Browse Prior Art Database

PLANARIZED MR HEAD

IP.com Disclosure Number: IPCOM000013051D
Original Publication Date: 2000-Feb-01
Included in the Prior Art Database: 2003-Jun-12

Publishing Venue

IBM

Abstract

In the present day thin film magnetic head fabrication there are several processing issues to be considered. The most critical of the process parameters is the dimension of the second ferromagnetic pole tip which defines the size of the written track. That size is the one that controls the "Track per Inch" or track recording density. The pole tip fabrication deals with a lithographic step where the size of the final pole tip is obtained in a thick layer of photoresist . It is extremely important that all the key parameters controlling the photo exposure of said resist layer are controlled and maintained nominal through wafer to wafer processing. Besides the classical photo parameters of exposure dose, focus, and the development cycle the most important parameter left is resist coating uniformity and mean thickness. The thin film heads are made with structures such as Cu coils and resist insulators defining the back edge of the pole region. The back edge slanted profile may be of a thickness between 5 to 10 um at the top of the insulation stack for single layer coils. The entire insulation stack (with the coils within) also is fabricated atop two other thick structures. One is the First Shield of the read head (S1) generally of a thickness of 2 to 4um. The second one (which is placed just under the insulation stack) is the Second Shield (S2) and first ferromagnetic layer (P1) of a merged write head. This last one generally of a plated permalloy of a thickness between 3 to 4um. The combined step between the first and second shields (P1) could be as thick as 6um or more. If we combine the step problem just described with the insulation stack behind the pole tip region we obtain a very dramatic source for resist coating non-uniformities. The resist coating variation does result in pole tip size variation or impossibility of fabrication altogether. We can try methods of stack height reduction by altering the design of the inductive head such as was done in the Low Profile and/or Sunken Designs. In both cases the emphasis was to lower the sensitivity of the insulation stack to produce uneven resist coatings. Still a very large source of coating non-uniformity resides with the S1/S2 processing step. By modifying the photoresist mask used to insulate the coils (under and on top) we can simultaneously add the same photoresist layer at the coils region as well as surrounding the step part of the S1 and S2 shields. After photoresist insulation hard bake cycle (240 Deg. C for 7 hours.) the resist covering the edges of the steps (1 um overlap), and extending from 10 to 30 um into the lower regions of the alumina, produces a sloping surface that merges, in a very gradual way, the two dissimilar planes. Because of the sloping profile the photoresist, as spun in order to obtain its final thickness, the photoresist can freely flow from the lower surface to the higher surface were the pole tips are going to be fabricated. The illustration shows the actual sloping profiles with the cross section of the steps where one can observe the planarizing effect of the peripheral resist coating. Figure 1 is an example of the steps versus focal plane variation at the moment of pole exposure. In Figure 2 a cross section of an inductive head is exemplified where the steps were planarized. From methodical evaluation of product wafers, produced with and without planarization, we found that the biggest improvement is obtained when full planarization is used (both S1 and S2/P1). The pole tip resist thickness was measured by cross sectioning using Focus Ion Beam at the Zero Throat Region of the insulation stack (beginning of the insulation) and using SEM metrology. The resist thickness Sigma variation is presented in Table 1