Controlling Self Assembly and Topology at Micro-Nano Length-scales

induced flow in a droplet deployed on a hydrophilic substrate leads to the formation of the famous coffee-ring pattern 15. Alternatively ..... This le...
1 downloads 5 Views 3MB Size
Subscriber access provided by UNIV OF DURHAM

Interface Components: Nanoparticles, Colloids, Emulsions, Surfactants, Proteins, Polymers

Controlling Self Assembly and Topology at Micro-Nano Lengthscales using Contact Free Mixed Nanocolloid Droplet Architecture Lijun T Raju, Shubhankar Chakraborty, Binita Pathak, and Saptarshi Basu Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.8b00790 • Publication Date (Web): 16 Apr 2018 Downloaded from http://pubs.acs.org on April 16, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Controlling Self Assembly and Topology at Micro-Nano Length-scales using Contact Free Mixed Nanocolloid Droplet Architecture Lijun T. Rajua, Shubhankar Chakrabortya, Binita Pathaka, Saptarshi Basua* a

Department of Mechanical Engineering, Indian Institute of Science Bangalore, Bangalore, Karnataka − 560012, India

Abstract Spatially varying ordering of colloids of multiple sizes at micro-nano scales finds application in different industrial processes including manufacturing of photonic crystals. In this work, we showcase a unique physics based architecture through which we have been able to control the morphology of the precipitates evolving out of the drying of a contact free droplet at micro to nano length-scales. We show that by varying the relative concentration of the larger sized colloids, one can modulate evaporation, subsequent particle transport and particle ordering at the droplet interface, thereby controlling the rates of certain instabilities like buckling. In this way, we have produced Evaporation Induced Self Assembly (EISA) structures (devoid of any substrate effect) with striking topological and surface features. Furthermore, we proved that these instabilities can be further tuned using measured amount of external heating through alteration of the evaporation rates. Notwithstanding, we also quantified that the ordering of the mixed colloids varies, in a spatial sense, across the droplet surface, exhibiting unique patterns, porosity and lattice arrangements, all at the nanoscale. The results assure that fine tuning of macroscale parameters like heating rate and particle loading can be used to fine tune micro-nanoscale features in a droplet based high throughput bottom up framework. INTRODUCTION

Droplets of colloidal suspension form an integral part of many industrial applications such as pharmaceutics, food, drugs and surface coating, among others 1,2,3,4. Evaporation of the liquid phase and self-assembly of particles produce final structures with different morphologies. The properties of these structures can be modified by manipulating the drying characteristics of the functional droplets. Such structural modulation is particularly crucial to applications

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

such as DNA chips, photonic crystals, semiconductors and surface patterning, to name a few 5,6,7,8,9,10,11,12

. Although droplet-based architecture is a simplistic and inexpensive fabrication

technique, it involves complex interfacial transport phenomena and interparticle dynamics. Several numerical and experimental studies have been conducted in the past decade to investigate the dynamics of droplet drying13,14,15,16. Many droplet drying investigations are carried out for sessile droplet configuration in which the evaporation of liquid phase leads to the formation of intricate patterns on the substrate15,17. These deposition patterns can be altered by using external perturbations such as heat and electric charge18,19. Other methods to alter the residual patterns involve variation in initial composition, type, size and shape of particles suspended in the droplet

20

. Some recent experiments have also demonstrated the

possibility of using a reactive nanofluid droplet (as opposed to chemically inert systems), with nanoparticles of different shapes and sizes, to obtain varied surface patterns21,22. Morphologies of the final structure also changes with the properties of the substrate 23. The substrate properties (such as hydrophobicity) alters the internal flow dynamics which subsequently affects the arrangement of particles inside the droplet. For instance, evaporation induced flow in a droplet deployed on a hydrophilic substrate leads to the formation of the famous coffee-ring pattern

15

. Alternatively, a buoyancy driven flow field is observed in a

droplet placed on a hydrophobic substrate which leads to the formation of a porous shell at 24,25

the droplet surface

. The shell is formed mainly due to collision and agglomeration of

particles. The visco-elastic shell eventually undergoes a sol-gel transition and leads to structural instabilities in the droplet. Buckling is the most commonly observed geometric instability in such drying colloidal droplets. The onset of buckling is marked by small indentations on the droplet surface which grows and lead to the formation of striking topological features.

The phenomenon of buckling has been previously explored in sprays as well as in a single droplet mode

23,24,26

. The buckling dynamics can be controlled by varying properties of the

particles and initial concentration in the colloidal suspension. Buckling usually occurs due to capillary pressure developed due to evaporation of liquid through the porous shell. Therefore, manipulation of the evaporation rate or the pore size alters the dynamics of the buckling phenomena. In this study, we provide insights into the drying characteristics of a droplet of a colloidal suspension of multiple sized particles in a contact-free environment. Previously, we have explored the morphological changes in a droplet of colloids with surfactants which was levitated using a standing acoustic wave 27,28. It was shown that the addition of different types ACS Paragon Plus Environment

Page 2 of 30

Page 3 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

of particles induces inhomogeneity in shell and leads to the formation of final structures which are drastically different from the droplet of a single colloidal suspension27,28. In the current work, we intend to investigate the effect of adding similar particles of two different sizes (binary mixture of colloids). Inclusion of nanoparticles of similar surface chemistry (and shape), but with disparate sizes, lead to varied arrangement of agglomerates inside the droplet. The phenomena of buckling which is prominent in a single colloidal system can be suppressed completely by varying the composition of larger particles in the binary mixture. We incorporate polystyrene particles of two different sizes (200 nm and 860 nm) in various initial concentrations in the droplet. Apart from buckling modulation, relative concentrations of the particles lead to striking spatial ordering at the micro-nanoscale as well as sizedependent segregation of particles for a range of different cases. Voronoi reconstruction performed on the final structure reveals the distinct arrangement of particles at the air-liquid interface. The size-based segregation can be utilized to induce differential functional properties to the final structures, which is crucial to industrial applications such as fabrication of multi-layered pharmaceutical products29, photonic crystals with controlled morphologies or for creation of encapsulated systems, for use in textile, food, cosmetic and agricultural industries30. Therefore, the methodology proposed in this paper can be scaled up as a unique high throughput fabrication technique for such industrial applications. EXPERIMENTAL METHODS

In the present study, colloidal droplets of initial diameter ~ 415 ± 20  were suspended around the nodes of an ultrasonic levitator (Tec5, 100 kHz, 154 dB). The droplets (less than 1mm diameter) are deployed into the pressure nodes of the levitator by the standard technique of putting a metered liquid drop on the face of the transducer from a syringe/pipette and allowing the dispersion to atomise. The atomised dispersion gets collected near the pressure nodes forming droplets. Thereafter, the solvent in the levitated droplet was allowed to evaporate, with and without external thermal irradiation (irradiation intensities of I = 0.16 MW m-2 and 0.0 MW m-2 respectively). A tunable CO2 laser (Synrad 48, beam diameter of 3.5 mm and wavelength of 10.1 µm) was employed for the purpose of externally heating the droplet at I = 0.16 MW m-2. A high speed camera (SA5 coupled with a 5x zoom lens, circular field of view of ~885µm diameter, resolution- 0.9µm/pixel) was used at 1000 fps (for I = 0.0 MWm-2) and 250fps (for I = 0.16 MW m-2) to monitor the dynamics of the drying droplets. The temperature was measured at the droplet surface using an infrared camera (FLIR SC5200 at 500 fps, field of view ~ 2000µm x 1600µm, resolution –12.4µm/pixel) which was

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

synchronized with the high speed camera and the laser using a delay generator. The set-up is demonstrated schematically in Figure 1.

We used colloidal dispersions (1% by weight) of polystyrene particles (Thermo Scientific™ Fluoro-Max Red Aqueous Fluorescent Particles) of similar surface chemistry but two different sizes (diameters): 200 nm and 860 nm respectively. Aqueous suspension of the two particles were mixed at different proportions or volumetric particle number fractions, defined as , , 



(1)

  

where N860 and N200 are the number of particles of diameter 860nm and 200nm respectively, per unit volume of the dispersion. We used Nf,860,vol = 0.0, 0.01, 0.05, 0.11, 0.50 and 1.0 for our studies.

Figure 1. Experimental Setup

ACS Paragon Plus Environment

Page 5 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

RESULTS AND DISCUSSION Global Evaporation Timescales Drying of a colloidal droplet is a complex interplay of several thermophysical processes like inter-particle collision, agglomeration, buckling, and precipitation which leads to different morphological structures. The initial drying phase of a colloidal droplet is almost similar to that of a pure fluid which shows a continuous decay in size of the droplet (Figure 2). The rate of evaporation is enhanced with an increase in the laser heating intensity (t~180s for I = 0 MWm-2 and t~15s for I = 0.16 MWm-2). For the case of I = 0 MWm-2, the diameter  



regression curves (Figure 2) follow a linear   −  ∗  relation (where d and d0 are the 



instantaneous and initial droplet diameters respectively, t is the time which is normalized by a characteristic time scale,∗ ), which is a characteristic of diffusion-limited evaporation of acoustically levitated droplets as shown by Yarin et. al31. Yarin et. al31 determined the surface

averaged Sherwood Number, Sh, for an acoustically levitated droplet to be #$

ℎ  !" & #%

where , 

-.

'

(2)

()*+

/0 1

, is the gas particle velocity amplitude, Aoe is the effective pressure

amplitude of the incident acoustic field, 23 is the unperturbed gas density, c0 is the sound velocity in the gaseous medium, 4 is the angular frequency of the levitator, Dm is the mass

diffusion coefficient for water vapour in air.

Aoe is related to SPL (Sound Pressure Level, in dB) by the relation31 56  20 log< (> ? ) + 74 The vaporisation rate of a liquid droplet is given by

(3) 32

HI

D  ρ 4πr    2 K L 2M NO ℎ PQ(1 + ,S ) HJ

(4)

where r is the droplet radius, 2 is the liquid density and BM is the Spalding mass transfer number, defined as ,S 

TUV WTXV

(5)

ef g  2  KL h  ie h

jUklm 

+ 4KL  ℎ nn opnqf − pr s + 2 4KL 

f

ℎ  M

(8)

where Ieff is the effective intensity of the laser irradiation on the droplet, Aproj is the projection of the irradiated droplet surface to a plane normal to the laser, Tsurf and T∞ are the temperatures at near field and far away (ambient) from the surface of drop respectively, cp is the specific heat of the liquid, hloss is the heat transfer coefficient corresponding to heat loss from droplet surface to surrounding, hfg is the latent heat of vaporization. The droplet surface temperature attains a steady state (wet bulb limit) at ~0.4s, which corresponds to only ~2% of the droplet lifetime (~15s). Hence, we can take

jUklm 

 0 for most part of the evaporation

process. Comparing the first and last terms in the energy balance, we get d? ∗ KL  ~ 2 4KL 

f 

~ d? /(4 2 ℎM )

u ~

.mm

/Y vmw

∗ 

ℎ  M

(9) (10)



#/Y vmw .mm

f

(11) u

(12)

Thus, we get a linear d~t relation when the amount of energy incident on the drop is the limiting phenomenon for evaporation.

ACS Paragon Plus Environment

Page 7 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Hence, from Figure 2, we can conclude that the case of I = 0.0 MWm-2 corresponds to the diffusion-limited evaporation process while the case of I = 0.16 MWm-2 corresponds to irradiation limited evaporation. However, as mentioned earlier, the presence of colloidal particles seems to slightly alter the exact timescale when the diameter regression freezes (shell formation). This leads to a small scatter in the data (Figure 2).

Figure 2. (a) Variation of square of normalized diameter of droplet (normalized by initial diameter) with normalized time (normalized by characteristic time ∗ scale for mass diffusion limited evaporation in acoustic levitator) (b) Variation of normalized diameter of droplet (normalized by initial diameter) with normalized time (normalized by characteristic time scale for irradiation limited evaporation) tII*. Representative shadowgraphy images of

droplet diameter regression for Nf,vol,860=0.05, at (c) t/∗ = 0, (d) t/∗ = 0.45, (e) t/∗ = 0.7.

Final agglomerate structure for (f) Nf,vol,860=0.0 (buckling with deep indentaion), (g)Nf,vol,860=0.11 (buckling with shallow indentaion), (h) Nf,vol,860=1.0 (no buckling) (scale bar 100 μm)

ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Global Precipitation Dynamics and Buckling Evaporation of liquid aggravates collision among the particles leading to the formation of agglomerates. Depending on the nature of evaporation (natural drying or laser irradiated) and the initial composition of the colloidal droplet (varying values of Nf, 860,vol), a variety of final structures with interesting morphologies were obtained, as shown in Figure 3 (the SEM scanning electron microscopy, images of the final structures). For volumetric number ratio

Nf,860,vol = 0 (only smaller particles present in the system) severe buckling is observed, irrespective of the heating conditions. The tendency of buckling tends to decrease as the number of larger particles is increased. This trend is similar for both heated and non-heated cases.

ACS Paragon Plus Environment

Page 8 of 30

Page 9 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 3. Different structures of agglomerates under

SEM

(~1500X

magnification),

showcasing the different buckling regimes. The formation of these varying structures is attributed to the formation of a visco-elastic, nanoporous shell at the droplet surface and the subsequent onset and growth of buckling instability. Such a viscoelastic shell is formed at the surface of an evaporating colloidal droplet whenever ample time is not available for radial homogenization of particleconcentration throughout the droplet volume (diffusion time scale >> evaporation time scale) as evaporation proceeds. In other words, whenever Peclet Number (defined as the ratio of diffusion timescale to the convection timescale) is much higher than unity, an accumulation of particles occur at the liquid-air interface forming the above mentioned shell. Peclet Number Estimation: In our case, tevap = l/vevap is the convection timescale, where vevap is the rate of droplet surface regression and l = rdrop is the characteristic length-scale, chosen to be equal to the initial radius of the droplet. Thus, ACS Paragon Plus Environment

Langmuir 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

5x 

yzmm

.V{|



(  /*)

( / .V{| )



.V{| *



fyl}| yl}| *

Page 10 of 30



(13)

where, D is the diffusion coefficient, estimated using the Stokes-Einstein relation N

~ j

(14)

%€|{lz‚Y.

where kB is the Boltzmann constant, T is the average temperature of the solvent in Kelvins, µ is the viscosity of the solvent (water) and dparticle is the particle diameter. Peclet number for four different cases are shown in Table 1. The high values of Peclet Number clearly indicate that the formation of shell near the droplet surface is highly favoured in the experiments reported here. Table 1. Peclet Number (Pe) Heat Supplied (MWm-2) Particle Diameter Only 200 nm Only 860 nm

0.00

0.16

~10

~10h

~4 × 10

~4 × 10h

The shell so formed is however inhomogeneous (minimum thickness at the poles) as a result of the flow field inside droplet (due to acoustic streaming)34. It should be noted here that the flow field due to acoustic streaming was observed to consist of concentric circular streamlines.35 We conjecture that since the streamlines are circular, it doesn’t lead to any radial re-homogenization or mixing, and thereby doesn’t offer any impedance to shell formation. However, a variation in shell thickness indicates that the flow field leads to inhomogeneities in the polar / zenith (θ) direction. Evaporation of liquid through the pores of the shell forms nano-menisci in the interstitial spaces. The capillary pressure developed at these menisci triggers buckling instability in this shell having non-uniform thickness. Buckling is initiated as the capillary pressure (Pcap) exceeds the threshold value (Pth). The capillary pressure depends upon the permeability of the shell (k), rate of evaporation of the liquid phase through the porous shell (J), viscosity of the liquid phase (µ) and the shell thickness (h) which is given as: Pcap ∼µJh /k

(15)

ACS Paragon Plus Environment

Page 11 of 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

The permeability (k) depends upon the porosity (or packing fraction) as well as the size of particles forming the porous shell. Similarly, thickness of the shell also depends on the

packing of the particles in the shell (quantified later). The critical buckling pressure (Pcrit) depends on the shell thickness (h), the Young’s Modulus (Y), and the radius of the shell (Rcav), given as: Pcrit ~ 4 Y (h / Rcav)2

(16)

The ratio of the two pressures is given by …‚{|

…‚lz

~ 

 €†‡‚{V

#T~




Q’  1 +