Angewandte
Research Articles
Chemie
diffraction peak from PbI (marked with *) indicates that the
further reduced PL intensity and shortened TPPL lifetimes of
the samples after passivation illustrate that the charge
transport dynamics between the perovskite and HTM layers
are also improved.
2
(
oFPEA) PbI perovskite was generated separately on the 3D
2 4
perovskite film. We estimated the FHWM of the (110)
diffraction of the 3D perovskite phase for all the samples and
found that the C-FPD/oFPEAI sample has a smaller FHWM
than the C/oFPEAI sample, suggesting again an increased
crystallite size (Figure S5).
To investigate the chemical interaction between FPD and
perovskite as well as the defect-passivation at GBs and the
surface, we performed high-resolution X-ray photoelectron
spectroscopy (XPS) analysis. Firstly, the characteristic fluo-
rine element from the FPD and oFPEAI was identified in the
corresponding samples confirming the existence of the
passivating agents. The single peaks located in the F 1s
spectra at binding energy (BE) of 687.2 eV belong to fluoro-
substitution on the phenyl group of oFPEAI (Figure S9a, blue
Atomic force microscope (AFM) and SEM measurements
were further examined to scrutinize the surface of the
perovskite films (Figure S6). The SEM results of the C/
oFPEAI and C-FPD/oFPEAI samples agreed with previous
reports showing an in situ formed 2D perovskite layer over
the 3D perovskite surface. To further get the information of
surface crystals shown in SEM, the microscale analysis of the
surface was carried out on all the samples of four conditions
by confocal laser scanning fluorescence microscopy (CLSM).
This measurement can map the emissive chemical constitu-
ents spatially by exploiting the differences in their emission
[
36]
and green). However, the signal of F 1s was not detected for
the C-FPD sample (Figure S9a, red), which may be associated
with the low FPD concentration or illustrating the inexistence
of the FPD molecule on the surface of the perovskite film. As
shown in Figure 3a, the BE at 143.5 and 138.7 eV were
[
41]
2+
spectra. For pristine 3D perovskite film, we found remnant
assigned to 4f , 4f of divalent Pb , respectively, and the
5/2
7/2
[41,42]
PbI phases (shown as green color)
presenting at the GBs
two weaker peaks at 141.8 and 137.0 eV were associated with
2
0
[45]
0
0
of perovskites (shown as red color). (Figure S7a) In contrast,
metallic Pb . We calculated the intensity ratio of Pb / (Pb +
2
+
the CLSM images of the C-FPD film show less PbI phase and
Pb ) for three modified samples and the pristine to observe
a notable tendency (Table S3). Obviously, this ratio in the
perovskite films with FPD (C-FPD and C-FPD/oFPEAI) was
reduced to 4.4% and 6.5%, respectively, indicating that the
2
lightened dark regions at GBs. (Figure S7d) This indicates
that the FPD can passivate the trap states near the surface and
along the GBs, resulting in enhanced overall film quality. In
the CLSM images of the C/oFPEAI and C-FPD/oFPEAI
films (Figure S7g–m), the (oFPEA) PbI phase forms a con-
2
+
FPD could better passivate the noncoordinated Pb thereby
reducing the formation of metallic lead. Interestingly, as
shown in Figure 3a,b, the BE of Pb 4f core level and I 3d core
level shifted about 0.3 eV to lower energy for oFPEAI treated
perovskites (C/oFPEAI and C-FPD/oFPEAI), indicating that
the thin 2D perovskite layer facilitates charge transfer at the
interface because the core levels are closer to the Fermi
2
4
tinuous layer showing as evenly distributed highly emissive
spots on the 3D perovskite, which is in agreement with our
[
36]
previous reports.
To get insight into the passivation effect of the DP
strategy, we monitored the steady-state PL and time-resolved
photoluminescence (TRPL) decay of the perovskite films
with different compositions. It is known that the typical
characteristics for successful defect passivation include higher
PL intensity and longer TRPL lifetime because of the reduced
[20,46]
level.
To further evaluate the passivation effect of different
molecules, we conducted depth profiling XPS spectra for Pb
4f of the pristine perovskite (C) and C-FPD/oFPEAI samples.
As shown in Figure 3c,d, sample C shows the vertical
composition uniformity for the Pb element. For the C-FPD/
oFPEAI sample, since the formed 2D perovskite layer is as
thin as 20 nm, the depth profile of XPS spectra shows a clear
transition into the underlying 3D perovskite after Ar etching
for 30 s (1 cycle, as shown in Figure S9b). After 2 cycles of Ar
[43,44]
non-radiative recombination sites in the perovskite films.
As shown in Figure 2b, we found that the PL intensity of the
C-FPD and C/oFPEAI samples obviously increased com-
pared with the pristine 3D perovskite film (C), indicating that
the recombination in the perovskite layer is effectively
suppressed. Besides, for the samples with oFPEAI treatment
0
0
0
2+
(
C/oFPEAI and C-FPD/oFPEAI), new peaks corresponding
etching, the Pb intensity ratio (Pb / (Pb + Pb )) in sample C
reached 13.4%, which is higher than that of the C-FPD/
oFPEAI sample (6.2%). This clearly showed that the metallic
to the (oFPEA) PbI phase at 506 nm were measured when
2
4
illuminating from the top side of the film. While from the
bottom side, only one peak at 768 nm could be observed
0
Pb was reduced due to the successful passivation of FPD on
2
+
(
Figure S8), indicating that the (oFPEA) PbI formed only on
perovskite GBs defects (i.e., noncoordinating Pb ).
2
4
the top of the 3D perovskite. This was echoed by the
significantly prolonged average TRPL lifetime from 17.3 to
The passivation of defects should considerably influence
the VOC of the devices due to the consequently enhanced
carrier concentration and suppressed non-radiative recombi-
nation. To evaluate the effect of this DP strategy on the
photovoltaic performance, we fabricated complete solar cells
based on perovskite films with and w/o the corresponding
passivating procedures. The champion device with pristine
perovskite (C) shows a higher PCE of 21.22% under reverse
scan (Figure 4a). The solar cells based on C-FPD and C/
oFPEAI perovskites exhibited improved VOC of 1.11 and
1.13 V and PCE of 21.62% and 22.52%, respectively (seen in
Table 1). After optimization, the dually-passivated (based on
5
5.8, and 129.3 ns when FPD and oFPEAI were introduced,
respectively, as shown in Figure 2d, implying suppressed
charge trapping in C-FPD and C/oFPEAI films.
As expected, the C-FPD/oFPEAI sample with the DP
strategy showed the most intensive PL and longest average
TRPL lifetime (513.3 ns, Table S2), which confirms the
synergetic passivation effect of the two passivators on differ-
ent defect sites on perovskites. We also did PL and TRPL
measurements on the samples with the HTM layers deposited
on the four kinds of perovskites. As shown in Figure 2c, e, the
8
ꢀ 2021 Wiley-VCH GmbH
Angew. Chem. Int. Ed. 2021, 60, 8303 – 8312