journal article·2007-08-01

Molecular Self-Assembly at Bare Semiconductor Surfaces: Characterization of a Homologous Series of n-Alkanethiolate Monolayers on GaAs(001)

Christine L. McGuiness, Daniel Blasini, John P. Masejewski, Sundararajan Uppili, Orlando M. Cabarcos, Detlef Smilgies, David L. Allara

ACS Nano· Volume 1· Issue 1
Publisher page

Abstract

Structural trends for a homologous series of n-alkanethiolate self-assembled monolayers (SAMs), C(n)H(2n+1)S- with 12 < or = n < or = 19, on GaAs(001), studied by a combination of grazing incidence X-ray diffraction and infrared spectroscopy, along with ancillary probes, show an overall decay in organization with decreasing n, with the largest changes occurring below n = 15-16. The long-chain monolayers form a mosaic structure with < or =10 nm domains of molecules organized in an incommensurate pseudo-hcp arrangement with nearest neighbor distances of 4.70 and 5.02 A, a 21.2 A(2) area per chain, two chains per subcell in a herringbone packing with a chain tilt angle of 14 degrees , and preferential domain alignment along the substrate [110]([110]) step edge direction. In contrast, for n < 14 no evidence of translational ordering is seen and the alkyl chains exhibit a loss of conformational ordering and coverage relative to the n > 16 cases. A 4'-methyl-biphenyl-4-thiolate companion SAM shows evidence for ordered structures but with lattice parameters close to those expected for a structure commensurate with the intrinsic GaAs(001) square lattice. These trends are explained on the basis of competitions between lattice, interfacial, and intermolecular forces controlling the nanoscale structures of the SAMs. Overall these results provide an important aspect of understanding the effects of SAM formation on surface properties such as electronic and chemical passivation.
Source: openalex_inverted_index · Quality: baseline

Research context

OpenAlex topics: Molecular Junctions and Nanostructures · Quantum Dots Synthesis And Properties · Nanowire Synthesis and Applications

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

Departments of Chemistry and Materials Science, The Pennsylvania State University, 104 Chemistry Building, University Park, Pennsylvania 16801-6300

Department of Chemistry, Cornell University, Baker Laboratory, Ithaca, New York 14853-1301

Cornell High Energy Synchrotron Source, 200L Wilson Lab, Ithaca, New York 14853.

References (100+)

Show source-provided references
1. Ulman A. · Chem. Rev. · 1996 · 96 · 153310.1021/cr9502357 ↗
2. Love J. C. · Chem. Rev. · 2005 · 105 · 110310.1021/cr0300789 ↗
3. Allara D. L. · Langmuir · 1985 · 1 · 4510.1021/la00061a007 ↗
4. Schlotter N. E. · Chem. Phys. Lett. · 1986 · 132 · 9310.1016/0009-2614(86)80702-3 ↗
5. Brzoska J. B. · Langmuir · 1994 · 10 · 436710.1021/la00023a072 ↗
6. Parikh A. N. · J. Phys. Chem. · 1994 · 98 · 757710.1021/j100082a031 ↗
7. Allara D. L. · Langmuir · 1995 · 11 · 235710.1021/la00007a007 ↗
8. Nuzzo R. G. · J. Am. Chem. Soc. · 1987 · 109 · 73310.1021/ja00237a017 ↗
9. Nuzzo R. G. · J. Am. Chem. Soc. · 1987 · 109 · 235810.1021/ja00242a020 ↗
10. Laibinis P. E. · J. Am. Chem. Soc. · 1991 · 113 · 715210.1021/ja00019a011 ↗
11. Sheen C. W. · J. Am. Chem. Soc. · 1992 · 114 · 151410.1021/ja00030a076 ↗
12. McGuiness C. L. · J. Am. Chem. Soc. · 2006 · 128 · 523110.1021/ja058657d ↗
13. Porter M. D. · J. Am. Chem. Soc. · 1987 · 109 · 355910.1021/ja00246a011 ↗
14. Love J. C. · J. Am. Chem. Soc. · 2003 · 125 · 259710.1021/ja028692+ ↗
15. Li Z. Y. · Langmuir · 2003 · 19 · 674410.1021/la034245b ↗
16. Lim H. · Langmuir · 2004 · 20 · 74310.1021/la035404u ↗
17. Schvartzman M. · J. Vac. Sci. Technol., B · 2003 · 21 · 14810.1116/1.1532026 ↗
18. Yamamoto H. · Langmuir · 1999 · 15 · 864010.1021/la990467r ↗
19. Gu Y. · Langmuir · 1995 · 11 · 184910.1021/la00006a003 ↗
20. Tanzer T. A. · Appl. Phys. Lett. · 1999 · 75 · 279410.1063/1.125152 ↗
21. Dubois L. H. · J. Chem. Phys. · 1993 · 98 · 67810.1063/1.464613 ↗
22. Chidsey C. E. D. · J. Chem. Phys. · 1989 · 91 · 442110.1063/1.456776 ↗
23. Alves C. A. · J. Am. Chem. Soc. · 1992 · 114 · 122210.1021/ja00030a015 ↗
24. Poirier G. E. · Chem. Rev. · 1997 · 97 · 111710.1021/cr960074m ↗
25. Camillone N. · J. Chem. Phys. · 1993 · 98 · 350310.1063/1.464071 ↗
26. Fenter P. · Langmuir · 1991 · 7 · 201310.1021/la00058a008 ↗
27. Dhirani A. · Langmuir · 1995 · 11 · 260910.1021/la00007a046 ↗
28. Fenter P. · Phys. Rev. Lett. · 1993 · 70 · 244710.1103/PhysRevLett.70.2447 ↗
29. Samant M. G. · Langmuir · 1991 · 7 · 43710.1021/la00051a001 ↗
30. Camillone N. · J. Chem. Phys. · 1994 · 101 · 1103110.1063/1.467854 ↗
31. Poirier G. E. · Langmuir · 1994 · 10 · 285310.1021/la00021a001 ↗
32. Poirier G. E. · Langmuir · 1999 · 15 · 116710.1021/la981374x ↗
33. Tour J. M. · Acc. Chem. Res. · 2000 · 33 · 79110.1021/ar0000612 ↗
34. Delamarche E. · J. Phys. Chem. B · 1998 · 102 · 332410.1021/jp980556x ↗
35. Kreutz T. C. · Appl. Phys. Lett. · 2003 · 83 · 421110.1063/1.1625422 ↗
36. Razeghi M. · IEEE J. Sel. Top. Quant. Electron. · 2000 · 6 · 134410.1109/2944.902188 ↗
37. Sze S. M. · 1981
38. Tsukamoto S. · Appl. Phys. Lett. · 1994 · 65 · 219910.1063/1.112760 ↗
39. Sugiyama M. · Surf. Sci. · 1997 · 385 · L91110.1016/S0039-6028(97)00388-9 ↗
40. Sugiyama M. · Phys. Rev. B · 1994 · 50 · 490510.1103/PhysRevB.50.4905 ↗
41. Pelzel R. I. · Surf. Sci. · 2000 · 470 · L8110.1016/S0039-6028(00)00855-4 ↗
42. Oigawa H. · Jpn. J. Appl. Phys. 2. · 1989 · 28 · L34010.1143/JJAP.28.L340 ↗
43. Oigawa H. · Jpn. J. Appl. Phys. 2 · 1991 · 30 · L32210.1143/JJAP.30.L322 ↗
44. McGuiness C. L. · J. Phys. Chem. C. · 2007 · 111 · 422610.1021/jp065173a ↗
45. Jun Y. · Langmuir · 2006 · 22 · 362710.1021/la052473v ↗
46. Nesher G. · J. Phys. Chem. B · 2006 · 110 · 1436310.1021/jp062181i ↗
47. We note that an incorrect EAL (36Å) of the Ga 3d electrons through an alkanethiolate SAM was usedto calculate the absolute thickness of the alkanethiolate SAMs inthis work, in affect decreasing the calculated SAM thickness by ∼2Å from the real value. Furthermore, thicknesses were only calculatedfrom the attenuation of the Ga 3d signals without consideration ofthe As 3d attenuations.
48. Nesher G. · J. Am. Chem. Soc. · 2007 · 129 · 73410.1021/ja065399r ↗
49. Frey S. · Langmuir · 2001 · 17 · 240810.1021/la001540c ↗
50. Leung T. Y. B. · Surf. Sci. · 2000 · 458 · 3410.1016/S0039-6028(00)00385-X ↗
51. Adlkofer K. · J. Phys. Chem. B · 2003 · 107 · 1173710.1021/jp0356719 ↗
52. Adlkofer K. · J. Phys. Chem. B · 2003 · 107 · 58710.1021/jp0267813 ↗
53. Shaporenko A. · J. Phys. Chem. B · 2004 · 108 · 1796410.1021/jp040474a ↗
54. Shaporenko A. · Langmuir · 2003 · 19 · 499210.1021/la020909q ↗
55. Mars C. K. · 2000
56. Sullivan, P.M.; McGuiness, C.L.Unpublished results.
57. TheC18andC12monolayers have previously beenpeak fit using high resolution XPS, see ref44for detailed fits. Our fits closely match thosefound for the high resolution method.
58. Samples were transferred throughambient conditions within 1 min to the XPS analysis chamber. Thisis equates to the same amount of time the etched samples are exposedto ambient conditions before immersion in the alkanethiol solution.
59. Heister K. · Surf. Sci. · 2003 · 529 · 3610.1016/S0039-6028(03)00299-1 ↗
60. Biebuyck H. A. · Langmuir · 1994 · 10 · 182510.1021/la00018a034 ↗
61. Zharnikov M. · Langmuir · 2000 · 16 · 269710.1021/la991034r ↗
62. These FWHM and BE shifts have beenattributed to changes in polarizability of the alkyl chains, whichin turn changes their ability to screen the electrostatic interactionsbetween the substrate electrons and the C 1s core hole.(59-61)These effects are not reflective of a change in chemical state ofthe C atoms. In addition effects such as a decrease in the conformationaland orientational order of the molecules have been considered forshort alkyl chains.(59-61)Similar FWHM-chain length correlations have alsobeen observed for biphenyl alkyl molecules assembled on Ag and Au.(129)
63. The theoreticaldependence of the C 1s intensity as a function of SAM thickness isgiven by the equation:I⁢ = ∫0zdρoe−(z/λ)dzwherez= distance into thealkyl chain matrix from the S-C interface towards the vacuum interface,zd= the total thickness of the alkyl chain matrix,ρo= intrinsic C 1s photoelectron emission densityper unit thickness (dI/dz)owith no overlayer attenuation, and λ = C 1s photoelectronattenuation length for a fully dense alkyl chain matrix = 35.4 Å.130Expressed in terms of a ratio:In/I18⁢ = [∫0ndCρoe−n(dC/λ)d(ndC)][∫018dCρoe−n(dC/λ)d(ndC)]⁢ = [∫0ndCe−nα dn][∫018dCe−nαdn] = (e−nα −1)/(e−18α−1)wheredC= thicknessof the alkyl matrix per C atom, α =dC/λ, andn= number of C atoms in the alkylchain but treated as a continuous variable for the integration. Withthe experimental value of 23.8 (± 0.2) Å for theC18SAM thickness12and correcting for the1.94 Å S-atom thickness contribution,(87)dC= [(23.8 – 1.94)/18]= 1.21 Å and α = 0.0342 which givesIn/I18= (e−0.0342n−1)/(e−0.615−1) = 2.18(e−0.0342n−1)
64. If thedecrease in monolayer coverage were to result completely in a lowervolume density structure (e.g.,a void content of ∼13% in the case of theC12SAM) the corresponding value of λ would increase. Giventhe established relationship that λ ∝ (density)−1/2,131the corrections to the derived surface coverages would be minor,e.g., the void content estimated for theC12SAM would change from 13% to ∼14%.
65. Petrovykh D. Y. · Langmuir · 2006 · 22 · 257810.1021/la050928a ↗
66. The thicknessof the each monolayer, used for the IR modeling, was determined fromtheoretical thicknesses predicted by chain tilt angles determinedby NEXAFS(12)and scaled by the length ofeachn-alkanethiol.
67. To achievea fit within this range of tilt angles, monolayer thicknesses of 10–12Å were needed, which is significantly thinner than the thickness(∼14–16 Å) predicted from extended all-transchains with this range of chain tilts.
68. Parikh A. N. · J. Chem. Phys. · 1992 · 96 · 92710.1063/1.462847 ↗
69. Becausethe IRS spectra are the result of an average of all molecular environments,the assumption was made that 50% of the chains had one twist (Ψ= 43°) and 50% of the chains had a second twist, rotated 90°(Ψ = 133°) to the first. An all-transextendedchain was assumed. The specific values of the twist had consequentialeffects on the fit of the IRS spectra and specifically the ratio ofthe d−and d+peak intensities. ThusΨ = 43° and Ψ = 133° were unique answers tothe best fit simulation. Using this new model, the subtle differencein the ra−modes observed experimentally was also observed in the simulatedspectra; seeSupporting Information.
70. Additionally,the effect ofgauchedefects at the terminal methylgroup position on the fit of the 2 chain per unit cell model to ra−modes wereconsidered.(10)It was found thatintroducing up to ∼50% totalgauchedefects(including bothgauche(1) andgauche(2) defects) improved the simulated IRS spectra fit to the experimentaldata. However, unlike alkanethiolates on Au surfaces, the improvementto the IRS spectral fit is only asubtleeffect (seeSupporting Informationfor details).
71. Additionalin plane, sharp (Δqxy≈ 0.01 Å−1) Bragg peaks atqxy= 1.39 and 1.68 Å−1on theC18SAMs and atqxy= 1.29 Å−1on the MBT SAMs were also observed from time to time for differentmonolayer samples. At present their origin is unknown, but given thelarge difference in Δqxycompared to the main monolayer reflections, these peaks donot appear to be related to the main monolayer structure. Furthermore,these reflections were sometimes observed to have ring-like structurestypically associated with powder diffraction pattern from unorientedcrystallites. The degree of cleanliness of the samples may be thesource of these additional peaks. Most GIXRD experiments of self-assembledmonolayers have been conducted underinsituvacuum conditions where extensive sputtering, annealing, and heatingsteps are taken to ensure a clean surface. In contrast, no such stepswere taken for these experiments. In one study of alkaneselenolateson Au{111} where the additional cleaning steps were not taken, someof the same spurious peaks were also observed.(75)It has also been observed during these experiments thaton one bare native oxide sample that had been reused similar reflectionswere observed. It should be noted, however, that these peaks werenot originally present on a fresh, native oxide surface and resultfrom the monolayer formation process (seeSupportingInformation).
72. Full radialmaps along each direction are included inSupportingInformation.
73. The Scherrerequation,L= 0.89λ/BcosθBwhere λ is the X-ray wavelength, θBis the Bragg reflection of the monolayer, andBis the FWHM of θB, was used todetermined the domain size.
74. A rangeof domain sizes (90 to >1000 Å) have been observed for alkanethiolateson Au surface depending on the preparation conditions. Annealing thesurfaces leads to larger domain sizes. See, for example, refs25,28,29, and133.
75. Samant M. G. · Langmuir · 1992 · 8 · 161510.1021/la00042a020 ↗
76. Samant M. G. · Langmuir · 1993 · 9 · 108210.1021/la00028a033 ↗
77. Kaganer V. M. · J. Chem. Phys. · 1995 · 102 · 941210.1063/1.468809 ↗
78. Kaganer V. M. · Rev. Mod. Phys. · 1999 · 71 · 77910.1103/RevModPhys.71.779 ↗
79. Ocko B. M. · Phys. Rev. Lett · 2005 · 94
80. Tidswell I. M. · J. Chem. Phys. · 1991 · 95 · 285410.1063/1.460886 ↗
81. Xue Q. K. · Prog. Surf. Sci. · 1997 · 56 · 110.1016/S0079-6816(97)00033-6 ↗
82. Xue Q. K. · Thin Solid Films · 1996 · 282 · 61810.1016/0040-6090(96)08710-X ↗
83. Biegelsen D. K. · Phys. Rev. B · 1994 · 49 · 542410.1103/PhysRevB.49.5424 ↗
84. Wang X. · Surf. Interface Anal. · 1996 · 24 · 56410.1002/(SICI)1096-9918(19960916)24:9<564::AID-SIA148>3.0.CO;2-1 ↗
85. It is unlikelythat a lack of scattering centers due to a decreased chain lengthis the primary reason for no observation of monolayer reflectionsfor the C12SAMs. Other GIXRD chain lengths studies ofalkanethiols on Au {111} surfaces have observed diffraction from chainlengths as short as nine methylene units. See, for example, refs28and134.
86. Hecht E. · 2001
87. Shi J. · Chem. Phys. Lett. · 1995 · 246 · 9010.1016/0009-2614(95)01085-N ↗
88. Althoughthe molecular tilt angle can be directly calculated from the ratioof theIexp/Iisoof the Z modes, whereIexpis the experimentalintensity andIisois the intensity fromthe simulated isotropic film of the same thickness as the SAM, attemptsto use this method results in deviations in the molecular tilt angleof ±25°. This is most likely due to the fact that approximatedipole moment directions are used in this analysis and were not exclusivelyoriented along thex,y, orzdirection.
89. Krapchetov D. A. · Langmuir · 2005 · 21 · 588710.1021/la0503000 ↗
90. Nuzzo R. G. · J. Chem. Phys · 1990 · 93 · 76710.1063/1.459528 ↗
91. Ungar G. · J. Phys. Chem · 1983 · 87 · 68910.1021/j100227a032 ↗
92. Snyder R. G. · J. Mol. Spectrosc · 1961 · 7 · 11610.1016/0022-2852(61)90347-2 ↗
93. Kraack H. · Science · 2002 · 298 · 140410.1126/science.1078372 ↗
94. Kraack H. · Langmuir · 2004 · 20 · 537510.1021/la049977y ↗
95. Kraack H. · Langmuir · 2004 · 20 · 538610.1021/la0497954 ↗
96. Salem L. · J. Chem. Phys · 1962 · 37 · 210010.1063/1.1733431 ↗
97. Dubois L. H. · J. Am. Chem. Soc · 1990 · 112 · 57010.1021/ja00158a013 ↗
98. 2006
99. Voznyy O. · J. Phys. Chem. B · 2006 · 110 · 2361910.1021/jp064675l ↗
100. Vogel V. · J. Chem. Phys. · 1986 · 84 · 520010.1063/1.450635 ↗