Since its inception about three decades ago, magnetic force microscopy (MFM) has become a primary imaging technique to analyze submicrometer domains [Saenz 1987, Proksch 1995]. The magnetic data storage industry utilizes MFM extensively to image magnetic recording media and devices [Rice 1994, Svedberg 2002]. Accordingly, MFM analysis is instrumental in facilitating the understanding of magnetic recording phenomena. While conventional MFM probes have been particularly influential in the analysis of magnetically recorded bits, their use in the analysis of fully energized modern magnetic transducers has been limited because of their relatively low coercivity. The magnetic field generated by modern magnetic transducers can reach more than 10 kOe at a distance less than a few nanometers above the write-pole tip. Consequently, the magnetization of the probe can change as it scans across an energized write-pole tip, making interpretation of the MFM images difficult.
While relatively high-coercivity probes have been demonstrated in the past, they have either not been tested on magnetic recording transducers or used only to evaluate transducers with dc currents up to 3 mA; meanwhile, modern magnetic recording transducers operate at write currents above 25 mA. Presently, commercially available high-coercivity MFM probes are marketed with coercivity values higher than 5 kOe and are composed of FePt or CoPt with a thickness of 45 nm, such as the Asylum Research Magnetic Force Microscopy-High Coercivity probes. Similarly, MFM probes with coercivity greater than 5 kOe have been fabricated on tip-less carbon nanocones, using FePt films [Chen 2009]. Liou and Yao  present their experimental work on the development of high-coercivity MFM probes, using CoPt composition, but do not characterize the probe's coercivity or provide data for dc write currents higher than ±3 mA. Liou  have used superparamagnetic MFM tips to characterize magnetic transducers, but only for dc write current of 3 mA. Others have used FePt-based compositions to enhance the coercivity and/or lateral resolution of MFM probes via unconventional techniques, such as focused ion beam milling of FePt-coated plateau probes [Amos 2009], FePt/FeCo/FePt exchange-spring trilayer tips [Rheem 2005], and multidomain MFM probes [Amos 2008]; Albeit, neither has demonstrated sufficient coercivity to analyze fully energized magnetic transducers. Hence, these probes may be limited in their usefulness for characterizing modern magnetic recording transducers.
The present letter surmounts this limitation by fabricating ultrahigh coercivity (UHC) MFM probes that can withstand the high magnetic fields (9–11 kOe) modern recording writers can generate [Ito 2002], thus enabling the acquisition of 2-D magnetic field gradient maps of fully energized magnetic transducers via MFM analysis. Furthermore, the magnetic moment of these probes has been tailored in order to significantly reduce the effect of their stray magnetic fields on the analyzed specimen. These newly developed MFM probes are also suitable for the analysis of magnetic thin films and structures under externally applied magnetic fields [Mohanty 2005] and may also be of great importance in advancing nano-MRI via magnetic resonance force microcopy [Rugar 2004, Degen 2009].
EXPERIMENT AND RESULTS
To attain coercivity larger than 10 kOe, the deposition conditions for FePt (55/45) L10 thin films are first optimized on silicon substrates. The composition and thickness of the adhesion (Ta), seed (CrRu and MgO), and magnetic layers along with the sputtering pressure (3–10 mTorr) and substrate temperature (500 °C–700 °C) of the magnetic layer are tuned for optimal results. Next, selected compositions are sputter deposited on several silicon probes in conjunction with silicon witness samples. In this letter, two types of newly developed MFM probes are presented: low moment (LM) with ∼2.3 × 10−13 emu and extremely LM (ELM) with ∼4.5 × 10−14 emu (80% less magnetic moment), both exhibiting coercivity larger than 11 kOe. The LM and ELM probes are composed of CrRu(15 nm)/MgO(6 nm)/FePt(25 nm) and CrRu(25 nm)/MgO(6 nm)/FePt(5 nm), respectively. The magnetic composition for the LM and ELM probes is deposited at 700 °C under 5 mTorr and at 500 °C under 10 mTorr, respectively. The incorporation of both CrRu and MgO to the compositions facilitates the L10 ordering of the FePt film. The L10 ordering was confirmed via X-ray diffraction analysis on the witness silicon samples. In order to estimate the magnetic moment of each type of MFM probe, the volume of the magnetic compositions is calculated based on the assumption of a semispherical active magnetic region and is then multiplied by an approximated composition magnetization (MS) of ∼ 750 emu · cm−3. The calculations are based on an initial radius of curvature (ROC) of 10 nm for the silicon probes. The volume integration starts from the seed layer ROC (R1) and ends on the magnetic layer ROC (R2), resulting in the following formulation: mtot∼ 2π/3(R23 − R13)MS, where mtot is the total calculated magnetic moment. The first selected composition has a coercivity of about 10.5 kOe for the out-of-plane direction, as measured by vibrating sample magnetometer on a witness silicon substrate.
To determine their coercivity, the probes are saturated in one direction and used to take a reference image of a demagnetized perpendicular bit patterned media (BPM). The probes are then removed from system, subjected to an applied magnetic field of 1 kOe in the reverse direction, and another MFM image of the BPM is taken for comparison with the reference. The process is repeated for increasing applied magnetic fields until the color coding of the MFM image inverts, an indication that the magnetization of the probes has switched. This technique is conducted to characterize both the newly developed and commercially available MFM probes from Nanosensors (NS), the point probe plus–LM–MFM–reflex coating (PPP-LM-MFMR) probes.
Atomic force microcopy (AFM) and MFM imaging is performed using a dimensions 3100 scanning probe microcopy system by Veeco Instruments. The tapping/lift-mode technique is applied for all MFM imaging. The phase shift in the mechanical response of the probe, due to probe–sample interaction, is monitored. The lift-start/lift-scan heights are set to 50/30 and 50/10 nm for the BPM and magnetic writer analysis, respectively. The resonance frequency of the cantilevers is around 70 kHz with a force constant of ∼2.8 N/m.
Fig. 1. On the left, coercivity field analysis of a conventional MFM probe by NS: (a) AFM image of a series of seven magnetic bits, taken with the NS MFM probe, (b) MFM images of the magnetic bits after applying a magnetic field of 25 kOe, (c) and −1 kOe along the probe's axis of symmetry. On the right, coercivity field analysis of the newly developed LM-UHC MFM probe: (d) AFM image of the same series of magnetic bits as shown on the left, taken with the LM-UHC MFM probe, (e) MFM images of the magnetic bits after applying a magnetic field of −10 kOe, (f) −11 kOe, (g) and −12 kOe along the probe's axis of symmetry.
View All | Next
A. Probes’ Coercivity Analysis
First, the coercivity of the NS MFM probes is investigated. One of the probes is magnetized along its axis of symmetry with a magnetic field of ∼25 kOe, followed by AFM/MFM imaging of the BPM [see Fig. 1(a) and (b)]. The MFM image shown in Fig. 1(c) is performed after a magnetic field of 1 kOe has been applied in the reverse direction. As is clearly evident from the MFM images, the magnetization of the commercial MFM probe is already reversed by a magnetic field of 1 kOe. Similar results are obtained for four other NS MFM probes. Next, the coercivity of the newly developed MFM probes is determined by the same technique. Since the coercivity in the out-of-plane direction of the FePt thin-film composition on silicon witness samples has been measured to be higher than 10.5 kOe, the incremental magnetic field analysis starts from a magnetic field of 10 kOe. To avoid redundancy, the MFM image taken after the probe was saturated with a magnetic field of 25 kOe is omitted. Fig. 1(e)– (g) reveals that the coercivity of the newly developed MFM probes lies between 11 and 12 kOe, at least an order of magnitude higher than the coercivity of the commercially available MFM probes. Similar results are obtained for four other probes. Hence, from here on these probes will be referred to as LM-UHC MFM probes. To summarize, the commercially available NS MFM probes are suited for analyzing magnetic structures that emanate less than 1 kOe of stray magnetic fields. On the other hand, the newly developed LM-UHC MFM probes can be used to investigate specimens that emanate stray magnetic fields higher than 11 kOe and are therefore suitable for characterizing fully energized modern perpendicular magnetic recording writers.
B. Modern Magnetic Recording Transducer Analysis
The commercial and LM-UHC MFM probes have been used to analyze a modern perpendicular magnetic writer by performing head saturation and remanence tests on a shielded write-pole tip [Kanai 2003, Xiao 2005]. The transducer exhibits both a front magnetic shield and a trapezoidal write-pole tip. The front shield contributes to a larger magnetic field gradient and a better magnetic field angle for sharper bit transitions and improved grain switching, respectively [Mallary 2002]. The trapezoidal shape of the write-pole tip eliminates the erasure of adjacent tracks when recording at relatively large skew angles, resulting in higher magnetic tracks density [Ito 2002, Kanai 2007].
The MFM probes by NS are initially used to perform a write-pole saturation test. The saturation test is most commonly performed using a spinstand system with which various signal parameters are analyzed as a function of applied write currents. For example, the write currents may be varied from 1 to 45 mA, while the average giant or tunneling magnetoresistance readback signal of a recorded track is measured for each write current step. A plot of the write current versus the average readback SNR is generated. In most cases, the write current at which the SNR plateaus is considered the saturation current of the magnetic transducer under investigation. As can be seen from Fig. 2, a write current of only 5 mA is already sufficient to generate magnetic stray fields greater than the coercivity of the NS probe, and therefore the magnetization of the probe switches right when the probe is scanning above the write-pole tip. For the 25 mA write current, an even more pronounced magnetization switching of the probe is evident when the probe reaches the region of the write-pole tip. MFM line scans along the write-pole tip down-track and cross-track directions further emphasize the magnetization switching of the probe [see Fig. 3(a) and (b)]. The probe is thus unsuitable for characterizing dc energized magnetic transducers.
Fig. 2. MFM analysis of a modern perpendicular magnetic transducer, using a commercially available MFM probe by NS. The hollow arrows point to the location of the write-pole tip.
Previous | View All | Next
Fig. 3. MFM line scans along the write-pole tip (a) down-track and (b) cross-track directions. Both micrographs reveal the switching of the probe's magnetization as the probe scans above the write-pole tip.
Previous | View All | Next
The newly developed LM-UHC MFM probes are then used to perform the same saturation test with dc write currents ranging from 0–45 mA. As can be seen from Fig. 4, the probe maintains its polarization of magnetization throughout the entire test. When the dc write current polarity is changed from 45 to −45 mA [see Fig. 4(h) and (i)], only the color coding of the MFM image is inverted; hence, reaffirming that the probe exhibits high anisotropy and high coercivity. In order to accurately determine the minimum dc write current that is required to saturate the write-pole tip, the MFM response is plotted for different dc write currents along two orthogonal lines: (a) cross-track is marked with a black-dotted line and (b) down-track is marked with a red-dotted line [see Fig. 4(h)]. For the down-track analysis, the phase difference between the minimum and maximum of each MFM line scan is measured and then compared [see Fig. 5(a)]. For the cross-track direction, the full-width–half-maximum (FWHM) is measured for each MFM line scan and then compared [see Fig. 5(b)]. The MFM line-scan investigation of both directions reveals that the maximum force interaction between the MFM probe, and hence the maximum magnetic field of the write-pole generates, occurs for dc write currents above 25 mA. Consequently, the MFM images for the dc write currents between 25 and 45 mA have been omitted from the presented figures and MFM line scans.
Fig. 4. MFM analysis of a modern perpendicular magnetic transducer, using the newly developed UHC-LM MFM probe. The sequence of MFM images has been taken to perform a write-pole saturation test by applying a range of dc write currents.
Previous | View All | Next
Fig. 5. MFM line scans of the write-pole tip taken along its (a) down-track direction and (b) cross-track direction. The magnetic gradient distance between the write-pole and front shield is ∼75 nm, and the FWHM of the magnetic field profile is ∼120 nm.
Previous | View All | Next
Two standard parameters in the characterization of modern magnetic recording transducers are the distance along the down-track direction (connecting the write-pole tip and the front shield) at which the magnetic recording field drops from a maximum to a minimum and the FWHM of the magnetic field profile along the cross-track direction at the recording side of the write-pole tip. The former relates to the sharpness of bit transitions and the latter to the track width. Minimization of both parameters is vital for higher density magnetic recording. The MFM line scans for the write current of 45 mA shows that for the particular head under investigation, the magnetic gradient distance along the down-track direction is ∼75 nm, and the FWHM of the magnetic field profile along the cross-track direction is ∼120 nm.
For proper magnetic recording, the stray magnetic fields emanating from a write-pole tip should diminish when the transducer is not excited [Kaoa 2005]. Otherwise, the writer can erase recorded information during the read-back process. Thus, writer remanence tests are of great importance in characterizing magnetic recording transducers. As evident from Fig. 4(a), a nonzero magnetic flux is detected above the write-pole tip when the transducer is not energized. Nonetheless, since an attractive force is also detected from the front shield (right side of the MFM image), one cannot conclude with certainty that the magnetic stray fields emanating from the MFM probe did not nucleate the magnetic domains in the write-pole tip. In order to confirm that the particular magnetic writer under investigation exhibits a nonzero remanence magnetization, ELM-UHC MFM probes are fabricated with ∼80% less magnetic moment than the LM-UHC probes.
The ELM FePt MFM probes have a coercivity of about 15 kOe for the out-of-plane direction, as measured by Kerr rotation on a witness silicon substrate. The same material composition has been sputter deposited on five silicon cantilevers. To examine the remanence magnetization of the writer-pole tip, the probes were used to obtain a set of four MFM images taken at specific dc write currents in the following order: −45, 0, 45, and 0 mA. Fig. 6 reveals that the magnetic writer exhibits a nonzero remanence magnetization and that the ELM-UHC probes are well tailored to minimize the probe-induced magnetic interaction. As a result, this particular writer cannot be used to store information on magnetic recording media for which the switching field distribution lies below the magnetic field generated under its remanence state.
Fig. 6. MFM analysis of a modern perpendicular magnetic transducer, using the newly developed UHC-ELM MFM probe. The sequence of MFM images has been taken to perform a write-pole remanence test by applying a range of dc write currents.
Previous | View All
This work was supported by the National Science Foundation/Nanomond under Contract IIP-0712445 and by the Department of Defense/Defense Microelectronics Activity under Contract H94003-09-2-0901.