MAGNETIC SUSCEPTIBILITY AND MAGNETIC HYSTERESIS LOOP OF SOME
CARBONYL IRON POWDERS USED IN MAGNETORHEOLOGICAL FLUIDS
ANTONIO J. F. BOMBARD
Escola Federal de Engenharia de Itajubá, Av. BPS, 1303, Itajubá, MG, CEP 37.500-903, BRAZIL
E-mail: [email protected]
INÉS JOEKES
Instituto de Química, Unicamp, C.P. 6154, Campinas, SP, CEP 13.083-970, BRAZIL
E-mail: [email protected]
MARCELO KNOBEL
Instituto de Física Gleb Wataghin, Unicamp, C.P. 6165, Campinas, SP, CEP 13.083-970, BRAZIL
E-mail: [email protected]
Keywords: magnetic properties, carbonyl iron powders, magnetorheological fluids
Abstract
Some commercial carbonyl iron powders, from BASF AG, were characterized through their
magnetic susceptibility and saturation magnetization. The saturation magnetization increases in the
order: CC < HQ < OX < CS < SM. Although all the powders have 96% minimum of iron content,
this result suggests that an investigation at higher magnetic fields, under simultaneous shear is very
important. The reversibility and the reproducibility of the MR effect were satisfactory, confirming
that carbonyl iron powders are good ferromagnetic materials to prepare magnetorheological fluids.
1
Introduction
Magnetorheological fluids (or magnetorheological suspensions, in a physical-chemistry
context) are considered a new material, although their applications were first reported by
Rabinow at the end of the forties. [1] Other interesting applications about MRFs, pictures
showing a microscopic representation and the macroscopic effect of a MRF, can be found at
the web site of Lord Rheonetic magnetic fluids. [2]
MRF contain basically four types of substances: a ferromagnetic powder (such as
carbonyl iron powder, cobalt, or nickel); a carrying fluid (mineral or silicone oils, in general);
a dispersant (to avoid or minimize particle coagulation); and gel-forming additives (as
hydrophobic silica powder, for example). [3]
The rheological properties of such fluids are changed by the application of magnetic fields
in a reversible way. For example, under a field of 250 kA/m, they can present yield stresses of
up to 100 kPa1. Besides, the change of the semi-solid state for fluid state and vice-versa
happens in ~ 5 milliseconds. [4]
These characteristics allow the use of these fluids in controllable mechanical devices, such
as conventional shock absorbers, mountings, brakes and clutches [5]; dampers to reduce
damages in civil engineering constructions caused by earthquakes [6]; polishing in optics [7];
reduction of vibrations in helicopters [8]; and biomedical applications, [9,10] among others.
Since the active material inside MRF is a ferromagnetic powder, it is necessary to
evaluate their magnetic properties in order to better understand the MRF formulation
development.
1
To give some idea of the sizes of these units, the Earth’s magnetic field is typically H = 56 A/m (0.7 Oe) and the most formulations of
toothpaste shows yield stresses around 40 Pa.
2 Experiment
2.1 Magnetic measurements
The powders, without any treatment, were weighted and the magnetization hysteresis loop
were measured using a SQUID (superconducting quantum interference device) type
magnetometer (Quantum Design MPMS XL7). From the hysteresis loop data, magnetic
susceptibility was calculated and other magnetic parameter were obtained.
2.2 Rheological measurements
The yield stress, viscosity curves, and the MR effect (considered in this work as the viscosity
increase when the magnetic field is applied) were measured using a controlled stress
rheometer (Paar Physica MCR 300), with the plate-plate configuration (50 mm) and 1 mm
gap. The magnetic field was applied through a Helmholtz coil system and the magnetic flux
density (B) was measured with a Hall effect probe and a portable gaussmeter. In order to yield
reproducible results, the experiments were performed using triplicate sample. For each
sample, triplicate measurement were obtained.
3 Results and discussion
3.1 Magnetic measurements
The evaluation of the magnetization is important, because several parameters of the
mechanical design of devices using MRFs directly depend on this intrinsic property. Carlson
et al. [5] showed that the minimum active fluid volume V in valve mode devices, like a
damper, is related with the yield stress σy and the viscosity η of the MRF through the
equation 1, where ‘c’ is a factor that depends on the device type, ∆P is the pressure drop and
Q is the flow rate.
V=
12  η  ∆Pσ y
c 2  σ y2  ∆Pη

Q∆Pσ
y


(1)
On the other hand, Shulman et al [11], and Rosensweig [12] associated the yield stress σy with
the magnetic susceptibility χ of the fluid, as showed in the equation 2:
σ y ∝ ϕµ 0 H 02 f (ϕ , χ ) ,
(2)
where ϕ is the volumetric fraction of the magnetic powder, µ0 is the vacuum permeability, H0
is the applied magnetic field strength, and f (ϕ,χ) is a function involving various terms on ϕ
and χ. Since the magnetic susceptibility is defined by the ratio between magnetization M and
the field strength H, the magnetic response of the powders plays an important role in the
performance of MRFs, as demonstrated in previous studies [13].
Figure 1 shows the magnetization M (emu/g) as a function of the applied magnetic field
strength H (Oersted). The results show that the magnetization of the powders increases in the
following order: CC < HQ < OX < CS < SM, but it is difficult to see the difference between
the curves at lower magnetic fields. So, as pointed by Carlson et al. [14], a graphic of M2 vs.
H (square of the magnetization vs. magnetic field) show the best point to energize the MRF,
if a secant line and tangent to the curve is traced. The associated abscissa value is the best
point to work and to design the magnetic circuit of the MR device. Figure 2 shows this type of
plot, applied to the five BASF carbonyl iron powders studied in the present work. Figure 3
shows the magnetic susceptibility χ of the powders as a function of the field strength H in an
expanded view (between 10 and 6000 Oersted) that is the usual field region for practical
devices using MRF’s [14].
1000
HQ
CS
OX
SM
CC
M (emu/g)
100
300
10
200
100
1000
1
10
100
10000
1000
60000
10000
100000
H (Oersted)
Figure 1. Magnetization (M) of the BASF carbonyl iron powders as a function of the applied magnetic field
strength (H).
0,3
70000
HQ
CS
OX
SM
CC
HQ
CS
OX
SM
CC
1E-1
40000
30000
2
M (emu/g)
2
50000
χ = M/H (emu/g.Oe)
60000
20000
1E-2
10000
0
0
1000
2000
3000
4000
5000
6000
7000
8000
H (Oersted)
Figure 2. The square of the magnetization as a
function of applied magnetic field strength of five
BASF carbonyl iron powders: CC, HQ, OX, CS and
SM.
0,003
10
100
1000
H (Oersted)
10000
100000
Figure 3. Magnetic susceptibility χ as a function of the
applied magnetic field strength H of five different
carbonyl iron powders.
Figure 2 shows that for the powders CS and SM, the best point is around 2000 Oe, but for the
powders OX, HQ and CC, this point is ~ 3800 Oersted. Based on this result, we can propose
that MRFs based on CS and SM powders can show better performance than MRFs prepared
with HQ, CC or OX powders at the same mass fraction. The MR effect must be larger at the
same field or the MR effect can be set to the same mechanical response, but with lower power
consumption. At this point we are trying to confirm experimentally this prediction.
Figure 3 shows that the magnetic susceptibility increases in the following order: CC < HQ <
OX < SM. It was expected that the larger the magnetic susceptibility, the larger the yield
stress, according to the equation 2. So, one should expect that a MRF prepared with SM
powder had showed a larger yield stress, compared with CC powder, for example. However,
our preliminary results, working at low fields, did not confirm this prediction [15]. Up to now,
we do not have a good explanation for this contradictory result.
3.2 Rheological measurements
In order to evaluate the reversibility of the MR effect, a MRF was prepared according to the
composition listed on the tables 1 and 2. The viscosity was measured under a constant shear
stress, above the yield stress, during 15 minutes, turning the applied magnetic field ‘on’
(constant magnetic induction = 300 Gauss) and ‘off’. The results are shown in figure 4.
Table 1: Composition of the Gel base.
Oil (Texaco ‘C1’)
90 ml
Oleic acid
16 ml
Hydrophobic silica
4,7354 g
(Cab-O-Sil TS 610)
Table 2: MRF formulation.
BASF carbonyl iron 20,0007 g 66,1% w/w
powder ‘SM’
Gel base
10,2576 g 33,9% w/w
τ = 100 Pa
1000
B = 300 G
Viscosity (Pa.s)
100
10
1
off
1E-1
0
100
200
300
400
500
600
700
800
900
1000
Time (s)
Figure 4. Viscosity measured as a function of the time, under constant shear stress = 100 Pa, and alternating
applied magnetic field ‘on’ (B = 300 Gauss) and ‘off’. MRF formulated with BASF carbonyl iron powder, grade
‘SM’ at 66% w/w.
Figure 4 shows that the MR effect (here, the viscosity increase when the magnetic field is
turned ‘on’, or the viscosity decrease when the field is turned ‘off’) is almost perfectly
reversible, and the effect has good reproducibility. The viscosity increases more than 5000
times, under 300 Oersted.
Table 3: Composition of the three similar MRF based on BASF iron powder ‘CL’.
Formulation
BASF iron powder ‘CL’
Gel base
% w/w iron
A
20,0131 g
10,2237 g
66,19
B
20,0713 g
10,0507 g
66,63
C
20,0676 g
10,0991 g
66,52
To check if our results are reliable, three similar MRF were also prepared, with the same
amount of carbonyl iron powder and all the other constituents of the formulation. The details
of the formulations are described in table 3. The viscosity curves as functions of the shear rate
in the range of 10-4 to 103 s-1 were measured under two magnetic field strengths. Each data set
represents the results measured using a different sample. The result is shown in figure 5.
1000000
CL-A
CL-B
CL-C
100000
300 G
Viscosity (Pa.s)
10000
1000
CL-A
CL-B
CL-C
100 G
100
10
1
1E-1
1E-4
1E-3
1E-2
1E-1
1
10
100
1000
Shear Rate (1/s)
Figura 5: Viscosity curves of a magnetorheological fluid formulated with BASF carbonyl iron powder grade
‘CL’. The points represent triplicate of sample and triplicate of measurement. The lower data points were
measured under B = 100 Gauss, and the upper data under B = 300 Gauss.
As expected, figure 5 shows that the viscosity increases with the magnetic field and decreases
when the shear rate increase (MRF are shear thinning). The MR effect is more pronounced at
lower shear rates. The reproducibility is quite well at 300 gauss and a bit worst at 100 gauss.
4 Conclusion
The results about the magnetic properties suggest that the MRF’s based on BASF carbonyl
iron powders ‘CS’ and ‘SM’ must have a better performance than the other grades, as ‘CC’,
‘HQ’ and ‘OX’. To confirm this prediction, is necessary more investigation, specially at
higher magnetic field strengths and with simultaneous shear. The MR effect of MRF
formulation ‘SM’ showed a reversible behaviour. MRF formulation ‘CL’ showed good
reproducibility.
5 Acknowledgements
We’d like to thank to Prof. Dr. Maria Regina Alcântara (IQ-USP) for the use of Physica MCR
300 rheometer. We’d like also to thank to BASF AG, FAPESP and FAPEMIG, Brazil, (grant
TEC-2271/96), for the financial support. Dr. Al W Friederang is gratefully acknowledged.
References
[1] Rabinow J, “The magnetic fluid clutch”, AIEE Trans., 67 (1948) pp. 1308 - 1315.
[2] http://www.mrfluid.com/
[3] Rogers CA , Ashour O and Kordonsky W. , “Magnetorheological fluids: materials,
characterization, and devices”, J. Intelligent Material Systems and Structures, 7 (1996) pp.
123–130.].
[4] Kormann C, Laun HM, Richter HJ, “MRFs with nano-sized magnetic particles”, Int. J.
Modern Phys. B, 10 (1996) pp. 3167- 3172.
[5] Jolly MR, Bender JW, Carlson JD, “Properties and applications of commercial
magnetorheological fluids”, J. Intell. Material Syst. and Struct., 10 (1999) pp. 5–13.
[6] Wilson EK, "Earthquakes and Chemistry: Two - faced fluids may put a damper on quake
motions in buildings" Chem. Eng., 77(45) November 9, 1999 pp. 27.[7] Kordonski W, Golini
D, "Progress update in magnetorheological finishing“, Int. J. Modern Phys. B, 13 (1999) pp.
2205 - 2212.
[8] Kamath GM, Wereley NM, Jolly MR, "Characterization of magnetorheological helicopter
lag dampers" J. Am. Helicopter Soc.; 44(3) (1999) pp. 234 - 248.
[9] Liu J, Flores GA, Sheng RS; "In-vitro investigation of blood embolization in cancer
treatment using magnetorheological fluids"; J. Mag. Magnetic Materials; 225(1-2) (2001) pp.
209-217.
[10] Gubbels PA, Lokhoff NM, Bakels AP; “Medical electrical lead containing a magnetic
fluid”, EP 0 891 786 A2; European Patent Office; 1999.
[11] Shulman ZP; Kordonsky VI; Zaltsgendler EA; Prokhorov IV; Khusid BM and Demchuk
SA, “Structure, physical properties and dynamics of magnetorheological suspensions”; Int. J.
Multiphase Flow; 12(6); 1986; pp. 935 – 955.
[12] Rosensweig RE; “On magnetorheology and electrorheology as states of unsymmetric
stress”; J. Rheology; 39; 1995; pp. 179 – 192.
[13] Margida AJ, Weiss KD, Carlson JD, “Magnetorheological materials based on iron alloy
particles”; Int. J. Modern Phys.B; 10 (23-24); (1996) pp. 3335-3341.
[14] Carlson JD, Chrzan M, “Magnetorheological fluid dampers”, US Patent 5,277,281;
January 11, 1994.
[15] Bombard AJF, Knobel M, Joekes I, “Evaluation of a magnetorheological suspension in a
shock absorber and in a magnetorheometer”; 8th International Conference on
Electrorheological (ER) Fluids and Magnetorheological (MR) Suspensions; Nice, France;
July 9-13, 2001 (to appear).