- Email: [email protected]
Effect of gas mass flux on cryogenic liquid met breakup
Effect of gas mass flux on cryogenic liquid jet breakup* R.D. Ingebo National Aeronautics and Space Administration, Lewis Research Center, Cleveland, OH 44135, USA A scattered-light scanning instrument developed at NASA Lewis Research Center was used to measure the characteristic drop size of clouds of liquid nitrogen droplets. The instrument was calibrated with suspensions of monosized polystyrene spheres. In this investigation of the mechanism of liquid nitrogen jet disintegration in a high-velocity gas flow, the Sauter mean diameter, D32, w a s found to vary inversely with the nitrogen gas mass flux raised to the power 1.33. Values of 032 varied from 5 to 25/zm and the mass flux exponent of 1.33 agrees well with theory for liquid jet breakup in high-velocity gas flows. The loss of very small droplets due to the high vaporization rate of liquid nitrogen was avoided by sampling the spray very close to the atomizer, i.e. 1.3 cm downstream of the nozzle orifice. The presence of high velocity and thermal gradients in the gas phase also made sampling of the particles difficult. As a result, it was necessary to correct the measurements for background noise produced by both highly turbulent gas flows and thermally induced density gradients in the gas pase.
Keywords: Cryogenic liquids; sprays; light-scattering instrument
Nomenclature Atomizer orifice area (cm 2) Diameter of ith drop (cm) D32 Sauter mean drop diameter, E niD~/Y,niD~ (cm) n Number of droplets V Fluid velocity (cm s -~) W Weight flow of fluid (g s -~)
Axial downstream spray sampling distance (cm) Density of fluid (g cm -3)
Ao
Di
Optical particle-sizing instruments are needed that will overcome the many difficult problems encountered in the study of cryogenic sprays. It is very difficult to obtain reproducible drop size data for very fine liquid nitrogen particles formed by the disintegration of cryogenic liqid jets in two-fluid atomizers. One of the most severe measurement problems is the loss of small drops due to their high vaporization rate. Therefore, cryogenic sprays should be sampled very close to the atomizer orifice. In a cryogenic liquid nitrogen spray, the drops are formed at a temperature very close to their boiling point, i.e. 77 K, which is low compared with the atomizing gas temperature of 295 K. This not only makes small drops evaporate very quickly but also the temperature gradient in the gas phase surrounding the drops presents further difficulties. Thermal gra-
*Paper presented at the 1991 Space Cryogenics Workshop, 1 8 - 2 0 June 1991, Cleveland, OH, USA
Subscripts c g
Acoustic Nitrogen gas
dients tend to scatter the laser beam owing to large variations in the refractive index and this produces background noise when light-scattering measurements are being made. To overcome these difficulties, the scattered-light scanner used earlier as an optical particle sizing instrument I to study water sprays was further developed at NASA Lewis Research Center. Improvements to the instrument are discussed in detail in a recent paper 2 and as a result it was possible in this study to measure characteristic drop sizes of liquid nitrogen sprays with the scattered-light scanner. Several investigators have made experimental measurements of the drop size of water and fuel sprays and correlated it with relative velocity, i.e. gas velocity relative to liquid velocity, and also with liquid properties 3 7. Some correlations agree and some do not agree very well with atomization theory. This is attributed to the fact that measurement techniques and drop sizing instruments have not yet been developed and standardized to the extent that good agreement
0011 - 2275/92/020191 - 0 3 ~? 1992 Butterworth Heinemann Ltd
Cryogenics 1992 Vol 32, No 2
191
Effect of gas mass flux on liquid jet breakup: R.D. Ingebo might be expected. In this study, the entire spray crosssection was sampled at a distance of • = 1.3 cm downstream of the atomizer orifice, whereas some investigators have used sampling distances of the order of 10-20 cm.
Apparatus and procedure A pneumatic two-fluid nozzle was mounted in the centre of the test section as shown in Figure 1, and a detailed diagram of it is shown in Figure 2. Two nozzles were tested, with orifice areas of 0.112 and 0.264 cm 2. At a temperature of 77 K measured with an iron-constantan thermocouple, liquid nitrogen (LN2) was axially injected into the gas stream at a flow rate of 29 g s -~, as indicated by a turbine flow meter. The atomizing gas, i.e. gaseous nitrogen at 298 K, was admitted into the outer tube and disintegrated the LN2 jet along the centre line of the atomizer. The nitrogen gas flow rate was varied over the range 1.3-4.5 g s -~ as measured with a 0.51 cm diameter sharp-edge orifice. The optical system of the scattered-light scanner is shown in Figure 1. The instrument measures scattered light as a function of scattering angle by repeatedly sweeping a variable-length slit in the focal plane of the collecting lens. The data obtained are scattered-light energy as a function of scattering angle, relative to the laser beam axis. This method of particle size measure-
ment is similar to that by Swithenbank et al. 8 and it was described in detail by Buchele 9. Calibration was accomplished with five sets of monosized polystyrene spheres having diameters of 8, 12, 25, 50 and 100/zm. Since the sprays were sampled very close to the atomizer orifice, they contained a relatively large number of very small drops. As a result, the light-scattering measurements required correction for multiple scattering as described 9,a° for the case of high concentrations of droplets. Drop size measurements were also corrected as described 9 to include Mie scattering theory when very small characteristic droplet diameters, i.e. < 10 #m, were measured.
Experimental results Two pneumatic two-fluid atomizers were used with liquid and gaseous nitrogen to determine the effect of nitrogen gas flow rate on the characteristic drop size of a cryogenic spray. Also, the effect of gas mass flux on the characteristic drop size was investigated to determine a correlating expression for the two nozzles.
Effect of nitrogen gas flow rate on characteristic drop size The entire spray cross-section was sampled at an axial distance of:e = 1.3 cm, where • is the distance from the
4000 AIR
l LIOLIIDNITROGEN~
~
NITROGENGAS~
ATOMIZER,DO = 0.sGq CM F LIC~4T-SCATTER I N/~G
0 01 c~ DIARETER S.O c~ DIk~qETER CM x. a'l I~ LAiR
U
~
/
SPRAY '
" //
~
.~ 7.5 c,q DI,eOIETEA
\] ~
II
I,.- 7.5 CMDIARETER
L'PHOTO
/yo/
t-.
'5 v
2000
c~
1000
Figure 1 Atmospheric pressure test section and optical path of scattered-light scanner
INSIDE Dl ARETER 0.564 c~--,,\
NI TROGEN 6AS - ~ \ \
"-..
,
--
800
60O
",,
IX\I
40O
,,
,
L LIOUID NITROGEN
, .ozz,E o ,F,cE DINqETER
O.663 c~ - /
Figure 2
Diagram of pneumatic two-fluid atomizer
192 Cryogenics 1992 Vol 32, No 2
/
i
I
I
I
5
6
7
8
i
I
9 10 Wg/Ao (gcN-2S "1)
20
Figure 3 Variation of reciprocal of Sauter mean diameter (D3-21) with gas flow rate per unit area (Wg/Ao). NozzleAo: O, M1, 0.112 cm2; n , No. 4, 0.264 cm 2
Effect of gas mass flux on liquid jet breakup: R.D. Ingebo
atomizer orifice to the centre line of the laser beam. The liquid nitrogen flow rate was held constant at 29.0 g s -~. The characteristic drop diameter or Sauter mean diameter, D32, was measured for the two atomizers and plotted against nitrogen gas flow rate, Wg. The 33following relationship was obtained: D~ ~ - Wg at a downstream distance of ,~ = 1.3 cm. This expression agrees well with that given by atomization theory for liquid jet disintegration in the regime of aerodynamic stripping, i.e. in high-velocity gas streams i i. Correlation of nitrogen gas mass flux with
D32
The reciprocal of the Sauter mean diameter, D3~ 1, is plotted against atomizing gas flow rate per unit area, Wg/Ao, in Figure 3. From this plot, the following relationship is obtained: D3~ l = 59(Wg/Ao)133, where 59 is the correlating coefficient for the two nozzles. Also, it should be noted that Wg/Ao is equal to the gas mass flux, p~ V~, where p~ and Vc are gas density and acoustic velocity, respectively. Concluding remarks The fluid mechanics of liquid nitrogen jet breakup is much more difficult to study than the atomization of fuel or water jets, primarily because the surface temperatures of the liquid nitrogen jets used in this study were always near the boiling point of liquid nitrogen, i.e. = 77 K. Since the atomizing gases were at room temperature, 293 K, this created large temperature gradients that deflected the laser beam and caused beam steering to occur as characteristic drop diameter was measured with the scattered-light scanner. This problem was minimized
by moving the laser beam away from the slit in the scattered-light scanner optical system.
References lngebo, R.D. Experimental and theoretical effects of nitrogen gas flowrate on liquid jet atomization J Propulsion Power (1988) 4 406-411 lngebo, R.D. and Buehele, D. R. Scattered-light Scanner Measurements of Cryogenic Liquid-Jet Breakup NASA TM-102432 (1990) Kim, K.Y. and Marshall, W.R., J r Drop size distributions from pneumatic atomizers A1ChE J (1971) 17 575-584 Lorenzetto, G.E. and Lefebvre, A.H. Measurements of drop size on a plain-jet airblast atomizer AIAA J (1977) 15 1006-1010 Nukiyama, S. and Tanasawa, Y. Experiments on the atomization of liquids by means of an air stream, parts I l l - I V Trans Soc Mech Eng Jpn (1939) 5 (18) 6 3 - 7 5 Weiss, M.A. and Worsham, C.H. Atomization in high velocity airstreams Am Rocket Soc J (1959) 29 252-259 Wolf, H.E. and Andersen, W.H. Aerodynamic break-up of liquid drops, in: Proceedings of the 5th International Shock Tube Symposium (Eds Slawasky, Z.I., Moulton, J.F., Jr and Filler, W.S.) Naval Ordance Lab., White Oak, MD, USA (1965) 1145-1169 (available from NTIS, Ad-638011) Swithenbank, J., Beer, J.M., Taylor, D.S., Abbot, D. and McCreath, G.C. A laser diagnostic technique for the measurement of droplet and particle size distribution, in: Experimental Diagnostics in Gas Phase Combustion Systems (Eds Zinn, B.T. and Boulman, G.T.), Progress in Astronautics and Aeronautics Vol 53, AIAA, New York, USA (1977) 421-447 9 Buchele, D.R. Particle Sizing by Weighted Measurements of Scattered Light NASA TM- 100968 10 Felton, P.G., Hamidi, A.A. and Aigail, A.K. Measurement of drop size distribution in dense sprays by laser diffraction, in: ICLASS-85; Proceedings of the Third International Conference on Liquid Atomisation and Spray Systems Vol 2 (Eds Eisenklam, P. and Yule, A.) Institute of Energy, London, UK (1985) IVA/4/1-IVA/4/I 1, 11 Adelberg, M. Mean drop size resulting from the injection of a liquid jet into a high-speed gas stream A1AA J (1968) 6 1143-1147
Cryogenics 1992 Vol 32, No 2
193
Keywords: Cryogenic liquids; sprays; light-scattering instrument
Nomenclature Atomizer orifice area (cm 2) Diameter of ith drop (cm) D32 Sauter mean drop diameter, E niD~/Y,niD~ (cm) n Number of droplets V Fluid velocity (cm s -~) W Weight flow of fluid (g s -~)
Axial downstream spray sampling distance (cm) Density of fluid (g cm -3)
Ao
Di
Optical particle-sizing instruments are needed that will overcome the many difficult problems encountered in the study of cryogenic sprays. It is very difficult to obtain reproducible drop size data for very fine liquid nitrogen particles formed by the disintegration of cryogenic liqid jets in two-fluid atomizers. One of the most severe measurement problems is the loss of small drops due to their high vaporization rate. Therefore, cryogenic sprays should be sampled very close to the atomizer orifice. In a cryogenic liquid nitrogen spray, the drops are formed at a temperature very close to their boiling point, i.e. 77 K, which is low compared with the atomizing gas temperature of 295 K. This not only makes small drops evaporate very quickly but also the temperature gradient in the gas phase surrounding the drops presents further difficulties. Thermal gra-
*Paper presented at the 1991 Space Cryogenics Workshop, 1 8 - 2 0 June 1991, Cleveland, OH, USA
Subscripts c g
Acoustic Nitrogen gas
dients tend to scatter the laser beam owing to large variations in the refractive index and this produces background noise when light-scattering measurements are being made. To overcome these difficulties, the scattered-light scanner used earlier as an optical particle sizing instrument I to study water sprays was further developed at NASA Lewis Research Center. Improvements to the instrument are discussed in detail in a recent paper 2 and as a result it was possible in this study to measure characteristic drop sizes of liquid nitrogen sprays with the scattered-light scanner. Several investigators have made experimental measurements of the drop size of water and fuel sprays and correlated it with relative velocity, i.e. gas velocity relative to liquid velocity, and also with liquid properties 3 7. Some correlations agree and some do not agree very well with atomization theory. This is attributed to the fact that measurement techniques and drop sizing instruments have not yet been developed and standardized to the extent that good agreement
0011 - 2275/92/020191 - 0 3 ~? 1992 Butterworth Heinemann Ltd
Cryogenics 1992 Vol 32, No 2
191
Effect of gas mass flux on liquid jet breakup: R.D. Ingebo might be expected. In this study, the entire spray crosssection was sampled at a distance of • = 1.3 cm downstream of the atomizer orifice, whereas some investigators have used sampling distances of the order of 10-20 cm.
Apparatus and procedure A pneumatic two-fluid nozzle was mounted in the centre of the test section as shown in Figure 1, and a detailed diagram of it is shown in Figure 2. Two nozzles were tested, with orifice areas of 0.112 and 0.264 cm 2. At a temperature of 77 K measured with an iron-constantan thermocouple, liquid nitrogen (LN2) was axially injected into the gas stream at a flow rate of 29 g s -~, as indicated by a turbine flow meter. The atomizing gas, i.e. gaseous nitrogen at 298 K, was admitted into the outer tube and disintegrated the LN2 jet along the centre line of the atomizer. The nitrogen gas flow rate was varied over the range 1.3-4.5 g s -~ as measured with a 0.51 cm diameter sharp-edge orifice. The optical system of the scattered-light scanner is shown in Figure 1. The instrument measures scattered light as a function of scattering angle by repeatedly sweeping a variable-length slit in the focal plane of the collecting lens. The data obtained are scattered-light energy as a function of scattering angle, relative to the laser beam axis. This method of particle size measure-
ment is similar to that by Swithenbank et al. 8 and it was described in detail by Buchele 9. Calibration was accomplished with five sets of monosized polystyrene spheres having diameters of 8, 12, 25, 50 and 100/zm. Since the sprays were sampled very close to the atomizer orifice, they contained a relatively large number of very small drops. As a result, the light-scattering measurements required correction for multiple scattering as described 9,a° for the case of high concentrations of droplets. Drop size measurements were also corrected as described 9 to include Mie scattering theory when very small characteristic droplet diameters, i.e. < 10 #m, were measured.
Experimental results Two pneumatic two-fluid atomizers were used with liquid and gaseous nitrogen to determine the effect of nitrogen gas flow rate on the characteristic drop size of a cryogenic spray. Also, the effect of gas mass flux on the characteristic drop size was investigated to determine a correlating expression for the two nozzles.
Effect of nitrogen gas flow rate on characteristic drop size The entire spray cross-section was sampled at an axial distance of:e = 1.3 cm, where • is the distance from the
4000 AIR
l LIOLIIDNITROGEN~
~
NITROGENGAS~
ATOMIZER,DO = 0.sGq CM F LIC~4T-SCATTER I N/~G
0 01 c~ DIARETER S.O c~ DIk~qETER CM x. a'l I~ LAiR
U
~
/
SPRAY '
" //
~
.~ 7.5 c,q DI,eOIETEA
\] ~
II
I,.- 7.5 CMDIARETER
L'PHOTO
/yo/
t-.
'5 v
2000
c~
1000
Figure 1 Atmospheric pressure test section and optical path of scattered-light scanner
INSIDE Dl ARETER 0.564 c~--,,\
NI TROGEN 6AS - ~ \ \
"-..
,
--
800
60O
",,
IX\I
40O
,,
,
L LIOUID NITROGEN
, .ozz,E o ,F,cE DINqETER
O.663 c~ - /
Figure 2
Diagram of pneumatic two-fluid atomizer
192 Cryogenics 1992 Vol 32, No 2
/
i
I
I
I
5
6
7
8
i
I
9 10 Wg/Ao (gcN-2S "1)
20
Figure 3 Variation of reciprocal of Sauter mean diameter (D3-21) with gas flow rate per unit area (Wg/Ao). NozzleAo: O, M1, 0.112 cm2; n , No. 4, 0.264 cm 2
Effect of gas mass flux on liquid jet breakup: R.D. Ingebo
atomizer orifice to the centre line of the laser beam. The liquid nitrogen flow rate was held constant at 29.0 g s -~. The characteristic drop diameter or Sauter mean diameter, D32, was measured for the two atomizers and plotted against nitrogen gas flow rate, Wg. The 33following relationship was obtained: D~ ~ - Wg at a downstream distance of ,~ = 1.3 cm. This expression agrees well with that given by atomization theory for liquid jet disintegration in the regime of aerodynamic stripping, i.e. in high-velocity gas streams i i. Correlation of nitrogen gas mass flux with
D32
The reciprocal of the Sauter mean diameter, D3~ 1, is plotted against atomizing gas flow rate per unit area, Wg/Ao, in Figure 3. From this plot, the following relationship is obtained: D3~ l = 59(Wg/Ao)133, where 59 is the correlating coefficient for the two nozzles. Also, it should be noted that Wg/Ao is equal to the gas mass flux, p~ V~, where p~ and Vc are gas density and acoustic velocity, respectively. Concluding remarks The fluid mechanics of liquid nitrogen jet breakup is much more difficult to study than the atomization of fuel or water jets, primarily because the surface temperatures of the liquid nitrogen jets used in this study were always near the boiling point of liquid nitrogen, i.e. = 77 K. Since the atomizing gases were at room temperature, 293 K, this created large temperature gradients that deflected the laser beam and caused beam steering to occur as characteristic drop diameter was measured with the scattered-light scanner. This problem was minimized
by moving the laser beam away from the slit in the scattered-light scanner optical system.
References lngebo, R.D. Experimental and theoretical effects of nitrogen gas flowrate on liquid jet atomization J Propulsion Power (1988) 4 406-411 lngebo, R.D. and Buehele, D. R. Scattered-light Scanner Measurements of Cryogenic Liquid-Jet Breakup NASA TM-102432 (1990) Kim, K.Y. and Marshall, W.R., J r Drop size distributions from pneumatic atomizers A1ChE J (1971) 17 575-584 Lorenzetto, G.E. and Lefebvre, A.H. Measurements of drop size on a plain-jet airblast atomizer AIAA J (1977) 15 1006-1010 Nukiyama, S. and Tanasawa, Y. Experiments on the atomization of liquids by means of an air stream, parts I l l - I V Trans Soc Mech Eng Jpn (1939) 5 (18) 6 3 - 7 5 Weiss, M.A. and Worsham, C.H. Atomization in high velocity airstreams Am Rocket Soc J (1959) 29 252-259 Wolf, H.E. and Andersen, W.H. Aerodynamic break-up of liquid drops, in: Proceedings of the 5th International Shock Tube Symposium (Eds Slawasky, Z.I., Moulton, J.F., Jr and Filler, W.S.) Naval Ordance Lab., White Oak, MD, USA (1965) 1145-1169 (available from NTIS, Ad-638011) Swithenbank, J., Beer, J.M., Taylor, D.S., Abbot, D. and McCreath, G.C. A laser diagnostic technique for the measurement of droplet and particle size distribution, in: Experimental Diagnostics in Gas Phase Combustion Systems (Eds Zinn, B.T. and Boulman, G.T.), Progress in Astronautics and Aeronautics Vol 53, AIAA, New York, USA (1977) 421-447 9 Buchele, D.R. Particle Sizing by Weighted Measurements of Scattered Light NASA TM- 100968 10 Felton, P.G., Hamidi, A.A. and Aigail, A.K. Measurement of drop size distribution in dense sprays by laser diffraction, in: ICLASS-85; Proceedings of the Third International Conference on Liquid Atomisation and Spray Systems Vol 2 (Eds Eisenklam, P. and Yule, A.) Institute of Energy, London, UK (1985) IVA/4/1-IVA/4/I 1, 11 Adelberg, M. Mean drop size resulting from the injection of a liquid jet into a high-speed gas stream A1AA J (1968) 6 1143-1147
Cryogenics 1992 Vol 32, No 2
193