Application of Macro-Synthetic-Fiber-Reinforced Sprayed Concrete for Primary Lining Support on Highway Tunnel Project in Bosnia and Herzegovina

The application of macro synthetic fiber reinforced sprayed concrete is seen by many design engineers as the alternative to steel reinforced sprayed concrete. The first applications of macro synthetic fiber reinforced sprayed concrete in Bosnia and Herzegovina were applied during the construction of the Sarajevo By-pass Motorway. The Project included a Motorway tunnel-system of three (3) double tubed, two-lane tunnels receptively 860m,

The carriageway width is 7.0m, and consists of two lanes of 3.50m.

Underground Conditions
The Underground conditions (geological formations) on the location of Tunnel T3 were diverse and complex. Three (3) basic types of rock were found (Sedimentary, Metamorphic and Magmatic rocks) [2]. The Metamorphic rocks were from Paleozoic and the Dolomites were from Devon. In-between two Magmatic layers of Quartz Porphyries were also found ( Figure 2).

Tunnel excavation and support elements (Main Design of the Project)
In accordance to the Main Design documentation in the Project the excavation works were divided into three (3) phases. The top heading, bench and the invert arch excavation phase. In better underground conditions the excavation works were divided in top heading and bench phases only [3]. The Main Design defined rock categories, predicting the round lengths and defining support elements (system) for each of the rock category respectively (Table  1).

Tunnel T3 excavation works
At the early beginning of the Tunnel T3 Project it was obvious that the predicted rock mass categories given in the Main Design are different compared to the actual underground conditions on site. At the entrance portal and the tunnel precut it was identified that the Main Design was predicting 4 th (IV) rock mass category and actually 5 th (V) rock mass category was present. Consequently it was necessary to apply other support elements, demanding more time (require longer installation time) then predicted. Due to the underestimation of underground conditions inevitably delays in the excavation works in Tunnel 3 were going to occur. Only undertaking appropriate measures it was possible to mitigate delays and finish the Project within the expected program of work.
According to the geological predictions approximately two thirds of the Tunnel T3 was located either in dolomitic limestone (approx. 900m) or quartz-porphyric metamorphic rock (approx. 900m). In such underground conditions it was possible to modify the conventional support elements (systems) and apply fiber reinforced sprayed concrete instead of mesh reinforced sprayed concrete. It has been estimated that the changes in these support elements would probably save 2 to 3hours per round length (step of excavation) [4]. The length of the tunnel rock mass categories where fiber reinforced sprayed concrete could be used was estimated to be 1800m in length. Average round length (step of excavation) was estimated to be 2.5m. 1800m/2.5m = 720 steps 720 (steps) x (2-3hours-shorter installation time) =makes up to 1440 to 2160hours

Calculation
In order to implement this idea a 60m test field in Tunnel T3 was formed to compare the behavior of the support elements in real underground conditions. The behavior of mesh reinforced sprayed concrete (cross section 24.300-24.330; 30m) and fiber reinforced sprayed concrete (cross section from 24.330m to 24.360m ; 30 m)lining had to be tested. Measurement profiles were installed (measuring the movements in the primary lining for a period of time until the settlements stabilize) at 24.315m (cross section with mesh reinforced sprayed concrete) and 24.345m (cross section with fiber reinforced sprayed concrete).

The updated Q-System
The relation between the rock-mass quality and the support elements (supporting system) using fiber reinforced sprayed concrete is defined by the updated Q-System by Grimstad and Barton (Figure 3). Identifying the rock-mass quality (Q) for certain rocks the supporting elements (system) can be predicted using the updated Q-System.
For the intended underground conditions in Tunnel T3 where fiber reinforced sprayed concrete could have been used, the rockmass quality index was ranging between 1 and 10 (D-Poor, C-Fair, B-Good). Considering the Span or Height ratio (for ESR =1) being10 the expected reinforcement categories 4, 5 and 6 would need energy absorption capacity between E=500-700J ( Figure 4).

The test field of macro-syntheticfiber reinforced sprayed concretein tunnel T3
The interaction of the rock-mass and the tunnel primary lining could be observed during the construction, making real scale measurements. Measuring the displacements of the rock-mass the impact on the tunnel support elements could be observed in real time. According to the NATM concept, measurement profiles are installed systematically at characteristic sections along the tunnel. The aim of the monitoring is to predict and to prove the adequacy of the supporting system. This approach was also applied in the test field with the aim to confirm the adequacy of the fiber reinforced sprayed concrete lining in Tunnel T3. The back calculation analysis was used to determine the relevant mechanical characteristics of the rock-mass and to refine the support measures. The test field of a 60m section of the both mesh and fiber-reinforced sprayed concrete lining was installed to examine if the fiber reinforced sprayed concrete primary lining could be equally efficient as the conventional primary lining when used in similar geological conditions [5]. In particular the bending capacity had to be assessed this way. To quantify the efficiency of the fiber reinforced sprayed concrete primary lining, the results from the test section had to be compared with the results of the measurement profiles featuring the conventional primary lining for the same overburden, and very similar geological conditions.

The test setup
The test setup was as follows: i.
In to the already used sprayed concrete mixture 4kg of polypropylene fibers were added v.
Measuring the profiles on cross sections at 24.315m (mesh reinforced sprayed concrete) and 24.345m (cross sectionfiber reinforced sprayed concrete) was based on the geological mapping and data collected by the geologist on the site (RMR=47). The geotechnical parameters were assumed using the Hoek & Marinos [6] method taking into account all the relevant indicators of the rock mass behavior including the favorableness of the rock layering. Relevant parameters (σci=40 MPa, GSI=42, mi=10, D=0, 5, MR=500, γ= 0, 025 MN/m3, Tunnel Depth=65m) where analyzed using the software ("Roc Lab") to acquire Mohr-Coulomb strength parameters for numerical analysis (c=0, 274MPa; phi=42, 86deg; E=1744, 22MPa). After the backcalculation the following results were computed (Figure 6 & 7): The bending moments (Md, max=8, 78kNm/m) and axial forces (Nd, max=489, 80N/m) are plotted on the interaction diagrams (Figure 7) of the mesh reinforced sprayed concrete for the installed sprayed concrete C25/30 and reinforcement steel mesh type-Q 188. The predicted bending capacity of the primary lining in the B1 rock mass category should not exceed 20kNm/m.

Conclusion
The macro-synthetic fiber reinforced sprayed concrete samples have shown less energy absorption capacity in comparison with the mesh reinforced sprayed concrete samples determined by the EFNARC plate test (EN 14485-5;2006). The required energy absorption capacity according to the updated Q-System rock classification by Grimstad & Barton [2] was between E=500-700J see ( Figure 3) and the tested energy absorption capacity of tested specimens using macro-synthetic fiber in the test field was of an average of E=740J see ( Figure 5). It could be concluded that the energy absorption capacity of the macro-synthetic fiber reinforced sprayed concrete was still of satisfactory capacity and could be applied for the support system in category B1.
The test filed results show a very good post cracking performance of the macro-synthetic fiber reinforced sprayed concrete at deflections of about 5mm. It was measured that the deflections of 5mm usually occur during the first hours after excavation.
Analyzing the energy absorption test results at the deflection of 5mm it can be concluded that both the mesh and fiber sprayed concrete capacity absorb almost the same force F=45kN see (Figure 4 & 5). Considering these results the fiber sprayed concrete provides more safety during the first hours of excavation due to simple installation.
Analyzing the results of the back calculation the Md, max=8,78kNm/m at the total displacement of 10mm is much more smaller compared to the predicted bending capacity of 20kNm/m in the primary lining for B1 rock mass according to ÖNORM B2203. The measurement profiles on cross sections at 24.315m and 24.345m showed almost the same deflections for both the mesh and fiber sprayed concrete primary lining. So it can be concluded that the fiber sprayed concrete lining can be equally efficient. The in-situ identified geological rock mass categories and the applied support systems are shown in Table 2. Comparing the geological rock mass categories from the Main Design (Table 1) and the categories identified during the excavation obviously shows that the share of 4th (IV) rock category was increased and on the other hand, the shares of 2nd (II) and 3th (III) geological rock mass categories were reduced.
After all, in Tunnel T3 almost 1600m of fiber reinforced sprayed concrete was applied. The net saving time using macro synthetic fiber in the sprayed concrete lining was 70days. This fully compensated the lost time for the increased length of 4th (IV) rock category that was not predicted by the Main Design. The Project was successful accomplished on time, with no delays. The application of fiber sprayed concrete using engineering know-how and experience played a dominant role in this successful construction project. Macro synthetic fiber sprayed concrete and fiber sprayed concrete in general should be considered as a powerful tool in tunnel engineering projects.