Subscriber access provided by University of Sunderland
Article
Photoluminescence detection of surface oxidation processes on InGaN/GaN nanowire arrays Konrad Maier, Andreas Helwig, Gerhard Muller, Jörg Schörmann, and Martin Eickhoff ACS Sens., Just Accepted Manuscript • DOI: 10.1021/acssensors.8b00417 • Publication Date (Web): 15 Oct 2018 Downloaded from http://pubs.acs.org on October 17, 2018
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 9 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
Photoluminescence detection of surface oxidation processes on InGaN/GaN nanowire arrays Konrad Maier†, Andreas Helwig†, Gerhard Müller§, Jörg Schörmann∥, Martin Eickhoff‡,* †Airbus
Group Innovations, D-81663 Munich, Germany
§Department
Germany
of Applied Sciences and Mechatronics, Munich University of Applied Sciences, D-80335 Munich,
∥Institute
of Experimental Physics I, Justus Liebig University Giessen, D-35392 Giessen, Germany
‡Institute
of Solid State Physics, University of Bremen, D-28359 Bremen, Germany
KEYWORDS: III-nitride semiconductors, nanowires, photoluminescence, chemisorption, hydrocarbons, alcohols, H2O, O2.
ABSTRACT: InGaN/GaN nanowire arrays (NWA) exhibit efficient photoluminescence (PL) in the green spectral range, which extends to temperatures well beyond 200°C. Previous work has shown that their PL is effectively quenched when oxidizing gas species such as O2, NO2, and O3 abound in the ambient air. In the present work we extend our investigations to reducing gas species, in particular to alcohols and aliphatic hydrocarbons with C1 to C3 chain lengths. We find that these species give rise to an enhancing PL response which can only be observed when the NWAs are operated at elevated temperature and in reactive synthetic air backgrounds. Hardly any response can be observed when the NWAs are operated in inert N2 backgrounds, neither at room temperature nor at elevated temperature. We attribute such enhancing PL response to the removal of quenching oxygen and the formation of enhancing water adsorbates as hydrocarbons interact with oxygen species co-adsorbed on the heated InGaN surfaces.
Due to the bandgap variability of the III-nitride system, heterojunction nanowires can be grown in which photoluminescence excitation and emission spectra are well separated. In particular, this allows for the realization of InGaN/GaN nanowire arrays (NWA) in which electronhole pairs can be photo-generated by near UV-LED light within the GaN barrier regions and green luminescence light to be emitted from the InGaN well regions. As demonstrated in previous works, the PL emission emerging from such NWAs extends to temperatures well beyond room temperature and is sensitive to the chemical environments in which the arrays are operated. Such sensitivity has been observed in gaseous, liquid and biochemical environments1–3. Overall, these favorable properties make InGaN/GaN NWAs a promising technology platform for opto-chemical transducers. Regarding gaseous environments, previous work has shown that oxidizing air constituents such as O2, NO2 and O3 lead to efficient PL quenching4. Adsorbed H2O, on the other hand, was found to play an interesting double role both as a quencher and as an enhancer of the native PL response5, depending on the measurement conditions. In all these cases, the photo-activated interaction between adsorbed molecules and the semiconductor nanostructures takes place at the non-polar lateral sidewalls of the NWAs without any intentional coatings. In
the past several years, however, an increasing body of observations has revealed that III-nitride surfaces, like silicon6 and silicon carbide7–9, also tend to form thin layers of natural oxide10–15. As a consequence, the surfacechemical interactions occurring at the non-polar InGaN/GaN sidewalls are likely to bear some resemblance to more conventional metal oxide (MOX) nanowire assemblies which have been reported to exhibit chemically sensitive photo-luminescence as well16–19. In the present work we expand on our previous research on InGaN/GaN NWAs, focusing on their hydrocarbon response. Hydrocarbons, overall, form a huge group of reducing gas species whose interactions with naturally oxidized InGaN/GaN surfaces has not yet been comprehensively characterized. Here, we confine our investigations to pure hydrocarbons and to alcohols with C1 to C3 chain lengths. Among the wide variety of functional groups that can abound on hydrocarbon cores, OH functional groups appeared to be of particular interest as alcohols, similar to water vapor, can support autoprotolysis reactions of the form20,21 2 𝑅𝑂𝐻↔𝑅𝑂 ― + 𝑅𝑂𝐻2 + (R: hydrocarbon backbones; 𝑂𝐻, 𝑂 ― , 𝑂𝐻2 + : functional groups)
ACS Paragon Plus Environment
ACS Sensors which suggests that these species might give rise to a similarly complex PL response behavior as adsorbed water vapor. In contrast to this expectation we find that all kinds of hydrocarbons do give rise to surprisingly similar PL response behavior, namely PL enhancement when InGaN/GaN NWAs are operated at elevated temperature and in chemically reactive synthetic air (SA) environments but not in chemically inert N2 backgrounds. Based on these results we propose that hydrocarbons, in general, are detectable in an indirect manner via surface oxidation processes. In particular we propose that enhancing PL responses occur due to two cooperative effects, the first being the removal of quenching oxygen adsorbates and the second the formation of enhancing water-related adsorbates which form as the hydrocarbons undergo surface oxidation with the co-adsorbed oxygen species.
Growth and properties of InGaN/GaN nanooptical probes The optochemical transducers used in this investigation are self-assembled InGaN/GaN nanowire arrays (NWAs) as depicted in Fig. 1. They were grown by plasma-assisted molecular beam epitaxy (PAMBE) on (111) silicon substrates with (000-1) orientation along the growth axis exposing non-polar m-planes as lateral sidewalls (Figs.1 a,b). The vertical and lateral extensions of these nanowires are indicated in the scanning electron microscopy (SEM) images of Figs.1c and d.
well-separated from the shorter-wavelength PL excitation light. With the chosen In/Ga ratio (0.28) green PL light is obtained (Fig.2a). Furthermore, as there are random variations in the In/Ga ratio (0.18 < In/Ga < 0.28)22 inside the InGaN well region, the latter contains quantum-dot like In-rich regions with slightly different bandgaps. As a consequence, photo-generated electron-hole pairs are efficiently separated into different regions until charge carrier densities are reached which flatten out the built-in potential profiles. In this way, premature recombination is prevented and large carrier concentrations can be built up, which allow luminescence light to be emitted at temperatures of 200°C and beyond4. Further, due to small band bending perpendicular to the growth axes1,22, both electrons and holes can diffuse to the non-polar side walls of the InGaN wells, where they are able to interact both with native surface states as well as with molecular adsorbates. wavelength [nm] 600 550 500 450 400 4K
a)
PL intensity [a.u.]
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 9
b)
100
TNW
10
1
375 K
HeCd 325 nm T = 4 .. 375 K
0.1 2.0 2.2 2.4 2.6 2.8 3.0 3.2 energy [eV]
PL maximum 10
100 T [K]
1000
Figure.2. (a) PL emission spectra of GaN/InGaN transducers measured at different temperatures; (b) Temperature dependence of the integrated peak PL intensity.
Figure 1. (a) Nanowire heterostructures on Si substrates with axial stacking of GaN and InGaN sections, (b) crystallographic directions and planes relative to the GaN/InGaN nanowire growth direction; (c) top view SEM of self-assembled nanowire ensemble grown by PAMBE; (d) cross sectional SEM of self-assembled nanowire ensemble grown by PAMBE.
Although chemical sensitivity of uniform GaN nanowires had been previously observed1, the use of vertically structured GaN/InGaN nanowires has several advantages: Firstly, due to the smaller bandgaps of the InGaN well regions, intense commercial UV LED light sources ( ~ 365 nm) can be used for the photo-generation of charge carriers. Secondly, due to the bandgap tunability of the InGaN well regions, the emission wavelengths of the photo-luminescence (PL) light can be shifted into ranges
Whereas our previous work was dealing with those PL intensity changes that are induced by the adsorption of oxidizing gases (O2, NO2, O3)4 and by water vapor (H2O)5, our concern here is on those PL intensity changes that are induced by reducing gas species, particularly alcohols and aliphatic hydrocarbons with carbon chain lengths ranging from C1 to C3.
Gas test measurements The measurement system for the characterization of the optochemical response is displayed in Fig.3a and the spectral characteristics of its optical components are summarized in Fig.3b. The light of a near-UV LED light source ( ~ 365 nm), used for luminescence excitation, is reflected onto the InGaN/GaN nanowire array placed on the ground plane and focused onto the NWA by a lens system. The same lens system captures the luminescence light (green arrow in Fig.3a) and focusses it onto the detector window of a compact photomultiplier tube (PMT) integrated into the top lid of the sensor system.
ACS Paragon Plus Environment
Appropriate filters were used to separate the UV excitation and the green luminescence light. In order to allow measurements at different transducer temperatures, the NWA samples were mounted on a ceramic heater substrate carrying a screen-printed platinum meander on its backside. This Pt heater meander simultaneously served as a heater and as an indicator for the adjusted heater temperature. These latter temperature readings are regarded as reasonable approximations to the actual temperature of the NWA samples. The total volume of the sensor chamber amounted to approximately 5 cm3. With the total gas flow rate of 500 sccm/min this allowed gas exchange times in the order of one second.
with the test gases supplied from gas cylinders or from a bubbler, the overall gas flow was maintained at 500 sccm and controlled with the help of an independent mass flow controller. In this way, any effects potentially emerging from variable air flows were ruled out.
Figure 4. Schematics of the gas test rig, featuring test gas cylinders and a vapor saturation bottle for producing high concentrations of alcohols, either diluted in SA or in N2.
Gas response of InGaN/GaN nano-optical probes General response behavior
The PL response tests towards different hydrocarbons and alcohols were carried out using a custom-designed gas test rig with a set of mass flow controllers as shown in Fig.4. Aliphatic hydrocarbons were supplied from gas cylinders containing test gas/SA mixtures with concentrations up to 1%. Lower concentrations, down to 5% of the cylinder concentration, were obtained by dilution with a second gas line of pure synthetic air. Low concentrations of alcohols in SA were supplied from gas cylinders with initial concentrations of 100 ppm and by further dilution with appropriate flows of SA. Higher concentrations of alcohols were provided by a vapor saturation bottle (bubbler) by forcing gas streams of either synthetic air or nitrogen through the liquid alcohol inside the bottle. With a saturation vapor pressure of 5.8kPa of EtOH23 maximum concentrations of 5-6% EtOH in synthetic or nitrogen could be obtained by simply passing SA or N2 through the vapor saturation bottle. Lower concentrations of EtOH were obtained by dilution with a second stream of pure background gas (SA or N2). Throughout all gas tests, either
Background SA
resp. [%]
Figure.3. (a) Experimental arrangement for measuring the PL emission spectra of InGaN/GaN transducers under variable gas atmospheres and at different NWA operation temperatures; (b) spectral characteristic of the optical components indicated in Fig.3a.
In Fig.5 two sequences of PL response patterns to admixtures of increasing concentrations of ethane, methane and ethanol to the background of synthetic air (80% N2, 20% O2) are displayed. All hydrocarbon species, and in particular ethanol, give rise to clearly observable PL enhancement effects when the NWAs are operated at elevated temperature (T = 120°C) and in backgrounds of synthetic air. In order to demonstrate consistency with our previously reported results on the oxidizing gas response4, we show on the far right-hand side the quenching response to increasing concentrations of NO2.
resp. [%]
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
flow [sccm]
Page 3 of 9
Ethane Methane
a)
4 2
Ethanol NO2
0 -2
T=30°C
Popt=70 mW
T=120°C
Popt=70 mW
b)
4 2 0 -2
c)
100 50 0
0
2
4
6
8
10
12
14
16
time [h] Figure 5. PL response of an InGaN/GaN NWA towards increasing concentrations of ethane, methane, ethanol and
ACS Paragon Plus Environment
ACS Sensors
Ethanol response The results reported in Fig.5a are consistent with preliminary findings, namely that reducing gas species such as H2 and ethene do not have any sizeable influence on the PL response of InGaN/GaN NWAs when these are operated at room temperature4,24. However, the fact that small PL enhancement effects can already be observed at room temperature when the NWAs are exposed to ethanol with its additional OH functional group has not been reported before. The second important finding is that the PL enhancement is increased when the NWAs are operated at elevated temperatures. A third important observation is that oxygen backgrounds are necessary to actually observe PL enhancement effects. This latter fact is evidenced by the data presented in Fig.6.
6 4 2 0
a) Popt=70 mW
6 4 2 0
b) Popt=70 mW
6 4 2 0
c) Popt=70 mW
6 4 2 0
d)
1.5
e)
Fig.7a shows how the ethanol response scales with the ethanol concentration for different NWA temperatures. This figure, in particular, demonstrates that the concentration-dependence of the PL response follows the typical behavior of the recently published Langmuir adsorption and recombination (LAR) model4 over almost 5 orders of magnitude in ethanol concentration. Fig.7b further shows that the PL response is moderately increased as the PL excitation power is increased. Background SA
EtOH in N2 T=30°C
EtOH in SA T=30°C
a) Popt = 70 mW
T [°C] 2 30 50 80 1 120
Popt=70 mW
30
90
20 70 200
T
Eads= 0.71 eV
150
210
270
time [min]
concentration [ppm]
Figure.7. (a) Concentration- and temperature-dependence of the ethanol enhancement effect. Full lines represent fits to the LAR model. The Eads values given close to these lines represent extracted LAR adsorption energies; (b) dependence of the ethanol enhancement effect on PL excitation light intensity.
Figure.6. PL response of an InGaN/GaN NWA towards staircases of ethanol concentration steps applied in backgrounds of N2 (a, b) and SA (c, d). Panel (e) shows the timing of the ethanol and background gas flows.
1.2 1.0
Eads [eV]
In Fig.6, PL response data to staircases of ethanol concentration steps are shown, both applied in a background of inert N2 and in a more reactive background of synthetic air. Again, room-temperature and elevated temperature (T = 120°C) exposures are compared. The very small quenching responses towards EtOH, when applied in inert N2 backgrounds, show that EtOH adsorbates by themselves form inefficient recombination centers, orders of magnitude less efficient than oxidizing species such as O2, NO2 and O34. When EtOH is applied in reactive SA backgrounds, EtOH consistently results in enhancing responses, particularly at elevated temperature. The short PL quenching transients upon start-up and the PL
Popt
10-1 100 101 102 103 104 105 10-1 100 101 102 103 104 105
concentration [ppm]
0.5
TNWA = 50°C
Eads= 0.67 eV
0 EtOH in SA T=120°C
b)
Popt [mW]
Ead =0.8 s 8 eV
EtOH in N2 T=120°C
response [%]
3
1.0
0.0
enhancement overshoots upon termination of the EtOH exposure pulses, which can be observed at room temperature (Fig.6c), are indications for the involvement of water or water-related molecules which could potentially be formed as reactions between EtOH adsorbates and co-adsorbed O2 take place. Such waterrelated effects have already been discussed in our previous work5 and their possible contribution to the EtOH response is discussed in more detail below.
Ea =0 ds .77 eV
NO2 diluted in synthetic air (SA). Panel (c) shows the timing of the individual test gas flows.
conc. (%) resp. [%] resp. [%] resp. [%] resp. [%]
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 9
O2 NO2
0.8
O3 H2O
0.6
EtOH
0.4 0.2 0.0
0
100
200
300
400
500
T [K] Figure.8. LAR adsorption energy of EtOH as a function of NWA temperature. Adsorption energies for previously investigated species are shown for comparison.
ACS Paragon Plus Environment
Page 5 of 9
Fig.8 lists those values for the LAR adsorption energies Eads that have been extracted from the calibration curves in Fig.7. There the values of Eads for EtOH are also compared to those previously obtained for O2, NO2, O3 and H2O4,24. As can be seen, the LAR energies for EtOH are slightly higher than those for H2O and O2 and lower than those for the oxidizing air contaminants NO2 and O3. Interestingly, all data sets follow the same linear trends, indicating a strengthening adsorption as the NWA operation temperature is increased.
shows that the whole range of saturated hydrocarbons with C1 to C3 chain lengths (i.e. CH4, C2H6 and C3H8) does produce enhanced PL responses when the NWAs are operated at elevated temperatures and in SA background. The right-hand panel shows that this also holds true when saturated and unsaturated hydrocarbons with the same carbon chain length are used (C3H8, C3H6, C3H4). In this latter case it can also be observed that unsaturated hydrocarbons, with double C-C bonds, start to produce enhancing responses already at room temperature.
Background SA
resp. [%]
6 4 2 0
CH3-OH Popt=70 mW
resp. [%]
Background N2
6 4 2 0
C2H5-OH
6 4 2 0
C3H7-OH
Background SA: CH3-OH
30°C 120°C
resp. [%]
Similar data as for ethanol were also obtained for other alcohols. In particular these confirmed that clearly visible PL enhancement effects can only be observed at elevated NWA temperatures and in SA backgrounds. The importance of temperature and SA backgrounds for PL enhancements is demonstrated in the summary plot of Fig.9.
2
2
resp. [%]
Other alcohols
resp. [%]
3
resp. [%]
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
Increasing chain length:
Methane (CH4)
Ethane (C2H6)
Propene (C3H6)
Propane (C3H8)
Propyne (C3H4)
100
100
1 0 3 2 1 0 1000
concentration [ppm]
C2H5-OH
Propane (C3H8)
30°C P =70 mW opt 120°C
1 0 3
Increasing bond order:
1000
concentration [ppm]
Figure 10. Response of InGaN/GaN NWA to aliphatic hydrocarbons operated in SA backgrounds: (left) scaling of PL enhancement effect with hydrocarbon chain length; (right) scaling with bond order, i.e. single and double C-C bonds.
C3H7-OH
Discussion 1000
10000
concentration [ppm]
1000
10000
concentration [ppm]
Figure.9. Impact of background gas and NWA temperature on the PL enhancement effect of alcohols with C1 to C3 chain lengths.
While all alcohols give rise to very small quenching effects in N2 backgrounds, both at room temperature and at elevated temperatures, large enhancements are observed when the alcohols are diluted in reactive SA backgrounds and when the NWAs are operated at 120°C. Considering the room temperature response data in SA background, it is seen that methanol is more reluctant in producing enhancing PL responses than ethanol and propanol. This latter effect can possibly be attributed to the lower C-H bond strength in longer-chain alcohols25.
Aliphatic hydrocarbons Similar PL enhancement effects as with alcohols were observed with pure, aliphatic hydrocarbons, not carrying any functional groups. Fig.10 summarizes PL response data for such hydrocarbons. The left-hand panel of this figure
Overall, the above results demonstrate that hydrocarbons give rise to a PL response provided the InGaN/GaN NWAs are operated at elevated temperatures and in reactive backgrounds of synthetic air. These observations suggest that hydrocarbon species are detected in an indirect manner via reaction products that are formed via the interaction of hydrocarbon adsorbates with co-adsorbed oxygen species. Such interactions constitute oxidation processes, which - in multi-step processes - are able to transform the initial hydrocarbons into a species-dependent number of H2O and CO2 molecules. Surface combustion is the firmly established process underlying the detection of combustible gases on heated metal oxide (MOX) surfaces. This process of gas response has initially been proposed in reference26 and subsequently been confirmed by different authors27–29. The dominant end products of surface combustion, i.e. H2O and CO2 have also been directly identified using methods of in-operando spectroscopy30. Depending on the kind of analytes and MOX materials, surface combustion temperatures typically range between 200°C to 500°c.
ACS Paragon Plus Environment
Turning to the first question of similarity, we note that in the past several years an increasing body of observations has emerged that points to the fact that III-nitride surfaces, like silicon6 and silicon carbide7-9, also tend to form thin layers of natural oxide after exposure to air ambients10–15. Another indication for the similarity of oxide and oxidized III-nitride surfaces is the similarity of the chemically sensitive photo-luminescence that has been observed on MOX nanowire assemblies16–19 and on InGaN/GaN optochemical transducers. A first and obvious answer to the second question is that any kind of surface oxidation process that might take place on III-nitride surfaces will remove quenching oxygen adsorbates and thus result in an enhanced PL. The level of PL response that will ultimately occur depends on the kinds of new adsorbates that are formed in the course of the oxidation processes. In case these oxidation processes proceed up to the final end products of surface combustion, i.e. to H2O and CO2, the ultimate PL response will depend on the quenching or enhancing properties of both end products. Gas sensing tests with CO2 consistently yielded zero response, which means that CO2 either immediately desorbs or that it exhibits extremely small electron and hole trapping cross sections which render them inefficient as surface recombination centers. Further, as enhancing PL responses could not be observed, this nil result also rules out the possibility that CO2 adsorbates are able to passivate native surface recombination centers on the InGaN/GaN surfaces. Much more interesting than CO2 is H2O. As demonstrated in Fig.11, H2O consistently produces enhancing PL responses when applied to heated InGaN/GaN surfaces. As shown there, such enhancing responses can be observed not only in reactive backgrounds of synthetic air but also in inert N2. Unlike the alcoholic species discussed above, H2O adsorbates obviously have a PL enhancing capability of their own. In a previous publication we have presented evidence that this PL enhancing property derives from the fact that adsorbed water molecules can become photo-electrochemically dissociated at heated and UV-illuminated III-nitride surfaces and that the emerging positive and negative dissociation products are able to bind to the photoactivated Lewis acid-base pairs which abound at such surfaces5. Overall, the above arguments reveal that enhancing PL responses can derive from two effects: (i) the removal of quenching O2 adsorbates, and (ii) the adsorption and dissociation of H2O molecules formed in the surface oxidation processes.
resp. [%]
Invoking a similar detection mechanism to be operative on heated InGaN/GaN surfaces, two key questions arise: the first concerns the similarity of InGaN/GaN to MOX surfaces, while the second relates to the kinds of surface oxidation processes that do take place at moderately heated InGaN surfaces and how these manage to produce enhancing PL responses.
6 4 2 0
a)
6 4 2 0
b) Popt=70 mW
r.h. [%]
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 9
resp. [%]
ACS Sensors
80 60 40 20 0
T=120°C
H2O in N2
T=120°C
H2O in SA
c)
1
3
5
7
9
11
time [min] Figure 11: PL response of an InGaN/GaN NWA towards a staircase of H2O concentration steps as applied in N2 (a) and in SA (b) backgrounds. NWA temperature and excitation light conditions are indicated in the insets. Panel (c) shows the timing of the H2O vapor flow.
This hypothesis can be further tested by considering the numbers of O2 adsorbates that are removed and the numbers of H2O molecules that are formed when the different hydrocarbon species become completely converted into the universal end products of surface combustion. In Fig.12 these numbers are compared to the PL responses that were observed upon exposure to 2000ppm of each of these species. The comparison shows that the PL response observed does scale with the numbers of oxygen adsorbates removed and H2O molecules formed when the groups of singly bonded aliphatic hydrocarbons and alcohols are considered. Comparing both groups, larger responses are observed in the group of alcohols than in the group of single-bonded hydrocarbons when the same numbers of reaction products are expected to be formed. As very similar responses would be expected in case both groups would become completely converted to H2O and CO2, full combustion at moderately heated and UV-illuminated III-nitride surfaces appears to be unlikely. Another observation that points into the same direction is that higher PL-responses are observed in the groups of double-bonded as compared to single bonded hydrocarbons. We therefore conclude that at moderately heated and UV-illuminated III-nitride surfaces surface oxidation reactions can be observed that generate speciesspecific reaction products with overall surface state passivating properties but that these processes do not proceed up to the universal end products of combustion, i.e. H2O and CO2. This latter conclusion is also supported by the observations reported in Fig.8. There, we have shown that the EtOH-related reaction products do not bind with the LAR adsorption energies characteristic of H2O molecules but with somewhat higher energies.
ACS Paragon Plus Environment
5 4
a)
3 2 1
C H 4 C 2H 6 C 3H 8 C 3H 8 C 3H 6 C 3H 4 C H 3O H C 2H 5O H C 3H 7O H
0
10 8
b)
6 4
excitation light to the NWAs and luminescence light back to the light detector, InGaN/GaN nano-optical probes can also be operated in environments heavily affected by electromagnetic interference; (ii) concerning the limited size of the luminescence response, a significant new observation is that selected area growth of GaN nanowires allows the individual nanowire probes to be grown in the form of regular arrays with wire-to-wire distances in the order of hundreds of nanometers up to several micrometers. Such sparse arrays not only allow better media access but also for a more efficient light coupling by exploiting photonic crystal effects32.
AUTHOR INFORMATION
2
C H 4 C 2H 6 C 3H 8 C 3H 8 C 3H 6 C 3H 4 C H 3O H C 2H 5O H C 3H 7O H
O2 consumed + H2O formed
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 PL response [%]
Page 7 of 9
Figure 12: (a) enhancing PL response of InGaN/GaN NWAs when exposed towards 2000ppm of different hydrocarbons in a background of synthetic air; (b) number of PL-quenching O2 adsorbates consumed plus numbers of PL-enhancing H2O adsorbates formed assuming complete combustion of the individual hydrocarbon species.
Looking towards future perspectives of InGaN/GaN NWAs as gas sensing materials there are two groups of relevant comments. The first relates to basic research and the second to future applications. Concerning basic research, several activities are underway to fully exploit the potential of chemically sensitive luminescent probes. These include: (i) improvements in materials quality and refinements in the optical readout periphery to allow measurements at temperatures up and beyond the 200°C range where complete surface combustion processes are likely to take place; (ii) controlled oxidation of InGaN well surfaces and/0r coating with very thin oxide films in the sense of coreshell nanowires to attain information on adsorption processes at other types of gas sensitive semiconductor materials; (iii) combination of photo-luminescence probing with diffuse reflectance infrared Fourier transform (DRIFT) spectroscopy30 to relate luminescence changes to changes in the surface chemical bonding.
Corresponding Author *
[email protected] Author Contributions The manuscript was written through contributions of all authors. / All authors have given approval to the final version of the manuscript.
Funding Sources Authors from the Justus-Liebig-Universität acknowledge financial support within the LOEWE program of excellence of the Federal State of Hessen (project initiative STORE-E).
REFERENCES (1)
(2)
(3)
(4)
(5)
Concerning applications, two comments are in order: (i) a key motivation for investigating nano-optical chemical probes was the possibility of media separation. This subject has been addressed in a previous publication31. There, it was shown that by growing InGaN/GaN NWAs on transparent sapphire substrates the electrical readout periphery could be completely removed from the spot of sensing where the sensitive materials are in contact with the medium to be sensed. By employing light fibers for carrying UV
(6) (7) (8) (9)
Teubert, J.; Becker, P.; Furtmayr, F.; Eickhoff, M. GaN Nanodiscs Embedded in Nanowires as Optochemical Transducers. Nanotechnology 2011, 22 (27), 275505 DOI: 10.1088/0957-4484/22/27/275505. Teubert, J.; Paul, S.; Helwig, A.; Müller, G.; Eickhoff, M. Group III-Nitride Chemical Nanosensors with Optical Readout. In Gas Sensing Fundamentals; Kohl, C.-D., Wagner, T., Eds.; Springer Series on Chemical Sensors and Biosensors; Springer Berlin Heidelberg, 2014; Vol. 15, pp 311–338 DOI: 10.1007/5346_2014_58. Paul, S.; Maier, K.; Das, A.; Furtmayr, F.; Helwig, A.; Teubert, J.; Monroy, E.; Müller, G.; Eickhoff, M. III-Nitride Nanostructures for Optical Gas Detection and PH Sensing. In SPIE Defense, Security, and Sensing; George, T., Islam, M. S., Dutta, A. K., Eds.; International Society for Optics and Photonics, 2013; p 87250K DOI: 10.1117/12.2015221. Maier, K.; Helwig, A.; Müller, G.; Becker, P.; Hille, P.; Schörmann, J.; Teubert, J.; Eickhoff, M. Detection of Oxidising Gases Using an Optochemical Sensor System Based on GaN/InGaN Nanowires. Sensors Actuators B Chem. 2014, 197, 87–94 DOI: 10.1016/j.snb.2014.02.002. Maier, K.; Helwig, A.; Müller, G.; Hille, P.; Teubert, J.; Eickhoff, M. Photoluminescence Probing of Complex H2O Adsorption on InGaN/GaN Nanowires. Nano Lett. 2017, 17 (2), 615–621 DOI: 10.1021/acs.nanolett.6b03299. Sze, S. M.; Lee, M.-K. Semiconductor Devices, Physics and Technology, 3rd ed.; John Wiley & Sons Inc: New York, 2012. Roy, J.; Chandra, S.; Das, S.; Maitra, S. Oxidation Behaviour of Silicon Carbide-a Review. Rev. Adv. Mater. Sci 2014, 38 (1), 29–39. Schalwig, J.; Kreisl, P.; Ahlers, S.; Muller, G. Response Mechanism of SiC-Based MOS Field-Effect Gas Sensors. IEEE Sens. J. 2002, 2 (5), 394–402 DOI: 10.1109/JSEN.2002.806214. Zangooie, S.; Arwin, H.; Lundström, I.; Lloyd Spetz, A. Ozone Treament of SiC for Improved Performance of Gas Sensitive Schottky Diodes. Mater. Sci. Forum 2000, 338–342
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
(10)
(11)
(12) (13)
(14) (15)
(16)
(17)
(18)
(19)
(20)
(21)
(342), 1085–1088 DOI: 10.4028/www.scientific.net/MSF.338342.1085. Schalwig, J.; Müller, G.; Karrer, U.; Eickhoff, M.; Ambacher, O.; Stutzmann, M.; Görgens, L.; Dollinger, G. Hydrogen Response Mechanism of Pt–GaN Schottky Diodes. Appl. Phys. Lett. 2002, 80 (7), 1222–1224 DOI: 10.1063/1.1450044. Weidemann, O.; Hermann, M.; Steinhoff, G.; Wingbrant, H.; Lloyd Spetz, A.; Stutzmann, M.; Eickhoff, M. Influence of Surface Oxides on Hydrogen-Sensitive Pd:GaN Schottky Diodes. Appl. Phys. Lett. 2003, 83 (4), 773 DOI: 10.1063/1.1593794. Shalish, I.; Shapira, Y.; Burstein, L.; Salzman, J. Surface States and Surface Oxide in GaN Layers. J. Appl. Phys. 2001, 89 (1), 390–395 DOI: 10.1063/1.1330553. Garcia, M. A.; Wolter, S. D.; Kim, T.-H.; Choi, S.; Baier, J.; Brown, A.; Losurdo, M.; Bruno, G. Surface Oxide Relationships to Band Bending in GaN. Appl. Phys. Lett. 2006, 88 (1), 013506 DOI: 10.1063/1.2158701. Winnerl, A.; Garrido, J. A.; Stutzmann, M. GaN Surface States Investigated by Electrochemical Studies. Appl. Phys. Lett. 2017, 110 (10), 101602 DOI: 10.1063/1.4977947. Wang, L.; Bu, Y.; Li, L.; Ao, J.-P. Effect of Thermal Oxidation Treatment on PH Sensitivity of AlGaN/GaN Heterostructure Ion-Sensitive Field-Effect Transistors. Appl. Surf. Sci. 2017, 411, 144–148 DOI: 10.1016/j.apsusc.2017.03.167. Faglia, G.; Baratto, C.; Sberveglieri, G.; Zha, M.; Zappettini, A. Adsorption Effects of NO2 at Ppm Level on Visible Photoluminescence Response of SnO2 Nanobelts. Appl. Phys. Lett. 2005, 86 (1), 011923 DOI: 10.1063/1.1849832. Lettieri, S.; Santamaria Amato, L.; Maddalena, P.; Comini, E.; Baratto, C.; Todros, S. Recombination Dynamics of Deep Defect States in Zinc Oxide Nanowires. Nanotechnology 2009, 20 (17), 175706 DOI: 10.1088/0957-4484/20/17/175706. Valerini, D.; Cretì, A.; Caricato, A. P.; Lomascolo, M.; Rella, R.; Martino, M. Optical Gas Sensing through Nanostructured ZnO Films with Different Morphologies. Sensors Actuators B Chem. 2010, 145 (1), 167–173 DOI: 10.1016/j.snb.2009.11.064. Pallotti, D. K.; Amoruso, S.; Orabona, E.; Maddalena, P.; Lettieri, S. Biparametric Optical Sensing of Oxygen by Titanium Dioxide. Sensors Actuators B Chem. 2015, 221, 515– 520 DOI: 10.1016/j.snb.2015.06.116. Fonrodona, G.; Ràfols, C.; Bosch, E.; Rosés, M. Autoprotolysis in Aqueous Organic Solvent Mixtures. Water/Alcohol Binary Systems. Anal. Chim. Acta 1996, 335 (3), 291–302 DOI: 10.1016/S0003-2670(96)00329-7. Mollin, J.; Pavelek, Z.; Schneiderová, A.; Vičar, J.; Šimánek,
(22)
(23) (24)
(25)
(26)
(27) (28) (29) (30)
(31)
(32)
Page 8 of 9 V.; Lasovský, J. Autoprotolysis Constants and Activity Ratios of the Lyate Ions in Water-Alcohol Mixtures. Collect. Czechoslov. Chem. Commun. 1983, 48 (8), 2156–2164 DOI: 10.1135/cccc19832156. Kehagias, T.; Dimitrakopulos, G. P.; Becker, P.; Kioseoglou, J.; Furtmayr, F.; Koukoula, T.; Häusler, I.; Chernikov, A.; Chatterjee, S.; Karakostas, T.; et al. Nanostructure and Strain in InGaN/GaN Superlattices Grown in GaN Nanowires. Nanotechnology 2013, 24 (43), 435702 DOI: 10.1088/09574484/24/43/435702. Kretschmer, C. B.; Wiebe, R. Liquid-Vapor Equilibrium of Ethanol--Toluene Solutions. J. Am. Chem. Soc. 1949, 71 (5), 1793–1797 DOI: 10.1021/ja01173a076. Maier, K.; Helwig, A.; Müller, G.; Hille, P.; Teubert, J.; Eickhoff, M. Competitive Adsorption of Air Constituents as Observed on InGaN/GaN Nano-Optical Probes. Sensors Actuators B Chem. 2017, 250, 91–99 DOI: 10.1016/j.snb.2017.04.098. Pauling, L. The Nature of the Chemical Bond and the Structure of Molecules and Crystals: An Introduction to Modern Structural Chemistry; Cornell University Press: New York, 1960. Windischmann, H.; Mark, P. A Model for the Operation of a Thin-Film SnO2 Conductance-Modulation Carbon Monoxide Sensor. J. Electrochem. Soc. 1979, 126 (4), 627 DOI: 10.1149/1.2129098. Williams, D. E. Conduction and Gas Response of Semiconductor Gas Sensors. In Solid State Gas Sensors; Moseley, P. T., Ed.; A. Hilger: Bristol, 1987; pp 71–123. Ihokura, K.; Watson, J. The Stannic Oxide Gas Sensor Principles and Applications; CRC Press: Boca Raton, 1994. Sze, S. M. Semiconductor Sensors; Wiley, 1994. Degler, D.; Barz, N.; Dettinger, U.; Peisert, H.; Chassé, T.; Weimar, U.; Barsan, N. Extending the Toolbox for Gas Sensor Research: Operando UV/Vis Diffuse Reflectance Spectroscopy on SnO2-Based Gas Sensors. Sensors Actuators B Chem. 2016, 224, 256–259 DOI: 10.1016/J.SNB.2015.10.040. Paul, S.; Helwig, A.; Müller, G.; Furtmayr, F.; Teubert, J.; Eickhoff, M.; Sumit, P. Opto-Chemical Sensor System for the Detection of H2 and Hydrocarbons Based on InGaN/GaN Nanowires. Sensors Actuators B Chem. 2012, 173 (0), 120–126 DOI: 10.1016/j.snb.2012.06.022. Winnerl, J.; Hudeczek, R.; Stutzmann, M. Optical Design of GaN Nanowire Arrays for Photocatalytic Applications. J. Appl. Phys. 2018, 123 (20), 203104 DOI: 10.1063/1.5028476.
ACS Paragon Plus Environment
Page 9 of 9 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
FOR TOC ONLY:
ACS Paragon Plus Environment
9
