The Search for Nanobubbles by Using Specular and Off-Specular

Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

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


The search for Nanobubbles by using specular and offLangmuir is published by the American specular Neutron Reflectivity 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


Philipp Gutfreund, Marco Maccarini, Andrew J.C. Dennison, and Max Wolff


Subscriber access provided by CORNELL UNIVERSITY Langmuir, Just Accepted Manuscript • LIBRARY

DOI: 10.1021/acs.langmuir.6b02087 • Publication Date (Web): 12 Aug 2016 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

Subscriber access by CORNELL UNIVERSITY LIBRARY Downloaded from provided http://pubs.acs.org on August 23, 2016


Subscriber access provided by CORNELL UNIVERSITY LIBRARY Just Accepted



“Just Accepted” manuscripts have been peer-reviewed and ac online prior to technical editing, formatting for publication and 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

Subscriber access provided by CORNELL UNIVERSITY LIBRARY Society provides “Just Accepted” as a free service to the


Subscriber access provided by CORNELL UNIVERSITY LIBRARY dissemination of scientific material as soon as possible after a


Subscriber access provided by CORNELL UNIVERSITY LIBRARY appear in full in PDF format accompanied by an HTML abstract


Subscriber access provided by CORNELL UNIVERSITY LIBRARY fully peer reviewed, but should not be considered the official ve


Subscriber access provided by CORNELL UNIVERSITY LIBRARY readers and citable by the Digital Object Identifier (DOI®). “Jus


provided by CORNELL UNIVERSITY LIBRARY toSubscriber authors.access Therefore, the “Just Accepted” Web site may not


access provided by CORNELL UNIVERSITY LIBRARY inSubscriber the journal. After a manuscript is technically edited and form


Subscriber access provided by CORNELL UNIVERSITY LIBRARY Accepted” Web site and published as an ASAP article. Note th


Subscriber access provided by CORNELL UNIVERSITY LIBRARY changes to the manuscript text and/or graphics which could


Subscriber access provided by CORNELL UNIVERSITY LIBRARY and ethical guidelines that apply to the journal pertain. ACS or consequences arising from the use of information containe 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

Page 1 of 25 4

RQ

4

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Langmuir D2 O as received N2 enriched D 2 O

2 -8

10

8 6 4

2

ACS Paragon Plus Environment

0.10

0.20

-1

Q (Å )

0.30

Langmuir D 2 O as received N2 enriched D 2 O Steitz et al. Doshi et al.

SLD (10 Å )

6

-6

-2

1 2 3 4 5 6 7 8 9 10 11 12 13

5 4 3 2 1

SiO2

OTS

Page 2 of 25

D2O

Si

0 ACS Paragon Plus Environment

0

20 40 60 distance from interface (Å)

80

Page 3 of 25

Langmuir

-2

Reflectivity

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

10

D2O

Yoneda peak

as received D2O gas enriched D2O after 11 h

-3

10

-4

10

-5

10


0.0

0.5

1.0 θf [º]

1.5

2.0

0

1

2

-1

Langmuir

0-2 h after solvent exchange

Page 4 of 25 50 40

φf [º]

30

0

20 10 0 ACS Paragon Plus Environment

-6

1

3

x10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

θf [º]

0

Page 5 of 25 1

2

Langmuir

θf [º]

r = 100 nm σ = 5%

3

50 40

φf [º]

30

10 0 ACS Paragon Plus Environment

-6

20

x10

1 2 3 4 5 6 7 08 9 10 11 12 13 14 15 16 -1 17 18

1

-5

Langmuir

Reflected Intensity

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Page 6 of 25

Data 0-2h after solvent exchange r = h = 50 nm σ = 5% r = h = 50 nm σ = 10% r = h = 100 nm σ = 5% r = h = 100 nm σ = 10% r = 50 nm h = 10 nm σ = 10% r = 50 nm h = 10 nm σ = 25%

3.5x10

3.0 2.5 2.0 1.5 1.0 0.5 0.0


0.4

0.6

| φf [º] |

0.8

1.0

Page 7 of 25

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 search for Nanobubbles by using specular and o-specular Neutron Reectivity Philipp Gutfreund, ∗ † Marco Maccarini, † ‡ Andrew J.C. Dennison, ,

,

¶ , †, §

and Max

Wol k †Institut Laue-Langevin, 38000 Grenoble, France ‡TIMC-IMAG, Université Joseph Fourier, 38706 Grenoble, France ¶Department of Chemistry, Technical University Berlin, 10623 Berlin, Germany §Department of Physics and Astronomy, University of Sheeld, S102TN Sheeld, UK kMaterials Physics, Department of Physics and Astronomy, Uppsala University, 75121

Uppsala, Sweden E-mail: [email protected]

Abstract We apply specular and o-specular neutron reection at the hydrophobic silicon/water interface to check for evidence of nanoscopic air bubbles whose presence is claimed after an ad hoc procedure of solvent exchange. Nanobubbles and/or a depletion layer at the hydrophobic/water interface have long been discussed and generated a plethora of controversial scientic results. By combining neutron reectometry (NR), o-specular reectometry (OSS) and grazing incidence small angle neutron scattering (GISANS), we studied the interface between hydrophobized silicon and heavy water before and after saturation with nitrogen gas. Our specular reectometry results can be interpreted by assuming a sub-molecular sized depletion layer and the o-specular measurements show no change with nitrogen super saturated water. This picture is consistent with 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 8 of 25

the assumption that, following the solvent exchange, no additional nanobubbles are introduced at signicant concentrations (if present at all). Furthermore, we discuss the results, in terms of the maximum surface coverage of nanobubbles that could be present on the hydrophobic surface compatibly with the sensitivity limit of these techniques.

Introduction The solid-water interface has been for many years the subject of fundamental scientic research due to the relevance of interfacial phenomena to biological processes in which water plays a major role.

1

Moreover, the push for miniaturization driven by rapidly evolving mi-

crouidic technologies

2

calls for a deeper understanding of the physical-chemical mechanisms

that govern the processes occurring at this interface. Experiments using dierent model hydrophobic surfaces at the solid-water interface showed controversial results as a liquid layer of reduced density was observed sandwiched between the hydrophobic surface and the bulk water.

3

In order to quantify the resulting depletion eects, the depletion distance

d2 4

was

introduced:

Z ρ(z) d2 = dz, 1− ρbulk where and

ρ(z)

ρbulk

(1)

denotes the density of the depleted liquid at a distance

represents the bulk liquid density.

d2

z

from the interface

reduces the smeared-out density prole

of the depletion to a step-like function that represents an equivalent layer of zero density. The results of neutron reectometry (NR) studies on spin-coated deuterated polystyrene (dPS)-D2 O interfaces showed a depletion distance of experiment by another group

6

d2 ≈ 2.6 Å, 5

however, a repeat of this

did not show this depletion when a freshly prepared PS lm

was not exposed to air before the measurement.

A similar scenario emerged with X-ray

reectivity (XRR) results on bulk water in contact with octadecyl-trichlorosilane (OTS) coated substrates.

7

A depletion distance of only

d2 = 1.1 Å

was observed although up to

half of the contribution may have arisen from the hydrogen termination of the hydrophobic

2


Page 9 of 25

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

coating, which is practically invisible to X-rays.

8

Even though this topic is still controversial,

the depletion between water and a hydrophobic surface, if present at all, might occur on a sub-molecular scale and may arise from preferred orientation of water molecules at the interface as suggested by molecular dynamics (MD) simulations and demonstrated for other liquids at hydrophobic surfaces.

7

and Raman-scattering

9

1013

The situation becomes increasingly complicated if water supersaturated with gas is used as the exceeding gas may condense in the form of a nanoscale layer at a hydrophobic interface.

14

These so-called nanobubbles were initially assumed to be the cause of the hydropho-

bic gap

5

and motivated further reectivity studies to probe the inuence of dierent gas

enrichments on the water depletion yielding controversial results.

15,16

A further argument of

controversy is the fact that theoretical calculations predicted the lifetime of nanobubbles to be in the

µs range 1719

and thus not observable on a laboratory time scale. Therefore the ob-

servation of nanobubbles has often been attributed to the invasive eect of the Atomic Force Microscopy (AFM) technique,

20,21

which was mainly used to show their presence. With the

introduction of a well established protocol to make nanobubbles reproducibly and at high surface coverage in this system,

22

the so-called solvent exchange technique, it became possi-

ble to deliberately produce nanobubbles which are stable on the time scale of hours or even days

23

by replacing ethanol by water.

Eventually, nanobubbles were observed with non-

invasive techniques like infrared spectroscopy

14

and optical microscopy.

24

However, though

it is currently accepted that nanobubbles may be articially produced locally at solid-liquid interfaces under certain circumstances

25

their (time and space averaged) concentration and

thus their inuence on mesoscopic quantities in microuidics or cell biology is still debated.

26

We believe this is due to the fact that the techniques used so far to examine nanobubbles are either local (various microscopy techniques) or they had no particular sensitivity to the shape of the bubbles, and quantifying only the mean surface coverage (infrared spectroscopy, specular reectivity). The reported size of nanobubbles (50 - 10000 nm) makes them ideal candidates for investi-

3


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 25

gations with grazing incidence small angle neutron scattering (GISANS), specular and ospecular neutron reectometry. These techniques operate in the reciprocal space and provide structural information averaged in time and over a macroscopic surface area in contrast to the aforementioned local microscopy techniques.

27

The investigation of buried interfaces, e.g.

solid-liquid interfaces, is mainly done by NR or high energy X-rays as those interfaces are rarely accessible by soft X-rays or visible light. Neutrons have a further advantage for soft matter interfaces since cold and thermal neutrons do not cause radiation damage of organic specimens nor they cause alterations of the samples due to radiation induced accumulation of charges on the surfaces.

28

Neutron scattering techniques can further augment their capabili-

ties thanks to isotopic substitution, which oers a powerful tool for contrast enhancement of the low atomic number elements typically composing soft matter.

29

O-specular scattering

(OSS), which is typically several orders of magnitude weaker than the specular reection is not routinely used, apart from synchrotron X-rays sources.

27

However, specic examples

where neutron OSS has been used are highly ordered systems like magnetic domains or gratings, especially in multilayered systems, water.

31

30

or for microphase separated polymers in heavy

Other examples include capillary waves,

and organic photovoltaics,

36

32

lipid bilayers,

33,34

polymer dewetting

35

which all prot from the high contrast of deuterated materi-

als. Grazing incidence small angle neutron scattering has recently experienced an increasing recognition

37

partly due to the unique possibility to use TOF in combination with GISANS

to measure depth dependent patterns in a single measurement analysis software.

38,39

and due to advances in

40

Previous GISANS measurements

41

on the interface between

dPS and D2 O were performed

on D22 at the Institut Laue-Langevin (ILL) in Grenoble, France, but the PS layer was unstable at higher temperatures used to enhance the appearance of nanobubbes. Therefore we used in this study silicon substrates coated with OTS that are stable at the relevant range of temperature and performed specular and o-specular NR and GISANS on N 2 enriched D2 O in order to cover the relevant lateral and out of plane length-scales of depletion layers

4


Page 11 of 25

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

and nanobubbles.

Experimental Details 3 The silanization of the single crystal silicon (100) block (80*50*20 mm , Siltronix, France) was performed according to reference water was 102

◦

42

. The advancing contact angle of

and the receding one 70

◦

Millipore

ltered

◦ (static contact angle was 93 ). The specular and

o-specular NR measurements, which took 35 min and 10 min, respectively were performed on the FIGARO horizontal sample plane reectometer

43

at the Institut Laue-Langevin using

a wavelength ( λ) range from 1.7 Å to 19 Å with a relative resolution of reectivity was measured at two reection angles (0.625 olution of

∆θ θ

= 3.3 %.

= 4.2 %.

The

◦ and 3.2 ) with a relative angular res-

The GISANS measurements, which were also performed on FIGARO

and took 2 h each used a wavelength resolution of range as before. The reection angle was set to angular divergence of

◦

∆λ λ

∆θi = 0.02◦

and

= 7.4 %

θi = 0.393◦

∆φi = 0.1◦ ,

tion corresponds to an angular spread of

∆λ λ

with the same wavelength

with a vertical and horizontal

respectively. The detector pixel resolu-

∆θf = 0.018◦

and

∆φf = 0.16◦ .

All resolutions are

given as Gaussian equivalent full width at half maximum (FWHM). The solid-liquid sample cell contained about 3 ml of liquid and was mounted with the liquid on top of the solid so that macroscopic bubbles would drift away from the interface under consideration. The measurements were taken according to the following procedure.

Firstly specular and

o-specuar reectivities were recorded on the OTS-D 2 O interface by using heavy water as received from

Sigma-Aldrich,

France (99.9 atom % deuteration).

Then the water was ex-

changed by 9 ml of ethanol and after 1 min the ethanol was again exchanged within 100 s by 12 ml D2 O which was saturated with nitrogen by bubbling it with

N2

for 30 min at a

◦ ◦ temperature of 5 C. The sample cell and the ethanol were kept at 45 C throughout the experiment. Fitting of the reectivity data was accomplished using a slab model with

5


Motot . 44

The

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 25

error in depletion distance, was determined as follows. The SLD and the thickness of the density depleted layer were allowed to vary simultaneously until the

χ2

increased by 5% as

compared to the best t. This was done because the thickness and SLD of this layer are highly correlated parameters in the data tting. Then the maximum and minimum values for the depletion distance were calculated within this parameter range.

The absolute

χ2

values for all NR ts varied between 4 and 5. The GISANS patterns were reduced as follows: In order to accumulate meaningful statistics the data was binned to wavelength bands of

∆λ/λ = 20 %.

For the wavelength shown here

(5 Å) this corresponds to a wavelength range of 4.5-5.5 Å . This results in a Gaussian equiv-

∆λ/λ = 15 %

alent wavelength resolution of

(FWHM). Although other wavelengths were

recorded and analyzed as well it turned out that only at 5 Å wavelength a GISANS signal outside the specular and direct beam region above background could be observed. As can be seen in the measured GISANS patterns the background was very low, on the order of

10−6

with most of the pixels having 0 counts after 2 h of acquisition. We therefore did not

subtract background from the GISANS images. The analysis of the GISANS patterns was done by comparing the absolute intensities to simulations performed with the software package BornAgain.

40

Detector sizes and pixels,

incident and reected angles and divergence as well as wavelength and wavelength resolution were matched to those used in the experiment. The simulations used layer parameters from the specular reectivity data to form a multilayer structure to which model nanobubbles of oblate shape with dierent spherical radius

r, height h and surface coverage σ

were added at

the water-OTS interface. The model used to describe the expected scattering from nanobubbles was the form factor of a truncated oblate spheroid with a scattering length of 0 (air) and the measured contact angle

θ = (180◦ − 93◦ ) = 87◦ .

Structure factor contributions were not included as high nanobubble coverage would be clearly visible by specular reectivity and would give rise to stronger scattering than that observed, so the simulations focussed on coverages below 25% (less than half that expected

6


Page 13 of 25

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

by random sequential adsorption). Polydispersity eects were not included and would correspond to the weighted sum of the images shown in the supporting information. In-plane roughness correlations of the respective interfaces were not included in the simulation as they did not produce scattering stronger than that of the nanobubbles in the range of realistic correlation lengths (

The Search for Nanobubbles by Using Specular and Off-Specular

Recommend Documents