Dimensional Tailoring of Hydrothermally Grown Zinc Oxide Nanowire

Subscriber access provided by UNIV OF CALIFORNIA SAN DIEGO LIBRARIES

Communication

Dimensional tailoring of hydrothermally-grown zinc oxide nanowire arrays Jayce J. Cheng, Samuel M. Nicaise, Karl K. Berggren, and Silvija Gradecak Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.5b04625 • Publication Date (Web): 26 Dec 2015 Downloaded from http://pubs.acs.org on December 27, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Nano Letters is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Dimensional tailoring of hydrothermally-grown zinc oxide nanowire arrays Jayce J. Cheng1, Samuel M. Nicaise2, Karl K. Berggren2, Silvija Gradečak1* 1

Department of Materials Science and Engineering, 2Department of Electrical Engineering and

Computer Science, Massachusetts Institute of Technology, 77 Massachusetts Ave, Cambridge, MA 02139, USA

Abstract. Hydrothermally-synthesized ZnO nanowire arrays are critical components in a range of nanostructured semiconductor devices. The device performance is governed by relevant nanowire morphological parameters that cannot be fully controlled during bulk hydrothermal synthesis due to its transient nature. Here, we maintain homeostatic zinc concentration, pH, and temperature by employing continuous flow synthesis and demonstrate independent tailoring of nanowire array dimensions including areal density, length, and diameter on device-relevant lengthscales. By applying diffusion/reaction-limited analysis, we separate the effect of local diffusive transport from the c-plane surface reaction rate and identify direct incorporation as the c-plane growth mechanism. Our analysis defines guidelines for precise and independent control of the nanowire length and diameter by operating in rate-limiting regimes. We validate its utility by using surface adsorbents that limit reaction rate to obtain spatially uniform vertical growth rates across a patterned substrate.

ACS Paragon Plus Environment

1

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 28

Keywords. Zinc oxide, nanowires, hydrothermal growth, nanowire density control, reactionlimited growth, morphological tailoring.

As zinc oxide (ZnO) nanostructures find an increasing use in mediating the interaction between optical and electronic device phenomena1–6, hydrothermal synthesis has emerged as a scalable alternative to vapor-phase techniques7 for producing vertically-aligned ZnO nanowire (NW) arrays1,2 with excellent crystal quality at the wafer scale8–11 and on a range of substrates, under relatively benign temperature and pH conditions. Growth8,12–15 and seeding conditions9,11,16 have been used to achieve limited variation of NW length, diameter, and areal density. However, future applications will require simultaneous tailoring of multiple devicedependent critical dimensions such as NW length in dye-sensitized solar cells1,17, spacing and NW diameter in quantum dot photovoltaics (QDPV)4, aspect ratio in piezoelectric devices6,18, or NW tapering rate in field emission14,19. Recent interest in selected area growth (SAG) of ZnO NW arrays20–23 further reinforces the need for uniform growth rates over an entire substrate despite varying pattern fill factor24,25. However, the growth rate in regions with varying NW areal densities can vary significantly because of the reactant diffusion26 that, coincidentally, varies on similar lengthscales (100 nm – 1 µm) as those required for mediating light and charge interaction.

ZnO NW arrays are currently optimized via a time-consuming empirical approach, often termed “cook-and-look”, where the desired NW morphology is obtained through an iterative growth-characterization loop during bulk synthesis27. To transform the NW growth into a more deterministic process, it is critical to identify both the dominant growth mechanisms and the effect of diffusive transport vs. reaction. In particular, identifying the c-plane growth mechanism


2

Page 3 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

is important for controlling the NW growth rate14, impurity interaction28,29, and dopant incorporation30. Generally, solution-phase crystal growth mechanisms can be determined by observing the facet growth rate-supersaturation relationship31–34 or special surface features12,26,35– 37

. Applying this approach to ZnO requires constant Zn2+ and OH- concentrations, but typical

bulk synthesis8,10,27 – where a solution containing fixed amounts of reactants is sealed and heated – is unsuitable because reactant supersaturation decreases during growth due to consumption. Despite efforts to account for variation of reactant supersaturation during growth9,33,38, identifying the growth mechanism has been challenging14,27,35,36, mostly due to diverse Zn2+ speciation14,35,39. In contrast, continuous flow (CF) synthesis maintains homeostatic reactant supersaturation17,32, which makes it ideal for rigorous investigation, and ultimately control, of the NW growth mechanisms.

Even if the global reactant concentration is kept constant, the local environment of individual NWs has to be considered because the reactant concentration may vary spatially in the vicinity of different NWs15. The effect of diffusive reactant transport on morphology has often been invoked, albeit qualitatively8,19,26,38. Boercker et al.40 showed that NW length varied as the inverse square root of NW density when operating in a diffusion-limited growth regime, and Lee et al.15 observed Gibbs-Thomson effects in nanowire arrays with non-uniform diameters. It is thus necessary to make a distinction between the NW growth behavior observed in sparse vs. dense NW arrays (i.e. arrays grown using discrete seeds vs. mutually contacting seeds, respectively), since reactant transport to NW growth planes may obscure the actual growth rate variation with respect to the reactant concentration measured in the bulk solution.


3

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 28

In this study, we approach dimensional tailoring of ZnO NW arrays by first formulating a method to separate the effects of diffusive transport and reaction rate on observed NW growth rate. We used dynamic deposition spincoating as a robust, scalable method to vary NW areal density, achieving large area uniformity while maintaining excellent vertical alignment. In contrast to previous studies, we show that c-plane growth rate depends on the total area ratio of c-plane exposed to solution, which is particularly pertinent to NW growth on sparse or patterned seed layers. We applied a diffusion/reaction-limited analytical model to separate the effects of diffusion and reaction, helping us to identify the direct incorporation growth mechanism relevant for ZnO NW c-planes. The diffusion/reaction-limited model further yields a set of design guidelines for obtaining uniform NW growth rates by choosing synthesis conditions that are stable and rate-limiting. We demonstrate that surface adsorbent-limited growth using citrate ions gives homogeneous c-plane growth rates irrespective of the NW spacing (i.e. areal density), which we precisely define using electron-beam lithography. These design guidelines account for the interplay between transport and reaction, and will greatly facilitate bottom-up design of solution-phase processes for low-dimension metal oxide nanocrystal growth.

To control NW spacing over several orders of magnitude, we used dynamic deposition spincoating (Figure 1), wherein a low concentration of a zinc acetate sol-gel precursor was dispensed onto the substrate after reaching the targeted spin speed (see Methods and Supporting Information). Subsequent dewetting and high temperature annealing formed discrete ZnO seeds of ~ 5 nm diameter (Figure 1a, inset), with seed densities that decrease controllably over an order of magnitude with increasing spin speed (Figure 1a and Supporting Information Figure S1). Hydrothermal synthesis of ZnO NW arrays from these sparse seed layers yielded NW arrays


4

Page 5 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

aligned normal to the substrate (Figure 1c – f), indicating that the seeds were well aligned with the c-axis normal to the substrate41 due to the high temperature annealing step11.

Figure 1. Control over areal density of ZnO NW arrays. (a) Seed density vs. spincoating speed. Control of ZnO seed density between 4 × 1010 to 1.7 × 1012 cm-2 by varying a 10 mM sol-gel seed solution spincoating speed from 2000 to 10000 revolutions per minute (rpm). Error bars are standard error and omitted where smaller than the symbol size. Inset: Scanning electron microscope (SEM) image showing discrete ZnO seeds ~ 5 nm in diameter on Si (100) substrate. (b) Linear relationship observed between NW density and seed density; NW density is about two orders of magnitude lower than seed density. Dashed line is a guide to the eye. (c) – (f) Cross sectional SEM images of density-controlled ZnO NW arrays. NW arrays grown on Si(100) substrates at [Zn2+] = 10 mM, 90oC for 60 min on seed layers spincoated at 4000, 6000, 8000 and 10000 rpm, respectively. All micrographs at same scale.


5

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 28

Sparse and dense ZnO seeded Si substrates were subjected to hydrothermal synthesis in a homebuilt continuous flow reactor that maintains homeostatic reactant concentration and temperature growth conditions (Supporting Information Figure S2). For the pH range 4 – 5 and zinc concentrations between 5 – 20 mM, speciation plots show that the dominant zinc species in the growth solution was [Zn(H2O)6]2+ (Supporting Information Figure S3). We obtained wellaligned, uniform NW arrays comparable to those grown using bulk (non-homeostatic) synthesis, producing both sparse and dense ZnO NW arrays (Figure 1c – f) with excellent optical quality, as determined using photoluminescence (Supporting Figure S4). On density-controlled sparse arrays, NW density varies linearly with the seed density (Figure 1b), indicating a narrow seed grain size distribution42. The NW density is two orders of magnitude lower than the seed density, which we attribute to the seed dissolution-growth equilibrium in the initial stages of growth19,42.

Based on the range of NW densities obtained and shown in Figure 1b, we were able to control average NW spacing between 70 – 600 nm. The capability to change NW spacing in this range (100 nm – 1 µm)4,43 is key to enhancing QDPV device performance; it also has practical importance because NW density affects the net consumption of Zn2+ growth units, which in turn has an impact on the local Zn2+ concentration at the crystal-solution interface and thus NW cplane growth rate.

To identify the effects of changing growth parameters on the NW arrays, we quantified NW length, diameter, and array density from multiple regions on the substrate (Supporting Information Figure S5). Notably, we found that the c-plane growth rate does not increase with decreasing NW density per se (as previously found for dense seed layers and by assuming nanowires with the same facet area)19,40, but rather with the total area ratio of c-planes exposed to


6

Page 7 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

growth solution (Figure 2a – c). To quantify the total area of the growing crystal facet, we defined the unitless c-plane area ratio S, given by:

=

(1)

where ρ is the NW areal density (cm-2) and A is the mean NW c-facet area (cm2). This value gives an instantaneous measurement of the area ratio of substrate consisting of c-planes upon which Zn2+ growth units are impinging (Figs. 2a and b).

We grew a total of 40 substrates encompassing a range of growth and seeding parameters. Sparse seed layer density was varied from 4 × 1010 cm-2 to 1.7 × 1012 cm-2, and dense seed layers deposited by spincoating or sputtering (Supporting Information Figure S1e and f) were also used to eliminate any possible effects of the seed deposition method5. All samples were grown at [Zn2+] = 10 mM and temperature of 90oC to ensure identical reactant transport conditions. The cplane growth rate increases with decreasing S (Figure 2c), i.e. NW arrays with less total c-facet area exposed to growth solution grow faster than those with greater c-facet area. Interestingly, NW density does not affect the growth rate; both sparse and dense arrays show the same behavior such that NW arrays with the same c-plane area ratio S have similar c-plane growth rates. Therefore, a dense array of thin NWs (Figure 2a) grows at the same rate as a sparse array of thick NWs (Figure 2b), since they have similar S (points A and B, respectively, in Figure 2c). To extend the range of accessible S values, five samples were grown with various concentrations of added Al3+ up to 0.1 mM and six samples were grown with added Mg2+ up to 0.08 mM, both of which inhibit m-plane growth14 and yielded NW arrays with S below 0.1. This allowed us to


7

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 28

confirm that the relationship between growth rate and S extends to c-plane area ratios down to S=0.02, the lowest value measured in our work.

Figure 2. Diffusion/reaction-limited analysis applied to ZnO NW growth. Top down SEM images of (a) high and (b) low NW density arrays with similar c-plane area ratios S, shaded in yellow. (c) c-plane growth rate vs. c-plane area ratio. High (A) and low (B) NW density arrays (shown in (a) and (b), respectively) are indicated. The crossover from reaction-limited to diffusion-limited growth (dashed line) takes place at S = 0.11. Error bars are standard error and omitted where smaller than the symbol size. (d) Schematic showing serial processes of diffusion and reaction in the vicinity of the NW array growth front. Zn2+ diffusion (dotted arrow), adsorption, and subsequent incorporation (solid arrow) into NW top faces (c-plane). The local variation in Zn2+ concentration in the vicinity of the c-facets is shown for diffusion- and reactionlimited growth scenarios (left and right inset, respectively). Arrow thickness represents process rate; thinner arrow refers to a “slow” process.


8

Page 9 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

To explain this relationship, we adapted the diffusion/reaction-limited analysis40 that relates the serial processes of consumption of Zn2+ growth units at the ZnO crystal growth front, parameterized by c- and m-plane growth rates, and Zn2+ diffusive transport from the bulk solution (Figure 2d). We modified the model by combining m-plane growth rate and NW density under the c-plane area ratio S (see Supporting Information) to obtain the following equation:

=

= ; ≡ + 1 +

(2)

where Rc-plane is the c-plane growth rate, k1 is the surface reaction rate for dehydration of [Zn(H2O)]2+ (surface adsorbed) to ZnO (crystal), ρ is ZnO molar density (4.20 × 1028 m-3), C∞ is the bulk solution concentration of Zn2+, D is the Zn2+ aqueous tracer diffusion coefficient (2.91 × 10-9 m2s-1)44, δ is the stagnant layer thickness (the thickness of fluid adjacent to the substrate surface without flow and with only diffusive transport), and S is the c-plane area ratio. The Thiele modulus measures the ratio of surface reaction rate to reactant diffusion to the crystal growth front – since they are serial processes (Figure 2d), determines the degree to which growth is reaction- or diffusion-limited. By solving the Thiele equation for = 1, the S value for which the growth regime changes from reaction-limited to diffusion-limited can be determined. In our case ([Zn2+] = 10 mM at 90oC), this crossover point is at S = 0.11, as depicted in Figure 2d. Therefore, when the c-plane area ratio is above 0.11, the large area of c-planes at which reactants are being incorporated cause the growth rate to be constrained by reactant transport to the NW growth front (Figure 2d left inset). However, growth of NW arrays with a c-plane area


9

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 28

ratio below 0.11 is reaction-limited, i.e. transport of reactants to the NW growth front is faster than incorporation of reactants (Figure 2d right inset).

We next show the utility of this analysis in two applications: (1) fundamentally, we extract the actual variation in c-plane growth rate with Zn2+ concentration and demonstrate that the cplane grows by the direct incorporation mechanism; (2) practically, we achieve excellent NW height uniformity over an electron-beam lithography patterned NW array with a graded pitch by inducing reaction-limited NW growth with c-plane citrate adsorbents.

As mentioned above, measuring the c-facet growth rate as a function of supersaturation would allow unique identification of the growth mechanism. However, changing the Zn2+ concentration causes the diffusion environment to change because NW density and diameter also increase with Zn2+ concentration. Therefore, the individual contributions of each process (diffusion vs. reaction) to measured c-plane growth rate are indistinguishable. The diffusion/reaction analysis gives us a tool to separate these two processes, effectively accounting for the effect of variation in S on measured c-plane growth rates as Zn2+ concentration is changed. In practice, we remove the effect of diffusion by multiplying the measured c-plane growth rate Rc-plane by S (see Supporting Information). This leaves the desired relationship between solely reaction rate and Zn2+ supersaturation, which we term the diffusion-corrected growth rate (Rc-plane S).

We grew 36 NW arrays at Zn2+ concentrations between 5 – 20 mM on sparse and dense spincoated and sputtered seed layers (Supporting Information Figure S8). The diffusioncorrected c-plane growth rates of these arrays show a linear relationship with Zn2+ concentration


10

Page 11 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

(Figure 3a), indicative of growth by direct addition of Zn2+ growth units to the c-plane. The direct incorporation growth rate given by classical crystal growth theory34 is

&' = ! "#$ % () * − ,$

(3)

where a is the unit cell dimension (3.2 Å), d is the distance between kink sites (32 Å)34, ν is the vibration frequency (1013 Hz), νatom is the Zn2+ hydrated volume (74.01 Å3)45, ∆U is the Zn2+ dehydration barrier, kT is available thermal energy (3.02 kJ/mol at 90oC), and Ceqm is the equilibrium Zn2+ concentration. Fitting our data with Equation 3 (Figure 3a) gives a Zn2+ dehydration barrier of ∆U = 65 kJmol-1, which is in the range of dehydration barrier values for transition metal ions (40 – 80 kJ/mol, depending on aqueous charge and size, among other factors34). Together with speciation calculations (Supporting Information Figure S3), this result strongly suggests that it is indeed the [Zn(H2O)6]2+ species crystallizing on the c-facet. We also considered the layer-by-layer (LBL) growth mechanism by calculating the growth rate32 as a function of Zn2+ concentration (Figure 3a), but it varies non-linearly and greatly overestimates the c-plane growth rate at higher Zn2+ concentrations.


11

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 28

Figure 3. Identifying the c-plane growth mechanism. (a) Diffusion-corrected c-plane growth rate vs. Zn2+ concentration. Results from twenty growth runs are binned in 5 mM intervals for clarity; individual experiment plots can be found in SI. Error bars omitted where smaller than the symbol size. A linear fit to the data is shown in red. The calculated growth rate for layer-by-layer mechanism (gray) is a poor fit to the experimental results due to the exponential relationship with supersaturation. (b) SEM image showing a typical roughened NW top face.

We further bolster our experimental results by recalling that the diffusion/reaction-analysis (Equation (2)) also shows a linear relationship between c-plane growth rate (Rc-plane) and Zn2+ concentration (C∞). Using Equation S10 for the case of diffusion-limited growth, we calculate the expected gradient

.

/0

of 4.33 × 10-36 m4s-1 for the growth rate-concentration plot, which is

remarkably close to the experimentally determined value of 4.19 × 10-36 m4s-1. The close agreement between our experiments and values predicted by both macroscale reactant transport mechanisms as well as atomistic-scale classical crystal growth theory confirms that these analytical models accurately describe the ZnO NW growth system.


12

Page 13 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

The linear relationship between growth rate and supersaturation is not uncommon in aqueous growth of other inorganic semiconductor crystals34,46. Discrete interface models46–48 predict linearity at high supersaturation, where adatom surface diffusion becomes less significant than volume diffusion. In all these cases, growth units enter directly into kinks – crystal positions of high surface energy – that are present on the surface in high density. We confirmed a high kink density by observing high c-plane surface roughness using high resolution SEM (Figure 3b), which has also been observed by others10.

Identifying the c-facet growth mechanism means that the growth rate can be controlled by changing the Zn2+ concentration. This result also has implications for interaction of the c-facet with impurities because dopant incorporation frequently takes place at the c-facet due to its higher surface energy compared to the m-facet30,49. Impurity adsorption also offers the opportunity for growth control through modification of the growth rate. The variation of facet growth rate with additive concentration strongly depends on the facet growth mechanism28,50, and so this understanding will allow further tuning of growth rates via impurity addition, which we show next.

Diffusion/reaction analysis suggests that growth under reaction-limiting conditions yields growth rates independent of the nanowire diffusion environment. To test this hypothesis, we used electron-beam lithography to define a set of NW rows with varied distances (Figure 4a – c) – thereby locally tuned nanowire environment – and used citrate ion adsorbents to achieve reaction-limited growth by limiting incorporation of Zn2+ into the c-facet surface51. Four samples were grown under varying citrate concentrations from 0 to 7.5 µM, displaying shorter NWs, i.e. reduced growth rate, with increasing citrate concentration (Figure 4a). More quantitatively, from


13

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 28

Figure 4e and by using Supporting Information Equation (S8), we showed that the surface reaction rate constant k1 decreases with increasing citrate concentration (Table I), indicating that citrate adsorbents are indeed limiting the c-plane growth rate. The stagnant layer thickness δ and Thiele modulus also decrease (Table I), all of which confirm reaction-limited growth regime with less dominant diffusion (Figure 3d, right inset). Notably, the reaction-limited growth achieved here using c-plane adsorbent yields superior NW length uniformity (decrease in length variation from 41% to 16%) across the patterned substrate (Figure 4d and Supporting Information Figure S9), according to our expectation. This independence of NW growth rate from pattern fill factor is particularly useful for device applications where NW length must be uniformly controlled over the entire substrate, regardless of the NW density.


14

Page 15 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 4. Achieving uniform nanowire growth rates. (a) Cross sectional SEM images of NW arrays with graded pitch varying from 150 nm to 830 nm grown with increasing citrate concentration. Arrays grown at 1.25, 2.5, 5 and 7.5 µM citrate (artificially colored) are superposed on NW grown without citrate, showing decrease in NW length and greater uniformity with increasing citrate concentration. All growths performed with bulk synthesis: equimolar zinc nitrate and HMTA concentration of 2.5 mM, at 90oC for 100 min. (b) Top down


15

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 28

SEM image of a typical patterned substrate. Example micrograph of the six arrays patterned on each substrate and grown at a certain citrate concentration. (c) Top down SEM image of the cleaved edge (red arrows) of a substrate. The measured pitch between individual rows is as labelled. (d) NW length for each citrate concentration vs. measured pitch. Error bars represent standard error. Six arrays were grown for each citrate concentration, as shown in (b). (e) Reciprocal c-plane growth rate Rc-plane versus c-plane area ratio for each growth and linear fits.

Table I. k1 and δ values extracted from Figure 4e Citrate concentration (µM)

Surface reaction rate constant k1 (ms-1)

Stagnant layer thickness δ (mm)

Thiele modulus 1

0

7.42 × 10-6

2.62

0.46

1.25

5.95 × 10-6

1.14

0.28

2.5

4.84 × 10-6

0.84

0.24

5

4.14 × 10-6

0.61

0.23

7.5

3.27 × 10-6

0.35

0.14

In summary, we outlined a way to independently tailor all dimensions of ZnO NW arrays. By controlling the NW density over an order of magnitude, we showed that the NW c-plane growth rate greatly accelerates as the c-plane area ratio decreases, which we attribute to a transition from a diffusion-limited to a reaction-limited growth regime. Our diffusion/reaction analysis yields a linear relationship between c-plane growth rate and Zn2+ concentration, which is indicative of a direct incorporation mechanism operative on the ZnO c-plane, corroborated by both the reactant transport model and classical crystal growth theory. This analysis allowed us to tune the c-facet growth rate by changing Zn2+ concentration and by impurity adsorption, giving precise control of


16

Page 17 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

growth rates critical for obtaining NW arrays with uniform and controlled length. We then applied our model to achieve reaction-limited growth and consequently spatially uniform c-plane growth rates regardless of the NW areal density. This study emphasizes the importance of tuning the growth conditions to operate under rate-limiting conditions for precise growth rate control. Having this set of NW array design guidelines will streamline processing by facilitating optimization of NW array morphology for a wide variety of specific device purposes.

Methods

Si (100) or (111) substrates were cleaned for seeding with (1) DI rinse, (2) 10 min sonication in acetone, (3) 10 min in boiling IPA (4) 10 min in piranha solution (5) DI rinse, and dried with N2. Spincoated seed layers were deposited from a sol-gel solution consisting of 1:1 zinc acetate (99.999%) to monoethanolamine in 2-methoxyethanol (all chemicals from Sigma-Aldrich). Dense seed layers were deposited with a 0.3 M seed solution at 4000 rotations per minute (rpm) and annealed at 150oC in air; this process was done twice to yield a ~30 nm thick seed layer. To obtain sparse seed layers, dynamic deposition was used: 50 µL of a 0.01 M seed solution was dispensed onto the substrate once it had reached the targeted spin speed. The substrate was then annealed at 400oC in air. Sputtered seed layers were RF deposited on an AJA Orion 5 at 3 mTorr, 12 sccm Ar, 100 W. CF synthesis was carried out at a 1:1 zinc nitrate to hexamethylene tetramine (HMTA) ratio in a reactor maintained at [Zn2+] error < 0.5 mM and temperature error < 1oC. (HMTA) was used as the hydroxide source for its excellent pH stability12,52 between 4 – 5. Bulk synthesis was carried out at a 1:1 zinc nitrate to HMTA ratio at 2.5 mM, with 20 mg in 3ml of polyethyleneimine added. Zn2+ concentration was measured by taking filtered aliquots that were characterized with inductive coupled plasma-atomic emission spectroscopy (ICP-


17

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 28

AES), on a HORIBA Jobim Yvon ICP-AES with zinc standard samples (1000 ± 1 ppm, SigmaAldrich) at the 202.551 nm peak. Seeded substrates underwent a 1 min O2 plasma clean and were immersed after the growth solution had equilibrated for 40 min. After growth, the substrate was removed, rinsed with DI water, and blown dry with N2. Imaging was performed on an FEI Helios scanning electron microscope on both top-down and cleaved cross-sectional NW arrays. Electron-beam lithography was carried out on an Elionix ELS-F125 system at 125 kV accelerating voltage, 0.2 pA current, and 0.16 – 0.8 µs nm-2 dwell time.

Supporting Information

Supporting information contains details regarding seed layer deposition, continuous flow synthesis, nanowire photoluminescence spectra, nanowire array characterization, diffusion/reaction analysis, nanowire growth mechanism study and surface-adsorbent reactionlimited growth. This material is available free of charge via the Internet at http://pubs.acs.org.

Corresponding Author

Address correspondence to: [email protected]

Author Contributions

S.G. and J.J.C. designed and carried out NW growth experiments. K.K.B. and S.M.N. designed and patterned electron-beam lithography templates for graded-pitch NW studies. All authors discussed the results and commented on the manuscript.


18

Page 19 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Acknowledgements

This work was supported by a Massachusetts Institute of Technology Energy Initiative grant. J.J.C. acknowledges support from the Agency for Science, Technology and Research, Singapore. S.M.N. and K.K.B. acknowledges support from the National Science Foundation Scalable Nanomanufacturing 12-544. Authors would like to thank Dr. Sehoon Chang and Sema Ermez for their help and advice.

References

(1)

Law, M.; Greene, L. E.; Johnson, J. C.; Saykally, R.; Yang, P. Nanowire Dye-Sensitized Solar Cells. Nat. Mater. 2005, 4, 455–459.

(2)

Chu, S.; Wang, G.; Zhou, W.; Lin, Y.; Chernyak, L.; Zhao, J.; Kong, J.; Li, L.; Ren, J.; Liu, J. Electrically Pumped Waveguide Lasing from ZnO Nanowires. Nat. Nanotechnol. 2011, 6, 506–510.

(3)

Jean, J.; Chang, S.; Brown, P. R.; Cheng, J. J.; Rekemeyer, P. H.; Bawendi, M. G.; Gradečak, S.; Bulović, V. ZnO Nanowire Arrays for Enhanced Photocurrent in PbS Quantum Dot Solar Cells. Adv. Mater. 2013, 25, 2790–2796.

(4)

Kramer, I. J.; Zhitomirsky, D.; Bass, J. D.; Rice, P. M.; Topuria, T.; Krupp, L.; Thon, S. M.; Ip, A. H.; Debnath, R.; Kim, H.-C.; et al. Ordered Nanopillar Structured Electrodes for Depleted Bulk Heterojunction Colloidal Quantum Dot Solar Cells. Adv. Mater. 2012, 24, 2315–2319.

(5)

Nicaise, S.; Cheng, J.; Kiani, A.; Gradecak, S.; Berggren, K. Control of Zinc Oxide Nanowire Array Properties with Electron Beam Lithography Templating for PV Applications. Nanotechnology 2015, 26, 075303–075311.

(6)

Wang, Z. L.; Song, J. Piezoelectric Nanogenerators Based on Zinc Oxide Nanowire Arrays. Science (80-. ). 2006, 312, 242–246.

(7)

Brewster, M. M.; Zhou, X.; Lu, M.-Y.; Gradečak, S. The Interplay of Structural and Optical Properties in Individual ZnO Nanostructures. Nanoscale 2012, 4, 1455–1462.

(8)

Vayssieres, L. Growth of Arrayed Nanorods and Nanowires of ZnO from Aqueous Solutions. Adv. Mater. 2003, 15, 464–466.


19

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 28

(9)

Tak, Y.; Yong, K. Controlled Growth of Well-Aligned ZnO Nanorod Array Using a Novel Solution Method. J. Phys. Chem. B 2005, 109, 19263–19269.

(10)

Greene, L. E.; Law, M.; Goldberger, J.; Kim, F.; Johnson, J. C.; Zhang, Y.; Saykally, R. J.; Yang, P. Low-Temperature Wafer-Scale Production of ZnO Nanowire Arrays. Angew. Chemie (International Ed. 2003, 42, 3031–3034.

(11)

Greene, L. E.; Law, M.; Tan, D. H.; Montano, M.; Goldberger, J.; Somorjai, G.; Yang, P. General Route to Vertical ZnO Nanowire Arrays Using Textured ZnO Seeds. Nano Lett. 2005, 5, 1231–1236.

(12)

Liu, X.; Jin, Z.; Bu, S.; Zhao, J.; Liu, Z. Growth of ZnO Films with Controlled Morphology by Aqueous Solution Method. J. Am. Ceram. Soc. 2006, 89, 1226–1231.

(13)

Chen, L.-Y.; Yin, Y.-T.; Chen, C.-H.; Chiou, J.-W. Influence of Polyethyleneimine and Ammonium on the Growth of ZnO Nanowires by Hydrothermal Method. J. Phys. Chem. C 2011, 115, 20913–20919.

(14)

Joo, J.; Chow, B. Y.; Prakash, M.; Boyden, E. S.; Jacobson, J. M. Face-Selective Electrostatic Control of Hydrothermal Zinc Oxide Nanowire Synthesis. Nat. Mater. 2011, 10, 596–601.

(15)

Lee, J. M.; No, Y.-S.; Kim, S.; Park, H.-G.; Park, W. Il. Strong Interactive Growth Behaviours in Solution-Phase Synthesis of Three-Dimensional Metal Oxide Nanostructures. Nat. Commun. 2015, 6, 6325–6332.

(16)

Huang, J.-S.; Lin, C.-F. Influences of ZnO Sol-Gel Thin Film Characteristics on ZnO Nanowire Arrays Prepared at Low Temperature Using All Solution-Based Processing. J. Appl. Phys. 2008, 103, 014304–014309.

(17)

Chen, L.-Y.; Yin, Y.-T. Facile Continuous Flow Injection Process for High Quality Long ZnO Nanowire Arrays Synthesis. Cryst. Growth Des. 2012, 12, 1055–1059.

(18)

Xu, S.; Qin, Y.; Xu, C.; Wei, Y.; Yang, R.; Wang, Z. L. Self-Powered Nanowire Devices. Nat. Nanotechnol. 2010, 5, 366–373.

(19)

Liu, J.; She, J.; Deng, S.; Chen, J.; Xu, N. Ultrathin Seed-Layer for Tuning Density of ZnO Nanowire Arrays and Their Field Emission Characteristics. J. Phys. Chem. C 2008, 112, 11685–11690.

(20)

Alenezi, M. R.; Henley, S. J.; Silva, S. R. P. On-Chip Fabrication of High Performance Nanostructured ZnO UV Detectors. Sci. Rep. 2015, 5, 8516–8524.

(21)

Zhang, D.; Wang, S.; Cheng, K.; Dai, S.; Hu, B.; Han, X.; Shi, Q.; Du, Z. Controllable Fabrication of Patterned ZnO Nanorod Arrays: Investigations into the Impacts on Their Morphology. ACS Appl. Mater. Interfaces 2012, 4, 2969–2977.


20

Page 21 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

(22)

Jung, M.-H.; Lee, H. Selective Patterning of ZnO Nanorods on Silicon Substrates Using Nanoimprint Lithography. Nanoscale Res. Lett. 2011, 6, 159–169.

(23)

Erdelyi, R.; Nagata, T.; Rogers, D. J.; Teherani, F. H.; Horvath, Z. E.; Labadi, Z.; Baji, Z.; Wakayama, Y.; Volk, J. Investigations into the Impact of the Template Layer on ZnO Nanowire Arrays Made Using Low Temperature Wet Chemical Growth. Cryst. Growth Des. 2011, 11, 2515–2519.

(24)

Kim, Y.-J.; Lee, C.-H.; Hong, Y. J.; Yi, G.-C.; Kim, S. S.; Cheong, H. Controlled Selective Growth of ZnO Nanorod and Microrod Arrays on Si Substrates by a Wet Chemical Method. Appl. Phys. Lett. 2006, 89, 163128–163130.

(25)

Wei, Y.; Wu, W.; Guo, R.; Yuan, D.; Das, S.; Wang, Z. L. Wafer-Scale High-Throughput Ordered Growth of Vertically Aligned ZnO Nanowire Arrays. Nano Lett. 2010, 10, 3414– 3419.

(26)

McPeak, K. M.; Baxter, J. B. ZnO Nanowires Grown by Chemical Bath Deposition in a Continuous Flow Microreactor. Cryst. Growth Des. 2009, 9, 4538–4545.

(27)

Xu, S.; Adiga, N.; Ba, S.; Dasgupta, T.; Wu, C. F. J.; Wang, Z. L. Optimizing and Improving the Growth Quality of ZnO Nanowire Arrays Guided by Statistical Design of Experiments. ACS Nano 2009, 3, 1803–1812.

(28)

Kubota, N.; Mullin, J. W. A Kinetic Model for Crystal Growth from Aqueous Solution in the Presence of Impurity. J. Cryst. Growth 1995, 152, 203–208.

(29)

Martins, P. M.; Rocha, F. A.; Rein, P. The Influence of Impurities on the Crystal Growth Kinetics according to a Competitive Adsorption Model. Cryst. Growth Des. 2006, 6, 2814–2821.

(30)

Tay, C. B.; Tang, J.; Nguyen, X. S.; Huang, X. H.; Chai, J. W.; Venkatesan, V. T.; Chua, S. J. Low Temperature Aqueous Solution Route to Reliable P - Type Doping in ZnO with K: Growth Chemistry , Doping Mechanism , and Thermal Stability. J. Phys. Chem. C 2012, 116, 24239–24247.

(31)

Chernov, A. A.; Rashkovich, L. N.; Mkrtchan, A. A. Solution Growth Kinetics and Mechanism - Prismatic Face of ADP. J. Cryst. Growth 1986, 74, 101–112.

(32)

Morin, S. A.; Bierman, M. J.; Tong, J.; Jin, S. Mechanism and Kinetics of Spontaneous Nanotube Growth Driven by Screw Dislocations. Science (80-. ). 2010, 328, 476–480.

(33)

Govender, K.; Boyle, D. S.; Kenway, P. B.; Brien, P. O. Understanding the Factors That Govern the Deposition and Morphology of Thin Films of ZnO from Aqueous Solution. J. Mater. Chem. 2004, 14, 2575–2591.


21

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 28

(34)

Markov, I. V. Crystal Growth for Beginners; 2nd ed.; World Scientific Publishing Co. Pte Ltd: Singapore, 2004.

(35)

Baruah, S.; Dutta, J. pH-Dependent Growth of Zinc Oxide Nanorods. J. Cryst. Growth 2009, 311, 2549–2554.

(36)

Sun, Y.; Riley, D. J.; Ashfold, M. N. R. Mechanism of ZnO Nanotube Growth by Hydrothermal Methods on ZnO Film-Coated Si Substrates. J. Phys. Chem. B 2006, 110, 15186–15192.

(37)

Li, Q.; Kumar, V.; Li, Y.; Zhang, H.; Marks, T. J.; Chang, R. P. H. Fabrication of ZnO Nanorods and Nanotubes in Aqueous Solutions. Chem. Mater. 2005, 17, 1001–1006.

(38)

Kozhummal, R.; Yang, Y.; Guder, F.; Hartel, A.; Lu, X.; Kucukbayrak, U. M.; MateoAlonso, A.; Elwenspoek, M.; Zacharias, M. Homoepitaxial Branching: An Unusual Polymorph of Zinc Oxide Derived from Seeded Solution Growth. ACS Nano 2012, 6, 7133–7141.

(39)

Richardson, J. J.; Lange, F. F. Controlling Low Temperature Aqueous Synthesis of ZnO. 1. Thermodynamic Analysis. Cryst. Growth Des. 2009, 9, 2570–2575.

(40)

Boercker, J. E.; Schmidt, J. B.; Aydil, E. S. Transport Limited Growth of Zinc Oxide Nanowires. Cryst. Growth Des. 2009, 9, 2783–2789.

(41)

Wu, W.-Y.; Yeh, C.-C.; Ting, J.-M. Effects of Seed Layer Characteristics on the Synthesis of ZnO Nanowires. J. Am. Ceram. Soc. 2009, 92, 2718–2723.

(42)

Xu, C.; Gao, D. Two-Stage Hydrothermal Growth of Long ZnO Nanowires for Efficient TiO 2 Nanotube-Based Dye-Sensitized Solar Cells. J. Phys. Chem. C 2012, 116, 7236– 7241.

(43)

Kramer, I.; Sargent, E. Colloidal Quantum Dot Photovoltaics: A Path Forward. ACS Nano 2011, 5, 8506–8514.

(44)

Oelkers, E. H.; Helgeson, H. C. Calculation of the Thermodynamic and Transport Properties of Aqueous Species at High Pressures and Temperatures: Aqueous Tracer Diffusion Coefficients of Ions to 1000°C and 5 Kb. Geochim. Cosmochim. Acta 1988, 52, 63–85.

(45)

Persson, I. Hydrated Metal Ions in Aqueous Solution: How Regular Are Their Structures? Pure Appl. Chem. 2010, 82, 1901–1917.

(46)

Chernov, A. A. Notes on Interface Growth Kinetics 50 Years after Burton, Cabrera and Frank. J. Cryst. Growth 2004, 264, 499–518.


22

Page 23 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

(47)

Burton, W. K.; Cabrera, N.; Frank, F. C. The Growth of Crystals and the Equilibrium Structure of Their Surfaces. Philos. Trans. R. Soc. A Math. Phys. Eng. Sci. 1951, 243, 299–358.

(48)

Bennema, P. The Importance of Surface Diffusion for Crystal Growth from Solution. J. Cryst. Growth 1969, 5, 29–43.

(49)

Pradel, K. C.; Wu, W.; Ding, Y.; Wang, Z. L. Solution-Derived ZnO Homojunction Nanowire Films on Wearable Substrates for Energy Conversion and Self-Powered Gesture Recognition. Nano Lett. 2014, 14, 6897–6905.

(50)

Sangwal, K. Effect of Impurities on Crystal Growth Processes. Prog. Cryst. Growth Charact. 1996, 32, 3–43.

(51)

Tian, Z. R.; Voigt, J. a; Liu, J.; McKenzie, B.; McDermott, M. J.; Rodriguez, M. a; Konishi, H.; Xu, H. Complex and Oriented ZnO Nanostructures. Nat. Mater. 2003, 2, 821–826.

(52)

McPeak, K. M.; Le, T. P.; Britton, N. G.; Nickolov, Z. S.; Elabd, Y. a; Baxter, J. B. Chemical Bath Deposition of ZnO Nanowires at near-Neutral pH Conditions without Hexamethylenetetramine (HMTA): Understanding the Role of HMTA in ZnO Nanowire Growth. Langmuir 2011, 27, 3672–3677.

Table of Contents Graphic and Synopsis Pattern-independent solution-phase zinc oxide nanowire growth by controllably shifting growth conditions into reaction-limited regime using surface adsorbents


23

Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1 170x113mm (300 x 300 DPI)


Page 24 of 28

Page 25 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 2 170x109mm (300 x 300 DPI)


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3 154x74mm (300 x 300 DPI)


Page 26 of 28

Page 27 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 4 170x192mm (300 x 300 DPI)


Nano Letters

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

TOC 79x24mm (300 x 300 DPI)


Page 28 of 28

Dimensional Tailoring of Hydrothermally Grown Zinc Oxide Nanowire

Recommend Documents