Stronger Lewis Base Antisolvents Improve
Perovskite Nanocrystal Stability
Junzhi Ye a,b#*, Charlie Nicholls a#, Woo Hyeon Jeong a, Dong Yoon Chung a, Ashish Gaurav a,
Kieran De-Ville a, Rui Xu c, Zongming Ni d, Qingyu Wang e, Xinyu Shene, Jieling Tan f, Eilidh
L. Quinn a, Maxime Atkinson a, Wei Zhang f, Haitao Zhao g, Henry J. Snaith e, Robert A. Taylor
e, Yunwei Zhang c, Robert L. Z. Hoye a*
a Department of Chemistry, University of Oxford, Inorganic Chemistry Laboratory, Oxford,
OX1 3QR, United Kingdom
b Institute of Polymer Optoelectronic Materials and Devices, Guangdong Basic Research
Centre of Excellence for Energy & Information Polymer Materials, State Key Laboratory of
Luminescent Materials and Devices, School of Materials Science and Engineering, South
China University of Technology, Guangzhou 510640, China
c School of Physics, Sun Yat-sen University, Guangzhou, 510275, China
d Wenzhou Institute of Technology, Wenzhou, 325000, China
e Clarendon Laboratory, Department of Physics, University of Oxford, Oxford, OX1 3PU,
United Kingdom
f Center for Alloy Innovation and Design (CAID), State Key Laboratory for Mechanical
Behavior of Materials, Xi’an Jiaotong University, Xi’an, 710049, China
g Research Centre for Materials Intelligent Manufacturing, State Key Laboratory of Ultraprecision
Machining Technology, Department of Electrical and Electronic Engineering, The
Hong Kong Polytechnic University, Hong Kong, China

Main:
🔹 Figure 1

Optical properties of CsPbI₃ nanocrystals after purification with different antisolvents.

(a) Steady-state absorption and photoluminescence (PL) spectra of nanocrystals in colloidal solution. Data obtained by Charlie Nicholls.
(b) Comparison of PL lifetime, PL quantum yield (PLQY), and ligand density after purification with no antisolvent, ethyl acetate, and acetonitrile. Ligand density was quantified using ¹H-NMR with residual toluene as internal reference. Data obtained by Junzhi Ye, Charlie Nicholls.
(c) Time-resolved photoluminescence (TRPL) decay of nanocrystals drop-cast on Si substrates. Data obtained by Junzhi Ye, Qingyu Wang
(d–f) Transmission electron microscopy (TEM) images of nanocrystals purified with no antisolvent, ethyl acetate, and acetonitrile, respectively. Data obtained by Jieling Tan.

🔹 Figure 2

Stability and surface chemistry evolution of CsPbI₃ nanocrystals under different purification conditions.

(a) Photographs of colloidal nanocrystal solutions stored in ambient air over time. Data obtained by Maxime Atkinson.
(b) Time evolution of PL intensity for LARP- and hot-injection (HI)-synthesized nanocrystals under ambient conditions. Charlie Nicholls, Ashish Gaurav
(c) FTIR spectra of nanocrystal solutions after purification. Dong Yoon Chung
(d) Zoomed FTIR spectra highlighting ligand-related vibrational modes. Charlie Nicholls, Dong Yoon Chung
(e) TEM images showing structural degradation (δ-phase formation) in samples purified without antisolvents. Jieling Tan.

🔹 Figure 3

Mechanism of antisolvent–nanocrystal surface interactions.

(a) Schematic illustration of surface chemistry evolution during purification and aging, comparing ethyl acetate and acetonitrile.
(b) ¹H-NMR spectra of ethyl acetate: comparison between pure solvent, supernatant, and nanocrystal samples. Charlie Nicholls
(c) ¹H-NMR spectra of acetonitrile: showing chemical shift changes due to interaction with nanocrystal surface. Charlie Nicholls

🔹 Figure 4

Experimental and theoretical investigation of antisolvent binding to nanocrystal surfaces.

(a) Pb 4f XPS spectra for nanocrystals purified with no antisolvent, ethyl acetate, and acetonitrile. Junzhi Ye
(b) N 1s XPS spectra showing additional peak corresponding to acetonitrile coordination. Junzhi Ye
(c) Electron localization function (ELF) for Pb–O interaction (ethyl acetate), indicating weak interaction. Rui Xu, Yunwei Zhang
(d) ELF for Pb–N interaction (acetonitrile), showing covalent bonding characteristics. Rui Xu, Yunwei Zhang

🔹 Figure 5

LED device performance of CsPbI₃ nanocrystals purified with different antisolvents.

(a) Current density–voltage (J–V) characteristics of devices. Woo Hyeon Jeong a, Dong Yoon Chung
(b) Luminance–voltage (L–V) curves. Woo Hyeon Jeong a, Dong Yoon Chung
(c) External quantum efficiency (EQE) versus current density. Woo Hyeon Jeong a, Dong Yoon Chung
Inset: Photograph of operating devices and normalized electroluminescence spectra with corresponding CIE coordinates. Woo Hyeon Jeong a, Dong Yoon Chung



SI:
🔹 Figure S1

Frequency of reported antisolvents used for nanocrystal purification from data mining through literature.
(a) Antisolvents used for ligand-assisted reprecipitation synthesis (LARP).
(b) Antisolvents used for hot injection synthesis.
(c) Antisolvents used for other synthesis methods.
(d) Total antisolvents frequency distribution for different types of synthesis methods.

🔹 Figure S2

Normalized steady-state absorption spectra and Elliott model fitting of CsPbI₃ nanocrystals.
(a) Normalized steady-state absorption spectra confirming similar optical density and nanocrystal concentration.
(b–d) Elliott model fitting of CsPbI₃ nanocrystals purified with:
(b) no antisolvent,
(c) ethyl acetate,
(d) acetonitrile.

🔹 Figure S3

Fitted photoluminescence (PL) spectra of CsPbI₃ nanocrystals synthesized by LARP and hot injection (HI) methods.
(a–c) LARP samples purified with: no antisolvent, ethyl acetate, and acetonitrile.
(d–f) HI samples purified with: no antisolvent, ethyl acetate, and acetonitrile.

🔹 Figure S4

Thin film X-ray diffraction (XRD) patterns of LARP-synthesized nanocrystals.
(a) No antisolvent
(b) Ethyl acetate
(c) Acetonitrile
The (100) peak splitting arises from face-down nanoplatelet superlattice formation. Enhanced double peaks indicate improved size uniformity and ordering after antisolvent treatment.

🔹 Figure S5

Nanocrystal size distribution from TEM analysis.
Box plots showing size distribution of LARP nanocrystals based on TEM images (20 data points per sample).

🔹 Figure S6

Fluence-dependent photoluminescence decay of LARP nanocrystals.
(a–c) PL decay curves for samples purified with no antisolvent, ethyl acetate, and acetonitrile.
(d) Extracted average PL lifetimes as a function of excitation fluence.

🔹 Figure S7

¹H NMR spectra of nanocrystals in deuterated toluene.
(a) Full spectra for samples treated with acetonitrile, ethyl acetate, and no antisolvent.
(b) Zoomed-in region showing hydrogen signals near C=C bonds in oleic acid and oleylamine ligands.

🔹 Figure S8

Analysis of ¹H NMR spectra.
(a) Spectra normalized using residual toluene peak.
(b) Comparison of acetonitrile sources.
(c) Comparison between fresh and aged nanocrystals after acetonitrile treatment.

🔹 Figure S9

FTIR spectra of nanocrystal thin films after antisolvent treatment.
Comparison of ethyl acetate and acetonitrile treated samples.
The highlighted region shows the C≡N stretching mode associated with acetonitrile binding.

🔹 Figure S10

LED device performance of CsPbI₃ nanocrystals after purification.
(a) Current density–voltage (J–V) curves
(b) Luminance–voltage (L–V) curves
(c) External quantum efficiency (EQE) vs current density
Inset: normalized electroluminescence spectra of the devices.