Cu-BDC SURMOF-2, which leads to lower effective values of n for the thin MOF films. The morphology21,22 of epitaxially grown MOF thin films is of crucial importance for any application or device functionality. MOF films have to reach a distinct thickness of approximately 80–120 nm to become compact,
091903-4
Redel et al.
Appl. Phys. Lett. 103, 091903 (2013)
FIG. 2. (a) Refractive index (n) change of the 3D HKUST-1 SURMOF at different heating temperatures over 18 h. (b) Refractive index (n) change of the laminar Cu-BDC SURMOF-2 material at different heating temperatures over 18 h. (c) and (d) Corresponding dielectric constants k from the determined refractive index (n) data of the HKUST-1 and Cu-BDC SURMOF.
continuous, smooth, and homogeneous. Figures S8a and S9 show SEM cross-sections of HKUST-1 and Cu-BDC. In addition, morphology studies have been performed after two full cycling studies on “heating and loading.” After this cycling process, HKUST-1 SURMOF films still show a high degree of homogeneity without huge cracks or defects of meso/macro porosity (see supplementary material23(b)). Inhomogeneity, cracks, defects, or detectable meso/macro porosity within the epitaxially grown HKUST-1 and
Cu-BDC SURMOF thin films will significantly lower the overall effective refractive index (n) through the inclusion of air voids with n ¼ 1. Previous work using scanning force microscopy, electrochemistry,19 and measurements of BET values have demonstrated that HKUST-1 thin films with a very high degree of homogeneity can be prepared in which defects and electrical short cuts across the films are virtually absent.19 In contrast, for the second type of Cu-BDC SURMOF-2 thin films, the surfaces are much rougher and
FIG. 3. (a) Variation of mass density as a function of linker length in the SURMOF-2 isoreticular series. (b) The refractive index (n) and dielectric constant (k) of the MOF-2 isoreticular series calculated from the experimental data of Cu-BDC at a wavelength of 750 nm using a simple model.
091903-5
Redel et al.
the defect density is considerably higher (see supplementary material23(b)). The observed refractive index (n) of Cu-BDC SURMOF-2 was found to be lower compared to that of the compact and continuous HKUST-1 SURMOF thin films. In conclusion, we report refractive index (n) measurements on epitaxially grown SURMOF thin films using liquid phase epitaxy (LPE). This yields a value of the refractive index (n) for heated HKUST-1 thin films of n ¼ 1.39 at a wavelength of 750 nm, which is in good agreement with the value8 obtained from theoretical calculations for the static dielectric constant corresponding to n ¼ 1.30 (k ¼ 1.7). It will be interesting to see in future experiments whether for the HKUST-1 thin films the experimental static dielectric constant is as low as predicted theoretically (k ¼ 1.7), a value substantially smaller than that of (k ¼ 2.33 6 0.05) reported for polycrystalline thin films made from another MOF, ZIF 8 at a frequency of 100 kHz.9 It should be noted that Cu-BDC SURMOF-2, one of the MOFs studied here, is a member of an isoreticular series of metal-organic frameworks6 with a fairly small pore-size (1.1 nm 1.1 nm, calculated density of 1.157 g cm3). When using longer organic ligands,6 e.g., containing 2 to 10 phenyl rings (see the recent work by Yaghi7 and coworkers) larger pore sizes in the SURMOFs can be realized (see supplementary material23(g)). For such large-pore SURMOFs, the value of n should drop further. Assuming that (n1) is proportional to the mass density and using Cu-BDC-SURMOF-2 as a reference, we yield for a linker with length l nðlÞ ¼ ððn1 1Þ=q1 Þ qðlÞÞ þ 1; where q(l) is the mass density for linker length l, q1 ¼ 1.569 g/cm3 is the density of the reference Cu-BDC SURMOF-2; with n1 ¼ 1.34 at a wavelength of 750 nm correspond to the refractive index (n) of Cu-BDC SURMOF-2 (linker length of 1.115 nm). A plot of the expected behavior is shown in Fig. 3(b) (assuming that k ¼ n2). Using this simple model, we predict a value of k 1.13 for a linker length of 5 nm (see supplementary material23(g)). More work on the interesting dielectric and optical properties of SURMOF thin films is required, as this material class possesses outstanding potential as ultra-low k dielectric thin film materials and coatings. We foresee a bright future of SURMOFs in microelectronic devices and optical sensors. E.R. thanks the Alexander von Humboldt (AvH) Foundation for a Feodor Lynen Postdoctoral ReturnFellowship. Z.W. thanks the Chinese Scholarship Council (CSC) for financial aid. S.W. and H.G. acknowledge financial support from the Landesstiftung Baden-W€urttemberg. This work was funded within the priority program SPP 1362
Appl. Phys. Lett. 103, 091903 (2013)
of the German Research Foundation (DFG). The authors are grateful to Dr. Peter Weidler for valuable discussions. 1
J. Veres, S. D. Ogier, S. W. Leeming, D. C. Cupertino, and S. M. Khaffaf, Adv. Funct. Mater. 13(3), 199–204 (2003). Y. Xu, Y. Tsai, and K. N. Tu, Appl. Phys. Lett. 75(6), 853–855 (1999); D. Shamiryan, T. Abell, F. Iacopi, and K. Maex, Mater. Today 7(1), 34–39 (2004); S. Walheim, E. Sch€affer, J. Mlynek, and U. Steiner, Science 283, 520–522 (1999). 3 H. Li, M. Eddaoudi, M. O’Keeffe, and O. M. Yaghi, Nature 402(6759), 276–279 (1999); O. M. Yaghi, G. M. Li, and H. L. Li, ibid. 378(6558), 703–706 (1995). 4 G. Ferey and C. Serre, Chem. Soc. Rev. 38(5), 1380–1399 (2009). 5 O. Shekhah, J. Liu, R. A. Fischer, and Ch. W€ oll, Chem. Soc. Rev. 40, 1081–1106 (2011). 6 J. Liu, B. Lukose, O. Shekhah, H. K. Arslan, P. Weidler, H. Gliemann, S. Br€ase, S. Grosjean, A. Godt, X. Feng, K. M€ ullen, I.-B. Magdau, T. Heine, and Ch. W€ oll, Sci. Rep. 2, 921, (2012). 7 H. X. Deng, S. Grunder, K. E. Cordova, C. Valente, H. Furukawa, M. Hmadeh, F. Gandara, A. C. Whalley, Z. Liu, S. Asahina, H. Kazumori, M. O’Keeffe, O. Terasaki, J. F. Stoddart, and O. M. Yaghi, Science 336(6084), 1018–1023 (2012). 8 K. Zagorodniy, G. Seifert, and H. Hermann, Appl. Phys. Lett. 97(25), 251905 (2010). 9 S. Eslava, L. Zhang, S. Esconjauregui, J. Yang, K. Vanstreels, M. R. Baklanov, and E. Saiz, Chem. Mater. 25, 27–33 (2013). 10 O. Shekhah, H. Wang, S. Kowarik, F. Schreiber, M. Paulus, M. Tolan, C. Sternemann, F. Evers, D. Zacher, R. A. Fischer, and Ch. W€ oll, J. Am. Chem. Soc. 129, 15118–15119 (2007). 11 S. Bundschuh, O. Kraft, H. K. Arslan, H. Gliemann, P. G. Weidler, and Ch. W€ oll, Appl. Phys. Lett. 101, 101910 (2012). 12 P. G. Weidler and F. Friedrich, Am. Miner. 92, 1130–1132 (2007); F. Friedrich, A. Steudel, and P. G. Weidler, Clays Clay Miner. 56(5), 505–510 (2008). 13 F. D. Bloss, Optical Crystallography, Monograph Series No. 5 (Mineralogical Society of America, 1999), p. 239. 14 G. Lu and J. T. Hupp, J. Am. Chem. Soc. 132(23), 7832–7833 (2010). 15 S. Hermes, F. Schroder, R. Chelmowski, Ch. W€ oll, and R. A. Fischer, J. Am. Chem. Soc. 127(40), 13744–13745 (2005); E. Biemmi, C. Scherb, and T. Bein, ibid. 129(26), 8054–8055 (2007); P. Horcajada, C. Serre, D. Grosso, C. Boissiere, S. Perruchas, C. Sanchez, and G. Ferey, Adv. Mater. 21(19), 1931–1935 (2009). 16 H. Gliemann and Ch. W€ oll, Mater. Today 15(3), 110–116 (2012). 17 H. K. Arslan, O. Shekhah, J. Wohlgemuth, M. Franzreb, R. A. Fischer, and Ch. W€ oll, Adv. Funct. Mater. 21, 4228–4231 (2011). 18 M. E. Silvestre, M. Franzreb, P. G. Weidler, O. Shekhah, and Ch. W€ oll, Adv. Funct. Mater. 23, 1210–1213 (2013). 19 A. Drag€asser, O. Shekhah, O. Zybaylo, C. Shen, M. Buck, Ch. W€ oll, and D. Schlettwein, Chem. Commun. 48, 663–665 (2012). 20 H. K. Arslan, O. Shekhah, D. C. F. Wieland, M. Paulus, C. Sternemann, M. A. Schroer, S. Tiemeyer, M. Tolan, R. A. Fischer, and Ch. W€ oll, J. Am. Chem. Soc. 133, 8158–8161 (2011). 21 O. Zybaylo, O. Shekhah, H. Wang, M. Tafipolsky, R. Schmid, D. Johannsmann, and Ch. W€ oll, Phys. Chem. Chem. Phys. 12(28), 8093–8098 (2010). 22 Y. Yoo and H. K. Jeong, Cryst. Growth Des. 10(3), 1283–1288 (2010). 23 See supplementary material at http://dx.doi.org/10.1063/1.4819836 for (a) Fig. S1 a setup employed for the fabrication of MOF thin films with the LPE spray method; for (b) Figs. S8a and S9 on SURMOF thin film morphology data; for (c) Figs. S6 and S7a-c on XRD data (out of plane) of HKSUT-1 and Cu-BDC SURMOF-2 thin films; for (d) Fig. S2 on the experimental heating equipment; for (e) Fig. S5 on IRRAS data of HKUST-1 and Cu-BDC SURMOFs; for (f) Figs. S4ab and Figs. S4c-e on Spectroscopy Ellipsometry (SE) measurements and for (g) Fig. S10 on the used organic linker lengths in the isoreticular series of MOF-2 (e.g., from 1.1 nm to 2.8 nm). 2
