Kinetics of Precipitation of Surfactants. I.

Anionic Surfactants
with Calcium and with Cationic Surfactants
Cheryl H. Rodriguez, Lori H. Lowery, John F. Scamehorn*, and Jeffrey H. Harwell
Institute for Applied Surfactant Research, University of Oklahoma, Norman, Oklahoma 73019

ABSTRACT: Isoperibol calorimetry was used to measure the

rates of precipitation for aqueous solutions of several anionic
surfactants with calcium and of anionic and cationic surfactants. A monomer concentration-dependent supersaturation
ratio was used to describe the relative rates of precipitation
for the surfactant systems studied. This supersaturation ratio
allows for the relative rates of precipitation of any surfactant
solution to be compared whether micelles are present in solution or not. In general, as the supersaturation ratio increases,
the rate of precipitation decreases and the induction time
decreases, both above and below the critical micelle concentration (CMC). The rate of precipitation of sodium dodecyl sulfate (SDS) with dodecyl pyridinium chloride is much slower
than the rate of precipitation of the anionic surfactants with
calcium for similar supersaturation ratios. The rate of precipitation of SDS with calcium is slightly faster than the rate of precipitation of sodium octyl benzene sulfonate for similar supersaturation ratios. Studies of precipitate crystals, conducted
using image analysis, showed that size and shape depended
on the initial supersaturation, the precipitating surfactant molecule, and the extent of aging (until an equilibrium size and
shape was reached). Also, differences in the appearance of
crystals formed from solutions above and below the CMC
were observed. These were most likely due to the difference
in supersaturation of these solutions. The crystals formed due
to precipitation of SDS with calcium at a concentration above
the CMC formed flat trapezoidal, rhombic, and hexagonal
shapes. These aged into clusters by 1 wk. For a solution that
was precipitated at concentrations beginning below the CMC,
the crystals began as elongated and rhombic flat plates and
aged into trapezoidal, rhombic, and needle-like structures.
Paper no. S1123 in JSD 4, 114 (January 2001).
KEY WORDS: Anionic surfactants, calorimetry, cationic surfactants, kinetics of precipitation, surfactant precipitation.

An important characteristic of ionic surfactants, which can

inhibit their use in many applications, is their tendency to
precipitate from aqueous solutions. In detergency, surfactants are used to aid in the removal of oily soil and in the
suspension of solids in a washing liquid. In hard water
(water containing multivalent cations), anionic surfactants
tend to precipitate, and then they are no longer available to
*To whom correspondence should be addressed at the University of
Oklahoma, The Energy Center, 100 E. Boyd, Norman, OK 73019-1004.
E-mail: scamehor@ou.edu
Copyright 2001 by AOCS Press

participate in the cleaning process. Builders are commonly

used to prevent precipitation in anionic surfactant detergency systems (1,2). In all-purpose laundry detergents,
washing and softening the clothes simultaneously in the
wash cycle is a goal. Anionic surfactants are generally necessary for the removal of some soil types, and cationic surfactants for fabric softening. These dissimilar surfactants can
precipitate together from solution leading to a residue and
a decrease in cleaning action. Other applications where
surfactant precipitation can be detrimental are surfactantbased separations (3) and enhanced oil recovery (4). Surfactant precipitation can be advantageous in some applications such as surfactant recovery by crystallization (5,6).
Because of the tendency for ionic surfactants to precipitate,
surfactant precipitation behavior and the ability to manipulate this behavior are extremely important.
Numerous research efforts have studied the thermodynamics of surfactant precipitation (e.g., Krafft temperature
or hardness tolerance studies) (4); but few studies have
been directed at understanding the kinetics of the precipitation reaction. Precipitation kinetics, not thermodynamics,
can dictate the influence of precipitation in processes which
utilize surfactants. For example in laundry detergency,
whether surfactants actually precipitate in a wash cycle (e.g.,
20 min) is important, not whether precipitation occurs at
equilibrium (which may not occur for much longer times).
In this work, the rates of precipitation of single-component anionic surfactant/calcium and anionic surfactant/
cationic surfactant systems are measured using calorimetry.
The effect of micelles on the precipitation rate is deduced
from experiments above and below the critical micelle concentration (CMC). Image analyses are used to characterize
the crystals formed in these systems. This is the first of a series of three papers. In the second paper, rates of precipitation of well-defined mixtures of anionic surfactants are presented. In the third paper, atomic force microscope scans
of these single components and mixed systems are presented and used to interpret rate results.

EXPERIMENTAL PROCEDURES
Materials. The two anionic surfactants used in this study
were sodium dodecyl sulfate (SDS) and sodium octyl benzene sulfonate (SOBS). The cationic surfactant used was

Journal of Surfactants and Detergents, Vol. 4, No. 1 (January 2001)

C.H. RODRIGUEZ ET AL.

dodecyl pyridinium chloride (DPC). Electrophoresis/highperformance liquid chromatography-grade SDS was at least
99% pure and was obtained from Fisher Scientific (Pittsburgh, PA). It was further purified by recrystallization from
water, then from methanol, followed by drying under a
vacuum at 30C. The SOBS was obtained from Aldrich (Milwaukee, WI) at a purity of 97%. The SOBS was recrystallized
first from methanol, and then from water. It was then rinsed
with cold methanol and dried under vacuum at 30C. Technical-grade DPC was purchased from Pfaltz & Bauer (Waterbury, CT). The DPC was purified by recrystallization from
an 80:20 mixture of petroleum ether/ethanol (Fisher Scientific). This was repeated three times, yielding white crystals when dried under vacuum at low heat. Reagent-grade
calcium chloride was obtained from Fisher Scientific and
was used as received. Water was double-deionized.
Methods. (i) Precipitation phase boundaries. For the SOBS/
CaCl2 system, a series of solutions was made for each surfactant concentration with varying CaCl2 concentrations. The
temperature of these solutions was first lowered to nearly
0C for 24 h to force precipitation (711). Surfactant solutions can stay supersaturated for long periods of time
(12,13) resulting in nonequilibrium apparent hardness tolerances. The temperature was then held constant at 30C
for 4 d while gently shaking the samples daily to ensure
equilibrium (8). For each series of solutions of increasing
calcium concentration, some samples would still contain
crystals at the end of 4 d, whereas others would have become clear. The average of the highest calcium concentration yielding clear solutions and the lowest calcium concentration yielding turbid solutions is considered the hardness
tolerance at that surfactant concentration (on the phase
boundary). The SDS/CaCl2 precipitation phase boundary
was obtained from Stellner and Scamehorn (8). The
SDS/DPC precipitation phase boundary was obtained from
Stellner et al. (10).
(ii) Calorimetry. A Tronac (Orem, UT) model 458/558
calorimeter was used in isoperibol mode to measure the
heat of reaction as a function of time. Isoperibol calorimetry is a nearly adiabatic process. However, a small amount of
heat is transferred from the reaction vessel to the water bath
and is added to the reaction vessel by the stirrer and thermistors. Over short lengths of time, this heat leak can be
modeled as a linear function of the reaction vessel temperature. The temperature of the water bath at 30C can be
maintained within 0.025C using a Tronac PTC-41 temperature controller. A diagram of the reaction vessel setup is
shown in Scheme 1. The rates of precipitation of SDS or
SOBS with calcium and of mixtures of SDS with DPC were
measured. Approximately 48 g of anionic surfactant solution was placed in the reaction vessel, and approximately 2
g of concentrated calcium chloride solution or concentrated DPC solution was injected into a soft glass ampoule,
which was then sealed with a Microflame (Foxboro, MA) butane torch and placed in the ampoule holder-stirrer. The reaction vessel was clamped into place, and the system allowed
to equilibrate with the water bath temperature. The am-

SCHEME 1
poule could then be broken with the hammer to allow instantaneous mixing of its contents with the solution in the
reaction vessel by vigorous stirring. A program written by
Lopata (14) to run the calorimeter for buret titration was
modified to run the calorimeter with the ampoule method
(15).
(iii) Heats of reaction. When an anionic surfactant solution
is mixed with a counterion, the apparent experimental heat
of reaction is the total heat released from the precipitation
reaction, dilution of the ampoule contents, dilution of the
reaction vessel contents, breaking of the ampoule, and in
some cases, micelle formation and dissociation. Additional
experiments must be done to determine these extraneous
heats. The heat of breaking the ampoule is measured by
breaking an ampoule containing water into the reaction vessel containing water. Heat of dilution of the ampoule solution is measured by breaking either the CaCl2 or DPC (below
the CMC) ampoule solution into water. The heat of demicellization of DPC is found by breaking an ampoule of micellar
DPC solution into water, and then subtracting the heat of dilution from the total heat obtained. The heat of dilution of
the reaction vessel solution is measured by breaking water
into the appropriate reaction vessel solution below the CMC.
The heat of micellization upon introduction of CaCl2 is
measured by breaking a CaCl2 ampoule solution into a surfactant solution that is above the CMC but outside the precipitation region. This results in the formation of additional
micelles without precipitation. Heat of mixed micelle formation occurs in the SDS/DPC system and can be measured by
breaking a DPC ampoule solution into an SDS solution that
is above the CMC and outside of the precipitation phase
boundary. Heats of surfactant precipitation are measured by
breaking an ampoule containing a counterion or cationic
surfactant solution into the reaction vessel surfactant solution. The decrease in the CMC upon addition of a counterion is calculated using a model presented by Stellner and
Scamehorn (9). A Microscribe 450 (San Jose, CA) chart
recorder was used to plot the temperature difference be-

Journal of Surfactants and Detergents, Vol. 4, No. 1 (January 2001)

ANIONIC SURFACTANT PRECIPITATION. PART I

tween the reaction vessel and the water bath, in voltage, vs.
time.
For each run, the average heat capacity, Cp, is obtained
by adding a known amount of heat for a known amount of
time before and after the reaction. Then, by using this heat
capacity, the overall heat occurring during a reaction, QT,
can be obtained:
QT = Cp T

[1]

where T is the total temperature change during a reaction

minus the temperature change due to the heat leak from
the reaction vessel. For systems that start above the CMC,
demicellization occurs as the precipitation reaction proceeds. The heat associated with this process is subtracted
from QT as a function of time. The concentration of micelles in solution can be found at each point during a
reaction using the same model as that by Stellner and
Scamehorn (9) along with a model of the precipitation reaction pathway (10,16,17). The heat due only to precipitation at each point along the reaction pathway can thus be
separated from all of the extraneous heats associated with a
calorimeter run.
(iv) Conductivity. Conductivity can be used to follow the
precipitation reactions of some of the systems studied and
to verify the calorimetric results qualitatively. As precipitation occurs, ions are removed from the system, which decreases the total conductivity of the system. An SDS solution
was placed in a side arm flask kept at 30C. The solution was
stirred and nitrogen was blown into the flask to reduce the
absorption of carbon dioxide into the solution (18). A concentrated, 30C CaCl2 or DPC solution was quickly added
to the SDS solution. An Orion (Beverley, MA) conductivity
meter model 1-1 with a total conductivity probe was used to
measure the conductivity of the solution as a function of
time during the precipitation reaction.
(v) Optical analysis. A qualitative study of the crystal sizes
and shapes at various times after mixing was conducted by
image analysis. The reactants were mixed at 30C. After mixing the reactants, crystal size and shapes were analyzed at 4
min of vigorous stirring via image analysis. The crystals were
then allowed to age for 1 d and 1 wk before analysis was done
with image analysis. A Nikon (Tokyo, Japan) microscope interfaced with a Gateway (North Sioux City, SD) computer
utilizing Optimas (Edmonds, WA) software was used.

FIG. 1. Initial conditions used for the precipitation reactions of sodium

dodecyl sulfate (SDS) with CaCl2 in relation to the SDS/CaCl2 precipitation phase boundary (from Ref. 8). Prec., precipitation.

the CMC. Below the CMC, as the total anionic surfactant

concentration increases, the concentration of calcium required for precipitation to occur decreases. The solubility
product relationship that describes this precipitation for a
monovalent surfactant with calcium is shown in Equation 2:
KSP = [S]2 [Ca2+] fS2 fCa

[2]

where KSP is the activity-based solubility product, [S ] is the

anionic surfactant concentration, and [Ca2+] is the calcium
ion concentration. The parameters fS and fCa are the activity coefficients of the surfactant and calcium, respectively.
Along the precipitation phase boundary and above the
CMC, a simultaneous equilibrium exists between the surfac-

RESULTS AND DISCUSSION

Precipitation phase boundaries. Establishment of precipitation
phase boundaries is necessary to determine the region
where precipitation will occur for each system, and they can
be used to determine the theoretical equilibrium condition
for each precipitation reaction (i.e., solution composition
when precipitation is complete). The precipitation phase
boundaries for SDS and SOBS with calcium are shown in
Figures 1 (8) and 2. Both experimental and theoretical
model results are shown on these phase boundaries. It is
well-known that hardness tolerance reaches a minimum at

FIG. 2.

Initial conditions used for the precipitation reactions of

sodium octyl benzene sulfonate (SOBS) with CaCl2 in relation to the
SOBS/CaCl2 precipitation phase boundary.

Journal of Surfactants and Detergents, Vol. 4, No. 1 (January 2001)

C.H. RODRIGUEZ ET AL.

tion. The parameter z is the ion valence, and the parameter

a is an empirical value based on the diameter of the ion. Values for A and B are tabulated (24), as well as for the constant a (24,25). The parameter I is the ionic strength of the
solution, defined as:
I = 0.5 ci zi2 = [NaS] + 3[CaCl2]

SCHEME 2
tant as monomer, in micelles, and in precipitate. This equilibrium for an anionic surfactant with Ca2+ is illustrated in
Scheme 2. There are also sodium ions in solution due to the
dissociation of the surfactant salt, which for clarity are not
shown in Scheme 2. Sodium was found not to precipitate
the anionic surfactants studied at the concentration levels
employed in this paper owing to a lower KSP with calcium
than with sodium (on the order of 1010 as opposed to
104). Precipitation occurs above the CMC when the surfactant monomer concentration, along with the unbound
counterion concentration, multiplied by appropriate activity coefficients exceeds the solubility product as shown in
Equation 3 (19):
KSP = ([S]mon)2 [Ca2+]un fS2 fCa

[3]

where [S]mon is the anionic surfactant monomer concentration and [Ca2+]un is the unbound calcium concentration
(calcium not bound to micelles).
Micelles act as sequestering agents for the calcium ions
(i.e., the calcium ions bind to the micelles). As more surfactant is added to the system, the additional surfactant tends
to form more micelles, reducing [Ca2+]un even further. This
results in a minimum in the precipitation phase boundary
at the CMC. From the precipitation phase boundaries in
Figures 1 and 2, the CMC for SDS is 0.0067 M (14) and for
SOBS is 0.012 M. These values are consistent with the CMC
values obtained from surface tension measurements: 0.0072
M for SDS and 0.012 M for SOBS (20). Several reviews have
discussed the thermodynamics (e.g., phase diagrams or
Krafft temperatures) of anionic surfactant precipitation
with metal ions (4,21,22).
The parameters fS and fCa in Equations 2 and 3 can be
described by the extended Debye-Huckel expression (23):
log f = A (z)2 I 0.5/(1 + B a I 0.5) 0.3 I

[4]

where f can be fS or fCa and the constants A and B are dependent on the solvent and the temperature of the solu-

[5]

where ci is the total concentration of ion i in solution, zi is

the valence of ion i, [NaS] is the total anionic surfactant
concentration in solution, and [CaCl2] is the total CaCl2
concentration in solution. Below the CMC, the surfactant
monomer can be treated as a simple, strong electrolyte in
the calculation of the ionic strength and activity coefficients. However, there is no universally accepted practice
for the calculation of activity coefficients in a micellar solution. In this study, the anionic surfactants and the CaCl2 in
the micellar solutions are treated as simple, strong electrolytes, as our group has done previously (11). Several
other methods have been proposed. One considers the micelles as a separate species in solution, contributing only a
portion of the actual micelle valence (a shielded micelle)
(26). Another treats the micelles as a separate phase, which
therefore does not contribute to the ionic strength of the
aqueous solution (27). Burchfield and Woolley (26) also
discuss the work of other researchers who have treated the
surfactant in solution as a simple, strong electrolyte as we
have done here.
If solution concentrations are not large enough to satisfy
the solubility product (Eq. 3) and the surfactant concentration lies above the eutectic point, or CMC, the solution will
contain monomer and micelles but will not precipitate at
equilibrium. If the solubility product is not satisfied (Eq. 2)
and the surfactant concentration is below the CMC, the surfactant will be present only as monomer. If a solution contains a concentration of surfactant and calcium that exceeds
the solubility product, then that solution is supersaturated
and, theoretically, will precipitate until the solution concentration lies somewhere along the precipitation phase
boundary (solubility product is exactly satisfied; Eq. 2 or 3
is valid). The supersaturated points inside the precipitation
phase boundaries in Figures 1 and 2 are the initial conditions for the precipitation reactions studied here.
The equilibrium relationships that exist in an anionic/
cationic surfactant solution are shown schematically in
Scheme 3 for a micellar solution in which precipitation has
occurred. In this case, both the sodium and chloride ions
can be bound onto the micelles, but for clarity they are not
shown in this figure. Cationic and anionic surfactant can be
present as monomer and in mixed micelles. Above the solubility product, anionic and cationic surfactant monomers
combine to form precipitate. The solubility product relationship for SDS-DPC precipitation is:
KSP = [S]mon [DP+]mon f2
+

[6]

where [DP ]mon is the monomer concentration of the dodecyl pyridinium ion and f is the mean activity coefficient
of the solution (10). Below the CMC, all of the surfactant is

Journal of Surfactants and Detergents, Vol. 4, No. 1 (January 2001)

ANIONIC SURFACTANT PRECIPITATION. PART I

SCHEME 3
present as monomer; and as the concentration of one of the
surfactants increases, a lesser concentration of the oppositecharged surfactant is needed to cause precipitate to form.
Above the CMC, there are an SDS-rich region and a DPCrich region as seen in Figure 3 from Stellner et al. (10).
Along the SDS-rich branch of the precipitation phase
boundary, as SDS-rich mixed micelles form, they act as sequestering agents for the surfactant monomers, and more
cationic surfactant is needed for precipitation. This same
trend is true for the DPC-rich branch. The points lying in-

side the precipitation phase boundary in Figure 3 represent

the initial conditions used to measure the kinetics of
SDS/DPC precipitation reactions. Owing in part to the
complex surfactant mixtures in detergent formulations, researchers have experimentally determined precipitation
phase boundaries for mixtures of anionic/cationic surfactants. Scowen and Leja (28) determined the phase boundaries for various alkyl quaternary ammonium bromides with
different anionic surfactants.
The initial conditions shown in Figures 13 are supersaturated solutions. Supersaturation is a measure of the excess
concentration of the reactants above the equilibrium solubility concentrations (2931). A supersaturated solution can
be prepared by cooling the solution below its saturation
temperature, evaporating the solvent in the solution, or by
direct mixing of two reactants which are of higher concentration than their equilibrium solubility (which is done in
this study) (29). The degree of supersaturation and the
presence of foreign materials, or seed crystals, affect the
ability of the supersaturated solution both to form nuclei
and to grow crystals. In this paper, a supersaturation ratio
(So) is used to describe the excess concentration of reactants above the precipitation phase boundary for an anionic
surfactant precipitating with Ca2+, as follows:
So = ([Ca2+]un ([S]mon)2 fCa fS2/KSP)1/3

[7]

Equation 7 only takes into consideration the monomer

surfactant concentration and the unbound calcium concentration. If the initial solution conditions exactly satisfy the
solubility product, then So = 1. Therefore, the more that So
exceeds unity, the more supersaturated the solution becomes. The initial presence of micelles removes the surfactant and calcium from the monomer phase. The supersaturation ratio for the SDS/DPC system is defined as:
So = ([DP+]mon ([S]mon) (f)2/KSP)1/2

[8]

Equation 8 is a special case of the general supersaturation

ratio for monovalent reactants (32).
The supersaturated solutions depicted in Figures 1 and 2
will theoretically precipitate until the surfactant and calcium concentrations in solution reach some point on the
precipitation phase boundary (solubility product is satisfied, or So = 1). Since the stoichiometry of the precipitate is
invariant, for defined initial conditions, as precipitation occurs the concentration of each precipitating species can be
calculated from material balances (defining the reaction
pathway). For calcium/SDS, the reaction pathway is described by Equations 911 (10,16,17):

FIG. 3. Initial conditions used for the precipitation reactions of SDS

with dodecyl pyridinium chloride (DPC) in relation to the SDS/DPC

precipitation phase boundary (from Ref. 10). For other abbreviation
see Figure 1.

2[NaS] = 2[S]unr + 2[S]ppt

[9]

[CaCl2] = [Ca2+]unr + [Ca2+]ppt

[10]

2[S]ppt = [Ca2+]ppt

[11]

where [NaS] is the initial surfactant concentration, [CaCl2]

is the initial CaCl2 concentration, [S]unr is the unreacted
(or unprecipitated) surfactant ion concentration, [S]ppt is

Journal of Surfactants and Detergents, Vol. 4, No. 1 (January 2001)

C.H. RODRIGUEZ ET AL.

the precipitated surfactant concentration, and [Ca2+]unr is

the unreacted calcium concentration. Equations 9 and 10
are material balances on surfactant and calcium, respectively. Equation 11 defines the stoichiometry of the precipitation reaction. Note that [S]unr includes surfactant in
both monomeric and micellar form and [Ca2+]unr includes
unbound and bound calcium. Combining these equations
gives the reaction pathway constant, D :
D = 2[NaS] [CaCl2] = 2[S]unr [Ca2+]unr

[12]

The constant D is calculated for the initial conditions. Then,

[S]unr is assumed to decrease as the precipitation reaction
proceeds. From Equation 12, [Ca2+]unr can be calculated
for every value of [S]unr given. This continues until the reaction pathway intersects the appropriate precipitation
phase boundary (either Eq. 2 or Eq. 3 is satisfied depending on whether micelles are present on the phase boundary). At each point along the reaction pathway, the unreacted component concentrations can be put into the
pseudo-phase separation model to calculate the concentration of surfactant in the micelles at that point. This allows
for the subtraction of the heat of demicellization from the
overall process heat as the surfactant precipitates. The reaction pathways for SOBS precipitating with calcium are
shown in Figure 4. Similar reaction pathways for SDS precipitating with calcium can be depicted.
The corresponding material balances and reaction pathway equation for a precipitation reaction of a monovalent
anionic surfactant with a monovalent cationic surfactant are
as follows:
[NaS] = [S]unr + [S]ppt

[13]

[DPC]i = [DP+]unr + [DP+]ppt

[14]

[S]ppt = [DP+]ppt

[15]

D = [S]unr [DP+]unr

[16]

FIG. 4. Reaction pathways for SOBS precipitating with calcium at several initial solution compositions. The four sets of data show how the
composition varies from an initial value to the end, where it intersects
the theory line. For abbreviation see Figure 2.

where [DPC]i is the initial DPC concentration, [DP+]unr is

the unreacted DPC concentration, and [DP+]ppt is the precipitated DPC concentration.
Nucleation.There is extensive literature describing crystallization as a general phenomenon. Most of these articles
or books address precipitation of inorganic salts, but they
can be extrapolated to other precipitation systems. We will
briefly review some aspects which are relevant to the precipitation of surfactant salts.
The induction period is the period between the attainment of supersaturation and the onset of precipitation, as
determined visually or by the measurement of some appropriate physical property. A solution can remain supersaturated for a long time. Also, concentrations may have to be
some distance above the precipitation phase boundary before spontaneous precipitation will occur in real situations.
Many authors have described a metastable zone directly
above a precipitation phase boundary in which spontaneous
nucleation does not readily occur (33). Many factors, such
as thermal history, mechanical action and presence of solid
particulates, affect the metastable zone width (34). Nucleation can be either homogeneous (spontaneous), heterogeneous, or secondary. Homogeneous nucleation occurs
when the nuclei are made up of the precipitating components. Usually, large supersaturation ratios are required for
homogeneous nucleation to occur (33,35,36). One of the
few studies on the kinetics of surfactant precipitation found
that, for calcium laurate below the CMC, a supersaturation
ratio of 5 was required for homogeneous nucleation (35).
Subnuclei, or embryos, are thought to form and dissipate
constantly in supersaturated solutions, and an energy barrier must be overcome in order for a subnucleus to form a
critical nucleus (the smallest nucleus that can grow into a
crystal). A critical nucleus can form as a result of random
free energy or concentration fluctuations in local regions
of a solution (29,37). The structure of a critical nucleus is
thought to be either a tiny replica of the crystal it will form
or a diffuse body of ions, not yet in the rigid lattice of a crystal. Since very few ions are thought to form a critical nucleus, it is likely that the critical nucleus does not have the
characteristics of the bulk crystal (35).
Secondary nucleation occurs when nucleation sites are
present due to the loss of weak outer layers or weak outgrowths of the crystals of the precipitating species, due
mostly to collisions with other crystals or the reaction vessel
hardware (36,38). There is an adsorbed layer of solute on
the surface of a growing crystal (38), which can be easily
separated to form new nuclei.
Heterogeneous nucleation occurs when small particles
present in the system act as nucleation sites for the deposition of the precipitating components. The most active nuclei are probably in the range of 0.1 to 1.0 m (36). In this
work, the shattered glass of the ampoule, the stirring rod,
thermistors, and reaction vessel walls could act as nucleation sites. Thorough discussions of the thermodynamics of
nucleation are given in reviews, including Nyvlt et al. (34)
and Adamson (39).

Journal of Surfactants and Detergents, Vol. 4, No. 1 (January 2001)

ANIONIC SURFACTANT PRECIPITATION. PART I

Crystal growth. Crystal growth can be either diffusioncontrolled or surface reaction-controlled. In diffusioncontrolled growth, the rate-limiting step is the diffusion of
the precipitating components from the bulk to the crystal
surface. A model that describes the diffusion mechanism
has been used to describe crystal growth of inorganic salt
precipitation (40) and anionic surfactant precipitation with
calcium (35).
There are three possible mechanisms for surface reaction-controlled growth: mononuclear growth, polynuclear
growth, and growth due to a screw dislocation. A new layer
of a crystal is started when the precipitating components
nucleate onto the crystal surface. The first site to become occupied on the surface is a high-energy site and is more difficult to fill than the adjacent sites. If the precipitating components nucleate onto the surface of a crystal very slowly
compared to the growth of a layer, then each layer is completed before a new layer begins. This is called mononuclear
growth and is the most constrained growth mechanism. If
surface nucleation is fast, new layers begin before the old layers are complete. This is called polynuclear growth. A screw
dislocation is a surface dislocation that forms when a slip in
a crystal plane occurs, pushing up part of the surface layer.
This creates a lower-energy site adjacent to the raised layer,
removing the need for surface nucleation. The Frank mechanism (41) describes this mechanism in which the crystal is
formed from a single layer, spiraling upward as the precipitating species occupy the sites at the step of the screw dislocation. A crystal forming due to a screw dislocation can continue growing without inhibition until the supersaturation
of the system is satiated. The Frank mechanism is widely accepted and has been seen in several instances (13,42). The
spiral pattern that forms is in many instances representative
of the molecular pattern in the lattice (43), and the step
heights have been found to be some multiple of the unit cell
height (42,43). Atomic force microscopy has been used to
determine the surface structure of Ca(DS)2 crystals (where
DS = dodecyl sulfate anion) precipitated from a 0.020 M
SDS, 0.01 M CaCl2 solution, and spiral growth was found
(20). The surface structure of Ca(OBS)2 crystals was also
studied for the same concentrations of surfactant and calcium, and many steps representing differing heights were
found on the surface being scanned (20).
Calorimetry. The average heat of micellization upon the
introduction of CaCl2 to an anionic surfactant solution is
812 cal/mol of surfactant-forming micelles. The heat of
dilution of the CaCl2 ampoule solutions was found to be
423 cal/mol CaCl2. The heat of dilution of the DPC ampoule solutions was +275 cal/mol DPC. The heat of demicellization of the DPC ampoule solution was found to be
+1700 cal/mol demicellized DPC. The heats of reaction vessel solution dilution and ampoule breaking were found to
be negligible. The average heat of reaction for SDS precipitating with Ca2+ is 7,500 cal/mol surfactant precipitated,
for SOBS precipitating with Ca2+ is 4,600 cal/mol surfactant precipitated, and for SDS precipitating with DPC is
20,000 cal/mol precipitate formed.

FIG. 5. Precipitation rate curves of SDS with CaCl2 below the critical
micelle concentration (CMC) at 30C. So, supersaturation ratio; for
other abbreviation see Figure 1.
Figure 5 shows the rate of precipitation of SDS with calcium for several initial conditions below the CMC, and Figure 6 for initial conditions above the CMC. In both cases, as
the initial supersaturation ratio is decreased, the overall rate
of precipitation decreases. Also in both cases, a decrease in
the initial supersaturation ratio results in an increase in the
induction time. Above the CMC, the initial supersaturation
ratio is based on the monomer SDS concentration and unbound calcium concentration (Eq. 7). Figure 7 shows a comparison of the rates of precipitation for a range of supersaturation ratios both above and below the CMC. The general

FIG. 6. Precipitation rate curves of SDS with CaCl2 above the CMC at
30C. For abbreviations see Figures 1 and 5.

Journal of Surfactants and Detergents, Vol. 4, No. 1 (January 2001)

C.H. RODRIGUEZ ET AL.

FIG. 7. Precipitation rate curves of SDS with CaCl2 at 30C. For abbreviations see Figures 1 and 5.
trend is an increase in the precipitation rate as the initial supersaturation ratio is increased, regardless of whether micelles are present. In all of the kinetic data shown in this
paper, mole fraction surfactant precipitated is the amount
precipitated divided by the amount precipitated at equilibrium; hence, this variable will always approach unity at long
enough times. The effect of increasing the degree of supersaturation on the precipitation rate of SOBS precipitating
with calcium is shown in Figure 8. The initial conditions for
each of these rate curves are above the CMC. The same
trend is seen here as was seen for the precipitation of SDS
with calcium. As the supersaturation ratio increases, the rate

of precipitation increases and the induction time decreases.

The rate of precipitation of SOBS with calcium is more than
1 min slower than for the precipitation of SDS with calcium
for an initial supersaturation ratio of 5.4.
The effect of increasing the degree of supersaturation on
the precipitation rate of SDS precipitating with DPC is
shown in Figure 9. The rate of precipitation for the
SDS/DPC surfactant precipitation is much slower than for
the SDS/CaCl2 and the SOBS/CaCl2 systems. The SDS/
DPC surfactant precipitation took more than 12 min compared with 3 min or less for the SDS/CaCl2 and SOBS/
CaCl2 surfactant precipitation rates for similar supersaturation ratios. Increasing the degree of supersaturation in the
SDS/DPC system increases the rate of precipitation. Also,
as seen for the precipitation of the two anionic surfactants
with calcium, as the supersaturation ratio decreases, the induction time increases.
The half-lives of the precipitation reactions decrease as
the initial supersaturation ratios increase. Figure 10 shows
this trend for each system studied. Reduced time is defined
in this paper as time divided by the half-life. Figures 1113
show the extent of precipitation plotted as a function of
reduced time for the SDS/CaCl2, SOBS/CaCl2, and SDS/
DPC systems, respectively. This results in a universal precipitation rate curve. For a given system, this universal rate
curve is remarkably independent of supersaturation or the
presence or absence of micelles.
Conductivity. Another method to measure the kinetics of
a precipitation reaction involving ionic species is to measure
the change in the conductivity of the solution with time.
When ions form complexes or are incorporated into crystal
lattices, the conductivity of the solution must decrease.
Turnbull (32) outlines a procedure to calculate the concentration of an ionic solution from conductivity measure-

FIG. 8. Precipitation rate curves of SOBS with CaCl2 above the CMC FIG. 9. Precipitation rate curves of SDS with DPC above the CMC at
at 30C. For abbreviations see Figures 2 and 5.

30C. For abbreviations see Figures 1, 3, and 5.

Journal of Surfactants and Detergents, Vol. 4, No. 1 (January 2001)

ANIONIC SURFACTANT PRECIPITATION. PART I

FIG. 10. Half-life as a function of initial supersaturation ratio for all

systems studied at 30C. For abbreviations see Figures 13 and 5.

ments. A qualitative comparison between the reaction kinetics as determined by isoperibol calorimetry and the conductivity of these same solutions is shown in Figures 1416
(for SDS/CaCl2) and 17 (for SDS/DPC). One of the most
important variables in the determination of the kinetics of
precipitation is the agitation of the solution. Agitation for
the conductivity experiments was provided by a magnetic
stir bar. The stirring rate was arbitrarily set at 240 rev/min
compared to 400 rev/min in the calorimeter. Owing to variations in the shape and size of both the reaction vessels and
the stirring mechanisms of the two experimental methods,
the exact agitation in the calorimeter could not be simu-

FIG.11. Extent of precipitation plotted against reduced time for SDS

with CaCl2 at 30C. For abbreviations see Figures 1 and 5.

FIG. 12. Extent of precipitation plotted against reduced time for

SOBS with CaCl2 above the CMC at 30C. For abbreviations see Figures 2 and 5.

lated in the conductivity experiments. Therefore, slight discrepancies may be due to these inconsistencies in the conditions rather than inaccuracies in the calorimetric method
for determining precipitation rates. Figure 14 compares
conductivity and calorimeter runs for an SDS/CaCl2 system
below the CMC for an initial supersaturation ratio of 3.7.
Figures 15 and 16 show the comparisons between calorimeter runs and conductivity for SDS/CaCl2 systems above the
CMC. Other researchers have followed the precipitation

FIG. 13. Extent of precipitation plotted against reduced time for SDS
with DPC above the CMC at 30C. For abbreviations see Figures 1, 3,
and 5.

Journal of Surfactants and Detergents, Vol. 4, No. 1 (January 2001)

10

C.H. RODRIGUEZ ET AL.

FIG. 14. Comparison of precipitation rate curves between isoperi- FIG. 16. Comparison of precipitation rate curves between isoperibol
bol calorimetry and conductivity for SDS/CaCl2 below the CMC at
So = 3.7 and 30C. For abbreviations see Figues 1 and 5.

calorimetry and conductivity for SDS/CaCl2 above the CMC at So =

7.0 and 30C. For abbreviations see Figures 1 and 5.

kinetics of BaSO4 using conductivity and have reported similar results (31,32). The shapes of the curves from calorimetry and from conductivity are very similar for Figures 1416
(though opposite in direction due to the detection methods). These results verify that the precipitation is being
measured by calorimetry, not some spurious phenomenon
or artifact of the experimental procedure.
A comparison of the calorimetry results with the conductivity for an SDS/DPC system is shown in Figure 17. For this
system, the presence of swamping electrolyte (0.15 M NaCl)
in the solution made the change in the conductivity due to
precipitation very small and therefore difficult to detect.
The calorimetric and conductivity results show approxi-

mately qualitative agreement even for this high-ionic

strength system.
Optical analysis of crystals. Image analysis and microscopy
were used to obtain visual information about the crystals
from single surfactant solutions at various lengths of time
after precipitation took place. Table 1 relates initial anionic
surfactant and counterion concentrations to aging time, initial supersaturation, and the initial relation of each solution
to the CMC. The solutions listed in this table correspond to
the image analysis pictures shown in Figures 1823. Image
analysis at 40 magnification was used to view Ca(DS)2 and
Ca(OBS)2 crystals at 4 min after the surfactant and calcium
had been mixed together. These pictures are shown in Fig-

FIG. 15. Comparison of precipitation rate curves between isoperibol FIG. 17. Comparison of precipitation rate curves between isoperibol

calorimetry and conductivity for SDS/CaCl2 above the CMC at So =

4.5 and 30C. For abbreviations see Figures 1 and 5.

calorimetry and conductivity for SDS/DPC above the CMC at So = 5.1

and 30C. For abbreviations see Figures 1, 3, and 5.

Journal of Surfactants and Detergents, Vol. 4, No. 1 (January 2001)

ANIONIC SURFACTANT PRECIPITATION. PART I

11

TABLE 1
Time of Analysis, Initial Supersaturation, and Initial
Relation to Critical Micelle Concentration (CMC)
for Image Analysis Solutionsa
Initial anionic
surfactant
concentration

Initial counterion
concentration

Time

So

Relation
to CMC

010 M SDS
0.075 M SOBS
0.010 M SDS
0.075 M SOBS
0.001 M SDS
0.001 M SDS
0.004 M SDS
0.004 M SDS
0.0003 M SDS
0.0003 M SDS

0.008 M CaCl2
0.010 M CaCl2
0.008 M CaCl2
0.010 M CaCl2
0.008 M CaCl2
0.008 M CaCl2
0.007 M CaCl2
0.007 M CaCl2
0.015 M DPC
0.015 M DPC

4 min
4 min
1 wk
1 wk
4 min
1 wk
1d
1 wk
1d
1 wk

6.59
5.54
6.59
5.54
1.90
1.90
4.58
4.58
3.83
3.83

Above
Above
Above
Above
Below
Below
Below
Below
Above
Above

a
SDS, sodium dodecyl sulfate; SOBS, sodium octyl benzene sulfonate;
DPC, dodecyl pyridinium chloride; So, supersaturation ratio.

ures 18 and 19 for 0.01 M SDS with 0.008 M CaCl2 and 0.075
M SOBS with 0.01 M CaCl2, respectively. The crystals from
the pure SDS solution are mostly trapezoidal and rhombic
in shape, with a few hexagonal shapes. Many crystals from
the pure SOBS solution are elongated flat plates, many with
jagged edges. Some broken-edged trapezoidal shapes are
present as well. Image analysis was also used to view the crystals after aging for 1 wk in solution as shown in Figures 20
and 21 at 40 magnification. The crystals from 0.01 M SDS
solution and 0.008 M CaCl2 are mostly clusters as shown in
Figure 20. The crystals from 0.075 M SOBS solution and
0.01 M CaCl2 are long, clear, and needle-like. The initial
conditions for Figures 1820 are below the CMC. Figures 22
and 23 show crystals from a 0.001 M SDS and 0.008 M CaCl2
solution aged for 4 min and 1 wk, respectively. These conditions are above the CMC. The Ca(DS)2 crystals aged for
4 min are mostly elongated with a few rhombic shapes.

FIG. 19. Image analysis picture of Ca(OBS)2 crystals at 40 precipitated from a 0.075 M SOBS/0.010 M CaCl2 solution; taken 4 min after
mixing at 30C. OBS, octyl benzene sulfonate; for other abbreviation
see Figure 2.
Crystals aged for 1 wk are trapezoidal, rhombic, with some
needle-like shapes. There are some broken edges as well, as
seen in Figure 22. Several microscope pictures of Ca(DS)2
crystals (not included here) have also been taken for an initial SDS concentration of 0.004 M and 0.007 M CaCl2 (approximately at the CMC). After 1 d, mostly needle-like,
trapezoidal, and rhombic shapes are seen at 10 and 40
magnification. The crystal sizes range from 10 to 150 m.
After 1 wk, the crystals have not changed significantly in size
or shape. Development of crystals below the CMC is different from the development of crystals above the CMC. Crystals forming in a less supersaturated system have been seen
previously to develop differently from crystals forming in a

FIG. 18. Image analysis picture of Ca(DS)2 crystals at 40 precipitated FIG. 20. Image analysis picture of Ca(DS)2 crystals at 40 precipitated
from a 0.010 M SDS/0.008 M CaCl2 solution; taken 4 min after mixing
at 30C. For abbreviations see Figures 1 and 3.

from a 0.010 M SDS/0.008 M CaCl2 solution after 1 wk at 30C. For abbreviations see Figures 1 and 3.

Journal of Surfactants and Detergents, Vol. 4, No. 1 (January 2001)

12

C.H. RODRIGUEZ ET AL.

FIG. 21. Image analysis picture of Ca(OBS)2 crystals at 40 precipitated from a 0.075 M SOBS/0.010 M CaCl2 solution after 1 wk at 30C.
For abbreviations see Figures 2 and 19.

FIG. 23. Image analysis picture of Ca(DS)2 crystals at 40 precipitated

more supersaturated system (29). It is uncertain if the differences seen in this paper between the crystals formed in
micellar solutions and submicellar solutions are related to
the presence of micelles. Lee and Robb (44) presented
an electron micrograph of Ca(DS)2 crystals formed under
unspecified conditions. Several shapes were seen in this micrograph including trapezoidal and rhombic shapes, which
are similar to some of the crystals seen in this project. The
sizes ranged between 1 and 10 m.

Crystals from a 3.0 104 M SDS, 1.5 103 M DPC, and

0.15 M NaCl solution at 1 d old and 1 wk old at 40 magnification are very thin and plate-like. The crystals have very
irregular shapes that look similar to the clumped-together
SDS/CaCl2 crystals above the CMC after 1 wk. The 1-d-old
and 1-wk-old SDS/DPC crystals range in diameter from less
than 15 m to about 20 to 50 m, with no significant
changes occurring within this time span.
Modeling. The only other papers in the literature of which
we are aware that address kinetics of surfactant precipitation
directly (35,44,45) are in general agreement with our results
on the rate of precipitation of the Ca(DS)2 system. The rate
of precipitation increased as supersaturation increased, as reported in those papers, consistent with results reported here.
A model using the diffusion mechanism to describe crystal growth was unsuccessful in describing the kinetics of precipitation in this paper. However, the structures for Ca(DS)2
and Ca(OBS)2 crystals are quite different from atomic force
micrographs. So, more sophisticated future kinetic models
must incorporate these crystal structure aspects.

from a 0.001 M SDS/0.008 M CaCl2 solution after 1 wk at 30C. For abbreviations see Figures 1 and 3.

ACKNOWLEDGMENTS

FIG. 22. Image analysis picture of Ca(DS)2 crystals at 40 precipitated

from a 0.001 M SDS/0.008 M CaCl2 solution, taken 4 min after mixing
at 30C. For abbreviations see Figures 1 and 3.

Financial support for this work was provided by the industrial sponsors of the Institute for Applied Surfactant Research including
Akzo Nobel Chemicals Inc., Albemarle Corporation, Amway Corporation, Clorox Company, Colgate-Palmolive, Dial Corporation,
Dow Chemical Company, DowElanco, E.I. DuPont de Nemours &
Company, Halliburton Services Corporation, Henkel Corporation,
Huntsman Corporation, ICI Americas Inc., Kerr-McGee Corporation, Lever Brothers, Lubrizol Corporation, Nikko Chemicals,
Phillips Petroleum Company, Pilot Chemical Company, Procter &
Gamble Company, Reckitt Benckiser North America, Schlumberger Technology Corporation, Shell Chemical Company, Sun
Chemical Corporation, Unilever Inc., and Witco Corporation.

Journal of Surfactants and Detergents, Vol. 4, No. 1 (January 2001)

ANIONIC SURFACTANT PRECIPITATION. PART I

REFERENCES
1. Coons, D., M. Dankowski, M. Diehl, G. Jakobi, P. Kuzel, E.
Sung, and U. Trabitizsch, Performance in Detergents, Cleaning Agents, and Personal Care Products 5.1 Detergents, in Surfactants in Consumer Products, Theory, Technology, and Applications, edited by J. Falbe, Springer-Verlag, Berlin, 1987, p. 197.
2. Lynn, J.L., Jr., Detergency, in Kirk-Othmer Encyclopedia of Chemical Technology, 4th edn., Vol. 7, John Wiley and Sons, New York,
1991, p. 1072.
3. Dunn, R.O., J.F. Scamehorn, and S.D. Christian, Use of Micellar-Enhanced Ultrafiltration to Remove Dissolved Organics
from Aqueous Streams, Sep. Sci. Technol. 20:257 (1985).
4. Scamehorn, J.F., and J.H. Harwell, Precipitation of Surfactant
Mixtures, in Mixed Surfactant Systems, edited by K. Ogino and
M. Abe, Marcel Dekker, New York, 1993, p. 283.
5. Brant, L.W., K.L. Stellner, and J.F. Scamehorn, Recovery of
Surfactant from Surfactant-Based Separations Using a Precipitation Process, in Surfactant-Based Separation Processes, edited by
J.F. Scamehorn and J.H. Harwell, Marcel Dekker, New York,
1989, Vol. 33, p. 323.
6. Yin, Y., J.F. Scamehorn, and S.D. Christian, Recovery of a Dialkyl Diphenyl Ether Disulfonate Surfactant from Surfactant
Flush Solutions by Precipitation, in Surfactant-Enhanced Subsurface Remediation: Emerging Technologies, edited by D.A. Sabatini,
R.C. Knox, and J.H. Harwell, ACS Symposium Series, Vol. 594,
American Chemical Society, Washington, DC, 1995, p. 231.
7. Stellner, K.L., and J.F. Scamehorn, Surfactant Precipitation in
Aqueous Solutions Containing Mixtures of Anionic and Nonionic Surfactants, J. Am. Oil Chem. Soc., 63:566 (1986).
8. Stellner, K.L., and J.F. Scamehorn, Hardness Tolerance of
Anionic Surfactant Solutions: 1. Anionic Surfactant with
Added Monovalent Electrolyte, Langmuir, 5:70 (1989).
9. Stellner, K.L., and J.F. Scamehorn, Hardness Tolerance of
Anionic Surfactant Solutions: 2. Effect of Added Nonionic Surfactant, Langmuir 5:77 (1989).
10. Stellner, K.L., J.C. Amante, J.F. Scamehorn, and J.H. Harwell,
Precipitation Phenomena in Mixtures of Anionic and Cationic
Surfactants in Aqueous Solutions, J. Colloid Interface Sci. 123:186
(1988).
11. Amante, J.C., J.F. Scamehorn, and J.H. Harwell, Precipitation of
Mixtures of Anionic and Cationic Surfactants: II. Effect of Surfactant Structure, Temperature, and pH, Ibid. 144:243 (1991).
12. Peacock, J.M., and E. Matijevic, Precipitation of Alkylbenzene
Sulfonates with Metal Ions, Ibid. 77:548 (1980).
13. Fan, X.J., P. Stenius, N. Kallay, and E. Matijevic, Precipitation
of Surfactant Salts II. The Effect of Nonionic Surfactants on
Precipitation of Calcium Dodecyl Sulfate, Ibid. 121:571 (1988).
14. Lopata, J.J., A Study of the Thermodynamic Properties of Surfactant Mixtures: Mixed Micelle Formation and Mixed Surfactant Adsorption, Ph.D. Dissertation, The University of Oklahoma, Norman, 1992.
15. Lowery, L.H., The Kinetics of Surfactant Precipitation, M.S.
Thesis, The University of Oklahoma, Norman, 1994.
16. Matheson, K.L., M.F. Cox, and D.L. Smith, Interactions Between Linear Alkylbenzene Sulfonates and Water Hardness
Ions. I. Effect of Calcium Ion on Surfactant Solubility and Implications for Detergency Performance, J. Am. Oil Chem. Soc.
62:1391 (1985).
17. Matheson, K.L., Detergency Performance Comparison Between LAS and ABS Using Calcium Sulfonate Precipitation
Boundary Diagrams, Ibid. 62:1269 (1985).
18. Rizkalla, E.N., Kinetics of the Crystallisation of Barium Sulphate, J. Chem. Soc., Faraday Trans. 1, 79:1857 (1983).
19. Scamehorn, J.F., Precipitation of Mixtures of Anionic Surfactants, in Mixed Surfactant Systems, edited by P.M. Holland and
D.N. Rubingh, ACS Symposium Series, Vol. 501, American
Chemical Society, Washington, DC, 1992, p. 392.

13

20. Rodriguez, C.H., The Thermodynamics and Kinetics of Anionic Surfactants and Surfactant Mixtures, Ph.D. Dissertation,
The University of Oklahoma, Norman, 1997.
21. Fan, X.J., M. Colic, N. Kallay, and E. Matijevic, Precipitation of
Surfactant Salts III. Metal Perfluoroalkanecarboxylates, Colloid
Polym. Sci. 266:380 (1988).
22. Hrust, V., N. Kallay, and D. Tezak, Precipitation and Association of Silver Laurate in Aqueous Solutions, Ibid. 263:424
(1985).
23. Davies, C.W., Ion Association, Butterworths, London, 1962,
p. 41.
24. Klotz, I.R., and R.M. Rosenberg, Chemical Thermodynamics: Basic
Theory and Method, 4th edn., Krieger Publishing Co., Malabar,
1991, p. 440.
25. Robinson, R.A., and R.H. Stokes, Electrolyte Solutions, 2nd edn.,
Butterworths, London, 1959, p. 174.
26. Burchfield, T.E., and E.M. Woolley, Model for Thermodynamics of Ionic Surfactant Solutions: 1. Osmotic and Activity Coefficients, J. Phys. Chem. 88:2149 (1984).
27. Dharmawardana, U.R., S.D. Christian, E.E. Tucker, R.W. Taylor, and J.F. Scamehorn, A Surface Tension Method for Determining Binding Constants for Cyclodextrin Inclusion Complexes of Ionic Surfactants, Langmuir 9:2258 (1993).
28. Scowen, R.V., and J. Leja, Spectrophotometric Studies on Surfactants. I. Interactions Between Cationic and Anionic Surfactants, Can. J. Chem. 45:2821 (1967).
29. Lieser, K.H., Steps in Precipitation Reactions, Angew. Chem. Internat. Edit. 8:188 (1969).
30. Konak, A.R., Derivation and Use of a Generalized Rate Equation for Crystallization and Precipitation, Kristall. Technik.
9:243 (1974).
31. Johnson, R.A., and J.D. ORourke, The Kinetics of Precipitate
Formation: Barium Sulfate, J. Am Chem. Soc. 76:2124 (1954).
32. Turnbull, D., The Kinetics of Precipitation of Barium Sulfate
from Aqueous Solution, Acta Metallurg. 1:684 (1953).
33. Walton, A.G., The Formation and Properties of Precipitates, Wiley,
New York, 1967, p. 1.
34. Nyvlt, J., O. Shnel, M. Matuchov, and M. Broul, The Kinetics
of Industrial Crystallization, Elsevier, New York, 1985, p. 35.
35. Clarke, D.E., R.S. Lee, and I.D. Robb, Precipitation of Calcium
Salts of Surfactants, Faraday Disc. Chem. Soc. 61:165 (1976).
36. Mullin, J.W., Crystallization, 3rd edn., Butterworth-Heinemann,
Boston, 1993, p. 180.
37. Mullin, J.W., Crystallization, 3rd edn., Butterworth-Heinemann,
Boston, 1993, p. 8.
38. Randolph, A.D., and M.A. Larson, Theory of Particulate Processes,
2nd edn., Academic Press, San Diego, 1988, p. 109.
39. Adamson, A.W., Physical Chemistry of Surfaces, 5th edn., Wiley,
New York, 1990, p. 364.
40. Nielsen, A.E., Kinetics of Precipitation, MacMillan, New York,
1964, p. 29.
41. Frank, F.C., The Influence of Dislocations on Crystal Growth,
Discussions Faraday Soc. 5:48 (1949).
42. Rawles, R., Atomic Force Microscopy Reveals Details of Protein
Crystallization, Chem. Eng. News (August 26):32 (1996).
43. Walton, A.G., The Formation and Properties of Precipitates, Wiley,
New York, 1967, p. 44.
44. Lee, R.S., and Robb, I.D., Precipitation of Calcium Surfactants:
Part 2, J. Chem. Soc. Faraday Trans.1 75:2116 (1979).
45. Lee, R.S., and Robb, I.D., Precipitation of Calcium Surfactants:
Part 3, Ibid. 75:2126.(1979).
[Received December 16, 1998; accepted December 1, 2000]

Cheryl Rodriguez is a scientist in the Corporate Technology Department of the Clorox Services Company. She received her B.S. and
Ph.D. in chemical engineering at the University of Oklahoma. Her

Journal of Surfactants and Detergents, Vol. 4, No. 1 (January 2001)

14

C.H. RODRIGUEZ ET AL.

awards include the 1995 Ralph H. Potts Memorial Fellowship from

the American Oil Chemists Society (AOCS) and an Outstanding
Paper Presentation at the 1996 AOCS Annual Meeting.
Lori Lowery is an engineer with Shell Oil Co. She received her
B.S. at the University of Kansas and her M.S. at the University of
Oklahoma, both in chemical engineering.
John Scamehorn holds the Asahi Glass Chair in Chemical Engineering and is Director of the Institute for Applied Surfactant Research at the University of Oklahoma. He received his B.S. and
M.S. at the University of Nebraska and his Ph.D. at the University
of Texas, all in chemical engineering. Dr. Scamehorn has worked
for Shell, Conoco, and DuPont, and has been on a number of editorial boards for journals in the area of surfactants and of separa-

tion science. He has coedited four books and coauthored over 140
technical papers. His research interests include surfactant properties important in consumer product formulation and surfactantbased separation processes.
Jeffrey Harwell holds the Conoco/DuPont Professorship in
Chemical Engineering and is Associate Dean of Engineering at the
University of Oklahoma. He received his B.S. at Texas A&M University in chemistry, his M.S. at Texas A&M University in chemical engineering, and his Ph.D. at the University of Texas in chemical engineering. Dr. Harwell has coedited three books and coauthored over 100 technical papers. His research interests include the
use of surfactants in environmental remediation, microemulsion
formulation, and nanotube production.

Journal of Surfactants and Detergents, Vol. 4, No. 1 (January 2001)