May_EDFA_Digital

edfas.org

ELECTRONIC DEVICE FAILURE ANALYSIS | VOLUME 19 NO. 2

momenta between the lab frame and the annihilation

frame was measured by finding the opening angle of the

emitted photons. Detectors were placed several meters

to either side of the sample at angles varying from 180°.

This is called angular correlation of annihilation radiation

(ACAR). Increasingly shallower anglesmeasure an increas-

ing momentum difference between the zero momentum

frame and the lab frame transverse to the direction of

travel of the emitted photons. The major drawback of

ACAR is extremely low count rates; this is due to the

small solid angle of the detector, which is a long distance

from the sample being studied. For this reason, Doppler-

broadening spectroscopic techniques are now used.

One measures the momentum difference in-line with the

emitted photons. In this case, one photon will have more

energy and appear blue-shifted due to the electron’s origi-

nal momentum in that direction, and the photon moving

in the opposite directionwill be red-shifted symmetrically

about the 511 keV peak. The detector can be placed right

beside the sample, which increases the detection rate by

orders ofmagnitude. Using twodetectors, thephotons can

be observed in coincidence, and background detections

can be dramatically reduced.

A number of detectors are placed as physically close to

the target as possible. The configuration of these detec-

tors varies depending on their purpose. Typically, at least

one cryogenically cooled high-purity germanium (HPGe)

detector is used because of its high energy resolution, and

often it is paired with another HPGe, or with a NaI(Tl) or

BaF

detector. The HPGe detectors have an energy reso-

lution of approximately 2.4 eV at 511 keV, and they are

absolutely necessary for Doppler-broadening spectrosco-

pies. The NaI(Tl) and BaF

detectors have a lower energy

resolution but a higher efficiency, and theyworkwell with

time-of-flight measurements or as coincidence counters.

CONCLUSION

Positron spectroscopies offer a level of precision for

defect detection and identification not available for any

other spectroscopic technique. Unfortunately, traditional

positron beams have a lower intensity than is viable for

time-sensitive commercial uses. To make the positron

beam a more useful tool for the semiconductor industry,

theDFPGhas beendeveloped, which increases the intensi-

ty of a traditional beam through a number of approaches.

An intense source of positrons for a beam can be

created by using a combination of a larger source, pre-

cisely etched moderator meshes, and an electrostatic

lensing system. It is designed to image moderated posi-

trons emitted from a thin, ring-shaped source. The size

and shape of the source allows for a larger amount of

radioactive material and therefore more positrons while

minimizing self-absorption and thermal breakdown. The

positrons emitted by the radioactive isotopes are focused

by the DFPG into a small spot of only a few millimeters.

REFERENCES

1. A. Ore and J.L. Powell:

Phys. Rev.,

1949,

75,

pp. 1696-99.

2. S. Berko and J. Mader:

Appl. Phys.,

1975,

pp. 287-306.

3. B.Y. Tong:

Phys. Rev. B,

1971,

pp. 1436-39.

4. R.M. Nieminen and J. Olivia:

Phys. Rev. B,

1980,

22,

pp. 2226-47.

5. C. Hugenschmidt et al.:

J. Phys.: Conf. Series,

2013,

443.

6. I.J. Rosenberg, A.H. Weiss, and K.F. Canter:

Phys. Rev. Lett.,

1980,

44,

pp. 1139-42.

7. B. Nielsen, K.G. Lynn, T.C. Leung, G.J. Van der Kolk, and L.J. Van

Ijzendoorn:

Appl. Phys. Lett.,

1990,

56,

pp. 728-30.

8. R. Krause-Rehberg and H.S. Leipner:

Positron Annihilation in Semi-

conductors,

Springer, 1999.

9. Z. von Borries:

Physik,

1942,

119,

p. 498.

10. P.J. Schultz and K.G. Lynn:

Rev. Mod. Phys.,

1988,

60,

pp. 701-81.

11. P. Coleman:

Positron Beams and Their Applications,

World Scientific,

2000.

12. A.P. Mills, Jr.:

Rev. Mod. Phys.,

1988,

60,

p. 704.

13. K.G. Lynn and H. Lutz:

Rev. Sci. Instrum.,

1980,

51,

p. 977.

14. S. Pendyala and J. McGowen:

J. Electron Spectrosc. Relat. Phenom.,

1980,

19,

p. 161.

15. A. Vahanem et al.:

Phys. Rev. B,

1987, p. 4506.

16. P.J. Schultz et al.:

Phys. Rev. B,

1983,

27,

p. 6626.

17. W. Anwand, J.M. Johnson, M. Butterling, W. Wagner, W. Skorupa, and

G. Brauer:

J. Phys.: Conf. Series,

2013,

443.

18. A.I. Williams et al.:

J. Appl. Phys.,

2015,

118,

p. 105302.

19. A.R. Koymen et al.:

Nucl. Instrum. Methods Phys. Res. A,

1999,

422,

pp. 479-83.

ABOUT THE AUTHORS

Manfred Fink

is a professor at the University of Texas at Austin in the Atomic, Molecular, and

Optical Physics Group. He began his career by studying electron diffraction and has expanded his

research to include novel uses of Raman spectroscopy to detect cancer, predict earthquakes, and

diagnose lactose intolerance in newborns; neutrino physics; and slow positron beam physics. Dr.

Fink has held a number of visiting professorships at universities in Germany and has been the chair

of the Texas chapter of the Humboldt Foundation.