829
Langmuir 1989,5,824-833
A Further LEED Study for the Surface Structure Designated Cu(100)-c( 2 x2)-N H.C.Zeng and K. A. R. Mitchell* Department of Chemistry, University of British Columbia, 2036 Main Mall, Vancouver, British Columbia, Canada V6T 1Y6 Received December 14,1988. I n Final Form: February 11,1989
A multiple-scatteringanalysis of LEED intensities has been made for the N on Cu(100) surface structure. Following indications from a recent study (Surf. Sci. 1987,188,599),the present work emphasizes models with N chemisorbing at half-monolayercoverage in the expected hollow sites on the metal surface but with consideration of relaxation effects, both vertical and lateral, in the metallic structure. The best correspondence in the comparison of experimental and calculated I(E) curves is reached with N held 0.06 A above the top copper layer, while the Cu atoms in the second layer directly below N are depressed by 0.09 A compared with the other atoms in that layer. The first-to-second Cu-Cu interlayer separation is 1.85 A. Further small lateral displacements (magnitude 0.14 A or less) consistent with the p4g diperiodic s ace group may occur, even though the required extra diffracted beams for a (2X2)-typeLEED pattern ave not been reported yet for this surface.
R
Introduction Detailed structural information for nitrogen chemisorbed on well-characterized surfaces of transition metals remains sparse so far, although for oxygen chemisorption appreciable adsorbate-induced relaxations and reconstructions have been identified.l-s A recent report by Heskett et al. pointed to the possibility of a N-induced reconstruction a t the (110) surface of copper, with the formation of a surface layer of copper nitride.' An earlier investigation of phonon spectra from the 42x2) surface formed by N chemisorbed on Cu(100) suggested a structure related to that of bulk Cu3N,5although a subsequent LEED analysis also noted some significant differences from this bulk structure.s For example, in Cu3N each N atom is surrounded octahedrally by six Cu atoms,' whereas a t the Cu(100) surface N apparently chemisorbs with essentially five coordination.6 To a first approximation, the Cu(100) surface presents a rigid array of adsorption sites to a chemisorbing N atom, although with chemisorption relaxations in the metallic structure could occur, particularly since the energy associated with the chemisorption interaction is large compared with that in the individual Cw-Cu interactions. Indeed, with thermodynamic data for bulk copper and copper nitride as a rough guide! the former can be estimated to exceed the latter by a factor of over 40. The previous LEED investigation of the Cu(100)-c(2X 2)-N surface structure reported that the N atoms incorporate deeply into the expected fourfold hollow (40sites to become closely coplanar with the topmost copper layer? Each N atom then bonds to four neighboring Cu atoms in the top copper layer and to one directly below in the second copper layer. The topmost Cu-Cu interlayer spacing was indicated to be expanded by about 8% from the bulk value; the possibility of further second-order re(1) Kramer, H.M.; Bauer, E. Surf. Sci. 1980,92, 53. (2) Bader, M.; Puschmann, A.; Ocal, C.; Haaee, J. Phys. Rev. Lett. 1986,57, 3273. (3) Zeng, H. C.; McFarlane, R. A.;Sodhi, R. N. S.;Mitchell, K. A. R. Can. J. Chem. 1988,66,2054. (4) Heskett, D.; Baddorf, A.; Plummer, E. W. Surf. Sci. 1988,195,94. (5) Franchy, R.;Wuttig, M.; Ibach, H. 2.Phys. B 1986, 64, 453. (6)Zang,H. C.; Sodhi, R. N. S.;Mitchell,K. A. R. Surf. Sci. 1987,188, 599. (7) Juza, R. 2.Anorg. Chem. 1941, 243,118. (8) Rossini, F. D.; Wagman, D. D.; Evans, W. H.; Levine, S.;Jaffe, I. Selected Values of Chemical Thermodynamic Properties; National Bureau of Standards: Washington, DC, 1952.
Table I. Optimized Structural Parametern (in A) model dN1 dl2 d22 dzs 4F1 4F2
0 0.06
1.98 1.85
0 0.09
1.81 1.76
laxations was noted, although that feature could not be explored in the previous study? Subsequently, we gained access to the Cray X-MP/22 computer at the University of Toronto to enable a further investigation of the details of the Cu(100)-c(2X2)-N surface structure, and the results obtained provide the basis for the present paper. This work uses the conventional procedures of LEED crystallography;specifically, intensity versus energy (I(@) curves from experiment are compared with those from multiple-scattering calculations in order to find the geometrical model in the calculations that gives the best correspondence with the experimental data?JO The basic types of models considered here are indicated in Figure 1. The experimental intensities used are based entirely on those measured in the previous study;s the data set comprises three beams designated (l,O), (l,l),and (1/ 2,1/2) at normal incidence and seven beams designated (o,o), ( o , ~(i,o) , (i,i), ~ / 2 , i p ) 0/2,1/2), , and (i/2,3/2) at an off-normal direction, where the polar angle is 15' and the azimuthal direction is [Ool]with respect to crystal ax-. The basic methods and parameters used for the multiple-scattering calculations" also follow those specified previously, except where some minor changes are specifically noted below.
Results and Discussion The previous study noted the possibility, when a halfmonolayer of N atoms sinks deeply into hollow 4f adsorption sites of the (100) surface of copper, that those second-layer Cu atoms directly below the N atoms may be pushed down toward the third Cu layer. Such a relaxation tendency would not apply to the other set of Cu atoms in the second layer, which are not below the N atoms. Discounting any lateral relaxations a t this stage, Figure 2 indicates the structural parameters needed to a depth of (9) Determination of Surface Structure by LEED; Marcus, P. M., Jona, F., Eds.; Plenum: New York, 1984. (10)Van Hove, M. A.; Weinberg, W. H.; Chan, C. M. Low-Energy Electron Diffraction;Springer-Verlag: Berlin, 1986. (11) Van Hove, M. A.; Tong,S. Y. Surface Crystallography by LEEO Springer-Verlag: Berlin, 1979.
0743-746318912405-0829$01.50/0 0 1989 American Chemical Society
830 Langmuir, Vol. 5, No. 3, 1989
e 4F1
Zeng and Mitchell
4F3
4F2
RS1
RS2
OCTl
KT2
Figure 1. Basic models considered in this work for the N on Cu(100) chemisorption structure. In each case, Cu atoms are represented by open circles and the N atom by a solid circle. Arrows indicate lateral displacements of magnitude A.
7 d d N 1
1st layer
d12
2nd layer
0
- f d * 2
0
p23
3rd layer
0
0
Figure 2. Interlayer spacings represented by d considered for. the 4F2 model of the Cu(100)-c(2x2)-N surface structure. The subscriptsidentify the layers involved, e.g., N, 1,2, ...,respectively, for the N layer, the fmt Cu layer, the second Cu layer, ...;all other Cu-Cu interlayer spacings are fixed at the bulk value (1.81 A).
three metal layers for defining such a system with 42x2) translational symmetry. Table I specifies values of these parameters found for the two related models designated 4Fl and 4F2; the first is constrained with dN1and dB equal to zero, while the second includes these parameters in the optimization. All other interlayer separations are fixed at the metallic bulk value of 1.81 A. The discussion in the previous paragraph assumes that with N close to coplanar with the topmost Cu layers the parameter dz2 shown in Figure 2 would be greater than zero. In the first instance, this was suggested by bond length considerations? but to provide an independent assessment a series of multiple-scattering calculations were made for normal incidence with dN1= 0.00 A, d12= 1.98 A, dzz = -0.24(0.04)-0,00 A, and da = 1.81 A. The R-factor analysis showed the correspondence of experimental and calculated I(E)curves became progressively less favorable as dz2 became more negative. Therefore, for the 4F2 model, the optimization over the full intensity data set was done only for positive values of d22. Specifically, the optimization was carried out iteratively over the following ranges of four parameters: dN1 = 0.00-0.12 A, d12 = 1.74-2.01 A, d22 = 0.00-0.24 A, and d23 1.75-1.89 A. Eleven iterations were made varying one parameter, or sometimes two parameters, while the others were fixed at
values indicated by the previous cycle. The optimal sets of parameters for the models 4F1 and 4F2 reported in Table I were found by minimizing the basic LEED R-fador first introduced by Zanazzi and JonaI2 but later modified by Van Hove and Koestner.13 Further parallel evaluations of the experimental-calculation comparison of I(E) curves were also made with the LEED R-factor proposed by Pendry;14 the final results were similar in each case, although the modified Zanazzi-Jona index tended to home more consistently to the fmal result. Basically thisR-factor was more sensitive to the structural changes; for example, from the first iteration with dN1and d12equal to zero to the final iteration reported in Table I for the 4F2 model, the Pendry R-fador reduced by only about 10% (from 0.41 to 0.37) whereas the modified Zanazzi-Jona R-factor reduced by twice as much in percentage terms (from 0.34 to 0.27). With the latter R-fador, the optimal value of the muffin tin zero constant (V0J changed from -6.0 eV after the first iteration to -7.2 eV after the 11th iteration. Figure 3 reports comparisons of I @ ) curves for the 10 measured beams with those calculated for the 4F1 and 4F2 models according to the geometrical parameters in Table I. The situation for the 4F1 model corresponds very closely to that covered in the earlier analysis (a slightly more precise value for d12could be given here since the incrementa in the spacing considered in these cslculations (often 0.02 A) were much smaller than the 0.15-Aincrement used previously). Generally, it can be said that the fractional-order beams from calculation give a good account of the measured I ( E ) curves and that the discrepancies are mainly for the integral beams. Certainly the extension from 4F1 to 4F2 improves the correspondence for the integral beams, according to both a visual analysis and the R-factor analyses, although the match is still not complete. At this stage, the multiple-scattering calculations for the model 4F2 were repeated for a range of polar angles of incidence (specificallyfrom 13" to 17O), but the best match between experiment and calculation remained for the previously reported value of 15': and therefore we had no evidence that the discrepancies in Figure 3 could be associated with this aspect of the measurement. We also calculated sets of N phase shifts (to 1 = 7) from superposition potentials independently of the set first used and provided by Moritz,15although this again led to no change in the overall level of correspondence shown in Figure 3. That suggests some refinement in the geometrical model may still be appropriate; in the following, the emphasis remains in considering changes in the ordered structural arrangements, although it is recognized that in principle disorder effects could also have a role.16 Five further basic types of geometrical models (Figure 1)are explicitly considered here for the normal incidence data, although to assess them most efficiently from the computational point of view they are first seen as an extension from the model 4F1 with dN1 and dz2both equal to zero, rather than from the more general model 4F2. Four of these models include lateral relaxations in the topmost layer to accommodate the deep penetration of N atoms into the 4f hollows of the Cu(100) surface. In the model termed 4F3, the topmost Cu atoms undergo a lateral displacement of magnitude A consistently with the p4g (12) Zanazzi,E.; Jona, F. Surf. Sci. 1977,62, 61. (13) Van Hove, M. A.; Koestner, R. J. In Determination of Surface
Structure by LEED, Marcus,P. M., Jona, F., E&.; Plenum: New York, 1984; p 357. (14) Pendry, J. B. J . Phys. C 1980,13,937. (15) Imbihl, R.; Behm, R. J.; Ertl, G.; Moritz, W. Surf. Sci. 1982, 123, 129. (16) Henzler, M. Appl. Phys. A 1984, 34,205.
Langmuir, Vol. 5, No. 3, 1989 831
LEED Study of C~(100)-~(2X2)-N
20
I
do
I
lh ' lb '
2bo'
'
I , \, , , , (-1 llbeam
e :15'
(-v2 Y2)
0+1s l F -&
do
'
1O l ' lb ' ENERGY (eV) LVJ
Figure 3. Ten experimentalI(E) curves measured for the Cu(100)-c(2X2)-N surface structure and compared with those calculated for the 4F1 and 4F2 models with the geometrical parameters reported in Table I. diperiodic space group as first recognized for the Ni(100)-(2X2)-C surface" and later found applicable also to the corresponding N surface.18 These models correspond to half-monolayer coverage, and although only c(2x2) LEED patterns have been reported for N chemisorbed on C U ( ~ O O )in, ~principle ~ the intensities for the other half-order beams, such as (1J2,O) and (1,1/2), could be undetectably low if A is insufficiently large. For the 4F3 model type, most emphasis was given to the particular structural form where, after the lateral displacement A, N adsorbs on the sites which maintain four equal bond lengths to neighbors in the topmost Cu layer. This seems consistent with the previous indications that the metallic structure needs to relax so as to give more room for N,B although another possibility has the N adsorbing on those other sites, derived from the regular 4f sites, which with non-zero A give two shorter bonds and two longer bonds to the topmost Cu 1ayer.l' This was considered for the normal incidence data with A = 0.05 and 0.10 A, dN1and dz both zero, and d, = 1.81(0.05)-2.16 A, but the matching of the calculated I(E) curves to those from experiment turned out to be slightly less favorable than for those calculated for A equal to zero. (17) Onuferko, J. H.; Woodruff, D.P.; Holland, B. W. Surf. Sci. 1979, 87, 367. (18) D a m , W.; Lehwald, 5.;Ibach, H. Surf. Sci. 1986,178, 528. (19) Lee, R. N.; Farneworth, H. E. Surf. Sci. 1966,3,461.
In the models designated RS1 and RS2, the N sublayer plus the topmost Cu layer in 4F1 undergo a rigid registry shift in the [Oll] and [OOl] directions, respectively; they keep the local NCu4 bonding unchanged from that in the 4F1 model, while allowing some increase in the length of the N-Cu bond to the second layer. For the two further models designated OCTl and OCT2, an additional layer of Cu atoms is effectively superimposed on the top of the 4F1 and 4F3 models, respectively, 80 that the N atoms are located in sixfold sites (to be similar to the octahedral coordination applicable in bulk Cu3N). The ranges of geometrical parameters considered in this analysis for the model types 4F3, RS1, RS2,OCT1, and OCT2 are detailed in Table 11. Some particular comparisons of I(E) curves calculated for these further models are shown in Figure 4. Overall, it is found that the models OCTl and OCT2 match less favorably to the experimental I(E) curves, compared with the situation for 4F1, although the models 4F3, RS1, and RS2 can still give reasonable agreements provided A is not too large. Indeed, for the models RS1 and RS2, the calculated I(E) curves are essentially identical at normal incidence for given A in the range up to 0.30 A. In general, values of A greater than 0.10 A affect the calculated curves, especially for the integral beams, although the interlayer spacing dI2appears as the dominant factor in determining the basic level of correspondence between calculation and experiment. The four-domain registry shift models RS1
832 Langmuir, Vol. 5, No. 3, 1989
Zeng and Mitchell
Table 11. Ranges of Structural ParametersD(in A) Considered for Five Additional Models 4F3
RS1
dN1 = 0.00 A = 0.14(0.13)-0.53, 0.05,0.20 d12 = 1.81(0.05)-2.16 d a = 1.81
dN1 = 0.00 dlz = 1.98 d , = 1.81
dN1 = 0.06, 0.12
dN1
A = 0.20 dlz 1.81(0.05)-2.16 d a = 1.81
A = 0.30(0.03)-0.48 dlz lM(0.05)-1.96 d a = 1.81
A
0.00(0.08)-0.56
RS2
OCTl
OCT2
dN2 = 0.00 dN2 = 0.00 A 0.15(0.15)-0.60 d12 = d2, = 1.81(0.10)-2.51 d12 = d a = lM(O.05)-2.16 dlz = 1.81(0.05)-2.16 dw = 1.81 A = 0.14(0.13)-0.53,0.05, 0.02 d a = 1.81 d g = 1.81 dN1
0.00
0.00
A is defined in Figure 1. Interlayer spacings are represented by d , and the subscripts identify the layers involved, e.g., N, 1, 2, ... for the N layer, first Cu layer, second Cu layer, ... All other Cu-Cu interlayer spacings are fixed at the bulk value (1.81 A).
ENERGY (eV) Figure 4. Comparison of three experimentalZQ curves measured for normal incidence from Cu(100)-~(2XZ)-Nwith those calculated for some articular N coplanar models: (a) 4F1 with d12= 1.96 A; (b) 4F3 with d12= 1.96 A, A = 0.14 A; (c) RS1 with d12 = 1.96 A, A = 0.30 (d) 4F1 with d12= 1.86 A; (e) 4F3 with d12= 1.86 A, A = 0.14 A; (f) RS1 with d12 = 1.86 A, A = 0.30 A; (g) OCTl with d12
x;
d a = 1.91 A; (h) OCT2 with d12 = d a = 1.91 A, A = 0.14 A.
and RS2 have some curiosity intereat, although there does not appear to be strong evidence for their relevance, compared with the 4F3-type model, for which the match-up of intensity curves is at least as good provided A is 0.14 A or less. The best correspondence at normal incidence for the 4F3 model with A equal to 0.14 A occurs for d12 close to 1.96 A. In this specific model, the calculated intensity for the (1,1/2) beam, summed over the energy range considered, is less than one-tenth that for the (1/ 2,1/2) beam. However, as A increases beyond 0.14 A for the model 4F3, the intensity of (1,1/2), and other beams symmetry, goes up rapidly. which are absent for ~(2x2)
Concluding Remarks The calculated I(E) curves for the 4F2 model shown in Figure 2 summarize the best correspondence with the experimental data available so far for the Cu(100)-c(2X2)-N surface structure. In this model, with dN1 = 0.06 A, d12 = 1.85 A, and dzz = 0.09 A, each N atom bonds to four Cu atoms in the fist metal layer and to one Cu atom in the second layer, with Cu-N bond lengths of 1.81and 2.00 A, respectively. The average bond length of 1.85 A is within 0.01 A of that predicted from a bond order-bond length relation calibrated with structural information for bulk CuSN? This therefore supports the view that this structure is basically reasonable from a chemical perspective. Nevertheless, the correspondence between I(@ curveafrom experiment and calculation is not exact, and further displacements in the metal structure seem very likely. A possible extension from the model 4F2 is to one designated 4F4, which is indicated in Figure 5. These models are related in the first instance as 4F3 to 4F1, although insofar as the Cu atoms in the second metal layer (which are
4F4 Figure 5. Lateral relaxations possible for the model of the N on Cu(100) surface structure designated 4F4.
directly below N atoms) are displaced downwards, a further p4g-type displacement can occur in the third layer. However, in this latter case,the overall C w C u interactions seem best favored by the lateral displacements in the first and third metal layers having opposite senses. The structural details for N chemisorption on the Cu(100) surface appear to differ from thase for 0 chemisorption, where the (2d2Xd2)45' surface translational symmetry now seems well established. The details for the latter structure have been controversial: although at this stage the best correspondence in the LEED intensity
Langmuir 1989,5, 833-838 analysis has been obtained with a missing-row model for the metal surface and with 0 bonding essentially to four neighboring Cu atoms.20 The differing coordinations for 0 and N at the Cu(100) surface are not inconsistent with the fact that valence considerations cause different structural arrangements for the bulk Cu(1) compounds, (20) Zeng, H. C.; McFarlane, R. A.; Mitchell, K. A. R. Surf. Sci. 1989, 208,Ll.
833
namely, CuzO and Cu3N. In both cases, the structures found by LEED for the chemisorption systems are broadly consistent with bond order-bond length relations applied to the corresponding bulk compounds. Acknowledgment. We are grateful to the Natural Sciences and Engineering Research Council of Canada for ,supporting this research. Registry No. Cu, 7440-50-8; N2, 7727-37-9.
Chain-Substituted Lipids in Monomolecular Films. Effect of Polar Substituents on Molecular Packing F. M. Menger,* S. D. Richardson, M. G. Wood, Jr., and M. J. Sherrod Department of Chemistry, Emory University, Atlanta, Georgia 30322 Received November 14,1988. In Final Form: February 5, 1989 A series of highly purified fatty acids and phospholipids each possessing a polar chain-substituent(hydroxy or keto) at varying locations (carbons 8, 10, 12, and 16 for the fatty acids and carbons 4,6,8, 10, and 12 for the phospholipids) on an 18-carbon chain were synthesized. Pressure-area isotherms revealed how these molecules pack in monomolecular films. Most of the fatty acids and phospholipids exhibited pressure-area curves indicative of "looping" conformations where both the polar substituent and polar head group contact the aqueous subphase. As the pressure was increased,the polar substituentswere forced out of the aqueous interface, and the chains assumed more vertical conformations. Pressure-area isotherms for the hydroxylated fatty acids showed unusually small molecular areas in the condensed state owing to the presence of hydrogen bonding. A phospholipid disubstituted at the 12 position with a keto group gave a molecular area of only 21.6 A2/molecule at 35 dynlcm; this is consistent with two vertical chains, one in the water and one in the air. Isotherms reflected a strong dependence on the position of the polar substituent along the chain. Introduction Over the past few years, there has been a growing interest in developing artificial membranes to serve as controlled drug delivery systems. Designing such synthetic membranes requires an understanding of how the individual molecules pack together. In an effort to learn more about "packing" behavior, we examined a series of synthetic fatty acids and phospholipids bearing diverse substituents on their hydrocarbon chains. Earlier, we published the effects of hydrocarbon branches (methyl, nbutyl, phenyl) on the packing behavior of fatty acids and phospho1ipids.l The present study deals with the effects of polar substituents (hydroxy and keto) at various locations along the chains. Stearic acid derivatives were synthesized with a hydroxy substituent a t the 8-, lo-, and 12-, or 16-positions, and distearoylphosphatidylcholines were furnished with a keto group at the 4-, 6-, 8-, lo-, and 12-positions (Scheme I). Both chains of the phospholipids, or only chain 2, were modified in this manner. Owing to the large number of compounds, it is necessary to adopt a short-hand notation. Thus, when both octadecanoyl chains possess a keto group at C4,the phosphatidylcholine will be designated (1,2)-PC-4K; a lipid having a C8 keto group on only the second chain will be called (2)-PC-8K. Packing properties of the chain-modified lipids in monomolecular films were examined by means of a Wilhelmy-type film balance. (1) Menger, F. M.; Wood,M. G., Jr.; Richardson, S.; Zhou, Q.; Elrington, A. R.; Sherrod, M. J. J. Am. Chem. SOC.1988,110,6191.
0743-7463/89/2405-0833$01.50/0
(2)- PC-8K
Recently, Cadenhead? Tachibana? Zografi? and Nagarajan5 have published pressure-area isotherms of hy(2) Kellner, B. M. J.; Cadenhead,D. A. J . Colloid Interface Sci. 1978, 63,452. ( 3 ) Tachibana,T.;Yoshizumi,T.; Hon, K. Bull. Chem. SOC.Jpn. 1979, 52, 34.
0 1989 American Chemical Society
