Assessing the impact of spatial coherence on the sinusoidal linear Fresnel zone plate’s depth of focus Authors: Maryam Farahiyan, Rasoul Aalipour Published: 2024 ABSTRACT AbstractThis research examines, using both theory and experimentation, how a light beam’s spatial coherence affects a sinusoidal linear Fresnel zone plate’s depth of focus. We generate a one-dimensional partially coherent Gaussian Schell-model beam from a coherent laser beam through putting a rotating diffuser near to the common focal plane of two cylindrical lenses in a 2f-system. By adjusting the beam spot size on the diffuser, one may modify the coherence width of the produced beam. Our findings demonstrate that when the coherence width of the partly coherent beam drops, the zone plate’s depth of focus increases. Not only is it claimed that the experimental results validate the theoretical conclusions, but a mathematical link is also established between the experimental parameter used to manipulate the light coherence and the theoretically obtained coherence width. This topic has many applications in the field of optical, microscopy, optical trapping, imaging systems and biomedical.Subject terms: Optics and photonics, Physics Introduction The ability of a light beam to interfere with a delayed or spatially shifted version of itself is considered as coherence. This property of the light changes as the light beam propagates through space1. So, coherence is closely related to interference and diffraction phenomena. Coherent light sources in ideal imaging systems are usually point or linear, while in practice, each light source has a spatial extent, causing the light produced by these sources to not be spatially coherent. Therefore, the study of incoherent and partially coherent light beams is a significant topic of research related to imaging, measuring and diagnostic systems2–11. In this regard, various methods have been developed to measure spatial coherence, among which Young’s double slits experiment and its generalized models are significant in terms of history and their developing applications1. Also, in the last decade, several spatial coherence measurement methods based on the analysis of diffraction patterns have been presented12,13. We have also recently presented a method for measuring spatial coherence using phase step diffraction patterns and its application in measuring source width14. In order to conduct deeper studies on partial coherent sources, we need to know how the properties of the beams change for different sources. These light sources are found in nature or easily produced in the laboratory. One of the common sources that produces partial coherent light is the Gaussian Schell-model source. The Gaussian Schell-model is defined by a Gaussian mean intensity function and a normalized Gaussian correlation or degree of coherence function2. Simulations and experimental methods of producing such beams have been studied by many researchers from different perspectives15–23. The properties of the Gaussian Schell-model light beams in the focus of various lenses, especially the study of their depth of focus (DOF) have attracted the attention of some researchers24. DOF is a lens optics concept that measures the tolerance of placement of the image plane to the lens2. It has been studied in various literature in terms of imaging, optical microscopy, digital camera, optical metrology, bio-medical, and so on25–30. As we know from geometrical optics, the image of an object is not ideal due to the existence of aberrations in imaging systems. This is because the DOF of the lens, which determines the longitudinal range of image formation, is strongly influenced by the aberrations of the optical system31,32. So in optical systems and especially in optical microscopes, depth of focus and elimination of aberrations are important issues. One of the ways to avoid the aberration effects is to use the diffractive lenses instead of the refracted lenses. Fresnel zone plate (FZP) is a type of diffractive lens that includes a wide range of spatial frequencies, and can produce multiple foci. In recent years, FZPs have been used in systems that are difficult or impossible to perform by refractive optics33. FZP is composed of transparent and opaque zones. When the light beams are diffracted from the transparent zones, they interfere constructively on the plane perpendicular to the optical axis passing through the center of the plate and form the maximum intensity at the focal point of the FZP. FZPs are made in various shapes such as circular, square, petal, and linear. The linear Fresnel zone plate (LFZP) acts like a cylindrical lens and produces a linear focus34,35. To consider FZPs, the coherent light beams have often been used, and less attention has been paid to the light beams with low coherence36. In this paper, we study the focusing properties of a sinusoidal linear Fresnel zone plate (SLFZP) under illumination of a partially coherent light beam. Specifically, we measure the DOF of the SLFZP and establish a relationship between the DOF of the zone plate and the transverse coherence width of an illuminated Schell-model partially coherent beam. The paper consists of two main theoretical and experimental sections, in each of which the requirements of the study and the results are given. Theory Paraxial focusing of partially coherent light beam by sinusoidal linear Fresnel zone plate In Fig. 1, a monochromatic partially coherent parallel light beam is focused by a sinusoidal linear Fresnel zone plate (SLFZP) with transmission function A monochromatic partially coherent parallel light beam is focused by an SLFZP. where is the coordinate at the zone plate plane, λ and f respectively are the wavelength of incident light and the focal length of the SLFZP. Suppose denotes the field in front of the SLFZP. Then the field distribution immediately behind the plate is given by the expression37 Within the framework of Fresnel diffraction, the field distribution at any point (x, z) behind the plate becomes where is the coordinate at the observation plane and k = 2π/λ is the wavenumber. On substituting Eq. (2) into Eq. (3), we find that By use of the second-order statistical properties of partially coherent light beams, the mutual intensity of the fields on a screen at distance z from the SLFZP is given by38 The angular brackets denote the average taken over an ensemble of statistical realizations of the beam and is the mutual intensity of the partially coherent light incident to the plate. On the other hand, the mutual intensity can be written in terms of the measurable quantities of the second-order statistical properties, Here, stands for the intensity of the th field, and is the correlation coefficient37. We assume that the statistics of the incident light beams are stationary, and hence, we can rewrite Eq. (6) as follows38 where and . By considering Eq. (5) and using Eq. (7), the intensity of the partially coherent light beam on the observation plane in the focal region, with , takes the following form14,39 The inner integral, {⋯}, seems to be simply the autocorrelation integral of that we call quadratic phase multiplicative transmission function (QPT). We have already shown that the Fourier transform of the autocorrelation integral of QPT gives the intensity of diffraction of coherent light beam from the field just to the right of the aperture39. Using the discrete Fourier transform (DFT) method39, the normalized intensity of diffraction of coherent light beam from the SLFZP in the focal region is expressed as follows Equation (8) states that the intensity on the screen is given by the Fourier transform of the product of the autocorrelation of QPT with the correlation coefficient of the incident light beam to the SLFZP. On the other hand, the convolution theorem states that the Fourier transform of the product of two functions equals the convolution integral of their Fourier transforms39. So, we can write Eq. (8) as the convolution where ∗ denotes one-dimensional convolution and . Finally, taking Fourier transform twice from the both sides of Eq. (10), the intensity of the partially coherent light beam on the observation plane in the focal region can be written as follows In deriving this equation, we have used the convolution and Fourier integral theorems35,39. Gaussian–Schell model beams Suppose that the SLFZP is illuminated by a one dimensional Gaussian Schell–model (1D-GSM) beam. In that case the intensity and correlation coefficient of the beam can be written as follows37 The parameters A and B are position-independent constants, and σ and δ respectively are the source width and the coherence width. Now, substituting these relations for 1D-GSM beam in Eq. (7) and then inserting the resultant in Eq. (11), we can calculate the intensity distribution of the 1D-GSM beam in the focal region of the SLFZP. Figure 2 shows the intensity distributions calculated in the -plane in the focal region for different coherence widths of the 1D-GSM beam. In calculating, we supposed that the wavelength of the beam is and the intensity distribution across the beam on the plane just before the SLFZP is uniform. It means that the source width of the beam in this plane is big enough (). Also, the focal length of the SLFZP was considered to be . As the images show, the focal width and the depth of focus (DOF) extend as the coherence width decreases. These results open a door to the more detailed study of the DOF of the SLFZP under illumination of the partially coherent light beam. It is discussed in detail in the next subsection. The calculated intensity distributions of the 1D-GSM beam with the wavelength of in the focal region of the SLFZP with the focal length of in the plane for the various coherence widths of the beam, (a) δ = 0.15 mm, (b) δ = 0.12 mm, (c) δ = 0.08 mm, and (d) δ = 0.05 mm. Depth of focus According to Fig. 2, when the SLFZP is illuminated with a 1D-GSM beam in normal incidence, the main parameters of the focus, i.e. the focal width, , and the DOF40 extend as the coherence width, δ, of the 1D-GSM beam decreases. In order to show the variation of the DOF of the SLFZP by the coherence width of the 1D-GSM beam, we calculated the focal width and so the DOF of the SLFZP using where FWHM stands for the full width at half maximum of the calculated intensity in Eq. (11) at for the 1D-GSM beam in the focal region. Figure 3 illustrates the calculated DOF of the SLFZP versus the coherence width. In this figure the solid curve is a fitted exponential function. The DOF of the SLFZP with the focal length of versus the coherence width of the 1D-GSM beam with the wavelength of . The circs are the calculated DOF and the solid curve is the fitted exponential function. This study shows some significant results. At first, when the SLFZP is illuminated by a partially coherent 1D-GSM beam, its DOF decreases exponentially with the coherence width of the beam. In other words, one can achieve an extended DOF by using the spatially low coherent light beams. Second, it is possible to estimate the coherence width of the beam by measuring the DOF of the SLFZP. Experiment Procedure of experiment To prepare the SLFZP, we used the method of printing on transparent film41. We designed this zone plate with a focal length of 300 mm for an illuminating beam with a wavelength of 633 nm using MATLAB software. Then, print it on transparent film using a phonix2250 printer with a resolution of 3600dpi. For generating the 1D-GSM beam, we arranged a 2f-system using two cylindrical lenses under illumination of the parallel beam of a He–Ne laser. after passing through the cylindrical lens, we have 1D- light beam. Next, installed a rotating diffuser near the common focal plane of the lenses42–44. In order to change the spatial coherence width of the beam, we controlled the beam spot size on the diffuser. This work was done by changing the diffuser distance from the common focal plane of the lenses. According to Fig. 4, the Gaussian beam of a He–Ne laser with the wavelength of 633 nm is expanded using a beam expander to fit the dimensions of the provided SLFZP. Then, we generate a 1D-GSM beam by arranging the desired 2f-system composed of two cylindrical lenses with the focal lengths of and installing the rotating diffuser, mounted on an optical rail, at distance from the common focal plane of the cylindrical lenses. Therefore, by changing the distance , the size of the light spot covered by the diffuser is controlled and so a 1D-GSM beam with different coherent widths is generated. The prepared SLFZP is placed after the second lens. The cylindrical lenses were chosen so that their focal length is large enough so that the Gaussian beam that illuminates the zone plate has a relatively large width. After the SLFZP, a CCD camera ( pixels with pixel size) is placed on an optical rail until the camera moves easily around the focal plane of the SLFZP and records the intensity patterns. The experimental setup. BE, CL, RD, SLFZP, and CCD stand respectively for the beam expander, cylindrical lens, rotating diffuser, sinusoidal linear Fresnel zone plate, and camera. Results and discussion The 1D-GSM beams were created by placing the rotating diffuser at different distances from the common focal plane of cylindrical lenses. In each diffuser distance, the intensity patterns of the beam were recorded at the successive planes by the intervals of from the focal plane of the SLFZP until to a distance of using the CCD camera. According to the similar studies done in Ref. 37, at first we considered delta values from 0.2 to 0.5 mm. But later, to have better graphs and a view farther from the focal points, we decided to check the distances away from the focal points as well. The distances are very smaller than , therefore the distances investigated experimentally are in accordance with the approximation that the theoretical part requires. Figure 5 shows the typical recorded patterns on the successive planes by the intervals, , from the focal plane of the SLFZP for the four different diffuser distances. As the figure shows by moving away from the focal plan the wave-front expands and so the full width of the intensity patterns get wider. This confirms that it follows the Gaussian beam propagation mechanism. However, the remarkable result that can be extracted from this figure and is a manifestation of the purpose of the study is the broadening of the intensity patterns by increasing the distance of the diffuser from the focal plan of the collimating lens. In the following, by analyzing the obtained results, we will discuss the reason for it. Intensity Patterns of the 1D-GSM beam with the wavelength of 633 nm that have been recorded on the planes at successive intervals, , from the focal plane of the SLFZP for the diffuser distance, (a) (b) , (c) , and (d) . After recording the intensity patterns, we calculated the intensity profile on every pattern. Stacking these intensity profiles sequentially results in an intensity distribution in the plane. Figure 6 illustrates the typical intensity distributions of the 1D-GSM beam in -plane from the focal plane of the SLFZP until the distance from that for the four different diffuser distances. The stacked intensity distributions on the recorded patterns of the 1D-GSM beam on -plane from the focal plane of the SLFZP to the distance of 12 mm from that for the diffuser distances, (a) , (b) , (c) , and (d) . In order to show that the experimental results in Fig. 6 approve the theoretical results in Fig. 2, we draw the intensity profiles on the corresponding intensity patterns in Figs. 2 and 6. Figure 7 illustrates the corresponding intensity profiles which show the coincidence between the theoretical and experimental results. The Normalized intensity profiles plotted theoretically in blue and experimentally in red for the mm, For the corresponding points of Figs. 2 and 6. Next, we calculated the DOFs of the SLFZP for the 1D-GSM beams with the various coherence widths (for the difference diffuser distances) by evaluating the beam waists from the intensity profiles on the patterns at (focal plane). We have plotted the DOFs of the SLFZP versus the diffuser distance, in Fig. 8. The solid curve in this plot is an exponential function that has been fitted on the evaluated DOFs of the SLFZP. The theoretical depth of focus and measured DOFs of the SLFZP versus the distance of the rotating diffuser from the common focal plane of the cylindrical lenses, . Figure 8 shows that the DOF of the SLFZP increases with the diffuser distance as an exponential function. If we want to justify this behavior, we must consider the generation method of the spatially partial coherent beam in our experimental setup and the theoretical results in the previous section. As shown in Fig. 4, we generated the 1D-GSM beam using a rotating diffuser and controlled the coherence width of the beam by changing the distance of the diffuser from the common focal plane of the cylindrical lenses. It is easy to understand that by increasing the distance of the diffuser from the focal plane of the two lenses, the incident rays to the diffuser cover a larger cross-sectional area of the diffuser. In other words, the light rays experience more points of the rotating diffuser. As a result, the spatial correlation of the incident rays to different points due to the inhomogeneity of these points decreases relative to each other. We can even quantify this result. If we compare the fitting curves obtained for the depth of focus in the theory and experiment sections, we can achieve a mathematical relationship between the coherence width and the diffuser distance from the focal plane of the two lenses as follows It should be noted that this formula is valid in the case that the intensity distribution on the plane just before the SLFZP is uniform. In other words, the source width on this plane, , should be large enough. Conclusion Under the light of a spatially partial coherent Gaussian Schell model beam, the depth of focus of a linear sinusoidal Fresnel zone plate is examined. Our results demonstrate that when exposed to such a uniformly distributed beam of light, the zone plate’s depth of focus increases as spatial coherence decreases. We found a mathematical link between the coherence width and the effective experimental parameter in the coherence change, in addition to the experimental data validating the theoretical research. This article’s findings can be applied to improve optical trapping efficiency for tiny systems and biomedical applications under irradiation by incoherent light. Author contributions M. F. contributed to conducting experiments, writing, original draft preparation and software. R. A. contributed to conceptualization of this research, methodology and supervision. All authors have read and agreed to the published version of the manuscript. Data availability All data analyzed during the current study are available from the corresponding author on reasonable request. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement All data analyzed during the current study are available from the corresponding author on reasonable request.