Insight into diffusive and convective processes affecting gold nanoparticles microclustering by multiphoton photoreduction

Journal Pre-proof Insight into diffusive and convective processes affecting gold nanoparticles microclustering by multiphoton photoreduction Tiziana Ritacco, Pasquale Pagliusi, Michele Giocondo

PII:

S0927-7757(20)31520-X

DOI:

https://doi.org/10.1016/j.colsurfa.2020.125927

Reference:

COLSUA 125927

To appear in:

Colloids and Surfaces A: Physicochemical and Engineering Aspects

Received Date:

25 September 2020

Revised Date:

13 November 2020

Accepted Date:

16 November 2020

Please cite this article as: Ritacco T, Pagliusi P, Giocondo M, Insight into diffusive and convective processes affecting gold nanoparticles microclustering by multiphoton photoreduction, Colloids and Surfaces A: Physicochemical and Engineering Aspects (2020), doi: https://doi.org/10.1016/j.colsurfa.2020.125927

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2020 Published by Elsevier.

Insight into diffusive and convective processes affecting gold nanoparticles microclustering by

ro

-p

Tiziana Ritacco1,2, Pasquale Pagliusi1,2*, Michele Giocondo2

of

multiphoton photoreduction

re

(1) Department of Physics, University of Calabria, Ponte P. Bucci 31C, Rende (CS), 87036, Italy

lP

(2) CNR Nanotec– Institute of Nanotechnology, S.S. Cosenza, 87036 Rende (CS), 87036, Italy

ur na

*E-mail: [email protected]

Jo

Graphical asbtract

1

of ro -p

Highlights



ur na



re



Size and density control of GNPs clusters created via multi-photon direct laser writing are achieved by tuning the waiting time (WT) between consecutive exposures. The role of chloroauric ion and water diffusion on the morphology of the GNPs aggregates is pointed out. WT large enough to restore ionic concentration in the exposed volume, but shorter than rehydration characteristic time, produces compact homogenous GNPs clusters. For longer WT, multiple pressure shockwaves occur upon local dehydration/rehydration, causing shattered, inhomogeneous and wider GNPs clusters.

lP



Jo

ABSTRACT

Multi-photon direct laser writing (MP-DLW) in polymeric matrices doped with a tetrachloroauric acid (HAuCl4) water solution allows creating clusters of gold nanoparticles (GNPs) inside the focus figure of a tightly focused ultrafast laser beam. The key physical phenomena involved in the process are analyzed, with the aim to assess the limits and potential of this promising technology. Multi Photon Absorption (MPA) triggers the photo-reduction of

2

AuCl4– ions and the consequent creation of GNPs in the spotted volume. Thermal electronic decays lead to a local abrupt increase of the temperature, which influences the morphology of the created structures. At the same time, two different effects take place, related to the dehydration of the polymeric matrix, and the concentration gradient of the gold precursor upon localized photoreduction. Given their different timescales, these phenomena allow for controlling the GNPs density and size dispersity when a given energy dose is delivered in multiple exposures,

of

tuning the delay between consecutive laser exposures. A simple yet effective experiment to

-p

ro

estimate the temperature distribution at the micron-scale is also proposed.

ABBREVIATIONS

re

HAuCl4 tetrachloroauric acid; GNPs Gold Nanoparticles; LSPR Localized Surface Plasmon Resonance; MNPs Metallic Nanoparticles; MP-DLW Multi-Photons Direct Laser Writing; MPA

lP

Multi-Photon Absorption; MPPR Multi-Photon Photo-Reduction; PVA Polyvinyl Alcohol; SEM

ur na

Scanning Electron Microscopy; voxel Volume pixel; HWHM half-width at half-maximum.

KEYWORDS: gold nanoparticles, 3d laser printing, multi-photon photo-reduction, diffusive

Jo

phenomena

In the field of nanotechnologies, Multi-Photon Direct Laser Writing (MP-DLW) is among the

most advanced optical techniques for creating arbitrarily complex 3D structures in organic resists, featuring details well below the diffraction limit.[1-3] This technique has been also used in “resists” containing a UV-photosensitive metallic precursor that is reduced upon exposure

3

(Multi-Photon Photo-reduction, MPPR), with the aim to create 3D assemblies of metallic nanoparticles (MNPs).[4-14] Due to their peculiar properties, i.e. localized surface plasmon resonance (LSPR) and in some case biocompatibility (e.g. gold nanoparticles), the possibility to create objects and patterns of MNPs by MP-DLW is attracting an increasing interest in many scientific and technologic fields, such as plasmonics, optical metamaterials, sensing and security devices.[15-25] Recently, this fabrication technique has been harnessed to produce platforms for

of

SERS[26-27] and thermo-plasmonics[28] which could be employed in sensing devices and in the

ro

medical field respectively. With respect to other techniques, MP-DLW allows for the direct

integration of MNPs clusters in optical devices, with nanometric precision. However, the low

-p

density, high polydispersity and irregular shapes of the MNPs created by MPPR are hindering

re

their application in high-sensitive smart-platforms, where control on the MNPs size and density is mandatory. Despite the vast body of scientific literature devoted to this fabrication method,[29fundamental insights on the collateral effects of MPPR of metal precursors in soft materials,

lP

32]

as hydro-gels or polymers, are still missing.

ur na

Other authors analyzed the mechanism of MPPR of MNPs in solid systems, such as photosensitive glasses. In particular, they observed that the MNPs clustering depends on the ionic diffusion and heat transfer, which they tuned in function of the laser power and repetition rate.[33-35]

Jo

Here, we report the analysis of the relevant physical phenomena involved in the MPPR of

tetrachloroauric acid (HAuCl4) in aqueous systems,[36] elucidating the role of thermal and diffusive processes in the creation and clustering of gold nanoparticles (GNPs). We already observed that the GNPs size distribution is affected by the exposure time.[32] The aim of this work is to assess the potential and the limitations of MPPR by DLW and to determine the

4

parameters which allow a better control on the size dispersity and density of the created GNPs clusters, in order to tune their plasmonic properties for the specific applications. When a femto-second pulsed NIR laser is tightly focused at the interface between an aqueous solution of polyvinyl alcohol (PVA) and HAuCl4 and a glass substrate,[4] MPPR occurs inside the focus region, where the multi-photon absorption (MPA) threshold is overcome (voxel).[36-38] As a consequence, Au0 atoms are created, acting as seeds for the GNPs growth, and an ionic

of

concentration gradient arises at the voxel surface. The ions surrounding the voxel are also

ro

thermally reduced because of the heat released through non-radiative relaxations, and, depending on the delivered energy density, a thermal pressure wave can be triggered at the same time,

-p

which is responsible for the local dehydration of the irradiated area.

re

From both our experiments and literature, [4,31-32] it turns out that the GNPs created upon a single laser pulse train exposure, in the 10-3–10-1s range, are polydisperse in size and their

lP

density is considerably lower than the expected one.[36] It suggests low probability of the MPPR or seeds diffusion/aggregation. The growth of the GNPs is ruled by an autocatalytic reaction

ur na

occurring on their surface,[39-40] where the remaining chloroauric ions AuCl4- are reduced, and this process is governed by the Fickian ionic diffusion. According to this model, we demonstrate a consistent increase in the GNPs cluster density by delivering the energy dose in multiple laser exposures, with a waiting time (WT) between them of the order of the ionic diffusion

Jo

characteristic time (~10-2s). During this time interval, the initial AuCl4- concentration in the voxel is recovered before each exposure, resulting in a cumulative effect for the growth of the GNPs heaps, whose radial size remains below the micron scale. Longer WTs are comparable to the typical rehydration time of the voxel and, under these conditions, the GNPs clusters are considerably larger and scattered. We suppose that in this regime, the rapid heating of the

5

irradiated volume generates a pressure wave due to local dehydration at each exposure, since rehydration of the exposed volume is allowed during the WT. Albeit a pressure wave also occurs in the single exposure, evidently, the cumulative effect of several pressure waves is relevant to the GNPSs clusters morphology. The local temperature distribution and the disruptive effect of the multiple thermal pressure waves as a function of the delivered dose, the WT and the number

of

of exposures, have been experimentally investigated.

ro

RESULTS AND DISCUSSION

Our sample consists of a PVA\HAuCl4\H2O film drop casted on top of a 170 m thick

-p

coverslip. Its bottom surface is oil coupled to a 1.4 NA objective of an inverted microscope,

re

focusing a femtosecond pulsed IR laser (780 nm wavelength) at the film-coverslip interface to a 350nm half-width at half-maximum (HWHM) beam waist (see Materials and Methods).

lP

With the aim to assess the role of chloroauric ions diffusion and water displacement on the density, size and morphology of the GNPs cluster created by MPPR, we delivered the energy

ur na

dose in a single, 10-1 s, exposure or in multiple (100), 10-3 s, exposures, each consisting of a femtosecond pulse train with 80 MHz repetition rate. WTs between exposures were chosen in the 10-1-10 s range. In fact, according to the Fick’s second law, we estimate the chloroauric ion diffusion time in the order of 10-2 s, whereas the diffusion time of water in the PVA matrix is one

Jo

order of magnitude larger.[41-44]

In Figure 1, we compare the SEM images of the GNPs spots created with low (1.0 mW, A-C) or high (25 mW, D-F) laser power, for single (A, D) and multiple (B, C, E, F) exposures, by

6

varying the WT (10-1 and 10 s). At low laser power and WT = 10-1 s (Figure 1B), compact aggregates of GNPs are created, with a considerably higher density and lower polydispersity compared to the single exposure (Figure 1A). For WT  1 s, we observe the onset of a new regime in which larger inhomogeneous clusters are created. In particular, at WT = 10 s (Figure 1 C), the GNPs cluster is shattered and features a GNPs ring. This trend becomes even more

of

evident by increasing the laser power to 25 mW, when the higher MPPR probability and voxel size leads to the creation of considerably larger GNPs clusters. In this case, the gold ablation

ro

threshold is reached in the voxel,[45] whose center is almost empty, presenting only few small (<

-p

15 nm) GNPs. At the same time, the presence of gold islands of irregular shape suggests that high temperature causes the GNPs coalescence and influences the cluster morphology.[46] The

re

single exposure creates an outermost dense annulus of small GNPs and fused gold structures in the inner region, with only few smaller GNPs inside (Figure 1D). Delivering the same dose in

lP

multiple exposures with WT = 10-1 s leads to GNPs of growing size toward the outer region of the cluster where gold islands are present (Figure 1E). For WT = 10 s (Figure 1F), the radius of

ur na

the GNPs cluster increases drastically, from 1.0 µm to 5.5 µm, and we observe an inversion on the size distribution. Larger fused gold structures occur around the central hole in the coverslip, whose radius is comparable to the beam waist, surrounded by an external annulus of GNPs of

Jo

radially decreasing size.

The experimental findings described above can be interpreted in term of the chloroauric ions and water diffusion, as illustrated in Figure 2. During the laser exposure (Figure 2A), because of the concentration gradient occurring upon MPPR, AuCl4- ions migrate toward the voxel. By applying the second Fick’s law to describe the AuCl4- diffusion (see Materials and Methods), we estimated the characteristic diffusion time of the AuCl4- ion in water, for the voxel size of

7

interest (radius 350 nm), i of the order of 10-2s. When the exposure is longer than i, AuCl4- ions migrating toward the voxel are photo-reduced as they reach its border. As a result, the seeds in the inner part of the voxel are not fed anymore and the cluster features a high density of large GNPs on the border and very few small GNPs inside. The longer is the single exposure time, the more evident is this effect. This is the typical distribution reported by many authors in creating GNPs patterns through MP-DLW 4,5,28. Vice versa, when the energy dose is delivered through

of

multiple exposures, with WT in the order of the AuCl4- ions diffusion time i or larger (Figure

ro

2B), the initial ionic concentration is recovered in the voxel before the following laser pulse train is delivered, and dense cluster of small GNPs with low polydispersity is created. A GNPs cluster

-p

figuring these characteristics is obtained by keeping the laser power at 1.0 mW, as shown in the

re

SEM image of Figure 1B. The cluster is dense, compact, and presents homogenous GNPs of (30 ± 10) nm diameter, while in all the other cases the GNPs diameter varies in the range 10-400 nm.

lP

On the other hand, fast heating of the exposed volume, caused by MPA, could produce a rapid water vaporization and expansion. As a consequence of the resulting pressure wave, the

ur na

surrounding liquid is ejected and dehydration of the PVA matrix occurs around the voxel. This phenomenon is analogous to that occurring in biological tissues upon laser surgery.[47] If WT is large enough to ensure the complete rehydration of the irradiated area (Figure 2C), a pressure wave may occur at each exposure, resulting in the blast of the created GNPs (Figures 1C and

Jo

1F). Assuming that the dehydrated area is comparable to the GNPs cluster size (radius 5.5 µm, see Figure 1F), the characteristic diffusion time w of water in the PVA matrix has been estimated by the Fickian diffusion w1s, two orders of magnitude higher than i. According to the results depicted in Figure 1 and the scenarios described above, low power multiple exposures, with a WT large enough to restore ionic concentration, but shorter than

8

rehydration characteristic time, appears as the most suitable condition to produce compact homogenous GNPs clusters by MPPR DLW.

In order to demonstrate the onset of blasting waves in the HAuCl4 aqueous solution as a consequence of multiple laser exposures, we investigate their cumulative effect on polymeric

of

fences, located at micron distance from the irradiated area (Figure 3A). The barriers are 3D

ro

printed on the glass substrate by MP-DLW in a commercial photoresist (IP-L 780, Nanoscribe, Germany) and the HAuCl4 doped solution is drop-casted on top of them (details in Materials and

-p

Methods). The 25 mW laser beam is then focused at the center of each fence, and the overall

re

energy dose is delivered in a single or in 100 exposures, by varying the WT. Confocal microscopy analysis carried out on the samples before and after irradiation is reported in Figure

lP

3. Although the minimum distance from the laser focus is 10 m, the pressure waves following irradiations can deform and, in some cases, destroy the polymeric barriers. In fact, the polymeric

ur na

barriers are deformed even after the single exposure (Figure 3B), but, in cases of multiple exposures, the destructive effect increases with the WT (Figure 3C-F). It is evident already at WT<w , for the barriers with four channels (second row from the top), which allow the faster motion of the water “jets” (Figure 3D). For longer WT, equal or larger than w (Figure 3E and

Jo

3F, respectively), the effect becomes more relevant and leads to the destruction of the barriers. In cases of multiple exposures, GNPs clusters are visible as bright dots (Figures 3C-E) or rings (Figure 3F) at the center of the barriers.

9

This result demonstrates that the blasting wave originates since the very first exposure, even if its effects pile up and become more prominent under multiple exposures. When WT is of the order of w or larger, each pulse train concurs with a pressure wave at its full power, shattering the created little heap, ejecting the GNPs and, eventually, destroying the polymeric matrix. In order to confirm the thermal origin of the pressure waves, it is important to estimate the temperature reached during MPPR. A coarse temperature mapping has been obtained using a

of

thermographic paper as the matrix for the chloroauric solution, instead of PVA (Figure 4).

ro

A temperature calibration of the thermographic paper, in term of the gray shades and patterns it turns upon heating, has been performed with a hot plate (Figure 4A-E) and a temperature

-p

controlled metallic tip (Figure 4F-I). To maintain similar condition to the experiments in Figure

re

1, the laser is focused through the glass coverslip, upon which the wetted thermographic paper lays. No darkening is observed upon laser exposures at 25 mW if the thermographic paper is

lP

wetted with bare water, confirming that laser absorption by the paper itself is negligible. On the other hand, when the thermographic paper is wetted with the chloroauric solution, the produced

ur na

“gray-scale” patterns show that the temperature in the spot center overcomes 250 °C, in agreement with the theoretical predictions obtained with a thermodynamic approach,[33,47] depending on whether the energy is delivered through a single 10-1 s exposure (Figure 4J), or through 100, 10-3 s long, exposures and WT = 10 s (Figure 4K), respectively. Figure 4L shows

Jo

the overlay between the temperature field of Figure 4K and the SEM image of the spot printed with the same exposure parameters (Figure 1F). Inside the red circle (diameter: ~5 µm) the average temperature is higher than 250 °C, which is enough to melt GNPs. [45-46]. Between the orange and red circles, the average temperature is around 150 °C, which triggers the thermal seeding of GNPs on a wide area and create a carpet of 5-10 nm GNPs around the cluster. Inside

10

the outermost ring, between the yellow and orange circles, the average temperature is about 100 °C.

At this stage, it is evident that, in order to obtain small size and well-shaped GNPs structures by MP-DLW, one should minimize both the thermal effects and the diffusion of the created

of

GNPs. The first requirement can be fulfilled by keeping the laser power at the lowest level, just above the MPPR threshold. The latter can be achieved through a proper choice of the polymeric

-p

ro

matrix, as for instance PEG-DA.[9]

re

Conclusion

In this work we elucidate some relevant physical phenomena involved in the MP-DLW of

lP

GNPs cluster by MPPR in aqueous solution of the metallic precursor, highlighting the role of the key control parameters and the limits of this technique at the same time. We point out the effects

ur na

of the ions and water diffusion on the morphology of the GNPs aggregates, i.e. GNPs size, dispersion and density, fundamental in applications exploiting plasmonic phenomena. We demonstrate that the control on the GNPs growth and clustering is improved when the energy dose is delivered in multiple exposures. Tuning the WT on the order of the AuCl4- ions diffusion

Jo

time or larger, allows to restore the initial chloroauric ions concentration in the voxel between consecutive exposures, and therefore to control the density, size and polydispersity of the created GNPs. However, when the WT is equal or larger than the rehydration characteristic time, the effects of pressure waves originating in the laser focus become destructive, strongly affecting the GNPs cluster morphology. A thermographic paper, wetted with the tetrachloroauric acid

11

solution, has allowed the temperature mapping at the microscale in the area where the GNPs cluster is created, as a function of the delivered energy. Temperatures well above 250 °C are achieved in the voxel, high enough to cause the GNPs coalescence.

Corresponding Author

ro

*(Word Style “FA_Corresponding_Author_Footnote”). * (Word Style

of

AUTHOR INFORMATION

“FA_Corresponding_Author_Footnote”). Give contact information for the author(s) to whom

MATERIALS AND METHODS

re

-p

correspondence should be addressed.

lP

Samples preparation: For the first set of experiments (Figures 1 and 3), we prepared a solution of tetrachloroauric acid tetrahydrate (HAuCl4•3H2O, molecular weight: 393.83 g/mol, density:

ur na

3.9 g/cm3) (7.5 mM) and of polyvinyl alcohol (PVA, molecular weight: 13.000–23.000 g/mol) (18 mM), in ultrapure Milli-Q water (18.2 MΩ cm). The mixture presents an absorption peak at 400 nm and no significant absorbance above 500 nm. Therefore, it is compatible with our MPDLW system based on a 780 nm erbium pulsed laser source (see below). Samples were prepared

Jo

by drop-casting 50 µL (70 mg) of the chloroauric solution on a glass coverslip (22 x 22 x 0.13 mm3). The drop is left to dry by 90 minutes, until only 0.4 mg of solution are left on the glass substrate, with the following composition: 42 wt. % of tetrachloroauric acid, 5.5 wt. % of PVA and 52.5 wt. % of deionized water. The obtained film thickness, measured by AFM, is about 1.5 µm.

12

The cylindrical-like and parabolic barriers in Figure 3 were created by MP-DLW in IP-L 780 (Nanoscribe GMBH, Germany), drop-casted on a glass coverslip. The unexposed material was washed through 25 minutes bath in PGMEA (Propylene glycol methyl ether acetate; Alfa Aesar, IL, USA) and a 5 minutes bath in IPA (Isopropyl alcohol; Alfa Aesar, IL, USA). The barriers were then rinsed with deionized water and left to dry for 3 hours. 50 μL of the chloroauric solution were drop-casted upon them on the glass substrate.

of

The experiments for the temperature mapping (Figure 4) were carried out by wetting 1 cm2

ro

chips of thermo-graphic paper with a 7.5 mM solution of tetrachloroauric acid tetrahydrate

(HAuCl43H2O) in Milli-Q water (18.2 MΩ cm), enclosed between two glass coverslips, to limit

-p

the water evaporation.

re

All the samples were prepared and experiments were performed under the laboratory conditions, at controlled temperature (22 ± 1 °C) and humidity (45% RH).

lP

MP-DLW System: The GNPs structures were created using the Nanoscribe “Photonic Professional GT”, which incorporates an erbium laser (Toptica “Femto Fiber pro NIR”) with the

ur na

following specifications: central wavelength 780 nm; pulse duration 100 fs; repetition rate 80 MHz. The laser beam is focused on the sample through a 63X (N.A. 1.4) immersion oil objective. The transversal position is controlled through a galvo scanner. Considering the beam waist of 350nm (HWHM), the pulse fluence is about 3×10 J/m2 and 8×102 J/m2, for the used

Jo

average beam power 1.0 mW and 25 mW, respectively. Diffusion model: When the laser is focused inside the HAuCl4 mixture, seeds of gold are

created because of the MPPR. This process, ruled by electronic transitions, has characteristic time of ~ 10-13 s, much smaller than the exposure time.[36] During the laser exposure, we consider that the AuCl4- ions are fully depleted inside of the voxel and a concentration gradient arises at

13

the voxel edge. Similarly, rapid dehydration of the polymer matrix occurs over a wider volume. The fundamental differential equation of diffusion in an isotropic medium is the second Fick’s law: 𝜕𝐶(𝑟⃗, 𝑡) = 𝐷 ∇2 𝐶(𝑟⃗, 𝑡) 𝜕𝑡

(1)

where D and C are the diffusion coefficient and the concentration of the diffusing substance (i.e.

of

AuCl4- and water), respectively. Given the spherical symmetry of the system, diffusion is radial

𝜕𝐶(𝑟, 𝑡) 𝜕 2 𝐶(𝑟, 𝑡) 2 𝜕𝐶(𝑟, 𝑡) = 𝐷( + ) 𝜕𝑡 𝜕𝑟 2 𝑟 𝜕𝑟

ro

and Equation 1 takes the form:

(2)

-p

If the initial concentration inside the spherical depleted region is zero, i.e. C(r
for r≥R, the solution of Equation 2 is:[48-49] ∞

re

where R is the radius of the sphere, while concentration is assumed constant outside, C(r,t)=C0

(−1)𝑛 2𝑅 𝑛𝜋𝑟 𝐷𝑛2 𝜋 2 𝑡 𝐶(𝑟, 𝑡) = 𝐶0 (1 + ∑ sin ( ) exp (− )) 𝜋𝑟 𝑛 𝑅 𝑅2

lP

(3)

𝑛=1

The concentration at the center is obtained by the limit of Equation 3 as 𝑟 → 0

ur na

∞

𝐶(0, 𝑡) = 𝐶0 (1 + 2 ∑(−1)𝑛 exp (− 𝑛=1

𝐷𝑛2 𝜋 2 𝑡 )) 𝑅2

(4)

The solutions can be expressed in terms of two dimensionless parameters 𝐷𝑡⁄𝑅 2 and 𝑟⁄𝑅 ,

Jo

whose graphs are reported in Figure 5.

The initial uniform concentration 𝐶0 of the diffusing substance (i.e. AuCl4- ions and water) in

the depleted volume is recovered with a characteristic time 𝜏 ~ 0.3 𝑅 2 ⁄𝐷

(5),

for which the concentration at the center (𝑟 = 0) reaches about 90% of 𝐶0 (Figure 5A).

14

Diffusion coefficients can be estimated by Stokes-Einstein equation 𝐷 = 𝐾𝐵 𝑇⁄6𝜋𝑎𝜂, where KB is the Boltzmann constant, a the radius of the diffusing species, and η the viscosity of the medium. For both the AuCl4- ion and water in the PVA aqueous solution, they are of the same order of magnitude (𝐷~10−11 𝑚2 𝑠 −1 ).[42-43] Therefore, characteristic times (5) for chloroauric ions diffusion and rehydration mainly differ because of the diverse dimension of the depleted volume. While photoreduction process is restricted to the laser spot (𝑅𝑖 ~0.35 𝜇𝑚), dehydration

of

of the PVA matrix occurs over the micron-sized GNP cluster (𝑅𝑤 ~5.5 𝜇𝑚). According to these

ro

values, the time dependence of the concentration at the voxel center 𝐶(0, 𝑡) in equation (4) is

re

𝜏𝑖 ≅ 4 × 10−3 𝑠 and 𝜏𝑤 ≅ 0.9 𝑠, respectively.

lP

AUTHOR INFORMATION Corresponding Author

-p

reported in Figure 5B and 5C, for both the AuCl4- ion and water, whose characteristic times are

ur na

*E-mail: [email protected] Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval

Jo

to the final version of the manuscript.

Declaration of interests

☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

15

ACKNOWLEDGMENT The research is supported by the CNR project “Material and processes BEYOND the NANO scale - Beyond Nano”, cod. PONa3-00362. The authors acknowledge the experimental support of Dr. G. Desiderio, for SEM

Jo

ur na

lP

re

-p

ro

of

characterization.

16

REFERENCES [1]

S. Engelhardt. Direct Laser Writing. In Laser Technology in Biomimetics; Schmidt, V.,

Belegratis, M. R., Eds.; Springer-Verlag: Berlin/Heidelberg, 2013. [2]

G. Von Freymann, A. Ledermann, M. Thiel, I. Staude, S. Essig, K. Busch, M. Wegener.

of

Three-Dimensional Nanostructures for Photonics. Adv. Funct. Mater. 2010, 20, 1038.

M. Roehrig, M. Thiel, M. Worgull, H. Hölscher. 3D Direct Laser Writing of Nano and

-p

[3]

ro

https://doi.org/10.1002/adfm.200901838

https://doi.org/10.1002/smll.201200308

K. Kaneko, H.B. Sun, X.M. Duan, S. Kawata. Two-Photon Photoreduction of Metallic

lP

[4]

re

Microstructured Hierarchical Gecko-Mimicking Surfaces. Small 2012, 8, 3009.

Nanoparticle Gratings in a Polymer Matrix. Appl. Phys. Lett. 2003, 83, 1426.

[5]

ur na

https://doi.org/10.1063/1.1601302

I. Izquierdo-Lorenzo, S. Jradi, P.M. Adam. Direct Laser Writing of Random Au

Nanoparticle Three-Dimensional Structures for Highly Reproducible Micro-SERS

Jo

Measurements. RSC Adv. 2014, 4, 4128. https://doi.org/10.1039/C3RA46220J [6]

R. Kuladeep, L. Jyothi, S. Chakradhar, D.N. Rao. Fabrication of Metal Nanostructures in

a Polymer Matrix Using Femtosecond Laser Writing Technique. Opt. Eng. 2014, 53, 071823. https://doi.org/10.1117/1.OE.53.7.071823

17

[7]

Y.Y. Cao, N. Takeyasu, T. Tanaka, X.M. Duan, S. Kawata. 3D Metallic Nanostructure

Fabrication by Surfactant-Assisted Multiphoton-Induced Reduction. Small 2009, 5, 1144. https://doi.org/10.1002/smll.200801179 [8]

S. Tabrizi, Y.Y. Cao, B.P. Cumming, B. Jia, M. Gu. Functional Optical Plasmonic

Resonators Fabricated via Highly Photosensitive Direct Laser Reduction. Adv. Opt. Mater.

T. Wloka, S. Czich, M. Kleinsteuber, E. Moek, C. Weber, M. Gottschaldt, K. Liefeith,

ro

[9]

of

2016, 4, 529. https://doi.org/10.1002/adom.201500568

U.S. Schubert. Microfabrication of 3D-hydrogels via two-photon polymerization of poly(2-ethyl-

re

https://doi.org/10.1016/j.eurpolymj.2019.109295

-p

2-oxazoline) diacrylates. Eur. Polym. J., 2020, 122, 109295.

[10] S. Tabrizi, Y.Y. Cao, H. Lin, B.H. Jia. Two-photon reduction: A cost-effective method

lP

for fabrication of functional metallic nanostructures. Sci. China-Phys. Mech. Astron. 2017, 60, 034201. https://doi.org/10.1007/s11433-016-0447-6

ur na

[11] A. Ishikawa, T. Tanaka. Two-Photon Fabrication of Three-Dimensional Metallic Nanostructures for Plasmonic Metamaterials. JLMN 2012, 7, 11. 10.2961/jlmn.2012.01.0002 [12] E. Blasco, J. Müller, P. Müller, V. Trouillet, M. Schön, T. Scherer, C. Barner-Kowollik,

Jo

M. Wegener. Fabrication of Conductive 3D Gold-Containing Microstructures via Direct Laser Writing. Adv. Mater. 2016, 28, 3592. https://doi.org/10.1002/adma.201506126 [13] T. Baldacchini, A.C. Pons, J. Pons, C.N. LaFratta, J.T. Fourkas, Y. Sun, M.J. Naughton.

Multiphoton laser direct writing of two-dimensional silver structures. Opt. Express 2005, 13, 1275. https://doi.org/10.1364/OPEX.13.001275

18

[14] N. Tosa, J. Bosson, G. Vitrant, P. Baldeck, O. Stephan. Fabrication of Metallic Nanowires by Two-Photon Absorption. Sci. Study Res. 2006, 7, 899. [15] V. Amendola, R. Pilot, M. Frasconi, O.M. Maragò M.A. Iati. Surface Plasmon Resonance in Gold Nanoparticles: a Review. J. Phys. Condens. Matte. 2017, 29, 203002. https://doi.org/10.1088/1361-648X/aa60f3

of

[16] S.A. Maier. Plasmonics: Fundamentals and Applications, Springer Science & Business

ro

Media LLC, USA, 2007.

-p

[17] U. Kreibig, M. Vollmer. Optical Properties of Metal Clusters, Springer, Berlin, 1995. [18] A.J. Haes, C.L. Haynes, A.D. McFarland, G.C. Schatz, R.P. Van Duyne, S. Zou.

lP

https://doi.org/10.1557/mrs2005.100

re

Plasmonic Materials for Surface-Enhanced Sensing and Spectroscopy. MRS Bull. 2005, 30, 368.

[19] Y. Xia, N.J. Halas. Shape-Controlled Synthesis and Surface Plasmonic Properties of

ur na

Metallic Nanostructures. MRS Bull. 2005, 30, 338. https://doi.org/10.1557/mrs2005.96 [20] K. Saha, S.S. Agasti, C. Kim, X. Li, V.M. Rotello Gold Nanoparticles in Chemical and Biological Sensing. Chem. Rev. 2012, 112, 2739. https://doi.org/10.1021/cr2001178

Jo

[21] E. Di Fabrizio, S. Schlücker, J. Wenger, R. Regmi, H. Rigneault, G. Calafiore, M. West, S. Cabrini, M. Fleischer, N.F. Van Hulst, M.F. Garcia-Parajo, A. Pucci, D. Cojo, C.A.E. Hauser, M. Ni. Roadmap on Biosensing and Photonics With Advanced Nano-Optical Methods. J. Opt. 2012, 18, 063003. https://doi.org/10.1088/2040-8978/18/6/063003

19

[22] S. Zeng, K.T. Yong, I. Roy, X.Q. Dinh, X. Yu, F. Luan. A Review on Functionalized Gold Nanoparticles for Biosensing Applications. Plasmonics 2011, 6, 491. https://doi.org/10.1007/s11468-011-9228-1 [23] G.E. Lio, G. Palermo, R. Caputo, A. De Luca. Opto-mechanical control of flexible plasmonic materials. J. Appl. Phys., 2019, 125, 082533. https://doi.org/10.1063/1.5055370

of

[24] G.E. Lio, G. Palermo, A. De Luca, R. Caputo. Tensile control of the thermal flow in

ro

plasmonic heaters realized on flexible substrates. J. Chem. Phys., 2019, 151, 244707. https://doi.org/10.1063/1.5130725

-p

[25] A. Royon, K. Bourhis, M. Bellec, G. Papon, B. Bousquet, Y. Deshayes, T. Cardinal, L.

re

Canioni. Silver Clusters Embedded in Glass as a Perennial High Capacity Optical Recording Medium. Adv. Mater. 2010, 22, 5282. https://doi.org/10.1002/adma.201002413

lP

[26] V.G. Ivanov, N. D. Todorov, L.S. Petrov, T. Ritacco, M. Giocondo, E.S. Vlakhov. Strong Surface Enhanced Raman Scattering From Gold Nanoarrays Obtained by Direct Laser

ur na

Writing. Journal of Physics: Conference Series 2016, 764. https://doi.org/10.1088/17426596/764/1/012023

[27] M.R. Lee, H.K. Lee, Y. Yang, C.S.L. Koh, C.L. Lay, Y.H. Lee, I.Y. Phang, X.Y. Ling.

Jo

Direct Metal Writing and Precise Positioning of Gold Nanoparticles within Microfluidic Channels for SERS Sensing of Gaseous Analytes. ACS Appl. Mater. Interfaces 2017, 9, 39584. https://doi.org/10.1021/acsami.7b11649

20

[28] G. Palermo, T. Ritacco, D.M. Aceti, L. Pezzi, M. Giocondo, A. De Luca. Photo-Thermal Effects in 1D Gratings of Gold Nanoparticles. Crystals 2017, 7, 14. https://doi.org/10.3390/cryst7010014 [29] S. Maruo, J.T. Fourkas. Recent Progress in Multiphoton Microfabrication. Laser Photon. Rev. 2008, 2, 100. https://doi.org/10.1002/lpor.200710039

of

[30] W.E. Lu, Y.L. Zhang, M.L. Zheng, Y.P. Jia, J. Liu, X.Z. Dong, Z.S. Zhao, C.B. Li, Y.

ro

Xia, T.C. Ye, X.M. Duan. Femtosecond Direct Laser Writing of Gold Nanostructures by Ionic Liquid Assisted Multiphoton Photoreduction. Opt. Mater. Express 2013, 3, 1660.

-p

https://doi.org/10.1364/OME.3.001660

re

[31] Y.Y. Cao, X.Z. Dong, N. Takeyasu, T. Tanaka, Z.S. Zhao, X.M. Duan, S. Kawata. Morphology and Size Dependence of Silver Microstructures in Fatty Salts-Assisted Multiphoton

009-5270-7

lP

Photoreduction Microfabrication. Appl. Phys. A 2009, 96, 453. https://doi.org/10.1007/s00339-

ur na

[32] T. Ritacco, L. Ricciardi, M. La Deda, M. Giocondo. Controlling the optical creation of gold nanoparticles in a PVA matrix by direct laser writing. J. Eur. Opt. Soc-Rapid., 2016, 11. https://doi.org/10.2971/jeos.2016.16008

Jo

[33] M. Bellec, A. Royon, B. Bousquet, K. Bourhis, M. Treguer, T. Cardinal, M. Richardson, L. Canioni. Beat the diffraction limit in 3D direct laser writing in photosensitive glass. Opt. Express 2009 17, 10304. https://doi.org/10.1364/OE.17.010304 [34] J.C. Desmoulin, Y. Petit, L. Canioni, M. Dussauze, M. Lahaye, H.M. Gonzalez, E. Brasselet, T. Cardinal. Femtosecond laser structuring of silver-containing glass: Silver

21

redistribution, selective etching, and surface topology engineering. J. Appl. Phys. 2015, 118, 213104. https://doi.org/10.1063/1.4936233 [35] A.A. Khalil, J.P. Bérubé, S. Danto, J.C. Desmoulin, T. Cardinal, Y. Petit, R. Vallée, L. Canioni. Direct laser writing of a new type of waveguides in silver containing glasses. Sci. Rep. 2017, 7, 11124. https://doi.org/10.1038/s41598-017-11550-0

of

[36] M. Sakamoto, M. Fujistuka, T. Majima. Light as a Construction Tool of Metal

ro

Nanoparticles: Synthesis and Mechanism. J. Photoch. Photobio. C 2009, 10, 33. https://doi.org/10.1016/j.jphotochemrev.2008.11.002

-p

[37] J. Kimling, M. Maier, B. Okenve, V. Kotaidis, H. Ballot, A. Plech, Turkevich Method for

https://doi.org/10.1021/jp061667w

re

Gold Nanoparticle Synthesis Revisited. J. Phys. Chem. B 2006, 110, 15700.

lP

[38] V. Tribuzi, D.S. Corrêa, W. Avansi, C. Ribeiro, E. Longo, C.R. Mendonça. Indirect Doping of Microstructures Fabricated by Two-Photon Polymerization With Gold Nanoparticles.

ur na

Opt. Express 2012, 20, 21107. https://doi.org/10.1364/OE.20.021107 [39] G. Festag, A. Steinbrück, A. Csáki, R. Möller, W. Fritzsche. Single particle studies of the autocatalytic metal deposition onto surface-bound gold nanoparticles reveal a linear growth

Jo

Nanotechnology 2006, 18, 015502. https://doi.org/10.1088/0957-4484/18/1/015502 [40] K.R. Brown, L.A. Lyon, A.P. Fox, B.D. Reiss, M.J. Natan. Hydroxylamine Seeding of

Colloidal Au Nanoparticles. 3. Controlled Formation of Conductive Au Films. Chem. Mater. 2000, 12, 2. https://doi.org/10.1021/cm980066h

22

[41] F. Pan, J. Ma, L. Cui, Z. Jiang. Water vapor/propylene sorption and diffusion behavior in PVA–P(AA-AMPS) blend membranes by GCMC and MD simulation. Chem. Eng. Sci. 2009, 64, 24, 5192. https://doi.org/10.1016/j.ces.2009.08.026 [42] D.J.T. Hill, A.K. Whittaker, Zainuddin. Water diffusion into radiation crosslinked PVA– PVP network hydrogels. Radiat. Phys. Chem. 2011, 80, 2, 213.

of

https://doi.org/10.1016/j.radphyschem.2010.07.035

ro

[43] Y. Ōyanagi, M. Matsumoto. Viscosity of moderately concentrated aqueous solutions of polyvinyl alcohol. Journal of Colloid Science 1962, 17, 426. https://doi.org/10.1016/0095-

re

-p

8522(62)90054-5

[44] A. Yamasaki, T. Iwatsubo, T. Masuoka, K. Mizoguchi. Pervaporation of ethanol/water

lP

through a poly(vinyl alcohol)/cyclodextrin (PVA/CD) membrane. J. Membr. Sci. 1994, 89, 111.

ur na

https://doi.org/10.1016/0376-7388(93)E0217-8

[45] A. Plech, V. Kotaidis, M. Lorenc, J. Boneberg. Femtosecond laser near-field ablation from gold nanoparticles. Nat. Phys. 2006 2, 44. https://doi.org/10.1038/nphys191 [46] H. Sun, M. Yu, X. Sun, G. Wang, J. Lian. Effective Temperature Sensing by Irreversible

Jo

Morphology Evolution of Ultrathin Gold Island Films. J. Phys. Chem. C 2013, 117, 3366. https://doi.org/10.1021/jp310405b [47] B. Majaron, P. Plestenjak, M. Luka. Thermo-Mechanical Laser Ablation of Soft Biological Tissue: Modeling the Micro-Explosions. Appl. Phys. B 1999, 69, 71. https://doi.org/10.1007/s003400050772

23

[48] J. Comyn. Introduction to polymer permeability and the mathematics of diffusion, Polymer permeability, Springer, Dordrecht 1985.

Jo

ur na

lP

re

-p

ro

of

[49] J. Crank The mathematics of diffusion, Clarendon Press, Oxford 1975.

24

of ro -p re

Jo

ur na

lP

Figure 1. Back-scattering SEM imaging of GNPs clusters created by MPPR at 1.0 mW (A-C) and 25 mW (D-F), with an overall exposure time of 10-1s. (A) Single, 10-1s, exposure: a cluster of GPNs is created. Some thermally reduced GNPs are recognizable outside the voxel. (B) 100, 10-3 s long, exposures and WT = 10-1 s: a larger and denser cluster of GNPs is created by allowing for the replenishment of the initial ionic concentration after each pulse train. (C) 100, 10-3 s long, exposures and WT = 10 s: thermal effects become evident and the GNPs cluster is shattered. At higher laser power (i.e. 25 mW), thermal effects are amplified, and the GNPs clusters become even larger. (D) Single, 10-1s, exposure yields a continuous ionic diffusion toward the exposed area and the ions are reduced at its edge. Smaller, thermally reduced, GNPs are found in the outer ring. (E) 100, 10-3 s long, exposures and WT = 10-1s, generate an aggregate of GNPs, whose size grows outward, consistently with a weak pressure waves regime. (F) 100, 10-3 s long, exposures and WT = 10 s: the local heating creates a full pressure wave at each pulse train, that radially ejects GNPs, with the smallest ones further away. Due to the high temperature reached during the MP-DLW, GNPs coalesce and create gold islands of irregular shape.

25

of ro

Jo

ur na

lP

re

-p

Figure 2. Scheme of the GNPs creation by MP-DLW. (A) Single exposure of 10-1s: the GNPs are created immediately inside of the focus figure, while the migrating ions are photo-reduced at its border. As a result, the cluster is strongly polydisperse, with a denser edge of bigger GNPs, while only few small GNPs occur at its center. (B) 100 exposures of 10-3 s and WT = 0.1s: the initial ionic concentration inside of the focus figure is restored before the successive exposure, thus new seeds and GNPs are created in the voxel, leading to low polydispersity. (C) 100 exposures of 10-3 s and WT = 10 s, thermal effects become evident and the cluster is shattered.

26

of ro -p re lP

Jo

ur na

Figure 3. (A) Confocal microscopy imaging of a matrix of 4x3 barriers (height 5 µm) created in IP-L 780. The first row are cylinders hindering the in-plane water displacement; the remaining structures have four small and one large channels to permit lateral water flow. The HAuCL4 doped solution is drop-casted and a single spot of GNPs is printed at the base of each structure, at laser power 25 mW, by delivering a single exposure (B) or 100 exposures with WTs << w (C), < w (D), ~w (E) and > w (F).

27

of ro -p re lP ur na

Jo

Figure 4. (A)-(E) Optical images in reflection mode, of thermographic paper heated on a hot plate at controlled temperature. The paper darkens at temperatures ranging from 60 °C to 100 °C, due to the phase transition of the dye. Above 100 °C, the dye turns in specific patterns according to the temperature. (F)-(I) Thermographic paper heated by a metal tip at controlled temperature. At 100 °C a black spot, whose size is comparable with the diameter of the tip, is surrounded by a grey area, due to the convection/diffusion of the water ejected during the exposure. Increasing the temperature, evaporation leads to an expulsion of the dye, and therefore a white area can be seen around the spot. Above 250 °C, an internal white area and a black ring surround a grey spot, whose size is comparable to the tip diameter. (J)-(K) Thermographic paper wetted by HAuCl4water solution is irradiated with a single 10-1s exposure (J), and 100, 10-3s long, exposures with WT = 10s (K), at laser power 25 mW. The yellow central spot of GNPs is visible, due to the light scattering from the clusters of gold nanoparticles. (l) Overlay between the pattern in (K) and the spot observed by SEM imaging (Figure 1F), created in the same laser condition. The

28

Jo

ur na

lP

re

-p

ro

of

thermographic paper allows a temperature mapping of the spot: aggregates of GNPs are inside the red circle (diameter ~5 µm), where T > 250 °C. Between the orange (diameter ~20 µm) and the red circles, the average temperature is T ~ 150 °C. Between the yellow (diameter ~40 µm) and the orange circles, the average temperature is T ~ 100 °C.

Figure 5. (A) Concentration distribution at various times in a sphere with initial null concentration and surface concentration C0. Numbers on curves are values of Dt/R2. (B) Time dependence of the AuCl4- ion concentration at the voxel center (D=10-11 m2s-1, Ri=0.35 µm). (C)

29

Jo

ur na

lP

re

-p

ro

of

Time dependence of the water concentration at the voxel center, assuming a dehydrated area of radius Rw=5.5 µm.

30