Detection of the impurities in Nd doped phosphate laser glass by x-ray fluorescence technique
Atul Kumar*, M. P. Kamath, A. S. Joshi, P. A. Naik, P. D. Gupta
Laser Plasma Division
A. K. Singh, M. K. Tiwari
Indus Synchrotrons Utilization Division
Raja Ramanna Centre for Advanced Technology, Indore 452013, M.P., India
*Email: katul@rrcat.gov.in

Introduction:

The advances in laser glass composition and manufacturing have enabled the development of third generation high energy high power (HEHP) lasers like the National Ignition facility (NIF)1 that are being used as drivers for fusion energy studies. These laser glasses are phosphate based because of their low intensity dependent refractive index n2 and have a compositions equivalent to LHG-8 of M/s Hoya, Japan or LG-770 glass of M/s Schott Ltd2. The manufacturing and handling procedures introduce many unwanted impurities in these glasses, leading to lower laser induced damage threshold and lower fluorescence life time. The x-ray fluorescence (XRF) is a powerful technique3 which can identify the elemental composition of such glasses in a non-destructive manner. In the XRF process, the specimen is excited with high energy photons, which can cause individual atoms to ionize. When these atoms undergo relaxation, the electrons relocate from a high energy orbital to a lower one. During this process, fluorescence radiation is emitted. The emitted radiation has lower energy compared to the incident source and it depends on the specific electron transitions of the individual elements. Thus, the resulting x-ray fluorescence spectrum bears the signature of the elements present in the specimen.
In the present study, we have performed XRF analysis on different laser glass samples and raw materials used for the fabrication of these glasses namely aluminum metaphosphate (AMP), barium metaphosphate (BMP) and potassium metaphosphate (KMP). The advantage of the this study is that it can indicate to the presence of most of the impurities like transition metal impurities (of interest to us) in one measurement, if their concentrations are of the order of few ppm (by weight) or more. Our studies on the glasses and the raw materials indicate that the raw materials are pure to the desired levels but the glasses fabricated using the raw materials show many impurities. These impurities are expected to enter during handling of the raw materials and from the glass-cullet during fabrication process. This study will help in improving the quality of the glass by repeatedly doing the XRF analysis after every stage, so that faulty handling can be identified.

Method of fabrication of the phosphate laser glass:

A two arm HEHP Nd:phosphate glass laser chain is operational at Raja Ramanna Centre for Advanced Technology (RRCAT), Indore4. A joint collaborative project is undertaken with Central Glass and Ceramic Research Institute (CGCRI), Kolkata to develop the process technology for the fabrication of the phosphate laser glasses equivalent to LHG-8 glass of M/s Hoya, to overcome the existing embargo. The phosphate laser glass is fabricated as a two step process. In the first step, the sintered raw material (AMP, BMP and KMP and Nd2O3 in appropriate fractions) is melted in a silica crucible. The glass-cullet produced as a result of melting in the silica crucible is cast. The cast glass is powdered and loaded in a bottom pouring platinum crucible for optical quality, and chemical and physical homogenization. During these steps in the fabrication, the impurities may be added to the glass at several stages, namely: a) mixing of the raw materials, b) impurities coming from the silica crucible, c) impurities coming from the glass-cullet cast, d) impurities coming from the platinum crucible, e) impurities coming during annealing and fine annealing of the glass.
Six samples of laser glass (named LG-1 to LG-6) were fabricated at CGCRI for the studies. XRF analysis was done on different fabricated samples of laser glasses (LG-1 to LG-6) and raw material samples made from AMP, BMP and KMP, to identify the elemental impurities in laser glass and raw materials. In addition to this, the sensitivity of the XRF technique was also ascertained using small amounts of platinum deliberately added to the glass samples.

Experiments and Methodology

XRF analysis was performed at the BL-16 beamline of Indus-2 synchrotron facility5 on fabricated glass samples and the raw materials after the first step of melting in silica crucible. The glasses had platinum oxide dissolved in them because of its solubility in the laser glass. Platinum comes into the glass as an unavoidable impurity during homogenization of the glass in the platinum crucible. Metallic platinum absorbs at the lasing wavelength (1054 nm) and reduces the laser induced damage threshold (LIDT) of the laser glass. Oxygen bubbling is done in platinum crucible to reduce the OH bond impurity in the hygroscopic phosphate glass. The oxygen bubbling also oxidizes the platinum metal to its oxide, thereby reducing the metallic platinum in the glass. The other impurities occur because of the poor quality of the silica crucible and/or cast of glass-cullet in addition to poor handling during the mixing of the raw materials in the first step of melting.
The glasses with different amounts of platinum oxide were: LG-1(5 ppm Pt), LG-2 (10 ppm Pt), LG-3(25 ppm Pt), LG-4 (5 ppm Pt with O2 bubbling), LG-5 (10 ppm Pt with O2 bubbling) and LG-6 (25 ppm Pt with O2 bubbling). Varying platinum oxide content helped in ascertaining the sensitivity of the technique to Pt impurity. In the experiment, an x-ray beam was incident on sample for excitation and scattered spectrum was recorded by the detector shown in Fig. 1. The samples were excited using 14 keV and 17 keV synchrotron x-rays, selected by a double-crystal monochromator system. The analyses have been done on all the raw powder materials (for qualitative analysis) and on the pallets (for quantitative analysis). 15 mm dia. pallets were formed using a semi-automatic pallet machine. One such pallet is shown in sample holder in Fig. 1.

Figure 1: Raw specimens analyzed by XRF measurements
The captured data was analyzed using the PyMCA software 6. The calibration of the measured fluorescence spectrum was achieved by selecting the ‘K’ lines of various elements present in the laser glass samples, and in the raw material samples. The method of spectrum calibration however needs a proper comparison with a standard doped sample as a confirmatory test to account for the variation in the actual composition of the fabricated laser glass samples and that measured with XRF. In the present study, XRF method was used to qualitatively see the presence of different impurities, strictly from the yes/no point of view, rather than the full quantitative estimation.

Results and discussion:

Figure 2 shows a typical spectrum of the fabricated LG-1 glasses and the raw material BMP. The impurities found in the fabricated glasses are listed in Table 1. The silicon impurity occurs because of melting in silica crucible. Sr impurity seen in BMP as Ba and Sr are chemically close to each other. For the same reason, Rb impurity was found in KMP. However, there no impurities were seen in AMP. The transition metal impurities however may come either from during handling of raw materials or through the casts of glass-cullet. This will be ascertained in future by casting the glass in casts made of different materials like graphite, aluminum, and stainless steel. The studies of impurities after the casting of the glass will give idea about its origin. In addition to this, the XRF technique was found to be sensitive to ~ 5 ppm of Pt impurities.

Figure 2: a) Typical spectrum of a) LG-1 fabricated glass, and b) Barium meta-phosphate specimen (raw material)

Sample
 
Elements

5ppm Pt

10ppm Pt

25ppm Pt

LG-1
 
( in ppm )

LG-4
(with O2 )
( in ppm )

LG-2
 
( in ppm )

LG-5
(with O2 )
( in ppm )

LG-3
 
( in ppm )

LG-6
(with O2 )
( in ppm )

Detected Elements

P

489890

449680

367020

421230

388300

437580

K

112000

112000

112000

112000

112000

112000

Ba

110220

87320

61910

64930

71500

63170

Si

3520

23230

9890

20640

11820

5970

Transition metal impurities

Fe

495

229

155

1540

344

269

Ni

11

6.1

3.6

21.6

7.1

3.3

Cu

37

19.9

7.3

50.2

19.2

8.7

Zn

64

29.6

15.4

100

28.8

23.5

Sb

1420

693

177

307

133

565

Pt

4.5

3.3

4.2

6.4

9.7

8.3

Pb

40.5

14.8

22.4

48.1

19.7

22.1

 
Table 1: List of the impurities present in the glass samples

Summary:

We have presented XRF analysis on different laser glasses (LG-1 to LG-6) and raw materials (AMP, BMP and KMP) used for the fabrication of such glasses. The impurities in these glasses may be entering during handling or coming from the silica crucible, or the cast for the glass-cullet. The XRF method was also found to be sensitive up to ~ 3 ppm of platinum, as was observed in studies of LG-4 glass. In contrast, the raw materials were found to be free from transition element impurities. Further studies to improve the quality of glass are required to be undertaken.
Acknowledgement:
We acknowledge the contribution of Dr. R. Sen, Dr. Annapurna and other colleagues of the glass division, CGCRI Kolkata for preparation of the phosphate laser glass samples for the XRF studies.
References:

L. L. Seaver, LLNL report (925) 423 (Oct. 2010)
J. H. Campbell and T. I. Suratwala, J. Non-Cryst. Solids 318, 2638 (2000).
M. Mantler and M. Schreiner, X-Ray Spectrum. 29, 3 (2000)
A.S. Joshi et al, EPJ Web of Conferences 59, 08001 (2013).
M. K. Tiwari, P. Gupta, A. K. Sinha, S. R. Kane, A. K. Singh, S. R. Garg, C. K. Garg, G. S. Lodha and S. K. Deb, J. Synchrotron Rad. 20, 386 (2013).
V.A. Solé, E. Papillon, M. Cotte, Ph. Walter, J. Susini, Spectrochimica Acta B 62, 63 (2007)