banner image
No data available.
Please log in to see this content.
You have no subscription access to this content.
No metrics data to plot.
The attempt to load metrics for this article has failed.
The attempt to plot a graph for these metrics has failed.
Integrated on-chip inductors using magnetic material (invited)
Rent this article for
View: Figures


Image of FIG. 1.
FIG. 1.

Optical microscope images of copper/polyimide based inductors fabricated in a CMOS process. Top: stripe with magnetic material. Middle: stripe with slotted magnetic material. Bottom: four-turn elongated spiral inductor with magnetic material.

Image of FIG. 2.
FIG. 2.

Cross-sectional SEM images of thick copper wires surrounded by CoZrTa magnetic material forming an inductor that is integrated on top of a seven-metal level CMOS process. Two levels of CoZrTa magnetic material were deposited around the thick inductor wires using magnetic vias to complete the magnetic circuit.

Image of FIG. 3.
FIG. 3.

magnetic hysteresis loops of the hard and easy axes of magnetization for Co–4.5%Ta–4.0%Zr (at. %). The slope corresponds to the permeability. The Inset shows that the coercivity is only resulting in minimal hysteretic losses when used to build inductors. The magnetostriction coefficient of the CoZrTa alloy was 0.2 ppm strain.

Image of FIG. 4.
FIG. 4.

Permeability vs applied magnetic field of amorphous CoZrTa as a function of film thickness. A drop in permeability with increasing film thickness occurs because of the growing demagnetization field caused by the increasing film thickness.

Image of FIG. 5.
FIG. 5.

Real and imaginary components of the complex permeability spectra vs film thickness. The imaginary component becomes large with increasing thickness. Dashed lines correspond to modeled spectra.

Image of FIG. 6.
FIG. 6.

The complex permeability spectra of laminated vs unlaminated thick CoZrTa. Eight laminations that were thick each were formed and eliminated the multiple resonant modes.

Image of FIG. 7.
FIG. 7.

Measurements of long transmission lines with different linewidths. The arrow show the comparison between transmission lines with two levels of CoZrTa vs without magnetic material. The magnetic alloy consists of four laminations that are thick.

Image of FIG. 8.
FIG. 8.

Measurements of inductors surrounded with magnetic material using different magnetic via widths.

Image of FIG. 9.
FIG. 9.

Measurements and simulations of the (a) sheet inductance and (b) shunt inductance for inductors with different via widths. Measurements show a decreasing shunt inductance with decreasing via width whereas the sheet inductance does not change.

Image of FIG. 10.
FIG. 10.

time constant (inductance vs ac resistance) of a family of different inductors using thick copper or thick aluminum metallization with two layers of CoZrTa except as indicated. The diagonal lines indicate the corresponding factors. Simulations of the copper inductors are shown as dashed lines.

Image of FIG. 11.
FIG. 11.

Inductance vs frequency of spiral inductors. The inductance is well over causing a roll-off from the resonant frequency of the inductor. The copper wires are wide. The sizes of the two-turn and six-turn spiral are and , respectively and the dc resistances are 0.86 and , respectively.

Image of FIG. 12.
FIG. 12.

Plots calculated using an analytical model of the quality factor. (a) factor vs film thickness. (b) factor vs the number of laminations for a magnetic film.

Image of FIG. 13.
FIG. 13.

Quality factor of long lines with two levels of thick CoZrTa using four laminations vs eight laminations. Increasing the laminations not only reduces the eddy current losses but also reduces the inductance causing a net shift in the peak quality factor.


Article metrics loading...


Full text loading...

This is a required field
Please enter a valid email address
752b84549af89a08dbdd7fdb8b9568b5 journal.articlezxybnytfddd
Scitation: Integrated on-chip inductors using magnetic material (invited)