Sol–gel microencapsulation of oil phase with Pickering and nonionic surfactant based emulsions

    Sol-gel microencapsulation of oil phase with pickering and nonionic surfactant based emulsions Chlo´e Butstraen, Fabien Sala¨un, Eric Devaux PII: DOI: Reference:

S0032-5910(15)00513-6 doi: 10.1016/j.powtec.2015.06.055 PTEC 11098

To appear in:

Powder Technology

Received date: Revised date: Accepted date:

9 May 2015 23 June 2015 27 June 2015

Please cite this article as: Chlo´e Butstraen, Fabien Sala¨ un, Eric Devaux, Sol-gel microencapsulation of oil phase with pickering and nonionic surfactant based emulsions, Powder Technology (2015), doi: 10.1016/j.powtec.2015.06.055

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Sol-gel microencapsulation of oil phase with pickering and

T

nonionic surfactant based emulsions

IP

Chloé Butstraen a,b,*, Fabien Salaün a,b, Eric Devaux a,b, Université Lille Nord de France, 1 rue Lefèvre, F-59000 Lille, France ;

b

ENSAIT, GEMTEX, 2 allée Louise et Victor Champier, BP 30329, F-59056 Roubaix, France ;

SC R

a

NU

Mails : [email protected], [email protected], [email protected]

* Corresponding author: Chloé Butstraen, ENSAIT-GEMTEX, 2 allée Louise et Victor Champier, BP

MA

30329, 59056 Roubaix, France. Phone : +33 3 20 25 86 85. Fax : +33 3 20 27 25 97. E-mail:

AC

CE P

TE

D

[email protected]

1

ACCEPTED MANUSCRIPT Abstract Sol-gel polycondensation was used to encapsulate two different kinds of core with a silica

T

shell, i.e. castor oil which is considered as a model active agent, and bisphenol A bis(diphenyl

IP

phosphate), an insoluble liquid fire retardant. The influence of the nature of the emulsifier was

SC R

also studied, i.e. pickering emulsion based on the interface stabilization performed by the organization of solid nanoparticles was compared to a classical emulsion process using a nonionic surfactant. The influence of both cores and emulsifiers on the stability of emulsion was

NU

studied by granulometric analysis, optical microscopy and macroscopic morphology (from

MA

nacked-eye observations). The sol-gel encapsulation efficiency was assessed by Fouriertransform infrared spectroscopy (FT-IR) and scanning electron microscopy (SEM) analysis.

D

Finally, thermal stability of microcapsules was evaluated by thermogravimetric analysis

TE

(TGA). Results show that both pickering and classical emulsions processing allow successful sol-gel encapsulation of castor oil and bisphenol A bis (diphenyl phosphate) with a satisfying

CE P

thermal stability for textile application. However, the use of pickering emulsion with

AC

nanoparticles provides more highly stable emulsions and promotes silica shell formation.

Keywords: oil-in-water emulsion, Pickering emulsion, Sol-gel encapsulation.

2

ACCEPTED MANUSCRIPT 1. Introduction

T

Textile structures are widely used in various applications such as clothing, insulation,

IP

absorption, or filtration due to their remarkable properties combining flexibility, low weight

SC R

and mechanical behavior. It can be however interesting to bring further functionalities to these textiles in order to increase their added value. In this context, specific ennobling or finishing

NU

operations allow to add colors, water repellent or antimicrobial properties for example. Nevertheless, some active agents are very sensitive to the environment, or are in the liquid

MA

state and so unable to be linked efficiently on textile surfaces. In order to avoid any interaction with the environment, limit the reactivity and overcome volatility, the active agent

D

can be encapsulated. The handling is then facilitated by transforming liquids products into

CE P

TE

solid powders allowing its release when necessary [1].

Encapsulation consists in coating small droplets or particles of active agents in order to entrap it as a core material with a protective natural or synthetic polymeric shell. A wide range of

AC

applications including cosmetics, drugs, perfumes, pesticides release or fire retardancy can be achieved by microencapsulation in various textile fields [2]. Fire retardant properties are particularly requested in numerous textile applications because of the high flammability, corrosive and toxic gases liberation of most synthetic polymers during combustion, and because of the more drastic safety standards. Active capsules or particles can be fixed on textiles by various ways in order to provide a durable functionalization. They can be blended with the polymer during spinning [3] or incorporated onto the manufactured textile by conventional finishing process such as padding [4] or coating without modification of the intrinsic properties of the support material [5].

3

ACCEPTED MANUSCRIPT For environmentally friendly purpose, a new dry powder impregnation technology for nonwoven structures by functional microcapsules is used [6]. It consists in the dispersion of

T

microcapsules by an alternative electrostatic field in order to avoid water utilization and

IP

pollution by binders and capsules, and to reduce thermal energy consumption due to water

SC R

evaporation after binding. To efficiently implement this process, capsules need to reach some particular specifications. They have to be within the size of 10 to 100 µm to allow particles mobility and migration through the textile pores of the nonwoven. Then, they need a

NU

satisfying thermal stability in order to stay undamaged during the fixation on the fibers which

MA

is performed around 150°C. Regardless the impregnation process, capsules need to have sufficient mechanical properties to resist to stress during utilization. Silica shells or matrices are frequently used to encapsulate various types of active agents such as dyes [7], drugs [8],

TE

D

enzymes [9], bacteria [10] or fire retardant [3], and are widely used in cosmetics formulation because of their biocompatibility and limited toxicity [11]. Moreover, they exhibit high

CE P

mechanical properties [12], excellent chemical and thermal stabilities with limited release of toxic gases during decomposition. This is due to the excellent thermal stability and to the high

AC

heat resistance of Si-O-Si chains and of the relative crosslinked matrix [13]. Thus, silica particles are often used alone as fire retardant fillers [14]. As encapsulation shell, it offers a thermal insulation layer and a protective screen from further thermal degradation of the polymer residue, even at high temperature, and participates to char formation. It also improves thermal behaviour of fire retardants as ammonium polyphosphate by delaying the pyrolysis process in polyurethanes [15], and exhibits enhanced fire resistance and improved thermal stability encapsulating Bisphenol A bis (diphenyl phosphate) (BDP) in polyethylene terephthalate matrix [16]. Furthermore, sol-gel synthesis has the advantage of being carried out in very mild conditions, in aqueous solution, at low temperature, and under atmospheric

4

ACCEPTED MANUSCRIPT pressure.

T

In this study, two different actives (castor oil and BDP) have been encapsulated by silica

IP

shell. Castor oil, a renewable natural oil, has been encapsulated as a model compound. BDP is

SC R

a halogen-free aryl phosphate effective fire retardant used increasingly because of environmental and/or regulatory issues. It is widely used as a copolymer during polymerization because of its good thermal and hydrolytic stability. The low volatility allows

NU

long term stability and ageing [17], even if it tends to exude at high concentrations [18]. In

MA

this study, the purpose is the encapsulation of BDP to provide fire resistance to previously manufactured nonwoven textiles. Indeed, sol-gel encapsulation of this highly viscous colorless fluid improves polyethylene terephthalate thermal behavior and displays good flame

TE

D

retardancy for charring polymers as polycarbonates. [3,19].

CE P

A classical emulsion process, using a common surfactant reducing the interfacial tension and facilitating the droplets division is compared to pickering emulsion which is based on the

AC

utilization of nanoparticles with a strong anchoring at the interface and limiting the coalescence of the droplets to stabilize the oil-water interface. Polyoxyethylene (20) sorbitan monolaurate (Polysorbate 20 or Tween®20) is a non-ionic hydrophilic surfactant with a hydrophilic-lipophilic balance of 16.7. It is used as an oil-in-water emulsifier for its good stability and relative non-toxicity in many detergents and emulsification formulations from domestic to food or pharmaceutical applications. Aerosil R816 is a spherical powder of fumed silica particles with an average diameter of 12 nanometers. It corresponds to Aerosil 200 hydrophilic particles treated with hexadecylsilane. After high temperature treatment, particles are partially hydrophobic due to the grafting of dimethylsilyl groups on silica particles surface [20]. This strong SiO-Si linkage presents excellent thermal and chemical stability [21]. Getting of oil-in-water or water-

5

ACCEPTED MANUSCRIPT in-oil emulsions depends on the hydrophilic/lipophilic groups rario modifying their wettability

[22]. These particles are usually used in water-based coating systems or cosmetics and can be used for

T

oil-in-water emulsification [23]. Emulsions based on nanoparticles are prepared in the same way than

IP

classical emulsion, and are known to have an improved stability, longer than several months [24, 25].

SC R

In addition to its high stability performances, pickering emulsion is used in this work to insulate the interface between the two phases and avoid to any material transfer as proved by Arditty [26]. In

NU

addition, it improves the shell rigidity and acts as an interfacial barrier against deformation [20].

MA

The aim of this study is to encapsulate one FR compound, the BDP for textile functionalization. To handle the functionalization process, capsules need to be ranged

D

between 10 and 100 µm and to support temperature higher than 150°C. To manage this

TE

Pickering emulsion is combined with sol-gel process.

CE P

2. Materials and Methods

AC

2.1. Materials

BDP (Devan Chemicals, Belgium), castor oil (Sigma Aldrich, France) and cyclohexane (Sigma Aldrich) are used respectively as fire retardant, inert and removable core. Tetraethylorthosilicate (TEOS) (Sigma Aldrich) is used as shell material. Polyoxyethylene (20) sorbitan monolaurate (Tween®20) (Sigma Aldrich) is employed as an emulsifier and Evonik Aerosil R816 (Safic Alcan, France) is employed as a solid emulsion stabilizer for Pickering emulsions. Cetyltrimethylammonium bromide (CTAB) (Sigma Aldrich) is used as a structure-direction agent combined with Pickering emulsion. Formic acid and sodium

6

ACCEPTED MANUSCRIPT hydroxide purchased from Aldrich are used as pH control agents. All products are used

T

without any further purification before use.

SC R

IP

2.2. Preparation of microcapsules

Both castor oil and BDP are non-soluble in water and primarily dispersed into an aqueous

NU

solution during the emulsion process thanks to emulsifiers. Emulsification is performed by an important mechanical stirring leading to the shredding of the oil into small droplets.

MA

Preliminary tests have been performed in order to confirm castor oil and BDP droplets stabilization by Aerosil R816, to determine the particle content, and to confirm the extended

D

stability with this system. For both core (castor oil or BDP), emulsions have been prepared

TE

with 10 wt.-% of core and 0.5, 1 and 3 wt.-% of dried particles by mechanical stirring at 1000

CE P

rpm performed with a four titled blades propeller.

The preparation of microcapsules is illustrated in figure 1. The emulsion of 10 g of the core

AC

material is firstly prepared by mechanical stirring using a four titled blade propeller running at 1000 rpm into 100 ml of an aqueous phase containing 1 g of Tween®20 or 0.1 g of Aerosil R816 silica particles. 0.5 g of CTAB is added after the emulsion in the case of silica particles to promote hydrolysed silanol migration at the droplets surface [27]. 100 ml of 10 wt.-% tetraethoxysilane (TEOS) solution, previously hydrolyzed at pH 2.8, is then added dropwise in the emulsion. The mixture is kept under stirring at 45°C during 24 hours to initiate silane condensation. Sodium hydroxide aqueous solution (10 wt.-%) is gradually added up to pH 6, close to neutralization, in order to speed up the condensation and to shape a thick shell around the active substance. Capsules are then matured for 1 hour, filtered, rinsed with water and dried 24 hours at 50°C. A thin powder is then obtained. 7

ACCEPTED MANUSCRIPT

Cyclohexane, a highly volatile solvent, has been encapsulated by the same preparation

T

method and evaporated one night at 80°C to be used as blank for FT-IR comparison and TGA

SC R

IP

analysis.

NU

2.3. Analytical methods

MA

2.3.1. Particle size analysis and morphological characterization

Particle diameters were characterized with a laser-light blocking technique (AccusizerTM,

D

model 770, Particle sizing systems, Santa Barbara, CA, connected to C770 software version

TE

2.54). The particle size distribution, in number, was obtained one particle at a time, by

CE P

comparing the detected pulse heights with standard calibration curve, obtained from a set of uniform particles of known diameter. Measurements were performed directly in the preparation solution at room temperature without any dilution. Data obtained were presented

AC

as a size distribution curve and expressed as the mean particle diameter. The average values and standard deviations were calculated.

The microscopic aspects of the particles were observed by both optical microscopy (Axioskos Zeiss) equipped with a camera (IVC 800 12S) and scanning electron microscopy (JEOL 6490LV) under partial vacuum of 30 Pa at an accelerated voltage of 10-16 kV without any metal coating.

2.3.2. Chemical characterization

8

ACCEPTED MANUSCRIPT

The chemical structures of the shell polymer, the core and the encapsulated core were

T

analyzed by Fourier transform infrared spectroscopy (FT-IR). Samples were grounded and

IP

mixed with KBr to prepare pellets. Spectra were recorded in absorbance mode with 128 scans

SC R

and a resolution of 4 cm-1 by a Nicolet Nexus instrument connected to a computer.

NU

2.3.3. Thermal stability

MA

The thermogravimetric analysis (TGA) was carried out on a TA 2050 Instrument under nitrogen or air atmosphere at a purge rate of 50 ml/min. For each experiment, a sample of

D

approximately 10 mg was used. A heating rate of 10°C/min was applied, and the temperature

TE

was raised from 20 to 800°C. In addition to the weight loss percentage (TG) curves, the

CE P

derivative weight loss percentage (DTG) was calculated for each sample.

AC

3. Results and discussion

3.1. Emulsion of the cores

For the preparation of microcapsules, the first and limiting step consists in the emulsion process. Before shell formation, it is indeed necessary to shape the core droplets in the synthesis medium with satisfying size and stability. Preliminary experiments have been carried out using Tween®20 as surfactant to optimize the formation of stable oil-in-water emulsion for both cores. 1 g of Tween®20 was used to emulsify 10 g of both cores in 100 ml of the aqueous solution under a 1000 rpm stirring performed with a four titled blades 9

ACCEPTED MANUSCRIPT propeller. A monodispersed and stable emulsion with an average diameter of approximately 30 µm was obtained. To compare classical emulsion with pickering emulsion characteristics,

T

study of castor oil and BDP droplets shirring and stability was also carried out with Aerosil

IP

R816. 0.5, 1 and 3 wt.-% of fumed silica particles were used to stabilize the droplets obtained

SC R

in the same conditions: 1000 rpm stirring conducted during 30 minutes using a four titled blades propeller with 10 g of core added drop by drop into 100 ml of aqueous solution. Figure 2 displays optical microscopy photographs magnified by 10 of emulsions prepared in this way

MA

NU

and aged from 5 minutes to 1 week after the stirrer switches off.

In all cases, individual droplets from few micrometers to 500 µm are observed. For both cores, “bigger” droplets from 100 to 500 µm are observed with 0.5 wt.-% of particles.

TE

D

Differences of droplets size between 1 wt.-% and 3 wt.-% are not pointed out and most

CE P

droplets appear smaller than 100 µm.

In each case, stability appears largely satisfying, since after 1 week of storage, phases remain

AC

separated with very little rearrangement and size variation. Actually, size does not increase with the storage time. No coalescence or Ostwald ripening is observed after ageing. No limited coalescence phenomenon is observed after 5 minutes. Either the droplets coverage by nanoparticles is achieved, and size is determined by the stirring process efficiency or limited coalescence occurs in less than 5 minutes. Chevalier and Bolzinger [20] have described three distinct regimes depending on silica nanoparticles content in oil-in-water emulsions. In the first one, silica content is too small to stabilize the droplets and emulsion failed. In the second one, all the particles are anchored at the interface and droplets size depends on the particles amount, whereas in the third regime, size is controlled by the stirring process and the excess of particles stays aggregated in the aqueous solution, leading to a thickening of the solution.

10

ACCEPTED MANUSCRIPT In our experimental conditions, 0.5 wt.-% particles correspond to the second regime, whereas 1 wt.-% and 2 wt.-% may be attributed to the third one. Sizes being preferably comprised

T

between 10 and 100 µm in order to allow capsules migration through textile pores in the

IP

nonwoven structure, 1 or 2 wt.-% particles emulsions are preferred to the 0.5 wt.-% one. To

SC R

avoid the thickening of the solution and interactions between silanols and particles in excess during the shell synthesis, the smaller excess of particles is favored. In these conditions, the formulations based on 1 wt.-% of silica nanoparticles have been selected for both castor oil

MA

NU

and BDP encapsulation with our experimental parameters.

After storage at room temperature, creaming and/or sedimentation phenomena of the droplets

TE

D

were observed after few minutes, related to the difference in density between the two phases (figure 3). Indeed, for BDP which density is 1.284, all droplets sedimented with a size

CE P

segregation and larger droplets stay at the bottom of the test tube, whereas the smaller the droplet, the higher is its localization in the sedimented phase. On the contrary, density of

AC

castor oil being 0.955 which is really close to the aqueous phase density, limited creaming phenomenon occurs. Combined with silica nanoparticles which density is close to 2.2, droplets exhibit various densities depending on the silica content at the interface. This leads to the phase separation observed for the castor oil emulsions prepared with three silica contents. Thus, creaming and sedimentation occur in the same mixture for these emulsions and no size segregation is observed.

In spite of creaming and/or sedimentation, droplets stabilized by silica nanoparticles remain stable and no coalescence is observed during more than several weeks, whereas with classical surfactant, phase separation occurs during the few hours after the end of stirring. A gentle

11

ACCEPTED MANUSCRIPT manual shaking, sufficient to scatter again droplets in the medium, is performed before each

IP

T

microscopic observation.

SC R

Besides, with the same amount of castor oil or BDP, different sedimentation and/or creaming volumes are observed, in agreement with the literature [28-30]. In Song et al. study [28], it has also been shown that for high amount of particles (higher than 5 wt.-%), the creaming

NU

volume remains constant. According to French et al. [29], increasing particle content leads to

MA

a decrease of the emulsion propensity to aggregate. Indeed, for low particle amount, clusters of aggregated droplets are observed, and droplets are bridged due to the interface area that exceeds the surface able to be covered. When the particles amount increases, bridging occurs

TE

D

less and less and interface became smoother. For even higher particles content, free silica particles in the continuous phase prevent droplets aggregation and leads to the growing of the

CE P

creaming volume.

Figure 4 displays the size distribution, in number, of 10 g of castor oil and BDP emulsified

AC

with 1 g of Tween®20 or with 1 wt.-% of Aerosil R816 silica nanoparticles. Emulsion processing was carried out by a four blade propeller at 1000 rpm. Whatever the encapsulated oil phase and the kind of emulsion, the particles exhibit a size distribution ranged from 0.5 to 100 µm. The average size is close to 30 +/- 18 µm for all formulations. Moreover, 90 to 95% of the droplets have a diameter included between 10 and 100 µm as preferred in this case study.

Thus, emulsification of castor oil and BDP in an aqueous solution has been successfully performed with a classical non ionic surfactant and with solid silica nanoparticles. All

12

ACCEPTED MANUSCRIPT emulsions prepared present an average diameter around 30 µm and at least 90% of the

Silica shell formation and hardening

SC R

3.2.

IP

T

droplets have a size comprised between 10 and 100 µm.

NU

Silica particles were prepared by sol-gel encapsulation after previous emulsions. Sol-gel polymerization takes place in two steps, i.e. the hydrolysis of silica polymer is followed by its

MA

condensation in order to initiate Si-O-Si network formation. Hydrolysis consists in the substitution of an alkyl group by a hydroxyl on silicon tetraoxide in the presence of water to

D

form a silanol with liberation of an alcohol molecule. On the contrary, condensation consists

TE

in the reaction of two silanols to create a Si-O-Si bridge with liberation of one molecule of water. Hydrolysis is catalyzed under acidic and basic conditions with a minimum rate at pH 7

CE P

and an exponential increase for both low and high pH. On the contrary, condensation displays a minimum at pH 2, and a maximum around pH 7. Moreover, the pH choice allows

AC

controlling the size and the shape of particles. Actually, low pH increases the hydrolysis rate and inhibits condensation which is, in these conditions, the limiting step. This leads to small and uniform growth of capsules bringing to dense and homogeneous polymeric shells. In contrast, basic condition promotes fast condensation leading to inhomogeneous shell formation and aggregates [31]. Thus in this work, shell polymerization was firstly carried out under acidic conditions for 24 hours in order to promote the controlled growth of shell before pH neutralization leading to fast condensation and hardening of the shell as shown in figure 5. Moreover, a cationic surfactant Cetyltrimethylammonium bromide (CTAB) was added after emulsion completion and before hydrolyzed TEOS addition in the medium to assist silanols migration at the 13

ACCEPTED MANUSCRIPT droplets interface before condensation. Optical microscopy was also performed during the particles synthesis and photographs are shown for each step in figure 5. No significant

T

morphological modification is observed during the early shell formation conducted during the

IP

24 first hours. On the contrary as expected, pH neutralization leads to fast shell formation and

SC R

particles aggregation in the shape of clusters. This aggregation should even be accentuated by the presence of the cationic surfactant.

NU

In order to assess the chemical composition of synthesized silica microparticles, FT-IR studies were carried out. Results are shown in figure 6. Figure 6 (a) and (b) correspond respectively

MA

to castor oil and BDP based particles previously emulsified with Tween®20 and pickering emulsion. In figure 6 (a), spectrum of pure and encapsulated castor oil displays characteristic

TE

D

absorption bands. Two intense absorption peaks are indeed observed at 2922 and 2854 cm-1 and are attributed to C-H stretching vibrations. The band located between 3500 to 3200 cm-1

CE P

is attributed to O-H stretching vibration and the one at 1748 cm-1 to vibrations of the C-O groups of castor oil. Moreover, encapsulated oil presents at the same time characteristic peaks

AC

of oil and shell. Spectrum of capsules prepared with both emulsifiers, present symmetric and asymmetric Si-O-Si stretching vibrations at 1088 and 789 cm-1 respectively. The band around 3200-3500 cm-1 and the one at 1650 cm-1 are attributed to uncondensed silanols Si-OH vibrations. Likewise, bands of BDP and polysiloxane shell can each be observed in figure 6 (b). Characteristic bands of BDP can be distinguished in all spectra, i.e; 3000-3100 cm-1 band is attributed to aromatic C-H stretching vibration, and between 400 and 1600 cm-1 various distinctive peaks can be observed. The band at 1593 cm-1 is attributed to C=C stretching, the one at 1491 cm-1 to C-C stretching and the one at 841 cm-1 to C-H bending in aromatics groups. P=O stretching in pentavalent phosphorous compounds is observed at 1303 cm-1 and, aromatic-O and P-O stretching in pentavalent phenyl phosphate are noticed at 1188 and 956

14

ACCEPTED MANUSCRIPT cm-1 respectively [16]. This confirms the encapsulation of castor oil and BDP by sol-gel

IP

Drying of particles

SC R

3.3.

T

process to create a polysiloxane shell for the two kinds of emulsifiers.

Once silica shell is completely formed and hardened, microparticles are filtered, rinsed three

NU

times with water and dried 24 hours at 50°C. A thin powder is obtained. Water rinse allows removal of both silicon tetraoxide that did not react and of ethanol liberated during the

MA

hydrolysis. Dry powder has been observed by scanning electron microscopy and results are shown in figure 7. Aggregates from few micrometers to 500 µm can be observed. They are

D

composed of interconnected silica capsules. They can be formed by coagulation or

TE

crosslinking between particles during the synthesis, and particularly during the increase of pH leading to uncontrolled and heterogeneous condensation of silanols in the continuous

CE P

medium. Moreover, this particle networking could be even magnified by the huge quantity of cationic surfactant used to foster silanols migration towards the interface. Powders obtained

AC

from Tween®20 based emulsion appear rougher and more agglomerated than pickering based encapsulation. It seems also that powder made with nanoparticles presents a much bigger quantity of small particles leading to a higher size distribution.

Thermal stability of the capsules produced has been investigated by thermogravimetric analysis to check the ability of capsules to impregnate nonwoven structures. Figures 8 (a) and (b) display respectively thermograms of raw and encapsulated castor oil and BDP previously emulsified with Tween®20 or silica nanoparticles, and figures 8 (c) and (d) their respective derivative curves. Castor oil is completely degraded in one step with 5% of weight loss at 350°C, the maximum loss rate Tmax being observed for 404°C and 95% weight loss at 450°C 15

ACCEPTED MANUSCRIPT (Figure 8 a and c). Encapsulated castor oil after classical emulsion exhibits the same one step degradation profile with 5% weight loss up to 180°C attributed to the evaporation of water

T

and ethanol molecules trapped in the polymeric shell. A 21% weight residue is observed and

IP

is attributed to the condensed silica shell. On the contrary, encapsulated castor oil after

SC R

pickering emulsion exhibits a more complex degradation profile. Indeed, after the same 5% weight loss under 180°C attributed to trapped water and ethanol molecules liberation, another degradation step corresponding to 10 to 15% weight loss is seen between 180°C and 280°C.

NU

This small weight loss is also observed for the hollow silica shell prepared with cyclohexane

MA

which has been evaporated after synthesis. It has been attributed to the deshydroxylation of OH groups on surface of particles, and to the degradation of the CTAB cationic surfactant which occurs between 220°C and 320°C (unshown data). For the hollow capsules, the next

TE

D

progressive 10% weight loss, between 300°C and 600°C is attributed to isolated hydroxyls degradation [32]. Thus, the following degradation step of pickering capsules, occurring

CE P

between 280°C and 500°C, with a maximum weight loss rate occurring for 369°C mostly corresponds to the degradation of the castor oil core and to a lesser extent to isolated

AC

hydroxyls groups degradation as for hollow capsules. In this case, 27% of residue is observed which is 5% higher than for the classical emulsion. 5% being much more important than silica nanoparticles content, it can be assumed than nanoparticles promote dense silica shell formation at the oil and water interface, leading to an increase in its rigidity and thermal stability. This is in concordance with scanning electron microscopy analysis showing smoother and denser particles when prepared with pickering emulsions than prepared with a classical surfactant.

BDP degradation is shown in figure 8 (b) and (d). The thermal behaviour of BDP exhibits a complex mechanism of degradation which occurs in three steps. The first one matches 6%

16

ACCEPTED MANUSCRIPT weight loss between 200 and 310°C, the second and main one 82% between 310 and 460°C, and the last one 11% between 460 and 520°C. Finally, a BDP residue of 1% remains after

T

total degradation. The degradation of encapsulated BDP begins at lower temperature than row

IP

BDP for both emulsions and corresponds to silica shell degradation. For the classical

SC R

emulsion, one step is observed, i.e. 63% of the weight, correlated to BDP content, is lost between 200°C and 600°C, and residue represents 31% of the sample weight. With silica particles, the first 10 to 15% are attributed to trapped water and ethanol liberation, to surface

NU

OH deshydroxylation, and to CTAB degradation as for encapsulated castor oil. The main

MA

degradation correlated to BDP degradation is also extended and occurs from 250°C to 600°C. The 40% remaining residue are attributed to the shell. This higher residue weight is also explained by silica nanoparticles promotion of silica shell formation. For both classical and

TE

D

pickering emulsions, this extended degradation and higher residue amount are attributed to the

CE P

formation of a char due to BDP [33].

With Tween®20, thermal stability of castor oil is relatively unaffected by the emulsifier,

AC

whereas degradation of BDP occurs 100°C earlier due to the degradation of silica shell. With pickering emulsions for both cores, the degradation of the capsules begins at lower temperature than the degradation of row core, but thermal degradation temperature range is extended. Additionally, the presence of silica nanoparticles leads to an increase of the weight of residue which has been attributed to sol-gel process promotion during the silica shell formation. Nevertheless, for all capsules, with the two cores and two kinds of emulsion, thermal stability stays largely satisfying for nonwoven impregnation, and the amount of residue corresponding to condensed silica in the shell remains important.

4. Conclusion

17

ACCEPTED MANUSCRIPT

Silica shell has been successfully synthesized to encapsulate bisphenol A bis (diphenyl

T

phosphate) and castor oil for textile functionalization. Actually, a thin powder making

IP

handling easier is obtained from liquid actives. It was found that pickering emulsion based on

SC R

the utilization of silica nanoparticles designed to stabilize droplets interface during emulsion really increases the stability of the emulsions for more than a week. In order to emulsify 10 wt.-% of each core with a 1000 rpm stirring, 1 wt.-% of particles has been selected to avoid

NU

any thickening of the solution and alteration during the sol-gel process.

MA

The influence of the nature of emulsifier on the emulsion characteristic was considered by comparison of optical and macroscopic observations of droplets suspension and their influence on sol-gel encapsulated particles by Fourier Transform Infrared Spectroscopy,

TE

D

thermogravimetric analysis and scanning electron microscopy. It shows no particular incidence on the chemical structure of capsules. Nonetheless for both cores, thermal stability

CE P

is affected by nanoparticles, and the presence of CTAB leads to an earlier but more extended degradation temperature range compared with classical emulsion. Likewise, morphology of

AC

capsules produced with pickering emulsion defers and are smoother, denser and less aggregated.

ACKNOWLEDGMENTS

This work was supported by research grants FUI FOMOTEX program supported by Techtera, Up-Tex and with Fibroline company (Ecully, France) as lead manager.

18

ACCEPTED MANUSCRIPT REFERENCES

[1] S.S. Bansode, S.K. Banarjee, D.D. Gaikwad, S.L. Jadhav, R.M. Thorat,

IP

T

Microencapsulation: a review, Int. J. Pharm. Sci. Rev. Res. 1 (2010) 38-43.

SC R

[2] G. Nelson, Application of microencapsulation in textiles, Int. J. Pharm., 242 (2002) 55-62.

[3] F. Salaün, G. Creach, F. Rault, X. Almeras, Thermo-physical properties of polypropylene

NU

fibers containing a microencapsulated flame retardant, Polym. Adv. Technol. 24 (2013) 236-

MA

248.

[4] K. Son, D.I. Yoo, Y. Shin, Fixation of vitamin E microcapsules on dyed cotton fabrics,

TE

D

Chem. Eng. J.,239 (2014) 284-289.

CE P

[5] F. Salaün, E. Devaux, , P. Rumeau, P.O. Chapuis, S.K. Saha, S. Volz, Polymer nanoparticles to decrease thermal conductivity of phase change materials, Thermochim.

AC

Acta. 477 (2008) 25-31.

[6] J. Marduel, Process for impregnating a fibrous, filamentary and/or porous network with powder using electrodes subjected to an AC electric field, U. S. Patent 7534473 B2, 19 may 2009.

[7] T.Z. Ren, Z.Y. Yuan, B.L. Su, Encapsulation of direct blue dye into mesoporous silicabased materials, Colloids Surf. 300 (2007) 79-87.

19

ACCEPTED MANUSCRIPT [8] B. Chen, G. Quan, Z. Wang, J. Chen, L. Wu, Y. Xu, Hollow mesoporous silicas as a drug

T

delivery system for insoluble drugs, Powder Technol. 240 (2013) 48-53.

IP

[9] S.I. Matsuura, S.A. El-Safty, M. Chiba, E. Tomon, T. Tsunoda. T.A. Hanaoka, Enzyme

SC R

encapsulation using highly ordered mesoporous silica monoliths, Mater. Lett. 89 (2012) 184187.

NU

[10] S. Fennouh, S. Guyon, C. Jourdat, J. Livage, C. Roux, Encapsulation of bacteria in silica

MA

gels, C.R. Acad. Sci. Paris IIc 2 (1999) 625-630. [11] Y.W. Chen-Yang, Y.T. Chen, C.C. Li, H.C. Yu, Y.C. Chuang, J.H. Su, Y.T. Lin, Preparation of UV-filter encapsulated mesoporous

TE

D

silica with high sunscreen ability, Mater. Lett. 65 (2011) 1060-1062.

[12] L. Zhang, M. D’Acunzi, M. Kappl, G.K. Auernhammer, D. Vollmer, C.M. van Kats, A.

CE P

Van Blaaderen, Hollow silica spheres synthesis and mechanical properties, Langmuir 25(5)

AC

(2009) 2711-2717.

[13] J. Feng, J. Hao, J. Du, R. Yang, Flame retardancy and thermal properties of solid bisphenol A bis(diphenyl phosphate) combined with montmorillonite in polycarbonate, Polym. Degrad. Stab. 95 (2010) 2041-2048.

[14] F. Laoutid, L. Bonnaud, M. Alexandre, J.M. Lopez-Cuesta, P. Dubois, New prospects in flame retardant polymer materials: From fundamentals to nanocomposites, Mater. Sci. Eng., R 63 (2009) 100-125.

20

ACCEPTED MANUSCRIPT [15] B. Wang, H. Sheng, Y. Shi, W. Hu, N. Hong, W. Zeng, H. Ge, X. Yu, L. Song, Y. Hu, Recent advances for microencapsulation of flame retardant, Polym. Degrad. Stab. Xxx (2015)

IP

T

1-14.

SC R

[16] F. Salaün, G. Creach, F. Rault, S. Giraud, Microencapsulation of bisphenol-A (diphenyl phosphate) and influence of particle loading on thermal and fire properties of polypropylene

NU

and polyethylene terephthalate, Polym. Degrad. Stab. 98 (2013) 2663-2671.

MA

[17] S.V. Levchik, E.D. Weil, A review of recent progress in phosphorus-based flame retardants, J. Fire Sci. 24 (2006) 345-364.

TE

D

[18] S.V. Levchik, E.D. Weil, Flame retardancy of thermoplastic polyesters a review of the

CE P

recent literature, Polym. Int. 54 (2005) 11-35.

[19] K.H. Pawlowski, B. Schartel, Flame retardancy mechanisms of triphenyl phosphate,

AC

resorcinol bis(diphenyl phosphate) and bisphenol A bis(diphenyl phosphate) in polycarbonate/acrylonitrile/butadiene/styrene blends, Thermochim. Acta. 498 (2010) 92-99.

[20] Y.Chevalier, M.A. Bolzinger, Emulsions stabilized with solid nanoparticles: pickering emulsions, Colloids Surf., A 439 (2013) 23-34.

[21] H. Alloul, T. Roques-Carmes, T. Hamieh, A. Razafitianamaharavo, O. Barres, J. Toufaily, F. Villieras, Effect of chemical modification on surface free energy companents of Aerosil silica determined with capillary rise technique, Powder Technol. 246 (2012) 575-582.

21

ACCEPTED MANUSCRIPT [22] R. Lorentz, Y. Rahali, S. Issa, Y. Bensouda, G. Holtzinger, A. Aoussat, A.M. PenséLhéritier, One-pot synthesis of sub-micro organosilicate particles for the formulation of

IP

T

Pickering emulsions, Powder Technol. 264 (2014) 446-457.

silica and calcite, Particuology 8 (2010) 390-393.

SC R

[23] S. Wang, Y. He, Y. Zou, Study of pickering emulsions stabilized by mixed particles of

MA

Colloids Surf., A. 132 (1998) 257-265.

NU

[24] B.R. Midmore, Preparation of a novel silica-stabilized oil/water emulsion,

[25] K. Zhang, W. Wu, H. Meng, K. Guo, J.F. Chen, Pickering emulsion polymerization :

TE

D

preparation of polystyrene/nano-SiO2 composite microspheres with core-shell structure,

CE P

Powder Technol. 190 (2009) 393-400.

[26] S. Arditty, V. Schmitt, J. Giermanska-Kahn, F. Leal-Calderon, Materials based on solid-

AC

stabilized emulsions, J. Colloid Interface Sci. 275 (2004) 659-664.

[27] Y. Balguerie, S. Bone, Procédé d'encapsulation de produit lipophile ou hydrophile dans une membrane polysiloxane , European Patent : EP 2 080 552 B1, 16 January 2009.

[28] X. Song, Y. Pei, M. Qiao, F. Ma, H. Ren, Q. Zhao, Preparation and characterizations of Pickering emulsions stabilized by hydrophobic starch particles, Food Hydrocolloids 45 (2015) 256-263.

22

ACCEPTED MANUSCRIPT [29] D.J. French, P. Taylor, J. Fowler, P.S. Clegg, Making and breaking bridges in a

T

Pickering emulsion, J. Colloid Interface Sci. 441 (2015) 30-38.

IP

[30] J. Liu, D. Yin, S. Zhang, H. Liu, Q. Zhang, Synthesis of polymeric core/shell

SC R

microspheres with spherical virus-like surface morphology by Pickering emulsion, Colloids Surf., A. 466 (2015) 174-180.

NU

[31] R. Ciriminna, M. Sciortino, G. Alonzo, A. Schrijver, M. Pagliaro, From molecules to

MA

systems: sol-gel microencapsulation in silica-based materials, J. Am. Chem. Soc. 111 (2010) 765-789.

TE

D

[32] E.F. Vansant, P. Van De Voort, K.C. Vrancken, Characterisation and chemical

CE P

modification of the silica surface, Elsevier, Amsterdam, 1995.

[33] J. Feng, J. Hao, J. Du, R. Yang, Using TGA/FTIR TGA/MS and cone calorimetry to

AC

understand thermal degradation and flame retardancy mechanism of polycarbonate filled with solid bisphenol A bis(diphenyl phosphate) and montmorillonite, Polym. Degrad. Stab. 97 (2012) 605-614.

23

ACCEPTED MANUSCRIPT Figure captions :

T

Figure 1 – Overview of the preparation of microcapsules

SC R

IP

Prefered magnification : single column

Figure 2 – Emulsion’s stability according to the amount of particles by optical microscopy

Preferred magnification: double column

NU

(magnification by 10)

MA

Figure 3 – Effect of silica particles amount on castor oil and BDP emulsions’ macroscopic morphology

TE

D

Preferred magnification: single column

Aerosil R816

CE P

Figure 4 – Size distribution in number of castor oil and BDP emulsified with Tween®20 and

AC

Preferred magnification: single column

Figure 5 – Schematic representation and optical microscopy at the various steps of the sol-gel encapsulation Preferred magnification: single column

Figure 6 – FT-IR spectra of row and encapsulated castor oil (a) and BDP (b) Preferred magnification: single column

24

ACCEPTED MANUSCRIPT Figure 7 – SEM micrographs of microcapsules : Tween®20-Castor oil (a), Aerosil R816Castor oil(b), Tween®20-BDP (c) and Aerosil R816-BDP (d)

IP

T

Preferred magnification: single column

SC R

Figure 8 – TG curves of row and encapsulated castor oil (a) and BDP (b) and respectively their DTG curves (c) and (d)

AC

CE P

TE

D

MA

NU

Preferred magnification: double column

25

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

Figure 1

26

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

Figure 2

27

AC

Figure 3

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

28

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

Figure 4

29

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

Figure 5

30

AC

Figure 6

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

31

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Figure 7

32

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

Figure 8

33

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

Graphical Abstract

34

ACCEPTED MANUSCRIPT Highlights Silica shell particles was successfully prepared by sol-gel synthesis

o

1 wt.-% of silica nano particles allow preparation of highly stable emulsions

o

Utilization of nano particles conduce to smoother and denser particles

o

Both classical surfactant and nano particles lead to satisfying thermal stability

AC

CE P

TE

D

MA

NU

SC R

IP

T

o

35