Effect of Sulfide Precursor Selection on the Nucleation, Growth, and

Dec 6, 2016 - NUS Graduate School for Integrative Sciences and Engineering, National University of Singapore, 117456 Singapore. ‡ Singapore Institut...
0 downloads 0 Views 3MB Size
Subscriber access provided by University of Otago Library

Article

Effect of sulfide precursor selection on the nucleation, growth, and elemental composition of Cu2ZnSnS4 nanocrystals Tang Jiao Huang, Ryan Lee Guang-Ren, Xuesong Yin, Chunhua Tang, Guojun Qi, and Hao Gong Cryst. Growth Des., Just Accepted Manuscript • DOI: 10.1021/acs.cgd.6b01217 • Publication Date (Web): 06 Dec 2016 Downloaded from http://pubs.acs.org on December 10, 2016

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.

Crystal Growth & Design 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 25

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

Crystal Growth & Design

Effect of sulfide precursor selection on the nucleation, growth, and elemental composition of Cu2ZnSnS4 nanocrystals

Tang Jiao Huang,1,2,3 Ryan Lee Guang-Ren,3 Xuesong Yin,3 Chunhua Tang,3 Guojun Qi,2 and Hao Gong*,3

1 NUS Graduate School for Integrative Sciences and Engineering, National University of Singapore, 117456 Singapore 2 Singapore Institute of Manufacturing Technology (SIMTech), 2 Fusionopolis Way, #08-04, 138634 Singapore 3 Department of Materials Science and Engineering, National University of Singapore, 117576 Singapore * Corresponding author: [email protected]

Abstract A means to synthesize large quantities of Cu2ZnSnS4 (CZTS) nanocrystals (NCs) is necessary for large-scale production of solution-processed CZTS solar cells. We embarked on a pioneering attempt to investigate and understand the conditions necessary to synthesize CZTS NCs in a novel formamide solvent system using an easily scalable heat-up method and focused on the effect of sulfide precursor selection (thioacetamide vs. thiourea). Unlike previous reports, which studied metallic and binary compounds, the compound (CZTS) we investigated comprises more than one metallic element. This difference in composition made possible the observation that sulfide precursor selection strongly alters the elemental composition of the NCs, a vital consideration for multinary compounds of which their performance is strongly dependent on their elemental composition. We found also that sulfide

1 ACS Paragon Plus Environment

Crystal Growth & Design

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

precursor choice influences the nucleation and growth characteristics of the NCs but not the phase of the final product or the phase transformation during reaction. The reasons for the observed differences are investigated and presented. Based on these findings, multigram quantities of CZTS NCs with small size and suitable elemental composition, properties crucial for solution-processed CZTS solar cells, are synthesized successfully with high yields.

2 ACS Paragon Plus Environment

Page 2 of 25

Page 3 of 25

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

Crystal Growth & Design

1 Introduction Cu2ZnSnS4 (CZTS) is a direct band gap semiconductor highly suited for use as a light absorber in solar cells.1 The earth abundance and non-toxicity of its constituents, coupled with respectable photovoltaic conversion efficiencies, make CZTS-based solar cells promising candidates for terawatt-scale photovoltaic deployment.2,3

To realize economic photovoltaic deployment on such a large scale, CZTS-based solar cells need to be fabricated using a process that is low-cost, safe, and robust with good control over the film’s elemental composition. The latter is important because the efficiency of such solar cells is highly dependent on the elemental composition of the CZTS-based film.4,5 Recently, a means to synthesize solution-processable CZTS nanocrystals (NCs) cheaply, without using highly toxic and/or potentially explosive chemicals, and with good control over the film’s elemental composition was demonstrated using a formamide solvent system.6 In brief, suitable metal and sulfide precursors separately dissolved in formamide were rapidly mixed at an elevated temperature of 170°C, forming CZTS NCs. This method of synthesizing NCs is termed ‘hot-injection method’.7 The NCs thus synthesized can be used to form good quality, large-grained CZTS thin-films without a fine-grain underlayer.6

However, to synthesize CZTS NCs on a large-scale, the ‘heat-up method’ may be more suitable because it can be more easily scaled up compared to hot-injection.8 To date, the heatup method has been demonstrated to be able to synthesize NCs in a single reaction in a laboratory on a multigram scale, up to 40 g.9-17 As such, it is of interest to make pioneering efforts to investigate and understand the conditions necessary to form CZTS NCs in formamide using the heat-up method without compromising their properties, such as size and elemental composition.

3 ACS Paragon Plus Environment

Crystal Growth & Design

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

Due to their different reaction characteristics, modifying the previously optimized hotinjection synthetic procedure to enable heat-up synthesis of CZTS NCs is not straightforward. For the heat-up method, its reaction characteristics can be understood using the LaMer model. In this model, there are three main stages, namely, accumulation, nucleation, and growth.18 During accumulation, the precursors, first mixed at a lower temperature, are heated to higher temperatures, decomposing to form suitable reactants in the process. When the reactants accumulate till a critical concentration, whereupon the reactants are highly supersaturated, rapid nucleation followed by growth occurs. This presence of an accumulation stage is what distinguishes the heat-up method from the hot-injection method. Consequently, the reaction conditions for CZTS formation via the heat-up method are significantly different from that of hot-injection. For example, nucleation and growth may initiate at lower temperatures if the critical concentration of reactants is reached at a lower temperature. The differences between these two methods thus make detailed study and understanding of the processes taking place during heat-up synthesis necessary.

In light of the distinguishing difference between the two methods, we identified sulfide precursor conversion rate as one of the key parameters to consider during CZTS NC heat-up synthesis.19,20 For example, both theoretical and experimental works have reported that higher precursor conversion rates led to higher concentrations of NCs in the reaction medium.19,21-25 Recently, a comparative study of different sulfide precursors (thiourea and its derivatives) showed that thiourea derivatives with higher conversion rates resulted in higher NC concentrations and smaller final NC diameters.19 Slower conversion rates have been shown to result in larger volumes and rod aspect ratios for CdS nanorods19 and larger diameters, lengths, and aspect ratios for Bi2S3 nanowhiskers.26 It is clear, therefore, that precursor conversion rate has a marked influence on NC synthesis. However, these comparative studies

4 ACS Paragon Plus Environment

Page 4 of 25

Page 5 of 25

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

Crystal Growth & Design

on the effect of precursor conversion rate on NC synthesis were mainly performed on metallic and binary compounds and not on multinary compounds comprising three or more elements such as quaternary CZTS.19,21-25,27-29 Consequently, it is still unclear how the properties (e.g. elemental ratios) of these multinary compounds will be affected when different sulfide precursors that differ only in their chemical side groups are used. For instance, during the synthesis of multinary compounds, metal-ligand complexes are first formed when the metal ions and sulfide precursors are co-dissolved.26,30 It is possible that when a different sulfide precursor is used, the relative stability between different metal-ligand complexes in the reaction medium changes, altering the rate of release of metal ions and the equilibrium position of the reaction, in turn affecting the elemental composition of the NCs. Considering that small size and suitable elemental composition are crucial for solutionprocessed CZTS solar cells, understanding how these properties are affected is important.

Therefore, the focus of this study is to investigate and understand the influence of sulfide precursor selection on CZTS NC heat-up synthesis in a formamide solvent system. To do so, two types of sulfide precursors namely thiourea (TU) and thioacetamide (TAA), known to have different reactivities and differ only in one chemical side group, are investigated. We found significant differences in the nucleation and growth characteristics as well as the NC’s elemental composition. Ceteris paribus, samples synthesized using more reactive TAA initiated nucleation at a lower temperature, have smaller particles, and an elemental composition close to stoichiometric CZTS. The reasons for these differences are investigated and presented. In addition to previous studies that reported how changing sulfide precursors changed the precursor conversion rate, we found that doing so changes also the stability of metal-ligand complexes and this change must be taken into consideration if the elemental composition of quaternary CZTS NCs is to be well controlled.

5 ACS Paragon Plus Environment

Crystal Growth & Design

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 25

2 Experimental Section 2.1 Synthesis of NCs 10 mL of precursor solution (10 mL formamide, 99.0+%, Sigma-Aldrich; 2.4 mmol Cu(Ac)2.H2O, 99.0%, Fluka; 1.65 mmol SnCl2, 98%, Sigma-Aldrich; 1.5 mmol ZnCl2, 98+%, Sigma-Aldrich) and an excess of sulfide precursor salt (either 8.325 mmol TAA, 98%, Sigma-Aldrich or 8.325 mmol TU, 99.0+%, Sigma-Aldrich) were mixed in a three-necked flask under magnetic stirring and degassed under nitrogen for 30 min. The resultant solution was then heated to a set temperature at a rate of approximately 10°C/min. After each reaction, the flask was quenched to room temperature. The solution was mixed with 25 mL of ethanol and centrifuged at 10,000 rpm for 10 min. The NCs collected were redispersed in 5 mL of formamide, mixed with 20 mL of ethanol, and centrifuged a further two times. The reaction conditions can be found in Table 1. The yield of the reaction is > 90% for sample 170-30 and approximately 80% for sample TU170-30.

Table 1. Reaction conditions of NC samples. T is the set temperature. Sulfide source

T/oC

Time at T/min

TAA100-0

TAA

100

0

TAA140-0

TAA

140

0

TAA170-0

TAA

170

0

TAA170-30

TAA

170

30

TU140-0

TU

140

0

TU170-0

TU

170

0

TU170-30

TU

170

30

Sample

6 ACS Paragon Plus Environment

Page 7 of 25

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

Crystal Growth & Design

2.2 Characterization Scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDX) were performed by using a Zeiss SupraTM 40 FESEM equipped with an In-lens SE detector and an Oxford X-Max Silicon Drift Detector (50 mm2). SEM and EDX measurements were performed at an accelerating voltage of 5 and 15 kV respectively. Transmission electron microscopy (TEM) and selected area electron diffraction (SAED) were performed by using a JEOL JEM-2010F TEM at an accelerating voltage of 200 kV. X-ray diffraction (XRD) was performed by using a Bruker D8 Advance X-ray diffractometer using Cu Kα radiation (λ = 0.15406 nm). Raman spectroscopy was performed by using a Horiba Scientific LabRAM HR Evolution system using an Ar laser (λ = 514 nm).

7 ACS Paragon Plus Environment

Crystal Growth & Design

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

3 Results and Discussion 3.1 Sulfide precursors do not have a strong influence on the final phase and phase transformation of NCs SEM images (Figures S1a2 and S1b2) of samples obtained at the end of the reaction (TAA170-30 and TU170-30) showed that both samples comprised nanometer-sized particles only. TEM of the particles revealed that those synthesized using TAA sulfide precursor (TAA170-30) had a mean size of 10.9 ± 3.5 nm (standard deviation) while those synthesized using TU sulfide precursor (TU170-30) had a mean size of 61.5 ± 11.4 nm. HRTEM of both samples showed bright fringes with a periodicity corresponding to the (112) d-spacing of kesterite CZTS (JCPDS 01-075-4122). SAED also showed diffraction rings matching the (112), (220), and (312) planes of kesterite CZTS (Figure 1). The XRD and Raman peaks (Figures 2 and 3) of the same samples were dominated by the characteristic peaks of CZTS.31 These characterizations demonstrate the successful heat-up synthesis of kesterite CZTS NCs in a formamide solvent system. The results also show that the synthesizability of CZTS NCs is unaffected by the different properties between TAA and TU sulfide precursors.

8 ACS Paragon Plus Environment

Page 8 of 25

Page 9 of 25

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

Crystal Growth & Design

Figure 1. (1) TEM (top left inset: size histogram), (2) HRTEM, and (3) SAED of (a) TAA170-30 and (b) TU170-30.

9 ACS Paragon Plus Environment

Crystal Growth & Design

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

Phase transformation can be revealed by characterizing samples obtained at varying stages of synthesis. The Raman spectra of NC samples synthesized using TAA or TU showed that those synthesized at lower temperatures (TAA100-0 and TU140-0) were dominated by a Raman peak at approximately 475 cm-1 (Figure 3). This peak corresponds to Cu2-xS.31 As the temperature and reaction time were increased, the ratio of CZTS to Cu2-xS peaks increased (TAA140-0 and TU170-0). At higher temperatures and longer reaction times, the Cu2-xS peak was no longer detected, leaving only the characteristic Raman peaks of CZTS (TAA170-0, TAA170-30, and TU170-30). These findings suggest that Cu2-xS was first formed at lower temperatures and steadily consumed and that the amount of CZTS was smaller initially and increased as the temperature and reaction time increased. The similar trends observed for both sets of samples suggest that the different properties between TAA and TU sulfide precursors do not affect phase transformation during synthesis.

It is interesting to compare the phase transformations observed during the synthesis of our NCs with literature. It was reported that wurtzite CZTS formation initiates by the formation of copper sulfide (Cu2-xS),15,32,33 similar to our observation that Cu2-xS was the first species formed. However, our subsequent observations differ from the literature on wurtzite CZTS. Li et al. reported that the formation of wurtzite CZTS next proceeded by the asynchronous doping of Zn2+ and Sn4+ into Cu2-xS, with Zn2+ doped preferentially at first.15 Regulacio et al. reported that wurtzite CZTS nucleates and grows on the Cu2-xS seeds directly and subsequently converts the Cu2-xS into wurtzite CZTS.32 Tan et al. reported that it first proceeded by the diffusion of Sn4+ into Cu2-xS to form Cu3SnS4, followed by the diffusion of Zn2+ to form CZTS.33 In this work, EDX measurements (Figure 6), which showed that the NCs were Sn-rich and Zn-poor during the initial stages of growth, and XRD and Raman results (Figures 2 and 3), which did not detect a Cu3SnS4 phase, suggest that the formation of

10 ACS Paragon Plus Environment

Page 10 of 25

Page 11 of 25

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

Crystal Growth & Design

our kesterite CZTS NCs from the initially formed Cu2-xS nuclei is unlike previous reports.15,32,33 It most likely proceeds by the asynchronous diffusion of Zn2+ and Sn4+ into Cu2-xS, with Sn4+ (instead of Zn2+) doped preferentially at first, to form kesterite CZTS. The underlying reasons for this difference with the literature remain unclear.

Figure 2. XRD of NCs synthesized using (a) TAA or (b) TU sulfide precursor. Bottom: characteristic XRD peaks of kesterite CZTS (JCPDS 01-075-4122).

Figure 3. Raman spectra of NCs synthesized using (a) TAA or (b) TU sulfide precursor.

11 ACS Paragon Plus Environment

Crystal Growth & Design

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

3.2 Sulfide precursors have strong influence on nucleation and growth characteristics Although the phase of the final product and the phase transformation during reaction were similar for both sulfide precursors, differences in nucleation and growth characteristics were found.

Firstly, the lowest temperature at which NCs could be obtained was lower for TAA (100°C) than for TU (140°C). Additionally, the reaction solution that utilized TAA as the sulfide precursor initiated a color change at a lower temperature, turning brown at approximately 100°C and completely black at 110°C while the reaction solution utilizing TU turned brown at approximately 140°C and was completely black at 150°C. This suggests that nucleation initiated at a lower temperature when TAA was used instead of TU.

Secondly, the particles obtained when TAA was used were smaller in size, as evidenced from SEM and TEM imaging (Figures S1 and 1). A close inspection of SEM and TEM images showed that while NCs synthesized using TAA occurred singly, those synthesized using TU were grouped in tight clusters. There is a question of whether these clusters in the TU samples are stable NCs with a number of discrete grains or aggregates of normal clusters. We then carried out additional experiments where samples from the solution were taken out at different times of 3, 30, 60, 120, 240, 360, and 480 min during synthesis. We find that the particle (cluster) sizes are almost the same. Such results may suggest that the particles are stable and composed of discrete grains. However, we also notice from literature that the aggregation of normal clusters can be slowed down when the particle size is large enough.34,35 Therefore, it is difficult to differentiate whether the larger particle size in the TU samples is due to the poorer passivation of TU compared to TAA.

12 ACS Paragon Plus Environment

Page 12 of 25

Page 13 of 25

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

Crystal Growth & Design

Thirdly, the full width at half maximum (FWHM) of the XRD peaks of NCs synthesized using TAA were almost always larger (Figure 4). This suggests that the average crystal size when TAA was used was smaller than when TU was used.

Figure 4. Full width at half maximum (FWHM) of the XRD peaks of CZTS synthesized using TAA (■) or TU (●) sulfide precursor.

13 ACS Paragon Plus Environment

Crystal Growth & Design

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 5. (a) Sulfide dissociation reactions of TAA (left) and TU (right). (b) Proposed LaMer diagram representing the change in concentration of converted TAA and TU as a function of synthesis progress. [Monomer, M] is the concentration of converted sulfide precursor, RN is the rate of formation of nuclei, Region I is the accumulation stage, Region II is the nucleation stage, and Region III is the growth stage. The superscripts a and b refer to TAA and TU respectively.

14 ACS Paragon Plus Environment

Page 14 of 25

Page 15 of 25

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

Crystal Growth & Design

The question remained as to why these three differences were present. We noted that the key dissimilarity between the two types of samples laid in the choice of sulfide precursor. The differences observed might therefore be due to the different properties between TAA and TU.

We first considered the reaction mechanism of TAA and TU precursor conversion. The conversion of TAA is known to proceed via a first-order reaction with either a hydrogen ion (acid-catalyzed) or a hydroxyl ion (base-catalyzed).36,37 As the solvent we used was formamide, which has an acid dissociation constant (pKa) of -0.48 at 20°C (for comparison, the pKa of water at 20°C is 14.169),38 the conversion of TAA in our reaction proceeds via the acid-catalyzed reaction. It is also known that at elevated temperatures (90°C), this acidcatalyzed reaction mainly occurs via the thionamide form of TAA, not its tautomeric thiolimido form.37,39 Thus, the precursor conversion mechanism of TAA under our experimental conditions proceeds by means of its thionamide form via the acid-catalyzed reaction and its detailed reaction steps during conversion is shown in Figure 5a. Of the four steps shown in Figure 5a, the rate-limiting step is the bi-molecular reaction between the protonated TAA and a nucleophile (Nu) (Step 3).39

In light of the identity of the rate-limiting step, the difference in reactivities between TAA and TU can be understood. Comparing TAA with TU, the key dissimilarity lies in the identity of one of the chemical groups attached to the central carbon (Figure 5a). In TAA, one of the chemical groups attached the central carbon is a singly bonded methyl group (–CH3) and this is replaced by an amine group (–NH2) in TU. This is significant because in both protonated TAA and TU, the central carbon is bonded to a positively charged –SH2+ group that draws electrons away by inductive effect, making the central carbon susceptible to nucleophilic attack. In TU, the substitution of the methyl group with an amine group (which is a more

15 ACS Paragon Plus Environment

Crystal Growth & Design

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

effective electron donor due to resonance effect) reduces the electron deficiency of the central carbon. This makes nucleophilic attack of the central carbon more difficult and the energy barrier for the rate-limiting bi-molecular reaction (step 3) larger for TU than for TAA. As a consequence, the precursor conversion rate of TU is much slower than that of TAA.

The implications of a faster precursor conversion rate for TAA compared to TU can be understood using the LaMer diagram (Figure 5b). The critical concentration for homogeneous nucleation is represented in Figure 5b by the horizontal line (2). When the concentration of converted precursors is higher than line (2), nucleation occurs. Due to the faster precursor conversion rate of TAA, the concentration of converted TAA reactant becomes higher than line (2) at a lower temperature compared to TU. This explains why nucleation was observed to initiate at a lower temperature when TAA (100°C) was used instead of TU (140°C).

Despite the occurrence of nucleation (which consumes converted precursors), the concentration of converted precursor continues to increase at first. This is because initially, the rate of consumption of converted precursor by nucleation and growth is slower than its rate of production. Due to a faster precursor conversion rate for TAA, it can be expected that the maximum supersaturation level attained when TAA was used is higher and reached within a shorter duration after the initiation of nucleation (compared to TU). As a consequence, faster rates of nucleation leading to the formation of more nuclei and a faster rate of consumption of converted TAA are expected. For TU, the longer duration at which the reaction remains in the nucleation stage makes heterogeneous nucleation more likely, and this may be a contributing factor for the formation of larger particles. Additionally, because a lower maximum supersaturation level is attained (due to TU’s slower precursor conversion rate), the nucleation rate is slower. This causes fewer converted TU to be consumed via

16 ACS Paragon Plus Environment

Page 16 of 25

Page 17 of 25

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

Crystal Growth & Design

nucleation, leaving more converted TU available for growth. This may account for the larger crystal sizes obtained for TU than for TAA.

3.3 Sulfide precursors have strong influence on elemental composition

Fs afdas df

Figure 6. Ratio of metal ions for NCs synthesized using (a) TAA or (b) TU. The horizontal dotted lines represent the elemental ratios of stoichiometric CZTS.

Significant differences in elemental composition also exist between NCs synthesized using TAA or TU (Figure 6). For NCs synthesized using TAA, the elemental composition was markedly different from stoichiometric CZTS initially, being Zn-poor and Sn-rich. This may be due to the asynchronous diffusion of Zn2+ and Sn4+ into Cu2-xS, with Sn4+ doped preferentially at first. After maintaining the reaction at 170°C for 30 min, the composition of the CZTS NCs (TAA170-30) became similar to that of stoichiometric CZTS. Maintaining the reaction for a longer duration (60 min, TAA170-60) caused no significant change to the stoichiometry (compared to TAA170-30). The convergence of the measured NC composition towards that of stoichiometric CZTS at longer reaction durations is evidence that the diffusion of Zn2+ and Sn4+ into Cu2-xS is the main process by which kesterite CZTS is formed in our system since diffusion is a time-dependent process. Also, the stable CZTS NC

17 ACS Paragon Plus Environment

Crystal Growth & Design

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

stoichiometry despite increased reaction durations (from TAA170-30 to TAA170-60) is an advantage for CZTS NC heat-up synthesis because it facilitates good control over the NC’s elemental composition.

For NCs synthesized using TU, the elemental composition was also markedly different from stoichiometric CZTS initially (TU140-0 and TU170-0). However, unlike TAA, the NCs remained off-stoichiometric despite maintaining the reaction at 170°C for 30 min. Increasing the duration from 30 min to 60 min (TU170-60) and 120 min (TU170-120) did not cause the elemental composition of the NCs to approach that of stoichiometric CZTS. Rather, when normalized to Cu, it is clear that the NCs remained Zn-poor, though NCs maintained at 170°C for longer durations did become less Zn-poor (TU170-120) (Figure S2). This suggests that though the diffusion of Zn2+ and Sn4+ into Cu2-xS to form CZTS does occur, the rate of Zn2+ diffusion is much slower for the TU samples compared to the TAA samples.

There are two possible reasons for the much slower rate of Zn2+ diffusion observed in the TU samples. The first possibility is that the larger particle size in the TU samples increases the diffusion distance that Zn2+ has to travel to reach the centre of the particle. Due to the asynchronous diffusion of Zn2+ and Sn4+ into Cu2-xS, with Sn4+ doped preferentially at first, it is possible that this difference in diffusion rates is accentuated when the diffusion distance is increased. The second possibility is that the Zn-TU complex is more stable than the Zn-TAA complex. Consequently, the release of Zn2+ ions from the Zn-TU complex into the NCs occurs much more slowly.

To elucidate the cause, additional experiments were performed. To investigate the first possibility, the amount of TU sulfide precursor added was increased to increase TU

18 ACS Paragon Plus Environment

Page 18 of 25

Page 19 of 25

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

Crystal Growth & Design

concentration and the rate of TU conversion during synthesis. The other reaction conditions were the same as those used to synthesize TU170-30. While this adjustment did lead to smaller CZTS particles (Figures S3a-b), the amount of Zn present was not increased (Figure S3c). Hence, it is unlikely that the much slower rate of Zn2+ diffusion in the TU samples is due to larger particle size.

To investigate the second possibility, we increased the amount of Zn precursor added. Doing so increases the Zn-TU complex concentration, favoring the forward reaction involving the release of Zn2+ ions from the Zn-TU complex into the NCs. In our optimized procedure, where the Zn2+concentration was increased by 8 times, CZTS NCs with a composition similar to that of stoichiometric CZTS were obtained (Figures 6 and S4). (The TU concentration was also increased by three times to ensure that the amount of TU is in excess. The other parameters were the same as those used to synthesize TU170-30). The results thus show that the much slower rate of Zn2+ diffusion in the TU samples is due to the higher stability of the Zn-TU complex compared to the Zn-TAA complex. These findings are summarized in Figure 7.

Increasing the reaction volume to 50 ml yielded approximately 3 g of CZTS NCs which are of comparable quality to those of samples TAA170-30 and Zn8X, suggesting that the reaction is scalable (Figures 2a, 3a, S4, S5a, and S5b). CZTS NC inks were prepared and used to fabricate thin films of CZTS using the same method described previously.6 Annealed thin films exhibited sharp XRD and Raman peaks characteristic of large-grained CZTS and no secondary phases of CZTS were detected (Figure S5). This shows that the particles behave the same as those prepared previously.6

19 ACS Paragon Plus Environment

Crystal Growth & Design

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 7. Effect of reaction conditions on synthesized NCs.

3.4 Significance and possible applications of findings Our findings that sulfide precursors have strong influence on nucleation and growth characteristics and elemental composition have important implications for the heat-up synthesis of CZTS NCs as well as for other multinary compounds that have three or more elements in their structure. In this study, the use of less reactive TU formed larger particles and particle size decreased with increased starting TU concentration (Compare Figures S1b2, S3a, and S3b). The formation of large particles is disadvantageous if the CZTS NCs are to be used for solution-processing because larger particles have poorer dispersion stabilities. To avoid this, strategies including using sulfide precursors with faster reaction kinetics or increasing the amount of sulfide precursors added can be used.

With regard to elemental composition, when synthesizing multinary compounds comprising three or more elements, especially for those of which their performance is strongly dependent on their elemental composition, it is important to consider how changing sulfide precursors alters the relative stability of different metal-ligand complexes. In the case of quaternary

20 ACS Paragon Plus Environment

Page 20 of 25

Page 21 of 25

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

Crystal Growth & Design

CZTS, this change altered the rate of release of metal ions during synthesis, affecting its elemental composition. Understanding how the relative stability of different metal-ligand complexes is altered facilitates rational adjustment of elemental composition.

4 Conclusions This pioneering effort to investigate and understand the conditions necessary to synthesize CZTS NCs using a heat-up method in a formamide solvent system is successful. It is found that sulfide precursor choice does not affect the final phase and phase transformation of NCs. However, unlike previous reports, it is also found that changing sulfide precursors can influence the elemental composition of the products. The reasons are investigated and understood. Based on this understanding, CZTS NCs with small size and an elemental composition close to stoichiometric CZTS have been synthesized for both TAA and TU in multigram quantities with high yield. Our findings show that using sulfide precursors with different reactivities more than just changes NC concentration and size. Rather, in order for the other properties of the NCs to be kept unchanged, the influence of sulfide precursor selection on other nucleation and growth characteristics as well as on metal-ligand complex stability must also be considered, and the latter is especially important for multinary compounds comprising more than two elements.

21 ACS Paragon Plus Environment

Crystal Growth & Design

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

5 Acknowledgements This work is supported by Singapore Ministry of Education Academic Research Fund Tier 2 MOE2013-T2-2-138, grant number R284-000-125-112.

6 Supporting Information EDX, SEM, XRD, and Raman specta of nanocrystals. This material is available free of charge via the Internet at http://pubs.acs.org.

7 References (1) Katagiri, H.; Ishigaki, N.; Ishida, T.; Saito, K. Jpn. J. Appl. Phys. 2001, 40, 500-504. (2) Wadia, C.; Alivisatos, A. P.; Kammen, D. M. Environ. Sci. Technol. 2009, 43, 2072-2077. (3) Wang, W.; Winkler, M. T.; Gunawan, O.; Gokmen, T.; Todorov, T. K.; Zhu, Y.; Mitzi, D. B. Adv. Energy Mater. 2013, 4, 1301465. (4) Katagiri, H. In IEEE 3rd International Conference on Photonics Pulau Pinang, Malaysia, 2012, p 345-349. (5) Zhou, H.; Hsu, W.-C.; Duan, H.-S.; Bob, B.; Yang, W.; Song, T.-B.; Hsu, C.-J.; Yang, Y.

Energy Environ. Sci. 2013, 6, 2822-2838. (6) Huang, T. J.; Yin, X.; Tang, C.; Qi, G.; Gong, H. J. Mater. Chem. A 2015, 3, 1778817796. (7) Kwon, S. G.; Hyeon, T. Small 2011, 7, 2685-2702. (8) van Embden, J.; Chesman, A. S. R.; Jasieniak, J. J. Chem. Mater. 2015, 27, 2246-2285. (9) Timonen, J. V. I.; Seppälä, E. T.; Ikkala, O.; Ras, R. H. A. Angew. Chem. Int. Ed. 2011,

50, 2080-2084. (10) Park, J.; An, K.; Hwang, Y.; Park, J.-G.; Noh, H.-J.; Kim, J.-Y.; Park, J.-H.; Hwang, N.M.; Hyeon, T. Nature Mater. 2004, 3, 891-895.

22 ACS Paragon Plus Environment

Page 22 of 25

Page 23 of 25

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

Crystal Growth & Design

(11) Joo, J.; Yu, T.; Kim, Y. W.; Park, H. M.; Wu, F.; Zhang, J. Z.; Hyeon, T. J. Am. Chem.

Soc. 2003, 125, 6553-6557. (12) Steinhagen, C.; Panthani, M. G.; Akhavan, V.; Goodfellow, B.; Koo, B.; Korgel, B. A. J.

Am. Chem. Soc. 2009, 131, 12554-12555. (13) Kameyama, T.; Osaki, T.; Okazaki, K.-i.; Shibayama, T.; Kudo, A.; Kuwabata, S.; Torimoto, T. J. Mater. Chem. 2010, 20, 5319-5324. (14) Yang, H.; Jauregui, L. A.; Zhang, G.; Chen, Y. P.; Wu, Y. Nano Lett. 2012, 12, 540-545. (15) Li, M.; Zhou, W.-H.; Guo, J.; Zhou, Y.-L.; Hou, Z.-L.; Jiao, J.; Zhou, Z.-J.; Du, Z.-L.; Wu, S.-X. J. Phys. Chem. C 2012, 116, 26507-26516. (16) Chesman, A. S. R.; Duffy, N. W.; Peacock, S.; Waddington, L.; Webster, N. A. S.; Jasieniak, J. J. RSC Adv. 2013, 3, 1017-1020. (17) Chesman, A. S. R.; van Embden, J.; Duffy, N. W.; Webster, N. A. S.; Jasieniak, J. J.

Cryst. Growth Des. 2013, 13, 1712-1720. (18) LaMer, V. K.; Dinegar, R. H. J. Am. Chem. Soc. 1950, 72, 4847-4854. (19) Hendricks, M. P.; Campos, M. P.; Cleveland, G. T.; Jen-La Plante, I.; Owen, J. S.

Science 2015, 348, 1226-1230. (20) Tiong, V. T.; Zhang, Y.; Bell, J.; Wang, H. CrystEngComm 2014, 16, 4306-4313. (21) Sugimoto, T.; Shiba, F.; Sekiguchi, T.; Itoh, H. Colloids Surf. Physicochem. Eng.

Aspects 2000, 164, 183-203. (22) Sugimoto, T.; Shiba, F. Colloids Surf. Physicochem. Eng. Aspects 2000, 164, 205-215. (23) Abe, S.; Čapek, R. K.; De Geyter, B.; Hens, Z. ACS Nano 2012, 6, 42-53. (24) Owen, J. S.; Chan, E. M.; Liu, H.; Alivisatos, A. P. J. Am. Chem. Soc. 2010, 132, 1820618213. (25) Yordanov, G. G.; Dushkin, C. D.; Adachi, E. Colloids Surf. Physicochem. Eng. Aspects 2008, 316, 37-45.

23 ACS Paragon Plus Environment

Crystal Growth & Design

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

(26) He, R.; Qian, X.; Yin, J.; Zhu, Z. J. Cryst. Growth 2003, 252, 505-510. (27) Xiong, Y.; Cai, H.; Wiley, B. J.; Wang, J.; Kim, M. J.; Xia, Y. J. Am. Chem. Soc. 2007,

129, 3665-3675. (28) Lim, B.; Xiong, Y.; Xia, Y. Angew. Chem. Int. Ed. 2007, 46, 9279-9282. (29) Ren, J.; Tilley, R. D. J. Am. Chem. Soc. 2007, 129, 3287-3291. (30) Madarász, J.; Bombicz, P.; Okuya, M.; Kaneko, S. Solid State Ionics 2001, 141–142, 439-446. (31) Fernandes, P. A.; Salomé, P. M. P.; da Cunha, A. F. J. Alloys Compd. 2011, 509, 76007606. (32) Regulacio, M. D.; Ye, C.; Lim, S. H.; Bosman, M.; Ye, E.; Chen, S.; Xu, Q.-H.; Han, M.-Y. Chem. Eur. J. 2012, 18, 3127-3131. (33) Tan, J. M.; Lee, Y. H.; Pedireddy, S.; Baikie, T.; Ling, X. Y.; Wong, L. H. J. Am. Chem.

Soc. 2014, 136, 6684-6692. (34) Nagaraja, C. M.; Kaur, M. Mater. Lett. 2013, 111, 230-233. (35) Wang, W.; Shen, H.; He, X.; Li, J. J. Nanopart. Res. 2014, 16, 2437. (36) Swift, E. H.; Butler, E. A. Anal. Chem. 1956, 28, 146-153. (37) Butler, E. A.; Peters, D. G.; Swift, E. H. Anal. Chem. 1958, 30, 1379-1383. (38) Riddick, J. A.; Bunger, W. B.; Sakano, T. K. Organic Solvents: Physical Properties and

Methods of Purification; John Wiley & Sons: NY, USA, 1986. (39) Rosenthal, D.; Taylor, T. I. J. Am. Chem. Soc. 1957, 79, 2684-2690.

24 ACS Paragon Plus Environment

Page 24 of 25

Page 25 of 25

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

Crystal Growth & Design

For Table of Contents Use Only

Effect of sulfide precursor selection on the nucleation, growth, and elemental composition of Cu2ZnSnS4 nanocrystals

Tang Jiao Huang,1,2,3 Ryan Lee Guang-Ren,3 Xuesong Yin,3 Chunhua Tang,3 Guojun Qi,2 and Hao Gong*,3

1 NUS Graduate School for Integrative Sciences and Engineering, National University of Singapore, 117456 Singapore 2 Singapore Institute of Manufacturing Technology (SIMTech), 2 Fusionopolis Way, #08-04, 138634 Singapore 3 Department of Materials Science and Engineering, National University of Singapore, 117576 Singapore * Corresponding author: [email protected]

Pronounced differences are observed when quaternery Cu2ZnSnS4 nanocrystals are synthesized using different sulfide precursors. Mechanisms are proposed and validated.

25 ACS Paragon Plus Environment