Study on characterization of outgassing of graphite

Study on characterization of outgassing of graphiteSutanwi Lahiri, S. Mohapatra, K. K. Mishra, K. B. Thakur, A. V. Bapat et al. Citation: AIP Conf. Proc. 1538, 38 (2013); doi: 10.1063/1.4810029 View online: http://dx.doi.org/10.1063/1.4810029 View Table of Contents: http://proceedings.aip.org/dbt/dbt.jsp?KEY=APCPCS&Volume=1538&Issue=1 Published by the AIP Publishing LLC. Additional information on AIP Conf. Proc.Journal Homepage: http://proceedings.aip.org/ Journal Information: http://proceedings.aip.org/about/about_the_proceedings Top downloads: http://proceedings.aip.org/dbt/most_downloaded.jsp?KEY=APCPCS Information for Authors: http://proceedings.aip.org/authors/information_for_authors

Downloaded 02 Jul 2013 to 180.149.51.67. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions

Study on characterization of outgassing of graphite

Sutanwi Lahiria, S Mohapatraa, K K Mishrab, K B Thakura, A V Bapatb, V K Magoa, A K Dasa and L M Gantayetc

aLaser and Plasma technology Division, bPower Beam Equipment Development Section, cBeam Technology Development Group, Bhabha Atomic Research Centre (BARC), Mumbai-400085,India

Abstract. Graphite is a widely used structural material for high temperature and high vacuum applications in furnaces handling corrosive materials or plasmas. The porosity, density, dimensions, impurities along with the history of graphitization, handling and storage of graphite dictate the outgassing characteristics. Graphite plates of density 1.78 g/cc were outgassed up to 1200 oC in vacuum of the order of 5x10-5 mbar in an in-house facility. A sample of the degassed graphite was taken for Thermo-Gravimetric Analysis (TGA). The TGA of the outgassed sample shows continuous weight loss in the temperature range of 200 oC to 1200 oC, indicating a much higher temperature is required for complete removal of gases. However, it may be noted that total outgassing is a function of the dimensions of the sample. The small coupon used in the TGA studies may not exactly replicate the large graphite blocks that outgases from the surface as well as from the bulk. The behaviour of graphite plates used in the liquid metal handling furnace where the graphite was finally used further confirmed this observation. The outgassing shown by the degassed graphite can be attributed to the diffusion of the absorbed gases from the bulk as well as desorption of the chemisorbed species which requires higher activation energies and higher temperature.

Keywords: Degassing; Graphite; Thermo-Gravimetric Analysis PACS: 07.30.Kf; 81.05.uf; 81.20.Ym

INTRODUCTION

The importance of graphite as a structural material in vacuum furnaces is well established due to its refractory nature and resistance to corrosion. The porosity and the density of graphite depend on the graphitization route. The theoretical density of graphite is 2.26 g/cc. However, synthetic graphite hardly reaches the theoretical value of density [1, 4]. In spite of the difficulties in handling, assembly and fabrication, graphite is the best choice for handling of liquid metals and plasmas due to its chemical passivity. The porosity, density, dimensions, impurities along with the history of graphitization, handling and storage of graphite dictate the outgassing characteristics [3-7]. In the present study, GLM-S grade of graphite manufactured by Graphite India Limited has been degassed and used in a metal processing furnace. Table 1 gives the details of the bulk properties of the graphite used. Contaminants from outgassing of graphite degrade the performance of the vacuum system. Therefore, degassing at high temperatures is an essential pretreatment conducted before its use in the metal processing vacuum furnace. The outgassing studies of graphite plates under various conditions are presented in this paper.

EXPERIMENTAL PROCEDURE

The degassing furnace is a high-vacuum double-walled water-cooled vessel pumped using rotary, roots and diffusion pump combination. The hot-zone is designed for a temperature of 1500 C. Graphite having physical properties mentioned in Table 1 was outgassed to a temperature of 1200 oC in vacuum of the order of 5x10-5 mbar in this degassing facility. Subsequently, the behaviour of the degassed graphite was studied in a similar metal processing vacuum furnace or test furnace.

The graphite plates of typical dimensions of 1500 mm x 200 mm x 40 mm, were outgassed in a batch process of about 165 kg charge in the round the clock shift operation. The change in pressure with step input in heater power was studied. During initial stages when change in temperature was less than 1 oC per hour, steady state was assumed to have been reached. If the pressure in the chamber was less than 5x10-5 mbar at steady state, next step input in power was given. High temperature outgassing was carried out by increasing the power. However, temperature of the hearth was deliberately reduced to 600 oC for safe operations in night shifts. The furnace was operated continuously for about 100 hours.

Carbon Materials 2012 (CCM12)AIP Conf. Proc. 1538, 38-42 (2013); doi: 10.1063/1.4810029

© 2013 AIP Publishing LLC 978-0-7354-1162-3/$30.00

38

Downloaded 02 Jul 2013 to 180.149.51.67. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions

Figure 1 shows a typical variation of vacuum as a function of temperature for a given batch. As the temperature is increased from room temperature to 2000C, a peak of moisture is observed indicative of the presence of moisture and volatile impurities [2]. A marked deterioration in vacuum from an initial value of 1x10-6 mbar to 5x10-5 mbar is observed. The degassing of the graphite was carried out at this temperature for 4 hours till there was an improvement in vacuum. A sudden deterioration of vacuum or increase in pressure is observed as the graphite attains at a particular temperature. This increase in pressure of the degassing chamber due to outgassing of different species from graphite surface as well as bulk is called an outgassing burst. An outgassing burst was observed to be around 400 oC with a sudden increase in pressure from 10-5 mbar to 10-4 mbar. Isothermal degassing was carried out for 10 hours to improve vacuum to 6x10-6 mbar. Subsequent peak was observed at 600 oC. The temperature was maintained at 600 oC for a period of 5 hours to allow the vacuum of the chamber to recover and reach 10-5 mbar again. Beyond 6000C, the pressure of the chamber deteriorated with the increase in temperature and. Similar batches were carried out for degassing up to 1200oC, in which the temperatures outgassing peaks were observed at 400oC, 600oC and 1200oC. After the furnace had cooled below 50 oC, the degassed material was used for TG analysis. The graphite blocks were further used in a furnace, one month and one year after degassing.

TABLE (1).Bulk properties of the graphite

Property Name Specification Make Graphite India Limited Grade GLM -S grade Bulk Density 1.78 gm/cc (min) Apparent Porosity 18% (max) Ash content 0.2% w/w Max. grain size 0.8 mm (max)

0 1000 2000 3000 4000 5000 60000

200

400

600

800

1000

1200

Time (min)Temp

eratur

e of gr

aphite

(deg C

)

10-7

10-6

10-5

10-4

10-3

Vacuu

m (mb

ar)

FIGURE 1. Schedule of outgassing operation.

RESULTS AND DISCUSSION

Thermo-gravimetric Analysis

The TGA of a sample taken from the degassed graphite charge shows continuous weight loss up to a temperature of 12000C, indicating that a much higher temperature is required for complete removal of gases [4]. Figure 2 shows the variation of weight loss of the graphite coupon with temperature. The graphite pellet was essentially cylindrical in shape having a length of about 4 mm and diameter 3mm. The outgassing of a sample can take place from the surface as well as the bulk of the substrate. For a thick substrate, initial outgassing from the surface is followed by diffusion of species across the thickness. Since the thickness of the sample used for TGA (3mm) is negligible

39

Downloaded 02 Jul 2013 to 180.149.51.67. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions

compared to that of the degassed plates, surface and bulk outgassing of the coupon is simultaneous. Hence, the small coupon used in the TGA studies may not exactly replicate the large graphite blocks that outgases from the surface as well as from the bulk. The outgassing rate from a cylinder of diameter 3 mm is much faster compared to the outgassing from the surface of graphite plates (1500 mm x 200 mm x 40 mm). Moreover, because of its porosity, the effective surface area is much larger than the geometrical area and the ratio of the surface area to volume becomes smaller with increasing thickness of the material [3]. However, the total outgassing of the graphite plates when carried out over a period of 100 hours leads to the bulk outgassing of the materials. The outgassing shown by the degassed graphite pellet can be attributed to the diffusion of the absorbed gases from the bulk as well as desorption of the chemisorbed species which requires higher activation energies and higher temperature.

FIGURE 2. Continuous weight loss of graphite pellet during TG analysis It is well reported in the literature that degassing cycle is reproduced on reheating in a vacuum and the amount of

gas released is a function of the duration of the exposure to air. The ten weeks exposure to atmospheric air enables up to one-half the original amount of gas to be build up in graphite [2]. The degassed graphite was subsequently used in the test furnace 30 days (Run 1) after degassing and the reduction in the gas content of the degassed graphite was noted. As the degassed graphite was used in the test furnace shortly after degassing, the gas content of the graphite was found to be very low compared to the trend shown by the material in the degassing furnace. The graphite was heated up to 970 0C. No sharp peaks were observed and the vacuum was always better than 3x10-5

mbar as shown in Fig. 3.

0 2000 4000 60000

200

400

600

800

1000

1200

Time (min)

Gra

phite

tem

pera

ture

(deg

C)

2x10-5

4x10-5

6x10-5

8x10-510-4

Vacu

um (m

bar)

Graphite outgassing as a function of temperature - 2011

FIGURE 3. Outgassing characteristic of graphite, one month after degassing

0 200 400 600 800 10000

1

2

3

4

5

6

% W

eight

loss

Sample Temperature, deg C

40

Downloaded 02 Jul 2013 to 180.149.51.67. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions

0 2000 4000 6000 8000 10000

0

200

400

600

800

1000

1200

1400

Time (min)

Grap

hite t

empe

rature

(deg

C)

10-5

10-4

Graphite outgassing as a function of temperature - 2012

Vacu

um (m

bar)

FIGURE 4. Outgassing characteristic of graphite, one year after degassing

The graphite was then used in furnace after a year (Run 2) and heated to a temperature of 12100C. The behaviour

of the graphite was similar to that observed in the previous test. However, the graphite showed peaks of outgassing at temperatures in the range of 4000C to 6000C, as shown in Fig. 4. This is due to the fact that the physisorbed species on the graphite surface comes out. The absence such peaks at higher temperatures confirms that there has been no chemisorption of gases after the degassing activity, even after a period of one year.

For a system having a constant pumping speed, the integrated area under the curve showing the variation of outgassing rate with time will give the total outgassing from a given volume of material. The outgassing rate is given by:

dQ/dt = S ∆P The total outgassing from a given material over time t:

Where S is the average pumping speed, dQ/dt is the outgassing rate and P is the pressure of the vacuum furnace. Table 2 presents a comparison of the total outgassing load from the degassing furnace with that from the test

furnace. It is observed that the outgassing from the test furnace one year after degassing was almost equal to the load observed in the degassing furnace. The total gas load during degassing was found to be 1.2x10-7 mbar l/cm2, an order lower than that reported by Beitel [7]. In fact, a slight higher load is observed in the test furnace when used a year after degassing. This may be attributed to the absorption of moisture while reloading the graphite in the furnace. The sharp peaks observed in Run 2 up to a temperature of 6000C and the moisture peaks in the RGA data in this temperature range further confirms this observation. It was further observed that the residual gas content of the Run 2 showed a dominance of moisture (mass number= 18) as compared to the RGA data obtained during degassing and Run 1. The comparison is given in Table 2. The value of the moisture content reported here is on a higher side compared to the values reported in literature [2, 9]. This is attributed to the effect of outgassing of the chamber, higher relative humidity of the air, improper handling during loading/ unloading or inefficiency of the pumping system. In a nutshell, the present study will serve as a baseline data for estimating the load on the vacuum pumps due to the presence of the graphite and will be compared to the subsequent improvements made to minimize the outgassing load.

TABLE (2). Outgassing load observed

Run Volume of the furnace, m3

Effective pumping speed, l/s

Geometric area exposed, cm2

Total outgassing load, mbar l/cm2

Degassing run 4.54 7000 19560 1.02E-7 First test, 2011 9.42 17400 56000 8.48E-8 Second test, 2012 9.42 17400 56000 1.2E-7

41

Downloaded 02 Jul 2013 to 180.149.51.67. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions

TABLE (3). Comparison of the moisture content Temperature, 0C Fractional composition (% vol/vol) for amu=18

Degassing Run 1 Run 2 36 to 350 37.2 35.9 77.1

350 to 800 35.1 20.6 21.6 800 to 1250 2.25 2.0 11.8

CONCLUSIONS

The degassing of graphite is an essential prerequisite for use of graphite in liquid metal handling furnaces. The chamber pressure observed in the degassing furnace up to a temperature of 1200 oC gave an idea of its gas content of the graphite as obtained from Graphite India Limited. The degassing of graphite shows major degassing at substrate temperatures of 400 oC, 600 oC and 1200 oC. The subsequent use of graphite in vacuum test chamber within one month of outgassing showed the absence of any pressure surges in the vacuum chamber. On using the components wrapped in plastic bags and preserving in wooden crates, initial desorption of the physisorbed species to a temperature of 600 0C was observed. However, the absence of very sharp outgassing peaks rule out the possibilities of chemisorption of gases. Hence, it can be concluded that chemisorption of gases and other volatile impurities takes place during the graphitization process which are removed by vacuum-baking at high temperatures. Further chemisorption of species on exposure to atmosphere is found to be unlikely. However, the degassing is effective when the degassed components are used immediately. The gas load on the vacuum pumps can be minimized and the low vapour pressure of graphite compared to the other refractory metals like Ta, W, Nb, coupled with the chemical passivity make it an obvious choice in diverse areas of vacuum engineering.

ACKNOWLEDGMENT

The authors wish to thank Dr D Das from Chemistry Division for his help in carrying out TGA and all operating staff at degassing furnace.

REFERENCES

1. R. Carlson and J. Ferritto, US Patent No. 4,226,900 (7 October 1980). 2. E. N. Marmer and Y. P. Lyakhin, Poroshkovaya Metallurgiya 4 (124), 75-86 (1993). 3. Y. Watanabe, T. Tanabe and S. Imoto, Fusion Eng. Des. 9, 143-148 (1989). 4. W. Espe, “Materials of High Vacuum Technology- Volume 1”, New York: Pergamon Press, 1966, pp. 888. 5. S. Sukenbou and Y. Gamay, J. Vac. Sci. Technol. A 1(1), 96-97 (1983). 6. J. D. Herbert, A.E. Groome and R .J. Reid, J Vac Sci Technol. A 15, 1767 (1994). 7. G.A. Beitel, J Vac Sci. Technol.A 8(5), 647-657 (1971). 8. M. Dee and C. B. Ramsey, Nucl. Instr. and Meth. in Phys. Res. , 172, 449-453 (2000). 9. J. P. Redmond and P. L. Walker, Nature 186(4718), 72-74 (1960).

42

Downloaded 02 Jul 2013 to 180.149.51.67. This article is copyrighted as indicated in the abstract. Reuse of AIP content is subject to the terms at: http://proceedings.aip.org/about/rights_permissions