Cucurbit[7]uril-based high-performance catalytic microreactors

Fabrication of cucurbit[n]uril-based catalytic microreactors through the immobilisation of metallic nanoparticles onto microchannels.

The Agilent Eclipse Plus C18 5 µm 4.6 x 150 mm column was used to analyse the resultant products.

S.1.2 Preparation of metallic catalyst nanoparticles
The metallic catalyst NPs were synthesized by rapid reduction of corresponding high valence metallic acids/salts in the presence of sodium borohydride. The Au NPs were synthesized by firstly mixing 1 wt% sodium citrate (2 mL) with 1 wt% chloroauric acid (1 mL) in 100 mL water, stirring and cooling in an ice bath for 10 min. Then 0.1 M sodium borohydride (10 mL, pre-cooled) was added rapidly, stirring for 1 h. Similarly, the Pd NPs were synthesized by the rapid addition of sodium borohyride (0.1 M, 4 mL, pre-cooled) to the cooled mixtures of 1 wt% sodium citrate (2 mL) with 10 mM sodium tetrachloropalladate (1 mL) in 100 mL water.

S.1.4 Preparation of PDMS microfluidic devices
The microfluidic device was produced via standard soft lithography by pouring poly(dimethylsiloxane) (PDMS, 20 g) along with a crosslinker (Sylgard 184 elastomer kit, Dow Corning, pre-polymer : crosslinker = 10 : 1 by weight) onto a silicon wafer patterned with SU-8 photoresist. [2][3][4] It was then placed in vacuum for half an hour to remove dissolved gas. The PDMS was allowed to solidify at 90 • C for 12 h before being peeled off, after which inlets and outlets were generated using a biopsy punch. The enclosed microfluidic channels were formed by attaching the moulded PDMS replica onto a clean glass slide after exposure to oxygen plasma for 10 s in a Femto plasma cleaner.

S3
The microchannels could be fabricated to various shapes and length ( Figure S2 as examples), to satisfy the requirement of certain catalytic reactions, such as various inlets or gas protection. In this work, the PDMS microchannels with 40 µm diameter and 3 cm length was utilized. Figure S2: Pictures of example PDMS microchannels with various channel shapes and length.

S.1.5 Preparation of CB[7]-based catalytic microreactors
The CB[7]-based Au NPs catalytic microreactors were prepared as follows: (1) The blank PDMS microfluidic channels were activated by oxygen plasma for 10 s, after which the MV-silane@CB[7] solution (2.5 mM MV-silane, 1 mM CB[7], in 50 : 50 water : ethanol solvent) was injected and flowed through the microchannel at a flow rate of 300 µL h −1 for 2 h, followed by washing with water for 1 h at the same rate.
(2) The prepared Au NPs solution was injected and flowed through the microchannel at a rate of 200 µL h −1 for 1 h, followed by washing with water for 1 h at the same rate.
The CB[7]-based Pd NPs catalytic microreactors were prepared in the same approach, injecting the prepared Pd NPs solution in step (2) instead. Note that all the solutions were filtered using 200 nm filter tips before injecting into microchannels to avoid the blockage of the microchannels.
Control channels (A and B) were prepared by solely injecting MV-silane or CB[7] solutions in step (1).

S.2.3 SEM characterisation of microreactors
As seen in Figure S5       respectively. An increase of the loading NP density using increased concentration was observed. S10  catalyst instead. f The first run gave rise to 60% yield; however, later reaction cycles quickly decreased to 0% yield. g The yield of three Pd NPs reactions were determined from HPLC analysis.

S. 3.2.1 Reduction of nitrobenzene by NaBH 4
6 mL deionised water, 0.5 mL 1 mM nitrobenzene and 2 mL 50 mM NaBH 4 were mixed, injected and flowed through the microreactor, control channel (A) and blank microchannel, all at a flow rate of 200 µL h −1 . The bench reaction was carried out at the same experimental condition, using Au NP (6.8 ± 2.1 nm, 1 wt% 100 µL ) as the catalyst instead. The resultant products were analysed using UVvis spectroscopy. Yield was deduced from the reactant absorption peak (nitrobenzene, 270nm). The microreactor can be operated continuously for 300 h without detectable loss of the yield and TOF. Figure S13: UV-vis spectra of (a) Au NP catalytic microreactor before and after reaction, yielding 99% after 5 min. (b) Au NP control channel: the first run gave 50% yield (the blue dashed line), then subsequent reaction cycles gave 0% yield (the pink dashed line) likely due to severe leakage of the gold agglomerated particles as a function of salts presented in the reaction. (c) The bench reaction with Au NPs led to a yield of 50% after 5 min and 99% after 15 min. No further increase of the product peak (ii) was observed after 15 min. The peak at 530 nm was caused by the remaining Au NPs in the reaction solution. The Au NPs used for the bench reaction were difficult to be separated and recycled for further use.

S. 3.2.2 Reduction of nitrophenol by NaBH 4
Similar to previous reaction, the products were analysed using UV-vis spectroscopy. Figure S14: UV-vis spectra of (a) Au NP catalytic microreactor before and after reaction, yielding product in 99% after 5 min. (b) Au NP control channel: the first run gave 60% yield (the blue dashed line); however, the yield quickly decreased to 0% in subsequent reaction cycles (the pink dashed line). (c) The bench reaction with Au NPs led to a yield of 60% after 5 min and 99% till 20 min. The Au NPs used for the bench reaction were difficult to be separated and recycled for further use.

S. 3.3.2 Suzuki reaction 2
Similar to previous reaction, the products were analysed using HPLC. The yield was deduced from HPLC calibration line of the reactant (4-iodobenzoic acid). Figure S19: HPLC spectra of (a) reactants (dashed lines) and after reaction in Pd NP catalytic microreactor for 30 min, leading to 99% yield (solid line). (b) The Pd NP control channel did not produce any product after 30 min. Only the peaks of the reactants were observed. (c) The bench reaction with Pd NPs gave 50% yield after 30 min, however, with various byproducts (peaks at 17 and 27 min). The Pd NPs used in the bench reaction were difficult to be separated and recycled for further use.
The reaction mixture was injected and flowed through the preheated microreactor and control channel (A) at a flow rate of 100 µL h −1 , heating and reacting at 60 • C for 2 h. The bench reaction was carried out using Pd NP (3.7 ± 0.8 nm, 1 wt% 100 µL) to catalyse 2 mL reaction mixture at 60 • C for 4 h. The resultant products were analysed using HPLC. The yield was deduced from HPLC calibration line of the reactant (1-iodo-4-nitrobenzene). The microreactor can be operated continuously for 300 h without detectable loss of the yield and TOF. S17 Figure S20: HPLC spectra of (a) reactants (dashed lines) and after reaction in Pd NP catalytic microreactor for 2 h, leading to 85% yield (solid line). (b) The Pd NP control channel did not produce any product after 2 h. The peaks remained the same with the reactants. (c) The bench reaction with Pd NPs gave 10% yield after 4 h. Mole of catalyst = 4.14 x 10 9 ÷ N A = 6.89 x 10 −15 So TOF = (3.52 ± 0.04) × 10 −10 mol 6.89 × 10 −15 × 300s = 171 ± 2