Vapour-phase deposition of oriented copper dicarboxylate metal–organic framework thin films

Timothée Stassin a, Sabina Rodríguez-Hermida a, Benedikt Schrode b, Alexander John Cruz ac, Francesco Carraro d, Dmitry Kravchenko a, Vincent Creemers a, Ivo Stassen a, Tom Hauffman c, Dirk De Vos a, Paolo Falcaro d, Roland Resel b and Rob Ameloot *a
aCentre for Membrane Separations, Adsorption, Catalysis and Spectroscopy for Sustainable Solutions (cMACS), KU Leuven, Celestijnenlaan 200F box 2454, 3000 Leuven, Belgium. E-mail: rob.ameloot@kuleuven.be
bInstitute of Solid State Physics, Graz University of Technology, Petersgasse 16, 8010 Graz, Austria
cDepartment of Materials and Chemistry, Research group Electrochemical and Surface Engineering, Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussels, Belgium
dInstitute of Physical and Theoretical Chemistry, Graz University of Technology, Stremayrgasse 9/Z2, 8010 Graz, Austria

Received 5th July 2019 , Accepted 25th July 2019

First published on 25th July 2019


Copper dicarboxylate metal–organic framework films are deposited via chemical vapour deposition. Uniform films of CuBDC and CuCDC with an out-of-plane orientation and accessible porosity are obtained from the reaction of Cu and CuO with vaporised dicarboxylic acid linkers.


Metal–organic frameworks (MOFs) are microporous crystalline materials built from metal ion nodes connected by multitopic organic linkers. Because of their record-breaking specific surface area (up to 7500 m2 g−1) and functionalisable pore interior, MOFs have been extensively studied.1–3 Interestingly, the same properties that make MOFs high-performance materials for adsorptive separations also promise high potential in microelectronics.4–6 However, the integration of MOFs in microelectronic devices is rarely reported due to a lack of methods to deposit and pattern MOF thin films in a way that is compatible with current microfabrication protocols.4 MOFs are typically obtained through solvothermal treatment of an organic linker and metal salt dissolved in an organic solvent.1 Besides the environmental impact related to the use of solvents, the deposition of MOF films via wet protocols can lead to contamination and corrosion issues, for instance due to the presence of salts and particle formation.7

Alternative routes for the synthesis of MOF powders make use of non-salt metal precursors (e.g., metal oxides) and little to no solvent. In particular, thermochemical (by heating the linker and precursor together),8,9 mechanochemical (neat, ion-, or liquid-assisted grinding)10–13 and accelerated aging processes14,15 have been proven successful.16 Contact with bulk solvent can also be avoided by reacting a drop-casted synthesis solution containing metal salt and organic linker in a saturated solvent atmosphere, yielding highly oriented MOF films.17

Inspired by the vapour-phase deposition methods common in microfabrication, similar approaches were developed to grow MOF films. Stassen et al. introduced the concept of MOF chemical vapour deposition (MOF-CVD), in which a metal oxide precursor layer is first deposited and subsequently transformed to a MOF upon reaction with linker vapour.18 Directly reacting the surface in an alternating fashion with vaporised metal and linker precursors, has also been reported. This strategy resulted in non-porous, yet crystalline coordination compounds and amorphous layers that yield MOFs after a post-deposition crystallization step.19–21

Here we report the CVD of MOFs based on Cu(II) and the linkers 1,4-benzenedicarboxylic acid (H2BDC) and trans-1,4-cyclohexanedicarboxylic acid (H2CDC). The resulting CuBDC and CuCDC materials consist of 4-connected Cu(II) carboxylate chains that line one-dimensional pores of 5.3 and 3.5 Å in diameter, respectively.22,23 The MOF-CVD method for these materials consists of two steps: (i) vapour-phase deposition of Cu or CuO films as a metal source and (II) a solid–vapour reaction between this precursor and the vaporised organic linker (Fig. 1). Interestingly, the CuBDC and CuCDC MOF thin films grow with an out-of-plane orientation, with the pores normal to the surface.


image file: c9cc05161a-f1.tif
Fig. 1 Chemical vapour deposition of Cu-based MOF thin films (MOF-CVD), a two-step process: CuO or Cu precursor layer deposition and subsequent solid–vapour reaction of this layer with dicarboxylic acid linker vapour.

CuCDC and CuBDC films were obtained on silicon or glass substrates (Fig. S6 and S8, ESI). CuO and Cu precursor layers were deposited via physical vapour deposition (PVD), an established method available in any microfabrication facility. The surface of metallic Cu is terminated by a CuO layer. Thermal treatment, before or during linker exposure, will result in further oxidation (Fig. S3 and Table S4, ESI).24 CuO films of different thicknesses (∼15 nm and ∼100 nm) were placed together with excess H2BDC or H2CDC in a Schlenk tube and evacuated (∼10−1 mbar). To study the effect of relative humidity, water was added for a set of experiments to reach a relative humidity of 5% at the reaction temperature. The reaction was performed at 200 °C for 16 hours, followed by thermal activation. Solvent-free activation by thermal treatment has been demonstrated before in the MOF literature, with several examples of dicarboxylic acid linkers thermally desorbed from the MOF pores (e.g., H2BDC, fumaric acid).25–27

The crystallinity and phase of the resulting films were investigated by synchrotron grazing-incidence X-ray diffraction (GIXRD). The GIXRD data analysed using GIDVis28 was compared with patterns calculated for known crystal structures. Under all conditions phase-pure materials are obtained (Fig. 2). Reaction with H2CDC results in the known porous CuCDC MOF under both dry and humidified conditions. Reaction with H2BDC under dry conditions yields a phase that strongly resembles the isostructural CuBDC material, but with a slight peak shift to higher q values, i.e., a lower interplanar spacing. Reaction with H2BDC under humidified conditions results in a dense phase that strongly resembles the coordination polymer [Cu2(OH)2(BDC)] consisting of Cu(II) hydroxide layers connected by BDC,29 but again, peaks are shifted to higher q values. This material is further referred to as CP-CuBDC.


image file: c9cc05161a-f2.tif
Fig. 2 The reactive atmosphere composition (linker and relative humidity) determines the formed MOF structure. Reaction of CuO and Cu with H2CDC vapour under dry and humidified conditions results in the CuCDC structure. Reaction with H2BDC under dry conditions results in the CuBDC structure while humidified conditions yield the CP-CuBDC structure. Intensity distribution as a function of 2θ extracted from GIXRD patterns (coloured) and simulated patterns (black).

The MOF/CP formation was corroborated by ATR-FTIR spectroscopy (Fig. S13 and Table S7, ESI). The absence of a band around 1670 cm−1 indicates that no uncoordinated carboxylic acid is present. The vibrational bands at ∼1590 cm−1 and ∼1420 cm−1 are respectively attributed to the asymmetric and symmetric vibrations of metal-coordinated carboxylate groups. All films grown from thin CuO films have a homogeneous, mirror-like appearance and a colour ranging from deep blue to brown (Fig. 3 and Fig. S4, ESI). AFM images show homogeneous films with RRMS roughness values in the 10–30 nm range (Table S6, ESI).


image file: c9cc05161a-f3.tif
Fig. 3 Thin film morphology. Top left: Homogeneous, mirror-like CuCDC film. Top right: AFM images of CuCDC films deposited under dry and humidified conditions, and of CuBDC and CP-CuBDC films. Bottom: Tilted-view and cross-sectional SEM images of CuCDC film grown from 15 nm of CuO with no residual CuO visible (left) and from 100 nm of CuO layer showing incomplete CuO conversion (right). AFM and SEM scale bars are 400 and 200 nm, respectively. AFM colour scale: see Fig. S9 (ESI).

The total film thickness was determined by ellipsometry and cross-checked by high-resolution cross-sectional SEM, with less than 10% difference between both methods. The thickness for films grown from ∼15 nm CuO is 50 ± 3 nm for CP-CuBDC, 83 ± 5 nm for CuBDC, 87 ± 5 nm for CuCDC (dry) and 110 ± 8 nm for CuCDC (hum.) (Fig. S10, ESI). The thickness of the CuCDC, CuBDC and CP-CuBDC films is lower than expected based on the ratio of the density measured for the CuO film (6.5 g cm−3; Fig. S2 and Table S3, ESI) and calculated for the final materials based on their crystal structure. For this precursor thickness, complete oxide-to-MOF conversion would result in ∼80 nm CP-CuBDC, ∼190 nm CuCDC and ∼200 nm CuBDC films, i.e., an expansion factor of 5, 13 and 13, respectively. Although cross-sectional SEM does not show residual CuO, a few nm of unreacted precursor cannot be ruled out due to the instrument resolution (Fig. 3). When starting from thicker (∼100 nm) CuO, similar images reveal a ∼110 nm CuCDC film on top of ∼90 nm unreacted CuO (Fig. 3). These results suggest that only about the first 10 nm of CuO is converted to MOF (Fig. S11, ESI), leaving the remaining precursor unreacted. Similar self-limiting MOF growth has previously been observed for ZIF-8 CVD starting from ZnO layers.18 Although the linkers have a different volatility (higher for H2CDC) and acidity (higher for H2BDC), these differences cancel out and result in the same fraction of CuO converted (Fig. S12, ESI).

GIXRD indicates that all CuCDC films, independently of the precursor, are oriented with the (100) plane parallel to the substrate (Fig. 4 and Fig. S5, ESI). For H2BDC, the results are different: while the CuBDC MOF is also clearly (100)-oriented, CP-CuBDC shows a weaker and different preferred orientation that depends on the precursor. For CP-CuBDC films prepared from thermal CuO and PVD CuO, the 010 peak at q ≈ 0.64 Å−1 is found at approximately 20 degrees tilted from the vertical axis. For CP-CuBDC grown from PVD Cu, this off-specular tilt is even stronger. These results indicate a different contact plane than (100) and potentially the presence of multiple orientations (Fig. S7, ESI).


image file: c9cc05161a-f4.tif
Fig. 4 Out-of-plane orientation of CuCDC and CuBDC films. (a) GIXRD pattern for CuCDC (dry) overlaid with the Debye–Scherrer diffraction rings (grey lines) and Bragg peak position (red dots) calculated for a (100) orientation. (b) Normalised ATR-FTIR spectra of MOF-CVD thin films and bulk CuCDC powder synthesized solvothermally. The carboxylate bands are highlighted in grey. The differences in absorbance between the asymmetric and symmetric vibrations are indicative of the orientation of the carboxylate groups with respect to the substrate. (c) Schematic representation of (100)-oriented CuCDC and the resulting pore orientation perpendicular to the substrate highlighted in yellow.

image file: c9cc05161a-f5.tif
Fig. 5 Patterned Cu-based MOF thin films. Left: Schematic representation of the patterning sequence based on shadow mask CuO deposition. Right: Patterned Cu-based MOF thin films. Scale bar = 1 cm.

The (100) orientation of CuBDC and CuCDC films is confirmed by ATR-FTIR using p-polarised light since the absorbance of the asymmetric carboxylate vibration (∼1590 cm−1) is strongly enhanced compared to the symmetric one (∼1410 cm−1) (Fig. 4). The IR selection rules predict that vibrations normal to the surface give a reduced signal, whereas the signal from vibrations parallel to the surface is enhanced.30 For both CuBDC and CuCDC, the average dihedral angle between the (100) plane and the crystallographic plane defined by the carboxylate groups is 70° (Fig. S14 and Table S8, ESI). Since this angle is close to 90°, the observed difference in absorbance is explained by the (100) plane oriented parallel to the substrate and the resulting near-perpendicular orientation of the Cu–carboxylate bonds. The ratio of the symmetric and asymmetric carboxylate absorbance signals indicates a higher degree of orientation for CuCDC grown in a dry versus a humid atmosphere (8.7 and 6.2, respectively) and a higher degree of orientation for CuCDC than CuBDC (8.7 and 6.7, respectively) (Table S7, ESI).

Both CuBDC and CuCDC have one-dimensional pores parallel to the crystallographic a-axis. Therefore, (100)-oriented films of these materials have pores oriented perpendicular to the surface, thus ideally accessible for guest molecules (Fig. 4). Conversely, CuBDC films prepared by liquid-phase epitaxy display a (001) orientation with pores oriented parallel to the substrate (Fig. S15, ESI).31 To evaluate their porosity, CuCDC and CuBDC films were deposited by CVD on quartz crystal microbalance (QCM) crystals and the resonance frequency change was measured upon exposure to increasing methanol vapour concentrations. Both films take up significantly more methanol compared to a bare reference crystal, thus demonstrating the pore accessibility (Fig. 6).


image file: c9cc05161a-f6.tif
Fig. 6 Gravimetric methanol uptake in CuCDC and CuBDC films. Resonance frequency shift of QCM crystals coated with CuCDC (blue) and CuBDC (red) films and an uncoated reference (black) as a function of relative pressure. Error bars correspond to the 95% confidence interval based on at least 7 measurements.

The development of MOF-CVD methods will facilitate the integration of MOFs in microelectronics. While Stassen et al. demonstrated lift-off patterning for ZIF-8 MOF-CVD, we show the direct conversion of patterned CuO deposited by PVD through a shadow mask to patterned MOF films (Fig. 5).18

In conclusion, we expanded MOF-CVD to thin films of copper dicarboxylate MOFs. To the best of our knowledge, these are the first crystalline and oriented porous materials deposited entirely from the vapour phase. Future MOF-CVD research will target even more porous Cu-based MOFs, such as the well-known HKUST-1 based on Cu(II) paddlewheels linked by 1,3,5-benzenetricarboxylate.32

T. S. and I. S. thank the Research Foundation Flanders (FWO) for a SB-PhD fellowship and postdoctoral fellowship (1S53316N and 12L5417N). R. A. and P. F. acknowledge the funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreements no. 716472 and 771834, acronyms: VAPORE and POPCRYSTAL), and the Research Foundation Flanders (FWO) for funding in the projects G083016N, 1529618N and 1501618N. This study is a result from the lead project Porous Materials@Work (Graz University of Technology, Austria). We acknowledge the European Synchrotron Radiation Facility for provision of synchrotron radiation facilities and we would like to thank Oleg Konovalov and Andrey Chumakov for assistance in using beamline ID10. The authors are grateful to Steven De Feyter for access to AFM instrumentation and to Max L. Tietze for XPS data processing and CuO patterning.

Conflicts of interest

There are no conflicts to declare.

Notes and references

  1. A. Corma, H. García and F. X. Llabrés i Xamena, Chem. Rev., 2010, 110, 4606–4655 CrossRef CAS PubMed.
  2. J.-R. Li, R. J. Kuppler and H.-C. Zhou, Chem. Soc. Rev., 2009, 38, 1477–1504 RSC.
  3. I. M. Hönicke, I. Senkovska, V. Bon, I. A. Baburin, N. Bönisch, S. Raschke, J. D. Evans and S. Kaskel, Angew. Chem., Int. Ed., 2018, 57, 13780–13783 CrossRef PubMed.
  4. I. Stassen, N. Burtch, A. Talin, P. Falcaro, M. Allendorf and R. Ameloot, Chem. Soc. Rev., 2017, 46, 3185–3241 RSC.
  5. L. E. Kreno, K. Leong, O. K. Farha, M. Allendorf, R. P. Van Duyne and J. T. Hupp, Chem. Rev., 2012, 112, 1105–1125 CrossRef CAS PubMed.
  6. V. Stavila, A. A. Talin and M. D. Allendorf, Chem. Soc. Rev., 2014, 43, 5994–6010 RSC.
  7. I. Stassen, D. De Vos and R. Ameloot, Chem. – Eur. J., 2016, 22, 14452–14460 CrossRef CAS PubMed.
  8. I. Stassen, N. Campagnol, J. Fransaer, P. Vereecken, D. D. Vos and R. Ameloot, CrystEngComm, 2013, 15, 9308–9311 RSC.
  9. J.-B. Lin, R.-B. Lin, X.-N. Cheng, J.-P. Zhang and X.-M. Chen, Chem. Commun., 2011, 47, 9185–9187 RSC.
  10. K. Uzarevic, T. C. Wang, S.-Y. Moon, A. M. Fidelli, J. T. Hupp, O. K. Farha and T. Friščić, Chem. Commun., 2015, 52, 2133–2136 RSC.
  11. A. Pichon, A. Lazuen-Garay and S. L. James, CrystEngComm, 2006, 8, 211 RSC.
  12. S. Tanaka, K. Kida, T. Nagaoka, T. Ota and Y. Miyake, Chem. Commun., 2013, 49, 7884–7886 RSC.
  13. K. Imawaka, M. Sugita, T. Takewaki and S. Tanaka, Polyhedron, 2019, 158, 290–295 CrossRef CAS.
  14. X. Feng, C. Jia, J. Wang, X. Cao, P. Tang and W. Yuan, Green Chem., 2015, 17, 3740–3745 RSC.
  15. C. Mottillo, Y. Lu, M.-H. Pham, M. J. Cliffe, T.-O. Do and T. Friščić, Green Chem., 2013, 15, 2121–2131 RSC.
  16. S. L. James, A. LaZuen-Garay and A. Pichon, US Pat., US8466285B2, 2013 Search PubMed.
  17. E. Virmani, J. M. Rotter, A. Mähringer, T. von Zons, A. Godt, T. Bein, S. Wuttke and D. D. Medina, J. Am. Chem. Soc., 2018, 140, 4812–4819 CrossRef CAS PubMed.
  18. I. Stassen, M. Styles, G. Grenci, H. V. Gorp, W. Vanderlinden, S. D. Feyter, P. Falcaro, D. D. Vos, P. Vereecken and R. Ameloot, Nat. Mater., 2016, 15, 304–310 CrossRef CAS PubMed.
  19. L. D. Salmi, M. J. Heikkilä, E. Puukilainen, T. Sajavaara, D. Grosso and M. Ritala, Microporous Mesoporous Mater., 2013, 182, 147–154 CrossRef CAS.
  20. E. Ahvenniemi and M. Karppinen, Chem. Commun., 2016, 52, 1139–1142 RSC.
  21. K. B. Lausund and O. Nilsen, Nat. Commun., 2016, 7, 13578 CrossRef CAS PubMed.
  22. H. Kumagai, M. Akita-Tanaka, K. Inoue, K. Takahashi, H. Kobayashi, S. Vilminot and M. Kurmoo, Inorg. Chem., 2007, 46, 5949–5956 CrossRef CAS PubMed.
  23. C. G. Carson, G. Brunnello, S. G. Lee, S. S. Jang, R. A. Gerhardt and R. Tannenbaum, Eur. J. Inorg. Chem., 2014, 2140–2145 CrossRef CAS.
  24. I. Platzman, R. Brener, H. Haick and R. Tannenbaum, J. Phys. Chem. C, 2008, 112, 1101–1108 CrossRef CAS.
  25. T. Loiseau, C. Serre, C. Huguenard, G. Fink, F. Taulelle, M. Henry, T. Bataille and G. Férey, Chem. – Eur. J., 2004, 10, 1373–1382 CrossRef CAS PubMed.
  26. T. Stassin, S. Waitschat, N. Heidenreich, H. Reinsch, F. Puschkell, D. Kravchenko, I. Stassen, J. van Dinter, R. Verbeke, M. Dickmann, W. Egger, I. Vankelecom, D. De Vos, R. Ameloot and N. Stock, ChemRxiv DOI:10.26434/chemrxiv.8311235.v1.
  27. K. Barthelet, J. Marrot, D. Riou and G. Férey, Angew. Chem., Int. Ed., 2002, 41, 281–284 CrossRef CAS PubMed.
  28. B. Schrode, S. Pachmajer, M. Dohr, C. Röthel, J. Domke, T. Fritz, R. Resel and O. Werzer, J. Appl. Crystallogr., 2019, 52, 683–689 CrossRef CAS PubMed.
  29. S. Abdelouhab, M. François, E. Elkaim and P. Rabu, Solid State Sci., 2005, 7, 227–232 CrossRef CAS.
  30. V. P. Tolstoy, I. V. Chernyshova and V. A. Skryshevsky, Handbook of Infrared Spectroscopy of Ultrathin Films: Tolstoy/Infrared Spectroscopy, John Wiley & Sons, Inc., Hoboken, NJ, USA, 2003 Search PubMed.
  31. J. Liu, B. Lukose, O. Shekhah, H. K. Arslan, P. Weidler, H. Gliemann, S. Bräse, S. Grosjean, A. Godt, X. Feng, K. Müllen, I.-B. Magdau, T. Heine and C. Wöll, Sci. Rep., 2012, 2, 921 CrossRef PubMed.
  32. S. S.-Y. Chui, S. M.-F. Lo, J. P. H. Charmant, A. G. Orpen and I. D. Williams, Science, 1999, 283, 1148–1150 CrossRef CAS PubMed.
  33. T. F. Willems, C. H. Rycroft, M. Kazi, J. C. Meza and M. Haranczyk, Microporous Mesoporous Mater., 2012, 149, 134–141 CrossRef CAS.

Footnotes

Electronic supplementary information (ESI) available: Acknowledgments, materials and methods, Cu(II) precursor characterisation (ellipsometry, XRR, XPS), optical images, GIXRD patterns, SEM images, AFM investigation, film thickness by ellipsometry, thermogravimetric analysis, ATR-FTIR data and dihedral angles. See DOI: 10.1039/c9cc05161a
Crystal structure codes from the CSD: CuCDC (SIWGUB), CuBDC (ZUBKEO), CP-CuBDC (KAKSUL). Calculated pore diameters obtained by Monte Carlo sampling using Zeo++.33 Film expansion factor calculated from MOF/CP crystallographic density and experimental CuO density.

This journal is © The Royal Society of Chemistry 2019