copyright o reso by the e-"ri"fftT"'ffrljiffi.t.tl-,"#:J*rll33'fitolr*,r.ion or the copl.right owner.

Wet Chemical Approaches to the Chara eterization ofOrganic Surfaces: Self-Assembled llonolayers, Wetting, and

the Physical-Organic Chemistry' of the Solid-LiquidInterfac e

George \{. \ \ 'hi tesides' and

Department of Chemistr l ' . Hart,ard L'ni l .er-.r ir

Recei t ,ed Not 'ember

.Phvs ica l -organic methods are usefu l in s tudf ing the sur face chemist ry 'o f organic so l ids . These meth-ods complement rhe usual spect roscopic apprbaches in character iz ing ihe so l " id- l iqu id in ter face. Th ispaper focuses on t\ \ 'o 1gpics drau'n from physical-organic surface cl iemistr l ' : preparations of orderedorganic surfaces bv self 'assembl5' of organic molecul is on inorganic suppoit. 'and uses of

" 'ett i l f i1

cha.ra.cteriz. ing these and other surfaces. l \ Ionolal 'er f i lms prepired b1. chemisorption of alkaneth"iolsand d ia lkr ' ld isu l f ides c ,n go ld are the best character ized and mbst u ' ide l l ' . tua iea 6 i i f r . ie t f -assembleds1'stems- \ \ 'ett ing is u.nlqueiv valuable in characterizing surfaces for i is combination of high surfacesens i t iv i tv and appl icab i l i r l . to d isordered sur faces.

In t rod uc t ionThe interfacial chemistr l ' of organjc mater ja ls is an

important but underdeveloped subfiejd of surfacescience.I The relevance of th i i subject ranges from tech-nologl '^ (adhesion to polvmers,2 biocompat ib le ma-terials,3 fiber-matrix interactions in composiiesn) to basicscience (u 'et t ing of sol ids b} ' l iquids,s-e cel l -surface bio-

Pau l E . La ib in i s

, C antbr tc ige. . \ /o-sscchu-(€ i r -s 021 38

2 9 , 1 9 6 9

chemistry lo) . Organic surface science is less developedthan corresponding subf ie lds focused on inorganic sur-faces (metals and metal oxides) for several reasons. First,organic surfaces are usual l l ' less ordered, less stable ther-mal l l ' . and less amenable to character izat ion by' scatter-ing and di f f ract ion techniques than are cr5 'stal l ine inor-ganic surfaces. Second, many of the advances in surfacescience have rel ied on instrumental techniques der ivedfrom surface physics.rl Because organic materials ma-vbe sensitive to radiation damage, and because they andtheir damage products are often volatile, techniques requir-ing high vacuum are often not applicable to them. Finalll',

^ (1)Swalen, J. D. ; .Al lara, D. L. ; Andrade, J. D. ; Chandross, E. A. ;Garof f . S. ; Israelachvi l i , I . ; McCarthy, T. J. ; Murrav, R. ; pease, R. F. ;Rabol l J . F. ; \ \ '1 'nne, X. J. ; Yu, H. Langmuir 1962. ' .3. 932-950.

(2) Zisman, \ \ ' . .4. ln Handbook of Adhesi t ,es, 2nd ed. ; Skeist . L. Ed. ;Van Nostrand Reinhold: New York, lg?T: Chapter 3. Schonhorn, H.ln Poll,mer Sur/oce."; Clark, D. T., Feast, \\ '. i., Eds.; \l ' i lev: Chich-ester , 1978; Chapter i0. Leger, L. Ann. Chim. 'Por is)

198?, l2, t?5-1 6 1 .

(3) Ratner, B. D. ln Biomaterials: Interlacial phenomena and,Appl icot ions; Cooper, S. L. , Peppas, N. A. . Eds. ; Adr.ances in Chemis-tr1' 199; American Chemical Societl ': \\:ashing1.on. DC, 19gZ; Chapter 2,qp 9-2.3. fg/og9 and Interlacial Aspecti ol Biomedical poll,mers;Andrade, J. .D. , E-d. ; l l -unu1.-Neu.) 'ork, 198S. Andrade, J. O.; i f taal , ,Y. . /C". lo l l 'm. Scf . 1986, 79, t4g. Durrani . A. A. : Cfrapman. D. inl9l; yer Sur/oces a_nd Interfaces; Feast, \\.. J.. N{unro, H. S., Eds.; Wilev:Chichester , 1967; Chapre_r 10. Lundsrrom, L; Ivarsson, B. ;Jonsion, U. ;FJ"Jng-.l. l2_Polyryer Surfaces and Interfaces; Feasr, $'. J., Munro,H. S. , Eds. : \ \ ' i le1 ' : Chichester , 1987; Chaprer 11.

(4) Qelannal ' , F. ; Fro-"-en, L. : Derul t te ie, A. J. Mater. Scj . 1g67.22.l -16 . Schu l t z . J . ; Lav ie l l e , L . , {CS Srmp. Ser . t g89 , Jg l , 165 -202 .

(5) \ -oung, T. Phikts. Trans. r?. Soc. (London) 1g05, 95. 65-E;.(6 ) \ \ ' enze l , R . N . /nd . Eng . Chem. l936 .26 .99g-994 . t | enze l . R . N .

1 . P_hys . Chem.1949 ,53 . 1466-14G7. B racke , M. ; DeB isschop , F . : Joc . , s ,P. Prog. Col lo id Po\m. Sci . 1988, ?6,2St-ZSg.

(7) Neumann, A. W. Adu. Col lo id Inter face Sci . lg?{, 4, 105-191.Adamson, A. \\ '. In Phl,sico/ Chemistrl ' of Surf aces,4th ed.; \\ ' i ler: NewYork; 1982. Joannl ' ,_J. F. ; de Gennes, P.-G. J. Chem. ph1s. lg-g1, g l ,552-562. de Gennes. P.-G. Reu. Mod. Ph1,s. l965, bZ, 82?-869. Schu,aru,! . t1 ' . ; Garof f , S. Langmuir 1985, 1,219-230. Schr, , 'ar tz, L.W.;Garof f ,S. J. Col /o id Inter face Sci . 1965, i06,422-457. Pomeau, y. ; \ tanni-menus, J. J . Cct l lo id Inter lace Sci . lg55, j04,47i-4t8.

(8) Neogi , P; Mi l ler . C. A. J. Col lo id Inter face Scf . 1983, 92,SBg-49.- _ (9) Jcann)1, J { J . Chem. Ph1,s. , Ph1,s.-Chem. Bio/ . lg8Z, 84,197-9.N.ggi, P.;Irl i l ler, C. A. J. Colloid Int_erface Sci. 1982,86, b25-98. Lopez,J. ; Mi l ler , C. A. ; Rudensrein, F. J. Col lo id Inter lace Scr ' . 1926,56, 4 '60-6.

(10) Albersheim. P. ; Anderson-Prout l ' , A. J. Ann. Ret, . p lant ph.r ,s-io/ . 1975, 26,31-52. Denburg, J. L. Adt , . Comp. ph1,s io l . Biochim.r978 . 7 . t ( )5 -226 .

(11 ) S_omorja i , G A. ; Bent, B. E. Prog. Col lo id Pol l 'm. Sci . 1965, 20,38-56 Feldman, L. C-; N{aver, J. W. In Fundamentats o l Surface andThin Fi lm,,4nol1,s is; Elsevier Science: New York, 1986.

,e . 1990 Amer i can Chemica l Soc ie tv0743- 7463 I 90 I 24(t6-0087$02.50 i 0

E- i Lcngr 'u i r , \ 'o l 6 . l 'o . 1 . 1990

Scheme I . l \ Ie thods for Prepar ing Funct iona l izedOrganic Sur faces: Sel f 'Assembled } lono la) 'ers ,Langmuir -B lodget t F i lms, and Sur face 'Funct iona l izedPol l 'mers

Se//-Asse mbly

Langmuir-Blodgett

RsH Tnt------> s s s

oo..'ot

H

__ci-l _-^__---..-l | =HFLj I\- ^-7 I | -

. ..-C/ I | ->( f1- -Jt I\\_t/ I | __>( X- ->{ |bIZ U t5--{_j

Reactions at Exist ing lnterlaces

H2CrOa+

,; PE

the questions of interest in organic surface chemistrl '-the t1'pe. distribution, and reactivitv of functional groupsin the interface and the character of solvent-swollen sur-face gel la1'ers-are usuall-"- more qualitative than thosethat have been the focus of t radi t io inal surface ph1's icalchemistrl ' . Accordingly, the communitl ' of scientisLs con-cerned u' i th the surfaces of stable, wel l -ordered. inor 'ganic sol ids has been less at t racted to the area of unsta 'b le. d isordered organic surfaces.

\ \ -e have been developing ph1's ical-organic approachesto the study' of organic surfaces. l2 ' r3 Ph1's ical-organicchemistry is an area based on analogy rather than onabsolute numerical measures. Thus, for example. com-par isons of values of pK" for acidic funct ional groups insolut ion and on surfaces can reveal the character of thesurface. An absolute measure of a surface propert l ' can,of course, also be useful , but i t is more of ten a compari-son of that propertS' u'ith a related propert)' of some other,better def ined system, usual l f in solut ion, that is mostefficient in y' ielding results.

A strategl' for exploring organic surface chemistrl ' basedon the ph1'sical-organic approach inevitablf involves twoareas of research that have been relativell ' uncommon asprimary foci in modern surface science. First, since muchof physical-organic chemistry is based on studies in solu-t ion, the sol id- l iquid interface rather than the sol id-vacuum interface is of primary importance. Second, reac-tivit ies (or, more precisely, relotiue reactivit ies) are invalu-a b l e i n p h y s i c a l - o r g a n i c c h e m i s t r . v a s p r o b e s f o rcharacter iz ing structure and environment in solut ion; webelieve that reactivity-based approaches also proride valu-able information concerning interfaces.

Three t1'pes of sy'stems have been the principal sub-jects of research in organic surface chemistry (SchemeI): self-assembled monolayers (SAMs) derived from alkl ' l 't r ichlorosi lanes on si l icon/si l icon dioxidela-16 and from

(12) Bain, C. D. ; Whi tesides, G. M. Angew. Chem. Int . Ed. Engl .1989, I0t ,522-528. Whitesides, G. M.; Biebul 'ck, H. A. In Mct lecularRecogni t ion: Chemical and Biochemical Problems; Roberts, S. M., Ed. lSpecid Publication No.78;Ro1'aJ Society of Chemistry Cambridge, 1989;pp 270-285.

( t3) \1 'h i tesides, G. M.; Ferguson, G. S. Chemtracts 1988, / , 171-187.(14 ) Sag i r ' , J . l s r . J . Chem. 1979 , 76 ,339-345 ,346-353 . Sag iv , J . J .

Am. Chem. Soc. 1980, 102,92-98.(15) \ \ 'asserman, S. R. ; Tao, Y.-T. ; \ \ 'h i tes ides, G. M. Langrr ,u i r

1 9 E 9 , 5 , 1 0 7 4 - 1 0 8 ? .(16) \ \ 'asserman, S. R. ; S 'h i tesides, G. M.; Tidswel l , L M.; Ocko, B.

M . ; P e r s h a n , P . S . ; A x e , J . D . J . A m . C h e m . S o c . 1 9 8 9 , . 1 1 t , 5 8 5 2 - 5 8 6 1 .Tidswel l , I . M.; Ocko, B. M.; Pershan, P. S. ; \ \ 'asserman, S. R. ; \ \ 'h i tc-s ides, G. M.; Axe, J. D. Phys. Reu. B, in press.

f i lms.oras ' \ \ 'et chemical" methods are indispensab.e : : .

preparing and characterizing these systems. A u'et cirenr-ical method of analysis-wetting-has also proved remark'ably useful and sensitive in characterizing surface

L e c ! u r e s

AuAu

(17)Nuzzo, R. G.; Al lara, D. L. J. Am. Chem. Soc' 1983, /05, 4461-a+63. Nuzzo, R. G.; Fusco, F. A. ;Al lara, D. L. J. Am. Chem' Soc' 1987,t09, 2358-2368.

t is i t i , T. T.-T. ; \ {eaver, M. J. Am. Chem. Soc' 198{, t06 ' 610?-

6108. Li, T. T.-T.: \\ 'eaver, M. J. Am. Chem. Soc. 1984, 106, 1233-

1239. Finklea. H. O.; Avery, S. ; Lynch, M.; Furtsch,T. Langmuir 1987,

3 , 40F413 .' i ig ,Porter , \ { . D. : Br ight , T. B. ; Al lara, D. L ' ; Chidsey, C' E ' D' J '

Am. Chem. Soc. 1967, 109, 3559-3568'(20t Nuzzo, R. G.; Dubois, L. H. ; Al lara, D. L. J. Am' Chem' Soc' , in

press.- tCt r Nuzzo. R. G.:7*garski , B. R. ; Dubois, L. H. J. Am' Chem' Soc'

1 9 6 7 , l ( / 9 . ; 3 3 - ; 4 0 .(22r Chidser ' . C. E. D. ; L iu, G. 'Y. ; Rowntree, P. ; Scoles, G' J ' Chem'

Phr-c. 1969, 91, 1421-4423'i : 3 rT rough ton , E . B . ; Ba in , C . D . ; Wh i tes ides ' G . N{ . : Nuzzo , R G ' ;

Ai lara. D. L. : Portcr . M. D. Langmuir 1988, 4, 365-385'L: . i i Bain, C. D. ; Troughton, E. B. ; Tao, Y. 'T. ; Eval l , J . ; \ \ 'h i tes ides,

G. \1 . J . Am. Chem. Soc . 19 t9 , l / . 1 , 321-335 ., - ! rBa in . C . D . ; \ \ ' h i t es ides , G . M. J ' Am ' Chem. Soc . 19 t6 . i J0 .

5S9t-5t9s. Laib in is, P. E. ; Bain, C. D. l Nuzzo. R. G.: \4 'h i tesides, G.

\1. Unpublished results.(26i 'Laib in is, P. E. ;Janes, L. E. ; Pr ime, K L. ; Nuzzo. R G' : l \ .h i te '

s rdes , G . lV . Unpub l i shed resu l t s . P r ime , K . L . : La ib in i s . P E ' : Bu r -

baum. B. W.; Nuzzo, R. G.; \4 'h i tesides, G. \ { . Unpubl ished re- 'u l r '( 2? ) Ba in , C . D . ; Wh i t€s ides ' G . \ { . J AT , C ' ' en ' Soc l9 t6 ' i l 0 '

656H561. Bain, C. D. ; \ \ 'h i resides' G. \1 J .4n Chen' -soc l9E9'

i l . l , ?155-7164 .t i g )Ba in , C . D . ; Wh i t cs ides , G . I { J . Am. Chem. Soc . 1988 , l J0 .

3665-3666.(29 )Ba in , C . D . ; \4 'h i tes ides , G ] r { . Sc ience tH 'osh rn i t l on , DC) 1986 '

240,6243.( ' eOt Ba in , C . D . ; \ \ ' h i t es ides . G . \ { . J . Am Chem. Soc ' 1989 , l l l .

7164-71 ;5 .(31)Laibin is, P. E. ; Nuzzo, R. G.; \ \ 'h i tes ides, G' M' Unpubl ished

results.(32t Bain, C. D. : Biebul 'ck. H. A. ; \ \ 'h i tcs ides. G. N{ ' Langmuir 1989,

v r t a a t L t

( 3 3 i B a i n . 6 . p . ; \ \ ' h i t e s i d e s . G . \ 1 . L a n g m - i r 1 9 8 9 , 5 ' 1 3 ? 0 - 1 3 7 8 '(31 ) Ba in , g . p . ; \ \ ' h r res i c ies . G . \1 . J . P r r - r - ' . Chem.1989 ,93 , 16?0-

1 6 ; 3 .(35) Strong. L : \ \ 'h i tes ides. G \ { Langmuir 1986, 4, 546-558'iSOl t " iUin i r . P. E. ; \ \ -h i tesides, G. lv{ . ; Al lara, D. L. ; Nuzzo, R. G.

Unpubhshed resul ts.(3;) G" in.r . G. L. , Jr . Insolubie )+lonolayers; Inte rsc ience: New York,

tgOO. Roberts. G. G. AdL. Ph-r- ' . 1965, 34,4i5-512. Petcrson, I ' R ' J 'Mol . Electron. 1967. 3, 103- l i l . \ 'ander lwer, M.; Barraud, A. J. Mol 'Electron. 1966. 4, 20i-221. \ 'anderlwer, M, Thin Solid Films 1988'159,243-25L Bubeck, C. Thin Solid Films 1988, -160, 1-14' Agarwal,V. K. Ph-r . " . Today 1988, 4J, 40-46.

(36t Aielah. A.; Sherrington, D. C. Polymer 1983, 24, 1369-1386'Akelah. A. J. Chem. Technol . Biotechnol . , Chem. Technol . l9t4, 34A,

263-286. Akelah. A. J. Mater. Sci. 19t6, 21,2977-300]-(39)Breeis, D. N{. J. Mater. Sfl. 196t, 3,262-265. Bla-is, P'; Carls'

son, J. ; Csul log, G. W.; Wi les, D. M.J. Col lo id Inter !gce Sci '1971'47,636-649. Xato, K. J. Appt. Polym. Scr. 1974, ,8,308?-3094' Kato, K'J. Appl . Po\m. Sci . 1977, 21,2i35-2743.-

(i0) Rasmussen, J. R.; Stcdronsky, E. R-; \\ 'hitcsides, G. M' J' Am'Chem. Soc. 1977, 99,4i36-4i45.

(41)Rasmussen, J. R. ; Bergbrei t€r , D. E. ; Whi tcs ides, G. M' J ' Am'

Chem. Soc. 1977, 99,4i46-4i56.(42) Holmes-Farley, S. R.; Reamey, R. H.; McCarthy, T' J'; Deutch,

J.; \\ 'hitesides, G. M. Langmuir 1985, t, 725-740.(43) Holmes-Far ley, S. R. ; \ {h i tcs ides, G. M. Langmuir 1986' 2,266-

2 8 1 .(44 )Ho lmes-Far ley , S . R . ; \ 4h i tes ides , G . M. Langmur r 1987 ,3 ,62 -

?6 .(45) Holmes-Farley, S. R.; Nuzzo, R. G.; McCarthy, T. J.; White-

s ides, G. \1. Langmuir 1987, 3, 799-815.(46) Holmes-Far le i ' , S. R. ; Bain, C. D' ; Whi tcs ides, G. M. Langmuir

r 9 8 8 . 4 , 9 2 1 - 9 3 7 .(47) Wi ison, M.D.; Whi tesides, G. M. J. Am. Chem. Soc. 1988, l . l0 '

8?16-8719 .(48) U' i lson, M. D.; Ferguson, G. S. ; \ l h i tcs ides, G. M. J. Am' Chem'

Soc., in press. \\ ' i lson, "M.

D.; Ferguson, G. S.; Laibinis, P' E';\ \ 'h i tes ides. C. M. Unpubl ished resul ts.

The Langmu i r Lec tu res

Scheme I I . Go ld ( I ) A l ky l Th io l a tes A re t he P roduc t s o fReac t i ons o f bo th A l kane th io l s and D ia l kv l D i su l f i dess ' i th a Gold Sur face

R S H + A u l

RS'Au* 'Au l

R S S R + A u l

functionall luts'23-33'a5-a? "n6

structure2s-30'44'45 in theseclasses of materials. Traditional spectroscopic methods-especially XPS,ts'zs-34'36'45 polarized infrared externalr e fl ectan ce spectroscopy (P I ERS ), I 7're'20'23'25-27'3r'36

"n 6

opt i cal ellipsometn'l 5-l ?'r s'23-31 -ale, of course, also hi ghll'useful .

An important mot ivat ion under l f ing these studies inorganic surface science is to ans'*'er. for these materials.one of the fundamentai questions of molecular-level struc-ture of matter and its macroscopic ph1'sical properties.\\te have used organic materiais in these studies for tu'oreasons: organic mater ia ls are amenable to a wide rangeof variation in molecular-level structure: surfaces are acces-sible for character izat ion and studl ' . \ \ 'e have used uer-tabi l i ty as the ph1's ical properr) ' to be measured, becausei t is easi ly measured exper imental l l ' {e and brc,aCl l ' re le-vant technological lv. l -4 '50 One pr imarl ' focus of ourresearch can thus be reduced to a speci f ic quest ion: 'Hos

does the molecular-level structure of organic surfaces iniu-ence their wettabil ity by water?"

\\:e start this account with a brief description of me:.h-ods used in preparation of self-assembled monolayers andsurface-functionalized pollmers. \\:e continue s'itir a qual-itative discussion of the ph1'sical-organic chemistrl ' ofwetting and particularly of its use in measuring the deprhsof funct ional groups below the 'surface' of an organicsolid in contact with water. \\re close with an outl ine ofexper iments that extend these techniques for prepar inghomogeneous. self-assembled monolayers to systems com-bining sel f -assembl l ' u ' i th microl i thograph5' to generatepatterns in the plane of the monolal 'er .

Preparat ions and Substrates

Se l f -assembled mono la l 'e rs p repared by chemi -sorp t ion o f a lkaneth io ls o r d ia lk f ' l d isu l f ides ongold are easi l - r 'obtained b1, 'exposing a c lean gold surfacefor a few minutes at room temperature to a 0.1 mlr{ solu-t ion of the organosul fur compound.tT, ts '24 These s1's-tems are the best understood of the organic surfaces nowbei n g stu d i ed. I ?'1e-22'24's4'35 Surpri singly, und erstandin gof the reactions involved in interaction of the sulfur withthe gold is st i l l incomplete (Scheme I I ) . The organosul-fur species formed from both alkanethiols and dialkl ' ld isul f ides at the gold(0) surface is a gold(I) th io late (RS-Au*) -qdsorbed epi taxial l l ' on the gold(0) sub-strate.22'32'35 In the case of dialkr.l disulfides. this sur-face gold thiolate is undoubtedly foimed br.oxidative addi-t ion bf the sul fur-sul fur bondto gold(0i . Format ion ofa gold thiolate from a thiol requires loss of h1'drogen,

(49)Bigelow, W. C.; Picket t , D. L. ; Z isman, \ \ ' . A. J. Col lo id Sci .1946, t , 513-538. Johnson, R. E. ; Det t re. R. H. J. Col lo id Sci . 1965.20.1?3- l?6 . Neumann , A .W. ;Good , R . J . In Sur /oce and Co l lo id Sc ience ,Good, R. J. , Stromberg, R. R. , Eds. ; Plenum: New york, l9T9; Vol . 11,p p 3 1 - 9 1 .

. (50)_Markgra! , ! .4. Treatment Required t 'or Pr int ing u, i th l l 'a ter-bosed . lnAs; TAPPI Proc. , 19E?; p 333. Armsrrong, J. A-. ; \ \ 'h i tes ides.G. M. Chem. Eng. .Veus 19t6, 64 (Aug 11),22-ZS: Research Br iet ' ings.1986; Nat ional Academ-r ' ; \ \ 'ashingron, DC, 1986;Proc. .Nat l . ,4cod. Si i .L: .S.A. 1987. 84. 4665-4?48.

Langmuir , \ /o1.6, No. 1, 1990 89

but u'hether this hydrogen is lost as H2 or lost as waterb)' reaction with traces of oxidants in the s1'stem is notknou'n. In both cases, adsorption is fast but not diffu-sion-limited. The half-t ime for adsorption of both octa-decanethiol and dioctadecyl disulfide is zrr, - t h froma 1 !\{ solution in ethanol and faster for more concen-trated solutions.2a The energy of the bond between theorganic thiolate and the gold surface is high (approxi-mately 40-45 kcal/mol): desorption of these organosul-fur species from the surface is slow. Exchange of alk5'lthiolates on the gold surface with thiols or disulfidespresent in solution is relatively slow (requiring hours toda1's to proceed to significant conversion). Under vac-uum, the species that desorbs on heating is a dialkyldisul f ide.2t

It is unclear how close these systems are to thermody-namic equil ibrium. Certain observations-for example,the absence of large-scale ' islanding' of the componentsin a monolaver comprising a mixture of different orga-nosulfur compounds29-3r-strongly argues against wide-spread lareral mobi l i ty of the alkyl th io late moiet ies andthus against a s)'stem at equil ibrium. The preferenceobsen'ed for long-chain alkanethiols over short-chain onesin cc,npetr l ive adsorpt ions at gold surfaces2s-sl is whatone uould €rpect on the basis of thermodl 'namic consid-era: ic,ns. Thus. the s1'stem shoq's c lear evidence of k inet icccrni :o l . but rnal ' . nonetheless, produce resul ts that cor-relate u ' i th those expected for a s) 'stem at equi l ibr ium.

The structure of the monolal 'ers is better understoodthan the mechanisms b1' n 'h ich they' form. In br ief , anumber of studies nor+' indicate that long-chain alkane-thioiates adsorbed on gold const i tut€ monolal 'ers that areat iea-st quasi-crystalline.rts2as The a\'l chains are largel-vor exclusivelv t rans extended, s ' i th the axis of the chainti lted - 30o from the normal to the surface (Figure 1).is-ztThe suifur atoms rest in threefold hollows of the gold(I11)surface.35 Computer simulationsst suggest that the out-ermost ends of the chains show greater mobil ity than theinner parts and may have occasional gauche conforma-tions; the inner parts appear to be effectively crystal-I i n e .

Surface Chemistry of Surface-Oxidized Polyeth-1'lene.

'Polyethy'lene carboxylic acid" (PE-CO2H) hasbeen reviewed thoroughly.13'52 This material is formedb1' brief treatment of commercial, low-density polyeth-1'lene fi lm with chromic acid solutlon.3s'+0 The predom-inant reactive functional groups at the surface of PE-CO2H are carboxl' l ic acids, although other carbonyl-coniaining moieties are also present.ao Alcohois are notpresent in the oxidized interface in significant concen-trations. Oxidation is accompanied by extensive etch-ing of the surfaces if i t is prolonged, and the surfaces areprobably microscopicall l 'rough, even when the exposureto the chromic acid solution is short. The polar func-tionalit ies are distributed over an interfacial region 2-3nm thick.a2 There are undoubtedly differences in thedensity and distribution of functional groups in crystal-l ine and amorphous regions of the poly'mer.

Despite (or perhaps because o0 the disorder in the struc-ture of the functionalized interface of PE{OrH, this mate-rial and its deriva1l"ur4r'45-nE (obtained by the familiarreactions of organic synthetic chemistry) are particu-

(51) Harr is , J . ;Rice, S. A. J. Chem. Ph1's. 1988,88, 1298- i30E. Bare-man, J. P. ;Cardin i , G. ;Kle in, M.L. Mater. Ber, . Soc. Sl , .p.Proc. 1989,14l ,4I I -418; Ph: 's . Rec. Let t . 1988,60, 2L52-5; Chem. Phys. Let t .1988, t45, 4E3-8. Northrop, S. H. ; Cun' in, M. S. J. Phl 's . Chem. 1985,89, 4707-13.

(52i Ferguson, G. S.; \\ 'hitesides, G. M. In Modern Approaches toV'ettabil ity: Theorl' and Applications; Schroder, M., Loeb, G., Eds.,Plenum: Nen' \ 'ork, in press.

90 Langmuir . \ ' r , i 6 . . \ 'o . t . 1990

- , ^ r ^ r ^ \ ^ ' ^

Figure l. Schematic i l lustration of an alk5'l thiolate mono'la.,*er on Au(i11). The top figure, a top view, shows the infer-ence from electron diffraction experiments (ref 35): the adsorbedthiolates are epitaxiall5' located on the gold surface. The sul-fur atoms are located in the threefold hollow sites. The circlessurrounding the sulfur atoms are used to suggest the area par'allel to the gold surface occupied b1' the alkl ' l chain. The bot'tom figure, a side view, shows the orientation of the alky'l chainsinferred from polarized infrared external reflectance spectros-cop5' (PIERS) (ref 20). These alkl ' l chains exist largelf in anordered. trans, zigzag conformation canted at an angle of approx-imatell ' 30o from the normal to the surface. The presence ofgauche conformations in the alkl ' l chains cannot be ruled out.

larly useful substrates u'ith s'hich to explore propert)'-structure relat ionships in organic surface chemistr l ' . PE-CO2H is readill'available and easill'manipulated. It showscomplex phenomena such as surface reconstruct ion onheat ingas'52 or mechanical deformat ions3 that are notobserved with SAIt{s on rigid substrates. Despite the obvi-ous differences between PE-CO2H (and its derivatives)and sel f -assembled monolal 'ers of organic th io ls on gold.results obtained with it correlate surprisingll ' closell 'u' iththose obtained from the better defined but more diff i 'culty obtainable SAIr{s.26'a8

Phys ica l -Organ ic Chemis t ry o f Wet t ing

Wetting has been considered an archaic technique. capa-ble of f ielding l itt le detailed information about surfaces.Measurement of a contact angle of a l iquid on a solidsurface typicall l ' f ields a single number. The area of solidin contact wi th l iquid (-1 mm2) is large, and the con-tact angle averages contributions to solid-liquid and soiid-vapor f ree energies f rom the smal ler regions (10-100 A2)relevant to molecular-scale characterization of surfaces.The form of Young's equations {Figure 2) is such thatcontact angles can be interpreted onlf in terms of dif-ferences and ratios of surface-free energies, rather thanas a direct measure of 'ysl or 7sv. Many features of thetheoretical underpinnings of wetting remain unsatisf5'-ing. and the technique retains a strongly empir ical char-acter. For these reasons, the technique of contact anglemeasurement has been (correctl l ') considered to be aninformation-poor technique, at least relative to some ofthe surface spectroscopies (u 'hen theS'are appl icable). Inour hands, however, wett ing has proved invaluable for

A s t h e d r o p s p r e a d s :

1 ) T h e S V r n t e r l a c e d ? c ' c r ! ? r ' . r o , . 3 .

2) The LV interface incr?!r.r l .rnrrvt.o'e

3) The SL in te r lace inc re !ses / t r

c o s 0 o t r e e e n e r g y

Figure 2. Schematic i l lustration of a spreacirng ctop of .;q:::in contact with a solid surface, showing the rela"16n5 !r1:rec:'.the relevant parameters: the contact angle, ft the rcird t'rp< rinterfacial free energl', 'ysvi the l iquid,'r 'apor inte rfec;a- i:etenerg\ ' , 1, r ' : and the sol id/ l iquid interfacial f ree e:- . t :51. ' . 'Yodi's e{uation describes the ielationship between these pJ'eters lor a stationary drop at thermodynamic equil ibrruc'

probing the charact.er of solid-l iquid and solid+olid inler-faces involving organic solids-an area of investigationin u'hich surface spectroscopy has l imited usefulness. Therange of information obtained by contact angle measure-ments can be increased by measuring c-o-n-tact angles asa function of the pH of aqueous dropsls'23'33'42-46 (*con-

tact angle t i t rat ion-n2), by using probe l iquids other thanwater.25'26'30'31 and by examining both advancing andretreating contact angles.so'3l Like most techniques inph1's i cd--or g an i c c h e m istry, m ea-surem ent of contact anglesrei ies heavi l l ' on comparisons of measurements in s imi-lar s1'stems rather than on interpretation of absolute val-ues obtained from only one s5'stem.

Figure 2 summarizes the essential features of a first-ordei anall'sis of contact angles. The thermody:amic anai-

5'sis in this s1'stem is familiar: following application of adrop, the drop edge expands. This expansion increasesthe area of the l iquid-r'apor interface (always an ener-get ical l l ' unfavorable process) and decreases the area ofthe solid-\ 'apor interface (always a favorable process).The area of the sol id- l iquid interface also increases. Inpr incipie, the drop edge comes to rest u 'hen the changesin energies that accompan] ' these changes in areas bal 'ance. The relat ionship descr ib ing this equi l ibr ium si tu-ation is given b1'\ 'oung's equation.s This equation relatesthe cosine of the contact angle and a ratio of interfacialf reeenergies. I t assumesasl 'stem atthermodynamicequi-l ibrium and a smooth, homogeneous, unreactive surface.\\ 'e emphasize that, in using this approach, one shouldbe aware of its l imitations and approximations and notethat for man) s)'stems in organic surface chemistry thehysteresis in contact angle-that is, the difference betweenthe advancing and retreating contact angle-is far fromzero. A large value in hysteresis is commonly taken toindicate a s)'stem not at equil ibrium. Thus, we use con-tact angles because they are convenient, because thel'arevery sensitive to details of interfacial structure at the ang-strom scale (see below), and because they are appl icableto the characterization of solid-l iquid interfaces. \\ 'ebelieve these measurements at least correlate *' i th ther-modynamically' significant measures of surface and inter-facial free energies.

fn general, we follow Zisman and others in using cosd' rather than d", in our work, because cos d is propor-tional to interfacial free energJ-; we use advancing ratherthan receding conlact angles because the hysteresis inthese s1'stems is often very large, and d, may be zero evenu'hen d" is large.

Figure 3 summarizes data from measurements of con-

Lcngmui r Lec tures

C C S " + . S L = Y S V

s v ' ' s LC 3 a . . =a-

T h e L c i ; T - : . L € c i i a r € , i Langmuir , \ 'o1.6, No. 1, 1990 91

I

"sls[\$,

PE[>c:o] [coN" 0 rP E . H

P E I > C = O l I C O . C H 3 ]

P E [ > C H O H ] [ C H 2 O H ]

P E [ > C = O ] [ c o N " a

t , " t r , Q

P E [ > C = O ] l c o . H J

p|r

f!_Sufg Q.- Ch.ange in the advancing contact angle of water onPE-CO2H and der ivat ives as a funct ion of the pH of the water.Data fo i unfunct ional ized PE-H are included for reference.

tact angle on a number of der isat ives of PE-CO,H. Thesedataa6 provide a useful introduct ion to the ulse of u 'et-t ing in character iz ing the interface betu'sEn organic sol-ids and u 'a te r : the l ' demonst ra te severa l in te res t ing p l :e -nomena and h in t a t the use fu l sens i t i r . i t l - o f u .e t r ing icthe depth of funct ional groups belo\a ' the organic u.ale:interface. This f igure plots the advancing contact angieof buffered u'ater on a number of der ivat i r -es of pE-CO2H as a funct ion of the pH of th is water. Unfuncr i r , : , -a l ized pol l 'eth1' lene is, as one would expect, h l .drophobic( that is, the value of {9" on i t is large) and insensi t ive rothe pH of the drop. In comparison, PE-CO2H is jesshvdrophobic and shows a t ransi t ion f rom a more h1'dro-phobic state at low pH to a more h1'drophil ic stare arhigh pH. It is clear that this transition reflects the con-version of PE-CO,H groups to PE-CO,- groups as thedrop becomes more basic. This s1'stem still defies detailedanalysis because it exhibits large h1'steresis and becauseit is unclear how to account for the reaction of carboxS'-l ic acid groups u' i th h1'droxide ion in analyzing the angleat which the spreading of the drop stops. The contactalgle titration curves of tertiary amines also t1'picall l 'show inf lect ions; those of many pr imary amines do not.for reasons that are complicated.ao \\ 'e note that the val-ues of solution pH corresponding to the half-way pointsof these inflections are several pH units removed fromthe values of pK" of the corresponding functional groupsin homogeneous aqueous solution. A large part of thisdi f ference is probably at t r ibutable to the thermod5.-namic diff iculty of creating a charged species in the rel-atively low dielectric constant region represented b1' apolyethl' lene-u'ater interface,43'n6 but other factors mavalso contribute.

Perhaps the most interesting and unexpected observa-tion in the data of Figure 3 is the high value of conracta n g l e o f P E - C O N H C . H r . o t T h i s s u r f a c e i s v e r l 'hydrophobic-even more hydrophobic than pol1,eth5'i-ene i tsel f . The amide group is i tsel f very polar, and rhepresence of h igh densi t ies of amide groups at the surfaceof PE-CONHC6H' might have been expected to makethe surface of this material h5'drophil ic. The fact rhatthis surface is more h1'drophibic than u'functionalizedpolyethylene is probably a combination of two factors.

Figure .1. Schematic i l lustration of self-assembled monolayerson gold: lA) pure HS(CH2)rrOH f ie lds a high-energy surfaceconsisr ing of CHrOH groups, (Br_pure HS(CH?)2tCHr. f ie lds aJou'-energl ' surface presenr ing CH" groups. and (C) mixture ofthe tu 'o th io ls f ie lds a complex, d isordered interface.

First, the aromatic ring must effectivell ' completel5,' hidethe underlling polar amide groups from contact with water.Second, the roughness of the surfacea2 undoubtedly con-t r ibu tes to i t s apparent h1 'd rophob ic i t l ' .6 The f i rs tobsen'ation-that a phenS'l ring is sufficient to hide apolar amide group-suggests that a relativell 'small, non-polar group can bury a polar one and that wettabil itymust be sensitive to molecular-scale details of interfa-cial structure.

Although the thermodynamics and kinetics of spread-ing of u'ater on chemically and morphologically hetero-geneous organic surfaces are certainly sti l l notunderstood,i '8 it is clear that measurements of contactangle have a number of useful characteristics. First, con-tact angles are sens^itive to the polaritl'of functional groupsin the surface.ot-nt Second, thel ' respond to the state ofjc,nizai ion of rhese groups and can thus be used to detection;zabie funct ional i t \ ' . ' f4-4; Third, they are sensi t ive toic 'ca l c ie ta : l s o f s t ruc tu re a t the so l id - l iqu id in te r -t -ce .1- i ! Charac ter is t i cs o f u 'e t t ing tha t a re nou ' la rge lyignc,rei-especial l l h_\ 'steresis and kinet ics of spreading-have' .he pot€nt ia l to provide addi t ional informat ion, oncetheir measurement is ' * 'e l i control led and their interpre-ta t ion unders tood.

Sel f -Assembled tr Ionolayers of Organic Thiols onGold : S t ruc ture and Use in S tud f ing the Depth

Sensi t iv i ty of Wett ing

Our most use fu l exper iments2T-31 based on se l f -assembied monolal,ers have utiiized so-called mixed mono-la1'ers: that is, monolayers prepared by exposing a goldsubstrate to a solut ion containing a mixture of twoalkanethiols in concentrations chosen to give control overthe relative concentrations of these species on the sur-face. Figures 4 and 5 i l lustrate the degree of control thatcan be exercised in building monolayers by using self-assembly. In this representative experiment,2s we beginwith tu'o components: a long-chain alkanethiol terminat-ing in a methl' l group and a short-chain aikanethiol ter-minating in a hydroxyl group. We have prepared mono-layers from solutions containing these two componentsover a range of values of the ratio .R = [HS(CH2)ro-CH20H].or,,t ion/[HS(CH2)2rCH3]"orution. Physical mea-surements of relevant properties of the resulting mono-la5'ers are summarized in Figure 5. Pure monolayers ofthe long-chain alkanethiol (Figure 4) form a thick, hydro-phobic monolal'er containing no ox)'gen by XPS spec-troscopt'; pure monolal'ers of the short, h5'droxyl-termi-nated thiol form thinner monolayers, containing oxygenby XPS. The macroscopic property of the system-

t z v

R N I

)"-r***t

N,9 \I

(cH2): !

S H

nO H

( c H 2 ) r r

S H

v0

4 0

t ' rF /

9 2 L a n g m u i r , l / o 1 . 6 , I ' o . l , 1 9 9 0

100

80 e- t

60

1 r 0

F c lrlSC,oCH2OHl p'iscictl

Figure 5. Properties of monolal'ers formed bl' the adsorptionof mixtures of HS(CH2)rrOH and HS(CH2)21CH3 onto gold f romethanol ic solut ions. R is the rat io of the concentrat ions of thet$'o components in solution. Squares and circles represent datafrom tu'o separate experiments. The lines are provided as guidesto the e1'e. Upper f igure: e l l ipsometr ic th ickness. \ f iddle f ig 'ure: advancing contact angles of water (open s1'mbols t and hexa-decane (solid s1'mbols) obtained b-v rhe sessile drop technique.Lower f igure: areas of the Au(1f-72) (open s1'mbols) and O(ls lpeaks (solid s)'mbols) obtained b1' XPS. The vertical scale isarbi t rary. Data u 'ere col lected on a SSX-100 X-rav photoelec't ron spectrometer (Surface Science Instruments) u ' i th a mono-chromat ized Al Ka source. 100-eV pass energr ' . I 'mm X-ra1'spot, and 35o takeoff angle. The peaks u'ere fitted q'ith a svm-metr ical 90% Gaussian/10% Lorentzian prof i le,

\ r 'et tabi l i t -v by water (and also b1' hexadecane. HD)-correlates wel l wi th the th ickness and ox) 'gen content ofthe fi lm. Thus, by simpll ' r 'arf ing the value of .R from-1 to 100 i t is possible to control the th ickness of themonola) 'er (*1 A) f rom -25to *10 A and to control i tsn 'et tabi l i t5 ' by water f rom 0^) l10o to d" < 10o. Thisabi l i tS ' to control s imultaneously ' the composi t ion, struc-ture, and macroscopic propert l ' of these mixed mono-la1'er f i lms makes them ideal for studf ing structure-property relations in this class of mat,erials. The abil it i 'to control the fi lm thickness to a few angstroms by sim-ple organic s1'nthetic techniques is particularly notewor-thy'. To achieve this level of control b1' using the tech-niques of classical surface science-vapor-phase epitaxl',spu t te r ing , chemica l vapor depos i t ion , and o thers -requires sophisticated and often expensive experimentaltechniques.

The abil it-v- to prepare self-assembled monola-'-ers incor-porat ing a range of types of organic structures makes i tpossible to study the depth sensitivity of u'etting in somedetai l . Figure 6 shows a representat ive exper imentalapproach for the measurement of depth sensi t iv i ty. Inthis experiment,zs we have positioned a polar function-alit l ' , an etheral ox1'gen atom, at various distances fromthe monolal, 'er-water interface b-"- varf ing the length ofthe n-alk1' l chain, C,,H2' ,*1, in a ser ies of a lkanethiols ofstructure HS(CH2)r6OCnH2n*, . \\ 'ettabil it l ' by wat€r t 'ar-ies s igni f icant ly as CnHr, ,*r ranges from methl ' l to butvlbut is constant for alkyl groups larger than pentl ' l (at avalue approximately that expected for molecules contain-

Figure 6. \ fonolal 'ers formed b5' adsorpt ion of a lkane' .hrc i=terminated bv alk l ' l ethers onto gold. Upper f igure: Scherr .a: ici l lustrat ion of a monola,ver formed b1- adsorbing HS,CH-' , r -OtCH,t"CH" onto gold. Lower-f igure: Advancing contaci a: : ; ,esof qater {oi . g l l 'cerol (O). and hexadecane (r) on monc, ia le:sformed b1'adsorbing HS(CH2)r6O(CH2),CH, onto gold as a f ';:rc-tion of the length of the terminal alk]' l chain. PEG (poli ' teth-}' lene gll 'col)) is included for comparison as a surface in rr 'hichrhe ether l inkage is exposed to the contacting l iquid. The valueof the contact angle of \r 'ater on PEG is approximate, since PEGrapidll ' dissolves in the drop of water.

ing no ethereal ox1'gen). !\7e infer from this experimentthat the thickness of a hydrocarbon layer provided b1' aCn alki ' l chain is sufficient to hide the polar ox-vgen func-tionalit l ' from contact with l iquid water.

An obvious question in this type of experiment con-cerns the order in these functionalized monola5'ers, rel-ative to the order in monola)'ers derived from unfunc-tionalized alkanethiols. A point of particular interest con'cerns the order in that part of the monolal'er occupiedbl ' the group R (Figure 6). Is i t necessary to have a highdegree of order in this R group to achieve effective screen-ing of the ethereal ox1'gens from water? \\ 'e have begunto explore th is quest ion b1' examining a number of inter-faces having polar functional groups separated from thesolid-u'ater interface b}'r'arious means%'€ (Figure ?). Thisfigure summarizes results from several systems: mono-layers on gold incorporating an ether or amide group asthe 'h idden' polar funct ional i t l ' ,26 and amide der iva-tives of PE-COrH.ot The interest and surprise in thesedata are that all of these s1'stems show similar behavior.Thus, a range of organic surfaces, varf ing in order fromthe quasi-cr1'stall ine alkanethiolates (RS-Au) on gold tothe disordered functionalized polyrners PE-CONHR, allshow similar response in their wettabil ity -by water tothe length of the n-alk1'l group R. \\7e infer from thesedata that a high degree of order is not a prerequisite forburf ing the polar functional groups.

The high sensitivit-v of wetting to local details in struc-ture can be used as the basis for s5'nthesis of new t1'pesof materials shou'ing unusual and potentially useful behav-ior . Figure 8 shows the wettabi l i ty of a mater ia l , PE-CONHC6Ho-o-COrH, obtained bl ' convert ing the carbox'

I ' l ic acid groups of PE-CO,H to amides_by coupling withinthrani lc acld (o-aminobenzoic acid).a? At low pH, th ismater ia l is extremel l ' h1'drophobic-more hydrophobicthan unfunctionalized poly'eth1'lene itself. At high pH,i t is as h1'drophi l ic as any der ivat ive of PE-CO,H that

6I

Aq

The Langmuir Lectures

a i a l e r

: i r c e ' o

a F a r t c ? c a ^ Q

C H ,( C H : ) "

o- I-

tCx ,1 , "IIIS x

1 2 0

o

9 0

. n (

0.0

cos C' l

I C

- - - . . /4,

c ' /

r\ /t'

" \ (

4 0" t

C N

J U

l.

\ \ t '

> >

t 5 3 = E :E - i - d =

E l h e r

T h e L a n E r . ' - : r L e c t L ' 6 s

-= , / -PE coNH1CHr lnH r r

/ , - , .Au S(CH2)roO(CHz)nH , r t

A u ' S ( C H 2 ) 1 I C O N H ( C H 2 ) n H { o )

I 1 0 1 2 1 4 1 6 1 8 2 0

n

Au

Figure 7. Depth sensitivity of wetting by water on various inter-faces containing a submerged polar functionalit:*: advancing con-tact angles of u'ater as a function of the hydrocarbon chain lengthseparating the polar functional i t l ' and the sol id-] iquid inter-f ace. The pol l 'eth5' lene derivatives (r) were prepared b1'reac-t ion of PE-COCI u' i th alk5' lamines (refs 46-48); the thiolatemonolal 'ers on Au were prepared b5' adsorption of authenticsamp les o f HS(CH2) r6O(CH2) .H (o ) and HS(CH2)TCONH-(CH2)"H (o) and b1' reaction of monolavers prepared b1' adsorp-t ion o f HS(CH,) , , ,NHo onto go ld r . r ' i th a lkanov l ch lor ides (O){ re f s ?3 and 2a i . ' \ \ ' e 6e l i eve rhe da ta rep resen red b1 'o we refor monoiavers that were iess ordered than the other svslems.

Langntu i r , \ 'o1.6 , . \ 'o . 1 , 1990 93

Figure 9. Schematic i l lustrat ion of monolal 'ers formed b1'adsorpt i on o f m ix tu res o f HS tCH2) , ,OH and HS(CH2) rsOH on rogo ld :(A) pure monola l 'e r o f HS(CH"t ,oOH; (B-D) monolayers con-taining different mixtures of the t$'c, components; (E) pure mono-l a1 'e r o f HS(CH2) r lOH.

' o jII

" "1J L l- l

= l

."J:r" t

1 1 0 1 0 0 €

F = [ H S C T T O H ] [ H S C T s O H )

Figure 10. Properties of monolal'ers formed bi' the adsorp-t ion of mixtures of HS(CHr)rrOH and HS(CH2)rsOH onto goldfrom ethanolic solutions as a function of solution composition.R: ell ipsometric thickness (O), cosine of the advancing contactangle of wat.er (O). Inset: expanded plot of cos d. in the regionof the maximum h1'drophobicit l ' .

phil ic. In going from the lou'-pH to the high-pH forms,the elemental composition of the interface has not changed.Thus, the wettabil ity of this surface-functionalized poll '-mer must be due to relatively small changes in local con-format ion of the anthrani lamide moiet ies-perhaps nomore than a shift in the position of the aromatic ring bya few angstroms.

Figures 9 and 10 show another kind of behavior reflect-ing the high sensitivity of wetting to Iocal structure atthe solid-water interface. The system examined in theseexperiments2s'30 comprised mixed monolayers obtainedby chemisorption of short-chain and long-chain hydroxyl-terminated alkanethiols on gold (Figure 9). Our hi 'poth-esis was that pure monolayers of either short- or long-chain components would have highly ordered structurespresenting two-dimensional arra)'s of h1'drox1'l groups atthe solid-l iquid interface. Ir{onolal'ers derived from eitherprecursor u 'ould be polar and essent ia l ly indist inguish-able b1' u 'et tabi l i ty . In monolayers containing a mixtureof these components, wettabil ity of the s1'stem should

ni"i i ' ,?*. n ? 'l ' I l l ' ' i * 'I ' S H Y S H

IFI

I

II

/E : : S : -t

j-o u- . ]

llII1

?. "l

3lI

0 . , j

tffnt

0.8

a n

r ( gc

J U E

0

-\-\ \-^{ ),---

-/\ ,,' \., //

r l F:' 1 . 0

p HFigure 8. Right f igure: advancing contact angle of u'ater onPE-CO?H and anil ide derivatives as a function of the pH ofthe l iquid. Data for PE-H and PE-CONH2 are inciuded forreference. Left figure: proposed change in conformation for PE-anthranilate used in rationalizing the extraordinary change inwettabil it l 'observed between low and high pH (upper and lowerconformat ions. resoect ivelr ' ) .

we have observed. This extreme change in wettabil ityu ' i th pH is remarkable, and n' i l l , n 'e bel ieve, form thebasis for the design of pH-sensitive devices. Its detailedorigin has ncit been established. although \a'e believe it tobe due to a change iriconformation with pH. We h1'poth-esize thht at low pH the carboxl' l ic acid group is proto-nated, and the interface adopts a local structure thatexposes to water primarily the C-H bonds of the aro-matic ring. In this low-pH conformation, the surface wouldthus resemble the surface of l iquid benzene and wouldbe hydrophobic. It is possible that intramolecular h5'dro-gen bonding is responsible for the lou'-pH conformat ion,but we have no direct structural support for th is hi 'poth-esis. At h igh pH, u 'e h1'pothesize that the carboxi ' l ic acidgroups ionize, the surface reconstructs b) ' a rotat ion ofthe anthrani lamide moiet ies that exposes the resul t ingcarboxl'late ions to u'ater, and the interface becomes hvdro

4

91 Langmui r , \ ' o l .6 , . \ ' o . 1 , 1990

be qui te di f ferent provided that the mixture u 'as an int i -mate one-that is, that there \ r 'as no segregat ion of thet \^ 'o components into * is lands.- \ \ 'e expected, in thesemixed monolal'ers, that the portions of the monolal'erc lose to the goid u 'ould remain t ight ly packed and qua-si-cr1 'stal l ine but that the outer port ions of the mono-la1'er u'ould be disordered u'ith methS'lene units exposedto the contacting water. In these intermediate regionsof composi t ion, u 'hen the monolayer contained a mix-ture of the long- and short-chain th io lates, i t u 'ould thusbe expected to be significantly h1'drophobic. Figure 10shows both contact angles and thicknesses of pure andmixed monolayers. The plot of R versus thickness estab-lishes the range of concentrations of thiols in solutionfor which the composition of the monolal'er is in transi-tion (R = 6-7). In this transition region-u'hen the mono-layer would be expected to be maximal ly ' d isordered ini ts outer structure-the contact angle undergoes a s igni f -icant excursion from h1'drophil ic tou'ard h1'drophobic.

This peak in h1'drophobic i t l ' reconf i rms the sensi t i r ' -i t1 ' of u 'et tabi l i t l ' to local structure of the sol id- l iquidinterface. I t a lso provides an important piece of evi-dence that mixed monola) 'ers prepared by coadsorpt ionof tu 'o alkanethiols f rom solut ion do nof, in fact . formphase-segregated, ' is landed- mixtures but rather mono-la1'ers in which the two alkanethiolates are mixed later-al ly on the gold surface. \ \ 'e do not know the minimumsize of islands that would show the same behavior as puremonolal'ers of the components of '"r 'hich thel' were made.but i t seems l ikel l ' that mixtures of is lands having radi igreater than approximatell ' 100 nm (for monolal'ers 2 nmthick) would be di f f icul t to dist ineuish f rom extendedsheets of pure components.

Ex tens ions o f Se l f -Assembl l ' to Organ ic Sur facesHav ing Pat te rns in the P lane o f the Sur face

Our exper iments u ' i th sel f -assembled monolal 'ers ongold, and u' i th organic pol l 'mers. hare f ie lded s1'stemsin u 'h ich composi t ion and structure can be manipulatedconvenient ly along the perpendicular to the plane of theinterface. Clear l l ' , the degree of control over composi-t ion and structure is greater for sel f -assembled monolal ' -ers than for functionalized organic poll 'mers, but in bothcases there are strong similarit ies between the s5'stems.Neither, however, provides immediate opportunit) ' to pre-pare s5'stems in which structure can be controlled in theplane of the interface. We have recently developed (incollaboration with James Hickman and Mark \\ 'r ightonat MIT) a model s1'stem that combines microlithogra-phy with self-assembly.u' \\ 'e believe that this s1'stemdemonstrates a widely applicable method for the prepa-ration of organic surfaces having a useful degree of con-trol over composition and structure in the plane of thesurface.

The basis for this experimental technique is summa-rized in Figure 11. \\ 'e pattern a substrate of interest( in these exper iments, s i l icon ni t r ide) b1'deposi t ing th ingold and aluminum strips using conventional microlitho-graphic techniques.ss The aluminum oxidizes spontane-ously in air and presents aluminum oxide at the sol id-vapor interface;the gold remains clean (except for weakll 'adsorbed contaminants f rom the labora tory a tmo-sphere). We select two adsorbates, L, and Lr, such thatthese tu 'o adsorbates adsorb strongly and select ively on

(54) Laib in is, P. E. ; Hickman. J. J . ; \ \ ' r ighton, N{.l \ { . Science ( l l 'ashington, DC) 1989, 245, 845-&?.

(55) Xi t t lesen, G. P. ; \ \ 'h i re, H. S. ; \ \ ' r ighron, M.Soc. 1984, 106, 7389-?396.

S.; \\ 'hitesides, G.

S. J. Am. Chem.

L 1 = C I 1 C H 2 r r I S H

t 2 = C F 3 1 C F 2 s C O O H

Figure l l . Schematic i i iusira ' . jon c, f ' .he s imultaneous forma't ion of two independent (orthogonal , sel f -assembled monolay'-ers on gold and alumina surfaces bv exposure to a solution con-taining both a th io l and a carboxl ' l ic aciC.

gold and alumina. In the example given. Lr is an alkane-thiol and L, a fluorine-labeled alkanecarboxl'[c acid. Exposure of the substrate patterned u'ith gold and aluminum/alumina to a solut ion containing Lr and Lr resul ts in rep-licating the gold pattern with a self-assembled rnonoial'erd e r i v e d f r o m t h e a l k a n e t h i o l a n d t h e a l u r , i n u m , ralumina pattern with a self-assembled monolal'er derivedfrom the alkanecarboxyl ic acid.s Figure 12 suggesls rheselect iv i t l ' that can be observed in th is s) 'stem bv shou'-ing scanning Auger element maps for gold. a lurninum.sul fur . and f luor ine.

These s) 'stems are being extensively developed in ourlaborator l ' . and several combinat ions of substrates andadsorbates is now avai lable giv ing *orthogonal- sel f 'as 'sembled monolsS's1s.57 It is experimentally' straightfor-u'ard. using existing technology, to obtain features b1'microlithography having dimensions in the range 0.5-1.0 sm. Techniquesss'se presentl5' being developed (espe-

cialll' shadou' lithographl'm) promise to extend these tech-niques to features wi th dimensions of <100 A. Thus, b1'combining microlithography and self-assembly, we hopeeventually t,o be able to prepare interfaces in which u'ehave control over both the in-plane and perpendicular-to-plane composition and structure on scales of dimen-sions ranging from 2 to t00 A. \\Ie believe that theses1'stems u'i l l be useful in studf ing a wide range of phe-nomena in surface science and technologl' ' .

The Future

N{odern organic surface chemistry is at an early' stageof development. Several s1'stems are, hou'ever. now avail-able for studl'.

Sel f -assembled monolayers on metais and metaloxides are the best characterized and most control-1"61u.t2-36'56'6t They offer a range of materials that areflat and stable, and can be varied from highly orderedto disordered. Although preparation of most of these sys-tems is straightforward, it is unlikely that they wil l beused in applications requiring large areas of functional-ized interface.

(56)Allara, D. L.; Nuzzo, R. G. Langmuir 1985, t, 45-52 and refer'ences cited therein.

(57) Laibinis, P. E.; Folkers, J. P.; Whitpsides, G. M. Unpublishedresults.

(58) E-Beam and X-ray nanolithography Wilkinson, C. D. W.;Beau'mont, S. P. Springer Proc. Ph1's. 1986, /3, 36-50. Anderson, E. H.;Kern, D. P. ; Smith, H. l . Microelectron. Eng. 1987,6, 541-546. New'man, T. H.; Will iams, K. E.; Pease, R. F. W. J. Vac. Sci. Technol. B1967, 5, 88-91. Kuan, W.; Frank, C. W.; Fu, C. C.; Allee, D. R.; Mac-cagno, P. ; Pease, R. F. W. J. l lac Sci .Technol . B 1988,6,2274-22i9.

(59) Lithographl'by STlr{: Silver, R. M.; Ehrichs, E. E.;Delozal'rne,A. L. Appl . Phys. Let t . 1987, 5r , 24i-249. L in, C. W.; Fan, F. R. F. ;Bard, A. J. J . Electrochem. Soc.1987, 134,1038-1039. Cranston, D' H. ;Lin, C. W.; Bard. A. J. J. Electrochem. Soc. 198t, 135, ?85-786. McCord,M. A. ; Kern, D. P. ; Chang, T. H. P. J. Vac. Sci- Technol . B 1988, 6,187?- l 880 .

L 1 + L 2

The Langmuir Lectures

L1 L1 L1 L2 L2 L2

H*#A u A l

i s o o c t a n r

(60) Jones, E. T. T.; Chyan,Soc. 1987, 109, 5226-5228.

(61) Hickman, J. J . ; 7,ou, C. ;S. : Laib in is, P. E. ; Bain, C. D. ;1989 . i l t , i 27 t -?2 i2 .

O. M.; Wrighton, M. S. J. Am. Chem.

Ofer, D.; Han'e1', P. D.; Wrighton, M.Whitesides. G. M. J. Am. Chem. Soc.

The Lanpmu: , r Lec tu re s

SEM

AI

Au

0 1 0 2 0 3 0 4 0Dis tance (mic rons)

Figure 12. Scanning electron migograph (SEM) and scan-ning Auger element maps for an arral ' of four strips of gold(numbers 1, 3. 6, and E) and four of a luminum/alumina (num-bers 2,4.5, and ?) on a s i l icon ni t r ide substrate that u 'as exposedto a mixture of HS(CH2)rCl and CF.(CF2) 'CO2H in isooctane.The SEM and element maps are for the arra! r'iewed from above;the schematic of the device (the height of the strips is not draunto scale) is a side vie'*'.

Langmuir-Blodgett f i lms complement sel f -assem-bied monolal'ers. Although thel' are experimentall l ' morediff icult to u'ork u'ith than self-assembled monolal'ers(especially for highly ordered s1'stems, in which transferfrom the l iquid-air interface to a solid support runs therisk of introducing cracks and other t1'pes of defects intothe f i lm), thel 'are more easi l l 'appl ied to mult i layer s) 's-tems.

Funct ional ized polymer surfaces are. in general , d is-<irdered and complex. Large areas of these surfaces. are,however. readily available. N{oreover, because the under-l f ing pol l 'mer is permeable and deformable, these inter-faces display' a range of propert ies and phenomena onheating or mechanical deformation that cannot be observed

Langmuir , l /o1.6, No. 1, 1990 95

with rigid sil icon, glass, or metal substrates. The recon-struction of functionalized organic polymers is a fasci-nating and highly technologically relevant subject in itsown right.s2 It may also lead to new classes of materi-als. by introducing functional groups at the solid-l iquidinterface and subsequently diffusing these functionalgroups a feq' angstroms below the surface by heating orb-r' mechanical deformation.

\Vork to date has also established a range of tech-niques that seem immediately useful in studying organicinterfacial chemistry. Of the classical surface spectro-scopies, XPS is uniquely valuable in providing informa-tion about composition and about the location and char-acter of functional groups. Polarized infrared externalreflectance spectroscopy (PIERS) is exceptionally use-ful in the specialized task of characterizing the degree oforder in self-assembled monolayers containing methyl-ene chains. Opt ical e l l ipsometry and XPS provide com-plementarf information about thickness in monolayers.The ultimate usefulness of instrumental techniques instudf ing organic surfaces remains circumscribed by theintrinsic characteristics of these materials. Organic inter-faces are usuall l 'not sufficiently u'ell-ordered to providehi gh-q ualitf informati on b1' scattering techniques; organicmaterials damage readily on exposure to high-energ-v radi-at ion or part ic les and ma5 contaminate vacuum sys-tems. Organic materials are often electrical insulators,and surface charging can make experiments diff icult; thetechniques of vacuum physics are usually not applicableto the condensed interphases (both solid-l iquid and sol-id-solid) of primary interest in organic surface chemis-tr) '.

81' contrast, contact angle measurements provide a con-venient probe of the structure of the interface. The uti l-i t1 ' of th js method is based on i ts high sensi t iv i ty to localsurface structure and on its abil ity to detect certain typesof react i ons-especiall l ' t hose involving ionization of func-t ional gruups-that are u 'e l l understood in ph1's ical-or-ganic chemistr l ' . Contact angle measurements thus con-st i tu ie a f i rst step in using react iv i ty as a probe of sur-face siructure. \ \ 'et t ing has an addi t ional advantage asa probe of surfaces that it is instrumentally straightfor-u 'ard and inexpensive:as i t is , thus, accessible to scien-tists and technologists who might not have the financialor emotional resources to invest in the full panoply ofhigh-r'acuum surface spectrometry.

Aithough uncertainties remain concerning many aspectsof the organic surfaces now available, they are suffi-cientl l ' well defined that they can be used to discoverand studl 'new and complex phenomena. Adhesion andtribologl' are immediately accessible opportunities.62 Thecondensation of thin l iquid fi lms on polar interfacesGs isa subject of broad interest in a range of technologiesil(especially when the condensing l iquid is water) and isalso relevant to the extensive body of information devel-oping from applications of the surface force balances and

(62) Hintermann, H. E. J, Vac. Sci. TechnoL B 198{,2, 816-822. Mas-signon, D. J. Chim. Ph1,s. , Phys.-Chim. Bio l . 1987,84, 135-140.

(63) Der jagin, B. V. ; Churaev, N. V. in Flu id Inter facia l Phenome-non; Croxton, C. A., Ed.; \\ ' i ley: Chichester, 1986; Chapter 15. Beys-ens, I ) . ; Knobler , C. M. Phl 's . Reu. Let t . l986, 57, 1433-1436. Kuroda,Y.; Ki t taka, S. ; I r1 iura, K. ; Mor imotn,T. Langmuir 1988, 4,2L0-215and references therein.

(64) Sidemna, S. ; Moalem Maron, D. ln Eoi l ing Phenomeno; vanStralen. S. , Cole, R. , Eds. : Hemisphere: Washington, DC, l9?9; Vol . 2,pp 903-922. Huang, P. H. Sens. Actuators 1988, 13,329-337.

(65) lsraelachvi l i . J . N. ; Tabor, A. Proc. R. Soc. London, A 1982,33/ ,19. Israelachvi[,J. N. J.Colloid Interface Sci. 1986, Lm,263. Israelachvili,J . N. ; Chr istenson, H. K. Ph1' .s ico A 1986, 140,2i8.

2 3 4 5 6 7 8r t t t t t l

It{ E!lF! Fnr !'IilF#ii";iF : - ' T . . Y Ex l g r I I IE ' F J '

E* lr,i G,;.l l l t t t r

F

Si

S

96 Langmuir , \ /o1.6, No. l , 1994

atomic force microscope.66 The application of the pre-parative techniques and design criteria alreadl' availableto the preparation of new materials s1'stems displayingnew types of properties is just beginning.

These organic surface systems should also provide excel-lent substrates with q'hich to studl'certain phenomenain basic science.r2'l3 Wetting is providing an outstand-ingly useful probe with which to study the energetics ofthe solid-l iquid interface and to explore the kinetics andthermodlnamics of drop spreading.?-e The ability to con-trol organic fi lm thickness at the level of angstroms pro-vides, for the first t ime, the capabil ity of making ver)'thin hydrocarbon barriers with very high dimensiona-l con-trol for use in studying electron transfer.ls It should bepossible to use these s5'stems in electrochemistry to studl'the fundamentals of electron transport at interfaces andin the preparation of new t1'pes of microelectrochemicaldevices and electrochemicall5' based sensors. \\ ' i th fur-ther development of synthet ic techniques, i t should be

(66) I r {ar t in, Y. ; S ' i l l iams, G. C. ; \ f ickramasinghe, H. K. J. .4ppl .Ph- rs . 1987 , 61 ,4 i23 -4 i29 . Mar t i . O . ; R ib i , H . O . ; D rake . B . ;A lb rech r ,T. R. ; Quate, C. F. ; Hansma, P. K. Science ( tLtashington, DC) 1988.239,50-52. Mart i , O. ; Drake, B. ; Gould, S. ; Hansma, P. K. J. l /oc. Sci .Technol . , A 1988. 6.28i-290. Schneir , J . ;Mart i , O. ;Remmers, G.; Gla-ser, D. ; Sonnenfeld, R. ; Drake, B. ; Hansma. P. K. ; El lngs, \ ' . J . I 'oc.Sci . Technol . . /4 1988, 6, 2E3-286. Gould, S. ; Mart i , O. ; Drake, B. : Hel le-mans, L. ; Bracker, C. E. ; Hansma, P. K. ; Keder, N. L. ; Eddl ' , N{. N{. :Stucky, G. D. Narure 1988, 332,332-334. Burnham, N. A. ; Col ton. R.J. J. Vac. Sci. Technol., .4 19E9, 7, 2906-2913.

The Langmuir Lectures

possible to attach macromolecules to SAIMs and to beginto use them to model certain biological surfaces. In themore distant future, it may'be practical to build cavitieswith defined dimension into self-assembled monolal'ersand perhaps to use them for shape-selective recognitionand catah'sis.

Acknorx'ledgment. This work rests on the efforts ofa substantial group of outstanding co-workers. The self-assembled monolayers were developed by Colin Bain andBarry Troughton. PE-CO2H and its derivatives were stud-ied b5'Randy'Holmes-Far ley, J im Rasmussen, and GregFerguson. Mark Wilson carried out the experiments withthe anthranilamide of PE-CO2H. Mark Wrighton andJames Hickman (MIT) have been invaluable collabora'tors in a number of projects involving self-assembled monola1'ers. and Ralph Nuzzo (AT&T Bell Laboratories) hasbeen an indispensable contributor to our understandingof monolal'ers and to their characterization (especiall l 'w i th in f ra red techn iques) . The work has been sup-ported in part bv ONR and b1'DARPA (through the Uni-lers i t l ' Research Ini t iat ive). The Harvard I r {RL (sup-ported bv \SFt provided instrumental and research sup-port , and \SF has aiso contr ibuted to research directedtor i 'ard understanding the coordinat ion chemistry at sur 'faces indispensable to the design of sel f 'assembled mono-iavers .

II