Porous Nanophotonic Optomechanical Beams for Enhanced Mass

Subscriber access provided by UNIV AUTONOMA DE COAHUILA UADEC

Article

Porous Nanophotonic Optomechanical Beams for Enhanced Mass Adsorption Anandram Venkatasubramanian, Vincent T. K. Sauer, Jocelyn N WestwoodBachman, Kai Cui, Mike Xia, David S. Wishart, and Wayne K Hiebert ACS Sens., Just Accepted Manuscript • DOI: 10.1021/acssensors.8b01366 • Publication Date (Web): 03 Apr 2019 Downloaded from http://pubs.acs.org on April 17, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

Porous Nanophotonic Optomechanical Beams for Enhanced Mass Adsorption Anandram Venkatasubramanian1, 2, 3, Vincent T. K. Sauer1, 2, 3, Jocelyn N. Westwood-Bachman 1, 2, Kai Cui1, Mike Xia1, David S. Wishart1, 3, 4 and Wayne K. Hiebert*, 1, 2 1

Nanotechnology Research Centre, National Research Council of Canada, Edmonton, AB, Canada T6G 2M9 2Department

of Physics, University of Alberta, Edmonton, Alberta, Canada T6G 2E1

3Department

of Biological Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2R3

4Department

of Computing Science, University of Alberta, Edmonton, Alberta, Canada T6G 2E8

*To whom correspondence should be addressed. E-mail - [email protected]

Abstract We have developed a porous silicon nanocantilever for a nano-optomechanical system (NOMS) with a universal sensing surface for enhanced sensitivity. Using electron beam lithography, we selectively applied a V2O5/HF stain etch to the mechanical elements while protecting the silicon-oninsulator photonic ring resonators. This simple, rapid, and electrodeless approach generates tunable device porosity simultaneously with the mechanical release step. By controlling the porous etchant concentration and etch time, the porous etch depth, resonant frequency and the adsorption surface area could be precisely manipulated. Using this control, cantilever sensors ranging from non-porous to fully porous were fabricated and tested as gas-phase mass sensors of volatile organic compounds coming from a gas chromatography stream. The fully porous cantilever produced a dramatic 10-fold increase in sensing signal and a 6-fold improvement in limit of detection (LOD) compared to an otherwise identical non-porous cantilever. This signal improvement could be separated into mass responsivity increase and adsorption increase components. Allan deviation measurements indicate that a further 4-fold improvement in LOD could be expected upon speeding up characteristic peak response time from 1 s to 50 ms. These results show promise for performance enhancement in nanomechanical sensors for applications in gas sensing, gas chromatography, and mass spectrometry. Keywords: NOMS, porosity, gas chromatography, gas sensing, mass sensing

ACS Paragon Plus Environment

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 18

The application of nanomechanical beams for mass sensing has been around for more than a decade

1-9.

However, rapid improvements in nanomechanical mass sensitivity and a growing

awareness of its potential to compete with mass spectrometry has brought increased attention to this area of nanotechnology

1,5,8-10.

These improvements have been due to recent trends in

nanomechanical device miniaturization which reduce the device dimension and hence the mass of the nanomechanical beam. This favourably affects the mass sensitivity

11.

However, lowering the

device dimensions reduces the surface area for analyte adsorption which, in turn, decreases the adsorption capacity due to a reduction in the number of adsorption sites. There are a few ways to mitigate this reduction in adsorption capacity. One of them is to coat the nanomechanical beam with high surface area materials like metal organic frameworks, or polymers

12,13.

However, these may

also alter the chemical affinity of the surface. Another approach is to increase the surface area of the sensor by introducing porosity to the nanomechanical beam surface without affecting the chemical affinity of the surface. This work focuses on the latter approach. Previous work on porous nanomechanical beams has achieved an increase in the adsorption of volatile organic compounds (VOC) by a factor of 2X-3X

14-20

depending on the beam’s surface

chemistry. The use of porous nanomechanical devices is not only limited to mass sensing but has also been used for spectroscopic applications such as infrared (IR) calorimetry

20.

Across the

spectrum of applications for porous nanomechanical beams, changes in beam porosity have been principally realized by anodic aluminium oxide (AAO) or other electrode based porosity inducing techniques. However these methods are not applicable for structures devoid of electrodes (as in our case) wherein the mass sensor is based on a nanophotonics platform made from silicon-on-insulator. Stain etching 21 has an advantage over electrode etching methods as it does not require electrodes. In this paper, using lithographically controlled stain etching, we have produced locally porous nanomechanical structures of various porosities. We have compared the VOC adsorption measurements between non-porous devices, to these porous nanostructures. We found an order of magnitude increase in VOC sensing signal of fully porous beams compared to non-porous beams for the same concentration. By determining the effective densities, we were able to resolve this signal improvement into component effects of increased adsorption and increased mass responsivity. This is in contrast to the usual literature approach

14,16,17,20

of simply equating signal

increase with mass sensitivity improvement. The signal increase translates into a limit of detection (LOD) improvement by 6X. This improvement is consistent with a comparison of measured sensor noise floor between porous and non-porous devices. The noise measurement also suggests a further factor of 4 improvement in LOD would result for faster eluant peaks due to better noise performance of the porous device at shorter sampling times.


Page 3 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

Experimental Setup The nanophotonic – gas chromatography (GC) measurement setup used for measuring the gas uptake is similar to the one described in our previous work 8. For details on the operating principle, experimental setup of the nanophotonic optomechanical sensing system, please refer to our earlier paper 8. The device chip is patterned using an electron-beam lithography (EBL) resist (PMMA) to open windows that expose only the nanomechanical beam but protect the ring resonator (as shown in Figure 1 a,b). While opening windows in the resist, the clamped edge of the cantilever beam is protected and about 75% of the length from the free end of the cantilever beam is exposed to a stain etch solution. Following the patterning, the devices were exposed to stain etch solutions. While there are many stain etch chemistries 21, in this work the etch chemistry based on V2O5/HF was chosen due to its simple setup and rapid etch rates 22-25. A single time window was chosen for the stain etch to coincide with the amount of etch time needed to mechanically release the devices due to HF etching of the oxide support. In this manner, the technique allows creating the porosity and performing the mechanical release in a single step. The patterned devices were exposed to different stain etch chemistries by varying the mass of the oxidant (V2O5) in the solution in order to change the rate of stain etching in the given time window (for details about the etch procedure, refer to the supplementary materials section). Molarities of V2O5 ranged from 0 (no stain etching) to 314 mM (maximum stain etching rate). From the TEM plot shown in Figure 1c (see supplementary materials for sample preparation details), it is clear that as the molarity of the oxidant increases, the depth of the created pore region increases. Note that the finite thickness (130 nm) of the cantilever devices puts a limit on the maximum depth of the pore region; increasing the oxidant concentration beyond 245 mM for the process conditions and the nanomechanical beam dimensions caused complete etching of the silicon beam (and a loss of mechanical resonance) as is evident in Figure S1 a,b (supplementary material). Setting a maximum oxidant concentration at 228 mM of V2O5 (hereafter device 4, fully porous) gave us a device with maximum porosity and an etch depth of about 80 nm for the given etch time. Given the patterning and the access to etch zone, the porous layer is believed to extend throughout the device thickness of the exposed region. The other concentrations for cantilever tests were 174 mM V2O5 (hereafter device 3, moderately porous), 77 mM V2O5 (hereafter device 2, slightly porous) and a non-porous device (device 1, 0 mM V2O5). Following the exposure to stain etch chemistry, the EBL resist was stripped by simply immersing the resist-coated chip in acetone for 20 min followed by gently dip washing in 100% isopropanol and water in separate beakers. Without exposing it to air, the device was placed in a Buffered Oxide Etch (BOE) etch solution for 2.25 minutes to release the nonporous mechanical devices on the same chip for comparative measurements with the porous (etched) devices. The chip was then washed in piranha solution, dried using critical point dryer and ACS Paragon Plus Environment

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 18

then mounted on a piezo actuator housed in a pressure controlled vacuum chamber as described in8 . The non-porous nanomechanical device used for this experiment is a cantilever of dimensions 4 µm length, 130 nm wide and 220 nm deep. Based on the phase lock loop method (PLL) described in8 , the resonant frequency of the non-porous nanomechanical device was measured to be 10.45 MHz at atmospheric pressure and at a temperature of 298 K. Under the same conditions of pressure, temperature and drive, the resonant frequency of the porous nanomechanical devices were found to vary depending on the porosity of the nanomechanical beam with decreasing frequency for increasing porosity. These values are listed in the supplementary section in Table S1. Results and Discussion Vapour adsorption tests to demonstrate the increase in adsorption capacity were performed using five different volatile organic compound (VOC) mixtures containing benzene, toluene and xylene. The mixtures were sent through a standard gas chromatograph (GC) with output stream split to deliver eluant equally to the NOMS sensor chip and the GC flame ionization detector (FID). All samples used in the experiment were measured in mg/ml with hexane used as the solvent and their concentrations are listed in the supplementary section in Table S2. We measured the FID response from the Gas Chromatograph (Agilent 6890N network GC system), the NOMS mechanical response and the optical microring resonator response. For a detailed description of the significance of each of these responses, please refer to our earlier work8. The microring resonator probes the index of refraction of the gas and its response was used to calibrate the analyte gas-phase concentration in parts-per-million (ppm) at the chip surface following the procedure outlined in8 . The NOMS responds as a frequency shift Δf due to mass loading of analyte on the cantilever and is represented normalized to its resonant frequency f (in parts-per-million, ppm, Hz/Hz). Figure 2a shows the nanomechanical (NOMS) response to identical GC runs for the 4 different porosity devices. It is immediately clear that there was an enhancement for the porous device signals with the absolute signal increasing for increasing porosity. This is in contrast to the microring response (see Figure S2) that showed no dependence with respect to porosity, which was as expected since the rings were not stain etched. Figure 2, b-d confirmed the behaviour with detail of the NOMS peak signal vs. concentration for benzene, toluene, and xylene, respectively, for the 4 different porosities. Overall signal responsivity (the slope of frequency shift vs. concentration) improved with porosity. To further elucidate this point, we define a signal improvement ratio (SIR) for the NOMS signal, 𝑆𝐼𝑅 ≡

(∆𝑓/𝑓)𝑝𝑜𝑟𝑜𝑢𝑠 (∆𝑓/𝑓)𝑛𝑜𝑛 ― 𝑝𝑜𝑟𝑜𝑢𝑠


(1)

Page 5 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

and present the result in Figure 3a. SIR values sort neatly into 3 bins based on oxidant molarity (and therefore device porosity) for all analytes and concentrations. Improvement is just over 2X, 6X, and 10X for slightly (77mM), moderately (174 mM), and fully porous devices (228 mM), respectively. The fully porous cantilever produced an impressive 10-fold increase in VOC adsorption signal compared to its non-porous counterpart. This is significant considering previous attempts using porous NEMS structure based gas/ vapour phase sensing 14-20 achieved only a 2X-3X improvement in signal sensitivity. This kind of substantial increase should translate directly into a similar improvement in sensor performance and validates the efficacy of our sensing enhancement approach. Further analysis will confirm this enhancement while clarifying the mechanisms. The signal derives from mass loading on the sensor device and is generally described according to the following relationship for small mass changes: Δf/f = −Δm/(2Meff) . Two distinct effects therefore contribute to the signal increase for the porous devices compared to non-porous. First, increase in adsorption sites in the higher porosity structures leads to increased adsorption (an increase in the amount of analyte sticking at a given concentration) and an increase in mass loading Δm. Second, the effective mass Meff of the porous devices decreases due to the pore voids. This second effect can be cast as an increase in mass responsivity Rm of the porous sensors (one obtains more signal for the same mass load) which we define as (in units of ppm/ag) 𝑅𝑚 ≡ ―

(∆𝑓/𝑓) ∆𝑚

1

≡ 2𝑀𝑒𝑓𝑓

(2)

These two contributions can be separated using information about the device properties to extract the effective mass. In particular, the effective density (and porosity) is extracted. A least squares fitting procedure combined with COMSOL numerical modeling was undertaken (detailed in the supplementary) to determine the density of the porous cantilevers. Effective densities (porosities) of 79% (21%), 48% (52%), and 28% (72%) of the bulk density of silicon (2320 kg/m3) were found for devices 2, 3, and 4, respectively. The values produce an effective mass (= 0.245 x volumecantilever x ρeffective,cant) which is smaller than the effective mass of non-porous device by a factor of 1.27, 2.09 and 3.51 for device 2, 3 and 4, respectively. The result is that mass responsivity can be calibrated; we found that it grew with porosity starting at 8 ppm/ag for the non-porous device and ending at 28 ppm/ag for the fully porous one. The signal improvement between these two devices was a factor of 10.31 while the mass responsivity increase was only a factor of 3.51. Therefore we infer that the remaining factor of 2.89 must be accounted from an increased amount of mass Δm from increased adsorption onto the fully porous device. This discussion is summarized in Table 1. The inferred increased adsorption in the porous devices is plotted as the mass adsorption ratio (MAR) ((Δm)porous/(Δm)non-porous) in Figure 3b. There was increased adsorption on the porous devices, saturating at a ratio of about 3 for the moderately and fully porous devices. We postulate


ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 18

that these two devices may have similar surface areas. This is not unreasonable for the random pore morphology as an equilibrium could be reached where extra etching removes surface area as quickly as it adds surface area once the etch depth accesses the majority of the device volume. To confirm that the performance of porous devices is actually better than the non-porous one, we compare the four devices by their limit of detection for benzene. Benzene is chosen due to its weak adsorption affinity to minimize bias due to chemical affinity. The analyte concentration was lowered in steps until no obvious peak was evident at the expected elution time during visual inspection of the NOMS signal. This procedure is shown graphically in Figure 4a. The last concentration step with an obvious NOMS peak was considered the LOD and is summarized in Table 1 where we define the LOD improvement ratio 𝐿𝐼𝑅 ≡

𝐿𝑂𝐷𝑛𝑜𝑛 ― 𝑝𝑜𝑟𝑜𝑢𝑠 𝐿𝑂𝐷𝑝𝑜𝑟𝑜𝑢𝑠

. The LOD does improve

with porosity, by as much as a factor of 6 for the fully porous device. In fact, LOD values for toluene and xylene adsorption on device 4 (not shown) reached the remarkable levels of 60 ppb and 20 ppb, respectively. While this performance is impressive, the question remains as to why LIR values are different than SIR values with the former at slightly more than half the latter. Insight can be obtained by comparing the Allan deviation between the fully porous and non-porous devices. Allan deviation is defined as 𝜎(𝜏) ≡

( 〈(( ) 1

∆𝑓

2

𝑓

𝑛+1

―

𝑓

〉)

2

( )𝑛) ∆𝑓

, where the bar denotes averaging over the

n-th observation time interval τ and the denotes expectation value; it is a standard metric for estimating the minimum values of Δf /f that can be extracted out of noise. Figure 4b shows the measurements of Allan deviation for devices 1 and 4 (non-porous and fully porous) taken under identical conditions. For τ ~1s to 60s, the porous device has a poorer value of σ by about a factor of 2. This is a reasonable explanation for why LIR is about ½ the value of SIR because the time-scale of importance in Figure 2a is around 1 s. The long-τ behaviour in the graph is consistent with thermal drift from temperature fluctuations. Having a smaller thermal mass, the porous device is more susceptible to temperature changes. It is interesting to note that the two curves cross and the porous device performs better at short-τ. If eluent peaks could be sped up to be ~ 0.05 s (50ms) widths, the porous device would improve its LOD about another factor of 4. This improvement could generally be expected in moving from non-porous to porous devices because the dynamic range of the porous devices, defined as the power ratio of driven mechanical amplitude over noise amplitude, tended to outperform the non-porous device. The dynamic range is the most important feature in setting the noise levels for shorter sampling times26. A more detailed note on the effect of temperature fluctuations, thermal mass, dynamic range, quality factor of the nanomechanical device on the Allan deviation is given in the supplementary section. ACS Paragon Plus Environment

Page 7 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

Finally, we double check the consistency of the LOD measurements by comparing them to the Allan deviation. Table 2 shows σ and 𝛿𝑚(𝜏) ≡ 2𝑀𝑒𝑓𝑓𝜎(𝜏) along with the LOD expressed as mass (cf. Figure 4a) for the non-porous and fully porous devices. The ratios of these values between the two devices are also presented. The ratio of δm values at τ = 1 s (a ratio of about 0.5) is similar to the ratio of ΔmLOD as we expect. The 7x improvement in mass sensitivity for the porous device at short time scale is also evident. The mass LOD (ΔmLOD) is about 5 times that derived from Allan deviation, Δm ~ 5 δm, which is consistent with requiring visual identification of a clear peak out of the background. Conclusion In conclusion, using electrodeless porous etching, we have enhanced the signal sensitivity of nanomechanical mass sensors by 10-fold while preserving the universality of the NOMS sensing surface. These nanocantilevers were fabricated using a simple, controllable, and localized stain etching procedure on silicon cantilevers which protected photonic elements on the same chip. We determined that the signal enhancement combined effects of adsorption increase and mass responsivity increase. The enhancement delivered limit of detection improvement as large as a factor of 6. Noise analysis confirmed the findings and pointed to further gains available from speeding up the gas concentration characteristic time response, a feature readily achievable as well as desirable in standard and micro- gas chromatography. Further gains in specificity and adsorption uptake can be achieved by coating the porous nanomechanical device with appropriately chosen sensing layers, a path for future exploration. Given the promising results, an area of future work would be to quantitatively test the process variability to ascertain the chip to chip variability. Such results can be used for validating the performance enhancement of nanomechanical sensors for applications in gas sensing, gas chromatography, and mass spectrometry. Acknowledgement The authors would like to thank Alberta Innovates – Health Solutions (AIHS) (through the Collaborative Research Innovation Opportunities grant), Natural Sciences and Engineering Research Council of Canada and the Nanotechnology Research Centre of the National Research Council of Canada for their continued support and funding for this research. The fabrication of the devices was facilitated through CMC Microsystems, and post processing was done at the University of Alberta-NanoFab. The authors declare there are no sources of conflicting financial interest with respect to this manuscript. ACS Paragon Plus Environment

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 18

Supporting Information Supporting information section provided with this paper contains description of the stain etching procedure, tables listing the resonant frequency at different porosities, concentrations of the VOC samples, description of the procedure to estimate the effective mechanical properties using computational simulations. The supporting information is available free of charge at pubs.acs.org


Page 9 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

Figure 1 a) Schematic of the NOMS photonic sensor system coated with resist layer with the etch window allowing porous etching of the cantilever, b) Patterned electron beam resist exposing nanocantilevers while protecting ring resonators, c) Juxtaposed TEM images of porous etched silicon with various concentrations of V2O5. The chequered pattern due to porous silicon between the white borders can be clearly seen when compared to the uniform shade of non-porous silicon.


ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Benzene

Page 10 of 18

Toluene

Figure 2 a) Time traces of the fractional frequency shift (Δf*106/f) of the nanomechanical beam of different porosities (for Benzene and Toluene adsorption) filtered over a time period of 0.5 s for clearer visual presentation. The porosities are denoted by the molar concentration of the oxidant (V2O5) in solution. Increasing concentration implies increasing porosity. Data presented in (a) is for sample S4, where 1 µL was injected and a split ratio of 10:1 was maintained in the GC system. Peak fractional frequency shift response of the NOMS (in ppm, frequency shift) at different analyte concentrations for different porosities is presented in b) for Benzene, c) Toluene and d) Xylene.


Page 11 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

Figure 3 a) Signal improvement ratio (SIR) and b) Mass adsorption ratio (MAR) for nanomechanical beam for various analyte concentrations and different porosities. Only those analyte concentrations which can evoke a clear real-time response from the non-porous nanomechanical beam are represented here.


ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 18

Figure 4 a) Mass detected vs benzene gas phase concentration at different porosities. b) Comparison of Allan deviation between a non-porous (Device 1) and a fully porous mechanical beam (228mM porous, Device 4) under identical drive conditions. The Phase Lock Loop (PLL) bandwidth used was 500 Hz. The Allan deviations at two time stamps relevant to thermomechanical noise (τ = 50 ms) and gas adsorption time scale (τ = 1 s) is presented. The bumps in Allan deviation at sampling time less than < 50 ms are essentially harmonics of power line at multiples of 60 Hz.


Page 13 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

Table 1 Important gas uptake parameters for devices etched in different porous etchant concentrations. Limit of Detection (LOD) values (ppm, concentration) were measured experimentally with benzene. Effective mass of the non-porous mechanical beam (meff,nonporous) = 6.2 x 10-14 g. All the values are rounded to first decimal digit. SIR – Signal Improvement Ratio, Rm –

Responsivity, NRm – Normalized Responsitivity (Rm,porous/Rm,nonporous), IAR – Inferred Adsorption

Ratio (SIR/NRm), LIR – LOD Improvement Ratio. Device 1 - Non Porous, Device 2 - 77 mM (Slightly Porous), Device 3 - 174 mM (Moderately Porous), Device 4 - 228 mM (Fully Porous).

Device

SIR

1

-

2

Rm

LOD

NRm

IAR

8

1

1

11.4

1

2.2

10.2

1.3

1.8

9.5

1.2

3

6.3

16.8

2.1

3.0

3.2

3.5

4

10.3

28.1

3.5

2.9

1.9

6

(ppm/ag)


(ppm)

LIR

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 18

Table 2 Comparison of Mass sensitivity between devices etched in different porous etchant concentrations. Limit of Detection (LOD) values in mass were measured experimentally with benzene. All the values are rounded to first decimal digit/nearest integer. Allan deviation based mass sensitivity (δmAD,

(τ=1s),

δmAD,

(τ=0.05s)),

experimentally measured mass shift using benzene

(ΔmLOD) are expressed in zeptograms (zg). Device 1 - Non Porous, Device 4 - 228 mM (Fully Porous).

Device

στ=1s

στ=0.05s

x10-7

10-7

x

δmAD

δmAD

τ=1s

τ=0.05s

(zg)

(zg)

ΔmLOD (zg)

1

7

6

87

75

342

4

12

3

43

11

205

1.7

0.5

0.5

0.15

0.6

porous /nonporous


Page 15 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Sensors

TOC Graphic


ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 18

Reference 1

Chaste, J., Eichler, A., Moser, J., Ceballos, G., Rurali, R. & Bachtold, A. A nanomechanical mass

sensor

with

yoctogram

resolution.

Nat.

Nanotechnol.

7,

300-303,

doi:10.1038/nnano.2012.42 (2012). 2

Ekinci, K. L., Huang, X. M. H. & Roukes, M. L. Ultrasensitive nanoelectromechanical mass detection. Appl. Phys. Lett. 84, 4469-4471, doi:10.1063/.1.1755417 (2004).

3

Hanay, M. S., Kelber, S., Naik, A.K., Chi, D., Hentz, S., Bullard, E. C., Colinet, E., Duraffourg, L. & Roukes, M. L. Single-protein nanomechanical mass spectrometry in real time. Nat. Nanotechnol. 7, 602-608, doi:10.1038/nnano.2012.119 (2012).

4

Naik, A. K., Hanay, M. S., Hiebert, W. K., Feng, X. L. & Roukes, M. L. Towards singlemolecule

nanomechanical

mass

spectrometry.

Nat.

Nanotechnol.

4,

445-450,

doi:10.1038/nnano.2009.152 (2009). 5

Olcum, S., Cermak, N., Wasserman, S. C., Christine, K. S., Atsumi, H., Payer, K. R., Shen, W. J., Lee, J. C., Belcher, A. M., Bhatia, S. N. & Manalis, S. R. Weighing nanoparticles in solution at the attogram scale. Proc. Natl. Acad. Sci. U. S. A. 111, 1310-1315, doi:10.1073/pnas.1318602111 (2014).

6

Sage, E., Brenac, A., Alava, T., Morel, R., Dupre, C., Hanay, M.S., Roukes, M. L., Duraffourg, L., Masselon, C. & Hentz, S. Neutral particle mass spectrometry with nanomechanical systems. Nat. Commun. 6, doi:10.1038/ncomms7482 (2015).

7

Son, S., Kang, J. H., Oh, S., Kirschner, M. W., Mitchison, T. J. & Manalis, S. Resonant microchannel volume and mass measurements show that suspended cells swell during mitosis. J. Cell Biol. 211, 757-763, doi:10.1083/jcb.201505058 (2015).

8

Venkatasubramanian, A., Sauer, V. T. K., Roy, S. K., Xia, M. K., Wishart, D. S. & Hiebert, W. K. Nano-Optomechanical Systems for Gas Chromatography. Nano Lett. 16, 6975-6981, doi:10.1021/acs.nanolett.6b03066 (2016).

9

Yang, Y. T., Callegari, C., Feng, X. L., Ekinci, K. L. & Roukes, M. L. Zeptogram-scale nanomechanical mass sensing. Nano Lett. 6, 583-586, doi:10.1021/nl052134m (2006).

10

Li, M., Myers, E. B., Tang, H. X., Aldridge, S. J., McCaig, H. C., Whiting, J. J., Simonson, R. J., Lewis, N. S. & Roukes, M. L. Nanoelectromechanical Resonator Arrays for Ultrafast, Gas-Phase

Chromatographic

Chemical

Analysis.

Nano

Lett.

10,

3899-3903,

doi:10.1021/nl101586s (2010). 11

Sansa, M., Sage, E., Bullard, E. C., Gely, M., Alava, T., Colinet, E., Naik, A. K., Villanueva, L. G., Duraffourg, L., Roukes, M. L., Jourdan, G. & Hentz, S. Frequency fluctuations

in

silicon

nanoresonators.

Nat.

doi:10.1038/nnano.2016.19 (2016). ACS Paragon Plus Environment

Nanotechnol.

11,

552-+,

Page 17 of 18 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

12

ACS Sensors

Ellern, I., Venkatasubramanian, A., Lee, J. H., Hesketh, P. J., Stavila, V., Robinson, A. & Allendorf, M. HKUST-1 coated piezoresistive microcantilever array for volatile organic compound sensing. Micro Nano Lett. 8, 766-769, doi:10.1049/mnl.2013.0390 (2013).

13

Venkatasubramanian, A., Navaei, M., Bagnall, K. R., McCarley, K. C., Nair, S. & Hesketh, P. J. Gas Adsorption Characteristics of Metal-Organic Frameworks via Quartz Crystal Microbalance Techniques. J. Phys. Chem. C 116, 15313-15321, doi:10.1021/jp304631m (2012).

14

Boytsova, O., Klimenko, A., Lebedev, V., Lukashin, A. & Eliseev, A. Nanomechanical humidity detection through porous alumina cantilevers. Beilstein J. Nanotechnol. 6, 13321337, doi:10.3762/bjnano.6.137 (2015).

15

Harraz, F. A. Porous silicon chemical sensors and biosensors: A review. Sens. Actuator BChem. 202, 897-912, doi:10.1016/j.snb.2014.06.048 (2014).

16

Hwang, Y. H., Gao, F., Hong, A. J. & Candler, R. N. Porous Silicon Resonators for Improved

Vapor

Detection.

J.

Microelectromech.

Syst.

21,

235-242,

doi:10.1109/jmems.2011.2170819 (2012). 17

Hwang, Y. H., Kim, S. & Candler, R. N. Receptor - Coated Porous Silicon Resonators for Enhanced Sensitivity of Vapor Detection. 2011 Joint Conference of the Ieee International Frequency Control Symposium/European Frequency and Time Forum Proceedings IEEE International Frequency Control Symposium 922-925 (Ieee, 2011).

18

Lee, D., Kim, S., Jeon, S. & Thundat, T. Direct Detection and Speciation of Trace Explosives Using a Nanoporous Multifunctional Microcantilever. Anal. Chem. 86, 50775082, doi:10.1021/ac500745g (2014).

19

Lee, D., Yim, C. & Jeon, S. Nanomechanical analysis of the adsorption and desorption of water vapor on porous surfaces. RSC Adv. 5, 80925-80928, doi:10.1039/c5ra15850h (2015).

20

Lee, D., Zandieh, O., Kim, S., Jeon, S. & Thundat, T. Sensitive and selective detection of hydrocarbon/water vapor mixtures with a nanoporous silicon microcantilever. Sens. Actuator B-Chem. 206, 84-89, doi:10.1016/j.snb.2014.09.036 (2015).

21

Kolasinski, K. W. in Handbook of Porous Silicon

(ed Leigh Canham) 35-48 (Springer

International Publishing, 2014). 22

Kolasinski, K. W. & Barclay, W. B. in Pits and Pores 5: A Symposium in Honor of David Lockwood Vol. 50 ECS Transactions (eds R. Boukherroub et al.) 25-30 (Electrochemical Soc Inc, 2013).

23

Kolasinski, K. W. & Barclay, W. B. The Stoichiometry of Electroless Silicon Etching in Solutions

of

V2O5

and

HF.

Angew.

Chem.-Int.

doi:10.1002/anie.201300755 (2013). ACS Paragon Plus Environment

Edit.

52,

6731-6734,

ACS Sensors 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

24

Page 18 of 18

Kolasinski, K. W., Hartline, J. D., Kelly, B. T. & Yadlovskiy, J. Dynamics of porous silicon formation by etching in HF + V2O5 solutions. Mol. Phys. 108, 1033-1043, doi:10.1080/00268971003639258 (2010).

25

Kolasinski, K. W. & Yadlovskiy, J. in Physica Status Solidi C: Current Topics in Solid State Physics, Vol 8, No 6 Vol. 8 Physica Status Solidi C-Current Topics in Solid State Physics (eds A. Cantarero et al.) (Wiley-V C H Verlag Gmbh, 2011).

26

Roy, S. K., Sauer, V. T. K., Westwood-Bachman, J. N., Venkatasubramanian, A. & Hiebert, W. K. Improving mechanical sensor performance through larger damping. Science 360, doi:10.1126/science.aar5220 (2018).


Porous Nanophotonic Optomechanical Beams for Enhanced Mass

Recommend Documents