KP15 bulks, prepared by the gas-phase-transfer method, had a flat and smooth surface shown in Figure 1a. The X-ray diffraction patterns of the synthesized KP15 were both theoretically calculated and experimentally measured. The consistency between the two patterns shows that there is no impurity phase (Figure 1b), which confirms an excellent crystallization quality of the KP15 bulks.

Figure 1: (a) Optical microscopy result of KP15 bulks. (b) XRD results of KP15 bulks.

Measurement of the absorption coefficient and the Hansen solubility parameters for KP15

According to the Hansen’s theory [19], the dispersed concentration C of a KP15 dispersion prepared by liquid exfoliation can be expressed by Equation 1 as follows.

(1)

where δD is the intermolecular dispersion force, δH is the intermolecular hydrogen bond; δP is the intermolecular polar force; δA,D, δA,P, δA,H are the Hansen solubility parameters (HSPs) of the solute; and δB,D, δB,P, δB,H are the HSPs of the solvent. Therefore, to get a high concentration of KP15 in dispersion, the HSPs of the solvent for the exfoliation of KP15 should be close to those of KP15. A weighted average method was used to calculate the HSPs of KP15. The concentration of KP15 was used as a weight factor for each suspension. This way, the HSPs of KP15 can be expressed according to Equation 2 [19].

(2)

where δi,sol are the HSPs of the solvent and C is the concentration of the KP15 dispersions. The Lambert–Beer law (Equation 3) was then used to measure the concentration of the KP15 dispersions:

(3)

where A is the absorbance, K is the absorption coefficient of the material, b is the absorbing layer thickness (which in this work is the width of the cuvette, i.e., 1 cm), and C is the concentration of the KP15 dispersions. The absorbance A and the absorption coefficient K are related to the wavelength of the incident light. To determine A and K, it is necessary to choose a specific incident wavelength. The bandgap of bulk KP15 is approx. 1.75 eV [20]. However, according to our previous study, with thickness reduction of the KP15 nanowires, a surface-state luminescence at 693 nm gradually dominates in the KP15 nanowire [14]. This could affect light absorption properties of KP15 due to its decreased size.

To avoid the generation of concentration error caused by the absorbance influence of the surface state, a wavelength (800 nm) which is far away from the bandgap of KP15 bulk and surface state in the KP15 nanowires was chosen. Some dispersions for which we predetermined the concentration were prepared to fit and determine the absorption coefficient K. Solutions of five different concentrations of KP15 dispersions in butyrolactone were prepared by liquid exfoliation with a predetermined concentration. UV−visible absorption spectra results are shown in Figure 2. The concentration linearly varies with absorbance. The slope of this fitted linear equation is 3.86 ± 0.13. This means that the absorption coefficient of KP15 is 3.86 ± 0.13 mL·mg−1·cm−1.

Figure 2: Absorbance of predetermined KP15 dispersions exfoliated in butyrolactone. (a) Absorbance of different concentrations of predetermined KP15 dispersions exfoliated in butyrolactone. (b) Absorbance (800 nm) as a function of concentration of predetermined KP15 dispersions. The absorption coefficient (800 nm) is 3.86 ± 0.13 mL·mg−1·cm−1.

We selected 20 common solvents, including benzyl benzoate, toluene, ethyl acetate, acetone, alcohol, butyrolactone, N,N'-dimethylpropyleneurea, bromobenzene, cyclopentanone, N-dodecyl-2-pyrrolidone, glycol, vinyl acetate, hexane, isopropyl alcohol, N,N-dimethylformamide, O-phthalic dimethyl ester, dimethyl sulfoxide, N-methylpyrrolidone, water, and cyclohexanone. The HSPs of those solvents are listed in Table 1.

Table 1: Hansen parameters for the solvents [21].

solvent δD (MPa1/2) δP (MPa1/2) δH (MPa1/2) benzyl benzoate 20 5.1 5.2 toluene 18 1.4 2 ethyl acetate 15.8 5.3 7.2 acetone 15.5 10.4 7 alcohol 18.1 17.1 16.9 butyrolactone 18 16.6 7.4 N,N'-dimethylpropyleneurea 17.8 9.5 9.3 bromobenzene 19.2 5.5 4.1 cyclopentanone 17.9 11.9 5.2 N-dodecyl-2-pyrrolidone 17.5 4.1 3.2 glycol 17 11 26 vinyl acetate 16 7.2 5.9 hexane 14.9 0 0 isopropyl alcohol 15.8 6.1 16.4 N,N-dimethylformamide 17.4 13.7 11.3 O-phthalic dimethyl ester 18.6 10.8 4.9 dimethyl sulfoxide 18.4 16.4 10.2 N-methylpyrrolidone 18 12.3 7.2 water 15.8 8.8 19.4 cyclohexanone 17.8 8.4 5.1

Figure 3 exhibits the concentrations of KP15 dispersions exfoliated in different solvents. Cyclopentanone and butyrolactone were more suitable than the other solvents to exfoliate KP15. Figure 4 shows the relationship between the HSPs of different solvents and the concentration of the KP15 suspension. Based on Equation 2, the HSPs of KP15 were δD = 17.60 MPa1/2, δP = 11.19 MPa1/2, and δH = 8.95 MPa1/2. As long as the difference between the HSPs of KP15 and the HSPs of a given solvent is reduced, τ can be reduced with an improved exfoliation efficiency. Figure 5 shows the concentration of KP15 dispersions as a function of τ. When τ tends to zero, the concentration of the KP15 dispersion reaches the maximum value, which corresponds to the results of the aforementioned equation.

Figure 3: UV−visible spectrum of the KP15 dispersions using various solvents.

Figure 4: Concentration of KP15 dispersions as a function of the Hansen parameters. (a) Concentration of KP15 dispersions as a function of δD. (b) Concentration of KP15 dispersions as a function of δP. (c) Concentration of KP15 dispersions as a function of δH.

Figure 5: Concentration of KP15 dispersions as a function of τ, τ = [(δA,D − δB,D)2 + (δA,P − δB,P)2/4 + (δA,H − δB,H)2/4].

Liquid exfoliation of one-dimensional KP15

The HSPs obtained for KP15 were δD = 17.60 MPa1/2, δP = 11.19 MPa1/2, and δH = 8.95 MPa1/2. We chose a mixed solution containing water and acetone to exfoliate KP15. The HSPs of water were δD = 15.8 MPa1/2, δP = 8.8 MPa1/2, and δH = 19.4 MPa1/2. The HSPs of acetone were δD = 15.5 MPa1/2, δP = 10.4 MPa1/2, and δH = 7.0 MPa1/2. The HSP range of a mixed solution of water and acetone can cover the HSPs of KP15, however, both of them can be easily removed. The HSPs (δi) in a mixed solution containing water and acetone can be expressed by Equation 4.

(4)

where ϕi,comp is the volume fraction of the corresponding solvent and δi,comp is the HSPs of the solvent. The concentration of the KP15 dispersion can be measured by the Lambert–Beer law (Equation 3). As shown in Figure 6a, by tuning the volume fraction of acetone in the mixed solution, the HSPs of the mixed solution can be close to those of KP15, and the exfoliation efficiency can be clearly improved. The concentration values of the KP15 suspension in the solutions were 0.0268 mg·mL−1 (exfoliated in deionized water), 0.0079 mg·mL−1 (acetone), and 0.0236 mg·mL−1 (alcohol), respectively [13]. When the solvent mixture with a 50% volume fraction of acetone is used for stripping, the concentration of the KP15 dispersion finally increases to 0.0458 mg·mL−1. At this point, the parameter τ is close to the minimum value.

Figure 6: Results of KP15 dispersions exfoliated in acetone/water mixtures. (a) Absorbance of KP15 dispersions exfoliated in acetone/water mixtures with different acetone volume fractions. (b) KP15 suspension concentration and the calculated τ as a function of the acetone volume fraction.

The Raman result for the KP15 nanowires exfoliated in water−acetone mixed solution is shown in Figure 7c. At least 11 distinguishable Raman peaks located at 476.6, 453.0, 408.8, 378.3, 368.4, 354.1, 303.7, 288.5, 126.1, 114.1, and 90.7 cm−1 were seen and those Raman results were similar to the Raman modes of mechanically exfoliated KP15[11]. As shown in Figure 7d, Figure 7e, and Figure 8, the thinnest KP15 nanowires obtained by liquid exfoliation could reach 5.1 nm and had smooth boundaries. The thicknesses of 79% of the liquid-exfoliated KP15 nanowires were below 50 nm; the widths of 60.9% of the KP15 nanowires were below 100 nm. The sizes of the obtained KP15 nanowires were much smaller than those obtained in our previous studies [13]. Meanwhile, a strong temperature-dependent Raman response in exfoliated KP15 nanowires has been observed. That may help with non-invasive temperature measurements of KP15 nanodevices (details are demonstrated in Supporting Information File 1).

Figure 7: (a) Height distribution of KP15 nanowires after liquid exfoliation. (b) Cross sections of KP15 nanowires after liquid exfoliation. (c) Raman spectra of KP15 nanowires after liquid exfoliation. (d) Thickness histograms of KP15 nanowires after liquid exfoliation. (e) Width histograms of KP15 nanowires after liquid exfoliation.

Figure 8: Sizes of exfoliated KP15 nanowires. (a) Cross section of the KP15 nanowire marked in the upper right corner inset image. (b) Cross section of the KP15 nanowire marked in the upper right corner inset image. (c) Cross section of the KP15 nanowire marked in the upper right corner inset image. (d) Cross section of the KP15 nanowire marked in the upper right corner inset image.