Ultrathin palladium nanosheets with selectively controlled surface facets† †Electronic supplementary information (ESI) available: Synthetic details of the surfactants and ultrathin PdNSs as well as additional electron microscopic images. See DOI: 10.1039/c8sc00605a

Ultrathin two-dimensional palladium nanosheets with selectively exposed surface facets were controllably synthesized by designed functional surfactants.


Table Content
Supplementary Table S1. Summarization of all the sample information and corresponding synthetic conditions. Electronic Supplementary Material (ESI) for Chemical Science. This journal is © The Royal Society of Chemistry 2018 S2
H 2 PdCl 4 solution (10 mM) was prepared by dissolving 0.36 g of palladium (II) chloride in 20 mL of HCl solution (0.2 M) and further diluting to 200 mL with deionized water. All the reagents were of analytical reagent grade and used without further purification.

Synthesis of C n N-COOH (Br -) and C 22 N-COOH (Cl -).
In a typical synthesis of C 22 N-COOH (Br -), 3.76 g of N,Ndimethyldocosylamine (10 mM) and 1.24 g of bromoacetic acid (11 mM) were mixed in 50 ml of isopropanol and then refluxed under 95 o C for 24 h. After the removal of solvent by reduced pressure distillation, the crude product was washed with diethyl ether several times and dried in a vacuum oven overnight. The final white product of C 22 N-COOH (Br -) was accordingly obtained before use. The other surfactants with different alkyl length were synthesized by substituting N, Ndimethyldocosylamine with N,N-dimethyleicosylamine, N,N-dimethyloctadecylamine, N,N-dimethylhexadecylamine, N,N-dimethyltetradecylamine, respectively, via the similar procedures. C 22 N-COOH (Cl -) was synthesized by substituting bromoacetic acid with chloroacetic acid and following the above-mentioned procedures. The products were verified by 1 H NMR. Taking C  (m, 38H), 0.90 (t, J = 6.6 Hz, 3H).

Synthesis of C n -QA (Br -) and C 22 -QA (Cl -).
In a typical synthesis of C 22 -QA (Br -), 3.9 g of 1-bromodocosane (10 mM) and 3.54 g of trimethylamine (15 mM) were mixed in 150 ml of acetonitrile and further refluxed under 95 o C for 20 hours.
After cooling to room temperature, the solvent was removed by reduced pressure distillation. Then, the crude product was washed with diethyl ether three times and dried in a vacuum oven overnight. The cationic surfactants with different alkyl length (C 20 -QA (Br -), C 18 -QA (Br -), and C 16 -QA (Br -)) were obtained following the similar procedures by substituting 1-
2.3. Synthesis of C n -Py (Br -) and C 22 -Py (Cl -). In a typical synthesis of C 22 -Py (Br -), 3.9 g of 1-bromodocosane (10 mM) and 1.2 g of pyridine (15 mM) were mixed in 200 ml of acetonitrile and then refluxed under 95 o C for 20 hours. After cooling to room temperature, the solvent was removed by reduced pressure distillation. Then, the crude product was washed with diethyl ether and dried in a vacuum oven overnight. The pyridyl-type surfactants with different alkyl length (C 20 -Py (Br -), C 18 -Py (Br -), and C 16 -Py (Br -)) were also obtained following the similar procedures by substituting 1-bromodocosane with 1bromoeicosane, 1-bromooctadecane, 1-bromohexadecane, and 1-bromotetradecane, respectively.

Synthesis of ultrathin PdNSs
In a typical synthesis of the PdNSs{100}, 1.6 mL of H 2 PdCl 4 aqueous solution (10 mM) was added into a vial containing 5 mL of C 22 N-COOH (Br -) aqueous solution (0.05 mM) at room temperature. After homogeneous mixing, 1 mL of fresh AA aqueous solution (0.3 M) was injected into the above solution. The synthesis composition ratio is C 22 N-COOH (Br -): H 2 PdCl 4 : AA = 25: 1.6: 30. Then the vial was placed undisturbedly at 35 o C for several hours. After that, the black product was collected by centrifugation and washed several times with absolute ethanol, and then freeze-dried at -60 ºC. The Pd products synthesized by other surfactants were obtained via the similar procedures. Besides, CO-assisted synthesis of PdNSs was carried out by bubbling of CO gas into the reaction solution, instead of the addition of AA (take care when using toxic CO).

First-principles calculations
First-principles calculations were performed based on the generalized gradient approximation (GGA) with plane-wave basis sets and ultrasoft pseudopotentials, as implemented in the CASTEP code. The tolerance of the energy was set as 1 × 10 -3 eV/cell. The exchange-correlation energy and potential were described self-consistently using the Perdew, Burke, and Ernzerhof (PBE) functional. Brillouin zone integration was performed with variable number of k-points generated by Monkhorst-Pack algorithm, depending on the cell size and shape. In order to simplify the simulations, we used HCOO -*, Py-N + -Me* and Me 4 -N + * instead of C 22 N-COOH, C 22 -Py and C 22 -QA, respectively. For the supercells of Pd{100} and Pd{111}, the default value of k-point set is 3×3×1, while for the Pd{110} supercell, the default value of k-point set is 2×3×1. And the S4 interaction affinity between the functional head groups (FGs) of the surfactant and Pd planes were elucidated using the binding energy (ΔE b ), i.e., the difference between the total energy of the binding system (E T (FG/Pd{hkl})) and the sum of energy for the individual Pd plane (E T (Pd(hkl))) and functional groups (E T (FG)) as follows: The configuration of FGs under stable binding state is governed by the ΔE b of FG with Pd(hkl). A more negative ΔE b corresponds to the favorite facet for coupling with the specific FGs.

Electrochemical hydrogen evolution reaction (HER)
The electrocatalytic tests were performed on the CHI 660E electrochemical analyzer at room temperature. A three-electrodes system was used for all electrochemical tests, which consisted of a carbon rod as the counter electrode, a saturated calomel electrode as the reference electrode, and glassy carbon electrode (GCE, 0.07065 cm 2 ) as the working electrode. Typically, an ink of the catalysts was prepared by mixing 1 mg of catalysts, 4 mg of carbon black (Vulcan XC-72), 0.8 mL of ethanol and 0.2 mL of water. After sonicating for 30 min, 50 μL of Nafion solution was added and further sonicated for an additional 30 min. Then, 6 μL of the ink solution (~0.006 mg of catalyst) was dropped on the working electrode and dried at room temperature before test. Linear sweep voltammetry (LSV) was used to evaluate the electrochemical activity of different catalysts with a scan rate of 5 mV s -1 . All these results were obtained by IR compensation and all reported potentials in this work are referenced to the reversible hydrogen electrode (RHE).

Characterizations
The high-resolution thermal-field emission scanning electron microscope (

S5
respectively. 1 H NMR spectra were recorded on Avance III HD 400 spectrometer (400MHz), and the chemical shifts were reported in ppm relative to the residual deuterated solvent and the internal standard tetramethylsilane. The X-ray photoelectron spectra (XPS) were performed on a scanning X-ray microprobe (Thermo ESCALAB 250Xi) that uses Al Kα radiation. The binding energy of the C 1s peak (284.8 eV) was employed as a standard to calibrate the binding energies of other elements.
Small-angle X-ray scattering (SAXS) measurements were performed on a SAXSess mc2 apparatus with Cu Kα radiation (Anton Paar).

Fig. S1
The molecular structures and the corresponded abbreviations of the designed functional surfactants used in this work. The configuration of function groups (FGs) under stable binding state is governed by the ΔE b of FG with Pd(hkl). A more negative ΔE b corresponds to the favorite facet for coupling with the specific FGs. For instance, the HCOOonto Pd{100} plane presents the ΔE b of 2.79 eV (Fig. S3a), which is more negative than the HCOOonto Pd{110} and Pd{111} (Fig. S3b and c).

S7
This indicates that the surfactant with the FG of HCOOfavors to bind onto the Pd{100} plane as the efficient capping agent.
The chemisorption of HCOOonto Pd{100} inhibits the growth of PdNSs along Pd{100} plane, and thus facilitates the formation PdNSs{100} with {100}-exposed facets. However, due to the small difference of binding energy onto Pd{100} and Pd{110} planes, the mixed {100}/{110}-exposed facets were also observed when using the surfactants of C 22 N-COOH (Cl -), totally same to our experimental results. The results also indicate halide ion of Bralso assists the growth of PdNSs{100} with pure {100}-exposed crystal facet. Similarly, the preferential chemisorption of Py groups onto Pd{110} facets was accordingly confirmed (ΔE b of -1.48 eV for Pd{110} is lowest). However, due to steric hindrance and very weaker affinity of QA onto Pd facets, it is very difficult to distinguish their favorable exposed facets by simulation. The negatively charged PdBr 4 2would firstly interact with quaternary ammonium group in C 22 N-COOH (Br -) through the electrostatic interaction, which further self-assemble into the lamellar organic-inorganic hybrids. The presence of carboxyl groups could greatly stabilize the lamellar micelles due to the intermolecular hydrogen bonding, while the oxygen atoms in carbonyl groups and Brwould strongly interact and/or adsorb onto Pd {100} plane. Both the nanoconfined effect of lamellar micelles and preferentially chemisorbed planes of Pd determined the construction of ultrathin 2D Pd nanosheets with specific {100}-exposed facets. As indicated by TEM images of PdNSs, small piece of Pd nanosheets (<10 nm) were found at the initial stage (b), which epitaxially grew bigger only along the 2D plane direction (c-e). The irregular edges of the PdNSs intermediates (c, d) gradually grew smooth ones with square morphology, and the size also became bigger with increasing the reaction period.
The results indicated the epitaxial growth mechanism of the PdNSs, which also corresponded to the single-crystalline structure and square shape of the as-resulted nanosheets.     Well-defined PdNSs were still obtained using the surfactants of C 20 N-COOH (Br -) and C 18 N-COOH (Br -). Pd nanoplates were formed using C 16 N-COOH (Br -), while bulk Pd crystals with irregular morphology were synthesized using C 14 N-COOH (Br -) or the surfactants with shorter alkyl chains. C 18 N-COOH (Br -) could direct the formation of PdNSs, in comparison to nanoplates and bulk nanocrystals synthesized using C 18 -QA (Br -) or C 18 -Py (Br -). The results further testified that the carboxyl groups were powerful for the confined growth of 2D PdNSs, due to the strong chemisorption interactions between Pd and oxygen atoms in carboxyl groups. S25 Table S1. The summarization of all the sample information and corresponding synthetic conditions.