Liquid Marbles Prepared from pH-Responsive Sterically Stabilized

ARTICLE pubs.acs.org/Langmuir

Liquid Marbles Prepared from pH-Responsive Sterically Stabilized Latex Particles Syuji Fujii,*,† Motomichi Suzaki,† Steven P. Armes,§ Damien Dupin,§ Sho Hamasaki,† Kodai Aono,† and Yoshinobu Nakamura†,‡ †

Department of Applied Chemistry, Faculty of Engineering, and ‡Nanomaterials Microdevices Research Center, Osaka Institute of Technology, 5-16-1 Ohmiya, Asahi-ku, Osaka 535-8585, Japan § Department of Chemistry, The University of Sheffield, Dainton Building, Brook Hill, Sheffield S3 7HF, United Kingdom W Web-Enhanced bS Supporting Information b

ABSTRACT: Submicrometer-sized pH-responsive sterically stabilized polystyrene (PS) latex particles were synthesized by dispersion polymerization in isopropyl alcohol with a poly[2-(diethylamino)ethyl methacrylate]- (PDEA-) based macroinitiator. These PDEA
PS latexes were extensively characterized with respect to their particle size distribution, morphology, chemical composition, and pH-responsive behavior. Millimeterand centimeter-sized “liquid marbles” with aqueous volumes varying between 15 μL and 2.0 mL were readily prepared by rolling water droplets on the dried PDEA
PS latex powder. The larger liquid marbles adopted nonspherical shapes due to gravitational forces; analysis of this deformation enabled the surface tension to be estimated. Scanning electron microscopy and fluorescence microscopy studies indicated that flocs of the PDEA
PS particles were adsorbed at the surface of these water droplets, leading to stable liquid marbles. The relative mechanical integrity of the liquid marbles prepared from alkaline aqueous solution (pH 10) was higher than those prepared from acidic aqueous solution (pH 2) as judged by droplet roller experiments. These liquid marbles exhibited long-term stability (over 1 h) when transferred onto the surface of liquid water, provided that the solution pH of the subphase was above pH 8. In contrast, the use of acidic solutions led to immediate disintegration of these liquid marbles within 10 min, with dispersal of the PDEA
PS latex particles in the aqueous solution. Thus the critical minimum solution pH required for long-term liquid marble stability correlates closely with the known pKa value of 7.3 for the PDEA stabilizer chains. Stable liquid marbles were also successfully prepared from aqueous Gellan gum solution and glycerol.

’ INTRODUCTION Liquid marbles,1
3 which are typically millimeter-sized water droplets stabilized by adsorbed particles at gas
liquid interfaces, have attracted increasing attention in view of their potential applications in cosmetics, pharmaceuticals, and home and personal care products.4 These liquid-in-gas dispersed systems are usually prepared from relatively hydrophobic particles that adsorb at the gas
liquid interface. Most of the literature is concerned with surface-modified lycopodium powder,1a
c hydrophobic silica particles,1d
h or carbon black,1i,j although there have also been a few examples of polymer latexes being used as liquid marble stabilizers.2 In principle, such synthetic particles should be particularly attractive for preparing liquid marbles, since they can be readily designed with specific surface chemistries (and hence wettability) by use of various functional monomers. Moreover, it is possible to confer stimulus-responsive character, which is more difficult to achieve with inorganic particles. For example, we reported2c that sterically stabilized polystyrene (PS) latexes synthesized from poly[2-(diethylamino)ethyl methacrylate] (PDEA) macromonomer can be used to prepare pH-responsive liquid marbles. In this case PDEA is a well-known pH-responsive polymer with a pKa value of 7.3.5 r 2011 American Chemical Society

Recently, some of us described the facile synthesis of sterically stabilized latex particles by dispersion polymerization using a PDEA-based macroinitiator as an inistab (initiator þ colloidal stabilizer).6 This azo-based macroinitiator is more readily prepared than the PDEA macromonomer reported in our previous study,2c since its synthesis involves atom transfer radical polymerization, which can be conducted easily under mild conditions. In contrast, the PDEA macromonomer was prepared by oxyanioninitiated polymerization, which requires rigorously dried solvent. Herein we describe the synthesis of submicrometer-sized PS latex particles carrying PDEA colloidal stabilizer on their surface (PDEA
PS latex) of low polydispersity by dispersion polymerization with this PDEA-based macroinitiator. The PDEA
PS latex particles were characterized in terms of their particle size, morphology, and chemical composition by laser diffraction, scanning electron microscopy, elemental analysis, X-ray photoelectron spectroscopy, and 1H NMR spectroscopy. In their nonprotonoated form, the PDEA stabilizer chains are sufficiently Received: April 11, 2011 Revised: May 12, 2011 Published: June 02, 2011 8067

dx.doi.org/10.1021/la201317b | Langmuir 2011, 27, 8067–8074

Langmuir hydrophobic to ensure effective liquid marble formation upon addition of water droplets to the dried PDEA latex. The resulting liquid marbles were characterized in terms of their mechanical stability and morphology by use of a droplet roller apparatus2e and both scanning electron microscopy and fluorescence microscopy. Moreover, the PDEA stabilizer chains on the surface of the latex particles confer stimulus-responsive behavior on these liquid marbles.

’ EXPERIMENTAL SECTION Materials. The chemicals used for the synthesis of PDEA-based macroinitiator are the same as those used in our previous study.6 PS homopolymer (average molecular weight 45 000), R,R0 -azobis[isobutyronitrile] (AIBN), 2-(diethylamino)ethyl methacrylate (DEA, 99%), glycerol (g99.5%), and isopropyl alcohol (IPA, 99%) were purchased from Sigma
Aldrich. 2,20 -Azobis[2-methyl-N-(2-hydroxyethyl)propionamide] (VA-086) was kindly provided by Wako Chemicals, Japan. Gellan gum (Kelcogel F) was kindly donated from CPKelco. Milli-Q water (Millipore Corp.) with a specific resistance of 18.2  106 Ω 3 cm was used in all experiments.

Synthesis of PS Latex Particles Carrying PDEA Colloidal Stabilizer. Submicrometer-sized PDEA
PS latex particles were

synthesized by the dispersion polymerization of styrene with a PDEA60 macroinitiator (mean degree of polymerization = 60) and VA-086 azo initiator in IPA at 80 °C. The PDEA macroinitiator was synthesized by atom transfer radical polymerization of DEA with 2,20 -azobis[2-methylN-[2-(2-bromoisobutyryloxy)ethyl] propionamide] as an initiator in IPA at 25 °C and was purified by passage through a silica column.6a The PDEA macroinitiator (6.55 g), VA-086 (0.59 g), and styrene monomer (30.0 g) were added to IPA (300 mL) in a round-bottomed 500 mL flask containing a magnetic stirrer bar. The number of moles of azo groups used in this polymerization formulation was 1.0 mol % based on styrene monomer. The reaction mixture was vigorously stirred at room temperature until the macroinitiator, VA-086 initiator, and monomer had dissolved completely. The reaction mixture was purged with nitrogen gas for 30 min in order to remove dissolved oxygen. The styrene polymerization commenced after the reaction vessel was placed in an oil bath set at 80 °C. The reaction was allowed to proceed for 168 h with continuous stirring at 500 rpm under a nitrogen atmosphere. Pyrenelabeled PDEA
PS latex particles were also synthesized under the same conditions as described above with 1-pyrenylmethyl methacrylate as a comonomer (1.0 wt % based on styrene). These PDEA
PS latexes were purified via 10 centrifugation
redispersion cycles (typically 6000 rpm for 30 min) in a centrifuge (Hitachi CF16RXII centrifuge). Each supernatant was replaced with fresh IPA and then water, and the final purified latex was freeze-dried for 48 h. Characterization of PDEA
PS Latex. Particle Size Analysis by Laser Diffraction. A Malvern Mastersizer 2000 instrument equipped with a small volume Hydro 2000SM sample dispersion unit (ca. 150 mL including flow cell and tubing), a HeNe laser operating at 633 nm, and solid-state blue laser source operating at 466 nm were used to size the latex particles. The stirring rate was adjusted to 2000 rpm. It was confirmed that the latex size did not change under these measurement conditions, which indicated no aggregation. The raw data were analyzed by use of Malvern software. The mean particle diameter was taken to be the volume mean diameter (D4/3), which is mathematically expressed as D4/3 = Σ Di4Ni/Σ Di3Ni (where Di is the diameter of individual particles and Ni is the number of particles corresponding to the Di diameter). Scanning Electron Microscopy. The dried latexes were placed on an aluminum stub and sputter-coated with gold by use of an Au coater (SC701 Quick Coater, Elionix, Japan) in order to minimize sample-charging problems. Scaning electron microscopy (SEM) studies were conducted on a VE-8800 instrument (Keyence, Japan) operating at 5 kV.

ARTICLE

Chemical Composition. CHN elemental microanalyses were carried out on a CHN-Corder MT-5 (Yanaco, Japan). The PDEA contents of PDEA
PS latexes were determined by comparing the nitrogen content of PDEA homopolymer (prepared by free radical solution polymerization in toluene with AIBN initiator) to that of each latex. X-ray Photoelectron Spectroscopic Study. For X-ray photoelectron spectroscopy (XPS) analysis, powdered samples were pressed into pellets (13 mm diameter, prepared at 80 kN for 20 min) and mounted onto sample stubs with conducting tape. XPS measurements were carried out on an Axis Ultra spectrometer with a monochromated Al KR X-ray gun. The base pressure was below 1.0  10
8 Torr. Pass energies of 80 and 20 eV were employed for survey spectra and elemental core-line spectra, respectively. Quantification of the atomic percent composition was obtained from the high-resolution spectra by use of the manufacturer’s sensitivity factors. Spectra were referenced to the hydrocarbon component of the C1s signal at 285 eV. Preparation of Liquid Marbles. Water droplets were deposited onto the dried PDEA
PS latex powder bed with a micropipet [Nichipet EX, Nichiryo (2
20 μL) or Micropipet, Eppendorf research (1000 μL)]. By gently rolling the aqueous droplet on the powder bed, the liquid was entirely encapsulated by the PDEA
PS latex, resulting in a liquid marble. The volume of the dispensed water droplet was varied from 15 μL to 2 mL. Water droplets were also sprayed onto the powder bed via an atomist spray equipped with a spray head (of 1.3 mm diameter aperture) to prepare multiple liquid marbles simultaneously. Characterization of Liquid Marbles. Fluorescence Microscopy. Fluorescence microscopic studies were conducted using a Nikon Eclipse LV100 microscope operating with episcopic illumination. The fluorescence excitation (band gap between 530 and 560 nm) and emission wavelengths (band gap between 575 and 650 nm) were controlled by a 49004 ET-CY3 filter block (Chroma Technology Corp). Fluorescent images were captured by a Lumenera Infinity 2.1 chargecoupled device (CCD) camera. Droplets of 10 and 50 μL of an aqueous solution containing 2.0 wt % Gellan gum and 0.01 wt % rhodamine B dye at 60 °C were stabilized with fluorescent PDEA
PS latex to form liquid marbles. After cooling to room temperature, the dyed aqueous droplet phase formed a hydrogel, which enabled the liquid marble to be cut in half for direct observation of its interfacial latex overlayer. The sectioned liquid marble was placed onto a glass slide prior to fluorescence microscopic observations. High-Speed Camera. For observation of the motion of liquid marbles, a high-speed camera (VW-6000, Keyence) equipped with a zoom lens (VH-Z200, Keyence) and a free angle stand (VW-S200, Keyence) was used. The shutter speed was set at 2 ms and a frame rate of 497.723 was utilized. Droplet Roller Experiment. A droplet roller setup was used to investigate the mechanical properties of the liquid marbles, as described previously2e (Figure S1, Supporting Information). An aqueous solution of alizarin blue black B dye (0.2 wt %) was pumped by a Perista pump (SJ-1211 L, Atto Co.) through silicone tubing (1.0 mm inner diameter), and water droplets emerged at a constant rate from a circular aperture of 500 μm diameter. The dye was used to aid droplet observation. The vertical distance between the tip end and the powder bed slope was set to be 2.0 cm, and the inclination angle of the bed slope was fixed at 20°. The mean weight of a single water droplet was 13 mg. The height between the edge of the slope and the powder bed (placed below the slope) was varied from 0 to 20 cm to investigate the mechanical integrity of the liquid marbles. All surfaces and components were thoroughly washed with IPA and dried before use.

’ RESULTS AND DISCUSSION PDEA homopolymer is a weak polybase that is soluble in aqueous media below pH at around 7 due to protonation of its 8068

dx.doi.org/10.1021/la201317b |Langmuir 2011, 27, 8067–8074

Langmuir

ARTICLE

Table 1. Quantitative Surface Composition and Microanalytical Data XPS result sample PS homopolymer

Figure 1. (a) Digital photograph and (b) SEM image of dried PDEA
PS latex particles.

tertiary amine groups. At pH 8 or above, PDEA homopolymer has either very low or zero charge density, which results in precipitation in basic aqueous media. PDEA-based polymers have been used for syntheses of latex particles,6,7 microgels,8 and shell cross-linked micelles.9 The number-average molecular weight of the PDEA-based macroinitiator used in this study was determined to be 23 600 by 1H NMR. Their relatively narrow molecular weight distributions (Mw/Mn ∼ 1.13, as determined by gel-permeation chromatography) indicated well-controlled atom transfer radical polymerization at 25 °C, which ensures preservation of the azo functionality. Moreover, there was also good agreement between the targeted number-average molecular weight of 22 700 (calculated from the monomer/initiator molar ratio) and those determined experimentally from 1H NMR studies. Dispersion polymerization with the PDEA macroinitiator and VA-086 initiator was conducted at a rate of initiation, Ri, of 7.9  1013 mL
1 3 s
1.10 Ri is given by 2kdf[I], where kd is the rate constant of decomposition, f is the initiator efficiency, and [I] is the initiator concentration. Ri was calculated with the assumption that the kd value for the PDEA macroinitiator is equal to that of the VA-086 initiator (1.21  10
5 s
1 at 80 °C)6a and that f is 0.30, as reported for other macroinitiators.11 Thus 87% of the azo groups were estimated to decompose at 80 °C within 168 h. Dispersion polymerization of styrene with this PDEA macroinitiator led to a colloidally stable PDEA
PS latex with no coagulum. Since the PDEA chains are soluble in the IPA medium, this macroinitiator acts as an effective steric stabilizer during the latex synthesis. After drying, the purified PDEA
PS particles were obtained as a fine white powder (see Figure 1a). A typical SEM image of this dried latex is shown in Figure 1b. A relatively monodisperse, spherical morphology can be observed: analysis of this image (and similar images) indicated a number-average diameter (Dn) and polydispersity (Dw/Dn) of 430 nm and 1.08, respectively. Particle size distributions obtained for dilute dispersions of PDEA
PS latex in IPA obtained by laser diffraction are shown in Figure S2 (Supporting Information). D4/3 of this latex was measured to be 450 ( 170 nm, which agrees well with its Dn; this suggests a high degree of dispersion for the PDEA
PS latex particles in IPA. Figure S3 (Supporting Information) shows the XPS survey spectra recorded for PDEA
PS particles, PDEA homopolymer (prepared by solution polymerization with AIBN initiator), and PS homopolymer. Oxygen, carbon, and nitrogen were detected for both PDEA homopolymer and PDEA
PS latex. Given that the XPS sampling depth is typically only 2
5 nm, these observations provide good evidence that the grafted PDEA stabilizer is present at the surface of the PDEA
PS latex particles, as expected. Moreover, the intensity of the N1s signal

C, %

chemical composition

N, % O, %

100

C, %

N, %

H, %

92.29

7.82

PDEA homopolymer

83.52 4.56 11.92

64.53

7.55

10.35

PDEA
PS latex

89.54 2.38

91.38

0.18

7.77

8.08

obtained for the PDEA
PS particles can be compared to that of the PDEA homopolymer in order to estimate a surface coverage of approximately 52% for the PDEA stabilizer chains on the particle surface (Table 1). High-resolution C1s spectra recorded for PDEA
PS latex, PDEA homopolymer, and PS homopolymer are shown in Figure 2. In the C1s core-line spectra obtained for PDEA
PS latex and PDEA homopolymer, two components due to the two carbon atoms (CdO and C—O) of the methacrylic ester groups were observed in addition to that due to the hydrocarbon atoms (C
C and C
H) on the polymer backbone. This is direct evidence for the surface-grafted PDEA stabilizer chains. A “shakeup” satellite at approximately 290
292 eV is clearly visible in the core-line spectra of PDEA
PS latex and PS homopolymer. This feature has been previously assigned to a π
π* transition for the aromatic rings of the PS component.12 In the O1s core-line spectra of PDEA
PS particles and PDEA homopolymer, two components due to the two carbon atoms (CdO and C—O) of the methacrylic ester groups were observed. In principle, an additional component due to oxygen atoms of the VA-086 initiator fragments bonded to the PS chains as end groups might be detectable at 533 eV. Unfortunately, this signal could not be identified. By comparison of the nitrogen content of PDEA
PS latex to that of PDEA homopolymer (N = 7.55%), elemental microanalyses indicated a PDEA stabilizer content of around 2.4% (Table 1). From these XPS results and elemental analyses, it can be confirmed that the PDEA stabilizer is located at the surface of the PS latex particles, as expected. The transmittance (% T) of a 0.7 wt % aqueous PDEA
PS latex was determined at various pH values in order to assess its degree of dispersion; see Figure S4 (Supporting Information). Above pH 8, the relatively hydrophobic PDEA
PS particles adsorbed as flocs at the planar air
water surface and also onto the glass walls of the cell, indicating substantial aggregation under these conditions: % T values were typically around 80%. However, % T dropped sharply to around 5% at pH 7.1. Below pH 7.1, the PDEA chains become extensively protonated, which aids spontaneous latex dispersion and good colloid stability, leading to more turbid dispersions. The critical pH required for such dispersions lies close to the pKa of 7.3 determined by acid titration for PDEA homopolymer.5 Thus the PDEA chains located at the surface of the latex particles dictate their aqueous behavior and confer pH-tunable wettability. Laser diffraction particle size distributions obtained for dilute aqueous PDEA
PS latex at acidic and basic pH are shown in Figure 3. The D4/3 was 470 ( 130 nm at pH 3 with a relatively narrow particle size distribution, suggesting a high degree of dispersion. On the other hand, the PDEA
PS particles were appreciably flocculated at pH 10, since there is a significant increase in the apparent particle diameter (D4/3 ∼ 8.0 μm). These laser diffraction results are also in good agreement with observations made by optical 8069

dx.doi.org/10.1021/la201317b |Langmuir 2011, 27, 8067–8074

Langmuir

ARTICLE

Figure 2. XPS core-line C 1s, N 1s, and O 1s spectra obtained for PDEA
PS particles, PDEA homopolymer, and PS homopolymer.

Figure 3. Laser diffraction particle size distribution curves obtained for PDEA
PS latex particles at pH 3 and 10. The dotted curve shown at around 0.1 μm was obtained after pH cycling (from pH 3 to 10 and back to pH 3).

microscopy (Figure S5, Supporting Information): colloidally stable particles at pH 3 were barely detected, whereas micrometer-sized flocs were observed at pH 10. Adjusting the solution pH from 10 to 3 again led to redispersion of the PDEA
PS latex particles: the D4/3 was 450 ( 180 nm, which is almost the same as that obtained originally at pH 3. This indicates that the PDEA
PS particles can recover their colloidal stability. It was also confirmed that this dispersion
flocculation transition was reversible over at least five cycles. After centrifugal washing, the purified latex was dried to obtain a fine white powder. Individual liquid marbles were prepared by rolling a droplet of deionized water over the dried PDEA
PS powder. This powder immediately coated the droplet, rendering it both hydrophobic and nonwetting. These liquid marbles remained intact after transfer onto a glass slide (see Figure 4), PET film or paper. When a fully coated liquid marble (15 μL, pH 6.5) was allowed to fall onto the PET film from a height of 2 cm, it bounced and retained its original shape (see Web enhanced object). High-speed digital photography confirmed that elastic deformation of the liquid marble occurred on contact with the PET film. These liquid marbles clearly have significant surface roughness, which suggests that they are coated with latex flocs

Figure 4. Digital photographs obtained for liquid marbles prepared from PDEA
PS particles. The droplet volumes are (a, b) 15 μL and (c) 1.0 mL, respectively. These liquid marbles are observed from (a) top and (b, c) side.

rather than just a monolayer.2c,e When the liquid marble bounced on the PET film, some excess PDEA
PS latex became detached from its surface. Fluorescence microscopy studies were conducted for a liquid marble prepared with water-soluble rhodamine B (0.01 wt %) dissolved in a 2.0% aqueous Gellan gum solution at 60 °C in combination with pyrene-labeled PDEA
PS (Figure 5). Figure 5a shows the surface morphology of a 10 μL liquid marble. Both pink and blue regions are observed, which indicates that the liquid marble surface is heterogeneous and stabilized by flocs of PDEA
PS latex particles. The air phase appears dark, since it contains no dye. The surface of the underlying aqueous Gellan gum solution (pink color) can be observed beneath these PDEA
PS flocs, which suggests that the liquid marbles are not fully covered with latex particles. This surface morphology is consistent with that observed in the digital images recorded for liquid marbles, which have a grainy and relatively rough topography (Figure 4). On cooling to room temperature, the liquid marble was sectioned with a scalpel and examined by fluorescence microscopy. This was only possible due to the nature of the Gellan gum, which is a free-flowing 8070

dx.doi.org/10.1021/la201317b |Langmuir 2011, 27, 8067–8074

Langmuir

ARTICLE

whereas the 2.0 mL liquid marbles tended to break up during their transfer. For a 1.0 mL liquid marble with a height H of 5.27 ( 0.18 mm, the effective surface tension (γ) was calculated to be 68 ( 4.6 mN 3 m
1 by use of eq 1 (see also Supporting Information):1b,13 γ¼ Figure 5. Fluorescence microscopic images of liquid marbles prepared with pyrene-labeled PDEA
PS latex from an aqueous alkaline solution (pH 10) containing 2.0 wt % Gellan gum and 0.01 wt % rhodamine B at 60 °C. Note the pink regions due the rhodamine B-dyed water and the blue aggregates from the pyrene-labeled PDEA
PS latex powder. (a) Surface morphology of a 10 μL liquid marble. (b) After cooling to 25 °C, a 50 μL liquid marble was sectioned with a scalpel. One of the liquid marble hemispheres was then placed on a glass slide.

Figure 6. Diameter of the contact area, height, and width of liquid marbles placed on a glass substrate containing various volumes of water (pH 6.5). The solid curve indicates the theoretical behavior if it is assumed that the water droplets have a perfectly spherical morphology independent of water volume.

aqueous solution at 60 °C but forms a hydrogel at 20 °C. The rhodamine B within the hydrogel gives rise to a pink coloration, which is surrounded by blue aggregates ranging from 10 to 300 μm due to the pyrene-labeled PDEA
PS latex particles (Figure 5b). These “double label” fluorescence images clearly demonstrate that PDEA
PS latex acts as a liquid marble stabilizer and is only located at the surface of the water droplet. Figure 6 shows the relationship between the diameter of the contact area between liquid marble and the glass substrate and the dimensions (height and width) of liquid marbles containing varying amounts of water. The solid curve indicates the theoretical behavior if it is assumed that the water droplets have a perfectly spherical morphology independent of water volume. For small volumes (15 μL), the liquid marbles are almost spherical, but greater deformation is observed as the droplet volume is increased (see Figures 4b,c and Figure S6, Supporting Information). Larger droplet volumes led to greater deviations of the contact width and the liquid marble dimensions from the theoretical values expected for a sphere. Gravitational forces cause the liquid marbles to deviate from their ideal spherical shape. Liquid marbles could be prepared on the powder bed by use of up to 2.0 mL of water. For droplet volumes below 1.0 mL, liquid marbles could be easily transferred onto a glass substrate,

FgH 2 4

ð1Þ

where F is the density of water (1.00 g 3 cm
3) and g is the acceleration due to gravity (9.81 m 3 s
2). This effective surface tension is similar to those obtained for liquid marbles prepared from other hydrophobic particles such as poly(vinylidene fluoride) and polyethylene particles.13 The reduction in the effective surface tension should be due to the hydrophobic PDEA
PS latex coating on the water droplet surface.14 Aussilious and Qu
er
e1b rationalized the relationship between the quasi-spherical radius (R0) of the water droplet, the capillary length (k
1), and the deviation from spherical dimensions: for R0 , k
1, quasi-spherical shape; for R0 . k
1, puddle shape. Here, R0 = (3V/4π)1/3, where V is the volume of the water droplet, and k
1 = (γ/Fg)1/2. With the γ value of 68 mN 3 m
1, k
1 is calculated to be 2.6 mm. This result suggests that a 15 μL liquid marble (R0 = 1.53 mm) should have a near-spherical shape and larger liquid marbles (>100 μL or R0 > 2.88 mm) adopt a puddle shape, which agrees well with the results shown in Figure 4 and Figure S6 (Supporting Information). By spraying water droplets on the powder bed by use of an atomist spray, liquid marbles with a mean volume of 0.087 μL were obtained (see Figure S7, Supporting Information). Dn and Dw/Dn of these liquid marbles were determined to be 0.55 mm and 1.65, respectively. This spray method is a rather facile and efficient approach for the simultaneous preparation of many small liquid marbles. In order to investigate the effect of aqueous solution pH on the relative mechanical integrity of liquid marbles, the minimum kinetic energy required to destroy them was investigated by use of the purpose-built droplet roller apparatus; see Table 2 and Figure S1 (Supporting Information). At pH 10, the liquid marbles survived with their original volume preserved when dropped from a height of 4 cm or lower. Such liquid marbles accumulated intact on the powder bed. If no powder bed was utilized (just a powder slope), water droplets broke up even when released from zero height. This is presumably because merely rolling these water droplets down the powder slope does not lead to a sufficiently thick and/or contiguous latex coating to withstand their impact with the underlying substrate. In contrast, fully coated liquid marbles did not break up even after being dropped without any powder on the bed and slope. The powder bed is essential in order to obtain stable liquid marbles. Between heights of 5 and 7 cm, some fraction of the liquid marbles bounced intact on the powder bed, while the remaining liquid marbles disintegrated, typcally dividing into multiple droplets after their first collision with the powder bed (see Table 2 for further details). Above a height of 8 cm, every liquid marble broke up and formed smaller droplets. The small “daughter” droplets generated after break-up of the “parent” liquid marble were immediately stabilized by excess dried PDEA
PS latex as the former rolled over the powder bed. The percentage of fragmenting droplets and the maximum and minimum numbers of the surviving droplets after their vertical descent increased monotonically with increasing 8071

dx.doi.org/10.1021/la201317b |Langmuir 2011, 27, 8067–8074

Langmuir

ARTICLE

Table 2. Characterization of Mechanical Integrity of Liquid Marblesa pH 2b

pH 10b

no. of water droplets after droplet roller expt height, cm rolling, % broken, %

a

max

min

no. of water droplets after droplet roller expt rolling, % broken, %

max

min

0

100

0

1

1

100

0

1

1

2

100

0

1

1

100

0

1

1

3

100

0

1

1

100

0

1

1

4

90

10

2

1

100

0

1

1

5

10

90

4

1

50

50

4

1

6

10

90

5

1

30

70

4

1

7

0

100

5

2

10

90

5

1

8 10

0 0

100 100

6 10

2 2

0 0

100 100

5 7

2 3

12

0

100

11

5

0

100

9

5

15

0

100

16

10

0

100

12

6

20

0

100

21

11

0

100

17

10

Prepared from PDEA
PS latex by use of droplet roller equipment. b Ten droplets were examined.

Figure 7. Mean lifetimes required for destabilization of 15 μL liquid marbles prepared from PDEA
PS latex after being carefully placed on the surface of liquid water at various pH values. The percentages of liquid marbles that burst and shrink are shown next to the individual data points.

height. The impinging kinetic energy per liquid marble for the series of experiments conducted at a height of either 5 or 7 cm was estimated to range from 6.4  10
6 to 8.9  10
6 J. These values are 3.5
4.8 times larger than the liquid marble surface energy (1.8  10
6 J), which is calculated by use of its effective surface tension of 68 mN 3 m
1 and the liquid marble volume of 13 μL, and a spherical shape is assumed. The liquid marbles were less stable at pH 2 compared with those prepared from an aqueous solution at pH 10: partial break-up on impact commenced at a height of 4 cm, and every liquid marble broke into smaller droplets when released above a height of 7 cm. These liquid marbles can also be transferred intact onto the planar surface of liquid water. The hydrophobic PDEA
PS particles adsorbed at the liquid marble surface prevent diffusion of water between the droplet interior and the aqueous subphase. There are two possible fates for such liquid marbles: they either burst or gradually shrink. The effect of varying the solution pH on

Table 3. Long-Term Stability of Liquid Marbles at the Planar Air
Water Interfacea addition of acid (HCl, 11.7 M, 200 μL)b,c stand still

20 s 1 h 34 min

addition of water (pH 10.0, 200 μL)b

1 h 39 min

a

These mean lifetimes are the average of 10 experiments. b Water or acid was added 1 min after placing liquid marbles at the surface of liquid water. c pH 1.1 after HCl addition.

the stability of individual liquid marbles placed at the air
water interface was examined (see Figure 7). Here the destabilization time is defined as the time required for a given liquid marble to either burst or shrink after its careful transfer onto the water surface. Enhanced stabilities (mean lifetimes of at least 1 h) were achieved above pH 8.1, which indicates that the nonprotonated PDEA
PS particles are sufficiently hydrophobic to adsorb 8072

dx.doi.org/10.1021/la201317b |Langmuir 2011, 27, 8067–8074

Langmuir

ARTICLE

Figure 8. (a, b) Digital photographs and (c) an SEM image of liquid marbles prepared from 2.0 wt % aqueous Gellan gum solution. The images were taken (a) before and (b, c) after freeze-drying.

strongly at the air
water interface under these conditions. These liquid marbles eventually either shrank (due to slow evaporation of water from their interior) or burst on the water surface. These lifetimes were significantly longer than those observed for liquid marbles prepared from poly(dimethylsiloxane)-stabilized poly(2-vinylpyridine) particles (∼12 min)2e but are somewhat shorter than those observed for liquid marbles prepared from closely related PDEA
PS particles synthesized by emulsion polymerization from a PDEA macromonomer (∼ 250 min).2c In contrast, the liquid marbles described in the present study proved to be unstable below pH 7.1, with disintegration invariably occurring within 10 min (or less). This is because protonation of the PDEA chains leads to spontaneous desorption of the latex particles from the water droplets and hence immediate disintegration of the liquid marbles, with concomitant dispersal of the latex into the acidic solution. Control experiments confirmed that the dried PDEA
PS latex did not disperse in aqueous media at pH 8.1 or above: they merely remained adsorbed at the planar air
water interface. Clearly, the minimum pH required for long-term stability (over 1 h) of these liquid marbles correlates closely with the pKa value of PDEA. PDEA
PS latex confers the expected pH-responsive behavior on the liquid marbles (Table 3), which broke up within 20 s after addition of HCl to the aqueous solution (the final solution pH was 1.1). In contrast, the same-sized liquid marbles remained stable for 99 min (average of 10 experiments) after the addition of deionized water in the corresponding control experiments. This is because the PDEA
PS particles become highly hydrophilic on protonation, leading to their desorption from the droplet interface and the rapid disintegration of the liquid marbles. It is interesting to ask whether such PDEA
PS latexes can be used to prepare liquid marbles from solvents other than water. Recently, Murakami and Bismarck15 and also Takahara and coworkers2g reported the preparation of liquid marbles from oils as the droplet phase. This was achieved by use of fluorinated polymer particles. Murakami and Bismarck suggested that minimizing the wettability of the liquid on the dry powder is important for successful liquid marble formation: if the powder repels the liquid, stable oil marbles should be obtained. In this study, we prepared liquid marbles from either aqueous Gellan gum solution or glycerol as the droplet phase. An aqueous solution of Gellan gum (15 μL, 2.0 wt %), which is a free-flowing fluid at 80 °C, was dropped onto the powder bed and liquid marbles were prepared by rolling the cooling droplet at 25 °C, whereupon the aqueous Gellan solution forms a gel. These gelled liquid marbles retained their spherical morphology even after freeze-drying. SEM studies of such a fractured liquid marble indicated flocs adsorbed at the surface of the liquid marbles (see Figure 8). Glycerol marbles were also successfully prepared from

Figure 9. Digital photographs of a glycerol-based liquid marble stabilized with PDEA
PS particles. This droplet is placed (a) on a glass substrate and (b) at the planar air
water interface.

PDEA
PS particles (Figure 9). Glycerol remains a liquid up to 290 °C (its decomposition temperature) and the glycerol marbles retained their spherical shape on the glass substrate without any deformation for at least 1 week due to minimal solvent evaporation. Moreover, these glycerol liquid marbles can also float at the surface of liquid water for at least 16 min, although the density of glycerol (1.26 g cm
3)16 is significantly higher than that of water. These results suggest the possibility of novel transportation methods for viscous fluids (the viscosities of glycerol and water are 1.5 and 0.89  10
3 Pa 3 s,16 respectively).

’ CONCLUSIONS In summary, relatively monodispersed submicrometer-sized PS latex particles with pH-responsive PDEA steric stabilizer on their surface were successfully synthesized by dispersion polymerization from commercially available starting materials. The PDEA
PS latexes can be used to prepare pH-responsive liquid marbles with aqueous volumes varied from 15 μm to 2 mL. The larger liquid marbles adopted nonspherical shapes due to gravitational forces; analysis of this deformation enabled estimation of the surface tension of 68 mN 3 m
1. These liquid marbles were stabilized by flocs of PDEA
PS latex particles, as confirmed by fluorescence microscopy and SEM studies. The relative mechanical integrity of these liquid marbles was assessed by use of a purpose-built droplet roller apparatus: the relative mechanical integrity of the liquid marbles prepared from alkaline aqueous solution (pH 10) was higher than those prepared from acidic aqueous solution (pH 2). These liquid marbles exhibited longterm stability (over 90 min) when transferred onto the surface of liquid water, provided that the solution pH of the subphase was above pH 8. In contrast, the use of acidic solutions led to immediate disintegration of these liquid marbles within 10 min, with dispersal of the PDEA
PS latex particles in the aqueous solution. Finally, stable liquid marbles were also successfully prepared from aqueous 8073

dx.doi.org/10.1021/la201317b |Langmuir 2011, 27, 8067–8074

Langmuir Gellan gum solution and glycerol. Provided that their mean lifetimes can be extended significantly by further formulation optimization, the liquid marbles described in this study are promising candidates for potential applications in various fields as smart capsules for the delivery of water-soluble actives.

’ ASSOCIATED CONTENT

bS

Supporting Information. Details on calculation of surface tension of liquid marble, and seven figures as described in the text. This material is available free of charge via the Internet at http://pubs.acs.org/.

W b

Web Enhanced Feature. Depiction of fully coated liquid marble allowed to fall onto the PET film from a height of 2 cm.

’ AUTHOR INFORMATION Corresponding Author

*E-mail [email protected].

’ ACKNOWLEDGMENT We thank Wako Chemicals (Japan) and CPKelco (USA) for kind donation of the 2,20 -azobis[2-methyl-N-(2-hydroxyethyl)propionamide] initiator and Gellan gum (Kelcogel F), respectively. Reviewers are also thanked for their fruitful comments. This work was partially supported by Core-to-Core Program (Project 18004) promoted by the Japanese Society for the Promotion of Science.

ARTICLE

Colloids Surf., B 2010, 78, 193. (c) Fujii, S.; Suzaki, M.; Nakamura, Y.; Sakai, K.; Ishida, N.; Biggs, S. Polymer 2010, 51, 6240. (7) Homola, A.; James, R. O. J. Colloid Interface Sci. 1977, 59, 123. (8) (a) Amalvy, J. I.; Wanless, E. J.; Li, Y.; Michailidou, V.; Armes, S. P.; Duccini, Y. Langmuir 2004, 20, 8992. (b) Hayashi, H.; Iijima, M.; Kataoka, K.; Nagasaki, Y. Macromolecules 2004, 37, 5389. (9) (a) Liu, S.; Weaver, J. V. M.; Save, M.; Armes., S. P. Langmuir 2002, 18, 8350. (b) Fujii, S.; Cai, Y.; Weaver, J. V. M.; Armes, S. P. J. Am. Chem. Soc. 2005, 127, 7304. (10) (a) Shen, S.; Sudol, E. D.; El-Aasser, M. S. J. Polym. Sci., Part A: Polym. Chem. 1993, 31, 1393. (b) Okubo, M.; Fujii, S.; Maenaka, H.; Minami, H. Colloid Polym. Sci. 2002, 280, 183. (11) Walz, R.; B€omer, B.; Heitz, W. Makromol. Chem. 1977, 178, 2527. (12) (a) Lascelles, S. F.; Armes, S. P.; Zhdan, P.; Greaves, S. J.; Brown, A. M.; Watts, J. F.; Leadley, S. R.; Luk, S. Y. J. Mater. Chem. 1997, 7, 1349. (b) Fujii, S.; Matsuzawa, S.; Nakamura, Y.; Ohtaka, A.; Teratani, T.; Akamatsu, K.; Tsuruoka, T.; Nawafune, H. Langmuir 2010, 26, 6230. (13) Bormashenko, E.; Pogreb, R.; Whyman, G.; Musin, A. Colloids Surf., A 2009, 351, 78. (14) Newton, M. I.; Herbertson, D. L.; Elliott, S. J.; Shirtcliffe, N. J.; McHale, G. J. Phys. D: Appl. Phys. 2007, 40, 20. (15) Murakami, R.; Bismarck, A. Adv. Funct. Mater. 2010, 20, 732. (16) Lide, D. R., Ed. Handbook of Chemistry and Physics, 89th ed.; CRC Press: Boca Raton, FL, 2008.

’ REFERENCES (1) (a) Aussilious, P.; Qu
er
e, D. Nature 2001, 411, 924. (b) Aussilious, P.; Qu
er
e, D. Proc. R. Soc. A: Math. Phys. Eng. Sci. 2006, 426, 973. (c) McHale, G.; Elliott, S. J.; Newton, M. I.; Herbertson, D. L.; Esmer, K. Langmuir 2009, 25, 529. (d) Wang, W.; Bray, C. L.; Adams, D. J.; Cooper, A. I. J. Am. Chem. Soc. 2008, 130, 11608. (e) Forny, L.; Saleh, K.; Denoyel, R.; Pezron, I. Langmuir 2010, 26, 2333. (f) Carter, B. O.; Adams, D. J.; Cooper, A. I. Green Chem. 2010, 12, 783. (g) Bhosale, P. S.; Panchagnula, M. V. Langmuir 2010, 26, 10745. (h) Inoue, M.; Fujii, S.; Nakamura, Y.; Iwasaki, Y.; Yusa, S. Polym. J. 2011in press. (i) Dandan, M.; Erbil, H. Y. Langmuir 2009, 25, 8362. (j) Bormashenko, E.; Pogreb, R.; Musin, A.; Balter, R.; Whyman, G.; Aurbach, D. Powder Technol. 2010, 203, 529. (2) (a) Bormashenko, E.; Stein, T.; Whyman, G.; Bormashenko, Y.; Pogreb, R. Langmuir 2006, 22, 9982. (b) Gao, L. C.; McCarthy, T. J. Langmuir 2007, 23, 10445. (c) Dupin, D.; Armes, S. P.; Fujii, S. J. Am. Chem. Soc. 2009, 131, 5386. (d) Bormashenko, E.; Pogreb, R.; Whyman, G.; Musin, A.; Bormashenko, Y.; Barkay, Z. Langmuir 2009, 25, 1893. (e) Fujii, S.; Kameyama, S.; Armes, S. P.; Dupin, D.; Suzaki, M.; Nakamura, Y. Soft Matter 2010, 6, 635. (f) Tian, J.; Arbatan, T.; Li, X.; Shen, W. Chem. Commun. 2010, 46, 4734. (g) Matsukuma, D.; Watanabe, H.; Yamaguchi, H.; Takahara, A. Langmuir 2011, 27, 1269. (3) (a) Binks, B. P.; Murakami, R. Nat. Mater. 2006, 5, 865. (b) Binks, B. P.; Duncumb, B.; Murakami, R. Langmuir 2007, 23, 9143. (c) Fujii, S.; Murakami, R. KONA Powder Particle J. 2008, 26, 153. (d) McHale, G.; Newton, M. I. Soft Matter 2011, DOI: 10.1039/ C1SM05066D. (4) (a) Dampeirou, C. WO Patent 034917, 2005. (b) Lahanas, K. M. U.S. Patent 6290941, 2001. (5) B€ut€un, V.; Armes, S. P.; Billingham, N. C. Polymer 2001, 42, 5993. (6) (a) Fujii, S.; Kakigi, Y.; Suzaki, M.; Yusa, S.; Muraoka, M.; Nakamura, Y. J. Polym. Sci., Part A: Polym. Chem. 2009, 47, 3431. (b) Pi, M.; Yang, T.; Yuan, J.; Fujii, S.; Kakigi, Y.; Nakamura, Y.; Cheng, S. 8074

dx.doi.org/10.1021/la201317b |Langmuir 2011, 27, 8067–8074