An investigation into the adsorption mechanism of n-butanol by ZIF-8: A combined experimental and ab initio molecular dynamics approach

2-methylimidazole

methanol and dried in an oven at 80 o C for a further 3 days. Both samples were activated in a vacuum oven at 120 °C overnight, to remove residual solvent trapped inside the pores of the structure.
Powder X-ray diffraction data (PXRD) for spectra between 2 20-60 ° were collected on a Bruker D8 Discover in transmission geometry using monochromatic Co K α1 radiation (λ = 1.7890 ). A ṡ tep size of 0.015 ° was employed with a step time of 0.6 s. For low angle spectra, Powder X-ray diffraction data (PXRD) data were collected on a Bruker D8 Advance in reflectance geometry using monochromatic Cu K α1 radiation (λ = 1.5406 ) and a quartz calibrated Lynxeye detector over a 2 ṙ ange of 5 -35 ° with a step size of 0.015 °. A narrow primary focal slit of 0.2 mm was used to reduce the background caused by the Perspex sample holder and as a result a step time of 16 s was required for these measurements.
Combined thermogravimetric (TGA) and differential thermal analysis (DTA) data were collected on a TA Instruments SDT Q600 with two alumina crucibles (TA Instruments), one containing a reference and the other, the sample. A temperature range ramp method was applied, using an increasing temperature rate of 5 o C min -1 , over a range of 25 -800 o C. A flow of air was regulated at 100 ml min -1 during analysis.
Surface area and pore volume data were determined using nitrogen adsorption/desorption isotherms obtained on a Micrometritics ASAP 2020. The samples were degassed under a nitrogen vacuum for 4 hours at 150 o C prior to analysis. The weight of the sample after degassing was used in all further calculations. BET surface areas were calculated over a partial pressure range of 0.005-0.0005 (p/p o ). The total pore volume was measured with a single point on the adsorption isotherm at p/p 0 = 0.994 cm 3 g -1 .
Attenuated total reflectance infrared (ATR-IR) were collected using a Shimadzu IT Affinity-1 fitted with an ATR stage. The spectra were recorded between 400 -4000cm -1 with a resolution of 4 cm -1 for 64 scans, both background and sample spectra.
Raman (R) data were collected between 100 -4000 cm -1 using a Horiba Jobin Yvon HR LabRAM system in the backscatter configuration with a laser line at 633 nm originating from an argon ion laser.
The laser was focused onto the sample to a spot size of around 1 μm at 100 % power during analysis, as no sample decomposition was observed under these conditions. Ionisation Detector (BID). The column used was an Agilent DB-WAX column with a carrier gas of He flowing at a rate of 50 ml min -1 . Samples were diluted 10-fold in pure methanol before injection onto the column. The column was heated using a temperature programme of 35 C for 4 mins, followed by a ramp at 10 C min -1 to 95 C.
The equilibrium adsorption capacity, Qe, of ZIF-8 after 24 h is defined as grams of adsorbate per gram of adsorbent, calculated using equation 1: Where C o is the starting concentration of solution (g L -1 ); C e is the equilibrium concentration of solution (g L -1 ); V is the volume of solution (L); M is the mass of adsorbent (g). 1 The Langmuir model is then fitted to adsorption data to determine the maximum saturation capacity for n-butanol, as summarised in equation 2 below where: Q max is the saturation adsorption capacity (g g -1 ), k is the Langmuir coefficient (L g -1 ) and C e is the equilibrium concentration of the solution (g g -1 ). 2 The linear derivation of the Langmuir model is as follows in equation 3: where E ZIF+nBuOH is the energy of ZIF-8 with n-butanol adsorbed inside the cage, E ZIF is the energy of ZIF-8 without the presence of butanol and nE BuOH is the energy of n-butanol without the presence of the ZIF-8 cage. 7 Charge density analysis was carried out by computing the binding induced density difference utilising the CP2K tool "cubecruncher" to manipulate cube files. Powder X-ray diffraction data shown in Fig. S1(a)  Thermogravimetric analysis (TGA) carried out in air (Fig. S1(b) -ray diffraction (Fig. S2). The total observed weight losses are 64.3 ( 0.9) and 64.9 ( 0.5) for solvothermal and RT methods, respectively.
The slightly higher observed mass loss for the room temperature sample might be explained by small amounts of 2-methylimidazole starting material remaining in the sample, reflecting the lower onset decomposition temperature for this sample. Figure S2. Powder X-ray diffraction patterns collected with Co K α1 radiation (λ = 1.7889 Å) of the zinc oxide products obtained after thermogravimetric analysis of the ZIF-8 synthesised (i) solvothermally and (ii) by room temperature precipitation method. The hkl assignments are given for the reflections of zinc(II) oxide (ICSD 26170). 9 Elemental analysis data provided in Table S1 show that the compositions of ZIF-8 synthesised by both methods are very similar; with the sample synthesised at room temperature exhibiting only very slightly higher carbon, hydrogen and nitrogen contents. In both cases, the nitrogen content is lower than the theoretical value for ZIF-8 and slightly outside of the accepted error for this technique of 0.4 %. 10  Fig. S1(c) and Fig. S1(d) show characteristic bands of ZIF-8 for both samples with assignments made based on literature data; the spectra are almost identical for both samples. 11,12 The aromatic ν sym C-H is present at ca. 3135 cm -1 for both IR/R spectra, with an additional sharp band present at ca. 3115 cm -1 in the Raman spectra. The ν asym C-H of the methyl group is again present for both IR and R at ca. 2931 cm -1 . The broad band at 1583 cm -1 in both IR spectra is assigned to ν C-C and is Raman inactive. The Raman spectra have a relatively weak band at 1510 cm -1 that is also assigned to C-C stretching and is only weak in the IR spectra. The δ asym (CH 3  The total pore volume calculated for L-ZIF-8 is 0.66 cm 3 g -1 compared to 0.71 cm 3 g -1 ; the observed increase total pore volume is a result of a small contribution from interparticle meso-and macro-porosity (Fig S3(b)) present in the sample with small, uniform crystals.