Silver deposition precursor molecule trimethylphosphine(hexafluoroacetylacetonato)silver(I) [(hfac)AgP(CH3)3] was used to deposit silver onto water-modified (hydroxyl-terminated) solid substrates. A silicon wafer was used as a model flat surface, and water-predosed ZnO nanopowder was investigated to expand the findings to a common substrate material for possible practical applications. Following the deposition, oxygen plasma was used to remove the remaining organic ligands on a surface and to investigate its effect on the morphology of chemically deposited silver nanoparticles and films. A combination of microscopic and spectroscopic techniques including electron microscopy and x-ray photoelectron spectroscopy was used to confirm the change in the morphology of the deposited material consistent with Ostwald ripening as a result of plasma treatment. Particle agglomeration was observed on the surfaces, and the deposited metallic silver was oxidized to Ag2O following plasma treatment. The fluorine-containing ligands were completely removed. This result suggests that chemical vapor deposition can be used to deposit silver in a very controlled manner onto a variety of substrates using different topography methods and that the post-treatment with oxygen plasma is effective in preparing materials deposited for potential practical applications.

The process of depositing silver has been studied for many decades because of a variety of applications in fields such as catalysis,1,2 energy storage,3 and sensing.4 Controlling the deposition process at the molecular level has always been of great interest.5–7 Depositing the metal on semiconducting substrates is of particular interest since it allows for additional control of the electronic properties of the resulting systems. For example, the combination of silver and silicon has shown promising properties for energy conversion.8 ZnO-supported silver is also suggested to serve as an efficient photocatalyst as well as a key component in solar cells.9–11 For many of these applications, it is imperative to control the amount and morphology of deposited silver structures, and thus, chemical deposition is often preferred to other methods to produce uniform coatings or structures reliably. However, given the limited number of suitable silver deposition precursors, one major problem that has to be addressed in chemical deposition schemes is the removal of the surface-bound ligands left on the surfaces following the deposition procedures.12 To remove this potential source of contamination, additional thermal treatments are often necessary, bringing up the cost of the devices produced and influencing the quality of the metal/semiconductor interfaces.13,14

Oxygen plasma is used in this work to serve as an alternative approach to solve the ligand contamination issue. The experimental investigation presented here uses this method to remove the remaining surface-bound ligands from deposited silver nanoparticles and thin films on a flat silicon substrate and the surface of ZnO nanopowder. Oxygen plasma has been demonstrated before to be effective in alleviating surface contamination, especially for carbon- and fluorine-containing species.15 In addition, it was also reported that the oxygen plasma could cause Ostwald ripening for silver nanoparticles,16,17 meaning that the surface morphology of the deposited nanostructures and thin films that are often formed by the Volmer–Weber mechanism18 can be improved. Thus, both the cleanliness and the surface morphology of the chemically deposited silver structures could be controlled simultaneously.

Trimethylphosphine(hexafluoroacetylacetonato)silver(I) [Ag(hfac)P(CH3)3] is used as the silver precursor in this work. It has a number of attractive properties including stability in ambient air, making it an attractive candidate for metalorganic deposition methods.18–21 A hydroxyl terminated silicon surface was used as the initial model substrate, and the formation of silver nanoparticles with controlled size distribution was recorded. Following oxygen plasma treatment, Ostwald ripening is observed, as the silver nanoparticles agglomerate into larger structures. It is also found that the extent to which the surface silver nanoparticles undergo the Ostwald ripening depends on the initial silver nanoparticle density. Specifically, the surface with higher initial silver nanoparticle density undergoes more extensive Ostwald ripening compared to the one with lower initial silver nanoparticle densities, resulting in an additional control of the structures deposited. The second set of investigations utilizes ZnO nanopowder to examine the effect of the starting surface morphology on both the deposition process and the plasma treatment. Similar to the processes investigated on a flat silicon surface, Ostwald ripening is observed following the oxygen plasma treatment. However, on the ZnO surface, smaller particles are formed with a narrower size distribution for the comparable treatment process. In all cases, x-ray photoelectron spectroscopy (XPS) confirms the efficient removal of surface fluorine species accompanied by silver oxidation.

Figure 1 shows the structure of trimethylphosphine(hexafluoroacetylacetonato)silver(I) or (hfac)AgP(CH3)3 (Strem Chemicals, Inc., 99%) used for depositing silver in all the experiments described below. This solid precursor was used as purchased without any further treatment. A p-type double-side polished Si(100) wafer (Virginia Semiconductor) was used as the silicon substrate. The hydroxylated silicon surface was prepared following a modified Radio Corporation of America procedure, and the details of this process can be found elsewhere.22,23 Briefly, the Si(100) wafer was first kept in a solution of Milli-Q water [first-generation Milli-Q water system (Millipore) with 18 MΩ·cm resistivity], hydrogen peroxide (Fisher, 30% certified ACS grade), and ammonium hydroxide (Fisher, 29% certified ACS plus grade) (4:1:1 in volume) for 10 min at 80 °C in a water bath. Then, the sample was transferred to the hydrofluoric acid buffer solution (Fisher, 29% certified ACS plus grade) for 2 min. Following this step, a solution composed of Milli-Q water, hydrogen peroxide, and hydrochloric acid (Fisher, 37.3% certified ACS grade) (4:1:1 in volume) was used to react with the sample for 10 more minutes at 80 °C to produce an oxide-covered silicon surface with high density of hydroxyl groups. To prepare the water-predosed ZnO powder, pure ZnO powder (99.99% purity, Alfa Aesar) was first pressed onto a tungsten mesh using a hydraulic press. Then, the sample was placed into a high-vacuum chamber, annealed for 30 min at 500 °C to eliminate the impurities, and cooled back to room temperature, and water was introduced into the chamber for 2 min at ∼1 Torr. After the water was pumped down, the sample was briefly heated up to 200 °C to remove molecularly adsorbed water. The water-dosed ZnO has a high surface concentration of the hydroxyl groups so that the metal deposition processes are expected to be more efficient for the precursors containing the hfac moiety.12,24 Thus, the initial stages of the reaction on hydroxylated surfaces would be expected to proceed according to the following generalized mechanism:

(Hfac)AgP(CH3)3+surface-H=HfacH(gas)+surface-AgP(CH3)3surface-AgP(CH3)3=surface-Ag+P(CH3)3(gas).
Fig. 1.

(Color online) Structure of (hfac)AgP(CH3)3. Yellow = fluorine, white = hydrogen, black = carbon, red = oxygen, purple = silver, and green = phosphorus. A simplified structure is shown on the right.

Fig. 1.

(Color online) Structure of (hfac)AgP(CH3)3. Yellow = fluorine, white = hydrogen, black = carbon, red = oxygen, purple = silver, and green = phosphorus. A simplified structure is shown on the right.

Close modal

The deposition of the silver precursor was performed in a high-vacuum chamber with a base pressure of approximately 10−5 Torr. An organo-metallic precursor doser (McAllister Technical Services) was used to introduce the precursor into the chamber. The silver precursor was loaded into the doser and heated up to 100–110 °C to allow for the sublimation of the solid. The doser and the dosing procedure have been described in detail previously.25 The substrates (hydroxylated silicon and ZnO) were placed on a button heater that can maintain a temperature of up to 1000 °C. A thermocouple was attached to the shield of the button heater to measure its temperature.

The oxygen plasma treatment was performed in a plasma cleaner (Harrick plasma, PDC-32G). The silver/silicon samples were loaded into the cleaner and exposed to the oxygen plasma with different intensities and variable time periods as indicated in the text. Specifically, low intensity gives a 6.8 W RF coil output. Medium intensity gives a 10.5 W RF coil output. High intensity gives an 18 W RF coil output. The ZnO samples were also originally treated for 10 s with oxygen plasma at low intensity, but no discernible changes were observed. As a result, the exposure time was increased to 60 s with low intensity for all the ZnO samples discussed below.

A K-Alpha+ x-ray photoelectron spectrometer system from Thermo Scientific was used to perform XPS measurements. An Al Kα x-ray source (hν = 1486.6 eV) with a 35° electron emission angle (relative to sample surface) was used for all the measurements with a 0.1 eV resolution. The calibration of the resulting spectra was performed by setting 284.6 eV as the C 1s peak.26–29casaxps version 2.3.17 was used to process all the XPS spectra.

For morphology studies, a Zeiss Auriga 60 FIB/SEM at the W. M. Keck Electron Microscopy facility at the University of Delaware was used to perform all the scanning electron microscopy (SEM) and electron dispersive spectroscopy (EDS) investigations. The accelerating voltage was set to be 6 keV with the working distance of 5.0 mm. Focused ion beam work was performed at a 54° angle relative to the SEM detector with a current of 120 pA.

Atomic force microscopy (AFM) images were acquired in the tapping mode using a J-scanner scanning probe microscope (Multimode, NanoScope V). AFM tips were purchased from Budget Sensor. The tips have a resonant frequency of 300 kHz and a force constant of 40 N/m. gwyddion was used to process all the images.

Figure 2 shows the XPS spectra of the Ag 3d region following the exposure of the hydroxylated silicon surface to 10−5 Torr of (hfac)AgP(CH3)3 for 1 h at room temperature. Figure 2(a) presents the spectrum recorded immediately after deposition, before 10 s of oxygen plasma treatment. The position of the Ag 3d5/2 peak at 368.3 eV indicates the presence of metallic silver.18,30–32 Figure 2(b) shows the silver spectrum of the same surface following the plasma treatment. The recorded peaks are clearly broader, and the binding energy of Ag 3d5/2 this time shifts to 367.8 eV, implying the formation of surface oxide or Ag2O.33 The solid lines at the bottom of the spectra show the expected Ag 3d5/2 binding energies for metallic silver and Ag2O.30–33 The dashed lines indicate the experimentally obtained positions of the Ag 3d5/2 peaks. Based on the results in Fig. 2, two major conclusions can be made. First, the interaction between the hydroxylated silicon surface and the silver-containing precursor (hfac)AgP(CH3)3 leads to silver deposition in a metallic form following the 1-electron reduction of Ag(I) in the precursor, which is fully consistent with the previously published observations.18 Second, oxygen plasma oxidizes the metallic silver on the surface, as would be expected.

Fig. 2.

XPS comparison of the Ag 3d spectra before (a) and after (b) 10 s of oxygen plasma treatment on the hydroxylated silicon surface after the deposition of (hfac)AgP(CH3)3. The solid lines at the bottom show the expected Ag 3d5/2 binding energies for metallic silver and Ag2O. The dashed lines indicate the experimentally obtained values for the Ag 3d5/2 peak.

Fig. 2.

XPS comparison of the Ag 3d spectra before (a) and after (b) 10 s of oxygen plasma treatment on the hydroxylated silicon surface after the deposition of (hfac)AgP(CH3)3. The solid lines at the bottom show the expected Ag 3d5/2 binding energies for metallic silver and Ag2O. The dashed lines indicate the experimentally obtained values for the Ag 3d5/2 peak.

Close modal

XPS investigation of the Ag 3d spectral region unveils the chemical state change of silver deposited on the surface before and after the oxygen plasma treatment. However, it does not give any information about the surface morphology. To address this question, SEM is performed on the same sample before and after the plasma treatment. As shown in Fig. 3(a), the surface before the oxygen plasma treatment shows nanoparticles that are randomly located on the support. However, following a brief plasma treatment, this morphology is changed substantially. As shown in Fig. 3(b), plasma treatment causes the smaller particles to merge into substantially larger structures, apparently undergoing Ostwald ripening.16,17 The large particle labeled as “1” in Fig. 3(b) can serve as a good example. Note the absence of the small nanoparticles in the vicinity of this particle in the image. This observation supports the occurrence of Ostwald ripening. According to Fig. 3(a), the average particle size before plasma treatment is approximately 20 nm. The diameter of particle 1 in Fig. 3(b) is nearly 150 nm. To confirm this hypothesis, EDS is also performed at the center of particle 1 as shown in Fig. 3(b) to probe the element composition. The result shows the presence of silver in this structure (Fig. S1 in the supplementary material).35 

Fig. 3.

SEM comparison of the hydroxylated silicon surface before (a) and after (b) 10 s of oxygen plasma treatment following the deposition of (hfac)AgP(CH3)3. The number in (b) indicates the location where EDS was taken.

Fig. 3.

SEM comparison of the hydroxylated silicon surface before (a) and after (b) 10 s of oxygen plasma treatment following the deposition of (hfac)AgP(CH3)3. The number in (b) indicates the location where EDS was taken.

Close modal

As discussed in Sec. III A, Ostwald ripening clearly occurred as a result of the oxygen plasma treatment, and it leads to the agglomeration of the small silver nanoparticles into larger ones. If the amount of silver deposited (nanoparticle density) could be increased, it is possible that the Ostwald ripening process could be confirmed in more detail. Hence, a number of additional experiments have been conducted. First, silver deposition was performed again at ∼10−5 Torr for 1 h. However, this time the temperature of the substrate (hydroxylated silicon) was kept at 350 °C instead of room temperature. As confirmed in Fig. 4(a), the density of silver nanoparticles deposited at 350 °C is higher and the particles are larger in size compared to the ones deposited at room temperature [Fig. 3(a)]. Additional AFM and XPS information comparing the two surfaces (room temperature deposition vs 350 °C deposition) can be found in the supplementary material. Specifically, Fig. S2 shows that the deposition at room temperature leads to the formation of particles with a lower density and a smaller average height. The Ag 3d signals from the corresponding survey spectra also show a stronger silver intensity for the sample deposited at 350 °C, confirming that more silver is present on this surface. Figure S3 illustrates the height distributions of the nanoparticles deposited at two different temperatures. The nanoparticles deposited at room temperature have an average height of approximately 3 nm, while those deposited at 350 °C have an average height of 7 nm. After 10 s of plasma treatment, particle agglomeration was observed again, as shown in Fig. 4(b). However, compared to the particles in Fig. 3(b), the occurrence of Ostwald ripening was very clear this time. In addition, the average particle size of this sample after the treatment is larger (21 nm vs 13 nm) than that of the sample shown in Fig. 3(b) based on the AFM studies. The last observation is fully consistent with the previously published work.17 Namely, higher initial particle density results in a larger particle size following the oxygen plasma treatment. EDS was performed at the center of particle 1 as shown in Fig. 4(b), and the silver signal was detected again (Fig. S4).

Fig. 4.

(Color online) SEM images of the hydroxylated silicon surface before (a) and after (b) 10 s of oxygen plasma treatment. The silver nanoparticles were deposited on the surface at ∼10−5 Torr for 1 h at 350 °C. The number in (b) indicates the location where EDS was taken.

Fig. 4.

(Color online) SEM images of the hydroxylated silicon surface before (a) and after (b) 10 s of oxygen plasma treatment. The silver nanoparticles were deposited on the surface at ∼10−5 Torr for 1 h at 350 °C. The number in (b) indicates the location where EDS was taken.

Close modal

To grow a thicker silver film on the same surface for further comparison, (hfac)AgP(CH3)3 was dosed for 2 h at 10−2 Torr while keeping the substrate temperature at 350 °C. Figure 5 shows SEM images before [(a) and (b)] and after [(c) and (d)] 10 s of oxygen plasma treatment of the resulting surfaces. The morphology of the silver structures can be described as the relatively well-ordered silver structures likely grown initially by a Volmer–Weber nucleation mechanism.18 Clearly, surface morphology changes dramatically after the plasma treatment. The nanoparticles become much more disordered and are overall similar to porous silver.

Fig. 5.

(Color online) SEM comparison of the hydroxylated silicon surface with deposited silver nanoparticles at 10−2 Torr 350 °C for 2 h before [(a) and (b)] and after [(c) and (d)] 10 s of low intensity oxygen plasma treatment.

Fig. 5.

(Color online) SEM comparison of the hydroxylated silicon surface with deposited silver nanoparticles at 10−2 Torr 350 °C for 2 h before [(a) and (b)] and after [(c) and (d)] 10 s of low intensity oxygen plasma treatment.

Close modal

Based on the comparison of the results in Figs. 3–5, it is clear that the final morphology of the surface depends on the initial density of the silver nanoparticles. Specifically, the surface with a higher density of silver nanoparticles tends to be more susceptible to the oxygen plasma treatment, resulting in the formation of large clusters. Of course, partially it can be explained by the proximity of nanoparticles, making it easier for the Ostwald ripening process to occur.

Apart from Ostwald ripening, oxygen plasma also serves as an efficient technique to eliminate the surface contamination. The initial stages of silver deposition at room temperature were reported to have minimal fluorine contamination consistent with the proposed mechanism eliminating hfacH;18 however, fluorine contamination becomes pronounced for the formation of larger silver structures or films in a CVD regime (at elevated temperatures). That is why, in addition to SEM images, the samples deposited with silver at 10−2 Torr 350 °C for 2 h are also examined by XPS. As shown in Fig. 6, before the plasma treatment [Fig. 6(a)], the F 1s peak at 688.4 eV indicates the presence of −CF3 species in the hfac moiety of the precursor.12 This peak disappears completely following 10 s of plasma treatment [Fig. 6(b)], confirming the effective removal of the surface fluorine-containing species. The high-resolution XPS spectra of the Ag 3d region are also taken (Fig. S5), and they show the oxidation of the surface silver again following the oxygen plasma treatment, which is consistent with the result shown in Fig. 2 for low nanoparticle density samples.

Fig. 6.

F 1s XPS spectra before (a) and after (b) the oxygen plasma treatment for the sample shown in Fig. 5, (a) and (c), respectively.

Fig. 6.

F 1s XPS spectra before (a) and after (b) the oxygen plasma treatment for the sample shown in Fig. 5, (a) and (c), respectively.

Close modal

To gain a better understanding of the surface composition change before and after the plasma treatment, the XPS surveys of the sample prepared at 10−2 Torr 350 °C for 2 h are compared in Fig. 7. Consistent with Fig. 6(b), the F1s peak disappears following plasma treatment [Fig. 7(b)], confirming the removal of surface fluorine. At the same time, carbon concentration decreased approximately two-fold, indicating that oxygen plasma is also efficient in removing carbon contamination from the surface. It should be emphasized that the presence of any carbon contamination following plasma treatment may be caused by the impurities adsorbed on the sample during its brief exposure to ambient conditions during the transfer to the XPS instrument. This is why the removal of fluorine-containing species confirmed by these measurements is more indicative of the efficiency of plasma treatment for ligand removal.

Fig. 7.

Survey spectra of the surface (described in Fig. 5) before (a) and after (b) the plasma treatment.

Fig. 7.

Survey spectra of the surface (described in Fig. 5) before (a) and after (b) the plasma treatment.

Close modal

Additional studies were also performed to shed light on the formation process of the silver clusters during plasma treatment. The corresponding SEM images are shown in Fig. 8. The silver precursor is first introduced into the reaction chamber at 10−2 Torr 350 °C for 4 h to grow a continuous silver film, as shown in Figs. 8(a) and 8(b). This film is still not smooth, but the average thickness of the film deposited at these conditions is approximately 75 nm. Following the successful deposition of the film, the surface is treated with oxygen plasma for 30 s, and it is apparent that smaller nanoparticles, with diameters below 10 nm begin to form around the structures comprising the starting film [Fig. 8(c)]. When the surface is further treated with oxygen plasma for another 60 s, the density of the smaller nanoparticles clearly increases, as shown in Fig. 8(d). If the sample is additionally treated with plasma for 60 s at medium intensity, the density of the smaller nanoparticles increases further [Fig. 8(e)]. Eventually, if the sample is treated with plasma for 300 s at high intensity, silver nanorods are formed on the surface [Fig. 8(f)], which is consistent with the previous observations for the agglomeration process for silver deposited by evaporation.17 In other words, when the surface covered with a continuous nanostructured film of silver is exposed to the oxygen plasma, the initial step in the morphological change of the silver films deposited by the chemical deposition method is the formation of small silver nanoparticles. As the intensity of the plasma and the duration of exposure are increased, the small silver nanoparticles merge and grow into larger structures, and finally, silver nanorod formation is recorded. Thus, if compared with the results in Fig. 5, it is also important to note that the formation of the larger structures and nanorods will depend dramatically on the structure of the starting silver film since the sample in Fig. 5 with a smaller amount of silver and lower density of starting structures does not show the same behavior, as the samples covered with continuous silver films presented in Fig. 8.

Fig. 8.

(Color online) SEM images of the silver film deposited on the hydroxylated silicon surface for 4 h, at 10−2 Torr 350 °C (a), the side view showing the thickness of the silver deposited on the surface (b), the same surface after 30 s of oxygen plasma treatment at low intensity (c), the surface in (c) after 60 s of oxygen plasma treatment at low intensity (d), the surface in (d) after 60 s of oxygen plasma treatment at medium intensity (e), and the surface in (e) after 300 s of oxygen plasma treatment at high intensity (f).

Fig. 8.

(Color online) SEM images of the silver film deposited on the hydroxylated silicon surface for 4 h, at 10−2 Torr 350 °C (a), the side view showing the thickness of the silver deposited on the surface (b), the same surface after 30 s of oxygen plasma treatment at low intensity (c), the surface in (c) after 60 s of oxygen plasma treatment at low intensity (d), the surface in (d) after 60 s of oxygen plasma treatment at medium intensity (e), and the surface in (e) after 300 s of oxygen plasma treatment at high intensity (f).

Close modal

As mentioned above in the Introduction, ZnO can also be an interesting substrate to study the process of silver deposition and plasma post-treatment. Previous work12 shows that copper nanoparticles can be successfully grown on ZnO powder using chemical deposition with Cu(hfac)VTMS as a precursor. Building on this previous work, it is important to extend the knowledge of the deposition and plasma treatment based on flat surfaces to a more complex and possibly more practical substrate, ZnO surface, by using (hfac)AgP(CH3)3. Of course, it would be expected that the deposition process shown on silicon will also work on ZnO; however, it is also known that despite the chemical similarities, different substrates can behave very differently even for the same precursor due to the nature of the substrate itself.18,25,34

First, to rule out the possibility that oxygen plasma itself may affect the morphology of the ZnO nanopowder, pure ZnO nanopowder dosed with water according to the previously established procedure12–14 to prepare the hydroxyl-terminated surface is treated with oxygen plasma for 60 s. As shown in Fig. 9, the morphology of the water-predosed ZnO powder does not change noticeably before [Fig. 9(a)] and after [Fig. 9(b)] the exposure to the low intensity oxygen plasma for 60 s.

Fig. 9.

(Color online) SEM images of the water-predosed ZnO nanopowder (a) and the same nanopowder after 60 s of oxygen plasma treatment (b).

Fig. 9.

(Color online) SEM images of the water-predosed ZnO nanopowder (a) and the same nanopowder after 60 s of oxygen plasma treatment (b).

Close modal

After confirming that oxygen plasma itself does not affect the morphology of the pure water-predosed ZnO powder, silver deposition is performed onto the water-predosed ZnO. As shown in Fig. 10, before the plasma treatment [Fig. 10(a)], no nanoparticles can be discerned from the surface of the water-predosed ZnO immediately following the silver deposition, possibly because the silver nanoparticles are too small to be detected by the SEM. However, after 60 s of oxygen plasma treatment, nanoparticles with the size around 5–10 nm appear on the ZnO surface [Fig. 10(b)]. This is very likely the result of Ostwald ripening. Specifically, small nanoparticles that cannot be detected easily by SEM grow into larger surface structures that are easily observed. It is also worth noting that compared to the nanoparticles produced by Ostwald ripening on the hydroxylated silicon surface, the nanoparticles on the ZnO surface [based on Fig. 10(b)] appear to be much more evenly distributed in terms of particle size. Specifically, no large size structures are observed in Fig. 10(b). All the particles have the diameter between 5 and 10 nm.

Fig. 10.

(Color online) SEM images showing the water-predosed ZnO powder before (a) and after (b) 60 s of low intensity oxygen plasma treatment. The inset in each figure illustrates the zoom-in image of the boxed area. Following the plasma treatment, the morphology of the ZnO powder changes. Small nanoparticles (b) appear on the surfaces of the ZnO crystal regardless of the crystal orientation.

Fig. 10.

(Color online) SEM images showing the water-predosed ZnO powder before (a) and after (b) 60 s of low intensity oxygen plasma treatment. The inset in each figure illustrates the zoom-in image of the boxed area. Following the plasma treatment, the morphology of the ZnO powder changes. Small nanoparticles (b) appear on the surfaces of the ZnO crystal regardless of the crystal orientation.

Close modal

Apart from morphology investigations, the chemical state change of the silver on the surface is also observed by XPS. Consistent with the results acquired from the silicon samples, the surface silver undergoes oxidation following plasma treatment, as indicated by the binding energy shift of the Ag 3d5/2 peak from 368.5 eV before the plasma treatment [Fig. 11(a)] down to 367.5 eV after the plasma treatment [Fig. 11(b)]. The solid lines at the bottom of the spectra show the expected Ag 3d5/2 binding energies for metallic silver and Ag2O.30–33 The dashed lines indicate the experimental value of Ag 3d5/2.

Fig. 11.

XPS of the silver spectra on the water-predosed ZnO surface before (a) and after (b) 60 s of low intensity oxygen plasma treatment. The solid lines at the bottom show the expected Ag 3d5/2 binding energies for metallic silver and Ag2O. The dashed lines indicate the experimental value of Ag 3d5/2.

Fig. 11.

XPS of the silver spectra on the water-predosed ZnO surface before (a) and after (b) 60 s of low intensity oxygen plasma treatment. The solid lines at the bottom show the expected Ag 3d5/2 binding energies for metallic silver and Ag2O. The dashed lines indicate the experimental value of Ag 3d5/2.

Close modal

By using SEM, XPS, and AFM, the formation of silver nanoparticles is confirmed following the reaction of silver deposition precursor (hfac)AgP(CH3)3 with hydroxylated silicon and water-predosed ZnO surfaces. Following the deposition, Ostwald ripening is observed on both surfaces as a result of the oxygen plasma treatment. The process of agglomeration depends on the initial silver nanoparticle density, amount of silver, and plasma exposure time. This can be explained by the fact that the surface with denser silver nanoparticles has more available silver atoms as the source to undergo the Ostwald ripening process, making it easier for the silver nanoparticles to grow larger. In contrast, when the surface silver nanoparticles are less dense, fewer silver atoms are available for Ostwald ripening. Ostwald ripening occurs on ZnO nanopowder as well, leading to the formation of silver structures that are clearly observed following oxygen plasma treatment. Finally, it is also shown that oxygen plasma can be an effective way to remove fluorine- and carbon-containing surface contamination, with essentially all fluorine-containing surface adducts removed by 10 s of plasma treatment of the silicon sample. All these findings suggest that oxygen plasma is a practical and efficient way to tune the size, density, and purity of silver nanoparticles deposited by metalorganic silver deposition, both in a regime where the reaction is limited by the surface (room temperature) and the regime where regular CVD is reached (350 °C).

The authors acknowledge the donors of the Petroleum Research Fund, administered by the American Chemical Society, for partial support of this research. This work was also partially supported by the National Science Foundation [CHE1057374 and DMR1609973 (GOALI)]. Additional partial support was provided by the University of Delaware Research Foundation Strategic Initiatives (UDRF-SI) Grant. The authors acknowledge the NSF (9724307; 1428149) and the NIH NIGMS COBRE program (P30-GM110758) for partial support of activities in the University of Delaware Surface Analysis Facility.

1.
H.
Fang
,
Y.
Wu
,
J. H.
Zhao
, and
J.
Zhu
,
Nanotechnology
17
,
3768
(
2006
).
2.
I. E.
Wachs
and
R. J.
Madix
,
Surf. Sci.
76
,
531
(
1978
).
3.
Y.
Yu
,
L.
Gu
,
C. B.
Zhu
,
S.
Tsukimoto
,
P. A.
van Aken
, and
J.
Maier
,
Adv. Mater.
22
,
2247
(
2010
).
4.
B. H.
Zhang
,
H. S.
Wang
,
L. H.
Lu
,
K. L.
Ai
,
G.
Zhang
, and
X. L.
Cheng
,
Adv. Funct. Mater.
18
,
2348
(
2008
).
5.
M. S.
Rill
,
C.
Plet
,
M.
Thiel
,
I.
Staude
,
G.
Von Freymann
,
S.
Linden
, and
M.
Wegener
,
Nat. Mater.
7
,
543
(
2008
).
6.
S. T.
Dubas
,
P.
Kumlangdudsana
, and
P.
Potiyaraj
,
Colloids Surf., A
289
,
105
(
2006
).
7.
K.
Luo
,
T. P.
St Clair
,
X.
Lai
, and
D. W.
Goodman
,
J. Phys. Chem. B
104
,
3050
(
2000
).
8.
T. L.
Temple
,
G. D. K.
Mahanama
,
H. S.
Reehal
, and
D. M.
Bagnall
,
Sol. Energy Mater. Sol. Cell
93
,
1978
(
2009
).
9.
A.
Kim
,
Y.
Won
,
K.
Woo
,
C. H.
Kim
, and
J.
Moon
,
ACS Nano
7
,
1081
(
2013
).
10.
M. J.
Height
,
S. E.
Pratsinis
,
O.
Mekasuwandumrong
, and
P.
Praserthdam
,
Appl. Catal. B
63
,
305
(
2006
).
11.
R.
Georgekutty
,
M. K.
Seery
, and
S. C.
Pillai
,
J. Phys. Chem. C
112
,
13563
(
2008
).
12.
H.
Kung
,
Y.
Duan
,
M. G.
Williams
, and
A. V.
Teplyakov
,
Langmuir
32
,
7029
(
2016
).
13.
H.
Kung
and
A.
Teplyakov
,
J. Phys.: Condens. Matter
27
,
054007
(
2015
).
14.
H.
Kung
and
A.
Teplyakov
,
J. Catal.
330
,
145
(
2015
).
15.
M.
Morra
,
E.
Occhiello
, and
F.
Garbassi
,
Langmuir
5
,
872
(
1989
).
16.
K.
Morgenstern
,
G.
Rosenfeld
, and
G.
Comsa
,
Surf. Sci.
441
,
289
(
1999
).
17.
J.
Tang
,
P.
Photopoulos
,
A.
Tserepi
, and
D.
Tsoukalas
,
Nanotechnology
22
,
235306
(
2011
).
18.
Y.
Duan
,
S.
Rani
,
Y.
Zhang
,
C.
Ni
,
J. T.
Newberg
, and
A. V.
Teplyakov
,
J. Phys. Chem. C
121
,
7240
(
2017
).
19.
H. K.
Kim
,
H. C.
Jeong
,
K. S.
Kim
,
S. H.
Yoon
,
S. S.
Lee
,
K. W.
Seo
, and
I. W.
Shim
,
Thin Solid Films
478
,
72
(
2005
).
20.
Z.
Yuan
,
N. H.
Dryden
,
J. J.
Vittal
, and
R. J.
Puddephatt
,
Chem. Mater.
7
,
1696
(
1995
).
21.
N. H.
Dryden
,
J. J.
Vittal
, and
R. J.
Puddephatt
,
Chem. Mater.
5
,
765
(
1993
).
22.
K. A.
Perrine
and
A. V.
Teplyakov
,
Langmuir
26
,
12648
(
2010
).
23.
K. A.
Perrine
,
J.-M.
Lin
, and
A. V.
Teplyakov
,
J. Phys. Chem. C
116
,
14431
(
2012
).
24.
H.
Kung
and
A. V.
Teplyakov
,
J. Phys. Chem. C
118
,
1990
(
2014
).
25.
Y.
Duan
,
F.
Gao
, and
A. V.
Teplyakov
,
J. Phys. Chem. C
119
,
27018
(
2015
).
26.
A.
Bansal
,
X. L.
Li
,
S. I.
Yi
,
W. H.
Weinberg
, and
N. S.
Lewis
,
J. Phys. Chem. B
105
,
10266
(
2001
).
27.
J. A.
Haber
and
N. S.
Lewis
,
J. Phys. Chem. B
106
,
3639
(
2002
).
28.
W. J.
Gammon
,
O.
Kraft
,
A. C.
Reilly
, and
B. C.
Holloway
,
Carbon
41
,
1917
(
2003
).
29.
X.
Tan
,
Q.
Fan
,
X.
Wang
, and
B.
Grambow
,
Environ. Sci. Technol.
43
,
3115
(
2009
).
30.
M. P.
Seah
,
G. C.
Smith
, and
M. T.
Anthony
,
Surf. Interface Anal.
15
,
293
(
1990
).
31.
G.
Johansson
,
J.
Hedman
,
A.
Berndtsson
,
M.
Klasson
, and
R.
Nilsson
,
J. Electron. Spectrosc. Relat. Phenom.
2
,
295
(
1973
).
32.
V. K.
Kaushik
,
J. Electron. Spectrosc. Relat. Phenom.
56
,
273
(
1991
).
33.
S. W.
Gaarenstroom
and
N.
Winograd
,
J. Chem. Phys.
67
,
3500
(
1977
).
34.
Y.
Duan
and
A. V.
Teplyakov
,
J. Chem. Phys.
146
,
052814
(
2017
).
35.
See supplementary material at http://dx.doi.org/10.1116/1.4986208 for EDS spectrum of particle 1 in Fig. 3(b) and particle 1 in Fig. 4(b) confirming the presence of silver in the resulting structure, AFM-based analysis of particle size distribution for hydroxylated silicon surface after the deposition of the silver precursor at room temperature and 350 °C at ∼10−5 Torr for 1 h, additional XPS spectra.

Supplementary Material