HK1174670B - Electron radiation monitoring system to prevent gold spitting and resist cross-linking during evaporation - Google Patents

Description

Electron radiation monitoring system to prevent gold sputtering and resist cross-linking during evaporation

RELATED APPLICATIONS

The present application claims priority under 35U.S. C. § 119 (e) from U.S. provisional application No.61/303,040 entitled "electric vehicle monitoring SYSTEM TO present GOLD SPITTING AND resttcross LINKING locking event", filed 2/10, 2010, the entire contents of which are incorporated herein by reference.

Technical Field

The present invention relates generally to metal deposition systems and, more particularly, to systems and methods for detecting and/or correcting conditions caused by impurities in metal evaporation sources employed in electron beam metal evaporation/deposition.

Background

To form microchips for electronic devices, various steps in processing semiconductor wafers involve depositing one or more layers of metal on the semiconductor wafer. These metal films are used, for example, to form metal contacts or conductive vias. Metal films are typically deposited on semiconductor wafers by employing Chemical Vapor Deposition (CVD) systems or Physical Vapor Deposition (PVD) systems. PVD systems are generally divided into sputtering systems and evaporation systems.

In a sputtering system, an energetic ion beam, e.g., argon ions, is directed at a metal target in a vacuum chamber. The energetic ions impact the metal atoms to be released from the target. The released metal atoms propagate through the vacuum chamber and are deposited on one or more wafers also present in the vacuum chamber.

In an evaporation system (also referred to herein as an evaporation/deposition system), a metal source (also referred to herein as a small metal slug) is heated in a vacuum chamber, which in some systems is maintained at about 10 f^-7Torr until the metal melts and atoms evaporate from the metal source. The metal source may be heated by any of a number of methods, including, for example, resistive heating, or by directing an electron beam at the metal source. From metalsThe source evaporated metal atoms propagate through the vacuum chamber and are deposited on one or more semiconductor wafers also present in the vacuum chamber.

During deposition of metal on a semiconductor wafer, according to certain semiconductor manufacturing processes, the semiconductor wafer may be covered with a barrier material, conventionally referred to as a "mask," that covers areas of the wafer where it is not desired to form a metal film. For example, the mask may be formed from a patterned layer of photoresist (also referred to herein as "resist"). The open areas in the mask are formed in the areas where a metal film is desired to be deposited on the wafer. These open areas are formed, for example, by applying a layer of photoresist to the wafer and exposing the photoresist to light through a photolithographic mask that includes the pattern desired to be formed in the photoresist. The exposed photoresist polymerizes. A subsequent development step chemically removes the non-polymerized photoresist. The remaining photoresist is baked to remove the volatilized chemicals. It is desirable that the remaining photoresist be polymerized, without crosslinking, i.e., hardened. Aspects and embodiments of the methods and apparatus disclosed herein are not limited to semiconductor manufacturing processes that employ any particular mask formation process.

After the metal film is deposited, the mask and any metal deposited on the mask are removed in a process known as metal lift-off. What remains is the metal film formed on the semiconductor wafer in the areas not blocked by the mask.

In some semiconductor manufacturing processes, the metallized wafer is placed in a solvent, such as N-methyl pyrrolidone (NMP) or ethylene glycol, to undergo a wet strip process to dissolve the photoresist used as a mask to define the desired metallization pattern, strip unwanted metals, and form the desired portions of the circuit.

Most useful photoresists can crosslink if exposed to excessive heat or light. Crosslinked or hardened photoresists are not completely soluble in the wet stripping chemicals conventionally used in certain manufacturing processes. Thus, if the photoresist on the wafer becomes cross-linked prior to the lift-off process, photoresist residue will remain on the wafer after the lift-off process. Although photoresist residues can often be removed by reprocessing with more aggressive wet and/or dry stripping processes, the additional reprocessing steps negatively impact the production flow and manufacturing schedule.

Furthermore, if contaminants present on the semiconductor wafer, such as photoresist residue or nodules from metal "sputtering" as discussed below, are not detected on the wafer, the contaminants may cause further problems with downstream process steps. Such problems may include, for example, poor adhesion or planarity of subsequently deposited layers. These problems can lead to a reduction in line yield (the amount of wafers that are not rejected during manufacturing) and/or chip yield (the amount of active devices per wafer formed in the manufacturing process). Undetected contaminants may also lead to reliability problems, including device failure in the field.

Disclosure of Invention

This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

Applicants have found that certain photoresists can be crosslinked not only by exposure to heat or light, but also by bombardment with backscattered electron beams from the electron beam used to heat the small metal masses in the evaporator. In addition, applicants have found that the amount of crosslinking and metal "spitting" is related to the amount of impurities in the small metal nuggets.

According to an embodiment of the present invention, a method of detecting impurities in a small metal block disposed in an electron beam evaporator during an electron beam metal evaporation/deposition process is provided. The method includes monitoring a first electrical signal provided by an electrode disposed in a deposition chamber of the electron beam evaporator and physically separated from the slug during the electron beam metal evaporation/deposition process, detecting a change in the first electrical signal during the electron beam metal evaporation/deposition process, and indicating an increased impurity concentration in the slug in response to the detected change in the first electrical signal.

According to certain aspects, the act of detecting includes comparing the first electronic signal to a threshold value and determining that the impurity above a defined concentration is present in the metal nugget in response to the first electronic signal exceeding the threshold value by more than a predetermined amount.

According to certain aspects, the threshold value is determined by monitoring a second electrical signal provided by the electrode over a time period of the e-beam metal evaporation/deposition process, wherein the threshold value is determined from the second electrical signal.

According to certain aspects, monitoring at least one of the first electrical signal and the second electrical signal comprises monitoring at least one of a voltage reading and a current reading. According to a further aspect, monitoring at least one of the first electronic signal and the second electronic signal comprises monitoring at least one of a first series of periodic readings and a second series of periodic readings, respectively. According to a further aspect, the method further includes establishing a baseline mean and a baseline standard deviation for the second series of periodic readings. According to a further aspect of the method, determining that impurities above a defined concentration are present in the small metal slug includes: the determination is made in response to observing a first series of cycle readings from the electrode having a mean value that deviates from the baseline mean value by more than a predetermined amount and observing a first series of cycle readings from the electrode having a standard deviation that deviates from the baseline standard deviation by more than a predetermined amount. According to other aspects, the method further includes providing the first series of periodic readings and the second series of periodic readings to a computer system programmed to generate an alarm in response to the first series of periodic readings violating a set of Statistical Process Control (SPC) rules established based on the second series of periodic readings.

According to certain aspects, the method further comprises: in response to determining that the impurity is present in the small metal masses at a concentration in excess of the predetermined concentration, an indication is provided to the production control system that the electron beam metal evaporation/deposition system is not suitable for processing semiconductor production wafers.

According to certain aspects, the method further comprises directing an electron beam at a surface of the metal slug, wherein the act of monitoring the first and second electron signals comprises monitoring backscattered electrons from impingement of the electron beam with impurities in the metal slug. According to a further aspect, the method further comprises providing a change in at least one of a voltage of the electrode and a current flowing through the electrode by providing a backscattered electron beam to impinge on the electrode.

According to certain aspects, the method further comprises replacing the slug in response to determining that impurities above a defined concentration are present in the slug.

According to another embodiment of the present invention, a method is provided. The method includes depositing metal obtained from a slug on a semiconductor wafer in a vacuum chamber of an electron beam metal evaporation/deposition system during an electron beam metal evaporation/deposition process, monitoring an electrical signal provided by an electrode disposed in the vacuum chamber during the electron beam metal evaporation/deposition process, detecting a change in the electrical signal during the electron beam metal evaporation/deposition process, and at least one of stopping processing the semiconductor wafer on the electron beam metal evaporation/deposition system and performing preventative maintenance on the electron beam metal evaporation/deposition system in response to the change in the electrical signal indicative of an increased impurity concentration in the slug.

According to certain aspects, the method further comprises inspecting a semiconductor wafer being processed in the e-beam metal evaporation/deposition system while detecting the change in the electrical signal. According to a further aspect, the method further includes reprocessing the semiconductor wafer being processed in the e-beam metal evaporation/deposition system upon detecting a change in the electrical signal.

According to certain aspects, the method further comprises electrically isolating the electrode from ground.

According to certain aspects, performing preventative maintenance includes replacing small metal blocks.

According to certain aspects, monitoring the electrical signal provided by the electrode during the e-beam metal evaporation/deposition process includes monitoring an electrical signal generated by electrons backscattered on the electrode from impurities in the small metal mass.

According to certain aspects, an increase in yield is achieved by reducing the number of semiconductor wafers comprising crosslinked photoresist during processing in an e-beam metal evaporation/deposition system.

According to certain aspects, an increase in yield is achieved by reducing the number of semiconductor wafers including metal nodules resulting from metal sputtering during processing in an e-beam metal evaporation/deposition system.

According to another embodiment of the present invention, an electron beam metal evaporation/deposition system is provided. The electron beam metal evaporation/deposition system includes an electrode configured to be disposed within a vacuum chamber of the electron beam metal evaporation/deposition system and spaced from ground, the electrode configured to be disposed to have an unobstructed linear path between a portion of the electrode and a surface of the slug during operation of the electron beam metal evaporation/deposition system, and an electrical gauge coupled to the electrode, the electrode further configured to be disposed to not interfere with an electron path between the surface of the slug and a wafer processed in the electron beam metal evaporation/deposition system.

According to certain aspects, the electrical meter is at least one of a voltmeter and an ammeter.

According to certain aspects, the apparatus further includes a controller configured to receive a signal from the electrical meter, to detect a change in the signal relative to the baseline, and to alert an operator of the change in the signal.

Drawings

The figures are not intended to be drawn to scale. In the drawings, each identical or nearly identical component that is illustrated in various figures is represented by a like numeral. For purposes of clarity, not every component may be labeled in every drawing. In the drawings:

FIG. 1 is a Scanning Electron Microscope (SEM) image of photoresist residue on a surface of a semiconductor wafer;

FIG. 2 is an SEM image of photoresist residue in a cross-section of a semiconductor wafer;

FIG. 3 is a partial flow diagram of a conventional semiconductor manufacturing process flow;

FIG. 4 is a cross-sectional view of an electron-beam metal evaporation/deposition system including an electrode according to an embodiment of the present invention;

FIG. 5 is an isometric illustration of the electrode of FIG. 4 installed in a deposition chamber of an electron beam metal evaporation/deposition system;

FIG. 6 is a schematic diagram of an appliance meter electrically coupled to an electrode according to an embodiment of the present invention;

FIG. 7 illustrates a computer control system useful in one or more embodiments of the invention;

FIG. 8 illustrates a storage system that may be used in the computerized control system of FIG. 7 in accordance with one or more embodiments of the invention;

FIG. 9 is a flow chart of a portion of a semiconductor manufacturing process flow according to an embodiment of the present invention;

FIG. 10 is a plot of electron beam power and voltage readings from a test of an electrode according to an embodiment of the present invention installed in an electron beam metal evaporation/deposition system;

FIG. 11 is a plot of electron beam power and voltage readings from another experiment with an electrode according to an embodiment of the present invention installed in an electron beam metal evaporation/deposition system; and

FIG. 12 is a plot of electron beam power and voltage readings from yet another experiment of an electrode according to an embodiment of the present invention installed in an electron beam metal evaporation/deposition system.

Detailed Description

The invention is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the drawings. The invention is capable of other embodiments and of being practiced or of being carried out in various ways. Also, the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of "including," "comprising," or "having," "containing," "involving," and variations thereof herein, is meant to encompass the items listed thereafter and equivalents thereof as well as additional items.

The present invention is generally directed to systems and methods for detecting impurities in small metal masses employed in electron beam (e-beam) metal evaporation/deposition systems (also referred to herein as "evaporators", "e-beam evaporators" or "metal evaporators"). It is desirable to detect impurities in contaminated small metal pieces before processing many wafers through an evaporator equipped with these small metal pieces. This impurity can cause many forms of defects that can be observed on wafers processed by e-beam metal evaporation/deposition systems, or cannot be observed immediately, leading to failures during subsequent process steps or in operation.

For example, impurities such as carbon may be present in the nuggets. Possibly in combination with carbon during drawing and forging in the manufacture of small nuggets, where the oil acts as a lubricant. Poor clean room operation and inadequate handling techniques in the evaporator to remove the nuggets may also introduce carbon into the nuggets.

Impurities such as carbon in the gold slug cause the generation of a high energy backscattered electron beam when the slug is heated by the e-beam to evaporate the gold. These backscattered electron beams may impinge on areas of photoresist on the wafer to be processed by the e-beam metal evaporation/deposition system and cause photoresist cross-linking. The crosslinked photoresist as described above may not be completely removed in a subsequent photoresist stripping process, leaving photoresist residues, which require re-processing and/or additional cleaning of the wafer for removal. An example of a photoresist residue 10 formed on a semiconductor wafer according to this mechanism is shown in fig. 1 and 2.

The electron beam striking the molten slug also produces secondary electron emissions from the slug. The secondary electrons are formed by electrons in the electron beam which collide the electrons against atoms in the small metal block, or by electrons which are absorbed in the electron beam and then re-emitted from atoms in the small metal block. Secondary electrons typically have much lower energy than backscattered electrons and therefore do not contribute to photoresist crosslinking, if at all, as does a backscattered electron beam.

Carbon impurities in the molten nuggets float to the surface of the molten nuggets, forming a "skin (skin)". When an electron beam directed to the surface of the gold slug encounters carbon, some of the electrons from the electron beam are elastically backscattered. The backscattered electron beams are not able to efficiently transfer their energy to the small gold masses to melt them. The backscattered electron beam retains most, if not all, of the energy imparted to it as the electron beam is formed. The energy of the backscattered electron beam is about 10 kilovolts in a typical e-beam (e-beam) metal evaporator system. If the backscattered electron beams reach a portion of the photoresist on the wafer to be processed in the e-beam metal evaporation/deposition system in a sufficient amount, they may impart sufficient energy to the photoresist to cause that portion of the photoresist to become crosslinked.

The exact reason why the carbon particles in the molten nuggets tend to backscatter electrons from the applied electron beam is not fully understood. However, it has been found that the free electrons generated by a solid phase material when struck by an electron beam are significantly larger than when the material is in the liquid phase. When the electron beam strikes a solid source, it produces many energetic electrons. As the solid source melts, the emission level decreases. Due to its very high melting point, the carbon remains in solid phase on the molten nuggets at the temperatures typically employed in e-beam metal evaporation/deposition systems, and thus can effectively block the electron beam from reaching the gold in the nuggets and melt it, causing the electrons to elastically backscatter without being absorbed into the nuggets.

The sidewalls of the patterned photoresist, which is used as a mask for the metal deposition process, are exposed throughout the deposition process and are masked as the surface of the photoresist becomes covered with metal while the metal deposition process is performed. After the first few hundred angstroms of metal are deposited, the photoresist under the large metallization structure will be masked from further bombardment by the backscattered electron beam. Thus, the photoresist sidewalls are more cross-linked than the surface of the photoresist. This results in a linear or striped pattern of photoresist residue after chemically stripping the remaining photoresist, as shown in fig. 1.

Carbon impurities in the nuggets may also contribute to gold "spitting" where droplets of liquid gold are expelled from the molten nuggets. These molten droplets may deposit on the wafer to be processed in the e-beam metal evaporation/deposition system and may, in some cases, cause shorts between adjacent metal lines or other structures on the device to be formed on the wafer. For example, gold "spits" on the inside surface of the electrodes of metal-insulator-metal (MIM) capacitors can cause reliability problems. Gold particles deposited by gold "sputtering" can also damage the probe tips or expensive diaphragm probes used in microchip circuit testing.

In some semiconductor manufacturing processes, efforts have been made to reduce defects caused by carbon impurities in the small gold pieces employed in gold evaporators by adding tantalum to the small gold pieces. Tantalum adsorbs carbon, thus reducing the amount of carbon that is unconstrained to form a film on the surface of the metal slug and cause "sputtering" or electron back scattering. However, this approach is not without problems. Crucibles used in e-beam evaporators for holding small pieces of gold (or other metals) are typically formed from materials such as molybdenum, tungsten, silicon carbide, or carbon. Adding tantalum to the nuggets may induce wetting of the crucible holding the nuggets (wetting). If the crucible is wetted by the molten gold, the crucible may crack due to differential thermal shrinkage between the crucible material and the gold upon cooling. Furthermore, the addition of tantalum to small gold blocks does not lead to a spray-free process in all cases. It is therefore desirable to provide identification of contaminated patches before they cause a large number of defects on the wafer, rather than attempting to mitigate the effects of possible contaminants by, for example, applying an adsorbent material to the patches.

A typical semiconductor manufacturing process generally includes a series of process steps similar to those shown in the flow chart of fig. 3. The metal deposition process typically includes act 410-450 of figure 3. In step 410, the wafers are cleaned, for example, by immersing them in an acid solution such as hydrochloric acid. After the pre-deposition clean, the wafer is loaded in a metal evaporator (act 420) and a metal deposition recipe is run (act 430). Upon completion of the metal deposition, the wafer is removed from the metal evaporator (act 440), and an additional batch of pre-cleaned wafers is introduced into the evaporator (act 450).

The wafer receiving the metal deposition is subjected to a lift-off process (act 460) in which photoresist and/or additional metal deposition masks that may have been used are removed from the wafer along with the metal deposited on the masks. The wafer then typically proceeds to an inspection operation (act 470) where a portion of the wafer is inspected, or in some processes all of the wafer is inspected, such as by an automated optical inspection tool. In some processes, the inspection operation is performed manually. During inspection operation 470, a determination is made as to whether a defect, such as a resist residue or a metal "sputtered" nodule, is present on the processed wafer (act 480). If defects below the predetermined amount are observed, the processed wafer is transferred to a subsequent processing operation and the operation of the wafer in the vaporizer continues (act 490).

However, if an unacceptable amount of defects are observed on the inspected wafer, the vaporizer is no longer in service (act 500) and fault detection is performed (act 520). Wafers processed by a metal evaporator after being found to include a batch of wafers having defects from a metal deposition process will be suspect. If the defect found in the first bad batch does result from a problem such as contamination of the small metal blocks in the evaporator, it is likely that a batch processed after the first found bad batch will also have defects caused by the contaminated small metal blocks. Thus, these batches are also likely to need to be reprocessed or discarded. A sample of the wafer processed through the vaporizer after the wafer determined in act 480 to have an unacceptably high number of defects, or in some cases all of the wafers, is thus inspected (act 510).

A determination is made if these subsequently processed wafers also exhibit unacceptable levels of defects (act 530). If the wafers appear acceptable, they are forwarded for conventional post-processing (act 540). However, if these wafers exhibit an unacceptably high level of defects, a determination is made as to whether they can be reprocessed, such as by an additional cleaning operation (act 550), to remove the observed resist residue.

If the wafers are determined to be reprocessable, they are reprocessed (act 560) and then forwarded for subsequent processing (act 540). In some cases, the reprocessed wafers may be inspected again before being sent to subsequent processing. If it is determined in act 560 that the wafers can no longer be processed, for example, if they have an unacceptably high level of non-removable metal "spitting," the wafers are discarded (act 570).

In a typical semiconductor manufacturing process, many batches of wafers may be processed by a metal evaporator between the time that a bad slug begins to cause defects to appear on the wafer processed therethrough and the time that these defects are discovered in downstream inspection steps. Many batches of wafers may be affected before the problem of small metal blocks is discovered. The contaminated small metal blocks may therefore cause significant costs in time and throughput for reprocessing the wafers, for example to remove resist residues. Significant costs may also be incurred if the defects found on the wafer cannot be remedied by the reprocessing process, and the affected wafer must be discarded.

To facilitate reducing these potential losses, methods and apparatus have been developed to detect the presence of contaminated small metal lumps in the vaporizer in less time than in previously known processes (and in some embodiments, in real time during vaporizer operation). It has been found that by fabricating the electrode 510 (see fig. 4 and 5) to fit within the evaporator deposition chamber 505, the total electron radiation generated by small metal pieces (e.g., small gold pieces) during operation of the evaporator can be monitored. In some embodiments, the electrode 510 is electrically isolated from the evaporator's inner surface 515 and ground 550 by one or more insulators 520. In certain embodiments, the electrodes are electrically coupled to a high impedance voltmeter 710 and/or an ammeter (see fig. 6).

In certain embodiments, the electrode 510 and/or insulator 520 are cleaned or replaced along with the other pieces of the evaporator shroud during regularly scheduled preventative maintenance operations.

During operation, an e-beam 537 is generated by an electron gun 535 and directed to the small metal pieces in crucible 530. The electrons generated by the e-beam 537 striking the small metal slug will backscatter and strike the electrode 510, resulting in a negative voltage and/or current from the electrode. The measured voltage is proportional to the amount of backscattered electron beams generated and therefore to the amount of impurities in the small metal pieces, for example carbon impurities in the small gold pieces. The higher the carbon concentration in the nuggets, the greater the negative voltage on the electrode. Similarly, the higher the carbon concentration in the small gold lumps, the more backscattered electrons that strike the electrode and the greater the current generated. When the amount of carbon on the small gold lumps reaches a threshold value, resist cross-linking and/or gold sputtering will occur as indicated by the current generated by the electrodes and/or the voltage of the electrodes. Thus, the voltage on the electrodes and/or the current produced by the electrodes can be monitored and when the voltage and/or current reaches or exceeds a threshold, the evaporator is shut down to replace the slug. The voltage generated on the electrode may be a negative voltage in some embodiments, and thus exceeding the threshold may be a voltage exhibiting a more negative voltage than the threshold negative voltage.

As described above, without detecting the impurity (or impurities) in the small metal chunks during the metal deposition process, problems (e.g., gold sputtering and/or resist cross-linking) may not be detected for many lots until the affected wafer reaches a much later inspection in the manufacturing process. This can result in a significant amount of waste and lost revenue. Some embodiments of the present invention may identify the cause of these problems as soon as they begin to occur.

Embodiments of the present invention may provide an indication to an operator that the metal source in the vaporizer is appropriate to be replaced. In certain embodiments, the cue is given before problems such as resist cross-linking and/or metal sputtering occur. By monitoring the electrode voltage, or in some embodiments, the electrode current, an appropriate threshold voltage or current can be established that indicates a desire to replace the contaminated slug immediately.

As shown in fig. 4 and 5, in one embodiment, the electrode 510 is fabricated in the form of a copper plated ring. Copper is used as the electrode material for two reasons. First, copper is a very good conductor. Second, copper is a metal compatible with high vacuum systems. However, any vacuum compatible conductive metal (e.g., stainless steel) may be used in commercial systems, as the present invention is not limited to a particular type of electrode material.

In one embodiment, the insulator 520, which isolates the electrodes 510 from the evaporator's interior surface 515 and from ground, is comprised of alumina, an insulating ceramic material. In other embodiments, the insulator 520 is constructed from other ceramics, such as titanium dioxide, silicon dioxide (quartz), or conventional glass. In other embodiments, insulator 520 is formed from a plastic material such as PVDF. Any non-conductive material that is vacuum compatible and has sufficient mechanical strength to support the electrodes may be used for the insulator 520.

In the embodiment shown in fig. 4 and 5, the electrode 510 is disposed in the evaporator deposition chamber 505 such that, during operation of the evaporator, there is an unobstructed path between the small metal slug within the crucible 530 of the evaporator and the electrode 510. The electrode 510 is also positioned so that it does not obstruct the path 551 between the small metal blocks and the locations on the wafer support structure 540 where the wafer may be mounted during metal deposition in the vaporizer. Such positioning of the electrodes allows the electrodes to collect backscattered (and secondary) electrons from the small metal masses without blocking the metal deposition on the wafer.

Although shown as a copper plated ring, the electrodes may be formed in any shape and configuration in alternative embodiments. For example, in one embodiment, the electrodes are in the form of one or more loops of wire. In another embodiment, the electrodes 510 are formed by a plurality of plates within the evaporator deposition chamber. In yet another embodiment, electrode 510 is formed from a metal mesh. One skilled in the art can form the electrode 510 in any shape and size to fit within any particular type of evaporator.

In operation, the voltage and/or current generated across the electrode 510 during operation of the evaporator using known good (i.e., contamination minimization) small metal blocks is monitored to establish a baseline level of voltage and/or current. A plurality of voltage and/or current data points are sampled in time at a given frequency (e.g., one data point per half second, or in some embodiments, one data point per second). Other embodiments may employ data sampling frequencies at any convenient rate or within the capabilities of a data recording device used in conjunction with the electrodes. These data points are used to generate a baseline average, baseline range or standard deviation, or in some embodiments, both a baseline average and standard deviation or range of one or both of the voltage and current generated by the electrodes during operation of the vaporizer with known good small metal blocks.

Embodiments of the present invention can be used to inspect many types of small metal pieces for impurities, such as gold, aluminum, titanium, or any other metal that may be used in a metal evaporator. These different metals produce different amounts of back-scattered and secondary electrons when struck by the e-beam. The amount of backscattered and secondary electrons generated also varies depending on the particular type of evaporator employed and the intensity of the e-beam applied to the slug. In addition, the particular design, shape, positioning, and material (or materials) of the structure of a particular electrode affect the amount of electrons captured by the electrode. Thus, a baseline of voltage and/or current produced by a particular type of slug on a particular evaporator having a particular electrode configuration is different than a baseline produced on an evaporator having a different type of slug and/or electrode configuration. However, differences in the established baseline in the voltage and/or current generated by the cleaned metal slug and deviations from that baseline indicative of a contaminated slug may be detected regardless of the particular value(s) of the established baseline parameter(s).

The established baseline may also vary from one slug to another on the same evaporator, for example, due to the accumulation of metal on the electrodes over time or the size, shape or surface properties of different slugs. In some embodiments, the voltage and/or current baselines established for the electrodes in the vaporizer are periodically calibrated as (or after) small metal masses in the vaporizer lose mass due to vaporization, thus increasing the concentration of non-vaporized contaminant material as (or after) or both as the electrodes accumulate deposited metal.

In other embodiments, the parameters (e.g., mean, range, and/or standard deviation) of the voltage and/or current baselines may be substantially the same for evaporators having similar or identical vacuum chamber and electrode configurations. Thus, in certain embodiments, the parameters of the voltage and/or current baselines established on one evaporator may be applied to other similarly configured evaporators. Thus, in some embodiments, it is not necessary to establish voltage and/or current baselines for a particular type of small metal block for each individual evaporator. Conversely, voltage and/or current baselines established using known good metal nuggets (and, in some embodiments, known contaminated metal nuggets) on a representative vaporizer can provide this data to establish threshold and/or control limits for acceptable electrode voltage and/or current parameters, which can be used to monitor any of a group of similarly configured vaporizers for the presence of a possible contaminated metal nugget. In some embodiments, the manufacturer of the vaporizer will calibrate the vaporizer to produce specific parameters of electrode voltage and/or current when a small metal slug is well-performing and contaminated. The manufacturer calibration may in some embodiments reduce or eliminate the need for a user of the vaporizer to perform a baseline measurement to establish a control chart for electrical parameters measured on the electrodes, which may be used to distinguish between good metal pieces and contaminated metal pieces.

Once a baseline in voltage and/or current is established, deviations in the parameters of the baseline indicate possible contamination of the small metal pieces. For example, a downward shift in voltage and/or an increase in current to or from the electrode during operation of the evaporator would indicate a possible increase in the number of backscattered electrons generated and, therefore, a small metal slug of possible contamination. Similarly, an increase in standard deviation or reading range will, in certain embodiments, be indicative of a contaminated slug. In some embodiments, the voltage signal observed from the electrodes in an evaporator with a particular baseline mean and standard deviation can indicate both a downward shift in mean and an increase in standard deviation upon surface contamination of the small metal pieces.

Any of a number of other changes in the electrical signal measured at the electrodes may be used to provide an indication of possible contamination of the slug. For example, if the magnitude of the trend is not statistically possible while accounting for the inherent variation in readings, the trend of the voltage or current readings (i.e., the first derivative of the curve formed by the series of data points) may be indicative of a small metal slug of possible contamination. In other embodiments, a change in the moving average of a series of voltage and/or current readings (e.g., three or five consecutive readings) that is statistically unlikely to account for the inherent variation in readings may be indicative of a small metal slug of possible contamination. In a further embodiment, a change in the range of a series of data points that is statistically unlikely to take into account the inherent variation in readings (e.g., the last three or five consecutive readings compared to a series of previous readings) may be indicative of a small metal slug of possible contamination.

One of ordinary skill in the art of process control will be able to set control limits (e.g., statistical program control limits) around a set of baseline voltage and/or current readings that, when violated, will indicate a potentially contaminated metal slug. In some embodiments, control charts will be established for voltage and/or current readings from the electrodes, and data points for the voltage and/or current readings are plotted on these control charts. If the plotted data points violate one or more Statistical Program Control (SPC) rules, this will represent a small metal block of possible contamination.

In some embodiments, the control chart may establish and plot data points for monitoring against one or more Western Electric SPC rules (Western Electric SPC rules). These rules are as follows:

1) one point outside the upper or lower control limit

The upper and lower control limits are set at three standard deviations from the mean. If a point is outside any of these limits, there is only a 0.3% chance that it will be caused by normal processes.

2) Eight points on the same side of the mean

The chance that any given point falls above or below the average is equal. One dot has half the chance of the previous one falling on the same side of the average. The probability that subsequent points also fall on the same side of the mean is one quarter. The probability of eight dots on the same side of the average is only about 1%.

3) Eight point increase or decrease

The logic used here is the same as that used for the eight points on the "same side of the average". Sometimes, the rule changes to seven-point rises or falls.

4) Two of the three points are outside the warning limits

The warning limit is typically set at two standard deviations (i.e., two Σs) from the mean. The probability of any point falling outside the warning limit is only 5%. The chance of two of the three consecutive points falling outside the warning limit is only about 1%.

5) Four of the five points fall outside of one sigma

In the usual process, 68% of the points fall within one Σ of the average value, and 32% of the points fall outside it. The probability of four out of five points falling outside of one sigma is only about 3%.

6) Fourteen points alternate direction

The rule treats each pair of adjacent points as a unit. For all seven pairs, the chance that the second point is always higher (or always lower) than the previous point is only about 1%.

7) Fifteen continuous points are located in one sigma

In normal operation, 68% of the points fall within one Σ of the average. The probability of this is less than 1% for 15 consecutive points.

8) Eight continuous points are positioned outside one sigma

Since 68% of the points are located within one Σ of the average value, the probability that eight consecutive points fall outside one Σ line is less than 1%.

In other embodiments, a control chart may be employed in which violations of one or more Wheeler or Nelson SPC rules, which are well known to those familiar with statistical program control, may be used as an indicator of possible contaminating metal nuggets.

In some embodiments, the voltage and/or current from the electrodes in operating the vaporization system is automatically measured periodically by a voltmeter and/or galvanometer, and the results of the measurements are provided to a monitoring computer or controller programmed to issue an alarm if one or more of the parameters measured deviate above or below a threshold, deviate outside an acceptable range, or violate one or more SPC rules. In some embodiments, an acceptable threshold or range of measured parameters is predetermined by performing a baseline measurement on the electrode on a particular vaporizer from a particular small metal block being monitored.

In different embodiments, the monitoring computer or controller for monitoring the electrical parameter from the electrode 510 may be implemented in any of a number of forms. In one example, a computerized controller for embodiments of the systems disclosed herein is implemented using one or more computer systems 600 as exemplarily shown in fig. 7. Computer system 600 may be, for example, a general purpose computer, e.g., Intel-basedOr Core^TMProcessor, MotorolaProcessor, SunProcessor, Hewlett-PackardProcessors or any other type of processor or combination of such computers. Alternatively, the computer system may include specially-programmed, special-purpose hardware, such as an application-specific integrated circuit (ASIC) or a controller dedicated to a semiconductor wafer processing apparatus.

Computer system 600 may include one or more processors 602 typically connected to one or more storage devices 604, which may include, for example, any of one or more disk drive memories, flash memory devices, RAM storage devices, or other devices for storing data. Memory 604 is typically used for storing programs and data during operation of the controller and/or computer system 600. For example, memory 604 may be used to store historical data relating to measured electrical parameters of electrode 510 over a period of time as well as current sensor measurement data. Software, including program code to implement embodiments of the present invention, may be stored on a computer-readable and/or writable non-volatile recording medium (described further with reference to FIG. 8) and then copied into memory 604, where it may be executed by processor 602. Such programming code may be written in any of a number of programming languages, such as, for example, Java, Visual Basic, C, C #, or C + +, Fortran, Pascal, Eiffel, Basic, COBALL, or any variety of combinations thereof.

The components of computer system 600 may be connected by an interconnection mechanism 606, which may include one or more buses (e.g., between components integrated within the same device) and/or networks (e.g., between components located in respective sputtering devices). The interconnection mechanism typically enables information (e.g., data, instructions) to be exchanged between the systems 600.

The computer system 600 may also include one or more input devices 608, such as a keyboard, mouse, trackball, microphone, touch screen, and one or more output devices 610, such as a printing device, display screen, or speaker. The computer system may be electrically or otherwise connected to the electrical sensor 614, which may include, for example, one or more of an ammeter and voltmeter configured to measure an electrical parameter of the electrode 510. Additionally, computer system 600 may include one or more interfaces (not shown) that may connect computer system 600 to a communication network (in addition to or in lieu of a network that may be formed by one or more components of system 600). In some embodiments, the communication network forms part of a process control system of a semiconductor manufacturing line.

In accordance with one or more embodiments, one or more output devices 610 are coupled to another computer system or component for communication with computer system 600 over a communication network. Such a configuration allows one sensor to be located remotely from another sensor, or any sensor to be located at any significant distance from any subsystem and/or controller, while still providing data therebetween.

As exemplarily shown in fig. 8, the controller/computer system 600 may include one or more computer storage media, such as a readable and/or writable non-volatile recording medium 616, in which signals defining programs executed by one or more processors 620 (e.g., processor 602) may be stored. The medium 616 may be, for example, a disk memory or a flash memory. In typical operation, processor 620 will cause data (e.g., code implementing one or more embodiments of the invention) to be read from storage medium 616 into memory 618, which allows the information to be accessed by the processor(s) faster than medium 616. Memory 618 is typically a volatile random access memory such as a Dynamic Random Access Memory (DRAM) or a static memory (SRAM) or other suitable device that facilitates transfer of information to and from processor 620.

While computer system 600 is illustratively shown as a computer system upon which aspects of the present invention may be implemented, it should be understood that the present invention is not limited to implementation in the exemplary illustrated software or computer system. Indeed, rather than being implemented on a general purpose computer system, for example, the controller, or components or portions thereof, may alternatively be implemented as a dedicated system or dedicated Programmable Logic Controller (PLC) or distributed control system. Further, it should be understood that one or more features or aspects of the control system may be implemented in software, hardware, or a combination of software and hardware, or any combination thereof. For example, one or more algorithms executable by computer system 600 may be executed on separate computers, which in turn may communicate over one or more networks.

Fig. 9 shows a procedure for operating an evaporator comprising an electrode according to an embodiment of the invention. In the process of FIG. 9, the deposition operation includes acts 810 and 870. Acts 810 through 830 and 860 through 870 are substantially the same as acts 410 through 430 and 440 through 450, respectively, of fig. 3 described above. Unlike fig. 3, the routine of fig. 9 additionally includes an act of monitoring an electrical characteristic of the electrode (act 840). In one embodiment, the voltage from the electrodes is monitored, and the average and standard deviation of the monitored voltage is calculated. In another embodiment, the current from the electrodes is monitored, and the average and standard deviation of the monitored current is calculated. In act 850, the monitored electrical parameter is compared to a baseline previously established for the parameter, for which it is determined whether the measured parameter falls within acceptable tolerance limits.

If it is determined that the measured parameters are within the tolerances, the metallized recipe is complete, the wafer is unloaded (act 860), and transported to normal downstream processes (act 880), and a new batch of pre-cleaned wafers is introduced into the vaporizer (act 870). Alternatively, if in act 850 the measured parameters are found to be outside of tolerance, they are sent for inspection (act 900) when the wafer is unloaded (act 890). If it is determined that defects such as metal splatter and/or resist residue are present in the wafer at unacceptable levels (act 910), the wafer may be discarded or, in some embodiments, reprocessed if possible (act 930). The vaporizer that generated the out of tolerance signal and the defective wafer will stop production and undergo trouble shooting (act 940). If the small metal pieces in the evaporator are found to be contaminated, it is replaced before returning the evaporator to the production line for continued operation (act 940).

If wafers from the vaporizer that exhibit the out-of-tolerance parameter are found to have an acceptably low defect density in act 910, they are transferred for further processing, but as a precautionary measure, the vaporizer is subjected to a trouble shooting and any contaminated small metal pieces are replaced before additional wafers are processed by the vaporizer (act 940).

It should be understood that the various actions shown in FIG. 9 are merely exemplary. In different embodiments, different acts or more of the acts may be performed in different orders. In other embodiments, additional acts are included in the program, and in further embodiments, one or more of the illustrated acts are omitted or replaced.

Other embodiments of methods of operating evaporators comprising the electrodes described herein will be apparent to those skilled in the art. In an alternative embodiment, a positive voltage is applied to the electrodes. This positive voltage attracts electrons and thus increases the amount of electrons captured by the electrodes, making the device more sensitive to the presence of a backscattered electron beam. The positive charge on the electrode can also deflect (at least to some extent) the backscattered electron beam toward the sidewalls of the deposition chamber and away from the wafer in the wafer holder. The amount of deflection will vary with the amount of voltage applied to the electrodes. In certain embodiments utilizing positively biased electrodes, the average and/or standard deviation of the current from the electrodes may be used as the electrical parameter being monitored so that a constant voltage may be maintained across the electrodes. In other embodiments, the voltage applied to the electrodes may be made or allowed to vary over time. As with the previous embodiments, monitoring changes in current and/or voltage mean or standard deviation from baseline values may indicate small metal pieces of possible contamination. Upon receiving a signal from a controller or other system for monitoring the vaporizer indicative of a possible contaminated electrode, the operator can troubleshoot the vaporizer and replace the suspect small metal block if it is indeed needed.

In another embodiment, an electrical charge (positive or negative) may be applied to electrode 510. The charge on the electrodes can be measured at any time during operation of the evaporator. A change in the measured charge or a change in the rate of change of the charge on the electrode may be indicative of a potentially contaminating metal slug.

Examples of the invention

To study the source of backscattered electron radiation, a series of experiments were performed to compare the amount of energetic free electrons generated from different materials during the evaporation process in an e-beam metal evaporation/deposition system. The electrodes were fabricated to fit within the vacuum chamber of an e-beam metal evaporation/deposition system. The electrodes are provided to compare the number of energetic electrons generated by electron beams impinging on different materials. The electrodes are formed from copper plates bent into rings. The copper electrode plate is electrically isolated from ground by a ceramic insulator in the vacuum chamber. The electrodes were connected with copper wires to a high impedance voltmeter (Keithley 2420 power meter) with data logic capability, where the voltage signals from the electrodes were monitored and written to a data file. Since the device is not calibrated against a certain standard, the measured potential cannot be correlated with the actual amount of charge on the electrodes.

A Temescal FC2700 evaporator with 15KW power supply was used for the experiment. The recipe was generated with a 30 second ramp up to 45% constant power delivery cycle for melting the metal nuggets and a 30 second ramp up to 50% constant power delivery cycle for maintaining the temperature of the melted metal nuggets at the metal evaporation temperature. The hold times for both of the constant power periods are 30 seconds. Different gold melts from different small gold blocks were run with this recipe and the voltage collected on the electrodes was recorded.

The first experiment was performed with a small gold slug, which was estimated to be contaminated with 1ppm of carbon on the surface. The data obtained from this experiment are shown in fig. 10. In this figure, the electron beam power is represented by "beam power" data points and the voltage observed at the electrodes is represented by "electrode potential" data points. When the 10KV high voltage is on and the transmitter is idle, the electrode potential is 0V relative to ground ("part of the graph to the left of the point indicated at point a"). As soon as the beam emits current and power rises and the beam begins to appear on the gold melt, the SMU measures about-1.25V ("point indicated by" a "). As the power continues to rise to 45% of the maximum value (the data point between the point indicated by "a" and the point indicated by "B"), the voltage remains relatively unchanged. When the nugget begins to melt, the voltage suddenly drops to-0.5V (the point immediately before the point indicated by "B"). Further increasing the power to 50% causes the electrode voltage to drop to-0.4V (starting approximately at the point indicated by "C"). At point "D", the beam power applied to the slug used in the 1ppm carbon contaminated sample test was abruptly turned off and the electrode voltage was restored to 0V.

In fig. 10, the data points for "beam power" indicate that the beam is still at 50% power at point "D", however, this is an artifact of the data collection methodology. Data for the 1ppm carbon contamination sample test and for the 30ppm carbon contamination sample test (described below with reference to fig. 11) were taken in different runs and then combined. The "beam power" data points correspond more closely in time to the "electrode potential" data points for the 30ppm carbon contamination sample test, which is why the 30ppm carbon contamination sample test data points do not appear to return to 0V on the right hand side of the graph; the beam power was still on during the 30ppm carbon contamination sample test data point acquisition on the right hand side of the graph.

The 2Hz pattern was scanned with a circular electron beam melting recipe and the data was reported at a sampling rate of 1 second. The circular scanning movement of the beam corresponds to the voltage peaks recorded and shown in fig. 10 and 11. The beam focus changes as the beam sweeps across different portions of the surface of the slug. The beam focus shrinks when passing through uncontaminated gold and the beam spreads when passing through high carbon areas, causing variations in the backscattered electron radiation, each passing through a small metal block. When the beam passes through a portion of the molten gold that includes carbon contamination, the amount of electrons scattered is greater than when the beam passes through the molten gold in the uncontaminated region. The change in the amount of backscattered electrons as the electron beam passes through the "dirty" and "clean" areas of molten gold reflects the change in voltage level observed at the data points in fig. 10 and 11. For the data point for 1ppm carbon contaminated molten gold (fig. 10), a voltage difference of about 0.5V was observed between the data points obtained when the electron beam passed through the "clean" and "dirty" regions of molten gold. The voltage difference was about 1V for the 30ppm carbon contaminated molten gold data point (fig. 11), which was about twice the voltage difference observed for the 1ppm carbon contaminated molten gold data point.

When the experiment was repeated with static beams and 1ppm carbon contaminated small gold blocks, there was no fluctuation in voltage, although a similar trend in data was observed as shown in fig. 10. The data obtained in this static beam test is shown in fig. 12. In fig. 12, points "a", "B", "C", and "D" represent similar points shown in fig. 10. Fig. 12 also includes point "E" where as the electron beam power is decreased, the molten gold solidifies and the voltage from the electrode is reduced to about-1.6V. Fluctuations in the observed voltage do not occur because the electron beam remains focused on a single portion of the slug rather than sweeping the "clean" and "dirty" areas of the slug.

The experiment described with reference to fig. 10 using the same recipe as the 2Hz circular beam scan pattern was repeated with molten gold having about 30ppm of carbon on the surface. Data from this replicate experiment is shown in figure 11. A comparison between fig. 10 and 11 shows the voltage difference observed between the "clean" 1ppm carbon contaminated patch and the "dirty" 30ppm carbon contaminated patch. When the beam strikes a 30ppm carbon contaminated slug at point "A", the grounded electrode potential is-2.2V, which is about twice the value observed for a 1ppm carbon contaminated slug. Melting the 30ppm carbon contaminated briquette requires more power and longer time than melting the 1ppm carbon contaminated briquette (note that points "B" and "C" in fig. 11 are shifted to the right from points "B" and "C" in fig. 10).

The voltage developed across the electrode throughout the ramp-up and constant power cycles with 30ppm carbon contamination of the patch was accompanied by the same trend as the experiment with 1ppm carbon contamination of the patch, although the entire curve shifted towards more negative voltages, indicating that more electrons were collected by the electrode. When the 30ppm carbon contaminated nugget became the melt, the voltage dropped, but it stayed at a higher total negative voltage than the observed 1ppm carbon contaminated melted nugget. The range and standard deviation of the voltage readings from the melted 30ppm carbon contaminated nuggets was significantly greater than the 1ppm carbon contaminated nuggets, as can be seen by comparing fig. 10 with fig. 11.

It was found that small nuggets with surface carbon greater than about 30ppm would not melt completely even when the electron beam was at 90% of the beam power supplied from a 15KW power supply. This means that the level of carbon contamination is sufficient to reflect or backscatter so many electrons from the surface of the nugget that insufficient electrons can reach the nugget to impart enough energy to completely melt the nugget.

Although both the 1ppm and 30ppm carbon-contaminated nuggets appeared optically clean and shiny, SEM examination of the 30ppm carbon-contaminated nuggets revealed spots of carbon particles on the surface. No carbon particles were found in SEM examination of molten gold made from low carbon content material (< 1 ppm). EDX measurements of 30ppm carbon contaminated nuggets showed strong carbon signals indicating carbon contamination on the nuggets. EDX detection of <1ppm carbon contaminated nuggets showed a much weaker carbon peak than EDX detection of 30ppm carbon contaminated nuggets.

Using the electrode potential as a reference, the observed voltage of-0.4V was determined as the baseline voltage for well-cleaned molten gold, while voltages less than-0.8V were determined as molten gold indicative of carbon contamination.

From this data, a voltage of-0.8 volts can be established as the threshold voltage level using the evaporator configuration used in this experiment. If the voltage data point peak is observed to be below-0.8 volts, this represents a potentially contaminating piece of gold in the particular vaporizer employed in the experiment. It should be understood that other metals and/or evaporators may have different voltage thresholds. However, applicants have determined that differences between the cleaned and contaminated metal pieces with respect to the mean and standard deviation or deviation in the range of voltage data points observed on electrodes disposed in the deposition chamber of the e-beam evaporator will exhibit similar results to those observed in this experiment.

The test also shows that in a metal evaporator, the first nuggets with relatively low carbon contamination show a voltage drop across the electrodes provided in the deposition chamber of the metal evaporator that is lower than the voltage drop of the nuggets with higher amounts of carbon contamination. Furthermore, for metal evaporators operating with dirtier nuggets, the voltage readings at the electrodes fluctuate significantly more than with cleaner nuggets.

Having thus described several aspects of at least one embodiment of this invention, various alterations, modifications, and improvements will readily occur to those skilled in the art. Such alterations, modifications, and improvements are intended to be part of this disclosure, and are intended to be within the spirit and scope of the invention. Accordingly, the foregoing description and drawings are by way of example only.

Claims (21)

1. A method of detecting impurities in a metal slug disposed in an electron beam evaporator during an electron beam metal evaporation/deposition process, the method comprising:

monitoring a first electrical signal during the e-beam metal evaporation/deposition process, the first electrical signal being provided by an electrode disposed in a deposition chamber of the e-beam evaporator, the electrode being physically separated from the small metal block;

detecting a change in the first electrical signal during the e-beam metal evaporation/deposition process by comparing the first electrical signal to a threshold value;

determining that impurities above a defined concentration are present in the small metal slug in response to the first electrical signal exceeding a threshold by more than a predetermined amount; and

an increased concentration of impurities in the metal slug is indicated in response to determining that impurities above a defined concentration are present in the metal slug.

2. The method of claim 1, wherein the threshold is determined by monitoring a second electrical signal provided by the electrode over a period of time of an e-beam metal evaporation/deposition process; wherein the threshold is determined from the second electrical signal.

3. The method of claim 2, wherein monitoring at least one of the first electrical signal and the second electrical signal comprises monitoring at least one of a voltage reading and a current reading.

4. The method of claim 3, wherein monitoring at least one of the first electrical signal and the second electrical signal comprises monitoring at least one of a first series of periodic readings and a second series of periodic readings, respectively.

5. The method of claim 4, further comprising establishing a baseline mean and a baseline standard deviation of the second series of periodic readings.

6. The method of claim 5, wherein determining that impurities above the defined concentration are present in the small metal piece comprises: the determination is made in response to at least one of observing that an average of the first series of periodic readings from the electrode deviates from the baseline average by more than a predetermined amount and observing that a standard deviation of the first series of periodic readings from the electrode deviates from the baseline standard deviation by more than a predetermined amount.

7. The method of claim 6, further comprising providing the first series of periodic readings and the second series of periodic readings to a computer system programmed to generate an alert in response to the first series of periodic readings violating a set of statistical process control criteria established based on the second series of periodic readings.

8. The method of any of claims 1-7, further comprising providing an indication to a production control system that the electron beam metal evaporation/deposition system is not suitable for processing a semiconductor product wafer in response to determining that an impurity in excess of a defined concentration is present in the small metal piece.

9. The method of any of claims 1-7, further comprising directing an electron beam at a surface of the metal slug; the act of monitoring the first electrical signal includes monitoring backscattered electrons from an electron beam impinging with the impurity in the metal slug.

10. The method of any of claims 1-7, further comprising replacing the slug in response to determining that impurities above a defined concentration are present in the slug.

11. A method of detecting impurities in a metal slug disposed in an electron beam evaporator during an electron beam metal evaporation/deposition process, comprising:

depositing metal obtained from a small metal block during an electron beam metal evaporation/deposition process on a semiconductor wafer in a vacuum chamber of an electron beam metal evaporation/deposition system;

monitoring an electrical signal provided by an electrode disposed in the vacuum chamber during the electron beam metal evaporation/deposition process;

detecting a change in the electrical signal during the e-beam metal evaporation/deposition process by comparing the electrical signal to a threshold value;

determining that impurities above a defined concentration are present in the small metal slug in response to the electrical signal exceeding a threshold by more than a predetermined amount;

at least one of stopping processing of a semiconductor wafer on the e-beam metal evaporation/deposition system and performing preventative maintenance on the e-beam metal evaporation/deposition system is performed in response to determining that impurities above a defined concentration are present in the small metal slug.

12. The method of claim 11, further comprising inspecting a semiconductor wafer processed in the e-beam metal evaporation/deposition system upon detecting a change in the electrical signal.

13. The method of claim 12, further comprising reprocessing the semiconductor wafer processed in the e-beam metal evaporation/deposition system upon detecting the change in the electrical signal.

14. The method of any of claims 11-13, further comprising electrically isolating the electrode from ground.

15. The method of any of claims 11-13, wherein performing preventative maintenance includes replacing the small metal block.

16. The method of any of claims 11-13, wherein monitoring the electrical signal provided by the electrode during the e-beam metal evaporation/deposition process comprises monitoring an electrical signal generated by electrons backscattered on the electrode from impurities in the metal slug.

17. The method of any of claims 11-13, wherein the improvement in yield is achieved by reducing the number of semiconductor wafers including photoresist cross-linking during processing in the e-beam metal evaporation/deposition system.

18. The method of any of claims 11-13, wherein the improvement in yield is achieved by reducing the number of semiconductor wafers comprising metal nodules resulting from metal sputtering during processing in the e-beam metal evaporation/deposition system.

19. An electron beam metal evaporation/deposition system comprising:

an electrode configured to be disposed within a vacuum chamber of the e-beam metal evaporation/deposition system and isolated from ground, the electrode configured to be disposed to have an unobstructed linear path between a portion of the electrode and a surface of a small metal block during operation of the e-beam metal evaporation/deposition system, the electrode further configured to be disposed to not obstruct the linear path between the surface of the small metal block and a wafer disposed for processing in the e-beam metal evaporation/deposition system;

an electrical meter coupled to the electrode, the electrical meter configured to: monitoring a first electrical signal provided by the electrode during an electron beam metal evaporation/deposition process; and

a controller configured to compare the first electrical signal to a threshold and determine that impurities above a defined concentration are present in the metal slug in response to the first electrical signal exceeding the threshold by more than a predetermined amount.

20. The system of claim 19, wherein the electrical meter is at least one of a voltmeter and an ammeter.

21. The system of claim 19 or 20, wherein the controller is further configured to alert an operator of the change in the signal.