Multilayer growth

When a high (negative) overpotential is applied a second layer can begin to grow before the first one is completed. This leads to multilayer growth, which is only imperfectly understood, so we refrain from a further discussion. 

At higher overpotentials the nucleation rate increases faster than the step (Chapter 3) propagation rate, and the deposition of each layer proceeds with the formation of a large number of nuclei. This is the multinuclear multilayer growth. Armstrong and Harrison (13) have shown that initially, the theoretical current-time transient for the two-dimensional nucleation (Fig. 7.7) has a rising section, then passes through several damped oscillations, and finally, levels out to a steady state. 

Figure 7.7 also shows the theoretical i-t transients for the formation of successive layers under conditions of progressive nucleation. The theoretical current-time transient for three-dimensional nucleation is shown in Figure 7.8. The difference between 2D and 3D nucleation (Fig. 7.7 and 7.8) is in the absence of damped oscillations in the latter case. A comparison between the theoretical and experimental transients for the 2D polynuclear multilayer growth is shown in Figure 7.9. 

Figure 7.9. Experimental and theoretical (open circles, Monte Carlo simulation) current transients for polynuclear multilayer growth. (From Ref. 22, with permission from Annual Reviews, Inc.)...

By integrating Equations (3) and (4), neglecting the A term, with random initial conditions, mounds similar to those of the simulation can be obtained (Fig. 3). These mounds also coarsen in time. However, there has not been direct test of this equation as a description of multilayer growth. In particular, Eq. (4) was derived by fitting to Monte-Carlo data in the submonolayer regime. In this paper we show that certain aspects of multilayer growth by the Monte-Carlo model are well represented by Eq. (3) and (4). 

Schlenoff JB, Dubas ST (2001) Mechanism of polyelectrolyte multilayer growth charge overcompensation and distribution. Macromolecules 34 592-598... 

J. M. Phillips and T. R. Story, Commensurability Transitions in Multilayers A Response to Substrate-Induced Elastic Stress, Phys. Rev. B 42 (1990) 6944-6953 C. D. Hruska and J. Phillips, Observed Microscopic Structure in the Simulation of Multilayers, Phys. Rev. B 37 (1988) 3801-3804 J. M. Phillips and C. D. I oiska, Methane Adsorbed on Graphite. IV. Multilayer Growth at Low Temperatures, Phys. Rev. B 39 (1989) 5425-5435 J. M. Phillips, Layer by Layer Melting of Argon Films on Graphite A Computer Simulation Study, Phys. Lett. A 147 (1990) 54-58 J. M. Phillips, The Structure near Transitions in Thin Films, Langmuir 5 (1989) 571-575. 

Chapter 5). The measured overpotential dependence of the steady state current density of the transients (c.f. Fig. 6.15) indicates a multilayer growth mode generated by screw dislocations [4.62, 4.63, 4.68, 4,69]. It can be assumed that the screw dislocations of a real Au.Qikl) substrate are inherited by the growing ultra-thin Ag film, determining the growth mechanism under these experimental conditions. 

Figure 5.24 Oscillogram of a current transient at multinuclear multilayer growth in the standard system Ag (100)/AgNO3 15.45). Overvoltage / = - 14 mV current scale 2 (tA div" time scale 5 ms divV...

Figure 5.25 Experimental and theoretical current transients for polynuclear multilayer growth [5.45]. Theoretical solid line [5.40] dashed line [5.39] circles, Monte Carlo simulation [5.43] dotted line first monolayer according to eq. (5.15) with/ from the initial part of the transients. The shadowed area represents the range of variation of the experimental i-t curves obtained at different overpotentials on quasi-perfect Ag (100) faces in the standard system Ag (100)/AgNO3 [5.45, 5.46[. All transients are normalized to the i,max and fi,max of the first layer formation.

The results obtained in the system C xQikt)/C x clearly correspond to 3D Me phase formation on the native substrate involving spiral growth and/or 2D nucleation and multilayer growth. It is evident that experiments in such systems are solely carried out to demonstrate local metal deposition, but not for surface heterostructuring and modification. 

For 1-min deposition from 0.1 mM Os solution, only 39 5% of the islands are one layer high, and there is a significant multilayer growth even up to 5 monolayer height.19 In general, the tallest islands are the widest islands as well. One layer high islands typically... 

The multilayer growth appeared more significantly with Os, with 61 5% after only 1-min deposition from a 0.1 mM Os solution vs. only 24 3% multilayer growth on Ru/Pt(l 11) after 1-min deposition from a 1 mM Ru solution. There was a wider distribution of the island width and height for a single deposition of Os... 

Detailed measiuements using a number of different teehniques revealed that the subsequent multilayer growth on top of the reaehed eopper stmeture substrate is aeeompanied by a pronouneed dewetting leading to the formation of tall 3D islands with a bulk-like stmeture on top of the ehemisorbed wetting layer [61]. 

Increasing further the coverage results in multilayer growth and wetting transitions of CO which were investigated, for example, in Refs. 190 and 397. 

---