This might be explained by the thermodynamic balance among the three phases of matter;

at lower pressure, liquid will be evaporated at lower temperature

Dear Mohamed, you probably think the depth of the crater is increased 3 times because the recoil pressure of vapor increases also up to 3 times?

It's easy to confirm by measurements of mass loss in the sample. But no such data exists. The mass loss in vapors in EBW is insufficient to create high values of recoil pressure and a such deep keyhole.

Dear Rinat,

one simplified explanation is, that in vacuum the plasma density in welding crater is lowered and photons are penetrated deeply. In atmospheric pressure laser power is transferred to plasma in upper part of the welding crater and then plasma transferred power to the metal sample

Even if we agree with the absorption of laser radiation in plasma (though the vapor temperature is low enough), and with increased evaporation in a vacuum, the vapor pressure inside a crater must be in the range 0,1-1,0 ATM for the formation of a deep crater under the influence of evaporation (it was shown in a number of theoretical calculations). But according to measurements, the vapor pressure inside a cavity in electron-beam welding is only 0,001 ATM.   Hence, the evaporation seems not a main reason for the formation of a deep crater.  Perhaps the capillary effects may be aggravated in a vacuum thus intensify the melt removal from the beam impact zone. 

If you use a laser with a long  radiation wavelength (CO2 laser), to a depth of a crater at atmospheric pressure can affect the radiation screening by plasma of low-threshold air breakdown. Especially it should be marked in the processing of metals. See the books of Yu. P. Raizer.

Evaporation in the  bottom of the crater is only one part of reason for concrete pressure in the crater. As you mention, the pressure in the crater orifice and out of the crater is much more less in vacuum. So plasma density in the upper part of the crater is much times less in the case of vacuum, than in atmospheric pressure out of sample. And shielding of laser radiation penetration is an other one-less in the case of vacuum in the technology chamber and higher in the case of atmospheric pressure around it!.

 For K. N. Firsov. The low-threshold air breakdown for CO2 laser beam near metal surface corresponds to power density in the range 10-100 MW/cm2 and limits the excessive increasing the power density for laser welding. So, the power density for typical welding conditions is in the range 1-5 MW/cm2.   Thus we cannot take into account the low-threshold air (vapor) breakdown.

For Georgi Mladenov. This explanation is acceptable as one of possible a phenomenological picture that still (after 40 years!) requires consistent quantitative evidence and additional arguments against some contradictions. For example, according to experiments, the threshold of deep crater formation has not been accompanied by the increasing of evaporation. So there is no strong link between keyhole-effect and evaporation. As it is shown by many assessments the keyhole-effect could be the result of coupled capillary effects and which probably amplified in a vacuum. So it can be another picture.

The problem was intensively studied by Remy Fabbro and Koji Hirano.

The latest paper on this topic  http://scitation.aip.org/content/lia/journal/jla/27/2/10.2351/1.4913455

To Grigory Ermolaev. This article is devoted to the explanation (from the point of evaporation model) the causes of reducing the differences in welding depth in atmosphere and in vacuum at high welding speeds. That is one of possible explanation, if not pay attention to the contradictions of the evaporation model with other experiments.

This article you can find on RG https://www.researchgate.net/publication/273629767_Explanation_of_penetration_depth_variation_during_laser_welding_under_variable_ambient_pressure

Article Explanation of penetration depth variation during laser weld...

Rinat,

you mention evaporation, understanding total evaporation losses. In the my explanation we expect change of distribution of evaporation due to change of absorbed energy distribution- a concentration in a spot in the crater bottom has increased evaporation and on side wall of the crater- a decrease of the evaporation!

Of course, the intensity of evaporation depends on the angle of incidence of the beam on the metal surface. But this fact does not help us to understand the reason of cavity formation at the threshold conditions (the threshold for the effect of deep penetration of iron or steel - 0.5 MW / cm2). According to the evaporation model the cavity formation is due to allegedly sharp increase of evaporation. But in reality, there is no any sufficient increased evaporation at this point.

To Georgi Mladenov.

Laser welding with deep penetration is accompanied by the formation of pores in the root part of the seam. In accordance with evaporative model the pores are formed just in that place of seam weld where exist a high enough excess vapor pressure to provide the depth of penetration. But experiments show that the pores are filled with shielding gases. The protective gas is not able to get into the pores with the overcoming a large vapor pressure inside of them. This fact also indicates that there is no such a large pressure in the root part of seam where the pores are formed.

Dear Rinat,

I do not understand all of your arguments!

- How you know, that at threshold conditions in which is observed deep penetration there is no any sufficient increase evaporation in the bottom of the welding cavity? In the case of electron beam, in this point a self-focusing of beam by ion(gas)focusing is occurs. So in the laser welding some non-linear phenomena at photon transfer through crater can provide similar self-focusing of the beam-and a big increase of absorbed power density and evaporation.

-Pores in the root part of seam are generated after closing of the crater by molten metal.The observation of shielding gas in the pores really demonstrate, that not only evaporation is involved in mechanism of deep penetration.Plasma formation and mentioned probably self-focusing of laser beam in conditions of deep penetration are different for laser beam propagation through various gas mixtures.

To Georgi Mladenov ( I do not understand all of your arguments!)

A bit of history.

Historically, the evaporative mechanism of melt removal by the laser induced vapor recoil pressure was first proposed and justified for the beam power density of 100 MW/cm2 in 1973 (Batanov V A, Fedorov V B). This article makes no mention about of laser welding with the characteristical power density of about 1 MW/cm2. Simultaneously, S I Anisimov developed a theory of evaporation based on the representations of continuum mechanics. You also will not find any mention about the laser welding in that book, since the theory is focused on a very high power density (1 GW/cm2).

 In 1976, there were published the theoretical articles (Andrew & Athey, Klemens), in which a model of penetration based on the assumption that the hydrodynamic processes in welding with deep penetration are determined by the action the recoil pressure of vapor on melt removal, has been proposed. Apparently, the credibility of these works and the common desire to develop a theory of the laser welding process quickly was so great that the assumption became considered as allegedly established fact very soon. It is difficult now to explain by otherwise why evaporation model welding was not supported by reliable experimental evidence. In fact - the lower border of application area for the evaporation model of melt removal was lowered to 1 MW / cm2 without the pilot study.

Then the experiments began to appear with the results in contradiction with the evaporation model. Unfortunately, they were left without proper attention. But to no purpose. We all lost the time. And today it is very difficult to overcome the stereotype of thinking which was built during four decades. This maybe explains why it is so difficult sometimes to understand my arguments that I find very simple.

 I managed to find a number of experimental facts that contradict the evaporation model of deep penetration (some of these facts you can find in my article, some of them I hope to discuss in my future publications). It was not possible to find the experiments that conclusively confirm this model.

This contradictions raise doubts about the use of evaporation model for welding.

 I suggest thermocapillary model of deep penetration, which is based on experimental data and accurately describes them. This model with high accuracy meets all known experiments on different directions of laser-matter interaction. From my point of view that is the best confirmation.

In general my point of view is following. The understanding the real hydrodynamic mechanism of welding must be supported by appropriate experimental facts but the conclusions of computational and theoretical work  without experimental verification have to be used with great care.

 As for the influence of self-focusing beam on metal evaporation in welding, it should be borne in mind:

- Experimental measurement of parameters of evaporation cannot depend on assumptions about the presence or absence of self-focusing of the beam by welding (what is the reason to consider the spillover effects which may influence on evaporation for the very weak evaporation? ) ;

- We know that it is requires much higher power density (than in laser welding) and much denser plasma for the manifestation of the self-focusing of laser beam.

Sorry for a trivial question, but can the decreased heat deposition into metal be explained by convection cooling in air/gas, while in vacuum, the only mechanism (beyond conduction) is radiation - which is much less efficient??

I have studied this issue for many years. There were also many studies that have all made different partial responses. You will find a review of these studies in the article that was reported by G. Ermolaeev earlier in this discussion, and references attached to it. To answer the original question by R. Seidgazov, and trying to be brief, I will say that the improvement of the penetration depth is mainly the result of two phenomena: the first one is the decrease in the evaporation temperature (which is the order of 1000K) when the ambient pressure is reduced. This is of course a result of the equilibrium between the metal vapor and liquid (the well-known Clausius Clapeyron law). It is easy to show (using heat balance equations) that when the evaporation temperature decreases (which corresponds to the temperature of the walls of the keyhole), the incident laser power can be distributed over a greater depth.

The second effect explaining the increase of the keyhole depth is due to the decrease of the screening processes of the laser radiation through the metallic vapors (inside of the keyhole and externally, in the vapor plume) generated by the evaporation process: When a CO2 laser is used (10.6 micron wavelength), the laser radiation is absorbed by the effects of Inverse Bremsstrahlung, because the metal vapor is hot and ionized, and the wavelength is large (see reference 4 the aforementioned paper). When using the present solid state lasers (1.06 micron wavelength), absorption by Inverse Bremsstrahlung becomes negligible, but it is then the scattering processes (Rayleigh and Mie) that appear on the aggregates, clusters or micro-droplets inside the vapor, which become dominant and cause the propagation of laser radiation to be limited. When the ambient pressure decreases, such adverse effects decrease because of the decrease of the vapor density (electronic, for CO2 lasers or aggregates for solid state lasers), and therefore we gain a deeper penetration keyhole.  

I can also tell you about the limitation of the observed increase in the keyhole depth with the reduction of ambient pressure or the increase of the welding speed. These results are the subject of a paper that I will present to the next Icaleo Conference in October (I can send it to those who are interested, but I will also put it on RG).

To Remy Fabbro.

Thank you for your clear and concise vision of the challenge in terms of the evaporation model of welding. You are one of the leading researchers in this field. And your opinion means a lot to a scientific audience. Therefore, I would like to know your opinion, why there is no reliable experimental evidence of the evaporation model? Why they are no referenced and why they are undetectable?

I wanted to know your opinion concerning of the extremely low vapor pressure inside a keyhole, which is determined by direct measurements and confirmed by the other indirect evidences and which contradicts to the evaporation model?

As it is known, to form the deep keyhole under the influence of recoil pressure of vapor this value should be in the range of 0.1-1 atm. However:

1. Direct measurements show the pressure in the deep cavity only 310 N / m2 when welding Armco iron by electron beam of 7 kW (Verigin AM, Erokhin AA, BA Shavyrin. N., Reznichenko VF 1980).

2. When welding Al-alloy electron beam capacity 1,1 ÷ 3,85 kW pressure inside a keyhole is too low - in the range of 19 - 570 N / m2 (Bondarev AA, Voropai NM 1974)

3. Inside the pores which are formed in the root portion of the seam (i.e., in the region where exist the maximum vapor pressure), gas detectors indicate the presence of protective gases. How the shielding gases with low pressure can overcome ostensibly a very high vapor pressure to reach the root of seam and to appear inside a pores?

I think it is very interesting to know the opinion of the researcher, who wrote:

"It is clear that a complete understanding of the hydrodynamic phenomenon involved in the melt pool is essential for a real mastering of welding process. Today the action of the main driving forces emanating from the keyhole is not completely understood and we are rather far from a precise knowledge of the melt pool behavior. "- Fabbro R., Hamadou M., Coste F, 2004

To Rinat Seidgazov

It is easy to answer your question about the level of pressure inside the keyhole. Consider a keyhole in a quasi-steady state (which is not the case in reality, the latter being the subject of numerous instabilities, I can come back later), it stays opened only if the internal pressure generated by the evaporation process prevents him from closing. The pressures which tend to close it are the ambient pressure, which is transmitted through the liquid, and the closing pressure Pc defined with the surface tension (sigma) and the average radius R of the capillary (that is of the order of the radius of laser spot). As Pc = sigma/R, (sigma about 2N/m, and R typically 0.2 mm) it is seen that Pc is of the order of a few kPa (or N/mm2). Therefore the internal pressure is equal to the sum of the ambient pressure and Pc.

When under vacuum conditions, inside the keyhole there is only an evaporation  pressure equal to Pc that maintains it opened; Which is of the order of magnitude of the values you mentioned in 1 and 2 (which I would like to know how they were measured; I do not know these two quoted works). When the ambient pressure is the atmospheric pressure, the inside of the capillary has a pressure slightly greater than one atm. It is this internal pressure that varies with the ambient pressure that will define the evaporation temperature (by Clausius-Clapeyron law), and so its decrease with vacuum.

As regards the presence of pores, as shown previously, as there is a small pressure difference between the inside and outside of the keyhole (Pc), because of the capillary instabilities (due mainly to a inhomogeneous deposit of laser energy along the keyhole walls), one can understand that the external gas can be sucked because of these oscillations and the resulting very complex movements of the liquid bath. These instabilities of the melt pool are very easily seen using visible range or X-ray radiographies fast videos.

To Remy Fabbro.

I would like to give a quote from your publication:

“The different pressures that are acting on this surface are the recoil pressure Pr due to the vaporization process, which open the keyhole, and the two closing pressures: the surface tension Ps (=2.σ/D, σ surface tension and D local beam diameter) and the dynamical pressure Pdyn induced by the high velocity melt flow around the keyhole coming from the front keyhole wall. Typically

Pdyn=0.5 ρmV2melt , where ρm denotes to melt density.” (R. Fabbro, K. Chauf -2000)

Let us estimate the dynamic pressure value.

In experiments (Eriksson I., Powell J., Kaplan AFH. -2013) it was found that the value of the melt flow reaches V ~ 7,5 m/s (stainless steel, 6 kW beam power , focal spot size of 0.9 mm and a power density of ~ 1 MW / cm2). The corresponding dynamic pressure is up to 2 x 105 N / m2 (or 2 atm), and thus is the primary quantity in the balance of pressures.

If you believe to evaporation model of welding and such a rapid melt flow arises as a result of the recoil pressure vapor, then the value of the vapor pressure (as an overpressure relative to ambient pressure) should be not less then the dynamic pressure of the melt, that is, of 2 x 105 N / m2 (or 2 atm). But in reality (as shown by experimental data), it is less than 103 N / m2.

So we have the contradiction. The conclusion should be made - the recoil pressure of vapor is not a main reason of melt flow.

What is the real reason in this case?

My publications are dedicated to this issue. Such value of melt flow velocity can be reached due to the action of thermocapillary forces with the shear structure of flow. For the conditions of experiment mentioned above the estimation of thermocapillary flow rate using thermocapillary model of deep penetration yields a value of 7.1 m/s. This value very close to the measurements mentioned above, and this fact confirms the thermocapillary mechanism of melt flow during deep penetration welding.

I think that some capillary effects are amplified in a vacuum. Much points to this.

I am attaching the file (description for the invention) with the scheme of measurement of pressure inside a keyhole in the experiment 1 referred earlier. The measurement is carried out through the side opening (4) connected to a tensor or a piezoelectric transducer (5). The sectional area of the auxiliary hole is greater cross-sectional area of the electron beam at the control level.

To Rinat Seidgazov

There is no contradiction between the data you give. You only forget that these different experiments you mentioned are realized under very different conditions of welding speed and ambient pressure and so the pressures inside the keyhole (KH) are very different.

Let me recall the evolution of the keyhole geometry as a function of welding speed (this is detailed in several of our previous publication ( for example “Dynamic approach of the keyhole and melt pool behavior for deep penetration Nd-Yag Laser welding”). At low welding speed the KH is quite vertical, with strong instabilities along the KH walls. Laser power is distributed more or less homogenously along these walls and consequently the absorbed intensity is low. The velocity of the melt flowing around the KH is low, which results of the low welding speed and the corresponding low absorbed intensity. For this situation, as explained in my last response, the pressure P0 inside the KH is low. P0 is equal to the sum of ambient pressure Pa and the closing pressure Pc  (the dynamic pressure, scaling with Vmelt2  is negligible here). So P0 depends strongly of the ambient pressure. As said previously, under vacuum conditions, P0 is about Pc which is low (typically around kPa, which agrees with your quoted experiments 1 and 2 (diameters of KH to be precised?) ). And if Pa = 1 atm, Pa is slightly greater than 1 atm. It is only the difference between P0 and Pa (so equal to Pc) that makes the vapor plume expelled from the keyhole.

Now if the welding speed is increased, as also estimated in my previous papers, the inclination of the keyhole front wall (KFW) increases. So the absorbed laser intensity and consequently the corresponding surface temperature and recoil pressure Pr increase. For welding speeds of several m/min (depending of course of incident laser power and focal spot), this pressure can be much higher than 1 atm. The vapor is expelled like a jet, quite perpendicularly from the KFW surface. This corresponds to the experiments of Eriksson et al. you mention. In that case the melt flow can be expelled by this piston effect, sideways  and along the KFW, with the velocity of several m/s (meter/second!!) which is very high and corresponds effectively to recoil pressure of the order of few atm. These pressures are much greater than the closing pressure Pc, so Pc does not play anymore a role in the dynamics of the keyhole (at least around the KFW region). It also explains that why, under vacuum (when Pa=0), nothing is changed at high welding speeds! This recoil pressure Pr is not modified by the ambient pressure and as shown by different experiments, there is no increase of penetration depths or surface temperature when ambiant pressure is reduced (all these points are detailed in my next presentation at Icaleo Conference and corresponding submitted paper).

To my point of view, this scheme is very clear. I don’t think one needs thermocapillary forces. At high welding speed thermocapillary forces are negligible compared to the dynamic one generated by these gradient of pressures. However, at low welding speeds, many simulations have shown that they can play a role in the overall dynamics of the melt pool and its geometry, because the recoil pressure is very low.  Also, I cannot understand how their effect under vacuum could be “amplifed”.

Thanks for the reference (but it is in Russian) I can’t read it.

To Remy Fabbro

1. “At low welding speed… the velocity of the melt flowing around the KH is low… (the dynamic pressure, scaling with Vmelt2  is negligible here)”

What values of welding speed are called "low" here?

In the article ("Dynamic approach of the keyhole and melt pool behavior for deep penetration Nd-Yag Laser welding") I find the range (so called "Rosental") with a lowest welding speed (less than 5 m/min = 8.3 cm/c). As I think, that is a welding speed of the most widely used in industry. These values sufficiently close to the value of 10 cm / s, which was used in the experiments of Eriksson et al., in which the melt flow rate of 7.5 m/s is registered. The corresponding dynamic pressure is about of 2 atm and cannot be ignored.

To justify the neglecting of dynamic pressure, its value should be a lot less then pressure inside a keyhole (103 Pa) and should be about 102 Pa. These values correspond Pdyn flow velocity with values less than 0.15 m/s, which are observed at the beam power density less than 0.1 MW / cm2. This is considerably less then threshold value for keyhole formation (for steel 0.5 MW / cm2). Under these conditions, thermocapillary flow has the structure of the vortex flow with closed streamlines and with a viscous friction between the layers of counterflow (shown in attached figure).

In reality, the melt flow rate in the beam impact zone during deep penetration welding is higher than 5 m/s and generates a dynamic pressure nearly 1 atm.

2. "The vapor is expelled like a jet, quite perpendicularly from the KFW surface… In that case the melt flow can be expelled by this piston effect, sideways  and along the KFW, with the velocity of several m/s (meter/second!!) which is very high and corresponds effectively to recoil pressure of the order of few atm."

It is cannot be concluded that the recoil vapor pressure reached up to several atmospheres from the fact of observation flowing vapor jet perpendicular to the surface of a keyhole front wall (KFW). For the reaching by vapor pressure of several atmospheres, the velocity of the vapor should be close to the speed of sound in the vapor (about 800 m/s). But very low flow rates of vapor (10-50 m/s) are recorded in the experiments, indicating the weakness of evaporation.

3. "I don’t think one needs thermocapillary forces. At high welding speed thermocapillary forces are negligible compared to the dynamic one generated by these gradient of pressures. However, at low welding speeds, many simulations have shown that they can play a role in the overall dynamics of the melt pool and its geometry, because the recoil pressure is very low."

Of course, high-speed welding has its own features. Your article is dedicated to this study in which you are considering a welding speed up to 20 m/min. 

And also the thermocapillary forces have a natural limit to the rate of the melt removal depending of temperature gradient on the beam impact zone. The thermocapillary removal of melt can reach 10-15 m/s (for steel) and provides a welding speed up to 15-20 m/min. The crisis of the beam power dissipation on the front wall of the cavity with the thermocapillary melt removal mechanism emerges at these values. This crisis is overcome by the increased evaporation which accelerates the melt flow  by the vapor recoil pressure.

Your experiments demonstrate how this crisis emerges and then overcome. And perhaps it is no coincidence that analysis of the features of hydrodynamics of the melt in your experiments is limited in welding speed to 20 m/min.

4. "I cannot understand how their effect under vacuum could be “amplified”."

This question has not been studied yet. However, there is a few experiments, which indicate the possibility of a significant increase of some capillary effects in a vacuum. This may explain the reason for increasing the depth of penetration in a vacuum, and explain the effect that you saw in my article.

These experiments provide a basis for improving the capillary model of deep penetration welding.

A bit of history.

I attach the article (in English) where the evaporation mechanism was firstly proposed as the melt removal induced by vapor recoil pressure with the beam power density of 10 MW/cm2   (Batanov V A, Fedorov V B - 1973).

Many thanks to all who have expressed their views and participated in discussions. Special thanks to Georgi Mladenov and Remy Fabbro. Despite the fact that today we have not convinced each other, the discussion was essential.

Rinat,

My activity is in field of EB welding. Nevertheless laser welding is interesting subject for our  EB technology community. In first days of June 2016 in Varna, Bulgaria will held the next 12 International conference on Electron beam technologies EBT'2016. I invited you to joint us and to do an Invited lecture on laser welding. My e-mail is [email protected]. Please tray to prepare such report which will be of interest to our participant and hope you will find a synergy in discussions with EBT researchers.  Paper on laser welding were presented in former EBT conferences-for example from Kiev KPI&Paton electrical welding Institute team. We have participants from more than 10-12 developed countries from all the word-see the site of EBT'2014 conference in Varna. Prof. Georgi Mladenov

Dear Georgi Mladenov, heartfelt thank you for the invitation to such a representative conference, the chairman of which you are. Of course, I'd much rather spent the summertime  on the Bulgarian Black Sea coast with a lot of sunny pleasure and interesting personal contacts. But there is a feeling that all the hopes and dreams will be broken on the procedures for obtaining grant.

Dear Rinat,

Our conference have not sponsors, but let we be in touch in the first part of the next year.

I will give all need letters to you, if you apply for any grant, and will  help you to overcome partly the financial cost of your stay in Varna, utilizing hotel with all included in the price for accommodation and  omitting (waive) the Conference fee,too.

Prof..G.Mladenov

To Rinat Seidgazov.

Is it possible to obtain the copy of articles measuring directly the pressure inside of the keyhole? Or at least to get more complete reference? I was not able to find them.

I found one of these articles in my library. It is 1986 in Russian. I could not find similar experiments among the English-language articles. I hope you will be able to organize professional translation.

To Remy Fabbro/

Sorry to return to the discussion. Just a question appeared after reading the publication, where you interpret the effect.

If, as you think, the keyhole is formed as a result of the action of high vapor pressure, the only reason for increasing the channel depth can be the increasing of the vapor pressure inside a keyhole.

 In this case, it is necessary to recognize (following the logic of the evaporation model) that during welding under vacuum, the vapor pressure inside a keyhole increases, because a sharp increase of the welding depth is observed (experimental fact).

But it seems that you wanted to convince us to the opposite, insisting that the pressure inside a keyhole will decrease when its depth increases?

A Plasma (an ionic gas) is produced during Laser Welding, especially at high power levels, due to ionization by the laser beam. A plasma can absorb and scatter the laser beam and reduce the depth of penetration significantly. It is therefore necessary to suppress the plasma. The shielding gas for protecting the molten metal can be directed sideways to blow and deflected sideways to blow and deflect the plasma away from the beam path.

i.e.

decrease of the vapor density due to reduced pressure.

According to your guess, the vapour density in a vacuum decreases. With this, I can agree.

But I draw your attention to the fact that according to the generally accepted mechanism of keyhole formation by the action of vapour pressure, a decrease in the vapour density should also lead to a decrease in the penetration depth. But in fact, everything happens the other way around - the penetration depth increases in a vacuum significantly! This means that the mechanism of keyhole formation is not related to the vapour density in reality. I also agree with that. But the mechanism of the keyhole formation is found to be completely unexpected.

It is view point of material and evaporation temperature, which is reducing due to low pressure.

If the penetration depth increases in a vacuum due to the vapour pressure, then the vapour pressure should paradoxically increase (perhaps due to reducing of boiling temperature to low pressure) with a decrease in the external pressure. But such a paradox (vapour pressure increase with the decrease of external pressure) seems not possible.

I am not an expert of laser welding. Maybe I am giving a wrong answer. Vacuum means partial or no atmosphere. No heat dissipation in the atmosphere. Hence, more heat will be transferred in the welding zone to cause more depth of weld. This may be a minor reason, I don't know.

Lalit, thank you for your version. But the effect of absorption of laser radiation in the atmosphere is insignificant and cannot affect the penetration depth, changing it three times.

Hi Lalit,

laser welding is a rather fast process. Therefore the heating of the substrate due to the missing atmosphere can not explain the effect. And in terms of energy loss around and in the melt zone, this is governed by radition and not convection at these elevated temperatures.

Dear Doctor Rinat:

The penetration depth of laser welding had a great association with the shielding affect of plasma plume which could be dramatically reduced under a vacuum environment. Our group had carried out relevant in this direction and some related results were published as attached files. I hope they could help you.

Dear Xia.

Thanks for your interesting articles. They are very helpful to me. However, you probably agree with me that there does not yet answer the question about the reasons for the increase in penetration depth in a vacuum. You have shown very well that the laser power absorbed by a sample can increase by about 10% in a vacuum. But this cannot explain the increase in penetration depth by 4-10 times!

Dear Rinat.

I think the two main reason are the decreasing shielding effect for laser beam due to weaker metal vapor and another point that boiling temperature decreases under lower ambient pressure, so more surfaces can exceed the new boiling temperature then, stronger recoil pressure strengthen the driving force of keyhole formation.

Dear Roulin.

Thanks.

Publications are known that discuss these ideas. However, they are not convincing and here's why.

1. Due to the increase in the transparency of the plasma in a vacuum, the laser radiation power absorbed by the metal increases by only 10%. But the penetration depth increases by 4 times! That is, the effect of plasma transparency cannot explain such an increase in the penetration depth.

2. To clarify the effect of the boiling point, let's answer the question:

What happens to the pressure in the penetration channel when the ambient pressure decreases? There are two possible answers - the pressure increases or it decreases.

If in your opinion, the pressure in the penetration channel decreases in a vacuum, then it leads to a decrease in the boiling point and an increase in evaporation. I agree with that. However, there is a fact that the depth of penetration in vacuum increases by a factor of 4 or more compared to atmospheric conditions. Can we say that the vapour pressure in a vacuum is much higher than atmospheric? If you agree with this, then you have already abandoned the hypothesis about the significant effect of the boiling point on the penetration depth, and

should explain the paradoxical increase in pressure inside a cavity with a decrease in external pressure!

I think this is a dead end.

If your opinion is that the pressure inside a cavity is increasing, then the boiling point should be also increasing.

Following this logical physics, my opinion is that these two effects (the effect of the plasma and the change in the boiling point) do not explain the increase in the penetration depth in a vacuum.

Dear Rinat Seidgazov,

According to our recent numerical study as below, the deep penetration in high vacuum could be caused by the effect of recoil pressure, laser scattering and absorption by nanoparticles, where the effect of absorption was dominant over that of scattering in the process parameters applied.

The higher the vacuum around the target

the lower the convection heat transfer coefficients. The lower

the convection heat transfer coefficients the lower the “total” heat

transfer coefficient (this is due to extremely small value of

StefaneBoltzmann constant). In consequence we have the lower

the eigen-values from heat equation and therefore an important

increase of temperature as well as a flat spatial thermal field

distribution and a linear behavior of the time evolution of the same

thermal field.