Experimental Data Contributing to the Elusive Surface Charge of Inert Materials in Contact with Aqueous Media

colloids and interfaces

Article Experimental Data Contributing to the Elusive Surface Charge of Inert Materials in Contact with Aqueous Media

Antun Barišić 1, Johannes Lützenkirchen 2,* , Nikol Bebić 1, Qinzhi Li 3, Khalil Hanna 3 , Andrey Shchukarev 4 and Tajana Begović 1,*

1 Department of Chemistry, Faculty of Science, University of Zagreb, Horvatovac 102A, HR-10000 Zagreb, Croatia; [email protected] (A.B.); [email protected] (N.B.) 2 Institut für Nukleare Entsorgung, Karlsruher Institut für Technologie, P.O. Box 3640, DE-76021 Karlsruhe, Germany 3 Ecole Nationale Supérieure de Chimie de Rennes, Université de Rennes 1, CNRS, UMR 6226, F-35000 Rennes, France; [email protected] (Q.L.); [email protected] (K.H.) 4 Department of Chemistry, Umeå University, SE-90187 Umeå, Sweden; [email protected] * Correspondence: [email protected] (J.L.); [email protected] (T.B.); Tel.: +49-721-608-2-4023 (J.L.); +385-1-4606-154 (T.B.)

Abstract: We studied the charging of inert surfaces (polytetrafluoroethylene, i.e., PTFE; graphite; graphene; and hydrophobic silica) using classical colloid chemistry approaches. Potentiometric titrations showed that these surfaces acquired less charge from proton-related reactions than oxide minerals. The data from batch-type titrations for PTFE powder did not show an effect of ionic strength, which was also in contrast with results for classical colloids. In agreement with classical colloids, the electrokinetic results for inert surfaces showed the typical salt level dependence. In some cases, the point of zero net proton charge as determined from mass and tentatively from acid–base titration differed from isoelectric points, which has also been previously observed, for example by Citation: Barišić,A.; Lützenkirchen, Chibowski and co-workers for ice electrolyte interfaces. Finally, we found no evidence for surface J.; Bebić,N.; Li, Q.; Hanna, K.; contaminations of our PTFE particles before and after immersion in aqueous solutions. Only in the Shchukarev, A.; Begović,T. presence of NaCl-containing solutions did cryo-XPS detect oxygen from water. We believe that our Experimental Data Contributing to low isoelectric points for PTFE were not due to impurities. Moreover, the measured buffering at pH 3 the Elusive Surface Charge of Inert could not be explained by sub-micromolar concentrations of contaminants. The most comprehensive Materials in Contact with Aqueous Media. Colloids Interfaces 2021, 5, 6. explanation for the various sets of data is that hydroxide ion accumulation occurred at the interfaces https://doi.org/10.3390/ between inert surfaces and aqueous solutions. colloids5010006 Keywords: surface charge; zeta-potential; PTFE; graphite; graphene Received: 11 December 2020 Accepted: 20 January 2021 Published: 23 January 2021 1. Introduction Publisher’s Note: MDPI stays neutral Inert surfaces, i.e., surfaces that do not expose any surface functional groups to the with regard to jurisdictional claims in surrounding electrolyte solution, develop a pH-dependent charge, the origin of which published maps and institutional affil- has been at the center of some fierce debates [1,2]. The issue is also related to the inert iations. gas/electrolyte aqueous solution interfaces (Ray–Jones effect [3]), i.e., the observation of a minimum of the surface tension at millimolar concentration (in terms of both salt and acid-concentration), which has been debated since almost a century [4]. On the inert surfaces, the actual charge, its pH-dependence, and the occurrence of isoelectric points Copyright: © 2021 by the authors. in the pH range 2–4 are hardly contested [5], but amongst potential others the following Licensee MDPI, Basel, Switzerland. explanations for the origin of these observations have been put forward: This article is an open access article • Preferential adsorption of hydroxide ions with respect to protons [6]; distributed under the terms and • Charge transfer in interfacial water [7]; conditions of the Creative Commons • Attribution (CC BY) license (https:// Dissolved inorganic carbon as a source of contamination [8]; creativecommons.org/licenses/by/ • Dissolved charged surface-active contaminants [9]. 4.0/).

Colloids Interfaces 2021, 5, 6. https://doi.org/10.3390/colloids5010006 https://www.mdpi.com/journal/colloids Colloids Interfaces 2021, 5, 6 2 of 19

Unfortunately, the various features (and in recent years in particular the question of contaminant being present or not) have been debated in similar ways as the origin of the Ray–Jones effect. Thus, the roles of organic contaminations [10] or inorganic ions [11] have been refuted. Yet, the surface-active contaminations, when considered as a mixture of molecules with low and high pKa values at sub-micromolar concentrations, appear to be able to explain the charging of inert surfaces [11]. However, from our point of view, many of the attempts to explain the unexpected observations address selected aspects, such as zeta- potential measurements, but will probably be unable to yield a comprehensive explanation including other observations. For example, the trace concentrations of surface-active agents would not be able to explain the buffering effect of polytetrafluorethylene (PTFE), as well as other inert surfaces, at pH values as low as 3. A simple experiment involving the addition of powders of hydrophobic material into MilliQ water under inert gas atmosphere lowers the pH to values down to pH 3 if sufficient material is added. In the absence of a better explanation, we believe that currently this buffering can only be explained by involving the adsorption of the water ions (i.e., in particular the hydroxide ions, which lowers the pH). Overall, this explanation still appears to be able to phenomenologically account for all observations. As another example, the charge separation of ultrapure water droplets on cleaned PTFE surfaces in the absence of carbon dioxide, resulting in charged water droplets and charged PTFE surfaces, requires charge carriers [12]. The purest water used in the cited study still had a higher conductivity than ideal water, but the purer the water, the higher the charge on the water droplet [12]. Such charge separation has been related to electrets and the water ions again have been a preferred interpretation of charge separation on all kinds of surfaces that we experience in dry environments sometimes [13]. Yet the dilemma remains that even for the idea of preferential adsorption or excess accumulation of hydroxide ions at inert surfaces—which in our opinion explains both the buffering and the charge separation—an excessive affinity of named solute must be accepted. This strong affinity is in contradiction with many advanced molecular simula- tions [14,15], some of which suggest that the proton is favored at the interfaces of interest. In view of the contradictory conclusions, it is not unlikely that the four above-mentioned causes (and maybe others) are all simultaneously at play. The literature on the various topics is exhaustive and extending year by year [16–18], and it would be futile to even try and discuss all the information that is being accumulated. Instead, we focus in the next section of this introduction on how Emil Chibowski has been contributing to this field. His interest in contact angles [19–21], surface energies [22–24], and electrokinetics [25–27] brought him into contact with discussions about the origin of the charge that was also measured in electrolyte solutions on, for example, oil droplets [26–28], paraffin [25], or ice [29,30]. One conclusion from many of the published papers was that protons and hydroxide ions are charge-creating solutes, while hydroxide ions determine the zeta-potential. However, due to the small charge observed in many cases, this also opened the discussion about oriented water dipoles being responsible for the determined zeta-potentials [25]. More precisely, Chibowski with his coworkers showed that near hydrophobic surface potential may be created by immobilized and oriented water dipoles. Indeed, dipoles are mentioned in Hunter’s book on electrokinetics at a very early point [31] but are ignored subsequently because they cannot outcompete charged surfaces. In the present work, we report experimental work based on traditional colloid-chemical techniques. We investigated various inert surfaces in monovalent electrolyte (NaCl) solutions. The purpose was to contribute additional information to the discussion about the origin of the charge of inert surfaces. Such results need to be considered for a comprehensive understanding of the charging of surfaces that do not have intrinsic functional groups that could generate charge in the classical sense. More specifically, we present data obtained with different techniques (electrokinetics as well as potentiometric acid–base and mass titrations) and from different laboratories on different inert surfaces (PTFE, graphite, graphene, and hydrophobic silica) that all confirm previous findings with respect to low isoelectric points of inert materials. We took Colloids Interfaces 2021, 5, 6 3 of 19

great care to exclude carbon dioxide, which particularly for the potentiometric titration part is crucial. Moreover, mass titrations with powder with increasing mass content (γ) expose increasingly high surface area, which should make the experiment less prone to contamination effects (unless the powder has these contaminations adsorbed when immersed). To test the role of such contaminants, we carried out XPS measurements on the bare powder and after contact with deﬁned solutions. With the results obtained in this study on different solids and involving several laboratories, we continue to prioritize the interpretation of hydroxide ion adsorption with potential contributions from water dipoles that could even be the reason for the hydroxide ion afﬁnity. The main reason for this interpretation is that it explains the pH dependence, the buffering, and the low IEP (isoelectric point)at the same time. We doubt that on the basis of our results with high concentrations of powder, which we show is not contaminated before and after experiments (within what is detectable by XPS), that contaminants at low concentrations can cause pH values as low as 3, which we observed in suspensions of inert particles.

2. Materials and Methods 2.1. Materials and Methods at KIT Various inert materials were used. 1. Hydrophobic silica was obtained from Wacker. The brand HDK H20 (courtesy Wacker, referred to as HDK-H20 in this article) has a specific surface area of 127 m2/g (measured as all the other samples used at KIT (Karlsruhe Institute of Technology) by BET N2 (Brunauer, Emmett, and Teller (BET) method, Micromeritics, Gemini, employing liquid nitrogen after outgassing at 120 ◦C for several hours) and was used as a powder (HDK-H20P) or prepared as an aqueous suspension that was continuously stirred (HDK-H20S). 2. A sample of Hypersep/Hpercarb (Hypersep PGC, referred to as H-PGC in this article) was obtained from Thermo Scientific and had a specific surface area of 121 m2/g. According to the manufacturer, this material consists of 30 µm spherical particles of 100% porous (with no micropores) graphitic carbon, composed of flat sheets of hexagonally arranged carbon atoms with a fully satisfied valence. 3. The PTFE samples used were the same as in a previous paper [32]. PTFE Beads Microdispers-200 powder (referred to as PTFE-B-200 in the following) was obtained from Polysciences Inc. and its specific surface area was measured to be s ≈ 6 m2/g, i.e., a value slightly different from what was previously obtained with another charge of the product [32]. Flat plates were cut from a conventional PTFE sample to fit the measurement cell for streaming current/potential (20 × 10 mm). The NaCl solutions were prepared from dried salt. Solutions containing HCl and NaOH were used to adjust the pH. All experiments were performed at room temperature using MilliQ-water (18.2 µOhm/cm) to prepare the solutions under Ar atmosphere. Continuous potentiometric titrations were carried out using a Metrohm Titrando 907 controlled by the Tiamo software. A known mass of powder or volume of stock suspension was inserted into a titration vessel. The final known total liquid volume (50 mL) contained known amounts of background electrolyte salt. The pH was initially increased by adding a known volume of standardized NaOH. The acid-titration was started after 10 h stirring of the suspension under purified argon atmosphere. Known volumes of the HCl titrant of known concentration were added to the suspension under continuous stirring and in argon atmosphere, and the pH electrode readings were recorded as a function of titrant addition. Small volumes and waiting times between additions were applied. The surface area exposed in the titrations was at least 10 m2. The set-up was calibrated before each titration using an electrolyte solution of the same electrolyte medium and titrating with the same titrant solution. The relative surface charge density due to proton reactions was then calculated from the raw data. With the PTFE powder an alternative method was used, which consisted of preparing separate batches of known amount of powder in known volumes of solutions with constant background electrolyte and known amounts of acid (or Colloids Interfaces 2021, 5, 6 4 of 19

in some few cases base) added. All manipulations were carried out under puriﬁed argon and the closed tubes were then put on a shaker. After 2 days, the pH of the suspension was measured under Ar atmosphere and the surface charge calculated. The pH electrode was calibrated on a daily basis using at least 4 buffer solutions. Zeta-potential measurements were done using a Brookhaven Zeta-PALS set-up. The pH was measured as described in the PTFE batch titration procedure. Streaming potential/current measurements on PTFE flat surfaces were performed using the adjustable gap cell of the Anton Paar SurPass apparatus. The measurements were conducted with a maximum pressure of 300 mbar and a gap height of approximately 100 µm. The pH electrode and conductivity meter were calibrated using the SurPass software.

2.2. Materials and Methods at the University of Zagreb Measurements at the Zagreb laboratory were conducted on flat surfaces (PTFE, graphite, and graphene) and colloidal particles (graphite and carbon nanotubes (CNTs)). • Graphite: Graphite flat surfaces were prepared from a graphite rod (d = 6.15 mm, 99.9995%) and polished before measurements while synthetic graphite powder (99% purity, Alfa Aesar, Haverhill, MA, USA) was used as colloidal graphite particles. • Graphene: Substrates exposing monolayer graphene on quartz supports were bought from Graphenea (Dimensions: 10 mm × 10 mm, Coverage: 97%). • Carbon nanotubes (CNT): MWCNT (d = 8–15 nm, l = 10–50 µm, Timesnano, Chengdu, China) were used. The specific surface area of the colloidal particles was determined by the Brunauer, Emmett, and Teller (BET) method (Micromeritics, Gemini) using liquid nitrogen. Prior to the determination of surface area, powders were outgassed at 150 ◦C for 2 h. The specific surface areas were determined to be s = 18.7 ± 0.1 m2g−1 for graphite particles and s = 163.2 ± 0.7 m2g−1 for CNT particles. All experiments were performed at 25 ◦C in inert atmosphere (argon). The pH of the solutions was monitored, in all experiments, by a combined glass electrode with Ag/AgCl reference electrode. The combined glass electrode was calibrated at 25 ◦C using standard buffer solutions (pHbuff = 3, 7, 10) before each experiment. The point of zero charge (pHpzc) of the colloidal particles was determined by potentiometric mass titration. Dry colloidal particles (graphite or CNT) were added into aqueous solution containing 2 mmol dm−3 of NaCl. Between each addition, ultrasound was used to break aggregates and pH was measured. The colloidal particles were added into the solution until constant pH∞ = pHpzc was obtained [33]. The proton-related surface charge density of CNT as a function of pH was determined by means of potentiometric acid–base titration. For the blank titration, an aqueous elec- −3 trolyte solution (Ic = 2 mmol dm , pH0 = 3) was prepared by adding NaCl and HCl in 10 mL of water. The solution was titrated with NaOH and after each addition the pH was measured. A blank titration was conducted until pH > 9. Afterwards a suspension of carbon nanotubes (γ = 5 g cm−3) was prepared by adding 50 mg of carbon nanotubes to 10 mL of an aqueous electrolyte solution of the same composition as in the blank titration. After each addition of NaOH, the pH of the suspension was measured, and between measurements the suspension was treated with an ultrasound probe for 1 min. Again, potentiometric titration of CNT suspension was conducted until pH > 9. The electrophoretic mobilities of the colloidal particles and their average hydrodynamic diameters were obtained in parallel experiments on a Brookhaven 90Plus/BI-MAS. Initial suspensions were prepared by dispersing colloidal particles (γ(graphite) = 0.4 g dm−3, γ(CNT) = 0.05 g dm−3) in aqueous solution, which contained 1 mmol dm–3 NaCl and −3 NaOH (pH0 ≈ 11, Ic = 2 mmol dm ). The prepared suspensions were titrated with HCl(aq) solution. Electrophoretic mobilities at each pH are reported as average values of 10 runs. The average hydrodynamic diameters of the graphite colloidal particles were determined with dynamic light scattering measurements simultaneously with the electrophoretic measurements (results are presented in Appendix A.2). Colloids Interfaces 2021, 5, x FOR PEER REVIEW 5 of 19

solution. Electrophoretic mobilities at each pH are reported as average values of 10 runs.

Colloids Interfaces 2021, 5, 6 The average hydrodynamic diameters of the graphite colloidal particles were5 ofdetermined 19 with dynamic light scattering measurements simultaneously with the electrophoretic measurements (results are presented in appendix A.2). The electrokinetic potential for the flat surfaces (PTFE, graphite, and graphene) was Thedetermined electrokinetic using potential the streaming for the flat potential surfaces te (PTFE,chnique graphite, on the andSurPass graphene) apparatus was (Anton determinedPaar). using Flat thesurfaces streaming were fixed potential to the technique self-made on epoxy the SurPass carriers apparatusand were (Antonpretreated with Paar). Flatan surfacesaqueous weresolution fixed that to thecontained self-made 1 mmol epoxy dm carriers−3 NaCl and and were 1 mmol pretreated dm−3 with NaOH. The −3 −3 an aqueousNaCl/NaOH solution thatsolution contained (pH0 ≈ 111, mmol Ic = 2 dm mmolNaCl dm−3) and was 1 titrated mmol dmwith HCl(aq).NaOH. TheThe stream- −3 NaCl/NaOHing potential solution gap (pH cell0 ≈ width11, Ic was= 2 adjusted mmol dm to 100) was μm titrated while maximum with HCl(aq). cell pressure The was streaming200 potential mbar. gap cell width was adjusted to 100 µm while maximum cell pressure was 200 mbar. 2.3. Materials and Methods at the University of Umeå 2.3. Materials and Methods at the University of Umeå XPS spectra of the PTFE powders were recorded with a Kratos Axis Ultra DLD elec- XPS spectra of the PTFE powders were recorded with a Kratos Axis Ultra DLD tron spectrometer using a monochromated Al K-alpha source operated at 150 W, a hybrid electron spectrometer using a monochromated Al K-alpha source operated at 150 W, a lens system with magnetic lens providing an analysis area of 0.3 × 0.7 mm2, and a charge hybrid lens system with magnetic lens providing an analysis area of 0.3 × 0.7 mm2, and neutralizer. The binding energy scale was referenced to the C 1s line of Teflon, set at 292.5 a charge neutralizer. The binding energy scale was referenced to the C 1s line of Teflon, eV [34]. Processing of the spectra was accomplished with the Kratos software. Accuracy set at 292.5 eV [34]. Processing of the spectra was accomplished with the Kratos software. in BE determination was 0.1 eV, and in atomic ratio—8–10% rel. Data for the bare PTFE Accuracy in BE determination was 0.1 eV, and in atomic ratio—8–10% rel. Data for the bare powder were acquired at room temperature using the conventional XPS procedure. For PTFE powder were acquired at room temperature using the conventional XPS procedure. the PTFE samples equilibrated in aqueous solutions, we performed XPS measurements For the PTFE samples equilibrated in aqueous solutions, we performed XPS measurements under liquid nitrogen cooling using the fast-freezing sample preparation technique. The under liquid nitrogen cooling using the fast-freezing sample preparation technique. The fast-freezing protocol for cryo-XPS has been described in detail elsewhere [35]. fast-freezing protocol for cryo-XPS has been described in detail elsewhere [35].

3. Results3. Results and Discussion and Discussion 3.1. Potentiometric3.1. Potentiometric Acid–Base Acid–Base and Mass and Titration Mass Titration Figure1 Figureshows the1 shows potentiometric the potentiometric titration resultstitration for results PTFE-B-200, for PTFE-B-200, which were which con- were consistent atsistent low pH at betweenlow pH between results from results mass from titrations mass titrations and independent and independent potentiometric potentiometric (batch) titrations.(batch) titrations. The mass The titration mass titration results setresults a point set ofa point zero netof zero proton net chargeproton at charge pH 3. at pH 3. The dataThe coincided data coincided at the low at pH,the low irrespective pH, irrespective of whether of whether acid was acid added was or added not. With or not. With increaseincrease in pH, more in pH, negative more negative fundamental fundamental charge accumulated charge accumulated at the interfaces. at the interfaces. Unlike Unlike what wouldwhat be would observed be observed for an oxide for an surface, oxide thesurface, proton-related the proton-related surface chargesurface did charge not did not depend ondepend the salt on levelthe salt for level the PTFE for the sample. PTFE sample.

10 0 3456789 -10 pH -20 NaOH addition, 100 mM NaCl

2 -30 NaOH addition, 1 mM NaCl − MT (no HCl addition), 10 mM NaCl -40 MT (HCl addition), 10 mM NaCl MT (HCl addition), 100 mM NaCl -50 MT (no HCl added), 10 mM NaCl / mC m / mC H σ -60 -70 -80 -90 -100 -110

Figure 1. Proton-relatedFigure 1.surfaceProton-related charge density surface chargeof polytetraflu density oforethylene polytetraﬂuorethylene (PTFE)-B-200 (PTFE)-B-200obtained by obtainedmass titrations by (▲,,♦,) and independentmass titrations potentiometric (N,5,,/) and (batch) independent titrations potentiometric (■,●) in sodium (batch) chloride titrations aqueous (, •solution,) in sodium θ = chloride25 °C. aqueous solution, θ = 25 ◦C.

Proton-related surface charge densities of CNT obtained from continuous potentiometric acid–base titration (see AppendixA, Figures A1–A3) are presented in Figure2. The point of zero charge of CNT in sodium chloride aqueous solution was found to be at Colloids Interfaces 2021, 5, x FOR PEER REVIEW 6 of 19

Colloids Interfaces 2021, 5, 6 Proton-related surface charge densities of CNT obtained from continuous potentiom-6 of 19 etric acid–base titration (see Appendix A, Figures A1–A3) are presented in Figure 2. The point of zero charge of CNT in sodium chloride aqueous solution was found to be at pHpzc

=pH 4. Thepzc = values 4. The of values the proton-related of the proton-related surface surfacecharge density charge densityin the basic in the region basic were region higher were thanhigher the than values the obtained values obtained for PTFE for particles. PTFE particles.

100

0 456789 pH 2 − -100 / mC m / mC

H -200 σ

-300

-400

FigureFigure 2. 2. SurfaceSurface charge charge of of carbon carbon nanotube nanotube (CNT) (CNT) obta obtainedined by by potentiometric potentiometric acid–base acid–base titrations titrations inin sodium sodium chloride chloride aqueous aqueous solution. solution. Lines Lines repres representent third third order order polynomial polynomial fit ﬁt used used to to determine determine pHpzc. Ic = 2 mmol dm−3, −θ3 = 25 °C. ◦ pHpzc. Ic = 2 mmol dm , θ = 25 C.

FigureFigure 3 shows that the titrations titrations of the the hydrophobic hydrophobic surfaces, graphite (H-PGC), and hydrophobichydrophobic silica (HDK-H20P)(HDK-H20P) yieldedyielded much much less less proton proton release release than than a typical a typical hydrophilic hydro- philicsurface surface (HDK-H20S). (HDK-H20S). The latter The waslatter obtained was obtained from HDK-H20Pfrom HDK-H20P by suspension by suspension in MilliQ in MilliQwater andwater agitating and agitating for weeks. for weeks. Initially, Initially, the powder the powder (added (added at high at solid high concentration)solid concentration)was sticking was sticking to the air–water to the air–water interfaces, interfaces, and thus and the containerthus the container could not could even be not ﬁlled even with be filledwater. with However, water. withHowever, time, thewith material time, the must material have turnedmust have (at least turned partially) (at least hydrophilic, partially) hydrophilic,because the particlesbecause the became particles very became well dispersed very well (see dispersed Appendix (seeB, FigureAppendix A5), B, and Figure thus A5),the missingand thus volume the missing of water volume could of be water added. coul It wasd be observedadded. It thatwas this observed hydrophilic that this sample hy- drophilicbehaved likesample conventional behaved like silica. conventional The comparison silica. to The the hydrophobiccomparison to surfaces the hydrophobic shows that surfacesthe deprotonation shows that ofthe the deprotonation latter was clearly of the weaker latter was than clearly on oxidic weaker surfaces. than on Furthermore, oxidic surfaces.with bothFurthermore, hydrophobic with surfaces,both hydrophobic we observed surfaces, (i) positive we observed charges (i) at positive the lowest charges pH,i.e., at thebelow lowest pH pH, 4.5, andi.e., below (ii) changes pH 4.5, in and slope (ii) that changes are uncommon in slope that and are were uncommon not observed and inwere the notbatch observed titrations in the of batch the PTFE titrations particles. of the The PTFE continuous particles. titrationsThe continuous yielded titrations relative yielded surface relativecharges. surface The data charges. in Figure The 3data were in translatedFigure 3 were to absolute translated charge to absolute supposing charge that supposing the bare thatparticles the bare did particles not introduce did not any introduce charge, and any since charge, we knowand since the amountswe know of the titrants amounts added of titrantsto the system added forto the each system point, for we each can point, assume we that can the assume relative that charge the relative corresponded charge corre- to the spondedabsolute to charge the absolute for this case.charge for this case.

ColloidsColloids Interfaces Interfaces2021 2021, 5, ,5 6, x FOR PEER REVIEW 7 of 719 of 19

HDK-H20P HDK-H20S H-PGC 0 3456789

-100 pH

-200 2 −

-300 / mC m / mC H σ -400

-500

-600

Figure 3. Surface charge of graphite (▲, H-PGC), hydrophobic silica (■, HDK-H20P), and hydro- Figure 3. Surface charge of graphite (N, H-PGC), hydrophobic silica (, HDK-H20P), and hydrophilic philic silica (●, HDK-H20S) obtained by mass titrations in sodium chloride aqueous (Ic = 1 mmol silica (•, HDK-H20S) obtained by mass titrations in sodium chloride aqueous (I = 1 mmol dm−3) dm−3) solution, θ = 25 °C. c solution, θ = 25 ◦C. When the stirring of the HDK-H20S sample was stopped, part of the material floated When the stirring of the HDK-H20S sample was stopped, part of the material floated immediately on the surface of the suspension (see Appendix B, Figure A6). Stirring again immediately on the surface of the suspension (see AppendixB, Figure A6). Stirring again immersed all the material again into the suspension, which did not show any phase seg- immersed all the material again into the suspension, which did not show any phase segre- regation under the mechanical treatment. The observations that the hydrophobic HDK- gation under the mechanical treatment. The observations that the hydrophobic HDK-H20P H20P particles at some point become “hydrophilic” under continuous mixing with water particles(yielding atHDK-H20S) some point could become have “hydrophilic” been due to various under continuousissues. Among mixing other with potential water rea- (yield- sons,ing HDK-H20S) this observation could may have be related been due to the to variousnon-equilibrium issues. Amongnature of other the systems. potential In reasons,such thermodynamicallythis observation may unstable be related systems to the (particle non-equilibrium suspensions/dispersions), nature of the systems.kinetics (here, In such forthermodynamically example, affected unstable by the mixing systems time) (particle may influence suspensions/dispersions), the state of dispersions/suspen- kinetics (here, for sionexample, of initially affected hydrophobic by the mixing particles. time) Furthermore, may influence HDK-H20P the state particles of dispersions/suspension could be, at least toof some initially extent, hydrophobic water-dispersible, particles. but Furthermore, require some HDK-H20P specific mixing particles time could and mixing be, at leasten- to ergysome to extent, be suspended water-dispersible, in water. but require some specific mixing time and mixing energy to beComparing suspended the in water.determined surface charge densities at pH ≈ 9 for different materials, we foundComparing that the the proton-related determined surface char chargege density densities increased at pH ≈ as9 forthe differentsurface became materials, wemore found hydrophilic. that the The proton-related results of the surface acid–base charge titrations density are increased shown in asTable the 1 surface for this becamepH, andmore although hydrophilic. the salt The levels results were of different, the acid–base the data titrations in Figure are 1 shownsuggest in that Table this1 variationfor this pH, wasand not although relevant the for salt the levels hydrophobic were different, surfaces. the data in Figure1 suggest that this variation was not relevant for the hydrophobic surfaces. Table 1. Proton-related surface charge densities at pH 9 obtained in the potentiometric acid–base Tabletitrations 1. Proton-relatedon the hydrophobi surfacec and charge hydrophilic densities materials. at pH 9 obtained in the potentiometric acid–base titrations on the hydrophobic and hydrophilic materials. σH(at pH ≈ 9)/mC m−2 Ic/mmol dm−3 −2 −3 PTFE-B-200 σH(at−100 pH ≈ 9)/mC m Ic/mmol1 dm PTFE-B-200 −100 100 PTFE-B-200 −100 1 HydrophobicPTFE-B-200 silica HDK-H20P −70 −100100 100 HydrophobicGraphite (H-PGC) silica HDK-H20P −150 −70100 100 GraphiteCNT (H-PGC) −340 −1502 100 Hydrophilic silicaCNT HDK-H20S −490 −340100 2 Hydrophilic silica HDK-H20S −490 100

The point of zero charge (pHpzc) was determined for graphite and CNT particles using potentiometric mass titrations. Continuous additions of solid particles were made

Colloids Interfaces 2021, 5, x FOR PEER REVIEW 8 of 19

Colloids Interfaces 2021, 5, 6 8 of 19

The point of zero charge (pHpzc) was determined for graphite and CNT particles using potentiometric mass titrations. Continuous additions of solid particles were made to theto theNaCl(aq) NaCl(aq) solution, solution, and and it was it was assumed assumed that that the finally the ﬁnally obtained obtained constant constant value value of pH of pzc −3 −3 representedpH represented pH pH. Thepzc. constant The constant pH was pH reached was reached at mass at mass concentrations concentrations 30 g 30dm g dm in bothin cases.both cases.The measured The measured pH values pH values are arepresented presented in Figure in Figure 4. 4The. The point point of of zero zero charge charge ofof graphitegraphite was was pH pHpzcpzc == 5.6, 5.6, somewhat somewhat higher higher than than the the value value obtained obtained by by results results presented presented inin Figure Figure 3 (pH(pHpzcpzc == 5.1). 5.1). It It should should be be stressed stressed that that two two different different graphite graphite samples samples in in two two laboratorieslaboratories were were analyzed. analyzed. The The mass mass titratio titrationn curve curve determined determined in in Zagreb Zagreb (Figure (Figure 44)) showsshows that that addition addition of of dry dry grap graphitehite particles particles to to aqueous aqueous sodium sodium chloride chloride solution solution led led to to increasingincreasing pH pH values. values. This This result result might might sugg suggestest the the presence presence of of basic basic impurities impurities in in the the graphitegraphite sample. sample. However, However, the the surface surface charge charge density density curve curve (Figure (Figure 33)) has a wide plateau aroundaround the the point point of of zero zero charge, charge, which which makes makes an an accurate accurate determination determination of of pH pHpzcpzc difficult.difﬁcult.

6.0 Graphite CNT pH 5.5

5.0

4.5

4.0

0102030 − γ / g dm 3

−3 FigureFigure 4. 4. PotentiometricPotentiometric mass mass titration of graphite andand CNTCNT inin NaCl(aq)NaCl(aq) solution. solution.I cIc= = 22 mmolmmol dm , dmθ =−3 25, θ◦ =C. 25 °C.

TheThe point point of of zero charge ofof CNTCNT obtainedobtained with with two two different different methods, methods, namely, namely, poten- po- tentiometrictiometric mass mass titration titration (pH (pHpzcpzc= = 3.8) 3.8) and and potentiometric potentiometric acid-base acid-basetitration titration (pH(pHpzcpzc == 4.0), 4.0), waswas in in good good agreement. agreement. 3.2. Electrokinetic Potential of Particles 3.2. Electrokinetic Potential of Particles At pH values where the trend towards less negative charges was quite clear in titra- At pH values where the trend towards less negative charges was quite clear in titrations, the electrokinetic data continued to show persistent negative values for PTFE-B-200 tions, the electrokinetic data continued to show persistent negative values for PTFE-B-200 (Figure5). In both kinds of data, the scatter was relatively large. This has been observed in (Figure 5). In both kinds of data, the scatter was relatively large. This has been observed separate experiments as well, and has also been reported by others [36], being potentially in separate experiments as well, and has also been reported by others [36], being poten- attributable to the difﬁculty to disperse the hydrophobic material or to changes in size (i.e., tially attributable to the difficulty to disperse the hydrophobic material or to changes in formation of aggregates that might settle and affect the measurements; in the experiments size (i.e., formation of aggregates that might settle and affect the measurements; in the here, the size was not measured). In the paper by Marinova et al. [36], data for xylene experimentsshowed similar here, scattering the size was at high not pHmeasured). and the absenceIn the paper of a clear by Marinova trend to zeroet al. mobility [36], data at forlow xylene pH in showed 1 mmol dmsimilar−3 NaCl. scattering This has at high been observedpH and the in separateabsence experimentsof a clear trend as well to zero and −3 mobilityis attributed at low to thepH difﬁcultyin 1 mmol to dm disperse NaCl. the This hydrophobic has been observed material. in Additional separate experiments experiments ashave well been and carriedis attributed out where to the difficulty the PTFE to particles disperse were the hydrophobic initially dispersed material. in ethanol Additional and experimentssubsequently have transferred been carried to an aqueousout where solution, the PTFE resulting particles in betterwere initially dispersions dispersed [32]. This in ethanolyielded and similar subsequently results, but transferred the scatter ofto thean aqueous data persisted. solution, Even resulting with such in better a suspension, disper- sionsonce [32]. the amount This yielded of ethanol similar was results, decreased but the (as scatter checked of bythe IR data spectroscopy), persisted. Even the dispersionwith such abecame suspension, worse. once the amount of ethanol was decreased (as checked by IR spectroscopy), the dispersion became worse.

Colloids Interfaces 2021, 5, x FOR PEER REVIEW 9 of 19 ColloidsColloids Interfaces 2021,, 55,, 6x FOR PEER REVIEW 99 of of 19 19

0 0 234567891011 234567891011 pH pH -10 -10 / mV ζ / mV ζ -20 -20

-30 -30

-40 -40

-50 -50

Figure 5. Electrokinetic zeta-potential of PTFE-B-200 in aqueous sodium chloride solution Ic = 1 FigureFigure 5. 5. ElectrokineticElectrokinetic zeta-potential zeta-potential of PTFE-B- of PTFE-B-200200 in aqueous in aqueous sodium sodiumchloride chloridesolution I solutionc = 1 mmol dm−3, θ = 25−3 °C. ◦ mmolIc = 1 mmoldm−3, θ dm = 25 ,°C.θ = 25 C. The electrophoretic mobilities of the two different graphite samples and the CNT TheThe electrophoretic electrophoretic mobilities mobilities of of the the two two different different graphite graphite samples samples and and the the CNT CNT particles in NaCl(aq) suspension at various pH values are presented in Figure 6. The IEPs particlesparticles in in NaCl(aq) NaCl(aq) suspension suspension at atvarious various pH pH values values are arepresented presented in Figure in Figure 6. The6. IEPs The ofIEPs the ofgraphite the graphite samples samples were found were to found be about to be pH aboutiep = 4.2 pH (KTI= laboratory) 4.2 (KTI laboratory) and pHiep = and3.2 of the graphite samples were found to be about pHiep = 4.2 (KTIiep laboratory) and pHiep = 3.2 (ZagrebpH = laboratory). 3.2 (Zagreb laboratory).This shift is Thisopposite shift isof opposite that observed of that for observed the shift for of the pH shiftpzc obtained of pHpzc (Zagrebiep laboratory). This shift is opposite of that observed for the shift of pHpzc obtained for two graphite samples, which indicates the adsorption of anions on graphite surfaces. forobtained two graphite for two samples, graphite which samples, indicates which the indicates adsorption the adsorption of anions ofon anionsgraphite on surfaces. graphite Nevertheless,surfaces. Nevertheless, the isoelectric the isoelectric points of pointsthe graphite of the graphite and CNT and particles CNT particles were in were the same in the Nevertheless, the isoelectric points of the graphite and CNT particles were in the same rangesame as range other as inert other surfaces inert surfaces (3 < pH (3 < < 4). pH The < 4).IEP The of the IEP graphite of the graphite particles particles (Zagreb (Zagreb labor- range as other inert surfaces (3 < pH < 4). The IEP of the graphite particles (Zagreb laboratory)laboratory) was in was agreement in agreement with withmeasurements measurements of the of hydrodynamic the hydrodynamic radius radius as a as function a function of atory) was in agreement with measurements of the hydrodynamic radius as a function of pHof pH(see (see Appendix Appendix A, AFigure, Figure A4). A4 ). pH (see Appendix A, Figure A4).

Graphite Graphite 1 Graphite - KiT 1 Graphite - KiT CNT CNT

1 0 − 1 0 − 34567891011 34567891011 m s m s

μ pH

-1 μ pH 1

-1 − 1 − / V / V μ -2 μ -2

-3 -3

-4 -4

-5 -5

-6 -6

FigureFigure 6. 6. ElectrophoreticElectrophoretic mobilities mobilities of of graphite graphite (Zg), (Zg), graphite—KIT graphite—KIT (H-PGC), (H-PGC), and and CNT CNT in in aqueous aqueous Figure 6. Electrophoretic mobilities◦ of graphite (Zg), graphite—KIT (H-PGC), and CNT in aqueous sodiumsodium chloride chloride solution solution at 25 °C.C. Lines representrepresent thirdthird order order polynomial polynomial ﬁt fit used used to to determine determine pH iep. sodium chloride− solution3 at 25 °C. Lines represent third order−3 polynomial fit used to determine pHIc =iep 2. I mmolc = 2 mmol dm dmfor−3 graphitefor graphite and and CNT, CNT,Ic = I 1c mmol= 1 mmol dm dmfor−3 for graphite—KIT. graphite—KIT. pHiep. Ic = 2 mmol dm−3 for graphite and CNT, Ic = 1 mmol dm−3 for graphite—KIT.

Colloids Interfaces 2021, 5, 6 10 of 19 Colloids Interfaces 2021, 5, x FOR PEER REVIEW 10 of 19

ConcerningConcerning the the electrophoretic electrophoretic mobility mobility of of graphite graphite and and CNT CNT particles, particles, we we found found no no crossing in in the the studied studied pH pH range, range, which which would would allow allow us us to tofully fully compare compare these these datasets. datasets. Lower values values of of the the electrophoretic electrophoretic mobility mobility of graphite of graphite particles particles and andtheir their lower lower value value of pHof pHiep couldiep could be explained, be explained, as in as the in case the caseof PT ofFE PTFE electrokinetic electrokinetic potentials. potentials. Moreover, Moreover, the lowerthe lower electrophoretic electrophoretic mobility mobility of the ofgraphite the graphite particles particles could have could been have the been result the of asym- result of metricasymmetric counter counter ion association ion association on graphite. on graphite. Unlike PTFE, Unlike where PTFE, the where surface the is surface fully chem- is fully icallychemically inert and inert has and no hasdetectable no detectable contamination contamination (see later), (see graphite later), often graphite contains often oxygen contains groupsoxygen that groups attract that ions attract from ions aqueous from solution aqueous. It solution. could in Itparticular could in be particular possible bethat possible car- boxylatethat carboxylate groups exist groups in these exist carbon-based in these carbon-based particles. particles.The presence The of presence carboxylate of carboxylate groups shouldgroups result should in resultlow isoelectric in low isoelectric points aspoints well. as well. ForFor graphite graphite particles, particles, the the electrokinetic electrokinetic potential potential can can be be calculated calculated using using the the Smolu- Smolu- chowski equation, equation, which which then then allows allows comparison comparison between between flat flat graphite graphite samples samples and and graphite particles. particles. At At a neutral a neutral pH, pH, the calcul the calculatedated electrokinetic electrokinetic potential potential of graphite of graphite particlesparticles was was ζ ≈ −ζ45≈ mv, −45 which mv, which is much is muchmore morenegative negative than the than electrokinetic the electrokinetic potential potential of the of flatthe graphite flat graphite surface. surface. The electr Theokinetic electrokinetic potential potential of graphite of graphite particles particles is in the isrange in the of rangethe electrokineticof the electrokinetic potential potential for the for graphene the graphene surface. surface. Despite Despite this difference, this difference, the pH theiep pHfor iep graphitefor graphite particles particles and andflat graphite flat graphite was similar. was similar. It indicates It indicates that the that same the amount same amount of hy- of droniumhydronium ions ions is needed is needed to neutralize to neutralize the theinterfacial interfacial water water layer. layer. TheThe electrophoretic mobilitymobility ofof the the hydrophobic hydrophobic silica silica particles particles (HKD20-P) (HKD20-P) is presentedis pre- sentedin Figure in 7Figure. The IEP 7. The was IEP close was to close pH 3, to in pH the 3, range in the of range other of inert other surfaces. inert surfaces. As for the As PTFEfor theparticles, PTFE particles, there was there significant was significant scatter in scatter these data.in these data.

0.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 pH

− -0.4 m s μ

1 − -0.8 / V /

-1.2

-1.6

-2.0

−3 Figure 7. 7. ElectrophoreticElectrophoretic mobilities mobilities of ofhydrophobic hydrophobic silica silica particles particles in Icin = 1I cmmol= 1 mmol dm−3 dmaqueousaqueous sodium chloride chloride solution solution at at 25 25 °C.◦C.

Interestingly,Interestingly, the the electrokinetic electrokinetic data data sh showedowed a alower lower IEP IEP (for (for HDK-H20P, HDK-H20P, the the IEP IEP was was below pH pH 4, 4, for for H-PGC, H-PGC, it it was was at at pH pH 4.1) 4.1) than than what what we we inferred inferred for for the the point point of ofzero zero net net proton charge from the titrations. Again, this is comparable for some of the samples stud- studied iedin Zagreb in Zagreb and and at KITat KIT for for the the particles. particles. It It also also is is comparable comparable to to the the differences differences observedobserved by byChibowski Chibowski and and co-workers co-workers for for the the ice ice surface,surface, where an an IEP IEP near near pH pH 3.5 3.5 and and a point a point of of zero net proton proton charge charge at at pH pH 7 7[29] [29 ]was was obtained. obtained. The The same same authors authors reported reported an anisoelectric isoelectric point of 3.3, andand thethe pointpoint ofof zerozero netnet proton proton charge charge was was 6.3 6.3 for for hexadecane hexadecane [29 [29].]. While While this thisagrees agrees with with our our observations observations (i.e., (i.e., signiﬁcantly significantly higher higher points points of zeroof zero net net proton proton charge charge than thanisoelectric isoelectric points), points), for some for some samples, samples, we stress we stress again again that the that continuous the continuous titrations titrations yielded yieldedrelative relative charges charges and thus and the thus obtained the obtained points of points zero chargeof zero are charge subject are tosubject debate. to Moreover,debate. the calculation of the absolute charges is only possible through book-keeping of the charge.

Colloids Interfaces 2021, 5, x FOR PEER REVIEW 11 of 19

Colloids Interfaces 2021, 5, 6 11 of 19

Moreover, the calculation of the absolute charges is only possible through book-keeping of the charge. Since we know how much titrant was added to initially increase the pH, the pointSince of we zero know net how proton much charge titrant can was be addedcalculated. to initially This can increase be easily the pH,falsified thepoint by contami- of zero net proton charge can be calculated. This can be easily falsiﬁed by contamination of the nation of the NaOH solution by carbon dioxide, which is almost impossible to avoid. NaOH solution by carbon dioxide, which is almost impossible to avoid. Thus, the proton Thus, the proton balance is questionable as soon as base titrants are involved. balance is questionable as soon as base titrants are involved.

3.3. Electrokinetic Electrokinetic Potential Potential of Flat Surfaces The obtained values of the electrokinetic po potentialtential at various pH values for graphite and graphene flat ﬂat surfaces are presented in Figure8 8..

20 Graphite Graphene 10

/ mV 234567891011 ζ -10 pH

-20

-30

-40

-50

Figure 8. 8. ElectrokineticElectrokinetic potential potential of ofvarious various inert inert flat ﬂat surfaces surfaces in aqueous in aqueous NaCl NaCl solution. solution. Lines Lines −3−3 ◦ representrepresent third order polynomialpolynomial ﬁtfit used to determine pHiepiep.. IIc c== 2 2mmol mmol dm dm, θ, =θ 25= 25°C. C.

The determined determined pH pHpzcpzc forfor the the graphite graphite particles particles was wasmuch much higher higher (pHpzc (pH = 5,6,pzc Figure= 5,6, 4)Figure than4 determined) than determined values valuesof the ofpH theiep. pHThisiep suggests. This suggests that the that surface the surface of the of graphite the graphite particlesparticles contained contained oxygen oxygen groups, groups, as aspreviously previously discussed discussed in in the the context context of of electrophoretic electrophoretic measurements. TheThe presencepresence of of these these groups groups on theon surfacethe surface of the of graphite the graphite particles particles caused causedother ions other to ions have to affinity have affinity toward toward the surface, the surface, which inwhich turn in caused turn caused pHiep to pH beiep at to lower be at lowervalues values then pH thenpzc .pHpzc. Electrokinetic potentials potentials of of PTFE flat flat surface were also determined by means of streamingstreaming potential measurements in two labo laboratoriesratories and were in very good agreement (Figure9 9).). TheThe isoelectricisoelectric pointpoint was was at at pH pHiepiep == 3.1 3.1 and and did did not not depend depend on on sodium sodium chloride concentration.concentration. TheseThese resultsresults were were also also consistent consistent with with results results for isoelectricfor isoelectric point point and point and pointof surface of surface charge charge obtained obtained for PTFE for particles.PTFE particles. The electrokinetic The electrokinetic potential potential was negative was neg- at ativeneutral at pHneutral (pH pH≈ 7), (pH which ≈ 7), was which explained was explained by the stronger by the affinitystronger of affinity hydroxide of hydroxide ions (than ionshydronium (than hydronium ions) toward ions) the towa inertrd PTFE the inert surfaces. PTFE surfaces.

Colloids Interfaces 2021, 5, x FOR PEER REVIEW 12 of 19 Colloids Interfaces 2021, 5, 6 12 of 19

50 mM NaCl 5 mM NaCl 2.5 mM NaCl 20 2 mM NaCl 1 mM NaCl MQ water 0 234567891011

/ mV pH ζ -20

-40

-60

-80

Figure 9. Electrokinetic potential of PTFE flat surfaces in NaCl(aq) solution and pure water, θ = 25 ◦C. Figure 9. Electrokinetic potential of PTFE flat surfaces in NaCl(aq) solution and pure water, θ = 25 The results for 2 mM NaCl were obtained in Zagreb for a different source of PTFE. °C. The results for 2 mM NaCl were obtained in Zagreb for a different source of PTFE. Overall, the graphite surfaces did not show classical zeta-potentials, with the far Overall, the graphite surfaces did not show classical zeta-potentials, with the far too too low magnitudes observed with increasing distance from pHIEP, while conventional low magnitudes observed with increasing distance from pHIEP, while conventional behav- behavior was observed for the PTFE samples. ior was observed for the PTFE samples. The points of zero net proton charge matching the values of point of zero charge andThe isoelectric points of point zero indicate net proton that charge asymmetric matchi adsorptionng the values of of counter point of ions zero did charge not play and a isoelectricsignificant point role onindicate these that surfaces. asymmetric adsorption of counter ions did not play a significant roleSince on the these data surfaces. for PTFE were very consistent, we focused the XPS investigations on the PTFESince particles.the data for These PTFE are were discussed very consistent, in the following we focused section. the XPS investigations on the PTFE particles. These are discussed in the following section. 3.4. XPS Measurements 3.4. XPSTable Measurements2 contains the results of XPS measurements of the PTFE particles and cryo-XPS measurementsTable 2 contains on the the particles results after of XPS having measurements been in contact of the with PTFE a defined particles solution. and cryo-XPS Table3 measurementscontains the precise on the compositions particles after of having these solutions.been in contact with a defined solution. Table 3 contains the precise compositions of these solutions. Table 2. Atomic percentage from XPS measurements on PTFE particles (PTFE-B-200, denoted as PTp) Tableand cryo-XPS 2. Atomic data percentage after exposure from XPS of PTFE-B-200 measurements to the on solutions PTFE particles characterized (PTFE-B-200, in Table denoted3. In the table,as PTp)“bdl” and means cryo-XPS below data detection after exposure limit, and of “dl” PTFE-B- means200 at to detection the solutions limit (~0.1characterized atom %). in Table 3. In the table, “bdl” means below detection limit, and “dl” means at detection limit (~0.1 atom %). ID C F O ID C F O PTp 34.6 65.4 bdl PTp 34.6 65.4 bdl #1 34.4 65.6 bdl #1 #234.4 34.1 65.6 63.9 bdl 2.0 #2 #334.1 35.1 63.9 64.9 2.0 dl #3 #435.1 34.1 64.9 63.6 dl 1.3 #4 34.1 63.6 1.3

Colloids Interfaces 2021, 5, 6 13 of 19

Table 3. Solution compositions based on additions from stock solutions (NaCl, NaOH) for the samples used for the cryo-XPS measurements reported in Table2.

ID ρ (g/L) Na+ (mM) Cl− (mM) H+ (mM) OH− (mM) #1 4.7 0.00 0.00 0.00 0.00 #2 6.5 5.00 5.00 0.00 0.00 #3 4.3 1.00 0.90 0.00 0.10 #4 6.2 5.00 4.83 0.00 0.17

The results indicate that the oxygen content at the surface of the PTFE powder was below detection limit (bdl). Thus, there was no detectable impurity on the powder. When this powder was exposed to neat water, the composition of the retrieved powder did not change, i.e., the powder did not take up detectable amounts of persistent contaminants from MilliQ water. A similar result was obtained when the particles were exposed to a 100 µM NaOH solution (sample #3). Interestingly, in the presence of NaCl, oxygen was detected in the system. The signature of the oxygen concurred with the signal from water/ice. In these cases, also a signal from a hydrocarbon component in C 1s spectrum was recorded, which hinted at some adventitious carbon (i.e., a contaminant). Overall, no detectable contamination occurred in pure water, and thus we presume, on the basis of the recorded data, that the charging in such systems does not arise from contaminants. The problem now lies in the detection limits and how much would be required to cause the buffering of the concentrated suspensions at pH 3. Thus, would it be possible to detect the undetectable [17]? From previous mass titrations in KCl [32], the lowest pH should be reached with 200 g/L of the PTFE powder. A mass titration involving the PTFE powder at different concentrations in MilliQ water showed that at 200 g/L, the pH was 4 (see Figure 10). Assuming the XPS detection limit of 0.1% of a monolayer and further assuming for oxygen in particular that each contaminant oxygen can release one proton, we estimated that for such a system, a pH value of 4.3 should be obtained at most. This would support the idea that contaminations are not responsible for the drop, more so since for carboxylate groups, two oxygens release one proton. However, the theoretical detection limit is probably lower than what usually has to be assumed in practice. With 1% of a monolayer, for example, and assuming that two oxygens release one proton, the pH that Colloids Interfaces 2021, 5, x FOR PEER REVIEW 14 of 19 could be reached would be 3.6, i.e., lower than the value measured. Unfortunately, the results are not entirely conclusive, and thus the debate must continue.

pH 5.0

4.5

4.0

3.5

3.0 0 50 100 150 200 γ / g dm-3

FigureFigure 10. 10. ResultsResults from from mass mass titrations titrations of PTFE-B-200 of PTFE-B-200 powder powder in MilliQ in water, MilliQ θ water,= 25 °C.θ = 25 ◦C.

As for NaCl-containing solutions, the situation is even more ambiguous, since there is measurable oxygen. In this case, we cannot at all rule out the occurrence of contaminants (potentially from the salt) that would provoke the stronger decrease in pH. The initial idea of these experiments was to assess the presence of hydroxide ions in the fast-frozen samples. However, it turned out that most of the water disappeared, and no trace of hydroxide ions was found.

4. Conclusions In summary, we collected various new sets of experimental data for inert surfaces in electrolyte solutions. Table 4 summarizes the points of zero charge obtained in this study with different materials and various techniques in two different laboratories.

Table 4. Determined values of pHiep and pHpzc for different inert materials in aqueous NaCl solutions, θ = 25 °C. For the titration results in brackets, we indicate whether the results stem from potentiometric mass titrations (mt) or from acid–base titrations (pt).

Methods Streaming Poten- Titration Electrophoresis tial/Current pHiep pHpzc pHiep PTFE (Zg) 3.1 - -

sur- PTFE (KIT) 3.0–3.2 - -

faces Graphite (Zg) 3.3 - - Flat Graphene (Zg) 3.8 - - PTFE-B-200 (KIT) - 3.1 (mt) <3 Graphite (Zg) - 5.6 (mt) 3.2 Graphite-H-PGC (KIT) - 5.1 (mt) 4.2 3.8 (mt) CNT (Zg) - 4.0

Particles Particles 4.0 (pt) Hydrophobic silica HDK- - 5.0 (pt) 3–3.5 H20P (KIT)

Colloids Interfaces 2021, 5, 6 14 of 19

As for NaCl-containing solutions, the situation is even more ambiguous, since there is measurable oxygen. In this case, we cannot at all rule out the occurrence of contaminants (potentially from the salt) that would provoke the stronger decrease in pH. The initial idea of these experiments was to assess the presence of hydroxide ions in the fast-frozen samples. However, it turned out that most of the water disappeared, and no trace of hydroxide ions was found.

4. Conclusions In summary, we collected various new sets of experimental data for inert surfaces in electrolyte solutions. Table4 summarizes the points of zero charge obtained in this study with different materials and various techniques in two different laboratories.

Table 4. Determined values of pHiep and pHpzc for different inert materials in aqueous NaCl solutions, θ = 25 ◦C. For the titration results in brackets, we indicate whether the results stem from potentiometric mass titrations (mt) or from acid–base titrations (pt).

Methods Streaming Titration Electrophoresis Potential/Current pHiep pHpzc pHiep PTFE (Zg) 3.1 - - PTFE (KIT) 3.0–3.2 - - Graphite (Zg) 3.3 - -

Flat surfaces Graphene (Zg) 3.8 - - PTFE-B-200 (KIT) - 3.1 (mt) <3 Graphite (Zg) - 5.6 (mt) 3.2 Graphite-H-PGC (KIT) - 5.1 (mt) 4.2 3.8 (mt) CNT (Zg) - 4.0 4.0 (pt) Particles Hydrophobic silica HDK-H20P (KIT) - 5.0 (pt) 3–3.5 Hydrophilic silica HDK-H20S (KIT) - <3.5 (pt) -

In general, our electrokinetic data conﬁrm the low isoelectric points at pH < 4 for such systems. The effect of salt on the electrokinetic data was as expected. Titrations showed various features that were not expected—the batch-type titrations of PTFE-B-200 did not show any salt dependence of the proton-related surface charge, different from what was observed on hydrophilic surfaces with protonable groups, where higher salt content resulted in increased protonation or deprotonation for a given pH; the titration data for two other inert materials exhibited several changes in slope of the charging curve, which was again different from the typical shapes of titration curve for variable charge minerals. Finally, mass titrations in some systems yielded low end points, in the vicinity of the isoelectric point. However, notably for graphite and the hydrophobic silica, these end points and estimated points of zero charge from continuous titrations were clearly higher. At present, the different observations cannot be interpreted in a simple way. The higher mass titration end points could stem from the respective degrees of hydrophobicity. For example, the graphite samples could contain some (de)protonable groups or the hydrophobic silica could include some silica contribution. However, such groups, in particular carboxylate groups, should lead to low points of zero charge, while the values were higher than expected. Graphite data cannot be easily reconciled with points of zero charge varying between shape and origin of the substrate and the methods. Contrary to this, PTFE data were very self-consistent. The XPS investigation on the PTFE powder did not indicate the presence of contaminants. Even after contact with MilliQ water, no contaminations were detectable. On contact with NaCl containing solutions, new signals were found. Considering the XPS results, the possible number of contaminants on the PTFE-B-200 in terms of estimates of the theoretical XPS detection limit suggest that they should have been observed if the low pH of a given suspension of the particles in MilliQ water under argon was due to contamination from the surface. When using “practical” detection limits, which are higher than the theoretical ones, the required number of contaminants is in the range of possible contamination. It is important to note here that Colloids Interfaces 2021, 5, 6 15 of 19

the samples for XPS measurements were exposed to air. Overall, our results, although favoring the absence of contaminants, should not be considered as ultimate proof of the absence of contamination and their role in the charge of inert surfaces in the presence of aqueous solutions. The most comprehensive interpretation of the observations therefore from our point of view remains the adsorption of water ions.

Author Contributions: Conceptualization, J.L. and T.B.; methodology, A.B. and J.L.; validation, T.B.; formal analysis, A.B., N.B., A.S. and J.L.; investigation, A.B., J.L., N.B., Q.L. and A.S.; writing— original draft preparation, A.B. and J.L.; writing—review and editing, J.L., Q.L., K.H., A.S. and T.B.; visualization, A.B.; supervision, T.B., J.L. and K.H.; funding acquisition, T.B. All authors have read and agreed to the published version of the manuscript. Funding: This research was partly funded by the Croatian science foundation, grant number IP-2014-09-6972. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: The data presented in this study are available on request from the corresponding author. Acknowledgments: The authors are grateful to the Laboratory for Precipitation Processes, Ruder¯ BoškovićInstitute (Zagreb, Croatia), for BET measurements and the Department of Textile Chemistry Technology and Ecology, Faculty of Textile Technology (University of Zagreb), where streaming Colloids Interfaces 2021, 5, x FOR PEER REVIEW 16 of 19 current measurements were done. Conflicts of Interest: The authors declare no conflict of interest.

AppendixAppendix A A AppendixAppendix A.1. A.1. Acid–Base Acid–Base Titrations Titrations and and Determ Determinationination of of the the Proton-Related Proton-Related Surface Surface Charge Charge DensityDensity for for CNT CNT Particles Particles FigureFigure A1 A1 shows shows the the raw raw data data obtained obtained from from continuous continuous acid–base acid–base titrations titrations of of the the twotwo required required runs runs with with and and without without CNT CNT particles. particles.

pH 10

4 3.84 3

0 200 400 600 800 1000 1200 V(NaOH) / μL

FigureFigure A1. A1. Acid–baseAcid–base titration titration of ofaqueous aqueous NaCl NaCl solution solution with with (●) ( •and) and without without (■) (CNT) CNT particles. particles. γ −3 −3 ◦ =γ 5 =g 5dm g dm−3, Ic = ,2I cmmol= 2 mmol dm−3, dm θ = 25, θ°C.= 25 C.

ByBy applying applying the the appropriate appropriate lo logisticgistic function function to to the the data data obtained obtained by by acid–base acid–base titra- titra- tion,tion, the the dependence dependence of of the the surface surface charge charge de densitynsity on on pH pH can can be be co constructed.nstructed. The The appro- appro- priate logistic function is given by the expression: priate logistic function is given by the expression: (A− A ) pH=+12 A V 2 1(+ )p x 0

where A1 is the initial value (in our case initial pH value), A2 is the final value, p is power, and x0 is center (inflection point). These adjustable parameters are determined from the numerical adjustment of the function according to the data, while V represents the volume of the added base. A fit of the above function to experimental data from Figure A1 is shown in Figures A2 and A3.

pH 10

3 0 100 200 300 400 500 600 V (NaOH) / µL

Colloids Interfaces 2021, 5, x FOR PEER REVIEW 16 of 19

Appendix A Appendix A.1. Acid–Base Titrations and Determination of the Proton-Related Surface Charge Density for CNT Particles Figure A1 shows the raw data obtained from continuous acid–base titrations of the two required runs with and without CNT particles.

pH 10

4 3.84 3

0 200 400 600 800 1000 1200 V(NaOH) / μL

Figure A1. Acid–base titration of aqueous NaCl solution with (●) and without (■) CNT particles. γ = 5 g dm−3, Ic = 2 mmol dm−3, θ = 25 °C.

By applying the appropriate logistic function to the data obtained by acid–base titra- Colloids Interfaces 2021, 5, 6 tion, the dependence of the surface charge density on pH can be constructed. The appro-16 of 19 priate logistic function is given by the expression: − =+(A12 A ) pH A2 V p (A1 −1(+A2) ) pH = p + A2 1 + ( V ) x0 x0

wherewhere AA11 is the initial value (in our our case case initial initial pH pH value), value), A A22 isis the the final final value, value, pp isis power, power, andand xx00 isis centercenter (inflection(inflection point). These adjustable parameters parameters are are determined determined from from the the numericalnumerical adjustmentadjustment ofof the functionfunction according to the data, while V representsrepresents the the volume volume ofof thethe addedadded base. base. A A fit fit of theof the above above function function to experimental to experimental data fromdata Figurefrom Figure A1 is shownA1 is inshown Figures in FiguresA2 and A2A3 .and A3.

pH 10

3 Colloids Interfaces 2021, 5, x FOR PEER REVIEW 17 of 19 0 100 200 300 400 500 600 V (NaOH) / µL Figure A2. Fit of the applied logistic function with experimentally obtained data of acid–base titration Figure A2. Fit of the applied logistic function with experimentally obtained data of acid–base titra- of aqueous NaCl solution without CNT particles. tion of aqueous NaCl solution without CNT particles.

pH 10

3 0 200 400 600 800 1000 1200 V (NaOH) / µL FigureFigure A3.A3. Fit ofof appliedapplied logisticlogistic functionfunction withwith experimentallyexperimentally obtained obtained data data of of acid–base acid–base titration titration of CNTof CNT suspension suspension in aqueousin aqueous NaCl NaCl solution. solution.

TheThe surfacesurface charge density can now be calculatedcalculated from the difference in added base base volumevolume forfor titrationtitration withwith ((VVdd) and without (Vb)) CNT CNT particles particles according according to to the the following following equationequation [[37]:37]: Fc(Vb − Vd) σH = FcV()− V σ = sγV b d H sVγ where F denotes Faraday constant, c is starting concetration of NaOH, and s is speciﬁc surface area of CNT particles. Volumes V and V are calculated by choosing default pH where F denotes Faraday constant, c is startingb concetrationd of NaOH, and s is specific values and applying model parameters in previous equation. surface area of CNT particles. Volumes Vb and Vd are calculated by choosing default pH values and applying model parameters in previous equation.

Appendix A.2. Hydrodynamic Diameter of Graphite Particles Figure A4 shows the average hydrodynamic diameters of graphite particles as a function of pH. At pH > 7, the obtained values were between 1000 and 1300 nm. For pH < 6, aggregation of the graphite particles was observed, and the average hydrodynamic diameter was around 2000 nm with maximum at 2500 nm. The recorded maximum value of the hydrodynamic diameter occurred at a pH value close to the independently obtained pHiep.

2400 / nm avg d 2100

1800

1500

1200

900 34567891011 pH

Colloids Interfaces 2021, 5, x FOR PEER REVIEW 17 of 19

Figure A2. Fit of the applied logistic function with experimentally obtained data of acid–base titration of aqueous NaCl solution without CNT particles.

pH 10

3 0 200 400 600 800 1000 1200 V (NaOH) / µL Figure A3. Fit of applied logistic function with experimentally obtained data of acid–base titration of CNT suspension in aqueous NaCl solution.

The surface charge density can now be calculated from the difference in added base volume for titration with (Vd) and without (Vb) CNT particles according to the following equation [37]: FcV()− V σ = b d H sVγ

Colloids Interfaces 2021, 5, 6 where F denotes Faraday constant, c is starting concetration of NaOH, and s is specific17 of 19 surface area of CNT particles. Volumes Vb and Vd are calculated by choosing default pH values and applying model parameters in previous equation.

AppendixAppendix A.2.A.2. Hydrodynamic DiameterDiameter ofof GraphiteGraphite ParticlesParticles FigureFigure A4 showsshows the the average average hydrodynamic hydrodynamic diameters diameters of graphite of graphite particles particles as a func- as a functiontion of pH. of At pH. pH At > pH 7, the > 7, obtained the obtained values values were between were between 1000 and 1000 1300 and nm. 1300 For nm. pH For< 6, pHaggregation < 6, aggregation of the graphite of the graphite particles particles was observed, was observed, and the and average the average hydrodynamic hydrodynamic diam- diametereter was around was around 2000 nm 2000 with nm maximum with maximum at 2500 at nm. 2500 The nm. recorded The recorded maximum maximum value of valuethe hydrodynamic of the hydrodynamic diameter diameter occurred occurredat a pH value at a pH close value to the close independently to the independently obtained obtainedpHiep. pHiep.

2400 / nm avg d 2100

1800

1500

1200

Colloids Interfaces 2021, 5, x FOR PEER REVIEW 18 of 19

900 34567891011 pH Figure A4. Average hydrodynamic diameter of graphite particles. γ = 0.040 g− dm3 −3, Ic = 2 mmol −3 Figure A4. Average hydrodynamic diameter of graphite particles. γ = 0.040 g dm , Ic = 2 mmol dm , dm−3, θ = 25 °C. θ = 25 ◦C.

Appendix B Appendix B This appendix shows photos of the HDK-H20S system. Figure A5 shows the suspen- sionThis under appendix continuous shows stirring. photos It ofwas the well HDK-H20S dispersed, system. and there Figure was A5 no shows notable the difference suspen- sionto hydrophilic under continuous silica suspensions. stirring. It was Figure well dispersed, A6 shows and the there same was suspension no notable differencewith stirring to hydrophilicstopped. One silica can suspensions. clearly see a Figurelayer of A6 particle showss the at the same top suspension of the bottle, with above stirring the stopped.aqueous Onephase. can clearly see a layer of particles at the top of the bottle, above the aqueous phase.

FigureFigure A5.A5. ImageImage of of the the well-dispersed, well-dispersed, stirred stirred HDK-H2 HDK-H20S0S suspension. suspension. There There was no was visible no visible indication of non-dispersed powder. indication of non-dispersed powder.

Figure A6. Image of the same suspension as in Figure A5 when stirring was stopped. The powder on top of the suspension was clearly separated from the rest of the well-dispersed suspension. After stirring resumed, the state as shown in Figure A5 was retrieved.

References 1. Buch, V.; Milet, A.; Vácha, R.; Jungwirth, P.; Devlin, J.P. Water surface is acidic. Proc. Natl. Acad. Sci. 2007, 104, 7342–7347. 2. Beattie, J.K.; Djerdjev, A.M.; Warr, G.G. The surface of neat water is basic. Faraday Discuss. 2009, 141, 31–39, doi:10.1039/B805266B. 3. Jones, G.; Ray, W.A. The Surface Tension of Solutions of Electrolytes as a Function of the Concentration. I. A Differential Method for Measuring Relative Surface Tension. J. Am. Chem. Soc. 1937, 59, 187–198, doi:10.1021/ja01280a048. 4. Langmuir, I. Repulsive Forces between Charged Surfaces in Water, and the Cause of the Jones-Ray Effect. Science 1938, 88, 430– 432. 5. Healy, T.W.; Fuerstenau, D.W. The isoelectric point/point-of zero-charge of interfaces formed by aqueous solutions and nonpolar solids, liquids, and gases. J. Colloid Interface Sci. 2007, 309, 183–188, doi:10.1016/j.jcis.2007.01.048. 6. Zimmermann, R.; Freudenberg, U.; Schweiß, R.; Küttner, D.; Werner, C. Hydroxide and hydronium ion adsorption—A survey. Curr. Opin. Colloid Interface Sci. 2010, 15, 196–202, doi:10.1016/j.cocis.2010.01.002. 7. Poli, E.; Jong, K.H.; Hassanali, A. Charge transfer as a ubiquitous mechanism in determining the negative charge at hydrophobic interfaces. Nat. Commun. 2020, 11, 901, doi:10.1038/s41467-020-14659-5.

Colloids Interfaces 2021, 5, x FOR PEER REVIEW 18 of 19

Figure A4. Average hydrodynamic diameter of graphite particles. γ = 0.040 g dm−3, Ic = 2 mmol dm−3, θ = 25 °C.

Appendix B This appendix shows photos of the HDK-H20S system. Figure A5 shows the suspension under continuous stirring. It was well dispersed, and there was no notable difference to hydrophilic silica suspensions. Figure A6 shows the same suspension with stirring stopped. One can clearly see a layer of particles at the top of the bottle, above the aqueous phase.

Colloids Interfaces 2021, 5, 6 18 of 19 Figure A5. Image of the well-dispersed, stirred HDK-H20S suspension. There was no visible indication of non-dispersed powder.

FigureFigure A6.A6.Image Image ofof thethe samesame suspensionsuspension asas inin FigureFigure A5A5 whenwhen stirringstirring was stopped. The powder on top of the suspension was clearly separated from the rest of the well-dispersed suspension. on top of the suspension was clearly separated from the rest of the well-dispersed suspension. After After stirring resumed, the state as shown in Figure A5 was retrieved. stirring resumed, the state as shown in Figure A5 was retrieved. References References 1. Buch, V.; Milet, A.; Vácha, R.; Jungwirth, P.; Devlin, J.P. Water surface is acidic. Proc. Natl. Acad. Sci. 2007, 104, 7342–7347. 1.2. Buch,Beattie, V.; Milet,J.K.; A.;Djerdjev, Vácha, A.M.; R.; Jungwirth, Warr, P.;G.G. Devlin, The J.P.surface Water surfaceof neatis water acidic. isProc. basic. Natl. Faraday Acad. Sci. Discuss. USA 2007 2009, 104, ,141 7342–7347., 31–39, [CrossRefdoi:10.1039/B805266B.][PubMed] 2.3. Beattie,Jones, G.; J.K.; Ray, Djerdjev, W.A. The A.M.; Surface Warr, Tension G.G. The of surface Solutions of neat of Electrolytes water is basic. as a FaradayFunction Discuss. of the Concentration.2009, 141, 31–39. I. A [CrossRef Differential][PubMed Method] 3. Jones,for Measuring G.; Ray, W.A. Relative The Surface TensionTension. of J. SolutionsAm. Chem. of Soc. Electrolytes 1937, 59, as187–198, a Function doi:10.1021/ja01280a048. of the Concentration. I. A Differential Method 4. forLangmuir, Measuring I. Repulsive Relative Surface Forces between Tension. ChargedJ. Am. Chem. Surfaces Soc. 1937in Water,, 59, 187–198. and the [CauseCrossRef of the] Jones-Ray Effect. Science 1938, 88, 430– 4. Langmuir,432. I. Repulsive Forces between Charged Surfaces in Water, and the Cause of the Jones-Ray Effect. Science 1938, 88, 430–432. 5. [CrossRefHealy, T.W.;][PubMed Fuerstenau,] D.W. The isoelectric point/point-of zero-charge of interfaces formed by aqueous solutions and nonpo- 5. Healy,lar solids, T.W.; liquids, Fuerstenau, and gases. D.W. TheJ. Colloid isoelectric Interface point/point-of Sci. 2007, 309 zero-charge, 183–188, doi:10.1016/j.jcis.2007.01.048. of interfaces formed by aqueous solutions and nonpolar 6. solids,Zimmermann, liquids, and R.; Freudenberg, gases. J. Colloid U.; Interface Schweiß, Sci. R.;2007 Küttner,, 309, 183–188.D.; Werner, [CrossRef C. Hydroxide] and hydronium ion adsorption—A survey. 6. Zimmermann,Curr. Opin. Colloid R.; Freudenberg, Interface Sci. 2010 U.; Schweiß,, 15, 196–202, R.; Küttner, doi:10.1016/j.cocis.2010.01.002. D.; Werner, C. Hydroxide and hydronium ion adsorption—A survey. 7. Curr.Poli, Opin.E.; Jong, Colloid K.H.; Interface Hassanali, Sci. A.2010 Charge, 15, 196–202. transfer [asCrossRef a ubiquitous] mechanism in determining the negative charge at hydrophobic 7. Poli,interfaces. E.; Jong, Nat. K.H.; Commun. Hassanali, 2020 A., 11 Charge, 901, doi:10.1038/s41467-020-14659-5. transfer as a ubiquitous mechanism in determining the negative charge at hydrophobic interfaces. Nat. Commun. 2020, 11, 901. [CrossRef] 8. Yan, X.; Delgado, M.; Aubry, J.; Gribelin, O.; Stocco, A.; Boisson-Da Cruz, F.; Bernard, J.; Ganachaud, F. Central role of bicarbonate anions in charging water/hydrophobic interfaces. J. Phys. Chem. Lett. 2018, 9, 96–103. [CrossRef] 9. Uematsu, Y.; Bonthuis, D.J.; Netz, R.R. Charged surface-active impurities at nanomolar concentration induce Jones–Ray effect. J. Phys. Chem. Lett. 2018, 9, 189–193. [CrossRef] 10. Okur, H.I.; Drexler, C.I.; Tyrode, E.; Cremer, P.S.; Roke, S. The Jones–Ray Effect Is Not Caused by Surface-Active Impurities. J. Phys. Chem. Lett. 2018, 9, 6739–6743. [CrossRef] 11. Uematsu, Y.; Bonthuis, D.J.; Netz, R.R. Nanomolar Surface-Active Charged Impurities Account for the Zeta Potential of Hydrophobic Surfaces. Langmuir 2020, 36, 3645–3658. [CrossRef][PubMed] 12. Yatsuzuka, K.; Higashiyama, Y.; Asano, K. Electrification of polymer surface caused by sliding ultrapure water. Ieee Trans. Ind. Appl. 1996, 32, 825–831. [CrossRef] 13. McCarty, L.S.; Whitesides, G.M. Electrostatic Charging Due to Separation of Ions at Interfaces: Contact Electrification of Ionic Electrets. Angew. Chem. Int. Ed. 2008, 47, 2188–2207. [CrossRef][PubMed] 14. Vácha, R.; Horinek, D.; Berkowitz, M.L.; Jungwirth, P. Hydronium and hydroxide at the interface between water and hydrophobic media. Phys. Chem. Chem. Phys. 2008, 10, 4975–4980. [CrossRef][PubMed] 15. Jungwirth, P. Spiers Memorial Lecture Ions at aqueous interfaces. Faraday Discuss. 2009, 141, 9–30. [CrossRef] 16. Das, S.; Bonn, M.; Backus, E.H.G. The Surface Activity of the Hydrated Proton Is Substantially Higher than That of the Hydroxide Ion. Angew. Chem. Int. Ed. 2019, 58, 15636–15639. [CrossRef] 17. Duignan, T.T.; Peng, M.; Nguyen, A.V.; Zhao, X.S.; Baer, M.D.; Mundy, C.J. Detecting the undetectable: The role of trace surfactant in the Jones-Ray effect. J. Chem. Phys. 2018, 149, 194702. [CrossRef] 18. Nauruzbayeva, J.; Sun, Z.; Gallo, A.; Ibrahim, M.; Santamarina, J.C.; Mishra, H. Electrification at water–hydrophobe interfaces. Nat. Commun. 2020, 11, 5285. [CrossRef] 19. Krasowska, M.; Terpilowski, K.; Chibowski, E.; Malysa, K. Apparent contact angles and time of the three phase contact formation by the bubble colliding with teflon surfaces of different roughness. Phys. Probl. Min. Process 2006, 40, 293–306. 20. Terpilowski, K.; Krasowska, M.; Chibowski, E.; Malysa, K. Influence of teflon surface roughness on the contact angle and the bubble attachment in water. Pol. J. Chem. 2008, 82, 11–16. Colloids Interfaces 2021, 5, 6 19 of 19

21. Drelich, J.W.; Boinovich, L.; Chibowski, E.; Della Volpe, C.; Hołysz, L.; Marmur, A.; Siboni, S. Contact angles: History of over 200 years of open questions. Surf. Innov. 2019, 8, 3–27. [CrossRef] 22. Chibowski, E. Some problems of characterization of a solid surface via the surface free energy changes. Adsorpt. Sci. Technol. 2017, 35, 647–659. [CrossRef] 23. Chibowski, E. Apparent Surface Free Energy of Superhydrophobic Surfaces. J. Adhes. Sci. Technol. 2011, 25, 1323–1336. [CrossRef] 24. Chibowski, E.J. Surface free energy and wettability of silyl layers on silicon determined from contact angle hysteresis. Adv. Colloid Interface Sci. 2005, 113, 121–131. [CrossRef][PubMed] 25. Chibowski, E.; Wiacek, A.E.; Holysz, L.; Terpilowski, K. Investigation of the electrokinetic properties of paraffin suspension. 1. In inorganic electrolyte solutions. Langmuir 2005, 21, 4347–4355. [CrossRef][PubMed] 26. Wiacek, A.; Chibowski, E. Stability of oil/water (ethanol, lysozyme or lysine) emulsions. Colloids Surf. B Biointerfaces 2000, 17, 175–190. [CrossRef] 27. Wiacek, A.E.; Chibowski, E.; Wilk, K. Studies of oil-in-water emulsion stability in the presence of new dicephalic saccharide- derived surfactants. Colloids Surf. B Biointerfaces 2002, 25, 243–256. [CrossRef] 28. Wiacek, A.; Chibowski, E. Zeta potential, effective diameter and multimodal size distribution in oil/water emulsion. Colloids Surf. A Physicochem. Eng. Asp. 1999, 159, 253–261. [CrossRef] 29. Drzymala, J.; Sadowski, Z.; Holysz, L.; Chibowski, E. Ice/water interface: Zeta potential, point of zero charge, and hydrophobicity. J. Colloid Interface Sci. 1999, 220, 229–234. [CrossRef] 30. Kallay, N.; Cop, A.; Chibowski, E.; Holysz, L. Reversible charging of the ice-water interface II. Estimation of equilibrium parameters. J. Colloid Interface Sci. 2003, 259, 89–96. [CrossRef] 31. Hunter, R.J. Zeta Potential in Colloid Science: Principles and Applications; Academic Press: Waltham, MA, USA, 2013; Volume 2. 32. Preoˇcanin,T.; Selmani, A.; Lindqvist-Reis, P.; Heberling, F.; Kallay, N.; Lützenkirchen, J. Surface charge at Teflon/aqueous solution of potassium chloride interfaces. Colloids Surf. A Physicochem. Eng. Asp. 2012, 412, 120–128. [CrossRef] 33. Noh, J.S.; Schwarz, J.A. Estimation of the point of zero charge of simple oxides by mass titration. J. Colloid Interface Sci. 1989, 130, 157–164. [CrossRef] 34. Beamson, G.; Briggs, D. The XPS of Polymers Database; SurfaceSpectra: Manchester, UK, 2000. 35. Shchukarev, A.; Ramstedt, M. Cryo-XPS: Probing intact interfaces in nature and life. Surf. Interface Anal. 2017, 49, 349–356. [CrossRef] 36. Marinova, K.G.; Alargova, R.G.; Denkov, N.D.; Velev, O.D.; Petsev, D.N.; Ivanov, I.B.; Borwankar, R.P. Charging of Oil−Water Interfaces Due to Spontaneous Adsorption of Hydroxyl Ions. Langmuir 1996, 12, 2045–2051. [CrossRef] 37. Lützenkirchen, J.; Preoˇcanin,T.; Kovaˇcević,D.; Tomišić,V.; Lövgren, L. Potentiometric Titrations as a Tool for Surface Charge Determination. Croat. Chem. Acta 2012, 85, 391–417. [CrossRef]