| 摘要: | 本研究利用掃描式電子穿隧顯微鏡(in situ scanning tunneling microscopy，in situ-STM)和循環伏安法(cyclic voltammetry，CV)探討兩個主題：第一個部分是觀察3-巰基丙烷磺酸鈉鹽（3-Mercapto-1-propanesulfonic acid sodium salt，MPS）在0.1 M 過氯酸鉀 (pH 3) 溶液中，於鉑(111)電極上的吸附結構以及對電鍍銅的影響。從STM的結果可知，在-0.3 V ~ 0.1 V的電位區間，MPS可形成整齊的吸附結構：(2 × 2)和(√3 × √3)R30°且在-0.15 V時，可觀察到兩種結構同時存在；當電位大於0.1 V時，便會形成不整齊的吸附態，此為可逆的過程；相反的，MPS於0.1 M 過氯酸 (pH 1)溶液中，不論是調整電位或者濃度改變，均無法觀察到整齊的吸附結構1。由此可知，溶液的pH值、陽離子或是電場的調控均會影響MPS於鉑(111)電極上的吸附行為。鉑(111)電極先經MPS修飾後，銅會以隨機的方式於電極表面上成核，此與鉑(111)電極上鍍銅的成核位置不同，顯示在不同載體上銅具有不同的移動性。沉積過程中，銅是以二維層狀的方式進行沉積且MPS和氯分開吸附於不同區塊上，且MPS於UPD銅上會形成(√3 × √3)R30°的整齊結構，此現象與CV的結果相符卻與0.1 M 過氯酸下鍍銅的結果(三維島狀物)有很大的差異。實驗上，使用較小的過電壓(η< 20 mV)可進行薄層銅的沉積，過程中銅依然是以水平的方式成長，於薄層銅(1.8 ML)上MPS形成整齊的(3 × 3)結構，與pH 1溶液中所觀察到的結構相異1，表示溶液的pH值和銅膜厚度皆會影響MPS分子於銅膜上的覆蓋度以及空間的排列。由CV結果可知，鉑(111)電極經MPS修飾後對銅沉積有抑制的效果，且依舊可觀察到氯吸附於單層銅上的特徵峰，此外，在0.1 V出現的另一特徵峰，推測為MPS吸附於單層銅上造成，由此可知，MPS和Cl兩吸附物種彼此互為競爭的角色且從訊號峰分開的結果可知，兩物種是分開吸附於銅膜上，為非勻相吸附。兩者所佔的比例與濃度有相對的關係。 第二個部分是於銅薄膜修飾的鉑(111)電極上，觀察鎳、鈷的沉積模式以及結構的變化，另外加入緩衝劑-硼酸，觀察對鎳、鈷沉積行為的影響。兩金屬的沉積模式皆以二維層狀的方式成長，同層數的薄膜會結合在一起形成連續且平整的鎳膜或鈷膜，厚度可達十層且具有原子級的平坦度。沉積過程中，使用較小的過電壓(η< 20 mV)以觀察結構的轉變和氯的吸附。鈷金屬的成核方式是以1~2層的核島沉積於銅薄膜表面上的缺陷和台階邊緣；鎳則是以單原子層的島狀物優先沉積於缺陷周圍。成核過程中，銅薄膜電極的形貌沒有改變，表示鎳、鈷與銅載體間無混合的現象。藉由高解像STM觀察不同厚度的薄膜，其表面形貌和氯吸附層的原子結構。在第一層薄膜上，觀察到氯的吸附層呈現雙條紋狀結構(double-lined pattern)；第二層則是中空的環狀結構(hollow ring pattern);第三層之後，主要是以波浪狀結構(moiré pattern)和凹陷三角形(triangular depressions)的特殊結構存在。受到載體與沉積金屬原子大小不同和電場控制的因素導致金屬原子沉積於不同位置上，因此產生高度差不同的氯吸附層，然而隨著薄膜厚度增加，受到載體影響的程度則逐漸變小。薄膜剝除的位置從台階邊緣開始發生且以層狀剝離的模式進行。由實驗結果可知，硼酸可降低鎳沉積所需的過電壓，使其還原電位提前且沉積速度增加；然而，對於鈷的沉積無明顯提前的效果但可抑制氫氣的產生，此外，硼酸的加入不影響薄膜的成長模式。因此，硼酸不但可阻止電鍍過程中pH的變化，亦同時可催化鎳的還原反應。 In situ scanning tunneling microscopy (STM) and cyclic voltammetry (CV) were used to examine (I) the spatial structure of adsorbed 3-Mercapto-1-propanesulfonic acid (MPS) molecules and its effect on the electrodeposition of copper on a Pt(111) electrode in 0.1 M KClO4 + 1 mM HCl + 10-7 M MPS (pH 3). Two ordered MPS structures, Pt(111) - (2 × 2) (θ = 0.25) and (√3 × √3)R30° (θ = 0.33) structures were observed between -0.3 V ~ 0.1 V (vs Ag/AgCl). These MPS structures were no longer present at E > 0.1 V. Shifting the potential negatively could restore the ordered structures of (√3 × √3)R30° and (2 × 2). By contrast, the MPS adlayer seen in 0.1 M HClO4 was always disordered, regardless of the potential of Pt(111) electrode or adjusting the concentration of MPS. It is reasonable to state that pH, potential control, and/or countercations to the sulfonate group of the MPS admolecule could be important in guiding the adsorption of MPS molecules on Pt(111) electrode. Real-time STM imaging revealed random nucleation of copper adatoms on MPS-modified Pt(111), whch contrasts markedly with copper deposition on Pt(111). This difference implies that copper adatoms could have different mobility on the substrates, thereby rendering unlike nucleation and growth processes. Segregated domains of (√3 × √3)R30° - MPS and chloride, were observed atop a monolayer of Cu deposit. By contrast, the Cu deposit on MPS-modified Pt(111) in 0.1 M HClO4 was decidedly rough. With a small overpotential (η < 20 mV), multilayer copper was electroplated on Pt(111) in a layered manner, producing atomically smooth Cu deposit capped by patches of (3 × 3) MPS. pH and the thickness of Cu film could affect the coverage and spatial structure of MPS. Strongly adsorbed MPS molecules on the Pt(111) electrode could impede the rate of Cu2+ reduction, thereby inhibiting rather than accelerating electrodeposition of copper under the present conditions. The UPD peaks still appeared at the same potentials, but broadened noticeably and decreased in intensity. Meanwhile, a weak reduction feature emerged at 0.1 V, which is associated with Cu UPD at MPS-occupied domains on Pt(111). MPS and chloride anions could compete for surface sites in the potential region of Cu UPD. These CV results indicate that chloride and MPS adspecies segregated into different domains, rather than mixed homogeneously to form a different structure. (II) Nickel and cobalt thin film electrodeposited onto a Pt(111) – supported copper film in 0.1 M KClO4 + 1 mM HCl (pH 3) + 0.06 M NiCl2/0.04 M CoCl2. The role of boric acid, a buffer used frequently in the deposition of Ni and Co, was investigated. Both Ni and Co deposit were grown in 2D mode on Pt(111). The Pt(111) – supported copper film was atomically smooth up to 10 layers, onto which a highly ordered chloride adlattice was formed. By a small (< 20 mV) overpotential triggered random nucleation at surface defects of pit and step. Cobalt clusters measured several nanometer in diameter and 1-2 atoms in height were stable against prolonged STM imaging. For nickel, it’s used to deposit prior to the defects in a monolayer in height. The morphology of copper substrate was unchanged by nucleation, suggesting a minimal surface mixing at the Co or Ni/Cu interface. Deposition of cobalt/nickel proceeded subsequently in layers, whose atomic structures were deduced from the atomic-resolution STM images obtained with the chloride adlayer. On the first layer the chloride adlayer produced a pair-lined pattern, followed by a hollow ring pattern seen with Cl adsorbed on the second layer of metal film. Starting from the third layer and up to the 10th layer, the chloride adlayer looked alike, exhibiting a prominent moiré pattern along with minor triangular depressions. Cobalt/nickel atoms would have to occupy different sites on the Cu substrate, yielding unlike corrugation heights of the adsorbed chloride adatoms. As the film thickened, the Pt substrate exerted less influence on the structure of Ni. Raising the potential resulted in stripping of Ni, which appeared to start from step edges and proceeded in a layered mode. Adding boric acid in the electrolyte could shift the potential for Ni deposition to more positive values and promoted the reduction of Ni2+. However, boric acid had little effect on Co deposition; it clearly inhibited hydrogen evolution. It appears that boric acid acted not only as a buffer but also as a catalyst toward the reduction of Ni and Co. |
