Dilution Factor in Spectrochemical Analysis C.
B. POST, D. G. SCHOFFSTALL,
AND
GEORGE HURLEY
Metallurgical Department, The Carpenter Steel Company, Reading, Pa.
centration of alloying elements other than X can be spectrographically analyzed for X by the following steps: 1. Determining the reIative intensity ratio of the X and ironspectrum lines in the conventional manner 2. Multiplying the above intensity ratio of alloy and iron lines by the atomic dilution factor (loo - )‘ to get the corrected 100 intensity ratio X to Fe 3. Using the standardization curve determined from X - P, alloys to get the true % X in the alloy
The applicability of the atomic dilution factor in spectrochemical analysis has been studied, using two methods for exciting the spectra of ferrous alloys: a technique described b y Harler, Harvey, and Dietcrt (3) and chill-cast metal electrodes in a condensed spark circuit. The alternating current spark analytical procedure was found adequate for routine spectrochemical analysis. Dilution may be expressed in terms of atomic per cent dilution, permitting econ,omical and accurate analysis of a wide range of alloys without recourse to 4 series of standard samples or a number ol calibration curves for each alloy.
The calculation of the atomic percentages for the various elements in steel 22, Table 11, is given as follows:
H E variation of the iron spectrum intensity in ferrous alloys having different iron content,s has not been completely and satisfactorily discussed in spectrochemical analysis. The probable reason for lack of information on this subject is that commercial installations having the necessary precision are not called upon to handle the wide range of ferrous alloys which would make such a study necessary. At The Carpenter Steel Company Laboratories, a day’s run would include the analysis of such elements as manganese, silicon, copper, nickel, cobalt, molybdenum, chromium, vanadium, co~umbium,tantalum, a,ld in steels represented by the SBE, NE, stainless, high-speed, carbon and tool, and high-nickel (up to 80% nickel) grades of steels. Under these conditions it was necessary that a dilution factor be properly evaluated to correct for the decrease in intensity of the iron spectrum as the percentage of Table iron is decreased (internal-standard method of spectrochemical analysis) ; otherwise the number 1,:lement of standardization curves (concentration us. relative intensity of an alloy line to an iron line) would Copper be cumbersome, if not unworkable. lIangapese The following is the theoretical cdTcct of variaChromium tions in the iron content on the rolativc intensity Nickel of the iron spectrum.
If the sum
the atomic percentages Of constituents other t,han X is grouped together as Y, then (X) (Y) (Fe) = 100 (1) Of
+
+
c
IF*
2)
a standardization curve (% In using alloys where Y = 0, the second term on t6e left in Etuation 3 is not evaluated and so TX Fe(X) = K -
IF.
Concentration Range
Alloy Line
Internal Standard
‘;o
A.
A.
0.10-0.35
Cu 3273.96
F e 3265.62
0.23-1 50 8.00-27.00
>.In 2933.06
Cr
2879.3
F e 2920.69 Fe 2872.3
5.00-12.00
Xi
2992.59
0.13-0.70 0.10-0.70 0.12-0.80 0.50-1.50 1.15-2 .OO 0 06-1.20
Mn 2933.06 Ni 3414.7 Ab 2816.2 hlo 2644.4 Ni 3423,7 AI 3944.0
F e 2987.29 Fe Fe Fe Fe Fe Ni
2807.0 3428.2 2807.0 2644.0 3424.3 3973.56
Excitation D.C. arc, no l e ~ i h Sectored, 50% D.C. arc, no lelis D.C. a r c , n o lem Sectored, SOYo D.C. arc, no Ienu Sectored, 50% A.C. spark, no lens h.C. spark, no lem A.C. spark. n o lens A.C. spark, no lens A.C. spark, no lens A.C. spark, no lens
The applicability of the atomic dilution factor has been studied, using twomethods for exciting the spectra of ferrous alloys. The first makes use of a technique described by Hasler, Harvey, and Dietert ( 3 ) . hletal filings are placed on the platform of an especially designed carbon electrode and burned in the 220-volt direct current arc. I n this method the application of the atomic dilution factor is in accordance with the theoretical presentation. The second method uses chill-cast metal electrodes in a condensed spark circuit. The data of the high-voltage spark analysis show the application of the atomic dilution factor to be at variance with the method used in the direct current arc. For this reason the two methods are presented separately below. It is evident from the data that while the technique of applying the atomic dilution factor differs between the two methods, the required correction is numerically equal to the atomic dilution factor.
(3)
100
100 19
-
I. Wave Lengths of Spectrum Lines Used in A l l o y Analysis
(2)
(x)= K & (loo - )‘
IF^ 100
1.8175
-
the atomic weight of iron, the atomic dilution factor can be calculated with sufficient accuracy by subtracting the percentages by yeight from 100 to show the dilution of the base element-i.e., the dilution of the internal standard.
or IX
Atomic Percentage 0.43 0.53 0.83 19.25 8.25 70.90
The atomic dilution factor for this steel is 0.71 on the basis that the iron in the alloy is diluted atomically from 100% to 71.0% (1.00 to 0.71). Whore the atomic weights of the diluting elements approach
where (X) is the percentage concentration of the alloying element under consideration. Sow, assume that the relative intensities Of a pair of alloy and iron lines depend only on the atomic ProPortions Of the alloying element and iron; then
-+ K
Moles/ 100 grams 0,0079 0.0096 0.015 0.330 0.150 1,285
100 000
1Tanganese Nickel SIolybdenum Afoiybdenum Nickel Aluminum
(X) = K -Ix (Fe) IF^
So b y Weight/ Atomic Weight 0.095/12.0 0.53/54.9 0.42/28.06 18.23/52.0 8.89/58 7 71 8 4 / 5 5 . 8
Percentage by Weight 0.095 Mn 0.53 Si 0.42 Cr 18.23 Ni 8.89 Fe 71.835
T
(4)
APPARATUS AND METHOOS
expresses the ordinary standardization curve constructed by varying X in iron alloys of otherwise low alloy content. diluReferring now t o Equation 3, call (100 - y ) the 100 tion factor”, and we find that an alloy having an appreciable con-
SAMPLE PREPARATION. Cast electrodes are used for the condensed spark spectrum. The electrode diameter is 0.8 cm. ()/I( inch), and the electrodes are ground to a spherical dome on No. 120 emery paper. 412
ANALYTICAL EDITION
July, 1945
Table Steel
Type"
S O .
SAE 52100 C.S. Co. Solar SAE 4140 SAE 6120 C.S. Case-hardened thread gage 420-F 410 446 (C.S.S. No. 7) 303 321 4970 N i Steel 49% h'i Steel 49(& Xi Steel 30R Xi Steel 49% Ni Steel Invar Experimental S2YC y i Steel Experimental Experimental Experimentsi 302 307 308 403 Experimeiir a, 430 443 329 405 440 440 416 403 403 443 430 443 406 406
1 2 3 4 5
6 7 8 9 10
11 12 13 14
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49
347
50
51 52 53
;: 56 57 58 59 60 61
62
63 64 65 66 67 68
347 307 307 18-5Cr-Ni steel 347 347 347 303 303 Modified 306 304 307 308 2117
NE 8720 SAE 3140 414 Cr-\\- hot nurti steel NE 8740 Carpenter Hampdei. 18-4-1high speed Carpenter Stentor Carpenter Mn-Mo Hi-C-iIi-Cr Air-harii 721
II.
Typea and Chemical Analyses of Steels Used
C
.\in
Si
Cr
Si
c10
70
7'
c ,0
%
0.98 0.50 0.41 0.49 0.18 0.28 0.106 0.124 0 008 0.045 0.02 0.024 0.031 0.17 0.036 0.096 0.27 0.08 0.75 0.75 0.75 0.095 0 085 0.082 0.079 0:oio 0,080
0.057 0.07 1.03 1.11 0.10G
0.10 0.10B 0 123 0.100 0.130 0.096 0.112 0.063 0,065 0.086 0.154 0.102 0.048 0.057 0,0;77 0.076 0,084 0.186 0.073 0.061 0.152 0,085
0.37 0.38 0.16 0.55 0 39 2 0Y
0.74 0.90 0.G5 1.62 0.30 0 23 0.03
413
0.26 0.40 0.70 0.88 1.37 0.36 0.46 0.68 0.80 1.01 0.46 0.46 0.30 0.73 0.29 0.51 0.54 0.34 0.27 0.27 0.27 0.53 4.42 1.82 0.38 0.88 0.35 0.46 0.42 0.46 0.38 0.37 0.44 0.41
0.41 0.42 0.45 0.42 0.42 0.42
..
0.25 1.02 0.36 0.46 0.22 0.34 0.34 0.56 0.57 0.60 0.40 0.39 0.33 0.28 0.27 0.25 1.59 0.15 0.15 0.15 0.15 0.42
0.29 0:is 0'35 0.49 0.42 0.24 0.37 0.40 0.40 0.34 0.31 0.48 0.60 0.66 0.39 0.34
..
4:62 4.56 0.34 1.53
0:48 0.60
0:io
0164 0.52 1.10 ..
..
0.78 1.56 3:92 1.78 4.42 0.65 0.68 0.43 0.56 0.89 0.40 0.27 1.60 0.80 0.33 1 07 0.97 1.24
0.50 0,56 , .
0:27 0.29 0.10 0.28 0.44 0.68 0.33 0.44 0.21 0.22 0.28 0.32 2.79 0.27 n . 83
1.35 0.20
0.85 0.91 0.11 13.67 12 37 26.05 18 69 17.92 0.13 0.15 0.13 0.08 0.13 0.06 22.96 0.15 0.16 0.16 0.16 18.23 20.11 20.60 12.40 8.39 16.80 20.10 26.40 14.43 17.00 16.70 13.23 12.33 12.13 21.80
16.50 21.45 13.57 13.40 18.49 17.29 19.50 20.27 17.58 18.70 17.38 18.33 18.65 18.40 21.00 18.23 20.68 21.70 20.11 0.90 0.82 12.37 7.56 0.48 12.10 4.15 0.19 0.17 11.73 7.46 20,oo 14.00
0.18 0.10
...
0.06 0.05 0.15 0.46 0.12 7.70 10.94 49.00 48 5 5 48 81 29.74 49.03 35.59 12.09 32.56 0 02
0.02 0.02 8 89 9.82 10.20 0.60 . I .
0.15 0.18 4.80 0.18 0.21 0.48 0.13 0.40 0.40 0.21 0.16 0.16 0.35 0.38 10.74 10.20 9.95 9.68 5.01 11.58 9.76 10.69 8.85 8.86 11.39 8.51 9.88 9.95 9.82 1.73 1.37 0.27 0.21 0.49 0 48 0.45 0.13 0.07 0.30 1.40 77.00 57.00
,M0 70
cu
IV
A
Cb
0 10
%
R
R
...
..
..
0.13 0:49 0.18
..
..
..
..
0:09
0.20
.. .. .. .. .. ..
.. ..
..
..
0:io
..
o:o9 0 : io
0.13 0.07 0.08 0.12 0.08 0.11 ... 0.11 0.12 0.18 0 22 0.10 0.14 0.12 0 11
,.
, .
.. ..
.. ..
.. ..
..
... . . ,
.. 1'06 0:96
..
.. .. , . . ,
.. , .
, .
..
.. ..
... ...
...
.. .. ,.
..
.. .. .. ..
.. .. .. ..
..
...
.. ..
,.. ,..
. .
...
..
..
0 : i6
..
, . .
... ... ... ... .. ...
..
...
. .
... ... ...
..
...
o'i2 0.06 0.18 0.15 0.24 0.10 0.53
.. .. ..
...
0.94 0.76 0.87
..
..
.. ...
..
0.18
.. ..
...
...
..
..
..
..
...
.. ..
..
... ... ...
... ... ...
0.17 0.17
.. ..
. .
, . .
,.
.,
...
, .
1:46
.. ..
... ... ...
.. .. ..
,.
.. ..
4:13 4.98
..
0174
0.74
..
..
.. 0'71
.. ..
.. ..
.,
..
.. ..
.,.
..
,..
..
. , .
.. ..
7.011
, .
, .
,..
..
.. , .
..
..
.. ,.
..
.. ,.
18'00 ..
,. ,.
D.C. a i c experiments, actual niialyses of samples used A . C . spark experiments, typicnl analyses. a
Type of steels not otiierwi5e c,lassified given in .+IS1 numbers.
Filings for the direct current arc experiments were taken from rods, bars, strip, and cast electrodes. A total of approximately 50 mg. of filings was collected from a number of different points on the sample. These filings were mixed well and then 3 mg. of filings were placed on the concave platform of the lower electrode (3). -411 ,lower electrodes were machined from 0.8-cm. (5/16inch) ordinary spectrographic electrodes (Union Carbide and Carbon spectrographic grade). The electrodes were pre-arced 6 seconds after the current was adjusted to about 7 amperes before mounting the filings. SOURCE CONDITIOS~. The metallic cast electrodes for spark analysis are centered accurately on the optical axis by means of fixed electrode holders, arranged 2 mm. above and below the optical axis. A Baird Associates power panel supplies the 28,000volt condensed spark. A prespark of 30 seconds is employed and the exposure time is 45 seconds. The carbon electrodes for the direct current arc analysis are arranged 0.77 cm. above and 0.33 cm. below the optical axis, giving a total electrode space of 1.10 cm., with the source connected so that the lower electrode is positive. The arc is allowed
to vaporize the electrode until the entire platform is consumed and only the stub remains. This requires about 90 seconds. The arc is photographed during the entire arcing period to produce a single spectrogram. A grating spectrograph was used SPECTROGRAPH EMPLOYED. in these experiments. The grating is aluminum sputtered on Pyrex, ruled 15,000 lines to the inch, 3-meter focal length. The spectrograph was built a t The Carpenter Steel Company Laboratories and utilizes the Paschen mounting. Kodak Commercial sheet film (cut 5 X 25 cm., 2 X 10 inches) was used for the experiments on the Hasler-Dietert electrode (220-volt direct current arc), while 35-mm. Kodak Spectrum S o . 1 film was used for the cast electrodes in the high-voltage spark experiments. For spark spectra, the source gap is located 46.5 cm. from the slit. S o lens is used. A height-limiting aperture is placed 15 cm. from the slit. For direct current arc spectra the carbon electrodes are placed 120 cm. beyond the heightrlimiting aperture, which is in turn 13.0 cm. from the slit of the spectrograph for
I N D U S T R I A L A N D E N G I N E E R I N G CHEMISTRY
414
Table Ill. Analysis Steel NO.
1 2 3 4 5 6 7 9
I.\
