Experimental and numerical investigation into the non-explosive excavation of tunnels

Quan Zhang,Zhigang Tao,Chun Yang,Shan Guo,Manchao H,Chongyuan Zhang,Huiya Niu,Chao Wang,Shn Wang

a School of Mines,China University of Mining and Technology,Xuzhou,Jiangsu,221116,China

b State Key Laboratory of Geomechanics and Deep Underground Engineering,China University of Mining and Technology,Beijing,100083,China

c School of Resources and Safety Engineering,Central South University,Changsha,410083,China

d Department of Mining and Materials Engineering,McGill University,Montreal,QC,H3A 0E8,Canada

e Institute of Geomechanics,Chinese Academy of Geological Sciences,Beijing,100081,China

f College of Environmental Science and Engineering,Ocean University of China,Qingdao,266100,China

g School of Energy Science and Engineering,Henan Polytechnic University,Jiaozuo,454003,China

Keywords:Instantaneous expansion (IE)Non-explosive Tunneling Smooth tunnel cross-section Excavation-induced damage zone

ABSTRACT The use of explosives is restricted on some important holidays,and the handling of unexploded charge is very dangerous.Therefore,an innovative non-explosive technology called instantaneous expansion (IE)was developed for tunneling.IE,whose components are derived from solid wastes such as coal gangue and straw conduces to realizing the reuse of waste.Moreover,its cost is lower than explosives.Blind guns of IE are easy to treat with water.The IE tunneling method is classified into two categories,i.e.IE with a single fracture (IESF) and IE with multiple fractures (IEMF),which are used to form the tunnel crosssection directionally cross-section and to fragment the rocks inside the cross-section,respectively.In this study,the principle of IE tunneling was elaborated first.Then,tunneling experiments and numerical simulations were performed on IE,conventional blasting (CB) and shaped charge blasting (SCB) in comparison.The experimental and numerical results show that IE achieved the best performance of directional rock breaking and corresponded to the most minor excavation-induced damage zone of the surrounding rock.Besides,the tunnel cross-section created by IE was flat and smooth.Comparing IE with CB and SCB,the over/under-excavation area decreased by 64% and 17%,and the excavation-induced damage zone fell by 26% and 11%,respectively.The range of the loose circle is reduced,which is conducive to improving the long-term stability of the roadway.The research provides a safe and economical tunneling method with excellent application prospects.

1.Introduction

Drilling and blasting is the primary method of tunnel construction presently employed (Yan et al.,2012; Yilmaz and Unlu,2014).However,use of the traditional blasting method can result in over/under-excavation of the tunnel cross-section and damage to a large area of surrounding rock (Saiang,2010) that inevitably affects the stability and integrity of the tunnel.Therefore,precisely controlling the direction of the blasting cracks (He et al.,2021),obtaining a smooth excavation contour (He et al.,2003),and reducing the disturbance to surrounding rock are critical technical issues in tunnel excavation.Smooth blasting has been found to address these problems to a certain extent (Hu et al.,2014).However,smooth blasting produces many toxic gases and dust.Shaped charge hydraulic blasting has been proposed to optimize the tunnel working environment provided by traditional smooth blasting.This technology employs water-medium coupling instead of the airmedium coupling typically used in shaped charge blasting (SCB)(Zhang et al.,2021).This method can directionally shape the peripheral contour of the tunnel cross-section,reduce the amount of explosives required,increase the blasthole spacing,decrease the amount of harmful gases and dust generated (Ye et al.,2017),and ensure the safety of blasting.However,explosives are still required whether using conventional blasting (CB),smooth blasting,or shaped charge hydraulic blasting.Although blasting using explosive has many advantages,such as high power and a wide application range,its use is restricted in some holidays,and government approval procedures are complicated in China.

Scholars attempted to develop non-explosive blasting methods to excavate tunnels,considering the shortcoming associated with explosives.These methods include liquid CO2blasting,plasma blasting,and soundless chemical demolition agents.The principle of liquid CO2blasting is that CO2can be transformed from its liquid state to gaseous state in a very short time (40 ms),resulting in a volumetric expansion over 600 times the original liquid volume.The rapid increase in pressure generated by this expansion opens an energy-dissipating sheet that can fracture the rock (Cao et al.,2017).This generates no sparks and produces low vibration (Li et al.,2020).However,it is less powerful than explosives,and its operation is tedious.It is also challenging to achieve a remarkable directional rock-breaking effect.Plasma blasting uses electrical energy to excite an electrolyte solution,turning it into a plasma that vigorously oscillates to produce high temperature and high pressure (Cho et al.,2016; Riu et al.,2019).The associated cluster movement of ions and shock waves is formed quickly and used to break the rock(Ikkurthi et al.,2002).It has the advantages of simple operation,no toxic gas production,and low vibration generation.However,it is associated with high economic costs,complex equipment,and complicated power supply requirements.Soundless chemical demolition agents are composed of calcium oxide,silicate,and organic and inorganic additives (Hinze and Brown,1994; Natanzi et al.,2016; Laefer et al.,2018).When the demolition agent reacts with water to form calcium hydroxide,its volume increases,and the pressure surges instantaneously.Rock fractures when the generated pressure exceeds the tensile strength of the rock (Cho et al.,2018).Although the entire process generates no sparks,no toxic gas,and slight vibration and noise (étkin and Azarkovich,2006),its response time is too long (4-24 h).

A novel non-explosive rock-breaking technology called instantaneous expansion (IE) was developed in this study.First,the principle of IE tunneling was introduced.Then,an experimental study on CB,SCB and IE tunneling was carried out for comparison.Finally,a numerical study on these three methods was conducted to compare the directional rock-breaking performance and the size of the excavation-induced damage zone.

2.Principle of IE tunneling

2.1.IE technology

IE consists of four parts:a special splitting agent,a splitting tube,a coupling medium,and a current initiating device (Fig.1).The special splitting agent is composed of five materials(Fig.2),among which reducer,oxidant,additive,adhesive and catalyst account for 30%-65%,30%-40%,5%-30%,1.25%and 0.5%-3%of its composition by weight,respectively.The special splitting agent,which is nonexplosive,is safer than explosives (Zhang et al.,2022).In addition,its cost is about 1/3 of the price of explosives.The splitting tube falls into two types,i.e.IE with a single fracture (IESF) and IE with multiple fractures (IEMF).IESF produces two symmetrical cracks in a direction.The tube has two rows of energy-focusing holes,and the angle between which is 180.IEMF generates multiple(at least 3)cracks,and it does not have energy-focusing holes.Tunnel excavation employs IESF for directional forming of the tunnel cross-section and IEMF for fragmentation inside the tunnel cross-section.The coupling medium,which can be saltwater,asbestos and sand,helps enhance the power of IE and optimize the working environment.The current initiating device consists of an initiating head,a current initiator,and a wire,and is used to excite the special splitting agent.It is noteworthy that IE does not require a detonator to trigger,which represents a considerable improvement in safety over the methods employing explosives.

Fig.1.Schematic diagrams of IE components:(a)IESF,and(b)IEMF.1-splitting tube;2-special splitting agent;3-coupling medium; 4-current initiator;5-wire;6-initiating head; 7-energy-focusing hole.

Fig.2.Weight percentage of the five material components of the special slitting agent(Zhang et al.,2020).

2.2.Mechanism of directional rock fracture initiation and propagation

In IESF,a current initiator triggers the special splitting agent,which changes from a solid to gaseous CO2and water vapor(Fig.3).In the limited space of the blasthole,the gas pressure increases almost instantaneously,forming a high-density,high-speed,and high-pressure energy concentrating jet in the selected direction.The rock then produces an initial guiding crack under the action of this energy concentrating jet that provides a directional effect for the further expansion of the crack.Once the gas enters this initial crack,it has a“gas wedge”effect,resulting in a stress concentration in the direction of energy accumulation that continues to expand and extend in the direction of the initial crack.The crack continues to grow until the stress applied by the jet is less than the rock’s Type I fracture toughness.In the other directions,the rock rarely breaks due to the protection provided by the splitting tube wall.Therefore,IESF provides energy generation and directional fracturing while protecting the rock in the other directions,effectively controlling the damage to the surrounding rock.

Fig.3.Rock crack evolution process: (a) IESFs into the blastholes,(b) Initiating IESFs,(c) Crack initiation,(d) Crack propagation,and (e) Crack coalescence.rd is the radius of the blasthole,a is the generated crack length,p is the gas pressure in the blasthole,and p′ is the gas pressure in the crack.

The vertical stress σvand horizontal stress σHin the excavated tunnel cross-section are respectively given by

where γ is the bulk density of the overlying strata of the tunnel,H is the depth of the tunnel,and λ is the lateral pressure coefficient.

As shown in Fig.4,when the distance between two blastholes exceeds two times the blasthole diameter,the stresses generated in each blasthole do not interact.In the excavation process of a roadway tunnel,the spacing between peripheral blastholes is typically much larger than twice the blasthole diameter.Therefore,the directional initiation model of the crack can be calculated using a single blasthole.According to the theory of elasticity,the stress around a small hole can be expressed as

where σris the radial stress at a distance r from the center of the hole; and σθis the hoop stress at an angle θ with respect to the x axis.

The stress concentration acts along the centerline of the blasthole.At the wall of the blasthole (r=rd),this stress reaches its peak value,given by

The hoop stress is the smallest in the direction of energy application (θ=0 and 180 ).This is therefore where the hole is most likely to crack at hoop stress of

Fig.4.Mechanical model of directional rock breaking.

The gas pressure generated by the splitting agent in the direction of energy application is given by

where η is the expansion coefficient,ρ is the gas density,and v is the gas velocity.

When multiple blastholes along the contour of the tunnel are triggered simultaneously,the gas pressure generated by the splitting agent and hoop stress caused by the in situ stress on the hole wall should not be less than the dynamic tensile strength σtdof the rock,i.e.

Thus,combining Eqs.(4)-(6),we have

After the rock cracks at the blasthole wall,it propagates along the direction of energy accumulation due to tensile fracture failure(Fig.3d),which represents an open type(Type I)crack propagation.Therefore,the stress intensity factor generated at the crack tip must not be less than the dynamic fracture toughness of the rock,i.e.

The stress intensity factor (KI) at the crack tip is the superposition of the stress intensity factors produced by the in situ stress(KSI) and gas pressure (KGI),i.e.

The stress intensity factor of in situ stress at the crack tip is

where F is related to σH/σvand a/rd.

If the gas pressure p′in the crack is a function of the gas pressure p in the blasthole,then we have

The stress intensity factor of the gas pressure at the crack tip is

The stress intensity factor at the crack tip is

Thus,the directional propagation of cracks should meet

2.3.Directional fracture mechanism of the tunnel cross-section

The IESFs are installed in blastholes drilled in the tunnel face following the desired contour of the tunnel cross-section.Multiple blastholes are triggered simultaneously,forming a superimposed stress field.Because the IESF concentrates its energy solely in the direction of the energy-focusing holes,the rock can be fractured with less charge.Rock is a quasi-brittle material,and its tensile strength is far less than its compressive strength,thus,it is more prone to rapid tensile failure.The IESF applies this principle to initiate directional tensile failure in the rock,reducing the stress required for rock fracture as well as any disturbance to the surrounding rock.When the electric current is triggered,the highpressure gas in each blasthole causes the rock to crack.As the tensile stress is the largest along the line connecting adjacent blastholes,tensile failure is induced along this line,directionally expanding the cracks from each blasthole in the energy gathering direction.After extending to a certain distance,the cracks from adjacent blastholes connect to form a smooth tunnel cross-section contour.Thus,the use of IESF increases the ability to propagate cracks predominantly along the energy-focusing direction while suppressing the generation of secondary cracks in the non-energyfocusing direction.After applying IESF in the blastholes along the contour of the tunnel cross-section (Fig.5),the rock will be directionally fractured along the design tunnel contour,with no cracks generated outside of it due to the use of the splitting tube.Therefore,the over/under-excavation of the tunnel cross-section is reduced,and the stability of the surrounding rock is ensured.Thus,the resulting tunnel section will be smooth,illustrating the excellent shaping effect of IESF excavation.

3.Tunneling experiments

3.1.Engineering geology

A series of tunnels were excavated using various methods in the Linsheng Coal Mine,located in Sujiatun District,Shenyang City,Liaoning Province,China.The annual production capacity of the mine is 1.2 million ton,and the total length of rock and coal tunnel excavation is over 12,000 m.The mine is located in an asymmetrical dipping syncline structure that struck northeast and sloped southwest.The west wing is relatively gentle,generally 20-40.The test site is located in the material lane of the second mining area in the west wing,with a buried depth of 420 m.The shape of the tunnel cross-section is a straight-wall semicircular arch,3.2 m wide by 2.75 m high.The circulation length is 1800 mm,and the lithology of the surrounding rock is limestone.The physicomechanical parameters of limestone are shown in Table 1.

Table 1 Physico-mechanical parameters of limestone.

3.2.Excavation scheme

Fig.5.Effect and mechanism of IESF roadway excavation.

Excavations were conducted using three methods,i.e.CB,SCB and IE,to compare their tunnel excavation capabilities.Three CB excavations were carried out using an uncoupled continuous charge structure to excavate 1800 mm of tunnel footage.The explosive employed was a tertiary emulsion explosive typically applied in coal mines.Each explosive charge was 32 mm in diameter,200 mm long,and weighed 0.2 kg.The explosive charges were detonated by a detonator.The blasthole layout for the CB excavations is shown in Fig.6,and the blasthole parameters and explosive charge structure are provided in Table 2.The cutting holes adopted the oblique cut approach,and a total of six 38-mm diameter cutting holes were drilled into the tunnel face to a depth of 2000 mm at the angles shown in Fig.6 and defined in Table 2.Each cutting hole was equipped with five 0.2-kg charges,making a single-hole charge of 1 kg.A total of 13 auxiliary holes,38 mm in diameter and 1800 mm in length,were drilled in the face parallel to the longitudinal axis of the tunnel around the cutting hole pattern.Each auxiliary hole was equipped with four 0.2-kg explosive charges,making a single-hole charge of 0.8 kg.Eight bottom holes,38 mm in diameter and 1800 mm in length,were drilled along the bottom of the face.Each bottom hole contained four 0.2-kg explosive charges,making a single-hole charge of 0.8 kg.Finally,15 peripheral holes,38 mm in diameter and 1800 mm in length,were drilled around the periphery of the tunnel cross-section at 400 mm intervals at the angles shown in Fig.6 and defined in Table 2.Each hole contained three 0.2-kg explosive charges,making a total single-hole charge of 0.6 kg.The length of the chemical gun mud,used to ensure an appropriate working environment by reducing the concentration of toxic and harmful gases and dust,was 200 mm,and the length of the yellow mud used to seal each hole was 1000 mm (Fig.7).

Three SCB excavations were conducted to excavate 1800 mm of tunnel footage,using the same blasthole layout used for the CB excavations(Fig.6).The parameters and explosive charge structure for all SCB blastholes are provided in Table 3.The parameters and explosive charge structure of the cutting and auxiliary holes were also the same as those used for the CB excavations,except that the diameter of the peripheral and bottom holes was 48 mm.The larger diameter was required because the explosive charges were first inserted into energy-focusing tubes with an outer diameter of 42 mm and an inner diameter of 36.5 mm,and then placed into the holes (Fig.8).The energy-focusing tube had two rows of energyfocusing holes with an angle of 180.The energy-focusing holes were oriented along the perimeter of the tunnel cross-section.The charge structures in these holes were the same as that used for the CB excavations.The charge of a peripheral hole was 0.6 kg,and the total lengths of chemical gun mud and yellow mud in peripheral holes were 200 mm and 1000 mm,respectively.The charge of a bottom hole was 0.8 kg,and the lengths of the two kinds of mud were 200 mm and 800 mm,respectively.It is noteworthy that in order to reduce gun smoke,dust and toxic and harmful gases,a water pipe was opened prior to detonation to sprinkle the blasting area,and was closed 15 min after the explosion.

Three IE excavations were conducted,each excavating 1800 mm of tunnel footage.The hole layout for the IE excavations is shown in Fig.6,and the parameters and explosive charge/splitting agent structure of these holes are listed in Table 4.Because IE tunneling requires a free surface,the six cutting holes were still blasted with explosives using the same charge structure as that used in the CB and SCB excavations.Thirteen auxiliary holes,48 mm in diameter and 1800 mm in length,were equipped with IEMFs (Fig.1b),each of which had 0.8 kg of splitting agent with a length of 1000 mm.Thus,thesealing lengthof each holewas 800 mm.Fifteenperipheral holes,48 mm in diameter and 1800 mm in length,were loaded with IESFs(Fig.9),each of which had 0.6 kg of the splitting agent with a length of 800 mm.Thus,the sealing length of each hole was 1000 mm.Eight bottom holes,48 mm in diameter and 1800 mm in length,were equipped with IESFs,each of which had 0.8 kg of the splitting agent with a length of 1000 mm.Therefore,the sealing length of each bottom hole was 800 mm.The energy-focusing holes of the IESF were oriented to follow the contour of the tunnel cross-section.

Table 3 Blasthole and charge parameters for SCB excavation.

Fig.6.Blasthole layout (unit: mm).

Table 2 Blasthole and charge parameters for CB excavation.

Fig.7.Peripheral hole charge for CB.1-wire; 2-blast hole; 3-yellow mud; 4-chemical gun mud; 5-detonator; 6-explosive.

3.3.Excavation process

Fig.8.Peripheral hole charge for SCB: (a) Schematic diagram of the charge structure,and (b) Schematic diagram of the energy-focusing tube.1-wire; 2-blast hole; 3-yellow mud;4-chemical gun mud;5-detonator;6-explosive;7-energy-focusing tube; 8-tube wall; 9-energy-focusing hole.

For safety,the limestone on the two sides and the roof of the tunnel were inspected and processed prior to drilling.Additionally,the air pressure pipe and the water supply pipe were checked for leakage.An air-leg rock drill was used to drill the blastholes.The number,location,depth and diameter of the blastholes were drilled in accordance with Fig.6 and Tables 2-4 As the drilling quality directly affects the blasting footage and tunneling effect,the depth,angle and spacing of the blastholes were checked after drilling.To ensure the desired tunnel excavation effect,any rock powder or water in the blastholes was blown out with a pressure fan before loading them with the charges.Then,the IEMFs or explosives were loaded into the blastholes within the cross-section.Next,the IESFs or explosives were loaded into the peripheral holes in the manner shown in Figs.7-9.Wooden or bamboo cannon sticks were used to gently push the IEs or explosives into the blastholes.The explosives were then sealed with chemical gun mud and yellow mud,and the IEs were sealed by yellow mud.Finally,a gun stick was used to tamp the yellow mud to ensure the desired sealing effect,and all blastholes were connected in series for detonation.After the explosions,the rock debris was removed from the tunnel,realizing the excavation of the tunnel cross-section.

Table 4 Blasthole and charge parameters for IE excavation.

Fig.9.Peripheral hole charge for IESF.1-wire; 2-hole; 3-yellow mud; 4-coupling medium;5-initiating head;6-special splitting agent;7-energy-focusing tube.

3.4.Data collection and analysis

The effect of tunnel directional excavation is evaluated by the following parameters: the maximum over/under-excavation depths,the directional forming rate of the tunnel section (DFRTS),the semicircular blasthole remnant rate,and the blasthole utilization rate.

The over/under-excavation depths were measured by a laser profiler and a total station and their maximum values were recorded to quantify the accuracy of the blasting method.The specific measurement method is introduced as follows: first,the central axis of the tunnel was determined with a total station,and a point on it was selected to accurately locate the laser profiler.Next,the start and end measurement angles of the tunnel section and the number of measured points were set in the laser profiler,and the software could automatically complete the measurement of the current section.The shape and data of the measured cross-section could be displayed in real time and compared with the designed standard cross-section to obtain over/under-excavated values.A positive value indicated over-excavation while a negative value indicated under-excavation.

To evaluate the orientation effect of the peripheral contour,the DFRTS was defined as the ratio of the absolute value of the total over-and under-excavated areas to the designed cross-section area,and is expressed as

where ψ is the DFRTS,Sois the over-excavated area,Suis the underexcavated area,and Sdis the designed area of the section.Thus,the closer the DFRTS value to zero,the better the directional shaping effect provided by the excavation method.

Fig.10.Partial topography of tunnel cross-section excavated by CB.

Fig.11.Tunnel cross-section contours (unit: mm) formed by CB after the (a) first excavation,(b) second excavation,and (c) third excavation.

The semicircular blasthole remnant rate,which refers to the ratio of the number of semicircular blasthole remnants to the number of blastholes on the tunnel contour,was determined by visual inspection of the tunnel walls and was used to quantify the over/under-excavated range and directional forming effect of the tunnel.The higher the semicircular blasthole remnant rate,the smaller the range of over/under-excavation.

The blasthole utilization rate,which refers to the ratio of the actual length of the excavation to the blasthole depth,was measured by a ruler,and the measurement was used to quantify the effectiveness of each blasting method.The higher the blasthole utilization rate,the longer the tunneling length.

3.5.Experimental results and analysis

3.5.1.CB excavations

Serious over-and under-excavation was noted,and the crosssection contour was observed to be jagged and uneven (Fig.10) in all three CB excavations.The characteristics of the three excavations are shown in Fig.11.The maximum over-excavation depths of the straight tunnel wall sections were 282 mm,291 mm and 207 mm,with an average of 260 mm;the maximum over-excavation depths of the arched roof sections were 436 mm,485 mm and 450 mm,with an average of 457 mm; the maximum over/under-excavation depths of the straight floor sections was 146 mm (overexcavation),-178 mm (under-excavation) and-192 mm (underexcavation),with an average of 172 mm.The arched roof sections thus exhibited the largest deviation from the intended tunnel cross-section,whereas the straight floor sections exhibited the smallest.Indeed,it was observed that arches were more difficult to form using the CB method than straight lines.The designed area of the tunnel section evaluated in these excavations was 7,583,839 mm2,and three total over/under-excavation areas were 1,614,001 mm2,1,536,843 mm2and 1,833,650 mm2,with an average of 1,661,498 mm2.Therefore,based on Eq.(11),the DFRTS values were 21.28%,20.27% and 24.18%,with an average of 21.91%.

There were 23 blastholes (peripheral holes and bottom holes)around the excavated cross-section,but few semicircular blasthole remnants were on the tunnel walls.After three excavations,tunnel cross-section contours have 2,3 and 2 semicircular blasthole remnants,corresponding to semicircular blasthole remnant generation rates of 9%,13%and 9%,respectively,with an average of 10%.The blasthole utilization rates were 78%,80%and 75%,respectively,with an average of 78%.Note that the uneven tunnel cross-section contour resulting from the CB excavations required a considerable amount of shotcrete in the later stage,which increased the working time and cost of tunnel construction.Thus,tunneling using the CB method was found to be destructive to the surrounding rock,reducing its stability and complicating later efforts to provide support.

3.5.2.SCB excavations

The tunnel cross-section resulting from the SCB excavations was obviously superior to that resulting from the CB excavations,as the problem of over/under-excavation of the surrounding rock appears to have been addressed (Fig.12).The energy-focusing tube concentrated the release of blasting energy along the direction of the energy-focusing holes,which were oriented form a continuous crack creating a smooth cross-section contour.The surrounding rocks were protected in the other directions.Therefore,SCB effectively used the blasting energy,reduced the range within which the surrounding rock was loosened,maintained the integrity of the tunnel cross-section,and ensured the stability of the surrounding rock,making full use of its self-supporting capacity.

Fig.13.Tunnel cross-section contours (unit: mm) formed by SCB after the (a) first excavation,(b) second excavation,and (c) third excavation.

The results of the three SCB excavations are shown in Fig.13.The maximum over-excavation depths of the straight tunnel wall sections were 121 mm,130 mm and 126 mm,with an average of 126 mm;the maximum over-excavation depths of the arched roof sections were 198 mm,141 mm and 173 mm,with an average of 171 mm; and the maximum over/under-excavation depths of the straight floor sections were-87 mm (under-excavation),114 mm(over-excavation) and-97 mm (under-excavation),with an average of 99 mm.The three total over/under-excavation areas were 692,654 mm2,669,722 mm2and 832,841 mm2,with an average of 731,739 mm2,a decrease of 929,759 mm2compared to that for the CB excavations.Based on Eq.(11),the three DFRTS values were 9.13%,8.83% and 10.98% and the average DFRTS was 9.65%,a decrease of 12.26%compared to that for the CB excavations.

Semicircular blasthole remnants were clearly visible on the surface of the tunnel cross-section in a significant difference from the CB-excavated tunnel cross-section (Fig.12).There were 18,19 and 19 semicircular blasthole remnants after the SCB excavations,corresponding to semicircular blasthole remnant rates of 78%,83%and 83%,respectively,with an average of 81%.This represents an increase of 71% compared to that for the CB excavations.The blasthole utilization rates were 92%,91%and 91%,respectively,with an average of 91%,representing an increase of 13%compared to that for the CB excavations.Therefore,SCB effectively controlled the blasting effect to realize a tunnel cross-section that was basically consistent with the design cross-section,reducing the amount of debris generated and shotcrete required,thus improving efficiency while reducing costs.Additionally,it should be noted that this method was quite conducive to tunnel support and stability.

3.5.3.IE excavations

The tunnel cross-section surfaces created by the IESFs in the peripheral holes were observed to be smooth and regular,with well-formed arc and straight sections alike(Fig.14).This indicates a reduced damage range within the surrounding rock and improved stability.As shown in Fig.15,the over/under-excavation depths were smaller and the directional forming of the tunnel crosssection was better when using IE than that when using CB or SCB.For the three IE excavations,the maximum over-excavation depths of the straight tunnel wall sections were 66 mm,115 mm and 114 mm,with an average of 98 mm,which is 28 mm less than that for SCB and 162 mm less than that for CB.The maximum overexcavation depths of the arched roof sections were 119 mm,112 mm and 120 mm,with an average value of 117 mm,which is 54 mm less than that for the SCB excavations and 340 mm less than that for the CB excavations.The maximum over-excavation depths of the straight floor sections were 81 mm,82 mm and 77 mm,with an average maximum of 80 mm,which is 19 mm less than that for the SCB excavations and 92 mm less than that for the CB excavations.The total over/under-excavation areas were 633,495 mm2,641,660 mm2and 539,560 mm2,with an average of 604,905 mm2,which is 1,056,593 mm2less than that for the CB excavations and 126,834 mm2less than that for the SCB excavations.Therefore,use of IE results in less shotcrete being required to meet the tunnel cross-section than explosive blasting,decreasing the required labor and increasing the construction speed.Based on Eq.(11),the three DFRTS values were 8.35%,8.46%and 7.11%,with an average of 7.97%.The average DFRTS of the IE excavations was closer to zero than that of the CB or SCB excavations,indicating that its directional forming effect was better.

There were many semicircular blasthole remnants on the excavated tunnel surface(Fig.14).In all,21,20 and 21 semicircular blasthole remnants were visible on the tunnel surface,corresponding to semicircular blasthole remnants rates of 91%,87%and 91%,respectively,with an average of 90%.The rates were 80% and 9% higher than those for the CB and SCB excavations,respectively.Therefore,IE provided the best directional forming effect and the least damage to the surrounding rock.The blasthole utilization rates were 95%,96%and 97%,with an average of 96%,representing the increases of 18% and 5% over those for the CB and SCB excavations,respectively.Thus,use of IE is able to improve the actual tunneling length,helping to realize safe and efficient tunneling.

Fig.14.Partial topography of tunnel cross-section excavated by IE.

Fig.15.Tunnel cross-section contours (unit: mm) formed by IE after the (a) first excavation,(b) second excavation,and (c) third excavation.

4.Numerical simulation of peripheral hole

4.1.Principles

4.1.1.Constitutive equation of damage

The meso-structure mechanical parameters of the rock follow the Weibull distribution function.The element damage of rock is judged according to the maximum tensile stress criterion and the Mohr-Coulomb criterion(Wang et al.,2016).The maximum tensile stress criterion is expressed as

where F1is the function representing the state of stress,σ1is the maximum principal stress,and σtis the uniaxial tensile strength.The Mohr-Coulomb criterion (Tang et al.,2002) is expressed as

where F2is the function representing the state of stress,σ3is the minimum principal stress,φ is the angle of internal friction,and σcis the uniaxial compressive strength.

According to the elastic damage theory,the elastic modulus of rock gradually decreases as the damage develops,i.e.

where E0and E are the elastic moduli of rock before and after damage,respectively; and D is the damage variable which is written as (Chang et al.,2015):

where εtand εcare the corresponding maximum tensile and maximum compressive principal strains,respectively;ε1and ε3are the maximum and minimum principal strains,respectively; and n is the element damage evolution coefficient.A damage variable closer to 0 indicates slighter damage,and a damage variable closer to 1 indicates more severe damage.

4.1.2.Equation of load in the hole

In the cases of CB and SCB,the rock mass is subjected to the explosion stress wave to produce an initial crack first,and then the explosion gas further promotes crack expansion.The loading equation of the explosion stress wave(Gao et al.,2019) is

where Pdis the dynamic stress under the action of the explosion stress wave; P0is the peak stress which can be calculated by 140e6Q3/2; Q is the explosive quantity; gais the explosive delay rate; and t0is the loading time,which can be obtained by 0.81e-3Q1/3.

The explosion gas pressure(Zhu et al.,2013)can be expressed as

where Pgis the explosion gas pressure,Pg0is a constant related to the peak explosion gas pressure,s is the loading step,s0is a constant related to the loading step,and m is the homogeneous shape parameter.

IESF relies on the expansion and cracking induced by hightemperature and high-pressure gas,rather than the loading of explosion stress waves.The gas pressure loading is also expressed by Eq.(21).Because rock breaking by both explosives and IESF belongs to a solid-gas coupling problem,the multi-field coupling simulation software COMSOL is adopted.

Fig.16.Numerical models for (a) CB,(b) SCB,and (c) IESF excavations.

4.2.Schemes

Three numerical models (Fig.16) of CB,SCB and IESF are all 2 m×1 m in size.Each model has two holes,the spacing of which is 400 mm and diameter is 48 mm.The rock used is the limestoneencountered in roadway excavation,and its physico-mechanical parameters are shown in Table 1.The parameters of the energyfocusing tube are shown in Table 5,and the validity of the modeling method of the energy-focusing tube has been verified(Gao et al.,2019).For all the three models,displacement constraints were applied to the bottom and right side;horizontal in situ stress σ1was applied to the left side,with a magnitude of 15 MPa; and vertical in situ stress σ2was applied to the upper side,with a magnitude of 11 MPa.

Table 5 Parameters of the energy-focusing tube (Gao et al.,2019).

Fig.17.Detonation stress wave and detonation gas pressure generated by CB and SCB.

The difference in these three models is as follows: the detonation stress wave(Pd)and detonation gas pressure(Pg)generated by CB directly act on the hole wall without being oriented by the energy-focusing tube,while those generated by SCB split the rock directionally under the guidance of the energy-focusing tube.The thickness of the energy-focusing tube is 4 mm,and the width of the energy-focusing hole is 8 mm.For CB and SCB,the effect of detonation stress wave was dominant in 0-50 steps,and the effect of detonation gas pressure dominated in 50-100 steps (Fig.17).IESF,which does not generate any explosion stress waves,depends on the detonation gas pressure to fracture the rock directionally under the guidance of the energy-focusing tube.For IESF,the effect of explosive gas played a dominant role in 0-100 steps (Fig.18).

4.3.Results and analysis

The numerical results of the excavation effect of the peripheral holes using the three excavation methods,i.e.CB (Fig.19),SCB(Fig.20)and IESF(Fig.21),are to compare the directional formation effect of the surrounding contour of the roadway.After CB(Fig.19),the rock first produced multiple cracks under the action of detonation stress wave(0-50 steps),and then these cracks continued to expand under the action of detonation gas pressure(50-100 steps).Ultimately,multiple crack surfaces appeared near the two holes.The area around the hole was damaged most seriously,and the area farther away from the borehole underwent slighter damages.After SCB (Fig.20),the rock cracked along the direction of the energyfocusing hole under the guidance of the energy-focusing tube.In the last stage of the explosion stress wave effect (50 steps),fragmentation occurred in a small area around the hole wall.Subsequently,the detonation gas pressure promoted further expansion of these cracks.The area around the hole wall was damaged to some extent,but the damage range was much smaller than that of CB.Moreover,the cracks expanded directionally along the direction of the energy-focusing hole as a whole.After IESF (Fig.21),cracks were generated,expanded and connected directionally under the action of high-pressure gas.The area around the hole wall was scarcely damaged,and the damage degree was lower than those of CB and SCB.In summary,IESF produced a better directional effect than SCB,which was conducive to maintaining the original strength of the surrounding rock,reducing the range of loose circles,and promoting the stability of the roadway.

Fig.18.Gas pressure generated by IESF.

5.Discussion

5.1.Excavation-induced damage zone

To compare the surrounding rock damages caused by CB,SCB and IE,one section was selected for the acoustic test after each method was used for excavation.Each section has eight acoustic test holes,numbered S1-S8 (Fig.22).

The wave velocity-depth curves of all the acoustic test holes are illustrated in Fig.23.The abscissa is the depth from each measuring point in the acoustic test hole to the tunnel wall,with the depth at the orifice being 0 m.With the increase of hole depth,the wave speed increased sharply first and then remained basically stable.An unstable wave speed suggests that the surrounding rock is damaged,whereas a stable wave speed indicates that the surrounding rock is close to an unexcavated state and not damaged.Hence,the excavation-induced damage zones corresponding to the three methods were obtained (Table 6).For CB,the wave speed was stable in the range of 4543-5013 km/s.The damage depths of the eight holes were 2.2 m,2.4 m,2.6 m,2.3 m,2.2 m,2.2 m,2.4 m and 2 m,with an average damage depth of 2.3 m.For SCB,the wave speed fluctuated in the range of 4529-5107 km/s,and the average damage depth of the eight holes was 1.9 m.For IE,the wave speed ranged from 4560 km/s to 5161 km/s,and the average damage depth was 1.7 m which was 11% smaller than that of SCB and 26% smaller than that of CB.IE corresponded to a smaller damage depth because it broke the rock through detonation gas pressure without generating any detonation stress waves.The reduction of the excavation-induced damage range conduces to the long-term stability and safety of the roadway and can optimize the effective supporting length of the bolt.According to the excavation-induced damage depth,the excavation-induced damage zones corresponding to the three excavation methods were obtained (Fig.24).

Fig.19.Numerical results of the excavation effect of the peripheral holes using the CB method.

Fig.20.Numerical results of the excavation effect of the peripheral holes using the SCB method.

Fig.21.Numerical results of the excavation effect of the peripheral holes using the IESF method.

5.2.Comparison of three excavation methods

Fig.22.Layout of acoustic test holes.

A comparison of the three excavation methods is provided in Table 7.As can be observed,the use of IE decreased the depth and area of over/under-excavation,led to a shrinkage of the excavationinduced damage zone,and increased the number of semicircular blasthole remnants of peripheral holes,DFRTS,and blasthole utilization rate.The IESFs allowed the cracks in the rock to propagate along the tunnel cross-section contour.With the aid of the energyfocusing effect,it neatly and smoothly excavated the tunnel,following the designed cross-section.Furthermore,the use of IE eased the damage because shock waves were transferred into the surrounding rock in the non-energy-focusing direction.In this way,it retained the original strength and stability of the tunnel,facilitated the long-term stability of the surrounding rock after excavation,and realized rapid tunnel excavation.

5.3.Comparison of the performances of explosives,CO2 blasting and IE

The performances of explosives,CO2blasting and IE are compared in Table 8.Explosives contain nitric acid.After being triggered by a detonator,they generate detonation waves and explosive gas within 10-6-10-5s to break the rock.Its advantages include (1) great power and significant crushing effect,(2) low blasting cost (Wang et al.,2021) and wide application range,(3)simple blasting process and high blasting efficiency,(4)good effect of directional rock breaking,and(5)no need of a free face.It has the following six shortcomings: (1) Their use is restricted in some holidays,and government approval procedures are complicated at ordinary times in China; (2) Blind artillery handling is dangerous;(3)The results of the United Nations partition test,the Kenan test,and the time/pressure test are all“+“,i.e.they are rather dangerous to transport and use;(4)The considerable vibration and noise they produced affect the surrounding buildings and residents(Xu et al.,2017; Yang et al.,2017); (5) It will produce harmful gases such as NOx and CO (Harris and Mainiero,2008; Abdollahisharif et al.,2016; Torno and Torano,2020); and (6) The blasting will result in a large range of loose circle.

Table 6 Excavation-induced damage depths (unit:m) of three excavation methods.

Table 7 Comparison of the excavation effects of the three excavation methods.

Table 8 Performance comparison of explosives,CO2 blasting and IE.

Fig.23.Curves of wave speed versus depth: (a) CB,(b) SCB,and (c) IE.

Fig.24.Excavation-induced damage zones of three excavation methods.

CO2blasting transforms liquid CO2into gaseous CO2within 4 × 10-3s by means of chemical heating (Bai et al.,2020; Shang et al.,2022) and break the rock through the physical work of gas expansion (Hu et al.,2018).It has the following advantages: (1) It does not require strict approval from government departments and is easy to use; (2) Mild vibration and noise produced only exert a limited impact on construction workers and surrounding residents;(3)No harmful gas is produced(Yang et al.,2019);(4)The results of the United Nations partition test,the Kenan test and the time/pressure test are all“-“,i.e.it is safe to transport and use(Yang et al.,2020); and (5) It does not require a pyrotechnic warehouse for storage and can be managed easily.However,it also has some shortcomings: (1) The power is weaker than explosives; (2) The cost is much higher than explosives; (3) The blasthole is larger in diameter,which makes its construction more time-consuming;(4)The effect of directional rock breaking is poor; (5) The heating device contains hazardous chemical substances; and (6) A free surface is required for use.

IE,whose most components are derived from solid wastes,such as coal gangue and straw,conduces to realizing the reuse of waste and protecting the environment.It uses electric current to excite the fuse and triggers the expanding agent within 0.05-0.5 s.A large amount of CO2and water vapor is produced through chemical changes to break the rock.It has many advantages: (1) The cost is low,only 1/3 the cost of explosives;(2)No government approval is required,and it is allowed to be used during holidays in China;(3)The results of the United Nations partition test,the Kenan test and the time/pressure test are all“-“,i.e.it is safe to transport and use;besides,its construction procedure is simple; (4) It is green and environmentally friendly for reuse of solid wastes;(5)It possesses many functions,as it can not only produce multiple cracks in the rock but also directionally fracture the rock; (6) The range of the loose circle is smaller than that of explosives;(7)It is non-explosive and highly safe;(8)Blind guns are easy to handle and can be treated with water; and (9) IE saves about 2 min to install a hole than explosives for the following reason.In explosive blasting,the detonator and the explosive cannot be connected in advance; instead,they can only be connected when used on site.However,for IE,the special splitting agent and the fuze are pre-assembled and do not need to be connected at the time of use.The rest of the operation procedures are the same.It merely has a few disadvantages: (1) A free surface is required; and (2) A small amount of CO gas is generated.

6.Conclusions

In this study,a new tunneling technology called IE is proposed.The technology is divided into two categories,i.e.IESF and IEMF.IESF is used to directionally form the tunnel cross-section and IEMF is used inside the tunnel to fragment the internal rock.IE is obviously advantageous over traditional explosive blasting methods.Comparing IE excavations with CB and SCB excavations,the maximum over/under-excavation depth of the tunnel wall decreased by 62% and 22%,and the DFRTS increased by 14.8% and 2.54%,respectively.Both experimental and numerical results show that among the three excavation methods,IE achieves the best directional rock-breaking effect and corresponded to the smallest excavation-induced damage zone.Moreover,when IE is used for excavation,the resulting tunnel cross-section was smoother,which reduces the need for shotcrete,promotes work efficiency and improves the long-term stability of the tunnel.

Although the felt vibration of IE is milder than that of explosives,the specific magnitudes of explosive blasting and IE vibration failed to be quantitatively compared in this study.Data for comparison from this perspective will be collected from tunneling experiments in a future study.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by the Postgraduate Research &Practice Innovation Program of Jiangsu Province,China (Grant No.KYCX21_2368) that are gratefully acknowledged.