ORGANIC CHEMISTRY IN TWODIMENSIONS : SU RFACE-FU NCTIONALIZEDPOLYMERS AND SELF.ASSEMBLEDMONOLAYER FILMS1

George M. Whitesides and Gregory S. Ferguson,z Harvard University

Organic chemistry is largely derived from studies of the reactivity and prop-enies of moiecules in homogeneous solution. and much of the intuidon oforganic chemists is based on the behavior of molecules in solution. Surfacesand intertacesr (that is, quasi rwodimensional assemblies of molecules orfunctional groups) provide environmen$ that can be quite different from thoseof solutions. and chemical intuition derived from solution is often wrong whenapplied to processes occurring at surfaces. The centrai focus of our programin organic surface chemistry is on new science: that is. understanding andcontrolling the phenomena characteristic of surfaces. interfaces. and thinfilms. A charm of surface chemistry is. however. its ability to combine newscience with relevance to a wide range of technological problems.''s and wehope to contribute to these appiied areas as well.6

Underlying our program in surface chemistry is a broad interest in the prop-erties of organic surfaces as components of materials. In panicular. we hopeto develop the abiiity to ration alize and predict the macroscoprc propeniesof surtaces-wening, adhesion, friction-by knowing their microscopic, mo-lecular-level structures. The issue of stmcture/properry relationships in solidslies at the base of much of the current research in areas such as matenaisscience. condensed maner and device physics, and polymer physical chem-istry. Surface science spans these fields and is curently a research area ofparticularly great activitv.s'7-6 The appeal of surface chemistry as an avenueinto detailed understanding of the relations between microscopic and mac-roscopic properties of maner is that interfaces are more accessible to analysisand more easily modified by synthesis $att are the interiors of solids.

Organic chemistry has played a surprisingly small role in interfacial science.tAlthough organic chemistry offers, in principal, the ability to introduce a widerange of functional and structural groups into surfaces, in practice it has beendifrcult to rationalize, much less design and synthesize. ordered two-dimen-sional arrays of organic moieties.a We have taken a physical-organic approachto the study of organic interfacial chemistlv: We formulare a hypothesis re-

CF{EMTRACTS-ORGANIC CHEMISTRY 1: 171- 1 87 ( I 988 )0895-444518E/$.m O 1986 Chemtracts

iTAY/JUNE It'88 / SURFACEfUilCTTONALIZED PC'LYTERS ANO SELFASIIETBI-ED FIL'IS 171

lating molectrlar-scale stnrctlre to macroscopic ProPerty' synthuize and char-

orrrirc interfaca having structures aPProPriate to testing that hypothesis,

measure the'propenies it interest, and iruerpret the information concerning

strucrure *a pioperties in terms of the original hypothesis. The physical-

organic paradigFl ior the study of complex pailerns of strucnrre and reactiviry

is iundam"ntally a qualitadve one, often relying more on analogy than on

numerical calcuiations based on fundamental theory. It has, however, pro-

vided one of rhe most durable and useful methods of understanding compli-

cated systems. Physical-organic chemistry counts among its many successes

the correlation of organic structures with reactivities in solution, the ration-

alization of areas tu.l .t photochemistry and catalysis' and the inference of

the properties and stnrctures of reactive intermediates;D we believe it will

also be immensely valuable in understanding surfaces.

SYNTHESIS OF SURFACES AND INTERFACES

We have reiied on two seParate ryPes of experimental systems in our studies

(Scheme 1):

r-j

€

-

Self-Asscmbly

Oxidat ion

+

H e C r O lpE 4 -PE-CO2H"

1) PlasmaPE + -PE-.H"

2) BHe'

1. Slrfrct-Functirindbed Orgrnic Pollmcrs, Espcddly'Polyethylene Car'

boxylic Acidl (PE-CO2[I). Thesc s),stems are PrePared by .oxidizing poly-

ethytene (PE) films with chromic acid and using the carborylic acid groups

in6oduccd onto the surface as the staning point for more elaborate chemical

modification (Scheme 1).$s The chromic acid oxidation has the advantages

of restraining the funstionality to a very thin (less than 10 A in depth) layer

along the surfacc @ntour of the polyurer and of generating a set of func-

tionalities limited to carboxylic acids and ketones and/or aldehydes. PE-CO:H

is convenient to prepare and snrdy and is an excellent material for exploratory

snrdies. It also provides an entry into the examination of propenies rePre-

PC'LY'TERS AT0) SEU!TSSENI-ED FII.G / CHEII|TRACTS-ORGANIC CHEMISTRY

N=Au I HS-R-X

s-R-xJ-r'-r'

S, R-X

- Rr-x'

Si / SlOz ClrSl-R-X

Schcme 1. Scfiematic representation of the two mdtods used for production of functionalizEd surfaces' Left:

spontaneous self-assembly of an onented monolayer film by adsorption of organosulfur compounds on gold or

af'fyf ricfrbrosilanes on silicon/silicon dioxide. Rigtrc Oxidative funstionalizEtion of polydlylene ftlms.

172 SURFACE.FU}GNOil L.trED

sentadve of a "real" material. It is, however, a complex, microscopicallyheterogeneous and strucrurally illdefined material.rT2' self'Assembled Adsorbed Monolayer Films.l&p we and others have fo-ctsed on rwb classes of monollyersi organosulfur compounds (especiait), ;;-ganic thiols) adsorbed on gold,r5'{)-'rs and alkyl siloxane monolayen preparedby reaction of alkyl trichlorosilanes with surfaces containing hydroryl groupsand/or adsorbed water.'tsn Both of these systems, and otheis rilated to rhem,are excellent models for interprering the characterisrics of pE-CO2H and itsderivatives. Immersion of a silicon iuf.r coated wirh a thin film (:i000 i;of evaporated gold in a solution of a faqv thiol for t hour at room remperarureresults in the formation of a highly ordered, quasicrvstailine monolaver offatty thiol anached to tbe gold suriace by ruirui-gold coordi""',#i;l;r:1)' The essential Processes occurring during the adsorption and organzationof the thiol on the gold surface are stitl iniompleteli understood, but rhevare cenainly reiated to the familiar, if complex, coordination ;;;';?thiols and gold(O) or gold(I).sra

One of the most anracdve features of organic chemistry is the wide variarionin the strucure of organic molecules thaican be produced through synthesis.A challenge to our program in organic surface ihemistry has been ro bnngthese synthetic techniques to bear on two separare classes of problems insurface chemistry: first, the introduction of small fragments having desiredfunctionality onto surfaces through chemicai reacrion: second. the prepara-tion/assembly of these fragments in extended macroscopic arravs with controiover position and orientadon. The nvo approaches we have followed-oneleading to PE-CO2H and its derivatives. and the other to self-assembledorganic monolayers-are quite diferent. The former inroduces funcrionaigrouPs onto a preformed heterogeneous material (Scheme 2). This procedureis convenient and exPenmentati relevant to a broad range of polymer tech-nologies, but it requires the srudy and analysis of matenals rhat are intnnsicallvstructurallv ill-defined. The latter prepares well-defineo. approp";;.1:, fu;;:

PE-HII

III

I

CrO3/H2SO4

ROH<t---

H'

LiAtH4orBHg

-> PE-CH2OHPE-CO2B PE-C02H

l*,,II

PE-COCIJ rcoc'

PE-CH2OCOR

RoH // \

nnrxz/ \

PE-CO2R PE-CONHR

Sghcme 2 Representative reaction seguences used to convert the surfaceof polyethylene fitm (pE-H) to "polyethylene carborylic acid,,(pE-corH) andderivatives.

ttiAY/JUNE I98S / SURFACE-FTJilCNOilALJZED POLYTGRS Ar€ SEI.FAS{itr8LED FIL.TIS 17g

tionalized molecules, which are then allowed to self-assemble on a reactive

surface and form highly ordered two-dimensional stru$ures. Self-assembly

will, we beiieve, bec6me a mainstay of ordered monolayer formadon,"t and

will eventually prove invaluable in ratronal suategies for modification of the

propenies of interfaces. Prepanng these svstems is, however, expenmentally

roi. complex than generating functionalized polymer surfaces'

CHARACTERIZATION OF SURFACES

We have used rhe usuai array of specroscopic techniques to characterize

surtaces: ailenuared total reflectance-infrared (ATR-IR) and polarized in-

frared external reflective specrroscopy (PIERS), X-rav photoelectron sPec-

troscopy (xPS), elecrron spln resonance (ESR), fluorescence. electron

microscopy, and ellipsometry are all usefui (Table). We have, however, aiso

Table. selected Methods for Analysis of surtaccs and Intertaces

Technique Application and Depth Sensed

Scanning tunneling microscopy (STM)Low-angle X'raY scattenng

Electron microscopy (SEM, TEM);eiectron difiractton

Contact angle (HrO)X-ray photoelectron spectroscopy ;

(XPS); auger spectroscoPyEllipsometry

Attenuated total reflectance'infrareO I(ATH-|R) |

Polarized infrared extemal reflecttve )spectroscopy (PIERS) |

Rutherford backscattenng (RBS) I

Fluorescence spectroscopy

Electron spin resonance (ESR)spectroscopy

Individual atomic positions on surfacesElec-tron density map of the surface of

very flat solidsSurface morphology and degree of

crystalline orderPolanty of top -10 A'Atomic and chemical composition of top

-so ADetermination of film thickness with a

resolution of 2 A

Vibrational analysis of top -1000 A

Atomic composition as a function ofdepth with resolution of hunclredso f A

Assay for density of functionality aftercovalent attachment of fluorescentprooes

Location and mobility of paramagneticcenters (e.9., TMPO) in interfaces

been able to apply to problems in the physical-organic chemistry of surfaces

rwo techniqu"i ior cbaractenzation that are less familiar to the spectroscopiccommunity. Ttre first is the measurement and interpretation of liquid-soiidcontact angles. This technique has proven to be the most surface'sensitiveand most convenienr (if not the most easiiy interpreted) method that we have

available to charact eiue organic interfaces.!0-35'{r'4r It is especially useful in

characterizing the soli&water interface. The sccond technique involves stud-ies of chemicai reactivity at interfaces. This approach is especially useful when

applied using simple, high-yreld reactions that are well undentood in ho-mogeneous, liquid phase chemistry. Ionization and esterification of carboxylicasids and saponificltion of carboxylic acid esters are esPecially diagnostic.r

The combination of measurement of contact angle with studies of ionizationof functional groups has resulted in a technique we call "contact angle titra-tion":+hat is, study of the vanation in the contact angle with the pH of the

SURFACE-FUI{CflOilAUZED P(TLYIERS AXO SELFA$iEilBt€D FILilS / CHEMTRACTS-ORGANIC CHEMISTRY174

aqueoui drop (Scheme 3). This technique increases the information derivedfrom the measurement of contact angies. Traditional approaches to studyingcontact ange generate only nro.numben (the advancing conract angle 0"-anJthe receding contact angle g,)." Receding conracr angies are presendy vervdifficult to interpret. EEorts to characterize complex, heierogenious interfacesusing only advancing contact angles are unlikely to be verv broadly userul.By measunng contact angle as a function of pH, however, one can often inferthe existence, environment, and narure of ionizable groups present at theinterface.

Contact angle titration is based on the observation of variations in contactangle with pH at surfaces connining ionizable groups.rs This variadon plau-

eo

t20

90

60

30

ot?o

90

60

30

0

* O O O $ S o O o PE-CONHoPE.H

PE-COOCH3

PE-CH20H

IPE-CONHCHzCHzN)

\ PE[>C=OJ[CHzNH2J

PE.C OOH

o48t?pH

Sclrcmc 3. Dependence of 0. on pH for surface-functlonalized polyethylene film. Iop: Using unbufiered aqueoussofut tons.Buf iers: (D0.1 Mphosphatebuf ier ; (o) a i lo thers(0.05M),pH1,0.1 NHCI; gHz,male icac id;pH3,tartaric acid; pH 4, succinic acid; pH 5, acetic acid; pH 6, maleic acid: pH 7 and g, HEPES; pH g and 10, CHES,pH 11, Fiethylamine; pH 12, phosphate; pH 13,0.1 N NaOH. The crosshatctred and labeled "assorted bufiers"at pH 8 include data for phosphates MoPS, HEPES, TAps, TFlls, and triethanolamine.

ASSORTED BUFFERS

MAY/JUNE I9E8 / SURFACE.FIJi|CNOTAUZED PC'LYrER9 A}IO SELFASTiETBIID FILTS 175

sibly reflects ionization of the functional group: The charged form of an acid

or base is more rr'var"prrnic than the unclargla form. and the contact angle

with water is rower. Although this simpre expranation is fundamentally cor-

rect. and the technique is a ver.V useful one in companng acidities' its simplicity

hides a number of iomPlexides'rus

one interesting aspect of conuct angre tiration concerns the intuitive concept

of the ..quanuty" of a functional group Present at an interface' our inidal

inruition .on."Lirig ,rrrr".. functionai group chemistry was that' in most

sysrems likely r"i.i*Ji"d ."p"rimentatiy, th" number of functional groups

present on a rePresentative area of surface would be small compared with

the quantity of . r""g.nt Pr:s.elt in the volume of soiution used in expenments

on that s,rrface. Tffi Ueiiet is largely incorrect for measurements of contact

angles: The number of functional Eoups Present at high density on a surface

is comparable to that preseat in titutibni used for contact angle titration in

unbufiered systems. The difierence berween the titration curves obtained using

bufiered and unbuffered sorutions (Scheme 3) exempiifies the phenomenon.3z

Explanation of this observation helps to ciarif.v the concept of "concentration"

in a heterogeneous system consisung of a iurface *d: :ontaaing

liquid

phase. we .onsil; ,i. spreading of-an aqueous drop at an interface to be

determined in part by the extent -of

ionization of the functionality present at

that interface. Let us examine the "concenrration" of this functionaliqv in a

sysrem consistrng of a 1-p.l gtop in contact with a denvatized polvethvlene

surface (a l-p.l drop typically covers an area of -1 mm:)' The density of

functionat groups-"'" riti r*i".. can be in the order of 6 x 10r4/cm: for a

surface with typical roughness;33 at this density' the concentration of reagent

in solution in the contacting drop required to- react stoichiometricallv with

that funcrionatity is -0.1 m-M.rt For an unbufiered aqueous solution and a

monoprodc acidjbase reaction. a concentration of acid or base > 0' 1 mM

(i.e., pH < 4 or pH > 10) is thus require.d to.achieve a stoichiometnc reaction'

clearly, the difierence in contact angle duation.crrrves obtained using buffered

and unbuffered solutions is due ,o ,irf".. functionaliry that is itself sufficiently

concentrated in the system comprising surface and drop to bufier the pH of

the aqueous solurion in the ranie pH :-? Tlur. the qualitative idea that a .

monolayer of organic functiona-liry is insignificantly small in quantitv com-

pared with the fuictionaliry preseniin solution is incorrect, if one is concerned

with small volumes of soludon'

A second interesting issue concerns the detailed interpretadon of the contact

angle titration .o*Ir. h particular, we ask how should the solid-iiquid in-

terfacial free enerSy "Ysr be related to the functional groups present on the

surface? The fundamental relation connecdng the conracr angle to igterfacial

free energy tem$ is Young's equation (Eq' 1)'5t

Tsu

cos0 =

^lr s L

Jsv - 'Ysr

' lw( 1 )

176 ITRFACE.FUrCNOilALtrED POLYTERS AXO ITELFASSEMLED FII'IS / CHEMTRACTS{RGANIC CHEMISTFIY

For aqueous solutions constituted with appropriate buffers, the liquid-vaporinterfacial free enerily ̂ lw is the same as that for pure water. vanationJininterfacial free energies are thus related to the obsCrved value of g pr,,',*iivby the terrns 15v and "y51. These tenns, in turn, depend on a number of factors:&g tyP!, densiry, and distribution of funaional groups present at the solid-vaPor (liquid) interface; their extent of ionization; the roughness of the sur-face; tbe relative humidiry of the vapor.

As a fint approximation, we have proposed that the interfacial free energycan be expressed as a linear combinatibn of funcdonal group contnbutions.multiplied by the normalized fraction 9i of these groups on the surfacei2r4(Eq. 2). The parameters "yisr and 1,.5y refled intrinsic

(2a)

(2b)

hydrophiiiciry and group size or:rea.data with contact angles indicate that

Comparisons of infrared specroscopicthis gypg of analysis is approximatelv

=?=?

'Isr

"Isv

9i 1,se

9i lisv

Vo

o

S

oeI' a - A -

- r,lL911-tfl - v r r ! l E

! - A - v I

I I - + 7

: - - T I I II - - -

- I 4

_ _ _ I

?rs ry/sv

Figurc' scfiematic representation of an ideal 1np leftl and real (top right, bottomldrop of tiquid (L) in contactwith a solict (s) and vapor (v) with contact angle 0. The symbols inirre upper right picture represent (c) watermolecules' (A) dissolved solutes (phosphate, buner salts), io, e) polar surface groups (co2H, cor, c=o,. . .),(t) nonpolar surface grcups (CH;, 9H., .). t f,p of liquid (,bo6m; not drawn to scate), the ,.precu6or fitrn,,,extends microns beyond the edge of thi drop in ."rt",n orcumstances.

tI

MAY/JUNE l!)gE / SURFACE RJ[crloilAU"Fn POLYI|ERS atrto sELFAssEtBLED FtLfls 1TT

correct for PE-CO2H and some of its acidic derivatives,rs but that interactions

berween groups, ana perrraps interfacial heterogeneity, make the problem

more complex than can be discribed completely Jr*g tiris simple approach'35

Rn unaerltanding of these interactions and complexities remains to be es-

tablished.

Ath i rd important issueis themeaningofhysteres is in themeasurementofconuct

"ngl"r- For derivatines of PE-CO'H, 0" aPPear-s to provide a simple'

semiquantitadvely interpretable measure of interfacial group character and

density. connct angles are, however, a simple Parameter derived from ob-

servations of a complex realiry (Figure)' Advancing and receding contacl

angles difier on many surfaces, anOltt the derivatives of PE-CO2H (partic-

utarly polar derivatives) display very large hystereses in their contact ansies:

0, is frequently 0 even for'syste*i tt"t'ing fairly large vaiues of 0"' Large

hysteresis is usually interpr.,id -,o indicats a heterogeneous system far from

thermodynu.i. equilibrium.se Yet analyses of 0" based on Young's equation'

- .qu"rion *ru,n]ng th;odynamic equilibrium, seem to give interpretable

and reasonable results. It is not crear how one shourd rreat a system that is

nor at thermodynamic equilibnum. but for which physical measurements cor-

relate with those expeded based on physical-organic analogies to Processes

occurring at equilibrium in solution'

RESULTS

Both functionatized polvethvlene and its derivatives, and self-assembled mon'

olayer fiims. provide systems with which to examine reactions occumng at

interfaces and to test hypotheses concerning stnrcture/reactivit;- and struc-

rure/properry relationships. In so doing' we find that many of the results we

obtain can be rationaiized by anaiogy to phenomena in solution (often with

cbaracteristic difierences thai can uJinterpreted to comPare and contrast the

environments provided by homoseneous solutions and interfaces)' we also

frequently encoun,., un.ipectedlh.no*.na, which suggest that any models

of organic reactivity at interfaces. based exclusively on analogies with solution'

are not complete. The studies that follow provide examples'

Surface Acidities. Scheme 3 indicates that carboxylic acids and many' but

not all, amines show inflections in plots of 0, vs the pH of the drop used in

measuring the conract angie.s?r5 Asiuming that the midpoint of the inflectton

.orr.rporids to half-ionizltion of the functional group (an assumption sup-

ported by independent ATR-IR measurem.nts oJ.arboxylic acid surfaces),31

we infer that acidiries of functional groups at an interface and in solution are

very difierent. For example, the value of pH lol " solution in connct with a

surface required to achiive half-ionizarion of the carboxylic aod grouPs at

that surfari ott be as high as 12. What is the oriFn of this very large aPParent

decrease in acidity of &rUorylic acids (and con"sponding increase in the

apparent acidiry of ammoniurn-ions)? We believe thai the origin of these shifts

can ultimately be anributed to the locally low dielectric constant at the Poly-

ethylene-water interface,x{-CI but rationeli-a3isn of these anomalous values

of p& is not Yet comPlete.

Relstions between Func{ional Group Hydrophilicity and Wettebility of In'

terfaces. We assumed at the outset of our studies that more hydrophilic

interfacial groups (as measured by some convenient Parameter such as the

Hansch ,r, i"o*eiet',) would lead to more wettable surfaces. In fact, ex-

perimental observations reiadng wenability to functional group hydrophiiiciry

SUFFACE.FUI{CTIOIIAIJZED POLYTERS AI{D SEI-FASSETELED RtrIS / CHEMTRACTS{RGANIC CHEMISTRY178

difier significantly from those expected (Scheme a). As the value of n for thefunctional group on the surface decreases. 0o also decreases, but only up toa point. Beyond thar point, further increases in functionai group hydrophilicityresult in no further increase in wettability: that is. the hydrophilicirv of thesurface "sarurates."35 We postulate that the origin of this efiect lies in con-densation of water vapor at polar solid-vapor interfaces (Scheme 5). Nonpolarinterfaces condense relatively little water. All of our experiments invoivingcontad angJes with water nre carried out at LAUVo relarive humidiry in orderto Elssure that the system is as close to therrnodynamic equilibrium as possible.Polar functional groups at interfaces are undoubtedly associated with hy-drating warer adsorbed from the vapor phase. We postulate that. beyond acertain vaiue of the Hansch tr pzuameter. the polar surface functional groupsbecome completely surrounded by condensed. hydrating water, producing asolid-vapor interface whose polarity is essentially independent of the under-lying functionai group. Under these circumstances. the wettabilitv of thesurface is determined primarily by the area fraction of the surface convertedto polar funcdonaliry, and then hydrated by condensed water.

160

120

/ro|t t ' ?-o-

ooo

o o 6o

' 2 n '1

Schcme 4. Contact angles of water for derwatves of PE-CO.H, PE-R, witha range of fryctrophilicities of the trc,up R. lr is the Hansch parameter, a me€lsureof functional group hydrophilioty, derived trom the equilibrium constant forpartioning between aqueous and hyclrocarbon phases (rnset).

These observations and interpretations imply the existence of a thin. con-densed water film on polar surfaces. The nature of this film, and especiallythe relAtion of its stnrcnrre to that of bulk water, remains an important and

complex problem.

The Range of Interactions Determining Wetting. Scheme 4 displays an aston-ishing observation: Although a surface incorporating amides (PE'CONH:) is

relatively hydrophilic, the analogous primary amide PE-CONHC3H' is morehydrophobic than unfunctionalized polyethylene. Some of the apparent hy-

9Oo

-3. 4

\ (

MceHrzI t t {x ro -cxro8c ,r,?. /

/ /

?rr/ t o iC O N H C a H , . ^ \ /

lo /o 05, /PE-Hc - / , . \ o \ //o^\ 6 /3(:Ar/r**

II/TAY/JUNE 1988 I SURFACE.FI,NCTIOTAU:ZED FOLYTERS ANO SELFASSEilBLED FIL.TIS 179

oHz

Hro

P

Scheme 5. Schematic illustration of the degree of hydration of a functional

group p at the solid+rapor interface. When P is nonpolar, the equilibnum lies

io tne bft; when P is polar, it lies to the righl

drophobiciry of PE-CONHCTH7 and its analogs undoubtedly reflects the mt-

croscopic roughness of the surface of these. materials (generated during the

oxidanve surfice functionalization). Nonetheless. we find that it takes oniy

a small hydrophiiic or hydrophobic group to determine the wenability of a

surface. Furthermore, a smail hydrophobic group is capable of completelv

masking an underlying, intnnsically hydrophilic core functionaliry' Thus. for

example, replacement of a terminal CHi grouP in one of the well-defined.

seif-assemblid monolayer systems by a CH2OH group changes the monolaver

from being very hydrophobic to very hydrophilic,'r and reacvlation of the

terminal n-ydroiyt (CHTOCOR) once again makes it very hydrophobic- The

interactions that determine macroscopic wettability are, aPParently, verv short

in range.r0We believe, in fact, that measurement of contact angie is the most

surface-sensitive rechnique presently available for examining the sclid-liquid

interface. The great advantages of wening as a probe of surface stnrcture

(reladve, for eiample, to XPS) are that its measurement is very simple,

convenient, and ineipensive, and that it is inrinsically appiicable to the soli#

iiquid interface and to heterogeneous, noncrystalline surfaces. Its disadvan-

mges are that contact angle measurements are information Poor, that they

require a liqui&solid interface, and that their physical basis is complex and

still incompletely understood.

Designed Interfaces. The materials PE-CO-X are convenient but heteroge-

neous. The best characterized and strucnrrally best defined organic interfaces

now available are rhose formed by adsorbing long-chain alkvl thiols on gold,

or by atlowing long-chain alkyl trichlorosilanes to react with surface hydroxyl

groups and adsorbed water prescnt on the surface of glass or silica. Both of

these systems have the alkyl groups in completely rraru'extended confor-

mationi, provided that the terminal functional group is relatively srnall. For

organic ttriots on gold, the chains are tilted -30o from the normal to the metal

surfacc;€'rs for attcyl siloxanes on silicon/silicon dioxide, they are aPProxr-

mately perpendicoi"t to the subsrate surface (Scheme 6).e Transmissionelectron microscopy indicates that the thiol/gold system has at least micro-

srystralline order in the plane of the monolayer.azThese ordered monolayer systems permit an exquisite degree of control over

sttucture and dimensionality at the interface. As one example, consider a

monolayer formed by adsorption of HS(CH?),9OH on gold. Formation of

such a monolayer is experimintally very straightforward: one simply dips thegold-coated substrate into a solution of the c,,rrr-thioalcohol in a solvent such

as acetonitrile for t hour at room temperature, withdraws it, and washes it

briefly. At the conclusion of this procedure, the entire accessible surface of

suRFAcE-FuilcnoilALeED FoLynERs Ar{o SELFASSETELED FrLrs r CHEMTRACTS-ORGAN|C cHEMlsrFY180

it-IIIIII

LrI

IIL

OH'"OH

) )

) )

) )

) )a{-l'-o'l't lt l

tsl-o,sl

(o'''ro'lr

( (

a((aaa

Scheme 6. Schematic illustration of conformation and packing order in mon-olayers of organic thiols on gold and alkyl siloxanes on silicon/silicon dioxide.The monolayer is composed of three important regions: the head groups(portion binding to solid substrate), the polymethytene chains (for formation ofvan der Waals surface), and the tail groups (termrnal functronality that deter-mines the character of the soli#liquid and solid-vapor interfaces).

$!1c 7' Stylized illustrauons of monolayer structures.'r Proposed structures of (A) pure monotayer ofHs(cHJ'eoH; (B) monolayer composed of 50o/o HS(CHJ,eOH and 50o/o HS(CHJ,,OH; (C) pure monolayer ofHS(CHJ'OH' Struaures we believe do not occur in the systems studied here: (D) disordered monolayer and (E)monolayer containing a mixture of components and strowing phase separation into islands.

variabletunctionality

polymethylenechains

suPPort-organicinterface

\ H S H SAu Au Au

O H

looHtf,,r,, l l ,J-.I U '::"'SH

ITIAY/JUNE 1988 / SURFACEfiJrcnor r r?FF FoLyrcns Al|O SETFASSEilBLED FtLrS 1g1

the gold is covered with a uniform monolayer?3 A thick, and the exposed

surface is a densely packed monolayer of hydroryl groups. The thickness of

the monolayer is "aiity

controlled at the scale of angstroms by varying the

number of methylene groups in the thiol chain; the surface ProPerties ate

independently controllable through variations in the terminal functionai

group. If mixtures of rwo difierent terninally funcdonalized thiols are used'

monolayers can be made having the rwo mixed on the surface (Scheme 7)'

CURRENT PROBLEMS

The physicakrganic chemistry of surfaces promises to provide new materiais

based on rationil synthetic modificadon of surfaces and interfaces' new an-

alytical methods with which to characterize surfaces, and deeper levels of

understanding of familiar processes such as dissolution. wetting, adsorption.

and adhesiorioccurring arlnrerfaces and in solutions. The phenomena.being

obsewed are, however, usually more complex than those occurring in ho-

mogeneous solution, and are, consequently, still incompietely understood at

even the simplest levels. Tbe fieid presents a number of fascinating funda-

mentai problems in interfacial chemistry, among which we place the following:

1. Molecular-Level Order. How should the order in these svstems be de-

fined and measured? One advantage of a two-dimensional system is that

it is, in principle, less complex structurally than a three-dimensional

svstem: The components of a two-dimensional system are by definition

restricted to a plane rather than free to nanslate and rotate in three

dimensions. In practice. however, the problem of defining order in

surface -fu nction alize d po I ymers and seif- assem b le d monoi aye rs rem ains

very complex. All of ih.s. svstems are, in realiry, only quasi rwo-di-

mensionai. Matenals such as functionalized polyethylene are obviously

mictoscopically rough and heterogeneous and have functionaliry dis-

tributed nonuniformiy in a thin interfacial layer. Contact of these

systems with a liquid phase may result in interfacial swelling and

reconsrruction. Seli-assembled monolavers are bener defined strucrur-

ally, but even with these sysrems, subtle issues of order in the plane of

thi monolayer. ar the gold-monolayer and monolayer-liquid interfaces

and berween adja..niorganic molecules require the development of

new analytical techniques and new criteria for order.

2. Kinetics vs Thermodynamics. The extent to which any of the systems

currently studied :re ar thermodynamic equilibrium. and the influence

of deparnrres from equilibrium on their behavior, is almost completely' uncertain at Present.

3. Wetting. Despite interesting and provocative theoretical contributions

to the ih"ory'of wening inlenain idealized systems,e'6a-7t there is no

usefully detailed theory of wening relevant to real, misroscopically het-

.rog"n.ous surfaces. The current rationalizadon of hysteresis in the

measgrement of conracr angles is especially unsatisfaaory. Detaiied

examination of hysteresis, bJth theoretically and experimentally would

be particularly uieful, because hysteresis apPears to be very sensidve

to order; an understanding of the relation between interfacial stnrcture

and hysteresis might provide a new avenue of approach to this important

subject.4. IVlolecular Design of Monolayers. Essentially all work so far carried

out with self-asslmbled monolayers has focused on derivadves of fary

ltUFFACE-Ffrt{61OtrA!.trED FOLYUeRS At{O SELFASSETBLED RIJS t CHEMTRACTS-ORGANIC CHEMISTRY1&l

acids. These systems have the two virnres that they are easy to manip-ulate synthetically and that they do, for whatever reason, form *.it-ordered monolayers. They are, however. not stable at even modesdvelevated temperatures and have no strong intermolecular interactionscontributing to order in the plane of the monolayer or to thermal oroxidarive stability. It is important to develop molecrrlar stnrctures otherthan fattv acids that form ordered, stable rwodimensional sheet struc-nrres.

CONCLUSIONS

Organic chemistry at interfaces is a field offering major oppornrnities for borhthe conduct of basic science and the developmenr of new technology. It alsoprovides, through the synthesis of extended funcrionalized interfacej, a bridgebenveen the science of isolated molecules and the ssience and technology 6fmateriais. Since chemical reactivity and we$abiliw provide whar we beiievewiil prove to be invaluable probes of interfacial structure for organic svsrems.these s)'stems are particularly attractive for srudies aimed at understandinethe characteristics of soli&liquid interfaces.

Surface'funcdonaiized polymers (of which the best developed is PE-CO'H)are proving to be convenient svstems with which to conduct exploratorv work.They are easily prepared and manipulated. and because thev presenr solid-vaPor interfaces that have low surface free energies, thev are relativelv re-sistant to contamination by atmosphenc contaminanc. Funher. since theyare physically robust, surface-modified polvmers can be used to examinecomplex materials problems such as biocompadbilirv,D.s adhesion.srgas per-meation.s friction,s3 and the influence of bending, stretching,s and su*o..reconstructions on interfacial propenies.

Self-assembled monolayers will, we believe, prove to be the ultimare cor-nerstone of the basic science in organic surface chemisrry. They will cenainlyalso find technological application in areas such as promotion of adhesion,inhibition of corrosion, and control of friction, and they may prove importantin the production of sensors and microelectronic devices. The remarkableease with which very complex monolayer stnrcrures can be assembled frommolecules of very modest complexiry will be invaluable in studying the prop-ernes of organized molecular assemblies. The best defined of these suitemsis presently obtained by adsorption of orfunctionalized fatty thiols on gold.although organosilicon compounds on silicon dioxide and glass may ultimitelyprove equally ordered. Alkyl thiols on gold have as their major advantagethe comparibility of the thiol moiety with a wide range of organic functionalgrouPs, and the fact that these systerm lead to highly ordered monolayers.Silanes on silica are more economical, bener adapted to the formation ofmultilayer stnrctures, and more robust srnrctrrrally.

Given the astonishing sensitiviry of wenabiliry to local surface structure. itssrudy should provide a range of imponant new q?es of informadon aboutinterfaces, especially solid-liquid interfaces. Designing and interpreting theseexperirnents will require a physical-organic approach-the sysrems beingstudied are too complex to be defined using conventional, spectroscopy-basedphysical chemistry. Because werdng is directly relevant to a broad range oftechnological problems, these studies should be exceptionally valuable inapplicarions. The experimental techniques required to srudy wening are vervsimple. Surface science based on studies of wettabilir,v should rhus be acces-sible even to those without routine access to the instrumenmtion of high-

ir Y/JUNE l98E / suRFAcE-FuilctoNALuED poLyrEns Ailo sELFAssEilBLED Frlrs 1&l

ACKNOWLEDGMENTS

u, 4739 -$. Adams' N'K' Adv' Chem'

Chem. Ser', 1964, No'

16. Ishida.H. Rubber Chem'19tr7, 24, L.

PC'LYTERS AND SELHSSEilBLED

Scr., 1964, No. 43,S?;'-Adamson' A'W'' Ling' l ' Adv'

43. 57.

Tech. Rev.,Il|tnr 60, 497; Debe' M'K' Prog' Surl' Sci"

FIIJS / CHEMTRACTS-ORGANIC CHEMISTFY

vacuum physics. A more realistic theoretical basis for wening is desperately

needed. Current fieatments do not deal adequately with su.bjects such as

molecular-scaie heterogeneity, hysteress, . deviations from thermodynamic

equitibrium, *otl. na-rure 6i tnl interactions berween sorids and liquids.

The research carried out in our Eoup has been the.result of-skilled experi-

menrarion and ;;;G;rfrl and lreatiue design and interpretarion by a num-

ber of individuals, including Jim Rasmussen, Randy H-olmes-Fariey' Tom

McCarthy, Cotin il"in, n"rry Troughton'-Paul Laibinis' Hans Biebuyck' Lou

Suong, Stephen Wasserman, and l-luis M. Scarmoutzos' We are also grateful

toourco l leaguesPete rpen t ran(D iv i s ion : {npp l l :dSc ience ,Harvard ) ,Rarph Nuzzo (AT&T Bell Labor"rri.r), and Maiii wrighton (Mrr) for im-

pori"n, contributions to our understanding of surfaces.

REFERENCES AND NOTES

l . T h e w o r k d e s c r i b e d i n t h i s p a P e r w i r s s u p p o l l d l n p a r t b y t h e o 6 c e o f N a v a lResearch and the Defensc ndvlced Rescarch Projects Agency' and by the NSF

through g-nl. oruw (cT{E g54S702) and to the Harvard Materials Research

I-aboratory (DMR 86-14003)'

2. NIH Postdoctoral Fellow, 1988-1989'

3. The words ..sur{ace,. and ,.interface" have cleariy defined meanings oniy in the

contexl of an ideal system. a liquid and vapor in contact wtth a smooth' ho-

mogeneous solid. ln the real sysiems discussed here (espeoally functionalized

polyethylene films). these words should be understood as indicating the more

complicated norion of an "interphase reglon"' which incorporates the com-

piexines (roughness. inhomo;;;;,y. surfice reconstruction anC swelling' etc')

inevitab tY encountered'

4 . A r m s t r o n g , J . A . . W h i t e s i d e s ' G ' M ' C h e m ' E n g ' N e w 1 ' 1 9 t 6 ' ( ' T t R e s e a r c hBriefings 1986, Nauonal Academy of Screnci: Washington' DC' 1986: Proc'

Natl- Acad. Sci' USA' Lgtrl' U' 4ff'5'

5. Swalen. J.D., Allara. D.L., Andrade. J.D" chandross' E'A" Garoff ' s" lsrae'

l a c h v i l i . J . . M c C a r t t r y ' T . J . ' M u n a y . R . , P e a s e . R . F . . R a b o l t . J . F . . W v n n e .K'J" Yu' H' Langmutr u)itr7' 3' 932'

6. Holmes.Farley, S.R., Troughton. E.B., Whitesides. G.M., unpublished results.

? .Go lovchenko,J .A .Scrence,19E6,232,48 ;Tromp.R.M. ' -y3 ' : 'T .R.J . .Demuth .J . E . S c r c n c e , | ! | t 6 , 2 3 1 , 3 } a ; ' N o o n " o , J . R . . D a v i s . H . L . i b i d . , 3 1 0 : M a d e y ,T.E- ibid-, 316; Engel, T' ibid" 327;Celona' R'J" Pierce' D'T' ibid" 333;

comizzoli, R.8., Frankenthai. R.P.' Milner. P'c" Sinclair ' J'D' ibid" 340'

8. Somoriai, G. Ctumistry in Two Dimcnsions: surfaces' cornell universit'v Press:

Itbaca, New York, 1981; Gaines, G.L. Iruo[uble Monolayers ar Liquid'Gas

Irucrfaca; Intcrsciencc: Ncw York' 1966

g. Adanson, A.w. Physical ctumw of swfirca; John wiley & sons: New York'

L982.

10. Lsraelachvili, J. Intcrmoleanlar and Swfrce Forces; Academic: London' 1985'

11. Israelachvili. I. Proc. Natt- Acad Sci' U'S'A' 1!lt7' 84' {722; Israelachvili' J' 'Acc'

Chem- Rcs., 19F7 r 20, 4!5' and references therein'

12. Garofr,s. Proc. Natl. Acad- sci. u.s.A. l!tt7, E4,4729' and referenccs thcrein'

13.Rabolt ,J.F.MRsBul let in, ' ! f r I ,November16/Deccmber3l '39.

1 4 . D u b o i s , L . H . , Z c g a n k i ' B . R . , N u z z o , R . G . P r o c . N a t l . A c a d . S c i . U . 5 . A . | w '

184 SURFACFFUTCNOilALtrED

17. Dierkcr, S.B., Murray, c.A.. Lcgrange, J.D., schloner, N.E. chem- phys. Len.,l lrE1,137,453; Rabe, J.P., Swalen. J.D., Rabolt, J.F. /. Chem. phys., lgtrl,,86, L60l; Rabolt, J.F., Burns, F.C., Schloner, N.E., Swalen. J.D. J. Chem.Phys., 19t3,78,945: Rothberg, L., Higashi, G.S., Allara, D.L., Garofi, S.Chem. Phys. Len., l9ttrl, 133, 67.

18. Langmuir, I. J. Am. chem.Soc., 1917, J9, 1848; Blodgen, K.B. J. Am. chem.Soc. 1935,57, 1007; Blodgen, K.8., Langmuir,I. Phys. Rev. l!W,51,96{;for morc recent reviews of Langmuir-Blodgen s)rstems, see: Blinov, L.M.Rassan chcm. Rev. l9&i, 52, 773; Roberts. G.G. Adv. Phys. l9ES. 34, 475;Proceedings of the Second International Conference on l-angmuir-BlodgenFilms, in.Thin Solid Films, 19t5, vols. 1,32, 133, 134.

19. Sugi, M. /. Mol. Ehctrorucs, L9t5,1,3.20. Allara, D.L., lrfrrzrg, R.G., Langmuir 19E5, l, 45, 52.21. Hubbard, A.T. Acc. Chem. Res. 19t0. 13, LT|.22. Soriaga. M.P., Stickney, J.L., Hubbard, A.T. /. Mol. CataL 19E3, 21, 21I.23. Bard, A.J..l. Chem. Ed. 7-9l/F, &, 3A2.24. Faulkner, L.R. Chem. Eng. News llrK,62,28.7.5. Li, T.T.-T., Liu. H.Y., Weaver, M.J. J. Am. Chem. .Soc. 19E/' 106, L?33: Li.

T.T.-T., Weaver, M.J. /. Am. Chem. Soc. I-9E4, 106,6L07.26. Mori, K., Watanabe, A., Muroi, A., Nakamura, Y. I. Polymer Sci.: Pan A:

Polymer CIvm. tg?n,25, ?393.27. Fendler, J.H. Chem. Rev.,19'$7, 87,877; Ringsdorf. H., Schlarb.8.. Venzmer.

I. Angew. Chem. Int. Ed. Eng. L9ttrd,,27, II3.28. I-zhav, M., Addadi, L., Leiserowrtz. L. Proc. iYatl. Acad. Sci. U.S.A. L987,84,

4737, and rcferences therein.29. Lowry, T.H., fuchardson, K.S. Mechanbm and Theory in Organic Chemurm,

3rd ed., Harper and Row: New York, 1987.30. Rasmussen, J.R., Stedronsk],, E.R., 'Vhiresides, G.M. J. Am. Chem. Soc. L977,,

gg, 1736.

31. Rasmussen, J.R., Bergbreiter, D.E., Whiresides, G.M. J. Am. Chem. Soc. L977,99, 4746.

32. Holmes-Farley, S.R., Reamey, R.H., McCanhy, T.J., Deutch. J.. Whitesides,G.M. Langmurr 19E5. I,725.

33. Holmes-Farley, S.R., Whitesides, G.Iv[. Langmuir 19E6, 2, 266.34. Holmes-Farley, S.R., Whitesides, G.M. Langmuir l-9t7, 3,62.35. Holmes-Farley, S.R., Bain, C.D., Whjtesides, G.M. Langmuir, in press. 1988.36. For a similar procedure on polypropylene, see Lee, K.-W., McCanhy ,7.J. tVacro-

molecules UIEE, 21, 309.37. The surface region of PE-CO3H is rough on both the molecular and macroscopic

scales. It is an illdefined combinatron of carborylic acids. ketones and/oraidehydes, and methylene groups. Whether heterogeneity ensts on a largerscale of length, owing to uneven funcnonalization between crystdline andamorphous regions, is unknown. The surface regon swells to varying degrees,depending on the contacting solvent, and reconstnrcts on heating below themelting point.

38. Bigelow, W.C., Picken. D.L., Zisman, WA. J. Colloid Sci. 19<16, 1,5L3.39. Zisman, W.A. Adv. Chem. Scr. t3t64,, 43, L.40. Troughton,8.B., Bain, C.D., Whitesides, G.M., Nuzzo, R.G., Allara. D.L.,

Porter, M.D. Lsngrutir lllitt, 4,365.41. Bain, C.D.. Whitesides, G.M. Science U)tt, 24, 62.42. Strong, L., Whitesides, G.M. Langmuir, in press. 1988.43. Nu"o, R.G., Fusco, F., Allara, D.L. J. Am. Chem. Soc.,lW7,109,358; ]\{rrzze,

R.G., Allara. D.L. /. Am. Chem. Soc., 19&3, 105, ,148I.

44. f{"'os, R.G., Zegarski, 8.R., Dubois. L.H. /. Am. Chem. Soc. U)E7, 109, 733.

iiAY/JUNE 1I'88 / SURFACE.FIJI{CTIONAU:GD PC'LYIIERS AIIO SEI-FASSETEI-ED FILiIS 185

45. Am. Chem. Soc.

B.. Lundstroem, I.J. J. Colloi"d Interface

Langmuir l!W7 , 3, n9.

Am. Chem. Soc. TS186, 108,

Am. Chem. Soc., tn

monolavers. attributed to

to self-assemblv Processes'

46.

47 .48.

49.50.

51 .

52.

53.

60.

61.

65.6.

186 SURF^CE.FUTTCTIO}IALtrED PC'LYTERS ANO SELFATTSETALED FILIS / CHEMTRACTS-ORGAI'{IC CHEMISTRY

67. Kaelble. D.H. Phvsical Chemurm of Adhesion,' Wiley-Intersciencc: New York.1971, p. 170.

68. Gouid. R:F. , ed. Adv. Chem. 5rr. y)(4t,43.

69. Schwara, L.W., Garoff, S. Langmri 19t5. I, 219.70. Schwarrz. L.W.. Garoff. S. J. Coll. Int.Sci. 1985, 106, 4U,71. Cherry, B.W Polymer Surfaces, Cambridge Univenity Press: Cambridge, UK.

1981.72. Woodrufi, D.P. The Solid-Liquid [nterface; Cambridge Univeniry Press: Cam-

bridge, UK. 1973.73. Fones. M.A. J. Chem. Soc., Faradav Traru. I l!t92,78. L}L.74. Maiev, V.V., Gribanova, E.V. Dokl. Akad. Navk. SSSR 19t3, 2n, 413.75. Wolfram, E., Faust, R..Analq Univenitatie Scienriannr Budapest 19E0, 15i.76. Baszkin. A., Ter-Minassian-Saraga. L. J. Coll. Int. Sci. LyR,43, I9{J.7. Pomeau. Y., Vannimenus,.I. Coll. Int.Sci. 19t5, 104,477.78. Andrade, J.D.. Smith, L.M.. Gregonis. D.E. In Andradc. J.D., ed.. Surface and

Interfacial ,4specs of Biomedical Polymerc, vol. 1. Plenum: New York. 1985.79. Rarner. B.D. In "Biomaterials: lnterfaciai Phenomenoa and Appiicadons."

Cooper. S.L., Peppas, N.A., eds.. Advances in Chemisrry, 199, AmencanChemical Society: Washington DC. 1982. chap. 2. pp.9-8.

80. Durrani, A.A.. Chapman. D. In Feast. W.J.. Munro. H.S.. eds.. Polvmer Surfacesand Inrerfaces, Wiley: Chichester. UK. 1987. chap. 10: Lundstrom.I.. Ivarsson.B.. Jonsson. U.. Elwrng, H. Ib id. . chap. 11.

81. Wu. S. Polvmer Interface andAdhesion. Dekker: New York. 1982: Clark. D.T..Feast. W.J. Polvmer Surfacar,'Wiley: Chichester. UK I9T7: Zisman. W.A. InSkeist. I.. ed. , Handbook of Adhesive.s, Van Nostrand: New York.Ign. Chap.a1

82. Comvn. J.. ed.. Polvmer Permeabiliry, Elsevier: London. 1985.83. Bowden. F.P.. Tabor. D. The Ficrion and Lubncanon of Solids. Part II. Oxtord

Univeniry Press: London. L9&.84. Holmes-Farley, S.R.. Reamey, R.H.. McCanhy, T.J., Whiresides. G.M.. 'rn-

published results.

85. Scarmouuos. L.M.. Holmes-Farley, S.R.. McCanhv. T.J.. Whitesides. G.M..unpublished results.

MAY/JUNE 1988 ,' SURFACE.FUilCNOXALEED PCILYTERS AflD SELFAIISEilBLED FIL.iISi 187