Adsorption and controlled release of three kinds of flavors on UiO‐66

Abstract Delivery systems for controlled release of fragrances are significantly essential in the flavor and fragrance industry due to a limited life span (premature evaporation and degradation) of fragrance compounds. Recently, several adsorption materials such as porous materials have been developed in delivery systems for targeted fragrance release. In this work, UiO‐66, a member of metal–organic framework (MOF) family with high porosity and greater adsorbability, was selected as a prospective alternative to traditional porous adsorbents for controlled release of fragrances. Isophorone, eugenol, and β‐ionone with strong aroma are widely used as perfume flavors, soap flavor, cosmetic flavors, and even as a food‐flavoring agents, and were chosen as representative fragrances for adsorption and controlled release studies. The adsorption and release behavior of fragrances on UiO‐66 was evaluated by high‐performance liquid chromatography (HPLC). The UiO‐66 with high surface area (1,076 m2/g) achieved effective storage and controlled release for isophorone, eugenol, and β‐ionone. The adsorption rates of isophorone, eugenol, and β‐ionone can reach 99.4%, 99.9%, and 60.2%, respectively. Additionally, the release of these fragrances from UiO‐66 can sustain over 20 days. UiO‐66 exhibited higher release rate over eugenol with desorption rates of 95.2% than that of β‐ionone (52.6%) and isophorone (49.6%), respectively, suggesting a good adsorption‐release selectivity of UiO‐66 to different fragrances. This study further confirms the usability of UiO‐66 in fragrance release and extends the application of MOF porosity in aroma release.

Metal-organic frameworks (MOFs; He, Chen, Lü, & Liu, 2014;Lin, Liu, & Chen, 2016;Qiu, Feng, Zhang, Jia, & Yao, 2017) are an emergent class of porous materials consist of organic-inorganic structure, which shows more advantages than traditional support materials due to their unique hybrid structures (high surface area and porosity) and relatively milder synthesized conditions (in terms of solvent, pH, and temperature; Li et al., 2014;Mu, Jiang, Chao, Lou, & Chen, 2018;Yang et al., 2018). First, MOFs are compositionally and structurally diverse, allowing for the facile synthesis of MOFs of different compositions, shapes, sizes, and chemical properties.
Second, MOFs are intrinsically biodegradable as a result of relatively labile metal-ligand bonds, making it possible to rapidly degrade and clear the nanocarriers after the intended task is completed (Rocca, Liu, & Lin, 2011). Though the moderately low chemical and aqueous stability of MOFs has limited their scope for industrial applications, this drawback of MOFs is considered an advantage for delivery system in light industry or medicine applications, as the MOF particles can be biodegraded and eliminated from the body after the compounds are released (Orellana-Tavra et al., 2016). At present, there are some studies of MOFs used as nanocarriers for drug delivery (ibuprofen [Erucar & Keskin, 2016], alendronate [Golmohamadpour, Bahramian, Shafiee, & Mamani, 2018], doxorubicin hydrochloride [Bhattacharjee et al., 2018]), VOC removal (Xu et al., 2018;Zhang, Lv, Shi, Yang, & Yang, 2019;Zhu, Hu, Tong, Zhao, & Zhao, 2017), and sustained release of small molecule dyes  and cosmetic molecules (caffeine and urea; Erucar & Keskin, 2016). With recent success in assembling edible MOFs (CD-MOFs) using γ-cyclodextrin and potassium ion (Smaldone et al., 2010), it is expected that MOFs could be applied in the food industries in the near future (Wu et al., 2019). Most recently, MOFs have been investigated as novel allyl isothiocyanate (AITC) carrier for food safety and food industry applications, which allow the successful implementation of AITC into food systems by improving its stability while eliminating or reducing its negative organoleptic impact on various food matrices (Wang et al., 2016). One well-recognized challenge of MOFs is their poor aqueous stability, limiting their scope for practical applications.
However, this drawback of MOFs is considered an advantage for de-
Cyclohexane (AR) was obtained from Tianjin Fengchuan Chemical Reagent Technologies Co., Ltd. Isophorone, eugenol, and β-ionone (>98%) were provided by Yunnan Industrial of China Tobacco Industry Co., Ltd. The physical and chemical properties of the fragrance compounds are shown in Table S1.

| Preparation of UiO-66
The synthesis of UiO-66 followed the procedure reported by Silva, Luz, Xamena, Corma, and Garcia (2010) with slight modification. 0.625 g of ZrCl 4 and 0.615 g of 1,4-benzenedicarboxylic acid (H 2 BDC) were dissolved in 50 ml dimethylformamide (DMF). The solution was transferred into a stainless steel autoclave with Teflon-lined and heated at 120°C for 24 hr. When the autoclave was naturally cooled to room temperature, the solid was filtered out and washed with ethyl alcohol and then dried at 90°C. The dried product was impregnated with DMF for 6 hr, then filtered, washed three times with ethanol, and dried at 90°C. The solid was then transferred into a stainless steel autoclave with Teflon-lined and alcoholized with ethanol at 90°C for 5 hr, centrifuged, and washed with ethanol and acetone to obtain a solid product.
The final product was dried at 90°C.

| Characterizations
Nitrogen adsorption-desorption was taken on a NOVA 2000e gas sorption analyzer (Quantachrome Corp.) to determine the pore-size distributions, BET surface areas, and pore volumes. N 2 sorption isotherms were obtained at 77 K using a Micromeritics ASAP 2460.
Thirty milligram of the sample was activated at 140°C under primary vacuum overnight. The total pore volume and pore-size distribution were evaluated by the NLDFT (nonlocal density functional theory) method from the adsorption branch of the isotherm and using a cylindrical model. XRD experiments were conducted on a D/max-3B spectrometer with Cu Kα radiation, and scans were made in the 2θ range 0.1-5° with a scan rate 0.05°/min (low-angle diffraction), and in the 2θ range 5-90° with a scan rate of 10°/min (wide-angle diffraction). SEM analysis was taken on a FEIQuanta200FEG microscope with an accelerating voltage of 15 kV to observe the morphology of UiO-66. FTIR spectra were recorded using a Thermo Nicolet 8700 instrument in the range of 4,000-400 cm −1 . TGAs were carried out on a Mettler Toledo TGA/DSC/1600LF in temperature range 25-800°C with a 10°C/min heating rate under an N 2 atmosphere (20 ml/min).

| Quantitative evaluation by HPLC
Considering the accurate quantitative analysis of low concentration of fragrance solution, HPLC was selected to determine all fragrance solution concentrations calculated by corresponding peak areas. HPLC analysis was conducted on an Agilent 1260 Infinity instrument at 30°C with a 150 × 4.6 mm column (5 μm particles).

| Adsorption of fragrances in UiO-66
The adsorption experiments of UiO-66 toward isophorone, eugenol, and β-ionone were carried out at room temperature.
Typically, 100 ppm of spice solution was dispersed in 15 ml of solvents (ethanol and cyclohexane). Subsequently, different amounts The fragrance adsorption rates were calculated using the following equations: where C 0 and C t represent the fragrance concentrations (mg/L) at initial time t, respectively.

| Controlled release of fragrances in UiO-66
One hundred fifty milligram of UiO-66 was dispersed into 150 ml of cyclohexane solution containing 100 ppm of fragrances with stirring for 2 hr, filtered, and naturally dried to obtain the precipitation.
Aliquots (10 mg) of the solid precipitation were taken and redispersed into 5 ml of ethanol solution in certain time intervals (1 hr), and the intervals were adjusted to lager intervals (24 hr) after 24 hr.
The mixed solution was centrifuged and washed with 5 ml of ethanol for three times, and then, 15 ml of supernatant liquid was totally obtained and filtered with a needle (2.2 μm) and then examined by HPLC. The released capacity of UiO-66 was evaluated by remained fragrance amount or percent remained fragrance in adsorbent after elution calculated as follows: (1) Adsorption rate (%) = C 0 − C t ∕C 0 × 100 where M 0 (mg) is the initial mass of fragrances in the solution and M f (mg) is the mass of released fragrances in the final supernatant. M (g) represents the mass of the UiO-66 before loading fragrances.

| Characterizations of UiO-66
The structural properties of obtained UiO-66 were examined by N 2 adsorption-desorption, powder X-ray diffraction (XRD), and scanning electron microscopy (SEM), and the results were summarized in the supplementary. The XRD pattern of UiO-66 is shown in Figure S4.
The characteristic diffraction peaks appeared in 2θ between 3° and 60°, which were consistent with that in reported XRD pattern (Abid et al., 2012). The adsorption-desorption isotherm ( Figure S5a) and pore-size distribution ( Figure S5b) of UiO-66 implied that UiO-66 exhibited a typical type I isotherm (IUPAC classification), dominated by a large number of micropores range from 2 to 10 Å. The BET surface area, pore volume, and pore size of UiO-66 (Table S5) are 1,076 m 2 /g, 0.12 cm 3 /g, and 6.7 nm, respectively. The SEM images in Figure S6a,b confirmed that as-prepared UiO-66 is uniform intergrown particles and easily agglomerated without specific shape, corresponding to that reported by Lv, Liu, Xiong, Zhang, and Guan (2016). All these results demonstrated that the UiO-66 was successfully synthesized.
In order to explore the potential influence of the fragrances load-  Figure S4 shows that there is no significant change in domain UiO-66 crystalline structure after the fragrance (isophorone, eugenol, and β-ionone) adsorption and after release for 20 days. The FTIR spectra of UiO-66 in Figure S7 show characteristic bands corresponding to carboxylate (-O-C-O-) groups at 1,400 and 1,584 cm −1 and the peak at 3,400 cm −1 corresponding to the free water molecules in the pores.
All absorption peaks correlate well with the literature report . The signal is less intense for isophorone/UiO-66 (Zr) compared with that of pure UiO-66, while the signal intensity of β-ionone/UiO-66 and eugenol/UiO-66 was higher than that of pure UiO-66, respectively. This deviation is most probably due to the fragrance adsorption on UiO-66 result in the change in the dipoles. The TGA curve in Figure   S5 shows three stages of weight losses for as-synthesized materials which can be attributed to water departure (25-130°C), DMF or fragrance departure (150-400°C), and decomposition of ligand and collapse of UiO-66 structure (400-650°C). Meanwhile, UiO-66 after fragrance adsorption presents lower water loss, and the eugenol/UiO-66 further exhibited less ligand decomposition, suggesting that more or less fragrance adsorption on the surface of UiO-66 reduces the surface water evaporation, resulting in higher thermal stability of Zr-MOF.
All these results demonstrate the successful construction of fragrance/UiO-66. This indicates that the crystalline structure of UiO-66 was highly stable during adsorption process and remained intact in contact with fragrances, suggesting the promising potential of UiO-66 in fragrance adsorption and release applications.

| Effect of solvents
Isophorone, eugenol, and β-ionone were selected as typical fragrances to investigate the effect of solvents on adsorption process. influence on fragrance adsorption. All fragrances suffered great decrease in adsorption in ethanol solution than in cyclohexane solution, which may be attributed to a more polarity of ethanol solvent (Fang et al., 2017) and the formation of hydrogen bond between ethanol solvent and fragrance molecules causing a larger adsorption-free energy and a lower adsorption capacity. As illustrated in Comparatively, the adsorption rate of β-ionone was much lower (38.3%) due to space steric hindrance resulted by lateral chains of β-ionone during the adsorption process.

| Effect of adsorbent dosage
The effect of adsorbent dosage on fragrance adsorption in cyclohexane solution was investigated at pH 7 and the initial fragrance concentration of 100 ppm. As shown in Figure 2, with an increase in dose from 3 to 30 mg, the adsorption rate of isophorone, eugenol, and β-ionone significantly increased from 57.2% to 99.4%, 66.7% to 99.9%, and 44.4% to 60.2%, respectively, due to the increase in surface area and active sites enhancing the adsorption capacity. Nevertheless, the adsorption rate of isophorone, eugenol, and β-ionone reached maximum at 9, 6, and 18 mg of UiO-66, respectively, and then, no further increase in adsorption rate is observed after that on account of the decrease in the adsorption sites. Therefore, the optimum dose of UiO-66 was fixed as 9, 6, and 18 mg, respectively.

| Release capacity under different temperatures
The effect of different temperatures on release behavior of volatile fragrances from UiO-66 was investigated. The sustained release of UiO-66 at different temperatures (15-90°C) within 5 hr. The loading amount of isophorone and eugenol on UiO-66 increased sharply and then slowed down with the increase in temperature, which were from 48.6 to 53.3 mg/g and 71.4 to 83.4 mg/g, respectively, indicating a reduced release rate. It may be because the molecular motion accelerated with the increase in temperature, and the fragrances initially adsorbed in relative large pores can diffuse into smaller micropores. After the diffusion reached a stable equilibrium, the solid loading reached a stable value, whereas for β-ionone in Figure 3c, the loading amount of β-ionone on UiO-66 at 30°C with 15.2 mg/g was higher than that at 15°C with 13.7 mg/g, exhibiting slight decrease in release rate. Then, the delivery system displayed an obviously increasing release rate and decreasing loading amount with the increase in temperature from 30 to 90°C, which may be because the movement energy of molecules is greater than the weak surface adsorption energy at high temperature, causing a vast majority of β-ionone adsorbed on the surface of UiO-66 could be released. In summary, temperature acted as a kind of activator for β-ionone release, so that the cumulative release of β-ionone (from 13.7 to 5.4 mg/g) was 60.6% during the temperature range. The results indicate that the release of eugenol and β-ionone from UiO-66 is more controlled and effective than that of isophorone.

| Long-time release study
These flavors and fragrances can be slowly released over a long period of time. The order of desorption rate from the adsorbent is β-ionone with 95.2% > eugenol with 52.6% > isophorone with 49.6%.

| Possible adsorption and controlled released mechanism
As demonstrated in N 2 adsorption-desorption analysis, the sample UiO-66 ( Figure 6) has small pore sizes range from 2 to 10 Å.
The micromolecular and simple fragrances (Figure 7) can be adsorbed inner the pores or on the surface of UiO-66. The adsorption of fragrance molecules by adsorbents mainly belongs to physical adsorption and is mainly affected by the pore structure of adsorbents, structure of fragrance molecules, and interactions (hydrogen bonding and π-π interaction). The high specific surface area of UiO-66 can exhibit high adsorption capacity to the fragrance molecules. The adsorption amount of eugenol was the largest probably due to the hydrogen bonding and π-π interaction between the eugenol molecules and UiO-66. The lowest adsorption capacity of β-ionone may be attributed to its large molecular size and long branch occupying many adsorption sites during adsorption process. The smallest molecular size of isophorone can easily enter into the interior of the pores resulting in a relatively high adsorption capacity on UiO-66.
Based on the studies (Zhou et al., 2017), the desorption and release of fragrances on porous materials undergoes general diffusion, Knudsen diffusion, and surface diffusion. The higher the proportion of micropores in adsorbents is, the higher the proportion of surface diffusion in the release process is, and the better effect the sustained release can achieve. Therefore, UiO-66, mainly microporous structure, can be regarded as desirable porous material for sustained release. The difference in sustained release effect is similarly related to the structure of fragrance molecules and the interaction between the adsorbent and the fragrances. When the bulky fragrances accumulated on the surface of the material, rapid diffusion and desorption can be achieved during the sustained release process. On the other hand, small molecular size fragrances retained in interior of the pores of the material after adsorption and were not easily volatilized and released resulting in undesirable sustained release capacity. The interaction between fragrance and adsorbents (hydrogen bonding and π-π interaction) will increase the diffusion resistance of fragrance causing a relatively slower release.

| CON CLUS ION
As confirmed, UiO-66 in our work can be used for effective adsorption and controlled release of commonly used fragrances of isophorone, eugenol, and β-ionone. Different solvents and F I G U R E 6 Chemical structures of UiO-66 in two-dimensional (2D) view (Zr: blue, O: red, C: gray) F I G U R E 7 Chemical structures of fragrances in this work (isophorone, eugenol, β-ionone) adsorbent dosage had a great influence on the adsorption process.
In cyclohexane solutions, the maximum adsorption rates of UiO-66 toward isophorone, eugenol, and β-ionone were 99.4%, 99.9%, and 60.2%, respectively, rapidly approached at 30, 10, and 30 min, respectively. The results of slow release indicate that these flavors and fragrances can be slowly released over a long period of time.
Particularly, the release of eugenol and β-ionone from UiO-66 is more controlled and effective than that of isophorone. The order of desorption rate from the adsorbent is β-ionone with 95.2% > eugenol with 52.6% > isophorone with 49.6%. In addition, temperature had different influence on release behavior of these fragrances. Increasing temperature accelerates the adsorption of isophorone, eugenol converted to chemical adsorption, advertising to their slow release, while temperature rise will avail to the release of β-ionone from UiO-66.

CO N FLI C T O F I NTE R E S T
The authors declare that they do not have any conflict of interest.

E TH I C A L A PPROVA L
This study does not involve any human or animal testing.