Ligand‐dependent aggregation‐enhanced photoacoustic of atomically precise metal nanocluster

Atomically precise metal nanoclusters (MNCs), as a potential type of photoacoustic (PA) contrast agent, are limited in application due to their low PA conversion efficiency (PACE). Here, with hydrophilic Au25SR18 (SR = thiolate) as model NCs, we present a result that weakly polar solvent induces aggregation, which effectively enhances PA intensity and PACE. The PA intensity and PACE are highly dependent on the degree of aggregation, while the aggregation‐enhanced PA intensity (AEPA) positively correlates to the protected ligands. Such an AEPA phenomenon indicates that aggregation actually accelerates the intramolecular motion of Au NCs, and enlarges the proportion of excited state energy dissipated through vibrational relaxation. This result conflicts with the restriction of intramolecular motion mechanism of aggregation‐induced emission. Further experiments show that the increased energy of AEPA originates from the aggregation inhibiting the intermolecular energy transfer from excited Au NCs to their surrounding medium molecules, including solvent molecule and dissolved oxygen, rather than restricting radiative relaxations. This study develops a new strategy for enhancing the PA intensity of Au NCs, and contributes to a deeper understanding of the origin of the PA signal and the excited state energy dissipation processes for MNCs.

only considers two complete dissipation paths of excited state energy, radiative and nonradiative relaxations, and does not take excited state energy transfer (ESET, through electron transfer and resonance energy transfer) into account.Energy transfer is a very important pathway for the deactivation of excited MNCs, such as intramolecular energy transfer from the excited core to the chromophores in ligand, [6] and intermolecular energy transfer to adjacent molecules. [7]Furthermore, the emission of MNCs in solution strongly depend on the solvent, [8] indicating that the ESET to solvent [9] is a significant deactivation route for MNCs.We reviewed reports about AIE of MNCs [5a-5i] and found two fatal deficiencies: (i) the ESET processes from excited MNC are not taken into account; (ii) there is no convincing evidence to supports the restriction of nonradiative relaxation (e.g., the decrease of PA or photothermal signals, which originate from the energy dissipated through vibrational relaxation, one type of nonradiative relaxation).Such a conclusion is obtained based on the discussion of emission lifetime, time-resolved emission, and nanosecond transient absorption spectroscopy, among others.However, these experimental results that essentially reflect changes in excited states are inevitably regulated by the ESET processes.Thus, two essential issues arise: first, whether aggregation inhibits intramolecular motions?In other words, is the PA signal of monodisperse MNCs truly stronger than that of cluster aggregates of the same concentration?Second, when studying the deactivation of the excited state of cluster aggregates, can the proportion of excited state energy consumed through the ESET processes be ignored, or is it unchanged?
MNCs are composed of a limited number of metal atoms, ranging from several to thousands, and exhibit a discrete molecular-like electronic orbital structure.Altering the number of metal atoms (even adding or removing a single atom), [10] their arrangement, [11] or substituting mono-heteroatom [12] effectively regulate the photophysical properties of MNCs.Herein, the atomically precise spherical Au 25 SR 18 (SR is thiolate), which has an Au 13 icosahedral core that is capped with six Au 2 SR 3 staple motifs (Figure 1 inset), [13] was employed as model NCs to demonstrate the regulation of PA intensity by aggregation.Ethanol induced the aggregation of Au 25 Cys 18 which enhances its PA intensity and PACE was studied in detail.Then the prevalence of such an aggregation-enhanced PA intensity (as AEPA) in Au 25 SR 18 was examined and the ligand effect on it was discussed.Finally, the adding energy source of AEPA was revealed by comparing changes in emission, photocurrent, and the yield of reactive oxygen species (ROS) with and without aggregation.

Synthesis and characterization of Au 25 Cys 18
Au 25 Cys 18 (Cys is cysteine) was synthesized using a modified route reported previously [14] and purified through fractional precipitation (for experimental details please see the Supporting Information), [15] and the yield was about 40% based on Au 3+ ion.The purified sample clearly showed characteristic absorption peaks of Au 25 Cys 18 (Figure 1A, 400, 450, 670, and 770 nm), and had a blue-shifted saddle at 600 nm and an enlarged ratio of the peak at 670 nm to the saddle at 600 nm (∼2.5), comparable with the unpurified sample (610 nm and ∼1.7).Native polyacrylamide gel electrophoresis (PAGE, 30%) analysis revealed only one red band under visible light (Figure S2), and electrospray ionization mass spectrometry (ESI-MS, Figure 1B and Figure S3) identified two ion species that were all unambiguously attributed to Au 25 Cys 18 , indicating that the sample is highly pure.
PA imaging was recorded to analyze the PA intensity change due to variations in concentration under excitation at 670 nm.A tube had to be irradiated for 7 min to collect a complete PA signal, and the retested PA intensity did not significantly change, illustrating that Au 25 Cys 18 is stable under high-intensity laser irradiation and its PA intensity does not attenuate during the test, similar to that of Au 25 SG 18 reported before [4a] ; while, there does not seem to be a correlation between the PA intensity of Au 25 Cys 18 and the pH of the aqueous solution (Figure S4).The reconstructed PA images are shown in Figure 1C inset, with PA intensity gradually enhancing with the thickening of Au 25 Cys 18 aqueous solution from 0.2 to 1.0 mg mL −1 .There is a good linear dependence between PA intensity quantitatively obtained from reconstructed images and mass concentration in the same range (Figure 1C).Combined with the absorbance at 670 nm linearly correlated with Au 25 Cys 18 's concentration (the mass extinction coefficient is ∼1.38 L⋅g −1 ⋅cm −1 ), PA intensity is also linearly dependent on absorbance (Figure S5), indicating that the proportion of energy consumed through nonradiative relaxation does not alter with the concentration of Au NCs.

Aggregation-enhanced photoacoustic of Au 25 Cys 18
The dependence of PA intensity on the aggregation degree of Au 25 Cys 18 was examined in a solution with a concentration of 0.16 mg mL −1 (absorbance is 0.22 at 670 nm).5a-5c] The particle size in different solutions with increasing f e (the volume fraction of ethanol in the mixture of water and ethanol, f e = vol EtOH ∕vol (H 2 O+EtOH) )) was recorded using dynamic light scattering (DLS), as shown in Figure 2A and Figure S6.The particle size in the aqueous solution was about 5.8 nm, which increased to 245.7 nm with an increase of f e up to 50%, indicating that the addition of EtOH was not conducive to the existence of monodisperse Au 25 SR 18 NCs.Instead, they formed larger aggregates.When the f e exceeds 50%, the particle size decreased (approximately 57.1 nm when f e = 70%), indicating that the large aggregates transformed into small and dense particles.The particle sizes in solution with f e = 0% and 70% were also measured by transmission electron microscopy (TEM, Figure 2C,D).The average particle size in the aqueous solution was about 1.7 nm, as shown in Figure 2C, which was only 0.5 nm larger than the diameter of Au 25 SR 18 without take account of the organic chain (∼1.2 nm), [13] meaning that the particles existent in the aqueous solution were monodisperse Au 25 Cys 18 NCs.In Figure 2D, some large particles (diameter is about 35 nm) appear in the solution with f e = 70%.We can see that the internal density of these larger particles is not uniform since they are composed of many smaller particles (with a diameter similar to that of the particles in aqueous solution), indicating that these large particles were formed by the aggregation of Au 25 SR 18 , rather than the fusion of NCs.Another piece of evidence that adding ethanol induces cluster aggregation is the appearance of a very strong Rayleigh scattering in the solution with f e = 70%, compared to the aqueous solution (Figure 2C,D).UV-vis-NIR spectroscopy was used to follow the aggregation degree with increasing f e (Figure 2B) as the absorbance in the 300-600 nm range increased with elevating f e from 0% to 70%, indicating the formation of larger aggregates (giving rise to most of the background scattering).Additionally, no precipitate is obtained by centrifuging the aqueous solution of Au 25 Cys 18 , but Au 25 Cys 18 can be collected by centrifugation of the aqueous solution by adding two equivalents of EtOH.All these experimental results indicate that the aggregation degree of Au NCs in the solution is significantly related to f e .
PA imaging for Au 25 Cys 18 solutions with various f e (Figure 2E) was characterized under excitation at 670 nm.PA intensity has an S-shaped enhancement (Figure 2F, black line) with the increase of aggregation degree that slowly enhancing with f e increasing from 0% to 20%, followed by faster enhancement from 20% to 50%.After an f e value of 50%, the increase slows down again.PA intensity is 1250 at f e of 70%, about 3.3-fold enhancement compared with that of its aqueous solution (about 380).PA imaging for Au 25 Cys 18 solution with higher f e was not characterized anymore because of dark precipitates appearing on tube wall after setting 30 min for gel curing, which would result in the reduction of Au 25 Cys 18 's concentration and heterogeneous PA signal in tubes.
Relative PACE was used to reflect the proportion of the excited state energy losses via nonradiative relaxation.The absolute PACE was calculated by where E aco is the acoustic energy, E abs is the absorbed energy, E laser is the laser pulse energy, A is the absorbance. [16]The relative PACE was calculated by: where E 0 , A 0 , and I 0 are the acoustic energy, absorbance, and ultrasound intensity of Au 25 SR 18 dissolved in ultrapure water, E x , A x , and I x are those of Au 25 SR 18 dissolved in mixed solvent with f e = x.Here, ultrasound intensity was substituted for acoustic energy (which is linear related to the ultrasound intensity) as the acoustic energy was difficult to quantify on our instrument.Relative PACEs of Au 25 Cys 18 solutions with various f e were shown in Figure 2F (red line), we can see that the relative PACE monotonically enlarges with the increase of f e .When f e = 70%, relative PACE had a 3.1-fold enhancement.5a-5i] It is worth noting that the above conclusion did not take account of the changes of solvent thermal properties and light scattering of aggregates.Previous theoretical and experimental studies have shown evidence that rapid heat transfer from light absorber to adjacent substances is the main contribution to efficient PA conversion, [17] and heat transfer to the surroundings could be facilitated by increasing the heat capacity of the surrounding layer. [16]From Figure S7, it is observed that the heat capacity of H 2 O-EtOH mixed solvent diminishes with the increase of f e due to both density (ρ) and specific heat capacity (C) decreasing with enlarging f e , meaning that the changes of solvent thermal properties with increasing f e inhibit PA conversion of Au NCs, and PACE increasing with the increase of f e is not directly caused by the changes of solvent thermal properties.The light scattering at 670 nm produced by the NC aggregates results in the real light absorbance of aggregates being less than the observed one (Figure 2B inset and Figure S8), indicating that the actual enhancement of PACE is larger than the calculated one for aggregates.

Ligand effect on AEPA
To investigate the prevalence of AEPA in Au 25 SR 18 and correlate it with protected ligands, two strategies were employed to alter the ligands.On the one hand, we decorated the amino (─NH 2 ) groups in Au 25 Cys 18 with alkyl acyl (n-hexanoyl) and aryl acyl (benzoyl), respectively (Figure 3A, products named as C6 and BZ), [6a] then conducted PA imaging for the solutions of C6 and BZ with various f e at length.The successful reaction between benzoyl chloride and Au 25 Cys 18 was confirmed by ESI-MS (Figure 3B and Figure S9), with few ion species observed in the m/z range of 1550-1800 Da corresponding to the intact Au 25 Cys 18 ion decorated with 8-16 benzoyl groups.The FTIR spectrum (Figure S10) also show successful decoration of benzoyl groups on the NH 2 of cysteine in Au 25 Cys 18 .Product C6 was also characterized by ESI-MS under the same conditions, with only the ion species of intact Au 25 Cys 18 ion embellished with few (0-4) alkyl acyl groups appearing in the mass spectrum (Figure 3C and Figure S11).The product C6 had good solubility in a weak alkali solution, but no useful ion species was observed in the mass spectrum under weak alkali condition (pH = 9), except the ion species of Au 25 Cys 18 (Figure S12).We further characterized the FTIR spectrum of C6.By comparing the FTIR spectra of Au 25 Cys 18 , C6 and n-hexanoyl chloride (Figure 3D), we found that the only peak at 1550 and 1780 cm −1 , which corresponded to carbonyl groups (C═O) for Au 25 Cys 18 and n-hexanoyl chloride, respectively, was replaced by two peaks at 1590 and 1500 cm −1 attributed to two kinds of C═O in C6, indicating that the amino groups in cysteines had successfully reacted with n-hexanoyl chloride.
In addition, the worse solubility of C6 in the reaction solution and dilatory dissolution rate in water indicated the success of the amidation reaction.For C6 and BZ, the formation of larger cluster aggregates with the increase of f e was suggested by the hyperchromic shift of the UV-vis absorption spectrum in the range of 300-600 nm (Figure S13a and Figure S14a), and the samples were collected by centrifugation of the aqueous solution mixed with EtOH.The PA intensity of C6 enhanced with the increase of f e (Figure 3E and Figure S13c), and the relative PACE increased with enlarging f e due to the negligible change of absorbance at 670 nm (Figure S13b).When f e = 70%, the PA intensity and relative PACE were about 2.3-fold enhancement comparable to its aqueous solution.The PA intensity of BZ enhanced with f e increasing from 0% to 50% (about 1.9-fold when f e = 50%).7d] As the change of absorbance at 670 nm was negligible (Figure S14b), the PA intensity and relative PACE had a maximum at an f e value of 50% (1.9-fold), and 1.8-fold enhancement when f e = 70% (Figure 3E and Figure S14c).On the other hand, other ligand-protected Au 25 SR 18 , including Au 25 (MPA) 18 (MPA is 3-mercaptopropionic acid), Au 25 (p-MBA) 18 (p-MBA is 4-mercaptobenzoic acid), and Au 25 SG 18 , were directly synthesized via previous routes [14,18] and purified by fractional precipitation.As shown in Figure 3F and Figure S15, the PA intensity of Au 25 (MPA) 18 , Au 25 (p-MBA) 18 , and Au 25 SG 18 were enhanced by 2.7-, 2.1-, and 2.5-fold, respectively, when the f e value was at 50%, and increased further to 3.2-, 2.3-, and 3.1-fold when the f e value was 70%.The relative PACEs of Au 25 (MPA) 18 , Au 25 (p-MBA) 18 , and Au 25 SG 18 were enhanced 3.2-, 2.3-, and 3.1-fold due to the ignored differences of the 670 nm absorbance between the solutions with f e of 0% and 70%.Such results showed that all aforementioned hydrophilic Au 25 SR 18 NCs exhibited AEPA to different degree, and the AEPA was highly dependent on the protected ligands.Specifically, (i) short ligand chains endowed Au 25 SR 18 with strong AEPA as Au 25 Cys 18 and Au 25 (MPA) 18 had the strongest PA enhancement at f e = 70% among all the hydrophilic Au 25 SR 18 NCs mentioned above; (ii) elongated organic chains of the ligands inhabited AEPA because the AEPA of Au 25 Cys 18 was stronger than that of C6, BZ, and Au 25 SG 18 ; and (iii) aryl groups were a more significant factor than alkyl groups in diluting AEPA, as the AEPA of C6 was stronger than that of BZ, and the AEPA of Au 25 (MPA) 18 was stronger than that of Au 25 (p-MBA) 18 .
Previously, an aggregation-enhanced emission and photoacoustic phenomenon has been observed in a nano-in-nano (NANO 2 ) system, which is synthesized by hydrophobic Au 25 (SC n H 2n + 1 ) 18 encapsulated in discoidal phospholipid bicelles. [19]Compared to dispersed Au NCs under the same NC concentration, the emission, PA, and absorbance of NANO 2 have been enhanced by 20-60-, 3-, and ∼10-fold ,respectively, while the PACE has been reduced to ∼1/3.This suggests that the aggregation induced the enhancement of PA, but a reduction of PACE.The different phenomenon

2.4
The origin of AEPA Not considering the ESET processes, the excited MNC deactivation through two completive pathways, radiative and nonradiative relaxations, thus we assumed that the AEPA was originated from the increased restriction of radiative relaxation.To verify this hypothesis, the emission of Au 25 SR 18 solutions with and without aggregation (f e = 70% and 0%) were compared.Previous studies showed that the absorbance of Au 25 SR 18 in the range from 600 (the saddle) to 1000 nm corresponds to the transitions in the Au 13 icosahedral core, [13a] the NIR I window emission in the range of 700-800 nm arises from the Au 2 SR 3 staple motifs, while the emission in NIR II window (1000-1200 nm) originates from the Au 13 metal core. [20]Here, both the emission spectra in the NIR I (700-900 nm) and NIR II windows (900-1500 nm) were measured under excitation at 670 nm.As shown in Figure S21, no obvious emission peak appears in the NIR I window, as well as C6 and BZ, meaning that the energy barrier prevents the core-to-staple charge transfer. [20]However, the weak emission in NIR II window excited by 670 nm (Figure 5A and Figure S22) indicated that radiative decay is an important deactivation pathway for excited state of Au 13 core.The ignored variation of emission in the NIR I and II window between monodisperse Au 25 Cys 18 and its aggregates, as well as C3, C6, C12, and BZ (Figure S21 and Figure S22) and Au 25 SG 18 reported by Wu, [5b] indicated that the increased energy of AEPA does not come from the inhibition of radiative relaxation, and the ESET processes need to be responsible for AEPA.Note that the NIR II emission of C6 did not increase, as well as C3 and C12, as shown in Figure S23, meaning that modifying the linear chain on the ligand does not adjust the emission of Au 25 SR 18 .But bright emission of BZ in the range of 900-1100 nm indicates that rational surface engineering [6a] enhances the emission of Au 25 SR 18 .This enhancement of the emission is not due to the RIM mechanism because Au 25 Cys 18 and BZ had the same PA intensity (Figure 4), providing further evidence that the increased energy of Au 25 SR 18 consumed through radiative and nonradiative relaxations should be attributed to the restriction of ESET processes.
As nonluminous cysteine undergoes no structural changes during the aggregation process, the intramolecular ESET can be ignored.Photoinduced electron transfer (PET) is an intermolecular ESET process that transfers an excited electron from a donor to an acceptor, and the efficiency of electron transfer in the self-assembled film of metal NCs highly depends on their surface ligands. [21]To assess their PET efficiency, [22] we measured the photocurrent intensities of Au 25 SR 18 NCs.Cyclic voltammetry (CV) of ITO glasses coated with an aqueous solution of Au 25 SR 18 NCs were measured with and without illumination of 665 nm laser first.As no redox peak appeared at −0.3 V in the CV curves (Figure S24), the photocurrent was measured at this potential to avoid energy loss due to photoinduced redox reaction in the solution.Figure 5B shows an immediate increase in the current (about 0.41 μA, Table 1) when the light source is turned on, indicating a PET process from the cluster.We compared the photocurrents of ITO glasses coated with aqueous solution and mixture solution (f e = 70%) of Au 25 Cys 18 with equivalent concentration and volume, and no significant variation was observed (Figure S25), illustrating that the photocurrent is independent of the aggregation degree in the coating solution.This may be because the dried film does not inherit the aggregation structure in the coating solution, a view supported by the SEM pictures of two coating ITO glasses (Figure S26).We then compared the photocurrents of ITO glasses coated with 5 μL aqueous solutions for Au 25 Cys 18 , Au 25 SG 18 , C6, and BZ.Absorbance at 670 nm of We speculate that multiple factors have restricted the intermolecular ESET processes, leading to the AEPA of Au NCs.First, either monodispersed MNC or MNC aggregates is surrounded by solvent molecules in solution, and an increase in f e restricts the ESET process from excited MNCs to solvent molecules.9b,9c,23] Thus, in solutions with the same MNC concentration, the proportion of excited MNCs deactivated by solvent relaxation decreases with the increase of ethanol content in the solvent.Second, the formation of NC aggregates in solution implies that the average distance between excited MNC and surrounding medium molecules (including solvent molecule and dissolved oxygen) increases, which also limits the ESET processes from excited state NC to surrounding medium molecules.One piece of evidence is that the yield of ROS in aqueous solution is higher than that in the mixture solution (Figure 5C), although the concentration of dissolved oxygen in water is less than that in the mixture solvent. [24]The yield of ROS in aqueous solution and mixture solution (f e = 70%) of Au 25 Cys 18 were measured with 2′,7′-dichlorodihydrofluorescein (DCFH) as fluorescent indicator. [25]DCF, as the product of DCFH oxidized by ROS, exhibits strong emission centered at 530 nm under 490 nm light excitation.To eliminate the effects of solvent changes and light absorption of MNCs on the photoluminescence of DCF, the fluorescence tests were conducted in the supernatant obtained by preparing the reaction solution into solutions with the same f e (70%) and volume, and then removing the MNCs by centrifugation.However, the ESET process to dissolved oxygen does not have primary responsibility for AEPA, as the PA intensity of two consecutive tests on the same sample remains almost unchanged.Since the PA test on the sample solution is carried out in a sealed tube, if there is a significant proportion of energy transfer by this ESET process, the first test will consume a large amount of dissolved oxygen, resulting in an increase in the PA intensity of the second test due to insufficient dissolved oxygen for transferring excited state energy.9b] For clusters with a higher PET, the aggregation has a more pronounced inhibitory effect on the intermolecular ESET process.This results in a larger proportion of energy being dissipated through nonradiative relaxation, leading to a more prominent manifestation of AEPA in the clusters.

CONCLUSION
In summary, we employed hydrophilic Au 25 SR 18 as model NCs to investigate the regulation of PA intensity by aggregation.However, the result did not conform to previous hypotheses.First, aggregation was found to be an effective PA enhancement strategy, as hydrophilic Au 25 SR 18 generated strong PA signal upon aggregation induced by nonsolvent ethanol, and PA intensity and PACE dependent on the degree of aggregation, and the AEPA positively correlated to the protected ligands.Further results demonstrated that the enhanced energy of AEPA originated from the aggregation increasing the restriction of intermolecular ESET processes from excited Au NC to its surrounding solvent and dissolved oxygen molecules, rather than inhibiting radiative relaxation, and the ligand effect on AEPA was due to the protect ligands endowing Au NCs distinct ESET efficiency.However, although we confirmed the AEPA property of MNCs from an energy perspective, we acknowledge the necessity to further unravel the underlying physical mechanisms behind this phenomenon.General information about materials and PEGA method;

F I G U R E 1
Characterization of Au 25 Cys 18 (A) UV-vis-NIR spectra of the purified and unpurified Au 25 Cys 18 samples, inset is the coreshell framework of Au 25 SR 18 without organic chains, (B) negative model ESI-MS, (C) the linear dependence between the PA intensity and the concentration, inset is 3D reconstructed PA images of sample tubes (excited by 670 nm).

F
I G U R E 2 (A) Average sizes of aggregates measured using DLS and (B) UV-vis-NIR spectra in solution with different f e , inset is absorbance at 670 nm, (C, D) TEM and Rayleigh scattering of solution with f e = 0 and 70%, inset of (C) is the particle size of cluster in aqueous solutions measured by TEM, (E) 3D reconstructed PA images, and (F) PA intensity and relative PACEs (excited by 670 nm) of Au 25 Cys 18 dissolved in solvent with different f e .

F
I G U R E 3 (A) Schematic illustration of amidation reaction between amino groups in Au 25 Cys 18 and acyl chloride, ESI-MS of (B) BZ and (C) C6 (* is Au 25 Cys 18 ), (D) FTIR spectra of Au 25 Cys 18 , C6, and n-hexanoyl chloride, the purple area represents the carbonyl group vibration regions, (E) PA intensity of C6 and BZ, and (F) Au 25 SG 18 , Au 25 MPA 18 and Au 25 ( p -MBA) 18 in mixed solvent with different f e (excited by 670 nm).

F I G U R E 4
PA intensity of Au 25 (SR)18 protected by different ligands in aqueous solution (absorbance are 0.22 and 1.0 at 670 nm, excited by 670 nm). between Neih's report and our work may be due to the NCs aggregating through various routes, hydrophilic Au NCs self-aggregate together under nonsolvent induced, while NANO 2 is spontaneously formed via hydrophobic interactions between the lipid tails and the linear thiolates, and phospholipid bicelles may be playing an important role in excited MNCs deactivation.Additionally, Neih attributed this phenomenon to RIM mechanism without any further experimental evidence.Although hydrophilic Au 25 SR 18 protected by different ligands show varying degrees of AEPA and have clear ligand effects, careful observation reveals that the PA intensity of these six Au 25 SR 18 NCs (absorbance is ∼0.22) is almost identical (about 380, as shown in Figure 4 and Figure S16), indicating that the proportion of excited state energy of F I G U R E 5 (A) NIR II emission of Au 25 Cys 18 with and without aggregation, (B) the photocurrents of Au 25 SR 18 under 665 nm laser irradiation, and (C) the yield of ROS in Au 25 Cys 18 solution with and without aggregation.monodispersed Au 25 SR 18 in aqueous solution consumed via nonradiative relaxations is not correlated with the protect ligands.We further verify this conclusion from two other aspects.First, the PA intensity of these six Au 25 SR 18 in higher concentration solutions was compared; second, we modified the amino groups in Au 25 Cys 18 with n-propionyl chloride and n-dodecanoyl chloride, functionalized the carboxyl (─COOH) groups in Au 25 Cys 18 with ethylamine, n-pentylamine and benzylamine, and decorated the ─COOH of C3 and C6 with ethylamine and n-pentylamine (schemes in Figure S17, products named as C3, C12, C2N, C5N, PhN, C3 + C2N, and C6 + C5N), respectively, and compared their PA intensity under the same absorbance.As shown in Figure 4 and Figure S18-S20, the PA signals of these Au 25 SR 18 NCs show no regular change (enhancement or weakening) when their aqueous solution with an absorbance value at 670 nm is 1.0, although their protected ligands are ordered adjusting, demonstrating that PA intensity and relative PACE of monodispersed Au 25 SR 18 are independent of its ligands.The unordered change of PA intensity in Figure 4 may be caused by the following factors: (i) each sample is not processed completely consistently; (ii) it is difficult to prepare NC solutions with exactly the same absorption; (iii) the PA values are quantified from reconstructed 3D PA images, and the selection positions also cause a certain degree of error.

Table 1 ,
The photocurrent of Au 25 SR 18 in five cycles under 665 nm laser irradiation.revealed that current increases significantly under 665 nm laser excitation for all Au 25 SR 18 samples, indicating that irradiation helps to separate electron-hole pairs in excited Au 25 SR 18 NCs, and the various photocurrent intensities of Au 25 SR 18 NCs indicate that the PET efficiency highly depends on protected ligands.The photocurrent of Au 25 Cys 18 was larger than that of C6 (0.19 μA), BZ (0.14 μA), and Au 25 SG 18 (0.36 μA), suggesting that Au 25 Cys 18 has the highest PET efficiency, corresponding to its highest AEPA among them.The current of Au 25 Cys 18 was greater than that of Au 25 SG 18 followed by C6, than BZ, mirroring the order of their weakened AEPA and indicating a positive correlation between AEPA and intermolecular ESET efficiency for Au NCs.