Cocamidopropyl Betaine

Methods for producing CAPB and new analytical methods

for determining purity and side-product composition

B. Grüning, PhD, D. Këseborn and H.I. Leidreiter, PhD

Th. Coldschmidt AG, Esfien, Cermany

T
here are many detergent and foaming
mance than formulas without CAPB.4
surfactants available fortoiletries on the
The use of CAPB in efficacious, mild
market. However, only a handful of prod-
dishwashing liquids and oral hygiene prod-
ucts need to be considered to focus on the ucts are the latest examples ofCAPB's
raw materials that have a strong market. Of versa- tility. Its excellent toxicological
these, the most important are the anionic
profile, which has been documented by many
surfactants, such as lauiyl sulfates and
invesôgaôons and reports in the past few
lauryl
years, is one reason for its success in
ether sulfates.
dentifrice products.'°
A rangeofseoondaiy surfactants has been
Despite the widespread use of CAPB,
developed for use with these efficient and
oost-effective primary surfactants. In gen-
no oomprehensive documentaõon of its
eral, formulators use secondary surfactants syn- thesis, analysis and minor
to improve the properties of the primary oomponents ex- ists. By reviewing the
syntheôc process, we can also document its
surfactants and to opómize product perfor-
composiöon and by- products. We will also
mance. The most important secondary sur-
assess new analytica} methods, such as the
Actant is cocamidopropyl betaine (CAPB),
direct quantitaõve de- terminaõon of CAPB.
atso called coco fatty acid amidopropyl be-
taine.'"'17 Since its introduction into the Production Methods
mar- ket in the l9b0s, CAPB has bemme
essentia) in surface-active formulas.
AuiideJormotion: The methods for
pro- ducing fattyacid amidopropyl
Be•ieȚts: Combining primary surfactants
with CAPB reduces skin and mucus mem- betaines basi- cally follow a scheme that
bmne irritation,’ "”°'improves the condiôon- has shown validity for some time. Forming
ing properties of hair shampoos"" and pro- the fattyacid amide is the first step (Figure
duces a pleasant, smooth skin feel. 1-1); carboxymeth- ylaông the amide is the
Formulas that include CAPB thicken easier, seoond (Figure 1-2). For CAPB, the first
develop better foam and give a betłer step reacts 3- aminopropyl-
cleaning perfor- dimethylamine(DMAPA)with either fatty
acids, fatty acid methyl esters or directly
with natural fats (fatty acid glyceriri esters).
First presented in April l99ti at the Forum
Gosnieticuni in Salzburg, Austria. Pubhshed in
The predominant source oils used
Geman in PaJimerte tend fÒ'rrnefil, 77 ż44 (l98ö) — hydrogenated coconut oil and, occasions
ally, hydrogenated palm kernel oil — deter
 Reproduction in English or any other language of all or part of this anicle is strictly prohibited.
Voi )12.Februory
.'¥¥¥W0 Ø6tCCKAł•tl
1997 03ó1-4387/97/Œ02-ŒIó7S03.œ/ 19 7 Alluracl Publishing Corp. Cosmetlcs 8t Tolletrles• mogazIno/6y
BSTAINE, SyMTł-ÏBłs,ANALYsIS,
PURITY

/ormufocfdn cosmd Eco. en

 O CH mine the fatty acid composition of the betaine, which
II i corre- sponds to that of the oil. Table I -1 outlines the
R-C-OR' HON-CH CH CH -N typical fatty acid composition of CAPB.
-HOR’ CH To use DMAPA most efficiently and avoid large amounts
0 CH ofliijuid waste, you should recycle the amine, which is
II I partly distilled from the reaction medium together with the
R-C-NH-CH CH,CH -N metha- nol or water formed, during the transformation of
I fatty acid methyl esters or fatty acids.
CH
0 Auntie purr at • When CAPB synthesis starts from
II fat, glycerin forms as a by-product. As glycerin has no
R-C = fatty acid residue negative effect on most applications and may even be de-
R' = H, CH or derived from sired, a glycerin content of 2 to 3' o is norma) in betaine
glycerin solutions. Ifdesired, some glycerin can be removed from
Figure 1-1. Betaine production, amidation the amidoamine by phase separation, for a glycerin content
around ldc.
The DMAPA used in the amidation reaction, present in
excess, can be distilled out of the amidoamine to around
O CH 0.05% remaining. All amidoamines, whether derived from
II I !I fatty acid esters or fatty acids, contain small amounts of
R-C-NH-CH CH CH -N + CICH -C-ONa non- amidated fatty acid that are carried over into the
CH betaine.
Specialty f›etoines: Fatty acids and fatty acid methyl
O CH O esters with fatty acid compositions that correspond with
II I II natural oils can be used to produce betaines. Betaines with
R-C-NH- NH-CH CH CH -N(+) -CH -C-O(-) + NaCl
special fatty acid compositions can also be made, including
CH lauric acid amidopropyl betaine, caprylic/capric acid
amidopropyl betaine and betaines from topped fatty acids
Figure 1-2.Betaine production, carboxymethylation that contain no or only very few shortmhain
components.
Carboxyiriethylotion: The amidoamine is carboxy- methylated in aqueous medium with chloroacetic acid or its sodium salt.
The reaction leads directly to the betaine, normally obtained as a 30&o aqueous solution. Sodium chlo- ride, formed as a by-
product, is present in most betaine solutions at approximately 5' o. Apart from the reaction with the tertiary nitrogen atom, the
chloroacetic acid can also be hydrolyzed in a side reaction that forms glycolic acid. The glycolic acid content can reach levels of
more than ldc in the betaine
Table solution;
1-I. Fatty acidhowever, the level is typically considerably lower (about 0.1' o).
composition
During the past few years, of CAPBresearch on the production of betaines has primarily concentrated on improving the
carboxymethylation process. These efforts have focused on manufacturing purer products and minimizing by-products. Other
efforts have focused on increasing concentration
Raffle'aold W'efgfif % levels or reducing the water content in betaine solutions; these efforts all
{cfsa/n length}
primarily involve fine-tuning the carboxy- (appfiox vafuea} step.
methylation
8 7 Minimizing coritominonta: As the tertiary amines are
10 6 consumed during carboxymethylation, the pH value falls
12 49
14 19 considerably. This slows the reaction rate proportionally
16 9 because free amines are also removed from the reaction
18 10 medium by protonation as well as carboxyrnethylation. By
maintaining an alkaline pH throughout the reaction,
carboxymethylation can be carried out faster and more
completely, with less amidoamine in the final product.*
This technique can give amidoamine contents of less than
0.3&o in the betaine. Sometypes ofCAPB have
higheramidoamine contents, up to 3'Zo, for special
applications.
Chloroacetic acid and the dichloroacetic acid present in
chloroacetic acid in small amounts are both unwanted by-
products in betaines (as well as other zwitterionic or
ampho-

Figure 1-3.Flowablllty
é6/Cosmetics of aqueous
& Tolletrles• magazine betaine solutions Vol. T12. February 1997
teric surfactants, like ainphoglycinates) because of tlieir primarily hecausc aqueous betaine solutions form viscous.
toxicity. Cliloroacetic acid is almost completely depleted gel-like phases at slightly higher concentrations. In
during the carboxyinethylation reaction, but dichloroacetic contrast to the situation for sur actants like lauryl ether
acid is almost inert under the typical reaction conditions. sulfates, fluid mesophasesconsistingofCAPB,NaCl and
By subrnitting betaines to additional post-treatinent steps, water do not exisl at higher concentrations. The critical
tlie residual cliloroacetic acids can be reduced. This can lie concentrations above which non-fJowabIe gel-phases are
achieved either by reacting at alkaline pH or by additional formed partly depend upon the length of the fatty-acid
treatment with ammonia or amino acids that reduce die alkyl chains. Figure 1-3 show.s this
arriount of inonocliloroacetic acid." fnrsolutionscoinposedofstoichiometricainounts of fatty acid
Using sulfonating reagents is another possible way to amidopropyl 'etaine and sodium chloride.
minirnize the chloroacetic acids.°0 Final)y, both muno- and Numerous attempts have been made to obtain flowable
dicliloroacetic acid can be hydrolytically decomposed siin- betaine solutions with increased active matter; in some
ply by exposure to high temperaŁures (>120’C)." Levels of cases, by adding other surfaCtants. °’“3' Other ways to
less than 5 ppm chloroacetic acid and tess than 10 ppin achieve highly concentrated betaines include adding sol-
dicliloroacetic acid can be acliieved. There are, liowever, vents or using additional salts not normally contained in
vey rew products on the market that fiilfill CAPB purity betaine solutions, such as sodium nitrate, trimetliylglycine
requiremerits of airńdoamine content <0.3&o, monnchloru- (natural betaine: methanaminium, l -carboxy-N, N,N-
acetic acid <5 pprn and dichloroacetic acid < lU ppin. trimethyl-, innersalt)^ or nitrilutriacetate.
Sodium chtoride: Based on weight, the most iinportant A relatively simple way to uhtain more highly concen-
corriPonerits of tlie marketed aqueous CAPB solutions are trated betaine solutions is to adjust the free fatty acid
the betaine itself and sodium chloride. In generał, sodium content. By adding small amounts of fatty acid to the
chloride is left in tlie solution as it does not interfere with betaine with the amidoamine solution, you can produce
most applications. In fact, sodium chloride is desirable to betaine contents of 34-369'o and detergent-active
build viscosity in ready-to-use preparations, such as sharri substances, in- cluding fatty acids, of 36-38&o.’6
- poos. For special applications, there are also betaine prod- The increase in concentrations has an important
xn:te with rcduced salt conłents. These can be produced by positive side effect: such betaine solutions are
using solvents or niembrane separation processes. microbiologically stable and don't require preservatives.^
Flotuafzle CAPB eoncenfr‹itea: Most marketed CAPB By spray-drying aqueous betaine solutions, it is pos-
solutions cuntain approximately 309’o active inatter. This is sible to obtain highly concentrated betaine pro‹1ucts,
typically consisting of 80-85% fatty acid amidopropyl

70/Cosmetics & Toiletries" B•zine

Vol. 112, February 1997
 $gg mV

betaina

NaOH an
“ens
otte
d

25

Figure 1-4. Trtration of CAPB in

rnethanoVethyłane glycol menomethyl ether whh
perchlorlc acid in dioxane

betaine, 13-15a sodium chloride and 0.3 to 3.08b water.

The details of betaine production demonstrate that mar-
keted solutions will contain by-products. Determining the
types and amounts o(such incidental components helps us
characterize the betaine's quality.
Analytical Methods
Water feterminotiim: Karl-Fischer titration, as de-
scribed in several standard methods," is still the preferred
method to quickly determinethe exact water content. Using
modem titrators and reagents, water determination can
be carried out in a few minutes. Other methods, such as
oven drying at l05‘C or water determination by
extraction,’ are generally more time nsuming and show
a greater stan- dard deviation than Karl-Fischer titration.
In some cases a quick method to determine the water
content can be carried out witii (sugar-) refractometers, but
these instruments must be calibrated on products with a
known vrater eontent, determined by Karl-Fischer titration.
Fafty ncid amidopropył befoicie confenJ: Until re-
cently, the only reliable method to determine the cuntent or
fattyacid amidopropyl betaine (betaine content) in solutions
was using subtraction methods. Many attempts have been
made" to use direct tiŁration methr›ds, but, according to
our investigations, these methods lead to inaecurate
results. Many times the results are invalidated by other
protonable substances, such as free amidoamine, glycolic
acid, citric acid or trimethylglycine, during the titraóon
with acids, such as perchloric acid in acetic acid." Płatinga
et al describe a potentiometric titration with potassium
hydmxide in methyl isobutyl ketone. Tliis method is
relatively sluw; moreover, further interferences wit)i water
lead to a lack of aecuraey. Methods that include an
exłraction step can yield luw walues due to the short-chain
fatty acid amidopropyl betaines heing
only parfially measured.
lteccntly a inodified titration method has been propused
thata1lowsadifferentiatingtitraóonoffattyacidamidopn›pył
betaines by choosing tlie proper solvent. The basic ap-
proach and results, explained iii detail in reference 13,
am reviewed here.
A inixturc of methanol and ethylenc gjycol inonometliył
ether serves as the solvent for the CAPB. Tlie CAPB is
72/Cosrnefłcsg Tołkzłi1es* rnogazine VOI. 112, February 19g7
 re1ativestandarddeviañonof0.4&.Fromthechlorideconcen-
tration, the Sodium Chloride content is calculated.
Glqcerim Aspreviously mentioned, glycerinis formed
as a by-product during amidoamine production starting
from triglycerides. The most oonvenient way to
determine the glycerin in CAPB solutions is iodometric
titration .’ Com- pounds with vicinal OH-groups are
cleaved by periodate, then the periodate is reduced to
iodate:

CH OH-CHCH-CH OH 2 HCHO + HCOOH

+ 2 NaIO, —+ + 2 NaIO + HUO

The iodate forms free iodine with the addition of iodide.

The free iodine is titrated wlth sodium thiosulfate
using starch as indicator.
Free/ot?y ociz£- The problem of determining free fatty
acid concentration in CAPB solutions has still not
been completelysolved. Aspreviously described, the gas
Flguæ 1-5. oetermlnation of amidoamlne by HPLC chroma- tography (GC) method" extracts free fatty acids
(peak at 7.73 min) from the dried betaine using diethyl ether. After elution
through a silica gelcartridge, theeluateis neutralized with
tetramethyl- ammonium hydroxide in the presence of
phenolphthalein; this is then directly injected into a gas
chromatograph. The formation of fatty acid methyl
esters talres place in the injector of the gas
chromatograph. Pentadecanoic acid was used as the
internal standard. Apart from problems with the time-
50.1
consuming preparation of samples, this procedure has also
inexplicable problems with the reproducibility.
20.9
At this Sme, investigations are underway to
determine the free fatty acids using a new, high-pressure
hquid chro- matography (HPLC) procedure. This
8.8 ss CB.8 IS.8 28.0 2S.B -
method derivatizes the fatty acids with 2-
bromoacetophenone and identifies them by reversed-
phase (RP) HPLC analysis.
Paak at 10.52 min: dichloroacetlc acid The total free fatty acids can also be determined
13.77 min: glycolic acid
IB.18 min: monochloroaoetic acid
with nuclear magnetic resonanoe spectroscopy.
Many acidamidoamine: A variety of methods have also
Flgure1-6. Determlnation otmonochloroaceticacld, been discussed to determine residual amounts of free fatty
acidamidoamines.1'" Titration and chromatographic meth-
ods have been investigated most extensively; Käseborn
ex- plains in detail why existing ôtration methods
adjusted to its pure zwitterionic form with a sodium
cannot be satisfactorily applied to alkylamidopropyl
hydrox- ide/sodium acetate buffer. The titration uses a
betaines."
perchloric acid solution in 1,4 oxane. The titration step
Thin layer chromatography has shown to be a sucoessful
transforms the betaine into its cationic form. The choice
semi-quantitaôve method to determine free amidoamine. The
of the special methanol/ethylene glycol menomethyl
silica gel plate can be developed in a solvent mixture of
ether solvent mix- ture allows an almost selective titration.
chloro- form, methanoland ammonia(30/50/2), and the
The titration curves obtained can be evaluated by modern
amidoamine is detected by means of a bromophenol blue
titrators without problems (Figure 1-4)." Acoording to
soluôon.
our experience, there is a relative standard deviañon of
Amidoamine can be quantitafivelydeterminedby HPLC
approximately 0.69b using this method. without any problems using a special ion exchange
Gerhards et al have discussed alternative methods, in-
eolumn.• A 0.02 M phosphate buffer solution serves as
cluding titration with ion-sensitive electrodes, in their
eluent. Under thèse conditions, the amidoamine can be
pub- lications. " detected as asingie peak, not separated according to the
Sodium chloride determ - Determining the chIo- fatty acid composiôon. Figure 5 shows an example of
rideoonoentration usingpotentiometictitrationindilute nitric such a determination. This method is currently being
acid with silver nitrate is one of the most reliable
discussed by committees of the DGKbandGAT.*
analytical methods applied to fatty acid amidopropyl
-productocida: Becauseof tlietoxicologicaleffects of
betaine solutions.' The precipitation titration can be carried
out reliably with a 'ProTens: Dr. Herhst Haudelschemiker, Uörtadt Weiher, Germany
°Deutsche Gesellschaft ftir wissenschaRhiehe und auuj
y4/Cœmetics & Tolletrles^ magazine
angewandte Kosmeök (Fachgruppe AuaigtiJe)
•Gemeinsehaftsausschup Teugide der Oetsehen Gesellschaft ftir
Feltwisseuschaft und DIN

Vol. 112. FeDïuary J997

chloroaoetic acids, researchers needed to develop new ana-
lytical methodswith lower detection limits than themethods
previously used, mainly ion chmmatography with suppres- Vot ll2,Feb uogl997
sor techniques or capillary zone electrophoresis.
For the GC methods, according to Cetinkaya° and
Arens andSpilker,'theorganicacidsareoonverted tomethyl
esters and are detected using an electmn capture
detector.
HPLC detection can be carried out directly with
little sample preparation other than dilution on an
anion ex- change column.‘The elution is achieved with
0.01 M sulfii-
rieacid;thedetectioniscarriedoutwithaUVdetectorat20S
rim. The detection limits are approximately 5 ppm for
monochlomaoetic acid and lO ppm for dichloroacetie
aeld. With this method, glyoolicaeid can be measured at the
same time. Figure 6 shows a corresponding
chromatogram.
The GC methods have proved more reliable in
inter- laboratory tests, but the HPLC method turns
out to be oonsidembly less time-consumingand more
universal, sinoe it detects all three organic acids (glyoolic
and mono- arid diehloroacetie) simultaneously.
DMAPA Residual DMAPA is normally determined in the
amidoaminebefotebetaineproduction.Witharecent1ydevel-
oped method, however, it is also possible to detect
free DMAPA in the finalbetaine.'In boththe
amidoamineandthe betaine, DMAPA is analyzed by RP-
HPLC, with UV-detec- tion after derivatizing with phanyl
isotliiocyanate. This sensi- tive method is especially suitable
for the detection ofvary low

‘nminex HPX-87 H, Blo-Zad Laboratories, Munich, Germany

 concentrationsof DMAPA. In betaines, DMAPA
‹xxnirs only in the lower ppm-level. typknlJy from less
than 5 to 20 ppm. safe ncid compoeifion•With the
help of the analytical methods discussed above, it is Cosmetics & Tolietttes* magazine/75
possible to characterize the quality of the process by
which the fatty acid amidopropyl betaine was
produoed. The fatty acid composition is, how- ever, a
characteristic of the fatty raw materials used. There-
fore. the fatty acidcomposition of the raw material
used, not
the CAPB. is normally analyzed.
It is also possible to characterize the product by
deter- mining the fatty acid composition of the
betaine. Tlie free fatty acids can be obtained alter
cleaving the amide bond by heating in conoentrated
hydrochloric acid. After their con- version into methyl
esters by imown methods (with metha- noUsulfuric
acid, for instance), the fatty acids can be ana- lyzed
by gas chromatography.
These sample preparations can be avoided in two
other HPLC methods."" Using these methods,
aikyl distribu- tions can be obtained directlyaAar
sepamtion on areversed phase co1uzrin(RP 1&'7
pm)withmethano1/water(detection by UVat2l5 nm)
or usingion exchange chromatography. In the laNer,
the order of elution is reversed compared to the RP
chmmatography, with the longest-chain betaine
eluted first followed by shorter-chain derivatives.
Conclusion
The described analytical methods for the
determina- tion of the main components (water
and betaine), the by- products (sodium chloride,
glyoerinand fattyaeid) and trace
 components (such as fatty acid amidoamine, glycolic
acid, mono- and dichloroacetic acid) are reliable and can
be routinelyapplied.Thus the production ofCAPB and
related marketed forxns of betaine can be carried out in
acontrolled manner to ensure consistent high quality. If
required, these analytiea) methods can be supplemented by
modern HPLC methods to determine fatty acid
composition or by IH- and 'W-NMR spectroscopy. The
NMR methods repeatedly prove helpM to recognize
unexpected additiona) compo- nents; however, trace
components caimot be detected by NMR spectroscopy.
The further development of analytical methods is an
essential requirement for the continued imr vement of
CAPB production methmls and its quality as a marketed
ingredient so the cosmetic and detergent industries can
be oFered CAPB products that satisfy the highest
demands of quality and technical properties.

Address correspondence to B. Brüning c/o Conmetióa a

Yo/fetrfas magazine, 362 South Schmale Road, Carol Stream, IL
6018&2787 uSA.
1. M Arens and R Spilker. FETT/CAT BO 212-214 (1995)
Z. T Bóhmer, pri 'ate communication. submitted for publication.
3. Id Ce fiaya, Parf Kosni W 816 (1991)
4. Y Cf›evalier. £angmuir7 848 (1091)
s. osF — G»heitsmethoden H-III 3a (02), DIN 51777, ISO 4317,
DGK-AnaIysenmethoden W 010.1
6. DGF — Einheitsmethodan H-III Z (92). ISO 4318, DGK-
Anatysenmethoden W 011.1
7. DGF — Einheitsrnethoden H-III 9 (62)
8. DGF — Einheitsmethoden E-III 3b (79)
9. XD¢xtdngo,BM PhIIIips,JSanchez LeaI,TGarÖaandAGarbayo,
/r›tJ Gasmet SÖ13 26S (1091)
10. dGDominguez,FBaIaguer,JLPanaandCM PelajeroGoA'/scftm/l:#
/rdorzri/ort55 10 (1981)
J3 Donlcerbroek and CN Wang, Pfofi 2nd Warja fiuzfacfants
fiöngress Pafis v I\\ (198B) p 134
Fasal report onthe safety assessmentofcocamidopropytbetaine,
Ja» c»/i rax 1o sa (fest)
ta. R Gemards, I Juscofie, D KAaabom, S Keune and R Schulz,
Tenside Burt Det 33 8 (1996)
14. B Gr0nk+g, Proc3›dCESIO IntSuzlfictant Cfi›ngiess, v 3(1592) p
Z21
15.TJ Hall-Manning, GH Holland, G Rennie. P Ravetl, J Hinres,
DA
BasI‹etterandMDBanatLProc4fhCESIOIntfiurfacfantC‹x›gzess,

16. K Hoffmann, Jshrbuofi fifir den Pfakfiker. Me‹fi›g I c em Industúa

H. Ziogcows y (1973)
17. R HMtinger. Galcls ma infimnierl SZ so (1975)
18. D K&sebom, 7ego Cosmetïcs Spec/a/2
19. HI Leidzeiter, K Jenni and C Jorbandt, SÖF-W020 8fi2 4994\
20. JM PIaünqa,JJ Donkerbroek and FLlMulder,JADCS7067(1083)
(ptA¥shed by the Am Oil Chem )
2s. W Seidel. Parf f¢osm 71 712 (19g0)

Z2. V Bade, DE Z9.Z0.479

Z3. V Bade, DE 36.13.944
24. V Bade, DE 37a.S22
Z5. V Ba¢Io, DE 38.26.654
Z6. U Begoihn,W Foitzik, B Grúning, D Kdsebom and C
Weitemeyer, DE 4 07.3Bfi
27. WFótzgc,BGr0nk›g,DK8sebomandCWeitameyer,OE42.05.B80
28. J Hamann, FU Kóhle, K Stark and W Wehner, EU 656 346
ZS. T k4essenger, DE Mather and BM Philllps, US 4.243,549
30. H Seitz and R Vybiról, DE 42.32.157
31. G Uphues, P Neumann arxl A Behlar. DE 45.40.423
O GUphues, UPk›og, K Bschof, K Kenar and P SaldBk, DE .

76/Cocmettcs & Tolletrles• mogoztne

Vol. 7 12. Februory 19' 7