Oil organogel system for magnetorheological fluid

Abstract

We introduce a model colloid system comprised of carbonyl iron particles dispersed in oil organogel. The rheological properties of magnetorheological fluids with different particle concentrations were investigated. It was found that the system showed behavior expected from a structured material due to the network of the matrix. The microstructural changes investigated in hysteresis and thixotropic tests demonstrated a compacted chain structure can be obtained by proper shearing. The field-induced shear stress and dynamic modulus proved a promising magnetorheological performance of the suspension.

---

Magnetorheological fluid (MRF) falls in the category of “intelligent” or “smart” materials which is a suspension of micro-sized particles dispersed in a nonmagnetic carrier. The good reversibility and tunable rheology metric of the magneto-responsive colloid make it a promising material for scientific and commercial applications.^1–3 For example, this “intelligence” can be of great use in controllable dampers,^4,5 polishing devices,⁶ torque transducers^7,8 and sensors,⁹ etc.

On account of undesired sedimentation, the formation of a three-dimensional network of magnetic particles within the carrier is necessary to resist the settlement of dispersion induced by gravity. In this regard, the solid analogs of MRF including magnetorheological elastomers^10–12 or gels^13–15 have attracted tremendous attention of researchers for the advantage of stability since they can form strong interconnected networks impregnating magnetic particles. On exposure to an external magnetic field, the particles interact with the surrounding polymer chains and alter their positions.¹⁶ However, the induced dipole interaction between magnetic particles is relatively small compared with the high resistance of the matrix.¹⁷ This generally results in a small magnetoelastic response compared to the magnetorheological fluid. Building on the background, several strategies have been developed to control magnetorheological response of these soft materials, such as particle isotropic orientation,^18,19 exploitation of hyperbranched polymer²⁰ or highly swollen physical polymer matrix,^14,17,21 surface modification of magnetic particles,^22,23 as well as addition of plasticizer.²⁴ These recent developments suggest that an improved magnetorheological performance can be achieved provided high dispersibility and high mobility of magnetic particles in the matrix could be achieved.

With these considerations in mind, we present a new approach to dispersing the particle via the oil organogel formed by the low molecular weight gelator. One typical representative for the gelator molecule is 12-hydroxy stearic acid (HSA),^25–28 which can form soft solid in a very large variety of organic liquid by self-assembled fibrillar networks, at concentrations well below 1 wt%. Differing from the chemical network of polymer employed in magnetorheological elastomer or gel, the molecular networks are non-cross linked chemically, reversible thermally and thixotropic originating from the non-covalent interactions. Thus the soft organogel of HAS retains the mobility of magnetic particles and is of potential in the preparation of magnetorheological materials with respect to its gelation properties. In this paper, we focus on the rheological performance of MRF based on oil organogel system. The off-state viscosity, yield behavior, hysteresis, thixotropy as well as dynamic properties are analyzed. Besides the feasibility of organogel as a candidate carrier, this study may help to shed light on the design of other soft viscoelastic magnetorheological system through low molecular weight gelator.

Carbonyl iron powder (CI, OP quality, with a median particle size of 4–5 μm, >97.8% w/w iron content, supplied by BASF) was chosen as magnetic particles, and mineral oil (0.23 Pa s at 25 °C) was employed as carrier. 12-Hydroxy stearic acid (chemical grade, 99.5% purity) was obtained from Aladdin Company (China). All materials were used as received. MRF samples were prepared by heating the corresponding oil with 12 g L⁻¹ HSA to about 80 °C; homogenization at that temperature till HSA was completely melted. Then CI power was added. For a good dispersion, the mixtures were agitated by an ultrasonic processor (Digital Sonifier, Models 450) fitted with a tip 6.4 mm in diameter for 10 min, and then cooled at 0 °C for one hour of gelation to obtain MRFs. Morphological observation with a scanning electronic microscopy (SEM, Quanta FEG 250) was conducted at 15 kV. The sample was prepared as follows: 1 g MRF was immersed for 90 min in hexane to extract the oil. This operation was repeated 4 times until oil extraction was completed. Afterwards, the sample was dried at room temperature.

The magnetorheological experiments were carried out using an Anton Paar Physica MCR 302 rheometer with a magnetorheological device (MRD) in a plate–plate cell with 20 mm diameter and 1 mm gap. All experiments were run at 25 °C set to ±0.1 °C accuracy by a Peltier temperature control device. The samples were equilibrated in the measuring cell for at least 3 min prior to conducting measurements. Steady state flows were measured by applying a constant shear rate and measuring the viscosity until a steady state value was achieved at each shear rate. Measurements were done with increasing or decreasing shear rate via a log–log scale at different values of magnetic field. In the magnetic sweep experiments,²⁹ the shear rate was held constant, and the magnetic field was varied from 0 to 920 mT and back to 0 mT. The procedure was repeated for different values of shear rate of 0.1, 10 and 50 s⁻¹. In dynamic amplitude sweep measurements, an oscillatory strain of increasing amplitude from 0.001 to 100% was imposed on the sample at fixed frequency of 1 Hz under different values of the magnetic field. 3 Interval Thixotropy Test (3ITT) under osci-rot-osci mode was performed. In the first interval, called rest interval, the sample was under a very small oscillatory strain (γ = 0.005%) at 1 Hz frequency to obtain the reference storage modulus value for the original structure state. In the second interval, called load interval, structure decomposition took place under a high load with high shear rates 100 s⁻¹ for 60 s. In the third interval, called recovery interval, the strain came back to 0.005% again and the structure can recover. These intervals are more than enough for the sample to reach the steady state in each experimental condition.

Fig. 1(a) shows the SEM photograph for magnetic particles extracted from the suspension. It was observed that the particles were connected to each other indicating a feature that the HSA adhered against the surface of the magnetic particles. This microscope link bounded on the magnetic particles originated from the HSA assembled fibers, which was cross-linked by the hydrogen bonds.²⁴ Adhesive properties between particles and matrix would be an important factor for the rheological property of suspensions. Fig. 1(b) illustrates a picture of organogel containing 12 g L⁻¹ HSA (based on previous study, the mineral oil gelation occurred at low content of 2 g L⁻¹ HSA). The gel appears translucent and self-supported under the gravity. This was compared for responding MRF containing 10% particles, as shown in Fig. 1(c), which can still flow with the same inclination. Clearly, the addition of iron particles prevents HSA molecules to form a continuous gelled network. It deserves to be mentioned, the magnetic particles can be immobilized completely by the organogel system with high gelator content, but the resulted gel mixture is difficult to be manipulated. Thus a liquid like MRF with 12 g L⁻¹ HSA was employed for rheological tests.

Fig. 1 (a) SEM photograph for carbonyl iron particles extracted from MRF of 10% concentration. (b) Image of organogel containing 12 g L⁻¹ HAS. (c) Image of the responding MRF of 10% concentration.

As shown in Fig. 2(a), the shear viscosity measured in the absence of magnetic field indicated shear-thinning behavior for the MRFs of varying particle concentrations. At low shear rate, the viscosity was just likely coming out of the power-law region of the curve and flattening off towards infinite viscosity η_∞, without the obvious low shear resistant plateau. Thus, the power law model³⁰ was employed to determine the infinite viscosity η_∞ by fitting the steady shear viscosity profiles, and the obtained fitting constants (with R² > 0.99) were plotted in the inset of Fig. 2(a). It is observed that the high shear rate viscosity did not appear to increase dramatically with the particle concentrations and seems to reach a plateau at 10%. This suggests that the gel matrix of HSA can also greatly change the dependence of viscosity property on particle concentration of MRF.

Fig. 2 (a) Viscosity curves of MRFs containing different CI concentration in the absence of magnetic field. The infinite viscosity η_∞ on the CI concentration by fitting the steady shear viscosity profiles to power law is shown in the inset. (b) Dynamic yield stress of MRFs as functions of magnetic flux density B for different particle concentrations.

The dynamic yield stresses predicted according to Bingham plastic model are plotted as a function of applied magnetic flux density (Fig. 2(b)). With respect to the dependence on the magnetic field, the dynamic yield stresses increased with the intensity of the field. As indicated in ref. 31 and 32, effects of the matrix fluid yield stress must also be taken into consideration at low volume fractions of magnetic particles. In dilute dispersion, for which the average distance between particles is relatively large,¹⁴ inter-particle attractive forces may be insufficient to overcome the resistance of matrix fluid to form chain like structures. For the largest volume fraction examined in the present work, field-induced dynamic yield stress increased sub-quadratically with the external field due to the local saturation of the magnetized particles, and up to about 65 kPa, as observed in ref. 33. At a given field strength, the plastic viscosity increased with the particle concentration (not shown for brevity), also signifying the continuous enhanced structures. Generally, the organogel based on low molecular weight gelator allows for the formation of a week self-assembles microstructure that suppress particle sedimentation without compromising the magnetorheological activity under magnetic field. Moreover, the less dependence of particle concentration on the off-state viscosity makes it possible to prepare MRF with high volume fraction, resulting in high shear stress.

Shear induced microstructural changes were investigated in magnetic sweep tests. Shear stress hysteresis in magnetic field ramp-up and ramp-down at different constant shear rate were shown in Fig. 3. It's found there exist two types of hysteresis behaviors: type I, backward sweep curve lies above forward sweep curve; type II, backward sweep curve lies below forward sweep curve. Regarding to the behavior of type I, which is also observed in ref. 34 and 35 and obvious in MRFs of high particle concentration. On the remove of magnetic field, some chain structures formed at high field have no time to relax and the variation of shear stress lagged behind the excited magnetic field. The gel network is expected to maintain these structures and give rise to a hysteresis. Furthermore, the hysteresis loop area tended to shrink at high shear rates indicating that either the shear stress or the pre-tensioned particle structure in the previous magnetic ramp test relaxed quickly with the resulting smaller structures. Type II, as observed specially in MRFs with low particle concentration (<5%), was unexpected, implying the structure of shearing is different from the scenario of large particle loading. In dilute suspension, the stress increased with increasing field and tended to saturated at around 0.4 T, and slender chain structures were formed. However, the generated structures were not static, but rather, dynamically formed and destroyed to reach equilibrium. Once the field started to decrease, the broken term become more dominant, and then the structure appeared unbalance at fixed shear rate. The fall in shear stress happened during then. It is worth noting that in the low magnetic field the behavior of type I is displayed in all samples, indicating that the structures formed are not completely destroyed by the hydrodynamic forces due to the gel network. The microstructure evolution and rheological behavior of MRF may be more complex in reality depending on shear rate, magnetic field as well as matrix, and as a whole the magnetic sweep hysteresis results might be a variation or a combination of these two generic types according to our study.

Fig. 3 Magnetic field sweep experiments for MRFs of different CI concentrations at shear rates of 0.1, 10, and 50 s⁻¹.

Considering that the microstructure evolution of MR fluid is influence by the network of matrix fluid, the stress in response to shearing and magnetic fields was obtained in the shear rate ramp tests. We see a hysteresis formed by the loading and unloading curves in Fig. 4. In dilute suspension, the shear stress curves in shear rate ramp up and down overlapped in the absence of field, and were divergent distinctly under the field. This behavior may be easily understood in the term of microstructure arrangement. The formed fibrillar chains may undergo slight arrangements at different shear states and evolve into denser structures.³⁶ This evolution leaded to shear stress hysteresis. However for MRF of high concentration, the shear induced microstructure arrangement seemed have limit effect on the field-dependent stress, as displayed in Fig. 4(b). The shear stress hysteresis was obvious at low shear rate range in the absence of magnetic fields, indicating a thixotropic behavior (stress measured with increasing shear rate was higher than those measured with decreasing shear rate). The hysteresis in a low shear rate region demonstrates that the suspension structure was not recovered to that of the initial state. This discrepancy can be ascribed to disrupt of the gel network and a long relaxation of this material.

Fig. 4 Shear rate ramp test of MRFs of (a) 2.5% concentration and (b) 15% concentration in the absence and presence of magnetic field.

To confirm the thixotropic behavior of MRF observed above, the dynamic response to a step-wise change from one steady-state to another was employed as shown in Fig. 5. In the absence of field, the thixotropic behavior of MRF containing gel network implied a time dependent effect of the rheological properties of the suspensions, because after the application of high shear, the gel network of particles would be progressively rebuilt to reach equilibrium. The structure recovery ratio (calculated by the ratio between values of the G′ in the third time interval and the equilibrium value of the G′ obtained at the end of the first time interval) increased from 65% to 90% with the increasing of particle concentration. Thus, it is expected that high concentration will motivate a quick spatial arrangement in the gel network. On the other hand, the application of a magnetic field to MRFs showed a marked increase of storage modulus after high shearing, and the chain organization became stronger, making the sample more elastic. Comparing equilibrium value of the first time interval with the third, G′ increased by a factor of 10 approximately, as a consequence, corroborating the assumption of particle arrangements mentioned in the shear rate ramp tests.

Fig. 5 The thixotropic behavior of MRFs of different CI concentrations, initially at a low oscillatory strain, then subjected to a high shear rate followed by the original strain. (a) In the absence of magnetic field, (b) in the magnetic field of 0.2 T.

Fig. 6 shows results of strain sweep tests. Linear viscoelastic regime (LVR) was observed at a very small strain amplitude (10⁻¹ to 10⁻³) where the mesostructures remained undisturbed. In this region, the system was predominantly elastic with storage modulus (G′) greater than loss modulus (G′′). Even though as low concentration as 2.5% (not shown here for brevity), the solid like behavior still existed in the absence of field indicating an internal structure formed in suspension. The contribution of gel matrix would be responsible for this. As the strain exceeded certain amplitude, G′ began to decrease and fell below G′′ due to the particle movement and destruction of chain structures. The position of critical strain, where G′ and G′′ are equal, represents the transition between the viscoelastic-solid and the viscoelastic-liquid, changes depending on the strength of magnetic field,³⁷ because a larger magnetic field is able to retain a solid structure to higher strains, but the change was negligible in our measurements for all samples. Concerning to the loss modulus demonstrating the heat dissipation in suspensions during dynamic strain, G′′ showed a pronounced strain hardening followed by strain thinning at the high magnetic field. The overshot behavior may be regarded as coming from the balance between formation and destruction of chain structures.³⁸ Furthermore, the maximum G′′ increased almost linearly with the CI concentration up to 15%, as sketched in the inset of Fig. 6(a). Interestingly, a second pseudo-plateau of G′′ was observed at high strain. This was interpreted as a homogeneous rupture of particles networks.³⁹ To better understand the effect of magnetic field on the viscoelastic property, the CI concentration dependence of the plateau value of G′ in LVR is shown in Fig. 6(b). The storage modulus increased with either the CI concentration or the magnetic flux density. For MRF of 20% CI concentration, the field-on storage modulus increased a factor of six compared to that of field-off state, exhibiting a high dynamic efficiency.

Fig. 6 (a) Storage modulus (filled) and loss modulus (opened) vs. strain amplitude under different magnetic field for MRF of 10% concentration. The inset refers to the variation of Max. G′′ vs. CI concentration. (b) CI concentration dependence of storage modulus in the linear viscoelastic regime for MRFs under different magnetic fields. The inset shows the initial region of the field-off state.

Conclusions

The study presents the rheological properties of MRF with carbonyl iron particles interspersed in organogel system and provides an insight into the viscosity, yielding, hysteresis, thixotropy as well as dynamic behaviors in the absence or presence of magnetic field. Two types of hysteresis behaviors were observed. The thixotropic tests showed a compacted chain structure could be obtained by proper shearing. The MRF based on organogel combines the functionality of a weak linked 3-dimensional network matrix with the fine mechanical properties could be of potential interest for technological purposes. Furthermore, there is considerable flexibility in the choice of the low molecular weight gelator, which will open new possibility for design of soft magnetorheological materials.

Notes and references

1. J. de Vicente, D. J. Klingenberg and R. Hidalgo-Alvarez, Soft Matter, 2011, 7, 3701–3710.
2. I. Bica, Y. D. Liu and H. J. Choi, J. Ind. Eng. Chem., 2013, 19(2), 394–406.
3. Magnetorheology: Advances and Applications, ed. N. Werley, RSC, Cambridge, 2014.
4. D. Wang and W. H. Liao, Smart Mater. Struct., 2011, 20, 023001.
5. F. Imaduddin, S. A. Mazlan and H. A. Zamzuri, Mater. Des., 2013, 51, 575–591.
6. W. Kordonski and A. Shorey, Magnetorheological (MR) Jet Finishing Technology, J. Intell. Mater. Syst. Struct., 2007, 18, 1127–1130.
7. B. M. Kavlicoglu, F. Gordaninejad and X. Wang, Smart Mater. Struct., 2013, 22, 125030.
8. S. Dai, C. Du and G. Yu, J. Intell. Mater. Syst. Struct., 2013, 24, 1675–1682.
9. D. H. Reas and J. E. Martin, Adv. Funct. Mater., 2010, 20(10), 1577–1584.
10. Y. Fan, X. Gong, S. Xuan, L. Liu and X. Li, Ind. Eng. Chem. Res., 2013, 52, 771–778.
11. T. Mitsumata, S. Ohori, N. Chiba and M. Kawai, Soft Matter, 2013, 9(42), 10108–10116.
12. V. V. Sorokin, E. Ecker, G. V. Stepanov, M. Shamonin, G. J. Monkman, E. Y. Kramarenko and A. R. Khokhlov, Soft Matter, 2014, 10(43), 8765–8776.
13. R. R. Fuhrer, E. K. Athanassiou, N. A. Luechinger and W. J. Stark, Small, 2009, 5(3), 383–388.
14. T. Mitsumata, A. Honda, H. Kanazawa and M. Kawai, J. Phys. Chem. B, 2012, 116(40), 12341–12348.
15. Y. Xu, X. Gong and S. Xuan, Smart Mater. Struct., 2013, 22, 075029.
16. B. Mordina, R. K. Tiwari, D. K. Setua and A. Sharma, J. Phys. Chem. C, 2014, 118(44), 25684–25703.
17. H. N. An, B. Sun, S. J. Picken and E. Mendes, J. Phys. Chem. B, 2012, 116, 4702–4711.
18. J. Wu, X. L. Gong, L. Chen, H. Xia and Z. Hu, J. Appl. Polym. Sci., 2009, 114, 901–910.
19. A. Boczkowska, S. F. Awietjan, S. Pietrzko and K. J. Kurzydłowski, Composites, Part B, 2012, 43, 636–640.
20. V. G. Vasiliev, N. A. Sheremetyeva, M. I. Buzin, D. V. Turenko, V. S. Papkov, I. A. Klepikov, I. V. Razumovskaya, A. M. Muzafarov and E. Yu Kramarenko, Smart Mater. Struct., 2016, 25, 055016.
21. J. E. Kim and H. J. Choi, IEEE Trans. Magn., 2011, 47, 3173–3176.
22. I. A. Polunina, V. I. Roldughin, G. S. Matrosova, S. A. Sosnina and K. E. Polunin, Colloids Surf., A, 2004, 239, 95–99.
23. W. Q. Jiang, J. J. Yao, X. L. Gong and L. Chen, Chin. J. Chem. Phys., 2008, 21, 87–92.
24. J. Wu, X. L. Gong, Y. Fan and H. Xia, J. Appl. Polym. Sci., 2012, 123, 2476–2484.
25. P. Terech, D. Pasquier, V. Bordas and C. Rossat, Langmuir, 2000, 16, 4485–4494.
26. M. A. Rogers, A. J. Wright and A. G. Marangoni, Curr. Opin. Colloid Interface Sci., 2009, 14, 33–42.
27. J. Zou, D. S. Frost and L. L. Dai, Colloids Surf., A, 2012, 414, 477–485.
28. M. Burkhardt, L. Noirez and M. Gradzielski, J. Colloid Interface Sci., 2016, 466, 369–376.
29. L. J. Felicia and J. Philip, Langmuir, 2012, 29(1), 110–120.
30. D. I. Santiago-Quinonez and C. Rinaldi, Soft Matter, 2012, 8, 5327–5333.
31. J. P. Rich, G. H. McKinley and P. S. Doyle, Langmuir, 2012, 28(8), 3683–3689.
32. J. P. Rich, P. S. Doyle and G. H. McKinley, Rheol. Acta, 2012, 51, 579–593.
33. F. F. Fang, H. J. Choi and M. S. Jhon, Colloids Surf., A, 2009, 351, 46–51.
34. L. J. Felicia and J. Philip, Langmuir, 2013, 29, 110–120.
35. L. J. Felicia and J. Philip, Langmuir, 2015, 31(11), 3343–3353.
36. L. Shan, K. Chen, M. Zhou, X. Zhang, Y. Meng and Y. Tian, Smart Mater. Struct., 2015, 24, 105030.
37. J. Claracq, J. Sarrazin and J. P. Montfort, Rheol. Acta, 2004, 43(1), 38–49.
38. K. Hyun, S. H. Kim, K. H. Ahn and S. J. Lee, J. Non-Newtonian Fluid Mech., 2002, 107, 51–65.
39. I. Arief and P. K. Mukhopadhyay, J. Magn. Magn. Mater., 2016, 397, 57–63.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra24257j

---