Porphyrin based metal-organic frameworks: highly sensitive materials for optical sensing of oxygen in gas phase

Luminescence of PCN-224-based MOFs is efficiently quenched by molecular oxygen making them promising for optical sensing of oxygen in the gas phase.


S3
Metalation and deprotection of the methyl esters was done by a modified procedure: 2 70 mg (83 µmol) of H2TMCPP were dissolved in 50 ml benzonitrile in a Schlenk flask. Oxygen was removed by bubbling argon through the solution while vigorously stirring. 78 mg (165 µmol) cis-bis((benzonitrile)dichloroplatinum(II) were added and the solution was heated at 150 °C overnight. After the reaction was completed as indicated by UV-VIS spectroscopy, the solution was cooled to room temperature. The product was precipitated in 200 mL cyclohexane.
The crude product was filtered through silica gel and washed thoroughly with cyclohexane.
The product was eluted from the silica gel bed with THF. 100 mg crude product were obtained and directly used without any further purification in the next step. 100 mg crude Pt(II)TMCPP were dissolved in 50 mL THF and 3 mL 1M aq. NaOH were added.
The solution was stirred at 65 °C overnight and cooled to room temperature. After reaction completion indicated by TLC, the product was precipitated by dropwise addition of 1M HCl. Pd(II)TCPP was synthesized according to a reported procedure. 3 65 mg (82 µmol) of 5,10,15,20-tetrakis-(4-carboxyphenyl)-21,23H-porphyrin were dissolved in 4 mL DMF in a 10 mL microwave-suitable borosilicate vial. 60 mg (338 µmol) of palladium dichloride was added, the reaction vessel was closed and heated in a synthesis reactor at 155 °C for 15 minutes.
Completion of the reaction was controlled after cooling to room temperature by UV-VIS spectroscopy. The resulting solution was filtered to remove colloidal palladium, diluted with THF : diethyl ether (2:1 v/v), filtered and washed with 3 x 20 mL water. All solvents were removed and the product was dried at 65 °C in vacuum. 55 mg product (74.8%) were obtained as a red powder. 1

Synthesis of poly(trimethylsilylpropyne) (PTMSP)
Polymerization of TMSP was carried out in a dry box as follows. 4 TaCl5 (0.5 g, 1.4 mmol) was dissolved in 0.1 L of toluene. The mixture was stirred for 30 min at room temperature. 12 g (107 mmol) of TMSP were added to this catalyst solution. The mixture immediately turned dark brown and the solution solidified within 30 min. After 24 h the polymerization mixture was worked up by precipitation of the polymer in rapidly stirred 200 mL hot methanol. The polymer was washed several times with hot methanol and then dried to a constant weight. A yield of 95% (11.4 g) was achieved.

Pt(II)PCN-224
The measured data for this system were compared to known structures for two PCN-224 based MOFs, one MOF consisting of Ni(II) metalated TCPP (Ni(II)TCPP) and the other based on the metal-free porphyrin. 6 The reported structures (see Fig. S1) are cubic and belong to space group Im-3m (no. 229). To mimic the Pt(II)PCN-224 MOF we replaced the Ni atom in the literature structure by Pt.  by Pt atoms. The labels in white are the Laue indices.

S7
In the next step also the relative intensities of the peaks are compared between the model structures (Ni(II)PCN-224, Pt(II)PCN-224) and the measured data. For this comparison we integrated the intensity over the entire PXRD pattern (shown in Fig. S2 and S3) and these data are then normalized to the maximum intensity. As a logical next step these data are compared to calculated PXRD patterns of the model structures. These calculations were performed using the Mercury software package. [7][8][9][10][11] The obtained results are shown in Fig. S4 and S5. One can see that for both model systems Ni(II)PCN-224 and Pt(II)PCN-224 the relative intensities of all peaks agree well with the relative intensities determined in the experiment. Two peaks with rather weak experimental intensities (013 and 123) better agree to the Pt(II)PCN-224 data which can be directly explained by the larger atomic number of Pt and the resulting larger scattering factor. These peaks are hardly detectable for the Ni containing system, but can be measured for the Pt system, which is what we see in the experiment. Based on these observations one can conclude that the newly synthesized Pt(II)PCN-224 MOF adopts the structure of Ni(II)PCN-224, with Ni atoms being replaced by Pt atoms. Nevertheless, a refinement of the unit cell parameter (lattice vectors a=b=c) has been performed by using the program PowderCell. 12 The lattice parameter was varied and the resulting PXRD pattern was compared to the experimental one. This refinement procedure resulted in a lattice vector of 38.62 Å. For this lattice parameter we find a good agreement between experiment and model (see Fig. S6). Some structural information for this system is given in Table S1. The data for the model structures was calculated by using the Mercury software package. The scattering angle has been calculated for a wavelength of 1.5406 Å (Cu-Kα1).  S9

PCN-224:
The results for PCN-224 have been analyzed analogously to the steps described above for Pt(II)PCN-224. The synchrotron PXRD pattern is shown in Fig. S7 and the extracted line graph is presented in Fig. S8.
By comparing peak positions and relative intensities between the measured data and the data calculated for the literature structure one can conclude that the synthesized MOF exhibits the structure reported for PCN-224 in Ref [ 6 ]. Despite this good agreement it seems that all the experimental peaks are slightly shifted towards lower angles (see Fig. S8), which would correspond to larger unit cell parameters. As this is true for all observed peaks and as the system has cubic symmetry this could for example be rationalized by different temperatures of the measurement here and in Ref [ 6 ]. Isotropic volumetric expansion could lead to such shifts of the peaks. Refining the unit cell parameter based on the measured data by using the program PowderCell 12 and employing the same methodology as described above one can actually find a slightly enlarged lattice parameter of 38.61 Å. Nevertheless, we can safely conclude on the synthesized material exhibiting the literature reported structure, 6 with the refined lattice parameter of 38.61 Å. In Table S2 some structural information can be found for PCN-224. The data for the model structures was calculated by using the Mercury software package. The scattering angle has been calculated for a wavelength of 1.5406 Å (Cu-Kα1).

PXRD measurement of Pd(II)PCN-224
Powder x-ray diffraction (PXRD) measurements of Pd(II)PCN224 were performed with an PANalytical Empyrean system using a sealed copper tube together with a multilayer mirror for generating a parallelized and monochromatized primary X-ray beam. A wavelength of 1.542 Å was used. The scattered intensity was detected with a PIXcel detector operating in a onedimensional (1D) mode. The diffraction pattern was converted into a reciprocal space using the equation = 4 (Θ) , with q as the length of the scattering vector, λ as the used wavelength, and 2 as the angle between the primary and the scattered X-ray beam. The distance d between the real space planes depends on q as follows: = 2 .The intensity has been normalized to its maximum value within the considered q-range. The results are shown in Fig. S10 and S11.

Electrospinning of PAN micro/nanofibers
A 10% w/w solution of PAN in DMF was electrospun using a spinneret with a diameter of 0.8 mm. The humidity level was kept at 35% relative humidity throughout the electrospinning procedure and as a collector an aluminum foil covered square copper plate with an area of 16 cm² was used. Electrospinning was done using 25 kV DC voltage for 15 minutes. The nonwoven fibrous material was collected as a white, stable fiber matt. The PAN fiber matt was easily removed from the substrate with a pair of tweezers.
The PAN micro/nanofibers were then cross-linked by heating them with a controlled temperature increase of 2 °C/min from room temperature to 280 °C. The fibers were kept at 280 °C for 2 h. After cross-linking, the PAN nanofibers were slightly darker and completely insoluble in DMF. S17   According to the specification of the high purity N2 7.0(≥99.99999 % purity) from Linde (www.linde.com), residual oxygen of 30 ppb or less can be expected. This level of residual oxygen is within the purity obtained by purifying N2 6.0 with commercially available gas filters for O2 removal and can therefore considered to be sufficient for ultra-trace oxygen measurements.
 To ensure high precision of gas mixing, mass flow controllers (Voegtlin, red-y s GSC-A4 series, with highest accuracy (+-0.3 % of full scale and +-0.5 % of reading according to specification) were used.
 As a measurement chamber, a 10 mL Schlenk flask was equipped with a steel capillary. The sample to be measured was placed in a glass vial in the Schlenk tube and the sample was excited via optical fibers through the bottom wall of the Schlenk tube. S19 calibration gas (20ppm O2 in N2) and pure N2 from 20 ppm down to N2 to ensure oxygen-free conditions at the beginning of the calibration.
 Calibration required an equilibration phase between 1 and 3 calibration cycles to ensure stable calibration.

S20
Setup for temperature controlled calibrations

Immobilization of TMCPP in poly(1-trimethylsilyl-1-propyne) (PTMSP) and its sensing properties of the material
A stock solution of PTMSP with a concentration of 10 mg mL -1 in toluene was prepared. It was used to prepare the polymer "cocktail" containing 0.5% w/w of TMCPP in respect to the polymer which was knife-coated (25 µm-thick wet layer) on a clean PET support. and f can be found in the inlet.

Immobilization of TCPP on aluminum supported TLC silica gel and the sensing
properties of the material 10 mg silica gel was stirred with 0.5 mg TCPP in 5mL THF for 20 minutes. Afterwards the silica gel was filtered and washed 3x with 5 mL THF and 3x with 5 mL acetone until no TCPP was found in the filtrate. The slightly colored silica gel was dried at 70 °C under reduced pressure overnight. Calibrations were recorded as described.