Two-step micro cellular foaming of amorphous polymers in supercritical CO2

Microcellular foaming of amorphous rigid polymers, polymethylmethacrylate (PMMA) and polystyrene (PS) was studied in supercritical CO₂ (ScCO₂) in the presence of several types of additives, such as triblock (styrene-co-butadiene-co-methylmethacrylate, SBM and methylmethacrylate-co-butylacrylate-co-methylmethacrylate, MAM) terpolymers. This work is focused in the two-step foaming process, in which the sample is previously saturated under ScCO₂ being expanded in a second step out of the CO₂ vessel (e.g. in a hot oil bath) where foaming is initiated by the change of temperature near or above the glass transition temperature of the glass/polymer glassy system. Samples were saturated under high pressures of CO₂ (300 bar), at room temperature, for 16 h, followed by a quenching at a high depressurization rate (150 bar/min). In the last step, foaming was carried out at different temperatures (from 80 °C to 140 °C) and different foaming times (from 10 s to 120 s). It was found that cellular structures were controlled selecting either the additive type or the foaming conditions. Cell sizes are ranging from 0.3 μm to 300 μm, and densities from 0.50 g/cm³ to 1 g/cm³ depending on the polymers considered.     

Temperature influence and CO2 transport in foaming processes of poly(methyl methacrylate)–block copolymer nanocellular and microcellular foams

Fabricated by high-pressure or supercritical CO₂ gas dissolution foaming process, nanocellular and microcellular polymer foams based on poly(methyl methacrylate) (PMMA homopolymer) present a controlled nucleation mechanism by the addition of a methylmethacrylate–butylacrylate–methylmethacrylate block copolymer (MAM), leading to defined nanocellular morphologies templated by the nanostructuration of PMMA/MAM precursor blends. Influence of the CO₂ saturation temperature on the foaming mechanism and on the foam structure has been studied in 90/10 PMMA/MAM blends and also in the neat (amorphous) PMMA or (nanostructured) MAM polymers, in order to understand the role of the MAM nanostructuration in the cell growth and coalescence phenomena. CO₂ uptake and desorption measurements on series of block copolymer/homopolymer blend samples show a competitive behavior of the soft, rubbery, and CO₂-philic block of PBA (poly(butyl acrylate)) domains: fast desorption kinetics but higher initial saturation. This competition nevertheless is strongly influenced by the type of dispersion of PBA (e.g. micellar or lamellar) and a very consequent influence on foaming.CO₂ sorption and desorption were characterized in order to provide a better understanding of the role of the block copolymer on the foaming stages. Poly(butyl acrylate) blocks are shown to have a faster CO₂ diffusion rate than poly(methyl methacrylate) but are more CO₂-philic. Thus gas saturation and cell nucleation (heterogeneous) are more affected by the PBA block while cell coalescence is more affected by the PMMA phases (in the copolymer blocks + in the matrix).     

Solubility and diffusivity of CO2 in carboxylated polyesters

The solubility of CO₂ in saturated polyester resins at different temperatures (306 and 343 K) and pressures (0.1-30 MPa) has been measured using a magnetic suspension balance. The solubility data were used for estimating the binary diffusion coefficients. The results show a good solubility of CO₂ in polymers, up to 0.64 g CO₂/g polymer. The diffusion coefficients of supercritical CO₂ in polymers have generally high values and are in the range from 0.156 × 10⁻⁸ to 10.38 × 10⁻⁸ cm²/s. DSC and XRD analyses of the semi-crystalline polymer samples indicate that amorphous degree of polymers after exposure to CO₂ is increased. The observed structural effects are dependent on pressure, temperature and time of exposure to CO₂.     

Absorption of CO2 in high acrylonitrile content copolymers: dependence on acrylonitrile content

In continuation of our goal to determine the ability of CO₂ to plasticize acrylonitrile (AN) copolymers and facilitate melt processing at temperatures below the onset of thermal degradation, a systematic study has been performed to determine the influence of AN content on CO₂ absorption and subsequent viscosity reduction. Our previous report focused on the absorption of CO₂ in a relatively thermally stable 65 mol% AN copolymer. In this study, the ability for CO₂ to absorb in AN copolymers containing 85-98 mol% acrylonitrile was determined, and subsequent viscosity and equivalent processing temperature reductions were evaluated. Eighty five and 90 mol% acrylonitrile/methyl acrylate (AN/MA) copolymers were found to absorb up to 5.6 and 3.0 wt% CO₂, corresponding to reductions of T_g of 37 and 27 °C, and subsequent viscosity reductions of 61 and 56%, respectively. CO₂ absorption in these copolymers was found to occur immediately, in contrast to the time dependent absorption observed in the 65 mol% copolymer. An Arrhenius scaling analysis was used to determine the equivalent reductions in processing temperature resulting from the viscosity reductions, and reductions of up to 25 and 9 °C were observed for the 85 and 90 mol% AN copolymers. Based on the specific conditions used for absorption, no significant CO₂ uptake was observed for AN copolymers containing greater than 90 mol% acrylonitrile. Higher temperatures than those used here may be required to absorb CO₂ into AN copolymers containing greater than 90 mol% AN.     

Dispersion polymerization of 2-hydroxyethyl methacrylate stabilized by a hydrophilic/CO2-philic poly(ethylene oxide)-b-poly(1,1,2,2-tetrahydroperfluorodecyl acrylate) (PEO-b-PFDA) diblock copolymer in supercritical carbon dioxide

Dispersion polymerization of 2-hydroxyethyl methacrylate (HEMA) has been successfully performed in supercritical carbon dioxide at P=370 bar and T=65 °C with azobis(isobutyronitrile) as initiator and a hydrophilic/CO₂-philic poly(ethylene oxide)-b-poly(1,1,2,2-tetrahydroperfluorodecyl acrylate) (PEO-b-PFDA) block copolymer as steric stabilizer. The PEO-b-PFDA (2K/21K) block copolymer was synthesized by reversible addition-fragmentation chain transfer (RAFT) polymerization. Spherical particles of poly(HEMA) were obtained in the range of 200-400 nm diameter size with a narrow particle size distribution (D_w/D_n<1.1). The effect of the stabilizer concentration on the dispersion polymerization was investigated from 20 w/w% down to 3.5 w/w% versus HEMA. Precipitation polymerization in the absence of stabilizer lead to the formation of large aggregates of partially coalesced particles whereas discrete spherical particles of poly(HEMA) were obtained by dispersion polymerization even at low concentration of PEO-b-PFDA (3.5 w/w% versus HEMA).     

CO2-solubility of oligomers and polymers that contain the carbonyl group

Poly(vinyl acetate) (PVAc) is miscible with CO₂ over a broad range of molecular weights at 298 K. The cloud-point pressures needed to dissolve ∼5 wt% poly(methyl acrylate) (PMA) at 298 K are significantly greater than those needed to dissolve PVAc, even though a PMA repeat group has the same number of carbon, hydrogen, and oxygen atoms as in PVAc. This large difference in dissolution pressures is attributed to the lack of accessibility of the carbon dioxide to the carbonyl group in PMA. In addition, experimental data for poly(dimethyl siloxane) (PDMS) copolymers with readily accessible side groups suggest that an acetate group is slightly more CO₂-philic than an acrylate group. PVAc is more CO₂-soluble than other hydrocarbon homopolymers, including poly(propylene oxide) (PPO) and poly(lactide) (PLA). However, PVAc is significantly less miscible with CO₂ than PDMS and poly(fluoroalkyl acrylate) (PFA).     

Supercritical CO2 extraction of Helichrysum italicum: Influence of CO2 density and moisture content of plant material

Kinetics and selectivity of supercritical carbon dioxide (SC CO₂) extraction of Helichrysum italicum flowers were analyzed at pressures in the range of 10-20 MPa and temperatures of 40 °C and 60 °C (density of SC CO₂ from 290 to 841 kg/m³) and also at 10 MPa and 40 °C using flowers with different moisture contents (10.5% and 28.4%). Increased moisture content of H. italicum flowers resulted in enchased solubility of solute enabling decrease of SC CO₂ consumption necessary for achieving desired extraction yield. The most abundant compounds in the supercritical extracts are sesquiterpenes and waxes while monoterpenes and sesquiterpenes are the main constituents of essential oil obtained by hydrodistillation. The optimal set of working parameters with respect to extraction yield, SC CO₂ consumption and chemical composition of extract were defined related to moisture content of raw material and SC CO₂ density.     

Investigation of the nanocellular foaming of polystyrene in supercritical CO2 by adding a CO2‐philic perfluorinated block copolymer

Nanocellular foaming of polystyrene (PS) and a polystyrene copolymer (PS‐b‐PFDA) with fluorinated block (1,1,2,2‐tetrahydroperfluorodecyl acrylate block, PFDA) was studied in supercritical CO₂ (scCO₂) via a one‐step foaming batch process. Atom Transfer Radical Polymerization (ATRP) was used to synthesize all the polymers. Neat PS and PS‐b‐PFDA copolymer samples were produced by extrusion and solid thick plaques were shaped in a hot‐press, and then subsequently foamed in a single‐step foaming process using scCO₂ to analyze the effect of the addition of the fluorinated block copolymer in the foaming behaviour of neat PS. Samples were saturated under high pressures of CO₂ (30 MPa) at low temperatures (e.g., 0°C) followed by a depressurization at a rate of 5 MPa/min. Foamed materials of neat PS and PS‐b‐PFDA copolymer were produced in the same conditions showing that the presence of high CO₂‐philic perfluoro blocks, in the form of submicrometric separated domains in the PS matrix, acts as nucleating agents during the foaming process. The preponderance of the fluorinated blocks in the foaming behavior is evidenced, leading to PS‐b‐PFDA nanocellular foams with cell sizes in the order of 100 nm, and bulk densities about 0.7 g/cm³. The use of fluorinated blocks improve drastically the foam morphology, leading to ultramicro cellular and possibly nanocellular foams with a great homogeneity of the porous structure directly related to the dispersion of highly CO₂‐philic fluorinated blocks in the PS matrix. © 2012 Wiley Periodicals, Inc. J Appl Polym Sci, 2012     

Supercritical CO2-assisted preparation of ibuprofen-loaded PEG-PVP complexes

Stoichiometric ratios of poly(ethylene glycol) (PEG, M_w = 400) with poly(vinylpyrrolidone) (PVP, M_w = ±3.1 × 10⁴ and M_w = 1.25 × 10⁶ M_w) were prepared from ethanol cast solutions and in supercritical CO₂. The complex formation was studied via glass transition (T_g) analysis obtained from differential scanning calorimetry (DSC) thermograms. PEG-PVP blends were also loaded with ibuprofen. The molecular dispersion of ibuprofen, mechanism of interaction, the effect of CO₂ pressure and temperature and ageing of blends were also analysed with DSC, attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy and X-ray diffraction (XRD). T_g analysis indicated that supercritical CO₂ can facilitate the formation of stoichiometric PEG-PVP complexes. Processing of PEG-PVP blends with ibuprofen results in the molecular dispersion of ibuprofen mainly bonded to PVP carbonyl groups, without significant disruption of the PEG-PVP complex. Increasing process pressure results in extraction of some PEG fractions. Post-processing ATR-FTIR shifts in ibuprofen-PEG-PVP complexes is greater with supercritical CO₂ processing. These shifts are mainly attributed to atmospheric moisture absorption. Overall it was shown that, ibuprofen-loaded PEG-PVP complexes can be prepared from supercritical CO₂ processing showing similar characteristics to such complexes prepared from solution casting.     

Preparation of nanometer dispersed polypropylene/polystyrene interpenetrating network using supercritical CO2 as a swelling agent

Nanometer dispersed polypropylene/polystyrene (PP/PS) interpenetrating networks (IPNs) have been prepared by the radical polymerization and crosslinking of styrene (St) within supercritical (SC) CO₂-swollen PP substrates. In this method, monomer St, crosslinking agent divinyl benzene (DVB), and the initiator benzoyl peroxide were first impregnated into PP matrix using SC CO₂ as a solvent and swelling agent at 35.0 °C, and then the polymerization and crosslinking were carried out at 120 °C. The composition of the IPNs can be controlled by SC CO₂ pressure, concentrations of St and DVB in the fluid phase. Transmission electron microscopy shows that the PS is homogeneously dispersed in the IPNs and its phase size is in the range of 20-30 nm. The impact strength, tensile strength, and elongation-at-break of the PP/PS IPNs increase with increasing PS percentage in the IPNs.     

Solid-state supercritical CO2 foaming of PCL and PCL-HA nano-composite: Effect of composition, thermal history and foaming process on foam pore structure

In this work we investigated the solid-state supercritical CO₂ (scCO₂) foaming of poly(?-caprolactone) (PCL), a semi-crystalline, biodegradable polyester, and PCL loaded with 5 wt% of hydroxyapatite (HA) nano-particles.In order to investigate the effect of the thermal history and eventual residue of the crystalline phase on the pore structure of the foams, samples were subjected to three different cooling protocols from the melt, and subsequently foamed by using scCO₂ as blowing agent. The foaming process was performed in the 37-40 °C temperature range, melting point of PCL being 60 °C. The saturation pressure, in the range from 10 to 20 MPa, and the foaming time, from 2 to 900 s, were modulated in order to control the final morphology, porosity and pore structure of the foams and, possibly, to amplify the original differences among the different samples.The results of this study demonstrated that by the scCO₂ foaming it was possible to produce PCL and PCL-HA foams with homogeneous morphologies at relatively low temperatures. Furthermore, by the appropriate combination of materials properties and foaming parameters, we prepared foams with porosities in the 55-85% range, mean pore size from 40 to 250 μm and pore density from 10⁵ to 10⁸ pore/cm³. Finally, we also proposed a two-step depressurization foaming process for the design of bi-modal and highly interconnected foams suitable as scaffolds for tissue engineering.     

Polymeric nanoparticles from macroscopic crystalline monomers by facile solid-state polymerization in supercritical CO2

When macroscopic crystalline monomers were polymerized by a free-radical solid-state reaction in the presence of supercritical CO₂ (scCO₂), the resultant products were found to be composed of unexpected nanoparticle morphologies. In particular, the solid-state polymerization (SSP) of amino acid based monomers, acryloyl-β-alanine (ABA) and methacryloyl-β-alanine (MBA), initiated by azobisisobutyronitrile in scCO₂ (at 65 °C and 34.5 MPa), produced corresponding polymers having aggregated spherical architectures. The average diameters of the PABA and PMBA particles were measured to be 94 and 102 nm, respectively. In addition, high molecular weight polymers (PMBA, M_w = 3.8 × 10⁵ g/mol) with a high yield (∼96%) were obtained. The microscopic investigation revealed that a unique particle formation mechanism was involved in the SSP in which large sized crystalline monomers were chipped into small pieces during the initial stage of polymerization and subsequently converted into nanoscale objects after 24 h.     

Controlling the foam morphology of supercritical CO2-processed poly(methyl methacrylate) with CO2-philic hybrid nanoparticles

Highly CO₂-philic nanoparticles, octatrimethylsiloxy polyhedral oligomeric silsesquioxanes (POSS) are used to increase the affinity of poly(methyl methacrylate) (PMMA) to CO₂ in supercritical carbon dioxide (scCO₂) foaming, thus to improve its foaming performance and the foam morphology. PMMA and PMMA-POSS composite foams were produced based on the two-factorial design, at the upper and lower experimental conditions of pressure, temperature, processing time, and venting rate. The foams of PMMA-5% POSS composites exhibited smaller average pore sizes and higher pore densities than neat PMMA and PMMA-0.5% POSS composites. The smallest average pore diameter (0.3 μm) and the highest pore density (6.33 × 10¹² cm⁻³) were obtained with this composite processed at 35°C, 32 MPa, for 24 h and depressurized with fast-venting rate (0.4 MPa/s). ScCO₂ processing decreased the density of the polymer by more than 50%.     

Quantification of specific interactions between CO2 and the carbonyl group in polymers via ATR-FTIR measurements

We have used in situ ATR-FTIR measurements to provide estimates of the strength of specific interactions between carbon dioxide (CO₂) and carbonyl groups in polymers such as poly(methyl methacrylate) (PMMA), poly(vinyl acetate) (PVAc), poly(lactide) (PLA) and poly(lactide-co-glycolide) (PLGA). Polymer films were exposed to high pressure CO₂ and the carbonyl stretching vibration at 1700 cm⁻¹ and the CO₂ bending mode at 660 cm⁻¹ were studied. The observed shift in the carbonyl stretching band to higher wavenumber was attributed to dielectric effects according to the Kirkwood-Bauer-Magat (KBM) equation. On the other hand, the splitting of the CO₂ bending mode provided direct evidence of specific interactions between the polymer and CO₂. These interactions were quantified via an equilibrium constant for the association reaction between CO₂ and the carbonyl group.     

Absorption of CO2 and subsequent viscosity reduction of an acrylonitrile copolymer

Acrylonitrile (AN) copolymers (AN content greater than about 85 mol%) are traditionally solution processed to avoid a cyclization and crosslinking reaction that takes place at temperatures where melt processing would be feasible. It is well known that carbon dioxide (CO₂) reduces the glass transition temperature (T_g) and consequently the viscosity of many glassy and some semi-crystalline thermoplastics. However, the ability of CO₂ to act as a processing aid and permit processing of thermally unstable polymers at temperatures below the onset of thermal degradation has not been explored. This study concentrates on the ability to plasticize an AN copolymer with CO₂, which may ultimately permit melt processing at reduced temperatures. To facilitate viscosity measurements and maximize the CO₂ absorption, a relatively thermally stable, commercially produced AN copolymer containing 65 mol% AN was investigated in this research. Differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) indicated that CO₂ significantly absorbs into and reduces the T_g of the AN copolymer. Pressurized capillary rheometry indicated that the magnitude of the viscosity reduction is dependent on the amount of absorbed CO₂, which correlates directly to the T_g reduction of the plasticized material. Up to a 60% viscosity reduction was obtained over the range of shear rates tested for the plasticized copolymer containing up to 6.7 wt% CO₂ (31 °C T_g reduction), corresponding to as much as a 30 °C equivalent reduction in processing temperature. A Williams-Landel-Ferry (WLF) analysis was used to estimate the viscosity reduction based on the T_g reduction (and corresponding amount of absorbed CO₂) in the plasticized AN copolymer, and the predicted viscosity reduction based on using the universal constants was 34-85% higher than measured, depending on the amount of absorbed CO₂. Van Krevelen's empirical solubility relationships were used to calculate the expected absorbance levels of CO₂, and found to be highly dependent on the choice of constants within the statistical ranges of error of the Van Krevelen relationships.     

Partitioning of drug model compounds between poly(lactic acid)s and supercritical CO2 using quartz crystal microbalance as an in situ detector

Quartz crystal microbalance (QCM) was used as an in situ detector to investigate the potential application in the phase equilibrium determination of supercritical CO₂-drug-polymer systems. CO₂ solubility in two biodegradable polymers, poly(d,l-lactic acid) (d,l-PLA) and poly(l-lactic acid) (l-PLA) was primarily measured at 313.15 K and pressures up to 10.0 MPa. d,l-PLA showed a better CO₂ absorption ability due to its amorphous structure. Four drug model compounds of poor solubility in water, ibuprofen, aspirin, salicylic acid and naphthalene were selected as representatives for the examination of drug uptake in PLA matrices, as well as partition coefficient during supercritical impregnation. It was found that partition coefficients of drugs can reach as high as 10³-10⁴ orders of magnitude and greatly affected by the intermolecular interactions between drugs and PLA. Aspirin exhibited the best partitioning during the supercritical impregnation at pressures of 8.0-10.0 MPa due to the existence of carboxylic acid and acetyl groups. Drug partitioning is additionally related to the drug concentration in ScCO₂, i.e. salicylic acid showed little absorption in PLA according to its poor solubility in ScCO₂ at 7.5-8.0 MPa, whereas the well CO₂-soluble compound, naphthalene, exhibited a moderate partition coefficient although its polarity was different from l-PLA.     

Foaming behavior of isotactic polypropylene in supercritical CO2 influenced by phase morphology via chain grafting

Polystyrene (PS) and poly(methyl methacrylate) (PMMA) grafted isotactic polypropylene copolymers (iPP-g-PS and iPP-g-PMMA) with well-defined chain structure were synthesized by atom transfer radical polymerization using a branched iPP (iPP-B) as polymerization precursor. The branched and grafted iPP were foamed by using supercritical CO₂ as the blowing agent with a batch method. Compared to linear iPP foam, the iPP-B foams had well-defined close cell structure and increased cell density resulted from increased melt strength. Further incorporating PS and PMMA graft chains into iPP-B decreased the crystal size and increased the crystal density of grafted copolymers. In iPP-g-PS foaming, the enhanced heterogeneous nucleation by crystalline/amorphous interface further decreased the cell size, increased the cell density, and uniformized the cell size distribution. In contrast to this, the iPP-g-PMMA foams exhibited the poor cell morphology, i.e., large amount of unfoamed regions and just a few cells distributed among those unfoamed regions, although the crystal size and crystal density of iPP-g-PMMA were similar to those of iPP-g-PS. It was found that the iPP-g-PMMA exhibited PMMA-rich dispersed phase, which had higher CO₂ solubility and lower nucleation energy barrier than copolymer matrix did. The preferential cell nucleation within the PMMA-rich phase or at its interface with the matrix accounted for the poor cell morphology. The different effect of phase morphology on the foaming behavior of PS and PMMA grafted copolymers is discussed with the classical nucleation theory.     

CO2 gas-activated sintering of carbonated hydroxyapatites

Sintering compacts of carbonated hydroxyapatite (CHA) nanoparticles (3.4 wt% CO₃²⁻) in a CO₂ flow (4 mL/min) proceeded at a temperature which was more than 200 °C lower than that for hydroxyapatite in air (1150 °C). During heating from RT to 1200 °C (5 K/min) the rate of shrinkage of the CHA compacts showed a maximum thrice as high as that in air at about 929 °C. The shrinkage correlates with a mass loss caused by the release of CO₂ due to the thermal decomposition of CO₃²⁻ ions that substitute PO₄³⁻ ions in the CHA lattice. Firing the compacts in the CO₂ flow at 800 and 900 °C for 2 h resulted in an additional carbonatation on the B-sites and a further decrease in the sintering temperature to 890 °C. The compacts fired in the 900-1000 °C range became almost complete ceramics with high densities and mechanical properties close to those of medical implants. Firing at temperatures above 1000 °C resulted in an additional carbonatation on the A-sites. However, this led to a material with low densities and poor mechanical properties. A supposition has been proposed that the effect of CO₂ gas-activated sintering is a result of the intensification of the diffusion in the nanoparticles caused by CO₂ molecules entering the bulk from the CO₂ atmosphere and (or) releasing from the bulk due to the decomposition of carbonates on the B-sites in the lattice.     

Investigation of the degree of homogeneity and hydrogen bonding in PEG/PVP blends prepared in supercritical CO2: Comparison with ethanol-cast blends and physical mixtures

The degree of homogeneity and H-bond interaction in blends of low-molecular-mass poly(ethylene glycols) (PEG, M_w = 400, 600, 1000) and poly(vinylpyrrolidone) (PVP, M_w = 9 × 10³) prepared in supercritical CO₂, ethanol and as physical mixtures were studied by differential scanning calorimetry (DSC), Fourier-transform infrared (FTIR) spectroscopy and dynamic mechanical analysis (DMA) techniques. Homogeneity of samples prepared in supercritical CO₂ were greater than physically mixed samples, but slightly less than ethanol-cast samples. PEG-PVP H-bond interaction was higher for ethanol-cast blends when compared to blends prepared in supercritical CO₂. This reduced interaction was attributed to a combination of: (1) shielding of PEG-PVP H-bond interactions when CO₂ is dissolved in the blend; (2) rapidly reduced PEG and PVP chain mobility upon CO₂ venting, delaying rearrangement for optimum PEG-PVP H-bond interaction.     

Theoretical study on interaction between CO2 and carbonyl compounds: Influence of CO2 on infrared spectroscopy and activity of CO

The interactions between CO₂ and carbonyl compounds at different CO₂ pressures have been studied both experimentally and theoretically. In situ high-pressure FTIR on carbonyl compounds, i.e., acetaldehyde, acetone, and crotonaldehyde, in supercritical CO₂ have been measured at various CO₂ pressures varying from 6 to 22 MPa. In order to get insights into the mechanism, theoretical study has been conducted concerning the effect of CO₂ on frequency shift of CO in acetaldehyde, acetone, benzaldehyde, crotonaldehyde and cinnamaldehyde at different CO₂ pressures. It has been shown that the experimental frequency shifts can be well simulated by the theoretical model calculations using particular structures, in which a carbonyl compound interacts with a few CO₂ molecules, depending on the carbonyl compounds examined, except for acetaldehyde.The interaction energies between CO₂ and those carbonyl compounds are also given. In addition, the effect of CO₂ on hydrogenation of crotonaldehyde and benzaldehyde has been discussed by means of the local softness (s⁺) calculated at CO₂ pressures of 0-22 MPa, which can explain the reactivity difference in the crotonaldehyde and benzaldehyde hydrogenations in supercritical CO₂.