Journal of Hazardous Materials 107 (2004) 27–36

Modeling evaporation using models that are not

boundary-layer regulated
Merv F. Fingas∗
Emergencies Science and Technology Division, Environment Canada, Ottawa, Ont. Canada

Abstract

Experimentation shows that oil is not strictly air boundary-layer regulated. The fact that oil evaporation is not strictly boundary-layer
regulated implies that a simplistic evaporation equation suffices to describe the process. The following processes do not require consideration:
wind velocity, turbulence level, area, thickness, and scale size. The factors important to evaporation are time and temperature.
The equation parameters found experimentally for the evaporation of oils can be related to commonly available distillation data for the oil.
Specifically, it has been found that the distillation percentage at 180 ◦ C correlates well with the equation parameters. Relationships have been
developed enabling calculation of evaporation equations directly from distillation data:
percentage evaporated = 0.165(%D) ln(t) (1)

where %D is the percentage (by weight) distilled at 180 ◦ C and t is the time in minutes.
These equations were combined with the equations generated to account for the temperature variations:
percentage evaporated = [0.165(%D) + 0.045(T − 15)] ln(t) (2)

The results have application in oil spill prediction and modeling. The simple equations can be applied using readily available data such as
sea temperature and time. Old equations required oil vapour pressure, specialized distillation data, spill area, wind speed, and mass transfer
coefficients, all of which are difficult to obtain.
© 2003 Elsevier B.V. All rights reserved.

Keywords: Evaporation; Oil evaporation; Oil spills; Boundary-layer evaporation; Weathering

1. Introduction Scientific and quantitative work on water evaporation is

decades old [3,4]. The basis for the oil work in the literature
Evaporation is a very important process for most oil spills. is water evaporation. There are several fundamental differ-
In a few days, typical crude oils can lose up to 40% of ences between the evaporation of a pure liquid such as wa-
their volume [1]. Most oil spill behaviour models include ter and that of a multi-component system such as crude oil.
evaporation as a process and in the output of the model. Most obviously, the evaporation rate for a single liquid such
Despite the importance of the process, relatively little work as water is a constant with respect to time [3,4]. Evaporative
has been conducted on the basic physics and chemistry of loss, by total weight or volume, is not linear with time for
oil spill evaporation [2]. The difficulty with oil evaporation crude oils and other multi-component fuel mixtures [5].
is that oil is a mixture of hundreds of compounds and this Evaporation of a liquid can be considered as the move-
mixture varies from source to source and even over time. ment of molecules from the surface into the vapour phase
Much of the work described in the literature focuses on above it. The layer of air above the evaporation surface is
‘calibrating’ equations developed for water evaporation [2]. known as the boundary layer [6]. The characteristics of this
Furthermore, very little empirical data on oil evaporation air layer or boundary layer can influence evaporation. In the
has been published in the past. case of water, the boundary layer regulates the evaporation
rate. Air can hold a variable amount of water, depending
∗ Tel.: +613-998-9622; fax: +613-991-9485. on temperature, as expressed by the relative humidity. Un-
E-mail address: [email protected] (M.F. Fingas). der conditions where the boundary layer is not moving (no

0304-3894/$ – see front matter © 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.jhazmat.2003.11.007
28 M.F. Fingas / Journal of Hazardous Materials 107 (2004) 27–36

wind) or has low turbulence, the air immediately above the Much of the pioneering work for evaporation work was
water quickly becomes saturated and evaporation slows or performed by Sutton [7]. Sutton proposed an equation based
ceases. In practice, the actual evaporation of water proceeds largely on empirical work:
at a small fraction of the possible rate because of the sat- E = KCs U 7/9 d −1/9 Sc−r (4)
uration of the boundary layer. The boundary-layer physics
is then said to regulate the evaporation of water. This reg- where Cs is the concentration of the evaporating fluid
ulation manifests itself in the sensitivity of evaporation to (mass/volume), U the wind speed, d the area of the pool, Sc
wind or turbulence. When turbulence is weak or absent, the Schmidt number and r the empirical exponent assigned
evaporation can slow down by orders-of-magnitude. The values from 0 to 2/3. Other parameters are defined as above.
molecular diffusion of water molecules is at least 103 times The terms in this equation are analogous to the very generic
slower than turbulent diffusion [6]. Eq. (1), proposed above. The turbulence is expressed by a
Evaporation can be viewed as consisting of two compo- combination of the wind speed, U, and the Schmidt number,
nents, fundamental evaporation and regulation mechanisms. Sc. The Schmidt number is the ratio of kinematic viscosity
Fundamental evaporation is that process consisting of the of air (ν) to the molecular diffusivity (D) of the diffusing gas
evaporation of the liquid directly into the vapour phase in air, i.e., a dimensionless expression of the molecular dif-
without any regulation other than by the thermodynamics of fusivity of the evaporating substance in air. The coefficient
the liquid itself. Regulation mechanisms are those processes of the wind power typifies the turbulence level. The value
that serve to regulate the final evaporation rate into the en- of 0.78 (7/9) as chosen by Sutton, represents a turbulent
vironment. For water, the main regulation factor is the air wind whereas a coefficient of 0.5 would represent a wind
boundary-layer regulation discussed above. The boundary- flow that was more laminar. The scale length is represented
layer regulation is manifested by the limited rate of diffusion, by d and has been given an empirical exponent of −1/9.
both molecular and turbulent diffusion, and by saturation This represents, for water, a weak dependence on size. The
dynamics. Molecular diffusion is based on exchange of exponent of the Schmidt number, r, represents the effect of
molecules over the mean free path in the gas. The rate of the diffusivity of the particular chemical, and historically
molecular diffusion for water is about 105 slower than the was assigned values between 0 and 2/3 [7].
maximum rate of evaporation possible, purely from thermo- This expression for water evaporation was subsequently
dynamic considerations [6]. The rate for turbulent diffusion, used by those working on oil spills to predict and describe
the combination of molecular diffusion and movement with oil and petroleum evaporation. Much of the literature fol-
turbulent air, is on the order of 102 slower than that for max- lows the work of Mackay and co-workers [5,8]. Mackay and
imum evaporation. In fact, in the case of water, maximum co-workers adapted the equations for hydrocarbons using
evaporation is not known and has only been estimated by the evaporation rate of cumene. Data on the evaporation of
experiments in artificial environments or by calculation [3]. water and cumene have been used to correlate the gas phase
If the evaporation of oil was like that of water and was mass transfer coefficient as a function of wind speed and
boundary-layer regulated one could write the mass transfer pool size by the equation:
rate in semi-empirical form (also in generic and unitless Km = 0.0292U 0.78 X−0.78 Sc−0.67 (5)
form) as:
where Km is the mass transfer coefficient in units of mass
E ≈ KCTu S (3)
per unit time and X the pool diameter or the scale size of
where E is the evaporation rate in mass per unit area, K the evaporating area. Stiver and Mackay [5] subsequently de-
mass transfer rate of the evaporating liquid, presumed con- veloped this further by adding a second equation:
stant for a given set of physical conditions, sometimes de- km AP
noted as kg (gas phase mass transfer coefficient, which may N= (6)
RT
incorporate some of the other parameters noted here), C the
where N is the evaporative molar flux (mol/s), km the mass
concentration (mass) of the evaporating fluid as a mass per
transfer coefficient at the prevailing wind (m/s), A the area
volume, Tu a factor characterizing the relative intensity of
(m2 ), P the vapour pressure of the bulk liquid (Pa), R the
turbulence, and S a factor that relates to the saturation of the
gas constant (8.314 J/(mol K)), and T the temperature (K).
boundary layer above the evaporating liquid. The saturation
Thus, boundary-layer regulation was assumed to be
parameter, S, represents the effects of local advection on
the primary regulation mechanism for oil and petroleum
saturation dynamics. If the air is already saturated with the
evaporation. This assumption was never tested by exper-
compound in question, the evaporation rate approaches zero.
imentation, as revealed by the literature search [2]. The
This also relates to the scale length of an evaporating pool.
implications of these assumptions are that evaporation rate
If one views a large pool over which a wind is blowing, there
for a given oil is increased by:
is a high probability that the air is saturated downwind and
the evaporation rate per unit area is lower than for a smaller • increasing turbulence;
pool. It should be noted that there any many equivalent ways • increasing wind speed; and
of expressing this fundamental evaporation equation. • increasing the surface area of a given mass of oil.
 M.F. Fingas / Journal of Hazardous Materials 107 (2004) 27–36 29

These factors can then be verified experimentally to 3.1. Wind experiments

test whether oil is boundary-layer regulated or not. These
factors formed the basis of experimentation for this Experiments on the evaporation of oil with and with-
paper. out wind were conducted with Alberta Sweet Mixed Blend
(ASMB), gasoline, and with water. Water formed a base-
line data set since much is known about its evaporation be-
2. Experimental haviour [11]. Regressions on the data were performed and
the equation parameters calculated. Curve coefficients are
Detailed experimental results are given in the literature the constants from the best fit equation [Evap = a ln(t)], √t:
[9]. time in minutes, for logarithmic equations or Evap = a t,
Evaporation rate was measured by weight loss using an for the square root equations. Oils with few components
electronic balance. The balance was a Mettler PM4000. The evaporating at one time have a tendency to fit square root
weight was recorded using a Toshiba 3100, a serial cable to curves [12]. While data were calculated separately for per-
the balance and a modified version of the software program, centage of weight lost and absolute weight, the latter is usu-
‘Collect’ (Labtronics, Richmond, Ont.). ally used because it is more convenient. Both values show
Measurements were conducted in the following fashion. the small relative upward tendency with respect to wind ef-
A tared petri dish of defined size was loaded with a mea- fects. The plots of wind speed versus the evaporation rate
sured amount of oil. At the end of the experiment, vessels (as a percentage of weight lost) for each oil type, are shown
were cleaned and rinsed with dichloromethane and a new in Figs. 1–3. These figures show that the evaporation rates
experiment started. The weight loss dishes were standard for oils and even the light product, gasoline and water are
glass petri dishes from Corning. A standard 139 mm di- not increased by a significant amount with increasing wind
ameter (i.d.) dish was used for most experiments. For the speed. In most cases, there is a rise from the 0-wind level to
experiments in which area was a variable, dishes of other the 1 m/s level, but after that, the rate remains relatively con-
diameters were used. stant. The evaporation rate after the 0-wind value is nearly
The constant temperature chamber (room) used was a identical for all oils. The oil evaporation data can be com-
constant temperature model constructed in 1993. It could pared to the evaporation of water, as illustrated in Fig. 3.
maintain temperatures from −40 to +60 ◦ C and regulate the These data show the classical relationship of the water evap-
chosen temperature within ±1 ◦ C. oration rate correlated with the wind speed (evaporation
In experiments involving wind, air velocities were mea- varies as U0.78 , where U is wind speed). This, by itself,
sured using a Taylor vane anemometer (no model number would appear to indicate that the oils used here are some-
on the unit) and a Tadi, ‘Digital Pocket Anemometer’. what boundary-layer regulated, but only to the degree that
Details of these measurements are given in the literature the effect is seen in moving from 0-wind to 1 m/s, and not
[9]. These velocities were later confirmed using a hot wire thereafter.
anemometer and appropriate data manipulations of the
outputs. The anemometer was a Thermo Systems (TSI)
model 1053b, with power supply (TSI model 1051-1), 50
averaging circuit (TSI model 1047), and signal linear-
lizing circuit (TSI model 1052). The voltage from the
averaging circuit was read with a Fluke 1053 voltmeter. 40
The hot wire sensor (TSI model 1213-60) was angled at
45◦ .
Percent Evaporated

30
Evaporation data were collected on the Toshiba 3100
laptop computer and subsequently transferred to other com-
puters for analysis. Curve fitting was performed using the
20
software program “TableCurve”, Jandel Scientific Corpora-
tion, San Raphael, CA. wind = 0 m/s
Oils were taken from supplies of Environment Canada 10
wind = 1 m/s
wind = 1.6 m/s
and were supplied by various oil companies for environmen- wind = 2.1 m/s
wind = 2.6 m/s
tal testing. Properties of the oils can be found in standard
references [10]. 0

3. Results and discussion 0 500 1000 1500 2000 2500

Time - minutes
The results of the boundary-regulation experiments are
presented in the order of the experimental series. Fig. 1. Evaporation of ASMB with varying wind velocities.
30 M.F. Fingas / Journal of Hazardous Materials 107 (2004) 27–36

100 25

Gasoline
Evaporation Rate (%/min.or%/ln min.)

80 20
Percent Evaporated

60
15

40
Water (adjusted)
10
wind = 0 m/s
wind = 1 m/s ASMB
20 wind = 1.6 m/s
wind = 2.1 m/s
wind = 2.6 m/s 5

0
FCC Heavy Cycle
0

0 20 40 60 80 100 120 140 160 180 0 1 2

Time - minutes Wind Velocity - m/s

Fig. 2. Evaporation of gasoline with varying wind velocities. Fig. 4. Correlation of evaporation rates and wind velocity.

Fig. 4 shows the rates of evaporation compared to the wind All the above data show that oil is only slightly, if at all,
speed for all the liquids used in this study. This figure shows boundary-layer regulated, perhaps only affecting the very
the evaporation rates of all test liquids versus wind speed. initial rates after turbulence is applied.
The lines shown are those calculated by linear regression
using the graphics software, Sigma Plot. This clearly shows 3.2. Evaporation rate and area
that water evaporation rate increased, as expected, with in-
creasing wind velocity. The oils, ASMB and gasoline, do not Alberta Sweet Mixed Blend was also used to conduct a
show a significant increase with increasing wind speed. The series of experiments with varying evaporation area. The
increase may only be a result of the increase in evaporation in mass of the oil was kept constant so that the thickness of the
going from the 0-wind level to the other levels. In any case, oil would also vary. However, the greater the area, the lesser
they do not show the U0.78 relationship that water shows. the thickness and both factors would increase oil evapora-
tion if it were boundary-layer regulated. Data are illustrated
100 graphically in Fig. 5. These data show, at best, a very weak
correlation of thickness and area with evaporation rate. Be-
cause of the poor correlation between area and evaporation
80 rate, one can conclude that evaporation rate is not highly
correlated with area and that the evaporation of oil is not
boundary-layer regulated to any significant degree.
Percent Evaporated

60
3.3. Study of mass and evaporation rate

40 Alberta Sweet Mixed Blend oil was again used to conduct

a series of experiments with volume as the major variant. Al-
ternatively, thickness and area were held constant to ensure
wind = 0 m/s that the strict relationship between these two variables did
20 wind = 1 m/s
wind = 1.6 m/s
not affect the final regression results. Fig. 6 illustrates the
wind = 2.1 m/s relationship between evaporation rate and volume of evapo-
wind = 2.6 m/s
regression line ration material (also equivalent to mass of evaporating mate-
0
rial). This figure illustrates a strong correlation between oil
0 20 40 60 80 100 120 140 160 180
mass (or volume) and evaporation rate. This suggests no or
little boundary-layer regulation. It also shows that any ten-
Time - minutes
dencies observed in the area tests described above may have
Fig. 3. Evaporation of water with varying wind velocities. been due to volume/mass factors rather than area.
 M.F. Fingas / Journal of Hazardous Materials 107 (2004) 27–36 31

0.6

4
0.5
Evaporation Rate-g/ln min.

Evaporationrate - g/min.

3
0.4
wind

0.3
2

0.2

1
0.1 no wind

0 0.0
0 50 100 150 200
6 8 10 12 14 16
Area - cm2
Hydrocarbon Number
Fig. 5. Correlation of area with evaporation rate.
Fig. 7. Evaporation Rate of pure compounds.

3.4. Study of the evaporation of pure hydrocarbons—with cating boundary-layer regulation. Decane (carbon number
and without wind 10) shows a lesser effect and hexadecane (carbon number
16) shows a negligible difference between the two exper-
A study of the evaporation rate of pure hydrocarbons imental conditions. This experiment shows the extent of
was conducted to test the classic boundary-layer evapo- boundary-regulation and the reason for the small or negligi-
ration theory as applied to the hydrocarbon constituents ble degree of boundary-regulation shown by crude oils and
of oils. The evaporation rate data are illustrated in Fig. 7. petroleum products. Crude oil contains very little material
This figure shows that the evaporation rates of the pure with carbon numbers less than decane, often less than 3%
hydrocarbons have a variable response to wind. Heptane of its composition [11]. Even the more volatile petroleum
(hydrocarbon number 7) shows a large difference between products, gasoline and diesel fuel only have limited amounts
evaporation rate in wind and no wind conditions, indi- of compounds more volatile than decane, and thus are also
not strongly boundary-layer regulated.
5 Another evaluation of evaporation regulation is that of sat-
uration concentration, the maximum concentration soluble
in air. The saturation concentrations of water and several oil
components are listed in Table 1 [13]. This table shows that
4
Evaporation Rate - g/ln min.

saturation concentration of water is less than that of many

3 Table 1
Saturation concentration of water and hydrocarbons
Substance Saturation concentration at 25 ◦ C (g/m3 )a

2 Water 20
n-Pentane 1689
Hexane 564
Cyclohexane 357
1 Benzene 319
n-Heptane 196
Methylcyclohexane 192
Toluene 110
0 Ethybenzene 40
p-Xylene 38
0 20 40 60 80 100 120 140
m-Xylene 35
Weight - grams o-Xylene 29
a Values taken from Ullmann’s Encyclopedia.
Fig. 6. Correlation of mass with evaporation rate.
32 M.F. Fingas / Journal of Hazardous Materials 107 (2004) 27–36

use, and a way to accurately predict evaporation using other

5 readily available data is necessary. Two means to predict the
evaporation were developed. The first data type is to use the
Evaporation Rate %/ln or sq. rt. min

value of the slope (fitted with one parameter) at 15 ◦ C as a

4 basis for correlation. The assumption here is that the slopes
of the temperature parameters are similar, so that they can
be used as a predictor. It has already been noted that only
3 light and medium crude oils display similar slopes. How-
ever, it will be fruitful to test such a hypothesis. The other
observation noted is that the slope of the equation appears
2 to correlate with the magnitude of the evaporation rate at
15 ◦ C. The second type of data used to study evaporation are
distillation data. Distillation data are very common and are
1 often the only data used to characterize oils. This is because
the data are crucial in operating refineries. Crudes may even
be priced on the basis of their distillation data. New proce-
0 dures to measure distillation data are very simple, fast and
0 10 20 30 repeatable [10]. Two separate ways of using the distillation
data will be tried, first a portion of the curve, and second,
Temperature (Celsius)
the entire distillation curve slope.
Fig. 8. Variation of evaporation rates with temperature. The empirically measured parameters at 15 ◦ C were cor-
related with both the slopes and the intercepts of the tem-
common oil components. The saturation concentration of perature equations. Full details of this correlation are given
water is in fact, about two orders-of-magnitude less than the in the literature [1]. The resulting equations are:
saturation concentration of volatile oil components such as
percentage evaporated = [B + 0.045(T − 15)] (7)
pentane. This further explains why oil has a boundary-layer
limitation higher than that of water. where B is the equation parameter at 15 ◦ C and T the tem-
Fig. 8 shows the composite of all evaporation rates ver- perature in degrees Celsius.
sus temperature. The evaporation rates are the coefficients Distillation data were directly correlated to the evapora-
of the logarithmic equation except for diesel and Bunker tion rates determined by experimentation. The distillation
C Light, for which they are the coefficients of the square data are available in two forms, percent evaporated at a given
root equation. Fig. 8 shows that the evaporation rates (used temperature value (as used here) and as temperature at which
here interchangeably with equation parameters) are linear a fixed amount of material is lost. The percentage distilled
with respect to temperature. This confirms the theoretical at each temperature was correlated with the equation pa-
approaches discussed in the introduction above. These show rameter (sometimes referred to here as the evaporation rate).
the evaporation rates with curves fit by linear regression. The Detailed correlation data are given in the literature [1]. The
curves for the light crudes, ASMB, Brent, Arabian Light, optimal point, or point at which the regression coefficient is
Statfjord, and Gullfaks appear to be parallel. The curves maximum, was found to be 180 ◦ C by using peak functions.
for gasoline, Terra Nova crude, diesel, and Bunker C Light The percent mass distilled at 180◦ was used to calculate the
have different slopes than those for crude oils, which may relationship between the distillation values and the equation
be due to the unique properties of these liquids. Gasoline parameters. The equations used were derived from correla-
evaporates at a rapid rate and is composed of only lighter tions of the data.
crude components. Terra Nova is a heavier crude with a The data from those oils that were better fitted with
large wax component. Diesel is a refined fuel with medium square root equations—diesel, Bunker C Light and FCC
to heavy components remaining. Bunker C Light is a refined Heavy Cycle—were separated and calculated separately.
residual with a small amount of diesel as a diluent. The Since there are only three data points, the reliability and
evaporation rates of the latter two products are best fit with accuracy are lower than for the other set.
square root equations rather than logarithmic equations. The equations derived from the regressions are as follows:
Further examination of the temperature behaviour of oil For oils that follow a logarithmic equation:
evaporation was conducted by determining the equations
by which the evaporation rates, or equation parameters, percentage evaporated = 0.165(%D) ln(t) (8)
change with temperature. A series of correlations was per-
formed, between the evaporation rates, by both percentage For oils that follow a square root equation:
√
and weight loss, using a linear equation. percentage evaporated = 0.0254(%D) t (9)
The resulting finding that unique equations may be needed
for each oil is a significant disadvantage to practical end where %D is the percentage (by weight) distilled at 180 ◦ C.
Table 2
Empirical equations for predicting evaporation
Oil type Equation Oil type Equation Oil type Equation
√ √
Adgo %Ev = (0.11 + 0.013T) t FCC Medium Cycle %Ev = (−0.16 + 0.013T) t Orimulsion plus water %Ev = (3 + 0.045T) ln(t)
√
Adgo—long term %Ev = (0.68 + 0.045T) ln(t) FCC-VGO %Ev = (2.5 + 0.013T) t Oseberg %Ev = (2.68 + 0.045T) ln(t)
Alberta Sweet Mixed Blend %Ev = (3.24 + 0.054T) ln(t) Federated %Ev = (3.47 + 0.045T) ln(t) Panuke %Ev = (7.12 + 0.045T) ln(t)

M.F. Fingas / Journal of Hazardous Materials 107 (2004) 27–36

Amauligak %Ev = (1.63 + 0.045T) ln(t) Federated (new—1999) %Ev = (3.45 + 0.045T) ln(t) Pitas Point %Ev = (7.04 + 0.045T) ln(t)
Amauligak—f24 %Ev = (1.91 + 0.045T) ln(t) Garden Banks 387 %Ev = (1.84 + 0.045T) ln(t) Platform Gail (Sockeye) %Ev = (1.68 + 0.045T) ln(t)
Arabian Medium %Ev = (1.89 + 0.045T) ln(t) Garden Banks 426 %Ev = (3.44 + 0.045T) ln(t) Platform Holly %Ev = (1.09 + 0.045T) ln(t)
Arabian Heavy %Ev = (1.31 + 0.045T) ln(t) Gasoline %Ev = (13.2 + 0.21T) ln(t) Platform Irene—long term %Ev = (0.74 + 0.045T) ln(t)
√
Arabian Heavy %Ev = (2.71 + 0.045T) ln(t) Genesis %Ev = (2.12 + 0.045T) ln(t) Platform Irene—short term %Ev = (−0.05 + 0.013T) t
Arabian Light %Ev = (2.52 + 0.037T) ln(t) Green Canyon Block 109 %Ev = (1.58 + 0.045T) ln(t) Point Arguello—co-mingled %Ev = (1.43 + 0.045T) ln(t)
Arabian Light %Ev = (3.41 + 0.045T) ln(t) Green Canyon Block 184 %Ev = (3.55 + 0.045T) ln(t) Point Arguello Heavy %Ev = (0.94 + 0.045T) ln(t)
Arabian Light (2001) %Ev = (2.4 + 0.045T) ln(t) Green Canyon Block 65 %Ev = (1.56 + 0.045T) ln(t) Point Arguello Light %Ev = (2.44 + 0.045T) ln(t)
ASMB—Standard #5 %Ev = (3.35 + 0.045T) ln(t) Greenplus Hydraulic Oil %Ev = (−0.68 + 0.045T) ln(t) Point Arguello Light—b %Ev = (2.3 + 0.045T) ln(t)
ASMB (offshore) %Ev = (2.2 + 0.045T) ln(t) Greenplus Hydraulic Oil %Ev = (−0.68 + 0.045T) ln(t) Polypropylene Tetramer %Ev = (0.25)(t) (at 15 ◦ C)
Av Gas 80 %Ev = (15.4 + 0.045T) ln(t) Gullfaks %Ev = (2.29 + 0.034T) ln(t) Port Hueneme %Ev = (0.3 + 0.045T) ln(t)
√
Avalon %Ev = (1.41 + 0.045T) ln(t) Heavy Reformate %Ev = (−0.17 + 0.013T) t Prudhoe Bay (old stock) %Ev = (1.69 + 0.045T) ln(t)
Avalon J-34 %Ev = (1.58 + 0.045T) ln(t) Hebron MD-4 %Ev = (1.01 + 0.045T) ln(t) Prudhoe Bay (new stock) %Ev = (2.37 + 0.045T) ln(t)
Aviation Gasoline 100 LL ln(%Ev) = (0.5 + 0.045T) ln(t) Heidrun %Ev = (1.95 + 0.045T) ln(t) Prudhoe stock b %Ev = (1.4 + 0.045T) ln(t)
Barrow Island %Ev = (4.67 + 0.045T) ln(t) Hibernia %Ev = (2.18 + 0.045T) ln(t) Rangely %Ev = (1.89 + 0.045T) ln(t)
√ √
BCF-24 %Ev = (1.08 + 0.045T) ln(t) High Viscosity Fuel Oil %Ev = (−0.12 + 0.013T) t Sahara Blend %Ev = (0.001 + 0.013T) t
√
Belridge Crude %Ev = (0.03 + 0.013T) t Hondo %Ev = (1.49 + 0.045T) ln(t) Sahara Blend (long term) %Ev = (1.09 + 0.045T) ln(t)
Bent Horn A-02 %Ev = (3.19 + 0.045T) ln(t) Hout %Ev = (2.29 + 0.045T) ln(t) Sakalin %Ev = (4.16 + 0.045T) ln(t)
√ √
Beta %Ev = (−0.08 + 0.013T) t IFO-180 %Ev = (−0.12 + 0.013T) t Santa Clara %Ev = (1.63 + 0.045T) ln(t)
Beta—long term %Ev = (0.29 + 0.045T) ln(t) IFO-30 (Svalbard) %Ev = (−0.04 + 0.045T) ln(t) Scotia Light %Ev = (6.87 + 0.045T) ln(t)
√ √
Boscan %Ev = (−0.15 + 0.013T) t IFO-300 (old Bunker C) %Ev = (−0.15 + 0.013T) t Scotia Light %Ev = (6.92 + 0.045T) ln(t)
Brent %Ev = (3.39 + 0.048T) ln(t) Iranian Heavy %Ev = (2.27 + 0.045T) ln(t) Ship Shoal Block 239 %Ev = (2.71 + 0.045T) ln(t)
√ √
Bunker C—Light (IFO-250) %Ev = (.0035 + 0.0026T) t Jet A1 %Ev = (0.59 + 0.013T) t Ship Shoal Block 269 %Ev = (3.37 + 0.045T) ln(t)
Bunker C—long term %Ev = (−0.21 + 0.045T) ln(t) Jet Fuel (Anch) %Ev = (7.19 + 0.045T) ln(t) Sockeye %Ev = (2.14 + 0.045T) ln(t)
√ √
Bunker C—short term) %Ev = (0.35 + 0.013T) t Jet Fuel (Anch) short term %Ev = (1.06 + 0.013T) t Sockeye Co-mingled %Ev = (1.38 + 0.045T) ln(t)
√
Bunker C Anchorage %Ev = (−0.13 + 0.013T) t Issungnak %Ev = (1.56 + 0.045T) ln(t) Sockeye Sour %Ev = (1.32 + 0.045T) ln(t)
Bunker C Anchorage (long term) %Ev = (0.31 + 0.045T) ln(t) Isthmus %Ev = (2.48 + 0.045T) ln(t) Sockeye Sweet %Ev = (2.39 + 0.045T) ln(t)
√
California API 11 %Ev = (−0.13 + 0.013T) t Jet 40 Fuel %Ev = (8.96 + 0.045T) ln(t) South Louisiana %Ev = (2.39 + 0.045T) ln(t)
√ √
California API 15 %Ev = (−0.14 + 0.013T) t Jet A1 %Ev = (.59 + 0.013T) t South Pass Block 60 %Ev = (2.91 + 0.045T) ln(t)

33
 34
Table 2 (Continued )
Oil type Equation Oil type Equation Oil type Equation

Cano Limon %Ev = (1.71 + 0.045T) ln(t) Jet Fuel (Anch) %Ev = (7.19 + 0.045T) ln(t) South Pass Block 67 %Ev = (2.17 + 0.045T) ln(t)
√
Canola Oil Little Jet Fuel (Anch) short term %Ev = (1.06 + 0.013T) t South Pass Block 93 %Ev = (1.5 + 0.045T) ln(t)
Carpenteria %Ev = (1.68 + 0.045T) ln(t) Komineft %Ev = (2.73 + 0.045T) ln(t) South Timbalier Block 130 %Ev = (2.77 + 0.045T) ln(t)
√
Cat cracking feed %Ev = (−0.18 + 0.013T) t Lago %Ev = (1.13 + 0.045T) ln(t) Soybean oil Little
Chavyo %Ev = (3.52 + 0.045T) ln(t) Lago Treco %Ev = (1.12 + 0.045T) ln(t) Statfjord %Ev = (2.67 + 0.06T) ln(t)

M.F. Fingas / Journal of Hazardous Materials 107 (2004) 27–36

√ √
Combined oil/gas %Ev = (−0.08 + 0.013T) t Lucula %Ev = (2.17 + 0.045T) ln(t) Sumatran Heavy %Ev = (−0.11 + 0.013T) t
Compressor Lube Oil—new %Ev = (−0.68 + 0.045T) ln(t) Main Pass Block 306 %Ev = (2.86 + 0.045T) ln(t) Sumatran Light %Ev = (0.96 + 0.045T) ln(t)
√
Cook Inlet—Granite Point %Ev = (4.54 + 0.045T) ln(t) Main Pass Block 37 %Ev = (3.04 + 0.045T) ln(t) Taching %Ev = (−0.11 + 0.013T) t
Cook Inlet—Swanson River %Ev = (3.58 + 0.045T) ln(t) Malongo %Ev = (1.67 + 0.045T) ln(t) Takula %Ev = (1.95 + 0.045T) ln(t)
Cook Inlet Trading Bay %Ev = (3.15 + 0.045T) ln(t) Marinus Turbine Oil %Ev = (−0.68 + 0.045T) ln(t) Tapis %Ev = (3.04 + 0.045T) ln(t)
√
Corrosion Inhibitor Solvent %Ev = (−0.02 + 0.013T) t Marinus Valve Oil %Ev = (−0.68 + 0.045T) ln(t) Tchatamba Crude %Ev = (3.8 + 0.045T) ln(t)
Crude Castor oil Little Mars TLP %Ev = (2.18 + 0.045T) ln(t) Terra Nova %Ev = (1.36 + 0.045T) ln(t)
√
Cusiana %Ev = (3.39 + 0.045T) ln(t) Maui %Ev = (−0.14 + 0.013T) t Terresso 150 %Ev = (−0.68 + 0.045T) ln(t)
Delta West Block 97 %Ev = (6.57 + 0.045T) ln(t) Maya %Ev = (1.38 + 0.045T) ln(t) Terresso 220 %Ev = (−0.66 + 0.045T) ln(t)
Diesel Anchorage—Long %Ev = (4.54 + 0.045T) ln(t) Mayan crude %Ev = (1.45 + 0.045T) ln(t) Terresso 46 Industrial oil %Ev = (−0.67 + 0.045T) ln(t)
√
Diesel Anchorage—Short %Ev = (0.51 + 0.013T) t Mississippi Canyon Block 72 %Ev = (2.15 + 0.045T) ln(t) Thevenard Island %Ev = (5.74 + 0.045T) ln(t)
Diesel—long term %Ev = (5.8 + 0.045T) ln(t) Mississippi Canyon Block 194 %Ev = (2.62 + 0.045T) ln(t) Turbine Oil STO 90 %Ev = (−0.68 + 0.045T) ln(t)
√
Diesel Mobile 1997 %Ev = (0.03 + 0.013T) t Mississippi Canyon Block 807 %Ev = (2.05 + 0.045T) ln(t) Turbine Oil STO 120 %Ev = (−0.68 + 0.045T) ln(t)
√ √
Diesel (regular stock) %Ev = (.31 + 0.018T) t Nektoralik %Ev = (0.62 + 0.045T) ln(t) Udang %Ev = (−0.14 + 0.013T) t
Diesel fuel—Southern—long term %Ev = (2.18 + 0.045T) ln(t) Neptune Spar (Viosca Knoll 826) %Ev = (3.75 + 0.045T) ln(t) Udang (long term) %Ev = (0.06 + 0.045T) ln(t)
√
Diesel fuel—Southern—short term %Ev = (−0.02 + 0.013T) t Nerlerk %Ev = (2.01 + 0.045T) ln(t) Vasconia %Ev = (0.84 + 0.045T) ln(t)
√
Diesel Mobile 1997 long-term %Ev = (−0.02 + 0.013T) t Ninian %Ev = (2.65 + 0.045T) ln(t) Viosca Knoll Block 826 %Ev = (2.04 + 0.045T) ln(t)
Dos Cuadros %Ev = (1.88 + 0.045T) ln(t) Norman Wells %Ev = (3.11 + 0.045T) ln(t) Viosca Knoll Block 990 %Ev = (3.16 + 0.045T) ln(t)
√
Ekofisk %Ev = (4.92 + 0.045T) ln(t) North Slope—Middle Pipeline %Ev = (2.64 + 0.045T) ln(t) Voltesso 35 %Ev = (−0.18 + 0.013T) t
Empire Crude %Ev = (2.21 + 0.045T) ln(t) North Slope—Northern Pipeline %Ev = (2.64 + 0.045T) ln(t) Waxy Light and Heavy %Ev = (1.52 + 0.045T) ln(t)
√
Endicott %Ev = (0.9 + 0.045T) ln(t) North Slope—Southern Pipeline %Ev = (2.47 + 0.045T) ln(t) West Delta Block 30 w/water %Ev = (−0.04 + 0.013T) t
Esso Spartan EP-680 Industrial Oil %Ev = (−0.66 + 0.045T) ln(t) Nugini %Ev = (1.64 + 0.045T) ln(t) West Texas Intermediate %Ev = (2.77 + 0.045T) ln(t)
Eugene Island 224-condensate %Ev = (9.53 + 0.045T) ln(t) Odoptu %Ev = (4.27 + 0.045T) ln(t) West Texas Intermediate %Ev = (3.08 + 0.045T) ln(t)
Eugene Island Block 32 %Ev = (0.77 + 0.045T) ln(t) Olive Oil Little West Texas Sour %Ev = (2.57 + 0.045T) ln(t)
Eugene Island Block 43 %Ev = (1.57 + 0.045T) ln(t) Oriente %Ev = (1.32 + 0.045T) ln(t) White Rose %Ev = (1.44 + 0.045T) ln(t)
Evendell %Ev = (3.38 + 0.045T) ln(t) Oriente %Ev = (1.57 + 0.045T) ln(t) Zaire %Ev = (1.36 + 0.045T) ln(t)
√
FCC Heavy Cycle %Ev = (.17 + 0.013T) t Orimulsion 400—dewater %Ev = (3.6) ln(t) (at 15 ◦ C)
√
FCC Light %Ev = (−0.17 + 0.013T) t
 M.F. Fingas / Journal of Hazardous Materials 107 (2004) 27–36 35

These equations can be combined with the equations gen- 4. Conclusions

erated in previous work [1] to account for the temperature
variations: Oil evaporation is not strictly boundary-layer regulated.
For oils that follow a logarithmic equation: The results of the following experimental series have shown
the lack of boundary-layer regulation.
percentage evaporated = [0.165(%D) (1) A study of the evaporation rate of several oils with in-
+ 0.045(T − 15)] ln(t) (10) creasing wind speed shows that the evaporation rate does
not change significantly except for the initial step over
For oils that follow a square root equation: the 0-level wind. Water, known to be boundary-layer
regulated, does show a significant increase with wind
√ speed, U (Ux , where x varies from 0.5 to 0.78, depend-
percentage evaporated = [0.0254(%D) + 0.01(T −15)] t
ing on the turbulence level).
(11) (2) Increasing area does not significantly change the oil
evaporation rate. This is directly contrary to the predic-
where %D is the percentage (by weight) distilled at 180 ◦ C. tion resulting from boundary-layer regulation.
In addition, a large number of experiments were per- (3) The volume or mass of oil evaporating correlates with
formed on oils to directly measure their evaporation curves. the evaporation rate. This is a strong indicator of the
The empirical equations that result are given in Table 2. lack of boundary-layer regulation because with water,
Since the equations described above require only time volume (rather than area) and rate do not correlate.
and temperature (or at the very worst, the percentage of oil (4) Evaporation of pure hydrocarbons with and without
distilled at 180 ◦ C), it is relatively simple to apply these wind (turbulence) shows that compounds larger than
forms of equations. They can also be applied in models as nonane and decane are not boundary-layer regulated.
increments where t, the time, becomes the total time and Most oil and hydrocarbon products consist of com-
the previous evaporation is subtracted, e.g., if one was mod- pounds larger than these two and thus would not be
eling the evaporation of Alberta Sweet Mixed Blend oil expected to be boundary-layer regulated.
in the time step from 12 to 15 h. The equation is (from
The fact that oil evaporation is not strictly boundary-layer
Table 2):
regulated implies that a simplistic evaporation equation will
suffice to describe the process. The following factors do not
percentage evaporation = (3.24 + 0.054T) ln(t) (12)
require consideration: wind velocity, turbulence level, area,
thickness, and scale size. The factors important to evapora-
Substituting for the temperature of 15 ◦ C and with a time tion include time and temperature.
of 12 h or 720 min, we get a percentage of 26.65. With 18 h, The equation parameters found experimentally for the
we get a percentage of 27.72 with a difference of 1.07%, evaporation of oils can be related to commonly available dis-
the amount evaporated in the interval between 15 and 18 h. tillation data for the oil. Specifically, it has been found that
The variation of evaporation is illustrated in Fig. 9, which the distillation percentage at 180 ◦ C correlates well with the
shows the evaporation of two oils, diesel fuel and North equation parameters. Regression coefficients (r2 ) range from
Slope Crude, at two temperatures. 0.74 to 0.98, depending on the type of equation and the se-
lection of data [14]. Relationships have been developed that
allow evaporation equations to be calculated directly from
70
distillation data.
For oils that follow a logarithmic equation:
60

percentage evaporated = [0.165(%D)+0.045(T −15)] ln(t)

Diesel at 20 degrees
Percent Evaporated

50
Diesel at 5 degrees (13)
40 For oils that follow a square root equation:
North Slope at 20 degrees √
30 percentage evaporated = [0.0254(%D) + 0.01(T − 15)] t
(14)
20
North Slope at 5 degrees
10
References
0
0 20 40 60 80 100 120 [1] M.F. Fingas, in: Proceedings of the 1999 International Oil Spill
Time in Hours Conference, American Petroleum Institute, Washington, DC, 1999,
pp. 281–287.
Fig. 9. Comparison of evaporation curves. [2] M.F. Fingas, J. Hazard. Mater. 42 (1995) 157.
36 M.F. Fingas / Journal of Hazardous Materials 107 (2004) 27–36

[3] F.E. Jones, Evaporation of Water, Lewis Publishers, Chelsea, MI, [10] P.S. Jokuty, S. Whiticar, Z. Wang, M.F. Fingas, B. Fieldhouse,
1992. P. Lambert, J. Mullin, Properties of Crude Oils and Oil Prod-
[4] W. Brutsaert, Evaporation into the Atmosphere, Reidel Publishing ucts, Environment Canada Manuscript Report EE-165, Ottawa, Ont.,
Company, Dordrecht, Holland, 1982. 1999.
[5] W. Stiver, D. Mackay, Environ. Sci. Technol. 18 (1984) 834. [11] Z. Wang, M. Fingas, K. Li, J. Chromatogr. Sci. 32 (1994) 361.
[6] J.L. Monteith, M.H. Unsworth, Principles of Environmental Physics, [12] M.F. Fingas, J. Hazard. Mater. 56 (1997) 227.
Hodder and Stoughton, London, 1990. [13] Ullmann Encyclopedia, Ullmann Publishing, Hamburg, 1989–1995.
[7] O.G. Sutton, Proc. R. Soc. London, Ser. A 146 (1934) 701. [14] M.F. Fingas, in: Proceedings of the Nineteenth Arctic and Marine
[8] D. Mackay, R.S. Matsugu, Can. J. Chem. Eng. 51 (1973) 434. Oil Spill Program Technical Seminar, Environment Canada, Ottawa,
[9] M.F. Fingas, J. Hazard. Mater. 57 (1998) 41. Ont., 1996, pp. 29–72.