CA2223981A1 - Process for preparing amides of n-alkyl polyhydroxyalkyl amines - Google Patents

Abstract

N-alkyl polyhydroxy alkyl amines such as N-methyl glucamine having a Gardner Color of less than 1 are reacted with sources of fatty acyl groups such as methyl esters, anhydrides, and/or fatty acids that have greater than 98 %
transmittance at 460 nm in organic hydroxy solvents such as methanol to prepare N-alkyl polyhydroxy amine amides with good color. The N-alkyl polyhydroxyamines can be purified by crystallization, and/or subjected to reductive bleaching, to provide superior color. The reaction is preferably carried out at low temperature for short periods of time and with low catalyst levels to minimize formation of cyclic products. The resulting amide product can be further purified by treatment with anionic and cationic exchange resins to remove soap and amine impurities. The anionic ion exchange resin can be readily regenerated by acidifying it followed by washing with an organic solvent.

Description

W O 96/40619 PCT/U~GJ~7l24 PROCESS FOR PREPAR~NG A~DES OF N~ YL POLYHYDROXYALKY~
A~ES

lS FTF~n OF 1H~.INrVEI~IION
The present l~lio~l relates to an i-~-~lu~d process for ~ 8 amides of N-allyl polyllyd~o~y~llyl amines, ~Q~ret;~lly ones having good color and low levels of undes.l ble by-products.
BACKGROUND OF lnl~ INV13NTION
The m~mlf~lre of N-alkyl polyllyd~u~y~Ltcyl a-m-ines ~N-alkyl polyl.ydrc,~y amines), such a_ N-methylgl~ -r, and the fatty acid amides thereof, has been known for many years, and such m~t~ le are ~ ~b'e CQ~ ;a11Y. Recently, there ha been ocç~Q;~n to employ N-aLtcyl polyl-yd-u~y amines, e.g., in .r~ u~Q with, e.g., ~tty scid esters to pl~pale fatty acid polyhydlu~ amide dcl~ nte for use 2s in rlDA~ ,R products. It has been ~etf - .~ d that care Jnust be taken in ~ N-all~yl polyLy~l~osy amines and amides to provide the amides with the best color.The present i.l~l;on affords access to high quality polyhy-Lusy fatty acid amide sud~ct~. The present U1~nliOn provides means for pl~&ing N-allyl polyl-ydlu~y amine amides which are almost water white, esper~ y amides of N-~ lL~ ur~ lle in higb yields, and more ~pe~slly, amides of N-methyl~1u~ e co~ g low levels of cyclic m~tPri~ls as h~ closed BACKGROUND ART
~ cose ledù~ Am;rl~tinn proces~es are tlicrlosed in U.S. Pat. No.

2,016,962, Flint et al., issued October 8, 1935.
3s U.S. Pat. No. 1,98S,424, Piggott, issued Dec. 25, 1934, r1i5AlOses U~ 8 ~tex~le ~ by l~&CLi11,~ (a) the product of heating glucose and a~1UeUUg m~h~1&I1IInC in p~..~.lce of hydrogen and a hydrogc ~ catalyst under W 096/40619 PCT~US96/07124 ple..;,.ll~, with (b) an organic carboxylic acid such as stearic acid or oleic acid. The conf~ cA~;nn product, prepalcd at about 160~C, is said to be Npredc~ y, if not e,~,L~ ely, an amide" and is ass~.lcdly of the finrnAlllAA R-CO-NRl-CH2-(CHOH~4-CH20H wLP~in R is an allyl radical c~ g at least 3 carbon atoms, while Rl is s hydrogen or an allyl radical.
U.S. Pat. No. 2,016,962, issued Oct. 8, 1935"licrloses a process for pl~u glllcA...~ s and related products.
U.S. Pat. No. 2,703,798, ScLv~al4 issued March 8, 1955, asserts that compositions produced by l~a_lh~B fatty acids or acid anhydrides with N-10 alkylglll~-A~ s (p~,~ hly such as the process as taught by Piggott) have poorcolor and poor dete.~.-~ p-up~,.Lies. Thus, Scl-v~_~ teaches problems acsQc;stedwith fo~ g the crn~l~ncAtisn products of N-mcmnAlkyl~llrA~ Ps and fiatty acids, with respect to unde~ ~le color ch~cle~ ;~l ;rC~ and dcL,.~n~ prop~ lies.
According to S-,l-w~4 a~plo~;~..AI- Iy equim~lAr l lu~JulliùlL of N-lS mnnnAlkylglll~ . es can be reacted with fatty alkyl esters by heating at 140~C-230~
C, plef~ bly 160~C-180~C at normal, reduced or S~C Al~nS1~k-A-;C plC.~ S for a period ~sol...i~l~ in excess of one hourN during which time two initially ;~ c--;ble ph~ses merge to form a product said to be a useful dete.~,~l.
~llitable N mnncalkylgll~c~ es are ilh~ cd by N-methylg1-~c~ e, N-ethylglllrAmirle~ N isoplo~ llrAmins and N-butylglllc-a-minP~ Sllit~ e fatty allyl esters are illu~ ,d by the product of 1e&L~ , a C6-C30 fatty acid whLh an -lirhA~Lic nhnl e.g., methyl ester of lauric acid.
More recent ploce~_s include those desr-;l~ed in U.S. Patents: 5,334,764, Schi~Pl Connor"ehllm_tP~ and St. Laurent; 5,338,486, Cnnnnr, .erhP;1~pl~ and Kao;
2s 5,338,487, CQnnclr, !Srhe;1~Pl and Kao; and 5,380,892, Connor, Srh~o;bp1, and Kao, all of said patents being l~c~olahd herein by lGr~ ce.
Accord;l.~ to Thomas Hedley ~ Co. Ltd. (now Procter ~ Gamble Ltd.), British Pat. No. 809,060 p-~bli~h~d Feb. 18, 1959, the c~ .pc,~ c made by the process herein are useful as ~--- r, ~ for laundry d~t~ ls such as those having granular form.
Hildreth (supra) ~ use of the cc--.l~o~ e herein in the blc-' -m;ctry field as det~ agents for ~ hili7i~ plasma ~~ cs and EP-A 285,768, ~ l.r~
Dec. 10, 1988, d~ec.;1,rs apr~ ofthese co~ u~ as a !1~ 1. Thus, these c~ -l-o~ , or co~--l-o~ n~ , them, can be highly des.._~le ~
Yet aao~ ,r process for making co---pGs~Lol s co--~ g the amide co---~ùuilds 35 of this ~ll.~.Lol~ is inrl lded in the ~o~ d ~ los ~- ~; of u~p~uv~d l~
See EP-A 285,768. See also H K~lLL ~ , Tenside ~c---r,~ D~lL~ g 25 (1988) 8-13, inter alia for e~rlitir~nA-l AicrloCllr~ of ~-uces~s for maldng N-W O 96/40619 PCT~US96/07124 a~1~ Jc~ Fs All of the above patents and publie~tionc are h~co~G~led herein by ~rtl~.,c~.
SUMMARY OF THE INVENIION
The present de~lop..le~ relates to a series of uul,lo~ ,.c.lls relating to l,rocesscs for ~ palU~g amides of N-allyl PO1Y1~YdIUA~ amines ~N-sll~yl&l~lo polyols). Both the N-allyl POIYII~tIOAY smines and the source of fstty acyl groups, e.g., esters, used to foml the smides sre sele~led to have good color, the ~erl~tinn co~ ;o~5 are sFI~cled to avoid the formAtinn of color ,.~ c snd pr~u~ for color ~ .;A1c, and/or the smide product is treated with an ion e~ ~G resin, 10 ~I~Lw'cs of ion ~ ng~ resins, or cc ...h;.~AIiOnc thereo~, andlor a redur~ing "bleach"
to ~c~ , the best color amides. The CO~ 1 Q" of all of the Illly~ .l.e.lLs is .~,quu~d in order to achieve amides with the very best color for f~ 3 d~,t~w~t co..~o~ ~isnS~ especi~11y liquid dete~g~.lL comrositinn~ that are ~water white" and which contain low levels of cyclic rnAt~Ale lS The iu ;ol- provites a process for ~.ep&.. ~g polyl,~o~ tty acid amide ;,~.. r ,~ , c,o ~p i- ~ r ~ Lul~, a . . n l-"l sF1e~,~ed from the group co~ g of fatty acids, fatty acid anhydrides and filtty acid esters, Psper;s1ly fatty acid esters, hs ling greater t~an 98% hAn~ re at 460mn with an N-al~,ku,uuo polyol having a Gardner Color of less than 1 (<0.1 abso,l,~lue at 440nm), e.g., methyl esters orhi~e~id-s. CrystA1l;7Ati~n of said N-a11~1&1luhlo polyol csn be used to provide the ap~.~.plla~e pudty snd color. N-all ~ nu.o polyol vnth this Gardner Color is ~stable"
for three houas st 130~C. The N-aLI.~l&.uno polyol i_ C4'-8~ ,d stable if it has a Gardner Color of 4, or les~, a~er three hours under these ~n 1:l;n~8 A less purc N-all~ u~o polyol will be a dark brown afcer three hours under these Co~ ;o~8 2S Also, in o~der to ~ the be~t color amides, the d~ ~;on of N-all~12u.uno polyol should be carried out at b~ en about 110~C and about 160~C for a period of time of les_ than about three hours, more preferably at a l_.u~c.~u~, of from about 120~C to about 140~C for a pedod oftime of les~s than about one and a halfhours and even more p,_fe.~bl" at a t~,.ll~al~f of from about 130~C to about 135~C for a period of time of less than about one hour However, for co~ ~ f c;dl practice, good results can be ob~ d with de~d~a~l times of from about four to about eight hours, ~ bly from about five to about six hours, to t~,CC ~d~te CO~ c;al e~l~ p~ mit~tion~ A more pure N-al~ U.O polyol can be achieved by cryst~lli7~tion from an P~lleollC sol~ltion, either with, or without, an organic solvent 3S present The deh~.t~l N-all~yl~l~ulo polyol is then rea~ted with, c.g, fatty acid esters and ç~pecially l~ ce~ides~ to form fatty acid polyll~ Ar amide sur~ctants.

The reS~lting polyLydlo~y fatty acid amide ~~ ct~nt is then post treated with an ion ~ A~e resin, rï~ixture of ion e-~ ge resins, or CO~ ;On~ of ion e resins, and/or reduçir(g bleach such as NaBH4, etc., or Ly~o~ ;nn over a catalyst, as taught he.e~l~ler, and, optionally, co~ nc of ll~.a~ A
5 particularly c~ e post 1.~ is the hydro~en~tion of a sc~l~ti~-n of the polyl,~d.u~y fatty acid aïnide s~ ot~nt over a hydrog~n~tin~ catalyst like nickel, pS~n5~ m, copper cl--u-- ile, etc.
In a p.~ ,d process, the fatty acid ester is a Clû-Clg alkyl or alkenyl fatty acid methyl ester, or, I~;~ly~,~,.ide, and the N-alh~,k,,--u.o polyol is sf~ cl from N-0 methyl gl~lc~mine, N-methyl fmct~mine~ N-methyl In~lt~min~ and N-methyl glycerol amine.
DETA~ED I)ESCR~PTION OF 1~ INVENTION
The p.uces3es of this invention employ s~le ~e i reactants, N-alk~l~-~.o polyolsand sources of fatty acyl groups, with good color, espe~i~lly color that is he&t stable.
lS The ~colorN .'.f~ d to herein is the Gardner Color, e.g., of the N-alkyl&.. no PolyoL the N-alkyl&..-ino fatty acid amide, etc. NGardner Color" is the s~du~l Gardner ~ ,.--~-L known in the art. A Gardner Color reading near zero (sol~fion) ~ ,s_lt~ a nearly color1~c (Nw&ter-whiteN) s~ ti~n Gardner Colors below about 1 are ~ u~cd for the N-allylarï~ino polyol re~ct~nts, and it is pl f~ d 20 to have Gardner Colors dose to 0.
Gardner Color is d~t~ ed by ~O.C.S. (~ e ~c~ Oil ChPmict~ Society) Offlcial M~thod to la-64, entitled COLOR Gardner 1963 (Glass S ~ d~ds) est~hlich~d 1978 and revised 1982. The e~ p--. -l and ~ d~ds for de~ --A~g Ghrdner Color can be y~ d from Delta S~ntifi~, Box 5728, Long ~a~ New 2S York 20014, or from Gardner ~ lo-.~, Silver Spring, ~rylaD~ U.S.~ As used herein, the Gardner Color limits typically refer to the color ~ from the color bodie~ that are present, or which are the result of the de~ l,ed ~-1 ~tinns and not to d~ ldy added color ~
The odor cl~ ~'e~s of the N-alk~l~u...o polyol ~~4 and it~ amide, are 30 s~ lly free of amine or "fish" type odor (once any exces~ N-alL;I~u.~u is ~,,ov~d) and also i,~,bs~ lly free oftypical bro. ~lU~ sugar odors.
~he N-alh~l&..uno Polyols S ''e N-aLt ~,l~.uno polyols can be p~e,.~d by IJ-UC ~ C5 similar to those dr~ -ed in COlJe~ld;1113 U. S. Patent Appli~tir~n Serial No. 07/907,382, filed July 8, 35 1992, in the name of Junan Kao et al. for PROCESS FOR PREPARlNG N-ALKYLA~IES lN AQUEOUSIHYDROXY SOLV~TS, said ~ppl;c~ m being ~co.~o.~ed herein by .. f~.~lce, e~pe~lly page 6, line 4, to page 23, line 3 and WO 96/40619 PCT~US96/07124 EXAMPLES I - VL and IX - XIV. The pol~ .,y amine used to form the polyl.~d.u~ acid amide can be made by any process that wiU provide the desired colûr.
As ~ e~ he.~ n- ~, N-all yl~l~O polyols with good color are achic~.,d by 5 careful 5rle~ n of reaction cQ~.tl;,;c ~-c The reaction for the ~le~alaLiûn of the N-alkylamino polyols (also le~.-C~ to herein as "polyl-~Lw-y~-l"~es" or "N-alkyl polyl-~llo~ aminen) herein can be termed the ~R-l~ re~cti~ n, and is illustrated by the for~n~tit~n of N-m~Ll.ylgl.~ min~, ~I.e~
Rl is methyl.
10 Adduct Process In this f~rst v ~ ;QI~ of the R-l re~ctinn, the process involves pre-reacting the amine and red~ g sugar to form an adduct.
water and/or organic solvent, e.g., ~ h~o RlNH2 + glucose ~ Adduct + H20 The Adduct has the r, - ".. l~ (I) as foUows:
O
RlNH- ~- (CHOH)3 - CH- CH20H
catalyst Adduct + H2 ~ RlNHCH2(C~O~)~lCH2OH
The ~e~,lA~, solvents and catalysts used in the R-l l~ rtion are all well-known ...~f ~;~ tho ~gh not usually used in such ~ ied fonn for maldng det~rg~.~t ~--- r ~ 3 and are available, at least in some form, from a variety of CQ~ sources. The following are nn~ g . i'es of ~sterisle which can 2s be u~ed herein.
~ min~ M~ter~sJ - The '~N-aLt~ ~e~ used to form the N-aLIcylsmino polyols include p-~y amines of the forrm~ls RlNH2, wl.~ Rl is, for . , 'e, aL~cyl, e.g., Cl~1g, ~osre~slly Cl-C4 alkyl, or the co--- ~o~.~l;-.g l-ydnJ~.~ llcyls, e.g., Cl-C411~ A~ . F - ~tor include methyl, ethyl, propyl, l~dr~ and the likc.30 No~ p examples of amines usefi~l herein include methyl amine, e~yl amine, propyl amine, butyl amine, 2-l~ydr~ I'up~l amine, 2-L~Lo~y~ roy~ 2-L~ l amine; l-..~ell.o~.ol ~1, and methyl amine. The C1~3 aL~l~.llncs arc p.~-_d, and N- ~-~ e is most p ~f,.-~d. All such amines are jointly ~~
to herein as "N-alkyl amines." The amine can be either a~Luus or in a solvent, 3s e.g., s;~l~,c~ e solvent, of a CQ"C.. ~ ion offrom about 30~/ to about 90%")~fe. b3~, from about 40% to about 70%.
Polyllydlo~y Matenal - A p~ ;lled source of pol~,L~droA~ m~t~n~le usefill in W O96/40619 PCT~US96/07124 aU of the R-l re~ctionc co~ ,.;ce recl~lcin~ sugars or re~l~ring sugsr de~ivali~.cs. By "sugars" herein is meant re~l~ring sugars such as glllcosP, fructose, ~ n~ se~ lactose, m~ltose, xylose and the like. The term "sugars" herein also ine~ Pc glyceraldehyde.
Such "sugars" can include m~tPri~le which break down to form sugars, such as plant syrups such as cane syrups, corn syrups, potato starchderived sugar syrups, hydrolyzed wood pulp-derived sugars and the like. High ~uctose, high glucose andhigh m~ltose syrups are cc~ l and "1. r~ d, ~Cpecislly if their Gardner Color is S~ticf~r,tQry. The l~kul~ sugar m~te,n~l COlll~li~S, for this first v~ristion, an adduct with the amine such as ~ lh~l~luu-c. The species are clet~ -fd ( ~-e~-cd) by g.c.
0 analysis, (gas-liquid ~,lu~ graphy or "g.l.c.n) using Hewlett-Pach.. l 5890 Series 2 on column i~;e~l;on using DBl 15 meter 0.25 m film thickness n~ 250 m.
A particular alv~ c of the "Adduct" process is that the "Adduct" can be formed in the p-~s~ince of water. Acco-dingly, raw m~teri~l$ such as corn syrup, and the like, can be used as the sugar source. However, the sugar sol~ti~n can be lS plcp~n,d from g-~u.,llar, powdered, etc., sugar by dissolving the sugar in the solvent, prert.~bly ~q~eovs solvent. CO~c~ alions of sugar in the solvent, e.g., water, are typically from about 40% to about 90%, p~,f~ly from about 50% to about 70%
(TypicaUy, 71% is the upper limit.) It is highly uu~oll~ that the color ofthe starting sugar m~tf~ri~l, for plep~u.g, aU N-ah~l~-u--o polyols, be l~s than about one on the 20 Gardner Color scale, p.~,f~;~dl)ly less than about Gardner 0+, and more preferably about water white. Typical color mgtP islQ that are present in the starting sug. r mgt~riglQ neg,&L~ affect the catalyst . nd the .~Lon yield. The_e color ~
Iso co--' ~ ibule to the ~,~-lual color of the N-all~yl~.lu.o polyols. Such colors can be le.~u.~, if present, by proce.l~ such a "carbon ble~ B " in which the color 2s msteri~lQ are adsorbed. The sugar ~gterjgl i_ p-~,f~, bly hsn~llP~d wi~lwul e heating and/or under non~ 3 contlitinnQ to prevent degradation.
Of course, use of sug. rs having low Gardner Colors (e.g., 0 or <1, i.e., water-white syrups) to forrn the N-. ILyla~lfu~O polyols will help ensure that N-, Lkyl~ o polyols having desirably low Gardner Colors will be plù~ ced Stated otherwise, use 30 of low (~1) Gardner Color wgars (pl-,fe.ably white solids or water-white sohlti~nQ) and use of the l_~ioll so~ -e ~liQrlosed herein results in low Gardner Color N-alkyl~ulullo polyols.
~ stsl~yst - A variety of hydrog~c .~I;nn catalysts can be used in the R-l r~_Lon.
Tn-1~,ded among such catalysts are nickel (~ d when treated as ~
35 h~ller),pl ~ u ~,p~ m,iron,cobalt,l~ ,varioushydrogen~ti-nalloys, and the lilce. The catalyst used in the hydrog~c ~ step is p~ r a p~ ,ul~e nickel catalyst, Raney nickel, nickel, other nickel catalysts affi~ced to ~I,;.~ e W O 96/40619 PCT~US96/07124 m~tPri~lc such as silica or ~ min~ Catalysts which are easier to remove (e.g., by filtration) are plc;r~lled. Highly prere led catalysts herein cc--~ e "United Catalyst G49B,N HUnited Catalyst G96," and NUCI C46N particulate M catalysts supported onsilica, available from United Catalysts, Inc., Louisville, R~ u~L y, and Raney nickel s type catalysts from W.R Grace & Co., of R~ltimore~ Maryland, such as R~4200 and RA3 100.
Acll-c~ g good color also r~lu.l~,j o ~ 3 and 1llA;IIIil;n..~3 the activity of the pl~f~.led nickel catalysts inrlutlin~ any of the cG,-~ ~I;nn~l Raney nickel or ~uppGlLed~ nickel catalysts well-known in the art. ConvPntinn~l nickel under thelo tra~len~rlr RAN~Y NICKEL 4200 and 3200 (Grace Chemicals) are quite suitable for use herein. UCI (IJnited Catalyst, ~c.) G-96B and G49B and G49C are also r ~ -~1c W-lth respect to the nickel catalyst, it is beli~d that ~llOV..1~5 oxides of nickel from the catalyst ~r~ or ;--~l~edes ~ ~tl~tinn of nickel ions into the r~;libn mitieu, and thus results in the rO----~ of l~aclioll products having a lS dc~ bl~ low nickel cont~nt More~ r, it has been found that thc nickel catatyst pre-treated and ~ ,f~,l~ly post-treated with p..~ Pd llydr~en can be re-used in multiple ~bse~lv~ ~On~, thereby yielding a ~~I.s~ overall cost savings. In general, nickel c..~ ;.4 such as those that are CQ ~ c;&lly available, typicalty are co..~ ~ with, e.g., oxides of nickel, organic m~tPri~le, excess caustic, and/or 20 ~ min~ fines, P~rer-~lly after sh;~ ~ and storage. The nickel catalysts that are used in the processes herein are pl~,f~ly free of catalytic activity ;~ 9 .~ s ofnickel oxides, organic m~teri~l~, caustic, ~lllmin~ fines, etc. Tlie.. rO.." it is dcs.- ~le to wash the catatyst with one, or more, solvents to effect l~ Val of o.~ cs and/or water-soluble m~tPn~l~, to preferably lower the p~ and/or treat the catalyst with a 25 strong r~J~,e;-~g agent, c.g., h~ g~n gas under high ~ , and/or t~ Lule co~ , to destroy, or remove, the nickel oxides. Once the catalyst is "c~ nP~ "
the cataly~3t is d~ d under non-~ 3l,1s ~~, e.g., I~l .,gen gas, or, more desirably, a leJ~ gas, e.g., hydrogen. Any ~ to the normal .h-.e should desirably occur for 021ly short periods of time and while the tc.lIp~ lun~ U lOW The activity of the catalyst can be incl~d ~ A~.l.Ally by the re~lllctin~l, or removal, of these ;~ " ;1 ;p~A" even when they are present in very small ~ ,9~ The res~llting catalyst also provides amines, and ll-e.~,fc,l~ amides, with good color.
~ When the nickel catalyst is in contact with either adduct or N-allyl 35 pol~ ~o~Ollyl amine, the h~Logcn plc~ should be ~--A;--~ F~ to r..~:... 7 catalyst ~ t-;l;~ n ~ rly~ ahigh hydrogen pl~ e.g~, from about 100 psig to about 3500 psig, ~ f~ably from about 500 psig to about 1500 psig, and a t~ c.~Lu~e of from about 20~C to about 135~C, y~ere~ y from about 40~C to about 85~C, wiU reduce the level of nickel ion dissolved in the N-alkyl polyl-~ uAy~lkyl amine, and, by depositing the nickel back onto the catalyst, l~e.~e.~le its activity s A co ~ of hydrogen gas and s~ cled p.~ .. pe.~lui~ cnnrtitinnS
can reduce this ~ hiti7~tion and, in fact, reverse the process to deposit nickel and eg_nc.~e the catalyst Low~ g the soluble Ni content in the N-aLkyl POIYIIYdIOA~
amine product to less than about 10 ppm, prertl~bly less than about 5 ppm, more preferably less than about 2 ppm, wiU cfIt-;L~ l~.,n~ ~Le the catalyst.
0 When the catalyst is sepal~l.,d from the N-alkyl polyll~d~ûAy~Ucyl amine, the t~.llp~ila~ul~ should be less than about 135~C, pler~ bly less than about 85~C, and the separation, typicaUy filtration, should be ~cco~ ed under hydrogen pl~
Regen~.alion of catalyst can be achieved using the ~tep d~CC ;l ed for initial activation lS The N-aLyl.Y---no polyol .~_~-L herein, which is "~ y free of nickel", cc~ c no more than about 20 parts per m~llion (ppm) nickel, and pr~,f~.~ly less than about S ppm nickel (Ni~). Nlckel can be CG--~.~ll~ llle~d by C~ .I;nn~l atomic ~bsolylioll s~ Jscoy~ using diluted samples (5/1 dilution to tlL.~ce).
Solvent - FG~ ofthe adduct in the R-l process is co-~ ntly camed out in water and/or orgsnic solvent, ecpec~ y polar, most p~ r L~d~u~y _olvents.
Typical; ~ 'e9 of organic solventc useful herein in the fc~rm9tinn of the &l~ulle-~~lgar adduct include ...~,l1.AI101 (plertll d), eth9nnl, l-plUpanOl, iSO-plUp&l~Ol, the butAnc~
lel~e glycol, 1,2-propylene glycol (~lerN~,d), 1,3-propylene glycol, gly~,~ul and 2s the lilce The amine itself can also r.~ as a solvent, typically at mole ratios of A.~.;..P, ~p,.Ar offrom about 4:1 to about 30:1 The hydrcg, ~AI;n~ reaction of the R-l I~Lon can also be carried ûut in the pl~o~CP ûf an organic or aqueous solvent which dissolves the adduct HydrogPnAtinn solvents are, cGl-~iellLlr, polar, P~pe~sl1y l-~L~Ay, solvents, i e, of 30 t_e same type as those m~ntiQn~d above for use in the fiCl~9tic~n of the adduct. When S~ A11Y al~d-uu~ organic solvent is used, the umc~cLcd amine is ~-,..,o._d with the water after the adduct ~ step. However, when an P~ e solvent is used, the amine and solvent are not ~.nuv~id until the catalyst r~,~o~ step.
Water is the p.~,r~ ,d solvent for the l.~rlLo~ reaction l~th~nnl is a 3s pr~ ,d organic solvent for use in the hydrogen-s-~Qn l~lion.
General R-l Reaction Col~dilions - R~(Ctinn CQ.~ ;nl-~ for the R-l reaction are as follows. Step (a) - Adduct fo-l--alion - Step (a) of the process is _ g _ ~l~fc.ably carried out at a t~n~ e of from about 0~C to about 80~C, l,r~ bly from sbout 10~C to about 60~C, for processes l~titi~r~g organic hr~U~y solvent and below about 70~C, ~ bly less than about 50~C, more plef~d~bly less thsn about 30~C, more prtrel~bly from about 15~C to about 25~C, for ~queol~s solvents.
s The reaction time uset for adduct fc~ will typically be on the order of from a few ...~.-"~,s to about 20 hours, ~epen~;uB son~ al on the reaction le...l)e.alul~; chosen and/or the ratio of amtne to sugar. In general, for the organic solvent, lower lea~;l;oil te..lpe.~ s in the range of 0~C-80~C require longer reaction times, and vic~.~. In general, for the organic solvent, over a l)refe.r~d 10~C-60~C
1G&~;LiO11 t~ C.aLui~, range, good adduct yields, e.g., more than about 90%, prufe-~bly more than about 95%, are a~ e1 in 1-10 hours for the organic solvent.For the lower reaction le.l~y~.alul~ range, 0-70~C, pl~,f~ly 0-30~C, that gives good color, especially in water, the r~ ti~n time can also be as nwch as 10 hours, but, typically, eq~ ;hrillm is ~b~ Y reached within about four hours or less, IS especislly with higher n-~ r;~--gPr ratios. The t~ p~.~lur~ and reaction time are s~lected to give an adduct with a Gardner Color of pr~r~bly less than about 1.
Good adduct color is "~C~-~ r for OblA l~ B good l a tions and color in any ~I ~S~ dlu~f I~'liOIl and ~ ~ catalyst activity. Below a Gardner Color of about 1, the res~lting N-allyl polyl.y~ r arnine, and c.~ e~ ently the resulting amide, has good color. The color bodies can bc ~ d by, e.g., carbon bl: & -hi~ as used for the sugar SQIUtiOn The adduct also has a very low level of glucose. The glucose leveL as a p~ of the adduct is pler~ bly less than about 1%, and more prl fe.~bly less than about OIlC ~lf of one percent. fflucose i~ ,.es with the L~Lu~u~ I- &_lion 2s step to form the N-allyl polyllrJl~.Ar armine. Excess a2nine can also help reduce the glu~ose level and ~--:--:-~ e rc~ of sorbitol during h~,~ug~ I~AI;OI~
In general, the t~nlJ~aLw~ will rise during adduct fo~ ~---I;on since the l~&_Lon t-~,nic. Th~erore"~ g te.-~pc.alu-~s below about 3û~C, as l~.luh~,d in batch p~ùce~ , involves providing cooling for the ~ nte and/or the l~.;liOn rnix.
T~l~ u~s above about 50~C require l~~_Lion times of less than about 10 ~ 5 to avoid ~ce~&~e color ro-.--AIi~ Such short times are normally not feasible except in a CQ~ vO~iS l.a.~l;ol~ Even with such a co l;~ u-J~ reaction, back-rnLcing should be ".:l-: ": ~A, e.g., by use of plug flow con-litinne~ to avoid ~ , e-l~o~ of the adduct to higher t~--p~uies. Ide Llly, the adduct is p.u..l~ reacted with L~og~,n 3s to fo~m the c~ .on-~ 3 N-allyl POIYII~ UAY amine to ~ , degra~l~tion Hûwever, le.~lp~ lul~ below about 30~C, ~l~f~.~bly Iess than about 20~C, allow one to handle and/or store the adduct for at least several hours, which fA~litAt~s the use W O 96/40619 PCTrUS96/07124 of batch processes. At 0~C, the adduct is stable for 24 hours.
Surface l-~ e.aL~s~ e.g., when pr~h~ g the adduct for the hydrogen reaction process, should be .~ ed below about 100~C, pref~.dbly below about 70~C.
S pcf~ nt con~ ons can vary. Molar ratios of ~ ne s~ r not greater than about 7:1 are pr~r~ bly used herein, ~lthou~h ratios up to about 30:1 can be used when the amine is used as a solvent, at least in part. Generally the desired adduct form~tirm is achieved at a mole ratio of ~mine~ g~r vith an excess of arnine, e.g., mole ratios of >1:1, prtr~.~ly greater thsn about 1.1:1, snd the lil~e, e.g., 0 greater than about 1.3 :1. Typical ~a~ l~lL concf~ lions in the water and/or h~/dlUA~
solvent are in the 10-80%, typically 40-50% (wt.) range. Adduct forrn~tinn can be carried out at ~I---n~h- ~ ;c or s~t.~ osl)h~cpl~ s.
Step (b) Reaction with Hydrogen - Step (b) should be nrC~ plicl~e~ so as to avoid the prolo~ exposure of the adduct to the catalyst when the hydrogen 5 pre.,~ r~, is less than about 500 psig, and prl r~ly the L~dluge.~ p-.,~e should be at least about 1000, and more pref._.ably at least sbout 1500 psig. K~eping this time below about one hour, and pref;~ below about a half hour, ~ s the amount of catalyst metal, e.g., nickeL that is Coll~_.Led to water soluble ion. Such ions are unde~ ~le for a variety of ressons ir~ ing their affectt on color rc~ l;nn 20 ;~ o~ with other ~ t ~ ;~1~, ssfety, etc.
Step (b) can be carried out in either a slurry process or a fixed bed. Step (b) is pl~ abl~, carried out at a te.~ of from about 20~C to about 120~C, pl~t ably from about 50~C to about 100~C for organic hyJ~uAy solvent plOC~iC5 Step (b) is p.~vf~_bly carried out in two stages for nqueolls solver~ pluc~s~s The 2s first stage is at a ttvpc~lu r that is low enough to avoid ru~,i,.,- of the c~ -ol~ti~g reduced sugar, e.g., sorbitol in the case of glllcosP, and other ed l"~-u-lucts. Typically this is from about 20~C to about 70~C, more ,,.,f~ly from about 40~C to about 65~C, and even more preferably from about 50~C to about 60~C. ~ the second stage, a~er the rfv ~ (hydrcg~ AI;.~I~) of the adduct to the N-alkyl poly}l~ LuAy amine is at least about 80% r~ -, ' 'e prtvfv.~ly at least about 90% co~ i kle more p~lLrv~ at least about 95% ~ , the le.l-pe.alule is raised to at least about 75~C, plefe ably at least about 80~C, and up to about 135~C, p.lvfv.a~l~ 130~C, so that the le---~ adduct and any other m~qten~le that may form color bodies are .--;-~ d and the adduct is at least about 95%, pr~f,rv.~ly at least about 98%, more prtvfv.~bly at least about 99.9~/ co~v.l~d to the co .~on&~g N~ cyl amino polyol. T}li8 second ~tage is r~Q .~ to the pr~&~lion of N-aL~cyl polyhydoxy amine with good stable color upon l-P~ing Heat stability is ,ro~cd for the N-alkyla~ o polyol by using excess amine in the prep~a~ion step and a higher te.llpe.aLI~l~ at the heat l,eA~ step.
During Step (b) it is highly pl~f~.l.d to avoid loc~li7ed o~.-l-e~ e.g., at the surface of the heating flr-~ or heat ~ clle.~r. Such surface or "skinH
s telllp ~alw~s should be below about 180~C, p-~ f~bly below about 100~C, snd even more ~ f~ably less than about 70~C, during the first stage and less than about 100~C
during the second stage.
The reaction with hydrogen is prefe.ably carried out with limited initial water when the solvent is an organic h~l~u~ solvent, ~tthou~h even then, water (e.g., up to 1:1 wt. H2O-~I~ohnl) can be present. Optional water removat from the adduct pr~àl~t in Step (a) can be f ~. I~d by use of drying agents, or by simply i.L.;ppin3 water and sohrent from the adduct, and then .~ solving the adduct in fresh water-free solvent. The 1.~ reaction can typically be mn, for; 'e, at t~npc~a~w~ of 20~C-120~C at 50-1,000 psi or, for . . '- at 50~C-90~C at 100-500 psi for periods of 0.1-3S hours, generatty 0.~-8 hours, typicatly 1-3 hours when the organic solvent is used.
When the solvent co~ c water, the hydrog~ ;n~ r~ Lon is done in t vo stages as .]~- ssed before.
The adduct/solvent se' l~isn used in the L~.Log~ reaction is typically at a 10-80%, typically 40-50%, (wt.) solute level.
It will be &~cd that the s~ l;n" of l.~ ,xcll l~Lol~ cQn~litinnc will depend so~ ...hdl on the type of pl~we e~ - to the fc" .~ or, so the above-noted r~lion CO~ 5 can be varied without dep~L,.~ from this i,l~..l~ion. However, as noted before, the hy~ e.l pres;,~e pl~f~ly should be 2s above about 500, p~ubly 1000, more pref~.ably about 1500, psig when the adduct and the catalyst, e~re~lly the plefe.ret nickel catalyst, are both present. Use of lower ~,r~ down to about 100 psig will require either a s~A-~e step to remove M ion, or more prolons~ed post lr~ n~ as ~I;c,~55/~ hc.~ --An- r, to whieve verylow Nl CQnt~nt IIy~oge~ reaction catalyst levels are typically from about 1% to about 100%, preferably fronn about 2% ~referably about 5%) to about 30% ~lef~ably 20%) more p.ef~bly from about 5% (~ re-~bly 10%) to about 15% (p.~ft;r~ly about 20%) solids by weight, c~ t~od based on wt. catalyst:wt. reduç;rg sugar Step (c) F;~ g - The catalyst is then sep6l~let from the product a~er the iOl~ is completed. The catalyst is l~llu~cd from the product of Step (c) which is then pr~f~bly dried by cryst~ 7~tinn or by solvent/water i,~ ,pihLg, or by means of , W O96/40619 PCT~US96/07124 L~ dry-ing agents. This helps prevent reversion to the sugar starting m~tPri5~1 Step (c), when it involves solvent/water stripping, is preferably done in a wiped film c~a~o~tor.
Steps (a)-(c) of the R-l process are preferably con~1ctecl under non-oYi~li7i-~gs contliti~nc (e.g., H2 or inert gas) to provide good color. Catalyst removal in the Step (c) process is done plerelably under hydrogen plCSa~llc; to prevent M (catalyst)tliccQ1l1tion or at least under inert con-litionc ~lucose Addition Process Another suitable process for plepa,il,g the POlY~ OAY amine utilizes glucose 0 a(ldition (The n(~ cose .A~ itionn process) after p~ g the catalyst and amine in a simplified reaction which can achieve good results so long as the glucose is added under a hydrogen plei.~e of at least about lO0 psig, l,l.,r~ bly at least about 500 psig, and more pief~,.~ly at least about lO00 psig, at a tc.~ .atul~, of less than about 80~C, pr~r~lr less than about 70~C, most ~lef~l~ less than about 60~C. The lS m~tPri~lc and the cc~nAitiQnQ for the l~ ;"d- r of the rcacliol~ are the same as ~let~i1Pd above for the adduct process.
The pr~alaLon of the N-allylaminol polyols by either of the processes can be cQn-lucted in any well-stirred pl~.S;.~, vessel suitable for c~nr~1cting l-~
rP~it~ne In a ~ll~e.~e.l~ mode, for the "t~T1~cose ~ titionn process a ~ ,S;,ul~,20 reactor ~-vith a s~,p~ate storage ft,S_.vulliS employed. The l~e3~ 0i~ (which, itseLfi can be p~ Pd) c~ cA~cs with the reactor via suitable pipes, or the like. In use, a stirred slur~y of the nickel catalyst is first nr1e-AnPA,n ;",~1 ~d:-~ being treated with l-~dlu~en to remove traces of nickel oxides. This can be CC,~ ntly done in the reactor. (~1l-, "Al;~L~ly, if the i-~A....rA~jl..~. has access to an oxide-free source of 25 nickel catalyst, ~ 1 with H2 is ~ eCe ~ . However, for most mAm1fiA~lnng p~oc~es some trace of oxides will iç~i~ly be present, so the H2 r -1 iS p~efe-l~d.) Mer removal of excess slurry ..~r~A;.~... (water) the N-alkyl amLne is introduced into the reactor. Thc.c~lc[, the sugar is introduced from the storage l~Oi. into the reactor either under h~g~l pf ~ u~ or by means of a 30 high p.~c ~ p:-~g system, and the lea~iol~ is allowed to plocecd. The progress of the l_Lon can be rn-~-- lor~d by perioAirAlly l~llU~inp, ~ '9 of the r~lion ll~ule and analyzing for ull~ cled sugar using gas cl~u..~AIography (~g.c."), or by heating the sarnple to about lO0~C for 30-60 r - ~ r s in a sealed vial to check for co1Or stability. Typically, for a reaction of about 8 liters (ca. 2 gallons) size the initial stage (to 95% of reducibles being d~ ed) l~ Ul~S about 60 .. : ~ , de~ 8 sOlll~,..l.al on catalyst level and t~ lp~.~lure. The le.llpe.alule ofthe reaction mixture can then be raised to , ' the reaction (to 99.9% of the ~ b'~ being W O 96/40619 PCT~US96/07124 d~p'cted).
Cryst~ 7ation of Polyl,ydru~y~ f-s The color quality, ~labil;Ly, and/or purity of the N-alkylamino polyol can be further i~lp~u~d by a process of cryst~l1i7~tion of the N-alkyl~,....o polyol from an s aqueov~ sol~ltion or water/organic solvent l,~Lulc. Cryst~11i7~tion is carried out by cooling the aqueous n~ul~ of the N-alkylamino polyol from Step (b) to 0-10~C, ormore, preferably by col~ç~ Lillg the a~lucuus llli~lUI~; to about 70% solids prior to cooling, and most ~ r~.dSly by adding from about 10 to about 200 parts of an organic solven~, e.g., ~nf ll.~nol ~ceton~o., etc. either to the aqueous feed sol~ltion, or, 0 most plerc.ably, to the co~c~ aled sQl~ltion Highly pure crystals of the N-alkyl~l-illo polyol form which can be ;CQ1~ted from the S-~P~ .I sol~tion by filtration and/or CPnt~ifilg~tion To obtain the purest crystals ~o~ lc, the filter cake, or c~ ;r.-g~ cake, should be washed with from about 0.25 to about 1.25 parts of chilled (0-5~C) solvent. The wet cake can then be employed to produce polyl.y-llu~y 5 f~Ltb acid amides with ~ ~ccd color. The crys~lli7~ti~n method provides a ;,~ll~li~lgl~r illl~.ro.~ amide product.
F~----alion of Polyl-y~l-u~y Fatty Acid Amides The N-al~ o polyol co- ~ u~ Ae ~.~,p~cd by either of the above l.~.~ l;n~
and having the rc~ui-. d Gardner Color can be used in an overall process for p~ii~g 20 polyl-ydru~ fatty acid amide s~ r~ which inrl~ld~C an amide-rt~ g reaction co...~ reac~ a source of fatty acyl groups such as fatty acids, fatty acid anhydrides and fatty acid esters, especially fatty acid esters, having greater than 98%
~ at 460nm with an N-alL~ no polyol having a Gardner Color of less than 1 (<0.1 abs. at 440nm), more p ru~ bly esters which have been ~ictilled in the 25 p.es~ ce of from about 0.05% to about 2% alkali metal oxide, e.g., those pr~&~;d in the rur.,go~ manner, in an organic hydruAy solvent in the ple3~Ce of base catalyst.
The fc-tnntinn of such s~r~ ! with high purity and low color is an espe~;slly be~- r~e ~1 result of such a process when an organic ~ ~UA.Y solvent is used, since the det~,~.l fnnnlllstor can pump and/or incG.~û.ale the poly~-~dru~ fatty acid amide 30 ler~tinn product plus the ~~_Lion solvent such as 1,2-1,r~p&l~e diol, (propylene glycol), glycerol, or alcohol (e.g., in liquid d~le.~ ) directly into the final de~e~genl forrmll~tinn This offers eco~ c advantages in that a inal solvent ~ nuval step is rendered ~ c~ ,c .y, particularly where u~hol-s glycols or ethanol are used.
~ The polyl.~o~u-~-e products of either of the aforesaid R-l rea~ionc~
3s p.~,ft; ~bly with water ;,-~l,st~n;~lly re.no~cd, can be ~rther employed in an amide-rO~ g reaction which is ~e~ ted herein as the HR-2" reaction. A typical R-2 amide-fo. ~ ~ :tinn herein can be illu~llaLCd as follows:

R2COOMe + R3N(H)cH2(cBHaoseH)c~atca~y2sto~ eH,g.~ ;d~

R2C(o)N(R3)CH2(CHoH~4CH20H + MeOH
S wLer_.n each R2 is Clo-C20 allyl and each R3 is Cl-C4 allyl, Cl-C4 alkoxyallcyl, or h~dr~ lkyl group.
Thus, the process herein can ~nco~ s an overall process for p~ ~u~g polyl-~dlo~ fatty acid amide S~r~ option~lly c4"~r~;C~g an R-1 process as dc 5~ d above and then r~_lu,~ the polyl.r.lluA~une having a color of less than 0 Gardner 1 vith a fatty acid ester having at least 98% ll~c-~ ce at 460Dm in an organic Lydlu~.y solvent (pl~,fe.~bly, .~.CII.An~l) in the pl~,sc~ce of a base catalyst to fo~n the polyl,yd.oAy fatty acid amide - ~- r~ l~..l (at a ten,~ u~ of from about 40~C to about 135~C for a time of less than about three hours, more pl~f~,.~ly at a c.,l?c,.~Lule of from about 40~C to about 100~C, and even more plcfe.~l~ at a tc.~lp~,.aLule offrom about 50~C to about 80~C for a time of less than about 2 hours,);
and optionally, l~OVUU2 said solvent. The rPclllting amide pi~du.l is treated with ion ~ ch~ ,e resin, more plefe.~bl)r a ~lule of acid and base resins, or, optionally, with ~J~ g bleach to provide a pro.lu~l that is r ~ lly "water white~.
In a more l"~.f~-ed ernhodi n~nt, the amidc ~ rt~nt is treated first with acid ion e ~ resin to convert any soap to fatty acid and remove any residual amine that has not been uj"~s,.led to amide. Then thc amide ~ is treated with base ion e-~ g~ resin to remove the fatty acid. Both resin_ remove part of any color bodies that have already fnrrn~d R-2, or the co...l~ inn of R-l and R-2 leP.~ 8 herein, can be used to 25 prep~ polyllyd~uA~ fatty acid amide ,,...1;.~ ofthe 1~ (Il) as follows:
R2 C(O) - N~ ) _ Z
~1,~,.~.: each R1 is ~ Cl-C4 L~d-oc&~ L cl-c4 alkoxyalkyl, or h~uA~,Lkyl, e.g., 2-Ly~lluA~ l, 2-}~LuA~ o~yl, etc., p-~ ..ably Cl-C4 alkyl, more preferably Cl orC2 allcyl, most pr~ Cl alkyl (i.e., methyl) or ~ LhOAY~ICYI; and R2 i_ a Cs-C3 130 L~oc~l moiety, p..,~.dljl~ straight chain C7-Clg atkyl or alkenyl, more pr~,fe.~l~ straight chsin Cg-C17 aLIcyl or alkenyl, most p efe.~bly straight chsin Cl 1-C17 aLkyl or alkenyl, or ~1 Alu~ thereof, and Z is a polyl,~Lu.~ d uc& l,yl moiety having a linear l",Lu~ chain with at lesst 3 l.~LuA~L directly co~ d to the chsin, or an allwA~ d~i~raLi~e (p~ bly ~lhUA~/htCd or pr~po~la~ed) thereof.
35 Z l,.ef~ will be derived from a 5~d~1C;'~g sugsr in a reductive pmi-~tion re&Lioll;
more preferably Z is a glycityl moiety. Z p-~f~l~ will be ~-'xted from the group crl~ g of -CH2~CHOH)n-CH20H, -C~CH20H) (CHOH)n-CH20H, CH2--(CHOH)2(CHO~)(CHOH)-CH20X where n is an integer from 3 to 5, .i.cl~ , and R' is H or a cyclic mono- or poly- 5~rçh~n~t~, and alkc,~l&l~l d~ aLi~3 thereof.Most pr~,~wl~d are glycityls wh~ n is 4, particularly -CH2-(CHOH~4-CH20.
Mi~lwe., ofthe above Z moiefiçs are de~
s In Formula (II), Rl can be, for F~ r~ N-methyl, N-ethyl, N-propyl, N-isop.u~L N-butyl, N-isobutyl, N_2 1~Yd-UAYeLI~YI~ N-1-n1~LLUAY1I1U~ or N-2-h,~.ll OAY~I U~
R2-CO-N< can be, for ~ , co.- do~ ~I~,f.de, oleamide, lauramide, dl" carn~sn~ t-~ .;de, tallowamide, etc.
Z can be l-de~A~lucityl, 2-deuAy~u~,~i yl, l-deo~y.l~ltityl, l-deu~la_Lilyl, 1-deoxygalactityl, l-deu~yll~m,Lyl, l-deu~ to~riotityl, etc.
The following ~ , catalysts and solvents can coll~ueilLly be used in the R-2 reaction herein, and are listed only by way of e~mpl;fi~ ~tion and not by way of lim;tPtirn Such m~t~ are all well known and are luul;nely available from a lS variety of co~ c- ~,ial sources.
~e~.c~ ls - Various fatty esters can be used in the R-2 r~ inrl~ g mono-, di- and t~i-esters (i.e., ll;~c~,.;des). Methyl esters, ethyl esters, and the l~ce are all quite s~ '~ The pOlyllyd~uA~ l~lllc Ic~ include re~ct~ntc available ~om the above-dr~ ;l,cd R-1 rea~;lion, such as N-alkyl and N-L~druAy~llyl pol~ .lluAyal,.in~s with the N ~ b~l;luent group such as CH3-, C2Hs-, C3H7-, HOCH2CH2-, snd the li1ce. ~POIYIIYdIOA~ UI~eS available from the R-l reaction are plefc~ not GQ"IA~ Ied by the pl~3e.nce of residual z---u-~--l, of metallo hydrog~n~tion catalysts, ~lthm~gh a few parts per million te.g., 10-20 ppm] can be present.) Mi~ of the ester and I~uAlurt s of the polyl,~ d~ ~lul~e re~ csn 2s also be used.
Ca~ s - The cstslysts used in the R-2 le~-tinn are basic m~t~ tc such as the ~lL...;f~ d), hydro ndes (less prefe.l~d due to poCQ;~1e hydrolysis re~innc)~
C~~ t~'~, and the like. ~f~lt;d ~ Yi~le catalysts include the alkali metal Cl-C4~llrnY~1es such as sodium m~hsYirle~ pol~ u~;~c, and the l;ke. The catalysts 30 can be ~ d sepd~lel~ from the leclion nu~ , or can be gell~.al~d in situ using an alkali metal such as so~illm For in situ ge.l~.~Lon, e.g., sodium metal in the nl solvent, it is p-efe.lcd that the other 1~ ntc not be present until catalyst g~~ lionis complete. The catalyst typically is used at a level of about 5-8 mole %
- ofthe ester react~nt Mi,-lul~s of catalysts can also be used.
3s Solvents -The organic }-,~dlUA,~r solvents used in the R-2 reaction in~ .o, for exa~ple""f ~l~Anol ~thAnnl ~lop~lol, iso-~,lo~allol, the I~IA~nl~ glycerol, 1,2-propylene glycol, 1,3-propylene glycol, and the l~ce. luPll~Al~nl iS 8 pl~f~ f~ d alcohol solvent and 1,2-propylene glycol is a ~ r d diol solvent. Mixtures of solvents can also be used.
General R-2 Reaction Conditions - It is also an optional objective herein to prepare the desired pro.lucLs while .~ the formAtinn of ~;y~ d by-products, ~ ~ ~ OH

s e.g., OH OH
ester amides and color bodies. ~acti~n tc.~ Ul.,S bdow about 135~C, typically in the range offrom about 40~C to about 100~C, pfCf~..~;l~50~C to 80~C, are used to achieve this objective, esret~;Ally in batch p,ucesscs where reaction times are typically on the order of about 0.5-2 hours, or even up to 6 hours. SG11~_~l1al higher 0 telll~Lulf s can be tolerated in cG~ vv!~ procf sse~, where reQ;~lf~n~e times can be shorter.
Plllificn~iol~ ofthe POIYl-Yd.UAY Fattv Acid Amide The POIYI1~d~A~r fatty acid amide ~~fiA~Ant p~epa~ d by the p~uce~s herein is very pure and has good color. However, for products that are not colored, or which 5 are clear, even purer, less colored S~ tAntS are "~ L~. Acco,d,n~, the PO1Y1~Y~OA~ fatty acid amide S~rfActAnt iS p,ef~.~bly post treated with an ion AI~gf. resin, ~lliAIUl~ of ion ~ -nllAI~gf resins, or c~ ;on~ of ion f .~
resins, and/or ~ ~luc: ~5 bleach such as NaBH4, etc., or hydrog~nAti~ n over a catalyst.
Tlf nl--.-~nl with ion e-~ ,f resins can be very e~t~ ., if the ~ II. n~ iS
20 c~efully carried out. Since the minor CO--~ present are both cationic innature, e.g., amines, and/or anionic in nature, e.g., soaps and/or fatty acids, it is desirable to treat with both anionic and rAAtit~nic (~cidic and basic) ion e-~ ge resins. A p_lL-,uLu~ is to treat a s~ ti~-n ofthe polyl,yLuAy fattyacid amide ~fActAnt first with acid ion P~ ge resin to remove the amine and 2s convert any fatty acid soap to fatty acid and then treat with base ion ~ ,e resin to remove the fatty acid.
Another particularly ~cli~-, post h l is the hydrogf~nAtion of a sol~tinn of the polyl,yd,oAy fatty acid amide s-~ II over a hydrog~n-otinn catalyst like nickel, pollo~ n~, copper cl~o" ile, etc. Su~ gly, the l~yLog~-~AI;o~ is c ~c~ in 30 el;~ n~ g color bodies and color body p~ ,U~ will~ou~ ad~ ely A l;~ g the structure ofthe sl-rfoctont The hydro~en-o-tinn is typically carried out in a batch reactor. A catalyst, typic. lly of either nickel or pollo~ m is sluITied in a s -l-ltinn of the polyl,yLuAy fatty W O 96/40619 PCT~US96/07124 .

acid amide s~ ct~nt and reacted under c~ n~itinns that will achieve the desired i,n~ cm~;llL. Typical reaction con~litiQn~ are h~Lu~e.l ~r~l~; of from about 150to about 1000, pr~lably from about 300 to about 500, psi; tC..~ alu~t; of from about 50 to about 120, p~f~ably from about 50 to about 65, ~C, to limit pole.lhal s soap r~ iOn; and reaction time of from about one to about four, p.cr~.ably from about one 1:o about two hours.
The color ofthe ~ r~ -.l is ~nea;~ as % ~ A.~ on at 420 n~nnl..tt~
against a 50/50 ll~ We, by weight, of .--~ ll.A~ tilled water blank. The ~ t~nt is diluted to ~0% by weight with the blank solllt;nn and read in a ;.~echvphoto,n~ te~ .
0 Typical color of cn---~ ,;al prodllction varies from about 55% to about 70%
h~ n, as .-lea~d above. For clear products, the .-- n;.. , L~ cc;~
should be at least about 70%.
The catalyst loading to achieve 70% h~ epen~ on the type of catalyst used, and the desired level of color .,ll~rovG,lle.~l. For nickel catalysts, the 15 loading ranges from about 2% to about 10%"~r~rel~ from about 2% to about 5%, c,~lesse~ as weight of catalyst based upon the n~ t~nt in sol~ltinn These levels of catalyst will raise the hl-n~ n from about 40%-48% to about 70% with 2%
catalyst and to about 80-85% with 10% catalyst. Post L~LO~t-~ I;nn with p~ m catalyst produces ~lrPrjt~r color with less catalyst. P~1lAt~ m catalyst usage ranges 20 from about 0.005% to about 0.15% with r~ 8 h~n~ of from about 85%
to about 90% when starting with colors having IIA-~ nC of about 60%. For c,q"~,p~. ;er~l~, a ~ nn of sbout 42% was raised to about 75% by nickel catalystand to about 93% by p~ ilm catalyst, using con~lil;n~c of about 120~C and about 360 psi hydrogen.
Another optinn~l reductive bl~c'i~ step utilizes a reduring m~tPri~l such as;
NaBH4, LiAI~, etc. It has been found that the pH should be from about 10 to about 10.9, pl~f~ably from about 10.1 to about 10.6, more pl~f~l~ about 10.4. This pH
range ~,lo~,ides ~ycellpnt bl~ ~hi~ at a good rate wilLuu~ ~ces~, creation of fatty acid soap by hydrolysis of the amide.
The following . ~ S are intPnrled to i~ e the practice of the R-2 reac~on using the N-polyllyd~oAy~l~illes pl~an,d by the above-~iiQr~osed R-l - reaction ( vith H20 having been ~.IIU~el;i). It is des.l ble to use conc~ ~ion ranges of the ll l,t~ntc and solvent to plc~v;de a "70% c~nG~ ~ed" (with respect to - I P ~tAntQ) ~ iO~ W~. This 70% cnn~ dt~ u~lw~ provides eYrP1l~nt 35 results, in that high yields of the desired polyl~ ln~r fatty acid amide prv~luc~ are secured rapidly. Indeed, jnrl;C~tinn~ sre that the r~ihun is ~lb t~nl;~lly complete within one hour, or le_s. The cQ~ nc~ of the r~lion u~lu.e at the 70~/0 c~ aLion level provides ease of h~Antlli~ However, even better results are secured at the 80% and 90% con~ aLion levels, in that ch,o,,.ulo~aphy data Ale that even less of the unde~red ~cli~ed by-products are formed at these higher COI~C~ ~1 ~ alions. At the higher co~c~ aLiOlls the reaction systems are s so---~ more ~liffic~llt to work with~ and require more effi~ent stirring (due to their initial th~ ne~), and the like, at least in the early stages of the reaction. Once the reaction pr~ceeds to any ap~-~ iab'c extent, the viscos.ly of the reaction system decleases and ease of mixing i,-~ ases.
All pC,.Cf'~ 3f ~, ratios and propo.lions herein are by weight, unless oLllc. ~;se 0 ~pec-r:ed All limits and ~ ';CAI values herein are a~pro~ AIe unless olh~, w;~ stated.
FX~MPLE I
Standard Reaction A reaction ."-b~lu~e con~i~ti~g of about 214 g C12 fatty acid methyl ester lS (Procter & Gamble methyl ester CE1295); about 195 g N-methyl-D-~ mine, dry powder; about 10.8 g 25% sodium l-.l}-ylale, and about 37.7 g propylene glycol as a solvent is used. The reaction ~ressel co~ a one liter, four neck, round-bottom flask .~_lo., one 300 mm coiled Co~ - -, one 250 ml round-bottom flask; several ada~tu-~, one agitator with a v ~' le speed motor; one mantle c~ rcle~l to a Therm-20 O-Watchg) for le ll~. alulC control; and a ~_cuulll water &s~llalor for va~iuulll.
The methyl ester is added to the reactor and, with ~~ , is heated to about 60~C. The propylene glycol and the N-methyl g31l~Amin~ (puwdc,.~,d), are added with ~lffi~i~P,nt ~ n to keep the solids ~spPn~le~l The twl~c~alu e is raised to about 80~C and a v Illlm of about 100 mmHg abs. is created, if more than about 0.1%
25 ..~;;,~ is present, to e~ A-IG the moi;.lu-e. The p,~ is raised with n.liu~enand the sodium .ll~ lylalc is added. The le,ll~,.alulC, is set at about 80~C and the time is set at zero. The pre~e is reduced app..,~ ly every thirty ~ s from, ~PP~ 500 to 350 to 200 to 100 mmHg. The pr~ .e is again raised with l~l ùgwl and a sample is taken for GC analysis.
The above i.landald reaction results in about 20~600 ppm cyclic rnAteriAAI
which is con~ Pred undesirable. In one standard reacfion, the level of cyclic is 250 ppm while the percent col-~e.:i,on is 91%; lc w~ t,he reaction twlllJ~.alul~ to about 70~C lowers the cyclic level to about 80 ppm and the CG~ On to about 88%;
lu~._.h~g the reaction time to about one hour dec.eases the cyclic mAtPri~le to about 3s 50 ppm and the co~ on to about 89%; cutting the catalyst level in half reduces the cyclic ~tPrj~l~ to about 90 ppm and raises the CGll~w~Ol to about 93%; rw..ovlll~
the .~ nol in 30 ~ les l-,.luces the cyclic m~tPri~le to less than about 50 ppm and - 19- , raises the co,l~e.~on to about 90%; and red~ g the ~uum to a ~ of about 200 mn~Ig reduces the cyclic m~teri~le to about 40 ppm while r~io~ng the COI.~oll to about 87%.
l~oducing the time to remove the ~ l and redl~;r~g the vacuum have the s most c~ impact onred~l~ing cyclic formation.
Color ull~Jr~.,.,wnl is obesined by using rÇ~ct~ntc with better color. The methyl ester and POIY1IYdrOA~r amine should both have a Gartner color of less than about 1, the amine being the most ~ o~ . Using an excess of amine in the R-l n, e.g., about 100% excess and/or higher heat ~ 'P-'I te~ e.a~-,s provide 0 illlplu.~.d amine color. Use of a cryst~lli7~tinn step ~ es the color even more.
The amide is p~.fw~bly treated with an ion ~ g4 resin, or, more p~fe.a~ly, with both anionic and c~ti~n;C ion ~ ~ol.~ resins, to rernove color b~odies. This l is P~ ~,pli~hF d as follows.
F~A~IplF ~
lS An overall process at the 80% Ic~l2ult co-~c~ aLion level for the amide sy"~ s~s is as follows.
A reaction ..~lu.e co~ of about 84.87 g C12 f~tty acid methyl ester ~Procter ~ Gamble methyl ester CE1270), about 7~ g N-methyl polyl,yd,o~~ c per Example L above, about 1.04 g sodium ~--F ~ and a total of about 39.96 g20 methyl alcohol (ca. 20% by wt. of .~clio~ UrC) iS used. The ,~lion vessel cc....l" ;c~ s a ;,~ld~.l reflux set-up fitted with a drying tube, col-dF-~e ~ and ...Fcl-~nir~l stirring blade. The N-methylglllc~ r/~ nnl h heated with stirring under l~,l,ùgcn (reflux). After the sol-~tion has lc~cl-ed the desired ~l-yc~alurt~ the ester and sodium meth~Yi~e catalyst are added. The reaction ~.~lure is .~ d at 25 reflux for about 6 hours. The l~aclion is Ç~5f~ lly t: F'-l in about 1.5 hours.
After removal of the ,-~ ol, the l-,cu._.ed pro-lu.;l weighs about 105.57 grams.CL.)~ography inrlirinte~s the pr~_~ellce of only traces of u"de~d ester-amide by-prcducts, and no de ~ b~ cl;~edl by-pr~,dL~
While the ru-~u,--g d:Crlo..~,~c generally relates to a solvent-~ciQt~ method 30 for plCp~L N-methyl pOl~ Lor.r amines, such 8S N-methylgl~ -:--r, ss well as their fatty scid amide d~;~aLi~es using fatty methyl esten, it is to be understood that ~ v~ ti~nc are available which do not depart from the spint snd scope of this .lLion. Thus, I~ g sugars such as ~uctose, g~1~cto5~ nl~G5~, msltos~ and tose, as well as sugar sources such as high d~hu~ corn syn~p, high liu~;lûS:e corn 3s syrup and high m~ltose corn syrup, snd the lice, can be used to prepare the polyl,~dro~,l-ne tn~ten~l (i.e., to replace ~ A~ ) ofthe reJ~
S~ ;ly, a wide variety of f~ts and oils (trigl~ ,.ides) can be used herein in place of the fatty esters ~mplified above and can provide an unobvious ro~ in the degree of compl~t~n~cs For ~ ?I~ fats and oils such as soyl~l oil, eullo~cp~ed oil, sunflower oil, tallow, lard, safflower oil, corn oil, canola oil, peanut oil, fish oil, .apcsced oil, and the like, or hardened (I.~droy~ ed) forms s thereof, can be used as the source of l~i~c~.ide esters for use in the present process.
When the hi~ ly.~ ides are used, the reaction p-~,ceeds to be closer to ce . ' ~inn and there are less by-products to be le...o.~d. SperifirA-lly~ greater than about 95%
c~ fi~n is possible. F~re led triglycerides are palm kernel oil, coconl~f oil, palm oil, and ta11ow.
lo Purification The ~CllrfAC~f~ntc produced by the processes Jic~l~ sed above are ;,.u~.;si..~
pure. However, for ~,.~&lalion of very clear products, even greater purity is le.luil~.d. The~efole, it has been found l~eC~ to treat he ~rfAAr~f~nt product herein by at least one of h~ ; sPlo~ed from the group col.~ of reductive b'e~.ehin~ and ion ~ h ~A I I ~
l~e~ucl;~e ble~chi~ is well known as a method of re~ ring/~ g color bodies and/or color body pre~iul~ that are co~ .led to color bodies later by action of light, oxygen, hlle~a.,lion with other m~t-ori~l~, etc. However, in order to treat the N-alkyl polyl~Lu~.~r amine amide ~rf~r,fAnt herein, it is ~ce~ to tske p~CC~ inn~
20 to avoid soap fC,~ AIi~ as fticrl~ scd ~.c.n&ll~
The use of l-~ûg~.l and hydrog, ~,~I;n~ cstslyst csn also provide good reductive b'-- ' in~ wilhout ~ DD;~_ soap rO....~I;O~- ~lthnu~h this techni~lue usually is more comrlic~ted and re~lu,r~s special e~ rc..cd hydro~;f n~tion cstslysts sre those ~e~ . ;t-ed h~.c ~.t-. fv~ c.
2s It will be ~p.ec;ated thst the tn~mlf~lre of dct~ lrf~rt~nt~ from such l~...._b'el~_au-~eDis 8ni~po~L~Ladva~ ~e of the present process. The present proCess is p&~ r useful when pl~i~ the longer-chsin (e.g., Clg) and 7III~ ~ed fatty acid PO1YI~ AY amides, since the l~-lah~ly mild reaction tf...~.~ureD and contl;tinnc herein afford the desired products with ..~ l by-product form~tisn A pre-formed pofion of the polyl-~dlu,.~ fstty acid smide Yll f~ -.l csn be used to sssist initi~tinn of the R-2 amide-fo. .--:~g reaction when L~.,~;des or the longer-chain methyl esters are used as f~ ~.C~
Fx~MPT F m p"~;~r_l;o~ of N-methyl g11~c~minf pn~ceeds as follows.
3s A~P1'O~ 2500 g of q~ oll~ sol~ltinn cQ~ g about 45% by weight c~--- ~f!~w~ grade N-methyl ~ ris cl~g~d to a rotary c.~o.~or where it is heated to about 71~C under about 27.5~ of Hg v m until about 957 g Of W O 96/40619 PCT~US96/07124 con~ q~e is c~ cted colr._spo~ldmg to a solids CO~ aLiO-~ in the e~yolalor residue of about 75%. The residue is mixed with about 660 g of al~h~dlous ...~ Oand cooled rapidly to about 1-2~C using an ice bath ~Le.cupoll N-methyl glllc-Am;r.
cryst~lli7Ps yielding a white slurry. An app~ .rt~l~ 1100 g portion of slury is s cl~ed to a Waring blender where it is n~ixed for about 3-4 ~ cs before being filtered using a R~rhner funnel. The sample was filtered to dryness before beingwashed twice with about 165 g ~liq~loP of chilled (about 5~C) .~ OI and once with about 330 g of chilled ~--~ nl The final cake yields about 438 g of pu~iffed N-methyl gh~cA~ at about 16% volatiles for a yield of about 83% of the solids in the io original feed.
The following table illu~ .tcs the color and heat stability .."~,o~e."c,~
g~,n~.aLed by this procedure. The pl-nfiPd crystals are dissolved in ~ t;lle~ water to yield the same cQ..cr~ aLon of solids as the original feed. Color is ~ d on the . I~AS as percent hA~ IAnce using a Milton Roy Spe~ ol).c 21D sl,e~,t,u~ t- r inan about 21 cm cell at about 420 nm. The ~ " 'es are also tested for heat stability by s bje~tin~ the mAteriA1 to about 180~C in an oil bath under an inert ~ os~; ., for about 1 hour. The treated samples are re-diluted to about 50% con~ alion to mske up for any water lost during heat l~ n~ and the .ubse~ nt colors are read.
S~qn~le Original Feed Purified C~ystals Initi~ color 71.9% T 94.8% T
Color after heat ~ 18.8% T 89.0% T
Example IV-A
2s (Amide P'~)&aLiOn with Non-Cry~tAl1;7Pd Amine) n r~ s~ tion (about 332.62 g) of G~P ~ de N-methyl ~A ~ CC~ n~ about 54% by weight solids is cl~c;l to a ~ d&~d one liter flaslc set up cc~ ~ a . . ~ A ~ ing blade, cQI~1t ~ ~ and I CC~;~ ~.
Over the course of about one hour and twenty ~ , the soll~tinn is graduslly heated to about 132~C snd the pl~ss~llc iS reduced to about 66 cm Hg ~r Clll~m to remove the water which is cc~n~1P;l~ed and co1l~cted in the recei~
To the dried N-methyl gl-~cAmin-P~ is added about 201.71 g of Procter &
Gamble CE-1295 methyl ester and about 37.20 g propylene glycol. After stirring, about 15.01 g sodium ~ u~:de sol~ltinn (about 25% by wgt. in ~ nl) snd about 3s 14 g .~ nl are added to the reactor and the time is l.,co--led. The mi~cture is allowed to cool to about 85~C as I~ O1 is ~ lPd offunder al~ s~ ;c plcs~e.
After about 30 ~ q no more ~ ol is visibly A~ ;n,~2 SO vacuum is slowly applied to the reaction vessel to strip out the ~ n~ ",f I~ OI and drive the reaction to comrl~ion When the ~acuunl reaches about 66 mm Hg without fio~min~ the reaction is comp~ er breaking the V~;UUlll vith n il,ogen, about 126.86 g water and about 74.60 g ethanol are added to the ~ u,~. The res ~lting glucose amide so' -tinn is dark yellow in color and ~1~eE~SU1eS about 54.9%
5 Tl,~ ee at about 420 nm.
FY~ , 'e IV-B
(Amide Cryst~ 7~tion with CrysPlli7pd Amine) A reaction l-~Lule CO~ c~ g of about 121.0 g of purified N-methyl g~ min~
filter cake from FY~mrle m (about 16% volatiles), about 112.1 g of Procter &
0 Gamble C~1295 methyl ester and about 19.7 g propylene glycol are ch& cd to a one liter r~_Lion vessel e~luil ~,ed with ,-,P~ l stirring blade, cc~len~- and leCei~
The .,~lu-, is heated with stir ing to about 80~C and held under a slight ~s ~lllm for about 30 }--;-~ cs to remove any residual l~oi~lul~ and .~ --nl from the filter cake.
After brealcing ~ uu", with l~ill'u~Cn~ about 8.4 g of about 25% sodium lS - Il.n~;~e sol~tinn is chal~.,d to the reactor and the time lecGlded. M~ l isallowed to distill off and is collected in the lCcG,~w. After about one hour, v llm is slowly applied to strip out the ~ ---,g ~-- ~ ol and drive the reaction to ~,c ~'otion After about two and one half hours ts~rget Va~,UU~ iS ;~ ,d and no more ~ sllnl is rlictilling Vacuum is broken with nitrogen and about 65.1 g ~ictill~od water and about 39.5 g ethanol are added to the mi~cture. The resultant glucose amide so1 ~tinn has a very pale yellow tint and ".ez~ed about 88.9%
Tl;...~.~-;ll~--ce at about 420 nm.
FXA~LE V
The rc~en~,.nlioll of strong base anionic ion ~ e resin after exhaustion by 25 polyl.~JruAr amide elution is conflucted as follows.
F.th~nnli~ HCl Sol.~tion iS p~ d by adding about 27.4 g of cQ~F-.~-nlct (about 36.5 wgt%) HCl to about 972.6 g 3A eth~r~nl A tilute caustic s~ tinn is plepalct by dissolving about 15.3 g of NaOH
pellets (assay = about 98%) into about 1484.7 g of .li~illed water.
About four l,un~ct-fifty ml. of c-1- ~-,~ A.. b. .l:le IRA-410 resin is packed in a 500 ml ~ntlualed d'~F-'~ cylinder and is washed with about one liter warm ~lictill~d water to remove residual amide. The resin is washed with about one liter of about 5% Fth~noli~ HCl so1 ltinn (prc~)arcd as dP~ ed above) to acidify, followed by washing with about one liter of ethanol to ~s , '~ ~ the removal offiltty acid. The 3~ resin is then washed with about one liter of warm ~ictillF~d water to l~hydlnte the res n.
The resin is then l~enF-atcd by slowly eluting about 11/2 liters of about 5%

~ql-eous NaOH sol~ltion through the resin. T}se res?n is then washed wsth ~ictilled water until the pH is about 8.
The ~en~.~lion of strong wid e~tionic ion ~ resin after ~ inn by polyl-~Lo~ amsde elution proceeds as follows:
F.th~nn~;~; HCl sol .I;nn is p~s~d by adding about 27.4 g of c~
(about 36.5 wgt%) HCl to about 972.6 g 3A ~th~nnl About four l~ul-~r~ d-fifty ml. of ~ e~ 120 Plus strong acid c~tinnic resin is packed in a 500 ml g,ndu~ed d;e~e~-c;-~g cylirsder, ~ d irs about a 50~C heating tape and is washed with about one siter warm .lictill~od water to remove 0 t,-,e s~s?~hl~l amide product. The resir~ is a~ ed by elut?ng about one "ter of eth~nnlie HCl and is then washed with warm dictill~od water to l~hydl~e the resin.
Regcne.~l;on is c ~ d by slowly eluting an ~dd;ti-~n~l liter of about 5%
~queous HCl lhluugll the resin. The resin is then washed with ~icti1led water until the pH is a~pr~;...A~Cl~ 5.
lS FX ~PLE ~1 About two l.~lJ~ t mls of l~5CIl~ .alcd ~ c~ 120 Plus from Example VII is packed in about a 250 ml ~~ cd c~l;,ldcl wl~ped with a heating tspe set at about 50~C. About two l~ 1 grams of glucose amide which is pl~,d from cryctA1li7ed N-methyl g~l~c-Amine in P~ ~ ~ dallcc with Example IV-B is eluted II~IUU~
20 the resin and is coll~cted in about 200 g ~ otQ
About 1800 g of eluate from the c~tinn;C column is then eluted lhluugll about 200 mls of r~e.~ G IR~10 strong anionic resin from Example VII.
This column's ~e~ .alun, is also I~ d at about 50~C with the aid of an electric heat tspe. The eluate is cQllPcted in sixteen, about 100 g A-liq~lote 2s Before resin l-~ , analysis of the glucose amide in~ir~t~e the following al~pl~ quality and cQ~.pos ~inn Tln~ rc at about 360 nm = 74.1%
N-methyl g~ Amin~ = 2.8%
Fatty A~id/Methyl Ester = 4 9%
30 C;l~lcose amide = 55.6%
Ester Amide ~ 0.2%
~ After resin LICA~ , both the color quality and r~n~pos;tinn of the product are greatly Ullpl~
3s Tln~ A~ce at 360 nm = 93.3%
N-methyl ~ J "';"f 5 0.1%
Fatty Acid/Methyl Ester = 0.6%
~ cose amide = 55.5%
Ester Amide = 0.1%

W O 96/40619 PCT~US96/07124 FXAMpLE VII
A second method for r~cll~.alion of strong base anionic ion ~ e resin after ~ by polyl~ydluAy amide elution is conAIlcted as follows.
FthAAn-lic HCl sQ1~ltion is ple~ d by adding about 27.4 of cQI~c~ lcd s (about 36.5 wgt ~/O) HCl to about 972.6 g 3A eth~An~l A dilute sol~ltion of about 7 mole ethoxylated lauryl alcohol is p.~p~ed by dissolving about 9 g of ell,o,-~lale in about 9 g of ethanol and about 1482 g of warm, Aictil1ed water.
About four hund,cd-f~y ml of ~-l-n~,ted resin, is packed in a 500 ml 0 grad~lated d;~r~ g cylinder, w~?ped in a heating tape and held at about 50~C.The resin is washed with about one liter of warm AiCtill~.d water to remove the residual amide. About one liter of warm, about 5% aqueous HCl is eluted through the resin to acidify. The column is allowed to set for about two hours at about 50~C
to allow the fatty acid to migrate to the surface of the resin. The column is back wa~hed with about 11/2 liters of warm ethoxylate SQl~tion to remove the fatty acid from the col~lmn The resin is then .~,g~ lcd by slowing eluting about 11/2 liters of about 5%
aq.,~us NaOH sollltion through the resin. The resin is then washed with distilled water until the pH is about 8.
The c~Ati~nic resin is l~e.~.,-alcd in the same manner as desc,il,ed in Example VII.
When glucose amide pl~cd in the manner A~-il.ed in F~~~~ e IV-A, having an amber color and ~ g about 32.1% T~ Ce at about 360 nm, is passed ILUugl~ these ion ~ ge resins, the color hl~JlU~_5 to a pale straw color 25 .~.rA~ about 82.2% T.a--~ ce at about 360 nm.
EXAMPLE VIII
N-methyl glllcAminP, with good color stability and which ~-l,s~q~-Pntly produces good qu. lity glucose amide is ple~d in the following manner.
APP14~ r a two gallon autoclave is chu~cd with about 360g of Grace 30 4200 Raney nickel catalysts as a 50% ;,.~ Q-OI~ in water, about 920g of 50% methyl amine and about 1000g water. The reactor is pr~ e;i to about 1500 psig with 1,~ Log~. The reactor co~ are heated to about 50~C while stir~ing. To this is cl~ecl about 2600g of ClearSweat~ 99DE corn syrup and the co-~ arc reacted at about 50~C for about two hours. Fresh hydrogen is added to II~ I the p,~ ,e 3s as it is C~ '''F~ by the reaction. A sample is rl ."u~. d from the reactor at the end of about two hours and its co,~.PGS ~ n~e~td to be appr~ - ;--IAIe1~.

W O 96/40619 PCT~US96/07124 N-methy~ rAmin~ = 95 o%
n-~uco~yl~ne = 1.0%
glucose = 1.0%
sorbitol = o 9%
. s This mAtçriA-1 was Ught yellow in color and upon s~bse~ ent reaction to glucose amide in accoldance with the procedure desrribed in FYAmple IV-A results in a product that is very dark in color.
The reaction ~Lule lc~ in the autoclave is now ~.lI,;e~,le~ to a 0t~llpc-alule ill.;l~,ase from about 50~C to about 100~C over the course of about 60 r..;~ while hydrogen p~,s,~e was "-A~ ed at about 1500 psig. After about 100~C is reached, the reactor is quicldy cooled under LydloO_.I pr~,~ulc by intro~uring cooling water to the reactor coil. When the ~-~lurt has cooled to about 30-50~C, the mAt~ is L~cl.alged from the reactor. Its co...l os ~;r~ is lS ap~lu~ y.
N-methyl g1llrA-min~ = 97 3%
n-gluco~l~ c non-clEtectn~'e glucose non-~:e - ~'F
sorbitol = 0.8%
This water white mAt~riAI is used to produce glucose amide in acco-dsl~cc with the p.oce.lu-l_ used in FYAmple IV-A and results in a product that is pale yellow in color.
FXAMPL}~ lX
AMlDES MADE ~OM CRYST~T ~ ~7Fn NMG AND BASE TREAl~ ESI~R
About 49.1 kg of Procter & Gamble C~1295 methyl ester is cl~S~ to a 72 liter t1;Atil1Atinn flask e~luipped with a con~ r and .~ c~._. . About 900 g of sodium 1--J~ scl~ tir~n (about 25% by weight in .-- ~ 'O1) iS added to the ester. At an30 absolute p~l'G' of about less than 10 mm of Hg., the ester is heated to about140~C. The ~ tillAte iS CQn'~ and coll~cted in the ~ r. The first about 618 g coll~ted in the ,.,cei~,r is discarded; the le-n~ g tlicti1lAte is coll~cted as a 'water white', low odor methyl laurate.
About 175.0 g of n-methyl gl~ e cry~tals purified in &conl~ce with 3s Example m are dissolved in water to produce about 375.0 g of n~eol~c sohl~ionThis sDl-~ti~n is d~ed to a ~lal~d&ld one liter reaction flask set up co~.~r.;~ a A1 stirnng blade, co~d~n~r" and recG;~,r. Over the course of about two hours and forty ,~ , the sol~ltion is gradually heated to about 130~C and the pl~e is reduced to about 26 inches of Hg v~uw~ to remove the water which is condenced and coll~cted in the .ecei~.er.
To the de~rL~Lcd n-methyl gll~rAm;ne is added about 195.9 g of the ~ d methyl laurate d~s_ ;l,cd above and about 36.5 g of propylene glycol. A~er stir~ing, about 14.5 g of sodium ~ >~i~le sol~ltion (about 25% by weight in ...- 1~ 1) is s added to the reactor and the time is recorded. The ~l~lu-~ is allowed to cool to about 85~C as ~--~ ol is tlictilled o~under Alo~nsph~ ;c p~ . After about 30 ",~ no more ,~ OI is visibly ~ tillir~g SO ~ m iS slowly applied to the reaction vessel to strip out the ~ O1 and drive the reaction to r,c ~p~tion When the v reaches about 25 inches of Hg wilhou~ es;.;~, 0 r~ the reaction is complete. Afier breaking the ~ ,UUlll with l~illug~, about123.0 g water and about 72.3 g ethanol are added to the r,~lur~. The resllltir~gglucose amide sol~ltion is ~ater white' in color and ~ ,S as 95% T1A--~ 11 at 420nm.
FXAMPT F X
AMIDES PREPARED USING TRIGLYCERIDES
Tl;gl~,.;de reAA~tAntC include CRISCO~ oil, palm oiL sun~lower oiL canola oil, F~l oil, coccmlt oil palm ~teA~;~ e, and the co..~pondi.,~ hy~'~ l oils. The catalysts are aLkali metal salts of monohydric Alcl~hr~ls or polyh~d,~.A~ coho~ e.g., sodium ...~ c~ The r~;liol1 l..F.J;~ iS a .~ o.. c ---- ri c~ e.g., NEODOL~
10-8 or 23-3, or GENAPOL 26-L-5.
The .~_lion is con~ucted in a melt. The N-methylg~ Amine at a mole ratio of from about 2.3:1 to about 2.9:1 based upon the trigl~.;dc, nol~ on ~ fRrtAAnt and triglycende are co-melted at about 120-140~C under vacuum in about 30 mimltes.
About 7.5 mole%, based upon the N-methyl Bh~ e of sodium m~hnYirle is 25 added to the reaction lluAIu~. The l~&_lion mixture beG~ rs h.~9~~ ~r~_ in s~ c The reaction ~IU~ lur~. is ;--...-P-~ e1Y cooled to about 85~C. The reaction ule is ...~ -rd under V.._uul~l for about 1-2 hours .,nd is w , '~t~ at this point.
In an A1t~rnAte process, the N-~ lgl~,c~ e is muced at room t~"~ lu,~;
with the l~ nic ~-- rr~ L~.,e ;de, and catalyst. The mixture is heated to 85-30 90~C under, A~ el~, vacuum and nitrogen. The reaction ~lU~Lul~s become clearin one to one and a half hours. The reaction n~lul~ are ~ F~ at about 85~C
for about 2-3 hours.
More spe~Aifi~AAAlly~ about 127.45 g of N-methylgluc~ e l~owder is added to a 500 ml three-necked, round-bottcm flask e~luipped with an internal ~ t~r, 3s ~ m line, nitrogen line, and "~erl-A~ AI stirrer. The N-methylgl-,~A---:-~e is melted at about 130-140~C and dried under vacuum. Hardened palm kernel oil (about 156.41 g) is added to a s~_~.le 500 ml three-necked, round-bottom flask e.luil~ed with an internal ~ ...n~ AtPr and a vacuum line. The hardened palm kernel oil ismelted at about 130-140~C and dried under ~a. uuln. The dried ha..lened palm kernel oil and about 31.54 g propylene glycol are added to the N-methylgluAAmine with n~ixing. About 1.76 g sodium mPtho~n~le as a 25% I-~lu~e with ~ AI~nl is added to s this l~ UI'~, with mixing and the ".P~ is ,~ .w~d by vacuum. The ..~lu ~ is h~ Gg~ Irol~c in about 1.5 ~ IG~ at which time, cooling is applied. The "~lu~ is cooled to about 90~C in about seven ...;-~ g and .-.A;--I~;~-rd at this t~.-pe~ ul~ for about 85 ,..:....l~c The ll~lul~ is pouret out and the analysis was done by gas c~ '10~ap}'~r.
R~.-lu~lk5 the water from the ~~ aelh ~ s the r~ ;nll of fatty acid.
~rtr~.~bly, the water level is less than about 0.1%.
FXAMPLE X~
TREAI~ OF AMIDE WITH BOROHYDRIDE
About two hundred grams of a glucose smide are sdded to a one liter, three-lS rlP~ P~1, resction flask fitted with a ~ n~ Pr on a top load bAl~nce. The reactor is ~,~d to a heated mantle and cc~ P iled to a ...P-I.Ani~AI sti rer.
The t~llpc.~luie is raised to, and ~ at, about 38~C lL,c,-~ ou( the period. About 1.23 g of co~ ,;al sodium bo,ohyd,ide and about 0.20 g of pu~d~;~l sodium borohydride are added to the reactor.
There is about 0.49 g of sodium hydroxide in the borol~Lide, which raises the pH from about 8.7 to about 10.4. The starting color of the amide is albout S4%
llu~ cs ~n at 420 nAn-~.... tr;~ and after about two hours of I~A~ r.,l the transmission is about 76%. The fi,lal pH of the sollltinn is lo~ d to about 8 with 31% I,~Loel-'n- ;c acid.
The pH of 10.4 results in Lc,eas~d pro~luAtinn of soap, but a pH of more than about 10 is r~u~d for bor~J,;dc ~l~ililr. U"hcsled N~ ll-yl~ c~m;ne amide typically has a soap content of about 3.09. l~e pH~soap content of borohydride trcated N-methylgll~r-Amin~ amide varies appro~ Iy as follows: 10.1/3.14;
10.3/3.16; 10.6/3.17; and 11.0/3.41. As a result, the pH should be less than about 30 10.9 turing lr~l-- ~~1 Fx~MpLE XII
Polyll~ld,u~y fatty acid amide S~rfAr~tS~nt SOI,I1tinn as in F ~ 'S II before ~JU~ , having a % 1,~ nA--Iics ~ below about 70~/4 is treated with L ~ ug~n in ahigh pl~_ stirred reactor, heated by an internal coil co~ d to a steam/water mu~ng apl~&,~lus. The ~ iI so o co c, &l)p~-~ ~ 60% s ~rfA"tAnt 22% water, 12% c~hAnr,I and 6% propylene glycol. About 1000 g of the soI~ltinn is slu~ied with about 1.2 g of p~IIadi~lm catalyst (5% p~ lm on car'oon) wetted to W O 96/40619 PCT~US96/07124 about 50% moisture. The reactor is sealed and the agitator started at about 500 rpm.
The reactor is ~~pealedly (five times) slowly pl~ cl to about 200 psi and then slowly vented. The reactor is then p,~ ed to about 400 psi and the agitator increased to about 1200 rpm. The te.~ lu,~, is raised to about 66~C and the s reaction carned out for about two hours and the product filtered under hydrogen p-~su-e to remove catalyst. The % Il~ ''Q~is now more than about 80%.

Claims (9)

WHAT IS CLAIMED IS;

1. A process, useful in the preparation of amides of N-alkylamino polyols, selected from the group consisting of:
(1) a process carried out under non-oxidizing conditions for preparing amides ofN-alkylamino polyols comprising reacting a source of fatty acyl groups selected from the group consisting of fatty acids, fatty acid anhydrides, fatty acid esters, and mixtures thereof, preferably fatty acid ester, having greater than 98% transmittance at 460nm with an N-alkylamino polyol, preferably N-alkylglucamine, having a GardnerColor of less than 1, preferably an N-alkylamino polyol that has had its purity improved by a process of crystallization of the N-alkylamino polyol from an aqueous solution or water/organic solvent mixture and recovering the N-alkylamino polyol, said crystallization preferably being carried out by cooling an aqueous mixture of the N-alkylamino polyol to 0-10°C, and isolating highly pure crystals of the N-alkylamino polyol from the supernatant solution by filtration and/or centrifugation, and more preferably wherein an aqueous mixture of said N-alkylamino polyol is concentrated to at least about 70% solids prior to cooling and then adding from about 10 to about 200 parts of an organic solvent to the concentrated solution, and optionally, but preferably wherein the resulting filter cake, or centrifuge cake, is washed with from about 0.25 to about 1.25 parts of chilled, 0-5°C solvent, the reaction preferably being carried out in an organic hydroxy solvent, preferably selected from the group consisting of: methanol, ethanol, propanol, iso-propanol, butanol, glycerol, 1,2-propylene glycol, 1,3-propylene glycol, and mixtures thereof, in the presence of a base catalyst, the catalyst level for the preferred fatty acid ester being at a level of from about 5 to about 8 mole % of the fatty acid ester, at atemperature of from about 40°C to about 135°C, preferably from about 50°C to about 80°C, for a period of time that is less than about three hours, preferably less than about two hours;
(2) a process for removing an impurity selected from the group consisting of amine, fatty acid, and mixtures thereof, preferably amine, and optionally fatty acid, and more preferably N-alkylamino polyol, from aqueous detergent surfactant, preferably N-alkyl polyhydroxy amine amide, solution comprising treating said surfactant solution with ion exchange resin;
(3) a process for removing color body, color body precursor, or mixtures thereoffrom N-alkyl polyhydroxy amine amide comprising treating said N-alkyl polyhydroxy amine amide with reducing bleach, preferably a borohydride the treatment preferably being carried out at a pH of from about 10 to about 10.9, more preferably from about 10.1 to about 10.6,;
(4) a process for regenerating a strong base ion exchange resin containing fatty acyl anion groups comprising acidifying the resin to form fatty acids corresponding to the fatty acyl anion groups and removing the fatty acids by dissolving them in organic solvent; and (5) mixtures of said processes.

2. The process according to Claim 1 (1) wherein the resulting N-alkyl polyhydroxy amine amide is treated with ion exchange resin, preferably a mixture of acid and base resins, and then, optionally, but preferably, said N-alkyl polyhydroxy amine amide is treated first with acid ion exchange resin to convert any soap that is present to fatty acid and remove any residual amine that may be present and then is treated with base ion exchange resin to remove the fatty acid.

3. The process according to Claim 1 (1) or Claim 2 wherein said N-alkyl polyhydroxy amine amide is treated with reducing bleach.

4. The process according to Claim 1 (1), Claim 2, or Claim 3 wherein the pressure is held at a vacuum of less than about 200 mm Hg and the solvent is removed from the reaction product in less than about one hour, preferably less than about one-half hour.

5. The process of Claim 1 (2) wherein said impurity comprises fatty acid soap and said solution is first treated with acid ion exchange resin to remove the amine and convert the fatty acid soap to fatty acid and then is treated with base ion exchange resin to remove the fatty acid.

6. The process of Claim 1 (3) wherein said reducing bleach is hydrogen and the treatment is carried out in the presence of hydrogenation catalyst, preferably selected from the group consisting of: nickel and palladium catalysts.

7. The process of Claim 1 (4) wherein the fatty acyl groups contain from about 6to about 30 carbon atoms, preferably from about 10 to about 20 carbon atoms, andmore preferably from about 12 to about 16 carbon atoms.

8. The process of Claim 1 (4) or Claim 7 wherein the solvent is ethanol.

9. The process of Claim 1, carried out under non-oxidizing conditions, for preparing amides of N-alkylamino polyols comprising at least one step selected from the group consisting of: (1) reacting a source of fatty acyl groups selected from the group consisting of fatty acids, fatty acid anhydrides, fatty acid esters, and mixtures thereof having greater than 98% transmittance at 460nm with an N-alkylamino polyol having a Gardner Color of less than 1; (2) removing from an aqueous solution of said amides of N-alkylamino polyols an impurity selected from the group consisting ofamine, fatty acid, and mixtures thereof by a step comprising treating said solution with ion exchange resin and then, when said ion exchange resin is a strong strong base ion exchange resin containing fatty acyl anion groups, regenerating said strong strong base ion exchange resin by a process comprising acidifying the resin to form fatty acids corresponding to the fatty acyl anion groups and removing the fatty acids by dissolving them in organic solvent; and (3) removing color body, color body precursor, or mixtures thereof from said amides of N-alkylamino polyols comprising the step of treating said amides of N-alkylamino polyols with reducing bleach.