Rings and loops in perflurosurfactants viscoelastic solutions

Colloids and Surfaces A: Physicochem. Eng. Aspects 483 (2015) 150–154

Contents lists available at ScienceDirect

Colloids and Surfaces A: Physicochemical and Engineering Aspects journal homepage: www.elsevier.com/locate/colsurfa

Rings and loops in perflurosurfactants viscoelastic solutions I. Ionita-Abutbul a , L. Abezgauz a , D. Danino a , H. Hoffmann b,∗ a b

Technion – Israel Institute of Technology, Haifa 32000, Israel University of Bayreuth, BZKG, Gottlieb-Keim-Str. 60, 95448 Bayreuth, Germany

h i g h l i g h t s

g r a p h i c a l

a b s t r a c t

• In perfluorosurfactants, electric birefringence detects 4 relaxation times around the 2nd CMC, the onset of the viscosity peak in micellar solutions. • Cryo-TEM correlates this complex behavior with the formation of small micellar rings and loops, an effective way to avoid micellar endcaps.

a r t i c l e

i n f o

Article history: Received 23 June 2015 Accepted 10 July 2015 Available online 27 July 2015 Keywords: Second CMC Micelles Viscosity Electric birefringence Cryo-TEM

a b s t r a c t The structure of micellar solutions has been a topic of intense research. Of particular interest is the relation between rheological properties and the nanostructure of the corresponding micelles. At compositions surrounding the 2nd CMC, the onset of micellar growth, spherical micelles typically transition into short rodlike micelles, and these assemblies continue to grow into long, entangled threadlike micelles, avoiding micellar ends. Another possibility to avoid unfavorable micellar ends is to form small closed structures—rings and loops. Here, we disclose with cryo-TEM the formation of these assemblies near the 2nd CMC, for a perfluorosurfactant C8 F17 SO3 N(C2 H5 )4 , and we correlate the observations with electric birefringence (EB) data that show 4 separate processes near that critical composition. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Many surfactants form highly viscoelastic solutions already at low surfactant concentrations. Cryo-TEM, SANS and SAXS measurements have shown that such solutions contain elongated micelles called wormlike or threadlike micelles. The elastic properties are due to the formation of an entanglement network of the linear threadlike micelles. The rheological behavior of many such systems has been studied in detail [1–6]. It was established that they often behave like Maxwell fluids, meaning that the zero-shear viscosity

∗ Corresponding author. http://dx.doi.org/10.1016/j.colsurfa.2015.07.026 0927-7757/© 2015 Elsevier B.V. All rights reserved.

◦ is given by the product of the shear modulus and the structural relaxation time [7]. ◦ = G◦ × ␶s

(1) ◦

rises usually from the water visWith increasing concentration cosity, continuously, over many orders of magnitude according to a power law equation ◦ ∼ (C/C x )˛

(2)

having a high power law exponent ␣ of around 10 [7]. This was also found for the C8 F17 SO3 N(C2 H5 )4 , a commercially available anionic perflurosurfactant. This surfactant has a CMC of around 1 mM. Conductivity data showed that the tetraethylammonium counter-ions bind strongly to the micelles [8]. The zero-shear viscosity begins to rise strongly

I. Ionita-Abutbul et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 483 (2015) 150–154

100000

n1 = -1.8*10

-8

0.47 ms

1000

in mPas

n1 = -3.1*10

-8

n1 = -3.3*10

-8

0.14 ms

10000

o

151

I

II

1 ms

0.4 ms

III

100

n1 = -3.5*10

0.61 ms

-8

n1 = -3.6*10

1 ms -8

n1 = -8.1*10

-8

10 0.75 ms

1

0.8 ms

0.56 ms

1

10

1 ms

100

c (C8F17SO3N(C2H5)4) in mM Fig. 1. Zero-shear viscosity of C8 F17 SO3 N(C2 H5 )4 aqueous solutions as a function of concentration at T = 25 ◦ C.

after the 2nd CMC [9–11], the concentration where spherical micelles transition to rodlike; this occurs at around 7.5 mM (Fig. 1, the onset of the viscosity rise) [12]. With further increase in the concentration (region II in Fig. 1) the zero-shear viscosity increases more than four orders of magnitude, and passes over a maximum at 60 mM. The increase is thought to be due to continuous elongation of the wormlike micelles, and the formation of a network of linear, entangled micelles [13]. The power-law increase begins around the overlap of the small rodlike micelles. At further increase, a decrease in the viscosity is observed (region III in Fig. 1), generally attributed to mechanisms such as the formation of branched micelles (see [1], and references therein), or a decrease in the micellar size [14]. In this work, we focus on the onset of the viscosity rise, namely, the region of transition from spherical to elongated micelles. Two experimental methods are combined: electric birefringence (EB), and direct-imaging cryo-transmission electron microscopy (cryoTEM). The first method can provide information on the length of the micelles at the overlap concentration [12]. The measurements were reported previously but they turned out to be surprising, and could not be well explained. Thus, in this work a correlation with cryoTEM is used to show directly the structural building units at the transition zone for explaining the unexpected behavior identified in the EB measurements. 2. Experimental 2.1. EB measurement Experiments were done at the quadratic mode at a constant field E = 11.9 × 104 V/m, at 25 ◦ C, and constant laser wavelength of 632 nm, at concentrations around the onset of the viscosity rise, namely, between 5 mM and 14 mM. High-voltage pulses of short rise and decay times were applied, and the time constant of the decay, and the amplitude of the stationary value, were measured. 2.2. Cryo-TEM analysis Cryo-TEM samples were prepared in a controlled environment vitrification system (CEVS) always at a controlled temperature (25 ◦ C) and at saturation. A 6 ␮l drop of the suspension was placed on a 400-mesh TEM copper grid covered with a perforated carbon film. To remove excess solution and produce a thin liquid film the drop was blotted manually. The blotted samples were allowed to stand in the CEVS for 10 s to relax from shearing effects caused by the blotting. The relaxed samples were then plunged into liquid ethane (−183 ◦ C) to form vitrified specimens and transferred

2 ms

4 ms

Fig. 2. EB signals from C8 F17 SO3 N(C2 H5 )4 at constant field E = 11.9 × 104 Vm−1 at 25 ◦ C for increasing concentrations of (from top left to bottom right): 5, 8, 10, 12, 13 and 14 mM. Pulse lengths (0.14 ms to 0.8 ms) are indicated by arrows to show when the E-field was switched on and off, respectively.

to liquid nitrogen (−196 ◦ C) for storage. Vitrified specimens were examined at temperatures below −175 ◦ C using a Gatan 626 cryo holder either in a Tecnai T12 G2 TEM (FEI, Netherlands) or a Philips CM120 TEM operating at 120 kV. Images were recorded on a Gatan MultiScan 791 camera or Gatan UltraScan 1000 using the DigitalMicrograph software (Gatan, U.K.) in the low-dose imaging mode to minimize beam exposure and electron-beam radiation damage, as described [15,16]. 3. Results and discussion EB measurements were performed at 6 concentrations between 5 and 14 mM surfactant, from below the 2nd CMC to way pass it, where the micelles are expected to be already long (see Fig. 1). Specifically, it was expected that a rotation time could be observed according to the equation for ␶ which would increase as rot =

4l3 kB T

(3)

However, within this concentration range in which the viscosity was measured, four separate processes were detectable [12]. Typical signals are shown in Fig. 2. A simple signal was observed only in the very dilute concentration region. This relaxation process  1 was associated with the rotation of the small rodlike micelles. However, unexpectedly, the time constant did not increase in time, and instead a second process was noted. Both processes could be measured over an extended concentration region. Finally, at the concentration at which the viscosity began to rise strongly a third effect appeared. The time constant for this effect, again, did not depend very much on the concentration. Within a small concentration region, all three processes could be observed simultaneously as shown in Fig. 3. At even higher concentrations, a fourth process was detected, which could be associated with the structural relaxation time and which also could be measured with oscillating rheology. All four processes are shown on Fig. 4. While ␶2 has the opposite sign in comparison to ␶1 , and ␶3 was only visible within a small concentration region of a factor 2, the other processes were visible over a large concentration region. Obviously, these results are very different than what is expected and much more information was in the electric birefringence results than in the rheological results. The data did not behave as expected from theoretical consideration, and has been explained as follows: ␶1 is the rotation of small rods in the electric field parallel to the field. ␶2 is explained by the alignment of small rods perpendicular to the electric field [17]. ␶3 is assumed to be due

152

I. Ionita-Abutbul et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 483 (2015) 150–154

300

n in a. u.

250 200 pulse length 0.3 ms

150 100 50 0,0

0,1

0,2

0,3

0,4

t in ms Fig. 3. EB experiment with three different effects of opposite sign (70 mM C8 F17 CO2 N(CH3 )4 ; E = 4.6 × 105 V/m and T = 20 ◦ C.

10 1 0,1

4

(-)

t in sec

0,01 1E-3

3

1E-4

2

(-)

(+)

1E-5 1E-6 1E-7

1

1

(-) 10

100

1000

c (C8F17SO3N(C2H5)4 in mM Fig. 4. Four relaxation time constants ␶1 to ␶4 in aqueous solutions of C8 F17 SO3 N(C2 H5 )4 as a function of concentration, at 25 ◦ C.

to a hindered rotation of rods that are somewhat longer than the rods. And finally, ␶4 is the structural relaxation time which controls the viscosity behavior. Overall, the EB results indicated that the solutions would contain rodlike micelles, that would not grow with the concentration over a large concentration range, a result that is not in agreement with theoretical consideration. Cryo-TEM is a powerful method to probe the nanostructure of micelles and to detect structural transitions e.g., the 2nd CMC [18] and micellar elongation/shortening [14], as well as to capture short lived intermediates such as branching [16,19]. Cryo-TEM analysis of C8 F17 SO3 N(C2 H5 )4 solutions was done at the same concentration range surrounding the 2nd CMC, where the viscosity begins to rise. The results are shown in Fig. 5. The 5 mM solution shows still globular micelles and these coexist with few short cylindrical micelles with a relatively small axial ratio. These objects could have been the small rods from which the electric birefringence signal was coming from. At the concentration of 7 mM the micrographs show that closed rings start to form. Micellar rings are not very common objects, although in essence they are, like micellar elongation, an efficient way to avoid micellar ends. At 8 mM, many rings and long wormlike micelles are present in the solution. The rings have a diameter of ∼50 nm. It is therefore likely that the birefringence signal comes from these objects. Many closed structures are strongly deformed, but most of them are spherical. The situation is changed considerably at slightly higher concentrations of 10 mM and 12 mM surfactant, where branches start to appear. Deformed and un-deformed rings are still present. In addition, long wormlike micelles are present which are longer than 1000 nm. Only few of the micelles have normal end caps; the majority of

micelles present a hybrid structure with rings connected to a main micellar body. This means that the end has folded back on the wormlike micelles and has made a connection with the micelles, a branching point. Branching points which are not located at the end of the wormlike micelles are also visible. The long wormlike micelles are highly bent, presenting straight sections of about 50 nm. This length is also the diameter of the largest un-deformed rings and the length could therefore represent the persistence length of the wormlike micelles. No individual objects are visible that can be correlated with the ␶3 -process, it is therefore likely that the ␶3 -process is not given by the rotation of a free rod but by a section of the wormlike micelles that represents the persistence length of the worms. This interpretation is in agreement with the fact that the ␶3 -process is also present at much higher concentrations in the viscoelastic region in which the ␶4 -process can be observed both by the electric birefringence and by rheological measurements. This process is due to the deformation and the relaxation of the entanglements in the network. With increasing concentration the structural relaxation time and the zero-shear viscosity for this process pass over a maximum at a concentration of 50 mM. It is likely that the mechanism for the relaxation process changes at this concentration from one relaxation process to another. In other systems it was observed that the switch of a mechanism for the structural relaxation time occurs when the linear wormlike micelles form branching points. The deformed network of the wormlike micelles can relax with a reptation mechanism while the network that is cross-linked with fluid branching points can relax with sliding of the branching points. Two branching points can coalesce and release mechanical strain in the system. It is however unlikely that this switch of mechanism occurs in the present system because the branching of the wormlike micelles is already present at concentrations before the viscosity maximum. It is more likely that the mechanism switches from a reptationcontrolled to a kinetically-controlled mechanism for which the strain in the system is released by a breaking of the wormlike micelles. Such a mechanism has recently been observed for the system CTAB/NaSal [20]. Specifically, for compositions to the left of the first viscosity maximum, that is, NaSal/CTAB ratios less than 0.6 the system relaxes by a reptation mechanism while at a NaSal/CTAB ratio of 1 it relaxes by a kinetically controlled mechanism. The two mechanisms can easily be distinguished, by adding glycerol to the aqueous phase. Glycerol does not influence the kinetically controlled mechanism, but it strongly affects the reptation mechanism. The addition of glycerol has an effect on the interaction energy between the wormlike micelles and therefore on the inter diffusions of the wormlike micelles. For situations in which the wormlike micelles can make sticky contacts the effect of glycerol seems to be especially large. Sticky contacts can have large adhesion energies from the hydrophobic surface of the micelles. C8 F17 SO3 N(C2 H5 )4 micelles are covered with a shell of the counterions and on contact the micelles would form a hydrophobic bond. This bond has to be broken first before the wormlike micelles can reptate freely again. The addition of glycerol reduces the attraction between the micelles and therefore their stickiness. As long as the reptation time is shorter than the kinetic-process, the structural relaxation time will be controlled by the reptation process. When the kinetic process becomes shorter than the reptation process the system relaxes by the kinetic process. This could be the case at the viscosity maximum. The question that is to be answered next is why does the structural relaxation time becomes shorter with increasing concentration. The answer could be that a strand of the wormlike micelles has to be broken between the entanglement points. With increasing concentration this distance becomes shorter and the breaking of

I. Ionita-Abutbul et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 483 (2015) 150–154

153

Fig. 5. Cryo-TEM micrographs of the structures formed by C8 F17 SO3 N(C2 H5 )4 at increasing concentrations left to the viscosity peak. (A) At a low concentration of 5 mM spherical micelles are noted, already coexisting with short wormlike micelles. (B) At 7 mM micelles elongation led to closed rings formation. (C) The closing of the edges into rings increased at 8 mM where many isolated rings were present. (D) At 10 mM long wormlike micelles were observed and branches started to appear. (E) At 12 mM both an increase in the length of the wormlike micelles and junctions are observed. (F) Numerous junctions are evident, leading to the beginning of a network at 30 mM. Bar = 100 nm.

the worms could be due to the coalescence of two entanglement points.

4. Summary This paper discusses the anomaly in the EB measurements of a perfluorosurfactant with bulky tetraethylammonium counterions. Structurally, the results can be explained by the formation of micel-

lar loops, coexisting micellar linear micelles. Similar structures were reported for a tetrameric surfactant 12-3-12-4-12-3-12 [21], and for some bulky polymers but no EB measurements or detained rheology were done for those system. This may be explained as follows: at low concentrations, the rigidity of the threadlike micelles precludes their bending, so that they have to incur the energetic penalty associated with ends. As the concentration increases, the length of the threadlike micelles increases and so they can bend and

154

I. Ionita-Abutbul et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 483 (2015) 150–154

close into loops; but because the system is still dilute the micelles cannot become too long or form networks. Increasing the concentration further brings the micelles closer to each other, enabling entanglements and even the formation of branch points. Four relaxation processes with a similar behavior could also be observed for the double chain surfactant C16 C8 N(CH3 )2 Br [22]. Both the ␶1 and the ␶3 -processes were detected over a large concentration region. It is therefore likely that the ␶1 -process is due the alignment of deformed rings from wormlike micelles. Such rings would then mainly occur in the concentration region in which the zero-shear viscosity begins to rise. At higher concentrations they are replaced by entangled wormlike micelles that give rise to the processes ␶3 and ␶4 . A few rings are often trapped in the network [14]. The ␶3 -process is difficult to observe experimentally with rheology measurements because the process is hidden below the long relaxation time of ␶4 that leads to a high viscosity of the system. Oscillating rheological measurements at higher frequencies as normally available show however that such a process is present. The combined analysis of cryo-TEM and EB provided better understanding of the transition at the 2nd CMC, showing also that electric birefringence measurements can provide much deeper understanding of the dynamic behavior of viscoelastic surfactant solutions than rheological measurements. Acknowledgments We acknowledge the support of the Russell Berrie Nanotechnology Institute (RBNI), and the Israel Science Foundation (both in Israel) for supporting the cryo-TEM work. References [1] S.R. Raghavan, G. Fritz, E.W. Kaler, Wormlike micelles formed by synergistic self-assembly in mixtures of anionic and cationic surfactants, Langmuir 18 (10) (2002) 3797–3803. [2] C. Moitzi, N. Freiberger, O. Glatter, Viscoelastic wormlike micellar solutions made from nonionic surfactants: structural investigations by SANS and DLS, J. Phys. Chem. B 109 (33) (2005) 16161–16168. [3] H. Rehage, H. Hoffmann, Rheological properties of viscoelastic surfactant systems, J. Phys. Chem. 92 (16) (1988) 4712–4719. [4] H. Rehage, H. Hoffmann, Viscoelastic surfactant solutions – model systems for rheological research, Mol. Phys. 74 (5) (1991) 933–973.

[5] C. Oelschlaeger, A. Schopferer, F. Scheffold, N. Willenbacher, Linear-to-branched micelles transition: a rheometry and diffusing wave spectroscopy (DWS) study, Langmuir 25 (2) (2009) 716–723. [6] S.A. Rogers, M.A. Calabrese, N.J. Wagner, Rheology of branched wormlike micelles, Curr. Opin. Colloid Interface Sci. 19 (6) (2014) 530–535. [7] M.E. Cates, Flow behaviour of entangled surfactant micelles, J. Phys.: Condens. Matter 8 (47) (1996) 9167–9176. [8] H. Hoffmann, B. Tagesson, Influence of substituted ammonium-ions on the thermodynamics and kinetics of micelles of perfluorated octanesulfonate, Z. Phys. Chem. Frankf. 110 (1) (1978) 113–134. [9] S. May, A. Ben-Shaul, Molecular theory of the sphere-to-rod transition and the second cmc in aqueous micellar solutions, J. Phys. Chem. B 105 (3) (2001) 630–640. [10] N. Dan, S.A. Safran, Junctions and end-caps in self-assembled non-ionic cylindrical micelles, Adv. Colloid Interface Sci. 123 (2006) 323–331. [11] P. Lo Nostro, B.W. Ninham, Hofmeister phenomena: an update on ion specificity in biology, Chem. Rev. 112 (4) (2012) 2286–2322. [12] M. Angel, H. Hoffmann, U. Kramer, H. Thurn, The electric birefringence anomaly in colloidal and micellar systems, Ber. Bunsen Ges. Phys. Chem. Chem. Phys. 93 (2) (1989) 184–191. [13] F. Kern, F. Lequeux, R. Zana, S.J. Candau, Dynamical properties of salt-free viscoelastic micellar solutions, Langmuir 10 (6) (1994) 1714–1723. [14] L. Ziserman, L. Abezgauz, O. Ramon, S.R. Raghavan, D. Danino, Origins of the viscosity peak in wormlike micellar solutions. 1. Mixed catanionic surfactants. A cryo-transmission electron microscopy study, Langmuir 25 (18) (2009) 10483–10489. [15] D. Danino, Cryo-TEM of soft molecular assemblies, Curr. Opin. Colloid Interface Sci. 17 (6) (2012) 316–329. [16] D. Danino, A. Bernheim-Groswasser, Y. Talmon, Digital cryogenic transmission electron microscopy: an advanced tool for direct imaging of complex fluids, Colloid Surf. A 183 (2001) 113–122. [17] H. Hoffmann, D. Grabner, Electric birefringence anomaly of solutions of ionically charged anisometric particles, Adv. Colloid Interface Sci. 216 (2015) 20–35. [18] L. Abezgauz, K. Kuperkar, P.A. Hassan, O. Ramon, P. Bahadur, D. Danino, Effect of hofmeister anions on micellization and micellar growth of the surfactant cetylpyridinium chloride, J. Colloid Interface Sci. 342 (1) (2010) 83–92. [19] D. Danino, Y. Talmon, H. Levy, G. Beinert, R. Zana, Branched thread-like micelles in an aqueous-solution of a trimeric surfactant, Science 269 (5229) (1995) 1420–1421. [20] H. Hoffmann, R. Abdel-Rahem, Influence of co-solvent on the rheological behavior of aqueous viscoelastic surfactant solutions, Colloid Polym. Sci. 288 (6) (2010) 603–612. [21] M. In, V. Bec, O. Aguerre-Chariol, R. Zana, Quaternary ammonium bromide surfactant oligomers in aqueous solution: self-association and microstructure, Langmuir 16 (1) (2000) 141–148. [22] H. Hoffmann, U. Kramer, H. Thurn, Anomalous behavior of micellar solutions in electric birefringence measurements, J. Phys. Chem. 94 (5) (1990) 2027–2033.