SURFACTANT SYSTEMS
`Their chemistry, pharmacy and biology
`
`Page 1 of 98
`
`SENJU EXHIBIT 2043
`INNOPHARMA v. SENJU
`IPR2015-00903
`
`

`
`SURFACTANT SYSTEMS
`Their chemistry, pharmacy and biology
`
`D. Attwood
`DepaJ~ment of Pharmacy ·
`University of Manchester
`A. T. Florence
`Department of Pharmacy
`University of Strathclyde
`
`LONDON NEW YORK
`CHAPMAN AND HALL
`
`Page 2 of 98
`
`

`
`First published 1983 by
`Chapman and Hall Ltd
`11 New Fetter Lane, London EC4P 4EE
`Puhlished in the USA by
`Chapman and Hall
`733 Third Avenue, New York NY 10017
`© 1983 D. Attwood and A. T. Florence
`Printed in Great Britain by
`J. W Arrowsmith Ltd, Bristol
`
`ISBN 0 412 14840 4
`
`All rights reserved. No part of this book may be reprinted, or
`reproduced or utilized in a:ny form by any electronic, mechan(cid:173)
`ical or other means, now known or hereafter invented; inClud(cid:173)
`ing photocopying and recording, or in any information storage
`and retrieval system, without permission in wr.iting from the
`Publisher.
`
`British Library Cataloguing in Publication Data
`
`Attwood, D.
`Surfactant systems.
`1. Sl.,lrface active agents
`I. Title
`II. Florence, A. T.
`668' .1
`TP994
`
`ISBN 0-412-14840-4
`
`Page 3 of 98
`
`

`
`6 Pharmaceutical aspects of
`solubilization
`
`I
`/ ,
`
`6.1 Introduction
`Solubilization in surfactant solutions above the critical micelle concentration
`offers one approach to the formulation of poorly soluble drugs in solution form
`[1].
`The objective of this chapter and of Chapter 7 is to review the state of the art of
`solubilization in surfactant systems with emphasis on the consequences of a
`surfactant presence in pharmaceutical formulations. In particular, emphasis will
`be placed on the effect of surfactants on bioavailability and the toxicity of
`formulations for neglect of these topics will, on the one hand, prevent the
`realization of the potential of surfactant systems and, on the other, might.lead to
`the unwise use of surfactants in formulations. Some attempt will be made to place
`the topic in perspective and to answer the question as to the real value of
`surfactant solubilization in pharmaceutical formulation.
`It is perhaps true that micellar solubilization has not made much impact on
`drug formulation. There are relatively few marketed products which could be
`considered to be isotropic solutions of drug and surfactant in either the UK or the
`USA, although surfactants are present in many formulations as minor adjuvants
`and to that extent their presence and influence is perhaps hidden .
`. The limiting factors in the use of solubilizers as effective formulation aids are (i)
`the finite capacity of the micelles for the drug, (ii) the possible short- or long-term
`adverse effects of the surfactant on the body, and (iii) the concomitant
`solubilization of other ingredients such as preservatives, flavouring and colouring
`matter in the formulation with consequent alterations in stability and effective(cid:173)
`ness. Nonetheless,-there is scope for development simply because there is a need
`for agents to increase the solubility of poorly soluble drugs even if only at the
`stage·of ·pharmacological evaluation where, indeed, surfactants are used often
`Without due regard to the implications. The use of co-solvents and surfactants to
`· solve:problems of low solubility has the advantage that the drug entity can be
`used without chemical modification and hence toxicological data do not have to
`be repeated as would be the case when alternative approaches are used to produce
`more soluble compounds. Some caution has, however, to be adopted in the
`
`293
`
`Page 4 of 98
`
`

`
`294
`
`Surfactant systems
`
`interpretation of animal pharmacology and toxicology on formulations which
`differ from the final marketed product, especially if the final preparation contains
`surfactant but early test formulations do not or vice versa. Surfactants, as we will
`see in Chapter 7, are not inert substances some having distinctive pharmaco(cid:173)
`logical actions. There is a demonstrable need for the development of less toxic
`surfactants; the polyoxyethylene- polyoxypropylene block co-polymers which
`will be discussed later seem to have fewer side effects than conventional
`surfactants and seem to be worthy of further investigation.
`With the development of new dosage form technology in which control of drug
`release is achieved, it is conceivable that micellar systems will find some place
`because of the ability or the micellar phase to alter the transport properties of
`solubilized drug molecules. One can envisage the deliberate addition of
`surfactants to drug rese.rvoirs to control the exit rate of drugs from polymeric
`devices. This will be explained in Chapter 7.
`This chapter is restricted mainly to aqueous systems and the solubilization of
`water-insoluble and poorly soluble drug entities and pharmaceutical additives,
`and, because of the lesser toxicity of non-ionic surfactants, it will concentrate on
`non-ionic surfactant systems. Wherever possible cited work refers to systems
`which have potentia] utility in pharmacy as there is a danger that all our
`knowledge is gained on model systems (of toxic ionic surfactants which are used
`because they are available in a pure state, benzene or similar well-defined solutes,
`and other unacceptable additives such as propanol) while we remain blissfully
`unaware of bow to solve the real problems that arise [1].
`As we have seen in Chapter 1, the range of available surfactants is wide, and so,
`too, are the mechanisms of solubilization and the effects the surfactants have on
`the solubilized material. Examples are known of enhanced drug activity and of
`inactivation, of increased stability, and instability; the interactions of the
`surfactants with components of the body must also be considered. In the case of
`insoluble drugs, the presence of micelles may enhance their activity th rough
`solubilization and transport to the site of action, a process wh ich otherwise might
`have been a slow one. This has, of cou rse, dire consequences in the case of
`carcinogens: normally insoluble carcinogenic substances which may be ingested
`may become very active in combination with surfactants, and, as the latter are
`taken in increasing amounts in food (non-ionics in bread is one example), this is a
`problem which warrants further study. Drugs which are meant to act on the
`intestinal mucosa, such as sulpbaguanidine, might be inadvertently solubilized.
`There is the problem, especially with non-ion ic su rfactants, of interactions with
`preservatives in pharmaceuticals and consequent loss ·of biological action.'
`.Some drugs themselves are surface-active and form micelles, W.hile S·Hrface
`activity may. not, in all cases, be the cause of their biological activity; •it .must. in
`some way influence it apd modify their interaction with the .components•of
`dosage forms or the components of the body. Surface-active drugs and surfactant
`molecules will interact .to form mixed micelles at sufficiently high coneentrations;
`a phenomenon which has implications for the thermodynamic activity and
`possibly. the biological activity of the .drug molecule.
`
`Page 5 of 98
`
`

`
`Pharmaceutical aspects of solubilization
`
`295
`
`Since 1964 there have been several comprehensive reviews of solubilization in
`surfactant systems, notably those by Swarbrick [2], Mulley [3], Sjoblom [4],
`Droseler and Voight [5], Elworthy et al. [6], and Florence [1]. These reviews
`together cite over a thousand sources primarily concerned with pharmaceutical
`applications. Other major publications which deal with micellar systems
`implicating solubilized species include Cordes [7], Fendler and Fendler [8] and
`the collections of papers edited by Mittal contain several contributions on the
`topic [9].
`In this chapter the solubilization of a number of classes of drugs and
`pharmaceutical products will be dealt with; in some cases the division into
`sections has had to be somewhat arbitrary, but, as far as possible, compounds
`with similar structures, such as the steroids, have been dealt with as a group.
`
`6.2 Solubilization of drugs
`
`6.2.1 Antibacterial compounds
`
`(A) PHENOLIC COMPOUNDS
`Solutions of cresol with soap were eady pharmaceutical examp~s of solubilized
`systems. Phenol itself is soluble in water to the extent of 7.7% (w/v), but it has
`djsadvantages; the alternatives, cresol, ch1orocresol, chlo roxylenol, and thymol.
`are much less soluble in water,and their use as disinfectants has led to the need for
`fprmulation in surfactant solutions.
`Solution of cresol with soap (lysol) is a saponaceous solution containin~ 50%
`v;/v cresol. Its monograph specifies no particular soap, although activity of the
`preparation depends to a large extent on the type of soap employed. Although
`still used, the absence of strict standards for lysol, the widely varying phenol
`fractions used in its preparation, and the varying properties of the soaps make it
`a~ unsatisfactory solution. The high toxicity of phenol and the cresols has
`mitigated against their more widespread use. Emphasis is now being placed on
`th¢ir chlorinated derivatives, chloroxylenol and chlorocresol. Chloroxylenol is a
`potent, non-irritant bactericide oflow toxicity. It has, however, a low solubility in
`water, 0.031 g ml- 1 at 20° C [10]; the official preparation, Solution of
`Chloroxylenol B.P., contains 5% vjv chloroxylenol with terpineol in an alcoholic
`soap solution. A modification of this, claimed to be less alkaline, has been
`described by Lloyd and Clegg [11]. There are numerous commercial formu(cid:173)
`lations with a wide spread of Rideal- Walker coefficitmts.
`Mu!ley and Metcalf (12] have carried out detailed i'nvestigations of the phase
`behaviour of non-ionic detergent systems containing ehloroxy,lenol (4-chloro-
`3,5-xylenol). Two of their phase diagrams are reproduced in Fig. 6.1. The
`surfactant C6H 13 (OCH 2CH 2)60 H is an efficient solubilizcr above its C MC,
`which is approximately 3% w/w, but high concentrations are required to form
`isotropic
`liquids contamrng
`reasonable quantities of chloroxylenoJ.
`C6lin(OCH,2CH2 hOH requires concentrations above 50 % w/ w to achieve an
`isotropic solution, and this compound probably acts more as a hydrotrope than
`
`Page 6 of 98
`
`

`
`296
`
`Surfactant systems
`
`as a micellar solubilizer in this concentration region. The former detergent forms
`only small aggregates of about 13 monomers in aqueous solution [13]. To obtain
`systems suitable for use it is essential to increase the alkyl chain length;
`C 10H21 (OCH 2CH2 )6 0H has a low CMC and its micelles are reasonably large,
`containing 73 monomers at 25°C [14]. Isotropic micellar systems are formed at
`lower concentrations of detergent than for the shorter alkyl-chain homologues.
`However, there is a concomitant increase in the complexity of the phase diagram
`with the formation of liquid crystalline phases (Fig. 6.1).
`There is apparently no evidence from these phase diagrams for the existence of
`simple phenol- glycol chain complexes: the liquid which separates at a solubility
`limit is a solution of variable composition. Ultraviolet spectroscopy shows that a
`hydrogen-bonded complex between the phenolic hydroxyl group and the other
`oxygens of cetomacrogol lOOO (C 16H 33 (OCH2 CH2h3 - 240H) is formed when
`chloroxylenol is solubilized by this commercial non-ionic surfactant. The
`
`X"'6-(cid:173)
`x=2-----
`
`IL+S
`
`Water
`
`Chloroaylenol
`
`Figure 6.1 The upper phase diagram (after Mulley and Metcalf (12]) illustrates the
`phases existent in the system c6 H IJ (OCHl CHl >x OH: 4-chloro-3,5-xyleool: water at
`20° C. The dotted line represents the behaviour when x = 2and the solid lineswberex = 6.
`The lower diagram shows the much more complex behaviour in the system:
`C 10 H2 1 (OCH1 CH1 )6 OH: ehloroxylenol: water. JL = isotropic liquid; LC = liquid
`crystalline; S = solid 2L = two liquids (immiscible.)
`
`Page 7 of 98
`
`

`
`Pharmaceutical aspects of solubilization
`
`297
`
`solubility of the chloroxylenol is directly proportional to the surfactant
`concentration, above the CMC [ 15]. However, rough determinations of the
`solubilities of resorcinol and phenol in cetomacrogol solutions, varying in
`concentration from 1 to 20%, showed these compounds to be less soluble than in
`water, although their solubility was proportional to detergent concentration
`[16].
`Solutions of phenols in ionic systems exhibit similar behaviour. An initial fall in
`the solubility of 2-hydroxyphenol and 4-benzylphenol in potassium laurate
`solutions was noted below the CMC of the soap [17]. Few workers have
`commented on this 'insolubilization'; compounds with very low water solubility
`possibly do not show this property. That it is not restricted to phenols is shown by
`the results of Heller and Klevens [18] for ethyl benzene in potassium laUt·ate.
`Ethyl benzene has a solubility in water similar to that of 4-benzylphenol.
`The binding of series of phenols, cresols and xylenols to the non-ionic
`surfactant cetomacrogol 1000 can be described by a Langmuir adsorption
`isotherm [19]
`
`x = K 1K 2c/(1 + K 1c)
`where x is the solute bound (mmol g- 1 micelle), c is the concentration of free
`unionized solute (mmol), K 1 is the binding constant (1 mmol- 1
`) and K 2 the solute
`bound at hypothetical saturation (mmol g - 1 micelle). The combined parameter
`K 1 K 2 is specific for each system and may be defined as the distribution coefficient
`of the solubilizate at infinitely dilute solubilizate concentration (P0 ), Azaz and
`Don brow (19] assert. Its value characterizes ideal behaviour both in the aqueous
`and micellar phases hence strictly would be subject to activity corrections.
`Binding capacity is inversely related to the water solubility of the pheno~ cresol
`and xylenol, as can be seen in Table 6.1.
`Values of P 0 in 0.1 M NaCI are also shown for a few compounds in this Table. A
`log-log plot of binding capacity and aqueous solubility yields a straight line. Azaz
`
`Table 6.1 Aqueous solubility and distribution coefficient at infinite dilution (P 0 ) between
`cetomacrogol and water of phenols at 25° C
`
`Compound
`
`Phenol
`a-Cresol ..
`p-Cresol
`m-Cresol
`2,4-Xylenol
`2,6-Xylenol
`3,5-Xylenol
`3,4-Xylenol
`2,3-Xylenol
`2,5-Xylenol
`
`Solubility in water
`(moll- 1 x 1 03
`)
`
`P0 in water•
`
`Solubility in 0.1 M P0 in 0.1 M
`NaCI (moll - 1
`NaCI•
`X 103)
`
`1000
`240
`199
`142
`51.0
`49.5
`40.0
`39.0
`37.4
`29.0
`
`42.0
`79.5
`76.4
`85.1
`125
`114
`132
`151
`169
`197
`
`233
`188
`133
`
`117
`80.6
`
`190
`
`• Units: (1/g) x 103 = 1/ 1000g or dimensionless units assuming density of cetomacrogol is unity at
`25•c. Measured in 2% cetomacrogol. From Azaz and Donbrow [19].
`
`Page 8 of 98
`
`

`
`298
`
`Surfactant systems
`
`and Donbrow's work has supported earlier work [20--23] which demonstrated
`that in unsaturated systems the binding 'constants' of solubilizates to surfactants
`are concentration-dependent and not, in fact, constant as some authors have
`assumed (e.g. [24-26] ). This variation may have important practical implications
`in formulation.
`A wider range of 34 benzoic acid derivatives has been studied in detail by
`Tomida et al. [27]. Using a solubility method these workers obtained saturation
`solubilities of the benzoic acid derivatives in Brij 35 (a polyoxyethylene lauryl
`ether) over a range of concentrations. Solubility ratios, calculated as the solubility
`in the surfactant solution/solubility in HCl, were a linear function of surfact-ant
`concentration allowing the calculation of a partition coefficient P m which can by
`defined as
`
`(6.1)
`
`where Cm and Ca are the concentrations in the micellar and aqueous pha:s~s,
`respectively. Pm is obtained from the solubility data as
`st S = (Pm-l)vC5 + 1
`
`a
`where S1 is the tota l solubility of solubilizate in the presence of surfactant at
`concentration C5,Sa is the solubility in the absence of surfactant and vis the
`partial molar volume of the .surfactant. Some of the extensive data is reproduced
`in Table 6.2 for the ortho, para and meta substituents. The data are consistent
`with the findings of Azaz and Donbrow: the order of aqueous solubilities is
`always ortho >meta> para and the order of Pm is the opposite except for the
`hydroxybenzoic acids for which the ortbo compound was solubilized roost,
`followed by para and meta compounds. Patel and Foss [21 J obtained the
`
`(6.2)
`
`Tilble 6.2 Aqueous s.ol!Jbilities, Sa, and partition coeffici~nts of b~nz9ic acids, P m>.~e~weep
`aqueous and micellar phases obtained from solubility meth0d•
`:·.J
`"i
`:·
`
`Substituent
`
`or tho
`
`Sa(moll- 1)
`
`Pm
`
`4.05 x w- 2
`8.66 x w- 3
`5.29 X 10- 3
`1.75 X 10- 3
`6.55 X 10- 3
`.2.50 X 10- 2
`1.08 X 10- 2
`2.53 x w- 2
`2.57 x w- 2
`
`43.1
`99.8
`150
`271
`120
`32.0
`116
`47.5
`
`22.4
`
`H
`F
`Cl
`Br
`I
`CH3
`OCH3
`OH
`N02
`CN
`COOH
`
`• From [27].
`
`meta
`
`)
`
`Sa(moll- 1
`2.61 x w- 2
`1.65 x w- 3
`1.92 x to- 3
`1.36 X 10- 3
`2.74 X 10-4
`6.13 x w- 3
`1.18 x w- 2
`5.71 x w- 2
`1.57 x w- 2
`2.35 x w- 3
`4.42 X }Q- 4
`
`par !I
`
`Pm
`
`Sa(moll- 1
`
`)
`
`Pm
`
`57.4
`91.2 4.98 x w- 3
`3.48 x w- 4
`346
`1.42 X 10-4
`505
`9.16 x 10-6
`1150
`2.23 ·X 10- 3
`166
`1.30x10- 3
`72.8
`38.3 4.17xl0- 2
`1.01 X 10- 3
`96.8
`5.60 x w- 3
`50.1
`6.50 X 10-S
`155
`
`94.6
`446
`634
`908
`163
`109
`42.2
`117
`57.0
`69.8
`
`Page 9 of 98
`
`

`
`Pharmaceutical aspects of solubilization
`
`299
`
`magnitude of interaction of hydroxy-, chloro- a nd · aminobenzoic acids in
`polysorbate 80 and cetomacrogol 1000; hydroxy and amino derivatives showeJ
`the order of interaction to be ortho > para > meta. Substitution of a hydroxy
`group in the ortho position results in more affinity for any surfacLant than para
`and meta substituents. This can be explained by the fact that the intramolecular
`hydrogen bonding increases the proton-donating nature of the carboxylic group.
`The greater the dissociation constant of the acid group the greater the hydrogen
`bonding to the oxyethylene groups in the micelle.
`Plots of log P m versus log P octanol for the 34 compounds studied produced three ·
`groupings of results [27] to which the following equations applied.
`n
`s
`r
`log Pm = 0.9211og P octanol +0.118 21 0.986 0.080
`(6.3)
`log Pm = 0.88llog Poctanol + 0.392 5 0.999 0.014
`(6.4)
`log Pm = 0.968Jog P octanol + 0.600 3 0.999 0.036
`(6.5)
`where n is the number of points used in the regression, r is the correlation
`coefficient, and sis the standard deviation. Equation 6.3 applies to the majority of
`the compounds studied; Equation 6.4 to the nitro and cyano derivatives and the
`last equation to the compounds with dicarboxylic groups. The intercept values of
`the three groups are quite different; it is believed that the magnitude of the
`intercept is a reflection of the site of solubilization in the micelle. The closer the
`environment is to the nature of octanol used in the partitioning studies to obtain
`Poc1, the closer the intercept should be to zero. A negative intercept (for salicylic
`acid derivatives) has been identified with solubilization in the hydrocarbon core.
`The site of solubilization while of little practical importance in the design of
`pharmaceutical formulations is of more than academic interest as the position of
`the solubilizate in the micelle may determine its stability and reactivity towards
`attacking species in the continuous phase (see Chapter 11).
`To obviate the problem of the different affinities of ionized and unionized
`species for micelle, Tomida eta/. [27] carried out their investigations at a pH such
`that ionization was suppressed. pH is rarely as low as that in this work and its
`influence on solubilization must be considered. Although the hydrogen ion
`, concentration can influence the solution properties of non-ionic surfactants [28],
`the principal influence on uptake is exercised through the effect of pH on the
`equilibrium between ionized and unionized drug or solute species. This effect has
`been studied in most detail by Collett and Koo [29]. Increasing pH leads to a
`decrease in the micellar uptake of organic acids because of increasing solute
`solubility in the aqueous phase through increased ionization. This effect is clearly
`seen in Fig. 6.2 when the results of uptake of 4-chlorobenzoic acid between pH 3
`and 4.40 are plotted as a ratio of its solubility in water of the appropriate pH, i.e.
`as the solubility ratio R. Considering the micellar species to form a phase or
`pseudophase allows a simple quantitative measure of the interaction between
`solubilizate and micelle. The conce~tration of solubilizate in the micelle is related
`to its concentration in the aqueous phase by a partition coefficient as defined in
`Equation 6.1.
`
`Page 10 of 98
`
`

`
`300
`
`Surfactant systems
`
`12
`
`pH
`
`3.0
`
`3.4
`
`3.62
`
`4.0
`
`4.15
`
`4.4
`
`3
`2
`Polysorbate cancn
`
`
`
`4 X 163 ( mol ( 1
`)
`
`Figure 6.2 The influence of polysorbate 20 concentration and pH on the solubility ratio,
`R, of 4-chlorobenzoic acid. From Collett and Koo [~9].
`
`Curved Scatcbard plots for the interaction of propyl p-hydroxybenzoate
`(propyl paraben) with four polyoxycthylenc dodecyl ethers are shown in Fig. 6.3
`[31]. The primary class of binding sites exhibited a high affinity and a low
`capacity for the preservative while the secondary sites bad a low affinity and large
`binding ?apacity. Thus
`
`(.6.6)
`
`Analysis of these has allowed n1 , n2 , K 1 and K 2 to be estimated and related to the
`nature of the bindmg 'process in the micelles, e·specially in respect of the
`interaction with the hydrophilic polyoxyeth'ylene layer':(32].
`Qf great practicaH.mportahce are ~he effects of additives on the binding of
`preservative molec::ules . to· surfactant micelles. ·Blanchard et al., [33] have
`confirmed the negiigible ·effect of sorbitol on the interaction of phenolic
`preservatives with polysorbate 80 using a Scatchard approach: The sorbitol is
`probably too hydrophilic to interact with the micelle aiid thus does not compete
`for binding sites. Similar conclusions were reached by Shimamoto and Mirna [34]
`studying ·the effects 'Of glycerol, propylene glycol and ·1,3-butylerre glycol on
`paraben~non-ionic surfactant interactions. These polymers' had little effect on th'e
`binding of preservatives to the ptimary binding sites located ·at the core/ PEG
`boundary of the micelle but they were 'thought to decrease binding at· the
`secondary sites, 1,3-butylene glycol being most effective in displacing the
`preservatives: These secondary sites are reckoned to be non-speciijc and located
`in the PEG layer. As materials such as 1,3-butylene glycol may penetrate"the PEG
`region they would probably displace sol ubilizate molecules. It would seem that
`displacement from the primary site would require imich greater structural
`specificity (see discussion on interaction of preservative mixtures with micelles·,
`
`Page 11 of 98
`
`

`
`Pharmaceutical aspects of solubilization
`
`301
`
`7
`
`0
`
`0
`
`0 0
`
`0 ·1
`
`r
`
`0·2
`
`Figure 6.3 Scatchard plots for the interaction of propyl paraben with polyoxyethylene
`dodecyl ethers, n being the number of oxyethylene units.
`0 n = 15; • n = 20; 6 n = 30; 'n = 50
`From [31] with permission.
`
`below). Any displacement of preservative from the micelle is likely to increase the
`preservative activity of the formulation.
`The effects of added electrolytes on solubilized systems are discussed in
`Chapter 5. In Table 6.1 it can be seen that the addition of sodium chloride to a
`non-ionic system increases the P m of the s'Olubilizate. An electrolyte can have a
`dual effect, fi.rst on the properties of the · surfactant and secondly on the
`solubilizate. If the electrolyte salts out the solubilizate Pm will increase, an effect
`observed with non-ionic surfactant systems whose micelles would· be increased in
`size by such electrolytes. In ionic surfactant systems ·the effect can be more
`complex. The addition of electrolyte to an ionic surfactant results in a decrease in
`CMC, increase in micellar size and a decrease in effective charge per monomer,
`probably leading to a greater concentration of head groups and a more rigid
`micellar interior [35] which might result in decreased uptake of solubilizate into
`the micellar core. Uptake of methyl and ethyl para ben is increased by the addition
`of 10 mM NaCI to sodium Iaury! sulphate [36] (see Table 6.3). As both electrolyte
`and the presence of paraben lowers the surfactant CMC, analysis of the results
`produced the unexpected conclusion that for all three compounds the partition
`coefficient to the micellar phase is reduced on addition of electrolyte. This is a
`problem which occurs and recurs in detailed studies of mechanisms of
`solubilization, beipg clearest when pH effects are studied. Generally the
`formulator is interested in total solubility which includes solubility in the aqueous
`and micellar phases. While the partitioning of a species to the micellar phase
`might be reduced, its increased solubility in the aqueous phase may compensate
`for this. In spite of the partition coefficient of the methyl, ethyl and butyl para ben
`
`Page 12 of 98
`
`

`
`f
`
`302
`
`Surfactant systems
`
`Table 6.3 Solubilization of alkylparabens in water and in 40 m M sodium
`Iaury! sulphate solution at 2r C*.
`
`Alkylparaben
`
`Solubility (mmoll - 1
`
`) in
`
`Water
`
`40mM NaLS
`
`40 mM NaLS and
`50mM NaCI
`
`Methylparaben
`Ethylparabcn
`Butylparaben
`
`14.5
`5.4
`1.1
`
`• From (36, 38).
`
`33.9
`22.7
`24.3
`
`31.6
`21.9
`26.7
`
`increasing towards the micellar phase from 1.2 through 3.2 to 21, respectively, in
`40 mM sodium Iaury! sulphate (NaLS) the total solubility is still highest for
`methyl para ben with a solubility limit of 33.9 mM. Ethyl para ben has the lowest
`solubility (22.7 mM)and butyl para ben has a solubility of24.3 mM in 40 mM NaLS
`[38]. In a series such as the alkyl parabens their different locations in the micelle
`may be another factor complicating a ready understanding of the observation; the
`effect of the paraben on CMC which follows the order butyl > ethyl
`> methylparaben is of little importance when the total surfactant concentration
`is 40 mM as in the investigations in question but would obviously be important it
`surfactant concentrations close to the concentration (Fig. 6.4 shows this effect).
`Uptake of solubilizate into surfactant micelles changes the physical state of the
`micelle (see Section 5.5). Sometimes the change in shape may result in drastic
`changes in the physical properties of the system as a whole:...this may influence its
`use. The effect of additive on the cloud point of non-ionic surfactants
`
`9
`
`8
`
`~
`
`"' l: 7
`u 6
`~
`u
`
`)(
`
`5
`
`4
`
`0
`
`Figure 6.4 The CMCs of sodium Iaury! sulphate solutions in the presence of alkyl
`parabens. e, methylparaben; 0, ethylparaben; x, butylparaben. From Goto and Endo
`[37] with permission.
`
`Page 13 of 98
`
`

`
`•
`
`Pharmaceutical aspects of solubilization
`
`303
`
`t100
`.....
`c: ·o
`,
`0. 60
`0 u 40
`
`:::)
`
`0·05
`
`0 ·2
`
`Figure 6.5 Relation between the cloud point and P0 values for phenol and its homologues
`at concentrations of 0.05 moll- 1 in 2 % cetomacrogol I 000. From Don brow and Azaz
`[ 43]. Values of P 0 from Table 6.1.
`
`(39--42] is of some practical importance as the cloud point may be lowered
`below room temperature. Fig. 6.5 shows the effect of a range of phenols on
`the cloud point of cetomacrogol solutions (cetomacrogol 1000 B.P. is
`C 16H 3 3 (0CH2CH2) 22 _ 240H) where the relation between cloud point and the
`distribution coefficient of phenols, cresols and xylenols between micelles and
`water is demonstrated. The effect of a phenol on the cloud point is inversely
`related to its hydrophilicity. As the cloud point is thought to be due to the growth
`of the non-ionic micelles with increasing temperature, the binding of solute to the
`micellar structure could explain the lowering of the cloud point if the surfactant
`monomer-solute complex is more hydrophobic than the surfactant monomer
`itself.
`Phenol- water systems display critical solution temperatures (CST). Addition
`of fatty acid soaps generally causes a lowering of CST. 3 % sodium oleate lowers
`the CST of the phenol- water mixture from "' 65 to 0° C [ 45]. 1 % lowers it to
`43° C and 1 % sodium stearate lowers it to 49.1 o C [ 46]. Prins [ 44] in
`investigations on cetyltrimethylammonium bromide (CTAB)- phenol- watet
`mixtures, found striking effects caused by the detergent on the CST of the
`phenol- water system. Table 6.4 gives the concentrations of CTAB and phenol
`which have in admixture with water a CST of 20° C. The figures were obtained
`from a study of the phase diagram of the system.
`
`Table 6.4 Concentrations of CTAB and phenol
`having a critical solution temperature of 20° c•
`
`CTAB ( % wfw)
`
`Phenol ( % w/w)
`
`8.0
`17.5
`28.0
`34.0
`40.5
`41.5
`40.0
`
`• From Prins [ 44].
`
`11.0
`16.8
`15.2
`31.7
`35.6
`41.0
`45.1
`
`Page 14 of 98
`
`

`
`•
`
`304
`
`Swfactant systems
`
`Cetyltrimethylammonium bromide has a more complicated action than the fatty
`acid soaps. Although it generally enhances the mutual solubility of phenol and
`water, at some levels it has been shown to decrease the mutual solubility. Addition
`of more CTAB causes the mutual solubility to increase again until at 48% CTAB,
`complete miscibility is attained. Prins has gone some way to explaining this
`behaviour. When phenol is added to an aqueous solution of CTAB consisting of
`spherical micelles the phenol induces the formation of rod-shaped aggregates (see
`simplified phase diagrams in Fig. 6.6).
`
`CTAB
`
`,.~;:~~ Spherical micelles
`Rod-shaped micelles
`
`~= Molecular dispersion
`
`cetyltrimethylammonium
`the
`representa.tion of
`Figure 6.6 A diagrammatic
`bromide- phenol- water system. (After Prins [ 44] ). Horizontal arrow shows increasing
`phenol COI-lCentrations and passage from solutions containing spherical micelles, .through
`solutions containing asymmetric micelles to a molecular dispersion on breakdown of the
`micelles.
`
`The rod-shaped micelles can solubilize large quantities of phenol, but they
`reach a point (b) where they disintegrate, forming a molecular dispersion which
`results in a loss of mutual solubility. All systems containing more than 48 %
`CT AB are completely miscible. Examination of the interfacial tension curve of
`the co-existent phases in the water- phenol- CTAB system (phenol:water 40:60 by
`weight) shows that the curve mimics the solubility behaviour. A minimum at
`about 4% is followed by an increase in interfacial tension, which falls again after
`15% concentration, falling to 0 mN m- 1 at about 40%.
`
`(B) INTERACTION OF PRESERVATIVE MIXTURES WITH SURFACTANTS.
`In many formulations more than one solute will be a potential solubilizate
`whether or not this is desired. As discussed in Chapter 5, the effect, if any, of one
`solute on the solubilization of another will depend on the mechanisms of
`solubilization. If solubilization of one solute occurs at specific 'sites' within the
`micelles then molecules with similar binding affinities might compete for the
`available sites leading to a decreased solubilization of each. In some cases one
`solute might induce a reorganization of the micelle structure and allow increased
`uptake; both mechanisms might operate such that maxima and minima are seen
`in the plots of solubility versus the concentration of second solubilizate [ 4 7] (see
`Fig. 5.23). Benzoic acid, for example, increases the solubility of methyl para ben in
`
`Page 15 of 98
`
`

`
`•
`
`Pharmaceutical aspects of solubilization
`
`305
`
`cetomacrogol solutions, but dichlorophenol decreases its solubilization [ 47].
`Chloroxylenol reduces the solubility of methylparaben and methylparaben
`reduces the solubility of chloroxylenol in cetomacrogol, there being no effect on
`mutual solubilities in the absence of surfactant.
`The distribution of a solubilizate between micelles and the aqueous phase does
`not obey necessarily a simple partition law when a second solubilizate is present
`[ 48]. A non-linear increase in solubilization with increasing surfactant concen(cid:173)
`tration has been found with a second solubilizate present. Fig. 6.7 shows the
`change in micellar partition coefficient of o-hydroxybenzoic acid in the presence
`of increasing levels of benzoic acid when polysorbate 80 is the solubilizer [ 48]. At
`1 % surfactant there is a marked decrease in parrition coefficient, but at 3% there
`is little change. Nalidixic acid does not alter the micellar distribution coefficient of
`o-hydroxybenzoic acid but chloramphenicol reduces the distribution. Nalidixic
`acid has no detectable effects on the cloud point of the polysorbate soiutions,
`suggesting that it does not alter micellar structure.