Patterned electromagnetic alignment of magnetic nanowires

Abstract

A combination of electromagnetic alignment and topological pattern assisted alignment to position magnetic nanowires, which is referred to as the Patterned Electromagnetic Alignment (PEA), is developed and examined. Electrodeposited, FeNiCo nanowires with different lengths were used as the test nanomaterial, and the microscale grooved surface was formed by UV nanoimprint lithography. The accuracy of the PEA with FeNiCo nanowires was evaluated by measuring the deviation angle from the direction of the magnetic field line for different magnetic field strengths and nanowire lengths, and a statistical alignment distribution was reported for different nanowire length groups. The results were compared with those of the electromagnetic alignment on flat surfaces and in grooved-patterned substrates without electromagnetic alignment. Overall, the deviation angle for the PEA was lower than that for the electromagnetic alignment when all other experimental conditions were identical, indicating that the alignment accuracy along the direction of the magnetic field lines was enhanced in the presence of surface micro grooves. This can be attributed to the fact that, upon attachment of nanowires to the substrate surface, the surface micro grooves in the PEA add additional deterministic characteristics to the otherwise stochastic nature of the nanowire deposition and solvent evaporation processes compared to the sole electromagnetic alignment.

Graphical abstract

The paper presents a new method to improve the alignment of magnetic nanowires by combining electromagnetic alignment with microscale surface groove patterns, which was named patterned electromagnetic alignment (PEA).

1. Introduction

Ferromagnetic nanowires have an inherently larger remnant magnetization compared to nanoparticles due to their high aspect ratio shape, that permits its use in low field environments for cell manipulation [1, 2]. In circuits, Ni nanowires have been used in a tunable stop-band notch filter [3], and they hold promise in the field of microfluidics, as an actuation element for micro-valves, pumps, and mixers [4]. As metal electrodes for electrochemical detection in miniature devices, typically noble metal nanowires (e.g., Pt, Au, Ag) have been used, as comprehensively reviewed by Gencoglu and Minerick [5]. There is not only a need to find a less expensive replacement, but transition elements that are ferromagnetic also provide the added benefit to magnetically manipulate the nanowires into a device [6–9].

Ji et al. [10] recently reviewed the methods to generate interconnects of nanowires in devices, that includes a two-step process where the nanowire is aligned and then pinned to remain in place. Efforts to align and position template-grown nanowires can be classified into three types depending on the forces used to control the orientation of the nanowires: forces by fluid flow, field forces, and surface forces (or surface patterns). In the first type, preferential orientation of nanowires was induced via either evaporation of the solvent [11–13], spontaneous spreading/wetting/flow of the nanowire solution [14–19] or brushing of the nanowire solution [20]. Alignment of the nanowires on an area as large as 4 cm² have been demonstrated by using spontaneous spreading of the droplets of the nanowire solution [15]. Drawbacks of using fluid flow included the high average misalignment angle (typically more than 20°) of the nanowires [15] and the difficulty in apply a uniform capillary or shear force to the nanowires [21]. Also, the large shear force upon brushing led to the deformation or breaking of the nanowires [20]. In the second type, either optical [22–24], electric [25–29] or magnetic [30–32] field force was used to control the orientation of nanowires. An optical tweezer could align nanowires into arbitrary patterns generated by computer aided design. However, this method is limited to a small scale assembly of nanowires [24]. The applied electric or magnetic fields between two large electrodes or magnets could align a large number of nanowires suspended in fluid. Beside the presence of a misalignment, chaining of more than one nanowires inside the electric [28] or magnetic field [32] was observed, in particular when a high concentration of nanowire solution was used. The last type of alignment techniques employs surface forces via chemical [31, 33–36], topological surface patterns [37], and shear force induced by surface [38, 39], to assemble nanowires within specific locations of the substrate. These techniques overall show a low ratio of aligned nanowires when they are not combined with other techniques [34, 40]. A successful example in this type of alignment techniques is the use of strain release on highly stretchable substrates where an accurate alignment of nanowires on a flexible substrate as large as 2 inch × 6 inch was achieved [41]. However, this method requires high elasticity of the substrate as well as application of large forces between the elastic substrate and the nanowires, which significantly limits the applicability of this method.

High precision and large scale alignment of nanowires has been achieved by combining more than one of the alignment techniques mentioned above. Examples include alignment using the sites of patterned microscale electrodes [42] or nanoscale magnet arrays templates [43] (and magnetic trapping using microscale magnet arrays [1]) and nanoscale topographical grooves patterns combined with capillary force [44]. The first two examples demonstrated a micron level precision, but they required complicated low throughput fabrication processes for microelectrodes or micro/nanomagnet arrays [1, 42, 43]. The method combining surface chemical patterns with capillary force showed that short nanowires with the length less than 100 nm can be positioned within the chemically-patterned trenches with high angular as well as location accuracies [44]. However, the applicability of this method for nanowires longer than several microns needs to be further studied, which would make it easier to connect the aligned nanowires to external micro to macroscale electrodes.

In this paper, a new method is presented that combines a magnetic field force with topological microscale grooves to control and achieve high precision alignment of magnetic nanowires with various lengths. Although individually both methods have been demonstrated in the literature, [45–50] the combined effect can significantly improve the alignment accuracy for nanowires. To differentiate this method with others, it is referred to as patterned electromagnetic alignment (PEA).

2. Experimental

2.1. Nanowires fabrication and preparation

FeNiCo composite nanowires were pulsed electrodeposited in a nanoporous alumina template with 0.2 μm diameter pores and up to 30 μm in length. Details on the electrodeposition process can be found in [51]. After the electrodeposition, the alumina template was dissolved with a 2 M NaOH solution overnight. The dissolved nanowires were then rinsed with water and ethanol three times. After dissolving the alumina template, the nanowires were kept in ethanol. Adding a surfactant such as cetyltrimethylammonium bromide (CTAB) could help reducing the aggregation of the nanowires. However, based on our observations, there were still a significant amount of aggregated nanowires even in the presence of surfactant, especially at large nanowires concentrations [52]. Some nanowires were broken during the preparation process, so that the nanowires had different lengths. Figure S1 of the Supporting Information presents the detailed distribution of lengths of all the nanowires used in this study. The diameter of nanowires cannot be determined accurately with optical micrographs due to its resolution limitation. However, SEM images of selected nanowires showed that the diameter is in the range of 250 – 580 nm. The large variation of the nanowire diameter is attributed to both the irregularity of the AAO template and also due to the subsequent treatment in NaOH that is used to dissolve the membrane. The NaOH promotes the formation of oxide, as expected from Pourbaix [53], and with the formation of an oxide the diameter size increases owing to the larger specific volume compared to the metallic counterparts.

2.2. Substrate preparation for electromagnetic and PEA

A schematic of the process steps is shown in Figure 1. Blank Si substrates for the electromagnetic alignment were squares of 1.5 cm × 1.5 cm (Figure 1a). Since the reorientation of the nanowires were affected by parameters (e.g. stochastic surface and fluid flow forces) other than the deterministic magnetic field force, the nanowires easily deviated from the direction of magnetic field line after the evaporation of the solvent. Groove-patterned substrates for the PEA were fabricated via nanoimprint lithography into a double layer of 1 μm dextran and 2.5 μm polyurethane acrylate (PUA) coated on the Si substrates. The width of the grooves was 1.4 μm and their period was 2.71 μm (Figure 1b). The patterned area was a 3 mm × 3 mm square in the center of the substrate in order to enhance alignment of the nanowires.

Figure 1.

Schematic steps for (a) electromagnetic alignment (nanowires deposited onto blank silicon substrates) and (b) PEA on a microscale grooved surface formed by UV nanoimprint lithography into a double resist layer of dextran and PUA.

2.3. Electromagnetic and PEA

The electromagnet used in this study and aligned metal particles on a blank white paper on the electromagnet are shown in Figure S2 of the Supporting Information. The electromagnetic alignment and PEA of magnetic nanowires were performed in separate experiments to determine the accuracy of the alignment versus applied magnetic field strength, using both the blank substrate for the electromagnetic alignment (Figure. 1a) and the patterned substrate for the PEA (Figure. 1b), with identical conditions. Four samples were prepared and analyzed for both electromagnetic alignment and PEA experiments at each applied magnetic field strength.

An aliquot of 10 μL of the same nanowire solution was used for each experiment. The nanowire solution was dispensed on the substrate while the substrate was placed on the electromagnet. The magnetic field was then induced by the electromagnet which was connected to a (BK 1735 DC) power supply. The current was constant during each experiment.

2.4. Inspection and Characterization

A 3 mm × 3 mm square in the middle for each sample was inspected by optical microscopy and scanning electron microscopy (SEM). This area corresponds to the patterned area for the PEA samples. The deviation angle (θ_D) and the length of the nanowires (l) were measured manually from the optical micrographs. The deviation angle was defined as the angle between the nanowire direction and the direction of the magnetic field lines (or the direction of the grooves). The lengths and angles of nanowires were measured with a computer desktop ruler using optical micrographs taken with a 40X objective. Prior to using this method, the lengths of three nanowires determined using scanning electron micrographs were compared with those measured by optical micrographs. The lengths determined by optical micrographs were consistently larger by 0.7 ± 0.2 μm. Thus, overall measurement error in the determination of individual nanowires is less than 1 μm. For the nanowire deviation angles, the measurement error was estimated to be ~ 2°.

3. Results

3.1. Electromagnetic alignment and PEA

Figure 2(a) and (b) presents SEM images of the nanowires in group and individually, respectively, dispersed on a blank Si substrate. Not only is there a distribution of diameter size but also length. In the analysis of alignment accuracy, the nanowire lengths were grouped with the following ranges for better observation of the general trend: 0–10 μm, 11–20 μm, and 21–30 μm (i.e. 0 < l ≤ 10 μm, 10 < l ≤ 20 μm, and 20 < l ≤ 30 μm, respectively).

Figure 2.

SEM images of FeNiCo nanowires (a) in group and (b) individually on blank Si substrate after they are released from the AAO membrane.

Figures 3 and 4 show micrographs of nanowire alignment after the ethanol from the nanowire solution have evaporated. In Figure 3, only the electromagnetic alignment was used and in Figure 4 the combined PEA method was employed. The applied magnetic field in both cases was 70 Oe. The image in the center of each figure corresponds to an optical micrograph showing an overview of the substrate. The nanowires in the optical micrograph for the electromagnetic alignment (Figure 3), are circled with red lines as a guide. The large spot on the sample in Figure 3 corresponds to impurities remaining after evaporation of solvent. In Figure 4, the nanowires aligned by PEA method appear longer due to head-to-tail connections to each other. Such head-to-tail alignment of nanowires have been observed when field forces were used [28, 32]. In the PEA case, this behavior may be related to the modulation of the magnetic field strength at the surface, leading to an increase in the probability for more than one nanowires to encounter each other within a single groove. The SEM images in Figure 3 and 4 show examples of individual nanowires found on the substrate. The direction of the magnetic field lines is parallel to the y direction for both alignment methods. The orientation of nanowires was determined using the optical micrographs. Overall, most nanowires for both alignment methods show preferential directions along the magnetic field lines. However, for the electromagnetic alignment one can still observe nanowires with orientations far away from the magnetic field lines (marked by the yellow numbers 5–8). On the other hand, most nanowires are aligned along both the field lines and the groove direction for the PEA.

Figure 3.

An optical micrograph (center) and SEM micrographs (right and left) of FeNiCo nanowires using electromagnetic alignment on Si substrates at a magnetic field strength of 70 Oe. The numbers in the SEM image corresponds to the very same nanowire marked (in red, (1–4) in the optical micrograph. The nanowires marked with the yellow numbers (5–8) were oriented far away from the magnetic field lines. The applied magnetic field lines are parallel to the y direction in the optical micrograph. Individual nanowires in the optical micrograph are circled for clarity.

Figure 4.

An optical micrograph (center) and SEM micrographs (right and left) of FeNiCo nanowires using PEA on the microgrooves formed in a double resist layer of dextran and PUA on Si substrates at a magnetic field strength of 70 Oe. The numbers in the SEM image corresponds to the very same nanowire marked in red, (1–4) in the optical micrograph. The applied magnetic field lines are parallel to the y direction in the optical micrograph.

3.2. Statistical Analysis

In order to quantitatively determine the accuracy of the alignment, statistics were taken by measuring the deviation angle (θ_D) from the direction of the magnetic field line for all the nanowires found in a square area of 3 mm × 3 mm in the middle of each substrate (which corresponds to the pattern area for the PEA). We plotted the normalized angular distribution diagrams of nanowires versus applied magnetic field strength from 0 to 90 Oe. For each applied magnetic field strength and nanowire length group, the number of nanowires within a certain deviation angle group was divided by the total number of nanowires found in the same sample area. The error bars were determined as one standard deviation of the normalized values of the four samples of each distribution. Statistics were taken for different groups of the nanowire length of 0–10 μm, 10–20 μm and 20–30 μm. Figures S3 and S4 in the Supporting Information show the entire results of the diagrams for different ranges of nanowire lengths and different magnetic field strengths. As representative results, Figure. 5(a) and (b) shows the statistical diagrams for the nanowire length of 0–10 μm for the electromagnetic alignment and PEA methods, respectively. The results of the alignment on samples without an applied magnetic field (i.e, 0 Oe), but with patterned groove alignment was compared to the PEA results. For quantitative analysis, the deviation of aligned nanowires from the direction of the magnetic field lines as well as the full width at half maximum (FWHM) value from each fitted Gaussian curve was determined and used. Also, the percentage of the total number of nanowires with a deviation angle less than 5° from the direction of the field lines was calculated.

Figure 5.

Normalized number of nanowires for the nanowire length group of 0–10 μm as a function of the deviation angle for different magnetic field strengths for (a) electromagnetic alignment and (b) PEA methods. For each angular distribution diagram, the corresponding fitted Gaussian curve is included.

The summary of these two values and their corresponding standard deviations as well as the total number of individual nanowires found for a certain group of nanowire length are presented in Table 1 and 2. Overall, the FWHM value of the fitted Gaussian curve decreases as the magnetic field strength increases for both alignment methods, which indicates that more nanowires are aligned close to the direction of the field line with the magnetic field strength. The FWHM value decreases consistently from 180° without magnetic field (randomly oriented) to 15 ± 1° at 90 Oe for the PEA. For the electromagnetic alignment, on the other hand, the FWHM value decreases only to 32 ± 11° at the same magnetic field, which is more than two times greater than that for the PEA. Also, the percentage of the nanowires with less than 5° deviation angles for the PEA is more than four time higher than the electromagnetic alignment at all the applied magnetic field strengths except for 0 Oe.

Table 1.

The full width at half maximum (FWHM) values and the standard deviations obtained by the Gaussian curve fitting from the angular distribution diagrams for different magnetic field strengths and nanowire length groups for both electromagnetic alignment and PEA. Also included in [ ] are the total number of nanowires for each group of magnetic field strength and nanowire length.

FWHM	Magnetic Field Strength (Oe)	0	10	30	50	70	90
	Nanowires Length (μm)
Electro-magnetic alignment	0–10	180 ± 95 [24]	180 ± 94 [12]	63 ± 19 [26]	54 ± 19 [18]	63 ± 16 [35]	32 ± 11 [23]
	10–20	180 ± 94 [22]	93 ± 51 [15]	60 ± 16 [52]	38 ± 8 [40]	24 ± 4 [42]	44 ± 9 [41]
	20–30	180 ± 95 [76]	139 ± 68 [86]	32 ± 5 [123]	39 ± 7 [95]	25 ± 4 [91]	24 ± 4 [94]
PEA	0–10	180 ± 96 [16]	63 ± 42 [12]	17 ± 4 [44]	15 ± 3 [32]	15 ± 2 [78]	15 ± 1 [56]
	10–20	180 ± 93 [61]	67 ± 19 [68]	22 ± 4 [76]	17 ± 2 [71]	17 ± 2 [77]	17 ± 2 [76]
	20–30	180 ± 94 [105]	96 ± 48 [98]	28 ± 7 [90]	18 ± 5 [99]	17 ± 2 [146]	17 ± 3 [144]

Table 2.

The percentages of the number of nanowires with the deviation angles less than 5° and their standard deviations obtained from the angular distribution diagrams for different magnetic field strengths and nanowire length groups for both electromagnetic alignment and PEA.

Percentage for │θD│< 5°	Magnetic Field Strength (Oe)	0	10	30	50	70	90
	Nanowires Length (μm)
Electro-magnetic alignment	0–10	0 ± 0	5 ± 10	12 ± 16	7 ± 8	8 ± 16	7 ± 13
	10–20	3 ± 6	17 ± 33	7 ± 6	20 ± 7	27 ± 14	13 ± 13
	20–30	1 ± 3	9 ± 5	16 ± 12	19 ± 13	30 ± 14	39 ± 22
PEA	0–10	0 ± 0	11 ± 13	50 ± 64	41 ± 30	63 ± 20	58 ± 27
	10–20	7 ± 9	20 ± 23	31 ± 27	35 ± 26	49 ± 13	52 ± 17
	20–30	3 ± 3	19 ± 15	23 ± 14	27 ± 9	40 ± 13	41 ± 8

Figures 6(a) and (b) show the angular distribution diagrams of the nanowires for the three length groups of 0–10 μm, 10–20 μm and 20–30 μm at a magnetic field strength of 90 Oe for the electromagnetic alignment and PEA, respectively. For PEA, the FWHM value slightly decreases from 17 ± 3° for long nanowires of the 20–30 μm length group to 15 ± 1° for short nanowires of the 0–10 μm length group by ~ 10%. However, this value fluctuates between 24 ± 4° and 44 ± 9° for the nanowires with different lengths for electromagnetic alignment. For each group of magnetic field strength and nanowire length, the percentage of nanowires with less than 5° deviation angle is higher in the PEA (above 40% for all of the length ranges) than that in the electromagnetic alignment.

Figure 6.

Normalized number of nanowires at a magnetic field strength 90 Oe as a function of deviation angle for different nanowire length groups for (a) electromagnetic alignment and (b) PEA methods. For each angular distribution diagram, the corresponding fitted Gaussian curve is included.

Figure 7(a) and (b) show 3-dimentional (3-D) graphs of the FWHM values for applied magnetic field strengths (0–90 Oe) and nanowire length groups for the electromagnetic alignment and PEA, respectively. The 3-D graph clearly shows the general trend of increasing the alignment accuracy (or decreasing the FWHM values) with increasing the magnetic field strength and with decreasing the length of nanowires. Another important to note is that more systematic trends are seen for the PEA compared to those for the electromagnetic alignment. This can be attributed to the fact that, while the process for nanowires to approach the flat substrate surface and the re-orientation during solvent evaporation is stochastic, the surface groove patterns in the PEA add deterministic characteristics during both nanowire deposition and solvent evaporation processes. This leads to the decrease in the FWHM values for the PEA.

Figure 7.

3-D graphs for FWHM values versus applied magnetic field strength and nanowires length group for the (a) electromagnetic alignment and (b) PEA.

4. Discussion

Overall, the alignment accuracy of shorter nanowires was higher than that for longer nanowires for the PEA. One of the main mechanism of this trend is the dynamics of the shorter nanowires suspended inside the solution under an applied magnetic field. The positioning of magnetic nanowires on a substrate surface is a two-step process: in the first step nanowires in suspension rotate towards the direction of the magnetic field line, which is followed by the second step of approaching to the surface via surface-nanowire interaction. The suspended and/or deposited nanowires may experience re-orientation during capillary force of fluid flow and evaporation of the solvent. Once the nanowire adheres (sticks) to the substrate by the surface forces between the nanowire and substrate, the electromagnetic field was unable to rotate the nanowire any more.

Ignoring the nanowire-surface interaction, the rotational velocity of nanowires in the first step can be modeled by balancing the magnetic and hydrodynamic forces applied to a nanowire [54, 55]. The results show that the rotational velocity of shorter nanowires is higher than that of the longer nanowires [54, 55]. Thus, at all applied field strengths shorter nanowires require a shorter rotation time than the longer nanowires to align with the direction of the magnetic field line before approaching and attached to the bottom substrate. This mechanism results in a better alignment accuracy for shorter nanowires than longer ones. Also for both longer and shorter nanowires, the efficiency of the PEA is higher than the electromagnetic alignment at the high (90 Oe) applied magnetic field strengths.

In the second step of surface-nanowire interaction, the forces exerted by a surface on a depositing nanowire broadly consist of adhesion and friction, both of which depend on the material properties such as surface energies and Young’s moduli, and the interfacial properties such as the friction coefficient and the contract area [56]. The presence of surface grooves changes the interfacial properties and makes adhesion and friction anisotropic. When a nanowire is deposited into the trench of a microgroove, the effective contact area may increase depending on the location of the nanowire in the trench, leading to an increase in both adhesion and friction. In addition, the geometric interlocking significantly increases the friction in the direction perpendicular to the groove, which increases pinning of the nanowire against any shear force such as evaporation of the solvent [15]. This effect becomes more significant as the depth of the groove increases up to the diameter of the nanowire. On the other hand, when a nanowire is deposited on the protrusion of a microgroove, the effective contact area may decrease, leading to a decrease in adhesion and friction.

It is worthwhile to mention the effect of the microchannel depth on the alignment accuracy. For a given magnetic field strength, there would be a critical depth of the microchannel. Above this critical depth, the pinning of nanowires in the microchannel trench is so large that nanowires once deposited in the trench of the microchannel cannot escape upon external force such as capillary force and evaporation of solvent. In this regime, the alignment accuracy will not change with the depth of the microchannel. On the other hand, when the depth is below the critical depth, nanowires in the microchannel trench can escape upon external force, resulting in a decrease in the alignment accuracy. In this regime, the alignment accuracy may depend on the depth of microchannels. The depth of the microchannel used to align nanowires should be determined in consideration of the design of the final device as well as compatibility of the subsequent manufacturing process steps.

The effect of the viscosity of solvent would be negligible in PEA. The main mechanism to align nanowires in this study is the magnetic field with the support of surface patterns, which help pin the nanowires along the groove direction after the nanowires approach the surface. Increasing the viscosity of the nanowire solution will result in an increase in the nanowire rotation time by the electromagnetic field because of the elevated resistive hydraulic force, especially for longer nanowires. This will adversely affect both electromagnetic alignment and PEA.

Using different solvents will modify (1) the time for nanowires to rotate upon application of electromagnetic field, (2) the time for nanowires to reach the substrate surface, (3) the time for solvent evaporation, and (4) the magnitude of shear force applied during evaporation of solvent. However, the average nanowire rotation time by the electromagnetic field used in this study is in the order of tens of milliseconds (based on the Figure S5), significantly shorter than the time for nanowires to reach the surface and the time for solvent evaporation which are in the order of minutes. Thus, the effect of solvent on the nanowire rotation is negligible. The different shear force applied by different solvents upon evaporation may affect reorientation of nanowires, which will vary the alignment accuracy depending on the depth of the grooves. It should also be noted that the main reason to select ethanol as the solvent for these nanowires was to prevent any further oxidation before the nanowires final assembly.

The results indicate that the PEA is a highly powerful technique which systematically aligns individual nanowires of various lengths with higher accuracy compared to only electromagnetic alignment or only groove pattern alignment. The PEA technique can be used to place not only nanowires, but also nanotubes or other magnetic particulates. The less than perfect 100 % alignment accuary presented here demonstrates the challenge in alignment when the nanomaterial is not homogenous. Nanowires with more uniform feature sizes would be expected have higher alignment accuracy and less alignment distribution and thus larger FWHM values. This technique can also be applied to curved substrates in combination with the technology to produce micro/nanoscale patterns on curved substrate for flexible electronics. The PEA technique can also be used in diverse applications such as large scale parallel nanowires for optoelectronics on glass [57] and silicon [58, 59] substrates, and high performance transistors on silicon based substrates [60, 61].

5. Conclusion

A new alignment method (PEA) is presented that combines electromagnetic alignment with microscale surface groove patterns to realize large scale, high precision alignment of magnetic nanowires. Compared to only an electromagnetic alignment, or only a grooved pattern alignment, the PEA of FeNiCo nanowires demonstrated a systematic enhancement of the alignment accuracy, the percentage of the nanowires having less than a 5° deviation angle from the direction of the magnetic field lines for the PEA remained above 40% at the highest applied magnetic field. For both the PEA and electromagnetic only alignment methods, the alignment accuracy increased with increasing the magnetic field strength and decreasing the nanowire length. The PEA that is developed in this work has the potential to be impletemented in processed to build sensing elements with pre-grown magnetic nanowires.

Highlights.

• High precision nanowire alignment technique combining electromagnetic alignment with microscale grooves is demonstrated, which is named patterned electromagnetic alignment (PEA).

• The patterned grooves induce anisotropic interfacial properties and help pin the nanowires on the surface along the magnetic field.

• The alignment accuracy of PEA increases with the magnetic field strength and with decreasing the length of nanowires.

• The PEA technique is applicable to place not only nanowires, but also nanotubes or other magnetic particulates.