Ionic liquids as potential carriers of low viscosity magneto-rheologicalfluids

Carlos Guerrero-Sanchez,~ Armando Ortiz~Alvarado,a Ulrich S. Schubert*a.~~aLaboratory of Macromolecular Chemistry and Nanoscience, Eindhoven University of Technology,

P. 0. Box 513, 5600 MB Eindhoven, The Netherlands;bDutch Polymer Institute, P. 0. Box 902, 5600 AX Eindhoven, The Netherlands;

Cloniqa Technologies, Eindhoven, The Netherlands;dLaboratory of Organic and Macromolecular Chemistry, Friedrich-Schiller-University Jena,

Humboldtstr. 10, D-07743 Jena, Germany

ABSTRACT

Based on the latest investigations on the formulation of new magneto-rheological fluids, it is envisioned that the use ofionic liquids as carriers of magneto-rheological fluids will open new possibilities of applications for these smart fluidsdue to the fact that their physical and chemical properties can be fine-tuned in a broad range. This contribution addressesone potentially important advantage of magneto-theological fluids which use ionic liquids as novel carriers. Inconnection with this, magneto-rheological fluids with a low viscosity in the off-state without compromising otherproperties of the formulations (e. g., sedimentation of the dispersed magnetic particles, liquid state of the carriers in abroad range of temperatures) are often required for specific applications. In this regard, ionic liquids of low viscosity canbe very useful in the development of such magneto-rheological fluids. Thus, this contribution reports on the magnetorheological properties of iron(Il, 111) oxide particles dispersed in the ionic liquid 1-ethyl-3-methylimidazoliumthiocyanate (a low viscosity ionic liquid) in the temperature range from 20 °C to 80 °C. The experimental results haverevealed that the apparent viscosity of the dispersion slightly changes with the temperature when a constant magneticfield is applied and its value mainly depends on the shear rate and the strength of the magnetic field. The viscosity of thedispersion remains practically unmodified with both the temperature and the magnetic field intensity as the magneticsaturation of the material is reached; in this regime the viscosity will only depend on the applied shear rate. In contrast,the yield stress values of the dispersion as well as the corresponding shear stress vs. shear rate curves have shown aninverse behavior with temperature for a constant magnetic field.

Keywords: Low viscosity magnetorheological fluids, smart materials, low viscosity ionic liquids, composite materials,magnetoresponsive fluids

1. INTRODUCTION

Magneto-rheological (MRFs) are dispersions of magnetic particles in a liquid carrier, which can reversibly andinstantaneously change from a liquid state to a semi-solid or plastic state in the absence/presence of a magnetic field.Moreover, in their on-state (i. e., semi-solid state), these fluids show a viscoplastic behavior which is characterized by amagnetic field-dependent yield stress. These main characteristics render MRFs useful for several controlledelectromechanical applications.~ The properties of MRFs are mainly defined by the composition and nature of itsmagnetic particles, its dispersion additives, and its fluid carrier. Thus, considerable efforts have been taken to optimizethe composition of these fluids for specific applications.~’1 Regarding the magnetic material, particles obtained fromcarbonyl iron are mainly preferred in commercial magneto-rheological formulations. Moreover, the dispersion of themagnetic particles in the carrier liquids is achieved through the use of different additives which inhibit their settling andagglomeration. The dispersion additives can vary widely and may include substances such as organoclays, polymer

*u.s.schubert@tue.nl; phone +31 402475303; fax +31 402474186; www.schubert-group.com

c.guerrcro~ioniqa.com; phone +31 40 247 3269; fax +31 40 247 4186; www.ioniqa.com

modified metal oxides, carboxylate soaps or other thixotropic and surface active agents.’~ The selection of a suitableliquid carrier during the formulation of MRFs is another significant aspect influencing the behavior and characteristics ofthese smart fluids. ~n this regard, suitable liquid carriers are mainly selected based on their rheological properties (liquidcarriers with a relatively low viscosity) and thermal stability. So far, a relatively narrow range of liquid carriers has beenutilized for the preparation of MRFs, which includes petroleum based oils, silicon oils, mineral oils, synthetichydrocarbon oils, glycols and other alcohols, and ethers)~ This important aspect may restrict the use of MRFs inspecific applications, for instance in terms of viscosity and chemical compatibility with other substances or materials insome magneto-rheological devices (e. g., MRFs in medical devicest5~ and magneto-rheological mounts~’1). Based on thelimitations of conventional MRFs, the use of ionic liquids (ILs) has been recently proposed for the preparation of MRFswith enhanced properties.17”1 This is mainly due to the interesting and intriguing properties shown by ILs (c. g.,negligible vapor pressure and flammability, remarkable chemical and physical stabilities in a broad range oftemperatures, liquid state in a range of approximately 400 °C, possibility of tuning their properties by varying the natureof their ions, etc.)i121 Thus, it is thought that ILs will play an important role in the preparation of advanced MRFs in thecoming years, which might help to overcome current limitations of conventional MRFs. From this point of view, MRFsbased on ILs might also open new possibilities of applications (and/or improve the existing ones) for these smart fluidsbecause their physical and chemical properties can be fine-tuned in a broad range by varying the nature of the iLs and/ortheir corresponding additives. Nevertheless, this new approach to prepare MRFs is still in its infancy, and additionalinvestigations must be conducted on this topic in order to reach an acceptable understanding of the fundamentalscientific aspects of these new MRFs and of their “potential” limitations in terms of applications. Hence, thiscontribution addresses the magneto-rheological properties of iron(Il, III) oxide particles dispersed in the hydrophilic IL1-ethyl-3-methylimidazolium thiocyanate (EMI-SCN) — a low viscosity IL — in the temperature range from 20 °C to80 °C. The relatively low viscosity and remarkable stability of EMI-SCN could yield very suitable MRFs forapplications where a low viscosity in the -off-state is required (e. g., MRFs in medical devices151).

2. METHODOLOGYThe MRF investigated in this contribution was prepared and characterized following similar procedures as reportedelsewhereJ7~ Iron(I1, Ill) oxide (magnetite) powder (Aldrich, <5 urn, 98%, density 4.8-5.1 g cm ~ (25 °C)) was dispersedin the IL 1-ethyl-3-methylimidazolium thiocyanate (EMI-SCN) (Aldrich, ?95%). EMI-SCN is a hydrophilic IL with adensity and viscosity values of 1.11 g cm3 (20 °C) and 22 mPa s (room temperature), respectively. The MRF used in thisinvestigation had a content of 40% wt of magnetite particles dispersed in the IL EMI-SCN. The dispersion process wasperformed in a cylindrical polyethylene container using polyethylene stirring paddles. The mixing process was achievedby mechanical stirring at a rate of 2400 rpm for 15 mm at 20 °C. Rheological measurements of the pure IL EMI-SCNwere carried out under steady shear (300 rpm) and under several shear rates at the desired temperatures using a PhysicaMCR 301 rheometer (Anton Paar) coupled with a convection oven (CTD-450) using a cone and plate measuring system(CPSO, 50 mm in diameter). Magneto-rheological measurements of the prepared MRF were carried out at the desiredtemperatures under steady shear and under several shear rates using the aforementioned Physica MCR 301 rheomctercoupled with a commercial magneto-rheological device (MRDI 80/IT magneto-rheological cell). The homogeneousmagnetic field was oriented perpendicular to the shear flow direction. A circular parallel plate measuring system(PP2OIMR made of nonmagnetic metal to prevent the occurrence of radial component of magnetic forces on the shaft ofthe measuring system) with a diameter of 20 mm and a gap of 1 mm between the plates was used. Before performing themagneto-rheological measurements, the prepared MRF was additionally re-mixed in order to ensure the homogeneity ofthe dispersion (e. g., no supernatant clear layer formation (sedimentation of the dispersed magnetic particles) wasobserved for at least two days).

3. RESULTS AND DISCUSSIONThe first investigations in this study correspond to the rheological measurements of the liquid carrier (EMI-SCN) of theprepared MRF. Figs. la and lb show that EMI-SCN has a Newtonian behavior1131 in the temperature range from 20 °C to80 °C and at the investigated shear rate range; the shear stress increases linearly with the shear rate, and therefore theviscosity of the fluid remains relatively independent (constant) of the shear rate.

a20C 035C ~ 50C A65C 080’C •20C 035.C A5~C A65C 05Qc

‘aa.

•a 0S A0 0

Fig. I. Temperature effect on the rheological characteristics of the hydrophilic ionic liquid I -ethyl-3-methylimidazoliumthiocyanate (EMI-SCN). a) Shear stress vs. shear rate, and b) viscosity vs. shear rate both in the temperature range from20 °C to 80 °C.

In Fig. 2a the variation of the viscosity with temperature for the hydrophilic IL EMI-SCN is shown for a broad range oftemperature (from -60 °C to 200 °C). Note that at room temperature conditions (—20 °C) this IL has a viscosity value of22 mPa s, which turns this fluid in a good candidate to be investigated as a liquid carrier for the preparation of lowviscosity MRFs. Even though the dependence of the viscosity with temperature shows a non-linear behavior in thetemperature range shown in Fig. 2a, for shorter temperature intervals the viscosity (rj) of EMI-SCN might obey anArrhenius model in good agreement as shown in Fig. 2b for the temperature range from 20 °C to 80 °C(i. e., 20 °C = 0.00341 KL and 80 °C = 0.00283 K’~’).

*

a.a

0C,

>

C,

0.

0 -4.5C.C

Fig. 2. a) Variation of the viscosity with temperature for the hydrophilic ionic liquid I-ethyl-3-methylimidazoliumthiocyanate (EMI-SCN). b) An Arrhenius model is fitted to the obtained experimental measurements of viscosity vs.temperature for EMI-SCN in range from 20 °C to 80 °C (i. e., 20 °C = 0.00341 K’ and 80 °C = 0.00283 K’.

The flow curves of the investigated MRF at several temperatures and under the influence of a magnetic field of differentintensities (H) are displayed in Fig. 3. Note that the shape of the curves resembles more to a pseudo-plastic system with ayield stress than to the Bingham model131 conventionally used to describe MRFs under the presence of a magnetic field.This is also demonstrated in Fig. 3 (bottom right) for a selected case of the obtained experimental data. In this specificcase, a pseudo-plastic model (e. g., Ostwald (power law)),~31 and the Bingham model are fitted and compared to theexperimental data. From this latter plot, it is clear that the pseudo-plastic model is more suitable to predict theexperimental data than the Bingham model; similar findings were obtained for the rest of the obtained experimental data.

EMI-SCNa)

aa.

0

a(0

100

10

1

0.1

0.01

a0A0

0.1

a.

0.01

0 ~ ~‘ 0’’’’[~ 0;

b) . :

• S a a a ;. . a

o 0 0 0 0 ,o P 0

‘A A A ‘A 4 AA A 4 A A. 4, .40 0 9 0 .0 0 0

10 100Shear Rate [lie]

• a. a a a,..

O 0, 0 .0 0 0

A A A A A A1’’

A A A ,A A. A,,0 0, 0 0 0 9

1000 10 100 1000Shear Rate [lid

100000

10000

1000

100

10

‘EMI.scH:.3.5

.4

EMI-SCN

a)

I P . I I.

b)

.5

.80 .60 -40 -20 0 20 40 60 80 100 120 140 160 180 200 220T rci

In q 2433.51T - 12.112

0.0028 0.0029 0.003 0.0031 0.0032 0.0033 0.0034 0.0035

in~

120020 C

120035 .C

a.800

U)~ 600C

U)

400

200

0

~ci°

~. B

50.C

4

o• H 0 KNmO H =73 KNmA H = 145 KNm

8HZ1SKNmo H ° 284 KNm

• H = 351 KNm

I

88

8A

ci .0ci 0

- CH=OKNmOH73KNmaHal45KNm

• 8H2I5KNmoH2B4~(fiJm

•H~3S1KNm

B B B ‘B B B

100 200 300 400 500 600Shear Rate tile]

• 88

A

o ° I~l

•HOKNmOH73KNm

*H=i45KNm

8 H = 213 KNm0H2B4KPJm

•H=3SIKA/m

S a

300 400Shear Rate (liz]

ci

• H 0 KNm0 H a 73 KPJm

* H a 145 KNm8 H 215 KNma H 284 KNm

• H a 351 KNm

B

84

• 0

B H- 0 K.SJm0H73K.&JmA H. 145 KNm

• 8H215KPJm• OH.2B4KNm

B H • 351 KPJm

I

• Cap data— Blngham

4 Ostwa/d

Fig. 3. Flow curves of the investigated magneto-rheological fluid (MRF) at different temperatures and intensities of amagnetic field (H). Bottom right plot compares the predictions of a pseudo-plastic model (Ostwald (power law model))and the Bingham model with the experimental data of the MRF measured at 80°C and H 215 KA m’.

Yield stresses of the MRF at the investigated conditions (different temperatures and magnetic field intensities) wereestimated by using a yield stress analysis method incorporated in the software package of the utilized commercialrheometer. This analysis method estimates the yield stress by calculating the flexion point in a stress-strain curve in alogarithmic plot. The evaluation places a regression on the input data and checks for the point with the largest distance tothe regression curve; this point is taken as the yield point. The yield stresses of the MRF obtained by this latter methodare displayed in Fig. 4. As observed in the plots of Figs. 3 and 4, the shear stress vs. shear rate curves as well as the

1000 ••~ 88A

~

B• •

.8

A

0

B

0

B

.8 8

~ 8 A8

A~ci

ci ci. ci~pciO ci

BBB B B I B B

0 100 200 300 400 500 600Shear Rate (1/si

88

A

0 100 200 300 400Shear Rate ti/SI

65.c

8

ci ci

500 600

0

! !~poOci ci

1000

600I,C

~ 600itC.0~° 400

200

0

1000

900

800

~ 700

600

~i 500C~ 400U)

300

200

100

0

700

600

500aa.

~ 400

U)~ 300C

U)200

100

0

1000

900

800

a~ 700

600

“ 500C

.~ 400U)

300

200

100

0

600

500

C

9~.400aC

~ 300itC.0~“ 200

100

8A

A A

ci• ci ci

I

80CC

0

.88 8

A

100 200

8A

0

500 600

ocic ci 0;

80 C H 215KAIm

0 100 200 300 400Shear Rate (us]

500 600 0 75 150 225 300 375

Shear Rate (1/al

450 525 600

estimated yield stress values of the investigated MRF reveal an inverse behavior with temperature for a fixed magneticfield intensity.

900 V 900

800 8000

700 0 700 ~

~‘600 V s o 600 A •H’OKPJm— V — o 0H74KNm~ soo 035C C~ 500 A~A146t(~D.VJm

0 0 A50C — 0 0

U, 0 ~°4Q0 ~I+~215KNm400 0 ~65C ~ °H284KAIm

300 V V V 080 C ~ 300 0 • H 351 KA1m

200 V V 200

100 100

0 0 •~0 50 100 150 200 250 300 350 400 ~ 20 25 30 35 40 45 50 55 60 65 70 75 80 85

Magnetic I’IeId LKA!rn] Temperature rc

Fig. 4. Yield stresses of the investigated magneto-rheological fluid (MRF) as a flinction of temperature and of magnetic fieldintensity (H).

Furthermore, the experimental results of this study have revealed that the apparent viscosity of the investigated MRFslightly changes with the temperature when a constant magnetic field is applied, and its value mainly depends on theshear rate and the strength of the magnetic field. Moreover, the viscosity of the MRF remains practically unmodifiedwith both the temperature and the magnetic field intensity as the magnetic saturation of the material is reached; in thisregime the viscosity will only depend on the applied shear rate. These effects can be observed in the plots of Fig. 5.

4. CONCLUSIONS

In this contribution, the magneto-rheological properties of magnetite particles dispersed in the hydrophilic IL EMI-SCNwere investigated in the temperature range from 20 °C to 80 °C. The use of ILs for the preparation of MRFs allowedvery conveniently to perform detailed temperature dependence magneto-rheological studies in this kind of dispersionsdue to the fact that ILs are in the liquid state in a broad range of temperatures, and they have a remarkable thermalstability and negligible vapour pressure. in addition, it was demonstrated that ILs of low viscosity may be useful in theformulation of MRFs, where a low viscosity in the off-state of these fluids is required without sacrificing other properties(e. g., sedimentation of the dispersed magnetic particles, liquid state of the carriers in a broad range of temperatures). Theexperimental results have revealed that the apparent viscosity of the dispersion slightly changes with the temperaturewhen a constant magnetic field is applied and its value mainly depends on the shear rate and the strength of the magneticfield. Moreover, the apparent viscosity of the investigated dispersion remains practically unmodified with both thetemperature and the magnetic field intensity as the magnetic saturation of the material is reached; in this regime theviscosity will only depend on the applied shear rate. In contrast, the yield stress values of the dispersion as well as thecorresponding shear stress vs. shear rate curves have shown an inverse behavior with temperature for a constant magneticfield.

The possibilities to utilize ILs as a novel carries of MRFs seem to be unlimited due to the fact that, so far, there arearound 300 lLs commercially available, and more than one million of new ILs could be easily synthesized, covering aremarkable broad range of potential properties.~21 This is without considering the mixture of two or more ILs to obtainmore suitable (or the ultimate) liquid carriers for MRFs. Moreover, the use of ILs as both carriers and surfactants tostabilize diverse heterogeneous systems~41 — including magnetic particles~71’1 — has been demonstrated recently. This factmight lead to avoid the use of additives and/or stabilizing agents in MRFs formulations, which may diminish theresponse of the fluids to an applied magnetic field. Furthermore, it has been described that several polymers andpolymeric surfactants can be used in combination with ILs to form stable heterogeneous systems,~5~61 which could leadto the development of novel magneto-rheological polymers (or elastomers) and composite materials.~7~9~

100 H = 0 KAIm10.000 H=74I~AJm

HI46KAIm

~II00~II

00~~I

_0~1I_~

0~a100~i* i

- - --

‘III000 ~ ~

001001

0~II- 0~0~~l-

0010011

• 0~I~

II.00011

00 I•001

001001

0

Fig. 5. Viscosity curves of the investigated magneto-rheological fluid (MRF) as a function of the shear rate at differenttemperatures and intensities of a magnetic field (H).

Based on the new approach recently proposed in the literature~7~ for the formulation of new MRFs based on ILs, it isenvisioned that ILs will play an important role in the preparation of MRFs with improved properties in the coming years,which might help to overcome current limitations of conventional MRFs formulations. From this point of view, MRFsbased on ILs might also open new possibilities of applications (and/or improve the existing ones) for these smart fluidssimply because their physical and chemical properties can be fine-tuned in a broad range by varying the nature of the ILsand/or their corresponding additives. Even though this new approach to prepare MRFs seems very promising, this “new”field of research is still in a very early stage, and additional investigations must be performed in order to reach anacceptable understanding of the fundamental scientific aspects of these new MRFs and of their effects and “potential”limitations in terms of applications (e. g., in engineering devices or, perhaps, in medical therapies). Future work in this

I~ip..s•B.B

BB

BB

V 00

00~!

1

0.01

10,000

100

0.1 1

100

0

10 100 1000

Shear rate [lIsi

to

a.

00L0>

toaa.

00

>

to

a.

00

>

0.1 1 10 100 1000

Shear rate [1151

10,000 VHa 215 KAhn

0011

011111

100 -0 111111

0111

V V ill.

a.

0

>

cca.

to0

>

to‘5a.

to0005

>

•20 ‘C035 ‘C

50 ‘CA 65’C080 ‘C

• 20 ‘C035 ‘C

50 CA 65 C080 ‘C

cc 20 ‘C035 ‘CA 50 ‘C

A 65 ‘c080 ‘C

0.1 VI • 10

Shear rate Ills]

100 1000

10.000 - H a 284 KAIm

•20 ‘C035 ‘c‘50 ‘CA 65 ‘C080 ‘C

S 20 ‘C035 ‘C

~ 50 ‘CABS ‘C080 ‘C

B 20 ‘C

a35 ‘C*50 ‘C

ASS ‘C080 ‘C

100

0.1 1 10 100 1000

Shear rate [ifs]

Ha 351 KAIm10,000

100

I

0111

oh

0.1 1 10 100 1000

Shear rate [Ifs]

0.1 1 10 100 1000

Shear rate [lid

direction might include an exhaustive screening of ILs currently known and/or commercially available — as well as someof their mixtures — for the preparation and detailed characterization (in terms of properties and applications) of these newMRFs formulations.

ACKNOWLEDGEMENTS

This research is supported by the Technology Foundation STW, applied science division of NWO and the technologyprogramme of the Ministry of Economic Affairs. The authors also thank the Dutch Polymer Institute for financialsupport.

REFERENCES

Goncalves, F. D., Koo J. H. and Ahmadian, M., “A review of the state of the art in magnetorheological fluidtechnologies part I: MR fluid and MR fluid models,” The Shock and Vibration Digest 38, 203-2 19 (2006).

121 Choi, J. S., Park, B. J., Cho, M. S. and Choi, H. J, “Preparation and magnetorheological characteristics of polymer

coated carbonyl iron suspensions,” J. Magn. Magn. Mater. 304, e374-e376 (2006).131 López-López, M. T., Zugaldia, A., Gonzalez-Caballero, F. and Durán J. D. G., “Sedimentation and redispersion

phenomena in iron-based magnetorheological fluids,” J. Rheol. 50, 543-560 (2006).~ López-López, M. T., Durán, J. D. G., Delgado, A. V. and Gonzalez-Caballero, F., “Stability and magnetic

characterization of oleate-covered magnetite ferrofluids in different nonpolar carriers,” 3. Colloid Interface Sci. 291,144-151 (2005).

~ Gudmundsson, K. H., Jonsdottir, F. and Olafsson, S., “The viscosity of magneto-rheological fluids in a prosthetic

knee actuator,” 11th International Conference on New Actuators (ACTUATOR 2008), 481-484, June 9-11(2008),Bremen, Germany.

16] Barber, D. F., Carison, J. D. and Wilder, R., “Prototype MR mounts utilizings glycol-based MR fluids,” 11th

international Conference on New Actuators (ACTUATOR 2008), 830, June 9-11(2008), Bremen, Germany.171 Guerrero-Sanchez, C., Lara-Ceniceros, T., Jimenez-Regalado, E., Rasa, M. and Schubert, U. S.,

“Magnetorheological fluids based on ionic liquids,” Adv. Mater. 19, 1740-1747 (2007).~ Guerrero-Sanchez, C., Lara-Ceniceros, T., Jimenez-Regalado, E., Rasa, M. and Schubert, U. S., “Ionic liquids as

carriers of magnetorheological fluids,” 11th International Conference on Magnetic Fluids (5P7), July 23-27 (2007),Kosice, Slovakia.

191 Dodbiba, G., Park, H. S., Okaya, K. and Fujita, T., “Investigating magnetorheological properties of a mixture of twotypes of carbonyl iron powders suspended in an ionic liquid,” J. Mang. Magn. Mater. 320, 1322-1327 (2008).

Ito] Guerrero-Sanchez, C., Lara-Ceniceros, T., Erdmenger, T., Jimenez-Rega]ado, E. and Schubert, U. S., “Recent

developments on magnetorheological fluids based on ionic liquids,” 11th International Conference onElectrorheological Fluids and Magnetorheological Suspensions, August 25-29 (2008), Dresden, Germany.

~ Akin, F., Gradi, .1. and Peukert, W., “First Studies on the Rheological Behavior of Suspensions in Ionic Liquids,”

Chern. Eng. Technol. 29, 1347-1354 (2006).(12] Deetlefs, M. and Seddon, K. R., “Ionic liquids: fact and fiction,” Chim. Oggi 24(2), 16-23 (2006).fl31 Macosko, C. W., [Rheology: Principles, Measurements, and Applications], VCH Publishers, New York (1994).fl41 Guerrero-Sanchez, C., Erdmenger, T., ~ereda, P., Wouters, D. and Schubert, U. S. “Water-soluble ionic liquids as

novel stabilizers in suspension polymerization reactions: engineering polymer beads,” Chem. Fur. J. 12, 9036-9045(2006).

1151 Winterton, N., “Solubilization of polymers by ionic liquids” J. Mater. Chem. 16, 428 1-4293 (2006).fl61 Ueki, T. and Watanabe, M., “Macromolecules in ionic liquids: progress, challenges, and opportunities,”

Macromolecules 41, 3739-3749 (2008).I~~1 Hoibrey, J. D., Chen, 3., Turner, M. B., Swatloski, R. P., Spear, S. K. and Rogers, R. 0., “Applying ionic liquids for

controlled processing of polymer materials,” ACS Symp. Ser. 913, 71-87 (2005).~ Swatloski, R. P., Holbrey, J. D., Weston, J. L. and Rogers, R. D. “Preparation of magnetic composites using ionic

liquids,” Chim. Oggi 24(2), 3 1-35 (2006).

P91 Sun, N., Swatloski, R. P., Maxim, M. L., Rahman, M., Harland, A. G., Haque, A., Spear, S. K., Daly, D. T. and

Rogers, R. D. “Magnetite-embedded cellulose fibers prepared from ionic liquid,” J. Mater. Chcm. 18, 283-290(2008).