Abstract
In this study, possibilities of scaling up the synthesis of zinc oxide (ZnO) nanorods (NRs) by the hydrothermal method have been explored. It was found that batches yielding several grams can easily be made using common and easily available materials. Further, a printable composition was fabricated by mixing the obtained ZnO NRs into a common solvent-based screen printable varnish. Scanning electron microscope, high-resolution transmission electron microscope, X-ray diffraction, UV–vis spectroscopy analysis of the scaled up batch indicated that the ZnO nanostructures were of NRs shape, well crystalline and having less defects. Using the ZnO NRs-based printable composition a device fabrication on a flexible substrate was demonstrated, producing a flexible light-emitting device being highly tolerant to bending.
Similar content being viewed by others
Introduction
Zinc oxide (ZnO) has emerged as a very promising material for electronics and optoelectronics applications [1]. It has a rich family of nanostructures (NSs), such as nanorods (NRs) [2], nanotubes [3], and nanowalls [4], which exhibit superior properties in chemical, electric, and optical fields [5]. ZnO can function in a variety of practical applications, including optoelectronic devices, e.g., UV sensors [6], solar cells [7], and light-emitting diodes (LEDs) [5, 8].
Several approaches including very sophisticated and expensive deposition technique such as metal-organic chemical vapor deposition (MOCVD) [9] and sputter deposition techniques [10], have been used to synthesize ZnO NRs. These methods are of batch type and difficult to scale up for large volume production. Alternatives to those highly expensive methods are provided by relatively cheap wet chemical method such as sol–gel [11] and resistive heating, e.g., vapor liquid solid (VLS) [12] have been used to synthesize ZnO NSs. However, these processes are also yielding rather small amount of ZnO material. In most cases, ZnO NRs have been synthesized on substrates but the use of these substrates always limits the scale-up production [13] and normally the synthesis is easier to control in small scale reactor. A demand for low-cost large-area devices for lighting requires scalable production of low-cost materials that can be deposited by cheap and scalable methods. As printing and coating are low cost methods for materials deposition, the objective here is the manufacturing of a printable composition with a high concentration of ZnO NSs. A test print on a semi-automatic screen printer requires at least a few grams to test printability. Here, we have scaled up the preparation of ZnO NRs by a hydrothermal method to produce a few grams per batch. The scale up is accomplished by increasing the reaction volume and the area onto which ZnO NSs can grow and precipitate. This method is already proven to be successful and by varying the molar concentration ZnO NRs of various aspect ratios (length/diameter) can be obtained [14].
Flexibility of electronic devices is desirable for both manufacturing and application reasons. A variety of flexible organic electronic components have been realized already. Likewise, flexibility of inorganic electronic devices is desired as well, for a wide range of applications, such as photovoltaic devices, [15, 16] flexible batteries, [17], and LEDs [18]. While there has been much progress toward the development of flexible organic LEDs, a similar level of maturity has not yet been achieved in flexible inorganic LEDs. One approach to improve the mechanical stability of flexible devices the electronic industry is moving to a different approach such as printed electronics (PE). Printed electronics has a potential to develop new devices, e.g., light weight flexible LEDs. In this approach, the expensive material and complex method is replaced with a cheap material and low temperature, low-cost manufacturing techniques, such as inkjet printing, roll to roll coasting, and screen printing to print devices at large scale. Until now, research has been focused in PE to organic materials, much less effort has been devoted to use inorganic semiconductor. Despite this, some inorganic NSs such as NRs, nanoparticles, and nanowires have an extra advantage for printing; it can be easily fabricated in large amount by solution method which make it very suitable for PE [19].
Experimental details
Scaling up the production of surface grown and precipitated ZnO NRs was done by arranging a stack of glass plates supported on a plastic frame (Fig. 1a) immersed in the nutrient growth solution in a plastic container. The reaction vessel for the synthesis of ZnO NRs consists of a 10 L rectangular polypropylene plastic box (Fig. 1b). The rack for glass plates was manufactured from plastic pipes (polypropylene compound), HFIR 20 and 11 from Dietzel GmbH, Austria and consists of four corner pipes having a diameter of 5 mm, and supporting pipes with a diameter of 12 mm, onto which glass slides are supported. The frame is assembled by fixing the supporting and structural pipes into holes drilled in the corner pipes, and fixing their position by plastic cable ties. In this way, the frame for supporting a stack of glass slides was manufactured without using glues, adhesives, or metal connectors, from materials that are cheap and stable to the conditions of ZnO NRs growth, preparation and washing. Glass slides were prepared by cutting window glass pane into 220 × 140 mm2 pieces.
Before synthesis, the plastic box, the rack, and the glass panes were cleaned with acetone, iso-propanol, and deionized water (DI). A seed layer on the glass panes, prepared by the Pacholski et al. method [20] was spray coated on each slide to enhance the density and crystallinity of the ZnO NRs. A 10 L nutrient growth solution containing equimolar (0.075 M) zinc nitrate hexahydrate (Zn(NO3)2·6H2O) and hexamethylenetetramine (HMT, C6H12N4) was prepared and after continuous stirring for half an hour it was poured in the plastic box. The rack along with the seeded glass plates was lowered into the nutrient solution and was placed in the ordinary laboratory oven at 95 °C for several hours. After terminating the reaction the growth plates were withdrawn from the vessel (Fig. 1c) and placed in the oven at 100 °C until they were dried. Finally, from each growth plate the grown NRs were mechanically scraped by a glass microscopy slide (Fig. 1d). They were collected, washed with acetone, deionized water several time, and then dried. The morphological and structural studies of products were examined at 10 kV using a JEOL JSM-6301F field emission scanning electron microscope (FE-SEM), high-resolution transmission electron microscope (HRTEM), and X-ray diffraction (XRD). Optical characterization was performed on a PerkinElmer Lambda 900 UV–vis spectrometer.
To fabricate the LEDs, we have investigated different organic dispersant for ZnO NRs to get a potentially printable ink. After finalizing the organic dispersant, specified amount of ZnO NRs was dispersed in the organic dispersant, yielding a printable ink. A layer of this ink was printed on the patterned substrate layers. The patterned layers were prepared as follows. First, we have printed a pattern of PEDOT: PSS either on paper or plastic substrate and then a poly (9,9-dioctylfluorenyl-2,7-diyl) (PFO) layer was spin coated using spin speed 2,000 rpm for 30 s. For printing the ZnO NRs we have tested two method (1) stencil printing and (2) screen printing. In stencil printing, ink is placed in a stencil spacer in the shape of adhesive tape defining the area to be printed, and is spread in the space by squeezing with a flat edged object. Stencil print is considered as a first order printability check, and provides samples for proof of principle.
A screen printing paste was manufactured by dispersing ZnO NRs (5.166 g) in a solvent-based screen printing varnish, Plastijet XG 383 (2.5 g, 50%), manufactured by Fujifilm Sericol Ltd., and a small amount (1.1 g) of a retarder, Plastijet ZV 558, was added. The mixture was stirred using an Ultra Turrax™ Tube dispenser before printing.
Screen printing was performed using a nylon web 120-34 (wires per inch and wire diameter in mils) for both electrodes and ZnO NRs paste, using a semi-automatic flat-bed screen printing machine, Screen-printer TIC SCF 550, Eickmeyer. The printed substrates were dried at 90 °C in an oven after printing. Finally, we applied aluminum contacts and proceeded for the measurements. Electroluminescence (EL) spectra were acquired on an Andor Shamrock 303iB spectrometer supported with Andor-Newton DU-790 N CCD.
Results and discussions
Large scale synthesis of ZnO NRs and their characterization
The hydrothermal method is used to perform the synthesis of ZnO NRs. The grown materials from each glass slide were mechanically scraped as shown in Fig. 1d. ZnO growth by low-temperature hydrothermal method is substrate independent and produces high-quality NRs. We have used glass slide because in such a large amount solution plastic sheets were not possible to sink in the solution. After collecting the product it was carefully cleaned, dried out and weighted as a result more than 6 g of ZnO NRs were obtained.
The ZnO NRs morphology was first observed using SEM as shown in Fig. 2a. All the top surfaces of the grown ZnO NRs were of hexagonal shape with an average diameter of 550 nm, and an average length of 2.2 μm. In this process, the size of the NRs can be tailored by controlling the hydrothermal conditions such as precursor concentrations, growth temperature, and growth time [14, 16, 21].
An individual ZnO NR was characterized by using the HRTEM and selected area electron diffraction (SAED) to reveal their structural properties (Fig. 3a). The image illustrates that the ZnO NR has a high-quality single crystal structure with a lattice spacing of 0.52 nm and follows [002] growth direction (inset of Fig. 3a, right). The corresponding SAED pattern in the inset of Fig. 3a, left is showing that the rod is single crystal and can be readily indexed [002] or c-axis growth direction of ZnO NR. This is obvious, since the fastest growth direction for the ZnO NRs is along the c-axis or [001] direction.
The crystallinity of the corresponding product of ZnO NRs was further confirmed by XRD as shown in Fig. 3b. The sample for measurement was prepared by spreading the synthesized ZnO NRs on a substrate, which was then used as a sample holder for XRD. The observed diffraction peaks in the recorded XRD pattern is in good agreement with the values available in the JCPDS card 36-1451for hexagonal ZnO with wurtzite structure. Usually the relative intensity of peak (002) preferably appear stronger for the case of aligned ZnO NRs grown on a substrate, in our case the XRD pattern slightly differed and resemble to the ZnO powder in which (002) peak weakens while (101) peak intensity appeared stronger. This is because XRD signal originates from the whole sample which is dominated by the rods randomly distributed and lying down on the substrate. Except ZnO no other evident diffraction peak was observed, suggesting that the large amount of prepared nanomaterials belongs to high-quality ZnO.
The optical properties of the synthesized ZnO NRs were examined by UV–vis spectrometer at room temperature and the results are shown in Fig. 4. For the measurement, the synthesized ZnO NRs were well dispersed in ethanol and the resultant solution was used for measurement. The typical UV–vis absorption spectrum has a peak near the band edge in the exciton absorption region at about 377 nm and blue shifted relative to bulk excitonic absorption (380 nm) which could be assigned to strongly bound exciton [22].
Printed LEDs
The performance of the ZnO/polymer LEDs is strongly dependent on the processing conditions and environment [23]. Past studies on ZnO/polymer-based devices have relied on ZnO NRs grown within the solution on top of the polymer spin-coated substrates [24]. The aqueous nutrient solutions at elevated temperature will degrade the polymer films and as a result the subsequent device performance will be greatly affected [25]. Furthermore, we also used a paper substrate which cannot survive in the solution environment. Here, we have employed the printed method to avoid aqueous environment in addition to other advantages like large area printing.
Schematic diagram of the printed ZnO NRs/polymer LED is depicted in Fig. 5a. As shown in the schematic diagram the printed LED is composed of a hole transport layer (HTL) of PEDOT: PSS printed on PET substrate, a PFO layer, ZnO NRs printed layer and Al top electrode. The PEDOT: PSS was often used as a HTL in different devices because it has a high-hole mobility [26].
Figure 5b shows the current versus voltage (I–V) relationship for the printed ZnO LED consisting of PEDOT: PSS/PFO/ZnO NRs/Al. The hybrid LED showed a rectifying behavior of the diode with a turn-on voltage of about 3.9 V and low leakage current of ~1 nA in the reverse bias of 6 V. The plastic substrate containing the printed LEDs was rolled many times to examine any possible degradation of both the printed films and/or the LEDs as clearly shown in the optical photograph of Fig. 2b–d. Interestingly, the ZnO NRs/polymer structure was still adhered with no observable changes in the I–V characteristics were seen by the rolling/bending process.
The room temperature electroluminescence (EL) spectrum of the printed LED obtained at an injection voltage of 16 V is shown in Fig. 5c. The EL reveals a broad emission band covering the visible spectrum range from 415 to 630 nm. Two blue shoulder peaks at 450 and 480 nm can clearly be observed in the EL spectrum, the first peak at 450 nm is attributed to the excitonic transition in PFO [27]. The second peak at 480 nm represents a first-order vibronic transition of PFO [27]. On the other hand, the EL spectrum demonstrated a broad green emission centered at 532 nm as shown in Fig. 5c that can be ascribed to the deep level defects in ZnO. It is known that ZnO NRs grown by the solution method present a large number of defects [28]. The PFO also has defect emission in this range due to the formation of the fluorenone agglomerates or so-called keto defect during electro-(photo) oxidation [29]. The reason for this could be when a fluorenone defect is formed, the alkyl chain is lost, favoring local interchain interaction between the PFO backbone [29]. But the presence of fluorenone defects only may not be sufficient to produce the green band emission. The interchain interactions and cross linking of the polymers are also proposed to be required for the appearance of the green band [30]. These defects can easily be produced during synthesis and functioning of the device. The emission of the device suggested that the carrier recombination takes place in the PFO/ZnO NRs interface as recently reported [31]. The interaction between the ZnO NRs surface and PFO molecules makes the PFO link up with ZnO. This leads to PFO/ZnO interface producing a strong defect emission [32]. As a result the device exhibits a broad EL spectrum in the range 415–630 nm and gives the cold white light. The color quality of the printed LED was investigated with the Commission International DE L′Eclairage (CIE) as shown in Fig. 5d. The resultant light has a cold white impression (bluish) with color coordinate: x = 0.2574 and y = 0.3961 and a correlated color temperature (CCT) of 8367 K.
Conclusions
Here, we have demonstrated that the synthesis of ZnO NRs can be scaled up by a simple and low cost hydrothermal method. We have obtained more than 6 g of ZnO NRs per batch. With a suitable ratio of the solvent, retarder, and ZnO NRs materials we were able to manufacture printable composition producing continuous and smooth layers by screen printing. Printing of the ZnO NRs on a polymer printed flexible substrate was realized for fabricating a functional electroluminescent device and was operational even after continuous bending of the substrate. The approach described here can be applied to a wide range of printing inorganic\polymer LEDs on a paper or plastic substrates and that it can be extended to other semiconductor devices. Although further optimization of the printing device is necessary to lower the elevated turn-on voltage and to achieve a very high-performance LED, our screen printing approach provides a potential path way for simple and low-cost printed devices.
References
Ozgur U, Alivov YI, Liu C, Teke A, Reshchikov MA, Dogan S, Avrutin V, Cho SJ, Morkoc H (2005) J Appl Phys 98: 41301. doi:10.1063/1.1992666
Ko SH, Lee D, Kang HW, Nam KH, Yeo JY, Hong SJ, Grigoropoulos CP, Sung HJ (2011) Nano Lett 11:666. doi:10.1021/nl1037962
Zhou JZ, Guo HH, Lin ZH, Feng ZF, Lin LL (2009) J Phys Chem C 113:12546. doi:10.1021/jp902607c
Israr MQ, Sadaf JR, Nur O, Willander M, Salman S, Danielsson B (2011) Appl Phys Lett 98:253705. doi:10.1063/1.3599583
Willander M, Nur O, Zhao QX, Yang LL, Lorenz M, Cao BQ, Zuniga Perez J, Czekalla C, Zimmermann G, Grundmann M, Bakin A, Behrends A, Al-Suleiman M, El-Shaer A, Che Mofor A, Postels B, Waag A, Boukos N, Travlos A, Kwack HS, Guinard J, Le Si Dang D (2009) Nanotechnology 20:332001. doi:10.1088/0957-4484/20/33/332001
Seminario JM, Gimenez AJ, Yanez-Limon JM (2011) J Phys Chem C 115:282. doi:10.1021/jp107812w
Wei M, Han JB, Fan FR, Xu C, Lin SS, Duan X, Wang ZL (2010) Nanotechnology 21: 035102
Zhang Y, Zhang XM, Lu MY, Chen LJ, Wang ZL (2009) Adv Mater 21:2767. doi:10.1002/adma.200802686
Wu JJ, Wong DKP (2007) Adv Mater 19:2015. doi:10.1002/adma.200602052
Ting JM, Chen MT (2006) Thin Solid Films 494:250. doi:10.1016/j.tsf.2005.08.134
Shi SB, Xu JP, Zhang XS, Li L (2011) J Appl Phys 109. doi:10.1063/1.3586243
Zhao QX, Klason P, Willander M (2007) Appl Phys A 88:27. doi:10.1007/s00339-007-3958-0
Yuan FL, Peng H, Bai LY, Li JL, Chen YF (2007) J Phys Chem C 111:194. doi:10.1021/jp065390b
Wang ZL, Xu S, Lao C, Weintraub B (2008) J Mater Res 23:2072. doi:10.1557/Jmr.2008.0274
Rogers JA, Baca AJ, Yu KJ, Xiao JL, Wang SD, Yoon J, Ryu JH, Stevenson D, Nuzzo RG, Rockett AA, Huang YG (2010) Energy Environ Sci 3:208. doi:10.1039/b920862c
Yu H, Zhang Z, Han M, Hao X, Zhu F (2005) J Am Chem Soc 127:2378. doi:10.1021/ja043121y
Gaikwad AM, Whiting GL, Steingart DA, Arias AC (2011) Adv Mater 23:3251. doi:10.1002/adma.201100894
Zainelabdin A, Zaman S, Amin G, Nur O, Willander M (2010) Nanoscale Res Lett 5:1442. doi:10.1007/s11671-010-9659-1
Toyama T, Takeuchi H, Yamaguchi D, Kawasaki H, Itatani K, Okamoto H (2010) J Appl Phys 108. doi:10.1063/1.3493157
Pacholski C, Kornowski A, Weller H (2002) Angew Chem Int Ed Engl 41:1188
Wang M, Ye CH, Zhang Y, Hua GM, Wang HX, Kong MG, Zhang LD (2006) J Cryst Growth 291:334. doi:10.1016/j.jcrysgro.2006.03.033
Yang Z, Zong XL, Ye ZZ, Zhao BH, Wang QL, Wang P (2010) Biomaterials 31:7534. doi:10.1016/j.biomaterials.2010.06.019
Olson DC, Shaheen SE, Collins RT, Ginley DS (2007) J Phys Chem C 111:16670
Bolink HJ, Coronado E, Sessolo M (2009) Chem Mater 21:439. doi:10.1021/Cm8031362
Olson DC, Lee YJ, White MS, Kopidakis N, Shaheen SE, Ginley DS, Voigt JA, Hsu JWP (2007) J Phys Chem C 111:16640. doi:10.1021/Jp0757816
Chen LJ, Manekkathodi A, Lu MY, Wang CW (2010) Adv Mater 22:4059. doi:10.1002/adma.201001289
Wadeasa A, Nur O, Willander M (2009) Nanotechnology 20:065710
Djurisic AB, Leung YH (2006) Small 2:944. doi:10.1002/smll.200600134
Montilla F, Mallavia R (2007) Adv Funct Mater 17:71. doi:10.1002/adfm.200600141
Zhao W, Cao T, White JM (2004) Adv Funct Mater 14:783. doi:10.1002/adfm.200305173
Wadeasa A, Tzamalis G, Sehati P, Nur O, Fahlman M, Willander M, Berggren M, Crispin X (2010) Chem Phys Lett 490:200. doi:10.1016/j.cplett.2010.03.050
Lee CY, Wang JY, Chou Y, Cheng CL, Chao CH, Shiu SC, Hung SC, Chao JJ, Liu MY, Su WF, Chen YF, Lin CF (2009) Nanotechnology 20(42): 425202
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Amin, G., Sandberg, M.O., Zainelabdin, A. et al. Scale-up synthesis of ZnO nanorods for printing inexpensive ZnO/polymer white light-emitting diode. J Mater Sci 47, 4726–4731 (2012). https://doi.org/10.1007/s10853-012-6342-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10853-012-6342-1