Open Access
Add to BibSonomy
Abstract
Optoacoustic biomedical imaging combines the high spatial resolution of the ultrasound imaging with the specificity of the optical absorption spectroscopy techniques. It is being used in various scenarios such as anatomical, functional and molecular imaging. Typically light sources for this imaging technique is based on solid state lasers since they can produce high energy short optical pulses. However, they are bulky, expensive and the imaging speed is limited because their low pulse repetition rate. High power diode lasers (HPDLs) are a promising alternative for imaging small volume absorbers as they are compact, affordable and allow high repetition rates. However, HPDLs provide relative low peak optical power compared to solid state lasers. Therefore, imaging systems based on diode lasers require much longer pulse duration resulting in lower in-depth resolution and optoacoustic conversion efficiency. HPDLs need dedicated fast electronics to generate short optical pulses. In this work, we have designed, built and test a pulsed diode laser driver based on RF power MOSFETs, specifically considering the optimization of the current pulse in order to maximize the optical peak power, achieving current pulses of more than 900 A with a duration of 50 ns. We have studied the operation of a low cost HPDL out of the manufacturers datasheet ratings without noticeable degradation at high current (> 250 A) and short pulse duration (< 60 ns). We have obtained an optical peak power of 750 W and a energy per pulse of 31.2 µJ at 40 ns optical pulse duration. The optoacoustic images obtained in this operation regime shown a clear enhancement respect to the ones obtained in standard operation of the HPDL.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
In recent years, biophotonics have concentrated great scientific and business interest. Thanks to recent technological advances, more information can be extracted from the interaction between light and biological tissues, giving rise to more advanced and precise diagnostic techniques. In particular, the optoacoustic effect is a physical phenomenon that consists in the conversion of ultrashort pulses of light absorbed by chromophores present in a tissue into pressure waves that propagate though the medium. This phenomenon can be used in biomedicine to perform Optoacoustic Imaging Techniques (OITs) such as Optoacoustic Tomography (OAT) [1–3] and Optoacoustic microscopy [4–6] benefiting from the main advantages of optical imaging techniques, i.e. the high contrast and the possibility of spectroscopic analysis, and ultrasonic imaging techniques, namely the high spatial resolution and penetration depth. The OITs can provide oxygen saturation, structural and functional information, biomarkers and molecular detection, hemodynamic and gene expression monitoring, etc [7,8]. It can be found applications of OITs in oncology, cardiology, dermatology, etc [3,9].
OIT requires laser sources capable of emitting short duration pulses, less than few hundreds of nanoseconds, with energy between several $\mu$J to few mJ. The most commonly type of lasers used in OIT are solid-state lasers (Nd:YAG or Ti:Sapphire) and dye lasers [10,11]. Although these lasers meet and even exceed these requirements, they have drawbacks such as low pulse repetition rate of few tens of hertz that limits the framerate of the optoacoustic image system, elevated cost, periodical maintenance, and bulky size. The lack of affordable laser sources is slowing down the penetration of the OIT into the clinical practice. High Power Diode Lasers (HPDLs) can potentially be more suitable laser sources for commercial implementation of optoacoustic biomedical imaging applications compared to the commonly used solid state lasers due to their lower cost and compact size. In addition, they are capable of higher repetition rates what reduces the total acquisition time needed for obtaining optoacoustic images. Pulsed HPDLs has been used for OIT [12–17]. However, the commercially available HPDLs are designed and characterized to operate with much longer pulse duration and lower peak powers compared to what usually is required in the common OIT applications [18]. The use of HPDLs to produce short duration optical pulses for these applications requires the development of an specific high-speed driver electronics that can deliver peak current in the order of hundreds of amperes with nanosecond to tens of nanoseconds rise times. The typical laser diode drivers are based on the generation of pulses through an RLC network that stores the energy. This kind of drivers are not able to modulate the pulse width since the time constant of the circuit is fixed [19]. However, with the appearance of Gallium Nitride transistors (eGaN FETs), with better switching characteristics than silicon MOSFETS [20], it has been possible to obtain new variable drivers capable to produce pulses of tens of amps with less than 10 ns pulse duration [21].
The compliance voltage of the laser diode driver is mainly determined by the HPDL parasitic inductance due to packaging. The maximum voltage of the driver limits the rise slope of the current pulse. It is important to select an appropriate HPDLs packaging in order to minimize this inductance. The maximum current that a HPDL can handle, and therefore the maximum peak power it can deliver, is limited by the catastrophic optical damage (COD), a phenomenon by which the laser device is damaged and its emission power greatly decreases [22]. Typically, the manufacturers only specify the maximum current of the HPDLs in normal operation conditions with a certain pulse width and the repetition rate. These maximum current specifications are very conservative when the pulse duration is reduced from the normal operation, typically 100 ns. It is possible to operate beyond the maximum current specification without damage and increase the peak power and therefore the energy per pulse [23].
In this work we present a pulse driver with variable pulse width designed in accordance with the requirements of optoacoustic imaging, being able to produce pulses up to 950 A with pulse width of 50 ns. This driver was paired with a HPDL using the overdrive technique to increase the optical peak power and the peak current beyond its maximum ratings using short pulse duration with stable operation. We have reached a peak power of 752,7 W with a pulse duration of 40 ns. Optoacoustic images have been obtained using a wideband ultrasonic sensor at different pulse widths in order to demonstrate the improvement that this new HPDL operating regime entails. The image obtained by adapting the pulse width to the bandwidth of the transducer presents a better contrast than the image obtained with a larger pulse width and limited peak power, even though the pulse energy of the latter is 160% higher than that of the first.
2. Diode laser driver for high current short pulse generation
A current driver has been designed and implemented with capacity to generate high current pulses ($\sim$1000 A) and short pulse durations (up to 10 ns). It is based on a high-speed N-channel power MOSFET (DE275-101N30A, IXYS Corp., Milpitas, CA, USA) which is controlled through a MOSFET driver (IXRF631, IXYS Corp.). The MOSFET driver input is compatible with TTL or CMOS logic levels. A capacitor bank stores the energy during the transistor cut-off periods. The stored energy is delivered to the laser when the power MOSFET is activated generating a high current pulse. A pulse monitoring port is implemented to measure the current of the pulses. The Fig. 1(a) shows the block diagram of the pulse driver and Fig. 1(b) the electrical schematic.
With this driver configuration the maximum current pulse is:
where $V_{H}$ is the power supply voltage, $V_{laser}$ is the forward voltage of the HPDL and $R_{eq}$ is the equivalent resistance of the discharge loop: being $ESR$ the equivalent series resistance of the energy storage capacitors ($C_L$) , $R_{DSon}$ is the drain-source resistance of the power MOSFETs, $R_L$ is the current limiting resistor, $R_d$ is the dynamic parasitic resistance of the laser diode and $R_m$ is the resistor at the monitoring port.The minimum capacitance for the generation of high current pulses which ensures that the current pulse maintain an adequate amplitude is given by the following equation:
where $t_p$ is the pulse duration.The current pulse shape follows the next expression:
where $I_{pmax}$ is the peak current of the pulse, $t_p$ is the pulse duration, $C_L$ is the total capacitance.The rise time of the current pulse is limited by the parasitic inductance present in the packaging of the laser diode and the current path. The minimum rise time can be estimated using the following expression [24]:
where $L_p$ is the total parasitic inductance, $I_{pmax}$ is the peak current and $V$ is the supply voltage.The MOSFET driver and the power MOSFET dynamic characteristics have a very critical role in the generation of pulse current with fast transients. The MOSFET driver activates the power MOSFET. The main parameter of this device is the current that can supply into the power MOSFET gate. MOSFETs need an amount of charge to be injected into the gate electrode to turn on, called total gate charge. The MOSFET driver current required to turn-on the power MOSFET is given by:
where $I$ is the MOSFET driver current, $Q_g$ is the gate charge and $T_{d(on)}$ is the turn-on time.Other important parameters of the MOSFET driver are the rise time and fall time that have a direct dependence with the capacity gate $(C_{iss})$ of the power MOSFET, being necessary that this capacity is the smaller possible. To turn on properly a power MOSFET it is required a very low impedance path between the MOSFET driver and the power supply. The most common method to achieve this low impedance is to bypass the power supply at the driver with a capacitance value much larger than the load capacitance. Usually, this is done by placing two or three different types of by-passing capacitors, with complementary impedance curves, very close to the driver itself. These capacitors should be carefully selected for low inductance, low resistance, and high pulse current service. The MOSFET driver must be able to drain all the current into an adequate grounding system. There are two paths for returning current that need to be considered: one between the MOSFET driver and power MOSFET and another between the MOSFET driver and its power supply. Both of these paths should be as low as possible in resistance and inductance and therefore as short as practical. The MOSFET driver IXRFD631 has a capacity to inject up to 30 A pulse current with very fast rise and fall times (4 ns).
The power MOSFET must have the capacity to withstand high pulse current (hundreds of amperes), fast rise and fall times and high operating voltage. Parameters such as total gate capacity ($C_{iss}$), gate charge ($Qg$), resistance ($R_g$ and $R_{ds}$) should be as small as possible, also the packaging has to be adequate to dissipate a large amount of power and reduce the inductances ($L_d$, $L_g$, $L_s$) to the minimum. The purposed design has two power MOSFETs to maximize the current and the transition speed of the pulses, as well as to balance the load and the dissipation. The disposition of the MOSFETS in the PCB and their discharge paths have to be perfectly symmetrical to ensure that the current pulses generated by each one are perfectly synchronized. The theoretical pulse current that the two power MOSFETs (DE275-101N30A) can provided is 480 A with maximum pulse duration in the range of microseconds, a maximum $V_{DSS}$ of 100V and rise and fall times equal to 5 ns and 8 ns, respectively. The MOSFET presents a low $R_{DS(on)}$ equal to 60 $m\Omega$.
The capacitors responsible for providing the energy of the pulses must be arranged in parallel to minimize the $ESR$ and be as close as possible to the laser. A total of 43 capacitors of 1.2 $\mathrm {\mu }$F have been arranged in the driver with a total storage capacitance ($C_L$) of 51.6 $\mathrm {\mu }$F.
With short pulse duration and high currents, the effects of stray inductance ($L_p$) are very important [25], limiting the rise times of the pulses and the maximum current. It can be reduced some of this $L_p$[26] taking special considerations on the PCB design. The $V_{cc}$ to ground current path defines the loop that generates the inductive term. This loop was kept as short and wide as possible and treated as coplanar transmission lines to avoid issues related to the output lead inductance. Furthermore, the output ground traces must provide a balanced symmetric coplanar return for optimum operation.
A monitoring port has been implemented through multiple resistors ($R_m$), giving a ratio in the pulse measured of 80 A/V in 50 $\Omega$.
Finally, a snubber ($C_{snb}$, $R_{snb}$) and a filter ($C_m$) are implemented to improve the shape of the pulse current. To stabilize the operating temperature of the power MOSFET we have added passive cooling and a temperature monitoring through a thermocouple. Two clamping diodes have been placed in parallel with the HPDL to reduce the inductive voltage overshoot and protect the device.
Taking into account all the previous considerations, a compact driver was designed and manufactured. Figure 2 shows the driver with the HPDL.
3. Driver and HPDL results
The following experiments test the capability of the driver to generate short high pulse current either in short-circuit operation and with a laser connected. The pulse width and the repetition rate are controlled by an external pulse generator. A fast oscilloscope is used to record the signal from the monitor port. When the driver is tested with a HPDL, an ultra-high speed photodiode is used to obtain the shape of the emitted optical pulse and a power meter is used to measure the average optical power. Finally, the emitted pulse peak power is obtained by taking into account the recorded shape of the pulse and the average optical power.
3.1 Short circuit characterization
Figure 3(a) shows the current pulse obtained shorting the output of the driver with different pulse width, 1 kHz of repetition rate, 100 V power supply and 20 V in power MOSFET gate. The maximum current peak obtained is 944 A at 50 ns doubling the maximum pulse current of the driver specified by the manufacturer to 480 A (240 A each MOSFET) for pulse width up to 300 $\mu$s and 2$\%$ duty cycle. For shorter pulse widths the peak current is limited by the path inductance ($\approx$3 nH) and the power MOSFET. When the pulse width increases and we approach to the voltage limit of the Power MOSFET ($V_H$= 100 V), the pulse current slope decreases due to the saturation in the curve transfer of the MOSFET, limiting the maximum current (Fig. 3(a) and (b). After the main current pulse there is a ripple that are typical in this operation conditions [27]. This ripple is reduced using the clamping diodes mentioned in section 2. Table 1 summarize the main characteristics of the electrical pulses in short circuit.
Fig. 3. (a) Short-circuit current pulses at different pulse width for $V_H$ = 100 V, and (b) I-V driver characteristic curve for different pulse width.
Table 1. Results with short-circuit for $V_H$ = 100 V
3.2 High power diode laser
In order to increase the output power of HPDL devices it is common to place multiple optical cavities in a single package. The HPDL used in this work is the SPLS4L90A (OSRAM Opto Semiconductors, Munich, Germany) formed by an array of four laser chips, each chip has three stripe lines, giving a total of twelve emitting cavities. Each optical cavity has the capacity to emit a maximum optical peak power of 40 W, therefore, this device has a total optical peak power of 480 W with a current of 160 A, the emission wavelength is 905 nm. The HPDL packaging is SMT QFN with low inductance (< 2nH). The drawback of this strategy to increase peak power is that the more cavities that are added to increase output power, the lower the beam quality.
In our previous work [23], we have explored the ability of HPDLs to increase their peak power by reducing the pulse width beyond the manufacturer maximum ratings. Figure 4 shows the safe operation zone of HPDL. The area under the line indicates the possible power values that can be achieved using smaller pulses without damaging the device. In this graph it is considered 100 $\%$ peak power the one corresponding to a pulse width of 100 ns, that is the manufacturers normal operation condition for the SPLS4L90A. In order to improve the optical peak power the driver has to provide, for shorter pulses, enough current to reach the limits of the safe region under the curve. As an example, for 50 ns pulses the maximum optical peak power would be 141.41 $\%$ of the power at 100 ns giving as a result a value of 678.76 W and the current necessary to reach this value will be approximately 226 A.
The HPDL SPLS4L90A has been tested using the designed driver following a procedure of several steps. First, the normal operation of the laser was checked, i.e., at nominal current, pulse width and 1 kHz repetition rate for correct operation assessment. We obtained the energy per pulse measuring the average power using a thermopile (S401C Thorlabs, Newton, NJ, USA) knowing the pulse repetition rate. The optical pulse shape was recorded using a fast APD photodiode (APDC5658, Hamamatsu Photonics, Hamamatsu, Japan) and an oscilloscope. Combining both measurements, we monitored online the peak power of the pulses. From this starting point, the injected current, and therefore the peak power, were increased in steps while the current pulse width was decreased. During each step the pulse width (FWHM) and the optical peak power are monitored continuously for 60 second to check the stability. The optical peak power was raised and the pulse width decreased in steps until either the maximum optical peak power or the maximum driver voltage was reached.
Figure 5 depicts a comparison between the current pulse and the 40 ns width optical pulse achieved. The rise of the optical pulse perfectly follows the electrical pulse, but the optical pulse fall time is much slower than the rise time, with a constant value for the descent slope of around 4.9 A/ns (conversion of optical power to current), widening the optical pulse duration. The electrical pulse duration is 18 ns whereas the optical pulse duration is 40 ns. The long fall time is due to the disconnection of the HPDL from the circuit during the off state of the MOSFET in combination with the capacitance of the HPDL. In the driver the four HPDL chips are connected in parallel what result in a total parasitic capacitance equal to the add up of the individual chips capacitance. This capacitance prevents the HPDL from being discharged by the carrier recombination quickly.
As shown by Fig. 6(a), we reach the critical point with a pulse duration of 40 ns, the optical peak power is 752,7 W at 258.4 A. For shorter pulse widths the driver cannot supply enough current to reach the maximum peak power limited by the maximum working voltage (100 V), from this point the current slope is not fast enough to reach higher currents due to the combined inductance of the driver ($\sim$3 nH) and the HPDL ($\sim$2 nH). Figure 6(b) shows the shape of optical pulses achieved with different pulse width and the Table 2 summarizes the main parameters achieved.
Table 2. HPDL results. The maximum peak power reached is 752.7 W at 40 ns pulse width, the peak power decreases with smaller pulse widths since the drive does not supply enough current due to the current slope is limited by the inductance.
The current values obtained for widths less than 40 ns are similar to the theoretical ones since the driver is limiting by the maximum voltage. To further reduce the pulse width increasing the current either the inductance of the HPDL package has to be reduced or the power transistor has to be changed to other one that can withstand higher maximum voltage. For values greater than 40 ns the limiting factor for the optical peak power is the maximum current that the laser can withstand without suffering COD. For this reason, the driver current was limited to keep the peak optical power below the threshold value of the COD by reducing the current control voltage ($V_H$).
Regarding the operating frequency, it is possible to operate beyond 1 kHz, but the duty cycle of the HPDL is very restrictive (typically 0.01%). For example, for a pulse of 40 ns, the maximum frequency would be 2.5 kHz. Experimentally the driver was tested up to 2 kHz without failures due to COD or driver thermal issues.
The temperature of the driver and the laser was monitored during the operation using a thermocouple and an infrared camera (FLIR ONE Pro LT, Teledyne FLIR LLC, Oregon, US). Figure 7(a) and (b) shows the infrared camera image with overdrive and nominal operation respectively. Power MOSFETs do not suffer from heating due to the low average current, rising from 25 degrees (room temperature) to 28 degrees in operation. Regarding the HPDL (bright area at the top of image), we can see changes in its temperature, being higher in nominal operation, given the higher energy per pulse and therefore average power than in overdrive operation where the energy is lower, in both cases the temperature does not limit the maximum power. The high brightness shown by the transistor pads is due to the high emissivity of the metal.
Fig. 7. (a) Thermal image of driver and HPDL with overdrive operation (752.7 W optical peak power, 40 ns pulse width) (b) Thermal image of driver and HPDL with nominal operation (480W optical peak power, 100 ns pulse width).
4. Optoacoustic imaging
We have investigated the change in the quality of the optoacoustic image reconstructions respect to the pulse width of the optical pulses used to illuminate the target. We have employ the proposed current driver and the same HPDL characterized in the previous section. To appreciate any improvement in the frequencies contents it is required to use an UltraSonic Transducer (UST) with a wide bandwitdh enough. We have used a 200 $\mathrm {\mu }$m hydrophone (Precision Acoustics Ltd, Dorchester, UK) with a calibrated response from 1 to 25 MHz. We will use an optical pulse width of 20 ns (25 MHz) and 100 ns (10 MHz), which is the maximum optical pulse width of the HPDL.
The schematic of the optoacoustics imaging system, which was used for signal and imaging study, is depicted in Fig. 8. The target is placed in a petri dish filled with water and it is illuminated by the HPDL from the bottom side. The container is held by a 3D translation stage (TS) that allows to scan a plane above the target. The UST is precisely positioned in the center of the illumination in order to maximized the optoacoustic signals. The laser is attached to the developed driver which is triggered using an external pulse generation (PG) at a frequency of 1 kHz. For imaging, we used stop-and-go scan, the step distance between measures is 200 $\mu$m with a total of measured points of 1425. The signals are amplified 20 dB (Amp) and then captured using an oscilloscope (OS) with an average of 64 signals. The captured signal are boxcar filtered and decimated before the image reconstruction to improve the SNR and reduce the computer calculation burden. We have used a 3D FFT reconstruction for a planar sensor algorithm [28] for the reconstruction of the optoacoustic images. The method uses a k-space algorithm which performs (1) a Fourier transform on the data along both t, y, and z dimensions (into wavenumber-frequency space), (2) a mapping, based on the dispersion relation for a plane wave in an acoustically homogeneous medium, from wavenumber-frequency space to wavenumber-wavenumber space, and finally (3) an inverse Fourier transform back from the wavenumber domain to the spatial domain. The result is an estimation of the initial acoustic pressure distribution from which the acoustic waves are propagated.
The target is made out of laser printing the letters ’GOTL’ in a acetate sheet, Fig. 10(a). Two scan using 40 ns and 100 ns pulse widths were performed. The distance from the target to the transducer is 6.75 mm (4.5 $\mu$s). An example of the measured acoustics signals and their spectra obtained for both pulse widths at the same scanning point are depicted in Fig. 9. We observe that both the pressure and the bandwidth are greater for the signals generated with the shorter pulse width. The optoacoustic conversion is more efficient with smaller pulse widths due to the fulfilment of the stress confinement condition [29]. In this setup, as the optical pulses are not short enough, the optoacoustic signal results from the convolution of the optical pulse and the illuminated absorber shape.
Fig. 10. (a) ’GOTL’ laser printing in a acetate sheet (b) Ink GOTL optoacoustic imaging with HPDL nominal operation (465 W,100 ns) , (c) Ink GOTL optoacoustic imaging with HPDL enhanced operation (700 W,40 ns)), (d) Profile GOTL central horizontal line amplitude.
The corresponding reconstructed images are shown in Fig. 10(b) and Fig. 10(c). We can appreciate a clear improvement of the reconstructed image with the smaller pulse width. Figure 10(d) shows the amplitude profile of the reconstructions along the dashed line, showing a greater difference between the maximum and minimum amplitudes for the smaller pulse width indicating an improvement of the contrast from 51% to 83%. In addition, a better approximation to the dimensions of the real object is also observed. In Table 3 we compare the dimensions of the object with the ones obtained from the reconstructions. For large dimensions the two reconstructions have similar results but for small dimensions (for example the thickness of the letter L), the pulse of shorter duration is a 25.77% closer to the real value.
Table 3. Optoacoustics image dimensions
5. Conclusions
In this article, we have designed and manufactured a pulse current driver for generation of high peak current pulses with variable pulse width in the range of tens of nanoseconds for optoacoustic imaging generation. By adapting the optical emission bandwidth to the actual ultrasound receiver spectrum response the image quality of the reconstruction can be enhanced. The maximum current reached in short-circuit operation is 944 A for a 50 ns pulse width. A commercial HPDL have been tested achieving a maximum peak power of 752.7 W at 40 ns pulse width in stable operation. This value is 153$\%$ higher than the manufacturers maximum peak current given for a 100 ns pulse duration. In order to demonstrate the imaging improvement, two images have been carried out with the HPDL at different pulse width, 40 ns and 100 ns, observing that the image with the shortest pulse presents greater bandwidth and pressure generated, which leads to a reconstructed image with greater contrast with an improvement of 63% .
Funding
Comunidad de Madrid (S2018/NMT-4333 MARTINLARA-CM, PEJD-2018-PRE/IND-8162).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but maybe obtained from the authors upon reasonable request.
References
1. J. Xia, J. Yao, and L. Wang, “Photoacoustic tomography: Principles and advances,” Prog. Electromagn. Res. 147, 1–22 (2014). [CrossRef]
2. V. Neuschmelting, N. C. Burton, H. Lockau, A. Urich, S. Harmsen, V. Ntziachristos, and M. F. Kircher, “Performance of a multispectral optoacoustic tomography (msot) system equipped with 2d vs. 3d handheld probes for potential clinical translation,” Photoacoustics 4(1), 1–10 (2016). [CrossRef]
3. L. V. Wang and S. Hu, “Photoacoustic tomography: In vivo imaging from organelles to organs,” Science 335(6075), 1458–1462 (2012). [CrossRef]
4. S. Jeon, J. Kim, D. Lee, J. W. Baik, and C. Kim, “Review on practical photoacoustic microscopy,” Photoacoustics 15, 100141 (2019). [CrossRef]
5. S. M. Maswadi, B. L. Ibey, C. C. Roth, D. A. Tsyboulski, H. T. Beier, R. D. Glickman, and A. A. Oraevsky, “All-optical optoacoustic microscopy based on probe beam deflection technique,” Photoacoustics 4(3), 91–101 (2016). Special Issue: Photoacoustic Microscopy. [CrossRef]
6. M. Erfanzadeh, P. D. Kumavor, and Q. Zhu, “Laser scanning laser diode photoacoustic microscopy system,” Photoacoustics 9, 1–9 (2018). [CrossRef]
7. Y. Zhang, M. Jeon, L. Rich, H. Hong, J. Geng, Y. Zhang, S. Shi, T. Barnhart, P. Alexandridis, J. D. Huizinga, M. Seshadri, W. Cai, C. Kim, and J. Lovell, “Non-invasive, multimodal functional imaging of the intestine with frozen micellar naphthalocyanines,” Nat. Nanotechnol. 9(8), 631–638 (2014). [CrossRef]
8. C. Kim, C. Favazza, and L. Wang, “In vivo photoacoustic tomography of chemicals: High-resolution functional and molecular optical imaging at new depths,” Chem. Rev. 110(5), 2756–2782 (2010). [CrossRef]
9. M. Jeon, J. Kim, and C. Kim, “Multiplane spectroscopic whole-body photoacoustic imaging of small animals in vivo,” Med. Biol. Eng. Comput. 54(2-3), 283–294 (2016). [CrossRef]
10. K. H. Song, C. Kim, K. Maslov, and L. V. Wang, “Noninvasive in vivo spectroscopic nanorod-contrast photoacoustic mapping of sentinel lymph nodes,” Eur. J. Radiol. 70(2), 227–231 (2009). Advances in Molecular Imaging. [CrossRef]
11. R. Esenaliev, I. Larina, K. Larin, D. Deyo, M. Motamedi, and D. Prough, “Optoacoustic technique for noninvasive monitoring of blood oxygenation: A feasibility study,” Appl. Opt. 41(22), 4722 (2002). [CrossRef]
12. R. Kolkman, W. Steenbergen, and T. van Leeuwen, “In vivo photoacoustic imaging of blood vessels with a pulsed laser diode,” Lasers Med. Sci. 21(3), 134–139 (2006). [CrossRef]
13. T. J. Allen and P. C. Beard, “Pulsed near-infrared laser diode excitation system for biomedical photoacoustic imaging,” Opt. Lett. 31(23), 3462–3464 (2006). [CrossRef]
14. P. K. Upputuri and M. Pramanik, “Performance characterization of low-cost, high- speed, portable pulsed laser diode photoacoustic tomography (pld-pat) system,” Biomed. Opt. Express 6(10), 4118–4129 (2015). [CrossRef]
15. L. Zeng, G. Liu, Y. Di-Wu, and X. Ji, “Portable optical-resolution photoacoustic microscopy with a pulsed laser diode excitation,” Appl. Phys. Lett. 102(5), 053704 (2013). [CrossRef]
16. K. Daoudi, P. Berg, O. Rabot, A. Kohl, S. Tisserand, P. Brands, and W. Steenbergen, “Handheld probe integrating laser diode and ultrasound transducer array for ultrasound/photoacoustic dual modality imaging,” Opt. Express 22(21), 26365–26374 (2014). [CrossRef]
17. A. Hariri, A. Fatima, N. Mohammadian, S. Mahmoodkalayeh, M. A. Ansari, N. Bely, and M. R. N. Avanaki, “Development of low-cost photoacoustic imaging systems using very low-energy pulsed laser diodes,” J. Biomed. Opt. 22(7), 075001 (2017). [CrossRef]
18. S. B. Gawali, L. Leggio, M. Sánchez, S. Rodríguez, E. Dadrasnia, D. C. Gallego, and H. Lamela, “Combining high power diode lasers using fiber bundles for beam delivery in optoacoustic endoscopy applications,” in Semiconductor Lasers and Laser Dynamics VII, vol. 9892K. Panajotov, M. Sciamanna, A. Valle, and R. Michalzik, eds., International Society for Optics and Photonics (SPIE, 2016), pp. 310–316.
19. A. Stylogiannis, L. Prade, A. Buehler, J. Aguirre, G. Sergiadis, and V. Ntziachristos, “Continuous wave laser diodes enable fast optoacoustic imaging,” Photoacoustics 9, 31–38 (2018). [CrossRef]
20. M. Danilovic, Z. Chen, R. Wang, F. Luo, D. Boroyevich, and P. Mattavelli, “Evaluation of the switching characteristics of a gallium-nitride transistor,” in 2011 IEEE Energy Conversion Congress and Exposition, (2011), pp. 2681–2688.
21. J. Glaser, “How gan power transistors drive high-performance lidar: Generating ultrafast pulsed power with gan fets,” IEEE Power Electron. Mag. 4(1), 25–35 (2017). [CrossRef]
22. H. Wenzel, P. Crump, A. Pietrzak, X. Wang, G. Erbert, and G. Tränkle, “Theoretical and experimental investigations of the limits to the maximum output power of laser diodes,” New J. Phys. 12(8), 085007 (2010). [CrossRef]
23. M. Sánchez, D. Gallego, and H. Lamela, “Overdrive short pulse high peak power diode lasers without catastrophic optical damage,” IEEE Photonics J. 13(4), 1–7 (2021). [CrossRef]
24. Picolas, “Impedance of laser diodes inductive behaviour,” Application Note.
25. Z. Chen, D. Boroyevich, and R. Burgos, “Experimental parametric study of the parasitic inductance influence on mosfet switching characteristics,” in The 2010 International Power Electronics Conference - ECCE ASIA -, (2010), pp. 164–169.
26. J. Nissinen and J. Kostamovaara, “A 4 a peak current and 2 ns pulse width cmos laser diode driver for high measurement rate applications,” in 2013 Proceedings of the ESSCIRC (ESSCIRC), (2013), pp. 355–358.
27. Y. Xiao, H. Shah, T. Chow, and R. Gutmann, “Analytical modeling and experimental evaluation of interconnect parasitic inductance on mosfet switching characteristics,” in Nineteenth Annual IEEE Applied Power Electronics Conference and Exposition, 2004. APEC ’04., vol. 1 (2004), pp. 516–521 Vol.1.
28. B. E. Treeby and B. T. Cox, “k-Wave: MATLAB toolbox for the simulation and reconstruction of photoacoustic wave fields,” J. Biomed. Opt. 15(2), 021314 (2010). [CrossRef]
29. T. Lee, H. W. Baac, Q. Li, and L. Guo, “Efficient photoacoustic conversion in optical nanomaterials and composites,” Adv. Opt. Mater. 6(24), 1800491 (2018). [CrossRef]
