Supporting Information Supercritical Carbon Dioxide as Reaction Medium for Selective Hydrogenation of Fluorinated Arenes

The selective hydrogenation of fluorinated arenes with polar functionalities to valuable fluorinated cyclohexanes is achieved using supercritical CO2 as solvent in combination with Rh NPs immobilized on molecularly modified silica as catalysts.


Reactors and reactor inlets
The catalytic tests involving scCO 2 were performed using an in-house engineered 30 mL autoclave (Alloy C-276, material No. 2.4819, T max = 200°C, P max = 400 bar), which was equipped with a Teflon inlet and a Teflon cap. (The actual volume including the upper part is 34 mL measured by CO 2 and the volume excluding the Teflon inlet, Teflon cap and magnetic stirrer is ca. 24.5 mL). Electronic Supplementary Material (ESI) for Green Chemistry. This journal is © The Royal Society of Chemistry 2022 stirred for 1 h at room temperature under argon atmosphere and the support color changed from white to yellow/brown. The solvent was carefully removed under reduced pressure and the impregnated Support was transferred to an autoclave. The autoclave was pressurized with 50 bar H 2 (at r.t.) and heated to 100 °C for 18 h. A grey/black powder was obtained, indicating the formation of NPs. Theoretical metal loading = 0.1 mmol Rh/g Support.

Catalytic reactions
In a typical experiment, 5 mg of catalyst (1 wt.% Rh), CaO (typically 4.5 mol% as compared to the substrate) and the chosen mass of substrate were weighed in an antistatic weighing boat and transferred into a Teflon inlet. Covering one component with the others should be avoided. Finally, tetradecane was added as an internal standard directly into the Teflon inlet. The autoclave was first pressurized with 55 bar H 2 (controlled by a digital pressure meter), and then 12.3 g CO 2 (≙ 0.5 g mL -1 ) was added by gravimetric dosage using a balance and a high pressure compressor. Hereby it is important to control the gas flow rate in a very careful manner to avoid spreading the chemicals placed inside the Teflon inlet into the reactor parts. Purging of the pipes with the corresponding gas turned out to be important. The reaction mixture was heated to the reaction temperature while stirring at 500 rpm. After the reaction, the autoclave was cooled down in an ice bath and depressurized very slowly (ca. 45 min-60 min). The organic part inside the autoclave was extracted with acetone. (Sometimes, the gas phase was bubbled through 2 mL acetone solvent, which was cooled with a dry ice-acetone bath of -50 to -60°C to check if there was any organic residue in the CO 2 phase, this process usually lasted ca. 8 hours). Acetone solutions were analyzed separately by GC-FID and GC-MS. (There were nearly no organic compounds detected in the collected gas phase solution in all cases.). Selectivity and yields are given as average values with standard deviations from series of n experiments (n = 3 to 25).
The use of fresh well sealed Teflon inlets (see Figure S1 b) was found beneficial to avoid spreading material out of the container while pressurizing, as this may prevent catalyst and substrates from being in contact which might cause possible fluctuations in the reaction outcome. It is thus also advised to perform the flushing and depressurizing processes with extreme caution when using similar reactor setups to the one described here. We note that using reactors made from Hastelloy steel were found to be possible alternatives without the need for the Teflon inlets, albeit the formation of even small amounts of HF must be considered in appropriate safety measures.

Analytics
Gas chromatography (GC) was performed on a Shimadzu GC2030 equipped with an FID-detector. Gas chromatography coupled with a mass spectrometer (GC-MS) were performed on a Shimadzu QP2020. The determination of the yields was done by injecting the reaction mixture into the GC. The product identifications was achieved by injecting the pure products or by GC-MS.
Different conditions and equipment settings were used for the different substrates.

Synthetic approach evaluation
4-fuorocyclohexan-1-ol was selected as a target product for this evaluation. The selective hydrogenation approach we propose in this study was systematically compared to a typical conventional method. The detailed synthetic routes are shown in Figures S9 and S10. For both synthetic approaches, the starting point is a widely available and cheap commercial compound, which preparation is thus not included in the evaluation.
From the green chemistry principles, five parameters were chosen to rank the pathways, i.e. the number of steps (Steps), the atom economy (AE), the overall reaction yield (Y), the hazardous nature of the reagents (Safety) and the economical aspect (difference of value between the product and the starting substrate Eco).
The AE was determined using the following formula: The Safety parameter was evaluated qualitatively by ranking the different pathways on a scale from one (= most hazardous) to five (= least hazardous) based on the hazardous nature [3] of the used chemicals as shown in Table S3.
The parameter Eco is based on the difference in price between the starting materials and the desired product. Our approach provides a better difference and the Eco parameter is set arbitrarily to 5. For the conventional pathway, the addition of value is still attractive, and Eco was thus set to 4.
Parameters Y and Steps are elucidated on Figures S9 and S10.