Impact of Digestion on Surface Microstructures of Microplastics

Recently, there have been more reports and concerns about microplastics from various media and journals in the world. Except for land and water pollutions, much attention has been given to the impact of microplastics on animal and human health. However, previous studies have shown that digestion could not affect the particles of various microplastics due to their stable behavior. This work focuses on the impact of articial digestion on the surface microstructures of microplastics from well recognized popular sources such as PP, PE, PET, PS and PVC. SEM and AFM were used to study the impact of articial digestion on the surface morphologies of microplastics; while ATR-FTIR and XPS were used to investigate the impact of articial digestion on the chemical properties of the surface of the microplastics. There were no physical differences observed by both SEM and AFM. There were no signicant chemical differences detected by both FTIR and XPS after treatment. The slight differences of resolved C1s spectra for PS and PET samples detected by XPS should be further investigated. The study results show that the digestion system could not decompose the microplastics. Generally, plastic particles are widely considered to be inert due to their low chemical reactivity.


Introduction
Plastics pollutions are traditionally considered as environmental issues [1]; they are more or less seen as pollution problems [2,3]. More attention has been given to health issues recently in relation to pollution from microplastics [4,5]. There have been a lot of news and reports about microplastics from various media and scienti c journals recently in the world. The reports show that microplastics cause pollution in the soil [6], air [7], fresh water and sea [8]. For example, the tea made from the popular plastic tea bags in the market contains up to tens of billions of microplastic particles [9], and washing clothes can cause the release of micro bers to water [10], and more than 90% of the 39 different salt brands worldwide contain microplastics [11]. According to media reports, researchers have discovered "traces" of plastic particles in the Arctic and are shocked by the number. Although the Arctic is inaccessible, there are more than 10,000 plastic particles per liter of snow [12]. Recent researches have shown that various food products were contaminated with microplastic particles, indicating a widespread exposure [13][14][15][16][17][18][19][20]. Microplastics can be enriched in food chain and also introduced into foodstuffs through the air, indicating the ubiquitous presence of plastic particles [21,22].
There is no uniform de nition of microplastic size, but a well acceptable size range is between 100 nm and 5 mm. Plastic particles below 100 nm are commonly de ned as nano-plastics [23,24]. The particles below 150 µm in size are considered to be systemically bioavailable to humans, and particles below 4 µm can be taken up by intestinal cells [24,25]. Hanvey et al [26] have tried initially to standardize analytical methods in order to better estimate human exposure. So far, how microplastics affect human health is still far from being fully understood. Plastic materials are usually considered as very unreactive chemically due to their large molecular sizes. A potential bioreactivity of microplastic particles is therefore more likely to be expected due to their increased surface-to-volume ratio with decreasing particle sizes due to their chemical properties. However, more recently, Stock et al [27,28] systematically investigated the impact of the gastrointestinal passage on the physicochemical particle characteristics of the ve most produced plastics (PE, PP, PVC, PS and PET). The results showed that all the plastic particles were highly resistant to the arti cial digestive juices, and the main stages of the human gastrointestinal tract did not decompose the particles.
In this work, it is hypothesized that arti cial digestion may affect only the surface of the microplastics or the whole particles. This work focuses on the surface microstructures and chemical properties of these microplastics, particularly the variations in their atom scale and chemicals. The impact of arti cial digestion on the microplastics was studied by SEM, AFM, XPS and ATR-FTIR. This is essential for a reliable future risk assessment of microplastics.

Preparation of stock solutions
In vitro simulating digestion process used in this work was based on Minekua [29] and Andre Brodkorb [30]. The simulated digestion consists of three phases: oral phase, gastric phase and intestinal phase.  Table 1 below.

Digestion treatment
All the electrolyte stock solutions (SSF, SGF and SIF) were rstly pre-heated to 37°C to achieve human body temperature. Dispersed 0.5 g of microplastics (PP, PE, PS, PET, and PVC) powder into 5 ml water respectively, then mix microplastics solutions with 5 ml SSF electrolyte stock solution, added salivary amylase into the mixture to reach 75U mL − 1 in the nal mixture. It was incubated and then mixed in a water bath with integrated shaker at 37°C for 5 minutes. Oral bolus was mixed with 10 ml SGF electrolyte stock solution, then added porcine pepsin to achieve 2000 U/ml in the nal digestion mixture. The hydrochloric acid standard solution(1N) was used to adjust the pH to 3.0. It was incubated and mixed at 37 ℃ for 2 h. Add 20 ml SIF electrolyte stock solution into gastric chyme to achieve a nal ratio 1:1(vol/vol). Adjust the pH to 7.0 by adding sodium hydroxide solution (0.5N). Add the bile salt and pancreatin to the SIF/gastric chyme to reach a nal concentration of 100 U/mL and 10Mm. Verify the pH and adjust to 7. Incubated and mixed the mixture at 37 ℃ for 2 h as gastric phases. The mixture was sieved and rinsed with ultrapure water 3 times, then dry the sample for later use.

Scanning electron microscope (SEM)
The surface morphologies of the microplastic samples were studied by using the scanning electron microscope (Zesis EVO 18, German). The microplastics powders were held by a quadrilateral brass stub using a conductive resin. The samples were coated with Au under high vacuum conditions. The SEM observation was operated at an acceleration voltage of 10 KV and 2000 X, 10000 X magni cation.

Atomic force microscopy (AFM)
The surface topography was studied by atomic force microscopy (Bruker-JPK Bio AFM, Germany). Particle solutions were dropped on monocrystalline silicon wafer and dried at room temperature. Then, they were placed on the sample stage. The standard mode for determining the height of the surface structures was applied for surface analysis. AFM imaging was conducted in an area of 30 30 µm 2 .

Fourier transform infrared spectroscopy (FTIR)
FTIR was used to study the molecular structure of the particles and their chemical bonds. Spectrometer FTIR equipped with an attenuated total re ectance (ATR) was used to study the chemical properties of the microplastics. The ATR-FTIR spectra were observed using a Thermo Scienti c iS50 ABX, USA. FTIR spectra were obtained at a resolution of 4 cm − 1 with 64 accumulations in the range of 4000 − 600 cm − 1 .

X-ray photoelectron spectroscopy (XPS)
The powder samples were placed on non-conductive tape to the sample holder and moved into a spectrometer fast entry lock chamber. The chemical compositions of the samples' surfaces and element oxidation states were determined by using X-ray photoelectron spectroscopy (XPS, Axis Ultra DLD, KratOs, UK). XPS spectra were operated using Al Kα radiation (E = 1486.6 eV). All the binding energies were calibrated with respect to the C 1s peak at 284.8 eV [31].

Impact of arti cial digestion on the surface morphologies of microplastics
The effect of the arti cial digestion on the surface morphologies of the microplastics was studied using SEM and AFM. Figure 1 shows the SEM morphologies of the various microplastics before and after digestive treatment. It is seen that except for the PS microplastics with spherical particles, all the other microplastic materials (PE, PP, PET and PVC) originally had irregular shape particles; their diameters were between 1 and 20 µm, which were well fed into the microplastic scale. PVC had the smallest particle size and clustered together. PS, PE and PP had reasonable smooth surface, while many deformations and bers were observed on the surface of PET particles. After arti cial digestion, all the microplastic particles kept their original particle size and morphologies. There is no damage phenomenon on the surfaces of the various microplastics, such as hole, ber, rough surface or corrosion marks.
The surface images of individual PS particles were enlarged to study the impact of arti cial digestion on the sample's surfaces in detail. The particles before and after treatment show similar morphologies: reasonable smooth surface with some marks probably from processing. There are no differences or damages observed. All the other microplastics have similar phenomena: there is no morphological difference observed, no holes, deformation stress, ber, rough surface or corrosion marks was detected.
This indicates the arti cial digestion could not damage these microplastics.
AFM provides high-resolution images that show the microstructure and surface morphology of molecules directly by measuring the interactive forces between the probe and atom. Fig. 2 shows the typical AFM images of PET based on a, a+) height retrace image; b, b+) amplitude Retrace image and c, c+) 3D image respectively.AFM revealed the surface characteristics and particle sizes of the samples. The results indicate that the arti cial digestion generally did not affect the surface roughness of the samples. There were no signi cant differences between the surface structure and roughness of the PET microplastics before and after treatment.
The height retrace image was used to present the image in detail. Figure 3 shows the height retrace image of the various microplastics before and after treatment. It is seen that there is no observable difference. There are no damage phenomena such as holes, ber, rough surface or corrosion marks observed. The results corresponded with SEM observation (see Fig. 1). Both results indicated the arti cial digestion could not change the particle sizes and morphologies as well as the surface microstructures of the samples.

Impact of arti cial digestion on the chemical composition of the samples' surface
The impact of arti cial digestion on the chemical composition of the samples' surfaces was studied by ATR-FTIR and XPS (see Fig.4). The ATR-FTIR curves provide all the microplastics before and after treatment. The study particularly focused on the ranges of 4000−600 cm −1 wave-lengths. The spectrum of PE particles before and after treatment showed characteristic molecular vibrations at 2950 cm −1 , 1460 cm −1 and 720 cm −1 , representing the C-H bond. The spectrum of PP particles before and after treatment presented peaks at 1460 cm −1 and 1378 cm −1 referred to -CH 2 bending vibration and -CH 3 bending vibration. The bands at 970 cm −1 and 1155 cm −1 were assigned to [CH 2 CH(CH 3 ) n ] functional group. The FTIR spectrum of PVC particles before and after treatment displayed characteristic bands at around 2920 cm −1 (-CH 2 asymmetric stretching vibration mode), 1430 cm −1 (-CH bending vibrations bond of CH 2 ), 1250 cm −1 (C-H bending vibration near Cl), 1100 cm −1 (C-C stretching vibration), and 960 cm −1 (-CH waging vibration). The peak in the range of 600-700 cm −1 was attributed to the C-Cl stretching vibration. In the FTIR spectrum of the PS particles before and after treatment, the peaks in the range of 2900-3000 cm −1 were related to -CH 2 asymmetric and symmetric stretching vibration; and the bands at 1600 cm −1 , 1493 cm −1 , 1460 cm −1 and 760 cm −1 were assigned to aromatic rings. The FTIR spectrum of PET particles before and after treatment showed a characteristic C=O at approximately 1715 cm −1 and a typical C-O-C stretching vibration at approximately 1100 cm −1 and 1250 cm −1 ; the bands at 725 cm −1 and 875 cm −1 were assigned to aromatic rings. Based on the above results, it is seen that there is no observable difference for all the samples before and after treatment. There are no new peaks, weakened peaks or shift peaks detected, indicating no chemical reaction was detected.
As organic polymers, carbon and oxygen were used to present the variation of the chemical impact by XPS. Both carbon and oxygen elements can be clearly detected for all samples. There are no new peaks, weakened peaks or shift peaks detected, indicating no chemical reaction was detected. Table 2 lists the detailed data. It is seen that the atomic percentages of carbon to oxygen on the various microplastics display no remark difference before and after arti cial digestion treatment. The results corresponded with ATR-FTIR results. As carbon-based materials, the carbon peaks C1s of these samples were deconvoluted according to their binding energy to study the impact in detail [32][33][34]. Figure5 shows the deconvolutedcarbon peaks detected by XPS for the various microplastics. The peak C1s for the microplastics were deconvoluted into 2 or 3 peaks at corresponding binding energy, which represents different chemical bonds of the carbon element. C1 s spectra were deconvoluted at the peaks of 284.8 eV (C-C/ C-H), 286 eV (C-O), 289 eV (C=O) and 287 eV (C-Cl) respectively. It is seen that there are no obvious changes in the position of binding energies, peak number and shape of the deconvoluted carbon spectra for each microplastic before and after treatment. There is no prominent change of the intensity for the C-C, C-O bonds in PE, PVC and PP microplastics, which con rmed that arti cial digestion has no impact on the microplastics. The results correspond with ATR-FTIR analysis. It is observed that the peak areas of C=O for PS and PET samples are slightly different after treatment. The peak area of C-O for PET also decreased. The uctuation of the proportions of C=O and C-O bonds can be attributed to the additives in the polymer or the partial oxidation of the surface of the microplastic PET [35]. It could also be the error of measurement or divided processing and calculation since XPS is a quality other than quantity technology in many cases. The details will be investigated in our future work. Plastic materials are usually considered as chemically unreactive due to their large molecules even in micro-size particles. The results demonstrated a high resistance of all the plastic particles to the arti cial digestive juices, and the main stages of the human gastrointestinal tract did not decompose the particles. Previous articles that report the adverse effect of digestive juice on microplastics are limited, and the research of consequent toxicity is in its infancy as well. Stock et al [28] showed that human gastrointestinal tract did not have adverse effect on particle sizes and shapes of microplastics. They revealed that the uptake of microparticles by intestinal epithelial cell line is dependent on the material and size of particles; it is only when exposed to high concentrations that is more than dietary exposure, microplastics would lead to cytotoxic effects. The cellular uptake of small amounts of PS microplastics, and PS particles do not pose health risk to mammals under the corresponding experimental conditions. It is expected that the nanoparticles from plastics could be harmful to man's health since the nanoparticles can penetrate into cells [36]. Recently, Wang et al [37] investigated that nano-size polystyrene which was transformed by arti cial gastric uid greatly damaged hepatocytes. Actually, all materials can be in nano-size naturally and their effect on human health are complex and not understood. Hence, the present study did not discuss the impact of nano-particles on man's health.
Finally, it should be pointed out that all the microplastics used in this work are arti cially made through granulating virgin materials as done in most previous reports. Practically, it is hard to collect the real microplastics from nature and identify their years. It is expected that the actual microplastics in nature may have different molecular weight and molecular weight distribution; they might even have slightly different chemicals, such as less plasticizer content or contain some by products through decomposition. The future work should consider this factor.

Conclusion
Both SEM and AFM studies show that the arti cial digestion could not change the particle size and morphologies as well as surface microstructures of all the microplastics PE, PP, PVC, PS and PET used in this work. There is no observable difference or damage phenomenon observed, such as holes, deformation strength, ber, rough surface or corrosion marks. The results indicate that arti cial digestion could not change these microplastics. The impact of arti cial digestion on the chemical composition of Typical AFM images of PET microplastics before (a, b, c) and after (a+, b+, c+) arti cial digestion treatment: height retrace image (a, a+), amplitude Retrace image (b, b+) and 3D images (c, c+) Deconvoluted carbon peaks detected by XPS for the various microplastics before and after arti cial digestion treatment