We describe fabrication of microstructures by two-photon polymerization using bursts of femtosecond laser pulses. With the aid of an acousto-optic modulator driven by a function generator, two-photon polymerization is performed at variable burst repetition rates. We investigate how the time between the bursts of laser pulses influences the ultimate dimensions of lines written in a photosensitive resin. We observe that when using the same laser fluence, polymer lines fabricated at different burst repetition rates have different dimensions. In particular, the widths of two-photon polymerized lines become smaller with decreasing burst repetition rates. Based on the thermal properties of the resin and experimental writing conditions, we attribute this effect to localized heat accumulation.
©2012 Optical Society of America
The ability to fabricate complex three-dimensional microstructures with great accuracy and with sub-micron feature sizes is at the present time readily achieved by employing two-photon polymerization (TPP) . The relevance of this technical advancement in the field of microfabrication is demonstrated by the multitude of applications where TPP is successfully implemented. Whether creating microstructures as unique prototypes or producing patterns otherwise impossible to realize with different microfabrication techniques, researchers have used TPP in diverse fields including bioengineering, photonics, microfluidics, and micro-electrical mechanical systems [2–9].
The vast majority of TPP research so far has been concentrated on either applications or methods for minimizing the smallest feature size that can be polymerized . The latter aspect in particular has experienced great interest from the scientific community, since several groups have demonstrated that feature size equal or smaller than 100 nm can be reproducibly obtained by TPP. In some of these articles small feature sizes are achieved by proximity effects, gentle excitation fields, and shorter excitation wavelengths [11–15]. In other ones the authors borrowed an ingenious idea from the world of nonlinear microscopy and have applied it to TPP with promising results [16–18]. Although these studies have highlighted how flexible TPP is in micro- and nano-patterning, they have also revealed that the mechanism that leads to the generation of the reactive species that start polymerization upon nonlinear absorption of light is perhaps more complex than initially thought.
Indeed, even if experiments described in several recent works yielded results that are attractive for their potential applications, the same experiments have also produced unexpected outcomes. In one case for example, photosensitive resins were found producing feature sizes by TPP that are inversely dependent on exposure time . This astonishing finding is contrary to the common experience where patterns created by TPP are larger when exposure times are longer. In another case, state-of-the-art three-dimensional microstructures were written in commercial photosensitive resins using a continuous-wave laser centered at 532 nm . The most surprising part of this work is not only the quality of the microstructures created that is comparable or in some cases better than the quality of similar microstructures fabricated using ultra-short pulsed lasers, but also the fact that the mechanism of microfabrication was proven to be based on two-photon absorption even if the laser average power used in the experiments was only around 10 mW. These works in conjunction with others [16, 21, 22], points to a mechanism for the generation of active species capable to start polymerization that is different to the well-accepted process that occurs via linear absorption of light .
In this model, when a photon of appropriate energy (usually near UV) impinges on a photosensitive resin (hereafter resin) that can be polymerized by a radical mechanism, photoinitiator molecules get excited to the first electronic state. After non-radiative decay, photoinitiators undergo an intersystem crossing to a triplet state. The lifetime of this state is quite short (~100 ps) and rapidly produces two reactive species (radicals) through a homolytic bond cleavage that can start polymerization by reacting with saturated and/or unsaturated olefins. This energetic pathway has been experimentally characterized for photoinitiator molecules excited by one-photon absorption [24, 25]. In TPP however, photoinitiators may follow different energetic routes. To explain unexpected results such as the ones described earlier, authors have suggested (although not experimentally proven yet) energetic possibilities for photoinitiators including formation of long-lived excited states, pairs of solvated electrons and radical dications, and successive excited states absorptions with consequential heating of the resin. In two previous reports the authors have emphasized their attentions on the photophysics and photochemistry of TPP when using tight focusing conditions and excitation fields delivered by either low repetition rate amplified fs lasers or high repetition rate ps lasers. In the first case (fs laser), it was found that radical generation was controlled mostly by avalanche ionization . In the second case (ps laser), it was found that photopolymerization was induced by either a linear or nonlinear absorption process depending on the laser repetition rate .
Although it might be impossible to arrive at a universal mechanism for TPP due to the great variety of materials and experimental conditions thus far used, we seek in this article to explore TPP in the traditional and most common settings. That is, excitation is provided by a femtosecond (fs) pulsed laser oscillator with high repetition rate (80 MHz) and the resin is composed of a mixture of acrylic monomer and a commercially available photoinitiator. Since most published works in the field of TPP employ similar excitation conditions and materials, we believe the findings of this study to be most relevant for better understanding and eventually optimizing the process of TPP. From now on, the term TPP will be used to generally describe any photopolymerization process that is started by nonlinear absorption of light.
Using sequences of bursts of fs laser pulses as excitation, and varying the temporal distance between these bursts, we can distinguish thermal from photochemical contributions in TPP. The inspiration for this work came from previous research where a similar method was implemented for understanding the mechanism that leads to the fabrication of waveguides within bulk glasses by fs direct laser writing [27, 28]. Although the mechanism of light absorption and the magnitude of heat accumulation are different between waveguide writing and TPP, we believe that with a burst mode approach TPP can also be investigated leading to a better understanding of the several steps that renders this process so unique in the fabrication of three-dimensional microstructures.
2. Methods and materials
Although a detailed description of the experimental setup employed for TPP can be found elsewhere [29, 30], a brief account of it is presented here. The source of light is a Ti:sapphire oscillator delivering 100 fs pulses at a constant repetition rate (rp) of 80 MHz and with center wavelength of 800 nm (Spectra-Physics MaiTai DeepSeeTM). The laser beam is focused into the resin by means of a 40x microscope objective lens with a numerical aperture of 0.75. Microfabrication is performed by keeping the focused laser beam fixed while the sample is moved in predetermined geometries with the aid of a computer-controlled, three-axis translational stage assembly (Newport Corp. Laser µFAB).
TPP excitation from the 80 MHz train of fs laser pulses is converted to an excitation with a series of bursts of fs laser pulses by means of a distinct device that intercepts the laser beam before entering the microfabrication workstation (Fig. 1a ). In this system, an acousto-optic modulator (AOM) works as a gate transmitting only a certain number of fs pulses (Fig. 1b). A pair of fast photodiodes (PD) is used to monitor the temporal profile of the pulse sequences with the aid of an oscilloscope. Both the temporal width of the bursts (Δt) and their repetition rate (Rp) are controlled by a digital function generator (FG). A square type function from the FG is used to create the bursts of fs laser pulses. By positioning an iris after the AOM, only the diffracted beam is allowed to reach the microfabrication workstation. Throughout this work, Δt is kept constant at 1 µs while Rp is varied from 0.5 MHz to 0.05 MHz. Thus, each burst contains 80 laser pulses and the distance between consecutive bursts (T = 1/Rp – Δt) ranges from 1 µs to 19 µs.
To evaluate the effect of fs laser burst mode writing in TPP, we fabricated a series of suspended lines across support walls at a height of 20 µm. Each line is made only by one laser pass. In this way, the full width of the written lines can be measured precisely by using a scanning electron microscope (SEM) to eliminate the confusion, and possible source of error, that arises from the truncation effect when writing microstructures on the surface of a substrate . The dependence of the polymerized line widths as a function of exposure time is examined for each Rp by varying the writing scans speed at constant laser average power. The conditions used in the experiments described in this work are summarized in Table 1 . Each letter in this table corresponds to one set of polymerized line widths versus laser scan speed. The velocity of the stage used in all set of data (A to E) in Table 1 is varied from 10 µm/s to 100 µm/s. For each scan speed, three lines are written and their average value is taken as the line width for that particular combination of laser average power and writing scan speed. Under these experimental conditions, the volume defined by the focused laser beam is always struck by a large number of bursts. For example, the numbers of bursts per volume when writing at 50 µm/s are approximately 13000 and 1300 for Rp of 0.5 MHz and 0.05 MHz, respectively.
A variable attenuator is used at the exit of the fs laser to select and maintain a constant level of laser average power going into the AOM setup. When lowering Rp by means of the digital function generator, the laser average power going into the microfabrication workstation decreases by a factor that is proportional to the ratio of the new repetition rate to the original one. The energy per burst and the energy per pulse remain constant throughout this work and they are 100 nJ/burst and 1.25 nJ/pulse, respectively.
The negative tone resin used for TPP consists of two components, a photoinitiator (2-benzyl-2-dimethylamino-4’-morpholinobutyrophenone, Sigma-Aldrich) and a branched monomer (pentaerythritol triacrylate, Sigma-Aldrich). The photoinitiator is used in a concentration of 0.6% in weight; the monomer constitutes the remaining part of the resin. The starting materials are used as received without any further purification. Prior to use, the resin is mixed until a homogenous solution is obtained. The resulting viscous liquid is then applied onto flat glass substrates by drop casting. TPP is started at the resin/substrate interface to ensure that microstructures do not move during the writing process and above all that they may survive the subsequent developing step. After completion of the writing process, the unsolidified part of the resin is washed away in a bath of isopropyl alcohol first and acetone later, revealing the desired microstructures on the glass substrate.
3. Results and discussions
Scanning electron micrographs of test samples used to study the effect of scan velocity on polymerized line widths are shown in Fig. 2 . A top view of one of these microstructures is presented in Fig. 2(a) where it can be observed that while the majority of the suspended lines are straight, some are either deformed or completely missing. This occurrence is quite common when performing dosimetry experiments in TPP. It is explained by taking into consideration several materials factors such as shrinkage, surface tension generated during drying of the solvent used in the developing step, and (for some lines) poor mechanical strength [32–34]. Rows of horizontal parallel lines in Fig. 2 were made using different experimental conditions as described in Table 1. In particular, going from the left row to the right one, the order of the writing conditions is E, A, B, C, and D. The microfabrication scan speed is increased in the vertical direction. Lines at the bottoms of each row are made with lower scan speed than lines at the top.
The widths of the suspended lines are retrieved by magnified images such as the one in the inset of Fig. 2(a). In this case, three lines were made under writing conditions E (Table 1) and scan velocity of 60 µm/s. The average value for the width of these three lines is 0.50 ± 0.05 µm. Figure 2(b) is a tilted view of a test sample. It reveals both the 10 µm thick vertical walls that are used as supports and the horizontal lines that run across them. It can be observed that the latter ones are anchored to the top of the support walls and therefore they are suspended over the substrate.
The widths of polymerized lines as a function of writing speeds are shown in Fig. 3(a) . Five different trends are plotted corresponding to the five experimental conditions as described in Table 1. Increasing writing speed corresponds to diminishing exposure time, thus leading to smaller polymerized line widths. Furthermore, smaller features are obtained when using lower laser average powers. When writing at 40 µm/s for example, the polymerized line widths changes from 2.22 µm to 0.50 µm when the laser average power is decreased from 50 mW to 5 mW due to the change in bursts repetition rate. Within the experimental conditions used in this study, the fabricated lines that emerged from the developing step range from straight to wavy. The mechanical properties of microstructures fabricated by TPP depend on the exposure conditions that, in consequence, control the polymer degree of cross-linking. Stronger microstructures correspond to more extensive cross-linking. When writing at speeds of 100 µm/s for example, only lines fabricated with experimental conditions A, B, C, and D survived the developing step while the line fabricated with experimental condition E is washed away. Furthermore, because of the different exposure conditions, we expect the mechanical properties of lines fabricated using the same writing speed in Fig. 3 to be different; even in the case when line widths happen to be similar, such as the lines produced using writing speed of 100 µm/s and experimental conditions C and D. Thus, the data shown in Fig. 3 summarizes and compares only microstructures dimensions and not their mechanical properties.
The smallest line width is 0.44 ± 0.03 µm and it is obtained with 5.0 mW laser average power, 70 µm/s writing speed, and Rp = 0.05 MHz. This value is almost three times smaller than the diffraction limited spot size (beam diameter = 2ω0 = 1.22 λ/NA = 1.3 µm). The ability to break the diffraction limit in TPP is explained by taking into consideration the threshold effect [11, 35]. For the resin to undergo a phase transformation from liquid to solid that produces features with enough structural integrity to survive the developing step, a minimum density of active species (i.e. radicals, cations) need to be formed at the focal point of the laser beam following light absorption. Only then can polymerization be successfully sustained for creating microstructures by TPP. As a consequence of this phenomenon, a threshold light intensity is formed. By setting only the very top of the laser Gaussian beam intensity over this threshold, features with dimensions smaller than the one predicted by optical diffraction are achieved.
By using the threshold effect, it can be demonstrated also that the relationship between the width (W) of the polymerized lines and the writing speed (v) follows the equation:36]. The square of the polymerized line widths versus scan speeds is displayed in Fig. 3(b) using a semilogarithmic plot. The solid lines are linear regressions and demonstrate the high fidelity with which our data follow the trend predicted by Eq. (1). Fth can be retrieved from Fig. 3(b) when writing speeds approach zero. The value we obtain for it is (32 ± 9) kJ/cm2. This number is consistent with the laser fluence threshold measured in a resin containing the same photoinitiator used in our system .
Surprisingly the rate at which the square of the polymerized line widths decreases with increasing writing speed (slopes of lines in Fig. 3(b)) varies going from experimental conditions A to E. In particular, this slope diminishes when using lower bursts repetition rates and consequentially lower laser average powers. For data acquired at P = 5mW and Rp = 0.05 MHz for examples, lines made at different writing speed show almost no widths variance, that is feature size is independent of light exposure (at least under the writing speeds used in this study). Based on Eq. (1), the slope of the data shown in Fig. 3(b) is given by 2ω02. Hence, the only effect that lowering laser average power should have is to shift the data set on the ordinate axis. If the threshold model used to obtain Eq. (1) is describing fully the TPP process, then one would expect Fig. 3(b) to show a series of parallel lines with matching negative slopes and different intercepts in the ordinate axis. Although this effect has been experimentally observed in a commercial resin excited by a train of pulses from a fs laser oscillator [11, 38], the threshold effect model obviously fails in explaining completely our results. Hence, some additional phenomena must be considered in combination with the threshold effect to interpret the unexpected behavior shown in Fig. 3(b).
The work performed in this paper differs from previous experiments in the method with which energy is deposited into the resin (e.g. 80 MHz train of fs pulses versus bursts of fs pulses at variable repetition rates). Therefore to compare results from this work with previous experiments one must look at lines created using identical net fluence (NF). Doing so will enable us to investigate how a defined amount of energy influences the ultimate line width created by TPP when it is deposited into the resin under different time regimes. Net fluence is defined as:Fig. 4 can be constructed. In it, radial dimensions of lines written with the same NF (48 kJ/cm−2) are plotted in function Rp. While almost a flat dependence is observed at large Rp, a sharp decrease in line width is observed for Rp that range from 0.05 MHz to 0.10 MHz which corresponds to times T between 9 µs and 19 µs. This observation suggests that the ultimate effect produced in TPP after light absorption is not only dependent on the amount of energy deposited into the system but it also depends on how this energy is deposited over time. Choosing a set of data for lines fabricated with the constant Rp/v different than 5·103 µm−1, generates a similar trend as the one shown in Fig. 4 with a sharp decrease in line width for times T between 9 µs and 19 µs as well.
We evaluate two different potential mechanisms for investigating the physical reasons that can justify the trend in Fig. 4. While the first possibility takes into account the accumulation and diffusion of matter, the second one considers the accumulation and diffusion of heat. The only species that can accumulate at the focal volume during the repetitive arrival of bursts of fs laser pulses are either radicals or photoinitiator molecules in an intermediate (long-lived) excited state. The existence of the latter ones has not been experimentally proved yet, but it has been postulated to elucidate the mechanism of recent findings in resolution augmentation through photo-induced deactivation photolithography . In both cases (radicals and excited photoinitiators) the diffusion constants in unpolymerized resin at room temperature are quite low (10−9 – 10−6 cm2/s) and they will tend to contract during polymerization because of the change in the medium viscosity from liquid to solid [39–41]. Thus, we believe that this factor plays a small or insignificant role in the line width variations observed in Fig. 4. It will take in fact longer than tens of microseconds for radicals and/or excited photoinitiators to diffuse out of the focal volume.
We believe the situation becomes physically more realistic if heat accumulation is considered as part of the TPP process. In this model, the effect of the localized temperature increase due to the repetitive arrivals of burst of fs laser pulses must be characterized by the material cooling time (tc) defined as tc = (2ω0)2/χ where χ is the resin temperature diffusion constant. Assuming that the value of χ for the material of this study is similar to the one measured for polymethylmethacrylate (1.07·10−3 cm2/s) [42–44], we can infer a cooling time of 16 µs under the employed excitation conditions. Since heat can augment the feature size produced by photo-polymerization (either directly by decomposing the photoinitiator into radicals, or indirectly by affecting the rate constants of the polymerization mechanism) we would expect a change in polymerized line widths at values of T comparable to tc. Specifically, for T < tc the width of polymerized lines should increase due to heat accumulation while for T > tc heat should have enough time between bursts to escape the volume within the focused laser beam and consequentially have no effect on polymerized line widths. The vertical line in Fig. 4 corresponds to tc and it demarks the experimental data in exactly the aforementioned way. Widths of polymerized lines written with the same NF get bigger as T values become smaller. At shortest T they reach a plateau and at T = tc their rate of growth is maximum. These results strongly suggest that heat accumulation plays an important role in TPP, and are consistent with measurements performed on microstructures made using picosecond lasers with variable repetition rates .
One effect that has not been taken into consideration during this discussion is polymerization shrinkage. This phenomenon is particularly relevant in TPP where high degrees of polymer cross-linking are needed to fabricate robust microstructures. Polymer cross-linking, and thus shrinkage, depends on the total amount of energy per volume deposited within the resin. Since the points of the plot in Fig. 4 correspond to line widths of microstructures fabricated with the same NF, the overall trend of the represented data (with the resultant conclusions) is not affected by polymerization shrinkage.
Depending on the writing conditions, we demonstrate that heat accumulation contributes to the ultimate dimensions of the polymerized features in TPP. Although only a detailed description of the photoinitiator energetic path from the ground state to the formation of radicals can reveal exactly how local heating occurs during TPP, a plausible mechanism is the repetitive absorption and subsequent non-radiative decay of the photoinitiator when in an excited state .
By using a sequence of bursts of fs laser pulses from a high repetition rate oscillator, we have elucidated the mechanism of TPP in an acrylic based resin containing a commercially available photoinitiator. In particular, by varying the repetition rate of the bursts we have shown that the ultimate dimensions of the written microstructures depend in part on heat induced polymerization. Thus, in conjunction with the more traditional view of a pure photo-induced chemical reaction, it is important to consider also the effect of heat accumulation when performing TPP. This is particularly true in applications where the highest spatial accuracy in three-dimensional microstructures is required, and in applications where the smallest feature size and resolution are sought after. In these cases, a burst mode excitation approach when performing TPP is attractive in order to eliminate heat accumulation effects. Future work will focus on determining whether performing TPP with bursts of fs laser pulses can deliver microstructures with smaller features size and eventually improve writing resolution than the ones obtained with the traditional CW train of 80 MHz fs laser oscillator. Furthermore, it will be interesting to apply in a systematic approach the writing methodology described in this article to a series of resins containing photoinitiators belonging to different class of molecules and having different values of two-photon cross-sections.
References and links
1. S. Maruo and J. T. Fourkas, “Recent progress in multiphoton microfabrication,” Laser Photonics Rev. 2(1-2), 100–111 (2008). [CrossRef]
2. P. Tayalia, C. R. Mendonca, T. Baldacchini, D. J. Mooney, and E. Mazur, “3D Cell-migration studies using two-photon engineered polymer scaffolds,” Adv. Mater. (Deerfield Beach Fla.) 20(23), 4494–4498 (2008). [CrossRef]
3. F. Klein, B. S. Richter, T. Striebel, C. M. Franz, G. Freymann, M. Wegener, and M. Bastmeyer, “Two-component polymer scaffolds for controlled three-dimensional cell culture,” Adv. Mater. (Deerfield Beach Fla.) 23(11), 1341–1345 (2011). [CrossRef]
4. M. Deubel, G. von Freymann, M. Wegener, S. Pereira, K. Busch, and C. M. Soukoulis, “Direct laser writing of three-dimensional photonic-crystal templates for telecommunications,” Nat. Mater. 3(7), 444–447 (2004). [CrossRef] [PubMed]
5. L. Li, E. Gershgoren, G. Kumi, W.-Y. Chen, P. T. Ho, W. N. Herman, and J. T. Fourkas, “High-performance microring resonators fabricated with multiphoton absoprtion polymerization,” Adv. Mater. (Deerfield Beach Fla.) 20(19), 3668–3671 (2008). [CrossRef]
6. G. Kumi, C. O. Yanez, K. D. Belfield, and J. T. Fourkas, “High-speed multiphoton absorption polymerization: fabrication of microfluidic channels with arbitrary cross-sections and high aspect ratios,” Lab Chip 10(8), 1057–1060 (2010). [CrossRef] [PubMed]
7. J. Wang, Y. He, H. Xia, L. G. Niu, R. Zhang, Q. D. Chen, Y. L. Zhang, Y. F. Li, S. J. Zeng, J. H. Qin, B. C. Lin, and H. B. Sun, “Embellishment of microfluidic devices via femtosecond laser micronanofabrication for chip functionalization,” Lab Chip 10(15), 1993–1996 (2010). [CrossRef] [PubMed]
9. R. A. Farrer, C. N. LaFratta, L. Li, J. Praino, M. J. Naughton, B. E. A. Saleh, M. C. Teich, and J. T. Fourkas, “Selective functionalization of 3-D polymer microstructures,” J. Am. Chem. Soc. 128(6), 1796–1797 (2006). [CrossRef] [PubMed]
10. Y. L. Zhang, Q.-D. Chen, H. Xia, and H.-B. Sun, “Designable 3D nanofabrication by femtosecond laser direct writing,” Nano Today 5(5), 435–448 (2010). [CrossRef]
11. T. Tanaka, H.-B. Sun, and S. Kawata, “Rapid sub-diffraction-limit laser micro/nanoprocessing in a threshold material system,” Appl. Phys. Lett. 80(2), 312–314 (2002). [CrossRef]
12. S. Juodkazis, V. Mizeikis, K. K. Seet, M. Miwa, and H. Misawa, “Two-photon lithography of nanorods in SU-8 photoresist,” Nanotechnology 16(6), 846–849 (2005). [CrossRef]
13. D. Tan, Y. Li, F. Qi, H. Yang, Q. Gong, X. Dong, and X. Duan, “Reduction in feature size of two-photon polymerization using SCR500,” Appl. Phys. Lett. 90(7), 071106 (2007). [CrossRef]
14. S. H. Park, T. W. Lim, D. Y. Yang, N. C. Cho, and K. S. Lee, “Fabrication of a bunch of sub-30-nm nanofibers inside microchannels using photopolymerization via a long esposure technique,” Appl. Phys. Lett. 89(17), 173133 (2006). [CrossRef]
15. W. Haske, V. W. Chen, J. M. Hales, W. Dong, S. Barlow, S. R. Marder, and J. W. Perry, “65 nm feature sizes using visible wavelength 3-D multiphoton lithography,” Opt. Express 15(6), 3426–3436 (2007). [CrossRef] [PubMed]
16. L. Li, R. R. Gattass, E. Gershgoren, H. Hwang, and J. T. Fourkas, “Achieving lambda/20 resolution by one-color initiation and deactivation of polymerization,” Science 324(5929), 910–913 (2009). [CrossRef] [PubMed]
17. J. Fischer, G. von Freymann, and M. Wegener, “The materials challenge in diffraction-unlimited direct-laser-writing optical lithography,” Adv. Mater. (Deerfield Beach Fla.) 22(32), 3578–3582 (2010). [CrossRef] [PubMed]
18. J. Fischer and M. Wegener, “Ultrafast polymerization inhibition by stimulated emission depletion for three-dimensional nanolithography,” Adv. Mater. (Deerfield Beach Fla.) 24(10), OP65–OP69 (2012). [CrossRef] [PubMed]
19. M. P. Stocker, L. J. Li, R. R. Gattass, and J. T. Fourkas, “Multiphoton photoresists giving nanoscale resolution that is inversely dependent on exposure time,” Nat. Chem. 3(3), 225–227 (2011). [CrossRef] [PubMed]
20. M. Thiel, J. Fischer, G. von Freymann, and M. Wegener, “Direct laser writing of three-dimensional submicron structures using a continuous-wave laser at 532 nm,” Appl. Phys. Lett. 97(22), 221102 (2010). [CrossRef]
21. J. F. Xing, X. Z. Dong, W. Q. Chen, X. M. Duan, N. Takeyasu, T. Tanaka, and S. Kawata, “Improving spatial resolution of two-photon microfabrication by using photoinitiator with high initiating efficiency,” Appl. Phys. Lett. 90(13), 131106 (2007). [CrossRef]
22. M. Malinauskas, A. Zukauskas, G. Bickauskaite, R. Gadonas, and S. Juodkazis, “Mechanisms of three-dimensional structuring of photo-polymers by tightly focussed femtosecond laser pulses,” Opt. Express 18(10), 10209–10221 (2010). [CrossRef] [PubMed]
23. C. Decker, “Photoinitiated curing of multifunctional monomers,” Acta Polym. 45(5), 333–347 (1994). [CrossRef]
24. S. Jockusch, I. V. Koptyug, P. F. McGarry, G. W. Sluggett, N. J. Turro, and D. M. Watkins, “A Steady-State and Picosecond Pump-Probe Investigation of the Photophysics of an Acyl and a Bis(acyl)phosphine Oxide,” J. Am. Chem. Soc. 119(47), 11495–11501 (1997). [CrossRef]
25. C. S. Colley, D. C. Grills, N. A. Besley, S. Jockusch, P. Matousek, A. W. Parker, M. Towrie, N. J. Turro, P. M. W. Gill, and M. W. George, “Probing the Reactivity of Photoinitiators for Free Radical Polymerization: Time-Resolved Infrared Spectroscopic Study of Benzoyl Radicals,” J. Am. Chem. Soc. 124(50), 14952–14958 (2002). [CrossRef] [PubMed]
26. M. Malinauskas, P. Danilevičius, and S. Juodkazis, “Three-dimensional micro-/nano-structuring via direct write polymerization with picosecond laser pulses,” Opt. Express 19(6), 5602–5610 (2011). [CrossRef] [PubMed]
27. S. M. Eaton, H. B. Zhang, P. R. Herman, F. Yoshino, L. Shah, J. Bovatsek, and A. Arai, “Heat accumulation effects in femtosecond laser-written waveguides with variable repetition rate,” Opt. Express 13(12), 4708–4716 (2005). [CrossRef] [PubMed]
28. R. R. Gattass, L. R. Cerami, and E. Mazur, “Micromachining of bulk glass with bursts of femtosecond laser pulses at variable repetition rates,” Opt. Express 14(12), 5279–5284 (2006). [CrossRef] [PubMed]
29. T. Baldacchini, M. Zimmerley, E. O. Potma, and R. Zadoyan, “Chemical mapping of three-dimensiona microstructures fabricated by two-photon polymerization using CARS microscopy,” in Proc. of SPIE, 2009), 72010Q–72011.
30. T. Baldacchini, M. Zimmerley, C. H. Kuo, E. O. Potma, and R. Zadoyan, “Characterization of Microstructures Fabricated by Two-Photon Polymerization Using Coherent Anti-Stokes Raman Scattering Microscopy,” J. Phys. Chem. B 113(38), 12663–12668 (2009). [CrossRef] [PubMed]
31. H.-B. Sun, T. Tanaka, and S. Kawata, “Three-dimensional focal spots related to two-photon excitation,” Appl. Phys. Lett. 80(20), 3673–3675 (2002). [CrossRef]
32. S. Maruo, T. Hasegawa, and N. Yoshimura, “Single-anchor support and supercritical CO2 drying enable high-precision microfabrication of three-dimensional structures,” Opt. Express 17(23), 20945–20951 (2009). [CrossRef] [PubMed]
33. H. B. Sun, T. Suwa, K. Takada, R. P. Zaccaria, M. S. Kim, K. S. Lee, and S. Kawata, “Shape precompesation in two-photon laser nanowriting of photonic lattices,” Appl. Phys. Lett. 85(17), 3708–3710 (2004). [CrossRef]
34. A. Ovsianikov, J. Viertl, B. Chichkov, M. Oubaha, B. MacCraith, I. Sakellari, A. Giakoumaki, D. Gray, M. Vamvakaki, M. Farsari, and C. Fotakis, “Ultra-low shrinkage hybrid photosensitive material for two-photon polymerization microfabrication,” ACS Nano 2(11), 2257–2262 (2008). [CrossRef] [PubMed]
36. C. Martineau, R. Anemian, C. Andraud, I. Wang, M. Bouriau, and P. L. Baldeck, “Efficient initiators for two-photon induced polymerization in the visible range,” Chem. Phys. Lett. 362(3-4), 291–295 (2002). [CrossRef]
37. J. Serbin, A. Egbert, A. Ostendorf, B. N. Chichkov, R. Houbertz, G. Domann, J. Schulz, C. Cronauer, L. Fröhlich, and M. Popall, “Femtosecond laser-induced two-photon polymerization of inorganic-organic hybrid materials for applications in photonics,” Opt. Lett. 28(5), 301–303 (2003). [CrossRef] [PubMed]
38. W. H. Teh, U. Durig, G. Salis, R. Harbers, U. Drechsler, R. F. Mahrt, C. G. Smith, and H. J. Guntherodt, “SU-8 for real three-dimensional subdiffraction-limit two-photon microfabrication,” Appl. Phys. Lett. 84(20), 4095–4097 (2004). [CrossRef]
39. N. Fang, C. Sun, and X. Zhang, “Diffusion-limited photopolymerization in scanning micro-stereolithography,” Appl. Phys., A Mater. Sci. Process. 79(8), 1839–1842 (2004). [CrossRef]
40. A. Pikulin and N. Bityurin, “Spatial resolution in polymerization of sample features at nanoscale,” Phys. Rev. B 75(19), 195430 (2007). [CrossRef]
41. I. Sakellari, E. Kabouraki, D. Gray, V. Purlys, C. Fotakis, A. Pikulin, N. Bityurin, M. Vamvakaki, and M. Farsari, “Diffusion-assisted high-resolution direct femtosecond laser writing,” ACS Nano 6(3), 2302–2311 (2012). [CrossRef] [PubMed]
42. J. Morikawa, A. Orie, T. Hashimoto, and S. Juodkazis, “Thermal diffusivity in femtosecond-laser-structured micro-volumes of polymers,” Appl. Phys., A Mater. Sci. Process. 98(3), 551–556 (2010). [CrossRef]
43. L. Flach and R. P. Chartoff, “A process model for nonisothermal photopolymerization with a laser light source. I: basic model development,” Polym. Eng. Sci. 35(6), 483–492 (1995). [CrossRef]
44. J. Morikawa, A. Orie, T. Hashimoto, and S. Juodkazis, “Thermal and optical properties of femtosecond-laser-structured PMMA,” Appl. Phys., A Mater. Sci. Process. 101(1), 27–31 (2010). [CrossRef]
45. J. Fischer and M. Wegener, “Three-dimensional optical laser lithography beyond the diffraction limit,” Laser Photonics Rev. (2012).