Polymer 255 (2022) 125111

Contents lists available at ScienceDirect

Polymer
journal homepage: www.elsevier.com/locate/polymer

Synthesis of pH-Sensitive polydopamine capsules via pickering emulsions

stabilized by cellulose nanocrystals to study drug release behavior
Mastoure Shirjandi, Vahid Haddadi-Asl *, Elahe Abdollahi, Fatemeh Khanipour
Department of Polymer Engineering and Color Technology, AmirKabir University of Technology, P.O. Box 15875-4413, Tehran, Iran

A R T I C L E I N F O A B S T R A C T

Keywords: Stimuli-responsive carriers capable of recognizing and responding to physiological changes (external stimuli such
Polydopamine nanocapsules as photo, temperature, pH, etc.) play an essential role in targeting drug delivery. In this study, Hollow poly­
Pickering emulsion dopamine (PDA) nanocapsules was developed with a considerably small mean droplet size of 397 nm templated
Cellulose nanocrystal
by Pickering emulsions stabilized by cellulose nanocrystals for drug encapsulation. These PDA drug delivery
pH-responsive system
Sustained release
systems were characterized with scanning electron microscope, dynamic light scattering, and fourier transform
infrared techniques. To study the drug release behavior, the typical chemotherapeutic agent, doxorubicin (dox),
was loaded into PDA capsules, and drug loading efficiency of 76.14% and encapsulation efficiency 73.43% was
obtained. The maximum release percentages of dox after 48 h were 19.01, 39.64, and 80.76% in the solution
with pH 7.4, 6.8, and 5.5, respectively. To achieve controlled drug release in the acidic environment of cancer
tissue, dox should be released in an acidic media and blocked in the neutral media. These PDA capsules can work
as stimuli-sensitive gatekeepers to control the release of drug molecules in response to the pH stimulus. It is
worth noting that this is the first work that was able to prepare and investigate drug loading and release
properties in PDA capsules formed on emulsion droplets stabilized by cellulose nanocrystals with a size of less
than 500 nm.

1. Introduction [11]. The shell materials are versatile, which are usually composed of
biodegradable polymers, such as polyvinyl alcohol (PVA) [12], poly(D,
Cancer is the second leading cause of death worldwide. Conventional L-lactide) (PLA) [13], chitosan [14], alginate [15], and polyurethane
chemotherapy is the most commonly used approach in cancer treatment (PU) [16]. Many mechanisms can initiate changes in a capsule shell wall
based on applying small toxic chemotherapeutic molecules [1]. that results in the release of capsule contents [17], such as pH [18],
Chemotherapy has many disadvantages, including drug toxicity, limited temperature [19], light [20], etc. Also, Grafting of responsive polymers
targeting, poor drug solubility, and low drug stability in a physiological onto different surfaces results in a smart surface which changes its
environment [1,2]. However, anticancer drugs do not distinguish cancer physico-chemical properties in the presence of stimuli [21–24]. Of these
cells from normal cells, causing serious side effects to damage normal stimuli, pH-triggered system is most promising, as pH values vary in
cells. Also, most anti-cancer drugs bind extensively to plasma proteins different tissues and intracellular compartments [11]. Also, carriers
and red blood cells, resulting in decreased efficacy and increased should have good stability against various environments, minimal
toxicity. For these side effects to be minimized, they should be released adsorption on plasma proteins and red blood cells, and appropriate size.
only at the cancer cells without adsorption on plasma proteins and red Furthermore, to be effectively used as anti-cancer drugs, they should
blood cells [3]. To cope with these challenges, drugs were incorporated show pH responsiveness, because cancer cells have a lower pH (<6.8)
in carriers involves capsules [4], nanoparticles [5,6], micelles [7], compared with normal cells [3]. In this respect, numerous pH-sensitive
hydrogels [8], emulsions [9], and dendrimers [10]. In this respect, the polymeric carriers, especially polymer capsules, have been developed
use of capsules in drug delivery carriers has generated considerable in­ for sustained drug delivery [6,11]. Strategies for local and target de­
terest because of its high carrier capacity and controllable drug release livery of drug to cancer cells via trigger pH-controlled drug release may

Abbreviations: PDA, polydopamine; Dox, doxorubicin; CNC, cellulose nanocrystal; DA, Dopamine hydrochloride; EE, encapsulation efficiency; LE, drug loading
efficiency.
* Corresponding author.
E-mail address: [email protected] (V. Haddadi-Asl).

https://doi.org/10.1016/j.polymer.2022.125111
Received 24 April 2022; Received in revised form 19 June 2022; Accepted 25 June 2022
Available online 6 July 2022
0032-3861/© 2022 Elsevier Ltd. All rights reserved.
M. Shirjandi et al. Polymer 255 (2022) 125111

facilitate targeted drug delivery in cancer tissues, result in reducing the Dopamine hydrochloride (DA) was purchased from Sigma Aldrich. So­
toxicity came from side effects of drug and increasing the efficacy [6]. dium hydroxide (NaOH), hydrochloric acid (HCl), and Tris (hydrox­
pH had a remarkable effect on the release rate and cumulative release ymethyl)aminomethane hydrochloride (Tris) were purchased from
amount of drug. The faster release of drug from drug-loaded capsules at Merck. Doxorubicin hydrochloride (Dox) was purchased from Acte. All
lower pH was due to the conversation of hydrophobic tertiary amines chemicals were used without additional purification unless stated
into hydrophilic groups. These results suggested that upon internaliza­ otherwise.
tion by cells, the micelles were capable to retain most drug under normal
physiological conditions and be triggered to release drug under the 2.2. Methods
acidic environment [25].
Inspired by the adhesive versatility of mussels, polydopamine (PDA) 2.2.1. Preparation of CNC
was first investigated by Messersmith et al. and has subsequently
attracted significant attention in the polymer community [26]. PDA 2.2.1.1. Preparation of CNC from waste paper powder. In this paper, CNC
displays many attractive properties, such as good adhesion, biodegrad­ was prepared via acid hydrolysis of waste paper powder (CNC-WP) with
ability, and UV resistance [27]. Furthermore, the cytotoxicity of poly­ a similar method reported by Lei et al. [47]. Preparing CNC was initiated
dopamine in vitro and in vivo is negligible [5,28] which means that by mixing 2 g waste paper powder (dried in a vacuum oven at 50 ◦ C for
polydopamine has excellent biocompatibility features. As a result of 30 min) with H2SO4 (100 ml, 63% w/w) stirred at 45 ◦ C for 1 h. After the
these beneficial properties, polydopamine and its hybrid materials have allotted time, the hydrolysis was quenched by adding 10-fold excess cold
been applied in various areas such as batteries [29], adsorbents [30], water (1000 ml) to the reaction mixture followed by keeping in an ice
anti-corrosion protection [31], tissue engineering [32] and drug de­ bath. The suspension was left undisturbed until CNCs began to sediment
livery [33,34]. (about 1 h). After removing the supernatant, the remaining suspension
PDA can be easily synthesized and forms thin coatings onto a wide was centrifuged for 5 min at 7000 rpm. Once again, the supernatant
range of surfaces via oxidative self-polymerization of its monomer in removed and the solution was washed with water, sodium bicarbonate
alkaline conditions [34]. Templating is a commonly employed approach (2% w/v), and finally with distilled water to reach solution with pH 7
for synthesizing PDA capsules, followed by removing the template core. [48]. Afterward, turbid supernatant was separated and ultrasound 1 h
The template can be hard core material (e.g. Calcium carbonate (CaCO3) with a UP100H (Germany) probe sonicator from Hielscher Ultrasonics to
particles [35], polystyrene (PS) spheres [36], silica (SiO2) particles [31]) obtain CNC suspension. Finally, these suspensions were dried in the
that can form very stable and monodisperse capsules, but this method vacuum oven at 50 ◦ C for 24 h.
relies on etching the hard templates using harsh acid/solvent. To over­
come this disadvantage, soft templating substances have been used by 2.2.1.2. Preparation of CNC from cotton. In order to hydrolyze CNCs by
many researchers (eg. emulsion droplets), which can be removed using sulfuric acid from cotton (CNC–C), a well-studied procedure was
without using harmful solvents [37]. Surfactants are often toxic to employed [45,49]. At first, 87.5 mL of 63% w/w sulfuric acid was
biological tissues, and the removal of surfactants is challenging and pre-heated to 45 ◦ C and then was added to the 5 g cotton and the so­
costly [38]. Pickering emulsions are surfactant-free emulsions stabilized lution was stirred for 1 h by mechanical stirrer. The remainder of the
by amphiphilic solid particles that are more stable, efficient, sustainable, reaction was performed by following the same procedure as described
and biocompatible than classic emulsions [39,40]. Many types of solid for preparing CNC-WP.
particles, such as silica [41], Halloysite nanotubes (HNT) [42], and
cellulose nanocrystals (CNCs) [39,43], have been used to stabilize 2.2.2. Preparation of O/W pickering emulsions stabilized by unmodified
Pickering emulsions. In particular, CNCs have been demonstrated as CNCs
excellent Pickering emulsifiers due to their high aspect ratio, sustain­ The emulsion was prepared according to the same method which was
ability, biocompatibility [21,38]. Cellulose nanocrystals have been hy­ stated by Varanasi et al. [50] with minor modification. Practically, The
drolyzed from various cellulose sources such as microcrystalline CNC suspensions with varying concentrations were prepared by
cellulose [44], cotton [45], bacterial cellulose [40], and waste paper dispersing unmodified CNCs in deionized water via deionized water
[46]. sonication for 10 min. To achieve an electrostatic screening on the
In the present work, Pickering emulsion droplets have been used as highly negatively charged surface of CNCs, the ionic strength of the
soft templates for the direct coating of polydopamine to generate aqueous phase was adjusted by adding 50 mM NaCl [51]. The CNC
submicron-size polydopamine capsules. The drug loading and release suspension and oil phase (toluene or hexadecane) [52,53] were mixed
properties of these carriers were also studied for the first time. CNCs with the volume ratio of 4:1, and the O/W Pickering emulsion was
were synthesized from cellulosic sources by acid hydrolysis and the one prepared by sonicating the mixture using the probe ultrasonic for 20
with the less length was chosen as an emulsifier to prepare oil-in-water min. The obtained Pickering emulsion was kept at room temperature for
Pickering emulsions, which act as a template to produce PDA capsules 24 h and was subsequently used as the stock emulsion.
for drug encapsulation. The advantages of this system include: (1) CNC
is a biocompatible emulsifier for oil-in-water emulsions; (2) the drug 2.2.3. Preparation of PDA capsules
(Dox) is encapsulated in the capsules after DA polymerization. So, it is The capsules were prepared by depositing polydopamine (PDA) on
suitable for loading both hydrophilic and hydrophobic drugs. (3); PDA the Pickering emulsion template by oxidative self-polymerization in
shell can control the release of toxic drug (Dox) in the physiological alkaline conditions. Briefly, the emulsion was diluted at the desired
environment and release it in the acidic environment of tumor tissue. ratio, and Tris was dissolved in the dispersion to adjust the pH to 8.5.
Finally, dopamine was added to the system, and the coating polymeri­
2. Experimental zation was performed for 24 h at room temperature under a stirrer at
500 rpm [54]. At the end of the reaction time, the solution was centri­
2.1. Materials fuged and washed with distilled water to remove excess monomers.
Washing was continued until the supernatant became colorless. To
Waste paper powder and cotton are used as cellulosic sources. Sul­ remove the oil inside the capsules, the isolated product was placed in a
furic acid (H2SO4, 99%) and sodium bicarbonate that were used to vacuum oven for 24 h.
prepare CNC, were purchased from Merck and Dr. Mojalali, respectively.
N-hexadecane and Toluene were purchased from sigma-Aldrich and
Merck, respectively and used as oil phase to prepare an emulsion.

2
M. Shirjandi et al. Polymer 255 (2022) 125111

2.2.4. Loading and release of dox particle size and its distribution. Prior to the test the samples were
The 1, 2, and 4 mg PDA capsules were mixed into Dox buffer solution diluted with distilled water prepared by dispersing the dried powder in
(4, 8, and 10 mL drug buffer solution at concentrations of 40, 80, 100, water and homogenized to yield a uniform dispersion. DLS was con­
and 150 ppm, pH 8.5) with the ultrasonic probe for 30 min. The solu­ ducted on 0.005 wt% CNC dispersion using a Cordouan Tech device
tions were placed in a vacuum oven for 30 min to improve drug uptake (France).
into hollow structure inside of the capsule. Afterward the solutions were The morphology of the nanoparticles was investigated using a field
stirred in the dark environment for 24 h. The product was centrifuged emission scanning electron microscope (FE-SEM) (TESCAN MIRA3,
and then the free Dox was washed away from the supernatant. To Czech Republic). The FE-SEM specimens were prepared by drying a
calculate the Dox loading efficiency (LE) and encapsulation efficiency dispersion drop on a clean aluminum foil in the air. Fourier transform
(EE), the content of the original Dox and the residual Dox in the su­ infrared (FTIR) spectra were obtained using the Bomem (MB 102,
pernatant were determined by UV–Vis absorption spectroscopy using Canada) FTIR spectrophotometer within a range of 400–4000 cm− 1
the intensity of the peak at the wavelength of 480 nm, which is the using a resolution of 4 cm− 1. An average of 32 scans has been reported
characteristic absorption of free Dox. The LE and EE were calculated by for each sample. The cell path length was kept constant during all the
eqs. (1) and (2), respectively. experiments. Samples were prepared on a KBr pellet in vacuum desic­
m0 − m1 cators under a pressure of 0.01 Torr.
LE(%) = × 100 (1) The zeta potential of CNC suspension in deionized water (0.01% w/
mc
w) was determined at 25 ◦ C by a Nano Zetasizer equipment (WALLIS
m0 − m1 Instruments, Cordouan Tech, France). The zeta potentials were deter­
EE(%) = × 100 (2)
m0 mined from 3 measurements consisting of 12 repeated runs per mea­
surement. Contact angle is another important parameter to examine the
where, m0, m1, and mc are the mass of the original drug, the mass of wettability of nanoparticles and predict the type of emulsion. Sample
residual drug in the supernatant, and the mass of the capsule used in was prepared at 3.0 wt%, which were further coated on a glass slide by
loading, respectively. Experiments were carried out in triplicate, and the drop casted to obtain a film. The angle of the resulting drop was
average values were reported. measured using ImageJ software.
In vitro release behavior of Dox was assessed using the sample and To evaluate the stability, the emulsions were diluted up to 20 times
separate (SS) method. The cumulative release of the Dox from the cap­ with water. The emulsions were visualized by an inverted optical mi­
sules was studied in conical tubes containing 5 ml PBS solution at croscope (Leica DMBX). The creaming index (CI%), defined as the ratio
various pH. 1 mg PDA-Dox capsules were dispersed in PBS with pH = of the emulsion volume to the total liquid volume, was quantified after
7.4, 6.8 and 5.5. Then the suspensions were shaken at a constant tem­ 24 h.
perature of 37 ◦ C and 500 rpm. At predetermined time intervals, the The morphology of PDA capsules was characterized by Seron
sample vial was centrifuged at 7000 rpm for 5 min, 4 mL media was AIS2100 scanning electron microscopy (SEM). The capsule diameter
withdrawn, and fresh PBS was added. The released amount of Dox was was measured using an image analysis “ImageJ” software. A total of at
calculated by comparing the absorbance of the drug at 480 nm by least 50 droplets were calculated and the mean diameter was reported as
UV–vis spectroscopy with the previously measured calibration curves the average diameter. UV–vis spectrum was recorded in the UV2601
with a dilution series. Calculation of the corrected concentration of Rayleigh spectrophotometer with a wavelength of 480 nm.
released Dox was based on the eq. (3).

v ∑
t− 1 3. Results and discussion
Cc = Ct + Ct (3)
V 0
3.1. Characterization of synthesized CNCs
where Cc is the corrected concentration at time t, Ct is the apparent
In this work, waste paper powder and cotton were used as cellulose
concentration at time t, v is the volume of sample taken (4 mL), and V is
sources, which are more accessible compared to other cellulosic sources.
the total volume of the release fluid (5 mL). Finally, the cumulative
Cellulose fibers’ structure contains crystalline as well as amorphous
percent of drug released from the samples was plotted against time. All
regions. With using sulfuric acid, the amorphous regions were quickly
the data were recorded within 48 h.
digested. Under employing specific conditions, i.e., 60 min at 45 ◦ C,
The release kinetics of Dox from PDA capsules was analyzed using
most amorphous regions were removed, resulting in producing the cel­
the Noyes–Whitney equation and application of the Fick’s law, Higuchi
lulose nanocrystals in the reaction solution [33,37]. The particle size
and Korsmeyer–Peppas models described by Eqs. (4)–(6):
distributions of CNC samples prepared from two different cellulose
Mt sources were investigated by DLS analysis. As indicated in Fig. 1(a and
= 1 − e− kF t
(4)
M∞ b), the average particle size reported for the CNC-WP was 45.38 nm, but
the spectrum of the CNC–C sample exhibited a major peak at 59.2 and a
Mt minor peak at 438.25 nm with an intensity of 74% and 26%, respec­
= kt0.5 (5)
M∞ tively. Isolation of CNC from waste paper resulted in the dissolution of
cellulose parts, and the nanostructures showed minor sizes. On the other
Mt
= ktn (6) hand, CNC–C showed larger sizes, which is due to the breakdown of
M∞
weaker interactions of cellulose fibers caused by the mechanical forces
where Mt/M∞ is the fraction of Dox released at time t, kF is the first-order during acid hydrolysis. The geometrical dimensions of CNCs prepared
rate constant independent of the drug concentration, k is kinetic rate from various source materials are summarized in Table 1. The CNCs
constant, and n is an index that provides information on the release prepared with the waste paper source in this work were slightly different
mechanism; i.e. Fickian diffusion (n < 0.43), non-Fickian or anomalous from the values reported in previous works [33,37,40,49]. This differ­
diffusion (0.43 < n < 0.85), and case II transport (n > 0.85) [54,55]. ence is probably due to the use of a higher concentration of sulfuric acid
along with utilizing ultrasonic. As a result of the high mechanical energy
supplied by ultrasonication, nanoparticles can disperse better in sus­
2.3. Characterization pension and have more contact with oxygen, resulting in a slight
negative charge on the particle surfaces caused by partial oxidation and
A dynamic light scattering (DLS) instrument was used to analyze reduced particle size.

3
M. Shirjandi et al. Polymer 255 (2022) 125111

Fig. 1. characrization of CNC; (a,b) DLS spectra ((a) CNC-WP and (b) CNC–C); (c) FTIR spectra; (d) FESEM images of cellulose nanocrystals extracted from
wastepaper; (e) measurement of the water contact angle of CNC.

examining the stability of cellulose nanocrystal dispersions in aqueous

Table 1
media [61]. The Zeta Potential measured for CNCs dispersed in water
Geometrical dimensions of CNCs prepared from various source material.
was − 29.07 mV, indicating the formation of stable suspensions and good
Source Rh (nm) Reference electrostatic stability against aggregation caused by the presence of
Cotton 179.3 [57] sulfate groups at the surface of CNCs. This value is in good agreement
Waste paper 56–140 [47] with the reported zeta potential range of CNCs in the previous study
Microcrystalline cellulose 107 [44]
[58]. Also, Contact angle measurement is widely used to evaluate the
Bacterial cellulose nanocrystals 855 [40]
wettability of hydrophilic or hydrophobic materials [59]. The contact
angle value for CNCs was about 49◦ , which confirms the hydrophilic
On the other hand, the yield percentage, -which is defined as the nature of these nanoparticles due to the high density of hydroxyl groups
ratio of the resulting weight to the initial material, of CNC-WP (46.53%) (Fig. 1(e)). Contact angel of 49◦ and hydrophilic nature suggest that the
was higher than CNC–C (15.6%) due to the presence of more cellulose CNC can form an oil-in-water emulsion. The measured contact angle, in
fibers in waste papers than cotton. Danial et al. previously reported this case, is higher than the value reported in the literature [47,54]. This
33–50% yield percentage for CNCs obtained from wood pulp [56]. The can be attributed to the introduction of sulfate groups into the cellulose
purpose of synthesizing cellulose nanocrystals in this study was to sta­ crystals through hydrolysis by sulfuric acid. The sulfate groups intro­
bilize emulsion droplets, which later served as capsule templates. duced to the outer surfaces of cellulose during acid hydrolysis caused
Nanoparticles of smaller sizes and with more uniform size distributions dehydration of cellulose fiber resulting in a reduction of hydrophilicity
result in smaller and more uniform emulsion droplets and, consequently, and an increase in the contact angle.
final capsules; therefore, there is more tendency to use nanoparticles
with smaller sizes. Hence, CNC-WP was selected to stabilize the pick­ 3.2. Pickering emulsion preparation
ering emulsion due to its smaller size, single peak size distribution, and
more efficient reaction. Since the contact angle between 30 and 150◦ lead to the irreversible
FTIR analysis was utilized to identify the structure of the prepared adsorption properties of the Pickering emulsion, in which the particle
nanoparticles (Fig. 1(c)). The broad IR band, due to the OH stretching repulsion energy is several times greater than the thermal energy due to
vibrations, centered at 3392 cm− 1, was slightly shifted to higher Brownian motion [62]. Hence, with a contact angle of 49◦ and the Zp of
wavenumbers with respect to pure cellulose (3370 cm− 1), due to a − 29 mV, produced CNCs can be considered suitable materials for
change in intramolecular hydrogen bonds in the crystalline moieties of emulsion stabilization. For Pickering emulsion formation, the surface
nanocellulose [58]. The C–H stretching at 2912 cm− 1 was observed. The charge plays an important role not only for the colloidal properties of
hydrolysis process leads to a change in the intensity of the peaks in the solid particles but also for the adsorption of solid particles onto the in­
crystalline and amorphous regions. In the FTIR spectra of the CNC, the terfaces between two immiscible liquids. When the solid particles
band at 1450 cm− 1, attributed to CH2 bending vibration, has been possess a high Zp, the particles tend to repel each other instead of
described as a “crystalline absorption band” [58]. Additionally, the peaks adsorbing firmly onto the o/w interfaces. Reducing the Zp to a
at 1150-1160 cm− 1 and at 1033 cm− 1 indicate the presence of cellulose low-charged region, on the other hand, causes aggregation of the solid
crystals and the sulfate functional groups attached to CNC, respectively particles, which further strengthens the particles network in the
[59,60]. continuous phase and improves the stabilization of the emulsion. In
FE-SEM was used to study the morphology of the nanoparticles. As general, the stability of a picking emulsion depends on various param­
illustrated in Fig. 1(d), the crystalline particles were almost homoge­ eters such as particle concentration, surface activity, and ionic strength
neous and indicated rod-like structures of CNCs. Analysis of the FE-SEM [63]. The aqueous phase for all tested emulsions consisted of cellulosic
images using ImageJ software revealed an average length of 40 nm and nanocrystalline particles dispersed in water at the required concentra­
width of 10 nm. The obtained data, were less than the values reported tion without further treatment. Hexadecane or toluene was added af­
from the FE-SEM test in other studies for acid-treated cellulose nano­ terward as an oil phase. The biphasic system was then sonicated,
crystals (100–300 nm) [56]. Subsequently, these particles exhibited resulting in a very stable oil in water (O/W) emulsion. The oil to water
crystalline, elongated, nanometer-sized characteristics. ratio was kept at 1:4 in all cases.
Measurement of the zeta potential (Zp) is an important parameter in To investigate the effect of solvent polarity on the stability of the

4
M. Shirjandi et al. Polymer 255 (2022) 125111

Pickering emulsions, toluene and hexadecane were used to formulate to coalesce together, and a layer of oil formed on the top of the emulsion.
the emulsions containing 3 wt% of CNCs. The creamy O/W emulsion Increasing the particle concentration assisted in increasing the emul­
was formed with both solvents. Fig. 2 (a, b) show the hexadecane and sions’ viscosity and reducing the size of the emulsion droplets [53]. With
toluene in water emulsions stabilized by 3 wt% CNC after 24 h, increasing the concentration of CNC to 2 and 3 wt%, the emulsion vol­
respectively. It appeared that the toluene-in-water emulsion creamed ume increased. It is worth noting that even after a few days, no phase
faster than hexadecane-in-water emulsion at the same nanoparticle separation or instability mechanisms were observed in the emulsions
concentration. The creaming process is due to the density difference stabilized by 2 and 3 wt% CNC. Furthermore, the creaming rate, which is
between the oil and water. To have a better perspective regarding the associated with Stokes’s law, decreased with increasing CNC concen­
stability of the emulsions, the creaming index, which is the percentage tration. This trend has been previously observed in several Pickering
height of the emulsion phase over the total mixture, was visually eval­ emulsions systems [38,53]. Also, increase in the concentration of the
uated. The creaming index (CI%) for hexadecane and toluene in water CNC was led to increase in the CI%. Higher nanoparticle concentrations
emulsion was calculated 100 and 69.7%, respectively. The differences in produced emulsions with larger emulsion volumes and smaller droplet
CI could be related to the polarity of the organic phase. The polarity of sizes, which is consistent with the results of previous studies [38,53,65].
each solvent is presented by its dielectric constant ε and dipole moment In addition, at high concentrations of CNC, the particles tended to form a
μ; i.e., toluene (ε = 2.38; μ = 0.43), and hexadecane (ε = 2.05; μ = 0.06). three-dimensional interconnected structure through hydrogen bonding,
By increasing the polarity of the dispersed phase used in the emulsion, which trapped the emulsified oil droplets in this network. As a result, the
the surface tension of the interface was reduced, which reduces the droplets were prevented from joining together, and the O/W emulsion
separation energy of the particles from the interface of the oil and water; became stable [64]. The volume of the emulsified layer with 3 wt% of
thus the stability was decreased. The polarity of toluene is higher than CNC was higher and the droplet size was smaller than the similar work
that of hexadecane, and as a result, the tendency of the toluene system performed by Varanasi et al. [50]. The smaller size of droplets is due to
become unstable was more. Optical microscope images with 100x the small scale of the cellulose nanoparticles used in this study. Also, the
magnification show that smaller emulsion droplet sizes were formed smaller surfactant formed the smaller and more stable emulsion
when hexadecane was used as the organic phase (Fig. 2 c,d). In contrast, droplets.
toluene droplets with ununiformed size distribution and relatively large The emulsions consisting of 2 and 3 wt% of nanoparticles were quite
size were dispersed in the aqueous phase. This behavior was mainly similar in appearance. In order to compare them microscopically; their
attributed to the more slowly coagulation of the smaller oil droplets than optical microscope images were examined. As shown in Fig. S1 (a,b), the
larger droplets due to the effect of gravity [43,53,64]. Therefore, it can two systems included small droplets and uniform droplet size distribu­
be concluded that hexadecane is a more suitable organic phase than tion and their microscopic images indicated no significant differences.
toluene, and for this reason, hexadecane in water Pickering emulsion Therefore, it can be concluded that 2 wt% of surfactant was enough to
stabilized with CNCs was used in later stages. stabilize the hexadecane in water emulsion. And at this concentration,
Moreover, the effect of particle concentration on the characteristics all hexadecane droplets were effectively emulsified. As the entire surface
of emulsions was detected. In each case, the CI% was calculated and of the oil droplets was coated with CNC, increasing the nanoparticle
reported in Table S1. The emulsion could not be formed when there were concentration beyond 2 wt% would not significantly affect emulsion
no nanoparticles in the system. The addition of CNCs resulted in the stability, droplet size, or size distribution.
formation of emulsions with their irreversible adsorption on the O/W Due to the emulsion’s involvement in the polydopamine capsule’s
interface to the minimum required coverage. Emulsions were formed at preparation, 24 h after emulsification and before adding dopamine to
concentrations of 0.5 and 1 wt% of nanoparticles, but the emulsion the system, the emulsion was examined with an optical microscope at
volume was low. This can be due to the insufficient surfactant concen­ magnifications 100 and 500 (Fig. S1-c,d). It is apparent that droplet size
tration to completely cover the oil droplets, which caused some droplets and size distribution of the system was relatively remained unchanged.
Thereby, 2 wt% loading content of CNC is sufficient to stabilize the
hexadecane in water emulsion, and using more nanoparticles is
unnecessary.
The type of emulsion can be determined by using the drop test. In this
test, one drop of the emulsion was added to both pure water and pure
hexadecane, and dispersibility were observed. The oil in the water
emulsion was immediately dispersed in water, while the water-in-oil
emulsion was dispersed rapidly in the organic phase. As highlighted in
Fig. S2, the emulsion droplet was fully dispersed in water and confirmed
the type of oil-in-water emulsion, which was expected considering the
fact that the contact angle of water with cellulose nanocrystals was less
than 90◦ .
Noteworthy, this study is the first report of the preparation of
emulsion stabilized by unmodified cellulose nanocrystals with a size of
less than 200 nm. Varanasi et al. reported 2.5 μm as the size of hex­
adecane droplets in water emulsion stabilized with 3 wt% of cellulose
nanocrystals [50]. Also, Tang et al. with using 1.5% cinnamoyl
chloride-modified cellulose nanocrystals achieved a droplet size of 3.8
μm [54] and Chen et al. produced 1.5 μm emulsion droplets with un­
modified cellulose [65].

3.3. Preparation of PDA capsules on pickering emulsion droplets

In order to synthesize PDA capsules, the stock emulsion was diluted

at the desired ratio to synthesize PDA capsules, followed by introducing
Fig. 2. Toluene (a,c) and hexadecane in water emulsion stabilized by 3 wt% different amounts of DA and Tris powder to the mixture, according to
CNC (scale bar = 50 μm). Table 2. By reducing the number of droplets, fewer interactions between

5
M. Shirjandi et al. Polymer 255 (2022) 125111

Table 2 and PDA-5 with different dopamine concentrations (based on the values
Amounts of different components in the formation of polydopamine capsules. mentioned in Table 2), were prepared, and the resulting structures were
Formulation Emulsion Deionized water Dopamine Tris compared. According to Fig. 4, in the formulation containing less
code (ml) (ml) (mg) (mg) dopamine, the spheres from dopamine polymerization on the emulsion
PDA-1 0.2 80 960 581.4 droplets’ surface were distributed evenly in the system, and no other
PDA-2 0.2 40 480 290.7 composition was observed. Whereas in the formulation containing more
PDA-3 0.2 20 240 145.4 dopamine, the spheres were joined together by another compound. The
PDA-4 0.4 20 240 145.4 effect of dopamine concentration on the morphology of polydopamine
PDA-5 0.4 20 480 290.7
capsules can be explained as follows: When the dopamine concentration
was low, the spontaneous polymerization was slow, and it could take
droplets would occur, thereby reducing the chance of droplets coa­ place on the surface of oil droplets homogeneously to form a compact
lescing when the dopamine polymerizes [66]. Therefore, the emulsion and thick PDA shell. However, as the concentration of dopamine
was diluted before adding dopamine and Tris. Prior to polymerization, increased, some dopamine could rapidly polymerize to form lots of tiny
the emulsion was white; however, as soon as dopamine was added, the cores that would develop into nanoparticles after further growth pro­
color changed, indicating that it was polymerizing. The PDA capsules cesses in the solution. Therefore, less dopamine would polymerize onto
were obtained by removing the oil drop by vacuum oven. the surface of oil droplets producing a thin PDA shell even under a
In order to investigate the effect of emulsion concentration on higher dopamine concentration. As a result, the capsules are bonded
capsule structure, 4 formulations PDA-1, PDA-2, PDA-3, and PDA-4, together in the environment via PDA nanoparticles. Therefore, PDA-4
were prepared and examined by SEM with a different dilution ratio of was selected as the optimal formulation according to the previous
emulsion and constant dopamine concentration for all samples, ac­ studies.
cording to Table 2. SEM images showed that in the PDA-1 and PDA-2 Further examination using SEM showed that the prepared PDA
formulations, no clear and single structure was observed (Fig. 3 (a,b)). capsules showed a spherical and regular structure tended to agglomerate
Because of the low amount of oil droplets, the polymerization took place due to the adhesion of polydopamine (Fig. 4(a)). These capsules had a
in the aqueous phase before the dopamine reached the droplet surface. smooth surface; While in most capsules prepared on the hard template,
Polydopamine spheres were evident in the formulation PDA-3, which due to the use of harsh solvents to remove the template, the final
considering their size compared to emulsion droplets, the polydopamine structure of the capsule were wrinkled [35,64,67]. Therefore, it is
shell developed on top of the droplets (Fig. 3(c)). Also, for PDA-4, the apparent that the use of a mini-emulsion process lead to the formation of
formation of regular spheres with a smooth surface and relatively uni­ relatively uniform PDA capsules with control of shell thickness and
form size was confirmed (Fig. 3(d)). The mean particle size for PDA-3 particle distribution. As shown in Fig. 4(a), polydopamine capsules with
and PDA-4 capsules were calculated using ImageJ software 542.26 ± a diameter between 300 and 400 nm were uniformly formed with intact
1.87 and 252 ± 0.0012, respectively, showing that smaller capsules shells, which confirmed the successful preparation of hollow polydop­
belong to PDA-4. A reduction in the amount of dispersed phase would amine capsules. The size of these capsules was smaller than the values
lead to a smaller number of droplets (assuming the droplet size was mentioned in similar articles [35,53,68], which indicates the advantage
unaffected), which resulted in a considerable amount of polymer coating of using nanoparticles in the stabilization of emulsions. These nano­
on each droplet. So, the shell thickness was significantly more remark­ particles cause smaller droplets than usual and smaller capsules.
able, and consequently, the size of the final capsule was smaller for the DLS test was used to investigate the average size and size distribution
experiments with fewer oil droplets. Also, the successful removal of of capsules. According to the results of Fig. S3, this test also confirmed
hexadecane oil droplets without additional steps was confirmed. The the relatively uniform distribution and an average size of 397 nm for the
hollow structure with a smooth inner surface (marked with red arrows) prepared capsules.
is clearly shown in these images. Because of the small size of the cap­ FTIR was used to investigate the composition of chemical groups of
sules, the PDA-4 formulation was selected for later stages. the capsules (Fig. 5). The presence of relatively high absorption peaks of
To study the effect of dopamine concentration, 2 formulations PDA-4 the capsules was because of the production of several compounds [64].
The presence of characteristic peaks at wavelengths 3392 and 2925
cm− 1 is related to the hydroxyl group and the symmetric C–H vibration
of CNC. With the introduction of polydopamine on the surface of CNC,
some changes occurred in peaks in the range of 1200–1600 cm− 1. The
vibration at 1517 cm− 1 is related to the N–H bond [69,70] and the ab­
sorption peak at 1633 cm− 1 can be attributed to the aromatic ring [71].
Meanwhile, the characteristic peaks related to bending and phenolic
tensile vibration of C–O–H can be found in 1386 cm− 1 and 1296 cm− 1,
respectively. The characteristic peak of C–O vibration can also be seen at
1037 cm− 1 [68]. Therefore, the presence of polydopamine in the shell of
the capsules was confirmed.

Fig. 3. SEM images of a) PDA-1, b) PDA-2, c) PDA-3 and d) PDA-4 (scale bar =
1 μm). Fig. 4. SEM images of a) PDA-4 and b) PDA-5 (scale bar = 1 μm).

6
M. Shirjandi et al. Polymer 255 (2022) 125111

After the drug loading, a typical absorption peak at 480 nm for Dox
was decreased compared to the initial drug, indicating successful drug
loading in the capsules. The mass of capsules, drug concentration, and
volume of drug solution were changed, and their effect on LE and EE was
investigated. The UV–Vis intensity of Dox was plotted as a function of a
predetermined concentration of Dox. Fig. S4 illustrates that the intensity
value increases linearly with increasing concentration of Dox (x) (R2 =
0.9974).
In capsules, uptake includes adsorption and encapsulation [74]. It
has been widely accepted that the large number of aromatic ring
structures in PDA can lead to the π-π stacking and hydrophobic in­
teractions with Dox and the subsequent multilayer adsorption. The in­
ternal hollow cavities of PDA capsules might enhance these adsorption
interactions. Besides, the contribution of electrostatic attraction cannot
be easily ruled out, as PDA is negatively charged at the loading pH of 8.5
[76].
To evaluate the drug loading capacity of PDA capsules, Dox was
mixed with different masses of PDA capsules (pH = 8.5) for 24 h. LE and
EE were calculated for each formulation and reported in Table 3. In
PDA-Dox1 formulation, the drug solution became almost colorless at the
end of the loading time; But many capsules remained suspended in the
environment after loading; This is probably because the decrease in the
Fig. 5. FTIR spectra of PDA capsules. concentration of the drug in the adsorption medium is leading to reduce
the drug concentration difference within inside and outside of the cap­
The formation of PDA hollow capsules suggests dominant self- sules; as a result, the absorption driving force is lost, and the drug was
polymerization of dopamine at the interfaces in oil-in-water emul­ not loaded in some capsules. Therefore, the maximum loading capacity
sions. It should be rationalized that the surfaces of O/W emulsion of the capsules was not used. In PDA-Dox2, suspended capsules were less
droplets have a pH much higher than that of the bulk water phase by than PDA-Dox1, and no suspended capsules were observed in PDA-
several orders of magnitude and could significantly accelerate oxidation Dox3. According to the values reported in Table 3, it can be said that
and self-polymerization of dopamine. Because the pKa of the amine with decreasing empty capsules and increasing the ratio of drug to the
group of dopamine is 8.9, the high basicity of the surfaces of new alkane capsule, the encapsulation efficiency has been reduced, whereas the
droplets can readily deprotonate the PDA chains formed atop and turn loading efficiency was increased.
them more hydrophobic. Thus, the hydrophobic interaction between Under the same Dox concentration (100 ppm) and blank PDA
deprotonated PDA chains would be taken into account for selective microcapsule amount (1 mg), an increase in drug loading efficiency from
growth of PDA shells on the surfaces of alkane-in-water emulsion 14 to 74% and encapsulation efficiency from 47 to 73% was observed
droplets, as before shown by researchers [72]. with the volume change of Dox solution from 4.0 to 10.0 ml. As the
initial Dox concentration increased from 40 to 100 ppm at a constant
temperature, the LE was increased. For the PDA capsules with a constant
3.4. Drug loading mass, their surface adsorption sites can effectively contact nearly all the
Dox molecules at the lower Dox concentration. However, the available
As has been extensively investigated in other reports [47], PDA adsorption sites on the capsules could attain saturation at the higher Dox
capsules are biocompatible, increasing drug bioavailability and facili­ concentration. Therefore, the encapsulation efficiency exhibited a
tating targeted delivery. In this study, as an initial test of the obtained declining trend with the increase of the initial Dox concentration. Of
PDA capsules for drug delivery application, PDA capsules were loaded note, when the initial Dox concentration was 100 ppm, the encapsula­
with a model drug. The loading and release behavior of such drug de­ tion efficiency could be up to 73% with an adsorption capacity of 741.6
livery systems was investigated in detail. The small molecules loading mg/g, suggesting the PDA capsules are a high-efficiency capsule for drug
mainly depends on the encapsulation, and its loading capacity relies on
the competition of the driving forces of concentration gradient, the Table 3
electrostatic interaction, hydrophobic interaction between Dox and Amounts of different components in drug loading.
polydopamine aromatic structures, and the mass transfer resistance Sample Capsule Drug Drug EE LE
traversing the PDA film [11,34,73–77]. Moreover, large aromatic rings code mass (mg) concentration volume (%) (%)
in PDA make it possible to load chemical drugs or dyes on their surface (ppm) (ml)
via π–π stacking and/or hydrophobic–hydrophobic interactions. PDA PDA- 4 100 4 83.2 7.07
molecules have many amino and phenolic groups; they are negatively Dox1
charged because of deprotonation of the phenolic group at high pH, PDA- 2 100 4 55.7 9.65
while at low pH, they are positively charged due to protonation of the Dox2
PDA- 1 100 4 46.7 14.4
amino group. Therefore, the PDA particles incorporate cationic and Dox3
anionic molecules at high and low pH, respectively [74]. Due to the PDA- 1 100 8 63.5 50.8
significant anti-cancer effect, Dox was selected as a model drug to Dox4
evaluate the application of PDA capsules as a drug carrier. It is positively PDA- 1 100 10 73.43 74.16
Dox5
charged at pH 7.4 (pKa = 8.6) [11]. Dox molecules were diffused and
PDA- 1 150 10 45.8 60.8
encapsulated inside the interior of PDA capsules that own a negative Dox6
phenolic group, resulting in a higher drug payload and loading effi­ PDA- 1 80 10 32.4 23.8
ciency. The encapsulation efficiency (EE) and loading efficiency (LE) Dox7
were calculated by measuring the absorbance of the supernatant at 480 PDA- 1 40 10 36.7 15.6
Dox8
nm by UV–Vis spectra.

7
M. Shirjandi et al. Polymer 255 (2022) 125111

encapsulation.
From Table 3, it can be concluded that the desired drug loading ef­
ficiency (more than 70%) can be achieved at a feeding Dox concentra­
tion of 100 ppm, Dox solution volume of 10 mL, and blank PDA capsules
amount 1 mg. PDA capsule with a high Dox loading of 741.6 mg g− 1
(Dox: PDA, w/w) was used for our followed experiments. High loading
content in these capsules implies that hollow structure has a larger
storage capacity than nonhollow (i.e., nanoparticle and core-shell)
structure owing to the large surface areas with inner and outer sides.
As revealed before, PDA possesses a negative surface charge, and the
electrostatic interaction between the PDA shell and Dox induces a high
loading efficiency for Dox. Besides, the strong π–π stacking also plays a
role in the high loading amount. Fig. 6. Release profile of Dox from PDA capsule at different pH values.

3.5. Drug release

Table 4
Kinetic parameters of Dox released from PDA capsules.
Considering the zwitterionic of PDA films as previously reported
[11], these capsules could provide a sizeable inner cavity for enhanced k R2
drug loading and respond to the intracellular pH gradients for controlled First-order model 0.385 0.978
drug release. In vitro drug release study was performed in PBS solution Higuchi model 16.68 0.77
over 48 h at pH 7.4, 6.8, and 5.5 representing the physiological pH in Korsmeyer–Peppas model 43.77 0.854
normal tissues, extracellular cancer environment, and tumor cells con­
dition. The cumulative release percent of drug from the capsules was
plotted versus time to obtain the drug release profile (Fig. 6). The drug
release process in PDA capsules could be divided into three periods.
During the initial release phase, the diffusion of the outer and inner
surface-associated drug resulted in the rapidly increasing drug content
in PBS. In the next step, water entered the polymer; it is possible that
Dox release can occur by diffusion through pores resulting in a slow drug
release. In the final erosion-controlled release phase, the lower amount
of drug present in the capsules reduced drug release. The drug release
phases agreed with findings for BTZ release in chitosan-ferulic acid
microcapsules [14]. The amount of the Dox released from the PDA
capsule within this 48 h was 19.01 ± 2.08 at pH 7.4 compared with
39.64 ± 6.16 and 80.76% ± 1.09 of the Dox was released from capsules
at pH 6.8 and 5.5 respectively. At pH 5.5, a simulated tumor cells con­
dition, Dox released within 48 h, was over 4-folder faster than it was at
physiological pH. These results further confirm the release of Dox from
this polymeric delivery system is pH-dependent, with the fastest release
at pH 5.5 and the slowest release at pH 7.4, which allow the Dox activity
to be recovered in cancer tissue that has acidic media. At the same time,
it renders Dox release and becomes inactive in normal tissues.
To further investigate the Dox release mechanism, the release pro­
files for the capsules were fitted to the First-order, Higuchi and Kors­
meyer–Peppas models. The fitted parameters and R2 from models are Fig. 7. The model fit of release amounts versus cumulative times.
shown in Table 4 and Fig. 7. In Korsmeyer–Peppas model, the index (n)
was lower than 0.43, indicating that the release of Dox from PDA cap­ pH-dependent, with the fastest release at pH 5.5 and the slowest release
sules could be described by the Fickian diffusion mechanism, where the at pH 7.4. These results showed that this PDA-Dox platform is suitable
Dox release is diffusion controlled. for drug delivery applications and maximizes the anti-cancer effect.

4. Conclusion Declaration of competing interest

To summarize, PDA hollow capsules were used as a carrier for Dox The authors declare that they have no known competing financial
via the assembly of dopamine on the soft Pickering emulsion template. interests or personal relationships that could have appeared to influence
Pickering emulsion stabilized by CNC represents a surfactant-free pro­ the work reported in this paper.
cess to fabricate capsule as CNC is sustainable, biocompatible, and dis­
plays excellent emulsifying capability. Small CNCs with a uniform size Data availability
distribution have been prepared from waste paper, which effectively
stabilized Pickering emulsion. PDA capsules were formed on the surface The data that has been used is confidential.
of hexadecane in water Pickering emulsion droplets that stabilized by 2
wt% CNC. The effect of monomer and emulsion concentration was Appendix A. Supplementary data
investigated in capsule structure, and the optimal formulation was
selected to study the loading and release properties of these drug de­ Supplementary data to this article can be found online at https://doi.
livery systems. The results of SEM showed uniform morphology and org/10.1016/j.polymer.2022.125111.
hollow structure. PDA capsules could load Dox in high loading and
encapsulation efficiency (>70%). Dox release from these capsules was

8
M. Shirjandi et al. Polymer 255 (2022) 125111

References [24] S. Hajebi, A. Abdollahi, H. Roghani-mamaqani, M. Salami-kalajahi, Temperature-

responsive poly(N-isopropylacrylamide) nanogels: the role of hollow cavities and
different shell cross-linking densities on doxorubicin loading and release, Langmuir
[1] N. Avramovic, B. Mandic, A. Savic-Rodojevic, T. Simic, Polymeric nanocarriers of
36 (2020) 2683–2694, https://doi.org/10.1021/acs.langmuir.9b03892.
drug delivery systems in cancer therapy, Pharmaceutics 12 (2020) 298–315,
[25] X. Yang, W. Wu, J. Li, Z. Hu, N. Wang, X. Yu, A facile strategy to construct
https://doi.org/10.3390/pharmaceutics12040298.
fluorescent pH-sensitive drug delivery vehicle, Polymer 197 (2020), 122496,
[2] S.D. koker, R. Hoogenboom, B.D. Geest, Polymeric multilayer capsules for drug
https://doi.org/10.1016/j.polymer.2020.122496.
delivery, Chem. Soc. Rev. 41 (2012) 2867–2884, https://doi.org/10.1039/
[26] H. Lee, S.M. Dellatore, W.M. Miller, P.B. Messersmith, Mussel-inspired surface
C2CS15296G.
chemistry for multifunctional coatings, Science 318 (2007) 426–430, https://doi.
[3] A.K. Mishra, J. Lim, J. Lee, S. Park, Y. Seo, H. Hwang, J.K. Kim, Control drug
org/10.1126/science.1147241.
release behavior by highly stable and pH sensitive poly(N-vinylpyrrolidone)-block-
[27] D. Hauser, D. Septiadi, J. Turner, A. Petri-Fink, B. Rothen-Rutishauser, From
poly 4-vinylpyridine) copolymer micelles, Polymer 213 (2021) 1–8, https://doi.
bioinspired glue to medicine: polydopamine as a biomedical material, Materials 13
org/10.1016/j.polymer.2020.123329.
(2020) 1730, https://doi.org/10.3390/ma13071730.
[4] M. Graziano, D.W. Charig, N. Di Trani, A. Grattoni, G. De Pasquale, Design and
[28] X. Zhou, T. Chang, S. Chen, T. Liu, H. Wang, J.F. Liang, Polydopamine decorated
characterization of AM sensorized capsules for drug delivery devices, Infinite
orlistat loaded nanoparticles with enhanced cytotoxicity against cancer cell lines,
Science (2021) 2–3, https://doi.org/10.18416/AMMM.2021.2109504.
Mol. Pharm. 16 (2019) 2511–2521, https://doi.org/10.1021/acs.
[5] Y. Liu, K. Ai, J. Liu, M. Deng, Y. He, L. Lu, Dopamine-Melanin colloidal
molpharmaceut.9b00116.
nanospheres: an efficient near-infrared photothermal therapeutic agent for in vivo
[29] W. Jiang, X. Yang, J. Zhang, G. Zhang, Polydopamine-based materials applied in Li-
cancer therapy, Adv. Mater. 25 (2013) 1353–1359, https://doi.org/10.1002/
ion batteries: a Review, J. Mater. Sci. 56 (2021) 19359–19382, https://doi.org/
adma.201204683.
10.1007/s10853-021-06536-3.
[6] H. Hemmatpour, V. Haddadi-asl, H. Roghani-mamaqani, Synthesis of pH-sensitive
[30] J. Fu, Q. Xin, X. Wu, Z. Chen, Y. Yan, S. Liu, M. Wang, Q. Xu, Selective adsorption
poly (N, N-dimethylaminoethyl methacrylate)-grafted halloysite nanotubes for
and separation of organic dyes from aqueous solution on Polydopamine
adsorption and controlled release of DPH and DS drugs, Polymer 65 (2015)
microspheres Jianwei, J. Colloid Interface Sci. 461 (2015) 292–304, https://doi.
143–153, https://doi.org/10.1016/j.polymer.2015.03.067.
org/10.1016/j.jcis.2015.09.017.
[7] Z. Zhang, Y. Xue, P. Zhang, W. Zhang, Hollow polymeric capsules from POSS-based
[31] L. Cheng, C. Liu, H. Wu, H. Zhao, F. Mao, L. Wang, A mussel-inspired delivery
block copolymer for photodynamic therapy, Macromolecules 49 (2016)
system for enhancing self-healing property of epoxy coatings, J. Mater. Sci.
8440–8448, https://doi.org/10.1021/acs.macromol.6b02414.
Technol. 80 (2021) 36–49, https://doi.org/10.1016/j.jmst.2020.10.075.
[8] A.I. Rezk, F.O. Obiweluozor, G. Choukrani, C. Hee, C. Sang, Drug release and
[32] J. Yan, R. Wu, S. Liao, M. Jiang, Y. Qian, Applications of Polydopamine-modified
kinetic models of anti-cancer drug (BTZ) from a pH-responsive alginate
scaffolds in the peripheral nerve tissue engineering, Front. Bioeng. Biotechnol. 8
polydopamine hydrogel: towards cancer chemotherapy, Int. J. Biol. Macromol. 141
(2020), 590998, https://doi.org/10.3389/fbioe.2020.590998.
(2019) 388–400, https://doi.org/10.1016/j.ijbiomac.2019.09.013.
[33] W. Zong, Y. Hu, Y. Su, N. Luo, X. Zhang, Q. Li, Polydopamine-coated liposomes as
[9] S. Hiranphinyophat, A. Otaka, Y. Asaumi, S. Fujii, Y. Iwasaki, Particle-stabilized
pH-sensitive anticancer drug carriers, J. Microencapsul. 33 (2016) 257–262,
oil-in-water emulsions as a platform for topical lipophilic drug delivery, Colloids
https://doi.org/10.3109/02652048.2016.1156176.
Surf., B 197 (2021), 111423, https://doi.org/10.1016/j.colsurfb.2020.111423.
[34] Z. Wang, Y. Duan, Y. Duan, Application of Polydopamine in tumor targeted drug
[10] I. Kumar, S. Dhiman, P. Palia, P. Kumar, N. Sharma, Dendrimers: potential drug
delivery system and its drug release behavior, J. Contr. Release 290 (2018) 56–74,
carrier for novel drug delivery system, Asian J. Pharm. 9 (2021) 70–79, https://
https://doi.org/10.1016/j.jconrel.2018.10.009.
doi.org/10.22270/ajprd.v9i2.945.
[35] L. Zhang, J. Shi, Z. Jiang, Y. Jiang, S. Qiao, J. Li, R. Wang, Bioinspired preparation
[11] F. Cheng, J. Zhang, F. Xu, L. Hu, J. Zhu, pH-sensitive Polydopamine nanocapsules
of Polydopamine microcapsule for multienzyme system construction, Green Chem.
for cell imaging and drug delivery based on Folate receptor targeting, J. Biomed.
13 (2011) 300–306, https://doi.org/10.1039/C0GC00432D.
Nanotechnol. 9 (2013) 1155–1163, https://doi.org/10.22270/ajprd.v9i2.945.
[36] J. Cui, C. Ma, Z. Li, L. Wu, W. Wei, M. Chen, B. Peng, Z. Deng, Polydopamine-
[12] L. Russi, C.D. Gaudio, 3D printed multicompartmental capsules for a progressive
functionalized polymer particles as templates for mineralization of hydroxyapatite:
drug release, Ann. 3D Print. Med. 3 (2021), 100026, https://doi.org/10.1016/j.
biomimetic and in vitro bioactivity, RSC Adv. 6 (2016) 6747–6755, https://doi.
stlm.2021.100026.
org/10.1039/C5RA24821C.
[13] S.R. Abulateefeh, M.Y. Alkawareek, A.M. Alkilany, Tunable controlled release drug
[37] J. Cui, Y. Yan, G.K. Such, K. Liang, C.J. Ochs, A. Postma, F. Caruso, Immobilization
delivery system based on mononuclear aqueous core-polymer shell microcapsules,
and intracellular delivery of an anticancer drug using mussel-inspired
Int. J. Pharm. 558 (2019) 291–298, https://doi.org/10.1016/j.
polydopamine capsules, Biomacromolecules 13 (2012) 2225–2228, https://doi.
ijpharm.2019.01.006.
org/10.1021/bm300835r.
[14] C. Li, K. Fang, W. He, K. Li, Y. Jiang, J. Li, Evaluation of chitosan-ferulic acid
[38] Z. Zhang, M. Cheng, M.S. Gabriel, Â.T. Neto, J.D.S. Bernardes, R. Berry, K.C. Tam,
microcapsules for sustained drug delivery: synthesis, characterizations, and release
Polymeric hollow microcapsules (PHM) via cellulose nanocrystal stabilized
kinetics in vitro, J. Mol. Struct. 1227 (2021), 129353, https://doi.org/10.1016/j.
Pickering emulsion polymerization, J. Colloid Interface Sci. 555 (2019) 489–497,
molstruc.2020.129353.
https://doi.org/10.1016/j.jcis.2019.07.107.
[15] S. Joshi, R. Sahu, V.A. Dennis, S.R. Singh, Nanofiller-enhanced soft non-Gelatin
[39] J.O. Zoppe, R.A. Venditti, O.J. Rojas, Pickering emulsions stabilized by cellulose
Alginate capsules for modified drug delivery, Pharmaceuticals 14 (2021) 355,
nanocrystals grafted with thermo-responsive polymer brushes, J. Colloid Interface
https://doi.org/10.3390/ph14040355.
Sci. 369 (2012) 202–209, https://doi.org/10.1016/j.jcis.2011.12.011.
[16] A.Y. Niu, F.J. Stadler, J. Song, S. Chen, Facile fabrication of polyurethane
[40] I. Kalashnikova, B. Cathala, I. Capron, New Pickering emulsions stabilized by
microcapsules carriers for tracing cellular internalization and intracellular pH-
bacterial cellulose nanocrystals, Langmuir 27 (2011) 7471–7479, https://doi.org/
triggered drug release, Colloids Surf. B Biointerfaces 153 (2017) 160–167, https://
10.1021/la200971f.
doi.org/10.1016/j.colsurfb.2017.02.018.
[41] J. Frelichowska, M. Bolzinger, Y. Chevalier, Pickering emulsions with bare silica,
[17] A.P. Esser-kahn, S.A. Odom, N.R. Sottos, S.R. White, S. Moore, Triggered release
Colloids Surf. A Physicochem. Eng. Asp. 343 (2009) 70–74, https://doi.org/
from polymer capsules, Macromolecules 44 (2011) 5539–5553, https://doi.org/
10.1016/j.colsurfa.2009.01.031.
10.1021/ma201014n.
[42] G. Cavallaro, S. Milioto, L. Nigamatzyanova, F. Akhatova, R.F. Fakhrullin,
[18] Q. Zheng, T. Lin, H. Wu, L. Guo, P. Ye, Y. Hao, Q. Guo, J. Jiang, F. Fu, G. Chen,
G. Lazzara, G. Lazzara, Pickering emulsion gels based on halloysite nanotubes and
Mussel-inspired polydopamine coated mesoporous silica nanoparticles as pH-
ionic biopolymers: properties and cleaning action on marble surface, ACS Appl.
sensitive nanocarriers for sustained release, Int. J. Pharm. 463 (2014) 22–26,
Nano Mater. 2 (2019) 3169–3176, https://doi.org/10.1021/acsanm.9b00487.
https://doi.org/10.1016/j.ijpharm.2013.12.045.
[43] C. Tang, S.B. Spinney, Z. Shi, B. Peng, J. Luo, K. Michael, C. Tam, Amphiphilic
[19] D. Xiao, W. Liang, Z. Xie, J. Cheng, Y. Du, J. Zhao, A temperature-responsive
cellulose nanocrystals for enhanced Pickering emulsion stabilization, Langmuir 34
release cellulose-based microcapsule loaded with chlorpyrifos for sustainable pest
(2018) 12897–12905, https://doi.org/10.1021/acs.langmuir.8b02437.
control, J. Hazard Mater. 403 (2021), 123654, https://doi.org/10.1016/j.
[44] E. Zeinali, V. Haddadi-asl, H. Roghani-mamaqani, Nanocrystalline cellulose grafted
jhazmat.2020.123654.
random copolymers of N -isopropylacrylamide and acrylic acid synthesized by
[20] B. Poinard, S. Zhan, Y. Neo, E. Li, L. Yeo, H. Peng, S. Heng, K. Neoh, J. Chen,
RAFT polymerization: effect of different acrylic acid contents on LCST behavior,
Y. Kah, Polydopamine nanoparticles enhances drug release for combined
RSC Adv. 4 (2014) 31428–31442, https://doi.org/10.1039/C4RA05442C.
photodynamic and photothermal therapy, ACS Appl. Mater. Interfaces 10 (2018)
[45] O.M. Vanderfleet, M.S. Reid, J. Bras, L. Heux, J.G. Mohan, K.R.P. Emily, Insight
21125–21136, https://doi.org/10.1021/acsami.8b04799.
into thermal stability of cellulose nanocrystals from new hydrolysis methods with
[21] Z. Abousalman-rezvani, P. Eskandari, H. Roghani-mamaqani, Synthesis of
acid blends, Cellulose 26 (2018) 507–528, https://doi.org/10.1007/s10570-018-
coumarin-containing multi-responsive CNC-grafted and free copolymers with
2175-7.
application in nitrate ion removal from aqueous solutions, Carbohydr. Polym. 225
[46] W. Lei, X. Zhou, C. Fang, Y. Li, Y. Song, C. Wang, Z. Huang, New approach to
(2019), 115247, https://doi.org/10.1016/j.carbpol.2019.115247.
recycle office waste paper : reinforcement for polyurethane with nano cellulose
[22] E. Zeinali, V. Haddadi-Asl, H. Roghani-Mamaqani, Synthesis of dual thermo-and
crystals extracted from waste paper, Waste Manag. 95 (2019) 59–69, https://doi.
pH-sensitive poly (N-isopropylacrylamide-co-acrylic acid)-grafted cellulose
org/10.1016/j.wasman.2019.06.003.
nanocrystals by reversible addition-fragmentation chain transfer polymerization,
[47] W. Lei, C. Fang, X. Zhou, Q. Yin, S. Pan, R. Yang, Cellulose nanocrystals obtained
J. Biomed. Mater. Res. 106 (1) (2018) 231–243, https://doi.org/10.1002/jbm.
from office waste paper and their potential application in PET packing materials,
a.36230, 2018.
Carbohydr. Polym. 181 (2018) 376–385, https://doi.org/10.1016/j.
[23] P. Eskandari, Z. Abousalman-rezvani, Carbon dioxide-switched removal of nitrate
carbpol.2017.10.059.
ions from water by cellulose nanocrystal-grafted and free multi-responsive block
[48] M. Ioelovich, Optimal conditions for isolation of nanocrystalline cellulose particles
copolymers, J. Mol. Liq. 318 (2020), 114301, https://doi.org/10.1016/j.
michael Ioelovich, Nanosci. Nanotechnol. 2 (2012) 9–13, https://doi.org/10.5923/
molliq.2020.114301.
j.nn.20120202.03.

9
M. Shirjandi et al. Polymer 255 (2022) 125111

[49] E.D. Cranston, D.G. Gray, Morphological and optical characterization of [64] Z. Li, H. Wu, M. Yang, D. Xu, J. Chen, H. Feng, Y. Lu, L. Zhang, Y. Yu, W. Kang,
polyelectrolyte multilayers incorporating nanocrystalline cellulose, Stability mechanism of O/W Pickering emulsions stabilized with regenerated
Biomacromolecules 7 (2006) 2522–2530, https://doi.org/10.1021/bm0602886. cellulose, Carbohydr. Polym. 181 (2018) 224–233, https://doi.org/10.1016/j.
[50] S. Varanasi, L. Henzel, L. Mendoza, R. Prathapan, G. Garnier, Pickering emulsions carbpol.2017.10.080.
electrostatically stabilized by cellulose nanocrystals, Front. Chem. 6 (2018) [65] Q. Chen, T. Liu, C. Tang, Tuning the stability and microstructure of fine Pickering
409–418, https://doi.org/10.3389/fchem.2018.00409. emulsions stabilized by cellulose nanocrystals, Ind. Crop. Prod. 141 (2019),
[51] X. Li, J. Li, J. Gong, Y. Kuang, L. Mo, T. Song, Cellulose nanocrystals (CNCs) with 111733, https://doi.org/10.1016/j.indcrop.2019.111733.
different crystalline allomorph for oil in water Pickering emulsions, Carbohydr. [66] Y. Zhai, J.J. Whitten, P.B. Zetterlund, A.M. Granville, Synthesis of hollow
Polym. 183 (2018) 303–310, https://doi.org/10.1016/j.carbpol.2017.12.085. Polydopamine nanoparticles using miniemulsion templating, Polymer 105 (2016)
[52] F. Cherhal, F. Cousin, I. Capron, Structural description of the interface of Pickering 276–283, https://doi.org/10.1016/j.polymer.2016.10.038.
emulsions stabilized by cellulose nanocrystals, Biomacromolecules 17 (2016) [67] J. Sun, C. Su, X. Zhang, J. Li, W. Zhang, N. Zhao, J. Xu, S. Yang, Responsive
496–502, https://doi.org/10.1021/acs.biomac.5b01413. complex capsules prepared with polymerization of Dopamine, Hydrogen-bonding
[53] C. Tang, Y. Chen, J. Luo, M. Yin, L. Zengqian, J. Tang, Z. Zhang, B. Peng, K.C. Tam, assembly, and catechol dismutation, J. Colloid Interface Sci. 513 (2018) 470–479,
Pickering emulsions stabilized by hydrophobically modified nanocellulose https://doi.org/10.1016/j.jcis.2017.10.021.
containing various structural characteristics, Cellulose 26 (2019) 7753–7767, [68] M. Feng, S. Yu, P. Wu, Z. Wang, S. Liu, J. Fu, Rapid, High-efficient and selective
https://doi.org/10.1007/s10570-019-02648-x. removal of cationic dyes from wastewater using hollow Polydopamine
[54] C. Tang, Y. Li, J. Pun, A. Sameh, M. Osman, K.C. Tam, Polydopamine microcapsules: Isotherm, kinetics, thermodynamics and mechanism, Appl. Surf.
microcapsules from cellulose nanocrystal stabilized Pickering emulsions for Sci. 542 (2021), 148633, https://doi.org/10.1016/j.apsusc.2020.148633.
essential oil and pesticide encapsulation, Colloids Surf., A: Physiochem. Eng. [69] S. Liu, Y. Chen, C. Liu, L. Gan, X. Ma, Polydopamine-coated cellulose nanocrystals
Aspects 570 (2019) 403–413, https://doi.org/10.1016/j.colsurfa.2019.03.049. as an active ingredient in Poly (vinyl alcohol) films towards intensifying packaging
[55] A. Tran, K. Shim, T.V. Thi, J. Kook, S. Soo, A. An, S. Lee, Targeted and controlled application potential, Cellulose 26 (2019) 9599–9612, https://doi.org/10.1007/
drug delivery by multifunctional mesoporous silica nanoparticles with internal s10570-019-02749-7.
fluorescent conjugates and external polydopamine and graphene oxide layers, Acta [70] C. Wang, J. Bai, Y. Liu, X. Jia, X. Jiang, Polydopamine coated Selenide
Biomater. 74 (2018) 397–413, https://doi.org/10.1016/j.actbio.2018.05.022. Molybdenum: a new photothermal nanocarrier for highly effective chemo-
[56] W.H. Danial, Z.A. Majid, M. Nazlan, M. Muhid, S. Triwahyono, M.B. Bakar, photothermal synergistic therapy, ACS Biomater. Sci. Eng. 2 (2016) 2011–2017,
Z. Ramli, The reuse of wastepaper for the extraction of cellulose nanocrystals, https://doi.org/10.1021/acsbiomaterials.6b00416.
Carbohydr. Polym. 118 (2015) 165–169, https://doi.org/10.1016/j. [71] H. Wen, H. Zhou, L. Hao, H. Chen, H. Xu, X. Zhou, Enzyme cum pH dual-responsive
carbpol.2014.10.072. controlled release of Avermectin from functional Polydopamine microcapsules,
[57] J. Paulo, S. Morais, M. De Freitas, M. De, M. De Souza, L. Dias, D. Magalhães, Colloids Surf. B Biointerfaces 186 (2020), 110699, https://doi.org/10.1016/j.
A. Ribeiro, Extraction and characterization of nanocellulose structures from raw colsurfb.2019.110699.
cotton linter, Carbohydr. Polym. 91 (2013) 229–235, https://doi.org/10.1016/j. [72] H. Xu, X. Liu, D. Wang, Interfacial basicity-guided formation of Polydopamine
carbpol.2012.08.010. hollow capsules in pristine O/W emulsions− toward understanding of emulsion
[58] L. Bergamonti, M. Potenza, A.H. Poshtiri, A. Lorenzi, A.M. Sanangelantoni, template roles, Chem. Mater. 23 (2011) 5105–5110, https://doi.org/10.1021/
L. Lazzarini, P. Lottici, C. Graiff, Ag-functionalized nanocrystalline cellulose for cm2028417.
paper preservation and strengthening, Carbohydr. Polym. 231 (2020), 115773, [73] B. Yu, D.A. Wang, Q. Ye, F. Zhou, W. Liu, Robust Polydopamine nano/
https://doi.org/10.1016/j.carbpol.2019.115773. microcapsules and their loading and release behavior, Chem. Commun. (2009)
[59] A. Agostinho, D. Meirelles, A. Letícia, R. Costa, R.L. Cunha, The stabilizing effect of 6789–6791, https://doi.org/10.1039/B910679K.
cellulose crystals in O/W emulsions obtained by ultrasound process, Food Res. Int. [74] Q. Liu, B. Yu, W. Ye, F. Zhou, Highly selective uptake and release of charged
128 (2020), 108746, https://doi.org/10.1016/j.foodres.2019.108746. molecules by pH-responsive Polydopamine microcapsules, J. Macromol. Sci. 11
[60] J.N. Putro, S.P. Santoso, F.E. Soetaredjo, S. Ismadji, Y. Ju, Nanocrystalline cellulose (2011) 1227–1234, https://doi.org/10.1002/mabi.201100061.
from waste paper: Adsorbent for azo dyes removal, Environ. Nanotechnol. Monit. [75] Q. Li, J. Yang, Y. Chen, X. Zhou, D. Chen, Y. Li, X. Zhu, Hyaluronic acid
Manag. 12 (2019), 100260, https://doi.org/10.1016/j.enmm.2019.100260. methotrexate conjugates coated magnetic Polydopamine nanoparticles for
[61] M. Shahrousvand, F. Ahmadi, E. Shahrousvand, A. Babaei, High aspect ratio multimodal imaging-guided multistage targeted chemo-photothermal therapy,
Phospho-Calcified rock candy-like cellulose nanowhiskers of wastepaper Mol. Pharm. 15 (2018) 4049–4062, https://doi.org/10.1021/acs.
applicable in osteogenic differentiation of hMSCs, Carbohydr. Polym. 175 (2017) molpharmaceut.8b00473.
293–302, https://doi.org/10.1016/j.carbpol.2017.08.001. [76] Y. Xing, J. Zhang, F. Chen, J. Liu, K. Cai, Mesoporous Polydopamine nanoparticles
[62] L.E. Low, S.P. Siva, Y.K. Ho, E.S. Chan, B.T. Tey, Recent advances of with co-delivery function for overcoming multidrug resistance via synergistic
characterization techniques for the formation, physical properties and stability of chemo-photothermal therapy, Nanoscale 9 (2017) 8781–8790, https://doi.org/
Pickering emulsion, Adv. Colloid Interface Sci. 277 (2020) 102–117, https://doi. 10.1039/C7NR01857F.
org/10.1016/j.cis.2020.102117. [77] F. Chen, Y. Xing, Z. Wang, X. Zheng, J. Zhang, K. Cai, Nanoscale Polydopamine
[63] Y. Zhou, S. Sun, W. Bei, M.R. Zahi, Q. Yuan, H. Liang, Preparation and (PDA) meets Π− Π interactions: an interface-directed coassembly approach for
antimicrobial activity of oregano essential oil Pickering emulsion stabilized by mesoporous nanoparticles, Langmuir 32 (2016) 12119–12128, https://doi.org/
cellulose nanocrystals, Int. J. Biol. Macromol. 112 (2018) 7–13, https://doi.org/ 10.1021/acs.langmuir.6b03294.
10.1016/j.ijbiomac.2018.01.102.

10