Method and device for detecting boron and phosphorus contents in boron and phosphorus co-doped silicon-based thin film

Abstract

A method and device for detecting boron and phosphorus contents in boron and phosphorus co-doped silicon-based thin films, the method comprising: placing a wafer to be measured containing a boron and phosphorus co-doped silicon-based thin film in an etching solution for etching; measuring thicknesses of the boron and phosphorus co-doped silicon-based thin film before and after etching to obtain a thickness change; and determining a ratio of the boron and phosphorus contents in the boron and phosphorus co-doped silicon-based thin film based on the thickness change. The above scheme can reduce the detection cost and improve the convenience of detection of the boron and phosphorus contents in the boron and phosphorus co-doped silicon-based thin film.

Description

Technical Field

[0001] This invention relates to the field of semiconductor technology, and specifically to a method and apparatus for detecting the boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films. Background Technology

[0002] In the semiconductor field, boron-phosphorus co-doped silicon-based thin films (such as boron-phosphorus silicon glass BPSG) are an important dielectric material and are widely used in semiconductor integrated circuit manufacturing, mainly as metal front dielectric (PMD) layers, planarization layers and passivation layers.

[0003] The boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films directly affects the film's key physicochemical properties, including etching rate, reflow characteristics, hygroscopicity, film stress, and ion trapping ability. Therefore, in semiconductor manufacturing processes, accurately detecting the boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films is of great significance for process monitoring and ensuring device yield.

[0004] However, existing methods for detecting boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films are generally costly and lack convenience for on-site testing. Summary of the Invention

[0005] The problem this invention aims to solve is to reduce detection costs and improve the convenience of detection in the process of detecting boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films.

[0006] To address the aforementioned problems, this invention provides a method for detecting the boron and phosphorus content in a boron-phosphorus co-doped silicon-based thin film. The method includes: placing a wafer containing the boron-phosphorus co-doped silicon-based thin film into an etching solution for etching; measuring the thickness of the boron-phosphorus co-doped silicon-based thin film before and after etching to obtain the thickness change; and determining the ratio of phosphorus to boron content in the boron-phosphorus co-doped silicon-based thin film based on the thickness change.

[0007] In one possible embodiment, the method further includes: establishing a model of the correspondence between the etching rate and the ratio of boron to phosphorus content before etching; Determining the ratio of phosphorus to boron content in the boron-phosphorus co-doped silicon-based thin film based on the thickness change includes: determining the etching rate based on the thickness change, and inputting the determined etching rate into the correspondence model to obtain the ratio of phosphorus to boron content in the boron-phosphorus co-doped silicon-based thin film.

[0008] In one possible embodiment, the correspondence model is: Y=K V; Where Y represents the ratio of phosphorus to boron content in the boron-phosphorus co-doped silicon-based thin film; K is the proportionality coefficient; and V is the etching rate.

[0009] In one possible embodiment, the method further includes: when the ratio of phosphorus to boron content in the boron-phosphorus co-doped silicon-based thin film exceeds a preset ratio range, adjusting the boron or phosphorus content based on the etching rate and forming a new wafer to be tested according to the adjusted content.

[0010] In one possible embodiment, the correspondence model is determined by the following method: obtaining multiple sets of boron-phosphorus co-doped silicon-based thin film standard samples with known boron and phosphorus content ratios; placing each standard sample in the etching solution and etching under the same conditions, measuring the thickness of each standard sample before and after etching, and determining the etching rate; using the known boron content, known phosphorus content, and measured etching rate of each set of standard samples as calibration data, fitting and establishing the correspondence model.

[0011] In one possible embodiment, the etching is performed at a set temperature, which is 15–35°C.

[0012] In one possible embodiment, the wafer to be tested is a wafer that has undergone a reflow process.

[0013] In one possible embodiment, the etching solution is a buffered oxide etching solution.

[0014] In one possible embodiment, the boron-phosphorus co-doped silicon-based thin film is boron-phosphorus silicon glass.

[0015] This invention also provides a device for detecting the boron and phosphorus content in a boron-phosphorus co-doped silicon-based thin film. The device includes: an operation unit adapted to place a wafer containing the boron-phosphorus co-doped silicon-based thin film into an etching solution for etching; a measurement unit adapted to measure the thickness of the boron-phosphorus co-doped silicon-based thin film before and after etching to obtain the thickness change; and a determination unit adapted to determine the boron and phosphorus content in the boron-phosphorus co-doped silicon-based thin film based on the thickness change.

[0016] Compared with the prior art, the technical solution of the embodiments of the present invention has the following advantages: The present invention utilizes an etching solution to etch a silicon wafer containing a boron-phosphorus co-doped thin film. The thickness of the film before and after etching can be measured to obtain the thickness change. This thickness change can then be used to determine the phosphorus to boron content ratio in the film. The main equipment used in the entire testing process consists of an etching solution bath and a film thickness measuring instrument, both standard equipment on semiconductor production lines. This eliminates the need for expensive dedicated analytical equipment such as SIMS, XRF, or FTIR, significantly reducing equipment procurement and maintenance costs. Furthermore, the testing process is simple, requiring only four steps: thickness measurement, etching, thickness measurement again, and calculation. The sampling and testing procedures are convenient, allowing for rapid acquisition of etching rate data to infer the change in the boron-phosphorus content ratio. This makes it suitable for high-frequency in-line monitoring on production lines. Compared to SIMS and XRF methods, which require removing the wafer from the production line for testing with dedicated equipment, this invention significantly shortens the testing cycle and increases the monitoring frequency. Furthermore, the etching rate is highly sensitive to changes in boron and phosphorus content; even small fluctuations in the doping ratio can cause significant rate differences, making it easier to detect upstream process drift and achieve early warning. Etching rate monitoring is directly related to the selectivity and uniformity of the wet cleaning / etching process. Monitoring the etching rate essentially means monitoring the actual process behavior the product is undergoing, which is more instructive for engineering than simply measuring component content. When an abnormal etching rate occurs, it can help pinpoint problems with the flow rates of source gases such as silane, phosphine, and borane in the chemical vapor deposition (CVD) process, and optimize the process window by adjusting boron first, then phosphorus, providing a basis for adjusting the boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films during CVD. Attached Figure Description

[0017] Figure 1 This is a flowchart of a method for detecting the boron and phosphorus content in a boron-phosphorus co-doped silicon-based thin film according to an embodiment of the present invention; Figures 2 to 7 This is a schematic diagram of the conventional process for a semiconductor structure containing a boron-phosphorus co-doped silicon-based thin film; Figure 8 This is a schematic cross-sectional view of a semiconductor structure containing a boron-phosphorus co-doped silicon-based thin film before etching, according to an embodiment of the present invention. Figure 9 This is a schematic diagram of the cross-sectional structure of a semiconductor structure containing a boron-phosphorus co-doped silicon-based thin film after etching, according to an embodiment of the present invention. Figure 10 This is a flowchart illustrating the establishment of a correspondence model in an embodiment of the present invention; Figure 11 This is a schematic diagram of the calibration curve of a correspondence model Y=K×V in an embodiment of the present invention; Figure 12This is a flowchart of another method for detecting boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films in an embodiment of the present invention; Figure 13 This is a schematic diagram of the structure of a device for detecting the boron and phosphorus content in a boron-phosphorus co-doped silicon-based thin film according to an embodiment of the present invention; Figure 14 This is a schematic diagram of the etching rate and the corresponding phosphorus and boron content ratio of a standard sample according to an embodiment of the present invention. Detailed Implementation

[0018] The boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films directly affects the film's key physicochemical properties, including etching rate, reflow characteristics, hygroscopicity, film stress, and ion trapping ability. For example, too low a boron content will lead to increased film reflow temperature and poorer planarization; too high a phosphorus content will increase film hygroscopicity and may cause phosphorus to diffuse into the active region of the device, resulting in decreased device reliability.

[0019] Currently, the main methods used in the industry for monitoring the boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films include secondary ion mass spectrometry (SIMS), X-ray fluorescence spectrometry (XRF), and Fourier transform infrared spectroscopy (FTIR). Among them: Secondary ion mass spectrometry (DIMS) uses a primary ion beam to sputter the sample surface, collects the generated secondary ions, and analyzes them according to their mass-to-charge ratio to obtain information on the depth distribution of elements and isotopes. This method has the advantage of high-precision absolute quantification, but the equipment used is expensive, resulting in high detection costs. Furthermore, it requires removing the wafer from the production line and placing it on specific equipment for detection, making it unsuitable for high-frequency detection on the production line.

[0020] X-ray fluorescence spectrometry (XRF) is used for quantitative analysis of elements in thin films by exciting them with X-rays to produce characteristic fluorescence. However, boron has a low atomic number (Z=5), resulting in low fluorescence yield and low X-ray energy. It is also easily absorbed by air and the thin film itself, requiring a dedicated multilayer analyzer and extended counting time. This high equipment requirement leads to high detection costs. Furthermore, it also requires removing the wafer from the production line to the dedicated equipment for detection, making it unsuitable for high-frequency detection on the production line.

[0021] Fourier transform infrared spectroscopy indirectly and quantitatively estimates the boron and phosphorus content by measuring the absorption intensity of chemical bond vibrations in boron-oxygen and phosphorus-oxygen bonds in thin films. This method is an indirect quantitative approach, relying on calibration curves and requiring the prior establishment of reliable boron and phosphorus concentration calibration models. Without standards, its absolute accuracy is insufficient. Changes in annealing conditions, densification, and stress states alter bond angles and bond strengths, leading to variations in absorption coefficients. This causes readings of the same concentration to drift under different process conditions, necessitating the establishment of separate calibration libraries for different processes. This results in poor process adaptability and high maintenance costs for the calibration libraries.

[0022] The three methods mentioned above each have their own technical characteristics, but they share common shortcomings: high overall testing costs and insufficient convenience for on-site testing. Therefore, the industry needs a method for detecting boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films that is easy to operate, has low equipment costs, and is suitable for rapid on-site testing.

[0023] To address the aforementioned problems, this invention provides a method for detecting the boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films. This method utilizes the high sensitivity of the etching rate of boron-phosphorus co-doped silicon-based thin films in BOE etching solution to changes in boron and phosphorus doping content. By measuring the change in film thickness before and after etching, the ratio of phosphorus to boron content is indirectly obtained. This method achieves boron and phosphorus content monitoring that is simple to operate, has low equipment cost, and is suitable for rapid on-site detection.

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0025] Reference Figure 1 This invention provides a method for detecting the boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films, the method comprising the following steps: Step 11: Place the wafer to be tested containing a boron-phosphorus co-doped silicon-based thin film in an etching solution for etching; Step 12: Measure the thickness of the boron-phosphorus co-doped silicon-based thin film before and after etching to obtain the thickness change. Step 13: Determine the ratio of phosphorus to boron content in the boron-phosphorus co-doped silicon-based thin film based on the thickness change.

[0026] Boron-phosphorus co-doped silicon-based thin films are dielectric films formed by simultaneously introducing boron and phosphorus dopants into a silicon dioxide (SiO2) network. A typical example is boron-phosphorus silicon glass (BPSG). In semiconductor integrated circuit manufacturing, BPSG is widely used as a pre-metal dielectric (PMD) layer, planarization layer, and passivation layer. The boron and phosphorus content directly affects key properties of the film, such as etching rate, reflow characteristics, hygroscopicity, film stress, and ion trapping ability. Therefore, accurately monitoring the boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films is crucial for process control and ensuring device yield.

[0027] The core technical principle of this invention lies in the fact that the etching rate of boron-phosphorus co-doped silicon-based thin films in fluorine-containing etching solutions (such as buffered oxide etching solutions, BOE) is highly sensitive to the doping content of boron and phosphorus, and their effects on the etching rate are opposite. Specifically: The introduction of boron will cause some degree of breakage in the silicon-oxygen network, but when boron is introduced in a tricoordinate form, it will increase the degree of network crosslinking, and boron will react with fluoride ions (F... -Boron exhibits a certain tendency to complex, resulting in a relatively dense film structure. Therefore, increasing the boron content typically reduces the etching rate of boron-phosphorus co-doped silicon-based films in BOE solution, i.e., boron has a "decelerating effect" on the etching rate.

[0028] The introduction of phosphorus disrupts the network structure of silicon dioxide, increases the number of unbridged oxygen bonds and the porosity of the film, making it easier for hydrofluoric acid and fluoride ions to penetrate into the film. Therefore, the increase in phosphorus content significantly increases the etching rate of boron-phosphorus co-doped silicon-based films in BOE solution, i.e., phosphorus has an "accelerating effect" on the etching rate.

[0029] The final measured etching rate of the boron-phosphorus co-doped silicon film to BOE solution is the net result of the superposition of the "decelerating effect" of boron and the "accelerating effect" of phosphorus. Since the two effects act in opposite directions, the etching rate actually reflects the coupled result of the combined effects of boron and phosphorus. When the ratio of phosphorus to boron content (i.e., the P / B ratio) in the boron-phosphorus co-doped silicon film is at a specific level, under fixed process conditions (fixed temperature, fixed etching solution concentration, etc.), the etching rate of the boron-phosphorus co-doped silicon film to this film is a fixed value. In other words, under fixed process conditions, there is a definite correspondence between the BOE etching rate of the boron-phosphorus co-doped silicon film and the P / B ratio.

[0030] Based on the above principle, this invention measures the thickness change of a boron-phosphorus co-doped silicon-based thin film before and after etching in an etching solution, calculates the etching rate, and then uses a pre-established model of the correspondence between the etching rate and the ratio of boron to phosphorus content to indirectly obtain the ratio of phosphorus to boron content in the boron-phosphorus co-doped silicon-based thin film, thereby achieving rapid monitoring of boron and phosphorus content.

[0031] It should be noted that, since the BOE etching rate is the coupled result of the combined effects of boron and phosphorus, this invention cannot quantify the absolute content of boron and phosphorus separately and precisely. However, the purpose of this invention is not to obtain the absolute content of boron and phosphorus, but to monitor whether the ratio of phosphorus to boron content is within the range required by the product, thereby determining whether the upstream CVD deposition process is stable. This already meets the actual needs of production line process monitoring.

[0032] Based on the above principles, steps 11 to 13 will be described in detail: Regarding step 11, the wafer to be tested, containing a boron-phosphorus co-doped silicon-based thin film, is placed in an etching solution for etching. The role of the etching solution is to selectively etch the boron-phosphorus co-doped silicon-based thin film, thereby reducing its thickness. The etching solution should be selected to meet the following requirements: it should be able to etch the boron-phosphorus co-doped silicon-based thin film at a controllable rate, and the etching rate should be sufficiently sensitive to changes in the boron and phosphorus doping content.

[0033] In a preferred embodiment of the present invention, the etching solution is a buffered oxide etchant (BOE). BOE is a mixed solution of hydrofluoric acid (HF) and ammonium fluoride (NH4F), with commonly used volume ratios including NH4F:HF = 6:1, 10:1, 20:1, etc. NH4F acts as a buffer in the solution, maintaining a stable HF concentration, making the etching rate more controllable and stable than that of a pure HF solution. BOE is a conventional wet etching solution used in semiconductor production lines, which are typically equipped with BOE etching tanks, thus eliminating the need for additional dedicated equipment.

[0034] During the etching process, it is necessary to maintain constant etching conditions, including the temperature, concentration, and stirring conditions of the etching solution. The etching temperature is typically controlled within the range of 15–35°C, and temperature fluctuations should be minimized to ensure the repeatability of the etching rate. The etching time T is a pre-set fixed value, and the etching time T should not completely etch the boron-phosphorus co-doped silicon-based film; that is, the film should retain a certain thickness after etching to facilitate post-etching thickness measurement. Once the etching time T is determined, it should remain unchanged in all subsequent testing processes to ensure comparability between different batches.

[0035] In one embodiment of the present invention, the wafer to be tested is immersed in BOE etching solution and kept there for a set time T before being removed. After removal, the etched wafer is cleaned and dried to remove residual etching solution and reaction products from the surface. Cleaning is typically done with deionized water, and drying can be done by nitrogen blowing or rotary drying. The purpose of the cleaning and drying steps is to ensure the accuracy of film thickness measurement and to avoid interference from residues with the measurement results.

[0036] In a preferred embodiment of the present invention, the wafer under test is a wafer processed by a reflow process. The reflow process is an important step in the fabrication of boron-phosphorus co-doped silicon-based thin films, typically performed after film deposition. The film is annealed at a high temperature of 800–950°C in a nitrogen (N2) or water-containing atmosphere to soften and flow the glassy film, achieving surface planarization. The reflow process alters the film's density, stress state, and bonding structure, thus affecting its etching behavior. Therefore, to ensure the accuracy and consistency of the test results, the wafer under test should be in the same reflow process state as during calibration. In this embodiment, the wafer under test is a wafer prepared under fixed CVD deposition and reflow conditions, processed by a reflow process.

[0037] The conventional process flow for preparing boron-phosphorus co-doped silicon-based thin films is as follows: First, refer to Figure 2 A semiconductor substrate 21 (such as a silicon wafer) is prepared, on which device structures such as transistors have been fabricated. In some embodiments, an oxide layer 22 may be formed on the substrate 21.

[0038] Then, refer to Figure 3 A boron-phosphorus co-doped silicon-based thin film was deposited on a substrate using chemical vapor deposition (CVD) 23. The CVD deposition can employ sub-atmospheric pressure chemical vapor deposition (SACVD) or plasma-enhanced chemical vapor deposition (PECVD) routes. Taking the SACVD route as an example, tetraethoxysilane (TEOS) is used as the silicon source, ozone (O3) as the oxidant, triethylboron (TEB) as the boron source, and triethyl phosphate (TEPO) as the phosphorus source during the deposition process. The deposition temperature is typically 400–500℃.

[0039] Taking the PECVD route as an example, silane (SiH4) is used as the silicon source, nitrous oxide (N2O) as the oxidant, diborane (B2H6) as the boron source, and phosphine (PH3) as the phosphorus source. The deposition temperature is typically 300–500℃. After deposition, the typical doping amounts of boron and phosphorus are approximately 2–4 wt% for boron and 4–9 wt% for phosphorus, respectively.

[0040] Subsequently, referring to Figure 4 A reflow annealing process is performed to soften and flow the thin film, achieving surface planarization. After reflow annealing, a flat boron-phosphorus co-doped silicon-based thin film 24 is formed on the surface of substrate 21.

[0041] Reference Figure 5 A photoresist layer 25 is formed on a flat boron-phosphorus co-doped silicon-based thin film 24. (Refer to...) Figure 6 Using photoresist layer 25 as a mask, the flat boron-phosphorus co-doped silicon-based thin film 24 and substrate 21 are etched. (Refer to...) Figure 7 Remove the photoresist layer 25.

[0042] The detection method of this invention is applicable to boron-phosphorus co-doped silicon-based thin films after reflow annealing in the above-described process. On the production line, samples are taken from product wafers of boron-phosphorus co-doped silicon-based thin films grown in CVD equipment. Wafers prepared under fixed CVD deposition and reflow conditions can be used as the wafers to be tested.

[0043] In a specific embodiment of the present invention, the boron-phosphorus co-doped silicon-based thin film is boron-phosphorus silicon glass (BPSG). BPSG is currently the most commonly used boron-phosphorus co-doped silicon-based thin film in semiconductor manufacturing, and the method of the present invention is particularly suitable for monitoring the boron to phosphorus content ratio in BPSG. However, the scope of protection of the present invention is not limited to BPSG; it is also applicable to other silicon-based thin films containing both boron and phosphorus dopants.

[0044] Regarding step 12, the thickness of the boron-phosphorus co-doped silicon-based thin film before and after etching is measured to obtain the thickness change.

[0045] Specifically, refer to Figure 8Before etching, the thickness of the boron-phosphorus co-doped silicon-based thin film 33 on the wafer under test was measured using a film thickness measuring instrument to obtain the film thickness value D before etching. F To improve the representativeness of the measurement results, measurements should be taken at multiple locations on the wafer surface, with as many test points as possible. The average value of all test points should be taken as the pre-etch film thickness D. F The wafer under test also includes a substrate 31 and an oxide layer 32 formed on the surface of the substrate 31. A boron-phosphorus co-doped silicon-based thin film 33 is located on the surface of the oxide layer 32.

[0046] After etching is completed, refer to Figure 9 The boron-phosphorus co-doped silicon-based thin film 33 is etched to a certain thickness, forming the etched boron-phosphorus co-doped silicon-based thin film 34. At this point, the same film thickness measurement instrument can be used to measure the film thickness at the same location on the wafer under test as before etching, obtaining the etched film thickness value D. B The number of test points after etching is the same as before etching, and the average value of all test points is taken as the film thickness D after etching. B .

[0047] The difference in film thickness before and after etching is the thickness change ΔD = D. F -D B Because the etching process thins the film, D F >D B Therefore, the thickness change ΔD is a positive value.

[0048] In one embodiment of the present invention, the thickness of the boron-phosphorus co-doped silicon-based thin film is measured using at least one of an ellipsometry, a reflectance spectrometer, or a profilometer.

[0049] Ellipsometry calculates film thickness and refractive index by measuring the change in polarization state of polarized light after reflection from a thin film surface and using an optical model for fitting. It offers advantages such as non-contact, non-destructive operation, and high precision, making it suitable for accurate film thickness measurement. Reflectance spectrometers calculate film thickness by measuring the reflection spectrum of the thin film surface and using spectral interference fringes; they also offer non-contact and rapid measurement. Profilometers directly measure surface height differences by scanning the thin film surface with a mechanical probe, and can be used for direct film thickness measurement. All of these instruments are standard equipment on semiconductor production lines and require no additional procurement.

[0050] When selecting a film thickness measuring instrument, ensure that the instrument is calibrated and that the same instrument and measurement parameters are used throughout the entire testing process to eliminate systematic errors between instruments. The film thickness measuring instrument should maintain consistent measurement conditions before and after etching, including measurement location, measurement parameters, and calibration status.

[0051] Regarding step 13, the ratio of phosphorus to boron content in the boron-phosphorus co-doped silicon-based thin film is determined based on the thickness change.

[0052] Specifically, the etching rate V is first calculated based on the thickness change. The etching rate V is the ratio of the thickness change ΔD to the etching time T, i.e.: V = ΔD / T = (D F - D B ) / T (1) Among them, D F D represents the film thickness before etching. B The value represents the film thickness after etching, and T represents the etching time.

[0053] Then, the calculated etching rate V is substituted into the pre-established correspondence model to obtain the ratio Y of phosphorus and boron content in the boron-phosphorus co-doped silicon-based thin film.

[0054] In specific implementation, such as Figure 10 As shown, the model relating etching rate to the ratio of boron to phosphorus content in this invention was established using the following calibration method: Step 101: Obtain standard sample.

[0055] Specifically, product wafers of boron-phosphorus co-doped silicon-based thin films grown on the production line's CVD equipment can be used to prepare multiple sets of standard samples under fixed CVD deposition and reflow conditions. The number of samples, n, should be sufficiently large to ensure the statistical reliability of the calibration results; in the embodiments of this invention, n ≥ 20. The standard samples are labeled with numbers 1, 2, 3…n.

[0056] Step 102: Measure the known ratio of boron to phosphorus content.

[0057] Specifically, a calibrated X-ray fluorescence spectrometer (XRF) can be used to measure the boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films on each standard sample, and the ratios of phosphorus and boron content in each standard sample, Y1, Y2, Y3...Y, can be calculated. n The purpose of using XRF here is to obtain a reference value for the known boron-to-phosphorus ratio for calibration. XRF is used as a benchmark measurement tool during the calibration stage and is no longer needed in daily production line monitoring.

[0058] Step 103: Measure the etching rate.

[0059] Specifically, the etching rate of boron-phosphorus co-doped silicon-based thin films of each standard sample can be tested using the following methods: First, a calibrated film thickness measurement instrument was selected to test the pre-etch thickness D of the boron-phosphorus co-doped silicon-based thin film on a standard sample. F (The more test points the better, and the mean is defined as D) FThen, the standard sample is immersed in a BOE etching solution under fixed conditions (fixed temperature, fixed concentration, etc.) for etching time T. The time T is chosen such that it does not completely etch the boron-phosphorus co-doped silicon film, and the value of T remains constant in subsequent processes. After etching, the post-etching film thickness D of the boron-phosphorus co-doped silicon film at the same location on the standard sample is measured. B (with D) F If the number of test points is the same, the mean is defined as D. B ).

[0060] Combining formula (1), the etching rates V1, V2, V3...V of each standard sample can be obtained. n .

[0061] Step 104: Establish a correspondence model.

[0062] Specifically, the ratios of boron to phosphorus content in each standard sample, Y1, Y2, Y3...Y, can be calculated using XRF measurements. n The etching rates V1, V2, V3...V of each standard sample n Based on the data and the corresponding relationship model Y=K×V, the proportionality coefficients K1, K2, K3...K corresponding to each standard sample are calculated respectively. n The average value is taken as the proportionality coefficient K.

[0063] Therefore, a corresponding relationship model is established: Y = K × V = K × (D F - D B ) / T (2) Where Y is the ratio of phosphorus to boron content in the boron-phosphorus co-doped silicon-based thin film, K is the proportionality coefficient determined through calibration, and D... F D represents the film thickness before etching. B The value represents the film thickness after etching, and T represents the etching time.

[0064] This model establishes a linear relationship between the etching rate and the ratio of boron to phosphorus content. Under fixed process conditions (fixed etching solution type, concentration, temperature, etching time, etc.), the etching rate V can be calculated by measuring the film thickness DF and DB of the wafer before and after etching. Then, the ratio of phosphorus to boron content Y can be obtained through the model Y=K×V.

[0065] like Figure 11 As shown, during the calibration process, the etching rate V is plotted on the x-axis, and the boron to phosphorus content ratio Y is plotted on the y-axis. This plots n sets of calibration data (V1, Y1), (V2, Y2), ..., (V...) are then used. n Y nPlotting this on a coordinate system, the scaling factor K can be obtained through linear fitting. In practical applications, due to process fluctuations and measurement errors, the calibration data may not fall exactly on a straight line, but by taking K1 to K... n The average value or linear fitting using the least squares method can be used to obtain the statistically stable proportional coefficient K.

[0066] It should be noted that the correspondence model is established under fixed process conditions, meaning that the type, concentration, temperature, etching time, and CVD deposition and reflow conditions of the wafer under test must all remain consistent. If the process conditions change (such as changing the etching solution ratio, adjusting the etching temperature, or altering the CVD deposition parameters), the model should be recalibrated to update the correspondence model.

[0067] In a preferred embodiment, refer to Figure 12 The method may further include: step 14, when the ratio of phosphorus to boron content in the boron-phosphorus co-doped silicon-based thin film exceeds a preset ratio range, adjusting the boron or phosphorus content based on the etching rate and forming a new wafer to be tested according to the adjusted content.

[0068] Specifically, the preset ratio range is a safe range for the boron and phosphorus content ratio set according to product specifications. For example, for a BPSG film of a certain product, the product specifications require a boron content of 2.5–5.0 wt% and a phosphorus content of 4.0–6.0 wt%. The preset range of the boron and phosphorus content ratio Y can be determined based on product specifications and process experience.

[0069] When the boron to phosphorus content ratio Y exceeds the preset range, it indicates that the flow rate of the boron or phosphorus source in the upstream CVD deposition process may be deviated, requiring process adjustment. The specific adjustment strategy is as follows: First, determine the source of the deviation based on the direction of change in the etching rate V. If the etching rate V is high, it indicates that the "accelerating effect" of phosphorus is dominant, which may indicate a problem of low boron content or high phosphorus content. If the etching rate V is low, it indicates that the "decelerating effect" of boron is dominant, which may indicate a problem of high boron content or low phosphorus content.

[0070] Then, the process window is optimized according to the order of "adjusting boron first, then phosphorus." That is, the flow rate of the boron source gas (such as TEB or B2H6) during CVD deposition is adjusted first, and the etching rate is observed to return to the normal range after adjustment. If adjusting the boron source alone still cannot bring the ratio Y back to the preset range, then the flow rate of the phosphorus source gas (such as TEPO or PH3) is adjusted. The advantage of this adjustment strategy is that boron adjustment has a relatively large impact on the film reflow characteristics and stress; adjusting boron first allows for fine-tuning while ensuring the physical properties of the film. Directly adjusting phosphorus may cause other problems such as changes in hygroscopicity.

[0071] After the adjustment is completed, a new wafer to be tested is formed according to the adjusted CVD deposition process parameters, and the detection method of the present invention is executed again for verification until the boron and phosphorus content ratio Y returns to the preset range.

[0072] Through the aforementioned process feedback and adjustment mechanism, this invention can not only detect the boron to phosphorus content ratio in boron-phosphorus co-doped silicon-based thin films, but also provide quantitative basis for CVD process adjustment, achieving closed-loop process control. When an abnormal etching rate occurs, the flow rate of source gases such as silane, phosphine, and borane in the CVD deposition process can be traced back to quickly identify the cause of the process abnormality and shorten the process debugging cycle.

[0073] Reference Figure 13 This invention also provides a device 130 for detecting the boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films. The detection device 130 may include: an operation unit 131, a measurement unit 132, and a determination unit 133. Wherein: The operation unit 131 is adapted to place the wafer to be tested, which contains a boron-phosphorus co-doped silicon-based thin film, into an etching solution for etching.

[0074] The measuring unit 132 is adapted to measure the thickness of the boron-phosphorus co-doped silicon-based thin film before and after etching, and to obtain the thickness change.

[0075] The determining unit 133 is adapted to determine the ratio of phosphorus to boron content in a boron-phosphorus co-doped silicon-based thin film based on the thickness change.

[0076] Specifically, the operation unit 131 may include an etching solution tank (such as a BOE tank), a wafer clamping mechanism, a temperature control module, and a timing module. The etching solution tank is used to hold the etching solution, the wafer clamping mechanism is used to fix and immerse or remove the wafer to be tested, the temperature control module is used to maintain a constant etching solution temperature, and the timing module is used to control the etching time T. In an actual production line, the operation unit can reuse the existing BOE wet etching equipment on the production line.

[0077] The measurement unit 132 can specifically be an ellipsometry, reflectance spectrometer, or profilometer, etc., for measuring film thickness. The measurement unit 132 should have multi-point measurement capabilities, enabling it to measure film thickness at multiple locations on the wafer surface and calculate the average value. The measurement unit 132 should use the same measurement parameters and calibration status before and after etching.

[0078] The determination unit 133 may specifically include a data processing module, internally storing a pre-established correspondence model. The determination unit receives the pre-etching film thickness value D output by the measurement unit. F and the film thickness D after etching B Calculate the thickness change ΔD=D F -D B The etching rate V = ΔD / T is used, and then the etching rate V is substituted into the corresponding relationship model to calculate the ratio Y of phosphorus and boron content. The determining unit may also include a display module and an alarm module, which will issue an alarm when the ratio Y exceeds a preset range.

[0079] In one embodiment of the present invention, the determining unit 133 is also adapted to output process adjustment suggestions when the ratio of phosphorus and boron content exceeds a preset ratio range, including the suggested adjustment of the direction and amplitude of the boron or phosphorus source flow rate, to assist the operator in process debugging.

[0080] The following uses borosilicate glass (BPSG) as a specific example to illustrate the complete operation process of the detection method of the present invention.

[0081] Example 1: Establishment of a Correspondence Model

[0082] Seven standard samples, numbered 1# to 7#, were prepared from silicon wafers containing BPSG grown on a CVD device on the production line. These samples were obtained under fixed CVD deposition conditions (TEOS / O3 / TEB / TEPO system, deposition temperature 450℃, deposition pressure 8 Torr) and fixed reflow conditions (900℃, N2 atmosphere, 30 minutes).

[0083] The boron and phosphorus contents in BPSG films on 20 standard samples were measured using a calibrated XRF instrument (wavelength dispersive X-ray fluorescence spectrometer), and the phosphorus to boron content ratios Y1 to Y7 of each sample were calculated.

[0084] A calibrated ellipsometry was selected as the film thickness measurement instrument. For each standard sample, the BPSG film thickness before etching was measured to obtain the pre-etch film thickness value D for each standard sample. F .

[0085] The BOE etching solution (NH4F:HF = 6:1, temperature 25℃) was poured into the etching tank. The standard sample was immersed in the BOE solution for 60 seconds (T = 60 seconds) to ensure that the BPSG film was not completely etched after etching. After removing the sample, it was rinsed with deionized water and dried with nitrogen.

[0086] Using the same ellipsometry, the BPSG film thickness after etching of standard samples was measured, and the film thickness D of each standard sample after etching was obtained. B .

[0087] Calculate the etching rate V for each standard sample. i =(D F -D B ) / T (i=1~7). The etching rate V value and Y value of each standard sample are as follows: Figure 14 As shown. Subsequently, K can be calculated based on the model Y=K×V. i =Y i / V i (i=1~7) and take the average value K=(K1+K2+…+K7) / 7, which gives the proportionality coefficient K. Based on this, establish the corresponding relationship model Y=K×V=K×(D F -D B ) / T.

[0088] Example 2: Daily Production Line Monitoring

[0089] Once the correspondence model is established, it can be used for rapid monitoring of the boron to phosphorus content ratio in BPSG films on the daily production line. The specific operation procedure is as follows: (1) Sampling: Take one wafer from the BPSG product wafer grown by the CVD equipment on the production line as the wafer to be tested. The wafer is prepared under the same CVD deposition and reflow conditions as the calibration, and has been processed by the reflow process.

[0090] (2) Thickness measurement before etching: Using a calibrated ellipsometry, the thickness of the BPSG film was measured at 81 test points on the surface of the wafer to be tested, and the average value was taken as the film thickness value D before etching. F .

[0091] (3) Etching: Immerse the wafer to be tested in BOE etching solution (NH4F:HF=6:1, temperature 25℃) for 60 seconds (T=60 seconds, the same as the calibration time). After removal, rinse with deionized water and dry with nitrogen.

[0092] (4) Thickness measurement after etching: Using the same ellipsometry, the thickness of the BPSG film after etching was measured at the same 81 test points, and the average value was taken as the thickness value DB after etching.

[0093] (5) Calculate the etching rate: V = (D F-D B ) / T.

[0094] (6) Calculate the boron-to-phosphorus ratio: Substitute V into the corresponding relationship model Y=K×V to obtain the ratio Y of phosphorus and boron content.

[0095] (7) Judgment and feedback: Compare Y with the preset ratio range. If Y is within the preset range, the CVD deposition process is judged to be normal and production continues; if Y exceeds the preset range, the flow rate of boron or phosphorus source gas in the CVD process is adjusted according to the strategy of "adjusting boron first and then phosphorus", and a new wafer to be tested is formed according to the adjusted process parameters. Steps (1) to (7) are re-executed for verification.

[0096] The entire inspection process only requires four steps: "thickness measurement - etching - thickness measurement - calculation". It can be completed using the existing BOE etching tank and ellipsometry on the production line, without the need for expensive specialized equipment such as SIMS, XRF or FTIR. The inspection time for a single test can be controlled within 30 minutes, meeting the needs of high-frequency online monitoring on the production line.

[0097] Compared with the SIMS method, the present invention has the following advantages: SIMS is a destructive test, and the sample cannot be used after the test. Moreover, the equipment is extremely expensive and the test cycle is long. The present invention is a non-destructive or micro-destructive test. The wafer can still be used after etching if the film thickness allows. In addition, it uses conventional production line equipment, and the detection speed is fast and the cost is low.

[0098] Compared with the XRF method, the present invention has the following advantages: XRF has low detection sensitivity for boron (Z=5), requires a dedicated multilayer analyzer and long counting time, and has high equipment configuration and maintenance costs; the present invention does not directly measure the elemental fluorescence signal, but indirectly reflects the boron-phosphorus content ratio through the etching rate, thus avoiding the problem of difficult boron detection, and the equipment cost is much lower than that of XRF.

[0099] Compared with the FTIR method, this invention has the following advantages: FTIR is an indirect quantitative method that relies on calibration curves, and changes in annealing conditions, compaction, and stress state can cause the absorption coefficient to drift. It requires the establishment of calibration libraries for different processes, resulting in poor process adaptability. Although this invention also requires calibration when establishing the model, once the model is established, only the film thickness change needs to be measured in daily monitoring. The operation process is simple, and the etching rate directly reflects the behavior of the film in the actual wet process, which is more meaningful for engineering guidance than the chemical bond absorption information of FTIR.

[0100] In summary, this invention utilizes the high sensitivity of the etching rate of boron-phosphorus co-doped silicon-based thin films in BOE etching solution to changes in boron and phosphorus doping content. By measuring the change in film thickness before and after etching, the ratio of phosphorus to boron content is indirectly obtained, achieving boron-phosphorus content monitoring that is simple to operate, low in equipment cost, and suitable for rapid on-site detection. This invention is particularly suitable as a high-frequency online monitoring method for CVD deposition of BPSG thin films on semiconductor production lines, enabling timely detection of process drift and providing quantitative basis for process adjustments.

[0101] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. A method for detecting the boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films, characterized in that, include: The wafer to be tested, containing a boron-phosphorus co-doped silicon-based thin film, is placed in an etching solution for etching. The thickness of the boron-phosphorus co-doped silicon-based thin film before and after etching was measured to obtain the thickness change. The ratio of phosphorus to boron content in the boron-phosphorus co-doped silicon-based thin film is determined based on the thickness change.

2. The method for detecting boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films as described in claim 1, characterized in that, Also includes: Before etching, a model is established to show the relationship between the etching rate and the ratio of boron to phosphorus content. Determining the ratio of phosphorus to boron content in the boron-phosphorus co-doped silicon-based thin film based on the thickness change includes: The etching rate is determined based on the thickness change, and the determined etching rate is input into the corresponding relationship model to obtain the ratio of phosphorus and boron content in the boron-phosphorus co-doped silicon-based thin film.

3. The method for detecting boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films as described in claim 2, characterized in that, The correspondence model is as follows: Y=K V; Where Y represents the ratio of phosphorus to boron content in the boron-phosphorus co-doped silicon-based thin film; K is the proportionality coefficient; and V is the etching rate.

4. The method for detecting boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films as described in claim 3, characterized in that, Also includes: When the ratio of phosphorus to boron content in the boron-phosphorus co-doped silicon-based thin film exceeds a preset ratio range, the boron or phosphorus content is adjusted based on the etching rate, and a new wafer to be tested is formed according to the adjusted content.

5. The method for detecting boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films as described in claim 2, characterized in that, The correspondence model was determined using the following method: Multiple sets of standard boron-phosphorus co-doped silicon-based thin films with known boron-phosphorus content ratios were obtained. Each standard sample was placed in the etching solution and etched under the same conditions. The thickness of each standard sample before and after etching was measured and the etching rate was determined. Using the known boron content, known phosphorus content, and measured etching rate of each set of standard samples as calibration data, the corresponding relationship model was established by fitting.

6. The method for detecting boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films as described in any one of claims 1 to 5, characterized in that, The etching is performed at a set temperature, which is 15–35°C.

7. The method for detecting boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films as described in any one of claims 1 to 5, characterized in that, The wafer to be tested is a wafer that has undergone a reflow process.

8. The method for detecting boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films as described in any one of claims 1 to 5, characterized in that, The etching solution is a buffered oxide etching solution.

9. The method for detecting boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films as described in any one of claims 1 to 5, characterized in that, The boron-phosphorus co-doped silicon-based thin film is boron-phosphorus-silicon glass.

10. A device for detecting the boron and phosphorus content in boron-phosphorus co-doped silicon-based thin films, characterized in that, include: An operating unit is adapted to place a wafer to be tested containing a boron-phosphorus co-doped silicon-based thin film in an etching solution for etching. The measurement unit is adapted to measure the thickness of the boron-phosphorus co-doped silicon-based thin film before and after etching, and to obtain the thickness change. The determining unit is adapted to determine the boron and phosphorus content in the boron-phosphorus co-doped silicon-based thin film based on the thickness change.

Images