Microstructural Characteristics of Eri Silk Fibre in Different Layers of Cocoon

ABSTRACT Five layers, i.e. floss, upper, middle, lower and pelade, exist in cocoon shell from outer to inner portion. An attempt has been made for assessment of microstructural characteristics of eri silk fiber in different layers of cocoons. Wide angle x- ray diffraction assessment revealed increase of crystallinity as well as degree of orientation and decrease in crystal size besides lattice spacing remains at par from floss to pelade layers. So, there is an improvement of single fibre tenacity and initial modulus from outer to inner position of cocoon’s shell. AFM assessment depicts increase in surface roughness of silk fibre from floss to pelade portion as observed from surface topography and root mean square average (Rq) of height deviation values. Descending trend of single fibre length and diameter were observed from floss to pelade layers along with significant reduction in lower and pelade layers vis-à-vis floss, upper and middle layers. Fibre quality index (FQI) can be measured from average single fibre length, fibre fineness and tenacity in five different layers. Expression has been derived for prediction of yarn tenacity from FQI and yarn count and good correlation exists between predicted and actual values.


Introduction
Eri is a domesticated non-mulberry silkworm Philosamia ricini under order Lepidoptera and family Saturnüdae. The silkworm is fed with the leaves of castor (Ricinus communis) which is primary food plant. Assam, Meghalaya, Nagaland, etc., of north-east India are the major producers of eri silk cocoons. Eri rearing is very popular in China, Thailand, Madagascar and Egypt. Eri cocoons are very soft and fibrous. Mulberry, tasar and muga cocoons can be used for reeling whereas eri cocoons are unreelable due to discontinuous filament as these are open mouthed. So, it is used for spinning (Jolly et al. 1979). Mulberry is well known for its luster, beauty, refinement, sensuality and elegance; but the importance of eri silk is next to mulberry for soft and warm characteristics.
Studies on structural characteristics reported that eri cocoons are soft and fibrous and the cocoon shells consist of multiple thin and paper like layers with existing several air gaps between layers (Komatsu 1980;. Analysis of cross-section images of eri silk fibers and its secondary structures revealed that, eri silk fibers consist of porous and non-porous structure which may be due to impact of temperatures of the place (Prasong, Wilaiwan, and Yaowalak 2011). Evaluation of physical, structural, mechanical and thermal properties of eri and mulberry silk fibres reported no difference between places in Ethiopia beside 20 to 25°C temperature difference for water exhaustion and degradation (Melesse et al. 2020). From the outer to inner layers of cocoon shell, the fibres show a slight initial increase of linear density and then exhibits decrease in mulberry, tasar, eri and muga varieties. Mulberry is the finest in linear density followed by eri, muga and tasar. Crosssection of mulberry silk fibre is triangular as against the rectangular structure of non-mulberry silk fibres. Highest density along with birefringence values revealed more oriented compact structure for mulberry vis-à-vis non-mulberry silks (Gupta, Rajkhowa, and Kothari 2000). Strength of silk fibre is proportional to crystalline area which is equal to ellipse area determined by microscopical parameters. Muga silk fibre exhibits greater area followed by eri and tasar for which muga silk fibre has maximum tenacity (Divakara, Somashekar, and Roy 2009). Eri silk fibroin showed little shoulder peaks for the decomposition of each component indicates two steps for weight loss (Srisuwan, Narkkong, and Srihanam 2009). Tenacity of silk fibre reduces with increase of crystal size for mulberry and nonmulberry varieties (Bhat and Nadiger 1980).
Saturnüdae silks (eri, tasar and muga) produce more highly defined structural transition as compared to Bombyx mori (Malay et al. 2016). Quantitative estimation by FTIR and 13 C solid state NMR reported that mulberry has higher β-sheet content in silk fibroin as compared to three wild varieties. B. mori silk has largest nano crystallites size along fiber axis with maximum orientation. The β-sheet structure has influence on tensile characteristics of silk fibers besides toughness has less correlation (Guo et al. 2018). Comparative analysis between mulberry as well as wild non-mulberry silks revealed that, wild silks are significantly different in terms of thermal stability, degradation, bound water sustaining ability as well as their molecular mobility during glass transitions. Sericin of non-mulberry cocoons has less thermal stability but better water exchange mobility and so, the same can be utilized in biomedical applications (Mazzi et al. 2014). Un degummed Kenyan eri silk fibre has crystallinity about 17% which can be improved to about 30% after degumming (Oduor et al. 2021). XRD assessment reported that crystalline index is about 60% for eri and tasar against 72% for mulberry (Vyas and Shukla 2016). Mulberry silk fibroin produced at the beginning and end stage of each instar of silkworm has different properties. Tensile characteristics are better at end stage of each instar of silkworm attributed to higher β-sheet content and crystallinity assessed by FTIR and XRD (Peng et al. 2019). Synchrotron radiation FTIR microscopy analysis revealed that eri silk fibre has about 23% β-sheet, 10% of α-helix and 67% of random coil. The values are 20, 19 and 14% for muga and 15, 14 and 72% for tropical tasar silk fibroin. Crystallinity ranges between 47 to 53% for nonmulberry vis-à-vis 55% for B. mori besides Hermann's orientation factors are 0.87 and 0.85 for muga and tasar against 0.92 for eri and mulberry respectively (Fang et al. 2016).
Various studies reported about microstructural characteristics of mulberry and tasar silk fibers in different layers of cocoons (Chattopadhyay et al. 1997;Das et al. 2005). Structure and physical properties of silkworm cocoons were evaluated for 25 diverse types with focus on tensile and compressive characteristics along with gas permeation through cocoon walls (Chen, Porter and Vollrath 2012). Macro characterization along with evaluation of amino acid composition for mulberry, tasar, muga and eri reported maximum fibre length exists for mulberry and minimum for eri and muga. Descending trend of degumming loss, filament denier and moisture regain was found from outer to inner layers of cocoon whereas inverse trend was evaluated for fibroin density. SEM micrographs illustrate micro voids in cross-section for three non-mulberry silk fibre with diminishing trend of cross-sectional area of both mulberry and non-mulberry varieties. For mulberry silk, glycine, alanine and serine constitute about 82% of amino acid vis-à-vis 73% for non-mulberry silks with maximum proportion of alanine. Substantial proportion of amino acids with bulky side groups present in tasar, eri and muga with higher ratios of hydrophilic to hydrophobic amino acids (Sen and Babu 2004a). Similar study on structure property correlations revealed ascending trend of tenacity and initial modulus from outer to inner layers. The change of microstructural parameters has very good correlation with properties. Stress relaxation and inverse stress relaxation evaluation exhibits molecular relaxation in the retracted fibre (Sen and Babu 2004b). Also, significant differences of dye uptake were found between mulberry and wild silks as well as in different layers of cocoons. Higher dye uptake occurred for mulberry as compared to tasar, muga and eri. Among wild silk varieties, tasar fibre absorbs more dye. Descending trend of dye uptake from outer to inner layers of cocoons was noticed due to change of fibre characteristics (Sen and Babu 2004c). Wide angle x-ray scattering assessment of mulberry and non-mulberry silk fibers revealed that higher oriented molecular structure for mulberry whereas non-mulberry silk fibres have unique properties and the structure consists of extended polypeptide chains bonded together by lateral N-H-O hydrogen bonds to form anti-parallel chain pleated sheets (Kawahara 1998;Marsh, Corey, and Pauling 1955;Urs et al. 2015). Structural assessment of alanine and glycine residues of eri silk fibres following solid state NMR ( 15 N & 13 C) reported that fraction of alanine at oriented domains in silk fibres as 75% and that of glycine as 65% (Asakura et al. 1999). Also, secondary structure determined by solid state NMR of eri silk fibre opined that the most probable conformation in native fibres is an anti-parallel β-sheet but the fibroin produced from liquid protein of silk gland appears to be primary α-helical (Beek et al. 2000). Descending trend of eri silk single fibre denier and proportion of silk fibres' weight were found from floss to pelade layers. But the single fibre tenacity increases from outer to inner layers beside breaking elongation remains at par (Chattopadhyay, Chakraborty, and Chatterjee 2018;Sen and Babu 2004a). Strength of eri silk fibre can be enhanced by about 15% with superior ultraviolet protection factor (UPF) by producing flat fibre web instead of cocoon spinning. For this purpose, silkworm larvae of 5 th instar were fed by castor leaves sprayed with nano TiO 2 and graphene oxide (GO) . Assessment of natural pigment in silk cocoons' shells reported about accumulation of pigments in sericin layer surface which are not durable. NMR analysis revealed presence of all-trans-lutein, and other five components as all-trans-neoxanthin, all-trans-violaxanthin, all-trans-α-carotene, all-trans-β-carotene and 9-cis d-carotene as evaluated by UV spectra and EI test data (Ma et al. 2016). Another study opined that yellow pigments belong to lipochrome or carotenoids and green pigments are flavones in mulberry cocoons whereas brown pigments in wild silk are soluble chromogen which turns brown due to oxidation (Bergmann 1939). For combed cotton compact yarns, prediction expression has been established for yarn tenacity based on 50% span length (mm), fibre bundle strength (g/tex), micronaure (μg/inch) and yarn count (tex) (Chellamani, Kumar, and Vittopa 2009). Similarly, for 100% wool worsted yarns, prediction of breaking strength was estimated based on diameter (micron), yarn count (tex) and fibre strength (g/tex) (Holdaway 1965).
From the literature cited, it is revealed that very limited studies were carried out about the microstructural characteristics like Wide Angle X-Ray Diffraction (WAXD) and Atomic Force Microscopy (AFM) of eri silk fibers in different layers of cocoons and its' influence on fibre quality. Also, no research was found for assessment of single fibre length and diameter in different layers of eri cocoon as well as prediction of yarn tenacity from single fibre quality parameters. This research works pertains with a view to these objectives.

Materials
Eri silk cocoons of white variety from Borduar, Kamrup, Assam (India) and brick red variety from Kokrajhar, Boroland Territorial Council, Assam (India), were procured for this study. The shell weight was 0.45 g and 0.50 g respectively for white and brick red eri cocoons. Sodium carbonate and neutral soap Extran MA02 (pH 6 to 8) of laboratory reagent grade from Merck, Germany, were utilized for degumming of eri cocoons.

Degumming of eri silk cocoons
White and brick red eri silk cocoons each 250 g without pupae were degummed using 10% sodium carbonate and 10% neutral soap concentrations on weight of material with one-hour boiling followed by hot and cold washing (Chattopadhyay, Chakraborty, and Chatterjee 2017). After drying of degummed eri cocoons using hot air dryer for 20 min at 105 ± 3°C temperature (BSI (British Standards Institution) 1974); the different quality parameters' assessments were carried out. The degumming loss was estimated at about 10.5% for white and 12.0% for brick red eri silk cocoons.

Extraction of silk fibers from different layers of eri cocoons
It is quite difficult to segregate different layers from eri silk cocoons without degumming due to presence of sericin and wax (Komatsu 1980). So, withdrawal of fibers from different layers of cocoons was carried out after degumming followed by drying. The fibres from the floss layer were withdrawn at first followed by upper, middle, lower and pelade (Chattopadhyay, Chakraborty, and Chatterjee 2018). The schematic diagram of different layers in cocoon along with extraction process sequence are illustrated in Figure 1. The images of silk fibres (Chattopadhyay, Chakraborty, and Chatterjee 2016) in different layers of cocoon for both varieties are shown in Figure 2. During extraction of fibres from different layers of cocoons intensive care was taken so that no stretch was imposed on silk fibres.

Conditioning of fiber samples
Degummed eri silk fiber samples were conditioned by keeping at standard atmospheric conditions i.e., 65 ± 2% relative humidity and 27 ± 2°C temperature for 24 hours (ASTM 2015a) subjected for assessment of X-ray Diffraction (XRD), Atomic Force Microscopy (AFM), denier, tenacity, initial modulus, length and diameter.

X-ray diffraction of silk fiber
X-ray wide angle scattering (WAXS) assessment of eri silk fibers was done by Rigaku X-Ray Analyser, Japan (Smart Lab), with angle of incidence from 10 to 90°.The X-ray used was nickel filtered CuK α radiation of wavelength 1.5406 Å. For this purpose, about 1 mm length of fibre snippet was cut for each sample precisely. Crystallinity (%), degree of orientation (f c ), distance between lattice planes/ lattice spacing (d) and crystal size (D) of fibre in different layers were estimated following various earlier research studies (Guo et al. 2018;Malay et al. 2016;Oduor et al. 2021;Peng et al. 2019;Sen and Babu 2004b) and the expressions as mentioned below: where A c is the integrated area of crystalline phase A a is the integrated area of amorphous phase. (A c + A a ) is the total area under the curve Degree of orientation f c (%) following Hermann's orientation function where α is the angle between the fiber axis and crystallographic axis Distance between lattice planes/lattice spacing (d) following Bragg's law where d = distance between lattice planes/lattice spacing n = an integer λ = x-ray wavelength θ = angle of incidence with lattice plane Crystal size D hkl following Scherrer's expression where D hkl = crystal size/thickness perpendicular to lattice plane (hkl) K = constant depend upon crystal shape λ = x-ray wavelength B = full width at half max (FWHM) or integral breadth θ B = diffraction angle of the (hkl) reflection For crystallinity (%) estimation, the area under crystalline region and total area under the curve were done using Origin21 software.

Assessment of single fiber tensile characteristics and denier
Single fiber tensile characteristics like breaking load, tenacity, initial modulus and breaking elongation in different layers of eri silk cocoons were measured using Lenzing Vibrodyn instrument (ASTM (American Society for Testing and Materials) 2014). For tenacity estimation, single fibre denier was assessed by Lenzing Vibroscope instrument (ASTM 2012a).

Atomic force microscopy (AFM)
Atomic force microscope (AFM) Mode Tap 190 AIG, Asylum Research MFP3D equipped with scanner Z axis-0.25 nm as well as X and Y axis-0.25 nm was used for this study. The fibers were immobilized on stubs with adhesive tapes for imaging in atmospheric condition under non-contact (tapping) mode with using tips from Budget Sensors. The frequency, k value, length and scan range were maintained as 190 kHz, 48 N/m, 225 μm and 1 Hz during assessment. Root mean square (RMS) of roughness data were obtained using the size 2 μm X 2 μm specimens.

Assessment of fiber length
Length of fiber from each layer of cocoon was measured by following Oil Plate Method (BISFA (International Bureau for standardization of manmade fibres) 1998) since both very long and short length fibers exist. About 250 mg conditioned fibre sample was taken so that the mass contains for about 2000 fibres (considering 2.0 denier fineness and 550 mm average length). The fibres were spread over a velvet cloth. Individual fibre was taken by tweezer and put on a glass plate covered with thin layer of paraffin oil. The fibre in crimped condition straightened gently by finger and tweezer. After straightening, the length of fibre was measured by scale with marks of 1 mm intervals. Total 50 individual fibres were assessed by this technique and statistical parameters (average, standard deviation etc.) were estimated.

Assessment of single fiber diameter
The diameter of eri silk fiber in different layers was estimated by Optical Fibre Diameter Analyzer (OFDA), Make BSC Electronics, Australia, Model OFDA 2000 following standard procedure (ASTM 6500-00 2012b). At first, fibre samples were cut into snippets of length about 1.8 mm using guillotine. Then the fibre snippets were put into slide spreader which individualize and it is placed below the laser scanner. By the movement of glass slide in each direction horizontally the fibre diameter being measured through image analysis technique. Average fibre diameter (micron), standard deviation and frequency distribution of diameter were obtained through computer interface.

Scanning electron microscopy (SEM) analysis
Surface morphology assessment of fiber samples extracted from different layers of cocoon was carried out using Scanning Electron Microscope (JEOL, Japan, Model JSM-6390 LV) with resolution of 100 μm at 10 KV. At first the samples were spur coated using platinum and then micrographs were taken with proper magnification so that clarity of the images exist.

Prediction of yarn tenacity from single fiber characteristics
From earlier study, it was found that about 65%, 25%, 5%, 3% and 2% proportion of fibers exists in floss, upper, middle, lower and pelade layers respectively in eri cocoons. Also, there is descending trend of single fibre denier along with ascending trend of single fibre tenacity were revealed (Chattopadhyay, Chakraborty, and Chatterjee 2018). Considering the silk fibers remain in yarn cross section as per the proportion, the weighted average of single fibre tenacity can be derived from the following expression Average single fibre tenacity T ¼ 0:65tl 1 þ 0:25tl 2 þ 0:05tl 3 þ 0:03tl 4 þ 0:02tl 5 where tl 1 , tl 2 , tl 3 , tl 4 and tl 5 are the single fiber tenacity in floss, upper, middle, lower and pelade layers respectively. Similarly, overall single fibre denier can be estimated by following expression D ¼ 0:65dl 1 þ 0:25dl 2 þ 0:05dl 3 þ 0:03dl 4 þ 0:02dl 5 where dl 1 , dl 2 , dl 3 , dl 4 and dl 5 are the single fiber denier in floss, upper, middle, lower and pelade layers respectively. The weighted average values for single fibre tenacity and denier are given in Table 1. Yarn samples were prepared for 40 Nm (25 tex), 60 Nm (16.70 tex) and 80 Nm (12.50 tex) counts following standard process parameters in long staple worsted system (N. Schlumberger and Cie 1994).
For each trial, average fiber length Hauteur (H) was assessed by standard procedure (IWTO (International Wool Textile Organization) 2011) of finisher gill box draw frame sliver and 10 trials were conducted for each yarn count. Hauteur is the cross section biased average length which is commonly used for setting of machine parameters in long staple worsted system of yarn production (Lee 1993). It is revealed from earlier research studies (Chellamani, Kumar, and Vittopa 2009;Holdaway 1965) that fibre strength, fibre length and fibre fineness are the three major parameters which have maximum influence on yarn tenacity. Among these, fibre length and strength are directly proportional whereas fibre fineness is inversely proportional. So, a parameter "fibre quality index" has been formulated as following expression.
Fibre quality index FQI ð Þ ¼ k: Single fibre length mm ð Þ � Single fibre tenacity g=tex ð Þ Fibre fineness tex ð Þ where k is the constant. The following equation has been derived based on linear programming technique for prediction of yarn tenacity.
Yarn tenacity g=tex ð Þ ¼ 2:14 p FQI À 0:07 yarn count tex ð Þ þ 0:98 The predicted values for yarn tenacity were compared with the actual measured yarn tenacity values. Actual yarn tenacity was assessed by using the Uster Tensorapid 4 following standard procedures (ASTM D 2256/2256 M 2015b) keeping gauze length 50 mm and traverse speed 500 mm/min and 50 readings were taken for each yarn sample.

Analysis of test results
Experimental data were analyzed using MINITAB 19 statistical software for different quality parameters. Analysis of X-ray spectrograms were carried out using ORIGIN 21 software to estimate crystallinity (%), degree of orientation, lattice spacing and crystal size.

X-ray diffraction (×RD) of silk fiber
The X-ray diffraction spectrograms of eri silk fiber in different layers for white and brick red cocoons are illustrated in Figure 3. It is observed from Figure 3 that for both white and brick red eri silk varieties, the intensity of peak increases from outer to inner layers along with decrease in peak width. This phenomenon establishes the fact that the crystallinity improves silk fibroin from outer (floss and upper) to inner (middle, lower and pelade layers). Between 2θ angle 15 to 25°, two peaks were observed in five different layers of both eri cocoons and maximum intensity exists in lower and pelade layers. Between 20 to 30° angle one peak was noticed in five layers with ascending trend of intensity from middle to lower and pelade layers. This may be due to change of α-helix to β-sheet structure of fibroin molecules by the higher extrusion pressure through the spinneret of silkworm as the volume of liquid silk reduces during formation of middle, lower and pelade layers. In case of brick red eri cocoon, very high intensity peaks at 15° and 20° angle of diffraction for lower and pelade layers which may be due to more extrusion pressure of  brick red eri silkworm during persuasion. As the neutral fibroin passing from the middle gland to spinneret through anterior gland; the fluid contacts with mild acidic solution and becomes coagulated and dehydrated. The coagulated fluid emerges through the spinneret as silk fiber. This is the turning point and by the movement of silkworm head, the coagulated silk becomes transformed into fibers due to dragging action. Since the volume of liquid silk is less at lower and pelade layers, so surface cohesion due to dragging action results maximum orientation of fibroin molecules (Iizuka 1980a;Iizuka et al. 1994;Ogihara 1938). Earlier study reported that correlation co-efficient is − 0.87 between initial modulus and single fibre denier and − 0.96 between crystallinity (%) and single fibre denier which denote that initial modulus as well as crystallinity are higher in finer fibres (Iizuka 1980b). The crystallinity (%), degree of orientation (%), lattice spacing and crystal size of silk fibres in five different layers are given in Table 2. It is revealed from Table 2 that the extent of silk fiber crystallinity as well as degree of orientation improve from outer to inner layers of eri cocoon although significant difference not exists which is in accordance of earlier research study (Sen and Babu 2004b). The crystallinity (%) estimated between 31 to 38% whereas degree of orientation as 92 to 97%. Earlier research reported that eri silk fibre crystallinity as 33% with orientation factor 95% whereas it is about 45% and 98% respectively for mulberry (Guo et al. 2018;Malay et al. 2016;Peng et al. 2019). The crystallinity of Kenyan eri silk fibre reported as 29% (Oduor et al. 2021). Hence, it can be opined that the orientation of fibre molecules is enhanced due to higher extrusion pressure from spinneret of silkworm and thereby increase of tensile characteristics (Chattopadhyay et al. 1997;Iizuka et al. 1994). No difference being noticed for lattice spacing between five different layers. The effect of double orientation in the macromolecular structure is less apparent on the layer line reflections than on the equatorial reflections which is great enough in some cases to permit a direct correlation of certain layer line reflection with equatorial reflection having the same values of d hkl (Marsh, Corey, and Pauling 1955). Also, it was found that there is descending trend of crystal size from outer to inner layers of silk cocoon. Significant difference exists between floss as well as upper layers to middle, lower and pelade layers Babu 2004a, 2004b). For non-mulberry silks, 47% alanine, 27% glycine and 11% serine exist and silk belongs to P 212121 with difference in β-structure and the axis along c-axis. Alanine concentration estimated in helical region whereas glycine is found in non-helical portions (Iizuka 1980b). Quantification of the abundance of poly (ala) residues revealed that S. c. ricini contained the highest level about 45% of the total length of tandem repeat (Malay et al. 2016). So, due to change of crystallinity as well as degree of orientation, the proportion of amino acid may vary from outer to inner layers causes decrease in crystal size. The tenacity of silk fibre decreases with increase of crystal size and vice-versa (Bhat and Nadiger 1980;Gupta, Rajkhowa, and Kothari 2000). Crystal size is inversely proportional to fineness of the silk fibre (Somashekarappa et al. 2005). The stress-strain curves of eri silk fibre in five different layers of both white and brick red cocoons are illustrated in Figure 4. It is revealed that the stress of silk fibre is enhanced from middle layer to inner layer of cocoon for both varieties. This is due to better orientation of

JOURNAL OF NATURAL FIBERS
fibroin molecules as observed from x-ray diffraction assessment. Also, ascending trend of silk fibre tenacity as well as initial modulus was found from outer to inner layer ( Figure 5) of both eri cocoons. These findings are in accordance to earlier research works carried out for mulberry, eri, tasar and muga fibers (Chattopadhyay et al. 1997;Chattopadhyay, Chakraborty, and Chatterjee 2018;Sen and Babu 2004a). X-ray diffraction assessment revealed that no difference of crystallinity and degree of orientation between two cocoon varieties whereas ascending trend was noticed from floss to pelade layers which may be the reason for significant difference of tenacity and initial modulus between layers. Between 15° and 20° angle of diffraction, very high intensity peaks were observed in case of brick red eri which may be the reason for higher initial modulus of single fibre.

Atomic force microscopy (AFM) of eri silk fibers in different layers
The AFM images of eri silk fiber surface topography in five different layers of cocoon are illustrated in Figures 6 and 7 for white and brick red varieties respectively. It is observed that the surface roughness of eri silk fibre increases from floss to pelade layers. The root means square average of height deviation "R q " values of eri silk fibre in different layers are given in Table 3.  From Table 3, it can be depicted that the root mean square (R q ) value of surface roughness increases from outer to inner layers. As the cocoon formation process propagates, the volume/quantity of liquid silk is reduced and thereby more pressure is exerted through the spinneret by silkworm larva. Due to lack of extrusion pressure uniformity, the surface smoothness varies. This may be the reason for higher R q values in case of middle, lower and pelade layers vis-à-vis floss and upper layers. Silk fibers from brick red eri cocoons has higher R q value in inner layers which may be attributed by higher extrusion pressure through spinneret during cocoon formation. Studies reported that due to presence of scales on surface of wool fibre; R q value is more as compared to cotton and polyester fibers (Koc 2015).

Single fiber length
The single fiber length in five different layers of white and brick red eri silk cocoons along with analysis of variance (ANOVA) are mentioned in Table 4.
It is revealed from Table 4 that significant reduction of average fiber length (5% & 1% confidence level) in lower and pelade layers as compared to floss, upper and middle layers for both cocoon varieties whereas no difference between two types. During initiation of spinning, higher volume of liquid silk is present in worm's gland and thereby it throws large amount of silk through spinneret. With continuation of cocoon spinning process, the volume gets reduced along with quantum of silk fibers. Before termination due to existence of very little volume of liquid silk in worm's gland; the emergence of fibres gets reduced substantially during formation of lower and pelade layers (Gupta, Rajkhowa, and Kothari 2000;Iizuka 1980a). This may be the reason for lower average length of fibres in lower and pelade layers.

Single fiber diameter
The single fiber diameter measured by OFDA is illustrated in Figure 8 and SEM images in Figures 9 and 10 respectively for white and brick red varieties. It is noted from Figures 8, 9 and 10 that the diameter of fibre exhibits descending trend from floss to pelade layers. Similar observations were reported about reduction of single fibre denier for mulberry, tasar and eri silk cocoons (Chattopadhyay et al. 1997;Chattopadhyay, Chakraborty, and Chatterjee 2018;Sen and Babu 2004b). Due to propagation of cocoon spinning process, the volume of liquid silk in gland gets reduced and so quantum of fibre emerged from spinneret of silkworm becomes lower (Iizuka 1980a) and hence decreasing trend of diameter is revealed.

Tenacity of eri silk yarn estimated from fiber characteristics
The single fiber length Hauteur (H) (mm), yarn count (tex) and fibre quality index (FQI) are given in Table 5 which were used for estimation of yarn tenacity by the expression (8). The predicted values and actual values of yarn tenacity are illustrated in Figure 11. It was found that good correlation exists between actual and predicted values of yarn tenacity with correlation coefficient 0.85. The difference between actual and predicted values of yarn tenacity is evaluated as minimum as 1.50%. So, it can be opined that the expression can be well utilized for prediction of yarn tenacity from fibre quality index and yarn count. Fibre quality index (FQI) can be estimated from fibre length, tenacity and fineness as per the Equation (7). Since there is an ascending trend of fibre tenacity and descending trend of fibre fineness being found from outer (floss) layer to inner layer (pelade), so weighted average values for these parameters can be considered according to the different proportion of silk fibers present in five different layers of eri cocoon.

Conclusions
For both white and brick red eri silk, the intensity of peak increases from outer to inner layers along with reduction of peak width and thereby silk fibroin crystallinity as well as degree of orientation improve. Between 15 to 25° angle of diffraction, two peaks were found in five layers and maximum intensity exists in lower and pelade. Also, between 20 and 30° angle, one peak present in each layer with ascending trend of intensity from middle to lower and pelade layers. This may be due to change of α-helix to β-sheet structure of fibroin molecules by higher extrusion pressure through spinneret of silkworm during formation of middle, lower and pelade layers as the volume of liquid silk reduces. No difference was observed for lattice spacing in five layers. Descending trend of silk fibroin crystal size from outer to inner layers of cocoon was revealed with significant difference between floss as well as upper to middle, lower and pelade layers. Due to change of crystallinity and orientation, the proportion of amino acid varies which causes decrease in crystal size. So, ascending trend of tenacity and initial modulus was observed from floss to pelade positions. Surface topography assessed by AFM depicts the increase of surface roughness from floss to pelade layers. Also, the root mean square (R q ) value of surface roughness enhances from outer to inner portions of eri cocoon. Substantial reduction for volume of liquid silk occurs during formation of inner layers of cocoon and thereby more pressure is exerted through spinneret by silkworm larva. Due to lack of extrusion pressure uniformity, the surface roughness varies which may be the reason for higher R q values in inner    layers. The length of single fiber exhibits descending trend from outer to inner layers as the volume of liquid silk in gland gets reduced during propagation of cocoon formation. Significant difference estimated in lower and pelade layers vis-à-vis floss, upper and middle of both varieties whereas no difference between two types. Also, decreasing trend of single fibre diameter was found from outer to inner layer of cocoons as obtained from OFDA and SEM assessment. Eri spun silk yarn tenacity can be predicted from fibre quality index (FQI) and yarn count. The FQI value is estimated from average fibre length as well as single fibre tenacity and fineness in different layers of eri cocoon. Good correlation coefficient 0.85 exists between predicted and actual yarn tenacity values.