Top Banner
Number Density and Diameter Control of Chemical Bath Deposition of ZnO Nanorods on FTO by Forced Hydrolysis of Seed Crystals Venkata Manthina, ,§ Tulsi Patel, and Alexander G. Agrios ,§,Department of Civil & Environmental Engineering, University of Connecticut, 261 Glenbrook Rd Unit 3037, Storrs, Connecticut 06269 § Center for Clean Energy Engineering, University of Connecticut, 44 Weaver Rd, Storrs, Connecticut 06269 Department of Chemical & Biomolecular Engineering, University of Connecticut, Unit 3222, 191 Auditorium Rd, Storrs, Connecticut 06269 ZnO nanorods have been studied extensively due to facile syn- thesis and useful optoelectronic properties for applications in nanoscale devices. In a common two-step procedure, an ethan- olic Zn 2+ precursor solution is used to deposit ZnO seed crys- tals on a substrate, which is then immersed in an aqueous Zn 2+ precursor solution to grow the nanorods. Here, a forced hydrolysis technique was employed based on additions of water and heat to the seed precursor solution before depositing the seeds on commercial fluorine-doped tin oxide (FTO)/glass sub- strates. ZnO nanorods were then grown from these seeds by chemical bath deposition. Analyses showed that the forced hydrolysis resulted in an increase in seed crystallite size and a decrease in the number of seeds deposited. With increasing seed size, the number density of nanorods decreased, while the length and diameter of each rod increased. These findings offer a simple method for exerting control over the number density of ZnO nanorods that is compatible with the rough FTO sur- face, unlike other methods that require smoother substrates. I. Introduction ZnO is a functional material with many applications due to its electrical, 1 photochemical, 2 thermoelectric, 3 piezoelectric, 4,5 and optical 6 properties. ZnO can be grown into a wide variety of nanostructures like nanoparticles, nanorods, nanowires, nanobelts, nanotubes, and so forth due to its anisotropic growth and ease of crystallization. One-dimensional growth of ZnO is studied extensively due to its potential use in solar cells, 7,8 chemical sensors, 9 field effect transistors, 10 light emit- ting diodes, 11 piezoelectric nanogenerators, 12 and laser diodes. 13 Many useful properties of ZnO nanorods are dependent on the rod diameter, length, and number density. The surface band bending of the ZnO nanorods is dependent on the diame- ter of the ZnO nanorods. 14 The performance of the photoelect- rochemical cells for water splitting and photovoltaics depend on the rod morphology. 3,15 In hybrid solar cells, a low density of ZnO nanorods on the photoanode is needed to provide void space for P3HT:PCBM infiltration. 16 For applications in sur- face acoustic wave devices growth along the a-axis is required. 17 In the fabrication of composite nanostructures such as nanoparticle-coated ZnO nanorods, tuning the density of the nanorods is required for balancing total surface area against effective coverage of the nanoparticles throughout the length of the nanorod. 1820 The optoelectronic and gas sensing properties of the ZnO are dependent on the shape of nanostructures. 21 ZnO nanorods can be grown in situ from a variety of sub- strates through procedures such as chemical bath deposition (CBD), 22,23 electrodeposition, 2426 chemical vapor deposi- tion, 27,28 thermal deposition, 29 and pulse laser deposition. 30 Among these CBD has the advantage of being low tempera- ture, low cost, suitable for large area processing for indus- trial applications, and environmentally benign. CBD is a two-step process, beginning with the deposition of a “seed layer” of ZnO nanocrystals on the substrate, from which nanorods are subsequently grown in the “chemical bath,” an aqueous solution of a zinc precursor with additives to mod- ify the growth. In CBD, increasing the concentrations of precursors in the chemical bath increases both nanorod diameter and length, but has little effect on the number density of the rods. As precursor concentrations are increased, this results in the rods fusing together. To obtain larger rods while maintaining space between them, control over the number density is required. Some density control has been achieved by control- ling the thickness of seed layers made by physical methods such as sputtering. 31 Other methods have been devised that work on smooth surfaces like indium tin oxide (ITO), 32 glass, and Si substrates, 31 but we have found them ineffective on the rougher surface of fluorine-doped tin oxide (FTO), which is preferred to ITO in applications requiring tolerance of high temperatures. Here, we show how the deposition of seed crystals on FTO from an ethanolic precursor solution can be modified by low-temperature thermal treatment combined with controlled additions of water to yield control over the number density of the resulting ZnO nanorods. II. Experimental Procedure (1) Reagents and Materials Zinc nitrate hexahydrate, hexamethylenetetramine (HMTA), polyethyleneimine, and ethanol were purchased from Sigma- Aldrich (St. Louis, MO) and were ACS grade. SnO 2 :F glass (FTO, transmission >80% in the visible spectrum; sheet resis- tance 8 O/) was purchased from Hartford Tec Glass (Hart- ford City, IN). (2) ZnO Seed Crystal Preparation The formation of ZnO seeds began with the dissolution of 5 mM zinc acetate dihydrate in ethanol by stirring for 3 h at room temperature. 33 Varying amounts of deionized water were added. The solution was stirred for another 3.5 h and R. Riman—contributing editor Manuscript No. 34016. Received October 24, 2013; approved December 23, 2013. Author to whom correspondence should be addressed. e-mail: [email protected] 1028 J. Am. Ceram. Soc., 97 [4] 1028–1034 (2014) DOI: 10.1111/jace.12819 © 2014 The American Ceramic Society J ournal
7
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: ZnO nanorod density control

Number Density and Diameter Control of Chemical Bath Deposition of ZnONanorods on FTO by Forced Hydrolysis of Seed Crystals

Venkata Manthina,‡,§ Tulsi Patel,¶ and Alexander G. Agrios‡,§,†

‡Department of Civil & Environmental Engineering, University of Connecticut, 261 Glenbrook Rd Unit 3037,Storrs, Connecticut 06269

§Center for Clean Energy Engineering, University of Connecticut, 44 Weaver Rd, Storrs, Connecticut 06269

¶Department of Chemical & Biomolecular Engineering, University of Connecticut, Unit 3222, 191 Auditorium Rd,Storrs, Connecticut 06269

ZnO nanorods have been studied extensively due to facile syn-

thesis and useful optoelectronic properties for applications in

nanoscale devices. In a common two-step procedure, an ethan-olic Zn2+ precursor solution is used to deposit ZnO seed crys-

tals on a substrate, which is then immersed in an aqueous

Zn2+

precursor solution to grow the nanorods. Here, a forcedhydrolysis technique was employed based on additions of water

and heat to the seed precursor solution before depositing the

seeds on commercial fluorine-doped tin oxide (FTO)/glass sub-

strates. ZnO nanorods were then grown from these seeds bychemical bath deposition. Analyses showed that the forced

hydrolysis resulted in an increase in seed crystallite size and a

decrease in the number of seeds deposited. With increasing

seed size, the number density of nanorods decreased, while thelength and diameter of each rod increased. These findings offer

a simple method for exerting control over the number density

of ZnO nanorods that is compatible with the rough FTO sur-

face, unlike other methods that require smoother substrates.

I. Introduction

ZnO is a functional material with many applications due toits electrical,1 photochemical,2 thermoelectric,3 piezoelectric,4,5

and optical6 properties. ZnO can be grown into a wide varietyof nanostructures like nanoparticles, nanorods, nanowires,nanobelts, nanotubes, and so forth due to its anisotropicgrowth and ease of crystallization. One-dimensional growthof ZnO is studied extensively due to its potential use in solarcells,7,8 chemical sensors,9 field effect transistors,10 light emit-ting diodes,11 piezoelectric nanogenerators,12 and laserdiodes.13

Many useful properties of ZnO nanorods are dependent onthe rod diameter, length, and number density. The surfaceband bending of the ZnO nanorods is dependent on the diame-ter of the ZnO nanorods.14 The performance of the photoelect-rochemical cells for water splitting and photovoltaics dependon the rod morphology.3,15 In hybrid solar cells, a low densityof ZnO nanorods on the photoanode is needed to provide voidspace for P3HT:PCBM infiltration.16 For applications in sur-face acoustic wave devices growth along the a-axis isrequired.17 In the fabrication of composite nanostructuressuch as nanoparticle-coated ZnO nanorods, tuning the densityof the nanorods is required for balancing total surface area

against effective coverage of the nanoparticles throughout thelength of the nanorod.18–20 The optoelectronic and gassensing properties of the ZnO are dependent on the shape ofnanostructures.21

ZnO nanorods can be grown in situ from a variety of sub-strates through procedures such as chemical bath deposition(CBD),22,23 electrodeposition,24–26 chemical vapor deposi-tion,27,28 thermal deposition,29 and pulse laser deposition.30

Among these CBD has the advantage of being low tempera-ture, low cost, suitable for large area processing for indus-trial applications, and environmentally benign. CBD is atwo-step process, beginning with the deposition of a “seedlayer” of ZnO nanocrystals on the substrate, from whichnanorods are subsequently grown in the “chemical bath,” anaqueous solution of a zinc precursor with additives to mod-ify the growth.

In CBD, increasing the concentrations of precursors in thechemical bath increases both nanorod diameter and length,but has little effect on the number density of the rods. Asprecursor concentrations are increased, this results in therods fusing together. To obtain larger rods while maintainingspace between them, control over the number density isrequired. Some density control has been achieved by control-ling the thickness of seed layers made by physical methodssuch as sputtering.31 Other methods have been devised thatwork on smooth surfaces like indium tin oxide (ITO),32 glass,and Si substrates,31 but we have found them ineffective onthe rougher surface of fluorine-doped tin oxide (FTO), whichis preferred to ITO in applications requiring tolerance ofhigh temperatures. Here, we show how the deposition of seedcrystals on FTO from an ethanolic precursor solution can bemodified by low-temperature thermal treatment combinedwith controlled additions of water to yield control over thenumber density of the resulting ZnO nanorods.

II. Experimental Procedure

(1) Reagents and MaterialsZinc nitrate hexahydrate, hexamethylenetetramine (HMTA),polyethyleneimine, and ethanol were purchased from Sigma-Aldrich (St. Louis, MO) and were ACS grade. SnO2:F glass(FTO, transmission >80% in the visible spectrum; sheet resis-tance 8 O/□) was purchased from Hartford Tec Glass (Hart-ford City, IN).

(2) ZnO Seed Crystal PreparationThe formation of ZnO seeds began with the dissolution of5 mM zinc acetate dihydrate in ethanol by stirring for 3 h atroom temperature.33 Varying amounts of deionized waterwere added. The solution was stirred for another 3.5 h and

R. Riman—contributing editor

Manuscript No. 34016. Received October 24, 2013; approved December 23, 2013.†Author to whom correspondence should be addressed. e-mail: [email protected]

1028

J. Am. Ceram. Soc., 97 [4] 1028–1034 (2014)

DOI: 10.1111/jace.12819

© 2014 The American Ceramic Society

Journal

Page 2: ZnO nanorod density control

then transferred into a preheated oven at 80°C for 12 h.After cooling to room temperature, this finished seed precur-sor solution was drop-cast on a precleaned FTO substrate,which was allowed to dry for 30 s and then rinsed with etha-nol. The substrate was then heated on a titanium hot plateat 350°C for 20 min to form the ZnO seed layer. The proce-dure was repeated two additional times to increase the num-ber of seeds on the substrate. Some samples, with 0.9M H2Oadded to the seed precursor solution, were aged for 24 hbetween the 80°C heating and drop-casting steps to examinethe effect of increased time for crystal growth on subsequentZnO morphology.

(3) ZnO Nanorod SynthesisNanorods were grown from the ZnO seeds by CBD. Theseeded FTO substrates were placed in 100 mL of an aqueoussolution of 50 mM zinc nitrate hexahydrate and HMTA and6 mM polyethyleneimine in a 100-mL glass bottle at 90°Cfor 24 h.

(4) CharacterizationThe morphology of ZnO seed crystals and nanorods wasinvestigated by scanning electron microscopy using an FEI(Hilsboro, OR) Quanta FEG250 SEM in high vacuum modeand characterized by X-ray powder diffraction (XRD) usinga Bruker (Billerica, MA) D8 Advance X-ray diffractometerusing CuK radiation (k = 0.154 178 nm) at a scanning rateof 0.04°/s in the 2h range from 10° to 90°. The data wereanalyzed using the Debye–Scherrer equation to determine theZnO seed particle size. Photoluminescence measurementswere performed using a Horbia Jobin Yvon (Kyoto, Japan)Fluorolog-3 spectrofluorometer with an excitation wave-length of 350 nm and a scan rate of 1 nm/s.

(5) Zn QuantificationZnO-seeded FTO substrates were etched for 6 h with 1%HNO3(aq) to dissolve the zinc. No seeds could be seen bySEM after this etching step. The etched zinc solution waskept in polypropylene vials (Corning Centristar, Acton, MA)and analyzed by inductively coupled plasma–mass spectrome-try (ICP–MS) using an Agilent Technologies (Santa Clara,CA) 7700 series instrument run in He mode and in a matrixof 1% HNO3(aq).

III. Results

(1) Forced Hydrolysis of ZnO SeedsTypical CBD of ZnO nanorods begins with the formation ofsmall (2–3 nm) seed crystals of ZnO by distributing a zinc(II)salt on a substrate, then heating it in air. We increased thesize of the seeds by forced hydrolysis: the addition of con-trolled amounts of water to the ethanolic solution of zinc(II)acetate dihydrate followed by heating.34 Hydrolysis formszinc hydroxides by the following stoichiometry:35

ZnðCH3COOÞ2þ2H2O!HO�Zn�OHþ2CH3COOH

Crystal growth of ZnO can be visualized as Zn(OH�)42� ions

gathering into zinc oxyhydroxy clusters.36

In this work, the concentration of water added to the seedprecursor solution, [H2O]SPS, was varied from 0 to 1M. Itshould be noted that even without added water, some wateris always present from the zinc acetate dihydrate salt.

Scanning electron microscopy images of ZnO seed crystalsdeposited on FTO are shown in Fig. 1. The seed crystals canbe distinguished as the smaller features in the image; thelarge-scale features are from the roughness of the FTO sur-face. It can be seen that as [H2O]SPS was increased from 0 to

1M, the size of the seeds increased, while the number ofseeds decreased.

The properties of the seeds were examined by powderX-ray diffraction data, presented in Fig. 2. All diffractionpeaks can be attributed either to the FTO substrate or tocrystalline ZnO (space group: P63mc(186); a = 0.3249 nm,c = 0.5206 nm), in agreement with the standard pattern forZnO (JCPDS 065-2880). For seeds made using increasingamounts of water, the degree of peak broadening decreases,indicating larger crystallite sizes.

We calculated crystallite size based on the full width athalf maximum of the XRD peak for ZnO (111) using theScherrer equation, with the results shown in Fig. 3. Whenseeds were prepared with neither added water nor heat treat-ment, the resulting size of the seeds was 4–5 nm. Heating theseed solution in an 80°C oven for 12 h increased the seed sizeto 25 nm. Addition of water further increased the seed sizeto as much as 35 nm for a seed precursor solution with 1Mwater.

ZnO nanorods were grown on the ZnO-seeded FTO sub-strates described in the preceding section by immersing themin an aqueous chemical bath for 24 h at 90°C. XRD data forthe nanorods on FTO are shown in Fig. 4. All diffractionpeaks can be attributed to FTO or to wurtzite ZnO, in agree-ment with the standard pattern for 1-D ZnO (JCPDS 070-8070). The intensity of the peak assigned to the (002) planeindicates that ZnO nanorods were formed through elonga-tion along the c-axis perpendicularly to the substrate.

Optical characterization by photoluminescence spectros-copy (see Fig. 5) indicates bandgap excitation similar to bulkZnO at 380 nm, indicating no quantum confinement effects.The atom ratio of Zn:O (0.99:1), analyzed by energy disper-sive X-ray spectroscopy, equals the stoichiometric ZnO ratio.

Scanning electron microscopy images (Fig. 6) of the sam-ples produced by CBD confirm that the ZnO grew in a nano-rod morphology under all tested concentrations of water inthe seed precursor solution, in agreement with the XRDdata. The diameter, length, and number density of the nano-rods were determined from these images. From these data,we calculated the aspect ratio (ratio of length to diameter)and roughness factor (ratio of surface area to projectedarea). All of these results are shown in Fig. 7. Both the

(a) (b)

(c) (d)

Fig. 1. Scanning electron microscopy image of ZnO seeds on theFTO substrate with different concentrations of water in the seedprecursor solution: (a) 0, (b) 0.5, (c) 0.8, and (d) 1M. Some seedcrystals are highlighted by yellow circles.

April 2014 Forced Hydrolysis of ZnO Seeds 1029

Page 3: ZnO nanorod density control

diameter [Fig. 7(a)] and length [Fig. 7(b)] of the nanorodsincreased monotonically with increasing water concentrationin the seed precursor solution (i.e., with larger ZnO seeds),while the number of nanorods per unit area [Fig. 7(c)]decreased.

(A) ZnO Seed Quantification: The amount of zinc onZnO-seeded FTO substrates was quantified by dissolving thezinc as Zn2+ in water with 1% nitric acid and analyzing thezinc by ICP–MS. Although equal amounts of seed solutionare drop-cast on each substrate, the subsequent ethanol rinsemeans that the amount of Zn remaining on the substrate canvary. Assuming that all Zn came from ZnO seed crystals,taking the XRD crystallite size the typical seed crystal sizeand assuming spherical crystals, we estimated the numberdensity of seed crystals on each substrate. Our analysesshowed that the amount of Zn etched from each seeded sam-ple decreased with increasing amounts of water in the seedprecursor solution. Seed size increases along the same trend,and the net result is a decreasing number density of seedcrystals as water is added to the precursor solution. Theresults are shown in Table I. For comparison, the numberdensity of nanorods grown from a given seed layer prepara-tion is also given. Clearly, each seed does not become a

nanorod, but the number of nanorods correlates to the num-ber of seeds.

(B) ZnO Seed Solution Aging: Some ZnO seed pre-cursor solutions were allowed to stand for 24 h after theforced hydrolysis to allow growth of seed crystals over time.The aged seeds as deposited on FTO are shown in Fig. 8(a).CBD from aged seeds resulted in a variety of rod shapes ofwhich examples are shown in Figs. 8(b) and (c).

IV. Discussion

(1) Density of ZnO NanorodsSeeds are used before the crystal growth of the nanorod topromote heterogeneous nucleation as homogenous nucleation

(a)(b)

Fig. 2. (a) X-ray diffraction pattern of ZnO nanoseeds on FTO substrate and (b) magnification of the ZnO (111) peak.

Fig. 3. Change in crystallite size (by X-ray powder diffraction) withthe concentration of the water in the seed precursor solution.

Fig. 4. X-ray diffraction pattern of ZnO nanorods on FTO.

1030 Journal of the American Ceramic Society—Manthina et al. Vol. 97, No. 4

Page 4: ZnO nanorod density control

requires a high activation energy barrier. Thus, seeds allowcrystal growth to take place at a low saturation ratio.37 Theseed layer is of crucial importance in determining the type ofmorphology and the dimensions of ZnO nanostructuresgrown by CBD.38 In this work, the seeds were altered by theaddition of varying amounts of water in the ethanolic seedprecursor solution. The seed size increases as more water isadded, presumably by accelerating the hydrolysis of zinc ace-tate to form ZnO. Furthermore, the addition of waterincreases the polarity of the precursor solution, which isknown to affect the morphology of crystals formed by homo-geneous nucleation.39

The number density of the nanorods [Fig. 7(c)] is obtainedfrom the corresponding SEM images (Fig. 6). The numberdensity decreased sharply with the addition of 0.5M H2O tothe ethanolic seed precursor solution, compared to a samplemade with no added water. Further additions of waterresulted in modest reductions in the rod number density.Water leads to a smaller number density of seed crystals,which could be expected to result in a lower number ofnanorods. Notably, however, the seeds greatly outnumber

the rods as was shown in Table I. According to the nucle-ation theory, ZnO nanorods nucleate at the seed/solutioninterface. The nucleation process occurs on the free surfaceor the grain boundary in between the ZnO seed nanoparti-cles.31,40–43 The nucleation process depends on the diameterand orientation of the ZnO seed nanoparticles.41 Our resultsindicate that grain-boundary nucleation between two ZnOgrains in the ZnO seed layer is the mechanism for the growthof the ZnO nanorods in agreement with previous reports.44

Reports on ZnO nanorods grown from sputtered seed lay-ers have given contradictory results.45 Ghayour et al. indi-cated a decrease in nanorod number density with increasingthickness of the seed layer.46 Liu et al., on the other hand,found sharply increasing rod density with seed layer thicknessfor thin seed layers (<3.5 nm).31 Using solution-processedseed layers, Ma et al. could alter the nanorod density byadjusting the concentration of the seed precursor solutionand the rotational speed of spin-coating used to deposit it.32

We are not aware of any prior work using forced hydrolysisof the seeds to adjust nanorod density.

(2) Nanorod MorphologyIn CBD, ZnO nanorod growth is limited by mass transportof the ions in the precursor solution.47 The average diameterand length and of the nanorods obtained from the corre-sponding SEM images are plotted in Figs. 7(a) and (b). Wehave seen an increase in the diameter and length withdecreasing density of the ZnO nanorods. Our results are inagreement with previous reports that diameter of the nano-rod increases with increasing seed size48 and decreasing seeddensity.48,49

This trend has been explained by theoretical modeling. Inthe work of Boercker et al.,47 when seeds were depositedover a defined rectangular area, rods that grew at the edge ofthe rectangle were much longer and wider than those at thecenter. This was explained by modeling based on nanorodgrowth being limited by mass transport of Zn2+ from thebulk solution, with the diffusion layer much larger than therod length. Rods in the center of the rectangle grew in anarea of depressed precursor concentration, while those at theedge had close access to bulk solution. In the present results,

Fig. 5. Room-temperature photoluminescence of ZnO nanorods.

(a) (b) (c)

(d) (e) (f)

Fig. 6. Scanning electron microscopy images of ZnO nanorods grown on FTO with different concentrations of water in the seed precursorsolution: (a) 0, (b) 0.5, (c) 0.6, (d) 0.7, (e) 0.8, (f) 0.9M.

April 2014 Forced Hydrolysis of ZnO Seeds 1031

Page 5: ZnO nanorod density control

a high density of growing nanorods depresses the steady-stateconcentration of Zn2+ at the location of the rods, while alower density of nanorods results in higher local [Zn2+] andhigher growth rates at each rod.

Figures 7(d) and (e) show how the roughness factor (sur-face area per projected area, calculated from the nanoroddimensions and number density) and aspect ratio (lengthdivided by diameter) change as a function of [H2O]SPS, theconcentration of water added to the seed precursor solution.Figure 7(f) correlates the nanorod number density to the roddiameter. The changes in nanorod diameter and length are

not proportional to each other, as the aspect ratio decreasesmonotonically with increasing [H2O]SPS. However, the rough-ness factor, after decreasing by about a factor of two for0.5M water versus no added water in the seed precursorsolution, stays roughly constant with [H2O]SPS for0.5M ≤ [H2O]SPS ≤ 1.0M. This is to say that with furtheradditions of water beyond 0.5M to the seed precursor solu-tion, the resulting increase in rod size approximately compen-sates for the decrease in rod number density to maintainabout the same surface area of ZnO nanorods per area ofsubstrate. The decrease in the density of the ZnO nanorods

(a) (b) (c)

(d) (e) (f)

Fig. 7. Variation in (a) diameter, (b) length, (c) number density, (d) roughness factor, and (e) aspect ratio (note logarithmic y-axis) of ZnOnanorods versus water concentration in the seed precursor solution, and (f) nanorod density versus diameter.

Table I. Density of Seeds with Different Water Concentrations in Seed Solution

Concentration of water added to seed solution (M) Amount of ZnO loaded (lg/cm2) Density of seeds (lm�2) Density of nanorods (lm�2)

0 1.64 � 0.11 651 � 45 20.10.5 0.97 � 0.20 210 � 44 3.210.8 0.42 � 0.03 65.3 � 4.1 1.011.0 0.30 � 0.03 29.4 � 3.2 0.203

(a) (b) (c)

Fig. 8. Scanning electron microscopy images of (a) accumulated ZnO seeds deposited on an FTO substrate and (b) and (c) ZnO nanorodsgrown from the seeds in (a).

1032 Journal of the American Ceramic Society—Manthina et al. Vol. 97, No. 4

Page 6: ZnO nanorod density control

with the increase in the size of the seeds is due to decrease inthe density of the grain boundaries with the increase in thesize of the particle.

(3) ZnO Seed Solution AgingAllowing a heated precursor solution to age for 24 hresulted in spherical agglomerates of seed crystals about800 nm in diameter [see Fig. 8(a)]. This can be attributed tocollision coalescence of ZnO crystallites forming misorientedattachments that fail to reorient themselves due to the lowdriving force at room temperature.50 Instead of a single largecrystal, the result is a large agglomeration of crystallites.The many different exposed facets on the exterior of thisseed provide multiple nucleation sites, resulting in radialgrowth of nanorods. Growth of ZnO from these seeds bythe usual CBD therefore produced starburst [Fig. 8(b)] andball [Fig. 8(c)] shapes. Thus, heating and addition of wateraccelerate the growth of ZnO nanoseeds and result in sepa-rate nanorods with reduced density, whereas extended agingat room temperature instead produces agglomerates that donot result in nanorod arrays. Ball shaped 3-D structures(average diameter ca. 20 lm) as shown in Fig. 8(c) are simi-lar to the structures that were grown on sputtered seedsusing Al3+ as additive.51

V. Conclusion

This work has demonstrated the use of forced hydrolysis forexerting control over the number density of ZnO nanorodsgrown by solution-based methods compatible with roughFTO substrates. Addition of water to and gentle heating ofthe ethanolic ZnO seed precursor solution results in increasedsize of ZnO crystallites that are deposited at a lower numberdensity on the substrate. CBD on these seeds results in anarray of nanorods that have lower number density but largerdimensions (diameter and length). With increased water addi-tion, the aspect ratio declines monotonically, but the totalnanorod surface area (and, therefore, the roughness factor)remains nearly constant as the effects of increased rod sizeand decreased number roughly cancel. By contrast, aging aheated seed precursor solution at room temperature resultsin different type of particle agglomeration giving clusters thatinduce radial growth of nanorods giving “starburst” shapes.The forced hydrolysis method enables large-scale and low-cost fabrication of ZnO nanorod arrays with controllednumber density, even on rough FTO substrates.

Acknowledgments

This material is based upon work supported by the National Science Founda-tion under grant no. CBET-1332022 and by the University of ConnecticutResearch Foundation. T. P. was supported by the National Science Founda-tion under grant no. EEC-1062955. The authors acknowledge the Universityof Connecticut Center for Clean Energy Engineering for use of the XRD andthe environmental scanning electron microscope, and the University ofConnecticut Department of Materials Science & Engineering for use of thespectrofluorimeter. The authors thank Neila Seda, Hongwei Luan andDr. Timothy Vadas for help with quantifying Zn2+ using ICP–MS.

References

1I. Beinik, M. Kratzer, A. Wachauer, L. Wang, R. T. Lechner, C. Teichert,C. Motz, W. Anwand, G. Brauer, X. Y. Chen, X. Y. Hsu, and A. B. Djurisic,“Electrical Properties of ZnO Nanorods Studied by Conductive Atomic ForceMicroscopy,” J. Appl. Phys., 110 [5] 052005–7 (2011).

2G. V. Elmore and H. A. Tanner, “The Photochemical Properties of ZincOxide,” J. Phys. Chem., 60 [9] 1328–9 (1956).

3Y. Kinemuchi, M. Mikami, K. Kobayashi, K. Watari, and Y. Hotta,“Thermoelectric Properties of Nanograined ZnO,” J. Electron. Mater., 39 [9]2059–63 (2010).

4M. Li, Y. J. Su, W. Y. Chu, L. J. Qiao, A. A. Volinsky, and G. Kra-vchenko, “Local Piezoelectric Effect on Single Crystal ZnO Microbelt Trans-verse I-V Characteristics,” Appl. Phys. Lett., 98 [8] 082105, 3pp (2011).

5I. K. Bdikin, J. Gracio, R. Ayouchi, R. Schwarz, and A. L. Kholkin,“Local Piezoelectric Properties of ZnO Thin Films Prepared by RF-Plasma-

Assisted Pulsed-Laser Deposition Method,” Nanotechnology, 21 [23] 235703(2010).

6N. Gopalakrishnan, L. Balakrishnan, K. Latha, and S. Gowrishankar,“Influence of Substrate and Film Thickness on Structural, Optical and Elec-trical Properties of ZnO Thin Films,” Cryst. Res. Technol., 46 [4] 361–7(2011).

7M. Law, L. E. Greene, A. Radenovic, T. Kuykendall, J. Liphardt, andP. Yang, “ZnO�Al2O3 and ZnO�TiO2 Core�Shell Nanowire Dye-SensitizedSolar Cells,” J. Phys. Chem. B, 110 [45] 22652–63 (2006).

8D. Wu, Z. Gao, F. Xu, J. Chang, W. Tao, J. He, S. Gao, and K. Jiang,“Hierarchical ZnO Aggregates Assembled by Orderly Aligned Nanorods fordye-Sensitized Solar Cells,” CrystEngComm, 15 [6] 1210–7 (2013).

9K. S. Weißenrieder and J. M€uller, “Conductivity Model for SputteredZnO-Thin Film gas Sensors,” Thin Solid Films, 300 [1–2] 30–41 (1997).

10S.-W. Chung, J.-Y. Yu, and J. R. Heath, “Silicon Nanowire Devices,”Appl. Phys. Lett., 76 [15] 2068–70 (2000).

11P. Yang, H. Yan, S. Mao, R. Russo, J. Johnson, R. Saykally, N. Morris,J. Pham, R. He, and H. J. Choi, “Controlled Growth of ZnO Nanowires andTheir Optical Properties,” Adv. Funct. Mater., 12 [5] 323–31 (2002).

12Z. L. Wang and J. Song, “Piezoelectric Nanogenerators Based on ZincOxide Nanowire Arrays,” Science, 312 [5771] 242–6 (2006).

13A. B. Djuri�si�c, A. M. C. Ng, and X. Y. Chen, “ZnO Nanostructures forOptoelectronics: Material Properties and Device Applications,” Prog. QuantumElectron., 34 [4] 191–259 (2010).

14C.-Y. Chen, J. R. D. Retamal, I. W. Wu, D.-H. Lien, M.-W. Chen, Y.Ding, Y.-L. Chueh, C.-I. Wu, and J.-H. He, “Probing Surface Band Bendingof Surface-Engineered Metal Oxide Nanowires,” ACS Nano, 6 [11] 9366–72(2012).

15S. A. Vanalakar, S. S. Mali, R. C. Pawar, N. L. Tarwal, A. V. Moholkar,J. H. Kim, and P. S. Patil, “Photoelectrochemical Properties of CdS SensitizedZnO Nanorod Arrays: Effect of Nanorod Length,” J. Appl. Phys., 112 [4]044302–7 (2012).

16D. Q. Yun, X. Y. Xia, S. Zhang, Z. Q. Bian, R. H. Liu, and C. H.Huang, “ZnO Nanorod Arrays with Different Densities in Hybrid Photovol-taic Devices: Fabrication and the Density Effect on Performance,” Chem.Phys. Lett., 516 [1–3] 92–5 (2011).

17J. S. Wang and K. M. Lakin, “C-Axis Inclined ZnO Piezoelectric ShearWave Films,” Appl. Phys. Lett., 42 [4] 352–4 (1983).

18V. Manthina, J. P. Correa Baena, G. Liu, and A. G. Agrios, “ZnO–TiO2

Nanocomposite Films for High Light Harvesting Efficiency and Fast ElectronTransport in Dye-Sensitized Solar Cells,” J. Phys. Chem. C, 45, 23864–70(2012).

19Y. Cui, A. Yu, H. Pan, X. Zhou, and W. Ding, “Catalytic Outgrowth ofSnO2 Nanorods from ZnO–SnO2 Nanoparticles Microsphere Core: Combus-tion Synthesis and Gas-Sensing Properties,” CrystEngComm, 14 [21] 7355–9(2012).

20V. Manthina, J. P. Correa Baena, G. Liu, and A. G. Agrios, “ZnO–TiO2

Nanocomposite Films for High Light Harvesting Efficiency and Fast ElectronTransport in Dye-Sensitized Solar Cells,” J. Phys. Chem. C, 116 [45] 23864–70(2012).

21S.-H. Yu, L. R. MacGillivray, and C. Janiak, “Nanocrystals,” CrystEng-Comm, 14 [22] 7531–4 (2012).

22P. O’Brien, T. Saeed, and J. Knowles, “Speciation and the Nature of ZnOThin Films From Chemical Bath Deposition,” J. Mater. Chem., 6 [7] 1135–9(1996).

23L. Vayssieres, K. Keis, A. Hagfeldt, and S.-E. Lindquist, “Three-Dimen-sional Array of Highly Oriented Crystalline ZnO Microtubes,” Chem. Mater.,13 [12] 4395–8 (2001).

24T. Yoshida, M. Tochimoto, D. Schlettwein, D. W€ohrle, T. Sugiura, andH. Minoura, “Self-Assembly of Zinc Oxide Thin Films Modified with Tetra-sulfonated Metallophthalocyanines by One-Step Electrodeposition,” Chem.Mater., 11 [10] 2657–67 (1999).

25B. O’Regan, V. Sklover, and M. Gratzel, “Electrochemical Deposition ofSmooth and Homogeneously Mesoporous ZnO Films from Propylene Carbon-ate Electrolytes,” J. Electrochem. Soc., 148 [7] C498–505 (2001).

26L. Xu, Q. Liao, J. Zhang, X. Ai, and D. Xu, “Single-Crystalline ZnONanotube Arrays on Conductive Glass Substrates by Selective Disolution ofElectrodeposited ZnO Nanorods,” J. Phys. Chem. C, 111 [12] 4549–52 (2007).

27M. H. Huang, Y. Wu, H. Feick, N. Tran, E. Weber, and P. Yang, “Cata-lytic Growth of Zinc Oxide Nanowires by Vapor Transport,” Adv. Mater., 13[2] 113–6 (2001).

28R. R. Bacsa, J. Dexpert-Ghys, M. Verelst, A. Falqui, B. Machado, W. S.Bacsa, P. Chen, S. M. Zakeeruddin, M. Graetzel, and P. Serp, “Synthesis andStructure–Property Correlation in Shape-Controlled ZnO Nanoparticles Pre-pared by Chemical Vapor Synthesis and Their Application in Dye-SensitizedSolar Cells,” Adv. Funct. Mater., 19 [6] 875–86 (2009).

29J. Jie, G. Wang, Y. Chen, X. Han, Q. Wang, B. Xu, and J. G. Hou, “Syn-thesis and Optical Properties of Well-Aligned ZnO Nanorod Array on anUndoped ZnO Film,” Appl. Phys. Lett., 86 [3] 031909, 3pp (2005).

30D. Bekermann, A. Gasparotto, D. Barreca, L. Bovo, A. Devi, R. A.Fischer, O. I. Lebedev, C. Maccato, E. Tondello, and G. Van Tendeloo,“Highly Oriented ZnO Nanorod Arrays by a Novel Plasma Chemical VaporDeposition Process,” Cryst. Growth Des., 10 [4] 2011–8 (2010).

31J. Liu, J. She, S. Deng, J. Chen, and N. Xu, “Ultrathin Seed-Layer forTuning Density of ZnO Nanowire Arrays and Their Field Emission Character-istics,” J. Phys. Chem. C, 112 [31] 11685–90 (2008).

32T. Ma, M. Guo, M. Zhang, Y. Zhang, and X. Wang, “Density-ControlledHydrothermal Growth of Well-Aligned ZnO Nanorod Arrays,” Nanotechnol-ogy, 18 [3] 035605 (2007).

April 2014 Forced Hydrolysis of ZnO Seeds 1033

Page 7: ZnO nanorod density control

33L. Spanhel and M. A. Anderson, “Semiconductor Clusters in the sol-gelProcess: Quantized Aggregation, Gelation, and Crystal Growth in Concen-trated Zinc Oxide Colloids,” J. Am. Chem. Soc., 113 [8] 2826–33 (1991).

34E. H. Otal, M. Granada, H. E. Troiani, H. C�anepa, and N. E. Wals€oe deReca, “Nanostructured Colloidal Crystals from Forced Hydrolysis Methods,”Langmuir, 25 [16] 9051–6 (2009).

35Y. Hu and H.-J. Chen, “Preparation and Characterization of Nanocrystal-line ZnO Particles From a Hydrothermal Process,” J. Nanopart. Res., 10 [3]401–7 (2008).

36W.-J. Li, E.-W. Shi, W.-Z. Zhong, and Z.-W. Yin, “Growth Mechanism andGrowth Habit of Oxide Crystals,” J. Cryst. Growth, 203 [1–2] 186–96 (1999).

37L. Vayssieres, K. Keis, S.-E. Lindquist, and A. Hagfeldt, “Purpose-BuiltAnisotropic Metal Oxide Material: 3D Highly Oriented Microrod Array ofZnO,” J. Phys. Chem. B, 105 [17] 3350–2 (2001).

38A. M. Lockett, P. J. Thomas, and P. O’Brien, “Influence of SeedingLayers on the Morphology, Density, and Critical Dimensions of ZnO Nano-structures Grown by Chemical Bath Deposition,” J. Phys. Chem. C, 116 [14]8089–94 (2012).

39H. L. Cao, X. F. Qian, Q. Gong, W. M. Du, X. D. Ma, and Z. K. Zhu,“Shape- and size-controlled synthesis of nanometre ZnO from a simple solu-tion route at room temperature,” Nanotechnology, 17, 3632–6 (2006).

40W. Wu, G. Hu, S. Cui, Y. Zhou, and H. Wu, “Epitaxy of Vertical ZnONanorod Arrays on Highly (001)-Oriented ZnO Seed Monolayer by a Hydro-thermal Route,” Cryst. Growth Des., 8 [11] 4014–20 (2008).

41S.-C. Liou, C.-S. Hsiao, and S.-Y. Chen, “Growth Behavior and Micro-structure Evolution of ZnO Nanorods Grown on Si in Aqueous Solution,”J. Cryst. Growth, 274 [3–4] 438–46 (2005).

42S.-W. Chen and J.-M. Wu, “Nucleation Mechanisms and Their Influenceson Characteristics of ZnO Nanorod Arrays Prepared by a HydrothermalMethod,” Acta Mater., 59 [2] 841–7 (2011).

43C.-C. Lin, S.-Y. Chen, and S.-Y. Cheng, “Nucleation and Growth Behav-ior of Well-Aligned ZnO Nanorods on Organic Substrates in Aqueous Solu-tions,” J. Cryst. Growth, 283 [1–2] 141–6 (2005).

44C.-S. Hsiao, C.-H. Peng, S.-Y. Chen, and S.-C. Liou, “Tunable Growth ofZnO Nanorods Synthesized in Aqueous Solutions at low Temperatures,”J. Vac. Sci. Technol., A, 24 [1] 288–91 (2006).

45S. Xu and Z. Wang, “One-Dimensional ZnO Nanostructures: SolutionGrowth and Functional Properties,” Nano Res., 4 [11] 1013–98 (2011).

46H. Ghayour, H. R. Rezaie, S. Mirdamadi, and A. A. Nourbakhsh, “TheEffect of Seed Layer Thickness on Alignment and Morphology of ZnO Nano-rods,” Vacuum, 86 [1] 101–5 (2011).

47J. E. Boercker, J. B. Schmidt, and E. S. Aydil, “Transport LimitedGrowth of Zinc Oxide Nanowires,” Cryst. Growth Des., 9 [6] 2783–9 (2009).

48W.-Y. Wu, C.-C. Yeh, and J.-M. Ting, “Effects of Seed Layer Characteristicson the Synthesis of ZnO Nanowires,” J. Am. Ceram. Soc., 92 [11] 2718–23 (2009).

49W. Liang, B. D. Yuhas, and P. Yang, “Magnetotransport in Co-DopedZnO Nanowires,” Nano Lett., 9 [2] 892–6 (2009).

50H. Zeng, P. Liu, W. Cai, X. Cao, and S. Yang, “Aging-Induced Self-Assembly of Zn/ZnO Treelike Nanostructures From Nanoparticles andEnhanced Visible Emission,” Cryst. Growth Des., 7 [6] 1092–7 (2007).

51X. Yan, Z. Li, C. Zou, S. Li, J. Yang, R. Chen, J. Han, and W. Gao,“Renucleation and Sequential Growth of ZnO Complex Nano/Microstructure:From Nano/Microrod to Ball-Shaped Cluster,” J. Phys. Chem. C, 114 [3]1436–43 (2010). h

1034 Journal of the American Ceramic Society—Manthina et al. Vol. 97, No. 4