George R. Harrison, "Intensity Relations in the Spectra of Titanium I. Line Intensities in the Stronger Multiplets of Ti I and Ti II," J. Opt. Soc. Am. 17, 389-416 (1928)
The relative intensities of the lines in 26 of the stronger multiplets of the spectrum of normal titanium, Ti I, and in 8 of the stronger multiplets due to the singly ionized atom, Ti II, were measured by a general method of photographic photometry which is described in detail. In Ti I, 58% of the multiplets were found to obey the intensity formulas to well within 5%, while in the 42% of abnormal multiplets, 71% of the lines were found normal, with 16% abnormally weak and 13% abnormally strong. In Ti II, 62% of the multiplets measured were normal, while of the abnormal 38%, 61% of the lines were normal, with 21% abnormally weak and 18% abnormally strong. Of all the lines measured, 86% appeared normal in intensity, while 7.6% were too weak and 6.4% were too strong. The departures from the predicted intensities are often fairly large, generally being of the order of 10% to 40%. No definite correlation was found between intensity abnormalities and depatures from Lande’s interval rule or the normal Zeeman patterns.
You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
Variation indicates unknown line or band lap. Found entirely normal on some plates.
Corrected for lap a3F′2–a3D′1, sum of two being normal.
Intervals normal.
Second order satellites too weak to measure, but look normal.
Remaining lines too weak to measure.
Ghost lap correction. a5G intervals abnormal.
Table 9
(36) a4P′–a4D′ Class V.
λ
a4P′
a4D′
Ik
Ic
Im
Z
4330.71
3
2
0
3
5
n?
4320.95
2
1
1
5
8
n
4314.98
1
1
40
25
Lap
n?
4312.88
3
3
35
27
17 ±5
n
4307.89
2
2
40
32
28
n
4301.93
1
2
15
25
25
n?
4300.05
3
4
60
120
132
n?
4290.23
2
3
50
63
44
n?
Note disagreement between measurements and King’s estimates, being in some cases on opposite sides of the theoretical values. Great variation of λ4313 seems to indicate that unknown line is superposed on it. There is some indication of this also on λ4308. a4P′ intervals abnormal; a4D′ intervals abnormal.
Table 10
(60) b4F′–a4G′ Classes III, IV & V.
λ
b4F′
a4G′
Ik
Ic
Im
3513.08
4
3
tr
1
—
3500.33
3
3
2
29
22
3491.05
2
3
10
180
180
3489.74
4
4
2
38
28
3477.18
3
4
15
241
245
3476.99
5
5
tr
29
3461.50
4
5
20
321
322
3444.31
5
6
30
420
410
Intervals practically normal.
No Zeeman measurements.
Corrected for lap of a 5F′1–b5G′2.
As self-reversal was never completely eliminated from this multiplet, it is possible that the last line in the table should be 42 and the one above it 35, though that this is less probable than the values given above is indicated by comparison with self-reversal curves of neighboring multiplets photographed simultaneously. a3D′ intervals abnormal.
Table 22
(138) a3G–b3H Classes III & IV.
λ
a3G
b3H
Ik
Ic
Im
Z
4925.42
4
4
5
10
10
n?
4915.24
5
5
5
10
10
n?
4913.63
3
4
20
194
190
n
4899.93
4
5
20
240
240
n
4885.09
5
6
20
296
300
n
Intervals practically normal.
Table 23
(141) a3H′–a3I′ Classes III & IV.
λ
a3H′
a3I′
Ik
Ic
Im
4893.06
5
5
2
14.3
—
4882.34
6
6
2
14.3
25
4870.14
5
6
20
500
540
4868.28
4
5
18
421
465
4856.01
6
7
20
594
594
Large background corrections, and numerous underlying bands. a3E′ intervals abnormal, a3I′ intervals normal.
Variation indicates lap by unknown line.
Accuracy not high because of large background correction.
Intervals very abnormal.
Table 25
(153) a3H′–c3H Class III.
λ
a3H′
c3H
Ik
Ic
Im
4769.78
6
5
4
3.2
4.3
4766.33
5
4
4
3.2
4.4
4759.28
6
6
25
114
114
4758.13
5
5
25
93
94
4747.69
5
6
3
3.2
3.4
4742.80
4
4
20
78
48
4734.68
4
5
3
3.2
3.2
Large background and band corrections. Long wave-length satellites are lapped with bands, but look really stronger than short ones. a3H′ intervals abnormal; c3H intervals normal.
Line lap correction.
Self-reversal was not eliminated enough to make sure satellites were correct relative to main lines, but curves look normal.
Intervals normal.
Obtained assuming same self-reversal as No. 320, which arises from same lower state and a very close upper state to this multiplet. Corrected for ghost laps. Varying amounts self-reversal used, but never completely removed.
Accuracy very low, due to strong bands present when self-reversal was low, and to numerous line laps and close lines.
Intervals d3D′ somewhat abnormal.
Variable relative to 3379.20 due to underlying bands, but appears equal to it.
Exact lap with a2F′3–a2G′4, sum of two being about 20% below sum estimated, though this deviation may be spurious due to difficulties of measurement.
Estimated from self-reversal curves.
This multiplet was difficult to measure, due to underlying bands, close lines necessitating use of narrow slit, and numerous laps.
Intervals practically normal.
Variation indicates unknown line or band lap. Found entirely normal on some plates.
Corrected for lap a3F′2–a3D′1, sum of two being normal.
Intervals normal.
Second order satellites too weak to measure, but look normal.
Remaining lines too weak to measure.
Ghost lap correction. a5G intervals abnormal.
Table 9
(36) a4P′–a4D′ Class V.
λ
a4P′
a4D′
Ik
Ic
Im
Z
4330.71
3
2
0
3
5
n?
4320.95
2
1
1
5
8
n
4314.98
1
1
40
25
Lap
n?
4312.88
3
3
35
27
17 ±5
n
4307.89
2
2
40
32
28
n
4301.93
1
2
15
25
25
n?
4300.05
3
4
60
120
132
n?
4290.23
2
3
50
63
44
n?
Note disagreement between measurements and King’s estimates, being in some cases on opposite sides of the theoretical values. Great variation of λ4313 seems to indicate that unknown line is superposed on it. There is some indication of this also on λ4308. a4P′ intervals abnormal; a4D′ intervals abnormal.
Table 10
(60) b4F′–a4G′ Classes III, IV & V.
λ
b4F′
a4G′
Ik
Ic
Im
3513.08
4
3
tr
1
—
3500.33
3
3
2
29
22
3491.05
2
3
10
180
180
3489.74
4
4
2
38
28
3477.18
3
4
15
241
245
3476.99
5
5
tr
29
3461.50
4
5
20
321
322
3444.31
5
6
30
420
410
Intervals practically normal.
No Zeeman measurements.
Corrected for lap of a 5F′1–b5G′2.
As self-reversal was never completely eliminated from this multiplet, it is possible that the last line in the table should be 42 and the one above it 35, though that this is less probable than the values given above is indicated by comparison with self-reversal curves of neighboring multiplets photographed simultaneously. a3D′ intervals abnormal.
Table 22
(138) a3G–b3H Classes III & IV.
λ
a3G
b3H
Ik
Ic
Im
Z
4925.42
4
4
5
10
10
n?
4915.24
5
5
5
10
10
n?
4913.63
3
4
20
194
190
n
4899.93
4
5
20
240
240
n
4885.09
5
6
20
296
300
n
Intervals practically normal.
Table 23
(141) a3H′–a3I′ Classes III & IV.
λ
a3H′
a3I′
Ik
Ic
Im
4893.06
5
5
2
14.3
—
4882.34
6
6
2
14.3
25
4870.14
5
6
20
500
540
4868.28
4
5
18
421
465
4856.01
6
7
20
594
594
Large background corrections, and numerous underlying bands. a3E′ intervals abnormal, a3I′ intervals normal.
Variation indicates lap by unknown line.
Accuracy not high because of large background correction.
Intervals very abnormal.
Table 25
(153) a3H′–c3H Class III.
λ
a3H′
c3H
Ik
Ic
Im
4769.78
6
5
4
3.2
4.3
4766.33
5
4
4
3.2
4.4
4759.28
6
6
25
114
114
4758.13
5
5
25
93
94
4747.69
5
6
3
3.2
3.4
4742.80
4
4
20
78
48
4734.68
4
5
3
3.2
3.2
Large background and band corrections. Long wave-length satellites are lapped with bands, but look really stronger than short ones. a3H′ intervals abnormal; c3H intervals normal.
Line lap correction.
Self-reversal was not eliminated enough to make sure satellites were correct relative to main lines, but curves look normal.
Intervals normal.
Obtained assuming same self-reversal as No. 320, which arises from same lower state and a very close upper state to this multiplet. Corrected for ghost laps. Varying amounts self-reversal used, but never completely removed.
Accuracy very low, due to strong bands present when self-reversal was low, and to numerous line laps and close lines.
Intervals d3D′ somewhat abnormal.
Variable relative to 3379.20 due to underlying bands, but appears equal to it.
Exact lap with a2F′3–a2G′4, sum of two being about 20% below sum estimated, though this deviation may be spurious due to difficulties of measurement.
Estimated from self-reversal curves.
This multiplet was difficult to measure, due to underlying bands, close lines necessitating use of narrow slit, and numerous laps.
Intervals practically normal.