Introduction

Plastic particles in the size range of 1 μm-5 mm are known as microplastics (MPs)1, which accumulate in the soil, the sea, and other environments2,3. Thus, MPs enter the human body through the ingestion of food and may cause disease4. Interactions between MPs and other organic pollutions in the environment, such as polycyclic aromatic hydrocarbons (PAHs)5, may lead to combined ecotoxicity6,7,which represents a more serious threat to human the health and life of many organisms. Investigating the adsorption effect of PAHs on MPs and establishing the related adsorption kinetic and thermodynamic models will enable quantitative analysis of the mechanism governing PAH adsorption on MPs8.

The 3D morphology of MP particles is an important factor that affects their adsorption of PAHs9. MP particles may be rough or smooth, have large or small pores, and be in a glassy or rubber state, resulting in different PAH loading capacities and behaviors. Besides focusing on morphology, research on PAHs adsorption by MPs has primarily focused on modeling the adsorption effect over time based on quantitative measurements of the changing PAH concentration10,11,12. Measuring the physical morphology in situ, while measuring the concentration changes in the solution environment would enable a more accurate and comprehensive mechanistic analysis of PAH adsorption by MPs in real-time.

However, the 3D morphology of MPs and the concentration of PAHs cannot be measured simultaneously using available methods because these factors are evaluated using different technical means. For example, scanning electron microscopy (SEM)13,14, atomic force microscopy (AFM)15, and optical interferometric microscopy have been applied to measure the 3D morphology of MPs. SEM is the most common technique for MP morphological measurements, although the resolution of AFM can be at the nanometer scale16. However, SEM requires spraying gold on the surface of microplastics, while AFM requires physical contact and scanning. These methods are limited in terms of enabling real-time and in-situ observations of the physical structure while detecting chemical changes. Digital holographic microscopy (DHM) is a real-time, dynamic, non-contact measurement technology in the realm of optical interferometric microscopy. This method has been used for 3D morphological measurements of MPs and to detect and position the density of MPs in seawater17,18,19,20,21,22. It is non-destructive technique and free of markers. The measuring precision can be nanometer scale. Commonly, changes in PAHs concentration due to adsorption by MPs can be monitored using ultraviolet spectrophotometry23, fluorescence spectrophotometry24, liquid chromatography25, or other detection methods. Therefore, the simultaneously measurement of 3D morphology of MP particles in solution and the concentration of solution is not feasible.

In fact, the refractive index (RI) of a liquid is directly related to the concentrations of its constituent elements, i.e., its composition26. Traditionally, RI measurements have relied on geometric optics, e.g., the Abbe refractometer27, the V-prism method, or a wave optics method involving interference28, an optical frequency comb29, or surface plasmon resonance30. Geometric optics-based methods rely on measuring angles, the accuracy of which is limited by the angular measurement sensors. DHM is an interferometric method that can determine the RI by reconstructing the material phase, thus exhibiting advantages over other interferometry methods, i.e., real-time dynamic and in situ measurements. However, using the traditional DHM method for RI measurements can lead to complexity or inaccuracies due to the two-fluid setup31, the microfluidic device32, or the dual-wavelength sources33. In our previous work, the average phase difference method was proposed to simplify the setup and improve the measuring precision of RI measurement34. The application of phase difference method (PDM) in chemical analysis has not been explored practically. Yet the chemical analyzing performances between PDM based on DHM and common techniques are not compared.

This report proposes a method based on digital holography that can be used to simultaneously measure the 3D morphology of MPs in a PAHs solution and the dynamic PAHs concentration. The experimental setup is a traditional single-wavelength and off-axis DHM system with a simple triangular cuvette. The PAHs concentration is monitored by RI measurement. Comparing the PAHs concentration results obtained by the PDM and by fluorescence spectrophotometry confirm that the PDM achieves a higher signal-to-noise ratio (SNR) and superior real-time performance. Experimental morphological observations of polypropylene (PP), polystyrene (PS), and Polyethylene (PE) MPs and kinetic analysis of naphthalene adsorption further verify that the proposed PDM based on DHM offers an in situ, real-time, and quantitative measurement technique for analyzing the mechanism governing PAH adsorption by MPs.

Methodology

Measuring the 3D morphology of MP particles

In single-wavelength DHM, the phase information of the sample is determined using Eqs. (1),

$$\phi {\text{=}}arc\tan \left\{ {\frac{{\operatorname{Im} \{ IFT\{ FT({I_{ho\log ram}}) \cdot W \cdot H\} \} \cdot {{{\text{\{ IFT\{ FT}} {I_{reference}}\\cdot W \cdot H{\text{\} \} }}}^*}}}{{\operatorname{Re} \{ IFT\{ FT({I_{ho\log ram}}) \cdot W \cdot H\} \} \cdot {{{\text{\{ IFT\{ FT}} {I_{reference}}\\cdot W \cdot H{\text{\} \} }}}^*}}}} \right\}$$
(1)

where FT and IFT represent the Fourier transform and inverse Fourier transform, respectively; Ihologram and Ireference denote the holograms recorded with and without the sample, respectively; W is the low-pass filtering window used to eliminate real image artifacts; H is the angular spectrum reconstruction factor, also known as the spatial frequency transfer function; and ∗ indicates the conjugate term. The imaginary (IM) and real (RE) parts of the reconstructed wavefront are also used in the computation:

$$H\left( {{f_x},{f_y}} \right)=\exp \left[ {jkd\sqrt {1 - {{\left( {\lambda {f_x}} \right)}^2} - {{\left( {\lambda {f_y}} \right)}^2}} } \right]$$
(2)

where fx and fy are the coordinates in the frequency spectrum, λ is the wavelength, d is the reconstructing distance, and k is the wavenumber.

The axial dimension AD of an MP particle is expressed in Eqs. (3),

$$AD=\frac{{\phi \lambda }}{{2\pi n}}$$
(3)

where n is the coupled refractive index of the solution and particle. Because the solution and particle are both homogeneous, the phase of the MP particle is directly related to its axial depth. Therefore, the 3D morphology and structure of MP particles can be revealed by the phase.

Measuring the PAHs solution concentration

The RI of a liquid is linear with respect to the concentration. The PDM proposed herein can determine the RI of the solution. Instead of using a complicated setup (e.g., a two-fluid setup, microfluidic device, or dual-wavelength sources), a simple triangular cuvette with a known geometry and a single-wavelength DHM system was used for the PDM34. The relative errors of PDM ranged from 0.03 to 0.55%. The precision of PDM is 0.0014–0.2647%. Figure 1a shows the triangular cuvette.

Fig.1
Fig.1
Full size image

(a) Image and (b) optical path diagram of triangular cuvette holding measured liquid.

As shown in Fig. 1b, two parallel rays (Ray 1 and Ray 2) pass through the triangular cuvette. The incident points of Ray 1 and Ray 2 are A(XA, YA) and B(XB, YB) in the recording plane. Based on Snell’s Law, the phase difference △φ between the two rays at distance D can be calculated using Eqs. (4),

$$\begin{gathered} \Delta \phi =f(\theta ,{\theta _1},D,{n_{air}},{n_{liquid}}) \hfill \\ =2\pi (D*{n_{air}}*sin(\theta ))/({(1 - sin{(\theta )^2}*{(sin({\theta _1})+{({n_{liquid}}^{2}/{n_{air}}^{2} - sin{(\theta )^2})^{(1/2)}})^2})^{(1/2)}}*(1 \hfill \\ - ({n_{air}}^{2}*sin{({\theta _1})^2})/{n_{liquid}}^{2}{)^{(1/2)}}) - (D*{n_{air}}*sin(\theta ))/(\cos (\theta - arcsin(sin(\theta )*(sin({\theta _1}) \hfill \\ +{({n_{liquid}}^{2}/{n_{air}}^{2} - sin{({\theta _1})^2})^{(1/2)}})))*{(1 - sin{(\theta )^2}*{(sin({\theta _1})+{({n_{liquid}}^{2}/{n_{air}}^{2} - sin{({\theta _1})^2})^{(1/2)}})^2})^{(1/2)}})/\lambda \hfill \\ \end{gathered}$$
(4)

where θ is the geometric parameter of the cuvette, θ1 is the incident angle, and nair and nliquid are the RIs of air and the tested liquid, respectively. D is the corresponding distance between pixel point A and pixel point B in object space.

Two liquids with known RIs were used to calibrate the unmeasurable θ1, which enables the determination of nliquid. In this study, purified water (RI is 1.3330, 20℃) and anhydrous ethanol (RI is 1.3611, 20℃) were utilized as calibrating liquids.

According to the electron theory of Lorentz and Lambert’s law, the relationship between adsorption coefficient K of solution and extinction coefficientηis

$$K=\frac{{4\pi }}{\lambda }\eta \approx \frac{{4\pi \gamma \omega }}{{\lambda ({\omega _0}^{2} - {\omega ^2})}}n - \frac{{4\pi \gamma \omega }}{{\lambda ({\omega _0}^{2} - {\omega ^2})}}+\frac{{\pi \gamma \omega }}{{\lambda ({\omega _0}^{2} - {\omega ^2})}}\frac{{{N^2}{e^4}}}{{2{\varepsilon _0}^{2}{m^2}}}\frac{{{{({\omega _0}^{2} - {\omega ^2})}^2} - {\gamma ^2}{\omega ^2}}}{{{{[{{({\omega _0}^{2} - {\omega ^2})}^2}+{\gamma ^2}{\omega ^2}]}^2}}}$$
(5)

λ is the wavelength of transmitted light, e is the charge of electrons, ω0 is the natural frequency of electrons, ω is the frequency of the light wave field, m is electron mass, c is the speed of light in a vacuum, γ is classic damping coefficient of radiation. The adsorption coefficient K is linear to the concentrationδof the solution. Therefore, the concentration is

$$C=\frac{{4\pi }}{{\alpha \lambda }}\eta \approx \frac{{4\pi r\omega }}{{\alpha \lambda ({\omega _0}^{2} - {\omega ^2})}}n - \frac{{4\pi r\omega }}{{\alpha \lambda ({\omega _0}^{2} - {\omega ^2})}}+\frac{{\pi r\omega }}{{\alpha \lambda ({\omega _0}^{2} - {\omega ^2})}}\frac{{{N^2}{e^4}}}{{2{\varepsilon _0}^{2}{m^2}}}\frac{{{{({\omega _0}^{2} - {\omega ^2})}^2} - {r^2}{\omega ^2}}}{{{{[{{({\omega _0}^{2} - {\omega ^2})}^2}+{r^2}{\omega ^2}]}^2}}}$$
(6)

α is a constant independent of the concentration. From the Eqs. (5) and (6), it is concluded that when the frequency of the optical field is a constant, the refractive index of the solution is linear to concentration. Based on the fitting experiments on standard salt solutions and anhydrous ethanol, the concentration of solution C was revealed using Eq. (7):

$${n_{liquid}}=1.33305+0.00296C$$
(7)

Kinetic model of adsorption

The adsorption at time t, denoted as qt, represents the mass of the adsorbate (adsorbed substance) per unit mass of the adsorbent (MP particles) at a given time. which can be calculated using Eqs. (8),

$${q_t}=\frac{{({C_0} - {C_t})V}}{m}$$
(8)

where C0 is the initial concentration (mg/L), Ct is the concentration at time t(mg/L), V is the volume of the solution (L), and m is the mass of the MP sample (g).

The experimental data were fitted to pseudo-first-order and pseudo-second-order kinetic models. The kinetic constants (k1, k2) and their respective correlation coefficients (R2) were calculated to measure the adsorption rate and the suitability of both kinetic models in this study.

Adsorption isotherms

The equilibrium characteristics of the adsorption process, the maximum adsorption capacity and the adsorption strength constants are elucidated by different isotherm models, which are built by fitting the experimental data. The commonly used isotherm models are Langmuir and Freundlich models.

The Langmuir model is expressed as:

$$\frac{{\text{1}}}{{{q_e}}}=\frac{1}{{{q_{\hbox{max} }}{K_L}}}+\frac{1}{{{q_{\hbox{max} }}}}{C_e}$$
(9)

The Freundlich model is given by:

$${q_e}={K_F}C_{e}^{{1/n}}$$
(10)

Where qe denotes the equilibrium adsorption capacity, qmax represents the maximum adsorption capacity, and KL and KF are the adsorption strength constants.

Adsorption thermodynamics

Calculating the thermodynamic parameters, such as Gibbs free energy (ΔG), enthalpy change (ΔH), and entropy change (ΔS), provide a better understanding of the energetics driving the adsorption mechanism and the influence of temperature on adsorption behavior.

ΔG can be calculated using Eqs. (11),

$$\Delta G{\text{=}}\ RT\ln {K_C}$$
(11)

where KC is the equilibrium distribution coefficient, R is the ideal gas constant, and T is the absolute temperature (K).

Similarly, ΔH and ΔS can be derived from the Van’t Hoff equation,

$$\ln {K_C}=\frac{{\Delta S}}{R}\ \frac{{\Delta H}}{{RT}}$$
(12)

which allows evaluations of the heat of adsorption and the disorder associated with the adsorption process.

Specific surface area analysis

The specific surface area (SSA), defined as the total surface area per unit mass of a solid (m²·g⁻¹), is a key parameter for characterizing microplastics (MPs). To investigate the effect of SSA on adsorption performance, the SSA of PS, PP, and PE microplastics was determined by multipoint Brunauer-Emmett-Teller (BET) analysis using a BET Surface Area Analyzer. The adsorption capacities were further normalized by SSA to assess the adsorption efficiency per unit surface area, calculated by dividing the adsorption capacity by the SSA as Eq. (13):

$$q_{{\text{e}}}^{{{\text{norm}}}}=\frac{{{q_e}}}{{SSA}}$$
(13)

where \(q_{e}^{{norm}}\) represents the adsorption capacity normalized by SSA.

Experimental results and discussion

Experimental setup

Fig.2
Fig.2
Full size image

Experimental setup.

The experimental setup was a typical off-axis DHM system (Fig. 2). The laser beam from the laser diode module (Thorlabs, LDM670, = 670 nm) is split into an object beam and a reference beam. The tested liquid is in the triangular cuvette (Purshee Experiment, PXGX_QX_SJX). The object beam is magnified by microscopic objective MO1 (Olympus, 20×, NA = 0.4). The reference beam is magnified by microscopic objective MO2 (Olympus, 10×, NA = 0.25). CCD (Charge Coupled Device, Thorlabs, DCC1545M, 5.2 μm×5.2 μm, 1080pixels×1440pixels) records the hologram. To ensure a constant geometric index in PDM, the triangular cuvette should be fixed in the holding mechanism.

Experimental results

(1) 3D morphologies of MP particles

The three types of MP particles tested in this study, i.e., PS, PP, and PE (Huaian Ruixiang New Material Co., LTD), had particle sizes ranging from 50 to 100 μm. The PAH used for the adsorption experiments was naphthalene (TCI, Shanghai, Chemical Industry Development Co., LTD, purity > 99%).

In this study, DHM was applied for the 3D phase reconstruction of three types of MP particles to analyze their morphological characteristics.

The hologram of PE particle in naphthalene solution is shown in Fig. 3a. With intense interference fringes, the carrier frequency is high enough to separate real image term from zero-order term. The frequency spectrum of hologram is shown in Fig. 3b, while the spectrum after low-pass operation is presented in Fig. 3c. The phase of particle is reconstructed from separated real image frequency spectrum. Meanwhile, the RI of naphthalene is calculated by Eq. (4) based on the averaging phase difference of region A and region B, which is referred in Fig. 3a. It should be noted that the two regions are randomly selected and with same area.

Fig.3
Fig.3
Full size image

The hologram of PE particle in naphthalene solution (a), frequency spectrum images before (b) low-pass filtering (c).

The morphologies of PE, PP, PS particles in naphthalene solution are shown in Fig. 4. The holograms of three types of the above particles are shown in Figs. 4a, c and e. The three-dimensional morphologies of three types of MP particles are exhibited by the phase images shown in Figs. 4b, d and f.

The morphological maps obtained by DHM highlight structural differences among the three MP particles’ surfaces: the PE particles exhibit significant surface wrinkles; the PP particles have smooth and flat surfaces; and the PS particles have a generally smooth surface with slight wrinkles. These SEM observations are consistent with the morphological features reported previously33, further validating the reliability and accuracy of DHM for morphological analysis.

Fig.4
Fig.4
Full size image

The holograms and phases of PE (a, b), PP (c, d) and PS(e, f) particles.

(2) PAH adsorption on MPs

Before the adsorption experiments, the MP particles were subjected to ultrasonic cleaning in deionized water to remove surface contaminants. The clean particles were then dried at 50 °C and stored.

Naphthalene was prepared as a 2.0 mg/L stock solution using a standard reference material. The stock solution was diluted to obtain various concentrations, as required by the experimental protocol.

a. Calibration

To further validate the proposed PDM, the naphthalene concentration was also measured by fluorescence spectrophotometry (PerkinElmer, LS-55) for comparison. The relationships between the naphthalene concentration and the fluorescence intensity and the RI were established via calibration experiments. The naphthalene solution concentrations ranged from 0.1 to 2.0 mg/L; the measurement was repeated 50 times. To compare the consistency of PDM and fluorescence spectroscopy in concentration measurement, based on the calibration results (0.2-2.0 mg/L naphthalene solution), the concentration response curves of both methods were fitted, as shown in Fig. 5a. The normalized data of Fig. 5a is shown in Fig. 5(b). The normalized response slopes of PDM and fluorescence spectroscopy are 0.6225 and 0.6286 respectively, with a difference of less than 1%, which indicates that the two methods have statistical consistency in their responses to concentration changes, and can be used as complementary validations for each other.

Fig.5
Fig.5
Full size image

The comparison between PDM and fluorescence spectroscopy in concentration measurement. (a) The concentration response curves of PDM and fluorescence spectroscopy. (b) The normalized data of Fig. 5a.

The fit accuracy (R²) and residual sum of squares (RSS) were also evaluated. The SNRs (Eq. 14)35 of the two methods were compared to evaluate their sensitivity and anti-noise capability,

$$SNR{\text{=}}\frac{{\Delta {\text{y}}}}{{{\sigma _{residual}}}}$$
(14)

where Δy represents the signal variation range, and σ residual is the standard deviation of the residuals from the fitting process.

The fitting model for fluorescence spectrophotometry is given in Eq. (15):

FI = 11.20475 + 550.3167 × concentration (15).

The calibration results are provided in Table 1. The SNRs of fluorescence intensity and refractive index at each concentration point in Fig. 5(a) were calculated by Eq. (14). The average value \(\overline {{SNR}}\)of SNRs is shown in Table 1.

Table 1 Calibration results.
Full size table

From Table 1, the \(\overline {{SNR}}\)of the PDM is approximately 38.02% higher than that of fluorescence spectrophotometry.

b. Adsorption kinetics

The naphthalene concentration variations with time during adsorption were measured by PDM and fluorescence spectrophotometry. The nliquid corresponding to various concentrations of naphthalene was also measured as part of the experiment and is presented in Table 2. Additionally, the measurement density and temporal trends of the refractive index in adsorption are illustrated in Fig. 6.

The experiments were conducted in nine groups, each with 20 measurements. The measured data were fitted to pseudo-first-order and pseudo-second-order kinetic models to analyze the kinetic characteristics and adsorption rate.

Table 2 Naphthalene concentration and corresponding refractive index values at different time Intervals.
Full size table
Fig.6
Fig.6
Full size image

Measurement density and temporal trends of refractive index in adsorption.

The fitting results for the experimental data are presented in Table 3. The adsorption processes for all three MPs followed the pseudo-second-order kinetic model, with a high degree of correlation (R2 > 0.96), exhibiting clear time dependence and eventually reaching adsorption equilibrium. These results indicated complex interactions between MPs and PAH (naphthalene) molecules during the adsorption process, where chemical adsorption has a dominant role.

Table 3 Comparison of adsorption kinetic parameters for PS, PP, and PE microplastics.
Full size table

Notably, among the three MPs, PS had the highest k₂ (fastest adsorption rate) and adsorption capacity (i.e., greatest Qe), as shown in Fig. 7. This phenomenon reflects the characteristics of PS, which has abundant active sites and a large specific surface area. The aromatic ring structure of PS therefore provides numerous adsorption sites and enhances the affinity between naphthalene molecules and its surface, thus rendering it the most effective for naphthalene adsorption.

Fig.7
Fig.7
Full size image

Adsorption kinetic curves of microplastics for naphthalene and at 25 °C (pH = 7.0):(a) fluorescence spectrophotometry (FS), (b) PDM.

In contrast, PP and PE exhibit slower adsorption rates and lower adsorption amounts due to their surface structures. PP shows a significantly lower adsorption rate and adsorption amount than PS, indicating that its adsorption process is mainly driven by chemical adsorption. This could be attributed to the relatively smooth surface of PP and its lower surface energy; these features reduce the opportunities for interactions with naphthalene molecules, thereby slowing down the adsorption rate. PE has a slightly higher adsorption capacity than PP. The slightly wrinkled surface structure of PE provides more adsorption sites, which enhance its ability to adsorb naphthalene, although its adsorption rate is still lower than that of PS.

The specific surface areas (SSA) of three types of microplastic particles were measured by BET Surface Area Analyzer (CIQTEK, V-Sorb 2800P). BET measurements revealed SSA of PS, PP, and PE were 0.54, 0.12, and 0.10 m²/g, respectively. Normalizing the adsorption capacity \(q_{{\text{e}}}^{{{\text{norm}}}}\)by SSA according to Eq. 13 yielded values of 3.44, 13.50, and 15.70 mg/m², indicating that the faster adsorption rate of PS cannot be solely attributed to its surface area. Instead, the presence of aromatic phenyl groups in PS likely facilitates \(\pi - \pi\)interactions with naphthalene, providing an additional driving force for adsorption beyond the contribution of surface area alone.

To further validate the broad applicability of the proposed method beyond naphthalene, supplementary adsorption kinetics experiments using phenanthrene were performed on PP, PE, and PS MPs via the PDM technique. The results are shown in Fig. 8. The adsorption kinetics profiles obtained from the PDM method closely corresponded with those measured by the reference method throughout the entire adsorption process, with both approaches converging to comparable equilibrium adsorption capacities. These findings demonstrate that the PDM technique can reliably and accurately characterize the dynamic adsorption behaviors of various PAHs.

Fig.8
Fig.8
Full size image

Adsorption kinetic curves of microplastics for phenanthrene at 25 °C (pH = 7.0): (a) fluorescence spectrophotometry (FS), (b) PDM.

A binary solution was prepared by mixing 25 mL of 2 mg/L naphthalene and 25 mL of 2 mg/L phenanthrene, which was subsequently exposed to microplastic particles for adsorption tests. As further illustrated in Fig. 9, in the mixed solution system, the adsorption capacities of the three types of MPs were all lower than the theoretical additive values obtained from the single-solute systems, indicating the presence of pronounced competitive effects. Meanwhile, adsorption capacities determined by both the PDM method and fluorescence spectrophotometry showed a high degree of consistency. This agreement not only validates the reliability of PDM measurements but also demonstrates its applicability to dynamic monitoring in mixed-contaminant systems, thereby further underscoring the environmental relevance of the present study.

Fig.9
Fig.9
Full size image

Comparative adsorption kinetics of microplastics in binary naphthalene–phenanthrene system at 25 °C (pH = 7.0): (a) fluorescence spectrophotometry, (b) PDM.

The fluorescence method and the PDM method yielded highly consistent results in terms of the adsorption kinetics parameters, thus validating the accuracy and reliability of PDM for monitoring the adsorption processes of MPs. These results also support the use of PDM in microplastic adsorption research, particularly for dynamic monitoring with high temporal resolution, where it exhibits significant advantages.

c. Adsorption isotherms

Before analyzing the adsorption isotherms, the nliquid corresponding to various concentrations of naphthalene was measured. The experiments were conducted in nine groups, each with 20 measurements. To visualize the distribution characteristics of refractive index across different naphthalene concentrations, a kernel density scatter plot is presented in Fig. 10. These data, presented in Table 4, were essential for understanding the concentration-dependent properties of the system, which helped inform the analysis of the adsorption process.

Table 4 Naphthalene concentration and corresponding refractive index values.
Full size table
Fig.10
Fig.10
Full size image

Measurement density and temporal trends of refractive index in adsorption.

Based on the assumptions of the Langmuir model, the adsorption process is considered to involve monolayer adsorption with equal adsorption sites. The experimental data (Table 5) showed that both fluorescence spectrophotometry and PDM provided a good fit to the Langmuir model, with R² >0.97. The slightly higher KL obtained by PDM suggested a more precise description of the adsorption process, particularly in terms of capturing the distribution of adsorption sites. For PP and PE MPs, the results obtained by fluorescence spectrophotometry and PDM were very similar, although PDM slightly overestimated the values.

Table 5 Langmuir and freundlich isotherm fitting parameters.
Full size table

The Freundlich model, which is suitable for describing non-ideal adsorption processes with multi-layer adsorption and heterogeneous adsorption sites, also provided a good fit for all three MPs (Table 5). Both methods indicated that the adsorption processes for all three MPs exhibited heterogeneity. For example, the Kf values and adsorption index (n) values indicated relatively uniform adsorption sites and strong adsorption capacity. For PP, the Kf and the n measured by PDM were higher than those obtained by fluorescence spectroscopy, suggesting that PP has more diverse surface adsorption sites, and PDM is more precise in revealing these details. The results for PE followed a similar trend.

As is shown in Fig. 11a, the adsorption amounts determined by PDM and fluorescence spectroscopy are highly consistent. However, in certain MP samples, PDM yielded slightly higher adsorption amounts, particularly at lower PAH concentrations, for example, at a naphthalene concentration of 1.0 mg/L (Fig. 11b). These results further confirm the higher sensitivity and precision of PDM, especially in capturing small-scale adsorption changes that might be missed by fluorescence spectroscopy at low concentrations.

Fig.11
Fig.11
Full size image

(a)Adsorption isotherms obtained by two different methods, (b)Adsorption quantities at low concentrations.

d. Adsorption thermodynamics

The data from thermodynamic adsorption experiments are presented in Fig. 12. Overall, the trends in terms of thermodynamic parameters were consistent between fluorescence spectroscopy and PDM.

Fig.12
Fig.12
Full size image

Thermodynamic parameters for adsorption: (a) ΔS, (b) ΔH, (c) ΔG.

Both methods indicated that the adsorption process of PS is exothermic (ΔH < 0). As the temperature increases, the ΔG value gradually decreases, indicating that the spontaneity of the adsorption process increases. Meanwhile, the adsorption capacity of PS increases with increasing temperature, suggesting that higher temperatures are favorable for the adsorption of naphthalene by PS MPs. This phenomenon is likely related to the surface characteristics of PS and its aromatic ring backbone structure, i.e., higher temperatures activate more adsorption sites and promote intermolecular interactions.

For PP, the ΔH value is negative, indicating that the adsorption process is exothermic. Similar to PS, the increasingly negative ΔG value with rising temperature suggests that higher temperatures enhance adsorption spontaneity. Although the process is primarily driven by hydrophobic interactions and van der Waals forces, the enhanced spontaneity implies that temperature influences surface interactions and improves adsorption efficiency. The relatively smooth and low-energy surface of PP restricts intermolecular interactions; however, at elevated temperatures, increased molecular motion and improved accessibility of adsorption sites facilitate more effective adsorption.

The thermodynamics of PE adsorption indicate an endothermic process, consistent with its increasing adsorption capacity at higher temperatures. Unlike PS and PP, PE exhibits a positive ΔH (ΔH > 0), confirming its endothermic nature and suggesting that adsorption is thermally activated. This trend is reinforced by the increasingly negative ΔG value, signifying greater spontaneity at elevated temperatures. The irregular surface morphology of PE, characterized by wrinkles, likely provides additional adsorption sites, while the flexibility of its molecular chains further influences the adsorption mechanism. The lower ΔS values suggest a more ordered adsorption configuration, possibly due to solvent-mediated interactions or structural rearrangement upon adsorption. Consequently, PE exhibits a stronger adsorption affinity at higher temperatures, consistent with its endothermic behavior.

e. Discussion

Based on the experimental results, PDM extends the capability of digital holographic microscopy, enabling simultaneous measurement of three-dimensional morphological changes in samples immersed in liquid environments and variations in the concentration of the liquid medium. Although the three-dimensional morphology of microplastic particles remained unchanged during the adsorption of polycyclic aromatic hydrocarbons (PAHs), using PDM, the measured 3D morphology can help to distinguish the type of microplastics. Experiments involving the adsorption of mixed PAH solutions onto microplastics demonstrated that the PDM still performs effectively in measuring concentration changes. Thus, the PDM shows potential for inferring the mechanisms of PAH adsorption by microplastics in their original environment.

Theoretically, different substances possess distinct molecular structures and dielectric constants, resulting in unique refractive indices. This should also apply to composite materials, which are expected to exhibit specific refractive index values. However, directly identifying medium types based on refractive indices obtained via PDM requires further experimental validation.

Conclusions

This report proposes a phase difference method (PDM) based on digital holographic microscopy for simultaneously measuring the 3D morphologies of MP particles and the concentration variations of a PAH solution, thereby providing comprehensive information about the adsorption behavior of PAHs on MPs. Importantly, the physical morphology measurements and chemical concentration measurements are conducted in real-time and in situ. The PDM results were consistent with fluorescence spectroscopy experiments in terms of analyzing the adsorption kinetics, isotherms, and thermodynamics. The 38% higher SNR of the PDM relative to the fluorescence spectrophotometer highlighted the applicability of the PDM for dynamic monitoring with high temporal resolution for characterizing adsorption kinetic analysis. In isotherm analysis, the PDM also had significantly higher sensitivity than fluorescence spectroscopy at low PAH concentrations, particularly in terms of capturing the distribution of adsorption sites. In thermodynamic analysis, the three types of MP particles differed significantly owing to their distinct 3D morphologies, further confirming that simultaneous measurements of physical morphology and chemical concentration are essential for evaluating adsorption processes. Moreover, these results validated the accuracy and reliability of the PDM for monitoring the adsorption behaviors of MPs.

This study demonstrates that PDM offers a novel approach for microplastic adsorption research, owing to its high resolution, sensitivity, and dual observation ability. The significance of PDM lies in the simultaneous, in-situ and real-time measurement of environment and morphology, which not only provides data for the study of the mechanism of microplastic adsorption of polycyclic aromatic hydrocarbons, but enables in-situ measurement data for various theoretical research on such as physiological, pathological or toxicological responses of biological cells caused by environmental changes, etc.