Characterization and Gel Properties of Low-Molecular-Weight Carrageenans Prepared by Photocatalytic Degradation

Low-molecular-weight carrageenan has attracted great interest because it shows advantages in solubility, absorption efficiency, and bioavailability compared to original carrageenan. However more environment-friendly and efficient methods to prepare low-molecular-weight carrageenan are still in great need. In the present study, a photocatalytic degradation method with only TiO2 has been developed and it could decrease the average molecular weight of κ-carrageenan to 4 kDa within 6 h. The comparison of the chemical compositions of the degradation products with those of carrageenan by FT-IR, NMR, etc., indicates no obvious removement of sulfate group, which is essential for bioactivities. Then 20 carrageenan oligosaccharides in the degradation products were identified by HPLC-MSn, and 75% of them possessed AnGal or its decarbonylated derivative at their reducing end, indicating that photocatalysis is preferential to break the glycosidic bond of AnGal. Moreover, the analysis results rheology and Cryo-SEM demonstrated that the gel property decreased gradually. Therefore, the present study demonstrated that the photocatalytic method with TiO2 as the only catalyst has the potential to prepare low-molecular-weight carrageenan with high sulfation degree and low viscosity, and it also proposed the degradation rules after characterizing the degradation products. Thus, the present study provides an effective green method for the degradation of carrageenan.

Recently, various biological activities of κ-carrageenan such as antiviral, antioxidant, anticoagulant, and anti-obesity effects have attracted widespread attention [3][4][5][6]. However, κ-carrageenan has a large molecular weight, poor solubility, and low bioavailability, which greatly limit its application range [7]. As the degradation products of carrageenan, carrageenan oligosaccharides exhibited diverse biological activities as well as other remarkable properties, especially better solubility and higher absorption [8]. In addition, carrageenan remarkable properties, especially better solubility and higher absorption [8]. In addition, carrageenan oligosaccharides also have potential as an antimicrobial agent [9], antifreeze agent [10], and drug delivery carrier [11].
Therefore, it is meaningful to develop the degradation methods to prepare low-molecular-weight carrageenan with excellent solubility so as to improve its biological activities and expand its application range. The reported degradation methods for κ-carrageenan include acid, enzymatic, microwave-assisted, ultrasonic-assisted, and oxidative degradation [12][13][14]. The operation process of acid degradation polysaccharide is relatively simple and costs less, but it is difficult to master the conditions of the acid degradation polysaccharide process, and easy to damage the structure of the polysaccharide, such as the sulfate groups, thus affecting its biological activities [15]. The enzymatic approach is high-cost and inefficient, which has limited its industrial-scale applications. Although the microwave-assisted and ultrasonic-assisted degradation are simple and non-pollution, their production costs are also high due to their necessary equipment [16]. Therefore, it is essential to develop a high-efficiency, environmentally friendly, and low-cost degradation method, which can retain the sulfate groups of κ-carrageenan.
Photocatalytic degradation, as an advanced oxidation process, has made rapid progress since the discovery of the first water splitting reaction using a semiconductor in 1972 by Fujishima and Honda. As shown in Figure 1, the photocatalytic process begins from the semiconductor band gap. When an electron (e − ) in a semiconductor valence band is excited to transition to the conduction band, photogenerated holes (h + ) in the valence band will be generated in response, thus yielding electron-hole (e − and h + ) pairs inside the semiconductor [17]. As a result, the excited electrons will produce free radicals to impel the photocatalytic reactions. Among various semiconductor catalysts, titanium dioxide (TiO2) is one of the best photocatalysts for inducing a series of reductive and oxidative reactions on its surface, as TiO2 is the most active photocatalyst under the photon energy of 300 nm < λ < 390 nm, and it remains stable after several catalytic cycles [18]. Besides this, TiO2 catalyst also has many functional properties, such as its strongly thermal stability and resistance to chemical damage, which promote its wide application in photocatalytic technology [19]. Previous research revealed that photocatalysis showed a good performance in the treatment of organic pollutants due to its advantages of fast reaction speed, low cost, and "green chemistry" technology [20]. At present, photocatalysis technology is applied in not only the removement of organic pollutants, but also anti-virus, anti-bacteria, etc. [21,22]. Photocatalytic degradation shows the advantages of high-efficiency, energy-saving, Previous research revealed that photocatalysis showed a good performance in the treatment of organic pollutants due to its advantages of fast reaction speed, low cost, and "green chemistry" technology [20]. At present, photocatalysis technology is applied in not only the removement of organic pollutants, but also anti-virus, anti-bacteria, etc. [21,22]. Photocatalytic degradation shows the advantages of high-efficiency, energy-saving, ecofriendly, and sustainability [23]. More importantly, it is more suitable for sulfated polysaccharide degradation because it does not obviously remove the sulfate group, which is critical for the bioactivities [24]. Photocatalysis with the addition of H 2 O 2 could degrade chondroitin sulfate [25] and fucoidan [26]. However, there is no report on the successful photocatalysis degradation of sulfated polysaccharides without H 2 O 2 to date.
In the present study, a photocatalytic method with only TiO 2 as photocatalyst was developed for κ-carrageenan degradation, and the resulting degradation products were detected by a series of analytical methods to reveal changes of κ-carrageenan in the structure, composition, rheological property, and microstructure during photocatalytic degradation.

Photocatalytic Degradation of κ-Carrageenan
A CEL-HXF300-T3 system (Zhongjiao Jinyuan Technology Co., Ltd., Beijing, China) was used for the photocatalytic reaction, and it was configured with a 300 W Xenon lamp whose wavelength range is 300 nm-2500 nm (PerkinElmer, Waltham, MA, USA).
κ-carrageenan (1 g) was dissolved in water (200 mL), magnetic stirring at 60 • C for 6 h, and placed in the refrigerator at 4 • C overnight to obtain κ-carrageenan solution (0.5% w/w). The κ-carrageenan solution was added into the photocatalytic reactor, and TiO 2 (5%), the combination of H 2 O 2 (300 mM) and TiO 2 (5%), or H 2 O 2 (300 mM) were added while stirring (the addition amounts of TiO 2 and H 2 O 2 were based on the results optimized by our team previously [26]). During the photocatalytic experiment, the solution was continuously stirred and exposed to a Xenon lamp for 6 h, and the samples degraded for 2 h, 4 h, and 6 h. After centrifugation at 10,000 rpm for 20 min, the supernatants were lyophilized individually, which were named as DC-2, DC-4, and DC-6, respectively.

Chemical Composition Analysis
The total sugar content was measured by phenol-sulfuric acid method using galactose as the standard [27]. The reducing sugar content was measured by 3,5-dinitrosalicylic acid (DNS) assay using galactose as the standard [28]. The uronic acid content was determined by m-hydroxy biphenyl method using D-galacturonic acid as the standard [29]. The sulfated polysaccharide content was determined by the metachromatic assay with 1,9-dimethylmethylene blue (DMB) using pure κ-carrageenan as the standard [30]. The 3,6-anhydro galactose content was measured by resorcinol method with a minor modification [31]. The total sulfate group content was determined by the barium chloride-gelatine method using anhydrous potassium sulfate as the standard [32]. The free sulfate content was determined through ion chromatography using a DIONEX IonPac AS23 column (4.0 mm × 250 mm) at 30 • C and 4.5 mM Na 2 CO 3 and 0.8 mM NaHCO 3 were utilized as mobile phases.
The sulfate content = Total sulfate content − Free sulfate content (1)

Molecular Weight Determination
The molecular weights (Mw) of samples were examined by high performance gel permeation chromatography (HPGPC), which was configured with a refractive index detector (Waters 2414, Milford, MA, USA) and a TSK G5000PWXL chromatographic column (7.8 mm × 300 mm) was used for separation. The sample was eluted with 100 mM ammonium acetate at a flow rate of 0.4 mL/min. A series of dextran standards with molecular weights of 5, 25, 50, 410, and 670 kDa were applied as standards.

Thin Layer Chromatography (TLC)
Each 2 µL of polysaccharide samples and standards (glucose, lactose, and β-cyclodextrin) were spotted separately onto a 20 cm × 20 cm plastic TLC plate (Macherey, Düren, Germany), and the plate was performed with the solvent system of n-butanol: formic acid: water (2:1:1, v/v/v) at 25 • C. Then, the plate was stained with orcinol-sulfuric acid reagent which consisted of 50 mg of orcinol and methanol: H 2 SO 4 (5:1, v/v), and then heated at 120 • C to visualize the bands.

Fourier-Transform Infrared (FT-IR) Spectroscopic Analysis
The carrageenan and its degradation products (2 mg) were individually pressed into 1 mm pellets with KBr (100 mg). The FTIR spectra of sample pellets were recorded in a dry atmosphere at room temperature (21 ± 1 • C) by using a Spectrum One-B FTIR Spectrometer (PerkinElmer, Waltham, MA, USA) in the range of 500 to 4000 cm − 1 . Furthermore, a background spectrum was subtracted automatically by using pure dried KBr to eliminate the interferences of carbon dioxide and water in the atmosphere.

Mass Spectrometry Analysis
DC-6 was further fractionated with a molecular weight cut-off (MWCO) of 3 kDa to obtain κ-carrageenan oligosaccharides, which were further subjected to mass spectrometry analysis according to our previously reported method with minor modification [33]. Briefly, the oligosaccharides were labeled with 1-phenyl-3-methyl-5-pyrazolone (PMP) at first, and then the products were analyzed by HPLC-MS n . LXQ mass spectrometer (Thermo Fisher, Pittsburgh, PA, USA) coupled with a photodiode array detector (PDA) and an electrospray ionization (ESI) source were employed. The oligosaccharide derivatives were separated with a TSKgel-Amide-80 column (4.6 mm × 150 mm, 3 µm) and eluted with 20 mM ammonium acetate-acetonitrile (5:95, v/v) at 0.2 mL/min. For the electrospray ion source, the spray voltage was set at 4.5 kV, the capillary voltage was set at 37 V, and the capillary temperature was 275 • C [34]. All data were acquired and processed through XCalibur software.

Rheological Measurement
The dynamic rheological measurement of the samples was obtained with a DHR-2 rheometer (TA Instruments, New Castle, DE, USA). A 60-mm diameter aluminum plate geometry was used for measurement at the distance of 500 µm. First of all, about 2 mL of the sample was taken and placed on the sample plate, the excess sample was scraped off, coated with silicone oil, and covered with a moisturizing cap to minimize measurement errors. Before the test, the initial temperature equilibrium of the samples was maintained for 5 min. A strain sweep program (0.001-100% at 25 • C and 1 Hz) was performed to determine the linear viscoelastic region. The time sweeps were conducted in the scope of 0-160 s at 4 • C and 25 • C, and the frequency sweeps were carried out in the range of 0.01-10% at 4 • C and 25 • C.

Cryogenic Scanning Electron Microscopy (Cryo-SEM)
The microstructure of the samples was carried out by using a Cryo-SEM (SU8010, Hitachi, Japan) configured with the PP3010T preparation system (Quorum Technologies Ltd., East Sussex, UK). Briefly, an appropriate amount of sample was loaded onto rivet and frozen quickly in liquid nitrogen slush. After that, the fractured samples were further sublimated at −60 • C for 30 min and coated with gold for 30 s. At last, the samples were transmitted to the SEM chamber for observation [35].

Statistical Analysis
All the results were presented as mean ± SD. The statistical analysis was performed using one-way ANOVA multiple-comparisons tests by SPSS 19.0 statistical software. p < 0.05 was considered to be statistically significant between the groups.

Photocatalytic Degradation of Carrageenan
In order to reveal the molecular weight change of κ-carrageenan in the photocatalytic reaction, HPGPC and TLC were used to analyze the hydrolysates during the photocatalytic reaction. We compared the degradation effects of 300 mM H 2 O 2 (Figure 2A), the combination of 300 mM H 2 O 2 and 5% TiO 2 ( Figure 2B), and 5% TiO 2 ( Figure 2C) on carrageenan. Comparison of the molecular distribution of products collected at 2 h and 4 h indicates that the combination of H 2 O 2 and TiO 2 possesses the highest degradation efficiency, followed by using H 2 O 2 alone, then TiO 2 . However, all of the three conditions could degrade carrageenan to 3~6 kDa after 6 h. Notably, it is unexpected that a similar outcome of 6 h degradation could also be achieved by using only TiO 2 . TLC analysis ( Figure 2D) also demonstrated that oligosaccharides gradually accumulated by using TiO 2 alone.
The microstructure of the samples was carried out by using a Cryo-SEM (SU Hitachi, Japan) configured with the PP3010T preparation system (Quorum Technol Ltd., East Sussex, UK). Briefly, an appropriate amount of sample was loaded onto and frozen quickly in liquid nitrogen slush. After that, the fractured samples were fu sublimated at −60 °C for 30 min and coated with gold for 30 s. At last, the samples transmitted to the SEM chamber for observation [35].

Statistical Analysis
All the results were presented as mean ± SD. The statistical analysis was perfo using one-way ANOVA multiple-comparisons tests by SPSS 19.0 statistical softwar 0.05 was considered to be statistically significant between the groups.

Photocatalytic Degradation of Carrageenan
In order to reveal the molecular weight change of κ-carrageenan in the photocat reaction, HPGPC and TLC were used to analyze the hydrolysates during the photo lytic reaction. We compared the degradation effects of 300 mM H2O2 (Figure 2A), the bination of 300 mM H2O2 and 5% TiO2 ( Figure 2B), and 5% TiO2 ( Figure 2C) on carr nan. Comparison of the molecular distribution of products collected at 2 h and 4 h cates that the combination of H2O2 and TiO2 possesses the highest degradation effici followed by using H2O2 alone, then TiO2. However, all of the three conditions coul grade carrageenan to 3~6 kDa after 6 h. Notably, it is unexpected that a similar out of 6 h degradation could also be achieved by using only TiO2. TLC analysis (Figur also demonstrated that oligosaccharides gradually accumulated by using TiO2 alone TiO2 as photocatalyst could generate electron-hole (e − and h + ) pairs under the and the excited electrons can further generate free radicals. H2O2, which could also duce radicals, is widely applied in the degradation of polysaccharides [36]. We TiO 2 as photocatalyst could generate electron-hole (e − and h + ) pairs under the light, and the excited electrons can further generate free radicals. H 2 O 2 , which could also produce radicals, is widely applied in the degradation of polysaccharides [36]. We have reported the photocatalytic degradation of fucoidan and chondroitin sulfate with TiO 2 as well as H 2 O 2 , and the products of 3~5 kDa were obtained after a 6 h reaction [25,26]. As a strong oxidizing agent, H 2 O 2 could promote the oxidation degradation [37], but it is harmful to health and unfriendly to environment [38]. Moreover, its addition could increase the production cost.
In the present study, 4 kDa degradation product of κ-carrageenan was produced through a 6 h reaction without H 2 O 2 . Of note, fucoidan and chondroitin sulfate could not be degraded obviously within 6 h (data not published). The efficient degradation of κ-carrageenan by photocatalytic reaction without H 2 O 2 may be attributed to its structure.
Notably, the results showed that sulfated polysaccharides can be effectively degraded by photocatalytic degradation method with only TiO 2 , which is a significant discovery.

Structure Changes of Carrageenan by Photocatalytic Degradation
In order to monitor the chemical structure changes of κ-carrageenan during the photocatalytic reaction, colourimetry, FT-IR spectroscopy, and NMR spectroscopy were applied. With the extension of photocatalytic time, the reducing sugar content increased from 0.85% to 14.41% ( Figure 3A), indicating the effective degradation by photocatalysis, which was consistent with molecular weight change. At the same time, no significant difference in sulfate group content was observed between κ-carrageenan and its degradation products ( Figure 3B). However, the total sugar content, sulfated polysaccharide content, and 3,6anhydro galactose content all decreased by around 20% after 6 h degradation ( Figure 3C-E), which reflected the breakdown proportion during the degradation. Of note, the uronic acid content showed an increasing trend, suggesting the byproduct as uronic acid ( Figure 3F).
ful to health and unfriendly to environment [38]. Moreover, its addition could inc the production cost.
In the present study, 4 kDa degradation product of κ-carrageenan was prod through a 6 h reaction without H2O2. Of note, fucoidan and chondroitin sulfate cou be degraded obviously within 6 h (data not published). The efficient degradation carrageenan by photocatalytic reaction without H2O2 may be attributed to its stru Notably, the results showed that sulfated polysaccharides can be effectively degrad photocatalytic degradation method with only TiO2, which is a significant discovery

Structure Changes of Carrageenan by Photocatalytic Degradation
In order to monitor the chemical structure changes of κ-carrageenan during the tocatalytic reaction, colourimetry, FT-IR spectroscopy, and NMR spectroscopy we plied. With the extension of photocatalytic time, the reducing sugar content incr from 0.85% to 14.41% ( Figure 3A), indicating the effective degradation by photocata which was consistent with molecular weight change. At the same time, no significan ference in sulfate group content was observed between κ-carrageenan and its degrad products ( Figure 3B). However, the total sugar content, sulfated polysaccharide co and 3,6-anhydro galactose content all decreased by around 20% after 6 h degradation ure 3C-E), which reflected the breakdown proportion during the degradation. Of the uronic acid content showed an increasing trend, suggesting the byproduct as u acid ( Figure 3F). The FT-IR spectra of κ-carrageenan, DC-2, DC-4, and DC-6 were presented in F 4. The absorption band at 1642 cm −1 was attributed to the stretching vibration of group [39]. The presence of signals at 1245 cm −1 and 850 cm −1 were assigned, respect The FT-IR spectra of κ-carrageenan, DC-2, DC-4, and DC-6 were presented in Figure 4. The absorption band at 1642 cm −1 was attributed to the stretching vibration of C=O group [39]. The presence of signals at 1245 cm −1 and 850 cm −1 were assigned, respectively, to O=S=O symmetric vibration and to -O-SO 3 stretching vibration at C-4 of the galactose ring. Importantly, the peak position, shape, and intensity of the presence of signals at 850 cm −1 showed that sulfate group was retained during the photocatalytic reaction. The absorption band at 929 cm −1 was attributed to the C-O-C vibration of the 3,6-anhydro-Dgalactose residue [40]. Furthermore, the absorption band at 1054 cm −1 was attributed to the stretching vibration of C-O-C on the sugar ring, and this peak was the characteristic absorption peak of pyranose ring. Notably, although great similarities were observed between κ-carrageenan and its degradation products, an absorption band at 1740 cm −1 corresponding to the stretching vibration of C=O grew with the extension of photocatalysis degradation, indicating the formation of carboxyl or aldehyde group [41,42]. the stretching vibration of C-O-C on the sugar ring, and this peak wa absorption peak of pyranose ring. Notably, although great similarities tween κ-carrageenan and its degradation products, an absorption ban responding to the stretching vibration of C=O grew with the extensio degradation, indicating the formation of carboxyl or aldehyde group To further explore the possible structural changes caused by phot tion, 1 H NMR, 13 C NMR, and HSQC spectra ( Figure 5) of DC-6 wer cross peaks at 4.63/102.1 ppm and 5.08/94.5 ppm in the HSQC spectr as anomeric proton/carbon signals of Gal4S and AnGal, respectively chemical shifts were also assigned in Table 1. Comparison of the NMR of κ-carrageenan [43,44] indicates that DC-6 has no obviously struc worth mentioning that no carbon signal was observed at 66.4 ppm co C-4 position of galactose without sulfate substitutions [45], indicating phurization occurs in the photocatalytic degradation. To further explore the possible structural changes caused by photocatalysis degradation, 1 H NMR, 13 C NMR, and HSQC spectra ( Figure 5) of DC-6 were investigated. The cross peaks at 4.63/102.1 ppm and 5.08/94.5 ppm in the HSQC spectrum were identified as anomeric proton/carbon signals of Gal4S and AnGal, respectively. Other 1 H and 13 C chemical shifts were also assigned in Table 1. Comparison of the NMR signals with those of κ-carrageenan [43,44] indicates that DC-6 has no obviously structural changes. It is worth mentioning that no carbon signal was observed at 66.4 ppm corresponding to the C-4 position of galactose without sulfate substitutions [45], indicating no obvious desulphurization occurs in the photocatalytic degradation. To further explore the possible structural changes caused by ph tion, 1 H NMR, 13 C NMR, and HSQC spectra ( Figure 5) of DC-6 w cross peaks at 4.63/102.1 ppm and 5.08/94.5 ppm in the HSQC spec as anomeric proton/carbon signals of Gal4S and AnGal, respective chemical shifts were also assigned in Table 1. Comparison of the NM of κ-carrageenan [43,44] indicates that DC-6 has no obviously stru worth mentioning that no carbon signal was observed at 66.4 ppm C-4 position of galactose without sulfate substitutions [45], indicati phurization occurs in the photocatalytic degradation.

Identification of Carrageenan Oligosaccharides Produced by Photocatalytic Degradation
The oligosaccharides in DC-6 were characterized by ESI-MS n in positive ion mode after PMP derivatization, and their extracted ion chromatograms and MS 2 spectra were shown in Figure 6. The oligosaccharides were identified mainly based on their pseudomolecular ion and product ions. As illustrated in Figure 6A   Similarly, oligosaccharides from disaccharide to hexasaccharide constructed with the alternate AnGal and Gal residues were found according to Figure 6D-J.
In the present study, a total of 20 oligosaccharides in the photocatalytic degradation products were identified, including di-, tri-, tetra-, penta-, hexa-, and hepta-saccharides (Table 2), and their relative yields could be reflected by their chromatogram peak areas. It is worth noting that sulfate content in the identified oligosaccharides is not as high as it has been determined in Section 3.2. This is due to the fragmentation through loss of sulfate groups and sensitivity problems of mass spectroscopy detection, particularly when analyzing oligosaccharides substituted with many sulfate groups [46,47]. The possible formation process of dAnGal was proposed in Figure 7. After the break of glycosidic bond of AnGal, the hexatomic ring of AnGal is opened. Then the new reducing end is subjected to a rapid oxidation to form the carboxyl group. Next, the decarbonation and oxidation reactions occur subsequently to generate dAnGal.
In the present study, a total of 20 oligosaccharides in the photocatalytic degradation products were identified, including di-, tri-, tetra-, penta-, hexa-, and hepta-saccharides (Table 2), and their relative yields could be reflected by their chromatogram peak areas. It is worth noting that sulfate content in the identified oligosaccharides is not as high as it has been determined in Section 3.2. This is due to the fragmentation through loss of sulfate groups and sensitivity problems of mass spectroscopy detection, particularly when analyzing oligosaccharides substituted with many sulfate groups [46,47]. The possible formation process of dAnGal was proposed in Figure 7. After the break of glycosidic bond of AnGal, the hexatomic ring of AnGal is opened. Then the new reducing end is subjected to a rapid oxidation to form the carboxyl group. Next, the decarbonation and oxidation reactions occur subsequently to generate dAnGal.  Moreover, the photocatalytic depolymerization rule was proposed in Figure 8 according to the degradation products. The major proportion (57.4%) of the degradation products demonstrated AnGal at their reducing end, and 17.5% was its decarbonylated derivative, dAnGal, at the reducing end. This suggests that AnGal's glycosidic bond is more ready to be attacked by free radicals. Notably, the rule of photocatalytic depolymerization was different from those of other depolymerization methods. Carrageenase could degrade κ-carrageenan to yield even-numbered oligosaccharides with AnGal at the non-reducing end and Gal at the reducing end [48], while acid hydrolysis could produce odd-numbered oligosaccharides with Gal at both the reducing and non-reducing ends [49]. In the present study, similar amounts of odd and even numbers of κ-carrageenan oligosaccharides were obtained by photocatalytic degradation, and more AnGal or its derivative were observed at the reducing ends. Thus, the degradation result of photocatalytic reaction was different from those of enzymatic degradation and acid hydrolysis.
Moreover, the photocatalytic depolymerization rule was proposed in Figure 8 according to the degradation products. The major proportion (57.4%) of the degradation products demonstrated AnGal at their reducing end, and 17.5% was its decarbonylated derivative, dAnGal, at the reducing end. This suggests that AnGal's glycosidic bond is more ready to be attacked by free radicals. Notably, the rule of photocatalytic depolymerization was different from those of other depolymerization methods. Carrageenase could degrade κ-carrageenan to yield even-numbered oligosaccharides with AnGal at the nonreducing end and Gal at the reducing end [48], while acid hydrolysis could produce oddnumbered oligosaccharides with Gal at both the reducing and non-reducing ends [49]. In the present study, similar amounts of odd and even numbers of κ-carrageenan oligosaccharides were obtained by photocatalytic degradation, and more AnGal or its derivative were observed at the reducing ends. Thus, the degradation result of photocatalytic reaction was different from those of enzymatic degradation and acid hydrolysis.

Rheological Properties of Carrageenan Degradation Products
In order to monitor the predominant elasticity and gel-like behavior of carrageenan during the photocatalytic degradation, the elastic (G′) and viscous (G″) moduli of carrageenan, DC-2, DC-4, and DC-6 in the time range of 0-160 s and the frequency sweep of 0.1-10 Hz were depicted. As shown in Figure 9, κ-carrageenan at 0.5% (w/w) could form a weak gel at 4 °C, and did not show obvious gel properties at 25 °C. At the same time, its degradation products, including DC-2, DC-4, and DC-6 all failed in the gel formation at both 4 °C and 25 °C. Moreover, their performance had not been improved with the change of frequency. It has been revealed that 3,6-anhydro-D-galactose residues allow a helicoidal secondary structure, which plays a crucial role in the formation of carrageenan gels [50]. Although the photocatalytic reaction did not break the structure blocks of carrageenan obviously, its gel property was deprived with the decrease of molecular weight. This is due to the length and the degree of entanglement of the molecular chain, which will be

Rheological Properties of Carrageenan Degradation Products
In order to monitor the predominant elasticity and gel-like behavior of carrageenan during the photocatalytic degradation, the elastic (G ) and viscous (G") moduli of carrageenan, DC-2, DC-4, and DC-6 in the time range of 0-160 s and the frequency sweep of 0.1-10 Hz were depicted. As shown in Figure 9, κ-carrageenan at 0.5% (w/w) could form a weak gel at 4 • C, and did not show obvious gel properties at 25 • C. At the same time, its degradation products, including DC-2, DC-4, and DC-6 all failed in the gel formation at both 4 • C and 25 • C. Moreover, their performance had not been improved with the change of frequency. It has been revealed that 3,6-anhydro-D-galactose residues allow a helicoidal secondary structure, which plays a crucial role in the formation of carrageenan gels [50]. Although the photocatalytic reaction did not break the structure blocks of carrageenan obviously, its gel property was deprived with the decrease of molecular weight. This is due to the length and the degree of entanglement of the molecular chain, which will be decreased during the degradation, resulting in the failure of gel formation. It follows that a sufficient molecular weight of polysaccharide is critical for the gel properties [51]. decreased during the degradation, resulting in the failure of gel formation. It follows that a sufficient molecular weight of polysaccharide is critical for the gel properties [51].

Cryo-SEM Microstructure of Carrageenan Degradation Products
Cryo-SEM has long been used for direct observation of the retained microstructure of the samples [52]. The microstructure of κ-carrageenan, DC-2, DC-4, and DC-6 were characterized by Cryo-SEM, and the results were shown in Figure 10. The gel pore wall thickness and pore area were determined with ImageJ; κ-carrageenan, DC-2, DC-4, and DC-6 had a pore wall thickness of 0.29 ± 0.06 μm, 0.16 ± 0.01 μm, 0.14 ± 0.02 μm, and 0.12 ± 0.01 μm, respectively. In addition, the pore area of κ-carrageenan was 21.50 ± 4.18 μm 2 while the pore structures of the degradation products were fragmented.

Cryo-SEM Microstructure of Carrageenan Degradation Products
Cryo-SEM has long been used for direct observation of the retained microstructure of the samples [52]. The microstructure of κ-carrageenan, DC-2, DC-4, and DC-6 were characterized by Cryo-SEM, and the results were shown in Figure 10. The gel pore wall thickness and pore area were determined with ImageJ; κ-carrageenan, DC-2, DC-4, and DC-6 had a pore wall thickness of 0.29 ± 0.06 µm, 0.16 ± 0.01 µm, 0.14 ± 0.02 µm, and 0.12 ± 0.01 µm, respectively. In addition, the pore area of κ-carrageenan was 21.50 ± 4.18 µm 2 while the pore structures of the degradation products were fragmented. decreased during the degradation, resulting in the failure of gel formation. It follows that a sufficient molecular weight of polysaccharide is critical for the gel properties [51].

Cryo-SEM Microstructure of Carrageenan Degradation Products
Cryo-SEM has long been used for direct observation of the retained microstructure of the samples [52]. The microstructure of κ-carrageenan, DC-2, DC-4, and DC-6 were characterized by Cryo-SEM, and the results were shown in Figure 10. The gel pore wall thickness and pore area were determined with ImageJ; κ-carrageenan, DC-2, DC-4, and DC-6 had a pore wall thickness of 0.29 ± 0.06 μm, 0.16 ± 0.01 μm, 0.14 ± 0.02 μm, and 0.12 ± 0.01 μm, respectively. In addition, the pore area of κ-carrageenan was 21.50 ± 4.18 μm 2 while the pore structures of the degradation products were fragmented.  Overall, κ-carrageenan showed a typical gel network structure, which had large pores and thick and homogeneous pore walls. Notably, the pore walls of degradation products became thinner, and imperfectly linked with some filamentous structures. It could be noticed that with the extension of photocatalytic time, the degradation products became irregular and broken. The microstructure features are in line with the result that the degradation products would lose their gel-like behavior with the increase of degradation time.

Conclusions
The present study developed a photocatalytic method for carrageenan degradation and by using TiO 2 as the only catalyst, this method could decrease the average molecular weight of κ-carrageenan to 4 kDa within 6 h. Notably, no significant loss of sulfate groups in the reaction was observed, which is critical for its application in preparing sulfated degradation products. Moreover, the degradation rules were proposed according to the degradation product characterization, and the glycosidic bond of AnGal was found to be more susceptible to break. In addition, the present study demonstrated the decreasing gel property of carrageenan alone with the photocatalytic degradation. Thus, the present study indicates the promising application of photocatalytic degradation with TiO 2 to prepare sulfated low-molecular-weight carrageenan, which may further broaden the application of carrageenan.