Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

On the stability of magnetic colloids

On the stability of magnetic colloids 10.2478/v10063-008-0026-3 NALES UNIVERSITATIS MARIAE CURIE-SKLODOWSKA LUBLIN ­ POLONIA VOL. LXIII, 3 SECTIO AA 2008 M. T. López-López, G. R. Iglesias, J. D. G. Durán, F. González-Caballero1 Department of Applied Physics, Faculty of Sciences, University of Grada, c/Campus de Fuentenueva s/n 18071 Grada. Spain 1 Maria Curie-Sklodowska University, Lublin, Pold In this paper, the preparation of fluids belonging to ferrofluid (FF) d magnetorheological (MR) categories is described. Furthermore, the effect of different stabilizing additives (thixotropic agents, surfactts, magnetic noparticles) d carrier liquids on the sedimentation of these fluids is alyzed. These studies are conducted using: (i) optical d electromagnetic induction techniques for monitoring the stability of dilute d concentrated suspensions, respectively; (ii) contact gle d pendt drop measurements to explain the observed stability in different carriers. The results of these experiments are alyzed in detail d magnetic fluids with improved stability properties are prepared accordingly. 1. INTRODUCTION Magnetic colloids are a group of materials that exhibit the remarkable property of chging their flow properties under the application of external magnetic field [1]. They c be classified into: (i) ferrofluids (FF), which are stable colloidal dispersions of ferro- or ferrimagnetic noparticles in a carrier liquid [2]; d (ii) magnetorheological (MR) fluids, which are dispersions of micron-sized magnetic particles [3]. No-sized particles (diameter 10 nm) used in FF are magnetically single-domain d, therefore, they possess a perment magnetic dipole moment. Because of the existence of magnetic interactions between these particles in the presence of external magnetic field, the structures formed in these fluids c largely affect their flow behavior. However, even under the application of high magnetic fields, FF mifest only a Paper dedicated to Professor Emil Chibowski on the occasion of his 65th birthday M. T. López-López, G. R. Iglesias, J. D. G. Durán d F. González-Caballero relatively modest magnetoviscous response d do not develop a yield stress [1, 4, 5]. On the contrary, micron-sized ferro- or ferrimagnetic particles are magnetically multidomain d attain large magnetic moments under the application of rather weak magnetic fields. As a consequence, MR fluids mifest a high magneto-viscous response characterized by a high yield stress [6, 7]. Due to these properties, MR fluids are field-responsive materials with a broad rge of technological applications. In order to prevent particle aggregation in FF based on non-polar liquids due to v der Waals attraction, d the subsequent gravitational settling, it is necessary to coat the particles with long-chain molecular species (e.g. fatty acids) [4]. In this paper, a series of orgic carriers, with increasing dielectric constts, are tested to determine the adequate coupling between the hydrocarbon tail d the carrier molecules that produce stable ferrofluids. MR fluids are typically formulated using high density materials such as iron, iron alloys or metal oxides (ceramic ferrites) dispersed in low-density liquids. Therefore, their stabilization against sedimentation arises as importt challenge, facing the technological applications of these field-responsive materials [8]. Approaches to improve the stability include: (i) addition of thixotropic agents (e.g. carbon fibers d silica noparticles) [6, 9]; (ii) addition of surfactts (e.g. oleic acid or stearate salts) [10, 11]; (iii) mixing magnetic noparticles [12, 13]; d (iv) use of viscoplastic media or water-in-oil emulsions as continuous phases [14, 15]. In this paper, we report on the possibility of stabilizing iron-MR fluids using different surfactts (oleic acid, aluminum stearate) as well as using thixotropic agents (silica noparticles). In addition, we describe the preparation of magnetic fluid composites (MFC) by dispersing micron-sized iron particles in ferrofluids d we alyze their stability properties. Due to the opaqueness of the MR fluids d MFC studied, the use of classical optical methods is not always suitable to characterize the sedimentation behavior of these systems. Therefore, we have also used electromagnetic method to measure the sedimentation rate in these suspensions [16]. Using this method, the effect of the surfactt or silica concentration (in MR fluids) d of the no-magnetite volume fraction (in MFC) on the stability of the suspensions is alyzed d discussed. 2. FERROFLUIDS 2.1. Ferrofluid Preparation d Particle Morphology The ferrofluids consisted of magnetite noparticles covered by oleate molecules dispersed in different orgic media. The magnetite particles are synthesized by coprecipitation of Fe(II) d Fe(III) salts in aqueous solutions, d the surfactt used to stabilize the suspensions was oleic acid. This fatty acid imposes a steric barrier between the oleate-covered magnetite particles that overcomes the v der Waals d magnetic attractions, avoiding particle aggregation in non-polar carriers. Fig. 1. TEM picture of the magnetite noparticles remaining in suspension after centrifugation at 12000×g. Liquid carrier: dodece. Bar length 20 nm. (From [17]) Figure 1 shows a TEM picture of the particles. The average particle diameter after centrifugation at 12000×g is shown in Table 1. The particle diameter in the dichloromethe ferrofluid is significtly smaller th in the other carriers. In this ferrofluid a clear gravitational settling is observed as a consequence of particle coagulation, leading to rather diluted suspensions in which only tiny particles remained. The dielectric constts of the liquids used in this work, measured with a Dekameter DK 300 apparatus (WTW Germy), are: 1.8 (mineral oil), 1.8 (kerosene), 2.0 (dodece), 2.2 (carbon tetrachloride), 4.8 (chloroform), d 9.1 (dichloromethe). The bulk crystal structure of the synthesized noparticles was alyzed by X-ray diffraction d the data showed excellent coincidence with the reference lines for magnetite. 2.2. Interfacial Free Energy of Interaction d Stability The interfacial free energy of interaction between oleate-covered magnetite noparticles immersed in the different carriers was calculated from previous estimation of the surface free energy of the solid material d the surface tension of the liquid media, following the v Oss' approach [18]. With this aim contact M. T. López-López, G. R. Iglesias, J. D. G. Durán d F. González-Caballero gle d pendt drop measurements were performed. The results are summarized in Tables 1 d 2. Tab. 1. Surface free energy components (mJ/m2) of magnetite noparticles, d surface tension components (mJ/m2) of the carrier liquids at 20.0 ± 0.1 ºC. LW: v der Waals component, +: electron-acceptor parameter; -: electron-donor parameter. Last column: average diameter (nm) (DTEM), obtained from TEM pictures, of particles dispersed in the indicated carrier liquids after centrifugation at 12000×g. (From [17]). Material Oleate-covered magnetite SOLID Magnetite Kerosene Mineral oil Dodece CCl4 Chloroform CH2Cl2 Water 49.3 ± 0.2 25 24 ± 3 25.35 27.0 27.15 53 ± 8 21.8 0.17 ± 0.01 0 0 0 0 0 0 25.5 55.4 ± 0.3 0 0 0 0 0 0 25.5 --7.8 ± 0.3 7.15 ± 0.25 7.9 ± 0.3 6.8 ± 0.3 7.8 ± 0.3 5.14 ± 0.14 - LW 25.3 ± 0.6 0.15 ± 0.22 5.2 ± 0.3 DTEM - CARRIER TOT Tab. 2. Total free energy of interaction, G SLS , between oleate-covered magnetite noparticles in the indicated carriers, d the corresponding Lifshitz-v der Waals d LW AB acid-base contributions ( G SLS ; GSLS ). All qutities in mJ/m2. (From [17]). Carrier Kerosene Mineral oil Dodece CCl4 Chloroform Dichloromethe Water LW G SLS AB GSLS TOT G SLS - 0.002 ± 0.007 - 0.05 ± 0.04 - 0.0001 ± 0.001 - 0.06 ± 0.04 - 0.05 ± 0.04 - 9.3 ± 0.5 - 0.26 ± 0.09 -4±3 -4±3 -4±3 -1±5 -4±3 -4±3 - 52 ± 10 -4±3 -4±3 -4±3 -1±5 -4±3 - 13 ± 3 - 52 ± 10 This thermodynamic alysis demonstrates that the observed stability of the suspensions in liquids with r < 5 is well correlated with the very low lyophobic (or acid-base) attraction between the particles, which c be easily surmounted by thermal agitation, since the v der Waals attraction is negligible. On the contrary, for liquids with r > 9, the suspensions become unstable because of the combined action of the v der Waals d lyophobic attractions, the latter being domint for very polar solvents. 2.3. Magnetic Properties of the Particles d the Ferrofluids The magnetization, M, of the solid powder d the ferrofluids was measured at 20 ºC as a function of the magnetic field strength, H. Figure 2 shows these curves. (a) (b) MO-4 K-4 MO-3 K-1 CL-2 M (kA/m) M (kA/m) 0 -200 -400 -1500 -1000 -500 0 500 1000 1500 0 -10 -20 -1500 -1000 -500 0 500 H (kA/m) H (kA/m) Fig. 2. Magnetization data of: (a) the synthesized oleate-magnetite noparticles (powder); (b) the indicated ferrofluids (see sample identification in Table 3). The lines correspond to the best fit to the Lgevin function (Equation 2). (From [17]). The saturation magnetization of the dry powder is Ms = 4.05 × 105 A/m d its initial magnetic susceptibility i = 5.7. In addition, the synthesized magnetite has negligible coercivity d remence, as expected for a superparamagnetic material above the blocking temperature. From these values, the magnetic particle diameter, Dm, c be determined by mes of [19]: 18kT i Dm = µ M 2 0 s 1/ 3 (1) where k is the Boltzmn constt, T the absolute temperature, d µ0 the magnetic permeability of vacuum. The calculated magnetic diameter (Dm = 7.60 ± 0.03 nm) is slightly smaller th that estimated from TEM pictures (DTEM = 7.8 nm for ferrofluid in kerosene). The difference between Dm d DTEM c be attributed to the presence of a magnetically "dead" layer on the particle surface, as frequently assumed in the literature [20]. The magnetic moment of each particle, m, c be obtained using the Lgevin function [21]: 1 M = M s coth - µ 0 mH kT (2) M. T. López-López, G. R. Iglesias, J. D. G. Durán d F. González-Caballero Fitting this function to the magnetization data in Figure 2a, we obtain a magnetic moment m = (1.38 ± 0.03)×10-19 Am2 = (14900 ± 300)µB per particle, where µB is the Bohr magneton. The saturation magnetization d the initial magnetic susceptibility of the ferrofluids are included in Table 3. The magnetic diameter d the magnetic moment of the particles in the different ferrofluids were determined in a similar way to that followed for dry particles using Equations 1 d 2. The data in Table 3 show that Dm d m for particles immersed in the ferrofluids are slightly larger th those obtained in the dry powder. The significt difference in the magnetic properties (Dm, m) in solid state d in suspension c be attributed to a weak structuration favored by the combined action of both interfacial d magnetic attractions between the particles immersed in the carrier liquid. Tab. 3. Saturation magnetization (Ms) d initial magnetic susceptibility (i) of the indicated ferrofluid samples (carrier / magnetite volume fraction ). The magnetic diameter (Dm) d magnetic moment (m) of the particles in the ferrofluids were obtained by fitting the magnetization data in Figure 2b to Equations 1 d 2. (From [17]). Sample Carrier/ (%) K-1 Kerosene/ 1.1 ± 0.1 K-4 Kerosene/ 4.2 ± 0.1 MO-3 Mineral oil/ 2.8 ± 0.1 MO-4 Mineral oil/ 4.5 ± 0.1 CL-2 Chloroform/ 1.2 ± 0.1 Ms (A/m) 5520 ± 140 20400 ± 500 13800 ± 300 22000 ± 600 5740 ± 140 i × 103 337.43 ± 0.12 401.07 ± 0.25 306.33 ± 0.19 360.53 ± 0.09 260.7 ± 0.3 Dm (nm) 13.5 ± 0.4 9.12 ± 0.07 9.55 ± 0.11 8.60 ± 0.06 12.0 ± 0.3 m (10-19 Am2) 2.02 ± 0.05 1.697 ± 0.024 2.19 ± 0.06 1.48 ± 0.05 1.68 ± 0.03 3. STABILITY OF MAGNETORHEOLOGICAL FLUIDS 3.1. Sedimentation Behavior in Diluted Suspensions A. Experimental. Iron powder of HQ quality (BASF, Germy) was used without further treatment. The iron particles are spherical d polydisperse (average diameter 930 ± 330 nm). Silicone oil of viscosity 35.1 ± 0.3 mPa·s d density 954 kg·m-3 (Fluka, Germy) d mineral oil of viscosity 39.58 ± 0.16 mPa.s d density 854 kg.m-3, both supplied by Fluka (Germy), were used as liquid carriers. Oleic acid (purity 90%) (Sigma-Aldrich, Germy), d silica noparticles 7 nm in diameter (Aerosil-300®, Degussa-Hüls, Germy) were used as stabilizing additives. Iron-silica suspensions were prepared as follows: (1) iron, silica d silicone oil were mixed in a polyethylene container; (2) the mixture was shaken d then introduced in a sonifier (Brson model 450, USA). The preparation of iron-OA suspensions involved the following steps: (1) different oleic acid-mineral oil solutions were prepared; (2) iron was added to the selected oil solution; (3) the suspensions were shaken d sonified (Brson model 450, USA); d (4) the samples were stirred (50 rpm; 24 h; 25 ºC) to allow the adsorption of the additive. No significt chges in the viscosity of the carrier were observed as silica noparticles or oleic acid content increased in the concentration rge employed. The sedimentation behavior of the suspensions was inferred from the evolution of the optical absorbce of the suspensions. With this aim, a Spectronic 601 spectrophotometer (Milton Roy, USA) at 590 nm of wavelength was used. All studied suspensions contained 1.25 g/L of iron (volume fraction = 0.017%). Samples were placed in square cuvettes with 1 cm of light path. Those were placed in such a way that the center of the light beam struck 1.5 cm above their bottom. B. Effect of oleic acid. In Figure 3 the normalized optical absorbce, = A/A0 (A0 is the initial absorbce) is plotted as a function of time for suspensions containing 0.017 vol% of iron d different concentrations of oleic acid (OA). Due to the high density d size of the particles, shows a clear tendency to decrease with time, as a consequence of the progressive disappearce of the iron particles from the lighted region as sedimentation proceeds. The experiments for OA concentration below 16 mM (concentration high enough to complete the first OA adsorbed monolayer, see [22]) show that there is no significt chge in the stability of the suspensions for this concentration rge. Therefore, it seems that steric repulsion does not play y importt role on the sedimentation behavior of iron suspensions at low OA concentration (< 16 mM). However, at high OA concentration ( 31 mM) there is a clear decrease of the settling rate, since the slope of the absorbce vs time curve decreases with the concentration of OA. Finally, at OA concentrations higher th 1.57 M, there is not y progressive improvement of the stability. The decrease of the sedimentation rate as OA concentration increases must be originated by the progressive OA covering of iron particles. As a consequence, the stabilization mechism must be originated by the progressive structuration of the OA molecules in adsorbed multilayers. This behavior drives steric repulsion among the OA-covered iron particles d, therefore, a decrease of settling rate. M. T. López-López, G. R. Iglesias, J. D. G. Durán d F. González-Caballero 0 m mol/L AO 0.3 m mol/L AO 3.1 m mol/L AO 16 m mol/L AO Time (hours) 3.13 mol/L OA 1.57 mol/L OA 627 m mol/L OA 235 m mol/L OA 31 m mol/L OA 0 m mol/L OA Time (hours) Fig. 3. Normalized absorbce, = A/A0, as a function of time for suspensions containing 0.017 vol% of iron d the indicated OA concentration. (From [22]). C. Effect of silica noparticles. These experiments were carried out for suspensions containing 0.017 vol% of iron d silica noparticles concentration up to 24.3 mM. The results of these tests are shown in Figure 4. As c be observed, suspensions with silica concentration below 2.4 mM settle rather quickly. This behavior is a direct consequence of the existence of iron-silica adhesion, which provokes a growing of the particle size d, therefore, a faster settling rate th in the absence of silica. Notice that the existence of heterogeneous aggregation among silica d magnetic particles in similar systems was proved earlier [23, 24]. On the other hd, for silica concentration higher th 2.4 mM the formation of a network of silica particles, by interparticle hydrogen bonding, that imparts a gel-like structure to the suspension [6, 9], is the domint phenomenon. For 4.8, 7.3 d 12.2 silica concentration the absorbce increases at the beginning of the experiment d then it falls down. This c be explained considering that, at the beginning, the iron-silica gel scatters the light that otherwise would strike the detector of the spectrophotometer. However, after a time these iron-silica structures break d fall down, since silica concentration is still too low. Finally, as c be seen in Figure 4, for silica concentration of 24.3 mM the iron-silica network is stiff enough to hold in suspension the entire iron load (absorbce remains constt with time). 0 mmol/L silica 1.25 mmol/L silica 2.4 mmol/L silica 4.8 mmol/L silica 7.3 mmol/L silica 12.2 mmol/L silica 24.3 mmol/L silica t (hours) Fig. 4. Normalized absorbce, = A/A0, as a function of time. All suspensions contained 0.017 vol% of iron d the silica concentration indicated. (From [22]). 3.2. Sedimentation Behavior in Concentrated Suspensions A. Experimental. Iron powder was used as magnetic particles. Kerosene (SigmaAldrich, Germy) was used as liquid carrier. Oleic acid (OA), aluminum stearate (AlSt) (technical quality) (Sigma-Aldrich, Germy) d silica noparticles were used as stabilizing additives. Iron-OA d iron-AlSt suspensions were prepared as described above for diluted iron-OA suspensions. Iron-silica suspensions were also prepared in the same way described above for diluted suspensions. The method used for measuring the particle sedimentation rate in magnetorheological fluids is based on the time evolution of the electromotive force induced in a coil that surrounds the sample. A detailed description of the experimental setup as well as details of the fundamentals of the method c be found in [16]. Briefly, the sedimentation behavior of the suspensions was estimated from the time evolution of the dimensionless induced potential . (t ) 1 - (t ) (3) where (t) is the instteous iron volume fraction in the region surrounded by the sensing coil. M. T. López-López, G. R. Iglesias, J. D. G. Durán d F. González-Caballero B. Comparison between the stabilization effects of OA d AlSt. In order to compare the stabilization efficiency of OA d AlSt, in Figure 5 are represented vs time for iron d iron/OA or AlSt suspensions. As observed, decreases faster with time the larger the OA or AlSt concentration. This c be explained by considering that, in the absence of OA (or AlSt), aggregation must be present because of v der Waals interaction d magnetic attraction due to the weak remnt magnetization of the iron particles. Presumably, large aggregates are formed that sp the walls of the test tube (1 cm in diameter). It is precisely the friction with the walls that must hinder their gravitational settling. When surfactts are added, decreases faster th in their absence, this is in fact indication of diminished aggregation: individual particles or small aggregates sediment more easily th large flocculi spning the tube. Similar trends are obtained for OA d AlSt suspensions, although AlSt concentration approximately seven times higher th that added in iron/OA suspensions is required to produce the same chge in the stability properties, indicating that OA is a more efficient surfactt th AlSt concerning the stabilization against irreversible iron aggregation in oil-based MR fluids. C. Effect of silica noparticles. Whereas OA d AlSt are surfactts that avoid iron particle aggregation by mes of steric repulsion, silica noparticles create a gel-like structure that hinders particles settling. Therefore, it is worth to alyze the effect of silica noparticles on the settling behavior of concentrated suspensions d to compare it with that of OA d AlSt. Figure 6, where the dimensionless induced potential is plotted as a function of time, shows clearly that, as expected, the sedimentation rate decreases because the carrier is progressively thickened as the silica concentration increases. For 83 mM silica concentration the thickening effect is not enough to produce y noticeable improvement on the stability properties. For higher concentrations (up to 167 mM silica) the gel formed significtly reduces the sedimentation rate, but it is not yet sufficiently thick to maintain the iron particles in suspension. Only for a concentration as high as 333 mM, the gel completely avoids particle settling, d is practically constt in the time interval studied. without additive 2.5 mmol/L OA 3.1 mmol/L OA 19 mmol/L AlSt 21 mmol/L AlSt t (s) Fig. 5. vs time for suspensions containing 10 % iron volume fraction d the indicated initial concentrations of oleic acid (OA) d aluminum stearate (AlSt). 0 mmol/L silica 83 mmol/L silica 117 mmol/L silica 167 mmol/L silica 333 mmol/L silica t (s) Fig. 6. Dimensionless increment of the induced potential () as a function of time for suspensions containing 10 % iron volume fraction, d the indicated silica concentrations. (From [25]). M. T. López-López, G. R. Iglesias, J. D. G. Durán d F. González-Caballero 4. STABILITY OF MAGNETIC FLUID COMPOSITES 4.1. Preparation of Magnetic Fluid Composites The magnetic fluid composites (MFC) studied in this paper are composed of micron-sized iron particles (BASF, Germy) dispersed in ferrofluids, prepared as described in paragraph 2, which contain oleate-covered magnetite dispersed in kerosene. To prepare these MFC, proper amounts of iron d ferrofluids were mixed, d the mixtures were shaken d finally immersed in a Brson sonifier. All the suspensions contained the same iron volume fraction = 10 %, while the magnetite volume fraction in the ferrofluids () rged from 0 to 24 %. The sedimentation behavior of these suspensions was studied by the electromagnetic method described earlier. 4.2. Sedimentation Study Figure 7 shows the sedimentation behavior of iron/magnetite suspensions. Our reference is the curve in the absence of magnetite ( = 0). First of all, we observe that the addition of a small amount of Fe3O4 implies increase in sedimentation rate (see curves for = 0.03 d 0.06), as compared with the suspension of iron in pure kerosene ( = 0). This fact c be attributed to the irreversible aggregation between iron particles in the absence of Fe3O4 noparticles, which provokes the formation of big aggregates that spread all over the tube slowing the gravitational settling. When ferrofluid carriers are used, the irreversible aggregation is prevented. Now, let us consider only the curves corresponding to suspensions containing magnetite. As observed, increase in magnetite volume fraction implies a decrease in the sedimentation rate that, in principle, could be attributed to the increase in the viscosity of the ferrofluid carriers. In order to check this hypothesis, a new sedimentation experiment was carried out using a = 10 % iron suspension in silicone oil. The result of this experiment is shown in Figure 8. As c be seen, the sedimentation rate in silicone oil ( = 62.3 mPas) is even slightly higher th in the ferrofluid with = 24% ( = 40.4 mPas). Therefore, it seems that the progressive stabilization of the suspensions as the magnetite content increases cnot be exclusively ascribed to increase in the drag force on the settling iron particles. As a consequence, the high stabilization achieved using a ferrofluid as continuous medium, could be associated to some kind of internal structuration of the particles in the suspension. This structure would avoid the aggregation between iron particles, favored by v der Waals d magnetic attractions. = 0.12 = 0.24 = 0.18 0.4 0.2 0.0 -0.2 0 = 0.03 =0 = 0.06 Time (s) Fig. 7. Dimensionless increment of the induced potential () vs time for iron suspensions (iron volume fraction = 0.1) with different ferrofluid carriers. The magnetite volume fraction of the ferrofluids, , is indicated. (From [16]). (dimensionless) (a) = 0.24 (b) silicone oil 0.4 0.2 (c) = 0.18 0.0 -0.2 0 1000 2000 3000 4000 5000 Time (s) Fig. 8. Similar to Figure 11, using the following carriers: (a) magnetite/kerosene ferrofluid ( = 0.24; = 40.4 mPas), (b) silicone oil ( = 62.3 mPas), d (c) magnetite/kerosene ferrofluid ( = 0.18; = 20.4 mPas). (From [16]). 4.3. Electron Microscopy In order to investigate y possible internal structure of the suspensions, TEM pictures were also taken from particles extracted from diluted (1:1000) M. T. López-López, G. R. Iglesias, J. D. G. Durán d F. González-Caballero samples. Figure 9a corresponds to particles extracted from a suspension that initially contained 10 vol% of iron. As c be seen, some iron aggregates persist even after diluting the suspension. Figure 9b corresponds to initial iron ( = 10%) - magnetite ( = 24%) suspension. In this case, iron aggregates have disappeared, d a "halo" of magnetite noparticles surrounds each particle of iron. This is presumably because of the magnetic attraction between the perment magnetic moment of the single-domain magnetite particles d the induced magnetic moment of the iron particles. Summarizing, the use of extremely bimodal iron-magnetite suspensions has been shown to be efficient way to slow down settling d to prevent irreversible iron aggregation in MR fluids. Fig. 9. TEM pictures of diluted suspensions (1:1000 as compared with those used in sedimentation experiments): (a) = 0.1, = 0; d (b) = 0.1, = 0.24. (From [16]). 5. CONCLUSIONS It is possible to prepare stable ferrofluids in highly non-polar liquid carriers by chemical co-precipitation of ferric d ferrous ions in the presence of oleic acid. The thermodynamic alysis demonstrated that the loss of stability when the dielectric constt of the carrier liquid is increased is a consequence of the combined action of v der Waals d solvation (lyophobic) forces. The magnetization alysis of the ferrofluids indicates that there exists some degree of particle structuration, induced by both interfacial d magnetic attractions between the particles dispersed in the liquid phase, even in carriers with a very low dielectric constt. It is feasible to stabilize concentrated iron suspensions against irreversible aggregation processes in oil media by mes of the adsorption of fatty acids (oleic acid) or salts (aluminum stearate) on the magnetic particles. The adsorbed molecules impart the needed steric barrier to hinder agglomeration among iron particles. The addition of silica noparticles to stabilize concentrated iron suspensions in oil carriers against aggregation d sedimentation processes is a very efficient mechism under rest conditions. Unfortunately, under shearing, particle settling is facilitated by the breakage of the silica network d compact sediments, which could make difficult the redispersion, are created. It is possible to stabilize magnetorheological fluids against aggregation d sedimentation processes by using a ferrofluid as carrier fluid. The magnetic attraction between the noparticles (magnetically single-domain) d the micron-sized particles (magnetically multi-domain) favors the formation of clouds of magnetite noparticles around each iron one. Acknowledgments. The authors wish to express their appreciation d deep recognition to the scientific merits d hum personality of Prof. Emil Chibowski. He undoubtedly constitutes a role model for young scientists. Fincial support by Ministerio de Educación y Ciencia (Spain) d FEDER funds (EU) under Project No. MAT2005-07746-C02-01, d Junta de dalucía (Spain) under Project FQM410 are gratefully acknowledged. One of the authors (M.T. López-López) also acknowledges fincial support by Secretaría de Estado de Universidades e Investigación (Ministerio de Educación y Ciencia, Spain) through its Postdoctoral Fellowship Program (No. EX2006-0467). 6. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Annales UMCS, Chemia de Gruyter

On the stability of magnetic colloids

Loading next page...
 
/lp/de-gruyter/on-the-stability-of-magnetic-colloids-HCud04ofK4

References (27)

Publisher
de Gruyter
Copyright
Copyright © 2008 by the
ISSN
0137-6853
eISSN
2083-358X
DOI
10.2478/v10063-008-0026-3
Publisher site
See Article on Publisher Site

Abstract

10.2478/v10063-008-0026-3 NALES UNIVERSITATIS MARIAE CURIE-SKLODOWSKA LUBLIN ­ POLONIA VOL. LXIII, 3 SECTIO AA 2008 M. T. López-López, G. R. Iglesias, J. D. G. Durán, F. González-Caballero1 Department of Applied Physics, Faculty of Sciences, University of Grada, c/Campus de Fuentenueva s/n 18071 Grada. Spain 1 Maria Curie-Sklodowska University, Lublin, Pold In this paper, the preparation of fluids belonging to ferrofluid (FF) d magnetorheological (MR) categories is described. Furthermore, the effect of different stabilizing additives (thixotropic agents, surfactts, magnetic noparticles) d carrier liquids on the sedimentation of these fluids is alyzed. These studies are conducted using: (i) optical d electromagnetic induction techniques for monitoring the stability of dilute d concentrated suspensions, respectively; (ii) contact gle d pendt drop measurements to explain the observed stability in different carriers. The results of these experiments are alyzed in detail d magnetic fluids with improved stability properties are prepared accordingly. 1. INTRODUCTION Magnetic colloids are a group of materials that exhibit the remarkable property of chging their flow properties under the application of external magnetic field [1]. They c be classified into: (i) ferrofluids (FF), which are stable colloidal dispersions of ferro- or ferrimagnetic noparticles in a carrier liquid [2]; d (ii) magnetorheological (MR) fluids, which are dispersions of micron-sized magnetic particles [3]. No-sized particles (diameter 10 nm) used in FF are magnetically single-domain d, therefore, they possess a perment magnetic dipole moment. Because of the existence of magnetic interactions between these particles in the presence of external magnetic field, the structures formed in these fluids c largely affect their flow behavior. However, even under the application of high magnetic fields, FF mifest only a Paper dedicated to Professor Emil Chibowski on the occasion of his 65th birthday M. T. López-López, G. R. Iglesias, J. D. G. Durán d F. González-Caballero relatively modest magnetoviscous response d do not develop a yield stress [1, 4, 5]. On the contrary, micron-sized ferro- or ferrimagnetic particles are magnetically multidomain d attain large magnetic moments under the application of rather weak magnetic fields. As a consequence, MR fluids mifest a high magneto-viscous response characterized by a high yield stress [6, 7]. Due to these properties, MR fluids are field-responsive materials with a broad rge of technological applications. In order to prevent particle aggregation in FF based on non-polar liquids due to v der Waals attraction, d the subsequent gravitational settling, it is necessary to coat the particles with long-chain molecular species (e.g. fatty acids) [4]. In this paper, a series of orgic carriers, with increasing dielectric constts, are tested to determine the adequate coupling between the hydrocarbon tail d the carrier molecules that produce stable ferrofluids. MR fluids are typically formulated using high density materials such as iron, iron alloys or metal oxides (ceramic ferrites) dispersed in low-density liquids. Therefore, their stabilization against sedimentation arises as importt challenge, facing the technological applications of these field-responsive materials [8]. Approaches to improve the stability include: (i) addition of thixotropic agents (e.g. carbon fibers d silica noparticles) [6, 9]; (ii) addition of surfactts (e.g. oleic acid or stearate salts) [10, 11]; (iii) mixing magnetic noparticles [12, 13]; d (iv) use of viscoplastic media or water-in-oil emulsions as continuous phases [14, 15]. In this paper, we report on the possibility of stabilizing iron-MR fluids using different surfactts (oleic acid, aluminum stearate) as well as using thixotropic agents (silica noparticles). In addition, we describe the preparation of magnetic fluid composites (MFC) by dispersing micron-sized iron particles in ferrofluids d we alyze their stability properties. Due to the opaqueness of the MR fluids d MFC studied, the use of classical optical methods is not always suitable to characterize the sedimentation behavior of these systems. Therefore, we have also used electromagnetic method to measure the sedimentation rate in these suspensions [16]. Using this method, the effect of the surfactt or silica concentration (in MR fluids) d of the no-magnetite volume fraction (in MFC) on the stability of the suspensions is alyzed d discussed. 2. FERROFLUIDS 2.1. Ferrofluid Preparation d Particle Morphology The ferrofluids consisted of magnetite noparticles covered by oleate molecules dispersed in different orgic media. The magnetite particles are synthesized by coprecipitation of Fe(II) d Fe(III) salts in aqueous solutions, d the surfactt used to stabilize the suspensions was oleic acid. This fatty acid imposes a steric barrier between the oleate-covered magnetite particles that overcomes the v der Waals d magnetic attractions, avoiding particle aggregation in non-polar carriers. Fig. 1. TEM picture of the magnetite noparticles remaining in suspension after centrifugation at 12000×g. Liquid carrier: dodece. Bar length 20 nm. (From [17]) Figure 1 shows a TEM picture of the particles. The average particle diameter after centrifugation at 12000×g is shown in Table 1. The particle diameter in the dichloromethe ferrofluid is significtly smaller th in the other carriers. In this ferrofluid a clear gravitational settling is observed as a consequence of particle coagulation, leading to rather diluted suspensions in which only tiny particles remained. The dielectric constts of the liquids used in this work, measured with a Dekameter DK 300 apparatus (WTW Germy), are: 1.8 (mineral oil), 1.8 (kerosene), 2.0 (dodece), 2.2 (carbon tetrachloride), 4.8 (chloroform), d 9.1 (dichloromethe). The bulk crystal structure of the synthesized noparticles was alyzed by X-ray diffraction d the data showed excellent coincidence with the reference lines for magnetite. 2.2. Interfacial Free Energy of Interaction d Stability The interfacial free energy of interaction between oleate-covered magnetite noparticles immersed in the different carriers was calculated from previous estimation of the surface free energy of the solid material d the surface tension of the liquid media, following the v Oss' approach [18]. With this aim contact M. T. López-López, G. R. Iglesias, J. D. G. Durán d F. González-Caballero gle d pendt drop measurements were performed. The results are summarized in Tables 1 d 2. Tab. 1. Surface free energy components (mJ/m2) of magnetite noparticles, d surface tension components (mJ/m2) of the carrier liquids at 20.0 ± 0.1 ºC. LW: v der Waals component, +: electron-acceptor parameter; -: electron-donor parameter. Last column: average diameter (nm) (DTEM), obtained from TEM pictures, of particles dispersed in the indicated carrier liquids after centrifugation at 12000×g. (From [17]). Material Oleate-covered magnetite SOLID Magnetite Kerosene Mineral oil Dodece CCl4 Chloroform CH2Cl2 Water 49.3 ± 0.2 25 24 ± 3 25.35 27.0 27.15 53 ± 8 21.8 0.17 ± 0.01 0 0 0 0 0 0 25.5 55.4 ± 0.3 0 0 0 0 0 0 25.5 --7.8 ± 0.3 7.15 ± 0.25 7.9 ± 0.3 6.8 ± 0.3 7.8 ± 0.3 5.14 ± 0.14 - LW 25.3 ± 0.6 0.15 ± 0.22 5.2 ± 0.3 DTEM - CARRIER TOT Tab. 2. Total free energy of interaction, G SLS , between oleate-covered magnetite noparticles in the indicated carriers, d the corresponding Lifshitz-v der Waals d LW AB acid-base contributions ( G SLS ; GSLS ). All qutities in mJ/m2. (From [17]). Carrier Kerosene Mineral oil Dodece CCl4 Chloroform Dichloromethe Water LW G SLS AB GSLS TOT G SLS - 0.002 ± 0.007 - 0.05 ± 0.04 - 0.0001 ± 0.001 - 0.06 ± 0.04 - 0.05 ± 0.04 - 9.3 ± 0.5 - 0.26 ± 0.09 -4±3 -4±3 -4±3 -1±5 -4±3 -4±3 - 52 ± 10 -4±3 -4±3 -4±3 -1±5 -4±3 - 13 ± 3 - 52 ± 10 This thermodynamic alysis demonstrates that the observed stability of the suspensions in liquids with r < 5 is well correlated with the very low lyophobic (or acid-base) attraction between the particles, which c be easily surmounted by thermal agitation, since the v der Waals attraction is negligible. On the contrary, for liquids with r > 9, the suspensions become unstable because of the combined action of the v der Waals d lyophobic attractions, the latter being domint for very polar solvents. 2.3. Magnetic Properties of the Particles d the Ferrofluids The magnetization, M, of the solid powder d the ferrofluids was measured at 20 ºC as a function of the magnetic field strength, H. Figure 2 shows these curves. (a) (b) MO-4 K-4 MO-3 K-1 CL-2 M (kA/m) M (kA/m) 0 -200 -400 -1500 -1000 -500 0 500 1000 1500 0 -10 -20 -1500 -1000 -500 0 500 H (kA/m) H (kA/m) Fig. 2. Magnetization data of: (a) the synthesized oleate-magnetite noparticles (powder); (b) the indicated ferrofluids (see sample identification in Table 3). The lines correspond to the best fit to the Lgevin function (Equation 2). (From [17]). The saturation magnetization of the dry powder is Ms = 4.05 × 105 A/m d its initial magnetic susceptibility i = 5.7. In addition, the synthesized magnetite has negligible coercivity d remence, as expected for a superparamagnetic material above the blocking temperature. From these values, the magnetic particle diameter, Dm, c be determined by mes of [19]: 18kT i Dm = µ M 2 0 s 1/ 3 (1) where k is the Boltzmn constt, T the absolute temperature, d µ0 the magnetic permeability of vacuum. The calculated magnetic diameter (Dm = 7.60 ± 0.03 nm) is slightly smaller th that estimated from TEM pictures (DTEM = 7.8 nm for ferrofluid in kerosene). The difference between Dm d DTEM c be attributed to the presence of a magnetically "dead" layer on the particle surface, as frequently assumed in the literature [20]. The magnetic moment of each particle, m, c be obtained using the Lgevin function [21]: 1 M = M s coth - µ 0 mH kT (2) M. T. López-López, G. R. Iglesias, J. D. G. Durán d F. González-Caballero Fitting this function to the magnetization data in Figure 2a, we obtain a magnetic moment m = (1.38 ± 0.03)×10-19 Am2 = (14900 ± 300)µB per particle, where µB is the Bohr magneton. The saturation magnetization d the initial magnetic susceptibility of the ferrofluids are included in Table 3. The magnetic diameter d the magnetic moment of the particles in the different ferrofluids were determined in a similar way to that followed for dry particles using Equations 1 d 2. The data in Table 3 show that Dm d m for particles immersed in the ferrofluids are slightly larger th those obtained in the dry powder. The significt difference in the magnetic properties (Dm, m) in solid state d in suspension c be attributed to a weak structuration favored by the combined action of both interfacial d magnetic attractions between the particles immersed in the carrier liquid. Tab. 3. Saturation magnetization (Ms) d initial magnetic susceptibility (i) of the indicated ferrofluid samples (carrier / magnetite volume fraction ). The magnetic diameter (Dm) d magnetic moment (m) of the particles in the ferrofluids were obtained by fitting the magnetization data in Figure 2b to Equations 1 d 2. (From [17]). Sample Carrier/ (%) K-1 Kerosene/ 1.1 ± 0.1 K-4 Kerosene/ 4.2 ± 0.1 MO-3 Mineral oil/ 2.8 ± 0.1 MO-4 Mineral oil/ 4.5 ± 0.1 CL-2 Chloroform/ 1.2 ± 0.1 Ms (A/m) 5520 ± 140 20400 ± 500 13800 ± 300 22000 ± 600 5740 ± 140 i × 103 337.43 ± 0.12 401.07 ± 0.25 306.33 ± 0.19 360.53 ± 0.09 260.7 ± 0.3 Dm (nm) 13.5 ± 0.4 9.12 ± 0.07 9.55 ± 0.11 8.60 ± 0.06 12.0 ± 0.3 m (10-19 Am2) 2.02 ± 0.05 1.697 ± 0.024 2.19 ± 0.06 1.48 ± 0.05 1.68 ± 0.03 3. STABILITY OF MAGNETORHEOLOGICAL FLUIDS 3.1. Sedimentation Behavior in Diluted Suspensions A. Experimental. Iron powder of HQ quality (BASF, Germy) was used without further treatment. The iron particles are spherical d polydisperse (average diameter 930 ± 330 nm). Silicone oil of viscosity 35.1 ± 0.3 mPa·s d density 954 kg·m-3 (Fluka, Germy) d mineral oil of viscosity 39.58 ± 0.16 mPa.s d density 854 kg.m-3, both supplied by Fluka (Germy), were used as liquid carriers. Oleic acid (purity 90%) (Sigma-Aldrich, Germy), d silica noparticles 7 nm in diameter (Aerosil-300®, Degussa-Hüls, Germy) were used as stabilizing additives. Iron-silica suspensions were prepared as follows: (1) iron, silica d silicone oil were mixed in a polyethylene container; (2) the mixture was shaken d then introduced in a sonifier (Brson model 450, USA). The preparation of iron-OA suspensions involved the following steps: (1) different oleic acid-mineral oil solutions were prepared; (2) iron was added to the selected oil solution; (3) the suspensions were shaken d sonified (Brson model 450, USA); d (4) the samples were stirred (50 rpm; 24 h; 25 ºC) to allow the adsorption of the additive. No significt chges in the viscosity of the carrier were observed as silica noparticles or oleic acid content increased in the concentration rge employed. The sedimentation behavior of the suspensions was inferred from the evolution of the optical absorbce of the suspensions. With this aim, a Spectronic 601 spectrophotometer (Milton Roy, USA) at 590 nm of wavelength was used. All studied suspensions contained 1.25 g/L of iron (volume fraction = 0.017%). Samples were placed in square cuvettes with 1 cm of light path. Those were placed in such a way that the center of the light beam struck 1.5 cm above their bottom. B. Effect of oleic acid. In Figure 3 the normalized optical absorbce, = A/A0 (A0 is the initial absorbce) is plotted as a function of time for suspensions containing 0.017 vol% of iron d different concentrations of oleic acid (OA). Due to the high density d size of the particles, shows a clear tendency to decrease with time, as a consequence of the progressive disappearce of the iron particles from the lighted region as sedimentation proceeds. The experiments for OA concentration below 16 mM (concentration high enough to complete the first OA adsorbed monolayer, see [22]) show that there is no significt chge in the stability of the suspensions for this concentration rge. Therefore, it seems that steric repulsion does not play y importt role on the sedimentation behavior of iron suspensions at low OA concentration (< 16 mM). However, at high OA concentration ( 31 mM) there is a clear decrease of the settling rate, since the slope of the absorbce vs time curve decreases with the concentration of OA. Finally, at OA concentrations higher th 1.57 M, there is not y progressive improvement of the stability. The decrease of the sedimentation rate as OA concentration increases must be originated by the progressive OA covering of iron particles. As a consequence, the stabilization mechism must be originated by the progressive structuration of the OA molecules in adsorbed multilayers. This behavior drives steric repulsion among the OA-covered iron particles d, therefore, a decrease of settling rate. M. T. López-López, G. R. Iglesias, J. D. G. Durán d F. González-Caballero 0 m mol/L AO 0.3 m mol/L AO 3.1 m mol/L AO 16 m mol/L AO Time (hours) 3.13 mol/L OA 1.57 mol/L OA 627 m mol/L OA 235 m mol/L OA 31 m mol/L OA 0 m mol/L OA Time (hours) Fig. 3. Normalized absorbce, = A/A0, as a function of time for suspensions containing 0.017 vol% of iron d the indicated OA concentration. (From [22]). C. Effect of silica noparticles. These experiments were carried out for suspensions containing 0.017 vol% of iron d silica noparticles concentration up to 24.3 mM. The results of these tests are shown in Figure 4. As c be observed, suspensions with silica concentration below 2.4 mM settle rather quickly. This behavior is a direct consequence of the existence of iron-silica adhesion, which provokes a growing of the particle size d, therefore, a faster settling rate th in the absence of silica. Notice that the existence of heterogeneous aggregation among silica d magnetic particles in similar systems was proved earlier [23, 24]. On the other hd, for silica concentration higher th 2.4 mM the formation of a network of silica particles, by interparticle hydrogen bonding, that imparts a gel-like structure to the suspension [6, 9], is the domint phenomenon. For 4.8, 7.3 d 12.2 silica concentration the absorbce increases at the beginning of the experiment d then it falls down. This c be explained considering that, at the beginning, the iron-silica gel scatters the light that otherwise would strike the detector of the spectrophotometer. However, after a time these iron-silica structures break d fall down, since silica concentration is still too low. Finally, as c be seen in Figure 4, for silica concentration of 24.3 mM the iron-silica network is stiff enough to hold in suspension the entire iron load (absorbce remains constt with time). 0 mmol/L silica 1.25 mmol/L silica 2.4 mmol/L silica 4.8 mmol/L silica 7.3 mmol/L silica 12.2 mmol/L silica 24.3 mmol/L silica t (hours) Fig. 4. Normalized absorbce, = A/A0, as a function of time. All suspensions contained 0.017 vol% of iron d the silica concentration indicated. (From [22]). 3.2. Sedimentation Behavior in Concentrated Suspensions A. Experimental. Iron powder was used as magnetic particles. Kerosene (SigmaAldrich, Germy) was used as liquid carrier. Oleic acid (OA), aluminum stearate (AlSt) (technical quality) (Sigma-Aldrich, Germy) d silica noparticles were used as stabilizing additives. Iron-OA d iron-AlSt suspensions were prepared as described above for diluted iron-OA suspensions. Iron-silica suspensions were also prepared in the same way described above for diluted suspensions. The method used for measuring the particle sedimentation rate in magnetorheological fluids is based on the time evolution of the electromotive force induced in a coil that surrounds the sample. A detailed description of the experimental setup as well as details of the fundamentals of the method c be found in [16]. Briefly, the sedimentation behavior of the suspensions was estimated from the time evolution of the dimensionless induced potential . (t ) 1 - (t ) (3) where (t) is the instteous iron volume fraction in the region surrounded by the sensing coil. M. T. López-López, G. R. Iglesias, J. D. G. Durán d F. González-Caballero B. Comparison between the stabilization effects of OA d AlSt. In order to compare the stabilization efficiency of OA d AlSt, in Figure 5 are represented vs time for iron d iron/OA or AlSt suspensions. As observed, decreases faster with time the larger the OA or AlSt concentration. This c be explained by considering that, in the absence of OA (or AlSt), aggregation must be present because of v der Waals interaction d magnetic attraction due to the weak remnt magnetization of the iron particles. Presumably, large aggregates are formed that sp the walls of the test tube (1 cm in diameter). It is precisely the friction with the walls that must hinder their gravitational settling. When surfactts are added, decreases faster th in their absence, this is in fact indication of diminished aggregation: individual particles or small aggregates sediment more easily th large flocculi spning the tube. Similar trends are obtained for OA d AlSt suspensions, although AlSt concentration approximately seven times higher th that added in iron/OA suspensions is required to produce the same chge in the stability properties, indicating that OA is a more efficient surfactt th AlSt concerning the stabilization against irreversible iron aggregation in oil-based MR fluids. C. Effect of silica noparticles. Whereas OA d AlSt are surfactts that avoid iron particle aggregation by mes of steric repulsion, silica noparticles create a gel-like structure that hinders particles settling. Therefore, it is worth to alyze the effect of silica noparticles on the settling behavior of concentrated suspensions d to compare it with that of OA d AlSt. Figure 6, where the dimensionless induced potential is plotted as a function of time, shows clearly that, as expected, the sedimentation rate decreases because the carrier is progressively thickened as the silica concentration increases. For 83 mM silica concentration the thickening effect is not enough to produce y noticeable improvement on the stability properties. For higher concentrations (up to 167 mM silica) the gel formed significtly reduces the sedimentation rate, but it is not yet sufficiently thick to maintain the iron particles in suspension. Only for a concentration as high as 333 mM, the gel completely avoids particle settling, d is practically constt in the time interval studied. without additive 2.5 mmol/L OA 3.1 mmol/L OA 19 mmol/L AlSt 21 mmol/L AlSt t (s) Fig. 5. vs time for suspensions containing 10 % iron volume fraction d the indicated initial concentrations of oleic acid (OA) d aluminum stearate (AlSt). 0 mmol/L silica 83 mmol/L silica 117 mmol/L silica 167 mmol/L silica 333 mmol/L silica t (s) Fig. 6. Dimensionless increment of the induced potential () as a function of time for suspensions containing 10 % iron volume fraction, d the indicated silica concentrations. (From [25]). M. T. López-López, G. R. Iglesias, J. D. G. Durán d F. González-Caballero 4. STABILITY OF MAGNETIC FLUID COMPOSITES 4.1. Preparation of Magnetic Fluid Composites The magnetic fluid composites (MFC) studied in this paper are composed of micron-sized iron particles (BASF, Germy) dispersed in ferrofluids, prepared as described in paragraph 2, which contain oleate-covered magnetite dispersed in kerosene. To prepare these MFC, proper amounts of iron d ferrofluids were mixed, d the mixtures were shaken d finally immersed in a Brson sonifier. All the suspensions contained the same iron volume fraction = 10 %, while the magnetite volume fraction in the ferrofluids () rged from 0 to 24 %. The sedimentation behavior of these suspensions was studied by the electromagnetic method described earlier. 4.2. Sedimentation Study Figure 7 shows the sedimentation behavior of iron/magnetite suspensions. Our reference is the curve in the absence of magnetite ( = 0). First of all, we observe that the addition of a small amount of Fe3O4 implies increase in sedimentation rate (see curves for = 0.03 d 0.06), as compared with the suspension of iron in pure kerosene ( = 0). This fact c be attributed to the irreversible aggregation between iron particles in the absence of Fe3O4 noparticles, which provokes the formation of big aggregates that spread all over the tube slowing the gravitational settling. When ferrofluid carriers are used, the irreversible aggregation is prevented. Now, let us consider only the curves corresponding to suspensions containing magnetite. As observed, increase in magnetite volume fraction implies a decrease in the sedimentation rate that, in principle, could be attributed to the increase in the viscosity of the ferrofluid carriers. In order to check this hypothesis, a new sedimentation experiment was carried out using a = 10 % iron suspension in silicone oil. The result of this experiment is shown in Figure 8. As c be seen, the sedimentation rate in silicone oil ( = 62.3 mPas) is even slightly higher th in the ferrofluid with = 24% ( = 40.4 mPas). Therefore, it seems that the progressive stabilization of the suspensions as the magnetite content increases cnot be exclusively ascribed to increase in the drag force on the settling iron particles. As a consequence, the high stabilization achieved using a ferrofluid as continuous medium, could be associated to some kind of internal structuration of the particles in the suspension. This structure would avoid the aggregation between iron particles, favored by v der Waals d magnetic attractions. = 0.12 = 0.24 = 0.18 0.4 0.2 0.0 -0.2 0 = 0.03 =0 = 0.06 Time (s) Fig. 7. Dimensionless increment of the induced potential () vs time for iron suspensions (iron volume fraction = 0.1) with different ferrofluid carriers. The magnetite volume fraction of the ferrofluids, , is indicated. (From [16]). (dimensionless) (a) = 0.24 (b) silicone oil 0.4 0.2 (c) = 0.18 0.0 -0.2 0 1000 2000 3000 4000 5000 Time (s) Fig. 8. Similar to Figure 11, using the following carriers: (a) magnetite/kerosene ferrofluid ( = 0.24; = 40.4 mPas), (b) silicone oil ( = 62.3 mPas), d (c) magnetite/kerosene ferrofluid ( = 0.18; = 20.4 mPas). (From [16]). 4.3. Electron Microscopy In order to investigate y possible internal structure of the suspensions, TEM pictures were also taken from particles extracted from diluted (1:1000) M. T. López-López, G. R. Iglesias, J. D. G. Durán d F. González-Caballero samples. Figure 9a corresponds to particles extracted from a suspension that initially contained 10 vol% of iron. As c be seen, some iron aggregates persist even after diluting the suspension. Figure 9b corresponds to initial iron ( = 10%) - magnetite ( = 24%) suspension. In this case, iron aggregates have disappeared, d a "halo" of magnetite noparticles surrounds each particle of iron. This is presumably because of the magnetic attraction between the perment magnetic moment of the single-domain magnetite particles d the induced magnetic moment of the iron particles. Summarizing, the use of extremely bimodal iron-magnetite suspensions has been shown to be efficient way to slow down settling d to prevent irreversible iron aggregation in MR fluids. Fig. 9. TEM pictures of diluted suspensions (1:1000 as compared with those used in sedimentation experiments): (a) = 0.1, = 0; d (b) = 0.1, = 0.24. (From [16]). 5. CONCLUSIONS It is possible to prepare stable ferrofluids in highly non-polar liquid carriers by chemical co-precipitation of ferric d ferrous ions in the presence of oleic acid. The thermodynamic alysis demonstrated that the loss of stability when the dielectric constt of the carrier liquid is increased is a consequence of the combined action of v der Waals d solvation (lyophobic) forces. The magnetization alysis of the ferrofluids indicates that there exists some degree of particle structuration, induced by both interfacial d magnetic attractions between the particles dispersed in the liquid phase, even in carriers with a very low dielectric constt. It is feasible to stabilize concentrated iron suspensions against irreversible aggregation processes in oil media by mes of the adsorption of fatty acids (oleic acid) or salts (aluminum stearate) on the magnetic particles. The adsorbed molecules impart the needed steric barrier to hinder agglomeration among iron particles. The addition of silica noparticles to stabilize concentrated iron suspensions in oil carriers against aggregation d sedimentation processes is a very efficient mechism under rest conditions. Unfortunately, under shearing, particle settling is facilitated by the breakage of the silica network d compact sediments, which could make difficult the redispersion, are created. It is possible to stabilize magnetorheological fluids against aggregation d sedimentation processes by using a ferrofluid as carrier fluid. The magnetic attraction between the noparticles (magnetically single-domain) d the micron-sized particles (magnetically multi-domain) favors the formation of clouds of magnetite noparticles around each iron one. Acknowledgments. The authors wish to express their appreciation d deep recognition to the scientific merits d hum personality of Prof. Emil Chibowski. He undoubtedly constitutes a role model for young scientists. Fincial support by Ministerio de Educación y Ciencia (Spain) d FEDER funds (EU) under Project No. MAT2005-07746-C02-01, d Junta de dalucía (Spain) under Project FQM410 are gratefully acknowledged. One of the authors (M.T. López-López) also acknowledges fincial support by Secretaría de Estado de Universidades e Investigación (Ministerio de Educación y Ciencia, Spain) through its Postdoctoral Fellowship Program (No. EX2006-0467). 6.

Journal

Annales UMCS, Chemiade Gruyter

Published: Jan 1, 2008

There are no references for this article.