等式中标度律的有效性。 (1) 已首先考虑文献(14 种配置)和新模拟(9 种配置)的结果进行测试。特别是,从文献中获取的配置是直径 d 的水合二氧化硅纳米孔 P =8.13 或 11.04 nm(参见 [24] 中的补充表 S1);直径d的唯一磁铁矿纳米颗粒 p =1.27 或 1.97 nm 分别浸入 6 或 7 nm 边长的立方水箱中(参见 [24] 中的补充表 S4);单独的 1AKI 或 1UBQ 蛋白分别浸入边长为 7.03 或 6.32 nm 的立方水箱中(参见 [24] 中的补充表 S10);直径d的水合二氧化硅纳米孔 P =8.13 nm 填充有 2、4、8 或 16 个直径 d 的磁铁矿纳米颗粒 p =1.97 nm(参见 [24] 中的补充表 S2)或 16 个直径 d 的磁铁矿纳米颗粒 p =1.27 nm(见[24]中的补充表S3);和直径 d 的水合二氧化硅纳米孔 P =11.04 nm 填充有 36 或 66 个直径 d 的磁铁矿纳米颗粒 p =1.27 nm 或 20 个直径 d 的磁铁矿纳米颗粒 p =1.97 nm(见[24]中的补充表S3)。此外,新的模拟装置是直径 d 的水合二氧化硅纳米孔 P =8.13 nm 填充了一个 1UBQ 蛋白和一个直径 d 的水合二氧化硅纳米孔 P =11.04 nm 填充 2、3 或 9 个 1AKI 蛋白,或 2、7、9 或 12 个 1UBQ 蛋白。
图 2 显示了先前列出的不同系统的水自扩散系数的结垢行为,即本体水 (DB =2.60×10
−9
m
2
/s)、水合二氧化硅纳米孔、溶剂化蛋白质和磁铁矿纳米粒子,以及填充有蛋白质或纳米粒子的水合二氧化硅纳米孔。正如预期的那样,二氧化硅纳米孔内的水显示出减少的自扩散,与由缩放参数 θ 表示的纳米限制程度的增加相一致
*
.悬浮分子(纳米粒子和蛋白质)对水的自扩散系数显示出类似的影响。在图 2 中,实线对应于方程。 (1) 与 DC /DB ≈0,这代表了假设纳米受限水分子没有流动性因此无法扩散的极限情况。相反,虚线对应于具有 D 更现实值的相同方程 C =0.39×10
−9
m
2
/s,如在 [24] 中报告的模拟中观察到的:该模型能够准确地恢复模拟结果 (R
2
=0.93),因此证实了方程的良好预测能力。 (1) 也适用于新的模拟配置。
<图片>
Implications of the Hypothesis
Inspired by the regular nanoporous structure of diatom algae frustules, in this work, we have presented a new concept for measuring the concentration of nanoparticles or biomolecules dispersed in water. The regular structure of the algae frustules can be artificially reproduced by nanoporous silica tablets, whose pore size, thickness, and shape should be precisely tuned to optimize the selective uptake of particles. The proposed nano-metering method relies on the effect of those nanoparticles or biomolecules on the self-diffusion coefficient of water nanoconfined within the tablet’s pores, and consists in the following steps:
1.
Synthesize porous tablet with a controlled size distribution of nanopores.
2.
Let the nanopores of the tablet fill with the solution containing the particles to be detected via capillary imbibition and particle diffusion, achieving equilibrium conditions between the nanopores and the surrounding solution.
3.
Remove the tablet from the solution and measure the self-diffusion coefficient of water in the hydrated nanopores filled with the particles, e.g., by QENS or D-MRI techniques.
4.
Correlate the measured self-diffusion coefficient of water with the particle concentration by means of Eqs. 1 to (5). The solvent accessible surface of nanopore and particles (SAS ) and their mean characteristic length of nanoconfinement (\(\bar {\delta }\)) should be computed from molecular dynamics or taken from available databases.
Molecular dynamic simulations and evidence from the literature have been employed to assess the feasibility of the proposed nano-metering protocol. Hydrated nanopores filled with different concentrations of iron-oxide nanoparticles or proteins have been analyzed, finding agreement between the computed and predicted self-diffusion coefficient of nanoconfined water, thus allowing to estimate the particle concentration. A preliminary analysis of the mechanisms involved in the nanopores filling has been also carried out. Because of the different time scales, two different phenomena have been considered separately:the imbibition of a dry tablet by pure water, driven by capillarity, and the particle diffusion through the hydrated pores, driven by concentration gradient. Results show that the leading characteristic time in the filling process is the time required for particles to diffuse into the hydrated pores; however, the estimated filling time does not exceed 1 h even in case of the thickest tablets considered (1 mm), therefore not compromising the practical feasibility of the nano-metering protocol.
Although the proposed nano-metering method has shown promising results from a numerical point of view, the actual experimental implementation may have to face some additional issues. First, the interaction between the pore surface and particles could be non-negligible and thus alter the filling process (e.g., pores clogging). This effect could generate a bias between the actual concentration of the particles in the bulk solution and the one measured within the pores. Such an issue could be solved by an accurate selection of the surface properties of the pores, which should not interact with the particles to be detected. Second, the current experimental techniques could have difficulty to measure the water diffusivity with a single-nanopore resolution. This issue could be mitigated by measuring the average self-diffusion coefficient over hundreds or thousands of nanopores, which could also provide a better statistical sampling in case of inhomogeneous particle filling throughout the tablet. Third, the uncertainty of the nano-metering protocol should be assessed by experiments. The configurations studied by molecular dynamics have revealed prediction errors up to ± 50% :this error range could be eventually reduced by considering larger statistical samples, both in terms of time (multiple measures) and space (averages over hundreds or thousands of pores). Fourth, the optimal diameter of the nanopores should be determined on the basis of the expected size and concentration of the particles to be detected. On the one hand, the pore size should be chosen to avoid low θ
*
(e.g., θ
*
should be> 0.2), since this could lead to negligible variations of the self-diffusivity of water that could be eventually below the resolution of the QENS or D-MRI techniques; on the other hand, high levels of water nanoconfinement should be avoided as well (e.g., θ
*
should be <0.8), to limit the risk of pore clogging or particle aggregation/segregation and thus biased concentration results.
In conclusion, further research is needed to validate experimentally the original nano-metering protocol discussed in this work. However, the presented numerical results prove the potential of the idea, which may pave the way to a completely new class of detection processes of emerging nanopollutants in water or biomolecules. In perspective, the microscopic size of the metering devices, e.g., nanoporous silica tablets, may allow automation of the nano-metering process through lab-on-a-chip devices.