Influence of the wetting behavior and surface energy on the dispersibility of multi-wall carbon nanotubes

Accepted Manuscript Title: Influence of the Wetting Behavior and Surface Energy on the Dispersibility of Multi-Wall Carbon Nanotubes Author: A. Dresel U. Teipel PII: DOI: Reference:

S0927-7757(15)30287-9 http://dx.doi.org/doi:10.1016/j.colsurfa.2015.10.027 COLSUA 20235

To appear in:

Colloids and Surfaces A: Physicochem. Eng. Aspects

Received date: Revised date: Accepted date:

14-7-2015 9-10-2015 17-10-2015

Please cite this article as: A.Dresel, U.Teipel, Influence of the Wetting Behavior and Surface Energy on the Dispersibility of Multi-Wall Carbon Nanotubes, Colloids and Surfaces A: Physicochemical and Engineering Aspects http://dx.doi.org/10.1016/j.colsurfa.2015.10.027 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Influence of the Wetting Behavior and Surface Energy on the Dispersibility of MultiWall Carbon Nanotubes A. Dresel, U. Teipel Technische Hochschule Nürnberg, Forschergruppe “Partikeltechnologie und Rohstoffinnovationen“, Wassertorstrasse 10, 90489 Nürnberg, Germany, [email protected]

GRAPHICAL ABSTRACT

Highlights 

The wetting behavior of carbon nanotubes was studied with liquid penetration method



The surface energy with disperse and polar fraction were derived



Carbon nanotubes exhibited a significant disperse and polar component



Good dispersibility in fluids with similar disperse and polar surface energies



Evidence of the influence of particle interactions by covalent functionalization

Abstract Carbon nanotubes (CNTs) possess extraordinary particle properties which makes them ideal for the development of innovative polymer composite materials. Because of the tendential agglomerate forming of CNTs, their homogenous exfoliation in organic or inorganic phases is often required and can be obtained by mechanical stressing of the agglomerates. However, the CNT-exfoliation is only achievable in few fluids. Therefore, the CNT-functionalization which means the covalent attachment and modification of the particle surface with functional groups or molecules is applied to alter particle interactions and in this way, their dispersing behavior. The poor CNT-dispersibility in fluids and CNT-dispersion strategies as the functionalization show the importance of the characterization of their particle interactions. In order to describe CNT/fluid-interactions the wetting behavior was characterized by contact angle measurements with the capillary liquid penetration method. The CNT-surface energies regarding Owens and Wendt could be estimated from the measured contact angles. Thereby, non-modified CNTs exhibited a good interaction with molecules of longer alkyls and even polar liquids and showed further a significant dispersive and polar fraction of the surface energy. The determined surface energies were used for the explanation and estimation of the CNT-dispersibility in liquids. Successful dispersions could be achieved with fluids of similar dispersive and polar fractions compared with those of the particles. Furthermore, ethylaminefunctionalized CNTs possessed an enhanced polar component of their surface energy which demonstrated that the CNT-fluid interactions can be controlled by covalent CNTfunctionalization.

Keywords: Carbon Nanotubes, Functionalization, Wetting Behavior, Surface Energy, Dispersion

1

Introduction

Carbon nanotubes (CNTs) are tubular carbon particles with shells of one or more graphite layers distinguished as single-wall (SWNT) and multi-wall (MWNT) [1]. These particles possess extraordinary properties as a high aspect ratio [2], enormous mechanical strength [3], electrical [4] and thermal conductivity [5] combined with low material densities [6]. These properties make carbon nanotubes very attractive for use in the development of innovative materials in a variety of fields [2,7]. However, the design of CNT-applications is challenging since carbon nanotubes strongly agglomerate and form bundles, that means the parallel alignment of individual CNTs, driven by Van-der-Waals attraction and their huge aspect ratios [8,9]. Therefore, the application of CNTs in innovative developments often requires their dispersion in organic or inorganic fluids which can be achieved by the mechanical stressing of the agglomerates [10]. The ultrasound dispersion is thereby frequently applied [8– 11]. However, the carbon nanotube exfoliation is only possible in few liquids whereby the CNT-dispersion is often enhanced by physical or chemical functionalization. Physical functionalization denotes the adsorption of amphiphilic molecules or macromolecules [12] while the chemical modification is the covalent bonding of functional groups or molecules [13] onto CNT-surface. With these methods interparticle and particle-fluid interactions are aimed to be altered [10]. The different CNT-dispersion strategies display the importance of the characterization of particle/fluid-interactions of CNTs concluding changes by functionalization techniques regarding the CNT-dispersibility. In literature, for example, H. T. Ham et al. 2005 [14] compared the dispersion of SWNTs in different liquids and aqueous surfactant systems with fluid parameters. They offered a strong influence of the dispersive component of Hansen solubility parameters [15] whereas high polar and hydrogen bonds could be neglected. In contrast, S. D. Bergin 2009 et al. [16] estimated Hansen parameters of < δD > = 17.8 MPa1/2, < δP > = 7.5 MPa1/2 and < δH > = 7.6 MPa1/2 for solvents with good dispersibility properties for SWNTs. However they concluded that the full understanding of the fluid dispersibility of carbon nanotubes remains to be understood [16]. The science of wetting provides suitable models for the explanation of CNT/fluid-interactions since it characterizes the interaction behavior of fluids on solid surfaces. In literature, the CNT-wetting is frequently investigated through contact angles measurements on CNTbuckypapers or cushions by the sessile drop method [17,18]. A. Sobolkina et al. [17] determined water contact angles on buckypapers of different types of carbon nanotubes.

However, they mentioned that the contact angle can only be roughly quantified due to the surface roughness. A precise method for the characterization of the wetting of one individual CNT could be derived from atomic force microscopy (AFM) [19]. On that microscopic scale also transmission and scanning electron microscopy (TEM and SEM respectively) were applied to contact angle estimations of melted metal or polymer droplets on carbon nanotubes [20,21]. Besides, a powder immersion test showed that CNTs can be generally wetted with liquids of surface tensions lower than 200 mNm-1 [21–23] which can even fill the inner side of open tubes by their capillarity [24,25]. B. A. Kakade and V. K. Pillai [26] investigated the wetting behavior of CNT-applications by the sessile drop method on thin films with water which indicated the change from hydrophobic to hydrophilic films by the incorporation of functionalized CNTs. Furthermore, the theoretical consideration of the wetting from a thermodynamic point of view indicated a significant influence of the CNT-diameter on theoretical contact angles [27]. Another important attribute for the characterization of particle/fluid-interactions provides the surface energy with disperse and polar components. Regarding F. M. Fowkes [28], the surface energy can be recognized as magnitude for the acting intermolecular forces which are distinguished in different types, namely in dispersive and polar contributors. Dispersive interactions, also denoted as London dispersion forces, are caused by the affinity of fluctuating dipoles to induced dipoles in other molecules. Polar interactions comprise interactions caused by hydrogen bonds, between dipoles (Keesom forces) as well as dipoles and polarizable molecules (Debye forces). The solid free surface energy γsv is then given by the sum of a polar (p) and disperse (d) component as in equation (1) [28,29]:

γ

sv

p d = γ +γ sv sv

(1)

The solid surface energy can be derived from contact angle measurements whereof the OWRK-model [30–32] is frequently used in colloidal science. Literature values for surface energies of non-modified CNTs are exemplarily shown in table 1. The OWRK-surface energy from contact angles determined by microscopy characterization is of similar magnitude whereas the surface energy derived from the sessile drop method is significantly higher. However, the shown investigations offer always the presence of a dispersive and polar fraction. The surface energy regarding Zisman [33,33] showed a wide region due to strong variations of the determined contact angles by the applied microscopy contact angle estimation.

In comparison to the methods above, the inverse gas chromatography (IGC) can be applied for the determination of the CNT-surface energy and characterization of their interaction behavior. Thereby, the dispersive surface energy is usually determined. For example, surface energies for pristine carbon nanotubes with 94 Jm-2 [34] or even higher than 120 Jm-2 [35] can be found in literature. O. Diaz et al. [35] mentioned an influence of the contribution of highenergy sites outweighed by IGC and of the increased interaction potential in micropores on that higher dispersive surface energies. In contrast to that the wetting behavior provides an average surface energy [35]. In conclusion, that shows the possibilities which are provided by the science of wetting for the characterization of CNT/fluid-interactions and the surface energy. However, it indicates also the problematic of a precise analysis of the wetting of disperse systems since direct contact angle measurements cannot be easily applied. In this work we present the liquid penetration method for the characterization of the wetting of carbon nanotube powders. Contact angle analysis on non-modified and functionalized carbon nanotubes were conducted how the surface energies of the CNT-materials could be derived. The CNT-surface energy was then applied for the characterization of CNT/fluid-interactions and the description of CNTdispersibility in liquid phases.

2

2.1

Material and Methods

Carbon-Nanotube Materials

Multi-wall carbon nanotubes have been used for the investigation of the wetting properties. Particularly, Baytubes® C150P (MWNTp) with carbon purity of > 95 % and Baytubes® C150HP (MWNThp) with carbon purity of > 99% (Bayer Material-Science AG, Leverkusen, Germany) were available. Length and diameter of MWNTp could be determined after their dispersion in N-methyl-2-pyrrolidone with a rotor-stator disperser (see chapter 2.5) from transmission electron microscopy analysis (images not drawn here). The diameter of the tubes was between 3.8 and 24.6 nm whereas the MWNTp-length varied in a wide range with a minimum and maximum length of 0.024 and 2.4 µm. The specific surface area of that material could be determined by multi-BET measurements with 198.3 m2g-1 [36]. In comparison to these non-functionalized MWNTs, the wetting behavior of modified MWNTs with covalently attached ethylamine molecules could be investigated. The functionalized MWNTs have been produced and characterized from our project partner the BASF in Ludwigshafen by a wet chemical functionalization process [37]. Source material for

the functionalization was the MWNTp. The ethylamine functionalization was characterized by the nitrogen fraction determined with X-ray photoelectron spectroscopy (XPS) and elementary analysis (EA). Elementary analysis showed a nitrogen fraction of < 1 % and the XPS a fraction of 0.1 at-% in the modified material. Besides that, a significant fraction of oxygen could be detected with 1.8 % and 0.9 at-% from EA and XPS respectively. The detected oxygen may be caused from remaining metal catalysts that can be applied in MWNT-synthesis [38]. The applied CNT-Materials and its specification are summarized in table 2.

2.2

Wetting and Dispersing Fluids

The used fluids with fluid density ρl, viscosity ηl and surface tension γlv with disperse and polar components regarding equation (2) [28] can be taken from table 3. The data are given with respect of a reference temperature between 20°C and 25°C so that the temperature influence is negligible. It can be observed from the table that fluids with diverse dispersive and polar properties were used. Particularly, the influence on the wetting behavior of the fluid surface tension was investigated with purely dispersive n-alkanes. In comparison to that, the impact of small polar fractions of the surface tension was analyzed with different n-alcohols which possess related molecule structures. The MWNT-wetting behavior of common organic and inorganic solvents as for example acetone and water as well as CNT-dispersing fluids like N-methyl-2-pyrrolidon and dimethylformamide was investigated.

p d γ = γ +γ lv lv lv

2.3

(2)

Liquid Penetration Method - Tensiometric Measurements

The wetting behavior was characterized with the liquid penetration method. A common apparatus for the penetration method is shown schematically in Figure 1. It comprises of a sample holder which connects the particles with a balance and the wetting liquid in the container beneath. The liquid contacts the bulk powder via a filter and penetrates through it driven by the powder capillarity. During the experiment the fluid mass increase is detected in dependence of the penetration time.

The relationship between the fluid mass ml in the powder and the penetration time t is given by the modified sorption equation (3) [41] which is derived from the Washburn equation [42]. 2 2 2 ρ * [A *  c* r  ] * γlv * cosΘ ml = l *t 2* η

(3)

l

In equation (3) are the properties of the wetting liquid. The expression in the square brackets is the capillary or material constant which is given by the free cross sectional area A in the powder, the orientation factor c and average radius r considering the porous structure and capillarity [43,44]. The material constant depends mainly on the powder properties and its packing density which means it is the crucial parameter for this method and requires a reproducible adjustment [41,45]. The material constant has to be determined with a wetting liquid assumed to wet the powdery material completely. If it is known, the contact angle Θ can be determined from the penetration curves as in equation (3). The liquid penetration measurements have been conducted with a tensiometer (Tensiometer K100, Krüss GmbH, Hamburg, Germany). Cylindrical sample holders with an inner diameter of 10 mm were applied for the measurements at a temperature of 298 K. The MWNT-samples were compressed to achieve reproducible constant fluid mechanics during the liquid penetration. Additionally, a certain amount of the CNT-powders were divided into equal smaller portions using a rotating sample divider (repro, Bürkle GmbH, Bad Bellingen, Germany) and a sample division process. Sufficient compaction of the CNT-powders was obtained using a stamping volumeter (STAV 2003, J. Engelmann Ludwigshafen Company, Germany) according to DIN ISO 787-11 which is usually applied to determine the tap density of powders. The powders were compressed with 25 stamps while a compression weight of approximately 10 grams was placed on top of the powder during tapping. The resulting CNTsample was one centimeter high. The average mass of non-modified MWNTs was 0.16 g and of the modified MWNTs 0.15g per sample. The non-functionalized MWNTs were dried at 150 °C and functionalized particles at 120 °C for 120 min to remove eventually adsorbed moister and stored afterwards in a desiccator.

2.4

Cryo Scanning Electron Microscopy

Cryo scanning electron microscopy (CryoSEM) was conducted on wetted non-modified MWNTs in order to get insight of the fluid state in the MWNT-powder and agglomerates. The MWNTp-powder was wetted with water by suspending the particles in the fluid. The water

soaked powder and agglomerates were poured into a sample holder and flash frozen in liquid nitrogen. A fracture plane was created by hitting the frozen MWNT sample with a metal stick. The fracture plane was analyzed with a scanning electron microscope (LEO 1530 VP Gemini, Carl Zeiss AG, Oberkochen, Germany) without sublimation of the frozen fluid from the sample.

2.5

Dispersion and Characterization of Carbon Nanotubes

In comparison to the wetting behavior, carbon nanotubes were dispersed in different liquids. Therefore, the CNTs were dispersed batchwise in a total volume of 340 ml with a rotor-stator disperser (Ultra Turrax® T 25 digital, IKA®, Staufen, Germany) at a concentration of 0.01 wt.-% and a temperature of 25 °C. The dispersing time for each dispersion was constant at 60 min and the circumferential speed of the dispersing tool (S 25 KV-18 G, IKA®, Staufen, Germany) was set to the maximum of 24.4 ms-1. The resulting dispersions were then characterized by their sedimentation and decomposition behavior in a centrifugal field. Light extinction profiles of the dispersions were monitored while centrifuging in an analytical centrifuge (LumiSizer®, LUM GmbH, Berlin, Germany) [46]. The measuring apparatus is shown in figure 2 and comprises of a light source (1) which applies parallel NIR-light along the dispersion in a cuvette (2) which is located on the rotor. The detector (3) monitors the space and time resolved extinction profiles (4-5) which characterizes the segregation of the dispersion. For the measurements, polyamide cuvettes were used with a width of 10 mm to minimize effects of the sample cell on the particle movement. In order to characterize polydisperse systems the centrifugation was conducted with a rapid stepwise increase of the centrifugal acceleration from 13 x g to 2325 x g (at a radius of 130 mm). Afterwards, it was kept constant at the maximum acceleration of 2325 x g for a long period of time.

3 3.1

Results and Discussion Penetration Measurements and Reproducibility

Penetration curves are shown in figure 3 by the fluid mass increase in dependence of the penetration time for the wetting of MWNTp with ethylene glycol. The diagram shows that the penetration process continues approximately 80 s wherein the liquid mass in the CNT-powder increases from 0 to 1.1 g. At longer wetting times, the fluid mass remains constant and shows thus a complete liquid penetration. The diagram contains three single measurements that show clearly the good compliance of their slopes. It demonstrates that the crucial parameter of the

method, constant fluid dynamics during the penetration, could be achieved through the underlying sample preparation technique which is the essential basis of this method.

3.2

Liquid Penetration and Determination of the Capillary Constant

For the penetration method the capillary constant (as in equation 3) has to be determined with a fluid that can be assumed as a complete wetting liquid of the solid surface (cos Θ = 1). In general, dispersive liquids with small surface tensions can often be applied [41,44]. For example, figure 4a shows penetration curves of n-hexane, n-heptane, low viscous polydimethylsiloxane (2.79 mPas) and ethylene glycol on MWNTp. In figure 4b, the liquid penetration of diiodomethane, dimethyl sulfoxide and water are additionally shown which were used for determining the surface energy. The relatively fast liquid penetration through the bulk can be seen by the termination of fluid mass increase after at longest 100 s. Especially the reasonable good wetting of the pristine CNT-material with water is noticeable which is indicated by its rapid capillary penetration. The beginning of penetration is characterized by the non-linear mass increase with the root of the time which is transferred to a homogenous penetration regarding the undelaying penetration model at longer wetting times. The capillary constant could be determined from each first linear segment drawn in each curve and given with the slope and R2 value of the linear regression.

The capillary constants were determined from three single measurements for each fluid and can be taken from table 4 as the arithmetic average with its standard deviations s. The very small standard deviation corresponds thereby with the good reproducibility of the penetration measurements and hence, the capillary constants. The material constant of 1.21x10-4 cm5 determined with ethylene glycol is significant higher compared with the liquids in table 4. In addition, the capillary constant of ethylene glycol was the highest of all used fluids in this work (see chapter 2.2).

The capillary constant represents the porosity and cavity structure in the powder. A high value of the constant represents a good penetration of the liquid into the open spaces caused by a good wetting and thus, a good detection of the porosity and capillary structure of the powder. The capillary constant of the MWNT-samples is therefore best detected with ethylene glycol. Even if ethylene glycol possesses polar properties represented by the polar fraction of the surface tension (see chapter 2.2), it occupies an outstanding wetting ability of the underlying

MWNTp-material theoretically composed of graphitic structures. This indicates an influence of the polar fluid surface tension in addition to the dispersive properties.

3.3

MWNT-Wetting Behavior

The wetting behavior of the non-functionalized MWNTs has been further investigated by the wetting with dispersive linear hydrocarbons (n-alkanes) of rising chain lengths with increasing surface tensions as wells as monovalent linear alcohols with an additional polar component. The results are shown by the cosine of the contact angle (open squares: n-alkanes, open rhombs: n-alcohols) in dependence of the liquid surface tension in figure 5 (a). Average values of three single measurements with its standard deviation bars are shown in the diagram. The figure shows a tendential increase of the cosine of the contact angle with increasing liquid surface tension and hence, molecule chain length. The cos(Θ) of n-alkanes increases from 0.30 (n-hexane) to 0.63 (n-hexadecane) and n-alcohols from 0.52 (ethanol) to 0.90 (1-octanol) whereby higher fluctuations occur at higher surface tensions. It demonstrates significant influence of the molecule structure of the liquid on the MWNT-wetting since the observed behavior of molecules with varying alkane or alkyl chains. This shows a contrary behavior compared with commonly observed wetting phenomena on solid and polymer surfaces which display poorer wetting with increasing liquid surface tension according W. A. Zisman [33]. Furthermore, the n-alcohols demonstrate better wetting than n-alkanes at comparable surface tensions. This observed behavior of the MWNTs exhibits a complex nature of the wetting of carbon nanotubes where besides the surface tension dispersive as well as polar fluid properties influence CNT/fluid-contact.

The observed abnormal wetting phenomenon of n-alkanes and n-alcohols is caused by a better adsorption on the CNT-outer shell of the fluids with more CH2- or OH-structures. The CNTwetting might be slightly influenced by structural defects within the graphitic lattices, amorphous carbon on the outer sidewall, remaining catalysts or even foreign atoms as oxygen which are integrated in the carbon nanotube construction. Suchlike deviations of MWNTs from their ideal structure are well reported in literature [47,48]. However, R. Tessonnier et al. [48] reported exemplarily an oxygen/carbon ratio situated at the CNT-surface by XPS analysis of approximately 2 x 10-3 for Baytubes C150P which is reasonable low. Besides, a homogeneous CNT-structure, well ordered side walls and a high graphitic character (observed from TEM and Raman spectroscopy) of the pristine Baytubes C150P were mentioned. Furthermore, the determined contact angles in this work represent an average measure for

CNT/fluid-interactions how such local high-energy sides were not overestimated. Therefore, the enhanced fluid adsorption might also be driven by the delocalized aromatic π-system [38] of CNTs which possibly enables the forming of hydrogen bonds between the CNT π-electron cloud and an accessible hydrogen bridge donor [49,50]. In particular, the formation of very week C-H…π bonds [51] may explain the enhanced stacking of longer alkane or alkyl chains and hence, the observed wetting behavior. Besides, the generally superior wetting of nalcohols might be caused by O-H…π interactions which are stronger than C-H…π bonds [51]. Analogies for the formation of such weak hydrogen bonds as C-H…π bonds where for example, methyl structures acting as hydrogen-donor with aromatic molecule structures as it is known from organic chemistry, crystallography and structural biology [51]. The wetting of MWNTp with liquids of high surface tensions and/or polar components (compare chapter 2.2) is shown in figure 5 (b). The figure shows the partial wetting of polar fluids as well as the purely dispersive diiodomethane with a high surface tension by cosine contact angles between 0.239 for acetonitrile and maximal 0.528 for dimethyl sulfoxide. The figure implies an enhanced wetting with increasing surface tensions below 45 mNm-1 of the polar organic fluids and drops down to 0.389 for the wetting of the MWNTp with the polar inorganic water. If this is compared with the polar component of the surface tension (chapter 3.2), it implies a tendentially better wetting of the investigated organic fluids with an increasing polar fraction of the surface tension and a maximum liquid polarity that is combined with the best wetting of the drawn fluids. Additionally, a good wetting of the dispersive polydimethylsiloxane can be observed from figure 5b which is also better as of nalkanes or n-alcohols of similar surface tension. This might be caused by an enhanced adsorption of long molecules on the CNTs by a better π-stacking. This behavior intensifies that dispersive as well as polar components of the liquid surface tension determine the wetting behavior of the investigated MWNTs. The MWNT-wetting with the investigated polar fluids might also be determined by interacting of the fluid with the π system of the carbon nanotubes.

3.4

CNT-Wetting with Polar Fluids

The observed reasonable good wetting of MWNTs with polar fluids and particularly with water can also be found in the literature. For example A. H. Barber et al. [19] characterized the wetting behavior of one single MWNT based on an atomic force microscopy technique and determined a contact angle with water of 80.1° (cosΘ = 0.172). Moreover, a theoretical approach for the modeling of droplets on fibers from a thermodynamic point of view resulted

in a dependency of the contact angles on the fiber diameter. The contact angle Θ of water on CNTs was < 90° for CNT-diameters > 2.2 nm [27]. Also other carbon modifications as graphite and carbon fibers can show a good wetting from polar liquids [32]. In order to get insight of the behavior of polar fluids especially of water, CryoSEM analysis was conducted. Figure 6 shows two different magnifications of the frozen CNT/water-fracture plane. The homogenously filled area between the particles can be seen from figure (a). This indicates that a good water penetration through the powder and the agglomerates took place and no hollows were formed between the individual MWNTs. Figure (b) shows further that all the gaps between the CNTs are filled with the wetting fluid which indicates a good CNT/water-contact. Thus, the CryoSEM analysis verifies the measured and observed wetting behavior of polar fluids on the investigated MWNTs and hence, the good wetting of the MWNTp-material with polar fluids.

3.5

Free Surface Energy of Carbon Nanotubes

The surface energies have been derived from the contact angle measurements of diiodomethane as a purely dispersive liquid with high surface tension, dimethyl sulfoxide as fluid with dispersive and polar components and water with dominating polar properties. Thus, a wide range of possible particle/fluid-interactions could be considered. The contact angles with standard deviation and the solid surface energy with dispersive and polar component can be taken from table 5. The contact angles of the non-functionalized MWNTs are given as the mean from three and of the ethylamine-functionalized MWNTs as the mean of two single measurements. It can be taken from table 5 that the contact angle on MWNTp of diiodomethane is 66.7°, of dimethyl sulfoxide is 58.1° and of water is 67.1°. The contact angles of all three test liquids on the purified MWNThp with higher carbon purity (> 99 %) are slightly higher which means that impurities from the synthesis, e. g. catalyst, influence the wetting behavior slightly. Especially in the non-modified MWNTs, catalyst particles can remain on the carbon nanotubes surface [38]. In comparison to the non-functionalized MWNTs, the ethylaminefunctionalization led to a significant decrease of the wetting angle of the dimethyl sulfoxide down to 45.5° and water to 40.5° while the contact angle of diiodomethane is with 72.9° only slightly higher. In particular, the ethylamine molecules effect an enhanced contact to polar fluids while the wetting of the dispersive fluid of the MWNTs remains nearly unchanged and

hence, proves that the wetting of carbon nanotubes itself can be specifically altered by the covalent attachment of molecules and functional groups on the particle surface. The surface energy can be derived from the measured contact angles if the interfacial tension γsl regarding Owens and Wendt [30] as in equation (4) is combined with Young equation [52] as it was shown by Rabel [31] and Kaelble [53].

γ

 d d p p  = γ +γ - 2 γ * γ + γ * γ  sl lv sv lv sv   lv sv

(4)

The substitution results in OWRK-equation (5) which corresponds to a linear form whereby the surface energy can be derived graphically as it is shown in figure 7. Therein, the OWRKplots for the MWNTp (closed rhombuses), MWNThp (open rhombuses) and the MWNTethylamine (open dots) are drawn. The OWRK-plots for the pristine and purified MWNTs are similar due to the weak influence of the purification on the wetting. The higher slope of the linear regression line of the functionalized MWNTs is caused by the enhanced wetting of the polar test fluids. The data plots are in good agreement to the underlying model due to the reasonable good linear regressions of the data points how the solid surface energies could be determined.

1+ cosΘ  * γlv d 2* γlv

p = γsv *

p γlv d + γsv d γlv

(5)

The total surface energy of MWNTp was 35.97 mNm-1 with a dispersive and polar component of 19.85 and 16.12 mNm-1 respectively. The purified MWNThp showed similar surface energy and components that are negligible smaller than those of the unpurified material which indicates a slight influence of remaining local synthesis catalysts on the derived properties. However, the significant polar component in both powders results thereby from the reasonable good wetting of the non-functionalized MWNTs with polar liquids. Besides, A. H. Barber et al. [19] found from atomic force microscopy analysis of the wetting of an individual carbon nanotube in certain liquids similar OWRK-properties with a surface energy of 27.8 mNm-1, a dispersive component of 17.6 mNm-1 and a polar fraction of 10.2 mNm-1. The slight differences to this work may be caused by different test fluids (polydimehtylsiloxane, polyethylene glycol, glycerol, water). However, the analogies of the surface energies confirm the capillary penetration method for characterizing the CNT-wetting properties directly on the bulk material.

The surface energy in this work involves a dispersive and polar fraction. This might be caused by the delocalized π-system of the carbon nanotubes which influences the CNT/fluidinteraction. The dispersive fraction considers thereby the interaction of polarizable fluid molecules with the CNT-outer shell. The polar fraction contributes with the interactions between permanent dipoles and the π-system as well as the formation of week hydrogen bridges. The contribution of the different components on the CNT/fluid-interaction causes the complex interaction and wetting behavior and hence, the restricted CNT-dispersibility in fluids. Thereby, one might expect that a good CNT-dispersion could be achieved if the surface energy fractions of the fluid are similar to those of the carbon nanotube. The ethylamine-functionalized MWNTs revealed a significant higher total surface energy of 54.49 mNm-1, a similar dispersive fraction of 16.69 mNm-1 and an increased polar component of 37.80 mNm-1 if compared to its source material (MWNTp). This demonstrates an enormous influence of the CNT-functionalization on its surface properties despite that small degree of the ethylamine functionalization. It means that the interactions between carbon nanotubes and a liquid can be intensively controlled by a covalent CNT-modification. Particularly, the ethylamine functionalization causes a slight decrease of dispersive forces on the particle/fluidinteractions but a significant enhancement of the interdependency of the MWNTs to liquids with higher polarity.

3.6

MWNT-Surface Energy and Dispersibility

In order to analyze the derived surface energy of the MWNTs regarding their dispersibility the free surface energy was plotted as a function of dispersive and polar fractions. Figure 8 shows the area for the surface energy considered as a continuous function that illustrates the theoretical pairs of dispersive and polar fractions with the resulting surface energy. The estimated surface energies of the MWNTs are highlighted as dots where the non-modified MWNTs correspond to the labels (1) and (2) for the MWNTp and MWNThp respectively and hence, show similar surface properties. The surface energy of the ethylamine-functionalized MWNTs (3) shows their significantly changed surface state raised by their functional groups. Surface tensions of different liquids are included in the diagram and marked by the triangles (4-13). A good particle/fluid-interaction can be considered if the dispersive and polar component of the particle and the liquid are of similar magnitude. However, the investigated fluids show that only few liquids possess similar values as the MWNTs. Only N-methyl-2pyrrolidon (8) and NN-dimethylformamide (7) are closed to the non-functionalized MWNTs.

In correlation to the observed wetting behavior and surface energies, MWNTs were dispersed in certain liquids. The resulting dispersions were characterized regarding their decomposition behavior in a centrifugal field. Figure 9 shows therefore the transmission of dispersions in dependence of the sedimentation time. Particularly, the centrifugal sedimentation analysis of MWNTp-dispersions in water (open dots), acetone (open squares), ethanol (open triangles), ethylene glycol (filled dots), 1-octanol (filled triangles) and N-methyl-2-pyrrolidone (open rhombuses) are shown. The sharp increase of the transmission with an increasing sedimentation shows the fast decomposition of the dispersions of the MWNTp in water, acetone, ethanol, 1-octanol and ethylene glycol in the centrifugal field. This behavior indicates a poor CNT-dispersion and hence, CNT-exfoliation in those liquids. In comparison to that, the centrifugation of the MWNT-dispersion in N-methyl-2-pyrrolidone resulted in a significant smaller increase of the transmission. This slight increase of the transmission might be caused by the sedimentation of a smaller amount of remaining agglomerates and indicates further a higher degree of finely dispersed MWNTs. The enhanced dispersibility of the nonmodified MWNTs in N-methyl-2-pyrrolidon in contrast to the other fluids can be explained by the surface energies. The surface energy properties of that solvent reach nearly the surface state of the MWNTs whereas the other dispersing fluids show greater differences. Even if 1octanol is a good wetting fluid of the non-modified MWNTs, the CNT-dispersibility in this fluid is poor which can be referred to the polar fraction of the fluid that is obviously lower than of the pristine MWNTs. This observations lead to the general formulation that successful dispersions of non-functionalized CNTs can be expected for fluids with similar dispersive and polar components and thus, a total surface energy of the same magnitude as the particles. However, a good CNT-wetting which means a small contact angle cannot be directly referred to a good CNT-exfoliation as it is exemplarily shown from the unstable MWNTp-dispersion in ethylene glycol and 1-octanol.

4

Conclusion

The liquid penetration method could be used for the characterization of the wetting behavior of carbon nanotubes. In particular, contact angles of various fluids with dispersive and polar properties on different multiwall carbon nanotubes were reproducibly determined how the complexity of the CNT-wetting could be described. An enhanced wetting of carbon nanotubes with dispersive n-alkanes and slightly polar linear monovalent alcohols of high chain lengths as well as the reasonable good CNT-wetting with organic and inorganic polar fluids could be shown. This indicated the influence of induced dipole and permanent dipole interactions as

well as the formation of week hydrogen bonds between the fluid molecules and the delocalized π-system of the carbon nanotubes. The free surface energy with dispersive and polar fractions could be derived from the contact angles of fluids which possessed a wide range of dispersive and polar properties. The non-modified carbon nanotubes displayed a significant dispersive and polar fraction of the free surface energy which involves the observed and different kinds of CNT/fluid-interactions. Besides, the effect of a covalent functionalization on the CNT-wetting behavior could be proved by significantly changed contact angles and surface energies. In particular, an ethylamine-functionalization achieved an enhanced wetting of polar fluids and hence, a substantial increased polar component of the CNT surface energy. Furthermore, the CNT-surface energy with the dispersive and polar fraction correlated well with the carbon nanotubes dispersibility. A good CNT-dispersion could be achieved if the carbon nanotube and the dispersing fluid occupied dispersive and polar fractions and hence, total surface energies, of similar magnitudes since the CNT/fluid-interaction is in a suitable adjustment. However, due to the complex nature of CNTs, the wetting behavior which means the contact angle could not be directly referred to the CNT-dispersibility.

5

Acknowledgment

The authors would like to thank the Federal Ministry of Education and Research (BMBF) of Germany for the financial support of the underlying work and projects. We would particularly like to thank our project partner, the BASF SE in Ludwigshafen in Germany, for supplying and characterizing the ethylamine-functionalized carbon nanotubes.

6

References

[1] S. Iijima, Helical microtubules of graphitic carbon, Nature 354 (1991) 56–58. [2] De Volder, Michael F L, S.H. Tawfick, R.H. Baughman, A.J. Hart, Carbon nanotubes: present and future commercial applications, Science 339 (2013) 535–539. [3] B. Peng, M. Locascio, P. Zapol, S. Li, S.L. Mielke, G.C. Schatz, H.D. Espinosa, Measurements of near-ultimate strength for multiwalled carbon nanotubes and irradiation-induced crosslinking improvements, Nat Nanotechnol 3 (2008) 626–631. [4] B.Q. Wei, R. Vajtai, P.M. Ajayan, Reliability and current carrying capacity of carbon nanotubes, Appl. Phys. Lett. 79 (2001) 1172. [5] P. Kim, L. Shi, A. Majumdar, P. McEuen, Thermal Transport Measurements of Individual Multiwalled Nanotubes, Phys. Rev. Lett. 87 (2001). [6] C. Laurent, E. Flahaut, A. Peigney, The weight and density of carbon nanotubes versus the number of walls and diameter, Carbon 48 (2010) 2994–2996. [7] A. Jorio, G. Dresselhaus, M.S. Dresselhaus, Carbon nanotubes: Advanced topics in the synthesis, structure, properties, and applications, Springer, Berlin, New York, 2008. [8] J.N. Coleman, Liquid-Phase Exfoliation of Nanotubes and Graphene, Adv. Funct. Mater. 19 (2009) 3680–3695. [9] J. Hilding, E.A. Grulke, Zhang, Z. George, F. Lockwood, Dispersion of carbon nanotubes in liquids, J. Dispersion Sci. Technol. 24 (2003) 1–41. [10] L. Vaisman, H.D. Wagner, G. Marom, The role of surfactants in dispersion of carbon nanotubes, Advances in Colloid and Interface Science 128-130 (2006) 37–46. [11] R. Rastogi, R. Kaushal, S.K. Tripathi, A.L. Sharma, I. Kaur, L.M. Bharadwaj, Comparative study of carbon nanotube dispersion using surfactants, Journal of Colloid and Interface Science 328 (2008) 421–428. [12] V.C. Moore, M.S. Strano, E.H. Haroz, R.H. Hauge, R.E. Smalley, J. Schmidt, Y. Talmon, Individually Suspended Single-Walled Carbon Nanotubes in Various Surfactants, Nano Lett. 3 (2003) 1379–1382. [13] X.-L. Xie, Y.-W. Mai, Zhou, X., P., Dispersion and alignment of carbon nanotubes in polymer matrix: A review, Materials Science and Engineering: R: Reports 49 (2005) 89– 112. [14] H.T. Ham, Y.S. Choi, I.J. Chung, An explanation of dispersion states of single-walled carbon nanotubes in solvents and aqueous surfactant solutions using solubility parameters, Journal of Colloid and Interface Science 286 (2005) 216–223.

[15] C.M. Hansen, The three dimensional solubility parameter and solvent diffusion coefficient and their importance in surface coating formulation: Their importance in surface coating formulation, Danish Technical Press., Copenhagen, 1967. [16] S.D. Bergin, Z. Sun, D. Rickard, P.V. Streich, J.P. Hamilton, J.N. Coleman, Multicomponent Solubility Parameters for Single-Walled Carbon Nanotube−Solvent Mixtures, ACS Nano 3 (2009) 2340–2350. [17] A. Sobolkina, V. Mechtcherine, C. Bellmann, V. Khavrus, S. Oswald, S. Hampel, A. Leonhardt, Surface properties of CNTs and their interaction with silica, Journal of Colloid and Interface Science 413 (2014) 43–53. [18] Y.C. Hong, D.H. Shin, S.C. Cho, H.S. Uhm, Surface transformation of carbon nanotube powder into super-hydrophobic and measurement of wettability, Chemical Physics Letters 427 (2006) 390–393. [19] A. Barber, S. Cohen, H. Wagner, Static and Dynamic Wetting Measurements of Single Carbon Nanotubes, Phys. Rev. Lett. 92 (2004). [20] S. Nuriel, L. Liu, A.H. Barber, H.D. Wagner, Direct measurement of multiwall nanotube surface tension, Chemical Physics Letters 404 (2005) 263–266. [21] Erik Dujardin, Thomas W. Ebbesen, Ajit Krishnan, Michael M. J. Treacy, Wetting of Single Shell Carbon Nanotubes, Adv. Mater. 10 (1998) 1472–1475. [22] E. Dujardin, T.W. Ebbesen, H. Hiura, K. Tanigaki, Capillarity and wetting of carbon nanotubes, Science 265 (1994) 1850–1852. [23] T.W. Ebbesen, Wetting, filling and decorating carbon nanotubes, Journal of Physics and Chemistry of Solids 57 (1996) 951–955. [24] Y. Gogotsi, J.A. Libera, A. Güvenç-Yazicioglu, C.M. Megaridis, In situ multiphase fluid experiments in hydrothermal carbon nanotubes, Appl. Phys. Lett. 79 (2001) 1021. [25] D. Schebarchov, S.C. Hendy, Uptake and withdrawal of droplets from carbon nanotubes, Nanoscale 3 (2011) 134. [26] B.A. Kakade, V.K. Pillai, Tuning the Wetting Properties of Multiwalled Carbon Nanotubes by Surface Functionalization, J. Phys. Chem. C 112 (2008) 3183–3186. [27] A.V. Neimark, Thermodynamic equilibrium and stability of liquid films and droplets on fibers, Journal of Adhesion Science and Technology 13 (1999) 1137–1154. [28] F.M. Fowkes, Attractive Forces at Interfaces, Ind. Eng. Chem. 56 (1964) 40–52. [29] F.M. Fowkes, Additivity of Intermolecular Forces at Interfaces. I. Determination of the Contribution to Surface and Interfacial Tension of Dispersion Forces in Various Liquids, J. Phys. Chem. 67 (1963) 2538–2541.

[30] D.K. Owens, R.C. Wendt, Estimation of the surface free energy of polymers, J. Appl. Polym. Sci. 13 (1969) 1741–1747. [31] W. Rabel, Einige Aspekte der Benetzungstheorie und ihre Anwendung auf die Untersuchung und Veränderung der Oberflächeneigenschaften von Polymeren, Farbe und Lack 77 (1971) 997–1005. [32] D.H. Kaelble, P.J. Dynes, L. Maus, Surface Energy Analysis of Treated Graphite Fibers, The Journal of Adhesion 6 (1974) 239–258. [33] W.A. Zisman, Relation of the Equilibrium Contact Angle to Liquid and Solid Constitution, in: F.M. Fowkes (Ed.), Contact Angle, Wettability, and Adhesion, AMERICAN CHEMICAL SOCIETY, WASHINGTON, D.C, 1964, pp. 1–51. [34] R. Menzel, A. Bismarck, M.S. Shaffer, Deconvolution of the structural and chemical surface properties of carbon nanotubes by inverse gas chromatography, Carbon 50 (2012) 3416–3421. [35] E. Díaz, S. Ordóñez, A. Vega, Characterization of nanocarbons (nanotubes and nanofibers) by Inverse Gas Chromatography, J. Phys.: Conf. Ser. 61 (2007) 904–908. [36] U. Teipel (Ed.), Mikrostrukturen von Carbon Nanotubes (CNT), Fraunhofer Verlag, 2011. [37] C. Wigbers, Brinks, Marion, K., J.-P. Melder (BASF SE, Christof Wigbers, Marion K. Brinks, Johann-Peter Melder) WO 2012/156442 A1, 2012. [38] A. Krüger, Neue Kohlenstoffmaterialien: Eine Einführung, Teubner, Wiesbaden, 2007. [39] G. Ström, M. Fredriksson, P. Stenius, Contact angles, work of adhesion, and interfacial tensions at a dissolving Hydrocarbon surface, Journal of Colloid and Interface Science 119 (1987) 352–361. [40] J. Schultz, K. Tsutsumi, J.-B. Donnet, Surface properties of high-energy solids, Journal of Colloid and Interface Science 59 (1977) 277–282. [41] U. Teipel, I. Mikonsaari, Determining Contact Angles of Powders by Liquid Penetration, Part. Part. Syst. Charact. 21 (2004) 255–260. [42] E.W. Washburn, The Dynamics of Capillary Flow, Phys. Rev. 17 (1921) 273–283. [43] K. Grundke, A. Augsburg, On the determination of the surface energetics of porous polymer materials, Journal of Adhesion Science and Technology 14 (2000) 765–775. [44] A. Siebold, A. Walliser, M. Nardin, M. Oppliger, J. Schultz, Capillary Rise for Thermodynamic Characterization of Solid Particle Surface, Journal of Colloid and Interface Science 186 (1997) 60–70.

[45] K. Grundke, T. Bogumil, T. Gietzelt, H.-J. Jacobasch, D.Y. Kwok, A.W. Neumann, Wetting measurements on smooth, rough and porous solid surfaces, in: H.-J. Jacobasch (Ed.), Interfaces, Surfactants and Colloids in Engineering, Steinkopff, Darmstadt, 1996, pp. 58–68. [46] D. Lerche, T. Sobisch, Consolidation of concentrated dispersions of nano- and microparticles determined by analytical centrifugation, Powder Technology 174 (2007) 46–49. [47] K.A. Wepasnick, B.A. Smith, J.L. Bitter, D. Howard Fairbrother, Chemical and structural characterization of carbon nanotube surfaces, Analytical and bioanalytical chemistry 396 (2010) 1003–1014. [48] J.-P. Tessonnier, D. Rosenthal, T.W. Hansen, C. Hess, M.E. Schuster, R. Blume, F. Girgsdies, N. Pfänder, O. Timpe, D.S. Su, R. Schlögl, Analysis of the structure and chemical properties of some commercial carbon nanostructures, Carbon 47 (2009) 1779– 1798. [49] C.A. Hunter, J.K.M. Sanders, The nature of .pi.-.pi. interactions, J. Am. Chem. Soc. 112 (1990) 5525–5534. [50] T. Steiner, Die Wasserstoffbrücke im Festkörper, Angew. Chem. (2002) 50–80. [51] G.R. Desiraju, Hydrogen Bridges in Crystal Engineering: Interactions without Borders, Acc. Chem. Res. 35 (2002) 565–573. [52] T. Young, An Essay on the Cohesion of Fluids, Philosophical Transactions of the Royal Society of London 95 (1805) 65–87. [53] D.H. Kaelble, Dispersion-Polar Surface Tension Properties of Organic Solids, The Journal of Adhesion 2 (1970) 66–81.

7

Glossary

carbon nanotubes contact angle

tubular carbon particles with graphitic shells angle formed by a fluid drop on a solid surface between the liquid and the surface

dispersion

distribution of at least an insoluble condensed phases in a

continuous matrix dispersing

separation and homogeneous distribution of particles in a

continuous phase dispersibility functionalization

assertion about the dispersing behavior of a material physical or covalent attachment of functional groups or

molecules liquid penetration method

characterization technique for the wetting behavior of

disperse systems surface energy

measure for the energy which is required to enlarge the surface of a liquid or solid

wetting

behavior of fluids on solid surfaces

8

List Of Figures Figure 1: Scheme of the capillary penetration method for the characterization of the wetting behavior of disperse systems Figure 2: Measuring assembly of the analytical centrifuge for the characterization of the particle sedimentation behavior: light source (1), cuvette including the sample (2), detector (3), space- and time- resolved extinction profiles (4-5) [46] Figure 3: Liquid penetration curves and reproducibility from the wetting of MWNTp with ethylene glycol Figure 4: Wetting curves with liner regression lines, slope and correlation coefficient of liquids on MWNTp for the comparison of capillary constants and determination of the surface energy Figure 5: Wetting of MWNTp with n-alkanes/n-alcohols (a) and with polar and dispersive fluids of high surface tensions (b) Figure 6: Cryo scanning electron microscopy images of MWNTp after the wetting of the powder with water Figure 7: OWRK-plot for the different MWNT-materials for the wetting with diiodomethane, dimethyl sulfoxide and water to determine the solid surface energy Figure 8: Free surface energy with dispersive and polar fractions of different MWNTs and liquids regarding the OWRK-model [30] Figure 9: Transmission of MWNTp-dispersions of different liquids in dependence of the sedimentation time in an analytical centrifuge

9

List Of Tables

Table 1:

Surface energy characterization of non-functionalized carbon nanotubes given in literature

Table 2:

Carbon nanotube-materials and specification

Table 3:

Properties of wetting and dispersing fluids

Table 4:

Capillary constants of different liquids on MWNTp

Table 5:

Contact angels of test fluids and free surface energy with dispersive and polar components of the carbon nanotubes regarding the OWRK-model [30]

FIGURE 1

FIGURE 2

FIGURE 3

FIGURE 4(a)

FIGURE 4(b)

FIGURE 5(a)

FIGURE 5(b)

FIGURE 6(a)

FIGURE 6(b)

FIGURE 7

FIGURE 8

FIGURE 9

Table 1:

Surface energy characterization of non-functionalized carbon nanotubes given in literature

Method for the contact angle characterization

AFM-method SEM-images of polymer droplets on CNTs (nanocomposites)

CNTs

MWN T MWN T

Sessile drop method on

MWN

CNT-cushions

T

TEM-images of melted

SWN

metal droplets on CNTs

T

γ sv

d γ sv

p γ sv

[mNm-

[mNm-

[mNm-

1]

1]

1]

OWRK

27.8

17.6

10.2

[19]

OWRK

45.3

18.4

26.9

[20]

OWRK

82.6

4.8

77.8

[18]

Zisman

40-80

Model

Ref.

[21]

Table 2:

Carbon nanotube-materials and specification

CNT-Material

Carbon

Functionalization

Purity1) MWNTp

(Baytubes® > 95 %

None

(Baytubes® > 99 %

None

C150P) MWNThp C150HP) MWNT-ethylamine

Not specified

ethylamine < 1 % N, 1.8 % O (Elementary analysis)1) 0.1 at-% N, 0.9-at-% O(XPS)1)

1)

Data and characterization of the CNT-materials from the producer and/or project partner

BASF

Table 3:

Properties of wetting and dispersing fluids [mNm-

d γlv [mNm-

p γlv

1

1

[mNm-

ρl

ηl

γlv

[gcm-3]

[mPas]

1

n-hexane

0.661

0.326

18.4

18.4

0

2)

n-heptane

0.684

0.409

20.4

20.4

0

2)

n-octane

0.703

0.542

21.8

21.8

0

2)

n-decane

0.73

0.929

23.9

23.9

0

[39]

n-dodecane

0.749

1.35

25.4

25.4

0

[39]

n-hexadecane

0.773

3.34

27.6

27.6

0

[39]

methanol

0.792

0.577

22.7

16

6.7

2)

ethanol

0.789

1.162

22.1

17.5

4.6

2)

1-butanol

0.81

2.544

24.93

4)

4)

2)

1-hexanol

0.814

4.578

25.81

4)

4)

2)

1-octanol

0.827

9.12

27.6

21.3

6.3

[40]

ethylene glycol

1.11

21.81

48

29

19

2)

polydimethylsiloxane 0.9

2.79

19.4

19.4

0

2) 3)

acetone

0.79

0.322

23.3

16.5

6.8

acetonitrile

0.78

0.316

28.66

4)

4)

dimethyl sulfoxide

0.5

0.5

44

36

8

2)

N-methyl-2-

1.023

1.68

40.79

29.21

11.58

2)

dimethylformamide

0.949

0.899

37.1

29

8.1

2)

diiodomethane

3.325

2.762

50.8

50.8

0

[39]

water

0.998

1.002

72.8

26

46,8

[39]

Fluid

]

]

Reference

]

pyrrolidone

2)

Database Tensiometer K100, Krüss GmbH, 3)Specification sheet 19.05.2011, 4106 Silikonöl

M3, Carl Roth, 4)unknown

Table 4:

Capillary constants of different liquids on MWNTp

Fluid



[cm ]

n-hexane

3.77x10-5

8.94x10-7

n-heptane

4.07x10-5

3.07x10-6

polydimethylsiloxane

6.06x10-5

5.74x10-6

ethylene glycol

1.21x10-4

2.35x10-6

diiodomethane

4.80x10-5

2.48x10-6

dimethyl sulfoxide

6.44x10-5

7.26x10-7

water

4.62x10-5

2.43x10-6

A2 * c * r 

5

s [cm5]

Table 5:

Contact angels of test fluids and free surface energy with dispersive and polar components of the carbon nanotubes regarding the OWRK-model [30]

Θ±s

MWNTp

MWNThp

MWNT-ethylamine

66.7° ± 1.27°

70.7° ± 2.29°

72.9° ± 0.57°

58.1° ± 0.43°

66.1° ± 1.60°

45.5° ± 8.06°

67.1° ± 1.24°

75.0° ± 0.65°

40.5° ± 0.14°

]

35.97

30.20

54.49

d -1 γsv [mNm ]

19.85

17.94

16.69

p -1 γsv [mNm ]

16.12

12.26

37.80

diiodomethane dimethyl sulfoxide water γsv [mNm

-1