semi合集-English.pdf - 第7354页

SEMI MF1391-0704 © 2004 6 6 10.6 For measurements at r oom temperature, locate the wavenum ber correspondi ng to the maxim u m absorbance in the region from 603 to 607 cm − 1 . For measurements at cryogenic te mperatures…

Figure 1

Carbon-Only Absorbance Spectrum

10.4.1 For double-beam instruments, position the

sample specimen in the sample beam and the reference

specimen in the reference beam and collect and plot the

carbon-only absorbance spectrum over the range of at

least 700 to 500 cm

−1

. Use a resolution of 2 cm

−1

better, at 605 cm

−1

, at room temperature, and a

resolution of 1 cm

−1

or better, at 607.5 cm

−1

, at cryo-

genic temperatures (below 80 K). Take spectra with the

sample chamber in a vacuum, dry nitrogen, or dry air.

Position the specimen so that the infrared beam passes

through the center of the specimens.

10.4.2 For single-beam (computer-assisted)

instruments, first collect a background spectrum with

the beam empty, then collect sample and reference

spectra sequentially under the same conditions as given

in Section 10.4.5. Ratio the sample and reference

spectra to the background and convert to absorbance.

Multiply the reference absorbance spectrum by the ratio

of the test specimen thickness to the reference specimen

thickness and subtract this normalized absorbance from

the test specimen absorbance to obtain the carbon-only

spectrum.

NOTE 3: If the user has interactive arithmetic capability,

subtract the reference absorbance spectrum from the sample

absorbance spectrum in a ratio that provides the best flatness

of the non-peak baseline between 560 and 640 cm

−1

. This

eliminates the 2-phonon band due to silicon and gives a

carbon-only spectrum.

10.5 With regard to the plot of the carbon-only

absorbance spectrum over the range from 500 to 700

−1

, define the baseline by drawing a straight line

from 560 to 640 cm

−1

. Use the average in the regions

from 550 to 570 cm

−1

, and 630 to 650 cm

−1

, to define

the endpoints of the straight line.

10.6 For measurements at room temperature, locate the

wavenumber corresponding to the maximum

absorbance in the region from 603 to 607 cm

−1

. For

measurements at cryogenic temperatures (below 80 K),

locate the wavenumber corresponding to the maximum

absorbance in the region from 605 to 609 cm

−1

10.6.1 Record the value of that wavenumber, to four

significant figures, as W

. Record the maximum

absorbance as A

, the absorbance of the absorption

peak.

10.6.2 Record the baseline absorbances, A

, as the

value of the baseline defined in 10.5 at W

10.6.3 Record both A

and A

to three significant

figures.

10.7 Determine and record the full width at half

maximum (FWHM) of the peak. If the FWHM of the

peak is greater than 6 cm

−1

(room temperature) or 3

−1

(cryogenic temperatures), the spectrum was not

properly obtained, and the instrument needs further

adjustment, or the specimen is under stress. See

Section 3.4 for a list of factors that influence FWHM.

Adjust the instrument in order to achieve the proper

FWHM and repeat the procedure.

10.8 Record the apodization function and the number of

zero-fills used (FT-IR instruments) or the spectral slit

width for dispersive infrared (DIR) instruments.

11 Calculation

11.1 Calculate the absorption coefficient, α, using the

expression:

)(

03.23

−=α (1)

where:

X = test specimen thickness, mm,

= the peak value of the carbon-only absorbance

spectrum, and

= the baseline value of the carbon-only absorbance

spectrum.

NOTE 4: This equation does not include a correction factor

for multiple reflections. Such a correction factor is not neces-

sary, due to the very strong lattice absorption at 610 cm

−1

11.2 For measurements at room temperature, calculate

the substitutional carbon content in atoms/cm

or parts

per million atomic (ppma) as follows:

6 Inoue, N., Stso. T. Nozaki, T., Endo, K., and Mizauma, K., “High

Reliability Infrared Measurement of Oxygen and Carbon in Silicon,”

in Emerging Semiconductor Technology, ASTM STP 960, D. C.

Gupta and P. H. Langer, eds., (ASTM, Philadelphia, 1987), pp. 365–

377.

ppma64.1

atoms/cm102.8ContentCarbon

316

α=

α×=

(2)

11.3 For measurements at cryogenic temperature,

calculate the content of substitutional carbon in

atoms/cm

or ppma as follows:

ppma74.0

atoms/cm107.3ContentCarbon

316

α=

α×=

(3)

NOTE 5: The calibration factor used in this test method was

determined as result of a study carried out in Japan by the

Silicon Technologies Committee of the Japanese Electronic

Industries Development Association (JEIDA), now JEITA,

the mid-1980’s. Further data analysis to eliminate recognized

systematic error was carried out by the ASTM subcommittee

assigned for this effort. The factor resulting from this study

has an uncertainty of ± 0.4 × 10

atoms/cm

or ±0.08

ppma·cm for the room temperature measurements. The

calibration factor for cryogenic measurements was calculated

using the ratio given in Kolbesen and Mladcnović.

The

uncertainty of the cryogenic temperature factor is therefore

± 0.2 × 10

atoms/cm

or ± 0.04 ppma·cm.

12 Report

12.1 Report the following information:

12.1.1 The instrument used, the operator, and the date

of the measurements,

12.1.2 Identification of test and reference specimens,

12.1.3 For room temperature measurements, the

nominal temperature of the sample chamber. For

cryogenic measurements, the nominal temperature of

the sample holder,

12.1.4 Thickness of test and reference specimen,

12.1.5 Location and size of the illuminated area on the

specimen,

12.1.6 For FT-IR instruments: (a) the apodization

function used, (b) the amount of zero-filling, and (c) the

number of data/cm

−1

. For dispersive instruments: (a)

the scan time/cm

−1

, (b) the spectral slit width, and (c)

the resolution at 600 cm

−1

12.1.7 Spectral full width at half maximum of the

absorption peak,

12.1.8 W

, wavenumber in cm

−1

of the absorption peak,

12.1.9 The absorption coefficient due to substitutional

carbon, α, in cm

−1

12.1.10 Carbon concentration, in ppma or in

atoms/cm

, and

12.1.11 The conversion coefficient used.

7 Kolbesen, B. O., and Mladcnović, T., Kristall und Technik 15(1),

K1–K3 (1980).

13 Precision

13.1 A round robin was carried out by JEIDA (now

JEITA

). The results of this 21-laboratory study are

shown in Table 1. Figure 2 shows that the sample

standard deviation increases with concentration value.

13.2 A single-laboratory investigation of the precision

of carbon measurement at cryogenic temperature with a

Fourier transform spectrophotometer was conducted

using a single sample analyzed once a day over a 30-

day period (30 measurement values). From this single-

laboratory, multiple-operator, multiple-day, single-

sample investigation of precision of carbon

measurement, the precision was found to be 0.02 ppma

(2 standard deviations). Nominal carbon content of the

sample was 0.16 ppma.

14 Bias

14.1 The bias of this test method cannot be evaluated

because there are no available reference standards

suitable for evaluating bias.

15 Keywords

15.1 carbon; infrared absorption; infrared spectroscopy;

silicon; single crystal silicon

Table 1 JEIDA Round Robin Results

Sample

Number

Number of

Measurements

Mean Value of

Concentration o

Substitutional

Carbon

Sample Standard

Deviation

1 21 2.79 0.204

2 21 3.89 0.342

3 21 0.23 0.059

4 21 0.35 0.050

5 21 0.74 0.100

6 21 0.93 0.081

7 21 1.03 0.079

8 21 1.25 0.168

9 21 0.41 0.083

10 21 0.84 0.123

11 20 0.80 0.081

12 20 0.94 0.135

13 20 0.99 0.095

14 20 0.30 0.097

15 20 0.46 0.084

16 20 0.71 0.070

17 20 1.07 0.108

18 20 1.46 0.145

19 20 3.41 0.276

20 20 0.31 0.051

21 20 0.45 0.073

22 20 0.51 0.099

23 20 0.74 0.116

24 20 0.16 0.055

25 20 0.62 0.100

26 20 0.87 0.131