Fabrication of nanodiamonds/polyaniline nanocomposite for bilirubin adsorption in hemoperfusion

Abstract

Carbon-based nanomaterials have been explored as effective adsorbents to remove bilirubin in hemoperfusion therapy. However, developing carbon-based absorbents with both high adsorption capacity and good hemocompatibility remains a challenge in clinical applications. In this study, an efficient adsorbent for bilirubin removal was fabricated by grafting polyaniline (PANI) onto nanodiamonds (NDs). The nanocomposite ND-PANI had negligible effect on the hemolytic activity, confirming its excellent blood compatibility. The adsorption results revealed that the ND-PANI had high adsorption capacities (947 mg/g) and rapid adsorption rate toward bilirubin. Moreover, it exhibited efficient bilirubin adsorption in bovine serum albumin (BSA) solution, indicating its potential for practical application. Additionally, the adsorption kinetics and isotherms were systematically analyzed and modeled, thereby offering insights into the possible adsorption mechanism. Our findings suggest that the ND-PANI could be used as an efficient sorbent for the bilirubin removal, offering a promising avenue for blood purification application.

Keywords:

• Nanodiamonds
• bilirubin
• adsorption
• hemoperfusion

1. Introduction

Bilirubin is a toxic metabolite of hemoglobin in human blood. It serves as a crucial indicator of liver function, as it is commonly conjugated with glucuronic acid in the liver and finally excreted from the body [1,2]. In cases of liver failure, the bilirubin cannot be eliminated from blood by liver detoxification. The excess bilirubin will accumulate in brain and other tissues, causing irreversible damage to the nervous system and further leading to hepatic coma and even death [3].

At present, the hemoperfusion is one of the most effective therapies to alleviate the pain of patients with liver failure, which removes the bilirubin in the blood plasma by utilizing adsorbents [4,5]. A variety of promising materials have been developed for bilirubin adsorption, including polymers [6], metal-organic frameworks [7], activated charcoal [8], carbon nanomaterials [9], etc. Among these emerging adsorbents, carbon nanomaterials have attracted significant attention due to their inherent advantages, such as good chemical stability, high surface activity, and ease of functionalization [10]. However, limited adsorption capability and poor hemocompatibility continue to restrict their clinical application in blood purification procedures.

Nanodiamonds (NDs), a relatively new member of carbon material, have found applications in the fields of machining, environmental engineering, and biomedicine due to the excellent mechanical, optical and physico-chemical properties [11–13]. There are a number of oxygen-containing active groups at the outer surface of NDs, including hydroxyl group, ester group, carboxyl group, providing a platform for precise control over the surface state of NDs through chemical modification [14,15]. In light of their high surface area and abundant functional groups, NDs were recently explored as efficient adsorbent to remove organic dyes and heavy metal ions from wastewater bodies [16–18]. Importantly, it shows lower cytotoxicity compared to other analogous carbon-based materials (nanotubes, graphene oxide, etc.) [19–21], thereby highlighting their potential applicability in biomedical devices. However, to the best of our knowledge, the NDs-based adsorbent for bilirubin removal has never been reported in the previous literature.

Bilirubin is recognized as a lipophilic pigment with a tetrapyrrole dicarboxylic acid, and it can strongly bind to albumin in the bloodstream [22,23]. Considering the structural characteristics of bilirubin, surface modification of the NDs adsorbent is urgently required to enhance the adsorption affinity toward bilirubin.

Herein, a composite of ND and polyaniline (ND-PANI) for bilirubin adsorption was fabricated by in situ polymerization of aniline monomer on the surface of NDs (Figure 1). The structural and surface chemical properties of the ND-PANI were investigated by transmission electron microscopy (TEM), Fourier-transform infrared spectroscopy (FT-IR), zeta potential, and other techniques, revealing the interactions between the adsorbent and bilirubin molecules. Additionally, the blood compatibility of the ND-PANI composite was assessed through hemolytic assay analysis. The results reveal that ND-PANI exhibited remarkable adsorption capacity and rapid adsorption for bilirubin. Furthermore, the possible adsorption mechanism was proposed based on the adsorption kinetics and isotherms studies.

Figure 1. Schematic illustration of the preparation process of the ND-PANI nanocomplex.

2. Experimental section

2.1. Materials

The NDs (3–5 nm) were purchased from Henan Yuxing New Material Co. Ltd. Sulfuric acid (H₂SO₄, AR), nitric acid (HNO₃, AR), ethylenediamine (99%), hydrochloric acid (HCl, AR), aniline (99.5%), ammonium peroxydisulfate (APS, 99%), red blood cells (RBC, 6%) were purchased from Macklin Reagent Co. Ltd. (Shanghai, China).

2.2. Amination of NDs

The carboxylated NDs (ND-COOH) were firstly prepared according to the procedure reported previously [24]. First, 0.2 g pristine NDs were annealed at 425°C for 5 h. The annealed NDs were added in 60 mL strong acid solution (H₂SO₄:HNO₃ = 3:1, vol/vol), then stirred at 80°C for 24 h. The acid-treated NDs were washed with deionized water for three times and then dried in air to obtain the ND-COOH. 200 mg of ND-COOH was dispersed in 40 mL of ethylenediamine solution. The mixture was refluxed at 80°C for 24 h, followed by centrifugation and washing with deionized water. The products were dried under vacuum and labeled as ND-NH₂ (Figure 1).

2.3. Preparation of ND-PANI

30 mg ND-NH₂ were mixed with 10 mL of 0.1 M HCl and stirred for 1 h. Then, 68 mg of APS were added, being stirred for another 1 h. 14 mg of aniline was dispersed in 3 mL of 1 M HCl, and then added dropwise into the ND-NH₂/APS solution under stirring. The mixture as subjected to constant stirring for 24 h in the ice bath. The ND-PANI was obtained by lyophilization after washed with deionized water (Figure 1).

2.4. Hemolytic assay of ND-PANI

The hemolysis (see Glossary) assay was conducted according to the methods previously reported [7]. 20 μL of red blood cells (RBCs) suspension was mixed with 980 μL of ND-PANI/PBS with different concentrations (50, 100, 200, 400, 800, and 1600 ug/mL). Phosphate buffered saline (PBS, pH = 7.4) was used as a negative control, and deionized water was used as a positive control. After incubation at 37°C for 180 min, the mixtures were centrifuged at 2000 rpm for 10 min. The absorbance of the supernatant was measured at 540 nm (hemoglobin) by using the UV–Vis spectrometer, and the hemolysis ratio was calculated using Equation (1)(1) $$\begin{array}{c}\text{Hemolysis ratio}\hspace{0.17em}\text{(\%) =}\frac{A-A_n}{A_p-A_n}\hspace{0.17em}\hspace{0.17em}\times \hspace{0.17em}\hspace{0.17em}100\text{\%}\end{array}$$(1) : (1) $$\begin{array}{c}\text{Hemolysis ratio}\hspace{0.17em}\text{(\%) =}\frac{A-A_n}{A_p-A_n}\hspace{0.17em}\hspace{0.17em}\times \hspace{0.17em}\hspace{0.17em}100\text{\%}\end{array}$$(1) where A, A_n, and A_p are the absorbance at 540 nm of the ND-PANI/PBS, negative control, and positive control, respectively.

2.5. Adsorption kinetics and isotherms of ND-PANI

To evaluate the adsorption kinetics of ND-PANI, 1 mg of ND-PANI was accurately added into 10 mL of bilirubin/PBS solution with different concentrations. The samples were shaken (180 rpm) at 37°C for different times (10, 20, 40, 80, and 120 min). Then, the samples were centrifuged, and the absorbance of the supernatant was measured at 435 nm using an UV–Vis spectrophotometer. All experiments were conducted in dark and repeated for three times.

The amount of bilirubin adsorbed by ND-PANI (q_t) at given time (t) was calculated using Equation (2)(2) $$\begin{array}{c}{q}_t\text{=}\frac{\left(C_0-C_t\right)V}{m}\end{array}$$(2) : (2) $$\begin{array}{c}{q}_t\text{=}\frac{\left(C_0-C_t\right)V}{m}\end{array}$$(2) where C₀ and C_t are the bilirubin concentrations (mg/mL) in the initial solution and at time t, respectively; V is the volume of the bilirubin solution (mL), and m is the mass of ND-PANI (g).

The experimental data were fitted by first-order and second-order models as following: (3) $$\begin{array}{c}\hspace{0.17em}\mathit{\text{ln}}\hspace{0.17em}\left(q_e‐q_t\right)\text{=}\hspace{0.17em}\mathit{\text{ln}}\hspace{0.17em}{q}_{\text{e,cal,1}}‐k_{1}t\end{array}$$(3) (4) $$\begin{array}{c}\frac{t}{q_t}\text{=}\frac{1}{k_2{q}_{\text{e,cal,}\hspace{0.17em}2}^2}+\frac{t}{q_{\text{e,cal,2}}}\end{array}$$(4) where q_e and q_t are adsorbed bilirubin at equilibrium and given times t (mg/g), q_{e, cal,1} and q_{e, cal,2} are the calculated adsorption capacities according to the first-order and second-order kinetic model, k₁ and k₂ are the rate constants of the pseudo-first and pseudo-second order equation, respectively.

For the isotherms study, 1 mg of ND-PANI was added 10 mL of bilirubin/PBS solution with different concentrations of 0.02, 0.04, 0.06, 0.08, 0.10 mg/mL, then shaken at 37°C (180 rpm) for 40 min to reach adsorption equilibrium. The absorbance of the supernatant was measured to determine the concentration of bilirubin.

The adsorption capacity of ND-PANI at equilibrium (q_e) was calculated using Equation (5)(5) $$\begin{array}{c}{q}_e\text{=}\frac{\left(C_0-C_e\right)V}{m}\end{array}$$(5) : (5) $$\begin{array}{c}{q}_e\text{=}\frac{\left(C_0-C_e\right)V}{m}\end{array}$$(5) where C₀ and C_e are the bilirubin concentrations of in the initial solution and at equilibrium (mg/L), respectively. V is the solution volume (L), and m is the mass of ND-PANI (g).

The Langmuir, Freundlich and Liu adsorption models were employed to fit experimental data, the expression of which are Equations (6)(6) $$\begin{array}{c}{q}_e\text{=}\frac{q_m{K}_L{C}_e}{\text{1+}{K}_L{C}_e}\end{array}$$(6) –(8)(8) $$\begin{array}{c}{q}_e\text{=}\frac{q_m{\left(K_{\text{Liu}}{C}_e\right)}^n}{\text{1+}{\left(K_{\text{Liu}}{C}_e\right)}^n}\end{array}$$(8) : (6) $$\begin{array}{c}{q}_e\text{=}\frac{q_m{K}_L{C}_e}{\text{1+}{K}_L{C}_e}\end{array}$$(6) (7) $$\begin{array}{c}{q}_e\text{=}{K}_F{C}_e^n\end{array}$$(7) (8) $$\begin{array}{c}{q}_e\text{=}\frac{q_m{\left(K_{\text{Liu}}{C}_e\right)}^n}{\text{1+}{\left(K_{\text{Liu}}{C}_e\right)}^n}\end{array}$$(8) where q_e is the amount of adsorbed bilirubin at equilibrium (mg/g); q_m is maximum amount of bilirubin adsorbed on ND-PANI; K_L, K_F, and K_Liu are the model constants related to adsorption affinity and/or adsorption capacity; n is the exponential term related to the heterogeneity of the adsorption sites.

2.6. Adsorption in albumin solutions

Bovine serum albumin (BSA) was used to mimic the real blood environment. To evaluate the adsorption of bilirubin in BSA solutions, 10 mg of ND-PANI was added into bilirubin/BSA solution with different concentration of 0.02, 0.04, 0.06, 0.08 mg/mL (pH = 7.4). The concentration of BSA is set to 40 mg/mL, which is close to the concentration of the blood or plasma environment in hemoperfusion settings. The solutions were shaken (180 rpm) at 37°C for various times (40, 80, 120, and 240 min). The absorbance of the supernatant was measured to determine the concentration of bilirubin.

2.7. Characterization

TEM was performed on a JEOL-2010 microscope (JEOL, Japan) operating at 200 kV. Thermal gravimetric (TG) analysis was performed using a NETZSCH STA 449F5 instrument (NETZSCH Group, Germany). X-ray diffraction (XRD) patterns of the samples were recorded in a Rigaku D/MAX-3B diffractometer (Cu Kα1, 0.154 nm). The absorbance spectra were recorded at a UH4150 UV − Vis spectrophotometer (Hitachi). FT-IR spectroscopy was recorded by a Thermo Scientific Nicolet iZ 10 spectrometer using the KBr tablet method. The dynamic light scattering (DLS) and zeta potential were measured on Zetasizer Nano ZS90 (Malvern).

3. Results and discussion

3.1. Morphology and structure

The morphology and structure of the ND-NH₂ and ND-PANI were characterized by TEM. As Figure 2(a) shows, the ND-COOH display a spheroidal shape with diameters ranging from 5 to 10 nm. Upon modification with PANI, the ND-PANI particles agglomerated into a sheet-like structure, with a layer of amorphous shell covered on the surface (Figure 2(b)). Furthermore, distinct lattice fringes with a spacing of 0.208 nm are observed in the HRTEM (Figure 2(c)), which can be attributed to the (111) facet of diamond.

Figure 2. TEM image of (a) ND-NH₂ and (b) ND-PANI. (c) HRTEM image of ND-PANI. (d) TG curves of ND-COOH, ND-NH₂, and ND-PANI.

TG analysis was performed up to 1000°C for ND-COOH, ND-NH₂, and ND-PANI (Figure 2(d)). The TG curves revealed a weight loss occurring around 100°C, which were related to vaporization of the adsorbed water molecules. In the temperature range of 500–800°C, the weight loss was mainly due to the thermal decomposition of surface groups such as–COOH,–NH₂, and–CH–. Notably, an additional weight loss of 3.0 wt% was observed for ND-PANI between 300 and 800°C compared to ND-NH₂. This increase in weight loss can be attributed to the thermal decomposition of polyaniline in the composite.

XRD spectrum was performed to reveal the structure of ND-PANI. As shown in Figure 3(a), two distinct peaks at 43° and 75° were assigned to (111) and (220) planes of NDs, respectively. A broad band at lower angles was also observed, indicative of the presence of an amorphous structure in the polyaniline component. The surface functionalization and chemical structure were characterized using FT-IR spectra. Figure 3(b) presents the FT-IR spectra of the ND-NH₂, conventionally prepared PANI, and ND-PANI. In the spectra of the ND-NH₂, the peaks at 1695 and 3540 cm⁻¹ was ascribed to C = O and–NH₂ stretching mode of the amide group (–CONH₂). The spectra of the PANI showed characteristic peaks at about 1590 and 1370 cm⁻¹ coming from C = N stretching mode of the quinonoid rings and C–N stretching mode. For the ND-PANI specimen, the peaks corresponding to C = O (1695 cm⁻¹), C–N (1370 cm⁻¹), and C = N (1590 cm⁻¹) stretching mode were observed simultaneously. The broad bands in the range of 3000–3400 cm⁻¹ were ascribed to the amine (N–H) stretching vibrations. These findings indicate the NDs have been successfully modified with polyaniline.

Figure 3. (a) XRD patterns of ND-PANI. (b) FT-IR spectra of ND-NH₂, conventional PANI, and ND-PANI. (c) Hydrodynamic diameter distribution and (d) zeta potential of ND-COOH, ND-NH₂, and ND-PANI.

In addition, DLS analysis was conducted to examine the size distribution and colloidal stability of as-synthesized nanoparticles. The average particle sizes of ND-COOH and ND-NH₂ in aqueous solution were determined to be 60 nm and 80 nm, respectively. After linking with polyaniline, the mean size of ND-PANI increased to about 500 nm (Figure 3(c)). It should be noted that these sizes appear larger compared to the TEM results obtained from TEM, which is probably due to the aggregation of nanoparticles. The surface charge of an adsorbent plays a vital role in influencing its adsorption performance. Thus, the zeta potential of ND-COOH, ND-NH₂, ND-PANI was measured to assess the surface charge variation during the modification process. As illustrated in Figure 3(d), ND-COOH exhibited a strong negative potential of–38.57 mV due to the deprotonation of–COOH. By contrast, ND-NH₂ displayed a potential of 16.07 mV, resulting from the protonation from–NH₂ to–NH₃⁺.The zeta potential of ND-PANI slightly decreased to 9.38 mV. Notably, bilirubin molecules are negatively charged under physiological conditions (pH 7.4) due to the existence of two carboxyl groups [25]. Thus, it can be inferred that ND-PANI may have a favorable bilirubin adsorption capacity as a result of electrostatic interaction.

3.2. Hemolysis assay of ND-PANI

Hemocompatibility is recognized as an essential factor for the performance of blood-contacting devices. Carbon nanomaterials (carbon nanotubes, graphene) usually exhibit low blood compatibility and might have unfavorable effects on RBC membranes. This can trigger the release of hemoglobin and further lead to blood coagulation and thrombus formation. Thereby, an in vitro hemolysis assay was carried out to evaluate the hemocompatibility of ND-PANI. PBS and deionized water were used as the positive and negative control, respectively. As shown in Figure 4(a), the hemolysis ratios of ND-PANI with a concentration range of 50–1600 ug/mL were below 3%, which is lower than the safety level of 5% for clinical application. Notably, no hemolysis effect of ND-PANI was detected even at a high ND-PANI dosage of 1600 μg/mL, surpassing the concentration employed in the bilirubin adsorption test. These results indicate that the ND-PANI exhibits a negligible risk of hemolysis for hemoperfusion application.

Figure 4. (a) Hemolysis ratio and hemolysis photographs of ND-PANI with different concentrations. (b) Absorption spectra of ND-PANI, bilirubin, and ND-PANI after bilirubin adsorption (ND-bilirubin). (c) changes in absorbance of bilirubin with different concentrations and (d) the corresponding fitted line plot.

Figure 4(b) shows the absorption spectrum of ND-PANI, bilirubin, and ND-PANI after bilirubin adsorption (ND-bilirubin). The characteristic absor­ption peak of bilirubin at 435 nm was clearly observed in the ND-bilirubin specimen, suggesting the bilirubin molecules were adsorbed on the surface of ND-PANI. Moreover, the intensity of absorption peak of 435 nm gradually increased with a rise in bilirubin concentration (Figure 4(c)). The absorbance exhibited a good linear correlation with the bilirubin concentration (Figure 4(d)), thus enabling further investigation into the bilirubin adsorption process by ND-PANI.

3.3 Adsorption kinetics and isotherms of ND-PANI

The time-dependent adsorption of bilirubin by ND-PANI in PBS solution was firstly carried out with different bilirubin concentrations. As shown in Figure 5(a), adsorption of bilirubin was quite rapid in the first 20 min, and the adsorption equilibrium was reached within 40 min. The removal efficiency was more than 95% (Figure 5(b)), and the adsorption capacities for bilirubin as high as 947 mg/g. These results indicate that the bilirubin could be removed effectively by ND-PANI.

Figure 5. (a) The time-dependent adsorption of bilirubin in PBS solution. (b) Bilirubin removal efficiency with different initial concentrations. (c) The adsorption capacity of bilirubin bovine in BSA solution (40 mg/L).

There is a certain amount of albumin in blood or plasma, which is considered to be a natural carrier of bilirubin through conjugation. To mimic the removal of bilirubin in hemoperfusion, adsorption performance of ND-PANI toward bilirubin was investigated in BSA solution (40 mg/L). As shown in Figure 5(c), the adsorption capacity of bilirubin by ND-PANI decreased in the presence of BSA, since the albumin was competing with ND-PANI for bilirubin adsorption. Nevertheless, the ND-PANI still showed a high adsorption capa­city of 107–225 mg/g at 40 min for different bilirubin concentrations (0.02, 0.04, 0.06, and 0.08 mg/mL). With the adsorption time prolonging to 240 min, adsorption capacity increased to 139–259 mg/g. Even with the competition from BSA, the ND-PANI still demonstrates considerable effectiveness in removing bilirubin compared to other adsorption materials (Table S1).

To gain insights into the adsorption process, pseudo-first-order and pseudo-second-order adsorption reaction models were applied to fit the kinetic experimental data [26,27]. Linear fitting results were depicted in Figure 6(a,b), and the values of k and q_e, along with the correlation coefficient (R²) values, were calculated and listed in Table 1 (C₀ = 0.1 mg/mL). The value of R² for the pseudo-second-order model (R² = 0.9989) was higher than that from the pseudo-first-order model (R² = 0.9830). Moreover, the calculated adsorption capacity (q_{e, cal} = 1282.7 mg/g) using pseudo-second-order model was closer to the experimental value (q_{e, exp} = 947.3 mg/g). Therefore, the pseudo-second-order model better described the adsorption process, indicating that chemisorption played a dominating role in the adsorption of bilirubin on ND-PANI [28].

Figure 6. Kinetic studies of bilirubin adsorption on ND-PANI and the fitted curves by (a) pseudo-first and (b) pseudo-second models. (c) Bilirubin adsorption isotherms by ND-PANI and the curve fitted by Langmuir (dashed line), Freundlich (dotted line), and Liu (solid line) models.

Table 1. Adsorption kinetic parameters of bilirubin by ND-PANI (C₀= 0.1 mg/mL).

Samples	qe (mg/g)	Pseudo-first-order			Pseudo-second-order
		qe,cal,1 (mg/g)	k1 (1/min)	R^2	qe, cal,2 (mg/g)	k2 (g/mg⋅min)	R^2
ND-PANI	947.3	256.8	0.048	0.9830	1282.7	1.289	0.9989

An adsorption isotherm describes the relationship between the adsorbate uptake (q_e) of the adsorbent and the equilibrium concentration (C_e) in the solution, which is important to understand the adsorption mechanism and interaction between adsorbent and adsorbate. The experimental data were fitted by Langmuir, Freundlich, and Liu isotherm models, as shown in Figure 6(c). The maximum adsorption capacity (q_m) and the corresponding adsorption constant (K, n, and R²) were calculated for each model (Table 2). The Langmuir and Freundlich isotherm models were established based on the assumptions of monolayer adsorption and infinite adsorption capacity, respectively. In contrast, the Liu isotherm model is a combination of the Langmuir and Freundlich isotherm models, which predicts that the adsorbent active sites cannot have the same energies and that adsorbate molecules tend to occupy specific sites [29–31].

Table 2. Parameters of Langmuir, Freundlich and Liu adsorption isotherm models for bilirubin adsorption on ND-PANI.

Liu				Langmuir			Freundlich
qm (mg/g)	KLiu (L/mg)	n	R^2	qm (mg/g)	KL (L/mg)	R^2	KF (mg/g)(L/mg)^1/n	n	R^2
1293.9	0.299	1.491	0.9900	1908.9	0.145	0.9755	231.34	0.789	0.9820

Considering that in situ polymerization of aniline on the surface of NDs could be uneven, the active sites of ND-PANI may have varying binding energies, thereby indicating a heterogeneous adsorption of bilirubin. As evidenced by the higher R² values, the Liu isotherm model presented the better fit than either Langmuir or Freundlich model. Furthermore, the maximum bilirubin adsorption capacity was calculated to be 1293.9 mg/g based on the Liu model, which is closer to the experimental result.

4. Conclusions

In summary, a bilirubin adsorbent (ND-PANI) based on NDs was proposed by grafting polyaniline onto NDs to enhance the adsorption performance. The FT-IR spectra and zeta potential analysis revealed that the electrostatic interaction between amine groups of ND-PANI and carboxyl groups of bilirubin played a crucial role in facilitating bilirubin adsorption. Furthermore, a hemolytic assay demonstrated the high hemocompatibility of the synthesized ND-PANI. The batch experiments show rapid adsorption rate with high adsorption capacities in aqueous solution (947 mg/g) and albumin solutions (225 mg/g). The adsorption kinetics and adsorption isotherms studies were conducted, which suggested that the adsorption process of bilirubin on ND-PANI occurred in a heterogeneous manner, with chemical adsorption primarily governing the adsorption process. These findings provided a new strategy to fabricate bilirubin adsorbent for hemoperfusion, highlighting the potential application of NDs in clinical blood purification therapy.

Author contributions

Futao Wang: Conceptualization, Methodology, Validation, Data curation, Writing—Original Draft; Xiangyun Zheng: Data curation, Methodology, Writing—Original Draft, Visualization; Qi Zhao: Methodology, Formal analysis, Investigation, Writing—Reviewing and Editing; Yuchen Feng: Software, Validation; Guanyue Gao: Formal analysis, Visualization; Dalibor M. Stanković: Formal analysis, Writing—Review & Editing; Jinfang Zhi: Conceptualization, Writing-Reviewing and Editing, Super­vision, Funding acquisition.

GLOSSARY

APS	=	ammonium peroxydisulfate
BSA	=	bovine serum albumin used as protein supplement in biochemical experiments
Hemolysis	=	rupture or destruction of red blood cells, leading to the release of hemoglobin into the surrounding fluid or medium
NDs	=	nanodiamonds
ND-COOH	=	carboxylated NDs
PBS	=	phosphate buffered saline
PANI	=	polyaniline
RBCs	=	red blood cells

Disclosure statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Additional information

Funding

The authors thank the financial support from National Natural Science Foundation of China (Nos. 21874143, 22375210, and U21A2070).