FURNACE ATMOSPHERE MEASUREMENT

Abstract

A method of determining the concentration of a species in a portion of a furnace atmosphere is described. The method comprises the steps of measuring first, second and third intensities of electromagnetic radiation in the furnace at first, second and third wavelengths respectively. The third wavelength is selected to be representative of absorption of electromagnetic radiation by the species. A fourth intensity of electromagnetic radiation is calculated, being an estimate of the intensity of electromagnetic radiation in the furnace at the third wavelength absent any absorbing species in the furnace atmosphere. The third intensity and the fourth intensities are used to determine a parameter that is proportional to the concentration of absorbing species in the portion of the furnace atmosphere. Apparatus for carrying out the method is also described.

Claims

1.-32. (canceled)

33. A method of determining the concentration of a species in a portion of a furnace atmosphere comprising the steps: (i) measuring a first intensity I.sub.1 of electromagnetic radiation in the furnace at a first wavelength λ.sub.1; (ii) measuring a second intensity I.sub.2 of electromagnetic radiation in the furnace at a second wavelength λ.sub.2; (iii) measuring a third intensity I.sub.3 of electromagnetic radiation in the furnace at a third wavelength λ.sub.3, the third wavelength being selected to be representative of the absorption of electromagnetic radiation by the species; (iv) calculating a fourth intensity I.sub.4 at the third wavelength using the first and second intensities, the fourth intensity being an estimate of the intensity of electromagnetic radiation in the furnace at the third wavelength absent any absorbing species in the furnace atmosphere; and (v) using the third intensity and the fourth intensity to determine a parameter that is proportional to the concentration of absorbing species in the portion of the furnace atmosphere.

34. The method according to claim 33, wherein at step (v) the parameter is determined using the difference between the third intensity and the fourth intensity.

35. The method according to claim 33, wherein the measurement steps (i), (ii) and (iii) measure the respective intensity through a measurement path length and wherein the parameter or the third and fourth intensities are corrected for the measurement path length.

36. The method according to claim 33, wherein the measurement steps (i), (ii) and (iii) measure the respective intensity through a measurement path length and wherein the first intensity, the second intensity and the third intensity are corrected for the measurement path length.

37. The method according to claim 35, wherein the correction for the measurement path length is made in accordance with the Lambert-Beer Law.

38. The method according to claim 33, wherein the parameter is used to determine a concentration of the species in the portion of the furnace atmosphere.

39. The method according to claim 38, wherein the concentration is a qualitative concentration of the species in the portion of the furnace atmosphere or a quantitative concentration of the species in the portion of the furnace atmosphere.

40. The method according to claim 38, wherein the parameter is used to determine the quantitative concentration of the species in the portion of the furnace atmosphere by using an absorption coefficient for the particular species at the third wavelength, the absorption coefficient indicating the absorption strength of a unit quantity of the species per unit length.

41. The method according to claim 33, wherein the first and second intensities ((I.sub.1, I.sub.2) are used to determine the temperature inside the furnace, the temperature being determined by calculating the gradient δI/δ λ and comparing the calculated gradient with the theoretical gradient at that wavelength, the determined temperature in the furnace then being used to calculate a blackbody spectrum, thereby allowing a calculation of an intensity at the third wavelength.

42. The method according to claim 33, wherein the fourth intensity is calculated by linear interpolation between the first intensity at the first wavelength and the second intensity at the second wavelength to obtain a predicted intensity at the third wavelength.

43. The method according to claim 33, wherein the first wavelength and the second wavelength are shorter than the third wavelength.

44. The method according to claim 33, wherein the first wavelength is shorter than the third wavelength and the second wavelength is longer than the third wavelength.

45. The method according to claim 33, wherein at least a portion of the electromagnetic radiation in the furnace has a wavelength between 400 nm and 800 nm.

46. The method according to claim 33, wherein the species is a metal species.

47. The method according to claim 46, wherein the metal species is an alkali metal species.

48. The method according to claim 46, wherein the metal species is sodium.

49. The method according to claim 48, wherein the third wavelength is between 585 nm and 595 nm, preferably between 588 nm and 591 nm.

50. The method according to claim 48, wherein the first wavelength is between 500 nm and 584.9 nm, preferably between 500 nm and 575 nm, even more preferably between 565 nm and 575 nm.

51. The method according to claim 48, wherein the second wavelength is between 500 nm and 584.9 nm, preferably between 575.1 nm and 584.9 nm. more preferably between 578 nm and 584 nm.

52. The method according to claim 48, wherein the second wavelength is between 596 nm and 700 nm.

53. The method according to claim 33, wherein the measured intensity at more than two wavelengths is used to determine the intensity at the third wavelength.

54. The method according to claim 33, wherein the furnace is a glass making furnace, preferably a furnace making soda-lime-silica glass or borosilicate glass.

55. An apparatus for measuring the concentration of a species in a portion of a furnace atmosphere, the species absorbing electromagnetic radiation in the furnace, the apparatus comprising: collection means for collecting electromagnetic radiation from inside the furnace; coupling means for directing the collected electromagnetic radiation to a suitable measuring means, the measuring means being configured to measure the intensity of the collected electromagnetic radiation at one or more wavelengths; and computing means, in particular a computer, configured to take the output from the measuring means to determine the amount of absorption of electromagnetic radiation inside the furnace at a measurement wavelength, the measurement wavelength being selected to be representative of the absorption of electromagnetic radiation in the furnace atmosphere by the species.

56. The apparatus according to claim 55, wherein the collection means and/or coupling means comprises one or more optical fibre and/or one or more mirror.

57. The apparatus according to claim 55, wherein the coupling means comprises a jacket for locating in a hole in a wall section or roof section of the furnace.

58. The apparatus according to claim 55, further comprising cooling means for cooling the collection means and/or the coupling means.

59. The apparatus according to claim 58, wherein the cooling means comprises one or more fluid coolant.

60. The apparatus according to claim 59, wherein the cooling means comprises at least one water cooling circuit and/or at least one air purge.

61. The apparatus according to claim 58, wherein the cooling means comprises one or more electrical cooler, in particular a Peltier cooler.

62. The apparatus according to claim 55, wherein the measuring means comprises a spectrometer.

63. The apparatus according to claim 55, wherein the measuring means is sensitive to electromagnetic radiation having a wavelength between 380 nm and 800 nm.

64. The apparatus according to claim 55, wherein the species is a metal species, preferably an alkali metal species, more preferably sodium.

Description

[0075] The invention will now be described with reference to the following figures:

[0076] FIG. 1 shows a spectral distribution from inside a float glass making furnace.

[0077] FIG. 2 shows a schematic representation of a furnace showing a measurement position for measuring the intensity of electromagnetic radiation inside the furnace.

[0078] FIG. 3 shows the intensity variation with wavelength for light inside the furnace; and

[0079] FIG. 4 shows a comparison of the total sodium concentration determined using a method according to the present invention compared with a conventional extractive technique.

[0080] FIG. 1 shows the spectral distribution in the refiner end of a float glass making furnace. The temperature at the measuring position is around 1400° C. Axis 2 is the wavelength in nm, and axis 4 is the measured power in watts (W). As can be seen there is an absorption 6 due to sodium atoms in the atmosphere at around 589.5 nm.

[0081] Up to around 600 nm to 640 nm, there is reasonable agreement between the spectrum that is measured and that which is calculated from Planck's Law (other than the absorption around 589.5 nm due to sodium in the furnace atmosphere).

[0082] FIG. 2 shows a schematic representation of a glass making furnace to show the location of the measurement probe for making measurements of the intensity of visible light inside the furnace.

[0083] The glass making furnace 10 has a feed end 12 in which glass making raw materials are fed into the furnace 10 for conversion into molten glass. The furnace has a first burner section 14 and a second burner section 16 for providing the necessary heat into the raw materials for conversion to molten glass. The molten glass exits the furnace through outlet 18 and travels in the direction of arrow 20 to other parts of the glass making furnace, for example a float bath, rollers or container mould.

[0084] The furnace 10 has a rectangular configuration with two substantially parallel side walls 21, 23 substantially parallel to the direction of arrow 20. At the outlet end is an end wall 22. The furnace has a bottom and a roof (both not shown). All the walls/roof/bottom are made of suitable refractory materials. Furnaces for making/processing other materials have a similar construction.

[0085] Located in a hole in the end wall 22 of the furnace 10 is a suitably water cooled jacket 30. The water cooled jacket 30 is an elongate tubular member that houses an optical fibre suitably configured to view across the furnace in the direction of the arrow 24 along the line l-l′. That is, the optical fibre in the water cooled jacket 30 is configured to receive electromagnetic radiation i.e. visible light from within the furnace that travels along the measurement path length l-l′ counter to the direction of arrow 24. The water cooled jacket may include an air purge and/or Peltier cooler for additional cooling. The direction 24 may be substantially parallel to the end wall 22 or at an angle thereto, in particular perpendicular thereto.

[0086] The optical fibre 32 is coupled to a suitable spectrometer 34, for example a Thorlab CSS 100 (commercially available from www.thorlabs.de).

[0087] The wavelength dependent intensity response of the spectrometer may be calibrated by measuring the intensity response to a calibrated black body source at different temperatures.

[0088] The spectrometer 34 is in electrical communication with computer 38 via a suitable cable 36. The computer 38 may also be used to control the spectrometer 34 and exchange data therewith. The computer 38 is used to calculate the sodium concentration in the atmosphere of the furnace 10 based on the wavelength dependent intensity measurements made by the spectrometer 34.

[0089] In this particular example the furnace 10 was a float glass making furnace and the water cooled jacket was located in the refiner end of the furnace. The temperature in this region is around 1400-1500° C.

[0090] Light from inside the furnace 10 is collected by the suitably configured optical fibre located in the water cooled jacket 30, the optical fibre being suitably coupled to the spectrometer 34.

[0091] The spectrometer is used to measure the intensity of light at two wavelengths, 572 nm and 582 nm. It is possible to measure across a wavelength range and to use two wavelengths of interest. It is possible to use more than two wavelengths. Preferably the intensity at each of the two wavelengths is measured at the same time, although they could be measured one after the other.

[0092] In a variant to the arrangement shown in FIG. 2, the water cooled jacket 30 may be located in a hole in the sidewall 23, such that the water cooled jacket is substantially perpendicular to the sidewall 23. The optical fibre may be configured to collect light from within the furnace in a direction parallel to the longitudinal axis of the water cooled jacket, and not in a direction perpendicular to the longitudinal axis of the water cooled jacket as shown in FIG. 2. That is, the water cooled jacket may be located in a hole in the sidewall 23 and the optical fibre may be arranged to view across the furnace in the direction of arrow 24.

[0093] In another variant, light from inside the furnace exits a hole in one of the furnace walls, for example the sidewall or end wall or roof, and is guided via suitably positioned mirrors and/or lenses for coupling to the spectrometer.

[0094] FIG. 3 shows the intensity spectrum for a measurement of electromagnetic radiation inside the furnace 10. The axis 40 is the intensity and the axis 42 is the wavelength in nm.

[0095] The intensity variation with wavelength across the region 570 nm to 610 nm, shown as line 44, was measured using a Thorlab CSS-100 spectrometer.

[0096] As can be seen in FIG. 3, the absorption due to sodium atoms in the furnace atmosphere is clearly visible at about 589 nm.

[0097] The intensity at 572 nm (dotted line a-a′) and 582 nm (dotted line b-b′) was used to determine an intensity at 589 nm (dotted line c-c′) absent any absorbing species at 589 nm.

[0098] The wavelength of 589 nm is representative of the wavelength at which elemental sodium atoms absorb visible electromagnetic radiation.

[0099] The predicted intensity at 589 nm was determined by interpolation from the measured intensities at 572 nm and 582 nm. The interpolation may be based on the Plank equation, or some other algorithm using the measured intensities. More than two measured intensities may be used in the interpolation to obtain the intensity at 589 nm absent any absorbing species.

[0100] In the example shown in FIG. 3, linear interpolation (dashed line 46) was used to determine an intensity at 589 nm. The intensity at 589 nm determined in this way is an estimate of the intensity at 589 nm without there being any absorption due to sodium atoms in the furnace atmosphere (or at least in the portion of the furnace atmosphere under investigation).

[0101] Having used the measured intensities to determine an estimated intensity at 589 nm without any sodium atoms in the furnace atmosphere, this is then compared with the actual measured intensity at 589 nm, clearly seen as trough 48 in FIG. 3. The absorption due to sodium atoms is then determined by subtraction i.e. the measured intensity at 589 nm is subtracted from the estimated intensity at 589 nm to give a difference δI.sub.589nm.

[0102] The difference δI.sub.589 nm is a measure of the absorption due to sodium atoms in the furnace atmosphere at the measuring position 28 (with reference to FIG. 2) i.e. δI.sub.589nm is proportional to the concentration of sodium atoms in the portion of the furnace atmosphere under investigation.

[0103] It will be evident that the absorption of light in the 589 nm region due to sodium atoms will also depend upon the measurement path length, i.e. essentially the distance in metres between the measuring point 28 and the sidewall 21 (shown as line l-l′ in FIG. 2). It is possible to correct the absorption intensity for the measurement path length by assuming the absorption follows the well-known Lambert-Beer Law. Although the Lambert-Beer Law only holds for homogeneous media, in the present method the correction for the measurement path length using the Lambert-Beer Law is acceptable for practical purposes.

[0104] It is possible to use a more complex analysis to correct for path length if necessary.

[0105] The intensity difference δI.sub.589nm corrected for the measurement path length is also proportional to the concentration of sodium atoms in the portion of the furnace atmosphere under investigation.

[0106] To obtain the concentration of sodium atoms in the portion of the furnace atmosphere under investigation the difference δI.sub.589nm is corrected for path length using the well-known Lambert-Law and the quantity so determined can be divided by the absorption coefficient for sodium atoms at 589 nm. A spectrometer calibrated absorption coefficient at 589 nm of 0.01 was used for sodium atoms.

[0107] The total sodium concentration in the furnace atmosphere includes contributions from sodium atoms and sodium hydroxide. The ratio of sodium to sodium hydroxide is temperature dependent and can be determined based on thermodynamic calculations. Such calculations are known to a person skilled in the art.

[0108] Given that the absorption at 589 nm is due only to the sodium atoms, to get an actual measurement of the total concentration of sodium species in the atmosphere it is necessary to determine the concentration of all sodium species. Practically, only sodium hydroxide was considered as another sodium species in the furnace atmosphere.

[0109] Using thermodynamic considerations it was possible to estimate the ratio of sodium to sodium hydroxide at a given temperature in gaseous form. Such calculations showed that the sodium hydroxide to sodium ratio varied between 20:1 at 1500° C. and 10:1 at 1600° C.

[0110] By assuming a spectrometer calibrated absorption coefficient at 589 nm of 0.01 for sodium atoms, and a ratio of sodium to sodium hydroxide of 0.05, it was possible to compare the total sodium concentration in the atmosphere determined using a method according to the present invention with a conventional extractive method.

[0111] FIG. 4 shows the total sodium concentration (in ppm) for three measurements, TEST A, TEST B and TEST C determined using a conventional extractive technique and a method according to the present invention. The average of the three tests is labelled “MEAN”. Axis 50 on FIG. 4 is the total sodium concentration in ppm.

[0112] A conventional extractive technique was used as follows. A sample was extracted from the furnace atmosphere in the vicinity of the measuring position 28 over a period of twenty minutes and the sodium content determined using well known wet chemical techniques. The results are shown as the dark columns in FIG. 4.

[0113] The spectrometer was used to collect wavelength dependent intensity data over each extraction period and subsequently averaged. For each TEST A, B, C, the intensity data at two wavelengths 572 nm and 582 nm was used to linearly interpolate a predicted intensity at 589 nm. The measured intensity at 589 nm was then subtracted from the measured intensity at 589 nm. This intensity difference δI.sub.589nm was then corrected for the measurement path length using the known position of the water cooled jacket 30 relative to the side wall 21.

[0114] To obtain a total sodium concentration in the portion of the furnace atmosphere under investigation, the intensity data corrected for measurement path length was then divided by the absorption coefficient for sodium atoms at 589 nm. A spectrometer calibrated absorption coefficient of 0.01 was used. Finally, it was assumed that the ratio of sodium to sodium hydroxide was 0.05 at the measurement temperature of around 1500° C.

[0115] As FIG. 4 shows, in view of the number of assumptions and simplifications, the variation in the total sodium concentration measured using the extractive technique (the darker columns in FIG. 4) and the method according to the present invention (the lighter columns in FIG. 4) varied by less than 1 ppm. Typically the difference was within 3 ppm of the total sodium content determined by the conventional extractive technique.

[0116] It will be readily apparent that the measuring position could be at any position within the furnace, such as in the burner sections 12, 14 or in the furnace roof The temperature at the other measuring positions may be different.

[0117] There may be more than one measuring point such that the variation of sodium in the furnace atmosphere at different locations in the furnace can be determined, each measurement being used to determine the sodium concentration in a portion of the furnace atmosphere.

[0118] In the present example a float glass making furnace was used, but the method is applicable to other glass making furnaces, such as those making rolled glass, glass for containers and glass sheets for substrates. Also the method is applicable to other furnaces where it is desirable to measure the sodium content of the furnace atmosphere. The method may also be used for other species that absorb electromagnetic radiation inside the furnace.