Coassembly Kinetics of Graphene Oxide and Block Copolymers at the

Aug 16, 2017 - Beijing Advanced Innovation Center for Soft Matter Science and Engineering, Beijing University of Chemical Technology, Beijing 100029, ...
0 downloads 14 Views 3MB Size
Download PDF
Subscriber access provided by UNIVERSITY OF ADELAIDE LIBRARIES

Article

Co-Assembly Kinetics of Graphene Oxide and Block Copolymers at the Water/Oil Interface Dayong Chen, Zhiwei Sun, Thomas P. Russell, and Lihua Jin Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b02009 • Publication Date (Web): 16 Aug 2017 Downloaded from http://pubs.acs.org on August 17, 2017

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 free 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 accessible to all readers and 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.

Langmuir 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 33

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

Co-Assembly Kinetics of Graphene Oxide and Block Copolymers at the Water/Oil Interface

Dayong Chen1,*, Zhiwei Sun2, Thomas P. Russell2,3,4, Lihua Jin1 1Department

of Mechanical and Aerospace Engineering,

University of California, Los Angeles 2Department

of Polymer Science and Engineering,

University of Massachusetts, Amherst 3Beijing

Advanced Innovation Center for Soft Matter Science and Engineering, Beijing University of Chemical Technology, Beijing 100029, China

4

WPI—Advanced Institute for Materials Research (WPI-AIMR) Tohoku University, 2-1-1 Katahira, Aoba Sendai 980-8577, Japan *Corresponds to [email protected]

Abstract The co-assembly kinetics of graphene oxide (GO) nanosheets and diblock copolymers at the water/toluene interface is probed by tracking the dynamic interfacial tension using pendant drop tensiometry. The diblock copolymer significantly enhances the surfactancy of the GO nanosheets at the interface. It is found that diblock copolymers rapidly adsorb to the water/toluene interface and enhance the adsorption affinity of GO nanosheets to the interface. The continuous adsorption of GO at the interface leads to a random loose packing state, at which the adsorbed GO and diblock copolymers start to form an elastic film. After this transition, GO continues to adsorb to the interface, however, at a much slower speed, yielding a more solid-like elastic film in the long time equilibrium limit. Key words: graphene oxide, interfacial assembly, kinetics

1

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

Introduction Graphene oxide (GO) is the product of chemical exfoliation of graphite and has been known for more than a century.1 A recent surging interest is to use GO as a precursor for mass-production of graphene (or reduced GO) for advanced applications including energy storage, separation membranes, flexible electronics, and bio-devices.1 Essentially a graphene oxide particle is a graphene sheet derivatized by carboxylic acid at the edges, and phenol hydroxyl and epoxide groups mainly on the basal plane. Therefore, a GO particle is mostly hydrophilic on the edges, and contains hydrophobic islands on the basal plane in less or non-oxidized regions.2 This attribute allows GO to be used as a surfactant to stabilize water/air interface and water/oil interfaces.2, 3, 4, 5, 6, 7, 8, 9, 10 GO has been used for preparing ultra-stable Pickering emulsions11, oil recovery6, 12, and for dispersing carbon materials3. GO has also been used as a surfactant in miniemulsion polymerization systems.13, 14, 15 Due to the nature of hydrophilic edges and a hydrophobic basal plane, the surfactancy of a GO particle strongly depends on its particle size.3 Larger GO particles have been found to have a stronger propensity to adsorb at the water/air interface, while smaller GO particles, being more hydrophilic, remain suspended in the aqueous phase.3 One method to enhance the surfactancy of GO is to use an end-functionalized polymer or an amphiphilic diblock copolymer, in which the hydrophilic block can interact favorably with GO at the interface, while the hydrophobic block can dissolve in the oil phase. Previously, Sun et al. demonstrated that graphene oxide can be trapped at water/oil interfaces and jammed into a solid thin film using both an end-functionalized polymer or a diblock copolymer, polystyrene-b-poly(2vinylpyridine) (PS-b-P2VP), as a ligand.16 However, the assembly kinetics of GO and diblock copolymer at the water/oil interface has not been carefully analyzed yet. Here, we provide a detailed mechanistic understanding of the co-assembly kinetics.

Experiment Graphene oxide was obtained from the Graphene Supermarket in a concentrated aqueous solution (5 g/L) with an average size of 500 to 700 nm. PS-b-P2VP with a molecular weight of Mn = 19,500 (polystyrene block: 16,000 and poly(2-vinylpyridine:

2

ACS Paragon Plus Environment

Page 2 of 33

Page 3 of 33

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

3,500) and polydispersity index (PDI) of 1.05 was obtained from Polymer Source and used without further purification. Anhydrous toluene from Sigma-Aldrich was used as received. Aqueous GO solutions of various concentrations were prepared from the concentrated GO stock solution and sonicated for 5 min before use. An interface was made by injecting a pendant drop of DI water or the aqueous GO dispersion into a toluene solution containing PS-b-P2VP block copolymers. As illustrated in Figure 1, both the assembly process of the block copolymers at the toluene-water interface, and the coassembly of the block copolymer and GO were recorded by monitoring the interfacial tension change with time. The interfacial tension between water and the organic solvent was measured at

with a tensiometer (Dataphysics OCA 15 plus) by the pendant

drop method. The volume of the droplets was controlled to be 10 µL.

Results and Discussions To enhance the surfactancy of hydrophilic GO Sun et al. developed a simple route by adding an end-functionalized homopolymer or a diblock copolymer, PS-b-P2VP, that interacts favorably with the hydrophilic parts of GO.

16

It was found that the GO can be

effectively trapped at the toluene/water interface and jammed into a solid thin film by electrostatic interactions or hydrogen bonding with a quaternized amine unit. The chemical structures of graphene oxide and PS-b-P2VP, and the interaction between carboxyl groups in GO and the pyridine unit of PS-b-P2VP are shown in Figure 1(a). The pendant drop tensiometry experiment is schematically illustrated in Figure 1(b). Specifically, a pendant drop of the aqueous GO solution was suspended in a toluene solution containing PS-b-P2VP. Both GO nanosheets and PS-b-P2VP molecules diffused to the interface and interacted at the interface. The assembly process was recorded by monitoring the interfacial tension change with time. Figure 2(a) shows the dynamic interfacial tension as different concentrations of GO were introduced into the aqueous phase. When there was no PS-b-P2VP in the toluene phase, GO alone only slightly reduced the interfacial tension between toluene and water, showing that the GO was not an effective surfactant for the toluene/water interface. When a water droplet without GO

3

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 33

was suspended in a toluene solution of PS-b-P2VP, a rapid reduction in the interfacial tension was observed, showing that this diblock copolymer is an effective surfactant for the toluene/water interface. When GO is dispersed in the aqueous droplet suspended in a toluene solution of PS-b-P2VP, a similarly rapid reduction in the interfacial tension was observed. As the GO concentration increased, the apparent equilibrium value of interfacial tension also increased and the equilibrium value was reached at an even shorter time. In the long time equilibrium, the adsorbed GO and PS-b-P2VP diblock copolymer at the interface formed a solid film that buckled when the volume of the droplet was decreased, decreasing the interfacial area, and compressing the assembly at the interface. (Figure 2(b)). To study the assembly kinetics, Sun et al. used the buckling of the interfacial film upon compression to determine the surface coverage ratio.16 The coverage of GO on the droplet surface was found to increase with time during assembly and could reach as high as 90% coverage at equilibrium. They showed the assembled film can withstand large shrinkage and remain intact even when the droplet was completely withdrawn back into the needle.16 TEM characterization of the assembled film showed GO sheets oriented parallel to the oil/ water interface, forming a mosaic or tiling across the entire interface. The thickness of the GO film was not uniform. Most parts of the assembly are a single layer of GO, though there were some areas where there was an overlap of the GO sheets. This morphology is similar to that of GO films formed when GO nanosheets are compressed on a basic subphase.5 These results show that co-assembly of GO and diblock copolymer at the water/oil interface can provide a simple route to produce large area of GO films. However the co-assembly kinetics is not fully understood. Sun at al. have used a few different types of block copolymers, including polystyrene-b-poly(4vinylpyridine), poly(styrene-r-2-vinylpyridine), and polystyrene-b-poly(2-vinylpyridine) of different molecular weight and compositions.

16

In the following analysis of the

assembly kinetics, we focus on the experimental results for PS-b-P2VP with a molecular weight of Mn = 19,500 (polystyrene block: 16,000 and poly(2-vinylpyridine: 3,500) and polydispersity index (PDI) of 1.05. However, the analysis is general, and can be applied to other systems containing different block copolymers or random copolymers, as long as the concentration of polymer is below the critical micelle concentration (CMC).

4

ACS Paragon Plus Environment

Page 5 of 33

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

Self-assembly of block copolymers at the water/oil interface We first consider the self-assembly of diblock copolymers at the water/oil interface. Amphiphilic diblock copolymers have long been used as giant surfactant molecules to stabilize oil and water interface,17 leading to the formation of ultra stable oil-in-water or water-in-oil emulsions, and giant polymersome vesicles.18, 19, 20 When a selective solvent, a good solvent for only one block, is used, very stable micelles can form.21, 22 In the experiments of Sun et al. PS-b-P2VP was dissolved in toluene as the oil phase.16 Toluene is a marginally selective solvent for the polystyrene block.23, 24 For PS-b-P2VP (with M = 19,500 , PS: 16,000 and P2VP: 3,500), the range of copolymer concentrations (0.1 n

.3 g/L) studied is lower than the CMC.23 Therefore only PS-b-P2VP unimers are present in the oil phase and the hydrodynamic radius of the PS-b-P2VP molecule is estimated to be Rh ~ 4 nm .23

For molecules of this size, the bulk diffusion constant D can be

estimated from the Einstein-Stokes equation D = k b T / 6πη R h ≈ 1 × 10 −10 m 2 /s , where k T = 4.11 × 10−21 J is the thermal energy, and η = 0.56 mPa⋅s is the viscosity of toluene at b

. Figure 3(a) shows that when a pendant drop of water of volume 10 µL was suspended in the toluene solution of PS-b-P2VP, within 1 s the interfacial tension decreased significantly and within several hundreds of seconds, the equilibrium interfacial tension was achieved. As the bulk concentration of PS-b-P2VP increased, the equilibrium interfacial tension decreased. The adsorption kinetics of PS-b-P2VP at the toluene/water interface can be a diffusion controlled process, where the diffusion of the PS-b-P2VP molecules to the interface is the rate limiting step.25 The adsorption can also be kinetically limited or governed by a mixed diffusion-kinetic mechanism, where the adsorption of PS-b-P2VP from the subsurface (an imaginary plane, a few molecular diameters below the interface.) to the interface can be hindered by an adsorption energy barrier or the molecules have to rearrange to the right configuration to adsorb to the interface.25 First, we consider a simple diffusion controlled process by applying the Ward-Tordai equation:26

Γ(t ) = 2c p

Dt D t − 2 ∫ csd π π 0

5

( t − τ ),

ACS Paragon Plus Environment

(1)

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 6 of 33

where Γ(t ) is the PS-b-P2VP surface coverage at the interface, cp is the bulk PS-b-P2VP concentration, c s is the concentration of PS-b-P2VP in the subsurface, D is the diffusion

coefficient of PS-b-P2VP unimer, and τ is a dummy variable of integration.26 This equation can only have analytical solutions in the two limiting cases: the initial adsorption stage, and the close to the equilibrium adsorption stage.25 For the initial stage, the adsorption of PS-b-P2VP to the interface is dilute, so the linear Henry isotherm can be applied to relate surface tension with the surface coverage:

γ (t)− γ 0 = −RTΓ(t ). t→0

(2)

Combining the Henry isotherm with Equation 1 leads to

γ t→0 (t ) = γ 0 − 2RTc p

Dt π

(3)

In line with Equation 3, we re-plot the dynamic interfacial tension as a function of t 1/2 in Figure 3(b). We note the missing data at the very early stage of adsorption. For a toluene/water interface without any PS-b-P2VP molecules present, the interfacial tension is γ 0 = 32 mN/m (Figure 2). The value of neat water/toluene interface energy is slightly lower than the reported value (36 mN/m) in literature,27 due, possibly, to impurities. Pendant drop tensiometry is known to capture dynamic surface tension from a few seconds.25, 28 From the injection of the water drop to the capture of the first data points, the waiting time is ~1-2 s. We can obtain the diffusion constant by measuring the slope of curves in Figure 3b at t = 0 . The diffusion constant D is estimated to be 1×10−9 −1×10−8 m2 /s , which is one to two orders of magnitude higher than that expected from the Einstein-Stokes equation. This apparent ‘superdiffusivity’ is a result of a convective flow. Such flow can be induced at the initial time of the drop formation.28, 29 The experimental difficulties to quantitatively analyze data from the pendant drop tensiometry have been pointed out by other researchers.29, 30 For these reasons, fitting the data using Equation 3 can be misleading.

6

ACS Paragon Plus Environment

Page 7 of 33

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

Next we analyze the experimental data at the close to the equilibrium adsorption stage. For this stage, the surfactant surface coverage can be related to the surface energy by the Gibbs adsorption isotherm:

1 dγ Γ=− ⋅ RT dlnc p

(4)

For this adsorption stage, the concentration in the subsurface ( c s ) can be given by25:



π cs = cp − Γ 4Dt

(5)

Combining Equation 4, Equation 5 and Equation 1 at the long time limit leads to31

RTΓ 2eq

γ t→∞ (t) = γ eq + cp

π 4Dt

(6)

In Figure 4(a), interfacial tension γ is plotted against t −1/2 . For each curve, a linear fit was performed as t −1/2 approached 0 and γ approached γ eq . The intercept is γ eq and the slope is

1 RTΓ 2eq π /4D . cp

In Figure 4(b), we plot γ eq as a function of polymer

concentration log(c p ). Following Eastoe25 and Xu32, we use an empirical quadratic fit for γ eq vs. log(c p ). Using the Gibbs adsorption isotherm (Equation 4), we can obtain the equilibrium surface coverage Γ eq as a function of the bulk concentration cp , as shown in the inset of Figure 4(b). Given the slope of each fitted line in Figure 4(a), the value of surface coverage Γ eq , and the bulk concentration cp , we can now calculate the diffusion constant D ~ 1 × 10 − 14 m 2 /s , which is 4 orders of magnitude smaller than predicted by the Einstein-Stokes equation. This indicates that at the close to the equilibrium adsorption stage, a diffusion-controlled mechanism is inadequate to explain the observations. A kinetically limited adsorption or a mixed diffusion-activation process is expected, since the surface is crowded with PS-b-P2VP molecules. For more PS-b-P2VP to adsorb to the

7

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 8 of 33

interface, the molecules must overcome an energy barrier related to the rearrangement of PS-b-P2VP molecules at the interface or a change in the configuration of PS-b-P2VP molecules at the interface. If we consider the calculated diffusion constant as an effective diffusion constant given by a mixed diffusion-activation mechanism, the effective diffusion coefficient Deff = 1 × 10−14 m 2 /s takes into account this activation barrier, and is related to the physical diffusion coefficient D = 1 × 10 − 10 m 2 /s by an Arrhenius-type relationship25:

Deff = D⋅e − ∆G/RT

(7)

The energy barrier ∆G is calculated to be 22.4 KJ/mol, which is plausible given the high molecular weight of PS-b-P2VP in use.25 We do note that this analysis is oversimplified. A kinetically limited adsorption relating the surface coverage of surfactant molecules and the contact time has been resolved.33,

34, 35, 36

However, lacking the relationship

between the dynamic surface tension and the surface coverage prevents us from fitting the results here. We note that while Equation (3) and Equation (6) are derived for a planar interface, they can be used in our current experiments for the following reason. The equilibrium surface coverage Γ eq and the bulk block copolymer concentration c p together define a

4 depletion length hp according to 4πr 2Γ eq = πc p (r + hp )3 − r 3  , where r is the radius of 3 the drop.37 Using results from Figure 4b, we find that hp is on the order of 0.01 mm, two orders of magnitude lower than the radius r ≈ 1.3 mm for a drop of 10 µL. Therefore the influence of interface curvature is minimal.

Co-assembly of Graphene Oxide and Block Copolymer at the toluene/Water Interface We now consider the co-assembly of graphene oxide and diblock copolymer at the water/oil interface. First, graphene oxide nanosheets used in the experiments have an average size of L = 500− 700 nm and a thickness of d ≈ 1 nm . Such 2D nanosheets are

8

ACS Paragon Plus Environment

Page 9 of 33

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

found to have a hydrodynamic radius RGO = (5.9 ± 2.2)× L2/3 .38 Therefore, the hydrodynamic radius of graphene oxide particles is estimated to be RGO ≈ 400 nm . We note that this value is significantly higher than the effective radius of an approximated sphere. For non-spherical particles, the effective hydrodynamic radius, reff , is often approximated as the radius of a sphere of volume equal to the volume of the particle. Then, approximating the nanosheets as discs with thickness d and length (diameter) L allows us to write reff = (3/16)1/3 d 1/3L2/3 ≈ 40 nm . This large discrepancy is believed to be due to different bending stiffness of the particles.39 The diffusion constant of GO ( R ≈ 400 nm ) in GO

water is therefore expected to be DGO = 5× 10−13 m2 /s .

We first consider the interfacial c0-assembly of PS-b-P2VP and GO for a fixed PS-bP2VP bulk concentration of 0.1 g/L, while the GO bulk concentration is changed from 0 to 0.3 g/L. The dynamic interfacial tension is plotted against the contact time on a semilog scale in Figure 5(a). It is evident that when the GO concentration is lower than 0.08 g/L, the dynamic interfacial tension decays following the same trend as that when there is only PS-b-P2VP adsorbed at the interface. When the bulk concentration of GO is above 0.09 g/L, the dynamic interfacial tension decays initially following the same trend, however it exhibits a sharp transition at a certain point, above which the interfacial tension barely changes with time. The transition time decreases as the GO concentration is increased. Therefore, a higher GO concentration seems to lead to an earlier transition, and therefore a higher apparent equilibrium surface tension. The sudden change in slope indicates some critical phenomenon. For the pendant drop tensiometry, the shape of the drop is fitted with the Young-Laplace equation to yield the surface tension.28 In the Young-Laplace equation, only pressure, surface tension and surface curvatures are considered and the shape of the interface is assumed to be Laplacian (i.e. the interface is assume to be fluidic). Prior to this transition, the GO nanosheets are sparsely absorbed at the oil and water interface, the evolution of the shape of the drop and consequently the interfacial tension is mainly determined by the adsorption of PS-b-P2VP, and is barely affected by the adsorption of GO nanosheets. When GO nanosheets interfacial coverage reaches a certain critical value, their edges start to overlap, and an elastic film will begin to form at the interface.5 This transition 9

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 10 of 33

point corresponds to the onset of an elastic interface. The existence of an elastic film at the interface has been confirmed in the later stage by experiments, as shown in Figure 2 and TEM images by Sun et al.16 Once an elastic film has formed, further evolution of the shape of the drop will cause the elastic film to be locally stretched or compressed. Therefore, further deformation of the drop is penalized and the Young-Laplace analysis of the pendant drop fails to track any further adsorption of PS-b-P2VP molecules and GO nanosheets at the interface.40 The formation of an elastic film is due to the jamming of randomly adsorbed graphene nanosheets at the toluene/water interface when a critical areal fraction is reached. For the system with GO concentration of 0.12 g/L GO and PS-b-P2VP concentration of 0.1 g/L, the critical transition time is ~60 s from Figure 5 (a).

Figure 6(a) shows the surface fraction of GO coverage as a function of the contact time. At the critical transition time, we can find the critical areal fraction Ac ≈ 0.65 . The interfacial coverage of GO ( A ) was determined by the interfacial buckling method. Specifically, to induce an interfacial buckling, a pendant drop of aqueous GO solution was first injected into the toluene phase, and subsequently the drop volume was reduced after certain time. The state prior to the volume reduction was the “free” adsorption state. Then, as the volume of the drop was reduced, the GO jammed and close-packed, forming a buckled film at the interface. Therefore, the coverage ( A ) of GO on the drop surface in the “free” state could be estimated from the ratio of the interfacial area when the buckling was observed to that in the “free state”. This areal coverage can be used to approximate the relative interfacial coverage as in the Langmuir adsorption isotherm: . We note after the critical point Ac , the adsorption of GO doesn’t stop. It A ≈ Γ GO / Γ GO max still gradually increases and finally equilibrates at Ae ≈ 0.85 . However this late stage of adsorption does not significantly change the shape of the pendant drop. We note that a similar phenomenon has been observed on the adsorption of amphiphilic protein molecules at the air/water interface.40,

41, 42

These two areal fractions Ac and Ae may correspond to the random very loose packing limit (close to 0.6543) and the random close packing limit (2D random close packing occurs at relative coverage values of 85 ±

10

ACS Paragon Plus Environment

Page 11 of 33

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

1% for polydisperse disks.44,

45, 46),

respectively, for granular particles packing and

jamming in 2D. Different from traditional granular particles, which are typically so heavy that their thermal motion is negligible, the adsorbed GO nanosheets can still diffuse or rearrange in plane under thermal fluctuation. Therefore, random close packing limit can be approached over time. Between Ac and Ae limits, the interfacial film



is probably highly viscoelastic. The co-assembly of GO nanosheets and diblock copolymers at the toluene/water interface changed the dissipation and static friction among GO sheets. More studies, for instance, using techniques such as fluorescence recovery after photobleaching47, 48, interfacial dilational rheology49, or interfacial shear rheology50 are called for to observe the relaxation of the adsorbed GO nanosheets in plane. When the bulk concentration of GO is below 0.08 g/L, no sharp transition is observed in the dynamic interfacial tension (Figure 5(a)), indicating that no interfacial jamming is reached. Indeed, in Figure 6(a), the equilibrium areal fraction of GO adsorption is Ae ≈ 0.45 , lower than the critical areal fraction of Ac ≈ 0.65 for the onset of interfacial jamming.

At the initial stage of GO nanosheets adsorption, the interface is almost empty of GO coverage. Therefore desorption of GO can be neglected. The interfacial coverage of GO can be calculated directly from Fick’s diffusion equation and mass conservation25:

Γ GO = 2cGO

DGOt π

(8)

In Equation 8, the influence of drop curvature is neglected since the length scale defined by the diffusion,



DGOt , is much smaller than the radius of the drop r . In deriving

Equation 8, we also assume that the interface has already been covered with a high population of PS-b-P2VP molecules so there is an ample number of active sites from the PS-b-P2VP at the interface permitting the adsorption of GO. In Figure 6(b), we re-plot the normalized GO surface coverage (for the GO concentration of 0.12 g/L) as a function of t 1/2 . The maximum coverage of graphene oxide at the interface is taken to be the

11

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 12 of 33

inverse of the particle size, Γ GO = 3× 10−6 nm −2 for an average GO particle diameter of max 600 nm, a molecular weight of 1.9 × 108 Dalton can be estimated, given that the carbon atom density on graphene is 38.2 nm-2 and assuming the carbon to oxygen ratio is 3:1.51, 52

Linear fitting of the initial stage of GO absorption in Figure 6(b) using Equation 8,

allows us to obtain the diffusion constant of GO in water DGO ≈ 3.9× 10−13 m2 /s . We note that this value is very close to the expected diffusion constant of GO from the EinsteinStokes equation, confirming that at the initial GO adsorption stage, the diffusion of GO to the subsurface is the rate limiting step for GO assembly at the PS-b-P2VP present toluene/water interface.

Interfacial jamming will occur at the same critical areal

fraction Ac ≈ Γ GO / Γ GO ≈ 0.65 . c max

Thus, for water drops containing different bulk

concentrations of GO in contact with a toluene solution containing 0.1 g/L PS-b-P2VP, according to Equation 8, the jamming point is reached at a surface coverage of −2 . It Γ GO = 2cGO DGOt c / π . In Figure 5(b), we plot the jamming time t c as a function of cGO c is evident that the interfacial jamming occurs at roughly a constant surface coverage,

since all the data points collapse onto a linear trend that goes through the origin. This further confirms that the early state GO adsorption has diffusion controlled kinetics and the onset of interfacial jamming is responsible for the transition observed on the decaying of the dynamic interfacial tension. Next, we consider the condition of a water drop with fixed GO bulk concentration (0.08 g/L) in contact with toluene containing different bulk concentrations of PS-b-P2VP. The evolution of the dynamic interfacial tension is plotted in Figure 7. When the diblock copolymer concentration is 0.1 g/L, the dynamic interfacial tension gradually decreases with the contact time and doesn’t level off at 1100 s. We know from Figure 6 (a) that the equilibrium GO areal coverage is Ae ≈ 0.45 , lower than the jamming limit. When the diblock copolymer concentration is 0.2 g/L or above, the dynamic interfacial tension rapidly plateaus in