The Solubility of Oxygen and Ozone in Liquids Rubin Battino, Timothy R. Rettich,” and Toshihiro Tominaga” Department of Chemistry, Wright State University, Dayton, Ohio 45435, ‘This review covers the solubility ofoxygen and ozonein liquids asa function of tempera ture and pressure. Solubility data for individual eyctems were critically evaluated and recommended or tentative values presented in many cases. The trend of solubilities in homologous series or related solvents is discussed. Liquids include water; seawater; aqueous salt solutions; mixed solvents; hydrocarbons; organic compounds containing ‘oxygen, halogen, sulfur, nitrogen, or silicon; olive oil; and human blood. For ozone, only its solubility in water is presented, ‘ey words: aqueous stutions; pioogical mas; gs snuou ‘ganic solvents; oxygen; ozone; seawater; wate. Contents. 1. Introduction 164 4.8. Oxygen in Organic Compounds Containing 2. Methods of Measurement 164 Sulfur, Nitrogen, or Silicon 3. Treatment of the Solubility Data 165 49. Biological Fluids. 3.1. Solubility Units. : 165 49.2, Olive Oil 32. Reliability Estimates 165 49.8. Human Blood 33. Fitting Equations. 165 4.10. Miscellaneous Fluids 34. Thermodynamic Functions. 1635. Oume Solubility Data 4. Oxygen Solubility Data up to 200 KP: 165 5.1. Ozonein Water. 4.1, Oxygen in Water 165 6 Concluding Observations. 4.L.a. Oxygen/Water ae a Recommended 7. Acknowledgments Reference SYSteM vvnvennneninnone 167 8, Reference. 4.2. Oxygen in Seawater... 167 43. Oxygen in Aqueous Salt Solutions... 167 43.a. Acids. 168 43.b, Base8 ons 168 4.3.c. Group TA Cations. 169 4.3.0. Group 1A Cations. 199 43.e. Other Cations 170 44, Oxygen in Organic Solvents Plus Water... 170 45. Oxygen in Hydrocerbone 170 4S, Saturated 170 45b. Cyclic. im Soc, ArOMBtIC ne m 46, Onygenin Orie Compounds ‘Containing Onygen : m 4.6.2. Alcohols. m ob. Ketones, Ethers, Acids, Esters nc 173 47. Oxygen in Organic Compomnds Containing Halogen 114 Chemiiry Departs, inols Wekyen University, Bloomington, Mi On ave fom Okayama University of Seon, Okayams, Japan. © 1983by the US. Secretary of Commerc on behalf ofthe United Stats. ‘hie copyright is asgned fo the American Ititute of Physics andthe Reprints avallale rom ACS; see Repeat List at back of ise (047-2699/29/020169-16/45.00 163 List of Tables Solubility of oxygen in water forthe temperature range 273.15-348.15 K and partial pressure of gas of 101.325 kPa.. 2. Thermodynamic functions for diaslving Gaygese at 101.325 kPa pressure in water in the tempera- ture range 273.15-348.15K.. Solubility f oxygen in wate forthe temperate range 348.15-573.15 K and partial pressure of 101.325 KPa rn Coefficients for Eq. (19) for several solubility units for oxygen in seawater. Values of Cin mL- oxygen (STP) per liter seawater for selected temperatures and salinities. Sctsclenow sll lft parameters for hydioclo~ ric and sulfuric aC Setschenow salt effect parameters for 1.0 rol L!niteie acid at 283 15-993.15 K . Setschenow salt effect parameters for various al- kali metal hydroxides at 298.15 K Values fh for alkali metal halidesat 248.13 K Values of Ky for alkali metal salts at 248.15 K. vs 176 176 176 176 176 176 171 7 7 166 166 16 167 161 168 ws 168 169 169 4. Phys. Chem. Rot. Data, Vol. 12, No.2, 1963164 BATTINO, RETTICH, AND TOMINAGA 11, Values for salt effect parameters in Eq. (21) for "* 21, Themolefraction solubility ofoxygeninfiveoxy- Group HA metal salts... 169 gen containing solvents at 101.325 kPa partial 12, Valuesfk. for selected sulfate salts a 248423 pressure of gas (10'x,) m4 170 22. Mole fraction solubility of oxygen in organic 13, Mole faction abso oxygen in lal? compounds containing halogen at 101.325 kPa ater minutes at 101.325kPs partial pressure of partial pressure of gas (10*r,) i oxygen 170 23. Mole fraction solubility of oxygen in organi 14. Mole fraction solubility of oxygen in saturated compounds containing halogen at 298.15 K and hydrocarbons at 101.325 kPa partial pressure of 101.325 kPa partial pressure of ga 175 ne suienine 170 24Mole fraction solubility of oxygen in Uiganic lity of oxygen in cyclic hy- ‘compounds containing sulfur, nitrogen, or silicon = = ‘two standard deviation from a smooth curve. The 149 se- a8 as lected data points in the remaining 15 papers were then 2. an smoothed by Fa. (5) sus B08 Tn addition to values for the mole fraction, the Ostwald 93 =o cocficientsand Cvaluesare shown in Table . Thesolubility ss mat term Cis defined by the equation vo. a2 1 {eV 4.689% 10% ED an Cis = at Ho) me = ‘where the mole fraction x, fs at 101.325 Pa partial pressure ae ‘ofoxygen, 7 isthemolar volume of pure waterattempera- 3.5 aan ture 7, and the ratio of oxygen in air is assumed to be ns >a 20.94%. Thus Cis the number of milters of gas at STP 3s.» ao an from ait dissolved in one liter of solvent. Table 1 also in-——yo.5 a on cludes a8 a convenience the values forthe molar volume of ys as on pure water at selected temperatures. These values are often ‘used to caleulate solubility in unite other than those already tabulated, and were obtained from Kell's equation for the density of pure water.* For temperatures below 348 K, the ‘iting equations are Inx, = — 6421517 4 83,91236/r-4.23.24323 Inf, (11) In L = ~58.51653 + 84.5991/r-+ 2441285 Info, (12) In C= — 5216764 + 84,59929/r 4 2341230107}. (13) The 15 sources for the solubility data used in Table 1 are ‘Refs. 17, and 19-32, The data used from each are shown ‘below by (reference number—number of data points used: we ee | aan ay "ti i 9, etn tte 0 5) cate 1 4. Phys. Chem. Rot. Data, Vol 12, No 2, 1983 (1721, (1937, (20~26, @2i—14), 22-2}, 2315}, (4,059) O14, OFAN, ORM, O81) CA) 1-3, 62-2) iting all 149 points by Bq. (5) gave a stanaara devi ation of + 0.36%. Considering the variety of both chemical ‘and physical techniques used fo measure te solubility, and the numberof datapoints, this ft shows an amazing consi tency. The recommended values in Table are believed tobe accurate about 0.3%. (See next seston From the codfiients dy Ay, and Ay in Ea, (It), the thermodynamic functions can be calculated according t0 Eqs. (6)-(9}. Note that for a three term fit, the value for AC’; isconstant at 203 J mol" K~ "ovr this temperature range within the precision ofthe data. The Gibbs energy, enthalpy, and entropy changes are listed in Table 2 “Thesolublty ofoxygea in water at emperatures above 348 K were investigated by Stephan et al.” and Pray et al.™* ‘Their data were fitted by Clever and Han,” together with lower temperature data to yield Inxj = ~ 54.0411 + 68.8961/r-4 18.554 In(rh (14) te 3 Ao ape at oe tere as 5 = = ps pow a‘SOLUBILITY OF OXYGEN AND OZONE IN LIQUIDS 167 ‘Values for the mole fraction of dissolved oxygen at selected temperatures are shown in Table 3. The precision of these values is poorer than those below 348 K, and is estimated to be + 2.0%. 4.1.4. Onygen/Water asa Recommended Reference System ‘Werecommend the system of oxygen dissolved in water asa reference system since it s the most commonly mea- Sured and most precisely known gas solubility. While the data in Table {summarize th bes avalablesolubity values derived from many sources and thus are appropriate for gen- tral use recent work shows much better precision thn the 3.3% woovited with Table 1. Therefore new methods with high precision should use the smoothed values fom Benson etal.” for comparison. They proposed the following iting in hota Pana AK -lnx, T1814 + 5.59617 10°/(T /K) = 1.049668 10°17 7K? (13) ‘or when the data are retitted by 4g. (> In.xy = — 4.21517 + 83.91236/r + 23.24323 In(r). (16) ‘The standard deviation from the fitting equation is 40.0179, which is by far the most precise measurement of | gas solubility to date. Benson et af" determined a mole fraction x, of 2,301 10" at 298.19 K and 101.325 kPa partial pressure ‘of oxygen. This value is 0.35% higher tha that given in ‘Table 1, but the historical trend has been towards higher ‘mole fraction values. Since this is the most precise and probably mast accurate gas solubility measurement avail- able, all future tests of methods should include values for the system of oxygen in water for comparison. Smoothed values from Eq. (15) are given in Table 1. Mortimer” recently presented a critical review of oxy- ‘gen solubilities in fresh waters with many tables, nomo- _Eraphs, etc. His recommendations are based on the Work of ‘Benson etal.” 4.2. Oxygen in Seawater ‘The solubility of oxygen in seawater has been studied neatly #8 widely #9 the solubility in pune wates, but wil less ‘agreement among the data provided by different research- ‘ers. The most reliable data appear to be those from Carpen- or and Murray and Riley, 28 with moaencomante hatuinen 0 and 36°C and 0% to 40% salinity. vane) sai srs ease} can | sus] st sc fo ‘Thedificulties with seawater measurements are both in analysis and interpretation. The measurement of oxygen in seawater was often complicated due to the loss of iodine and Interferences within the chemical methods ot analysis. Kor this reason, much ofthe early work was excluded. The multi- tude of components in seawater also precluded a treatment of the data via Sotchenow's equation (whisk follows) for oon, centration effects for each ionic species: wl ‘5, and S} are the solubilities and y, and y} are the activity coefficients of the gas in the salt solution and pure solvent, respectively. C, isthe electrolyte concentration and i, is the salt effect parameter. ‘As a simplification, Weiss” proposed the following equation: In C= Ay + Aa/r + Asn} tar + SB + Bar + Br], (18) where solubility Cis given by Ea, (10) and S is salinity in parts per thousand. The 4 terms in Eq (18) resemble the normal Weiss Eq, (5), while the additional B terms fit the sauinty dependence. in teir analysis of oxygen soluoiity in seawater, Chen and Carpenter" showed that Eq. (19) was superior to Eq, (18) Im C= Ay + Ay/(E PR) + AstO{L 78) + AAT /R) + S[B, + By/T /K)] + ByS(19) ‘Theconcentration term Ccan have the different unitsshown in Table 4, with the corresponding coefficients. ‘The values generated from Eq, (19) withthe coeficients in Table 4 (sce Table 5) are valid only between 9 and 35°C, ‘below 400 ppt salinity and relatively low pressures. Solubi- lity studies with higher salinity (marine brines) by Kinsman etal,” are imprecise and unconfirmed and thus are not in- 4.3. Oxygen in Aqueous Salt Solutions The “saltingin” or “salting-out” of a nonelectrolyte soluteis a common phenomenon. One way of expressing the effect ofa dissolved electrolyte on the solubility of oxygen in waters the Setschenow salt parameter k. Ifthe concentr a wenn [acme | mie eas 5 zosm | race | za.s6 5 cesim | -esmser | easter al sam | Soe | ate % aww | gent | need tion ofthe electrolyte isc (mol L~") andthe gas solubility is 4: Phys Chem. Ret Data, VoL 12, No.2, 1085168 BATTINO, RETTICH, AND TOMINAGA expressed a the Bunsen coeficient in pure water and in ‘he electrolyte sotuion, te sai ener parameter is Koce = (I/e3 ope (20) ‘Other units of concentration and ga sclubilty will result in ‘equivatent salt ettect parameters ‘The Setschenow equation (20) is an approximation ‘based upon a large solbilty diference between the electro Iyteand nnnelecrolyte solute. This pond approximation or oxygen dissolved in water, Plosoflogaal versus, ate ‘often linear over a modest range of electrolyte coneenira~ tions. However at high concentrations of eletroyte, these plors can become highly curve, Except fr thehigh pressure studiesby Bruhn, Gerlach, ‘and Pawlck, the solubility of oxygen was determined at presuaree near 101 395Pa, Whenever possible the india ‘ual author's value for a” is used instead of those suggested calier inthis work in the hope that the rato (a°%a} might cancel out any systematic errors ‘ne electroiytes are discussed in the following groups: acids, bases, Group IA salts, Group IIA salts, and others. ‘Due to the scatter observed in these data, and the frequent Jack of any similar stvioa. the values given in thie set should be considered only as tentative aa Aatde ‘Theacids considered are hydrochloric, sulfuric, and ni- ti. The values for the Setschenow sat eect parameter for the first two acids are shown in Table 6 and are from Gefiskon The eal ffoct parameter appears tn Ancreane ‘with temperature and also with increasing sulfuric acid con centration. Gefcken, who investigated concentrations fom (02 t0 2.6 mol L~, noted only a slight decrease, but data from Bob show more pronounced flo of kz, for con ‘centrations ranging upto 18 mol L~”. ‘The solubility of oxygen in .0 mol L" aqueous suf ric acid was measured over a wide temperature range by ‘Bruha, Gerlach, and Pavlek.*" Values for Ky remained ‘early constant at 0.10 L mol" for temperatures between 323 and 523K. ‘Wile acid isaease Uf spec Imeret. VALUES Og OE 41,0m0l L~solutionare shown n Table 7. The most exten- sive study was by Pogrebnays, Usov, and Baranov™ who ‘measured oxygen solubility from 283 o 333 K for concen- ‘rations of nitric aid from 0 to 12 mol L~?. For the lower {temperature and concentration regions, nitric acid behaves like hydrochlori and sulfuric acid, salting out oxyzen. But ‘athigher concentrations at the same teaspeatus, allay in fet is observed. At 323 K, the salting-in phenomenon ‘occurs above 10 mol L~', and at 333 K, it occurs above 8 ‘mol L~!, At mole factions of nitric aid below 0.16, the 4: Phys. Chem. Ret. Data, Vo 12,Mo. 2, 1963 solubility of oxygen decreases with increasing temperature. ‘Athigher mole fractions, oxygen becomes mote soluble with Increasing temperature. 430. ares Strong bases such as Group TA hydroxides have apro- ‘nounced effect on the solubility of oxygen. Khomutoy and ‘Konnik examined allthealkalimetal hydrorides and the results ae summarized in Table 8. Te two results fr lith- jum hydroride reflect two diferent linear salting-out re- sions, the lower concentration one being proportionaly more effective-Recent work by Broden and Simonson" show a similar pattern for sodium hydroxide. A plot of opia?/a| versus, between 001 and 0.10 mol Lis linear ‘at with an anomalously high slope of 0.68 L mol" At Iigher concentrations of sodium hydroxide, Gefcken* and Bruhn, Gerlach, and Pawlek, calculate slopes near 0.180in agreement with Khomatow and Komi: The values ony computed by Yasunishi* are somewhat lower. Khomutoy and Konnik’s value for potassium hydrox- ide is supported by the results of Gelfcken Shoor et al*® ‘and Roaster and Ape'baum, "with ao evidence yet found formoce than one slope. The valu off, proposed by Mac- Asthur" is lower than all the others and can be rejected. Khamntav ana Konnile® paper i the only ghlchod em search for rubidium and cesium hydroxide sal eects, ‘Asexpected,a weak base such as ammonium hydroxide has a very small sat effect parameter, nearly constant at ‘0.006 L-mol despite varying coneentratio from 0 to 8 wea‘SOLUBILITY OF OXYGEN AND OZONE IN LIQUIDS 169 ‘43.0. Group 1A Cations ‘With the single exception of high concentrations of lth- {um fuoride, al alkali metal halides act to ealt-out dissolved oxygen. As previously noted with lithium hydroxide, two distinct regions of linear salt effects exist for lithium halides except lithium iodide. The valves for kx» and the related concentration ranges are shown in Table 9. Results from several investigations are shown for sodium and potassium chloride. The general trend for the alkali metal halides is toward a smaller salt ellect parameter when increasing the size of either the cation or anion. ‘Other common salts of the alkali metals are listed in ‘Table 10. Lithium nitrate all-in oxygen at higher concen: trations, and along with lithium perchlorate shows two li car regions of salt effects. The multiply charged sulfate, phosnhate. and carbonate have a hivher ionic strength and show larger salt effect parameters than do perchlorate and nitrate. A report of oxygen solubility in so- dium acetate solution of varying pH was made by Guseva et al, but without suficient information to calculate ky. ‘The higher temperature and pressure study by Broden and Simonson shows two regions for sodium bicarbonate salt “fect, At 329 KC and pressures about 1 MPa higg Was extial t0.03 L mol” for concentrations up to 0.1 mol L~?, but uog Was equal to 0.14 L mol~" at higher concentrations. The Broden and Simonson values should be considered tentative. ‘Mixtures of electrolytes including Group IA cations have been studied to avery limited extent. Murray, Riley, ‘and Wilson”? observed a salting-out effect for a mixture of sodium iodide and sodium hydroxide equivalent to 0.180 Lmol~" of total electrolyte concentration, whichis the tena- tive value for sodium hydroxide italf. A similar result was ‘observed for sodium sulfate and sulfuric acid by Kobe and fe ae a gg fat aah ne tts scree em aeaite ecthcaeasataS soot ate meme SER ne 0 Ye Hg i a nt te a EY oa eee eee cont A) | on Kenton® and for sodium sulfate, sodium dithionite, and so- «dium hydroxide in solution by Jhavert and Sharma ** 4.24. Group HA Cations Five salts of alkaline earth metals have been examined for their effect on the solubility of dissolved oxygen. They are _magesium chloride, calcium chloride, barium chloride, mag nesium sulfate, and calcium nitrate. At 298.15 K and con- centrations below about 2 mol L~?, Yasunishi*® found the Kua Values to be 0.200, 0.273, 0.204, 0.194, and 0.270 Lmol™", respectively. When results from concentrations ranging from 2to4 mol L~' were included inthe data base, a better fit was achieved for magnesium chloride, magnesium, sullate, and caletum chloride using the equation: te (e)"(2) em Speci vs regen a Table 1 Seo Siwy te we fw (BT md eH, we Fe 4 Phys. Chem. Ret. Data, Vol. 12, No.2, 1983170 430. Other Cations Bruhn, Gerlach, and Pawlek"® reported salt effect pa rameters for various salts at elevated temperature and pres- sure. The results are summarized in Table 12, Alltheelectro- Iytes salt-out oxygen, with lite effect of temperature. ‘Morrison and Johnstone" have shown that tetraethyl ‘ammonium bromide has a mild salting-n effect, with k,., ‘equal to — 0.04 L mol~'. As expected, the highly charged electrolyte AlySO,); has a large salt effect parameter equal 100/99 L mol". tne same stuay reports tnat concentra tions of aluminum chloride up to 1.7 mol L~? can be fitted by Ea, (20) to yield keg of 0.274 L mol~'. But including concentrations up to 2.2 mol L~, Eq, (21) provides a better fit with K°,.4 = 0.303 and k's, = 0.0859. Murray, Riley, ‘and Wilson? have shown that manganese chloride and man- anese sulfate have salt effect parameters of 0.202 and 0.194 L mol", respectively Colloidal solutions of hydrated iron oxide,” uranyl u- ride,” and uranyl sulfate’ solutions have been used in oxygen solubility tests, but the salt effect parameters were not obtained. Mixed solutions of uranyl sulfate, sulfuric acid, and copper sulfate” showed no obvious salting-out ef- fects, but this may be duc tothe low concentrations used in the study. a 4.4. Oxygen in Organic Solvents Plus Water The solubility of oxygen in mixtures of water with polar ‘organic solvents (usually alcohols) has been determined by a 4. Phys. Chem. Ref. Data, Vol. 12, No.2, 1985, BATTINO, RETTICH, AND TOMINAGA number of investigators. Tokunaga” has determined oxygen solubilities in methanol/water, ethanol/water, I-propanol/ ‘water, and 2-propanol/water mixtures over the whole com- position range and at 273.15, 293.15, and 313.15K. Cargill™ ‘determined oxygen solubilities in ethanol/water and t-bu- tanol/water mixtures over the whole composition range and in the temperature range of 277.3 to 3344 K. Cargill paid particular attention to solubilities at low concentrations of alcohols. We consider Cargill's data to be the most reliable For the ethan water mintuicy and ploveit uu calculate values of the solubilities in terms of the mole fraction at 101.325 kPa partial pressure of oxygen in Table 13. The ex iensve data reported by Toleunaga may be fonnd in Refi. 1 m5. 4.5. Oxygen in Hydrocarbons 45a, saturate Hyarocarbons Smoothed values of mole fraction solubility for n-pen- tan, n-hexane, n-heptane, n-octane, n-nonane, n-decane, ana’2,zMtrimetnyipentane are given im 1abie 14. AKO shown are the coeficients Ay and A, of Eq, () the standard deviation 9, andthe numberof datapoints used. In the case ‘of the normal hydrocarbons, the male fraction solubilities are a very weak function ofthe temperature (4H1j=0). Al- ‘though the coefcient 4, of Eq (5}is large for n-hexane, n- heptane and n-nonane, these are considered to be artical ‘because of the narrow temperature range studied and the experimental uncertainty. The dependence of the mole frac- tion solubility on the solvent chain ength is not clear. From sundecane to noctadecane, the Makranczy ea. values” decrease as a function of hydrocarbon chain length, while ‘the Blanc and Batiste values show the opposite trend. This {demonstrated in Fig 1 Infact, leant aquarc analysis of 26 te te at aca af ee eet eeSOLUBILITY OF OXYGEN AND OZONE IN LIQUIDS m eat 22 4 109 x cue | Solbilty of oxygen inthe mllane. The number inthe gure ‘eferto dats taken from the following references: 1 (Ree $95 2 (Be. 63 (Ref: 4 Re 625 Ref 63 6[Ref. and 7 (Rel. oF pairs of data forall of the normal hydrocarbons in Table 14 sive do —Ua82, and 1W'o = 087. trimethylpentane, the mote fraction solubilities are 20% to 30% higher than those forthe normal hhydrocecbons. In addition, both the Kobatake and Hilde brand* and Kretschmer et a results show a decrease in solubility as the temperature increases AH; <0) Followingis a brief description of orginal data foreach solvent. ‘n-Hexane: smoothed values were obtained by using Naumenko*! and Makranczy eta values both at 208.15, K, and Guerry’s values" at 293.19 and 298.13 K. n-Heptane: The values of Guerry® at 293.15 and 298.15, Thomsen and Gjaldback® at 298.15, 98.25, and 307.95 K, jams at 298.15 K, and Blane and Batite! at 303.15 K were used s-Octane: Although the values of Wilcock etal are preferred, smoothed values were obtained by using those of ‘Thomsen and Gjaldback® and Tams, in addition to those of Wilcock eral 2.24Trimethylpentane: The four values of Kobatake and Hildebrand over the temperature range of 282.87 to 303.36 K give 4y~ — 65102, 4,— 1.8995, and 10% = 0001. The four values of Kretschmer eta over the tem- perature range of 248.15 to 323.15 K (which are lower than those of Kobatake and Hildebrand by 8.7%) give Ao = — 6.6138, 4, = 1.9369, and 10% = 0.53. The smoothed values were obttined by using the data from both of these papers along with those of Baldwin and Daniel and yams. n-Nonane: The Thomsen and Gjaldback values® at 2980 and 298.15 K and the Blane and Batiste value at 303.18 K were weed a-Decane: The values of Wilcock et al at 283.15, 298.11, and 313.48K givedy = — 6.2238, 4, = 0.3043, and 107 = 0.08. Although these values are prefered, emoothed values were obtained by using them along with the Iams’ value at 298.15 K.® ‘45. Cycle Hydrocarbons ‘Smoothed values of mole fraction solubility for a group of eyslis hydrocarhone are given in Table 15. Mole Faction solubility increases in the order cyclohexene < cyclooctane ~cyclohexane < methyleyclohexane ~cis-1,2-dimethyleyc- Iohexane < trans-1,2-dimethyleyclohexane~ 1,3-dimethyl- ‘yclohexane cis and trans mixture) ~ 1,4-dimethyleyclohex- ane (cis and zrans mixture). Solublites in these solvents are Tower than those in n-alkanes and higher than those in ben- ‘The following is a summary of the original data and how they were processed. Cyclohexane: Wilhelm and Battino® report sight points over the temperature range of 283.47 to 313.06 K, which give y= — 6.592, 4, = — 0.280, and 10% =0.10. Wild ef al.” report 11 points over the temperature range of 312.2t0 371.2 K, which give 4, = — 6312, 4, = — 1.278, and 10's = 0.12. Guerry* reports two points at 293.15 and J. Phys. Chem, Ref. Data, Vo. 12, No.2, 1988,172 BATTINO, RETTICH, AND TOMINAGA 298.15 K. All of the data of these three groups except for 1 point of Wild ef al, at 293.2 K were used to obtain the smoothed values. Cyclohexene: Only Guerry® reports values at 293.15 and 298.15 K. Methyleyclohexane; _cis-1,2-Dimethyleyclohexane; trans-1,2-Dimethyleyclohexane; mixture of 59 mol % cis- 1,3-Dimethyleyclohexane and 41 mol % trans-1,3-Dimeth- yleyclohexane; mixture of 70 mol % cis-1,4-Dimethyleyclo- hhexane and 30 mo! % srans 1,4-Dimethyleyclohexane: Only the data of Geller et al. are avsilabe.”* ‘Cyclooctane: Only the data of Wilenek er al. are avail- able” 4.5, Aromatic Hydrocarbons ‘The mole fraction solubility increases when an alkyl substituent is introduced on the Benzene ring, Le, from 8.10 10~* for benzene to 15.69 10~* for (1-methylpro- pyllbenzene, both at 298.15 K. This is shown in Tables 16 fand 17. In benzene, the solubility increases as temperature rises, and this corresponds to Ai; = 1.48 KJ mol! at 298.15 K. In methylbenzene and 1,4-dimethylbenzene, the temperature dependence of the solubility i wot lage cou pared to experimental uncertainty (477; ~0). “The following i a summary of how the data were han- ed Benzene: Horiut reported sx values in the tempera: ‘ture range of 283.15 to 333.15 K. Schlapfer et al." reported seven values the temperature range 283 to 343 K. Byrne et al reported four values from 310.6 t0 310.7 K. These arein satisfactory agreement and smoothed values were obtained bby using six values of Hort, seven values of Schlapfer eral. ang unre values of Byrne er af. Both the Naumnenko valve”* ‘298.15 K (which is 8.3% higher than the smoothed value) and the Morgan and Pyne value” at 298.15 K (which is 6.49% lower) were omitted Meer rrr el 4 Phvs. Chem. Ret. Data, Vo. 12, No.2, 1983, Table 17. ole fraction solubility of oxygen ‘ip aronatie hydrocarbons at 298.15 © Se Hagia re oo bk 2-Dinothy 1beozone ary 1,3-Dinetnyibenzene 11.96 unyibenzene 12.20 (-aotny lev )oenzene 13.88 Temetnyl A propyltenzane whee ‘autyivenzene sao (t-metnypropr Joenzene 15.69 aren Methylbenzene (Foluene}: Only the Field etal. data” ‘were used to obtain smoothed values. The Naumenko et a. value” at 298.15 K is 16% higher than the smoothed value of Field etal. and was omitted. 14-Dimethylbenzene (p-Xyleue}: Sia values of Fisher over the temperature range of 303.2 to 353.2 K” anda value ‘of Naumenko et al at 298.15 K were used to obtain the smoothed values. Other alkylbenzenes: Only Naumenko eral: values” at 298.15 K are available and they are shown in Table 16. 48. Oxygen in Organic Compounds Containing ‘Oxygen 45,.Aleohole ‘The smoothed values of the mole fraction solubilities are shown in Table 18. For 1-propanol, I-pentanol, 1-hex- anol, I-nonanol, I-undecanol, and I-dodecanol, only the ‘Makranczy et al. values are available However, their val- ues are systematically low when compared with those ob- tained by other groups and were thus consequently omitted. ‘The solubility increases as function of carbon number (see Fig. 2). The temperature dependence ofthe solubility is not ery lage butt decreas as the temperate iss king Toe 2eunethgtet-propauel ‘The following summarizes how the original data were handled ‘Methanol: Only values of Kretschmer et al. at 248.15, 273.15, 298.15, and 323.15 K® were used to obtain smoothed values. The Makranczy etal. value at 298.15 K"™ (which is 8% lower then that of Kretschmer et al, was ig- nored. Ethanol: At least eight groups have reported oxygen solubility i cul The urigiual data weal wy obtain the smoothed values areas follows: four values of Kretschmer et al over the temperature ange of 248.15 to 323.15 K were weighted tun times: vight walnne af Schlager ot 4973 to 343 K; six values of Timofeev,” 273 to 290.2 K with the value at 296.55 K omitted a value of Shchukarey and Tol- macheva" at 277.15 K with the omission of three values at higher temperatures; and a value of Naumenko™ at 298.15SOLUBILITY OF OXYGEN AND OZONE IN LIQUIDS 173 some hte retin shanty of mp te 81 ra rae ee) va 0 Es wT K. Both Metschi’s value! and Makranczy etal’ value" at 298.15 K were omitted; they are lower than the smoothed value by 12% and 25%, respectively. 2-Propanol: Only the values of Kretschmer et al. at 15, and 323.15 K” are available. |-Butanol: The smoothed solubility values are based on the Kreschmer ef al. data.*” The Makranczy etal. value at 298.15 K® (which is 27% lower than Kreschmer etal.) was ignored. '2 Methyl 1 propanol: Only Batting al. data are avail. able, 274.09 to 327.96 K. | -Heptanol: Solubility data are available only at 298.15 K. The Tams’ value is preferred to that of Makranczy et al in view of the trends in solubility with alcohol carbon ‘number. 1-Octanol: The values of Wilcock et al. at 283.35 and 298.13 K,'* Ijams at 298.15 K,”” and Makranczy ef al. at 298.15 K™ are in satisfactory agreement. However, only the ‘values of Wilcock eal. were used to obtain smoothed values. 273. TT [AY 0 4, 8 mrmmmemmnse uae for prope the itgle or 2a proper, ‘td the cos for jcebexanc. 10*x, 1 -Decanol: The smoothed values are based on the val- ues of Wilcock et al. at 282.74, 298.10, and 313.56 K.” The value of Makranczy eral. at 298.15 K* is 22% lower than that of Wilcock eal. at the same temnerature and was reject ed. Cyclohexanol: Only Cauguil reports a value at 299.15 K: 46h Ketones, Ethers, Acid, and Esters For 2-propanone (acetone), 1,1-oxybisethane (diethyl ether), and acetic acid, methyl ester (methyl acetate), solubi- lites have heen measured over a wide range of temperature (194.8 0318 K}, Smoothed solubility values for these three solvents are shown in Table 19. The solubility shows a mini- mum as a function of temperature for each of the solvents Smoothed values for other Ketones are shown in Table 20. For ketones, the solubility increases as a function ofthe car~ ‘bon number, i.e, from 8.40%10™ for 2-propanone to 11.74x 10 “Tor 2-nexanone at 298.13 K. Mole fraction so- Iubility in moneycic ethers (10%x, = 19 at 298.15 K for I,1- ‘oxybisethane and 1,1-oxybispropane) is as high as those in ‘normal alkanes: The solubility in oyeio others io about one half of the values in normal alkanes. Some data forthe solu- bility of oxygen in several ethers and an ester are given in Table 21 ‘The following summarizes the data treatment. 2-Propanone acetone}: Nine values of Horiuti, 194.85 to 313,15 K”; three values of Kretschmer etal, 248.15 to 298.19 K; and 21 values of Bub and Hillebrand, 298.2 t0 318 K* are in satisfactory agreement. All ofthe data of the three groups except for seven points from Bub and Hille- ‘brand, were used o obtain amoothed values. MetzohI's value at 298.15" is21% lower than the smoothed value and was omitted. 1,1-Onybisethane (diethylether): Six values of Horiut, 195.05 to 293.15 K,” a value of Schlapfer etal. at 293 K.”* 4. Phys. Chem. Ret. Data, Vol. 12,No. 2, 1989,174 BATTINO, RETTICH, AND TOMINAGA. and a value of Christoff at 273.15 K™ are in satisfactory ‘agreement ‘The smonthed values are based on these valuoe ‘Acetic acid, methylester (methyl acetate): Nine values of Horiuti 194.75 to 313.15 K,” and a value of Schlapfer al. at 293 K” arein satisfactory agreement and were used to ‘obtain the smoothed values. 2-Butanone, 2-Pentanone, 2-Hexanone: Only Bub and Hillebrand’s data are available. Allof the data were used to ‘oorain smoothen values for each solvent. ‘Cyclohexanone: Only Guerry’s data are available. 1,-Oxybispropane (dipropyl ether); tetrahydrofuran; 14 dioxane tetrahydro 2H pyrany oootioaoidy othyl cotor oe om Tame [he Phys. Chem. Ret. Data, Vol. 12, No.2, 1985 va (ethyl acetate}: Only one group reports solubility values for ‘cach solvent except for 1,4dioxane where Guerry's value is 16% lower than the value of Schlapfer eal. at 293.15K. 4.7, oxygen in organic Gompounas Containing Halogen ‘Smoothed values of mole fraction solubilities are shown ‘m Table 22 as a tunction ot temperature. ‘Ihe enthalpy of solution is nearly zero for carbon tetrachloride, positive for chlorobenzene (AH; = — 973 to — 489 Jmol~! at 298.15 K,), and negative for the perfluorocompounds including per- fluorotributylamine in Table 24: AH} = — 973 to — 2313 Jmol" at 298.15 K}. For most ofthe solvents the solubility Was measured only at 298.15 K. These data are chown in Table 23. ‘The following is a brief review of the original literature for the solvents listed in Table 21 wuss | sem | ron or | oe roe ‘ets ett fo teres rm‘SOLUBILITY OF OXYGEN AND OZONE IN LIQUIDS 175 ee oom | ord ort sa | opt roca | cae | ort weomee | tat | gga ft saune aa a ane cat? | aot ota convo | mat vette cata | net Stott sutaegt | at epee cate |e spat ute wa eat cua | eo oper semen | carat cairgtst | at seam wat | eyes | ut Seacoast. Sutera ven, 2035.5 Samattore trop erfermty pte, “eee or pre, "atcsntettl etre nites oe pent ree 8 Tae, tener eit son yer Sos ey Mer on Carbon tetrachloride: Horiuti” reports seven values over the temperature range of 273.15 to 333.15 K. The smoothed valics were obtained from these values. Both val ues of Metschl"* and Naumenko® at 298.15 K are lower than that of Horiuti by 19% and 17%, respectively. Chlorobenzene: Nine values of Horiuti” over the tem- perature range of 273.15 to 353.15 K were used to obtain the smoothed values. The Naumenko etal.”*point at 298.15 Kis 14% higher than Horiuti's value at the same temperature. Hexaftuorobenzene: The smoothed values were ob- tained by using four values of Evans and Battino™” in the ‘temperature range between 282.92 and 297.81 K. The value of Wesseler ef al. at 298.15 K is 4.1% lower than that of Evans and Battino, Perfluorobutylperfluorotetrahydrofuran; perfluoro-1- isopropoxy hexane; and perfiuoro-1,4-diisopropoxy butane: Only the values of Tham etal. are available.”* 4.8, Oxygen in Organic Compounds Containing Sulfur, Nitrogen, or Silicon Inmost ofthe solvents, the solubility was reported only ‘2208.18 K. In pyerolidine, pyridine, and piperidine, colubi lites have been reported also at 293.15 K, but these differ by ‘no more than 1.2% from the respective values at 298.15 K. Only for perfluorotributylamine and octamethyleyclotetra- siloxane are solubilities available for a wider temperature range. Smoothed values are given in Table 24. In dimethyl- sufloxide, two other groups report solubility values: Those of ‘Batra ana Foley”® are 31% lower than that of Dymond”; ‘Chaenko et al determined the solubility by using an elec- ‘rochemical sensor and their value is 5.9 times higher than Dymond’s value”? when converted to 1 atm of oxvaen nartin! error aot sae oan fea anne of sae ons of ey Ale ae he at, Be monn. ‘4. Phys. Chem. Ret. Data, Vol. 12, No.2, 1983178 RATTINO, RETTICH, AND TOMINAGA pressure, For perfluorotrbutylamine, Wesseler eta and Zande: report solubility values, but only the Kobatake and Hildebrand values" were used to obtain the smoothed val- ues. For nitromethane and L-propyl nitrate, solubilities were reported by Chaenko et al and Amster and Levy." but these are not included in Table 24. 4.9. Oxygen in Biological Fluids Because of its biological importance the solubility of ‘oxygen has been determined in many biological fluids. Some of these measurements are of reasonable precision, but a reat many are of poor precision. Also, itis difficult to exact- ly characterize many ofthe biological fuids used. In terms of| examining trends and having a rough knowledge of solubi- lity, the data are reasonable. However, we will only report on {wo tuids here—solubilities m olive oll and in human blood. For the other fluids (including animal and vegetable oils, lipids and lipid solutions, aqueous protein solutions, serum rplaoma, tiauc, alcoholic solutions, sugar solutivu, aud ‘wines, among others) we refer you to Ref. 1 ‘494, Solubility in Ove Ot ‘The solubility of oxygen in olive ol has been deter- mined by four groups,””*" but by far the most precise ‘work was that done by Battino and co-workers.” Table 25 tsives the smoothed mole fractions and Ostwald coefficients ‘Obtained from the following equation: [RAT mol! KIT Rn, = 1240.9 + 39.809(7 /K} (22) ‘The density of olive oil is given by p/tg.em~ 0.000.468: (¢/°C) and its molecular weight is 884-445, gmol~'. The standard deviation of the solubility is about 05%. 8 oe ona ‘49, Solubitty in Human Blood ‘The solubility of oxygen was determined in human loud of approximately 15.5 g/100 mil of hensoglobin by four ‘groups **!%"°2 There was substantial agreement and in ‘the range 273 to 310 K their data can be represented by the following equation to about 4% for ane standard deviation in the Bunsen coeficient: Ina = 1.9685-0.018 638(7'/K) 3) ‘Av 310.13 K the smoothed value ofthe Bunsen coeficient is 0.0221. 4. Phvs. Chem. Ref. Data, Vol 12. No.2 1983, 4.10. Oxygen Solubility in Miscellaneous Fluids See Ref, 1 forthe solubility of oxygen in variety of ‘miscellaneous fluids. These data are not included here for the fluids are special, poorly characterized, or would need excessive space to fully characterize by physical properties. Included in the abovesited reference are solubilities in HF, $05, N,O,,N; + CO; mineral cl oils, hydrocarbon fuels, petroleum, liquid parafin, surfactants; and above 0.2 MPa solubilities in Cly, N0, and CO;. 5. Ozone Solubilities ‘Except for the solubiity of ozone m water, there are very few measurements of ozone solubilities. In this section ‘we will only deal with the solubility of ozone in water. For solubilities of oronc in helogenatia solvents, CCl, satcated ‘with water and various concentrations of nitric acid, various ‘concentrations of sulfuric acid, and a group of miscellaneous solvents (acetic acid. acetic anhyeiride. trifuarnacetic acid, ‘methyl acetate, propionic acid, acetic acid/water mixtures, acetic acid + propionic acid, and (Cl, + CHCl,CHC!, mix- tures) see Battino." 5.1. Ozone in Water ‘The determination of ozone solubilities in pure water is complicated because ozone decomposes irreversibly in wa- ter: The relevant presentation is 20,a)-920,f0q)-220,0a)-»30 ea) ‘Therelevant gas solubility is step 1, step is reversible, and step 3 may or may not attain equilibrium conditions. Unless extreme care is taken the solubility via step 1 may not be ‘what is actually determined. ‘An analysis and solution ofthe problem is provided by Roth and his co-workers." Their approach involved determining ozone concentrations in the gas phase in equi- librium with the ozone-saturated liquid (water) phase and ddoine this as a function of time so that the decomposition rate could also be determined. Since ozone solubility de- ‘pends on the pH of the aqueous solution, solubilities were ‘determined in buffered water, the pH being adjusted by otSOLUBILITY OF OXYGEN AND OZONE IN LIQUIDS 7 H,SO, oF NaOH. In the temperature range 276.5 0333.2 K (measurements also at 293.2 and 313.2 K) they found the following equation to best represent their data: Ky = 3.842X10"(OH }®™expl — 2428/17 /K)}. (25) Ky is the Heary’s law constant [see Bq, (14), and [OH] is the hydroxide concentration in mol L~'. The index of deter- mination was 0.84 with a standard error at estimate of 0.20. ‘Table 26 gives smoothed values obtained from Eq, (25) at tree temperatures and at integra values of pH from | to 10. ‘These values are considered to be the best currently avail- able, but are necessarily of poor precision. For a detailed discussion of other results of ozone solu- bility in water see Refs. 1, 103, and 104, 6. Conciuaing Upservations As befits its importance the solubility of oxygen has been measured in a wide variety of solvents and solutions. ‘This paper shows some gaps in the coverage. Definitive stu- dies need to be carried out in homologous series of n-alkanes and L-alkanols over a range of temperatures. On the other luau, yuilesciable valucs arcu availall fin vayges sul bilities in water and seawater. Additional measurements are needed in heavy water, however." For perspective we have prepared Fig. 3 which shows oxygen solubilities as a mole fraction at 101.325 kPa partial pressure of gas in some six- teen solvents as a function of temperature. Finally, if experi- ‘menters carried out their determinations over a wider range ‘of temperatures, this would result in more reliable thermo- dynamic data 1615, rage Lig et @ xe 4 a eo 4 3 Lj ——— 4 _—— 7 (eae ae eke cee 33 05 38 oa 1OOKTIK) FFigune 5. Mole fraction alublites st 101.325 LPs parti pressure of xy- 20s methane 2—ethano, hema: dort, ‘ethyl acct, 6 carbon tetrachloride, 7—eyeobexane: esl ether, 9—moctane, 10—herauorobenzene, 11-224. ‘eimatbyentang, 12—octamethyleyltetrasionane; per. Aorotoutyamine, 1—foorocarbon Aub, FCO, 15—ea xin and 16~carori-D. 7. Acknowledgments We wish to express our appreciation for grant No. NB SINADA 2006 from The Ofice of Standard Reference Data ofthe US. National Bureau of Standards. We also acknowl- eg the assistance of C-J, Chou, T. Park, and B. Ramali om. 8. References “Oxygen and Orone, TUPAC Solsbilty Data Sei etal by R. Bano {Pergamon Oxford, 198), VoL. ‘RP Madam and KA: Kobe, Chem. Re. 25, 191941, 2R Patio and HL. Cher, Chen Rev 6395 0566) “+H. L. Gover and R Batino, “The Slob of Gass Liga" in Soins and Slats, ete by MR 1. Dack (Wily, New York, ‘bioyou sari pp SP °E Wii sad R. Baio, Che. Re. 73,1973. °F, Wile, Bain, ad R. Wisk, Chem, Re. 7,219 (1977, "WG Soluble ase and Ligue, New York, 170, ‘Wi Gerad, Ga Slbiti, Widespread Apliatons (Pergamon, Ox- for, 1980p 7 "ALS KerteO, Len, .Y. Matos, “Sobtiliy”in Experi ‘ei Theron f Nerang Fite Vol Wnt by 8 Veer {Bernt London, 1973) pp 785-748. "Sous of Inorganic and Orie Compound ited by H. Stephen ‘and Stephen, Vols and 2 (cll, New Yor, 1983, "AS Solio org and Metal Organic Compound, Sed ‘5 Van Nostrand, New York. 140), ‘Landolt Bot: 2. Tel, Bane anon chywuthe Spine ‘Det, 1960 1 Band eck, 4 Tel Warmth BandtllG isp der Abortion von Gani Pasion Singer, Ber in) interatonal Cra! Table, ite by . W. Wash (McCaw il, New Yor 1900 ‘]M. Past Molla Thermadynanief ii Phae Eiri {Prenioe al Eales Gly Now Sony, 1095p. 32. LH Hira M Prasat and RL Sct, Repolar and Related Solutions (Van Nota Reba, New Yor, 1970p. 28 "SL.H. Dymond and EB. Smith The Vr! Cafe Pare Gases ond ‘Max odo, Carndon, Orford 19) "BB Besson nd Kran, Chm. hy 64, 69 (1976. NG. S. Kel J. Chem. Eng. Dats 15,119 (1970, "7B. B. Benson, D. Knut, and M.A. Pete, J Sola. Chem. 8 688 nny 2.HL Carpenter, Lino. Ocenog 1, 26(196. 2E Doula. Phy: Chem. 6, 169 (1964, 2G, Gefen, Z Physik Chem. 4, 257 (194 20 8 Klos and BB. Benson) Marine Res 24481960, Livingston R Morgan and AH. Richa.) Pye Chem, 34,2356 (1330, MCLN. Mara, JP Riley, nd TR. Wilton, Deep Seu Re 15,237 1968) ON rn and J. Riley, Des Sea Ret. 16,311 (1969) PD. M. Now an B.E, Conway, Chem Isramenttion 5, 71974. 6.6. Power Ap Phyl 4, 458 (1968) 76.6. Pomecand Steg 3. App. Phyl. 2,148 1970. iL a: Pay and EF. Stepan, U's Atom Energy Commision, BM- or 25H, ten, Limnol Oceanog 3 423 (198 2W.Timofen 2 Physik. Chem 6 14 (1850, SELL Span 8 Hoel 8-DeondT A.Pey, BMEIO67 (1956, HLA: Pay, LB, Shih, and BH. Minch, Ind Eg. Che. 4 1146959 ‘SILL, Chava C1. Han, “The Saat atin Water frm 60 {0 Kin Thormodynamicsof deta Sen ith nda Appice ans ACS Symposia Seis 138 eed by 6. A. Newman (American {Chena Soiy, Washington,DC, 1980p. 13-36 CHL Mosier, Internationale Vercniane fr Teortche nd “Angewandte Linadlog Communication No 22, "The Oxygen Cow 4 Phys. Chom. Ret. Data, Vol. 12, No.2, 1983178 BATTINO, RETTICH, AND TOMINAGA lent of Airsturated Fresh Waters over Ranges of temperature and At- _mospherc Pressure of Linnologcal Interest,” Statgar, 1981 RLF. Weis, Dep Sea Res. 17,811 (1970) MCT. Chon and 1H Carpenter, bic mancrn (1078) D.1.1. Kinsman, M.Boardnan, and M. Bore in th Salt Symposium, cited by A. H. Coogan (Northen Geological Sock, Cleveland, Oo, 1973, Vol p. 328. “6G, Brun, J. Gesach. and F, Pawlek. Z. Ano. Alle Chem. 837,68 (1963). ‘$C: Bohr, 2. Physik. Chem. 71, 47 (191 °V. L Popreboaya, A. P. Uso, and A. V. Baranoy, Zh. Vee. Khim, Obschebet 17,344 (1972) ‘N-E Khomutoy and F-L-Koaaik, Zh. Fiz. Khim. 48 620(1976). “A. Broden and R Simonson, Sven. Pappertida #1, 411978). “A. Yasui, J. Chem. Hog. Jpn. 10, 891977} Kagaku Kogaia Rom Duns 185 (1978, “S:K Shor, KD. Waker, and K. E Gubbing 2. Phys, Chem. 8, 312 (1969), “M.D. Knasterand L.A. Apebaum, Zh, Fie. Khin, 3,223 (1960); Russ J.Phys. Chem 38,120 (1964), 120.01 Mowat, -Fuyo Chew 20, 499(1910 “TA. Mishning,O. Adceva, and. K-Boshowskaya, Materaly Vr [Nauchn. asked: Geol as. 46,93 196), MS. Gusev, E. Konik, R-N. Sat A. A. Kur'nin, nd I. Mor- dsthvioty Pie Rastvvoy 105 (172) IC A. Kobe and F-H. Kenton Ind, Eng. Chem, Anal, E410, 76(1938) 245. Thaver and M.M. Sharma, Chem. Bag, Se. 23, 11968. A. Bocken and G. Hertaberg,Z. Paik. Chez. 198, 1(1950 GAP T Mowsaon aa NT. Tose, Thom Se 4 (1985) °K, Shkoltkova, Uh. Zap. Leningrad Gos Uni, Se. Khim. Nauk. Nes, 64 (1959). C.J, Anderson, R.A. Keeler, and. J Klac, J Chem. Eng- Data 7,290 (196, £3. Tokunaga, J. Chem. Eng, Dats 20, 4 (1979. ‘3. W. Carp J. Chem. So. Faraday Trans. 172, 296(1976. 1, Markranry, K. Megyery-Balog, L Rus, an L. Pty, Hang. Ind ‘Chem 269 (1996, ‘°C: Blinc and M, Baste, Bul Cet. Rech Pau SNPA 4,235 (1970) ‘KK. Newmenko, Candidate's thesis Leningrad, 1970, ‘©D. Guery, PAD. thes, Vanderbilt Univer, 1948, (SES Thomsen and J.C. Galdbaek, Acta Chet. Scand 17, 127 (1963). “RJ. whet, R- Bano, ana W-F. Danforth, Cbem Thermo 1, arr s78) ‘£C. Cams, PLD. thesis, Vanderbilt Universi, 1981 SY. Kobatake and. H. Hildebrand, Phys. Chem. 653311961, 6. B. Kectchmce, J. Nowakowsha, and R, Wit Ind. Bag. Cc 38, 506 (1946, ‘RR. aldwin and S.G. Daniel J. ast. Peto. 39,105 (195353. App hems ano) 2 1619594. ‘Wii aed 8 Dettin, 3 Chom. Thermedya: 6119 (1979, 7%, D. Wil T. Seca, and O.E. Potter, Chem. Bag. J. 18,209 (1978, E.B. Geller, R. Battno, and E. Wileln, J. Chem. Thermodya. 8 197 1976, PRI. Wick. R.attnn and F. Wile. ¥ Chom Thermos 8.118 (977. "PL Horii Sci Papers. Phys. Chem. Research (Tokyo 1,125(1931)- 4. Phys. Chem. Ret. Data, Vol. 12, No.2, 1983 P,Schlaper,T. Avdykowski, and A. Bukowisci, Schweiz, Arch. An- gem. Wis Teoh 15,299 (194), LB. Byre,R Batino, and W.F. Danforth J. Chem, Thermos 6,245 usr YIN. K. Noumenko, M. M. Mahia, ad V. B. Aleskoek, Zh. Pik. ‘Khim, (Leningrad) 42, 252 (1969 3 Appl. Chem. 2, 2575199). LR. Morgan and H.R Pyne, J Phy: Chem. 38, 2085 (1930, MLR. Field, E Wilhelm, and R: Rati, 1 Chom Thorman 6 997 tox "0. . Fisher, J. Chem. Eng Data 8, 26 (1963, [Sex corsecton, bid 8, 5111963). "J Mairancy, L. Rus, and K, Balog-Menvey, Hung. Ind, Chem. 7,41 usr. 1G, Cai J Chim. Phys. 2, 531927, "S.A. ShehukarevandT. A. Tolmacheva, Zh, Stok, Khim. 9,21 (1968; 4 tract. Chem. 8, 16 1968) ‘3. Met J. Phys, Chem, 417 (1924, "R, Battine,F.D. Evags, W. F. Danforth and E, Wilhela, J. Chem. ‘Thermodyn 3,743 (1971 ‘$0. K Bub and W. A. Hillbran, 3. Chem. Bog. Data 24,315 1979 “A Gis 2 PRY: Ue. 9, 4900914) "FD, Evans and R Battin, J. Chom, Thenmodya. 3, 7531971. ‘NE. P Wessel, Ris and. C. Clark, J Florine Chem. 9, 137 (1977, "MK. Thar, RD, Walker, and J Model J Chem. Eng, Data 18 385 or. *D. D. Lawson, J. Moacanin KV. Scherer. F.Teranovs, snd J. Ingham, J. Fvrine Chem. 12,221 (1978 23, Chr Gaba, Acta Chem, Sand. 6623 (1950, {BW Tat ana BT Pay Tham Sing Date £7,268 (1073 EI.H. Dymond, J. Phys Chem. 7, 18291967, *'N.V. Chnenko, G.[ Sabhora, N’K. Naumenko, and. A. Keds ‘Zhi Khim 83,1189 (1979; Rose J Pye. Chen. $3,133 1979, ‘PR. Zander, Res Exp, Med. 164, 97 (1974, A. BL Amster and JB. Levy, ARS. 28, 70 (1959, "*D_ Davison, P. Eggleton, and P. Fopai, Quart J Expt Physio. 37,91 (1953), "5P.1.Keand RG. Ackman, J. Am. Oil, Chem. Soe $0,429 (1973). "°K. Bato, FD. Evans, aad W.F. Danforth, Am, Oli Chr. So. 45, 350(1968) "SC. Christofides and J. Heley-Whyte, J. Appl Physio. 27, 582(1969, "RJ. W. Roughton and J. W. Sevraghaus 3. App. Physio. 38, 861 ast {SC Cristofries and. Hedley Wate J. App: Physiol. 0, 815 (1976, DE. Sulivan, PLD. Thass, Vanderbilt University, 1979 ‘5, A Roth and DE. Sulivan Ind. Eng. Chem. Fundam 20, 137(1981) Nat added npr: Congrorc and Walco cpa ee unpgee ‘DO system. Their data follow as temperatre (17, aad nl aston soluiity att atn pat presure of gas 10% 278.15, 0.3729, 283.15, (02980; 293.15, 0.2673, 298.15, 02489: 303.15, 02368, matog. 216,161 (18 "Noe add i pray. See E. Wilelm's atte in iron ond Ait ia TUPAC Solbilty Data Sere etited by R. Battin Pergnmon, Oxford, 1009), Val IN for gore detonate Ota eeicen ari larly a applied to high precision measurements.