MOF@chitosan Composites with Potential Antifouling Properties for Open-Environment Applications of Metal-Organic Frameworks

Metal-organic frameworks (MOFs) are promising materials for a myriad of applications because of their easy synthesis and large variability through the organic linker. For open-environment applications, the organic content can, however, give rise to fouling, that is, biofilm formation. Biofilms can destroy the MOF and reduce the sorption capacity. Therefore, it is necessary to formulate MOFs for open-environment application to avoid the growth of microorganisms. Chitosan is a polysaccharide biopolymer, obtained from chitin shells of shrimps by alkaline deacetylation, and has known fungistatic properties. Here, chitosan is used as a matrix for MOF@chitosan composites with different aluminum-based MOFs to implement the fungistatic effect of chitosan to MOFs. The obtained composites with the highest possible MOF loadings of up to 90% were tested according to DIN EN ISO 846 to examine the fungistatic material properties against the fungi Chaetomium globosum and Aspergillus falconensis.


Introduction
Metal-organic frameworks (MOFs) consist of metal clusters, also called secondary building units (SBUs), and bridging organic ligands, also called linkers [1,2]. MOFs are microporous materials with highly tunable properties through their broad variability of metals and linkers [3,4] and a wide field of potential applications [5], such as gas separation [6,7], and storage [8], drug delivery [9,10], catalysis [11], heat transformation, etc. [12][13][14][15][16][17][18][19][20]. Still, for realistic applications, the initially microcrystalline MOF powders must be formulated or shaped into composites with polymers [21][22][23][24][25]. Furthermore, for the application of porous MOFs under ambient, open-environment conditions in the presence of air and moisture, not only must their hydrothermal stability be considered but also their resistance to antifouling needs to be addressed. A particularly relevant case for the antifouling behavior of MOFs is their use for cyclic water sorption from air for dehumidification [26], including water-harvesting. A recently advocated potential open-environment application is the use of porous materials to capture and release ('harvest') water vapor under atmospheric conditions without external power sources as a promising methodology for obtaining drinking water in arid or desert regions of the world. The potential of metal-organic frameworks to adsorb and desorb water for fresh water production in regions with medium or high humidity conditions during the night was probably first investigated by Kim et al. [27] and Trapani et al. in 2016 [28], and followed-up by Yaghi and others in 2017, until now [29,30].
Powder X-ray diffractometry (PXRD) was performed at ambient temperature on a Bruker D2 phaser (BRUKER, Billerica, MA, USA) using Cu-Kα radiation (λ = 1.54182 Å) between 5° < 2θ < 50° with a scan rate of 0.0125 s −1 (300 W, 30 kV, 10 mA) and a step size of 0.15 and 0.05 °per step, giving a typical total measurement time of 6 min for a diffractogram. In Figures S3 and S4, the PXRDs of the neat MOFs were also measured with a Rigaku Miniflex 600 (RIGAKU, Tokio; Japan) using Cu-Kα radiation (λ = 1.54182 Å) between 5° < 2θ < 50° with a scan rate of 0.083 s −1 (600 W, 40 kV, 15 mA) and a step size of 0.01 °per step, giving a total measurement time of 10 min for a diffractogram. Analyses of the diffractograms were carried out with Match 3.11 software. All PXRD patterns are collected in Section S5 in the Supplementary Information. The broadening of the PXRDs, obtained on the Bruker D2 phaser, is related to the short measurement time, which was selected due to the high number of samples.
Thermogravimetric analysis (TGA) was measured on a Netzsch TG209 F3 Tarsus (NE-TZSCH, Selb, Germany) device under synthetic air atmosphere, ramping 10 K min −1 to 600 °C. TGA curves are given in Section S7 in the Supplementary Information SEM images were acquired on a JEOL JSM-6510 Advanced electron microscope (JEOL, Akishima, Japan) with a LaB6 cathode at 5-20 keV. The microscope was equipped with an Xflash 410 (Bruker, Billerica, MA, USA) silicon drift detector. SEM images are presented in Section S8 in the Supplementary Information Surface areas (Brunauer-Emmett-Teller [77], BET) were determined by nitrogen (purity 99.999%, 5.0) sorption experiments at T = 77 K, using liquid nitrogen and ca. 20-50 mg of sample and performing on a Quantachrome Autosorp 6 (QUANTACHROME, Odelzhausen, Germany) instrument within a partial pressure range of pp0 -1 = 10 -3 − 1 bar. Each sample was degassed under a vacuum (<10 -2 mbar) at 120 °C for ca. 3 h prior to measurement. All surface areas (BET) were calculated from five adsorption points applying Rouquerol plots (r > 0.998). For pressure ranges of five-point BET calculations of each MOF, see Section S9 in the Supplementary Information. All N2 sorption isotherms are shown in Section S9 in the Supplementary Information   Figure 1. Schematic formation of (a) aluminum fumarate (Alfum) from Al 3+ and fumarate and (b) MIL-160 from Al 3+ and 2,5-furandicarboxylate (see Section S1, Figures S1 and S2, Supplementary Information for further details).

Materials and Instrumentation
All chemicals were used as received by suppliers (see Section S1 in the Supplementary  Information) Table S1 in the Supplementary Information).
Powder X-ray diffractometry (PXRD) was performed at ambient temperature on a Bruker D2 phaser (BRUKER, Billerica, MA, USA) using Cu-K α radiation (λ = 1.54182 Å) between 5 • < 2θ < 50 • with a scan rate of 0.0125 s −1 (300 W, 30 kV, 10 mA) and a step size of 0.15 and 0.05 • per step, giving a typical total measurement time of 6 min for a diffractogram. In Figures S3 and S4, the PXRDs of the neat MOFs were also measured with a Rigaku Miniflex 600 (RIGAKU, Tokio; Japan) using Cu-Kα radiation (λ = 1.54182 Å) between 5 • < 2θ < 50 • with a scan rate of 0.083 s −1 (600 W, 40 kV, 15 mA) and a step size of 0.01 • per step, giving a total measurement time of 10 min for a diffractogram. Analyses of the diffractograms were carried out with Match 3.11 software. All PXRD patterns are collected in Section S5 in the Supplementary Information. The broadening of the PXRDs, obtained on the Bruker D2 phaser, is related to the short measurement time, which was selected due to the high number of samples.
Thermogravimetric analysis (TGA) was measured on a Netzsch TG209 F3 Tarsus (NETZSCH, Selb, Germany) device under synthetic air atmosphere, ramping 10 K min −1 to 600 • C. TGA curves are given in Section S7 in the Supplementary Information. SEM images were acquired on a JEOL JSM-6510 Advanced electron microscope (JEOL, Akishima, Japan) with a LaB 6 cathode at 5-20 keV. The microscope was equipped with an Xflash 410 (Bruker, Billerica, MA, USA) silicon drift detector. SEM images are presented in Section S8 in the Supplementary Information.
Surface areas (Brunauer-Emmett-Teller [77], BET) were determined by nitrogen (purity 99.999%, 5.0) sorption experiments at T = 77 K, using liquid nitrogen and ca. 20-50 mg of sample and performing on a Quantachrome Autosorp 6 (QUANTACHROME, Odelzhausen, Germany) instrument within a partial pressure range of pp 0 −1 = 10 −3 − 1 bar. Each sample was degassed under a vacuum (<10 −2 mbar) at 120 • C for ca. 3 h prior to measurement. All surface areas (BET) were calculated from five adsorption points applying Rouquerol plots (r > 0.998). For pressure ranges of five-point BET calculations of each MOF, see Section S9 in the Supplementary Information. All N 2 sorption isotherms are shown in Section S9 in the Supplementary Information.
FT-IR spectra were measured in KBr-mode on a Bruker TENSOR 37 IR (BRUKER, Billerica, MA, USA) spectrometer in the range of 4000-400 cm −1 . All IR spectra are depicted in Section S6 in the Supplementary Information.
Supercritical drying was carried out using an ethanol-washed sample still dispersed in ethanol in an automated critical point dryer Leica EM CPD300 (LEICA, Wetzlar, Germany) which was set to perform 99 exchange cycles of CO 2 at slow speed and 100% stirring.
The samples for the tests for microbial metabolism were prepared on a laminar air flow bench of the type Laminar Air Flow (HERAEUS, Hanau, Germany). For this, petri dishes, needles, and an inoculating loop (SARSTEDT, Nümbrecht, Germany) were used. The tests were carried out in an Incu-Line Incubator (VWR, Radnor, PA, USA) and examined with an eyepiece microscope (WILD HEERBRUGG, Heerbrugg, Switzerland).

Chitosan Bead Synthesis
Chitosan beads were synthesized according to the literature of Bodmeier et al. and Wu et al. [78,79]. Chitosan solutions were prepared with different concentrations by dissolving chitosan flakes in 2 wt-% (weight-%) acetic acid. Thus, the following four concentrations of 6, 20, 30, and 40 g L −1 were obtained. For preparing beads, the chitosan solutions were taken up with a 1 mL syringe and added dropwise to a 10 wt-% Na 5 P 3 O 10 solution. Subsequently, the particles were washed with water and dried overnight (80 • C, 1-10 mbar) (further details in Section S2, Supplementary Information).

Synthesis of MOF@chitosan
MOF@chitosan beads were synthesized according to the literature of Zhuo et al. [58]. Different MOF loadings were prepared, up to a maximum of 90 wt-%. Composite materials were prepared with chitosan solutions of 6 or 20 g L −1 (corresponding to 10 and 20 or 3 and 6 mL of solution, cf. Table 1). The solid MOF was added into the chitosan solution (chitosan dissolved in 2 wt-% acetic acid) and stirred for 3 h. This mixture, drawn in portions of 1 mL in a 1 mL syringe and from there the portion of 1 mL, was added quickly dropwise (within 30 s) into a 10 wt-% Na 5 P 3 O 10 solution with constant stirring at room temperature for 15 min. The synthesized beads were collected and washed with Milli-Q water and dried overnight (80 • C, 1-10 mbar) (further details in Table 1 and Section S3, Supplementary Information). Figure 2 illustrates the synthesis process.

Synthesis of MOF@PVA
MOF@PVA beads were synthesized according to a synthesis of Khabzina et al. [80] with MOF loadings of up to 80 wt-%. A PVA solution (60 g L −1 ) was prepared by dissolving PVA powder in water at 80 °C for 24 h by constant stirring until a homogeneous solution was obtained. Then the MOF (480 mg) was added into the PVA solution (120 mg in 2 mL water) and stirred for 24 h to get a homogeneous suspension. The MOF@PVA suspension was, dropwise, added into liquid nitrogen to form composite beads. These beads were collected and stored in acetone for 24 h, followed by washing three times with acetone (150 mL) and drying overnight (80 °C, 1-10 mbar) (further details in Table 1 and Section S3, Supplementary Information).

Synthesis of MOF@PVA
MOF@PVA beads were synthesized according to a synthesis of Khabzina et al. [80] with MOF loadings of up to 80 wt-%. A PVA solution (60 g L −1 ) was prepared by dissolving PVA powder in water at 80 • C for 24 h by constant stirring until a homogeneous solution was obtained. Then the MOF (480 mg) was added into the PVA solution (120 mg in 2 mL water) and stirred for 24 h to get a homogeneous suspension. The MOF@PVA suspension was, dropwise, added into liquid nitrogen to form composite beads. These beads were collected and stored in acetone for 24 h, followed by washing three times with acetone (150 mL) and drying overnight (80 • C, 1-10 mbar) (further details in Table 1 and Section S3, Supplementary Information).

Synthesis of MOF@Silikophen ®
MOF@Silikophen ® composites were synthesized according to the literature of Jeremias et al. [37]. Silikophen ® P50/X (10 mL) was dissolved in xylene (30 mL). MOF tablets were prepared by using an IR-press with 2 tons of surface pressure (further details in Section S3, Supplementary Information). MOF tablets were dipped into the Silikophen ® solution, removed after 20 s, and dried at RT in air over night. Afterwards, the tablets were tempered according to the following program: 2 h/50 • C; 2 h/105 • C; 3 h/250 • C; and then cooled to RT (further details in Table 1 and Section S3, Supplementary Information).

Antifouling Tests
The procedure is based on method A of DIN EN ISO 846 (10/1997) (testing for resistance to fungi) [76], using a low-carbon nutrient medium. The composition of the nutrient medium can be taken from Table S5 in the Supplementary Information. The mineral salts are dissolved in 2000 mL of water and form the basis for the fungus tests. A total of 1000 mL of this stock mineral salt solution is mixed with 0.1 g of a non-toxic wetting agent (Tween80). The carbon source-free (incomplete) culture medium for the actual tests is obtained by adding 10 g agar (20 g L −1 ) to 500 mL of the stock mineral salt solution. The fungus tests were carried out according to the following protocol: First, the samples were immersed for 1 min in an ethanol-water mixture (70:30), air dried, and stored in a closed container at RT until use. The test specimens were disinfected before the test by means of water-steam sterilization (121 • C, 2 bar). The complete culture medium was filled after heating in sterile petri dishes and cured. The samples were placed as flat as possible on the incomplete nutrient medium. Spores of the fungi to be used were taken from the respective spore turf and combined with the mineral salt solution with a wetting agent additive to form a spore suspension. The spore suspensions were placed on the incomplete nutrient media and the petri dishes were sealed. The samples were incubated with the fungi at 24 ± 1 • C for up to four weeks.
The preparation of a spore suspension was carried out with the following fungi: Chaetomium globosum and Aspergillus falconensis (further details in Section S4, Supplementary Information).
Method A of the DIN EN ISO 846 [76] test describes five fungi (Aspergillus niger, Penicillium funiculosum, Paecilomyces varioti, Gliocladium virens, and Chaetomium globosum). Our labs only support risk group 1 organisms, according to the TRBA 460 "Classification of fungi in risk groups"; therefore, we decided to test our composites with two fungi classified as risk group 1 organisms: Aspergillus falconensis (as a substitute for Aspergillus niger, which belongs to risk group 2) and Chaetomium globosum. To better assess whether the used chitosan is digested or overgrown, the two mentioned fungi were examined individually (unlike the DIN EN ISO standard). Figure 3 shows two examples of two used composites that did not pass the overgrow tests. Due to the morphological appearances of the fungi, they could be distinguished easily from the composites.
All samples were tableted to obtain a uniform shape and surface area for the fungi growth and to facilitate visual assessment. This ensured that fungi growth would not be influenced by the shape or surface area of powders versus granules with larger and smaller outer-surface areas. However, during surface sterilization, some tablets degraded. Pieces of such degraded tablets were still used.
The evaluation of the samples was carried out by means of visual classification, which is also specified in the DIN EN ISO standard (cf. Table 2). Over the time of 15 or 27 days, photographic images of the antifouling tests were taken 1 to 6 days apart (Section S11, Supplementary Information). The basis for the assessment was always the final image of the test series (day 15 for Chaetomium globosum and day 27 for Aspergillus falconensis; Section S11, Supplementary Information) and the associated microscopic observation of the growth intensity (Table 2). Growth experiments with Aspergillus falconensis were carried out in triplicate to ensure reproducibility. Solids 2022, 2, FOR PEER REVIEW 7 All samples were tableted to obtain a uniform shape and surface area for the fungi growth and to facilitate visual assessment. This ensured that fungi growth would not be influenced by the shape or surface area of powders versus granules with larger and smaller outer-surface areas. However, during surface sterilization, some tablets degraded. Pieces of such degraded tablets were still used.
The evaluation of the samples was carried out by means of visual classification, which is also specified in the DIN EN ISO standard (cf. Table 2). Over the time of 15 or 27 days, photographic images of the antifouling tests were taken 1 to 6 days apart (Section S11, Supplementary Information). The basis for the assessment was always the final image of the test series (day 15 for Chaetomium globosum and day 27 for Aspergillus falconensis; Section S11, Supplementary Information) and the associated microscopic observation of the growth intensity (Table 2). Growth experiments with Aspergillus falconensis were carried out in triplicate to ensure reproducibility.

Results and Discussion
The aluminum fumarate (Alfum) used for the composite materials was supplied by BASF as Basolite ® A520. The synthesis of MIL-160 was based on a procedure by Cadiau et al. [42], except that twice the amount of sodium hydroxide was used to achieve complete deprotonation of the 2,5-furandicarboxylic acid. For comparison with the MOF@chitosan composites, both Alfum and MIL-160 were characterized by PXRD ( Figures S3 and S4 The used chitosan (medium molecular weight) was purchased from Sigma Aldrich, with a specified molecular weight dispersion of 190 to 310 kDa and a degree of deacetylation of 75-85% (from the data sheet of the manufacturer). The chitosan was received as flakes, which have a white-beige to brownish color. The crosslinked chitosan particles with different chitosan concentrations had BET-surface areas between 144 and 233 m 2 g −1 .

MOF@chitosan Composites
The soluble chitosan was crosslinked in the presence of the MOF, and to precipitate the MOF@chitosan composite Glutaraldehyde (1 mL of an 83 g L −1 solution) and sodium triphosphate (Na 5 P 3 O 10 , 10 wt-% solution) were tested as crosslinkers. Glutaraldehyde bonds connect covalently to chitosan ( Figure S42, Supplementary Information), whereas Na 5 P 3 O 10 connects the chitosan chains by ionic bonds through the protonated amino group (Figure 2). Na 5 P 3 O 10 proved to be the better crosslinker, due to its non-toxicity and easier handling, regarding synthesis, compared to glutaraldehyde. Hence, the use of glutaraldehyde was not pursued any further, also because of its toxicity. Furthermore, the syntheses of the composite materials with glutaraldehyde required considerably more time for the gelation and aging so that only monoliths, no beads, could be obtained and MOF loadings up to 90% could not be achieved (Section S2 and S3, Supplementary Information).
The composite formation does not affect the particle size of the MOF. The MOF peak widths in the PXRD patterns (albeit already high due to instrument effects; see below) did not change. The higher background is related to the amount of amorphous chitosan inside the composite. There should be no change in MOF particle size because ultrasound, which is known to decrease the particle size, was not used for composite formation.

Alfum@chitosan
For the Alfum@chitosan composites, several syntheses were carried out in which the MOF content was systematically varied, and two different chitosan concentrations (6 and 20 g L −1 ) were used. The MOF content was 60, 80, and 90 wt-% in relation to the all the composite material. Even higher loading was not possible because the volume of the chitosan solution eventually became too low and no longer produced a homogeneous dispersion; therefore, it became impossible to add it dropwise into the Na 5 P 3 O 10 solution. The diffraction patterns of the Alfum@chitosan composites confirm the retention of MOF crystallinity (Figures 4 and S5, Supplementary Information).   Figure S5 (Supplementary Information). The broad contribution from amorphous chitosan is only seen at lower MOF wt-% ( Figure S5). The broadening of the PXRDs is related to the short measurement time of 6 min on the Bruker D2 phaser, which was used due to the large number of samples and is not related to a low crystallinity of the MOF (cf. Figure S3 in the Supplementary Information with the sharper reflexes measured on a Rigaku Miniflex). The PXRD of Alfum was obtained from the purchased MOF from BASF.
The pore accessibility of Alfum within the Alfum@chitosan composite materials was investigated by nitrogen and water sorption ( Figures 5, S29 and S37, Supplementary Information).  Figure S5 (Supplementary Information). The broad contribution from amorphous chitosan is only seen at lower MOF wt-% ( Figure S5). The broadening of the PXRDs is related to the short measurement time of 6 min on the Bruker D2 phaser, which was used due to the large number of samples and is not related to a low crystallinity of the MOF (cf. Figure S3 in the Supplementary Information with the sharper reflexes measured on a Rigaku Miniflex). The PXRD of Alfum was obtained from the purchased MOF from BASF.
The pore accessibility of Alfum within the Alfum@chitosan composite materials was investigated by nitrogen and water sorption ( Figures 5, S29 and S37, Supplementary Information).
of samples and is not related to a low crystallinity of the MOF (cf. Figure S3 in the Supplementary Information with the sharper reflexes measured on a Rigaku Miniflex). The PXRD of Alfum was obtained from the purchased MOF from BASF.
The pore accessibility of Alfum within the Alfum@chitosan composite materials was investigated by nitrogen and water sorption (Figures 5, S29 and S37, Supplementary Information). The porosity, that is, BET-surface areas, pore volumes and water uptake values decrease disproportionately high with the decrease in the Alfum in the composite. Table 3 lists the porosity data in comparison with the calculated (expected) values, which are mass-corrected for the wt-% of the MOFs and the chitosan in the composite. The measured values also become much lower than the calculated (expected) BET-surface areas and the water uptake values. Thus, the composite materials with 60% MOF have a significantly lower surface than the materials with 80% and 90% MOF. Only when the composite was dried supercritically for activation prior to the sorption measurements the expected BETsurface area could be obtained. During the normal thermal/vacuum drying process, the The porosity, that is, BET-surface areas, pore volumes and water uptake values decrease disproportionately high with the decrease in the Alfum in the composite. Table 3 lists the porosity data in comparison with the calculated (expected) values, which are mass-corrected for the wt-% of the MOFs and the chitosan in the composite. The measured values also become much lower than the calculated (expected) BET-surface areas and the water uptake values. Thus, the composite materials with 60% MOF have a significantly lower surface than the materials with 80% and 90% MOF. Only when the composite was dried supercritically for activation prior to the sorption measurements the expected BETsurface area could be obtained. During the normal thermal/vacuum drying process, the beads of the Alfum@chitosan composite shrink considerably (decrease in diameter from 3.5 mm to 3.0 mm (20 g L −1 ) and 2.5 mm to 2.4 mm (6 g L −1 ), Figure S41, Supplementary Information) whereas during supercritically drying the beads retain the original size but become mechanically fragile. Hence, the solvent plays a role of a pore-forming agent by creating an interfacial volume between the MOF and the chitosan layer. When the solvent is removed by thermal/vacuum drying, the chitosan layer apparently covers and blocks part of the pores. Supercritical drying retains the interfacial volume and, thereby, prevents pore blocking but at the expense of mechanical stability. The beads break apart. The composite material with 90% MOF and a chitosan concentration of 6 g L −1 provides the best result with a BET-surface area of 964 m 2 g −1 .
Scanning electron microscopy (SEM) images with energy-dispersive X-ray spectroscopic (EDX) element mapping ( Figure 6) reveal a very good superposition, which suggests that both aluminum and phosphorus (from cross-linked chitosan) are evenly distributed in the bead.
A magnified view of the surface of the beads does not allow differentiation between the agglomerated particles of Alfum and the chitosan since Alfum does not have a welldefined morphology.  (1), pore volumes (2), and the water uptake values (3) were derived as the sum of the mass-weighted S BET , V pore , or H 2 O uptake of the MOFs and S BET , V pore , or H 2 O uptake of the chitosan with the following three formulas: S calc BET = wt-% MOF * S MOF BET + wt-% Chitosan * S Chitosan BET (1); V calc pore = wt-% MOF * V MOF pore + wt-% Chitosan * V Chitosan pore (2); H 2 O uptake calc = wt-% MOF * H 2 O uptake MOF + wt-% Chitosan * H 2 O uptake Chitosan (3). with S BET (Alfum) = 988 m 2 g −1 , S BET (MIL-160) = 1186 m 2 g −1 , etc. c d.s. = dried supercritically for activation prior to sorption measurements.

MIL-160@chitosan
The MIL-160 composite materials were prepared and characterized like the composite materials with Alfum. The same percentages of MOF and chitosan were used to ensure comparability. The PXRDs of the composite materials point out the reflexes of the MIL-160. Only the diffractogram of the spheres with 60% has a stronger background due to the high amorphous content of chitosan, shown in the ESI. The PXRDs of 80% and 90% composite materials do not have a strong background and correlate with the MOF content, shown in Figure 7 for the composite material with 90% MIL-160. In addition, the reflections match the MOF and the diffractogram has no impurities. pore blocking but at the expense of mechanical stability. The beads break apart. The composite material with 90% MOF and a chitosan concentration of 6 g L −1 provides the best result with a BET-surface area of 964 m 2 g −1 .
Scanning electron microscopy (SEM) images with energy-dispersive X-ray spectroscopic (EDX) element mapping ( Figure 6) reveal a very good superposition, which suggests that both aluminum and phosphorus (from cross-linked chitosan) are evenly distributed in the bead. Figure 6. SEM images of Alfum80@chitosan at different magnifications (left: overview, right: close-up). EDX-element mapping for aluminum and phosphorus (bottom) for the particle in the overview at top left. The dark features in the element maps, that is, the lower amount of Al and P in the center and to the upper right of the bead are due to X-ray blocking by the groove in the bead surface so that the generated element-specific X-rays cannot be detected. . EDX-element mapping for aluminum and phosphorus (bottom) for the particle in the overview at top left. The dark features in the element maps, that is, the lower amount of Al and P in the center and to the upper right of the bead are due to X-ray blocking by the groove in the bead surface so that the generated element-specific X-rays cannot be detected.  Figure S6 (Supplementary Information). The broad contribution from amorphous chitosan is only seen at lower MOF wt-% ( Figure S6). The broadening of the PXRDs are related to the short measurement time of 6 min on the Bruker D2 phaser, which was used due to the large number of samples and is not related to a low crystallinity of the MOF (cf. Figure S4 Figure S6 (Supplementary Information). The broad contribution from amorphous chitosan is only seen at lower MOF wt-% ( Figure S6). The broadening of the PXRDs are related to the short measurement time of 6 min on the Bruker D2 phaser, which was used due to the large number of samples and is not related to a low crystallinity of the MOF (cf. Figure S4 in the Supplementary Information with the sharper reflexes measured on a Rigaku Miniflex).
The sorption properties were investigated analogously to the other composite materials with nitrogen sorption and water sorption. The nitrogen sorptions are shown in Figure 8. related to the short measurement time of 6 min on the Bruker D2 phaser, which was used due to the large number of samples and is not related to a low crystallinity of the MOF (cf. Figure S4 in the Supplementary Information with the sharper reflexes measured on a Rigaku Miniflex).
The sorption properties were investigated analogously to the other composite materials with nitrogen sorption and water sorption. The nitrogen sorptions are shown in Figure 8. The nitrogen adsorption isotherms and the associated BET-surface areas correlate with the percentage of MOF (cf. Table 3). Thus, the composite materials with 60% MOF have a significantly lower surface area than the materials with 80% and 90% MOF. The composite material with 90% MOF and a chitosan concentration of 6 g L −1 gives the best result with a BET-surface area of 1068 m 2 g −1 . The associated water sorption isotherms are shown in Figure 8 and in the ESI and were measured for the composite materials with an MOF content of 80% and 90%. The S-shaped water sorption isotherm with little-to-no uptake at low p p0 -1 and a sudden rise in a small p p0 -1 interval correspond to Type V isotherms [77] and are typical for the adsorption of water vapor at MOFs [15]. The nitrogen adsorption isotherms and the associated BET-surface areas correlate with the percentage of MOF (cf. Table 3). Thus, the composite materials with 60% MOF have a significantly lower surface area than the materials with 80% and 90% MOF. The composite material with 90% MOF and a chitosan concentration of 6 g L −1 gives the best result with a BET-surface area of 1068 m 2 g −1 . The associated water sorption isotherms are shown in Figure 8 and in the ESI and were measured for the composite materials with an MOF content of 80% and 90%. The S-shaped water sorption isotherm with little-to-no uptake at low p p 0 −1 and a sudden rise in a small p p 0 −1 interval correspond to Type V isotherms [77] and are typical for the adsorption of water vapor at MOFs [15].
The maximum adsorption of the composite material is above the maximum uptake of MIL-160. This can be attributed to the polymer because it has a higher water uptake at a relative pressure of 0.9. This higher uptake is characteristic for polymers, and it is due to swelling. To verify the cross-linking of the composite and the MOF in the material, an EDX mapping for aluminum and phosphorus was performed and presented in Figure 9.
The EDX mapping shows that, in the complete area of the sphere, aluminum, and thus MOF and phosphorus, and thus cross-linked chitosan, are present. The lower amounts of Al and P in the middle right of the mapping is due to the particle geometry. The hollow in the middle of the particle blocks the X-rays, and they cannot be detected. In addition to the analytical methods described, TGA and IR, shown in the ESI, were recorded on selected samples.
The maximum adsorption of the composite material is above the maximum uptake of MIL-160. This can be attributed to the polymer because it has a higher water uptake at a relative pressure of 0.9. This higher uptake is characteristic for polymers, and it is due to swelling. To verify the cross-linking of the composite and the MOF in the material, an EDX mapping for aluminum and phosphorus was performed and presented in Figure 9. close-up). EDX-mapping measurement for aluminum and phosphorus (bottom). The dark features in the element maps, that is, the lower amounts of Al and P in the center and to the upper right of the bead are due to X-ray blocking by the groove in the bead surface so that the generated elementspecific X-rays cannot be detected.
The EDX mapping shows that, in the complete area of the sphere, aluminum, and thus MOF and phosphorus, and thus cross-linked chitosan, are present. The lower amounts of Al and P in the middle right of the mapping is due to the particle geometry. The hollow in the middle of the particle blocks the X-rays, and they cannot be detected. In addition to the analytical methods described, TGA and IR, shown in the ESI, were recorded on selected samples.

MOF@PVA Composites and MOF@Silikophen ® Composites
The MOF@PVA composite materials, as well as the Silikophen ® composite materials, were used as reference materials for the antifouling tests of MOF@chitosan. Silikophen ® is a phenylmethylpolysiloxane resin and, unlike the chitosan and PVA polymers, was not used as a matrix but as a coating. For coating, tablets of the two MOFs were pressed and immersed for 20 s in a Silikophen ® -xylene mixture, then dried in the air, transferred to a Büchi glass oven, and tempered. The PVA and Silikophen ® composites were characterized by PXRD (Figures S8 and S9), SEM ( Figures S22-S25), N2-sorption ( Figures S32 and S33), H2O-sorption ( Figure S40), and TGA ( Figure S16).

Antifouling Tests
The evaluation of the samples was conducted by visual examination according to DIN EN ISO 846. 76 Chaetomium globosum grows faster than Aspergillus falconensis and can Figure 9. SEM images of MIL-160(90)@chitosan at different magnifications (left: overview; right: close-up). EDX-mapping measurement for aluminum and phosphorus (bottom). The dark features in the element maps, that is, the lower amounts of Al and P in the center and to the upper right of the bead are due to X-ray blocking by the groove in the bead surface so that the generated element-specific X-rays cannot be detected.

MOF@PVA Composites and MOF@Silikophen ® Composites
The MOF@PVA composite materials, as well as the Silikophen ® composite materials, were used as reference materials for the antifouling tests of MOF@chitosan. Silikophen ® is a phenylmethylpolysiloxane resin and, unlike the chitosan and PVA polymers, was not used as a matrix but as a coating. For coating, tablets of the two MOFs were pressed and immersed for 20 s in a Silikophen ® -xylene mixture, then dried in the air, transferred to a Büchi glass oven, and tempered. The PVA and Silikophen ® composites were characterized by PXRD ( Figures S8 and S9

Antifouling Tests
The evaluation of the samples was conducted by visual examination according to DIN EN ISO 846 [76]. Chaetomium globosum grows faster than Aspergillus falconensis and can fully overgrow composites within 15 days (Section S11, Supplementary Information), so that incubation with Chaetomium globosum was stopped after this time while incubation with Aspergillus falconensis was continued for 27 days. Table 4 lists the results of the microbial metabolism tests with Aspergillus falconensis. None of composites exhibited an inhibitory effect toward Chaetomium globosum. Fully overgrown and colonized composites with Chaetomium globosum were observed in all samples (Section S11, Supplementary Information).  fully overgrow composites within 15 days (Section S11, Supplementary Information), so that incubation with Chaetomium globosum was stopped after this time while incubation with Aspergillus falconensis was continued for 27 days. Table 4 lists the results of the microbial metabolism tests with Aspergillus falconensis. None of composites exhibited an inhibitory effect toward Chaetomium globosum. Fully overgrown and colonized composites with Chaetomium globosum were observed in all samples (Section S11, Supplementary Information).   In all figures, the growth of the fungus is clearly recognizable and due to the green color of the fungus it can be distinguished from the white composite material (cf.). Visually, no growth of the fungus can be seen on top of the material surface in Figure 10. This is also confirmed by the microscopic image. The fungus grows to the vertical edges of the composite material but does not overgrow it. An overgrowth of the composite material, as it is needed for the categories 2-5 (Table 2), can be seen for Alfum and crosslinked chitosan alone ( Figure 11) and the PVA and Silikophen ® composites ( Figure 12). It is noteworthy that the separate Alfum MOF and chitosan did not show an inhibitory effect towards Aspergillus falconensis. Only the combination of Alfum and chitosan exhibited a higher resistance to the fungus. The distance from the start of fungus growth to the composite materials plays a major role and may explain the slight differences between the three replicate experiments. It is difficult to ensure that the composite materials always grow the same distance from where the fungus is placed on the nutrient medium because the fungus does not grow in a defined direction. During incubation, we ensured that all tablets had a chance to get in contact with the fungus by adding water to prevent the nutrient medium from running dry and making it useless for examination over the time of 15 or 27 days. The reference composite material Alfum@PVA shows a low degree of overgrowth, and it can also be clearly seen that the sample is not central to fungal growth. The composite Alfum@Silikophen ® is assigned to category 5, as over 50% of the sample surface is covered with the fungus. From this it can be concluded that the Silikophen ® binder has no anti-fouling effect and thus serves well as a comparison material.
We observed that the Aspergillus falconensis can experience two quite different growth states. The fungus can strongly overgrow the Alfum@PVA and MIL-160@Silikophen ® composites. For the MOF@chitosan composite, no growth is visible under microscopic observation, hence the material can be considered fungistatic and does not serve as a nutrient for microorganisms.

Conclusions
In this work, we were able to synthesize porous MOF@chitosan composite materials with different loadings. Alfum and MIL-160 were embedded in a chitosan matrix for the In all figures, the growth of the fungus is clearly recognizable and due to the green color of the fungus it can be distinguished from the white composite material (cf.). Visually, no growth of the fungus can be seen on top of the material surface in Figure 10. This is also confirmed by the microscopic image. The fungus grows to the vertical edges of the composite material but does not overgrow it. An overgrowth of the composite material, as it is needed for the categories 2-5 (Table 2), can be seen for Alfum and crosslinked chitosan alone ( Figure 11) and the PVA and Silikophen ® composites ( Figure 12).
It is noteworthy that the separate Alfum MOF and chitosan did not show an inhibitory effect towards Aspergillus falconensis. Only the combination of Alfum and chitosan exhibited a higher resistance to the fungus. The distance from the start of fungus growth to the composite materials plays a major role and may explain the slight differences between the three replicate experiments. It is difficult to ensure that the composite materials always grow the same distance from where the fungus is placed on the nutrient medium because the fungus does not grow in a defined direction. During incubation, we ensured that all tablets had a chance to get in contact with the fungus by adding water to prevent the nutrient medium from running dry and making it useless for examination over the time of 15 or 27 days.
The reference composite material Alfum@PVA shows a low degree of overgrowth, and it can also be clearly seen that the sample is not central to fungal growth. The composite Alfum@Silikophen ® is assigned to category 5, as over 50% of the sample surface is covered with the fungus. From this it can be concluded that the Silikophen ® binder has no antifouling effect and thus serves well as a comparison material.
We observed that the Aspergillus falconensis can experience two quite different growth states. The fungus can strongly overgrow the Alfum@PVA and MIL-160@Silikophen ® composites. For the MOF@chitosan composite, no growth is visible under microscopic observation, hence the material can be considered fungistatic and does not serve as a nutrient for microorganisms.

Conclusions
In this work, we were able to synthesize porous MOF@chitosan composite materials with different loadings. Alfum and MIL-160 were embedded in a chitosan matrix for the first time. The chitosan cross-linking was carried out in situ with Na 5 P 3 O 10 , whereby the chitosan concentrations were systematically varied in order to ensure a homogeneous dispersion necessary for the synthesis and thus to obtain the desired properties of the composite materials. Spherical particles could be obtained by the very reproducible drop casting method.
In addition, tests were performed on the microbial metabolism of the composites. An anti-fouling effect of the chitosan composites against Aspergillus falconensis could be detected whereas Chaetomium globosum proved to be more aggressive and could overgrow the chitosan composite materials. Further, the chitosan composite materials show a significantly higher resistance to Aspergillus falconensis than the composites with PVA and Silikophen ® . All MOF@chitosan composite materials could be classified as "inert" or "fungistatic" against Aspergillus falconensis. Another interesting observation was the behavior of pure MOF tablets where MIL-160 had a higher resistance than Alfum. In follow-up work, this suggests the need to investigate different MOF linkers and Al-MOFs with a wider variety of linkers for their fungistatic effects.