In this report, we describe the design, construction, and operation of a cell-phone-based Raman and emission spectral detector, which when coupled to a diffraction grating and cell-phone camera system provides means for the detection, recording, and identification of chemicals, drugs, and biological molecules, in situ by means of their Raman and fluorescence spectra. The newly constructed cell-phone spectrometer system was used to record Raman spectra from various chemicals and biological molecules including the resonance enhanced Raman spectra of carrots and bacteria. In addition, we present the quantitative analysis of alcohol–water Raman spectra, performed using our cell-phone spectrometer. The designed and constructed system was also used for constructing Raman images of the samples by utilizing a position scanning stage in conjunction with the system. This compact and portable system is well suited for in situ field applications of Raman and fluorescence spectroscopy and may also be an integrated feature of future cell-phones.

Cell-phone cameras allow one to record and store a picture within seconds. The compact cameras in modern day cell-phones are also increasingly becoming at par, even surpassing the quality and sensitivity of the traditional, stand-alone camera systems. This incredible advance in quality and low light sensitivity of cell-phone cameras is partly due to the use of back-illuminated CMOS sensors and advances in noise removal software pre-processing of the captured images. This outstanding quality and sensitivity of cameras in modern day cell-phones can also be utilized as sensitive light detectors for various types of sensing. There have already been several successful attempts to utilize cell-phone cameras as portable digital light microscope systems.1–3 In addition, there are several cell-phone applications that can detect heart beats by analyzing the scattered light-emitting diode (LED) light from a finger placed over the camera lens. In previous communications,4,5 we described the design, construction, and operation of absorption and emission systems, which could be performed at remote areas, in order to detect and identify bacteria, pathogens, and molecules. In another previous communication,6 we utilized the compact monochrome CCD cameras, such as the ones used in the closed-circuit television (CCTV) cameras, to record the Ultraviolet (UV) fluorescence emitted from bacteria after excitation with Ultraviolet-C (UVC) LED light. We also correlated the damage inflicted, by Ultraviolet (UV) light, to the bacterial cells and the consequent decrease in the recorded fluorescence intensity owing to bacterial death. UV fluorescence can also be used to detect extremely low concentrations of other fluorescing bio-molecules such as proteins and amino acids, tryptophan and tyrosine.

In this paper, we investigated the use of cameras used in modern day cell-phones in conjunction with advanced spectroscopic detection devices. The motivation for such applications was to use them as a means for utilizing the continuously improving camera sensitivity and picture quality of the modern day cell-phone camera for scientific purposes such as handheld cell-phone sized spectrometers, for the detection of pathogens and harmful chemicals through Raman and fluorescence spectroscopies, in remote areas and places where laboratory spectrometers are unavailable or cannot be used because of size and power requirements.

Raman and fluorescence spectroscopies are two powerful means for the detection and study of the structure and reaction mechanism of a molecule and biological species, such as proteins, amino and nucleic acids, and bacteria or viruses. These fast and non-invasive techniques based on scattering and emission of light provide a fingerprint of the molecule and its structure. Fluorescence spectroscopy is a very sensitive method for detecting the small concentrations of biological molecules, while the recorded Raman spectra provide a means for determining the vibrational modes of the molecule, under study, and, consequently, the structure of its molecular bonds. A combination of these techniques provides means for determining the characteristic structure(s) of complex chemical and biological species. Raman spectroscopy may also be utilized for bio-molecules such as DNA and RNA, which either do not fluoresce or emit very low intensity fluorescence. Raman spectroscopy has also been utilized in numerous, diverse areas and applications, such as concentration of pharmaceutical drugs and other mixtures,7–10 determining the content of alcoholic fermentation in yeast11 and solid fat content in milk;12 determination of crystallinity of cellulose,13 identification of chemical isomers,14–16 and species concentration in cryogenic fuels for space industry17 are just a few of the applications of Raman spectroscopy. To that effect, several Raman instruments are utilized to identify drugs, harmful gases, and chemicals in the environment and industrial plants.18–21 

Raman spectroscopy has been applied for the detection and identification of biological molecules by excitation with ultraviolet and visible light in order to increase the signal intensity, thus generating enhanced resonance Raman spectra.22–27 

Several notable attempts have been made to build cost-effective Raman spectrometers;28–30 however, they still cost several thousands of USD, with the majority designed for laboratory-based experiments rather than for in-field testing. Our pocket sized designed and constructed Raman spectrometer system that utilizes right angle geometry to record Raman and resonance Raman spectra of molecules and biological species that contain colored pigments can be constructed for much lower cost (see Table I in the  Appendix). The detector of our Raman device, described in this paper, was a cell-phone CCD/CMOS sensor camera, suggesting that this inexpensive pocket Raman system has the potential of being an integral part of ubiquitous cell-phones and thus makes possible the identification of chemicals and pathogens in situ, within minutes.

Most commercial Raman spectrometers utilize backscattered geometry [Fig. 1(a)]; this geometry results in intense Raman spectra; however, they have the disadvantage of being difficult to remove the noise imposed by Rayleigh scattering and back reflected excitation light. To eliminate back reflected light, in this geometry, dichroic mirrors and expensive Rayleigh cutoff filters are required. In contrast, using the transmission form [Fig. 1(b)], even though the dichroic mirrors may be eliminated, the signal recorded is masked by the intense excitation laser line. Consequently, transmission Raman systems require high quality laser line rejection filters and laser cleanup filters to eliminate excitation laser stray modes. One drawback of the Rayleigh rejection filters is that they may also limit the wavelength range in the low Raman shift regions (<200 cm−1).

FIG. 1.

Raman spectroscopy geometries: (a) backscattered geometry, (b) transmission geometry, and (c) right angle geometry.

FIG. 1.

Raman spectroscopy geometries: (a) backscattered geometry, (b) transmission geometry, and (c) right angle geometry.

Close modal

The design of our cell-phone-based Raman system presented in this paper utilizes a 90° excitation, and the geometry of the Raman signal collection is shown in Fig. 1(c). This cell-phone sized instrument reduces the intensity of Rayleigh scattering falling on the detector, resulting in an efficient removal of Rayleigh scattering. This right angle (90°) excitation geometry also has the advantage of being easier to use for the analysis of samples where a bulk property is to be measured, such as liquids or suspensions, vs microscopic level qualities. This is so since the scattering light collection optics can collect light from a wider collection area when it is in the right angle geometry. This would be especially effective when the excitation laser emits an intense collimated beam or a focused beam with a longer depth of focus. In this manner, the macroscopic size of the sample is interrogated by the laser beam, thereby providing an average Raman spectrum of the sample in the presence of microscopic heterogeneities. Low cost diode laser and simple plastic lenses were utilized for the construction of our excitation system. The design and construction are simple. The identification and detection of pathogens, such as colored bacteria, solvents, and even impurities in food are easily detected using our system.

Figure 2 displays a schematic representation of our cell-phone Raman spectrometer and Fig. 3 displays a photo of the constructed system with the major components marked. A diode laser, emitting at 532 nm and 50 mW, is used as the excitation source; other wavelength diode lasers may also be used. The laser excitation beam is focused onto the sample by means of a lens (∼10 mm focal length and 4 mm diameter), and the Raman scatter light, from the sample, is collected by using a two-lens optical system, where the first lens collects and collimates the light, while the second focuses it with a matching f-number onto the spectrometer slit. The constructed spectrometer utilizes a transmission grating (1000 lines/mm), and a lens located inside the spectrometer collects and collimates the light onto the grating, thus increasing the intensity of the recorded spectra. It is interesting to note that this system operates effectively even without a collimating lens; however, the recorded Raman intensity is much lower.

FIG. 2.

Schematic diagram of the designed system.

FIG. 2.

Schematic diagram of the designed system.

Close modal
FIG. 3.

Photo of the constructed system: top view (a) and side view (b).

FIG. 3.

Photo of the constructed system: top view (a) and side view (b).

Close modal

Most of the Raman spectra presented in this paper were acquired by using Google Pixel 3a (model XL) and Google pixel XL cell-phones. Our system also performed well using other cell-phones such as Nokia Lumina 1020 and Motorola Moto G. As mentioned earlier, a diode laser, emitting at 532 nm and 50 mW (model: 1875-532D-50-5V obtained from Laserland), was utilized as the light source for the scattered Raman spectra. The transmission grating that disperses the spectrum had a groove density of 1000 lines/mm and was obtained from Rainbow Symphony. The laser focusing lens, fluorescence collection lens, and focusing lens were all obtained from Laserland. The collimating lens used in the spectrometer was ∼25 mm in diameter, with a 60 mm focal length, and was obtained from Thorlabs. Raman spectra of common solvents such as ethanol, acetone, isopropyl alcohol, and methanol were recorded in a 1 cm path-length quartz cuvette. The cuvette was placed in front of the focused excitation laser beam, and the emitted Raman spectra were recorded at right angle geometry, as shown in Fig. 2. For recording the Raman spectrum of opaque/solid objects, such as carrots and bacterial pellet, the sample was placed at the focal point at an angle of 45° to the incoming excitation laser beam, thereby directing the scattered Raman radiation toward the collection lens. A Rayleigh line cutoff filter was employed in all cases when Raman spectra of opaque objects were recorded. This Rayleigh filter was not normally utilized when the samples studied were clear and transparent.

The cell-phone was placed behind the transmission grating with the camera directly facing the grating. Proper alignment of the camera was ensured by placing a cuvette filled with a dilute solution of Rhodamine 6G dye, as the sample, and recording its intense fluorescence spectrum dispersed by the transmission grating. The Rhodamine 6G solution cuvette is then replaced by a cuvette filled with the sample solution whose Raman spectrum has to be recorded. The Rayleigh scatted laser excitation light from the solution could be seen with the cell-phone camera through the transmission grating. The autofocus mechanism in the cell-phone camera system was used in order to focus onto this Rayleigh scattered laser line, and the spectra were acquired in the night-sight mode. The night-side mode provides an exposure time of up to ∼180 s by averaging several short exposures. Alternatively, other camera applications in the cell-phone such as the HD camera were also used. These applications allow for manual focusing, International Organization for Standardization (ISO) sensitivity of the camera, and exposure times up to a few seconds. Depending on the noise in the recorded spectra and lower acquisition times, spectra can be averaged to obtain a higher signal-to-noise ratio (SNR). For example, utilizing the HD camera app using the Google pixel XL smartphone, the maximum exposure time was limited to 0.7 s. In such a case, we acquired 20 spectra and averaged them in order to improve the signal-to-noise ratio.

For the case where a Rayleigh line cutoff filter is employed in the optical path, a highly scattering solution such as micro-particle suspension was utilized to detect weak Rayleigh scattered light through the cell-phone camera, and then, the detected light was focused onto the camera system. Once the proper focus is achieved, for spectrum acquisition, the micro-particle suspension can then be replaced with the sample.

The recorded spectra were rotated, when necessary, in order to vertically display all the spectral lines. The vertical axis of the spectra was binned by using the median value of all the pixels in the vertical axis, in order to remove noise (salt and pepper noise, hot pixels, and other noise) inherent in the acquired spectra. The resulting one-dimensional spectra (wavelength, λ vs intensity) were scaled in the vertical direction, and the spectral intensity was subsequently plotted vs wavenumber. The pixel to Raman shift wavenumber calibration was performed using the known Raman bands of ethanol. Spectral rotation, binning, scaling, and intensity profile plots were performed using the ImageJ software. Raman spectra were not corrected for the variations of the quantum efficiency (intensity response) of the CMOS sensor as a function of the wavelength (instrument response correction). Figure 4 shows the effect of spectral processing in removing the noise.

FIG. 4.

Effect of spectral processing in removing the noise. The spectrum shown is the Raman spectrum of isopropanol in the fingerprint wavenumber region, acquired with our system.

FIG. 4.

Effect of spectral processing in removing the noise. The spectrum shown is the Raman spectrum of isopropanol in the fingerprint wavenumber region, acquired with our system.

Close modal

The absorption spectra of samples, when required, were recorded by using a Shimadzu UV160 spectrophotometer. The carotenes were extracted from the raw carrots in the acetone solution. The HD camera app was utilized for recording the Raman spectra in order to determine the ethanol percentage in a water solution. Spectral acquisition time was 3.9 s, the ISO value was 7100, and the manual focusing function was employed to obtain proper focus.

The aperture of a cell-phone camera lens is ∼2.5 mm. Even when the camera lens is placed directly in front of the grating, 2000 grooves are illuminated (1000 grooves per mm grating), making the resolving power, R value, of the system 2000. The theoretical resolution of the system is, thus, given by

Δλ=λ/R=0.25nm
(1)

at 500 nm. The cell-phone camera consists of a 12 megapixel sensor. The dispersion of the spectrum on the sensor was ∼0.3 nm per pixel. The resolution of the system is, however, limited by the input slit size, which should not be made too narrow because the amount of light reaching the transmission grating and subsequently to the cell-phone camera detector in the system is very small. In our optimum system, we obtained a spectral resolution ∼50 cm−1, with ∼150 μm slit (Fig. 5).

FIG. 5.

Calculation of the resolution of the system with an ethanol Raman spectrum. Acquired Raman spectrum of ethanol (a), intensity plot of the fingerprint region of the ethanol Raman spectrum (b), and Gaussian fit of the 885 cm−1 Raman peak (c).

FIG. 5.

Calculation of the resolution of the system with an ethanol Raman spectrum. Acquired Raman spectrum of ethanol (a), intensity plot of the fingerprint region of the ethanol Raman spectrum (b), and Gaussian fit of the 885 cm−1 Raman peak (c).

Close modal

In Fig. 5, the FWHM (Full Width at Half Maximum) was calculated by making a Gaussian fit,

y=Ae(xc)22σ2,
(2)

using the 885 cm−1 peak of ethanol using ImageJ software. FWHM was then calculated as follows:

FWHM=2σ2ln2
(3)

=49.67 cm−1, where σ is the standard deviation of the Gaussian fit.

Typical Raman spectra of several organic molecules were recorded using the cell-phone Raman spectrometer, shown in Fig. 6. Figure 6(a) shows the processed spectra, from the pictures and Figs. 6(b)6(d), which show the Raman spectra of ethanol, isopropyl alcohol, and methanol, respectively.

FIG. 6.

Typical Raman spectra of different chemicals recorded by using the cell-phone Raman system: (a) processed spectra from the captured picture for ethanol, isopropyl alcohol, and methanol; (b)–(d) show the intensity plots of the Raman spectra of ethanol, isopropyl alcohol, and methanol.

FIG. 6.

Typical Raman spectra of different chemicals recorded by using the cell-phone Raman system: (a) processed spectra from the captured picture for ethanol, isopropyl alcohol, and methanol; (b)–(d) show the intensity plots of the Raman spectra of ethanol, isopropyl alcohol, and methanol.

Close modal

The 532 nm excitation wavelength used for recording enhanced Raman spectra by our system is close to the absorption band of several common biological pigments such as carotenes (Fig. 7), which cause resonance enhancement31 of the Raman spectral intensities of the biological molecules that contain these pigments.

FIG. 7.

Typical absorption spectrum of carotene pigments in carrot.

FIG. 7.

Typical absorption spectrum of carotene pigments in carrot.

Close modal

The resonance enhanced Raman spectra of carrots recorded using the same cell-phone spectroscopic device is shown in Fig. 8. The carrot Raman spectra were converted from color to a monochrome image to avoid the intensity artifact on the fluorescence continuum due to transmission variances associated with the Red, Green, and Blue (RGB) Bayer filters used.

FIG. 8.

Resonance enhanced Raman spectrum of carrots recorded by using our cell-phone system. The acquired spectrum (a) and the intensity plots showing prominent carotenoid bands associated with C=C stretching at ∼1500 cm−1 and C–C stretching at ∼1140 cm−1 (b).

FIG. 8.

Resonance enhanced Raman spectrum of carrots recorded by using our cell-phone system. The acquired spectrum (a) and the intensity plots showing prominent carotenoid bands associated with C=C stretching at ∼1500 cm−1 and C–C stretching at ∼1140 cm−1 (b).

Close modal

In addition, the Raman spectra of bacteria, such as Micrococcus luteus and Serratia marcescens, which contain color pigments that absorb in the 532 nm region, are shown in Fig. 9.

FIG. 9.

Resonance enhanced Raman spectrum of Serratia marcescens recorded with our cell-phone spectroscopic device.

FIG. 9.

Resonance enhanced Raman spectrum of Serratia marcescens recorded with our cell-phone spectroscopic device.

Close modal

In order to test for quantitative precision of this device, we measured the ratio of ethanol’s 2935 cm−1 band to water’s 3400 cm−1 band at various known concentrations of ethanol in water. Such determination is important when the ratio of active ingredients in consumer products is needed. The experimental data are shown in Fig. 10.

FIG. 10.

Plot of the ratio of the 2950 cm−1 band of ethanol to the 3400 cm−1 Raman band of water as a function of ethanol percentage in water solution.

FIG. 10.

Plot of the ratio of the 2950 cm−1 band of ethanol to the 3400 cm−1 Raman band of water as a function of ethanol percentage in water solution.

Close modal

This novel system which we designed and described in this paper may also be used to record Raman spectra by scanning the sample in both the vertical and horizontal directions. The Raman spectra thus recorded display a linear Raman spectral image, which can be scanned in the vertical direction, thus providing a 2D Raman spectral image. Figure 11 shows the Raman spectral image of two cuvettes: one containing distilled water and the other ethanol placed next to each other. The water in red color is derived from the intensity of the water Raman band at 3400 cm−1, whereas green color intensity corresponds to the ethanol Raman band at 2935 cm−1.

FIG. 11.

Optical image (a) of two cuvettes containing ethanol and distilled water and the corresponding Raman image (b) of the scanned region on the right.

FIG. 11.

Optical image (a) of two cuvettes containing ethanol and distilled water and the corresponding Raman image (b) of the scanned region on the right.

Close modal

The cell-phone Raman spectrometer was compared to the HORIBA benchtop Raman spectrometer system (HORIBA XploRA), one of the most sensitive industrial Raman spectrometers, with distilled water as the test sample. The Signal-to-Noise Ratio (SNR) of the recorded Raman spectra was compared for the two instruments. The integration time for both the Horiba Raman spectrometer and cell-phone Raman spectrometer was set at 3.9 s. The grating selected in the HORIBA Raman spectrometer was 1200 lines/mm groove density, and the transmission grating in the cell-phone Raman spectrometer had a groove density of 1000 lines/mm. The excitation wavelength of both the spectrometers was 532 nm. The excitation power was 25 mW in the Horiba Raman system and 50 mW in the cell-phone Raman spectrometer. The spectrometer input slit was 200 µm in the Horiba Raman system, and it was 150 µm in the cell-phone Raman spectrometer. The excitation numerical aperture (NA) was 0.25 for Horiba and nearly the same for the cell-phone Raman spectrometer. The parameter chosen for the comparison was their Signal-to-Noise Ratio (SNR). The data of those experiments are shown in Fig. 12. The SNR of the Horiba system was roughly one order of magnitude better than the cell-phone system. We found that the SNR of the cell-phone system improves, nearly by a factor of 2, when a single RGB channel was used for analysis [Fig. 12(b)], Red (R) channel in this case. This is attributed to the fact that the Raman spectrum signal falls entirely in the red channel, and if we also include other channels, it results in the addition of noise from other channels, which does not contribute to the Raman signal. The noise depends, to a great extent, on the CCD/CMOS sensors used in the cell-phone cameras, which still suffer from large noise compared to the specialized CCD detectors utilized in the benchtop spectrometer instruments. This rather high noise in the cell-phone sensor may be attributed to the thermal factors (of uncooled detectors), small pixel sizes, and relative inferior CMOS sensors compared to specialized CCD sensors. Another contributing factor to the low intensity signal recorded by using the cell-phone instrument is the relatively lower diffraction efficiency of the transmission grating used in our cell-phone Raman spectrometer.

FIG. 12.

Comparison of the Signal-to-Noise Ratio (SNR) for the water Raman spectrum obtained from the Horiba Raman spectrometer and cell-phone Raman spectrometer. (a) The comparison when intensities from all the three color channels, R, G, and B, are combined in the cell-phone camera Raman system and (b) when intensity of only the R channel (where the spectrum falls) is considered.

FIG. 12.

Comparison of the Signal-to-Noise Ratio (SNR) for the water Raman spectrum obtained from the Horiba Raman spectrometer and cell-phone Raman spectrometer. (a) The comparison when intensities from all the three color channels, R, G, and B, are combined in the cell-phone camera Raman system and (b) when intensity of only the R channel (where the spectrum falls) is considered.

Close modal

In this study, we utilized cell-phone camera technology to record the Raman spectra of molecules, which provide a new, novel means for recording Raman, enhanced Raman, and other optical spectra, such as fluorescence. By recording the Raman spectra of biological molecules, this device provides a simple, reliable, and inexpensive method for the identification of molecules, bacteria, viruses, and other disease causing pathogens in situ. This instrument is intended to provide Raman and other spectral recording technologies as a commonplace for global positioning system (GPS) sensing or camera technology in a cell-phone.

The successful demonstration of the cell-phone camera-based Raman spectrometer system can be improved by eliminating several challenges in the cell-phone camera used as the spectroscopic sensor. Fortunately, most of these challenges can be rather easily overcome. One challenge was the fact that it is hard for lens-based systems to focus perfectly different wavelengths of light onto the image sensor. This effect might have played a role in degrading the spectral resolution of our system; however, this lack of tight focus was not noticeable, especially in the fingerprint region of the Raman spectra (400–1800 cm−1). Small defocusing was observed in the red region of the spectrum; however, even there, the effect was not large enough to prevent the accurate identification of many chemical and biological species studied.

A second challenge is the rather limited dynamic range of the cell-phone captured images (8-bit per color channel). This limited dynamic range minimizes the simultaneous recording of strong and weak Raman lines. This problem can be overcome by using one of the several HDR (High Dynamic Range) camera applications, which enhance the dynamic range of the captured images by combining multiple images at different exposures.

A third possible challenge may be caused by using the red, green, and blue Bayer filters, which may distort the intensity profile of smooth continua. To overcome this problem, the image may be converted to a monochrome image by using an appropriate algorithm, which takes into account the spectral response of the individual Bayer filters.

In the cell-phone Raman spectroscopic system design, presented here, we simplified the excitation and emission geometry by using the minimum number of optical components for recording Raman spectra and their easy integration to modern cell-phone technology. This inexpensive yet accurate recording pocket Raman system has the potential of being an integral part of ubiquitous cell-phones that will make it possible to identify chemical impurities and pathogens, in situ within minutes.

This work was supported, in part, by the Welch Foundation under Grant No. 1501928, the Air Force Office of Scientific Research under Grant No. FA9550-20-1-0139, and the Texas A&M Engineering Experiment Station (TEES). The authors would like to thank Professor Maria King for providing bacteria and valuable discussions.

The data that support the findings of this study are available within the article.

TABLE I.

Cost breakdown of various components for the design of the open access Raman spectroscopy system.

ComponentCost (USD)Source
Diode laser, 532 nm, 35 amazon.com 
50 mW 
Small focusing and 10 amazon.com 
collimating lenses 
Transmission 0.4 rainbowsymphonystore.com 
grating 
Cardboard/glue/  
boxes, etc. 
Total 50.5 + cell-phone  
ComponentCost (USD)Source
Diode laser, 532 nm, 35 amazon.com 
50 mW 
Small focusing and 10 amazon.com 
collimating lenses 
Transmission 0.4 rainbowsymphonystore.com 
grating 
Cardboard/glue/  
boxes, etc. 
Total 50.5 + cell-phone  
1.
H.
Zhu
and
A.
Ozcan
, “
Wide-field fluorescent microscopy and fluorescent imaging flow cytometry on a cell-phone
,”
J. Visualized Exp.
174
,
e50451
(
2013
).
2.
N. A.
Switz
,
M. V.
D’Ambrosio
, and
D. A.
Fletcher
, “
Low-cost mobile phone microscopy with a reversed mobile phone camera lens
,”
PLoS One
9
(
5
),
e95330
(
2014
).
3.
C. W.
Pirnstill
and
G. L.
Coté
, “
Malaria diagnosis using a mobile phone polarized microscope
,”
Sci. Rep.
5
(
1
),
13368
(
2015
).
4.
R.
Li
,
U.
Goswami
,
M.
Walck
,
K.
Khan
,
J.
Chen
,
T. C.
Cesario
, and
P. M.
Rentzepis
, “
Hand-held synchronous scan spectrometer for in situ and immediate detection of live/dead bacteria ratio
,”
Rev. Sci. Instrum.
88
(
11
),
114301
(
2017
).
5.
R.
Li
,
U.
Goswami
,
M.
King
,
J.
Chen
,
T. C.
Cesario
, and
P. M.
Rentzepis
, “
In situ detection of live-to-dead bacteria ratio after inactivation by means of synchronous fluorescence and PCA
,”
Proc. Natl. Acad. Sci. U. S. A.
115
(
4
),
668
673
(
2018
).
6.
D.
Dhankhar
,
R.
Li
,
A.
Nagpal
,
J.
Chen
,
A.
Krishnamoorthi
, and
P. M.
Rentzepis
, “
A novel approach for remote detection of bacteria using simple charge-coupled device cameras and telescope
,”
Rev. Sci. Instrum.
91
(
7
),
074106
(
2020
).
7.
C. J.
Strachan
,
T.
Rades
,
K. C.
Gordon
, and
J.
Rantanen
, “
Raman spectroscopy for quantitative analysis of pharmaceutical solids
,”
J. Pharm. Pharmacol.
59
(
2
),
179
192
(
2007
).
8.
A.
Heinz
,
M.
Savolainen
,
T.
Rades
, and
C. J.
Strachan
, “
Quantifying ternary mixtures of different solid-state forms of indomethacin by Raman and near-infrared spectroscopy
,”
Eur. J. Pharm. Sci.
32
(
3
),
182
192
(
2007
).
9.
L.
Saerens
,
L.
Dierickx
,
B.
Lenain
,
C.
Vervaet
,
J. P.
Remon
, and
T. D.
Beer
, “
Raman spectroscopy for the in-line polymer–drug quantification and solid state characterization during a pharmaceutical hot-melt extrusion process
,”
Eur. J. Pharm. Biopharm.
77
(
1
),
158
163
(
2011
).
10.
W.
Gong
,
R.
Shi
,
M.
Chen
,
J.
Qin
, and
X.
Liu
, “
Quantification and monitoring the heat-induced formation of trans fatty acids in edible oils by Raman spectroscopy
,”
J. Food Measurement and Characterization.
13
,
2203
2210
(
2019
).
11.
A.
Picard
,
I.
Daniel
,
G.
Montagnac
, and
P.
Oger
, “
In situ monitoring by quantitative Raman spectroscopy of alcoholic fermentation by Saccharomyces cerevisiae under high pressure
,”
Extremophiles
11
(
3
),
445
452
(
2007
).
12.
C. M.
McGoverin
,
A. S. S.
Clark
,
S. E.
Holroyd
, and
K. C.
Gordon
, “
Raman spectroscopic prediction of the solid fat content of New Zealand anhydrous milk fat
,”
Anal. Methods
1
(
1
),
29
38
(
2009
).
13.
U. P.
Agarwal
,
R. S.
Reiner
, and
S. A.
Ralph
, “
Cellulose I crystallinity determination using FT–Raman spectroscopy: Univariate and multivariate methods
,”
Cellulose
17
(
4
),
721
733
(
2010
).
14.
S.
Mahapatra
,
Y.
Ning
,
J. F.
Schultz
,
L.
Li
,
J.-L.
Zhang
, and
N.
Jiang
, “
Angstrom scale chemical analysis of metal supported trans- and cis-regioisomers by ultrahigh vacuum tip-enhanced Raman mapping
,”
Nano Lett.
19
(
5
),
3267
3272
(
2019
).
15.
G. F.
Bailey
and
R. J.
Horvat
, “
Raman spectroscopic analysis of the cis/trans isomer composition of edible vegetable oils
,”
J. Am. Oil Chem. Soc.
49
(
8
),
494
498
(
1972
).
16.
G. L.
Johnson
,
R. M.
Machado
,
K. G.
Freidl
,
M. L.
Achenbach
,
P. J.
Clark
, and
S. K.
Reidy
, “
Evaluation of Raman spectroscopy for determining cis and trans isomers in partially hydrogenated soybean oil
,”
Org. Process Res. Dev.
6
(
5
),
637
644
(
2002
).
17.
V. S.
Tiwari
,
R. R.
Kalluru
,
F. Y.
Yueh
,
J. P.
Singh
,
W. S.
Cyr
, and
S. K.
Khijwania
, “
Fiber optic Raman sensor to monitor the concentration ratio of nitrogen and oxygen in a cryogenic mixture
,”
Appl. Opt.
46
(
16
),
3345
3351
(
2007
).
18.
L.
Harper
,
J.
Powell
, and
E. M.
Pijl
, “
An overview of forensic drug testing methods and their suitability for harm reduction point-of-care services
,”
Harm Reduct. J.
14
(
1
),
52
(
2017
).
19.
K.
Dégardin
,
A.
Guillemain
, and
Y.
Roggo
, “
Comprehensive study of a handheld Raman spectrometer for the analysis of counterfeits of solid-dosage form medicines
,”
J. Spectrosc.
2017
,
3154035
.
20.
P. C.
Kumar
and
J. A.
Wehrmeyer
, “
Stack gas pollutant detection using laser Raman spectroscopy
,”
Appl. Spectrosc.
51
(
6
),
849
855
(
1997
).
21.
T. T. X.
Ong
,
E. W.
Blanch
, and
O. A. H.
Jones
, “
Surface enhanced Raman spectroscopy in environmental analysis, monitoring and assessment
,”
Sci. Total Environ.
720
,
137601
(
2020
).
22.
K. A.
Britton
,
R. A.
Dalterio
,
W. H.
Nelson
,
D.
Britt
, and
J. F.
Sperry
, “
Ultraviolet resonance Raman spectra of Escherichia Coli with 222.5–251.0 nm pulsed laser excitation
,”
Appl. Spectrosc.
42
(
5
),
782
788
(
1988
).
23.
C. R.
Johnson
,
M.
Ludwig
,
S.
O’Donnell
, and
S. A.
Asher
, “
UV resonance Raman spectroscopy of the aromatic amino acids and myoglobin
,”
J. Am. Chem. Soc.
106
(
17
),
5008
5010
(
1984
).
24.
Y. C.
Cao
,
R.
Jin
, and
C. A.
Mirkin
, “
Nanoparticles with Raman spectroscopic fingerprints for DNA and RNA detection
,”
Science
297
(
5586
),
1536
1540
(
2002
).
25.
J. M.
Benevides
,
S. A.
Overman
, and
G. J.
Thomas
, Jr.
, “
Raman, polarized Raman and ultraviolet resonance Raman spectroscopy of nucleic acids and their complexes
,”
J. Raman Spectrosc.
36
(
4
),
279
299
(
2005
).
26.
B.
Lorenz
,
C.
Wichmann
,
S.
Stöckel
,
P.
Rösch
, and
J.
Popp
, “
Cultivation-free Raman spectroscopic investigations of bacteria
,”
Trends Microbiol.
25
(
5
),
413
424
(
2017
).
27.
R.
Li
,
D.
Dhankhar
,
J.
Chen
,
A.
Krishnamoorthi
,
T. C.
Cesario
, and
P. M.
Rentzepis
, “
Identification of live and dead bacteria: A Raman spectroscopic study
,”
IEEE Access
7
,
23549
23559
(
2019
).
28.
A.
Arbildo L.
,
E. H.
Montoya R.
, and
O. R.
Baltuano E.
, “
A homemade cost effective Raman spectrometer with high performance
,”
J. Lab. Chem. Educ.
3
(
4
),
67
75
(
2015
).
29.
W. R. C.
Somerville
,
E. C.
Le Ru
,
P. T.
Northcote
, and
P. G.
Etchegoin
, “
High performance Raman spectroscopy with simple optical components
,”
Am. J. Phys.
78
(
7
),
671
677
(
2010
).
30.
C.
Mohr
,
C. L.
Spencer
, and
M.
Hippler
, “
Inexpensive Raman spectrometer for undergraduate and graduate experiments and research
,”
J. Chem. Educ.
87
(
3
),
326
330
(
2010
).
31.
D. P.
Strommen
and
K.
Nakamoto
, “
Resonance Raman spectroscopy
,”
J. Chem. Educ.
54
(
8
),
474
(
1977
).