Electronic Supplementary Information Degradation of ZIF-8 in Phosphate Buffered Saline Media

Miriam de J. Velásquez-Hernández, Raffaele Ricco, Francesco Carraro, Ted Limpoco, Mercedes Linares-Moreau, Erich Leitner, Helmar Wiltsche, Johannes Rattenberger, Hartmuth Schröttner, Philipp Frühwirt, Eduard M. Stadler, Georg Gescheidt, Heinz Amenitsch, Christian J. Doonan, Paolo Falcaro a Institute of Physical and Theoretical Chemistry, Graz University of Technology, 8010 Graz, Austria. b Department of Chemistry, The University of Adelaide, 5005 Adelaide, South Australia, Australia. c Insitute of Analytical Chemistry and Food Chemistry, Graz University of Technology, 8010, Graz, Austria. d Graz Centre for Electron Microscopy (ZFE), 8010 Graz, Austria. e Institute of Inorganic Chemistry, Graz University of Technology, 8010 Graz, Austria. f Oxford Instruments GmbH Asylum Research, 65205 Wiesbaden, Germany. E-mail: christian.doonan@adelaide.edu.au, paolo.falcaro@tugraz.at


Gas Chromatography-Mass Spectrometry (GC-MS) analysis of 2-methylimidazole
The GC-MS analyses of the mother liquors were carried out on a gas chromatographic system with a single quadrupole mass selective detector (Shimadzu QP 2010 Ultra, Shimadzu Europe GmbH, Duisburg, Germany) with an AOC 5000 autosampler.
Aliquots of one microliter was injected in split mode (split ratio 1:20) at 250°C injection port temperature with a specific liner for aqueous injection (4 mm precision liner, Topaz, Restek, Belfonte, USA).
Separation was done on a polar column (Phenomenex ZB Wax 20 m*0.18 mm inner diameter and 0.18 µm film thickness, Phenomenex LTD, Aschaffenburg, Germany) with the following temperature ramp starting at 60°C (1 min) and ramped to 240°C with a temperature ramp of 15°C min -1 .Helium was used as carrier with a linear velocity of 35 cm/s in the constant flow mode.The quadrupole was operated in the scan mode with a scan range from 50-150 amu with a scan rate of 5 scans per second.Interface and ion source temperature were set to 230°C and 200°C respectively.Multiplier voltage was reduced by 200 voltage relative to tune value to avoid detector saturation.
Six different standard solutions of 2-methylimidazole (1 mM, 2 mM, 3 mM, 4 mM, 5 mM, and 6 mM) were analyzed with the established GC-MS method to generate the corresponding calibration curve.The repeatability of the analytical method was tested by conducting three repeated analyses of the standard solutions.

P Nuclear Magnetic Resonance Spectroscopy (NMR)
31 P NMR spectrum of the PBS (10 mM) prepared in D2O was collected on a 200 MHz Bruker spectrometer (time = 0).Afterwards, the micro-ZIF-8 particles (0.5 mg mL -1 ) were soaked into this sample, and the corresponding 31 P NMR spectra were recorded at different incubation times (1 h and 24 h).
To explain the decrease in signal associated with PBS without any newly appearing resonances we performed a tritration experiment with Na2HPO4 and ZnCl2.To facilitate NMR detection we increased the concentration of phosphate in solution to ca. 100 mM.
We prepared an NMR sample containing 113 mM phosphate (11.2 mg Na2HPO4 in 700 µL D2O, pH ≈ 8) and stepwise added 20 µL of an 878 mM solution of ZnCl2 in D2O.This results in Zn 2+ concentrations of 0, 24, 47, and 69 mM.A 31 P spectrum was recorded after each addition (Figure S1).After adding Zn 2+ the solution becomes immediately cloudy.The

Powder X-ray diffraction (PXRD)
PXRD patterns of the micro-and nano-ZIF-8 samples were collected on a Rigaku powder diffractometer equipped with D/teX Ultra 250 detector and using CuKα radiation (λ = 1.5406Å).The scan speed was 3 deg min -1 and the step 0.01°.

Scanning electron microscopy (SEM)
SEM micrographs were collected using Tescan VEGA 3 SEM with tungsten source filament working at 20 kV.The samples were deposited on conductive carbon tape and subsequently coated with Au.The EDX elemental mapping was performed using a Zeiss Sigma 300 VP (Field Emission Gun) with an Oxford X-Max N 80mm² EDX Detector.

Time-resolved Small Angle X-Ray Scattering (SAXS)
Time-resolved SAXS have been collected on the SAXS beamline at the ELETTRA synchrotron light source.We operated at photon energy of 8 keV covering the range of momentum transfer, q = 4π sin(θ)/λ, between 0.12 and 6.3 nm -1 .The stability of the MOF was monitored by measuring the MOF suspension in a 1 mm glass capillary.All the experiments were performed at RT.The acquisition started 15 minutes after the dispersion of ZIF-8 in PBS with a time resolution of 3 minutes (detector: Pilatus3 1M, Dectris Ltd, Baden, Switzerland; sample to detector distance: 1260 mm, as determined with a silver behenate calibration sample).Prior the experiment, PBS 10 mM solution was measured in order to assess and subtract the background from the data.The resulting two-dimensional images were radially integrated to obtain a 1D pattern of normalized intensity versus scattering vector q.
line shape is deteriorated and the 31 P integral steadily decreases.Due to different counterions, concentration and pH the 31P shift may be different compared to the PBS experiment.Importantly, no new resonances appear in the spectrum.

Figure S1 .
Figure S1.NMR 31 P control experiment to prove the decrease of the phosphate signal due to the formation of insoluble zinc phosphate by-products.AFM experimental section AFM samples were prepared as follows.The ZIF-8 polycrystalline film was prepared by inducing the growth of ZIF-8 on a Bovine Serum Albumin (BSA) film supported on Si (100).The BSA film was obtained with the following procedure: first Si was treated in basic Piranha solution at 60°C for 30 minutes and then exposed to a toluene solution of (3-Aminopropyl)triethoxysilane (ATPES, 10 μL in 5 mL).Then, APTES-functionalized Si was exposed to a mixture of BSA (1 mg in 0.5 mL of 15 mM MES buffer at pH=6) and 1-Ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC, 1 mg in 100 μL of 15 mM MES buffer at pH=6) at RT for 2 h.The sample was then rinsed and stored in fresh 15 mM MES buffer at 4°C.The polycrystalline ZIF-8 film was grown by exposing the BSA/APTES film to 1 mL of a freshly prepared 40 mM of Zn(OAc)2•2H2O and 0.64 mM of HmIM water solution at RT for 30 minutes.Finally, the sample was rinsed with ethanol and dried under Nitrogen flux.AFM experiments were performed using an Asylum Research (Oxford Instruments) model Cypher ES in tapping mode, with blueDrive photothermal excitation.The probes used were Olympus Instruments AC 40 silicon nitride cantilevers with silicon

Figure S2 .
Figure S2.Time-resolved solution SAXS of nano-ZIF-8 particles dispersed in DI water.The intensity of the (011) diffraction peak of sod-ZIF-8 particles dispersed in DI water is almost unchanged after a 5 h exposure and this is highlighted in the inset.